External Review Draft | EPA/600/R-15/047a | June 2015 | www.epa.gov/hfstudy
United States
Environmental Protection
Agency
              Assessment of the Potential
              Impacts of Hydraulic Fracturing
               or Oil and Gas on Drinking
              Water Resources
 Office of Research and Development
 Washington, D.C.

-------
DRAFT- DO NOT CITE OR QUOTE                              EPA/600/R-15/047a
                                                   External Review Draft

           United States
           Environmental Protection                          WWW.epa.
           Agency
 Assessment of the Potential Impacts of Hydraulic
    Fracturing for Oil and Gas on Drinking Water
                          Resources
                              NOTICE

     THIS DOCUMENT IS AN EXTERNAL REVIEW DRAFT, for review purposes only. It has not
     been formally disseminated by EPA. It does not represent and should not be construed to
     represent any Agency determination or policy.
                     Office of Research and Development
                    U.S. Environmental Protection Agency
                         Washington, DC 20460

-------
Hydraulic Fracturing Drinking Water Assessment
                                        DISCLAIMER
This document is an external review draft. This information is distributed solely for the purpose of
pre-dissemination peer review under applicable information quality guidelines. It has not been
formally disseminated by EPA. It does not represent and should not be construed to represent any
Agency determination or policy. Mention of trade names or commercial products does not
constitute endorsement or recommendation for use.
              This document is a draft for review purposes only and does not constitute Agency policy.
June 2015                                     ii                    DRAFT—DO NOT CITE OR QUOTE

-------
Hydraulic Fracturing Drinking Water Assessment
Contents
Contents	iii
List of Tables	x
List of Figures	xii
List of Acronyms/Abbreviations	xvi
Preface	xx
Authors, Contributors, and Reviewers	xxi
  Authors	xxi
  Contributors	xxii
  Reviewers	xxiii
  Acknowledgements	xxiv
Executive Summary	ES-1
  What is Hydraulic Fracturing?	ES-1
  Scope of the Assessment	ES-3
  Approach	ES-5
  Proximity of Current Activity and Drinking Water Resources	ES-5
  Major Findings	ES-6
    Water Acquisition	ES-6
    Chemical Mixing	ES-10
    Well Injection	ES-13
    Flowback and Produced Water	16
    Wastewater Management and Waste Disposal	ES-19
  Key Data Limitations and Uncertainties	ES-22
    Limitations in Monitoring Data and Chemical Information	ES-22
    Other Contributing Limitations	ES-23
  Conclusions	ES-23
  References for Executive Summary	ES-24
1.  Introduction	1-1
  1.1.    Background	1-1
  1.2.    Scope	1-1
  1.3.    Approach	1-6
    1.3.1.    EPA Hydraulic Fracturing Study Publications	1-6
    1.3.2.    Literature and Data Search Strategy	1-6
    1.3.3.    Literature and Data Evaluation Strategy	1-7
    1.3.4.    Quality Assurance and Peer Review	1-8
  1.4.    Organization	1-9
  1.5.    Intended Use	1-11
  1.6.    References for Chapter 1	1-12
2.  Hydraulic Fracturing, Oil and Gas Production, and the U.S. Energy Sector	2-1
  2.1.    What is Hydraulic Fracturing?	2-1
  2.1.    Hydraulic Fracturing and the Life of a Well	2-8
               This document is a draft for review purposes only and does not constitute Agency policy.
June 2015                                        iii                    DRAFT—DO NOT CITE OR QUOTE

-------
Hydraulic Fracturing Drinking Water Assessment
     2.1.1.     Site and Well Development	2-9
     2.1.1.     Hydraulic Fracturing	2-12
     2.1.2.     Fluid Recovery, Management, and Disposal	2-17
     2.1.3.     Oil and Gas Production	2-18
     2.1.4.     Site and Well Closure	2-19
  2.2.    How Widespread is Hydraulic Fracturing?	2-20
     2.2.1.     Number of Wells Fractured per Year	2-25
     2.2.2.     Hydraulic Fracturing Rates	2-27
  2.3.    Trends and Outlook for the Future	2-28
     2.3.1.     Natural Gas (Including Coalbed Methane)	2-28
     2.3.2.     Oil	2-31
  2.4.    Conclusion	2-32
  2.5.    References for Chapter 2	2-33
3.   Drinking Water Resources in the United States	3-1
  3.1.    Current and Future Drinking Water Resources	3-1
  3.2.    The Proximity of Drinking Water Resources to Hydraulic Fracturing Activity	3-3
  3.3.    Conclusion	3-11
  3.4.    References for Chapter 3	3-12
4.   Water Acquisition	4-1
  4.1.    Introduction	4-1
  4.2.    Types of Water Used	4-2
     4.2.1.     Source	4-2
     4.2.2.     Quality	4-4
     4.2.3.     Provisioning	4-5
  4.3.    Water Use Per Well	4-6
     4.3.1.     Hydraulic Fracturing Water Use in the Life Cycle of Oil and Gas	4-6
     4.3.2.     National Patterns of Water Use Per Well for Fracturing	4-6
     4.3.3     Factors Affecting Water Use Per Well	4-7
  4.4.    Cumulative Water Use and Consumption	4-8
     4.4.1.     National and State Scale	4-8
     4.4.2.     County Scale	4-9
  4.5.    Potential for Water Use Impacts by State	4-15
     4.5.1.     Texas	4-17
     4.5.2.     Colorado and Wyoming	4-28
     4.5.3.     Pennsylvania, West Virginia, and Ohio	4-32
     4.5.4.     North Dakota and Montana	4-36
     4.5.5.     Oklahoma and Kansas	4-39
     4.5.6.     Arkansas and Louisiana	4-41
     4.5.7.     Utah, New Mexico, and California	4.44
  4.6.    Chapter Synthesis	4-47
     4.6.1.     Major Findings	4-47
     4.6.2.     Factors Affecting Frequency or Severity of Impacts	4-49
     4.6.3.     Uncertainties	4-50

               This document is a draft for review purposes only and does not constitute Agency policy.
June 2015                                          iv                     DRAFT—DO NOT CITE OR QUOTE

-------
Hydraulic Fracturing Drinking Water Assessment
     4.6.4.     Conclusions	4-51
  4.7.    References for Chapter 4	4-54
5.   Chemical Mixing	5-1
  5.1.    Introduction	5-1
  5.2.    Chemical Mixing Process	5-3
  5.3.    Overview of Hydraulic Fracturing Fluids	5-6
     5.3.1.     Water-Based Fracturing Fluids	5-11
     5.3.2.     Alternative Fracturing Fluids	5-11
     5.3.3.     Proppants	5-13
  5.4.    Frequency and Volume of Hydraulic Fracturing Chemical Use	5-13
     5.4.1.     National Frequency of Use of Hydraulic Fracturing Chemicals	5-15
     5.4.2.     Nationwide Oil versus Gas	5-19
     5.4.3.     State-by-State Frequency of Use of Hydraulic Fracturing Chemicals	5-19
     5.4.4.     Volumes of Chemicals Used	5-24
  5.5.    Chemical Management and Spill Potential	5-28
     5.5.1.     Storage	5-30
     5.5.2.     Hoses and Lines	5-34
     5.5.3.     Blender	5-36
     5.5.4.     Manifold	5-36
     5.5.5.     High-Pressure Fracturing Pumps	5-37
     5.5.6.     Surface Wellhead for Fracture Stimulation	5-39
  5.6.    Spill Prevention, Containment, and Mitigation	5-41
  5.7.    Overview of Chemical Spills Data	5-42
     5.7.1.     EPA Analysis of Spills Associated with Hydraulic Fracturing	5-42
     5.7.2.     Other Spill Reports	5-47
  5.8.    Fate and Transport of Chemicals	5-50
     5.8.1.     Potential Paths	5-52
     5.8.2.     Physicochemical Properties	5-53
     5.8.3.     Mobility of Chemicals	5-55
     5.8.4.     Transformation Processes	5-64
     5.8.5.     Fate and Transport of Chemical Mixtures	5-64
     5.8.6.     Site and Environmental Conditions	5-65
     5.8.7.     Peer-Reviewed Literature on the Fate and Transport of Hydraulic Fracturing Fluid Spills5-66
     5.8.8.     Potential and Documented Fate and Transport of Documented Spills	5-66
  5.9.    Trends in Chemicals Use in Hydraulic Fracturing	5-70
  5.10.  Synthesis	5-71
     5.10.1.   Summary of Findings	5-71
     5.10.2.   Factors Affecting the Frequency or Severity of Impacts	5-72
     5.10.3.   Uncertainties	5-73
     5.10.4.   Conclusions	5-74
  5.11.  References for Chapter 5	5-77
6.   Well Injection	6-1
  6.1.    Introduction	6-1

               This document is a draft for review purposes only and does not constitute Agency policy.
June 2015                                          v                     DRAFT—DO NOT CITE OR QUOTE

-------
Hydraulic Fracturing Drinking Water Assessment
  6.2.    Fluid Migration Pathways Within and Along the Production Well	6-3
     6.2.1.     Overview of Well Construction	6-3
     6.2.2.     Evidence of the Existence of Fluid Movement Pathways or of Fluid Movement	6-11
  6.3.    Fluid Migration Associated with Induced Fractures within Subsurface Formations	6-27
     6.3.1.     Overview of Subsurface Fracture Growth	6-28
     6.3.2.     Migration of Fluids through Pathways Related to Fractures/Formations	6-31
  6.4.    Synthesis	6-50
     6.4.1.     Summary of Findings	6-51
     6.4.2.     Factors Affecting Frequency and Severity of Impacts	6-53
     6.4.3.     Uncertainties	6-55
     6.4.4.     Conclusions	6-57
  6.5.    References for Chapter 6	6-58
7.   Flowback and Produced Water	7-1
  7.1.    Introduction	7-1
     7.1.1.     Definitions	7-2
  7.1.    Volume of Hydraulic Fracturing Flowback and Produced Water	7-2
     7.1.1.     Flowback of Injected Hydraulic Fracturing Fluid	7-3
     7.1.2.     Produced Water	7-10
  7.2.    Flowback and Produced Water  Data Sources	7-13
  7.3.    Background on  Formation Characteristics	7-15
  7.4.    Flowback Composition	7-16
     7.4.1.     General  Characteristics	7-16
     7.4.2.     Temporal Changes in Flowback Composition	7-16
     7.4.3.     Total Dissolved Solids Enrichment	7-17
     7.4.4.     Radionuclide Enrichment	7-18
     7.4.5.     Leaching and Biotransformation of Naturally Occurring Organic Compounds	7-19
  7.5.    Produced Water Composition	7-22
     7.5.1.     Similarity of Produced Water from Conventional and Unconventional Formations	7-22
     7.5.2.     Variability in Produced Water Composition Among Unconventional Formation Types.7-22
     7.5.3.     General  Water Quality Parameters	7-25
     7.5.4.     Salinity and Inorganics	7-25
     7.5.5.     Metals	7-26
     7.5.6.     Naturally Occurring Radioactive Material (NORM) and Technologically Enhanced Naturally
     Occurring Radioactive Material (TENORM)	7-27
     7.5.7.     Organics	7-28
     7.5.8.     Reactions within Formations	7-28
  7.6.    Spatial Trends	7-29
  7.7.    Spill Impacts  on Drinking Water Resources	7-30
     7.7.1.     Produced Water Management and Spill Potential	7-30
     7.7.2.     Spills of Hydraulic Fracturing Flowback and Produced Water from Unconventional Oil and Gas
     Production	7-31
     7.7.3.     Case Studies of Potentially Impacted Sites	7-36
     7.7.4.     Roadway Transport of Produced Water	7-39


               This document is a draft for review purposes only and does not constitute Agency policy.
June 2015                                        vi                     DRAFT—DO NOT CITE OR QUOTE

-------
Hydraulic Fracturing Drinking Water Assessment
     7.7.5.     Studies of Environmental Transport of Released Produced Water	7-40
     7.7.6.     Coalbed Methane	7-41
     7.7.7.     Transport Properties	7-42
  7.8.    Synthesis	7-43
     7.8.1.     Summary of Findings	7-44
     7.8.2.     Factors Affecting the Frequency or Severity of Impacts	7-44
     7.8.3.     Uncertainties	7-45
     7.8.4.     Conclusions	7-46
  7.9.    References for Chapter 7	7-48
8.   Wastewater Treatment and Waste Disposal	8-1
  8.1.    Introduction	8-1
  8.2.    Volumes of Hydraulic Fracturing Wastewater	8-2
     8.2.1.     National Level Estimate	8-4
     8.2.2.     Regional/State and Formation Level Estimates	8-4
     8.2.3.     Estimation  Methodologies and Challenges	8-9
  8.3.    Wastewater Characteristics	8-10
     8.3.1.     Wastewater	8-10
     8.3.2.     Constituents in Residuals	8-12
  8.4.    Wastewater Management Practices	8-12
     8.4.1.     Underground Injection	8-20
     8.4.2.     Centralized Waste Treatment Facilities	8-23
     8.4.3.     Water Reuse for Hydraulic Fracturing	8-27
     8.4.4.     Evaporation	8-31
     8.4.5.     Publicly Owned Treatment Works	8-33
     8.4.6.     Other Management Practices and Issues	8-35
  8.5.    Summary and Analysis of Wastewater Treatment	8-38
     8.5.1.     Overview of Treatment Processes for Hydraulic Fracturing Wastewater	8-38
     8.5.2.     Treatment of Hydraulic Fracturing Waste Constituents of Concern	8-38
     8.5.3.     Design of Treatment Trains for CWTs	8-49
  8.6.    Potential Impacts on Drinking Water Resources	8-58
     8.6.1.     Bromide and Chloride	8-59
     8.6.2.     Radionuclides	8-62
     8.6.3.     Metals	8-65
     8.6.4.     Volatile  Organic Compounds	8-66
     8.6.5.     Semi-Volatile Organic Compounds	8-67
     8.6.6.     Oil and Grease	8-67
  8.7.    Synthesis	8-67
     8.7.1.     Summary of Findings	8-68
     8.7.2.     Factors Affecting the Frequency or Severity of Impacts	8-71
     8.7.3.     Uncertainties	8-72
     8.7.4.     Conclusions	8-73
  8.8.    References for Chapter 8	8-75
               This document is a draft for review purposes only and does not constitute Agency policy.
June 2015                                         vii                    DRAFT—DO NOT CITE OR QUOTE

-------
Hydraulic Fracturing Drinking Water Assessment
9.   Identification and Hazard Evaluation of Chemicals across the Hydraulic Fracturing Water Cycle
     9-1
  9.1.    Introduction	9-1
  9.2.    Identification of Chemicals Associated with the Hydraulic Fracturing Water Cycle	9-3
     9.2.1.     Chemicals Used in Hydraulic Fracturing Fluids	9-4
     9.2.2.     Chemicals Detected in Flowback and Produced Water	9-4
  9.3.    Toxicological and Physicochemical Properties of Hydraulic Fracturing Chemicals	9-4
     9.3.1.     Selection of Toxicity Values: Reference Values (RfVs) and Oral Slope Factors (OSFs)	9-5
     9.3.2.     Physicochemical Properties	9-7
     9.3.3.     Summary of Selected Toxicological and Physicochemical Property Data for Hydraulic
     Fracturing Chemicals	9-7
     9.3.4.     Additional Sources of Toxicity Information	9-8
  9.4.    Hazard Identification of Reported Hydraulic Fracturing Chemicals	9-10
     9.4.1.     Selection of Additional Chemicals for Hazard Identification	9-11
     9.4.2.     Hazard Identification Results	9-11
  9.5.    Hazard Identification and Hazard Evaluation of Selected Subsets of Hydraulic Fracturing Chemicals
         9-16
     9.5.1.     Selection of Chemicals for Hazard Evaluation	9-16
     9.5.2.     Multi Criteria Decision Analysis (MCDA) Framework for Hazard Evaluation: Integrating
     Toxicity, Occurrence, and Physicochemical Data	9-19
     9.5.3.     Hazard Evaluation Results	9-23
     9.5.4.     Summary of Chemicals Detected in Multiple Stages of the Hydraulic Fracturing Water Cycle
              9-33
  9.6.    Synthesis	9-35
     9.6.1.     Summary of Findings	9-35
     9.6.2.     Factors Affecting the Frequency or Severity of Impacts	9-37
     9.6.3.     Uncertainties	9-37
     9.6.4.     Conclusions	9-39
  9.7.    References for Chapter 9	9-41
  9.8.    Annex	9-43
     9.8.1.     Calculation of Physicochemical Property Scores (MCDA Hazard Evaluation)	9-43
     9.8.2.     Example of MCDA Score Calculation	9-44
10.  Synthesis	10-1
  10.1.  Major Findings	10-1
     10.1.1.   Water Acquisition (Chapter 4)	10-2
     10.1.2.   Chemical Mixing (Chapter 5)	10-5
     10.1.3.   Well Injection (Chapter 6)	10-8
     10.1.4.   Flowback and Produced Water (Chapter 7)	10-11
     10.1.5.   Wastewater Management and Waste Disposal (Chapter 8)	10-14
  10.2.  Key Data Limitations and Uncertainties	10-17
     10.2.1.   Limitations in monitoring data and chemical information	10-17
     10.2.2.   Other Contributing Limitations	10-19
  10.3.  Conclusions	10-19
  10.4.  Use of the Assessment	10-20

               This document is a draft for review purposes only and does not constitute Agency policy.
June 2015                                        viii                    DRAFT—DO NOT CITE OR QUOTE

-------
Hydraulic Fracturing Drinking Water Assessment
  10.5.  References for Chapter 10	10-21
           See Separate Supplemental Document for Appendices
Appendix A. Chemicals Identified in Hydraulic Fracturing Fluids and/or Flowback and Produced
           Water	A-l
Appendix B. Water Acquisition Tables	B-l
Appendix C. Chemical Mixing Supplemental Tables and Information	C-l
Appendix D. Designing, Constructing, and Testing Wells for Integrity	D-l
Appendix E. Flowback and Produced Water Supplemental Tables and Information	E-l
Appendix F. Wastewater Treatment and Waste Disposal Supplemental Information	F-l
Appendix G. Identification and Hazard Evaluation of Chemicals across the Hydraulic Fracturing
           Water Cycle Supplemental Tables and Information	G-l
Appendix H. Description of EPA Hydraulic Fracturing Study Publications Cited in This Assessment
           	H-l
Appendix I. Unit Conversions	1-1
Appendix J. Glossary	J-l
             This document is a draft for review purposes only and does not constitute Agency policy.
June 2015                                     ix                   DRAFT—DO NOT CITE OR QUOTE

-------
Hydraulic Fracturing Drinking Water Assessment
List of Tables
Table 1-1. Stages of the hydraulic fracturing water cycle have various potential effects on drinking water
          resources	1-4
Table 1-2. Criteria developed for the five factors used to evaluate literature and data cited in this assessment.
          	1-8
Table 1-3. Research questions addressed by this assessment	1-9
Table 4-1. Percentage of injected water volume that comes from reused hydraulic fracturing wastewater in
          various states, basins, and plays	4-3
Table 4-2. Annual average hydraulic fracturing water use and consumption in 2011 and 2012 compared to
          total annual water use and consumption in 2010, by county	4-11
Table 4-3. Estimated proportions of hydraulic fracturing source water from surface and ground water.4-19
Table 4-4. Brackish water use as a percentage of total hydraulic fracturing water use in Texas' main hydraulic
          fracturing areas, 2011	4-20
Table 5-1. Examples of common additives, their function, and the most frequently used chemicals reported to
          FracFocus for these additives	5-9
Table 5-2. Chemicals reported to FracFocus 1.0 from January 1, 2011 to February 28, 2013 in 10% or more
          disclosures, with the percent of disclosures for which each chemical is reported and the top four
          reported additives for the chemical	5-16
Table 5-3. The percentage of disclosures of the 20 most commonly reported chemical by state, where a
          chemical is reported in at least three states	5-21
Table 5-4. Example list of chemicals and volumes used in hydraulic fracturing	5-24
Table 5-5. Examples of typical hydraulic fracturing equipment and their functions	5-29
Table 5-6. Estimations of spill rates	5-49
Table 5-7. Ranking of the 20 most mobile organic chemicals, as determined by the largest log Kov/, with
          CASRN, percent of wells where the chemical is reported from January 1,2011 to February 28,
          2013 (U.S. EPA, 2015b), and physicochemical properties (log Kov/, solubility, and Henry's Law
          constant) as estimated by EPI Suite™	5-56
Table 5-8. Ranking of the 20 least mobile organic chemicals, as determined by the largest log Kov/, with  CASRN,
          percent of wells where the chemical is reported from January 1,2011 to February 28, 2013 (U.S.
          EPA, 2015b), and physicochemical properties (log Kov/, solubility, and Henry's Law constant) as
          estimated by EPI Suite™	5-59
Table 5-9. The 20 chemicals reported most frequently nationwide for hydraulic fracturing based on reported
          FracFocus 1.0 disclosures (U.S. EPA, 2015b), with EPI Suite™ physicochemical parameters where
          available, and estimated mean and median volumes of those chemicals, where density was
          available	5-62
Table 6-1. Results of studies of PA DEP violations data that examined well failure rates	6-21
Table 6-2. Comparing the approximate depth and thickness of selected U.S. shale gas plays and coalbed
          methane basins	6-33
Table 6-3. Modeling parameters and scenarios investigated by Reagan et al. (2015)	6-41
Table 7-1. Data from one company's operations indicating approximate total water use and approximate
          produced water volumes within 10 days after completion of wells (Mantell, 2013b)	7-4
Table 7-2. Additional short-, medium-, and long-term produced water estimates	7-5
Table 7-3. Flowback and long-term produced water characteristics for wells in unconventional formations
          (U.S. EPA, 2015e)	7-6
Table 7-4. Compiled minimum and maximum concentrations for various geochemical constituents in
          unconventional shale gas, tight gas, andCBM produced water (Alley et al., 2011)	7-23

               This document is a draft for review purposes only and does not constitute Agency policy.
June 2015                                        x                     DRAFT—DO NOT CITE OR QUOTE

-------
Hydraulic Fracturing Drinking Water Assessment
Table 7-5. Concentration ranges (mg/L) of several classes of naturally occurring organic chemicals in
           conventional produced water worldwide (reported in Neff, 2002)	7-28
Table 8-1. Estimated volumes (millions of gallons) of wastewater based on state data for selected years and
           numbers of wells producing fluid	8-7
Table 8-2. Hydraulic fracturing wastewater management practices in recent years	8-14
Table 8-3. Distribution of active Class IID wells across the United States	8-21
Table 8-4. Number, by state, of CWT facilities that have accepted or plan to accept wastewater from hydraulic
           fracturing activities	8-25
Table 8-5. Estimated percentages of reuse of hydraulic fracturing wastewater	8-29
Table 8-6. Studies of removal efficiencies and influent/effluent data for various processes and facilities.8-40
Table 8-7. Examples of centralized waste treatment facilities	8-51
Table 9-1. Sources of selected toxicityRfVs and OSFs	9-5
Table 9-2. List of the most frequently used chemicals in hydraulic fracturing fluids, with their respective
           federal chronic RfVs where available	9-11
Table 9-3. List of the 20 most mobile chemicals used in hydraulic fracturing fluid, with their respective federal
           chronic RfVs where available	9-13
Table 9-4. List of the 20 least mobile chemicals used in hydraulic fracturing fluid, with their respective federal
           chronic RfVs where available	9-14
Table 9-5. Thresholds  used for developing the toxicity score, occurrence score, and physicochemical
           properties score in this MCDA framework	9-22
Table 9-6. Toxicological properties of the 37 chemicals used in hydraulic fracturing fluid that were identified
           for hazard evaluation and MCDA analysis	9-24
Table 9-7. MCDA results for 37 chemicals used in hydraulic fracturing fluid	9-27
Table 9-8. Toxicological properties of the 23 chemicals detected in flowback and produced water that were
           identified for hazard evaluation and MCDA analysis	9-30
Table 9-9. MCDA results for 23 chemicals in hydraulic fracturing flowback and produced water	9-31
Table 9-10. List of the 23 chemicals with federal chronic RfVs identified to be used in hydraulic fracturing
           fluids  and detected in the flowback/produced water stage of the hydraulic fracturing water cycle.
           	9-34
               This document is a draft for review purposes only and does not constitute Agency policy.
June 2015                                         xi                      DRAFT—DO NOT CITE OR QUOTE

-------
Hydraulic Fracturing Drinking Water Assessment
List of Figures
Figure ES-1. Schematic cross-section of general types of oil and gas resources and the orientations of
           production wells used in hydraulic fracturing	ES-2
Figure ES-2. The stages of the hydraulic fracturing water cycle	ES-4
Figure ES-3. Water budgets representative of practices in the Marcellus Shale in the Susquehanna River Basin
           in Pennsylvania (a) and the Barnett Shale in Texas (b)	ES-8
Figure 1-1. Conceptualized view of the stages of the hydraulic fracturing water cycle	1-3
Figure 2-1. Schematic cross-section of general types of oil and gas resources and the orientations of
           production wells used in hydraulic fracturing	2-3
Figure 2-2. Shale gas and oil plays in the lower 48 United States	2-6
Figure 2-3. Tight gas plays in the lower 48 United States	2-7
Figure 2-4. Coalbed methane fields in the lower 48 United States	2-7
Figure 2-5. Generalized timeline and summary of activities that take place during the operational phases of an
           oil or gas well site operation in which hydraulic fracturing is used	2-8
Figure 2-6. Pulling drill pipe onto the drilling platform	2-11
Figure 2-7. Sections of surface casing lined up and being prepared for installation at a well site in Colorado.
           	2-12
Figure 2-8. Hydraulic fracturing operation in Troy, PA	2-13
Figure 2-9. Two wellheads side-by-side being prepared for hydraulic fracturing at a well site in Pennsylvania.
           	2-14
Figure 2-10. Water tanks (blue, foreground) lined up for hydraulic fracturing at a well site in central
           Arkansas	2-16
Figure 2-11. Impoundment on the site of a hydraulic fracturing operation in central Arkansas	2-18
Figure 2-12. Aerial photograph of a well pad and  service road in Springville Township, Pennsylvania. 2-21
Figure 2-13. Aerial photograph of hydraulic fracturing activities near Williston, North Dakota	2-21
Figure 2-14. Landsat photo showing hydraulic fracturing well sites near Frierson, Louisiana	2-22
Figure 2-15. Landsat photo showing hydraulic fracturing well sites near Pinedale, Wyoming	2-23
Figure 2-16. Location of horizontal wells that began producing oil or natural gas in 2000,2005, and 2012,
           based on data from Drillinglnfo (2014a)	2-24
Figure 2-17. Trends in U.S. oil and gas production	2-29
Figure 2-18. Historic and projected natural gas production by source (trillion cubic feet)	2-29
Figure 2-19. Natural gas  prices  and oil and gas drilling activity, 2008-2012	2-30
Figure 2-20. (a) Production from U.S. shale gas plays, 2000-2014, in billion cubic feet per day; (b) Production
           from U.S. tight oil plays, 2000-2014	2-31
Figure 2-21. U.S. petroleum and other liquid fuels supply by source, past and projected future trends (million
           barrels per day)	2-32
Figure 3-1. Geographic variability in drinking water sources for public water systems	3-2
Figure 3-2. Proximity of hydraulically fractured wells relative to populated areas	3-5
Figure 3-3. Temporal trends (2000-2013) in the  number and percent of hydraulically fractured wells located
           within populated areas	3-6
Figure 3-4. Location and number of public water  system (PWS) sources located within 1 mile of a
           hydraulically fractured well	3-8
               This document is a draft for review purposes only and does not constitute Agency policy.
June 2015                                         xii                     DRAFT—DO NOT CITE OR QUOTE

-------
Hydraulic Fracturing Drinking Water Assessment
Figure 3-5. The location of public water system sources within 1 mile of hydraulically fractured wells.. 3-9
Figure 3-6. Co-occurrence of hydraulic fracturing activity and populations supplied by private water systems.
           	3-10
Figure 4-1. Annual average hydraulic fracturing water use in 2011 and 2012 by county (U.S. EPA, 2015b).4-13
Figure 4-2. (a) Annual average hydraulic fracturing water use in 2011 and 2012 compared to total annual
           water use in 2010, by county, expressed as a percentage; (b) Annual average hydraulic fracturing
           water consumption in 2011 and 2012 compared to total annual water consumption in 2010, by
           county, expressed as a percentage	4-14
Figure 4-3. Locations of wells in the EPA FracFocus project database, with respect to U.S. EIA shale plays and
           basins (EIA, 2015; US. EPA, 2015b)	4-17
Figure 4-4. Major U.S. EIA shale plays and basins for Texas (EIA, 2015)	4-18
Figure 4-5. Annual average hydraulic fracturing water use in 2011 and 2012 compared to (a) fresh water
           available and (b) total water (fresh, brackish, and wastewater) available, by county, expressed as
           a percentage	4-23
Figure 4-6. (a) Estimated annual surface water runoff from the USGS; (b) Reliance on ground water as
           indicated by the ratio of ground water pumping to stream flow and pumping	4-25
Figure 4-7. Major U.S. EIA shale plays and basins for Colorado and Wyoming (EIA, 2015)	4-28
Figure 4-8. Major U.S. EIA shale plays and basins for Pennsylvania, West Virginia, and Ohio (EIA, 2015).4-32
Figure 4-9. Major U.S. EIA shale plays and basins for North Dakota and Montana (EIA, 2015b)	4-36
Figure 4-10. Major U.S. EIA shale plays and basins for Oklahoma and Kansas (EIA, 2015)	4-40
Figure 4-11. Major U.S. EIA shale plays and basins for Arkansas and Louisiana (EIA, 2015b)	4-43
Figure 4-12. Major U.S. EIA shale plays and basins for Utah, New Mexico, and California (EIA, 2015). ...4-45
Figure 5-1. Factors governing potential impact to drinking water resources	5-1
Figure 5-2. Hydraulic fracturing site showing equipment used on-site during the chemical mixing process.5-3
Figure 5-3. Overview of a chemical mixing process of the hydraulic fracturing water cycle	5-4
Figure 5-4. Example fracturing fluid decision tree for gas and oil wells	5-8
Figure 5-5. Estimated median volumes for chemicals reported in at least 100 FracFocus disclosures by
           February 28, 2013 for use in hydraulic fracturing from January 1, 2011 to February 28, 2013.5-27
Figure 5-6. Typical hydraulic fracturing equipment layout	5-30
Figure 5-7. Metal and high-density polyethylene (HOPE) chemical additive units	5-32
Figure 5-8. A worker adjusts hoses at a hydraulic fracturing site near Mead, Colorado	5-35
Figure 5-9. Manifold (pointed to by the white arrow)	5-37
Figure 5-10. High-pressure pumps on  either side of the manifold	5-38
Figure 5-11. Multiple fracture heads	5-39
Figure 5-12. Distribution of the causes of 151 hydraulic fracturing-related spills of chemicals and fracturing
           fluid	5-44
Figure 5-13. Percent distribution of sources of 151 hydraulic fracturing-related spills of chemicals or
           fracturing fluid	5-45
Figure 5-14. Total volume of fluids spilled for 151 hydraulic fracturing-related spills of chemicals and
           fracturing fluid, by spill  source	5-46
Figure 5-15. Number of hydraulic fracturing-related spills of chemicals or fracturing fluid that reported
           whether an environmental receptor was reached	5-47
Figure 5-16. Fate and transport schematic for a spilled hydraulic fracturing fluid	5-51
Figure 5-17. Histograms of physicochemical properties of chemicals used in the hydraulic fracturing process.
           	5-54
               This document is a draft for review purposes only and does not constitute Agency policy.
June 2015                                        xiii                     DRAFT—DO NOT CITE OR QUOTE

-------
Hydraulic Fracturing Drinking Water Assessment
Figure 5-18. Histograms of physicochemical properties of confidential chemicals used in the hydraulic
           fracturing process	5-55
Figure 5-19. Fate and Transport Spill Example: Case 1	5-67
Figure 5-20. Fate and Transport Spill Example: Case 2	5-68
Figure 5-21. Fate and Transport Spill Example: Case 3	5-69
Figure 6-1. Overview of well construction	6-5
Figure 6-2. The various stresses to which the casing will be exposed	6-7
Figure 6-3. Potential pathways for fluid movement in a cemented wellbore	6-12
Figure 6-4. Hydraulic fracture planes (represented as ovals), with respect to the principal subsurface
           compressive stresses: 5V [the vertical stress), SH [the maximum horizontal stress), and Sh (the
           minimum horizontal stress)	6-29
Figure 6-5. Conceptualized depiction of potential pathways for fluid movement out of the production zone:
           (a) induced fracture overgrowth into over- or underlying formations; (b) induced fractures
           intersecting natural fractures; and (c) induced fractures intersecting a transmissive fault. 6-37
Figure 6-6. Induced fractures intersecting an offset well (in a production zone, as shown, or in overlying
           formations into which fracture growth may have occurred)	6-43
Figure 6-7. Well communication (a frac hit) via induced fractures intersecting another well or its fracture
           network	6-43
Figure 7-1. Fraction of injected hydraulic fracturing fluid recovered from six vertical (top) and eight
           horizontal (bottom) wells completed in the Marcellus Shale	7-8
Figure 7-2. Example of flowback and produced water from the Marcellus Shale, illustrating rapid decline in
           water production and cumulative return of approximately 30% of the volume of injected fluid.
           	7-9
Figure 7-3. Percent of injected fluid recovered for Marcellus Shale wells  in West Virginia (2010-2012).7-10
Figure 7-4. Barnett Shale monthly water-production percentiles (5th, 30th, 50th, 70th, and 90th) and number of
           wells with data (dashed line)	7-12
Figure 7-5. Barnett Shale production data for approximately 72 months	7-12
Figure 7-6. TDS concentrations measured through time for injected fluid (at 0 days), flowback, and produced
           water samples from four Marcellus Shale gas wells in three southwestern Pennsylvanian counties.
           	7-18
Figure 7-7. Total radium and TDS concentrations measured through time for injected (day 0), flowback, and
           produced water samples from mutually exclusive  Greene County, PA, Marcellus Shale gas wells.
           	7-19
Figure 7-8. (a) Chloride (Cl) and (b)  DOC concentrations measured through time for injected (day 0),
           flowback, and produced water samples obtained from three Marcellus Shale gas wells from a
           single well pad in Greene County, PA used for hydraulic fracturing	7-21
Figure 7-9. Histograms of physicochemical properties of 86 organic chemicals identified in produced water
           (physicochemical properties estimated by EPI  Suite™)	7-43
Figure 8-1. Produced and flowback water volumes and produced gas volumes from unconventional wells in
           Pennsylvania from July of 2009 through June of 2014	8-3
Figure 8-2. Wastewater quantities in the western United States (billions of gallons per year)	8-5
Figure 8-3. Schematic of wastewater management strategies	8-13
Figure 8-4. Percentages of Marcellus Shale wastewater managed via various practices for (top) the second
           half of 2009 and first half of 2010 (total estimated volume of 216 Mgal), and (bottom) 2013 (total
           estimated volume of 1.3  billion gallons)	8-18
Figure 8-5. Management of wastewater in Colorado in regions where hydraulic fracturing is being performed.
           	8-19
               This document is a draft for review purposes only and does not constitute Agency policy.
June 2015                                        xiv                     DRAFT—DO NOT CITE OR QUOTE

-------
Hydraulic Fracturing Drinking Water Assessment
Figure 8-6. Lined evaporation pit in the Battle Creek Field (Montana)	8-32
Figure 8-7. Oil and gas wastewater volumes discharged to POTWs from 2001-2011 in the Marcellus Shale.
           	8-34
Figure 8-8. Full discharge water process used in the Pinedale Anticline field	8-50
Figure 9-1. Overall representation of the selected RfVs and OSFs, occurrence data, and physicochemical data
           available for the 1,173 hydraulic fracturing chemicals identified by the EPA	9-7
Figure 9-2. Fraction of chemicals with at least one data point in each ACToR data class	9-10
Figure 9-3. The two subsets of chemicals selected for hazard evaluation included 37 chemicals used in
           hydraulic fracturing fluid, and 23 chemicals detected in flowback or produced water	9-18
Figure 9-4. Overview of the MCDA framework applied to the hazard evaluations	9-20
Figure 10-1. Water budgets representative of practices in the Marcellus Shale in the Susquehanna River Basin
           in Pennsylvania (a) and the Barnett Shale in Texas (b)	10-3
               This document is a draft for review purposes only and does not constitute Agency policy.
June 2015                                         xv                     DRAFT—DO NOT CITE OR QUOTE

-------
Hydraulic Fracturing Drinking Water Assessment
List of Acronyms/Abbreviations
Acronym      Definition
2BE          2-butoxyethanol
ACToR        Aggregated Computational
              Toxicology Resource
              database
AMEC         AMEC Environment &
              Infrastructure, Inc.
ANRC         Arkansas Natural Resources
              Commission
AO           administrative order
AOGC         Arkansas Oil and Gas
              Commission
API           American Petroleum
              Institute
ATSDR        Agency for Toxic Substance
              and Disease Registry
AWWA        American Water Works
              Association
BLM          Bureau of Land Management
BTEX         benzene, toluene,
              ethylbenzene, and xylenes
CARES        Casella Altela Regional
              Environmental Services
CASRN        chemical abstract services
              registration number
CBI           confidential business
              information
CBM          coalbed methane
CCST         California Council on Science
              and Technology
CDWR        Colorado Division of Water
              Resources
CFR          Code of Federal Regulations
CICAD        Concise  International
              Chemical Assessment
              Document
Acronym
CM
COGCC

CWCB

CWT
CWTF

DBNM
DBF
DecaBDE
DfE
DI
DMR
DNR

DO
DOC
DOE
DOGGR
DOJ
DOT

DRO
EERC


EIA

EPA
Definition
chemical mixing
Colorado Oil and Gas
Conservation Commission
Colorado Water
Conservation Board
centralized waste treatment
centralized water treatment
facility
dibromochloronitromethane
disinfection by-products
decabromodipheyl ether
Design for the Environment
Drilling Info, Inc.
Discharge Monitoring Report
Department of Natural
Resources
dissolved oxygen
dissolved organic carbon
U.S. Department of Energy
California Department of
Conservation's Division of
Oil, Gas & Geothermal
Resources
U.S. Department of Justice
U.S. Department of
Transportation
diesel range organics
Energy and Environmental
Research Center, University
of North Dakota
U.S. Energy Information
Administration
U.S. Environmental
Protection Agency
             This document is a draft for review purposes only and does not constitute Agency policy.
June 2015                                  xvi                  DRAFT—DO NOT CITE OR QUOTE

-------
Hydraulic Fracturing Drinking Water Assessment
Acronym       Definition
EPAOW        U.S. Environmental
               Protection Agency's Office of
               Water
EPI            estimation programs
               interface
ERCB          Energy Resource
               Conservation Board
ERG           Eastern Research Group
ESN           Environmental Services
               Network
FRS           fluids recovery services
GES           Groundwater &
               Environmental Services, Inc.
GNB           Government of New
               Brunswick
GRO           gasoline range organics
GTI            Gas Technology Institute
GWPC         Ground Water Protection
               Council
HBCD          hexabromocyclododecane
HOPE          high-density polyethylene
HF            hydraulic fracturing
HHBP         Human Health Benchmarks
               for Pesticides
HISA          Highly Influential Scientific
               Assessment
HUG           hydrological unit code
IAEA          International Atomic Energy
               Agency
IARC          International Agency for
               Research on Cancer
IOGCC         Interstate Oil and Gas
               Compact Commission
IPCC           Intergovernmental Panel on
               Climate Change
IPCS           International Programme on
               Chemical Safety
Acronym       Definition
IRIS           Integrated Risk Information
               System
IUPAC         International Union of Pure
               and Applied Chemistry
KWO          Kansas Water Office
LOAEL         lowest observed adverse
               effect level
MCDA         multicriteria decision
               analysis
MCL           maximum contaminant level
MC LG          maximum containment level
               goal
MCOR         Marcellus Center for
               Outreach and Research
MIT           mechanical integrity test
MRL           minimum risk level
MSC           Marcellus shale coalition
MT GWIC       Montana Ground Water
               Information Center
MTBE          methyl tert-butyl ether
MVR           mechanical vapor
               recompression
NAS           National Academy of
               Sciences
NDDMR        North Dakota Department of
               Mineral Resources
NDDOH        North Dakota Department of
               Health
NDSWC        North Dakota State Water
               Commission
NETL          National Energy Technology
               Laboratory
NGO           non-governmental
               organization
NM OCD        New Mexico Oil
               Conservation Division
             This document is a draft for review purposes only and does not constitute Agency policy.
June 2015                                   xvii                  DRAFT—DO NOT CITE OR QUOTE

-------
Hydraulic Fracturing Drinking Water Assessment
Acronym       Definition
NM OSE        New Mexico Office of the
               State Engineer
NOAEL         no observed adverse effect
               level
NORM         naturally occurring
               radioactive material
NPC           National Petroleum Council
NPDES         National Pollution Discharge
               Elimination System
NRC           National Resource Council
NTP           U.S.  National Toxicology
               Program
NYSDEC        New York State Department
               of Environmental
               Conservation
O&G           oil and gas
ODNR         Ohio Department of Natural
               Resources
ODNR,         Ohio Department of Natural
DMRM         Resources, Division of
               Mineral Resources
               Management
OEPA          Ohio Environmental
               Protection Agency
ORD           Office of Research and
               Development
OSF           oral slope factor
OSHA          Occupational Safety & Health
               Administration
OSWER        Office of Solid Water and
               Emergency Response
OWRB         Oklahoma Water Resources
               Board
PA DCNR       Pennsylvania Department of
               Conservation and Natural
               Resources
PA DEP        Pennsylvania Department of
               Environmental Protection
Acronym       Definition
PAH           polycyclic aromatic
               hydrocarbon
PFBC          Pennsylvania Fish and Boat
               Commission
PDL           positive determination letter
PMF           Positive Matrix Factorization
POD           point-of-departure
POTW         publicly owned treatment
               work
PPRTV         provisional peer-reviewed
               toxicity value
QA            quality assurance
QAPP          quality assurance project
               plan
QC            quality control
QSAR          Quantitative Structure
               Activity Relationship
RfD            reference dose
RfV            reference value
RO            reverse osmosis
SAB           Science Advisory Board
SAIC           Science Applications
               International Corporation
SDWA         Safe Drinking Water Act
SDWIS         safe drinking water
               information system
SEECO         Southern Electrical
               Equipment Company
SGEIS          supplemented generic
               environmental impact
               statement
SHS MSC       statewide health standards
               for medium-specific
               concentrations
SMCL          secondary maximum
               contaminant level
             This document is a draft for review purposes only and does not constitute Agency policy.
June 2015                                   xviii                 DRAFT—DO NOT CITE OR QUOTE

-------
Hydraulic Fracturing Drinking Water Assessment
Acronym       Definition
SPE            Society of Petroleum
               Engineers
SRB            Susquehanna River basin
SRBC          Susquehanna River Basin
               Commission
STO            Statoil
STRONGER     State review of oil and
               natural gas environmental
               regulations
SVOC          semi-volatile organic
               compounds
SWE           Southwestern Energy
TARM         TerrAqua Resource
               Management
TEA           tert-butyl alcohol
TDS            total dissolved solids
TENORM       technologically enhanced
               naturally occurring
               radioactive material
THM           trihalomethane
TIPRO         Texas Independent
               Producers and Royalty
               Owners Association
TMDL         total maximum daily load
TOC            total organic carbon
TPH           total petroleum
               hydrocarbons
TSS            total suspended solids
TTHM         total trihalomethane
TWDB         Texas Water Development
               Board
TXRRC         Texas Railroad Commission
UCRB          Upper Colorado River basin
UIC            underground injection
               control
UOG           unconventional oil and gas
Acronym       Definition
USGAO         U.S. Government
               Accountability Office
USGS          U.S. Geological Survey
UWS           Universal Well Services
VOC           volatile organic compounds
WAWSA       Western Area Water Supply
               Authority
WFR           Well File Review
WHO          World Health Organization
WRF           Water Research Foundation
WVDEP        West Virginia Department of
               Environmental Protection
WWTP         wastewater treatment plant
WYOGCC       Wyoming Oil and Gas
               Conservation Commission
             This document is a draft for review purposes only and does not constitute Agency policy.
June 2015                                   xix                  DRAFT—DO NOT CITE OR QUOTE

-------
      Hydraulic Fracturing Drinking Water Assessment
      Preface

 1    The U.S. Environmental Protection Agency (EPA) is conducting a study of the potential impacts of
 2    hydraulic fracturing for oil and gas on drinking water resources. This study was initiated in Fiscal
 3    Year 2010 when Congress urged the EPA to examine the relationship between hydraulic fracturing
 4    and drinking water resources in the United States. In response, EPA developed a research plan
 5    (Plan to Study the Potential Impacts of Hydraulic Fracturing on Drinking Water Resources] that was
 6    reviewed by the Agency's Science Advisory Board (SAB) and issued in 2011. A progress  report on
 7    the study (Study of the Potential Impacts of Hydraulic Fracturing on Drinking Water Resources:
 8    Progress Report), detailing the EPA's research approaches and next steps, was released in late 2012
 9    and was followed by a consultation with individual experts convened under the auspices of the SAB.

10    The EPA's study includes the development of several research projects, extensive review of the
11    literature and technical input from state, industry, and non-governmental organizations as well as
12    the public and other stakeholders. A series of technical roundtables and in-depth technical
13    workshops were held to help  address specific research questions and to inform the work of the
14    study. The study is designed to address research questions posed for each stage of the hydraulic
15    fracturing water cycle:

16        •   Water Acquisition: What are the possible impacts of large volume water withdrawals
17            from ground and surface waters on drinking water resources?
18        •   Chemical Mixing: What are the possible impacts of surface spills of hydraulic fracturing
19            fluid on or near well pads on drinking water resources?
20        •   Well Injection: What are the possible  impacts of the injection and fracturing process on
21            drinking  water resources?
22        •   Flowback and Produced Water: What are the possible impacts of surface spills  of flowback
23            and produced water on or near well pads on drinking water resources?
24        •   Wastewater  Treatment and Waste Disposal: What are the possible impacts of
25            inadequate treatment of hydraulic fracturing wastewaters on drinking water resources?
2 6    This report, Assessment of the Potential Impacts of Hydraulic Fracturing for Oil and Gas on Drinking
27    Water Resources, includes both the literature review and results from the research projects
28    conducted as part of the EPA's study. It will undergo independent, external peer review  in
29    accordance with Agency policy and all of the peer review comments received will be considered in
3 0    the development of the final report.

31    The EPA's study will contribute to the understanding of the potential impacts of hydraulic
3 2    fracturing activities for oil and gas on drinking water resources and the factors that may influence
33    those impacts. The study will  help facilitate and inform  dialogue among interested stakeholders,
34    including Congress, other Federal agencies, states, tribal government, the international community,
35    industry, non-governmental organizations, academia, and the general public.
                   This document is a draft for review purposes only and does not constitute Agency policy.
      June 2015                                     xx                   DRAFT—DO NOT CITE OR QUOTE

-------
Hydraulic Fracturing Drinking Water Assessment
Authors, Contributors, and Reviewers

Authors

William Bates, USEPA-Office of Water, Washington, DC

Glen Boyd, The Cadmus Group, Inc., Seattle, WA

Susan Burden, USEPA-Office of Research and Development, Washington, DC

Lyle Burgoon, USEPA- Office of Research and Development, Research Triangle Park, NC

Christopher M. Clark, USEPA-Office of Research and Development, Washington, DC

Maryam Cluff, Student Services Contractor to USEPA under contract EP-13-H-000438-Office of
Research and Development, Washington, DC

Rebecca Daiss, USEPA-Office of Research and Development, Washington, DC

Jill Dean, USEPA-Office of Research and Development, Washington, DC

Deniz Inci Demirkanli, The Cadmus Group, Inc., Arlington, VA

Megan M. Fleming, USEPA-Office of Research and Development, Washington, DC

Jeffrey Frithsen, USEPA-Office of Research and Development, Washington, DC

Kenneth Klewicki, The Cadmus Group, Inc., Arlington, VA

Christopher D. Knightes, USEPA-Office of Research and Development, Athens, GA

Sandie Koenig, The Cadmus Group, Inc., Helena, MT

Jonathan Koplos, The Cadmus Group, Inc., Waltham, MA

Stephen D. LeDuc, USEPA-Office of Research and Development, Washington, DC

Claudia Meza-Cuadra, Student Services Contractor to USEPA under contract EP-13-H-000054-Office
of Research and Development, Washington, DC

Brent Ranalli, The Cadmus Group, Inc., Waltham, MA

Caroline E. Ridley, USEPA-Office of Research and Development, Washington, DC

Shari Ring, The Cadmus Group, Inc., Arlington, VA

Alison Singer, Student Services Contractor to USEPA under contract EP-13-H-000474-Office of
Research and Development, Washington, DC

John Stanek, USEPA- Office of Research and Development, Research Triangle Park, NC

M. Jason Todd, USEPA-Office of Research and Development, Washington, DC

Mary Ellen Tuccillo, The Cadmus Group, Inc., Waltham, MA


             This document is a draft for review purposes only and does not constitute Agency policy.
June 2015                                  xxi                  DRAFT—DO NOT CITE OR QUOTE

-------
Hydraulic Fracturing Drinking Water Assessment
Jim Weaver, USEPA-Office of Research and Development, Ada, OK

Anna Weber, The Cadmus Group, Inc., Arlington, VA

Larke Williams, USEPA-Office of Research and Development, Washington, DC

Liabeth Yohannes, Student Services Contractor to USEPA under contract EP-14-H-000455-Office of
Research and Development, Washington, DC

Erin Yost, ORISE Fellow to USEPA under interagency agreement DW8992 2983 with US DOE - Office
of Research and Development, Research Triangle Park, NC

Contributors

Natalie Auer, The Cadmus Group, Inc., Arlington, VA

Kevin Blackwood, Student Services Contractor to USEPA under contract EP-13-C-000133-Office of
Research and Development, Ada, OK

Jeanne  Briskin, USEPA-Office of Research and Development, Washington, DC

Rob Dewoskin, USEPA- Office of Research and Development, Research Triangle Park, NC

Linda Hills, The Cadmus Group, Inc., Helena, MT

Christopher Impellitteri, USEPA-Office of Research and Development, Cincinnati, OH

Richard Judson, USEPA- Office of Research and Development, Research Triangle Park, NC

Matt Landis, USEPA- Office of Research and Development, Research Triangle Park, NC

Ralph Ludwig, USEPA-Office of Research and Development, Ada, OK

John Martin, The Cadmus Group, Inc., Waltham, MA

Ashley  McElmury, Student Services Contractor to USEPA under contract EP-12-C-000025-Office of
Research and Development, Ada, OK

Gary Norris, USEPA- Office of Research and Development, Research Triangle Park, NC

Kay Pinley, Senior Environmental Employment Program under agreement CQ-835363 with NCCBA,
USEPA-Office of Research and Development, Ada, OK

Susan Sharkey, USEPA-Office of Research and Development, Washington, DC

Sarah Solomon, Student Services Contractor to USEPA under contract EP-D-15-003-Office of
Research and Development, Washington, DC

Holly Wooten, The Cadmus Group, Inc., Arlington, VA

Jie Xu, Student Services Contractor to USEPA under contract EP-13-C-00120-Office of Research and
Development, Ada, OK
             This document is a draft for review purposes only and does not constitute Agency policy.
June 2015                                   xxii                  DRAFT—DO NOT CITE OR QUOTE

-------
Hydraulic Fracturing Drinking Water Assessment
Reviewers
USEPA internal technical reviewers
Lisa Biddle, Office of Water, Washington, DC
Frank Brock, Region 2, New York, NY
Kyle Carey, Office of Water, Washington, DC
Brian D'Amico, Office of Water, Washington, DC
Tim Elkins, Region 5, Chicago, IL
Malcolm Field, Office of Research and Development, Washington, DC
Greg Fritz, Office of Chemical Safety and Pollution Prevention, Washington, DC
Mohamed Hantush, Office of Research and Development, Cincinnati, OH
Jana Harvill, Region 6, Dallas, TX
Charles Hillenbrand, Region 2, New York, NY
Mark W. Howard, Office of Solid Waste and Emergency Response, Washington, DC
Junqi Huang, Office of Research and Development, Ada, OK
Thomas Johnson, Office of Research and Development, Washington, DC
Jeff Jollie, Office of Water, Washington, DC
James Kenney, Office of Enforcement and Compliance Assurance, Washington, DC
Kristin Keteles, Region 8, Denver, CO
Bruce Kobelski, Office of Water, Washington, DC
Stephen Kraemer, Office of Research and Development, Athens, GA
Paul Lewis, Office of Chemical Safety and Pollution Prevention, Washington, DC
Chris Lister, Region 6, Dallas, TX
Barbara Martinez, ORISE Fellow to USEPA- Office of Research and Development, Washington, DC
Damon McElroy, Region 6, Dallas, TX
Keara Moore, Office of Water, Washington, DC
Nathan Mottl, Office of Chemical Safety and Pollution Prevention, Washington, DC
Greg Oberley, Region 8, Denver, CO
Mike Overbay, Region 6, Dallas, TX
              This document is a draft for review purposes only and does not constitute Agency policy.
June 2015                                    xxiii                  DRAFT—DO NOT CITE OR QUOTE

-------
Hydraulic Fracturing Drinking Water Assessment
Steve Platt, Region 3, Philadelphia, PA
Dave Rectenwald, Region 3, Philadelphia, PA
Meredith Russell, Office of Water, Washington, DC
Greg Schweer, Office of Chemical Safety and Pollution Prevention, Washington, DC
Steve Souders, Office of Solid Waste and Emergency Response, Washington, DC
Kate Sullivan, Office of Research and Development, Athens, GA
Scott Wilson, Office of Water, Washington, DC
Nathan Wiser, Office of Research and Development, Denver, CO
Acknowledgements
Contract support was provided by The Cadmus Group, Inc. under contract EP-C-08-015.
              This document is a draft for review purposes only and does not constitute Agency policy.
June 2015                                    xxiv                  DRAFT—DO NOT CITE OR QUOTE

-------
Hydraulic Fracturing Drinking Water Assessment                                     Executive Summary
Executive Summary
            This document is a draft for review purposes only and does not constitute Agency policy.
June 2015                                                    DRAFT—DO NOT CITE OR QUOTE

-------
      Hydraulic Fracturing Drinking Water Assessment                                         Executive Summary
      Executive Summary
 1    Since the early 2000s, oil and natural gas production in the United States has been transformed
 2    through technological innovation. Hydraulic fracturing, combined with advanced directional
 3    drilling techniques, made it possible to economically extract oil and gas resources previously
 4    inaccessible. The resulting surge in production increased domestic energy supplies and brought
 5    economic benefits to many areas  of the United States.

 6    The growth in domestic oil and gas production also raised concerns about potential impacts to
 7    human health and the environment, including potential effects on the quality and quantity of
 8    drinking water resources. Some residents living close to oil and gas production wells have reported
 9    changes in the quality of drinking water and assert that hydraulic fracturing is responsible for these
10    changes. Other concerns include competition for water between hydraulic fracturing activities and
11    other water users, especially in areas of the country experiencing drought, and the disposal of
12    wastewater generated from hydraulic fracturing.

13    The U.S. Congress urged the U.S. Environmental Protection Agency (EPA) to study the relationship
14    between hydraulic fracturing and drinking water. This report synthesizes available scientific
15    literature and data to assess the potential for hydraulic fracturing for oil and gas to change the
16    quality or quantity of drinking water resources, and identifies factors affecting the frequency or
17    severity of any potential changes. This report can be used by federal, tribal, state, and local officials;
18    industry; and the public to better understand and address any vulnerabilities of drinking water
19    resources to hydraulic fracturing activities.

      What is Hydraulic Fracturing?

20    Hydraulic fracturing is a stimulation technique used to increase oil and gas production from
21    underground rock formations. Hydraulic fracturing involves the injection of fluids under pressures
22    great enough to fracture the oil- and gas-producing formations. The fluid generally consists of
23    water, chemicals, and proppant (commonly sand). The proppant holds open the newly created
24    fractures after the injection pressure is released. Oil and gas flow through the fractures and up the
25    production well to the surface.

26    Hydraulic fracturing has been used since the late 1940s and, for the first 50 years, was mostly used
27    in vertical wells in conventional formations.l Hydraulic fracturing is still used in these settings, but
28    the process has evolved; technological developments (including horizontal and directional drilling)
29    have led to the use of hydraulic fracturing in unconventional hydrocarbon formations that could not
30    otherwise be profitably produced (see Figure ES-1). These formations include:
      1 Conventional formations often allow oil and natural gas to flow to the wellbore without hydraulic fracturing and
      typically contain trapped oil and natural gas that migrated from other subsurface locations. Hydraulic fracturing can be
      used to enhance oil and gas production from these formations. In unconventional formations, hydraulic fracturing is
      needed to extract economical quantities of oil and gas.
                    This document is a draft for review purposes only and does not constitute Agency policy.
      June 2015                                     ES-1                   DRAFT—DO NOT CITE OR QUOTE

-------
      Hydraulic Fracturing Drinking Water Assessment
                                                                      Executive Summary
 1
 2
 3
 4
 5
 6
 7
 8
 9
10
11
12
Shales. Organic-rich, black shales are the source rocks in which oil and gas form on
geological timescales. Oil and gas are contained in the pore space of the shale. Some shales
contain predominantly gas or oil; many shale formations contain both.
Tight formations. "Tight" formations are relatively low permeability, non-shale,
sedimentary formations that can contain oil and gas. Like in shales, oil and gas are
contained in the pore space of the formation. Tight formations can include sandstones,
siltstone, and carbonates, among others.
Coalbeds. In coalbeds, methane (the primary component of natural gas) is generally
adsorbed to the coal rather than contained in the pore space or structurally trapped in the
formation. Pumping the injected and native water out of the coalbeds after fracturing
serves to depressurize the coal, thereby allowing the methane to desorb and flow into the
well and to the surface.
      Figure ES-1. Schematic cross-section of general types of oil and gas resources and the
                orientations of production wells used in hydraulic fracturing.
                Shown are conceptual illustrations of types of oil and gas wells. A vertical well is producing from a
                conventional oil and gas deposit (right). In this case, a gray confining layer serves to "trap" oil (green)
                or gas (red). Also shown are wells producing from unconventional formations: a vertical coalbed
                methane well (second from right); a horizontal well producing from a shale formation (center); and a
                well producing from a tight sand formation (left). Note: Figure not to scale. Modified from USGS
                (2002) and Newell (2011).
                    This document is a draft for review purposes only and does not constitute Agency policy.
      June 2015                                     ES-2                   DRAFT—DO NOT CITE OR QUOTE

-------
      Hydraulic Fracturing Drinking Water Assessment                                         Executive Summary


 1    The combined use of hydraulic fracturing with horizontal (or more generically, directional) drilling
 2    has led to an increase in oil and gas activities in areas of the country with historical oil and gas
 3    production, and an expansion of oil and gas activities to new regions of the country.

      Scope of the Assessment

 4    We defined the scope of this assessment by the following activities involving water that support
 5    hydraulic fracturing (i.e., the hydraulic fracturing water cycle; see Figure ES-2):1

 6         •   Water acquisition: the withdrawal of ground or surface water needed for hydraulic
 7            fracturing fluids;
 8         •   Chemical mixing: the mixing of water, chemicals, and proppant on the well pad to create
 9            the hydraulic fracturing fluid;
10         •   Well injection: the injection of hydraulic fracturing fluids into the well to fracture the
11            geologic formation;
12         •   Flowback and produced water: the return of injected fluid and water produced from the
13            formation (collectively referred to as produced water in this report) to the surface, and
14            subsequent transport for reuse, treatment, or disposal; and
15         •   Wastewater treatment and waste disposal: the reuse, treatment and release, or
16            disposal of wastewater generated at the well pad, including produced water.
17    This assessment reviews, analyzes, and synthesizes information relevant to the potential impacts of
18    hydraulic fracturing on drinking water resources at each stage of the hydraulic fracturing water
19    cycle. Impacts are defined as any change in the quality or quantity of drinking water resources.
20    Where possible, we identify the mechanisms responsible or potentially responsible for any impacts.
21    For example, a spill of hydraulic fracturing fluid is a mechanism by which drinking water resources
22    could be impacted.

2 3    Drinking water resources are defined within this report as any body of ground water or surface
24    water that now serves, or in the future could serve, as a source of drinking water for public or
25    private use. This is broader than most federal and state regulatory definitions of drinking water and
26    encompasses both fresh and non-fresh bodies of water. Trends indicate that both types of water
27    bodies are currently being used, and will continue to be used in the future, as sources of drinking
28    water.

29    This assessment focuses on the potential impacts from activities in the hydraulic fracturing water
30    cycle on drinking water resources. We do this so federal, tribal, state, and local officials; industry;
31    and the public can better understand and address any vulnerabilities of drinking water resources to
32    hydraulic fracturing activities. We do not address other concerns raised about hydraulic fracturing
33    specifically or about oil and gas exploration and production activities more generally. Activities that
      1 In this assessment, we refer to the "EPA" when referencing other EPA studies. If a conclusion or analysis was done
      specifically by the authors of this assessment, we refer to it and its findings in the first person.
                    This document is a draft for review purposes only and does not constitute Agency policy.
      June 2015                                     ES-3                   DRAFT—DO NOT CITE OR QUOTE

-------
      Hydraulic Fracturing Drinking Water Assessment
Executive Summary
 1    are not considered include: acquisition and transport of constituents of hydraulic fracturing fluids
 2    besides water (e.g., sand mining and chemical production) outside of the stated water cycle; site
 3    selection and well pad development; other infrastructure development (e.g., roads, pipelines,
 4    compressor stations); site reclamation; and well closure. A summary and evaluation of current or
 5    proposed regulations and policies is beyond the scope of this report. Additionally, this report does
 6    not discuss the potential impacts  of hydraulic fracturing on other water users (e.g., agriculture or
 7    industry), other aspects of the environment (e.g., seismicity, air quality, or ecosystems), worker
 8    health or safety, or communities.  Furthermore, this report is not a human health risk assessment. It
 9    does not identify populations that are exposed to chemicals, estimate the extent of exposure, or
10    estimate the incidence of human health impacts.
                                              Well Injection
                Chemical Mixing
                                                                    Flow/back and Produced Water
                                                                           Wastewater Treatment and Waste Disposa
          Water Acquisition
      Figure ES-2, The stages of the hydraulic fracturing water cycle.
                Shown here is a generalized landscape depicting the activities of the hydraulic fracturing water cycle
                and their relationship to each other, as well as their relationship to drinking water resources. Arrows
                depict the movement of water and chemicals. Specific activities in the "Wastewater Treatment and
                Waste Disposal" inset are (a)underground injection control (UIC) well disposal, (b) wastewater
                treatment and reuse, and (c) wastewater treatment and discharge at a centralized waste treatment
                (CWT) facility. Note: Figure not to scale.
                     This document is a draft for review purposes only and does not constitute Agency policy.
      June 2015                                      ES-4                   DRAFT—DO NOT CITE OR QUOTE

-------
      Hydraulic Fracturing Drinking Water Assessment                                          Executive Summary
      Approach

 1    This assessment relies on relevant scientific literature and data. Literature evaluated included
 2    articles published in science and engineering journals, federal and state government reports, non-
 3    governmental organization (NGO) reports, and industry publications. Data sources examined
 4    included federal- and state-collected data sets, databases maintained by federal and state
 5    government agencies, other publicly-avail able data and information, and data, including
 6    confidential and non-confidential business information,  submitted by industry to the EPA.: The
 7    relevant literature and data complement research conducted by the EPA under its Plan to Study the
 8    Potential Impacts of Hydraulic Fracturing on Drinking Water Resources (hereafter referred to as the
 9    "Study Plan") and published by scientific journals or as peer-reviewed EPA reports; those articles
10    and reports are cited throughout this assessment The research topic areas and projects described
11    in the Study Plan were designed to meet the data and information needs of this assessment and
12    were developed with substantial expert and public input

      Proximity of Current Activity and Drinking Water  Resources

13    Thousands of wells are drilled and fractured every year  in the United States, with activities
14    concentrated in specific locations. We estimate 25,000-30,000 new wells were drilled and
15    hydraulically fractured annually in the United States between 2011 and 2014. Additional, pre-
16    existing wells (wells more than one year old that may or may not have been hydraulically fractured
17    in the past) were also likely fractured. Hydraulic fracturing took place in at least 25 states between
18    1990 and 2013. The EPA's analysis of disclosures made to FracFocus 1.0 (hereafter "FracFocus")
19    contained wells from 20 of these states.2 Almost half of these wells were in Texas. Colorado was a
20    distant second, while Pennsylvania and North Dakota were third and fourth, respectively. Hydraulic
21    fracturing activities were further localized within the 20 states. Of the approximately 1,500
22    counties or county equivalents in these 20 states, slightly over 400 contained all of the wells
23    disclosed to FracFocus during this time period. In Colorado, over 85% of the hydraulically fractured
24    wells disclosed were located in two counties. The price of gas and oil may cause short term
25    volatility in the number of wells drilled and fractured per year, yet hydraulic fracturing is expected
26    to continue to expand and drive an increase in domestic oil and gas production in coming decades.

27    Hydraulically fractured  wells can be located near residences and drinking water resources.
28    Between 2000 and 2013, approximately 9.4 million people lived within one mile of a hydraulically
      1 Some information provided to the EPA in response to two separate information requests to service companies and well
      operators was claimed as confidential business information.
       FracFocus is a publicly accessible website fwww.fracfocus.org) managed by the Ground Water Protection Council and
      the Interstate Oil and Gas Compact Commission where oil and gas production well operators may disclose information
      voluntarily or pursuant to state requirements about the ingredients used in hydraulic fracturing fluids at individual
      wells. The EPA analyzed disclosures from FracFocus 1.0 for over 38,000 oil and gas production wells hydraulically
      fractured between January 1,2011 and February 28,2013. A disclosure refers to data submitted for a specific oil and gas
      production well for a specific fracture date. Most wells had only one disclosure, but a small number of wells (876 wells)
      had multiple disclosures. For the purposes of this Executive Summary, we equate disclosures with wells when discussing
      this study.
                    This document is a draft for review purposes only and does not constitute Agency policy.
      June 2015                                     ES-5                   DRAFT—DO NOT CITE OR QUOTE

-------
      Hydraulic Fracturing Drinking Water Assessment                                         Executive Summary


 1    fractured well. Approximately 6,800 sources of drinking water for public water systems were
 2    located within one mile of at least one hydraulically fractured well during the same period. These
 3    drinking water sources served more than 8.6 million people year-round in 2013.

 4    Although proximity of hydraulic fracturing activities to a drinking water resource is not in of itself
 5    sufficient for an impact to occur,, it does increase the potential for impacts. Residents and drinking
 6    water resources in areas experiencing hydraulic fracturing activities are most likely to be affected
 7    by any potential impacts, should they occur. However, hydraulic fracturing can also affect drinking
 8    water resources outside the immediate vicinity of a hydraulically fractured well; a truck carrying
 9    wastewater could spill or a release of inadequately treated wastewater could have downstream
10    effects.

      Major Findings

11    From our assessment, we conclude there are above and below ground mechanisms by which
12    hydraulic fracturing activities have the potential to impact drinking water resources. These
13    mechanisms include water  withdrawals in times of, or in areas with, low water availability; spills of
14    hydraulic fracturing fluids and produced water; fracturing directly into underground drinking
15    water resources; below ground migration of liquids and gases; and inadequate treatment and
16    discharge of wastewater.

17    We did not find evidence that these mechanisms have led to widespread, systemic impacts on
18    drinking water resources in the United States. Of the potential mechanisms identified in this report,
19    we found specific instances where one or more mechanisms led to impacts on drinking water
2 0    resources, including contamination of drinking water wells. The number of identified cases,
21    however, was small compared to the number of hydraulically fractured wells.

22    This finding could reflect a  rarity of effects on drinking water resources, but may also be due to
23    other limiting factors. These factors include: insufficient pre- and post-fracturing data on the quality
24    of drinking water resources; the paucity of long-term systematic studies; the presence of other
25    sources of contamination precluding a definitive link between hydraulic fracturing activities and an
26    impact; and the inaccessibility of some information on hydraulic fracturing activities and potential
27    impacts.

28    Below, we provide a synopsis of the assessment's key findings, organized by each stage of the
29    hydraulic fracturing water cycle. We provide answers to the research questions presented in the
30    Study Plan and Chapter 1. While come citations are provided here, individual chapters should be
31    consulted for additional detail and citations.

            Water Acquisition
32    Water is a major component of nearly all hydraulic fracturing operations. It typically makes up
33    almost 90% or more of the  fluid volume injected into a well, and each hydraulically fractured well
34    requires thousands to millions of gallons of water. Cumulatively, hydraulic fracturing activities in
35    the United States used on average 44 billion gal of water a year in 2011 and 2012, according to the
                   This document is a draft for review purposes only and does not constitute Agency policy.
      June 2015                                    ES-6                  DRAFT—DO NOT CITE OR QUOTE

-------
      Hydraulic Fracturing Drinking Water Assessment                                           Executive Summary


 1    EPA's analysis of FracFocus disclosures. Although this represents less than 1% of total annual water
 2    use and consumption at this scale, water withdrawals could potentially impact the quantity and
 3    quality of drinking water resources at more local scales.:

         Research Questions:  Water Acquisition

          •   What are the types of water used for hydraulic fracturing?
 4    Water for hydraulic fracturing typically comes from surface water, ground water, or reused
 5    hydraulic fracturing wastewater. Hydraulic fracturing operations in the eastern United States
 6    generally rely on surface  water, while operations in the more semi-arid to arid western states
 7    generally use mixed supplies of surface and ground water. In the Marcellus Shale in Pennsylvania,
 8    for example, most water used for hydraulic fracturing originates from surface water, whereas
 9    surface and ground water are used in approximately equal proportions in the Barnett Shale in
10    Texas  (see Figure ES-3a,b). In areas that lack available surface water (e.g., western Texas), ground
11    water  supplies most of the water needed for hydraulic fracturing.

12    Across the United States,  the vast majority of water used in hydraulic fracturing is fresh, although
13    operators also make use of lower-quality water, including reused hydraulic fracturing wastewater.2
14    Based on available data, the median reuse of wastewater as  a percentage of injected volumes is 5%
15    nationally, with the  percentage varying by location.3 Available data on reuse trends indicate
16    increased reuse of wastewater over time in both Pennsylvania and West Virginia. Reuse as a
17    percentage of injected volumes is lower in other areas, including regions with more water stress,
18    likely because of the availability of disposal wells. For example, reused wastewater is
19    approximately 18% of injected volumes in the Marcellus Shale in Pennsylvania's Susquehanna
20    River Basin, whereas it is approximately 5% in the Barnett Shale in Texas (see Figure ES-3a,b).
      1 Water use is water withdrawn from ground- or surface water for a specific purpose, part or all of which may be returned
      to the local hydrologic cycle. If no water is returned, water use equals water consumption. Water consumption is water
      that is removed from the local hydrologic cycle following its use (e.g., via evaporation, transpiration, incorporation into
      products or crops, consumption by humans or livestock) and is therefore unavailable to other water users (Maupin etal..
      2014). In the case of hydraulic fracturing, water can be consumed by the loss of injected water to subsurface zones or via
      underground disposal of wastewaters, among other means.
      2 In this assessment, hydraulic fracturing "wastewater" refers to both produced water and any other water generated as a
      hydraulic fracturing site. As used in this assessment, the term "wastewater" is not intended to constitute a term of art for
      legal or regulatory purposes.
      3 Reused wastewater as a percentage of injected water differs from the percentage of wastewater that is managed through
      reuse, as opposed to other wastewater management options. For example, in the Marcellus in Pennsylvania,
      approximately 18% of injected water is reused produced water, while approximately 70% of wastewater or more is
      managed through reuse (Figure ES-3a).
                     This document is a draft for review purposes only and does not constitute Agency policy.
      June 2015                                      ES-7                   DRAFT—DO NOT CITE OR QUOTE

-------
Hydraulic Fracturing Drinking Water Assessment
                                                            Executive Summary
                     Water Source
                                                   Well             Wastewater Management

                                                      420,000 -1.3 million gal produced
                                                                     Reuse   CWT  • UIC disposal well
                                                        Most of the injected fluid is retained
                                                        in the subsurface: Produced water
                                                        volumes over 10 years are
                                                        approximately 10-30% of the original
                                                        injected fluid volume
              I Surf ace Water   Reused Water   Ground Water
                     Water Source
                                                   Well
                                                                   Wastewater Management
                                                        3.9 - 4.5 million gal produced
                                                       Produced
                                                       water volumes
                                                       overs
                                                       years are
                                                       approximately
                                                       the same as
                                                       the original
                                                       injected fluid
                                                       volume
5%
             • Surface Water   Reused Water   Ground Water
                                                                        Reuse  • UIC disposal well
Figure ES-3. Water budgets representative of practices in the Marcellus Shale in the
            Susquehanna River Basin in Pennsylvania (a) and the Barnett Shale in Texas (b).

            Pie size and arrow thickness represent the relative volume of water as it flows through the hydraulic
            fracturing water cycle. Wastewater going to a centralized waste treatment (CWT) facility may be
            either discharged to surface water or reused. Wastewater going to an underground injection control
            (UIC) well is disposed of below ground. These examples represent typical water management practices
            as depicted for the most recent time period reviewed by this assessment. They do not represent any
            specific well. Note: Values for Marcellus Shale are specific to the Susquenhanna River Basin, except for
            the produced water volumes. The longest-term measurement available was from the West Virginia
            portion of the the Marcellus Shale.
                This document is a draft for review purposes only and does not constitute Agency policy.

June 2015                                          ES-8                      DRAFT—DO NOT CITE OR QUOTE

-------
      Hydraulic Fracturing Drinking Water Assessment                                         Executive Summary


         •   How much water is used per well?
 1    The national median volume of water used per hydraulically fractured well is approximately 1.5
 2    million gal (5.7 million L), according to the EPA's analysis of FracFocus disclosures. This estimate
 3    likely represents a wide variety of fractured well types, including vertical wells that generally use
 4    much less water per well than horizontal wells. Thus, published estimates for horizontal shale gas
 5    wells are typically higher (e.g., approximately 4 million gallons (Vengosh etal.. 2014]]. There is also
 6    wide variation within and among states and basins in the median water volumes used per well,
 7    from more than 5 million gal (19 million L] in Arkansas, Louisiana and West Virginia to less than 1
 8    million gal (3.8 million L] in California, New Mexico, and Utah, among others. This variation results
 9    from several factors, including well length, formation geology, and fracturing fluid formulation.

         •   How might cumulative water withdrawals for hydraulic fracturing affect drinking water
             quantity?
10    Cumulatively, hydraulic fracturing uses billions of gallons of water each year at the national and
11    state scales, and even in some counties. As noted above, hydraulic fracturing water use and
12    consumption are generally less than 1% of total annual water use and consumption at these scales.
13    However, there are a few counties in the United States where these percentages are higher. For
14    2011 and 2012, annual hydraulic fracturing water use was 10% or more compared to 2010 total
15    annual water use in 6.5% of counties with FracFocus disclosures analyzed by the EPA, 30% or more
16    in 2.2% of counties, and 50% or more in 1.0% of counties. Consumption estimates followed the
17    same general pattern. In these counties, hydraulic fracturing is a relatively large user and consumer
18    of water.

19    High fracturing water use or consumption alone does not necessarily result in impacts to drinking
20    water resources. Rather, impacts result from the combination of water use or consumption and
21    water availability at local scales.  In our survey of published literature, we did not find a case where
22    hydraulic fracturing water use or consumption alone caused a drinking water well or stream to run
23    dry. This could indicate an absence of effects or a lack of documentation in the literature we
24    reviewed. Additionally, water availability is rarely impacted by just one use or factor alone. In
25    Louisiana, for example, the state  requested hydraulic fracturing operations switch from ground to
26    surface water, due to concerns that ground water withdrawals  for fracturing could, in combination
27    with other uses, adversely affect drinking water supplies.

28    The potential for impacts to drinking water resources from hydraulic fracturing water withdrawals
29    is highest in areas with relatively high fracturing water use and low water availability. Southern
30    and western Texas are two locations where hydraulic fracturing water use, low water availability,
31    drought, and reliance on declining ground water has the potential to affect the quantity of drinking
32    water resources. Any impacts are likely to be realized locally within these areas. In a detailed case
33    study of southern Texas, Scanlonetal. (2014] observed generally adequate water supplies for
34    hydraulic fracturing, except in specific locations. They found excessive drawdown of local ground
35    water in a small proportion (approximately 6% of the area] of the Eagle Ford Shale. They suggested
3 6    water management, particularly  a shift towards brackish water use, could minimize potential future
37    impacts to fresh water resources.
                   This document is a draft for review purposes only and does not constitute Agency policy.
      June 2015                                    ES-9                  DRAFT—DO NOT CITE OR QUOTE

-------
      Hydraulic Fracturing Drinking Water Assessment                                         Executive Summary


 1    The potential for impacts to drinking water quantity due to hydraulic fracturing water use appears
 2    to be lower—but not eliminated—in other areas of the United States. Future problems could arise if
 3    hydraulic fracturing increases substantially in areas with low water availability, or in times of water
 4    shortages. In detailed case studies in western Colorado and northeastern Pennsylvania, the EPA did
 5    not find current impacts, but did conclude that streams could be vulnerable to water withdrawals
 6    from hydraulic fracturing. In northeast Pennsylvania, water management, such as minimum stream
 7    flow requirements, limits the potential for impacts, especially in small streams. In western North
 8    Dakota, ground water is limited, but the industry may have sufficient supplies of surface water from
 9    the Missouri River system. These location-specific examples emphasize the need to focus on
10    regional and local dynamics when considering potential impacts of hydraulic fracturing water
11    acquisition on drinking water resources.

         •   What are the possible impacts of water withdrawals for hydraulic fracturing on water quality?
12    Water withdrawals for hydraulic fracturing, similar to all water withdrawals, have the potential to
13    alter the quality of drinking water resources. Ground water withdrawals exceeding natural
14    recharge rates decrease water storage in aquifers, potentially mobilizing contaminants or allowing
15    the infiltration of lower quality water from the land surface or adjacent formations. Withdrawals
16    could also decrease ground water discharge to streams, potentially affecting surface water quality.
17    Areas with large amounts of sustained ground water pumping are most likely to experience
18    impacts, particularly drought-prone regions with limited ground water recharge.

19    Surface water withdrawals also have the potential to affect water quality. Withdrawals may lower
2 0    water levels and alter stream flow, potentially decreasing a stream's capacity to dilute
21    contaminants. Case studies by the EPA show that streams can be vulnerable to changes in water
2 2    quality due to water withdrawals, particularly smaller streams and during periods of low flow.
2 3    Management of the rate and timing of surface water withdrawals has been shown to help mitigate
24    potential impacts of hydraulic fracturing withdrawals on water quality.

           Chemical Mixing
25    Hydraulic fracturing fluids are developed to perform specific functions, including: create and
2 6    extend fractures, transport proppant, and place proppant in the fractures. The fluid generally
27    consists of three parts: (1) the base fluid, which is the largest constituent by volume and is typically
28    water;  (2) the additives, which can be a single chemical or a mixture of chemicals; and (3) the
29    proppant Additives are chosen to serve a specific purpose (e.g., adjust pH, increase viscosity, limit
30    bacterial growth). Chemicals generally comprise a small percentage (typically 2% or less) of the
31    overall injected fluid volume. Because over one million gallons of fluids are typically injected per
32    well, thousands of gallons of chemicals can be potentially stored on-site and used during hydraulic
3 3    fracturing activities.

34    On-site storage, mixing, and pumping of chemicals and hydraulic fracturing fluids have the potential
35    to result in accidental releases, such as spills or leaks. Potential impacts to drinking water resources
36    from spills of hydraulic fracturing fluids and chemicals depend on the characteristics of the spills,
37    and the fate, transport, and the toxicity of chemicals spilled.


                   This document is a draft for review purposes only and does not constitute Agency policy.
      June 2015                                    ES-10                  DRAFT—DO NOT CITE OR QUOTE

-------
      Hydraulic Fracturing Drinking Water Assessment                                         Executive Summary


         Research Questions: Chemical Mixing

         •   What is currently known about the frequency, severity, and causes of spills of hydraulic
             fracturing fluids and additives?
 1    The frequency of on-site spills from hydraulic fracturing could be estimated for two  states, but not
 2    for operations nationally or for other areas. Frequency estimates from data and literature ranged
 3    from one spill for every 100 wells in Colorado to between approximately 0.4 and 12.2 spills for
 4    every 100 wells in Pennsylvania.1 These estimates include spills of hydraulic fracturing fluids and
 5    chemicals, and produced water reported in state databases. Available data generally precluded
 6    estimates of hydraulic fracturing fluid and/or chemical spill rates separately from estimates of an
 7    overall spill frequency. It is unknown whether these spill estimates are representative of national
 8    occurrences. If the estimates are representative, the number of spills nationally could range from
 9    approximately 100 to 3,700 spills annually, assuming 25,000 to 30,000 new wells are fractured per
10    year.

11    The EPA characterized volumes and causes of hydraulic fracturing-related spills identified from
12    selected state and industry data sources. The spills occurred between January 2006  and April 2012
13    in 11 states and included 151 cases in which fracturing fluids or chemicals spilled on or near a well
14    pad. Due to the methods used for the EPA's characterization of spills, these cases were likely a
15    subset of all fracturing fluid and chemical spills  during the study's time period. The reported
16    volume of fracturing fluids or chemicals spilled  ranged from 5 gal to more than 19,000 gal (19 to
17    72,000 L), with a median volume of 420 gal (1,600 L) per spill. Spill causes included  equipment
18    failure, human error, failure of container integrity, and other causes (e.g., weather and vandalism).
19    The most common cause was equipment failure, specifically blowout preventer failure, corrosion,
20    and failed valves. More than 30% of the 151 fracturing fluid or chemical spills were from fluid
21    storage units (e.g., tanks, totes, and trailers).

         •   What are the identities and volumes of chemicals used in hydraulic fracturing fluids, and how
             might this composition vary at a given site and across the country?
22    In this assessment, we identified a list of 1,076 chemicals used in hydraulic fracturing fluids. This is
23    a cumulative list over multiple wells and years. These chemicals include acids,  alcohols, aromatic
24    hydrocarbons, bases, hydrocarbon mixtures, polysaccharides, and surfactants. According to the
25    EPA's analysis of disclosures to FracFocus, the number of unique chemicals per well ranged from 4
26    to 28, with a median of 14 unique chemicals per well.

27    Our analysis indicates  that chemical use varies and that no single chemical is used at all well sites
28    across the country, although several chemicals are widely used. Methanol, hydrotreated light
29    petroleum distillates, and hydrochloric acid were reported as used in 65% or more of wells,
30    according to FracFocus disclosures analyzed by the EPA. Only 32 chemicals, excluding water,
31    quartz, and sodium chloride, were used in more than 10% of wells according to the EPA's analysis
      1 Spill frequency estimates are for a given number of wells over a given period of time. These are not annual estimates nor
      are they for the lifetime of a well.
                    This document is a draft for review purposes only and does not constitute Agency policy.
      June 2015                                    ES-11                  DRAFT—DO NOT CITE OR QUOTE

-------
      Hydraulic Fracturing Drinking Water Assessment                                         Executive Summary


 1    of FracFocus disclosures. The composition of hydraulic fracturing fluids varies by state, by well, and
 2    within the same service company and geologic formation. This variability likely results from several
 3    factors, including the geology of the formation, the availability and cost of different chemicals, and
 4    operator preference.

 5    Estimates from the EPA's database developed from FracFocus suggest median volumes of
 6    individual chemicals injected per well ranged from a few gallons to thousands of gallons, with an
 7    overall median of 650 gal (2,500 L) per chemical per well. Based on this overall median and
 8    assuming 14 unique chemicals are used per well, an estimated 9,100 gal (34,000 L) of chemicals
 9    may be injected per well. Given that the number of chemicals per well ranges from 4 to 28, the
10    estimated volume of chemicals injected per well may range from approximately 2,600 to 18,000 gal
11    (9,800 to 69,000 L).

          •   What are the chemical, physical, and toxicological properties of hydraulic fracturing chemical
             additives?
12    Measured or estimated physicochemical properties were obtained for 453 chemicals of the total
13    1,076 chemicals reported in hydraulic fracturing fluids. We could not estimate physicochemical
14    properties for the inorganic chemicals or mixtures. The 453 chemicals have a wide range of
15    physicochemical properties.

16    Properties affecting the likelihood of a spilled chemical reaching and impacting a drinking water
17    resource include mobility, solubility, and volatility. Of the 453 chemicals for which physicochemical
18    properties were available, 18 of the top 20 most mobile ones were reported in the EPA's FracFocus
19    database for 2% or less of wells. Choline chloride and tetrakis (hydroxymethyl) phosphonium were
20    exceptions and were reported in 14% and 11% of wells, respectively. These two chemicals appear
21    to be relatively more common, and, if spilled, would move quickly through the environment with
22    the flow of water. The majority of the 453 chemicals associate strongly with soils and organic
23    materials, suggesting the potential for these chemicals to persist in the environment as long-term
24    contaminants. Many of the 453 chemicals fully dissolve in water, but their aqueous solubility varies
2 5    greatly. Few of the chemicals volatilize, and thus a large proportion of most hydraulic fracturing
2 6    chemicals tend to remain in water.

27    Oral reference values and oral slope factors meeting the criteria used in this assessment were not
28    available for the majority of chemicals used in hydraulic fracturing fluids, representing a significant
29    data gap for hazard identification.l'2 Reference values and oral slope factors are important for
30    understanding the potential human health effects resulting from exposure to a chemical. Chronic
31    oral reference values and/or oral slope factors from selected federal, state,  and international
32    sources were available for 90 (8%) of the 1,076 chemicals used in hydraulic fracturing fluids. From
      1A reference value is an estimate of an exposure to the human population (including susceptible subgroups) for a given
      duration that is likely to be without an appreciable risk of adverse health effects over a lifetime. Reference value is a
      generic term not specific to a given route of exposure.
       An oral slope factor is an upper-bound, approximating 95% confidence limit, on the increased cancer risk from a lifetime
      oral exposure to an agent.
                    This document is a draft for review purposes only and does not constitute Agency policy.
      June 2015                                    ES-12                  DRAFT—DO NOT CITE OR QUOTE

-------
      Hydraulic Fracturing Drinking Water Assessment                                         Executive Summary


 1    U.S. federal sources alone, chronic oral reference values were available for 73 chemicals (7%) of the
 2    1,076 chemicals, and oral slope factors were available for 15 chemicals (1%). Of the 32 chemicals
 3    reported as used in at least 10% of wells in the EPA's FracFocus database (excluding water, quartz,
 4    and sodium chloride), seven (21%) have a federal chronic oral reference value. Oral reference
 5    values and oral slope factors are a key component of the risk assessment process, although
 6    comprehensive risk assessments that characterize the health risk associated with exposure to these
 7    chemicals are not available.

 8    Of the chemicals that had values available, the health endpoints associated with those values
 9    include the potential for carcinogenesis, immune system effects, changes in body weight, changes in
10    blood chemistry, cardiotoxicity, neurotoxicity, liver and kidney toxicity, and reproductive and
11    developmental toxicity. However, it is important to note that evaluating any potential risk to human
12    populations would require knowledge of the specific chemicals that are present at a particular site,
13    whether or not humans are exposed to those chemicals and, if so, at what levels and for what
14    duration, and the toxicity of the chemicals. Since most chemicals are used infrequently on a
15    nationwide basis, potential exposure is likely to be a local or regional issue, rather than a national
16    issue. Accordingly, consideration of hazards and risks associated with these chemical additives
17    would be most useful on a site-specific basis and is beyond the scope of this assessment.

         •   If spills occur, how might hydraulic fracturing chemical additives contaminate drinking water
             resources?
18    There are several mechanisms by which a spill can potentially contaminate drinking water
19    resources. These include overland flow to nearby surface water, soil contamination and eventual
2 0    transport to surface water, and infiltration and contamination of underlying ground water. Of the
21    151 spills characterized by the EPA, fluids reached surface water in 13 (9% of 151) cases and soil in
22    97 (64%) cases. None of the spills of hydraulic fracturing fluid were reported to have reached
23    ground water. This could be due to an absence of impact; however, it can take several years for
24    spilled fluids to infiltrate soil and leach into ground water. Thus, it may not be immediately
25    apparent whether a spill has reached ground water or not

26    Based on the relative importance of each of these mechanisms, impacts have the potential to occur
27    quickly, be delayed short or long periods, or have a continual effect over time. In Kentucky, for
28    example, a spill impacted a surface water body relatively quickly when hydraulic fracturing fluid
29    entered a creek, significantly reducing the water's pH and increasing its conductivity (Papoulias
30    and Velasco. 2013).

           Well Injection
31    Hydraulic fracturing fluids are injected into oil or gas wells under high pressures. The fluids flow
32    through the well (commonly thousands of feet below the surface) into the production zone (i.e., the
33    geologic formation being fractured) where the fluid injection pressures are sufficient to create
34    fractures in the rock.

35    There are two major subsurface mechanisms by which the injection of fluid and the creation and
36    propagation of fractures can lead to contamination of drinking water resources: (1) the unintended

                   This document is a draft for review purposes only and does not constitute Agency policy.
      June 2015                                   ES-13                   DRAFT—DO NOT CITE OR QUOTE

-------
      Hydraulic Fracturing Drinking Water Assessment                                        Executive Summary


 1    movement of liquids or gases out of the production well or along the outside of the production well
 2    into a drinking water resource via deficiencies in the well's casing or cement, and (2) the
 3    unintended movement of liquids or gases from the production zone through subsurface geologic
 4    formations into a drinking water resource. Combinations of these two mechanisms are also
 5    possible.

        Research Questions: Well Injection

         •   How effective are current well construction practices at containing fluids—both liquids and
             gases—before, during, and after fracturing?
 6    Production wells are constructed to access and convey hydrocarbons from the formations in which
 7    they are found to the surface, and to isolate fluid-bearing zones (containing oil, gas, or water) from
 8    each other. Typically, multiple casings are emplaced and cemented along the wellbore to protect
 9    and isolate the oil and/or natural gas from the formations it must travel through to reach the
10    surface.

11    Below ground drinking water resources are often separated from the production well using casing
12    and cement. Cemented surface casing, in particular, is an important well construction feature for
13    isolating drinking water resources from liquids and gases that may move through the subsurface. A
14    limited risk modeling study of selected injection wells in the Williston Basin in North Dakota
15    suggests that the risk of aquifer contamination from leaks inside the well to the drinking water
16    resource decreases by a factor of approximately one thousand when surface casing extends below
17    the bottom of the drinking water resource [Michie and Koch.  1991]. Most wells used in hydraulic
18    fracturing operations have casing and a layer of cement to protect drinking water resources, but
19    there are exceptions: a survey conducted by the EPA of oil and gas production wells hydraulically
20    fractured by nine oil and gas service companies in 2009 and 2010 estimated that at least 3% of the
21    wells (600 out of 23,000 wells) did not have cement across a portion of the casing installed through
22    the protected ground water resource identified by well operators. The absence of cement does not
23    in and of itself lead to an impact However, it does reduce the overall number of casing and cement
24    barriers fluids must travel through to reach ground water resources.

25    Impacts to drinking water resources from subsurface liquid and gas movement may occur if casing
26    or cement are inadequately designed or constructed, or fail. There are several examples of these
27    occurrences in hydraulically fractured wells that have or may have resulted in impacts to drinking
28    water resources. In one example, an inner string of casing burst during hydraulic fracturing, which
29    resulted in a release of fluids on the land surface and possibly into the aquifer near Killdeer, North
30    Dakota. The EPA found that, based on the data analysis performed for the study, the only potential
31    source  consistent with conditions  observed in two impacted monitoring wells was the blowout that
32    occurred during hydraulic fracturing (U.S. EPA, 2015)]. In other examples, inadequately cemented
33    casing has contributed to impacts to drinking water resources. In Bainbridge, Ohio, inadequately
34    cemented casing in a hydraulically fractured well contributed to the buildup of natural gas and high
35    pressures along the outside of a production well. This ultimately resulted in movement of natural
36    gas into local drinking water aquifers (Bair etal.. 2010: ODNR. 2008). In the Mamm Creek gas field


                   This document is a draft for review purposes only and does not constitute Agency policy.
      June 2015                                   ES-14                  DRAFT—DO NOT CITE OR QUOTE

-------
      Hydraulic Fracturing Drinking Water Assessment                                        Executive Summary


 1    in Colorado, inadequate cement placement in a production well allowed methane and benzene to
 2    migrate along the production well and through natural faults and fractures to drinking water
 3    resources fScience Based Solutions LLC. 2014: Crescent. 2011: COGCC. 20041 These cases illustrate
 4    how construction issues, sustained casing pressure, and the presence of natural faults and fractures
 5    can work together to create pathways for fluids to migrate toward drinking water resources.

 6    Fracturing older wells may also increase the potential for impacts to drinking water resources via
 7    movement of gases and liquids from the inside of the production well or along the outside of the
 8    production well to ground water resources. The EPA estimated that 6% of 23,000 oil and gas
 9    production wells were drilled more than 10 years before being hydraulically fractured in 2009 or
10    2010. Although new wells can be designed to withstand the stresses associated with hydraulic
11    fracturing operations, older wells may not have been built or tested to the same specifications and
12    their reuse for this purpose could be of concern. Moreover, aging and use of the well can contribute
13    to casing degradation, which can be accelerated by exposure to corrosive chemicals, such as
14    hydrogen sulfide, carbonic acid, and brines.

         •   Can subsurface migration of fluids—both liquids and gases—to drinking water resources
             occur, and what local geologic or artificial features might allow this?
15    Physical separation between the production zone and drinking water resources can help protect
16    drinking water. Many hydraulic fracturing operations target deep formations such as the Marcellus
17    Shale or the Haynesville Shale (Louisiana/Texas), where the vertical distance between the base of
18    drinking water resources and the top of the shale formation may be a mile or greater. Numerical
19    modeling and microseismic studies based on a Marcellus Shale-like environment suggest that
20    fractures created during hydraulic fracturing are unlikely to extend upward from these deep
21    formations into shallow drinking water aquifers.

22    Not all hydraulic fracturing is performed in zones that are deep below drinking water resources.
23    For example, operations in the Antrim Shale (Michigan) and the New Albany Shale
24    (Illinois/Indiana/Kentucky) take place at shallower depths (100 to 1,900 ft or 30 to 579 m), with
25    less vertical separation between the formation and drinking water resources. The EPA's survey of
26    oil and gas production wells hydraulically fractured by nine service companies in 2009  and 2010
27    estimated that 20% of 23,000 wells had less than 2,000 ft (610 m) of measured distance between
28    the point of shallowest hydraulic fracturing and the base of the protected ground water resources
29    reported by well operators.

30    There are also places in the subsurface where oil and gas resources and drinking water resources
31    co-exist in the same formation. Evidence indicates that hydraulic fracturing occurs within these
32    formations. This results in the introduction of fracturing fluids into formations that may currently
33    serve, or in the future could serve, as a source of drinking water for public or private use. According
34    to the data examined, the overall frequency of occurrence of this practice appears to be low, with
35    the activity generally concentrated in some areas in the western United States. The practice of
36    injecting fracturing fluids into a formation that also contains a drinking water resource  directly
37    affects the quality of that water, since some of the fluid likely remains in the formation following
38    hydraulic fracturing. Hydraulic fracturing in a drinking water  resource is a concern in the short-

                    This document is a draft for review purposes only and does not constitute Agency policy.
      June 2015                                   ES-15                 DRAFT—DO NOT CITE OR QUOTE

-------
      Hydraulic Fracturing Drinking Water Assessment                                        Executive Summary


 1    term (should there be people currently using these zones as a drinking water supply) and the long-
 2    term (if drought or other conditions necessitate the future use of these zones for drinking water).

 3    Liquid and gas movement from the production zone to underground drinking water resources may
 4    also occur via other production wells or injection wells near hydraulic fracturing operations.
 5    Fractures created during hydraulic fracturing can intersect nearby wells or their fracture networks,
 6    resulting in the flow of fluids into those wells. These well communications, or "frac hits," are more
 7    likely to occur if wells are close to each other or on the same well pad. In the Woodford Shale in
 8    Oklahoma, the likelihood of well communication was less than 10% between wells more than 4,000
 9    ft (1,219 m) apart, but rose to nearly 50% between wells less than 1,000 ft (305 m) apart (Ajani and
10    Kelkar. 2012). If an offset well is not able to withstand the stresses applied during the hydraulic
11    fracturing of a neighboring well, well components may fail, which could result in a release of fluids
12    at the surface from the offset well. The EPA identified incidents in which surface spills of hydraulic
13    fracturing-related fluids were attributed to well communication events.

14    Older or inactive wells—including oil and gas wells, injection wells, or drinking water wells—near a
15    hydraulic fracturing operation may pose an even greater potential for impacts. A study in Oklahoma
16    found that older wells were more likely to be negatively affected by the stresses applied by
17    hydraulic fracturing in neighboring wells (Ajani and Kelkar. 2012). In some cases, inactive wells in
18    the vicinity of hydraulic fracturing activities may not have been plugged properly—many wells
19    plugged before the 1950s were done so with little or no cement. The Interstate Oil and Gas Compact
20    Commission estimates that over one million wells may have been drilled in the United States prior
21    to a formal regulatory system being in place, and the status and location of many of these wells are
22    unknown (IOGCC, 2008). State programs exist to plug identified inactive wells, and work is on-
23    going to identify and address such wells.

           Flowback and Produced Water
24    Water, of variable quality, is a byproduct of oil and gas production. After hydraulic fracturing, the
25    injection pressure is released and water flows back from the well. Initially this water is similar to
26    the hydraulic  fracturing fluid, but as time goes on the composition is affected by the characteristics
27    of the formation and possible reactions between the formation and the fracturing fluid. Water
28    initially produced from the well after hydraulic fracturing is sometimes called flowback in the
29    literature, and the term appears in this assessment  However, hydraulic fracturing fluids and any
30    formation water returning to the surface are often referred to collectively as produced water. This
31    definition of produced water is used in this assessment

32    The amountof produced water varies, but typically averages 10% to 25% of injected volumes,
33    depending upon the amount of time since fracturing and the particular well (see Figure ES-3a).
34    However, there are exceptions to this, such as in the Barnett Shale in Texas where the total volume
35    of produced water can equal or exceed the injected volume of hydraulic fracturing fluid (see Figure
36    ES-3b). Flow rates are generally high initially, and then decrease over time throughout oil or gas
37    production.
                   This document is a draft for review purposes only and does not constitute Agency policy.
      June 2015                                    ES-16                 DRAFT—DO NOT CITE OR QUOTE

-------
      Hydraulic Fracturing Drinking Water Assessment                                        Executive Summary


 1    Impacts on drinking water resources have the potential to occur if produced water is spilled and
 2    enters surface water or ground water. Environmental transport of chemical constituents in
 3    produced water depends on the characteristics of the spill (e.g., volume and duration), the
 4    composition of spilled fluids, and the characteristics of the surrounding environment

        Research Questions: Flowback and Produced Water

         •   What is currently known about the frequency, severity, and causes of spills of flow/back and
             produced water?
 5    Surface spills of produced water from hydraulically fractured wells have occurred. As noted in the
 6    Chemical Mixing section above, the frequency of on-site spills from hydraulic fracturing activities
 7    could be estimated for two states, but not nationally. Estimates of spill frequencies at hydraulic
 8    fracturing sites in Colorado and Pennsylvania, including spills of produced water, ranged from
 9    approximately 0.4 to 12.2 spills per 100 wells. Available data generally precluded estimates of
10    produced water spill rates separately from estimates of overall spill frequency. Away from the well,
11    produced water spills from pipelines and truck transport also have the potential to impact drinking
12    water resources.

13    The EPA characterized spill volumes and causes for 225 cases in which produced water spilled on
14    or near a well pad. These spills occurred between January 2006 and April 2012 in 11 states. The
15    median reported volume per produced water spill was 990 gallons (3,750 L), more than double that
16    for spills of hydraulic fracturing fluids and chemicals. The causes of produced water spills were
17    reported as human error, equipment failure, container integrity failure, miscellaneous causes (e.g.,
18    well communication), and unknown causes. Most of the total volume spilled (74%) for all 225 cases
19    combined was caused by a failure of container integrity.

         •   What is the composition of hydraulic fracturing flowback and produced water, and what
             factors might influence this composition?
20    A combination of factors influence the composition of produced water, including: the composition
21    of injected hydraulic fracturing fluids, the type of formation fractured, subsurface processes, and
2 2    residence time. The initial chemical composition of produced water primarily reflects the chemistry
23    of the injected fluids. At later times, the chemical composition of produced water reflects the
24    geochemistry of the fractured formation.

25    Produced water varies in quality from fresh to highly saline, and can contain high levels of major
26    anions and cations, metals, organics, and naturally occurring radionuclides. Produced water from
27    shale and tight gas formations typically contains high levels of total dissolved solids (TDS) and ionic
28    constituents (e.g., bromide, calcium, chloride, iron, potassium, manganese, magnesium, and
29    sodium). Produced water also may contain metals (e.g., barium, cadmium, chromium, lead, and
30    mercury), and organic compounds such as benzene. Produced water from coalbed methane
31    typically has much lower TDS levels compared to other produced water types, particularly if the
32    coalbed was deposited under fresh water conditions..
                   This document is a draft for review purposes only and does not constitute Agency policy.
      June 2015                                   ES-17                  DRAFT—DO NOT CITE OR QUOTE

-------
      Hydraulic Fracturing Drinking Water Assessment                                        Executive Summary


 1    We identified 134 chemicals that have been detected in hydraulic fracturing produced water. These
 2    include chemicals added during the chemical mixing stage, as well as naturally occurring organic
 3    chemicals and radionuclides, metals, and other constituents of subsurface rock formations
 4    mobilized by the hydraulic fracturing process. Data on measured chemical concentrations in
 5    produced water were available for 75 of these 134 chemicals.

 6    Most of the available data on produced water content are for shale and coalbed methane
 7    formations, while less data are available for tight formations, such as sandstones. The composition
 8    of produced water must be determined through sampling and analysis, both of which have
 9    limitations—the former due to challenges in accessing production equipment, and the latter due to
10    difficulties identifying target analytes before analysis and the lack of appropriate analytical
11    methods. Most current data are for inorganic chemicals, while less data exist for organic chemicals.
12    Many more organic chemicals were reported as used in hydraulic fracturing fluid than have been
13    identified in produced water. The difference may be due to analytical limitations, limited study
14    scopes, and undocumented subsurface reactions.

         •   What are the chemical, physical, and toxicological properties of hydraulic fracturing flowback
             and produced water constituents?
15    The identified constituents of produced water include inorganic chemicals (cations and anions, i.e.,
16    metals, metalloids, non-metals, and radioactive materials), organic chemicals and compounds, and
17    unidentified materials measured as total organic carbon and dissolved organic carbon. Some
18    constituents are readily transported with water (i.e., chloride and bromide), while others depend
19    strongly on the geochemical conditions in the receiving water body (i.e., radium and barium), and
20    assessment of their transport is based on site-specific factors. We were able to obtain actual or
21    estimated physicochemical properties for 86 (64%) of the 134 chemicals identified in produced
22    water.

23    As in the case of chemicals in hydraulic fracturing fluid, chemical properties that affect the
24    likelihood of an organic chemical in produced water reaching and impacting drinking water
25    resources include: mobility, solubility, and volatility. In general, physicochemical properties suggest
26    that organic chemicals in produced water tend to be less mobile in the environment Consequently,
27    if spilled, these chemicals may remain in soils or sediments near spill sites. Low mobility may result
28    in smaller dissolved contaminant plumes in ground water, although these chemicals can be
29    transported with sediments  in surface water or small particles in ground water. Organic chemical
30    properties vary with salinity, and effects depend on the nature of the chemical.

31    Oral reference values and/or oral slope factors from selected federal, state, and international
32    sources were available for 83 (62%) of the 134 chemicals detected in produced water. From U.S.
33    federal sources alone, chronic oral reference values were available for 70 (52%) of the 134
34    chemicals, and oral slope factors were available for 20 chemicals (15%). Of the chemicals that had
35    values available, noted health effects include the potential for carcinogenesis, immune system
36    effects, changes in body weight, changes in blood chemistry, pulmonary toxicity, neurotoxicity, liver
37    and kidney toxicity, and reproductive and developmental toxicity. As noted above, evaluating any
3 8    potential risk to human populations would require knowledge of the specific chemicals  that are

                   This document is a draft for review purposes only and does not constitute Agency policy.
      June 2015                                   ES-18                 DRAFT—DO NOT CITE OR QUOTE

-------
      Hydraulic Fracturing Drinking Water Assessment                                       Executive Summary


 1    present at a particular site, whether or not humans are exposed to those chemicals and, if so, at
 2    what levels and for what duration, and the toxicity of the chemicals. The chemicals present in
 3    produced water can vary based on the formation and specific well, due to differences in fracturing
 4    fluid formulation and formation geology. Accordingly, consideration of hazards and risks associated
 5    with these chemicals would be most useful on a site-specific basis and is beyond the scope of this
 6    assessment

         •   If spills occur, how might hydraulic fracturing flowback and produced water contaminate
             drinking water resources?
 7    Impacts to drinking water resources from spills or releases of produced water depend on the
 8    volume, timing, and composition of the produced water. Impacts are more likely the greater the
 9    volume of the spill, the longer the duration of the release, and the higher the concentration of
10    produced water constituents (i.e., salts, naturally occurring radioactive material, and metals).

11    The EPA characterization of hydraulic fracturing-related spills found that 8% of the 225 produced
12    water spills included in the study reached surface water or ground water. These spills tended to be
13    of greater volume than spills that did not reach a water body. A well blowout in Bradford County,
14    Pennsylvania spilled an estimated 10,000 gal (38,000 L) of produced water into a tributary of
15    Towanda Creek, a state-designated trout fishery. The largest volume spill identified in this
16    assessment occurred in North Dakota, where approximately 2.9 million gal (11 million L) of
17    produced water spilled from a broken pipeline and impacted surface and ground water.

18    Chronic releases can and do occur from produced water disposed in unlined pits or impoundments,
19    and can have long-term impacts. Ground water impacts may persist longer than surface water
20    impacts because of lower flow rates and decreased mixing. Plumes from unlined pits used for
21    produced water have been shown to persist for long periods and extend to nearby surface water
22    bodies.

           Wastewater Management and Waste Disposal
2 3    Hydraulic fracturing generates large volumes of produced water that require management. In this
24    section we refer to produced water and any other waters generated onsite by the single term
25    "wastewater." Clark and Veil (2009) estimated that, in 2007, approximately one million active oil
26    and gas wells in the United States generated 2.4 billion gal per day (9.1 billion L per day) of
27    wastewater. There is currently no reliable way to estimate what fraction of this total volume can be
28    attributed to hydraulically fractured wells. Wastewater volumes in a region can increase sharply as
29    hydraulic fracturing activity increases.

3 0    Wastewater management and disposal could affect drinking water resources through multiple
31    mechanisms, including: inadequate treatment of wastewater prior to discharge to a receiving
32    water, accidental releases during transport or leakage from wastewater storage pits, unpermitted
3 3    discharges, migration of constituents in wastewaters following land application, inappropriate
34    management of residual materials from treatment, or accumulation of wastewater constituents in
35    sediments near outfalls of centralized waste treatment facilities (CWTs) or publicly owned
36    treatment works (POTWs) that have treated hydraulic fracturing wastewater. The scope of this

                   This document is a draft for review purposes only and does not constitute Agency policy.
      June 2015                                   ES-19                  DRAFT—DO NOT CITE OR QUOTE

-------
      Hydraulic Fracturing Drinking Water Assessment                                        Executive Summary


 1    assessment excludes potential impacts to drinking water from the disposal of hydraulic fracturing
 2    wastewater in underground injection control (UIC) wells.

        Research Questions: Wastewater Management and Waste Disposal

         •   What are the common treatment and disposal methods for hydraulic fracturing wastewater,
             and where are these methods practiced?
 3    Hydraulic fracturing wastewater is managed using several options, including: disposal in UIC wells
 4    (also called disposal wells); through evaporation ponds; treatment at CWTs, followed by reuse or
 5    by discharge to either surface waters or POTWs; reuse with minimal or no treatment; and land
 6    application or road spreading. Treatment of hydraulic fracturing wastewater by POTWs was used in
 7    the past in Pennsylvania. This decreased sharply following new state-level requirements and a
 8    request by the Pennsylvania Department of Environmental Protection (PA DEP) for well operators
 9    to stop sending Marcellus Shale wastewater to POTWs (and 15 CWTs) discharging to surface
10    waters.

11    Wastewater management decisions are generally based on the availability and associated costs
12    (including transportation) of disposal or treatment facilities. A survey of state agencies found that,
13    in 2007, more than 98% of produced water from the oil and gas industry was managed via
14    underground injection (Clark and Veil. 2009). Available information suggests that disposal wells are
15    also the primary management practice for hydraulic fracturing wastewater in most regions in the
16    United States (e.g., the Barnett Shale; see Figure ES-3b). The Marcellus Shale region is a notable
17    exception, where most wastewater is reused because of the small number of disposal wells in
18    Pennsylvania (see Figure ES-3a). Although this assessment does not address potential effects on
19    drinking water resources from the use of disposal wells, any changes in cost of disposal or
20    availability of disposal wells would likely influence wastewater management decisions.

21    Wastewater from some hydraulic fracturing operations is sent to CWTs, which may discharge
22    treated wastewater to surface waters, POTWs, or back to well operators for reuse in other
23    hydraulic fracturing operations. Available data indicate that the use of CWTs for treating hydraulic
24    fracturing wastewater is greater in the Marcellus Shale region than other parts of the country. Most
25    of the CWTs accepting hydraulic fracturing wastewater in Pennsylvania cannot significantly reduce
26    TDS, and many of these facilities provide treated wastewater to well operators for reuse and do not
27    currently discharge treated wastewater to surface water.

28    Reuse of wastewater for subsequent hydraulic fracturing operations may require no treatment,
29    minimal treatment, or more extensive treatment. Operators reuse a substantial amount (ca. 70-
30    90%) of Marcellus Shale wastewater in Pennsylvania (see Figure ES-3a). Lesser amounts of reuse
31    occur in other areas (e.g., the Barnett Shale; see Figure ES-3b). In certain formations, such as the
32    Bakken Shale in North Dakota, there is currently no indication of appreciable reuse.

33    In some cases, wastewater is used for land applications such as irrigation  or road spreading for
34    deicing or dust suppression. Land application has the potential to introduce wastewater
35    constituents  into surface water and ground water due to runoff and migration of brines. Studies of

                   This document is a draft for review purposes only and does not constitute Agency policy.
      June 2015                                   ES-20                  DRAFT—DO NOT CITE OR QUOTE

-------
      Hydraulic Fracturing Drinking Water Assessment                                        Executive Summary


 1    road spreading of conventional oil and gas brines have found elevated levels of metals in soils and
 2    chloride in ground water.

         •   How effective are conventional POTWs and commercial treatment systems in removing
             organic and inorganic contaminants of concern in hydraulic fracturing wastewater?
 3    Publicly owned treatment works using basic treatment processes are not designed to effectively
 4    reduce TDS concentrations in highly saline hydraulic fracturing wastewater—although specific
 5    constituents or constituents groups can be removed (e.g., metals, oil, and grease by chemical
 6    precipitation or other processes). In some cases, wastewater treated at CWTs may be sent to a
 7    POTW for additional treatment and discharge. It is blended with POTW influent to prevent
 8    detrimental effects on biological processes in the POTW that aid in the treatment of wastewater.

 9    Centralized waste treatment facilities with advanced wastewater treatment options such as reverse
10    osmosis, thermal distillation, or mechanical vapor recompression, reduce TDS concentrations and
11    can treat contaminants currently known to be in hydraulic fracturing wastewater. However, there
12    are limited data on the composition of hydraulic fracturing wastewater, particularly for organic
13    constituents. It is unknown whether advanced treatment systems are effective at removing
14    constituents that are generally not tested for.

         •   What are the potential impacts from surface water disposal of treated hydraulic fracturing
             wastewater on drinking water treatment facilities?
15    Potential impacts to drinking water resources may occur if hydraulic  fracturing wastewater is
16    inadequately treated and discharged to surface water. Inadequately treated hydraulic fracturing
17    wastewater may increase concentrations of TDS, bromide, chloride, and iodide in receiving waters.
18    In particular, bromide and iodide are precursors of disinfection byproducts (DBPs) that can form in
19    the presence of organic carbon in drinking water treatment plants or  wastewater treatment plants.
20    Drinking water treatment plants are required to monitor for certain types of DBPs, because some
21    are toxic and can cause cancer.

2 2    Radionuelides can also be found in inadequately treated hydraulic fracturing wastewater from
2 3    certain shales, such as the Marcellus. A recent study by the PADEP (2015b) found elevated radium
24    concentrations in the tens to thousands of picocuries per liter and gross alpha and gross beta in the
25    hundreds to thousands of picocuries per liter in effluent samples from some CWTs receiving oil and
26    gas wastewater. Radium, gross alpha, and gross beta were also detected in effluents from POTWs
27    receiving oil and gas wastewater (mainly as effluent from CWTs), though at lower concentrations
28    than from the CWTs. Research in Pennsylvania also indicates the accumulation of radium in
29    sediments and soils affected by the outfalls of some treatment plants that have handled oil and gas
30    wastewater, including Marcellus Shale wastewater, and other wastewaters (PADEP, 2015b:
31    Warner etal., 2013a). Mobilization of radium from sediments and potential impacts on downstream
32    water quality depend upon how strongly the radium has sorbed to sediments. Impacts may also
33    occur if sediment is resuspended (e.g., following storm events). There is no evidence of radionuclide
34    contamination in drinking water intakes due to inadequately treated hydraulic fracturing
35    wastewater.
                   This document is a draft for review purposes only and does not constitute Agency policy.
      June 2015                                   ES-21                  DRAFT—DO NOT CITE OR QUOTE

-------
      Hydraulic Fracturing Drinking Water Assessment                                        Executive Summary


 1    Hydraulic fracturing wastewaters contain other constituents such as barium, boron, and heavy
 2    metals. Barium in particular has been documented in some shale gas produced waters. Little data
 3    exist on metal and organic compound concentrations in untreated and treated wastewaters in
 4    order to evaluate whether treatment is effective, and whether there are potential downstream
 5    effects on drinking water resources when wastewater is treated and discharged.

      Key Data Limitations and Uncertainties

 6    This assessment used available data and literature to examine the potential impacts of hydraulic
 7    fracturing from oil and gas on drinking water resources nationally. As part of this effort, we
 8    identified data limitations and uncertainties associated with current information on hydraulic
 9    fracturing and its potential to affect drinking water resources. In particular, data limitations
10    preclude a determination of the frequency of impacts with any certainty. These limitations and
11    uncertainties are discussed in brief below.

           Limitations in Monitoring  Data and Chemical Information
12    While many activities conducted as part of the hydraulic fracturing water cycle take place above
13    ground, hydraulic fracturing itself occurs below ground and is not directly observable. Additionally,
14    potential mechanisms identified in this assessment may result in impacts to drinking water
15    resources that are below ground (e.g.,  spilled fluids leaching into ground water). Data that could be
16    used to characterize the presence, migration, or transformation of chemicals in the subsurface
17    before, during, and after hydraulic fracturing were found to be scarce relative to the number of
18    hydraulically fractured oil and gas production wells. Specifically, local water quality data needed to
19    compare pre- and post-hydraulic fracturing conditions are not consistently collected or readily
20    available. The limited amount of data collected before and during hydraulic fracturing activities
21    reduces the ability to determine whether hydraulic fracturing affected drinking water resources in
22    cases of alleged contamination.

23    Information (identity, frequency of use, physicochemical and toxicological properties, etc.) on the
24    chemicals associated with the hydraulic fracturing water cycle is not complete and limits
2 5    understanding of potential impacts on drinking water resources. Well operators claimed at least
26    one chemical as confidential at more than 70% of wells reported to FracFocus and analyzed by the
27    EPA. The identity of these chemicals, and other chemicals in produced water, are needed to
28    understand their properties and would also help inform what chemicals to test for to  establish
29    baseline conditions and to test for in the event of a suspected drinking water impact. Of the 1,173
30    total chemicals identified by the EPA in hydraulic fracturing fluid and flowback and produced
31    water, 147 have chronic oral reference values and/or oral slope factors from the sources that met
32    the selection criteria for inclusion in this assessment. Because the majority of chemicals identified
33    in this report do not have chronic oral reference values and/or oral slope factors, risk assessors at
34    the local and regional level may need to use alternative sources of toxicity information that could
35    introduce greater uncertainties.
                   This document is a draft for review purposes only and does not constitute Agency policy.
      June 2015                                    ES-22                 DRAFT—DO NOT CITE OR QUOTE

-------
      Hydraulic Fracturing Drinking Water Assessment                                        Executive Summary


           Other Contributing Limitations
 1    We found other limitations that hamper the ability to fully assess the potential impacts of hydraulic
 2    fracturing on drinking water resources nationally. These include the number and location of
 3    hydraulically fractured wells, the location of drinking water resources, and information on changes
 4    in industry practices. The lack of a definitive well count particularly contributes to uncertainties
 5    regarding total water use or total wastewater volume estimates, and would limit any kind of
 6    cumulative impact assessment Lack of specific information about private drinking water well
 7    locations and the depths of drinking water resources in relation to hydraulically fractured rock
 8    formations and well construction features (e.g., casing and cement) limits the ability to assess
 9    whether subsurface drinking water resources are isolated from hydraulically fractured oil and gas
10    production wells. Finally, this assessment is a snapshot in time, and the industry is rapidly changing
11    (e.g., the number of wells fractured, the location of activities, and the chemicals used). It is unclear
12    how changes in industry practices could affect potential drinking water impacts in the future.
13    Consideration of future development scenarios was not a part of this assessment, but such an
14    evaluation could help establish potential short- and long-term impacts to drinking water  resources
15    and how to assess them.

      Conclusions

16    Through this national-level assessment, we have identified potential mechanisms by which
17    hydraulic fracturing could affect drinking water resources. Above ground mechanisms can affect
18    surface and ground water resources and include water withdrawals at times or in locations of low
19    water availability, spills of hydraulic fracturing fluid and chemicals or produced water, and
20    inadequate treatment and discharge of hydraulic fracturing wastewater. Below ground mechanisms
21    include movement of liquids and gases via the production well into underground drinking water
22    resources and movement of liquids and gases from the fracture zone to these resources via
23    pathways in  subsurface rock formations.

24    We did not find evidence that these mechanisms have led to widespread, systemic impacts on
25    drinking water resources in the United States. Of the potential mechanisms identified in this report,
26    we found specific instances where one or more of these mechanisms led to impacts on drinking
27    water resources, including contamination of drinking water wells. The cases occurred during both
28    routine activities and accidents and have resulted in impacts to surface or ground water.  Spills of
29    hydraulic fracturing fluid and produced water in certain cases have reached drinking water
30    resources, both surface and ground water. Discharge of treated hydraulic fracturing wastewater has
31    increased contaminant concentrations in receiving surface waters. Below ground movement of
32    fluids, including gas, most likely via the production well, have contaminated drinking water
33    resources. In some cases, hydraulic fracturing fluids have also been directly injected into  drinking
34    water resources, as defined in this assessment, to produce oil or gas that co-exists in those
35    formations.

36    The number  of identified cases where drinking water resources were impacted are small relative to
37    the number of hydraulically fractured wells. This could reflect a rarity of effects on drinking water


                    This document is a draft for review purposes only and does not constitute Agency policy.
      June 2015                                   ES-23                  DRAFT—DO NOT CITE OR QUOTE

-------
      Hydraulic Fracturing Drinking Water Assessment                                           Executive Summary


 1    resources, or may be an underestimate as a result of several factors. There is insufficient pre- and
 2    post-hydraulic fracturing data on the quality of drinking water resources. This inhibits a
 3    determination of the frequency of impacts. Other limiting factors include the presence of other
 4    causes of contamination, the short duration of existing studies, and inaccessible information related
 5    to hydraulic fracturing activities.

 6    This state-of-the-science assessment contributes to the understanding of the potential impacts of
 7    hydraulic fracturing on drinking water resources and the factors that may influence those impacts.
 8    The findings in this assessment can be used by federal, state, tribal, and local officials; industry; and
 9    the public to better understand and address any vulnerabilities of drinking water resources to
10    hydraulic fracturing activities. This assessment can also be used to help facilitate and inform
11    dialogue among interested stakeholders, and support future efforts, including: providing context to
12    site-specific exposure or risk assessments, local and regional public health assessments, and
13    assessments of cumulative impacts of hydraulic fracturing on drinking water resources overtime
14    or over defined geographic areas of interest  Finally, and most importantly, this assessment
15    advances the scientific basis for  decisions by federal, state, tribal, and local officials, industry, and
16    the public, on how best to protect drinking water resources now and in the future.


      References for Executive Summary

      Ajani, A: Kelkar, M. (2012). Interference study in shale plays. Paper presented at SPE Hydraulic Fracturing
         Technology Conference, February 6-8, 2012, The Woodlands, TX.

      Bair. ES: Freeman. DC: Senko. IM. (2010). Subsurface gas invasion Bainbridge Township, Geauga County, Ohio.
         (Expert Panel Technical Report). Columbus, OH: Ohio Department of Natural Resources.
         http://oilandgas.ohiodnr.gov/resources/investigations-reports-violations-reformstfTHR

      Clark, CE: Veil, IA. (2009). Produced water volumes and management practices in the United States (pp. 64).
         (ANL/EVS/R-09/1). Argonne, IL: Argonne National Laboratory.
         http://www.circleofblue.org/waternews/wp-
         content/uploads/2010/09/ANL EVS  R09  produced water volume report 2437.pdf

      COGCC. Colorado Oil and Gas Conservation Commission Order No. 1V-276. (2004).
         https://cogcc.state.co.us/orders/orders/lv/276.html

      Crescent (Crescent Consulting, LLC). (2011). East Mamm creek project drilling and cementing study.
         Oklahoma City, OK. http://cogcc.state.co.us/Library/PiceanceBasin/EastMammCreek/ReportFinal.pdf

      EIA (Energy Information Administration).  (2015a). Glossary. Available online at
         http://www.eia.gov/tools/glossary/

      IOGCC (Interstate Oil and Gas Compact Commission). (2008). Protecting our country's resources: The states'
         case, orphaned well plugging initiative. Oklahoma City, OK: Interstate Oil and Gas Compact Commission
         (IOGCC). http://iogcc.myshopify.com/products/protecting-our-countrys-resources-the-states-case-
         orphaned-well-plugging-initiative-2008

      Maupin. MA: Kenny. IF: Hutson. SS:  Lovelace. IK: Barber. NL: Linsey. KS. (2014). Estimated use of water in the
         United States in 2010. (USGS Circular 1405). Reston, VA: U.S. Geological Survey.
         http://dx.doi.org/10.3133/cirl405

      Michie. TW: Koch. CA. (1991). Evaluation of injection-well risk management in the Williston Basin. J Pet Tech
         43: 737-741. http://dx.doi.org/10.2118/20693-PA
                    This document is a draft for review purposes only and does not constitute Agency policy.

      June 2015                                     ES-24                   DRAFT—DO NOT CITE OR QUOTE

-------
Hydraulic Fracturing Drinking Water Assessment                                            Executive Summary
Newell. R. (2011). Shale gas and the outlook for U.S. natural gas markets and global gas resources.
   Presentation presented at US EIA presentation at OECD Meetings, June 21, 2011, Paris, France.

ODNR. DMRM. (Ohio Department of Natural Resources, Division of Mineral Resources Management). (2008).
   Report on the investigation of the natural gas invasion of aquifers in Bainbridge Township of Geauga
   County, Ohio. Columbus, OH: ODNR.
   http://oilandgas.ohiodnr.gov/portals/oilgas/pdf/bainbridge/report.pdf

PA PEP (Pennsylvania Department of Environmental Protection). (2015b). Technologically enhanced
   naturally occurring radioactive materials (TENORM) study report. Harrisburg, PA.
   http://www.elibrary.dep.state.pa.us/dsweb/Get/Document-105822/PA-DEP-TENORM-
   Study Report Rev. 0 01-15-2015.pdf

Papoulias. DM: Velasco. AL. (2013). Histopathological analysis offish from Acorn Fork Creek, Kentucky,
   exposed to hydraulic fracturing fluid releases. Southeastern Naturalist 12: 92-111.

Scanlon. BR: Reedy. RC: Nicot. IP. (2014). Will water scarcity in semiarid regions limit hydraulic fracturing of
   shale plays? Environmental Research Letters  9. http://dx.doi.Org/10.1088/1748-9326/9/12/124011

Science Based Solutions LLC. (2014). Summary of hydrogeology investigations in the Mamm Creek field area,
   Garfield County. Laramie, Wyoming. http://www.garfield-county.com/oil-gas/documents/Summary-
   Hydrogeologic-Studies-Mamm%2 OCreek-Area-Feb-10-2014.pdf

U.S. EPA (U.S. Environmental Protection Agency). (2015J). Retrospective case study in Killdeer, North Dakota:
   study of the potential impacts of hydraulic fracturing on drinking water resources [EPA Report]. (EPA
   600/R-14/103). Washington, D.C.

USGS (U.S. Geological Survey). (2002). Natural gas production in the United States [Fact Sheet]. (USGS Fact
   Sheet FS-113-01). Denver, CO.

Vengosh. A: lackson. RB: Warner. N: Darrah. TH:  Kondash. A. (2014). A critical review of the risks to water
   resources from unconventional shale gas development and hydraulic fracturing in the United States.
   Environ Sci Technol 48: 36-52. http://dx.doi.org/10.1021/es405118y

Warner. NR: Christie. CA: lackson. RB: Vengosh. A. (2013a). Impacts of shale gas wastewater disposal on
   water quality in western Pennsylvania. Environ Sci Technol 47:11849-11857.
   http://dx.doi.org/10.1021/es402165b
               This document is a draft for review purposes only and does not constitute Agency policy.

June 2015                                      ES-25                   DRAFT—DO NOT CITE OR QUOTE

-------
Hydraulic Fracturing Drinking Water Assessment                               Chapter 1 - Introduction
                              Chapter 1
Introduction
            This document is a draft for review purposes only and does not constitute Agency policy.
June 2015                                                  DRAFT—DO NOT CITE OR QUOTE

-------
      Hydraulic Fracturing Drinking Water Assessment                                     Chapter 1 - Introduction
      1. Introduction

      1.1.  Background
 1    Since the early 2000s, oil and natural gas production in the United States has been transformed
 2    through the technological innovations of hydraulic fracturing and directional drilling. Hydraulic
 3    fracturing is a stimulation technique used to increase production of oil and gas. It involves the
 4    injection of fluids under pressures great enough to fracture the oil- and gas-production formations.
 5    Hydraulic fracturing in combination with advanced directional drilling techniques has made it
 6    possible to economically extract hydrocarbons from unconventional resources, such as shale, tight
 7    formations, and coalbeds.1 It can also enhance production from conventional resources. The surge
 8    in use of hydraulic fracturing and associated technologies has significantly increased domestic
 9    energy supplies (see Chapter 2) and brought economic benefits to many areas of the United States.

10    The growth in domestic oil and gas exploration and production- the direct result of the expanded
11    use of hydraulic fracturing- has also raised concerns about its potential for impacts to human
12    health and the environment. Specific concerns have been raised by the public about the effects of
13    hydraulic fracturing on the quality and quantity of drinking water resources. Some residents living
14    close to oil and gas production well sites report changes in the quality of ground water resources
15    used for drinking water and assert that hydraulic fracturing is responsible for these changes. Other
16    concerns include competition for water between hydraulic fracturing operations and other water
17    users, especially in areas of the country experiencing drought, and the disposal of waste water
18    generated from hydraulic fracturing. In response to public concerns, the U.S. Congress urged the
19    U.S. Environmental Protection Agency (EPA) to study the relationship between hydraulic fracturing
20    and drinking water (H.R. Rep. 111-316. 20091. In 2011, the EPA published its Plan to Study the
21    Potential Impacts of Hydraulic Fracturing on Drinking Water Resources (U.S. EPA. 2011c: hereafter
22    Study Plan). The research described in the Study Plan began the same year. In 2012, the EPA issued
2 3    Potential Impacts of Hydraulic Fracturing on Drinking Water Resources: Progress Report (U.S. EPA,
24    2012f: hereafter Progress Report] in order to update the public on the status of the research being
25    conducted under the Study Plan. In this report, we review and synthesize scientific literature,
26    including the publications resulting from the EPA's research and information provided by
27    stakeholders, to assess the potential for hydraulic fracturing for oil and gas to change the quality or
28    quantity of drinking water resources. This report also identifies factors affecting the frequency or
29    severity of any potential impacts.

      1.2.  Scope
30    This assessment focuses on hydraulic fracturing in onshore oil and gas wells in the contiguous
31    United States; limited available information on hydraulic fracturing in Alaska is included. To the
      1 Unconventional resources is an umbrella term for oil and natural gas that is produced by means that do not meet the
      criteria for conventional production. What has qualified as unconventional at any particular time is a complex function of
      resource characteristics, the available exploration and production technologies, the economic environment, and the scale,
      frequency, and duration of production from the resource (see Text Box 2-2).
                    This document is a draft for review purposes only and does not constitute Agency policy.
      June 2015                                     1-1                    DRAFT—DO NOT CITE OR QUOTE

-------
      Hydraulic Fracturing Drinking Water Assessment                                      Chapter 1 - Introduction


 1    extent possible, this assessment addresses hydraulic fracturing in all types of oil- and gas-bearing
 2    formations in which it is conducted, including shale, so-called 'tight' formations (e.g., certain
 3    sandstones, siltstones, and carbonates), coalbeds, and conventional reservoirs. It tends to focus on
 4    hydraulic fracturing in shale, which reflects the relatively large amount of literature and available
 5    data on hydraulic fracturing in this type of geologic formation.

 6    The scope of activities examined in this assessment is defined by the hydraulic fracturing water
 7    cycle. This cycle encompasses activities involving water that support hydraulic fracturing and
 8    consists of five stages: (1) acquisition of water needed to create hydraulic fracturing fluids; (2)
 9    mixing of water and chemicals on the well pad to create hydraulic fracturing fluids; (3) injection of
10    hydraulic fracturing fluids into the well to fracture the geologic formation; (4) management of
11    flowback and produced water on the well pad and in transit for reuse, treatment, or disposal; and
12    (5) reuse, treatment and discharge, or disposal of hydraulic fracturing wastewater (see Figure
13    l-l).i.2.3-4

14    Activities within the hydraulic fracturing water cycle can take place on or near the well pad or some
15    distance away. On-site activities include mixing and injecting hydraulic fracturing fluids and
16    capturing flowback and produced water. Water withdrawals and wastewater treatment and
17    disposal may occur in the  same watershed, adjacent watersheds, or watersheds many miles away
18    from the production site.

19    This assessment focuses on impacts on drinking water resource quantity and quality. Consistent
20    with the Study Plan (U.S. EPA. 2011c], drinking water resources are defined broadly  within this
21    report as any body of ground water or surface water that now serves, or in the future could serve,
22    as a source of drinking water for public or private use. This is broader than most regulatory
23    definitions of "drinking water" and encompasses both fresh and non-fresh bodies of water, since
24    trends indicate both types of water bodies are now and in the future will be used as sources of
25    drinking water (see Chapter 3). We note that drinking water resources provide not only water that
26    individuals actually drink but also  water used for many additional purposes such as cooking and
27    bathing.

28    We assess potential effects on drinking water resources from business-as-usual operations as well
29    as from accidents and unintended releases that may occur during the hydraulic fracturing water
30    cycle (see Table 1-1).
      1 Hydraulic fracturing fluids are engineered fluids, typically consisting of a base fluid, additives, and proppants, that are
      pumped under high pressure into the well to create and hold open fractures in the formation.
      2 Flowback is defined multiple ways in the literature. In general, it is either fluids predominantly containing hydraulic
      fracturing fluid that return from a well to the surface or a process used to prepare the well for production (see Chapter 7).
      3 Produced water is water that flows from oil and gas wells.
      4 Hydraulic fracturing wastewater is flowback and produced water that is managed using practices that include but are
      not limited to reuse in subsequent hydraulic fracturing operations, treatment and discharge, and injection into disposal
      wells (see Chapter 8).
                    This document is a draft for review purposes only and does not constitute Agency policy.
      June 2015                                      1-2                    DRAFT—DO NOT CITE OR QUOTE

-------
Hydraulic Fracturing Drinking Water Assessment
                                                                                        Chapter 1 - Introduction
                                                              Well Injection
                          Chemical Mixing
                                                                                        Flowback and Produced Water
                   Water Acquisition
                                                             Wastewater Treatment and Waste Disposal
Figure 1-1. Conceptualized view of the stages of the hydraulic fracturing water cycle.
           Shown here is a generalized landscape depicting the activities of the hydraulic fracturing water cycle and their relationship to each other, as well as
           their relationship to drinking water resources. Activities may take place in the same watershed or different watersheds and close to or far from
           drinking water resources. Drinking water resources are any body of ground water or surface water that now serves, or in the future could  serve, as
           a source of drinking water for public or private use. Arrows depict the movement of water and chemicals. Specific activities in the "Wastewater
           Treatment and Waste Disposal" inset are (a) underground injection control (UIC) well disposal, (b) wastewater treatment and reuse, and (c)
           wastewater treatment and discharge at a centralized waste treatment (CWT) facility. Note: Figure not to scale.
June 2015
This document is a draft for review purposes only and does not constitute Agency policy.

                                   1-3
DRAFT—DO NOT CITE OR QUOTE

-------
      Hydraulic Fracturing Drinking Water Assessment
Chapter 1 - Introduction
      Table 1-1. Stages of the hydraulic fracturing water cycle have various potential effects on
                drinking water resources.
                The potential effects addressed in this assessment, and how they are related to the activities within
                each stage, are summarized here.
Water cycle stage
Water acquisition
Chemical mixing
Well injection
Flowback and
produced water
Wastewater
treatment and
waste disposal
Activities or processes potentially
affecting drinking water resources
Water withdrawals
Spills of hydraulic fracturing fluids
Subsurface migration of hydraulic
fracturing fluids or formation fluids
Spills of flowback or produced water
Discharge of untreated or inadequately
treated wastewater and inappropriate
disposal of waste solids
Potential drinking water effects
addressed in this assessment
Quality
Ground
water
X
X
X
X
X
Surface
water
X
X
X
X
X
Quantity
Ground
water
X




Surface
water
X




 1    As part of the assessment, we evaluated immediate, near-term, and long-term effects on drinking
 2    water resources. For example, we considered how surface spills of hydraulic fracturing fluids may
 3    potentially have immediate or near-term impacts on neighboring surface water and shallow ground
 4    water quality (see Chapters 5 and 7). We also considered how the potential release of hydraulic
 5    fracturing fluids in the subsurface may take years to impact ground water resources, because
 6    liquids and gas often move slowly in the subsurface (see Chapter 6). Additionally, effects may be
 7    detected near the activity or at some distance away. For instance, we considered that, depending on
 8    the constituents of treated hydraulic fracturing wastewater discharged to a stream and the flow in
 9    that stream, drinking water resource quality could be affected a significant distance downstream
10    (see Chapter 8).

11    This assessment focuses predominantly on activities supporting a single well or multiple wells on a
12    single well pad, accompanied by a more limited discussion of cumulative activities and the effects
13    that could result from having many wells on a landscape. Studies of cumulative effects are generally
14    lacking, but we use the scientific literature to address this topic where possible.1

15    We address mechanisms for impacts as well as impacts of hydraulic fracturing for oil and gas on
16    drinking water resources. In general, a mechanism is the means or series of events that links an
17    activity to an impact, while an impact is the end result of a mechanism and represents a change in
18    the entity of interest Specific definitions used in this assessment are provided below.
      1 Cumulative effects refer to combined changes in the environment that can take place as a result of multiple activities
      over time and/or space.
                    This document is a draft for review purposes only and does not constitute Agency policy.
      June 2015                                      1-4                   DRAFT—DO NOT CITE OR QUOTE

-------
      Hydraulic Fracturing Drinking Water Assessment                                     Chapter 1 - Introduction


 1        •   A mechanism is a means or series of events by which an activity within the hydraulic
 2            fracturing water cycle has been observed to change the quality or quantity of drinking
 3            water resources.
 4        •   A suspected mechanism is a means or series of events by which hydraulic fracturing
 5            activities could logically have resulted in an observed change in the quality or quantity of
 6            drinking water resources. Available evidence may or may not be  sufficient to determine if
 7            it is the only mechanism that caused the observed change.
 8        •   A potential mechanism is a means or series of events by which hydraulic fracturing
 9            activities could logically or theoretically (for instance, based on modeling) change the
10            quality or quantity of drinking water resources but one that has not yet been observed.
11        •   An impact is any observed change in the quality or quantity of drinking water resources,
12            regardless of severity, that results from a mechanism.
13        •   A potential impact is any change in the quality or quantity of drinking water resources
14            that could logically occur as the result of a mechanism or potential mechanism but has not
15            yet been observed.
16    Potential mechanisms and impacts, as well as suspected mechanisms, are  addressed because data
17    required to document mechanisms and impacts may be inaccessible, incomplete, or nonexistent. In
18    addition, evidence maybe insufficient to isolate the contribution of hydraulic fracturing to changes
19    in the quality or quantity of drinking water resources from other human activities occurring
20    nearby. We anticipate that our understanding of mechanisms and impacts will be advanced as the
21    scientific community continues to evaluate potential health and environmental effects of hydraulic
22    fracturing.

23    In this assessment, we also identify and discuss factors affecting the frequency or severity of
24    changes to avoid a simple inventory of all specific situations in which hydraulic fracturing might
25    alter drinking water quality or quantity. This allows knowledge about the  conditions under which
26    effects are likely or unlikely to occur to be applied to new circumstances (e.g., a new area of oil or
27    gas development where hydraulic fracturing is expected to be used) and could inform the
28    development of strategies to prevent impacts. Although no attempt has been made in this
29    assessment to identify or evaluate comprehensive best practices for states, tribes, or the industry,
30    we describe ways to avoid or reduce the  impacts of hydraulic fracturing activities as they have been
31    reported in the scientific literature. A summary and evaluation of current or proposed regulations
3 2    and policies is beyond the scope of this report

3 3    For this assessment, we did not conduct site-specific predictive modeling to quantitatively estimate
34    environmental concentrations of contaminants in drinking water resources, although modeling
35    studies conducted by others are described. Further, this report is not a human health risk
36    assessment. It does not identify populations that are exposed to chemicals or other stressors  in the
37    environment, estimate the extent of exposure, or estimate the incidence of human health impacts
38    (see Chapter 9).
                   This document is a draft for review purposes only and does not constitute Agency policy.
      June 2015                                     1-5                  DRAFT—DO NOT CITE OR QUOTE

-------
      Hydraulic Fracturing Drinking Water Assessment                                     Chapter 1 - Introduction


 1    This assessment focuses on the potential impacts from activities in the hydraulic fracturing water
 2    cycle on drinking water resources. It does not address all concerns that have been raised about
 3    hydraulic fracturing nor about oil and gas exploration and production more generally. Activities
 4    that are not considered include acquisition and transport of constituents of hydraulic fracturing
 5    fluids besides water (e.g., sand mining and chemical production); site selection and well pad
 6    development; other infrastructure development (e.g., roads, pipelines, compressor stations); site
 7    reclamation; and well closure. We consider these activities to be outside the scope of the hydraulic
 8    fracturing water cycle and, therefore, their impacts are not addressed in this assessment
 9    Additionally, this report does not discuss the potential impacts of hydraulic fracturing on other
10    water uses  (e.g., agriculture or industry), other aspects of the environment (e.g., air quality or
11    ecosystems), worker health or safety, or communities.

      1.3. Approach
12    This assessment relies on scientific literature and data that address topics within the scope of the
13    hydraulic fracturing water cycle. Scientific journal articles and peer-reviewed EPA reports that have
14    been published containing results from the EPA's hydraulic fracturing study comprise one set of
15    applicable literature. Other literature evaluated includes articles published in science and
16    engineering journals, federal and state government reports, non-governmental organization (NGO)
17    reports, and oil and gas industry publications. Data sources examined include federal- and state-
18    collected data sets, databases curated by federal and state government agencies, other publicly
19    available data and information, and data including confidential and non-confidential business
20    information submitted by industry to the EPA.*

      1.3.1.  EPA Hydraulic Fracturing Study Publications
21    The research topic areas and projects described in the Study Plan were developed with substantial
2 2    expert and public input, and they were designed to meet the data and information needs of this
23    assessment As such, published, peer-reviewed results of the research conducted under the Study
24    Plan are incorporated and cited frequently throughout this assessment As is customary in
25    assessments that synthesize a large body of literature and data, the results of EPA research are
26    contextualized and interpreted in combination with the other literature and data described in
27    Section 1.3.2. The articles and EPA reports themselves that give complete and detailed project
28    results can be found on the EPA's hydraulic fracturing website (www.epa.gov/hfstudy). For ease  of
29    reference, a description of the individual projects, the type of research activity they represent (i.e.,
30    analysis of existing data, scenario evaluation, laboratory study, or case study), and the
31    corresponding citations of published articles and EPA reports that are referenced in this
3 2    assessment can be found in Appendix H.

      1.3.2.  Literature and Data Search Strategy
33    The EPA used a broad search strategy to identify approximately 3,700 sources of scientific
34    information that could be applicable to this assessment. This search strategy included both
      1 Information was provided to the EPA by nine hydraulic fracturing service companies in response to a September 2010
      information request and by nine oil and gas well operators in response to an August 2011 information request.
                    This document is a draft for review purposes only and does not constitute Agency policy.
      June 2015                                     1-6                    DRAFT—DO NOT CITE OR QUOTE

-------
      Hydraulic Fracturing Drinking Water Assessment                                     Chapter 1 - Introduction


 1    requesting input from scientists, stakeholders, and the public about relevant data and information,
 2    and thorough searching of published information and applicable data.x

 3    Over 1,400 articles, reports, data, and other sources of information were obtained through outreach
 4    to the public, stakeholders, and scientific experts. The EPA requested material through many
 5    venues, as follows. We received recommended literature from the Science Advisory Board (SAB),
 6    the EPA's independent federal scientific advisory committee, from its review of the EPA's draft
 7    Study Plan; its consultation on the EPA's Progress Report [U.S. EPA. 2012Q: and during an SAB
 8    briefing on new and emerging information related to hydraulic fracturing in fall 2013. Subject
 9    matter experts and stakeholders also recommended literature through a series of technical
10    workshops and roundtables organized by the EPA between 2011 and 2013. In addition, the public
11    submitted material to the SAB during the SAB review of the draft Study Plan, Progress Report, and
12    briefing on emerging information, as well as in response to a formal request for data and
13    information posted in the Federal Register (EPA-HQ-ORD-2010-0674) in November 2012. The
14    submission deadline was extended from April to November 2013 to provide the public with
15    additional opportunity to provide input to the EPA.

16    Approximately 2,300 additional sources were identified by conducting searches for material that
17    could be applicable to the assessment via online scientific databases and federal, state, and
18    stakeholder websites. We searched these databases and websites in particular for (1) materials
19    addressing topics not covered by the documents submitted by experts, stakeholders, and the public
20    as noted above, and (2) newly emerging scientific studies. Multiple targeted and iterative searches
21    on topics determined to be within the scope of the assessment were conducted until fall 2014. After
22    that time, we largely included newer literature as it was recommended to  us during our internal
23    technical reviews or as it came to our attention and was determined to be  important for filling a gap
24    in information. In many cases, our searches uncovered the same material submitted by the public,
25    but approximately 2,300 new sources were also identified.

      1.3.3.  Literature and Data Evaluation Strategy
2 6    We evaluated the literature and data identified in the search strategy above using the five
27    assessment factors outlined by the EPA Science Policy Council in A Summary of General Assessment
2 8    Factors for Evaluating the Quality of Scientific and Technical Information [U.S. EPA. 20031. The
29    factors are (1) applicability and utility, (2) evaluation and review, (3) soundness, (4) clarity and
30    completeness, and (5) uncertainty and variability. Table 1-2 lists these factors along with the
31    specific criteria for each that were developed for this assessment We first evaluated all materials
32    for applicability. If "applicable" under the criteria, the reference was evaluated on the basis of the
33    other four factors.

34    Our objective was to consider and then cite literature in the assessment that fully conforms to all
35    criteria defining each assessment factor. However, the preponderance of literature on some topics
36    did not fully conform to  some aspects of the outlined criteria. For instance, there were many white
      1 This study did not review information contained in state and federal enforcement actions concerning alleged
      contamination of drinking water resources.
                   This document is a draft for review purposes only and does not constitute Agency policy.
      June 2015                                     1-7                   DRAFT—DO NOT CITE OR QUOTE

-------
      Hydraulic Fracturing Drinking Water Assessment
Chapter 1 - Introduction
 1    papers and reports in technical areas in which independent peer review is not standard practice or
 2    is not well documented. Therefore, we included references in the assessment that were not peer-
 3    reviewed but that addressed topics not found in the peer-reviewed literature, that provided useful
 4    background information, or that corroborated conclusions in the peer-reviewed literature.
      Table 1-2. Criteria developed for the five factors used to evaluate literature and data cited in
                this assessment.
                Criteria are consistent with those outlined by the EPA's Science Policy Council (U.S. EPA, 2003). Criteria
                are incorporated into the Quality Assurance Project Plans for this assessment (U.S. EPA, 2014g,
                2013d).
Factor
Applicability
Review
Soundness
Clarity/completeness
Uncertainty/variability
Criteria
Document provides information useful for assessing the potential pathways for
hydraulic fracturing activities to change the quality or quantity of drinking water
resources, identifies factors that affect the frequency and severity of impacts, or
suggests ways that potential impacts may be avoided or reduced.
Document has been peer-reviewed.
Document relies on sound scientific theory and approaches, and conclusions are
consistent with data presented.
Document provides underlying data, assumptions, procedures, and model parameters,
as applicable, as well as information about sponsorship and author affiliations.
Document identifies uncertainties, variability, sources of error, and/or bias and
properly reflects them in any conclusions drawn.
      1.3.4.  Quality Assurance and Peer Review
 5    The use of quality assurance (QA) and peer review helps ensure that the EPA conducts high-quality
 6    science that can be used to inform policymakers, industry, and the public. QA activities performed
 7    by the EPA ensure that the agency's environmental data are of sufficient quantity and quality to
 8    support the data's intended use. The EPA prepared a programmatic Quality Management Plan [U.S.
 9    EPA. 2014h] for all of the research conducted under the EPA's Study Plan, including the review and
10    synthesis of the scientific literature in this assessment. The hydraulic fracturing Quality
11    Management Plan describes the QA program's organizational structure; defines and assigns QA and
12    quality control (QC) responsibilities; and describes the processes and procedures used to plan,
13    implement, and assess the effectiveness of the quality system. The broad plan is then supported by
14    more detailed QA Project Plans (QAPPs). For instance, the QAPPs developed for this assessment
15    provide the technical approach and associated QA/QC procedures for our data and literature search
16    and evaluation strategies introduced in Section 1.3.2 and 1.3.3 [U.S. EPA. 2014g. 2013d]. A QA audit
17    was conducted by the QA Manager during the preparation of this assessment in order to verify that
18    the appropriate QA procedures, criteria, reviews, and data verification were adequately performed
19    and documented. Identifying uncertainties is another aspect of QA; uncertainty, including data gaps
20    and data limitations, is discussed throughout this assessment
                   This document is a draft for review purposes only and does not constitute Agency policy.
      June 2015                                     1-8                   DRAFT—DO NOT CITE OR QUOTE

-------
      Hydraulic Fracturing Drinking Water Assessment                                     Chapter 1 - Introduction


 1    This report is classified as a Highly Influential Scientific Assessment (HISA), defined by the Office of
 2    Management and Budget (OMB) as a scientific assessment that (1) could have a potential impact of
 3    more than $500 million in any year or (2) is novel, controversial, or precedent-setting or has
 4    significant interagency interest [OMB, 2004]. The OMB describes specific peer review requirements
 5    for HISAs. The EPA often engages the SAB as an external federal advisory committee to conduct
 6    peer reviews of high-profile scientific matters relevant to the agency. Members of an ad hoc panel,
 7    the same panel that was convened under the auspices of the SAB to provide comment on the
 8    Progress Report, will also provide comment on this assessment.1 Panel members were nominated
 9    by the public and chosen to create a balanced review panel based on factors such as technical
10    expertise, knowledge, experience, and absence of any real or perceived conflicts of interest

      1.4. Organization
11    This assessment begins with a general description of hydraulic fracturing activities and the role of
12    hydraulic fracturing in the oil and gas industry in the United States (see Chapter 2). It follows with a
13    characterization of drinking water resources in the continental United States, with a focus on areas
14    in which we estimate hydraulic fracturing has taken place over the time period of 2000-2013 (see
15    Chapter 3).

16    Chapters 4 through 8 are organized around the stages of the hydraulic fracturing water cycle (see
17    Figure 1-1) and address the potential for activities conducted during those stages to change the
18    quality or quantity of drinking water resources. Each of the stages is covered by a separate chapter.
19    There is also a chapter devoted to an examination of the properties of chemicals and constituents
20    that have been or may be used in hydraulic fracturing fluids or present in flowback and produced
21    water (see Chapter 9).

22    Each chapter addresses research questions developed under the Study Plan, as data and
23    information allow (see Table 1-3). Concise answers appear in text boxes at the end of each chapter.
24    The final chapter provides major conclusions and a synthesis of information presented across the
25    assessment It also highlights significant gaps in information that contribute to uncertainties about
26    those conclusions (see Chapter 10).

      Table 1-3. Research questions addressed by this assessment.
                Each chapter addresses research questions developed under the Study Plan. Chapters 2 and 3 develop
                background on hydraulic fracturing and drinking water resources, respectively.
      Chapter and water cycle stage   Research questions
      1 Information about this process is available online at http://yosemite.epa.gov/sab/sabproduct.nsf/
      02ad90bl36fc21ef85256eba00436459/b436304ba804e3f885257a5b00521b3b!ODenDocument.
                    This document is a draft for review purposes only and does not constitute Agency policy.
      June 2015                                     1-9                   DRAFT—DO NOT CITE OR QUOTE

-------
Hydraulic Fracturing Drinking Water Assessment
                                              Chapter 1 - Introduction
 Chapter and water cycle stage
Research questions
 Chapter 4 - Water Acquisition
        What are the types of water used for hydraulic fracturing?
        How much water is used per well?
        How might cumulative water withdrawals for hydraulic
        fracturing affect drinking water quantity?
        What are the possible impacts of water withdrawals for
        hydraulic fracturing on water quality?
 Chapter 5 - Chemical Mixing
        What is currently known about the frequency, severity, and
        causes of spills of hydraulic fracturing fluids and chemical
        additives?
        What are the identities and volumes of chemicals used in
        hydraulic fracturing fluids, and how might this composition
        vary at a given site and across the country?
        What are the chemical and physical properties of hydraulic
        fracturing chemical additives?
        If spills occur, how might hydraulic fracturing chemical
        additives contaminate drinking water resources?
 Chapter 6 - Well Injection
        How effective are current well construction practices at
        containing fluids- both liquids and gases- before, during,
        and after fracturing?
        Can subsurface migration of fluids- both liquids and gases-
        to drinking water resources occur, and what local geologic
        or artificial features might allow this?
 Chapter 7 - Flowback and
 Produced Water
        What is currently known about the frequency, severity, and
        causes of spills of flowback and produced water?
        What is the composition of hydraulic fracturing flowback
        and produced water, and what factors might influence this
        composition?
        What are the chemical and physical properties of hydraulic
        fracturing flowback and produced water constituents?
        If spills occur,  how might hydraulic fracturing flowback and
        produced water contaminate drinking water resources?
              This document is a draft for review purposes only and does not constitute Agency policy.
June 2015                                    1-10                   DRAFT—DO NOT CITE OR QUOTE

-------
      Hydraulic Fracturing Drinking Water Assessment
                                             Chapter 1 - Introduction
      Chapter and water cycle stage
Research questions
      Chapter 8 - Wastewater
      Treatment and Waste Disposal
        What are the common treatment and disposal methods for
        hydraulic fracturing wastewater, and where are these
        methods practiced?
        How effective are conventional publicly owned treatment
        works and commercial treatment systems in removing
        organic and inorganic contaminants of concern in hydraulic
        fracturing wastewater?
        What are the potential  impacts from surface water disposal
        of treated hydraulic fracturing wastewater on drinking
        water treatment facilities?
      Chapters - Hazard Evaluation
      of Chemicals Across the Water
      Cycle Stages
     •   What are the toxicological properties of hydraulic
        fracturing chemical additives?
     •   What are the toxicological properties of hydraulic
        fracturing flowback and produced water constituents?
      1.5.  Intended Use
 1    We expect that this report, as a synthesis of the science, will contribute to the understanding of the
 2    potential impacts of hydraulic fracturing on drinking water resources and the factors that may
 3    influence those impacts. The data and findings in this report can be used by federal, tribal, state,
 4    and local officials; industry; and the public to better understand and address any vulnerabilities of
 5    drinking water resources to hydraulic fracturing activities.

 6    We expect this report will be used to help facilitate and inform dialogue among interested
 7    stakeholders, including Congress, other federal agencies, states, tribal governments, the
 8    international community, industry, NGOs, academia, and the general public. Additionally, the
 9    identification of knowledge gaps will promote greater attention to these areas by researchers.

10    We also expect this report may support future assessment efforts. For instance, we anticipate that it
11    could contribute context to site-specific exposure or risk assessments of hydraulic fracturing, to
12    regional public health assessments, or to assessments of cumulative impacts of hydraulic fracturing
13    on drinking water resources over time or over defined geographic areas of interest

14    Finally, and most importantly, this assessment advances the scientific basis for decisions by federal,
15    state, tribal, and local officials; industry; and the public on how best to protect drinking water
16    resources now and in the future.
                    This document is a draft for review purposes only and does not constitute Agency policy.
      June 2015                                     1-11                  DRAFT—DO NOT CITE OR QUOTE

-------
Hydraulic Fracturing Drinking Water Assessment                                        Chapter 1 - Introduction
1.6.  References for Chapter 1

H.R. Rep. 111-316. Department of the Interior. Environment, and Related Agencies Appropriation Act. 2010:
   Conference report (to accompany H.R. 2996), (2009). http://www.gpo.gov/fdsys/pkg/CRPT-
   lllhrpt316/pdf/CRPT-lllhrpt316.pdf

OMB (U.S. Office of Management and Budget). (2004). Final information quality bulletin for peer review.
   Washington, DC: US Office of Management and Budget (OMB).
   http://www.whitehouse.gov/sites/default/files/omb/assets/omb/memoranda/fy2005/m05-03.pdf

U.S. EPA (U.S. Environmental Protection Agency). (2003). A summary of general assessment factors for
   evaluating the quality of scientific and technical information [EPA Report]. (EPA/100/B-03/001).
   Washington, DC: U.S. Environmental Protection Agency, Office of Research and Development.
   http://www.epa.gov/spc/assess.htm

U.S. EPA (U.S. Environmental Protection Agency). (2011c). Plan to study the potential impacts of hydraulic
   fracturing on drinking water resources [EPA Report]. (EPA/600/R-11/122). Washington, DC: U.S.
   Environmental Protection Agency, Office of Research and Development.
   http://www2.epa.gov/hfstudy/plan-study-potential-impacts-hydraulic-fracturing-drinking-water-
   resources-epa600r-11122

U.S. EPA (U.S. Environmental Protection Agency). (2012f). Study of the potential impacts of hydraulic
   fracturing on drinking water resources: Progress report. (EPA/601/R-12/011). Washington, DC: U.S.
   Environmental Protection Agency, Office of Research and Development.
   http://nepis.epa.gov/exe/ZyPURL.cgi?Dockey=P100FH8M.txt

U.S. EPA (U.S. Environmental Protection Agency). (2013d). Supplemental programmatic quality assurance
   project plan for work assignment 5-83 technical support for the hydraulic fracturing drinking water
   assessment. Washington, D.C. http://www2.epa.gov/sites/production/files/documents/literature-
   review-qappl.pdf

U.S. EPA (U.S. Environmental Protection Agency). (2014g). Quality assurance project plan - Revision no. 2:
   Data and literature evaluation for the EPA's study of the potential impacts of hydraulic fracturing (HF) on
   drinking water resources [EPA Report]. Washington, D.C.

U.S. EPA (U.S. Environmental Protection Agency). (2014h). Quality management plan- Revision no. 2: Study of
   the potential impacts of hydraulic fracturing for oil and gas on drinking water resources [EPA Report].
   Washington, D.C. http://www2.epa.gov/hfstudy/quality-management-plan-revision-no-2-study-
   potential-impacts-hvdraulic-fracturing-oil-and
               This document is a draft for review purposes only and does not constitute Agency policy.

June 2015                                       1-12                    DRAFT—DO NOT CITE OR QUOTE

-------
Hydraulic Fracturing Drinking Water AssessmentChapter 2-Hydraulic Fracturing, OH and Gas Production, andthe U.S. Energy Sector
                          Chapter 2
Hydraulic Fracturing, Oil and Gas Production, and
the U.S. Energy Sector
          This document is a draft for review purposes only and does not constitute Agency policy.

June 2015                                           DRAFT—DO NOT CITE OR QUOTE

-------
      Hydraulic Fracturing Drinking Water AssessmentChapter 2-Hydraulic Fracturing, OH and Gas Production, andthe U.S. Energy Sector


      2. Hydraulic Fracturing,  Oil and Gas Production,  and the
          U.S.  Energy Sector
 1    This chapter provides general background information useful for understanding the in-depth
 2    technical chapters that follow. We describe the process and purpose of hydraulic fracturing and the
 3    situations and settings in which it is used (Section 2.1). Then, to place hydraulic fracturing in the
 4    context of well site operations, we describe activities from site assessment and selection through
 5    production to site closure. This helps illustrate the intensive nature of activities during the
 6    relatively short hydraulic fracturing phase during the life of a production well (Section 2.2). Finally,
 7    we characterize the prevalence of hydraulic fracturing in the United States, its importance in the oil
 8    and gas industry today and into the future, and its role in the U.S. energy sector (Sections 2.3 and
 9    2.4).

      2.1. What is Hydraulic Fracturing?
10    Hydraulic fracturing is a stimulation technique used to increase production of oil and gas. Hydraulic
11    fracturing involves the injection of fluids under pressures great enough to fracture the oil- andgas-
12    production formations. Hydraulic fracturing fluid transfers the pressure generated by equipment at
13    the surface into the subsurface to create fractures, and it carries and places the proppant into the
14    fractures so that they remain "propped" open after the injection pumping pressure is terminated
15    (Gupta and Valko. 2007). Oil and gas can then flow through the fractures into the well and through
16    the well to the surface. Hydraulic fracturing has been used since the late 1940s and for the first
17    almost 5 0 years was used in vertical wells in conventional hydrocarbon reservoirs.x Hydraulic
18    fracturing is still used in these settings, but the process has evolved; technological developments
19    have led to the use of hydraulic fracturing in low-permeability (unconventional) hydrocarbon
20    reservoirs that could not otherwise be profitably produced (see Text Box 2-1). Wells stimulated by
21    hydraulic fracturing may be vertical, deviated, or horizontal in orientation (see Figure 2-1), and
22    they may be newly drilled or older at the time the fracturing is done.

23
      1A conventional reservoir is a reservoir in which buoyant forces keep hydrocarbons in place below a sealing caprock.
      Reservoir and fluid characteristics of conventional reservoirs typically permit oil or natural gas to flow readily into
      wellbores. The term is used to make a distinction from shale and other unconventional reservoirs, in which gas might be
      distributed throughout the reservoir at the basin scale, and in which buoyant forces or the influence of a water column on
      the location of hydrocarbons within the reservoir are not significant.
                    This document is a draft for review purposes only and does not constitute Agency policy.
      June 2015                                     2-1                   DRAFT—DO NOT CITE OR QUOTE

-------
      Hydraulic Fracturing Drinking Water AssessmentChapter 2-Hydraulic Fracturing, OH and Gas Production, andthe U.S. Energy Sector
      Text Box 2-1. Is Hydraulic Fracturing "New"?

 1    Hydraulic fracturing in one form or another has been in use since the late 1940s, when a fracturing technique
 2    was patented by the Stanolind Oil and Gas Company and licensed to the Halliburton Oil Well Cementing
 3    Company. There are precedents that go back even further: reports from the early days of the oil and gas
 4    industry in the mid-19th century show producers trying to increase production by pumping fluids or
 5    dropping explosives into wells (Montgomery and Smith. 20101. Throughout its history, hydraulic fracturing
 6    has been used as a production technique to increase, or "stimulate," production from a well (some hydraulic
 7    fracturing methods are used to stimulate production in water wells, which is outside the scope of this report).

 8    The groundwork for the transformation to modern hydraulic fracturing was laid in the 1970s and early
 9    1980s, when a coalition of private companies, government agencies, and industry groups began sponsoring
10    research into shale gas development technologies. During that period, Congress began to offer tax incentives
11    to induce producers to apply the developing technologies in the field [Wang and Krupnick. 2013: EIA. 2011a:
12    Yergin. 2011]. The first horizontal wells were drilled in the mid-1980s in the Austin Chalk oil-bearing
13    formation in Texas (Pearson. 2011: Haymond. 19911. Directional drilling and other emerging technologies
14    matured in the late 1990s. In 2001, the Mitchell Energy company found a way to economically fracture the
15    Barnett Shale in Texas. The company was bought by Devon Energy, a company with advanced experience in
16    horizontal drilling. In 2002, seven wells were drilled and developed in the Barnett Shale using both horizontal
17    drilling and hydraulic fracturing. Fifty-five more wells were completed in 2003 [Yergin. 2011]. The
18    techniques were rapidly adopted and further developed by others. By 2003/2004, modern hydraulic
19    fracturing in the Barnett Shale was producing more gas than all other shale gas wells in the rest of the country
20    (mostly shallow shale gas production in the Appalachian and Michigan Basins, see Section 2.4.1) (DOE,
21    2011b: Montgomery and Smith. 2010). By 2005, the new techniques were being used in low-permeability
22    hydrocarbon plays outside of Texas, and modern hydraulic fracturing soon became the industry standard,
23    driving the surge in U.S. production of natural gas.

24    Despite the long history of hydraulic fracturing, the culmination of technical innovations in the early 2000s
25    represent an appreciable change. These innovations have made hydraulic fracturing economical enough to
26    become standard practice in the oil and gas industry. Modern hydraulic fracturing (sometimes referred to as
27    high-volume hydraulic fracturing) is characterized by the use of long horizontal wells and higher volumes of
28    more complex mixtures of water, proppants, and chemical additives for injection as compared to earlier
29    fracturing practices. Wells are often deep and long: shale gas production wells are commonly 5,000 to 13,500
30    ft (1,524 to 4,115 m) deep with long horizontal sections of 2,000 to 5,000 ft (610 to 1,524 m) or more in
31    length. Other important advances occurred in oil and gas geophysical survey techniques (such as downhole
32    telemetry and 3D seismic imaging)  (Wang and Krupnick. 2013: EIA. 2011a). Hydraulic fracturing continues to
33    be conducted in vertical production wells as well as conventional reservoirs using some of these newer
34    techniques. Modern hydraulic fracturing has made it possible to extract resources in previously untapped
3 5    hydrocarbon-bearing geologic settings, altering and expanding the geographic range of oil and gas production
36    activities.
                     This document is a draft for review purposes only and does not constitute Agency policy.
      June 2015                                       2-2                    DRAFT—DO NOT CITE OR QUOTE

-------
      Hydraulic Fracturing Drinking Water AssessmentChapter 2-Hydraulic Fracturing, OH and Gas Production, andthe U.S. Energy Sector
      Figure 2-1. Schematic cross-section of general types of oil and gas resources and the
                orientations of production wells used in hydraulic fracturing.
                Shown are conceptual illustrations of types of oil and gas wells. A vertical well is producing from a
                conventional oil and gas deposit (right). In this case, a thin, gray confining layer serves to "trap" oil
                (green) or gas (red). Also shown are wells producing from unconventional formations: a vertical
                coalbed methane well (second from right); a horizontal well producing from a shale formation
                (center); and a deviated well producing from a tight sand formation (left). Note: Figure not to scale.
                Modified from USGS (2002) and Newell (2011).

 1    Historically, oil and gas have been extracted from conventional reservoirs that develop when
 2    hydrocarbons formed in deeper geologic source formations migrate until they accumulate
 3    underneath an impermeable layer (see Figure 2-1). Extraction practices vary. In settings where a
 4    reservoir is permeable enough and under enough pressure to yield a relatively high rate of
 5    hydrocarbon flow into a well, the economic extraction of oil and/or gas may be as simple as using a
 6    drilled well to enable hydrocarbons to flow to the surface under the natural pressure of the
 7    reservoir. In other cases, producers may inject water and/or carbon dioxide under pressure into
 8    the reservoir via one or more nearby wells to help move and enhance production of the oil and gas.
 9    But essentially, producers are drawing on hydrocarbons that have already accumulated in a
10    relatively accessible form.

11    Hydraulic fracturing is one of several methods used to enhance production from oil and gas
12    reservoirs. It is distinct from other methods of hydrocarbon extraction (known generally as
13    enhanced recovery techniques) that involve injecting fluids to influence either reservoir pressure,
14    fluid viscosity, or both. The primary purpose of hydraulic fracturing is to increase the surface area
15    of the reservoir rock by creating fractures that are propped open, allowing the hydrocarbon to  flow
16    from the rock through the fractures to the well and through the well up to the surface.
                    This document is a draft for review purposes only and does not constitute Agency policy.
      June 2015                                     2-3                    DRAFT—DO NOT CITE OR QUOTE

-------
      Hydraulic Fracturing Drinking Water AssessmentChapter 2-Hydraulic Fracturing, OH and Gas Production, andthe U.S. Energy Sector


 1    Hydraulic fracturing, in conjunction with horizontal and directional drilling, has made it possible to
 2    economically extract oil and gas from "unconventional" geologic formations (see Text Box 2-2),
 3    such as the relatively low permeability shales in which oil and gas form (see Figure 2-1). With
 4    modern horizontal drilling techniques, producers can, for example, drill a single well that follows
 5    the contours of a relatively thin, horizontal shale formation. Such drilling allows fracturing to be
 6    conducted in a long horizontal section of the well that accesses an extensive portion of the oil-  or
 7    gas-bearing formation. Unconventional formations include:

 8       •   Shales. Organic-rich black shales are the source rocks in which oil and gas form on geologic
 9           timescales. Shales have very low permeability, and the hydrocarbons are contained in the
10           pore space in the shales. Some shales produce predominantly gas and others predominantly
11           oil;  often there will be some coproduction of gas from oil wells and coproduction of liquid
12           hydrocarbons from gas wells (USGS. 2013a: EIA. 2011a).
13       •   Tight formations. "Tight" sands (sandstones), siltstone, carbonates, etc., are  relatively low
14           permeability, non-shale, sedimentary formations that can contain hydrocarbons. The
15           hydrocarbons are contained in the pore space of the formations. There is a continuum in
16           permeability between "tight" formations which require hydraulic fracturing to be produced
17           economically and sandstone (and other) formations that do not In the literature, "tight gas"
18           is generally distinguished from "shale gas," while oil resources from shale and tight
19           formations are frequently lumped together under the label "shale oil" or "tight oil"
20           fSchlumberger. 2014: USGS. 2014al
21       •   Coalbeds. Hydraulic fracturing can be used to extract methane (the primary component of
22           natural gas) from coal seams. In coalbeds, the methane is adsorbed to the coal surface
23           rather than contained in pore space or structurally trapped in the formation.  Pumping the
24           injected and formation water out of the coalbeds after fracturing serves to depressurize the
25           coal, thereby allowing the methane to desorb and flow into the well and to the surface
26           (USGS. 2000).
      Text Box 2-2. "Conventional" Versus "Unconventional."

27    The terms "conventional" and "unconventional" are widely used in the literature to distinguish types of oil
28    and gas reservoirs, plays, wells, production techniques, and more. In this report, the terms are used to
29    distinguish different types of hydrocarbon resources: "conventional" resources are those that can
30    economically be extracted using long-established technologies, and "unconventional" resources are those
31    whose extraction has become economical only with the advances that have occurred in modern hydraulic
32    fracturing (often coupled with directional drilling) in recent years.

33    Note that as modern hydraulic fracturing has become industry standard, the word "unconventional" is less
34    apt than it once was to describe these resources. In a sense, "the unconventional has become the new
35    conventional" (NETL, 2013).
36    Although the goal of stimulation by hydraulic fracturing is the same wherever it is employed, the
37    way it is accomplished varies due to a number of factors. General location and geologic conditions,


                    This document is a draft for review purposes only and does not constitute Agency policy.
      June 2015                                     2-4                   DRAFT—DO NOT CITE OR QUOTE

-------
      Hydraulic Fracturing Drinking Water AssessmentChapter 2-Hydraulic Fracturing, OH and Gas Production, andthe U.S. Energy Sector


 1    whether the well is existing or newly drilled, the proximity of the well to infrastructure and raw
 2    materials, operator preferences, and other factors can affect how a hydraulic fracturing operation is
 3    designed and carried out Technological advances have made it possible to drill deeper and longer
 4    horizontal wells, to conduct fracturing through longer portions of the well, and to place multiple
 5    wells on a single well pad [NETL. 2013: Montgomery and Smith. 2010]. Many facets of hydraulic
 6    fracturing-related technology have changed since they were first pioneered (see Text Box 2-1). How
 7    hydraulic fracturing  is practiced now (especially in the long horizontal wells) is different from how
 8    it was conducted during the first decades of its use. As operators gain experience with both
 9    evolving and new technologies, practices will continue to change.

10    The following three maps show the locations of major shale oil and gas resources, tight gas
11    resources, and coalbed methane resources, respectively, in the continental United States (see
12    Figure 2-2, Figure 2-3, and Figure 2-4). These maps represent resources that are being exploited
13    now or could be exploited in the future. Hydraulic fracturing continues to be used to enhance
14    production in conventional reservoirs  (not shown), although it is uncertain how often this occurs.

15    The formations hydraulically fractured for gas or oil vary in their depth below the surface. For
16    example, the Marcellus Shale (found primarily in Pennsylvania, New York, and West Virginia) is
17    found at depths  of 4,000 to 8,500 ft (1,200 to 2,600 m), the Barnett Shale (Texas) is found at depths
18    of 6,500 to 8,500 ft (2,000 to 2,600 m), andthe Haynesville-Bossier Shale (Louisiana and Texas) is
19    found at depths  of 10,500 to 13,500 ft  (3,200 to 4,100 m) fNETL. 20131 These represent some of
20    the largest gas-producing shale formations or shale plays. However, some other plays are
21    shallower. Parts of the Antrim (Michigan), Fayetteville (Arkansas), and New Albany (Indiana and
22    Kentucky) shale plays, for example, are less than 2,000 ft (600 m) deep (NETL. 2013: GWPC and
23    ALL Consulting.  2009). Exploitation of thin coal seams often takes place close to the surface as well.
24    In the San Juan Basin (New Mexico), coal seams are 550 to 4,000 ft (170 to 1,200  m) deep; in the
25    Powder River Basin (Wyoming and Montana) they are 450 to greater than 6,500 ft  (140 to 2,000 m)
26    deep, and in the  Black Warrior Basin (Alabama and Mississippi) depths can range from the ground
27    surface to 3,500 ft (1,100 m) (ALL Consulting. 2004). See Chapter 6 for more information on the
28    depths of these formations and plays.
                   This document is a draft for review purposes only and does not constitute Agency policy.
      June 2015                                    2-5                   DRAFT—DO NOT CITE OR QUOTE

-------
Hydraulic Fracturing Drinking Water AssessmentChapter 2-Hydraulic Fracturing, OH and Gas Production, andthe U.S. Energy Sector
                                Lower 48  states shale plays
                                        fcSS±»tioR^
                                    BMHI

                                                                     Conasauge


                                                           Fbl« *  "ftis*
                                  _   .      fcwn    Haynesville-
                                             B,Iim-n   Bossier TS-L*-*-'S s^i Ba»m
Figure 2-2. Shale gas and oil plays in the lower 48 United States.

           Source: EIA(2015b).
               This document is a draft for review purposes only and does not constitute Agency policy.

June 2015                                        2-6                     DRAFT—DO NOT CITE OR QUOTE

-------
Hydraulic Fracturing Drinking Water AssessmentChapter 2-Hydraulic Fracturing, OH and Gas Production, andthe U.S. Energy Sector
                                         Major Tight Gas Plays, Lower 48 States
Figure 2-3. Tight gas plays in the lower 48 United States.

           Source: EIA(2011b).
                                    Coalbed Methane Fields, Lower 48 States
                 «Coo»Ba>
                  flew
                                                                                     if
                                                &
                          W*:     ^
                          "'BUCK MtM^  i $3™ ^°a^
                                   i Basin
                         •Irwtt I
                  V    VBnn V

    1^-SW CO«M•."

iPi  *•!«!,; zi
                                         r«,-x
        Source Energy I
        Updated April 8. 200&
                               data from USGS and vanous pubfestied s
Figure 2-4. Coalbed methane fields in the lower 48 United States.

           Source: EIA(2011b).
               This document is a draft for review purposes only and does not constitute Agency policy.

June 2015                                          2-7                     DRAFT—DO NOT CITE OR QUOTE

-------
      Hydraulic Fracturing Drinking Water AssessmentChapter 2-Hydraulic Fracturing, OH and Gas Production, andthe U.S. Energy Sector


      2.1.   Hydraulic Fracturing and the Life of a Well
 1    Hydraulic fracturing itself is a relatively short-term process, with the timeframe for a typical
 2    fracturing treatment being two to 10 days during which fluids are injected into the well to fracture
 3    the oil- and gas-bearing geologic formations [Halliburton. 2013: NYSDEC. 2011). However, it is a
 4    period of intense activity— the most activity that takes place at a well site during its existence.

 5    In this section, we briefly describe some of the supporting and ancillary activities that take place at
 6    the well site, from initial site development through production and ultimately to closure (see Figure
 7    2-5). This time period likely ranges from years to decades, depending on factors such as rate of
 8    depletion of the oil or gas, cost of production, and the price of oil and gas. The rate of oil and gas
 9    depletion in the reservoir is somewhat uncertain in unconventional formations because there is
10    relatively little history on which to base predictions.

11    The overview of well operations presented in this section is broad and is provided to illustrate
12    common activities and describe some specific operational details. The details of well preparation,
13    operations, and closure vary from company to company, from play to play, from jurisdiction to
14    jurisdiction, and from well to well. The various activities involved in well development and
15    operations can be conducted by the well owner  and/or operator, owner/operator representatives,
16    service companies, or other third parties contractors working for the well owner.
3 OUT years
Site assessment
and development
• Surveys, core
sampling, and
test well logging
• Permitting
• Road, well pad,
and pit
construction

==^
Well drilling and
construction
• Drillingand
installation of
casing and
cement







"-- 	 	 -_.
Hydraulic fracturing
• Setup of water tanks,
pumps, blenders,
command center, and
other equipment on site
• Transport of water,
chemicals, and
proppantto site
• Well preparation,
including perforation, if
necessary
• Injection of fracturing
fluids in multiple stages
• Removal of equipment
Fluid recovery and Oil and Gas
management Production
• Recovery, storage, • Capture of oil
treatment, and/or gas
and/or reuse, or
disposal of
wastewater

Site and well
closure
• Well plugging
• Site
reclamation







      Figure 2-5. Generalized timeline and summary of activities that take place during the
                operational phases of an oil or gas well site operation in which hydraulic fracturing
                is used.
                Relative duration of phases is approximate.
                    This document is a draft for review purposes only and does not constitute Agency policy.
      June 2015                                     2-8                   DRAFT—DO NOT CITE OR QUOTE

-------
      Hydraulic Fracturing Drinking Water AssessmentChapter 2-Hydraulic Fracturing, OH and Gas Production, andthe U.S. Energy Sector


      2.1.1.  Site and Well Development
 1    Numerous activities occur to assess and develop the site and to drill and construct the production
 2    well before hydraulic fracturing and production can occur.

      2.1.1.1. Site Assessment and Development
 3    Identifying a geologically suitable well site requires integrating data from geophysical surveys
 4    (including seismic surveys) that help to delineate subsurface features with other geologic
 5    information from rock core samples. Cores may be obtained while drilling exploratory wells or test
 6    holes. Core samples provide firsthand information on the characteristics of the oil- or gas-bearing
 7    formation, such as porosity, permeability, and details about the quantities and qualities of the
 8    hydrocarbon resource. Drilling rates and drill cuttings help identify the strata being drilled through
 9    and can help confirm and correlate stratigraphy and formation depths, including the depths of
10    water-bearing formations.1 Well logging (also known as wireline logging) is especially useful
11    combined with core analysis for understanding the properties of formations (Kundert and Mullen,
12    2009).2

13    Logistical factors involved in the selection of the well drilling site include topography; proximity to
14    facilities such as roads, pipelines, and water sources; well spacing considerations; well setback
15    requirements; potential for site erosion; location relative to environmentally sensitive areas; and
16    proximity to populated areas (Drohan and Brittingham. 2012: Arthur etal.. 2009a). Before
17    developing the site and initiating well drilling, the oil and gas company (or their representative)
18    obtains a mineral rights lease, negotiates with landowners, and applies for a drilling permit from
19    the appropriate state and local authorities. During the project, leases and permissions are also
20    needed for other activities including performing seismic surveys and drilling exploratory holes
21    (Hyne. 2012). This initial site assessment phase of the process may take several months (King.
22    2012).

23    Site preparation is necessary to enable equipment and supplies to reach the well area. Typically, the
24    site is surveyed first, and then an access road may need to be built to accommodate truck traffic
25    (Hyne. 2012). The operator then levels and grades the site to manage drainage and to allow
2 6    equipment to be hauled to and placed on site. Next, the operator may excavate and grade several
27    impoundments or storage pits near the well pad. In some cases, steel tanks may be used to hold
2 8    fluids instead of, or in addition to, pits. The pits may hold water intended for drilling fluids,
29    materials generated during drilling such as used drilling mud and drill cuttings, or the flowback and
30    produced waters after fracturing (Hyne. 2012). Pit construction is generally governed by local
31    regulations; federal regulations may also apply on federal and Indian Country. In some areas,
32    regulations may require pits to be lined to prevent fluid seepage into the shallow subsurface or may
      1 Drill cuttings are ground rock produced by the drilling process.
      2 Well logging consists of a continuous measurement of physical properties in or around the well with electrically
      powered instruments to infer formation properties. Measurements may include electrical properties (resistivity and
      conductivity), sonic properties, active and passive nuclear measurements, measurements of the wellbore, pressure
      measurement, formation fluid sampling, sidewall coring tools and others. Measurements may be taken via a wireline,
      which is a wire or cable that is used to deploy tools and instruments downhole and that transmits data to the surface.
                    This document is a draft for review purposes only and does not constitute Agency policy.
      June 2015                                     2-9                    DRAFT—DO NOT CITE OR QUOTE

-------
      Hydraulic Fracturing Drinking Water AssessmentChapter 2-Hydraulic Fracturing, OH and Gas Production, andthe U.S. Energy Sector


 1    prohibit pits altogether. Some sites have piping along the surface of the well pad or in the shallow
 2    subsurface that delivers water used for hydraulic fracturing, removes flowback and produced
 3    water, or transports the oil and gas once production begins [Arthur etal.. 2009a).

 4    After site and well pad preparation, drill rigs and associated equipment (e.g., the drill rig platform,
 5    drilling mud system components, generators, chemical storage tanks, blowout preventer, fuel
 6    storage tanks, cement pumps, drill pipe, and casing) are moved on and off the pad at the different
 7    stages of well drilling and completion. During drilling and completion, well pads can range in size
 8    from less than an acre to several acres depending on the scope of the operations [King. 2012:
 9    NYSDEC. 20111.

10    Well Drilling and Construction

11    Construction of the production well involves the drilling of the hole (or wellbore), along with the
12    installation and cementing of a series of casing strings to support the wellbore and isolate and
13    protect both the hydrocarbons being produced and any water-bearing zones through which the
14    well passes.1 In certain settings, some portions  of the well can be completed as open holes.2 Details
15    on these and other well construction activities are presented in Chapter 6 and Appendix D.

16    The operator begins drilling by lowering and rotating the drill string, which consists of the drill bit,
17    drill pipe (see  Figure 2-6), and drill collars (heavy pieces of pipe that add weight to the bit). The
18    drill pipe attaches to the drill bit, rotating and advancing the bit; as drilling advances, new sections
19    of pipe are added at the surface, enabling the drilling to proceed deeper (Hyne. 2012). A drilling
20    fluid is circulated during drilling.3 The drilling fluid, which may be water-based or oil-based, is
21    pumped down to the drill bit, where it cools and lubricates the drill bit, counterbalances downhole
22    pressures, and lifts the drill cuttings to the surface (King. 2012).

23    Although all wells are initially drilled vertically, finished well orientations include vertical, deviated,
24    and horizontal. The operator selects the well orientation that will provide access to the targeted
25    zone(s) within a formation and that will align the well with existing fractures and other geologic
26    structures to optimize production. Deviated wells may be "S" shaped or continuously slanted.
27    Horizontal wells have lateral sections oriented approximately 90 degrees from the vertical portion
28    of the well. In wells completed horizontally, the lengths of these laterals can range from 2,000 to
29    5,000 ft (610 to 1,524 m) or more (Hyne. 2012: Miskimins. 2008: Bosworthetal.. 1998).4
30    Horizontal wells are instrumental in accessing productive areas of thin and laterally extensive oil-
31    and gas-bearing shales. Although the portion of hydraulically fractured wells that are horizontal is
32    growing, in some areas, such as California, hydraulic fracturing is still primarily conducted in
33    vertical wells fCCST. 20151.
      1 Casing is steel pipe that is lowered into a wellbore. Casing extends from the bottom of the hole to the surface.
      2 An open hole completion is a well completion that has no casing or liner set across the reservoir formation, allowing the
      produced fluids to flow directly into the wellbore.
      3 Drilling fluid is any of a number of liquid and gaseous fluids and mixtures of fluids and solids (as solid suspensions,
      mixtures, and emulsions of liquids, gases, and solids) used when drilling boreholes [Schlumberger. 2014).
      4 A lateral is a horizontal section of a well.
                    This document is a draft for review purposes only and does not constitute Agency policy.
      June 2015                                     2-10                   DRAFT—DO NOT CITE OR QUOTE

-------
      Hydraulic Fracturing Drinking Water AssessmentChapter 2-Hydraulic Fracturing, OH and Gas Production, andthe U.S. Energy Sector
      Figure 2-6. Pulling drill pipe onto the drilling platform.
                Source: Joshua Doubek, Wikicommons, CC-BY-SA-3.0.

 1    The drilling and well construction proceeds with repeated steps (the drill string is lowered, rotated,
 2    drilled to a certain depth, pulled out, and then the casing is lowered into the hole, set, and
 3    cemented). Successively smaller diameters of casing are used as the hole is drilled deeper (see
 4    Figure 2-7). Selection and installation of the casing strings is important for several purposes,
 5    including isolating hydrocarbon reservoirs from nearby aquifers, isolating over-pressured zones,
 6    and transporting hydrocarbons to the surface (Hyne. 2012). Newly installed casing strings are
 7    cemented in place before drilling continues (or before the well is completed in the instance of the
 8    production casing). The cement protects the casing from corrosion by formation fluids, stabilizes
 9    the casing and the wellbore, and prevents fluid movement along the well between the outside of the
10    casing and wellbore (Renpu, 2011). The well can be cemented continuously from the surface down
11    to the production zone of the well. Partially cemented wells are also possible with, for example,
12    cement from the surface to some distance below the deepest fresh water-bearing formation and
13    perhaps cement across other deeper formations. Chapter 6 and Appendix D contain more details on
14    casing and cement.
                    This document is a draft for review purposes only and does not constitute Agency policy.
      June 2015                                     2-11                  DRAFT—DO NOT CITE OR QUOTE

-------
      Hydraulic Fracturing Drinking Water AssessmentChapter 2-Hydraulic Fracturing, OH and Gas Production, andthe U.S. Energy Sector
      Figure 2-7. Sections of surface casing lined up and being prepared for installation at a well site
                in Colorado.
                Photo credit: Gregory Oberley (U.S. EPA).

 1    When drilling, casing, and cementing are finished, the well can be completed in the production zone
 2    in several ways. The production casing may be cemented all the way through the production zone
 3    and perforated prior to hydraulic fracturing in the desired locations. Alternatively, operators may
 4    use an open hole completion, in which the casing is set just into the production zone and cemented.
 5    The remainder of the wellbore within the production zone is left open with no cement [Hyne,
 6    2012). Once all aspects of well construction are completed, the operator can remove the drilling rig,
 7    install the wellhead, and prepare the well for stimulation by hydraulic fracturing and subsequent
 8    production.

      2.1.1.  Hydraulic Fracturing
 9    Hydraulic fracturing is typically a short, intense, repetitive process requiring specialized equipment
10    and (for high volume horizontal wells)  large amounts of water, chemicals, and proppant. Machinery
11    and equipment are often brought to the site mounted on trucks and remain that way during use.
12    Tanks, totes, and other storage containers of various sizes holding water and chemicals are also
13    transported and installed on site.  Figure 2-8 shows a well pad prepared for hydraulic fracturing
14    with the necessary equipment and structures.
                   This document is a draft for review purposes only and does not constitute Agency policy.
      June 2015                                     2-12                  DRAFT—DO NOT CITE OR QUOTE

-------
      Hydraulic Fracturing Drinking Water AssessmentChapter 2-Hydraulic Fracturing, OH and Gas Production, andthe U.S. Energy Sector
      Figure 2-8. Hydraulic fracturing operation in Troy, PA.
                Site with all equipment on site in preparation for injection. Source: NYSDEC (2011).

      2.1.1.2. Injection Process
 1    Prior to injection, hydraulic fracturing fluids are mixed using specialized feeding and mixing
 2    equipment The mixing is generally performed mechanically on a truck-mounted blender and is
 3    electronically monitored and controlled by the operator in a separate van (see Chapter 5).
 4    Numerous hoses and pipes are used to transfer hydraulic fracturing fluid components from storage
 5    units to the mixing equipment and ultimately to the wellhead.

 6    A wellhead assembly is temporarily installed on the wellhead during the fracture treatment to
 7    allow high pressures and volumes of proppant-laden fluid to be injected into the well. Pressures
 8    required for fracturing can vary widely depending on depth, formation pressure, and rock type.
 9    Fracturing pressures have been reported ranging from 4,000 psi to 12,000 psi [Ciezobka and Salehi,
10    2013: Abou-Saved etal., 2011: Thompson, 2010]. The pressure during fracturing is measured using
11    pressure gauges, which can be installed at the surface and/or downhole [Ross and King. 2007).
12    Figure 2-9 shows two wellheads side-by-side being prepared for fracturing.
                    This document is a draft for review purposes only and does not constitute Agency policy.
      June 2015                                     2-13                  DRAFT—DO NOT CITE OR QUOTE

-------
     Hydraulic Fracturing Drinking Water AssessmentChapter 2-Hydraulic Fracturing, OH and Gas Production, andthe U.S. Energy Sector
     Figure 2-9. Two wellheads side-by-side being prepared for hydraulic fracturing at a well site in
               Pennsylvania.
               Photo credit: Mark Seltzer (U.S. EPA).

1    The entire length of the well in the production zone is not fractured all at once; instead, shorter
2    lengths or segments of the well in the production zone are isolated and fractured in "stages" [Lee et
3    al.. 2011). Each stage of a fracturing job can consist of phased injection of different fluids consisting
4    of varying components (i.e., chemicals and additives). These different fluids (1) remove excess
5    drilling fluid or cement from the formation (often using acid) (GWPC and ALL Consulting, 2009),
6    (2) initiate fractures ("pad fluid" without proppant), (3) carry the proppant (Hyne, 2012), and
7    (4) flush the wellbore to ensure that all proppant-laden fluids reach the fractures. Each phase
                   This document is a draft for review purposes only and does not constitute Agency policy.
     June 2015                                     2-14                   DRAFT—DO NOT CITE OR QUOTE

-------
      Hydraulic Fracturing Drinking Water AssessmentChapter 2-Hydraulic Fracturing, OH and Gas Production, andthe U.S. Energy Sector


 1    requires moving up to millions of gallons of fluids around the site through various hoses and lines,
 2    blending the fluids, and injecting them at high pressures down the well.

 3    The total number of stages depends on the formation properties and the orientation and length of
 4    the well. As technology has improved, the lengths of laterals in horizontal wells and the numbers of
 5    stages per well have tended to increase [NETL. 2013: Pearson etal.. 2013]. The number of stages
 6    per well can vary, with several sources suggesting that between 10 and 20 is typical [GNB, 2015:
 7    Lowe etal.. 2013). The full range reported in the literature is much wider, with one source
 8    documenting between 1 and 59 stages per well [Pearson et al., 2013] and others reporting values
 9    within this range [NETL. 2013: STO. 2013: Allison etal.. 20091. For more details on hydraulic
10    fracturing stages, see Chapter 5, Section 5.2.

11    The induced fractures are designed to achieve the optimum drainage of hydrocarbons from the
12    reservoir formations. Engineers can design fracture systems using modeling software that requires
13    a significant amount of data on formation permeability, porosity, in situ stress, mineralogy, and
14    geologic barrier locations, among other factors [Holditch. 2007). Microseismic monitoring during
15    fracturing can be used to characterize the horizontal and vertical extent of the fractures created and
16    assist with the design of future fracturing jobs [Cipollaetal., 2011]. Post-fracture monitoring of
17    pressure or tracers can also help characterize the results of a fracturing job. More details of
18    injection, fracturing, and related monitoring are provided in Chapter 6 and Appendix D.

      2.1.1.3. Fracturing Fluids
19    The fracturing fluids injected into the well serve a variety of purposes and require chemical
20    additives to perform properly (see Chapter 5, Section 5.3]. Depending on the geologic setting,
21    reservoir geochemistry, production type, proppant size, and other factors, operators typically
22    choose to use one of several common types of fracturing fluid systems [Arthur etal.. 2014:
23    Spellman,  2012: Gupta and Valko, 2007]. Water-based fracturing fluids are the most common, but
24    other fluid types can be used such as: foams or emulsions made  with nitrogen, carbon dioxide, or
25    hydrocarbons; acid-based fluids; and others [Montgomery, 2013: Sabaetal., 2012: Gupta and
2 6    Hlidek. 2009: Gupta and Valko. 2007: Halliburton. 19881 The most common water-based fluid
2 7    systems are slickwater formulations, which are typically used in very low permeability reservoirs,
2 8    and gelled fracturing fluids, which can be used in reservoirs with higher permeability [Barati and
29    Liang. 2014).1-2 More details of hydraulic fracturing fluid systems are discussed in Section 5.3.
30    Importantly, chemical usage in the industry is continually changing as processes are tested and
31    refined by companies. Shifts in fluid formulations are driven by economics, technological
32    developments, and concerns about environmental and health impacts.
      1 Slickwater is a type of fracturing fluid that consists mainly of water with a very low portion of additives like polymers
      that serve as friction reducers to reduce friction loss when pumping the fracturing fluid downhole [Barati and Liang.
      2014).
      2 Gelled fluids are fracturing fluids that are usually water-based with added gels to increase the fluid viscosity to aid in the
      transport of proppants [Spellman. 2012: Gupta and Valko. 2007).
                    This document is a draft for review purposes only and does not constitute Agency policy.
      June 2015                                     2-15                   DRAFT—DO NOT CITE OR QUOTE

-------
      Hydraulic Fracturing Drinking Water AssessmentChapter 2-Hydraulic Fracturing, OH and Gas Production, andthe U.S. Energy Sector


 1    The largest constituent of a typical hydraulic fracturing fluid is water (see Figure 2-10). The water
 2    sources used for hydraulic fracturing base fluid include ground water, surface water, treated
 3    wastewater, and reused flowback or produced water from other wells [URS Corporation. 2011:
 4    Blauch, 2010: Kargbo etal., 2010].* The water may be brought to the production well site via trucks
 5    or piping, or it may be locally sourced (for example, pumped from a local river or obtained from a
 6    water well tapping local ground water). Selection of water sources depends upon availability, cost,
 7    quality of the water, and the logistics of delivering it to the site. Chapter 4 provides additional
 8    details on water acquisition and the amount of water used for hydraulic fracturing.
      Figure 2-10. Water tanks (blue, foreground) lined up for hydraulic fracturing at a well site in
                central Arkansas.
                Photo credit: Martha Roberts (U.S. EPA).

 9    Proppants are, by volume, second to the base fluid in the hydraulic fracturing fluid system. Silicate
10    minerals, most notably quartz sand, are the most commonly used proppants. Increasingly, silicate
11    proppants are being coated with resins that help prevent development and flowback of particles or
12    fragments of particles. Ceramic materials, such as those based on calcined (heated) bauxite or
13    calcined kaolin (mullite) are also used as proppants due to their high strength and resistance to
14    crushing and deformation (Beckwith. 2011).
      1 Base fluid is the fluid into which additives and proppants are mixed to formulate a hydraulic fracturing fluid.
                    This document is a draft for review purposes only and does not constitute Agency policy.
      June 2015                                     2-16                  DRAFT—DO NOT CITE OR QUOTE

-------
      Hydraulic Fracturing Drinking Water AssessmentChapter 2-Hydraulic Fracturing, OH and Gas Production, andthe U.S. Energy Sector


 1    Additives comprise relatively small percentages of hydraulic fracturing fluid systems, generally
 2    constituting <2.0% of the fluid [GWPC and ALL Consulting. 2009]. The EPA analyzed additive data
 3    in the EPA FracFocus project database 1.0 and estimated that hydraulic fracturing additives in 2Oil
 4    and 2012 totaled 0.43% of the total amount of fluid injected for hydraulic fracturing [U.S. EPA,
 5    2015a]. Note that this small percentage can total tens of thousands of gallons of chemical additives
 6    for a typical high-volume hydraulic fracturing job (see Chapter 5, Section 5.4 for details on additive
 7    volumes). A given  additive may consist of a single chemical ingredient, or it may have multiple
 8    ingredients. The mix of chemicals used in any particular fracturing job is influenced by the
 9    properties of the target formation, the amount and type of proppant that needs to be carried,
10    operator preference, and to some degree, by local or regional availability of chemicals and potential
11    interactions between chemicals [King, 2012]. Chapter 5 includes details on the number, types, and
12    estimated quantities of chemicals that can be used in hydraulic fracturing.

      2.1.2.  Fluid Recovery, Management, and Disposal
13    When the injection pressure is reduced atthe endofthe fracturing process, the direction of fluid
14    flow reverses, with some of the injected hydraulic fracturing fluid flowing into the well and to the
15    surface along with some naturally-occurring fluids from the production zone [NYSDEC, 2011]. The
16    fluid is  initially a portion of the injected fluid, which decreases over the first few weeks or months
17    until produced water originating from the fractured oil- or gas-bearing rock formation
18    predominates. This recovery of produced water continues over the life of the well [Barbotetal..
19    2013]. Chapter 7 presents descriptions and discussions of the composition and quantities of fluids
20    recovered at the well, referred to as flowback and produced water.

21    The hydraulic fracturing flowback and produced water (sometimes referred to as hydraulic
22    fracturing wastewater], as well as any other liquid waste from the well pad itself (e.g., rainwater
23    runoff), is typically stored on-site in impoundments (see Figure 2-11] or tanks. This wastewater can
24    be moved offsite via truck or pipelines. The majority of these hydraulic fracturing wastewaters
25    nationally are managed through disposal into deep Class II injection wells regulated under the
26    Underground Injection Control (UIC] program under the Safe Drinking Water Act (see Chapter  8].
27    Other management strategies include treatment followed by discharge to surface water bodies, or
28    reuse for subsequent fracturing operations either with or without treatment (U.S. EPA. 2012f: U.S.
29    GAP. 2012]. Decisions  regarding wastewater management are driven by factors such as cost
3 0    (including costs of storage and transportation], availability of facilities for treatment, reuse, or
31    disposal, and regulations (Rassenfoss. 2011]. Wastewater management is yet another aspect of
32    fracturing-related oil and gas production that is changing significantly. Chapter 8 contains details of
33    the treatment, reuse and recycling, and disposal of wastewater.
                   This document is a draft for review purposes only and does not constitute Agency policy.
      June 2015                                    2-17                   DRAFT—DO NOT CITE OR QUOTE

-------
      Hydraulic Fracturing Drinking Water AssessmentChapter 2-Hydraulic Fracturing, OH and Gas Production, andthe U.S. Energy Sector
      Figure 2-11. Impoundment on the site of a hydraulic fracturing operation in central Arkansas.
                Photo credit: Caroline E. Ridley (U.S. EPA).

      2.1.3.  Oil and Gas Production
 1    After hydraulic fracturing, equipment is removed and partial site reclamation may take place if
 2    drilling of additional wells or laterals is not planned [NYSDEC. 2011]. Operators may dewater, fill
 3    in, and regrade pits that are no longer needed. Parts of the pad may be reseeded, and the well pad
 4    may be reduced in size (e.g., from 3 to 5 acres (1 to 2 hectares) during the drilling and fracturing
 5    process to 1 to 3 acres (0.4 to 1 hectares) during production) (NYSDEC, 2011).

 6    Wells may be shut-in immediately after completion if there is no infrastructure to receive the
 7    product or if prices are unfavorable. Prior to bringing a well into production, the operator typically
 8    runs a production test to determine the maximum flow rate the well can sustain and to optimize
 9    equipment settings (Hyne. 2012: Schlumberger. 2006). Such tests may be repeated throughout the
10    life of the well. During production, monitoring (e.g., mechanical integrity testing, corrosion
11    monitoring), including any compliance with state monitoring requirements, may be conducted to
12    enable operators to be sure that the well is operating as intended.

13    In the case of gas wells, the produced gas typically flows through a flowline to a separator that
14    separates the gas from water or any liquid hydrocarbons (NYSDEC. 2011). The finished gas is sent
15    to a compressor station where it is compressed to pipeline pressure and sent to a pipeline for sale.
16    Production at oil wells proceeds similarly, although oil/water or oil/water/gas separation occurs
                   This document is a draft for review purposes only and does not constitute Agency policy.
      June 2015                                    2-18                  DRAFT—DO NOT CITE OR QUOTE

-------
      Hydraulic Fracturing Drinking Water AssessmentChapter 2-Hydraulic Fracturing, OH and Gas Production, andthe U.S. Energy Sector


 1    most typically on the well pad, no compressor is needed, and the oil can be hauled (by truck or
 2    train) or piped from the well pad.

 3    During the life of the well it may be necessary to perform workovers to maintain or repair portions
 4    or components of the well and replace old equipment Such workovers involve ceasing production
 5    and removing the wellhead, and may include cleaning out sand or deposits from the well, repairing
 6    casing, replacing worn well components such as tubing or packers, or installing or replacing lift
 7    equipment to pump hydrocarbons to the surface [Hyne. 2012). In some cases, wells may be
 8    recompleted after the initial construction, with re-fracturing if production has decreased [Vincent,
 9    2011]. Recompletion also may include additional perforations in the well at a different interval to
10    produce from a different formation than originally done, lengthening the wellbore, or drilling new
11    laterals from an  existing wellbore.

12    As of 2012, Shires andLev-On [2012] suggested that the rate of re-fracturing in natural gas wells
13    was about 1.6%. Analysis for the EPA's 2012 Oil and Gas Sector New Source Performance Standards
14    indicated a re-fracture rate of 1% for gas wells [U.S. EPA. 2012d]. In the EPA's Inventory of U.S.
15    Greenhouse Gas Emissions and Sinks [U.S. EPA. 2015g]. the number of gas wells that were re-
16    fractured in a given year as a percent of the total existing population of hydraulically fractured
17    producing gas wells in a given year ranges from 0.3% to 1% across the 1990-2013 period.

      2.1.1.4. Production Rates and Duration
18    The production life of a well depends on a number of factors, such as the amount of hydrocarbons
19    in place, the reservoir pressure, production rate, and the economics of well operations. It may be as
20    short as three or four years in deep-water, high-permeability formations and as long as 40 to 60
21    years in onshore tight gas reservoirs [Ross and King. 2007). In hydraulically fractured wells in
22    unconventional reservoirs, production is often characterized by a rapid drop followed by a slower
23    decline compared to conventional hydrocarbon production wells [Patzeketal., 2013]. However,
24    most modern, high-volume fractured wells are less than a decade old. Consequently, there is a
25    limited historical basis to determine the full extent of the production decline [Patzeketal., 2013]
2 6    and to ultimately determine how much they will produce.

      2.1.4.  Site and Well Closure
2 7    Once a well reaches the end of its useful life, it is plugged, and the well site is closed. If a wellbore is
2 8    not properly plugged, fluids from higher pressure zones may eventually migrate through the
29    wellbore to the surface or to other zones such as fresh water aquifers [NPC, 2011b]. Plugging is
30    usually performed according to state regulations governing the locations and materials for plugs
31    [Calvertand Smith, 1994]. Operators typically use cement plugs placed across fresh water
32    formations and oil or gas formations [NPC. 2011b]. Some surface structures can be left in place, and
3 3    the local topography and land cover are restored to predevelopment conditions to the extent
34    possible, per state regulations. The wellhead and any surface equipment are removed.
3 5    Impoundments are dewatered, filled in, and graded.  The well casing is typically cut off below the
36    surface and a steel plate or cap is emplaced to seal the top of the casing and wellbore [API, 2010a],
37    although there may also be an aboveground marker  used in some locations. Some states require
3 8    notification of the landowner or a government agency of the location of the well.

                   This document is a draft for review purposes only and does  not constitute Agency policy.
      June 2015                                    2-19                   DRAFT—DO NOT CITE OR QUOTE

-------
      Hydraulic Fracturing Drinking Water AssessmentChapter 2-Hydraulic Fracturing, OH and Gas Production, andthe U.S. Energy Sector


      2.2.   How Widespread is Hydraulic Fracturing?
 1    Hydraulic fracturing activity in the United States and worldwide is substantial. One industry
 2    cumulative estimate stated that by the time of writing in 2010, close to 2.5 million fracture
 3    treatments had been performed globally [Montgomery and Smith. 2010).  In 2002, the Interstate Oil
 4    and Gas Compact Commission (IOGCC) stated that close to 1 million wells had been hydraulically
 5    fractured in the United States since the 1940s [IOGCC. 20021. A recent U.S. Geological Survey
 6    (USGS) publication analyzed 1 million hydraulically fractured wells and 1.8 million hydraulic
 7    fracturing treatment records from the United States from 1947 to 2010 [USGS. 2015). Although
 8    some form of hydraulic fracturing has been used for more than 60 years, the technological
 9    advancements  that combined hydraulic fracturing and directional drilling in the early 2000s
10    resulted in the  new era of modern hydraulic fracturing, which uses higher volumes of fracturing
11    fluids than were typically used in prior decades. Modern hydraulic fracturing is typically associated
12    with horizontal wells producing from unconventional shale reservoirs, but hydraulic fracturing
13    continues to be done in vertical wells in conventional reservoirs also. This ongoing mix of
14    traditional and modern hydraulic fracturing activities makes estimates of the total number of
15    hydraulic fracturing wells challenging.

16    The following series of images illustrates hydraulic fracturing activities and the scale of those
17    activities in the United States. Figure 2-12 (taken in Springville Township, in northeastern
18    Pennsylvania)  and Figure 2-13 (taken near Williston, in northwestern North Dakota) show
19    individual well pads in the context of the local landscape. Landsat images  in Figure 2-14 and Figure
20    2-15 provide satellite views of areas in northwest Louisiana and southeast Wyoming, respectively,
21    where hydraulic fracturing activities currently occur as identified by the well pads in the images.
22    These images serve to illustrate activity at a wider scale, though they are not representative of all
23    hydraulic fracturing activities in the eastern or western United States. The light red circles around
24    some of the well pads identify them as hydraulic fracturing wells that were reported by well
25    operators to the FracFocus registry (as summarized in the EPA FracFocus project database 1.0)
26    (U.S. EPA. 2015b). (The FracFocus well locations reflect information in the EPA FracFocus project
27    database for well operations reporting hydraulic fracturing activities between January 2011 and
28    February 2013. The Landsat images are from a later period, July and August of 2014, so additional
29    well pads in the images now may be represented in the FracFocus registry.)
                   This document is a draft for review purposes only and does not constitute Agency policy.
      June 2015                                    2-20                  DRAFT—DO NOT CITE OR QUOTE

-------
Hydraulic Fracturing Drinking Water AssessmentChapter 2-Hydraulic Fracturing, OH and Gas Production, andthe U.S. Energy Sector
Figure 2-12. Aerial photograph of a well pad and service road in Springville Township,
          Pennsylvania.
          Image © T Henry Fair / Flights provided by LightHawk.
Figure 2-13. Aerial photograph of hydraulic fracturing activities near Williston, North Dakota.
          Image © T Henry Fair / Flights provided by LightHawk.
              This document /s a draft for review purposes only and does not constitute Agency policy.
June 2015                                      2-21                   DRAFT—DO NOT CITE OR QUOTE

-------
Hydraulic Fracturing Drinking Water AssessmentChapter 2-Hydraulic Fracturing, OH and Gas Production, andthe U.S. Energy Sector
                  Hydraulic fracturing
                ; wells reported to
                  FracFocus
Figure 2-14. Landsat photo showing hydraulic fracturing well sites near Frierson, Louisiana.

           Source: Imagery from USGS Earth Resources Observation and Science, Landsat 8 Operational Land
           Imager (scene LC80250382014232LGNOO) captured August 20, 2014 and accessed on May 1, 2015
           from USGS's EarthExplorer (http://earthexplorer.usgs.gov/).
           Inset imagery from USDA National Agriculture Imagery Program (entity M 3209351_NE
           15_1_20130703_20131107) captured July 3, 2013 and accessed May 1, 2015 from USGS's
           EarthExplorer (http://earthexplorer.usgs.gov/).
           FracFocus well locations are from the EPA FracFocus project database 1.0 (U.S. EPA, 2015b).
               This document is a draft for review purposes only and does not constitute Agency policy.
June 2015                                        2-22                    DRAFT—DO NOT CITE OR QUOTE

-------
     Hydraulic Fracturing Drinking Water AssessmentChapter 2-Hydraulic Fracturing, OH and Gas Production, andthe U.S. Energy Sector
                      Hydraulic fracturing
                    O wells reported to
                      FracFocus
                        1 Miles
                                  /
     Figure 2-15. Landsat photo showing hydraulic fracturing well sites near Pinedale, Wyoming.

                Source: Imagery from USGS Earth Resources Observation and Science, Landsat 8 Operational Land
                Imager (scene LC80370302014188LGNOO) captured July 7, 2014 and accessed May 1, 2015 from
                USGS's EarthExplorer (http://earthexplorer.usgs.gov/).
                Inset imagery from USDA National Agriculture Imagery Program (entity M 4210927_NW
                12_1_20120623_20121004) captured June 23, 2012 and accessed May 1, 2015 from USGS's
                EarthExplorer (http://earthexplorer.usgs.gov/).
                FracFocus well locations are from the EPA FracFocus project database 1.0 (U.S. EPA, 2015b).


1    The maps in Figure 2-16 show recent changes nationally in the geography of oil and gas production
2    through the increased use of horizontal drilling, which occurs together with hydraulic fracturing.
                    This document is a draft for review purposes only and does not constitute Agency policy.
     June 2015                                       2-23                    DRAFT—DO NOT CITE OR QUOTE

-------
     Hydraulic Fracturing Drinking Water AssessmentChapter 2-Hydraulic Fracturing, OH and Gas Production, andthe U.S. Energy Sector


1    Some traditional oil- and gas-producing parts of the country, such as Texas, have seen an expansion
2    of historically strong production activity as a result of the deployment of horizontal drilling and
3    modern hydraulic fracturing. Pennsylvania, a century ago one of the leading oil- and gas-producing
4    states, has seen a resurgence in oil and gas activity. Other states currently experiencing a steep
5    increase in production activity, such as North Dakota, Arkansas, and Montana, have historically
6    produced less oil and gas and are therefore undergoing new development.
                                                                       2000
                                                                       2005
                                                                       2012
                                                                     New well locations in
                                                                     select years (number
                                                                     of wells)
                                                                      •  2000    (341)
                                                                      •  2005   (1,809)
                                                                      •  2012  (14,560)
                                                                        EIA shale basins
     Figure 2-16. Location of horizontal wells that began producing oil or natural gas in 2000, 2005,
               and 2012, based on data from Drillinglnfo (2014a).
                   This document is a draft for review purposes only and does not constitute Agency policy.
     June 2015                                     2-24                   DRAFT—DO NOT CITE OR QUOTE

-------
      Hydraulic Fracturing Drinking Water AssessmentChapter 2-Hydraulic Fracturing, OH and Gas Production, andthe U.S. Energy Sector


      2.2.1.  Number of Wells Fractured per Year
 1    We estimate that from roughly 2011 to 2014, approximately 25,000 to 30,000 new oil and gas wells
 2    were hydraulically fractured each year. Additional, pre-existing wells (wells more than one year old
 3    that may or may not have been hydraulically fractured in the past) were also likely fractured each
 4    year. Since the early 2000s, the percentage of all hydraulically fractured wells that are either
 5    horizontal or deviated has steadily grown. Our estimates are based on data detailed below from
 6    several public and private sector organizations that track drilling and various aspects of hydraulic
 7    fracturing activity. There is no complete database or registry of wells that are hydraulically
 8    fractured in the United States. Another source of uncertainty is the rate at which relatively new
 9    hydraulic fracturing wells are re-fractured or the rate at which operators use older, existing wells
10    for hydraulic fracturing. Future trends in the number of wells hydraulically fractured per year will
11    be affected by the cost of well operation and the price of oil and gas. Scenarios of increasing, flat,
12    and decreasing hydraulic fracturing activity all appear to be possible [Weijermars. 2014).

13    The number of wells reported to the FracFocus registry provides a low estimate of the number of
14    hydraulically fractured wells.x As of early April 2015, the FracFocus registry reported receiving
15    information on a cumulative total of approximately 95,000 fracturing jobs, or roughly 22,400 per
16    year over the 51-month period from January 2011 through March 2015 [GWPC. 2015]. In a more
17    detailed review of FracFocus data from 2011 and 2012, the EPA found there were approximately
18    14,000 and 22,500 fracturing jobs reported to the FracFocus website in those years, respectively,
19    across 20 states fU.S. EPA. 2015a). These 2011 and 2012 numbers are likely underestimates of
2 0    wells hydraulically fractured annually, in part because FracFocus reporting was voluntary for most
21    states for at least a portion of 2011 to 2012  (though the increase from 2011 to 2012 in part reflects
22    more states requiring reporting to the registry). Hydraulic fracturing practices may alternately (or
23    in addition to FracFocus) be tracked by states. Compared to state records of hydraulic fracturing
24    from North Dakota, Pennsylvania, and West Virginia in 2011 and 2012, we found thatthe count of
25    wells based on records submitted to FracFocus was an underestimate of the number of fracturing
26    jobs in those states by an average of approximately 30% (see Text Box 4-1).

27    An additional estimate of the number of hydraulically fractured wells can be obtained from
28    Drillinglnfo, a commercial database compiling data from individual state oil and gas agencies
29    (Drillinglnfo. 2014a). The data indicate an increase in the number of new hydraulically fractured
30    wells drilled each year, from approximately 12,800 in 2000 to slightly more than 21,600 in 2005, to
31    nearly 23,000 in 2012. The number of new horizontal wells (which are likely all hydraulically
32    fractured) show a significant increase, from 344 (about 1% of all new production wells) in 2000, to
33    1,810 in 2005, to 14,560 (nearly 41% of all new production wells) in 2012 (see Figure 2-16).
      1 The FracFocus registry was developed by the Ground Water Protection Council and the Interstate Oil and Gas Compact
      Commission. Oil and gas well operators can use the FracFocus registry to disclose information about hydraulic fracturing
      well locations, and water and chemical use during hydraulic fracturing operations. Submission of information to
      FracFocus was initially voluntary (starting in January 2011), but now about half of the 20 states represented in FracFocus
      have enacted reporting requirements for well operators that either mandate reporting to FracFocus or allow it as one
      reporting option. FracFocus data are discussed in more detail in Chapter 4 (regarding water volumes) and Chapter 5
      (regarding chemical use). For more information see www.fracfocus.org and U.S. EPA(2015a).
                    This document is a draft for review purposes only and does not constitute Agency policy.
      June 2015                                    2-25                   DRAFT—DO NOT CITE OR QUOTE

-------
      Hydraulic Fracturing Drinking Water AssessmentChapter 2-Hydraulic Fracturing, OH and Gas Production, andthe U.S. Energy Sector


 1    Because Drillinglnfo data do not directly report whether a well has been hydraulically fractured, we
 2    relied on properties of the well and the oil or gas producing formation to infer which wells were
 3    hydraulically fractured and when. First, we assumed that all horizontal wells were hydraulically
 4    fractured in the year they started producing. Second, we assumed that all wells within a shale,
 5    coalbed, or low-permeability formation, regardless of well orientation, were hydraulically fractured
 6    in the year they started producing.*

 7    We used well-specific data provided by oil and gas well operators to the EPA to supplement our
 8    estimates of hydraulic fracturing using Drillinglnfo data [U.S. EPA, 2015o]. Matching wells in each
 9    dataset using API well numbers, we found that 80% of 171 newly drilled wells known to be
10    fractured in 2009 and 2010 according to their well files were correctly identified as fractured using
11    well and formation properties in Drillinglnfo.2 We did not correctly identify all of the vertical or
12    deviated wells that were known to be fractured. (We were unable to identify wells for which
13    hydraulic fracturing was inferred using the properties in Drillinglnfo but were not fractured.) This
14    comparison suggests thatthe estimates of hydraulically fractured wells from Drillinglnfo are likely
15    underestimates.

16    Another source of estimates is from a U.S. Geological Survey publication that reviewed data from
17    the commercial IHS database of U.S. oil and gas production and well data [USGS. 2015]. The study
18    period was from 1947 through 2010. The authors estimated a total of approximately 277,000
19    hydraulically fractured wells between 2000 and 2010 (compared to close to 212,000 during the
20    same time period estimated based on Drillinglnfo data). This is roughly 25,000 wells per year over
21    that time period. Approximately three-quarters of these wells were vertical. Reflecting advances in
22    directional drilling technology over the decade ending in 2010, the percentage of total wells
23    fractured that were horizontal or deviated wells grew from less than 10% to  over 60%.

24    Well counts tracked by Baker Hughes provide another estimate of new wells  fractured annually.
25    Since 2012, this oilfield service company has published a quarterly count of new we 11s spudded;  it
26    includes only new inland U.S. wells "identified to be significant consumers of oilfield services and
27    supplies."3 A reported total of 36,824 oil and gas wells were spudded in the United States in 2012,
28    with new wells per quarter fluctuating between about 8,500 and 9,500 (Baker Hughes, 2014b).
29    While 100% of new wells are probably not hydraulically fractured (see below for estimates of
3 0    hydraulic fracturing rates in new wells), a count of new wells also does not include hydraulic
31    fracturing taking place in older, existing wells.
      1 The assignment of formation type (shale, coalbed, low-permeability, or conventional) for each well was based on a
      crosswalk of information on basin/play provided in Drillinglnfo f2014a) with expert knowledge of those basins/plays at
      EIA f2012a). If formation type could not be determined, it was considered conventional by default. This is similar
      methodology to that used by the EPA for its greenhouse gas inventory [U.S. EPA. 2013c).
      2 An API well number is a unique identifying number given to each oil and gas well drilled in the United States. The system
      was developed by the American Petroleum Institute.
      3 To spud a well is to start the well drilling process by removing rock, dirt, and other sedimentary material with the drill
      bit.
                    This document is a draft for review purposes only and does not constitute Agency policy.
      June 2015                                     2-26                   DRAFT—DO NOT CITE OR QUOTE

-------
      Hydraulic Fracturing Drinking Water AssessmentChapter 2-Hydraulic Fracturing, OH and Gas Production, andthe U.S. Energy Sector


 1    Data collected under the EPA's Greenhouse Gas Reporting Program (GHGRP) provide information
 2    on completions and workovers with hydraulic fracturing (i.e., re-fracturing) of gas wells. Data
 3    reported to GHGRP for years 2011 to 2013 suggest that 9-14% of the gas wells reported to be
 4    hydraulically fractured in each year were pre-existing wells undergoing re-fracturing [U.S. EPA,
 5    20146].1 The GHGRP requirements do not include reporting of re-fracturing in oil wells, and other
 6    data sources for information specifically on re-fracturing of existing oil wells compared to initial
 7    fracturing of oil wells were not identified. For comparison, an EPA survey of an estimated 23,200 oil
 8    and gas production wells that were hydraulically fractured by nine oil  and gas service companies in
 9    2009 and 2010 suggests that 42% of the wells were pre-existing (i.e., more than one year old) when
10    they were hydraulically fractured (U.S. EPA, 2015o). Differences in data (including data from
11    different years and data from gas wells only (GHGRP) versus oil and gas wells, for instance),
12    definitions, and assumptions used to estimate the percentage of pre-existing wells hydraulically
13    fractured in a year could account for the different results.

14    In summary, determination of the national scope of hydraulic fracturing activities in the United
15    States is complicated by a lack of a centralized source of information and the fact that well and
16    drilling databases each track different information. There is also uncertainty about whether
17    information sources are representative of the nation, whether they include data for all production
18    types, whether they represent only modern (high volume) hydraulic fracturing, and whether they
19    include activities in both conventional and unconventional reservoirs.  Taking these limitations into
20    account, however, it is reasonable to assume that between approximately 25,000 and 30,000 new
21    wells (and, likely, additional pre-existing wells) were hydraulically fractured each year in the
2 2    United States from about 2 011 to 2 014.

      2.2.2.  Hydraulic Fracturing Rates
2 3    Estimates of hydraulic fracturing rates, or the proportion of all oil and gas production wells that are
24    associated with hydraulic fracturing, also indicate widespread use of the practice. Based on an
25    assessment described above of data from Drillinglnfo (2014a), hydraulic fracturing rates have
26    increased over time. From 2005 to 2012, rates of hydraulic fracturing increased from 57% to 64%
27    of all new production wells (including oil wells, gas wells, and wells producing both oil and  gas).

28    In 2009, industry consultants stated that hydraulic fracturing was used on nearly 79%  of all wells
29    and more than 95% of "unconventional" wells (IHS, 2009). A 2010 article in an industry publication
3 0    noted "some believe that approximately 60% of all wells drilled today  are fractured" (Montgomery
31    and Smith. 2010). Of 11 important oil and gas producing states that responded to anIOGCC survey
32    (Arkansas, Colorado, Louisiana, New Mexico, North Dakota, Ohio, Oklahoma,  Pennsylvania, Texas,
3 3    Utah, and West Virginia), ten estimated that 78% to 99% of new oil and gas wells in their states
34    were hydraulically fractured in 2012; Louisiana was the one exception, reporting a fracturing rate
35    of 3.9% in 2012 (IOGCC. 2015). Although estimates of fracturing rates  are variable, largely ranging
36    from near 60% to over 90% (as described above), they are often higher for gas wells than they are
37    for oil wells. A2010to2011 industry survey of 20 companies involved in natural gas production
      1 The GHGRP reporting category that covers re-fracturing is "workovers with hydraulic fracturing."


                   This document is a draft for review purposes only and does not constitute Agency policy.
      June 2015                                    2-27                  DRAFT—DO NOT CITE OR QUOTE

-------
      Hydraulic Fracturing Drinking Water AssessmentChapter 2-Hydraulic Fracturing, OH and Gas Production, andthe U.S. Energy Sector


 1    found that 94% of the wells that they operated were fractured; among those, roughly half were
 2    vertical and half were horizontal [Shires and Lev-On. 2012].

      2.3.   Trends and Outlook for the Future
 3    Fossil fuels are the largest source of all energy generated in the United States. They currently
 4    comprise approximately 80% of the energy produced [EIA. 2014Q. However, the mix of fossil fuels
 5    has shifted in recent years. Coal, the leading fossil fuel produced by the U.S. since the 1980s, has
 6    experienced a significant decrease in production. In 2007, coal accounted for approximately 33% of
 7    U.S. energy production, and by 2013 it decreased to approximately 24% [EIA, 2014F). On the other
 8    hand, natural gas production has risen to unprecedented levels, and oil production has resurged to
 9    levels not seen since the 1980s (see Figure 2-17). Oil went from accounting for 15% of U.S. energy
10    production to 19% between 2007 and 2013, and natural gas (both dry and liquid)  went from 31%
11    to 35% (EIA. 2014Q.

12    Below, we discuss recent and projected shifts in oil and natural gas production that can primarily
13    be attributed to hydraulic fracturing and directional drilling technologies.

      2.3.1.  Natural Gas (Including Coalbed Methane)
14    Natural gas production in the United States peaked in the early 1970s, reached those levels again in
15    the mid-1990s, and between the mid- to late-2000s has increased to even higher levels (see Figure
16    2-17). The recent increase in total gas production has been driven almost entirely by shale gas (see
17    Figure 2-18).

18    As natural gas prices fell between 2008 and 2012 (EIA. 2014e). drilling of new natural  gas wells
19    declined markedly (EIA, 2014g] (see Figure  2-19). Nevertheless, natural gas production is expected
20    to increase over the coming decades (see Figure 2-18). EIA(2013b] predicts that shale gas
21    production will more than double between 2011 and 2040 and that the portion of total natural gas
22    production represented by shale gas will increase from one-third to one-half. The EIA projects
23    steady growth in the development of tight gas as well (about a 25% increase in production over the
24    30-year period) and delayed growth in the development of coalbed methane resources, for  which
25    production is not expected to increase again until sufficiently high natural gas prices are realized
26    around 2035. Overall, the EIA projects thatthe share of U.S. natural gas production from shales,
27    tight formations, and coalbeds will increase  from 65% in 2011 to nearly 80% in 2040.
                   This document is a draft for review purposes only and does not constitute Agency policy.
      June 2015                                    2-28                  DRAFT—DO NOT CITE OR QUOTE

-------
Hydraulic Fracturing Drinking Water AssessmentChapter 2-Hydraulic Fracturing, OH and Gas Production, andthe U.S. Energy Sector

or
TO ***
ns •* "*B • ****. y
i5S 71- .* *. /
| /v'" /"\-,*% IX
| "° : ' ^x' '.... •
3 .'..- /' -. .•
• • ' •••
C 1 r _• ^
5 1 / /
1 1Q ;/*' f
/
5/



A fi
- 3,5
3.0
- 2.5
- 2.0
- 1.5
- 1.0
U.S
n n





0
•S
~
1
c
o
15



                         — — — U.S. Natural Gas Gross Withdrawals (trillion cubic feet)
                         	U.S. Field Production of Crude Oil (billion barrels)
Figure 2-17. Trends in U.S. oil and gas production.
           Source: EIA (2013d) and EIA(2014d).
                               History
2012
Projections
                    1990       2000       2010       2020      2030       2040
Figure 2-18. Historic and projected natural gas production by source (trillion cubic feet).
           Source: EIA(2014a).
               This document is a draft for review purposes only and does not constitute Agency policy.
June 2015                                        2-29                    DRAFT—DO NOT CITE OR QUOTE

-------
      Hydraulic Fracturing Drinking Water AssessmentChapter 2-Hydraulic Fracturing, OH and Gas Production, andthe U.S. Energy Sector
                2003    2004    2005   2006   2007    2008    2009    2010    2011   2012
                       	U.S. Natural Gas Wellhead Price (Dollars per Thousand Cubic Feet)
                       — — — U.S. Natural Gas Rotary Rigs in Operation (Count)
                       ^^^U.S. Crude Oil Rotary Rigs in Operation (Count)

      Figure 2-19. Natural  gas prices and oil and gas drilling activity, 2008-2012.
                Source: EIA(2014e). EIA(2014g). and EIA(2013b).
 1    Shale gas production varies by play (see Figure 2-20a). Until 2010, the Texas Barnett Shale was the
 2    play with the most production. Although production from the Barnett Shale is still significant,
 3    production has increased sharply in other plays. By 2012, production from the Haynesville play (on
 4    the Louisiana/Texas border) surpassed that in the Barnett play, and by 2013 the Marcellus Shale
 5    (in the Appalachian Basin underlying Pennsylvania, West Virginia, and other states) was the play
 6    with the most production. Because technically recoverable resources are an order of magnitude
 7    higher in the Marcellus than in any other U.S. shale gas play, it is likely that the Marcellus Shale will
 8    be very active in shale gas production for the foreseeable future (EIAJ_2fllla).1

 9    In the 1970s, most tight gas production in the United States was in the San Juan Basin centered  in
10    New Mexico. As modern hydraulic fracturing came into common usage in the mid-2000s, the lead in
11    tight gas production shifted to Texas (especially East Texas) and the Rocky Mountain states (Vidas
12    and Hugman. 20081
      1 Technically recoverable resources represent the volumes of oil and natural gas that could be produced with current
      technology, regardless of oil and natural gas prices and production costs [EIA. 2013c).
                    This document is a draft for review purposes only and does not constitute Agency policy.
      June 2015                                     2-30                   DRAFT—DO NOT CITE OR QUOTE

-------
      Hydraulic Fracturing Drinking Water AssessmentChapter 2-Hydraulic Fracturing, OH and Gas Production, andthe U.S. Energy Sector
                         U.S. dry shale gas production
                              billion cubic feet per day
            • Marcel I us (PA & WV)
            • Haynesville(LA&TX)
            • Eagle Ford (TX)
            • Fayetteville(AR)
            • Barnett (TX)
            • Woodford (OK)
              Bakken (ND)
            • Antrim (Ml, IN, & OH)
            • Utica(OH, PA&VW)
              Rest of US 'shale'
                                                 40
35
30
25
20
15
10
i Eagle Ford (TX)
 Bakken (MT&ND)
i Spraberry (TX & NM Permian)
• Bonespring (TX & NM Permian)
 Wolfcamp (TX & NM Permian)
 Delaware (TX & N M Permian)
 Yeso-Glorieta (TX & NM Permian)
 Niobrara-Codell (CO, WY)
iHaynesville
iUtica(OH, PA&WV)
• Marcellus
i Woodford (OK)
 Granite Wash (OK & TX)
i Austin Chalk (LA &TX)
 Monterey (CA)
                             U.S. tight oil production
                          million barrels of oil per day
                                                4.5
                                                                                                 0.0
                                           <£>
                             (a)
                           (b)
      Figure 2-20. (a) Production from U.S. shale gas plays, 2000-2014, in billion cubic feet per day;
                (b) Production from U.S. tight oil plays, 2000-2014.
                Tight oil includes oil from shale and other tight formations, plus lease condensate from natural gas
                production. Source: EIA(2012c).

 1    Modern coalbed methane production techniques were pioneered in the Black Warrior Basin in
 2    Alabama and in the San Juan Basin [Vidas and Hugman. 2008]. With the use of hydraulic fracturing,
 3    most coalbed methane production in the United States now comes from the San Juan Basin and
 4    from Rocky Mountain Basins (e.g., the Uinta-Piceance Basin in Colorado and Utah and the Powder
 5    River Basin centered in Wyoming) [Vidas and Hugman, 2008].

      2.3.2.  Oil
 6    The EIA data indicate that as drilling activity for natural gas declined between 2008 and 2012,
 7    drilling for oil increased by a similar order of magnitude (see Figure 2-19]. Figure 2-21 shows past
 8    and projected future trends in U.S. oil production and importation (EIA. 2013a]. Note that this
 9    graph shows production and importation in millions of barrels (bbl] per day. The current surge in
10    tight oil production is expected to continue until the latter part of the current decade and then
11    taper, while conventional oil production is projected to remain fairly level. However, downward
12    trends in the price of oil since mid-2014 are not reflected in these projections.
                    This document is a draft for review purposes only and does not constitute Agency policy.
      June 2015                                      2-31                   DRAFT—DO NOT CITE OR QUOTE

-------
      Hydraulic Fracturing Drinking Water AssessmentChapter 2-Hydraulic Fracturing, OH and Gas Production, andthe U.S. Energy Sector
                         Historic
                                                2012
Projected
          1970
                   1980
                            1990
                                     2000
                                              2010
                                                       2020
                                                                 2030
                                                                            Net Petroleum &
                                                                            Biofuel Imports
                                                                            Other
                                                                            Natural Gas
                                                                            Plant Liquids
                                                                         17% Tight Oil
                                                                            Production
                                                                            Crude Oil Production
                                                                         23% (Excluding Tight Oil)
                                                                         2040
      Figure 2-21. U.S. petroleum and other liquid fuels supply by source, past and projected future
                trends (million barrels per day).
                Source: EIA(2013a).

 1    Like shale gas production, tight oil production varies by play (Figure 2-20b). The Bakken Shale play,
 2    centered in western North Dakota, is important for shale oil production with production increasing
 3    from 123 million bbl (20 billion L) in 2011 to 213 million bbl (34 billion L) in 2012. Proved reserves
 4    in the Bakken have increased from almost 2 billion to over 3 billion bbl (316 billion L to 503 billion
 5    L). The Eagle Ford play in Texas is another major area of shale oil activity, with production
 6    increasing from 71 million bbl (11 billion L) in 2011 to 210 million bbl (33 billion L) in 2012, and
 7    proved reserves increasing from 1.25 billion to 3.4 billion bbl (199 billion to 536 billion L) (EIA,
 8    2014b].  Oil production from the Eagle Ford surpassed that from the Bakken in 2013 (EIA, 2014h].
 9    Among other shale oil plays that might become important in future domestic U.S. oil production, the
10    Niobrara (centered in Colorado) and Austin Chalk (in Texas, Louisiana, and Mississippi) are
11    believed to have quantities of recoverable resources on the same order of magnitude as the Bakken
12    and Eagle  Ford plays fEIA. 2012bj.

      2.4.  Conclusion
13    Since about 2005, the combination of hydraulic fracturing and horizontal drilling pioneered in the
14    Barnett Shale have become widespread in the oil and gas industry. Hydraulic fracturing is now a
15    standard industry practice and has significantly contributed to a surge in U.S. production of both oil
16    and gas.  Modern hydraulic fracturing has resulted in additional types of geological formations being
17    tapped, and sometimes these formations are located in regions of the country new to intensive oil
                    This document is a draft for review purposes only and does not constitute Agency policy.
      June 2015                                     2-32                  DRAFT—DO NOT CITE OR QUOTE

-------
      Hydraulic Fracturing Drinking Water AssessmentChapter 2-Hydraulic Fracturing, OH and Gas Production, andthe U.S. Energy Sector


 1    and gas exploration and production. In other areas, the improved techniques have made possible a
 2    resurgence of production.

 3    An estimated 25,000 to 30,000 new wells drilled in the United States were hydraulically fractured
 4    as a production-enhancing technique in each year from 2011 to 2014. Additional pre-existing wells
 5    were also fractured. Since the early 2000s, the percentage of all hydraulically fractured wells that
 6    are either horizontal or deviated has steadily grown. Reserves of oil and gas that are now accessible
 7    with modern hydraulic fracturing are considerable, and if technical improvements outpace
 8    depletion of oil and gas resources, the quantity of resources that are deemed economically and
 9    technically recoverable may continue to grow. Given current trends, it appears likely that hydraulic
10    fracturing will continue to play an important role in the oil and gas industry, and the United States'
11    energy portfolio, in the decades ahead.

      2.5.   References for Chapter 2

      Abou-Sayed, IS: Sorrell, MA: Foster, RA: Atwood, EL: Youngblood, DR. (2011). Haynesville shale development
         program: From vertical to horizontal. Paper presented at North American Unconventional Gas Conference
         and Exhibition, June 14-16, 2011, The Woodlands, TX.

      ALL Consulting (ALL Consulting, LLC). (2004). Coal bed methane primer: New source of natural gas and
         environmental implications. Tulsa, OK: U.S. Department of Energy, National Petroleum Technology Center.
         http://bogc.dnrc.mt.gov/PDF/Web%20Version.pdf

      Allison. D: Folds. PS: Harless. PI: Howell. M: Vargus. GW: Stipetich. A. (2009). Optimizing openhole completion
         techniques for horizontal foam-drilled wells. Paper presented at SPE Eastern Regional Meeting, September
         23-25,2009, Charleston, WV.

      API (American Petroleum Institute). (2010a). Isolating potential flow zones during well construction
         [Standard] (1st ed.). (RP 65-2). Washington, DC: API Publishing Services.
         http://www.techstreet.com/products/preview/1695866

      Arthur. ID: Bohm. B: Cornue. D. (2009a). Environmental considerations of modern shale gas development.
         Paper presented at SPE Annual Technical Conference and Exhibition, October 4-7, 2009, New Orleans, LA.

      Arthur. ID: Layne. MA: Hochheiser. HW: Arthur. R. (2014). Spatial and statistical analysis of hydraulic
         fracturing activities in U.S. shale plays and the effectiveness of the FracFocus chemical disclosure system.
         In 2014 SPE hydraulic fracturing technology conference. Richardson, TX: Society of Petroleum Engineers.
         http://dx.doi.org/10.2118/168640-MS

      Baker Hughes. (2014b). Well count. U.S. onshore well count [Database]. Houston, TX: Baker Hughes, Inc.
         Retrieved from http://phx.corporate-ir.net/phoenix.zhtml?c=79687&p=irol-wellcountus

      Barati, R: Liang, IT. (2014). A review of fracturing fluid systems used for hydraulic fracturing of oil and gas
         wells. J Appl Polymer Sci Online pub. http://dx.doi.org/10.1002/app.40735

      Barbot. E: Vidic. NS: Gregory. KB: Vidic. RD. (2013). Spatial and temporal correlation of water quality
         parameters of produced waters from Devonian-age shale following hydraulic  fracturing. Environ Sci
         Technol 47: 2562-2569.

      Beckwith, R. (2011). Proppants: Where in the world. J Pet Tech 63: 36-41.

      Blauch. ME. (2010). Developing effective and environmentally suitable fracturing fluids using hydraulic
         fracturing flowback waters. Paper presented at SPE Unconventional Gas Conference, February 23-25,
         2010, Pittsburgh, PA.
                    This document is a draft for review purposes only and does not constitute Agency policy.

      June 2015                                      2-33                    DRAFT—DO NOT CITE OR QUOTE

-------
Hydraulic Fracturing Drinking Water AssessmentChapter 2-Hydraulic Fracturing, OH and Gas Production, andthe U.S. Energy Sector
Bosworth. S: El-Saved. HS: Ismail. G: Ohmer. H: Stracke. M: West. C: Retnanto. A. (1998). Key issues in
   multilateral technology. Oilfield Rev 10:14-28.

Calvert. DG: Smith. DK. (1994). Issues and techniques of plugging and abandonment of oil and gas wells.
   Paper presented at SPE Annual Technical Conference and Exhibition, September 25-28,1994, New
   Orleans, LA.

CCST (California Council on Science and Technology). (2015). An  independent scientifc assessment of well
   stimulation in California, Volume 1: Well stimulation technologies and their past, present, and potential
   future use in California. Sacramento, CA. http://www.ccst.us/publications/2015/2015SB4-vl.pdf

Ciezobka, I: Salehi, I. (2013). Controlled hydraulic fracturing of naturally fractured shales: A case study in the
   Marcellus Shale examining how to identify and exploit natural fractures. (SPE-164524-MS). Ciezobka, J;
   Salehi, I. http://dx.doi.org/10.2118/164524-MS

Cipolla. C: Weng. X: Mack. M: Ganguly. U: Gu. H: Kresse. 0: Cohen.  C. (2011). Integrating microseismic mapping
   and complex fracture modeling to characterize hydraulic fracture complexity. Paper presented at SPE
   Hydraulic Fracturing Technology Conference, January 24-26,2011, The Woodlands, TX.

DOE (U.S. Department of Energy). (2011b). Shale gas Applying technology to solve americas energy
   challenges. Available online at http://groundwork.iogcc.org/topics-index/shale-gas/topic-
   resources/shale-gas-applying-technology-to-solve-americas-energy-challe

Drillinglnfo. Inc.. (2014a). DI Desktop June 2014 download [Database]. Austin, TX: Drillinglnfo. Retrieved
   from http://info.drillinginfo.com/

Drohan. PI: Brittingham. M. (2012). Topographic and soil constraints to shale-gas development in the
   northcentral Appalachians. Soil Sci Soc Am J 76: 1696-1706. http://dx.doi.org/10.2136/sssaj2012.0087

EIA (Energy Information Administration). (2011a). Review of emerging resources: U.S. shale gas and shale oil
   plays. United States Department of Energy, http://www.eia.gov/analysis/studies/usshalegas/

EIA (Energy Information Administration). (2011b). Shale gas and oil plays, lower 48 States [Map]. Available
   online at http://www.eia.gov/pub/oil gas/natural gas/analysis  publications/maps/maps.htm

EIA (Energy Information Administration). (2012a). Formation crosswalk. Washington, DC:  U.S. Energy
   Information Administration.

EIA (Energy Information Administration). (2012b). Today in energy: Geology and technology drive estimates
   of technically recoverable resources. Washington, DC: U.S. Energy Information Administration.
   http://www.eia.gov/todayinenergy/detail.cfm?id=7190

EIA (Energy Information Administration). (2012c). What is shale  gas and why is it important? [December 5].
   Washington, DC: U.S. Energy Information Administration.
   http://www.eia.gov/energy in brief/article/about shale gas.cfm

EIA (Energy Information Administration). (2013a). Analysis & projections: AE02014 early  release overview.
   Release date: December 16, 2013. Washington, DC: U.S. Energy Information Administration.
   http://www.eia.gov/forecasts/aeo/er/executive summary.cfm

EIA (Energy Information Administration). (2013b). Annual energy outlook 2013 with projections to 2040.
   (DOE/EIA-0383). Washington, DC: U.S. Energy Information Administration.
   http://www.eia.gov/forecasts/archive/aeol3/pdf/0383(2013).pdf

EIA (Energy Information Administration). (2013c). Technically recoverable shale oil and shale gas resources:
   an assessment of 137 shale formations in 41 countries outside the United States (pp. 730). Washington,
   D.C.: Energy Information Administration, U.S. Department of Energy.
   http://www.eia.gov/analysis/studies/worldshalegas/

EIA (Energy Information Administration). (2013d). U.S. field production of crude oil. Release date: September
   27, 2014. Washington, DC: U.S. Energy Information Administration.
   http://www.eia.gov/dnav/pet/hist/LeafHandler.ashx?n=pet&s=mcrfpusl&f=a


               This document is a draft for review purposes only and does not constitute Agency policy.
June 2015                                       2-34                    DRAFT—DO  NOT CITE OR QUOTE

-------
Hydraulic Fracturing Drinking Water AssessmentChapter 2-Hydraulic Fracturing, OH and Gas Production, andthe U.S. Energy Sector
EIA (Energy Information Administration). (2014a). Annual energy outlook 2014 with projections to 2040.
   (DOE/EIA-0383(2014)). Washington, D.C.: U.S. Energy Information Administration.
   http://www.eia.gov/forecasts/aeo/pdf/0383(2014).pdf

EIA (Energy Information Administration). (2014b). Natural gas. U.S. crude oil and natural gas proved
   reserves. With data for 2012. Table 2. Principal tight oil plays: oil production and proved reserves, 2011-
   12. Release date: April 10, 2014. Washington, DC: U.S. Energy Information Administration.
   http://www.eia.gov/naturalgas/crudeoilreserves/

EIA (Energy Information Administration). (2014d). Natural gas. U.S. natural gas gross withdrawals.
   Washington, DC: U.S. Energy Information Administration.
   http://www.eia.gov/dnav/ng/ng prod sum a EPGO FGW mmcf a.htm

EIA (Energy Information Administration). (2014e). Natural gas: Natural gas prices [Database]. Washington,
   DC: U.S. Energy Information Administration. Retrieved from
   http://www.eia.gov/dnav/ng/ng pri sum dcu nus a.htm

EIA (Energy Information Administration). (2014f). October 2014 month energy review. (DOE/EIA-
   0035(2014/10)). Washington, D.C.: U.S. Energy Information Administration.
   http://www.eia.gov/totalenergy/data/monthly/archive/00351410.pdf

EIA (Energy Information Administration). (2014g). Petroleum & other liquids. Crude oil and natural gas
   drilling activity. Washington, DC: U.S. Energy Information Administration.
   http://www.eia.gov/dnav/pet/pet crd drill si a.htm

EIA (Energy Information Administration). (2014h). Today in energy: Tight oil production pushes U.S. crude
   supply to over 10% of world total. Washington, DC: U.S. Energy Information Administration.
   http://www.eia.gov/todayinenergy/detail.cfm?id=15571.

EIA (Energy Information Administration). (2015b). Lower 48 states shale plays. Available online at
   http://www.eia.gov/oil gas/rpd/shale gas.pdf

GNB (Government of New Brunswick). (2015). FAQs hydraulic fracturing (fraccing). New Brunswick, Canada.
   http://www2.gnb.ca/content/dam/gnb/Corporate/pdf/ShaleGas/en/FAO HydraulicFracturing.pdf

Gupta, DVS: Hlidek, BT. (2009). Frac fluid recycling and water conservation: A case history. In 2009 Hydraulic
   fracturing technology conference. Woodlands, Texas: Society of Petroleum Engineers.
   http://dx.doi.org/10.2118/119478-MS

Gupta. DVS: Valko. P. (2007). Fracturing fluids and formation damage. In M Economides; T Martin (Eds.),
   Modern fracturing: enhancing natural gas production (pp. 227-279). Houston, TX: Energy Tribune
   Publishing Inc.

GWPC (Groundwater Protection Council). (2015). FracFocus - chemical disclosure registry. Available online at
   http://fracfocus.org/

GWPC and ALL Consulting (Ground Water Protection Council (GWPC) and ALL Consulting). (2009). Modern
   shale gas development in the United States: A primer. (DE-FG26-04NT15455). Washington, DC: U.S.
   Department of Energy, Office of Fossil Energy and National Energy Technology Laboratory.
   http://www.gwpc.org/sites/default/files/Shale%20Gas%20Primer%202009.pdf

Halliburton. (1988). Primer on Hydraulic Fracturing. Provided to EPA on March 2, 2011. Available at Docket
   ID: EPA-HQ-ORD-2010-0674-1634. (HESI-3031). Halliburton.
   http://www.regulations.gov/#!documentDetail:D=EPA-HO-ORD-2010-0674-1634

Halliburton. (2013). Hydraulic fracturing 101. Available online at
   http://www.halliburton.com/public/projects/pubsdata/hydraulic fracturing/index.html

Haymond. D. (1991). The Austin Chalk - An overview. HGS Bulletin 33: 27-30, 32, 34.
               This document is a draft for review purposes only and does not constitute Agency policy.

June 2015                                       2-35                    DRAFT—DO NOT CITE OR QUOTE

-------
Hydraulic Fracturing Drinking Water AssessmentChapter 2-Hydraulic Fracturing, OH and Gas Production, andthe U.S. Energy Sector
Holditch. SA. (2007). Chapter 8: Hydraulic fracturing. In JD Clegg (Ed), Petroleum engineering handbook (pp.
   IV-323 - IV-366). Richardson, TX: Society of Petroleum Engineers. http://store.spe.org/Petroleum-
   Engineering-Handbook-Volume-IV-Production-Operations-Engineering-P61.aspx

Hyne. Ml. (2012). Nontechnical guide to petroleum geology, exploration, drilling and production. In
   Nontechnical guide to petroleum geology, exploration, drilling and production (3 ed.). Tulsa, OK: PennWell
   Corporation.

IHS (Global Insight). (2009). Measuring the economic and energy impacts of proposals to regulate hydraulic
   fracturing. Prepared for American Petroleum Institute. Lexington, MA: IHS Global Insight.
   http://s3.amazonaws.com/propublica/assets/natural gas/ihs gi hydraulic fracturing taskl.pdf

IOGCC (Interstate Oil and Gas Compact Commission). (2002). States experience with hydraulic fracturing. A
   survey of the interstate oil and gas compact commission. IOGCC (Interstate Oil and Gas Compact
   Commission). http://energyindepth.org/docs/pdf/IOGCC%20Hydraulic%20Fracturing%20Study%2007-
   2002.pdf

IOGCC (Interstate Oil and Gas Compact Commission). (2015). Groundwork: hydraulic fracturing - state
   progress. Available online at http://groundwork.iogcc.ok.gov/topics-index/hydraulic-fracturing/state-
   progress

Kargbo. DM: Wilhelm. RG: Campbell. PI. (2010). Natural gas plays in the Marcellus Shale: Challenges and
   potential opportunities. Environ Sci Technol 44: 5679-5684. http://dx.doi.org/10.1021/es903811p

King. GE. (2012). Hydraulic fracturing 101: What every representative, environmentalist, regulator, reporter,
   investor, university researcher, neighbor and engineer should know about estimating frac risk and
   improving frac performance in unconventional gas and oil wells. SPE Hydraulic Fracturing Technology
   Conference, February 6-8, 2012, The Woodlands, TX.

Kundert. D: Mullen. M. (2009). Proper evaluation of shale gas reservoirs leads to a more effective hydraulic-
   fracture stimulation. Paper presented at SPE Rocky Mountain Petroleum Technology Conference, April 14-
   16, 2009, Denver, CO.

Lee. PS: Herman. ID: Elsworth. D: Kim. HT: Lee. HS. (2011). A critical evaluation of unconventional gas
   recovery from the marcellus shale, northeastern United States. K S C E Journal of Civil Engineering 15:
   679-687. http://dx.doi.org/10.1007/sl2205-011-0008-4

Lowe. T: Potts. M: Wood. D. (2013). A case history of comprehensive hydraulic fracturing monitoring in the
   Cana Woodford. In 2013 SPE annual technical conference and exhibition. Richardson, TX: Society of
   Petroleum Engineers. http://dx.doi.org/10.2118/166295-MS

Miskimins. IL. (2008). Design and life cycle considerations for unconventional reservoir wells. In 2008 SPE
   Unconventional Reservoirs Conference. Richardson, TX: Society of Petroleum Engineers.
   http://dx.doi.org/10.2118/114170-MS

Montgomery, C. (2013). Fracturing fluid components. In A Bunder; J McLennon; R Jeffrey (Eds.), Effective and
   Sustainable Hydraulic Fracturing. Croatia: InTech. http://dx.doi.org/10.5772/56422

Montgomery. CT: Smith. MB. (2010). Hydraulic fracturing - History of an enduring technology. J Pet Tech 62:
   26-32.

NETL (National Energy Technology Laboratory). (2013). Modern shale gas development in the United States:
   An update. Pittsburgh, PA: U.S. Department of Energy. National Energy Technology Laboratory.
   http://www.netl.doe.gov/File%20Library/Research/Oil-Gas/shale-gas-primer-update-2013.pdf

Newell, R. (2011). Shale gas and the outlook for U.S. natural gas markets and global gas resources.
   Presentation presented at US EIA presentation at OECD Meetings, June 21, 2011, Paris, France.

NPC (National Petroleum Council). (2011b). Plugging and abandonment of oil and gas wells. (Paper #2-25).
   Washington, DC: National Petroleum Council  (NPC).
               This document is a draft for review purposes only and does not constitute Agency policy.

June 2015                                       2-36                   DRAFT—DO NOT CITE OR QUOTE

-------
Hydraulic Fracturing Drinking Water AssessmentChapter 2-Hydraulic Fracturing, OH and Gas Production, andthe U.S. Energy Sector
NYSDEC (New York State Department of Environmental Conservation). (2011). Revised draft supplemental
   generic environmental impact statement (SGEIS) on the oil, gas and solution mining regulatory program:
   Well permit issuance for horizontal drilling and high-volume hydraulic fracturing to develop the Marcellus
   shale and other low-permeability gas reservoirs. Albany, NY: NY SDEC.
   http://www.dec.ny.gov/energy/75370.html

Patzek. TW: Male. F: Harder. M. (2013). Gas production in the Barnett Shale obeys a simple scaling theory.
   PNAS 110: 19731-19736. http://dx.doi.org/10.1073/pnas.1313380110

Pearson. CM: Griffin. L: Wright. CA: Weijers. L. (2013). Breaking up is hard to do: creating hydraulic fracture
   complexity in the Bakken central basin. In 2013 SPE hydraulic fracturing technology conference.
   Richardson, TX: Society of Petroleum Engineers. http://dx.doi.org/10.2118/163827-MS

Pearson. K. (2011). Geologic models and evaluation of undiscovered conventional and continuous oil and gas
   resourcesUpper Cretaceous Austin Chalk, U.S. Gulf Coast. In US Geological Survey Scientific Investigations
   Report. (20125159). U.S. Geological Survey, http://pubs.usgs.gov/sir/2012/5159/

Rassenfoss. S. (2011). From flowback to fracturing: Water recycling grows in the Marcellus shale. J Pet Tech
   63: 48-51.

Renpu. W. (2011). Advanced well completion engineering (Third ed.). Houston, TX: Gulf Professional
   Publishing.

Ross. D: King. G. (2007). Well completions. In MJ Economides; T Martin (Eds.), Modern fracturing: Enhancing
   natural gas production (1 ed., pp. 169-198). Houston, Texas: ET Publishing.

Saba. T: Mohsen. F: Garry. M: Murphy. B: Hilbert. B. (2012). White paper: Methanol use in hydraulic fracturing
   fluids. (1103844.000 0101 0711 TS26). Maynard, MA: Exponent.

Schlumberger (Schlumberger Limited).  (2006). Fundamentals of formation testing. Sugar Land, Texas.
   http://www.slb.eom/~/media/Files/evaluation/books/fundamentals formation testing overview.pdf

Schlumberger (Schlumberger Limited).  (2014). Schlumberger oilfield glossary. Available online at
   http://www.glossary.oilfield.slb.com/

Shires, T: Lev-On, M. (2012). Characterizing pivotal sources of methane emissions from unconventional
   natural gas production - summary and analysis of API and ANGA survey responses. Washington, DC:
   American Petroleum Institute. American  Natural Gas Alliance.

Spellman.  FR. (2012). Environmental impacts of hydraulic fracturing. In Environmental impacts of hydraulic
   fracturing. Boca Raton, Florida: CRC  Press.

STO  (Statoil).  (2013).  Shale facts: drilling and hydraulic fracturing, how it's done, responsibly. (Global
   Version, April 2013). Stavanger, Norway.
   http://www.statoil.com/no/OurOperations/ExplorationProd/ShaleGas/FactSheets/Downloads/Shale Dr
   illingHydraulicFacturing.pdf

Thompson. AM. (2010) Induced fracture detection in the Barnett Shale, Ft. Worth Basin, Texas. (Master's
   Thesis). University of Oklahoma, Norman, OK.

U.S. EPA (U.S.  Environmental Protection Agency). (2012d). Oil and natural gas sector: standards of
   performance for crude oil and natural gas production, transmission, and distribution. Background
   supplemental technical support document for the final new source performance standards. Washington,
   D.C. http://www.epa.gov/airquality/oilandgas/pdfs/20120418tsd.pdf

U.S. EPA (U.S.  Environmental Protection Agency). (2012f). Study of the potential impacts of hydraulic
   fracturing on drinking water resources: Progress report. (EPA/601/R-12/011). Washington, DC: U.S.
   Environmental Protection Agency, Office of Research and Development.
   http://nepis.epa.gov/exe/ZvPURL.cgi?Dockev=P100FH8M.txt
               This document is a draft for review purposes only and does not constitute Agency policy.

June 2015                                       2-37                    DRAFT—DO NOT CITE OR QUOTE

-------
Hydraulic Fracturing Drinking Water AssessmentChapter 2-Hydraulic Fracturing, OH and Gas Production, andthe U.S. Energy Sector
U.S. EPA (U.S. Environmental Protection Agency). (2013c). Inventory of U.S. greenhouse gas emissions and
   sinks: 1990-2011. Washington, DC: U.S. Environmental Protection Agency, Office of Atmospheric
   Programs. http://www.epa.gov/climatechange/Downloads/ghgemissions/US-GHG-Inventory-2013-
   Main-Textpdf

U.S. EPA (U.S. Environmental Protection Agency). (2014e). Greenhouse gas reporting program, Subpart W -
   Petroleum and natural gas systems. Washington, D.C.

U.S. EPA (U.S. Environmental Protection Agency). (2015a). Analysis of hydraulic fracturing fluid data from the
   FracFocus chemical disclosure registry 1.0 [EPA Report]. (EPA/601/R-14/003). Washington, D.C.: Office
   of Research and Development, U.S. Environmental Protection Agency.
   http://www2.epa.gov/hfstudy/analysis-hydraulic-fracturing-fluid-data-fracfocus-chemical-disclosure-
   registry-1-pdf

U.S. EPA (U.S. Environmental Protection Agency). (2015b). Analysis of hydraulic fracturing fluid data from the
   FracFocus chemical disclosure registry 1.0: Project database [EPA Report]. (EPA/601/R-14/003).
   Washington, D.C.: U.S. Environmental Protection Agency, Office of Research and Development.
   http://www2.epa.gov/hfstudy/epa-project-database-developed-fracfocus-l-disclosures

U.S. EPA (U.S. Environmental Protection Agency). (2015g). Inventory of U.S. greenhouse gas emissions and
   sinks: 1990-2013. (EPA 430-R-15-004). Washington, D.C.
   http://www.epa.gov/climatechange/Downloads/ghgemissions/US-GHG-Inventory-2015-Main-Text.pdf

U.S. EPA (U.S. Environmental Protection Agency). (2015n). Review of well operator files for hydraulically
   fractured oil and gas production wells: Well design and construction [EPA Report]. (EPA/601/R-14/002).
   Washington, D.C.: Office of Research and Development, U.S. Environmental Protection Agency.

U.S. GAP (U.S. Government Accountability Office). (2012). Energy-water nexus: Information on the quantity,
   quality, and management of water produced during oil and gas production. (GAO-12-156). Washington,
   D.C. http://www.gao.gov/products/GAO-12-156

URS Corporation. (2011). Water-related issues associated with gas production in the Marcellus shale:
   Additives use flowback quality and quantities regulations on-site treatment green technologies alternate
   water sources water well-testing. (NYSERDA Contract PO Number 10666).

USGS (U.S. Geological Survey). (2000). Coal-bed methane: Potential and concerns [Fact Sheet]. (Fact Sheet
   123-00). http://pubs.usgs.gov/fs/fsl23-00/fsl23-00.pdf

USGS (U.S. Geological Survey). (2002). Natural gas production in the United States [Fact Sheet]. (USGS Fact
   Sheet FS-113-01). Denver, CO.

USGS (U.S. Geological Survey). (2013a). Map of assessed shale gas in the United States, 2012.
   http://pubs.usgs.gov/dds/dds-069/dds-069-z/

USGS (U.S. Geological Survey). (2014a). Energy glossary and acronym list. Available online at
   http://energy.usgs.gov/GeneralInfo/HelpfulResources/EnergyGlossary.aspxtft

USGS (U.S. Geological Survey). (2015). Trends in hydraulic fracturing distributions and treatment fluids,
   additives, proppants, and water volumes applied to wells drilled in the United States from 1947 through
   2010data analysis and comparison to the literature. (U.S. Geological Survey Scientific Investigations
   Report 20145131). Reston, VA. http://dx.doi.org/10.3133/sir20145131

Vidas, H: Hugman, B. (2008). Availability, economics, and production potential of North American
   unconventional natural gas supplies. (F-2008-03). Washington, DC:  The INGAA Foundation, Inc.
   http://www.ingaa.org/File.aspx?id=7878

Vincent. M. (2011). Restimulation of unconventional reservoirs: when are refracs beneficial? Journal of
   Canadian Petroleum Technology 50: 36-52. http://dx.doi.org/10.2118/136757-PA
               This document is a draft for review purposes only and does not constitute Agency policy.

June 2015                                       2-38                    DRAFT—DO NOT CITE OR QUOTE

-------
Hydraulic Fracturing Drinking Water AssessmentChapter 2-Hydraulic Fracturing, OH and Gas Production, andthe U.S. Energy Sector
Wang. Z: Krupnick. A. (2013). A retrospective review of shale gas development in the United States. What let
   to the boom? (RFF DP 13-12). Washington, DC: Resources for the Future.
   http://www.rff.org/RFF/documents/RFF-DP-13-12.pdf

Weijermars. R. (2014). US shale gas production outlook based on well roll-out rate scenarios. Appl Energ 124:
   283-297. http://dx.doi.0rg/10.1016/j.apenergy.2014.02.058

Yergin. D. (2011). The quest: energy, security and the remaking of the modern world. In The quest: energy,
   security and the remaking of the modern world. New York, NY: Penquin Press.
               This document is a draft for review purposes only and does not constitute Agency policy.

June 2015                                        2-39                    DRAFT—DO NOT CITE OR QUOTE

-------
Hydraulic Fracturing Drinking Water Assessment          Chapter 3-Drinking Water Resources in the UnitedStates
                           Chapter 3
Drinking Water Resources in the United States
          This document is a draft for review purposes only and does not constitute Agency policy.
June 2015                                            DRAFT—DO NOT CITE OR QUOTE

-------
      Hydraulic Fracturing Drinking Water Assessment             Chapter 3 - Drinking Water Resources in the UnitedStates


      3. Drinking Water Resources in the United  States
 1    Consideration of how and where hydraulic fracturing activities potentially impact drinking water
 2    resources requires an understanding of both the activities and the potentially impacted resources.
 3    In Chapter 2, we provided background on hydraulic fracturing and in this chapter, we provide an
 4    overview of drinking water resources in the United States. We describe the use of these resources,
 5    including patterns in current use and trends for future use (Section 3.1). We then characterize the
 6    spatial distribution of hydraulically fractured wells and current surface and ground water supplies
 7    throughout the United States (Section 3.2) to evaluate where potential impacts of hydraulic
 8    fracturing on drinking water resources may occur.

      3.1.  Current and Future Drinking Water Resources
 9    In this assessment, drinking water resources are defined broadly as any body of ground water or
10    surface water that now serves, or in the future could serve, as a source of drinking water for public
11    or private use. Drinking water resources provide not only water that individuals actually drink but
12    also water used for many additional purposes such as cooking and bathing. Our definition of
13    drinking water resources includes both fresh and non-fresh bodies of water.

14    The average American uses about 90 gal (341 L) of drinking water per day for indoor and outdoor
15    purposes (e.g., drinking, food preparation, washing clothes and dishes, flushing toilets, and
16    watering lawns or gardens (Maupinetal.. 2014: AWWA. 1999). Drinking water is supplied to
17    households by either public water systems (PWSs) or private water systems (private ground water
18    wells and surface water intakes).1 In 2011, approximately 270 million people (86% of the
19    population) in the United States relied on water supplied to their homes by one of the more than
20    51,000 community water systems (Maupinetal.. 2014: U.S. EPA. 2013b).2 These systems provided
21    households with nearly 24 billion gal (91 billion L) of water per day (Maupinetal.. 2014).3 In areas
22    without service by PWSs, approximately 43 million people (14% of the population) relied on
23    private sources for drinking water, and private water systems account for about 3.6 billion gal (14
24    billion L) of daily water withdrawals (Maupinetal., 2014).

25    Drinking water resources can be surface waters such as rivers, streams,  lakes, or reservoirs, as well
26    as ground water aquifers. In 2011, approximately 70% of the population receiving drinking water
27    from PWSs relied on surface water, and 30% relied on ground water (U.S. EPA. 2013b). However,
      i Public water systems (PWSs) provide water for human consumption from surface or ground water through pipes or
      other infrastructure to at least 15 service connections or serve an average of at least 25 people for at least 60 days a year
      [U.S. EPA. 2012e). Private (non-public) water systems serve fewer than 15 connections and fewer than 25 individuals
      (U.S. EPA. 1991).
      2 The EPA categorizes public water systems as either community water systems, which supply water to the same
      population year-round, or non-community water systems, which supply water to at least 25 of the same people at least six
      months per year, but not year-round. Approximately 101,000 non-community water systems provide water to non-
      residential facilities (e.g., schools, small businesses, churches, and campgrounds (U.S. EPA. 2013b).
      3 The U.S. Geological Survey (USGS) compiles data in cooperation with local, state, and federal environmental agencies to
      produce water-use information aggregated at the county, state, and national levels. Every five years, data at the county
      level are compiled into a national water use census and state-level data are published. The most recent USGS water use
      report was released in 2014, and contains water use estimates from 2010 (Maupinetal.. 2014: USGS. 2014b).
                    This document is a draft for review purposes only and does not constitute Agency policy.
      June 2015                                      3-1                   DRAFT—DO NOT CITE OR QUOTE

-------
     Hydraulic Fracturing Drinking Water Assessment
Chapters - Drinking Water Resources in the United States
1    the relative importance of surface and ground water sources for supplying drinking water varies
2    geographically (see Figure 3-1). Most larger PWSs rely on surface water and are located in urban
3    areas [U.S. EPA. 2011b). whereas most smaller PWSs rely on ground water and are located in rural
4    areas [U.S. EPA, 2014),  2013b]. In fact, more than 95% of households in rural areas obtain their
5    drinking water from ground water aquifers [U.S. EPA, 2011b].
        Drinking water sources* used by public water systems,
        as a percentage of all sources in a state
          J > 75% surface water sources
             50 - 75% surface water sources
             50 - 75% ground water sources
             > 75% ground water sources
             States with hydraulically fractured wells	
       *Drinking water sources include the following public water system sources: infiltration galleries, intakes, reservoirs, springs, and wells.
       Sources: Drillinglnfo (2014); ESRI (2010); U.S. Environmental Protection Agency (2011, 2013); U.S. Census Bureau (2010)
       Projection: North America Albers Equal Area Conic
                                                                                                  Copyright:© 2014
     Figure 3-1. Geographic variability in drinking water sources for public water systems.
                 The relative importance of surface and ground water as drinking water sources varies by state. The
                 public water system sources used in this analysis include infiltration galleries, intakes, reservoirs,
                 springs, and wells. States with hydraulically fractured wells were identified from Drillinglnfo data.

     The future availability of drinking water resources that are considered fresh in the United States
     will be affected by changes in climate and water use [Georgakakos etal.. 2014: U.S. Global Change
                     This document is a draft for review purposes only and does not constitute Agency policy.
     June 2015                                          3-2                      DRAFT—DO NOT CITE OR QUOTE

-------
      Hydraulic Fracturing Drinking Water Assessment             Chapter 3 - Drinking Water Resources in the UnitedStates


 1    Research Program, 2009].l Since 2000, about 30% of the total area of the contiguous United States
 2    has experienced moderate drought conditions and about 20% has experienced severe drought
 3    conditions [National Drought Mitigation Center. 2015: U.S. EPA. 2015r). Declines in surface water
 4    resources have already led to increased withdrawals and cumulative net depletions of ground
 5    water in some areas [Castle etal., 2014: Georgakakos etal., 2014: Konikow, 2013a: Famiglietti etal.,
 6    2011]. Other sources of water that might not be considered fresh, such as wastewater from sewage
 7    treatment plants, brackish (containing 3,000-10,000 mg/L IDS) and saline (containing more than
 8    10,000 mg/L TDS) surface and ground water, as well as seawater (containing about 35,000 mg/L
 9    TDS) are also increasingly being used to meet water demand. Through treatment or desalination,
10    these water sources can reduce the use of high-quality, potable fresh water for industrial processes,
11    irrigation, recreation, and toilet flushing (i.e., non-potable uses). In addition, in 2010, approximately
12    355 million gal per day (1.3 billion L per day) of treated wastewater was reclaimed through potable
13    reuse projects (NRG. 2012). Such projects use reclaimed wastewater to augment surface drinking
14    water resources or to recharge aquifers that supply drinking water to PWSs (NRG, 2012: Sheng,
15    2005).

16    An increasing number of states are developing new water supplies to augment existing water
17    through reuse of reclaimed water, recycling of storm water, and desalination (U.S.  GAP. 2014). Most
18    desalination programs currently use brackish water, although plans are underway to  expand the
19    use of seawater for desalination in some states. States with the highest installed capacity for
20    desalination include Florida, California, Arizona, and Texas (Cooley etal.. 2006). It is likely that
21    various water treatment technologies will continue to expand drinking water resources beyond
22    those currently being considered for use as drinking water. Therefore, these potential future
23    sources are also considered drinking water resources in this assessment

      3.2. The Proximity of Drinking Water Resources to Hydraulic Fracturing Activity
24    The colocation of hydraulic fracturing activities with surface and ground water increases the
25    potential for impacts to current and future drinking water resources (Vengoshetal.. 2014: Entrekin
26    etal.. 2011). In this section, we analyze the aboveground proximity of hydraulically fractured well
2 7    sites, drinking water resources (including the location of surface water bodies and ground water
28    wells that supply public water systems), and populated areas.2

29    To determine the spatial relationship between hydraulically fractured wells and populated areas,
30    we analyzed the locations of the approximately 273,000 oil and gas wells that were hydraulically
31    fractured in 25 states between 2000 and 2013 (see Chapter 2) with respect to where people live
32    (i.e., census blocks).3 Nationwide, approximately 9.4 million people lived within one mile of a
      1 Fresh water qualitatively refers to water with relatively low TDS that is most readily available for drinking water
      currently.
      2 The vertical proximity of ground water resources to geologic formations and hydraulic fracturing operations is
      addressed in Chapter 6.
      3 In the analyses in this chapter, we only include the oil and gas production wells that we identified were hydraulically
      fractured using criteria outlined in Chapter 2 and that began producing between 2000 and 2013. The well data found in
      Drillinglnfo may not represent the full year for 2013 since the frequency with which Drillinglnfo updates the database
      varies by state. The final update performed by Drillinglnfo for 2013 ranges by state from June 2013 to December 2013.
                    This document is a draft for review purposes only and does not constitute Agency policy.
      June 2015                                      3-3                   DRAFT—DO NOT CITE OR QUOTE

-------
      Hydraulic Fracturing Drinking Water Assessment             Chapter 3 - Drinking Water Resources in the UnitedStates


 1    hydraulically fractured well for some period of time between 2000 and 2013 [Drillinglnfo, 2014a:
 2    U.S. Census Bureau. 2010]: more than 5.7 million people lived within half a mile of a hydraulically
 3    fractured well.

 4    We then analyzed trends in the proximity of hydraulically fractured wells to highly populated areas.
 5    For this analysis, we considered metropolitan areas (areas with more than 50,000 people) and
 6    micropolitan areas (areas with 10,000 to 49,999 people)  (U.S. Census Bureau, 2013C].1
 7    Approximately 81,300 (30%) of new wells hydraulically fractured between 2000 and 2013 were
 8    located within a metropolitan or micropolitan area (see Figure 3-2) (Drillinglnfo, 2014a: U.S.
 9    Census Bureau. 2013c: U.S. EPA. 2013b: ESRI. 2010). From 2000 to 2008, the number of new wells
10    hydraulically fractured per year within metropolitan and micropolitan areas increased 300%; the
11    proportion of wells hydraulically fractured per year in metropolitan and micropolitan areas almost
12    doubled over the same eight-year period (see Figure 3-3).2 From 2008 to 2012, however, the
13    number of wells hydraulically fractured per year in metropolitan and micropolitan areas decreased
14    by about half in comparison to the peak of approximately 10,000 wells in 2 008 (see Figure 3-3),
15    whereas hydraulic fracturing in areas outside of metropolitan and micropolitan areas increased or
16    remained relatively constant (Drillinglnfo, 2014a: U.S. Census Bureau, 2013b).
      i Metropolitan and micropolitan combined statistical areas are geographic entities delineated by the Office of
      Management and Budget. Specifically, a metropolitan combined statistical area is a core urban area of 50,000 or more
      people while a micropolitan combined statistical area is an urban core of at least 10,000, but less than 50,000, people [U.S.
      Census Bureau. 2013c). These terms are referred to as metropolitan and micropolitan areas in this assessment.
      2 For comparison, the Drillinglnfo data indicate an increase in the number of wells estimated to be hydraulically fractured
      each year, regardless of location, from approximately 12,800 in 2000 to slightly more than 21,600 in 2005 to nearly
      23,000 in 2012, the last year for which complete data are available.
                    This document is a draft for review purposes only and does not constitute Agency policy.
      June 2015                                      3-4                   DRAFT—DO NOT CITE OR QUOTE

-------
Hydraulic Fracturing Drinking Water Assessment
                                                                     Chapters - Drinking Water Resources in the United States

                           Proximity of hydraulkalty fractured wells relative to metropolitan and micropolitan areas
                                 Hydraulically fractured wells within metropolitan and micropolitan areas
                                 Hydraulically fractured wells outside of metropolitan and micropolitan areas
                              | Populated census blocks within 1 mile of at least one well hydraulically fractured well
                                 Metropolitan and micropolitan areas without hydraulically fractured wells
                            urces: Drillinglnfa (2014); ESRI (2010); U.S. Census Bureau (2013)
                           roiectlon: North America Albers Equal Area Conic
Figure 3-2. Proximity of hydraulically fractured wells relative to populated areas.
             The estimates of hydraulically fractured wells from  2000 to 2013 developed from the Drillinglnfo data were based on several assumptions
             described in Chapter 2.
June 2015
This document is a draft for review purposes only and does not constitute Agency policy.
                                        3-5
DRAFT—DO NOT CITE OR QUOTE

-------
      Hydraulic Fracturing Drinking Water Assessment
                                                                        Chapters - Drinking Water R
12000
c
TO
ro 10000
"o
Q.
o
E J3 8000
.£ E
-D =
(X) ^
03 vi
<-> ra
O tt)
^ ra 6000
"QJ c
OJ g.
= 0
^ b 4000
3 i
1 2000
1_
T3
X
0
20







0 5









c
0
o o
' • .
) *









c
o
)

• ,





)






o
0 0



	

0
, •
4




1



c
4


00 2002 2004 2006 2008 2010 2012
c
ra
c
- 80% =
a.
o
"S •jr
60% ^ a.
ra vi
u ro
O 
-------
      Hydraulic Fracturing Drinking Water Assessment                                                 Chapters - Drinking Water R


 1    Between 2000 and 2013, approximately 6,800 PWS sources had a hydraulically fractured well
 2    within a 1 mile radius. Most of these PWS sources were located in Colorado, Louisiana, Michigan,
 3    North Dakota, Ohio, Oklahoma, Pennsylvania, Texas, and Wyoming (see Figure 3-5). These PWS
 4    sources had an average of seven fractured wells and a maximum of 144 fractured wells within that
 5    one mile proximity. They also supplied water to 3,924 PWSs—1,609 of which are community water
 6    systems—that served more than 8.6 million people year-round in 2013 [U.S. EPA. 2014): U.S.
 7    Census Bureau. 2013a: U.S. EPA. 201313).1

 8    We also analyzed the location of hydraulically fractured wells relative to populations where a high
 9    proportion (>30%, or twice the national average) obtain drinking water from private systems
10    (private ground water wells and surface water intakes).2 Between 2000 and 2013, approximately
11    3.6 million people obtained drinking water from private systems in counties with at least one
12    hydraulically fractured well fDrillinglnfo. 2014a: USGS. 2014bl and approximately 740,000 people
13    obtained drinking water from private supplies in counties with more than 400 fractured wells
14    (Drillinglnfo, 2014a: USGS, 2014b) (see Figure 3-6).3 These counties were located in Colorado,
15    Kentucky, Michigan, Montana, New Mexico, New York, Oklahoma, Pennsylvania, Texas, and
16    Wyoming (see Figure 3-6).
      i All PWS types were included in the locational analyses performed. However, only community water systems were used
      to calculate the number of customers obtaining water from a PWS with at least one source within 1 mile of a hydraulically
      fractured well. If non-community water systems are included, the estimated number of customers increases by 533,000
      people fU.S.EPA.2012e).
      2 There is no national data set of private water systems. The USGS estimates the proportion of the population reliant on
      private water systems, referred to as the "self-supplied population," by county, based on estimates of the population
      without connections to a public water system [Maupin et al.. 2014). We used the USGS estimates for this analysis.
      3 Approximately 14% of the U.S. population is supplied by private water systems [Maupin et al.. 2014). In this analysis, we
      only considered counties in which more than double the national average—that is, at least 30% of the county's
      population—was supplied by private water systems.
                    This document is a draft for review purposes only and does not constitute Agency policy.
      June 2015                                      3-7                                                         DRAF'

-------
Hydraulic Fracturing Drinking Water Assessment
                                                                 Chapters - Drinking Water Resources in the United States
                            '
                        Number of public water system (PWS) sources located
                       | within 1 mile of a hydraulically fractured well

                             1-3
                          -,   4-8
                          •   9 22

                          •   23-10	

                       Points indicate the location of hydraulically fractured wells; point color indicates
                       the number of hydraulically fractured wells within 1 mile of each PWS source.
                       The following PWS sources were included in this analysis: infiltration galleries, intakes,
                       reservoirs, springs, and wells.
                       Sources: Dtillinglnfo (2014); U.S. Environmental Protection Agency (2014); ESRI (2010)
                       Projection: North America Albers Equal Area Conic
Figure 3-4. Location and number of public water system (PWS) sources located within 1 mile of a hydraulically fractured well.

             Points indicate the location of hydraulically fractured wells; point color indicates the number of hydraulically fractured wells within 1 mile of each
             PWS source. The following PWS sources were included in this analysis: infiltration galleries, intakes, reservoirs, springs, and wells. The estimates of
            wells hydraulically fractured from 2000 to 2013 developed from the Drillinglnfo data were based on several assumptions described in Chapter 2.
June 2015
This document is a draft for review purposes only and does not constitute Agency policy.

                                      3-8
DRAFT—DO NOT CITE OR QUOTE

-------
Hydraulic Fracturing Drinking Water Assessment
                                                                  Chapters - Drinking Water Resources in the United States
                        Number of hydraullcally fractured wells within
                        1 mile of a public water system (PWS) source
                             1-8
                         •  9-24
                         •  25-48
                         •  49 -144
                       Points indicate the location of public water system (PWS) sources; point color Indicates
                       the number of hydraulically fractured wells within l mile of each PWS source.
                       The following PWS sources were included in this analysis: infiltration galleries, intakes,
                       reservoirs, springs, and wells. Note that some PWS systems have more than one source.
                       Sources: Drillinglnfo (2014); U.S. Environmental Protection Agency (2014); ESRI (2010)
                       Projection: North America Albers Equal Area Conic
Figure 3-5. The location of public water system sources within 1 mile of hydraulically fractured wells.
             Points indicate the location of public water system (PWS) sources; point color indicates the number of hydraulically fractured wells within 1 mile of
             each PWS source. The estimates of wells hydraulically fractured from 2000 to 2013 developed from the Drillinglnfo data were based on several
             assumptions described in Chapter 2.
June 2015
This document is a draft for review purposes only and does not constitute Agency policy.
                                      3-9
DRAFT—DO NOT CITE OR QUOTE

-------
Hydraulic Fracturing Drinking Water Assessment
                                                               Chapters - Drinking Water Resources in the United States

                       Number of hydraulically fractured
                       wells per county
                            1-400
                         	] 401  1400
                        H 1101 - 3400
                        B 3401 - 8600
                        B 8601 • 16051
                            Counties with 30% or more of
                        —I population supplied by private water
                        _l systems and more than 400
                            hydraulicaliy fractured wells
                      Sources: Drillinglnfo (2014); U.S. Geological Survey (2014); ESRI (2010); U.S. Census Bureau (2010)
                      Projection: North America Albers Equal Area Conic
Figure 3-6. Co-occurrence of hydraulic fracturing activity and populations supplied by private water systems.
            Color indicates the number of hydraulically fractured wells per county. The estimates of wells hydraulically fractured from 2000 to 2013 developed
            from the Drillinglnfo data were based on several assumptions described in Chapter 2. Counties with more than 400 hydraulically fractured wells
            and in which at least 30% of the population is supplied by private water systems are outlined in blue.
June 2015
This document is a draft for review purposes only and does not constitute Agency policy.
                                    3-10
DRAFT—DO NOT CITE OR QUOTE

-------
      Hydraulic Fracturing Drinking Water Assessment                                                Chapters - Drinking Water R


      3.3. Conclusion
 1    The evaluation of potential hydraulic fracturing impacts on drinking water resources in the United
 2    States depends on an understanding of how the country's current and future drinking water needs
 3    are and will be met The U.S. population requires sufficient drinking water resources—that is,
 4    bodies of fresh or non-fresh surface or ground water that now serve, or in the future could serve, as
 5    a source of water for drinking water for public or private use—to meet everyday needs. Currently,
 6    most people in the United States rely on water supplied to their homes via public water systems,
 7    and most of this water comes from fresh surface water bodies. Shortages in fresh water availability
 8    in the United States, especially in the western United States, have already led some states to
 9    augment their water supplies with other water sources (e.g., brackish and saline surface and
10    ground water, seawater, and reclaimed wastewater), suggesting that additional water bodies may
11    provide drinking water as the quantity and quality of existing sources change.

12    The colocation of hydraulic fracturing activities with drinking water resources increases the
13    potential for these activities to affect the quality and quantity of current and future drinking water
14    resources. While close proximity of hydraulically fractured wells to drinking water resources does
15    not necessarily indicate that an impact has or will occur, information about the relative location of
16    wells and water supplies is an initial step in understanding where potential impacts might occur.

17    Millions of people live in areas where their drinking water resources are located near hydraulically
18    fractured wells. While most hydraulic fracturing activity from 2000 to 2013 did not occur in close
19    proximity to public water supplies, a sizeable number  of hydraulically fractured wells (21,900)
20    were located within 1 mile of at least one PWS source  (e.g., infiltration galleries, intakes, reservoirs,
21    springs and ground water wells). Approximately 6,800 sources of drinking water for public water
22    systems, serving more than 8.6 million people year-round, were located within 1 mile of at least one
23    hydraulically fractured well. An additional 3.6 million people obtain drinking water from private
24    systems in counties with at least one hydraulically fractured well and in which at least 30% of the
2 5    population is reliant on private water systems.
                    This document is a draft for review purposes only and does not constitute Agency policy.
      June 2015                                     3-11                  DRAFT—DO NOT CITE OR QUOTE

-------
Hydraulic Fracturing Drinking Water Assessment
Chapters - Drinking Water R
     Text Box 3-1. Major Findings
     Current and future drinking water resources

          •   Most of the U.S. population (270 million in 2011, or 86%) relies on water supplied to their
             homes through a public water system, 70% of which comes from surface water and 30% of
             which comes from ground water.
          •   An estimated 14% of the U.S. population relies on private water systems for drinking water.
          •   An increasing number of states are developing new drinking water supplies via reuse of
             reclaimed water, recycling of storm water, and desalination. These new supplies can augment
             existing water sources.
          •   Most of the U.S. population (270 million in 2011, or 86%) relies on water supplied to their
             homes through a public water system, 70% of which comes from surface water and 30% of
             which comes from ground water.
          •   An estimated 14% of the U.S. population relies on private water systems for drinking water.
          •   An increasing number of states are developing new drinking water supplies via reuse of
             reclaimed water, recycling of storm water, and desalination. These new supplies can augment
             existing water sources.
     Proximity of drinking water resources to hydraulic fracturing activity

          •   Nationwide, while most hydraulic fracturing activity from 2000 to 2013 did not occur in close
             proximity to public water supplies, a sizeable number of hydraulically fractured wells
             (21,900) were located within 1 mile of at least one PWS source.
          •   The distance between wells that were hydraulically fractured between 2000 and 2013 and the
             nearest source supplying a PWS ranged from 0.01 to 41 miles, with an average distance of 6.2
             miles (9.9 km).
          •   An estimated 6,800 public water system sources were located within 1 mile of a hydraulically
             fractured oil and gas well between 2000 and 2013. These PWS sources supplied water to
             3,924 public water systems and served more than 8.6 million people year-round in 2013.
          •   Approximately 9.4 million people lived within 1 mile of at least one hydraulically fractured oil
             and gas well between 2000 and 2013.
          •   Approximately 3.6 million people obtain drinking water from private systems in counties with
             at least one hydraulically fractured well and in which at least 30% of the population (i.e.,
             double the national average) is reliant on private water systems.
3.4.  References for Chapter 3
AWWA (American Water Works Association). (1999). Residential end uses of water. In PW Mayer; WB
   DeOreo (Eds.). Denver, CO: AWWA Research Foundation and American Water Works Association.
   http://www.waterrf.org/PublicReportLibrarv/RFR90781 1999 241A.pdf
               This document is a draft for review purposes only and does not constitute Agency policy.
June 2015                                       3-12                   DRAFT—DO NOT CITE OR QUOTE

-------
Hydraulic Fracturing Drinking Water Assessment                                                   Chapters - Drinking Water R


Castle. SL: Thomas. BF: Reager. IT: Rodell. M: Swenson. SC: Famiglietti. IS. (2014). Groundwater depletion
   during drought threatens future water security of the Colorado River Basin. Geophys Res Lett 41: 5904-
   5911. http://dx.doi.org/10.1002/2014GL061055

Cooley. H: Gleick. PH: Wolff. G. (2006). Desalination, with a grain of salt: A California perspective. Oakland, CA:
   Pacific Institute for Studies in Development, Environment, and Security, http://www.pacinst.org/wp-
   content/uploads/2013/02/desalination report3.pdf

Drillinglnfo, Inc.. (2014a). DI Desktop June 2014 download [Database]. Austin, TX: Drillinglnfo. Retrieved
   from http://info.drillinginfo.com/

Entrekin, S: Evans-White, M: lohnson, B: Hagenbuch, E. (2011). Rapid expansion of natural gas development
   poses a threat to surface waters. Front Ecol Environ 9: 503-511. http://dx.doi.org/10.1890/110053

ESRI (Environmental Systems Research Institute Inc.). (2010). US states shapefile. Redlands, California.
   Retrieved from http://www.arcgis.com/home/item.html?id=la6cae723afl4f9cae228bl33aebc620

Famiglietti. IS: Lo. M: Ho. SL: Bethune.  I: Anderson. Kl: Syed. TH: Swenson. SC: de Linage. CR: Rodell. M.
   (2011). Satellites measure recent rates of groundwater depletion in California's  Central Valley. Geophys
   Res Lett 38: L03403. http://dx.doi.org/10.1029/2010GL046442

Georgakakos. A: Fleming. P: Dettinger. M: Peters-Lidard. C: Richmond. TC: Reckhow. K: White. K: Yates. D.
   (2014). Water resources. In JM Melillo; TC Richmond; GW Yohe (Eds.), Climate change impacts in the
   United States (pp. 69-112). Washington, D.C.: U.S. Global Change Research Program.
   http://www.globalchange.gov/ncadac

Konikow. LF. (2013a). Groundwater depletion in the United States (1900-2008): U.S. Geological Survey
   Scientific Investigations Report 2013-5079. Reston, VA: U.S. Geological Survey.
   http://pubs.usgs.gov/sir/2013/5079

Maupin. MA: Kenny. IF: Hutson. SS: Lovelace. IK: Barber. NL: Linsey. KS. (2014). Estimated use of water in the
   United States in 2010. (USGS Circular 1405). Reston, VA: U.S. Geological Survey.
   http://dx.doi.org/10.3133/cirl405

National Drought Mitigation Center. (2015). U.S. drought monitor. Available online at
   http://droughtmonitor.unl.edu/Home.aspx (accessed February 27, 2015).

NRC (National Research Council). (2012). Water reuse: Potential for expanding the nations water supply
   through reuse of municipal wastewater. Committee on the Assessment of Water Reuse as an Approach for
   Meeting Future Water Supply Need. Washington, DC: The National Academies Press.
   http://www.nap.edu/openbook.php7record id=13303

Sheng. Z. (2005). An aquifer storage and recovery system with reclaimed wastewater to preserve native
   groundwater resources in El Paso, Texas. J Environ Manage 75: 367-377.
   http://dx.doi.0rg/10.1016/j.jenvman.2004.10.007

U.S. Census Bureau. (2010). Special release - census blocks with population and housing unit counts, 2010
   TIGER/Line shapefiles [Computer Program]. Suitland, MD: U.S. Census Bureau, Geography Division.
   Retrieved from https://www.census.gov/geo/maps-data/data/tiger-line.html

U.S. Census Bureau. (2013a). Annual estimates of the resident population: April 1, 2010 to July 1, 2013.
   Suitland, MD: U.S. Census Bureau, Population Division.
   http://factfinder2.census.gov/faces/tableservices/jsf/pages/productview.xh tml?src=bkmk

U.S. Census Bureau. (2013b). Cartographic boundary shapefiles metropolitan and micropolitan statistical
   areas and related statistical areas (Combined statistical areas, 500k). Suitland, MD. Retrieved from
   https://www.census.gov/geo/maps-data/data/cbf/cbf msa.html

U.S. Census Bureau. (2013c). Metropolitan and micropolitan statistical areas main. Available online at
   http://www.census.gov/population/metro/ (accessed January 12, 2015).
               This document is a draft for review purposes only and does not constitute Agency policy.

June 2015                                       3-13                    DRAFT—DO NOT CITE OR QUOTE

-------
Hydraulic Fracturing Drinking Water Assessment                                                   Chapters - Drinking Water R


U.S. EPA (U.S. Environmental Protection Agency). (1991). Manual of individual and non-public water supply
   systems [EPA Report]. (EPA 570/9-91-004). Washington, D.C.

U.S. EPA (U.S. Environmental Protection Agency). (2011b). Ground water cleanup at Superfund Sites [EPA
   Report]. (EPA 540-K-96 008). Washington, DC: U. S. Environmental Protection Agency, Office Water.
   http://www.epa.gov/superfund/health/conmedia/gwdocs/brochure.htm

U.S. EPA (U.S. Environmental Protection Agency). (2012e). Public drinking water systems: facts and figures.
   Washington, DC: U.S. Environmental Protection Agency, Office of Water.
   http://water.epa.gov/infrastructure/drinkingwater/pws/factoids.cfm

U.S. EPA (U.S. Environmental Protection Agency). (2013b). Drinking water and ground water statistics, fiscal
   year 2011. Washington, DC: U.S. Environmental Protection Agency, Office of Water.
   http://water.epa.gov/scitech/datait/databases/drink/sdwisfed/upload/epa816rl3003.pdf

U.S. EPA (U.S. Environmental Protection Agency). (2014J). Safe drinking water information system (SDWIS).
   Data obtained from the Office of Water [Database]. Washington, D.C.:  Office of Water. Retrieved from
   http://water.epa.gov/scitech/datait/databases/drink/sdwisfed/index.cfm

U.S. EPA (U.S. Environmental Protection Agency). (2015r). WaterSense: water supply in the U.S. Available
   online at http://www.epa.gov/WaterSense/pubs/supply.html (accessed January 12,2015).

U.S. GAP (U.S. Government Accountability Office). (2014). Freshwater: Supply concerns continue, and
   uncertainties complicate planning. Report to Congressional requesters. (GAO-14-430). Washington, DC:
   U.S. Government Accountability Office (GAO). http://www.gao.gov/assets/670/663343.pdf

U.S. Global Change Research Program. (2009). Global climate change impacts in the United States. New York,
   NY: Cambridge University Press, http://downloads.globalchange.gov/usimpacts/pdfs/climate-impacts-
   reportpdf

USGS (U.S. Geological Survey).  (2014b). Estimated use of water in the United States, county-level data for
   2010. Reston, VA. http://water.usgs.gov/watuse/data/2010/

Vengosh. A: lackson. RB: Warner. N: Darrah. TH: Kondash. A. (2014). A critical review of the risks to water
   resources from unconventional shale gas development and hydraulic fracturing in the United States.
   Environ Sci Technol 48: 36-52. http://dx.doi.org/10.1021/es405118v
               This document is a draft for review purposes only and does not constitute Agency policy.

June 2015                                       3-14                   DRAFT—DO NOT CITE OR QUOTE

-------
Hydraulic Fracturing Drinking Water Assessment
                            Chapter 4
Water Acquisition
           This document is a draft for review purposes only and does not constitute Agency policy.
June 2015                                              DRAFT—DO NOT CITE OR QUOTE

-------
      Hydraulic Fracturing Drinking Water Assessment
      4.  Water Acquisition
      4.1.  Introduction
 1    Water is a crucial component of nearly all hydraulic fracturing operations, making up
 2    approximately 90% or more of fluid injected into a well [U.S. EPA. 2015a: GWPG and ALL
 3    Consulting. 2009). Given that at least 25,000 to 30,000 wells maybe fractured each year
 4    (Chapter 2), and that each well requires thousands to millions of gallons of water (Section 4.3), the
 5    potential exists for effects on the quantity of drinking water resources. Large volume water
 6    withdrawals also could alter the quality of drinking water resources by decreasing dilution of
 7    pollutants by surface waters, or in the case of ground water, allowing the infiltration of lower-
 8    quality water from the land surface or adjacent formations.

 9    In this chapter, we consider potential effects of water acquisition for hydraulic fracturing on both
10    drinking water resource quantity and quality, and where possible, identify factors that affect the
11    frequency or severity of impacts. We define drinking water resources broadly, to include not just
12    currently designated drinking waters, but waters that could in the future be used as drinking water
13    sources (see Chapter 1). Although most available data and literature pertain to water use, we
14    discuss water consumption where possible.1

15    We provide an overview of the types of hydraulic fracturing water used (Section 4.2); the amount of
16    water used per well (Section 4.3); and cumulative water use  and consumption estimates
17    (Section 4.4).2 We then discuss these three factors for 15 states where hydraulic fracturing
18    presently occurs and consider the potential for hydraulic fracturing water withdrawals to affect
19    water quantity and quality in localities within those states (Section 4.5). We primarily discuss
20    results at the state and county level because data are most available at these scales. Moreover,
21    states and localities often differ in industry activity, formation type, and water availability,  all of
22    which affect potential impacts.3 Lastly, we provide a synthesis that summarizes major findings,
23    factors affecting the frequency or severity of impacts, uncertainties, and conclusions (Section 4.6).
       Water use is water withdrawn for a specific purpose, part or all of which may be returned to the local hydrologic cycle.
      Water consumption is water that is removed from the local hydrologic cycle following its use (e.g., via evaporation,
      transpiration, incorporation into products or crops, consumption by humans or livestock), and is therefore unavailable to
      other water users (Maupin et al.. 2014). Hydraulic fracturing water consumption can occur through evaporation from
      storage ponds, the retention of water in the subsurface through imbibition, or disposal in Underground Injection Control
      (UIC) Class II injection wells.
      2 In this chapter, cumulative annual water use or water consumption refers to the amount of water used or consumed by
      all hydraulic fracturing wells in a given area per year.
      3 There is no standard definition for water availability, and it has not been assessed recently at the national scale [U.S.
      GAP. 2014). Instead, a number of water availability indicators have been suggested [e.g.. Roy et al.. 2005). Here,
      availability is most often used to qualitatively refer to the amount of a location's water that could, currently or in the
      future, serve as a source of drinking water [U.S. GAP. 2014). which is a function of water inputs to a hydrologic system
      (e.g., rain, snowmelt, groundwater recharge) and water outputs from that system occurring either naturally or through
      competing demands of users. Where specific numbers are presented, we note the specific water availability indicator
      used.
                     This document is a draft for review purposes only and does not constitute Agency policy.
      June 2015                                       4-1                    DRAFT—DP NPT CITE PR QUPTE

-------
      Hydraulic Fracturing Drinking Water Assessment
      4.2.  Types of Water Used
 1    Water used for hydraulic fracturing generally comes from surface water (i.e., rivers, streams, lakes,
 2    and reservoirs), ground water aquifers, or reused hydraulic fracturing waste water.1'2*3 These
 3    sources can vary in their initial water quality and in how they are provisioned to hydraulic
 4    fracturing operations. In this section, we provide an overview of the sources (Section 4.2.1), water
 5    quality (Section 4.2.2), and provisioning of water (Section 4.2.3) required for hydraulic fracturing.
 6    Detailed information on the types of water used by state and locality is presented in Section 4.5.

      4.2.1.  Source
 7    Whether water used in hydraulic fracturing originates from surface or ground water resources is
 8    largely determined by the amount of water needed and the type of locally available water sources.
 9    Water transportation costs can be high, so the industry tends to acquire water from nearby sources
10    if available (Nicotetal.. 2014: Mitchell et al.. 2013a: Kargbo etal.. 2010). Surface water is typically
11    available to supply most of the water needed in the eastern United States, whereas mixed supplies
12    of surface and ground water are used in the more semi-arid to arid western states. In western
13    states that lack available surface water resources, ground water supplies the majority of water
14    needed for fracturing unless alternative sources, such as reused waste water, are available and
15    utilized. Local policies also may direct the industry to seek withdrawals from designated sources
16    (U.S. EPA, 2013a):  for instance, some states have encouraged the industry to seek water
17    withdrawals from  surface water rather than ground water due to concerns over aquifer depletion
18    (Section 4.5).

19    The reuse of waste water from past hydraulic fracturing operations can reduce the need for fresh
20    surface or ground water and offset total new water withdrawals for hydraulic fracturing.4*5 Based
21    on available data, the median reuse of wastewater as a percentage of injected volume is 5%
22    nationally, butthis percentage varies by location (Table 4-1).6-1
      i Throughout this chapter we sometimes refer to "reused hydraulic fracturing wastewater" as simply "reused
      wastewater" because this is the dominant type of wastewater reused by the industry. When referring to other types of
      reused wastewater not associated with hydraulic fracturing (e.g., acid mine drainage, wastewater treatment plant
      effluent) we specify the source of the wastewater.
      2 We use the term "reuse" regardless of the extent to which the wastewater is treated [Nicotetal.. 2014): we do not
      distinguish between reuse and recycling except when specifically reported in the literature.
      3 We use "wastewater" as a general term to include both flowback and produced water that may be reused in hydraulic
      fracturing; we do not distinguish between flowback and produced water except when specifically reported in the
      literature.
      4 Hydraulic fracturing wastewater may be stored on-site in open pits, which may also collect rainwater and runoff water.
      We do not distinguish between the different types  of water that are collected on-site during oil and gas operations, and
      assume that most of the water collected on-site at well pads is hydraulic fracturing wastewater.
      5 We use the term "fresh water" to qualitatively refer to water with relatively low TDS that is most readily and currently
available for drinking water. We do not use the term to imply an exact TDS limit.
6 Throughout this chapter, we preferentially report medians where possible be
values than averages. Where medians are not available, averages are reported.
      6 Throughout this chapter, we preferentially report medians where possible because medians are less sensitive to outlier
                     This document is a draft for review purposes only and does not constitute Agency policy.
      June 2015                                       4-2                    DRAFT—DO NOT CITE OR QUOTE

-------
      Hydraulic Fracturing Drinking Water Assessment
 1    The reuse of waste water for hydraulic fracturing is limited by the amount of water that returns to
 2    the surface during production [Nicotetal.. 2012]. In the first 10 days of well production, 5% to
 3    almost 50% of injected fluid volume may be collected, with values varying across geologic
 4    formations (see Chapter 7, Table 7-1). Longer duration measurements are rare, but between 10%
 5    and 30% of injected fluid volume has been collected in the Marcellus Shale in Pennsylvania over 9
 6    years of production, while over 100% has been collected in the Barnett Shale in north-central Texas
 7    over six years of production (see Chapter 7, Table 7-2). Assuming that 10% of injected fluid volume
 8    is collected in the first 30 days and the reuse rate is 100%, it would take 10  wells to produce
 9    enough water to hydraulically fracture a new well. As more wells are hydraulically fractured in a
10    given area, the potential for waste water reuse increases.

11    Besides hydraulic fracturing wastewater, other wastewaters may be reclaimed for use in hydraulic
12    fracturing. These may include acid mine drainage, wastewater treatment plant effluent, and other
13    sources of industrial and municipal wastewater (Nicotetal.. 2014: Ziemkiewicz etal.. 2013).
14    Limited information is available on the extent to which these other wastewaters are used.
      Table 4-1. Percentage of injected water volume that comes from reused hydraulic fracturing
                wastewater in various states, basins, and plays.
                States listed by order of appearance in the chapter. See Section 4.5 for additional discussion of reuse
                practices by state and locality and variation over time where data are available.
State, basin, or play
Available estimate
Year of estimate
(NA = not available)
      i This chapter examines reused wastewater as a percentage of injected volume because reused wastewater may offset
      total fresh water acquired for hydraulic fracturing. In contrast. Chapter 8 of this assessment discusses the total percentage
      of the generated wastewater that is reused rather than managed by different means (e.g., disposal in Class II injection
      wells). This distinction is sometimes overlooked, which sometimes leads to a misrepresentation of the extent to which
      wastewater is reused to offset total fresh water used for hydraulic fracturing.
                    This document is a draft for review purposes only and does not constitute Agency policy.
      June 2015                                      4-3                    DRAFT—DO NOT CITE OR QUOTE

-------
     Hydraulic Fracturing Drinking Water Assessment
State, basin, or play
Texas— Barn ett Shale
Texas— Eagle Ford Shale
Texas-TX-LA-MS Salt Basin"
Texas— Permian Basin (far west portion)
Texas— Permian Basin (Midland portion)
Texas— Anadarko Basin
Colorado— Garfield County, Uinta-Piceance Basin
Colorado— Wattenberg Field, Denver-Julesburg Basin
Pennsylvania — Marcellus Shale, Susquehanna River Basin
West Virginia— Marcellus Shale, Statewide
California— Monterey Shale, Statewide
Overall Mean'1
Overall Median1
Available estimate
5%a
0%a
5%a
0%a
2%a
20%a
100%c
0%d
18%e
15%f
4%g
15%
5%
Year of estimate
(NA= not available)
2011
2011
2011
2011
2011
2011
NA
NA
2012
2012
2014


      a Estimated percentage of recycling/reused water in 2011 (Nicot et al., 2012).
      b Nicot et al. (2012) refer to this region of Texas as the East Texas Basin.
      c Based on industry practices reported in U.S. EPA(2015c).
       Reflects an assumption of reuse practices by Noble Energy in the Wattenberg Field of Colorado's Denver-Julesburg Basin, as
      reported by Goodwin et al. (2014).
      e Volume of flowback injected as a percentage of total water injected, 2012 (Hansen et al., 2013). This is the most recent
      estimate available. For 2008 to 2011, reuse as a percentage of injected volume averaged 13%, with a median of 8%, according
      to U.S. EPA(2015c).
      f Reused fracturing water as a percentage of total water used for hydraulic fracturing, 2012, calculated from data provided by
      the West Virginia PEP (2014).
      5 Reported  data on planned hydraulic fracturing operations as described in 249 well stimulation notices submitted during the
      first half of January 2014 to CCST (2014). Of these notices, 4% indicated planned use of produced water (sometimes blended
      with fresh water) for fracturing, while  96% indicated planned use of only fresh water.
       The overall mean is not weighted by the number of wells in a given state, basin, or play.
      'The overall median is not weighted by the number of wells in a given state, basin, or play.
     4.2.2.  Quality
1    Water quality is an important consideration when sourcing water for hydraulic fracturing. Fresh
2    water is often preferred to maximize hydraulic fracturing fluid performance and to ensure
3    compatibility with the geologic formation being fractured. This finding is supported by the EPA's
4    analysis of disclosures to FracFocus 1.0 (hereafter the EPA FracFocus report) [U.S. EPA. 2015a], as
5    well as by regional analyses from Texas [Nicot etal., 2012] and the Marcellus [Mitchell etal.,
                     This document is a draft for review purposes only and does not constitute Agency policy.

     June 2015                                          4-4                      DRAFT—DO NOT CITE OR QUOTE

-------
      Hydraulic Fracturing Drinking Water Assessment
 1    2013aj.1-2 Fresh water was the most commonly cited water source by companies included in an
 2    analysis of nine hydraulic fracturing service companies on their operations from 2005 to 2010 [U.S.
 3    EPA. 2013a). Three service companies noted that the majority of their water was fresh because it
 4    required minimal testing and treatment [U.S. EPA, 2013a].3 The majority of the nine service
 5    companies recommended testing for certain water quality parameters (pH and maximum
 6    concentrations of specific cations and anions) in order to ensure compatibility among the water,
 7    other fracturing fluid constituents, and the geologic formation [U.S. EPA, 2013a].

 8    The reuse of hydraulic fracturing waste water may be limited by water quality. As a hydraulically
 9    fractured well ages, the waste water quality begins to resemble the water quality of the geologic
10    formation and may be characterized by high TDS [Goodwin etal.. 2014]. High concentrations of
11    TDS and other individual dissolved constituents in wastewater, including specific cations (calcium,
12    magnesium, iron, barium, strontium), anions (chloride, bicarbonate, phosphate, and sulfate), and
13    microbial agents, can interfere with hydraulic fracturing fluid performance by producing scale in
14    the wellbore or by interfering with certain chemical additives in the hydraulic fracturing fluid (e.g.,
15    high TDS may inhibit the effectiveness of friction reducers) (Gregory etal., 2011: North Dakota
16    State Water Commission, 2010). Due to these limitations, wastewater may require treatment to
17    meet the level of water quality desired in the hydraulic fracturing fluid formulation. Minimal
18    treatment or blending of wastewater and fresh water is sometimes done to dilute high TDS or other
19    constituents. Fresh water typically makes up the largest proportion of the base fluid when blended
20    with water sources of lesser quality (U.S. EPA. 2015a].4 However, direct reuse of wastewater with
21    minimal or no treatment is sometimes possible with higher-quality wastewater (U.S. EPA, 2015c]
22    (Section 4.5.2). No data are currently available to characterize the relative frequency of reuse done
2 3    with treatment, minimal treatment, or no treatment

      4.2.3.   Provisioning
24    Water for hydraulic fracturing is typically either self-supplied by the  industry or purchased from
25    public water systems.5 Self-supplied water for fracturing generally refers to permitted direct
      i FracFocus is a national hydraulic fracturing registry for oil and gas well operators to disclose information about
      hydraulic fracturing well locations, and water and chemical use during hydraulic fracturing operations developed by the
      Ground Water Protection Council and the Interstate Oil and Gas Compact Commission [U.S. EPA. 2015a). The registry was
      originally established in 2011 for voluntary reporting. However, six of the 20 states discussed in this assessment required
      disclosure to FracFocus at various points between January 1,2011 and February 28,2013, the time period analyzed by
      the EPA; another three of the 20 states offered the choice of reporting to FracFocus or the state during this same time
      period (see Appendix Table B-5 for states and disclosure start dates) [U.S. EPA. 2015a).
      2 Of all disclosures to FracFocus that indicated a source of water for the hydraulic fracturing base fluid, 68% listed "fresh"
      as the only source of water used. Note, 29% of all disclosures considered in the EPA's FracFocus report included
      information on the source of water used for the base fluid [U.S. EPA. 2015a).
      3 Service companies did not provide data on the percentage of fresh water versus non-fresh water used for hydraulic
      fracturing fU.S. EPA. 2015a1.
      4 In FracFocus disclosures indicating that fresh water was used in any combination with "recycled," "produced," or
      "brine," the median concentration of fresh water across all states ranged from 69% to 93% [U.S. EPA. 2015a).
      5 According to Section 1401(4) of the Safe Drinking Water Act, a public water system is defined as system that provides
      water for human consumption from surface or ground water through pipes or other infrastructure to at least 15 service
      connections, or an average of at least 25 people, for at least 60 days per year. Public water systems may either be publicly
      or privately owned.
                     This document is a draft for review purposes only and does not constitute Agency policy.
      June 2015                                       4-5                    DRAFT—DO NOT CITE OR QUOTE

-------
      Hydraulic Fracturing Drinking Water Assessment
 1    withdrawals from surface or ground water or the reuse of waste water. Nationally, self-supplied
 2    water is more common, although there is much regional variation [U.S. EPA. 2015a: CCST. 2014:
 3    Mitchell et al.. 2 013a: Nicotetal.. 2012). Public water systems encompass a variety of water
 4    suppliers [U.S. EPA, 2015c]. Water purchased from municipal public water systems can be provided
 5    either before or after treatment [Nicotetal., 2014]. Water for hydraulic fracturing is also
 6    sometimes purchased from smaller private entities, such as local land owners [Nicotetal., 2014].

      4.3.  Water  Use Per Well
 7    In this section, we provide an overview of the amount of water used per well during hydraulic
 8    fracturing. We discuss water use in the life cycle of oil and gas operations (Section 4.3.1], national
 9    patterns and associated variability (Section 4.3.2], as well as the factors affecting water use per well
10    including well  length, geology, and fracturing fluid formulation (Section 4.3.3]. More detailed state-
11    and locality-specific information on water use per well is provided in Section 4.5.

      4.3.1.  Hydraulic Fracturing Water Use in the Life Cycle of Oil and Gas
12    Water is needed throughout the life cycle of oil and gas production and use, including both at the
13    well for processes such as well pad preparation, drilling, and fracturing (i.e., the upstream portion],
14    and later for end uses such as electricity generation, home heating, or transportation (i.e., the
15    downstream portion] (Jiang etal.. 2014: Laurenzi and Tersey. 2013]. Most of the water used and
16    consumed in the upstream portion of the life cycle occurs during hydraulic fracturing (Jiang etal..
17    2014: Clark etal., 2013: Laurenzi and Tersey, 2013].* Water use per well estimates in this chapter
18    focus on hydraulic fracturing in the upstream portion of the oil and gas life cycle.2

      4.3.2.  National Patterns of Water Use Per Well for Fracturing
19    Hydraulic fracturing for oil and gas requires a large volume of water to create sufficient pressures.
20    According to the EPA's project database of disclosures to FracFocus 1.0 (hereafter the EPA
21    FracFocus project database], the median volume of water used per well, based on
       Laurenzi and lersey (2013) reported that hydraulic fracturing accounted for 91% of upstream water consumption,
      based on industry data for 29 wells in the Marcellus Shale. (91% was calculated from their paper by dividing hydraulic
      fracturing fresh water consumption (13.7 gal (51.9 L)/Megawatt-hour (MWh)) by total upstream fresh water
      consumption (15.0 gal (56.8 L)/MWh) and multiplying by 100). Similarly, liangetal. (2014) reported that 86% of water
      consumption occurred at the fracturing stage for the Marcellus, based on Pennsylvania Department of Environmental
      Protection (PA DEP) data on 500 wells. The remaining water was used in several upstream processes (e.g., well pad
      preparation, well drilling, road transportation to and from the wellhead, and well closure once production ended). Clark
      etal. (2013) estimated lower percentages (30%-80%) of water use at the fracturing stage for multiple formations.
      Although their estimates for the fraction of water used at the fracturing stage may be low due to their higher estimates for
      transportation and processing, the estimates by Clark etal. (2013) similarly illustrate the importance of the hydraulic
      fracturing stage in water use, particularly in terms of the upstream portion of the life cycle.
      2 When the full life cycle of oil and gas production and use is considered (i.e., both upstream and downstream water use),
      most water is used and consumed downstream. For example, in a life cycle analysis of hydraulically fractured gas used for
      electricity generation. Laurenzi and lersey (2013) reported that only 6.7% of water consumption occurred upstream
      (15.0 gal (56.8 L)/MWh), while 93.3% of fresh water consumption occurred downstream for power plant cooling via
      evaporation (209.0 gal (791.2 L)/MWh).
                     This document is a draft for review purposes only and does not constitute Agency policy.

      June 2015                                       4-6                    DRAFT—DO NOT CITE OR QUOTE

-------
      Hydraulic Fracturing Drinking Water Assessment
 1    37,796 disclosures nationally, was 1.5 million gal (5.7 million L) (U.S. EPA. 2015b1.1 There was
 2    substantial variability around this median, however, with 10th and 90th percentiles of 74,000 and 6
 3    million gal (280,000 and 23 million L) per well, respectively.2 Even in specific basins and plays,
 4    water use per well varied widely. Water injected also can vary within a single field; Laurenzi and
 5    Jersey (2013] reported volumes for the Wattenberg Field of the Niobrara play ranging from 1 to 6
 6    million gal (3.8 to 23 million L) per well (10th to 90th percentile).

      4.3.3  Factors Affecting Water Use Per Well
 7    Water use varies depending on many factors, including well length, geology, and the composition of
 8    the fracturing fluid.

 9    Well length: Well length is a principal driver of the amount of water used per well. Increases in well
10    length affect total water volumes injected primarily by allowing a larger fracture volume to be
11    stimulated (Economides etal.. 2013]. Fracture volume is the volume of the fractures in the geologic
12    formation that fill with hydraulic fracturing fluid. The total volume of injected fluid equals fracture
13    volume plus the volume of the wellbore itself, plus any fluid lost due  to "leakoff" or other
14    unintended losses.3 Thus, as wells get longer, fracture, well, and total volumes all increase. This is
15    particularly evident in longer horizontal wells versus vertical wells. For example, median water use
16    in horizontal gas wells was over 35 times higher than in vertical gas wells (2.9 million gal vs. 82,000
17    gal (11 million L vs. 310,000 L), respectively) between the years 2000 and 2010 (USGS. 20151.

18    Geology: Geologic characteristics also influence the amount of water  used per well. There are three
19    major formation types: shales, tight sands, and coalbeds (see Chapter 2).  Reported differences in
2 0    water use for shales versus tight sands are rare. However, Nicotetal. (2012) reported that total
21    water use in tight sand formations is less  than half of that of shale in  Texas, although results were
2 2    not reported per well.

23    In contrast to hydrocarbons from shales and tight sands, coalbed methane (CBM) comes from coal
24    seams that often have a high initial water content and tend to occur at much shallower depths (U.S.
25    EPA. 20151]. Thus, dewatering is often necessary to stimulate production of CBM. In addition,
26    geologic pressures are lower (leading to higher permeability] and well lengths are shorter, all of
2 7    which result in lower water use per well.  Water use per well in CBM  operations can be lower by an
28    order of magnitude or more compared to operations in shales or tight sands. For example, Murray
29    (2013] reported water use across formations in Oklahoma, and found that water use in the
30    CBM-dominated Hartshorn Formation was much lower than in the shale  gas-dominated Woodford
31    Formation.
      i Water use data from the EPA's FracFocus project database were obtained from disclosures made to FracFocus 1.0.
      Although disclosures were made on a per well basis, a small proportion of the wells were associated with more than one
      disclosure (i.e., 876 out of 37,114, based on unique API numbers) [U.S. EPA. 2015b). For the purposes of this chapter, we
      discuss water use per disclosure in terms of water use per well.
      2 Although the EPA FracFocus report shows 5th and 95th percentiles, we report 10th and 90th percentiles throughout this
      chapter to further reduce the influence of outliers.
      3 Leakoff is the fraction of the injected fluid that infiltrates into the formation (e.g., through an existing natural fissure) and
      is not recovered during production. See Chapter 6 for more information about leakoff.
                    This document is a draft for review purposes only and does not constitute Agency policy.
      June 2015                                     4-7                    DRAFT—DO NOT CITE OR QUOTE

-------
      Hydraulic Fracturing Drinking Water Assessment
 1    Fracturing Fluid Type: The majority of wells use fracturing fluids that consist mostly of water [U.S.
 2    EPA. 2015a: Yang etal.. 2013: GWPC and ALL Consulting. 20091. The EPA inferred that more than
 3    93% of reported disclosures to FracFocus used water as a base fluid [U.S. EPA. 2015a). The median
 4    reported concentration of water in the hydraulic fracturing fluid was 88% by mass, with 10th and
 5    90thpercentiles of 77% and 95%, respectively. Only roughly 2% of disclosures (761 wells) reported
 6    the use of non-aqueous substances as base fluids, typically either liquid-gas mixtures of nitrogen
 7    (643 disclosures, 84% of non-aqueous formulations) or carbon  dioxide (83 disclosures, 11% of
 8    non-aqueous formulations). Both of these formulations still contained substantial amounts of
 9    water, as water made up roughly 60% (median value) of fluid in them (U.S. EPA, 2015a). Other
10    formulations were rarely reported. Non-aqueous formulations are discussed further in Chapter 5.

      4.4. Cumulative Water Use and Consumption
11    In this section we provide an overview of cumulative water use  and consumption for hydraulic
12    fracturing at the national, state, and county scales. We then compare these values to total water use
13    and consumption. We discuss both use and consumption because hydraulic fracturing is both a user
14    and consumer of water. Water use refers to water withdrawn for a specific purpose, part or all of
15    which may be returned to the local hydrologic cycle. Water consumption refers to water that is
16    removed from the local hydrologic cycle following its use, and is therefore unavailable to other
17    users (Maupinetal.. 2014). Hydraulic fracturing water consumption can occur through such means
18    as evaporation from storage ponds, the retention of water in the subsurface through imbibition, or
19    disposal in UIC Class II injection wells. In the latter two cases, the water consumed is  generally
20    completely removed from the hydrologic cycle. In this section, water consumption estimates are
21    derived from USGS water use data, and therefore both use and consumption are presented with the
22    published water use numbers being first.

      4.4.1. National and  State Scale
23    Cumulatively, hydraulic fracturing uses and consumes billions of gallons  of water each year in the
24    United States, but at the national or state scale, it is a relatively small user (and consumer) of water
25    compared to total water use and consumption. According to the EPA's FracFocus project database,
26    hydraulic fracturing used 36 billion gal (136  billion L) of water in 2011, and 52 billion gal (197
27    billion L) in 2012; therefore, hydraulic fracturing used an annual average of 44 billion gal (167
28    billion L) of water in 2011 and 2012 across all 20 states in the project database (U.S. EPA. 2015a. b).
29    Cumulative national water use for hydraulic  fracturing can also  be estimated by multiplying the
30    water use per well by the number of wells hydraulically fractured. If the median water use per well
31    (1.5 million gal)  (5.7 million L) from the EPA's FracFocus project database is multiplied by 25,000
32    to 30,000 wells fractured annually (see Chapter 2), cumulative national water use for hydraulic
33    fracturing is estimated to range from 37.5 to 45.0 billion gal (142 to 170 billion L) annually. Other
34    calculated estimates have ranged higher than this, including estimates of approximately  80 billion
35    gal (300 billion L) fVengosh etal.. 20141 and 50-72 billion gal (190-273 billion L) fU.S. EPA. 2015cl.
36    These estimates are higher due to differences in the estimated water use per well and the number
37    of wells used as multipliers. For example, (Vengosh et al., 2014) derived the estimate of
38    approximately 80 billion gal (300 billion L) by multiplying an average of 4.0 million gal (15 million
                   This document is a draft for review purposes only and does not constitute Agency policy.
      June 2015                                    4-8                   DRAFT—DO NOT CITE OR QUOTE

-------
      Hydraulic Fracturing Drinking Water Assessment
 1    L) per well (estimated for shale gas wells) by 20,000 wells (the approximate total number of
 2    fractured wells in 2012).:

 3    All of these estimates of cumulative water use for hydraulic fracturing are small relative to total
 4    water use and consumption at the national scale. For example, in the combined 20 states where
 5    operators reported water use to FracFocus in 2011 and 2012 (U.S. EPA. 2015b), annual hydraulic
 6    fracturing water use and consumption averaged over those two years was less than 1% of total
 7    annual water use and consumption in 2010 (see Appendix Table B-l).2'3

 8    At the state scale, hydraulic fracturing also generally uses billions of gallons of water cumulatively,
 9    but accounts for a low percentage of total water use or consumption. Of all states, operators in
10    Texas used the most water cumulatively (47% of cumulative water use reported in the EPA
11    FracFocus project database) (U.S. EPA. 2015b) (see Appendix Table B-l). This was due to the large
12    number of wells in that state. Over 94% of reported cumulative water use occurred in just seven of
13    the 20 states in the EPA FracFocus project database: Texas, Pennsylvania, Arkansas, Colorado,
14    Oklahoma, Louisiana, and North Dakota (U.S. EPA. 2015b). Hydraulic fracturing is a small
15    percentage when compared to total water use (<1%) and consumption (<3%) in each individual
16    state (see Appendix Table B-l). Other studies have shown similar results, with hydraulic fracturing
17    water use and consumption ranging from less than 1% of total use in West Virginia (West Virginia
18    PEP. 2013). Colorado (Colorado Division of Water Resources: Colorado Water Conservation Board:
19    Colorado Oil and Gas Conservation Commission. 2014). and Texas (Nicotetal.. 2014: Nicot and
20    Scanlon. 2012). to approximately 4% in North Dakota (North Dakota State Water Commission.
21    2014).

      4.4.2.  County Scale
2 2    Cumulative water use and consumption for hydraulic fracturing is also relatively small in most, but
23    not all, counties in the United States (see Table 4-2, Figure 4-1, and Figure 4-2a,b). Reported
      1 This could result in an overestimation because the estimate of 20,000 wells was derived in part from FracFocus, and
      these wells are not necessarily specific to shale gas; they may include other types of wells that use less water (e.g., CBM).
      The estimate of 1.5 million gal (5.7 million L) per well based on the EPA FracFocus project database likely leads to a more
      robust estimate when used to calculate national cumulative water use for hydraulic fracturing because it includes wells
      from multiple formation types (i.e., shale, tight sand, and CBM), some of which use less water than shale gas wells on
      average (U.S. EPA.2015b).
      2 The USGS compiles water use estimates approximately every five years in the National Water Census including the 1995
      Census in Solley etal. (1998): 2005 Census in Kenny etal. (2009): and 2010 Census in Maupinetal. (2014). The 2010
      version is the most updated version available. The Census includes uses such as public supply, irrigation, livestock,
      aquaculture, thermoelectric power, industrial, and mining at the national, state, and county scale. The 2010 Census
      included hydraulic fracturing water use in the mining category; there was no designated category for hydraulic fracturing
      alone.
       Percentages were calculated by averaging annual water use for hydraulic fracturing in 2011 and 2012 for a given state
      or county (U.S. EPA. 2015b). and then dividing by 2010 USGS total water use (Maupinetal.. 2014) and multiplying by 100.
      Note, the annual hydraulic fracturing water use reported in FracFocus was not added to the 2010 total USGS water use
      value in the denominator, and is simply expressed as a percentage compared to 2010 total water use or consumption.
      This was done because of the difference in years between the two datasets, and because the USGS 2010 Census (Maupin
      etal.. 2014) included hydraulic fracturing water use estimates in their mining category. This approach is consistent with
      that of other literature on this topic; see Nicot and Scanlon (2012). See footnotes for Appendix Table B-l and Table 4-2 for
      description of the consumption estimate calculations.
                     This document is a draft for review purposes only and does not constitute Agency policy.
      June 2015                                       4-9                    DRAFT—DO NOT CITE OR QUOTE

-------
      Hydraulic Fracturing Drinking Water Assessment
 1    fracturing water use in FracFocus in 2011 and 2012 was less than 1% compared to 2010 USGS total
 2    water use in 299 of the 401 reporting counties [U.S. EPA. 2015b) (see Figure 4-2a and Appendix
 3    Table B-2). However, hydraulic fracturing water use was 10% or more compared to total water use
 4    in 26 counties, 30% or more in nine counties, and 50% or more in four counties (see Table 4-2 and
 5    Figure 4-2a). McMullen County in Texas had the highest percentage at over 100% compared to
 6    2010 total water use.1 Total consumption estimates followed the same pattern, but with more
 7    counties in the higher percentage categories (hydraulic fracturing water consumption was 10% or
 8    more compared to total water consumption in 53 counties; 30% or more in 25 counties; 50% or
 9    more in 16 counties; and over 100% in four counties) (see Table 4-2 and Figure 4-2b). Note,
10    estimates based on the EPA's FracFocus project database may form an incomplete picture of
11    hydraulic fracturing water use in a given state or county because the  majority of states with data in
12    the project database did not require disclosure to FracFocus during the time period analyzed (U.S.
13    EPA, 2015a]. We conclude that this likely does not substantially alter the overall patterns observed
14    in Figure 4-2a,b (see Text Box 4-1 for further details).

15    These percentages depend both upon the absolute water use and consumption for hydraulic
16    fracturing and the relative magnitude of other water uses and consumption in that state or county.
17    For instance, a rural county, with a small population, might have relatively low total water use prior
18    to hydraulic fracturing.2 Also, just because water is used in certain county does not necessarily
19    mean it originated in that county. While the cost of trucking water can be substantial (Slutz etal..
20    2012), and the industry tends to acquire water from nearby sources when possible (see Section
21    4.2.1), water can also be piped in from more distant, regional supplies. Despite these caveats, it is
22    clear that hydraulic fracturing is generally a relatively small user (or consumer) of water at the
23    county level, with the exception of a small number of counties where  water use and consumption
24    for fracturing can be high relative to other uses and consumption.
       Estimates of use or consumption exceeded 100% when hydraulic fracturing water use averaged for 2011 and 2012
      exceeded total water use or consumption in that county in 2010.
      2 For example, McMullen County, Texas mentioned above contains a small number of residents (707 people in 2010,
      according to the U.S. Census Bureau (2014)).
                   This document is a draft for review purposes only and does not constitute Agency policy.
      June 2015                                    4-10                  DRAFT—DO NOT CITE OR QUOTE

-------
Hydraulic Fracturing Drinking Water Assessment
Table 4-2. Annual average hydraulic fracturing water use and consumption in 2011 and 2012
           compared to total annual water use and consumption in 2010, by county.

           Only counties where hydraulic fracturing water was 10% or greater compared to 2010 total water use
           are shown  (for full table see Appendix Table B-2). Annual average hydraulic fracturing water use data
           in 2011 and 2012 from the EPA's FracFocus project database (U.S.  EPA. 2015b). Total annual water use
           data in 2010 from the USGS (Maupin et al., 2014). States listed by order of appearance in the chapter.
State
Texas
Pennsylvania
West Virginia
Ohio
County
McMullen
Karnes
La Salle
Dimmit
Irion
Montague
De Witt
Loving
San Augustine
Live Oak
Wheeler
Cooke
Susquehanna
Sullivan
Bradford
Tioga
Lycoming
Doddridge
Carroll
Total annual
water use in
2010 (millions
ofgal)3
657.0
1861.5
2474.7
4073.4
1335.9
3989.5
2394.4
781.1
1131.5
1916.3
6522.6
4533.3
1617.0
222.7
4354.5
2909.1
5854.6
405.2
1127.9
Annual average
hydraulic fracturing
water use in 2011
and 2012
(millions of gal)b
745.9
1055.2
1288.7
1794.2
411.4
925.3
546.6
138.4
182.1
294.0
858.0
454.3
751.3
66.5
1059.4
566.3
704.6
78.5
152.7
Hydraulic
fracturing
water use
compared to
total water use
(%)C
113.5
56.7
52.1
44.0
30.8
23.2
22.8
17.7
16.1
15.3
13.2
10.0
46.5
29.9
24.3
19.5
12.0
19.4
13.5
Hydraulic fracturing
water consumption
compared to total
water consumption
(%)c'd
350.4
120.1
93.7
81.3
74.5
77.8
48.6
94.1
50.8
40.1
21.5
29.9
123.4
79.8
78.2
47.3
33.8
69.4
37.3
              This document is a draft for review purposes only and does not constitute Agency policy.

June 2015                                      4-11                    DRAFT—DO NOT CITE OR QUOTE

-------
Hydraulic Fracturing Drinking Water Assessment
State
North Dakota
Arkansas
Louisiana
County
Mountrail
Dunn
Burke
Divide
Van Buren
Red River
Sabine
Total annual
water use in
2010 (millions
ofgal)3
1248.3
1076.8
394.2
806.7
1587.8
1606.0
1522.1
Annual average
hydraulic fracturing
water use in 2011
and 2012
(millions of gal)b
449.4
309.5
63.6
102.2
899.6
569.6
395.2
Hydraulic
fracturing
water use
com pa red to
total water use
(%)C
36.0
28.7
16.1
12.7
56.7
35.5
26.0
Hydraulic fracturing
water consumption
compared to total
water consumption
(%)c'd
98.3
43.1
40.8
18.6
168.8
83.2
76.6
 a County-level data accessed from the USGS website (http://water.usgs.gov/watuse/data/2010/) on November
 11, 2014. Total water withdrawals per day were multiplied by 365 days to estimate total water use for the year
 (Maupin etal., 2014).
 b Average of water used for hydraulic fracturing in 2011 and 2012 as reported to FracFocus (U.S. EPA, 2015b).
 c Percentages were calculated by averaging annual water use for hydraulic fracturing reported in FracFocus in
 2011 and 2012 for a given state or county (U.S. EPA. 2015b), and then dividing by 2010 USGS total water use
 (Maupin etal., 2014) and multiplying by 100.
 d Consumption values were calculated with use-specific consumption rates predominantly from the USGS,
 including 19.2% for public supply, 19.2% for domestic use, 60.7% for irrigation, 60.7% for livestock, 14.8% for
 industrial uses, 14.8% for mining (Solley et al., 1998), and 2.7% for thermoelectric power (USGS, 2014h). We used
 rates of 71.6% for aquaculture (from Verdegem and Bosma, 2009) (evaporation per kg fish + infiltration per
 kg)/total water use per kg); and 82.5% for hydraulic fracturing (consumption value calculated by taking the
 median value for all reported produced water/injected water percentages in Tables 7-1 and 7-2 of this
 assessment and then subtracting from 100%). If a range of values was given, the midpoint was used. Note, this
 aspect of consumption is likely a low estimate since much of this produced water (injected water  returning to the
 surface) is not subsequently treated and reused, but rather disposed of in UIC Class  II injection wells—see
 Chapter 8).
                This document is a draft for review purposes only and does not constitute Agency policy.

June 2015                                        4-12                     DRAFT—DO NOT CITE OR QUOTE

-------
Hydraulic Fracturing Drinking Water Assessment
                                                                                  Chapter 4 - Water Acquisition
         Annual Average
         Water Use in Mgal
         (number of counties)
         ^| > 500 (27)
         ^| 100-500(60)
         ^ 10-100(86)
             1-10(115)
             ElAshale basins
n             State reporting
             requirement

Figure 4-1. Annual average hydraulic fracturing water use in 2011 and 2012 by county (U.S. EPA, 2015b).
           Source: (U.S. EPA, 2015b). Water use in millions of gallons (Mgal). Counties shown with respect to major U.S. Energy Information Administration
           (EIA) shale basins (EIA, 2015b). Orange borders identify states that required some degree of reporting to FracFocus 1.0 in 2011 and 2012.
June 2015
This document is a draft for review purposes only and does not constitute Agency policy.
                                 4-13
DRAFT—DO NOT CITE OR QUOTE

-------
Hydraulic Fracturing Drinking Water Assessment
                  Hydraulic fracturing water
                  consumption compared to total wa
                  consumption (number of counties)
Figure 4-2. (a) Annual average hydraulic fracturing water use in 2011 and 2012 compared to
           total annual water use in 2010, by county, expressed as a percentage; (b) Annual
           average hydraulic fracturing water consumption in 2011 and 2012 compared to
           total annual water consumption in 2010, by county, expressed as a  percentage.

           Annual average hydraulic fracturing water use data in 2011 and 2012 from the EPA's FracFocus project
           database (U.S. EPA. 2015b). Total annual water use data in 2010 from the USGS (Maupin et al.. 2014).
           See Table 4-2 for descriptions of calculations for estimating consumption. Counties shown with
           respect to major U.S. EIA shale basins (EIA, 2015b). Orange borders identify states that required some
           degree of reporting to FracFocus 1.0 in 2011 and 2012. Note: Values over 100% denote counties
           where the annual average hydraulic fracturing water use or consumption in 2011 and 2012 exceeded
           the total annual water use or consumption in that county in 2010.
               This document is a draft for review purposes only and does not constitute Agency policy.

June 2015                                      4-14                   DRAFT—DO NOT CITE OR QUOTE

-------
       Hydraulic Fracturing Drinking Water Assessment
       Text Box 4-1. Using the EPA's FracFocus Project Database to Estimate Water Use for Hydraulic
       Fracturing.

 1     FracFocus is a national hydraulic fracturing registry managed by the Ground Water Protection Council and the Interstate
 2     Oil and Gas Compact Commission (GWPC.2015). The registry was established in 2011 for voluntary reporting. However,
 3     six of the 20 states discussed in this assessment required disclosure to FracFocus at various points between January 1,
 4     2011 and February 28,2013, the time period analyzed by the EPA; another three of the 20 states offered the choice of
 5     reporting to FracFocus or the state during this same time period [U.S. EPA. 2015a). Estimates based on the EPA's
 6     FracFocus project database likely form an incomplete picture of hydraulic fracturing water use because most states with
 7     data in the project database (14 out of 20) did not require disclosure to FracFocus during the time period analyzed [U.S.
 8     EPA. 20153).

 9     Cumulative water use for fracturing is a function of the water use per well and the total number of wells fractured. For
10     water use per well, we found seven literature values for comparison with values from the EPA's FracFocus project
11     database. On average, water use estimates per well in the project database were 77% of literature values (the median was
12     86%); Colorado's Denver Basin was the only location where the project database estimate as a percentage of the
13     literature estimate was low (14%) (see Appendix Table B-3). In general, water use per well estimates from the EPA's
14     FracFocus project database appear to provide a reasonable approximation for most areas for which we have data, with
15     the exception of the Denver Basin of Colorado.

16     For number of wells, we compared data in the EPA's FracFocus project database to numbers available in state databases
17     from North Dakota, Pennsylvania, and West Virginia (see Appendix Table B-4). These were the state databases from
18     which we could distinguish hydraulically fractured wells from total oil and gas wells. On average, we found that the EPA
19     FracFocus project database included 67% of the wells listed in state databases for 2011 and 2012 (see Appendix Table B-
20     4). Unlike North Dakota and Pennsylvania, West Virginia did not require operators to report fractured wells to FracFocus
21     during this time period, possibly explaining its lower reporting rate. Multiplying the average EPA FracFocus project
22     database values of77% for water use per well and 67% for well counts yields 52%. Thus, the EPA FracFocus project
23     database estimates for water use could be slightly over half of the estimates from these three state databases during this
24     time period. These values are based on a small sample sizes (7 literature values and 3 state databases) and should be
2 5     interpreted with caution. Nevertheless, these numbers at the very least suggest that estimates based on the EPA's
26     FracFocus project database may form an incomplete picture of hydraulic fracturing water use during this time period.

27     To assess how this might affect hydraulic fracturing water use estimates in this chapter, we doubled the water use value
28     in the EPA's FracFocus project database for each county, an adjustment much higher than any likely underestimation.
29     Even with this adjustment, fracturing water use was still less than 1% of 2010 total water use in the majority of U.S.
30     counties (299 counties without adjustment versus 280 counties with adjustment). The number of counties where
31     hydraulic fracturing water use was 30% or more of 2010 total county water use increased from nine to 21 with the
32     adjustment.

3 3     These results indicate that most counties have relatively low hydraulic fracturing water use, relative to total water use,
34     even when accounting for likely underestimates. Since consumption estimates are derived from use, these will also follow
35     the same pattern. Thus, potential underestimates based on the EPA's FracFocus project database likely do not
36     substantially alter the overall pattern shown in Figure 4-2. Rather, underestimates of hydraulic fracturing water use
37     would mostly affect the percentages in the small number of counties where fracturing already constitutes a higher
38     percentage of total water use and consumption.


       4.5.  Potential for Water Use Impacts by  State
39     High fracturing water use or consumption alone does not necessarily result in impacts to drinking
40     water resources. Rather, impacts most often result from the combination of water use and water
41     availability at a given withdrawal point Where water availability is high compared to water
                       This document is a draft for review purposes only and does not constitute Agency policy.
       June 2015                                          4-15                     DRAFT—DO NOT CITE OR QUOTE

-------
      Hydraulic Fracturing Drinking Water Assessment
 1    withdrawn for hydraulic fracturing, this water use can be accomodated. However, where water
 2    availability is low compared to use, hydraulic fracturing withdrawals are more likely to impact
 3    drinking water resources. Water management, such as the type of water used or the timing or
 4    location of withdrawals, can modify this relationship. All of these factors can vary considerably by
 5    location.

 6    Besides potential water quantity effects, water withdrawals for hydraulic fracturing have the
 7    potential to alter the quality of drinking water resources. This possibility is not unique to the oil and
 8    gas industry, as any large-volume water withdrawal has the potential to affect water quality.
 9    Although there is little research that specifically connects water withdrawals for hydraulic
10    fracturing to potential water quality impacts, multiple studies have described the impact of drought
11    or industrial withdrawals on water quality [Georgakakos etal., 2014: Whitehead etal., 2009:
12    Murdoch etal.. 2000: Schindler. 1997}. For instance, in the absence of controls, surface water
13    withdrawals can lower water levels and alter stream flows, potentially decreasing a stream's
14    capacity to dilute contaminants [Mitchell etal.. 2013a: Entrekin etal.. 2011: NYSDEC. 2011: van
15    VlietandZwolsman. 2008: IPCC. 2007: Environment Canada. 2004: Murdoch etal.. 20001.
16    Furthermore, ground water withdrawals exceeding natural recharge rates may lower the water
17    level in aquifers, potentially mobilizing contaminants or allowing the infiltration of lower-quality
18    water from the land surface or adjacent formations [USGS, 2003: Jackson etal., 2001].

19    In the following section, we assess the potential for water quantity and quality impacts by location,
20    organized by state. We focus our discussion on the 15 states that account for almost all disclosures
21    reported in the EPA FracFocus project database [U.S. EPA. 2015b]: Texas (Section 4.5.1); Colorado
22    and Wyoming (Section 4.5.2);  Pennsylvania, West Virginia, and Ohio (Section 4.5.3); North Dakota
23    and Montana (Section 4.5.4); Oklahoma and Kansas (Section 4.5.5); Arkansas and Louisiana
24    (Section 4.5.6); and Utah, New Mexico, and California (Section 4.5.7).l Each section describes the
25    extent of hydraulic fracturing activity in that state or group of states; the type of water used in
26    terms of source, quality, and provisioning; and the water use per well. We then discuss cumulative
27    estimates and the potential for impacts to drinking water resources in the context of water
28    availability.

29    We have ordered the states by the number of hydraulically fractured wells reported, and combined
30    states with similar geographies or activity. Most of the available data did not allow us to assess the
31    potential for impacts at a finer resolution than the county scale. Any potential adverse impacts are
32    most likely to be observed locally at a particular withdrawal point Therefore, our analysis most
33    often suggests where the potential for impacts exists, but does not indicate where impacts will
34    occur at the local scale. Where possible, we utilize local-scale case studies in southern Texas,
35    western Colorado, and eastern Pennsylvania to provide details at a much finer resolution, and offer
36    insight into whether any impacts from water acquisition for hydraulic fracturing were realized in
37    these areas.
      1 We do not highlight the remaining five states included in the EPA FracFocus project database because of low reported
      activity: Virginia (90 disclosures), Alabama (55), Alaska (37), Michigan (15), and Mississippi (4).
                    This document is a draft for review purposes only and does not constitute Agency policy.
      June 2015                                     4-16                  DRAFT—DO NOT CITE OR QUOTE

-------
     Hydraulic Fracturing Drinking Water Assessment
     4.5.1.  Texas
1    Hydraulic fracturing in Texas accounts for the bulk of the activity reported nationwide, comprising
2    48% of the disclosures in the EPA FracFocus project database fU.S. EPA. 2015bl (see Figure 4-3 and
3    Appendix Table B-5). There are five major basins in Texas: the Permian, Western Gulf (includes the
4    Eagle Ford play), Fort Worth (includes the Barnett play), TX-LA-MS Salt (includes the Haynesville
5    play), and the Anadarko (see Figure 4-4); together, these five basins contain 99% of Texas' reported
6    wells (see Appendix Table B-5).
                                                             Black Warrior
                                                                S^
                                                                  ^
     Figure 4-3. Locations of wells in the EPA FracFocus project database, with respect to U.S. EIA
               shale plays and basins (EIA, 2015; US. EPA, 2015b).
               Note: Hydraulic fracturing is conducted in geologic settings other than shale; therefore, some wells on
               this map are not associated with any EIA shale play or basin. (EIA. 2015b: U.S. EPA. 2015b).
                   This document is a draft for review purposes only and does not constitute Agency policy.
     June 2015                                     4-17                   DRAFT—DO NOT CITE OR QUOTE

-------
      Hydraulic Fracturing Drinking Water Assessment
                                                                        TX-LA-MS
                                                                        Salt Marsh
                                                                    HavnesviUe-Bossitr
                                                                           Play
      Figure 4-4. Major U.S. EIA shale plays and basins for Texas (EIA, 2015).
                Source: (EIA, 2015b)

 1    Types of water used: What is known about water sources in Texas largely comes from direct surveys
 2    and interviews with industry operators and water suppliers [Nicot et al., 2014; Nicot et al., 20121.
 3    Overall, ground water is the dominant source throughout most of the state [Nicot et al., 2014; Nicot et
 4    al., 2012] (see Table 4-3). The exception is the Barnett Shale, where both surface and ground water
 5    are used in approximately equal proportions.

 6    Hydraulic fracturing in Texas uses mostly fresh water [Nicot etal.. 2012).* The exception is the far
 7    western portion of the Permian Basin, where brackish water makes up an estimated 80% of total
 8    hydraulic fracturing water use. Brackish water is used to a lesser extent in the Anadarko Basin and
 9    the Midland portion of the Permian Basin (see Table 4-4). Reuse of wastewater as a percentage of
10    total water injected is generally very low (5% or less) in all major basins and plays in Texas, except
11    for the Anadarko Basin in the Texas Panhandle, where it is 20% (Nicot etal.. 2012] (see Table 4-1).
      1 The EPA FracFocus report shows that "fresh" was the only source of water listed in 91% of all disclosures reporting a
      source of water in Texas [U.S. EPA. 2015a). However, 19% of Texas disclosures included information related to water
      sources [U.S. EPA.2015a).
                    This document is a draft for review purposes only and does not constitute Agency policy.
      June 2015                                     4-18                   DRAFT—DO NOT CITE OR QUOTE

-------
Hydraulic Fracturing Drinking Water Assessment
Table 4-3. Estimated proportions of hydraulic fracturing source water from surface and
            ground water.
            States listed by order of appearance in the chapter.
Location
Texas— Barn ett Shale
Texas— Eagle Ford Shale
Texas-TX-LA-MS Salt Basin0
Texas— Permian Basin
Texas— Anadarko Basin
Pennsylvania — Marcellus Shale, Susquehanna River Basin
West Virginia— Statewide, Marcellus Shale
Oklahoma— Statewide
Louisiana— Haynesville Shale
Surface water
50%a
10%b
30%b
0%b
20%b
78%d
91%e
63%f
87%g
Ground water
50%a
90%b
70%b
100%b
80%b
22%d
9%e
37%f
13%g
 3 Nicotetal. (2014).
 bNicotetal. (2012).
 c Nicotetal. (2012) refer to this region of Texas as the East Texas Basin.
 d Estimated proportions are for 2011 (U.S. EPA, 2015c).
 e Estimated proportions are for 2012, the most recent estimate for a full calendar year available from West Virginia PEP
 (2014). Data from the West Virginia DEP show the proportion of water purchased from commercial brokers as a separate
 category and do not specify whether purchased water originated from surface or ground water. Therefore, we excluded
 purchased water in calculating the relative proportions of surface and ground water shown in Table 4-3 (West Virginia DEP.
 2014).
 f Proportion of surface and ground water permitted in 2011 by Oklahoma's 90-day provisional temporary permits for oil and
 gas mining. Temporary permits make up the majority of water use permits for Oklahoma oil and gas mining (Taylor. 2012).
 g Data from October 1, 2009, to February 23, 2012, for 1,959 Haynesville Shale natural gas wells (LA Ground Water Resources
 Commission. 2012).
                This document is a draft for review purposes only and does not constitute Agency policy.

June 2015                                           4-19                     DRAFT—DO NOT CITE OR QUOTE

-------
      Hydraulic Fracturing Drinking Water Assessment
      Table 4-4. Brackish water use as a percentage of total hydraulic fracturing water use in Texas'
                 main hydraulic fracturing areas, 2011.
                 Adapted from Nicot et al. (2012).a
Play
Barnett Shale
Eagle Ford Shale
Texas portion of the TX-LA-MS Salt Basin"
Permian Basin— Far West
Permian Basin— Midland
Anadarko Basin
Percent
3%
20%
0%
80%
30%
30%
       a Nicot et al. (2012) present the estimated percentages of brackish, recycled/reused, and fresh water relative to total
       hydraulic fracturing water use so that the percentages of the three categories sum to 100%.
        Nicot et al. (2012) refer to this region of Texas as the East Texas Basin.
 1    The majority of water used in Texas for hydraulic fracturing is self-supplied via direct ground or
 2    surface water withdrawals [Nicot etal., 2014]. Less often, water is purchased from local
 3    landowners, municipalities, larger water districts, or river authorities [Nicotetal.. 2014).

 4    Water use per well: Water use per well varies across Texas' basins, with reported medians of
 5    3.9 million gal (14.8 million L) in the Fort Worth Basin, 3.8 million gal (14.4 million L) in the
 6    Western Gulf, 3.3 million gal (12.5 million L) in the Anadarko, 3.1 million gal (11.7 million L) in the
 7    TX-LA-MS Salt, and 840,000 gal (3.2 million L) in the Permian (see Appendix Table B-5). Relatively
 8    low water use in the Permian Basin, which contains roughly half the reported wells in the state, is
 9    due to the abundance of vertical wells, mostly for oil extraction (Nicotetal., 2012].

10    Water use per well is increasing in most locations in Texas. In the Barnett Shale, water use per
11    horizontal well increased from a median of 1.25  million gal (4.73 million L) in 2001 to 4.7 million
12    gal (17.8 million L) in 2012, as the number of wells and horizontal lengths increased (Nicotetal.,
13    2014). Similar increases in lateral length and water use per well were reported for the
14    Texas-Haynesville, East Texas, Anadarko, and most of the Permian Basin (Nicotetal., 2012: Nicot
15    andScanlon. 2012].1
      1 It should be noted that energy production also increases with lateral lengths, and therefore, water use per unit energy
      produced—typically referred to as water intensity—may remain the same or decline despite increases in per-well water
      use [Nicot etal.. 2014: Laurenzi and lersey. 2013).
                    This document is a draft for review purposes only and does not constitute Agency policy.
      June 2015                                      4-20                   DRAFT—DO NOT CITE OR QUOTE

-------
      Hydraulic Fracturing Drinking Water Assessment
 1    Cumulative water use/consumption: Cumulative water use and consumption for hydraulic fracturing
 2    can be significant in some Texas counties. Texas contains five of nine counties nationwide where
 3    operators used more than 1 billion gal (3.8 billion L) of water annually for hydraulic fracturing, and
 4    five of nine counties nationwide where fracturing water use in 2011 and 2012 was 30% or more
 5    compared to total water use in those counties in 2010 (see Table 4-2, Figure 4-2a, and Appendix
 6    Table B-21'1'2

 7    According to detailed county-level projections, water use for hydraulic fracturing is expected to
 8    increase with oil and gas production in the coming decades, peaking around the year 2030 (Nicotet
 9    al.. 2012). The majority of counties are expected to have relatively low cumulative water use for
10    fracturing in the future, but cumulative hydraulic fracturing water use could equal or exceed 10%,
11    30%, and 50% compared to 2010 total county water use in 30, nine, and three counties,
12    respectively, by 2030 (see Appendix Table B-7). Thus, potential hydraulic fracturing water
13    acquisition impacts in Texas may be most likely to occur over the  next 15-25 years as water
14    demand for fracturing is highest

15    Potential for impacts: Of all locations surveyed in this chapter, the potential for water quantity and
16    quality impacts due to hydraulic fracturing water use appears to be highest in western and
17    southern Texas. This area includes the Anadarko, the Western Gulf (Eagle Ford play), and the
18    Permian Basins. According to Ceres (2014], 28% and 87% of the wells fractured in the Eagle Ford
19    play and Permian Basin, respectively, are in areas of high to extremely high water stress.3 A
20    comparison of hydraulic fracturing water use to water availability at the county scale also suggests
21    the potential for impacts (see Text Box 4-2 and Figure 4-5). The Texas Water Development Board
22    estimates that overall demand for water (including water for hydraulic fracturing) out to the year
23    2060 will outstrip supply in southern and western Texas (TWDB,  2012). Moreover, the state has
24    experienced moderate  to extreme drought conditions for much of the last decade (National Drought
25    Mitigation Center. 2015). The 2012 Texas State Water Plan emphasizes that "in serious drought
26    conditions, Texas does not and will not have enough water to meet the needs of its people, its
27    businesses, and its agricultural enterprises" (TWDB. 2012).

28
      1 Texas also contains 10 of the 25 counties nationwide where hydraulic fracturing water consumption was greater than or
      equal to 30% of 2010 total water consumption (see Table 4-2).
      2 Nicotand Scanlon (2012) found similar variation among counties when they compared hydraulic fracturing water
      consumption to total county water consumption for the Barnett play. Their cumulative consumption estimates ranged
      from 581 million gal (2.20 billion L) in Parker County to 2.7 billion gal (10.2 billion L) in Johnson County, representing
      19.3% and 29.7% compared to total water consumption in those counties, respectively. Fracturing in Tarrant County, part
      of the Dallas-Fort Worth area, consumed 1.6 billion gal (6.1 billion L) of water, 1.4% compared to total county water
      consumption (Nicot and Scanlon. 2012).
      3 Ceres (2014) compared well locations to areas categorized by a water stress index, characterized as follows: extremely
      high (defined as annual withdrawals accounting for greater than 80% of surface flows); high (40-80% of surface flows);
      or medium-to-high (20-40% of surface flows).
                    This document is a draft for review purposes only and does not constitute Agency policy.
      June 2015                                     4-21                   DRAFT—DO NOT CITE OR QUOTE

-------
      Hydraulic Fracturing Drinking Water Assessment
      Text Box 4-2. Hydraulic Fracturing Water Use as a Percentage of Water Availability Estimates.

 1    Researchers at Sandia National Laboratories assessed county-level water availability across the continental
 2    United States [Tidwell et al.. 2013]. Assessments of water availability in the United States are generally
 3    lacking at the county scale, and this analysis—although undertaken for siting new thermoelectric power
 4    plants—can be used to assess potential impacts of hydraulic fracturing.

 5    The authors generated annual availability estimates for five categories of water: unappropriated surface
 6    water, unappropriated ground water, appropriated water potentially available for purchase, brackish
 7    groundwater, and wastewater from municipal treatment plants (Tidwell et al., 20131. In the western United
 8    States, water is generally allocated by the principle of prior appropriation—that is, first in time of use is first
 9    in right. New development must use unappropriated water or purchase appropriated water from vested
10    users. In their analysis, the authors assumed 5% of appropriated irrigated water could be purchased; they
11    also excluded wastewater required to be returned to streams and the wastewater fraction already reused.
12    Given regulatory restrictions, they considered no fresh water to be available in California for new
13    thermoelectric plants.

14    Combining their estimates of unappropriated surface and ground water and appropriated water potentially
15    available for purchase, we  derived a fresh water availability estimate for each county (except for those in
16    California)  and then compared this value to reported water use for hydraulic fracturing fU.S. EPA. 2015b). We
17    also added the estimates of brackish and wastewater to fresh water estimates to derive estimates of total
18    water availability and did a similar comparison. Since the water availability estimates already take into
19    account current water use for oil and gas operations, these results should be used only as indicator of areas
2 0    where shortages might arise in the future.

21    Overall, hydraulic fracturing water use represented less than 1% of fresh water availability in over 300 of the
22    395 counties analyzed (see Figure 4-5a). This result suggests that there is ample water available at the county
23    scale to accommodate hydraulic fracturing in most locations. However, there was a small number of counties
24    where hydraulic fracturing water use was a relatively high percentage of fresh water availability. In 17
25    counties, fracturing water  use actually exceeded the index of fresh water  available; all of these counties were
26    located in the state of Texas and were associated with the Anadarko, Barnett, Eagle Ford, and Permian
27    basins/plays (see Figure 4-4). In Texas counties with relatively high brackish water availability, hydraulic
28    fracturing water use represented a much smaller percentage of total water availability (fresh + brackish +
29    wastewater) (see Figure 4-5b). This finding illustrates that potential impacts can be avoided or reduced in
30    these counties through the use of brackish water or wastewater for hydraulic fracturing; a case study in the
31    Eagle Ford play in southwestern Texas confirms this (see Text Box 4-3).
                     This document is a draft for review purposes only and does not constitute Agency policy.
      June 2015                                        4-22                    DRAFT—DO NOT CITE OR QUOTE

-------
Hydraulic Fracturing Drinking Water Assessment
  Text Box 4.2 (continued): Hydraulic Fracturing Water Use as a Percentage of Water
               Availability Estimates.
  Figure 4-5. Annual average hydraulic fracturing water use in 2011 and 2012 compared to
            (a) fresh water available and (b) total water (fresh, brackish, and wastewater)
            available, by county, expressed as a percentage.

            Counties shown with respect to major U.S. EIA shale basins (EIA, 2015b). Orange borders identify
            states that required some degree of reporting to FracFocus 1.0 in 2011 and 2012. Data from U.S.
            EPA(2015b) and Tidwell et al. (2013); data from Tidwell et al. (2013) supplied from the U.S.
            Department of Energy (DOE) National Renewable Energy Laboratory on January 28, 2014 and
            available upon request from the U.S. DOE Sandia National Laboratories. The analysis by Tidwell et
            al. (2013) was done originally for thermoelectric power generation. As such, it was assumed that
            no fresh water could be used in California for this purpose due to regulatory restrictions, and
            therefore no fresh water availability data were given for California (a). The total water available
            for California is the sum of brackish water plus wastewater only (b).
               This document is a draft for review purposes only and does not constitute Agency policy.

June 2015                                       4-23                    DRAFT—DO NOT CITE OR QUOTE

-------
      Hydraulic Fracturing Drinking Water Assessment
 1    Surface water availability is generally low in western and southern Texas (Figure 4-6a), and both
 2    fracturing operations and residents rely heavily on ground water (Figure 4-6b). Similar to trends
 3    nationally, ground water aquifers in Texas have experienced substantial declines caused by
 4    withdrawals (Konikow. 2013b: TWDB. 2012: George etal.. 20111. Ground water in the Pecos Valley,
 5    Gulf Coast, and Ogallala aquifers in southern and western Texas is estimated to have declined by
 6    roughly 5,11, and 43 cubic miles (21, 45.5, and 182 cubic kilometers), respectively, between 1900
 7    and 2008 (Konikow, 201 Sb].1 The Texas Water Development Board expects ground water supply in
 8    the major aquifers to decline by 30% between 2010  and 2060, mostly due to declines in the
 9    Ogallala aquifer (TWDB, 2012].2 Irrigated agriculture is by far the dominantuser of water from the
10    Ogallala aquifer (USGS, 2009], but fracturing operations, along with other uses, now contribute to
11    the aquifer's depletion.
      i The estimate of total net volumetric groundwater depletion for the Gulf Coast aquifer is the sum of the individual
      depletion estimates for the north (Houston area), central, and southern (Winter Garden area) parts of the Texas Gulf
      Coast aquifer. Ground water depletion from the Carrizo-Wilcox aquifer is included in the estimate for the southern
      portion of the Gulf Coast aquifer (Konikow. 2013b).
      2 TWDB (2012) defines ground water supply as the amount of ground water that can be produced given current permits
      and existing infrastructure. By contrast, TWDB (2012) defines ground water availability as the amount of ground water
      that is available regardless of legal or physical availability. Total ground water availability in Texas is expected to  decline
      by approximately  24% between 2010 and 2060 (TWDB. 2012).
                     This document is a draft for review purposes only and does not constitute Agency policy.

      June 2015                                      4-24                   DRAFT—DO NOT CITE OR QUOTE

-------
Hydraulic Fracturing Drinking Water Assessment
               Ratio of Ground Water Pumping
               to Stream Flow + Pumping
                  <0.12    m 0.59-0.70
0,12-0.25

0,25-0.37

0,37-0.59
                          0.70-1.00

                          No Data
Figure 4-6. (a) Estimated annual surface water runoff from the USGS; (b) Reliance on ground
           water as indicated by the ratio of ground water pumping to stream flow and
           pumping.

           Estimates for Figure 4-6a were calculated at the 8-digit hydrological unit code (HUC) scale by dividing
           annual average daily stream flow (from October 1, 2012 to September 30, 2013) by HUC area. Data
           accessed from the USGS (USGS. 2014g). Higher ratios (darker blues) in Figure 4-6b indicate greater
           reliance on ground water. Figure redrawn from Tidwell et al. (2012), using data provided by the U.S.
           Department of Energy's Sandia National Laboratories on December 12, 2014.
               This document is a draft for review purposes only and does not constitute Agency policy.

June 2015                                        4-25                    DRAFT—DO NOT CITE OR QUOTE

-------
      Hydraulic Fracturing Drinking Water Assessment
 1    Extensive ground water pumping can induce vertical mixing of high-quality ground water with
 2    recharge water from the land surface that has been contaminated by nitrate or pesticides, or with
 3    lower-quality ground water from underlying geologic formations [USGS. 2009: Konikow and Kendy.
 4    2005]. Ground water quality degradation associated with aquifer pumping is well documented in
 5    the southern portion of the Ogallala aquifer in the Texas panhandle. The quality of ground water
 6    used by many private, public supply, and irrigation wells is poorest in the aquifer's southern
 7    portion, with elevated concentrations of TDS, chloride, nitrate, fluoride, manganese, arsenic, and
 8    uranium [Chaudhuri and Ale. 2014a: USGS. 2009. 2007). Elevated levels of these constituents result
 9    from both natural processes and human activities such as ground water pumping [Chaudhuri and
10    Ale, 2014a: USGS, 2009]. Similar patterns of ground water quality degradation (i.e.,  salinization and
11    contamination) have also been observed in other Texas aquifers.1

12    Ground water withdrawals for hydraulic fracturing, along with irrigation and other uses, may
13    contribute to water quality degradation associated with intensive aquifer pumping  in western and
14    southern Texas. Areas with numerous high-capacity wells and large amounts of sustained ground
15    water pumping are most likely to experience ground water quality degradation associated with
16    withdrawals [USGS, 2009, 2007]. Given that Texas is prone to drought conditions, ground water
17    recharge is limited, making the already declining aquifers in southern and western Texas especially
18    vulnerable to further ground water depletion and resulting potential impacts to ground water
19    quality [USGS. 2009: Tackson etal.. 2001].

20    This survey of the available literature and data points to the potential for impacts in southern and
21    western Texas, but generally does not indicate whether impacts will occur at the local scale around
2 2    specific withdrawal points. An exception is a case study in the Eagle Ford play of southwestern
2 3    Texas that compared water demand for hydraulic fracturing with water supplies at the scale of the
24    play, county, and one square mile [Scanlon et al.. 2014]. The authors observed generally adequate
25    water supplies for hydraulic fracturing, except in specific locations, where they found excessive
26    drawdown of local ground water in a small proportion (~6% of the area] of the Eagle Ford play
27    (see Text Box 4-3].
      i Persistent salinity has also been observed in west Texas, specifically in the southern Ogallala, northwest Edwards-
      Trinity (plateau), and Pecos Valley aquifers, largely due to prolonged irrigational ground water pumping and ensuing
      alteration of hydraulic gradients leading to ground water mixing [Chaudhuri and Ale. 2014b). High levels of ground water
      salinization associated with prolonged aquifer depletion have also been documented in the Carrizo-Wilcox and southern
      Gulf Coast aquifers, underlying the Eagle Ford Shale in south Texas [Chaudhuri and Ale. 2014b: Konikow. 2013b: Boghici.
      2009). Further, elevated levels of constituents, including nitrate, lead, fluoride, chloride, sulfate, iron, manganese, and
      TDS, have been reported in the Carrizo-Wilcox aquifer [Boghici. 2009).
                    This document is a draft for review purposes only and does not constitute Agency policy.
      June 2015                                     4-26                   DRAFT—DO NOT CITE OR QUOTE

-------
      Hydraulic Fracturing Drinking Water Assessment
      Text Box 4-3. Case Study: Water Profile of the Eagle Ford Play, Texas.

 1    Researchers from the University of Texas published a detailed case study of water supply and demand for
 2    hydraulic fracturing in the Eagle Ford play in southwestern Texas [Scanlon et al.. 2014]. This effort assembled
 3    detailed information from state and local water authorities, and proprietary industry data on hydraulic
 4    fracturing, to develop a portrait of water resources in this 16-county area.

 5    Scanlon et al. [2014] compared water demand for hydraulic fracturing currently and over the projected play
 6    life (20 years] relative to water supply from ground water recharge, ground water storage (brackish and
 7    fresh], and stream flow. Using detailed ground water availability models developed by the Texas Water
 8    Development Board, they reported that water demand for hydraulic fracturing in 2013 was 30% of annual
 9    ground water recharge in the play area, and over the 20-year play lifespan it was projected to be 26% of
10    groundwater recharge, 5-8% of fresh groundwater storage, and 1% of brackish ground water storage. The
11    dominant water user in the play is irrigation (62 to 65% of water use, 53 to 55% of consumption], as
12    compared with hydraulic fracturing (13% of water use and 16% of consumption]. At the county level,
13    projected water demand for hydraulic fracturing over the 20-year period was low relative to freshwater
14    supply (ranging from 0.6-27% by county, with an average of 7.3%]. Similarly, projected total water demand
15    from all uses was low relative to supply, excluding two counties with high irrigation demands (Frio, Zavala],
16    and one county with no known ground water supplies (Maverick].

17    Although supply was found to be sufficient even in this semi-arid region, there were important caveats
18    especially at sub-county scales. The researchers found no water level declines over much of the play area
19    assessed (69% of the play area], yet in some areas they estimated ground water drawdowns of up to 50 feet
20    (12% of the play area], and in others of 100 feet or more (6% of the play area]. This was corroborated with
21    well monitoring data that showed a sharp decline in water levels in several ground water monitoring wells
22    after hydraulic fracturing activity increased in 2009. The researchers concluded that any impacts in these
23    locations could be minimized if brackish ground water were used. Projected hydraulic fracturing water use
24    represents less than 1% of total brackish ground water storage in the play area. By contrast, they concluded
25    there is limited potential for reuse of wastewater in this play because of small volumes available (less than or
2 6    equal to 5% of hydraulic fracturing water requirements].

27    The potential for water quantity and quality effects appears to be lower in north-central and
28    eastern Texas, in areas including the Barnett and Haynesville plays. Residents obtain water for
29    domestic use—which includes use of water for drinking—from a mixture of ground water and
30    surface water sources  (see Appendix Table B-6). Counties encompassing Dallas and Fort Worth rely
31    mostly on publically-supplied surface water [TWDB, 2012] (see Appendix Table B-6].

32    Although the Trinity, the major aquifer in northeast Texas, is projected to decline only slightly
33    between 2010 and 2060 (TWDB.  2012], Beneetal. (2007] estimate that hydraulic fracturing
34    ground water withdrawals will increase from  3% of total ground water use in 2005 to 7%-13% in
35    2025, suggesting the potential for localized aquifer drawdown and potential impacts  to water
36    quality. Additionally, ground water quality degradation associated with aquifer drawdown has been
37    documented in the Trinity and Woodbine aquifers underlying much of the Barnett play, with both
38    aquifers showing high levels of salinization (Chaudhuri and Ale, 2013].
                    This document is a draft for review purposes only and does not constitute Agency policy.
      June 2015                                     4-27                   DRAFT—DO NOT CITE OR QUOTE

-------
      Hydraulic Fracturing Drinking Water Assessment
 1    Overall, the potential for impacts appears higher in western and southern Texas, compared to the
 2    northeast part of the state. Impacts are likely to be localized drawdowns of ground water, as shown
 3    by a detailed case study of the Eagle Ford play (see Text Box 4-3). Scanlonetal. [2014] suggested
 4    that a shift towards brackish water use could minimize potential future impacts to fresh water
 5    resources. This finding is consistent with our county-level data (see Text Box 4-2).

      4.5.2. Colorado and Wyoming
 6    Colorado had the second highest number of disclosures in the EPA FracFocus project database,
 7    (13% of disclosures) (see Figure 4-3 and Appendix Table B-5). We combine Colorado and Wyoming
 8    because  of their shared geology of the Denver Basin (including the Niobrara play) and the Greater
 9    Green River Basin (see Figure 4-7). There are three major basins reported for Colorado: the Denver
10    Basin; the Uinta-Piceance Basin; and the Raton Basin. Together these basins contain 99% of
11    reported wells in the state, although the bulk of the activity in Colorado is in the Denver Basin (see
12    Appendix Table B-5). Fewer wells (roughly 4% of disclosures) are present in Wyoming. There are
13    two major basins reported for Wyoming (Greater Green River and Powder River) that together
14    contain 86% of activity in the state (see Appendix Table B-5).
      Figure 4-7. Major U.S. EIA shale plays and basins for Colorado and Wyoming (EIA, 2015).
               Source: (EIA, 2015b)
                   This document is a draft for review purposes only and does not constitute Agency policy.
      June 2015                                    4-28                  DRAFT—DO NOT CITE OR QUOTE

-------
      Hydraulic Fracturing Drinking Water Assessment
 1    Types of water used: Water for hydraulic fracturing in Colorado and Wyoming comes from both
 2    ground water and surface water, as well as reused wastewater [Colorado Division of Water
 3    Resources: Colorado Water Conservation Board: Colorado Oil and Gas Conservation Commission.
 4    2014: BLM, 2013b]. The only publicly available information on water sources for each state is a list
 5    of potential sources; it does not appear that either state provides more specific information on
 6    water sources for hydraulic fracturing. In the Uinta-Piceance Basin of northwestern Colorado, the
 7    EPA [2015c] reports that most of the fresh water used for fracturing comes from surface water,
 8    although fresh water sources make up a small proportion of the total water used. In the Denver
 9    Basin (Niobrara play) of southeastern Wyoming, qualitative information suggests that ground
10    water supplies much of the water used for fracturing, although no data were available to
11    characterize the ratio of ground water to surface water withdrawals [AMEC, 2014:  BLM, 2013b:
12    Tyrrell. 20121

13    Non-fresh water sources (e.g., industrial and municipal wastewater, brackish ground water, and
14    reused hydraulic fracturing wastewater) are sometimes listed as potential alternatives to fresh
15    water for fracturing in both Colorado and Wyoming [Colorado Division of Water Resources:
16    Colorado Water Conservation Board: Colorado Oil and Gas Conservation Commission, 2014: BLM,
17    2013b): no data are available to show the extent to which these non-fresh water sources are used at
18    the state or basin level. In northwest Colorado's Garfield County (Uinta-Piceance Basin), the EPA
19    (2015c) reports that fresh water is used solely for drilling and that reused wastewater supplies
20    nearly all the water for hydraulic fracturing (see Table 4-1). This estimate of reused wastewater as
21    a percentage of injected volume is markedly higher than in other locations and results from the
22    geologic characteristics of the Piceance tight sand formation, which has naturally high water
23    content and produces large volumes of relatively high-quality wastewater (U.S. EPA. 2015c).

24    In contrast, a study by Goodwin etal. (2014) assumed no reuse of wastewater  for hydraulic
25    fracturing operations  by Noble Energy in the Denver-Julesburg Basin of northeastern Colorado (see
26    Table 4-1). It is unclear whether this assumption is indicative of reuse practices of other companies
27    in the Denver-Julesburg Basin. The difference in reused wastewater rates  reported by the EPA
28    (2015c) and Goodwin etal. (2014) may indicate an east-west divide in Colorado (i.e., low reuse in
29    the east versus high reuse in the west), due at least in part to differences in wastewater volumes
30    available for reuse. However, further information is needed to adequately characterize reuse
31    patterns in Colorado.

32    Water Use per Well: Water use per well varies across Colorado, with median values of 1.8 million,
33    400,000, and 96,000 gal (6.8 million, 1.5 million, and 363,000 L) in the Uinta-Piceance, Denver, and
34    Raton Basins, respectively according to the EPA FracFocus project database (see Appendix Table B-
35    5). Low water volumes per well are reported in Wyoming (see Appendix Table B-5). Low volumes
36    reported for the Raton Basin of Colorado and the Powder River Basin of Wyoming are due to the
37    prevalence of CBM extraction in these locations  fU.S. EPA. 20151: USGS. 2014dl

38    More difficult to explain are the low volumes reported for the Denver Basin in  the EPA FracFocus
39    project database. These values are lower than any other non-CBM basin reported in Appendix Table
40    B-5. Goodwin etal. (2014) report much higher water use per well in the Denver Basin, with a

                   This document is a draft for review purposes only and does not constitute Agency policy.
      June 2015                                    4-29                  DRAFT—DO NOT CITE OR QUOTE

-------
      Hydraulic Fracturing Drinking Water Assessment
 1    median of 2.8 million gal (10.6 million L) (although only usage for the Wattenberg Field was
 2    reported). Indeed, the 10th-90thpercentiles (2.4-3.8 million gal) (9.1 to 14.4 million L) from
 3    Goodwin etal. (2014) are almost completely above those from the EPA FracFocus project database
 4    for the Denver Basin (see Appendix Table B-5).1 It is difficult to draw clear conclusions because of
 5    differences in scale (i.e., field in Goodwin versus basin in the project database) and operators (i.e.,
 6    Noble Energy in Goodwin versus all in the project database). However, it seems plausible that the
 7    EPA FracFocus project database may be incomplete for estimating the amount of water used per
 8    well in the Denver Basin.

 9    Trends in water use per well are generally lacking for Colorado, with the exception of those
10    reported by Goodwin etal. (2014). They found that water use per well is increasing with well
11    length in the Denver Basin; however, they also observed that water intensity (gallons of water per
12    unit energy extracted)  did not change, since energy recovery increased along with water use.

13    Cumulative water use/consumption: Hydraulic fracturing operations in Colorado cumulatively use
14    billions of gallons of water, but this amount is a small percentage compared to total water used or
15    consumed at the county scale. Operators in both Garfield and Weld Counties, located in the Uinta-
16    Piceance and Denver Basins, respectively, use more than 1 billion gal (3.8 billion L) annually.
17    Fracturing water use and consumption in these counties exceed those in all  other Colorado counties
18    combined (see Appendix Table B-2), but the water used for hydraulic fracturing in Garfield and
19    Weld counties is less than 2% and 3% compared to 2010 total water use and consumption,
20    respectively. In comparison, irrigated agriculture accounts for over 90% of the water used in both
21    counties (Maupinetal.. 2014: Kenny etal.. 2009). Overall, hydraulic fracturing accounts for less
22    than 2% compared to 2010 total water use in all Colorado counties represented in the EPA
23    FracFocus project database (see Appendix Table B-2). Water use estimates based on the EPA
24    FracFocus project database may be low relative to literature and state estimates (Text Box 4-1), but
25    even if estimates from the project database were doubled, hydraulic fracturing water use and
26    consumption would still be less than 4% and 5.5% compared to 2010 total water use and
27    consumption, respectively, in each Colorado county.

28    In Wyoming, reported water use for hydraulic fracturing is small compared to Colorado (see
29    Appendix Table B-l). Fracturing water use and consumption did not exceed 1% of 2010 total water
30    use and consumption, respectively, in any county (see Appendix Table B-2). Unlike Colorado,
31    Wyoming did not require disclosure to FracFocus during the time period analyzed by the EPA (U.S.
32    EPA. 2015al  (see Appendix Table B-5).

33    The Colorado Division of Water Resources et al. (2014) project that annual water use for hydraulic
34    fracturing in the state will increase by approximately 16% between 2012 and 2015, but demand in
      1 Different spatial extents might explain these differences, since Goodwin etal. (2014) focus on 200 wells in the
      Wattenberg Field of the Denver Basin; however. Weld County is the center of activity in the Wattenberg Field, and the EPA
      FracFocus project database contains 3,011 disclosures reported in Weld County, with a median water use per of 407,442
      gal (1,542,340 L), similar to that for the basin as a whole.
                   This document is a draft for review purposes only and does not constitute Agency policy.
      June 2015                                    4-30                  DRAFT—DO NOT CITE OR QUOTE

-------
      Hydraulic Fracturing Drinking Water Assessment
 1    later years is unclear. Even with an increase of 16% or more, hydraulic fracturing would still
 2    remain a relatively small user of water at the county scale in Colorado.

 3    Potential for impacts: The potential for water quantity and quality impacts appears to be low at the
 4    county scale in Colorado and Wyoming, because fracturing accounts for a low percentage of total
 5    water use and consumption (see Figure 4-2a,b). This conclusion is also supported by the
 6    comparison of hydraulic fracturing water use to water availability at the county scale (see Text Box
 7    4-2 and Figure 4-5a,b). However, counties in Colorado and Wyoming may be too large to detect the
 8    potential for impacts, and local scale studies help provide details at a finer resolution. In a multi-
 9    scale case study in western Colorado, the EPA (2015c] also did not observe any impacts in the
10    Upper Colorado River Basin. Due to the high reuse rate of wastewater, they did not identify any
11    locations where fracturing currently contributed to locally high water use intensity. They did
12    conclude, however, that future water use effects were possible (see Text Box 4-4).


      Text Box 4-4. Case Study: Impact of Water Acquisition for Hydraulic Fracturing on Local Water
      Availability in the Upper  Colorado River Basin.

13    The EPA (2015c) conducted a case study to explore the impact of hydraulic fracturing water demand on water availability
14    at the river basin, county, and local scales  in the semi-arid Upper Colorado River Basin (UCRB) of western Colorado. The
15    study area overlies the Piceance geologic basin with natural gas in tight sands. Water withdrawal impacts were quantified
16    using a water use intensity index (i.e., the  ratio between the volume of water withdrawn at a site for hydraulic fracturing
17    and the volume of available water). Researchers obtained detailed site-specific data on hydraulic fracturing water usage
18    from state and regional authorities, and estimated available water supplies using observations at USGS gage stations and
19    empirical and hydrologic modeling.

20    They found that water supplies accessed for oil and gas demand were concentrated in Garfield County, and most fresh
21    water withdrawals were  concentrated within the Parachute Creek watershed (198 mi2). However, fresh water makes up a
22    small proportion of the total water used for fracturing due to large quantities of high-quality wastewater produced from
23    the Piceance tight sands.  Fresh water is used only for drilling, and the water used for fracturing is reported to be 100%
24    reused wastewater (see Table 4-1). Due to the high reuse rate. The EPA (2015c) did not identify any locations in the
2 5    Piceance play where fracturing contributed to locally high water use intensity.

26    Scenario analyses demonstrated a pattern of increasing potential impact with decreasing watershed size in the UCRB. The
2 7    EPA (2015c)  examined hydraulic fracturing water use intensity under the current rates of both directional (S-shaped) and
2 8    horizontal drilling. They showed that for the more water-intensive horizontal drilling, watersheds had to be larger to
29    meet the same index of water use intensity (0.4) as that for directional drilling (100 mi2 for horizontal drilling, as
30    compared to  30 mi2 for directional drilling). To date, most wells have been drilled directionally into the Piceance tight
31    sands, although a trend toward horizontal drilling is expected to increase annual water use per well by about 4 times.
32    Despite this increase, total hydraulic fracturing water use is expected to remain small relative to other users. Currently,
3 3    irrigated agriculture is the largest water user in the UCRB.

34    Greater water demand could occur in the future if the water-intensive oil shale extraction industry becomes economically
35    viable in the region. Projections  for oil shale water demand indicate that the industry could increase water use for energy
36    extraction in Garfield and Rio Blanco counties.
                      This document is a draft for review purposes only and does not constitute Agency policy.

      June 2015                                        4-31                    DRAFT—DO NOT CITE OR QUOTE

-------
      Hydraulic Fracturing Drinking Water Assessment
 1    East of the Rocky Mountains in the Denver Basin, sub-county effects may be possible given the
 2    combination of high hydraulic fracturing activity and low water availability, but lack of available
 3    data and literature at this scale limits our ability to assess the potential for impacts in this location.
 4    Ceres [2014] concludes that all fractured wells in the Denver Basin are in high or extremely high
 5    water-stressed areas. Furthermore, the development of the Niobrara Shale  in southeast Wyoming
 6    occurs in areas already impacted by high agricultural water use from the Ogallala aquifer, including
 7    the state's only three ground water control areas, which were established as management districts
 8    in the southeast portion of the state in response to declining ground water levels [AMEC. 2014:
 9    Wyoming State Engineer's Office, 2014: Tyrrell, 2012: Bartos and Hallberg,  2011]. Ground water
10    withdrawals for hydraulic fracturing may have the potential to contribute to water quality
11    degradation particularly in these areas.

12    Overall, the potential for impacts appears low at the county scale in Colorado and Wyoming, but
13    sub-county effects may be possible particularly east of the Rocky Mountains in the Denver Basin.
14    Lack of available data and literature at the local scale limits our ability to assess the potential for
15    impacts  in this location.

      4.5.3. Pennsylvania, West Virginia, and Ohio
16    Pennsylvania had the third most disclosures in the EPA FracFocus project database (6.5% of
17    disclosures] (see Appendix Table B-5 and Figure 4-3]. We combine West Virginia and Ohio with
18    Pennsylvania because they share similar geology overlying the Appalachian Basin (including the
19    Marcellus, Devonian, and Utica stacked plays] (see Figure 4-8]; however, much less activity is
20    reported in these two states (see Appendix Table B-5].
                                         Marcellus, Devonian,
                                           and Utica Plays
      Figure 4-8. Major U.S. EIA shale plays and basins for Pennsylvania, West Virginia, and Ohio
                (EIA, 2015).
                Source: (EIA, 2015b).
                    This document is a draft for review purposes only and does not constitute Agency policy.
      June 2015                                     4-32                  DRAFT—DO NOT CITE OR QUOTE

-------
      Hydraulic Fracturing Drinking Water Assessment
 1    Types of water used: Surface water is the primary water source for hydraulic fracturing in
 2    Pennsylvania, West Virginia, and Ohio [Mitchell et al.. 2013a: SRBC. 2013: West Virginia PEP. 2013:
 3    Ohio EPA. 2012b]. Available data for Pennsylvania are specific to the Susquehanna River Basin
 4    (SRB), where hydraulic fracturing water is sourced mostly from surface water [SRBC, 2013] (see
 5    Table 4-3). The industry also uses mostly surface water in West Virginia [West Virginia PEP, 2014,
 6    2013] (see Table 4-3]. Although specific data are not available, state reports indicate that most
 7    water for hydraulic fracturing in Ohio's Marcellus or Utica Shale formations is sourced from nearby
 8    surface water bodies fOhio EPA. 2012b: STRONGER. 2011bl

 9    Given that surface water is the primary water source, the water used for hydraulic fracturing is
10    most often fresh water in all three  states. In both Pennsylvania's SRB and throughout West Virginia,
11    most water for hydraulic fracturing is self-supplied via direct withdrawals from surface water and
12    ground water fU.S. EPA. 2015a: West Virginia PEP. 20131 Operators also purchase water from
13    public water systems, which may include a variety of commercial water brokers (West Virginia
14    PEP. 2014: SRBC. 2013: West Virginia PEP. 2013]. Municipal supplies may be used as well,
15    particularly in urban areas of Ohio (STRONGER. 2011b1.

16    Reused hydraulic fracturing wastewater accounted for an estimated 18% and 15 % of total water
17    used for fracturing in 2012 in Pennsylvania's SRB and West Virginia, respectively (West Virginia
18    PEP. 2014: Hansenetal.. 2013: SRBC. 2013] (see Table 4-1]. Available data indicate increased reuse
19    of wastewater over time in this region likely due to the lack of nearby disposal options; from 2010-
20    2012 reused wastewater as a percentage of injected water volume ranged from 10% to 18% and
21    6% to 15% in Pennsylvania's SRB and West Virginia, respectively (West Virginia PEP, 2014: Hansen
22    etal.. 2013]. In Ohio's Marcellus and Utica Shales, reuse of wastewater is reportedly uncommon
23    (STRONGER, 2011b], potentially due to the prevalence of disposal wells in Ohio (see Chapter 8].

24    Aside from reused hydraulic fracturing wastewater,  other types of wastewaters reused for
2 5    hydraulic fracturing may include wastewater treatment plant effluent, treated acid mine drainage,
2 6    and rainwater collected at various well pads (West Virginia PEP. 2014: SRBC. 2013: West Virginia
27    PEP. 2013: Ziemkiewicz etal.. 2013: Ohio EPA. 2012bl No data are available on the frequency of
28    use of these other wastewaters.

29    Water Use per Well: Operators in these three states reported the third, fourth, and fifth highest
30    median water use nationally in the EPA FracFocus project database, with 5.0, 4.2, and
31    3.9 million gal (18.9,15.9, and 14.8 million L] per well in West Virginia, Pennsylvania, and Ohio,
32    respectively (U.S. EPA. 2015b] (see Appendix Table B-5]. Hansenetal. (2013] report similar water
33    use estimates for Pennsylvania and West Virginia (see Appendix Table B-5]. This correspondence is
34    not surprising, as these estimates are also based on FracFocus data (via Skytruth]. For 2011, the
35    year overlapping with the time frame of the EPA FracFocus report (U.S. EPA, 2015a], Mitchell et al.
36    (2013a] report an average of 2.3 million gal (8.7 million L] for vertical wells (62 wells] and
37    4.6 million gal (17.4 million L] for horizontal wells (612 wells] in the Pennsylvania portion of the
38    Ohio River Basin, based on records from PA PEP. The weighted average water use per well was
39    4.4 million gal (16.7 million L], similar to results based on the EPA FracFocus project database
40    listed above.

                   This document is a draft for review purposes only and does not constitute Agency policy.
      June 2015                                   4-33                   DRAFT—DO NOT CITE OR QUOTE

-------
      Hydraulic Fracturing Drinking Water Assessment
 1    Cumulative water use/consumption: In this tri-state region, highest cumulative water use for
 2    hydraulic fracturing is in northeastern Pennsylvania counties. On average, operators in Bradford
 3    County reported over 1 billion gal (3.8 billion L) used annually in 2011 and 2012 for fracturing;
 4    operators in three other counties (Susquehanna, Lycoming, and Tioga Counties) cumulatively
 5    reported 500 million gal (1.9 billion L) or more used annually (see Table 4-2). On average,
 6    hydraulic fracturing water use is 3.2% compared to 2010 total county water use for counties with
 7    disclosures in the EPA FracFocus project database in these three states (see Table 4-2 and
 8    Appendix Table B-2). Susquehanna County in Pennsylvania has the highest percentages relative to
 9    2010 total water use (47%) and consumption (123%).

10    Potential for impacts: Water availability is higher in Pennsylvania, West Virginia, and Ohio than in
11    many western states, reducing the likelihood of impacts to drinking water quantity and quality. At
12    the county scale, water supplies appear adequate to accommodate this use (Tidwell etal.. 2013)
13    (see Text Box 4-2 and Figure 4-5a,b).

14    However, impacts could still occur at specific withdrawal points. In a second, multi-scale case study,
15    EPA researchers concluded that individual streams in this region can be vulnerable to typical
16    hydraulic fracturing water withdrawals depending on stream size, as defined by contributing basin
17    area (U.S. EPA. 2015c) (see Text Box 4-5). They observed infrequent (in less than 1% of
18    withdrawals) high ratios of hydraulic fracturing water consumption to stream flow (high
19    consumption-to-stream flow events). Passby flows can reduce the frequency of high consumption-
20    to-stream flow events, particularly in the smallest streams (U.S. EPA. 2015C).1
      1A passby flow is a prescribed, low stream flow threshold below which withdrawals are not allowed. The SRBC uses
      passby flows to protect streams in the Susquehanna River Basin, an area including much of eastern Pennsylvania [U.S.
      EPA. 2015cl.
                    This document is a draft for review purposes only and does not constitute Agency policy.
      June 2015                                     4-34                  DRAFT—DO NOT CITE OR QUOTE

-------
      Hydraulic Fracturing Drinking Water Assessment
      Text Box 4-5. Case Study: Impact of Water Acquisition for Hydraulic Fracturing on Local Water
      Availability in the Susquehanna River Basin.

 1    The EPA (2015c) conducted a second case study analogous to that in the UCRB (see Text Box 4-4), to explore
 2    the impact of hydraulic fracturing water demand on water availability at the river basin, county, and local
 3    scales in the SRB in northeastern Pennsylvania. The study area overlies the Marcellus Shale gas reservoir.
 4    Water withdrawal impacts were quantified using a water use intensity index (see Text Box 4-4). Researchers
 5    obtained detailed site-specific data on hydraulic fracturing water usage from state and regional authorities,
 6    and estimated available water supplies using observations at USGS gage stations and empirical and
 7    hydrologic modeling.

 8    Most water for fracturing in the SRB is self-supplied from rivers and streams with withdrawal points
 9    distributed throughout a wide geographic area. Public water systems provide a relatively small proportion of
10    the water needed.  Reuse of wastewater makes up approximately 13% to 18% of injected fluid volume on
11    average, as  reported by the EPA (2015c) for 2008 to 2011 and Hansen et al. (2013) for 2012, respectively
12    (see Table 4-1). The Susquehanna River Basin Commission (SRBC) regulates water acquisition for hydraulic
13    fracturing and issues permits that set limits on the volume, rate, and timing of withdrawals at individual
14    withdrawal points; passby flow thresholds halt water withdrawals during low flows.

15    The EPA (2015c) demonstrated that streams can be vulnerable from typical hydraulic fracturing water
16    withdrawals depending on their size, as denned by contributing basin area. Small streams have the potential
17    for impacts (i.e., high water use intensity) for all or most of the year. The EPA (2015c) showed an increased
18    likelihood of impacts in small watersheds (less than 10 mi2). Furthermore, they showed that in the absence of
19    passby flows, even larger watersheds (up to 600 mi2) could be vulnerable during maximum withdrawal
20    volumes and infrequent droughts. However, high water use intensity calculated from observed hydraulic
21    fracturing withdrawals occurred at only a few withdrawal locations in small streams; local high water use
2 2    intensity was not found at the majority of withdrawal points.

2 3    Without management of the rate and timing of withdrawals, surface water withdrawals for
24    hydraulic fracturing have the potential to affect both water quantity and quality [Mitchell etal..
25    2013a). Potential effects are generally applicable, but are especially relevant in this region because
26    surface water is the primary water source for hydraulic fracturing in Pennsylvania, West Virginia,
27    and Ohio. Of greatest concern are small, unregulated streams, particularly under drought
28    conditions or during seasonal low flows [U.S. EPA, 2015c: Vengosh etal., 2014: Mitchell etal.,
29    2013a: Vidic etal..  2013:  Rahm and Riha. 2012: Rolls etal.. 2012: Kargbo etal.. 2010: McKay and
30    King. 2006). Surface water quality impacts may be of concern if a pollution discharge point (e.g.,
31    sewage treatment plant, agricultural runoff, or chemical spill) is immediately downstream of a
32    hydraulic fracturing withdrawal [U.S. EPA. 2015c: NYSDEC. 2011).* Water quality impacts
      1 Aside from direct surface water withdrawals, unmanaged withdrawals from public water systems can cause cross-
      contamination if there is a loss of pressure, allowing the backflow of pollutants from tank trucks into the distribution
      system. The state of Ohio has issued a fact sheet relevant to this potential concern, intended specifically for public water
      systems providing water to oil and gas companies [Ohio EPA. 2012a). To prevent potential cross-contamination, Ohio
      requires a backflow prevention device at cross-connections. For example, bulk loading stations that provide public supply
      water directly to tank trucks are required to have an air-gap device at the cross-connection to prevent the backflow of
      contaminants into the public water system [Ohio EPA. 2012a).
                     This document is a draft for review purposes only and does not constitute Agency policy.
      June 2015                                      4-35                   DRAFT—DO NOT CITE OR QUOTE

-------
      Hydraulic Fracturing Drinking Water Assessment
 1    associated with reduced water levels may also include possible interference with the efficiency of
 2    drinking water treatment plant operations, as increased contaminant concentrations in drinking
 3    water sources may necessitate additional treatment and ultimately impact drinking water quality
 4    [Water Research Foundation. 2014: Benottletal.. 20101.1

 5    Overall, there appears to be adequate surface water for hydraulic fracturing, but there is the
 6    potential for impacts to both drinking water quantity and quality, particularly in small streams, if
 7    withdrawals are not managed [U.S. EPA. 2015c).

      4.5.4.  North Dakota and Montana
 8    North Dakota was fourth in the number of disclosures in the EPA FracFocus project database (5.9%
 9    of disclosures) (see Appendix Table B-5 and Figure 4-3). We combine Montana with North Dakota
10    because both overlie the Williston Basin (which contains the Bakken play, shown in Figure 4-9),
11    although many fewer wells are reported for Montana (see Appendix Table B-5). The Williston Basin
12    is the only basin with significant activity reported for either state, though other basins are also
13    present in Montana (e.g., the Powder River Basin).
      Figure 4-9. Major U.S. EIA shale plays and basins for North Dakota and Montana (EIA, 2015b).
                Source: (EIA, 2015b).
      1 For instance, an increased proportion of organic matter entering a treatment plant may increase the formation of
      trihalomethanes, byproducts of the disinfection process formed as chlorine react with organic matter in the water being
      treated [Water Research Foundation. 2014).
                    This document is a draft for review purposes only and does not constitute Agency policy.
      June 2015                                     4-36                  DRAFT—DO NOT CITE OR QUOTE

-------
      Hydraulic Fracturing Drinking Water Assessment
 1    Types of water used: Hydraulic fracturing of the Bakken play underlying much of western North
 2    Dakota and northeastern Montana depends on both ground and surface water resources. Surface
 3    water from the Missouri River system provides the largest source of fresh water in the center of
 4    Bakken oil development [North Dakota State Water Commission. 2014: EERC. 2011. 2010: North
 5    Dakota State Water Commission, 2010]. Apart from the Missouri River system, regional surface
 6    waters (i.e., small streams) do not provide a consistent supply of water for the oil industry due to
 7    seasonal stream flow variations. Sufficient stream flows generally occur only in the spring after
 8    snowmelt [EERC. 2011). Ground water from glacial and bedrock aquifer systems has traditionally
 9    supplied much of the water needed for Bakken development, but concerns over limited ground
10    water supplies have led to limits on the number of new ground water withdrawal permits issued
11    [Ceres. 2014: Plummer etal.. 2013: EERC. 2011. 2010: North Dakota State Water Commission.
12    20101

13    The water used for Bakken development is described as mostly fresh. The EPA FracFocus report
14    shows that "fresh" was the only source of water listed in almost all disclosures reporting  a source of
15    water in North Dakota [U.S. EPA. 2015a].1 Reuse of Bakken wastewater is limited due to its quality
16    of high TDS, which presents challenges for treatment and reuse. However, the industry is
17    researching treatment technologies for reuse of this wastewater [Ceres. 2014: EERC. 2013.  2011).

18    Water for hydraulic fracturing is commonly purchased from municipalities or other public water
19    systems in the region. The water is often delivered to trucks at water depots or transported directly
20    to well pads via pipelines fEERC. 2011).

21    Water Use per Well: Water use per well is intermediate compared with other areas, with a median
22    of 2.0 and 1.6 million gal (7.6 and 6.1 million L) per well in the Williston Basin in North Dakota and
23    Montana, respectively according to the EPA's FracFocus project database (see Appendix Table B-5).
24    The North Dakota State Water Commission reports similar volumes (2.2 million gal (8.3 million L)
25    per well on average for North Dakota) in a summary fact sheet (North Dakota State Water
26    Commission. 2014).2

27    A presentation by the North Dakota Department of Mineral Resources (NDDMR) suggests that
28    Bakken wells require an average of 600 gal (2,300 L) per day of "maintenance water" in addition to
29    the initial water for hydraulic fracturing (North  Dakota Department of Mineral Resources. 2013).3
30    This extra water is reportedly needed because of the relatively high salt content of Bakken brine,
31    potentially leading to salt buildup, pumping problems, and restriction of oil  flow. According to the
32    NDDMR, maintenance water can contribute to large additional volumes over a typical well life span
33    (6.6-8.8 million gal (25-33 million L) over 30-40 years). It is unclear whether this phenomenon is
34    restricted to the Bakken play.
      i However, 25% of North Dakota disclosures included information related to water sources [U.S. EPA. 2015a).
      2_The fact sheet is a stand-alone piece, and it is not accompanied by an underlying report.
       The NDDMR's presentation that mentions the issue of maintenance water was later picked up and reported on by
      National Geographic fhttp://news.nationalgeographic.com/news/energy/2013/ll/131111-north-dakota-wells-
      maintenance-water/) and by Ceres (2014). Peer-reviewed studies on the Bakken also report on maintenance water [e.g..
      Scanlon etal.. 2014). but they refer to the same original sources.
                   This document is a draft for review purposes only and does not constitute Agency policy.
      June 2015                                    4-37                  DRAFT—DO NOT CITE OR QUOTE

-------
      Hydraulic Fracturing Drinking Water Assessment
 1    Cumulative water use/consumption: Cumulative water use for fracturing in this region is greatest in
 2    the northwestern corner of North Dakota. In counties with 2011 and 2012 disclosures to FracFocus,
 3    fracturing water use averaged approximately 123 million gal (466 million L) per county annually in
 4    the two-state area, with use in McKenzie and Williams Counties in North Dakota exceeding
 5    500 million gal (1.9 billion L) per year (see Appendix Table B-2). There are four counties where
 6    2011 and 2012 average hydraulic fracturing water use was 10% or more of 2010 total water use.
 7    Mountrail and Dunn Counties showed the highest percentages. Outside of North Dakota's northwest
 8    corner, the rest of the state and Montana showed little cumulative water use from hydraulic
 9    fracturing (see Table 4-2 and Appendix Table B-2).

10    Potential for impacts: In this region, there are concerns about over-pumping ground water
11    resources, but the potential for impacts appears to be low provided the Missouri River is
12    determined to be a sustainable and usable source. This finding of a low potential for impacts is also
13    supported by the comparison of hydraulic fracturing water use to water availability at the county
14    scale (see Text Box 4-2 and Figure 4-5a,b.) This area is primarily rural, interspersed with small
15    towns. Residents use a mixture of surface water and ground  water for domestic use depending on
16    the county, with most water supplied by local municipalities (see Appendix Table B-6).

17    The state of North Dakota and the U.S. Army Corps of Engineers concluded that ground water
18    resources in western North Dakota are not sufficient to meet the needs of the oil and gas industry
19    (U.S. Army Corps of Engineers. 2011: North Dakota State Water Commission. 20101 All users
20    combined currently withdraw approximately 6.2 billion gal (23.5 billion L) of water annually in an
21    11-county region in western North Dakota, already stressing ground water supplies (U.S. Army
22    Corps of Engineers. 2011]. By contrast, the total needs of the oil and gas industry are projected to
23    range from approximately 2.2 and 8.8 billion gal (8.3 and 33.3 billion L) annually by the year 2020
24    (U.S. Army Corps of Engineers. 2011).

2 5    Due to concerns for already stressed ground water supplies,  the state of North Dakota limits
26    industrial ground water withdrawals, particularly from the Fox Hills-Hell Creek aquifer (Ceres,
27    2014: Plummer et al.. 2013: EERC. 2011. 2010: North Dakota State Water Commission. 20101
28    Currently, the oil industry is the  largest industrial user of water from the Fox Hills-Hell Creek
29    aquifer in western North Dakota (North Dakota State Water  Commission, 2010]. Many farms,
30    ranches, and some communities in western North Dakota rely on flowing wells from this artesian
31    aquifer, particularly in remote areas that lack electricity for pumping; however, low recharge rates
32    and prolonged withdrawals throughout the last century have resulted in steady declines in the
33    formation's hydraulic pressure (North Dakota State Water Commission, 2010]. Declines in
34    hydraulic pressure do not appear to be associated with impacts to ground water quality; rather, the
35    state is concerned with maintaining flows for users through  conservation  (North Dakota State
36    Water Commission. 2010].

37    To reduce pressure on ground water, the state  is encouraging the industry to seek surface water
38    withdrawals from the Missouri River system, which if used, may be an adequate resource. The
39    North Dakota State Water Commission concluded the Missouri River and its dammed reservoir,
40    Lake Sakakawea, are the only plentiful and dependable water supplies for the oil industry in

                   This document is a draft for review purposes only and does not constitute Agency policy.
      June 2015                                   4-38                  DRAFT—DO NOT CITE OR QUOTE

-------
      Hydraulic Fracturing Drinking Water Assessment
 1    western North Dakota [North Dakota State Water Commission. 20101. In 2011, North Dakota
 2    authorized the Western Area Supply Project, by which Missouri River water (via the water
 3    treatment plant in Williston, North Dakota) will be supplied to help meet water demands, including
 4    for oil and gas development, of the state's northwest counties [WAWSA, 2011]. Industrial surface
 5    water withdrawals are presently allowed in Lake Sakakawea on a temporary and controlled basis
 6    while the U.S. Army Corps of Engineers conducts a multi-year study to determine whether surplus
 7    water is available to meet the demands of regional municipal and industrial users [U.S. Army Corps
 8    of Engineers. 2011).

      4.5.5.  Oklahoma and Kansas
 9    Oklahoma had the fifth most disclosures in the EPA FracFocus project database (5.0% of
10    disclosures) (see Appendix Table B-5, and Figure 4-3). Three major basins— the Anadarko, which
11    includes the Woodford play; the Arkoma, which includes the Fayetteville play; and the Ardmore,
12    which includes the Woodford play—contain 67%  of the disclosures in Oklahoma (see Figure 4-9
13    and Appendix Table B-5). Few wells were reported for Kansas (Kansas disclosures comprise 0.4%
14    of the EPA FracFocus project database), but because of the shared geology of the Cherokee Platform
15    across the two states,  we group Kansas with Oklahoma. Oklahoma and Kansas were two of the
16    three states where a large fraction of wells were not associated with a basin defined by the U.S. EIA
17    (U.S.  EPA. 2015b) (see Appendix Table B-5).1

18    Types of water used: Water for hydraulic fracturing in Oklahoma and Kansas comes from both
19    surface and ground water (Kansas Water Office, 2014: Taylor,  2012). Data on temporary water use
20    permits in Oklahoma (which make up the majority of water use permits for Oklahoma oil and gas
21    mining) show that, in  2011, approximately 63% and 37% of water for hydraulic fracturing came
22    from surface and ground water, respectively (Taylor. 2012] (see Table 4-3). General water use in
23    Oklahoma follows an east-west divide, with the eastern half dependent on surface sources and the
24    western half relying heavily on ground water (OWRB, 2014). Water obtained for fracturing is
25    assumed to fit this pattern as well. No data are available on the proportion of hydraulic fracturing
26    water that is sourced from surface versus ground  water resources in Kansas.

27    For both Oklahoma and Kansas, no data are available to describe the extent to which reused
28    wastewater is used as a percentage of total injected volume. However, the quality of Oklahoma's
29    Woodford Shale wastewater has been described as low in TDS, and thus reuse could reduce the
30    demand for fresh water fKuthnertetal.. 20121
      1 Alaska was the other state in the EPA FracFocus project database where the U.S. EIA shale basins did not adequately
      describe well locations, with all 37 wells in Alaska not associated with a U.S. EIA basin. For all other states, U.S. EIA shale
      basins captured 86%-100% of the wells in the EPA FracFocus project database [U.S. EPA. 2015b).
                   This document is a draft for review purposes only and does not constitute Agency policy.
      June 2015                                    4-39                  DRAFT—DO NOT CITE OR QUOTE

-------
      Hydraulic Fracturing Drinking Water Assessment
      Figure 4-10. Major U.S. EIA shale plays and basins for Oklahoma and Kansas (EIA, 2015).
               Source: (EIA, 2015b)

 1    Water Use per Well: State-level estimates of median water use per well in Oklahoma include 2.6
 2    million gal (9.8 million L) and 3 million gal (11 million L) [U.S. EPA(2015b] and, Murray [2013],
 3    respectively]. Water use for hydraulic fracturing increased from 2000 to 2011, driven by volumes
 4    required for fracturing horizontal wells across the state [Murray, 2013]. Within the state there are
 5    wide ranges in water use for different formations. According to the EPA FracFocus project database,
 6    the Ardmore and Arkoma Basins of Oklahoma, had the highest median water use in the country,
 7    with medians of 8.0 and 6.7 million gal (30.3 and 25.4 million L) per well, respectively; whereas the
 8    Anadarko Basin had lower median water use per well and higher disclosure counts (3.3 million gal
 9    (12.5 million L), 935 disclosures) (see Appendix Table B-5). Wells not associated with a U.S. EIA
10    basin had a median of 1.9 million gal (7.2 million L) per well (592 disclosures) (see Appendix Table
11    B-5). It is not clear why lower water volumes were reportedly used in unassociated wells, but
12    Oklahoma has several CBM deposits in the eastern part of the state where very low water use has
13    been reported (Murray. 2013). Median water use per well in Kansas was 1.5 million gal (5.7 million
14    L), focused mostly in a five-county area in the south-central and southwest portions of the state
15    (see Appendix Table B-5).

16    Cumulative water use/consumption: Cumulatively, operators reported using an average of
17    71.9 million gal (272.2 million L) of water annually in Oklahoma counties with disclosures; in
                   This document is a draft for review purposes only and does not constitute Agency policy.
      June 2015                                    4-40                  DRAFT—DO NOT CITE OR QUOTE

-------
      Hydraulic Fracturing Drinking Water Assessment
 1    Kansas, this value is only 3.5 million gal (13.2 million L) (see Appendix Table B-2). Average
 2    hydraulic fracturing water use in 2011 and 2012 did not exceed 10% of 2010 total water use in any
 3    county in Oklahoma or Kansas (see Appendix Table B-2). However, there were six counties in
 4    Oklahoma (Alfalfa, Canadian, Coal, Pittsburg, Rogers Mills, and Woods) where fracturing water
 5    consumption exceeded 10% of 2010 total county water consumption.

 6    Potential for impacts: The potential for effects on drinking water resources appears to be low in
 7    Oklahoma and Kansas, since hydraulic fracturing water use and consumption are generally low as a
 8    percentage of total water use and consumption. This finding is generally supported by the
 9    comparison of cumulative fracturing water use to water availability at the county scale (see Text
10    Box 4-2 and Figure 4-5a,b). If impacts to water quantity or quality do occur, however, they are more
11    likely to happen in western Oklahoma than in the eastern half of the state or Kansas. Of the six
12    Oklahoma counties where fracturing consumption exceeded 10% of 2010 water consumption,
13    three (Alfalfa, Canadian, and Roger Mills) are in  the western half of the state where surface water
14    availability is lowest (Figure 4-6a). Surface water is fully allocated in the Panhandle and West
15    Central regions, encompassing much of the state's northwestern quadrant (OWRB. 2014). As a
16    result, residents generally rely on ground water in western Oklahoma (see Appendix Table B-6),
17    and it is likely that fracturing does as well.

18    Projecting out to 2060, Oklahoma's Water Plan concludes that aquifer storage depletions are likely
19    in the Panhandle and West Central regions due to over-pumping, particularly for irrigation (OWRB.
20    2014). Ground water depletions are anticipated to be small relative to storage, but will be the
21    largest in summer months and may lead to higher pumping costs, the need for deeper wells, lower
22    water yields, and detrimental  effects on water quality (OWRB, 2014). Drought conditions are likely
23    to exacerbate this problem, and Oklahoma's Water Plan specifically mentions the potential for
24    climate change to affect future water supplies in the state (OWRB. 2014). In the adjacent Texas
25    Panhandle, future irrigation needs may go unmet (TWDB. 2012). and this may be the case in
26    western Oklahoma as well.

2 7    Aquifer depletions in western Oklahoma may be associated with ground water quality degradation,
2 8    particularly under drought conditions. The central portion of the Ogallala aquifer underlying the
29    Oklahoma Panhandle  and western Oklahoma contains elevated levels of some constituents (e.g.,
3 0    nitrate) due to over-pumping, although generally it is of better quality than the southern portion of
31    the aquifer (USGS. 2009). Additional ground water withdrawals for hydraulic fracturing in western
32    Oklahoma may add to these water quality issues, particularly in combination with other  substantial
33    water uses (e.g., irrigation)  (USGS, 2009).

      4.5.6. Arkansas and Louisiana
34    Arkansas and Louisiana were  ranked seventh and tenth in the number of disclosures in the EPA
35    FracFocus project database, respectively (see Appendix Table B-5). Hydraulic fracturing activity in
36    Louisiana occurs primarily in the TX-LA-MS Salt Basin, which contains the Haynesville play; activity
37    in Arkansas is dominated by the Arkoma Basin, which contains the Fayetteville play (Figure 4-11).
                   This document is a draft for review purposes only and does not constitute Agency policy.
      June 2015                                   4-41                  DRAFT—DO NOT CITE OR QUOTE

-------
      Hydraulic Fracturing Drinking Water Assessment
 1    Types of water used: Surface water is reported as the primary source of water for hydraulic
 2    fracturing operations in both Arkansas and Louisiana fANRC. 2014; LA Ground Water Resources
 3    Commission. 2012; STRONGER. 20121 Quantitative information is lacking for Arkansas on the
 4    proportion of water sourced from surface versus ground water. However, data are available for
 5    Louisiana, where an estimated 87% of water for hydraulic fracturing in the Haynesville Shale is
 6    sourced from surface water [LA Ground Water Resources Commission, 2012] (see Table 4-3). In
 7    2008, during the early stages of development, hydraulic fracturing in Louisiana relied heavily on
 8    ground water from the Carrizo-Wilcox aquifer, although concerns for the sustainability of ground
 9    water resources have more recently prompted the state to encourage surface water withdrawals
10    [LA Ground Water Resources Commission, 2012].

11    The EPAFracFocus report suggests that significant reuse of wastewater may occur in Arkansas to
12    offset total fresh water used for hydraulic fracturing; 70% of all disclosures reporting a water
13    source indicated a blend of "recycled/surface," whereas only 3% of disclosures reporting a water
14    source noted "fresh" as the exclusive water source [U.S. EPA. 2015a].1According to Veil [2011],
15    Arkansas' Fayetteville Shale wastewater is of relatively good quality (i.e., low TDS], potentially
16    facilitating reuse. Data are generally lacking on the extent to which hydraulic fracturing wastewater
17    is reused to offset total fresh water use in Louisiana.
      1 93% of Arkansas disclosures included information related to water sources [U.S. EPA. 2015a).


                   This document is a draft for review purposes only and does not constitute Agency policy.
      June 2015                                     4-42                  DRAFT—DO NOT CITE OR QUOTE

-------
     Hydraulic Fracturing Drinking Water Assessment
     Figure 4-11. Major U.S. EIA shale plays and basins for Arkansas and Louisiana (EIA. 2015b).
               Source: (EIA.2015bl.

1    Water Use per Well: Arkansas and Louisiana have the highest median water use per well in the
2    nation, at 5.3 million and 5.1 million gal (20.1 million and 19.3 million L), respectively based on the
3    EPA FracFocus project database (see Appendix Table B-5).1

4    Cumulative water use/consumption: On average, hydraulic fracturing operations cumulatively use
5    408 million gal (1.54 billion L) of water each year in Arkansas counties reporting activity, or 9.3%
6    of 2010 total county water use (26.9% of total county consumption) (see Appendix Table B-2). In
7    2011 and 2012, five counties dominated fracturing water use in Arkansas: Cleburne, Conway,
8    Faulkner, Van Buren, and White Counties (see Appendix Table B-2). Van Buren, which is sparsely
9    populated and thus has relatively low total water use and consumption, is by far the county highest
     According to STRONGER (2012) and STRONGER (2011a). both states require disclosure of information on water use per
     well, but this has not been synthesized into state level reports.
                   This document is a draft for review purposes only and does not constitute Agency policy.
     June 2015                                     4-43                  DRAFT—DO NOT CITE OR QUOTE

-------
      Hydraulic Fracturing Drinking Water Assessment
 1    in hydraulic fracturing water use and consumption relative to 2010 total water use and
 2    consumption (56% and 168%, respectively) (see Table 4-2).

 3    In Louisiana, fracturing water use is concentrated in six parishes in the far northwestern corner of
 4    the state, associated with the Haynesville play.1 On average in 2011 and 2012, hydraulic fracturing
 5    used 117 million gal (443 million L) of water annually per parish, representing approximately 3.6%
 6    and 10.8% of 2010 total water use and consumption, respectively (see Appendix Table B-2).
 7    Operators in De Soto Parish used the most water (over 1 billion gal (3.8 billion L) annually).
 8    Fracturing water use and consumption was highest relative to 2010 total water use and
 9    consumption (35.5% and 83.2%, respectively) in Red River Parish (see Table 4-2). These numbers
10    may be low estimates since Louisiana required disclosures to the state or FracFocus and Arkansas
11    required disclosures to the state, but not FracFocus, during the time period analyzed (U.S. EPA,
12    2015al (see Appendix Table  B-5).

13    Potential for impacts: Water availability is generally higher in Arkansas and Louisiana than in states
14    farther west, reducing the potential for impacts to drinking water quantity and quality (Figure 4-6a,
15    Text Box 4-2, and Figure 4-5). There are, however, concerns about over-pumping of ground water
16    resources in northwestern Louisiana. Prior to 2008, most operators in the  Louisiana portion of the
17    Haynesville Shale used ground water, withdrawing from the Carrizo-Wilcox,  Upland Terrace, and
18    Red River Alluvial aquifer systems (LA Ground Water Resources Commission. 2012). To mitigate
19    stress on ground water, the state issued a water use advisory to the oil and gas industry that
20    recommended Haynesville Shale operators seek alternative water sources to the Carrizo-Wilcox
21    aquifer, which is predominantly used for public supply (LDEQ. 2008). Operators then transitioned
22    to mostly surface water, with a smaller ground water component (approximately 12% of all
23    fracturing water used) (LA Ground Water Resources Commission, 2012). Of this ground water
24    component, the majority (approximately 74%) still came from the Carrizo-Wilcox aquifer (LA
25    Ground Water Resources Commission, 2012).

26    Although the potential for hydraulic fracturing withdrawals to affect water supplies and water
27    quality in the aquifer appears greatly reduced, it is not entirely eliminated. Despite Louisiana's
28    water use advisory, a combination of drought conditions and higher than normal withdrawals (for
29    all uses, not solely hydraulic  fracturing) from the Carrizo-Wilcox and Upland Terrace aquifers
30    caused several water wells to go dry in July 2011. In August 2011, a ground water emergency was
31    declared for southern Caddo Parrish (LA Ground Water Resources Commission. 2012). There are
32    hydraulic fracturing wells in southern Caddo Parrish  (U.S. EPA, 2015b), and so it is possible that
33    fracturing withdrawals contributed to the problem of declines in ground water in this instance.

      4.5.7.  Utah,  New Mexico,  and California
34    Together, Utah, New Mexico, and California accounted for approximately 9% of disclosures in the
35    EPA FracFocus project database (3.8%, 3.1% and 1.9% of disclosures, respectively) (see Appendix
36    Table B-5 and Figure 4-3). Almost all reported hydraulic fracturing in Utah and California were in
      1 Louisiana is divided into parishes, which are similar to counties in other states.
                   This document is a draft for review purposes only and does not constitute Agency policy.
      June 2015                                    4-44                  DRAFT—DO NOT CITE OR QUOTE

-------
      Hydraulic Fracturing Drinking Water Assessment
 1    the Uinta-Piceance Basin (99%) and San Joaquin Basin (95%), respectively. Activity in New Mexico
 2    mostly occurs in the Permian and San Juan Basins, which together comprised 96% of reported
 3    disclosures in that state (see Figure 4-12).
      Figure 4-12. Major U.S. EIA shale plays and basins for Utah, New Mexico, and California
                (EIA, 2015).
                Source: (EIA, 2015b).

 4    Types of water used: Of these three states, California has the most information available on the
 5    sources of water used for hydraulic fracturing. Most current and proposed fracturing activity is
 6    focused in Kern County in the San Joaquin Basin, where well stimulation notices indicate that
 7    operators depend mainly on surface water purchased from nearby irrigation districts (CCST. 2014).
 8    California irrigation districts receive water allocated by the State Water Project, and deliveries may
 9    be restricted or eliminated during drought years (CCST, 2014).l In addition to publicly-supplied
10    surface water, operators also may self-supply a smaller proportion of water from on-site ground
11    water wells (CCST, 2014). Operators use primarily fresh water for hydraulic fracturing (96% of well
      i The California State Water Project is water storage and distribution system maintained by the California Department of
      Water Resources, which provides water for urban and agricultural water suppliers in Northern California, the San
      Francisco Bay Area, the San Joaquin Valley, the Central Coast, and Southern California [California Department of Water
      Resources. 2015).
                    This document is a draft for review purposes only and does not constitute Agency policy.
      June 2015                                      4-45                   DRAFT—DO NOT CITE OR QUOTE

-------
      Hydraulic Fracturing Drinking Water Assessment
 1    stimulation notices reported); reused wastewater (sometimes blended with fresh water) is used in
 2    small amounts relative to total water use (4% of well stimulation notices reported) (CCST. 2014) (see
 3    Table 4-1).

 4    The source, quality, and provisioning of water used for hydraulic fracturing in Utah and New
 5    Mexico are not well characterized. The 2010 New Mexico water use report summarizes
 6    withdrawals for a variety of water use categories. In 2010, mining water use (which includes water
 7    used for oil and gas production) consisted of 26% and 74% of surface and ground water
 8    withdrawals, respectively (NM OSE. 2013). Assuming that hydraulic fracturing follows the same
 9    pattern as other mining water uses (e.g., for metals, coal, geothermal), water for hydraulic
10    fracturing in New Mexico would be supplied primarily by ground water withdrawals. To our
11    knowledge, no data are available to characterize the source of water for hydraulic fracturing
12    operations in Utah. In addition, no data are available to describe the extent to which reused
13    wastewater is used as a proportion of total water injected for either Utah or New Mexico.

14    Water use per well: Median water use per well in Utah, New Mexico, and California is lower than in
15    other states in the EPA FracFocus project database: Utah ranks 13th (approximately 302,000 gal
16    (1.14 million L)), New Mexico ranks 14th (approximately 175,000 gal (662,000 L)), and California
17    ranks 15th (approximately 77,000 gal (291,000 L)) out of the 15 states (see Appendix Table B-5). A
18    likely explanation for the low water use per well in Utah and New Mexico is the prevalence of CBM
19    in the Uinta (Utah) and San Juan (New Mexico) Basins. Low water use per well in California is
20    attributed to the prevalence of vertical wells and the use of crosslinked gels. Vertical wells
21    dominate because the complex geology precludes long horizontal drilling and fracturing (CCST.
22    20141

23    For California, the California Council on Science and Technology (CCST) reports average water use
24    per well of 130,000 gal (490,000 L), which agrees with the state average of approximately 131,700
25    gal (498,500 L) according to the EPA FracFocus project database (CCST, 2014) (see Appendix Table
26    B-5); this is expected because estimates from CCST are also based on data submitted to FracFocus.

27    Cumulative water use/consumption: Operators in Utah, New Mexico, and California report using low
28    cumulative amounts  of water compared to most other states (see Appendix Table B-l). Only four
29    counties (Duchesne and Uintah Counties in Utah, and Eddy and Lea Counties in New Mexico)
30    required more than 50 million gal (189 million L) annually (see Appendix Table B-2). Fracturing
31    water use and consumption did not exceed 1% of 2010 total water use and consumption in any
32    county.

33    Potential for impacts: The potential for water quantity and quality impacts from hydraulic
34    fracturing water withdrawals in Utah, New Mexico, and California appears to be low at present (see
35    Text Box 4-2 and Figure 4-5a,b). Hydraulic fracturing does not use or consume much water
36    compared to other users or consumers in these states. As in other states, this does not preclude
37    sub-county effects, and this finding of low potential for impacts could change if fracturing activities
38    increase beyond present levels. This is  particularly the case because these states generally have low
39    surface water availability (see Figure 4-6a) and high ground water dependence (see Figure 4-6b),
                   This document is a draft for review purposes only and does not constitute Agency policy.
      June 2015                                   4-46                  DRAFT—DO NOT CITE OR QUOTE

-------
      Hydraulic Fracturing Drinking Water Assessment
 1    and have experienced frequent periods of drought over the last decade [National Drought
 2    Mitigation Center. 2015].

      4.6. Chapter Synthesis
 3    In this chapter we examine the potential for water acquisition for hydraulic fracturing to affect
 4    drinking water quantity and quality. The potential for impacts largely depends on water use,
 5    consumption, and availability. Water management—in terms of the type of water used, the timing
 6    or location of water withdrawals, or other factors—also can play a role. Because all of these factors
 7    vary considerably from place-to-place, any impacts that occur will be location-specific and occur at
 8    the spatial scale of the specific drinking water resource (i.e., the particular stream, watershed, or
 9    local ground water aquifer). Therefore, it is important to consider the potential for hydraulic
10    fracturing impacts by location.

11    We examine the potential for impacts by considering (1) the types  of water used for hydraulic
12    fracturing; (2) the amounts of water used per well;  (3) cumulative estimates of water used and
13    consumed for hydraulic fracturing; and (4) a state-by-state assessment of the potential for impacts
14    based on water use, consumption, and availability. We often could not assess the potential for
15    impacts at a finer resolution than the county scale due to lack of available local-scale data for  most
16    areas. Thus, our assessment suggests areas that are more likely than others to experience  impacts,
17    but does not necessarily indicate that these impacts will occur. Three case studies (southern Texas,
18    western Colorado, and eastern Pennsylvania), provide an in-depth examination at finer scales, and
19    we rely on those where possible (see Text Boxes 4-3, 4-4, and 4-5).

      4.6.1.  Major Findings
20    Water for hydraulic fracturing typically comes from surface water, ground water, or reused
21    wastewater. Because trucking can be a major expense, operators often use water sources as close to
22    well pads as possible. Operators usually self-supply surface or ground water directly, but also may
23    obtain water secondarily through public water systems or other suppliers. Hydraulic fracturing
24    operations in the eastern United States generally rely on surface water, whereas operations in more
25    semi-arid to arid western states use mixed surface and ground water supplies. In areas that lack
26    available surface water (e.g., western Texas), ground water supplies most of the water needed for
27    fracturing unless alternative sources, such as reused wastewater, are available and utilized.

28    The vast majority of water used for hydraulic fracturing nationally comes from fresh water sources,
29    although some operators also use lower-quality water (e.g., hydraulic fracturing wastewater,
30    brackish ground water, or small proportions of acid mine drainage and wastewater treatment plant
31    effluent). The use of non-fresh sources can reduce competition for current drinking water
32    resources. Nationally, the  proportion of reused wastewater is generally low as a percentage of
33    injected volume; based on available data, the median reuse of wastewater as a percentage of
34    injected volume is 5% nationally, but this percentage varies by location (see Table 4-1).1 Available
      1 Note that reused water as a percentage of total water injected differs from the percentage of wastewater that is reused
      (see Section 4.2 and Chapter 8).
                   This document is a draft for review purposes only and does not constitute Agency policy.
      June 2015                                    4-47                  DRAFT—DO NOT CITE OR QUOTE

-------
      Hydraulic Fracturing Drinking Water Assessment
 1    data on reuse trends indicate increasing reuse of wastewater over time in both Pennsylvania and
 2    West Virginia, likely due to the lack of nearby disposal options. Reuse as a percentage of water
 3    injected appears to be low in other areas, likely in part because of the relatively high availability of
 4    disposal wells (see Chapter 8).

 5    The median amount of water used per hydraulically fractured well, based on national disclosures to
 6    FracFocus, is approximately 1.5 million gal (5.7 million L) of water (U.S. EPA, 2015a, b). This
 7    estimate represents a variety of fractured well types, including types that use much less water per
 8    well than horizontal shale gas wells. Thus, published estimates for horizontal shale gas wells are
 9    typically higher (e.g., approximately 4 million gal (15 million L) per well (Vengosh etal., 2014]].
10    There is also wide variation within and among states and basins in the median water volumes
11    reported per disclosure, from more than 5 million gal (19 million L] in Arkansas and Louisiana to
12    less than 1 million gal (3.8 million L] in Colorado, Wyoming, Utah, New Mexico, and California (U.S.
13    EPA, 2015b]. This variation results from several factors, including well length, formation geology,
14    and fracturing fluid formulation (see Section 4.3.3].

15    Cumulatively, hydraulic fracturing uses billions of gallons of water every year at the national and
16    state scales, and even in some counties. When expressed as a percentage compared to total water
17    use or consumption at these scales, however, hydraulic fracturing water use and consumption is
18    most often a small percentage, generally less than 1%. This percentage may be higher in specific
19    areas. Annual hydraulic fracturing water use was 10% or more compared to 2010 total water use in
20    6.5% of counties with FracFocus disclosures in 2011 and 2012, 30% or more in 2.2% of counties,
21    and 50% or more in 1.0% of counties (U.S. EPA. 2015a]. Consumption estimates followthe same
22    general pattern, but with slightly higher percentages in each category. In these counties, hydraulic
23    fracturing represents a relatively large user and consumer of water.

24    High hydraulic fracturing water use  or consumption alone does not necessarily result in impacts to
25    drinking water resources. Rather, the potential for impacts depends on both water use or
26    consumption and water availability at a given withdrawal point Our state-by-state assessment
27    examines the intersection between water use or consumption and availability at the county scale.
28    This approach suggests  where the potential for impacts exists, but does not indicate where impacts
29    will occur at the local scale. Where possible, we use local-scale case studies in Texas, Pennsylvania,
30    and Colorado to provide details at finer spatial scales.

31    In our survey of the published literature, we did not find a case where hydraulic fracturing water
32    use by itself caused a drinking water well or stream to run dry. This could indicate an absence of
33    hydraulic fracturing effects on water availability; alternatively, it could reflect that these events are
34    not typically documented in the types of literature we reviewed. Water availability is rarely
35    impacted by just one use or factor alone. For example, drinking water wells in an area overlapping
36    with the Haynesville Shale in northwest Louisiana ran out of water in 2011, due to higher than
37    normal withdrawals and drought (LA Ground Water Resources Commission. 2012]. Hydraulic
38    fracturing water use in the area may have contributed to these conditions, along with other  water
39    uses and the lack of precipitation. Other impacts to drinking water quantity or quality (e.g.,
                   This document is a draft for review purposes only and does not constitute Agency policy.
      June 2015                                    4-48                   DRAFT—DO NOT CITE OR QUOTE

-------
      Hydraulic Fracturing Drinking Water Assessment
 1    declining aquifer levels, decreased stream flow, increased pollutant concentrations) also may occur
 2    before wells and streams actually go dry.

 3    The potential for impacts due to hydraulic fracturing water withdrawals is highest in areas with
 4    relatively high fracturing water use and low water availability. Southern and western Texas are two
 5    locations where hydraulic fracturing water use combined with low water availability, drought, and
 6    reliance on declining ground water sources has the potential to affect the quantity and quality of
 7    drinking water resources. Fracturing withdrawals combined with other intensive uses, particularly
 8    irrigation, could contribute to ground water quality degradation. Any impacts are likely to be
 9    realized locally within these areas. In a detailed case study of southern Texas, Scanlonetal. [2014]
10    observed generally adequate water supplies for hydraulic fracturing, except in specific locations.
11    They found excessive drawdown of local ground water in a small proportion (~6% of the area) of
12    the Eagle Ford play. They suggested water management, particularly a shift towards brackish water
13    use, could minimize potential future impacts to fresh water resources (see Text Box 4-3). County-
14    level data confirm that high brackish water availability in Texas may help offset hydraulic
15    fracturing water demand (see Text Box 4-2).

16    Comparatively, the potential for hydraulic fracturing water acquisition impacts to drinking water
17    quantity and quality appears to be lower—but not entirely eliminated—in other areas of the United
18    States. Detailed case studies in western Colorado and northeastern Pennsylvania did not show
19    impacts, despite indicating that streams could be vulnerable to water withdrawals from hydraulic
20    fracturing (U.S. EPA. 2015c). High wastewater reuse rates in western Colorado eliminated the need
21    for more fresh water withdrawals. In northeast Pennsylvania, water withdrawals for hydraulic
22    fracturing could result in high water consumption-to-stream flow events, but water management
23    (e.g., passby flows) limited the potential for impacts, especially on small streams (U.S. EPA, 2015c).
24    In western North Dakota, ground water is limited, but the industry may have sufficient supplies of
25    surface water from the Missouri River system. These location-specific examples emphasize the
26    need to focus on regional and local dynamics when considering the potential impacts of hydraulic
27    fracturing water acquisition on drinking water resources.

      4.6.2.  Factors Affecting Frequency or Severity of Impacts
28    The potential for hydraulic fracturing water use to affect drinking water resource quantity or
29    quality depends primarily on the amount of water used or consumed versus water availability at a
30    given withdrawal point Potential impacts to drinking water resources reflect all uses, including
31    hydraulic fracturing demands, compared to available water. Areas with high water use, low water
32    availability, slowly replenishing sources, and/or episodic water  shortages (e.g., seasonal or longer-
33    term droughts) are more vulnerable to potential impacts. Areas with high water availability relative
34    to existing uses, high rainfall distributed throughout the year, or high storage capacity, are less
3 5    likely to be affected.

3 6    Water management can alter this  dynamic between water use and availability. The type of water
37    used (e.g., fresh, brackish, reused hydraulic fracturing wastewater, other wastewaters) is a major
38    factor that can either increase or decrease the potential for impacts. Replacing a fresh water source
39    with another type of water can reduce the demand for fresh water and decrease potential

                   This document is a draft for review purposes only and does not constitute Agency policy.
      June 2015                                    4-49                   DRAFT—DO NOT CITE OR QUOTE

-------
      Hydraulic Fracturing Drinking Water Assessment
 1    competition for drinking water. Brackish ground water use may reduce the demand for fresh water
 2    and decrease competition for drinking water currently, but this may change if desalinization for
 3    drinking water becomes more prevalent in the future (see Chapter 3).

 4    The timing and location of water withdrawals can also affect the potential for impacts, particularly
 5    for surface water withdrawals. Withdrawing water from small streams is more likely to result in a
 6    high-consumption-to-stream flow event than removing water from larger streams [U.S. EPA,
 7    2015c). Withdrawals during periods of low stream flow are also more likely to result in impacts
 8    than withdrawals during high flow periods. Hydraulic fracturing operations may have the ability to
 9    withdraw water during periods of high stream flow, and store it for future use during drier periods.

      4.6.3.  Uncertainties
10    There are several uncertainties inherent in our assessment of hydraulic fracturing water use and
11    potential effects on drinking water quantity and quality. The largest uncertainties stem from the
12    lack of literature and data on this subject at local scales, and the question of whether any impacts
13    would be documented in the types of literature we reviewed.

14    We used a state-by-state approach to identify areas where potential impacts are likely, based on
15    relatively high fracturing water use and low water availability. Typically, only data at the county-
16    scale were available. Because impacts occur at smaller spatial scales (i.e., at water withdrawal
17    sites), our assessment suggests the potential for impacts, but does not indicate whether impacts
18    will occur. In only a few places could we use local case studies to determine if potential impacts
19    were realized; these  case studies show that local factors can greatly affect whether drinking water
20    resources are impacted.

21    In our survey of the published literature, we did not find a case where hydraulic fracturing water
22    use alone caused a drinking water well or stream to run dry. This could indicate an absence of
23    hydraulic fracturing  effects on water availability, or it could reflect that these events are not
24    typically documented in the types of literature we reviewed. Water availability is rarely impacted
25    by just one use or factor alone. These issues may have limited our findings.

26    Other uncertainties arise from data limitations regarding the volume and types of water used or
27    consumed for hydraulic fracturing, future water use projections, and water availability estimates.
28    There are no nationally consistent data sources, and therefore water use estimates must be based
29    on multiple, individual pieces of information. For example, in their National Water Census, the USGS
30    includes hydraulic fracturing in the broader category of "mining" water use, but hydraulic
31    fracturing water use  is not reported separately (Maupinetal., 2014]. There are locations where
32    annual average hydraulic fracturing water use in 2011 and 2012 exceeded total mining water use in
33    2010, and one county where it exceeded all water use (U.S. EPA, 2015b: Maupinetal., 2014]. This
34    could be due to a rapid increase in hydraulic fracturing water use, differences in methodology
35    between the two databases (i.e., the USGS 2010 National Water Census and the EPA FracFocus
36    project database), or both.
                   This document is a draft for review purposes only and does not constitute Agency policy.
      June 2015                                    4-50                  DRAFT—DO NOT CITE OR QUOTE

-------
      Hydraulic Fracturing Drinking Water Assessment
 1    The EPA FracFocus project database represents the most extensive database currently available to
 2    estimate hydraulic fracturing water use. However, estimates based on the project database form an
 3    incomplete picture of hydraulic fracturing water use because most states with data in the project
 4    database did not require disclosure to FracFocus during the time period analyzed [U.S. EPA, 2015a]
 5    (see Text Box 4-1). We conclude that this likely does not change the overall hydraulic fracturing
 6    water use patterns observed across the United States, but could affect our assessment of the
 7    potential impacts in specific locations.

 8    Hydraulic fracturing water use data are often provided in terms of water use per well. While this is
 9    valuable information, the potential impacts of water acquisition for hydraulic fracturing could be
10    better assessed if data were also available at the withdrawal point If the total volume, date, and
11    location of each water withdrawal were documented, the quality of the water used and potential
12    effects on availability could be better estimated. For example, surface withdrawal points could be
13    aggregated by watershed to estimate effects on downstream flow. Alternatively, if the location and
14    depth of ground water pumping were documented, these could be aggregated to assess effects on a
15    given aquifer. Some of this information is available in disparate forms, but the lack of nationally
16    consistent data on water withdrawal locations, timing, and amounts—data that are publicly
17    available, easy  to access, and easy to analyze—limits our assessment of hydraulic fracturing water
18    use.

19    Future hydraulic fracturing water use is also a source of uncertainty. Because water withdrawals
20    and potential impacts are  concentrated in certain localized areas, water use projections need to
21    match this scale. Projections are available for Texas at the county scale, but more information at the
22    county or sub-county scale is needed in other states with high hydraulic  fracturing activity and
23    water availability concerns (e.g., northwest North Dakota, eastern Colorado). Due to a lack of data,
24    we generally could not assess future cumulative water use and the potential for impacts in most
25    areas of the country, nor could we examine these in combination with other relevant factors (e.g.,
26    climate change, population growth).

      4.6.4.  Conclusions
2 7    Water acquisition for hydraulic fracturing has the potential to impact drinking water resources by
28    affecting drinking water quantity and quality (see Text Box 4-6). In our survey of the published
29    literature, we did not find a case where hydraulic fracturing water use by itself caused a drinking
30    water well or stream to run dry. However, the potential for impacts to drinking water quantity and
31    quality exists and is highest in areas with relatively high fracturing water use and low water
32    availability. Southern and western Texas are two locations where the potential appears highest due
33    to the combined effects of high hydraulic fracturing activity, low water availability, drought, and
34    reliance on declining ground water sources. Even in locations where water is generally plentiful,
35    localized impacts can still  occur in certain instances. Excessive ground water pumping can cause
36    localized drawdowns; surface water withdrawals can affect stream flow, particularly in smaller
37    streams or during low flow periods. These findings emphasize the need to focus on regional and
38    local dynamics when examining potential impacts  of hydraulic fracturing water acquisition on
39    drinking water quantity and quality.
                   This document is a draft for review purposes only and does not constitute Agency policy.
      June 2015                                    4-51                  DRAFT—DO NOT CITE OR QUOTE

-------
      Hydraulic Fracturing Drinking Water Assessment
      Text Box 4-6. Research Questions Revisited.

 1    What are the types of water used for hydraulic fracturing?

 2    •   Water for hydraulic fracturing typically comes from surface, ground water, or reused wastewater.
 3        Operators often use water sources as close to well pads as possible as trucking is a major expense.
 4        Operators usually self-supply surface or ground water directly, but also may obtain water secondarily
 5        through public water systems or other suppliers. Hydraulic fracturing operations in the eastern United
 6        States generally rely on surface water, whereas operations in more semi-arid to arid western states use
 7        mixed surface and ground water supplies. In areas that lack available surface water (e.g., western Texas),
 8        ground water supplies most of the water needed for fracturing unless alternative sources, such as reused
 9        wastewater, are available and utilized.

10    •   The vast majority of water used nationally comes from fresh water sources, although some operators also
11        use lower-quality water (e.g., hydraulic fracturing wastewater, brackish ground water, or small
12        proportions of acid mine drainage and wastewater treatment plant effluent). The use of non-fresh
13        sources can reduce competition for current drinking water resources. Nationally, the proportion of
14        reused wastewater is generally low as a percentage of injected volume; based on available data, median
15        reuse of wastewater across all basins and plays is 5% of injected volume (see Table 4-1). Available data
16        on reuse trends indicate increasing reuse of wastewater over time in both Pennsylvania and West
17        Virginia, likely due to the lack of nearby disposal options. Reuse as a percentage of water injected appears
18        to be low in other areas, likely in part because of the relatively high availability of disposal wells (see
19        Chapter 8).

2 0    How m uch water is used per well?

21    •   The median amount of water used per hydraulically fractured well, based on national disclosures to
22        FracFocus, is approximately 1.5 million gal (5.7 million L) of water (U.S. EPA. 2015a. b). This estimate
23        represents a variety of fractured well types. There is also wide variation within and among states and
24        basins in the median water volumes reported per disclosure, from more than 5 million gal (19 million L)
25        in Arkansas and Louisiana to less than 1 million gal (3.8 million L) in Colorado, Wyoming, Utah, New
26        Mexico, and California (U.S. EPA. 2015b). This variation results from several factors, including well
27        length, formation geology, and fracturing fluid formulation (see Section 4.3.3).

28    •   Trends indicate that water use per well is increasing in certain locations as horizontal well lengths
29        increase. This may not, however, increase water use per unit energy extracted.
                     This document is a draft for review purposes only and does not constitute Agency policy.
      June 2015                                        4-52                    DRAFT—DO NOT CITE OR QUOTE

-------
      Hydraulic Fracturing Drinking Water Assessment
 1    How might cumulative water withdrawals for hydraulic fracturing affect drinking water quantity?
 2    •   Cumulatively, hydraulic fracturing uses billions of gallons of water every year at the national and state
 3        scales, and even in some counties. When expressed as a percentage compared to total water use or
 4        consumption at these scales, however, hydraulic fracturing water use and consumption is most often a
 5        small percentage, generally less than 1%. This percentage may be higher in specific areas. Annual
 6        hydraulic fracturing water use was 10% or more compared to 2010 total water use in 6.5% of counties
 7        with FracFocus disclosures in 2011 and 2012, 30% or more in 2.2% of counties, and 50% or more in
 8        1.0% of counties [U.S. EPA. 2015a). Consumption estimates follow the same general pattern, but with
 9        slightly higher percentages in each category. In these counties, hydraulic fracturing represents a
10        relatively large user and consumer of water.
11    •   High hydraulic fracturing water use or consumption alone does not necessarily result in impacts to
12        drinking water resources. Rather, the potential for impacts depends on both water use or consumption
13        and water availability at a given withdrawal point. Our state-by-state assessment examines the
14        intersection between water use or consumption and availability at the county scale. This approach
15        suggests where the potential for impacts exists, but does not indicate where impacts will occur at the
16        local scale. Local-scale case studies help provide details at finer spatial scales.
17    •   In our survey of the published literature, we did not find a case where hydraulic fracturing water use by
18        itself caused a drinking water well or stream to run dry. This could indicate an absence of hydraulic
19        fracturing effects on water availability, or it could reflect that these events are not typically documented
20        in the types of literature we reviewed. Water availability is rarely impacted by just one use or factor
21        alone. For example, drinking water wells in an area overlapping with the Haynesville Shale in northwest
22        Louisiana ran out of water in 2011, due to higher than normal withdrawals and drought (LA Ground
23        Water Resources Commission. 2012]. Hydraulic fracturing water use in the area may have contributed to
24        these conditions, along with other water uses and the lack of precipitation. Other impacts to drinking
25        water quantity or quality (e.g., declining aquifer levels, decreased stream flow, increased pollutant
26        concentrations) also may occur before wells and streams actually go dry.
27    •   The potential for impacts due to hydraulic fracturing water withdrawals is highest in areas with
2 8        relatively high fracturing water use and low water availability. Southern and western Texas are two
29        locations where hydraulic fracturing water use combined with low water availability, drought, and
3 0        reliance on declining ground water sources has the potential to affect the quantity of drinking water
31        resources. Any impacts are likely to be realized locally within these areas. In a detailed case study of
32        southern Texas, Scanlon etal. (2014) observed generally adequate water supplies for hydraulic
33        fracturing, except in specific locations. They found excessive drawdown of local ground water in a small
34        proportion (~6% of the area) of the Eagle Ford play. They suggested water management, particularly a
35        shift towards brackish water use, could minimize potential future impacts to fresh water resources (see
36        Text Box 4-3). County-level data confirm that high brackish water availability in Texas may help offset
37        hydraulic fracturing water demand (see Text Box 4-2).
38    •   The potential for hydraulic fracturing water acquisition impacts to drinking water quantity and quality
39        appears to be lower—but not entirely eliminated—in other areas of the United States. Detailed case
40        studies in western Colorado and northeastern Pennsylvania did not show impacts, despite indicating that
41        streams could be vulnerable to water withdrawals from hydraulic fracturing (U.S. EPA. 2015c). High
42        wastewater reuse rates  in western Colorado eliminated the need for more fresh water withdrawals. In
43        northeast  Pennsylvania, water withdrawals for hydraulic fracturing could result in high water
44        consumption-to-stream flow events, but water management (e.g., passby flows) limited the potential for
45        impacts, especially on small streams (U.S. EPA, 2015c). In western North Dakota, ground water is limited,
46        but the industry may have sufficient supplies of surface water from the Missouri River system. These
                     This document is a draft for review purposes only and does not constitute Agency policy.
      June 2015                                       4-53                    DRAFT—DO NOT CITE OR QUOTE

-------
      Hydraulic Fracturing Drinking Water Assessment
 1        location-specific examples emphasize the need to focus on regional and local dynamics when considering
 2        the potential impacts of hydraulic fracturing water acquisition on drinking water resources.

 3    What are the possible impacts of water withdrawals for hydraulic fracturing on water quality?

 4    •   Water withdrawals for hydraulic fracturing, similar to all water withdrawals, have the potential to alter
 5        the quality of drinking water resources. Ground water withdrawals exceeding natural recharge rates
 6        decrease water storage in aquifers, potentially mobilizing contaminants or allowing the infiltration of
 7        lower-quality water from the land surface or adjacent formations. Withdrawals could also decrease
 8        ground water discharge to streams, potentially affecting surface water quality. Areas with numerous
 9        high-capacity wells and large amounts of sustained ground water pumping are most likely to experience
10        impacts, particularly in drought-prone regions with limited ground water recharge.

11    •   Surface water withdrawals also have the potential to affect water quality. Withdrawals may lower water
12        levels and alter stream flow, potentially decreasing a stream's capacity to dilute contaminants. Case
13        studies by the EPA show that streams can be vulnerable to changes in water quality due to water
14        withdrawals, particularly smaller streams and during periods of low flow fU.S. EPA. 2015c). Management
15        of the rate and timing of surface water withdrawals can help mitigate potential impacts of fracturing
16        withdrawals on water quality.

17    •   Like water quantity effects, any effects of water withdrawals on water quality will likely occur nearest the
18        withdrawal point, again emphasizing the need for location specific assessments.
      4.7.  References for Chapter 4

      AMEC. Hinckley.. HDR. (AMEC Environment & Infrastructure, Inc, Hinckley Consulting, HDR Engineering, Inc).
         (2014). Hydrogeologic study of the Laramie County control area. Prepared for the Wyoming State
         Engineers Office. Cheyenne, WY: Wyoming State Engineer's Office, http://seo.wyo.gov/seo-
         files/Final%20Draft%20Corrected%20Stamped.pdf?attredirects=0&d=l

      ANRC (Arkansas Natural Resources Commission). (2014). Non-riparian water use certification. Available
         online at http://anrc.ark.org/divisions/water-resources-management/non-riparian-water-use-
         certification-program/

      Bartos. TT: Hallberg. LL. (2011). Generalized potentiometric surface, estimated depth to water, and estimated
         saturated thickness of the high plains aquifer system, Marchjune 2009, Laramie County, Wyoming.
         Available online at http://pubs.usgs.gov/sim/3180/

      Bene. PG: Harden. B: Griffin. SW: Nicot. IP. (2007). Northern Trinity/Woodbine aquifer groundwater
         availability model: Assessment of groundwater use in the Northern Trinity aquifer due to urban growth
         and Barnett shale development. (TWDB Contract Number: 0604830613). Austin, TX: R. W. Harden &
         Associates, Inc.
         http://www.twdb.state.tx.us/groundwater/models/gam/trnt n/TRNT N Barnett Shale  Reportpdf

      Benotti. Ml: Stanford. BD: Snyder. SA. (2010). Impact of drought on wastewater contaminants in an urban
         water supply. J Environ Qual 39:1196-1200.

      BLM (Bureau of Land Management). (2013b). Hydraulic fracturing white paper, appendix e. Casper, WY:
         Bureau of Land Management, Wyoming State Office.
         http://www.blm.gOV/pgdata/etc/medialib/blm/wy/information/NEPA/og/2014/02feb.Par.49324.File.d
         at/vlAppE.pdf
                     This document is a draft for review purposes only and does not constitute Agency policy.

      June 2015                                      4-54                   DRAFT—DO NOT CITE OR QUOTE

-------
Hydraulic Fracturing Drinking Water Assessment
Boghici. R. (2009). Water quality in the Carrizo-Wilcox aquifer, 19902006. (Report 372). Austin, TX: Texas
   Water Development Board.
   http://www.twdb.texas.gov/publications/reports/numbered reports/doc/R372Carrizo-Wilcox.pdf

California Department of Water Resources. (2015). California state water project overview. Available online
   athttp://www.water.ca.gov/swp/ (accessed February 20,2015).

CCST (California Council on Science and Technology). (2014). Advanced well stimulation technologies in
   California: An independent review of scientific and technical information. Sacramento, CA.
   http://ccst.us/publications/2014/2014wst.pdf

Ceres (Coalition for Environmentally Responsible Economies). (2014). Hydraulic fracturing & water stress:
   water demand by the numbers. Boston, Massachusetts, https://www.ceres.org/issues/water/shale-
   energy/shale-and-water-maps/hydraulic-fracturing-water-stress-water-demand-by-the-numbers

Chaudhuri. S: Ale. S. (2013). Characterization of groundwater resources in the Trinity and Woodbine aquifers
   in Texas. Sci Total Environ 452: 333-348. http://dx.doi.Org/10.1016/j.scitotenv.2013.02.081

Chaudhuri, S: Ale, S. (2014a). Long term (1960-2010) trends in groundwater contamination and salinization
   in the Ogallala aquifer in Texas. J Hydrol 513: 376-390. http://dx.doi.Org/10.1016/j.jhydrol.2014.03.033

Chaudhuri. S: Ale. S. (2014b). Temporal evolution of depth-stratified groundwater salinity in municipal wells
   in the major aquifers in Texas, USA. Sci Total Environ 472: 370-380.
   http://dx.doi.0rg/10.1016/j.scitotenv.2013.10.120

Clark. CE: Horner. RM: Harto. CB. (2013). Life Cycle Water Consumption for Shale Gas and Conventional
   Natural Gas. Environ Sci Technol 47:11829-11836. http://dx.doi.org/10.1021/es4013855

Colorado Division of Water Resources: Colorado Water Conservation Board: Colorado Oil and Gas
   Conservation  Commission. (2014). Water sources and demand for the hydraulic fracturing of oil and gas
   wells in Colorado from 2010 through 2015 [Fact Sheet], http://cewc.colostate.edu/2012/02/water-
   sources-and-demand-for-the-hydraulic-fracturing-of-oil-and-gas-wells-in-colorado-from-2010-through-
   20157
Economides. Ml: Hill. A. d: Ehlig-Economides. C: Zhu. D. (2013). Petroleum production systems. In Petroleum
   production systems (2nd ed.). Englewood Cliffs, NJ: Prentice Hall.

EERC (Energy and Environmental Research Center, University of North Dakota). (2010). Bakken water
   opportunities assessment phase 1. (2010-EERC-04-03). Grand Forks, ND: Energy and Environmental
   Research Center. http://www.undeerc.org/bakken/pdfs/FracWaterPhaseIreport.pdf

EERC (Energy and Environmental Research Center, University of North Dakota). (2011). Bakken water
   opportunities assessment phase 2: evaluation of brackish groundwater treatment for use in hydraulic
   fracturing of the Bakken Play, North Dakota. (2011-EERC-12-05). Grand Forks, ND: Energy and
   Environmental Research Center. http://www.undeerc.org/Water/pdf/BakkenWaterOppPhase2.pdf

EERC (Energy and Environmental Research Center, University of North Dakota). (2013). BakkenSmart:  water
   [Fact Sheet]. Grand Forks, ND: Energy and Environmental Research Center.
   http://www.undeerc.org/bakken/pdfs/NDIC-NDPC-Water-Fact-Sheet.pdf

EIA (Energy Information Administration). (2015b). Lower 48 states shale plays. Available online at
   http://www.eia.gov/oil gas/rpd/shale gas.pdf

Entrekin. S: Evans-White. M: lohnson. B: Hagenbuch. E. (2011). Rapid expansion of natural gas development
   poses a threat to surface waters. Front Ecol Environ 9: 503-511. http://dx.doi.org/10.1890/110053

Environment Canada. (2004). Threats to Water Availability in Canada, http://www.ec.gc.ca/inre-
   nwri/default.asp?lang=En&n=OCD66675-l
               This document is a draft for review purposes only and does not constitute Agency policy.

June 2015                                       4-55                    DRAFT—DO NOT CITE OR QUOTE

-------
Hydraulic Fracturing Drinking Water Assessment
Georgakakos. A: Fleming. P: Dettinger. M: Peters-Lidard. C: Richmond. TC: Reckhow. K: White. K: Yates. D.
   (2014). Water resources. In JM Melillo; TC Richmond; GW Yohe (Eds.), Climate change impacts in the
   United States (pp. 69-112). Washington, D.C.: U.S. Global Change Research Program.
   http://www.globalchange.gov/ncadac

George. PG: Mace. RE: Petrossian. R. (2011). Aquifers of Texas. (Report 380). Austin, TX: Texas Water
   Development Board.
   http://www.twdb.state.tx.us/publications/reports/numbered reports/doc/R380 AquifersofTexas.pdf

Goodwin. S: Carlson. K: Knox. K: Douglas. C: Rein. L. (2014). Water intensity assessment of shale gas resources
   in the Wallenberg field in norlheaslern Colorado. Environ Sci Technol 48: 5991-5995.
   hllp://dx.doi.org/10.1021/es404675h

Gregory. KB: Vidic. RD: Dzombak. DA. (2011). Water managemenl challenges associated wilh Ihe production
   of shale gas by hydraulic fracluring. Elemenls 7:181-186.

GWPC (Groundwaler Protection Council). (2015). FracFocus - chemical disclosure regislry. Available online al
   hllp://fracfocus.org/

GWPC and ALL Consulting (Ground Water Protection Council (GWPC) and ALL Consulting). (2009). Modern
   shale gas developmenl in Ihe United Slates: A primer. (DE-FG26-04NT15455). Washington, DC: U.S.
   Deparlmenl of Energy, Office of Fossil Energy and National Energy Technology Laboratory.
   hllp://www.gwpc.org/siles/defaull/files/Shale%20Gas%20Primer%202 009.pdf

Hansen. E: Mulvaney. D: Belcher. M. (2013). Water resource reporting and water foolprinl from Marcellus
   Shale developmenl in Wesl Virginia and Pennsylvania. Durango, CO: Earlhworks Oil & Gas Accountability
   Project hllp://www.downslreamslralegies.com/documenls/reporls publicalion/marcellus wv pa.pdf

IPCC (Intergovernmental Panel on Climate Change). (2007). Climate change 2007: Imparts,  adaptation and
   vulnerability. Cambridge, UK: Cambridge University Press. hllp://www.ipcc.ch/ipccreporls/ar4-wg2.hlm

lackson. RB: Carpenter. SR: Dahm. CN: Mcknight DM: Naiman. Rl: Poslel. SL: Running. SW. (2001). Water in a
   changing world. Ecol Appl 11:1027-1045. hllp://dx.doi.org/10.1890/1051-
   0761(2001)011[1027:WIACW]2.0.CO:2

liang, M: Hendrickson, CT: Vanbriesen, IM. (2014). Life Cycle Water Consumption and Waslewaler Generation
   Imparts of a Marcellus Shale Gas Well. Environ Sci Technol 48:1911-1920.
   hllp://dx.doi.org/10.1021/es4047654

Kansas Water Office. (2014). How is water used in oil and gas exploration in Kansas? Topeka,  KA.
   hllp://www.kwo.org/aboul us/BACs/KWIF/rpl Hydraulic%20Fracluring KS Water FAQ 03082012 fina
   1 ki.pdf

Kargbo. DM: Wilhelm. RG: Campbell. PI. (2010). Nalural gas plays in Ihe Marcellus Shale: Challenges and
   potential opportunities. Environ Sci Technol 44: 5679-5684. http://dx.doi.org/10.1021/es903811p

Kenny. IF: Barber. NL: Hulson. SS: Linsey. KS: Lovelace. IK: Maupin. MA. (2009). Estimated use of water in Ihe
   United Stales in 2005. (Circular 1344). Reslon, VA: U.S. Geological Survey.
   hllp://pubs.usgs.gov/circ/1344/

Konikow. LF. (2013a). Groundwaler depletion in Ihe United Stales (19002008). (USGS Scientific
   Investigations Report 2013). Reslon, VA: U.S. Geological Survey. hllp://pubs.usgs.gov/sir/2013/5079

Konikow. LF: Kendy. E. (2005). Groundwaler depletion: A global problem. Hydrogeo J13: 317-320.
   hllp://dx.doi.org/10.1007/sl0040-004-0411-8

Kulhnerl. N: Werline. R: Nichols. K. (2012). Water reuse and recycling in Ihe oil and gas induslry: Devons
   water managemenl success. Presentation presenled al 2nd Annual Texas Water Reuse Conference, July
   20,2012, Forth Worth, TX.
              This document is a draft for review purposes only and does not constitute Agency policy.

June 2015                                     4-56                   DRAFT—DO NOT CITE OR QUOTE

-------
Hydraulic Fracturing Drinking Water Assessment
LA Ground Water Resources Commission (Louisiana Ground Water Resources Commission). (2012).
   Managing Louisianas Groundwater Resources: An interim report to the Louisiana Legislature. Baton
   Rouge, LA: Louisiana Department of Natural Resources.
   http:// dnr.louisiana.gov/index. cfm?md=pagebuilder&tmp=home&pid=907

Laurenzi. II: Jersey. GR. (2013). Life cycle greenhouse gas emissions and freshwater consumption of Marcellus
   shale gas. Environ Sci Technol 47: 4896-4903. http://dx.doi.org/10.1021/es305162w

LDEQ (Louisiana Department of Environmental Quality). (2008). Ground water use advisory: Commissioner
   of conservation recommends wise water use planning in the Haynesville Shale.
   http:// dnr.louisiana.gov/index. cfm?md=newsroom&tmp=detail&aid=509

Maupin. MA: Kenny. IF: Hutson. SS: Lovelace. IK: Barber. NL: Linsey. KS. (2014). Estimated use of water in the
   United States in 2010. (USGS Circular 1405). Reston, VA: U.S. Geological Survey.
   http://dx.doi.org/10.3133/cirl405

McKay. SF: King. Al. (2006). Potential ecological effects of water extraction in small, unregulated streams.
   River Research and Applications 22:1023-1037. http://dx.doi.org/10.1002/rra.958

Mitchell. AL: Small. M: Gasman. EA. (2013a). Surface water withdrawals for Marcellus Shale gas development:
   performance of alternative regulatory approaches in the Upper Ohio River Basin. Environ Sci Technol 47:
   12669-12678. http://dx.doi.org/10.1021/es403537z

Murdoch. PS: Baron. IS: Miller. TL. (2000). Potential effects of climate chance on surface-water quality in
   North America. J Am Water Resour Assoc 36: 347-366.

Murray. KE. (2013). State-scale perspective on water use and production associated with oil and gas
   operations, Oklahoma, U.S. Environ Sci Technol 47: 4918-4925. http://dx.doi.org/10.1021/es4000593

National Drought Mitigation Center. (2015). U.S. drought monitor. Available online at
   http://droughtmonitor.unl.edu/Home.aspx (accessed February 27, 2015).

Nicot. IP: Reedy. RC: Costley. RA: Huang. Y. (2012). Oil & gas water use in Texas: Update to the 2011 mining
   water use report. Nicot, JP; Reedy, RC; Costley, RA; Huang, Y.
   http://www.twdb.state.tx.us/publications/reports/contracted reports/doc/0904830939 2012Update M
   iningWaterUse.pdf

Nicot. IP: Scanlon. BR. (2012). Water use for shale-gas production in Texas, U.S. Environ Sci Technol 46: 3580-
   3586. http://dx.doi.org/10.1021/es204602t

Nicot. IP: Scanlon. BR: Reedy. RC:  Costley. RA. (2014). Source and fate of hydraulic fracturing water in the
   Barnett Shale: a historical perspective. Environ Sci Technol 48: 2464-2471.
   http://dx.doi.org/10.1021/es404050r

NM OSE (New Mexico Office of the State Engineer). (2013). New Mexico water use by categories 2010.
   (Technical Report 54). Santa Fe, NM: New Mexico Office of the State Engineer, Water Use and
   Conservation Bureau.
   http://www.ose.state.nm.us/Pub/TechnicalReports/TechReport%2054NM%20Water%20Use%20by%2
   OCategories%20.pdf

North Dakota Department of Mineral Resources. (2013). North Dakota Department of Mineral Resources:
   Government Finance Interim Committee 12/12/2013. Presentation presented at Department of Mineral
   Resources: Update on the Status of Oil and Gas Development in the State, 12/12/2013, Bismarck,  ND.

North Dakota State Water Commission. (2010). Water appropriation requirements, current water use, &
   water availability for energy industries in North Dakota: a 2010 summary. Bismarck, ND.
   http://www.swc.nd.gov/4dlink9/4dcgi/GetContentPDF/PB-1800/W&E%20RPT%20FinalR.pdf

North Dakota State Water Commission. (2014). Facts about North Dakota (racking and water use. Bismarck,
   ND. http://www.swc.nd.gov/4dlink9/4dcgi/GetContentPDF/PB-2419/Fact%20Sheet.pdf
               This document is a draft for review purposes only and does not constitute Agency policy.

June 2015                                      4-57                   DRAFT—DO NOT CITE OR QUOTE

-------
Hydraulic Fracturing Drinking Water Assessment
NYSDEC (New York State Department of Environmental Conservation). (2011). Revised draft supplemental
   generic environmental impact statement (SGEIS) on the oil, gas and solution mining regulatory program:
   Well permit issuance for horizontal drilling and high-volume hydraulic fracturing to develop the Marcellus
   shale and other low-permeability gas reservoirs. Albany, NY: NY SDEC.
   http://www.dec.ny.gov/energy/75370.html

Ohio EPA (Ohio Environmental Protection Agency). (2012a). Considerations for public water systems prior to
   providing raw or treated water to oil and natural gas companies.
   http://www.epa.state.oh.us/Portals/0/general%20pdfs/Considerations%20for%20Public%20Water%2
   OSystems%2 OPrior%2 Oto%2 OProviding%2 ORaw%2 Oor%2 OTreated%2 OWater%2 Oto%2 00il%2 Oand%2
   ONatural%20Gas%2 OCompanies.pdf

Ohio EPA (Ohio Environmental Protection Agency). (2012b). Ohios regulations: a guide for operators drilling
   in the Marcellus and Utica Shales. Columbus, OH.
   http://www.epa.state.oh.us/Portals/0/general%20pdfs/Ohio%20Regulations%20-
   %2 OA%2 OGuide%2 Ofor%2 00perators%2 ODrilling%2 Oin%2 Othe%2 OMarcellus%2 Oand%2 OUtica%2 OSh
   ales.pdf

OWRB (Oklahoma Water Resources Board). (2014). The Oklahoma comprehensive water plan. Available
   online at http://www.owrb.ok.gov/supply/ocwp/ocwp.php

Plummer. M: Wood. T: Huang. H: Guo. L: Reiten. I: Chandler. K: Metesh. I. (2013). Water needs and availability
   for hydraulic fracturing in the Bakken formation, eastern Montana. Presentation presented at 2013
   Technical Workshop, Water Acquisition Modeling: Assessing Impacts Through Modeling and Other Means,
   June 4, 2013, Arlington, VA.

Rahm. BG: Riha. SI. (2012). Toward strategic management of shale gas development: Regional, collective
   impacts  on water resources. Environ Sci Pol 17:12-23. http://dx.doi.Org/10.1016/j.envsci.2011.12.004

Rolls. Rl: Leigh. C: Sheldon. F. (2012). Mechanistic effects of low-flow hydrology on riverine ecosystems:
   ecological principles and consequences of alteration. Freshwater Science 31:1163-1186.
   http://dx.doi.0rg/10.1899/12-002.l

Roy. SB: Ricci. PF: Summers. KV: Chung. CF: Goldstein. RA. (2005). Evaluation of the sustainability of water
   withdrawals in the United States, 1995 to  2025. J Am Water Resour Assoc 41: 1091-1108.

Scanlon. BR: Reedy. RC: Nicot. IP. (2014). Will water scarcity in semiarid regions limit hydraulic fracturing of
   shale plays? Environmental Research Letters 9. http://dx.doi.Org/10.1088/1748-9326/9/12/124011

Schindler. DW. (1997). Widespread effects of climatic warming on freshwater ecosystems in North America.
   Hydrolog Process 11:1043-1067.

Slutz, I: Anderson, I: Broderick, R: Horner, P. (2012). Key shale gas water management strategies: An
   economic assessment tool. Paper presented at International Conference on Health, Safety and
   Environment in Oil and Gas Exploration and Production, September 11-13,2012, Perth, Australia.

Solley. WB:  Pierce. RR: Perlman. HA. (1998). Estimated use of water in the United States in 1995. (USGS
   Circular: 1200). U.S. Geological Survey, http://pubs.er.usgs.gov/publication/cirl200

SRBC (Susquehanna River Basin Commission). (2013). Comprehensive plan for the water resources of the
   Susquehanna River basin. Harrisburg, PA. http://www.srbc.net/planning/comprehensiveplan.htm

STRONGER (State Review of Oil and Natural Gas  Environmental Regulations). (2011a). Louisiana hydraulic
   fracturing state review. Oklahoma City, OK.
   http://www.strongerinc.org/sites/all/themes/stronger02/downloads/Final%20Louisiana%20HF%20Re
   view%203-2011.pdf

STRONGER (State Review of Oil and Natural Gas  Environmental Regulations). (2011b). Ohio hydraulic
   fracturing state review. Oklahoma City, OK.
   http://www.strongerinc.org/sites/all/themes/stronger02/downloads/Final%20Report%20of%202011
   %200H%20HF%20Review.pdf


              This document is a draft for review purposes only and does not constitute Agency policy.
June 2015                                      4-58                   DRAFT—DO NOT CITE OR QUOTE

-------
Hydraulic Fracturing Drinking Water Assessment
STRONGER (State Review of Oil and Natural Gas Environmental Regulations). (2012). Arkansas hydraulic
   fracturing state review. Oklahoma City, OK. http://www.aogc.state.ar.us/notices/AR HFR FINAL.pdf

Taylor. A. (2012). Watering the boom in Oklahoma: supplies, demands, and neighbors. Presentation
   presented at 2012 Kansas Water Issues Forums, February 29-March 1, 2012, Wichita and Hays, Kansas.

Tidwell. VC: Kobos. PH: Malczynski. L. enA: Klise. G: Castillo. CR. (2012). Exploring the water-thermoelectric
   power nexus. J Water Resour Plann Manag 138: 491-501. http://dx.doi.org/10.1061/rASCE) WR.1943-
   5452.0000222

Tidwell. VC: Zemlick. K: Klise. G. (2013). Nationwide water availability data for energy-water modeling.
   Albuquerque, New Mexico:  Sandia National Laboratories, http://prod.sandia.gov/techlib/access-
   control.cgi/2013/139968.pdf

TWDB (Texas Water Development Board). (2012). Water for Texas 2012 state water plan. Austin, TX.
   http://www.twdb.state.tx.us/waterplanning/swp/2012/index.asp

Tyrrell. P. (2012). Water needs for oil & gas well drilling and fracturing. Presentation presented at 85th
   Annual AWSE Fall Conference, Septermber 23-26,2012, Omaha, Nebraska.

Tyrrell. P. (2013). Wyoming update: water rights for hydraulic fracturing. Presentation presented at Summer
   172nd Western States Water Council Meeting, June 24-26,2013, Casper, Wyoming.

U.S. Army Corps of Engineers. (2011). Final Garrison Dam/Lake Sakakawea project, North Dakota surplus
   water report. Volume 1. Omaha, NE: The U.S. Army Corps of Engineers, Omaha District
   http://www.swc.nd.gov/4dlink9/4dcgi/GetSubContentPDF/PB-
   2811/Garrison%20Dam%20Lake%20Sakakawea%20Surplus%20Water%20Report.pdf

U.S. Census Bureau. (2014). American FactFinder. Available online at
   http://factfinder.census.gov/faces/nav/jsf/pages/index.xhtml

U.S. EPA (U.S. Environmental Protection Agency). (2013a). Data received from oil and gas exploration and
   production companies, including hydraulic fracturing service companies 2011 to 2013. Non-confidential
   business information source documents are located in Federal Docket ID: EPA-HQ-ORD2010-0674.
   Available at http://www.regulations.gov.

U.S. EPA (U.S. Environmental Protection Agency). (2015a). Analysis of hydraulic fracturing fluid data from the
   FracFocus chemical disclosure registry 1.0 [EPA Report]. (EPA/601/R-14/003). Washington, D.C.: Office
   of Research and Development, U.S. Environmental Protection Agency.
   http://www2.epa.gov/hfstudy/analysis-hydraulic-fracturing-fluid-data-fracfocus-chemical-disclosure-
   registry-1-pdf

U.S. EPA (U.S. Environmental Protection Agency). (2015b). Analysis of hydraulic fracturing fluid data from the
   FracFocus chemical disclosure registry 1.0: Project database [EPA Report]. (EPA/601/R-14/003).
   Washington, D.C.: U.S. Environmental Protection Agency, Office of Research and Development.
   http://www2.epa.gov/hfstudy/epa-project-database-developed-fracfocus-l-disclosures

U.S. EPA (U.S. Environmental Protection Agency). (2015c). Case study analysis of the impacts of water
   acquisition for hydraulic fracturing on local water availability [EPA Report]. (EPA/600/R-14/179).
   Washington, D.C.

U.S. EPA (U.S. Environmental Protection Agency). (20151). Retrospective case study in the Raton Basin,
   Colorado: study of the potential impacts of hydraulic fracturing  on drinking water resources [EPA Report].
   (EPA 600/R-14/091). Washington, D.C.

U.S. GAP (U.S. Government Accountability Office). (2014). Freshwater: Supply concerns continue, and
   uncertainties complicate planning. Report to Congressional requesters. (GAO-14-430). Washington, DC:
   U.S. Government Accountability Office (GAO). http://www.gao.gov/assets/670/663343.pdf

USGS (U.S. Geological Survey).  (2003). Ground-Water depletion across the nation, http://pubs.usgs.gov/fs/fs-
   103-037
               This document is a draft for review purposes only and does not constitute Agency policy.

June 2015                                       4-59                    DRAFT—DO NOT CITE OR QUOTE

-------
Hydraulic Fracturing Drinking Water Assessment
USGS (U.S. Geological Survey). (2007). Water-quality assessment of the high plains aquifer, 19992004.
   (Professional Paper 1749). Reston, VA. http://pubs.usgs.gov/pp/1749/downloads/pdf/P1749front.pdf

USGS (U.S. Geological Survey). (2009). Water quality in the high plains aquifer, Colorado, Kansas, Nebraska,
   New Mexico, Oklahoma, South Dakota, Texas, and Wyoming, 19992004. Reston, VA.
   http://pubs.usgs.gov/circ/1337/
USGS (U.S. Geological Survey). (2014d). Trends in major-ion constituents and properties for selected
   sampling sites in the tongue and powder river watersheds, Montana and Wyoming, based on data
   collected during water years 19802010. (Scientific Investigations Report 20135179). Reston, VA.
   http://pubs.usgs.gov/sir/2013/5179/

USGS (U.S. Geological Survey). (2014g). WaterWatch. Available online at http://waterwatch.usgs.gov/

USGS (U.S. Geological Survey). (2014h). Withdrawal and consumption of water by thermoelectric power
   plants in the United States, 2010. (Scientific Investigations Report 20145184). Reston, VA.
   http://dx.doi.org/10.3133/sir20145184

USGS (U.S. Geological Survey). (2015). Trends in hydraulic fracturing distributions and treatment fluids,
   additives, proppants, and water volumes applied to wells drilled in the United States from 1947 through
   2010data analysis and comparison to the literature. (U.S. Geological Survey Scientific Investigations
   Report 20145131). Reston, VA. http://dx.doi.org/10.3133/sir20145131

van Vliet. MTH: Zwolsman. JIG. (2008). Impact of summer droughts on the water quality of the Meuse river. J
   Hydrol 353: 1-17. http://dx.doi.Org/10.1016/i.ihydrol.2008.01.001
Veil. I A. (2011). Water management practices used by Fayetteville shale gas producers. (ANL/EVS/R-11/5).
   Washington, DC: U.S. Department of Energy, National Energy Technology Laboratory.
   http://www.ipd.anl.gov/anlpubs/2011/06/70192.pdf

Vengosh. A: Jackson. RB: Warner. N: Darrah. TH: Kondash. A. (2014). A critical review of the risks to water
   resources from unconventional shale gas development and hydraulic fracturing in the United States.
   Environ Sci Technol 48: 36-52.  http://dx.doi.org/10.1021/es405118y
Verdegem. MCI: Bosnia. RH. (2009). Water withdrawal for brackish and inland aquaculture, and options to
   produce more fish in ponds with present water use. Water Policy 11: 52-68.
   http://dx.doi.org/10.2166/wp.2009.003

Vidic, RD: Brantley, SL: Vandenbossche, IM: Yoxtheimer, D: Abad, ID. (2013). Impact of shale gas development
   on regional water quality [Review]. Science 340: 1235009. http://dx.doi.org/10.1126/science.1235009

Water Research Foundation. (2014). Water quality impacts of extreme weather-related events.
   http://www.waterrf.org/Pages/Projects.aspx?PID=4324
WAWSA (Western Area Water Supply Authority). (2011). Project progress report: western area water supply
   project: appendix N. Williston, ND. http://www.legis.nd.gov/assembly/62-
   2011/docs/pdf/wrlOlOllappendixn.pdf

West Virginia PEP (West Virginia Department of Environmental Protection). (2013). West Virginia water
   resources management plan. (Article 22-26). Charleston, WV.
   http://www.dep.wv.gov/WWE/wateruse/WVWaterPlan/Documents/WVWRMP.pdf
West Virginia PEP (West Virginia Department of Environmental Protection). (2014). Personal
   communication: email from Jason Harmon, West Virginia DEP to Megan Fleming, U.S. EPA with attachment
   of WV DEP fracturing water database. Available online
Whitehead. PG: Wade. Al: Butterfield. D. (2009). Potential impacts  of climate change on water quality  and
   ecology in six UK rivers. 40: 113-122. http://dx.doi.org/10.2166/nh.2009.078

Wyoming State Engineer's Office. (2014). Groundwater control areas and advisory boards. Available online at
   http://seo.wvo.gov/ground-water/groundwater-control-areas-advisorv-boards
               This document is a draft for review purposes only and does not constitute Agency policy.

June 2015                                       4-60                    DRAFT—DO NOT CITE OR QUOTE

-------
Hydraulic Fracturing Drinking Water Assessment
Yang. Y: Robart. Cl: Ruegamer. M. (2013). Analysis of U.S. Hydraulic Fracturing Design Trends. SPE Hydraulic
   Fracturing Technology Conference, The Woodlands, Texas, USA.

Ziemkiewicz. P: Donovan. I: Hause. I: Gutta. B: Fillhart. I: Mack. B: O'Neal. M. (2013). Water quality literature
   review and field monitoring of active shale gas wells: Phase II for Assessing Environmental Impacts of
   Horizontal Gas Well Drilling Operations. Charleston, WV: West Virginia Department of Environmental
   Protection.
               This document is a draft for review purposes only and does not constitute Agency policy.

June 2015                                        4-61                    DRAFT—DO NOT CITE OR QUOTE

-------
Hydraulic Fracturing Drinking Water Assessment
                            Chapter 5
Chemical Mixing
           This document is a draft for review purposes only and does not constitute Agency policy.
June 2015                                              DRAFT—DO NOT CITE OR QUOTE

-------
      Hydraulic Fracturing Drinking Water Assessment
Chapter 5 - Chemical Mixing
      5.  Chemical Mixing
      5.1.  Introduction
 1    This chapter addresses the potential for on-site spills of chemicals used in the chemical mixing
 2    process to affect the quality of drinking water resources. Chemical mixing is a complex process that
 3    requires the use of specialized equipment and a range of different additives to produce the
 4    hydraulic fracturing fluid that is injected into the well. The number, type, and volume of chemicals
 5    used vary from well to well based on site- and company-specific factors. Spills may occur at any
 6    point in the hydraulic fracturing process. Chemicals may spill from on-site storage and containment
 7    units; from interconnected hoses and pipes used to transfer chemicals to and from mixing and
 8    pumping units, and tanker trucks; and from the equipment used to mix and pressurize chemical
 9    mixtures that are pumped down the well. The potential for a spill to affect the quality of a drinking
10    water resource is governed by three overarching factors:  (1) fluid characteristics (e.g., chemical
11    composition and volume), (2) chemical management and spill characteristics, and (3) chemical fate
12    and transport (see Figure 5-1). This chapter is organized around the three factors.
                                           POTENTIAL
                                             IMPACT

      Figure 5-1. Factors governing potential impact to drinking water resources.
               Factors include (1) fluid characteristics (e.g., chemical composition and volume), (2) chemical
               management and spill characteristics, and (3) chemical fate and transport.
                   This document is a draft for review purposes only and does not constitute Agency policy.
      June 2015                                    5-1                  DRAFT—DO NOT CITE OR QUOTE

-------
      Hydraulic Fracturing Drinking Water Assessment                                   Chapter 5 - Chemical Mixing


 1    Section 5.2 provides an introductory overview of the chemical mixing process. The number and
 2    volume of chemicals used and stored on-site are affected by such variables as the type, size, and
 3    goals of the operation; formation characteristics; depth of the well; the length of the horizontal leg;
 4    and the number of fracturing phases and stages.

 5    Section 5.3 describes the different components of the hydraulic fracturing fluid, generally
 6    comprised of the base fluid, proppant, and additives, which may be either individual chemicals or
 7    mixtures. The composition of the hydraulic fracturing fluid is engineered to meet specific criteria.
 8    The total amount and types of additives vary according to the characteristics of the well, site
 9    geology, economics, availability, and the production goals [e.g.. Maule etal.. 2013]. Section 5.4
10    presents the wide range of different chemicals used and their classes, the most frequently used
11    chemicals nationwide and from state-to-state, and volumes used.1 Appendix A provides a list of
12    chemicals that the EPA identified as being used in hydraulic fracturing fluids based on eight
13    sources.

14    Sections 5.5 to 5.7 discuss how chemicals are managed on-site, how spills may occur, and the
15    different approaches for addressing spills. Section 5.5 describes how the potential impact of a spill
16    on drinking water resources depends upon chemical management practices,  such as storage, on-
17    site transfer, and equipment maintenance. Section 5.6 discusses spill prevention, containment, and
18    mitigation. A summary analysis of reported spills and their common causes at hydraulic fracturing
19    sites is presented in Section 5.7.

20    Section 5.8 discusses the fate and transport of spilled chemicals. Spilled chemicals may react and
21    transform into other chemicals, travel from the site of release to a nearby surface water, or leach
22    into the soils and reach ground water. Chemical fate and transport after a release depend on site
23    conditions, environmental conditions, physicochemical properties of the released chemicals, and
24    the volume of the release.

25    Section 5.9 provides an overview of on-going changes in chemical use in hydraulic fracturing, with
26    an emphasis on efforts by industry to reduce potential impacts from surface spills by using fewer
27    and safer chemicals. A synthesis and a discussion of limitations are presented in Section 5.10.

28    Factors affecting the frequency and severity of impacts to drinking water resources from surface
29    spills include size and type of operation, employee training and experience, standard operating
30    procedures, quality and maintenance of equipment, type and volume  of chemical spilled,
31    environmental conditions, proximity to drinking water resources, spill prevention practices, and
32    spill mitigation measures. Due to the limitations of available  data and the scope of this assessment,
33    it is not possible to provide a detailed analysis of all of the factors listed above. Data limitations also
34    preclude a quantitative analysis of the likelihood or magnitude of chemical spills or impacts. Spills
35    that occur off-site, such as those during transportation of chemicals or storage of chemicals in
36    staging areas, are out of scope. This chapter qualitatively characterizes the potential for impacts to
      1 Chemical classes are groupings of different chemicals based on similar features, such as chemical structure, use, or
      physical properties. Examples of chemical classes include hydrocarbons, pesticides, acids, and bases.
                    This document is a draft for review purposes only and does not constitute Agency policy.
      June 2015                                     5-2                    DRAFT—DO NOT CITE OR QUOTE

-------
      Hydraulic Fracturing Drinking Water Assessment
Chapter 5 - Chemical Mixing
 1    drinking water resources given the current understanding of overall operations and specific
 2    components of the chemical mixing process.

      5.2. Chemical Mixing Process
 3    An understanding of the chemical mixing process is necessary to understand how, why, and when
 4    spills that may affect drinking water resources might occur. This description provides a general
 5    overview of chemical mixing in the context of the overall hydraulic fracturing process [Carter etal..
 6    2013: Knappe and Fireline. 2012: Spellman. 2012: Arthur etal.. 2008].

 7    Figure 5-2 shows a hydraulic fracturing site during the chemical mixing process. The discussion
 8    focuses on the types of additives used at each phase of the process. While similar processes are
 9    used to fracture horizontal and vertical wells, a horizontal well treatment is described here because
10    it is likely to be more complex and because horizontal hydraulic fracturing has become more
11    prevalent over time with advances in hydraulic fracturing technology. A water-based system is
12    described because water is the most commonly used base fluid, appearing in more than 93% of
13    FracFocus disclosures between January 1, 2011 and February 28, 2012 [U.S. EPA, 2015a].
      Figure 5-2. Hydraulic fracturing site showing equipment used on-site during the chemical
                mixing process.
                Source: Industry source.
                   This document is a draft for review purposes only and does not constitute Agency policy.
      June 2015                                    5-3                   DRAFT—DO NOT CITE OR QUOTE

-------
      Hydraulic Fracturing Drinking Water Assessment
Chapter 5 - Chemical Mixing
 1    While the number and types of additives may widely vary, the basic chemical mixing process is
 2    similar across sites. The on-site layout of hydraulic fracturing equipment is also similar from site to
 3    site [BT Services Company. 2009). Equipment used in the chemical mixing process typically consists
 4    of chemical storage trucks, water supply tanks, proppant supply, slurry blenders, a number of high-
 5    pressure pumps, a manifold, surface lines and hoses, and a central control unit Detailed
 6    descriptions of specific additives and the equipment used in the process are provided in Sections
 7    5.3 and 5.5, respectively.

 8    The chemical mixing process begins after the drilling, casing, and cementing processes are finished
 9    and hydraulic fracturing equipment has been set up and connected to the well. The process can
10    generally be broken down into sequential phases with specific chemicals added at each phase to
11    achieve a specific purpose [Knappe and Fireline, 2012: Fink, 2003]. Phases may overlap. The
12    process for water-based hydraulic fracturing is outlined in Figure 5-3 below.
         Chemical Mixing Phases
Pre-Pad: Acid
and/or chemicals
may be injected
into clean wellbore

Steps repeated for each stage:
.^ Perforation: Initiate fractures with physical impact
_/ Pad: Inject base fluid and additives into the formation to grow fractures
Proppant: Inject base fluid, additives and proppant to prop open fractures
Flush: Fluid may be injected to facilitate hydrocarbon flow
//
' //
/ /
Hydraulic Fracturing Stages //
II






i

& \> V V
Chemicals & 	 K. Chemicals & •. 	 K
Proppant Injected L
Proppant
gradient

	 ^
Proppant Injected ' 	 /
Proppant
gradient

Chemicals & 	 K
Proppant Injected ' J
Proppant
gradient
-^^'n
ChemicalsS
I
n>
o
OJ
ff
00
1
lit
Proppant Injected
Proppant
gradient


_
'
,,
i= Horizontal Leg
      Figure 5-3. Overview of a chemical mixing process of the hydraulic fracturing water cycle.

                This figure outlines the chemical mixing process for a generic water-based hydraulic fracture of a
                horizontal well. The chemical mixing phases outline the steps in the overall fracturing job, while the
                hydraulic fracturing stages outline how each section of the horizontal well would be fractured
                beginning with the toe of the well, shown on left-side. The proppant gradient represents how the
                proppant size may change within each stage of fracturing as the fractures are elongated. The chemical
                mixing process is repeated depending on the number of stages used for a particular well. The number
                of stages is determined in part by the length of the horizontal leg. In this figure, four stages are
                represented, but typically, a horizontal fracturing treatment would consist of 10 to 20 stages per well
                (Lowe et al., 2013). Fracturing has been reported to be done in as many as 59 stages (Pearson et al.,
                2013).
                    This document is a draft for review purposes only and does not constitute Agency policy.

      June 2015                                      5-4                    DRAFT—DO NOT CITE OR QUOTE

-------
      Hydraulic Fracturing Drinking Water Assessment                                   Chapter 5 - Chemical Mixing


 1    The first phase of the process consists of the cleaning and preparation of the well. The fluid used in
 2    this phase is often referred to as the pre-pad fluid or pre-pad volume. Acid is typically the first
 3    chemical introduced. Acid, with a concentration of 3%-28% (typically hydrochloric acid, HC1), is
 4    used to adjust pH, clean any cement left inside the well from cementing the casing, and dissolve any
 5    pieces of rock that may remain in the well and could block the perforations. Acid is typically
 6    pumped directly from acid storage tanks or tanker trucks, without being mixed with other
 7    additives. The first, or pre-pad, phase  may also involve mixing and injection of additional chemicals
 8    to facilitate the flow of fracturing fluid introduced in the next phase of the process. These additives
 9    may include biocides, corrosion inhibitors, friction reducers, and scale inhibitors [Carter etal.,
10    2013: King. 2012: Knappe andFireline. 2012: Spellman.  2012: Arthur etal.. 20081.

11    In the second phase, a hydraulic fracturing fluid, typically referred to as the pad or pad volume, is
12    mixed, blended, and pumped down the wellbore to create fractures in the formation. The pad is a
13    mixture of base fluid, typically water,  and additives. The pad is designed to create, elongate, and
14    enlarge fractures along the natural channels of the formation when injected under high pressure
15    [Gupta and Valko. 2007]. A typical pad consists of, at minimum, a mixture of water and friction
16    reducer. The operator may also add other additives  (see U.S. EPA (2015a] and Table 5-1) used to
17    facilitate flow and kill bacteria (Carter etal.. 2013: King.  2012: Knappe andFireline. 2012:
18    Spellman,  2012: Arthur etal., 2008]. The pad is pumped into the formation through perforations in
19    the well casing (see Text Box 5-1].
20
21
22
23
24
Text Box 5-1. Perforation.

Prior to the injection of the pad, the well casing is typically perforated to provide openings through which the
pad fluid can enter the formation. A perforating gun is typically used to create small holes in the section of the
wellbore being fractured. The perforating gun is lowered into position in the horizontal portion of the well.
An electrical current is used to set off small explosive charges in the gun, which creates holes through the well
casing and out a short, controlled distance into the formation [Gupta and Valko, 20071.
25    In the third phase, proppant, typically sand, is mixed into the hydraulic fracturing fluid. The
26    proppant volume, as a proportion of the injected fluid, is increased gradually until the desired
27    concentration in the fractures is achieved. Gelling agents, if used, are also mixed in with the
28    proppant and base fluid in this phase to increase the viscosity and carry the proppant. Additional
29    chemicals may be added to gelled fluids, initially to maintain viscosity and later to break the gel
30    down into a more readily removable fluid. (Carter etal.. 2013: King. 2012: Knappe and Fireline.
31    2012: Spellman. 2012: Arthur etal.. 20081.

32    A final flush or clean-up phase may be conducted after the stage is fractured, with the primary
3 3    purpose of maximizing well productivity. The flush is a mixture of water and chemicals that work to
34    aid the placement of the proppant, clean out the chemicals injected in previous phases, and prevent
35    microbial growth in the fractures (Knappe and Fireline. 2012: Fink. 2003].

36    The second, third, and fourth phases are repeated multiple times in a horizontal well, as the
37    horizontal section, or leg, of the wellbore is typically fractured in multiple segments referred to as

                    This document is a draft for review purposes only and does not constitute Agency policy.
      June 2015                                     5-5                   DRAFT—DO NOT CITE OR QUOTE

-------
      Hydraulic Fracturing Drinking Water Assessment                                  Chapter 5 - Chemical Mixing


 1    stages. For each stage, the well is typically perforated and fractured beginning at the end, or toe, of
 2    the wellbore and proceeding backwards toward the vertical section. Each fractured stage is isolated
 3    before the next stage is fractured. The number of stages corresponds directly to the number of
 4    times the chemical mixing process is repeated at the site surface (see Figure 5-3). The number of
 5    stages depends upon the length of the leg [Carter etal.. 2013: King. 2012: Knappe and Fireline,
 6    2012: Spellman. 2012: Arthur etal.. 20081.

 7    The number of stages per well can vary, with several sources suggesting between 10 and 20 is
 8    typical [GNB, 2015: Lowe etal., 2013].' The full range reported in the literature is much wider, with
 9    one source documenting between 1 and 59 stages per well [Pearson et al., 2013] and others
10    reporting values within this range [NETL. 2013: STO. 2013: Allison etal.. 2009]. It also appears that
11    the number of stages per well has increased over time. For instance, in the Williston Basin the
12    average number of stages per horizontal well rose from approximately 10 in 2008 to 30 in 2012
13    [Pearsonetal.. 2013].

14    In each of these phases, water is the primary component of the hydraulic fracturing fluid, though
15    the exact composition of the fluid injected into the well changes over the duration of each stage. In
16    water-based hydraulic fracturing, water typically comprises between 90% and 94% of the
17    hydraulic fracturing fluid, proppant comprises 5% to 9%, and additives comprise the remainder,
18    typically 2% or less [Carter etal.. 2013: Knappe and Fireline. 2012: SWN. 2011]. The exception to
19    this typical fluid composition may be when a concentrated acid is used in the initial cleaning phase
20    of the fracturing process.

      5.3. Overview of Hydraulic Fracturing Fluids
21    Hydraulic fracturing fluids are formulated to perform specific functions: create and extend the
22    fracture, transport proppant, and place the proppant in the fractures [Montgomery. 2013:
23    Spellman, 2012: Gupta and Valko, 2007]. The hydraulic fracturing fluid generally consists of three
24    parts: [1] the base fluid, which is the largest constituent by volume, [2] the additives, which can be
25    a single chemical or a mixture of chemicals, and [3] the proppant Additives are chosen to serve a
26    specific purpose in the hydraulic fracturing fluid (e.g., friction reducer, gelling agent, crosslinker,
27    biocide] [Spellman. 2012]. Throughout this chapter, "chemical" is used to refer to individual
28    chemical compounds (e.g., methanol]. Proppants are small particles, usually sand, mixed with
29    fracturing fluid to hold fractures open so that the target hydrocarbons can flow from the formation
30    through the fractures and up the wellbore. The combination of chemicals, and the mixing and
31    injection process, varies based on a number  of factors as discussed below. The chemical
32    combination determines the amount and what type of equipment is required for storage and,
33    therefore, contributes to the determination of the potential for spills and impacts of those spills.

34    The particular composition of hydraulic fracturing fluids is selected by a design engineer based on
35    empirical experience, the formation, economics, goals of the fracturing process, availability of the
      1 The number of stages has been reported to be 6 to 9 in the Huron in 2009 [Allison etal.. 2009). 25 and up in the
      Marcellus [NETL. 2013). and up to 40 by STO (2013).
                   This document is a draft for review purposes only and does not constitute Agency policy.
      June 2015                                    5-6                   DRAFT—DO NOT CITE OR QUOTE

-------
      Hydraulic Fracturing Drinking Water Assessment                                     Chapter 5 - Chemical Mixing


 1    desired chemicals, and preference of the service company or operator [Montgomery, 2013: ALL
 2    Consulting. 2012: Klein etal.. 2012: Ely. 1989]. No single set of specific chemicals is used at every
 3    site. Multiple types of fracturing fluids may be appropriate for a given site and any given type of
 4    fluid may be appropriate at multiple sites. For the same type of fluid formulation, there can be
 5    differences in the additives, chemicals, and concentrations selected. There are broad criteria for
 6    hydraulic fracturing fluid selection based on the fracturing temperatures, formation permeability,
 7    fracturing pressures, and formation water sensitivity, as shown in Figure 5-4 [Gupta and Valko,
 8    2007: Elbel and Britt. 2000). One of the most important properties in designing a hydraulic
 9    fracturing fluid is the viscosity [Montgomery, 2013].x

10    Figure 5-4 provides a general overview of which fluids can be used in different situations. As an
11    example, crosslinked fluids with 25% nitrogen foam (titanate or zirconate crosslink + 25% N2)  can
12    be used in both gas and oil wells with high temperatures with variation in water  sensitivity.
      1 Viscosity is a measure of the internal friction of fluid that provides resistance to shear within the fluid, informally
      referred to as how "thick" a fluid is. For example, custard is thick and has a high viscosity, while water is runny with a low
      viscosity. Sufficient viscosity is needed to create a fracture and transport proppant [Gupta and Valko. 2007). In lower-
      viscosity fluids, proppant is transported by turbulent flow and requires more hydraulic fracturing fluid. Higher-viscosity
      fluids allows the fluid to carry more proppant, requiring less fluid but necessitating the reduction of viscosity after the
      proppant is placed [Rickman etal.. 2008: Gupta and Valko. 2007).
                     This document is a draft for review purposes only and does not constitute Agency policy.
      June 2015                                       5-7                    DRAFT—DO NOT CITE OR QUOTE

-------
Hydraulic Fracturing Drinking Water Assessment
                                                                                 Chapter 5 - Chemical Mixing
                                 Gas Well
                                                                     Oil Well
u
N
"

Less
|
Low pressure
or
No - water sensitiv
'

ss - 150°F -(More
7
c
Surface-delayed
borate c
corsslinkedguar +
or HPG
~I
., J kfW > 1000 md-ft + x,

Linear fluids





175°F


Low pressure
e Yes
or
No - water sensitive ~ Yes
~L
Foamed
fluids
0% - 75%
uality or
lowpH
rosslinked
25% CO2
Delays
titanate
crossli

Less -300°F1 More Less -| 250°F \ More
"L ~T
i \ i
LowpH
crosslinked Titanate
+ or
25% CO2 zirconate
Super crosslinked
foam +
id borate, Zirconate borate 25% Nj
or zirconate crosslinked + N
nkedguar HPG



> 300 ft .

-Yes

'
Low
No - Pressure
{See gas
well guide

Low
No PressL
1
Less -200°F- More
]
.
polymer cr<
emulsion



Very water sensitive _ ^
'
Low
Yes No • Pressu
T

lore


re Yes
T
P
Gelled oil Gelled oil + N2
re - Yes
A—,
Less -200°F- More


rconate 200°F
clinked Less~ -Mor
HPG
-] f-
Low pH Titanat
crosslinked cr
25% CO2

e
6 	 1
e or zirconate
osslinked
25% N2

Figure 5-4. Example fracturing fluid decision tree for gas and oil wells.
           Adapted from Elbel and Britt (2000).
June 2015
This document is a draft for review purposes only and does not constitute Agency policy.
                                 5-8
DRAFT—DO NOT CITE OR QUOTE

-------
     Hydraulic Fracturing Drinking Water Assessment
1    Table 5-1 provides a list of common types of additives, their functions, and the most frequently
2    used chemicals for each purpose based on the EPA's analysis of disclosures to FracFocus 1.0
3    (hereafter EPA FracFocus report; U.S. EPA[2015a]]. the EPA's project database of disclosures to
4    FracFocus 1.0 [hereafter EPA FracFocus database; U.S. EPA[2015b]], and other literature sources.
5    Additional information on more additives can be found in U.S. EPA f2015al.
     Table 5-1. Examples of common additives, their function, and the most frequently used
               chemicals reported to FracFocus for these additives.

               The list of examples of common additives was developed from information provided in multiple
               sources (U.S. EPA, 2015a, b; Stringfellow et al., 2014; Montgomery, 2013; Vidicetal., 2013; Spellman,
               2012; GWPC and ALL Consulting, 2009; Arthur et al., 2008; Gupta and Valko, 2007; Gidleyet al., 1989).
               The additive functions are based on information the EPA received from service companies (U.S. EPA,
               2013a).
Additives
Acid
Biocide
Breaker
Clay control
Corrosion
inhibitor
Crosslinker
Emulsifier
Foaming agent
Function
Dissolves cement, minerals, and clays to
reduce clogging of the pore space
Controls or eliminates bacteria, which
can be present in the base fluid and may
have detrimental effects on the
fracturing process
Reduces the viscosity of specialized
treatment fluids such as gels and foams
Prevents the swelling and migration of
formation clays in reaction to water-
based fluids
Protects the iron and steel components
in the wellbore and treating equipment
from corrosive fluids
Increases the viscosity of base gel fluids
by connecting polymer molecules
Facilitates the dispersion of one
immiscible fluid into another by reducing
the interfacial tension between the two
liquids to achieve stability
Generates and stabilizes foam fracturing
fluids
Chemicals reported in >20% of
FracFocus disclosures for additive3'15
Hydrochloric acid
Glutaraldehyde;
2,2-dibromo-3-nitrilopropionamide
Peroxydisulfuric acid diammonium salt
Choline chloride
Methanol; propargyl alcohol; isopropanol
Ethylene glycol; potassium hydroxide;
sodium hydroxide
2-Butoxyethanol;
polyoxyethylene(10)nonylphenyl ether;
methanol; nonyl phenol ethoxylate
2-Butoxyethanol; Nitrogen, liquid;
isopropanol; methanol; ethanol
                   This document is a draft for review purposes only and does not constitute Agency policy.

     June 2015                                      5-9                    DRAFT—DO NOT CITE OR QUOTE

-------
Hydraulic Fracturing Drinking Water Assessment
 Additives
Function
Chemicals reported in >20% of
FracFocus disclosures for additive3'15
 Friction reducer
Reduces the friction pressures
experienced when pumping fluids
through tools and tubulars in the
wellbore
Hydrotreated light petroleum distillates
 Gelling agent
Increases fracturing fluid viscosity
allowing the fluid to carry more proppant
into the fractures and to reduce fluid loss
to the reservoir
Guargum; hydrotreated light petroleum
distillates
 Iron control
 agent
Controls the precipitation of iron from
solution
Citric acid
 Nonemulsifier
Separates problematic emulsions
generated within the formation
Methanol; isopropanol; nonyl phenol
ethoxylate
 pH control
Affects the pH of a solution by either
inducing a change (pH adjuster) or
stabilizing and resisting change (buffer)
to achieve desired qualities and optimize
performance
Carbonic acid, dipotassium salt; potassium
hydroxide; sodium hydroxide; acetic acid
 Resin curing
 agents
Lowers the curable resin coated
proppant activation temperature when
bottom hole temperatures are too low to
thermally activate bonding
Methanol; nonyl phenol ethoxylate;
isopropanol; alcohols, C12-14-secondary,
ethoxylated
 Scale inhibitor
Controls or prevents scale deposition in
the production conduit or completion
system
Ethylene glycol; methanol
 Solvent
Controls the wettability of contact
surfaces or prevents or breaks emulsions
Hydrochloric acid
 a Chemicals (excluding water and quartz) listed as reported to FracFocus in more than 20% of disclosures for a given purpose
 when that purpose was listed as used on a disclosure. These are not necessarily the active ingredients for the purpose, but
 rather are listed as being commonly present for the given purpose. Chemicals may be disclosed for more than a single
 purpose (e.g., 2-butoxyethanol is listed as being used as an emulsifier and a foaming agent).
  Analysis considered 32,885 disclosures and 615,436 ingredient records that met selected quality assurance criteria,
 including: completely parsed; unique combination of fracture date and API well number; fracture date between January 1,
 2011, and February 28, 2013; valid CASRN; valid concentrations; and valid purpose. Disclosures that did not meet quality
 assurance criteria (5,645) or other, query-specific criteria were excluded from analysis.
               This document is a draft for review purposes only and does not constitute Agency policy.
June 2015                                         5-10                     DRAFT—DO NOT CITE OR QUOTE

-------
      Hydraulic Fracturing Drinking Water Assessment
 1    A general description of typical hydraulic fracturing fluid formulations nationwide is difficult
 2    because fracturing fluids vary from well to well. Based on the FracFocus report, the median number
 3    of chemicals reported for each disclosure was 14, with the 5th to 95th percentile ranging from four to
 4    28. The median number of chemicals per disclosure was 16 for oil wells and 12 for gas wells [U.S.
 5    EPA, 2015b]. Other sources have stated that between three and 12 additives and chemicals are
 6    used [Schlumberger. 2015: Carter etal.. 2013: Spellman. 2012: GWPC and ALL Consulting. 2009]. 1

 7    Water, the most commonly used base fluid for hydraulic fracturing, is inferred to be used as a base
 8    fluid in more than 93% of FracFocus disclosures. Alternatives to water-based fluids, such as
 9    hydrocarbons and gases, including carbon dioxide or nitrogen-based foam, may also be used based
10    on formation characteristics, cost, or preferences of the well operator or service company [ALL
11    Consulting, 2012: GWPC and ALL Consulting, 2009]. Non-aqueous base fluid ingredients were
12    identified in 761 [2.2%] of FracFocus 1.0 disclosures fU.S. EPA. 2015a). Gases and hydrocarbons
13    may be used alone or blended with water; more than 96% of the disclosures identifying non-
14    aqueous base fluids are blended [U.S. EPA, 2015a]. There is no standard method to categorize the
15    different fluid formulations [Patel etal., 2014: Montgomery, 2013: Spellman, 2012: Gupta and
16    Valko, 2007]. Therefore, we broadly categorize the fluids as water-based or alternative fluids.

      5.3.1.  Water-Based Fracturing Fluids
17    The advantages of water-based fracturing fluids are low cost, ease of mixing, and ability to recover
18    and recycle the water. The disadvantages are low viscosity, the narrowness of the fractures created,
19    and they may not provide optimal performance in water-sensitive formations (see Section 5.3.2]
20    [Montgomery. 2013: Gupta and Valko. 2007]. Water-based fluids can be as simple as water with a
21    few additives to reduce friction, such as "slickwater," or as complex as water with crosslinked
22    polymers, clay control agents, biocides, and scale inhibitors [Spellman, 2012].

2 3    Gels may be added to water-based fluids to increase viscosity, which assists with proppant
24    transport and results in wider fractures. Gelling agents include natural polymers, such as guar,
25    starches, and cellulose derivatives, which requires the addition of biocide to minimize bacterial
26    growth [Spellman. 2012: Gupta and Valko. 2007]. Gels may be linear or crosslinked. Crosslinking
27    increases viscosity without adding more gel. Gelled fluids require the addition of a breaker, which
28    breaks down the gel after it carries in the proppant, to reduce fluid viscosity to facilitate fluid
29    flowing back after treatment [Spellman, 2012: Gupta and Valko, 2007]. The presence of residual
30    breakers may make it difficult to reuse recovered water [Montgomery. 2013].

      5.3.2.  Alternative Fracturing Fluids

31    Alternative hydraulic fracturing fluids can be used for water-sensitive formations (i.e., formations
32    where permeability is reduced when water is added] or as dictated by production goals
33    (Halliburton. 1988]. Examples of alternative fracturing fluids include acid-based fluids; non-
34    aqueous-based fluids; energized fluids, foams or emulsions; viscoelastic surfactant fluids; gels;
      1 Sources may differ based on whether they are referring to additives or chemicals.


                   This document is a draft for review purposes only and does not constitute Agency policy.
      June 2015                                    5-11                   DRAFT—DO NOT CITE OR QUOTE

-------
      Hydraulic Fracturing Drinking Water Assessment
 1    methanol; and other unconventional fluids [Montgomery, 2013: Sabaetal., 2012: Gupta and Hlidek,
 2    2009: Gupta and Valko. 2007: Halliburton. 19881.

 3    Acid fracturing removes the need for a proppant and is generally used in carbonate formations.
 4    Fractures are initiated with a viscous fracturing fluid, and the acid (gelled, foamed, or emulsified) is
 5    added to irregularly etch the wall of the fracture and prop open the formation for a higher
 6    conductivity fracture [Spellman, 2012: Gupta and Valko, 2007].

 7    Non-aqueous fluids are used in water-sensitive formations. Non-aqueous fluids may also contain
 8    additives, such as gelling agents, to improve performance [Gupta and Valko. 2007]. The use of non-
 9    aqueous fluids has decreased due to safety concerns, and because water-based and emulsion fluid
10    technologies have improved [Montgomery. 2013: Gupta and Valko. 2007]. Methanol, for example,
11    was previously used as a base fluid in water-sensitive reservoirs beginning in the early 1990s, but
12    was discontinued in 2001 for safety concerns and cost [Sabaetal., 2012: Gupta and Hlidek, 2009:
13    Gupta and Valko, 2007]. Methanol is still used as an additive or in additive mixtures in  hydraulic
14    fracturing fluid formulations.

15    Energized fluids, foams, and emulsions minimize fluid leakoff, have high proppant-carrying
16    capacity, improve fluid recovery, and are sometimes used in water-sensitive formations [Barati and
17    Liang. 2014: Gu andMohanty. 2014: Spellman. 2012: Gupta and Valko. 2007: Martin and Valko.
18    2007].* However, these treatments tend to be expensive, require high pressure, and pose potential
19    health and safety concerns fMontgomery. 2013: Spellman. 2012: Gupta and Valko. 20071
20    Energized fluids are mixtures of liquid and gas [Patel etal., 2014: Montgomery, 2013]. Nitrogen
21    (N2] or carbon dioxide (C02], the gases used, make up less than 53% of the fracturing fluid volume,
22    typically ranging from 25% to 30% by volume [Montgomery. 2013: Gupta and Valko. 2007:
23    Mitchell, 1970]. Energized foams are liquid-gas mixtures, with N2 or C02 gas comprising more than
24    53% of the fracturing fluid volume, with a typical range of 70% to 80% by volume [Mitchell, 1970].
25    Emulsions are liquid-liquid mixtures, typically a hydrocarbon (e.g., condensate or diesel] with
26    water, with the hydrocarbon typically 70% to 80% by volume.2 Both water-based fluids, including
27    gels, and non-aqueous fluids can be energized fluids or foams.

28    Foams and emulsions break easily using gravity separation and are stabilized by using additives
29    such as foaming agents (Gupta and Valko. 2007]. Emulsions may be used to stabilize active chemical
30    ingredients or to delay chemical reactions, such as the use of carbon dioxide-miscible, non-aqueous
31    fracturing fluids to reduce fluid leakoff in water-sensitive formations (Taylor et al.. 2006].

3 2    Other types of fluids not addressed above include viscoelastic surfactant fluids, viscoelastic
3 3    surfactant foams, crosslinked foams, liquid carbon dioxide-based fluid, and liquid carbon dioxide-
34    based foam fluid, and hybrids of other fluids fKing. 2010: Brannon etal.. 2009: Curtice  etal.. 2009:
      1 Leakoff is the fraction of the injected fluid that infiltrates into the formation (e.g., through an existing natural fissure) and
      is not recovered during production [Economides et al.. 2007). See Chapter 6, Section 6.3 for more discussion on leakoff.
      2 Diesel is a mixture typically of C8 to C21 hydrocarbons.
                   This document is a draft for review purposes only and does not constitute Agency policy.
      June 2015                                    5-12                  DRAFT—DO NOT CITE OR QUOTE

-------
      Hydraulic Fracturing Drinking Water Assessment
 1    Tudoretal..20Q9: Gupta and Valko. 2007: Coulter etal.. 2006: Boyer etal.. 2005: Freddetal.. 2004:
 2    MacDonaldetal.. 20031.

 3    Alternative fluids have been developed to work in tight formations, shales, and coalbeds, where
 4    production is based on desorption of the natural gas, or in formations where the fracturing fluid
 5    must displace a fluid that is already in place.

      5.3.3.  Proppants
 6    Proppants are small particles carried down the well and into fractures by fracturing fluid. They hold
 7    the fractures open after hydraulic fracturing fluid has been removed [Brannon and Pearson, 2007].
 8    The propped fractures provide a path for the hydrocarbon to flow from the reservoir. Sand is most
 9    commonly used, but other proppants include man-made or specially engineered particles, such as
10    resin-coated sand, high-strength ceramic materials, or sintered bauxite [Schlumberger, 2014:
11    Brannon and Pearson, 2007]. Proppant types can be used individually or in combinations.

      5.4. Frequency and  Volume of Hydraulic Fracturing Chemical Use
12    This section highlights the different chemicals used in hydraulic fracturing and discusses the
13    frequency and volume of use. Based on the U.S. EPA analysis of the FracFocus 1.0 database (see Text
14    Box 5-2], we focus our analysis on individual chemicals rather than mixtures of chemicals used as
15    additives. Chemicals are reported to FracFocus by using the chemical name and the Chemical
16    Abstract Services Registration Number (CASRN], which is  a unique number identifier for every
17    chemical substance.1 The information on specific chemicals, particularly those most commonly
18    used, can be used to assess potential impacts to drinking water resources. The volume of chemicals
19    stored on-site provides information on the potential volume of a chemical spill.
      1 A CASRN and chemical name combination identify a chemical substance, which can be a single chemical (e.g.,
      hydrochloric acid, CASRN 7647-01-0) or a mixture of chemicals (e.g., hydrotreated light petroleum distillates (CASRN
      64742-47-8), a complex mixtures of C9to C16 hydrocarbons). For simplicity, we refer to both pure chemicals and
      chemical substances that are mixtures, which have a single CASRN, as "chemicals."
                   This document is a draft for review purposes only and does not constitute Agency policy.
      June 2015                                    5-13                  DRAFT—DO NOT CITE OR QUOTE

-------
      Hydraulic Fracturing Drinking Water Assessment
 1
 2
 3
 4
 5
 6
 7
 8
 9
10
11
12
13
14
15
16
17
18
19
20
Text Box 5-2. The FracFocus Registry and EPA FracFocus Report.

The Ground Water Protection Council (GWPC) and the Interstate Oil and Gas Compact Commission (IOGCC)
developed a national hydraulic fracturing chemical registry, FracFocus [www.fracfocus.org]. Well operators
can use the registry to disclose information about chemicals they use during hydraulic fracturing. The EPA
accessed data from FracFocus 1.0 from January 1, 2011 to February 28, 2013, which included more than
39,000 disclosures from 20 states that had been submitted by operators prior to March 1,2013.
Submission to FracFocus was initially voluntary and varied from state to state. During the timeframe of the
EPA's study, six of the 20 states with data in the project database began requiring operators to disclose
chemicals used in hydraulic fracturing fluids to FracFocus (Colorado, North Dakota, Oklahoma, Pennsylvania,
Texas, and Utah). Three other states started requiring disclosure to either FracFocus or the state (Louisiana,
Montana, and Ohio), and five states required or began requiring disclosure to the state (Arkansas, Michigan,
New Mexico, West Virginia, and Wyoming). Alabama, Alaska, California, Kansas, Mississippi, and Virginia did
not have reporting requirements during the period of the EPA's study.
Disclosures from the five states reporting the most disclosures to FracFocus (Texas, Colorado, Pennsylvania,
North Dakota, and Oklahoma) comprise over 78% of the  disclosures in the database; nearly half (47%) of the
disclosures are from Texas. Thus, data from these states are most heavily represented in the EPA's analyses.
The EPA's analysis may or may not be nationally representative.
The EPA summarized information on the locations of the wells in the disclosures, water volumes used, and
the frequency of use and concentrations (% by mass, reported as maximum ingredient concentration) of the
chemicals in the additives and the hydraulic fracturing fluid. Additional information can be found in the EPA
FracFocus report (U.S. EPA. 2015a).
21    The EPA compiled a list of 1,076 chemicals known to be have been used in the hydraulic fracturing
22    process (see a full list, methodology, and the source citations in Appendix A). The chemicals used in
23    hydraulic fracturing fall into different chemical classes and include both organic and inorganic
24    chemicals. The chemical classes of commonly used hydraulic fracturing chemicals include but are
25    not limited to:

26         •   Acids (e.g., hydrochloric acid, peroxydisulfuric acid, acetic acid, citric acid).
27         •   Alcohols (e.g., methanol, isopropanol, ethylene glycol, propargyl alcohol, ethanol).
28         •   Aromatic hydrocarbons (e.g., benzene, naphthalene, heavy aromatic petroleum solvent
29             naphtha).
30         •   Bases (e.g., sodium hydroxide, potassium hydroxide).
31         •   Hydrocarbon mixtures (e.g., petroleum distillates).
32         •   Polysaccharides (e.g., guar gum).
33         •   Surfactants (e.g., poly(oxy-l,2-ethanediyl)-nonylphenyl-hydroxy, 2-butoxyethanol).
34         •   Salts (e.g., sodium  chlorite, dipotassium carbonate).
                    This document is a draft for review purposes only and does not constitute Agency policy.
      June 2015                                      5-14                    DRAFT—DO NOT CITE OR QUOTE

-------
      Hydraulic Fracturing Drinking Water Assessment
 2
 3
 4
 5
 6
 7
 8
10
11
12
13
14
15
16

17
18
19
20
21
Text Box 5-3. Confidential Business Information (CBI)

This assessment relies in large part upon information provided to the EPA or to other organizations. The
submitters of that information (e.g., businesses that operate wells or perform services to hydraulically
fracture the well) may view some of the information as confidential business information (CBI), and
accordingly asserted CBI claims to protect such information. Information deemed to be CBI may include
information such as trade secrets or other proprietary business information, entitled to confidential
treatment under Exemption 4 of the Freedom of Information Act (FOIA) and other applicable laws. FOIA and
the EPA's CBI regulations may allow for information claimed as CBI provided to the EPA to be withheld from
the public, including in this document.

The EPA evaluated data from FracFocus 1.0, a national hydraulic fracturing chemical registry used and relied
upon by some states, industry groups and non-governmental organizations. A company submitting a
disclosure to FracFocus  may choose to not report the identity of a chemical it considers CBI. As part of the
EPA's analysis, more than 39,000 FracFocus 1.0 disclosures over the period January 1, 2013 to March 1,2013
were analyzed and more than 70% of disclosures contained at least one chemical designated as CBI. Of the
disclosures containing CBI chemicals, there was an average of five CBI chemicals per disclosure (U.S. EPA.
2015a). The prevalence  of CBI claims in FracFocus 1.0 limits completeness of the data set.

Consistent with the hydraulic fracturing study plan, data were submitted by nine service companies to the
EPA regarding chemicals used in hydraulic fracturing from 2005 to 2009. Because this submission was to the
EPA, the EPA was given the actual names and CASRNs of any chemicals the company considered CBI. This
included a total of 381 CBI chemicals, with a mean of 42 CBI chemicals per company and a range of 7 to 213
(U.S. EPA. 2013a).
      5.4.1.  National Frequency of Use of Hydraulic Fracturing Chemicals
22    The EPA reported that 692 chemicals were reported to FracFocus 1.0 for use in hydraulic fracturing
23    from  January 1, 2011, to February 28, 2013, with a total of 35,957 disclosures [U.S. EPA. 20153].!

24    Table 5-2 presents the 35 chemicals (5% of all chemicals identified in the EPA's study) that were
25    reported in at least 10% of the FracFocus 1.0 disclosures for all states reporting to FracFocus
26    during this time. This table also includes the top four additives that were reported to include the
27    given chemical in FracFocus disclosures from January 1, 2011 to February 28, 2013.
      1 The EPA reported that 692 chemicals were reported to FracFocus 1.0 for use in hydraulic fracturing from January 1,
      2011, to February 28,2013, with a total of 35,957 disclosures. Chemicals may be pure chemicals (e.g., methanol) or
      chemical mixtures (e.g., hydrotreated light petroleum distillates), and they each have a single CASRN. Of these 692
      chemicals, 598 had valid fluid and additive concentrations (34,675 disclosures). Sixteen chemicals were removed because
      they were minerals listed as being used as proppants. This left a total of 582 chemicals (34,344 disclosures).
                     This document is a draft for review purposes only and does not constitute Agency policy.
      June 2015                                       5-15                   DRAFT—DO NOT CITE OR QUOTE

-------
Hydraulic Fracturing Drinking Water Assessment
Table 5-2. Chemicals reported to FracFocus 1.0 from January 1, 2011 to February 28, 2013 in
          10% or more disclosures, with the percent of disclosures for which each chemical is
          reported and the top four reported additives for the chemical.
          For chemicals with fewer than four reported  additives, the table presents all additives (U.S. EPA,
          2015b).
No.
1
2
3
4
5
6
7
8
9
10
11
12
13
14
Chemical name3
Methanol
Hydrotreated light
petroleum distillates
Hydrochloric acid
Water
Isopropanol
Ethylene glycol
Peroxydisulfuric acid,
diammonium salt
Sodium hydroxide
Guar gum
Quartz6
Glutaraldehyde
Propargyl alcohol
Potassium hydroxide
Ethanol
CASRN
67-56-1
64742-47-8
7647-01-0
7732-18-5
67-63-0
107-21-1
7727-54-0
1310-73-2
9000-30-0
14808-60-7
111-30-8
107-19-7
1310-58-3
64-17-5
Percent of
disclosures'3
72%
65%
65%
48%
47%
46%
44%
39%
37%
36%
34%
33%
29%
29%
Chemical used in these additives
(four most common, FracFocus database)0
corrosion inhibitors, surfactants, non-
emulsifiers, scale control
friction reducers, gelling agents and gel
stabilizers, crosslinkers and related additives,
viscosifiers
acids, solvents, scale control, clean
perforations
acids, biocides, clay control, scale control
corrosion inhibitors, non-emulsifiers,
surfactants, biocides
crosslinkers and related additives, scale
control, corrosion inhibitors, friction reducers
breakers and breaker catalysts, oxidizer,
stabilizers, clean perforations
crosslinkers and related additives, biocides,
pH control, scale control
gelling agents and gel stabilizers, viscosifiers,
clean perforations, breakers and breaker
catalysts
breakers and breaker catalysts, gelling agents
and gel stabilizers, scale control, crosslinkers
and related additives
biocides, surfactants, crosslinkers and related
additives, sealers
corrosion inhibitors, inhibitors, acid inhibitors,
base fluid
crosslinkers and related additives, pH control,
friction reducers, gelling agents and gel
stabilizers
surfactants, biocides, corrosion inhibitors,
fluid foaming agents and energizers
              This document is a draft for review purposes only and does not constitute Agency policy.
June 2015                                     5-16                   DRAFT—DO NOT CITE OR QUOTE

-------
Hydraulic Fracturing Drinking Water Assessment
No.
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
Chemical name3
Acetic acid
Citric acid
2-Butoxyethanol
Sodium chloride
Solvent naphtha,
petroleum, heavy arorn/
Naphthalene
2,2-Dibromo-3-
nitrilopropionamide
Phenolic resin
Choline chloride
Methenamine
Carbonic acid,
dipotassium salt
1,2,4-Trimethylbenzene
Quaternary ammonium
compounds, benzyl-C12-
16-alkyldimethyl,
chlorides8
Poly(oxy-l,2-ethanediyl)-
nonylphenyl-hydroxy
(mixture)h
Formic acid
Sodium chlorite
Nonyl phenol ethoxylate
CASRN
64-19-7
77-92-9
111-76-2
7647-14-5
64742-94-5
91-20-3
10222-01-2
9003-35-4
67-48-1
100-97-0
584-08-7
95-63-6
68424-85-1
127087-87-0
64-18-6
7758-19-2
9016-45-9
Percent of
disclosures'3
24%
24%
21%
21%
21%
19%
16%
14%
14%
14%
13%
13%
12%
12%
12%
11%
11%
Chemical used in these additives
(four most common, FracFocus database)0
pH control, iron control agents, acids, gelling
agents and stabilizers
iron control agents, scale control, gelling
agents and gel stabilizers, pH control
surfactants, corrosion inhibitors, non-
emulsifiers, fluid foaming agents and
energizers
breakers/breaker catalysts, friction reducers,
scale control, clay control
surfactants, non-emulsifiers, inhibitors,
corrosion inhibitors
surfactants, non-emulsifiers, corrosion
inhibitors, inhibitors
biocides, clean perforations, breakers and
breaker catalysts, non-emulsifiers
proppants, biocides, clean perforations, base
fluid
clay control, clean perforations, base fluid,
biocides
proppants, crosslinkers and related additives,
biocides, base fluid
pH control, proppants, acids, surfactants
surfactants, non-emulsifiers, corrosion
inhibitors, inhibitors
biocides, non-emulsifiers, corrosion inhibitors,
scale control
surfactants, friction reducers, non-emulsifiers,
inhibitors
corrosion inhibitors, acids, inhibitors, pH
control
breakers/breaker catalysts, biocides, oxidizer,
proppants
non-emulsifiers, resin curing agents,
activators, friction reducers
                This document is a draft for review purposes only and does not constitute Agency policy.
June 2015                                          5-17                     DRAFT—DO NOT CITE OR QUOTE

-------
      Hydraulic Fracturing Drinking Water Assessment
No.
32
33
34
35
Chemical name3
Tetrakis(hydroxymethyl)p
hosphonium sulfate
Polyethylene glycol
Ammonium chloride
Sodium persulfate
CASRN
55566-30-8
25322-68-3
12125-02-9
7775-27-1
Percent of
disclosures'3
11%
11%
10%
10%
Chemical used in these additives
(four most common, FracFocus database)0
biocides, scale control, clay control
biocides, non-emulsifiers, surfactants, clay
control
friction reducers, crosslinkers and related
additives, scale control, clay control
breakers and breaker catalysts, oxidizer, pH
control
       a Chemical refers to chemical substances with a single CASRN, these may be pure chemicals (e.g., methanol) or chemical
       mixtures (e.g., hydrotreated light petroleum distillates).
       b Analysis considered 34,675 disclosures and 676,376 ingredient records that met selected quality assurance criteria,
       including: completely parsed; unique combination of fracture date and API well number; fracture date between January 1,
       2011, and February 28, 2013; valid CASRN; and valid concentrations. Disclosures that did not meet quality assurance criteria
       (3,855) or other, query-specific criteria were excluded from analysis.
       c Analysis considered 32,885 disclosures and 615,436 ingredient records that met selected quality assurance criteria,
       including: completely parsed; unique combination of fracture date and API well number; fracture date between January 1,
       2011, and February 28, 2013; valid CASRN; valid concentrations; and valid purpose. Disclosures that did not meet quality
       assurance criteria (5,645) or other, query-specific criteria were excluded from analysis.
        Hydrotreated light petroleum distillates (CASRN 64742-47-8) is a mixture of hydrocarbons, in the C9 to C16 range.
       e Quartz (CASRN 14808-60-7) the proppant most commonly reported, was also reported as an ingredient in other additives
       U.S. EPA(2015a).
        Heavy aromatic solvent naphtha (petroleum) (CASRN 64742-94-5) is mixture of aromatic hydrocarbons, in the C9 to C16
       range.
       5 Quaternary ammonium compounds, benzyl-C12-16-alkyldimethyl, chlorides (CASRN 68424-85-1) is a mixture of
       benzalkonium chloride with carbon chains between 12 and 16.
        Poly(oxy-l,2-ethanediyl)-nonylphenyl-hydroxy (mixture) (CASRN 127087-87-0) is mixture with varying length ethoxy links.
 1    There is no single chemical used at all wells across the nation. Methanol is the most commonly used
 2    chemical, reported at 72.1% of wells in FracFocus 1.0, and is associated with 33 types of additives,
 3    including corrosion inhibitors, surfactants, non-emulsifiers, and scale control [U.S. EPA. 2015b).
 4    Table 5-2 also shows the variability in different chemicals reported to FracFocus  1.0. The
 5    percentage of disclosures reporting a given chemical suggests the likelihood of that chemical's use
 6    at a site. Only three chemicals (methanol, hydrotreated light petroleum distillates, and hydrochloric
 7    acid) were used at more than half of the sites nationwide, and only 12 were used  at more thanone-
 8    third.

 9    In addition to providing information on frequency of use, FracFocus 1.0 data provides the
10    maximum concentration by mass of a given chemical in an additive. For example, for the most
11    frequently used chemical, methanol, the median maximum additive concentration reported in
12    FracFocus  disclosures is 30%, by mass, with a range of 0.44% to 100%  (5th to 95th percentile). This
13    suggests that methanol is generally used as part of a mixture of chemicals in the hydraulic
14    fracturing fluid, and may be stored in a mixture of chemicals or as pure methanol. This wide range
                      This document is a draft for review purposes only and does not constitute Agency policy.
      June 2015                                        5-18                    DRAFT—DO NOT CITE OR QUOTE

-------
      Hydraulic Fracturing Drinking Water Assessment
 1    of possible concentrations of methanol further complicates assessing the potential impact of spills,
 2    as the properties of the fluid will depend on the different chemicals present and on their
 3    concentrations. For all chemicals, spills of a highly concentrated chemical can have different
 4    potential impacts than spills of dilute mixtures.

      5.4.2.  Nationwide Oil versus Gas
 5    FracFocus 1.0 data also can elucidate the differences between the chemicals used for oil production
 6    and those used for gas production, providing a better understanding of potential spill impacts from
 7    each. Table C-l and C-2 in Appendix C present the chemicals reported in at least 10% of all gas (34
 8    chemicals) and oil (39 chemicals) disclosures nationwide.

 9    Many of the same chemicals are used for oil and gas, but some chemicals are used more frequently
10    in oil production and others more frequently in gas.1 For example, hydrochloric acid is the most
11    commonly reported chemical for gas wells (73% of disclosures); it is the fifth most frequently
12    reported chemical for oil wells (58% of disclosures). However, both oil and gas operators each
13    reports using methanol in 72% of disclosures. Methanol is the most common chemical used in
14    hydraulic fracturing fluids at oil wells and the second most common chemical in hydraulic
15    fracturing fluids at gas wells.

      5.4.3.  State-by-State Frequency of Use of Hydraulic Fracturing Chemicals
16    We conducted a state-by-state analysis of chemical use based on FracFocus 1.0 disclosures (U.S.
17    EPA. 2015b). Some states reported more disclosures than other states, because they have relatively
18    more hydraulic fracturing activity and/or greater numbers of disclosures to FracFocus 1.0.
19    Reporting can bias national numbers towards those states with a disproportionate number of
20    disclosures. For example, the EPA (2015a) reported that Texas had 16,405  of the 34,675
21    disclosures with parsed ingredients and valid CASRNs and concentrations,  making up almost half
22    (47%) of all disclosures for the 20 states reporting to FracFocus 1.0. We attempt to account for the
23    possible effect of having a large number of disclosures in Texas by looking at a compilation of the
24    top 20 chemicals reported to FracFocus for all states.

25    Table 5-3 presents and ranks chemicals reported most frequently to FracFocus 1.0 for each state
26    (U.S. EPA. 2015b). There are 94 unique chemicals comprising the top 20 chemicals for each state,
27    indicating similarity in chemical usage among states.

28    Methanol is reported in 19 of the 20 (95%) states. Alaska is the only state in which methanol is not
29    reported (based on the state's 20 disclosures to FracFocus). The percentage of disclosures
30    reporting use of methanol ranges from 38% (Wyoming) to 100% (Alabama, Arkansas).

31    Ten chemicals (excluding water) are among the 20 most frequently reported in 14 of the 20 states.
32    These chemicals are: methanol; hydrotreated light petroleum distillates; ethylene glycol;
      1 This separation was done solely based on whether it was an oil or gas disclosure. The analysis did not separate out
      reservoir factors, such as temperature, pressure, or permeability, which may be important factors for which chemicals are
      used.
                   This document is a draft for review purposes only and does not constitute Agency policy.
      June 2015                                    5-19                   DRAFT—DO NOT CITE OR QUOTE

-------
     Hydraulic Fracturing Drinking Water Assessment
1    isopropanol; quartz; sodium hydroxide; ethanol; guar gum; hydrochloric acid; and peroxydisulfuric
2    acid, diammonium salt1 These 10 chemicals are also the most frequently reported chemicals
3    nationwide.

4    By performing this analysis by state, we observed that methanol is used across the continental U.S.
5    (not Alaska), and there are 9 other chemicals that are frequently used across the U.S. Beyond those,
6    however, there are a number of different chemicals that are used in one state more commonly than
7    others and many may not be used at all in other states. This suggests that there is regional
8    variability in some chemicals and a common set of the same chemicals that are frequently used.
     1 Quartz was the most commonly reported proppant and also reported as an ingredient in other additives [U.S. EPA.
     20153).



                   This document is a draft for review purposes only and does not constitute Agency policy.
     June 2015                                     5-20                   DRAFT—DO NOT CITE OR QUOTE

-------
Hydraulic Fracturing Drinking Water Assessment
                                                                                Chapter 5 - Chemical Mixing
Table 5-3. The percentage of disclosures of the 20 most commonly reported chemical by state, where a chemical is reported in at
           least three states.

           The 20 most frequently reported chemicals were identified for all 20 states that reported to FracFocus 1.0 (U.S. EPA, 2015b). The chemicals were
           ranked by counting the number of states where that chemical was in the top 20. Chemicals were then ranked so that chemicals used most widely
           among the most states come first. Methanol is reported in 19 of 20 states, so methanol is ranked first. A chemical was only presented on the list if
           it were reported in at least three states, resulting in 33 chemicals. The full table of top 20 chemicals (91 chemicals) is presented in Appendix C.
Chemical name
Methanol
Distillates, petroleum,
hydrotreated light
Ethylene glycol
Isopropanol
Quartz
Sodium hydroxide
Ethanol
Guargum
Hydrochloric acid
Peroxydisulfuric acid,
diammonium salt
Propargyl alcohol
Glutaraldehyde
Naphthalene
2-Butoxyethanol
CASRN
67-56-1
64742-47-8
107-21-1
67-63-0
14808-60-7
1310-73-2
64-17-5
9000-30-0
7647-01-0
7727-54-0
107-19-7
111-30-8
91-20-3
111-76-2
Percentage of disclosures per state3
AL
100%

100%
100%




100%



100%
100%
AK

45%
100%
65%
100%
100%

50%

50%



100%
AR
100%
56%
22%
44%

21%
45%

99%

61%
55%


CA
39%
55%
60%

89%
69%

93%

83%




CO
63%
74%

57%
23%
22%
50%

53%



30%

KS
79%
90%
63%
25%
23%
28%
80%

85%
22%
71%
75%
42%

LA
59%
84%
34%
51%
37%
53%
42%
49%
57%
27%
30%
36%
29%

Ml
93%
100%
71%
79%





57%
36%

86%
79%
MS
75%
100%
75%

50%
50%
100%
50%



50%


MT
63%
60%
49%
64%
64%
54%

43%
23%
62%


49%

NM
91%
63%
45%
62%
68%
30%
47%
63%
79%
75%
68%
57%

37%
ND
53%
47%
36%
37%
46%
52%
27%
55%

55%


43%

OH
52%
84%
57%
49%
45%

60%
51%
99%
64%
49%
72%


OK
70%
70%
47%
42%
27%
27%
46%
25%
76%
39%
41%
55%
25%

PA
69%
60%
34%
31%

17%
16%
23%
96%

58%
34%

21%
TX
78%
66%
59%
48%
43%
46%
21%
43%
71%
54%
39%
40%

24%
UT
79%
75%
85%
53%
40%
37%

43%
85%
39%
36%

38%
53%
VA
61%

28%
54%


63%

86%

28%

9%
89%
WV
64%
82%
59%
31%
22%

54%
23%
96%
54%
58%
71%

26%
WY
38%
51%

43%
30%
57%
25%
69%

64%

22%


June 2015
This document is a draft for review purposes only and does not constitute Agency policy.

                                5-21
DRAFT—DO NOT CITE OR QUOTE

-------
Hydraulic Fracturing Drinking Water Assessment
                                                                                      Chapter 5 - Chemical Mixing
Chemical name
Citric acid
Sodium chloride
Solvent naphtha,
petroleum, heavy
aromatics
Quaternary ammonium
compounds, benzyl-C12-
16-alkyldimethyl,
chlorides
2,2-Dibromo-3-
nitrilopropionamide
Potassium hydroxide
Acetic acid
Choline chloride
Polyethylene glycol
1,2,4-Trimethylbenzene
Ammonium chloride
CASRN
77-92-9
7647-14-5
64742-94-5
68424-85-1
10222-01-2
1310-58-3
64-19-7
67-48-1
25322-68-3
95-63-6
12125-02-9
Diatomaceous earth, 91053.39.3
calcined
Didecyl dimethyl
ammonium chloride
Sodium chlorite
Sodium erythorbate
N,N-Dimethylformamide
7173-51-5
7758-19-2
6381-77-7
68-12-2
Percentage of disclosures per state3
AL




100%



100%







AK











100%




AR



28%






21%

24%

33%

CA











71%




CO

35%
33%

49%

21%
27%

27%
28%





KS
46%

43%
54%





40%




30%

LA

41%



34%
28%






35%


Ml




71%



50%







MS

50%

50%

100%






50%
100%


MT


70%

36%
60%


36%







NM
40%






34%



38%




ND

25%
49%


59%


29%
25%



24%


OH
66%


37%

73%

38%




32%


47%
OK
36%

31%
33%


24%








20%
PA
29%


16%
34%



29%

31%





TX
24%
21%
17%


39%
31%









UT
79%






52%



35%



33%
VA
80%
9%
9%

28%


57%

9%




13%

WV
41%
22%

22%






21%

21%

32%

WY

23%
35%



31%






23%


June 2015
This document is a draft for review purposes only and does not constitute Agency policy.
                                   5-22
DRAFT—DO NOT CITE OR QUOTE

-------
Hydraulic Fracturing Drinking Water Assessment
                                                                                           Chapter 5 - Chemical Mixing
Chemical name
Nonyl phenol ethoxylate
Poly(oxy-l,2-ethanediyl)-
nonylphenyl-hydroxy
(mixture)
Sodium persulfate
Tetramethylammonium
chloride
CASRN
9016-45-9
127087-87-0
7775-27-1
75-57-0
Percentage of disclosures per state3
AL




AK




AR




CA




CO

25%


KS

40%


LA




Ml




MS


100%

MT



44%
NM
30%



ND



29%
OH




OK




PA


16%

TX




UT
36%



VA
32%
9%


WV




WY


26%
26%
a Analysis considered 34,675 disclosures and 676,376 ingredient records that met selected quality assurance criteria, including: completely parsed; unique combination of
fracture date and API well number; fracture date between January 1, 2011, and February 28, 2013; valid CASRN; and valid concentrations. Disclosures that did not meet quality
assurance criteria (3,855) or other, query-specific criteria were excluded from analysis.
June 2015
This document is a draft for review purposes only and does not constitute Agency policy.
                                     5-23
DRAFT—DO NOT CITE OR QUOTE

-------
      Hydraulic Fracturing Drinking Water Assessment
      5.4.4.  Volumes of Chemicals Used
 1    Understanding the volume of chemicals used at each hydraulic fracturing site is important for
 2    understanding potential impacts of chemicals to drinking water resources, because the chemical
 3    volume governs how much will be stored on-site, the types of containers required, and the total
 4    amount that could spill. While the on-site operator has precise knowledge of the composition and
 5    volume of chemicals stored on-site, this information is generally not publicly available. We
 6    conducted a comprehensive review of publicly available sources and found two sources [OSHA.
 7    2014a. bj Sjolander et al.. 2 011 ] that identify specific chemicals used at a hydraulic fracturing site
 8    and provide information on volumes. These are presented in Table 5-4. The volume of chemicals
 9    totaled 7,500 gal (28,000 L)  and 14,700 gal (56,000 L) for the two sources, with a mean volume for
10    an individual chemical of 1,900 gal (7,000 L) and 1,225 gal (4,600 L), respectively. The range of
11    volumes for each chemical used is 30 to 3,690 gal (114 to 14,000 L).
      Table 5-4. Example list of chemicals and volumes used in hydraulic fracturing.
                Volumes are for wells with and unknown number of stages and at least one perforation zone. Every
                well and fluid formation is unique. Volumes may be larger for longer horizontal laterals and with a
                greater number of stages.
Ingredient
Water
Proppant
Acid
Friction
reducer
Surfactant
Potassium
chloride
Gelling agent
Scale
inhibitor
Examples

Sand
Hydrochloric acid or
muriatic acid
Polyacrylamide,
mineral oil
Isopropanol

Guar gum or
hydroxymethyl
cellulose
Ethylene glycol,
alcohol, and sodium
hydroxide
Siolander et al. (2011) '

Volume (gal) or
mass (Ibs)
4,000,000 gal
~ 1,500,000 lbsd
1,338 gal
2,040 gal


e

Percent
overall
94.62
5.17
0.03
0.05


e

Occupational Safety and Health
Administration
(OSHA, 2014a, b)b

Volume (gal)
2,700,000
285,300
3,690
2,640
2,550
1,800
1,680
1,290
Percent by
volume
90
9.51
0.123
0.088
0.085
0.06
0.056
0.043
                   This document is a draft for review purposes only and does not constitute Agency policy.
      June 2015                                        5-24               DRAFT—DO NOT CITE OR QUOTE

-------
Hydraulic Fracturing Drinking Water Assessment
Ingredient
pH buffer
Preservative
Crosslinker
Iron control
Corrosion
inhibitor
Biocide/
antimicrobial
agent
Gel-breaker
Examples
Carbonate
Ammonium persulfate
Borate salts
Citric acid
n,n-Dimethyl
formamide
Glutaraldehyde,
ethanol, methanol
Ammonium persulfate
All chemicals
Chemical Volume:
Mean
(full range)
Siolander et al. (2011V
Volume (gal) or
mass (Ibs)


e
e
e
2,040 gal
e
7,458 gal
1,864.5 gal
(1,338-
2,040 gal)
Percent
overall


e
e
e
0.05
e
0.21

Occupational Safety and Health
Administration
(OSHA. 2014a. b)b
Volume (gal)
330
300
210
120
60
30

14,700
1,225
(30-3,690)
Percent by
volume
0.011
0.01
0.007
0.004
0.002
0.001

0.49

 a Adapted from Penn State "Water Facts" publication entitled "Introduction to Hydrofracturing" (Siolander et al., 2011).
 Composite from two companies: Range Resources, LLC, and Chesapeake Energy, which released in July 2010 the chemistry
 and volume of materials typically used in their well completions and stimulations.
  Adapted from a table generated by the OSHA for use  in a training module (OSHA, 2014a, b).
 c As presented in Siolander et al. (2011); does not explicitly state percent by mass or by volume.
 d The Penn State publication presented proppant in pounds instead of gallons.
 e Listed as an ingredient, but no information on volume or percentage.
                 This document is a draft for review purposes only and does not constitute Agency policy.
June 2015                                                 5-25                  DRAFT—DO NOT CITE OR QUOTE

-------
       Hydraulic Fracturing Drinking Water Assessment
 1     Because of the limited information on chemical volumes publicly available, we estimated chemical
 2     volumes used across the nation based on the information provided in the FracFocus database.
 3     Figure 5-5 plots median estimated chemical volumes, ranked from high to low, with the range of 5th
 4     to 95th percentiles.1 Volumes used are presented for the 74 chemicals that were reported to
 5     FracFocus in at least 100 disclosures and for which density data were available.2 The estimated
 6     median volumes vary widely among the different chemicals, covering a range of near zero to 27,000
 7     gal (98,000 L). The mean of the estimated median volumes was  650 gal (2,500 L).3

 8     With the median chemical volume, we can estimate total chemical volume for all chemicals used.
 9     Based on the above mean of median chemical volumes of 650 gal (2,500 L) per chemical, and given
10     that the median number of chemicals used at a site is 14 (U.S. EPA. 2015a], an estimated 9,100 gal
       1 Volumes were estimated using FracFocus disclosures. The total hydraulic fracturing fluid volume reported was used to
       calculate the total fluid mass by assuming the fluid has a density of 1 g/mL. This is a simplifying assumption based on the
       fact that more than 93% of disclosures are inferred to use water as a base fluid. Water had a median concentration of 88%
       by mass in the fracturing fluid. Some disclosures reported using brine, which has a density between 1.0 and 1.1 g/mL.
       This would introduce at most an error of 10% for the fluid calculation (the difference of a chemical being present at 10
       versus 9 gal, 1,000 versus 900 gal). We also assume that the mass of chemicals present in calculating the total fluid mass
       is negligible. Given that <2% of the fluid volume is non-water chemicals, and assuming the density of which is 3 mg/L, the
       error introduced is approximately 6%. For reference, for the chemicals we are calculating volumes, chlorine dioxide is the
       densest at 2.757 mg/L. Chemical with densities less than 1 mg/L introduce approximately <1% error.
       Next, the mass of each chemical per disclosure was calculated. Each chemical is reported to FracFocus 1.0 as a maximum
       concentration by mass in the hydraulic fracturing fluid. This introduces error, as we only know that it is equal to or less
       than this mass fraction. In the U.S. EPA(2015a) EPA's analysis of the FracFocus 1.0 database, an additive is comprised of
       three chemicals with maximum ingredient concentration of 60% in the additive and a maximum concentration of 0.22%
       in the hydraulic fracturing fluid. Each of the three chemicals cannot be present at 60%. We have no way to know the
       actual proportions of each chemical in the additive and thus must calculate chemical mass based on the given information.
       Therefore, our calculations likely overestimate actual volumes. However, in some cases, the concentration in the additive
       that is given is less than 100% and only one chemical is listed in the additive. In these cases, it appears that the disclosure
       is reporting the concentration of that chemical in water. Hydrogen chloride is listed as the sole ingredient in the acid
       additive, and the maximum concentration is 40% by mass. In this case, the HC1 is diluted down to 40%, so the total
       volume would be underestimated.
       After all the chemical masses are calculated, the volume is calculated by dividing chemical mass by density.
       Given the limited information available, due to the limits of the FracFocus database and general lack of publicly available
       data, and despite the errors associated with these calculations, these calculations  provide context for the  general
       magnitude of volumes for each of the chemicals used on-site. These calculations are used to calculate median volumes for
       each chemical. These volume calculations are for the chemicals themselves, not the additives.
       Analysis considered 34,495 disclosures and 672,358 ingredient records that met selected quality assurance criteria,
       including: completely parsed; unique combination of fracture date and API well number; fracture date between January 1,
       2011, and February 28,2013; criteria for water volumes; valid CASRN; and valid concentrations. Disclosures that did not
       meet quality assurance criteria  (4,035) or other, query-specific criteria were excluded from analysis.
       2 Density data were gathered from Reaxys® and other sources as noted. Reaxys® (http://www.elsevier.com/online-
       tools/reaxys) is an online database of chemistry literature and data. Direct density source, as provided by Reaxys®, is
       provided in Table C-7 in Appendix C.
       3 Reporting records to FracFocus 1.0 was required in six of the 20 states between January 1,2011 and February 28,2013.
       An additional three states required disclosure to either FracFocus or the state, and five states required reporting to the
       state. Reporting to FracFocus 1.0 was optional in other states.  Some states changed their reporting requirements during
       the course of the study. The FracFocus 1.0 database therefore does not encompass all data on chemicals used in hydraulic
       fracturing. As stated in Text Box 4-2, this mix of voluntary versus mandatory disclosure requirements limits the
       completeness of data extracted  from FracFocus 1.0 for estimating hydraulic fracturing fluid. According to a comparison
       between FracFocus reported fluid volumes and literature values, water use per well was reported to be about 86% of the
       actual used (median of estimated values. See Chapter 4, Text Box 4-1). If the fluid volume is underreported, then
       estimated chemical volumes based on fluid volume would be similarly underestimated. Using the underreporting of 86%,
       then the estimated median chemical volume would be 760 gal.
                      This document is a draft for review purposes only and does not constitute Agency policy.

      June 2015                                             5-26                 DRAFT—DO NOT CITE OR QUOTE

-------
      Hydraulic Fracturing Drinking Water Assessment
 1    (34,000 L) of chemicals maybe used per well. Given that the number of chemicals per well ranges
 2    from 4 to 28 [U.S. EPA. 2015a], the total volume of chemicals per well may range from 2,600 to
 3    18,000 gal (9,800 to 69,000 L).

 4    Another way to estimate total volume of chemicals per well is to use the estimated median volume
 5    of 1.5 million gal (5.7 million L) of fluid used to fracture a well (see Chapter 4) (U.S. EPA. 2015a]
 6    and assume that up to 2% of that volume are chemicals added to base fluid (Carter etal., 2013:
 7    Knappe and Fireline. 2012). resulting in up to 30,000 gal (114,000 L) of chemicals used per well.
        100000
                                                                   Median of median volumes: 21 gal
                                                                   Mean of median volumes: 650 gal
                                                                   Maximum of median volumes: 27,000 gal
          0.01
         0.001 -=-
        0.0001 -=-

      Figure 5-5. Estimated median volumes for chemicals reported in at least 100 FracFocus
                disclosures by February 28, 2013 for use in hydraulic fracturing from
                January 1, 2011 to February 28, 2013.
                Shaded area represents the zone of 5% and 95% confidence limits. Derived from (U.S. EPA, 2015b).

 8    Using the mean of the median chemical volumes from disclosures in FracFocus 1.0, we can also
 9    estimate volume per additive and extrapolate to estimate on-site chemical storage. If we assume
10    three to five chemicals per additive, then total volume per additive stored on-site would
11    approximate 1,900 to 3,200 gal (7,400 to 12,000 L). On-site containers generally store 20% to
                   This document is a draft for review purposes only and does not constitute Agency policy.
      June 2015                                        5-27               DRAFT—DO NOT CITE OR QUOTE

-------
      Hydraulic Fracturing Drinking Water Assessment
 1    100% more additive volume than needed [Houston etal.. 2009: Malone and Ely. 2007]. This would
 2    suggest that 2,300 to 6,500 gal (8,800 to 25,000 L) per additive are stored on-site.

 3    The volume that may be released during a spill depends on where in the chemical mixing process
 4    the spill occurs. Spills from chemical or additive containers may result in higher volume spills than
 5    the estimated volumes in Figure 5-5 suggest However, if the spill is of the hydraulic fracturing fluid
 6    downstream of the blender, then the total volume of chemical spilled may be less than the
 7    estimated total volumes held on site.

      5.5.  Chemical Management and Spill Potential
 8    This section provides a description of the primary equipment used in the chemical mixing and well
 9    injection processes, along with a discussion of the spill vulnerabilities specific to each piece of
10    equipment. Equipment breakdown or failure can trigger a spill itself, and it can also lead to a
11    suspension of activity and the disconnection and reconnection of various pipes, hoses, and
12    containers. Each manipulation of equipment poses additional potential for a spill. The EPA found
13    that approximately one-third of chemical spills on or near the well pad related to hydraulic
14    fracturing resulted from equipment failure [U.S. EPA. 2015n). When possible, we describe
15    documented spills associated with or attributed to specific pieces of equipment, in text boxes in the
16    relevant subsections.

17    Hydraulic fracturing operations are mobile and must be assembled at each well site, and each
18    assembly and disassembly presents a potential for spills. Equipment used in the chemical mixing
19    and well injection processes typically consists of chemical storage trucks, oil storage tanks/tanker
20    trucks; a slurry blender; one or more high-pressure, high-volume fracturing pumps; the main
21    manifold; surface lines and hoses; and a central control unit There are many potential sources for
22    leaks and spills in this interconnected system.

23    Equipment varies in age and technological advancement depending upon service company
24    standards and costs associated with purchase and maintenance. Older equipment may have
25    experienced wear and tear, which may be a factor in spills caused by equipment failure. New
26    equipment may be more automated, reducing opportunities for human error. Information detailing
27    the extent of technological and age differences in fracturing equipment across sites and operators is
28    limited. Table 5-5 provides a description of typical hydraulic fracturing equipment.
                   This document is a draft for review purposes only and does not constitute Agency policy.
      June 2015                                       5-28              DRAFT—DO NOT CITE OR QUOTE

-------
     Hydraulic Fracturing Drinking Water Assessment
    Table 5-5. Examples of typical hydraulic fracturing equipment and their functions.
Equipment
Acid transport truck
Chemical storage truck
Base fluid tanks
Proppant storage units
Blender
High-pressure fracturing pumps
Manifold trailer with hoses and
pipes
Fracturing wellhead or frac head
Function
Transports acids to job sites, the truck has separate compartments for
multiple acids or additives.
Transport chemicals to the site in separate containment units or totes.
Chemicals are typically stored on and pumped from the chemical storage
truck.
Store the required volume of base fluid to be used in the hydraulic fracturing
process.
Hold proppant and feed it to the blender via a large conveyor belt.
Takes fluid (e.g., water) from the fracturing tanks and proppant (e.g., sand)
from the proppant storage unit and combines them with additives before
transferring the mixture to the fracturing pumps
Pressurize mixed fluids received from the blender and injected into the well.
A transfer station for all fluids. Includes a trailer with a system of hoses and
pipes connecting the blender, the high-pressure pumps, and the fracturing
wellhead.
A wellhead connection that allows fracture equipment to be attached to the
well.
1    While the primary equipment and layout is generally the same across well sites, the type, size, and
2    number of pieces of equipment may vary depending on a number of factors [Malone and Ely. 2007]:
3        •   The size and type of the fracture treatment;
4        •   The number of wells drilled per well pad;
5        •   The location, depth, and length of the fractures;
6        •   The volumes and types of additives, proppants, and fluids used; and
7        •   The operating procedures of the well operator and service company (e.g., some companies
8            require backup systems in case of mechanical failure, while others do not).
9    Figure 5-6 provides a schematic diagram of a typical layout of hydraulic fracturing equipment
                  This document is a draft for review purposes only and does not constitute Agency policy.
    June 2015                                       5-29               DRAFT—DO NOT CITE OR QUOTE

-------
      Hydraulic Fracturing Drinking Water Assessment
           Water Tanks
      /         /
III      II
 High Pressure Pumps
                                                                                   Flowback
                                                                                    Tanks
                                                       I
     11
     II
                                                  High Pressure Pumps

                                                      ''         ''
                                                           Manifold
                                                                                        Frac
                                                                                        Head
            	> low pressure lines   	»  high pressure lines

      Figure 5-6. Typical hydraulic fracturing equipment layout.
                This illustration shows how the various components of a typical hydraulic fracturing site fit together.
                Numbers of pumps and tanks vary from site to site. Some sites do not use a hydration unit as the gel is
                batch mixed prior to the treatment (Olson, 2011; BJ Services Company, 2009).

      5.5.1.  Storage
 1    This section provides an overview of publicly available information on storage and containment of
 2    chemicals used in the hydraulic fracturing process. Most public sources provide general
 3    information  on the types and sizes of containment units. While operators maintain a precise
 4    inventory of volumes of chemicals stored and used for each site, this information is typically not
 5    made public.

 6    The volumes of each chemical used are based on the size and site-specific characteristics of each
 7    fracture treatment Sites often store an excess of the design volume of chemicals for contingency
 8    purposes. Malone and Ely [2007] indicate that companies typically store an excess of 20% of the
 9    required chemical inventories on location. Houston etal. [2009] recommends maintaining an on-
10    site chemical reserve of 100% extra beyond what is necessary for the fracturing treatment
11    [Houston et al., 2009]. See Text Box 5-4 for documented spills from storage  units.
                    This document is a draft for review purposes only and does not constitute Agency policy.
      June 2015                                         5-30               DRAFT—DO NOT CITE OR QUOTE

-------
      Hydraulic Fracturing Drinking Water Assessment
 1
 2
 3
 4
 5
 6
 7
Text Box 5-4. Spills from Storage Units.

Of the 151 spills of chemicals, additives, or fracturing fluid discussed and evaluated in fU.S. EPA. 2015n) (see
Text Box 5-13 for more information), 54 spills were from storage units. Storage units include smaller totes or
tanks used for storing individual chemicals or additives, and larger tanks containing fracturing fluid. These
spills resulted from equipment failure, failure of storage integrity, or human error. Sixteen of these spills were
due to failure of container integrity, which includes holes and cracks in containers, demonstrating the need
for properly constructed and maintained storage units. The remaining spills from storage containers resulted
from human error or equipment malfunctions, or had an unknown cause.
      5.5.1.1. Hydraulic Fracturing Base Fluid Storage
 8    Base fluids used in hydraulic fracturing are typically stored on-site in large volume tanks. Non-
 9    water-based fluids may be stored in specialized containment units designed to prevent or minimize
10    releases. For example, nitrogen and carbon dioxide must be stored in compressed gas or cryogenic
11    liquid cylinders, as required by U.S. Department of Transportation (DOT) and OSHA regulations.
12    Due to the large volume of base fluid storage tanks (about 21,000 gal (80,000 L) (Halliburton.
13    1988), uncontrolled spills could damage other storage units and equipment, which could result in
14    additional spills. Fresh water used as a base fluid is generally not a source of concern for spills.
15    Reused wastewater, brine, and non-aqueous base fluids have the potential to adversely impact
16    drinking water resources in the event of a spill. An example of a documented spill of hydraulic
17    fracturing fluid is presented in Text Box 5-5.
18
19
20
21
22
Text Box 5-5. Spills of Fracturing Fluid Documented to Impact Surface Water.

In September 2009, two spills of hydraulic fracturing fluid occurred at the same site in Pennsylvania. A total
of approximately 7,350 gal (28,000 L) of fluid comprised of a mixture of water, gel, and friction reducer
leaked and migrated to Stevens Creek. While the causes of the spills are not clear, it appears that a
pressurized line may have broken and a seal may have failed (U.S. EPA. 2015n. Appendix B Line 307:
Lustgarten. 2009).
      5.5.1.2. Chemical Additive Storage
23    Additives are typically stored on-site in the containers in which they were transported and
24    delivered. The chemical additive trailer typically consists of a flatbed truck or van enclosure that
2 5    holds a number of chemical totes, described below, and is equipped with metering pumps that feed
26    chemicals to the blender. Depending on the size and type of the fracturing operation, there may be
27    one or more chemical additive trailers per site (ALL Consulting. 2012: NYSDEC. 2011). Additives
28    constitute a relatively small portion of fluids used in a hydraulic fracturing fluid, although they can
29    still be used in volumes ranging from the tens to tens of thousands of gallons.

3 0    The storage totes generally remain on the transportation trailers, but they also may be unloaded
31    from the trailers and transferred to alternative storage areas before use. Our investigation did not
3 2    find much information on how often, when, or why these transfers occur. Additional transfers and
                    This document is a draft for review purposes only and does not constitute Agency policy.
      June 2015                                         5-31               DRAFT—DO NOT CITE OR QUOTE

-------
      Hydraulic Fracturing Drinking Water Assessment
 1    movement can increase the likelihood of a spill. See Text Box 5-6 for documented spills from an
 2    additive storage unit.

      Text Box 5-6. Spill from Additive (Crosslinker) Storage Tote.

 3    During a tote transfer in Pennsylvania, a tote of crosslinker fell off a forklift, spilling approximately 15-20 gal
 4    (60-80 L) onto the well pad. The area was scraped clean with a backhoe and placed in a lined containment
 5    area fU.S. EPA. 2015n. Line 3091

 6    The most commonly used chemical totes are 200-400 gal (760-1,500 L) capacity polyethylene
 7    containers that may be reinforced with steel or aluminum mesh (NYSDEC. 2011]. Metal containers
 8    of the same capacity may also be used (ALL Consulting. 2012: UWS. 2008). The totes are typically
 9    equipped with bottom release ports, which enable the direct feed of the additives to the blending
10    equipment (NYSDEC, 2011]. Spills may occur if lines are improperly connected to these ports or if
11    the connection equipment is faulty.
      Figure 5-7. Metal and high-density polyethylene (HOPE) chemical additive units.
                The image on the left depicts metal totes (industry source). The image on the right depicts plastic
                totes (NYSDEC, 2011).

12    Certain additives require specialized containment units with added spill prevention measures. For
13    example, additives containing methanol may be subject to federal safety standards, and industry
14    has developed guidance on methanol's safe storage and handling (Methanol Institute. 2013].

15    Dry additives are typically transported and stored on flatbed trucks in 50 or 55 Ib (23 or 25 kg]
16    bags, which are set on pallets containing 40 bags each (NYSDEC. 2011: UWS. 2008: Halliburton.
17    1988]. Proppants are stored on-site in large tanks or bins with typical capacities of 220 to 440 Ib
18    (100 to 200 kg] (ALL Consulting. 2012: BT Services Company. 2009: Halliburton. 1988].
                   This document is a draft for review purposes only and does not constitute Agency policy.
      June 2015                                        5-32               DRAFT—DO NOT CITE OR QUOTE

-------
      Hydraulic Fracturing Drinking Water Assessment
      5.5.1.3. Acid Storage
 1    Acids are generally stored on-site in the containment units in which they are transported and
 2    delivered. Atypical acid transport truck holds 3,000 to 5,000 gal (11,400 to 19,000 L) of acid and
 3    can have multiple compartments to hold different kinds of acid [Arthur etal.. 2009b). Acids such as
 4    hydrochloric acid and formic acid are corrosive and can be extremely hazardous in concentrated
 5    form. Therefore, acid transport trailers and fracture tanks must be lined with chemical-resistant
 6    coating designed to prevent leakage and must meet applicable DOT regulatory standards (pursuant
 7    to 40 CFR 173) designed to prevent or minimize spills.

 8    Acid fracture treatments may use thousands of gallons (thousands of liters) of acid per treatment
 9    (Spellman. 2012). Given the large volumes used, failure of containment vessels during storage or
10    failure  of connections and hoses during pumping could result in high-volume acid spills. Eight spills
11    (out of 105 spills from state data sources) of acid or fracturing fluid containing acid were reported
12    to state data sources examined by the EPA (2015n). The median volume of these acid spills was
13    approximately 1,600  gal  (6,000 L) (Lines 240, 241,  248, 258, 264, 272, 278, and 281 in Appendix B
14    of U.S. EPA(2015n)).

15    Details of a documented acid spill are presented in Text Box 5-7.
16
17
18
19
Text Box 5-7. Spill of Acid from Storage Container.

In July 2014, over 20,000 gal (76,000 L) of hydrochloric acid spilled from a storage container when a flange
malfunctioned. The acid spilled into a nearby alfalfa field, where it was contained with an emergency berm
(Phillips. 2014: Wertz. 2 0141.There is no information on how much leached into soils or if the spill reached
drinking water resources.
      5.5.1.4. Gel Storage
20    Gels can be added to hydraulic fracturing fluid using either batch or continuous (also called "on-the-
21    fly") mixing systems. Gelling agents and gel slurries are stored differently on-site and may pose
2 2    different potential spill scenarios depending on whether the site is using batch or continuous
23    mixing processes (BJ Services Company. 2009).

      5.5.1.5. Batch Mixing
24    In a typical batch mixing process, powdered gelling agents and related additives (e.g., buffers,
25    surfactants, biocides) are mixed on-site with base fluid water in large tanks (typically 20,000 gal or
26    76,000 L) (BJ Services Company. 2009: Halliburton. 1988).

27    The number of gel slurry tanks used varies based on site-specific conditions and the size of the
28    fracture job. These tanks may be subject to leaks or overflow during the batch mixing process and
29    during storage prior to injection. One of the disadvantages of batch mixing is the need for multiple
30    suction hoses to draw pre-gelled fluids from storage tanks into the blender, which may increase the
31    potential for spills. Yeager and Bailey (2013) state that a drawback of batch mixing is the "fluid
32    spillage and location mess encountered when pre-mixing tanks," suggesting that small spills are not
                    This document is a draft for review purposes only and does not constitute Agency policy.
      June 2015                                        5-33               DRAFT—DO NOT CITE OR QUOTE

-------
      Hydraulic Fracturing Drinking Water Assessment
 1    uncommon during batch mixing. Details of a documented gel slurry spill are presented in Text Box
 2    5-8.
 3
 4
 5
 6
Text Box 5-8. Spill of Gel during Mixing.

On April 9, 2010, a company was mixing a gel slurry for an upcoming fracture job. The tank had developed a
crack, which allowed approximately 10,000 gal (38,000 L] of water mixed with 60 gal (230 L] of gel to leak
out. The mixture did not reach a water receptor, and absorbents were used to clean up the gel (U.S. EPA.
2015n. Appendix B Line 220].
      5.5.1.6. Continuous Mixing (On-the-Fly)
 7    In continuous mixing operations, powdered gels are typically replaced with liquid gel concentrates
 8    [Allen. 2013: BJ Services Company. 2009]. Operators prepare dilute gelling agents as needed using
 9    specialized hydration units [BT Services Company, 2009]. Liquid gel concentrates may be stored on-
10    site in single-purpose tanker trucks [Harms and Yeager. 1987]. but are more often stored in
11    specialized mixing and hydration units [Ayalaetal., 2006]. Continuous mixing requires less
12    preparation than batch mixing but typically requires more equipment [BT Services Company, 2009:
13    Browne and BD, 1999], which may increase the possibility for spills resulting from equipment
14    malfunctions or human error.

      5.5.2.  Hoses and Lines
15    High- and low-pressure hoses and lines are used to transfer hydraulic fracturing fluids from storage
16    units to specialized mixing and pumping equipment and ultimately to the wellhead. A discussion of
17    the different types of hoses and lines and possible points of failure is provided below. The following
18    photograph shows an example of hoses and lines at a hydraulic fracturing site.
                   This document is a draft for review purposes only and does not constitute Agency policy.
      June 2015                                        5-34               DRAFT—DO NOT CITE OR QUOTE

-------
      Hydraulic Fracturing Drinking Water Assessment
      Figure 5-8. A worker adjusts hoses at a hydraulic fracturing site near Mead, Colorado.
                Source: AP Photo/Brennan Linsley.

 1    Suction pumps and hoses move large volumes of base fluid to the blender. Incomplete or damaged
 2    seals in inlet or outlet connections can cause fluid leaks at the connection points. Improperly fitted
 3    seals also severely limit or eliminate suction lift, which may impair the suction pump and increase
 4    spill potential. Suction hoses themselves are susceptible to leaks due to wear and tear. Equipment
 5    providers recommend hoses be closely inspected to ensure they are in good operating condition
 6    prior to use [Upstream Pumping. 2015: BJ Services Company. 2009: Malone and Ely. 2007].

 7    Discharge hoses transfer additives from containment vessels or totes to the blender. Given the
 8    potential for concentrated chemicals to spill during transfer from storage totes to the blender, it is
 9    particularly important that these hoses are in good condition and that connector seals or washers
10    fit properly and are undamaged. Discharge hoses are also used to carry fracturing fluid pumped
11    from the blender via the low-pressure manifold to the high-pressure pumps. Proppant-heavy fluids
12    are pumped through discharge hoses at relatively low rates. If a sufficient flow rate is not
13    maintained, proppant may settle out, damaging pumps and creating potential for spills or leaks
14    [Upstream Pumping. 2015: BT Services Company. 2009: Malone and Ely. 20071.

15    High-pressure flow lines convey pressurized fluids from the high-pressure pumps into the high-
16    pressure manifold, and from the manifold into the wellbore. High-pressure flow lines are subject to
17    erosion caused by the high-velocity movement of abrasive, proppant-laden fluid. Curved sections of
18    flow lines (e.g., swivel joints) where abrasive fluids are forced to turn corners are particularly
19    subject to erosion and are more likely to develop stress cracks or other defects that may result in a
2 0    leak or spill. Safety restraints are typically used to prevent movement of flow lines in the event of
                   This document is a draft for review purposes only and does not constitute Agency policy.
      June 2015                                       5-35               DRAFT—DO NOT CITE OR QUOTE

-------
      Hydraulic Fracturing Drinking Water Assessment
 1    failure and to help control spills. High-pressure flow lines are pressure-tested to detect fatigue or
 2    stress cracks prior to the fracturing treatment [OSHA. 2015: BJ Services Company. 2009: Arthur et
 3    al.. 2008: Malone and Ely. 2007: Halliburton. 19881

 4    Nineteen spills of chemicals or fracturing fluids associated with leaks from hoses or lines had a total
 5    spill volume of 12,756 gal (48,300 L), with a median volume of 420 gal (1,600 L) (U.S. EPA. 2015n1.

      5.5.3.  Blender
 6    The blender is the central piece of equipment used to create the fracturing fluid for injection. It
 7    moves, meters, and mixes precise amounts of the base fluid, additives, and proppant and pumps the
 8    mixed slurry to high-pressure pumping equipment (BT Services Company. 2009: Malone and Ely.
 9    2007: Halliburton, 1988]. A typical blender consists of a centrifugal suction pump for pulling base
10    fluid, one or more chemical metering pumps to apportion the additives, one or more proportioners
11    to measure  and feed proppant, and a central agitator tank where fluid components are mixed
12    together.

13    The blending process is monitored to ensure that a uniform mixture  is maintained regardless of
14    injection rates and volumes. Excessive or reduced rates of flow during treatment can cause the
15    blender to malfunction or to shut down, which may result in spills. (Malone and Ely. 2007:
16    Halliburton. 1988). For aqueous hydraulic fracturing fluid blends, spills that occur downstream of
17    the blender will be a dilute mixture comprised primarily of water with a low concentration (less
18    than or equal to 2%) of chemicals. Details of a spill from a blender are presented in Text Box 5-9.
19
20
21
Text Box 5-9. Spill of Hydraulic Fracturing Fluid from Blender.

In May 2006, a blender malfunctioned during a fracture job in Oklahoma. Approximately 294 gal (1,100 L) of
fluid spilled into a nearby wheat field. The fluid consisted of hydrochloric acid, clay stabilizer, diesel, and
friction reducer. Contaminated soil was removed by the operator (U.S. EPA. 2015n. Appendix B Line 249].
      5.5.4.  Manifold
22    A trailer-mounted manifold and pump system functions as a central transfer station for all fluids
23    used to fracture the well. The manifold is a collection of low- and high-pressure pipes equipped
24    with multiple fittings for connector hoses. Fluids are pumped from the blender through the low-
25    pressure manifold hoses, which distribute it to high-pressure pump trucks. Pressurized slurry is
26    sent from the pump trucks through high-pressure manifold lines and into additional high pressure
27    lines that lead to the wellhead (Malone and Ely. 2007).
                   This document is a draft for review purposes only and does not constitute Agency policy.
      June 2015                                       5-36              DRAFT—DO NOT CITE OR QUOTE

-------
      Hydraulic Fracturing Drinking Water Assessment
      Figure 5-9. Manifold (pointed to by the white arrow).
               Source: Halliburton.

 1    Manifold and pump system components require varying amounts of manual assembly and undergo
 2    varying amounts of pre-testing [Malone and Ely. 2007). Improperly tested parts may be more likely
 3    to break or lose functionality, leading to a spill. In manifolds requiring more manual assembly, there
 4    may be more opportunities for human error. The EPA [U.S. EPA, 2015n] identified seven spills
 5    sourced from manifolds. Three of these spills, out of the 105 chemical or hydraulic fracturing fluid
 6    spills reported to state data sources, were fracturing fluid that resulted from either human error of
 7    equipment failure. These three spills were responsible for approximately 5,000 gal (19,000 L)  of
 8    spilled fluids (U.S. EPA. 2015n. Appendix B Lines 35.141.1601

      5.5.5. High-Pressure Fracturing Pumps
 9    High-pressure fracturing pumps take the fracturing fluid mixture from the blender, pressurize it,
10    and propel it down the well. Typically, multiple high-pressure, high-volume fracturing pumps are
11    needed for hydraulic fracturing [Upstream Pumping. 2015]. Such pumps come in a variety of sizes.
12    Bigger pumps move greater volumes of fluid at higher pressures; therefore, spills from these pumps
13    may be larger. Smaller pumps may require more operators and more maintenance [BT Services
14    Company. 2009). and therefore may result in more frequent spills.
                   This document is a draft for review purposes only and does not constitute Agency policy.
      June 2015                                       5-37              DRAFT—DO NOT CITE OR QUOTE

-------
      Hydraulic Fracturing Drinking Water Assessment
      Figure 5-10. High-pressure pumps on either side of the manifold.
               Source: http://drillingahead.com/roger-payne/gallery/14826/20000-psi-frac-near-
               caldwelltexas-2005#gallery img.

 1    The "fluid ends" of hydraulic fracturing pumps are the pump components through which fluids are
 2    moved and pressurized. Pump fluid ends must withstand enormous pressure and move a large
 3    volume of abrasive fluid high in solids content. They have multiple parts (e.g., seals, valves, seats
 4    and springs, plungers, stay rods, studs) that can wear out under the stress of high-pressure
 5    pumping [Upstream Pumping. 2015). Given the sustained pressures involved, careful maintenance
 6    of fluid ends is necessary to prevent equipment failure [Upstream Pumping, 2015: API, 2011].
 7    Details of a documented spill from a fracture pump are presented in Text Box 5-10.
      Text Box 5-10. Spill of Fluid from Fracture Pump.

 8    In December 2011, a fluid end on a fracture pump developed a leak, spilling approximately 840 gal (3,200 L)
      of fracturing fluid. A vacuum truck was used to recover the spilled fluid, and all affected soils were scheduled
10    to be neutralized and taken to a landfill at the end of the job, after removal of the equipment (U.S. EPA, 2015n,
11    Appendix B Line 14).
                   This document is a draft for review purposes only and does not constitute Agency policy.
      June 2015                                        5-38               DRAFT—DO NOT CITE OR QUOTE

-------
      Hydraulic Fracturing Drinking Water Assessment
      5.5.6.  Surface Wellhead for Fracture Stimulation
 1    A wellhead assembly, often referred to as a frac head or frac stack, is temporarily installed on the
 2    wellhead during the fracture treatment. The frac head assembly allows high volumes of high-
 3    pressure proppant-laden fluid to be injected into the formation [OSHA. 2015: Halliburton. 2014:
 4    Stinger Wellhead Protection, 2010]. The temporary frac head is equipped with specialized isolation
 5    tools so that the wellhead is protected from the effects of pressure and abrasion.
      Figure 5-11. Multiple fracture heads.
               Source: DOE/NETL

 6    As with all components of hydraulic fracturing operations, repeated and prolonged stress from
 7    highly pressurized, abrasive fluids may lead to equipment damage. The presence of minute holes or
 8    cracks in the frac head may result in leaks when pressurized fluids are pumped. In addition, surface
 9    blowouts or uncontrolled fluid releases may occur at the frac head because of valve failure or
10    failure of other components of the assembly. Details of a documented frac head failure are
11    presented in Text Box 5-11. Details on the Killdeer, ND, blowout and associated spill are presented
12    in Text Box  5-12.
                   This document is a draft for review purposes only and does not constitute Agency policy.
      June 2015                                        5-39              DRAFT—DO NOT CITE OR QUOTE

-------
      Hydraulic Fracturing Drinking Water Assessment
      Text Box 5-11. Spill from Frac Head Failure.

 1    On March 2,2011, a frac head failed during fracturing operations in Colorado. Approximately 8,400 gal
 2    (32,000 L) of slickwater fracturing fluid leaked. The majority of the spill was contained on-site, though a small
 3    amount ran off into a nearby cornrow. Some of the fluid was recovered, and saturated soils were scraped and
 4    stockpiled on the well pad [U.S. EPA, 2015n, Appendix B Line 751.
      Text Box 5-12. The Killdeer Case Study.

 5    In September 2010, a blowout occurred in the Franchuk 44-20 SWH well, in Dunn County near Killdeer, ND.
 6    Hydraulic fracturing fluids, oil, and flowback water spilled onto the land and possibly entered the Killdeer
 7    aquifer, which is a source of drinking water. The EPA investigated a reported blowout event at the Killdeer
 8    site as part of a retrospective case study. The study area is comprised of historical oil and gas production and
 9    current hydraulic fracturing. The discussion below was taken from the EPA Killdeer case study [U.S. EPA.
10    2015J).

11    Methods
12    Water quality samples were collected from three domestic wells, nine monitoring wells (installed by
13    Terracon), two supply wells, one municipal well, and one state well during three rounds in July 2011, October
14    2011, and October 2012. The geochemistry of water samples was investigated by analyzing major ions, trace
15    metals, methane/ethane gas concentrations, volatile organic compounds, semivolatile organic compounds,
16    glycol ethers, diesel- and gasoline-range organics, low-molecular-weight acids, and selected stable isotopes.
17    Data collected from this study were statistically compared with historical water quality data retrieved from
18    the literature and national water quality databases. To help determine whether hydraulic fracturing
19    processes were a cause of alleged impacts on water quality, detailed environmental record searches were
20    conducted to help identify other potential contaminant sources.

21    Results
22    Three study wells, NDGW09, NDGW08, and NDGW07, were excluded from the comparisons with historical
23    data. NDGW09 was excluded since it was screened in the Sentinel Butte aquifer so a comparison with
24    historical Killdeer aquifer water quality data was not appropriate. NDGW08 and NDGW07 had significant
25    differences in water quality compared to the remaining study wells. These two wells showed differences in
26    ion concentrations (e.g., chloride, calcium, magnesium, sodium, strontium) as well as tert-butyl alcohol (TEA).
27    The remaining study wells were then compared with historical data to determine if these wells represented
28    background water quality of the Killdeer aquifer. This comparison between the remaining study wells and
29    Killdeer aquifer historical water quality data indicated that these remaining study wells were in general
30    consistent with the historical background data and then used for the data analysis as background wells. There
31    were limited detections of other organic compounds in the study wells. In most cases, with the exception

32    of TEA, the detected organic compounds could not be directly linked to the blowout or hydraulic fracturing,
33    as these chemicals could have originated from other sources including vehicular traffic, generators used to
34    power well pumps, flaring of methane  from the pad production wells, and cement used to repair a well the
3 5    day prior to  sampling.
                     This document is a draft for review purposes only and does not constitute Agency policy.
      June 2015                                           5-40                DRAFT—DO NOT CITE OR QUOTE

-------
      Hydraulic Fracturing Drinking Water Assessment
 1
 2
 3
 4
 5
 6
 7

 8
 9
10
11
12
13
14
15
16
Comparisons of TEA between the study data and historical data could not be made since no historical data for
TEA were found for the Killdeer aquifer. TEA data were compared with the background study wells and were
found to be different. Based on the analysis of other potential sources of contamination, the EPA determined
that the only other potential sources of TEA were gasoline spills, leaky underground storage tanks, and
hydraulic fracturing fluids. The data from this study suggest the TEA resulted from the degradation of tert-
butyl hydroperoxide used during the hydraulic fracturing of the Franchuk well since MTBE and other
signature compounds associated with gasoline or fuels were not present in NDGW08 and NDGW07.

Conclusion
The analysis of data from NDGW008 and NDGW07 indicated that the main impact on water quality was from
briny water and TEA mixing with Killdeer aquifer water in these wells. In all cases, the fingerprinting
techniques used indicated that the impacts on NDGW07 and NDGW08 were consistent with deep formation
brines underlying the Killdeer study location. Based on the data analysis performed for the Killdeer case
study, the observed impacts on NDGW07 and NDGW08 were likely caused by the blowout that occurred at
the Franchuk 44-20. This evidence, along with the absence of another plausible candidate cause, strongly
suggests impact on a drinking water resource from the blowout during the hydraulic fracturing of the
Franchuk 44-20 SWH well.
      5.6. Spill Prevention, Containment, and Mitigation
17    Several factors influence spill prevention, containment, and mitigation, including Federal, State, and
18    local regulations and company practices. State regulations governing spill prevention, containment,
19    and mitigation at hydraulic fracturing facilities vary in scope and stringency [Powell. 2013: GWPC.
20    2009]. Employee training and equipment maintenance are also factors in effective spill prevention,
21    containment, and mitigation. Analysis of these factors was outside the scope of this assessment

22    Hydraulic fracturing operating companies themselves  may develop and implement spill prevention
23    and containment procedures. The American Petroleum Institute has a guidance document Practices
24    for Mitigating Surface Impacts Associated with Hydraulic Fracturing [API. 2011]. The document
25    describes practices currently used in the oil and natural gas industry to minimize potential surface
26    environmental impacts. As another example, the province of New Brunswick, Canada, released
27    rules for industry on responsible environmental management of oil and natural gas activities [GNB.
28    2013]. It was beyond the scope of this assessment to evaluate the efficacy of the practices in these
29    documents or the extent to which they are implemented.

30    Spill containment systems include primary, secondary, and emergency containment systems.
31    Primary containment systems are the storage units, such as tanks or totes, in which fluids are
32    intentionally kept Secondary containment systems, such as liners and berms installed during site
33    set-up, are intended to contain spilled fluids until they can be cleaned up. Emergency containment
34    systems, such as berms, dikes, and booms, can be implemented temporarily in response to a spill.

35    The EPA investigated spill containment and mitigation measures in an analysis of spills related to
36    hydraulic fracturing activities  [U.S. EPA. 2015n]. Of the approximately 25% of reports thatincluded
37    information  on containment, the most common types of containment systems referenced in the
38    hydraulic fracturing-related spill records included berms, booms, dikes, liners, and pits, though

                    This document is a draft for review purposes only and does not constitute Agency policy.
      June 2015                                        5-41               DRAFT—DO NOT CITE OR QUOTE

-------
      Hydraulic Fracturing Drinking Water Assessment
 1    many of the spill reports did not indicate specific containment measures. Some spills were reported
 2    to breach the secondary containment systems. Breaches of berms and dikes were most commonly
 3    reported.

 4    In cases where secondary containment systems were not present or were inadequate, operators
 5    sometimes built emergency containment systems. The most common were berms, dikes, and
 6    booms, but there were also instances where ditches, pits, or absorbent materials were used to
 7    contain the spilled fluid. Absorbent materials were generally used when small volumes (10-200 gal
 8    or 40-800 L) of additives or chemicals were spilled [U.S. EPA. 2015n]. There was not enough
 9    information to detail the regularity of emergency containment systems or their effectiveness.

10    Remediation is the action taken to clean up a spill and its affected environmental media. The most
11    commonly reported remediation activity, mentioned in approximately half of the hydraulic
12    fracturing-related spill records evaluated by the EPA, was removal of spilled fluid and/or affected
13    media, typically soil [U.S. EPA. 2015n). Other remediation methods reported by the EPA [U.S. EPA.
14    2015n) included the use of absorbent material, vacuum trucks, flushing the affected area with
15    water, and neutralizing the spilled material.

      5.7. Overview of Chemical Spills Data
16    Spills of hydraulic fracturing fluids have occurred across the country and have affected the quality
17    of drinking water resources fU.S. EPA. 2015n: Brantley etal.. 2014: COGCC. 2014: Gradient. 20131
18    Spills may infiltrate drinking water resources by reaching surface water, or by leaching into the
19    ground water. Potential impacts depend upon a variety of factors including the chemical spilled,
20    environmental conditions, and actions taken in response to the spill. However, due to a lack of
21    available data, little is known about the prevalence and severity of actual drinking water impacts.

      5.7.1.  EPA Analysis of Spills Associated with Hydraulic Fracturing
22    The EPA f2015nl (see Text Box 5-13 for additional information) evaluated 457 spills related to
23    hydraulic fracturing activities on or near the well pad. Of these spills, 151 spills were of chemicals,
24    additives, or fracturing fluids. Information in the spill reports included: spill causes (e.g., human
25    error, equipment failure), sources (e.g., storage tank, hose or line), volumes, and environmental
26    receptors.
27
28
29
30
31
32
33
34
Text Box 5-13. EPA Review of State and Industry Spill Data: Characterization of Hydraulic
Fracturing-Related Spills.

The EPA [2015n] used data gathered from select state and industry sources to characterize hydraulic
fracturing-related spills with respect to volumes spilled, materials spilled, sources, causes, environmental
receptors, containment, and responses. For the purposes of the study, hydraulic fracturing-related spills were
denned as those occurring on or near the well pad before or during the injection of hydraulic fracturing fluids
or during the post-injection recovery of fluids. Because the main focus of this study is to identify hydraulic
fracturing-related spills on the well pad that may reach surface or ground water resources, the following
topics were not included in the scope of this project: transportation-related spills, drilling mud spills, and
spills associated with disposal through underground injection control wells.
                    This document is a draft for review purposes only and does not constitute Agency policy.
      June 2015                                        5-42               DRAFT—DO NOT CITE OR QUOTE

-------
      Hydraulic Fracturing Drinking Water Assessment
 1    Data on spills that occurred between January 2006 and April 2012 were obtained from nine states with
 2    online spill databases or other data sources, nine hydraulic fracturing service companies, and nine oil and gas
 3    production well operators. The data sources used in this study contained over 36,000 spills. The EPA
 4    searched each spill report for keywords related to hydraulic fracturing (e.g., frac, glycol, flowback). Spill
 5    records from approximately 12,000 spills (33 percent of the total number of spills reviewed) contained
 6    insufficient information to determine whether the event was related to hydraulic fracturing. Of the spills with
 7    sufficient information, the EPA identified approximately 24,000 spills (66%) as not related to hydraulic
 8    fracturing on or near the well pad. The remaining 457 spills (approximately 1%) occurred on or near the well
 9    pad and were definitively related to hydraulic fracturing. These 457 spills occurred in 11 different states over
10    six years (January 2006 and April 2 012).

11    The EPA categorized spills according to the following causes: equipment failure, human error,
12    failure of container integrity, other (e.g., well communication, weather, vandalism), and unknown.
13    Figure 5-12 presents the percent distribution of chemical or fracturing fluid spills associated with
14    each cause. Over half of the spills were collectively caused by equipment failure (34%) and human
15    error (25%). Approximately one-quarter of the spill causes were unknown or not reported. A
16    report analyzing spills in Colorado is generally consistent with the EPA's findings (COGCC. 2014).*
17    Colorado found that equipment failure was the dominant spill cause, accounting for over 60% of
18    spills between 2010 and 2013, followed by human error accounting for 20%-25% of spills.
      1 The COGCC report included all materials related to oil and gas production and were therefore not specific to chemical
      and fracturing fluid spills.
                    This document is a draft for review purposes only and does not constitute Agency policy.
      June 2015                                         5-43               DRAFT—DO NOT CITE OR QUOTE

-------
     Hydraulic Fracturing Drinking Water Assessment
                                                                     Equipment failure
                                                                     I Failure of container integrity
                                                                     i Human error
                                                                     I Unknown
                                                                     I Other
     Figure 5-12. Distribution of the causes of 151 hydraulic fracturing-related spills of chemicals
               and fracturing fluid.
               Data from U.S. EPA(2015n).

1    Spills in the EPA Spills Report were also categorized by the following sources: storage, equipment,
2    well or wellhead, hose or line, and unknown. Figure 5-13 presents the percent distribution for the
3    chemical or fracturing fluid spills associated with each source. Storage units (e.g., chemical totes,
4    fracturing fluid tanks) were the predominant source of spills, accounting for 36% (54 spills) of spill
5    sources. Spills from storage units were predominantly caused by human error (39%), followed by
6    failure of container integrity  (30%). Spills from equipment were the next most common known
7    source (18%), followed by spills from hoses or lines (13%). Twenty-eight percent of spills had an
8    unknown source.
                   This document is a draft for review purposes only and does not constitute Agency policy.
     June 2015                                         5-44               DRAFT—DO NOT CITE OR QUOTE

-------
     Hydraulic Fracturing Drinking Water Assessment
                                                                            i Equipment
                                                                            I Hose or line
                                                                            iStorage
                                                                            I Well or wellhead
                                                                            •Unknown
     Figure 5-13. Percent distribution of sources of 151 hydraulic fracturing-related spills of
               chemicals or fracturing fluid.
               Data from U.S. EPA(2015n).

1    The reported total volume of 125 of 151 chemical or hydraulic fracturing fluid spills was
2    approximately 184,000 gal (697,000 L). The volume was unknown for 26 of these spills. The spills
3    ranged in volume from 5 to more than 19,000 gal (19 to 72,000 L), with a median volume of 420 gal
4    (1,600 L). The largest source of spills was storage containers, which released approximately 83,000
5    gal (314,000 L) of spilled fluid. Spills from wells or wellheads are often associated with high spill
6    volumes. Nine instances of spills at the well or wellhead were associated with approximately
7    46,000 gal (174,000 L) of spilled fluid (see Figure 5-14). The high pressure associated with frac
8    head blowouts has led to large, high-volume spills (see Text Box 5-11).
                   This document is a draft for review purposes only and does not constitute Agency policy.
     June 2015                                         5-45               DRAFT—DO NOT CITE OR QUOTE

-------
      Hydraulic Fracturing Drinking Water Assessment
                 90,000
                 80,000
                 70,000
              T3
              = 60,000
              '5,
               in
              2 50,000
              53
              'o 40,000
               in

              S3 30,000
               co
                 20,000

                 10,000

                     0
                          Equipment     Hose or line      Storage    Well or wellhead
                                                   Source of spill
Unknown
      Figure 5-14. Total volume of fluids spilled for 151 hydraulic fracturing-related spills of
                chemicals and fracturing fluid, by spill source.
                Data from U.S. EPA(2015n).

 1    Figure 5-15 presents the number of chemical or fracturing fluid spills that reached environmental
 2    receptors, by receptor type. Environmental receptors (i.e., surface water, ground water, soil) were
 3    identified for 101 of the 151 spills, or 67% of the spills in the EPA's analysis fU.S. EPA. 2015rQ. Soil
 4    was by far the dominant environmental receptor, with 97 spills reaching soil. Thirteen spill reports
 5    indicated that the spilled fluid had reached surface water. Nine spill reports identified both soil and
 6    surface water as a receptor. No spill report identified ground water as a receptor. The data contain
 7    few post-spill analyses, so ground water contamination may have occurred but have not been
 8    identified. Additionally, several years may be required for a spilled fluid to leach into the ground
 9    water and therefore impact on a ground water receptor may not be immediately apparent Storage
10    units were the predominant sources of spills reaching an environmental receptor.
                    This document is a draft for review purposes only and does not constitute Agency policy.
      June 2015                                         5-46               DRAFT—DO NOT CITE OR QUOTE

-------
      Hydraulic Fracturing Drinking Water Assessment
             120
             100
              80
              60
              40
              20
                                                   Surface Water                Ground Water

                                           Environmental Receptor

                                           • Yes • Unknown   No
      Figure 5-15. Number of hydraulic fracturing-related spills of chemicals or fracturing fluid that
                reported whether an environmental receptor was reached.
                "Unknown" refers to hydraulic fracturing related spill events for which environmental receptors were
                specified as unknown or not identified (positively or negatively).
                Data from U.S. EPA(2015n).

 1    Six spills from storage containers reached a surface water receptor. Thirty-eight of the spills from
 2    storage units reached a soil receptor. If a spill was confined to a lined well pad, for example, it might
 3    not have reached the soil, but most incident reports did not include whether the well pad was lined
 4    or unlined. Regarding spills of hydraulic fluids and chemicals from storage containers, 16 spills
 5    were due to failure of container integrity, which includes holes and cracks in containers, and
 6    overflowing containers as a result of human error or equipment malfunctions.

 7    The EPA analysis demonstrates that spills of chemicals, additives, and fracturing fluids do occur at
 8    well sites and reach both soil and surface water receptors.

      5.7.2.  Other Spill Reports
 9    Surface spills related to hydraulic fracturing activities are not well documented in the scientific
10    literature, though some evidence of spills and impacts to environmental media exists (e.g., U.S. EPA.
                    This document is a draft for review purposes only and does not constitute Agency policy.
      June 2015                                         5-47               DRAFT—DO NOT CITE OR QUOTE

-------
      Hydraulic Fracturing Drinking Water Assessment
 1    20151: Brantley etal.. 2014: Gross etal.. 2013: Papoulias and Velasco. 20131. For example,
 2    Papoulias and Velasco [2013] demonstrated that fracturing fluid spilled into surface water likely
 3    contributed to the distress and deaths of the threatened blackside dace fish in Kentucky. A variety
 4    of chemicals entered the creek and significantly reduced the stream's pH and increased stream
 5    conductivity. Using data from post-spill sampling reports in Colorado, Gross etal. [2013] identified
 6    concentrations of benzene, toluene, ethylbenzene, and xylene in ground water samples, which the
 7    authors attributed to numerous hydraulic fracturing-related spills. The COGCC [2014] published a
 8    report analyzing all spills reported to the state of Colorado between 2010 and 2013, and found that
 9    approximately 8% of them were related to hydraulic fracturing. Based on the EPA's analyses [U.S.
10    EPA, 2015n] and available scientific data, we estimate spill rates of chemicals and hydraulic
11    fracturing fluid range from 0.4 and 12.2 spills for every 100 wells. (See Text Box 5-14 for additional
12    information.]


      Text Box 5-14. Spill Rates.

13    Several studies have estimated the frequency of hydraulic fracturing-related spills. Three studies [Rahm et al..
14    2015: Brantley et al.. 2014: Gradient. 2013] calculated a spill rate for the Marcellus Shale in Pennsylvania
15    using reports from the Pennsylvania Department of Environmental Protection (PA DEP] Oil and Gas
16    Compliance Report Database, and here we estimate an on-site spill rate for Colorado. The PA DEP database
17    provides a searchable format based on Notices of Violations from routine inspections or investigations of spill
18    reports or complaints. Each study had different criteria for inclusion, presented in Table 5-6, resulting in a
19    range of rates even when using the same data source. Spills of hydraulic fracturing fluids, flowback/produced
20    water are estimated to occur at a rate of 0.4 per 100 wells fractured. Spills related to hydraulic fracturing
21    activity are estimated to occur at a rate between 3.3 to 12.2 spills per 100 wells installed (PA DEP data] (see
22    Table 5-6].

23    In its study of spills related to hydraulic fracturing, the EPA determined that spill reports from the Colorado
24    Oil and Gas Conservation Commission (COGCC] Information System were the most detailed spill reports from
25    among the nine state data sources investigated and generally provided more of the information needed to
26    determine whether a spill was related to  hydraulic fracturing (U.S. EPA. 2015n]. Here, we estimate the  spill
27    rate in Colorado by dividing the number of hydraulic fracturing-related spills identified by the EPA (U.S. EPA.
28    2015n. Appendix B](Appendix B in U.S. EPA, 2015b] by the number of wells hydraulically fractured in
29    Colorado for specific time periods between January 2006 and April 2012. We used three data sources to
30    estimate the number of wells: (1] there were 172 reported spills in Colorado for the 15,000 wells fractured
31    from January 2006 to April 2012 (Drillmglnfo.2012]. (2] there were 50 reported spills in Colorado for the
32    3,559 wells fractured from January 2011 to April 2012 (U.S. EPA. 2015b]. and (3] there were 41 reported
33    spills in Colorado for the 3,000 wells fractured from September 2009 to October 2010 (U.S. EPA. 2013a].
34    From these data we estimate an average of 1.3  reported spills on or near the well pad for every 100
3 5    hydraulically fractured wells.
36
                    This document is a draft for review purposes only and does not constitute Agency policy.
      June 2015                                         5-48               DRAFT—DO NOT CITE OR QUOTE

-------
Hydraulic Fracturing Drinking Water Assessment
Table 5-6. Estimations of spill rates.
            Spill rates from four different sources. Each source used different criteria to identify and include spills
            in their analysis.
Spill rate3
0.4
3.3
12.2
1.3
Data source
PADEPb
Media
PADEPb
PADEPb
COGCCC
Time period
2008-2013
2009-2012
2007-July 2013
Jan 2006-
May 2012
Inclusion criteria
Volume spilled
> 400 gal;
Spill reported to reach
water body.
"Unconventional"
well;
Spills with unknown
volumes not included,
Includes any spill
during HF activities
"Unconventional" well
based on
environmental
violation rates.
Specifically related to
hydraulic fracturing on
or near well pad
Information source
Brantley et al. (2014)d

Gradient (2013)e
Rahm et al. (2015)e

U.S. EPA(2013a)d

aSpill rate is the number of reported spills per 100 wells.
 PA DEP (http://www.depreportingservices.state.pa.us/ReportServer/Pages/ReportViewer.aspx7/Oil Gas/
OG Compliance)
CCOGCC (https://cogcc.state.co. us/cogis/lncidentSearch.asp)
 Spill rate is calculated as the number of spills per 100 wells fractured.
eSpill rate is calculated as the number of spills per 100 wells installed.
                 This document is a draft for review purposes only and does not constitute Agency policy.
June 2015                                                 5-49                  DRAFT—DO NOT CITE OR QUOTE

-------
      Hydraulic Fracturing Drinking Water Assessment
      Text Box 5-14 (Spill Rates), continued:
 1    The spill rates presented in Table 5-6 are based on spill reports found in two state data sources and are
 2    limited by both the spills reported in the state data sources and the inclusion criteria defined by each of the
 3    studies. Spills identified from state data sources are likely a subset of the total number of spills that occurred
 4    within a state for a specified time period. Some spills may not be recorded in state data sources because they
 5    do not meet the spill reporting requirements in place at the time of the spill. Additionally, the PA DEP Notices
 6    of Violation may include spills not specifically related to hydraulic fracturing, such as spills of drilling fluids.
 7    The inclusion criteria used by each of the studies affects which spills are used to calculate a spill rate. More
 8    restrictive criteria, such as only counting spills that were greater than 400 gallons, results in a lower number
 9    of spills being used for estimating spill rates, while less restrictive criteria, such as all spills from wells
10    marked unconventional in the PA DEP database, results in a greater number of spills being used for
11    estimating spill rates. Rahm etal. applied the least restrictive criteria of the four studies (i.e., spills from
12    unconventional wells) when identifying spills, while Brantley et al. applied more restrictive criteria (i.e., spills
13    of >400 gallons in which spilled fluids reached a surface water body). This may account for the different spill
14    rates calculated by these two studies.
15    Based on previous studies and the analysis here, hydraulic fracturing-related spills rates in Pennsylvania and
16    Colorado range from 0.4 and 12.2 per 100 wells. These numbers may not be representative of national spill
17    rates or rates in other regions. If this range is applied nationally however, assuming between 25,000 and
18    30,000 wells are fractured each year, we would expect between approximately 100 and 3,700 spills annually
19    from hydraulic fracturing.


      5.8.  Fate and Transport of Chemicals
2 0    This section provides an overview of fate and transport of the range of chemicals used in hydraulic
21    fracturing fluid, including the physicochemical properties of these chemicals, and an overview of
2 2    the potential for a spilled chemical to affect drinking water resources. A general overview of the
23    processes governing the fate and transport of a chemical spill is shown in Figure 5-16. A chemical
24    spill has the potential to migrate to and have  an impact on drinking water resources. Once spilled,
25    there are different paths that chemicals can travel and different processes they may undergo.
26    Chemicals may react and transform into other chemicals, volatilize, travel to surface water, leach
27    into and partition to soils, and/or reach ground water. The potential path and the  severity of the
28    impact of a spill depend on different factors, including the site conditions; environmental
29    conditions;  climate; weather;  and chemical properties, concentration, and volume of the release.
30    The point in the chemical mixing process where the spill occurs affects potential impact If the spill
31    occurs before chemicals are mixed into the base fluid, the chemicals will spill in their most
32    concentrated form. If the hydraulic fracturing fluid spills, then the chemicals will be more diluted,
33    and there may be effects on persistence and mobility due to  interactions among chemicals present
                    This document is a draft for review purposes only and does not constitute Agency policy.
      June 2015                                          5-50                DRAFT—DO NOT CITE OR QUOTE

-------
      Hydraulic Fracturing Drinking Water Assessment
                                                              Schematic of the Fate and Transport Processes
                                                                   Governing Potential Impacts of Spills to
                                                                              Drinking Water Resources
      Figure 5-16. Fate and transport schematic for a spilled hydraulic fracturing fluid.
                Schematic shows the potential paths and governing processes that spilled chemicals, which may lead
                to potential impacts to drinking water resources.

 1    For inorganic chemicals, the properties and processes governing fate and transport depend on pH,
 2    oxidation state, presence of iron oxides, soil organic matter, cation exchange capacity, and major ion
 3    chemistry [U.S. EPA. 1996).* Transport of these chemicals into ground water depends on the nature
 4    of ground water flow and flow through the unsaturated zone above the water table. Potential
 5    transformations of inorganic chemicals differ from those of organic chemicals.2 Some inorganic
 6    anions (i.e., nitrate, chloride, and bromide) move with their carrier liquid and are affected mostly by
 7    physical transport mechanisms. For many inorganic chemicals, transport is driven by the physical
 8    flow processes (advection and dispersion), sorption, and precipitation. The relative role of each of
 9    these depends on both chemical and environmental characteristics.3'4

10    Determining the fate and transport of organic chemicals and mixtures is a complex problem,
11    because of the many processes and different environmental media (air, soil, water) that can have an
      1 Cation exchange capacity is the total amount of cations (positively charged ions) that a soil can hold. For example, when
      metal ions like Ca2+ and Na+ pass through the soil, they adhere and remain attached to the soil.
      2 The unsaturated zone is also referred to as the vadose zone. Meaning "dry," the vadose zone is the soil zone above the
      water table that is only partially filled by water, hence "unsaturated zone."
      3 Advection is a mechanism for moving chemicals in flowing water, where a chemical moves along with the flow of the
      water itself.
      4 Sorption is the general term used to describe the partitioning of a chemical between soil and water and depends on the
      nature of the solids and the properties of the chemical.
                     This document is a draft for review purposes only and does not constitute Agency policy.
      June 2015                                          5-51               DRAFT—DO NOT CITE OR QUOTE

-------
      Hydraulic Fracturing Drinking Water Assessment
 1    impact Unlike inorganic chemicals, organic chemicals degrade, which may affect their movement
 2    and potential impact. Schwarzenbach etal. [2002] formalized a general framework for organic
 3    chemical transport, where transport and transformation depend on both the nature of the chemical
 4    and the properties of the environment. The fate and transport of organic compounds in soils has
 5    been presented in the literature (e.g., Bouchard etal., 2011: Rivettetal., 2011: Abriola and Finder,
 6    1985a. b] and in textbooks [e.g.. Domenico and Schwartz. 1997: Schnoor. 1996: Freeze and Cherry,
 7    19791.

      5.8.1.  Potential Paths
 8    Chemicals, additives, and hydraulic fracturing fluids that are released into the environment travel
 9    along different potential paths, as detailed in Figure 5-16. Liquids may flow overland to nearby
10    surface water or infiltrate the subsurface, where they may eventually reach the underlying ground
11    water or travel laterally to reach surface water. Movement can occur quickly or be delayed and
12    have a later or longer-term impact. Surface water and ground water gain or lose flow to each other,
13    and may transport chemicals in the process.

14    A dry chemical (e.g., gelling agents, biocides, friction reducers) released to the environment will
15    generally remain where it is  spilled. Any spill that is not removed could act as a long-term source of
16    contamination. Wind could cause the chemical to disperse, or rain could dissolve a soluble chemical.
17    Dissolved chemicals may infiltrate into soil or flow overland.  Insoluble chemicals and those sorbed
18    to soil particles could be mobilized by rain events via runoff and erosion.

      5.8.1.1. Movement across  the Land Surface
19    In low permeability soils, there may be little infiltration and greater overland flow. Higher
20    permeability soils will allow fluid to penetrate into the soil layer. In either case, some of the
21    chemicals in the fluid may sorb to the soil particles and the vegetation, and then these chemicals
22    may be mobilized during rainfall, runoff, or erosion. As rainwater percolates through the soil, it may
23    dissolve stored chemicals, which can then migrate toward ground water.

24    The type of release  is also important. If the spill is a slow leak, then the liquid may pond and the
25    affected area will expand slowly. If a more rapid release occurs, like a blowout or tank failure, then
26    momentum may result in greater overland movement and less soil infiltration during the event

      5.8.1.2. Movement through the Subsurface
27    The unsaturated and saturated zones are the two zones of soils below the ground surface.
28    Movement through the unsaturated zone is driven by the depth of ponding of the spilled  fluid,
29    gravity, and capillary properties of the subsurface.1 In fractured rock or highly permeable soils,
3 0    fluids may move quickly through the subsurface. In low permeability soil, the movement of the fluid
31    is slower. As chemicals pass  through the subsurface, some may sorb to soil or remain in the open
3 2    spaces between soil particles, effectively slowing their movement. Chemicals may again be
      1 Capillarity occurs because of the forces of attraction of water molecules to themselves (cohesion) and to other solid
      substances such as soils (adhesion).
                   This document is a draft for review purposes only and does not constitute Agency policy.
      June 2015                                       5-52              DRAFT—DO NOT CITE OR QUOTE

-------
      Hydraulic Fracturing Drinking Water Assessment
 1    mobilized during future precipitation events, resulting in infiltration towards ground water or
 2    movement through the unsaturated zone towards surface water.

 3    Fluids that move through the subsurface into the saturated zone will move in the direction of the
 4    flowing ground water. Generally, fluids travel further in systems with high ground water flow rates
 5    and high recharge (e.g., sandy aquifers in humid climates) than in systems with low flow and low
 6    recharge. Chemicals may sorb to suspended soil particles or complex with naturally occurring
 7    chemicals (e.g., dissolved organic carbon) and be transported with the flowing water.1 These
 8    mechanisms can mobilize sparingly soluble chemicals that would otherwise be immobile.

      5.8.2. Physicochemical Properties
 9    Three physicochemical properties are useful to describe the movement of chemicals in the
10    environment: (1) Kow, the octanol-water partition coefficient, (2) the aqueous solubility,  and (3) the
11    Henry's law constant.2 These properties describe whether a chemical will sorb to soil and organic
12    matter or stay in water (Kow), how much of a chemical may dissolve in water (aqueous solubility),
13    and whether a chemical will tend to remain in the water or volatilize (Henry's law constant).3

14    The Kow measures the relative hydrophobicity (chemical prefers to be in oil, log Kow >0) and
15    hydrophilicity (chemical prefers to be in water, log Kow <0) of a chemical. Aqueous solubility is the
16    maximum amount of a chemical that will dissolve in water in the presence of a pure chemical;
17    solubility generally serves as an upper bound on possible concentrations. The Henry's law constant
18    is the ratio of the  concentration of a chemical in air (or vapor pressure) to the concentration of that
19    chemical in water.

20    Estimates and measured values for physicochemical properties were obtained by using the
21    Estimation Program Interface Suite 4.1 (see Appendix C).4 Of the 1,076 chemicals the EPA listed as
22    used in hydraulic fracturing (see Appendix A), EPI Suite™ has estimated properties for 453 (42%).
23    EPI Suite™ does not have data available for the remaining 58% of the chemicals. The 453 chemicals
24    with physicochemical property data were chemicals with structures that are considered suitably
25    representative of the substance to compute properties within the constraints of EPI Suite™
26    software. Only unique defined organic desalted structures were submitted for property  calculation.
27    Figure 5-17  presents histograms of all 453 of the chemicals, sorted by four physicochemical
28    parameters: measured log Kow (n = 247, 23%), estimated log ftow(n=45 3, 42%) estimated log of the
      1 Complexation is a reaction between two chemicals that form a new complex, either through covalent bonding or ionic
      forces. This often results in one chemical solubilizingthe other.
      2 The octanol-water partition coefficient (Kow) represents the ratio of the solubility of a compound in octanol (a nonpolar
      solvent) to its solubility in water (a polar solvent) in a mixture of the two. The higher the Kow, the more nonpolar the
      compound.
      3 We present the physicochemical parameter values using logio because of the wide range of values that these parameters
      cover.
      4 EPI Suite™, version 4.1, http://www.epa.gov/opptintr/exposure/pubs/episuite.htm [U.S. EPA. 2012b). The EPI
      (Estimation Programs Interface) Suite™ is a Windows®-based suite of physicochemical property and environmental fate
      estimation programs developed by the EPA Office of Pollution Prevention and Toxics and Syracuse Research Corporation.
      EPI Suite™ provides estimates of physicochemical properties for organic chemicals and has a database of measured values
      for physicochemical properties when available. EPI Suite™ cannot estimate parameters for inorganic chemicals.
                    This document is a draft for review purposes only and does not constitute Agency policy.
      June 2015                                         5-53               DRAFT—DO NOT CITE OR QUOTE

-------
      Hydraulic Fracturing Drinking Water Assessment
 1    aqueous solubility (n = 453, 42%), and estimated log of Henry's law constant (at 25°C, n = 453,
 2    42%).
                  , Most mobile
                                            Least nobj|e
50

40

30

20

10
                -10  -8   -6  -4   -2   0   2   4   6   8  10
                            Measured log KOB,|L/kg]
                                                           LOO
                                                              _. Most mobile
                                                                                       Least mobile  .
                                                           80
                                                         £ 40
                                                         I
                                                           20
                                            0 -         i
                                            -25  -20  -15  -10  -5   0   5   10  15   20  25
                                                        Estimated log X,,,,, [L/kg]
                   Least soluble
                                            Most soluble
                                                           I Mi
                                                               Stays in water
                                                                                  Readily escapes to air
                 -20  -15   -10  -5   0   5   10   15   20
                        Estimated log Solubility [mg/L @ 25C]
                                             -50  -40  -30  -20  -10  -0   10  20  30  40  50
                                              Estimated log Henry's Law Constant [atm mj mole'1 @ 25C]
      Figure 5-17. Histograms of physicochemical properties of chemicals used in the hydraulic
                 fracturing process.
                 Measured values of log Kow (upper left). Estimated physicochemical properties for log Kow (upper
                 right), log Solubility (lower left), and log Henry's law constant (lower right) for all chemicals.
                 Physicochemical properties (log Kow, solubility, and Henry's Law constant) estimated by EPI Suite™.

 3    The EPA also used EPI Suite™ to determine the physicochemical properties for 19 chemicals
 4    provided to the EPA as confidential business information (CBI) (See Text Box 5-3 for discussion on
 5    CBI).1 The CBI chemical physicochemical properties are plotted as histograms in Figure 5-18. The
 6    values of the physicochemical properties of known and CBI chemicals are similar, covering similar
 7    ranges centered on similar values, suggesting that even though these chemicals are not publicly
 8    known, their physicochemical properties are not appreciably different from the known chemicals.
 9    This means that their fate and transport would not be appreciably different than the chemicals that
10    are publicly known.
      1 Well operators may specify certain ingredients as confidential business information (CBI) and not disclose the chemicals
      used to FracFocus. The CASRNs of a range of CBI chemicals were provided to USEPAby 9 service companies.
                     This document is a draft for review purposes only and does not constitute Agency policy.
      June 2015                                           5-54                DRAFT—DO NOT CITE OR QUOTE

-------
      Hydraulic Fracturing Drinking Water Assessment
               3.5
                3
             S'
             ii 2.5
             c
             I  2
             I 1.5
             *  1
               0.5
                 . Most mobile
                                          Least mobile ,
                                                                                    Least mobile
I    I
                         J
                           _  3
                           T  2
                           I-
                         -505
                           Measured log K0[v[L/kg]
   -.
   7 '
_  6
oi
7  5
c
T  4
A
E  3
2  2
   I
                  Least Soluble
                -20   -15   -10   -5    0    5    10   15   20
                       Estimated log Solubility [mg/L  2SC]
                                      	Most Soluble ^          Stays in water	

                                      J          '!      iL
                              °25  -20  -15 -10  -5   0   5  10   15  20
                                        Estimated log Kolv[L/kg]
                                                    Readily escapes to air
                              -50  -40  -30  -20 -10  0   10  20  30  40  50
                               Estimated log Henry's Law Constant [atm m* mole -1 @ 25C]
      Figure 5-18. Histograms of physicochemical properties of confidential chemicals used in the
                hydraulic fracturing process.
                Source: (U.S. EPA, 2013a)
                Measured values of log Kow (upper left). Estimated physicochemical properties for log Kow (upper
                right), log solubility (lower left), and log Henry's law constant (lower right) for all chemicals.
                Physicochemical properties (log Kow, solubility, and Henry's Law constant) estimated by EPI Suite™.

      5.8.3.  Mobility of Chemicals
 1    Figure 5-17 shows the distribution of the three properties. The log Kow distribution demonstrates
 2    that the chemicals cover a wide range from the more mobile to the less mobile. The more
 3    hydrophilic chemicals are more mobile (i.e., they move with water). The more hydrophobic
 4    chemicals tend to associate with organic matter and soil particles and to be less  mobile in the
 5    environment, and they may serve as long-term sources of contamination. A large number of the
 6    chemicals fall near log Kow = 0, which indicates that these chemicals are likely to  associate roughly
 7    equally with organic or aqueous phases. However, overall the log Kow values are  skewed positively,
 8    indicating less mobile chemicals,  which may result in their being later-term or long-term sources of
 9    impact to drinking water. The log S values span a wide range from fully miscible to sparingly
10    soluble. Many of the chemicals have high aqueous solubilities, with a large number being fully
11    miscible. Most of the chemicals have log Henry's law constants less than 0, indicating that most are
12    not volatile. Once these chemicals dissolve into water they will tend to stay there rather than
13    volatilize. Therefore, volatilization does not generally serve as a loss process for most hydraulic
14    fracturing chemicals.

15    Table 5-7 and Table 5-8 present the 20 most mobile and least mobile chemicals, known to be used
16    in hydraulic fracturing fluids, respectively, as ranked by log Kow. These were taken from the list of
                    This document is a draft for review purposes only and does not constitute Agency policy.
      June 2015                                         5-55                DRAFT—DO NOT CITE OR QUOTE

-------
      Hydraulic Fracturing Drinking Water Assessment
 1    515 chemicals with estimated values for physicochemical properties. These tables also include
 2    values for aqueous solubility and Henry's law constant, as well as frequency of use, based on
 3    FracFocus disclosures fU.S. EPA. 2015a).

 4    Table 5-7 shows the chemicals that have the lowest log Kow and are, thus, the most mobile. These
 5    chemicals are fully miscible (i.e., they will mix completely with water), which means they may move
 6    through the environment at high concentrations, leading to greater severity of impact These
 7    chemicals generally have low volatility, based on their negative log Henry's law constants (i.e., will
 8    remain in water and will not be lost to the air). These chemicals will dissolve in water and move
 9    rapidly through the environment (e.g., via infiltration into the subsurface or via overland flow to
10    surface waters). Chemicals exhibiting this combination of properties have greater potential to cause
11    immediate impacts to drinking water resources. Most of the chemicals in the table were
12    infrequently reported (<2% of wells) in FracFocus (U.S. EPA. 2015a). However, choline chloride
13    (14% of wells), used for clay control, and tetrakis(hydroxymethyl)phosphonium sulfate (11% of
14    wells), a biocide, were more commonly reported.
      Table 5-7. Ranking of the 20 most mobile organic chemicals, as determined by the largest log
               Kow, with CASRN, percent of wells where the chemical is reported from January 1,
               2011 to February 28, 2013 (U.S. EPA, 2015b), and physicochemical properties (log
               Kow, solubility, and Henry's Law constant) as estimated by EPI Suite™.
               (U.S. EPA, 2015b)
               For organic salts, parameters are estimated using the desalted form.




Rank

1


2



3






Chemical Name
1,2-Ethanediaminium, N,N'-bis[2-
[bis(2-hydroxyethyl)methylammo
nio]ethyl]-N,N'-bis(2-hydroxyeth
yl)-N,N'-dimethyl-, tetrachloride
Phosphonic acid,
[[(phosphonomethyl)imino]bis
[2,1-ethanediylnitrilobis
(methylene)]]tetrakis-
Phosphonicacid,
[[(phosphonomethyl)imino]bis
[2,1-ethanediylnitrilobis
(methylene)]]tetrakis-, sodium
salt




CASRN

138879-94-4


15827-60-8



22042-96-2



Percent of
wells
(U.S. EPA,
2015b)a

2%


0.2%



0.07%




Estimated
Log Kow
(unitless)b

-23.19


-9.72



-9.72


Estimated
Water
Solubility
(mg/L@
25°C)C

1.00 x 10s


1.00 x 10s



1.00 x 10s


Estimated
Henry's Law
Constant
(atm mVmole
@ 25°C)d

2.33 x 10"35


NA



NA


                   This document is a draft for review purposes only and does not constitute Agency policy.
      June 2015                                       5-56               DRAFT—DO NOT CITE OR QUOTE

-------
Hydraulic Fracturing Drinking Water Assessment
Rank
4
5
6
7
8
9
10
11
12
13
14
15
16
Chemical Name
Phosphonic acid,
[[(phosphonomethyl)imino]bis
[2,1-ethanediylnitrilobis
(methylene)]]tetrakis-,
ammonium salt (l:x)
Phosphonic acid, (((2-[(2-
hydroxyethyl)(phosphonomethyl)
amino)ethyl)imino]bis(methylene
))bis-, compd. with 2-
aminoethanol
2-Hydroxy-N,N-bis(2-
hydroxyethyl)-N-
methylethanaminium chloride
N-(3-Chloroallyl)hexaminium
chloride
3,5,7-Triazatricyclo(3.3.1.1
(superscript 3,7))decane, l-(3-
chloro-2-propenyl)-, chloride, (Z)-
(2,3-dihydroxypropyl)
trimethylammonium chloride
Phosphonic acid,
[[(phosphonomethyl)imino]bis
[6,1-hexanediylnitrilobis
(methylene)]]tetrakis-
[Nitrilotris(methylene)]tris-
phosphonic acid pentasodium salt
Aminotrimethylene phosphonic
acid
Choline chloride
Choline bicarbonate
alpha-Lactose monohydrate
Lactose
CASRN
70714-66-8
129828-36-0
7006-59-9
4080-31-3
51229-78-8
34004-36-9
34690-00-1
2235-43-0
6419-19-8
67-48-1
78-73-9
5989-81-1
63-42-3
Percent of
wells
(U.S. EPA,
2015b)a
NA
NA
NA
0.02%
NA
NA
0.006%
0.5%
2%
14%
0.2%
NA
NA
Estimated
Log Kow
(unitless)b
-9.72
-6.73
-6.7
-5.92
-5.92
-5.8
-5.79
-5.45
-5.45
-5.16
-5.16
-5.12
-5.12
Estimated
Water
Solubility
(mg/L@
25°C)C
1.00 x 10s
1.00 x 10s
1.00 x 10s
1.00 x 10s
1.00 x 10s
1.00 x 10s
1.00 x 10s
1.00 x 10s
1.00 x 10s
1.00 x 10s
1.00 x 10s
1.00 x 10s
1.00 x 10s
Estimated
Henry's Law
Constant
(atm ms/mole
@ 25°C)d
NA
5.29 x 10"42
4.78 x 10"19
1.76 x 10"8
1.76 x 10"8
9.84 x 10"18
NA
1.65 x 10"34
1.65 x 10"34
2.03 x 10"16
2.03 x 10"16
4.47 x 10"22
4.47 x 10"22
                This document is a draft for review purposes only and does not constitute Agency policy.
June 2015                                              5-57                 DRAFT—DO NOT CITE OR QUOTE

-------
      Hydraulic Fracturing Drinking Water Assessment
Rank
17
18
19
20
Chemical Name
Tetrakis(hydroxymethyl)phospho
nium sulfate
Disodium
ethylenediaminediacetate
Nitrilotriacetamide
l,3,5-Triazine-l,3,5(2H,4H,6H)-
triethanol
CASRN
55566-30-8
38011-25-5
4862-18-4
4719-04-4
Percent of
wells
(U.S. EPA,
2015b)a
11%
0.6%
NA
0.2%
Estimated
Log Kow
(unitless)b
-5.03
-4.79
-4.75
-4.67
Estimated
Water
Solubility
(mg/L@
25°C)C
1.00 x 10s
1.00 x 10s
1.00 x 10s
1.00 x 10s
Estimated
Henry's Law
Constant
(atm mVmole
@ 25°C)d
9.17 x 10"13
l.lOxlO"16
1.61 x 10"18
1.08 x 10"11
       a Some of the chemicals in these tables have NA (not available) listed as the number of wells, which means that these
       chemicals have been used in hydraulic fracturing, but they were not reported to FracFocus program for the time period of the
       study (January 1, 2011, to February 28, 2013) (U.S. EPA. 2015b). Analysis considered 34,675 disclosures and 676,376
       ingredient records that met selected quality assurance criteria, including: completely parsed; unique combination of fracture
       date and API well number; fracture date between January 1, 2011, and February 28, 2013; valid CASRN; and valid
       concentrations. Disclosures that did not meet quality assurance criteria (3,855) or other, query-specific criteria were excluded
       from analysis.
       b Log Kow is estimated using the KOWWIN™ model, which uses an atom/fragment contribution method.
       c Water solubility is estimated using the WSKOWWIN™ model, which estimates a chemical's solubility from Kow and any
       applicable correction factors.
        Henry's Law constant is estimated using the HENRYWIN™ model using the bond contribution method.
 1    Table 5-8 shows the chemicals that have the highest log Kow and are, thus, the least mobile. The
 2    estimated aqueous solubilities for some of these chemicals are extremely low, with highest
 3    solubilities of <10  [ig/L. Therefore, the concentration of these chemicals dissolved in water will be
 4    low. The estimated Henry's law constants are more variable for these low-mobility chemicals.
 5    Chemicals with high log Kow values (>0)  and high Henry's law constants will sorb strongly to organic
 6    phases and solids and may volatilize. However, their strong preference for the organic or solid
 7    phase may slow or reduce volatilization. The chemicals with low Henry's law constants will readily
 8    sorb to organic phases and solids. Less mobile chemicals will move slowly through the soil and have
 9    potentially delayed and longer-term impacts to drinking water resources. Seven of the chemicals in
10    Table 5-8 were reported to FracFocus [U.S. EPA. 2015b]. Five were reported infrequently (<1% of
11    wells). Tri-n-butyltetradecylphosphonium chloride (6% of wells), used as a biocide, and
12    OlO-alpha-alkenes (8% of wells), a mixture of alpha-olefins with carbon numbers greater than 10
13    used as a corrosion inhibitor, were more commonly reported. The least mobile organic chemical is
                     This document is a draft for review purposes only and does not constitute Agency policy.
      June 2015                                           5-58               DRAFT—DO NOT CITE OR QUOTE

-------
     Hydraulic Fracturing Drinking Water Assessment
1    sorbitan, tri-(9Z)-9-octadecenoate, a mineral oil co-emulsifier (0.05% of wells), with an estimated
2    log Kow of 22.56.1
     Table 5-8. Ranking of the 20 least mobile organic chemicals, as determined by the largest log
               KOW, with CASRN, percent of wells where the chemical is reported from January 1,
               2011 to February 28, 2013 (U.S. EPA, 2015b), and physicochemical properties (log
               Kow, solubility, and Henry's Law constant) as estimated by EPI Suite™.
               Source: (U.S. EPA, 2015b)
               For organic salts, parameters are estimated using the desalted form.
Rank
1
2
3
4
5
6
7
8
9
10
11
12
Chemical Name
Sorbitan, tri-(9Z)-
9 octadecenoate
Fatty acids, C18-unsatd.,
dimers
Sorbitan sesquioleate
Tri-n-butyltetradecyl-
phosphonium chloride
Sodium bis(tridecyl)
sulfobutanedioate
1-Eicosene
D&C Red 28
C.I. Solvent Red 26
1-Octadecene
Alkenes, C>10 alpha-
Dioctyl phthalate
Benzene, C10-16-alkyl derivs.
CASRN
26266-58-0
61788-89-4
8007-43-0
81741-28-8
2673-22-5
3452-07-1
18472-87-2
4477-79-6
112-88-9
64743-02-8
117-84-0
68648-87-3
Percent of
wells
(U.S. EPA,
2015b)a
0.05%
NA
0.02%
6%
NA
NA
NA
NA
NA
8%
NA
0.5%
Estimated
Log Kow
(unitless)b
22.56
14.6
14.32
11.22
11.15
10.03
9.62
9.27
9.04
8.55
8.54
8.43
Estimated
Water
Solubility
(mg/L@
25°C)C
1.12 x 10"19
2.31 x 10'10
2.31 x 10"11
7.90 x 10"7
7.46 x 10"9
1.26 x 10"5
1.64 x 10"8
5.68 x 10"5
1.256 x 10"4
3.941 x 10"4
4.236 x 10"4
2.099 x 10"4
Estimated
Henry's Law
Constant
(atm m /
mole @
25°C)d
4.02 x 10"11
4.12xlO"08
7.55 x 10"12
2.61 x 10"1
8.51 x 10"11
1.89 x 101
6.37 x 10"21
5.48 x 10"13
1.07 x 101
8.09 x 10°
1.18xlO"5
1.78 x 10"1
     1 Sorbitan, tri-(9Z)-9-octadecenoate, CASRN 26266-58-0, is soluble in hydrocarbons and insoluble in water, listed as an
     effective coupling agent and co-emulsifier for mineral oil [Santa Cruz Biotechnology. 2015: ChemicalBook. 2010).
                   This document is a draft for review purposes only and does not constitute Agency policy.
     June 2015                                         5-59               DRAFT—DO NOT CITE OR QUOTE

-------
     Hydraulic Fracturing Drinking Water Assessment
Rank
13
14
15
16
17
18
19
20
Chemical Name
Di(2-ethylhexyl) phthalate
1-Octadecanamine, N,N-
dimethyl-
N,N-dimethyloctadecylamine
hydrochloride
Butyryl trihexyl citrate
1-Hexadecene
Benzo(g,h,i)perylene
Dodecylbenzene
Isopropanolamine
dodecylbenzene
CASRN
117-81-7
124-28-7
1613-17-8
82469-79-2
629-73-2
191-24-2
123-01-3
42504-46-1
Percent of
wells
(U.S. EPA,
2015b)a
NA
NA
NA
0.03%
NA
NA
NA
0.02%
Estimated
Log Kow
(unitless)b
8.39
8.39
8.39
8.21
8.06
7.98
7.94
7.94
Estimated
Water
Solubility
(mg/L@
25°C)C
1.132 x 10"3
8.882 x 10"3
8.882 x 10"3
5.56 x 10"5
1.232 x 10"3
7.321 x 10"4
1.015 x 10"3
1.015 x 10"3
Estimated
Henry's Law
Constant
(atm m3/
mole @
25°C)d
1.18xlO"5
4.51 x 10"3
4.51 x 10"3
3.65 x 10"9
6.10x10°
1.26 x 10"2
1.34 x 10"1
1.34 x 10"1
      a Some of the chemicals in these tables have NA (not available) listed as the number of wells, which means that these
      chemicals have been used in hydraulic fracturing, but they were not reported to FracFocus program for the time period of
      the study (January 1, 2011, to February 28, 2013) (U.S. EPA, 2015b). Analysis considered 34,675 disclosures and 676,376
      ingredient records that met selected quality assurance criteria, including: completely parsed; unique combination of fracture
      date and API well number; fracture date between January 1, 2011, and February 28, 2013; valid CASRN; and valid
      concentrations. Disclosures that did not meet quality assurance criteria  (3,855) or other, query-specific criteria were
      excluded from analysis.
      b Log Kow is estimated using the KOWWIN™ model, which uses an atom/fragment contribution method.
      c Water solubility is estimated using the WSKOWWIN™ model, which estimates a chemical's solubility from Kow and any
      applicable correction factors.
       Henry's Law constant is estimated using the HENRYWIN™ model using the bond contribution method.
1    Table 5-9 shows the EPI Suite™ estimated physicochemical property values of the 20 chemicals
2    most frequently reported to FracFocus nationwide, with estimated mean and median volumes
3    based on FracFocus data. Most have log Kow < 1, meaning that they are generally hydrophilic and
4    will associate with water. These chemicals also have very high solubilities, so they will be mobile in
5    the environment and go where the water goes. These chemicals have the potential for immediate
6    impacts to drinking water resources. Naphthalene has a measured log Kow = 3.3 with an estimated
7    solubility of 142.1 mg/L, which means it will be less mobile in the environment. Naphthalene will
8    sorb to particles and move slowly through the environment, and have the potential to act as long-
                    This document is a draft for review purposes only and does not constitute Agency policy.
     June 2015                                            5-60                DRAFT—DO NOT CITE OR QUOTE

-------
     Hydraulic Fracturing Drinking Water Assessment
1    term sources of contamination.1 All of these chemicals have low Henry's law constants, so they tend
2    not to volatilize.
     1 Chemicals may have the potential to be long-term sources of contamination because they move slowly through the
     environment. In this discussion, we are not accounting for biodegradation or other transformation processes, which may
     reduce the persistence of certain chemicals in the environment. Under the right conditions, for example, naphthalene is
     biodegradable, which may reduce or remove it from the environment, and thus may not be a long-term source of
     contamination.
                     This document is a draft for review purposes only and does not constitute Agency policy.

     June 2015                                             5-61                 DRAFT—DO NOT CITE OR QUOTE

-------
Hydraulic Fracturing Drinking Water Assessment
                                                                           Chapter 5 - Chemical Mixing
Table 5-9. The 20 chemicals reported most frequently nationwide for hydraulic fracturing based on reported FracFocus 1.0
          disclosures (U.S. EPA, 2015b), with EPI Suite™ physicochemical parameters where available, and estimated mean and
          median volumes of those chemicals, where density was available.

          Source: (U.S. EPA, 2015b)
          Excludes water, sodium chloride, and quartz. NA means that the physicochemical parameter is not provided by EPI Suite™. For organic salts,
          parameters are estimated using the desalted form.
Rank
1
2
3
4
5
6
7
8
9
10
11
Chemical Name
Methanol
Distillates, petroleum,
hyd retreated light
Hydrochloric acid
Isopropanol
Ethylene glycol
Peroxydisulfuric acid,
diammonium salt
Sodium hydroxide
Guargum
Glutaraldehyde
Propargyl alcohol
Potassium hydroxide
CASRN
67-56-1
64742-47-8
7647-01-0
67-63-0
107-21-1
7727-54-0
1310-73-2
9000-30-0
111-30-8
107-19-7
1310-58-3
Number Of
Wells Using
Chemical
(% of wells)
24,753 (72%)
22,463 (65%)
22,380 (65%)
16,039 (47%)
15,800 (46%)
14,968 (44%)
13,265 (39%)
12,696 (37%)
11,562 (34%)
11,410 (33%)
10,049 (29%)
Log Kow (unitless)
Estimated
-0.63
NA
NA
0.28
-1.2
NA
NA
NA
-0.18
-0.42
NA
Measured
-0.77
NA
NA
0.05
-1.36
NA
NA
NA
NA
-0.38
NA
Water
Solubility
Estimate From
Log Kow
(mg/L @ 25°C)
1.00 x 106
NA
NA
4.024 x 105
1.00 x 106
NA
NA
NA
1.672 x 105
9.355 x 105
NA
Henry's Law Constant
(atm mVmole @ 25°C)
Estimated,
Bond
Method
4.27 x 10"6
NA
NA
7.52 x 10"6
1.31 x 10"7
NA
NA
NA
1.10 x 10"7
5.88 x 10"7
NA
Estimated,
Group
Method 25
3.62 x 10"6
NA
NA
1.14xlO"5
5.60 x 10"11
NA
NA
NA
2.39 x 10"s
NA
NA
Measured
4.55 x 10"6
NA
NA
S.lOxlO"6
6.00 x 10"s
NA
NA
NA
NA
1.15 x 10"6
NA
Estimated Volume,
per disclosure
(gal)
Mean
1,218
NA
28,320
2,095
614
NA
551
NA
1,313
183
NA
Median
110
NA
3,110
55
184
NA
38
NA
122
2
NA
June 2015
This document is a draft for review purposes only and does not constitute Agency policy.

                               5-62
DRAFT—DO NOT CITE OR QUOTE

-------
Hydraulic Fracturing Drinking Water Assessment
                                                                                           Chapter 5 - Chemical Mixing






Rank
12
13
14
15

16

17
18
19
20






Chemical Name
Ethanol
Acetic acid
Citric acid
2-Butoxyethanol
Solvent naphtha,
petroleum, heavy
arom.
Naphthalene
2,2-Dibromo-3-
nitrilopropionamide
Phenolic resin
Choline chloride






CASRN
64-17-5
64-19-7
77-92-9
111-76-2

64742-94-5

91-20-3
10222-01-2
9003-35-4
67-48-1



M U /"**
Wells Using
Chemical
(% of wells)
9,861 (29%)
8,186 (24%)
8,142 (24%)
7,347 (21%)

7,108 (21%)

6,354 (19%)
5,656 (16%)
4,961 (14%)
4,741 (14%)


Log Kow (unitless)



Estimated
-0.14
0.09
-1.67
0.57

NA

3.17
1.01
NA
-5.16


Measured
-0.31
-0.17
-1.64
0.83

NA

3.3
0.82
NA
NA


Water
Solubility
Estimate From
Log Kow
(mg/L @ 25°C)
7.921 x 105
4.759 x 105
1.00 x 106
6.447 x 104

NA

1.421 x 102
2.841 x 103
NA
1.00 x 106

Henry's Law Constant
(atm mVmole @ 25°C)

Estimated,
Bond
Method
5.67 x 10"6
5.48 x 10"7
8.33 x 10"1S
9.79 x 10"s

NA

5.26 x 10"4
6.16xlO"14
NA
2.03 x 10"16
Estimated,
Group
Method 25
4.88 x 10"6
2.94 x 10"7
NA
2.08 x 10"s

NA

3.7 x 10"4
NA
NA
NA


Measured
5.00E-06
1.00 x 10"7
4.33 x 10"14
1.60 x 10"6

NA

4.4 x 10"4
1.91 x 10"s
NA
NA
Estimated Volume,
per disclosure
(gal)



Mean
831
646
163
385

NA

72
183
NA
2,131


Median
121
47
20
26

NA

12
5
NA
290
 Note: Analysis considered 34,675 disclosures and 676,376 ingredient records that met selected quality assurance criteria, including: completely parsed; unique combination of
 fracture date and API well number; fracture date between January 1, 2011, and February 28, 2013; valid CASRN; and valid concentrations. Disclosures that did not meet quality
 assurance criteria (3,855) or other, query-specific criteria were excluded from analysis.
June 2015
This document is a draft for review purposes only and does not constitute Agency policy.
                                     5-63
DRAFT—DO NOT CITE OR QUOTE

-------
      Hydraulic Fracturing Drinking Water Assessment
      5.8.4.  Transformation Processes
 1    It is important to understand the processes governing transformation of chemicals in the
 2    environment The transformation of a chemical reduces its concentration over time. Chemicals may
 3    completely degrade before reaching a drinking water resource. Transformation processes may be
 4    biotic or abiotic. The transformation process may transform a chemical into a less or more harmful
 5    chemical.

 6    Biodegradation is a biotic process where microorganisms transform a chemical from its original
 7    form into another chemical. For example, the biodegradation pathway of methanol is
 8    CH3OH^ CH20 -^ CHOOH -^ C02, or methanol -^ formaldehyde -^ formic acid -^ carbon dioxide.
 9    This pathway shows how the original chemical transforms through a series of steps until it becomes
10    the final product, carbon dioxide. Some chemicals are readily biodegraded, while others break
11    down slowly over time. Biodegradation is a highly site-specific process, requiring nutrients, a
12    carbon source, water, and an energy source. A highly biodegradable chemical could be persistent if
13    the conditions for biodegradability are not met. Conversely, a highly biodegradable chemical could
14    biodegrade quickly under the right conditions, before it can impact a water resource. The
15    relationship between mobility and biodegradability is complex, and therefore a variety of factors
16    can influence a particular compound's movement through the environment

17    Abiotic processes, such as oxidation, reduction, photochemical reaction, and hydrolysis, can
18    transform or break apart chemicals. In hydrolysis, for example, a water molecule substitutes for a
19    group of atoms. The typical results are products that are more polar than the original, and thus
20    have different physicochemical properties than the original compounds [Schwarzenbach et al.,
21    2002).1

      5.8.5.  Fate  and Transport of Chemical Mixtures
2 2    Chemicals at hydraulic fracturing sites are often present as mixtures, which may act differently in
23    the environment than individual chemicals  do. Individual chemicals can affect the fate and
24    transport of other chemicals in a mixture primarily by changing their solubility and biodegradation
25    rates.

26    Mixtures of chemicals may be more mobile  than individual chemicals due to cosolvency, which
27    increases solubility in the aqueous phase. Methanol and ethanol are examples of cosolvent alcohols
28    used frequently in hydraulic fracturing fluids [U.S. EPA. 2015aj. The presence of either greatly
29    increases BTEX solubility [Rasa etal.. 2013: Corseuil etal.. 2011: Heermann and Powers. 19981.2 By
30    increasing solubility, ethanol can affect the  fate and transport of other compounds. For example,
      i A polar molecule is a molecule with a slightly positive charge at one part of the molecule and a slightly negative charge
      on another. The water molecule, HzO, is an example of a polar molecule, where the molecule is slightly positive around the
      hydrogen atoms and negative around the oxygen atom.
      2 BTEX is an acronym for benzene, toluene, ethylbenzene, and xylenes. These chemicals are a group of single ringed
      aromatic hydrocarbons based on the benzene structure. These compounds are found in petroleum and are of specific
      importance because of their potential health effects.
                    This document is a draft for review purposes only and does not constitute Agency policy.
      June 2015                                     5-64                  DRAFT—DO NOT CITE OR QUOTE

-------
      Hydraulic Fracturing Drinking Water Assessment
 1    BTEX has been observed to travel farther in the subsurface in the presence of ethanol [Rasaetal.,
 2    2013: Corseuiletal..2011: Corseuiletal.. 2004: Powers etal.. 2001: Heermann and Powers. 1998].

 3    The presence of surfactants lowers fluid surface tension and increases solubility of organic
 4    compounds, and can mobilize less soluble/less mobile organic compounds. Two common
 5    surfactants reported to FracFocus 1.0 from January 1, 2011 to February 2013 were 2-
 6    butoxyethanol  (CASRN 111-76-2, 21% of disclosures) and poly(oxy-l,2-ethanediyl)-nonylphenyl-
 7    hydroxy (mixture) (CASRN 127087-87-0, 20% of disclosures). Additionally, surfactants can
 8    mobilize bacteria in the subsurface, which can increase the impact of pathogens on drinking water
 9    resources (Brown and Jaffe. 2001).

10    When chemicals are present as mixtures, one chemical may decrease or enhance the
11    biodegradability of another through inhibition or co-metabolism. The process of inhibition can
12    occur when multiple chemicals compete for the same enzyme, so only one chemical is degraded at a
13    time, which can ultimately slow biodegradation of each of the chemicals present. For example, the
14    biodegradation of ethanol and methanol may inhibit the biodegradation of BTEX or other organic
15    compounds present (Rasaetal.. 2013: Powers et al.. 2001). Co-metabolism may increase the
16    biodegradation rate of other compounds. For example, when methane or propane is present with
17    tetrachloroethylene, the enzyme  produced by bacteria to degrade methane also degrades
18    tetrachloroethylene (e.g., Alvarez-Cohen and Speitel, 2001 and references therein). For the
19    purposes of chemicals used in hydraulic fracturing, the presence of other chemicals in additives and
20    hydraulic fracturing fluids could result in increased or decreased biodegradation if the chemicals
21    are spilled. A chemical that may have otherwise been biodegradable may be inhibited and act as a
22    long-term source.

      5.8.6.  Site and Environmental  Conditions
2 3    Environmental conditions at and around the spill site affect the movement and transformation of
24    the chemical. We discuss the following: site conditions (e.g., proximity, land cover,  and slope), soil
25    conditions (e.g., permeability and porosity), and weather and climate.

26    The proximity of a spill to a drinking water resource, either laterally in the case of a surface water
2 7    body or downward for an aquifer, affects the potential for impact. Land cover will affect how
28    readily a fluid moves over land. For example, more rugged land cover such as forest will impede
29    flow, and an asphalt road will facilitate flow. A spill that occurs on or near a sloped site may move
30    overland faster, making it more likely to reach a nearby surface water body. Flatter surfaces would
31    result in a greater chance for infiltration to the subsurface, which could increase the potential for a
32    ground water impact.

33    Soil characteristics that affect the transport and transformation of spill chemicals include soil
34    texture (e.g., clay, silt, sand), permeability, porosity, and organic content.1'1 Fluids will move more
      1 Permeability of a soil describes how easily a fluid can move through the soil. Under a constant pressure, a fluid will move
      faster in a high permeability soil than the same fluid in a low permeability soil.
                   This document is a draft for review purposes only and does not constitute Agency policy.
      June 2015                                    5-65                  DRAFT—DO NOT CITE OR QUOTE

-------
      Hydraulic Fracturing Drinking Water Assessment
 1    quickly through permeable soil (e.g., sand) than through less permeable soil (e.g., clay). A soil with a
 2    high porosity provides more volume to hold water and spilled chemicals. Another important factor
 3    for a site is the organic content, of which there are two competing types: soil organic carbon and
 4    dissolved organic carbon. Each type of carbon acts as a strong substance for chemicals to adhere to.
 5    Soil organic carbon present in a solid phase, such as dead and decaying leaves and roots, is not
 6    mobile and slows the movement of chemicals through the soil. Dissolved organic carbon (DOC)
 7    moves with the water and can act as a shuttling mechanism to move insoluble chemicals across the
 8    surface and through the subsurface. Chemicals may also associate with particulates and colloids,
 9    which may act as an important transport mechanism.

10    Weather and climate conditions also affect the fate and transport of a spilled chemical. After a
11    spilled chemical stops moving, rainfall may remobilize the chemical. The amount, frequency, and
12    intensity of precipitation will impact volume,  distance, and speed of chemical movement.
13    Precipitation may carry chemicals downward or overland, and it can cause erosion, which may
14    move sorbed chemicals overland.

      5.8.7.  Peer-Reviewed Literature on the Fate and Transport of Hydraulic Fracturing Fluid Spills
15    There has been limited peer-reviewed research investigating the fate and transport of chemicals
16    spilled at hydraulic fracturing sites. Aminto and Olson (2012) modeled a hypothetical spill of
17    1,000 gal (3,800 L) of hydraulic fracturing fluid using equilibrium partitioning. The authors
18    evaluated how 12 chemicals typically used for hydraulic fracturing in the Marcellus Shale would
19    partition amongst different phases: air, water, soil, and biota.2 They presented a ranking of
20    concentrations for each phase. In water, they  showed that sodium hydroxide (a pH buffer), 4,4-
21    dimethyl oxazolidine (a biocide), hydrochloric acid (a perforation clean-up additive), and 3,4,4-
22    trimethyl oxazolidine (a biocide) had the highest simulated water concentrations; however, these
23    concentrations depended on the chemicals included in the simulated mixture and the
24    concentrations of each. Their analysis also suggested that after a spill, a large fraction would enter
25    the air and leave the soil; however, some constituents would be left behind in the water, soil, and
26    biota compartments, which could effectively act as long-term contamination sources. Aminto and
27    Olson (2012) only studied this one scenario. Other scenarios could be constructed with different
28    chemicals in different concentrations. These scenarios may result in different outcomes with
29    greater impacts.

      5.8.8.  Potential and Documented Fate and Transport of Documented Spills
30    There is limited information on the fate and transport of hydraulic fracturing fluids and chemicals.
31    In this section, we highlight the potential and  documented impacts for three documented spills (U.S.
32    EPA. 2015n). In each case, we provide the documented and potential paths (surface, subsurface, or
33    combination) and the associated fate and transport governing processes by which a spill has been

      i Porosity of a soil describes the amount of empty space for a given volume of soil. The porosity describes how much air,
      water, or hydraulic fluid a given volume of soil can hold.
      2 The chemicals they investigated included: sodium hydroxide, ethylene glycol, 4,4-dimethyl oxazolidine, 3,4,4-trimethyl
      oxazolodine, 2-amino-2-methyl-l-propanol, formamide, glutaraldehyde, benzalkonium chloride, ethanol, hydrochloric
      acid, methanol, and propargyl alcohol.
                    This document is a draft for review purposes only and does not constitute Agency policy.
      June 2015                                     5-66                  DRAFT—DO NOT CITE OR QUOTE

-------
      Hydraulic Fracturing Drinking Water Assessment
 1    documented to or has the potential to have an impact on drinking water resources. The three cases
 2    involve a tank overflow with a surface water impact, a human error blender spill with a soil impact,
 3    and an equipment failure that had no impact These three spills were chosen to highlight cases
 4    where there was a documented impact, a potential impact, and no impact.

 5    In the first documented spill, shown in Figure 5-19, a tank overflowed twice, releasing a total of
 6    7,350 gal (980 ft3 or 28 m3).1 The spilled fluid was documented as containing a friction reducer and
 7    gel. The spill traveled across the land surface, crossed a road, and then continued to a nearby
 8    stream. The spill affected wetlands and a stream, where fish were killed. The fish kill indicates that
 9    the chemicals present were in high enough concentrations to have an adverse impact.
                                                 Tank overflow spill with documented impact to surface water:
                                                           2 spills, 7,350 gal fluid: water, friction reducer, gel
                                                                                Documented Impact:
                                                                       Fluid reached wetlands and creek
                                                                               Fish stressed and killed

      Figure 5-19. Fate and Transport Spill Example: Case 1.
                Spills information from U.S. EPA(2015n).

10    For this first spill, the documented path was overland flow from the tank to the stream with a
11    documented, immediate impact. In addition to this documented path, there are potential paths for
12    potential impacts to drinking water resources. The spilled chemicals may have penetrated into the
13    soils or sorbed to soils and vegetation as the fluid moved across the ground towards the stream.
      1 We provide the total volume of the spill in gallons as well as cubic length (cubic feet and cubic meters), because it may be
      a little harder to visualize how far a volume of 7,300 gal might travel.
                    This document is a draft for review purposes only and does not constitute Agency policy.
      June 2015                                     5-67                  DRAFT—DO NOT CITE OR QUOTE

-------
      Hydraulic Fracturing Drinking Water Assessment
 1    Chemicals could be mobilized during later rainfall, runoff, or erosion events. Chemicals that
 2    infiltrated the subsurface could serve as long-term sources, as well as travel laterally across the
 3    unsaturated zone, or continue downwards to the ground water. Additionally, some chemicals could
 4    be lost to transformation processes. The lack of reported soil or ground water sampling data
 5    prevents the ability to know if these potential paths occurred or not.

 6    The second documented spill [U.S. EPA, 2015n, line 144], shown in Figure 5-20, occurred when a
 7    cap was left off the blender, and 504 gal (70 ft3 or 2 m3) of biocide and hydraulic fracturing fluid
 8    were released; 294 gal (39 ft3 or 1.1 m3) were retained by a dike with a lined secondary
 9    containment measure, demonstrating the partial effectiveness of this containment mechanism.
10    However, 210 gal (28 ft3 or 0.8 m3) did run off-site and were vacuumed up. There was no
11    documented impact to surface or ground water.
                                                            Cap left off of blender; biocide and water spill
                                                   504 gal spilled: 294 gal caught by dike, 210 gal ran off-site
      Figure 5-20. Fate and Transport Spill Example: Case 2.
               Spills information from U.S. EPA(2015n).

12    In this second case, the uncontained 210 gal could have infiltrated the subsurface, creating a
13    potential path to ground water. There is no documented information on the composition of the
14    spilled fluid. Highly mobile chemicals would have penetrated the soil more quickly than less mobile
15    chemicals, which would have sorbed to soil particles. As the chemicals penetrated into the soil,
16    some could have moved laterally in the unsaturated zone, or traveled downward to the water table
                   This document is a draft for review purposes only and does not constitute Agency policy.
      June 2015                                    5-68                  DRAFT—DO NOT CITE OR QUOTE

-------
      Hydraulic Fracturing Drinking Water Assessment
 1    and moved with the ground water. These chemicals could have served as a long-term source. These
 2    chemicals could have transformed into other chemicals with different physicochemical properties,
 3    and any volatile chemicals could have moved to the air as a loss process. As in the first case, there
 4    was no reported sampling of soil or groundwater, so there is no way to demonstrate whether
 5    chemicals did or did not follow this path.

 6    In the third documented spill [U.S. EPA. 2015n. line 188], shown in Figure 5-21, 630 gal (84 ft3 or
 7    2.4 m3) of crosslinker spilled onto the well pad when a hose wore off at the cuff. The spill was
 8    contained in the berm and an on-site vacuum truck was used to clean up the spill. No impact to soil
 9    or water was reported.
                                                            Hose worn at cuff of blender; crosslinker spill
                                                            630 gal spilled onto pad. Prevented infiltration
                                                                                   of fluid into soil.
                               Equipment failure:
                                   Hose worn off
                                                               Onsite vac-truck
                                                               remediated spill
      Figure 5-21. Fate and Transport Spill Example: Case 3.
                Spills information from U.S. EPA(2015n).

10    For this third case, we do not have any information on whether the well pad was lined or not. If the
11    site had a liner, the spill could have been fully contained, not infiltrated the subsurface, and been
12    fully cleaned up. Without a liner or if the liner was not completely successful (e.g., had a tear), the
13    potential paths would have been similar to those above in the second case, where the chemicals
14    may sorb to the soils and penetrate into the subsurface. There was no reported sampling of soil or
15    ground water to determine whether or not chemicals migrated into the soil.
                    This document is a draft for review purposes only and does not constitute Agency policy.
      June 2015                                     5-69                  DRAFT—DO NOT CITE OR QUOTE

-------
      Hydraulic Fracturing Drinking Water Assessment
      5.9.  Trends in Chemicals Use in Hydraulic Fracturing
 1    This section provides an overview of ongoing changes in chemical use in hydraulic fracturing, with
 2    an emphasis on efforts to reduce potential impacts from surface spills by using fewer and safer
 3    chemicals.

 4    Representatives from oil and gas companies, chemical companies, and non-profits are working on
 5    strategies to reduce the number and volume of chemicals used and to identify safer chemicals
 6    [Waldron, 2014]. Southwestern Energy Company, for example, is developing an internal chemical
 7    ranking tool [SWN, 2014], and Baker Hughes is working on a hazard ranking system designed for
 8    wide-scale external use [Baker Hughes, 2014a: Brannonetal., 2012: Daultonetal., 2012: Brannon
 9    etal.. 2011]. Environmental groups, such as the Environmental Defense Fund, are also developing
10    hazard rating systems [Penttilaetal.. 2013]. Typical criteria used to rank chemicals include
11    mobility, persistence, biodegradation, bioaccumulation, toxicity, and hazard characteristics. In this
12    report, toxicity and a methodology to rank chemical hazards of hydraulic fracturing chemicals is
13    discussed in Chapter 9.

14    The EPA has not conducted a comprehensive review of efforts to develop safer hydraulic fracturing
15    chemicals. However, the following are some specific examples of efforts that companies cite as part
16    of their efforts toward safer chemical use:

17        •  A renewable citrus-based replacement for conventional surfactants [Fisher, 2012]:
18        •  A crosslinked gel system comprised of chemicals designated as safe food additives by the
19           U.S. Food and Drug Administration (Holtsclawetal.. 20111:
20        •  A polymer-free gel additive fAl-Ghazal etal.. 20131:
21        •  A dry, hydrocarbon-free powder to replace liquid gel concentrate [Weinstein et al., 2009]:
22        •  Biodegradable polymers [Irwin, 2013]:
23        •  The use of ultraviolet light to control bacteria [Rodveltetal., 2013]:
24        •  New chelating agents that reduce the use of strong acids [LePage etal., 2013], and
25        •  The recovery and reuse of flowback and produced water as hydraulic fracturing fluids,
26           which may reduce need to add additional chemicals [Horn etal.. 2013].
27    In addition to efforts to address environmental concerns, the oil and gas industry continues to
2 8    research and develop less expensive and more effective fracturing fluid additives. A review of the
29    EPA's new chemicals program found that from 2009 to April 2015, the Agency received pre-
30    manufacturing notices (PMN] for about 110 chemicals that have the potential for use as hydraulic
31    fracturing fluid additives. Examples include chemicals intended for use as clay control agents,
32    corrosion inhibitors, gel crosslinkers, emulsifiers, foaming agents, hydrate inhibitors, scale
33    inhibitors, and surfactants. At the time of PMN submission, these chemicals were not in commercial
34    use in the United States. As of April 2015, the EPA had received 30 notices of commencement,
35    indicating that some of those chemicals are now used commercially.
                   This document is a draft for review purposes only and does not constitute Agency policy.
      June 2015                                   5-70                  DRAFT—DO NOT CITE OR QUOTE

-------
      Hydraulic Fracturing Drinking Water Assessment
 1    The FracFocus 1.0 data extracted by the EPA cannot be used to identify temporal trends in additive
 2    usage. A data set with a much longer duration of data collection would be needed to distinguish
 3    actual temporal trends from the normal diversity of chemicals in use as a result of geologic and
 4    geographic variability. However, the current FracFocus 1.0 database provides a point of comparison
 5    for use in the future.

      S.lO.Synthesis
 6    Chemical mixing is the process by which a base fluid, chemicals,  and proppant are mixed prior to
 7    injection into the well. This chapter addressed the potential for on-site spills of chemicals used in
 8    the hydraulic fracturing process to affect the quality of drinking  water resources, which is governed
 9    by three overarching factors:  (1) fluid characteristics, (2) chemical management and spill
10    characteristics, and (3) chemical fate and transport.

      5.10.1. Summary of Findings
11    Documented on-site chemical spills have occurred during the chemical mixing process and reached
12    soil and surface water receptors, with potential impacts to drinking water resources. The EPA
13    analysis  of 497 spills reports found no documented impacts to ground water from those particular
14    chemical spills, though there was little information on post-spill  testing and sampling [U.S. EPA,
15    2015n). The EPA's case study in Killdeer, ND strongly suggests that there was impact to ground
16    water, but it is unclear if the path was via the surface spill caused by the blowout [U.S. EPA. 2015)].
17    The EPA found 151 spills of chemicals or fracturing fluid on or near the well pad in a six-year time
18    period. The chemical spills were primarily caused by equipment failure (34%), closely followed by
19    human error (25%). The remaining spills were caused by a failure of container integrity, weather,
20    vandalism, well communication, or unknown causes. Reported spills cover a large range of volumes,
21    from five to 19,000 gal (19 to 72,000 L), with a median of 420 gal (1,700 L).

22    If a spilled fluid reaches a drinking water resource, the potential to affect the water quality is largely
23    governed by the fluid characteristics. A typical water-based fracturing fluid is composed of 90%-
24    94% water, 5%-9% proppant, and less than 2% chemical additives (Carter etal., 2013: Knappe and
25    Fireline. 2012). According to the EPA's analysis of disclosures to FracFocus 1.0, approximately 93%
26    of hydraulic fracturing fluids are inferred to use water as a base  fluid (U.S. EPA, 2015a). Non-
27    aqueous constituents, such as nitrogen, carbon dioxide, and hydrocarbons, are also used as base
28    fluids or used  in combination with water as base fluids.

29    The EPA has identified 1,076 unique chemicals used in hydraulic fracturing fluids. The chemicals
30    include acids,  aromatic hydrocarbons, bases, hydrocarbon mixtures, polysaccharides, and
31    surfactants. Of the 1,076 chemicals, 453 have physicochemical properties in the EPI Suite™
32    database. These chemicals range from fully miscible to insoluble, and from highly hydrophobic to
33    highly hydrophilic. The majority of the chemicals are not volatile.

34    According to the EPA's analysis of FracFocus, a median of 14 chemicals are used per well, with a
35    range of four to 28 (5th and 95th percentiles). The volumes used range from tens to tens of
36    thousands of gallons (tens to tens of thousands of liters) per well; therefore, operators typically
                   This document is a draft for review purposes only and does not constitute Agency policy.
      June 2015                                    5-71                  DRAFT—DO NOT CITE OR QUOTE

-------
      Hydraulic Fracturing Drinking Water Assessment
 1    store chemicals on-site in large volumes (typically 200 to 400 gal (760-1,500 L) totes), often in
 2    multiple containers. The ten most common chemicals (excluding quartz) are methanol,
 3    hydrotreated light petroleum distillates, hydrochloric acid, isopropanol, ethylene glycol,
 4    peroxydisulfuric acid diammonium salt, sodium hydroxide, guar gum, glutaraldehyde, and
 5    propargyl alcohol. These chemicals are present in multiple additives. Methanol was reported in
 6    72% of the FracFocus disclosures, and hydrotreated light petroleum distillates and hydrochloric
 7    acid were both reported in over half the disclosures (U.S. EPA. 2015b).

      5.10.2. Factors Affecting the Frequency or Severity of Impacts
 8    The potential for spills from the chemical mixing process to affect drinking water resources
 9    depends on three factors: fluid characteristics, chemical management and spill characteristics, and
10    chemical fate and transport. Specific factors affecting the frequency and severity of impacts include
11    size and type of spill, volume of chemicals spilled, type of chemicals and their properties,
12    combinations of chemicals spilled, environmental conditions, proximity to drinking water
13    resources, employee training and experience, quality and maintenance  of equipment, and spill
14    containment and mitigation.

15    The size and type of a fracturing operation, including the number of wellheads, the depth of the
16    well, the length of the horizontal leg, and the number of stages and phases, affect the likelihood and
17    potential impacts of spills. Larger operations may require larger volumes of chemicals, more
18    storage containers, more equipment, and additional transfers between  different pieces of
19    equipment.  Larger storage containers increase the maximum volume of a spill or leak from a
20    storage container, and additional transfers between equipment increase the possibility of human
21    error.

22    The type of chemical spilled governs how it will move and transform in the environment More
23    mobile chemicals move faster through the environment, causing a quicker impact More mobile
24    chemicals are also generally more soluble and may reach the drinking water resource at higher
25    concentrations. Less mobile chemicals will move more slowly, and may have delayed and longer-
26    term impacts, at lower concentrations. The severity of impact is also governed by how the chemical
27    adversely impacts water quality. Water quality impacts may range  from aesthetic effects (e.g., taste,
28    smell) to adverse health effects.

29    The environmental conditions at and around the spill site affect the fate and transport of a given
30    chemical. These conditions  include soil properties, climate, weather, and terrain. Permeable soils
31    may allow for rapid transport of the spilled fluid through the soil and into a nearby drinking water
32    resource. Precipitation can re-mobilize trapped chemicals and move them over land or through  the
33    subsurface.

34    The proximity of a spill to drinking water resources affects the frequency and severity of impact
35    The closer a spill is to a drinking water a resource, the higher potential to reach it Also, as a fluid
36    moves toward a drinking water resource, it may decrease in concentration, which will affect the
37    severity of an impact. More concentrated chemicals have the potential to have a bigger impact on
                   This document is a draft for review purposes only and does not constitute Agency policy.
      June 2015                                    5-72                  DRAFT—DO NOT CITE OR QUOTE

-------
      Hydraulic Fracturing Drinking Water Assessment
 1    water quality. The characteristics of the drinking water resource will also influence the magnitude
 2    of the impact of a spill.

 3    The most successful way to prevent impacts to drinking water resources is to prevent spills from
 4    occurring in the first place and to quickly and effectively contain spills. Effective spill containment
 5    and mitigation measures can prevent or reduce the frequency and severity of impacts. Spill
 6    containment measures include well pad containment liners, diversion ditches, berms, dikes,
 7    overflow prevention devices, drip pans, and secondary containers. These may prevent a spill from
 8    reaching soil and water receptors. Spill mitigation, including removing contaminated soils,
 9    vacuuming up spilled  fluids, and using sorbent materials may limit the severity of a spill.
10    Implementation of these measures varies from site-to-site and may not always be effective.

      5.10.3. Uncertainties
11    The lack of information regarding the composition of chemical additives and fracturing fluids,
12    containment and mitigation measures in use, and the fate and transport of spilled fluids greatly
13    limits our ability to assess potential impacts to drinking water resources.

14    There is no standard design for hydraulic fracturing fluids. Detailed information on the chemicals
15    used is limited, and volumes of chemicals stored on-site are  generally not publicly available. These
16    limitations in data preclude the ability to know what volumes of chemicals may be spilled.
17    FracFocus, which currently holds the most comprehensive information on water and chemicals
18    used in hydraulic fracturing fluids, identifies well-specific chemicals and the concentration of those
19    chemicals as a maximum percentage of the mass of fracturing fluid. Accuracy and completeness of
20    original FracFocus disclosure information was not verified. In applying the EPA-standardized
21    chemical list to the ingredient records in the FracFocus database, standardized chemical names
22    were assigned to  only 65% of the ingredient records from the more than 36,000 unique, fully
23    parsed disclosures. The remaining ingredient records could not be assigned a standardized
24    chemical name and were excluded from analyses [U.S.  EPA. 2015a].

25    Operators may specify certain ingredients as confidential business information (CBI) and not
26    disclose the chemical used. More than 70% of disclosures to FracFocus  contained at least one CBI
2 7    chemical. Of disclosures with at least one CBI chemical, the average number of CBI chemicals was
28    five. Approximately 11% of all ingredients were reported to FracFocus as CBI [U.S. EPA. 2015a). No
29    data are available in FracFocus for any chemical listed as CBI. Therefore, FracFocus CBI chemicals
30    are not included in analyses of volume, physical properties, or any other analysis in this
31    assessment, although  we were able to do limited physicochemical analysis of 19 CBI  chemicals.

32    Of the 1,076 hydraulic fracturing fluid chemicals identified by the EPA, 623 did not have estimated
33    physicochemical properties reported in the EPI Suite™ database. Knowing the chemical properties
34    of a spilled fluid is essential to predicting how and where it will travel in the environment Although
35    we can make some generalizations about the physicochemical properties  of these chemicals and
36    how spilled chemicals may move in the environment, the distribution of properties could change if
37    we obtained data for all known fracturing fluid chemicals (as well as for those listed as CBI).
                   This document is a draft for review purposes only and does not constitute Agency policy.
      June 2015                                    5-73                  DRAFT—DO NOT CITE OR QUOTE

-------
      Hydraulic Fracturing Drinking Water Assessment
 1    In order to determine the potential impact of a spill, the physicochemical properties, the site-
 2    specific environmental conditions, and proximity to drinking water resources must be known. This
 3    information is generally lacking.

 4    There is a lack of baseline surface water and ground water quality data. This lack of data limits our
 5    ability to assess the relative change to water quality from a spill or attribute the presence of a
 6    contaminant to a specific source.

 7    In addition to limited information on chemical usage, we cannot complete a thorough assessment of
 8    the potential impact of chemical spills due to limited information on actual spills. Data sources used
 9    in the EPA's spills analysis do not cover all states with hydraulic fracturing activity. The available
10    data provide limited information on the types and volumes of chemicals spilled, spill causes,
11    containment and mitigation measures, and sources of spills.  In addition, there is little available data
12    on impacts of spills, due to a lack of baseline data and incomplete documentation of follow-up
13    actions and testing.

14    In general, then, we are limited in our ability to fully assess potential impacts to drinking water
15    resources from chemical spills, based on available current information. To improve our
16    understanding we need: more information on the chemical composition of additives  and fracturing
17    fluid; the physicochemical properties of chemicals used; baseline monitoring and field studies of
18    spilled chemicals; drinking water resources quality conditions before and after hydraulic fracturing
19    is performed; detailed site-specific environmental conditions; more information on the
2 0    containment and mitigation measures and their effectiveness; and the types and volumes of spills.

      5.10.4. Conclusions
21    The chemical mixing stage of the hydraulic fracturing process has the potential to cause impacts to
2 2    drinking water resources by way of surface spills of chemicals and  fracturing fluids. There are
2 3    documented chemical spills at fracturing sites, but a lack of available data limits our ability to
24    determine impacts. Potential impacts to drinking water resources are governed by the fluid
2 5    characteristics, chemical management and spill characteristics, and the fate and transport of spilled
26    chemicals through the environment
                   This document is a draft for review purposes only and does not constitute Agency policy.
      June 2015                                    5-74                  DRAFT—DO NOT CITE OR QUOTE

-------
      Hydraulic Fracturing Drinking Water Assessment
      Text Box 5-16. Research Questions Revisited.

 1    What is currently known about the frequency, severity, and causes of spills of hydraulic fracturing fluids
 2    and additives?

 3    •   The frequency of on-site spills from hydraulic fracturing operations could be obtained for two states.
 4        Frequency estimates from data and literature ranged from 0.4 to 1.3 spills for every 100 wells
 5        hydraulically fractured in Pennsylvania and Colorado, respectively, and between 3.3 and 12.2 spills for
 6        every 100 wells installed in Pennsylvania (Rahmetal., 2015: U.S. EPA, 2015n: Brantley etal., 2014:
 7        Gradient. 20131.i These estimates include spills of hydraulic fracturing chemicals and fluids, and
 8        flowback and produced water reported in state databases. It is unknown whether these spill estimates
 9        are representative of national occurrences. Estimates of the frequency of on-site spills from hydraulic
10        fracturing operations were unavailable for other areas. If the estimates are representative, the number of
11        spills nationally could range from 100 to 3,700 spills annually, assuming 25,000 to 30,000 new wells are
12        fractured p er y ear.

13    •   In an analysis of spills, EPA characterized volumes and causes of hydraulic fracturing-related spills
14        identified from selected state and industry data sources. The spills occurred between January 2006 and
15        April 2012 in 11 states and included 151 cases in which fracturing fluid or additives spilled on or near a
16        well  pad [U.S. EPA. 2015n). These cases were likely a subset of all chemical and fracturing fluid spills
17        during the study's time period. The reported volume of chemicals or fracturing fluid spilled ranged from
18        5 gal to more than 19,000 gal (19 to 72,000 L), with a median volume of 420 gal (1,600 L) per spill. Spill
19        causes included equipment failure, human error, failure of container integrity, and other causes (e.g.,
20        weather and vandalism). The most common cause was equipment failure. Specific causes of equipment
21        failure included blowout preventer failure, corrosion, and failed valves. More than 30% of the chemical or
22        fracturing fluid spills characterized by the EPA came from fluid storage units (e.g., tanks, totes, and
23        trailers) (U.S. EPA. 2015n).
      1 Spill frequency estimates are for a given number of wells over a given period of time. These are not annual estimates nor
      are they for over a lifetime of the wells.
                     This document is a draft for review purposes only and does not constitute Agency policy.
      June 2015                                        5-75                   DRAFT—DO NOT CITE OR QUOTE

-------
      Hydraulic Fracturing Drinking Water Assessment
 1    What are the identities and volumes of chemicals used in hydraulic fracturing fluids, and how might this
 2    composition vary at a given site and across the country?

 3    •   In this assessment, we compiled a list of 1,076 chemicals used to formulate hydraulic fracturing fluids.
 4        These chemicals include acids, alcohols, aromatic hydrocarbons, bases, hydrocarbon mixtures,
 5        polysaccharides, and surfactants. This is a cumulative list over multiple wells and years. Operators used
 6        an median of 14 unique chemicals per well according to the EPA's analysis of disclosures to FracFocus
 7        [U.S. EPA. 2015a).

 8    •   Our analysis showed that chemical use varies and that no single chemical is used at all well sites across
 9        the country, although several chemicals are widely used. Methanol, hydrotreated light petroleum
10        distillates, and hydrochloric acid were reported in 65% or more of FracFocus disclosures analyzed by the
11        EPA  [U.S. EPA, 2015al. The composition of hydraulic fracturing fluids varies by state, by well, and within
12        the same service company and geologic formation. This variability likely results from several factors,
13        including the geology of the formation, the availability and cost of different chemicals, and operator
14        preference [U.S. EPA.2015a).

15    •   Estimates from the EPA's database developed from disclosures made to FracFocus suggest median
16        volumes of individual chemicals injected per well ranged from a few gallons to thousands of gallons, with
17        a median of 650 gal (2,500  L) per chemical per well [U.S. EPA, 2015bl. If 14 unique chemicals are used
18        per well, then an estimated 9,100 gal (34,000 L) of chemicals may be injected per well (U.S. EPA. 2015aj.

19    What are the chemical and physical properties of hydraulic fracturing chemical additives?

20    •   Measured or estimated physicochemical properties were obtained for 453 of the 1,076 chemicals
21        reported in hydraulic fracturing fluids. The wide variety of chemicals results in a wide range of
22        physicochemical properties.

23    •   Many hydraulic fracturing chemicals fully dissolve in water, but the aqueous solubilities range from fully
24        miscible to sparingly soluble.

25    •   The octanol-water partition coefficient  ranges from the highly hydrophilic to the highly hydrophobic.
2 6        Many chemicals used in hydraulic fracturing fluid fall in the middle of this range, suggesting that they will
2 7        divide equally between water and solid phase, so that they may move slower through the environment
28        than those that associate more with water. More chemicals will associate strongly with soils and organic
29        materials, suggesting the potential of these chemicals to be long-term contaminants if they are spilled.

30    •   There are few hydraulic fracturing chemicals that are volatile. Most hydraulic fracturing chemicals will
31        tend to remain in water as opposed to volatilizing to the air.
                     This document is a draft for review purposes only and does not constitute Agency policy.
      June 2015                                        5-76                    DRAFT—DO NOT CITE OR QUOTE

-------
      Hydraulic Fracturing Drinking Water Assessment
 1    •   The chemicals for which we know physicochemical properties are not necessarily the chemicals most
 2        frequently reported as used in hydraulic fracturing activities. Of the 453 chemicals for which
 3        physicochemical properties were available, 18 of the top 20 most mobile chemicals were included in 2%
 4        or less of disclosures fU.S. EPA. 2015b). However, two more common, but highly mobile chemicals,
 5        choline chloride and tetrakis (hydroxymethyl) phosphonium sulfate were reported in 14% and 11% of
 6        disclosures, respectively. These two chemicals are thus relatively more common, and, if spilled, their
 7        properties of high mobility means they would move quickly through the environment with the flow of
 8        water.

 9    If spills occur, how might hydraulic fracturing chemical additives contaminate drinking water
10    resources?

11    •   When chemicals are spilled, there are several paths by which a chemical could contaminate drinking
12        water resources. The chemical could flow overland to nearby surface water, penetrate into the soil that
13        could travel laterally and impact surface waters, or infiltrate and contaminate the underlying ground
14        water.

15    •   Of the 151 spills characterized by the EPA, fluids reached surface water in 13 (9%) and soil in 97 (64%)
16        of those cases. None of the spills reportedly reached ground water [U.S. EPA, 2015n), but it could take
17        several years for spilled fluids to infiltrate soil and leach into ground water. Thus, it may not be
18        immediately known whether a spill reaches ground water or not

19    •   The timing of a potential impact varies, but it could occur quickly, be delayed, have a continual impact
20        over time, or occur much later. Which path the spill takes depends on different conditions, such as
21        distance to a water receptor, spill volume, soil characteristics, and the physicochemical properties of the
22        chemical.
      5.11.References for Chapter 5
      Abriola. LM: Finder. GF. (1985a). A multiphase approach to the modeling of porous-media contamination by
         organic-compounds .2. Numerical-simulation. Water Resour Res 21:19-26.
      Abriola. LM: Finder. GF. (1985b). A multiphase approach to the modeling of porous media contamination by
         organic compounds: 1. Equation development. Water Resour Res 21:11-18.
         http://dx.doi.org/10.1029/WR021i001p00011
      Al-Ghazal, M: Al-Driweesh, S: Al-Shammari, F. (2013). First successful application of an environment friendly
         fracturing fluid during on-the-fly proppant fracturing. Paper presented at International Petroleum
         Technology Conference, March, 26-28, 2013, Beijing, China.
      ALL Consulting (ALL Consulting, LLC). (2012). The modern practices of hydraulic fracturing: A focus on
         Canadian resources. Tulsa, Oklahoma: ALL Consulting LLC.
      Allen, TE. (2013). Pregel blender prototype designed to reduce cost and environmental problems. (SPE-
         27708-MS). Allen, TE. http://dx.doi.org/10.2118/27708-MS
                     This document is a draft for review purposes only and does not constitute Agency policy.
      June 2015                                       5-77                    DRAFT—DO NOT CITE OR QUOTE

-------
Hydraulic Fracturing Drinking Water Assessment
Allison. D: Folds. PS: Harless. PI: Howell. M: Vargus. GW: Stipetich. A. (2009). Optimizing openhole completion
   techniques for horizontal foam-drilled wells. Paper presented at SPE Eastern Regional Meeting, September
   23-25,2009, Charleston, WV.

Alvarez-Cohen. L: Speitel. GE. (2001). Kinetics of aerobic cometabolism of chlorinated solvents.
   Biodegradation 12:105-126. http://dx.doi.Org/10.1023/A:1012075322466

Aminto. A: Olson. MS. (2012). Four-compartment partition model of hazardous components in hydraulic
   fracturing fluid additives. Journal of Natural Gas Science & Engineering 7: 16-21. http://dx.doi.org/DOI:
   10.1016/j.jngse.2012.03.006

API (American Petroleum Institute). (2011). Practices for mitigating surface impacts associated with
   hydraulic fracturing. Washington, DC: API Publishing Services.
   http://www.shalegas.energy.gov/resources/HF3 e7.pdf

Arthur. ID: Bohm. B: Coughlin. Bl: Layne. M. (2009b). Evaluating the environmental implications of hydraulic
   fracturing in shale gas reservoirs. Paper presented at SPE Americas E&P Environmental and Safety
   Conference, March 23-25, 2009, San Antonio, TX.

Arthur. ID: Bohm. B: Layne. M. (2008). Hydraulic fracturing considerations for natural gas wells of the
   Marcellus  shale. Paper presented at The Ground Water Protection Council Annual Forum, September 21-
   24, 2008, Cincinnati, OH.

Ayala. S: Barber. T: Dessinges. MN: Frey. M: Horkowitz. I: Leugemors. E: Pessin. I: Way. CH: Badry. R: Kholy. IE:
   Gait. A: Hjellesnet. M: Sock. D. (2006). Improving oilfield service efficiency.  Houston, TX: Schlumberger
   Limited.
   http://www.slb.eom/~/media/Files/resources/oilfield review/ors06/aut06/improving oilfield service
   efficiency.pdf

Baker Hughes. (2014a). Baker Hughes green chemicals program overview. Available online

Barati. R: Liang. IT. (2014). A review of fracturing fluid systems used for hydraulic fracturing of oil and gas
   wells. J Appl  Polymer Sci Online pub. http://dx.doi.org/10.1002/app.40735

Bl Services Company. (2009). BJ fracturing manual 2.0. service company confidential business information
   document (Revision No. 1 ed. ed.). Houston, TX.

Bouchard. D: Cornaton. F: Hohener. P: Hunkeler. D. (2011). Analytical modelling of stable isotope
   fractionation of volatile organic compounds in the unsaturated zone. J Contam Hydrol 119:  44-54.
   http://dx.doi.0rg/10.1016/j.jconhyd.2010.09.006

Boyer. CM: Glenn. SA: Claypool. BR: Weida. SD: Adams. ID: Huck. PR: Stidham. IE. (2005). Application of
   viscoelastic fracturing fluids in Appalachian Basin reservoirs (SPE 98068 ed.). Richardson, TX: Society of
   Petroleum Engineers. http://dx.doi.org/10.2118/98068-MS

Brannon, HP: Daulton, PI: Hudson, HG: lordan, AK.  (2011). Progression toward implementation of
   environmentally responsible fracturing processes (SPE-147534-MS ed.). Denver, CO: Society of Petroleum
   Engineers. http://dx.doi.org/10.2118/147534-MS

Brannon. HP: Daulton. PI: Post. MA: Hudson. HG: lordan. AK. (2012). The quest to exclusive use of
   environmentally responsible fracturing products and systems. Richardson, TX: Society of Petroleum
   Engineers. http://www.onepetro.org/mslib/app/Preview.do?paperNumber=SPE-152068-
   MS&societyCode=SPE

Brannon. HP: Kendrick. DE: Luckey. E: Stipetich. A.  (2009). Multistage Fracturing of Horizontal Shale Gas
   Wells Using >90% Foam Provides Improved Production. In 2009 SPE Eastern regional meeting: limitless
   potential/formidable challenges (SPE 124767 ed.). Richardson, TX: Society of Petroleum Engineers.
   http://dx.doi.org/10.2118/124767-MS

Brannon. HP: Pearson. CM. (2007). Proppants and fracture conductivity. In Modern fracturing - enhancing
   natural gas production (1st ed.). Houston, TX: Energy Tribune Publishing, Inc.


               This document is a draft for review purposes only and does not constitute Agency policy.
June 2015                                       5-78                    DRAFT—DO NOT CITE OR QUOTE

-------
Hydraulic Fracturing Drinking Water Assessment
Brantley. SL: Yoxtheimer. D: Arimand. S: Grieve. P: Vidic. R: Pollak. I: Llewellyn. GT: Abad. I: Simon. C. (2014).
   Water resource impacts during unconventional shale gas development: The Pennsylvania experience. Int J
   Coal Geol 126:140-156. http://dx.doi.Org/10.1016/i.coal.2013.12.017

Brown. DG: laffe. PR. (2001). Effects of nonionic surfactants on bacterial transport through porous media.
   Environ Sci Technol 35: 3877-3883. http://dx.doi.org/10.1021/es010577w
Browne. PI: BD. L. (1999). The development of a rapid hydration on-the-fly crosslinked water fracturing fluid.
   Paper presented at CSPG and Petroleum Society Joint Convention, Digging Deeper, Finding a Better
   Bottom Line, June 14 18,1999, Calgary, Canada.
Carter, KE: Hammack, RW: Hakala, IA. (2013). Hydraulic Fracturing and Organic Compounds - Uses, Disposal
   and Challenges. SPE Eastern Regional Meeting, Pittsburgh, Pennsylvania, USA.

ChemicalBook (ChemicalBook Inc.). (2010). Sorbitan trioleate. Available online at
   http://www.chemicalbook.com/chemicalproductproperty en cb4677178.htm (accessed April 6,2015).
COGCC (Colorado Oil and Gas Conservation Commission). (2014). Risk-based inspections: Strategies to
   address environmental risk associated with oil and gas operations. (OGCC2014PROJECT #7948). Denver,
   CO.
Corseuil.  HX: Kaipper. BI: Fernandes. M. (2004). Cosolvency effect in subsurface systems contaminated with
   petroleum hydrocarbons and ethanol. Water Res 38:1449-1456.
   http://dx.doi.0rg/10.1016/j.watres.2003.12.015

Corseuil.  HX: Monier. AL: Fernandes. M: Schneider. MR: Nunes. CC: Do Rosario. M: Alvarez. PI. (2011). BTEX
   plume dynamics following an ethanol blend release: geochemical footprint and thermodynamic
   constraints on natural attenuation. Environ Sci Technol 45: 3422-3429.
   http://dx.doi.org/10.1021/esl04055q

Coulter. GR: Gross. BC: Benton. EG: Thomson. CL. (2006). Barnett shale hybrid fracs - One operator's design,
   application, and results. Richardson, TX: Society of Petroleum Engineers.
   http://dx.doi.org/10.2118/102063-MS
Curtice. Rl: Salas. WDI: Paterniti. ML. (2009). To gel or not to gel? In 2009 SPE  annual technical conference &
   exhibition. Richardson, TX: Society of Petroleum Engineers. http://dx.doi.org/10.2118/124125-MS

Daulton. D: Post. M: McMahon. I: Kuc. B: Ake. C: Hughes. B: Hill. D. (2012). Global chemical evaluation process
   review to qualify regulatory and environmental characteristics for oilfield chemical products. Paper
   presented at SPE Annual Technical Conference and Exhibition, October 8-10, 2012, San Antonio, TX.

Domenico. PA: Schwartz. FW. (1997). Physical and chemical hydrology. In Physical and chemical
   hydrogeology (2nd ed.). Hoboken, NJ: Wiley.
Drillinglnfo. Inc.. (2012). DI Desktop August 2012 download  [Database]. Austin, TX. Retrieved from
   http: //info, drillinginfo. com/

Economides. Ml:  Mikhailov. DN: Nikolaevskiy. VN. (2007). On the problem of fluid leakoff during hydraulic
   fracturing. Transport in Porous Media 67: 487-499. http://dx.doi.org/10.1007/sll242-006-9038-7

Elbel. I: Britt. L. (2000). Fracture treatment design. In MJ Economides; KG Nolle (Eds.), Reservoir stimulation
   (3rd ed.). New York, NY: John Wiley & Sons.
Ely. IW. (1989). Chapter 7: Fracluring fluids and additives. In  JL Gidley;  SA Holdilch; DE Nierode; RW Vealch
   Jr (Eds.), Recenl advances in hydraulic fracluring (pp. 131-146). Richardson, TX: Society of Pelroleum
   Engineers.
Fink. IK. (2003). Oil field chemicals. In Oil field chemicals. Boston, MA: Gulf Professional Publishing.

Fisher. K. (2012). Green frac fluid chemislry optimizes well productivity, environmental performance
   [Magazine]. The American Oil and Gas Reporter, March 2012,4.
               This document is a draft for review purposes only and does not constitute Agency policy.

June 2015                                       5-79                   DRAFT—DO NOT CITE OR QUOTE

-------
Hydraulic Fracturing Drinking Water Assessment
Fredd. CN: Olsen. TN: Brenize. G: Ouintero. BW: Bui. T: Glenn. S: Boney. CL. (2004). Polymer-free fracturing
   fluid exhibits improved cleanup for unconventional natural gas well applications. Richardson, TX: Society
   of Petroleum Engineers. http://dx.doi.org/10.2118/91433-MS

Freeze. RA: Cherry. IA. (1979). Groundwater. In Groundwater. Upper Saddle River, NJ: Prentice Hall.

Gidley. IL: Holditch. SA: Nierode. DE: Veatch Jr.. RW. (1989). Recent advances in hydraulic fracturing.
   Richardson, TX: Society of Petroleum Engineers.

GNB (Government of New Brunswick). (2013). Responsible environmental management of oil and natural gas
   activities in New Brunswick - rules for industry. New Brunswick, Canada.
   http://www2.gnb.ca/content/dam/gnb/Corporate/pdf/ShaleGas/en/RulesforIndustry.pdf

GNB (Government of New Brunswick). (2015). FAQs hydraulic fracturing (fraccing). New Brunswick, Canada.
   http://www2.gnb.ca/content/dam/gnb/Corporate/pdf/ShaleGas/en/FAQ HydraulicFracturing.pdf

Gradient. (2013). National human health risk evaluation for hydraulic fracturing fluid additives. Gradient.
   http://www.energy.senate.gov/public/index.cfm/files/serve7File Jd=53a41a78-cQ6c-4695-a7be-
   84225aa7230f

Gross. SA: Avens. HI: Banducci. AM: Sahmel. I: Panko. IM: Tvermoes. BE. (2013). Analysis of BTEX
   groundwater concentrations from surface spills associated with hydraulic fracturing operations. J Air
   Waste Manag Assoc 63: 424-432. http://dx.doi.org/10.1080/10962247.2012.759166

Gu. M: Mohanty. KK.  (2014). Effect of foam quality on effectiveness of hydraulic fracturing in shales.
   International Journal of Rock Mechanics and Mining Sciences 70: 273-285.
   http://dx.doi.0rg/10.1016/j.ijrmms.2014.05.013

Gupta, DVS: Hlidek, BT. (2009). Frac fluid recycling and water conservation:  A case history. In 2009 Hydraulic
   fracturing technology conference. Woodlands, Texas: Society of Petroleum Engineers.
   http://dx.doi.org/10.2118/119478-MS

Gupta. DVS: Valko. P. (2007). Fracturing fluids and formation damage. In M Economides; T Martin (Eds.),
   Modern fracturing: enhancing natural gas production (pp. 227-279). Houston, TX: Energy Tribune
   Publishing Inc.

GWPC (Groundwater Protection Council). (2009). State oil and natural gas regulations designed to protect
   water resources. Morgantown, WV: U.S. Department of Energy, National Energy Technology Laboratory.
   http://www.gwpc.org/sites/default/files/state oil and gas regulations  designed to protect water resou
   rces O.pdf

GWPC and ALL Consulting (Ground Water Protection Council (GWPC) and ALL Consulting). (2009). Modern
   shale gas development in the United States: A primer. (DE-FG26-04NT15455). Washington, DC: U.S.
   Department of Energy, Office of Fossil Energy and National Energy Technology Laboratory.
   http://www.gwpc.org/sites/default/files/Shale%20Gas%20Primer%202009.pdf

Halliburton. (1988). Primer on Hydraulic Fracturing. Provided to EPA on March 2, 2011. Available at Docket
   ID: EPA-HQ-ORD-2010-0674-1634. (HESI-3031). Halliburton.
   http://www.regulations.gov/#!documentDetail:D=EPA-HQ-ORD-2010-0674-1634

Halliburton. (2014). Hydraulic fracturing 101. Available online at
   http://www.halliburton.com/public/projects/pubsdata/hydraulic fracturing/fracturing 101.html

Harms. WM: Yeager. R. (1987). Diesel-based gel concentrate reduces stimulation costs. Oil and Gas Journal 85:
   37-39.

Heermann. SE: Powers. SE. (1998).  Modeling the partitioning of BTEX in water-reformulated gasoline systems
   containing ethanol. J Contam Hydrol 34: 315-341. http://dx.doi.org/10.1016/S0169-7722(98)00099-0
               This document is a draft for review purposes only and does not constitute Agency policy.

June 2015                                       5-80                    DRAFT—DO NOT CITE OR QUOTE

-------
Hydraulic Fracturing Drinking Water Assessment
Holtsclaw. I: Loveless. D: Saini. R: Fleming. I. (20111. SPE 146832: Environmentally-focused crosslinked gel
   system results in high retained proppant-pack conductivity. Presentation presented at Society of
   Petroleum Engineers Annual Conference, November 2,2011, Denver, CO.

Horn. A: Hu. I: Patton. M. (2013). QA/QC of water blending enhances crosslinked gel completions. Available
   online at http://content.stockpr.com/hydrozonix/files/downloads/1013HEP-hydrozonix.pdf

Houston. N: Blauch. M: Weaver. D: Miller. PS: O'Hara. D. (2009). Fracture-stimulation in the Marcellus shale:
   Lessons learned in fluid selection and execution. In 2009 SPE eastern regional meeting: limitless
   potential/formidable challenges. Richardson, TX: Society of Petroleum Engineers.
   http://dx.doi.org/10.2118/125987-MS

Irwin. C. (2013). Hydraulic fracturing: A way to go greener? Available online at
   http://breakingenergy.com/2013/04/23/hydraulic-fracturing-a-way-to-go-greener/

King. GE. (2010). Thirty years of gas shale fracturing: what have we learned? Society of Petroleum Engineers.
   http://dx.doi.org/10.2118/133456-MS

King, GE. (2012). Hydraulic fracturing 101: What every representative, environmentalist, regulator, reporter,
   investor, university researcher, neighbor and engineer should know about estimating frac risk and
   improving frac performance in unconventional gas and oil wells. SPE Hydraulic Fracturing Technology
   Conference, February 6-8, 2012, The Woodlands, TX.

Klein. M: Kenealey. G: Makowecki. B. (2012). Comparison of hydraulic fracture fluids in multi-stage fracture
   stimulated horizontal wells in the Pembina Cardium formation. In 2012 SPE hydrocarbon economics and
   evaluation symposium. Richardson, TX: Society of Petroleum Engineers.
   http://dx.doi.org/10.2118/162916-MS

Knappe, D: Fireline, IL. (2012). Fracking 101: Shale gas extraction using horizontal drilling and hydraulic
   fracturing. Presentation presented at NCAWWA-WEA Annual Conference, November 14, 2012, Raleigh,
   NC.

LePage. I: De Wolf. C: Bemelaar. I: Nasr-El-din. HA. (2013). An environmentally friendly stimulation fluid for
   high-temperature applications. SPE Journal 16:  104-110. http://dx.doi.org/10.2118/121709-PA

Lowe, T: Potts, M: Wood, D. (2013). A case history of comprehensive hydraulic fracturing monitoring in the
   Cana Woodford. In 2013 SPE annual technical conference and exhibition. Richardson, TX: Society of
   Petroleum Engineers. http://dx.doi.org/10.2118/166295-MS

Lustgarten. A. (2009). Frack fluid spill in Dimock contaminates stream, killing fish. Available online at
   http://www.propublica.org/article/frack-fluid-spill-in-dimock-contaminates-stream-killing-fish-921

MacDonald. Rl: Frantz. IH: Schlotterbeck. ST: Adams. B: Sikorski. D.  (2003). An update of recent production
   responses obtained from Devonian shale and Berea wells stimulated with nitrogen foam (with proppant)
   vs. nitrogen-only, Pike Co., KY. Richardson, TX: Society of Petroleum Engineers.
   http://dx.doi.org/10.2118/84834-MS

Malone. M: Ely. IW. (2007). Execution of hydraulic fracturing treatments. In M Economides; T Martin (Eds.),
   Modern fracturing: enhancing natural gas production (pp. 323-360). Houston, TX: ET Publishing.

Martin. T: Valko. P. (2007). Hydraulic fracture design for production enhancement. In M Economides; T
   Martin (Eds.), Modern fracturing enhancing natural gas production. Houston, TX: ET Publishing.

Maule. AL: Makey. CM: Benson. EB: Burrows. II: Scammell. MK. (2013). Disclosure of hydraulic fracturing fluid
   chemical additives: analysis of regulations. New Solutions: A Journal of Environmental and Occupational
   Health Policy 23:167-187. http://dx.doi.Org/10.2190/NS.23.l.i

Methanol Institute. (2013). Methanol safe handling manual. Alexandria, VA.
   http://www.methanol.org/Health-And-Safety/Safe-Handling/Methanol-Safe-Hanlding-Manual.aspx

Mitchell. Bl. (1970) Viscosity of foam. (Doctoral Dissertation). The University of Oklahoma,
               This document is a draft for review purposes only and does not constitute Agency policy.

June 2015                                       5-81                    DRAFT—DO NOT CITE OR QUOTE

-------
Hydraulic Fracturing Drinking Water Assessment
Montgomery. C. (2013). Fracturing fluid components. In A Bunder; J McLennon; R Jeffrey (Eds.), Effective and
   Sustainable Hydraulic Fracturing. Croatia: InTech. http://dx.doi.org/10.5772/56422

NETL (National Energy Technology Laboratory). (2013). Modern shale gas development in the United States:
   An update. Pittsburgh, PA: U.S. Department of Energy. National Energy Technology Laboratory.
   http://www.netl.doe.gov/File%20Library/Research/Oil-Gas/shale-gas-primer-update-2013.pdf

NYSDEC (New York State Department of Environmental Conservation). (2011). Revised draft supplemental
   generic environmental impact statement (SGEIS) on the oil, gas and solution mining regulatory program:
   Well permit issuance for horizontal drilling and high-volume hydraulic fracturing to develop the Marcellus
   shale and other low-permeability gas reservoirs. Albany, NY: NY SDEC.
   http://www.dec.ny.gov/energy/75370.html

Olson. IE. (2011). Hydraulic fracturing overview. Presentation presented at Summer Institute B: Energy,
   Climate and Water in the 21st Century, TXESS Revolution, Texas  Earth and Space Science Revolution
   Professional Development for Educators, June, 2011, Austin, TX.

OSHA (Occupational Safety & Health Administration). (2014a). Personal communication: email exchanges
   between Tandy Zitkus, OSHA and Rebecca Daiss, U.S.  EPA. Available online

OSHA (Occupational Safety & Health Administration). (2014b). Personal communication: phone conversation
   between Tandy Zitkus, OSHA and Rebecca Daiss, U.S.  EPA. Available online

OSHA (Occupational Safety & Health Administration). (2015). Oil and gas well drilling and servicing etool:
   Well completion. Available online at
   https://www.osha.gov/SLTC/etools/oilandgas/well  completion/well completion.html

Papoulias. DM: Velasco. AL. (2013). Histopathological analysis offish from Acorn Fork Creek, Kentucky,
   exposed to hydraulic fracturing fluid releases. Southeastern Naturalist 12: 92-111.

Patel. PS: Robart. Cl: Ruegamer. M: Yang. A. (2014). Analysis of US hydraulic fracturing fluid system and
   proppant trends. In 2014 SPE hydraulic fracturing technology conference. Richardson, TX:  Society of
   Petroleum Engineers. http://dx.doi.org/10.2118/168645-MS

Pearson. CM: Griffin. L: Wright. CA: Weijers. L. (2013). Breaking up is hard to do: creating hydraulic fracture
   complexity in the Bakken central basin. In 2013 SPE hydraulic fracturing technology conference.
   Richardson, TX: Society of Petroleum Engineers. http://dx.doi.org/10.2118/163827-MS

Penttila, B: Heine, L: Craft, E. (2013). Manuscript in preparation assessing the hazard data gap for hydraulic
   fracturing chemicals. Penttila, B; Heine, L; Craft, E.

Phillips. A. (2014). Frackers spill Olympic pools worth of hydrochloric acid in Oklahoma. Available online at
   http://thinkprogress.org/climate/2014/07/31/3466283/olympic-pool-sized-hydrochloric-acid-spill-
   Oklahoma/

Powell, B. (2013). Secondary containment: regulations and best management practices in the Appalachian
   Basin. (AADE-13-FTCE-18). Houston, Texas: American Association of Drilling Engineers.

Powers. SE: Hunt. CS: Heermann. SE: Corseuil. HX: Rice. D: Alvarez. Pll. (2001). The transport and fate of
   ethanol BTEX in groundwater contaminated by gasohol. Environ Sci Technol 31: 79-123.
   http://dx.doi.org/10.1080/20016491089181

Rahm, BG: Vedachalam, S: Bertoia, LR: Mehta, D: Vanka, VS: Riha, SI.  (2015). Shale gas operator violations in
   the Marcellus and what  they tell us about water resource risks. Energy Policy 82:1-11.
   http://dx.doi.0rg/10.1016/i.enpol.2015.02.033

Rasa. E: Bekins. BA: Mackay. DM: de Sieves. NR: Wilson. IT: Feris. KP: Wood. IA: Scow. KM. (2013). Impacts  of
   an ethanol-blended fuel release on groundwater and fate of produced methane: Simulation of field
   observations. Water Resour Res 49: 4907-4926. http://dx.doi.org/10.1002/wrcr.20382
               This document is a draft for review purposes only and does not constitute Agency policy.

June 2015                                       5-82                    DRAFT—DO NOT CITE OR QUOTE

-------
Hydraulic Fracturing Drinking Water Assessment
Rickman. R: Mullen. Ml: Petre. IE: Grieser. WV: Kundert. D. (2008). A practical use of shale petrophysics for
   stimulation design optimization: all shale plays are not clones of the Barnett shale. In 2008 SPE annual
   technical conference & exhibition. Richardson, TX: Society of Petroleum Engineers.
   http://dx.doi.org/10.2118/115258-MS

Rivett. MO: Wealthall. GP: Dearden. RA: McAlary. TA. (2011). Review of unsaturated-zone transport and
   attenuation of volatile organic compound (VOC) plumes leached from shallow source zones [Review]. J
   Contam Hydrol 123:130-156. http://dx.doi.Org/10.1016/i.iconhyd.2010.12.013

Rodvelt. GD: Yeager. VI: Hyatt. MA. (2013). Case history: challenges using ultraviolet light to control bacteria
   in Marcellus completions. In 2011 SPE eastern regional meeting. Richardson, TX: Society of Petroleum
   Engineers. http://dx.doi.org/10.2118/149445-MS

Saba. T: Mohsen. F: Garry. M: Murphy. B: Hilbert. B. (2012). White paper:  Methanol use in hydraulic fracturing
   fluids. (1103844.000 0101 0711 TS26). Maynard, MA: Exponent.

Santa Cruz Biotechnology. (2015). Sorbitane trioleate (CAS 26266-58-0).  Available online at
   http://www.scbt.com/datasheet-281154-Sorbitane-Trioleate.html (accessed April 6, 2015).

Schlumberger (Schlumberger Limited). (2014). Schlumberger oilfield glossary. Available online at
   http://www.glossary.oilfield.slb.com/

Schlumberger (Schlumberger Limited). (2015). Stimulation. Available online at
   http://www.slb.com/services/completions/stimulation.aspx

Schnoor. IL. (1996). Environmental modeling: Fate and transport of pollutants in water, air, and soil. In
   Environmental modeling: Fate and transport of pollutants in water, air, and soil (1 ed.). Hoboken, NJ:
   Wiley-Interscience.

Schwarzenbach. RP: Gschwend. PM: Imboden. DM. (2002). Environmental Organic Chemistry. In
   Environmental organic chemistry (2 ed.). Hoboken,  NJ: John Wiley & Sons, Inc.

Sjolander. SA: Clark. I: Rizzo. D: Turack. I. (2011). Water facts #31: Introduction to hydrofracturing. University
   Park, PA: Penn State College of Agricultural Sciences - Cooperative Extension. http://www.shale-gas-
   information-platform.org/fileadmin/ship/dokumente/introduction to hydrofracturing-2.pdf

Spellman.  FR. (2012). Environmental impacts of hydraulic fracturing.  In Environmental impacts of hydraulic
   fracturing. Boca Raton, Florida: CRC Press.

Stinger Wellhead Protection. I. nc. (2010). Stinger Wellhead Protection. Houston, TX: Stinger Wellhead
   Protection, Inc. http://etdevelopers.com/design-preview/STS/product-
   catalog/STS Product Catalog 2010-SWP.pdf

STO  (Statoil). (2013). Shale facts: drilling and hydraulic fracturing, how it's done, responsibly. (Global
   Version, April 2013). Stavanger, Norway.
   http://www.statoil.com/no/OurOperations/ExplorationProd/ShaleGas/FactSheets/Downloads/Shale  Dr
   illingHydraulicFacturing.pdf

Stringfellow. WT: Domen. IK: Camarillo. MK: Sandelin. WL: Borglin. S.  (2014). Physical, chemical, and
   biological characteristics of compounds used in hydraulic fracturing. J Hazard Mater 275: 37-54.
   http://dx.doi.0rg/10.1016/j.jhazmat.2014.04.040

SWN (Southwestern Energy). (2011). Frac fluid whats in it? Houston, TX.
   http://www.swn.com/operations/documents/frac  fluid fact sheetpdf

SWN (Southwestern Energy). (2014). Field Site Visit at  Southwestern  Energy. Available online

Taylor. RS: Lestz. RS:  Loree. D: Funkhouser.  GP: Fyten. G: Attaway. D: Watkins. H. (2006). Optimized C02
   miscible hydrocarbon fracturing fluids. Calgary, Alberta, Canada: Petroleum Society of Canada.
   http://dx.doi.org/10.2118/2006-168
               This document is a draft for review purposes only and does not constitute Agency policy.

June 2015                                       5-83                    DRAFT—DO NOT CITE OR QUOTE

-------
Hydraulic Fracturing Drinking Water Assessment
Tudor. EH: Nevison. GW: Allen. S: Pike. B. (2009). Case study of a novel hydraulic fracturing method that
   maximizes effective hydraulic fracture length. In 2009 SPE annual technical conference & exhibition.
   Richardson, TX: Society of Petroleum Engineers. http://dx.doi.org/10.2118/124480-MS

U.S. EPA (U.S. Environmental Protection Agency). (1996). Soil screening guidance: technical background
   document, part 2 [EPA Report] (2nd ed.). (EPA/540/R-95/128). Washington, DC: U.S. Environmental
   Protection Agency, Office of Solid Waste and Emergency Response.
   http://www.epa.gov/superfund/health/conmedia/soil/pdfs/part 2.pdf

U.S. EPA (U.S. Environmental Protection Agency). (2012b). Estimation Programs Interface Suite for Microsoft
   Windows (EPI Suite) [Computer Program]. Washington DC: US Environmental Protection Agency.
   Retrieved from http://www.epa.gov/oppt/exposure/pubs/episuitedl.htm

U.S. EPA (U.S. Environmental Protection Agency). (2013a). Data received from oil and gas exploration and
   production companies, including hydraulic fracturing service companies 2011 to 2013. Non-confidential
   business information source documents are located in Federal Docket ID: EPA-HQ-ORD2010-0674.
   Available at http://www.regulations.gov.

U.S. EPA (U.S. Environmental Protection Agency). (2015a). Analysis of hydraulic fracturing fluid data from the
   FracFocus chemical disclosure registry 1.0 [EPA Report]. (EPA/601/R-14/003). Washington, D.C.: Office
   of Research and Development, U.S. Environmental Protection Agency.
   http://www2.epa.gov/hfstudy/analysis-hydraulic-fracturing-fluid-data-fracfocus-chemical-disclosure-
   registry-1-pdf

U.S. EPA (U.S. Environmental Protection Agency). (2015b). Analysis of hydraulic fracturing fluid data from the
   FracFocus chemical disclosure registry 1.0: Project database [EPA Report]. (EPA/601/R-14/003).
   Washington, D.C.: U.S. Environmental Protection Agency, Office of Research and Development.
   http://www2.epa.gov/hfstudy/epa-project-database-developed-fracfocus-l-disclosures

U.S. EPA (U.S. Environmental Protection Agency). (2015J). Retrospective case study in Killdeer, North Dakota:
   study of the potential impacts of hydraulic fracturing on drinking water resources [EPA Report]. (EPA
   600/R-14/103). Washington, D.C.

U.S. EPA (U.S. Environmental Protection Agency). (2015n). Review of state and industry spill data:
   characterization of hydraulic fracturing-related spills [EPA Report]. (EPA/601/R-14/001). Washington,
   D.C.: Office of Research and Development, U.S. Environmental  Protection Agency.

Upstream Pumping. (2015). Upstream pumping: Wellhead technology and services. Available online at
   http://upstreampumping.com/

UWS (Universal Well Services). (2008).  Environmental response plan for field operations. (PATT-EPA-
   0001060). Meadville, PA: Universal Well Services, Inc.

Vidic. RD: Brantley. SL: Vandenbossche. IM: Yoxtheimer. D: Abad.  ID. (2013).  Impact of shale gas development
   on regional water quality [Review]. Science 340:1235009. http://dx.doi.org/10.1126/science.1235009

Waldron. P. (2014). In search of greener fracking for natural gas. Available online at
   http://www.insidescience.org/content/search-greener-fracking-natural-gas/1791

Weinstein. I: Phillippi. M: Walters. HG. (2009). Dry-polymer blending eliminates need for hydrocarbon carrier
   fluids. In 2009 SPE/EPA/DOE E&P Environmental & Safety Conference. Richardson, TX: Society of
   Petroleum Engineers. http://dx.doi.org/10.2118/121002-MS

Wertz. I. (2014). Fracking site operator  faces contempt complaint after acid spill. Available online at
   http://stateimpact.npr.org/oklahoma/2014/08/14/fracking-site-operator-faces-contempt-complaint-
   after-acid-spill/

Yeager. RR: Bailey. DE. (2013). Diesel-based gel concentrate improves Rocky Mountain region fracture
   treatments. Richardson, TX: Society of Petroleum Engineers. http://dx.doi.org/10.2118/17535-MS
               This document is a draft for review purposes only and does not constitute Agency policy.

June 2015                                       5-84                   DRAFT—DO NOT CITE OR QUOTE

-------
Hydraulic Fracturing Drinking Water Assessment
                             Chapter 6
Well  Injection
           This document is a draft for review purposes only and does not constitute Agency policy.
June 2015                                                DRAFT—DO NOT CITE OR QUOTE

-------
      Hydraulic Fracturing Drinking Water Assessment
      6. Well Injection
      6.1.  Introduction
 1    To conduct hydraulic fracturing, fluids (primarily water, mixed with the types of chemicals and
 2    proppant described in Chapter 5) are injected into a well under high pressure.: These fluids flow
 3    under pressure through the well (sometimes thousands of feet below the surface), then exit the
 4    well and move into the formation, where they create fractures in the rock. This process is also
 5    known as a fracture treatment or a type of stimulation.2 The fractures, which typically extend
 6    hundreds of feet laterally from the well, are designed to remain within the production zone to
 7    access as much oil or gas as possible, while using no more water or chemicals than necessary to
 8    complete the operation.3

 9    Production wells are sited and designed primarily to optimize production of oil or gas, which
10    requires isolating water-bearing formations and those containing the hydrocarbons to be exploited
11    from each other. This isolation can also protect drinking water resources. Appropriately sited,
12    designed, constructed, and operated wells and hydraulic fracturing treatments can reduce the
13    potential for impacts to drinking water resources. However, problems with the well's components
14    or improperly sited, designed, or executed hydraulic fracturing operations (or combinations of
15    these) could lead to adverse effects on drinking water resources.

16    The well and the geologic environment in which it is located are a closely linked system, often
17    designed with multiple barriers (i.e., isolation afforded by the well's casing and cement and the
18    presence of multiple layers of subsurface rock) to prevent fluid movement between  oil/gas zones
19    and drinking water resources. Therefore, in this chapter we discuss (1) the well (including its
20    construction and operation) and (2) features in the subsurface geologic formations that could
21    provide or have provided pathways for migration of fluids to drinking water resources. If present
22    and in combination with the existence of a fluid and a physical force that moves the fluid, these
23    pathways can lead to impacts on drinking water resources throughout the life of the well, including
24    during and after hydraulic fracturing.4

25    Fluids can move via pathways adjacent to or through the production well that are created in
26    response to the stresses exerted during hydraulic fracturing operations (see Section 6.2). While
27    wells  are designed and constructed to isolate fluids and maximize the production  of oil and gas,
      i A fluid is a substance that flows when exposed to an external pressure; fluids include both liquids and gases.
      2 In the oil and gas industry, "stimulation" has two meanings—it refers to (1) injecting fluids to clear the well or pore
      spaces near the well of drilling mud or other materials that create blockage and inhibit optimal production (i.e., matrix
      treatment) and (2) injecting fluid to fracture the rock to optimize the production of oil or gas. This chapter focuses on the
      latter.
      3 The "production zone" (sometimes referred to as the target zone) refers to the portion of a subsurface rock zone that
      contains oil or gas to be extracted (sometimes using hydraulic fracturing). "Producing formation" refers to the larger
      geologic unit in which the production zone occurs.
      4 The primary physical force that moves fluids within the subsurface is a difference in pressure. Fluids move from areas of
      higher pressure to areas of lower pressure when a pathway exists. Density-driven buoyancy may also serve as a driving
      force. See Section 6.3.
                    This document is a draft for review purposes only and does not constitute Agency policy.
      June 2015                                     6-1                   DRAFT—DO NOT CITE OR QUOTE

-------
      Hydraulic Fracturing Drinking Water Assessment
 1    inadequate construction or degradation of the casing or cement can allow fluid movement that can
 2    change the quality of drinking water resources. Potential issues associated with wells may be
 3    related to the following:

 4         •   Casing (e.g., faulty, inadequate, or degraded casing or other well components, as
 5            influenced by the numbers of casings; the depths to which these casings are set;
 6            compatibility with the geochemistry of intersected formations; the age of the well;
 7            whether re-fracturing is performed; and other operational factors) and
 8         •   Cement (e.g., poor, inadequate, or degraded cement, as influenced by a lack of cement in
 9            key intervals; poor-quality cement; improper or inadequate placement of cement; or
10            degradation of cement over time).
11    Fluid movement can also occur via induced fractures and/or other features within subsurface
12    formations (see Section 6.3). While the hydraulic fracturing operation may be designed so that the
13    fractures will remain within the production zone, it is possible that, in the execution of the
14    hydraulic fracturing treatment, fractures can extend beyond their designed extent. Four scenarios
15    associated with induced fractures may contribute to fluid migration or communication between
16    zones:

17         •   Flow of injected and/or displaced fluids through pore spaces in the rock formations out of
18            the production zone due to pressure differences and buoyancy effects.
19         •   Fractures extending out of oil/gas formations into drinking water resources or zones that
20            are in communication with drinking water resources or fracturing into zones containing
21            drinking water resources.
22         •   Fractures intersecting artificial structures, including abandoned or active (producing)
23            offset wells near the well that is being stimulated (i.e., well communication) or abandoned
24            or active mines.1
25         •   Fractures intersecting geologic features that can act as conduits, such as existing
26            permeable faults and fractures.
27    In this chapter, we describe the conditions that can contribute to or cause the development of the
28    pathways listed above, the evidence for the existence of these pathways, and potential impacts or
29    impacts on drinking water resources associated with these pathways.

30    The interplay between the well and the subsurface features is complex, and sometimes it is not
31    possible to identify what specific element is contributing to or is the primary cause of an impact to
32    drinking water resources. For example, concerns have been raised regarding stray gas detected in
33    ground water in natural gas production areas (for additional information about stray gas, see
34    Sections 6.2.2 and 6.3.2.4).2 Stray gas migration is a technically complex phenomenon, because
      1 An abandoned well refers to a well that is no longer being used or cannot be used because of its poor condition.
      2 Stray gas refers to the phenomenon of natural gas (primarily methane) migrating into shallow drinking water resources
      or to the surface.
                   This document is a draft for review purposes only and does not constitute Agency policy.
      June 2015                                     6-2                   DRAFT—DO NOT CITE OR QUOTE

-------
      Hydraulic Fracturing Drinking Water Assessment
 1    there are many potential naturally occurring or artificially created routes for migration of gas into
 2    aquifers (including along production wells and via naturally existing or induced fractures), and it is
 3    challenging to determine the source of the natural gas and whether the mobilization is related to oil
 4    or gas production activities.

 5    Furthermore, identifying cases where contamination of drinking water resources occurs due to oil
 6    and gas production activities—including hydraulic fracturing operations—requires extensive
 7    amounts of site and operational data, collected before and after hydraulic fracturing operations.
 8    Where such data does exist and provides evidence of contamination, we present it in the following
 9    sections. We do not attempt to predict which of these pathways is most likely to occur or to lead to
10    a drinking water impact, or the magnitude of an impact that might occur as a result of migration via
11    any single pathway, unless the information is available and documented based on collected data.

      6.2.  Fluid  Migration Pathways Within  and Along the Production Well
12    In this section, we discuss pathways for fluid movement along or through the production well used
13    in the  hydraulic fracturing operation. While these pathways can form at any time within any well,
14    the repeated high pressure stresses exerted during hydraulic fracturing operations may make
15    maintaining integrity of the well more difficult [Council of Canadian Academies. 2014]. In Section
16    6.2.1, we presentthe purpose of the various well components and typical well construction
17    configurations. Section 6.2.2 describes the pathways for fluid movement that can potentially
18    develop within the production well and wellbore and the conditions that lead to pathway
19    development, either as a result of the original design of the well, degradation over time or use, or
20    hydraulic fracturing operations.

21    While we discuss casing and cement separately, it is important to note that these are related—
22    inadequacies in one of these components can lead to stresses on the other. For example, flaws in
23    cement may expose the casing to corrosive fluids. Furthermore, casing and cement work together in
24    the subsurface to form a barrier to fluid movement, and it may not be possible to distinguish
25    whether integrity problems are related to the casing, the cement, or both. For additional
26    information on well design and construction, see Appendix D.

      6.2.1.  Overview of Well Construction
27    Production wells are constructed to convey hydrocarbon resources from the reservoirs in which
28    they are found to the surface and also to isolate fluid-bearing zones (containing oil, gas, or fresh
29    water) from each other. Multiple barriers are often present, and they act together to prevent both
30    horizontal movement (in or out of the well) and vertical movement (along the wellbore from deep
31    formations to drinking water resources). Proper design and construction of the casing, cement, and
3 2    other well components in the context of the location of drinking water resources and maintaining
33    mechanical integrity throughout the life of a well are necessary to prevent migration of fracturing
34    fluids, formation fluids, and hydrocarbons into drinking water resources.
                   This document is a draft for review purposes only and does not constitute Agency policy.
      June 2015                                    6-3                   DRAFT—DO NOT CITE OR QUOTE

-------
      Hydraulic Fracturing Drinking Water Assessment
 1    A well is a multiple-component system that typically includes casing, cement, and a completion
 2    assembly, and it may be drilled vertically, horizontally, or in a deviated orientation.1 These
 3    components work together to prevent unintended fluid movement into, out of, or along the well.
 4    Due to the presence of multiple barriers within the well and the geologic system in which it is
 5    placed, the existence of one pathway for fluid movement does not necessarily mean that an impact
 6    to a drinking water resource has occurred or will occur.

 7    Casing primarily acts as a barrier to lateral movement of fluids, and cement primarily acts as a
 8    barrier to unintended vertical movement of fluids. Together, casing and cement are important in
 9    preventing fluid movement into drinking water resources, and are the focus of this section. Figure
10    6-1 illustrates the configurations of casing and cement that may occur in oil and gas production
11    wells, including the types of casing strings that may be present, the potential locations of cement,
12    and other features. The figure depicts an idealized representation of the components of a
13    production well; it is important to note that there is a wide variety in the design of hydraulically
14    fractured oil and gas wells in the United States [U.S. EPA. 2015o], and the descriptions in the figure
15    or in this chapter do not represent every possible well design.

      6.2.1.1.  Casing
16    Casing is steel pipe that is placed into the drilled wellbore to  maintain the stability of the wellbore,
17    to transport the hydrocarbons from the subsurface to the surface, and to prevent intrusion of other
18    fluids into the well and wellbore [Hyne, 2012: Renpu, 2011]. A long continuous section of casing is
19    referred to as a casing string, which is composed of individual lengths of casing (known as casing
20    joints) that are threaded together using casing collars. In different sections of the well, multiple
21    concentric casing strings (of different diameters) can be used, depending on the construction of the
22    well.

23    The presence of multiple layers of casing strings can isolate and protect geologic zones containing
24    drinking water. In addition to conductor casing, which prevents the hole from collapsing during
2 5    drilling, one  to three other types of casing may be also present in a well. The types of casing include
2 6    (from largest to smallest diameter) surface casing, intermediate casing, and production casing
27    fGWPC. 2014: Hyne. 2012: Renpu. 20111 One or more of any of these types of casing may be
28    present in a well. Surface casing often extends from the wellhead down to the base (bottom) of the
29    drinking water resource to be protected. Wells also may be constructed with liners, which are
30    anchored or suspended from inside the bottom of the previous casing string, rather than extending
31    all the way to the surface, and production tubing, which is used to transport the hydrocarbons to
32    the surface.
      1 For the purposes of this assessment, a well's "orientation" refers to the direction in which the well is drilled, and
      "deviation" is used to indicate an orientation that is neither strictly vertical nor strictly horizontal. However, in industry
      usage, "deviation" is also used as a generic term to indicate well orientation.
                    This document is a draft for review purposes only and does not constitute Agency policy.
      June 2015                                     6-4                   DRAFT—DO NOT CITE OR QUOTE

-------
Hydraulic Fracturing Drinking Water Assessment
      • 12 =
                               12;
Wells used in hydraulic fracturing operations are constructed with
multiple layers of casing and cement to isolate fluid-bearing zones from
each other. While there is no "typical well," some of the more common
elements are presented in this figure. For additional information on the
design and construction of wells, see Appendix D.
Wells are constructed with multiple layers of casing that maintain
wellbore stability, transport hydrocarbons to the surface, and prevent
intrusion of fluids into the well and wellbore (1).
Surface casing (2) extends from the wellhead (3) down to or below the
base (bottom) of the ground water resource (4) to be protected. The
surface casing is cemented from its base to the surface to isolate the
ground water resource and prevent fluid movement. (Wells may also
have a conductor casing (not shown) to prevent unconsolidated material
from collapsing into the wellbore.) The  cement shoe (5) controls the
placement of cement and prevents it from flowing back into the casing
after it has been placed.
When used, intermediate casing (6) can reduce pressure on weak
formations or allow better control of over-pressured formations, and it
extends from the surface through the formation(s) of concern.
The production casing (7) extends to the end of the wellbore in the
production zone (8)—which, in many cases, is thousands of feet below
the surface—and is cemented in place.  In some wells, a production liner is
used in place of production casing. The  production liner is hung from the
next largest casing string by a hanger (9) that is attached to the casing and
typically is cemented in place from the surface. Tubing (10), when used,
conveys hydrocarbons to the surface; it is installed after hydraulic
fracturing operations and is not cemented in place. If the well has an
open hole completion (not shown), the production casing extends just
into the production zone and the entire length of the wellbore through
the production zone  is uncased.
At the bottom of the production casing, a packer (11), a mechanical
sealing device, may beset at the lower  end of the tubing to create and
seal off the annulus (12), the space between the tubing and casing (or
between two casing strings), and to  keep fluid from migrating along the
wellbore. Perforations (13) may be made through the casing and cement,
using small explosive charges; fracturing is conducted through the
perforations. Fracturing fluids are pumped into the well through the
tubing, if present, or through the production casing (if the well has an
open hole completion).
Cement (14) isolates fluid-containing formations from each other,
protects the casing from exposure to formation fluids, and adds strength
to the casing.

 Figure 6-1. Overview of well construction.
                  This document is a draft for review purposes only and does not constitute Agency policy.

June 2015                                               6-5                        DRAFT—DO NOT CITE OR QUOTE

-------
      Hydraulic Fracturing Drinking Water Assessment
 1    Among the wells represented by the Well File Review (see Text Box 6-1), between one and three
 2    casing strings were present (the Well File Review did not evaluate conductor casings). A
 3    combination of surface and production casings was most often reported, followed by a combination
 4    of surface, intermediate, and production strings. All of the production wells used in hydraulic
 5    fracturing operations in the Well File Review had surface casing, while approximately 39% of the
 6    wells (an estimated 9,100 wells) had intermediate casing, and 94% (an estimated 21,900 wells) had
 7    production casing (U.S. EPA. 2015o).

      Text Box 6-1. The Well File Review.

 8    The EPA conducted a survey of onshore oil and gas production wells that were hydraulically fractured by nine
 9    oil and gas service companies in the continental United States between approximately September 2009 and
10    September 2010. The Review of Well Operator Files for Hydraulically Fractured Oil and Gas Production Wells:
11    Well Design and Construction [U.S. EPA, 2015o), referred to as the Well File Review, presents the results of the
12    survey and describes, for these wells: well design and construction characteristics, the relationship of well
13    design and construction characteristics to drinking water resources, and the number and relative location of
14    well construction barriers (i.e., casing and cement) that can block pathways for potential subsurface fluid
15    movement.

16    The results of the survey are based on information provided by well operators for a statistically
17    representative sample of 323 hydraulically fractured oil and gas production wells. The EPA did not attempt to
18    independently and systematically verify data supplied by well operators. Consequently, results from analyses
19    based on these data are of the same quality as the supplied data.

20    Results of the survey are presented as rounded estimates of the frequency of occurrence of hydraulically
21    fractured production well design or construction characteristics with 95% confidence intervals. The results
22    are statistically representative of an estimated 23,200 (95% confidence interval: 21,400-25,000) onshore oil
23    and gas production wells hydraulically fractured in 2009 and 2010 by the nine service companies.

24    Hydraulic fracturing operations impose a variety of stresses on the well components. The casing
25    should be designed with sufficient strength to withstand the stresses it will encounter during the
26    installation, cementing, fracturing, production,  and postproduction phases of the life of the well.
27    These stresses, illustrated in Figure 6-2, include burst pressure (the interior pipe pressure that will
28    cause the casing to burst), collapse pressure (the pressure applied to the outside of the casing that
29    will cause it to collapse), tensile stress (the stress related to stretching exerted by the weight of the
30    casing or tubing being raised or lowered in the hole), compression and bending (the stresses that
31    result from pushing along the axis of the casing or bending the casing), and cyclical stress (the
32    stress caused by frequent or rapid changes in temperature or pressure). Casing strength can be
33    increased by using high-strength alloys or by increasing the thickness of the casing. In addition, the
34    casing must be resistant to corrosion from contact with the formations and any fluids that might be
35    transported through the casing, including hydraulic fracturing fluids, brines, and oil or gas.
                    This document is a draft for review purposes only and does not constitute Agency policy.
      June 2015                                      6-6                    DRAFT—DO NOT CITE OR QUOTE

-------
     Hydraulic Fracturing Drinking Water Assessment
                                           Tensile
                                                      Compressive
                                                          Collapse
                                               b
                                                 M
Burst
                         Note: Figure not to scale
                                                       Stresses Exerted on Well Casings
     Figure 6-2. The various stresses to which the casing will be exposed.
               In addition to the stresses illustrated, the casing will be subjected to bending and cyclical stresses.
               Source: U.S. EPA(2012c).

     6.2.1.2. Cement
1    Cement is one of the most important components of a well for providing zonal isolation and
2    reducing impacts on drinking water. Cement isolates fluid-containing formations from each other,
3    protects the casing from exposure to formation fluids, and provides additional strength to the
4    casing. The strength of the cement and its compatibility with the formations and fluids encountered
5    are important for maintaining well integrity through the life of the well.

6    The cement does not always need to be continuous along the entire length of the well in order to
7    protect drinking water resources; rather, protection of drinking water resources depends on a good
8    cement seal across the appropriate subsurface zones, including all fresh water- and hydrocarbon-
9    bearing zones. One study in the Gulf of Mexico found that there was no breakdown in isolation
                   This document is a draft for review purposes only and does not constitute Agency policy.
     June 2015                                      6-7                   DRAFT—DO NOT CITE OR QUOTE

-------
      Hydraulic Fracturing Drinking Water Assessment
 1    between geologic zones with pressure differentials as high as 14,000 psi as long as there was at
 2    least 50 ft (15 m) of high-quality cement between the zones [King and King. 2013].

 3    Most wells have cement behind the surface casing, which is a key barrier to contamination of
 4    drinking water resources. The surface casings in nearly all of the wells used in hydraulic fracturing
 5    operations represented in the Well File Review (93% of the wells, or an estimated 21,500 wells)
 6    were fully cemented.1 None of the wells studied in the Well File Review had completely uncemented
 7    surface casings.

 8    The length and location of cement behind intermediate and production casings can vary based on
 9    the presence and locations of over-pressured formations, formations containing fluids, or
10    geologically weak formations (i.e., those that are prone to structural failure when exposed to
11    changes in subsurface stresses). State regulations and economics also play a role; 2 5 out of 2 7 oil
12    and gas producing states surveyed by the Ground Water Protection Council require some minimum
13    amount of cementing on the production casing above the producing zone (GWPC. 2014). In general,
14    the intermediate casings of the wells studied in the Well File  Review were fully cemented; while
15    among production casings, about half we re partially cemented, about a third were fully cemented,
16    and the remainder were either uncemented or their cementing status was undetermined. Among
17    the approximately 9,100 wells represented in the Well File Review that are estimated to have
18    intermediate casing, the intermediate casing was fully cemented in an estimated approximately
19    7,300 wells (80%) and partially cemented in an estimated 1,700 wells (19%). Production casings
20    were partially cemented in 47% of the wells, or approximately 10,900 wells (U.S. EPA. 2015o).

21    The Well File Review also estimated the number of wells with a  continuous cement sheath along the
22    outside of the well. An estimated 6,800 of the wells represented in the study (29%) had cement
23    from the bottom of the well to the ground surface, and approximately 15,300 wells (66%) had one
24    or more uncemented intervals between the bottom of the well and the ground surface. In the
25    remaining wells, the location of the top of the cement was uncertain, so no determination could be
26    made regarding whether the well had a continuous cement sheath along the outside of the well
27    fU.S. EPA. 2015oj.

28    A variety of methods are available for placing the cement, evaluating the adequacy of the cementing
29    process and the resulting cement job, and repairing any identified deficiencies. Cement is most
3 0    commonly emplaced by pumping the cement down the inside of the casing to the bottom of the
31    wellbore and then up the space  between the outside of the casing and the formation (or the next
32    largest casing string). This method is referred to as the primary  cement job and can be performed
33    as a continuous event in a single stage (i.e., "continuous cementing") or in multiple stages (i.e.,
34    "staged cementing"). Staged cementing may be used when, for example, the estimated weight and
      i The Well File Review defined fully cemented casings as casings that had a continuous cement sheath from the bottom of
      the casing to at least the next larger and overlying casing (or the ground surface, if surface casing). Partially cemented
      casings were defined as casings that had some portion of the casing that was cemented from the bottom of the casing to at
      least the next larger and overlying casing (or ground surface), but were not fully cemented. Casings with no cement
      anywhere along the casing, from the bottom of the casing to at least the next larger and overlying casing (or ground
      surface), were defined as uncemented.
                   This document is a draft for review purposes only and does not constitute Agency policy.
      June 2015                                     6-8                  DRAFT—DO NOT CITE OR QUOTE

-------
      Hydraulic Fracturing Drinking Water Assessment
 1    pressure associated with standard cement placement could damage weak zones in the formation
 2    [Crook. 20081.

 3    Deficiencies in the cementing process can result from poor centering or lost cement.: Poor
 4    centering of the casing within the wellbore can cause uneven cement placement around the
 5    wellbore, leading to the formation of thin weak spots that are prone to creation of uncemented
 6    channels around the casing and loss of integrity [Kirksey, 2013]. If any deficiencies or defects in the
 7    primary cement job are identified, remedial cementing may be performed.

 8    In over 90% of wells studied in the Well File Review, the casing strings were cemented using
 9    primary cement methods. Secondary or remedial cementing was used on an estimated
10    4,500 casings (8%) most often on surface and production casings and less often on intermediate
11    casings. The remedial cementing techniques employed in these wells included cement squeezes,
12    cement baskets, and pumping cement down the annulus [U.S. EPA. 2015o]. See Appendix D for
13    more information on remedial cementing techniques.

14    A variety of logs are available to  evaluate the quality of cement behind the well casing. Among wells
15    in the Well File Review, the most common type of cement evaluation log run was a standard
16    acoustic cement bond log [U.S. EPA. 2015o]. Standard acoustic cement bond logs are used to
17    evaluate both the extent of the cement placed along the casing and the cement bond between the
18    cement, casing, and wellbore.2 Cement bond indices calculated from standard acoustic cement bond
19    logs on the wells in the Well File Review showed a median bond index of 0.7 just above the
20    hydraulic fracturing zone; this value decreased to 0.4 over a measured distance of 5,000 ft (1,524
21    m) above the hydraulic fracturing zone [U.S. EPA. 2015o]. While standard acoustic cement bond
22    logs can give an average estimate of bonding, they cannot alone indicate zonal isolation, because
23    they may not be properly run or calibrated [Boydetal., 2006:  Smolen, 2006].  One study of 28 wells
24    found that cement bond logs failed to predict communication between formations 11% of the time
25    [Boydetal., 2006]. In addition, they cannot discriminate between full circumferential cement
26    coverage by weaker cement and lack of circumferential coverage by stronger cement [King and
27    King. 2013: Smolen. 2006]. A few studies have compared cement bond indices to zonal isolation,
28    with varying results. For example, Brown etal. [1970] showed that among 16 South American wells
29    with varying casing size and cement bond indices, a cemented 5.5 in (14 cm] diameter casing with a
30    bond index of 0.8 along as little as five feet can act as an effective seal. The authors also suggest that
31    an effective seal in wells having calculated bond indices differing from 0.8 are expected to have an
32    inverse relationship between bond index and requisite length of cemented interval, with longer
33    lengths needed along casing having a lower bond index. Another study recommends that wells
34    undergoing hydraulic fracturing should have a given cement bond over an interval that is three
35    times the length that would otherwise be considered adequate for zonal isolation (Fitzgerald etal.,
      i Lost cement refers to a failure of the cement or the spacer fluid used to wash the drilling fluid out of the wellbore to be
      circulated back to the surface, indicating that the cement has escaped into the formation.
      2 Cement bond logs are used to calculate a bond index, which varies between 0 and 1, with 1 representing the strongest
      bond and 0 representing the weakest bond.
                   This document is a draft for review purposes only and does not constitute Agency policy.
      June 2015                                    6-9                  DRAFT—DO NOT CITE OR QUOTE

-------
      Hydraulic Fracturing Drinking Water Assessment
 1    1985]. Conversely, King and King [2013] concluded that field tests from wells studied by Flournoy
 2    andFeaster [1963] had effective isolation when the cement bond index ranged from 0.31 to 0.75.

      6.2.1.3. Well Orientation
 3    A well can be drilled and constructed with any of several different orientations: vertical, horizontal,
 4    and deviated. The well's orientation can be important, because it affects the difficulty of drilling,
 5    constructing, and cementing the well. In particular, as described below, constructing and cementing
 6    horizontal wells present unique challenges [Sabins, 1990]. In a vertical well, the wellbore is vertical
 7    throughout its entire length, from the wellhead at the surface to the production zone. Deviated
 8    wells are drilled vertically but are designed to deviate from the vertical direction at some point
 9    such that the bottom of the well is at a significant lateral distance away from the point in the
10    subsurface directly under the wellhead. In a horizontal well, the well is drilled vertically to a point
11    known as the kickoff point, where the well turns toward the horizontal, extending into and parallel
12    with the approximately horizontal targeted producing formation (see  Figure 6-1].

13    The use of horizontal wells, particularly in unconventional reservoirs, is increasing [Drillinglnfo,
14    2014b: Valko. 2009]. Among wells evaluated in the Well File Review (i.e., over the period of
15    September 2009 to September 2010], about 66% were vertical, 11% were horizontal, and about
16    23% were deviated wells (U.S. EPA, 2015O].1 This is generally consistent with information available
17    in industry databases—of the approximately 16,000 oil and gas wells used in hydraulic fracturing
18    operations in 2009 (one of the years for which the data for the Well File Review were collected],
19    39% were vertical, 33% were horizontal, and 28% were either deviated or the orientation was
20    unknown (Drillinglnfo. 2014b]. Note that among natural gas wells used in hydraulic fracturing
21    operations, 49.5% of wells in the Drillinglnfo database were horizontal. The use of horizontal wells
22    in hydraulic fracturing operations has also been steadily increasing; in 2012, 63.7% of all wells used
23    in hydraulic fracturing operations were horizontal, compared to just 4% in 2003 (Drillinglnfo,
24    2014b]. See Figure 2-16 for a map presenting the locations of horizontal wells in the United States.

      6.2.1.4. Well Completion
2 5    Another important aspect of well construction is how the well is completed into the production
2 6    zone, because the well's completion is part of the system of barriers and must be intact to provide a
27    fully functioning system.2 A variety of completion configurations  are available. The most common
28    configuration is for casing to extend to the end of the wellbore and be cemented in place (U.S. EPA,
29    2015o: George etal.. 2011: Renpu. 2011]. Before hydraulic fracturing begins, perforations are made
30    through the casing and cement into the production zone. It is through the  perforated casing and
31    cement that hydraulic fracturing is conducted. In some cases, a smaller temporary casing, known as
32    a frac string, is inserted inside the production casing to protect it from the high  pressures imposed
33    during hydraulic fracturing operations. Another method of completion is an open hole completion,
      1 The Well File Review considered any non-horizontal well in which the well bottom was located more than 500 ft (152
      m) laterally from the wellhead as being deviated.
      2 Completion is a term used to describe the assembly of equipment at the bottom of the well that is needed to enable
      production from an oil or gas well. It can also refer to the activities and methods (including hydraulic fracturing) used to
      prepare a well for production following drilling.
                    This document is a draft for review purposes only and does not constitute Agency policy.
      June 2015                                     6-10                   DRAFT—DO NOT CITE OR QUOTE

-------
      Hydraulic Fracturing Drinking Water Assessment
 1    where the production casing extends into the production zone and the entire length of the drilled
 2    horizontal wellbore through the production zone is left uncased. With open hole completions, the
 3    entire production zone can either be fractured all at once in a single stage, or in stages using a
 4    casing string set on formation packers, which separate the annulus into stages.1 A special casing
 5    with retractable sleeves is used to fracture each stage separately. Among wells represented in the
 6    Well File Review, an estimated 6% of wells  (1,500 wells) had open hole completions, 6% of wells
 7    (1,500 wells) used formation packers, and the rest were cased and cemented (U.S. EPA. 2015o).

 8    In some cases, wells may be recompleted after the initial construction, with re-fracturing if
 9    production has decreased (Vincent. 2011). Recompletion also may include additional perforations
10    in the well at a different interval to produce from a new formation, lengthening the wellbore, or
11    drilling new laterals from an existing wellbore.

      6.2.2.  Evidence of the Existence of Fluid Movement Pathways or of Fluid Movement
12    The following sections describe the pathways for fluid movement that can develop within the
13    production well and wellbore. We also describe the conditions that lead to the development of fluid
14    movement pathways and, where available, evidence that a pathway has allowed fluid movement to
15    occur within the casing or cement, and—in  the case of sustained casing pressure (see Section
16    6.2.3)—a combination of factors within the casing and cement. (See Figure 6-3 for an illustration of
17    potential fluid movement pathways related to casing and cement) We describe available
18    information regarding the rate at which these pathways have been identified in hydraulic fracturing
19    wells or, where  such information does not exist, present the results of research on oil and gas
20    production wells in general or on injection wells.2 Insufficient publicly accessible information exists
21    to determine whether wells intended for hydraulic fracturing are constructed differently from
22    production wells where no fracturing is conducted. However, given the applicability of well
23    construction technology to address the subsurface conditions encountered in hydraulic fracturing
24    operations and production or injection operations in general, this information is considered
25    relevant to the research questions (see Section 6.4).

26    While new wells can be specifically designed to withstand the stresses associated with hydraulic
27    fracturing operations, older wells, which are sometimes used in hydraulic fracturing operations,
28    may not have been designed to the same specifications. Where older wells were not designed
29    specifically to withstand the stresses associated with hydraulic fracturing, their reuse for this
30    purpose could be a concern. Frac strings, which are specialized pieces of casing inserted inside the
31    production casing, may be used to protect older casing during fracturing. However, the effect of
32    fracturing on the cement on the production casing in older wells is unknown. One study on re-
33    fracturing of wells noted that the mechanical integrity of the well was a key factor in determining
34    the success or failure of the fracturing (Vincent, 2011). An estimated 6% of wells (1,400 wells)
      1 A formation packer is a specialized casing part that has the same inner diameter as the casing but whose outer diameter
      expands to make contact with the formation and seal the annulus between the casing and formation, preventing
      migration of fluids.
      2 An injection well is a well into which fluids are being injected (40 CFR 144.3).
                    This document is a draft for review purposes only and does not constitute Agency policy.
      June 2015                                     6-11                  DRAFT—DO NOT CITE OR QUOTE

-------
     Hydraulic Fracturing Drinking Water Assessment
1    represented in the Well File Review were more than 10 years old before they were fractured
2    between 2009 and 2010 [U.S. EPA. 2015oj.
     Figure 6-3. Potential pathways for fluid movement in a cemented wellbore.

               These pathways include: (1) casing/tubing leak into a permeable formation, (2) migration along an
               uncemented annulus, (3) migration along microannuli between the casing and cement, (4) migration
               through poor cement, or (5) migration along microannuli between the cement and formation. Note:
               the figure is not to scale and is intended to provide a conceptual illustration of pathways that may
               develop within the well.
                   This document is a draft for review purposes only and does not constitute Agency policy.

     June 2015                                      6-12                   DRAFT—DO NOT CITE OR QUOTE

-------
      Hydraulic Fracturing Drinking Water Assessment
 1    Note that there are also potential issues related to where older wells are sited. For example, some
 2    wells may be in areas with naturally occurring subsurface faults or fractures that could not be
 3    detected or fully characterized with the technologies available at the time of construction. See
 4    Section 6.3.2.

      6.2.2.1. Pathways Related to Well Casing
 5    High pressures associated with hydraulic fracturing operations can damage the casing and lead to
 6    fluid movement that can change the quality of drinking water resources. The casing string through
 7    which fracturing fluids are injected is subject to higher pressures during fracturing operations than
 8    during other phases in the life of a production well. To withstand the stresses created by the high
 9    pressure of hydraulic fracturing, the well and its components must have adequate strength and
10    elasticity. If the casing is compromised or is otherwise not strong enough to withstand these
11    stresses (see Figure 6-2), a casing failure may result. If undetected or not repaired, casing failures
12    will serve as pathways for fracturing fluids to leak out of the casing.1 Below we present indicators
13    that pathways along the casing are present or allowing fluid movement

14    Fracturing fluids or fluids naturally present in the subsurface could flow into other zones in the
15    subsurface if inadequate or no cement is present and the pressure in the casing is greater than the
16    formation pressure. As we describe below, pathways for fluid movement associated with well
17    casing may be related to the original design or construction of the well, degradation of the casing
18    over time, or problems that can arise through extended use as the casing succumbs to stresses.

19    Casing failure may also occur if the wellbore passes through a structurally weak geologic zone that
20    fails and deforms the well casing. Such failures are common when drilling through zones containing
21    salt [Renpu. 2011). This type of well damage may also be possible if hydraulic fracturing causes
22    stress failure along a fault Investigation of a well following a seismic event in England that was
23    attributed to hydraulic fracturing found that the casing had been deformed by the stress from the
24    formation [De Pater and Baisch, 2011]. While it is not known if the casing deformation occurred
25    before  or after the seismic event, such damage is possible if mechanically weak formations are
26    present The changes in the pressure field in the portions of the formation near the wellbore during
27    hydraulic fracturing can also cause mechanically weak formations to fail, potentially damaging the
28    well. Palmer etal.  [2005] demonstrated through modeling that hydraulic fracturing within coal that
29    had a low unconfined compressive strength could cause shear failure of the coalbeds surrounding
30    the wellbore.

31    Corrosion in uncemented zones is the most common cause of casing failure. This can occur if
32    uncemented sections of the casing are exposed to corrosive substances such as brine or hydrogen
33    sulfide [Renpu. 2011]. Corrosion commonly occurs at the collars that connect sections of casing or
      1 Internal mechanical integrity tests (MITs), such as casing inspection logs or caliper logs, annulus pressure monitoring,
      and pressure testing, can provide early warning of a problem, such as a leak, within the casing. It is important to note that
      if a well fails an MIT, this does not mean there is a failure of the well or that drinking water resources are impacted. An
      MIT failure is a warning that something needs to be addressed, and a loss of integrity is an event that may result in fluid
      movement from the well if remediation is not performed.
                    This document is a draft for review purposes only and does not constitute Agency policy.
      June 2015                                     6-13                   DRAFT—DO NOT CITE OR QUOTE

-------
      Hydraulic Fracturing Drinking Water Assessment
 1    at other places where other equipment is attached to the casing. Corrosion at collars may
 2    exacerbate problems with loose or poorly designed connections, which are another common cause
 3    of casing leaks fKing and King. 2013: Brufatto etal.. 20031 Watson and Bachu f20091 found that
 4    66% of all casing corrosion occurred in uncemented well sections, as shown in Pathway 1 of Figure
 5    6-3.

 6    Aging and use of the well contribute to casing degradation. Casing corrosion and degradation can
 7    occur over time and with extended use. Also, exposure to corrosive chemicals such as hydrogen
 8    sulfide, carbonic acid, and brines can accelerate corrosion. Therefore, the potential for fluid
 9    migration related to compromised casing tends to be higher in older wells. Ajani and Kelkar [2012]
10    studied wells in Oklahoma and found a correlation between well age and integrity, specifically in
11    wells spaced between 1,000 and 2,000 ft (305 and 610 m) from a well being fractured, the
12    likelihood of impact to the well rose from approximately 20% to 60% as the well's age increased
13    from 200  days to over 600 days. Age was also found to be a factor in well integrity problems in a
14    study of wells in the Gulf of Mexico [Brufatto etal., 2003]. The studies mentioned did not look for
15    evidence of such fluid movement pathways, however. Because of this potential for degradation of
16    well components in older wells and the related potential for fluid movement, reentering older wells
17    for re-fracturing may contribute to the development of fluid movement pathways within those
18    wells.

19    As noted above, the casing and cement work together to strengthen the well and provide zonal
2 0    isolation.  Uncemented casing does not necessarily lead to fluid migration. However, migration can
21    occur if the casing in an uncemented zone fails during hydraulic fracturing operations or if the
2 2    uncemented section is in contact with fluid-containing zones (including the zone being fractured).
23    Sections of well casing that are uncemented may allow fluid migration into the annulus between the
24    casing and formation. Fluid is free to migrate between formations in contact with the uncemented
2 5    well section in any uncemented annulus without significant hole sloughing or wellbore swelling. If
26    the uncemented section extends through a drinking water resource,  fluid migration into the
27    drinking water resource may occur.

28    Other well integrity problems have been found to vary with the well environment, particularly
29    environments with high pressures and temperatures. Wells in high pressure/high temperature
30    environments, wells with thermal cycling, and wells in corrosive environments can have life
31    expectancies of less than 10 years (King and King. 2013].

32    The depth of the surface casing relative to the base of the drinking water resource to be protected is
33    an important factor in protecting the drinking water resource. In a limited risk modeling study of
34    selected injection wells in the Williston Basin, Michie and Koch (1991] found the risk of aquifer
35    contamination from leaks from the inside of the well to the drinking water resource was 7 in
36    1,000,000 injection wells if the surface casing was set deep enough to cover the drinking water
37    resource,  andthatthe risk increased to 6 in 1,000 wells if the surface casing was notset deeper
38    than the bottom of the drinking water resource. An example where surface casing did not extend
39    below drinking water resources comes from an investigation of 14 drinking water wells with
40    alleged water  quality problems in the Wind River and Fort Union formations in Wyoming

                   This document is a draft for review purposes only and does not constitute Agency policy.
      June 2015                                   6-14                 DRAFT—DO NOT CITE OR QUOTE

-------
      Hydraulic Fracturing Drinking Water Assessment
 1    [WYOGCC, 2014]. The state found that the surface casing of oil and gas wells was shallower than the
 2    depth of 3 of the 14 drinking water wells. Some of the oil and gas wells with shallow surface casing
 3    had elevated gas pressures in their annuli [WYOGCC. 2014).

 4    During hydraulic fracturing operations near Killdeer, in Dunn County, North Dakota, the
 5    production, surface, and conductor casing of the Franchuk 44-20 SWH well ruptured, causing fluids
 6    to spill to the surface  (Jacob, 2011]. The rupture occurred during the 5th of 23 stages of hydraulic
 7    fracturing when the pressure spiked to over 8,390 psi (58 MPa). The casing failed, and ruptures
 8    were found in two locations along the production casing—one just below the surface and one at
 9    about 55 ft (17 m) below ground surface. The surface casing ruptured in three places down to a
10    depth of 188 ft (57 m), and the conductor casing ruptured in one place. Despite a shutdown of the
11    pumps, the pressure was still sufficient to cause fluid to travel through the ruptured casings and to
12    bubble up at the surface. Ultimately, nearly  168,000 gallons (636 m3) of fluids and approximately
13    2,860 tons (2,595 metric tons) of contaminated soil were removed from the site (Jacob, 2011].
14    Sampling of two monitoring wells in the drinking water aquifer identified brine contamination that
15    was consistent with mixing of local ground water with brine from Madison Group formations,
16    which the well had penetrated (U.S. EPA,  2015)]. Tert-butyl alcohol (TEA] was also found in the two
17    wells with brine contamination. The TEA was consistent with degradation of tert-butyl
18    hydroperoxide, a component of the fracturing fluid used in  the Franchuk well. The rupture
19    (blowout] was the only source consistent with findings of high brine and TEA concentrations in the
20    two wells.1 For additional information about impacts at the Killdeer site, see Text Box 5-12 in
21    Section 5.5 and Section 6.3.2.2.

22    Inadequate casing or  cement can respond poorly when blowout preventers activate. When blowout
23    preventers are activated, they immediately  stop the flow in the well, which can create a sudden
24    pressure increase in the well. If the casing or cement are not strong enough to withstand the
25    increased pressure when this occurs, well components can  be damaged (The Royal Society and the
26    Royal Academy of Engineering. 2012] and the potential for  fluid release and migration in the
27    subsurface increases. Blowouts can also occur during the production phase, and cause spills on the
28    surface that can affect drinking water resources; see Section 7.7.3.2.

29    While well construction and hydraulic fracturing techniques continue to change, the pressure- and
30    temperature-related stresses associated with hydraulic fracturing remain as factors that can affect
31    the integrity of the well casing. Several studies have evaluated the components of a well that can
32    affect well performance and integrity. Ingraffea et al. (2014] conducted a study of well integrity in
33    Pennsylvania production wells to assess overall trends in well integrity. This study identified a
34    significant increase in well integrity problems from 2009 to 2011 rising to 5% to 6% of all wells,
35    followed by a decrease beginning in 2012 to about 2% of all wells, a reduction of approximately 100
      i A well blowout is the uncontrolled flow of fluids out of a well. A blowout preventer (BOP) is casinghead equipment that
      prevents the uncontrolled flow of oil, gas, and mud from the well by closing around the drill pipe or sealing the hole [Oil
      and Gas Mineral Services. 2010). BOPs are typically a temporary component of the well, in place only during drilling and
      perhaps through hydraulic fracturing operations.
                   This document is a draft for review purposes only and does not constitute Agency policy.
      June 2015                                    6-15                  DRAFT—DO NOT CITE OR QUOTE

-------
      Hydraulic Fracturing Drinking Water Assessment
 1    violations among 3,000 wells. The rise in well integrity problems between 2009 and 2011 coincided
 2    with an increase in the number of wells in unconventional reservoirs.

 3    Emerging isotopic techniques can be used to identify the extent to which stray gas occurring in
 4    drinking water resources is linked to casing failure (see Text Box 6-2 for more information on stray
 5    gas). Darrahetal. [2014] used hydrocarbon and noble gas isotope data to investigate the source of
 6    gas in eight identified "contamination clusters." Seven of these clusters were stripped of
 7    atmospheric gases (Ar36 and Ne20) and were enriched in crustal gases, indicating the gas migrated
 8    quickly from depth without equilibrating with intervening formations. The rapid transport was
 9    interpreted to mean that the migration did not occur along natural fractures or pathways, which
10    would have allowed equilibration to take place. Possible explanations for the rapid migration
11    include transport up the well and through a leaky casing (Pathway 1 in Figure 6-3) or along
12    uncemented or poorly cemented intervals from shallower depths (Pathways 2 through 5 in Figure
13    6-3). In four Marcellus Shale clusters, gas found in drinking water wells had isotopic signatures and
14    ratios of ethane to methane that were consistent with those in the producing formation. The
15    authors conclude that this suggests that gas migrated along poorly constructed wells from the
16    producing formation,  likely with improper,  faulty, or failing production casings. In three clusters,
17    the isotopic signatures and ethane to methane ratios were consistent with formations overlying the
18    Marcellus. The authors suggest that this migration occurred from the shallower gas formations
19    along uncemented or  improperly cemented wellbores. In another Marcellus cluster in the study,
20    deep gas migration was linked to a subsurface well, likely from a failed well packer.
21
22
23
24
25
26
27
28

29
30
31
32
33
34
35
36
37
38
39
Text Box 6-2. Stray Gas Migration.

Stray gas refers to the phenomenon of natural gas (primarily methane) migrating into shallow drinking water
resources, into water wells, or to the surface (e.g., cellars, streams, or springs). Stray gas in the wellhead of a
production well is an indicator of an active wellbore pathway. Methane is not a regulated drinking water
contaminant, but it can initiate chemical and biological reactions that release or mobilize other contaminants,
and gas can accumulate to explosive levels when allowed to exsolve (degas) from ground water in closed
environments. Stray gas may originate from conventional and unconventional natural gas reservoirs, as well
as from coal mines, landfills, leaking gas wells, leaking gas pipelines, and buried organic matter (Baldassare.
20111.

Detectable levels of dissolved hydrocarbons (generally methane and/or ethane) exist in most oxygen-poor
aquifers, even in the absence of human activity (Gorody. 2012). Pre-drilling studies show that low levels of
methane are frequently found in water wells in northern Pennsylvania and New York (Kappel. 2013: Kappel
and Nystrom, 2012): one USGS study detected methane in 80% of sampled wells in Pike County, Pennsylvania
(Senior. 2014). The origin of methane in ground water can be either thermogenic (produced by high
temperatures and pressures in deeper formations, such as the gas found in the Marcellus Shale) or biogenic
(produced in shallower formations by bacterial activity in anaerobic conditions). Occurrence of thermogenic
methane in shallower formations depends upon the existence of a hydrocarbon source and pathways by
which the gas can migrate. Interactions between hydrocarbon production and natural systems must also be
considered. For example, Brantley et al. (2014) describe how northern Pennsylvania's glacial history may
help explain why stray gas is more common there than in the southern part of the state.
                    This document is a draft for review purposes only and does not constitute Agency policy.
      June 2015                                     6-16                   DRAFT—DO NOT CITE OR QUOTE

-------
      Hydraulic Fracturing Drinking Water Assessment
 1    Stray gas migration can be a technically complex phenomenon, in part because there are many potential
 2    routes for migration. The lack of detailed monitoring data, including a lack of baseline measurements prior to
 3    hydrocarbon production, often further complicates stray gas research. Examining the concentrations and
 4    isotopic compositions of methane and higher molecular weight hydrocarbons such as ethane and propane
 5    can aid in determining the source of stray gas [Tilley and Muehlenbachs. 2012: Baldassare. 2011: Rowe and
 6    Muehlenbachs. 1999]. Isotopic composition and methane-to-ethane ratios can help determine whether the
 7    gas is thermogenic or biogenic in origin and whether it is derived from shale or other formations [Gorody.
 8    2012: Muehlenbachs et al.. 2012: Barker and  Fritz. 1981]. Isotopic analysis can also be used to identify the
 9    strata where the gas originated and provide evidence for migration mechanisms [Darrah et al.. 2014].

10    However, determining the source of methane does not necessarily establish the migration pathway. Well
11    casing and cementing issues may be an important source of stray gas problems [lackson et al.. 2013b]:
12    however, other potential subsurface pathways are also discussed in the literature. Multiple researchers (e.g.,
13    lackson etal., 2013b: Molofsky etal., 2013: Revesz etal., 2012: Osborn etal., 2011] have described biogenic
14    and/or thermogenic methane in ground water supplies in Marcellus gas production areas, although the
15    pathways of migration are generally not apparent. The Osborn etal. [2011] study found that thermogenic
16    methane concentrations in well water increased with proximity to Marcellus Shale production sites, while
17    Molofsky et al. [2013] found the presence of gas to be more closely correlated with topography and elevation.
18    Similarly, Siegel et al. [In Press] found no correlation between methane in ground water and proximity to
19    production wells.

2 0    The EPA conducted case studies that included investigating stray gas in northeastern Pennsylvania and the
21    Raton Basin  of Colorado. In northeastern Pennsylvania, many  drinking water wells within the study area
22    were found to have elevated methane concentrations.  For some of the wells, the EPA concluded that the
23    methane (both thermogenic and biogenic] was naturally occurring background gas not attributable to gas
24    exploration activities. In others, it appeared that non-background methane had entered the water wells
25    following well drilling and hydraulic fracturing. In most cases, the methane in  the wells likely originated from
26    intermediate formations between the production zone and the surface; however, in some cases, the methane
27    appears to have originated from deeper layers such as those where the Marcellus Shale is found [U.S. EPA.
28    2015o]. The Raton Basin case study examined the Little Creek Field, where potentially explosive quantities of
29    methane were vented into a number of drinking water wells in 2007. The methane was found to be primarily
30    thermogenic in origin, modified by biologic oxidation [U.S. EPA. 20151]. Secondary biogeochemical changes
31    related to the migration and reaction of methane within the shallow drinking water aquifer were reflected in
3 2    the characteristics of the Little Creek Field ground water [U.S.  EPA. 20151].

3 3    While the sources of methane in the two studies could be determined with varying degrees of certainty,
34    attempts to definitively identify the pathways of migration have been inconclusive. In northeastern
35    Pennsylvania, while the sources could not be  definitively determined, the Marcellus Shale could not be
36    excluded as a potential source in some wells based on  isotopic signatures, methane-to-ethane ratios, and
37    isotope reversal properties [U.S. EPA. 2014J]. The Pennsylvania
                     This document is a draft for review purposes only and does not constitute Agency policy.
      June 2015                                       6-17                   DRAFT—DO NOT CITE OR QUOTE

-------
      Hydraulic Fracturing Drinking Water Assessment
 1    Department of Environmental Protection (PA DEP) cited at least two operators for failure to prevent gas
 2    migration at a number of wells within the study area. Evidence cited by the state included isotopic
 3    comparison of gas samples from drinking water wells, water bodies, and gas wells; inadequate cement jobs;
 4    and sustained casing pressure (although, under Pennsylvania law, oil or gas operators can be cited if they
 5    cannot disprove the contamination was caused by their well using pre-drilling samples). A separate study
 6    (Ingraffea et al.. 2014] showed that wells in this area had higher incidences of well integrity problems relative
 7    to wells in other parts of Pennsylvania. In the Raton Basin, the source of methane was identified as the
 8    Vermejo coalbeds. Based on modeling of the Raton Basin study area, the migration could be explained either
 9    by migration along natural rock features in the area or by migration caused by fracturing around the wellbore
10    (U.S. EPA, 201511. Because the gas production wells were shut in shortly after the incident began, the wells
11    could not be inspected to determine whether an integrity failure in the wellbore was a likely cause of the
12    migration.1

13    These two case studies illustrate the considerations involved with understanding stray gas migration and the
14    difficulty in determining sources and migration pathways. In order to more conclusively determine sources
15    and migration pathways, studies in which data are collected on well integrity and ground water methane
16    concentrations both before and after hydraulic fracturing operations, in addition to the kinds of data
17    summarized above, would be needed.

      6.2.2.2. Pathways Related to Cement
18    Fluid  movement may result from inadequate well design or construction (e.g., cement loss or other
19    problems that arise in cementing of unconventional wells) or degradation of the cement over time
20    (e.g., corrosion or the formation of microannuli), which may, if undetected and not repaired, cause
21    the cement to succumb to the stresses exerted during hydraulic fracturing.2'3 The well cement must
22    be able to withstand the  subsurface conditions and the stresses encountered during hydraulic
23    fracturing operations. In this section, we present indicators that pathways within the cement are
24    present or allowing fluid movement

25    Uncemented zones can allow fluids or brines to move into drinking water resources. An improper
26    cement job can fail to maintain zonal isolation in several ways. The first is by poor cement
27    placement If a fluid-containing zone is left uncemented, the open annulus between the formation
28    and casing can act as a pathway for migration of that fluid. Fluids can enter the wellbore along any
29    uncemented  section of the wellbore if a sufficient pressure gradient is present Once the fluids have
30    entered the wellbore, they can travel along the entire uncemented length of the  wellbore, unless the
31    wellbore sloughs in around the casing.
32    Because of its low density, gas will migrate up the wellbore if an uncemented wellbore is exposed to
33    a gas-containing formation. Gas may then be able to enter other formations (including drinking
      1 Shutting in a well refers to sealing off a well by either closing the valves at the wellhead, a downhole safety valve, or a
      blowout preventer.
      2 External MITs, such as temperature logs, noise logs, oxygen activation logs, and radioactive tracer logs, can indicate
      improper cementing or degradation of the cement.
      3 Microannuli are very small channels that form in the cement and that may serve as pathways for fluid migration to
      drinking water resources.
                    This document is a draft for review purposes only and does not constitute Agency policy.
      June 2015                                     6-18                   DRAFT—DO NOT CITE OR QUOTE

-------
      Hydraulic Fracturing Drinking Water Assessment
 1    water resources) if the wellbore is uncemented and the pressure in the annulus is sufficient to force
 2    fluid into the surrounding formation [Watson and Bachu. 2009]. The rate at which the gas can move
 3    will depend on the difference in pressure between the annulus and the formation [Wojtanowicz.
 4    2008]. Harrison [1985] also demonstrated that such uncemented annuli could result in the
 5    migration of gas into overlying formations open to the annulus.

 6    In several cases, poor or failed cement has been linked to stray gas migration (see Text Box 6-2]. A
 7    Canadian study found that uncemented portions of casing were the most significant contributors to
 8    gas migration [Watson and Bachu. 2009]. The same study also found that 57% of all casing leaks
 9    occurred in uncemented segments. In the study by Darrah et al. [2014] [see Section 6.2.2.1] using
10    isotopic data, four clusters of gas contamination were linked to cement issues. In three clusters in
11    the Marcellus and one in the Barnett, gas found in drinking water wells had isotopic signatures
12    consistent with intermediate formations overlying the producing zone. This suggests that gas
13    migrated from the intermediate units along the well annulus, along uncemented portions of the
14    wellbore or through channels or microannuli. In a study in Utah, Heilweil etal. [2013] used a
15    stream-based methane monitoring method to identify a case of potential gas migration near a well
16    with defective casing or cement

17    Cementing of the surface casing is the primary aspect of well construction that protects drinking
18    water resources. Most states require the surface casing to be set and cemented from the level of the
19    lowermost drinking water resource to the surface [GWPC. 2014]. Most wells—including those used
20    in hydraulic fracturing operations—have such cementing in place. Among the wells represented in
21    the Well File Review, surface casing, which was found to be fully cemented in 93% of wells,
22    extended below the base (i.e., the bottom, deepest, or lowermost part] of the protected ground
23    water resource reported by well operators in an estimated 55% of wells (12,600 wells].1 In an
24    additional 28% of wells (6,400 wells], the operator-reported protected ground water resources
25    were fully covered by the next cemented casing string. A portion of the annular space between
26    casing and the operator-reported protected ground water resources was uncemented in at least 3%
27    of wells (600 wells] (U.S. EPA. 2015o].2 Improper placement of cement can lead to cement integrity
28    problems. For example, an improper cement job can be the result of loss of cement during
29    placement into a formation with high porosity or fractures, causing a lack of adequate cement
30    across a water- or brine-bearing zone. Additionally, failure to use cement that is compatible with
31    the anticipated subsurface conditions, failure to remove  drilling fluids from the wellbore, and
32    improper centralization of the casing in the wellbore  can all lead to the formation of channels (i.e.,
33    small connected voids] in the cement during the cementing process [McDaniel et al., 2 014: Sabins,
34    1990]. If the channels are small and isolated, they may not lead to fluid migration. However, if they
      1 In the in the Well File Review, protected ground water resources were reported by well operators. For most wells
      represented in the Well File Review, protected ground water resources were identified by the well operators from state
      or federal authorization documents. Other data sources used by well operators included aquifer maps, data from offset
      production wells, open hole log interpretations done by operators, operator experience, online databases, and references
      to a general requirement by the oil and gas agency.
      2 The well files representing an estimated 8% of wells in the Well File Review did not have sufficient data to determine
      whether the operator-reported protected ground water resource was uncemented or cemented. In these cases, there was
      ambiguity either in the depth of the base or the top of the operator-reported protected ground water resource.
                    This document is a draft for review purposes only and does not constitute Agency policy.
      June 2015                                     6-19                   DRAFT—DO NOT CITE OR QUOTE

-------
      Hydraulic Fracturing Drinking Water Assessment
 1    are long and connected, extending across multiple formations or connecting to other existing
 2    channels or fractures, they can present a pathway for fluid migration. Figure 6-3 shows a variety of
 3    pathways for fluid migration that are possible from failed cement jobs.

 4    One example of how cement problems associated with hydraulic fracturing contributed to a
 5    contamination incident occurred in Bainbridge, Ohio. This incident was particularly well studied by
 6    the Ohio Department of Natural Resources [ODNR, 2008] and by an expert panel [Bair etal., 2010].
 7    The level of detail available for this case is not typically found in studies of such events but was
 8    collected because of the severity of the impacts and the resulting legal action. The English #1 well
 9    was drilled to a depth of 3,900 ft (1,189 m] below ground surface (bgs] with the producing
10    formation located between 3,600 and 3,900 ft (1,097 and 1,189 m] bgs. Overlying the producing
11    formation were several uneconomic formations that contained over-pressured gas (i.e., gas at
12    pressures higher than the hydrostatic pressure exerted by the fluids within the well].1 The original
13    cement design required the cement to be placed 700-800 ft (213-244 m] above the producing
14    formation to seal off these areas. During cementing, however, the spacer fluid failed to return to the
15    surface, indicating lost cement, and the cement did not reach the intended height.2 Despite the lack
16    of sufficient cement, the operator proceeded with hydraulic fracturing.

17    During the fracturing operation, about 840 gallons (3.2 m3] of fluid flowed up the annulus and out
18    of the well. When the fluid began flowing out of the annulus, the operator immediately ceased
19    operations. About a month later, there was an explosion in a nearby house where methane had
2 0    entered from an abandoned and unplugged drinking water well that was connected to the cellar
21    (Bair etal.. 2010]. In addition to the explosion, the over-pressured gas entering the aquifer resulted
22    in the contamination of 26 private drinking water wells with methane.

2 3    Contamination at Bainbridge was the result of inadequate cement  The ODNR determined that
24    failure to cement the over-pressured gas formations, proceeding with the fracturing operation
2 5    without verifying there was adequate cement, and the extended period during which the well was
26    shut in all contributed to  the contamination of the aquifer with stray gas (ODNR. 2008]. Cement
27    logs found the cement top was at 3,640 ft (1,109 m] bgs, leaving the uneconomic gas-producing
28    formations and a portion of the production zone uncemented. Hydraulic fracturing fluids flowing
29    out of the annulus provided an indication that the fracturing had created a path from the producing
30    formation to the well annulus. This pathway may have allowed gas from the producing formation
31    along with gas from the uncemented formation to enter the annulus. Because the well was shut in,
32    the pressure in the annulus could not be relieved, and the gas eventually traveled through natural
33    fractures surrounding the wellbore into local drinking water aquifers. (During the time the well
34    was shut in, natural gas seeped into the well annulus and pressure built up from an initial pressure
35    of 90 psi (0.6 MPa] to  360 psi (2.5 MPa]]. From the aquifer, the gas moved into  drinking water wells
36    and from one of those wells into a cellar, resulting in the explosive  accumulation of gas.
      1 Hydrostatic pressure is the pressure exerted by a column of fluid at a given depth. Here, it refers to the pressure exerted
      by a column of drilling mud or cement on the formation at a particular depth.
      2 Spacer fluid is a fluid pumped before the cement to clean drilling mud out of the wellbore.
                   This document is a draft for review purposes only and does not constitute Agency policy.
      June 2015                                    6-20                  DRAFT—DO NOT CITE OR QUOTE

-------
      Hydraulic Fracturing Drinking Water Assessment
 1    While limited literature is available on construction flaws in wells that have been used in hydraulic
 2    fracturing operations, several studies have examined construction flaws in oil and gas wells in
 3    general. One study that examined reported drinking water contamination incidents in Texas
 4    identified 10 incidents related to drilling and construction activities among 250,000 oil and gas
 5    wells [Kell. 2011]. The study noted that many of the contamination incidents were associated with
 6    wells that were constructed before Texas revised its regulations on cementing in 1969 (it is not
 7    clear how old the wells were at the time the contamination occurred). Because this study relied on
 8    reported incidents, it is possible that other wells exhibited integrity issues but did not result in
 9    contamination of a drinking water well or were not reported. Therefore, this should be considered
10    a low-end estimate of the number of well integrity issues that could be tied directly to drilling and
11    construction activities. It is important to note that the 10 contamination incidents identified were
12    not associated with wells that were hydraulically fractured [Kell. 2011).

13    Several investigators have studied violations information from the PA DEP online violation
14    database to evaluate the rates of and possible factors contributing to well integrity problems,
15    including those related to cement. The results of these studies are summarized in Table 6-1. While
16    all of the studies shown in the table used the same database, their results vary, not only because of
17    the different time frames, but also because they used different definitions of what violations
18    constituted an integrity problem or failure. For example, Considine etal. [2012] considered all
19    events resulting in environmental damage—including effects such as erosion—and found a
2 0    relatively high violation rate. Davies etal. [2014] and Ingraffea et al.  [2014] investigated violations
21    related to well integrity, while Vidicetal. [2013] looked only at well integrity violations that
22    resulted in fluid migration out of the wellbore; these more specific studies found relatively lower
23    violation rates. Olawoyin et al. [2013] performed a statistical analysis that weighted violations
24    based on risk and found that the most risky violations included those involving pits, erosion, waste
25    disposal, and blowout preventers.

      Table 6-1. Results of studies of PA DEP violations data that examined well failure rates.
Study
Violations investigated
Data timeframe
Key findings3
                    This document is a draft for review purposes only and does not constitute Agency policy.
      June 2015                                     6-21                  DRAFT—DO NOT CITE OR QUOTE

-------
      Hydraulic Fracturing Drinking Water Assessment
Study
Considine et al.
(2012)
Davies et al.
(2014)
Ingraffea et al.
(2014)
Vidicetal.
(2013)
Olawovin et al.
(2013)
Violations investigated
Violations resulting in
environmental damage
(3,533 wells studied)
Failure of one of the
barriers preventing fluid
migration (8,030 wells
studied)
Violations and inspection
records indicating
structural integrity loss
(3,391 wells studied)
Failure of cement/casing
that allowed fluid to leak
out (6,466 wells studied)
All violations (2,001 wells
studied)
Data timeframe
2008-2011
2005-2013
2000-2012
2008-2013
2008-2010
Key findings3
Of 845 environmental damage incidents (which
resulted in 1,144 violations), approximately 10%
were related to casing or cement problems. The
overall violation rate dropped from 52.9% of all
wells in 2008 to 20.8% of all wells in 2011.
Approximately 5% of wells received this type of
violation. The incident rate increased to 6.3%
when failures noted on forms, but not resulting
in violations, were included.
Wells in unconventional reservoirs experienced
a rate of structural integrity loss of 6.2%, while
the rate for conventional wells was 1%.
Approximately 3.4% of wells received this type
of violation.
Analysis of 2,601 violations from 65 operators
based on weighted risks found that potentially
risky violations increased 342% over the study
period, while total violations increased 110%.
      a Note: While all of these studies used the same database, their results vary because they studied different time frames and
      used different definitions of what violations constituted an integrity problem or failure.
 1    Because a significant portion of Pennsylvania's recent oil and gas activity is in the Marcellus Shale,
 2    many of the wells in these studies were most likely used for hydraulic fracturing. For example,
 3    Ingraffea et al. [2014] found that approximately 16% of the oil and gas wells drilled in the state
 4    between 2000 and 2012 were completed in unconventional reservoirs, and nearly all of these wells
 5    were used for hydraulic fracturing. Wells drilled in unconventional reservoirs experienced higher
 6    rates of structural integrity loss than conventional wells drilled during the same time period
 7    [Ingraffea etal., 2014]. The authors did not compare rates of structural integrity loss in
 8    conventional wells that were and were not used for hydraulic fracturing; they assumed that
 9    unconventional wells were hydraulically fractured and conventional wells were not

10    Violation rates resulting in environmental  pollution among all wells dropped from 52.9% in 2008
11    to 20.8% in 2011 [Considine etal.. 2012]. and the drop may be due to a number of factors.
12    Violations related to failure of cement or other well components represented a minority of all well
13    violations. Of 845 events that caused environmental contamination, including but not limited to
14    contamination of drinking water resources, Considine etal. [2012] found that about 10% (85
                    This document is a draft for review purposes only and does not constitute Agency policy.
      June 2015                                     6-22                  DRAFT—DO NOT CITE OR QUOTE

-------
      Hydraulic Fracturing Drinking Water Assessment
 1    events) were related to casing and cement problems. The rest of the incidents were related to site
 2    restoration and spills (the violations noted are confined to those incidents that caused
 3    environmental damage i.e., the analysis excluded construction flaws that did not have adverse
 4    environmental effects).

 5    Another source of information on contamination caused by wells is positive determination letters
 6    (PDLs) issued by the PA DEP. PDLs are issued in response to a complaint when the state determines
 7    that contamination did occur in proximity to oil and gas activities. The PDLs take into account the
 8    impact, timing, well integrity, and formation permeability; however, liability is presumed for wells
 9    within a given distance if the oil and gas operator cannot refute that they caused the contamination,
10    based on pre-drilling sampling [Brantley etal.. 2014).' Brantley et al. [2014) examined these PDLs,
11    and concluded that approximately 2 0 unconventional gas wells impacted water supplies between
12    2008 and 2012; this equates to 0.1% to 1% of the 6,061 wells spudded between  2008 and 2012 (it
13    is unclear exactly how many PDLs are linked to an individual well). While these oil and gas wells
14    were linked to contamination of wells and springs, the mechanisms for the impacts (including
15    whether fluids may have been spilled at the surface or if there was a pathway through the well or
16    through the subsurface rock formation to the drinking water resource) were not described by
17    Brantley et al. (2014). We did not perform a full and independent review of the PDLs for this
18    assessment

19    While the studies discussed above present possible explanations for higher violation incidences in
2 0    hydraulically fractured wells, it should be noted that other explanations that are not specific to
21    hydraulic fracturing are also possible. These could include different inspection protocols and
2 2    different formation types.

23    Cementing in horizontal wells, which are commonly hydraulically fractured, presents challenges
24    that can contribute to higher rates of integrity issues. The observation by Ingraffea et al. (2014) that
25    wells drilled in unconventional reservoirs experience higher rates of structural integrity loss than
26    conventional wells is supported by conclusions of Sabins (1990), who noted that horizontal wells
27    have more cementing problems because they are more difficult to center properly and can be
28    subject to settling of solids on the bottom of the wellbore. Cementing in horizontal wells presents
29    challenges that can contribute to higher rates of integrity issues.

30    Thermal and cyclic stresses caused by intermittent operation also may stress cement (King and
31    King. 2013: Ali etal.. 2009). Increased pressures and cyclic stresses associated with hydraulic
32    fracturing operations can contribute to cement integrity losses and, if undetected, small integrity
33    problems can lead to larger ones. Temperature differences between the (typically warmer)
34    subsurface environment and the (typically cooler) injected fluids, followed by contact with the
35    (typically warmer)  flowback water, can lead to contraction of the well materials  (both casing and
36    cement), which introduces additional stresses. Similar temperature changes may occur when
      1 Under Pennsylvania's Oil and Gas Act, operators of oil or gas wells are presumed liable if water supplies within 1,000 ft
      (305 m) were impacted within 6 months of drilling, unless the claim is rebutted by the operator; this was expanded to
      2,500 ft (762 m) and 12 months in 2012.
                   This document is a draft for review purposes only and does not constitute Agency policy.
      June 2015                                    6-23                  DRAFT—DO NOT CITE OR QUOTE

-------
      Hydraulic Fracturing Drinking Water Assessment
 1    multiple fracturing stages are performed. Because the casing and cement have different mechanical
 2    properties, they may respond differently to these stress cycles and debond.

 3    Several studies illustrate the effects of cyclic stresses. Dusseault et al. [2000] indicate that wells that
 4    have undergone several cycles of thermal or pressure changes will almost always show some
 5    debonding between cement and casing. Microannuli formed by this debonding can be conduits for
 6    gas migration, in particular because the lighter density of gas provides a larger driving force for
 7    migration through the microannuli than for heavier liquids.: One laboratory study indicated that
 8    microannuli on the order of 0.01 in (0.3 mm) could increase effective cement permeability from
 9    1 nD (1 x 10-21 m2) in good quality cement up to 1 mD (1 x IQ-IS m2) [Bachu and Bennion. 2009].
10    This six-order magnitude increase in permeability shows that even small microannuli can
11    significantly increase the potential for flow through the cement Typically, these microannuli form
12    at the interface between the casing and cement or between the cement and formation. Debonding
13    and formation of microannuli can occur through intermittent operation, pressure tests, and
14    workover operations [Dusseault etal., 2000].2 While a small area of debonding may not lead to fluid
15    migration, the microannuli in the cement that result from the debonding can serve as initiation
16    points for fracture propagation if re-pressurized gas enters the microannulus [Dusseault etal.,
17    20001

18    The Council of Canadian Academies [2014] found that the repetitive pressure surges that occur
19    during the fracturing process would make maintaining an intact cement seal more of a challenge in
20    wells that are hydraulically fractured. Wang and Dahi Taleghani [2014] performed a modeling
21    study that showed that hydraulic fracturing pressures could initiate annular cracks in cement.
22    Another study of well data indicated that cement failure rates are higher in intermediate casings
23    compared to other casings [McDaniel etal., 2014]. The failures occurred after drilling and
24    completion of wells, and the authors surmised that the cement failures were most likely due to
25    cyclical pressure stresses caused by drilling. Theoretically, such cyclical pressure events could also
26    be experienced during multiple stage hydraulic fracturing. Mechanical stresses associated with well
27    operation or workovers and pressure tests also may lead to small cracks in the cement, which may
28    provide migration pathways for fluid.

29    Corrosion can lead to cement failure. Cement can fail to maintain integrity as a result of degradation
30    of the cement after the cement is set Cement degradation can result from attack by corrosive brines
31    or  chemicals such as sulfates, sulfides, and carbon dioxide that exist in formation fluids [Renpu,
32    2011]. These chemicals can alter the chemical structure of the cement, resulting in increased
33    permeability or reduced strength and leading to loss of cement integrity over time. Additives or
34    specialty cements that can decrease cement susceptibility to specific chemical components are
35    available.
      1 Microannuli can also form due to an inadequate cement job, e.g., poor mud removal or improper cement placement rate.
      2 A workover refers to any maintenance activity performed on a well that involves ceasing operations and removing the
      wellhead. Depending on the purpose of the workover and the tools used, workovers may induce pressure changes in the
      well.
                   This document is a draft for review purposes only and does not constitute Agency policy.
      June 2015                                    6-24                  DRAFT—DO NOT CITE OR QUOTE

-------
      Hydraulic Fracturing Drinking Water Assessment
      6.2.2.3. Sustained Casing Pressure
 1    An example of how the issues related to casing and cement discussed in the preceding sections can
 2    work together and be difficult to differentiate can be seen in the case of sustained casing pressure.:
 3    Sustained casing pressure is an indicator that pathways within the well related to the well's casing,
 4    cement, or both allowed fluid movement to occur. Sustained casing pressure can result from casing
 5    leaks, uncemented intervals, microannuli, or some combination of the three, which can be an
 6    indication that a well has lost integrity. Sustained casing pressure can be observed when an annulus
 7    (either the annulus between the tubing and production casing or between any two casings) is
 8    exposed to a source of nearly continuous elevated pressure. Goodwin and Crook [1992] found that
 9    sudden increases in sustained casing pressure occurred in wells that were exposed to high
10    temperatures and pressures. Subsequent logging of these wells showed that the high temperatures
11    and pressures led to shearing of the cement/casing interface and a total loss of the cement bond.

12    Sustained casing pressure occurs more frequently in older wells and horizontal or deviated wells.
13    One study found that sustained casing pressure becomes worse as a well ages. Sustained casing
14    pressure was found in less than 10% of wells that were less than a year old, but was present in up
15    to 50% of 15-year-old wells [Brufatto etal.. 2003). While these wells may not have been
16    hydraulically fractured, the study demonstrates that older wells can exhibit more integrity
17    problems. Watson and Bachu [2009] found that a higher portion of deviated wells had sustained
18    casing pressure compared to vertical wells. Increased pressures, cyclic stresses [Syed and Cutler,
19    2010]. and difficulty in cementing horizontal wells [Sabins. 1990] also may lead to increased
20    instances of sustained casing pressure in wells where hydraulic fracturing occurs [Muehlenbachs et
21    al.. 2012: Rowe and Muehlenbachs. 1999].

22    Sustained casing pressure can be a concern for several reasons. If the pressures are allowed to build
23    up to above the burst pressure of the exterior casing or the collapse pressure of the interior casing,
24    the casing may fail. Increased pressure can also cause gas or liquids to enter lower-pressured
25    formations that are exposed to the annulus either through leaks or uncemented sections.
26    Laboratory experiments by Harrison [1985] demonstrated that over-pressurized gas in the annulus
27    could cause rapid movement of gas into drinking water resources if a permeable pathway exists
28    between the annulus and the ground water. Over-pressurization of the annulus can be avoided by
29    venting the annulus to the atmosphere.

30    In a few cases, sustained casing pressure in wells that have been hydraulically fractured may have
31    been linked to drinking water contamination, although it is challenging to definitively determine
32    the actual cause. In one study in northeastern Pennsylvania, hydrocarbon and isotope
33    concentrations were used to investigate stray gas migration into domestic drinking water [U.S. EPA,
      i Sustained casing pressure is pressure in any well annulus that is measurable at the wellhead and rebuilds after it is bled
      down, not caused solely by temperature fluctuations or imposed by the operator fSkierven etal.. 2011). If the pressure is
      relieved by venting natural gas from the annulus to the atmosphere, it will build up again once the annulus is closed (i.e.,
      the pressure is sustained). The return of pressure indicates that there is a small leak in a casing or through uncemented or
      poorly cemented intervals that exposes the annulus to a pressured source of gas that is actively being used. It is possible
      to have pressure in more than one of the annuli.
                    This document is a draft for review purposes only and does not constitute Agency policy.
      June 2015                                     6-25                   DRAFT—DO NOT CITE OR QUOTE

-------
      Hydraulic Fracturing Drinking Water Assessment
 1    20141]. While the composition of the gas in the water wells was consistent with that of the gas
 2    found in nearby gas wells with high casing pressures, other possible sources of the gas could not be
 3    ruled out. Several gas wells in the study area were cited by the PA DEP for having elevated casing
 4    annulus pressures. In another example in Alberta, Canada, 14% of wells drilled since 1971 have
 5    shown serious sustained casing flow, defined as more than 10,594 ft3 (300 m3)/day atpressures
 6    higher than 0.48 psi/ft (11 kPa/m) times the surface casing depth (Tackson and Dussealt, 2014].
 7    Another study in the same area found gas in nearby drinking water wells had a composition that
 8    was consistent with biogenic methane mixing with methane from nearby coalbed methane and
 9    deeper natural gas fields (Tilley and Muehlenbachs, 2012].

10    Adequate well design, detection (i.e., through annulus pressure monitoring], and repair of sustained
11    casing pressure reduce the potential for fluid movement. Watson and Bachu (2009] found that
12    regulations that required monitoring and repair of sustained casing vent flow or sustained casing
13    pressure had a positive effect on lowering leak rates. The authors also found that wells initially
14    designed for injection experienced sustained casing pressure less often than those that were
15    retrofitted (Watson and Bachu. 2009].

16    Another study in Mamm Creek, Colorado, obtained similar results. The Mamm Creek field is in an
17    area where lost cement and shallow, gas-containing formations are common. A number of wells in
18    the area have experienced sustained casing pressure, and methane  has been found in several
19    drinking water wells along with seeps into local creeks and ponds. In one well, four pressured gas
2 0    zones were encountered during well drilling and there was a lost cement incident, which resulted
21    in the cement top being more than 4,000 ft (1,219 m] lower than originally intended. Due to high
22    measured bradenhead pressure (661 psi, or 4.6 MPa], cement remediation efforts were
23    implemented (Crescent. 2011: COGCC. 2004]. The operator of this well was later cited by the
24    Colorado Oil and Gas Conservation Commission (COGCC] for causing natural gas and benzene to
25    seep into a nearby creek. The proposed route of contamination was contaminants flowing up the
26    well annulus and then along a fault. The proposed contamination route appeared to be validated
27    because, once remedial cementing was performed on the well, methane and benzene levels in the
28    creek began to drop (Science Based Solutions LLC, 2014]. In response to the incident, the state
29    instituted requirements to identify and cement above the top of the highest gas-containing
30    formation and to monitor casing pressures after cementing.

31    Not every well that shows positive pressure in the annulus poses a potential problem. Sustained
32    pressure is only a problem when it exceeds the ability of the wellbore to contain it or when it
33    indicates downhole communication problems (TIPRO, 2012]. A variety of management options are
34    available for managing such pressure including venting, remedial cementing, and use of kill fluids in
35    the annulus (TIPRO. 2012].l
      1 A kill fluid is a weighted fluid with a density that is sufficient to overcome the formation pressure and prevent fluids
      from flowing up the wellbore.
                   This document is a draft for review purposes only and does not constitute Agency policy.
      June 2015                                   6-26                  DRAFT—DO NOT CITE OR QUOTE

-------
      Hydraulic Fracturing Drinking Water Assessment
      6.3.  Fluid Migration Associated with Induced Fractures within Subsurface
            Formations
 1    In this section, we discuss potential pathways for fluid movement associated with induced fractures
 2    and subsurface geologic formations (outside of the well system described in Section 6.2). We
 3    examine the potential for fluid migration into drinking water resources by evaluating the
 4    development of migration pathways within subsurface formations, the flow of injected and
 5    formation fluids, and important factors that affect these processes.:

 6    Fluid movement requires both a physical conduit (e.g., the permeable matrix pore volume or a
 7    fracture in the rock) and a driving force.2 In subsurface rock formations, fluid movement is driven
 8    by the existence of a hydraulic gradient depending on elevation and pressure, which is also
 9    influenced by fluid density, composition, and temperature (Finder and Celia. 2006). Pressure
10    differentials in the reservoir and density-driven fluid buoyancy are the key forces governing fluid
11    migration during and after hydraulic fracturing operations. Pressure differentials depend upon the
12    initial conditions  within these formations and are directly influenced by pressures that are created
13    by injection or production regimes. Buoyancy depends  on density differences among and between
14    gases and liquids, and it causes fluid migration when and where these density differences exist
15    along with a pathway (Finder and Gray, 2008).

16    As hydraulic fracturing takes place, injected fluids leaving the well create fractures within the
17    production zone and enter the formation through the newly created fractures. Unintended fluid
18    migration may result from this fracturing process. Migration pathways to drinking water resources
19    could develop as a result of changes in the subsurface flow or pressure regime associated with
20    hydraulic fracturing; via fractures that extend beyond the intended formation or that intersect
21    existing natural faults or fractures; or via fractures that intersect offset wells or other artificial
22    structures (Jackson etal.. 2013c). These subsurface pathways may facilitate the migration of fluids
23    by themselves or in conjunction with the well-based pathways described in Section 6.2. Fluids
24    potentially available for migration include both fluids that are injected into the well (including
25    leakoff) and formation fluids (including brine or natural gas).3
      1 A subsurface formation (also referred to as a "formation" throughout this section) is a mappable body of rock of
      distinctive rock type(s), including the rock's pore volume (i.e., the void space within a formation that fluid flow can occur,
      as opposed to the bulk volume which includes both pore and solid phase volume), with a unique stratigraphic position.
      2 Permeability (i.e., intrinsic or absolute permeability) of formations describes the ability of water to move through the
      formation matrix, and it depends on the rock's grain size and the connectedness of the void spaces between the grains.
      Where multiple phases of fluids exist in the pore space, the flow of fluids also depends on relative permeabilities.
      3 Leakoff is the fraction of the injected fluid that infiltrates into the formation and is not recovered during production
      (Economides et al.. 2007).
                    This document is a draft for review purposes only and does not constitute Agency policy.
      June 2015                                     6-27                   DRAFT—DO NOT CITE OR QUOTE

-------
      Hydraulic Fracturing Drinking Water Assessment
 1    The potential for subsurface fluid migration into drinking water resources can be evaluated during
 2    two different time periods [Kim and Moridis. 2015]:

 3         1.  Following the initiation of fractures within the reservoir prior to any production, when the
 4            injected fluid, pressurizing the formation, flows through the fractures and the fractures
 5            grow into the reservoir. Fluid leaks off into the formation, allowing the fractures to close
 6            except where they are held open by the proppant [Adachi et al., 2007].
 7         2.  During the production period, after fracturing is completed and pressure in the fractures is
 8            reduced, and hydrocarbons (along with produced water] flow from the reservoir into the
 9            well.
10    Note that these two time periods vary in duration. As described in Chapter 2, the first period of
11    fracture creation and propagation (i.e., the hydraulic fracturing itself] is a relatively short-term
12    process, typically lasting 2 to 10 days,  depending on the number  of stages in the fracture treatment
13    design. On the other hand, operation of the well for production covers a substantially longer period
14    (depending on many factors such as the amount of hydrocarbons in place and economic
15    considerations], and can be as long as  40 or 60 years in onshore tight gas reservoirs (Ross and King,
16    2007].

17    The following discussion of potential subsurface fluid migration into drinking water resources
18    focuses primarily on the physical movement of fluids and the factors that affect this movement.
19    Section 6.3.1 describes the basic principles of subsurface fracture creation, geometry, and
2 0    propagation, to provide context for the discussion of potential  fluid migration pathways in Section
21    6.3.2. Geochemical and biogeochemical reactions among injected fluids, formation fluids, subsurface
22    microbes, and rock formations are another important component of subsurface fluid migration and
23    transport. See Chapter 7 for a discussion of the processes that affect pore fluid biogeochemistry and
24    influence the chemical and microbial composition of flowback  and produced water.

      6.3.1.  Overview of Subsurface Fracture Growth
25    Fracture initiation and growth is a highly complex process due to the heterogeneous nature of the
26    subsurface environment It depends on the geomechanical characteristics of rock formations, fluid
27    properties, pore pressures, and subsurface stress fields. As shown in Figure  6-4, fracture formation
28    is controlled by the three in situ principal compressive  stresses: the vertical stress (Si/], the
29    maximum horizontal  stress (Sn), and the minimum horizontal stress (S/,]. During hydraulic
30    fracturing, pressurized fluid injection creates high pore pressures around the well. When the
31    pressure exceeds the  local least principal stress and the tensile strength of the rock, failure results
32    and fractures form fZoback. 2010: FiaeretaL 20081

33    Fractures propagate (increase in length] in the direction of the maximum principal stress, which is
34    perpendicular to the direction of the least principal stress. Deep in the subsurface, the maximum
35    principal stress is generally in the vertical direction because the overburden (the weight of
36    overlying rock] is the largest single stress.  Therefore, in deep formations, the local least principal
37    stress is the minimum horizontal stress (S/,], and the principal  fracture orientation is expected to be
38    vertical. This is the scenario illustrated in Figure 6-4. At shallower depths, where the rock is

                   This document is a draft for review purposes only  and does not constitute Agency policy.
      June 2015                                    6-28                  DRAFT—DO NOT CITE OR QUOTE

-------
     Hydraulic Fracturing Drinking Water Assessment
1    subjected to less pressure from the overburden, more fracture propagation is expected to be in the
2    horizontal direction. Using tiltmeter data from over 10,000 fractures in various North American
3    reservoirs, Fisher and Warpinski [2012] found that fractures at approximately 4,000 ft (1,220 m)
4    below the surface or deeper are primarily vertical (see below for more information on tiltmeters).
5    Between approximately 4,000 and 2,000 ft (1,220 and 610 m), fracture complexity increases, and
6    fractures at approximately 2,000 ft (610 m) or shallower are primarily horizontal (Fisher and
7    Warpinski. 2012].1 Horizontal fracturing can also occur in deeper settings in some less-common
8    reservoir environments where the principal stresses have been altered by salt intrusions or similar
9    types of geologic activity (Tones and Britt, 2009].
     Figure 6-4. Hydraulic fracture planes (represented as ovals), with respect to the principal
               subsurface compressive stresses: Sv (the vertical stress), SH (the maximum
               horizontal stress), and S/, (the minimum horizontal stress).
               At depths greater than approximately 2,000 ft (610 m), the principal fracture orientation is expected
               to be vertical, as fractures propagate in the direction of SH.
     1 Fracture complexity is the ratio of horizontal-to-vertical fracture volume distribution, as defined by Fisher and
     Warpinski (2012). Fracture complexity is higher in fractures with a larger horizontal component.
                   This document is a draft for review purposes only and does not constitute Agency policy.
     June 2015                                     6-29                   DRAFT—DO NOT CITE OR QUOTE

-------
      Hydraulic Fracturing Drinking Water Assessment
 1    In addition to the principal subsurface stresses, other geomechanical and reservoir characteristics
 2    and operational factors affect fracture creation, geometry, and propagation.1 These include initial
 3    reservoir pressure and saturation, injected fluid pressure or injection rate, geomechanical rock
 4    characteristics, reservoir heterogeneity, tensile strength, fluid type within fractures, and reservoir
 5    permeabilities [Kim and Moridis. 2015]. Fracture creation is a complex process that involves
 6    interactions between multiple properties. For example, as described by Daneshy [2009], fracture
 7    height depends on a combination of parameters and processes including the material properties of
 8    geologic formations, pore pressures, stress differences in adjacent formations, shear failure
 9    (slippage] at the  fracture tip, and the reorientation of the fracture as it crosses an interface between
10    formations. Injection rates, the initial water saturation of the formation, and the type of fluid
11    injected also have effects on fracture creation and propagation [Kim and Moridis, 2015, 2013].

12    Numerical modeling techniques have been developed to describe fracture creation and propagation
13    and to provide a  better understanding of this complex process [Kim and Moridis, 2013]. Modeling
14    hydraulic fracturing in shale or tight gas reservoirs requires integrating the physics of both flow
15    and geomechanics to account for fluid flow, fracture propagation, and dynamic changes in pore
16    volume and permeability. Some important flow and geomechanical parameters included in these
17    types of advanced models are: permeability, porosity, Young's modulus, Poisson's ratio, and tensile
18    strength, as well  as heterogeneities associated with these parameters.2 Some investigations using
19    these models have indicated that the vertical propagation of fractures (due to tensile failure] may
20    be limited by shear failure, which increases the permeability of the formation and leads to greater
21    leakoff. These findings demonstrate that elevated pore pressure can cause shear failure, thus
22    further affecting  matrix permeability, flow regimes, and leakoff (Daneshy. 2009]. Computational
23    investigations have also indicated that slower injection rates can increase the amount of leakoff
24    (Kim and Moridis. 2013].

25    In addition to their use in research settings, analytical and numerical modeling approaches are used
26    by oil and gas companies to design hydraulic fracturing treatments and predict the extent of
27    fractured areas (Adachietal.. 2007]. Specifically, modeling techniques are used to assess the
28    treatment's sensitivity to critical parameters such as injection rate, treatment volumes, fluid
29    viscosity, and leakoff. The industry models range from simpler (typically two-dimensional]
30    theoretical models to computationally more complicated and  accurate three-dimensional models.

31    In addition to computational approaches, monitoring of hydraulic fracturing operations can provide
32    insights into fracture development Monitoring techniques involve both operational monitoring
33    methods and "external" methods that are not directly related  to the production operation.
34    Operational monitoring refers to the monitoring of parameters including pressure, flow rate, fluid
35    density, and additive concentrations using surface equipment and/or downhole sensors (Eberhard,
36    2011]. This monitoring is conducted to ensure that the operation is proceeding as planned and to
      i Fracture geometry refers to characteristics of the fracture such as height and aperture (width).
      2 Young's modulus, a ratio of stress to strain, is a measure of the rigidity of a material. Poisson's ratio is a ratio of
      transverse-to-axial (or latitudinal-to-longitudinal) strain, and it characterizes how a material is deformed under pressure.
      See Zoback(2010) for more information on the geomechanical properties of reservoir rocks.
                    This document is a draft for review purposes only and does not constitute Agency policy.
      June 2015                                     6-30                   DRAFT—DO NOT CITE OR QUOTE

-------
      Hydraulic Fracturing Drinking Water Assessment
 1    determine if operational parameters need to be adjusted. Interpretation of pressure data can be
 2    used to better understand fracture behavior [e.g.. Kim and Wang. 2014]. Anomalies in operational
 3    monitoring data can also indicate whether an unexpected event has occurred, such as
 4    communication with another well (see Section 6.3.2.3).

 5    The volume of fluid injected is typically monitored to provide information on the volume and extent
 6    of fractures created [Fiewelling etal., 2013]. However, numerical investigations have found that
 7    reservoir gas flows into the fractures immediately after they open from hydraulic fracturing, and
 8    injection pressurizes both gas and water within the fracture to induce further fracture propagation
 9    [Kim and Moridis, 2015]. Therefore, the fracture volume can be larger than the injected fluid
10    volume. As a result, simple estimation of fracture volume based on the amount of injected fluid may
11    underestimate the growth of the vertical fractures, and additional information is needed to
12    accurately predict the extent of fracture growth.

13    External monitoring technologies can also be used to collect data on fracture characteristics and
14    extent during hydraulic fracturing and/or production. These monitoring methods can be divided
15    into near-wellbore and far-field techniques. Near-wellbore techniques include the use of tracers,
16    temperature logs, video logs, or caliper logs [Holditch. 2007]. However, near-wellbore techniques
17    and logs only provide information for, at most, a distance of two to three wellbore diameters from
18    the well and are, therefore, not suited for tracking fractures for their entire length [Holditch, 2007].
19    Far-field methods, such as microseismic monitoring or tiltmeters, are used if the intent is to
20    estimate fracture growth and height across the entire fractured reservoir area. Microseismic
21    monitoring involves placing one or more geophones in a position to detect the very small amounts
22    of seismic energy generated during subsurface fracturing [Warpinski, 2009].' Monitoring these
23    microseismic events gives an idea of the location and size of the fracture network, as well as the
24    orientation and complexity of fracturing [Fisher and Warpinski. 2012]. Tiltmeters, which measure
25    extremely small deformations in the earth, can be used to determine the direction and volume of
26    the fractures and, within certain distances from the well, to estimate their dimensions [Lecampion
27    etal.. 2005].

      6.3.2.  Migration of Fluids through Pathways  Related to Fractures/Formations
28    As noted above, subsurface migration of fluids requires a pathway, induced or natural, with enough
29    permeability to allow fluids to flow, as well as a hydraulic gradient physically driving the fluid
30    movement. The following subsections describe and evaluate potential pathways for the migration
31    of fracturing fluids, hydrocarbons, or other formation fluids from producing formations to  drinking
32    water resources. They also present cases where the existence of these pathways has been
33    documented. As described above, potential subsurface migration pathways for fluid flow out of the
34    production formation are categorized as follows: (1] flow of fluids into the production zone via
35    induced fractures and out of the production zone via flow through the formation,  (2] fracture
36    overgrowth outof the production zone, (3] migration via fractures intersecting offsetwells and
      1 Typical microseismic events associated with hydraulic fracturing have a magnitude on the order of -2.5 (negative two
      and half) [Warpinski. 2009).
                   This document is a draft for review purposes only and does not constitute Agency policy.
      June 2015                                    6-31                  DRAFT—DO NOT CITE OR QUOTE

-------
      Hydraulic Fracturing Drinking Water Assessment
 1    other artificial structures, and (4) migration via fractures intersecting other geologic features.
 2    Although these four potential pathways are discussed separately here, they may act in combination
 3    with each other or in combination with pathways along the well (as discussed in Section 6.2) to
 4    affect drinking water resources.

 5    In many cases (depending on fracture depth, height, and direction), the distance between the
 6    producing formation and the drinking water resource is one of the most important factors affecting
 7    the possibility of fluid migration between these formations (Reagan et al.. 2 015: Tacksonetal..
 8    2013c). This distance varies substantially among shale gas plays, coalbed methane plays, and other
 9    areas where hydraulic fracturing takes place in the United States (see Table 6-2). Many hydraulic
10    fracturing operations target deep shale zones such as the Marcellus or Haynesville/Bossier, where
11    the vertical distance between the top of the shale formation and the base of drinking water
12    resources may be 1 mile (1.6 km) or greater. This is reflected in the Well File Review, which found
13    that the largest proportion of wells used for hydraulic fracturing—an estimated 6,200 wells
14    (27%)—had 5,000 to 5,999 ft (1,524 to 1,828 m) of measured distance along the wellbore between
15    the induced fractures and the reported base of protected ground water resources  (U.S. EPA,
16    2015O).1 However, as shown in Table 6-2, operations in the Antrim and the New Albany plays take
17    place at relatively shallower depths, with distances of 100 to 1,900 ft (30 to 579 m) between the
18    producing formation and the base of drinking water resources. The Well File Review indicated that
19    20% of wells used for hydraulic fracturing (an estimated 4,600 wells) were located in areas with
20    less than 2,000 ft (610 m) between the fractures and the base of protected ground water resources
21    (U.S. EPA, 2015o). In coalbed methane plays, which are typically shallower than shale gas plays,
22    these separation distances can be even smaller. For example, in the Raton Basin of southern
23    Colorado and northern New Mexico, approximately 10% of coalbed methane wells have less than
24    675 ft (206 m) of separation between the gas wells' perforated intervals and the depth of local
25    water wells. In certain areas of the basin, this distance is less than 100 ft (30 m) (Watts. 2006).

2 6    Some hydraulic fracturing operations are conducted within formations that contain drinking water
27    resources (see Table 6-2). One example of hydraulic fracturing taking place within a geologic
28    formation that is also used as a drinking water source is in the Wind River Basin in Wyoming
29    fWYOGCC. 2014: Wright etal.. 20121 Vertical gas wells in this area target the lower Eocene Wind
30    River Formation and the underlying Paleocene Fort Union Formation, which consist of interbedded
31    layers of sandstones, siltstones, and mudstones. The Wind River Formation also serves as the
32    principal source of domestic, municipal,  and agricultural water in this rural area. Hydraulic
33    fracturing in rock formations that meet a state or federal definition of an underground source of
34    drinking water is also known to take place in coalbed methane operations in the Raton Basin (U.S.
35    EPA. 20151). in the Powder River Basin of Montana and Wyoming (as described in Chapter 7), and
36    in several other coalbed methane plays. In one field in Alberta, Canada, there is evidence that
37    fracturing in the same formation as a drinking water resource (in combination with well integrity
      1 In the Well File Review, measured depth represents length along the wellbore, which may be a straight vertical distance
      below ground or may follow a more complicated path, if the wellbore is not straight and vertical.
                   This document is a draft for review purposes only and does not constitute Agency policy.
      June 2015                                    6-32                   DRAFT—DO NOT CITE OR QUOTE

-------
     Hydraulic Fracturing Drinking Water Assessment
1    problems; see Section 6.2.2.2) led to gas migration into water wells [Tilley and Muehlenbachs,
2    2012]. However, no information is available on other specific incidents of this type.
     Table 6-2. Comparing the approximate depth and thickness of selected U.S. shale gas plays
               and coalbed methane basins.

               Shale data are reported in GWPC and ALL Consulting (2009) and NETL(2013); coalbed methane data
               are reported in ALL Consulting (2004) and U.S. EPA (2004). See Figures 2-2 and 2-4 in Chapter 2 for
               information on the locations of these basins, plays, and formations.
Basin/play/formation3
Approx. depth
(ft [m] below surface)
Approx. net thickness
(ft [m])
Distance between top of
production zone and base of
treatable water (ft [m])
Shale plays
Antrim
Barnett
Eagle Ford
Fayetteville
Haynesville- Bossier
Marcellus
New Albany
Woodford
600 to 2,200
[183 to 671]
6,500 to 8,500
[1,981 to 2,591]
4,000 to 12,000
[1,219 to 3,658]
1,000 to 7,000
[305 to 2, 134]
10,500 to 13,500
[3,200 to 4, 115]
4,000 to 8,500
[1,219 to 2,591]
500 to 2,000
[152 to 610]
6,000 to 11,000
[1,829 to 3,353]
70 to 120
[21 to 37]
100 to 600
[30 to 183]
250
[76]
20 to 200
[6 to 61]
200 to 300
[61 to 91]
50 to 200
[15 to 61]
50 to 100
[15 to 30]
120 to 220
[37 to 67]
300 to 1,900
[91 to 579]
5,300 to 7,300
[1,615 to 2,225]
2,800 to 10,800
[853 to 3,292]
500 to 6,500
[152 to 1,981]
10, 100 to 13, 100
[3,078 to 3,993]
2, 125 to 7,650
[648 to 2,332]
100 to 1,600
[30 to 488]
5,600 to 10,600
[1,707 to 3,231]
Coalbed methane basins
Black Warrior
(Upper Pottsville)
Powder River
(Fort Union)
Raton (Vermejo
and Raton)
San Juan (Fruitland)
0 to 3,500
[0 to 1,067]
450 to >6,500
[137 to 1,981]
< 500 to > 4, 100
[< 152 to > 1,250]
550 to 4,000
[168 to 1,219]
< 1 to > 70
[21]
75
[23]
10 to >140
[3 to >43]
20 to 80
[6 to 24]
As little as zero
As little as zerob
As little as zerob
As little as zero
                   This document is a draft for review purposes only and does not constitute Agency policy.

     June 2015                                      6-33                   DRAFT—DO NOT CITE OR QUOTE

-------
      Hydraulic Fracturing Drinking Water Assessment
Basin/play/formation3
Approx. depth
(ft [m] below surface)
Approx. net thickness
(ft [m])
Distance between top of
production zone and base of
treatable water (ft [m])
       a For coalbed methane, values are given for the specific coal units noted in parentheses.
       b Formation fluids in producing formations meet the definition of drinking water in at least some areas of the basin.
 1    The overall frequency of occurrence of hydraulic fracturing in aquifers that meet the definition of
 2    drinking water resources across the United States is unknown. Some information, however, that
 3    provides insights on the occurrence and geographic distribution of this practice is available.
 4    According to the Well File Review, an estimated 0.4% of the 23,200 wells represented in that study
 5    had perforations used for hydraulic fracturing that were placed shallower than the base of the
 6    protected ground water resources reported by well operators [U.S. EPA, 2015O].1 An analysis of
 7    produced water composition data maintained by the U.S. Geological Survey (USGS) provides insight
 8    into the geographic distribution of this practice. The USGS produced water database contains
 9    results from analyses of samples of produced water collected from more than 8,500 oil and gas
10    production wells in unconventional formations (coalbed methane, shale gas, tight gas, and tight oil)
11    within the continental United States.2 Just over 5,000 of these samples, which were obtained from
12    wells located in 37 states, reported total dissolved solids (TDS) concentrations. Because the
13    database does not track whether samples were from wells that were hydraulically fractured, we
14    selected samples from wells that were more likely to have been hydraulically fractured by
15    restricting samples to those collected in 1950 or later and to those that were  collected from wells
16    producing from tight gas, tight oil, shale gas, or coalbed methane formations.  This yielded 1,650
17    samples from wells located in Alabama, Colorado, North Dakota, Utah, and Wyoming.3'4 The TDS
18    concentrations among these samples ranged from approximately 90  mg/L to 300,000 mg/L.
19    Samples from approximately 1,200 wells in Alabama, Colorado, Utah, and Wyoming reported TDS
20    concentrations at or below 10,000 mg/L. This analysis, in conjunction with the result from the Well
21    File Review, suggests that, while the overall frequency of occurrence may be low, the activity may
22    be concentrated in some areas of the country.
      1 The 95% confidence interval reported in the Well File Review indicates that this phenomenon could have occurred in as
      few as 0.1% of the wells or in as many as 3% of the wells.
      2 We used the USGS Produced Water Geochemical Database Version 2.1 (USGS database v 2.1) for this analysis
      [http://energy.cr.usgs.gov/prov/prodwat/). The database is comprised of produced water samples compiled by the USGS
      from 25 individual databases, publications, or reports.
      3 See  Chapter 2, Text Box 2-1, which describes how commercial hydraulic fracturing began in the late 1940s.
      4 For this analysis, we assumed that produced water samples collected in  1950 or later from shale gas, tight oil, and tight
      gas wells were from wells that had been hydraulically fractured. To estimate which coal bed methane wells had been
      hydraulically fractured, we matched API numbers from coal bed methane wells in the USGS database v 2.1 to the same API
      numbers in the commercial database Drillinglnfo, in which hydraulically fractured wells had been identified by EPA using
      the assumptions described in Section 2.3.1. Wells with seemingly inaccurate (i.e., less than 12 digit) API numbers were
      also excluded. Only coalbed methane wells from the USGS database v 2.1 that matched API numbers in the Drillinglnfo
      database were retained for this analysis.
                    This document is a draft for review purposes only and does not constitute Agency policy.
      June 2015                                      6-34                   DRAFT—DO NOT CITE OR QUOTE

-------
      Hydraulic Fracturing Drinking Water Assessment
      6.3.2.1.  Flow of Fluids Out of the Production Zone
 1    One potential pathway for fluid migration out of the production formation into drinking water
 2    resources is flow of injected fluids (or displacement of formation fluids due to injection) through
 3    the formation matrix during or after a hydraulic fracturing treatment. In deep, low-permeability
 4    shale and tight gas settings and where induced fractures are contained within the production zone,
 5    flow through the production formation has generally been considered an unlikely pathway for
 6    migration into drinking water resources [Jackson etal.. 2013c]. However, there is limited
 7    information available on the fate of injected fluids that are not recovered during production (i.e.,
 8    leakoff) or displaced formation fluids for cases where hydraulic fracturing takes place within or
 9    close to drinking water resources.

10    Leakoff into shale gas formations may be as high as 90% or more of the injected volume (see
11    Section 7.2 and Table 7-2). The actual amount of leakoff depends on the amount of injected fluid,
12    the hydraulic properties of the reservoir (e.g., permeability), the capillary pressure near the
13    fracture faces, and the period of time the well is shut in following hydraulic fracturing before the
14    start of production (Kim etal.. 2014: Byrnes.  2011). '-2 However, despite the potentially large
15    volume of fluid that may be lost into the formation, the flow of this fluid is generally controlled or
16    limited by processes such as imbibition by capillary forces and adsorption onto clay minerals
17    (Duttaetal.. 2014: Dehghanpour etal.. 2013: Dehghanpour etal.. 2012: Roychaudhuri etal.. 2011).3
18    It has been suggested that these processes can sequester the fluids in the producing formations
19    permanently or for geologic time scales (Engelder, 2012: Byrnes, 2011).

20    A limited number of studies in the literature have evaluated a combination of certain conditions
21    that can facilitate migration of fluids despite these processes.. Myers (2012b) suggests that
22    migration of injected and/or formation fluids into the overburden may be possible in cases where
23    there is a significant vertical hydraulic gradient, sufficient permeability, density-driven buoyancy,
24    and the displacement of formation brines by large volumes of injected fluid. Flewelling and Sharma
25    (2014) note that, for migration to occur, an upward hydraulic gradient would be necessary,
26    particularly for brine that is denser than the ground water in the overlying formations; in the case
27    of natural gas, though, buoyancy would provide an upward flux A limited number of studies in the
28    literature have evaluated a combination of certain conditions that can facilitate migration of fluids
29    despite these processeSi Myers (2012b) suggests that migration of injected and/or formation fluids
30    into the overburden may be possible in cases where there is a significant vertical hydraulic
31    gradient, sufficient permeability, density-driven buoyancy, and the displacement of formation
      1 Relative permeability is a dimensionless property allowing for the comparison of the different abilities of fluids to flow
      in multiphase settings. If a single fluid is present, its relative permeability is equal to 1, but the presence of multiple fluids
      generally inhibits flow and decreases the relative permeability (Schlumberger. 2014).
      2 Shutting in the well after fracturing allows fluids to move farther into the formation, resulting in a higher gas relative
      permeability near the fracture surface and improved gas production [Bertoncello etal.. 2014).
      3 Imbibition is the displacement of a nonwetting fluid (i.e., gas) by a wetting fluid (typically water). The terms wetting or
      nonwetting refer to the preferential attraction of a fluid to the surface. In typical reservoirs, water preferentially wets the
      surface, and gas is nonwetting. Capillary forces arise from the differential attraction between immiscible fluids and solid
      surfaces; these are the forces responsible for capillary rise in small-diameter tubes and porous materials. These
      definitions are adapted from Pake (1978).
                    This document is a draft for review purposes only and does not constitute Agency policy.
      June 2015                                     6-35                  DRAFT—DO NOT CITE OR QUOTE

-------
      Hydraulic Fracturing Drinking Water Assessment
 1    brines by large volumes of injected fluid. Flewelling and Sharma [2014] note that for migration to
 2    occur, an upward hydraulic gradient would be necessary, particularly for brine that is denser than
 3    the ground water in the overlying formations; in the case of natural gas, though, buoyancy would
 4    provide an upward flux [Vengosh etal., 2014]. Some natural conditions could create this upward
 5    hydraulic gradient in the absence of any effects from hydraulic fracturing [Flewelling and Sharma,
 6    2014]. However, these natural mechanisms have been found to cause very low flow rates over very
 7    long distances, yielding extremely small vertical fluxes in sedimentary basins—corresponding to
 8    some estimated travel times of 100,000 to 100,000,000 years across a 328 ft (100 m] thick layer
 9    with about 0.01 nD (1 x 10~23 m2] permeability [Flewelling and Sharma, 2014]. Furthermore,
10    fracturing fluid would likely be sequestered in the immediate vicinity of the fracture network due to
11    capillary tension [Engelder, 2012].

12    Over-pressurization of producing formations due to the injection of large amounts of fluid during
13    hydraulic fracturing may support the upward hydraulic gradient for fluid migration [Myers,
14    2012b]. Myers' modeling results suggest that significant pressure buildup that occurs at the
15    location of fluid injection may not return to pre-hydraulic fracturing levels for up to a year.
16    However, these findings have been disputed in the literature due to certain suggested limitations of
17    the original study (e.g., extensive simplification of the model, lack of accurate characterization of
18    regional flow, misrepresentation of saturation conditions in shale formations], and they have been
19    found to be physically implausible given the hydrogeologic characteristics of actual sedimentary
20    basins fCohen etal.. 2013: Fie welling etal.. 2013: VidicetaL 2013: Saiers and Earth. 20121 Some
21    researchers have also suggested that pressure perturbations due to hydraulic  fracturing operations
22    are localized to the immediate vicinity of the fractures, due to the very low permeabilities of shale
23    formations [Flewelling and Sharma, 2014]. However, there are emerging studies indicating that
24    pressure impacts of hydraulic fracturing operations may extend farther than the immediate vicinity
25    and may create risk of induced seismicity [Skoumaletal.. 2015]. Following hydraulic fracturing
26    operations, a large-scale depressurization would be expected over the longer term due to
27    hydrocarbon production, which may counteract any short-term localized pressure effects of
28    hydraulic fracturing during production and cause fluids to flow primarily toward the fracture
29    network [Flewelling and Sharma, 2014].

30    In responses to these critiques, Myers [2013. 2012a] states that they do not prove his original
31    hypothesis or findings wrong, but instead highlight the need for complex three-dimensional
32    modeling and detailed data collection for improving the understanding of the process and risks to
33    drinking water resources. Myers  [2013, 2012a] argues that, given the large volume of hydraulic
34    fracturing operations  in formations such as the Marcellus, these formations would have to hold
35    very large volumes of water that would be imbibed into the shale. Furthermore, he notes that
36    migration of these fluids into overlying formations may be facilitated by existing fractures or out-of-
37    zone fracturing (as discussed in the following sections].

      6.3.2.2. Fracture Overgrowth out of the Production Zone
3 8    Fractures that extend out of the intended production zone into another formation or an unintended
39    zone within the same  formation could provide a potential fluid migration pathway into drinking
                   This document is a draft for review purposes only and does not constitute Agency policy.
      June 2015                                    6-36                  DRAFT—DO NOT CITE OR QUOTE

-------
      Hydraulic Fracturing Drinking Water Assessment
 1    water resources [Tacksonetal., 2013c]. This migration could occur either through the fractures
 2    themselves or in connection with other permeable subsurface features or formations (see Figure
 3    6-5). Such "out-of-zone fracturing" is undesirable from a production standpoint and may occur as a
 4    result of inadequate reservoir characterization or fracture treatment design [Eisner etal., 2006].
 5    Some researchers have noted that fractures growing out of the targeted production zone could
 6    potentially contact other formations, such as higher conductivity sandstones or conventional
 7    hydrocarbon reservoirs, which may create an additional pathway for potential migration into a
 8    drinking water resource [Reagan etal.. 2015). In addition, fractures growing out of the production
 9    zone could potentially intercept natural, preexisting fractures (discussed in Section 6.3.2.4) or
10    active or abandoned wells near the well where hydraulic fracturing is performed (discussed in
11    Section 6.3.2.3).
                             44-4^1  Naturally Occurring Fractures /^jjj^urf
      Figure 6-5. Conceptualized depiction of potential pathways for fluid movement out of the
                production zone: (a) induced fracture overgrowth into over- or underlying
                formations; (b) induced fractures intersecting natural fractures; and (c) induced
                fractures intersecting a transmissive fault.
                Thickness and depth of production formation are important site specific factors for each operation.

12    The fracture's geometry (see Section 6.3.1) affects its potential to extend beyond the intended zone
13    and serve as a pathway to drinking water resources. Vertical heights of fractures created during
                    This document is a draft for review purposes only and does not constitute Agency policy.
      June 2015                                     6-37                  DRAFT—DO NOT CITE OR QUOTE

-------
      Hydraulic Fracturing Drinking Water Assessment
 1    hydraulic fracturing operations have been measured in several U.S. shale plays, including the
 2    Barnett, Woodford, Marcellus, and Eagle Ford, using microseismic and microdeformation field
 3    monitoring techniques [Fisher and Warpinski. 2012). These data indicate typical fracture heights
 4    extending from tens to hundreds of feet. Davies etal. [2012] analyzed this data set and found that
 5    the maximum fracture height was 1,929 ft (588 m) and that 1% of the fractures had a height greater
 6    than 1,148 ft (350 m). This may raise some questions about fractures being contained within the
 7    producing formation, as some Marcellus fractures were found to extend for at least 1,500 ft
 8    (477 m), while the maximum thickness of the formation is generally 350 ft (107 m) or less  (MCOR.
 9    2012]. However, the majority of fractures were found to have heights less than 328 ft (100 m],
10    suggesting limited possibilities for fracture overgrowth exceeding the separation between  shale
11    reservoirs and shallow aquifers (Davies etal., 2012]. This is consistent with modeling results found
12    by Kim and Moridis (2015] and others, as described below. Where the producing formation is not
13    continuous horizontally, the lateral extent of fractures may also become important. For example, in
14    the Fisher and Warpinski (2012] data set, fractures were found to extend to horizontal lengths
15    greater than 1,000 ft (305m].
16    Results of National Energy Technology Laboratory (NETL] research in Greene County,
17    Pennsylvania, are generally consistent with those reported in the Fisher and Warpinski (2012]  data
18    set Microseismic monitoring was used at six horizontal Marcellus Shale wells to identify the
19    maximum upward extent of brittle deformation caused by hydraulic fracturing (Hammacketal.,
20    2014]. At three of the six wells, fractures extending between 1,000 and 1,900 ft (305 and 579 m]
21    above the Marcellus Shale were identified. Overall, approximately 40% of the microseismic events
22    occurred above the Tully Limestone, the formation overlying the Marcellus Shale that is sometimes
23    referred to as an upper barrier to hydraulic fracture growth. However, all microseismic events were
24    at least 5,000 ft (1,524 m] below drinking water aquifers, as the Marcellus Shale is one of the
25    deepest target formations (see Table 6-2], and no impacts to drinking water resources or another
26    local gas-producing interval were identified. See Text Box 6-3 for more information on the  Greene
27    County site.

28    Similarly, in Dunn County, North Dakota, there is evidence of out-of-zone fracturing in the Bakken
29    Shale (U.S. EPA. 201511. Atthe Killdeer site (see Section 6.2.2.1 and Chapter 5, Text Box 5-12],
30    fracturing fluids and produced water were released during a rupture of the casing at the Franchuk
31    44-20 SWH well. Water quality characteristics at two  monitoring wells located immediately
32    downgradient of the Franchuk well reflected a mixing of local Killdeer Aquifer water with deep
33    formation brine. Ion and isotope ratios used for brine fingerprinting suggest that Madison Group
34    formations (which directly overlie the Bakken in the Williston Basin] were the source of the brine
35    observed in the Killdeer Aquifer, and the authors concluded that this provides evidence for out-of-
36    zone fracturing. Industry experience also indicates that out-of-zone fracturing may be fairly
37    common in the Bakken and that produced water  from many Bakken wells has Madison Group
38    chemical signatures (Arkadakskiy and Rostron. 2013b. 2012b: Peterman etal.. 2012].
                   This document is a draft for review purposes only and does not constitute Agency policy.
      June 2015                                   6-38                  DRAFT—DO NOT CITE OR QUOTE

-------
      Hydraulic Fracturing Drinking Water Assessment
      Text Box 6-3. Monitoring at the Greene County, Pennsylvania, Hydraulic Fracturing Test Site.

 1    Monitoring performed at the Marcellus Shale test site in Greene County, Pennsylvania, evaluated fracture
 2    height growth and zonal isolation during and after hydraulic fracturing operations [Hammack et al.. 2014].
 3    The site has six horizontally drilled and two vertical wells that were completed into the Marcellus Shale.
 4    Pre-fracturing studies of the site included a 3D seismic survey to identify faults, pressure measurements, and
 5    baseline sampling for isotopes; drilling logs were also run. Hydraulic fracturing occurred April 24 to May 6,
 6    2012, and June 4 to 11, 2012. Monitoring at the site included the following:

 7    •   Microseismic monitoring was conducted during four of the six hydraulic fracturing jobs on the site,
 8        using geophones placed in the two vertical Marcellus Shale wells. These data were used to monitor
 9        fracture height growth above the six horizontal Marcellus Shale wells during hydraulic fracturing.
10    •   Pressure and production data were collected from a set of vertical gas wells completed in Upper
11        Devonian/Lower Mississippian zones 3,800 to 6,100 ft (1,158 to 1,859 m] above the Marcellus. Data were
12        collected during and after the hydraulic fracturing jobs and used to identify any communication between
13        the fractured areas and the Upper Devonian/Lower Mississippian rocks.
14    •   Chemical and isotopic analyses were conducted on gas and water produced from the Upper
15        Devonian/Lower Mississippian wells. Samples were analyzed for stable isotope signatures of hydrogen,
16        carbon, and strontium and for the presence of perfluorocarbon tracers used in 10 stages of one of the
17        hydraulic fracturing jobs to identify possible gas or fluid migration to overlying zones fSharmaetal.,
18        2014a: Sharma et al.. 2014b].
19    As of September 2014, no evidence was found of gas or brine migration from the Marcellus Shale [Hammack
20    etal..2Q14]. although longer-term monitoring will be necessary to confirm that no impacts to overlying zones
21    have occurred [Zhang et al., 2014al.
22    Extreme vertical fracture growth is generally considered to be limited by layered geological
23    environments and other physical constraints [Fisher and Warpinski. 2012: Daneshy. 2009]. For
24    example, differences in in situ stresses in layers above and below the production zone can restrict
25    fracture height growth in sedimentary basins [Fisher and Warpinski. 2012). High-permeability
26    layers near hydrocarbon-producing zones can reduce fracture growth by acting as a "thief zone"
27    into which fluids can migrate, or by inducing a large compressive stress that acts on the fracture [de
28    Pater and Dong. 2009. as cited in Fisher and Warpinski. 2012]. Although these thief zones may
29    prevent fractures from reaching shallower formations or growing to extreme vertical lengths, it is
30    important to note that they do allow fluids to migrate out of the production zone into these
31    receiving formations, which could potentially contain drinking water resources. A volumetric
32    argument has also been used to discuss limits of vertical fracture growth; that is, the volumes of
33    fluid needed to sustain fracture growth beyond a certain height would be unrealistic [Fisher and
34    Warpinski. 2012]. However, as described in Section 6.3.1, fracture volume can be greater than the
35    volume of injected fluid  due to the effects of pressurized water combined with the effects of gas
36    during injection [Kim and Moridis, 2015]. Nevertheless, some numerical investigations suggest
37    that, unless unrealistically high pressures and injection rates are applied to an extremely weak and
38    homogeneous formation that extends up to the near surface, hydraulic fracturing generally induces


                    This document is a draft for review purposes only and does not constitute Agency policy.
      June 2015                                     6-39                   DRAFT—DO NOT CITE OR QUOTE

-------
      Hydraulic Fracturing Drinking Water Assessment
 1    stable and finite fracture growth in a Marcellus-type environment and the fractures are unlikely to
 2    extend into drinking water resources [Kim and Moridis. 2015].

 3    Modeling studies have identified other factors that affect the containment of fractures within the
 4    producing formation. As discussed above, additional numerical analysis of fracture propagation
 5    during hydraulic fracturing has demonstrated that contrasts in the geomechanical properties of
 6    rock formations can affect fracture height containment [Gu and Siebrits, 2008] and that geological
 7    layers present within shale gas reservoirs can limit vertical fracture propagation [Kim and Moridis.
 8    2015]. Modeling and monitoring studies generally agree that physical constraints on fracture
 9    propagation will prevent induced fractures from extending from deep zones directly into drinking
10    water resources [Kim and Moridis, 2015: Flewellingetal., 2013: Fisher and Warpinski, 2012].

11    Using a numerical simulation, Reagan etal. [2015] investigated potential short-term migration of
12    gas and water between a shale or tight gas formation and a shallower ground water unit. Migration
13    was assessed immediately after hydraulic fracturing and for up to a 2-year time period during the
14    production stage. The potential migration pathway was assumed to be a permeable fracture or fault
15    connecting the producing formation to the shallower ground water unit Such a pathway may be
16    either entirely hydraulically induced (due to fracture overgrowth in a case where the separation
17    distance is limited, as discussed below], or may be a smaller induced fracture connecting to a
18    natural, permeable fault or fracture (as discussed in Section 6.3.2.4]. For the purposes of this study,
19    the pathway was assumed to be pre-existing, and Reagan etal. (2015] did not model the fracturing
20    process itself.

21    The subsurface system evaluated in the modeling investigation included a horizontal well used for
22    hydraulic fracturing and gas production, the connecting fracture or fault between the producing
23    formation and the aquifer, and a shallow vertical water well in the aquifer (see Figure 6-5]. The
24    parameters and scenarios used in the study are shown in Table 6-3; two vertical separation
25    distances between the producing formation and the aquifer were investigated, along with a range of
26    production zone permeabilities and other variables used to describe four production scenarios. The
27    horizontal well was assigned a constant bottomhole pressure of half the initial pressure of the
28    target reservoir, not accounting for any over-pressurization from hydraulic fracturing. Over-
29    pressurization during hydraulic fracturing may create an additional driving force for upward
30    migration. Results of this investigation, which represents a typical production period, indicate a
31    generally downward water flow within the connecting fracture (from the aquifer through the
32    connecting fracture into the hydraulically induced fractures in the production zone] and some
33    upward migration of gas (Reagan etal., 2015]. In certain cases, gas breakthrough (i.e., the
34    appearance of gas at the base of the drinking water aquifer] was also observed. The key parameter
35    affecting migration of gas into the aquifer was the production regime, particularly whether gas
36    production, driving the fluid migration toward the production well, was occurring in the reservoir.
37    Simulations including a producing gas well showed only a few instances of breakthrough, while
38    simulations without gas production tended to result in breakthrough; these breakthrough times
39    ranged from minutes to 20 days. However, in all cases, the gas escape was limited in duration and
40    scope, because the amount of gas available for immediate migration toward the shallow aquifer was
                   This document is a draft for review purposes only and does not constitute Agency policy.
      June 2015                                    6-40                  DRAFT—DO NOT CITE OR QUOTE

-------
     Hydraulic Fracturing Drinking Water Assessment
1    limited to that initially stored in the hydraulically induced fractures after the stimulation process
2    and prior to production. These simulations indicate that the target reservoir may not be able to
3    replenish the gas available for migration in hydraulically induced fractures prior to production.

     Table 6-3. Modeling parameters and scenarios investigated by Reagan etal. (2015).
               This table illustrates the range of parameters included in the Reagan et al. (2015) modeling study. See
               Figure 6-5, Figure 6-6, and Figure 6-7 for conceptualized illustrations of these scenarios.
Model parameter or variable
Values investigated in model scenarios
All scenarios
Lateral distance from connecting feature to water well
Vertical separation distance between producing
formation and drinking water aquifer
Producing formation permeability range
Drinking water aquifer permeability
Initial conditions
Production well bottom hole pressure
Production regime
328 ft (100m)
656 ft (200 m);
2,625 ft (800 m)
lnD(lxlO"21m2);
100nD(lxlO"19m2);
lu.D(lxlO"18m2)
0.1D(lxlO"13m2);
lD(lxlO"12m2)
Hydrostatic
Half of the initial pressure of the producing formation
(not accounting for over-pressurization from hydraulic
fracturing)
Production at both the water well and the gas well;
Production at only the water well;
Production at only the gas well;
No production
Fracture pathway scenarios
Connecting feature permeability
lD(lxlO"12m2);
10D(lxlO"1:Lm2);
1,000 D (1 x 10"9 m2)
Offset well pathway scenarios
Lateral distance from production well to offset well
Cement permeability of offset well
33 ft (10 m)
1 u.D (1 x 10"18 m2);
lmD(lxlO"15m2);
lD(lxlO"12m2);
1,000 D (1 x 10"9 m2)
                   This document is a draft for review purposes only and does not constitute Agency policy.
     June 2015                                      6-41                   DRAFT—DO NOT CITE OR QUOTE

-------
      Hydraulic Fracturing Drinking Water Assessment
 1    Based on the results of the Reagan etal. [2015] study, gas production from the reservoir appears
 2    likely to mitigate gas migration, both by reducing the amount of available gas and depressurizing
 3    the induced fractures (which counters the buoyancy of any gas that may escape from the
 4    production zone into the connecting fracture). Production at the gas well also creates pressure
 5    gradients that drive a downward flow of water from the aquifer via the fracture into the producing
 6    formation, increasing the amount of water produced at the gas well. Furthermore, the effective
 7    permeability of the connecting feature is reduced during water (downward) and gas (upward)
 8    counter-flow within the fracture, further retarding the upward movement of gas or allowing gas to
 9    dissolve into the downward flow. In contrast, Reagan etal. (2015) found an increased potential for
10    gas release from the producing formation in cases where there is no gas production following
11    hydraulic fracturing. The potential for gas migration during shut-in periods following hydraulic
12    fracturing and prior to production may be more significant, especially when out-of-zone fractures
13    are formed. Without the producing gas well, the gas may rise via buoyancy, with any downward-
14    flowing water from the aquifer displacing the upward-flowing gas.

15    Reagan etal. (2015) also found that the permeability of a connecting fault or fracture may be an
16    important factor for the potential upward migration of gas (although not as significant as the
17    production regime). For the cases where gas escaped from the production zone, the maximum
18    amount of migrating gas depended upon the permeability of the connecting feature: the higher the
19    permeability, the larger the amount. The results also showed that lower permeabilities delay the
20    down ward flow of water from the aquifer, allowing the trace amount of gas that entered into the
21    fracture early in the modeled period to reach the aquifer, which was otherwise predicted to
22    dissolve in the water flowing downward in the feature. Similarly, the permeabilities of the target
23    reservoir, fracture volume, and the separation distance were found to affect gas migration, because
24    they affected the initial amount of gas stored in the hydraulically induced fractures. In contrast, the
25    permeability of the drinking water aquifer was not found to be a significant factor in their
26    assessment

      6.3.2.3. Migration via Fractures Intersecting with Offset Wells and Other Artificial Structures
27    Another potential pathway for fluid migration is one in which injected fluids or displaced
28    subsurface fluids move through newly created fractures into an offset well or its fracture network,
29    resulting in well communication (Jackson etal.. 2013c). This may be a concern, particularly in
30    shallower formations where the local least principal stress is vertical (resulting in more horizontal
31    fracture propagation) and where there are shallow drinking water wells in the same formation.

32    The offset well can be an abandoned, inactive, or active well; if the well has also been used for
33    hydraulic fracturing, the fracture networks of the two wells might intersect The situation where
34    hydraulic fractures unintentionally propagate into other existing, producing hydraulic fractures is
35    referred to as a "frac hit" and is known to occur in areas with a high density of wells (Tacksonetal..
36    2013a). Figure 6-6 provides a schematic to illustrate fractures that intercept an offset well, and
37    Figure 6-7 depicts how the fracture networks of two wells can intersect
                   This document is a draft for review purposes only and does not constitute Agency policy.
      June 2015                                    6-42                  DRAFT—DO NOT CITE OR QUOTE

-------
Hydraulic Fracturing Drinking Water Assessment
Figure 6-6. Induced fractures intersecting an offset well (in a production zone, as shown, or in
          overlying formations into which fracture growth may have occurred).

          This image shows a conceptualized depiction of potential pathways for fluid movement out of the
          production zone (not to scale).
Figure 6-7. Well communication (a frac hit) via induced fractures intersecting another well or
          its fracture network.

          This image shows a conceptualized depiction of potential pathways for fluid movement out of the
          production zone (not to scale).
              This document is a draft for review purposes only and does not constitute Agency policy.

June 2015                                     6-43                   DRAFT—DO NOT CITE OR QUOTE

-------
      Hydraulic Fracturing Drinking Water Assessment
 1    Instances of well communication have been known to occur and are described in the oil and gas
 2    literature. For example, an analysis of operator data collected by the New Mexico Oil Conservation
 3    Division (NM OCD) in 2013-2014 identified 120 instances of well communication in the San Juan
 4    Basin [Vaidyanathan, 2014]. In some cases, well communication incidents have led to documented
 5    production and/or environmental problems. A study from the Barnett Shale noted two cases of well
 6    communication, one with a well 1,100 ft (335 m) away and the other with a well 2,500 ft (762 m)
 7    away from the initiating well; ultimately, one of the offset wells had to be re-fractured because the
 8    well communication halted production (Craig etal.. 2012). In some cases, the fluids that intersect
 9    the offset well flow up the wellbore and spill onto the surface. The EPA (2015n] recorded 10
10    incidents in which fluid spills were attributed to well communication events (see Chapter 7 for
11    more information).1 The subsurface effects of frac hits have not been extensively studied, but these
12    cases demonstrate the possibility of fluid migration via communication with other wells and/or
13    their fracture networks. More generally, well communication events may indicate fracture behavior
14    that was not intended by the treatment design.

15    A well communication event is usually observed at the offset well as a pressure spike, due to the
16    elevated pressure from the originating well, or as an unexpected drop in the production rate (Lawal
17    etal.. 2014: Tackson etal.. 2013a). Ajani and Kelkar (2012) performed an analysis of frac hits in the
18    Woodford Shale in Oklahoma, studying 179 wells over a 5-year period. The authors used fracturing
19    records from the newly completed wells and compared them to production records from
20    surrounding wells. The authors assumed that sudden changes in production of gas or water
21    coinciding with fracturing at a nearby well were caused by communication between the two wells,
22    and increased water production at the surrounding wells was assumed to be caused by fracturing
23    fluid flowing into these offset wells. The results of the Oklahoma study showed that 24 wells had
24    decreased gas production or increased water production within 60 days of the initial gas
25    production at the nearby fractured well. A total of 38 wells experienced decreased gas or increased
26    water production up to a distance of 7,920 ft (2,414 m), measured as the distance between the
27    midpoints of the laterals; 10 wells  saw increased water production from as far away as 8,422 ft
28    (2,567 m). In addition, one well showed a slight increase in gas production rather than a decrease.2

29    Other studies of well communication events have relied on similar information. In the NM OCD
30    operator data set, the typical means of detecting a well communication event was through pressure
31    changes at the offset well, production lost at the offset well, or fluids found in the offset well. In
32    some instances, well operators determined that a well was producing fluid from two different
33    formations, while in one instance, the operator identified a potential well communication event due
34    to an increase in production from the offset well (Vaidyanathan. 2014). In another study, Tackson et
35    al. (2013a) found that the decrease in production due to well communication events was much
36    greater in lower permeability reservoirs. The authors note an example where two wells 1,000 ft
37    (305 m) apart communicated, reducing production in the offset well by 64%. These results indicate
38    that the subsurface interactions of well networks or complex hydraulics  driven by each well at a
      1 Line numbers 163,236,265,271,286,287, 375,376,377, and 380 in Appendix B of U.S. EPAf2015n).
      2 The numbers of wells cited in the study reflect separate analyses, and the numbers cited are not additive.
                   This document is a draft for review purposes only and does not constitute Agency policy.
      June 2015                                    6-44                  DRAFT—DO NOT CITE OR QUOTE

-------
      Hydraulic Fracturing Drinking Water Assessment
 1    densely populated (with respect to wells) area are important factors to consider for the design of
 2    hydraulic fracturing treatments and other aspects of oil and gas production.

 3    The key factor affecting the likelihood of a well communication event and the impact of a frac hit is
 4    the location of the offset well relative to the well where hydraulic fracturing was conducted [Ajani
 5    andKelkar. 2012]. In the Ajani and Kelkar [2012] analysis, the likelihood of a communication event
 6    was less than 10% in wells more than 4,000 ft (1,219 m] apart, but rose to nearly 50% in wells less
 7    than 1,000 ft (305 m] apart Well communication was also much more  likely with wells drilled from
 8    the same pad. The affected wells were found to be in the direction of maximum horizontal stress in
 9    the field, which correlates with the expected direction of fracture propagation.

10    Well communication may be more likely to occur where there is less resistance to fracture growth.
11    Such conditions may be related to existing production operations (e.g., where previous
12    hydrocarbon extraction has reduced the pore pressure, changed stress fields, or affected existing
13    fracture networks] or the existence of high-permeability rock units (Jackson etal.. 2013a]. As Ajani
14    and Kelkar (2012] found in the Woodford Shale, one of the deepest major shale plays (see Table
15    6-2], hydraulic fracturing treatments tend to enter portions of the reservoir that have already been
16    fractured as opposed to entering previously unfractured rocks, ultimately causing interference in
17    offset wells. Mukherjee etal. (2000] described this tendency for asymmetric fracture growth
18    toward depleted areas in low-permeability gas reservoirs due to pore pressure depletion from
19    production at offset wells. The authors note that pore pressure gradients in depleted zones would
20    affect the subsurface stresses. Therefore, depending on the location of  the new well with respect to
21    depleted zone(s] and the orientation of the existing induced fractures,  the newly created fracture
22    may be asymmetric, with only one wing of the fracture extending into the depleted area and
23    developing significant length and conductivity (Mukherjee etal., 2000]. The extent to which the
24    depleted area affects fracturing depends on factors such as cumulative production, pore volume,
25    hydrocarbon saturation, effective permeability, and the original reservoir or pore pressure
26    (Mukherjee etal., 2000]. Similarly, high-permeability rock types acting as thief zones may also
27    cause preferential fracturing due to a higher leakoff rate into these layers (Tacksonetal., 2013a].

28    In addition to location, the potential for impact on a drinking water resource also depends on the
29    condition of the offset well (see Section 6.2 for information on the integrity of well components]. In
30    their analysis, Ajani and Kelkar (2012] found a correlation between well communication and well
31    age: older wells were  more likely to be affected. If the cement in the annulus between the casing
32    and the formation is intact and the well components can withstand the stress exerted by the
33    pressure of the fluid, nothing more than an increase in pressure and extra production of fluids may
34    occur during a well communication event. However, if the offset well is not able to withstand the
35    pressure of the fracturing fluid, well components may fail, allowing fluid to  migrate out of the well.
36    The highest pressures most wells will face during their life spans occur during fracturing. In some
37    cases, temporary equipment is installed in wells during fracturing to protect the well against the
38    increased pressure. Therefore, many producing wells may not be designed to withstand pressures
39    typical of hydraulic fracturing (Enform, 2013] and may experience problems when fracturing
40    occurs in nearby wells. Depending on the location of the weakest point in the  offset well, this could
                   This document is a draft for review purposes only and does not constitute Agency policy.
      June 2015                                    6-45                  DRAFT—DO NOT CITE OR QUOTE

-------
      Hydraulic Fracturing Drinking Water Assessment
 1    result in fluid being spilled onto the surface, rupturing of cement and/or casing and hydraulic
 2    fracturing fluid leaking into subsurface formations, or fluid flowing out through existing flaws in the
 3    casing and/or cement (see Chapters 5 and 7 for additional information on how such spills can affect
 4    drinking water resources). For example, a documented well communication event near Innisfail,
 5    Alberta, Canada (see Text Box 6-4) occurred when several well components failed because they
 6    were not rated to handle the increased pressure caused by the well communication (ERCB. 2012).
 7    In addition, if the fractures were to intersect an uncemented portion of the wellbore, the fluids
 8    could potentially migrate into any formations that are uncemented along the wellbore.


      Text Box 6-4.  Well Communication at a Horizontal Well near Innisfail, Alberta, Canada.

 9    In most cases, well communication during fracturing may only result in a pressure surge accompanied by a
10    drop in gas production and additional flow of produced water or fracturing fluid at an offset well. However, if
11    the offset well is not capable of withstanding the high pressures of fracturing, more significant damage can
12    occur.

13    In January 2012, fracturing at a horizontal well near Innisfail in Alberta, Canada, caused a surface spill of
14    fracturing and formation fluids at a nearby operating vertical oil well. According to the investigation report by
15    the Alberta Energy Resources Conservation Board (ERCB. 20121. pressure began rising at the vertical well
16    less than two hours after fracturing ended at the horizontal well.

17    Several components of the vertical well facility—including surface piping, discharge hoses, fuel gas lines, and
18    the pressure relief valve associated with compression at the well—were not rated to handle the increased
19    pressure and failed. Ultimately, the spill released an estimated 19,816 gallons (75 m3) of fracturing fluid,
20    brine, gas, and oil covering an area of approximately 656 ftby 738 ft (200 m by 225 m).

21    The ERCB determined thatthe lateral of the horizontal well passed within 423 ft (129 m) of the vertical well
22    at a depth of approximately 6,070 ft (1,850 m) below the surface, in the same formation. The operating
23    company had estimated a fracture half-length of 262 to 295 ft (80 to 90 m) based on a general fracture model
24    for the field. While there were no regulatory requirements for spacing hydraulic fracturing operations in
25    place at the time, the 423 ft (129 m) distance was out of compliance with the company's internal policy to
26    space fractures  from adjacent wells at least 1.5 times the predicted half-length. The company also did not
27    notify the operators of the vertical well of the fracturing operations. The incident prompted the ERCB to issue
28    Bulletin 2012-02—Hydraulic Fracturing: Interwellbore Communication between Energy Wells, which outlines
29    expectations for avoiding well communication events and preventing adverse effects on offset wells.

30    In older wells  near a hydraulic fracturing operation, plugs and cement may have degraded over
31    time; in some cases, abandoned wells may never have been plugged properly. Before the 1950s,
32    most well plugging efforts were focused on preventing water from the surface from entering oil
33    fields. As a result, many wells from that period were abandoned with little or no cement (NPC.
34    2011b). This can be a significant issue in areas with legacy (i.e., historic) oil and gas exploration and
35    when wells are re-entered and fractured (or re-fractured) to increase production in a reservoir. In
36    one study, 18 of 29 plugged and abandoned wells  in Quebec were found to show signs of leakage
37    (Council of Canadian Academies. 2014). Similarly, a PA DEP report cited three cases where natural
38    gas migration  had been caused by well communication events with old, abandoned wells (PA DEP,
                    This document is a draft for review purposes only and does not constitute Agency policy.
      June 2015                                     6-46                   DRAFT—DO NOT CITE OR QUOTE

-------
      Hydraulic Fracturing Drinking Water Assessment
 1    2009b). The Interstate Oil and Gas Compact Commission flOGCC. 2008] estimates that over 1
 2    million wells may have been drilled in the United States prior to a formal regulatory system, and the
 3    status and location of many of these wells are unknown. Various state programs exist to plug
 4    identified orphaned wells, but they face the challenge of identifying and addressing a large number
 5    of wells.1 For example, as of 2000, PADEP's well plugging program reported that it had
 6    documented 44,700 wells that had been plugged and 8,000 that were in need of plugging, and
 7    approximately 184,000 additional wells with an unknown location and status [PA PEP. 2000]. A
 8    similar evaluation from New York State found that the number of unplugged wells was growing in
 9    the state despite an active well plugging program [Bishop, 2013].

10    The Reagan etal. [2015] numerical modeling study included an assessment of migration via an
11    offset well as part of its investigation of potential fluid migration from a producing formation into a
12    shallower ground water unit (see Section 6.3.2.2]. In the offset well pathway, it was assumed that
13    the hydraulically induced fractures intercepted an  older offset well with deteriorated components.
14    (This assessment can also be applicable to cases where potential migration may occur via the
15    production well-related pathways discussed in Section 6.2.] More specifically, this analysis was
16    designed to assess transport through deteriorating cement between the subsurface formations and
17    the outermost casing, through voids resulting from incomplete cement coverage, through breached
18    tubing, or in simpler well installations without multiple casings. The highest permeability value
19    tested for the connecting feature represented a case with an open wellbore. A key assumption for
20    this investigation was that the offset well was already directly connected to a permeable feature in
21    the reservoir or within the overburden. Similar to the cases for permeable faults or fractures
22    discussed in the previous section, the study investigated the effect of multiple well- and formation-
23    related variables on potential fluid migration (see Table 6-3].

24    Based on the simulation results, an offset well pathway may have a greater potential for gas release
25    from the production zone  into a shallower ground water unit than the fault/fracture pathway
26    discussed in Section 6.3.2.2 (Reagan etal.. 2015]. This difference is primarily due to  the total pore
2 7    volume of the connecting pathway within the offset well; the offset well pathway may have a
28    significantly lower pore volume compared to the fault/fracture pathway, which reduces possible
29    gas storage in the connecting feature and increases the speed of buoyancy-dependent migration.
30    However, as with the fault/fracture scenario, the gas available for migration in this case is still
31    limited to the gas that is initially stored in the hydraulically induced fractures. Therefore, any
32    incidents of gas breakthrough observed in this study were found to be  limited in both duration and
33    magnitude.

34    Reagan etal. (2015] found that production at the gas well (the well used for hydraulic fracturing]
35    also affects the potential upward migration of gas and its arrival times  at the drinking water
36    formation due to its effect on the driving forces (e.g., pressure gradient]. Similar to the
37    fault/fracture cases described in Section 6.3.2.2, production in the target reservoir appears to
38    mitigate upward gas migration, both by reducing the amount of gas that might otherwise be
      1 An orphaned well is an inactive oil or gas well with no known (or financially solvent) owner.


                   This document is a draft for review purposes only and does not constitute Agency policy.
      June 2015                                    6-47                  DRAFT—DO NOT CITE OR QUOTE

-------
      Hydraulic Fracturing Drinking Water Assessment
 1    available for upward migration and creating a pressure gradient toward the production well. Only
 2    scenarios without the mitigating feature of gas production result in any upward migration into the
 3    aquifer. This assessment also found a generally downward water flow within the connecting well
 4    pathway, which is more pronounced when the gas well is operating. The producing formation and
 5    aquifer permeabilities appear not to be significant factors for upward gas migration via this
 6    pathway. In addition, Reagan etal. [2015] found the permeability of the connecting offset well to be
 7    one of the main factors affecting the migration of gas to the aquifer and the water well. Very low
 8    permeabilities (less than 1 mD) lead to no migration of gas into the aquifer regardless of the
 9    vertical separation distance, whereas larger permeabilities present a greater potential for gas
10    breakthrough.

11    In the same way that fractures can propagate to intersect offset wells, they can also potentially
12    intersect other artificial subsurface structures including mine shafts or solution mining sites. No
13    known incidents of this type of migration have been documented. However, the Bureau of Land
14    Management (BLM) has identified over 28,000 abandoned mines in the United States and is adding
15    new mines to  its inventory every year [BLM. 2013a]. In addition, the Well File Review identified an
16    estimated 800 cases where wells used for hydraulic fracturing were drilled through mining voids,
17    and an additional 90 cases of drilling through gas storage zones or wastewater disposal zones [U.S.
18    EPA, 2015o]. The analysis suggests that emplacing cement within such zones may be challenging,
19    which, in turn, could lead to a loss of zonal isolation (as described in Section 6.2) and create a
2 0    pathway for fluid migration.

      6.3.2.4.  Migration via Fractures Intersecting Geologic Features
21    Potential fluid migration via natural fault or fracture zones in conjunction with hydraulic fracturing
22    has been recognized as a potential contamination hazard for several decades (Harrison. 1983].
23    While porous  flow in unfractured shale or tight sand formations is assumed to be negligible due to
24    very low formation permeabilities (as discussed in Section 6.3.2.1), the presence of natural
25    "microfractures" within tight sand or shale formations is widely recognized, and these fractures
26    affect fluid flow and production strategies. Naturally occurring permeable faults and larger scale
27    fractures within or between formations may allow for more significant flow pathways for migration
28    of fluids outof the production zone (Tacksonetal., 2013c: Myers, 2012a). Figure 6-4 illustrates the
29    concept of induced fractures intersecting with natural faults or fractures extending out of the target
30    reservoir.

31    Natural fracture systems have a strong influence on the success of a fracture treatment, and the
32    topic has been studied extensively from the perspective of optimizing treatment design (e.g.. Weng
33    etal.. 2011: DahiTaleghani and Olson. 2009: Vulgamore etal.. 2007). Small natural fractures,
34    known as "microfractures," could affect fluid flow patterns near the induced fractures by increasing
35    the effective contact area. Conversely, the natural microfractures could act as capillary traps for the
36    fracturing fluid during treatment (contributing to fluid leakoff) and potentially hinder hydrocarbon
37    flow due to lower gas relative permeabilities (Dahi Taleghani etal.. 2013). Rutledge and Phillips
38    (2003) suggested that, for a hydraulic fracturing operation in East Texas, pressurizing existing
39    fractures (rather than creating new hydraulic fractures) may be the primary process that controls
                   This document is a draft for review purposes only and does not constitute Agency policy.
      June 2015                                    6-48                   DRAFT—DO NOT CITE OR QUOTE

-------
      Hydraulic Fracturing Drinking Water Assessment
 1    enhanced permeability and fracture network conductivity at the site. Ciezobka and Salehi [2013]
 2    used microseismic data to investigate the effects of natural fractures in the Marcellus Shale and
 3    concluded that fracture treatments are more efficient in areas with clusters or "swarms" of small
 4    natural fractures, while areas without these fracture swarms require more thorough stimulation.
 5    However, there is very little attention given in the literature to studying unintended fluid migration
 6    during hydraulic fracturing operations due to existing micro fractures.

 7    In some areas, larger-scale geologic features may affect potential fluid flow pathways. As discussed
 8    in Text Box 6-2, baseline measurements taken before shale gas development show evidence of
 9    thermogenic methane in some shallow aquifers, suggesting that natural subsurface pathways exist
10    and allow for naturally occurring migration of gas over millions of years [Robertson etal.. 2012].
11    There is also evidence demonstrating that gas undergoes mixing in subsurface pathways
12    fBaldassareetal.. 2014: Molofsky etal.. 2013:  Fountain and Tacobi.  20001 Warner etal. f20121
13    compared recent sampling results to data published in the 1980s and found geochemical evidence
14    for migration of fluids through natural pathways between deep underlying formations and shallow
15    aquifers—pathways that the authors suggest could lead to contamination from hydraulic fracturing
16    activities. In northeastern Pennsylvania, there is evidence that brine from deep saline formations
17    has migrated into shallow aquifers over geologic time, preferentially following certain geologic
18    structures [Llewellyn, 2014]. As described in Chapter 7, karst features (created by the dissolution
19    of soluble rock] can also serve as a potential pathway of fluid movement on a faster time scale.

2 0    Monitoring data show that the presence of natural faults and fractures can affect both the height
21    and width of hydraulic fractures. When faults are present, relatively larger microseismic responses
2 2    are seen and larger fracture growth can occur, as described below. Concentrated swarms of natural
2 3    fractures within a shale formation can result in a fracture network with a larger width-to-height
24    ratio (i.e., a shorter and wider network] than would be expected in  a zone with a low degree of
25    natural fracturing (Ciezobka and Salehi. 2013].

26    A few studies have used monitoring data to  specifically investigate the effect of natural faults and
2 7    fractures on the vertical extent of induced fractures. A statistical analysis of microseismic data by
28    Shapiro etal. (2011] found that fault rupture from hydraulic fracturing is limited by the extent of
29    the stimulated rock volume and is unlikely to extend beyond the fracture network. (However, as
30    demonstrated by microseismic data presented by Vulgamore etal. (2007], in some settings the
31    fracture network can extend laterally for thousands of feet] In the Fisher and Warpinski (2012]
32    data set (see Section 6.3.2.2], the greatest fracture heights occurred when the hydraulic fractures
33    intersected pre-existing faults. Similarly, Hammacketal. (2014] reported that fracture growth seen
34    above the Marcellus Shale is consistent with the inferred extent of pre-existing faults at the Greene
35    County, Pennsylvania, research site (see Section 6.3.2.2 and Text Box 6-3]. The authors suggested
36    that clusters of microseismic events may have occurred where preexisting small faults or natural
37    fractures were present above the  Marcellus Shale. At a site in Ohio,  Skoumal etal. (2015] found that
38    hydraulic fracturing induced a rupture along a pre-existing fault approximately 0.6 miles (1 km]
39    from the hydraulic fracturing operation. Using a new monitoring method known as tomographic
40    fracturing imaging, Lacazette and Geiser (2013] also found vertical hydraulic fracturing fluid
                   This document is a draft for review purposes only and does not constitute Agency policy.
      June 2015                                    6-49                   DRAFT—DO NOT CITE OR QUOTE

-------
      Hydraulic Fracturing Drinking Water Assessment
 1    movement from a production well into a natural fracture network for distances of up to 0.6 miles
 2    (1 km). However, Daviesetal. [2013] questioned whether this technique actually measures
 3    hydraulic fracturing fluid movement

 4    Modeling studies have also investigated whether hydraulic fracturing operations are likely to
 5    reactivate faults and create a potential fluid migration pathway into shallow aquifers. Myers
 6    [2012a, 2012b] found that a highly conductive fault could result in rapid (<1 year) fluid migration
 7    from a deep shale zone to the surface (as described in Section 6.3.2.1). Other researchers reject the
 8    notion that open, permeable faults would coexist with hydrocarbon accumulation [Fiewelling etal.,
 9    2013), although it is unclear whether the existence of faults in low permeability reservoirs would
10    affect the accumulation of hydrocarbons because, under natural conditions, the flow of gas maybe
11    limited due to capillary tension. Results from another recent modeling study suggest that, under
12    specific circumstances, interaction with a conductive fault could result in fluid migration to the
13    surface only on longer (ca. 1,000 year) time scales [Gassiatetal., 2013). Rutqvist et al. [2013) found
14    that, while somewhat larger microseismic events are possible in the presence of faults, repeated
15    events and aseismic slip would amount to a total rupture length of 164 ft (50 m) or less along a
16    fault, not far enough to allow fluid migration between a deep gas reservoir and a shallow aquifer. A
17    follow-up study using more sophisticated three-dimensional modeling techniques also found that
18    deep hydraulic fracturing is unlikely to create a direct flow path into a shallow aquifer, even when
19    fracturing fluid is injected directly into a fault [Rutqvist etal., 2015). Similarly, a modeling study
20    that investigated potential fluid migration from hydraulic fracturing in Germany found potential
21    vertical fluid migration up to 164 ft (50 m) in a scenario with high fault zone permeability, although
22    the authors note this is likely an overestimate because their goal was to "assess an upper margin of
23    the risk" associated with fluid transport (Lange etal., 2013). More generally, results from Rutqvist
24    etal. (2013) indicate that fracturing along an initially impermeable fault (as is expected in a shale
25    gas formation) would result in numerous small microseismic events that act to prevent larger
26    events from occurring (and, therefore, prevent the creation of more extensive potential pathways).

2 7    Other conditions in addition to the physical presence of a pathway would need to exist for fluid
28    migration to a drinking water resource to occur. The modeling study conducted by (Reagan etal.,
29    2015) discussed in Section 6.3.2.2 indicates that, if such a permeable feature exists, the transport of
30    gas and fluid flow would strongly depend upon the production regime and, to a lesser degree, the
31    features' permeability and the separation between the reservoir and the aquifer. In addition, the
32    pressure distribution within the reservoir (e.g., over-pressurized vs. hydrostatic conditions) will
33    affect the fluid flow through fractures/faults. As a result, the presence of multiple natural and we 11-
34    based factors may increase the potential for fluid migration into drinking water resources. For
35    example, in the Mamm Creek area of Colorado (see Section 6.2.2.2), well integrity and drilling-
36    related problems likely acted in concert with natural fracture systems to result in a gas seep into
37    surface water and shallow ground water (Crescent, 2011).

      6.4. Synthesis
38    In the injection stage of hydraulic fracturing, operators inject fracturing fluids into a well under
39    high pressure. These fluids flow through the well and into the surrounding formation, where they

                    This document is a draft for review purposes only and does not constitute Agency policy.
      June 2015                                    6-50                  DRAFT—DO NOT CITE OR QUOTE

-------
      Hydraulic Fracturing Drinking Water Assessment
 1    increase pore pressure and create fractures in the rock, allowing hydrocarbons to flow through the
 2    fractures and up the well.

 3    The production well and the surrounding geologic features function as a system that is often
 4    designed with multiple elements that can isolate hydrocarbon-bearing zones and water-bearing
 5    zones, including drinking water resources, from each other. This physical isolation optimizes oil
 6    and gas production and can protect drinking water resources via isolation within the well (by the
 7    casing and cement) and the presence of multiple layers of subsurface rock between the target
 8    formations where hydraulic fracturing occurs and drinking water aquifers.

      6.4.1. Summary of Findings
 9    Potential pathways for impacts on drinking water (i.e., the movement of hydrocarbons, formation
10    brines, or other fracturing-related fluids into drinking water resources), may be linked to one or
11    more components of the well and/or features of the subsurface system. If present, these potential
12    pathways can, in combination with the high pressures under which fluids are injected and pressure
13    changes within the subsurface, have an impact on drinking water resources.

14    The potential for these pathways to exist or form has been investigated through modeling studies
15    that simulate subsurface responses to hydraulic fracturing, and demonstrated via case studies and
16    other monitoring efforts. In addition, the development of some of these pathways—and fluid
17    movement along them—has been documented.

18    It is important to note that the development of one pathway within this system does not necessarily
19    result in an impact to a drinking water resource. For example, if cracks were to form in the cement
20    of a well, the vertical distance between the production zone and a drinking water resource (and the
21    multiple layers of rock in between) could isolate and protect the drinking water aquifer if pressures
22    were insufficient to allow fluid movement to the level of the drinking water resource. Conversely, if
23    an undetected fault were present in a rock formation, intact cement within the production well
24    could keep fluids from migrating up along the well to the fault and protect drinking water
25    resources.

      6.4.1.1.  Fluid Movement via the Well
26    A production well undergoing hydraulic fracturing is subject to higher stresses during the relatively
27    brief hydraulic fracturing phase than during any other period of activity in the life of the well. These
28    higher stresses may contribute to the formation of potential pathways associated with the casing or
29    cement that can result in the unintentional movement of fluids through the production wellbore if
30    the well cannot withstand the stresses experienced during hydraulic fracturing operations (see
31    Section 6.2).

32    Multiple barriers within the well, including casing, cement, and a completion assembly, isolate
33    hydrocarbon-bearing formations from drinking water resources. However, inadequate
34    construction, defects in or degradation of the casing or cement, or the absence of redundancies such
35    as multiple layers of casing, can allow fluid movement, which can then affect the quality of drinking
36    water resources. Ensuring proper well design and mechanical integrity—particularly proper

                   This document is a draft for review purposes only and does not constitute Agency policy.
      June 2015                                    6-51                  DRAFT—DO NOT CITE OR QUOTE

-------
      Hydraulic Fracturing Drinking Water Assessment
 1    cement placement and quality—are important actions for preventing unintended fluid migration
 2    along the wellbore.

      6.4.1.2. Fluid Movement within Subsurface Geologic Formations
 3    Potential subsurface pathways for fluid migration include flow of fluids out of the production zone
 4    into formations above or below it, fractures extending out of the production zone or into other
 5    induced fracture networks, intersections of fractures with abandoned or active wells, and fractures
 6    intersecting with faults or natural fractures (see Section 6.3).

 7    Vertical separation between the production zone where hydraulic fracturing operations occur and
 8    drinking water resources, and lateral separation between wells undergoing hydraulic fracturing
 9    and other wells can reduce the potential for fluid migration that can impact drinking water
10    resources.

11    Well communication incidents or "frac hits" have been reported in New Mexico, Oklahoma, and
12    other locations. While some operators design fracturing treatments to communicate with the
13    fractures of another well and optimize production, unintended communication between two
14    fracture systems can lead to spills in the offset well and is an indicator of hydraulic fracturing
15    treatments extending beyond their planned design. Surface spills from well communication
16    incidents have been documented in the literature, which provides evidence for occurrence of frac
17    hits. Based on the available information, frac hits most commonly occur on multi-well pads and
18    when wells are spaced less than 1,100 ft (335 m) apart, butthey have been observed at wells up to
19    8,422 ft (2,567 m) away from a well undergoing hydraulic fracturing.

      6.4.1.3. Impacts to Drinking Water Resources
20    We identified an impact on drinking water resources associated with hydraulic fracturing
21    operations in Bainbridge, Ohio. Failure to cement over-pressured formations through which the
22    production well passed—and proceeding with the fracturing operation without adequate cement
23    and an extended period during which the well was shut in—led to a buildup of natural gas within
24    the well annulus and high pressures within the well. This ultimately resulted in movement of gas
25    from the production zone into local drinking water aquifers (see Section 6.2.2.2).

26    Casings at a production well near Kill deer, North Dakota, ruptured following a pressure spike
27    during hydraulic fracturing, allowing fluids to escape to the surface. Brine and tert-butyl alcohol
28    were detected in two nearby water wells. Following an analysis of potential sources, the only
29    potential source consistent with the conditions observed in the two impacted wells was the well
30    that ruptured. There is also evidence that out-of-zone fracturing occurred at the well (see Sections
31    6.2.2.1 and 6.3.2.2).
32    There are other cases where hydraulic fracturing could be a contributing cause to impacts on
33    drinking water resources, or where the specific mechanism that led to an impact on a drinking
34    water resource cannot be definitively determined. For example:

35         •   Migration of stray gas into drinking water resources involves many potential routes for
36            migration of natural gas, including poorly constructed casing and naturally existing or

                   This document is a draft for review purposes only and does not constitute Agency policy.
      June 2015                                    6-52                  DRAFT—DO NOT CITE OR QUOTE

-------
      Hydraulic Fracturing Drinking Water Assessment
 1            induced fractures in subsurface formations. Multiple pathways for fluid movement may be
 2            working in concert in northeastern Pennsylvania (possibly due to cement issues or
 3            sustained casing pressure) and the Raton Basin in Colorado (where fluid migration may
 4            have occurred along natural rock features or faulty well seals). While the sources of
 5            methane identified in drinking water wells in each study area could be determined with
 6            varying degrees of certainty, attempts to definitively identify the pathways of migration
 7            have generally been inconclusive (see Text Box 6-2).
 8         •   At the East Mamm Creek drilling area in Colorado, inadequate  placement of cement
 9            allowed the migration of methane through natural faults and fractures in the area. This
10            case illustrates how construction issues, sustained casing pressure, and the presence of
11            natural faults and fractures, in conjunction with elevated pressures associated with well
12            stimulation, can work together to create a pathway for fluids to migrate toward drinking
13            water resources (see Sections 6.2.2.2 and 6.3.2.4).
14    Additionally, some hydraulic fracturing operations involve the injection of fluids into formations
15    where there is relatively limited vertical separation from drinking water resources. The  EPA
16    identified an estimated 4,600 wells that were located in areas with less  than 2,000 ft (610 m) of
17    vertical separation between the fractures and the base of protected ground water resources.

18    There are places in the subsurface where oil and gas reservoirs and drinking water resources co-
19    exist in the same formation.  Evidence we examined suggests that some  hydraulic fracturing for oil
20    and gas occurs within formations where the ground water has a salinity of less than 10,000 mg/L
21    TDS. By definition, this results in the introduction of fracturing fluids into  formations that meet the
22    Safe Drinking Water Act (SDWA) salinity-based definition of a source of drinking water and the
23    broader definition of a drinking water resource developed for this assessment. According to the
24    data we examined, these formations are generally in the western United States.

25    The practice of injecting fracturing fluids into a formation that also contains a drinking water
26    resource directly affects the quality  of that water, since it is likely some  of that fluid remains in the
27    formation following hydraulic fracturing. Hydraulic fracturing in a drinking water resource may be
28    of concern in the short-term (where people  are currently using these zones as a drinking water
29    supply) or the long-term (if drought or other conditions necessitate the future use of these zones
30    for drinking water).

31    There are other cases in which production wells associated with hydraulic fracturing are alleged to
32    have caused drinking water contamination.  Data limitations in most of those cases (including the
33    unavailability of information in litigation settlements resulting in sealed documents) make it
34    impossible to definitively assess whether or not hydraulic fracturing was a cause of the
35    contamination in these cases.

      6.4.2.  Factors Affecting Frequency and Severity of Impacts
36    Proper cementing across oil-, gas-, or water-bearing zones prevents the movement of brines, gas, or
3 7    hydraulic fracturing fluids along the well into drinking water resources. The likelihood of
38    contamination is reduced when the  well is fully cemented across these zones; however, this is not

                   This document is a draft for review purposes only and does not constitute Agency policy.
      June 2015                                    6-53                   DRAFT—DO NOT CITE OR QUOTE

-------
      Hydraulic Fracturing Drinking Water Assessment
 1    the case in all hydraulically fractured wells, either because the cement does not extend completely
 2    through the base of the drinking water resource or the cement that is present is not of adequate
 3    quality. Fully cemented surface casing that extends through the base of drinking water resources is
 4    a key protective component of the well. Most, but not all, wells used in hydraulic fracturing
 5    operations have fully cemented surface casing.

 6    Deviated and horizontal wells, which are increasingly being used in hydraulic fracturing operations,
 7    may exhibit more casing and cement problems compared to vertical wells. Sustained casing
 8    pressure—a buildup of pressure within the well annulus that can indicate the presence of small
 9    leaks—occurs more frequently in deviated and horizontal wells compared to vertical wells. Cement
10    integrity problems can also arise as a result of challenges in placing cement in these wells, because
11    they are more challenging than vertical wells to center properly.

12    Older wells may exhibit more integrity problems compared to newer  wells, which may be an issue
13    if older wells are hydraulically fractured or re-fractured. Degradation of the casing and cement as
14    they age or the cumulative effects of stresses exerted on the well over time may result in changes in
15    well integrity. Integrity problems can also be associated with the inadequate  design of wells that
16    were constructed pursuant to older, less stringent requirements. Well components that are subject
17    to corrosive environments, high pressures, or other stressors tend to  have more problems than
18    wells without these additional stressors.

19    The extent of subsurface fluid migration within subsurface rock formations and the potential for the
2 0    development of pathways that can  adversely affect drinking water depend on site-specific
21    characteristics. These include the physical separation between the production zone and drinking
22    water resources, the geological and geomechanical characteristics of the formations, hydraulic
23    fracturing operational parameters, and the physical characteristics of any connecting feature (e.g.,
24    abandoned wells, faults, and natural fractures).

25    As noted above, vertical separation between the production zone and drinking water resources
26    protects drinking water. Additionally, the proximity of wells undergoing hydraulic fracturing to
27    other wells increases the potential  for the formation of pathways for fluids to move via these wells
28    to drinking water resources. For example, if there is a deficiency in the construction of a nearby
29    well (or  degradation of the well components), that well could serve as a pathway for movement of
30    fracturing fluids, methane, or brines that might affect a drinking water resource. If the fractures
31    were to intersect an uncemented portion of a nearby wellbore, the fluids could migrate along that
32    wellbore into any uncemented formations.

33    Fractures created during hydraulic fracturing can extend out of the target production zone. Out-of-
34    zone fracturing could be a concern  for fluid migration if the hydraulic fracturing operation is not
35    designed to address site-specific conditions, for example if the production zone is thin and fractures
36    propagate to unintended vertical heights, or if the production zone is  not horizontally continuous
37    and fractures extend to unintended horizontal lengths. The presence of natural faults or fractures
38    can affect the extent of hydraulic fractures. When faults are present, relatively larger microseismic
39    responses are seen during hydraulic fracturing, and larger fracture growth can occur than in the

                   This document is a draft for review purposes only and does not constitute Agency policy.
      June 2015                                    6-54                  DRAFT—DO NOT CITE OR QUOTE

-------
      Hydraulic Fracturing Drinking Water Assessment
 1    absence of natural faults or fractures. However, modeling studies indicate that fluid migration from
 2    deep production zones to shallow drinking water resources along natural faults and fractures or
 3    offset wells is unlikely. These studies indicate that, in both cases, gas available for migration is
 4    limited to the amount that existed in the fractures and pore space of connecting features following
 5    hydraulic fracturing prior to production. Following the completion of a hydraulic fracturing
 6    treatment, depressurization of the production formation surrounding the fractures due to
 7    hydrocarbon production would make upward fluid migration into drinking water resources
 8    unlikely to occur.

 9    Based on the information presented in this chapter, the increased deployment of hydraulic
10    fracturing associated with oil and gas production activities, including techniques such as horizontal
11    drilling and multi-well pads, may increase the likelihood that these pathways could develop. This, in
12    turn, could lead to increased opportunities for impacts on drinking water resources.

      6.4.3.  Uncertainties
13    Generally, less is known about the occurrence of (or potential for) impacts of injection-related
14    pathways in the subsurface than for other components of the hydraulic fracturing water cycle,
15    which can be observed and measured at the surface. Furthermore, while there is a significant
16    amount of information available on production wells in general, there is little information that is
17    specific to hydraulic fracturing operations and much of this data is not readily accessible, i.e., in a
18    centralized, national database.

      6.4.3.1. Limited A vailability of Information Specific to Hydraulic Fracturing Operations
19    There is extensive information on the design goals for hydraulically fractured oil and gas wells (i.e.,
20    to address the stresses imposed by high-pressure, high-volume injection), including from industry-
21    developed best practices documents. Additionally, based on the long history of oil and gas
22    production activities, we know how production wells are constructed and have performed over
23    time. Over the years, many studies have documented how these wells are constructed, how they
24    perform, and the rates at which they experience problems that can lead to the formation of
25    pathways for fluid movement However, because we do not know which of these wells were
26    hydraulically fractured, we cannot definitively determine whether the rates at which integrity
27    problems arise (or other data pertaining to oil and gas wells in general) directly correspond to
28    wells used in hydraulic fracturing operations.

29    Because wells that have been hydraulically fractured must withstand many of the same downhole
30    stresses as other production wells, we consider studies of the pathways for impacts to drinking
31    water resources in production wells to be relevant to identifying the potential pathways relevant to
32    hydraulic fracturing operations. However, without specific data on the as-built construction of wells
33    used in hydraulic fracturing operations, we cannot definitively state whether these wells are
34    consistently constructed to meet the stresses they may encounter.

35    There is also, in general, very limited information available on the monitoring and performance of
36    wells used in hydraulic fracturing operations. Published information is sparse regarding
37    mechanical integrity tests (MITs) performed during and after hydraulic fracturing, including MIT

                   This document is a draft for review purposes only and does not constitute Agency policy.
      June 2015                                    6-55                 DRAFT—DO NOT CITE OR QUOTE

-------
      Hydraulic Fracturing Drinking Water Assessment
 1    results, the frequency at which mechanical integrity issues arise in wells used for hydraulic
 2    fracturing, and the degree and speed with which identified issues are addressed. There is also little
 3    information available regarding MIT results for the original hydraulic fracturing in wells built for
 4    that purpose, for wells that are later re-fractured, or for existing, older wells not initially
 5    constructed for hydraulic fracturing but repurposed for that use.

 6    There are also a limited number of published monitoring studies or sampling data that provide
 7    evidence to assess whether formation brines, injected fluids, or gas move in unintended ways
 8    through the subsurface during and after hydraulic fracturing. Subsurface monitoring data (i.e., data
 9    that characterize the presence, migration, or transformation of fluids in the subsurface related to
10    hydraulic fracturing operations) are scarce relative to the tens of thousands of oil and gas wells that
11    are estimated to be hydraulically fractured across the country each year (see Chapter 2).

12    Information on fluid movement within the subsurface and the extent of fractures that develop
13    during hydraulic fracturing operations is also limited. For example, limited information is available
14    in the published literature on how flow regimes or other subsurface processes change at sites
15    where hydraulic fracturing is conducted. Instead, much of the available research, and therefore the
16    literature, addresses how hydraulic  fracturing and other production technologies perform to
17    optimize hydrocarbon production.

18    These limitations on hydraulic fracturing-specific information make it difficult to provide definitive
19    estimates of the rate at which wells used in hydraulic fracturing operations experience the types of
20    integrity problems that can contribute to fluid movement.

      6.4.3.2. Limited Systematic, Accessible Data on Well Performance or Subsurface Movement
21    While the oil and gas industry generates a large amount of information on well performance as part
22    of operations, most of this is proprietary or otherwise not readily available to states or the public in
23    a compiled or summary manner. Therefore, no national or readily accessible way exists to evaluate
24    the design and performance of individual wells or wells in a region, particularly in the context of
2 5    local geology or the presence of other wells and/or hydraulic fracturing operations. Many states
26    have large amounts of operator-submitted data, but information about construction practices or the
2 7    performance of individual wells is typically not in a searchable or aggregated form that would
2 8    enable assessments of well performance under varying settings, conditions, or timeframes.
29    Although it is collected in some cases, there is also no systematic collection, reporting, or  publishing
30    of empirical baseline (pre-drilling and/or pre-fracturing) and post-fracturing monitoring data that
31    could indicate the presence or absence of hydraulic fracturing-related fluids in shallow zones and
32    whether or not migration of those fluids has occurred. Ideally, data from ground water monitoring
33    are needed to complement theories  and modeling on potential pathways and fluid migration.

34    While some of the types of impacts described above may occur quickly (i.e., on the scale of days or
35    weeks, as with  integrity problems or well communication events), other impacts (e.g., in slow-
36    moving, deep ground waters) may only occur or be able to be detected on much longer timescales.
37    Given the surge in the number of modern high-pressure hydraulic fracturing operations dating
38    from the early 2000s, evidence of any fracturing-related fluid migration affecting a drinking water

                    This document is a draft for review purposes only and does not constitute Agency policy.
      June 2015                                   6-56                  DRAFT—DO NOT CITE OR QUOTE

-------
      Hydraulic Fracturing Drinking Water Assessment
 1    resource (as well as the information necessary to connect specific well operation practices to a
 2    drinking water impact) could take years to discover.

 3    The limited amount of information hinders our ability to evaluate whether—or how frequently—
 4    drinking water impacts are occurring (or the potential for these impacts to occur) or to tie possible
 5    impacts to specific well construction, operation, or maintenance practices. This also significantly
 6    limits our ability to evaluate the aggregate potential for hydraulic fracturing operations to affect
 7    drinking water resources or to identify the potential cause of drinking water contamination or
 8    suspected contamination in areas where hydraulic fracturing occurs.

      6.4.4.  Conclusions
 9    Fluids can migrate from the wellbore and surrounding subsurface formations due to inadequate
10    casing or cement, and via natural and man-made faults, fractures, and offset wells or mines (see
11    Text Box 6-5). To prevent fluid migration through the wellbore or through subsurface pathways,
12    wells must have adequate casing and cement, and induced fractures must not intersect existing
13    fractures or permeable zones that lead to  drinking water resources. Evidence shows that the quality
14    of drinking water resources may have been affected by hydraulic fracturing fluids escaping the
15    wellbore and surrounding formation in certain areas, although conclusive evidence is currently
16    limited.
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
Text Box 6-5. Research Questions Revisited.

How effective are current well construction practices at containing fluids—both liquids and gases—
before, during, and after fracturing?
•   Wells that were designed with uncemented intervals of casing across porous or permeable zones, wells in
    which cementing does not resist formation or operational stresses, and wells in which cementing does
    not meet design specifications have the potential to promote unintended subsurface fluid movement.
    Even in optimally designed wells, metal casings and cement can degrade over time, either as a result of
    aging or of exposure to stresses exerted over years of operations. See Section 6.2.2.2.
•   We have limited information on the degree to which wells are designed and constructed with the
    multiple layers of casing that can withstand hydraulic fracturing pressures and contact with injected and
    produced fluids. We also are lacking information about whether wells have suitable cements that can
    prevent fluid movement outside the wellbore and between the production zone and drinking water
    resources. We also do not have information on the degree to which mechanical integrity is verified before
    or after hydraulic fracturing operations. See Section 6.2.2.1.
Can subsurface migration of fluids—both liquids and gases—to drinking water resources occur and
what local geologic or artificial features might allow this?
•   The presence of artificial penetrations, especially poorly constructed offset wells or undetected
    abandoned wells, mines, or other subsurface structures, provides pathways that, in the presence of a
    driving force, could allow for fluid movement to shallow geologic zones such as drinking water resources.
    See Section 6.3.2.3.
                    This document is a draft for review purposes only and does not constitute Agency policy.
      June 2015                                     6-57                  DRAFT—DO NOT CITE OR QUOTE

-------
     Hydraulic Fracturing Drinking Water Assessment
1    •   Intersections of induced fractures with transmissive faults or naturally occurring fractures or
2        porous/permeable rock zones can allow fluids to move out of the targeted fracture areas. However,
3        modeling studies indicate that fluid migration from production zones to drinking water resources along
4        natural faults and fractures is unlikely. See Section 6.3.2.4.

5    •   Some hydraulic fracturing operations involve the injection of fluids into formations where there is
6        relatively limited vertical separation from drinking water resources. Other hydraulic fracturing is
7        performed within formations that meet the SDWA or state salinity-based definition of a source of
8        drinking water, in addition to the broader definition of a drinking water resource developed for this
9        assessment. See Section 6.3.2.
     6.5.  References for Chapter 6

     Adachi. I: Siebrits. E: Peirce. A: Desroches. I. (2007). Computer simulation of hydraulic fractures. International
        Journal of Rock Mechanics and Mining Sciences 44: 739-757.
        http://dx.doi.0rg/10.1016/j.ijrmms.2006.ll.006

     Ajani. A: Kelkar. M. (2012). Interference study in shale plays. Paper presented at SPE Hydraulic Fracturing
        Technology Conference, February 6-8, 2012, The Woodlands, TX.

     Ali. M: Taoutaou. S: Shafqat. AU: Salehapour. A: Noor. S. (2009). The use of self healing cement to ensure long
        term zonal isolation for HPHT wells subject to hydraulic fracturing operations in Pakistan. Paper
        presented at International Petroleum Technology Conference, December 7-9, 2009, Doha, Qatar.

     ALL Consulting (ALL Consulting, LLC). (2004). Coal bed methane primer: New source of natural gas and
        environmental implications. Tulsa, OK: U.S. Department of Energy, National Petroleum Technology Center.
        http://bogc.dnrc.mt.gov/PDF/Web%20Version.pdf

     Arkadakskiy. S: Rostron. B. (2012a). Stable isotope geochemistry helps in reducing out-of-zone hydraulic
        fracturing and unwanted brine production from the Bakken Reservoir. Available online at
        http://isobrine.com/resources/

     Arkadakskiy. S: Rostron. B. (2013a). Tracking out-of-zone hydraulic fracturing in the Bakken with naturally
        occuring tracers. Paper presented at GeoConvention 2013: Integration, May 6-10, 2013, Calgary, Alberta.

     Bachu. S: Bennion. DB. (2009). Experimental assessment of brine and/or C02  leakage through well cements
        at reservoir conditions. Int J Greenhouse Gas Control 3: 494-501.
        http://dx.doi.0rg/10.1016/i.iiggc.2008.ll.002

     Bair. ES: Freeman. DC: Senko. IM. (2010). Subsurface gas invasion Bainbridge  Township, Geauga County, Ohio.
        (Expert Panel Technical Report). Columbus, OH: Ohio Department of Natural Resources.
        http://oilandgas.ohiodnr.gov/resources/investigations-reports-violations-reformstfTHR

     Baldassare, F. (2011). The origin of some natural gases in Permian through Devonian Age systems in the
        Appalachian Basin and the relationship to incidents of stray gas migration. Presentation presented at
        Technical workshop for hydraulic fracturing study, chemical and analytical methods, February2425,2011,
        Arlington, VA.

     Baldassare. Fl: McCaffrey. MA: Harper. IA. (2014). A geochemical context for stray gas investigations in the
        northern Appalachian Basin: Implications of analyses of natural gases from Neogene-through Devonian-
        age strata. AAPG Bulletin 98: 341-372. http://dx.doi.org/10.1306/06111312178
                    This document is a draft for review purposes only and does not constitute Agency policy.

     June 2015                                       6-58                    DRAFT—DO NOT CITE OR QUOTE

-------
Hydraulic Fracturing Drinking Water Assessment
Barker. IF: Fritz. P. (1981). Carbon isotope fractionation during microbial methane oxidation. Nature 293:
   289-291. http://dx.doi.org/10.1038/293289aO

Bertoncello. A: Wallace. I: Honarpour. MM: Kabir. C: Blyton. CA. (2014). Imbibition and water blockage in
   unconventional reservoirs: Well management implications during flowback and early production. SPE
   Journal 17.

Bishop. RE. (2013). Historical analysis of oil and gas well plugging in New York: Is the regulatory system
   working? New Solutions: A Journal of Environmental and Occupational Health Policy 23:103-116.
   http://dx.doi.Org/10.2190/NS.23.l.g

BLM (Bureau of Land Management). (2013a). Abandoned mine lands: A new legacy. Washington, DC: U.S.
   Department of the Interior, Bureau of Land Management.
   http://www.blm.gov/pgdata/etc/medialib/blm/wo/MINERALS  REALTY  AND  RESOURCE PROTECTIO
   N /ami/ami documents.Par.81686.File.dat/AML NewLegacy.pdf

Bovd. D: Al-Kubti. S: Khedr. OH: Khan. N: Al-Navadi. K: Degouv. D: Elkadi. A: Kindi. ZA. (2006). Reliability of
   cement bond log interpretations compared to physical communication tests between formations. Paper
   presented at Abu Dhabi International Petroleum Exhibition and Conference, November 5-8, 2006, Abu
   Dhabi, UAE.

Brantley. SL: Yoxtheimer. D: Arjmand. S: Grieve. P: Vidic. R: Pollak. I: Llewellyn. GT: Abad. I: Simon. C. (2014).
   Water resource impacts during unconventional shale gas development: The Pennsylvania experience. Int J
   Coal Geol 126:140-156. http://dx.doi.Org/10.1016/j.coal.2013.12.017

Brown, HP: Grijalva, VE: Raymer, LL. (1970). New developments in sonic wave train display and analysis in
   cased holes. (SPWLA-1970-F). Brown, HD; Grijalva, VE; Raymer, LL.
   https://www.onepetro.org/conference-paper/SPWLA-1970-F

Brufatto. C: Cochran. I: Conn. L: El-Zeghaty. SZA. A: Fraboulet. B: Griffin. T: lames. S: Munk. T: lustus. F: Levine.
   IR: Montgomery. C: Murphy. D: Pfeiffer. I: Pornpoch. T: Rishmani. L. (2003). From mud to cement -
   Building gas wells. Oilfield Rev 15: 62-76.

Byrnes. AP. (2011). Role of induced and natural imbibition in frac fluid transport and fate in gas shales.
   Presentation presented at Technical Workshops for Hydraulic Fracturing Study: Fate & Transport, March
   28-29,2011, Arlington, VA.

Ciezobka. I: Salehi. I.  (2013). Controlled hydraulic fracturing of naturally fractured shales: A case study in the
   Marcellus Shale examining how to identify and exploit natural fractures. (SPE-164524-MS). Ciezobka, J;
   Salehi, I. http://dx.doi.org/10.2118/164524-MS

COGCC. Colorado Oil and Gas Conservation Commission Order No. 1V-276. (2004).
   https://cogcc.state.co.us/orders/orders/lv/276.html

Cohen. HA: Parratt. T: Andrews. CB. (2013). Comments on 'Potential contaminant pathways from
   hydraulically fractured shale to aquifers' [Comment]. Ground Water 51: 317-319; discussion 319-321.
   http://dx.doi.org/10.llll/gwat.12015

Considine. T: Watson. R: Considine. N: and Martin. I. (2012). Environmental impacts during Marcellus shale
   gas drilling: Causes, impacts, and remedies. (Report 2012-1). Buffalo, NY: Shale Resources and Society
   Institute. http://cce.cornell.edu/EnergyClimateChange/NaturalGasDev/Documents/UBSRSI-
   Environmental%20Impact%20Report%202012.pdf

Council of Canadian Academies. (2014). Environmental impacts of shale gas extraction in Canada. Ottawa,
   Ontario.
   http://www.scienceadvice.ca/uploads/eng/assessments%20and%20publications%20and%20news%20
   releases/Shale%20gas/ShaleGas fullreportEN.pdf

Craig. MS: Wendte. SS: Buchwalter. IL. (2012). Barnett shale horizontal restimulations: A case study of 13
   wells. SPE Americas unconventional resources conference, June 5-7, 2012, Pittsburgh, PA.
               This document is a draft for review purposes only and does not constitute Agency policy.

June 2015                                      6-59                   DRAFT—DO NOT CITE OR QUOTE

-------
Hydraulic Fracturing Drinking Water Assessment
Crescent (Crescent Consulting, LLC). (2011). East Mamm creek project drilling and cementing study.
   Oklahoma City, OK. http://cogcc.state.co.us/Library/PiceanceBasin/EastMammCreek/ReportFinal.pdf

Crook. R. (2008). Cementing: Cementing horizontal wells. Halliburton.

Dahi Taleghani. A: Ahmadi. M: Olson. IE. (2013). Secondary fractures and their potential impacts on hydraulic
   fractures efficiency. In AP In Bunger; J McLennan; R Jeffrey (Eds.), Effective and sustainable hydraulic
   fracturing. Croatia: InTech. http://dx.doi.org/10.5772/56360

Dahi Taleghani. A: Olson. IE. (2009). Numerical modeling of multi-stranded hydraulic fracture propagation:
   Accounting for the interaction between induced and natural fractures. In 2009 SPE Annual Technical
   Conference and Exhibition. Richardson, TX: Society of Petroleum Engineers.
   http://dx.doi.org/10.2118/124884-MS

Pake. LP. (1978). Fundamentals of reservoir engineering. Boston, MA: Elsevier.
   http://www.ing.unp.edu.ar/asignaturas/reservorios/Fundamentals%20of%20Reservoir%20Engineering
   %20%28LP.%20Dake%29.pdf

Daneshy, AA. (2009).  Factors controlling the vertical growth of hydraulic fractures. (SPE-118789-MS).
   Daneshy, AA. http://dx.doi.org/10.2118/118789-MS

Darrah. TH: Vengosh.  A: Jackson. RB: Warner. NR: Poreda. Rl. (2014). Noble gases identify the mechanisms of
   fugitive gas contamination in drinking-water wells overlying the Marcellus and Barnett Shales. PNAS 111:
   14076-14081. http://dx.doi.org/10.1073/pnas.1322107111

Davies. Rl: Almond. S: Ward. RS: lackson. RB: Adams. C: Worrall. F: Herringshaw. LG: Gluyas. IG: Whitehead.
   MA. (2014). Oil and gas wells and their integrity: Implications for shale and unconventional resource
   exploitation. Marine and Petroleum Geology 56: 239-254.
   http://dx.doi.0rg/10.1016/j.marpetgeo.2014.03.001

Davies. Rl: Foulger. GR: Mathias. S: Moss. I: Hustoft. S: Newport. L. (2013). Reply: Davies etal.  (2012),
   Hydraulic fractures:  How far can they go? Marine and Petroleum Geology 43: 519-521.
   http://dx.doi.0rg/10.1016/j.marpetgeo.2013.02.001

Davies. Rl: Mathias. SA: Moss. I: Hustoft. S: Newport. L. (2012). Hydraulic fractures: How far can they go?
   Marine and Petroleum Geology 37:1-6. http://dx.doi.Org/10.1016/j.marpetgeo.2012.04.001

De Pater. Cl: Baisch. S. (2011). Geomechanical study of the Bowland shale seismicity: Synthesis report.
   Nottingham, England: British Geological Survey.
   https://www.bucknell.edu/script/environmentalcenter/marcellus/default.aspx?articleid=MF08SXMW82
   CV1MOAXK7ZINIPP

Dehghanpour. H: Lan. 0: Saeed. Y: Fei. H: Oi. Z. (2013). Spontaneous imbibition of brine and oil in gas shales:
   Effect of water adsorption and resulting microfractures. Energy Fuels 27: 3039-3049.
   http://dx.doi.org/10.1021/ef4002814

Dehghanpour. H: Zubair. HA: Chhabra. A: Ullah. A. (2012). Liquid intake of organic shales. Energy Fuels 26:
   5750-5758. http://dx.doi.org/10.1021/ef3009794

Drillinglnfo. Inc.. (2014b). Drillinglnfo Inc. DI Desktop raw data feed [Database].

Dusseault. MB: Gray. MN: Nawrocki. PA. (2000). Why oilwells leak: Cement behavior and long-term
   consequences. Paper presented at SPE  International Oil and Gas Conference and Exhibition in China,
   November 7-10, 2000, Beijing, China.

Dutta, R: Lee, C, -H: Odumabo, S: Ye, P: Walker, SC: Karpyn, ZT: Ayala, LF. (2014). Experimental investigation
   of fracturing-fluid  migration caused by spontaneous imbibition in fractured low-permeability sands. SPE
   Reserv Eval Engin  17: 74-81.
               This document is a draft for review purposes only and does not constitute Agency policy.

June 2015                                       6-60                    DRAFT—DO NOT CITE OR QUOTE

-------
Hydraulic Fracturing Drinking Water Assessment
Eberhard. M. (2011). Fracture design and stimulation - monitoring. Presentation presented at Technical
   Workshops for the Hydraulic Fracturing Study: Well Construction & Operations, March 10-11, 2011,
   Arlington, VA.

Economides. Ml: Mikhailov. DN: Nikolaevskiy. VN. (2007). On the problem of fluid leakoff during hydraulic
   fracturing. Transport in Porous Media 67: 487-499. http://dx.doi.org/10.1007/sll242-006-9038-7

Eisner. L: Fischer. T: Le Calvez. IH. (2006). Detection of repeated hydraulic fracturing (out-of-zone growth) by
   microseismic monitoring. The Leading Edge (Tulsa) 25: 548-554. http://dx.doi.Org/10.1190/l.2202655

Enform. (2013). Interim industry recommended practice 24: fracture stimulation: Interwellbore
   communication 3/27/2013 (1.0 ed.). (IRP 24). Calgary, Alberta: Enform Canada.
   http://www.enform.ca/safety resources/publications/PublicationDetails.aspx?a=29&type=irp

Engelder. T. (2012). Capillary tension and imbibition sequester frack fluid in Marcellus gas shale [Letter].
   PNAS 109: E3625; author reply E3626. http://dx.doi.org/10.1073/pnas.1216133110

ERCB (Energy Resource Conservation Board). (2012). Midway Energy Ltd. hydraulic Fracturing incident:
   Interwellbore communication January 13,2012. (ERCB Investigation Report, Red Deer Field Centre).
   Calgary, Alberta: Energy Resources Conservation Board.

Fisher. M: Warpinski. N. (2012). Hydraulic fracture height growth: Real data. S P E Prod Oper 27: 8-19.
   http://dx.doi.org/10.2118/145949-PA

Fitzgerald. DP: McGhee. BF: McGuire. IA. (1985). Guidelines  for 90 % accuracy in zone-isolation decisions. J
   Pet Tech 37: 2013-2022. http://dx.doi.org/10.2118/12141-PA

Fiaer. E: Holt. RM: Horsrud. P: Raaen. AM: Risnes. R. (2008).  Petroleum related rock mechanics (2nd edition
   ed.). Amsterdam, The Netherlands: Elsevier.

Flewelling. SA: Sharma. M. (2014). Constraints on upward migration of hydraulic fracturing fluid and brine.
   Ground Water 52: 9-19. http://dx.doi.org/10.llll/gwat.12095

Flewelling. SA: Tymchak. MP: Warpinski. N. (2013). Hydraulic fracture height limits and fault interactions in
   tight oil and gas formations. Geophys Res Lett 40: 3602-3606. http://dx.doi.org/10.1002/grl.50707

Flournoy. RM: Feaster. IH. (1963). Field observations on the use of the cement bond log and its application to
   the evaluation of cementing problems. Richardson, TX: Society of Petroleum Engineers.
   http://dx.doi.org/10.2118/632-MS

Fountain. 1C: lacobi. RD. (2000). Detection of buried faults and fractures using soil gas analysis.
   Environmental and Engineering Geoscience 6: 201-208. http://dx.doi.Org/10.2113/gseegeosci.6.3.201

Gassiat. C: Gleeson. T: Lefebvre. R: Mckenzie. I. (2013). Numerical simulation of potential contamination of
   shallow aquifers over long time scales. Water Resour Res 49: 8310-8327.
   http://dx.doi.org/10.1002/2013WR014287

George. PG: Mace. RE: Petrossian. R. (2011). Aquifers of Texas. (Report 380). Austin, TX: Texas Water
   Development Board.
   http://www.twdb.state.tx.us/publications/reports/numbered reports/doc/R380 AquifersofTexas.pdf

Goodwin. Kl: Crook. Rl. (1992). Cement sheath stress failure. S P E Drilling & Completion 7: 291-296.
   http://dx.doi.org/10.2118/20453-PA

Gorody. AW. (2012). Factors affecting the variability of stray gas concentration and composition in
   groundwater. Environmental Geosciences 19:17-31. http://dx.doi.org/10.1306/eg.12081111013

Gu. H: Siebrits. E. (2008). Effect of formation modulus contrast on hydraulic fracture height containment. S P
   E Prod Oper 23:170-176. http://dx.doi.org/10.2118/103822-PA
               This document is a draft for review purposes only and does not constitute Agency policy.

June 2015                                       6-61                   DRAFT—DO NOT CITE OR QUOTE

-------
Hydraulic Fracturing Drinking Water Assessment
GWPC (Groundwater Protection Council). (2014). State oil and natural gas regulations designed to protect
   water resources. Morgantown, WV: U.S. Department of Energy, National Energy Technology Laboratory.
   http://www.gwpc.org/sites/default/files/files/Oil%20and%20Gas%20Regulation%20Report%20Hyperl
   inked%20Version%2 OFinal-rfs.pdf

GWPC and ALL Consulting (Ground Water Protection Council (GWPC) and ALL Consulting). (2009). Modern
   shale gas development in the United States: A primer. (DE-FG26-04NT15455). Washington, DC: U.S.
   Department of Energy, Office of Fossil Energy and National Energy Technology Laboratory.
   http://www.gwpc.org/sites/default/files/Shale%20Gas%20Primer%202009.pdf

Hammack. R: Harbert. W: Sharma. S: Stewart. B:  Capo. R: Wall. A: Wells. A: Diehl. R: Blaushild. D: Sams. I:
   Veloski. G. (2014). An evaluation of fracture growth and gas/fluid migration as horizontal Marcellus Shale
   gas wells are hydraulically fractured in Greene County, Pennsylvania. (NETL-TRS-3-2014). Pittsburgh, PA:
   U.S. Department of Energy, National Energy Technology Laboratory.
   http://www.netl.doe.gov/File%20Library/Research/onsite%20research/publications/NETL-TRS-3-
   2014 Greene-County-Site 20140915 1 l.pdf

Harrison. SS. (1983). Evaluating system for ground-water contamination hazards due to gas-well drilling on
   the Glaciated Appalachian Plateau. Ground Water 21: 689-700. http://dx.doi.Org/10.llll/j.1745-
   6584.1983.tb01940.x

Harrison. SS. (1985). Contamination of aquifers by overpressurizing the annulus of oil and gas wells. Ground
   Water 23: 317-324.

Heilweil. VM: Stolp. Bl: Kimball. BA: Susong. DP: Marston. TM: Gardner. PM. (2013). A stream-based methane
   monitoring approach for evaluating groundwater impacts associated with unconventional gas
   development. Ground Water 51: 511-524. http://dx.doi.org/10.llll/gwat.12079

Hoi ditch. SA. (2007). Chapter 8: Hydraulic fracturing. In JD Clegg (Ed.), Petroleum engineering handbook (pp.
   IV-323 - IV-366). Richardson, TX: Society of Petroleum Engineers. http://store.spe.org/Petroleum-
   Engineering-Handbook-Volume-IV-Production-Operations-Engineering-P61.aspx

Hyne, Nl. (2012). Nontechnical guide to petroleum geology, exploration, drilling and production. In
   Nontechnical guide to petroleum geology, exploration, drilling and production (3  ed.). Tulsa, OK: PennWell
   Corporation.

Ingraffea. AR: Wells. MT: Santoro. RL: Shonkoff. SB. (2014). Assessment and risk analysis of casing and
   cement impairment in oil and gas wells in Pennsylvania, 2000-2012. PNAS 111: 1095510960.
   http://dx.doi.org/10.1073/pnas.1323422111

IOGCC (Interstate Oil and Gas Compact Commission). (2008). Protecting our country's resources: The states'
   case, orphaned well plugging initiative. Oklahoma City, OK: Interstate Oil and Gas Compact Commission
   (IOGCC). http://iogcc.myshopify.com/products/protecting-our-countrys-resources-the-states-case-
   orphaned-well-plugging-initiative-2008

lackson. G: Flores. C: Abolo. N: Lawal. H. (2013a). A novel approach to modeling and forecasting frac hits in
   shale gas wells. Paper presented at EAGE Annual Conference & Exhibition incorporating SPE Europec,
   June 10-13, 2013, London, UK.

lackson, RB: Vengosh, A: Darrah, TH: Warner, NR: Down, A: Poreda, Rl: Osborn, SG: Zhao, K: Karr, ID. (2013b).
   Increased stray gas abundance in a subset of drinking water wells near Marcellus shale gas extraction.
   PNAS 110:11250-11255. http://dx.doi.org/10.1073/pnas.1221635110

lackson. RE: Dussealt. MB.  (2014). Gas release mechanisms from energy wellbores. Presentation presented at
   48th US Rock Mechanics/Geomechanics Symposium, June 1-4, 2014, Minneapolis, Minnesota.

lackson. RE: Gorody. AW: Mayer. B: Roy. IW: Ryan. MC: Van Stempvoort. DR. (2013c). Groundwater protection
   and unconventional gas extraction: the critical need for field-based hydrogeological research. Ground
   Water 51: 488-510. http://dx.doi.org/10.llll/gwat.12074
               This document is a draft for review purposes only and does not constitute Agency policy.

June 2015                                      6-62                   DRAFT—DO NOT CITE OR QUOTE

-------
Hydraulic Fracturing Drinking Water Assessment
lacob. R. (2011). Incident action plan, Franchuk 44-20 SWH incident. Piano, Texas: Denbury Onshore, LLC.

lones. IR: Britt. LK. (2009). Design and appraisal of hydraulic fractures. In Design and appraisal of hydraulic
   fractures. Richardson, TX: Society of Petroleum Engineers.

Kappel. WM. (2013). Dissolved methane in groundwater, Upper Delaware River Basin, Pennsylvania and New
   York (pp. 1-6). (2013-1167). U. S. Geological Survey, http://pubs.usgs.gov/of/2013/1167/pdf/ofr2013-
   1167.pdf

Kappel. WM: Nystrom. EA. (2012). Dissolved methane in New York groundwater, 19992011. (Open-File
   Report 20121162). Washington, DC: U.S. Geological Survey, http://pubs.usgs.gov/of/2012/1162/

Kell. S. (2011). State oil and gas agency groundwater investigations and their role in advancing regulatory
   reforms, a two-state review: Ohio and Texas. Ground Water Protection Council.
   http://fracfocus.org/sites/default/files/publications/state  oil  gas agency groundwater investigations o
   ptimized.pdf
Kim. GH: Wang. IY. (2014). Interpretation of hydraulic fracturing pressure in tight gas formations. Journal of
   Energy Resources Technology 136: 032903. http://dx.doi.Org/10.1115/l.4026460

Kim. I: Moridis. Gl. (2013). Development of the T+M coupled flowgeomechanical simulator to describe
   fracture propagation and coupled flowthermalgeomechanical processes in tight/shale gas systems.
   Computers and Geosciences 60:184-198. http://dx.doi.Org/10.1016/j.cageo.2013.04.023

Kim. I: Moridis. Gl. (2015). Numerical analysis of fracture propagation during hydraulic fracturing operations
   in shale gas systems. International Journal of Rock Mechanics and Mining Sciences 76:127-137.
Kim. I: Um. ES: Moridis. Gl. (2014). Fracture propagation, fluid flow, and geomechanics of water-based
   hydraulic fracturing in shale gas systems and electromagnetic geophysical monitoring of fluid migration.
   SPE Hydraulic Fracturing Technology Conference, February 4-6, 2014, The Woodlands, Texas.
King. G: King. D. (2013). Environmental risk arising from well-construction failure: Differences between
   barrier and well failure, and estimates of failure frequency across common well types, locations, and well
   age. SPE Prod Oper 28. http://dx.doi.org/10.2118/166142-PA
Kirksey, I. (2013). Optimizing wellbore integrity in well construction. Presentation presented at North
   American Wellbore Integrity Workshop, Octoberl6-17,2013, Denver, CO.

Lacazette, A: Geiser, P. (2013). Comment on Davies etal., 2012 Hydraulic fractures: How far can they go?
   Marine and Petroleum Geology 43: 516-518. http://dx.doi.Org/10.1016/j.marpetgeo.2012.12.008

Lange. T: Sauter. M: Heitfeld. M: Schetelig. K: Brosig. K: lahnke. W: Kissinger. A: Helmig. R: Ebigbo. A: Class. H.
   (2013). Hydraulic fracturing in unconventional gas reservoirs: risks in the geological system part 1.
   Environmental Earth Sciences 70: 3839-3853. http://dx.doi.org/10.1007/sl2665-013-2803-3
Lawal, H: Abolo, NU: lackson, G: Sahai, V: Flores, C. (2014). A quantitative approach to analyze fracture area
   loss in shale gas reservoirs. SPE Latin America and Caribbean Petroleum Engineering Conference, May 21-
   23, 2014, Maracaibo, Venezuela.
Lecampion. B: leffrey. R: Detournay. E. (2005). Resolving the geometry of hydraulic fractures from tilt
   measurements. Pure Appl Geophys 162: 2433-2452. http://dx.doi.org/10.1007/s00024-005-2786-4

Llewellyn. GT. (2014). Evidence and mechanisms for Appalachian Basin brine migration to shallow aquifers in
   NE Pennsylvania, USA. Hydrogeo J 22:1055-1066. http://dx.doi.org/10.1007/sl0040-014-1125-l

McDaniel, I: Watters, L: Shadravan, A. (2014). Cement sheath durability: Increasing cement sheath integrity to
   reduce gas migration in the Marcellus Shale Play. In SPE hydraulic fracturing technology conference
   proceedings. Richardson, TX: Society of Petroleum Engineers. http://dx.doi.org/10.2118/168650-MS
MCOR (Marcellus Center for Outreach and Research). (2012). Extent and thickness of Marcellus Shale.
   University Park, PA: Pennsylvania State University. Retrieved from
   http://www.marcellus.psu.edu/images/Marcellus thickness.gif


               This document is a draft for review purposes only and does not constitute Agency policy.
June 2015                                       6-63                    DRAFT—DO NOT CITE OR QUOTE

-------
Hydraulic Fracturing Drinking Water Assessment
Michie. TW: Koch. CA. (1991). Evaluation of injection-well risk management in the Williston Basin. J Pet Tech
   43: 737-741. http://dx.doi.org/10.2118/20693-PA

Molofsky. LI: Connor. IA: Wylie. AS: Wagner. T: Farhat. SK. (2013). Evaluation of methane sources in
   groundwater in northeastern Pennsylvania. Ground Water 51: 333-349.
   http://dx.doi.org/10.llll/gwat.12056

Muehlenbachs. L: Spiller. E: Timmins. C. (2012). Shale gas development and property values: Differences
   across drinking water sources. (NBER Working Paper No. 18390). Cambridge, MA: National Bureau of
   Economic Research, http://www.nber.org/papers/wl8390

Mukherjee, H: Poe jr., B: Heidt, I: Watson, T: Barree, R. (2000). Effect of pressure depletion on fracture-
   geometry evolution and production performance. SPE Prod Facil 15:144-150.
   http://dx.doi.org/10.2118/65064-PA

Myers. T. (2012a). Author's reply. Ground Water 50: 828-830. http://dx.doi.Org/10.llll/j.1745-
   6584.2012.00991.x

Myers, T. (2012b). Potential contaminant pathways from hydraulically fractured shale to aquifers. Ground
   Water 50: 872-882. http://dx.doi.Org/10.llll/i.1745-6584.2012.00933.x

Myers. T. (2013). Author's reply for comments on potential contaminant pathways from hydraulically
   fractured shale to aquifers' [Comment]. Ground Water 51: 319321.
   http://dx.doi.org/10.llll/gwat.12016

NETL (National Energy Technology Laboratory). (2013). Modern shale gas development in the United States:
   An update. Pittsburgh, PA: U.S. Department of Energy. National Energy Technology Laboratory.
   http://www.netl.doe.gov/File%20Library/Research/Oil-Gas/shale-gas-primer-update-2013.pdf

NPC (National Petroleum Council). (2011b). Plugging and abandonment of oil and gas wells. (Paper #2-25).
   Washington, DC: National Petroleum Council (NPC).

ODNR. DMRM.  (Ohio Department of Natural Resources, Division of Mineral Resources Management). (2008).
   Report on the investigation of the natural gas invasion of aquifers in Bainbridge Township of Geauga
   County, Ohio. Columbus, OH: ODNR.
   http://oilandgas.ohiodnr.gov/portals/oilgas/pdf/bainbridge/report.pdf

Oil and Gas Mineral Services. (2010). MineralWise: Oil and gas terminology. Available online at
   http://www.mineralweb.com/library/oil-and-gas-terms/

Olawoyin.  R: Wang. IY: Oyewole. SA. (2013). Environmental safety assessment of drilling operations in the
   Marcellus-shale gas development. SPE Drilling & Completion 28: 212-220.
   http://dx.doi.org/10.2118/163095-PA

Osborn. SG: Vengosh. A: Warner. NR: lackson. RB. (2011). Methane contamination of drinking water
   accompanying gas-well drilling and hydraulic fracturing. PNAS 108: 8172-8176.
   http://dx.doi.org/10.1073/pnas.1100682108

PA PEP (Pennsylvania Department of Environmental Protection). (2000). Pennsylvanias plan for addressing
   problem abandoned wells and orphaned wells. Harrisburg, PA: PADEP.

PA PEP (Pennsylvania Department of Environmental Protection). (2009b). Stray natural gas migration
   associated with oil and gas wells [draft report]. Harrisburg, PA.
   http://www.dep.state.pa.us/dep/subject/advcoun/oil gas/2009/Stray%20Gas%20Migration%20Cases.p
   df
Palmer. ID: Moschovidis. ZA: Cameron. IR.  (2005). Coal failure and consequences for coalbed methane wells.
   Paper presented at SPE annual technical conference and exhibition, October 9-12, 2005, Dallas, TX.
               This document is a draft for review purposes only and does not constitute Agency policy.

June 2015                                      6-64                   DRAFT—DO NOT CITE OR QUOTE

-------
Hydraulic Fracturing Drinking Water Assessment
Peterman. ZE: Thamke. I: Futa. K: Oliver. T. (2012). Strontium isotope evolution of produced water in the East
   Poplar Oil Field, Montana. Presentation presented at US Geological Survey AAPG annual convention and
   exhibition, April 23, 2012, Long Beach, California.

Finder. GF: Celia. MA. (2006). Subsurface hydrology. Hoboken, NJ: John Wiley & Sons, Inc.
   http://dx.doi.org/10.1002/0470044209

Finder. GF: Gray. WG. (2008). Essentials of multiphase flow and transport in porous media. Hoboken, NJ: John
   Wiley & Sons.

Reagan. MT:  Moridis. Gl: lohnson. IN: Keen. ND. (2015). Numerical simulation of the environmental impact of
   hydraulic fracturing of tight/shale gas reservoirs on near-surface groundwater: background, base cases,
   shallow reservoirs, short-term gas and water transport. Water Resour Res 51:1-31.
   http://dx.doi.org/10.1002/2014WR016086

Renpu. W. (2011). Advanced well completion engineering (Third ed.). Houston, TX: Gulf Professional
   Publishing.

Revesz, KM:  Breen, Kl: Baldassare, Al: Burruss, RC. (2012). Carbon and hydrogen isotopic evidence for the
   origin of combustible gases in water-supply wells in north-central Pennsylvania. Appl Geochem 27: 361-
   375. http://dx.doi.0rg/10.1016/j.apgeochem.2011.12.002

Robertson. 10: Chilingar. GV: Khilyuk. LF: Endres. B. (2012). Migration of gas from oil/gas fields. Energ Source
   Part A 34: 1436-1447. http://dx.doi.org/10.1080/15567030903077899

Ross. D: King. G. (2007). Well completions.  In MJ Economides; T Martin (Eds.), Modern fracturing: Enhancing
   natural gas production (1  ed., pp. 169-198). Houston, Texas: ET Publishing.

Rowe, D: Muehlenbachs, K.  (1999). Isotopic fingerprints of shallow gases in the Western Canadian
   sedimentary basin: tools for remediation of leaking heavy oil wells. Organic Geochemistry 30: 861-871.
   http://dx.doi.org/10.1016/S0146-6380(99)00068-6

Roychaudhuri. B: Tsotsis. TT: lessen. K  (2011). An experimental and numerical investigation of spontaneous
   imbibition in gas shales. Paper presented at SPE Annual Technical Conference and Exhibition, October 30 -
   November 2, 2011, Denver, Colorado.

Rutledge. IT: Phillips. WS. (2003). Hydraulic stimulation of natural fractures as  revealed by induced
   microearthquakes, Carthage Cotton Valley gas field, east Texas. Geophysics 68: 441-452.
   http://dx.doi.0rg/10.1190/l.1567214

Rutqvist. I: Rinaldi. AP: Cappa. F:  Moridis. Gl. (2013). Modeling of fault reactivation and induced seismicity
   during hydraulic fracturing of shale-gas reservoirs. Journal of Petroleum Science and Engineering 107: 31-
   44. http://dx.doi.0rg/10.1016/j.petrol.2013.04.023

Rutqvist. I: Rinaldi. AP: Cappa. F:  Moridis. Gl. (2015). Modeling of fault activation and seismicity by injection
   directly into a fault zone associated with hydraulic fracturing of shale-gas reservoirs. Journal of Petroleum
   Science and Engineering 127: 377-386.  http://dx.doi.Org/10.1016/i.petrol.2015.01.019

Sabins. F. (1990). Problems in cementing horizontal wells. J Pet Tech 42: 398-400.
   http://dx.doi.org/10.2118/20005-PA

Saiers. IE: Earth. E. (2012).  Comment on 'Potential contaminant pathways from hydraulically fractured shale
   aquifers'  [Comment]. Ground Water 50: 826-828; discussion 828-830. http://dx.doi.Org/10.llll/j.1745-
   6584.2012.00990.x

Schlumberger (Schlumberger Limited).  (2014). Schlumberger oilfield glossary. Available online at
   http://www.glossary.oilfield.slb.com/

Science Based Solutions LLC.  (2014). Summary of hydrogeology investigations in the Mamm Creek field area,
   Garfield County. Laramie, Wyoming. http://www.garfield-county.com/oil-gas/documents/Summary-
   Hvdrogeologic-Studies-Mamm%20Creek-Area-Feb-10-2014.pdf
               This document is a draft for review purposes only and does not constitute Agency policy.

June 2015                                       6-65                   DRAFT—DO NOT CITE OR QUOTE

-------
Hydraulic Fracturing Drinking Water Assessment
Senior. LA. (2014). A reconnaissance spatial and temporal baseline assessment of methane and inorganic
   constituents in groundwater in bedrock aquifers, pike county, Pennsylvania, 201213 (pp. i-106). (2014-
   5117). Senior, LA. http://pubs.usgs.gov/sir/2014/5117/support/sir2014-5117.pdf

Shapiro. SA: Kriiger. OS: Dinske. C: Langenbruch. C. (2011). Magnitudes of induced earthquakes and
   geometric scales of fluid-stimulated rock volumes. Geophysics 76: WC55-WC63.
   http://dx.doi.0rg/10.1190/geo2010-0349.l

Sharma, S: Bowman, L: Schroeder, K: Hammack, R. (2014a). Assessing changes in gas migration pathways at a
   hydraulic fracturing site: Example from Greene County, Pennsylvania, USA. Appl Geochem.
   http://dx.doi.0rg/10.1016/j.apgeochem.2014.07.018

Sharma. S: Mulder. ML: Sack. A: Schroeder. K: Hammack. R. (2014b). Isotope approach to assess hydrologic
   connections during Marcellus Shale drilling. Ground Water 52: 424433.
   http://dx.doi.org/10.llll/gwat.12083

Siegel. PI: Azzolina. NA: Smith. Bl: Perry. AE: Bothun. RL. (In Press) Methane concentrations in water wells
   unrelated to proximity to existing oil and gas wells in northeastern Pennsylvania. Environ Sci Technol.
   http://dx.doi.org/10.1021/es505775c

Skjerven. T: Lunde.  0: Perander. M: Williams. B: Farquhar. R: Sinet. I: Sasby. I: Haga. HB: Finnseth. 0: lohnsen.
   Si (2011). Norwegian Oil and Gas Association recommended guidelines for well integrity. (117, Revision
   4). Norway: Norwegian Oil and Gas Association.
   http://www.norskoljeoggass.no/Global/Retningslinjer/Boring/117%20-
   %20Recommended%20guidelines%20Well%20integrity%20rev4%2006.06.%2011.pdf

Skoumal. Rl: Brudzinski. MR: Currie. BS. (2015). Earthquakes induced by hydraulic fracturing in Poland
   Township, Ohio. Seismological Society of America Bulletin 105: 189-197.
   http://dx.doi.org/10.1785/0120140168

Smolen. II. (2006). Cased hole and production log evaluation. Tulsa, OK: PennWell Books.

Syed. T: Cutler. T. (2010). Well integrity technical and regulatory considerations for C02 injection wells. In
   2010 SPE international conference on health, safety & environment in oil and gas exploration and
   production. Richardson, TX: Society of Petroleum Engineers. http://dx.doi.org/10.2118/125839-MS

The Royal Society and the Royal Academy of Engineering. (2012). Shale gas extraction in the UK:  A review of
   hydraulic fracturing. London. http://www.raeng.org.uk/news/publications/list/reports/Shale  Gas.pdf

Tilley. Bl: Muehlenbachs. K (2012). Fingerprinting of gas contaminating groundwater and soil in a
   petroliferous region, Alberta, Canada. In RD Morrison; G O'Sullivan (Eds.), Environmental forensics:
   Proceedings of the 211INEF Conference (pp. 115-125). London:  RSC Publishing.
   http://dx.doi.org/10.1039/9781849734967-00115

TIPRO (Texas Independent Producers and Royalty Owners Association). (2012). Bradenhead pressure
   management. Austin, TX. http://www.tipro.org/UserFiles/BHP Guidance Final 071812.pdf

U.S. EPA (U.S. Environmental Protection Agency). (2004). Evaluation of impacts to underground sources of
   drinking water by hydraulic fracturing of coalbed methane reservoirs. (EPA/816/R-04/003). Washington,
   DC.: U.S. Environmental Protection Agency, Office of Solid Waste.

U.S. EPA (U.S. Environmental Protection Agency). (2012c). Geologic sequestration of carbon dioxide:
   underground injection control (UIC) program class VI well construction guidance [EPA Report]. (EPA 816-
   R-ll-020). Washington, D.C.
   http://water.epa.gov/type/groundwater/uic/class6/upload/epa816rll020.pdf

U.S. EPA (U.S. Environmental Protection Agency). (20141). Retrospective case study in northeastern
   Pennsylvania: study of the potential impacts of hydraulic fracturing on drinking water resources [EPA
   Report]. (EPA 600/R-14/088). Washington, D.C.
               This document is a draft for review purposes only and does not constitute Agency policy.

June 2015                                       6-66                    DRAFT—DO NOT CITE OR QUOTE

-------
Hydraulic Fracturing Drinking Water Assessment
U.S. EPA (U.S. Environmental Protection Agency). (2015J). Retrospective case study in Killdeer, North Dakota:
   study of the potential impacts of hydraulic fracturing on drinking water resources [EPA Report]. (EPA
   600/R-14/103). Washington, D.C.

U.S. EPA (U.S. Environmental Protection Agency). (20151). Retrospective case study in the Raton Basin,
   Colorado: study of the potential impacts of hydraulic fracturing on drinking water resources [EPA Report].
   (EPA 600/R-14/091). Washington, D.C.

U.S. EPA (U.S. Environmental Protection Agency). (2015n). Review of state and industry spill data:
   characterization of hydraulic fracturing-related spills [EPA Report]. (EPA/601/R-14/001). Washington,
   D.C.: Office of Research and Development, U.S. Environmental Protection Agency.

U.S. EPA (U.S. Environmental Protection Agency). (2015o). Review of well operator files for hydraulically
   fractured oil and gas production wells: Well design and construction [EPA Report]. (EPA/601/R-14/002).
   Washington, D.C.: Office of Research and Development, U.S. Environmental Protection Agency.

Vaidyanathan. G. (2014). Email communications between Gayathri Vaidyanathan and Ken Klewicki regarding
   the New Mexico Oil Conservation Division District 3  Well Communication Data. Available online

Valko. PP. (2009). Assigning value to stimulation in the Barnett Shale: A simultaneous analysis of 7000 plus
   production hystories and well completion records. Paper presented at SPE Hydraulic Fracturing
   Technology Conference, January 19-21,2009, The Woodlands, TX.

Vengosh. A: lackson. RB: Warner. N: Darrah. TH: Kondash. A. (2014). A critical review of the risks to water
   resources from unconventional shale gas development and hydraulic fracturing in the United States.
   Environ Sci Technol 48: 36-52. http://dx.doi.org/10.1021/es405118y

Vidic. RD: Brantley. SL: Vandenbossche. IM: Yoxtheimer. D: Abad. ID. (2013). Impact of shale gas development
   on regional water quality [Review]. Science 340:1235009. http://dx.doi.org/10.1126/science.1235009

Vincent. M. (2011). Restimulation of unconventional reservoirs: when are refracs beneficial? Journal of
   Canadian Petroleum Technology 50: 36-52. http://dx.doi.org/10.2118/136757-PA

Vulgamore. TB: Clawson. TD: Pope. CD: Wolhart. SL: Mayerhofer. Ml: Machovoe. SR: Waltman. CK. (2007).
   Applying hydraulic fracture diagnostics to optimize stimulations in the Woodford Shale.  Richardson, TX:
   Society of Petroleum Engineers. http://dx.doi.org/10.2118/110029-MS

Wang. W: Dahi Taleghani. A. (2014). Cement sheath integrity during hydraulic fracturing; an integrated
   modeling approach. In 2014 SPE hydraulic fracturing technology conference. Richardson, TX: Society of
   Petroleum Engineers. http://dx.doi.org/10.2118/168642-MS

Warner. NR: lackson. RB: Darrah. TH: Osborn. SG: Down. A: Zhao. K:  White. A: Vengosh.  A. (2012).
   Geochemical evidence for possible natural migration of Marcellus Formation brine to shallow aquifers in
   Pennsylvania. PNAS 109:11961-11966. http://dx.doi.org/10.1073/pnas.1121181109

Warpinski, N. (2009). Microseismic monitoring: Inside and out. J Pet Tech 61: 80-85.
   http://dx.doi.org/10.2118/118537-MS

Watson. TL:  Bachu. S. (2009). Evaluation of the potential for gas and C02 leakage along  wellbores. SPE
   Drilling & Completion 24:115-126. http://dx.doi.org/10.2118/106817-PA

Watts. KR. (2006). A Preliminary Evaluation of Vertical Separation between Production Intervals of Coalbed-
   Methane  Wells and Water-Supply Wells in the Raton Basin, Huerfano and Las Animas Counties, Colorado,
   1999-2004.15.

Weng, X: Kresse, 0: Cohen, C: Wu, R: Gu, H. (2011). Modeling of hydraulic fracture network propagation in a
   naturally fractured formation. Paper presented at SPE Hydraulic Fracturing Technology Conference,
   January 24-26,2011, The Woodlands, TX.

Wojtanowicz. AK. (2008). Environmental control of well integrity. In ST Orszulik (Ed.), Environmental
   technology in the oil industry (pp. 53-75). Houten, Netherlands: Springer Netherlands.
               This document is a draft for review purposes only and does not constitute Agency policy.

June 2015                                       6-67                    DRAFT—DO NOT CITE OR QUOTE

-------
Hydraulic Fracturing Drinking Water Assessment
Wright. PR: McMahon. PB: Mueller. DK: Clark. ML. (2012). Groundwater-quality and quality-control data for
   two monitoring wells near Pavillion, Wyoming, April and May 2012. (USGS Data Series 718). Reston,
   Virginia: U.S. Geological Survey. http://pubs.usgs.gov/ds/718/DS718 508.pdf

WYOGCC (Wyoming Oil and Gas Conservation Commission). (2014). Pavillion Field Well Integrity Review.
   Casper, Wyoming.
   http://wogcc.state.wy.us/pavillionworkinggrp/PAVILLION REPORT 1082014 Final Reportpdf

Zhang, L: Anderson, N: Dilmore, R: Soeder, PI: Bromhal, G. (2014a). Leakage detection of Marcellus Shale
   natural gas at an Upper Devonian gas monitoring well: a 3-d numerical modeling approach. Environ Sci
   Technol 48:10795-10803. http://dx.doi.org/10.1021/es501997p

Zoback. MD. (2010). Reservoir geomechanics. Cambridge, UK: Cambridge University Press.
               This document is a draft for review purposes only and does not constitute Agency policy.

June 2015                                       6-68                    DRAFT—DO NOT CITE OR QUOTE

-------
Hydraulic Fracturing Drinking Water Assessment                  Chapter 7 - Flowback and Produced Water
                           Chapter 7
Flowback and  Produced Water
          This document is a draft for review purposes only and does not constitute Agency policy.
June 2015                                           DRAFT—DO NOT CITE OR QUOTE

-------
      Hydraulic Fracturing Drinking Water Assessment                       Chapter 7 - Flowback and Produced Water


      7.  Flowback and Produced Water

      7.1.  Introduction
 1    Water is a byproduct of oil and gas production. After hydraulic fracturing is completed, either in its
 2    entirety or for a specified stage, the operator reduces the injection pressure. Water is allowed to
 3    flow back from the well to prepare for oil or gas production. This return-flow water may contain
 4    chemicals injected as part of the hydraulic fracturing fluid, chemicals characteristic of the
 5    formation,  hydrocarbons, and in-formation reaction and degradation products. Initially this water,
 6    called flowback, is mostly fracturing fluid, but as time goes on, it becomes more similar to the
 7    formation water. For formations containing saline water (brine), the salinity of the water increases
 8    as time passes, marking the  increased contact time with the formation and in some cases the flow of
 9    formation water itself. This later stage water is called produced water, a term which can also refer
10    to flowback and produced water collectively.

11    Flowback and produced water are stored and accumulated at the surface for eventual reuse or
12    disposal. Typical storage facilities include open air impoundments and closed containers such as
13    those shown in Figure 1-1. Produced water is collected and may be taken to disposal wells,
14    recyclers, waste water treatment plants, or in some cases the water may be left in pits to evaporate
15    or infiltrate. Flowback and produced water leaks can occur on the well pad as a result of human
16    error, failure of container integrity, equipment failure, communication between wells, pipeline
17    leaks, and blowouts.: Above-ground piping systems can connect multiple well pads to
18    impoundments, and piping or impoundments may leak. Much produced water is transported by
19    truck, and pad incidents leading to spills of produced water can occur when trucks are filled.
20    On-road accidents are also possible, some of which could release produced water loads to the
21    environment

22    Impacts to  drinking water resources can occur if spilled flowback or produced water enters surface
23    water bodies or aquifers. Environmental transport of chemical constituents depends on the
24    characteristics of the spill, the fluid (e.g., density, as for highly saline water), the chemicals, and the
25    environment Attenuation processes (e.g., dilution, biodegradation of organics) in surface water and
26    aquifers tend to reduce concentrations.

27    We begin this chapter with a review of definitions for flowback and  produced water. We then
28    discuss typical volumes of flowback and produced water on a per-we 11 basis. This information is
29    aggregated to the state and basin level in Chapter 8. The characteristics of hydraulically fractured
30    shale, tight, and coalbed methane (CBM) formations are described. Spatial and temporal trends on
31    composition of produced water are illustrated with examples from the literature and data compiled
32    for this report The processes controlling the chemical composition  of produced water are
33    described in Appendix E. The potential for impacts on drinking water resources of flowback and
34    produced water are described based on reported spill incidents, contaminant transport principles,
35    and field study examples. The chapter concludes with a discussion of uncertainties and knowledge
       For discussion of well communication, see Chapter 6.
                   This document is a draft for review purposes only and does not constitute Agency policy.
      June 2015                                   7-1                   DRAFT—DO NOT CITE OR QUOTE

-------
      Hydraulic Fracturing Drinking Water Assessment                       Chapter 7 - Flowback and Produced Water


 1    gaps, factors that influence the severity of impacts, and a synthesis based on the EPA research
 2    questions.

      7.1.1.  Definitions
 3    Multiple definitions exist for the terms flowback and produced water. These differing definitions
 4    indicate challenges in determining the distinctions between the two terms or indicate that different
 5    usage of the terms routinely occurs among various industry, private, and public groups. However,
 6    the majority of produced water definitions are fundamentally similar. The following definition is
 7    used in this report: water that flows from oil or gas wells.

 8    This definition is similar to the definition used by the American Petroleum Institute (API):
 9    "Produced water is any of the many types of water produced from oil and gas wells" [API. 2010b]:
10    the definition used by the Department of Energy (DOE): "Produced water is water trapped in
11    underground formations that is brought to the surface along with oil or gas" (DOE, 2004), and a
12    similar definition used by the American Water Works Association (AWWA): "Produced water is the
13    combination of flowback and formation water that returns to the surface along with the oil and
14    natural gas" (AWWA, 2013). Produced water can variously refer to formation water, a mixture of
15    spent hydraulic fracturing fluid and formation water or returned hydraulic fracturing fluid. Thus
16    the term produced water is used when a distinction between fracturing fluid and formation water is
17    not necessary.

18    In general, the term flowback refers either to fluids predominantly containing hydraulic fracturing
19    fluid that returns to the surface or to a process used to prepare the well for production. Because
20    formation water can contact and mix with injection fluids, the distinction between returning
21    hydraulic fracturing fluid and formation water is not clear. In the early stages of operation,
22    however, a higher concentration of chemical additives is expected and later, water that is typical of
23    the formation (Stewart. 2013a).  In most cases, a precise distinction between these waters is not
24    determined during operations.

25    Various definitions have been used for the term flowback The American Petroleum Institute
26    defined flowback as "the fracture fluids that return to  the surface after a hydraulic fracture is
27    completed," (API. 2010b) and the American Water Works Association used "fracturing fluids that
28    return to the surface through the wellbore after hydraulic fracturing is complete" (AWWA. 2013).
29    Other definitions include production of hydrocarbons from the well (Barbotetal., 2013: U.S. EPA,
30    2012Q. or a time period (USGS. 2014f: Haluszczaketal.. 2013: Warner etal.. 2013b: Hayes and
31    Severin. 2012a: Hayes. 2009). As mentioned above, flowback can also be defined as a process used
32    to stimulate the well for production by allowing excess liquids and proppant to return to the
33    surface. Because we use existing literature in our review, we do not introduce a preferred definition
34    of flowback, but rather we mention the assumptions used by the author(s) we discuss.

      7.1.    Volume of Hydraulic Fracturing Flowback and Produced Water
3 5    The characteristics and volume of flowback and produced water vary by well, formation, and time.
3 6    This section presents information on flowback and produced water volume over various time
37    scales, and where possible, on a per-well and per-formation basis.

                   This document is a draft for review purposes only and does not constitute Agency policy.
      June 2015                                    7-2                  DRAFT—DO NOT CITE OR QUOTE

-------
      Hydraulic Fracturing Drinking Water Assessment                       Chapter 7 - Flowback and Produced Water


 1    The amount of flowback from a well varies and depends on several types of factors, including:
 2    production, formation, and operational. Production factors include the amount of fluid injected,
 3    type of hydrocarbon produced (gas or liquid), and location within the formation. Formation factors
 4    include the formation pressure, interaction between the formation and injected fluid (capillary
 5    forces) and reactions within the reservoir. Operational factors include loss of mechanical integrity
 6    and subsurface communication between wells (U.S. GAP. 2012: Byrnes. 2011: DOE. 2011a: GWPC
 7    and ALL Consulting. 2009: Reynolds and Kiker. 2003). The latter two factors  might be indicated by
 8    an unexpected increase in water production (Reynolds and Kiker. 2003).

 9    The processes that allow gas and liquids to flow are related to the conditions along the faces of
10    fractures. Byrnes (2011) conceptualized fluid flow across the fracture face as being composed of
11    three phases. The first is characterized by forced imbibition of fluid into the reservoir and occurs
12    during and immediately following fracture stimulation. Second is an unforced imbibition following
13    stimulation where the fluid redistributes within the reservoir rock, due to capillary forces, when the
14    well is shut-in. The last phase consists of flow out of the formation when the well is opened and
15    pressure reduced in the borehole and fractures. The purpose of this phase is  to recover as much of
16    the injected fluid as possible (Byrnes, 2011) in order to reduce high water saturations at the
17    fracture face and eventually allow higher gas flow rates. The length of the last phase and
18    consequently, the amountof water removed depends on  factors such as the amount of injected
19    fluid, the permeability and effective permeability of the reservoir, capillary pressure properties of
20    the reservoir rock, the pressure near the fracture faces, and whether the well is flowing or shut in.1
21    The well can be shut in for varying time periods depending on operator scheduling, surface facility
2 2    construction or hookup, or other reasons.

      7.1.1.  Flowback of Injected Hydraulic Fracturing Fluid
23    Generally, the fluid that initially returns to the surface has been attributed to a mixture of the
24    injected fracturing fluid, its transformation products, and the natural formation water. In some
25    cases, as shown below, the amount of flowback is greater than the amount of injected hydraulic
26    fracturing fluid and the additional water comes from the  formation (Nicotetal.. 2014) or an
27    adjacent formation (Arkadakskiy  and Rostron. 2013a). Several authors used  geochemical  analyses
28    to postulate mixing between formation water and injected fluid in the Marcellus Shale (Engle and
29    Rowan. 2014: BarbotetaL 2013:  Haluszczaketal.. 20131: Rowan et al., 2015). These possible
30    explanations are summarized in a following section (see  Section 7.6.4). Salinity increases in
31    flowback from highly saline formations, so it is not possible to specify precisely the amount of
32    injected fluids that return in the flowback (GWPC and ALL Consulting. 2009). Rather, such
33    estimates relate the amount of produced water measured at a given time after fracturing as a
34    percentage of the total amountof injected fluid.

35    Estimates vary butin composite indicate on average that between 5% and 75% (see Table 7-1,
36    Table 7-2, and Table 7-3) of the volume of injected fracturing fluid may flow back to the  surface
      1 When multiple fluids (water, oil, gas) occupy portions of the pore space, the permeability to each fluid depends on the
      fraction of the pore space occupied by the fluid and the fluid's properties. As defined by Pake (1978). when this effective
      permeability is normalized by the absolute permeability, the resulting relationship is known as the relative permeability.
                   This document is a draft for review purposes only and does not constitute Agency policy.
      June 2015                                     7-3                   DRAFT—DO NOT CITE OR QUOTE

-------
     Hydraulic Fracturing Drinking Water Assessment
Chapter 7 - Flowback and Produced Water
1    after hydraulic fracturing is complete [U.S. EPA. 2015q: Vengoshetal.. 2014: Mantell. 2013b: Vidic
2    etal.. 2013: Minnich. 2011: Xuetal.. 2011). These data (see Table 7-1] illustrate thatthe formations
3    differ in their water requirements for hydraulic fracturing and generation of produced water over
4    the short term.: Low percentages of flowback are typical, as is the decrease of flowback volume
5    with time as the wells enter the production phase [Gregory etal.. 2011: McElreath. 2011: GWPC and
6    ALL Consulting. 2009]. Some formations produce higher volumes, as noted for the Barnett Shale in
7    Texas [Nicotetal., 2014] and discussed below.

     Table 7-1. Data from one company's operations indicating approximate total water use and
               approximate produced water volumes within 10 days after completion of wells
               (Mantell. 2013b).
Formation
Approx. total
average water use
per well (million gal)
Produced water within the first
10 days after completion
Low estimate
(million gal)
High or only
estimate
(million gal)
Produced water as a percentage of
average water use per well
Low estimate
(% of total
water use)
High or only
estimate (% of
total water use)
Gas shale plays (primarily dry gas)
Barnett3
Marcellus3
Haynesville
3.4
4.5
5.4
0.3
0.3
--
1.0
1.0
0.25
9
7
--
29
22
5
Liquid plays (gas, oil, condensate)
Mississippi Lime
Cleveland/Tonkaw
Niobrara
Utica
Granite Wash
Eagle Ford
2.1
2.7
3.7
3.8
4.8
4.9
--
0.3
0.3
0.3
0.3
0.3
1.0
1.0
1.0
1.0
1.0
1.0
--
11
8
8
6
6
48
37
27
26
21
20
     a Mantell (2011) reported produced water for the first 10 days at 500,000 to 600,000 gal for the Barnett, Fayetteville and
     Marcellus Shales.
      Flowback estimates may be based on specific time periods (e.g., the flowback during the first 10,15, or 30 days).
                   This document is a draft for review purposes only and does not constitute Agency policy.
     June 2015                                     7-4                   DRAFT—DO NOT CITE OR QUOTE

-------
Hydraulic Fracturing Drinking Water Assessment
Chapter 7 - Flowback and Produced Water
Table 7-2. Additional short-, medium-, and long-term produced water estimates.
Location-formation
Produced water as
percentage of injected fluid
Reference
Comment
Estimates without reference to a specific data set
Unspecified Shale
Marcellus Shale
ND-Bakken
5% - 35%
10% - 25%
25%
Haves (2011)

Minnich (2011)

EERC (2013)


Initial flowback

Estimates with reference to specific data evaluation
Short duration
Marcellus Shale
TX— Barnett
TX— Haynesville
AR— Fayetteville
10%
20%
5%
10%
Clark etal. (2013)

Clark etal. (2013)

Clark etal. (2013)

Clark etal. (2013)

0-10 days
0-10 days
0-10 days
0-10 days
Mid duration
WV-Marcellus
Marcellus Shale
8%
24%
Hansen etal. (2013)

Hayes (2011, 2009)
30 days
Average from 19 wells, 90 days
Long duration
TX— Barnett
WV-Marcellus
TX-Eagle Ford
~100%a
10% - 30%
<20%
Nicot etal. (2014)

Ziemkiewicz et al. (2014)

Nicot and Scanlon (2012)

72 months
Up to 115 months
Lifetime
Unspecified duration
PA— Marcellus
6%
Hansen etal. (2013)

 a Approximate median with large variability: 5*  percentile of 20% and 90*  percentile of 350%.
               This document is a draft for review purposes only and does not constitute Agency policy.
June 2015                                         7-5                     DRAFT—DO NOT CITE OR QUOTE

-------
     Hydraulic Fracturing Drinking Water Assessment
Chapter 7 - Flowback and Produced Water
     Table 7-3. Flowback and long-term produced water characteristics for wells in unconventional
              formations (U.S. EPA, 2015e).
              Source: (U.S. EPA, 2015q).
Resource
type
Shale
Tight
Drill type
Horizontal
Directional
Vertical
Horizontal
Directional
Vertical
Fracturing flu id
(million gal)
Median
4.0
1.6
1.2
2.2
0.60
0.31
Range
0.13-15
0.051-12
0.015-22
0.042-9.4
0.056-4.0
0.019-4.0
Number
of data
points
50,053
124
4,152
765
693
1,287
Long-term produced water (gal/day per well)
Shale
Tight
Horizontal
Directional
Vertical
Horizontal
Directional
Vertical
900
480
380
620
750
570
0-19,000
22-8,700
0-4,600
0-120,000
12-1,800
0-4,000
22,222
695
12,393
2,394
3,816
21,393
Flowback
(percent of fracturing fluid returned)
Median
6%
14%
24%
7%
6%
8%
Range
1% - 50%
4% -31%
7% - 75%
7% - 60%
0% - 60%
1% - 83%
Number of
data points
6,488
19
18
39
263
48

1    In the following subsections, we first discuss water produced during the flowback period, then
2    longer-term produced water.

     7.1.1.1. Produced Water during the Flowback Period
3    Data were collected from six vertical and eight horizontal wells in the Marcellus Shale of
4    Pennsylvania and West Virginia [Hayes. 2009]. The author collected samples of flowback after one,
                  This document is a draft for review purposes only and does not constitute Agency policy.
     June 2015                                     7-6                   DRAFT—DO NOT CITE OR QUOTE

-------
     Hydraulic Fracturing Drinking Water Assessment                        Chapter 7 - Flowback and Produced Water


1    five, and 14 days after hydraulic fracturing was completed, as well as a produced water sample 90
2    days after completion of the wells. Both the vertical and horizontal wells showed their largest
3    volume of flowback between one and five days after fracturing, as shown in Figure 7-1. The wells
4    continued to produce water, and at 90 days, samples were available from four each of the
5    horizontal and vertical wells. The vertical wells produced on average 180 bbl/day (7,600 gal/day or
6    29,000 L/day) and the horizontal wells a similar 200 bbl/day (8,400 gal/day or 32,000 L/day).
7    Results from one Marcellus Shale study were fitted to a power curve, as shown in Figure 7-2
8    (Ziemkiewicz etal.. 2014). These and the Hayes (2009) data show decreasing flowback with time.
                  This document is a draft for review purposes only and does not constitute Agency policy.
     June 2015                                     7-7                   DRAFT—DO NOT CITE OR QUOTE

-------
Hydraulic Fracturing Drinking Water Assessment
                        Chapter 7 - Flowback and Produced Water
                                     MSC Study FB volumes {Vertical Wells)
                                                                                   -Avert teal
                                                                                   -B Vertical
                                                                                   - H Vertical
                                                                                   -N Vertical
                                                                                   -QVertical
                                                                                   -SVertial
                                   20
 40        60
  Time (days)
 80
 100
                                    MSC Study FB volumes (Horizontal Wells)
                                                                                 -C Horizontal
                                                                                 -D Horizontal
                                                                                 -E Horizontal
                                                                                 — F Horizontal
                                                                                 -GHorizontal
                                                                                 -KHorizontal
                                                                                 -M Horizontal
                                                                                 -OHonzontal
                                  20
40        60
 Time (days)
80
100
Figure 7-1. Fraction of injected hydraulic fracturing fluid recovered from six vertical (top) and
            eight horizontal (bottom) wells completed in the Marcellus Shale.
            Data from Hayes (2009).
                This document is a draft for review purposes only and does not constitute Agency policy.
June 2015                                          7-8                      DRAFT—DO NOT CITE OR QUOTE

-------
     Hydraulic Fracturing Drinking Water Assessment
Chapter 7 - Flowback and Produced Water
                       60
                       40
                                V = 0.082xa268B
                                 R2 = 0.9882
                                                                            35%
                                                                            30%
                                                                            75%
                                                                            20%
                                                                            15%
                                                                            10%
                          ou^oiAo^ou^ou>au"iOu"iOu-tOu"iOu-iOiriau"i

                                         Month after initial frac
                           a flowback/produced water        o % flowback

     Figure 7-2. Example of flowback and produced water from the Marcellus Shale, illustrating
               rapid decline in water production and cumulative return of approximately 30% of
               the volume of injected fluid.
               Source: Ziemkiewicz et al. (2014). Reprinted with permission from Ziemkiewicz, P; Quaranta, JD;
               Mccawley, M. (2014). Practical measures for reducing the risk of environmental contamination in
               shale energy production. Environmental Science: Processes & Impacts 16:1692-1699. Reproduced by
               permission of The Royal Society of Chemistry. http://dx.doil.org/10.1039/C3EM00510K.

1    In West Virginia, water recovered at the surface within 30 days following injection or before 50% of
2    the injected fluid volume is returned to the surface is reported as flowback. Data from 271 wells in
3    the Marcellus Shale in West Virginia [Hansenetal.. 2013] reveals the variability of recovery from
4    wells in the same formation and that the amount of injected fluid recovered was less than 15%
5    from over 80% of the wells (see Figure 7-3].
                   This document is a draft for review purposes only and does not constitute Agency policy.
     June 2015                                      7-9                   DRAFT—DO NOT CITE OR QUOTE

-------
      Hydraulic Fracturing Drinking Water Assessment
Chapter 7 - Flowback and Produced Water
                                                 Percent recovered
      Figure 7-3. Percent of injected fluid recovered for Marcellus Shale wells in West Virginia
                (2010-2012).
                Source: Hansen et al. (2013). One data point showing 98% recovery omitted. Reprinted with
                permission from Hansen, E; Mulvaney, D; Betcher, M. (2013). Water resource reporting and water
                footprint from  Marcellus Shale development in West Virginia and Pennsylvania. Durango, CO:
                Earthworks Oil & Gas Accountability Project. Copyright 2013. Permission Downstream Strategies, San
                Jose State University, and Earthworks Oil & Gas Accountability Project.

 1    The amount of flowback water produced by wells within the first few days of fracturing varies from
 2    formation to formation. Wells in the Mississippi Lime and Permian Basin can produce 10 million gal
 3    (37.8 million L) in the first 10 days of production. Wells in the Barnett, Eagle Ford, Granite Wash,
 4    Cleveland/Tonkawa Sand, Niobrara, Marcellus, and Utica Shales can produce 300,000 to 1 million
 5    gal (1.14 to 3.78 million L) within the first 10 days; while Haynesville wells produce less, about
 6    350,000 gal (1.32 million L) (Mantell. 2013bj.

      7.1.2.   Produced Water
 7    During oil and gas production, other fluids which contain water are produced with hydrocarbons.
 8    Throughout this production phase at oil and certain wet gas production facilities, produced water is
 9    stored in tanks and pits that may contain free phase, dissolved phase, and emulsified crude oil in
10    the produced water.1 This crude oil can be present in the produced water container or pit, because
11    the crude oil is not efficiently separated out by the flow-through process vessels (such as three-
12    phase separators, heater treaters, or gun barrels) and passes through to these containers/pits. The
      1 Dry natural gas occurs in the absence of liquid hydrocarbons; wet natural gas typically contains less than 85% methane
      along with ethane and more complex hydrocarbons [Schlumberger. 2014).
                    This document is a draft for review purposes only and does not constitute Agency policy.
      June 2015                                     7-10                   DRAFT—DO NOT CITE OR QUOTE

-------
      Hydraulic Fracturing Drinking Water Assessment                       Chapter 7 - Flowback and Produced Water


 1    produced water containers and pits containing oil at production facilities are typically regulated
 2    under 40 CFR part 112, produced water containers or pits may also be subject to other applicable
 3    state and or local laws, regulations and/or ordinances.

 4    Lutz etal. [2013] evaluated data reported to the Pennsylvania Department of Environmental
 5    Protection (PA DEP) for the time period January 2000 to December 2011. The data were divided
 6    between conventional gas wells that might have been hydraulically fractured and hydraulically
 7    fractured completions in the Marcellus Shale. The conventional wells produced less drilling water,
 8    less flowback (when fractured), and less brine than the shale wells (Lutz etal., 2013: see Table Ij.
 9    The average amount of produced water per well was 136,000 gal (514,000 L) for the conventional
10    wells and 1.38 million gal (5.211 million L) for the shale wells. The produced water to gas ratio was
11    1.27 gal (4.8 L) water per MMBtu for the shale wells, which was 2.8 times lower than for
12    conventional wells. Both the produced water and gas produced per well decreased over the four-
13    year period covered by the study. In contrast, conventional oil wells tend to have increased volumes
14    of produced water as they age,  and in some cases, older wells may produce five times as much
15    water as new wells (U.S. GAP. 20121.

16    From experience in several shale formations, Mantell (2013b, 2011] characterized the amount of
17    produced water over the long term as high, moderate, or low. Wells in the Barnett Shale,
18    Cleveland/Tonkawa Sand, Mississippi Lime, and the Permian Basin can produce more than
19    1,000 gal (3,800 L) of water per million cubic feet (MMCF) of gas because of formation
20    characteristics. The most productive of these can be as high as 5,000 gal (19,000 L) per MMCF. As a
21    specific example, a high-producing formation in the western United States was described as
22    producing 4,200 gal (16,000 L) per MMCF for the life of the well (McElreath. 20111. The well was
23    fractured and stimulated with about 4 million gal (15 million L) of water and returned
24    60,000 gal (230,000 L) per day in the first 10 days, followed by 8,400 gal (32,000 L) per day in the
2 5    remainder of the first year.

2 6    Similarly, produced water from horizontal wells in the Barnett Shale decreased rapidly after the
27    wells began producing gas (Nicotetal.. 2014] (see Figure 7-4]. The data show a high degree of
28    variability, which was attributed by Nicotetal. (2014] to a few wells with exceptionally high water
29    production. When the produced water data were presented as the percentage of injected fluid, the
30    median exceeded 100% at around 36 months, and the 90th percentile was 350% (see Figure 7-5].
31    This means that roughly 50% of the wells were producing more water than was used in stimulating
32    production. Nicotetal. (2014] noted an inverse relationship between gas and water production but
33    did not identify the source or mechanism for the excess water. Systematic breaching of the
34    underlying karstic  Ellenburger  Formation was not believed likely; nor was operator efficiency or
35    skill. A number of geologic factors that could impact water migration were identified by (DOE,
36    2011a] in the Barnett Shale, including fracture height, aperture size, and density, fracture
37    mineralization, the presence of karst chimneys underlying parts of the Barnett Shale, and others,
38    but the impact of these on water migration was undetermined.
                   This document is a draft for review purposes only and does not constitute Agency policy.
      June 2015                                   7-11                  DRAFT—DO NOT CITE OR QUOTE

-------
Hydraulic Fracturing Drinking Water Assessment
                               Chapter 7 - Flowback and Produced Water
                     20.000
                   S. 15.000 -
                   1
                     10.000
                      5000
                                                                          12.000
                                                                          10,000
                                                                         - 8000   ra
                                                                          6000
                                                                         - 4000
                                                                         - 2000
                                                                                o
                                 12
                                               36
                                                      48
                                                             60
                                                                    72
                                             Months

Figure 7-4. Barnett Shale monthly water-production percentiles (5th, 30th, 50th, 70th, and 90th)
           and number of wells with data (dashed line).

           Source: Nicot et al. (2014). FP is the amount of water the flows back to the surface, commingled with
           water from the formation. Reprinted with permission from Nicot, JP; Scanlon, BR; Reedy, RC; Costley,
           RA. (2014). Source and fate of hydraulic fracturing water in the Barnett Shale: a historical perspective.
           Environ Sci Technol 48: 2464-2471. Copyright 2014 American Chemical Society.
                        200
|g150 H
8: s>
*— CO
                        100 -
II
Q
£
                         50 -
                                 - 12,000

                                 - 10,000

                                 - 8000

                                 - 6000

                                 - 4000

                                 - 2000

                                   0
                                                                              "CD

                                                                              "CD
                                                                              4—1
                                                                               O
                                                                               N
                                                            CD
                                                            .a
                                                            E
                                                            Z!
                                                            •2.
                              0
12   24   36    48   60   72
                                             Months

Figure 7-5. Barnett Shale production data for approximately 72 months.

           Source: Nicot et al. (2014). Flowback and produced water are reported as the percentage of injected
           fluid. The dashed line shows the number of horizontal wells included. Data for each percentile show
           declining production with time, but the median production exceeds 100% of the injected fluid. FP is
           the amount of water the flows back to the surface, commingled with water from the formation.
           Reprinted with permission from Nicot, JP; Scanlon, BR; Reedy, RC; Costley,  RA. (2014). Source and fate
           of hydraulic fracturing water in the Barnett Shale: a historical perspective. Environ Sci Technol 48:
           2464-2471. Copyright 2014 American Chemical Society.
               This document is a draft for review purposes only and does not constitute Agency policy.

June 2015                                        7-12                    DRAFT—DO NOT CITE OR QUOTE

-------
      Hydraulic Fracturing Drinking Water Assessment                       Chapter 7 - Flowback and Produced Water


 1    The Niobrara, Granite Wash, Eagle Ford, Haynesville, and Fayetteville Shales are relatively dry and
 2    produce between 200 and 2,000 gal (760 to 7,600 L) of produced water per MMCF (Mantell, 2013).
 3    The Utica and Marcellus Shales are viewed as drier and produce less than 200 gal (760 L)
 4    per MMCF. DOE (2011a] concluded that the characteristic small amount of produced water from
 5    the Marcellus Shale was due either to its low water saturation or low relative permeability to water
 6    (see Chapter 6). For dry formations, low shale permeability and high capillarity cause water to
 7    imbibe into the formation, where it is retained permanently (He, 2011]. Engelder (2012] estimated
 8    that more than half of the fracturing fluid could be captured within the Marcellus if imbibition
 9    drove fluid just 5 cm (2 in] deep into reservoir rocks across the fracture surfaces. This estimate is in
10    agreement with the generalized analysis presented by Byrnes (2011], who estimated depths of 5 to
11    15 cm (2 to 6 in].

12    After fracture of coalbeds, water is withdrawn to liberate gas. CBM tends to produce large volumes
13    of water early on: more in fact, than conventional gas-bearing formations (U.S. GAP, 2012].  Within
14    producing formations, water production can vary for unknown reasons (U.S. GAP, 2012]. Data show
15    that CBM production in the Powder River Basin produces 16 times more water than in the San Juan
16    Basin (U.S. GAP. 2012].

17    The EPA (2015q] reported characteristics of long-term produced water for shale and tight
18    formations (see Table 7-3]. For shale, horizontal wells produced more water (900 gal/day] than
19    vertical wells (380 gal/day]. Typically, this would be attributed to the longer length of horizontal
20    laterals than vertical wells, but the data were not normalized to these lengths. The formation-level
21    data used to develop Table 7-3 appear in Table E-l of Appendix E.

22    The EPA (2015q] reported that a general rule of thumb is that flowback occurring in the first 30
23    days of production is roughly equal to the long-term produced water for unconventional
24    formations. As a specific example, from Pennsylvania Marcellus Shale data, the EPA determined that
25    for vertical wells in unconventional formations, 6% of water came from drilling, 35% from
26    flowback, and 59% from long-term produced water; and for horizontal wells the corresponding
27    numbers were 9%, 33%, and 58%. These values deviate from the rule of thumb, because the
28    Marcellus Shale was believed to generate low levels of flowback relative to other formations (U.S.
29    EPA. 2015q].

      7.2.   Flowback and Produced Water Data Sources
30    Unlike the evaluation of hydraulic fracturing fluid itself where the chemical composition may be
31    disclosed, knowledge concerning flowback and produced water composition comes from
32    measurements made on samples.

33    A number of factors are involved in the proper sampling and analyzing of environmental media
34    (U.S. EPA. 2013e: ATSDR. 2005: U.S. EPA. 1992]. There may be significant issues obtaining samples,
35    because the specialized equipment used to contain high-pressure natural gas is not designed for
36    producing environmental samples (Coleman, 2011].
                   This document is a draft for review purposes only and does not constitute Agency policy.
      June 2015                                   7-13                  DRAFT—DO NOT CITE OR QUOTE

-------
      Hydraulic Fracturing Drinking Water Assessment                        Chapter 7 - Flowback and Produced Water


 1    To choose the correct analytical methods, it is necessary to have information on:

 2         •   Physical state of the sample.
 3         •   Identification of analytes of interest.
 4         •   Required sensitivity and quantitation limits.
 5         •   Analytical objective (i.e., unknown identification, monitoring).
 6         •   Required sample containers, preservation, and holding times.
 7    Because some components of hydraulic fracturing fluid are proprietary chemicals, and subsurface
 8    reaction products may be unknown, prior knowledge of the identity of analytes may not be
 9    available. Consequently, studies may be limited in their ability to determine the presence of either
10    unknown or proprietary constituents  contained in flowback or produced water simply because of
11    the lack of knowledge of the identities of the constituents.

12    After laboratory analysis, the results are evaluated according to quality criteria. Data may be judged
13    to meet applicable quality criteria as determined by the analytical methods or they may be
14    "flagged." Typically, encountered flags are non-detect, below reporting limit or diluted to meet
15    calibration requirements or because of matrix interference [e.g., Hayes, 2009].1 For produced
16    water, a primary interference is from high total dissolved solids (TDS). Interferences also arise from
17    agents which cause foaming and alter  surface tension [Coleman. 2011). Diluted samples result in
18    higher detection limits, and thus lessen ability to identify lower concentrations in samples.

19    Because of identified limitations in existing methods, the EPA developed new methods for some
20    reported components of hydraulic fracturing fluids, including ethanols and glycols [U.S. EPA.
21    2014k], certain nonionic surfactants [DeArmond and DiGoregorio, 2013a], and acrylamide
22    [DeArmond and DiGoregorio. 2013b].2'3 Each of these methods are applicable to ground and surface
23    waters, and the last [DeArmond and DiGoregorio. 2013b] to waters with TDS well above 20,000
24    mg/L.

25    Generally, analytical methods are impacted by elevated TDS and chloride concentrations, especially
26    inorganic and wet chemistry methods [Nelson etal.. 2014: U.S. EPA. 2014b: Coleman. 2011). Matrix
27    interference impacts standard analysis (EPA Method 8015) for glycols, resulting in high detection
28    limits (10,000 [ig/L to 50,000 [ig/L) (Coleman. 2011).

29    Produced water levels of naturally occurring radionuclides may be 1,000 to 10,000 times the levels
30    of activity found in typical environmental water samples (U.S. EPA. 2014b). The standard EPA
31    method (Method 900.0) for gross alpha and gross beta involves evaporation of the sample to a layer
32    of residue and analyzing emitted alpha and beta particles. The method has several noted
      1 Matrix interference occurs when components of the sample other than the analyte of interest have an effect on a
      measurement [IUPAC. 2014).
      2 The compounds included were: Diethylene Glycol, Triethylene Glycol, Tetraethylene Glycol, 2-Butoxyethanol and 2-
      Methoxyethanol.
      3 The compounds included were: C12-C16 and C18 alcohol ethoxylates, and alkylphenol ethoxylates.
                    This document is a draft for review purposes only and does not constitute Agency policy.
      June 2015                                     7-14                  DRAFT—DO NOT CITE OR QUOTE

-------
      Hydraulic Fracturing Drinking Water Assessment                       Chapter 7 - Flowback and Produced Water


 1    limitations, including known under representation of radium 228, and applicability to drinking
 2    water samples with low levels (<500 mg/L) of TDS. As discussed below, produced water can have
 3    much higher TDS levels. Because of these limitations, the EPA [2014b] developed an updated
 4    method for the detection of gross alpha and gross beta to reduce the matrix interferences, although
 5    further improvement is possible.1

 6    Due to the high ionic strength and dissolved solids concentration of flowback water from shale
 7    operations, Nelson et al. [2014] similarly found that traditional wet chemistry techniques (EPA
 8    Methods 903.0 and 904.0) inefficiently recover radium from samples, with radium-226 recovery
 9    sometimes less than 1% . This concern, which could lead to false negatives, was previously noted by
10    Demorest and Wallace [1992].  Nelson and coauthors demonstrated that an accurate assessment of
11    flowback radium levels can be performed through nondestructive high-purity germanium gamma
12    spectroscopy and emanation techniques. Studies reporting radium concentrations obtained directly
13    via wet chemistry techniques or studies reporting third-party radium data via wet chemistry
14    techniques may need to be evaluated appropriately as these techniques may underestimate the
15    total radium loads of produced water [Nelson etal., 2014].

16    Data have been generated from specific produced water studies [e.g.. Hayes. 2009] or compilations
17    from various sources, such as the USGS produced water database developed in 2002 [Breit, 2002]
18    and updated in 2014 [Blondes etal., 2014]. In this database, data were compiled from a variety of
19    sources, some of which we cite as examples below. The data that appear in this chapter and
20    Appendix E are drawn individually from scientific literature and published reports, where
21    necessary we have filled gaps with data from the USGS database.

      7.3.   Background on Formation Characteristics
2 2    Subsurface processes and  resulting flowback and produced water composition vary depending
23    upon the  mineralogy, geochemistry, and structure of formation solids, as well as, residence time
24    and other factors [Dahm etal.. 2011: Blauch etal.. 2009]. The mineralogy and structure of
25    formation solids are determined initially by deposition, when rock grains settle out of their
26    transporting medium [Marshak. 2004]. Generally, shale results from clays deposited in deep,
27    oxygen-poor marine environments, and sandstone results from sand deposited in shallow marine
28    environments [Alietal.. 2010: U.S. EPA. 2004]. Coal forms when carbon-rich plant matter collects in
29    shallow peat swamps. In the United States,  coal is formed in both freshwater and marine
30    environments [NRG, 2010]. In the northern Rocky Mountains, coal formed within freshwater
31    alluvial systems of streams, lakes, and peat swamps. In contrast are parts of the Black Warrior
32    formation, which were deposited in brackish and marine settings [Horsey, 1981].

33    Variation in produced water composition follows, in part, from differences in formations which are
34    related to geologic processes. After deposition, physical, chemical, and biological processes occur as
      1 The method developed for determining gross alpha (Th, U, and Po) by liquid scintillation is based on: manganese dioxide
      coprecipitation followed by group separation of thorium, uranium and polonium on TRU Resin, stripping with ammonium
      bioxalate, and pulse-shape discrimination liquid scintillation analysis. The average recovery was 74±11% of the known
      concentration of 230Th with recoveries which ranged from 57% to 104%.
                   This document is a draft for review purposes only and does not constitute Agency policy.
      June 2015                                    7-15                  DRAFT—DO NOT CITE OR QUOTE

-------
      Hydraulic Fracturing Drinking Water Assessment                       Chapter 7 - Flowback and Produced Water


 1    sediments and are consolidated and cemented into rocks in a process called diagenesis. These
 2    processes, which can also occur in existing sedimentary rocks, are caused by increased pressure,
 3    temperature, and reaction with mineral-rich ground water [Rushing etal.. 2013: Marshak. 2004).
 4    Diagenesis may either decrease or increase porosity and permeability through sediment
 5    compaction and mineral precipitation, or through grain and cement dissolution [Alietal., 2010:
 6    Schmidt and McDonald, 1979]. Temperature and pressure greatly affect the types and extent of
 7    subsurface reactions, influencing the solubility of formation solids, saturation of pore waters, and
 8    prevalence of precipitates [Rushing etal.. 2013).

      7.4.   Flowback Composition
 9    The composition of returning hydraulic fracturing fluid changes with increasing residence time.  In
10    this section, we present several examples from individual wells which demonstrate how
11    concentrations approach apparently asymptotic values during the first few days or weeks after
12    hydraulic fracturing.

      7.4.1.  General Characteristics
13    Several interacting factors that influence the composition of hydraulic fracturing flowback and
14    produced water are recognized in the scientific literature: (1] the composition of injected hydraulic
15    fracturing fluids, (2) the targeted geological formation and associated hydrocarbon products,
16    (3) the stratigraphic environment, and (4) subsurface processes and residence time [Barbotetal..
17    2013: Chapman  etal.. 2012: Dahm etal.. 2011: Blauch etal.. 20091.

18    By design, hydraulic fracturing exposes fresh, organic- and mineral-rich surfaces. Subsurface
19    interactions between injected hydraulic fracturing fluids, formation solids, and formation waters
20    follow. As residence time increases, allowing in situ interactions between injected fluids, formation
21    fluids, and formation solids, changes in the geochemical content of flowback occur such that it still
22    largely reflects that of injected  fluids, while later flowback and produced water reflect that of
23    formation-associated fluid [Rowan etal.. 2011).

      7.4.2.  Temporal Changes in Flowback Composition
24    Ionic loads, metals, naturally occurring radioactive material (NORM), and organics increase in
25    concentration as water production continues [Barbotetal., 2013: Murali Mohan et al., 2 013: Rowan
26    etal.. 2011). The causes include precipitation and dissolution of salts, carbonates, sulfates, and
27    silicates; pyrite oxidation; leaching and biotransformation of organic compounds; and mobilization
28    of NORM and trace elements. Multiple geochemical studies confirm this trend [Barbotetal., 2013:
29    Haluszczaketal.. 2013: Chapman etal.. 2012: Davis etal.. 2012: Gregory etal.. 2011: Blauch etal..
30    20091.

31    Concurrent precipitation of sulfates (e.g., BaS04) and carbonates (e.g., CaCOs] alongside decreases
32    in pH, alkalinity, and dissolved carbon load occur over time (Orem etal., 2014: Barbotetal., 2013:
33    Blauch etal.. 2009: Brinck and  Frost. 2007). Orem and colleagues showed that organics within CBM
34    produced waters also decrease over time, possibly due to the exhausting of coal-associated waters
35    through formation pumping (Orem etal., 2007]. Decreases in microbial abundance and diversity
36    also occur over time after hydraulic fracturing (Murali Mohan et al., 2 013: Davis etal., 2012].

                   This document is a draft for review purposes only and does not constitute Agency policy.
      June 2015                                    7-16                  DRAFT—DO NOT CITE OR QUOTE

-------
      Hydraulic Fracturing Drinking Water Assessment                       Chapter 7 - Flowback and Produced Water


 1    The primary dissolution of native and emplaced salts within the formation and the mobilization of
 2    in situ brines constitute the major subsurface processes that control TDS levels in flowback and
 3    produced water [Dresel and Rose. 2010: Blauch etal.. 2009).* Leaching of organics appears to be a
 4    result of injected and formation fluids associating with shale and coal strata [Orem etal., 2014].

      7.4.3.  Total Dissolved Solids  Enrichment
 5    To varying degrees, produced water is enriched in dissolved solids, and the enrichment is
 6    dependent upon residence time [Rowan etal., 2011]. As an example, TDS concentrations increased
 7    until a limit was reached in flowback and produced water samples from four Marcellus Shale gas
 8    wells in three southwestern Pennsylvanian counties [Chapman etal., 2012] (see Figure 7-6]. As is
 9    shown in Figure 7-7, TDS in flowback from both Westmoreland County wells became consistent
10    with TDS concentrations cited for typical seawater (35,000 mg/L] within three days, and became
11    consistent with TDS cited for brines (greater than 50,000 mg/L] within five days (Chapman etal.,
12    2012]. TDS concentrations during production exceeded 188,000 mg/L for one well in Greene
13    County. Chapman et al.'s findings are further substantiated by Hayes and colleagues' earlier report
14    of produced water TDS concentrations in 19 Marcellus Shale wells in Pennsylvania and West
15    Virginia (Hayes, 2009]. From an initial injected median value of less than 1,000 mg/L, TDS
16    concentrations increased to a median value exceeding 200,000 mg/L within 90 days (Hayes, 2009].
17    In the Marcellus Shale, the cation portion of TDS is typically dominated by sodium and calcium,
18    whereas the anion portion is dominated by chloride (Chapman etal.. 2012: Blauch etal.. 2009]. In
19    section 7.6.4, we note that there is disagreement over whether increased salinity in Marcellus Shale
20    produced water is due to dissolution of salts or mixing of formation water with hydraulic fracturing
21    fluid.
      1 Native salts are formed inside the rock matrix, and can include evaporite minerals such as halite (NaCl), polyhalite
      (K2Ca2Mg(S04)4«2H20), celestite (SrS04),anhydrite (CaS04), kieserite (MgS04«H20), or sylvite (KC1) [Blauch etal.. 2009).
      Hydrologic intrusion emplaces salts within formation pores and fractures [Blauch etal.. 2009).
                    This document is a draft for review purposes only and does not constitute Agency policy.
      June 2015                                     7-17                  DRAFT—DO NOT CITE OR QUOTE

-------
      Hydraulic Fracturing Drinking Water Assessment
Chapter 7 - Flowback and Produced Water

200,000
180,000
160,000
, 	 s
< 140,000
M
£
^120,000


-
t A A A A
• • A
•• A
^h
-







            ISI
            3
            ^ 100,000
            IjO

            1  80,000
5
"05
4— <
Q




60,000
40,000

20,000
: .•
•

f
n , , , , i ,
0 5
• Greene Co.
Westmoreland Co. 1
Westmoreland Co. 2
A Washington Co.
ii i i i i i i i i i i i i i i i
10 15 20 25 30
                                              Days Post-Fracturing
      Figure 7-6. TDS concentrations measured through time for injected fluid (at 0 days), flowback,
                and produced water samples from four Marcellus Shale gas wells in three
                southwestern Pennsylvanian counties.
                Data from Chapman etal. (2012).

      7.4.4.  Radionuclide Enrichment
 1    Injected fluids used in hydraulic fracturing typically do not contain radioactive material [Rowan et
 2    al., 2011].1 Shales and sandstones, however, are naturally enriched in various radionuclides, as
 3    described below [Sturchio etal.. 2001]. Radium in pore waters or adsorbed onto clay particles and
 4    grain coatings can dissolve and return within flowback [Langmuir and Riese, 1985]. Where data are
 5    available, radium and TDS produced water concentrations are positively correlated with time
 6    passed since hydraulic fracturing [Rowan etal., 2011: Fisher, 1998]. Radium remains adsorbed to
 7    mineral surfaces in low saline environments, and then desorbs with increased salinity into solution
 8    [Sturchio etal.. 2001]. Over the course of 20 days, Marcellus Shale produced waters from a gas well
 9    were enriched almost fourfold in radium and from another gas well were enriched over twofold in
10    TDS concentrations as residence time increased [Chapman etal., 2012: Rowan etal.,  2011] (see
11    Figure 7-7].
      1 Recycling produced water may introduce radioactive material into hydraulic fracturing fluid. See section 8.4.3 and PA
      DEPf2015b1.
                   This document is a draft for review purposes only and does not constitute Agency policy.
      June 2015                                    7-18                  DRAFT—DO NOT CITE OR QUOTE

-------
      Hydraulic Fracturing Drinking Water Assessment
Chapter 7 - Flowback and Produced Water

7,000
6,000

„ 5,000
£; 4,000
1 3,000
PS

: .. i
• • 1
- iBB" -
-• ;
o 2,000 1-
1000 I •Total Radium
[ 	
200,000
180,000
160,000 ^
140,000 £
120,000 3
100,000 ^
HI
80,000 "I
60,000 5
re
40,000 o
20,000
n
                                           10          15
                                           Days Post-Fracturing
 20
25
      Figure 7-7. Total radium and TDS concentrations measured through time for injected (day 0),
                flowback, and produced water samples from mutually exclusive Greene County,
                PA, Marcellus Shale gas wells.
                Data from Rowan etal. (2011) and Chapman etal. (2012).

      7.4.5.  Leaching and Biotransformation of Naturally Occurring Organic Compounds
 1    Many organics are known to leach naturally into formation water through association with shale,
 2    sandstone, and coal strata [Benko and Drewes. 2008: Orem etal.. 2007). Oremetal. [2014] show
 3    that formation and produced waters from shale plays that were not impacted by production
 4    chemicals contain an array of organic compound classes associated with the formation. When
 5    unconventional formations are hydraulically fractured, additional organics from the freshly
 6    fractured hydrocarbon-bearing formation and the chemical additives contribute to a large increase
 7    in flowback and produced water organic loads [Orem etal.. 2014).

 8    The nature of the in-situ hydrocarbons reflects the formation's thermal maturity and heavily
 9    influences the organic content of the produced water.1 The Marcellus Shale is largely considered a
10    mature formation and therefore consists of wet and dry gas [Barbotetal.. 2013: Repetskietal..
11    2008). Conversely, the Utica Shale is less thermally mature; available hydrocarbon resources
12    consist of oil, condensate, and gas [Repetski et al., 2008]. Additionally, some coals within the
13    eastern and west-central regions of the San Juan Basin produce little to no water during production,
14    due to the regional thermal maturity, hydrostratigraphy, and in situ trapping mechanisms [New
15    Mexico Bureau of Mines and Mineral Resources. 1994).
      1 With increasing subsurface temperature after burial, petroleum source material (kerogen) produces hydrocarbons in a
      sequence from methane (immature), to oil (more mature), to gas (mature). Gas is produced by thermal cracking of oil (PA
      DCNR.2015).
                   This document is a draft for review purposes only and does not constitute Agency policy.
      June 2015                                    7-19                   DRAFT—DO NOT CITE OR QUOTE

-------
      Hydraulic Fracturing Drinking Water Assessment                        Chapter 7 - Flowback and Produced Water


 1    Dissolved organic carbon (DOC) concentrations decrease from hydraulic fracturing through
 2    flowback in shales and coalbeds [Murali Mohan et al.. 2013: Orem etal.. 2007]. DOC sorption,
 3    dilution with injected or formation water, biochemical reactions, and microbial transformation (i.e.,
 4    biotransformation in the form of degradation or uptake) may all cause decreased concentrations of
 5    DOC during flowback. Organic chemical additives injected during hydraulic fracturing offer a novel
 6    carbon and energy source for biotic and abiotic reactions at depth. Injected organics include many
 7    sugar-based polymer formulations, most notably of galactose and mannose (i.e., guar gum used as a
 8    gelling agent); hydrocarbon distillates used in crosslinkers, friction reducers, and gelling agents;
 9    and ethyl and ether glycol formulations used in non-emulsifiers, crosslinkers, friction reducers, and
10    gelling agents. (Wuchter etal.. 2013: Arthur etal.. 2009b: Hayes. 2009).

11    DOC and chloride concentrations exhibit strongly correlated inverse temporal trends (Barbotetal.,
12    2013: Chapman et al.. 2012] for flowback and produced water samples obtained from three
13    Marcellus Shale wells from the same well pad in Greene County, Pennsylvania (Cluff etal., 2014], as
14    shown in Figure 7-8. Chloride concentrations increased five- to six-fold as a function of residence
15    time (i.e., cumulative volumes of produced water). These  chloride concentrations followed an
16    increasing linear trend during the first two weeks of flowback (see Figure 7-8a, inset) then began to
17    approach asymptotic levels later in production, indicating that injected fluids had acquired a brine
18    signature as a result of subsurface mixing, fluid-solid interactions, and mineral dissolution
19    processes.

20    DOC concentrations exhibit an inverse trend and decreased through flowback and production
21    (Figure 7-8b) (Cluff etal.. 2014). DOC levels decreased approximately twofold between injected
22    fluid and initial flowback samples (Figure 7-8b, inset). DOC concentrations decreased by 11-fold
23    over the study's time frame (nearly 11 months) and leveled off several months after hydraulic
24    fracturing, presumably as a result of in situ attenuation processes (Cluff etal.. 2014).
                   This document is a draft for review purposes only and does not constitute Agency policy.
      June 2015                                    7-20                   DRAFT—DO NOT CITE OR QUOTE

-------
Hydraulic Fracturing Drinking Water Assessment
Chapter 7 - Flowback and Produced Water
100,000
90,000
80,000

70,000
c?
ao 60,000


;
.
__„„--'
_ --"
-
ir

£ 50,000 V
'C fmm
-2 40,000 J
30,000 I
b
20,000 r
10,000 T
7o,pga -i
60,000

^ 50,000
£ 40,000

~ 30,000
°
g 20,000
10,000


A • •

•
' " •*"

' • Welll
• Well 2
A Well 3








0 5 	 -10" 15
	 	 Days Post-Fracturing
„ i .--,-.--:-;-- 	
0 50 100














150 200 250 300 350
Days Post-Fracturing
                                               (a)

500
400 i
OD
&3QQ
u
0
Q
200
100
0
0
1 •- J

- - " 500
^400 1
to
£300 1
0 200
Q
100
0
(

• Well 1
* • Well 2
A Well 3
A* M
• • ,
AA

) 5 ^lO---"" 15
Da,ys •Post-Fracturing
•


50 100 150 200 250 300 350
Days Post-Fracturing
                                              (b)

Figure 7-8. (a) Chloride (Cl) and (b) DOC concentrations measured through time for injected
           (day 0), flowback, and produced water samples obtained from three Marcellus
           Shale gas wells from a single well  pad in Greene County, PA used for hydraulic
           fracturing.

           Data from Cluff et al. (2014).  Reprinted with permission from Cluff, M; Hartsock, A; Macrae, J; Carter,
           K; Mouser, PJ. (2014). Temporal changes  in microbial ecology and geochemistry in produced water
           from hydraulically fractured Marcellus Shale Gas Wells.  Environ Sci Technol 48: 6508-6517. Copyright
           2014 American Chemical Society.
              This document is a draft for review purposes only and does not constitute Agency policy.

June 2015                                      7-21                   DRAFT—DO NOT CITE OR QUOTE

-------
      Hydraulic Fracturing Drinking Water Assessment                       Chapter 7 - Flowback and Produced Water


 1    Orem etal. [2014] conducted a temporal study of two coalbed wells over the course of one year.
 2    Their results suggest that organic compound concentrations decrease over time. This trend may be
 3    due to pumping of water to the surface, which may exhaust coal-associated produced water.
 4    Subsequent produced water would not be associated with the coal. This suggests that the early
 5    produced water would contain the highest organic load [Orem etal., 2014].

 6    As noted above, most literature addresses general organic parameters such as bulk total organic
 7    carbon (TOG] or DOC instead of individual organic compounds [Sirivedhin and Dallbauman. 2004].
 8    Emphasis on the prevalence of bulk organics as opposed to unique organics is due largely to the
 9    lack of analytical standards for many compounds and also a lack of knowledge regarding the types
10    of organics to test in produced water samples [Schlegel etal., 2013: Strong etal., 2013].

      7.5.   Produced Water Composition
11    In this section, we discuss the characteristics of aggregated produced water data without regard for
12    temporal changes. Similarities between conventional and unconventional produced water are
13    noted and the variability between formation types is described. As we discuss below, produced
14    water may contain a range of constituents, but in widely varying amounts. Generally, these may
15    include:

16        •   Salts, including those composed from chloride, bromide, sulfate, sodium, magnesium and
17            calcium.
18        •   Metals including barium, manganese, iron, and strontium.
19        •   Dissolved organics including BTEX and oil and grease.
20        •   Radioactive materials including radium (radium-226 and radium-228].
21        •   Hydraulic fracturing chemicals and their transformation products.
2 2    We discuss these groups of chemicals and then conclude by discussing variability within formation
2 3    types and within production zones.

      7.5.1.  Similarity of Produced Water from Conventional and Unconventional Formations
24    Unconventional produced water is reported to be similar to conventional produced waters in terms
25    of TDS, pH, alkalinity, oil and grease, TOG, and other organics and inorganics [Wilson, 2014:
2 6    Haluszczak etal.. 2013: Alley etal.. 2011: Hayes. 2009: Sirivedhin and Dallbauman. 2004]. Although
27    salinity varies in shales and tight formations, produced water is typically characterized as saline
28    [Lee andNeff. 2011: Blauch et al.. 2009]. Produced water is also enriched in major anions (e.g.,
29    chloride, bicarbonate, sulfate], cations (e.g., sodium, calcium, magnesium], metals (e.g., barium,
30    strontium], naturally occurring radionuclides (e.g., radium-226 and radium-228] (Chapman etal.,
31    2012: Rowan etal.. 2011]. and organics (e.g., hydrocarbons] (Orem etal.. 2007:  Sirivedhin and
32    Dallbauman. 2004].

      7.5.2.  Variability in Produced Water Composition Among Unconventional  Formation Types
33    Alley etal. (2011] compared geochemical parameters of shale gas, tight gas, and CBM produced
34    water. This comparison aggregated data on produced water from original analyses, peer-reviewed

                   This document is a draft for review purposes only and does not constitute Agency policy.
      June 2015                                  7-22                  DRAFT—DO NOT CITE OR QUOTE

-------
      Hydraulic Fracturing Drinking Water Assessment
Chapter 7 - Flowback and Produced Water
 1    literature, and public and confidential government and industry sources and determined the
 2    statistical significance of the results. As shown in Table 7-4, Alley etal. [2011] found that of the
 3    constituents of interest common to all three types of unconventional produced water (calcium,
 4    chloride, potassium, magnesium, manganese, sodium, and zinc):

 5       1)  Shale gas produced water had significantly different concentrations from those of CBM;
 6       2)  Shale gas produced water constituent concentrations were significantly similar to those of
 7           tight gas, except for potassium and magnesium; and
 8       3)  Five tight gas produced water constituent concentrations (calcium, chloride, potassium,
 9           magnesium, and sodium) were significantly similar to those of CBM (Alley etal.. 2011).
10    The degree of variability between produced waters of these three resource types is consistent with
11    the degree of mineralogical and geochemical similarity between shale and sandstone formations,
12    and the lack of the same between shale and coalbed formations (Marshak. 2004).
      Table 7-4. Compiled minimum and maximum concentrations for various geochemical
                constituents in unconventional shale gas, tight gas, and CBM produced water (Alley
                et al., 2011).
                Source: (Alley et al., 2011).
Parameter
Alkalinity
Ammonium-N
Bicarbonate
Conductivity
Nitrate
Oil and grease
PH
Phosphate
Sulfate
Radium-226
Aluminum
Arsenic
Boron
Barium
Unit
mg/L
mg/L
mg/L
uS/cm
mg/L
mg/L
sud
mg/L
mg/L
pCi/g
mg/L
mg/L
mg/L
mg/L
Shale gasa
160-188
-
ND-4,000
-
ND-2,670
-
1.21-8.36
ND-5.3
ND-3,663
0.65-1.031
ND-5,290
-
0.12-24
ND-4,370
Tight gasb
1,424
2.74
10-4,040
24,400
-
42
5-8.6
-
12-48
-
-
0.17
-
-
CBMC
54.9-9,450
1.05-59
-
94.8-145,000
0.002-18.7
-
6.56-9.87
0.05-1.5
0.01-5,590
-
0.5-5,290
0.0001-0.06
0.002-2.4
0.01-190
                   This document is a draft for review purposes only and does not constitute Agency policy.
      June 2015                                    7-23                  DRAFT—DO NOT CITE OR QUOTE

-------
     Hydraulic Fracturing Drinking Water Assessment
Chapter 7 - Flowback and Produced Water
Parameter
Bromine
Calcium
Cadmium
Chlorine
Chromium
Copper
Fluorine
Iron
Lithium
Magnesium
Manganese
Mercury
Nickel
Potassium
Sodium
Strontium
Uranium
Zinc
Unit
mg/L
mg/L
mg/L
mg/L
mg/L
mg/L
mg/L
mg/L
mg/L
mg/L
mg/L
mg/L
mg/L
mg/L
mg/L
mg/L
mg/L
mg/L
Shale gasa
ND-10,600
0.65-83,950
-
48.9-212,700
-
ND-15
ND-33
ND-2,838
ND-611
1.08-25,340
ND-96.5
-
-
0.21-5,490
10.04-204,302
0.03-1,310
-
ND-20
Tight gasb
-
3-74,185
0.37
52-216,000
0.265
0.539
-
0.015
-
2-8,750
0.525
-
0.123
5-2,500
648-80,000
-
-
0.076
CBMC
0.002-300
0.8-5,870
0.0001-0.01
0.7-70,100
0.001-0.053
ND-0.06
0.05-15.22
0.002-220
0.0002-6.88
0.2-1,830
0.002-5.4
0.0001-0.0004
0.0003-0.20
0.3-186
8.8-34,100
0.032-565
0.002-0.012
0.00002-0.59
      -, no value available; ND, non-detect. If no range, but a singular concentration is given, this is the maximum concentration.
      3 n = 541. Alley etal. (2011) compiled data from USGS (2006): Mclntosh and Walter (2005): Melntosh et al. (2002) and
      confidential industry documents.
      b n = 137. Alley etal. (2011) compiled data from USGS (2006) and produced water samples presented in Alley etal. (2011).
      c Alley etal. (2011) compiled data from Montana GWIC (2009): Thordsen etal. (2007): ESN Rocky Mountain (2003): Rice etal.
      (2000); Rice (1999); Hunter and Moser (1990).
       SU = standard units.
1    Shale gas produced water tends to be more acidic, as well as, enriched in strontium, barium, and
2    bromide. CBM produced water is highly alkaline, and it contains relatively low concentrations of
3    TDS (one to two orders of magnitude lower than in shale and sandstone). It also contains lower
4    levels of sulfate, calcium,  magnesium, DOC, sodium, bicarbonate, and oil and grease than typically
                    This document is a draft for review purposes only and does not constitute Agency policy.
     June 2015                                        7-24                    DRAFT—DO NOT CITE OR QUOTE

-------
      Hydraulic Fracturing Drinking Water Assessment                        Chapter 7 - Flowback and Produced Water


 1    observed in shale and sandstone produced waters [Alley etal., 2011: Dahmetal., 2011: Benko and
 2    Drewes. 2008: Van Voast. 2003 j.1

      7.5.3.  General Water Quality Parameters
 3    For this assessment, the EPA identified data characterizing the content of unconventional flowback
 4    and produced water in a total of 12 shale and tight formations and CBM basins. These formations
 5    and basins span 18 states. Note that in this subsection we treat all fluids as produced. As a
 6    consequence, the variability of reported concentrations is likely higher than if the data could be
 7    standardized to a specific point on the flowback-to-produced water continuum.

 8    For most formations, the amount of general water quality parameter data is variable (see Table E-2
 9    of Appendix E). Average pH levels range from 5.87 to 8.19, with typically lower values for shales.
10    Larger variations in average specific conductivity are seen among unconventional formations and
11    range from 213 microsiemens (n.S)/cm in the Bakken Shale to 184,800 [j.S/cm in Devonian
12    sandstones (see Table E-2 of Appendix E). Shale and tight formation produced waters are enriched
13    in suspended solids, as reported concentrations for total suspended solids and turbidity exceed
14    those of coalbeds by one to two orders of magnitude.

15    Of the data presented in Table E-3 of Appendix E, differences are evident between the Black
16    Warrior and the three western formations (Powder River, Raton, and San Juan). The Black Warrior
17    is higher in average chloride, specific conductivity, TDS, TOG and total suspended solids; and lower
18    in alkalinity and bicarbonate than the other three. These differences are due to the saline or
19    brackish conditions during deposition in the Black Warrior that contrast to the freshwater
2 0    conditions for the western basins.

21    The average dissolved oxygen (DO) concentrations of CBM produced water range from 0.39-1.07
22    mg/L (see  Table E-3 Appendix E). By comparison, well-oxygenated surface water can contain up to
23    10 mg/L DO at 15 °C fU.S. EPA. 2012a). Thus, coalbed produced water is either hypoxic (less than 2
24    mg/L DO) or anoxic (less than 0.5 mg/L DO) and could contribute to aquatic organism stress (USGS,
25    2010: NSTC. 2000).

      7.5.4.  Salinity and Inorganics
26    The TDS profile of unconventional produced water is dominated by sodium and chloride, with large
27    contributions to the profile from mono- and divalent cations (Sun etal.. 2013: Guerraetal.. 2011).
28    In order of relative abundance, the following inorganic ions are typically found in highly saline
29    conventional produced water: sodium, chloride, calcium, magnesium, potassium, sulfate, bromide,
30    strontium, bicarbonate, and iodide (Lee and Neff, 2011). Shale and sandstone produced water tend
31    to be characterized as sodium-chloride-calcium water types, whereas CBM produced water tends to
32    be characterized as sodium chloride or sodium bicarbonate water types (Dahm etal., 2011).
33    Elevated levels of bromide, sulfate, and bicarbonate are also present (Sun etal., 2013). Elevated
34    strontium and barium levels are characteristic of Marcellus Shale flowback and produced water
      1 Several had low representation in the Alley etal. (2011) data set, including the Appalachian Basin (western New York
      and western Pennsylvania), West Virginia, eastern Kentucky, eastern Tennessee, and northeastern Alabama.
                   This document is a draft for review purposes only and does not constitute Agency policy.
      June 2015                                    7-25                  DRAFT—DO NOT CITE OR QUOTE

-------
      Hydraulic Fracturing Drinking Water Assessment                       Chapter 7 - Flowback and Produced Water


 1    [Barbotetal.. 2013: Haluszczaketal.. 2013: Chapman etal.. 20121. Data representing shales and
 2    tight formations are presented in Table E-4 of Appendix E.

 3    Rowan etal. [2015] acknowledge that the origin of saline water produced from the Marcellus Shale
 4    is a matter of debate. One idea is that injected fluid returns at higher salinity, because of dissolving
 5    halite and other minerals found in shale [Blauchetal., 2009]. Blauch and colleagues hypothesized
 6    that salt layers bearing barium, calcium, iron, potassium, magnesium, sodium, and strontium likely
 7    dissolve and contribute to flowback and produced water salinity [Blauchetal.. 2009]. However,
 8    actual mixing of formation water and fracturing fluid was postulated by Haluszczaketal. [2013]
 9    from arguments based on the near-neutral pH and low levels of chloride and sulfate in the
10    Marcellus data from Hayes [2009], as well as, the relationship between chloride and bromide.  Engle
11    and Rowan determined that water chemistry during the first 90 days of production is controlled by
12    mixing of injected and formation waters and stimulation of bacterial sulfate reduction [Engle and
13    Rowan, 2014: Haluszczaketal., 2013].  Rowan etal. [2015] argue, based on an observed shift to
14    isotopically heavier water, that produced water actually contains formation water.: Alternately,
15    Barbotetal. [2013] concluded from analysis of Marcellus Shale produced water that mixing (with
16    formation water] alone could not explain the observed patterns in chloride concentrations.

17    Marcellus Shale produced water salinities range from less than 1,500 mg/L to over 300,000 mg/L,
18    as shown by Rowan etal. [2011]. By comparison, the average salinity concentration for seawater is
19    35,000 mg/L. The TDS concentration of CBM produced water can be as low as 500 mg/L ranging to
20    nearly 50.000 mg/L fDahm etal.. 2011: Benko and Drewes. 2008: Van Voast. 20031. Lower
21    dissolved solids are expected from CBM produced water, in part, because some coals developed in
22    fresh water environments [Bouska, 1981]. Dahm etal. [2011] report TDS concentrations from a
23    composite CBM produced water database (n = 3,255] for western basins that often are less than
24    5,000 mg/L (85% of samples]. In other cases, as for the Black Warrior basin, TDS can be higher
25    along with concentrations of species that contribute to TDS (See Table E-5 Appendix E], such as
26    calcium, chloride, and sodium.

      7.5.5.   Metals
27    The metals content of unconventional produced water varies by well and site lithology, but is
28    typically dominated by the same metals that are associated with conventional produced water.
29    Unconventional produced water may also contain low levels of heavy metals (e.g., chromium,
30    copper, nickel, zinc, cadmium, lead, arsenic, and mercury]  [Hayes. 2009]. Data illustrating metal
31    concentrations in produced water appear in Tables E-6 and E-7 of Appendix E.
      1 The produced water becomes isotopically heavier because of increased prevalence of Oxygen-18 in the water, compared
      to the more prevalent Oxygen-16.
                   This document is a draft for review purposes only and does not constitute Agency policy.
      June 2015                                   7-26                  DRAFT—DO NOT CITE OR QUOTE

-------
      Hydraulic Fracturing Drinking Water Assessment                        Chapter 7 - Flowback and Produced Water


      7.5.6.  Naturally Occurring Radioactive Material (NORM) and Technologically Enhanced
             Naturally Occurring Radioactive Material (TENORM)
      7.5.6.1.  formation Solids Levels of NORM
 1    Elevated uranium levels in formation solids have been used to identify potential areas of natural
 2    gas production for decades [Fertl and Chilingar, 1988]. Marine black shales are estimated to contain
 3    an average of 15-60 ppm uranium depending on depositional conditions [Fertl and Chilingar,
 4    1988]. Shales that bear significant levels of uranium include the Barnett in Texas, the Woodford in
 5    Oklahoma, the New Albany in the Illinois Basin, the Chattanooga Shale in the southeastern United
 6    States, and a group of black shales in Kansas and Oklahoma [Swanson, 1955].

      7.5.6.2.  Produced Water Levels of TENORM
 7    When exposed to the environment or concentrated NORM is termed technologically-enhanced
 8    naturally-occurring radioactive material (TENORM].1 Radioactive materials commonly present in
 9    shale and sandstone sedimentary environments include uranium, thorium, radium, and their decay
10    products. These are present in most unconventional produced water, but particularly so in
11    Marcellus Shale produced water [Rowanetal., 2011: Fisher,  1998]. Low levels of uranium and
12    thorium return during flowback, typically in the concentrated form of mineral phases or organic
13    matter, due to insolubility under prevailing reducing conditions encountered within shale
14    formations (Nelson etal.. 2014: Sturchio etal.. 20011

15    Conversely, radium,  a decay product of uranium and thorium, is known to be  relatively soluble
16    within the redox range encountered in subsurface environments [Sturchio etal.. 2001: Langmuir
17    andRiese, 1985]. Dissolved radium primarily  occurs as Ra2+, but it complexes with carbonate,
18    chloride, and sulfate ions as well [Sturchio etal.. 2001: Langmuir and Riese. 1985]. Ra2+ can also
19    substitute for various cations (e.g., Ba2+, Ca2+, and Sr2+] during mineral precipitation, as is
20    sometimes the case with barite or anhydrite precipitation (Rowanetal., 2011].

21    Data from the Marcellus Shale show that radium and TDS produced water concentrations are
22    positively correlated (Rowanetal.. 2011: Fisher. 1998]. This pattern is expected for other
23    formations because radium remains adsorbed to mineral surfaces in low salinity environments,
24    then desorbs as solution salinity increases (Sturchio etal., 2001]. Controlling for this TDS
25    dependence, Marcellus Shale produced water  contains statistically more radium than
26    non-Marcellus Shale produced water, with a median total radium content of 2,460 picocuries per
27    liter (pCi/L] (n = 52] compared to 1,011  pCi/L (n = 91], respectively (Rowan etal., 2011]. Radium
28    levels in Marcellus produced water are at several thousand picocuries per liter, with maximum
29    concentrations of total radium (radium-226 and radium-228], radium-226 and radium-228
30    reported at approximately 18,000, 9,000, and  1,300 pCi/L, respectively fRo wan etal.. 20111 (see
31    Table E-8 in Appendix E]. Data from the Pennsylvania TENORM produced water study (PA DEP,
      1 The U.S. EPA Office of Radiation rhttp://www.epa.gov/radiation/tenorm/) states that technologically enhanced
      naturally occurring radioactive material (TENORM) is produced when activities such as uranium mining, or sewage
      sludge treatment, concentrate or expose radioactive materials that occur naturally in ores, soils, water, or other natural
      materials. Formation water containing radioactive materials would contain NORM, because they are not exposed;
      produced water would contain TENORM because it has been exposed to the environment.
                   This document is a draft for review purposes only and does not constitute Agency policy.
      June 2015                                    7-27                   DRAFT—DO NOT CITE OR QUOTE

-------
      Hydraulic Fracturing Drinking Water Assessment
Chapter 7 - Flowback and Produced Water
 1    2015) showed similar elevated levels, and consistently showed higher medians in unconventional
 2    compared to conventional formations (Table E-8 in Appendix E).

      7.5.7. Organics
 3    The organic content of produced water varies by well and lithology, but consists of certain naturally
 4    occurring and injected organic compounds. These organics may be dissolved in water or, for the
 5    case of oil production, in the form of a separate or emulsified phase. Produced water organics can
 6    contain any of the following: (1) volatile organic compounds (VOCs) such as benzene and toluene,
 7    (2) semi-volatile organic compounds (SVOCs) such as phenols; and/or (3) non-VOCs such as
 8    macromolecular natural organic matter [Orem etal.. 2014: Hayes. 2009: Benko and Drewes. 2008:
 9    Orem etal.. 2007: Sirivedhin and Dallbauman. 2004). Table 7-5 presents data from naturally
10    occurring organic chemicals in produced water.
      Table 7-5. Concentration ranges (mg/L) of several classes of naturally occurring organic
               chemicals in conventional produced water worldwide (reported in Neff, 2002).
               Source: (Neff, 2002)
Chemical class
TOC
Total organic acids
Total saturated hydrocarbons
Total benzene, toluene, ethylbenzene, and xylenes (BTEX)
Total PAH
Total steranes/triterpanes
Ketones
Total phenols (primarily C0-C5 phenols)
Concentration range (mg/L)
<0.1->11,000
<0.001-10,000
17-30
0.068-578
0.040-3
0.140-0.175
1-2
0.400-23
11    Several classes of naturally occurring organic chemicals are present in conventional and
12    unconventional produced waters, with large concentration ranges (Lee and Neff. 2011]. In addition
13    to data on total organic carbon (TOC) and dissolved organic carbon (DOC) as indicators of the
14    presence of organics, specifically identified organics include saturated hydrocarbons, BTEX, and
15    polycyclic aromatic hydrocarbon (PAHs) (see Table E-9 of Appendix E). Data are lacking on the
16    presence and concentration of many other types of organic chemicals that might be present in
17    produced water, because of their use in hydraulic fracturing fluid. There are a number of reasons
18    for this difference,  some of which could be related to analytical limitations, limited focus of
19    produced water studies, and undocumented subsurface reactions.

      7.5.8.  Reactions within Formations
2 0    The introduction of hydraulic fracturing fluids into the target formation induces a number of
21    changes to formation solids and fluids that influence the chemical evolution and composition of
                   This document is a draft for review purposes only and does not constitute Agency policy.
      June 2015                                   7-28                 DRAFT—DO NOT CITE OR QUOTE

-------
      Hydraulic Fracturing Drinking Water Assessment                        Chapter 7 - Flowback and Produced Water


 1    flowback and produced water. These changes can result from physical processes (e.g., rock
 2    fracturing and fluid mixing) and geochemical processes (e.g., introducing novel, oxygenated fluids)
 3    that mobilize trace or major constituents into solution.

 4    The creation of fractures exposes new formation surfaces to interactions involving hydraulic
 5    fracturing fluids and existing formation fluids. Formations targeted for unconventional
 6    development are composed of detrital, cement, and organic fractions. For example, elements
 7    potentially available for mobilization when exposed via fracturing include calcium, magnesium,
 8    manganese, and strontium in cement fractions, and silver, chromium, copper, molybdenum,
 9    niobium, vanadium, and zinc in organic fractions. The storage or release of these elements in newly
10    exposed surfaces is variable and not well studied, in part due to the vast number of possible
11    interactions occurring continuously in the environment at the rock surface (Vine and Tourtelot,
12    19701

13    Contact with and physical mixing of hydraulic fracturing fluids with existing formation brines also
14    influences the geochemical evolution of produced water. For instance, Marcellus Shale brines have
15    high concentrations of bromide, calcium, chloride, magnesium, sodium, and strontium (Engle and
16    Rowan, 2014). Hydraulic fracturing fluid contains elevated levels of DOC, alkalinity, and sulfate
17    (Engle and Rowan, 2014). Consequently, flowback acquires a geochemical signature reflecting both
18    injected and formation fluids. Produced water containing bothAlthough some constituents of
19    hydraulic fracturing fluids are known to readily degrade in the environment, little is known
20    regarding how the subsurface degradation proceeds or how the constituents interact within a
21    complex matrix of organics (Mouser et al., In Press).

      7.6.   Spatial Trends
22    As was reported for the volume of produced water (see Section 7.2.2), the composition of produced
23    water varies spatially on a regional to local scale according to the geographic and stratigraphic
24    locations of each well within a hydraulically fractured formation (Bibby etal.. 2013: Lee and Neff,
25    2011). Spatial variability of produced water content occurs (1) between plays of different rock
26    sources (e.g., coal vs. sandstone), (2) between plays of the same rock type (e.g., Barnett Shale vs.
27    Bakken Shale), and (3) within formations of the same source rock (e.g., northeastern vs.
28    southwestern Marcellus Shale) (Barbotetal.. 2013: Alley etal.. 2011: Breit. 2002).

29    Geographic variability in produced water content has been established at a regional scale for
30    conventional produced water. As an example, Benko and Drewes (2008) demonstrate TDS
31    variability in conventional produced water among fourteen western geologic basins (e.g., Williston,
32    San Juan, and Permian Basins). Median TDS in these basins range from as low as 4,900 mg/L in the
33    Big Horn Basin to as high as 132,400 mg/L in the Williston Basin based on over 133,000 produced
34    water samples from fourteen basins (Benko and Drewes. 2008).*
      1 Data were drawn from the USGS National Produced Water Geochemical Database v2.0. Published updates made in
      October 2014 to the database (v2.1) are not reflected in this document.
                   This document is a draft for review purposes only and does not constitute Agency policy.
      June 2015                                    7-29                  DRAFT—DO NOT CITE OR QUOTE

-------
      Hydraulic Fracturing Drinking Water Assessment                       Chapter 7 - Flowback and Produced Water


 1    High IDS (more than 200,000 ppm) is common throughout the central portion of the United States
 2    in various basins. Low TDS (<10,000 ppm) was found in the basins of the Rocky Mountains, and
 3    sometimes in Texas and California. In other areas, there was a mixture of mid-range, which in the
 4    case of Illinois was correlated to the depth of producing zones [Breit, 2002].

 5    Data further illustrating variability within shale, tight-gas and coalbed formations at both the
 6    formation and local scales are presented and discussed in Section E-3 of Appendix E.

      7.7.    Spill Impacts on Drinking  Water Resources
 7    Surface spills of flowback and produced water from unconventional oil and gas production have
 8    occurred across the country and in some cases have caused impacts to drinking water resources, as
 9    described in this section. Released fluids, if not contained on-site, may flow into nearby surface
10    waters or infiltrate into ground water via soil. In this section, we first briefly describe the potential
11    for spills from produced water handling equipment Next, we address individually-reported spill
12    events. These have originated from pipeline leaks, well blowouts, well communication events,  and
13    leaking pits and impoundments. We then summarize several studies of aggregated spill data, most
14    of which are based on state agency spill reports. The section concludes with discussion of two  cases.

      7.7.1.  Produced Water Management and Spill Potential
15    Produced water is typically transported from the wellhead through a series of pipes or flowlines to
16    on-site storage or treatment units [GWPC and IOGCC,  2014]. Faulty connections at either end of the
17    transfer process or leaks or ruptures in the lines carrying the fluid can result in surface spills.

18    Recovered fluids may be transferred to surface impoundments for long-term storage and
19    evaporation. Surface impoundments are typically uncovered earthen pits that may or may not be
20    lined. Recovered fluids may overflow from surface impoundments due to improper pit design and
21    weather events.

22    Produced water that is to be treated or disposed of off-site is typically stored in storage tanks or
23    impoundments until it can be loaded into transport trucks for removal [Gilmore etal.. 2013]. Tank
24    storage systems are typically closed loop systems in which produced water is transported from the
25    wellhead to aboveground storage tanks through interconnecting pipelines [GWPC and IOGCC.
26    2014]. Failure of connections and lines during the transfer process or the failure of a storage tank
2 7    can result in a surface release of fluids.

28    Depending on its characteristics, produced water, may be recycled and reused on-site. It can be
29    directly reused without treatment (after blending with freshwater] or it can be treated on-site prior
30    to reuse (Boschee. 2014]. As with other flowback management options, these systems also present
31    spill potential during transfer of fluids.
                   This document is a draft for review purposes only and does not constitute Agency policy.
      June 2015                                   7-30                  DRAFT—DO NOT CITE OR QUOTE

-------
      Hydraulic Fracturing Drinking Water Assessment                       Chapter 7 - Flowback and Produced Water


      7.7.2. Spills of Hydraulic Fracturing Flowback and Produced Water from Unconventional Oil
            and Gas Production
      7.7.2.1.  Pipeline Leaks
 1    In some locations, pipelines are used to transport produced water. Aggregated information on
 2    pipeline leaks from the whole country is not available. This section, rather, contains examples of
 3    incidents that have occurred. A leak was detailed in a field report from PA PEP (2009a], which
 4    described a leak from a 90-degree bend in an overland pipe carrying a mixture of flowback and
 5    freshwater between two impoundments. Along a 0.4 mi (0.6 km) length of the impacted stream,
 6    168 fish and 6 salamanders were killed; beyond a confluence at 0.6 km with a creek no additional
 7    dead fish were found. The release was estimated at 250 bbl (11,000 gal or 40,000 L). In response to
 8    the incident, the pipeline was shut off, a dam was constructed for recovering the water, water was
 9    vacuumed from the stream, and the stream was flushed with fresh water (PA PEP, 2009a].

10    In January 2015, 70,000 barrels (2,940,000 gal or 11,130,000 L) of produced water containing
11    petroleum hydrocarbons (North Dakota Department of Health. 2015] were released from a broken
12    pipeline that crosses Blacktail Creek in Williams County, ND. The response included placing
13    adsorbent booms in the creek, excavating contaminated soil, removing oil-coated ice, and removing
14    produced water from the creek. The electrical conductivity and chloride concentration in water
15    along the creek, the Little Muddy River, and Missouri  River were found to be elevated above
16    background levels, as were samples taken from ground water recovery trenches.

17    More incidents from North Dakota are documented at the North Dakota Department of Health
18    (NDDOH) Environmental Health web site (see http://www.ndhealth.gov/EHS/Spills/]. For the
19    period from November, 2012 to November 2013, NDDOH reported 552 releases of produced water
20    which were retained within the boundaries of the production or exploration facility and 104 which
21    were not (see http://www.ndhealth.gov/ehs/foia/spills/ChartWebPageOG 20121101
22    20131111.pdf).

      7.7.2.2.   Well Blowouts
23    Fingerprinting of water from two monitoring wells in Killdeer, ND, was used to determine that
24    brine contamination in the two wells resulted from a well blowout during a hydraulic fracturing
25    operation. Although the target formation was the Bakken Shale, data indicated that the residual
26    signatures of the brine were characteristic of the overlying Madison limestone formation (U.S. EPA,
27    2015)]. Prior research into out-of-zone hydraulic fracturing of the Bakken formation indicated that
28    a large number of hydraulically fractured wells contain water that is external to the Bakken Zone
29    fArkadakskiy and Rostron. 2013a: Arkadakskiy and Rostron. 2012a: PetermanetaL 20121 The
30    Bakken wells that contained external water were found to all contain water from the Mississippian
31    Lodgepole formation (part of the Madison Group]. The average volume of external water was 34%
32    and the external water volume ranged from 10% to 100% (Arkadakskiy and Rostron, 2013a].
33    Another example of a well blowout associated with a hydraulic fracturing operation occurred in
34    Clearfield County, PA. The well blew out, resulting in an uncontrolled flow of approximately
35    35,000 gal (132,000 L] of brine and fracturing fluid, along with an unquantified amount of gas;
36    some of the fluids reportedly reached a nearby stream (Barnes, 2010]. The blowout occurred while
                   This document is a draft for review purposes only and does not constitute Agency policy.
      June 2015                                   7-31                  DRAFT—DO NOT CITE OR QUOTE

-------
      Hydraulic Fracturing Drinking Water Assessment                       Chapter 7 - Flowback and Produced Water


 1    the company was drilling out the plugs used to isolate one fracture stage from another. An
 2    independent investigation found that the primary cause of the incident was that the only blowout
 3    preventer on the well had not been properly tested. In addition, the company did not have certified
 4    well control experts on hand or a written pressure control procedure [Vittitow, 2010]. In North
 5    Dakota, a blowout preventer failed, causing a release of between 50 and 70 barrels per day (2,100
 6    gal/day or 7,900 L/day and 2,940 gal/day or 11,000 L/day) of flowback and oil [Reuters. 20141. A
 7    3-ft berm was placed around the well for containment. Frozen droplets of oil and water sprayed on
 8    a nearby frozen creek. Liquid flowing from the well was collected and trucked offsite. Multiple well
 9    communication events reported by the media have also led to flowback and produced water spills
10    ranging from around 700 to 35,000 gal (2,600 L to 130,000 L) (Vaidyanathan. 2013a].

      7.7.2.3. Leaks from Pits and Impoundments
11    Leaks of flowback and produced water from on-site pits and impoundments have caused releases
12    as large as 57,000 gal (220,000 L) and have caused surface and ground water impacts
13    (Vaidyanathan. 2013b: PBFC2011: PADEP 2010]. VOCs have been measured in groundwater near
14    the Duncan Oil Field in New Mexico downgradient of an unlined pit storing produced water (Sumi,
15    2004: Eiceman. 1986]. Aspects of environmental transport from unlined pits are discussed below in
16    Section 7.8.5.

17    Two of the EPA's retrospective case studies found potential impacts from produced water
18    impoundments. In the southwest Pennsylvania case study (U.S. EPA, 2015k], elevated chloride
19    concentrations and their timing relative to historical data suggested a recent ground water impact
20    to a private water well occurred near an impoundment. The water quality trends suggested that the
21    chloride anomaly was related to the impoundment, but site-specific data were not available to
22    provide definitive assessment of the causes(s] and the longevity of the impact Evaluation of other
23    water quality parameters did not provide clear evidence of flowback or produced water impacts.

24    In the Wise County, TX case study (U.S. EPA, 2015m], impacts to two water wells were attributed to
25    brine, but the data collected for the study were not sufficient to distinguish among four possible
26    sources, one of which was leaks from reserve pits and/or impoundments. The others were: brine
27    migrating from underlying formations along wellbores, brine migrating from underlying formations
28    along natural fractures, and brine migrating from a nearby brine injection well. Alternate sources
29    for the impacts were considered, including road salting, landfill leachate, septic tanks, and animal
30    wastes, but evaluation of data showed that these were not likely. A third well experienced similar
31    impacts, but a landfill leachate source could not be ruled out in that case. Ricther and Kreitler
32    (1993]  reviewed sources of salinity to ground water resources by evaluating reviewing major
33    sources, which included natural saline ground water, halite dissolution, sea-water intrusion, oil-
34    field brine, agriculture, saline seeps and road salt For each source Ricther and Kreitler (1993]
35    provided a state-by-state review of the potential occurrence, which can be used as a general guide
36    to potential sources of salt at a specific area of interest.

      7.7.2.4. Data Compilation Studies
37    Environmental impacts from hydraulic fracturing-related fluids have been explored to a limited
38    extent in recent scientific literature (Brantley etal.. 2014: Farag and Harper. 2014: Gross etal..

                   This document is a draft for review purposes only and does not constitute Agency policy.
      June 2015                                   7-32                  DRAFT—DO NOT CITE OR QUOTE

-------
      Hydraulic Fracturing Drinking Water Assessment                       Chapter 7 - Flowback and Produced Water


 1    2013: Olmsteadetal.. 2013: Papoulias and Velasco. 2013: Vidic etal.. 2013: Considine etal.. 2012:
 2    Rozell and Reaven. 20121.

 3    From an Oklahoma Corporation Commission database of almost 13,000 releases reported from
 4    1993 to 2003, Fisher and Sublette [2005] determined that the primary origins of produced water
 5    releases were leaks from lines, tanks, wellheads, with lesser numbers of releases from surface
 6    equipment, and pits. The most  common cause was overflows followed by illegal activity, storms, fire
 7    accidents and corrosion. For these types of releases, the median release volume ranged from 20 bbl
 8    (840 gal or 3,180 L) to 60 bbl (2,500 gal or 9,500 L), and the maximums from 200 bbl (8,400 gal or
 9    31,800  L) to 2,800 bbl (118,000 gal or 445,200 L).

10    As noted in Text Box 5-14 of Chapter 5, U.S. EPA(2015n] characterized hydraulic fracturing-related
11    spills. Of the spills related to hydraulic fracturing activities (457 spills), 225 (49%) were spills of
12    flowback and produced water.  These spills were characterized with respect to volumes, spilled
13    materials, sources, causes, environmental receptors, containment, and response. Most of the
14    produced water spills in the EPA study occurred in Colorado (48%) and Pennsylvania (21%).
15    Flowback and produced water  constitute 84% (approximately 2.0 million gal or 7.6 million L) of the
16    total volume of hydraulic fracturing-related spills as calculated from Appendix B of U.S. EPA
17    [2015I1J.1 Flowback and produced water spills were characterized by numerous low-volume spills;
18    half of the spills with reported  volumes were less than 1,000 gal (3,800 L), and few spills exceeded
19    10,000  gal (38,000 L). Of the volume of spilled flowback and produced water, 16% was recovered
20    for on-site use or disposal, 76% was reported as unrecovered, and 8% was unknown. The potential
21    impact  of the unknown and unrecovered volume on drinking water resources is unknown.

22    Known sources for flowback and produced water spills include storage containers (e.g., pits,
23    impoundments, or tanks), wells or wellheads, hoses or lines, and equipment. Storage containers
24    accounted for 58% of flowback and produced water spills. The fewest spills occurred from wells
25    and wellheads, but these spills  had the greatest spill volumes compared to all other sources.

26    The causes of these spills were human error (38%), equipment failure (17%), failures of container
27    integrity (13%), miscellaneous causes (e.g., well communication, well blowout), and unknown
28    causes.  Most of the volume spilled (74%), however, came from spills caused by a failure of
29    container integrity.

30    In some of the cases, spills reached environmental receptors: soil (141 spills), surface water (17
31    spills), and ground water (1 spill); of these spills, 13 reached both soil and surface water.
32    Consequently 146 unique produced water spills reached environmental  receptors, accounting for
33    65% of the 225 cases and accounting for approximately 422,000 gal (1.60 million L) of flowback
34    and produced water. Spills with known volumes that reached a surface water body ranged from
       Chemicals and products, fracturing fluid, fracturing water, equipment fluids, hydrocarbons, and unknown fluids
      constitute the additional 16% (approximately 360,000 gal or 1.4 million L) of the total volume of hydraulic
      fracturing-related spills as calculated from Appendix B of U.S. EPAf2015d).
                   This document is a draft for review purposes only and does not constitute Agency policy.
      June 2015                                    7-33                  DRAFT—DO NOT CITE OR QUOTE

-------
      Hydraulic Fracturing Drinking Water Assessment                       Chapter 7 - Flowback and Produced Water


 1    less than 170 gal (640 L) to almost 74,000 gal (280,000 L). In 30 cases, it is unknown whether a spill
 2    of flowback and produced water reached an environmental receptor of any type.

 3    Gross etal. (2013) analyzed the Colorado Oil and Gas Conservation Commission's database for
 4    ground water BTEX concentrations linked to hydraulic fracturing-related surface spills between
 5    July 2010 and July 2011 in Weld County, CO. Only spills with an impact on ground water were
 6    included in the study. The 77 reported spills accounted for less than 0.5% of nearly 18,000 active
 7    wells. Forty-six of the 77 spills consisted of produced water and oil. Of the remaining spills, 23
 8    consisted of only oil and 8 consisted of only produced water. Thus the results that follow include
 9    cases with no produced water spill. From these composited spills, benzene concentrations in 90%
10    of the ground water samples exceeded 5 [ig/L, the U.S. drinking water standard. Additionally, 30%
11    of toluene, 12% of ethylbenzene, and 8% of xylene sample concentrations exceeded 1 mg/L, 0.7
12    mg/L and 10 mg/L, respectively (Gross etal.. 2013).

13    Based on five spills for which volumes were reported, the average volume of a produced water spill
14    was 294 gal (1,110 L), ranging from 42 (160 L) to 1,176 gal (4,450 L) (Gross etal.. 2013). Spill areas
15    averaged 2,120 ft2, with an average depth of 7 ft Tank battery systems and production facilities
16    were the biggest volume sources of spills with ground water impacts. Equipment failure was the
17    most common cause of spills with ground water impacts. Shallow ground water within the study
18    area (Niobrara Shale within the Denver-Julesburg Basin) is the main source of water for residents
19    due to limited surface water availability. Of the 77 reported spills, secondary containment was
20    absent from 51 of them (Gross etal.. 2013).

21    As noted from the Colorado (Gross etal., 2013) and Oklahoma (Fisher and Sublette, 2005) studies,
22    oil releases may occur alongside produced water spills. Review of recent oil field incidents in North
23    Dakota also shows incidents with both produced water and oil releases
24    (http://www.ndhealth.gov/EHS/Spills/). Oil releases are characterized by a number of features
25    including their unique hydrocarbon composition and physical properties. Impacts can include:
26    surface runoff, infiltration into soils, formation of sheens and oil slicks on surface waters,
27    evaporation, oxidation, biodegradation, emulsion formation, and particle deposition (U.S.  EPA.
28    1999).

29    A statistical analysis  of oil and gas violations in Pennsylvania found that violations regarding
30    structurally unsound impoundments or inadequate freeboard (vertical distance from the surface
31    water level to the overflow elevation) were the second most frequent type of violation with 439
32    instances in the period from 2 008 to 2010 (Olawoyin et al.. 2013). In a study of pits and
3 3    impoundments in West Virginia, Ziemkiewicz etal. (2014) found common problems of slope
34    stability and liner deficiencies. Construction quality control and quality assurance were often
35    inadequate; the authors found a lack of field compaction testing, use of improper types of soil,
36    excessive slope lengths, buried debris, and insufficient erosion control (Ziemkiewicz etal.. 2014).

37    Brantley et al. (2014) reviewed PA DEP's online oil and gas compliance database for notices of
38    violation issued to companies developing unconventional gas resources. Between May 2009 and
39    April 2013, 8 spills of flowback and produced water ranging from more than 4,000 gal (15,000 L) to
                   This document is a draft for review purposes only and does not constitute Agency policy.
      June 2015                                    7-34                  DRAFT—DO NOT CITE OR QUOTE

-------
      Hydraulic Fracturing Drinking Water Assessment                        Chapter 7 - Flowback and Produced Water


 1    more than 57,000 gal (220,000 L) reached surface water resources. The spills typically resulted in
 2    local impacts to environmental receptors and required remediation and monitoring. However, the
 3    study indicated the likelihood of a leak or spill of hydraulic fracturing-related fluids was low (less
 4    than 1%, based on 32 large spills outof more than 4,000 complete wells). Due to lack of data,
 5    specific impacts to the eight receiving surface waters were not discussed, other than noting the
 6    produced water had contacted the surface water.

 7    The Brantley et al. (2014) analysis of the PA DEP positive determination letters written in response
 8    to water-user complaints illustrates the importance of pre-drilling sampling, as it is one criterion
 9    that allows operators to refute the presumption that drilling caused water supply impacts (see
10    Chapter 6). The importance of this pre-drilling sampling and analysis is highlighted by naturally
11    occurring exceedances of EPA secondary MCLs for manganese and iron in private wells in
12    Pennsylvania (Boyer etal.. 2011: Williams etal.. 19981 Boyer etal. (2011) state that more than
13    40% of private water wells in Pennsylvania fail to meet federal drinking water standards. Boyer et
14    al. (2011) analyzed pre-drilling samples from private water wells in northeastern and
15    southwestern Pennsylvania and showed that 20% (of 222 wells) failed the drinking water standard
16    for iron and 27% (of 203 wells) failed for manganese.1 Williams etal. (1998). in their evaluation of
17    over 200 wells in Bradford, Tioga, and Potter counties in northeastern Pennsylvania, indicate about
18    50% of the wells exceeded secondary MCLs for iron and manganese.2 According to Boyer et al.
19    (2011), higher concentrations of these constituents tend to be associated with the sodium chloride
20    (Na-Cl) type  ground water often found in valleys in zones of more restricted ground water flow
21    (portions of aquifers with low permeability). Saline water can be found at shallow depths in these
22    areas (Williams etal.. 1998).

23    As an example of another set of criteria for assessing sites potentially contaminated by hydraulic
24    fracturing activities, the EPA (2012f) developed an approach to study sites where the impacts to
25    drinking water resources and the potential sources of the impacts are unknown, but may have been
26    the object of water-user complaints. The approach is based on a tiered scheme where results from
27    each tier are used to refine activities in higher tiers. The four tiers were as follows:

28         •   Verify potential issue:
29                o  Evaluate existing data and information from operators, private citizens, state and
30                   local agencies, and tribes (as appropriate).
31                o  Conduct site visits.
32                o  Interview stakeholders and interested parties.
33         •   Determine approach for detailed investigations:
34                o  Conduct initial sampling of water wells, taps, surface water and soils.
      1 Percentage of other parameters failing standards: 17% of 233 wells for pH, 3% of 233 wells for TDS, <1% of 226 wells
      for chloride, 1% of 218 wells for Barium, <1% of 177 wells for sulfate, 33% of 125 wells for coliforms, 4% of 115 wells for
      arsenic, 8% of 122 wells for fecal coliforms, 32% of 102 samples for turbidity.
      2 Naturally occurring constituents occasionally exceeding EPA primary MCLs in this area include barium, combined
      radium-226 and radium-228, and arsenic.
                    This document is a draft for review purposes only and does not constitute Agency policy.
      June 2015                                     7-35                   DRAFT—DO NOT CITE OR QUOTE

-------
      Hydraulic Fracturing Drinking Water Assessment                       Chapter 7 - Flowback and Produced Water


 1                o  Identify potential evidence of drinking water contamination.
 2                o  Develop conceptual site model describing possible sources and pathways of the
 3                   reported or potential contamination.
 4                o  Develop,  calibrate, and test fate and transport model(s).
 5         •   Conduct detailed investigations to detect and evaluate potential sources of contamination:
 6                o  Conduct additional sampling of soils, aquifer, surface water, and produced water
 7                   pits/tanks where present
 8                o  Conduct additional testing, including further water testing with new monitoring
 9                   points, soil gas surveys, geophysical testing, well mechanical integrity testing, and
10                   stable isotope analyses.
11                o  Refine conceptual site model and further test exposure scenarios.
12                o  Refine fate and transport model(s) based on new data.
13         •   Determine the source(s) of any impacts to drinking water resources:
14                o  Develop multiple lines of evidence to determine the source(s) of impacts to
15                   drinking  water resources.
16                o  Exclude possible sources and pathways of the reported contamination.
17                o  Assess uncertainties associated with conclusions regarding the source(s) of
18                   impacts.
19    This tiered assessment strategy provides an outline for collecting data and evaluating lines of
20    evidence for determining whether impacts have occurred.

      7.7.3.  Case Studies of Potentially Impacted Sites
      7.7.3.1.  Flowback and Produced Water Release from an Illegal Discharge Impacts Surface
               and Ground Water in Lycoming County, Pennsylvania
21    An estimated 6,300 gal to  more than 57,000 gal (24,000 to 220,000 L) of Marcellus Shale produced
22    water was illegally discharged atXTO Energy Inc.'s Marquardtpad and flowed into the
23    Susquehanna River watershed in November 2010 [U.S. EPA. 2013g].1 Overland and subsurface flow
24    of released fluids affected proximal surface water, a subsurface spring, and soil. No impacts to
25    drinking water wells and springs within 1 mile of the release were observed at the last sampling
26    date (17 days post-spill). However, residual, soil-associated  produced water constituents could
27    reach drinking water resources in the future through surface runoff or infiltration to the ground
28    water (Science Applications International Corporation. 2010). The release, which occurred atXTO's
29    Marquardt 8537H well pad in Penn Township, Lycoming County, PA, was discovered after a routine
3 0    inspection by the Pennsylvania Department of Environmental Protection. Subsequent investigation
      1 Violations associated with this incident can be found at the Pennsylvania Department of Environmental Protection's Oil
      and Gas Compliance Report database found at http://www.portal.state.pa.us/portal/
      server.pt/community/oil and gas compliance report/20299 under the following inspection IDs: 1928978,1928992, and
      1929005.
                   This document is a draft for review purposes only and does not constitute Agency policy.
      June 2015                                    7-36                  DRAFT—DO NOT CITE OR QUOTE

-------
      Hydraulic Fracturing Drinking Water Assessment                       Chapter 7 - Flowback and Produced Water


 1    revealed that flowback and produced water had been discharging into surface waters for over
 2    two months after the fluid was released from multiple tanks with open or missing valves on
 3    November 16, 2010 fU.S. EPA. 2013gj.

 4    Geochemical characterization of this produced water indicated concentrations of barium, chloride,
 5    iron, manganese, and IDS above Pennsylvania's surface water quality standards, and above the
 6    statewide health standards for medium-specific concentrations (SHS MSCs) for ground water use in
 7    residential and nonresidential settings [Science Applications International Corporation. 2010). The
 8    produced water also contained elevated levels of bromide, calcium, sodium, and strontium, which
 9    lack state surface water quality standards and SHS MSCs [Science Applications International
10    Corporation. 2010].

11    Post-spill surface water delineations indicated that released fluids migrated to an unnamed stream
12    known as Tributary 19617. The released fluids migrated approximately 1,400 ft (427 m) overland
13    to a depression which contains a natural fracture with a hydrological connection to a spring
14    [Schmidley and Smith. 2011). The distance from the depression to the stream is approximately
15    600 ft (183 m). Released fluids also drained through surface soils into ground water, which was
16    then released in seeps to the spring and stream; elevated levels of barium, bromide, calcium,
17    chloride, sodium, strontium, and TDS resulted (U.S. EPA. 2013gj. Elevated levels of these
18    constituents, particularly barium, bromide, and strontium, were indicative of Marcellus Shale
19    flowback and produced water that had mixed with surface water (Brantley etal.. 2014). Barium and
20    chloride were the only dissolved constituents detected in the stream that exceeded state surface
21    water quality standards; the remaining constituents lack established state surface water quality
22    standards.

2 3    Results from XTO's temporal study of surface water quality confirmed impacts to the stream from
24    produced water. Surface water quality was characterized at the confluence of the stream and
25    spring, and at the stream's upstream and downstream segments, for 65 days post-spill (Science
26    Applications International Corporation. 2010]. Downstream barium and bromide levels were one to
27    two orders of magnitude greater than upstream levels through this period. In addition, stream
28    strontium levels were two to three orders of magnitude greater than upstream levels at this time.
29    Chloride was initially detected in the stream with concentrations exceeding state water quality
30    standards (Schmidley and Smith, 2011]. Average chloride concentrations for stream samples were
31    two orders of magnitude greater than upstream concentrations (PADEP, 2011c]. By January 2011,
32    stream chloride concentrations had dropped below the limit established by Pennsylvania's surface
33    water quality standards.

34    Delineation of chloride concentrations within on-site soil indicated soil impacts due to overland
35    flow of flowback and produced water (Science Applications International Corporation, 2010]. Five
36    hundred tons of affected soil was consequently excavated for off-site disposal. Chloride
37    concentrations decreased with increased distance from the spill site but remained elevated above
38    background levels even at distances of a few thousand feet (Science Applications International
39    Corporation. 2010]. Produced water constituents that were present in soil at concentrations above
40    background levels (i.e., barium, sodium, strontium] could be available for long-term runoff and

                   This document is a draft for review purposes only and does not constitute Agency policy.
      June 2015                                    7-37                  DRAFT—DO NOT CITE OR QUOTE

-------
      Hydraulic Fracturing Drinking Water Assessment                        Chapter 7 - Flowback and Produced Water


 1    infiltration. For instance, the continued presence of chloride in affected soils is likely due to
 2    precipitated chloride salts in soil pores from residual produced water [Science Applications
 3    International Corporation. 2010). which may leach later. Near-term sampling (up to 17 days after
 4    the spill) found no elevated constituents indicative of runoff or infiltration of produced water when
 5    XTO sampled 14 drinking water wells and springs within one mile of the well pad. XTO was ordered
 6    to create a storm water collection system for off-site disposal of impacted storm water and to
 7    establish on-site water monitoring wells to track long-term ground water quality between the well
 8    pad and the stream [Schmidley and Smith. 2011). Other cases of illegal dumping have been
 9    reported [Caniglia. 2014: U.S. EPA. 2013g: Paterra. 2011].

      7.7.3.2.  Flowback Fluid Reaches Towanda Creek Due to Well Blowout in Bradford County,
               Pennsylvania, Causing Short-Term Impacts
10    The Chesapeake Energy ATGAS 2H well, located in Leroy Township, Bradford County, PA,
11    experienced a wellhead flange failure on April  19, 2011, during hydraulic fracturing operations.
12    Approximately ten thousand gallons (38,000 L) of flowback fluids spilled into an unnamed tributary
13    of Towanda Creek, a state-designated trout stock fishery and a tributary of the Susquehanna River
14    (USGS. 2013b:  SAIC and GES. 2011). Chesapeake conducted post-spill surface and ground water
15    monitoring (SAIC and GES. 2011). In addition, the EPA, PADEP, and Chesapeake collected split
16    samples from seven private wells within the vicinity of the blowout. The EPA requested thatthe
17    Agency for Toxic Substances and Disease Registry (ATSDR) evaluate the environmental data
18    collected from  seven private wells to determine whether harmful health effects would be expected
19    from consuming or using the well water. Data from Pre-blowout private well samples, collected
2 0    approximately six months prior to drilling activity at the site, were included in the evaluation.

21    Between the pre- and post-blowout samples, ATSDR (2013] determined that there was factor of ten
22    increases in some analyte concentrations (methane, barium, calcium, chloride, magnesium,
23    manganese, potassium, and sodium) and a factor of 7 increase in iron concentration in one well
24    (RW04) near the site. Other wells showed elevated levels of certain analytes.1 ATSDR concluded
25    that although the available data suggested that the ground water near this site is impacted by gas
26    activities, the data for RW04 did not conclusively indicate an impact ATSDR (2013) concluded that
27    further evaluation is needed to characterize any relationship between the drinking water wells and
28    aquifers as a result of changes in site conditions. Further sampling would be required to determine
29    current impacts, trends, and chronic exposures to ground water constituents related to natural gas
30    activities.
      1 Elevated sodium levels were detected in 6 wells, levels in 5 of them (RW02, RW03, RW05, RW06, and RW07) may be of
      concern to sensitive subpopulations; while the last (RW04) would exceed the dietary guideline for both sensitive and the
      general population. ATSDR judged that elevated lithium concentration in two wells (RW04 and RW06) could be a concern
      to individuals undergoing lithium therapy. One well (RW02) showed elevated arsenic concentrations, but these were
      similar in the pre- and post-blowout samples. Gross alpha radiation levels were above the EPA maximum contaminant
      level in one well (RW03), and ATSDR did not expect adverse health effects from drinking this water. ASTDR did not expect
      adverse health effects for the user of five private wells (RW01, RW03, RW05, RW06 (excepting for possible lithium
      impacts) and RW07).
                    This document is a draft for review purposes only and does not constitute Agency policy.
      June 2015                                     7-38                  DRAFT—DO NOT CITE OR QUOTE

-------
      Hydraulic Fracturing Drinking Water Assessment                       Chapter 7 - Flowback and Produced Water


 1    Chesapeake concluded that there were short-term impacts to surface waters of a farm pond within
 2    the vicinity of the well pad, the unnamed tributary, and Towanda Creek following the event [SAIC
 3    and GES. 20111 The lower 500 feet of the unnamed tributary exhibited elevated chloride, IDS, and
 4    specific conductance, which returned to background levels in under a week. Towanda Creek
 5    experienced these same elevations in concentration, but only at its confluence with the unnamed
 6    tributary; elevated chloride, IDS, and specific conductance returned to background levels the day
 7    after the blowout [SAIC and GES. 20111.

      7.7.4.  Roadway Transport of Produced Water
 8    Accidents during transportation of hydraulic fracturing produced water are a possible mechanism
 9    leading to potential impacts to drinking water. Nationwide data are not available, however, on the
10    number of such accidents that result in impacts. An estimate of releases from truck transport of
11    produced water could be made as follows:

                                     Total number of wells  x Produced water volume per well
        Total number of truckloads =	
                                                Produced water volume per truck

12    Then the total distance traveled by all trucks is given by:

              Total distance traveled = Total number of truckloads  x Distance per truck

13    The number of crashes impacting drinking water resources can be estimated from:

      Total crashes impacting drinking water resources
                    = Fraction of crashes releasing waste that impacts drinking water resources
                    x Fraction of all crashes releasing waste x Crashes per distance

14    Estimates of all but one of the quantities in these calculations can be made from various literature
15    sources, which are described in Appendix E. A key parameter is the number of crashes of trucks per
16    distance traveled. In 2012, the U.S. Department of Transportation (DOT) estimated thatthe number
17    of crashes per 100 million highway miles driven of a type of large truck was 110, which is a
18    relatively small number. A key parameter that is unknown is the number of crashes which impact
19    drinking water resources, so definitive estimates of impacts to drinking water resources cannot be
20    made. Alternatively, as an upper bound on drinking water resource impacts, the number of crashes
21    which release waste can be estimated. Based on various assumptions and scenarios presented in
22    Appendix E, the number of crashes with releases is bounded by the low tens of events. At 20 m3per
23    truckload, the volumes are low relative to the typical volume of produced water.

24    Several limitations are inherent in this analysis, including differing rural road accident rates and
25    highway rates, differing produced water endpoints, and differing amounts of produced water
26    transported. Further, the estimates present an upper bound on impacts, because  not all releases
27    would reach or impact drinking water resources.
                   This document is a draft for review purposes only and does not constitute Agency policy.
      June 2015                                   7-39                 DRAFT—DO NOT CITE OR QUOTE

-------
      Hydraulic Fracturing Drinking Water Assessment                        Chapter 7 - Flowback and Produced Water


      7.7.5.  Studies of Environmental Transport of Released Produced Water
 1    In this subsection, we describe transport study results that illustrate how produced waters have
 2    been shown to be transported from historical disposal practices and spills. Over the history of oil
 3    production in the U.S., produced water disposal methods have evolved from land application to
 4    storage in unlined ponds to deep well injection [Whittemore. 2007], although some unlined pits
 5    continue in use where allowed by states. The changes in practice occurred because of pollution
 6    impacts: first to surface waters and aquatic biota, and then to ground water from disposal ponds.
 7    Evaluation of sites contaminated by these historic practices sheds light on the potential for
 8    transport of released produced water, as discussed below.

 9    Impacts to ground water might occur following a spill on land. When the liquid is highly saline, its
10    migration is affected by its high density and viscosity compared with that of fresh water. When
11    spilled flowback or produced water flows over land, a fraction of the liquid is subject to infiltration.
12    The fraction depends on the rate of release, surface cover (i.e., pavement, cracked pavement,
13    vegetation, bare soil, etc.), slope of the land surface, subsurface permeability, and the moisture
14    content in the subsurface.

15    The potential for impacts from produced water spills depends on the distance from the source to
16    receptor; the distance depends on local topography. One study investigated receptor distances on a
17    formation basis [Entrekinetal., 2011]. The distance between gas wells and drainage ways was
18    determined to average 273 m (890 ft] for the Marcellus Shale and 353  m (1160  ft] for the
19    Fayetteville Shale (Entrekinetal., 2011]. Some wells were much closer, being as close as 1 m (3.28
20    ft]. For one location in each formation a separate analysis gave a mean estimate of 153 m (500 ft]
21    for the Marcellus Shale and 130m (430 ft] for the Fayetteville Shale. The average distance to public
22    drinking water intakes was 15 km (9.32 mi). The average distance to public water supply wells was
23    37 km (23.0 mi) for the Marcellus Shale and 123 km (76.4 mi) for the Fayetteville Shale. As the
24    density of gas development increases the number of gas wells located close to drainage ways and
25    public water supply wells may also increase.

26    For example, Whittemore (2007) described a site with relatively little infiltration due to moderate
27    to low permeability of silty clay soil and low permeability of underlying shale units. Thus, most of
28    the historically surface-disposed produced water at the site flowed into surface drainages.
29    Observed historic levels of chloride in receiving waters resulted from the relative balance of
30    produced water releases and precipitation runoff, with high concentrations corresponding to low
31    stream flows. Persistent surface water chloride contamination was attributed to slow flushing and
3 2    discharge of contaminated ground water.

33    Transport from the land surface to the water table is further characterized in general by flow
34    through variably water-saturated media, preferential flow paths, fractures in clays, and
3 5    macropores. Preferential flow paths along microscale heterogeneities are known to exist and
3 6    dominate transport even after cycles of repeated drying and rewetting. The effect of flowback on
3 7    transport of colloids has recently been evaluated in laboratory sand columns. The authors found
3 8    that flowback increased the mobility of colloidal particles, which potentially serve as a source of
39    aquifer contamination (Sangetal., 2014].

                    This document is a draft for review purposes only and does not constitute Agency policy.
      June 2015                                    7-40                   DRAFT—DO NOT CITE OR QUOTE

-------
      Hydraulic Fracturing Drinking Water Assessment                        Chapter 7 - Flowback and Produced Water


 1    In another study, Ottonetal. [2007] reported on a site in Oklahoma where two abandoned pits
 2    were major sources for releases of produced water and oil. Saline water from the pits flowed
 3    through thin soils and readily percolated into underlying permeable bedrock. Deeper, less
 4    permeable bedrock was contaminated by salt water later in the history of the site, presumably due
 5    to fractures [Otton et al., 2007]. The mechanisms proposed were lateral movement through
 6    permeable sand bodies, vertical movement along shale fractures,  and possibly increased
 7    permeability from clay  flocculation and shrinkage due to the presence of highly saline water [Otton
 8    etal.. 20071

 9    Because it is denser than freshwater, saline produced water can migrate downward through
10    aquifers. Whittemore [2007] reported finding oilfield brine with a chloride concentration of 32,900
11    mg/L at the base of the  High Plains aquifer. Where aquifers discharge to streams, saline stream
12    water has been reported, although at reduced concentrations [Whittemore. 2007]. likely due to
13    diffusion within the aquifer and mixing with stream water. The stream flow rate, in part,
14    determines mixing of substances in surface waters. High flows are related to  lower chemical
15    concentrations, and vice versa, as demonstrated for bromide by States etal. [2013] for the
16    Allegheny River.

17    Generally, the deeper that brine can move into an aquifer, as impacted by the volume and timing of
18    the release, the longer the duration of contamination [Whittemore. 2007]. Kharakaetal. [2007]
19    reported on studies at a site in Oklahoma with one abandoned and two active unlined brine pits.
20    Produced water from these pits penetrated 3- to 7-m thick shale and siltstone units, creating three
21    plumes of high-salinity  water (5,000 to 30,000 mg/L TDS). The impact of these plumes on the
22    receiving water body (Skiatook Lake] was judged to be minimal, although the estimate was based
23    on a number of notably uncertain transport quantities [Ottonetal., 2007].

24    Chloride impacts from produced water spills were studied through scenario modeling releases of
25    100 bbl (4,200 gal or 15,900 L] and 10,000 bbl (420,000 gal or 1.59 million L] (API. 2005]. The
26    scenarios included transport through a homogeneous or heterogeneous unsaturated zone using the
27    HYDRUS-1D model (Simuneketal.. 1998] and mixing within the top portion of a shallow aquifer
28    using a specially developed spreadsheet model. The results of the scenario modeling indicated that
29    ground water quality is unlikely to be impaired for spills with small soil penetration depths, which
30    correspond to spills distributed over large areas. Large spills of 100,000 bbl (4.2 million gal or 15.9
31    million L] over sandy unsaturated zones were found to have a high potential  to impact ground
32    water quality (API. 2005]. Spills of less than 100 bbl (420 gal or 1,590 L] were not modeled and
33    were presumed to have low impacts based on the results from the larger spills. The results were
34    constrained by the underlying assumptions of HYDRUS-1D—that there were no preferential flow
35    paths, including fractured systems, systems with macropores, or fine scale heterogeneities. More
36    rapid and spatially extensive transport could occur in these settings.

      7.7.6. Coalbed Methane
37    A CBM produced water impoundment in the Powder River Basin of Wyoming was studied for  its
38    impact on ground water (Healy etal.. 2011: Healy etal.. 2008]. Infiltration of water from the
39    impoundment was found to create a perched water mound in the unsaturated zone above bedrock

                   This document is a draft for review purposes only and does not constitute Agency policy.
      June 2015                                   7-41                  DRAFT—DO NOT CITE OR QUOTE

-------
      Hydraulic Fracturing Drinking Water Assessment                       Chapter 7 - Flowback and Produced Water


 1    in a location with historically little recharge. The subsurface sediments were found to be highly
 2    heterogeneous both physically and chemically, which increased the complexity of studying the site.
 3    Elevated concentrations of IDS, chloride, nitrate, and selenium were found at the site. For example,
 4    IDS exceeded 100,000 mg/L in one lysimeter sample, while the concentration was 2,275 mg/L in a
 5    composite CBM produced water sample [Healy etal.. 2008]. Most of the solutes found in the ground
 6    water mound did not originate with the CBM produced water, but rather were the consequence of
 7    dissolution of previously existing salts and minerals. The mechanisms were thought to be gypsum
 8    dissolution, cation exchange, and pyrite dissolution. Data from other sites indicated that the study
 9    site's ground water chemistry may not be typical and that the same phenomena may not occur at
10    other sites in the basin [Healy etal., 2011].

      7.7.7.  Transport Properties
11    The identified constituents of flowback and produced water include inorganic chemicals in the form
12    of cations and anions (including various types of metals, metalloids, and non-metals, and
13    radioactive materials, among others] and organic chemicals, including identified compounds in
14    various classes, and unidentified materials measured as TOG and DOC. Environmental transport of
15    these chemicals depends on properties of the chemical and properties of the environment, and is
16    extensively discussed in Section 5.8.3. In this section we discuss the characteristics of transport for
17    inorganics and note that some inorganics may move with the water, while many of the others are
18    influenced by reactions. For organic chemicals identified in produced water, we discuss EPI Suite™
19    estimates of the main transport parameters identified in Chapter 5, while noting the influences of
2 0    salinity and temperature on these properties.

21    Transport of inorganic chemicals depends on the nature of ground water and vadose zone flow, and
22    potential reactions among the inorganic chemical, solid surfaces, and geochemistry of the water.
23    Some inorganic anions (i.e., chloride and bromide] move with their carrier liquid and are mostly
24    impacted by physical transport mechanisms: flow of water and dispersion. In addition to the flow-
25    related processes, transport of most inorganics is driven by three mechanisms related to
26    partitioning to the solid phase: adsorption, absorption, and precipitation. The effects of these
27    mechanisms depend on both chemical and environmental characteristics, including the surface
28    reactivity, solubility, and redox sensitivity of the contaminant;  and the type and abundance of
29    reactive mineral phases, and the ground-water chemistry (U.S. EPA. 2007]. Through the use of
30    transport models, the effects of physical transport mechanisms and chemical processes are
31    integrated. Examples of transport models for reactive metals include the Geochemist's Workbench
32    (Bethke. 2014] and Hydrus (Simuneketal.. 1998].

3 3    Properties of organic chemicals which tend to affect the likelihood that a chemical will reach and
34    impact drinking water resources if spilled include high chemical mobility in water, low volatility,
35    and high persistence in water. Using the EPA chemical database EPI Suite™, we were able to obtain
36    actual or estimated physicochemical properties for 86 of the 134 organic chemicals identified in
37    produced water and listed in Appendix A. A portion of these, 66, are used in the chemical mixing
38    stage (see Appendix Table C-8]. EPI Suite™ results were generated for solubility, octanol water
39    partition coefficient (Kow), and the Henry's constant (see Figure 7-9]. The log Kow values are of the
                   This document is a draft for review purposes only and does not constitute Agency policy.
      June 2015                                    7-42                  DRAFT—DO NOT CITE OR QUOTE

-------
      Hydraulic Fracturing Drinking Water Assessment
       Chapter 7 - Flowback and Produced Water
 1    identified organic chemicals skewed positively, indicating some of the chemicals have low mobility
 2    which may result in less extensive dissolved contaminant plumes in ground water. These
 3    compounds, however, have a higher tendency to sorb to particulate or colloidal materials and hence
 4    be transported in systems where particle transport is dominant, for example: colloid transport in
 5    ground water and sediment transport in surface water. The log Henry constant values tend to be
 6    below 0 indicating that at equilibrium the concentration in air is less than the concentration in
 7    water. This behavior is reflected in the log solubility plot, as the solubilities skew conversely toward
 8    high values.

 9    The EPI Suite™ results are constrained by their applicability to one temperature (25 °C), and
10    salinity (low). Temperature changes impact Henry's constant, Kow, and solubility, and depend on the
11    characteristics of the chemical and ions present (Borrirukwisitsaketal., 2012: Schwarzenbach et
12    al.. 2002). In some cases the effect changes exponentially with salinity (Schwarzenbach etal.. 2002).
13    Therefore, property values that depart from the EPI Suite™ values are expected for produced water
14    at elevated temperature and salinity.
             -10  -8  -6  -4  -2  0   2   4  6   8  10
                        Measured log K,,,, [L/kg]
                                                        20     More Mobile
                                                        15
                                                      ii
                                                      e
                                                      T 10
                                                         0  L
-10  -8  -6   -4  -2   0  2   4  6   8  10
           Estimate log Kow [L/kg]
                                                       20
                                                       10
                                                     I 5
           -10 -9-8-7-6-5-4-3-2-10 1 2 3  4  5 6 7 8  9 10
                    Estimated log Solubility [mg/L <§> 25C]
                                                         -10 -9-8-7-6-5-4-3-2-10 1  2  3  4 5 6 7  8  9 10
                                                              Estimated log Henry's Law Constant [atm m* mole ' @ 25C]
      Figure 7-9. Histograms of physicochemical properties of 86 organic chemicals identified in
                produced water (physicochemical properties estimated by EPI Suite™).

      7.8.   Synthesis
15    After hydraulic fracturing is completed, the operator reduces injection pressure and water is
16    allowed to flow back from the well to prepare for oil or gas production. The flowback water may
17    contain fracturing fluid, fluid from the surrounding formation, and hydrocarbons. Initially this
18    flowback is mostly fracturing fluid, but as time passes, the produced water becomes more similar to
19    the water in the formation. This water is stored at the surface for eventual reuse or disposal.
                    This document is a draft for review purposes only and does not constitute Agency policy.
      June 2015                                     7-43                  DRAFT—DO NOT CITE OR QUOTE

-------
      Hydraulic Fracturing Drinking Water Assessment                        Chapter 7 - Flowback and Produced Water


 1    Impacts to drinking water from flowback and produced water can occur if spilled flowback or
 2    produced water enters surface water bodies or aquifers.

      7.8.1.  Summary of Findings
 3    The volume and composition of flowback and produced water vary geographically, both within and
 4    between different formations with time and with other site-specific factors. High initial rates of
 5    flowback decrease as time continues. The amount of fracturing fluid returned to the surface varies,
 6    and typically averages 10% to 25%. In most cases, lower flow rates of produced water continue
 7    throughout gas production.1 In some formations (i.e., the Barnett Shale), the ultimate volume of
 8    produced water can exceed the volume of hydraulic fracturing fluid because of inflow of water.

 9    The composition of flowback changes with time as the hydraulic fracturing fluid contacts the
10    formation and mixes with the formation water. At the same time, reactions occur between the
11    constituents of the fracturing fluid and the formation. Although varying within and between
12    formations, shale and tight gas produced water typically contains high levels of TDS (salinity) and
13    associated ionic constituents (bromide, calcium, chloride, iron, potassium, manganese, and sodium).
14    Produced water may contain toxic materials, including barium, cadmium, chromium, lead, mercury,
15    nitrate, selenium, and BTEX. CBM produced water can have lower levels of salinity if its coal source
16    was deposited under freshwater conditions.

17    Flowback and produced water spills are known to have occurred across the country. The causes
18    identified for these spills  are container and equipment failures, human error, well  communication,
19    blowouts, pipeline leaks,  and illegal dumping. Spills due to trucking accidents are possible, but
20    accident rates in the United States suggest only a small number of such releases occur.

21    USGS studies of impacts of produced water disposal in unlined pits document the potential for
22    surface releases (in these cases over multiple years) that have led to ground water impacts.
23    Contaminant plumes can  be traced to high TDS water disposed of in the pits, or geochemical
24    reaction between infiltrating low-TDS water, in the case of CBM produced water, releasing existing
25    minerals from the unsaturated zone to ground water.

      7.8.2.  Factors Affecting the Frequency or Severity of Impacts
2 6    The potential of spills of flowback and produced water to affect drinking water resources depends
27    upon the release volume, duration, and composition. Larger spills of greater duration  are more
28    likely to reach a nearby drinking water resource than are smaller spills. The composition of the
29    spilled fluid will also impact the severity of a spill, as certain constituents are more likely to affect
30    the quality of a drinking water resource. Low-volume and short-duration spills are less likely to
31    cause impacts, (see Section 7.7.5).

32    Potential impacts to water resources from hydraulic fracturing-related spills are expected to be
3 3    affected by watershed and water body characteristics. For example, overland flow is affected by
34    surface topography and surface cover. Infiltration of spilled produced water reduces the amount of
      1 Note that increasing produced water flow rates are indicative of water breakthrough and declining oil production.


                   This document is a draft for review purposes only and does not constitute Agency policy.
      June 2015                                    7-44                  DRAFT—DO NOT CITE OR QUOTE

-------
      Hydraulic Fracturing Drinking Water Assessment                       Chapter 7 - Flowback and Produced Water


 1    water threatening surface water bodies. However, infiltration through soil may lead to ground
 2    water impacts. Produced water with lower constituent concentrations may produce lesser impacts,
 3    but the USGS studies of CBM produced water impoundments described in section 7.7.5 showed
 4    impacts caused by CBM water mobilizing existing minerals. The USGS studies on historical disposal
 5    of saline produced water in unlined pits detected ground water plumes discharging into surface
 6    water bodies. The potential impact to drinking water in such cases depends on the location of
 7    drinking water wells and the size of any surface water supply reservoir.

      7.8.3.  Uncertainties
 8    We first discuss data gaps in our overall knowledge of flowback and produced water; closing such
 9    gaps would enable us to better predict impacts on drinking water resources. Second, we present
10    uncertainties that, based on site-specific conditions, also affect our ability to determine potential
11    impacts.

      7.8.3.1.  General Data Gaps
12    Knowledge of volume and some compositional aspects of flowback and produced water are known
13    from published sources. Most of the available data on TENORM has come from the Marcellus Shale,
14    where these are typically high in comparison to the limited data available from other formations.
15    Only a few studies have attempted to sample and characterize the organic fraction of flowback and
16    produced water; some data are available for shale and CBM, but none are available for tight
17    formations. The reported organic chemical data from flowback likely does not capture the full range
18    of chemicals that may be present, either as original components of the hydraulic fracturing solution
19    or transformation products generated in the subsurface.

20    Characterization of produced water organics is limited by several factors. Development or use of
21    proper analytical procedures depends upon knowing the identities of injected chemicals. Because
22    the formulation of hydraulic fracturing fluids can contain proprietary chemicals, the exact
23    formulations are not available. In addition, subsurface transformations yield degradation products,
24    which themselves must have appropriate analytical methods. Further difficulties are due to matrix
25    interference from high-TDS produced water. These problems result in the need to develop new
26    methods for analyzing both organics and inorganics in produced water.

27    Nationwide data on spills of flowback and produced water are limited in two primary ways: the
28    completeness of reported data cannot be determined, and individual states' reporting requirements
29    differ [U.S. EPA, 2015n]. Despite various studies, the total number of spills occurring in the United
30    States, their release volumes and associated concentrations, can only be roughly estimated because
31    of underlying data limitations.

      7.8.3.2.  Uncertainties at Individual Spill Sites
3 2    Spills of flowback and produced water present many uncertainties that, in combination, limit our
3 3    ability to predict impacts on drinking water resources. Spills vary in volume, duration, and
34    composition. The spilled liquid could be fracturing fluid mixed with formation water in a
35    proportion that depends on the time that has passed since fracturing. As described in Section 7.7,
36    spills may originate from blowouts, well communication, aboveground or underground pipeline

                   This document is a draft for review purposes only and does not constitute Agency policy.
      June 2015                                    7-45                  DRAFT—DO NOT CITE OR QUOTE

-------
      Hydraulic Fracturing Drinking Water Assessment                       Chapter 7 - Flowback and Produced Water


 1    breaks, leaking impoundments, failed containers, human error (including illegal activities, failure to
 2    detect spills, and failure to report spills) or unknown causes. The difference between these causes
 3    affect the size and location of the spill. In addition, the factors governing transport of spilled fluid to
 4    a potential receptor vary by site: the presence and quality of secondary or emergency containment
 5    and spill response; the rate of overland flow and infiltration; the distance to a surface water body or
 6    drinking water well; and transport and fate processes. Impacts to drinking water resources from
 7    spills of flowback and produced water depend on environmental transport parameters, which can
 8    in principle be determined but are unlikely to be known or adequately specified in advance of a
 9    spill.

10    Because some constituents of flowback and produced water are constituents of natural waters (e.g.,
11    bromide in coastal surface waters) or can be released into the environment by other pollution
12    events (e.g., benzene from gasoline releases, or bromide from coal mine drainage), baseline
13    sampling prior to impacts is one way to increase certainty of an impact determination. Further
14    sampling and investigation may be used to develop the linkage between a release and a
15    documented drinking water impact Produced water spill response typically includes delineation of
16    the extent of oiled soils, sheens on water surfaces, and the extent of saline water. Extensive
17    characterization of produced water is typically not part of spill response, and therefore the
18    chemicals, and their concentrations, potentially impacting drinking water resources are not usually
19    known.

      7.8.4.  Conclusions
2 0    Flowback and produced water has the potential to affect the quality of drinking water resources if it
21    enters into a surface or ground water body used as a drinking water resource. This can occur
2 2    through spills at well pads, or during transport of flowback. Specific impacts depend upon the spill
23    itself, the environmental conditions surrounding the spill, water body and watershed
24    characteristics, and the composition of the spilled fluid. Flowback and produced water may contain
25    toxic constituents and can potentially render water unpalatable or unsafe to drink. Conclusive
2 6    determination of impacts  to water resources depends on commitment of resources to the
2 7    implementation of sampling, analysis and evaluation strategies

      Text Box 7-1. Research Questions Revisited.

2 8    What is currently known about the frequency, severity, and causes of spills of flowback and produced
29    water?

30    •  Surface spills of flowback and produced water from unconventional oil and gas production have occurred
31       across the country. Some produced water spills have affected drinking water resources, including a few
32       private drinking water wells. The majority of flowback and produced water spills are under 1,000
33       gallons. The causes identified for these are container and equipment failures, human error, well
34       communication, blowouts, pipeline leaks, and illegal dumping.

3 5    What is the composition of hydraulic fracturing flowback and produced water, and what factors might
3 6    influence this composition ?
                   This document is a draft for review purposes only and does not constitute Agency policy.
      June 2015                                    7-46                  DRAFT—DO NOT CITE OR QUOTE

-------
      Hydraulic Fracturing Drinking Water Assessment                          Chapter 7 - Flowback and Produced Water
 1    •   The composition of flowback and produced water must be determined through sampling and analysis,
 2        both of which have limitations. The former due to the ability to access through production equipment and
 3        the latter due to issues with identifying target analytes in advance of analysis and the existence of
 4        appropriate analytical methods.
 5    •   The geochemical content of water flowing back initially reflects injected fluids. After initial flowback,
 6        returning fluid geochemistry shifts to reflect the geochemistry of formation waters and formation solids.
 7    •   According to available literature and data, conventional and unconventional flowback and produced
 8        water content are often similar with respect to the occurrence and concentration of many constituents.
 9    •   The least statistical variability in produced water content is exhibited between shale gas and tight gas
10        produced water, and the most statistical variability is exhibited between shale gas and coalbed methane
11        produced water.
12    •   Much produced water is generally characterized as saline (with the exception of most coalbed methane
13        produced water) and enriched in major anions, cations, metals, naturally occurring radionuclides, and
14        organics.
15    •   Shale and coalbed produced water is enriched in benzene. Benzene is  a constituent of concern in
16        Marcellus Shale, Raton CBM, and San Juan CBM produced water. Shale produced water is more likely to
17        contain elevated average total BTEX levels than other unconventional produced water.
18    •   Typically, unconventional produced water contains low levels of heavy metals. Elevated strontium and
19        barium levels, however, are characteristic of Marcellus Shale flowback and produced water. CBM and, in
20        particular, shale produced water are likely to contain NORM levels of concern.
21    •   Composition data were limited. Most of the available data on produced water content were for shale
22        formations  and CBM basins, while little data were available for sandstone formations. Additionally, the
23        majority of data were for inorganics, and little data were available for organics. Many more organic
24        chemicals have been reported to have been used in hydraulic fracturing fluid than have been identified in
25        produced water. The difference may be due to analytical limitations, limited study scopes, and
26        undocumented subsurface reactions.
27    •   Hydraulic fracturing flowback and produced water composition is influenced by the composition of
28        injected hydraulic fracturing fluids, the targeted geological formation and associated hydrocarbon
29        products, the stratigraphic environment, and subsurface processes and residence time.
30    •   Spatial variability of produced water content occurs between plays of different rock sources (e.g., coal vs.
31        sandstone), between plays of the same rock type (e.g., Barnett Shale vs. Bakken Shale), and within
32        formations  of the same source rock (e.g., northeastern vs. southwestern Marcellus Shale).

3 3    What are the chemical and physical properties of hydraulic fracturing flowback and produced water
34    constituents?

35    •   The identified constituents of flowback and produced water include inorganic chemicals  (cations and
36        anions in the form of metals, metalloids, non-metals, and radioactive materials), organic chemicals and
37        compounds, and unidentified materials measured as TOG (total organic carbon) and DOC (dissolved
38        organic carbon). Some constituents are readily transported with water (i.e., chloride and bromide), while
39        others depend strongly on the geochemical conditions in the receiving water body (i.e., radium and
40        barium), and assessment of their transport is based on site-specific factors. Using the EPA chemical
                     This document is a draft for review purposes only and does not constitute Agency policy.
      June 2015                                       7-47                    DRAFT—DO NOT CITE OR QUOTE

-------
      Hydraulic Fracturing Drinking Water Assessment                         Chapter 7 - Flowback and Produced Water


 1        database EPI Suite, we were able to obtain actual or estimated physicochemical properties for 86 (64%)
 2        of the 134 chemicals identified in produced water.

 3    •   As in the case of chemicals in hydraulic fracturing fluid, chemical properties that affect the likelihood of
 4        an organic chemical in produced water reaching and impact drinking water resources in the short-term,
 5        include: high chemical mobility in water, low volatility, and high persistence in water. In general, EPI
 6        suite results suggest that organic chemicals in produced water tend toward lower mobility in water.
 7        Consequently these chemicals could remain in soils or sediments nearby spills. Low mobility may result
 8        in smaller dissolved contaminant plumes in ground water. Although these compounds are more likely to
 9        be transported associated with sediments in surface water or small particles in ground water. Organic
10        chemical properties vary with salinity and the effects depend on the nature of the chemical.

11    If spills occur, how might hydraulic fracturing flowback and produced water contaminate drinking
12    water resources?

13    •   Spills of flowback or produced water might impact drinking water resources if the spill or release is of
14        sufficient volume and duration, to reach the resource at a sufficient concentration. Impacts to in-use
15        drinking water depend on proximity to sources. Scientific literature and published reports have shown
16        that produced water spills have impacted drinking water resources.
      7.9.    References for Chapter 7

      Ali. SA: Clark. Wl: Moore. WR: Dribus. IR. (2010). Diagenesis and reservoir quality. Oilfield Rev 22:14-27.

      Alley, B: Beebe, A: Rodgers, I: Castle, IW. (2011). Chemical and physical characterization of produced waters
         from conventional and unconventional fossil fuel resources. Chemosphere 85: 74-82.
         http://dx.doi.0rg/10.1016/j.chemosphere.2011.05.043

      API (American Petroleum Institute). (2005). Modeling study of produced water release scenarios.
         (Publication Number 4734). Washington, D.C.: API Publishing Services.
         http://www.api.Org/~/media/files/ehs/environmental performance/4734.pdf?la=en

      API (American Petroleum Institute). (2010b). Water management associated with hydraulic fracturing.
         Washington, D.C.: API Publishing Services.
         http://www.api.Org/~/media/Files/Policy/Exploration/HF2 el.pdf

      Arkadakskiy. S..: Rostron. Bl. (2012a). Stable isotope geochemistry helps in reducing out-of-zone hydraulic
         fracturing and unwanted brine production from the Bakken Reservoir. Available online at
         http://isobrine.com/wp-content/uploads/2013/04/Arkadakskiy Rostron 20121.pdf

      Arkadakskiy. S: Rostron. B. (2013a). Tracking out-of-zone hydraulic fracturing in the Bakken with naturally
         occuring tracers. Paper presented at GeoConvention 2013: Integration Canadian Society of Petroleum
         Geologists, May 6-10, 2013, Calgary, AB, Canada.

      Arthur. ID: Bohm. B: Coughlin. Bl: Layne. M. (2009b). Evaluating the environmental implications of hydraulic
         fracturing in shale gas reservoirs. Paper presented at SPE Americas E&P Environmental and Safety
         Conference, March 23-25,2009, San Antonio, TX.

      ATS PR (Agency for Toxic Substances and Disease Registry). (2005). Public health assessment guidance
         manual (Update). Atlanta, GA: U.S. Department of Health and Human Services, Public Health Service.
         http://www.atsdr.cdc.gov/hac/PHAManual/toc.html
                     This document is a draft for review purposes only and does not constitute Agency policy.

      June 2015                                       7-48                   DRAFT—DO NOT CITE OR QUOTE

-------
Hydraulic Fracturing Drinking Water Assessment                          Chapter 7 - Flowback and Produced Water
ATSDR [Agency for Toxic Substances and Disease Registry). (2013). Health consultation: Chesapeake ATGAS
   2H well site, Leroy Hill Road, Leroy, Leroy Township, Bradford County, PA. Atlanta, GA: ATSDR, Division of
   Cummunity Health Investigations.
   http://www.atsdr.cdc.gov/HAC/pha/ChesapeakeATGASWellSite/ChesapeakeATGASWellSiteHC1028201
   3 508.pdf

AWWA (American Water Works Association). (2013). Water and hydraulic fracturing: A white paper from the
   American Water Works Association. Denver, CO.
   http://www.awwa.Org/Portals/0/files/legreg/documents/AWWAFrackingReport.pdf

Barbot. E: Vidic. NS: Gregory. KB: Vidic. RD. (2013). Spatial and temporal correlation of water quality
   parameters of produced waters from Devonian-age shale following hydraulic fracturing. Environ Sci
   Technol 47: 2562-2569.

Barnes. T. (2010). 2 drillers fined for Pennsylvania gas well blowout. Available online at http://www.post-
   gazette.com/news/environment/2010/07/14/2-drillers-fined-for-Pennsylvania-gas-well-
   blowout/stories/201007140241 (accessed March 3,2015).

Benko, KL: Drewes, IE. (2008). Produced water in the Western United States: Geographical distribution,
   occurrence, and composition. Environ Eng Sci 25: 239-246.

Bethke. CM.: Yeakel. S. (2014). The geochemists workbench. Release 10.0. GWB essentials guide (Version
   Release 10.0). Champaign, II: Aqueous Solutions, LLC. Retrieved from
   http://www.gwb.com/pdf/GWB10/GWBessentials.pdf

Bibby, Kl: Brantley, SL: Reible, DP: Linden, KG: Mouser, PI: Gregory, KB: Ellis, BR: Vidic, RD. (2013). Suggested
   reporting parameters for investigations of wastewater from unconventional shale gas extraction. Environ
   Sci Technol 47:13220-13221. http://dx.doi.org/10.1021/es404960z

Blauch. ME: Myers. RR: Moore. TR: Lipinski. BA. (2009). Marcellus shale post-frac flowback waters - where is
   all the salt coming from and what are the implications? In Proceedings of the SPE Eastern Regional
   Meeting. Richardson, TX: Society of Petroleum Engineers.

Blondes. MS: Cans. KD: Thordsen. II: Reidy. ME: Thomas. B: Engle. MA: Kharaka. YK: Rowan. EL. (2014). Data:
   U.S. Geological Survey National Produced Waters Geochemical Database v2.0 (Provisional) [Database]:
   U.S. Geological Survey:: USGS. Retrieved from
   http://energy.usgs.gov/EnvironmentalAspects/EnvironmentalAspectsofEnergyProductionandUse/Produ
   cedWaters.aspx#3822349-data

Borrirukwisitsak. S: Keenan. HE: Gauchotte-Lindsay. C. (2012). Effects of salinity, pH and temperature on the
   octanol-water partition coefficient of bisphenol A. IJESD 3: 460-464.
   http://dx.doi.org/10.7763/IIESD.2012.V3.267

Boschee.  P. (2014). Produced and flowback water recycling and reuse:  Economics, limitations, and
   technology. Oil and Gas Facilities 3:16-22.

Bouska. V. (1981). Geochemistry of coal. New York, NY: Elsevier.

Boyer. EW: Swistock. BR: Clark. I: Madden. M: Rizzo. DE. (2011). The impact of Marcellus gas drilling on rural
   drinking water supplies. Harrisburg, PA: The Center for Rural Pennsylvania.
   http://www.iogawv.com/Resources/Docs/Marcellus-drinking-water-2011.pdf

Brantley, SL: Yoxtheimer, D: Arjmand, S: Grieve, P: Vidic, R: Pollak, I: Llewellyn, GT: Abad, I: Simon, C. (2014).
   Water resource impacts during unconventional shale gas development:  The Pennsylvania experience. Int J
   Coal Geol 126:140-156. http://dx.doi.Org/10.1016/i.coal.2013.12.017

Breit. GN. (2002). USGS Produced waters database. Available online at
   http://energy.cr.usgs.gov/prov/prodwat/

Brinck. EL: Frost. CD. (2007). Detecting infiltration and impacts of introduced water using strontium isotopes.
   Ground Water 45: 554-568. http://dx.doi.Org/10.llll/i.1745-6584.2007.00345.x
               This document is a draft for review purposes only and does not constitute Agency policy.

June 2015                                       7-49                    DRAFT—DO NOT CITE OR QUOTE

-------
Hydraulic Fracturing Drinking Water Assessment                         Chapter 7 - Flowback and Produced Water
Byrnes. AP. (2011). Role of induced and natural imbibition in frac fluid transport and fate in gas shales.
   Presentation presented at Technical Workshops for Hydraulic Fracturing Study: Fate & Transport, March
   28-29,2011, Arlington, VA.

Caniglia. I. (2014). Youngstown contractor sentenced to 28 months for dumping fracking waste. Available
   online at http://www.cleveland.com/court-
   justice/index.ssf/2014/08/youngstown contractor sentence.html (accessed March 12, 2015).

Chapman, EC: Capo, RC: Stewart, BW: Kirby, CS: Hammack, RW: Schroeder, KT: Edenborn, HM. (2012).
   Geochemical and strontium isotope characterization of produced waters from Marcellus Shale natural gas
   extraction. Environ Sci Technol 46: 3545-3553.

Clark. CE: Horner. RM: Harto. CB. (2013). Life Cycle Water Consumption for Shale Gas and Conventional
   Natural Gas. Environ Sci Technol 47:11829-11836. http://dx.doi.org/10.1021/es4013855

Cluff. M: Hartsock. A: Macrae. I: Carter. K: Mouser. PI. (2014). Temporal changes in microbial ecology and
   geochemistry in produced water from hydraulically fractured Marcellus Shale Gas Wells. Environ Sci
   Technol 48: 6508-6517. http://dx.doi.org/10.1021/es501173p

Coleman. N. (2011). Produced formation water sample results from shale plays. Presentation presented at
   Technical Workshops for the Hydraulic Fracturing Study: Chemical & Analytical Methods, February 2425,
   2011, Arlington, VA.

Considine. T: Watson. R: Considine. N: and Martin.  I. (2012). Environmental impacts during Marcellus shale
   gas drilling: Causes, impacts, and remedies. (Report 2012-1). Buffalo, NY: Shale Resources and Society
   Institute. http://cce.cornell.edu/EnergyClimateChange/NaturalGasDev/Documents/UBSRSI-
   Environmental%20Impact%20Report%202012.pdf

Dahm, KG: Guerra, KL: Xu, P: Drewes, IE. (2011). Composite geochemical database for coalbed methane
   produced water quality in the Rocky Mountain  region. Environ Sci Technol 45: 7655-7663.
   http://dx.doi.org/10.1021/es201021n

Pake. LP. (1978). Fundamentals of reservoir engineering. Boston, MA: Elsevier.
   http://www.ing.unp.edu.ar/asignaturas/reservorios/Fundamentals%20of%20Reservoir%20Engineering
   %20%28LP.%20Dake%29.pdf

Davis. IP: Struchtemeyer. CG: Elshahed. MS. (2012). Bacterial communities associated with production
   facilities of two newly drilled thermogenic natural gas wells in the Barnett Shale (Texas, USA). Microb Ecol
   64: 942-954. http://dx.doi.org/10.1007/s00248-012-0073-3

DeArmond. PD: DiGoregorio. AL. (2013a). Characterization of liquid chromatography-tandem mass
   spectrometry method for the determination of acrylamide in complex environmental samples. Anal
   Bioanal Chem 405: 4159-4166. http://dx.doi.org/10.1007/s00216-013-6822-4

DeArmond. PD: DiGoregorio. AL. (2013b). Rapid liquid chromatography-tandem mass spectrometry-based
   method for the analysis of alcohol ethoxylates and alkylphenol ethoxylates in environmental samples. J
   Chromatogr A1305:154-163. http://dx.doi.Org/10.1016/j.chroma.2013.07.017

Demorest. PL: Wallace. ES. (1992). Radiochemical  determination of norm in produced water utilizing wet
   chemistry separation followed by radiochemical analysis. In JP Ray; Engelhardt; Fr (Eds.), Produced
   water: Technological/environmental issues and solutions (pp. 257-266). New York, NY: Plenum Press.
   http://dx.doi.org/10.1007/978-l-4615-2902-6 21

DOE (U.S. Department of Energy). (2004). A white paper describing produced water from production of crude
   oil, natural gas, and coal bed methane. Lemont,  IL: Argonne National Laboratory.
   http://seca.doe.gov/technologies/oil-gas/publications/oil pubs/prodwaterpaper.pdf

DOE (U.S. Department of Energy). (2011a). A comparative study of the Mississippian Barnett shale, Fort
   Worth basin, and Devonian Marcellus shale, Appalachian basin. (DOE/NETL-2011/1478).
   http://www.netl.doe.gov/technologies/oil-gas/publications/brochures/DOE-NETL-2011-
   1478%20Marcellus-Barnett.pdf


               This document is a draft for review purposes only and does not constitute Agency policy.
June 2015                                      7-50                   DRAFT—DO NOT CITE OR QUOTE

-------
Hydraulic Fracturing Drinking Water Assessment                         Chapter 7 - Flowback and Produced Water
Dresel. PE: Rose. AW. (2010). Chemistry and origin of oil and gas well brines in western Pennsylvania (pp.
   48). (Open-File Report OFOG 1001.0). Harrisburg, PA: Pennsylvania Geological Survey, 4th ser.
   http://www.marcellus.psu.edu/resources/PDFs/brines.pdf

EERC (Energy and Environmental Research Center, University of North Dakota). (2013). BakkenSmart: water
   [Fact Sheet]. Grand Forks, ND: Energy and Environmental Research Center.
   http://www.undeerc.org/bakken/pdfs/NDIC-NDPC-Water-Fact-Sheet.pdf

Eiceman, GA. (1986). Hazardous organic wastes from natural gas production, processing and distribution:
   Environmental fates. (WRRI report, no. 227). New Mexico: Water Resources Research Institute.
   http://wrri.nmsu.edu/publish/techrpt/abstracts/abs227.html

Engelder. T. (2012). Capillary tension and imbibition sequester frack fluid in Marcellus gas shale [Letter].
   PNAS 109: E3625; author reply E3626. http://dx.doi.org/10.1073/pnas.1216133110

Engle. MA: Rowan. EL. (2014). Geochemical evolution of produced waters from hydraulic fracturing of the
   Marcellus Shale, northern Appalachian Basin: A multivariate compositional data analysis approach.  Int J
   Coal Geol 126: 45-56. http://dx.doi.Org/10.1016/i.coal.2013.ll.010

Entrekin. S: Evans-White. M: lohnson. B: Hagenbuch. E. (2011). Rapid expansion of natural gas development
   poses a threat to surface waters. Front Ecol Environ 9: 503-511. http://dx.doi.org/10.1890/110053

ESN Rocky Mountain (Environmental Services Network Rocky Mountain). (2003). Produced gas and water
   testing of CBM gas wells in the Raton Basin. Golden, CO.
   https://cogcc.state.co.us/Library/RatonBasin/ESN%20Final Reportpdf

Farag. AM: Harper. DP. (2014). A review of environmental impacts of salts from produced waters on aquatic
   resources. Int J Coal Geol 126:157-161. http://dx.doi.Org/10.1016/j.coal.2013.12.006

Fertl. WH: Chilingar. GV. (1988). Total organic carbon content determined from well logs. SPE Formation
   Evaluation 3: 407-419. http://dx.doi.org/10.2118/15612-PA

Fisher. IB: Sublette. KL. (2005). Environmental releases from exploration and production operations in
   Oklahoma: Type, volume, causes, and prevention. Environmental Geosciences 12: 89-99.
   http://dx.doi.org/10.1306/eg.11160404039

Fisher. RS. (1998). Geologic and geochemical controls on naturally occurring radioactive materials (NORM) in
   produced water from oil, gas, and geothermal operations. Environmental Geosciences 5:139-150.
   http://dx.doi.0rg/10.1046/i.1526-0984.1998.08018.x

Gilmore. K: Hupp. R: Glathar. I. (2013). Transport of Hydraulic Fracturing Water and Wastes in the
   Susquehanna River Basin, Pennsylvania. J Environ Eng 140: B4013002.
   http://dx.doi.org/10.1061/rASCE)EE.1943-7870.0000810

Gregory. KB: Vidic. RD: Dzombak. DA. (2011). Water management challenges associated with the production
   of shale gas by hydraulic fracturing. Elements 7:181-186.

Gross. SA: Avens. HI: Banducci. AM: Sahmel. I: Panko. IM: Tvermoes. BE. (2013). Analysis of BTEX
   groundwater concentrations from surface spills associated with hydraulic fracturing operations. J Air
   Waste Manag Assoc 63: 424-432. http://dx.doi.org/10.1080/10962247.2012.759166

Guerra. K: Dahm. K: Dundorf. S. (2011). Oil and gas produced water management and beneficial use in the
   western United States. (Science and Technology Program Report No. 157). Denver, CO: U.S. Department of
   the Interior Bureau of Reclamation.

GWPC and ALL Consulting (Ground Water Protection Council (GWPC) and ALL Consulting). (2009). Modern
   shale gas development in the United States: A primer.  (DE-FG26-04NT15455). Washington, DC: U.S.
   Department of Energy, Office of Fossil Energy and National Energy Technology Laboratory.
   http://www.gwpc.org/sites/default/files/Shale%20Gas%20Primer%202009.pdf
               This document is a draft for review purposes only and does not constitute Agency policy.

June 2015                                       7-51                   DRAFT—DO NOT CITE OR QUOTE

-------
Hydraulic Fracturing Drinking Water Assessment                         Chapter 7 - Flowback and Produced Water
GWPC and IOGCC (Ground Water Protection Council and Interstate Oil and Gas Compact Commission).
   (2014). Fracturing fluid management. Available online at http://fracfocus.org/hydraulic-fracturing-how-
   it-works/drilling-risks-safeguards (accessed December 12,2014).

Haluszczak. LO: Rose. AW: Kump. LR. (2013). Geochemical evaluation of flowbackbrine from Marcellus gas
   wells in Pennsylvania, USA. Appl Geochem 28: 55-61.
   http://dx.doi.0rg/10.1016/j.apgeochem.2012.10.002

Hansen, E: Mulvaney, D: Betcher, M. (2013). Water resource reporting and water footprint from Marcellus
   Shale development in West Virginia and Pennsylvania. Durango, CO: Earthworks Oil & Gas Accountability
   Project, http://www.downstreamstrategies.com/documents/reports publication/marcellus wv pa.pdf

Hayes. T. (2009). Sampling and analysis of water streams associated with the development of Marcellus shale
   gas. Des Plaines, IL: Marcellus Shale Coalition, http://eidmarcellus.org/wp-
   content/ uploads/2 012/11/MSCommission-Report.pdf

Hayes. T. (2011). Characterization of Marcellus and Barnett shale flowback waters and technology
   development for water reuse. Paper presented at Hydraulic Fracturing Study: March 2011 Technical
   Workshop on Water Resource Management, March 29-30, 2011, Arlington, VA.

Hayes. T: Severin. B. (2012a). Characterization of flowback water from the the Marcellus and the Barnett
   shale regions. Barnett and Appalachian shale water management and reuse technologies. (08122-05.09;
   Contract 08122-05). Hayes, T; Severin, B. http://www.rpsea.org/media/files/project/2146b3aO/08122-
   05-RT-Characterization Flowback Waters Marcellus Barnett Shale Regions-03-20-12.pdf

He, Z. (2011). Flow of gas and water in hydraulically fractured shale gas reservoirs. EPA HF Workshop, March
   28-29,2011, Arlington, VA.

Healy, RW: Bartos, TT: Rice, CA: Mckinley, MP: Smith, BD. (2011). Groundwater chemistry near an
   impoundment for produced water, Powder River Basin, Wyoming, USA. J Hydrol 403: 37-48.
   http://dx.doi.0rg/10.1016/j.jhydrol.2011.03.042

Healy. RW: Rice. CA: Bartos. TT: Mckinley. MP. (2008). Infiltration from an impoundment for coal-bed natural
   gas, Powder River Basin, Wyoming: Evolution of water and sediment chemistry. Water Resour Res 44:
   n/a-n/a. http://dx.doi.org/10.1029/2007wr006396

Horsey. CA. (1981). Depositional environments of the Pennsylvanian Pottsville Formation in the Black
   Warrior Basin of Alabama. Journal of Sedimentary Research 51: 799-806.
   http://dx.doi.org/10.1306/212F7DB5-2B24-llD7-8648000102C1865D

Hunter. IA: Moser. PH. (1990). Ground water availability in Jefferson County, Alabama: geological survey
   special map 224. Tuscaloosa, AL: Geological Survey of Alabama.
   http://www.ogb.state.al.us/documents/pubs/onlinepubs/Special  Maps/SM224 .PDF

IUPAC (International Union of Pure and Applied Chemistry). (2014). Gold Book: matrix effect. Available
   online at http://goldbook.iupac.org/M03759.html (accessed April 8, 2015).

Kharaka. YK: Kakouros. E: Thordsen. II: Ambats. G: Abbott. MM. (2007). Fate and groundwater impacts of
   produced water releases at OSPER B site, Osage County, Oklahoma. Appl Geochem 22: 2164-2176.
   http://dx.doi.0rg/10.1016/j.apgeochem.2007.04.005

Langmuir. D:Riese.AC. (1985). THE THERMODYNAMIC PROPERTIES OF RADIUM. Geochim Cosmo Act 49:
   1593-1601.

Lee. K: Neff. I. (2011). Produced water: Environmental risks and advances in mitigation technologies. New
   York, NY: Springer, http://dx.doi.org/10.1007/978-l-4614-0046-2

Levis. E. (2011). Texas company pays $93,710 settlement for polluting Clearfield County Creek. Pennsylvania
   Fish and Boat Commission. https://www.fish.state.pa.us/newsreleases/2011press/eog-settlement.htm

Lutz. BD: Lewis. AN: Doyle. MW. (2013). Generation, transport, and disposal of wastewater associated with
   Marcellus Shale gas development. Water Resour Res 49: 647-656.


              This document is a draft for review purposes only and does not constitute Agency policy.
June 2015                                      7-52                   DRAFT—DO NOT CITE OR QUOTE

-------
Hydraulic Fracturing Drinking Water Assessment                          Chapter 7 - Flowback and Produced Water
Mantell. ME. (2011). Produced water reuse and recycling challenges and opportunities across major shale
   plays. Presentation presented at Hydraulic Fracturing Study: March 2011 Technical Workshop #4 on
   Water Resources Management, March 29-30, 2011, Arlington, VA.

Mantell. ME. (2013b). Recycling and reuse of produced water to reduce freshwater use in hydraulic fracturing
   operations. In Summary of the technical workshop on water acquisition modeling: Assessing impacts
   through modeling and other means (pp. A20-A27). Washington, D.C.: U.S. Environmental Protection
   Agency, http://www2.epa.gov/hfstudy/summary-technical-workshop-water-acquisition-modeling-
   assessing-impacts-through-modeling-and

Marshak. S. (2004). Essentials of geology (1st ed.). New York, NY: W.W. Norton and Company.

McElreath. D. (2011).  Comparison of hydraulic fracturing fluids composition with produced formation water
   following fracturing Implications for fate and transport. Presentation presented at Hydraulic Fracturing
   Study: March 2011 Technical Workshop on Fate and Transport, March 2829,2011, Arlington, VA.

Mclntosh. 1C: Walter. LM. (2005). Volumetrically significant recharge of Pleistocene glacial meltwaters into
   epicratonic basins: Constraints imposed by solute mass balances. Chem Geol 222: 292-309.
   http://dx.doi.0rg/10.1016/j.chemgeo.2005.07.010

Mclntosh. 1C: Walter. LM: Martini. AM. (2002). Pleistocene recharge to midcontinent basins: effects on salinity
   structure and microbial gas generation. Geochim Cosmo Act 66:1681-1700.
   http://dx.doi.org/10.1016/S0016-7037r01)00885-7

Minnich. K. (2011). A water chemistry perspective on flowback reuse with several case studies. Minnich, K.
   http://www2.epa.gov/sites/production/files/documents/10 Minnich - Chemistry 508.pdf

Montana GWIC (Montana Ground Water Information Center). (2009). Coal bed methane powder river basin:
   Montana Bureau of Mines and Geology Ground Water Information Center. Butte, MT.

Mouser. P: Liu. S: Cluff. M: McHugh. M: Lenhart. I: MacRae. I. (In Press) Biodegradation of hydraulic fracturing
   fluid organic additives in sediment-groundwater microcosms.

Murali Mohan. A: Hartsock. A: Hammack. RW: Vidic. RD: Gregory. KB. (2013). Microbial communities in
   flowback water impoundments from hydraulic fracturing for recovery of shale gas. FEMS Microbiol Ecol.
   http://dx.doi.org/10.llll/1574-6941.12183

Neff. IM. (2002). Bioaccumulation in marine organisms:  Effect of contaminants from oil well produced water.
   Amsterdam:  Elsevier.

Nelson. AW: May. D: Knight. AW: Eitrheim. ES: Mehrhoff. M: Shannon. R: Litman. R: Schultz. MK. (2014).
   Matrix complications in the determination of radium levels  in hydraulic fracturing flowback water from
   Marcellus Shale. 1: 204-208. http://dx.doi.org/10.1021/ez5000379

New Mexico Bureau of Mines and Mineral Resources. (1994). Coalbed methane in the upper cretaceous
   fruitland formation, San Juan Basin, New Mexico and Colorado. Socorro, NM: New Mexico Bureau of Mines
   and Mineral Resources, New Mexico Institute of Mining and Technology.
   https://geoinfo.nmt.edu/publications/monographs/bulletins/146/

Nicot. IP: Scanlon. BR. (2012). Water use for shale-gas production in Texas, U.S. Environ Sci Technol 46:  3580-
   3586. http://dx.doi.org/10.1021/es204602t

Nicot, IP: Scanlon, BR: Reedy, RC: Costley, RA. (2014). Source and fate of hydraulic fracturing water in the
   Barnett Shale: a historical perspective. Environ Sci Technol  48: 2464-2471.
   http://dx.doi.org/10.1021/es404050r

North Dakota Department of Health. (2015).  Oil field environmental incident summary, incident
   20150107160242. Available online at
   http://www.ndhealth.gov/EHS/FOIA/Spills/Summarv Reports/20150107160242 Summary Reportpdf
               This document is a draft for review purposes only and does not constitute Agency policy.

June 2015                                       7-53                    DRAFT—DO NOT CITE OR QUOTE

-------
Hydraulic Fracturing Drinking Water Assessment                         Chapter 7 - Flowback and Produced Water
NRG (National Research Council). (2010). Management and effects of coalbed methane produced water in the
   western United States. Washington, DC: National Academies Press.
   http://www.nap.edu/catalog.php7record id=12915

NSTC (National Science and Technology Council). (2000). Integrated assessment of hypoxia in the Northern
   Gulf of Mexico. Washington, DC: U.S. National Science and Technology Council, Committee on Environment
   and Natural Resources, http://oceanservice.noaa.gov/products/hypox final.pdf

Olawoyin, R: Wang, IY: Oyewole, SA. (2013). Environmental safety assessment of drilling operations in the
   Marcellus-shale gas development. S P E Drilling & Completion 28: 212-220.
   http://dx.doi.org/10.2118/163095-PA

Olmstead. SM: Muehlenbachs. LA: Shih. IS: Chu. Z: Krupnick. Al. (2013). Shale gas development impacts on
   surface water quality in Pennsylvania. PNAS 110: 4962-4967.
   http://dx.doi.org/10.1073/pnas.1213871110

Orem. W: Tatu. C: Varonka. M: Lerch. H: Bates. A: Engle. M: Crosby. L: Mcintosh. I. (2014). Organic substances
   in produced and formation water from unconventional natural gas extraction in coal and shale. Int J Coal
   Geol 126: 20-31. http://dx.doi.Org/10.1016/j.coal.2014.01.003

Orem. WH: Tatu. CA: Lerch. HE: Rice. CA: Bartos. TT: Bates. AL: Tewalt. S: Corum. MD. (2007). Organic
   compounds in produced waters from coalbed natural gas  wells in the Powder River Basin, Wyoming, USA.
   Appl Geochem 22: 2240-2256. http://dx.doi.Org/10.1016/j.apgeochem.2007.04.010

Otton. IK: Zielinski.  RA: Smith. BD: Abbott. MM. (2007). Geologic controls on movement of produced-water
   releases at US geological survey research Site A, Skiatook  lake, Osage county, Oklahoma. Appl Geochem 22:
   2138-2154. http://dx.doi.0rg/10.1016/j.apgeochem.2007.04.015

PA DCNR (Pennsylvania Department of Conservation and Natural Resources). (2015). Thermal maturation
   and petroleum generation. Available online at
   http://www.dcnr.state.pa.us/topogeo/econresource/oilandgas/marcellus/sourcerock index/sourcerock
   maturation/index.htm (accessed April 9, 2015).

PA PEP (Pennsylvania Department of Environmental Protection). (2009a). Inspection Report, inspection
   record #1835041, enforcement record #251134. Harrisburg, PA: Commonwealth of Pennsylvania
   Department of Environmental Protection, Oil and Gas Management Program.

PA PEP (Pennsylvania Department of Environmental Protection). (2010). DEP Fines Atlas Resources for
   drilling wastewater spill in Washington County. Available online at
   http://www.portal.state.pa.us/portal/server.pt/community/newsroom/14287?id=13595&typeid=l
   (accessed February 13, 2014).

PA DEP (Pennsylvania Department of Environmental Protection). (2011c). Surface water sample analytical
   results from XTO 308 response data from XTO February 3, 2011 CAWP addendum. Indiana, PA: XTO
   Energy.

PA DEP (Pennsylvania Department of Environmental Protection). (2015b). Technologically enhanced
   naturally occurring radioactive materials (TENORM) study report. Harrisburg, PA.
   http://www.elibrary.dep.state.pa.us/dsweb/Get/Document-105822/PA-DEP-TENORM-
   Study Report Rev. 0 01-15-2015.pdf

Papoulias. DM: Velasco. AL. (2013). Histopathological analysis offish from Acorn Fork Creek, Kentucky,
   exposed to hydraulic fracturing fluid releases. Southeastern Naturalist 12: 92-111.

Paterra. P. (2011). DEP shuts down Tri-County Waste Water over illegal dumping. Available online at
   http://triblive.eom/x/pittsburghtrib/news/regional/s 728516.html#axzz3UCvkvM7t (accessed March
   12,2015).

Peterman. ZE: Thamke. I: Futa. K: Oliver. T. (2012). Strontium isotope evolution of produced water in the East
   Poplar Oil Field, Montana. Presentation presented at US Geological Survey AAPG annual convention and
   exhibition, April 23, 2012, Long Beach, California.


              This document is a draft for review purposes only and does not constitute Agency policy.
June 2015                                      7-54                    DRAFT—DO NOT CITE OR QUOTE

-------
Hydraulic Fracturing Drinking Water Assessment                          Chapter 7 - Flowback and Produced Water
Repetski. IE: Ryder. RT: Weary. PI: Harris. AG: Trippie. MH. (2008). Thermal maturity patterns (CAI and %Ro)
   in upper ordovician and devonian rocks of the Appalachian Basin: A major revision of USGS map I917E
   using new subsurface collections. U.S. Geological Survey. http://pubs.usgs.gov/sim/3006/SIM3006.pdf

Reuters. (2014). UPDATE 2-oil well in North Dakota out of control, leaking. Available online at
   http://www.reuters.com/article/2014/02/14/energy-crude-blowout-idUSL2NOLI15820140214
   (accessed March 2,2015).

Reynolds, RR: Kiker, RD. (2003). Produced water and associated issues a manual for the independent
   operator. (Oklahoma Geological Survey Open-File Report 6-2003). Tulsa, OK: Oklahoma Geological Survey.
   http://karl.nrcce.wvu.edu/regional/pww/produced water.pdf

Rice. CA. (1999). Waters co-produced with coal-bed methane from the FerronSandstone in east central Utah:
   chemical and isotopic composition, volumes.and impacts of disposal [Abstract]. Geological Society of
   America Abstracts with Programs 31: A385.

Rice. CA: Ellis. MS: Bullock. IH. (2000). Water co-produced with coalbed methane in the Powder River basin,
   Wyoming: Preliminary compositional data.  (Open File-Report 00-372). Denver, CO: U.S. Department of the
   Interior, U.S. Geological Survey.

Ricther. BC: Kreitler. CW. (1993). Geochemical  techniques for identifying sources of ground-water
   salinization. Boca Raton, FL: CRC Press, http://www.crcpress.com/product/isbn/9781566700009

Rowan. EL: Engle. MA: Kirby. CS: Kraemer. TF.  (2011). Radium content of oil- and gas-field produced waters
   in the northern Appalachian Basin (USA): Summary and discussion of data. (Scientific Investigations
   Report 20115135). Reston, VA: U.S. Geological Survey, http://pubs.usgs.gov/sir/2011/5135/

Rowan. EL: Engle. MA: Kraemer. TF: Schroeder. KT: Hammack. RW: Doughten. MW. (2015). Geochemical and
   isotopic evolution  of water produced from Middle Devonian Marcellus shale gas wells, Appalachian basin,
   Pennsylvania. AAPG Bulletin 99:181-206. http://dx.doi.org/10.1306/07071413146

Rozell. PI: Reaven. SI.  (2012). Water pollution risk associated with natural gas extraction from the Marcellus
   Shale. Risk Anal 32:13821393. http://dx.doi.Org/10.llll/j.1539-6924.2011.01757.x

Rushing. IA: Newsham. KE: Blasingame. TA. (2013). Rock typing: Keys to understanding productivity in tight
   gas sands. SPE Unconventional Reservoirs Conference, February 1012, 2008, Keystone, Colorado, USA.

SAIC and GES (SAIC Energy, Environment & Infrastructure, LLC and Groundwater  & Environmental Services,
   Inc). (2011). ATGAS investigation initial site characterization and response, April 19, 2011 to May 2, 2011,
   ATGAS2H Well Pad, permit no. 37-015-21237, Leroy Township, Bradford County, PA. Harrisburg,
   Pennsylvania: Chesapeake Appalachia, LLC.
   http://www.chk.com/news/articles/documents/atgas initial  site characterization report final 0829201
   l.pdf

Sang. W: Stoof. CR: Zhang. W: Morales. VL: Gao. B: Kay. RW: Liu. L: Zhang. Y: Steenhuis. TS. (2014). Effect of
   hydrofracking fluid on colloid transport in the unsaturated zone. Environ Sci Technol 48: 8266-8274.
   http://dx.doi.org/10.1021/es501441e

Schlegel. ME: Mclntosh. 1C: Petsch. ST: Orem. WH: lones. EIP: Martini. AM. (2013).  Extent and limits of
   biodegradation by in situ methanogenic consortia in shale and formation fluids. Appl Geochem 28:172-
   184. http://dx.doi.0rg/10.1016/j.apgeochem.2012.10.008

Schlumberger (Schlumberger Limited).  (2014). Schlumberger oilfield glossary. Available online at
   http://www.glossary.oilfield.slb.com/

Schmidley, EB: Smith, Bl. (2011). Personal communication from Schmidley and Smith to DiCello: CAWP
   Addendum  EM Survey & Well Location; XTO Energy, Inc. Marquardt Release. Available online
               This document is a draft for review purposes only and does not constitute Agency policy.

June 2015                                       7-55                    DRAFT—DO NOT CITE OR QUOTE

-------
Hydraulic Fracturing Drinking Water Assessment                          Chapter 7 - Flowback and Produced Water
Schmidt. V: McDonald. DA. (1979). The role of secondary porosity in the course of sandstone diagenesis. In PA
   Schole; PR Schluger (Eds.), Aspects of diagenesis : based on symposia sponsored by the Eastern and by the
   Rocky Mountain Sections, The Society of Economic Paleontologists and Mineralogists (pp. 175-207). Tulsa,
   OK: The Society of Economic Paleontologists and Mineralogists (SEPM).
   http://archives.datapages.com/data/sepm sp/SP26/The Role of Secondary Porosity.html

Schwarzenbach. RP: Gschwend. PM: Imboden. DM. (2002). Environmental Organic Chemistry. In
   Environmental organic chemistry (2 ed.). Hoboken, NJ: John Wiley & Sons, Inc.

Science Applications International Corporation. (2010). XTO - Marquardt pad soil and water corrective action
   workplan. (XTO-EPA0001074). Indiana, PA: XTO Energy.

Simunek. I: Sejna. M: van Genuchten. MT. (1998). The HYDRUS-1D software package for simulating the one-
   dimensional movement of water, heat, and multiple solutes in variably-saturated media, Version 2.0,
   IGWMC-TPS-70. Available online

Sirivedhin. T: Dallbauman. L. (2004). Organic matrix in produced water from the Osage-Skiatook petroleum
   environmental research site, Osage county, Oklahoma. Chemosphere 57: 463-469.

States. S: Cyprych. G: Stoner. M: Wydra. F: Kuchta. I: Monnell. I: Casson. L. (2013). Marcellus Shale drilling and
   brominated THMs in Pittsburgh, Pa., drinking water. J Am Water Works Assoc 105: E432-E448.
   http://dx.doi.org/10.5942/jawwa.2013.105.0093

Stewart. DR.  (2013a). Analytical testing for hydraulic fracturing fluid water recovery and reuse.  In Summary
   of the technical workshop on analytical chemical methods (pp. B6-B10). Stewart, DR.
   http://www2.epa.gov/sites/production/files/documents/analytical-chemical-methods-technical-
   workshop-summary.pdf

Strong, L: Gould, T: Kasinkas, L: Sadowsky, M: Aksan, A: Wackett, L. (2013). Biodegradation in waters from
   hydraulic fracturing: chemistry, microbiology, and engineering. J Environ Eng 140: B4013001.
   http://dx.doi.org/10.1061/(ASCE)EE.1943-7870.0000792

Sturchio. NC: Banner. IL: Binz. CM: Heraty. LB: Musgrove. M. (2001). Radium geochemistry of ground waters
   in Paleozoic carbonate aquifers, midcontinent, USA. Appl Geochem 16:109-122.

Sumi, L. (2004). Pit pollution: Backgrounder on the issues, with a New Mexico case study. Washington, DC:
   Earthworks: Oil and Gas Accountability Project.
   http://www.earthworksaction.org/files/publications/PitReport.pdf

Sun. M: Lowry. GV: Gregory. KB. (2013). Selective oxidation of bromide in wastewater brines from hydraulic
   fracturing. Water Res 47: 3723-3731. http://dx.doi.Org/10.1016/j.watres.2013.04.041

Swanson. VE. (1955). Uranium in marine black shales of the United States. In Contributions to the geology of
   uranium and thorium by the United States Geological Survey and Atomic Energy Commission for the
   United Nations International Conference on Peaceful Uses of Atomic Energy, Geneva, Switzerland, 1955
   (pp. 451-456). Reston, VA: U.S. Geological Survey, http://pubs.usgs.gov/pp/0300/report.pdf

Thordsen. II:. Kharaka. YK: Ambats. G: Kakouros. E: Abbott. MM. (2007). Geochemical data from produced
   water contamination investigations: Osage-Skiatook Petroleum Environmental Research (OSPER) sites,
   Osage County, Oklahoma. (Open-File Report 2007-1055). Reston, VA: United States Geological Survey.

U.S. EPA (U.S. Environmental Protection Agency). (1992). Guidance for data useability in risk assessment
   (part A) - final. (Publication 9285.7-09A). Washington, D.C.
   http://www.epa.gov/oswer/riskassessment/datause/parta.htm

U.S. EPA (U.S. Environmental Protection Agency). (1999). Understanding oil spills and oil spill response [EPA
   Report]. (EPA 540-K-99-007). Washington, D.C.: U.S. Environmental Protection Agency, Office of
   Emergency and Remedial Response.
   http://www4.nau.edu/itep/waste/hazsubmap/docs/OilSpill/EPAUnderstandingOilSpillsAndOilSpillResp
   onsel999.pdf
               This document is a draft for review purposes only and does not constitute Agency policy.

June 2015                                       7-56                    DRAFT—DO NOT CITE OR QUOTE

-------
Hydraulic Fracturing Drinking Water Assessment                          Chapter 7 - Flowback and Produced Water
U.S. EPA (U.S. Environmental Protection Agency). (2004). Evaluation of impacts to underground sources of
   drinking water by hydraulic fracturing of coalbed methane reservoirs. (EPA/816/R-04/003). Washington,
   DC.: U.S. Environmental Protection Agency, Office of Solid Waste.

U.S. EPA (U.S. Environmental Protection Agency). (2007). Monitored natural attenuation of inorganic
   contaminants in ground water: volume Itechnical basis for assessment [EPA Report]. (EPA/600/R-
   07/139). Washington, D.C. http://nepis.epa.gov/Adobe/PDF/60000N4K.pdf

U.S. EPA (U.S. Environmental Protection Agency). (2012a). 5.2 Dissolved oxygen and biochemical oxygen
   demand. In Water Monitoring and Assessment, http://water.epa.gov/type/rsi/monitoring/vms52.cfm

U.S. EPA (U.S. Environmental Protection Agency). (2012f). Study of the potential impacts of hydraulic
   fracturing on drinking water resources: Progress report. (EPA/601/R-12/011). Washington, DC: U.S.
   Environmental Protection Agency, Office of Research and Development.
   http://nepis.epa.gov/exe/ZyPURL.cgi?Dockey=P100FH8M.txt

U.S. EPA (U.S. Environmental Protection Agency). (2013e). SW-846 on-line. Available online at
   http://www.epa.gov/epawaste/hazard/testmethods/sw846/online/index.htm (accessed April 8,2015).

U.S. EPA (U.S. Environmental Protection Agency). (2013g). XTO  Energy, Inc. Settlement. Available online at
   http://www2.epa.gov/enforcement/xto-energy-inc-settlement

U.S. EPA (U.S. Environmental Protection Agency). (2014b). Development of rapid radiochemical method for
   gross alpha and gross beta activity concentration in flowback and produced waters from hydraulic
   fracturing operations [EPA Report]. (EPA/600/R-14/107). Washington, D.C.
   http://www2.epa.gov/hfstudy/development-rapid-radiochemical-method-gross-alpha-and-gross-beta-
   activity-concentration

U.S. EPA (U.S. Environmental Protection Agency). (2014b). The verification of a method for detecting and
   quantifying diethylene glycol, triethylene glycol, tetraethylene glycol, 2-butoxyethanol and 2-
   methoxyethanol in ground and surface waters [EPA Report]. (EPA/600/R-14/008). Washington, D.C.
   http://www2.epa.gov/hfstudy/verification-method-detecting-and-quantifying-diethylene-glycol-
   triethylene-glycol

U.S. EPA (U.S. Environmental Protection Agency). (2015J). Retrospective case study in Killdeer,  North Dakota:
   study of the potential impacts of hydraulic fracturing on drinking water resources [EPA Report]. (EPA
   600/R-14/103). Washington, D.C.

U.S. EPA (U.S. Environmental Protection Agency). (2015k). Retrospective case study in southwestern
   Pennsylvania: study of the potential impacts of hydraulic fracturing on drinking water resources [EPA
   Report]. (EPA 600/R-14/084). Washington, D.C.

U.S. EPA (U.S. Environmental Protection Agency). (2015m). Retrospective case study in Wise County, Texas:
   study of the potential impacts of hydraulic fracturing on drinking water resources [EPA Report]. (EPA
   600/R-14/090). Washington, D.C.

U.S. EPA (U.S. Environmental Protection Agency). (2015n). Review of state and industry spill data:
   characterization of hydraulic fracturing-related spills [EPA Report]. (EPA/601/R-14/001). Washington,
   D.C.: Office of Research and Development, U.S. Environmental Protection Agency.

U.S. EPA (U.S. Environmental Protection Agency). (2015q). Technical development document for proposed
   effluent limitation guidelines and standards for oil and gas extraction. (EPA-821-R-15-003). Washington,
   D.C. http://water.epa.gov/scitech/wastetech/guide/oilandgas/unconv.cfm

U.S. GAP (U.S. Government Accountability Office). (2012). Energy-water nexus:  Information on the quantity,
   quality, and management of water produced during oil and gas production. (GAO-12-156). Washington,
   D.C. http://www.gao.gov/products/GAO-12-156
               This document is a draft for review purposes only and does not constitute Agency policy.

June 2015                                       7-57                    DRAFT—DO NOT CITE OR QUOTE

-------
Hydraulic Fracturing Drinking Water Assessment                          Chapter 7 - Flowback and Produced Water
USGS (U.S. Geological Survey). (2006). Produced Water Database [Database]: U.S. Geological Survey :: USGS.
   Retrieved from
   http://energy.usgs.gov/EnvironmentalAspects/EnvironmentalAspectsofEnergyProductionandUse/Produ
   cedWaters.aspx#3822110-overview

USGS (U.S. Geological Survey). (2010). Volatile organic compounds in the nation's ground water and drinking-
   water supply wells: Supporting information. Glossary. Available online at
   http://water.usgs.gov/nawqa/vocs/national assessment/report/glossary.html

USGS (U.S. Geological Survey). (2013b). National hydrography dataset: High-resolution flowline data: The
   national map. Retrieved from http://viewer.nationalmap.gov/viewer/

USGS (U.S. Geological Survey). (2014f). USGS investigations of water produced during hydrocarbon reservoir
   development [Fact Sheet]. Reston, VA. http://dx.doi.org/10.3133/fs20143104

Vaidyanathan. G. (2013a). Hydraulic fracturing: when 2 wells meet, spills can often follow. Available online at
   http://www.eenews.net/energywire/stories/1059985587 (accessed September 2, 2014).

Vaidyanathan, G. (2013b). XTO comes out swinging against 'unwarranted' criminal charges in Pa. E&E News
   0.

Van Voast. WA. (2003). Geochemical signature of formation waters associated with coalbed methane. AAPG
   Bulletin 87: 667-676.

Vengosh. A: lackson. RB: Warner. N: Darrah. TH: Kondash. A. (2014). A critical review of the risks to water
   resources from unconventional shale gas development and hydraulic fracturing in the United States.
   Environ Sci Technol 48:  36-52. http://dx.doi.org/10.1021/es405118y

Vidic, RD: Brantley, SL: Vandenbossche, IM: Yoxtheimer, D: Abad, ID. (2013). Impact of shale gas development
   on regional water quality [Review]. Science 340:1235009. http://dx.doi.org/10.1126/science.1235009

Vine. ID: Tourtelot.  EB. (1970). Geochemistry of black shale deposits; A summary report. Econ Geol 65: 253-
   272. http://dx.doi.org/10.2113/gsecongeo.65.3.253

Vittitow. IG. Sr. (2010). Well control incident analysis, EOG Resources Inc., Punxautawney Hunting Club 36H,
   Clearfield County, Pennsylvania. Bedrock Engineering.
   http://www.pahouse.com/EnvResources/documents/BEDROCK ENGINEERING PHC 36H Incident Repo
   rt Final.pdf

Warner. NR: Kresse. TM: Hays. PD: Down. A: Karr. ID: lackson. RB: Vengosh. A. (2013b). Geochemical and
   isotopic variations in shallow groundwater in areas of the Fayetteville Shale development, north-central
   Arkansas. Appl Geochem 35: 207-220.

Whittemore. DO. (2007). Fate and identification of oil-brine contamination in different hydrogeologic
   settings. Appl Geochem 22:  2099-2114. http://dx.doi.Org/10.1016/j.apgeochem.2007.04.002

Williams. IE: Taylor. LE: Low. PI. (1998). Hydrogeology and Groundwater Quality of the Glaciated Valleys of
   Bradford, Tioga, and Potter Counties, Pennsylvania. 98.

Wilson. B. (2014). Geologic  and baseline groundwater evidence for naturally occurring, shallowly sourced,
   thermogenic gas in northeastern  Pennsylvania. AAPG Bulletin 98: 373-394.
   http://dx.doi.org/10.1306/08061312218

Wuchter. C: Banning. E:  Mincer. Tl: Drenzek. Nl: Coolen. Ml. (2013). Microbial diversity and methanogenic
   activity of Antrim Shale formation waters from recently fractured wells. FMICB 4:1-14.
   http://dx.doi.org/10.3389/fmicb.2013.00367

Xu. B: Hill. AD: Zhu. D: Wang. L. (2011). Experimental evaluation of guar fracture fluid filter cake behavior.
   Paper presented at SPE Hydraulic Fracturing Technology Conference, January 24-26,  2011, The
   Woodlands, TX.
               This document is a draft for review purposes only and does not constitute Agency policy.

June 2015                                       7-58                    DRAFT—DO NOT CITE OR QUOTE

-------
Hydraulic Fracturing Drinking Water Assessment                          Chapter 7 - Flowback and Produced Water
Ziemkiewicz. P: Ouaranta. ID: Mccawley. M. (2014). Practical measures for reducing the risk of environmental
   contamination in shale energy production. Environ Sci Process Impacts 16:1692-1699.
   http://dx.doi.org/10.1039/c3em00510k
               This document is a draft for review purposes only and does not constitute Agency policy.

June 2015                                        7-59                    DRAFT—DO NOT CITE OR QUOTE

-------
Hydraulic Fracturing Drinking Water Assessment           Chapter 8 - Wastewater Treatment and Waste Disposal
                          Chapter 8
Wastewater Treatment and Waste Disposal
          This document is a draft for review purposes only and does not constitute Agency policy.
June 2015                                          DRAFT—DO NOT CITE OR QUOTE

-------
      Hydraulic Fracturing Drinking Water Assessment               Chapter 8 - Wastewater Treatment and Waste Disposal


      8.  Wastewater Treatment and Waste Disposal

      8.1. Introduction
 1    Hydraulic fracturing used for the development of oil and gas resources results in the production of
 2    wastewater containing a range of problematic or potentially problematic constituents (see Chapter
 3    7) and requiring management For the purposes of this assessment, hydraulic fracturing
 4    wastewater encompasses flowback and produced water (often referred to together as produced
 5    water) that is managed using any of a number of practices, including treatment and discharge,
 6    reuse, or injection into Class IID wells regulated under the Underground Injection Control (UIC)
 7    program under the Safe Drinking Water Act (SDWA) (see also Chapter 1).: In this chapter, the term
 8    "wastewater" is generally used. In limited cases where more specific information is provided about
 9    a wastewater (e.g., a source indicates that the wastewater is flowback), that information is also
10    noted.

11    Although wells producing from either unconventional or conventional oil and gas reservoirs
12    generate produced water during the course of their productive lifespan, wells conducting modern
13    high-volume hydraulic fracturing can generate a large volume of flowback water in the period
14    immediately after fracturing. Stakeholders reported to the U.S. Government Accountability Office
15    that flowback volumes could be 420,000 gal to 2.52 Mgal (10,000 to 60,000 bblor 1.59 million to
16    9.54 million L) per well per hydraulic fracture (U.S. GAP. 20121 (see Chapter 7.1.1 for more
17    information on produced water volumes per well in various geologic basins and plays). This
18    necessitates having a wastewater management strategy in place at the beginning of activities at the
19    well. Selection of management choices may depend upon the quality and volume of the fluids,
20    logistics, and economics.

21    Treatment and disposal strategies vary throughout the United States and may include underground
22    injection, on-site or offsite treatment for reuse in subsequent hydraulic fracturing operations, reuse
23    without treatment, or other uses. In some areas, wastewater may be applied to the land (e.g., for
24    irrigation) or held in pits for evaporation. The large majority of wastewater generated from all oil
25    and gas operations in the United States is disposed of via Class IID wells (Clark and Veil. 2009). As
26    hydraulic fracturing activity matures, costs of different disposal practices may change in various
27    regions due to factors such as regulations, available infrastructure, feasibility and cost of reuse
28    practices, and other concerns that are difficult to anticipate and quantify at the time of this
29    assessment

30    Over the past decade, the rapid increase in modern hydraulic fracturing activities has led to the
31    need to manage the associated wastewater. There has been a shift towards reuse in areas where
      1 The term "wastewater" is being used in this study as a general description of certain waters and is not intended to
      constitute a term of art for legal or regulatory purposes. This general description does not, and is not intended to, provide
      that the production, recovery, or recycling of oil, including the production, recovery, or recycling of produced water or
      flowback water, constitutes "wastewater treatment" for the purposes of the Oil Pollution Prevention regulation (with the
      exception of dry gas operations), which includes the Spill Prevention, Control, and Countermeasure rule and the Facility
      Response Plan rule, 40 CFR 112 et seq.
                   This document is a draft for review purposes only and does not constitute Agency policy.
      June 2015                                     8-1                   DRAFT—DO NOT CITE OR QUOTE

-------
      Hydraulic Fracturing Drinking Water Assessment              Chapter 8 - Wastewater Treatment and Waste Disposal


 1    there are relatively few Class IID wells (e.g., the Marcellus Shale region) and indications of interest
 2    in reuse in areas where access to water for fracturing is limited (e.g., Anadarko Basin in TX and OK).
 3    The term reuse is sometimes used to imply no treatment or basic treatment (e.g., media filtration)
 4    for the removal of constituents other than total dissolved solids (IDS), while recycling is sometimes
 5    used to convey more extensive treatment (e.g., reverse osmosis (RO)) to remove IDS (Slutz etal.,
 6    2012). In this document, the term "reuse" will be used to indicate use of wastewater for subsequent
 7    hydraulic fracturing, without regard to the level of treatment

 8    This chapter provides follow-on to Chapter 7, which discusses the composition and per-well
 9    volumes of produced water and the processes involved in its generation. In this chapter,
10    discussions are included on management practices for hydraulic fracturing wastewaters, available
11    wastewater production information, and estimated aggregate volumes of wastewater generated for
12    several states with active hydraulic fracturing (Section 8.2). As a complement to information on the
13    composition of wastewaters in Chapter 7, issues of concern associated with wastewater
14    constituents are also presented (Section 8.3). Management methods that are used in 2014-2015  or
15    have been used in recent years are described (Section 8.4). Information is then presented on the
16    types and effectiveness of treatment processes that are suitable for removal of constituents of
17    concern in hydraulic fracturing wastewaters, either in centralized waste treatment facilities (CWTs)
18    or on-site treatment; examples of CWTs are presented (Section 8.5 and Appendix F). With the
19    background provided in the earlier sections of the chapter, documented and potential impacts on
20    drinking water resources are discussed (Section 8.6), and a final synthesis discussion is then
21    provided (Section 8.7).

22    This chapter makes use of background information collected by the EPA's Office of Water (OW) as
23    part of its development of proposed effluent limitations guidelines and standards for wastewater
24    from unconventional oil and gas resources (U.S. EPA. 2015q). These are defined by guidelines and
25    standards as resources in low permeability formations including oil and gas shales, tight oil, and
26    low permeability sandstones and carbonates. Coalbed methane is beyond the scope of those
27    standards.  In this chapter we consider wastewater generated by hydraulic fracturing of those
28    unconventional oil and gas resources included in the background research done by OW in addition
29    to wastewater generated by hydraulic fracturing in coalbed methane plays and conventional
30    reservoirs.

      8.2. Volumes of Hydraulic Fracturing Wastewater
31    This section of Chapter 8 provides a general overview of aggregate wastewater quantities
32    generated in the course of hydraulic fracturing and subsequent oil and gas production, including
33    estimates at regional and state  levels. It also discusses methodologies used to produce these
34    estimates and the challenges associated with the preparation and use of available estimates.
35    (Chapter 7 provides a more in-depth discussion of the processes affecting produced water volumes
3 6    and presents some typical per-well values and temporal patterns.) Wastewater volumes most likely
3 7    will vary in the future as the amount and locations of hydraulic fracturing activities change and as
38    existing wells age and move into later phases of production. The volumes and management of
                   This document is a draft for review purposes only and does not constitute Agency policy.
      June 2015                                    8-2                   DRAFT—DO NOT CITE OR QUOTE

-------
Hydraulic Fracturing Drinking Water Assessment
                                                       Chapter 8 - Wastewater Treatment and Waste Disposal
 1    wastewater are important factors affecting the potential for wastewater to affect drinking water
 2    resources.

 3    The volume of wastewater generated is generally tied to the volume of oil and gas production; as
 4    operators increase hydrocarbon production, there will be a corresponding increase in wastewater
 5    volumes to be managed. For example, data from the Pennsylvania Department of Environmental
 6    Protection (PA DEP) [PA PEP, 2015a] (see Figure 8-1) show trends in volumes of wastewater
 7    compared to gas produced from wells in the Marcellus Shale in Pennsylvania. Although the data
 8    show some variation, they demonstrate a general correlation between wastewater and produced
 9    gas.
       18000000
          115000000
             12000000
          I  9000000
              eoooooo
              3000000
                                                                             2,500,000,000
                                                                             2,000,000,000
                                                                             1,500,000,000 —
                                                                             1,000,000,000
                                                                             500,000,000
                   '




                                     '  X'


                                                 '
                    Flowback_PW disposed of (bbl)
                                                   X
                                              X'X'

                                                                    '
                                                              Gas produced (mcf)
      Figure 8-1. Produced and flowback water volumes and produced gas volumes from
               unconventional wells in Pennsylvania from July of 2009 through June of 2014.
               Source: PA DEP (2015a).

1 0    Information presented in Chapter 7 highlights the initial rapid recovery of fluid in the first weeks
11    after fracturing (see Figure 7-2), with a subsequent substantial reduction in the volume of water
1 2    flowing through the well to the surface. This is followed by recovery of produced water during the
13    longer-term productive phase of the well's life. One source suggests that, as a general rule of thumb,
14    the amount of flowback produced in the days or weeks after hydraulic fracturing is roughly
15    comparable to the amount of long-term produced water generated over a span of years, which may
16    vary considerably among wells (IHS,  2013). Thus, on a local level, operators can anticipate a
17    relatively large volume of wastewater in the weeks following fracturing, with slower subsequent
18    production of wastewater.  Wells also generate some amount of drilling- fluid waste. Compared to
                   This document is a draft for review purposes only and does not constitute Agency policy.
      June 2015                                    8-3                  DRAFT— DO NOT CITE OR QUOTE

-------
      Hydraulic Fracturing Drinking Water Assessment              Chapter 8 - Wastewater Treatment and Waste Disposal


 1    produced water, however, drilling waste water can constitute a relatively small portion of the total
 2    wastewater produced (e.g., <10% in Pennsylvania during 2004-2013) [U.S. EPA. 2015q] and is not
 3    discussed further in this assessment

 4    In addition to variation in per-well wastewater volumes, aggregate wastewater production for an
 5    area or region will vary from year to year with hydraulic fracturing activity. For instance, the
 6    average annual volume of wastewater generated by all gas production (both shale gas and
 7    conventional) in Pennsylvania quadrupled from the 2001-2006 period to the 2008-2011 period.
 8    During the latter period, wastewater volume averaged 1.1 billion gal (26 million bbl or 4.2 billion L)
 9    per year (Wilson and Vanbriesen. 2012).

      8.2.1.  National Level Estimate
10    Clark and Veil (2009) estimated that in 2007, the approximately one million active oil and gas wells
11    in the United States generated approximately 2.4 billion GPD (57.4 million bbl/day; 9.1 billion
12    L/day) of wastewater; no newer comprehensive national-level estimate exists in the literature as of
13    April 2015. Note that this estimate is not limited to wastewater from hydraulic fracturing
14    operations. This national-level estimate is reported to represent total oil and gas wastewater (from
15    conventional and unconventional resources, and from wells hydraulically fractured and wells not
16    hydraulically fractured), but the authors note that it does not include the flowback water
17    component Although Clark and Veil (2009) conducted a state-by-state analysis, the report may
18    have underestimated production due to significant data limitations: 1) data based on a timeframe
19    preceding the dramatic increase in hydraulic fracturing activity seen in more recent years; 2)
20    estimates based on data that were collected and maintained in a variety of ways, making data
21    synthesis difficult; and 3) incomplete data (U.S. GAP. 2012).

      8.2.2.  Regional/State and Formation Level Estimates
2 2    The amount of wastewater generated in a given region varies widely depending upon the volume of
23    wastewater generated per well and the number of wells in the area. The factors influencing
24    wastewater production are discussed in Chapter 7, which also discusses differences among
25    formations in volumes recovered during flowback and long-term water production. Table 7-2
26    provides rates of produced water generation for a number of formations in the United States.

27    At an aggregate level, wastewater volumes and geographic variability have been assessed for oil
28    and gas production in several studies. A 2011 study by the Bureau of Land Management (BLM)
29    (Guerraetal..  2011) states that more than 80% of oil and gas wastewater is generated in the
30    western United States, including volumes from both conventional and unconventional resources.
31    The BLM report notes substantial contributions from coalbed methane (CBM) wells, in particular
32    those in the Powder River Basin (Wyoming). The authors state that Wyoming produces the second
33    highest volume of water among the western United States. Guerraetal. (2011) also highlight the
34    large portion of wells and wastewater associated with Texas (44% of U.S. produced water volume).
35    Although the authors do not identify all wastewater contributions from production involving
36    hydraulic fracturing, the regions with established oil and gas production are likely to have methods
37    and infrastructure available for management of hydraulic fracturing wastewater. Figure 8-2
38    summarizes the findings for these western states, demonstrating the wide variability in wastewater

                   This document is a draft for review purposes only and does not constitute Agency policy.
      June 2015                                    8-4                  DRAFT—DO NOT CITE OR QUOTE

-------
      Hydraulic Fracturing Drinking Water Assessment
Chapter 8 - Wastewater Treatment and Waste Disposal
 1    from state to state (likely reflecting differences in formation geology and oil and gas production
 2    activity).
              32,515
                              2,080     19,792
                                                  2,040
                                                   18,021|
                                                     1,603  |
                                                      1,733  |
                                                    19,221
                    I California
                    I Colorado
                    I Kansas
                    I Montana
                    I New Mexico
                    I Oklahoma
                    lTexas
                    I Utah
                    I Wyoming
                    I Remaining W. States
      Figure 8-2. Wastewater quantities in the western United States (billions of gallons per year).
               Source: Guerra et al. (2011).

 3    Table 8-1 presents estimates of the numbers of wells and volumes of hydraulic fracturing
 4    wastewater generated in North Dakota, Ohio, Pennsylvania, Texas, and West Virginia. The data
 5    shown in this table come from publicly available state websites and databases; data for West
 6    Virginia reference a report by Hansenetal. [2013] that compiled available flowback data from West
 7    Virginia. The reported volumes have been summed by year. These states are represented in Table
 8    8-1 because the produced water volumes were readily identifiable as associated with hydraulic
 9    fracturing activity. Differences in the years presented for the states are due to differences in data
10    availability from the state agency databases. However, the increases in the numbers of wells
11    producing wastewater and the volumes of wastewater produced are generally consistent with the
12    timing of the expansion of high-volume hydraulic fracturing and track with the increase in
13    horizontal wells seen in Figure 2-12.

14    Several states with mature oil and gas industries (California, Colorado, New Mexico Utah,
15    Wyoming) make produced water volumes publicly available by well as part of their oil and gas
16    production data, but they do not directly indicate which wells have been hydraulically fractured.
17    Some states (Colorado, Utah, and Wyoming) specify the producing formation along with volumes of
18    hydrocarbons and produced water. New Mexico provides data for separate basins as well as for the
19    entire state. Determining volumes of hydraulic fracturing wastewater for these states is challenging
20    because there is a possibility of either inadvertently including wastewater from wells not
21    hydraulically fractured or of missing volumes that should be included. Appendix Table F-l provides
                   This document is a draft for review purposes only and does not constitute Agency policy.
      June 2015                                   8-5                   DRAFT—DO NOT CITE OR QUOTE

-------
     Hydraulic Fracturing Drinking Water Assessment               Chapter 8 - Wastewater Treatment and Waste Disposal


1    estimates of wastewater volumes in these states in regions where hydraulic fracturing activity is
2    taking place along with notes on data limitations.

3    The data in Table 8-1 and Appendix Table F-l illustrate the challenges both for compiling a national
4    estimate of hydraulic fracturing wastewater as well as comparing wastewater production among
5    states due to dissimilar data types, presentation, and availability.
                   This document is a draft for review purposes only and does not constitute Agency policy.
     June 2015                                      8-6                    DRAFT—DO NOT CITE OR QUOTE

-------
Hydraulic Fracturing Drinking Water Assessment
                                                        Chapter 8 - Wastewater Treatment and Waste Disposal
Table 8-1. Estimated volumes (millions of gallons) of wastewater based on state data for selected years and numbers of wells
          producing fluid.
State
North
Dakota


Ohio


Pennsylvania



Basin
Williston


Appalachian


Appalachian



Principal
Lithologies
Shale


Shale


Shale



Data Type
Produced water


Wells
Primarily
flowback water


Wells
Flowback water



Wells
Produced water
Wells
2000
2


161
-


-
-



-
-
-
2004
3


152
-


-
-



-
-
-
2008
130


844
-


-
-



-
-
-
2010
790


2,033
-


-
92



334
90
1,035
2011
1,900


3,303
3


9
340



1,564
400
1,826
2012
4,500


5,036
29


86
410



1,622
730
3,665
2013
8,500


6,913
110


400
350



1,465
930
4,761
2014
9,700


8,039
-


-
210



895
440
4,889
Comments
From North Dakota Oil and Gas
Commission website.3 Data
provided for six members of the
Bakken Shale. Produced water
includes flowback (reports are
submitted within 30 days of well
completion.)

Data from Ohio DNR Division of
Oil and Gas.b Utica data for 2011
and 2012. Utica and Marcellus
data for 2013. Brine is noted by
agency to be mostly flowback.

Waste data from PA DEP.C 2nd
half of 2010 and first half of
2014. Data described as
unconventional as determined
by formation. Separate codes are
provided by PA DEP for flowback
and produced water.



June 2015
This document is a draft for review purposes only and does not constitute Agency policy.
                                8-7
DRAFT—DO NOTE CITE OR QUOTE

-------
Hydraulic Fracturing Drinking Water Assessment
                                                               Chapter 8 - Wastewater Treatment and Waste Disposal
State
Pennsylvania
cont.
Texas
West Virginia
Basin
Appalachian
cont.
Unspecified
(entire
state)
Appalachian
Principal
Lithologies
Shale,
cont.
Shale,
Sandstone
Shale
Data Type
Flowback and
produced
water
Wells
Flowback water
- injected
volumes
Flowback water
- Estimated
total disposed
2000

-


2004

-


2008

-


2010
180
1,232

120
2011
740
2,434
490
110
2012
1,100
4,039
2,200
59
2013
1,300
5,015
3,100

2014
650
5,150
2,000

Comments


Waste injection data from Texas
Railroad Commission.11 Monthly
totals are provided for entire
state. Oct - Dec for 2011, full
years for 2012 and 2013, and Jan
- Oct for 2014
Estimates from Hansen et al.
(2013).
 a North Dakota Industrial Commission. Department of Mineral Resources. Bakken Horizontal Wells By Producing Zone:
 https://www.dmr.nd.gov/oilgas/bakkenwells.asp.
 b Ohio Department of Natural Resources, Division of Oil and Gas Resources. Oil and Gas Well Production. http://oilandgas.ohiodnr.gov/productiontfARCHl.
 c Pennsylvania Department of Environmental Protection. PA DEP Oil and Gas Reporting Web site.
 https://www.paoilandgasreporting.state.pa.us/publicreports/Modules/Welcome/Agreement.aspx
 d Railroad Commission of Texas. Injection Volume Query.
 http://webapps.rrc.state.tx.us/H10/searchVolume.do:isessionid=l3cgVHhK9nkwPrC7ZcWNMgyzF9LCYyRlNmvDy3F1005wqXfcGNGN!1841197795?fromMain=yes&
 sessionld= 143075601021612. Texas state data provide an aggregate total amount of flowback fluid injected for the past few years. (Data on brine volumes injected do
 not differentiate hydraulically fractured wells and are therefore not presented here.) These values are interpreted as estimates of generated flowback water as based
 on reported quantities of "fracture water flow back" injected into Class IID wells.
June 2015
This document is a draft for review purposes only and does not constitute Agency policy.

                                   8-8
DRAFT—DO NOTE CITE OR QUOTE

-------
      Hydraulic Fracturing Drinking Water Assessment              Chapter 8 - Wastewater Treatment and Waste Disposal


      8.2.3.  Estimation Methodologies and Challenges
 1    Compiling and comparing data regarding wastewater production at the wide array of oil and gas
 2    locations in the United States presents challenges, and various approaches are used to estimate
 3    wastewater volumes, both at the state and national level. Data from state agency web sites and
 4    databases can be a ready source of information, whether publicly available and downloadable or
 5    provided directly by agencies upon request. However, due to sometimes significant differences in
 6    the types of data collected, mechanisms, formats, and definitions used, data cannot always be
 7    directly compared from state to state and can be difficult to aggregate at a national level. The
 8    inconsistences encountered in data searches for this assessment agree with recent conclusions by
 9    Malone etal.  [2015], who noted inconsistences among 10 states with unconventional oil and gas
10    activity in the accessibility, usability, completeness, accuracy, and cost of various types of data  (e.g.,
11    wells drilled, production, waste, Class IID wells).

12    One challenge associated with using state production data to estimate the volume of wastewater
13    nationally or regionally is the lack of consistency in data collection [U.S. GAP. 2012). Some states do
14    not include a listing of wastewater (usually listed as produced water volumes) in their publicly
15    available oil and gas production reports, while others do. State tracking of wastewater volumes may
16    or may not include information that helps in determining whether the producing well was
17    hydraulically fractured (e.g., an indicator of resource type or formation). It also might not be clear
18    whether volumes listed as produced water include the flowback component Some states (e.g.,
19    Colorado) include information on disposal and management methods along with production data,
20    and others do not

21    Given these limitations, some studies have generated estimates of wastewater volume using water-
22    to-gas and water-to-oil ratios along with the reports of hydrocarbon production (Murray, 2013).
23    The reliability of any wastewater estimates made using this method will need to be evaluated in
24    terms of the quality, timeframe, and spatial coverage of the available data, as well as the extent of
25    the area to which the estimates will be applied. Water-to-hydrocarbon ratios are empirical
26    estimates. Because these  ratios show a wide variation among formations, reliable data are needed
27    to formulate a ratio in a particular region.

2 8    Another approach to estimating wastewater volumes would entail multiplying per-well estimates
29    of flowback and produced water production rates by the numbers of wells in a given area.
30    Challenges associated with this approach include obtaining accurate estimates of the number of
31    new and existing wells, along with accurate estimates of per-well water production both during the
3 2    flowback period and during the production phase of the wells' lifecycle. In particular, it can be
3 3    challenging to correctly match per-well wastewater production estimates, which will vary by
34    formation, with counts of wells, which may or may not be clearly labeled by or associated with
3 5    specific formations. Temporal variability in wastewater generation would also be difficult to
36    capture and would add to uncertainty. Such an approach, however, may be attempted for order of
37    magnitude estimates if the necessary data are available and reliable.
                   This document is a draft for review purposes only and does not constitute Agency policy.
      June 2015                                    8-9                  DRAFT—DO NOTE CITE OR QUOTE

-------
      Hydraulic Fracturing Drinking Water Assessment              Chapter 8 - Wastewater Treatment and Waste Disposal


      8.3.  Wastewater Characteristics
 1    Along with wastewater volume, wastewater characteristics are important for understanding the
 2    potential impacts of management and disposal of hydraulic fracturing wastewater on drinking
 3    water resources. Chapter 7 provides in-depth detail on produced water chemistry. This section
 4    provides brief highlights of the important features of wastewater composition as well as the
 5    characteristics of the residuals produced during wastewater treatment

      8.3.1. Wastewater
 6    This section briefly discusses why the composition of hydraulic fracturing wastewaters needs to be
 7    considered when planning for wastewater management, especially if treatment or reuse are
 8    planned. Concerns associated with selected constituents are presented; treatment considerations
 9    associated with various wastewater constituents are included in Section 8.5.

      8.3.1.1.  Total Dissolved Solids and Inorganics
10    Wastewaters are generally high in total dissolved solids (TDS), especially waters from shale and
11    tight sandstone formations, with values ranging from less than 1,000 mg/L to hundreds of
12    thousands of mg/L (see Section 7.6.4 and Table 7-4). The TDS in wastewaters from shale
13    formations is typically dominated by sodium and chloride and may also include elevated
14    concentrations of bromide, bicarbonate, sulfate, calcium, magnesium, barium, strontium, radium,
15    organics, and heavy metals [Chapman etal.. 2012: Rowan etal.. 2011: Blauchetal.. 2009: Orem et
16    al.. 2007: Sirivedhin and Dallbauman. 2004). Within each play, the minimum and maximum values
17    shown in Table 7-4 suggest spatial variation that may need to be accommodated when considering
18    management strategies such as reuse or treatment. In contrast to shales and sandstones, TDS
19    values for wastewater from CBM formations are generally lower, with concentrations ranging from
20    approximately 250 mg/L to 39,000 mg/L [Benko and Drewes. 2008: Van Voast. 2003] (see
21    Appendix Table E-3). This results in fewer treatment challenges and a wider array of management
22    options.

23    Although TDS has a secondary maximum contaminant level (MCL) (secondary MCLs are non-
24    mandatory water quality standards) of 500 mg/L for aesthetic purposes, it is not considered a
2 5    health-based contaminant and is therefore not regulated under the EPA's National Primary
26    Drinking Water Regulations, although other standards may apply. For example, a maximum
27    concentration of 500 mg/L has been used by the state of Pennsylvania for some industrial
28    wastewater discharges. Constituents commonly found in TDS from hydraulic fracturing
29    wastewaters may have potential impacts on health or create burdens on downstream drinking
3 0    water treatment plants if discharged  at high concentrations to drinking water resources. Bromide,
31    for example, can contribute to the increased formation of disinfection by-products (DBFs) during
32    drinking water treatment (Hammer and VanBriesen. 2012): see Section 8.6.1.

33    Metals (e.g., barium, cadmium, chromium, lead, copper, manganese, nickel, thallium, and zinc)
34    present in TDS can be toxic to humans and aquatic life at certain concentrations. Health effects of
35    these metals can include kidney damage, liver damage, skin conditions, high blood pressure, and
36    developmental problems (U.S. EPA, 2015i). To ensure safe drinking water, the EPA has established
                   This document is a draft for review purposes only and does not constitute Agency policy.
      June 2015                                   8-10                 DRAFT—DO NOTE CITE OR QUOTE

-------
      Hydraulic Fracturing Drinking Water Assessment               Chapter 8 - Wastewater Treatment and Waste Disposal


 1    primary MCLs for a number of these constituents. MCLs and action levels for these metals vary from
 2    0.002 mg/L for thallium to 1.3 mg/L for copper [U.S. EPA. 20151). Cadmium has been found in
 3    produced water from tight gas formations at concentrations as high as 0.37 mg/L (the MCL is 0.005
 4    mg/L), and chromium has been found at concentrations up to 0.265 mg/L (the MCL is 0.1 mg/L)
 5    (see Table 7-4).

 6    Other constituents of concern among dissolved solids are chloride, sulfate, barium, and boron.
 7    Elevated concentrations of chloride and sulfate are of concern because of drinking water aesthetics,
 8    and the EPA has established secondary MCLs for both chloride and sulfate of 250 mg/L (U.S. EPA,
 9    20151: Hammer and VanBriesen. 2012). Barium has a primary MCL of 2 mg/L and has been found in
10    some shale gas produced waters at concentrations in the thousands of mg/L (see Table 7-4). Boron
11    is not regulated under the National Primary Drinking Water Regulations, but internal plant
12    specifications for one CWT (e.g., the Pinedale Anticline Facility) and waste discharge requirements
13    (WDR) permit for another (e.g., San Ardo Water Reclamation Facility) limit boron effluent
14    concentrations to 0.75 mg/L (Shafer. 2011: Webb etal.. 20091.

      8.3.1.2.  Organics
15    Less information is available about organic constituents in hydraulic fracturing waste waters than
16    about inorganic constituents, but there are several studies that include some analyses of organic
17    constituents. The organic content in flowback waters can vary based on the chemical additives used
18    and the formation but generally consists of polymers, oil and grease, volatile organic compounds
19    (VOCs), and semi-volatile organic compounds (SVOCs) fWalsh. 2013: Hayes. 20091. Examples of
20    other constituents detected include alcohols, naphthalene, acetone, and carbon disulfide (U.S. EPA.
21    20151) (see Appendix Table E-10).  Wastewater associated with CBM wells may have high
22    concentrations of aromatic and halogenated organic contaminants that that may require treatment
23    depending on how the waste water will be managed or disposed of (Pashinetal., 2014: Sirivedhin
24    and Dallbauman. 2004). Concentrations of BTEX (benzene, toluene, ethylbenzene, and xylenes), in
25    CBM produced waters are, however, lower than in shale produced waters (see Appendix Table E-9).

2 6    Certain organic compounds are of concern in drinking water because they can cause damage to the
2 7    nervous system, kidneys, and/or liver and can increase the risk of cancer if ingested over a period
28    of time (U.S. EPA, 2006). Some organics in chemical  additives are known carcinogens, including 2-
29    butoxyethanol (2BE), naphthalene, benzene, and polyacrylamide (Hammer and VanBriesen, 2012).
30    Many organics are regulated for drinking water under the National Primary Drinking Water
31    Regulations. Section 8.6.4 provides further discussion of documented or potential situations in
32    which organic constituents have or might reach drinking water resources.

      8.3.1.3.  Radionudides
3 3    Radionudides are constituents of concern in some hydraulic fracturing waste waters, with most
34    available data obtained for the Marcellus Shale in Pennsylvania (see Appendix Table E-8). Results
35    from a USGS report (Rowan etal., 2011) indicate that radium-226 and radium-228 are the
36    predominant radionuclides in Marcellus Shale wastewater, and they account for most of the gross
37    alpha and gross beta activity in the waters studied. There are limited data on radionuclides in
38    wastewater from formations other than the Marcellus Shale, but information on the naturally

                   This document is a draft for review purposes only and does not constitute Agency policy.
      June 2015                                   8-11                DRAFT—DO NOTE CITE OR QUOTE

-------
      Hydraulic Fracturing Drinking Water Assessment               Chapter 8 - Wastewater Treatment and Waste Disposal


 1    occurring radioactive material (NORM) in the formations themselves, in particular uranium and
 2    thorium, may suggest the potential for high levels of radionuclides in produced water, especially
 3    where IDS concentrations are also high. Sections 7.5.4 and 7.6.6 provide further information on
 4    radionuclides in formations and in produced waters.

 5    The primary radioactive contaminants found in hydraulic fracturing wastewaters (radium, gross
 6    alpha radiation, and gross beta radiation)  can increase the risk of cancer if consumed at elevated
 7    levels over time (U.S. EPA. 2015i). Therefore, the EPA has established drinking water MCLs for
 8    combined radium (radium-226 plus radium-228), gross alpha, and gross beta of 5 pCi/L, 15 pCi/L,
 9    and 4 millirems/year, respectively (see Section 8.6.2).

      8.3.2.  Constituents in Residuals
10    Depending on the water being treated and treatment processes used, treatment residuals may
11    consist of sludges, spent media (used filter materials), or brines. Residuals can include constituents
12    such as total suspended solids  (TSS), TDS, metals, radionuclides, and organics. The treatment
13    process tends to concentrate wastewater constituents in the residuals. As an illustration of the
14    degree of concentration that can take place, processes such as electrodialysis and mechanical vapor
15    recompression have been found to yield residuals streams with TDS concentrations in excess of
16    150,000 mg/L,  from treating waters with influent TDS concentrations of approximately 50,000  -
17    70,000 mg/L (Hayes etal.. 2014: Peraki and Ghazanfari. 2014).

18    Also, technologically enhanced naturally occurring radioactive material (TENORM) in wastewaters
19    may cause residual wastes to have elevated gamma radiation emissions  (Kappel etal.. 2013). * One
20    study calculated that typical solids produced by precipitation processes  designed to remove barium
21    and strontium from Marcellus  Shale wastewater would contain between 2,571 and 18,087 pCi/g of
22    radium in the barium sulfate precipitate (Zhang et al., 2014b). Another similar study using mass
23    balances calculated that sludge from a sulfate precipitation process would average a radium
24    concentration of 213 pCi/g in sludge (Silvaetal.. 2012). Silvaetal. (2012) estimated a radium-226
25    concentration of 58 pCi/g in sludge from lime softening processes, a level that would necessitate
26    disposal of low level radioactive waste.

      8.4. Wastewater Management Practices
27    Operators have several strategies  for management of hydraulic fracturing wastewaters (see Figure
28    8-3), with the most common choice being  disposal via Class IID  wells (Clark etal.. 2013: Hammer
29    and VanBriesen. 2012). Other practices include reuse in subsequent hydraulic fracturing operations
30    (with varying levels of treatment), treatment at a CWT (often followed by reuse), evaporation (in
31    arid regions), or in some cases, depending on state and local requirements, various other
32    wastewater management strategies (e.g., irrigation, which involves no discharge to waters of the
33    U.S.). The management methods shown in Figure 8-3 represent various strategies, not all of which
34    will happen together.
      1 Technologically enhanced naturally occurring radioactive materials (TENORM) are radionuclides that have been
      concentrated or enhanced as the result of human activity.
                   This document is a draft for review purposes only and does not constitute Agency policy.
      June 2015                                    8-12                 DRAFT—DO NOTE CITE OR QUOTE

-------
      Hydraulic Fracturing Drinking Water Assessment
Chapter 8 - Wastewater Treatment and Waste Disposal
 1    At one time, treatment of unconventional oil and gas waste waters at publicly owned treatment
 2    works (POTWs) was a common practice for wastewater management in the Marcellus region [Lutz
 3    etal.. 2013). However, this practice has been essentially discontinued following a request from PA
 4    DEP that, by May 19, 2011, oil and gas operators stop sending Marcellus Shale wastewater to 15
 5    POTWs and CWTs that discharged to surface waters [U.S. EPA. 2015h1.

 6    Each of these wastewater management strategies may potentially lead to an impact on drinking
 7    water resources during some phase of their execution. Such impacts may include accidental
 8    releases during transport (see Chapter 7), discharges of treated wastewaters from CWTs or POTWs
 9    where treatment for certain constituents has been inadequate, migration of constituents in
10    wastewaters that have been applied to land, leakage from on-site storage pits (see Chapter 7),
11    inappropriate management of residuals (e.g., leaching from landfills or land application), or
12    accumulation of constituents in sediments near outfalls of CWTs or POTWs that have treated
13    hydraulic fracturing wastewater.

14    A reliable census of nationwide wastewater management practices is difficult to assemble due to a
15    lack of consistent and comparable data among states, but Table 8-2 illustrates the variability in the
16    primary wastewater management methods using available qualitative and quantitative sources.
17    Disposal via underground injection predominates in most regions. Reuse is most prevalent in the
18    Appalachian Basin in Pennsylvania. Moderate reuse occurs in the Arkoma (OK, AR) and Anadarko
19    (OK, TX) basins, and use of CWTs occurs predominantly in Pennsylvania.

                                                                             Road spreading,
                                                                               other uses
                                                                               Surface
                                                                                water
                                                                              discharge
                                                                               Landfill
                                                                              (residuals)
      Figure 8-3. Schematic of wastewater management strategies.
                   This document is a draft for review purposes only and does not constitute Agency policy.
      June 2015                                   8-13                 DRAFT—DO NOTE CITE OR QUOTE

-------
Hydraulic Fracturing Drinking Water Assessment
                                                          Chapter 8 - Wastewater Treatment and Waste Disposal
Table 8-2. Hydraulic fracturing wastewater management practices in recent years.
           Source: (U.S. EPA, 2015q).
Basin
Michigan
Appalachian
Anadarko
Arkoma
Fort Worth
Permian
TX-LA-MS Salt
Formation
Antrim
Marcellus/Utica (PA)
Marcellus/Utica (WV)
Marcellus/Utica (OH)
Granite Wash
Mississippi Lime
Woodford; Cana;
Caney
Fayetteville
Barnett
Avalon/Bone Springs,
Wolfcamp, Spraberry
Haynesville
Resource type
Shale Gas
Shale Gas
Shale Gas/Oil
Shale Gas/Oil
Tight Gas
Tight Oil
Shale Gas/Oil
Shale Gas
Shale Gas
Shale/tight Oil/gas
Tight Gas
Reuse

XXX
XXX
XX
XX
X
X
XX
X
X
X
Injection
for disposal
XXX
XX
XX
XXX
XXX
XXX
XXX
XX
XXX
XXX
XXX
CWT
facilities

XX
X
X
xa

xa
xa
xa
xa

Notes

Limited Class IID wells in east

Reuse limited but is being evaluated

Few existing Class IID wells; new CWT
facilities are under construction
Reuse not typically effective due to
high TDS early in flowback and
abundance of Class IID wells

Reuse not typically cost effective due
to high TDS early in flowback and
abundance of Class IID wells
Available datab
Qualitative
Quantitative
Quantitative
Mixed
Mixed
Qualitative
Qualitative
Mixed
Mixed
Mixed
Mixed
June 2015
This document is a draft for review purposes only and does not constitute Agency policy.
                                8-14
DRAFT—DO NOTE CITE OR QUOTE

-------
Hydraulic Fracturing Drinking Water Assessment
                                                                Chapter 8 - Wastewater Treatment and Waste Disposal
Basin
West Gulf
Denver
Julesburg
Piceance;
Green River
Williston
Formation
Eagle Ford, Pearsall
Niobrara
Mesaverde/Lance
Bakken
Resource type
Shale Gas/Oil
Shale Gas/Oil
Tight Gas
Shale Oil
Reuse
X
X
X
X
Injection
for disposal
XXX
XXX
XX
XXX
CWT
facilities
X
X
X

Notes


Also managed through evaporation to
atmosphere in ponds in this region
Reuse limited but is being evaluated
Available datab
Mixed
Mixed
Qualitative
Mixed
 a CWT facilities in these formations are operator owned.
 b This column indicates the type of data on which EPA based the number of X's. In most cases, EPA used a mixture of qualitative and quantitative data sources along with
 engineering judgment to determine the number of X's.
 XXX—The majority (> 50%) of wastewater is managed with this management practice.
 XX—A moderate portion (> 10% and < 50%) of wastewater is managed with this management practice.
 X—This management practice has been documented in this location but for a small (< 10%) or unknown percent of wastewater.
June 2015
This document is a draft for review purposes only and does not constitute Agency policy.
                                   8-15
DRAFT—DO NOTE CITE OR QUOTE

-------
      Hydraulic Fracturing Drinking Water Assessment               Chapter 8 - Wastewater Treatment and Waste Disposal


 1    Management choices are affected by cost and a number of other factors, including the chemical
 2    properties of the wastewater; the volume, duration, and flow rate of the water generated; the
 3    logistical feasibility of various options; the availability of necessary infrastructure; federal, state,
 4    and local regulations; and operator discretion [U.S. GAP, 2012: NPC, 2011a]. The economics (such
 5    as transport, storage, and disposal costs) and availability of various treatment and disposal
 6    methods are of primary importance [U.S. GAP, 2012]. For example, as of early 2015, Pennsylvania
 7    has nine operating Class IID wells within the state, whereas Texas has nearly 7,900 [U.S. EPA,
 8    2015qj.

 9    The availability and use of management strategies may change in a region over time as oil and gas
10    development increases or decreases, changing the volumes of wastewater that need to be handled
11    on a local, state, and regional level (see Text Box 8-1 for more information on hydraulic fracturing
12    wastewater management in Pennsylvania). Figure 8-4 illustrates shifting wastewater management
13    practices in Pennsylvania over the last several years as shale gas development has proceeded in the
14    Marcellus Shale. On-site reuse (labeled as "Reuse HF" in Figure 8-4) has grown. Also, most CWT
15    managementof Marcellus wastewater in recent years has been at zero-discharge facilities (i.e., for
16    reuse) (an estimated 80% in 2012 and 90% in 2013) (PADEP.  2015a). Combined with the volumes
17    managed via on-site reuse, Pennsylvania reuse rates are approximately 85% to 80%. In contrast,
18    wastewater disposal data for Colorado (see Figure 8-5) show a steady use of injection wells
19    (injected on lease) since 2000, and an apparent decrease in the use of onsite pits (state data were
20    filtered for formations indicated in the literature to be targets for hydraulic fracturing).
                   This document is a draft for review purposes only and does not constitute Agency policy.
      June 2015                                    8-16                 DRAFT—DO NOTE CITE OR QUOTE

-------
      Hydraulic Fracturing Drinking Water Assessment               Chapter 8 - Wastewater Treatment and Waste Disposal
      Text Box 8-1. Temporal Trends in Wastewater Management - Experience of Pennsylvania.

 1    Gross natural gas withdrawals from shale formations in the United States increased 518% between 2007 and
 2    2012 [EIA. 2014c). This production increase has led to larger volumes of wastewater that require appropriate
 3    management [Vidic etal.. 2013: Gregory et al.. 2011: Kargbo etal.. 2010]. The rapid increase in wastewater
 4    generated from oil and gas wells used for hydraulic fracturing has led to many changes in the wastewater
 5    disposal practices in the oil and gas industry. Changes have been most evident in Pennsylvania, which has
 6    experienced more than a 1,400% increase in natural gas production since 2000 [EIA. 2014c).

 7    Lutz etal. [20131 estimated that total wastewater generation in the Marcellus region increased 570%
 8    between 2004 and 2013 and concluded that this increase has created stress on the existing wastewater
 9    disposal infrastructure. In 2010, in response to concerns over elevated TDS in the Monongahela River basin
10    and studies linking high TDS (and in particular high bromide levels) to elevated DBF levels in drinking water
11    systems [PA DEP. 2011a). PA DEP amended Chapter 95 Wastewater Treatment Requirements under the
12    Clean Streams Law for new discharges of TDS in wastewaters. This regulation is also known informally as the
13    2010 TDS regulation. The regulation disallowed any new indirect discharges (i.e., discharges to POTWs) of
14    hydraulic fracturing waste and set limits of treated discharges fromCWTs of 500 mg/LTDS, 250mg/L
15    chloride, 10 mg/L barium, and 10 mg/L strontium. Existing discharges were exempt.

16    In April 2011, PA DEP requested that oil and gas well operators transporting unconventional wastewater to
17    the eight CWTs and seven POTWs that were exempt from the 2010 TDS regulation voluntarily stop
18    discharging to these facilities. Follow-up letters from PA DEP to the owners of the wells specified that the role
19    of bromides from Marcellus Shale wastewaters in the formation of total trihalomethanes (TTHM)  was of
20    concern [PA DEP. 2011a).

21    Between early 2011 and late 2011, although reported wastewater flows more than doubled, Marcellus
22    drilling companies in Pennsylvania reduced their wastewater flows to CWTs that were exempt from the 2010
23    TDS regulation by 98%, and discharge to POTWs was Virtually eliminated' [Hammer and VanBriesen. 20121.

24    Along with the decreased discharges from POTWs, there has been increased reuse of wastewater in the
25    Marcellus Shale region. From 2008-2011,  reuse of Marcellus wastewater has increased, POTW treatment
26    volumes have decreased, tracking of wastewater has improved, and wastewater transportation distances
27    have decreased [Rahmetal.. 2013]. Maloney and Yoxtheimer [2012] analyzed data from 2011 and found that
28    reuse of flowback water increased to 90% by volume. Disposed flowback water comprised 8% of the total
29    volume. Brine water, which was defined as formation water, was reused (58%), disposed via injection well
30    (27%), or sent to industrial waste treatment plants (14%). Of all the fluid wastes in the analysis, brine water
31    was most likely to be transported to other states (28%). They also concluded that wastewater disposal to
32    municipal sewage treatment plants declined nearly 100% from the first half of 2011 to the second half.
                    This document is a draft for review purposes only and does not constitute Agency policy.
      June 2015                                      8-17                  DRAFT—DO NOTE CITE OR QUOTE

-------
Hydraulic Fracturing Drinking Water Assessment
Chapter 8 - Wastewater Treatment and Waste Disposal
           2009 - 2010
                                  0.5%
                                                                     uic

                                                                   • CWT

                                                                   • POTW

                                                                     Reuse Non-HF

                                                                   • Reuse HF

                                                                   • Road Spreading

                                                                     Other

                                                                   « Uncertain
                          4%
                 2013
                           0.0003%  0-5%
                                                                uic

                                                                I CWT

                                                                Reuse Non-HF

                                                                lReuse HF

                                                                I Road Spreading

                                                                Other
Figure 8-4. Percentages of Marcellus Shale wastewater managed via various practices for
          (top) the second half of 2009 and first half of 2010 (total estimated volume of 216
          Mgal), and (bottom) 2013 (total estimated volume of 1.3 billion gallons).
          "Reuse HF" indicates on-site reuse. Source: Waste data from PA PEP (2015a).
              This document is a draft for review purposes only and does not constitute Agency policy.
June 2015                                     8-18                  DRAFT—DO NOTE CITE OR QUOTE

-------
      Hydraulic Fracturing Drinking Water Assessment               Chapter 8 - Wastewater Treatment and Waste Disposal
             70%
             60%
             50%
             40%
             30%
             20%
             10%
              0%
                   2000  2001  2002  2003 2004  2005  2006  2007  2008  2009 2010 2011  2012  2013
                 Central disposal pit or well      —Commercial disposal facility     -  Injected on lease
                •Onsite pit                    —Surface discharge

      Figure 8-5. Management of wastewater in Colorado in regions where hydraulic fracturing is
                being performed.
                Source: Production data from COGCC f20151.

 1    Regulations also affect management options and vary geographically. At the Federal level, existing
 2    oil and gas effluent limitations guidelines and standards (ELGs) can be found under 40 CFR Part
 3    435. These ELGs apply to conventional and unconventional oil and gas extraction facilities in
 4    various subcategories (e.g., Offshore, Onshore, Stripper Wells), with the exception of CBM
 5    discharges, which are not subject to the existing regulations. Subpart C, the Onshore subcategory,
 6    prohibits the discharge of wastewater pollutants to waters of the U.S. from onshore oil and gas
 7    extraction facilities. This "zero-discharge standard" means that oil and gas produced water
 8    pollutants cannot be directly discharged to surface waters. Operators have met these regulations
 9    through underground injection, reuse, or transfer of produced water to POTWs and/or CWTs. West
10    of the 98th meridian  (the arid western portion of the continental United States), discharges of
11    wastewater from onshore oil and gas  extraction facilities may be permitted for direct discharge to
12    waters of the U.S. if the produced water has a use in agriculture or wildlife propagation when
13    discharged into navigable waters. Definitions in 40 CFR 435.51(c) explain that the term "use in
14    agricultural or wildlife propagation" means that (1) the produced water is of good enough quality
15    to be used for wildlife or livestock watering or other agricultural uses; and (2) the produced water
16    is actually put to such use during periods of discharge. The regulations at 40 CFR 435.52 specify
17    that the  only allowable discharge is produced water, with an oil and grease concentration not
18    exceeding 35 milligrams per liter (mg/L). The regulations prohibit the discharge of waste
19    pollutants into navigable waters from any source (other than produced water) associated with
                    This document is a draft for review purposes only and does not constitute Agency policy.
      June 2015                                     8-19                  DRAFT—DO NOTE CITE OR QUOTE

-------
      Hydraulic Fracturing Drinking Water Assessment               Chapter 8 - Wastewater Treatment and Waste Disposal


 1    production, field exploration, drilling, well completion, or well treatment (i.e., drilling muds, drill
 2    cuttings, produced sands).

 3    Unpermitted discharges of wastes related to hydraulic fracturing have been described in a number
 4    of instances. In Pennsylvania, discharges of brine into a storm drain that discharges to a tributary of
 5    the Mahoning River in Ohio. Analyses of the brine and drill cuttings that were discharged indicated
 6    the presence of contaminants, including benzene and toluene [U.S. Department of Justice, 2014]. In
 7    California, an oil production company periodically discharged hydraulic fracturing wastewaters to
 8    an unlined sump for 12 days. It was concluded by the prosecution that the discharge posed a threat
 9    to groundwater quality [Bacher, 2013]. These unauthorized discharges represent both documented
10    and potential impacts on drinking water resources. However, data do not exist to evaluate whether
11    such episodes are uncommon or whether they happen on a more frequent basis and remain largely
12    undetected.

13    The following section provides an overview of hydraulic fracturing wastewater management
14    methods, with some discussion of the geographic and temporal variations in practices. Discussion is
15    provided on common treatment and disposal methods including on-site storage, underground
16    injection, CWTs, reuse of hydraulic fracturing fluids, and evaporation methods. This section also
17    provides discussion on past treatment of hydraulic fracturing wastewater at POTWs. Other
18    management practices are also covered. Brief descriptions of treatment technologies applicable to
19    hydraulic fracturing wastewater are available in Appendix F.

      8.4.1.  Underground Injection
20    Oil and gas wastewater may be disposed of via Class II injection wells regulated under the
21    Underground Injection Control (UIC] Program under the Safe Drinking Water Act (SDWA]1. Class II
22    wells include those used for enhanced oil recovery (IIR], disposal (IID], and hydrocarbon storage
23    (HH). Nationwide, injection wells dispose of a large fraction of wastewater from the oil and gas
24    industry,  including wastewater associated with hydraulic fracturing. A 2009  study notes that the oil
25    and gas industry in the United States generated about 882 billion gal (21  billion bbl or 3.34 trillion
26    L] of produced water in  2007 (Clark and Veil. 2009]. More than 98% of this volume was managed
27    via some form of underground injection, with 40% injected into Class II wells. However, a good
28    national estimate of the  amount of hydraulic fracturing wastewater injected into Class II wells is
29    difficult to develop due to lack of available on data injection volumes specific to hydraulic fracturing
30    operations that are compiled and able to be compared among states. Also, wastewater management
31    methods are not well tracked in all states. Regional numbers of Class IID wells and generally low
32    reuse rates  (see Section  8.4.3], however, are consistent with Class IID wells being a primary means
33    of wastewater management in many areas with hydraulic fracturing activity.

34    This assessment does not address whether there are documented or potential impacts on drinking
3 5    water resources associated with the injection of hydraulic fracturing wastewaters into Class IID
3 6    wells. However, should the feasibility of managing hydraulic fracturing wastewater via
      1 States may be given federal approval to run a UIC program under Section 1422 or 1425 of SDWA.


                   This document is a draft for review purposes only and does not constitute Agency policy.
      June 2015                                    8-20                 DRAFT—DO NOTE CITE OR QUOTE

-------
      Hydraulic Fracturing Drinking Water Assessment
Chapter 8 - Wastewater Treatment and Waste Disposal
 1    underground injection be limited in any way or become less economically advantageous, operators
 2    will likely adjust their wastewater management programs to favor other local practices such as
 3    treatment and discharge or reuse. Any new wastewater management decisions would then have to
 4    be evaluated in terms of potential impacts on drinking water resources.

 5    The decision to inject hydraulic fracturing wastewater into Class IID wells depends, in part, on cost
 6    and on the proximity of the production well to the disposal well (and, therefore, transportation
 7    costs). For oil and gas producers, underground injection is usually the least expensive management
 8    strategy unless significant trucking is needed to transport the wastewater to a disposal well [U.S.
 9    GAP. 20121.

10    Class IID wells are not distributed uniformly among states due to differences in geology (including
11    depth and permeability of formations), permitting, and historical demand for disposal of oil and gas
12    wastewater. Table 8-3 shows the numbers of active Class IID wells across the United States, with
13    the total count at a little over 27,000. The greatest numbers of wells are found in Texas, Oklahoma,
14    and Kansas. For example, Texas has nearly 7,900 Class IID wells, with an estimated daily disposal
15    volume of approximately 400 million gal per day (MGD) (1.5 billion L/day) (see Table 8-3). This
16    large disposal capacity in Texas is consistent with the availability of formations with suitable
17    geology and the demand for wastewater disposal associated with a mature and active oil and gas
18    industry. In contrast, Class IID wells are a relatively small portion of Marcellus wastewater
19    managementin Pennsylvania (about 10% in 2013 and the firsthalf of 2014) fPADEP. 2015al
20    because the state has nine injection wells as of early 2015. Wastewater is generally transported out
21    of state when being managed through injection into Class IID wells. The local availability of Class
22    IID wells and the capacity to accept large volumes of wastewater may begin to be affected by recent
23    state actions concerning seismic activity associated with injection (U.S. EPA, 2014F).
      Table 8-3. Distribution of active Class IID wells across the United States.
               Source: U.S. EPA(2015q).

State
Nearby basins with
hydraulic fracturing
Number of active
Class IID wells
(2012-2014)
Average disposal
rate per well
(GPD/well)a
Total state
disposal rate
(MGD)
                   This document is a draft for review purposes only and does not constitute Agency policy.
      June 2015                                    8-21                  DRAFT—DO NOTE CITE OR QUOTE

-------
Hydraulic Fracturing Drinking Water Assessment
Chapter 8 - Wastewater Treatment and Waste Disposal
State
AK
OH
WV
PA
NY
VA
TN
MD
NC
KS
OK
AR
MO
CO
WY
UT
NE
TX
NM
IN
IL
KY
Ml
CA
Nearby basins with
hydraulic fracturing
North Slope
Appalachian
Multiple basins
Cherokee, Anadarko, Arkoma
Denver-Julesburg, Green
River, Piceance, Uinta
Fort Worth, Western Gulf,
Permian, San Juan, Raton
Illinois
Michigan
San Joaquin
Number of active
Class MD wells
(2012-2014)
45
188
66
9
10b
12
0
0
0
5,516
4,622°
611d
11
294
330
109
113
7,876
736
183
1,054
58
779°
826
Average disposal
rate per well
(GPD/well)a
182,000
8,900
7,180
6,380
3,530
17,500
0
0
0
20,900
35,900
30,900
1,270
50,200
e
74,400
18,100
54,200
48,600
3,580
e
1,750
16,600
77,800
Total state
disposal rate
(MGD)
8.2
1.7
0.47
0.057
0.035
0.21
0
0
0
120
170
19
0.014
15
e
8.1
2.0
430
36
0.66
e
0.10
13
64
                This document is a draft for review purposes only and does not constitute Agency policy.
June 2015                                          8-22                    DRAFT—DO NOTE CITE OR QUOTE

-------
      Hydraulic Fracturing Drinking Water Assessment
Chapter 8 - Wastewater Treatment and Waste Disposal
State
LA
MS
AL
ND
MT
SD
Nearby basins with
hydraulic fracturing
TX-LA-MS Salt
Williston
All other states (NV, FL, OR, IA, and WA)f
Total (not including missing states)
Number of active
Class IID wells
(2012-2014)
2,448
499
85
395
199
21
42
27,137
Average disposal
rate per well
(GPD/well)a
42,100
69,500
44,200
31,600
31,100
10,200
89,400
40,400
Total state
disposal rate
(MGD)
100
35
3.8
12
6.2
0.21
3.8
1,040
       a Typical injection volumes per well are based on historical annual volumes for injection for disposal divided by the number of
       active Class IID wells during the same year (primarily data from 2007 to 2013).
       bThese wells are not currently permitted to accept unconventional oil and gas extraction wastewater.
       c With the exception of Oklahoma and Michigan, wells on tribal lands have not been intentionally included. Wells on tribal
       lands may be counted if state databases contained them.
       d Only 24 of the 611 active Class II wells in Arkansas are in the northern half of the state, close to the Fayetteville formation.
       e Disposal rates and/or number of Class IID wells is unknown.
       These are states that have minimal oil and gas activity. The number of wells shown for these states may include all types of
       Class II wells (e.g., Class II enhanced recovery wells) and therefore is an upper estimate. All other states not listed in this table
       have minimal oil and gas activity and no active Class IID wells.
      8.4.2.  Centralized Waste Treatment Facilities
 1    A CWT facility is generally defined as a facility that accepts industrial materials (hazardous, non-
 2    hazardous, solid, or liquid) generated at another facility (off-site) for treatment and/or recovery
 3    (EPA. 2000). (A POTW treats local municipal wastewater.) As a group, CWTs that accept oil and gas
 4    wastewater offer a wide variety of treatment capabilities and configurations. The fate of treated
 5    effluent at CWTs also varies, and can include the following: reuse in fracturing operations, direct
 6    discharge (to a receiving water under a National Pollution Discharge Elimination System (NPDES)
 7    permit), indirect discharge (to a POTW), or a combination of these. Zero discharge facilities do not
 8    discharge to either surface water or a POTW; effluent is generally used for reuse, although
 9    evaporation or land application may also be done. Some CWTs may be configured so that they only
10    partially treat the waste stream if allowed by the end use  (a reuse application that does not require
11    TDS removal). Potential impacts on drinking water resources associated with treatment in CWTs
12    will depend upon whether the CWT treats adequately for  constituents of concern prior to discharge
13    to surface water or to a POTW, and whether treatment residuals are managed appropriately.

14    Clean Water Act (CWA) regulations only apply to facilities that discharge treated wastewater to
15    surface waters or POTWs. For zero-discharge facilities, Pennsylvania and Texas have adopted
                     This document is a draft for review purposes only and does not constitute Agency policy.
      June 2015                                       8-23                  DRAFT—DO NOTE CITE OR QUOTE

-------
      Hydraulic Fracturing Drinking Water Assessment               Chapter 8 - Wastewater Treatment and Waste Disposal


 1    regulations to control permitting. PA DEP issues permits (General Permit WMGR123) thatallow
 2    zero-discharge CWTs to treat and release water back to oil and gas industries for reuse (see the
 3    Eureka Resources Facility in Williamsport, PA listed in Table 8-7 as an example of a zero-discharge
 4    facility1). The Texas Railroad Commission (TXRRC) regulates and categorizes wastewater recycling
 5    facilities into different categories: off-lease commercial recycling facilities (capable of being moved
 6    from one location to another) and stationary commercial recycling facilities. The Texas regulations
 7    also promote oil and gas wastewater treatment for reuse and water sharing (see
 8    http://www.rrc.state.tx.us/rules/rule.php).

 9    Wastewater from hydraulically fractured wells can be transported by truck or pipeline to and from
10    a CWT (Easton. 2014): this may present a vulnerability for spills or leaks (see Chapter 7). The
11    treated wastewater from CWTs may be integrated with other sources of water (for example, treated
12    municipal wastewater, storm water drainage, or other treated industrial waste streams) for reuse
13    applications (Easton, 2014).

      8.4.2.1. Numbers and Locations of CWTs
14    Although there are CWTs serving hydraulic fracturing operations throughout the country, including
15    the Barnett and Fayetteville shale plays plus oil fields in Texas and Wyoming, historically the
16    majority have served Marcellus Shale operations. This  is likely because the low availability of
17    injection wells (Boschee. 2014) in Pennsylvania necessitates other forms of management. An EPA
18    study (U.S. EPA. 2015q) identified 73 CWT facilities that have either accepted or plan to accept
19    hydraulic fracturing wastewater (see Table 8-4). Of these, 39 are located in Pennsylvania. Most of
20    these are zero-discharge facilities; they do not discharge to surface waters  or POTWs, and they
21    often do not include TDS removal. According to EPA research (U.S.  EPA. 2015q). the number of CWT
22    facilities serving operators in the Marcellus and Utica Shales has increased since the mid-2000s as
23    the number of wells drilled in the Marcellus and Utica Shales has increased, growing from roughly
24    five CWTs in 2004 to over 40 in 2013. A similar trend has been noted for the  Fayetteville Shale
25    region in Arkansas, where there are fewer Class IID wells available relative to the rest of the state
26    (U.S. EPA. 2015q).

27    In other regions, a small number of newer facilities have emerged in the last several years, most
28    often with TDS removal capabilities. In Texas, for example, two zero-discharge facilities are
29    available to treat wastewater from the Eagle Ford (beginning in 2011 and 2013), both equipped
30    with TDS removal, and one zero-discharge facility with TDS removal is located in the Barnett Shale
31    region (operational beginning in 2008). In Wyoming, the four facilities in the region of the
32    Mesaverde/Lance formations (operations beginning between 2006 and 2012; two zero-discharge
33    and two with multiple discharge options) are all capable of TDS removal (U.S. EPA. 2015q).
      1 The facility is also permitted for indirect discharge to the Williamsport Sewer Authority.


                   This document is a draft for review purposes only and does not constitute Agency policy.
      June 2015                                    8-24                 DRAFT—DO NOTE CITE OR QUOTE

-------
     Hydraulic Fracturing Drinking Water Assessment
Chapter 8 - Wastewater Treatment and Waste Disposal
     Table 8-4. Number, by state, of CWT facilities that have accepted or plan to accept
               wastewater from hydraulic fracturing activities.
               Source: U.S. EPA(2015q).
State
AR
CO
ND
OH
OK
PA
TX
WV
WY
Formation(s)
served where
hydraulic
fracturing
occurs
Fayetteville
Niobrara,
Piceance Basin
Bakken
Utica, Marcellus
Woodford
Utica, Marcellus
Eagle Ford,
Barnett, Granite
Wash
Marcellus, Utica
Mesaverde,
Lance
Total
Zero-discharge
CWT facilities3
Non-TDS
removal
treatment
2
3(1)
0
10(7)
2
23
1
4(2)
0
45
TDS
removal
treatment
0
0
1(1)
0
0
7(3)
3
0
2
13
CWT facilities that
discharge to surface
water or POTWa
Non-TDS
removal
treatment
0
0
0
1
0
6
0
0
0
7
TDS
removal
treatment
0
0
0
0
0
0
0
0
0
0
Discharging CWT
facilities with multiple
discharge options3
Non-TDS
removal
treatment
0
0
0
0
0
0
0
1
0
1
TDS
removal
treatment
1
0
0
0
0
3(1)
0
1
2
7
Total
known
facilities
3
3
1
11
2
39
4
6
4
73
     a Number of facilities also includes facilities that have not yet opened but are under construction, pending permit approval, or
     in the planning stages. Facilities that are not accepting process wastewater from hydraulic fracturing activities but plan to in
     the future are noted parenthetically.
1    Because few states maintain a comprehensive list of CWT facilities and the count provided by the
2    EPA [U.S. EPA. 2015q] includes facilities that plan to accept unconventional oil and gas
3    wastewaters, the data in Table 8-4 do not precisely reflect the number of facilities currently
4    available for handling hydraulic fracturing wastewaters. Additional discussion of CWTs in
5    unconventional oil and gas fields are reviewed in the literature for areas including the Barnett
6    [Hayes and Severin. 2012b] and the Fayetteville [Veil. 2011] as well as other oil fields in Texas and
7    Wyoming [Boschee, 2014, 2012]. In addition, news releases and company announcements indicate
8    that new wastewater treatment facilities are being planned [Greenhunter, 2014: Geiver, 2013:
9    Purestream. 2013: Alanco.  2012: Sionix. 2011].
                   This document is a draft for review purposes only and does not constitute Agency policy.
     June 2015                                     8-25                  DRAFT—DO NOTE CITE OR QUOTE

-------
      Hydraulic Fracturing Drinking Water Assessment              Chapter 8 - Wastewater Treatment and Waste Disposal


 1    Based on oil and gas waste disposal information available from PA DEP [PA PEP, 2015a] dating
 2    back to 2009, the estimated volumes of Marcellus wastewater sent to CWTs range from
 3    approximately 113 Mgal (428 million L) in the latter half of 2009 and first half of 2010, to about
 4    183 Mgal (693 million L) in 2011, and about 252 Mgal (954 million L) in 2013. These constitute
 5    about 52% of the total wastewater volume in 2009-2010, about 25% in 2011, and 20% in 2013,
 6    indicating that although total amounts of wastewater have increased (see Table 8-1), the
 7    percentage managed through CWTs has decreased.

 8    Among the Marcellus wastewater sent to CWTs, an estimated 35% was sent to zero-discharge
 9    facilities in Pennsylvania (those with general permits) in the latter half of 2010, and 42% was sent
10    to facilities with NPDES permits (indicating that they can discharge to surface waters). About 23%
11    went to CWTs whose permit types were more difficult to ascertain, generally outside of
12    Pennsylvania. By 2013, the portion sentto zero-discharge facilities had risen to 90%, with about
13    5% sent to CWTs with NPDES permits and 5% sent to CWTs whose discharge permit type is not
14    clear. The high percentage sent to zero-discharge CWTs is consistent with the concerted focus on
15    reuse in Pennsylvania, although CWTs with NPDES permits also often provide treated wastewater
16    for reuse, further limiting discharges to surface waters. The waste records do not indicate if a CWT
17    has more than one permit type.

      8.4.2.2. Residuals Management
18    Certain treatment processes at CWTs produce liquid or solids residuals as a by-product of that
19    process. The residuals produced depend on the constituents in the treated water and the treatment
20    process used. Residuals can consist of sludges (from precipitation, filtration, settling units, and
21    biological processes); spent media (media requiring replacement or regeneration from filtration,
22    adsorption, or ion exchange processes); concentrated brines (from membrane processes and some
23    evaporation processes); and regeneration and cleaning chemicals (from ion exchange, adsorption,
24    and membrane processes) (Fakhru'1-Razi etal.. 2009). Residuals from CWTs can constitute  a
25    considerable fraction of solid waste in an oil or gas production area. Chiado (2014) found that solid
26    wastes from hydraulic fracturing in the Marcellus accounted for 5% of the weight of waste
27    deposited in landfills in the area, with some area landfills reaching as high as 60% landfill mass
2 8    coming from hydraulic fracturing activities.

2 9    Management of Solid Residuals

3 0    CWTs may apply additional treatment to solid residuals including thickening, stabilization (e.g.,
31    anaerobic digestion), and dewatering processes prior to disposal. The solid residuals are then
32    typically sent to a landfill, land  applied, or incinerated (Morillon et al.. 2002). Pollutants may
33    accumulate in sludge, which may limit land application as a disposal option. For example, wastes
34    containing TENORMs can be problematic due to the possibility of radon emissions from the landfill
35    (Walter etal., 2012). In some states, many landfills that are specifically permitted to accept
36    TENORM have criteria written  into their permits, including gamma exposure rate (radiation) levels
37    and radioactivity concentration limits. Most non-hazardous landfills have limits on maximum
38    radiation that can be accepted.  For example, Pennsylvania requires alarms to be set at all municipal
39    landfills, with a trigger set at 10 u,R/hr above background radiation (Pa Code Title 25, Ch. 273.223

                   This document is a draft for review purposes only and does not constitute Agency policy.
      June 2015                                    8-26                 DRAFT—DO NOTE CITE OR QUOTE

-------
      Hydraulic Fracturing Drinking Water Assessment              Chapter 8 - Wastewater Treatment and Waste Disposal


 1    c). Texas sets a radioactivity limit, requiring that any waste disposed by burial contains less than 30
 2    pCi/g radium or 150 pCi/g of other radionuclides (TX Code Ch 4 Section F Section 4.620). Some
 3    states have volumetric limitations on TENORM in their permits  (e.g., Colorado).

 4    Solid residual wastes have the potential to impact the quality of drinking water resources if
 5    contaminants leach to groundwater or surface water. In a recent study by PA DEP, radium was
 6    detected in leachate from 34 of 51 landfills, with radium-226 concentrations ranging from 54 to
 7    416 pCi/L, and radium-228 ranging from 2.5 to  1,100 pCi/L [PADEP. 2015b). Countess et al. [2014)
 8    studied the potential for barium, calcium, sodium, and strontium to leach from sludges generated at
 9    a CWT handling hydraulic fracturing wastewaters in Pennsylvania. Tests used various strong acid
10    solutions (to simulate the worst case scenario) and weak acid digestions (to simulate
11    environmental conditions).  The extent of leaching varied by constituent and by fluid type; the data
12    illustrate the possibility of leaching of these constituents from landfills.

13    Management of Liquid Residuals

14    Practices for management of liquid residual streams are generally the same as for untreated
15    hydraulic fracturing wastewaters, although the reduced volumes tend to lower costs (Hammer and
16    VanBriesen. 2012). Concentrations  of contaminants, however, will be higher. Liquids mixed with
17    other wastes can be disposed of in landfills if the liquid concentration is low enough. If the liquid is
18    not injected into a disposal well, treatment to remove salts would be required for surface water
19    discharge to meet NPDES permit requirements and protect the water quality for downstream users
20    (e.g., drinking water utilities) (see Section 8.6). Because some constituents of concentrated
21    residuals can pass through or impact municipal wastewater treatment processes (Linaric etal.,
22    2013: Hammer and VanBriesen. 2012). these residuals may not be appropriate for discharge to a
23    POTW. Elevated salt concentrations, in particular, can reduce or inhibit microbiological treatment
24    at municipal wastewater systems such as activated sludge treatment (Linaric et al., 2 013).

      8.4.3.  Water Reuse for Hydraulic Fracturing
25    Water reuse in hydraulic fracturing operations has increased in recent years, with wastewaters
26    being used to formulate hydraulic fracturing fluids for subsequent fracturing jobs (Boschee. 2014,
27    2012: Gregory etal., 2011: Rassenfoss, 2011). Wastewater may be reused after some form of
28    treatment (sometimes only settling), depending on the reuse water quality requirements, and it
29    may be supplied for use in hydraulic fracturing through various routes.  Reused water is discussed
30    in Chapter  4 of this report (Water Acquisition) as well as in this  chapter, though in a different
31    context. The water reuse rate described in this chapter is the amount or percentage of generated
32    wastewater that is managed by being provided to operators for use in additional hydraulic
33    fracturing operations. In contrast, Chapter 4 discusses reused wastewater as a source water and as
34    one part of the base fluid for new fracturing fluid.

35    Hydraulic fracturing wastewater reuse reduces  costs associated with other forms of wastewater
3 6    management, and the economic benefits and feasibility of reuse can be expected to figure into
37    ongoing wastewater management decisions. However, although reuse minimizes other forms of
38    wastewater management on a local and short-term basis (e.g., those involving direct or indirect


                   This document is a draft for review purposes only and does not constitute Agency policy.
      June 2015                                    8-27                DRAFT—DO NOTE CITE OR QUOTE

-------
      Hydraulic Fracturing Drinking Water Assessment              Chapter 8 - Wastewater Treatment and Waste Disposal


 1    discharge to surface waters), reuse can result in the accumulation of dissolved solids (e.g., salts and
 2    TENORMs) as the process returns water to the subsurface. For example, data from a PA DEP study
 3    [PADEP. 2015b) suggests that hydraulic fracturing fluids that include reused wastewater already
 4    contain radium-226 and radium-228. Eventually, wastewaters with a component that has been
 5    reused more than once will need to be definitively managed, either through treatment or injection.
 6    Residuals from treatment will also require proper management to avoid potential impacts on water
 7    resources (see Section 8.4.2.2) (Kappeletal., 2013).

      8.4.3.1. Factors in Considering Reuse
 8    In making the decision whether to manage wastewater via reuse, operators have several factors to
 9    consider (Slutz etal.. 2012: NPC. 2011a):

10        •   Wastewater generation rates compared to water demand for future fracturing operations,
11        •   Wastewater quality and treatment requirements for use in future operations,
12        •   The costs and benefits  of wastewater management for reuse compared with other
13            management strategies,
14        •   Available infrastructure and treatment technologies, and
15        •   Regulatory considerations.
16    Among these factors, costs may  be the most significant driver, weighing the costs of transportation
17    from the generating well to the treatment facility and to the new well against the costs for transport
18    to alternative locations (a disposal well or CWT). Trucking large quantities of water can be
19    relatively expensive (from $0.50 to $8.00 per barrel), rendering on-site treatment technologies and
20    reuse potentially economically competitive in some settings (Dahm and Chapman. 2014: Guerra et
21    al.. 2011). Also, logistics, including proximity of the water sources for aggregation, may be a factor
22    in implementing reuse. For example, Boschee (2014) notes that in the Permian Basin, older
23    conventional wells are linked by pipelines to a central disposal facility, facilitating movement of
24    treated water to areas where it is needed for reuse.

25    Regulatory factors may facilitate reuse. In 2013, the Texas Railroad Commission adopted rules
26    intended to encourage statewide water conservation. These rules facilitate reuse by eliminating the
27    need for a permit when operators reuse on their own lease or transfer the fluids to another
28    operator for use in hydraulic fracturing (Rushton and Castaneda. 2014). Data for the years after
29    2013 will allow evaluation of whether reuse increases.

30    Recommended compositional ranges for base fluid may shift in the future as fracturing fluid
31    technology continues to develop. Development of fracturing mixture additives that are brine-
32    tolerant have allowed for the use of high TDS wastewaters (up to tens of thousands of mg/L) for
33    reuse in fracturing (Tiemann et  al.. 2014: GTI. 2012: Minnich. 2011). Some new fracturing fluid
34    systems are claimed to be able to tolerate salt concentrations exceeding 300,000 mg/L  (Boschee,
35    2014). This greater flexibility in acceptable water chemistry can facilitate reuse both logistically
36    and economically by reducing treatment needs. Additional discussion of the water quality feasible
37    for reuse and examples of recommended constituent concentrations are included in Appendix F.


                   This document is a draft for review purposes only and does not constitute Agency policy.
      June 2015                                    8-28                 DRAFT—DO NOTE CITE OR QUOTE

-------
      Hydraulic Fracturing Drinking Water Assessment
Chapter 8 - Wastewater Treatment and Waste Disposal
 1    Reuse rates may also fluctuate with changes in the supply and demand of treated waste water and
 2    the availability of fresh water. Flowback may be preferable to later-stage produced water for reuse
 3    because it is typically generated in larger quantities from a single location as opposed to water
 4    produced later on, which is generated in smaller volumes over time from many different locations.
 5    Flowback water also tends to have lower IDS concentrations than later-stage produced water; in
 6    the Marcellus, IDS has been shown to increase from tens of thousands to about 100,000 mg/L
 7    during the first 30 days [BarbotetaL 2013: Maloney and Yoxtheimer. 2012] (see Chapter 7). The
 8    changing production rate and quality of waste waters generated in a region as more wells go into
 9    production need to be taken into account, as well as possible decreases in the demand for reused
10    water as plays mature [Lutz etal.. 2013: Hayes and Severin. 2012b: Slutz etal.. 20121.

      8.4.3.2. Reuse Rates
11    Reliable information on reuse practices throughout the United States is hampered by a limited
12    amountof data thatare available and represent different regions of the country. In Table 8-5,
13    estimates have been compiled from various literature sources. Reuse rates are highest in the
14    Appalachian Basin, associated primarily with the Marcellus Shale. Documentation of reuse practices
15    is also more readily available for that region than for other parts of the country.

16    A number of studies have estimated reuse rates for Marcellus wastewater. Although the reported
17    values can differ substantially (see Table 8-5), the data point to a steep increase in reuse since
18    2008, with rates increasing from 0% to 10% in 2008 to upwards of 90% in 2013. As an example, an
19    analysis of waste disposal information from the PA DEP for Marcellus wells in Pennsylvania
20    (Hansenetal.. 2013] reports an increase in reuse from 9% (7.17 million gal or 27.1 million L] of
21    total wastewater volumes in 2008 to 56% (343.79 million gal or 1.3014 billion L] in 2011. During
22    that same timeframe, the authors report that disposal via brine/industrial waste treatment plants
23    increased from 32% in 2008 to 70% in 2009, and then declined to 30% in 2011. Because some
24    industrial waste treatment plants can treat wastewater for reuse, some of the volumes indicated by
25    Hansenetal. (2013] as managed by this route may have ultimately been used for fracturing,
26    meaning that the 56% value for 2011 is most likely an underestimate.
      Table 8-5. Estimated percentages of reuse of hydraulic fracturing wastewater.
Play or Basin
Source and Year
2008
2009
2010
2011
2012
2013
East Coast
Marcellus, PA
Marcellus, PA
Marcellus, PA
Marcellus, WV
Marcellus, PA
Marcellus, PA
Marcellus, PA
Rahm etal. (2013)
Ma etal. (2014)

Shaffer et al. (2013)

Hansenetal. (2013)

Hansenetal. (2013)

Malonev and
Yoxtheimer (2012)

Tiemann etal. (2014)
9



9


8
15-20


6


25-48


88
20


67-80


73
56
71.6
72


90
65 (partial
year)


87

90





                   This document is a draft for review purposes only and does not constitute Agency policy.
      June 2015                                   8-29                DRAFT—DO NOTE CITE OR QUOTE

-------
     Hydraulic Fracturing Drinking Water Assessment
                                      Chapter 8 - Wastewater Treatment and Waste Disposal
Play or Basin
Marcellus, PA
Marcellus, PA
Marcellus, PA
Marcellus, PA
(SW region)
Marcellus, PA
(NE region)
Marcellus, PA
Source and Year
Rassenfoss (2011)

Wendel (2011)

Lutz et al. (2013)
Rahmetal. (2013)

Rahmetal. (2013)

Rahm and Riha (2014)

2008


2009


2010
~67 overall
(general
estimate)
96 (one
specific
company)
75-85
13 (prior to 2011)
~10
0

~15
0

-25-45
-55-70

2011

90
56
-70-80
-90-100
2012





2013





55-80 (general estimate - appears to
cover recent years)
Gulf Coast & Midcontinent
Fayetteville
West Permian
Midland
Permian
Anadarko
Barnett
Barnett
Eagle Ford
East Texas
Haynesville
Haynesville
Veil (2011)

Nicot et al. (2012)
Nicot et al. (2012)

Nicot et al. (2012)
Nicot et al. (2012)
Rahm and Riha (2014)

Nicot and Scanlon
(2012)
Nicot and Scanlon
(2012)
Argonne National
Laboratory (2014)
Rahm and Riha (2014)





















20 (single
company
target)










0
2
20
5










5 (general estimate - appears to
cover recent years)
0
5

20
(estimate
based on
interviews)




0
5 (general estimate - appears to
cover recent years)
West Coast & Upper Plains
Bakken
Argonne National
Laboratory (2014)





0
1    According to Maloney and Yoxtheimer [2012], about 331 million gal (7.9 million bbl or 1.25 billion
2    L) of flowback and about 3 81 milliongal (about9.1 million bbl or 1.4 billion L) of produced water
3    (excluding flowback) were generated in the Marcellus in 2011. For flowback and produced water
     June 2015
This document is a draft for review purposes only and does not constitute Agency policy.
                               8-30                  DRAFT—DO NOTE CITE OR QUOTE

-------
      Hydraulic Fracturing Drinking Water Assessment              Chapter 8 - Wastewater Treatment and Waste Disposal


 1    combined, about 72% was reused. Of the flowback, 90% was managed through reuse (other than
 2    road spreading). Of produced brine water, 55.7% was reused (with 11.6% treated in CWTs and
 3    27.8% injected into Class IID wells in Ohio). Reuse is higher in the northeastern part of the
 4    Marcellus; in the southwestern portion, easier access to Class IID wells in Ohio makes disposal by
 5    injection more feasible (Rahmetal.. 2013).

 6    Data from Marcellus wastewater management reports submitted to PA DEP (PA PEP, 2015a) were
 7    compiled for this assessment; the data suggest that rates of reuse for hydraulic fracturing (as
 8    indicated by a waste disposal method of either "Reuse Other than Road Spreading" or a zero-
 9    discharge CWT) increased from about 28% in the second half of 2010 to about 60% in 2011, 83%  in
10    2013, and 89% in the first half of 2014. These values maybe underestimates because wastewater
11    treated at facilities with NPDES permits can be provided to operators for reuse, and the permit
12    types for some facilities could not be determined. Among the forms of reuse, on-site reuse ("Reuse
13    Other than Road Spreading") has risen steadily over the past few years, from about 8% in the
14    second half of 2010 to about 48% in 2011, 62% in 2012, and nearly 70% in the first half of 2014.

15    Outside of the Marcellus region, a lower percentage of wastewater from hydraulic fracturing
16    operations is reused. According to published literature, in Texas in 2011, 0% to 5% of wastewater
17    was reused in most basins, with the exception of the Anadarko Basin (20%) (Nicot and Scanlon,
18    2012): see Table 8-5. Maetal. (2014) note that only a small amount of reuse is occurring in the
19    Barnett Shale. Reuse has not yet been pursued aggressively in New Mexico or in the Bakken (North
20    Dakota) (Argonne National Laboratory. 2014: LeBas etal..  2013). Other sources, however, indicate
21    growing interest in reuse, as evidenced in specialized conferences (e.g., "Produced Water Reuse
22    Initiative  2014" on produced water reuse in Rocky Mountain oil and shale gas plays), and available
23    state-developed information on reuse (e.g., fact sheet by the Colorado Oil and Gas Conservation
24    Commission) (Colorado Division of Water Resources: Colorado Water Conservation Board:
25    Colorado  Oil and Gas Conservation Commission. 2014). The fact sheet discusses piping and trucking
26    wastewater to CWTs in the Piceance  Basin to treat for reuse.

      8.4.4.   Evaporation
27    In drier climates of the western United States, natural evaporation may be an option for treatment
28    of hydraulic fracturing wastewater (see Figure 8-6). Production data from the California
29    Department of Conservation's Division of Oil, Gas, and Geothermal Resources (DOGGR) (California
30    Department of Conservation. 2015). for example, lists "evaporation-percolation" as the
31    management method for 23% to 30% of the wastewater from Kern County over the last few years.
32    However, data on volumes of wastewater managed are not readily available for all states where this
33    practice is employed.

34    Evaporation is a simple water management strategy that consists of transporting wastewater to a
35    pond or pit with a large surface area  and allowing passive evaporation of the water from the surface
36    (Clark and Veil. 2009). The rate of evaporation depends on the quality of the wastewater as well as
37    the size, depth, and location of the pond. Evaporation also depends on local humidity, temperature,
38    and wind (NETL. 2014). The residual brine or solid can be  disposed of in an underground injection
39    well or landfill (see Section 8.4.3.2 for more details). In colder, dry climates, a freeze-thaw

                   This document is a draft for review purposes only and does not constitute Agency policy.
      June 2015                                   8-31                DRAFT—DO NOTE CITE OR QUOTE

-------
      Hydraulic Fracturing Drinking Water Assessment
Chapter 8 - Wastewater Treatment and Waste Disposal
 1    evaporation method has been used to purify water from oil and gas wastewater [Boysenetal.,
 2    19991.
      Figure 8-6. Lined evaporation pit in the Battle Creek Field (Montana).
                Source: DOE (2006). Permission from ALL Consulting.

 3    Alternatively, operators may transport wastewater by truck to an off-site commercial facility.
 4    Commercial evaporation facilities exist in Colorado, Utah, New Mexico, and Wyoming [NETL. 2014:
 5    DOE, 2004]. Nowak and Bradish [2010] described the design, construction, and operation of two
 6    large commercial evaporation facilities in Southern Cross, Wyoming and Danish Flats, Utah. Each
 7    facility includes 14,000-gal (53,000 L] three-stage concrete receiving tanks, a sludge pond, and a
 8    series of five-acre (20,234 m2] evaporation ponds connected by gravity or force-main underground
 9    piping. The Wyoming facility, which opened in 2008, consists of two ponds with a total capacity of
10    approximately 84 million gal (2 million bbl or 320 million L). The Utah facility, open since 2009,
11    consists of 13 ponds with a total capacity of 218.4 million gal (5.2 million bbl or 826.7 million L].
12    Each facility receives 420,000 to 1.47 million gal (10,000 to 35,000 bbl or 1.6 million to 5.56 million
13    L] per day of wastewater from oil and gas production companies in the area. Evaporation pits are
14    subject to state regulatory agency approval and must meet state standards for water quality and
15    quantity (Boysenetal.. 2002]. Impacts on drinking water resources from evaporation pits might
16    arise if a pit is breached due to extreme weather or other factors affecting infrastructure and if
17    leaking wastewater reaches a surface water body; such events as related specifically to evaporation
18    pits appear not to have not been evaluated in the literature, and their prevalence is unknown.
                   This document is a draft for review purposes only and does not constitute Agency policy.
      June 2015                                    8-32                 DRAFT—DO NOTE CITE OR QUOTE

-------
      Hydraulic Fracturing Drinking Water Assessment              Chapter 8 - Wastewater Treatment and Waste Disposal


      8.4.5. Publicly Owned Treatment Works
 1    Prior to the development of unconventional resources, POTWs were used to treat wastewater and
 2    other wastes from conventional oil and gas operations in some eastern states. Although this is not a
 3    common treatment method for oil and gas wastes in the United States, the small number of
 4    injection wells for waste disposal in Pennsylvania drove the need for disposal alternatives [Wilson
 5    and Vanbriesen, 2012]. When development of the Marcellus Shale began, POTWs continued to be
 6    used to treat wastewater, including wastes originating from new unconventional oil and gas wells
 7    [Kappel etal.. 2013: Soeder and Kappel. 2009). However, unconventional wastewater from the
 8    Marcellus region is difficult to treat at POTWs due to elevated concentrations of halides, heavy
 9    metals, organic compounds, radionuclides, and salts [Lutz etal., 2013: Schmidt, 2013]. Most of
10    these constituents have the potential to pass through the unit treatment processes commonly used
11    in POTWs and can be discharged into receiving waters [Cusick. 2013: Kappel etal.. 2013). In
12    addition, research has found that sudden, extreme salt fluctuations can disturb POTW biological
13    treatment processes (Linaricetal., 2013: Lefebvre and Moletta, 2006]. In order to meetNPDES
14    requirements, POTWs used to blend the hydraulic fracturing wastewater with incoming municipal
15    wastewater. For example, Ferrar etal. [2013] note that, per PA DEP requirements, one facility could
16    only accept up to 1% of their influent volume from unconventional oil and gas wastewater per day.

17    The annual reported volume of oil and gas produced wastewater treated at POTWs in the Marcellus
18    Shale region peaked in 2008 and has since declined to virtually zero (see Figure 8-7]. This decline
19    has been attributed  to stricter discharge limits for TDS for POTWs in Pennsylvania and widespread
20    voluntary compliance on behalf of oil and gas operators with the May 2011 request from PA DEP to
21    cease sending Marcellus Shale wastewater to 15 treatment plants (including both POTWs and
22    CWTs] by May 19, 2011 (Rahm etal.. 2013]. To comply with the request, the oil and gas industry in
23    Pennsylvania accelerated the transition of wastewater deliveries from POTWs to CWTs for better
24    removal of metals and suspended solids (Schmidt. 2013]. However, treated effluent from CWTs
25    may be delivered to  POTWs for additional treatment assuming treatment processes at POTWs are
26    not adversely affected and the POTWs can continue to meet NPDES discharge limits (Hammer and
27    VanBriesen. 2012]. General Pretreatment Regulations and State and local regulations typically
28    govern the pre-treated water volumes and qualities that can be accepted by the POTW.

29    Although operators  stopped sending Marcellus Shale wastewater to  POTWs in May of 2011,
30    conventionally produced wastes have continued to be processed at POTWs in Pennsylvania,
31    although at small volumes (29 Mgal and 20 Mgal for the years 2010 and 2011, respectively] (Wilson
32    and Vanbriesen. 2012].
                   This document is a draft for review purposes only and does not constitute Agency policy.
     June 2015                                   8-33                DRAFT—DO NOTE CITE OR QUOTE

-------
      Hydraulic Fracturing Drinking Water Assessment              Chapter 8 - Wastewater Treatment and Waste Disposal
                 140

                 120

                 100

                  80

                  60

                  AC

                  2G
                       2001  2002  2003  2004  2005  2006  2007  2008   2009  2010  2011

                             • Marcellus (Mgal)              • Conventional (Mgal)

      Figure 8-7. Oil and gas wastewater volumes discharged to POTWs from 2001-2011 in the
               Marcellus Shale.
               Source: Lutz et al. (2013).

 1    At least one study has evaluated POTW effluent chemistry before and after the cessation of
 2    treatment of hydraulic fracturing wastewater. Ferrar etal. [2013] collected effluent samples from
 3    two POTWs and one CWT facility in Pennsylvania before and after the 2011 PA DEP request.
 4    Results from POTW effluent samples collected while the facilities were still treating Marcellus Shale
 5    wastewater showed that concentrations of several analytes (barium, manganese, strontium, TDS,
 6    and chloride) were greater than various drinking water and surface water criteria (i.e., EPA MCLs
 7    and secondary MCLs for drinking water, surface water quality standards for aquatic life, and/or
 8    surface water standards for human consumption of aquatic organisms). Results for effluent samples
 9    collected after the POTWs stopped receiving Marcellus wastewater showed a statistically
10    significant decrease in the concentrations of several of these constituents. In particular, one of the
11    two POTWs showed a decrease in average barium concentration from 5.99 mg/L to 0.141 mg/L, a
12    decrease in the average strontium concentration from 48.3 mg/L to 0.236 mg/L, and a decrease in
13    the average bromide concentration from 20.9 mg/L to <0.016 mg/L. Influent concentrations at the
14    other POTW were lower (0.55 mg/L for barium, 1.63 mg/L for strontium, and 0.60 for bromide),
15    but significant decreases in these constituents were also seen in the effluents (0.036 mg/L barium,
16    0.228 mg/L strontium, and 0.119 for bromide); this POTW had continued to accept conventional oil
17    and gas wastewater. The authors conclude that the decreases in the concentrations of the various
18    constituents indicate that the elevated concentrations in the first samplings can be attributed to the
19    contribution of wastewater from unconventional natural gas development.
                   This document is a draft for review purposes only and does not constitute Agency policy.
      June 2015                                   8-34                 DRAFT—DO NOTE CITE OR QUOTE

-------
      Hydraulic Fracturing Drinking Water Assessment              Chapter 8 - Wastewater Treatment and Waste Disposal


      8.4.6.  Other Management Practices and Issues
 1    Additional strategies for wastewater management in some states include discharging to surface
 2    waters and land application. Wastewater from CBM fracturing and production, in particular,
 3    generally has lower IDS concentrations than wastewater from other types of unconventional plays
 4    and lends itself more readily to beneficial use. Below is a discussion of these other management
 5    practices.

      8.4.6.1. Land Application, Including Road Spreading
 6    Land application has been done using brines from conventional oil and gas production. Road
 7    spreading can be done for dust control or de-icing. Although recent data are not available, an
 8    American Petroleum Institute (API) survey estimated that approximately 75.6 million gal (1.8
 9    million bbl or 286 million L) of wastewater was used for road spreading in 1995 (API. 2000). The
10    API estimate does not specifically identify hydraulic fracturing wastewater. There is no current
11    nationwide estimate of the extent of road spreading using hydraulic fracturing wastewater.

12    Road spreading with hydraulic fracturing wastewater is regulated primarily at the state level
13    (Hammer and VanBriesen. 2012] and is prohibited in some states. For example, with annual
14    approval of a plan to minimize the potential for pollution, PA DEP allows spreading of brines from
15    conventional wells for dust control or road stabilization. Hydraulic fracturing flowback, however,
16    cannot be used for dust control and road stabilization (PADEP. 2011b). In West Virginia, use of gas
17    well brines for roadway de-icing is allowed per a 2011 memorandum of agreement between the
18    West Virginia Division of Highways and the West Virginia Department of Environmental Protection,
19    but the use of "hydraulic fracturing return  fluids" is not permitted (Tiemann et al., 2014: West
20    Virginia DEP. 20111.

21    Concerns about road application center on contaminants such as barium, strontium, and radium. A
22    report from PA DEP analyzed several commercial rock salt samples and compared results with
23    contaminants found in Marcellus Shale flowback samples; the results noted elevated barium,
24    strontium, and radionuclide levels in Marcellus Shale brines compared with commercial rock salt
25    (Titler and Curry. 2011). Another study found increases in metals (radium, strontium, calcium, and
26    sodium) in soils ranging from 1.2 to 6.2 times the original concentration (for radium and sodium,
27    respectively), attributed to road spreading of wastewater from conventional oil and gas wells for
28    de-icing purposes (Skalaketal., 2014).

29    Potential impacts on drinking water resources from road spreading, have been noted by Tiemann et
30    al. (2014) and Hammer and VanBriesen (2012). These include potential effects of runoff on surface
31    water, or migration of brines to groundwater. Snowmelt may carry salts or other chemicals from
32    the application site, with the possibility of transport increasing if application rates are high or rain
33    occurs soon after application (Hammer and VanBriesen. 2012). Research on the impacts of
34    conventional road salt application has documented long-term salinization of both surface water and
35    ground water in the northern United States; by the 1990s, 24% of public supply wells in the  Chicago
36    area had chloride concentrations exceeding 100 mg/L (Kelly, 2008: Kaushal etal., 2005). When
37    conventional oil field brine was used in a controlled road spreading experiment, elevated chloride
38    concentrations were detected in shallow ground water (531 ppm in winter and 1,360 ppm in

                   This document is a draft for review purposes only and does not constitute Agency policy.
      June 2015                                   8-35                 DRAFT—DO NOTE CITE OR QUOTE

-------
      Hydraulic Fracturing Drinking Water Assessment              Chapter 8 - Wastewater Treatment and Waste Disposal


 1    summer [Bair and Digel, 1990]. The amount of salt contributed to drinking water resources due to
 2    road application of hydraulic fracturing waste waters has not been quantified.

 3    In managing solid wastes from oil and gas production, a study on land application of oilfield scales
 4    and sludges suggested that radium in samples became more mobile after incubation with soil under
 5    moist conditions, due to microbial processes and interactions with the soil and water [Matthews et
 6    al., 2006]. Overall, potential effects from land application on drinking water resources are not well
 7    understood.

 8    Additionally, drill cuttings must be managed; in some places they are left in the reserve pit (pit for
 9    waste storage], allowed to dry and, buried on-site [Kappeletal.. 2013]. More, commonly, however,
10    drill cuttings are disposed of in landfills [Chiado. 2014]: about half of Marcellus drill cuttings are
11    disposed of in Pennsylvania, while the rest are trucked to Ohio or West Virginia [Maloney and
12    Yoxtheimer. 2012].

      8.4.6.2. Management of Coalbed Methane Wastewater
13    Wastewater from CBM wells can be managed like other hydraulic fracturing wastewater discussed
14    above. However, the wastewater from CBM wells can also be of higher average quality (e.g., lower
15    TDS content] than wastewater from other hydraulically fractured wells, which makes it more
16    suitable for certain management practices and uses. A number of management strategies have been
17    proposed or implemented, with varying degrees of treatment required depending on the quality of
18    the wastewater and the requirements of the intended use (Hulme. 2005: DOE. 2003. 2002].
19    Although specific volumes managed through the various practices below are not well documented,
20    qualitative information and considerations for feasibility are available and presented below.

21    CBM wastewater quality, which can range from an average of nearly 1,000 mg/L TDS  in the Powder
22    River Basin to an average of about 4,700 mg/L in the San Juan Basin (see Appendix Table E-3],
23    plays a large role in how the wastewater is managed. In basins with higher TDS such as the San
24    Juan, Uinta, and Piceance, nearly all the wastewater is disposed of in injection wells. Wastewater
25    may also be injected for aquifer storage and recovery, with the intention of later recovering the
26    water for some other use  fDOE. 20031

27    Discharge to rivers and streams for wildlife, livestock, and agricultural use, a management option
28    governed by the CWA, may be permitted in some cases. To be discharged, the wastewater must
29    meet technology-based limitations established by the permit authority and any applicable water
30    quality water quality standards. Direct discharge to streams (with or without treatment] is possible
31    where wastewater is of higher quality. This is a more common method of wastewater management
32    in basins such as the Raton Basin in Colorado and the Tongue River drainage of the Powder River
33    Basin in Montana fNRC. 20101

34    Agricultural uses include livestock watering, crop irrigation, and commercial fisheries. Livestock
3 5    watering with CBM wastewater is a common practice, and irrigation is an area of active research
36    (e.g.. Engle etal.. 2011: NRC. 2010]. Irrigation with treated CBM wastewater would be most suitable
37    on coarse-textured soils, for cultivation of salt-tolerant crops (DOE, 2003]. NRC (2010] remarks
38    that "use of CBM produced water for irrigation appears practical and sustainable," provided that

                   This document is a draft for review purposes only and does not constitute Agency policy.
      June 2015                                   8-36                 DRAFT—DO NOTE CITE OR QUOTE

-------
      Hydraulic Fracturing Drinking Water Assessment              Chapter 8 - Wastewater Treatment and Waste Disposal


 1    appropriate measures are taken such as selective application, dilution or blending, appropriate
 2    timing, and rehabilitation of soils. Approximately 13% of CBM wastewater in the Powder River
 3    Basin in Wyoming, and 26% to 30% in Montana, is used for irrigation [NRG. 2010).

 4    As noted above, a degree of treatment is needed for some uses. Plumlee etal. [2014] examined the
 5    feasibility, treatment requirements, and cost of several hypothetical uses for CBM wastewater. In
 6    several cases, costs for these uses were projected to be comparable to or less than estimated
 7    disposal costs. In one case study CBM wastewater for stream augmentation or crop irrigation was
 8    estimated to cost between $0.26 and $0.27 a barrel and disposal costs ranged from $0.01 per barrel
 9    (pipeline collection system with impoundment) to $2.00 per barrel (hauling for disposal or
10    treatment).

11    The applicability of particular uses may be limited by ecological and regulatory considerations, as
12    well as the irregular nature of CBM wastewater production (voluminous at first, and then declining
13    and halting after a period of years). Legal issues, including overlapping jurisdictions at the state
14    level and, in western states, senior water rights claims in over-appropriated basins, may also
15    determine use of CBM wastewater (Wolfe and Graham. 2002).

      8.4.6.3. Other Documented Uses of Hydraulic Fracturing Wastewater
16    Uses of wastewater from shales or other hydraulically fractured formations face many of the same
17    possibilities and limitations as those associated with wastewater from CBM operations. The biggest
18    difference is in the quality of the water. Wastewaters vary widely in water quality, with TDS values
19    from shale sand tight sand formations ranging from less than 1,000 mg/L TDS to hundreds of
20    thousands of mg/L TDS (DOE. 2006). Wastewaters on the lower end of the TDS spectrum could be
21    reused in many of the same ways as CBM Wastewaters, depending on the concentrations of
22    potentially harmful constituents and applicable federal, state, and local regulations. High TDS
23    wastewaters have more limited uses, and pre-treatment may be necessary (Shaffer etal.,  2013:
24    Guerraetal.. 2011: DOE. 2006).

2 5    Documented potential uses for wastewater in the western United States include livestock watering,
26    irrigation, supplementing stream flow, fire protection, road spreading, and industrial uses, with
27    each having their  own water quality requirements and applicability (Guerra etal.. 2011).  Guerra et
28    al. summarized the least conservative TDS standards for five possible uses in the western United
29    States that include 500 mg/L for drinking water (the secondary MCL), 625 mg/L for groundwater
30    recharge, 1,000 mg/L for surface water discharge, 1,920 mg/L for irrigation, and 10,000 mg/L for
31    livestock watering. The authors estimated that wastewater from 88% of unconventional wells in
32    the western United States could be used for livestock watering without treatment for TDS removal
33    based on a maximum TDS concentration of 10,000 mg/L. Wastewater from 10% of unconventional
34    wells were estimated to meet the criterion of 1,000 mg/L TDS for surface water discharge (Guerra
35    etal.. 2011).

36    A2006 Department of Energy (DOE) study points out that the quality necessary for use in
37    agriculture depends on the plant or animal species involved. Other important factors include the
38    sodium adsorption ratio and concentrations of TDS, calcium, magnesium, and other constituents
                   This document is a draft for review purposes only and does not constitute Agency policy.
      June 2015                                   8-37                 DRAFT—DO NOTE CITE OR QUOTE

-------
      Hydraulic Fracturing Drinking Water Assessment              Chapter 8 - Wastewater Treatment and Waste Disposal


 1    [DOE, 2006]. The authors note that in the Bighorn Basin in Wyoming, low salinity wastewater is
 2    used for agriculture and livestock watering after minimal treatment to remove oil and grease [DOE,
 3    20061

      8.5. Summary and Analysis of Wastewater Treatment
 4    A variety of individual treatment techniques and combinations of techniques may be employed for
 5    removal of hydraulic fracturing wastewater constituents of concern. These include methods
 6    commonly employed in municipal wastewater treatment as well as more advanced processes such
 7    as desalination. Treatment technologies are selected based on the water quality of the wastewater
 8    to be treated and the effluent concentration required for the intended management method(s) (i.e.,
 9    reuse, discharge to POTW, and discharge to surface water body). For example, if reuse is planned,
10    the level of treatment will depend on the water quality needed to formulate the new fracturing
11    fluid.

12    This section discusses treatment technologies that are most effective for removing specific
13    hydraulic fracturing wastewater constituents. It provides information on the unit processes
14    appropriate for treating different types of constituents as well as challenges associated with their
15    use. Considerations when designing a treatment system are also discussed for both centralized and
16    on-site  (i.e., mobile) facilities.

      8.5.1.  Overview of Treatment Processes for Hydraulic Fracturing Wastewater
17    This section provides a brief overview of the treatment technologies used to treat hydraulic
18    fracturing wastewater; Appendix F provides more in-depth descriptions of these technologies.

19    The most basic treatment need for oil and gas wastewaters, including those from hydraulic
20    fracturing operations is separation to remove suspended solids and oil and grease, done using basic
21    separation technologies (e.g., hydrocyclones, dissolved air or induced gas flotation, media filtration,
22    or biological aerated filters). Other treatment processes that may be used include media filtration
23    after chemical precipitation for hardness and metals (Boschee. 2014). adsorption technologies,
24    including ion exchange (organics, heavy metals, and some  anions) (Igunnu and Chen,  2014], a
25    variety of membrane processes (microfiltration, ultrafiltration, nanofiltration, RO), and distillation
26    technologies. In particular, advanced processes such as RO or distillation methods (e.g., mechanical
27    vapor recompression (MVR)) are needed for significant reduction in TDS (Drewes etal.. 2009: LEau
28    LLC. 2008: Hamieh and Beckman. 2006). An emerging technology is electrocoagulation, which has
29    been used in mobile treatment systems to treat hydraulic fracturing wastewaters (Halliburton,
30    2014: Igunnu and Chen. 2014). Removal efficiencies for hydraulic fracturing wastewater
31    constituents by treatment technology are provided in Appendix F.

      8.5.2.  Treatment of Hydraulic Fracturing Waste Constituents of Concern
32    The constituents prevalent in hydraulic  fracturing wastewater include suspended solids, TDS,
33    anions (e.g., chloride, bromide, and sulfate), metals, radionuclides, and organic compounds (see
34    Section 8.3 and Chapter 7). If the end use of the wastewater necessitates treatment, a variety of
35    technologies can be employed. This section discusses effective unit processes for removing these
36    constituents and provides examples of treatment processes being used in the field as  well as

                   This document is a draft for review purposes only and does not constitute Agency policy.
      June 2015                                    8-38                 DRAFT—DO NOTE CITE OR QUOTE

-------
     Hydraulic Fracturing Drinking Water Assessment               Chapter 8 - Wastewater Treatment and Waste Disposal


1    emerging technologies. Table 8-6 provides an overview of influent and effluent results at various
2    CWTs for the constituents of concern listed in this section and the specific technology(ies) used to
3    remove them.
                   This document is a draft for review purposes only and does not constitute Agency policy.
     June 2015                                      8-39                  DRAFT—DO NOTE CITE OR QUOTE

-------
Hydraulic Fracturing Drinking Water Assessment
                                                           Chapter 8 - Wastewater Treatment and Waste Disposal
Table 8-6. Studies of removal efficiencies and influent/effluent data for various processes and facilities.





Constituents
of concern

TSS




IDS




Location and results


Pinedale Anticline Water
Reclamation Facility,
Wyoming
(Shafer, 2011)

Results not reported.




>99%
Inf. = 8,000 to
15,000 mg/L
Eff. = 41 mg/L
RO


Maggie Spain Water-
Recycling Facility, Barnett
Shale, Texas
(Haves et al., 2014)

90%
Inf. = 1,272 mg/L
Eff. = 9 mg/L
Chemical oxidation,
coagulation, and
clarification

99.7%
Inf. = 49,550 mg/L
Eff. = 171 mg/L
MVR(3 units in parallel)




Judsonia, Sunnydale,
Arkansas
(U.S. EPA, 2015f)

No influent data.
Eff.: <4 mg/L
Meets NPDES Permit
Settling, biological
treatment, and induced
gas flotation
Results not reported.
MVR




9-month study treating
Marcellus Shale waste
using thermal distillation
(Boschee. 2014: Bruff and
Jikich, 2011)

>90%
Inf.: 35 to 114 mg/L
Eff. :<3 to 3 mg/L
100 micron mesh bag filter

98%
Inf.: 22,350 to 37,600 mg/L
Eff.: 9 to 400 mg/L
Thermal distillation

San Ardo Water
Reclamation Facility, San
Ardo, California
(Conventional oil and gas)
(Dahm and Chapman. 2014;
Webb et al., 2009)

Results not reported.




97%
Inf. = 7,000 mg/L
Eff. = 180 mg/L
Ion exchange softening and
double-pass RO
June 2015
This document is a draft for review purposes only and does not constitute Agency policy.
                                8-40
DRAFT—DO NOTE CITE OR QUOTE

-------
Hydraulic Fracturing Drinking Water Assessment
                                                               Chapter 8 - Wastewater Treatment and Waste Disposal





Constituents
of concern

Anions












Location and results


Pinedale Anticline Water
Reclamation Facility,
Wyoming
(Shafer. 2011)

Chloride: >99%
Inf. = 3,600 to 6,750 mg/L
Eff. = 18 mg/L
Sulfate: 99%
Inf. = 10 to 100 mg/L
Eff. = non-detect
RO








Maggie Spain Water-
Recycling Facility, Barnett
Shale, Texas
(Haves et al.. 2014)

Sulfate: 98%
Inf. = 309 mg/L
Eff. = 6 mg/L
Chemical oxidation,
coagulation, clarification,
and MVR










Judsonia, Sunnydale,
Arkansas
(U.S. EPA, 2015f)

Sulfate:
No influent data.
Eff.: 12 mg/L
Meets NPDES Permit
MVR









9-month study treating
Marcellus Shale waste
using thermal distillation
(Boschee, 2014: Bruff and
Jikich, 2011)

Bromide: >99%
Inf.: 101 to 162.5 mg/L
Eff.: <0.1 to 1.6 mg/L
Chloride: 98%
Inf.: 9,760 to 16,240 mg/L
Eff.: 2.9 to 184.2 mg/L
Sulfate: 93%
Inf.: 20.4 to <100 mg/L
Eff.: <1 to 2.2 mg/L
Fluoride: 96%
lnf.:<2to<20mg/L
Eff. :<0.2 to 0.42 mg/L
Thermal distillation
San Ardo Water
Reclamation Facility, San
Ardo, California
(Conventional oil and gas)
(Dahm and Chapman. 2014:
Webb et al., 2009)

Chloride: >99%
Inf. = 3,400 mg/L
Eff. = 11 mg/L
Double-pass RO
Sulfate: 6%
Inf. = 133 mg/L
Eff. = 125 mg/L


Sulfuric acid is added after
RO to neutralize the pH so
no sulfate removal is
expected.
June 2015
This document is a draft for review purposes only and does not constitute Agency policy.
                                   8-41
DRAFT—DO NOTE CITE OR QUOTE

-------
Hydraulic Fracturing Drinking Water Assessment
                                                               Chapter 8 - Wastewater Treatment and Waste Disposal





Constituents
of concern

Metals





















Location and results


Pinedale Anticline Water
Reclamation Facility,
Wyoming
(Shafer. 2011)

Boron: 99%
Inf. = 15 to 30 mg/L
Eff. = non-detect
Ion exchange




















Maggie Spain Water-
Recycling Facility, Barnett
Shale, Texas
(Haves et al.. 2014)

Iron: >99%
Inf. = 28 mg/L
Eff. =0.1 mg/L
For iron, 90% attributed to
chemical oxidation,
coagulation, and
clarification

Boron: 98%
Inf. = 17 mg/L
Eff. = 0.4 mg/L

Barium: >99%
Inf. = 15 mg/L
Eff. =0.1 mg/L

Calcium: >99%
Inf. = 2,916 mg/L
Eff. = 3.2 mg/L






Judsonia, Sunnydale,
Arkansas
(U.S. EPA, 2015f)

Cobalt:
No influent data.
Eff. :<0.007 mg/L
Tin:
No influent data.
Eff. :<0.1 mg/L

Arsenic:
No influent data.
Eff. :<0.001 mg/L

Cadmium:
No influent data.
Eff. :<0.0001 mg/L

Chromium:
No influent data.
Eff. :<0.007 mg/L

Copper:
No influent data.
Eff. :<0029 mg/L

9-month study treating
Marcellus Shale waste
using thermal distillation
(Boschee, 2014: Bruff and
Jikich, 2011)

Copper: >99%
Inf. = <0.2 to <1.0 mg/L
Eff. = <0.02 to <0.08 mg/L
Zinc: inf below detect
Inf. = <0.2 to <1.0 mg/L
Eff. = <0.02 to 0.05 mg/L

Barium: >99%
Inf. = 260.5 to 405.5 mg/L
Eff. =<0.1 to 4.54 mg/L

Strontium: 98%
Inf. = 233 to 379 mg/L
Eff. = 0.026 to 3.93 mg/L

Iron:
Inf. = 13.9 to 22.9 mg/L
Eff. = <0.02 to 0.06 mg/L

Manganese: 98%
Inf. = 2 to 2.9 mg/L
Eff. = <0.02 to 0.04 mg/L
San Ardo Water
Reclamation Facility, San
Ardo, California
(Conventional oil and gas)
(Dahm and Chapman. 2014:
Webb et al., 2009)

Sodium: 98%
Inf. = 2,300 mg/L
Eff. = 50 mg/L
Boron: >99%
Inf. = 26 mg/L
Eff. =0.1 mg/L

RO with elevated influent
pH













June 2015
This document is a draft for review purposes only and does not constitute Agency policy.
                                   8-42
DRAFT—DO NOTE CITE OR QUOTE

-------
Hydraulic Fracturing Drinking Water Assessment
                                                               Chapter 8 - Wastewater Treatment and Waste Disposal





Constituents
of concern

Metals,
cont.






















Location and results


Pinedale Anticline Water
Reclamation Facility,
Wyoming
(Shafer. 2011)



























Maggie Spain Water-
Recycling Facility, Barnett
Shale, Texas
(Haves et al.. 2014)

Magnesium: >99%
Inf. = 316 mg/L
Eff. = 0.4 mg/L
Sodium: >99%
Inf. = 10,741 mg/L
Eff. = 14.3 mg/L
Strontium: >99%
Inf. = 505 mg/L
Eff. = 0.5 mg/L

MVR
















Judsonia, Sunnydale,
Arkansas
(U.S. EPA, 2015f)

Lead:
No influent data.
Eff.: <0.001 mg/L
Mercury:
No influent data.
Eff.: <0.005 mg/L
Nickel:
No influent data.
Eff.: 0.002 mg/L

Silver:
No influent data.
Eff.: <0.0002 mg/L
Zinc:
No influent data.
Eff.: 0.02 mg/L

Cyanide:
No influent data.
Eff.: <0.01 mg/L
Meets NPDES permit
except for TMDLs for
hexavalent chromium
and mercury

9-month study treating
Marcellus Shale waste
using thermal distillation
(Boschee, 2014: Bruff and
Jikich, 2011)

Boron: 97%
Inf. =<1 to 3. 12 mg/L
Eff. = 0.02 to 0.06 mg/L
Calcium: 98%
Inf. = 1,175 to 1,933 mg/L
Eff. = 0.36 to 22.2 mg/L
Lithium: 99%
Inf. = 9.1 to 14.3 mg/L
Eff. = non-detecttoO.13
ms/L
1 ' '&/ •-
Magnesium: 98%
Inf. = 109.8 to 176.8 mg/L
Eff. =<0.1 to 2.04 mg/L
Sodium: 98%
Inf. = 4,712 to 7,781 mg/L
Eff. = 0.37 to 87.9 mg/L

Arsenic: 82%
Inf. = <0.01 to 0.028 mg/L
Eff. = <0.005 mg/L
Titanium: 86%
Inf. = <0.01 to 0.037 mg/L
Eff. = <0.005 mg/L
San Ardo Water
Reclamation Facility, San
Ardo, California
(Conventional oil and gas)
(Dahm and Chapman. 2014:
Webb et al., 2009)

























June 2015
This document is a draft for review purposes only and does not constitute Agency policy.
                                   8-43
DRAFT—DO NOTE CITE OR QUOTE

-------
Hydraulic Fracturing Drinking Water Assessment
                                                                                   Chapter 8 - Wastewater Treatment and Waste Disposal
                                                                    Location and results
 Constituents
 of concern
Pinedale Anticline Water
Reclamation Facility,
Wyoming
(Shafer. 2011)
Maggie Spain Water-
Recycling Facility, Barnett
Shale, Texas
(Haves et al.. 2014)
Judsonia, Sunnydale,
Arkansas
(U.S. EPA. 2015f)
9-month study treating
Marcellus Shale waste
using thermal distillation
(Boschee, 2014: Bruff and
Jikich. 2011)
San Ardo Water
Reclamation Facility, San
Ardo, California
(Conventional oil and gas)
(Dahm and Chapman. 2014:
Webb et al.. 2009)
 Metals,
 cont.
                                                   Settling, biological
                                                   treatment, induced gas
                                                   flotation, and MVR
                                                   Thermal distillation
 Radio-
 nuclides
Results not reported.
Results not reported.
Not regulated under
permit - believed to be
absent.
Radium-226: 97% - 99%
Inf. = 130 to 162 pCi/L
Eff. = 0.224 to 2.87 pCi/L

Radium-228: 97% - 99%
Inf. = 45 to 85.5 pCi/L
Eff. = 0.259 to 1.32 pCi/L

Gross Alpha: 97% - >99%
Inf. = 161to664pCi/L
Eff. = 0.841 to 6.49 pCi/L

Gross Beta: 98% - >99%
Inf. = 79.7 to 847 pCi/L
Eff. = 0.259 to 1.57 pCi/L

Thorium 232: 71%-90%
Inf. =0.055 to 0.114 pCi/L
Eff. = 0.011 to 0.016 pCi/L

Thermal distillation
Results not reported.
June 2015
                      This document is a draft for review purposes only and does not constitute Agency policy.
                                                        8-44
                                                                          DRAFT—DO NOTE CITE OR QUOTE

-------
Hydraulic Fracturing Drinking Water Assessment
                                                               Chapter 8 - Wastewater Treatment and Waste Disposal





Constituents
of concern

Organics























Location and results


Pinedale Anticline Water
Reclamation Facility,
Wyoming
(Shafer. 2011)

O&G: 99%
Inf. = 50 to 2,400 mg/L
Eff. = non-detect

BTEX: 99%
Inf. = 28 to 80 mg/L
Eff. = non-detect
Gasoline range organics:
RO: 99%
Inf. = 88 to 420 mg/L
Eff. = non-detect
Diesel range organics: 99%
Inf. = 77 to 1,100 mg/L
Eff. = non-detect
Methanol: 99%
Inf. = 40 to 1,500 mg/L
Eff. = non-detect
Oil-water separator,
anaerobic and aerobic
biological treatment,
coagulation, flocculation,
flotation, sand filtration,

membrane bioreactor,
and ultrafiltration


Maggie Spain Water-
Recycling Facility, Barnett
Shale, Texas
(Haves et al.. 2014)

TPH: >80%
Inf. = 388 mg/L
Eff. = 4.6 mg/L

BTEX: 94%
Inf. = 3.3 mg/L
Eff. = 0.2 mg/L
TOC: 48%
Inf. = 42 mg/L
Eff. = 22 mg/L
MVR
















Judsonia, Sunnydale,
Arkansas
(U.S. EPA, 2015f)

Biochemical oxygen
demand:
No influent data.
Eff.: <2 mg/L

O&G:
No influent data.
Eff.: <5 mg/L
Benzo (k) fluoranthene:
No influent data.
Eff.: <0.005 mg/L
Bis (2-Ethylhexyl)
Phthalate:
No influent data.
Eff.: <0.001 mg/L
Butyl benzyl phthalate:
No influent data.
Eff.: <0.001 mg/L
Meets NPDES permit
Settling, biological
treatment, induced gas

flotation, and MVR



9-month study treating
Marcellus Shale waste
using thermal distillation
(Boschee, 2014: Bruff and
Jikich, 2011)

Acetone: 93%
Inf. = 8.71 to 13.8 mg/L
Eff. = 0.524 to 0.949 mg/L

Toluene: >80%
Inf. = 0.0083 to 0.0015
mg/L
Eff. = non-detect to 0.0013
___ /i
mg/L
Methane: >99%
Inf. =0.748 to 5.49 mg/L
Eff. = non-detect to 0.0013
___ /i
mg/L
DRO: 0 to 82%
Inf. =4 to 7.1 mg/L
Eff. = 0.99 to 4.9 mg/L
O&G: No removal
Thermal distillation






San Ardo Water
Reclamation Facility, San
Ardo, California
(Conventional oil and gas)
(Dahm and Chapman. 2014:
Webb et al., 2009)

Results not reported.























June 2015
This document is a draft for review purposes only and does not constitute Agency policy.
                                   8-45
DRAFT—DO NOTE CITE OR QUOTE

-------
      Hydraulic Fracturing Drinking Water Assessment              Chapter 8 - Wastewater Treatment and Waste Disposal


      8.5.2.1. Total Suspended Solids
 1    The reduction of TSS is typically required for reuse. Hydraulic fracturing waste waters containing
 2    suspended solids can plug the well and damage equipment if reused for other fracking operations
 3    [Tiemann et al.. 2014: Hammer and VanBriesen. 2012). For treated water that is discharged, the
 4    EPA has a secondary treatment standard for POTWs that limits TSS in the effluent to 30 mg/L (30-
 5    day average). In addition, most advanced treatment technologies require the removal of TSS prior
 6    to treatment to avoid operational problems such as membrane fouling/scaling and to extend the life
 7    of the treatment unit TSS can be removed by several processes, such as coagulation, flocculation,
 8    sedimentation, and filtration (including micro filtration and media and bag and/or cartridge
 9    filtration), and with hydrocyclones, dissolved air flotation, freeze-thaw evaporation,
10    electrocoagulation, and biological aerated  filters (Boschee, 2014: Igunnu and Chen, 2014: Drewes et
11    al.. 2009: Fakhru'1-Razi etal.. 2009) (see Appendix F).

12    Technologies that remove TSS have been employed in the Marcellus Shale (sedimentation and
13    filtration) (Mantell, 2013a): Utica Shale (chemical precipitation and filtration) (Mantell, 2013a):
14    Barnett Shale (chemical precipitation and  inclined plate clarifier, >90% removal) (Hayes etal.,
15    2014): and Utah (electrocoagulation, 90%  removal) (Halliburton. 2014). Details of examples of
16    operating treatment facilities are provided in Table 8-6.

      8.5.2.2. Total Dissolved Solids
17    The TDS concentration of hydraulic fracturing wastewater is a key treatment consideration, with
18    the TDS removal needed dependent upon the intended use of the treatment effluent POTW
19    treatment and basic treatment processes at a CWT (i.e., chemical precipitation, sedimentation, and
20    filtration) are not reliable methods for removing TDS. Reduction requires more advanced treatment
21    processes such as RO, nanofiltration, thermal distillation (including MVR), evaporation, and/or
22    crystallization (Olssonetal.. 2013: Boschee. 2012: Drewes etal.. 2009). RO and thermal distillation
23    processes can treat waste streams with TDS concentrations up to 45,000 mg/L and more than
24    100,000 mg/L, respectively (Tiemann etal.. 2014). As noted in section 8.5.1, pretreatment (e.g.,
25    chemical precipitation, flotation, etc.) is typically needed to remove constituents that may cause
26    fouling or scaling or to remove specific constituents not removed by a particular advanced process.
27    Extremely high TDS waters may require a  series of advanced treatment processes to remove TDS to
28    desired levels. However, the cost of treating high-TDS waters may preclude facilities from choosing
29    treatment if other options such as deep well injection are available and more cost-effective
30    (Tiemann et al., 2014). Emerging technologies such as membrane distillation and forward osmosis
31    are also showing promise for TDS removal and require less energy compared to other desalination
32    processes (Shaffer etal.. 2013).

33    Examples of facilities with advanced technologies and their effectiveness in reducing TDS
34    concentrations in hydraulic fracturing wastewaters from conventional and unconventional
35    resources are summarized in Table 8-6.

      8.5.2.3. Anions
36    Although chemical precipitation processes can reduce concentrations of multivalent anions such as
37    sulfate, monovalent anions (e.g., bromide and chloride) are not removed by basic treatment

                   This document is a draft for review purposes only and does not constitute Agency policy.
      June 2015                                   8-46                 DRAFT—DO NOTE CITE OR QUOTE

-------
      Hydraulic Fracturing Drinking Water Assessment              Chapter 8 - Wastewater Treatment and Waste Disposal


 1    processes and require more advanced treatment such as RO, nanofiltration, thermal distillation
 2    (including MVR), evaporation, and/or crystallization [Hammer and VanBriesen. 2012].

 3    Ion exchange and adsorption are effective treatment processes for removing fluoride but not
 4    typically the anions of concern in hydraulic fracturing wastewaters (bromide, chloride, sulfate)
 5    (Drewes et al.. 2009]. Emerging technologies applicable to TDS will typically remove anions.
 6    However, issues discussed above, such as the potential for scaling, still apply.

      8.5.2.4. Metals and Metalloids
 7    Chemical precipitation, including lime softening and chemical oxidation, is effective at removing
 8    metals (e.g., sodium sulfate reacts with metals to form solid precipitates such as barium sulfate]
 9    (Drewes etal.. 2009: Fakhru'1-Razi et al.. 2009]. However, as mentioned in Section 8.5.2.3, chemical
10    precipitation does not adequately remove monovalent ions (e.g., sodium, potassium], and the
11    produced solid residuals from this process typically require further treatment, such as de-watering
12    (Duraisamy etal., 2013: Hammer and VanBriesen, 2012]. Media filtration can remove metals if
13    coagulation/oxidation is implemented prior to filtration (Duraisamy etal., 2013]. Advanced
14    treatment processes such as distillation (with pH adjustment to prevent scaling], evaporation, RO,
15    and nanofiltration can remove dissolved metals and metalloids (Hayes etal.. 2014: Igunnu and
16    Chen, 2014: Bruffandlikich, 2011: Drewes etal., 2009]. However, if metal oxides are presenter
17    formed during treatment, they must be removed prior to RO and nanofiltration processes to
18    prevent membrane fouling (Drewes et al., 2009]. Also, boron is not easily removed by RO, achieving
19    less than 50% rejection (the percentage of a constituent captured and thus removed by the
20    membrane] at neutral pH (rejection is greater at higher pH values] (Drewes etal.. 2009]. Ion
21    exchange can be used to remove other metals such as calcium, magnesium, barium, strontium, and
22    certain oxidized heavy metals such as chromate and selenate (Drewes et al., 2009]. Adsorption can
23    remove metals but is typically used as a polishing step to prolong the replacement/regeneration of
24    the adsorptive media (Igunnu and Chen. 2014].

25    The literature provides examples of facilities able to reduce metal and metalloid concentrations in
2 6    conventional and unconventional hydraulic fracturing wastewaters, some of which are provided in
27    Table 8-6. The facilities in Table 8-6 have achieved removals of 98%-99% for a number of metals.
28    Other work demonstrating effective removal includes a 99% reduction in barium using chemical
29    precipitation (Marcellus Shale region] (Warner etal.. 2013a] and over 90% boron removal with RO
30    (atpH of 10.8] at two California facilities (Webb etal.. 2009: Kennedy/Tenks  Consultants. 2002].
31    However, influent concentration must be considered together with removal efficiency to determine
32    whether effluent quality meets the requirements dictated by end use or by regulations. In the case
33    of the facility described by Kennedy/Tenks Consultants (2002] the boron effluent concentration of
34    1.9 mg/L (average influent concentration of 16.5 mg/L] was not low enough to meet California's
35    action level of 1 mg/L.

36    Newer treatment methods for metals removal include electrocoagulation (Halliburton. 2014:
37    Gomes etal., 2009] and electrodialysis (Banasiak and Schafer, 2009]. Testing of electrocoagulation
38    has been performed in the Green River Basin (Halliburton, 2014] and the Eagle Ford Shale (Gomes
39    etal.. 2009]. While showing promising results in some trials, results of these early studies have

                   This document is a draft for review purposes only and does not constitute Agency policy.
      June 2015                                   8-47                DRAFT—DO NOTE CITE OR QUOTE

-------
      Hydraulic Fracturing Drinking Water Assessment              Chapter 8 - Wastewater Treatment and Waste Disposal


 1    illustrated challenges, with removal efficiencies affected by factors such as pH and salt content
 2    Membrane distillation has also shown promise in removing heavy metals and boron in wastewaters
 3    fCamachoetal.. 20131

      8.5.2.5. Radionudides
 4    Several processes (e.g., RO, nanofiltration, and thermal distillation) are effective for removing
 5    radionuclides [Drewes etal.. 2009). Ion exchange can be used to treat for specific radionuclides
 6    such as radium [Drewes etal., 2009]. Chemical precipitation of radium with barium sulfate has also
 7    been shown to be a very effective method for removing radium [Zhang et al.. 2014b).

 8    Data on radionuclide removals achieved in active treatment plants are scarce. The literature does
 9    provide some data from the Marcellus Shale region on use of distillation and chemical precipitation
10    (co-precipitation of radium with barium sulfate). The nine-month pilot-scale study conducted by
11    Bruff and Jikich (2011) showed that distillation treatment could achieve high removal efficiencies
12    for radionuclides (see Table 8-6), and Warner etal. (2013b) reported that a CWT achieved over
13    99% removal of radium via co-precipitation of radium with barium sulfate. However, in both
14    studies, radionuclides were detected in effluent samples, and the CWT was discharging to a surface
15    water body during this time (Warner etal.. 2013b: Bruff and Tikich. 2011): see Section 8.6.2.
16    Effluent from distillation treatment was found to contain up to 6.49 pCi/L for gross alpha (from 249
17    pCi/L prior to distillation) (Bruff and Tikich. 2011). Between 2010 and 2012, samples of wastewater
18    effluent from a western PA CWT contained a mean radium level of 4 pCi/L (Warner et al., 2013a).

      8.5.2.6. Organics
19    Because hydraulic fracturing wastewaters can contain various types of organic compounds that
2 0    each have different properties, specific treatment processes or series of processes are used to
21    target the various classes of organic contaminants. Effectiveness of treatment depends on the
22    specific organic compound and the technology employed (see Appendix F). It should be noted that
23    in many studies, rather than testing for several organic constituents, researchers often measure
24    organics in terms of biochemical oxygen demand and/or chemical oxygen demand, which are an
25    indirect measure of the amount of organic compounds in the water. Organic compounds  may also
26    be measured and/or reported in groupings such as total petroleum hydrocarbons (TPH) (which
27    include gasoline range organics (GROs) and diesel range organics (DROs)), oil and grease, VOCs
28    (which include BTEX), and SVOCs.

29    Based on examples found in literature, facilities have demonstrated the capability to treat for
30    organic compounds in hydraulic fracturing wastewaters using a single process or a series of
31    processes (Hayes etal.. 2014: Bruff and Jikich. 2011: Shafer. 2011] (see Table 8-6). The processes
32    can include anaerobic and aerobic biological treatment, coagulation, flocculation, flotation,
33    filtration, bioremediation, ultrafiltration,  MVR, and dewvaporation. Forward osmosis is an
34    emerging technology that may be promising for organics removal in hydraulic fracturing
35    wastewaters because it is capable of rejecting the same organic contaminants as commercially-
36    available pressure-driven processes (Drewes etal.. 2009).
                   This document is a draft for review purposes only and does not constitute Agency policy.
      June 2015                                   8-48                 DRAFT—DO NOTE CITE OR QUOTE

-------
      Hydraulic Fracturing Drinking Water Assessment              Chapter 8 - Wastewater Treatment and Waste Disposal


      8.5.2.7. Estimated Treatment Removal Efficiencies
 1    There are relatively few studies that have evaluated the ability of individual treatment processes to
 2    remove constituents from hydraulic fracturing wastewater and present the resulting water quality.
 3    Furthermore, although a specific technology may demonstrate a high removal percentage for a
 4    particular constituent, if the influent concentration of that constituent is extremely high, the
 5    constituent concentration in the treated water may still exceed permit limits and/or disposal
 6    requirements. Appendix Table F-4 presents the results of simple calculations pairing average
 7    hydraulic fracturing wastewater concentrations from Chapter 7 with treatment process removal
 8    efficiencies reported in the literature in Table F-2.

 9    As an example, radium in wastewater from the Marcellus Shale and Upper Devonian sandstones can
10    be in the thousands of pCi/L. With a 95% removal rate, chemical precipitation may result in effluent
11    that still exceeds 100 pCi/L. Distillation and reverse osmosis might produce effluent with
12    concentrations in the tens of pCi/L. A radium concentration of 120 pCi/L, however, could be
13    reduced to less than 5 pCi/L by RO or distillation. Wastewater with barium concentrations in the
14    range of 140 - 160 mg/L  (e.g., the Cotton Valley and Mesaverde tight sands) might be reduced to
15    concentrations under 5 mg/L by distillation and roughly 11-13 mg/L by RO. Barium concentrations
16    in the thousands of mg/L would be substantially reduced by any of several processes but might still
17    be relatively high and could exceed 100 m g/L. Table F-4 also illustrates the potential for achieving
18    low concentrations of organic compounds in wastewater treated with freeze-thaw evaporation or
19    advanced oxidation and precipitation.

20    This analysis is intended to highlight the potential impacts of influent concentration on treatment
21    outcome and to illustrate the relative capabilities of various treatment processes for an example set
22    of constituents. Removal efficiencies would differ and likely be greater with a full set of
23    pretreatment and treatment processes that would be seen in a CWT (see Table 8-6).

      8.5.3.  Design of Treatment Trains for CWTs
24    Based on the chemical composition of the hydraulic fracturing wastewater and the desired effluent
25    water quality, a series of treatment technologies will most likely be necessary. The possible
26    combinations of unit processes to formulate treatment trains are extensive. One report identified
27    41 different treatment unit processes that have been used in the treatment of oil and gas
28    wastewater and 19 unique treatment trains (combinations of unit processes) (Drewes et al.. 2009).
29    Fakhru'1-Razi et al. (2009) also provide examples of process flow diagrams that have been used in
30    pilot-scale and commercial applications for treating oil and gas wastewater. Figure 8-8 shows the
31    treatment train for the Pinedale Anticline Facility, which includes pretreatment for dispersed oil,
32    VOCs, and heavy metals and advanced treatment for removal of TDS, dissolved organics, and boron.
33    This CWT can either discharge to surface water or provide the treated wastewater to operators for
34    reuse.
                   This document is a draft for review purposes only and does not constitute Agency policy.
      June 2015                                    8-49                 DRAFT—DO NOTE CITE OR QUOTE

-------
      Hydraulic Fracturing Drinking Water Assessment
                                     Chapter 8 - Wastewater Treatment and Waste Disposal
            Produced Water
            Ui
   Oil/
  Water
Separator
Anaerobic
  Basin
Aeration
  Basin
Clarifier
                Sand
                Filter
                   Bioreactor
                 Membrane
                 Bioreactor
                  Reverse
                  Osmosis
                                                     Additional
                                                     Treatment
                                                    and Disposal
                Boron Ion
                Exchange
                                                                                 Frac Water
                                                                                 to Pipeline
                                                                                Distribution
                                                Clean Water
                                                Discharged to
                                                Environment
      Figure 8-8. Full discharge water process used in the Pinedale Anticline field.
                Source: Boschee (2012).

 1    Table 8-7 provides information on some CWTs in locations across the country and the processes
 2    they employ. The table also notes for each facility whether data are readily available on effluent
 3    quality. Comprehensive and systematic data on influent and effluent quality from a range of CWTs
 4    that treat to a variety of water quality levels is difficult to procure, rendering it challenging to
 5    understand removal efficiencies and resulting effluent quality, especially when a facility offers a
 6    range of water quality products (e.g., for reuse vs. discharge). For those facilities with NPDES
 7    permits, discharge monitoring report (DMR) data may be available for some constituents, although
 8    if the facility does not discharge regularly, these data will be sporadic.

 9    CWTs such as the Judsonia Central Water Treatment Facility in Arkansas, the Casella-Altela
10    Regional Environmental Services, and Clarion Altela Environmental Services (see Table 8-7)
11    facilities have NPDES permits and use MVR or thermal distillation for TDS removal. As of March
12    2015, the Pinedale Anticline Facility and the Judsonia Facility appear to be the only CWTs in Table
13    8-7 discharging to a surface water body.
                   This document is a draft for review purposes only and does not constitute Agency policy.
      June 2015                                    8-50                 DRAFT—DO NOTE CITE OR QUOTE

-------
Hydraulic Fracturing Drinking Water Assessment
                                                           Chapter 8 - Wastewater Treatment and Waste Disposal
Table 8-7. Examples of centralized waste treatment facilities.



Facility
Pinedale Anticline
Water Reclamation
Facility3













SEECO-Judsonia
Water Reuse
Recycling Facility











State
WY















AR












Description of Unit
Processes
Oil/water separation,
biological treatment,
aeration,
clarification, sand
filtration, bioreactor,
membrane
bioreactor, RO, and
ion exchange








Settling, biological
treatment, induced
gas flotation, and
MVR








Does CWT have
a NPDES permit
for discharge?
No - However,
facility is
permitted to
discharge under
40CFR435
Subpart E
(WY0054224).
Facility is
permitted to
discharge up to
25% of its
effluent stream




Yes-AR0052051










Does CWT
provide
effluent for
reuse?
Yes















Yes











Does CWT have
advanced process
for TDS removal?
Yes, RO (Boschee,
2014, 2012)














Yes, MVR











What is the status of the
facility as of January
2015?
The treatment plant
produces treated water
for reuse and for
discharge to surface
water.
The website indicates
the facility is in
operation and is
recycling to support
drilling operations and is
discharging to the New
Fork River
(http://hswater. squares
pace.com/pinedale-
anticline/).

The treatment plant
provides treated water
for reuse and for
discharge to surface
water. Based on DMR
data from late 2014-
early 2015, the system is
discharging treated
water to a surface water
body, though
intermittently.
Are effluent
quality data
available through
literature search?
Yes -DMR data
available on
Wyoming DEQ
website. Some
information can
also be obtained
from Shafer
(2011).








DMR data
available









June 2015
This document is a draft for review purposes only and does not constitute Agency policy.
                                 8-51
DRAFT—DO NOTE CITE OR QUOTE

-------
Hydraulic Fracturing Drinking Water Assessment
                                                               Chapter 8 - Wastewater Treatment and Waste Disposal



Facility
Eureka Resources -
Williamsport 2nd
Street Facility







Standing Stone
Facility, Bradford
County









State
PA









PA










Description of Unit
Processes
Settling, oil/water
separation, chemical
precipitation,
clarification, MVR.
Can treat with or
without IDS removal.




Settling, oil/water
separation, chemical
precipitation,
clarification, MVR,
crystallization





Does CWT have
a NPDES permit
for discharge?
No - However,
future plans to
install RO for
direct discharge
capability





Yes-PA0232351








Does CWT
provide
effluent for
reuse?
Yes









Yes









Does CWT have
advanced process
for TDS removal?
Yes, MVR









Yes, MVR,
crystallizer








What is the status of the
facility as of January
2015?
Per Erteletal. (2013),
the facility provides
treatment wastewater
for reuse and indirect
discharge.

The facility treats
entirely or almost
entirely hydraulic
fracturing wastewater.
The facility can provide
treated wastewater for
reuse and also has
received an NPDES
permit for direct
discharge.
The facility treats
hydraulic fracturing
wastewater.
Are effluent
quality data
available through
literature search?
No









No








June 2015
This document is a draft for review purposes only and does not constitute Agency policy.
                                   8-52
DRAFT—DO NOTE CITE OR QUOTE

-------
Hydraulic Fracturing Drinking Water Assessment
                                                               Chapter 8 - Wastewater Treatment and Waste Disposal



Facility
Wellington Water
Works













Casella Altela
Regional
Environmental
Services (CARES)
McKean Facility












State
CO














McKean
County,
PA













Description of Unit
Processes
Dissolved air
flotation, pre-
filtration,
microfiltration with
ceramic membranes,
activated carbon
adsorption. Water is
pumped to an aquifer
storage and recovery
well. Water is then
extracted and
treated with RO
(Alzahrani et al..
2013).

Pretreatment system
(not defined in
literature) and
thermal distillation











Does CWT have
a NPDES permit
for discharge?
Permit number
issued by CO
(61879)












Yes - PA0102288













Does CWT
provide
effluent for
reuse?
Yes














Yes














Does CWT have
advanced process
for TDS removal?
Yes, RO but only
after the water is
sent to an aquifer
storage and
recovery well










Yes -thermal
distillation













What is the status of the
facility as of January
2015?
PerStewart(2013b), the
facility is providing
treated wastewater for
reuse, for agricultural
use, to a shallow well to
augment the municipal
drinking water supply,
and for discharge to the
Colorado River.






The treatment plant is
capable of reuse and
recycle for fracturing
operations and surface
water discharge of
excess water. However,
the facility's website
indicates it is only
treating water for
reuse/recycle as of early
2015
( http ://ca resf orwater. co
m/location/cares-
mckean).

Are effluent
quality data
available through
literature search?
No














No -just NPDES
discharge
requirements











June 2015
This document is a draft for review purposes only and does not constitute Agency policy.
                                   8-53
DRAFT—DO NOTE CITE OR QUOTE

-------
Hydraulic Fracturing Drinking Water Assessment
                                                               Chapter 8 - Wastewater Treatment and Waste Disposal



Facility
Clarion Altela
Environmental
Services (CAES)
Facility










Terraqua Resource
Management (aka.
Water Tower Square
Gas Well Wastewater
Processing Facility)





Maggie Spain Water-
Recycling Facility







State
Clarion
County,
PA











Lycoming
County,
PA







Decatur,
TX






Description of Unit
Processes
Pretreatment system
(not defined in
literature) and
thermal distillation










Equalization tanks,
oil-water separation
via chemical addition
(sulfuricacid,
emulsion breaker),
pH adjustment,
coagulation,
flocculation, inclined
plate clarifier, sand
filtration
Settling, flash mixer
with lime and
polymer addition,
inclined plate
clarifier, surge tank,
MVR

Does CWT have
a NPDES permit
for discharge?
Yes - PA0103632













Yes - PA0233650
Permit pending
approval for
discharge to
stream (as of
4/17/2009)




No





Does CWT
provide
effluent for
reuse?
Yes













Yes









Yes






Does CWT have
advanced process
for TDS removal?
Yes- thermal
distillation












No - However,
TARM recognizes
that they can't
discharge, until
they install TDS
treatment




Yes - MVR






What is the status of the
facility as of January
2015?
The treatment plant
capable of reuse and
recycle for fracturing
operations and surface
water discharge of
excess water. However,
the facility's website
indicates it is only
treating water for
reuse/recycle as of early
2015
(http://caeswater.eom/t
echnology/).

According to its website
(last updated 2012), the
facility reuses/recycles
treated water for
fracturing operations
(http://www.tarmsolutio
ns.com/solutions/).



A 17-month pilot study
using a commercial-scale
mobile treatment facility
was concluded in 2011.
The status is unclear as
of early 2015.
Are effluent
quality data
available through
literature search?
No -just NPDES
discharge
requirements











No









Yes -Some
information can
be obtained from
Hayes et al.
(2014).

June 2015
This document is a draft for review purposes only and does not constitute Agency policy.
                                   8-54
DRAFT—DO NOTE CITE OR QUOTE

-------
Hydraulic Fracturing Drinking Water Assessment
                                                               Chapter 8 - Wastewater Treatment and Waste Disposal



Facility
Fountain Quail/NAC
Services - Kenedy







Purestream -
Gonzales facility




LINN Energy Fyre
Ranch -Granite Wash









State
Kenedy,
TX







Gonzales,
TX




Wheeler
County,
TX







Description of Unit
Processes
Oil-water separator,
coagulation,
flocculation,
sedimentation,
filtration, MVR.




Induced gas flotation
and MVR




Induced gas flotation
and MVR







Does CWT have
a NPDES permit
for discharge?
No








No





No







Does CWT
provide
effluent for
reuse?
Yes








Yes





Yes








Does CWT have
advanced process
for TDS removal?
Yes -MVR








Yes - MVR





Yes - MVR








What is the status of the
facility as of January
2015?
According to its website,
the facility
reuses/recycles treated
water for fracturing
operations
(http://www.aqua-
pure.com/operations/sh
ale/ford/ford. html).

Per Dahm and Chapman
(2014) commercial
operations deployed
March 2014 for
reuse/recycle for
fracturing operations.
AVARA system installed
for reuse/recycle in June
2014.
http://purestream.eom/i
ndex.php/water-
management/vapor-
recompression/photos-
and-videos
Are effluent
quality data
available through
literature search?
No








No





No







June 2015
This document is a draft for review purposes only and does not constitute Agency policy.
                                   8-55
DRAFT—DO NOTE CITE OR QUOTE

-------
Hydraulic Fracturing Drinking Water Assessment
                                                               Chapter 8 - Wastewater Treatment and Waste Disposal



Facility
Fluid Recovery
Service Josephine
Facility"











Fluid Recovery
Service Franklin
Facility












State
PA













PA













Description of Unit
Processes
Oil-water separator,
aeration, chemical
precipitation with
sodium sulfate, lime,
and a polymer,
inclined plate clarifier








Oil-water separator,
aeration, chemical
precipitation with
sodium sulfate, lime,
and a polymer,
inclined plate clarifier







Does CWT have
a NPDES permit
for discharge?
Expired -
PA0095273












Expired -
PA0101508










Does CWT
provide
effluent for
reuse?
No













No












Does CWT have
advanced process
for TDS removal?
No













No












What is the status of the
facility as of January
2015?
The facility claims to
have stopped accepting
Marcellus wastewater
September 30, 2011
(Ferraretal.,2013). It
treats conventional oil
and gas wastewater.
The facility will be
upgrading to include
evaporative technology
that will enable it to
attain monthly average
TDS levels of 500 mg/L
or less.
This facility is not
accepting wastewater
from hydraulic fracturing
operations as of January
2015.
The facility will be
upgrading to include
evaporative technology
that will enable it to
attain monthly average
TDS levels of 500 mg/L
or less.
Are effluent
quality data
available through
literature search?
Yes -Some
effluent results
obtained from
Ferrar et al.
(2013) and
Warner et al.
(2013a). Also
minimal DMR
data from the
EPA.




Minimal DMR
data from the
EPA.









June 2015
This document is a draft for review purposes only and does not constitute Agency policy.
                                   8-56
DRAFT—DO NOTE CITE OR QUOTE

-------
Hydraulic Fracturing Drinking Water Assessment
                                                               Chapter 8 - Wastewater Treatment and Waste Disposal
Facility
Hart Resources-
Creekside Facility15
State
PA
Description of Unit
Processes
Oil-water separator,
aeration, chemical
precipitation with
sodium sulfate, lime,
and a polymer,
inclined plate clarifier
Does CWT have
a NPDES permit
for discharge?
Expired -
PA0095443
Does CWT
provide
effluent for
reuse?
No
Does CWT have
advanced process
for TDS removal?
No
What is the status of the
facility as of January
2015?
This facility is not
accepting wastewater
from hydraulic fracturing
operations as of January
2015.
The facility will be
upgrading to include
evaporative technology
that will enable it to
attain monthly average
TDS levels of 500 mg/L
or less.
Are effluent
quality data
available through
literature search?
Minimal DMR
data from the
EPA.
a For Pinedale Anticline Water Reclamation Facility, surface water discharges are permitted under 40 CFR 435 Subpart E (beneficial use subcategory agricultural and wildlife
water) not 40 CFR 437 (the discharge permit for CWTs). For the purposes of this assessment, this facility is included with CWTs.
As of May 15, 2013, these facilities are under an administrative order (AO). According to the AO, these facilities must comply with a monthly effluent limit for IDS not to
exceed 500 mg/L. This will allow them to treat high-saline wastewaters typical of unconventional oil and gas operations. To meet the requirements of the AO, they have
applied to PADEP for a NPDES permit and are planning to install treatment for TDS.
June 2015
This document is a draft for review purposes only and does not constitute Agency policy.
                                   8-57
DRAFT—DO NOTE CITE OR QUOTE

-------
      Hydraulic Fracturing Drinking Water Assessment               Chapter 8 - Wastewater Treatment and Waste Disposal


      8.6. Potential Impacts on Drinking Water Resources
 1    Several articles have noted potential effects of hydraulic fracturing wastewater on water resources
 2    fVengoshetal..2Q14: OlmsteadetaL 2013: Rahmetal.. 2013: States etal.. 2013: VidicetaL 2013:
 3    Rozell and Reaven. 2012: Entrekinetal.. 2011). with one study using probability modeling
 4    indicating that water pollution risk associated with gas extraction in the Marcellus Shale is highest
 5    for the wastewater disposal aspects of the operation [Rozell and Reaven, 2012]. Whether drinking
 6    water resources are affected by hydraulic fracturing wastewater depends at least in part upon the
 7    characteristics of the wastewater, the form of discharge or other management practice, and the
 8    processes used if the wastewater is treated. Other site-specific factors (e.g., size of receiving water
 9    and volume of wastewater) determine the magnitude and nature of potential effects, but a thorough
10    exploration of local factors is beyond the scope of this assessment The majority of hydraulic
11    fracturing wastewater is either injected into a disposal well or, in the case of the Marcellus region,
12    reused for other hydraulic fracturing jobs. Potential impacts on drinking water resources may occur
13    on a local level through several routes: treated wastewater may be discharged directly from
14    centralized waste treatment facilities (CWTs) or indirectly from publicly owned treatment works
15    (POTWs) that receive CWT effluent; sediments in water bodies receiving effluent may accumulate
16    contaminants; spills  or leaks may be associated with on-site storage or transportation (see Chapter
17    7); and in previous years, hydraulic fracturing wastewater treated at POTWs was discharged to
18    surface waters.

19    It has been suggested that the most significant effects of hydraulic fracturing on surface water
2 0    quality are related to discharges of partially treated wastewater, although these effects vary
21    according to region (Kuwayamaetal.. 2015]. A recent study (Bowenetal.. 2015] concluded that
22    there is currently no clear evidence of national-level trends in surface water quality (as measured
23    by specific conductivity  and chloride] in areas where unconventional oil and gas production is
24    taking place. These authors note that available national level databases have limitations for
25    assessing this question.

26    Pits and impoundments associated with waste management may have impacts on drinking water
27    resources and are discussed in Chapter  7. In addition, unauthorized discharge of wastewater is a
28    potential mechanism for impacts on drinking water resources. Descriptions of several incidents and
29    resulting legal actions have been publicly reported. However, such events are not generally
30    described in the scientific literature, and the prevalence of this type of activity is unclear.

31    Important considerations regarding the potential impact of hydraulic fracturing wastewater on a
32    receiving waterbody include whether constituents in the wastewater are known to have health
33    effects, if they are regulated drinking water contaminants,  or if they may give rise to regulated
34    compounds. For some classes of constituents, such as disinfection by-product (DBF] precursors,
35    considerable research exists. For others, information is limited regarding their concentrations in
36    effluents and whether they are likely to affect drinking water at intakes. The following subsections
37    identify several classes of constituents known to occur in hydraulic fracturing wastewater, discuss
38    whether potential impacts are likely, and provide specific examples of information gaps.
                   This document is a draft for review purposes only and does not constitute Agency policy.
      June 2015                                    8-58                  DRAFT—DO NOT CITE OR QUOTE

-------
      Hydraulic Fracturing Drinking Water Assessment              Chapter 8 - Wastewater Treatment and Waste Disposal


      8.6.1.  Bromide and Chloride
 1    Bromide and chloride are two constituents commonly found in high-total dissolved solids (IDS)
 2    hydraulic fracturing wastewater. As noted in section 8.3.1.1, chloride is a regulated contaminant
 3    with a secondary MCL standard of 250 mg/L. Bromide is not regulated but is of concern due to its
 4    role in the formation of DBFs [Parker etal.. 2014: Krasner. 2009] (see Appendix F for information
 5    on DBF formation). High-TDS wastewaters from the Marcellus Shale can be of concern because the
 6    limited availability of underground injection for disposal can result in a higher rate of discharge of
 7    treated wastewater to surface waters compared to other parts of the country. In response to
 8    concerns in part over bromide in discharges, operators in Pennsylvania have discontinued the
 9    practice of sending wastewater from hydraulic fracturing operations to POTWs [States etal., 2013].
10    Also, CWTs have been shifting towards treatment of those wastewaters for reuse rather than
11    discharging to surface water bodies [Hammer and VanBriesen. 2012].

12    States etal. [2013] found a strong correlation between bromide concentrations in source water
13    from the Allegheny River in Pennsylvania and the percentage of brominated trihalomethanes in
14    finished drinking water. The authors noted that source water containing 50 |ig/L bromide resulted
15    in treated water with approximately 62% of its finished water total trihalomethanes consisting of
16    bromoform, dibromochloromethane, and bromodichloromethane. Source water containing 150
17    Mg/L bromide resulted in finished water TTHMs composed of approximately 83% brominated
18    species. Allegheny River bromide concentrations measured during the study ranged from less than
19    25 |ig/L to 299 |ig/L, with the highest bromide concentrations measured under low-flow
20    conditions. Industrial wastewater sites accounted for approximately 50% of the increase in
21    bromide load as water moved downriver.

22    In addition, a related constituent, iodide can be a constituent in hydraulic fracturing wastewater
23    (see Chapter 7). Although its effects have not been as well documented as those associated with
24    bromide (Xuetal.. 2008], iodide raises some of the same concerns (such as DBF formation] as
25    bromide does (Parker etal.. 2014: Krasner. 2009]. lodinated DBFs are not regulated by the EPA as
26    of early 2015.

27    As discussed in Section 8.5, removal of dissolved solids, including chloride and bromide, requires
28    advanced treatment processes such as reverse osmosis (RO], distillation, evaporation, or
29    crystallization. Unless the treatment plant receiving the high-TDS wastewater employs processes
30    specifically designed to remove these constituents, effluent discharge may contain high levels of
31    bromide and chloride. Drinking water treatment plants with intakes downstream of these
32    discharges may receive water with correspondingly higher levels of bromide and chloride and may
33    have difficulty complying with Safe Drinking Water Act (SDWA] regulations related to DBFs.

      8.6.1.1. Effects on Receiving Streams
34    Studies show that discharges from oil and gas wastewater treatment facilities can elevate TDS,
35    bromide, and chloride levels in receiving waters (States etal.. 2013: Wilson and Van Briesen. 2013].
36    Wilson and Van Briesen (2013] measured bromide, chloride, and other constituents at water
37    intakes downstream of wastewater discharges for three years in the Monongahela River in western
38    Pennsylvania. By evaluating water chemistry data in the context of flow measurements, the authors

                   This document is a draft for review purposes only and does not constitute Agency policy.
      June 2015                                   8-59                  DRAFT—DO NOT CITE OR QUOTE

-------
      Hydraulic Fracturing Drinking Water Assessment              Chapter 8 - Wastewater Treatment and Waste Disposal


 1    attributed an overall decrease in bromide concentrations from 2010 to 2012 to a decrease in
 2    bromide loading; they note that this is likely to be associated with a decrease in management of
 3    fossil fuel wastewater at treatment plants that discharge to surface water.

 4    Although treatment plant effluents will be diluted upon reaching the receiving water, the dilution
 5    may not be adequate to avoid water quality problems if there are existing pollutant loads in the
 6    waterbody from contributors such as acid mine drainage or power plant effluents [Ferrar etal.,
 7    2013). Warner etal. (2013a] evaluated effluent from the Josephine Brine Treatment Facility (which
 8    treated both conventional and unconventional oil and gas wastewater at the time of the study) and
 9    concluded that even a 500 to 3,000-fold dilution of the wastewater would not reduce bromide
10    levels to background. In addition, downstream levels of chloride in the receiving stream were
11    elevated, with a downstream value of 88 mg/L as compared to an upstream value of 18 mg/L.

12    A study by Hladiketal. [2014] focused on sampling at sites downstream and near the outfalls of
13    plants that treated oil and gas wastewater in Pennsylvania. The authors documented brominated
14    and iodinated DBFs (e.g., dibromochloronitromethane (DBNM); dibromoiodomethane) at the
15    outfalls of CWTs and POTWs and noted that this DBF signature was different than for those plants
16    that did not accept oil and gas wastewater. For example, concentrations of
17    dibromochloronitromethane ranged from 0.26 to 8.7 |ig/L, and dibromoiodomethane was
18    measured at 0.98 and 1.3 |ig/L; neither compound was  detected at an upstream site or at the outfall
19    of the POTW not accepting oil and gas wastewater. These brominated and iodinated compounds are
20    considered more toxic than other types of DBFs  (Richardson etal.. 2007). Hladik et al. note that
21    these elevated DBF levels could contribute to DBFs at downstream drinking water intakes and can
22    also be an indicator of the potential for more highly brominated and iodinated DPBs forming in
23    drinking treatment plants downstream of these discharges (Hladiketal., 2014]. The sites studied
24    by Hladiketal. (2014) received wastewater from both conventional and unconventional oil and gas
25    development

26    Research suggests that a relatively small portion of hydraulic fracturing wastewater effluent can
27    notably affect DBF  formation. In laboratory studies, Parker etal. (2014) diluted hydraulic fracturing
28    wastewater from the Marcellus and Fayetteville  shales  with Allegheny and Ohio River waters and
29    then disinfected the mixtures. In chlorinated samples containing as little as 0.01% hydraulic
30    fracturing wastewater, the THM composition shifted significantly away from chloroform species to
31    a greater representation of brominated and iodinated species.

3 2    Elevated concentrations of bromide in effluents can place a burden on downstream drinking water
33    treatment systems. States etal. (2013] studied influent and finished water at the Pittsburgh Water
34    and Sewer Authority (PWSA) drinking water system, concluding that elevated bromide in the
35    source water led to elevated total trihalomethanes (TTHM) formation in the treated drinking water.
36    The authors also noted a substantial increase in the percentage of brominated TTHMs (States etal..
37    2013], as discussed above. The utility modified their treatment process and proposed
38    improvements to their storage facilities to address the elevated TTHM levels in the distribution
39    system (Chester Engineers, 2012].
                   This document is a draft for review purposes only and does not constitute Agency policy.
      June 2015                                   8-60                  DRAFT—DO NOT CITE OR QUOTE

-------
      Hydraulic Fracturing Drinking Water Assessment              Chapter 8 - Wastewater Treatment and Waste Disposal


      8.6.1.2. Modeling
 1    The EPA's contaminant modeling shows that that the strategies most likely to reduce bromide
 2    impacts on downstream users include reducing effluent concentrations (e.g., discharging flowback
 3    versus produced water), discharging during higher stream flow periods, and using a pulsing or
 4    intermittent discharge. Weaver et al. [In Press] developed a computer model to estimate river and
 5    stream bromide concentrations after treated water discharges. The model utilizes existing data for
 6    bromide concentrations in produced water, flowback, and mixtures, combined with existing stream
 7    flow data from USGS stations in Pennsylvania. The model parameters include steady state versus
 8    transient inputs to receiving waters, high and low streamflow months, varying effluent
 9    concentration and types (produced, flowback, and mixed). For steady-state scenarios in the model,
10    bromide concentrations are lowest under high flow conditions with lower concentrations of
11    effluent (flowback and mixed water).

12    A source apportionment study conducted by the EPA considered the relative contributions of
13    bromide, chloride, nitrate, and sulfate from CWTs primarily treating hydraulic fracturing
14    waste water to the Allegheny River Basin and to water at two downstream public water system
15    intakes on the Allegheny River (U.S. EPA. 2015p). The Allegheny River and its tributaries receive
16    runoff and discharges containing an array of contaminants, including these anions. Contaminant
17    sources include discharges from CWTs for oil and gas wastewater, runoff from acid mine drainage
18    and mining operations, discharges from coal-fired electric power stations, industrial wastewater
19    treatment plant effluents, and POTW discharges. The Allegheny River is also the water supply for
20    thirteen public water systems that serve over 500,000 people in western Pennsylvania.

21    In Pennsylvania, wastewater produced from hydraulic fracturing of the Marcellus formation has
22    been mostly diverted from CWTs and POTWs that discharge to public waters in the state (Hammer
23    and VanBriesen, 2012). Wastewater produced from hydraulic fracturing of non-Marcellus
24    formations, however, continues to be sent to surface-discharging facilities on the Allegheny River.

2 5    The source apportionment study considered contributions of bromide, chloride, nitrate and sulfate
26    to public water supplies from CWTs and other upriver sources by: developing chemical source
27    profiles, or fingerprints, for discharges upstream of the public water system intakes; characterizing
28    water quality in the river upstream and downstream of the CWTs, electric generating stations, and
29    industrial facilities; characterizing the water quality at the public water system intakes; and
30    analyzing the sampling data collected with the EPA Positive Matrix Factorization (PMF) receptor
31    model in order to quantify relative contributions of contaminant sources to the anions found at the
3 2    public water system intakes. The study focused on low-flow conditions.

33    CWTs and coal-fired power plants with flue gas desulfurization were found to contribute bromide
34    to the two public water supply intakes. Although acid mine drainage also contributed bromide, its
35    contribution was minor (9% at one intake) compared to the contributions from the CWTs (89%
36    and 3 7% at the two intakes) and coal-fired power plants (50-59% atone intake). The CWTs, coal-
37    fired power plants, and acid mine drainage combined accounted for 88-89% of the bromide at one
38    intake and 96% of the bromide at the other intake.
                   This document is a draft for review purposes only and does not constitute Agency policy.
      June 2015                                   8-61                 DRAFT—DO NOT CITE OR QUOTE

-------
      Hydraulic Fracturing Drinking Water Assessment              Chapter 8 - Wastewater Treatment and Waste Disposal


      8.6.1.3. Summary
 1    Most drinking water treatment plants are not designed to address high concentrations of IDS
 2    (including bromide and iodide), limiting their options for restricting the formation of brominated
 3    and iodinated DBFs. Tighter restrictions on IDS in effluent from POTWs and CWTs have led to a
 4    reduction in in-stream bromide concentrations. Advanced treatment processes at CWTs such as
 5    reverse osmosis, distillation, evaporation, or crystallization can reduce chloride, bromide, and
 6    iodide in surface waters. Strategies such as reducing effluent concentrations, discharging during
 7    higher stream flow periods, and utilizing a pulsing or intermittent discharge could also reduce the
 8    potential impact of elevated TDS on drinking water treatment plants.

      8.6.2.  Radionuclides
 9    Potential impacts on drinking water resources from technologically enhanced naturally occurring
10    radioactive material (TENORMs) associated with hydraulic fracturing waste water may arise from a
11    number of sources, including: treated wastewater that does not have adequately reduced
12    radionuelide concentrations, accumulation of radionuelides in surface  water sediments
13    downstream of wastewater treatment plant discharge points, migration from soils that have
14    accumulated radionuclides from previous activities such as pits or land application, and inadequate
15    management of treatment plant solids that have accumulated radionuclides (such as filter cake).

16    In Pennsylvania  between 2007 and 2010, TENORM-bearing produced  wastewaters were sentto
17    POTWs, which are generally not required to monitor for radioactivity (Resnikoff etal., 2010).
18    Although the practice  of management of Marcellus waters via POTWs has declined, there is still
19    potential for input of radionuclides to surface waters via discharge of CWT effluents either directly
20    to surface waters or indirectly through discharge to POTWs.

21    Data regarding TENORM content in oil and gas wastes that are treated and discharged to surface
22    waters are limited. However, a recent study by the Pennsylvania Department of Environmental
23    Protection (PA DEP) (PADEP. 2015b) provides information that helps fill this data gap. The study
24    began in 2013 and examined radionuclide (radium-226, radium-228, K-40, gross alpha, gross beta)
25    levels at 29 wastewater plants in Pennsylvania that cover a range of both sources and treatment
26    plant types, including  POTWs, CWTs that treat oil and gas wastewaters and can discharge to surface
27    water or a POTW, and zero liquid discharge facilities treating oil and gas wastewater. Four of the 10
28    discharging CWTs sampled during the study discharged to surface water under a National Pollution
29    Discharge Elimination System (NDPES) permit, and the others discharged to POTWs. Six of the
30    POTWs in the study received effluent from a CWT along with municipal wastewater. The CWTs in
31    the study are not described as receiving exclusively Marcellus wastewater, but the study itself was
32    motivated by concerns over an increase in radionuclides in oil and gas wastes observed during the
33    expansion of Marcellus Shale production.

34    The POTWs receiving  influent from CWTs treating oil and gas wastewater (along with municipal
35    wastewater influent) had average effluent radium-226 concentrations of 103 pCi/L (unfiltered) and
36    129 pCi/L (filtered) (filtration is used to remove very fine particulates from the water). Those
37    POTWs not receiving influent from CWTs treating oil and gas wastewater effluent had higher
38    average radium-226 values in unfiltered samples (145 pCi/L) and lower values for filtered samples

                   This document is a draft for review purposes only and does not constitute Agency policy.
      June 2015                                   8-62                 DRAFT—DO NOT CITE OR QUOTE

-------
      Hydraulic Fracturing Drinking Water Assessment              Chapter 8 - Wastewater Treatment and Waste Disposal


 1    (47 pCi/L). For perspective, the maximum contaminant level (MCL) for radium-226 plus radium-
 2    228 is 5 pCi/L. For reference, radium-226 in river water generally ranges from 0.014 pCi/L to 0.54
 3    pCi/L (0.5 to 20 mBq/L) flAEA. 20141 The results of the POTW sampling are inconclusive as to
 4    whether the effluents from POTWs receiving CWT-treated oil and gas wastewater are routinely
 5    higher than the effluents from those without this type of influent

 6    For the CWTs in the PA DEP study, average radium-226 content in the effluents was an order of
 7    magnitude higher than in effluents from the POTWs (1,840 pCi/L unfiltered, 2,100 pCi/L, filtered).
 8    The effluent averages were similar to averages for the influent concentrations, although median
 9    concentrations in the effluents were much lower than in the influents. Effluent from zero-discharge
10    facilities averaged 2,610 pCi/L radium-226 and 295 pCi/L radium-228, although these effluents
11    would most likely be reused as fracturing fluid (PADEP, 2015b]. The authors do note a potential for
12    environmental effects from spills of influent or effluent from zero-discharge facilities.

13    Warner etal. (2013a] noted that if the activities of radium-226 and radium-228 inMarcellus brine
14    influent at the CWT they studied are similar to those reported by other researchers (Rowan etal..
15    2011). then the CWT achieved a 1,000-fold reduction in radium content via a process of radium
16    coprecipitation with barium sulfate. The detection of radium in effluents from this CWT (mean
17    values of 4 pCi/L of radium-226 and 2 pCi/L of radium-288)  even with what may be high treatment
18    removal efficiency underscores the fact that effluent concentrations depend not only upon the
19    treatment processes used but also the influent concentration.

2 0    An additional concern related to evaluation of radionuclide concentrations in wastewater is that the
21    high TDS content of hydraulic fracturing wastewater can result in poor recovery of chemical
2 2    constituents when using wet chemical techniques, leading to underestimations of constituent
23    concentrations. In particular, recovery for radium may be as  low as <1% (Nelson etal., 2014].
24    Underestimation of radium content may lead to failure in identifying an impact or potential impact
25    on drinking water resources.

26    In addition to concerns over of the potential for TENORM in discharges to surface  waters, there are
2 7    may be a legacy of accumulation of radionuclides in surface water sediments. Studies of effluent,
2 8    stream water, and stream sediment associated with a CWT in western Pennsylvania that has
29    treated both conventional and unconventional oil and gas wastewaters indicate that radium-226
30    levels in stream sediment samples at the point of discharge are approximately 200 times greater
31    than upstream and background sediments. This indicates the potential for accumulation of
32    contaminants in localized areas of wastewater discharge facilities (Warner etal.. 2013a]. Although
33    the CWT in question also accepted conventional oil and gas wastewater, Warner etal. (2013a]
34    observed thatthe radium-228/radium-226 ratio in the river sediments near the discharge (0.22-
35    0.27) is consistent with ratios in Marcellus wastewater. The authors interpret this as an indication
36    that the radium accumulated in the sediments originated from the discharge of treated
37    unconventional oil and gas wastewater. Another study, however, did not find elevated levels of
38    alkali earth metals (including radium) in sediments just downstream  of the discharge points of five
39    POTWs that had previously treated Marcellus wastewater (Skalaketal., 2014). Accumulation of
40    contaminants in sediment may be dependent on treatment processes and their removal rates for

                   This document is a draft for review purposes only and does not constitute Agency policy.
      June 2015                                   8-63                  DRAFT—DO NOT CITE OR QUOTE

-------
      Hydraulic Fracturing Drinking Water Assessment              Chapter 8 - Wastewater Treatment and Waste Disposal


 1    various constituents as well as stream chemistry and hydrologic characteristics. Contamination
 2    with radium-226 would be potentially be long lived; the half-life of radium-226 is approximately
 3    1,600 years, while the half-life of radium-228 is 5.76 years.

 4    The recent PA DEP study [PADEP. 2015b] found thatthe radium-226 content in sediments near the
 5    discharge points for POTWs receiving treated oil and gas effluent from CWTs (along with their
 6    municipal wastewater influent) exceeds typical background soil levels of approximately 1 to 2
 7    pCi/g of radium-226 and radium-228. The authors conclude that wastewater effluent is the most
 8    likely source for the radium in these samples. Results indicate an average of 9.00 pCi/g radium-226
 9    and 3.52 pCi/g radium-228 in sediments near outfalls of POTWs. Sediments at 4 CWTs receiving oil
10    and gas wastewater and that discharge to surface water have much higher average concentrations
11    of 84.2 pCi/g for radium-226 and 19.8 pCi/g for radium-228.  However, the concentrations of
12    radium in the sediments does not correlate with concentrations of radium in the effluents
13    suggesting that sorption over time affects the concentration of radium in the sediments [PADEP,
14    2015b].

15    The association of radium with sediments near discharge points is attributed to adsorption of
16    radium to the sediments, a process governed by factors such as the salinity of the water and
17    sediment characteristics. In particular, radium has a high affinity for iron and manganese
18    (hydr)oxides in sediment. Increased salinity promotes desorption of radium from sediments, while
19    lower salinity promotes adsorption, with radium adsorbing particularly strongly to sediments high
20    in iron and manganese (hydro)oxides [Porcellietal.. 2014: Gonneea et al.. 2008). Warner et al.
21    [2013a] speculate that the discharge of saline CWT effluent into less saline stream water facilitates
22    sorption of radium onto streambed sediments. The long-term fate of radium sorbed to sediments
23    depends upon changes in water salinity and the sediment properties, including any redox processes
24    that may affect iron and manganese minerals in the sediments.

25    Other solids may contain radionuclides; filter cake samples from treatment at POTWs were found
26    by PADEP [2015b] to have radium contents greater than typical soil concentrations, and they
27    exhibited a large variation. Filter cake from CWTs had radium concentrations higher than in POTW
28    filter cake. The authors conclude that although the risk to workers and the public from handling
29    and temporary storage of these materials is minimal, there may be environmental risks from spills
30    or long term disposal. There could be impacts on surface waters through spills or effects on ground
31    waters from landfill leachate.

3 2    Radionuclide accumulation in CWTs or POTWs may continue to affect the plant even after
33    discontinuing treatment of high radionuclide wastewater. Radium can adsorb onto scales in pipes
34    and tanks and will also co-precipitate calcium, barium, and strontium in sulfate minerals [USGS,
35    2014e]. Pipe scale in oil and gas production facilities can have radium concentrations as high as
36    154,000 pCi/g, although concentrations of less than 13,500 pCi/g are more common  [Schubert et
37    al.. 2014]. A similar issue, the potential for accumulation and possible release of radionuclides and
38    other trace inorganic constituents in water distribution systems has gained attention, with the
39    potential for drinking water concentrations to exceed drinking water standards [Water Research
40    Foundation, 2010].  Scale eventually removed from pipes or other equipment may end up in

                   This document is a draft for review purposes only and does not constitute Agency policy.
      June 2015                                   8-64                 DRAFT—DO NOT CITE OR QUOTE

-------
      Hydraulic Fracturing Drinking Water Assessment              Chapter 8 - Wastewater Treatment and Waste Disposal


 1    landfills and then leach into groundwater or run off to a surface water body [USGS, 2013c].
 2    Although barium sulfate phases are relatively insoluble, one study demonstrates that barium
 3    sulfate scales that were buried in soil could be reduced by microbially mediated processes, allowing
 4    release of co-precipitated elements such as radium due to leaching by rainwater [Swannetal.,
 5    2004]. Monitoring would be needed in order to ascertain the potential for accumulation and release
 6    of radionuclides from systems that have treated or continue to treat hydraulic fracturing
 7    wastewaters with elevated TENORM concentrations.

 8    Accumulation of radionuclides (potassium, thorium, bismuth, radium, and lead) has been evaluated
 9    in two pits in Texas that have stored fluids associated with hydraulic fracturing [Rich and Crosby,
10    2013]. Gamma radiation in these pits has been found to vary from 8 to 23 pCi/g, with beta radiation
11    varying from 6 to 1329 pCi/g [Rich and Crosby, 2013]. Although the study sample size was small,
12    the results suggest that radionuclides associated with sediments from some reserve pits could have
13    potential impacts on surface waters or ground waters. This could happen through migration of
14    affected sediments or soils to surface waters or through leaching to ground water.

15    Salt and radionuclide accumulation can occur near road spreading sites; one study in Pennsylvania
16    found a 20% increase in radium concentrations in soils near roads where wastewaters from
17    conventional operations had been spread for de-icing [Skalak et al.. 2014]. Accumulation of
18    radionuclides in soils near roads presents a vehicle for potential impacts on drinking water
19    resources. The extent to which hydraulic fracturing wastewater contributes to this depends upon
20    state-level regulations regarding whether hydraulic fracturing wastewater can be used for road
21    spreading.

22    Effluents and receiving waters can be monitored for radionuclides. Research suggests that radium-
23    226 and radium-228 are the predominant radionuclides in Marcellus Shale wastewater, and they
24    account for most of the gross alpha and gross beta activity in the waters studied [Rowanetal.,
25    2011]. Gross alpha and gross beta measurement may therefore serve as an effective screening
26    mechanism for overall radionuclide concentrations in hydraulic fracturing wastewater. This in turn
27    can help in evaluating management strategies. Portable gamma spectrometers allow rapid
28    screening of wastewater effluents. Sediments may also be measured for radionuclide
29    concentrations at discharge points.

      8.6.3.  Metals
3 0    Given the presence in hydraulic fracturing wastewaters of some heavy metals, as well as barium
31    and strontium concentrations that can reach hundreds or even thousands of mg/L (see Table 7-10],
32    surface waters maybe impacted if discharges from CTWs or POTWs indirectly receiving oil and gas
33    wastewater via CWTs are not managed appropriately. Spills may also affect surface waters.

34    Common treatment processes, such as coagulation, are effective at removing many metals (see
35    Section 8.5.2.4].  A request by the EPA for effluent sampling from seven facilities in Pennsylvania
36    treating oil and gas wastewaters revealed low to modest concentrations of copper (0-50 ug/L], zinc
37    (14-256 u.g/L], and nickel (8-22 u,g/L] (U.S. EPA. 2015d. e]. However, metals such as barium and
38    strontium have been found to range from low to elevated in some CWT effluents. For the year 2011,


                   This document is a draft for review purposes only and does not constitute Agency policy.
      June 2015                                   8-65                 DRAFT—DO NOT CITE OR QUOTE

-------
      Hydraulic Fracturing Drinking Water Assessment              Chapter 8 - Wastewater Treatment and Waste Disposal


 1    for example, effluent from a Pennsylvania CWT had average barium levels ranging from 9 to 98
 2    mg/L [PADEP. 2015a]. That facility was operating with a barium removal stage and was treating
 3    both conventional and hydraulic fracturing wastewater, although effluent concentrations dropped
 4    after May, 2011. The facility is scheduled to upgrade its TDS removal capabilities.

 5    Data collected by the EPA between October 2011 and February 2013 at seven Pennsylvania
 6    facilities indicate effluent barium concentrations ranging from 0.35 to 25 mg/L (median of 3.5 mg/L
 7    and average of 6.7 mg/L). Strontium concentrations ranged from 0.36 to 546 mg/L (median of 297
 8    mg/L and mean of 236 mg/L  (U.S. EPA. 2015e). A December 2010 effluent sampling effort in at a
 9    discharging CWT in Pennsylvania reported average barium and strontium concentrations of 27
10    mg/L and nearly 3,000 mg/L, respectively (eight samples from one plant) (Volz etal., 2011). The
11    facility treats conventional oil and gas waste waters, and it also received Marcellus wastewater until
12    September, 2011.

13    Limited data are available on metal concentrations in wastewater and treated effluents that are
14    directly discharged; additional information would be needed to assess whether there will be
15    downstream effects on drinking water utilities. NPDES discharge permits, which restrict TDS
16    discharge concentrations, will likely reduce metal effluent concentrations due to the additional
17    treatment necessary to minimize TDS.

      8.6.4.  Volatile Organic Compounds
18    Benzene is a common constituent in hydraulic fracturing wastewater, and it is of concern due to
19    recognized human health effects. A wide range of concentrations of BTEX compounds occurs in
20    wastewater from the Barnett and Marcellus shales. Natural gas formations generally produce more
21    BTEX than oil formations (Veil etal.. 2004). Generally, lower concentrations of BTEX occur in
22    wastewater from coalbed methane (CBM) production (see Appendix Table E-9). Processes such as
23    aeration or dissolved air flotation can remove volatile organic compounds (VOCs) during treatment,
24    but if treatment is not adequate, the VOCs may reach water resources. The average concentration of
25    benzene in a December 2010  sampling effort was 12 |ig/L in the discharge of a Pennsylvania CWT
26    (Volz etal.. 2011). The facility was  receiving wastewater from both conventional and
27    unconventional operations at that time. Ferrar etal. (2013) measured mean concentrations of
28    benzene, toluene, ethylbenzene, and xylene in effluents from the same facility, and mean
29    concentrations among the four compounds ranged from about 2 to 46 [ig/L. Concentrations were
30    lower for samples taken after May  19, 2011 than before, and the effect was considered statistically
31    significant. The treatment processes at this facility do not include aeration.

3 2    Leakage from pits or spills creates another potential route of entry to drinking water resources.
33    VOCs have been measured in  groundwater near the Duncan Oil Field in New Mexico, downgradient
34    of an unlined pit storing oil and gas wastewater (Sumi, 2004: Eiceman, 1986). VOCs and oil were
35    also found in groundwater about 213 feet (65m) downgradient from an unlined pit in Oklahoma
36    (Kharakaetal.. 2002).
                   This document is a draft for review purposes only and does not constitute Agency policy.
      June 2015                                   8-66                  DRAFT—DO NOT CITE OR QUOTE

-------
      Hydraulic Fracturing Drinking Water Assessment              Chapter 8 - Wastewater Treatment and Waste Disposal


      8.6.5.  Semi-Volatile Organic Compounds
 1    Little is known about the fate of the SVOC, 2-butoxyethanol (2-BE) (an antifoaming and anti-
 2    corrosion agent used in slick-water) [Volzetal.. 2011] or its potential impact on surface waters,
 3    drinking water resources, or drinking water systems. This compound is very soluble in water and is
 4    subject to bio degradation, with a half-life estimation of 1-4 weeks in the environment [Wess etal.,
 5    1998]. The EPA has not classified 2-BE (or other glycol ethers] for carcinogenicity. 2-BE was
 6    detected in the discharge of a Pennsylvania CWT at concentrations of 59 mg/L (Volz etal.. 2011].
 7    Ferrar etal. (2013] detected 2-BE in the effluents from a CWT in western Pennsylvania at average
 8    concentrations of 34 - 45 mg/L; the latter value was measured when the CWT was receiving only
 9    conventional oil  and gas wastewater. Data are lacking on 2-BE concentrations in surface waters that
10    receive treated effluents from hydraulic fracturing wastewater treatment systems.

11    Polycyclic aromatic hydrocarbons are another common group of semi-volatile organic compounds
12    (SVOCs] in oil  and gas wastewater. They have been detected in soils 164 feet (50 m] downgradient
13    of an unlined pit in New Mexico (Sumi. 2004: Eiceman. 1986]. PAHs were also found in birds in
14    wetlands fed by oil and gas wastewater discharges in Wyoming (Ramirez. 2002].

      8.6.6.  Oil and Grease
15    Oil and gas wastewater often contains oil and grease from the formation or from oil-based drilling
16    fluids. Typically, oil and grease are separated from the wastewater before discharge either by a heat
17    treatment or by allowing gravity separation followed by skimming. If these processes are
18    inefficient, oil  and grease may be integrated with the discharge to surface waters. For example, in
19    some cases, oil and grease are allowed to separate in pits, and water is then withdrawn from the
20    lower part of the pit with a standpipe. If the oil layer is allowed to drop to the level of the standpipe
21    or if the water is agitated, oil and grease may be discharged along with the water. Oil and grease are
22    also often dispersed in wastewater in the form of small droplets that are 4 to 6 microns in diameter.
23    These droplets can be difficult to remove using typical oil/water separators (Veil etal.. 2004]. In a
24    study by the U.S. Fish and Wildlife Service regarding permitted oil and gas discharges between
25    1996 and 1999 from Wyoming oil fields, 15% of the 62 discharges to Wyoming wetlands reviewed
26    showed visible oil sheens in the receiving water and 10 of the sites sampled exceeded discharge
27    limits of 10 mg/L of oil and grease (Ramirez, 2002].

      8.7.  Synthesis
2 8    Hydraulic fracturing operations produce fluids during the flowback and production phases
29    (collectively called wastewater] of a production well, along with liquid and solid treatment
3 0    residuals from treatment processes. A variety of management strategies may be considered, with
31    cost frequently a driving factor. Available information suggests that Class IID wells regulated under
32    the Underground Injection Control (UIC] Program are the most frequently used wastewater
3 3    management practice, but reuse, discharge after treatment, and various other uses are also
34    employed.
                   This document is a draft for review purposes only and does not constitute Agency policy.
      June 2015                                   8-67                  DRAFT—DO NOT CITE OR QUOTE

-------
      Hydraulic Fracturing Drinking Water Assessment               Chapter 8 - Wastewater Treatment and Waste Disposal


      8.7.1.  Summary of Findings
 1    Hundreds of billions of gallons of waste water are generated annually in the United States by the oil
 2    and gas industry, although national level estimates are difficult to reliably obtain. It is also difficult
 3    to produce a nationwide estimate of the amount of wastewater that is attributable specifically to
 4    hydraulic fracturing because some states do not specifically identify wastewater from hydraulic
 5    fracturing operations in their available wastewater data.

 6    The total amount of wastewater produced in an area corresponds generally to oil and gas
 7    production and, therefore, may increase  if hydrocarbon production increases in a region.
 8    Geographically, a large portion of oil and gas wastewater in the United States is reported to be
 9    generated in the western part of the country, including contributions from both conventional and
10    unconventional resources. For some states, estimates of hydraulic fracturing wastewater volumes
11    can be made using publicly available production or waste data. Annual estimates compiled in this
12    way range from hundreds of millions to billions of gallons of wastewater generated per state per
13    year. Direct comparisons among these state data are problematic, however, because of a great deal
14    of variability in state data collection, including differences in the years for which data are available,
15    and challenges in definitively identifying wells that have been hydraulically fractured (to
16    distinguish hydraulic fracturing wastewater from that generated from wells that are not
17    hydraulically fractured). Within a given state, however, estimated volumes in areas where
18    hydraulic fracturing is practiced extensively have generally increased over the last several years,
19    along with numbers of wells contributing to total wastewater volumes. For example, the data made
2 0    available by PA DEP illustrate that the total volume of wastewater generated correlates generally
21    with a significant increase in volume of hydrocarbon production and with the number of
22    production wells. As hydraulic fracturing activities increase and the number of wells increases,  the
2 3    amount of hydraulic fracturing wastewater generated is likely to increase.

      8.7.1.1. Wastewater Management Practices
24    Hydraulic fracturing wastewater is managed in a variety of ways, including disposal through Class
25    IID wells; minimal treatment and reuse in subsequent fracturing operations; more complete
26    treatment followed by discharge, disposal, or reuse; evaporation; and other uses such as irrigation
2 7    (when the wastewater quality is adequate). Unauthorized discharges of hydraulic fracturing
2 8    waste waters have been documented; such discharges could potentially impact drinking water
29    resources, but estimates of the frequency of occurrence cannot be developed with the available
30    data.

31    As of 2015, available information suggests that wastewater management practices involve
32    extensive use  of Class II wells to manage  wastewater from most of the major unconventional plays
33    in the United States, with the notable exception of the Marcellus Shale region in Pennsylvania. More
34    than 98% of wastewater in the oil and gas industry is estimated to be injected into Class II wells
35    annually (including wells for enhanced oil recovery and disposal) (Clark and Veil. 2009). Based on
36    data compiled from 2012 and 2014, there are about 25,000 Class IID wells in the United States (U.S.
37    EPA. 2015q). In particular, large numbers of active injection wells are found in Texas  (7,876 or
                   This document is a draft for review purposes only and does not constitute Agency policy.
      June 2015                                    8-68                  DRAFT—DO NOT CITE OR QUOTE

-------
      Hydraulic Fracturing Drinking Water Assessment              Chapter 8 - Wastewater Treatment and Waste Disposal


 1    29%), Kansas (5,516 or 20%), Oklahoma (4,622 or 17%), Louisiana (2,448 or 9%), and Illinois
 2    (1,054 or 4%).

 3    Use of Class IID wells is likely driven by the availability of Class IID wells within reasonable
 4    transportation distance and the cost of transporting (and injecting) the wastewaters. In the oil and
 5    gas industry, Class IID wells have generally been the most economically favorable wastewater
 6    management practice (U.S. GAP, 2012). In Pennsylvania, there are only nine Class IID wells as of
 7    February 2015, and a significant growth of gas production using hydraulic fracturing in the
 8    Marcellus is generating increasing amounts of wastewater. Treatment and reuse are becoming
 9    increasingly popular in the Marcellus Shale region and are in more widespread use in comparison
10    to other oil and gas producing parts of the country.

11    Reuse of hydraulic fracturing wastewater to formulate fracturing fluid in subsequent hydraulic
12    fracturing jobs varies considerably on a national level, and reliable estimates are not available for
13    all areas. As of 2014-2015, the greatest amount of reuse occurs in Pennsylvania, where the scarcity
14    of Class IID wells to receive Marcellus wastewater drives this practice. Recent estimates of
15    wastewater reuse in Pennsylvania range as high as 90% or more. Waste disposal data from the PA
16    PEP (2015a) indicate that much of the reuse happens on-site. Operators also report some reuse of
17    wastewater in other regions such as the Haynesville Shale, the Fayetteville Shale, the Barnett Shale,
18    and the Eagle Ford Shale, although at much lower volume percentages (about 5 - 20%) compared
19    with practices in the Marcellus Shale region. Increased reuse and recycling of hydraulic fracturing
20    wastewaters has the added benefit of providing an additional water supply for hydraulic fracturing
21    fluid formulation in areas where water scarcity is a concern. If, however, hydraulic fracturing
22    activity slows, demand for wastewater for reuse will also decrease, and other forms of wastewater
23    management will be needed.

24    The decision to reuse/recycle depends upon several factors, including the volume  and rate of
25    production of the wastewater and whether these are suitable for water needs for ongoing
26    fracturing activities in the area. The composition of the water, in particular the TSS and TDS
2 7    content, and whether the water quality can be accommodated in the fracturing practices in an area
2 8    can also influence reuse, including decisions about what type of pretreatment or treatment may be
29    needed to make reuse or recycling feasible.

30    Treatment facilities (either  centralized waste treatment facilities (CWTs) or systems designed for
31    on-site use) can be permitted to treat oil and gas wastewaters. Treatment can be followed by
32    discharge to a surface water body or to a POTW, or the treated effluent may be used for reuse. Most
33    CWTs treating hydraulic fracturing wastewater are located in Pennsylvania (39 facilities), and a
34    number of CWTs (11) are located in Ohio. More are under construction or pending approval. Most
35    are "zero-discharge" and do not have the treatment capacity to reduce TDS; their effluent is reused
36    for hydraulic fracturing. Specialized on-site, mobile, or semi-mobile treatment facilities can be used
37    by operators to handle wastewater without the expense of long-distance transportation and can be
3 8    customized to produce an effluent that meets the water quality needs of the intended disposal or
39    reuse plans.
                   This document is a draft for review purposes only and does not constitute Agency policy.
      June 2015                                    8-69                  DRAFT—DO NOT CITE OR QUOTE

-------
      Hydraulic Fracturing Drinking Water Assessment              Chapter 8 - Wastewater Treatment and Waste Disposal


 1    Treatment of hydraulic fracturing wastewaters by publicly owned treatment works (POTWs) was
 2    previously practiced in Pennsylvania. POTWs are not designed for the high TDS content of
 3    Marcellus wastewaters, and stricter discharge limits for TDS in Pennsylvania, as well as a positive
 4    response to a request from Pennsylvania DEP that operators stop sending Marcellus wastewater to
 5    POTWs and some CWTs, led to the practice being discontinued in 2011. (Some POTWs in
 6    Pennsylvania still accept oil and gas wastewaters from conventional operations, including
 7    conventional wells that have undergone hydraulic fracturing.)

 8    Management plans will necessarily need to change with time as hydraulic fracturing activities in a
 9    region change. The volumes of wastewater also change during the life of a well. The chemical
10    composition of the wastewater changes during the transition from the flowback period and into the
11    production phase. In addition, the demand for reused water to support ongoing fracturing activities
12    will change. Taken in aggregate, these factors may influence costs and choices associated with
13    hydraulic fracturing wastewater management, especially if Class IID wells are limited in a particular
14    area for any reason.

      8.7.1.2. Treatment and Discharge
15    One of the most frequently cited concerns  regarding hydraulic fracturing wastewater, especially
16    from shale plays and tight sand plays, is the high TDS content, which poses challenges for
17    treatment, discharge, and reuse. Conventional treatment processes such as sedimentation, filtration
18    methods, flotation, chemical precipitation  and ion exchange can remove constituents such as oil and
19    grease, major cations, metals, and TSS. Because these processes do not remove monovalent ions
20    (e.g., chloride, bromide, sodium), reducing TDS in these high-salinity wastewaters requires more
21    advanced processes such as reverse osmosis (RO), electrodialysis, and distillation methods.
22    Distillation methods appear to be the approach of choice for newer CWT facilities that are designed
23    to lower TDS. RO, while highly effective, does have limits to TDS concentrations (less than
24    approximately 40,000 mg/L) that it can treat (Shaffer etal.. 2013: Younos and Tulou. 2005).

25    Hydraulic fracturing wastewater discharged from treatment facilities without advanced TDS
26    removal processes has been shown to cause elevated TDS, bromide, and chloride levels in receiving
27    waters in Pennsylvania. Existing literature indicates that bromide and chloride are important
28    wastewater constituents with regard to potential burdens on downstream drinking water
29    treatment facilities. Bromide in particular  is of concern due to the formation of disinfection by-
30    products (DBFs) during disinfection. Some types of DBFs are regulated under SDWA's Stage 1 and
31    Stage 2 DBF Rules, but a subset of DBFs, including a number of chlorinated, brominated,
32    nitrogenous, and iodinated DBFs, are not regulated. Brominated DBFs (and iodinated DBFs) are
33    more toxic than other species of DBFs. Modeling suggests that very small percentages of hydraulic
34    fracturing wastewater in  a river used as a  source for drinking water treatment plants may cause a
35    notable increase in DBF formation.

36    Radionuclides (in particular radium-226 and radium-228) in some hydraulic fracturing
37    wastewaters pose concerns for the quality of discharges if they are not adequately treated. Possible
38    elevated radionuclide content in treatment residuals is also a consideration. In Marcellus Shale  gas
39    production wastewater, radium-226, radium-228, gross alpha, and gross beta are most cited as the

                   This document is a draft for review purposes only and does not constitute Agency policy.
      June 2015                                   8-70                  DRAFT—DO NOT CITE OR QUOTE

-------
      Hydraulic Fracturing Drinking Water Assessment              Chapter 8 - Wastewater Treatment and Waste Disposal


 1    radioactive constituents of concern, and concentrations can range up to thousands of pCi/L. Fewer
 2    data exist on uranium content in wastewaters, and data are also limited on radionuclide
 3    concentrations in wastewaters from other unconventional plays. A confounding issue in evaluating
 4    radium concentrations is underestimation when using traditional wet chemical methods with high-
 5    IDS waters. A variety of treatment processes can be used for removal of radium, ranging from
 6    conventional methods such as chemical precipitation and filtration to more advanced and costly
 7    techniques, such as reverse osmosis or distillation (including mechanical vapor recompression).
 8    Whether the effluent from such treatment contains elevated radium, however, will depend upon
 9    influent concentrations as well as treatment removal efficiency.

10    Other potential effects on drinking water resources may result from discharges or spills of
11    hydraulic fracturing wastewaters containing elevated concentrations of barium and other metals.
12    Again, the management strategy and treatment choices will affect the likelihood of such impacts.

      8.7.2.  Factors Affecting the Frequency or Severity of Impacts
13    On a regional scale, potential effects on drinking water resources from hydraulic fracturing
14    waste water will depend upon the mix of wastewater management strategies used, and potential
15    impacts may change through time if the quantity of hydraulic fracturing wastewater changes and
16    strategies to manage the wastewater change. For example, if use of Class IID wells becomes
17    restricted in parts of the country where they are currently commonly used, the emphasis may shift,
18    at least locally, from use of Class IID wells and towards the use of treatment and either  discharge  or
19    reuse. Although reuse delays the discharge of wastewater by directing it to ongoing fracturing
20    activities, reuse may ultimately concentrate constituents such as radionuelides (depending upon
21    the ratio of recycled to new water). If a stream of wastewater or portion of wastewater has been
22    used for more than one hydraulic fracturing event and is eventually intended for disposal, the
23    method of disposal will need to be appropriate for the quality of the wastewater.

24    Potential effects on drinking water resources from hydraulic fracturing wastewaters that undergo
25    treatment depend upon the quality and quantity of discharges to receiving waters (discharge could
26    occur directly after treatment at a CWT or indirectly after discharge to a POTW). Hydraulic
27    fracturing wastewater management can consider appropriate levels of treatment and blending so
28    that the resulting IDS content in a receiving water will not result in formation of DBFs  during
29    subsequent drinking water treatment and will not impair biological treatment processes.

30    The volumes of discharges relative to the receiving water body size are important local factors to
31    consider in evaluating whether elevated concentrations can be anticipated at downstream drinking
32    water intakes. Small drinking water systems drawing water from smaller streams in affected areas
33    would likely face greater challenges in dealing with high bromide and chloride levels in source
34    waters. Furthermore, other potential impacts on surface water and shallow ground water may exist
35    due to spills of either untreated wastewater  or effluent from zero-discharge CWTs, and there will be
36    site-specific factors such as distance to a water body or depth to the water table to consider (see
37    Chapter 5).
                   This document is a draft for review purposes only and does not constitute Agency policy.
      June 2015                                    8-71                  DRAFT—DO NOT CITE OR QUOTE

-------
      Hydraulic Fracturing Drinking Water Assessment              Chapter 8 - Wastewater Treatment and Waste Disposal


 1    Results from existing literature and recent PA DEP data suggest that cumulative impacts from
 2    radionuclides may occur in sediments at or near discharge points from facilities that treat and
 3    discharge oil and gas wastewater (or have done so in the past). There may be consequences for
 4    downstream drinking water systems if the sediments are disturbed or entrained due to dredging or
 5    flood events. Similarly, some organic constituents may not be removed during treatment, and
 6    potential effects on receiving waters and sediments will depend upon the properties of the specific
 7    constituents, their concentrations, and the treatment used.

 8    The possibility of radionuclides affecting receiving waters and sediments will depend upon the
 9    technologically enhanced naturally occurring radioactive material (TENORM) content of the
10    wastewater and the treatment processes used. Although radionuclide contamination at drinking
11    water intakes due to treated hydraulic fracturing fluid has not been detected, a recent PA DEP study
12    [PADEP. 2015b) has revealed radium in effluents from both CWTs handling oil and gas wastewater
13    and POTWs receiving effluent from such facilities. The concentrations in the CWT effluents were
14    considerably higher than in the POTW effluents. The site selection criteria for this study included
15    some Pennsylvania wastewater facilities whose influents include wastewater from unconventional
16    operations or where radioactivity was measured in the influent, sludges, or effluents (CWTs may
17    also receive conventional wastewater). In regions where unconventional plays are known to be
18    enriched in radionuclides, analysis of TENORMs in untreated hydraulic fracturing waste waters,
19    selection of appropriate treatment processes, and monitoring of TENORMs in treatment effluent
20    and receiving waters could help address potential impacts on drinking water resources. Gross  alpha
21    and gross beta measurements or gamma spectroscopic analyses could be used as initial screening
22    methods for radionuclides. Enrichment of TENORMs in waste products from treatment processes
23    also requires appropriate management to reduce potential impacts.

24    Other management strategies such as irrigation, road spreading, and evaporation are less
25    frequently employed for hydraulic fracturing wastewaters. Irrigation or land application may have
26    potential effects on surface waters depending upon the constituents in the wastewater (e.g., salts
2 7    and radionuclides), the distance from the site of application to a receiving water, and whether
2 8    storm water management measures exist that mitigate runoff. Distance to the water table,
29    precipitation, and the hydrogeologic properties of the soil and sediment will influence whether
3 0    migration of these constituents results in contamination of shallow ground water.

      8.7.3.  Uncertainties
31    A full understanding of the practices being used for management of hydraulic fracturing
3 2    wastewaters is limited by a lack of available data in a number of areas. It is difficult to assemble a
33    complete, national- or regional-level picture of wastewater generation and management practices
34    because the tracking and availability of data vary from state to state. Although some states provide
35    well-organized and relatively thorough data, not all states make such information available, and it
3 6    can be difficult to identify wastewater volumes specifically associated with hydraulic fracturing
37    activity (as compared to all oil and gas production activities). Such data would be needed to place
3 8    hydraulic fracturing wastewater in the broader context of all oil and gas wastewaters. Data are also
                   This document is a draft for review purposes only and does not constitute Agency policy.
      June 2015                                    8-72                  DRAFT—DO NOT CITE OR QUOTE

-------
      Hydraulic Fracturing Drinking Water Assessment              Chapter 8 - Wastewater Treatment and Waste Disposal


 1    generally difficult to locate for production volumes, chemical composition, masses, and
 2    management and disposal strategies for residuals.

 3    Among management practices, up-to-date information on the volumes of hydraulic fracturing
 4    wastewaters disposed of via underground injection in different states are not uniformly available.
 5    Without this information, it is difficult to assess whether disposal well capacity will become an
 6    issue in areas where hydraulic fracturing activity is expected to increase.

 7    Assessment of the potential effects of hydraulic fracturing on drinking water resources is also
 8    limited by relatively few data on effluent quality from CWTs that receive oil and gas wastewaters
 9    (including those associated with hydraulic fracturing) and POTWs that receive CWT effluents. If a
10    CWT can discharge to surface water (e.g., the CWT has a NPDES permit), some monitoring data may
11    be available that will provide information on effluent quality, but the list of monitored constituents
12    may be limited.

13    In evaluating the treatment effectiveness of full scale facilities, relatively few data exist on the
14    quality of both influents and effluents from treatment facilities,  although some manufacturers of
15    patented CWT systems publicize information on treatment effectiveness. A better understanding of
16    the pollutant removal capabilities of facilities would be helped by influent and effluent sampling,
17    timed so that effluent samples are representative of influent samples to the degree possible. There
18    are limited analyses of influent and effluent samples for a wide range of constituents associated
19    with hydraulic fracturing fluids and wastewaters (e.g., major cations and anions, radionuclides,
20    metals, VOCs, SVOCs, diesel range organics (DROs), and total petroleum hydrocarbons (TPH)).
21    Analyses are needed in which the methods are appropriate for the TDS content of the sample.
22    Radium in particular needs to be analyzed using a method suitable for high-salt samples, otherwise
23    concentrations may be underestimated. Continued work towards ensuring that analytical methods
24    exist for the highly complex matrixes often encountered with oil and gas wastewater would provide
25    better certainty in the results of chemical analyses.

26    Monitoring of surface waters, even screening with a simple TDS proxy such as conductivity, would
27    be needed to help assess how often hydraulic fracturing activities (including spills or discharges of
28    wastewater) affect receiving waters; such data are lacking except for some studies in the Marcellus
29    Shale region. Existing data  are also limited regarding legacy effects, such as accumulation of
30    contaminants in sediments at discharge points, soil accumulation due to application of de-icing
31    brines or salts from wastewater treatment, and handling of waste water treatment residuals.

      8.7.4.  Conclusions
32    Oil and gas operations in the United States generate billions of gallons of wastewater daily; this
3 3    includes wastewater associated with hydraulic fracturing activities,  although what portion of this
34    oil and gas wastewater is attributable to hydraulic fracturing operations is difficult to estimate due
35    to lack of consistent data regarding wastewater volumes. Available information indicates that the
36    majority of this water is injected into Class IID wells regulated under the Underground Injection
37    Control (UIC) program,  although in some areas of the country, wastewater is reused (either with
3 8    our without treatment)  for new hydraulic fracturing jobs. In the Marcellus Shale region in


                   This document is a draft for review purposes only and does not constitute Agency policy.
      June 2015                                    8-73                  DRAFT—DO NOT CITE OR QUOTE

-------
      Hydraulic Fracturing Drinking Water Assessment               Chapter 8 - Wastewater Treatment and Waste Disposal


 1    Pennsylvania, the majority of wastewater is currently reused. Wastewater may also be treated in a
 2    CWT and discharged to a surface water body or to a POTW, or in certain settings, used for various
 3    other uses (e.g., irrigation) if water quality allows. Impacts on drinking water resources may result
 4    from inadequate  treatment prior to discharge or spills. Particular constituents of concern in
 5    wastewater from hydraulic fracturing, especially in the Marcellus Shale region, include bromide and
 6    radionuclides. There is limited information regarding the influents and effluents from facilities that
 7    treat wastewater from hydraulic fracturing operations.


 8    Text Box 8-2. Research Questions Revisited.

 9    What are the common treatment and disposal methods for hydraulic fracturing wastewater, and
10    where are these methods practiced?
11    •   The majority of hydraulic fracturing wastewater in the United States is disposed of via underground
12        injection wells. As of 2014-2015, most states where hydraulic fracturing occurs have an adequate
13        number of Class IID injection wells regulated under the Underground Injection Control (UIC) program.
14        The Marcellus  Shale region, especially the northeastern region, is an exception. Wastewater treatment for
15        reuse is increasing in the Marcellus shale region and may continue to increase in western shale plays as
16        the practice becomes encouraged and economically favorable.
17    How effective are conventional POTWs and commercial treatment systems in removing organic and
18    inorganic contaminants of concern in hydraulic fracturing wastewater?
19    •   Publicly owned treatment works (POTWs) using basic treatment processes cannot effectively reduce
20        elevated total dissolved solids (TDS) concentrations in hydraulic fracturing wastewater. Centralized
21        waste treatment facilities (CWTs) with advanced treatment processes can remove TDS constituents with
22        removal efficiencies ranging from 97% to over 99% as demonstrated at facilities that use treatment
23        processes such as mechanical vapor recompression, distillation, and reverse osmosis (see Table 8-6).
24        Advanced treatment processes have been shown to remove certain contaminants found in hydraulic
25        fracturing wastewater (see Table 8-6). Indirect discharge, where wastewater is pretreated by a CWT and
26        sent to a POTW, may be an effective option for hydraulic fracturing wastewater treatment (with
27        restrictions on contaminant concentrations in the pretreated wastewater that is sent to a POTW). This
28        option would require careful planning to ensure that the pretreated wastewater blended with POTW
29        influent is of appropriate quality and quantity to prevent deleterious effects on biological processes in the
30        POTW or the pass-through of contaminants.
31    •   Facilities that treat wastewater for reuse and employ only basic treatment are unable to remove all
32        contaminants in hydraulic fracturing wastewater. Depending on the water quality requirements for a
33        particular site, these lower quality treated waters may be of adequate quality for reuse on subsequent
34        hydraulic fracturing operations (and will be less costly). Some organic compounds (BTEX, some alcohols,
35        2-butoxyethanol) may not be removed by the processes employed in CWTs if they don't include specific
36        processes that target these compounds (e.g., distillation, advanced oxidation, adsorption).
                    This document is a draft for review purposes only and does not constitute Agency policy.
      June 2015                                      8-74                   DRAFT—DO NOT CITE OR QUOTE

-------
      Hydraulic Fracturing Drinking Water Assessment                Chapter 8 - Wastewater Treatment and Waste Disposal
 1    What are the potential impacts on drinking water treatment facilities from surface water disposal of
 2    treated hydraulic fracturing wastewater?

 3    •   Inadequate bromide and iodide removal from treated hydraulic fracturing wastewater has the greatest
 4        potential to affect surface water quality and place a burden on downstream drinking water treatment
 5        facilities that use chlorine-based disinfection due to the formation of DBFs. Radionuclides, metals, and
 6        trace organic compounds in effluents from CWTs may also be of concern if present in treated wastewater
 7        or if they accumulate in sediments downstream of discharge points. These constituents have reached
 8        drinking water resources via some discharges, although sampling data for effluents and receiving waters
 9        are limited. As of 2014-2015, there is no evidence that these contaminants have affected drinking water
10        facilities, but data are lacking for concentrations of these constituents at drinking water intakes in
11        regions with hydraulic fracturing.


      8.8.  References for Chapter 8

      Alanco. (2012). New subsidiary Alanco Energy Services, Inc. to provide produced water disposal services to
         natural gas industry. Alanco. http://www.alanco.com/news 040912.asp

      Alzahrani. S: Mohammad. AW: Hilal. N: Abdullah. P: laafar. 0. (2013). Comparative study of NF and RO
         membranes in the treatment of produced water-Part I: Assessing water quality. Desalination 315:18-26.
         http://dx.doi.0rg/10.1016/j.desal.2012.12.004

      API (American Petroleum Institute). (2000). Overview of exploration and production waste volumes and
         waste management practices in the United States, http://www.api.org/environment-health-and-
         safety/environmental-performance/~/media/Files/EHS/Environmental  Performance/ICF-Waste-
         Survey-of-EandP-Wastes-2000.ashx

      Argonne National Laboratory. (2014). Water use and management in the Bakken shale oil play. (DOE Award
         No.: FWP 49462). Pittsburgh, PA: National Energy Technology Laboratory.
         http://www.ipd.anl.gov/anlpubs/2014/05/104645.pdf

      Bacher, D. (2013). Oil company fined $60,000 for illegally discharging fracking fluid. Available online at
         https://www.indybay.org/newsitems/2013/ll/17/18746493.php7show comments=l (accessed March
         6,2015).

      Bair. ES: Digel. RK. (1990). Subsurface transport of inorganic and organic solutes from experimental road
         spreading of oilfield brine. Ground Water Monitoring and Remediation 10: 94-105.

      Banasiak. LI: Schafer. AI. (2009). Removal of boron, fluoride and nitrate by electrodialysis in the presence of
         organic matter. J Memb Sci 334:101-109. http://dx.doi.Org/10.1016/j.memsci.2009.02.020

      Barbot, E: Vidic, NS: Gregory, KB: Vidic, RD. (2013). Spatial and temporal correlation of water quality
         parameters of produced waters from Devonian-age shale following hydraulic fracturing. Environ Sci
         Technol 47: 2562-2569.

      Benko. KL: Drewes. IE.  (2008). Produced water in the Western United States:  Geographical distribution,
         occurrence, and composition. Environ Eng Sci 25: 239-246.

      Blauch. ME: Myers. RR: Moore. TR: Lipinski. BA. (2009). Marcellus shale post-frac flowback waters - where is
         all the salt coming from and what are the implications? In Proceedings of the SPE Eastern Regional
         Meeting. Richardson, TX: Society of Petroleum Engineers.

      Boschee. P. (2012). Handling produced water from hydraulic fracturing. Oil and Gas Facilities 1: 23-26.

      Boschee. P. (2014). Produced and flowback water recycling and reuse: Economics, limitations, and
         technology. Oil and Gas Facilities 3:16-22.
                     This document is a draft for review purposes only and does not constitute Agency policy.

      June 2015                                       8-75                    DRAFT—DO NOT CITE OR QUOTE

-------
Hydraulic Fracturing Drinking Water Assessment                Chapter 8 - Wastewater Treatment and Waste Disposal
Bowen. ZH: Oelsner. GP: Cade. BS: Gallegos. TI: Farag. AM: Mott. DN: Potter. Cl: Cinotto. PI: Clark. ML: Kappel.
   WM: Kresse, TM: Melcher, CP: Paschke, SS: Susong, DP: Varela, BA. (2015). Assessment of surface water
   chloride and conductivity trends in areas of unconventional oil and gas development-Why existing
   national data sets cannot tell us what we would like to know. Water Resour Res 51: 704-715.
   http://dx.doi.org/10.1002/2014WR016382

Boysen. DB: Boysen. IA: Boysen. IE.  (2002). Creative Strategies for Produced Water Disposal in the Rocky
   Mountain Region. Paper presented at 9th Annual International Petroleum Environmental Conference,
   October 2002, Albuquerque, NM.

Boysen. IE: Harm. I A: Shaw. B: Fosdick. M: Grisanti. A: Sorensen. IA. (1999). The current status of commercial
   deployment of the freeze thaw evaporation treatment of produced water. Richardson, TX: Society of
   Petroleum Engineers. http://dx.doi.org/10.2118/52700-MS

Bruff. M: likich. SA. (2011). Field demonstration of an integrated water treatment technology solution in
   Marcellus shale. Paper presented at SPE Eastern Regional Meeting, August 17-19,2011, Columbus, OH.

California Department of Conservation. (2015). Monthly production and injection databases, statewide
   production and injection data [Database]. Sacramento, CA: California Department of Conservation,
   Division of Oil, Gas & Geothermal Resources. Retrieved from
   http://www.conservation.ca.gov/dog/prod injection db/Pages/Index.aspx

Camacho. LM. ar: Dumee. L: Zhang. I: Li. I: Duke. M: Gomez. I: Gray. S. (2013). Advances in membrane
   distillation for water desalination and purification applications. Water 5: 94-196.
   http://dx.doi.org/10.3390/w5010094

Chapman.  EC: Capo. RC: Stewart. BW: Kirby. CS: Hammack. RW: Schroeder. KT: Edenborn. HM. (2012).
   Geochemical and strontium isotope characterization of produced waters from Marcellus  Shale natural gas
   extraction. Environ Sci Technol 46: 3545-3553.

Chester Engineers. (2012). The Pittsburgh water and sewer authority 40 year plan. (PWSA Project No. R-
   Dl.10015-11). Pittsburgh, PA: The Pittsburgh Water and Sewer Authority.
   http://apps.pittsburghpa.gov/pwsa/PWSA 40-year Plan.pdf

Chiado. ED. (2014). The impact of shale gas/oil waste on  MSW landfill composition and operations. In CL
   Meehan; JM VanBriesen; F Vahedifard; X Yu; C Quiroga (Eds.), Shale energy engineering 2014 technical
   challenges, environmental issues, and public policy (pp. 412-420). Reston, VA: American Society of Civil
   Engineers, http://dx.doi.org/10.1061/9780784413654.044

Clark. CE: Horner. RM: Harto. CB. (2013). Life Cycle Water Consumption for Shale Gas and Conventional
   Natural Gas. Environ Sci Technol 47:11829-11836. http://dx.doi.org/10.1021/es4013855

Clark, CE: Veil, IA. (2009). Produced water volumes and management practices in the United States (pp. 64).
   (ANL/EVS/R-09/1). Argonne, IL: Argonne National Laboratory.
   http://www.circleofblue.org/waternews/wp-
   content/uploads/2010/09/ANL EVS R09 produced water volume report 2437.pdf

COGCC (Colorado Oil and Gas Conservation Commission). (2015). COGIS - all production reports to  date.
   Denver, CO. Retrieved from http://cogcc.state.co.us/

Colorado Division of Water Resources: Colorado Water Conservation Board: Colorado Oil and Gas
   Conservation Commission. (2014). Water sources and demand for the hydraulic fracturing of oil and gas
   wells in Colorado from 2010 through 2015 [Fact Sheet], http://cewc.colostate.edu/2012/02/water-
   sources-and-demand-for-the-hydraulic-fracturing-of-oil-and-gas-wells-in-colorado-from-2010-through-
   2015/

Countess. S: Boardman. G: Hammack. R:  Hakala. A: Sharma. S: Parks. I. (2014). Evaluating leachability of
   residual solids from hydraulic fracturing in the Marcellus shale. In Shale energy engineering 2014:
   Technical challenges, environmental issues, and public policy. Reston, VA: American Society of Civil
   Engineers, http://dx.doi.org/10.1061/9780784413654.012
               This document is a draft for review purposes only and does not constitute Agency policy.

June 2015                                       8-76                    DRAFT—DO NOT CITE OR QUOTE

-------
Hydraulic Fracturing Drinking Water Assessment                Chapter 8 - Wastewater Treatment and Waste Disposal
Cusick. M. (2013). EPA fines western PA treatment plants for Marcellus wastewater violations. Available
   online at http://stateimpact.npr.org/pennsylvania/2013/05/24/epa-fines-western-pa-treatment-plants-
   for-marcellus-wastewater-violations/ (accessed March 6, 2015).

Dahm. K: Chapman. M. (2014). Produced water treatment primer: Case studies of treatment applications.
   (S&T Research Project #1617). Denver CO: U.S. Department of the Interior.
   http://www.usbr.gov/research/projects/download product.cfm?id=1214.

DOE (U.S. Department of Energy). (2002). Handbook on best management practices and mitigation strategies
   for coalbed methane in the Montana portion of the Powder River Basin. Tulsa, OK: U.S. Department of
   Energy, National Energy Technology Laboratory. http://bogc.dnrc.mt.gov/PDF/BMPHandbookFinal.pdf

DOE (U.S. Department of Energy). (2003). Handbook on coalbed methane produced water: Management and
   beneficial use alternatives. Tulsa, OK: ALL Consulting, http://www.all-
   llc.com/publicdownloads/CBM BU Screen.pdf

DOE (U.S. Department of Energy). (2004). A white paper describing produced water from production of crude
   oil, natural gas, and coal bed methane. Lemont, IL: Argonne National Laboratory.
   http://seca.doe.gov/technologies/oil-gas/publications/oil pubs/prodwaterpaper.pdf

DOE (U.S. Department of Energy). (2006). A guide to practical management of produced water from onshore
   oil and gas operations in the United States. Washington, DC: U.S. Department of Energy, National
   Petroleum Technology Office.
   http://fracfocus.Org/sites/default/files/publications/a guide to practical management of produced wat
   er from onshore oil  and gas operations in the united states.pdf

Drewes. I: Cath. T: Debroux. I: Veil. I. (2009). An integrated framework for treatment and management of
   produced water - Technical assessment of produced water treatment technologies (1st ed.). (RPSEA
   Project 07122-12). Golden, CO: Colorado School of Mines.
   http://aqwatec.mines.edu/research/projects/Tech Assessment PW Treatment Tech.pdf

Duraisamy. RT: Beni. AH: Henni. A. (2013). State of the art treatment of produced water. In W Elshorbagy; RK
   Chowdhury (Eds.), Water treatment (pp. 199-222). Rijeka, Croatia: InTech.
   http://dx.doi.org/10.5772/53478

Easton, I. (2014). Optimizing fracking wastewater management. Pollution Engineering January 13.

EIA (Energy Information Administration). (2014c). Natural gas. U.S. Energy Information Administration:
   Independent statistics and analysis. Available online at http://www.eia.gov/naturalgas/

Eiceman. GA. (1986). Hazardous organic wastes from natural gas production, processing and distribution:
   Environmental fates. (WRRI report, no. 227). New Mexico: Water  Resources Research Institute.
   http://wrri.nmsu.edu/publish/techrpt/abstracts/abs227.html

Engle. MA: Bern. CR: Healy. RW: Sams. II: Zupancic. IW: Schroeder. KT. (2011). Tracking solutes and water
   from subsurface drip irrigation application of coalbed methaneproduced waters, Powder River Basin,
   Wyoming. Environmental Geosciences 18:169-187.

Entrekin. S: Evans-White. M: Johnson. B: Hagenbuch. E. (2011). Rapid expansion of natural gas development
   poses a threat to surface waters. Front Ecol Environ 9: 503-511. http://dx.doi.org/10.1890/110053

EPA (Environmental Protection Agency). (2000). Development document for effluent limitations guidelines
   and standards for the centralized waste treatment industry. (821R00020). Washington, DC: U.S.
   Environmental Protection Agency.

Ertel, D: McManus, K: Bogdan, I. (2013). Marcellus wastewater treatment: Case study. In Summary of the
   technical workshop on wastewater treatment and related modeling (pp. A56-A66). Williamsport, PA:
   Eureka Resources, LLC. http://www2.epa.gov/hfstudy/summary-technical-workshop-wastewater-
   treatment-and-related-modeling
               This document is a draft for review purposes only and does not constitute Agency policy.

June 2015                                      8-77                   DRAFT—DO NOT CITE OR QUOTE

-------
Hydraulic Fracturing Drinking Water Assessment                Chapter 8 - Wastewater Treatment and Waste Disposal
Fakhru'1-Razi. A: Pendashteh. A: Abdullah. LC: Biak. PR: Madaeni. SS: Abidin. ZZ. (2009). Review of
   technologies for oil and gas produced water treatment [Review]. J Hazard Mater 170: 530-551.

Ferrar. Kl: Michanowicz. PR: Christen. CL: Mulcahy. N: Malone. SL: Sharma. RK. (2013). Assessment of effluent
   contaminants from three facilities discharging Marcellus Shale wastewater to surface waters in
   Pennsylvania. Environ Sci Technol 47: 3472-3481.

Geiver. L. (2013). Frac water treatment yields positive results for Houston Co. Retrieved from
   http://www.thebakken.com/articles/20/frac-water-treatment-yields-positive-results-for-houston-co

Gomes. I: Cocke. D: Das. K: Guttula. M: Tran. D: Beckman: I. (2009). Treatment of produced water by
   electrocoagulation. Shiner, TX: KASELCO, LLC. http://www.kaselco.com/index.php/library/industry-
   white-papers

Gonneea. ME: Morris. PI: Dulaiova. H: Charette. MA. (2008). New perspectives on radium behavior within a
   subterranean estuary. Mar Chem 109: 250-267. http://dx.doi.Org/10.1016/j.marchem.2007.12.002

Greenhunter (Greenhunter Resources). (2014). Oillfield water management solutions. Available online at
   http://www.greenhunterenergy.com/operations/owms.htm

Gregory. KB: Vidic. RD: Dzombak. DA. (2011). Water management challenges associated with the production
   of shale gas by hydraulic fracturing. Elements 7:181-186.

GTI (Gas Technology Institute). (2012). Barnett and Appalachian shale water management and resuse
   technologies. (Report no. 08122-05.FINAL.l). Sugar Land, TX: Research Partnership to Secure Energy for
   America, RPSEA. https://www.netl.doe.gov/file%201ibrary/research/oil-
   gas/Natural%20Gas/shale%20gas/08122-05-final-report.pdf

Guerra, K: Dahm, K: Dundorf, S. (2011). Oil and gas produced water management and beneficial use in the
   western United States. (Science and Technology Program Report No. 157). Denver, CO: U.S. Department of
   the Interior Bureau of Reclamation.

Halliburton. (2014). Hydraulic fracturing 101. Available online at
   http://www.halliburton.com/public/projects/pubsdata/hydraulic fracturing/fracturing 101.html

Hamieh, BM: Beckman, JR. (2006). Seawater desalination using Dewvaporation technique: theoretical
   development and design evolution. Desalination 195: 1-13.
   http://dx.doi.0rg/10.1016/i.desal.2005.09.034

Hammer. R: VanBriesen. I. (2 012). In frackings wake: New rules are needed to protect our health and
   environment from contaminated wastewater. New York, NY: Natural Resources Defense Council.
   http://www.nrdc.org/energy/files/fracking-wastewater-fullreport.pdf

Hansen. E: Mulvaney. D: Betcher. M.  (2013). Water resource reporting and water footprint from Marcellus
   Shale development in West Virginia and Pennsylvania. Durango, CO: Earthworks Oil & Gas Accountability
   Project, http://www.downstreamstrategies.com/documents/reports publication/marcellus  wv pa.pdf

Hayes. T. (2009). Sampling and analysis of water streams associated with the development of Marcellus shale
   gas. Des Plaines, IL: Marcellus Shale Coalition, http://eidmarcellus.org/wp-
   content/uploads/2012/ll/MSCommission-Report.pdf

Hayes. T: Severin. BF. (2012b). Evaluation of the aqua-pure mechanical vapor recompression system in the
   treatment of shale gas flowback water - Barnett and Appalachian shale water management and reuse
   technologies. (08122-05.11). Hayes, T; Severin, BF. http://barnettshalewater.org/documents/08122-
   05.11-EvaluationofMVR-3-12-2012.pdf

Hayes. TD: Halldorson. B: Horner. P:  Ewing. I: Werline. IR: Severin. BF. (2014). Mechanical vapor
   recompression for the treatment of shale-gas flowback water. Oil and Gas Facilities 3: 54-62.

Hladik. ML: Focazio. Ml: Engle. M. (2014). Discharges of produced waters from oil and gas extraction via
   wastewater treatment plants are sources of disinfection by-products to receiving streams. Sci Total
   Environ 466. http://dx.doi.Org/10.1016/j.scitotenv.2013.08.008


               This document is a draft for review purposes only and does not constitute Agency policy.
June 2015                                       8-78                    DRAFT—DO NOT CITE OR QUOTE

-------
Hydraulic Fracturing Drinking Water Assessment                Chapter 8 - Wastewater Treatment and Waste Disposal
Hulme. D. (2005). CBM co-produced water management, disposal, treatment and use. Hulme, D.
   http://www.uwyo.edu/haub/ruckelshaus-institute/  files/docs/publications/2005-cbm-water-
   management-hulme.pdf

IAEA (International Atomic Energy Agency). (2014). The environmental behaviour of radium: revised edition.
   Vienna, Austria. http://www-pub.iaea.org/MTCD/Publications/PDF/trs476 web.pdf

Igunnu. ET: Chen. GZ. (2014). Produced water treatment technologies. International Journal of Low-Carbon
   Technologies 9:157-177. http://dx.doi.org/10.1093/ijlct/cts049

IHS (Global Insight). (2013). Americas new energy future: The unconventional oil and gas revolution and the
   US economy. Douglas County, Colorado.
   http://www.energyxxi.org/sites/default/files/pdf/Americas New Energy Future Phase3.pdf

Kappel. WM: Williams. IH: Szabo. Z. (2013). Water resources and shale gas/oil production in the Appalachian
   basin critical issues and evolving developments. (Open-File Report 20131137). Troy, NY: U.S. Geological
   Survey, http://pubs.usgs.gov/of/2013/1137/pdf/ofr2013-1137.pdf

Kargbo, DM: Wilhelm, RG: Campbell, PI. (2010). Natural gas plays in the Marcellus Shale: Challenges and
   potential opportunities. Environ Sci Technol 44: 5679-5684. http://dx.doi.org/10.1021/es903811p

Kaushal. SS: Groffman. PM: Likens. GE: Belt. KT: Stack. WP: Kelly. VR: Band. LE: Fisher. GT. (2005). Increased
   salinization of fresh water in the northeastern United States. PNAS 102: 13517-13520.
   http://dx.doi.org/10.1073/pnas.0506414102

Kelly. WR. (2008). Long-term trends in chloride concentrations in shallow aquifers near Chicago. Ground
   Water 46: 772-781. http://dx.doi.Org/10.llll/i.1745-6584.2008.00466.x

Kennedy/lenks Consultants. (2002). Evaluation of technical and economic feasibility of treating oilfield
   produced water to create a new water resource, http://www.gwpc.org/sites/default/files/event-
   sessions/Roger Funston PWC2002 O.pdf

Kharaka. YK: Kakouros. E: Abbott. MM. (2002). Environmental impacts of petroleum production: 1- The fate
   of inorganic and organic chemicals in produced water from the Osage-Skiatook Petroleum Environmental
   Research B  site, Osage County, OK. 9th International Petroleum Environmental Conference, October 22-25,
   2002, Albuquerque, NM.

Krasner. SW. (2009). The formation and control of emerging disinfection by-products of health concern
   [Review]. Philos Transact A Math Phys Eng Sci 367: 4077-4095. http://dx.doi.org/10.1098/rsta.2009.010

Kuwayama. Y: Olmstead. S: Krupnick. A. (2015). Water quality and quantity impacts of hydraulic fracturing.
   Current Sustainable/Renewable Energy Reports 2:17-24. http://dx.doi.org/10.1007/s40518-014-0023-4

LEau LLC. (2008).  Dew vaporation desalination 5,000-gallon-per-day pilot plant. (Desalination and Water
   Purification Research and Development Program Report No. 120). Denver, CO: Bureau of Reclamation, U.S.
   Department of the Interior. http://www.usbr.gov/research/AWT/reportpdfs/reportl20.pdf

LeBas. R: Lord. P: Luna. D: Shahan. T.  (2013). Development and use of high-TDS recycled produced water for
   crosslinked-gel-based hydraulic fracturing. In 2013 SPE hydraulic fracturing technology conference.
   Richardson, TX: Society of Petroleum Engineers. http://dx.doi.org/10.2118/163824-MS

Lefebvre. 0: Moletta. R. (2006). Treatment of organic pollution in industrial saline wastewater: a literature
   review [Review]. Water Res 40: 3671-3682. http://dx.doi.Org/10.1016/j.watres.2006.08.027

Linaric. M: Markic. M: Sipos. L. (2013). High salinity wastewater treatment. Water Sci Technol 68:1400-1405.
   http://dx.doi.org/10.2166/wst.2013.376

Lutz. BD: Lewis. AN: Doyle. MW. (2013). Generation, transport, and disposal of wastewater associated with
   Marcellus Shale gas development. Water  Resour Res 49: 647-656.
               This document is a draft for review purposes only and does not constitute Agency policy.

June 2015                                       8-79                    DRAFT—DO NOT CITE OR QUOTE

-------
Hydraulic Fracturing Drinking Water Assessment                Chapter 8 - Wastewater Treatment and Waste Disposal
Ma. G: Geza. M: Xu. P. (2014). Review of flowback and produced water management, treatment, and beneficial
   use for major shale gas development basins. Shale Energy Engineering Conference 2014, Pittsburgh,
   Pennsylvania, United States.

Malone. S: Kelso. M: Auch. T: Edelstein. K: Ferrar. K: lalbert. K. (2015). Data inconsistencies from states with
   unconventional oil and gas activity. J Environ Sci Health A Tox Hazard Subst Environ Eng 50: 501-510.
   http://dx.doi.org/10.1080/10934529.2015.992678

Maloney, KO: Yoxtheimer, DA. (2012). Production and disposal of waste materials from gas and oil extraction
   from the Marcellus shale play in Pennsylvania. Environmental Practice 14: 278-287.
   http://dx.doi.org/10.10170S146604661200035X

Mantell. ME. (2013a). Recycling and reuse of produced water to reduce freshwater use in hydraulic fracturing
   operations. Presentation presented at Water acquisition modeling: Assessing impacts through modeling
   and other means Technical workshop on wastewater treatment and related modeling, June 4, 2013,
   Arlington, VA.

Matthews. 1C:  Li. S: Swann. CT: Ericksen. RL. (2006). Incubation with moist top soils enhances solubilization of
   radium and other components from oilfield scale and sludge: Environmental concerns from Mississippi.
   Environmental Geosciences 13: 43-53.

Minnich. K. (2011). A water chemistry perspective on flowback reuse with several case studies. Minnich, K.
   http://www2.epa.gov/sites/production/files/documents/10 Minnich - Chemistry 508.pdf

Morillon. A: Vidalie. IF: Syahnudi. U: Suripno. S: Hadinoto. EK. (2002). Drilling and waste management; SPE
   73931. Presentation presented at The SPE International Conference on Health, Safety, and Environment in
   Oil and Gas Exploration and Production, March 20-22, 2002, Kuala Lumpur, Malaysia.

Murray, KE. (2013). State-scale perspective on water use and production associated with oil and gas
   operations, Oklahoma, U.S. Environ Sci Technol 47:  4918-4925. http://dx.doi.org/10.1021/es4000593

Nelson. AW: May. D: Knight. AW: Eitrheim. ES: Mehrhoff. M: Shannon. R: Litman. R: Schultz. MK. (2014).
   Matrix complications in the determination of radium levels in hydraulic fracturing flowback water from
   Marcellus Shale. 1: 204-208. http://dx.doi.org/10.1021/ez5000379

NETL (National Energy Technology Laboratory). (2014). Evaporation [Fact Sheet]. Pittsburgh, PA: US
   Department of Energy, http://www.netl.doe.gov/research/coal/crosscutting/pwmis/tech-desc/evap

Nicot, IP: Reedy, RC: Costley, RA: Huang, Y. (2012). Oil & gas water use in Texas: Update to the 2011 mining
   water use report. Nicot, JP; Reedy, RC; Costley, RA; Huang, Y.
   http://www.twdb.state.tx.us/publications/reports/contracted reports/doc/0904830939 2012Update M
   iningWaterUse.pdf

Nicot. IP: Scanlon. BR. (2012). Water use for shale-gas production in Texas, U.S. Environ Sci Technol 46: 3580-
   3586. http://dx.doi.org/10.1021/es204602t

Nowak. N: Bradish. I. (2010). High density polyethylene (HOPE) lined produced water evaporation ponds.
   Presentation presented at 17th International Petroleum and Biofuels Environmental Conference, August
   31-September 2, 2010, San Antonio, TX.

NPC (National Petroleum Council). (2011a). Management of produced water from oil and gas wells. (Paper
   #2-17). Washington, DC. http://www.npc.org/Prudent Development-Topic Papers/2-
   17 Management of Produced Water Paper.pdf

NRC (National Research Council). (2010). Management and effects of coalbed methane produced water in the
   western United States. Washington, DC: National Academies Press.
   http://www.nap.edu/catalog.php7record id=12915

Olmstead. SM: Muehlenbachs. LA: Shih. IS: Chu. Z: Krupnick. Al. (2013). Shale gas development impacts on
   surface water quality in Pennsylvania. PNAS 110: 4962-4967.
   http://dx.doi.org/10.1073/pnas.1213871110
               This document is a draft for review purposes only and does not constitute Agency policy.

June 2015                                       8-80                    DRAFT—DO NOT CITE OR QUOTE

-------
Hydraulic Fracturing Drinking Water Assessment                Chapter 8 - Wastewater Treatment and Waste Disposal
Olsson. 0: Weichgrebe. D: Rosenwinkel. KH. (2013). Hydraulic fracturing wastewater in Germany:
   composition, treatment, concerns. Environmental Earth Sciences 70: 3895-3906.
   http://dx.doi.org/10.1007/sl2665-013-2535-4

Orem. WH: Tatu. CA: Lerch. HE: Rice. CA: Bartos. TT: Bates. AL: Tewalt. S: Corum. MD. (2007). Organic
   compounds in produced waters from coalbed natural gas wells in the Powder River Basin, Wyoming, USA.
   Appl Geochem 22: 2240-2256. http://dx.doi.Org/10.1016/j.apgeochem.2007.04.010

PA PEP (Pennsylvania Department of Environmental Protection). (201 la). Letter from Pennsylvania
   Department of the Environment to US EPA Region 3 Administrator Shawn Garvin. Available online at
   http://www.epa.gov/region3/marcellus shale?#inforeqsbypadep

PA PEP. Road-spreading of brine for dust control and road stabilization. 43 Pa.B. 7377 § 78.70 (2011b).
   http://www.pabulletin.com/secure/data/vol43/43-50/2362b.html

PA PEP (Pennsylvania Department of Environmental Protection). (2015a). PA DEP oil & gas reporting
   website, statewide data downloads by reporting period, waste and production files downloaded for
   Marcellus/unconventional wells, July 2009 December 2014. Harrisburg, PA. Retrieved from
   https://www.paoilandgasreporting.state.pa.us/publicreports/Modules/DataExports/DataExports.aspx

PA DEP (Pennsylvania Department of Environmental Protection). (2015b). Technologically enhanced
   naturally occurring radioactive materials (TENORM) study report. Harrisburg, PA.
   http://www.elibrary.dep.state.pa.us/dsweb/Get/Document-105822/PA-DEP-TENORM-
   Study Report Rev. 0 01-15-2015.pdf

Parker, KM: Zeng, T: Harkness, I: Vengosh, A: Mitch, WA. (2014). Enhanced formation of disinfection
   byproducts in shale gas wastewater-impacted drinking water supplies. Environ Sci Technol 48:11161-
   11169. http://dx.doi.org/10.1021/es5028184

Pashin. 1C: Mcintyre-Redden. MR: Mann. SD: Kopaska-Merkel. DC: Varonka. M: Orem. W. (2014). Relationships
   between water and gas chemistry in mature coalbed methane reservoirs of the Black Warrior Basin. Int J
   Coal Geol 126: 92-105. http://dx.doi.Org/10.1016/j.coal.2013.10.002

Peraki. M: Ghazanfari. E. (2014). Electrodialysis treatment of flow-back water for environmental protection in
   shale gas development. In Shale gas development Shale energy engineering 2014. Reston, VA: American
   Society of Civil Engineers, http://dx.doi.org/10.1061/9780784413654.008

Plumlee. MH: Debroux. IF: Taffler. D: Graydon. IW: Mayer. X:  Dahm. KG: Hancock. NT: Guerra. KL: Xu. P:
   Drewes. IE: Cath. TY. (2014). Coalbed methane produced  water screening tool for treatment technology
   and beneficial use. 5: 22-34. http://dx.doi.Org/10.1016/j.juogr.2013.12.002

Porcelli. D: Kim. CK: Martin. P: Moore. WS: Phaneuf. M. (2014). Properties of radium. In The environmental
   behaviour of radium: revised edition. Vienna, Austria: International Atomic Energy Agency. http://www-
   pub.iaea.org/MTCD/Publications/PDF/trs476  web.pdf

Purestream (Purestream Services). (2013). Purestream services will begin commercial operations to treat
   Eagle Ford Shale produced and frac flow-back water in Gonzalez County, Texas. Retrieved from
   http://purestreamtechnology.com/index.php/component/content/article/72-press-releases/206-
   purestream-services-will-begin-commercial-operations-to-treat-eagle-ford-shale-produced-and-frac-
   flow-back-water-in-gonzalez-county-texas

Rahm. BG: Bates. IT: Bertoia. LR: Galford. AE: Yoxtheimer. DA: Riha. SI. (2013). Wastewater management and
   Marcellus Shale gas development: trends, drivers, and planning implications. J Environ Manage 120: 105-
   113. http://dx.doi.0rg/10.1016/j.jenvman.2013.02.029

Rahm. BG: Riha. SI. (2014). Evolving shale gas management: water resource risks, impacts, and lessons
   learned [Review]. Environ Sci Process Impacts  16:1400-1412. http://dx.doi.org/10.1039/c4em00018h

Ramirez. P. (2002). Oil field produced water discharges into wetlands in Wyoming. (97-6- 6F34). Cheyenne,
   WY: U.S. Fish and Wildlife Service, http://www.fws.gov/mountain-
   prairie/contaminants/papers/r6718c02.pdf


              This document is a draft for review purposes only and does not constitute Agency policy.
June 2015                                      8-81                   DRAFT—DO NOT CITE OR QUOTE

-------
Hydraulic Fracturing Drinking Water Assessment               Chapter 8 - Wastewater Treatment and Waste Disposal
Rassenfoss. S. (2011). From flowback to fracturing: Water recycling grows in the Marcellus shale. J Pet Tech
   63: 48-51.

Resnikoff. M: Alexandrova. E: Travers. I. (2010). Radioactivity in Marcellus shale: Report prepared for
   Residents for the Preservation of Lowman and Chemung (RFPLC). New York, NY: Radioactive Waste
   Management Associates.

Rich. AL: Crosby. EC. (2013). Analysis of reserve pit sludge from unconventional natural gas hydraulic
   fracturing and drilling operations for the presence of technologically enhanced naturally occurring
   radioactive material (TENORM). New Solutions: A Journal of Environmental and Occupational Health
   Policy 23:117-135. http://dx.doi.Org/10.2190/NS.23.l.h

Richardson. SD: Plewa. Ml: Wagner. ED: Schoeny. R: Demarini. DM. (2007). Occurrence, genotoxicity, and
   carcinogenicity of regulated and emerging disinfection by-products in drinking water: A review and
   roadmap for research [Review]. Mutat Res 636:178-242. http://dx.doi.Org/10.1016/j.mrrev.2007.09.00

Rowan. EL: Engle. MA: Kirby. CS: Kraemer. TF. (2011). Radium content of oil- and gas-field produced waters
   in the northern Appalachian Basin (USA): Summary and discussion of data. (Scientific Investigations
   Report 20115135). Reston, VA: U.S. Geological Survey, http://pubs.usgs.gov/sir/2011/5135/

Rozell. PI: Reaven. SI. (2012). Water pollution risk associated with natural gas extraction from the Marcellus
   Shale. Risk Anal 32:13821393. http://dx.doi.Org/10.llll/j.1539-6924.2011.01757.x

Rushton. L: Castaneda. C. (2014). Drilling into hydraulic fracturing and the associated wastewater
   management issues. Washington, WD: Paul Hastings, LLP. http://www.paulhastings.com/docs/default-
   source/PDFs/stay-current-hydraulic-fracturing-wastewater-management.pdf

Schmidt. CW. (2013). Estimating wastewater impacts from (racking. Environ Health Perspect 121: A117.
   http://dx.doi.org/10.1289/ehp.121-all7

Schubert. I: Rosenmeier. I: Zatezalo. M. (2014). A review of NORM/TENORM in wastes and waters associated
   with Marcellus shale gas development and production. In CL Meehan; JM Vanbriesen; F Vahedifard; X Yu;
   C Quiroga (Eds.), Shale energy engineering 2014: technical challenges, environmental issues, and public
   policy (pp. 492-501). Reston, VA: American Society of Civil Engineers.
   http://dx.doi.org/10.1061/9780784413654.052

Shafer. L. (2011). Water recycling and purification in the Pinedale anticline field: results from the anticline
   disposal project. In 2011  SPE Americas E&P health, safety, security & environmental conference.
   Richardson, TX: Society of Petroleum Engineers. http://dx.doi.org/10.2118/141448-MS

Shaffer. PL: Arias Chavez. LH: Ben-Sasson. M: Romero-Vargas Castrillon. S: Yip. NY: Elimelech. M. (2013).
   Desalination and reuse of high-salinity shale gas produced water: drivers, technologies, and future
   directions. Environ Sci Technol 47: 9569-9583.

Silva. IM: Matis. H: Kostedt. WL: Watkins. V. (2012). Produced water pretreatment for water recovery and salt
   production. (08122-36). Niskayuna, NY: Research Partnership to Secure Energy for America.
   http://www.rpsea.org/media/files/project/18621900/08122-36-FR-
   Pretreatment Water Mgt Frac Water  Reuse Salt-01-26-12.pdf

Sionix (Sionix Corporation). (2011). Sionix to build Bakken water treatment plant. Retrieved from
   http://www.rigzone.com/news/article pf.asp?a id=110613

Sirivedhin, T: Dallbauman, L. (2004). Organic matrix in produced water from the Osage-Skiatook petroleum
   environmental research site, Osage county, Oklahoma. Chemosphere 57: 463-469.

Skalak, Kl: Engle, MA: Rowan, EL: lolly, GD: Conko, KM: Benthem, Al: Kraemer, TF. (2014). Surface disposal of
   produced waters in western and southwestern Pennsylvania: Potential for accumulation of alkali-earth
   elements in sediments. Int I Coal Geol 126:162-170. http://dx.doi.Org/10.1016/i.coal.2013.12.001
               This document is a draft for review purposes only and does not constitute Agency policy.

June 2015                                       8-82                   DRAFT—DO NOT CITE OR QUOTE

-------
Hydraulic Fracturing Drinking Water Assessment                Chapter 8 - Wastewater Treatment and Waste Disposal
Slutz. I: Anderson. I: Broderick. R: Horner. P. (2012). Key shale gas water management strategies: An
   economic assessment tool. Paper presented at International Conference on Health, Safety and
   Environment in Oil and Gas Exploration and Production, September 11-13,2012, Perth, Australia.

Soeder. PI: Kappel. WM. (2009). Water resources and natural gas production from the Marcellus shale [Fact
   Sheet] (pp. 6). (U.S. Geological Survey, Fact Sheet 20093032). Soeder, DJ; Kappel, WM.
   http://pubs.usgs.gov/fs/2009/3032/pdf/FS2009-3032.pdf

States, S: Cyprych, G: Stoner, M: Wydra, F: Kuchta, I: Monnell, I: Casson, L. (2013). Marcellus Shale drilling and
   brominated THMs in Pittsburgh, Pa., drinking water. J Am Water Works Assoc 105: E432-E448.
   http://dx.doi.org/10.5942/iawwa.2013.105.0093

Stewart. DR. (2013b). Treatment for beneficial use of produced water and hydraulic fracturing flowback
   water. Presentation presented at US EPA Technical Workshop on Wastewater Treatment and Related
   Modeling For Hydraulic Fracturing, April 18, 2013, Research Triangle Park, NC.

Sumi. L. (2004). Pit pollution: Backgrounder on the issues, with a New Mexico case study. Washington, DC:
   Earthworks: Oil and Gas Accountability Project.
   http://www.earthworksaction.org/files/publications/PitReport.pdf

Swann. C: Matthews. I: Ericksen. R: Kuszaul. I. (2004). Evaluations of radionuclides of uranium, thorium, and
   radium with produced fluids, precipitates, and sludges from oil, gas, and oilfield brine injections wells.
   (DE-FG26-02NT15227). Washington, D.C.: U.S. Department of Energy.
   http://www.olemiss.edu/depts/mmri/programs/norm final.pdf

Tiemann, M: Folger, P: Carter, NT. (2014). Shale energy technology assessment: Current and emerging water
   practices. Washington, DC: Congressional Research Service, http://nationalaglawcenter.org/wp-
   content/uploads//assets/crs/R43635.pdf

Titler. RV: Curry. P. (2011). Chemical analysis of major constituents and trace contaminants of rock salt.
   Harrisburg, PA: Pennsylvania Department of Environmental Protection.
   http://files.dep.state.pa. us/water/Wastewater%20Management/WastewaterPortalFiles/Rock%20Salt%
   20Paper%20final%20052711.pdf

U.S. Department of lustice. (2014). Company owner sentenced to more than two years in prison for dumping
   tracking waste in Mahoning River tributary. Available online at
   http://www.iustice.gov/usao/ohn/news/2014/05auglupo.html (accessed March 4, 2015).

U.S. EPA (U.S. Environmental Protection Agency). (2006). National Primary Drinking Water Regulations:
   Stage 2 Disinfectants and Disinfection Byproducts Rule.
   http://water.epa.gov/lawsregs/rulesregs/sdwa/stage2/

U.S. EPA (U.S. Environmental Protection Agency). (2014f). Minimizing and managing potential impacts of
   injection-induced seismicity from class II disposal wells: Practical approaches [EPA Report]. Washington,
   D.C. http://www.epa.gov/r5water/uic/ntwg/pdfs/induced-seismicity-201502.pdf

U.S. EPA (U.S. Environmental Protection Agency). (2015d). DMR spreadsheet Pennsylvania wastewater
   treatment plants per Region 3 Information Request. Data provided by request. Washington, D.C.: Region
   3, U.S. Environmental Protection Agency.

U.S. EPA (U.S. Environmental Protection Agency). (2015e). Effluent data from Pennsylvania wastewater
   treatment plants per Region 3 Information Request. Data provided by request. Washington, D.C.: Region 3,
   U.S. Environmental Protection Agency.

U.S. EPA (U.S. Environmental Protection Agency). (2015f). EPA Enforcement and Compliance History. Online:
   Effluent Charts: SEECO-Judsonia Water Reuse Recycling Facility. Available online at
   http://echo.epa.gov/effluent-chartstfAR0052051

U.S. EPA (U.S. Environmental Protection Agency). (2015h). Key documents about mid-Atlantic oil and gas
   extraction. Available online at http://www.epa.gov/region3/marcellus shale/#aoinfoww (accessed May
   7,2015).


               This document is a draft for review purposes only and does not constitute Agency policy.
June 2015                                       8-83                    DRAFT—DO NOT CITE OR QUOTE

-------
Hydraulic Fracturing Drinking Water Assessment                Chapter 8 - Wastewater Treatment and Waste Disposal
U.S. EPA. National primary drinking water regulations public notification rule and consumer confidence
   report rule health effects language, (parts 141.201, and 141.151), (U.S. Government Publishing
   Office2015i). http://www.ecfr.gov/cgi-bin/text-
   idx?SID=4d25ec04bc44e54blefdf307855f3185&node=pt40.23.141&rgn=div5

U.S. EPA (U.S. Environmental Protection Agency). (2015p). Sources contributing bromide and inorganic
   species to drinking water intakes on the Allegheny river in western Pennsylvania [EPA Report].
   (EPA/600/R-14/430). Washington, D.C.

U.S. EPA (U.S. Environmental Protection Agency). (2015q). Technical development document for proposed
   effluent limitation guidelines and standards for oil and gas extraction. (EPA-821-R-15-003). Washington,
   D.C. http://water.epa.gov/scitech/wastetech/guide/oilandgas/unconv.cfm

U.S. GAP (U.S. Government Accountability Office). (2012). Energy-water nexus: Information on the quantity,
   quality, and management of water produced during oil and gas production. (GAO-12-156). Washington,
   D.C. http://www.gao.gov/products/GAO-12-156

USGS (U.S. Geological Survey). (2013c). National Water Information System (NWIS)  [Database]. Retrieved
   from http://waterdata.usgs.gov/nwis

USGS (U.S. Geological Survey). (2014e). U.S. Geological Survey national produced waters geochemical
   database v2.0 (PROVISIONAL). Available online at
   http://energy.usgs.gov/EnvironmentalAspects/EnvironmentalAspectsofEnergyProductionandUse/Produ
   cedWaters.aspx#3822349-data

Van Voast, WA. (2003). Geochemical signature of formation waters associated with coalbed methane. AAPG
   Bulletin 87: 667-676.

Veil, I A.  (2011). Water management practices used by Fayetteville shale gas producers. (ANL/EVS/R-11/5).
   Washington, DC: U.S. Department of Energy, National Energy Technology Laboratory.
   http://www.ipd.anl.gov/anlpubs/2011/06/70192.pdf

Veil. IA:  Puder. MG: Elcock. D: Redweik. Rl. (2004). A white paper describing produced water from production
   of crude oil, natural gas, and coalbed methane. Lemont, IL: Argonne National Laboratory.

Vengosh, A: lackson, RB: Warner, N: Darrah, TH: Kondash, A. (2014). A critical review of the risks to water
   resources from unconventional shale gas development and hydraulic fracturing in the United States.
   Environ Sci Technol 48:  36-52. http://dx.doi.org/10.1021/es405118y

Vidic. RD: Brantley. SL: Vandenbossche. IM: Yoxtheimer. D: Abad. ID. (2013). Impact of shale gas development
   on regional water quality [Review]. Science 340:1235009. http://dx.doi.org/10.1126/science.1235009

Volz. CD: Ferrar. K: Michanowicz. D: Christen. C: Kearney. S: Kelso. M: Malone. S. (2011). Contaminant
   Characterization of Effluent from Pennsylvania Brine Treatment Inc., Josephine Facility Being Released
   into  Blacklick Creek, Indiana County, Pennsylvania: Implications for Disposal of Oil and Gas Flowback
   Fluids from Brine Treatment Plants [Standard]. Volz, CD; Ferrar, K; Michanowicz, D; Christen, C; Kearney,
   S; Kelso, M; Malone, S. http://www2.epa.gov/hfstudy/contaminant-characterization-effluent-
   pennsylvania-brine-treatment-inc-josephine-facility

Walsh. IM. (2013). Water management for hydraulic fracturing in unconventional resourcesPart 1. Oil and
   Gas Facilities 2.

Walter, GR: Benke, RR: Pickett, DA. (2012). Effect of biogas generation on radon emissions from landfills
   receiving radium-bearing waste from shale gas development. J Air Waste Manag Assoc 62:1040-1049.
   http://dx.doi.org/10.1080/10962247.2012.696084

Warner. NR: Christie. CA: lackson. RB: Vengosh. A. (2013a). Impacts of shale gas wastewater disposal on
   water quality in western Pennsylvania. Environ Sci Technol 47:11849-11857.
   http://dx.doi.org/10.1021/es402165b
               This document is a draft for review purposes only and does not constitute Agency policy.

June 2015                                       8-84                    DRAFT—DO NOT CITE OR QUOTE

-------
Hydraulic Fracturing Drinking Water Assessment                Chapter 8 - Wastewater Treatment and Waste Disposal
Warner. NR: Kresse. TM: Hays. PD: Down. A: Karr. ID: lackson. RB: Vengosh. A. (2013b). Geochemical and
   isotopic variations in shallow groundwater in areas of the Fayetteville Shale development, north-central
   Arkansas. Appl Geochem 35: 207-220.

Water Research Foundation. (2010). Assessment of inorganics accumulation in drinking water system scales
   and sediments. Denver, CO. http://www.waterrf.org/PublicReportLibrary/3118.pdf

Weaver. IW: Xu. I: Mravik. SC.  (In Press) Scenario analysis of the impact on drinking water intakes from
   bromide in the discharge of treated oil and gas waste water. J Environ Eng.

Webb. CH: Nagghappan. L: Smart. G: Hoblitzell. I: Franks. R. (2009). Desalination of oilfield-produced water at
   the San Ardo water reclamation facility, Ca. In SPE Western regional meeting 2009. Richardson, TX:
   Society of Petroleum Engineers. http://dx.doi.org/10.2118/121520-MS

Wendel. K. (2011). Wastewater technologies critical for continued growth of Marcellus. Available online at
   http://www.ogfj.com/articles/print/volume-8/issue-ll/features/wastewater-technologies-critical-
   for.html (accessed March 9, 2015).

Wess, I: Ahlers, H: Dobson, S. (1998). Concise International Chemical Assessment Document 10: 2-
   Butoxyethanol. World Health Organization.
   http://www.who.int/ipcs/publications/cicad/cicad 10 revised.pdf

West Virginia PEP (West Virginia Department of Environmental Protection). (2011). Memorandum of
   agreement from the Division of Water and Waste Management to the Division of Highway:
   WVDOH/WVDEP Salt brine from gas wells agreement. Available online at
   http://www.dep.wv.gov/WWE/Documents/WVDOHWVDEP%20Salt%20Brine%20Agreement.pdf

Wilson. IM: Van Briesen. IM. (2013). Source water changes and energy extraction activities in the
   Monongahela River, 2009-2012. Environ Sci Technol 47: 1257512582.
   http://dx.doi.org/10.1021/es402437n

Wilson. IM: Vanbriesen. IM. (2012). Oil and gas produced water management and surface drinking water
   sources in Pennsylvania. Environmental Practice 14: 288-300.

Wolfe. D: Graham. G. (2002). Water rights and beneficial use of produced water in Colorado. Denver, CO:
   American Water Resources Association, http://www.gwpc.org/sites/default/files/event-
   sessions/Dick Wolfe PWC02 O.pdf

Xu, P: Drewes, IE: Heil, D. (2008). Beneficial use of co-produced water through membrane treatment:
   Technical-economic assessment. Desalination 225:139-155.
   http://dx.doi.0rg/10.1016/i.desal.2007.04.093

Younos. T: Tulou. KE. (2005). Overview of desalination techniques. Journal of Contemporary  Water Research
   & Education 132: 3-10. http://dx.doi.Org/10.llll/i.1936-704X.2005.mpl32001002.x

Zhang, T: Gregory, K: Hammack, RW: Vidic, RD. (2014b). Co-precipitation of radium with barium and
   strontium  sulfate and its impact on the fate of radium during treatment of produced water from
   unconventional gas extraction. Environ Sci Technol 48: 4596-4603.
   http://dx.doi.org/10.1021/es405168b
               This document is a draft for review purposes only and does not constitute Agency policy.

June 2015                                      8-85                   DRAFT—DO NOT CITE OR QUOTE

-------
Hydraulic Fracturing Drinking Water Assessment       Chapter 9 - Identification and Hazard Evaluation of Chemicals across the Hydraul
                         Chapter 9
Identification and Hazard Evaluation of Chemicals
across the Hydraulic Fracturing Water Cycle
          This document is a draft for review purposes only and does not constitute Agency policy.
June 2015                                         DRAFT—DO NOT CITE OR QUOTE

-------
      Hydraulic Fracturing Drinking Water Assessment          Chapter 9 - Identification and Hazard Evaluation of Chemicals across the Hydraul


      9.  Identification and Hazard  Evaluation of Chemicals

          across the Hydraulic Fracturing Water Cycle

      9.1. Introduction
 1    Chapters 4 through 8 of this assessment each present a stage of the hydraulic fracturing water cycle
 2    and the mechanisms by which activities in those stages produce potential impacts on drinking
 3    water resources. In contrast, this chapter presents and integrates what is known about chemicals
 4    across stages of the hydraulic fracturing water cycle (i.e., used in hydraulic fracturing fluids and
 5    detected in hydraulic fracturing wastewater). The discussion is focused on available information
 6    about (1] chronic toxicity values—specifically, the available noncancer oral reference values (RfVs)
 7    and cancer oral slope factors (OSFs)—of chemicals that could occur in drinking water resources;
 8    and (2) properties of chemicals that could affect their occurrence in drinking water resources (see
 9    Chapters 5 and 7).l-2 To the extent that information was available to do so, knowledge of
10    toxicological and chemical properties was combined to illustrate an approach that may provide
11    preliminary insights about the relative hazard potential that chemicals could pose to drinking water
12    resources.

13    Risk assessment and risk management decisions will be informed by the scientific information on
14    the toxicity of chemicals in hydraulic fracturing fluid and wastewater, which recent authors note is
15    incomplete (Goldstein etal.. 2014). The U.S. House of Representatives' Committee on Energy and
16    Commerce Minority Staff released a report in 2Oil noting that more than 650 products (i.e.,
17    chemical mixtures) used in hydraulic fracturing contain 29  chemicals that are either known or
18    possible human carcinogens or are currently regulated under the Safe Drinking Water Act (House
19    of Representatives. 2011). However, that report did not characterize the potential toxicity of many
20    of the other compounds known to occur in hydraulic fracturing fluids or wastewater. More recently,
21    Kahrilas etal. (2015] reviewed the toxicity and physicochemical properties of biocides used in
22    hydraulic fracturing. Stringfellow et al. (2014] examined the toxicity and physicochemical
23    properties of several classes of chemicals that are reportedly used in hydraulic fracturing;  however,
24    this study only reported acute toxicity (from lethal doses], which may differ from the effects of low-
25    dose, chronic exposure to these chemicals. Wattenberg et al. (In Press] assessed the acute and
26    chronic toxicity data that was available for 168 chemicals from the FracFocus database that had at
27    least 25 reports of use in North Dakota. The authors found that 113 of these chemicals had some
28    health hazard data available, but determined that there were significant data gaps, particularly with
29    regards to what is known about the potential chronic toxicity of these chemicals. Overall, available
      1 A reference value (RfV) is an estimate of an exposure for a given duration to the human population (including
      susceptible subgroups) that is likely to be without an appreciable risk of adverse health effects over a lifetime. RfV is a
      generic term not specific to a given route of exposure. In the context of this chapter, the term RfV refers to reference
      values for noncancer effects occurring via the oral route of exposure and for chronic durations, except where noted.
      Source: IRIS Glossary (U.S. EPA.2011d).
      2 An oral slope factor (OSF) is an upper-bound, approximating a 95% confidence limit, on the increased cancer risk from a
      lifetime oral exposure to an agent. This estimate, usually expressed in units of proportion (of a population) affected  per
      mg/kg day, is generally reserved for use in the low dose region of the dose response relationship, that is, for exposures
      corresponding to risks less than 1 in 100.
                   This document is a draft for review purposes only and does not constitute Agency policy.
      June 2015                                     9-1                   DRAFT—DO NOT CITE OR QUOTE

-------
      Hydraulic Fracturing Drinking Water Assessment          Chapter 9 - Identification and Hazard Evaluation of Chemicals across the Hydraul


 1    information indicates that there may be hundreds of chemicals associated with the hydraulic
 2    fracturing water cycle for which toxicological data is limited or unavailable.

 3    Furthermore, the potential public health impact of hydraulic fracturing processes is not well
 4    understood [Finkeletal.. 2013: Colbornetal.. 2011]. Potential public health implications are
 5    highlighted in the recent studies by McKenzie etal. [2014] and Kassotis etal. [2014], but as of early
 6    2015, there is a lack of published, peer-reviewed epidemiological or toxicological studies that have
 7    examined health effects resulting from water contamination due to hydraulic fracturing. However,
 8    numerous authors have noted that with the recent increase in hydraulic fracturing operations there
 9    may be an increasing potential for significant public health and environmental impacts via ground
10    and surface water contamination [Goldstein etal., 2014: Finkeletal., 2013: Korfmacher etal., 2013:
11    Weinhold. 2012].

12    This chapter provides a compilation of the chemicals used or released during the fracturing
13    process, and information about their potential health effects. The data are presented in this chapter
14    as follows.

15    Section 9.2 discusses how ten information sources, including the EPA's analysis of the FracFocus
16    database [U.S. EPA. 2015a], were used to create a list of chemicals used in or detected in various
17    stages of the hydraulic fracturing water cycle. This chemical list was initially presented in the EPA's
18    2012 interim progress report [U.S. EPA, 2012f], and has been updated in this assessment with
19    additional chemicals from FracFocus. The consolidated chemical list includes chemicals that are
20    reportedly added to hydraulic fracturing fluids in the chemical mixing stage, as well as fracturing
21    fluid chemicals, formation chemicals, or their reaction products that may be carried in flowback or
22    produced water. Although over half of the chemicals cited on this list are listed in the EPA
23    FracFocus database, this chapter is not meant to be interpreted as a hazard evaluation of the
24    chemicals listed in the EPA FracFocus  database alone.

25    Section 9.3 provides an overview of the methods that were used for gathering information on
26    toxicity and physicochemical properties for all chemicals that were identified in Section 9.2, and
2 7    outlines the number of chemicals that had available data on these properties. For toxicological data,
28    the primary focus is on peer-reviewed, selected chronic oral RfVs and OSFs. This section also
29    discusses additional potential sources of toxicity information: estimates of toxicity predicted using
30    Quantitative Structure Activity Relationship [QSAR] modeling, or toxicological information
31    available on the EPA's Aggregated Computational Toxicology Resource [ACToR] database. This
32    chapter is focused on potential human health hazards of chemicals for the oral route of exposure
33    (drinking water]; therefore, the toxicological properties and physicochemical ranking metrics
34    described herein (see below] do not necessarily apply to other routes of exposure, such as
35    inhalation or dermal exposure. In addition, this analysis is focused on individual chemicals rather
36    than mixtures of chemicals used as additives. Furthermore, the propensity for a chemical to pose a
37    physical hazard (e.g. the flammability  and explosive ness of stray gas methane] are not considered
38    here.
                   This document is a draft for review purposes only and does not constitute Agency policy.
      June 2015                                     9-2                   DRAFT—DO NOT CITE OR QUOTE

-------
      Hydraulic Fracturing Drinking Water Assessment          Chapter 9 - Identification and Hazard Evaluation of Chemicals across the Hydraul


 1    Many chemicals reported in hydraulic fracturing were identified as being of interest in previous
 2    chapters of this report This includes the most frequently used chemicals in hydraulic fracturing
 3    fluid (Chapter 5), the most and least mobile chemicals in hydraulic fracturing fluid (Chapter 5), and
 4    inorganic chemicals and pesticides that may be detected in flowback and produced water (Chapter
 5    7). The available selected chronic oral RfVs and OSFs for these chemicals are summarized in Section
 6    9.4.

 7    Section 9.5 describes the hazard identification and hazard evaluation of chemicals for which data
 8    was available on toxicity, occurrence, and physicochemical properties. ^2 For hazard identification,
 9    the selected chronic oral RfVs and OSFs and health effects for these chemicals are presented and
10    summarized. To illustrate one approach to integrate toxicity, occurrence and physicochemical
11    properties data to generate a hazard potential score, a multicriteria decision analysis (MCDA)
12    framework was developed. In this context, occurrence and physicochemical property data were
13    used as metrics to estimate the likelihood that a chemical could impact drinking water resources.
14    Chemicals considered in these hazard evaluations include a subset of chemicals from the FracFocus
15    database, as well as a subset of chemicals that have been detected in flowback and produced water.

16    In general, characterizing chemicals and their properties on a national scale is challenging and that
17    the use and occurrence of chemicals is likely to differ between geological basins and possibly on a
18    well-to-we 11 basis (see Chapters 5 and 7). Therefore, for the protection of public health at the
19    community level, chemical hazard evaluations may be most useful to conduct on a regional or site-
20    specific scale. This level of analysis is outside the scope of this report; however, the methods of
21    hazard evaluation presented here can also be applied on a regional or site-specific scale  in order to
2 2    identify chemicals that may present the greatest potential human health hazard.

      9.2.  Identification of Chemicals Associated with the Hydraulic Fracturing Water
            Cycle
23    As the initial step towards developing a hazard evaluation, the EPA compiled a list of chemicals that
24    are used in or released by hydraulic fracturing operations across the country (U.S. EPA, 2012f). Ten
25    sources of information (described in Appendix A) were used to develop this list This consolidated
26    list was used to compile two sublists: (1) a list of chemicals known to be used in hydraulic
27    fracturing fluids, and (2) a list of chemicals that are reported to have been detected in hydraulic
28    fracturing flowback and produced water. It is likely that, as industry practices change, chemicals
29    may be used or detected that are not included on these lists. In addition, those chemicals that are
30    considered proprietary and identified as confidential business information (CBI) by well operators
31    are not listed or considered.
      1 Hazard identification is a process for determining if a chemical or a microbe can cause adverse health effects in humans
      and what those effects might be. See Terms of Environment at:
      http://iaspub.epa.gov/sor internet/registry/termreg/searchandretrieve/termsandacronyms/search.do.
      2 Hazard evaluation is a component of risk assessment that involves gathering and evaluating data on the types of health
      injuries or diseases (e.g., cancer) that may be  produced by a chemical and on the conditions of exposure under which such
      health effects are produced. See Terms of Environment at:
      httD://iasDub.epa.gov/sor internet/registry/termreg/searchandretrieve/termsandacronvms/search.do.
                    This document is a draft for review purposes only and does not constitute Agency policy.
      June 2015                                     9-3                    DRAFT—DO NOT CITE OR QUOTE

-------
      Hydraulic Fracturing Drinking Water Assessment          Chapter 9 - Identification and Hazard Evaluation of Chemicals across the Hydraul


 1    In total, the EPA identified 1,173 chemicals as being used in hydraulic fracturing fluid and/or
 2    detected in flowback and produced water. The complete list of chemicals and associated data is
 3    available in Appendices A and B.:

      9.2.1.  Chemicals Used in Hydraulic Fracturing Fluids
 4    Of the 1,173 total chemicals, the EPA identified 1,076 chemicals as being used in hydraulic
 5    fracturing fluids. Of these, 692 chemicals were listed in the FracFocus database, and therefore had
 6    information available in order to calculate their nationwide frequency of use [U.S. EPA, 2015a].2
 7    Frequency of use for individual  chemicals ranged from low (481 chemicals on the list were used in
 8    less than 1% of wells nationwide) to very high (methanol was used in 73% of wells nationwide).
 9    Furthermore, only 32 chemicals (excluding water, quartz, and sodium chloride) were reported in at
10    least 10% of the disclosures nationwide (see Section 5.4 and Table 5-2). As noted previously, the
11    FracFocus database does not list or consider those chemicals identified as CBI. The EPA determined
12    that approximately 70% of the disclosures in the FracFocus database contain at least one CBI
13    chemical, and for those disclosures, the average number of CBI chemicals per disclosure was five
14    (see Section 5.4, Text Box 5-3). Additionally, as noted previously, approximately 35% of FracFocus
15    ingredient records were not able to be assigned standardized chemical names. These ingredient
16    records were excluded from the EPA's analysis (see Section 5.10).

      9.2.2.  Chemicals Detected in  Flowback and  Produced Water
17    Of the 1,173 total chemicals, 134 were identified as having been detected in flowback or produced
18    water. Included among these chemicals are naturally occurring organic compounds, metals,
19    radionuclides, and pesticides. As reported in Chapter 7, concentration data in flowback or produced
20    water are available for 75 of these 134 chemicals (see Appendix E), including inorganic
21    contributors to salinity (Tables  E-4 and E-5), metals (Tables E-6 and E-7), radioactive constituents
22    (Table E-8), and organic constituents (Tables E-9, E-10, and E-ll). For these chemicals with
23    concentration data, the measured concentrations spanned several orders of magnitude. For
24    instance, for organic chemicals in produced water from the Marcellus shale formation (Table E-10),
25    average or median measured concentrations ranged from 2.7 |ig/L for N-nitrosodiphenylamine to
26    400 |ig/L for carbon disulfide. According to the sources listed in Appendix A, 37 of the total 134
27    chemicals in flowback and produced water were also identified as being used in hydraulic
28    fracturing fluid.

      9.3. Toxicological and Physicochemical Properties of Hydraulic Fracturing
           Chemicals
29    Toxicological and physicochemical data were collected as available for each of the chemicals
30    identified in Appendix A. The criteria used to identify and select toxicity values, RfVs and OSFs
31    (Section 9.3.1), and the method used to generate physicochemical property data (Section 9.3.2) are
32    discussed below. A summary of the available data for these chemicals follows in Section 9.3.3. Other
      1 The list of 1,173 chemicals was finalized as of this 2015 draft assessment. There may be chemicals present in flowback
      and produced water that are not included on this list.
      2 The FracFocus frequency of use data presented in this chapter is based on 35,957 well disclosures.
                   This document is a draft for review purposes only and does not constitute Agency policy.
      June 2015                                     9-4                   DRAFT—DO NOT CITE OR QUOTE

-------
      Hydraulic Fracturing Drinking Water Assessment         Chapter 9 - Identification and Hazard Evaluation of Chemicals across the Hydraul


 1    possible sources of toxicological information, including QSAR-approaches and the EPA's ACToR
 2    database, are discussed in Section 9.3.4.

      9.3.1.  Selection of Toxicity Values: Reference Values (RfVs) and Oral Slope Factors (OSFs)
 3    Toxicity information spans a wide range with respect to extent, quality and reliability. Toxicological
 4    data may include assessments from various sources including state, national, international, private
 5    and academic organizations as well as toxicity information which has not been formalized into an
 6    assessment and may be found in the scientific literature and databases including results from
 7    guideline tests, high throughput screening assays, alternative assays, and QSAR models. The
 8    sources of toxicity values - specifically, chronic oral RfVs and OSFs - selected for the purposes of
 9    this chapter are based on criteria developed specifically for this report For many of the chemicals
10    used in hydraulic fracturing or found in flowback or produced water there may be relevant
11    information, including cancer and noncancer-related information, from one or  more sources that
12    were not evaluated in this chapter.

13    The sources of RfVs and OSFs selected for the purposes of this chapter met the  followingkey
14    criteria: 1) the body or organization generating or producing the peer-reviewed RfVs, peer-
15    reviewed OSFs, or peer-reviewed qualitative assessment must be a governmental or
16    intergovernmental body; 2) the data source must include peer-reviewed RfVs, peer-reviewed OSFs,
17    or peer reviewed qualitative assessments; 3) the RfVs, OSFs, or qualitative assessments must be
18    based on peer-reviewed scientific data; 4) the RfVs, OSFs, or qualitative assessments must be
19    focused on protection of the general public; and 5) the body generating the RfVs, OSFs, or
2 0    qualitative assessments must be free of conflicts of interest with respect to the chemicals for which
21    it derives reference values or qualitative assessments. More detail on these criteria for selection
2 2    and inclusion of data sources, as well as the full list of data sources that were considered for this
23    study, are available in Appendix G.

24    RfVs and OSFs available from the EPA IRIS, the EPA PPRTV program, ATSDR, and the EPA HHBP
25    program all met the criteria for selection and inclusion as a data source (see Table 9-1). An attempt
26    was made to identify and acquire RfVs and OSFs from all 50 states, but only the peer-reviewed state
27    RfVs and OSFs from California met the stringent selection criteria and were included because of the
28    state's extensive peer review process.1 One international source for RfVs, the World Health
29    Organization's (WHO) International Programme on Chemical Safety (IPCS) Concise International
30    Chemical Assessment Documents (CICAD), also met the selection criteria. The International Agency
31    for Research on Cancer  (IARC) and U.S. National Toxicology Program (NTP) Report on Carcinogens
32    also met the criteria and were used as additional sources for qualitative cancer classifications.

      Table 9-1. Sources of selected toxicityRfVs and OSFs.
      Source
Website
      1 State RfVs and OSFs are also publicly available from Alabama, Texas, Hawaii, and Florida, but they did not meet the
      criteria for consideration as sources for RfVs and OSFs in this report. See Appendix G for details.
                    This document is a draft for review purposes only and does not constitute Agency policy.
      June 2015                                     9-5                   DRAFT—DO NOT CITE OR QUOTE

-------
      Hydraulic Fracturing Drinking Water Assessment
Chapter 9 - Identification and Hazard Evaluation of Chemicals across the Hydraul
Source
EPA Integrated Risk Information System (IRIS)
Human Health Benchmarks for Pesticides (HHBP)
EPA Provisional Peer-Reviewed Toxicity Value
(PPRTV) database
Agency for Toxic Substances and Disease Registry
(ATSDR) Minimum Risk Levels
State of California Toxicity Criteria Database
International Programme on Chemical Safety (I PCS)
Concise International Chemical Assessment
Documents (CICAD)
Website
http://cfpub.epa.gov/ncea/iris/index.cfm7fuseaction
=iris.showSubstanceList

http://iaspub.epa.gov/apex/pesti cides/f?p=HHBP:h
ome
http://hhpprtv.ornl.gov/index.html

http://www.atsdr.cdc.gov/toxprofiles/index.aspttbo
okmarkOS

http://www.oehha.org/tcdb/index.asp

http://www.who.int/ipcs/publications/cicad/en/

 1    EPA generally applies federal RfVs and OSFs for use in human health risk assessments. Therefore,
 2    for the purpose of hazard evaluation and making comparisons between chemicals in this chapter,
 3    only federal chronic oral RfVs and OSFs from the EPA IRIS, the EPA PPRTV program, ATSDR, and
 4    the EPA HHBP program were used. Furthermore, when a chemical had an RfV and/or OSF from
 5    more than one federal source, a modification of the EPA Office of Solid Waste and Emergency
 6    Response (OSWER) Directive 9285.7-53 tiered hierarchy of toxicity values was applied to
 7    determine which value to use. A single RfV and/or OSF was selected from the sources in this order:
 8    IRIS, HHBP, PPRTV, and ATSDR.1 The RfVs considered from these sources included noncancer
 9    reference doses (RfDs) from the IRIS, PPRTV, and HHBP programs, and oral minimum risk levels
10    (MRLs) from ATSDR.2'3

11    Because there are relatively few OSFs available compared to RfVs, OSFs were excluded from
12    discussion in this chapter; however, all available OSFs are reported in Appendix G. The EPA
13    drinking water maximum contaminant levels (MCLs) were also excluded from this analysis because
14    they are treatment-based. MCLs are set as close to maximum containment level goal (MCLG) values
15    as feasible. However, MCL and MCLGs values are still reported in Appendix G for the sake of
16    completeness.
      1 The OSWER hierarchy indicates that sources should be used in this order: IRIS, PPRTV, and then other values. In this
      report, this hierarchy was followed, but HHBP values were used in lieu of an IRIS value for a few chemicals. See Appendix
      G for details.
      2ARfD is an estimate (with uncertainty spanning perhaps an order of magnitude) of a daily oral exposure to the human
      population (including sensitive subgroups) that is likely to be without an appreciable risk of deleterious effects during a
      lifetime. For the complete definition see Appendix G.
      3 An MRL is an estimate of daily human exposure to a hazardous substance at or below which the substance is unlikely to
      pose a measurable risk of harmful (adverse), noncancerous effects. MRLs are calculated for a route of exposure
      (inhalation or oral) over a specified time period (acute, intermediate, or chronic). MRLs should not be used as predictors
      of harmful (adverse) health effects. Chronic MRL: Duration of exposure is 365 days or longer.
                    This document is a draft for review purposes only and does not constitute Agency policy.
      June 2015                                      9-6                    DRAFT—DO NOT CITE OR QUOTE

-------
      Hydraulic Fracturing Drinking Water Assessment
Chapter 9 - Identification and Hazard Evaluation of Chemicals across the Hydraul
      9.3.2.  Physicochemical Properties
 1    As presented in Chapter 5, EPI Suite™ software was used to generate data on the physicochemical
 2    properties of chemicals on the consolidated list EPI Suite™ provides an estimation of
 3    physicochemical properties based upon chemical structure, and will additionally provide
 4    empirically measured values for these properties when they are available for a given chemical. For
 5    more details on the software and on the use of physicochemical properties for fate and transport
 6    estimation, see Chapter 5.

      9.3.3.  Summary of Selected Toxicological and Physicochemical Property Data for Hydraulic
             Fracturing Chemicals
 7    Figure 9-1 summarizes the availability of selected RfVs and OSFs and physicochemical data for the
 8    1,173 hydraulic fracturing chemicals identified by the EPA.
( If 1
Hydraulic
Occurrence
Fracturing
Chemical List
v y V y
/^ >
Toxicological &
Physiochemical
Data
v y

1,173 Chemicals Associated with Hydraulic Fracturing:
Summary of Available Data:
/- A
Used in
Hydraulic Fracturing
Fluid:
1,076 chemicals
V J

f ~\
Frequency of Use
(FracFocus):
692 chemicals
V J


(^ A
Detected in
Flowback
or Produced Water:
134 chemicals
V. J



f ~\
Measured Concentration
in Flowback or Produced
Water (Appendix E):
75 chemicals
V J

( Chronic Oral RfV or OSF:
RfV or OSF (all sources): 147 chemicals
Federal RfV or OSF : 126 chemicals
Federal RfV : 119 chemicals
I Federal OSF: 29 chemicals


f N
Physico-
chemical Data
(EPI Suite):
515 chemicals
V J
Summary of Data Gaps:
/" A
Hydraulic Fracturing Fluid:
Chemical list excludes
confidential business
information
V J


f A
Flowback/Produced
Water:
Few studies are available
V J

f ~\
Lacking Frequency of Use Data:
384 chemicals used in hydraulic fracturing fluids
V J

C ~\
Lacking Chronic
Oral RfV or OSF
(all sources):
1,026 chemicals
V J


f ~\
Lacking Physiochemical
Properties Data
(EPI Suite):
658 chemicals
V }
      Figure 9-1. Overall representation of the selected RfVs and OSFs, occurrence data, and
                physicochemical data available for the 1,173 hydraulic fracturing chemicals
                identified by the EPA.
 9    Of the 1,173 chemicals identified by the EPA, only 147 (13%) have federal, or state, or international
10    chronic oral RfVs and/or OSFs from sources listed in Table 9-1. Therefore, chronic RfVs and/or
11    OSFs from the selected sources are lacking for 87% of chemicals that the EPA has identified as
12    associated with hydraulic fracturing. All available chronic RfVs and OSFs from the sources listed in
13    Table 9-1 are tabulated in Appendix G. Chronic RfVs and OSFs for chemicals used in hydraulic
14    fracturing fluids are listed in Tables G-la through G-lc, and chronic RfVs and OSFs for chemicals
                   This document is a draft for review purposes only and does not constitute Agency policy.
      June 2015                                     9-7                   DRAFT—DO NOT CITE OR QUOTE

-------
      Hydraulic Fracturing Drinking Water Assessment          Chapter 9 - Identification and Hazard Evaluation of Chemicals across the Hydraul


 1    reported in hydraulic fracturing flowback and produced water are listed in Tables G-2a through G-
 2    2c.

 3    From the U.S. federal sources that were considered here, the availability of chronic RfVs and OSFs
 4    can be summarized as follows. Of the 1,173 chemicals on the consolidated list, a total of
 5    126 chemicals have federal chronic RfVs and/or OSFs. Of these 126 chemicals, 119 have federal
 6    chronic RfVs, and 29 have federal OSFs (see Figure 9-1). 22 chemicals have both a federal chronic
 7    RfV and a federal OSF, while 7 have a federal OSF only.

 8    Overall, when chemicals in hydraulic fracturing fluid and chemicals in flowback are considered
 9    separately, the availability of chronic RfVs and OSFs can be summarized as follows:

10       •   For the 1,076 chemicals used in hydraulic fracturing fluid, chronic RfVs from all of the
11           selected federal, state, and international sources were available for 90 chemicals (8.4%).
12           From the federal sources alone, chronic RfVs were available for 73 (6.8%), and OSFs were
13           available for 15 (1.4%).

14       •   For the 134 chemicals reported in flowback and produced water, chronic RfVs from all of
15           the selected federal, state, and international sources were available for 83 chemicals (62%).
16           From the federal sources alone, chronic RfVs were available for 70 chemicals (52%), and
17           OSFs were available for 20 (15%).

18    The IRIS database was the most abundant source of the federal chronic RfVs and OSFs. IRIS had
19    available RfDs for 77 of the total 1,173 chemicals,  and OSFs for 27 chemicals. Of the other federal
20    data sources, the PPRTV database had RfDs for 33 chemicals, and OSFs for 2 chemicals; the HHBP
21    database had RfDs for 11 chemicals, butdid nothave available OSFs for any of the chemicals; and
22    the ATSDR database had chronic oral MRLs for 27 chemicals.

23    In addition to these chronic values, many of the chemicals also have less-than-chronic federal oral
24    RfVs. Subchronic or acute federal RfVs were identified for 91 chemicals on the consolidated list,
25    including 55 chemicals used in hydraulic fracturing fluid (Table G-ld), and 56 chemicals reported in
26    flowback or produced water (Table G-2d). There were 8 chemicals that had less-than-chronic RfVs
27    but lacked a chronic RfV. All of these less-than-chronic RfVs were found on the PPRTV or ATSDR
28    databases; the IRIS database did not have less-than-chronic RfVs for any of these chemicals. These
29    values are not discussed in this report, but are provided in Appendix G as supporting information.

30    From the total list of 1,173 chemicals associated with hydraulic fracturing, EPI Suite™ was able to
31    generate data on physicochemical properties for 515 (44%) of the chemicals (see Appendix A). The
32    remaining 658 chemicals lacked the structural information necessary to generate an estimate.

      9.3.4.  Additional Sources of Toxicity Information
33    Because the majority of chemicals identified in this report do not have RfVs and/or OSFs from the
34    selected sources, it is likely that risk assessors at the local and regional level may turn to alternative
35    sources of toxicity information. This section discusses other publicly accessible sources of
36    toxicological data that are lower on the continuum of quality and reliability in comparison to the

                   This document is a draft for review purposes only and does not constitute Agency policy.
      June 2015                                    9-8                  DRAFT—DO NOT CITE OR QUOTE

-------
      Hydraulic Fracturing Drinking Water Assessment          Chapter 9 - Identification and Hazard Evaluation of Chemicals across the Hydraul


 1    selected RfVs and OSFs described above. Because the quality of these data is unknown for most
 2    chemicals, values from these data sources are not included in the hazard evaluation in this report.

      9.3.4.1. Estimated Toxicity Using Quantitative Structure Activity Relationships (QSAR)
 3    One potential source of toxicological information is QSAR software, which is able to provide
 4    estimates or predictions of toxicity based on chemical structure. QSAR models for toxicity have
 5    been used and evaluated in a number of previous studies published in the peer reviewed literature
 6    [Ruppetal.,2010: Venkatapathy et al.. 2004: Moudgal etal.. 2003]. A key advantage to QSAR
 7    models is that they are able to rapidly and inexpensively estimate toxicity values for chemicals.
 8    Compared to toxicological studies involving animals or in vitro methods, which have monetary,
 9    time, and ethical considerations associated with them, the QSAR method requires only information
10    on chemical structure in order to generate a toxicity estimation. These values may be of lower
11    quality and less reliable than values generated using traditional toxicological methods. However,
12    because they increase the available pool of toxicity information, QSAR estimates may potentially be
13    a useful resource for risk assessors that are  faced with evaluating potential exposures to data-poor
14    chemicals.

      9.3.4.2. Chemical Data Available from ACToR
15    An additional tool for obtaining toxicological information is the ACToR database.1 ACToR is a large
16    data warehouse developed by the EPA to gather and house large and disparate amounts of public
17    data on chemicals including chemical identity, structure, physicochemical properties, in vitro assay
18    results, and in vitro toxicology data [Judsonetal.. 2009]. ACToR contains data on over 500,000
19    chemicals from over 2,500 sources, covering many domains including hazard,  exposure, risk
20    assessment, risk management, and use. Data in ACToR is organized on several levels of "assays" and
21    "assay categories". The information available in ACToR ranges from the federal RfVs and OSFs
22    discussed in Section 9.3.1, which have undergone extensive peer review, to other toxicity values
23    and study and test results that have undergone little to no peer review.

24    The ACToR database was searched for information related to the total list of 1,173 chemicals
25    associated with hydraulic fracturing.2 For the purposes of this chapter, the database was first
2 6    searched for all of the assays and assay categories that had data on these chemicals. This initial
2 7    search was then filtered to only include the assay categories that are most relevant to toxicity via
28    the oral route of exposure (drinking water]. These assay categories were assigned into the
29    following nine data classes: carcinogenicity, dose response values, drinking water criteria,
30    genotoxicity/mutagenicity, hazard identification, LOAEL/NOAEL, RfD, slope factor, and water
31    quality criteria. The type of data and examples of the data sources included in these data classes can
32    be found in the ACToR database documentation.

33    When all assays and assay categories were considered, it was found that all but 2 8 of the total 1,173
34    chemicals had available data on ACToR. When only the relevant assays and assay categories were
35    considered, 642 (55%] of the chemicals were found to have data on ACToR. The fraction of
      1 The ACToR database is available at: http://actor.epa.gov.
      2 The ACToR database was queried for the total list of 1,173 chemicals on April 1,2015.
                   This document is a draft for review purposes only and does not constitute Agency policy.
      June 2015                                     9-9                   DRAFT—DO NOT CITE OR QUOTE

-------
      Hydraulic Fracturing Drinking Water Assessment
Chapter 9 - Identification and Hazard Evaluation of Chemicals across the Hydraul
 1    chemicals that had at least one data point in each of the nine ACToR data classes is shown in Figure
 2    9-2. As can be seen in Figure 9-2, about half of the chemicals had some information on water quality
 3    criteria, while fewer chemicals had information on the other classes of data.
                                   Fraction of Chemicals with Data by Class
                                                        Water Quality Criteria
                                               Slope Factor

                                              RfD
                                          LOAEL/NOAEL
                                        Hazard ID
                                        Genotoxicity/Mutagenicity
                                    Drinking Water Criteria
                                    Dose Response Values
                                 Carcinogenicity
                           i
                          0.0
 I
0.4
 I
0.8
                                 0.2           0.4          0.6

Figure 9-2. Fraction of chemicals with at least one data point in each ACToR data class.
 4    Focusing on the 1,026 chemicals that lacked a chronic RfV and/or OSF from the selected sources
 5    described in Section 9.3.1, 497 (48%) of these chemicals had available data on ACToR. Because
 6    ACToR has a significant amount of data on potential chemical hazards, including for some data-poor
 7    chemicals, ACToR might help to  fill data gaps in the ongoing effort to understand potential hazards
 8    of hydraulic fracturing chemicals. Since the quality of the non-peer reviewed values is not known,
 9    these data are not considered in the hazard evaluation.

      9.4. Hazard Identification of Reported Hydraulic Fracturing Chemicals
10    This section focuses on chemicals that were identified as being of particular interest in previous
11    chapters of this report, or which otherwise may be of particular interest to risk assessors. Federal
12    RfVs are identified for these chemicals as available.
                    This document is a draft for review purposes only and does not constitute Agency policy.
      June 2015                                     9-10                   DRAFT—DO NOT CITE OR QUOTE

-------
      Hydraulic Fracturing Drinking Water Assessment
Chapter 9 - Identification and Hazard Evaluation of Chemicals across the Hydraul
      9.4.1.  Selection of Additional Chemicals for Hazard Identification
 1    Four subsets of chemicals were identified as being of interest in Chapter 5 (Chemical Mixing) and
 2    Chapter 7 (Flowback and Produced water):

 3       1.  Chapter 5: The most frequently used chemicals in hydraulic fracturing fluid, defined as
 4           chemicals being reported to the FracFocus database in at least 10% of well disclosures [U.S.
 5           EPA. 2015a).

 6       2.  Chapter 5: The top 20 most and leastmobile chemicals from the EPA's analysis of the
 7           FracFocus database [U.S. EPA, 2015a), as determined based on the octanol-water partition
 8           coefficient (Kow) from EPI Suite™.

 9       3.  Chapter 7: Inorganic chemicals that may be returned to the surface in flowback and
10           produced water. This includes metals, inorganic ions, and naturally occurring radioactive
11           material (NORM).

12       4.  Pesticides occurring in flowback and produced water.

13    The hazard identification for these four subsets of chemicals is presented below.

      9.4.2.  Hazard Identification Results
      9.4.2.1. Most Frequently Used Chemicals in Hydraulic Fracturing Fluid (FracFocus)
14    Chapter 5 listed 35 chemicals that are reported to the FracFocus database in at least 10% of well
15    disclosures nationwide (U.S. EPA. 2015a) (Table 5-2). For 32 of these chemicals (water, quartz, and
16    sodium chloride were excluded from this analysis), only 7 chemicals (22%) have a federal chronic
17    RfV, as shown in Table 9-2. None of these 32 chemicals have available OSFs for cancer. For this
18    subset of chemicals, methanol was reported to be the most frequently used chemical in the
19    FracFocus analysis, followed by hydrotreated light petroleum distillates and hydrochloric acid, all
20    of which were reported in greater than 60% of disclosures. Ethylene glycol, isopropanol, and
21    peroxydisulfuric acid-diammonium salt are the only 3  additional chemicals to have been used in
22    greater than 40% of disclosures.
      Table 9-2. List of the most frequently used chemicals in hydraulic fracturing fluids, with their
                respective federal chronic RfVs where available.
                Chemicals are ordered in the table, from high to low, based on their frequency of use from FracFocus.
                Includes all chemicals reported to FracFocus in at least 10% of well disclosures, excluding water,
                quartz, and sodium chloride.
Chemical
Methanol
Distillates, petroleum, hydrotreated light
Hydrochloric acid
CASRN
67-56-1
64742-47-8
7647-01-0
RfV
Chronic RfD
(mg/kg-day)
2
-
-
Source
IRIS
-
-
                    This document is a draft for review purposes only and does not constitute Agency policy.
      June 2015                                     9-11                  DRAFT—DO NOT CITE OR QUOTE

-------
Hydraulic Fracturing Drinking Water Assessment
Chapter 9 - Identification and Hazard Evaluation of Chemicals across the Hydraul
Chemical
Ethylene glycol
Isopropanol
Peroxydisulfuric acid, diammonium salt
Guar gum
Sodium hydroxide
Propargyl alcohol
Glutaraldehyde
Ethanol
Potassium hydroxide
Acetic acid
Citric acid
2-Butoxyethanol
Solvent naphtha, petroleum, heavy arom.
Naphthalene
2,2-Dibromo-3-nitrilopropionamide
Choline chloride
Phenolic resin
Methenamine
Carbonic acid, dipotassium salt
1,2,4-Trimethylbenzene
Quaternary ammonium compounds, benzyl-C12-
16-alkyldimethyl, chlorides
Poly(oxy-l,2-ethanediyl)-nonylphenyl-hydroxy
(mixture)
Formic acid
Sodium chlorite
Nonyl phenol ethoxylate
Tetrakis(hydroxymethyl)phosphoniumsulfate
Polyethylene glycol
Ammonium chloride
Sodium persulfate
CASRN
107-21-1
67-63-0
7727-54-0
9000-30-0
1310-73-2
107-19-7
111-30-8
64-17-5
1310-58-3
64-19-7
77-92-9
111-76-2
64742-94-5
91-20-3
10222-01-2
67-48-1
9003-35-4
100-97-0
584-08-7
95-63-6
68424-85-1
127087-87-0
64-18-6
7758-19-2
9016-45-9
55566-30-8
25322-68-3
12125-02-9
7775-27-1
RfV
Chronic RfD
(mg/kg-day)
2
-
-
-
-
0.002
-
-
-
-
-
0.1
-
0.02
-
-
-
-
-
-
-
-
0.9
0.03
-
-
-
-
-
Source
IRIS

-
-
-
IRIS
-
-
-
-
-
IRIS
-
IRIS
-
-
-
-
-
-
-
-
PPRTV
IRIS
-
-
-
-
-
                This document is a draft for review purposes only and does not constitute Agency policy.
June 2015                                           9-12                      DRAFT—DO NOT CITE OR QUOTE

-------
     Hydraulic Fracturing Drinking Water Assessment
Chapter 9 - Identification and Hazard Evaluation of Chemicals across the Hydraul
     9.4.2.2. Most and Least Mobile Chemicals Used in Hydraulic Fracturing Fluid (FracFocus)
1    Chapter 5 lists the 20 most mobile chemicals (Table 5-7) and 20 least mobile chemicals (Table 5-8)
2    from the EPA's analysis of the FracFocus database (U.S. EPA. 2015a). For these lists, mobility was
3    determined based on Kow. For the 20 most mobile chemicals, no federal chronic RfVs or OSFs were
4    available (see Table 9-3). Similarly, for the 20 least mobile chemicals, only one chemical—di(2-
5    ethylhexyl) phthalate—had a federal chronic RfV available (see Table 9-4).
     Table 9-3. List of the 20 most mobile chemicals used in hydraulic fracturing fluid, with their
               respective federal chronic RfVs where available.
               Chemicals are ordered in the table by lowest estimated log Kow. None of these chemicals had federal
               chronic RfVs available.
Chemical
1,2-Ethanediaminium, N,N'-bis[2-[bis(2-
hydroxyethyl)methylammonio] ethyl] -N,N'-
bis(2-hydroxyethyl)-N,N'-dimethyl-,
tetrachloride
Phosphonic acid, [[(phosphonomethyl)imino]
bis[2,l-ethanediylnitrilobis
(methylene)]]tetrakis-
Phosphonic acid, [[(phosphonomethyl)imino]
bis[2,l-ethanediylnitrilobis
(methylene)]]tetrakis-, sodium salt
Phosphonic acid, [[(phosphonomethyl)imino]
bis[2,l-ethanediylnitrilobis
(methylene)]]tetrakis-, ammonium salt (l:x)
Phosphonic acid, (((2-[(2-
hydroxyethyl) (phosphono
methyl) a m i no) ethyl) imino] bis
(methylene))bis-, compel, with 2-aminoethanol
2-Hydroxy-N,N-bis(2-hydroxyethyl)-N-
methylethanaminium chloride
N- (3 -Chloroallyl) hexaminium chloride
3,5,7-Triazatricyclo(3.3.1.1(superscript
3,7))decane, l-(3-chloro-2-propenyl)-,
chloride, (Z)-
(2,3-dihydroxypropyl)trimethylammonium
chloride
Phosphonic acid, [[(phosphonomethyl) imino]
bis[6,l-
hexanediylnitrilobis(methylene)]]tetrakis-
CASRN
138879-94-4
15827-60-8
22042-96-2
70714-66-8
129828-36-0
7006-59-9
4080-31-3
51229-78-8
34004-36-9
34690-00-1
Log Kow
(unitless)
-23.19
-9.72
-9.72
-9.72
-6.73
-6.7
-5.92
-5.92
-5.8
-5.79
RfV
Chronic RfD
(mg/kg-day)
-
-
-
-
-
-
-
-
-
-
Source
--
--
--
--
--
--
--
--
--
--
                   This document is a draft for review purposes only and does not constitute Agency policy.
     June 2015                                      9-13                   DRAFT—DO NOT CITE OR QUOTE

-------
Hydraulic Fracturing Drinking Water Assessment
Chapter 9 - Identification and Hazard Evaluation of Chemicals across the Hydraul
Chemical
[Nitrilotris(methylene)]tris-phosphonicacid
pentasodium salt
Aminotrimethylene phosphonic acid
Choline chloride
Choline bicarbonate
alpha-Lactose monohydrate
Lactose
Tetrakis(hydroxymethyl)phosphoniumsulfate
Disodium ethylenediaminediacetate
Nitrilotriacetamide
l,3,5-Triazine-l,3,5(2H,4H,6H)-triethanol
CASRN
2235-43-0
6419-19-8
67-48-1
78-73-9
5989-81-1
63-42-3
55566-30-8
38011-25-5
4862-18-4
4719-04-4
Log Kow
(unitless)
-5.45
-5.45
-5.16
-5.16
-5.12
-5.12
-5.03
-4.76
-4.75
-4.67
RfV
Chronic RfD
(mg/kg-day)
-
-
-
-
-
-
-
-
-
-
Source
--
--
--
--
--
--
--
--
--
--
Table 9-4. List of the 20 least mobile chemicals used in hydraulic fracturing fluid, with their
           respective federal chronic RfVs where available.
           Chemicals are ordered in the table by highest estimated log Kow.
Chemical
CASRN
Log Kow
(unitless)
RfV
Chronic RfD
(mg/kg-day)
Source
               This document is a draft for review purposes only and does not constitute Agency policy.
June 2015                                        9-14                     DRAFT—DO NOT CITE OR QUOTE

-------
     Hydraulic Fracturing Drinking Water Assessment
Chapter 9 - Identification and Hazard Evaluation of Chemicals across the Hydraul
Chemical
Sorbitan, tri-(9Z)-9-octadecenoate
Fatty acids, C18-unsatd., dimers
Sorbitan sesquioleate
Tributyltetradecylphosphonium
chloride
Sodium bis(tridecyl)
sulfobutanedioate
1-Eicosene
D&C Red 28
C.I. Solvent Red 26
1-Octadecene
Alkenes, C>10 .alpha.-
Dioctyl phthalate
Benzene, C10-16-alkyl derivs.
Di(2-ethylhexyl) phthalate
1-Octadecanamine, N,N-dimethyl-
N,N-dimethyloctadecylamine
hydrochloride
Butyryl trihexyl citrate
1-Hexadecene
Benzo (g,h,i) perylene
Dodecylbenzene
Isopropanolamine dodecylbenzene
CASRN
26266-58-0
61788-89-4
8007-43-0
81741-28-8
2673-22-5
3452-07-1
18472-87-2
4477-79-6
112-88-9
64743-02-8
117-84-0
68648-87-3
117-81-7
124-28-7
1613-17-8
82469-79-2
629-73-2
191-24-2
123-01-3
42504-46-1
Log Kow
(unitless)
22.56
14.6
14.32
11.22
11.15
10.03
9.62
9.27
9.04
8.55
8.54
8.43
8.39
8.39
8.39
8.21
8.06
7.98
7.94
7.94
RfV
Chronic RfD
(mg/kg-day)
-
-
-
-
-
-
-
-
-
-
-
-
0.02
-
-
-
-
-
-
-
Source
-
-
-
-
-
-
-
-
-
-
-
-
IRIS
-
-
-
-
-
-
-
     9.4.2.3.  Flowback and Produced Water: Inorganics and NORM
1    In addition to a number of volatile and semi-volatile organic compounds presented below, Chapter
2    7 also discusses the appearance of inorganic constituents such as metals, inorganic ions, and
3    naturally occurring radioactive material (NORM) in flowback and produced water. A number of
4    metals detected in flowback and produced water that appear on the EPA's consolidated list and are
5    noted in Chapter 7 have federal RfVs and/or OSFs listed in Appendix G (Table G-2). These metals
6    and inorganic ions include:  iron, boron, chromium, zinc, arsenic, manganese, cadmium, and
7    strontium. These metals have oral RfVs based on a number of health effects including:
8    neurotoxicity, developmental and liver toxicity, hyperpigmentation and keratosis of the skin, and
9    decrements in blood copper status and enzyme activity. Chromium (VI) is classified as a known
                  This document is a draft for review purposes only and does not constitute Agency policy.
     June 2015                                    9-15                   DRAFT—DO NOT CITE OR QUOTE

-------
      Hydraulic Fracturing Drinking Water Assessment         Chapter 9 - Identification and Hazard Evaluation of Chemicals across the Hydraul


 1    human carcinogen by IARC and NTP, while arsenic is classified as known human carcinogen by the
 2    EPA, IARC, and NTP. Radionuclides, such as radium-226, radium-228, and uranium-238, which are
 3    naturally occurring in the formation may also return to the surface within produced water. Each of
 4    these radionuclides is classified as a known human carcinogen by the EPA and IARC.

      9.4.2.4. Flowback and Produced Water: Pesticides
 5    Lastly, it should be noted that a number of pesticides appear in the tables presented in Appendix G.
 6    These chemicals were reported as having been detected in analyses of hydraulic fracturing
 7    flowback and produced waters by several  of the 10 sources cited in Appendix A; however, there is
 8    much uncertainty about why they were detected. They could have migrated to the shale formation
 9    or to the rock surrounding the shale formation, or they could have migrated into source waters
10    used by the hydraulic fracturing operation. It is also possible that these are laboratory
11    contaminants.

      9.5. Hazard Identification and Hazard Evaluation of Selected Subsets of
           Hydraulic Fracturing Chemicals
12    As described in Section 9.4, the majority of chemicals identified in the previous chapters of this
13    report do not have RfVs and/or OSFs from the sources meeting the criteria described in Section
14    9.3.1. This lack of data creates a challenge  for hazard evaluation, because the potential human
15    health effects of these chemicals are difficult to determine. On the other hand, other chemicals
16    identified by the EPA have more data available, including chronic RfVs, data on occurrence, and
17    data on physicochemical properties. This section focuses on the hazard evaluation of these subsets
18    of chemicals that had data available.

19    When considering the hazard evaluation of chemicals in drinking water, it is important to
2 0    remember that toxicity is contingent upon exposure. All chemicals, including pure water, may be
21    toxic if they are ingested in large enough quantities. Therefore, in addition to data on health effects,
2 2    hazard evaluations must also consider data on potential chemical exposure. In the context of the
2 3    hazard evaluation presented in this section, chemical occurrence and physicochemical property
24    data were used as metrics to estimate the likelihood that the chemical could reach and impact
2 5    drinking water resources.

2 6    For the selected subsets of chemicals that had data available, this section discusses the known
27    toxicological properties based on selected RfVs (hazard identification), and then illustrates one
28    possible method for combining toxicity and exposure potential information for a more data-
29    informed hazard evaluation. Additionally,  this section presents a summary of chemicals that have
30    occurrence data across multiple stages of the hydraulic fracturing water cycle.

      9.5.1.  Selection of Chemicals for Hazard Evaluation
31    From the overall list of 1,173 chemicals identified in this assessment, subsets of chemicals were
32    selected for hazard evaluation if they met the following criteria:

33        1.  Had a federal chronic  oral RfV;
                   This document is a draft for review purposes only and does not constitute Agency policy.
      June 2015                                    9-16                  DRAFT—DO NOT CITE OR QUOTE

-------
      Hydraulic Fracturing Drinking Water Assessment          Chapter 9 - Identification and Hazard Evaluation of Chemicals across the Hydraul


 1         2.  Had available data on frequency of use (in hydraulic fracturing fluids) or measured
 2            concentrations (in flowback and produced water); and
 3         3.  Had available data on physicochemical properties.
 4    These criteria were selected for hazard evaluation for the following reasons:

 5         1.  Federal RfVs generally undergo more extensive independent peer review compared to
 6            other sources of RfVs. Additionally, as described above, there are many more chemicals
 7            with federal chronic RfVs than chemicals with federal OSFs. Therefore, although OSFs are
 8            discussed in the hazard evaluation, chronic RfVs were selected for illustrative purposes of
 9            making comparisons between chemicals.
10         2.  Data on frequency of use (in hydraulic fracturing fluids) or measured concentration (in
11            flowback or produced water) provide a metric to help assess the likelihood of chemical
12            occurrence in the hydraulic fracturing water cycle. Chemicals that are used more
13            frequently in hydraulic fracturing fluid have a greater likelihood of accidental release or
14            dissemination due to the fact that they are present at a greater number of we 11s
15            nationwide. Likewise, chemicals that occur at higher concentrations in flowback or
16            produced water may result in greater exposures. Frequency of detection in flowback or
17            produced water would also be a useful metric for this evaluation, but this information was
18            not available for these chemicals.
19         3.  Information on physicochemical properties enables the estimation of chemical persistence
20            and mobility in the environment This is discussed in more detail in Section 9.5.2 below.
21    For chemicals that are used in hydraulic fracturing fluids, the FracFocus database was the only
22    source with reliable information on the frequency of use (U.S. EPA. 2015a). For chemicals found in
23    flowback or produced water, data on measured concentration were only available for the 75
24    chemicals presented in Appendix E. Therefore, hazard evaluations were only conducted on
25    chemicals included in these two data sources. While the other data sources listed in Appendix A
26    provide useful information on the diversity of chemicals that may occur in the hydraulic fracturing
27    water cycle, hazard evaluation could not be conducted on these sources in the absence of data on
28    frequency of use or measured concentration.

29    Overall, 37 chemicals used in hydraulic fracturing fluid and 23 chemicals detected in flowback and
30    produced water met the selection criteria for hazard evaluation (see Figure 9-3).
                   This document is a draft for review purposes only and does not constitute Agency policy.
      June 2015                                    9-17                   DRAFT—DO NOT CITE OR QUOTE

-------
Hydraulic Fracturing Drinking Water Assessment
                                                          Chapter 9 - Identification and Hazard Evaluation of Chemicals
                                                                        Across the Hydraulic Fracturing Water Cycle
  1,076 chemicals used in hydraulic fracturing fluid:
                                                  134 chemicals detected in flowback and produced water:
                         73
                  chemicals with federal RfV
                   (oral, chronic duration
     chemicals with
     physicochemical
      property data
  chemicals with
Frequency of use data
  (from FracFocus)
                      Chemicals selected for
                         hazard evaluation
                    70
            chemicals with federal RfV
              oral, chronic duration
chemicals with
physicochemica
 property data
Figure 9-3. The two subsets of chemicals selected for hazard evaluation included 37 chemicals used in hydraulic fracturing fluid,
           and 23 chemicals detected in flowback or produced water.
June 2015
 This document is a draft for review purposes only and does not constitute Agency policy.
                                   9-18
                      DRAFT—DO NOT CITE OR QUOTE

-------
      Hydraulic Fracturing Drinking Water Assessment                                              Chapter 9 - Identification and Ht
                                                                                                  Across the Hyc

      9.5.2.  Multi Criteria Decision Analysis (MCDA) Framework for Hazard Evaluation: Integrating
             Toxicity, Occurrence, and Physicochemical Data
 1    Integration or combining of various types of data may provide insights on those chemicals that may
 2    be of greater concern than other chemicals to drinking water resources. For the purpose of this
 3    chapter, a structured but flexible Multi Criteria Decision Analysis (MCDA) approach was developed
 4    to integrate factors related to hydraulic fracturing such as chemical toxicity, occurrence, and
 5    physicochemical data. The approach described here is for illustrative purposes only, in order to
 6    demonstrate how combining of information may be informative. Alternative frameworks may be
 7    considered by risk assessors for similar analyses.

 8    In this illustration, a MCDA framework was developed and applied to each list of chemicals
 9    identified in Section 9.5.1 and depicted in Figure 9-3 (37 chemicals used in hydraulic fracturing
10    fluids, and 23 chemicals detected in flowback or produced water). The MCDA framework serves to
11    place the toxicity of these chemicals in the context of factors that may increase the likelihood of
12    impacting drinking water resources. In essence, this analysis serves to illustrate the circumstances
13    under which drinking water resources may be affected.

14    MCDA is a well-established analysis tool that is used to transparently integrate multiple lines of
15    evidence to support decision-making.  For example, MCDA has been adapted as a method of
16    selecting an optimal cleanup plan for a contaminated site (Linkov et al.. 2 011), and as a method of
17    integrating chemical hazard data across multiple studies  (Hristozovetal., 2014). The MCDA
18    framework employed  here is based on the method by Mitchell etal. (2013b). who developed a
19    protocol for ranking chemical exposure potential by integrating data on physicochemical properties
20    and commercial use. This method is similar to approaches used by the petroleum industry to
21    quantitatively rank the potential hazards of hydraulic fracturing chemicals (see Section 5.9).
22    Moreover, the underlying philosophy of this approach is similar to that of the EPA's Design for the
23    Environment (DfE) Program. The DfE's Alternatives Assessment Criteria for Hazard Evaluation (U.S.
24    EPA. 2011a) was developed as a tool for evaluating and differentiating among chemical hazards
25    based on toxicity and physicochemical properties. Recently, this criteria and framework have been
26    applied in the Alternatives Assessment for the Flame Retardant Decabromodiphenyl Ether
27    (DecaBDE) and Flame Retardant Alternatives for Hexabromocyclododecane (HBCD) (U.S. EPA,
28    2014a. d). Aspects of MCDA methods and the DfE's Program for Alternatives Assessment are
29    evident in the National Research Council (NRC)'s "A Framework to Guide Selection of Chemical
30    Alternatives" document (NRC. 2014).

31    The methodology used to illustrate a hazard evaluation MCDA for hydraulic fracturing is outlined
32    below, and schematic  of the model is shown in Figure 9-4. Under the MCDA framework, each
33    chemical was assigned three scores:

34         1. A toxicity score;
35         2. An occurrence score; and
36         3. A physicochemical properties score.
                   This document is a draft for review purposes only and does not constitute Agency policy.
      June 2015                                   9-19                  DRAFT—DO NOT CITE OR QUOTE

-------
      Hydraulic Fracturing Drinking Water Assessment
                 Chapter 9 - Identification and Ht
                             Across the Hyc
 1    The three normalized scores were summed to develop a total composite hazard potential score for
 2    each chemical. These scores serve as a relative ranking and a means of making comparisons across
 3    chemicals. These scores are not intended to define whether or not a chemical will present a human
 4    health hazard, or indicate that one chemical is safer than another. Rather, the scores serve as a
 5    qualitative metric to identify chemicals that may be more likely to present an impact to drinking
 6    water resources, given available data on chemical properties and occurrence.
                                 Total Hazard Potential Score
                                                                   Physicochemical
                                                                      Properties
                                                       Mobility
Volatility
Persistence
      Figure 9-4. Overview of the MCDA framework applied to the hazard evaluations.
      9.5.2.1. Toxicity Score
 7    The toxicity score was based upon the federal chronic RfV, which was determined from peer
 8    reviewed sources as described in Section 9.3.1. Within each dataset (chemicals used in hydraulic
 9    fracturing fluids, or chemicals detected in flowback and produced water), toxicity was ranked based
10    on quartiles, with each chemical assigned a toxicity score of 1 to 4 (see thresholds outlined in Table
11    9-5). Note that chemicals in the lowest quartile received the highest toxicity score as these
12    chemicals have lower RfVs than for other chemicals.

      9.5.2.2. Occurrence Score
13    This score was based on the frequency or concentration at which chemicals were reported within
14    the hydraulic fracturing water cycle. For chemicals used in hydraulic fracturing fluids, the
15    occurrence score was based on the nationwide number of well disclosures for each chemical from
16    the FracFocus database. For chemicals that were detected in hydraulic fracturing flowback and
17    produced water, the occurrence score was based on the average or median measured concentration
18    reported in Appendix E. If the measured concentration of a chemical was reported by multiple
                   This document is a draft for review purposes only and does not constitute Agency policy.
      June 2015                                    9-20                   DRAFT—DO NOT CITE OR QUOTE

-------
      Hydraulic Fracturing Drinking Water Assessment                                               Chapter 9 - Identification and Ht
                                                                                                    Across the Hyc

 1    studies in Appendix E, the highest of these reported average or median concentrations was used for
 2    this calculation. Note that these two metrics of chemical occurrence—frequency of use, and
 3    concentration—cannot be directly compared to one another. Therefore, FracFocus chemicals and
 4    flowback and produced water chemicals were considered separately for this MCDA hazard
 5    evaluation. Within each dataset (chemicals used in hydraulic fracturing fluids, or chemicals
 6    detected in flowback and produced water), chemical occurrence was ranked based on quartiles,
 7    with each chemical assigned an occurrence score of 1 to 4, as shown in Table 9-5.

      9.5.2.3. Physicochemical Properties Score
 8    This score was based upon inherent physicochemical properties which affect the likelihood that a
 9    chemical will reach and impact drinking water resources. The thresholds chosen for ranking
10    physicochemical properties, shown in Table 9-5, are based on previously published thresholds used
11    in the DfE Alternatives Assessment Criteria for Hazard Evaluation [U.S. EPA. 2011a], the EPA Office
12    of Pollution Prevention and Toxics Pollution Prevention (P2) Framework [U.S. EPA. 2005], and
13    Mitchell et al. (2013b). When refining EPI Suite™ physicochemical properties data for input into
14    this MCDA, empirically measured values were always used when available. If multiple estimated
15    values were available, the most conservative value (i.e., the value resulting in the highest score
16    according to Table 9-5) was used.

17    The total physicochemical properties score for each chemical was based upon three subcriteria:
18    mobility in water, volatility, and persistence. Chemical mobility in water was assessed based upon
19    three physicochemical properties: the octanol-water partition coefficient (K0w), the organic carbon-
2 0    water partition coefficient (Koc], and aqueous solubility.  Chemical volatility was assessed based on
21    the Henry's law constant, which describes partitioning of a chemical between water and air.
2 2    Chemical persistence was assessed based on estimated half-life in water, which describes how long
23    a chemical will persist in water before it is transformed  or degraded. Details on the evaluation and
24    physicochemical score calculation are provided in the Chapter Annex, Section 9.8.1. For each
25    chemical, the mobility score, volatility score, and persistence score (each on a scale of 1 to 4) were
26    summed to calculate a total physicochemical score.

      9.5.2.4. Final MCDA Score Calculations
27    Each raw score (toxicity, occurrence, or physicochemical properties), calculated as described above,
2 8    was standardized by scaling to the highest and lowest raw score within the set of chemicals. The
29    following equation was used:

•3U                                   ^>x_final — (^x "" ^*minj / (^max "" ^*minj

31    in which Sx is the raw score for a particular chemical x, Smax is the highest observed raw score within
32    the set of chemicals, and Smin is the lowest observed raw score within the set of chemicals. Sx_fmai is
33    the standardized score for chemical x. Each standardized score (toxicity, occurrence, or
34    physicochemical properties) falls on a scale of 0 to 1. These standardized toxicity, occurrence, and
35    physicochemical properties scores were summed to calculate a total hazard potential score for each
36    chemical. The total hazard potential scores fell on a scale of 0 to 3, with higher scores indicating
                   This document is a draft for review purposes only and does not constitute Agency policy.
      June 2015                                    9-21                  DRAFT—DO NOT CITE OR QUOTE

-------
      Hydraulic Fracturing Drinking Water Assessment
Chapter 9 - Identification and Ht
             Across the Hyc
 1    chemicals that are predicted to be more likely to affect drinking water resources. An example of
 2    MCDA score calculation can be found in the Annex, Section 9.8.2.

 3    In the MCDA approach illustrated in this chapter, each factor (toxicity, occurrence, physicochemical
 4    properties) was given equal weight in the calculation of the final hazard potential score. This was
 5    done in order to prevent subjectivity and avoid biasing the results based on any individual variable
 6    that was considered in this analysis. This approach is adaptable, however. Risk assessors may
 7    choose to apply alternative weights that place more or less emphasis on the various factors being
 8    considered, in order to reflect expert judgement of a variable's relative importance. This MCDA
 9    approach may also be adapted to include other variables of interest, such as carcinogenic potential,
10    which were not considered in the MCDA approach illustrated in this chapter.
      Table 9-5. Thresholds used for developing the toxicity score, occurrence score, and
                physicochemical properties score in this MCDA framework.
Score
1
2
3
4
Toxicity Score
Chronic RfV
(federal)
>3rd quartile
>2nd quartile to
<3rd quartile
>lst quartile to
<2nd quartile
lst quartile to
<2nd quartile
>lst quartile to
<2nd quartile
>2nd quartile to
<3rd quartile
>2nd quartile to
<3rd quartile
>3rd quartile
>3rd quartile
Physicochemical Properties Score
Mobility score:
Log Kow
LogK0c
Aqueous solubility
(mg/L)
>5
>4.4
<0.1
>3to5
>3. 4 to 4.4
>0.1to<100
>2to3
>2.4to3.4
> 100 to < 1000
<2
<2.4
>1000
Volatility score:
Henry's law
constant
>10"1
>10"3to<10"1
>10"5to<10"3
<10"5
Persistence score:
Half-life in water
(days)
<16
>16 to <60
>60 to <180
>180
                    This document is a draft for review purposes only and does not constitute Agency policy.
      June 2015                                     9-22                  DRAFT—DO NOT CITE OR QUOTE

-------
      Hydraulic Fracturing Drinking Water Assessment                                                  Chapter 9 - Identification and Ht
                                                                                                         Across the Hyc
      9.5.3.  Hazard Evaluation Results
 1    Discussed below are the results of the hazard evaluations for each subset of chemicals identified in
 2    Section 9.5.1. For each subset of chemicals selected for hazard evaluation, the information
 3    presented includes: the available federal chronic oral RfV (hazard identification), followed by
 4    highlights of MCDA analyses (hazard evaluation).

 5    For this MCDA illustration, the calculated toxicity scores, occurrence scores, physicochemical
 6    properties scores, and total hazard potential scores are provided for chemicals used in hydraulic
 7    fracturing fluids and chemicals detected in flowback/produced water, respectively. These
 8    individual scores make it possible to visualize the extent to which the total hazard potential ranking
 9    of each chemical is driven by each of the variables considered in the MCDA.

      9.5.3.1.  Hazard Identification: Chemical Used in Hydraulic Fracturing Fluid
10    As discussed above, a total of 37 chemicals used in hydraulic fracturing fluids were identified for
11    hazard evaluation using the selection criteria described in Section 9.5.1. Some of the chemicals
12    represented include the BTEX chemicals (benzene, toluene, ethylbenzene, xylenes) as well as
13    naphthalene, acrylamide, phenol, 1,2-propylene glycol, ethylene  glycol, 2-butoxyethanol, ethyl
14    acetate, and methanol.

15    These chemicals along with their primary noncancer toxicological properties, including the point-
16    of-departure (POD), total product of uncertainty factors applied, the  federal chronic RfV, and the
17    health effect basis for the RfV, are shown  in Table 9-6.1/2 As seen in Table 9-6, all of these chemicals
18    had RfDs available from IRIS, PPRTV, or HHBP. These chemicals induce a variety of adverse
19    outcomes including immune system effects, changes in body weight,  changes in blood chemistry,
20    cardiotoxicity, neurotoxicity, liver and kidney toxicity, and reproductive and developmental
21    toxicity. The RfD values within this suite of chemicals range from 0.001-20 mg/kg-day, with (E)-
22    crotonaldehyde having the lowest RfD (0.001 mg/kg-day) and 1,2-propylene glycol having the
23    highest (20  mg/kg-day).

24    Comparison of RfVs among a set of chemicals requires a more thorough examination. For instance,
25    to derive the final chronic RfD for a given chemical, a number of UFs  may be applied to the POD.
26    Briefly, UFs are applied to account for 5 areas of uncertainty: 1) intraspecies variability; 2)
      1 The point-of-departure (POD) is the dose-response point that marks the beginning of a low-dose extrapolation. This
      point can be the lower bound on dose for an estimated incidence or a change in response level from a dose-response
      model or a NOAEL or LOAEL for an observed incidence, or change in level of response. See http://www.epa.gov/iris/ for
      more information.
      2 An uncertainty factor is one of several (generally 10-fold) default factors used in operationally deriving the RfV from
      experimental data. The factors are intended to account for (1) variation in susceptibility among the members of the
      human population (i.e., inter-individual or intraspecies variability); (2) uncertainty in extrapolating animal data to
      humans (i.e., interspecies uncertainty); (3) uncertainty in extrapolating from data obtained in a study with less-than-
      lifetime exposure (i.e., extrapolating from subchronic to chronic exposure); (4) uncertainty in extrapolating from a LOAEL
      rather than from a NOAEL; and (5) uncertainty associated with extrapolation when the database is incomplete. See the
      IRIS Glossary at: http://www.epa.gov/iris/ for more information.
                    This document is a draft for review purposes only and does not constitute Agency policy.
      June 2015                                      9-23                   DRAFT—DO NOT CITE OR QUOTE

-------
      Hydraulic Fracturing Drinking Water Assessment
Chapter 9 - Identification and Ht
             Across the Hyc
 1    interspecies uncertainty; 3) extrapolation from a subchronic study; 4) extrapolating from a NOAEL;
 2    and 5) an incomplete database. A UF of 1, 3 (10°-5), or 10 can be applied for any of these areas of
 3    uncertainty depending upon the amount and/or type data available. The maximum total UF that
 4    can be applied is 3,000; RfDs are not derived for chemicals that invoke the application of a total UF
 5    >3,000 or involves the application of the full 10-fold UF in four or more areas of uncertainty [U.S.
 6    EPA. 2002a). Therefore, those chemicals with a lower total uncertainty factor generally have more
 7    reliable and robust health effect information. For example, although (E)-crotonaldehyde has the
 8    lowest RfD, chemicals such as acrylamide, benzene, and dichloromethane have RfDs within a factor
 9    of 10 (0.002-0.006 mg/kg-day) but with much less uncertainty reflected in their values. All three
10    latter chemicals have large data sets with reproducible effects, and dose estimated based on
11    physiologically based pharmacokinetic models (for acrylamide and dichloromethane) or have
12    available human health effect data (for benzene). Thus, a chemical with a low RfD may reflect high
13    uncertainty in the value and not necessarily be the most toxic.

14    Although only federal RfVs are considered in this hazard evaluation, eight of these chemicals also
15    have federal OSFs. These include acrylamide, benzyl chloride, 1,4-dioxane, 1,3-dichloropropene,
16    benzene, epichlorohydrin, aniline, and dichloromethane. Of these chemicals, acrylamide is the most
17    potent carcinogen. Acrylamide has an OSF of 0.5 per mg/kg-day and is classified as a likely human
18    carcinogen in IRIS (U.S. EPA. 2010).  Benzene is the only chemical listed as a known human
19    carcinogen and has a calculated OSF of 0.015 mg/kg-day (U.S. EPA, 2002b). The OSF values for each
20    of these chemicals can be found in Appendix G.
      Table 9-6. Toxicological properties of the 37 chemicals used in hydraulic fracturing fluid that
                were identified for hazard evaluation and MCDA analysis.
                Chemicals are ranked, from low to high, based on their respective federal chronic RfVs.
Chemical
(E) -Crotonaldehyde
Benzyl chloride
Propargyl alcohol
Acrylamide
Benzene
Epichlorohydrin
CASRN
123-73-9
100-44-7
107-19-7
79-06-1
71-43-2
106-89-8
Point of
departure
(mg/kg-
day)
3.4
6.4
5
0.053
1.2
6.25
Total
uncertain
ty factor
3000
3000
3000
30
300
1000
RfV
Chronic
RfD
(mg/kg-
day)
0.001
0.002
0.002
0.002
0.004
0.006
Noncancer effect
Forestomach lesions
Cardiotoxicity
Renal and
hepatotoxicity
Degenerative nerve
changes
Decreased
lymphocyte count in
humans
Decreased fertility
Source
PPRTV
PPRTV
IRIS
IRIS
IRIS
PPRTV
                   This document is a draft for review purposes only and does not constitute Agency policy.
      June 2015                                    9-24                  DRAFT—DO NOT CITE OR QUOTE

-------
Hydraulic Fracturing Drinking Water Assessment
Chapter 9 - Identification and Ht
               Across the Hyc
Chemical
Dichloromethane
Aniline
2-(Thiocyano
methylthio)benzo
thiazole
Furfural
Naphthalene
2 - (2 -Butoxyethoxy)
ethanol
1,4-Dioxane
Bisphenol A
1,3-
Dichloropropene
Toluene
Ethylenediamine
Ethylbenzene
2-Butoxyethanol
(EGBE)
Acetophenone
Didecyldimethyl
ammonium
chloride
CASRN
75-09-2
62-53-3
21564-17-
0
98-01-1
91-20-3
112-34-5
123-91-1
80-05-7
542-75-6
108-88-3
107-15-3
100-41-4
111-76-2
98-86-2
7173-51-5
Point of
departure
(mg/kg-
day)
0.19
7
3.8
30
71
81
9.6
50
3.4
238
9
97.1
1.4
423
10
Total
uncertain
ty factor
30
1000
300
3000
3000
3000
300
1000
100
3000
100
1000
10
3000
100
RfV
Chronic
RfD
(mg/kg-
day)
0.006
0.007
0.01
0.01
0.02
0.03
0.03
0.05
0.03
0.08
0.09
0.1
0.1
0.1
0.1
Noncancer effect
Hepatic effects
Splenic effects
Decreased body
weight gain;
decreased white blood
cells (WBC) and
plasma alanine
aminotransferase
(ALT)
Liver pathology
Decreased mean
terminal body weight
> 10%
Changes in red blood
cells (RBC)
Liver and kidney
toxicity
Reduced mean body
weight
Chronic irritation
Increased absolute
kidney weight
Liver and kidney
toxicity
Liver and kidney
toxicity;
histopathology
Hemosiderin
deposition in liver
(inhalation study)
General toxicity; NO
LOAEL identified
Clinical signs;
decreased total
cholesterol levels
Source
IRIS
PPRTV
HHBP
HHBP
IRIS
PPRTV
IRIS
IRIS
IRIS
IRIS
PPRTV
IRIS
IRIS
IRIS
HHBP
                This document is a draft for review purposes only and does not constitute Agency policy.
June 2015                                           9-25                      DRAFT—DO NOT CITE OR QUOTE

-------
Hydraulic Fracturing Drinking Water Assessment
Chapter 9 - Identification and Ht
               Across the Hyc
Chemical
Cumene
N,N-Dimethylform
amide
1-Butanol
Xylenes
Formaldehyde
Phenol
2-Methyl-l-
propanol
(Isobutanol)
Acetone
Ethyl acetate
Formic acid
Dodecylbenzenesul
fonic acid
Ethylene glycol
Hexanedioic acid
Methanol
Benzole acid
1,2-Propylene
glycol
CASRN
98-82-8
68-12-2
71-36-3
1330-20-7
50-00-0
108-95-2
78-83-1
67-64-1
141-78-6
64-18-6
27176-87-
0
107-21-1
124-04-9
67-56-1
65-85-0
57-55-6
Point of
departure
(mg/kg-
day)
110
96
125
179
15
93
316
900
900
277
50
200
470
43.1 mg/La
4.4
5200
Total
uncertain
ty factor
1000
1000
1000
1000
100
300
1000
1000
1000
300
100
100
300
100
1
300
RfV
Chronic
RfD
(mg/kg-
day)
0.1
0.1
0.1
0.2
0.2
0.3
0.3
0.9
0.9
0.9
0.5
2
2
2
4
20
Noncancer effect
Increased average
kidney weight in
female rats
Increase in ALT and
liver weight
Hypoactivity and
ataxia
Decreased body
weight; increased
mortality
Decreased weight gain
Decreased maternal
weight gain;
developmental
toxicity
Hypoactivity and
ataxia
Nephropathy
Mortality and body
weight loss
Reproductive effects
Decreased pup
weight; kidney
pathology
Kidney toxicity;
chronic nephritis
Decreased body
weight
Extra cervical ribs;
developmental
toxicity
No adverse effects
observed in humans
Reduced red blood
cell counts and
hyperglycemia
Source
IRIS
PPRTV
IRIS
IRIS
IRIS
IRIS
IRIS
IRIS
IRIS
PPRTV
HHBP
IRIS
PPRTV
IRIS
IRIS
PPRTV
                This document is a draft for review purposes only and does not constitute Agency policy.
June 2015                                           9-26                      DRAFT—DO NOT CITE OR QUOTE

-------
     Hydraulic Fracturing Drinking Water Assessment
Chapter 9 - Identification and Ht
              Across the Hyc
Chemical
CASRN
Point of
departure
(mg/kg-
day)
Total
uncertain
ty factor
RfV
Chronic
RfD
(mg/kg-
day)
Noncancer effect
Source
      1 POD based on internal methanol blood concentration using a PBPK model.
     9.5.3.2.  MCDA Results: Chemical Used in Hydraulic Fracturing Fluid

1    The hazard potential scores of the selected 37 chemicals used in hydraulic fracturing fluid are
2    presented in Table 9-7.
     Table 9-7. MCDA results for 37 chemicals used in hydraulic fracturing fluid.

                Chemicals are ranked, from high to low, based on total hazard potential score. See section 9.5.2 for
                details on the calculation.
Chemical
Propargyl alcohol
2-Butoxyethanol (EGBE)
N, N- Dimethylformamide
Acrylamide
Formaldehyde
Naphthalene
Benzyl chloride
1-Butanol
Epichlorohydrin
2 - (2 - B utoxyethoxy) ethanol
Methanol
Ethylene glycol
Formic acid
Didecyldimethylammoniu
m chloride
1,4-Dioxane
(E)-Crotonaldehyde
Aniline
Furfural
CASRN
107-19-7
111-76-2
68-12-2
79-06-1
50-00-0
91-20-3
100-44-7
71-36-3
106-89-8
112-34-5
67-56-1
107-21-1
64-18-6
7173-51-5
123-91-1
123-73-9
62-53-3
98-01-1
Physicochemical
properties score
1.00
1.00
1.00
1.00
1.00
0.67
0.67
1.00
0.67
1.00
1.00
1.00
1.00
0.33
1.00
0.67
1.00
1.00
Occurrence
score
1.00
1.00
1.00
0.67
1.00
1.00
0.67
0.67
0.67
0.67
1.00
1.00
1.00
1.00
0.33
0.33
0.00
0.00
Toxicity
score
1.00
0.67
0.67
1.00
0.33
0.67
1.00
0.67
1.00
0.67
0.00
0.00
0.00
0.67
0.67
1.00
1.00
1.00
Total hazard
potential score
3.00
2.67
2.67
2.67
2.33
2.33
2.33
2.33
2.33
2.33
2.00
2.00
2.00
2.00
2.00
2.00
2.00
2.00
                    This document is a draft for review purposes only and does not constitute Agency policy.

     June 2015                                       9-27                    DRAFT—DO NOT CITE OR QUOTE

-------
      Hydraulic Fracturing Drinking Water Assessment
Chapter 9 - Identification and Ht
             Across the Hyc
Chemical
1,2-Propylene glycol
Hexanedioic acid
Toluene
Phenol
2-Methyl-l-propanol
(Isobutanol)
Dichloromethane
Ethylenediamine
Bisphenol A
2 - (Thiocyanomethylthio)
benzothiazole
Benzene
Dodecylbenzenesulfonic
acid
Xylenes
Ethylbenzene
Benzoic acid
Acetophenone
1,3-Dichloropropene
Cumene
Ethyl acetate
Acetone
CASRN
57-55-6
124-04-9
108-88-3
108-95-2
78-83-1
75-09-2
107-15-3
80-05-7
21564-17-0
71-43-2
27176-87-0
1330-20-7
100-41-4
65-85-0
98-86-2
542-75-6
98-82-8
141-78-6
67-64-1
Physicochemical
properties score
1.00
1.00
0.33
1.00
1.00
0.67
1.00
1.00
0.67
0.67
0.33
0.33
0.33
1.00
0.67
0.67
0.00
0.67
0.67
Occurrence
score
0.67
0.67
0.67
0.33
0.33
0.00
0.00
0.00
0.00
0.00
1.00
0.67
0.33
0.33
0.00
0.00
0.33
0.33
0.33
Toxicity
score
0.00
0.00
0.67
0.33
0.33
1.00
0.67
0.67
1.00
1.00
0.00
0.33
0.67
0.00
0.67
0.67
0.67
0.00
0.00
Total hazard
potential score
1.67
1.67
1.67
1.67
1.67
1.67
1.67
1.67
1.67
1.67
1.33
1.33
1.33
1.33
1.33
1.33
1.00
1.00
1.00
 1    Of the chemicals in hydraulic fracturing fluid that were considered in this hazard evaluation,
 2    propargyl alcohol received the highest overall hazard potential score. Propargyl alcohol was used in
 3    33% of wells in the FracFocus database, making it one of the most widely used chemicals that was
 4    considered in this analysis, and it also had one of the lowest RfVs, with an RfD of 0.002 mg/kg-day.
 5    It is also hydrophilic and has relatively low volatility, indicating that it is likely to be readily
 6    transported in water. Given these properties, propargyl alcohol received the highest overall ranking
 7    across all of the metrics that were considered in the hazard evaluation.

 8    The other chemicals that fell in the upper quartile in terms of frequency of use received lower
 9    hazard potential scores relative to propargyl alcohol, due to lower estimated toxicity and/or
10    physicochemical properties that are less conducive to transport in water. Naphthalene, used in 19%
11    of wells on the FracFocus database, has an RfD of 0.02 mg/kg-day, and is expected to have
12    somewhat lower transport in water relative to other chemicals because it is moderately
13    hydrophobic and moderately volatile. Methanol (RfD of 2 mg/kg-day), ethylene glycol (RfD of 2
                    This document is a draft for review purposes only and does not constitute Agency policy.
      June 2015                                     9-28                   DRAFT—DO NOT CITE OR QUOTE

-------
      Hydraulic Fracturing Drinking Water Assessment                                               Chapter 9 - Identification and Ht
                                                                                                   Across the Hyc

 1    mg/kg-day), 2-butoxyethanol (RfD of 0.1 mg/kg-day), formic acid (RfD of 0.9 mg/kg-day), N,N-
 2    dimethylformamide (RfD of 0.1 mg/kg-day), and formaldehyde (RfD of 0.2 mg/kg-day)—which
 3    were used in 73%, 47%, 23%, 11%, 9%, and 7% of wells in the FracFocus database, respectively—
 4    are all expected to be highly mobile in water and have low volatility, but have higher RfVs
 5    compared to many of the other chemicals in the assessment. Didecyldimethylammonium chloride
 6    (RfD of 0.1 mg/kg-day), used in 8% of wells, is expected to have reduced mobility in water due to its
 7    more hydrophobic properties.

 8    In addition to propargyl alcohol, the other most toxic chemicals (occurring in the lowest quartile of
 9    RfVs) received moderate to high hazard potential scores overall. Acrylamide (RfD of 0.002 mg/kg-
10    day) is used in only 1% of wells, but has physicochemical properties that are very conducive to
11    transport in water, and therefore received one of the highest overall hazard potential scores. Benzyl
12    chloride (RfD of 0.002 mg/kg-day) and epichlorohydrin (RfD of 0.006 mg/kg-day) are used in 6%
13    and 1% of wells, respectively, but scored slightly lower than acrylamide with regards to their
14    physicochemical properties. Other chemicals, including (E)-crotonaldehyde (RfD of 0.001 mg/kg-
15    day), benzene (RfD of 0.004 mg/kg-day), dichloromethane (RfD of 0.006 mg/kg-day), aniline (RfD
16    of 0.007 mg/kg-day), furfural (RfD of 0.01 mg/kg-day), and 2-(Thiocyanomethylthio)benzothiazole
17    (RfD of 0.01 mg/kg-day), received lower overall scores because they are used more infrequently
18    (each in less than 0.1% of wells in the FracFocus database).

      9.5.3.3. Hazard Identification: Chemicals Detected in Flowback and Produced Water
19    As discussed above, a total of 23 chemicals detected in flowback and produced water were
20    identified for hazard evaluation using the selection criteria described in Section 9.5.1. Of these 23
21    chemicals, 10 chemicals overlap with the hazard evaluation of chemicals used in hydraulic
22    fracturing fluids. Because of this overlap, many of the effects noted in each hazard evaluation are
23    similar.

24    These chemicals, along with their POD, total products of uncertainty factors applied, federal chronic
25    RfVs, and the health effect bases for the RfVs, are shown in Table 9-8. As seen in Table 9-8, all of
26    these chemicals had RfDs available from IRIS, PPRTV, or HHBP. These chemicals induce a variety of
27    adverse outcomes, including immune system effects, changes in body weight, changes in blood
28    chemistry, pulmonary toxicity, neurotoxicity, liver and kidney toxicity, and reproductive and
29    developmental toxicity. The RfD values within this suite of chemicals range from 0.001-0.9  mg/kg-
30    day, with pyridine having the lowest RfD and acetone having the highest RfD. For this subset of
31    chemicals, 88% have an RfD within 2 orders of magnitude of each other and 78% have RfDs within
32    a factor of 10 (range of 0.01-0.1 mg/kg-day). Some of these chemicals include chloroform,
33    naphthalene, 1,4-dioxane, toluene, cumene, and ethylbenzene.

34    Although only federal RfVs are considered in this hazard evaluation, 2 of these chemicals—benzene
35    and 1-4-dioxane—also have federal OSFs . These chemicals are also included in the hazard
36    evaluation of chemicals used in hydraulic fracturing fluids, discussed above. 1,4-dioxane is a more
37    potent carcinogen compared to benzene. The OSF for 1,4-dioxane is 0.1 per mg/kg-day and  is
38    classified as likely to be a human carcinogen by  IRIS (U.S. EPA. 2013fj.
                   This document is a draft for review purposes only and does not constitute Agency policy.
      June 2015                                    9-29                  DRAFT—DO NOT CITE OR QUOTE

-------
Hydraulic Fracturing Drinking Water Assessment
Chapter 9 - Identification and Ht
              Across the Hyc
Table 9-8. Toxicological properties of the 23 chemicals detected in flowback and produced
           water that were identified for hazard evaluation and MCDA analysis.

           Chemicals are ranked, from low to high, based on their respective federal chronic RfVs. Chemicals in
           italics are also included in the hazard evaluation of chemicals used in hydraulic fracturing fluids.
Chemical
Pyridine
2-
Methylnaphthalene
Benzene
Chloroform
Naphthalene
Di(Z-ethylhexyl)
phthalate
2,4-Dimethylphenol
Pyrene
1,4-Dioxane
Fluorene
Fluoranthene
2-Methylphenol
(o-Cresol)
Toluene
Carbon disulfide
CASRN
110-86-1
91-57-6
71-43-2
67-66-3
91-20-3
117-81-7
105-67-9
129-00-0
123-91-1
86-73-7
206-44-0
95-48-7
108-88-3
75-15-0
Point of
departure
(mg/kg-
day)
1
3.5
1.2
12.9
71
19
50
75
9.6
125
125
50
238
11
Total un-
certainty
factors
1000
1000
300
1000
3,000
1000
3000
3000
300
3000
3000
1000
3000
100
RfV
Chronic
RfD
(mg/kg-
day)
0.001
0.004
0.004
0.01
0.02
0.02
0.02
0.03
0.03
0.04
0.04
0.05
0.08
0.1
Non-cancer effect
Increased liver weight
Pulmonary alveolar
proteinosis
Decreased lymphocyte
count in humans
Fatty cyst formation in
the liver; elevated
SGPT(orALT)
Decreased mean
terminal body weight >
10%
Increased relative liver
weight
Clinical signs;
hematological changes
Kidney effects
Liver and kidney
toxicity
Decreased RBC, packed
cell volume and
hemoglobin
Nephropathy; increased
liver weights;
hematological
alterations
Decreased body weights
and neurotoxicity
Increased absolute
kidney weight
Fetal toxicity and
malformations
Source
IRIS
IRIS
IRIS
IRIS
IRIS
IRIS
IRIS
IRIS
IRIS
IRIS
IRIS
IRIS
IRIS
IRIS
               This document is a draft for review purposes only and does not constitute Agency policy.

June 2015                                       9-30                    DRAFT—DO NOT CITE OR QUOTE

-------
     Hydraulic Fracturing Drinking Water Assessment
Chapter 9 - Identification and Ht
             Across the Hyc
Chemical
Cumene
Benzyl alcohol
Dibutyl phthalate
Ethylbenzene
Acetophenone
Diphenylamine
Xylenes
Phenol
Acetone
CASRN
98-82-8
100-51-6
84-74-2
100-41-4
98-86-2
122-39-4
1330-20-
7
108-95-2
67-64-1
Point of
departure
(mg/kg-
day)
110
143
125
97.1
423
10
179
93
900
Total un-
certainty
factors
1000
1000
1000
1000
3000
100
1000
300
1000
RfV
Chronic
RfD
(mg/kg-
day)
0.1
0.1
0.1
0.1
0.1
0.1
0.2
0.3
0.9
Non-cancer effect
increased average
kidney weight in female
rats
Effects on survival,
growth, and tissue
histopathology
Increased mortality
liver and kidney
toxicity; histopathology
General toxicity; no
LOAEL identified
Alterations in clinical
chemistry; increased
kidney, liver, and spleen
weights
Decreased body weight;
increased mortality
Decreased maternal
weight gain;
developmental toxicity
Nephropathy
Source
IRIS
PPRTV
IRIS
IRIS
IRIS
HHBP
IRIS
IRIS
IRIS
     9.5.3.4.  MCDA Results: Flow/back and Produced Water
1    The hazard potential scores of the selected 23 chemicals detected in flowback and produced water
2    are presented in Table 9-9.
     Table 9-9. MCDA results for 23 chemicals in hydraulic fracturing flowback and produced
               water.
               Chemicals are ranked, from high to low, based on total hazard potential score. See Section 9.5.2 for
               details on the calculation.
Chemical
Benzene
Pyridine
Naphthalene
CASRN
71-43-2
110-86-1
91-20-3
Physicochemical
properties score
0.75
0.75
0.75
Occurrence
score
1.00
1.00
0.67
Toxicity
score
1.00
1.00
1.00
Total hazard
potential score
2.75
2.75
2.42
                   This document is a draft for review purposes only and does not constitute Agency policy.
     June 2015                                      9-31                   DRAFT—DO NOT CITE OR QUOTE

-------
      Hydraulic Fracturing Drinking Water Assessment
Chapter 9 - Identification and Ht
             Across the Hyc
Chemical
2,4-Dimethylphenol
2-Methylnaphthalene
Chloroform
2-Methylphenol
Benzyl alcohol
Bis(Z-Ethylhexyl) Phthalate
Carbon Disulfide
Toluene
Acetone
Pyrene
Di-n-butyl Phthalate
1,4-Dioxane
Fluoranthene
Xylenes
Ethylbenzene
Phenol
Diphenylamine
Isopropylbenzene
Acetophenone
Fluorene
CASRN
105-67-9
91-57-6
67-66-3
95-48-7
100-51-6
117-81-7
75-15-0
108-88-3
67-64-1
129-00-0
84-74-2
123-91-1
206-44-0
1330-20-
7
100-41-4
108-95-2
122-39-4
98-82-8
98-86-2
86-73-7
Physicochemical
properties score
1.00
0.25
0.75
1.00
1.00
0.25
0.50
0.50
0.75
0.75
0.75
1.00
1.00
0.50
0.50
1.00
1.00
0.25
0.75
0.00
Occurrence
score
0.33
1.00
0.33
0.33
0.67
0.67
1.00
1.00
0.67
0.00
0.33
0.00
0.00
1.00
0.33
0.67
0.00
0.67
0.00
0.00
Toxicity
score
1.00
1.00
1.00
0.67
0.33
1.00
0.33
0.33
0.00
0.67
0.33
0.67
0.67
0.00
0.33
0.00
0.33
0.33
0.33
0.67
Total hazard
potential score
2.33
2.25
2.08
2.00
2.00
1.92
1.83
1.83
1.42
1.42
1.42
1.67
1.67
1.50
1.17
1.67
1.33
1.25
1.08
0.67
 1    The highest total hazard potential scores for chemicals in flowback and produced water went to
 2    benzene and pyridine, followed closely by naphthalene. These three chemicals all have RfVs that fell
 3    in the lowest (most toxic) quartile relative to other chemicals in the hazard evaluation (RfDs of
 4    0.004, 0.001, and 0.02 mg/kg-day, respectively). Benzene fell in the upper quartile of observed
 5    chemical concentrations (with a maximum reported average concentration of 680 [J.g/1; Barnett
 6    shale produced water, Table E-9), while pyridine and naphthalene fell in the second highest quartile
 7    (with maximum reported average concentrations of 413 and 238 [j.g/1, respectively; Barnett shale
 8    produced water, Table E-10). These three chemicals only scored moderately in terms  of their
 9    physicochemical properties, however, as all three are expected to have somewhat lower transport
10    in water compared to other chemicals in the assessment 2-Methylnaphthalene also fell in the
11    lowest quartile in terms of toxicity (RfD of 0.004 mg/kg-day) and the highest quartile  in terms of
12    concentration (average of 1,362 [J.g/1; Barnett shale produced water, Table E-10), but received a
13    slightly lower score than these chemicals with regards to physiochemical properties.
                   This document is a draft for review purposes only and does not constitute Agency policy.
      June 2015                                    9-32                  DRAFT—DO NOT CITE OR QUOTE

-------
      Hydraulic Fracturing Drinking Water Assessment                                               Chapter 9 - Identification and Ht
                                                                                                   Across the Hyc

 1    Other chemicals occurring in the upper quartile of flowback and produced water concentrations
 2    include toluene (average of 760 [J.g/1; Barnett shale produced water, Table E-9), xylenes (average of
 3    360 [J.g/1; Barnett shale produced water, Table E-9), and carbon disulfide (median of 400 [J.g/1;
 4    Marcellus shale produced water, Table E-10). These chemicals all received moderate hazard
 5    potential scores, as all have higher RfDs (lower toxicity) relative to many of the other chemicals in
 6    the hazard evaluation, and are all expected to have moderate transport in water relative to the
 7    other chemicals.

 8    Other chemicals with RfVs that fell in the lowest (most toxic) quartile in flowback and produced
 9    water include chloroform (RfD of 0.01 mg/kg-day), di(2-ethylhexyl)phthalate (RfD of 0.02 mg/kg-
10    day), and 2,4,-dimethylphenol (RfD of 0.02 mg/kg-day). Of these, di(2-ethylhexyl)phthalate was
11    detected at moderately high concentrations relative to other chemicals in the assessment (average
12    of 210 [J.g/1; Barnett shale produced water, Table E-10), but is expected to have reduced mobility in
13    water due primarily to its more hydrophobic properties. The rest are expected to have moderate to
14    high transport in water, but were detected at relatively lower average concentrations compared to
15    other chemicals in the assessment.

      9.5.4.  Summary of Chemicals Detected in Multiple Stages of the Hydraulic Fracturing Water
             Cycle
16    A number of chemicals with federal chronic RfVs that are used in hydraulic fracturing fluids were
17    also found to be present in flowback and produced water stages of the hydraulic fracturing water
18    cycle. The use of a chemical in hydraulic fracturing fluids, and subsequent presence in later  stages
19    of the hydraulic fracturing water cycle, is of particular interest in demonstrating which chemicals in
20    this dataset may be mixed, injected, and then detected downstream in the water cycle. This  section
21    focuses on that group of chemicals.

22    Based on the available information in our datasets, 23 chemicals overall had federal chronic RfVs
2 3    and were identified as being used in hydraulic fracturing fluids and detected in the
24    flowback/produced water stage of the hydraulic fracturing water cycle. These chemicals are shown
25    in Table 9-10. 10 of these chemicals were included in both the hazard evaluation of hydraulic
26    fracturing fluids (see Table 9-6 and Table 9-7) and the flowback and produced water hazard
27    evaluation (see Table 9-8 and Table 9-9). This means that these 10 chemicals had both frequency of
28    use data from FracFocus and a reported measured concentration in flowback and produced water
29    from Chapter 7 (Appendix E). These 10 chemicals included all of the BTEX chemicals, as well as
30    naphthalene, 1,4 dioxane, acetone, acetophenone, cumene, and phenol. The chemicals of this group
31    with the lowest chronic oral RfVs were benzene, naphthalene, and  1,4-dioxane. These chemicals all
32    have RfDs within an order of magnitude of each other and are known or likely human carcinogens.
33    The next chemical of this group—toluene—has an RfD 20 times greater than benzene. Overall,
34    benzene was the most toxic of the chemicals listed in Table 9-10.
                   This document is a draft for review purposes only and does not constitute Agency policy.
      June 2015                                    9-33                  DRAFT—DO NOT CITE OR QUOTE

-------
Hydraulic Fracturing Drinking Water Assessment
Chapter 9 - Identification and Ht
              Across the Hyc
Table 9-10. List of the 23 chemicals with federal chronic RfVs identified to be used in
           hydraulic fracturing fluids and detected in the flowback/produced water stage of
           the hydraulic fracturing water cycle.
Chemical
1,4-Dioxane
Acetone
Acetophenone
Benzene
Cumene
Ethylbenzene
Naphthalene
Phenol
Toluene
Xylenes
1,2-Propylene
glycol
Dichloromethane
Ethylene glycol
Formic acid
Methanol
Aluminum
Iron
Di(Z-ethylhexyl)
phthalate
Acrolein
Arsenic
Chlorine
Chromium (III)
Chromium (VI)
Zinc
CASRN
123-91-1
67-64-1
98-86-2
71-43-2
98-82-8
100-41-4
91-20-3
108-95-2
108-88-3
1330-20-7
57-55-6
75-09-2
107-21-1
64-18-6
67-56-1
7429-90-5
7439-89-6
117-81-7
107-02-8
7440-38-2
7782-50-5
16065-83-1
18540-29-9
7440-66-6
Used in
hydraulic
fracturing
fluids?
Y
Y
Y
Y
Y
Y
Y
Y
Y
Y
Y
Y
Y
Y
Y
Y
Y
Y
Y
Y
Y
Y
Y
Y
FracFocus
frequency
of use
data?
Y
Y
Y
Y
Y
Y
Y
Y
Y
Y
Y
Y
Y
Y
Y
Y
Y
-
-
-
-
-
-
-
Detected in
flowback or
produced
water?
Y
Y
Y
Y
Y
Y
Y
Y
Y
Y
Y
Y
Y
Y
Y
Y
Y
Y
Y
Y
Y
Y
Y
Y
Physicochemical
properties data
available?
Y
Y
Y
Y
Y
Y
Y
Y
Y
Y
Y
Y
Y
Y
Y
-
-
Y
Y
-
-
-
-
-
In hazard
evaluation?3
FF+FB
FF+FB
FF+FB
FF+FB
FF+FB
FF+FB
FF+FB
FF+FB
FF+FB
FF+FB
FF
FF
FF
FF
FF
No
No
FB
No
No
No
No
No
No
 a FF+FB: chemical in both the hydraulic fracturing fluid and flowback/produced water hazard evaluations; FF or FB: chemical in
 either the hydraulic fracturing fluid or flowback/produced water hazard evaluations. A dash indicates data for chemical not
 available.
               This document is a draft for review purposes only and does not constitute Agency policy.
June 2015                                         9-34                    DRAFT—DO NOT CITE OR QUOTE

-------
      Hydraulic Fracturing Drinking Water Assessment                                                Chapter 9 - Identification and Ht
                                                                                                   Across the Hyc

 1    An additional 6 chemicals were included in either the hazard evaluation of hydraulic fracturing
 2    fluids (see Table 9-6 and Table 9-7) or the flowback and produced water hazard evaluation (see
 3    Table 9-8 and Table 9-9), but not both. These chemicals were reported to be have been used in
 4    hydraulic fracturing fluids and detected in flowback/produced water, but lacked the occurrence
 5    data (frequency of use or a measured concentration) to support inclusion in both of these hazard
 6    evaluations. The remaining 8 chemicals reported to have been used in hydraulic fracturing fluids or
 7    detected in flowback/produced water were not included in either of the hazard evaluations
 8    presented above because they lacked one or more of the inclusion criteria. These chemicals include
 9    acrolein as well as several metals. Arsenic and acrolein have the lowest RfDs by an order of
10    magnitude and arsenic is classified as a known human carcinogen by the EPA, IARC, and NTP.
11    Chromium (VI) is also classified as a known human carcinogen by IARC and NTP.

      9.6. Synthesis
12    The overall objective of this chapter was to identify and provide information on the toxicological
13    properties of chemicals used in hydraulic fracturing and of hydraulic fracturing waste water
14    constituents, and to evaluate the potential hazard of these chemicals to drinking water resources.
15    Toward this end, the EPA developed a comprehensive list of 1,173 chemicals with reported
16    occurrence in the hydraulic fracturing water cycle, separating them into subsets based on whether
17    they were reported to have been used in hydraulic fracturing fluids or detected in flowback and
18    produced water. First, for each of these chemicals, RfVs and OSFs from selected federal, state, and
19    international sources were collected when available. Second, for subsets of chemicals that were
20    identified as being of interest in previous chapters of this report, federal chronic RfVs were used to
21    conduct an initial identification of the potential human health hazards inherent to these chemicals.
22    Finally, for other subsets of chemicals that had data available, an approach for a more data-
23    informed hazard evaluation was illustrated by integrating data on federal chronic RfVs, occurrence,
24    and physicochemical properties using an MCDA framework.

      9.6.1.  Summary of Findings
2 5    Across the industrial landscape, thousands of chemicals are used commercially that lack toxicity
26    data (Judsonetal.. 2009). Similarly, major knowledge gaps exist regarding the toxicity of most
27    chemicals used in hydraulic fracturing fluids or detected in flowback/produced water, impeding the
28    assessment of human health risks associated with drinking water resources affected by hydraulic
29    fracturing.

30    Of the 1,076 chemicals used in hydraulic fracturing fluids, chronic RfVs and/or OSFs from all of the
31    selected federal, state, or international sources were available for 90 chemicals (8.4%). From the
32    federal sources alone, chronic oral RfVs were available for  73 chemicals (6.8%), and OSFs were
33    available for  15 (1.4%). Potential hazards associated with these chemicals include carcinogenesis,
34    immune system effects, changes in body weight, changes in blood chemistry, cardiotoxicity,
35    neurotoxicity, liver and kidney toxicity, and reproductive and developmental toxicity.

36    Of the 134 chemicals that are reported to have been detected in hydraulic fracturing flowback or
37    produced water, chronic RfVs and/or OSFs from all of the selected federal, state, or international


                   This document is a draft for review purposes only and does not constitute Agency policy.
      June 2015                                    9-35                  DRAFT—DO NOT CITE OR QUOTE

-------
      Hydraulic Fracturing Drinking Water Assessment                                                Chapter 9 - Identification and Ht
                                                                                                   Across the Hyc

 1    sources were available for 83 chemicals (62%). From the federal sources alone, chronic RfVs were
 2    available for 70 chemicals (52%), and OSFs were available for 20 (15%). Potential hazards
 3    associated with these chemicals include carcinogenesis, immune system effects, changes in body
 4    weight, changes in blood chemistry, pulmonary toxicity, neurotoxicity, liver and kidney toxicity, and
 5    reproductive and developmental toxicity.

 6    Of the chemicals included in the hazard evaluations, benzene is the only one of these chemicals with
 7    an OSF that is classified as a known human carcinogen, while acrylamide was found to be the most
 8    potent likely human carcinogen. Several other chemicals, including 1,4-dioxane, dichloromethane,
 9    naphthalene, and ethylbenzene are also classified as possible, probable, or likely human
10    carcinogens.

11    Toxicity information spans a wide range with respect to extent, quality and reliability. The sources
12    of RfVs and OSFs selected for the purposes of this chapter are based on criteria developed
13    specifically for  this report For the total 1,173 chemicals identified on the EPA's list, federal, state,
14    and international chronic RfVs and/or OSFs that met stringent selection criteria were available for
15    147 (13%) of the chemicals. Several of the RfVs from selected sources were derived using UFs of up
16    to several orders of magnitude, indicating uncertainty when comparing chemicals for potential
17    toxicity and identifying the chemicals that may be more likely to present a human health hazard.
18    For  many of the chemicals used in hydraulic fracturing or found in flowback or produced water
19    there may be relevant information, including cancer and noncancer-related information, from one
20    or more sources that were not evaluated in this chapter.  In instances where toxicity data is not
21    available from selected sources, risk assessors may need to draw from alternative sources of hazard
22    information. The chapter discusses two available resources for consideration when RfVs and/or
23    OSFs are not available: QSAR-predicted toxicity data, and toxicity data from the EPA's ACToR
24    database. Oral toxicity data was available on ACToR for 642 (55%) of the chemicals. The
25    information available in the ACToR data warehouse ranges from the federal RfVs discussed in
26    Section 9.3.1, which have undergone extensive peer review, to RfVs and study and test results that
27    have undergone little to no peer review.

28    When considering the potential impact of chemicals on drinking water resources and human health,
29    it is  important to consider exposure as well as toxicological properties. The majority of chemicals
30    identified in this report lacked the necessary data to conduct such an assessment However,
31    integrating data on toxicity, occurrence, and physicochemical properties using an MCDA framework
32    enabled a more data-informed hazard evaluation on some chemicals. This  analysis highlighted
33    several chemicals that may be more likely than others to reach drinking water and create a
34    toxicological hazard. In hydraulic fracturing fluid, an example is propargyl alcohol. It was among the
35    chemicals with the lowest RfVs considered in this hazard evaluation, was used in 33% of wells in
36    the FracFocus database, and is water soluble with low volatility. In flowback and produced water,
37    examples of such chemicals include benzene, pyridine, and naphthalene. These chemicals were also
38    among those with the lowest RfVs considered in this hazard evaluation, are expected to be
39    relatively mobile in water, and were present at relatively high average concentrations in flowback.
                   This document is a draft for review purposes only and does not constitute Agency policy.
      June 2015                                    9-36                  DRAFT—DO NOT CITE OR QUOTE

-------
      Hydraulic Fracturing Drinking Water Assessment                                                Chapter 9 - Identification and Ht
                                                                                                    Across the Hyc

      9.6.2.  Factors Affecting the Frequency or Severity of Impacts
 1    When assessing chemical hazards, there are multiple pieces of information that could be taken into
 2    account This includes knowledge of the chemicals used at a given well site, the toxicological and
 3    physicochemical properties of these chemicals, the amount of fluid being used and recovered, the
 4    likelihood of well integrity failures, and the likelihood of spills and other unintentional releases.
 5    These topics were previously discussed in Chapters 5 through 8 of this report Because of the large
 6    volumes of fluid being injected, even chemicals representing a small percentage of the total fluid by
 7    mass may pose a potential for exposure in the event of a spill or leak.

 8    Overall, contamination of drinking water resources depends on site-, chemical-, and fluid-specific
 9    factors [Goldstein etal.. 2014], and the exact mixture and concentrations of chemicals at a site will
10    depend upon the geology and the company's preferences. Therefore, potential hazard and risk
11    considerations are best made on a site-specific, well-specific basis. While the MCDA results in this
12    chapter illustrate an approach to evaluate the relative hazards of these chemicals at the national
13    level, a site-specific hazard evaluation would be necessary in order to identify chemicals of concern
14    at the local level.

15    For example, consider (E)-crotonaldehyde, which is one of the more toxic chemicals considered in
16    the hazard evaluation of hydraulic fracturing fluids. (E)-crotonaldehyde is reportedly used in only
17    0.06% of wells in the FracFocus database, based on the EPA's analysis. If the FracFocus database
18    represents a fair sample of all of the wells across the country, then the likelihood of (E)-
19    crotonaldehyde contamination  on a nationwide scale is limited. However, this in no way diminishes
20    the likelihood of (E)-crotonaldehyde contamination at well sites where this chemical is used.
21    Therefore, potential exposures to more toxic but infrequently used chemicals are more of a local
22    issue, rather than a national one.

23    This is in contrast with methanol, which was reported in 73% of wells in the FracFocus database.
24    Methanol is soluble and relatively mobile in water, but has a higher RfV relative to other chemicals
25    in the hazard evaluation. Therefore, when considering chemical usage on a nationwide basis,
26    methanol may be expected to have a higher exposure potential compared to other chemicals, with a
27    moderate overall hazard potential due to its relatively high RfV.

      9.6.3.  Uncertainties
2 8    There are several notable uncertainties in the chemical and toxicological data that limit a
29    comprehensive assessment of the potential health impacts of hydraulic fracturing on drinking
30    water resources.

31    For the purposes of this chapter, the lack of RfVs and OSFs from the sources meeting stringent
32    selection criteria is the most significant data gap. For instance, of the 32 chemicals (excluding
33    water, quartz, and sodium chloride) that are used in >10% of wells nationwide  according to
34    FracFocus, federal chronic RfVs were only available for 7 chemicals. Without these reliable and peer
3 5    reviewed data, comprehensive hazard evaluation and hazard identification of chemicals is difficult,
36    and the ability to consider the potential cumulative effects of exposure to chemical mixtures in
                   This document is a draft for review purposes only and does not constitute Agency policy.
      June 2015                                    9-37                  DRAFT—DO NOT CITE OR QUOTE

-------
      Hydraulic Fracturing Drinking Water Assessment                                               Chapter 9 - Identification and Ht
                                                                                                   Across the Hyc

 1    hydraulic fracturing fluid, flowback, or produced water is limited. Consequently, potential impacts
 2    on drinking water resources and human health may not be assessed adequately.

 3    Another major uncertainty lies in the total list of chemicals that was compiled for this chapter. As
 4    discussed in Section 5.1.3, information is lacking on the chemicals that are used in hydraulic
 5    fracturing fluid formulation. CBI chemicals, which were present in approximately 70% of well
 6    records on the FracFocus database, were excluded from the EPA's analysis. The analysis also
 7    excluded ingredient records that were not able to be assigned standardized chemical names, which
 8    resulted in approximately 35% of FracFocus ingredient records being excluded from the report
 9    This lack of data limits the ability to more completely assess the impact of chemicals that are
10    potentially used with great frequency. Moreover, there may be a regional bias in the EPA's analysis
11    of FracFocus, as 78% of chemical disclosures in the FracFocus database came from five states, and
12    47% were from Texas. Despite these limitations, the FracFocus database remains the most
13    complete source for tracking hydraulic fracturing chemical usage in the United States, and therefore
14    was the best available source for the hazard evaluation in this chapter. Although the sources used to
15    compile the chemical list represented the best available data at the time of this study, it is possible
16    that some of these chemicals are no longer used at all, and many of these chemicals  may only be
17    used infrequently. Therefore, it may be possible that significantly fewer than 1,076 chemicals are
18    currently used in abundance. Consequently, having a better understanding of the chemicals and
19    formulations, including those that are CBI, along with their frequency of use and volumes, would
2 0    greatly benefit risk assessment and risk management decisions.

21    Additionally, the list of flowback and produced water chemicals identified in this chapter is  almost
2 2    certainly incomplete. Few studies to date have examined the chemical composition of flowback and
2 3    produced water, and the hazard evaluation in this chapter relied on data from the relatively small
24    number of studies that are presented in Appendix E of this assessment. As discussed in Chapter 7,
2 5    chemicals and their metabolites may go undetected simply because they were not included  in the
2 6    analytical methodology. Additionally, chemical analysis of flowback and produced water may be
2 7    challenging, because high levels of dissolved solids in flowback and wastewater can interfere with
2 8    chemical detection.  As a result, it is likely that there are chemicals of concern in flowback and
29    produced water that have not been detected or reported.

3 0    Finally, when considering the MCDA framework that was used to illustrate an approach for  hazard
31    evaluation, it should be noted that the physicochemical variables were chosen specifically to reflect
3 2    chemical mobility and persistence in water. While this framework draws attention towards those
3 3    chemicals that are most likely to be carried in water, it does not attempt to address the numerous
34    other physicochemical variables that may affect chemical exposure. For instance, as discussed in
35    Chapter 5, hydrophobic chemicals may act as long-term sources of pollution by sorbing to soils or
36    sediments. Additionally, volatile chemicals that dissipate into the air have the potential to pose air
37    pollution hazards, which are not considered in this drinking water assessment; or could potentially
38    be deposited in bodies of water that are distant from the hydraulic fracturing site. Furthermore, as
39    discussed in Chapter 5, chemical fate and transport will be influenced by environmental and site-
40    specific conditions. The fate of a chemical in a chemical mixture will be also influenced by the other
                   This document is a draft for review purposes only and does not constitute Agency policy.
      June 2015                                    9-38                  DRAFT—DO NOT CITE OR QUOTE

-------
      Hydraulic Fracturing Drinking Water Assessment                                                Chapter 9 - Identification and Ht
                                                                                                    Across the Hyc

 1    chemicals that are present in the mixture, and the relative concentrations of each. Although the
 2    assessment of these various scenarios is outside the scope of this report, the potential hazards
 3    associated with hydrophobic or volatile chemicals should not be discounted when interpreting the
 4    results of this hazard evaluation. It should be emphasized that the MCDA framework illustrated in
 5    this chapter represents just one method that can be used to integrate chemical data for hazard
 6    evaluation, and is readily adaptable to include different variables, different weights for the
 7    variables, and site-specific considerations.

      9.6.4.  Conclusions
 8    The EPA has identified 1,173 chemicals used or detected in the hydraulic fracturing water cycle.
 9    Toxicity-based chronic RfVs and/or OSFs from sources meeting selection criteria are not available
10    for the large majority (87%) of these chemicals. In addition, 56% of these chemicals do not have
11    physicochemical property data. Furthermore, 36% of the chemicals used in hydraulic fracturing
12    fluids lack data on their nationwide frequency of use, and very few studies have analyzed the
13    chemical composition of flowback and produced water. Given the large number of chemicals used
14    or detected in various stages of the hydraulic fracturing water cycle, as well as the large number of
15    hydraulic fracturing wells nationwide, this missing chemical information represents a significant
16    data gap. Because of these large data gaps for drinking water resources, it remains challenging to
17    fully understand the toxicity and potential health impacts for single chemicals as well as mixtures of
18    chemicals associated with hydraulic fracturing processes. This chapter provides an initial overall
19    assessment of the potential human health effects associated with hydraulic fracturing on a
20    nationwide basis. It also provides tools that may support risk assessment and risk management
21    decision making at the local and regional level.

22    The toxicological data, occurrence data, and physicochemical data compiled in this report provide a
23    resource for assessing the potential hazards associated with chemicals in the hydraulic fracturing
24    water cycle. Additionally, the MCDA framework presented herein illustrates one method for
25    integrating these data for hazard evaluation. While the analysis in this chapter is constrained to the
26    assessment of chemicals on a nationwide scale, this approach is readily adaptable for use on a
27    regional or site-specific basis.

28    This collection of data provides a tool to inform decisions about protection of drinking water
29    resources. Agencies may use  these results to prioritize chemicals for hazard assessment or for
30    determining future research priorities. Industry may use this information to prioritize chemicals for
31    replacement with less toxic, persistent, and mobile alternatives. A summary of the findings related
32    to the overall objective of this chapter and the research questions is presented in Text Box 9-1.
                    This document is a draft for review purposes only and does not constitute Agency policy.
      June 2015                                     9-39                  DRAFT—DO NOT CITE OR QUOTE

-------
      Hydraulic Fracturing Drinking Water Assessment
Chapter 9 - Identification and Ht
              Across the Hyc
      Text Box 9-1. Research Questions Revisited.

 1    What are the toxicological properties of hydraulic fracturing fluid chemical additives?
 2    •   In a nationwide assessment, the EPA identified 1,076 chemicals that are used in hydraulic fracturing
 3        fluids. This does not include chemicals classified as CBI, which the FracFocus database indicates are used
 4        in more than 70% of wells. Chronic RfVs and/or OSFs from selected federal, state, and international
 5        sources were available for 90 (8.4%) of these chemicals. From the federal sources alone, chronic RfVs
 6        were available for 73 chemicals (6.8%), and OSFs were available for 15 chemicals (1.4%). RfVs and OSFs
 7        were not available for the majority of chemicals that are used in hydraulic fracturing fluid, representing a
 8        significant data gap with regards to hazard identification. Of the chemicals that have selected RfVs, health
 9        effects include the potential for carcinogenesis, immune system effects, changes in body weight, changes
10        in blood chemistry, cardiotoxicity, neurotoxicity, liver and kidney toxicity, and reproductive and
11        developmental toxicity.

12    •   When considering the hazard evaluation of these chemicals on a nationwide scale, chemicals such as
13        propargyl alcohol stand out for their relatively lower RfVs, high frequency of use, and expected transport
14        and mobility in water. However, the FracFocus database indicates that most chemicals are used
15        infrequently on a nationwide scale; therefore, potential exposures to the majority of these chemicals are
16        more likely to be a local issue, rather than a national one. Accordingly, potential hazard and risk
17        considerations for hydraulic fracturing fluid chemical additives are best made on a site-specific, well-
18        specific basis.

19    What are the toxicological properties of hydraulic fracturing wastewater constituents?

20    •   This assessment identified 134 chemicals that are reported to have been detected in hydraulic fracturing
21        flowback or produced water. These include chemicals that are added to hydraulic fracturing fluids during
22        the chemical mixing stage, as well as naturally occurring organic chemicals, metals, naturally occurring
23        radioactive material, and other subterranean chemicals that may be mobilized by the hydraulic fracturing
24        process. Chronic RfVs and/or OSFs from selected federal, state, and international sources were available
25        for 83 (62%) of these chemicals. From the federal sources alone, chronic RfVs were available for 70
26        chemicals (52%), and OSFs were available for 20 chemicals (15%). Of the  chemicals that had selected
27        RfVs, health effects include the potential for carcinogenesis, immune system effects, changes in body
28        weight,  changes in blood chemistry, pulmonary toxicity, neurotoxicity, liver and kidney toxicity, and
29        reproductive and developmental toxicity.

30    •   In a hazard evaluation of flowback and produced water data, chemicals such as benzene, pyridine, and
31        naphthalene stood out for their relatively lower RfVs, high average concentrations, and expected
32        transport and mobility in water. However, the chemicals present in flowback and produced water are
33        likely to vary on a regional and well-specific basis as a result of geological  differences as well as
34        differences between hydraulic fracturing fluid formulations. Therefore, potential hazard and risk
35        considerations are best made on a site-specific basis.
                     This document is a draft for review purposes only and does not constitute Agency policy.
      June 2015                                        9-40                    DRAFT—DO NOT CITE OR QUOTE

-------
Hydraulic Fracturing Drinking Water Assessment                                                   Chapter 9 - Identification and Ht
                                                                                                      Across the Hyc


9.7.  References for Chapter 9

Colborn, T: Kwiatkowski, C: Schultz, K: Bachran, M. (2011). Natural gas operations from a public health
   perspective. Hum Ecol Risk Assess 17:1039-1056. http://dx.doi.org/10.1080/10807039.2011.605662

Finkel. M: Hays. I: Law. A. (2013). The shale gas boom and the need for rational policy. Am J Public Health
   103: 1161-1163. http://dx.doi.org/10.2105/AIPH.2013.301285

Goldstein. BD: Brooks. BW: Cohen. SD: Gates. AE: Honeycutt. ME: Morris. IB: Orme-Zavaleta. I: Penning. TM:
   Snawder, I. (2014). The role of lexicological science in meeting the challenges and opportunities of
   hydraulic fracturing. Toxicol Sci 139: 271-283. http://dx.doi.org/10.1093/toxsci/kfu061

House of Representatives  (U.S. House of Representatives). (2011). Chemicals used in hydraulic fracturing.
   Washington, D.C.: U.S. House of Representatives, Committee on Energy and Commerce, Minority Staff.
   http://democrats.energycommerce.house.gov/sites/default/files/documents/Hydraulic-Fracturing-
   Chemicals-2011-4-18.pdf

Hristozov. PR: Zabeo. A: Foran. C: Isigonis. P: Critto. A: Marcomini. A: Linkov. I. (2014). A weight of evidence
   approach for hazard screening of engineered nanomaterials. Nanotoxicology 8: 72-87.
   http://dx.doi.org/10.3109/17435390.2012.750695

ludson. R: Richard. A: Dix. PI: Houck. K: Martin. M: Kavlock. R: Dellarco. V: Henry. T: Holderman. T: Sayre. P:
   Tan. S: Carpenter. T: Smith. E. (2009). The toxicity data landscape for environmental chemicals [Review].
   Environ Health Perspect 117: 685-695. http://dx.doi.org/10.1289/ehp.0800168

Kahrilas. GA: Blotevogel. I: Stewart. PS: Borch. T. (2015). Biocides in hydraulic fracturing fluids: a critical
   review of their usage, mobility, degradation, and toxicity. Environ Sci Technol 49:16-32.
   http://dx.doi.org/10.1021/es503724k

Kassotis, CD: Tillitt, DE: Wade Davis, I: Hormann, AM: Nagel, SC. (2014). Estrogen and androgen receptor
   activities of hydraulic fracturing chemicals and surface and ground water in a drilling-dense region.
   Endocrinology 155: 897-907. http://dx.doi.org/10.1210/en.2013-1697

Korfmacher. KS: Jones. WA: Malone. SL: Vinci. LF. (2013). Public health and high volume hydraulic fracturing.
   New Solutions: A Journal of Environmental and Occupational Health Policy 23:13-31.
   http://dx.doi.Org/10.2190/NS.23.l.c

Linkov. I: Welle. P: Loney.  D: Tkachuk. A: Canis. L: Kim. IB: Bridges. T. (2011). Use of multicriteria decision
   analysis to support weight of evidence evaluation. RiskAnal 31:1211-1225.
   http://dx.doi.0rg/10.llll/i.1539-6924.2011.01585.x

McKenzie. LM: Guo. R: Witter. RZ: Savitz. DA: Newman. L: Adgate. IL. (2014). Birth  outcomes and maternal
   residential proximity to natural gas development in rural Colorado. Environ Health Perspect 122: 412-
   417. http://dx.doi.org/10.1289/ehp.1306722

Mitchell. I: Pabon.  P: Collier. ZA: Egeghy. PP: Cohen-Hubal. E: Linkov. I: Vallero. DA. (2013b). A decision
   analytic approach to exposure-based chemical prioritization. PLoS ONE 8: e70911.
   http://dx.doi.org/1371/journal.pone.0070911

Moudgal. Cl: Venkatapathy. R: Choudhury. H: Bruce. RM: Lipscomb. 1C. (2003). Application of QSTRs in the
   selection of a surrogate toxicity value for chemical of concern. Environ Sci Technol 37: 5228-5235.

NRC (National Research Council). (2014). A framework to guide selection of chemical alternatives.
   Washington, D.C.: The National Academies Press, http://www.nap.edu/catalog/18872/a-framework-to-
   guide-selection-of-chemical-alternatives

Rupp. B: Appel. KE: Gundert-Remy. U. (2010). Chronic oral LOAEL prediction by using a commercially
   available computational QSAR tool. Arch Toxicol 84: 681-688. http://dx.doi.org/10.1007/s00204-010-
   0532-x
               This document is a draft for review purposes only and does not constitute Agency policy.

June 2015                                       9-41                   DRAFT—DO NOT CITE OR QUOTE

-------
Hydraulic Fracturing Drinking Water Assessment                                                   Chapter 9 - Identification and Ht
                                                                                                      Across the Hyc


Stringfellow. WT: Domen. IK: Camarillo. MK: Sandelin. WL: Borglin. S. (2014). Physical, chemical, and
   biological characteristics of compounds used in hydraulic fracturing. J Hazard Mater 275: 37-54.
   http://dx.doi.0rg/10.1016/j.jhazmat.2014.04.040

U.S. EPA (U.S. Environmental Protection Agency). (2002a). A review of the reference dose and reference
   concentration processes. (EPA/630/P-02/002F). Washington, DC: U.S. Environmental Protection Agency,
   Risk Assessment Forum. http://cfpub.epa.gov/ncea/cfm/recordisplay.cfm?deid=51717

U.S. EPA (U.S. Environmental Protection Agency). (2002b). Toxicological review of benzene (noncancerous
   effects) [EPA Report]. (EPA/635/R-02/001F). Washington, DC.

U.S. EPA (U.S. Environmental Protection Agency). (2005). Pollution prevention (P2) framework [EPA Report].
   (EPA-748-B-04-001). Washington, DC: Office of Pollution Prevention and Toxics.
   http://www.epa.gov/oppt/sf/pubs/p2frame-june05a2.pdf

U.S. EPA (U.S. Environmental Protection Agency). (2010). Toxicological review of acrylamide (CAS No. 79-06-
   1) in support of summary information on the Integrated  Risk Information System (IRIS) [EPA Report].
   (EPA/635/R-07/008F). Washington, DC.

U.S. EPA (U.S. Environmental Protection Agency). (2011a). Design for the Environment program alternatives
   assessment criteria for hazard evaluation (version 2.0). Washington, D.C.
   http://www2.epa.gov/saferchoice/alternatives-assessment-criteria-hazard-evaluation

U.S. EPA (U.S. Environmental Protection Agency). (2011d). Terminology services (TS): Vocabulary catalog -
   IRIS glossary. Available online at
   http://ofmpub.epa.gov/sor internet/registry/termreg/searchandretrieve/glossariesandkeywordlists/se
   arch.do?details=&glossaryName=IRIS%20Glossary (accessed May 21, 2015).

U.S. EPA (U.S. Environmental Protection Agency). (2012f). Study of the potential impacts of hydraulic
   fracturing on drinking water resources: Progress report. (EPA/601/R-12/011). Washington, DC: U.S.
   Environmental Protection Agency, Office of Research and Development.
   http://nepis.epa.gov/exe/ZyPURL.cgi?Dockey=P100FH8M.txt

U.S. EPA (U.S. Environmental Protection Agency). (2013f). Toxicological review of 1,4-Dioxane (with
   inhalation update) (CAS No. 123-91-1) in support of summary information on the Integrated Risk
   Information System (IRIS) [EPA Report]. (EPA-635/R-11/003-F). Washington, DC.

U.S. EPA (U.S. Environmental Protection Agency). (2014a). Alternatives assessment for the flame retardant
   decabromodiphenyl ether (DecaBDE). Washington, D.C. http://www2.epa.gov/saferchoice/partnership-
   evaluate-flame-retardant-alternatives-decabde-publications

U.S. EPA (U.S. Environmental Protection Agency). (2014d). Flame retardant alternatives for
   hexabromocyclododecane (HBCD) [EPA Report]. (EPA/740/R-14/001). Washington, D.C.
   http://www2.epa.gov/saferchoice/partnership-evaluate-flame-retardant-alternatives-hbcd-publications

U.S. EPA (U.S. Environmental Protection Agency). (2015a). Analysis of hydraulic fracturing fluid data from the
   FracFocus chemical disclosure registry 1.0 [EPA Report]. (EPA/601/R-14/003). Washington, D.C.: Office
   of Research and Development, U.S. Environmental Protection Agency.
   http://www2.epa.gov/hfstudy/analysis-hydraulic-fracturing-fluid-data-fracfocus-chemical-disclosure-
   registry-1-pdf

Venkatapathy. R: Moudgal. Cl: Bruce. RM. (2004). Assessment of the oral rat chronic lowest observed adverse
   effect level model in TOPKAT, a QSAR software package for toxicity prediction. J chem inf comput sci 44:
   1623-1629. http://dx.doi.org/10.1021/ci049903s

Wattenberg, EV: Bielicki, IM: Suchomel, AE: Sweet, IT: Void,  EM: Ramachandran, G. (In Press) Assessment of
   the acute and chronic health hazards of hydraulic fracturing fluids. J Occup Environ Hyg.
   http://dx.doi.org/10.1080/15459624.2015.1029612
               This document is a draft for review purposes only and does not constitute Agency policy.

June 2015                                       9-42                   DRAFT—DO NOT CITE OR QUOTE

-------
      Hydraulic Fracturing Drinking Water Assessment                                                 Chapter 9 - Identification and Ht
                                                                                                     Across the Hyc

      Weinhold. B. (2012). The future of tracking: new rules target air emissions for cleaner natural gas production.
         Environ Health Perspect 120: a272-a279. http://dx.doi.org/10.1289/ehp.120-a272
      9.8. Annex

      9.8.1.  Calculation of Physicochemical Property Scores (MCDA Hazard Evaluation)
 1    Section 9.5.2 describes how physicochemical properties scores were based on three subcriteria:
 2    mobility, volatility, and persistence. These subcriteria scores were calculated as follows:

 3    Mobility score: Chemical mobility in water was assessed based upon three physicochemical
 4    properties: the octanol-water partition coefficient (Kow), the organic carbon-water partition
 5    coefficient [Koc], and aqueous solubility. Kow and aqueous solubility were previously discussed in
 6    Section 5.8.3. Koc is a partitioning coefficient that measures the amount of chemical that is adsorbed
 7    onto soil organic carbon per the amount of chemical that is dissolved in water. Like Kow, Koc is
 8    typically reported as a base-10 logarithm (log Koc}. From EPI Suite™, Koc was estimated using the
 9    MCI Method. Chemicals with low Kow and Koc values are hydrophilic, and thus are more likely to
10    move with water rather than sorbing to soils or sediments. Chemicals with high aqueous solubility
11    are also more likely to move with water. Therefore, chemicals with low Kow, low Koc, or high aqueous
12    solubility were ranked as having greater potential to affect drinking water resources. Using the
13    thresholds designated in Table 9-5, each of these properties was assigned a score of 1-4. The
14    highest of these three scores was designated as the mobility score for each chemical.

15    Volatility score: Chemical volatility was assessed based on the Henry's law constant, which was
16    previously discussed in Section 5.8.3. Chemicals with low Henry's law constants are less likely to
17    leave water via volatilization, and were therefore ranked as having greater potential to impact
18    drinking water. Using the thresholds designated in Table 9-5, the Henry's law constant for each
19    chemical was assigned a score of 1-4. This value was designated as the volatility score for each
20    chemical.

21    Persistence score: Chemical persistence was assessed based on estimated half-life in water, which
22    describes how long a chemical will persist in water before it is transformed or degraded. From EPI
23    Suite™, half-life in water was estimated using the Level III Fugacity model. Chemicals with longer
24    half-lives are more persistent, and were therefore ranked as having greater potential to affect
25    drinking water. Using the thresholds designated in Table 9-5, the half-life of each chemical was
26    assigned a score of 1-4. This value was designated as the persistence score for each chemical.

27    For each chemical, the mobility score, volatility score, and persistence score (each on a scale of 1 to
28    4) were summed to calculate a total physicochemical score. The total scores were then
29    standardized by scaling to the highest and lowest scores observed in the subset of chemicals, using
30    the equation described in Section 9.5.2.4.
                    This document is a draft for review purposes only and does not constitute Agency policy.
      June 2015                                     9-43                   DRAFT—DO NOT CITE OR QUOTE

-------
      Hydraulic Fracturing Drinking Water Assessment                                               Chapter 9 - Identification and Ht
                                                                                                    Across the Hyc

      9.8.2.  Example of MCDA Score Calculation
 1    For an example of how the MCDA scores were calculated, consider benzene. This demonstrates how
 2    the MCDA score was calculated for benzene in the hazard evaluation of chemicals used in hydraulic
 3    fracturing fluids:

 4         •   With regards to toxicity (Appendix G), benzene was found to have a federal RfD of 0.004
 5            mg/kg-day (source: IRIS). Within the entire set of chemicals in this hazard evaluation,
 6            federal RfDs ranged from 0.001 mg/kg-day [(E)-crotonaldehyde] to 20 mg/kg-day (1,2-
 7            propylene glycol). The RfD of benzene fell in the lowest (most toxic) quartile of these
 8            scores, and therefore was given a toxicity score of 4. When the results were standardized
 9            to the highest score (4) and lowest score (1) within the set of chemicals, benzene was
10            calculated to have a toxicity score of 1, as follows:
11                                          1 = (4-I)/(4-1)

12         •   Benzene was used in 0.0056% of wells nationwide (U.S. EPA. 2015a). This usage frequency
13            falls in the lowest quartile of chemicals, and therefore benzene was given an occurrence
14            score of 1. When the results were standardized to the highest score (4) and lowest score
15            (1) within the set of chemicals, benzene was calculated to have an occurrence score of 0, as
16            follows:
17                                          0 = (l-l)/(4-l)

18         •   Based on physicochemical properties, benzene received a mobility score of 4 (log Kow =
19            2.13; logKoc = 1.75; solubility =  2000 mg/1), a volatility score of 2 (Henry's law constant =
20            0.00555), and a persistence score of 2 (half-life in water =  37.5 days). These scores sum to
21            a total physicochemical properties score of 8. Within the entire set of chemicals in this
22            hazard evaluation, several chemicals received total scores of 9, which was the highest
23            observed score. Cumene received a total score of 6, which was the lowest observed score.
24            When the results were standardized to the high score (9) and low score (6) using the
25            equation above, benzene was calculated to have a physicochemical properties score of
26            0.67 as follows:
27                                      0.67 = (8-6)/(9-6)

28    To calculate the total hazard potential score for benzene, the physicochemical properties score,
29    toxicity score, and occurrence score were summed for a total of 1.67. These results can be seen in
30    Table 9-7, which shows the MCDA results for chemicals used in hydraulic fracturing fluid.
                   This document is a draft for review purposes only and does not constitute Agency policy.
      June 2015                                    9-44                  DRAFT—DO NOT CITE OR QUOTE

-------
Hydraulic Fracturing Drinking Water Assessment                                 Chapter 10- Synthesis
                             Chapter 10
Synthesis
            This document is a draft for review purposes only and does not constitute Agency policy.
June 2015                                                  DRAFT—DO NOT CITE OR QUOTE

-------
      Hydraulic Fracturing Drinking Water Assessment                                      Chapter 10 - Synthesis
      10.   Synthesis
 1    In this assessment, we examined the potential for hydraulic fracturing for oil and natural gas to
 2    change the quality or quantity of drinking water resources, and identified factors that affect the
 3    frequency or severity of potential impacts. Drinking water resources are defined broadly in this
 4    report as any body of ground water or surface water that now serves, or in the future could serve,
 5    as a source of drinking water for public or private use. We assessed potential effects on drinking
 6    water resources from both routine operations and potential accidents. Impacts were defined as any
 7    change in the quality or quantity of drinking water resources. Where possible, we identified the
 8    mechanisms responsible or potentially responsible for any impacts. For example, a spill of
 9    hydraulic fracturing fluid is a mechanism by which drinking water resources could be impacted.

10    We did this by following water through the hydraulic fracturing water cycle: (1) the withdrawal of
11    ground or surface water needed for hydraulic fracturing fluids; (2) the mixing of water, chemicals,
12    and proppant on the well pad to create the hydraulic fracturing fluid; (3) the injection of hydraulic
13    fracturing fluids into the well to fracture the geologic formation; (4) the management of flowback
14    and produced water, both on the well pad and in transit for reuse, treatment, or disposal; and (5)
15    the reuse, treatment and discharge, or disposal of hydraulic fracturing wastewater.

16    In this chapter, we summarize major findings of the assessment, organized by each stage of the
17    hydraulic fracturing water cycle (Section 10.1);  highlight key uncertainties related to these major
18    findings (Section 10.2); and discuss the assessment's overall conclusions (Section 10.3) and
19    potential uses (Section 10.4).

      10.1.Major Findings
20    From our assessment, we conclude there are above and below ground mechanisms by which
21    hydraulic fracturing activities have the potential to impact drinking water resources. These
22    mechanisms include water withdrawals in times of, or in areas with, low water availability; spills of
23    hydraulic fracturing fluids and produced water; fracturing directly into underground drinking
24    water resources; below ground migration of liquids and gases; and inadequate treatment and
25    discharge of wastewater.

26    We did not find evidence that these mechanisms have led to widespread, systemic impacts on
27    drinking water resources in the United States. Of the potential mechanisms identified in this report,
28    we found specific instances where one or more mechanisms led to impacts on drinking water
29    resources, including contamination of drinking water wells. The number of identified cases,
30    however, was small compared to the number of hydraulically fractured wells.

31    This finding could reflect a rarity of effects on drinking water resources, but may also be due to
32    other limiting factors. These factors include: insufficient pre- and post-fracturing data on the quality
33    of drinking water resources; the paucity of long-term systematic studies; the presence of other
34    sources of contamination precluding a definitive link between hydraulic fracturing activities and an
35    impact; and the inaccessibility of some information on hydraulic fracturing activities and potential
36    impacts.


                   This document is a draft for review purposes only and does not constitute Agency policy.
      June 2015                                    10-1                  DRAFT—DO NOT CITE OR QUOTE

-------
      Hydraulic Fracturing Drinking Water Assessment                                       Chapter 10 - Synthesis


 1    Below, we provide a synopsis of the assessment's key findings, organized according to each stage of
 2    the hydraulic fracturing water cycle. We provide answers to the research questions presented in
 3    the Study Plan and Chapter 1. Results from Chapter 9 (Hazard Evaluation and Identification of
 4    Chemicals across the Hydraulic Fracturing Water Cycle) are included in the Chemical Mixing and
 5    the Flowback and Produced Water sections. While some citations are provided here, individual
 6    chapters can be consulted for additional detail and citations.

      10.1.1. Water Acquisition (Chapter 4)
 7    Water is a major component of nearly all hydraulic fracturing operations. It typically makes up
 8    almost 90% or more of the fluid injected into a well, and each hydraulically fractured well requires
 9    thousands to millions of gallons of water. Cumulatively, hydraulic fracturing activities in the United
10    States used on average 44 billion gal of water a year in 2011 and 2 012, according to the EPA's
11    analysis of FracFocus 1.0 disclosures. Although this represents less than 1% of total annual water
12    use and consumption at this scale, water withdrawals could potentially impact the quantity and
13    quality of drinking water resources at more local scales.

         Research Questions: Water Acquisition

         •   What are the types of water used for hydraulic fracturing?
14    Water for hydraulic fracturing typically comes from surface water,  ground water, or reused
15    hydraulic fracturing waste water. Hydraulic fracturing operations in the eastern United States
16    generally rely on surface water, while operations in the more semi-arid to arid western states
17    generally use mixed supplies of surface and ground water. In the Marcellus Shale in Pennsylvania,
18    for example, most water used for hydraulic fracturing originates from surface water, whereas
19    surface and ground water are used in approximately equal proportions in the Barnett Shale in
20    Texas (see Figure 10-la,b). In areas that lack available surface water  (e.g., western Texas), ground
21    water supplies most of the water needed for hydraulic fracturing.

22    Across the United States, the vast majority of water used in hydraulic fracturing is fresh, although
23    operators also make use of lower-quality water, including reused hydraulic fracturing waste water.
24    Based on available data, the median reuse of hydraulic fracturing wastewater as a percentage of
25    injected volumes is 5% nationally, with the percentage varying by location.: Available data on reuse
26    trends indicate increased reuse of wastewater over time in both Pennsylvania and West Virginia.
27    Reuse as a percentage of injected volumes is lower in other areas, including regions with more
28    water stress, likely because of the availability of disposal wells. For example,  reused wastewater is
29    approximately 18% of injected volumes in the Marcellus Shale in Pennsylvania's Susquehanna
30    River Basin, whereas it is approximately 5% in the Barnett Shale in Texas (see Figure 10-la,b).
      i Reused wastewater as a percentage of injected volumes differs from the percentage of wastewater that is managed
      through reuse, as opposed to other wastewater management options. For example, in the Marcellus Shale in
      Pennsylvania, approximately 18% of injected water is reused produced water, while approximately 70% or more of
      wastewater is managed through reuse (see Figure 10-la).
                    This document is a draft for review purposes only and does not constitute Agency policy.
      June 2015                                     10-2                   DRAFT—DO NOT CITE OR QUOTE

-------
Hydraulic Fracturing Drinking Water Assessment
Chapter 10 - Syn th esis
                         Water Source
                                                   Well            Wastewater Management

                                                      420,000 -1.3 million gal produced d
                  • Surface Water   Reused Water  Ground Water
                                                                   Up to  10-30%
                                                                   15% e _
                                                                             60-90% '
                                                                   - Reuse   CWT • U 1C disposal well
                                                        Most of the injected fluid is retained
                                                        in the subsurface: Produced water
                                                        volumes over lOyearsare
                                                        approximately 10-30% of the original
                                                        injected fluid volume"
                         Water Source

                                 - 4.S million gal
                                                   Well
                                                                  Wastewater Management
                         47.5%
                                                        Produced
                                                        water volumes
                                                        over 3
                                                        years are
                                                        approximately
                                                        the same as
                                                        the original
                                                        injected fluid
                                                        volume"
                  • Surface Water   Reused Water  Ground Water -
                                                                             UIC disposal well
Figure 10-1. Water budgets representative of practices in the Marcellus Shale in the
            Susquehanna River Basin in Pennsylvania (a) and the Barnett Shale in Texas (b).

            Pie size and arrow thickness represent the relative volume of water as it flows through the hydraulic
            fracturing water cycle. Wastewater going to a centralized waste treatment (CWT) facility may be
            either discharged to surface water or reused. Wastewater going to an underground injection control
            (UIC) well is disposed of below ground. These examples represent typical water management practices
            as depicted for the most recent time period reviewed by this assessment. They do not represent any
            specific well. Sources for 10-la: (a) Table 4-1 (Hansen et al.,  2013); (b) Table 4-3 (U.S. EPA, 2015c); (c)
            Appendix Table B-5 (Hansen et al., 2013); (d) Table 7-2 [Ziemkiewicz etal.. 2014]—Note: produced
            water value from the West Virginia portion of the Marcellus; it provided the longest-term
            measurement of produced water volumes; (e) Figure 8-4 [PA PEP. 2015a] and Table 8-5 [Ma et al..
            2014: Shaffer etal.. 2013]. Sources for 10-lb: (a) Appendix Table B-5 (U.S. EPA, 2015a;  Nicot et al.,
            2012; Nicot etal., 2011); (b) Table 4-3 (Nicot et al., 2014); (c) Table 4-1 (Nicot et al., 2012); d: Table 7-2
            (Nicot etal., 2014); (e) Table 8-5 (Nicot et al., 2012);  (f) Calculated by subtracting reuse values from
            100% (see Table 8-5).
                This document is a draft for review purposes only and does not constitute Agency policy.

June 2015                                           10-3                      DRAFT—DO NOT CITE OR QUOTE

-------
      Hydraulic Fracturing Drinking Water Assessment                                       Chapter 10 - Synthesis


         •   How much water is used per well?
 1    The national median volume of water used per hydraulically fractured well is approximately 1.5
 2    million gal (5.7 million L), according to the EPA's analysis of FracFocus 1.0 disclosures. This
 3    estimate likely represents a wide variety of fractured well types, including vertical wells that
 4    generally use much less water per well than horizontal wells. Thus, published estimates for
 5    horizontal shale gas wells are typically higher (e.g., approximately 4 million gal (Vengosh etal.,
 6    2014]]. There is also wide variation within and among states and basins in the median water
 7    volumes used per well, from more than 5 million gal (19 million L] in Arkansas, Louisiana and West
 8    Virginia to less than 1 million gal (3.8 million L] in California, New Mexico, and Utah, among others.
 9    This variation results from  several factors, including well length, formation geology, and fracturing
10    fluid formulation.

         •   How might cumulative water withdrawals for hydraulic fracturing affect drinking water
             quantity?
11    Cumulatively, hydraulic fracturing uses billions of gallons of water each year at the national and
12    state scales, and even in some counties. As noted above, hydraulic fracturing water use and
13    consumption are generally  less than 1% of total annual water use and consumption at these scales.
14    However, there are a few counties in the United States where these percentages are higher. For
15    2011 and 2012, annual hydraulic fracturing water use was 10% or more compared to 2010 total
16    annual water use in 6.5% of counties with FracFocus 1.0 disclosures analyzed by the EPA, 30% or
17    more in 2.2% of counties, and 50% or more in 1.0% of counties. Consumption estimates followed
18    the same general pattern. For these counties, hydraulic fracturing is a relatively large user and
19    consumer of water.

20    High fracturing water use or consumption alone does not necessarily result in impacts to drinking
21    water resources. Rather, impacts result from the combination of water use/consumption and water
2 2    availability at local scales. In our survey of published literature, we did not find a case where
23    hydraulic fracturing water use or consumption alone caused a drinking water well or stream to run
24    dry. This could indicate an absence of effects or a lack of documentation in the literature we
25    reviewed. Additionally, water availability is rarely impacted by just one use or factor alone. In
26    Louisiana, for example, the  state requested hydraulic fracturing operations switch from ground to
27    surface water, due to concerns that ground water withdrawals for fracturing could, in combination
28    with other uses, adversely affect drinking water supplies.

29    The potential for impacts to drinking water resources from hydraulic fracturing water withdrawals
30    is highest in areas with relatively high fracturing water use and low water availability. Southern
31    and western Texas are two  locations where hydraulic fracturing water use, low water availability,
32    drought, and reliance on declining ground water has the potential to affect the quantity of drinking
33    water resources. Any impacts are likely to be realized locally within these areas. In a detailed case
34    study of southern Texas, Scanlonetal. (2014] observed generally adequate water supplies for
35    hydraulic fracturing, except in specific locations. They found excessive drawdown of local ground
36    water in a small proportion (approximately 6% of the area] of the Eagle Ford Shale. They suggested
                   This document is a draft for review purposes only and does not constitute Agency policy.
      June 2015                                    10-4                   DRAFT—DO NOT CITE OR QUOTE

-------
      Hydraulic Fracturing Drinking Water Assessment                                       Chapter 10 - Synthesis


 1    water management, particularly a shift towards brackish water use, could minimize potential future
 2    impacts to fresh water resources.

 3    The potential for impacts to drinking water quantity due to hydraulic fracturing water use appears
 4    to be lower—but not eliminated—in other areas of the United States. Future problems could arise if
 5    hydraulic fracturing increases substantially in areas with low water availability, or in times of water
 6    shortages. In detailed case studies in western Colorado and northeastern Pennsylvania, the EPA did
 7    not find current impacts, but did conclude that streams could be vulnerable to water withdrawals
 8    from hydraulic fracturing. In northeast Pennsylvania, water management, such as minimum stream
 9    flow requirements, limits the potential for impacts, especially in small streams. In western North
10    Dakota, ground water is limited, but the industry may have sufficient supplies of surface water from
11    the Missouri River system. These location-specific examples emphasize the need to focus on
12    regional and local dynamics when considering potential impacts of hydraulic fracturing water
13    acquisition on drinking water resources.

         •   What are the possible impacts of water withdrawals for hydraulic fracturing on water quality?
14    Water withdrawals for hydraulic fracturing, similar to all water withdrawals, have the potential to
15    alter the quality of drinking water resources. Ground water withdrawals exceeding natural
16    recharge rates decrease water storage in aquifers, potentially mobilizing contaminants or allowing
17    the infiltration of lower quality water from the land surface or adjacent formations. Withdrawals
18    could also decrease ground water discharge to streams, potentially affecting surface water quality.
19    Areas with large amounts of sustained ground water pumping are most likely to experience
2 0    impacts, particularly drought-prone regions with limited ground water recharge.

21    Surface water withdrawals also have the potential to affect water quality. Withdrawals may lower
22    water levels and alter stream flow, potentially decreasing a stream's capacity to dilute
23    contaminants. Case studies by the EPA show that streams can be vulnerable to changes in water
24    quality due to water withdrawals, particularly smaller streams and during periods of low flow.
2 5    Management of the rate and timing of surface water withdrawals has been shown to help mitigate
2 6    potential impacts of hydraulic fracturing withdrawals on water quality.

      10.1.2. Chemical Mixing (Chapter 5)
27    Hydraulic fracturing fluids are developed to perform specific functions, including: create and
2 8    extend fractures, transport proppant, and place proppant in the fractures. The fluid generally
29    consists of three parts: (1) the base fluid, which is the largest constituent by volume and is typically
30    water; (2) the additives, which can be a single chemical or a mixture of chemicals; and (3) the
31    proppant Additives are chosen to serve a specific purpose (e.g., adjust pH, increase viscosity, limit
32    bacterial growth). Chemicals generally comprise a small percentage (typically 2% or less) of the
33    overall injected fluid volume. Because over one million gallons of fluid are typically injected per
34    well, thousands of gallons of chemicals can be potentially stored on-site and used during hydraulic
35    fracturing activities.

3 6    On-site storage, mixing, and pumping of chemicals and hydraulic fracturing fluids have the potential
37    to result in accidental releases, such as spills or leaks. Potential impacts to drinking water resources

                   This document is a draft for review purposes only and does not constitute Agency policy.
      June 2015                                    10-5                   DRAFT—DO NOT CITE OR QUOTE

-------
      Hydraulic Fracturing Drinking Water Assessment                                       Chapter 10 - Synthesis


 1    from spills of hydraulic fracturing fluids and chemicals depend on the characteristics of the spills,
 2    and the fate, transport, and the toxicity of chemicals spilled.

         Research Questions: Chemical Mixing

         •   What is currently known about the frequency, severity, and causes of spills of hydraulic
             fracturing fluids and additives?
 3    The frequency of on-site spills from hydraulic fracturing could be estimated for two states, but not
 4    for operations nationally or for other areas. Frequency estimates from data and literature ranged
 5    from one spill for every 100 wells in Colorado to between approximately 0.4 and 12.2 spills for
 6    every 100 wells in Pennsylvania.1 These estimates include spills of hydraulic fracturing fluids and
 7    chemicals, and produced water reported in state databases. Available data generally precluded
 8    estimates of hydraulic fracturing fluid and/or chemical spill rates separately from estimates of an
 9    overall spill frequency. It is unknown whether these spill estimates are representative of national
10    occurrences. If the estimates are representative, the number of spills nationally could range from
11    approximately 100 to 3,700 spills annually, assuming 25,000 to 30,000 new wells are fractured per
12    year.

13    The EPA characterized volumes and causes of hydraulic fracturing-related spills identified from
14    selected state and industry data sources. The spills occurred between January 2006 and April 2012
15    in 11 states and included 151 cases in which fracturing fluids or chemicals spilled on or near a well
16    pad. Due to the methods used for the EPA's characterization of spills, these cases were likely a
17    subset of all fracturing fluid and chemical spills during the study's time period. The reported
18    volume of fracturing fluids or chemicals spilled ranged from 5 gal to more than 19,000 gal (19 to
19    72,000 L), with a median volume of 420 gal (1,600 L) per spill. Spill causes included equipment
20    failure, human error, failure of container integrity, and other causes (e.g., weather and vandalism).
21    The most common cause was equipment failure, specifically blowout preventer failure, corrosion,
22    and failed valves. More than 30% of the 151 fracturing fluid or chemical spills were from fluid
23    storage units (e.g., tanks, totes, and trailers).

         •   What are the identities and volumes of chemicals used in hydraulic fracturing fluids, and how
             might this composition vary at a given site and across the country?
24    In this assessment, we identified a list of 1,076 chemicals used in hydraulic fracturing fluids. This is
25    a cumulative list over multiple wells and years.  These chemicals include acids, alcohols, aromatic
26    hydrocarbons, bases, hydrocarbon mixtures, polysaccharides, and surfactants. According to the
27    EPA's analysis of disclosures to FracFocus 1.0, the number of unique chemicals per well ranged
28    from 4 to 28, with a median of 14 unique chemicals per well.

29    Our analysis indicates  that chemical use varies and that no single chemical is used at all well sites
30    across the country, although several chemicals are widely used. Methanol, hydrotreated light
      1 Spill frequency estimates are for a given number of wells over a given period of time. These are not annual estimates nor
      are they for the lifetime of a well.
                    This document is a draft for review purposes only and does not constitute Agency policy.
      June 2015                                     10-6                  DRAFT—DO NOT CITE OR QUOTE

-------
      Hydraulic Fracturing Drinking Water Assessment                                       Chapter 10 - Synthesis


 1    petroleum distillates, and hydrochloric acid were reported as used in 65% or more of wells,
 2    according to FracFocus 1.0 disclosures analyzed by the EPA. Only 32 chemicals, excluding water,
 3    quartz, and sodium chloride, were used in more than 10% of wells according to the EPA's analysis
 4    of FracFocus disclosures. The composition of hydraulic fracturing fluids varies by state, by well, and
 5    within the same service company and geologic formation. This variability likely results from several
 6    factors, including the geology of the formation, the availability and cost of different chemicals, and
 7    operator preference.

 8    Estimates from the EPA's database developed from FracFocus 1.0 suggest median volumes of
 9    individual chemicals injected per well range from a few gallons to thousands of gallons, with an
10    overall median of 650 gal (2,500 L) per chemical per well. Based on this overall median and
11    assuming 14 unique chemicals are used per well, an estimated 9,100 gal (34,000 L) of chemicals
12    may be injected per well. Given thatthe number of chemicals per well ranges from 4 to 28, the
13    estimated volume of chemicals injected per well may range from approximately 2,600 to 18,000 gal
14    (9,800 to 69,000 L).

          •   What are the chemical, physical, and toxicological properties of hydraulic fracturing chemical
             additives?
15    Measured or estimated physicochemical properties were obtained for 453 chemicals of the toal
16    1,076 chemicals reported in hydraulic fracturing fluids. We could not estimate physicochemical
17    properties for the inorganic chemicals or mixtures. The 453 chemicals have a wide range of
18    physicochemical properties.

19    Properties affecting the likelihood of a spilled chemical reaching and impacting a drinking water
20    resource, include: mobility, solubility, and volatility. Of the 453 chemicals for which
21    physicochemical properties were available, 18 of the top 20 most mobile ones were reported in the
22    EPA's FracFocus 1.0 database for 2% or less of wells. Choline chloride and tetrakis (hydroxymethyl)
23    phosphonium were exceptions and were reported in 14% and 11% of wells, respectively. These
24    two chemicals appear to be relatively more common, and, if spilled, would move quickly through
25    the environment with the flow of water. The majority of the 453 chemicals associate strongly with
26    soils and organic materials, suggesting the potential for these chemicals to persist in the
27    environment as long-term contaminants. Many of the 453 chemicals fully dissolve in water, but
2 8    their aqueous solubility varies greatly. Few of the chemicals volatilize, and thus a large proportion
29    of most hydraulic fracturing chemicals tend to remain in water.

30    Oral reference values and oral slope factors meeting the criteria used in this assessment were not
31    available for the majority of chemicals used in hydraulic fracturing fluids, representing a significant
32    data gap for hazard identification.u Reference values and oral slope factors are important for
      1 A reference value is an estimate of an exposure to the human population (including susceptible subgroups) for a given
      duration that is likely to be without an appreciable risk of adverse health effects over a lifetime. Reference value is a
      generic term not specific to a given route of exposure.
      2 An oral slope factor is an upper-bound, approximating 95% confidence limit, on the increased cancer risk from a lifetime
      oral exposure to an agent.
                    This document is a draft for review purposes only and does not constitute Agency policy.
      June 2015                                     10-7                   DRAFT—DO NOT CITE OR QUOTE

-------
      Hydraulic Fracturing Drinking Water Assessment                                       Chapter 10 - Synthesis


 1    understanding the potential human health effects resulting from exposure to a chemical. Chronic
 2    oral reference values and/or oral slope factors from selected federal, state, and international
 3    sources were available for 90 (8%) of the 1,076 chemicals used in hydraulic fracturing fluids. From
 4    U.S. federal sources alone, chronic oral reference values were available for 73 chemicals (7%) of the
 5    1,076 chemicals, and oral slope factors were available for 15 chemicals (1%). Of the 32 chemicals
 6    reported as used in at least 10% of wells in the EPA's FracFocus database (excluding water, quartz,
 7    and sodium chloride), seven (21%) have a federal chronic oral reference value. Oral reference
 8    values and oral slope factors are a key component of the risk assessment process, although
 9    comprehensive risk assessments that characterize the health risk associated with exposure to these
10    chemicals are not available.

11    Of the chemicals that had values available, the health endpoints associated with those values
12    include the potential for carcinogenesis, immune system effects, changes in body weight, changes in
13    blood chemistry, cardiotoxicity, neurotoxicity, liver and kidney toxicity, and reproductive and
14    developmental toxicity. However, it is important to note that evaluating any potential risk to human
15    populations would require knowledge of the specific chemicals that are present at a particular site,
16    whether or not humans are exposed to those chemicals and, if so, at what levels and for what
17    duration, and the toxicity of the chemicals. Since most chemicals are used infrequently on a
18    nationwide basis, potential exposure is likely to be a local or regional issue, rather than a national
19    issue. Accordingly, consideration of hazards and risks associated with these chemical additives
20    would be most useful on a site-specific basis and is beyond the scope of this assessment.

         •   If spills occur, how might hydraulic fracturing chemical additives contaminate drinking water
             resources?
21    There are several mechanisms by which a spill can potentially contaminate drinking water
22    resources. These include overland flow to nearby surface water, soil contamination and eventual
23    transport to surface water, and infiltration and contamination of underlying ground water. Of the
24    151 spills characterized by the EPA, fluids reached surface water in 13 (9% of 151) cases and soil in
25    97 (64%) cases. None of the spills of hydraulic fracturing fluid were reported to have reached
26    ground water. This could be due to an absence of impact; however, it can take several years for
27    spilled fluids to infiltrate soil and leach into ground water. Thus, it may not be immediately
28    apparent whether a spill has reached ground water or not

29    Based on the relative importance of each of these mechanisms, impacts have the potential to occur
30    quickly, be delayed short or long periods, or have a continual effect over time. In Kentucky, for
31    example, a spill impacted a surface water body relatively quickly when hydraulic fracturing fluid
32    entered a creek, significantly reducing the water's pH and increasing its conductivity (Papoulias
33    and Velasco.  20131

      10.1.3. Well  Injection (Chapter 6)
34    Hydraulic fracturing fluids are injected into oil or gas wells under high pressures.  The fluids flow
35    through the well (commonly thousands of feet below the surface) into the production zone (i.e., the
                   This document is a draft for review purposes only and does not constitute Agency policy.
      June 2015                                    10-8                   DRAFT—DO NOT CITE OR QUOTE

-------
      Hydraulic Fracturing Drinking Water Assessment                                      Chapter 10 - Synthesis


 1    geologic formation being fractured) where the fluid injection pressures are sufficient to create
 2    fractures in the rock.

 3    There are two major subsurface mechanisms by which the injection of fluid and the creation and
 4    propagation of fractures can lead to contamination of drinking water resources: (1) the unintended
 5    movement of liquids or gases out of the production well or along the outside of the production well
 6    into a drinking water resource via deficiencies in the well's casing or cement, and (2) the
 7    unintended movement of liquids or gases from the production zone through  subsurface geologic
 8    formations into a drinking water resource. Combinations of these two mechanisms are also
 9    possible.

        Research Questions: Well Injection

         •   How effective are current well construction practices at containing fluids- both liquids and
             gases - before, during, and after fracturing?
10    Production wells are constructed to access and convey hydrocarbons from the formations in which
11    they are found to the surface, and to isolate fluid-bearing zones (containing oil, gas, or water)  from
12    each other. Typically, multiple casings are emplaced and cemented along the wellbore to protect
13    and isolate the oil and/or natural gas from the formations it must travel through to reach the
14    surface.

15    Below ground drinking water resources are often separated from the production well using casing
16    and cement. Cemented surface casing, in particular, is an important well construction feature  for
17    isolating drinking water resources from liquids and gases that may move through the subsurface. A
18    limited risk modeling study of selected injection wells in the Williston Basin in North Dakota
19    suggests that the risk of aquifer contamination from leaks inside the well to the drinking water
20    resource decreases by a factor of approximately one thousand when surface  casing extends below
21    the bottom of the drinking water resource [Michie and Koch. 1991). Most wells used in hydraulic
22    fracturing operations have casing and a layer of cement to protect drinking water resources, but
23    there are exceptions: a survey conducted by the EPA of oil and gas production wells hydraulically
24    fractured by nine oil and gas service companies in 2009 and 2010 estimated that at least 3% of the
25    wells (600 out of 23,000 wells) did not have cement across a portion of the casing installed through
26    the protected ground water resource identified by well operators. The absence of cement does not
27    in and of itself lead to an impact However, it does reduce the overall number of casing and cement
28    barriers fluids must travel through to reach ground water resources.

29    Impacts to drinking water resources from subsurface liquid and gas movement may occur if casing
30    or cement are inadequately designed or constructed, or fail. There are several examples of these
31    occurrences in hydraulically fractured wells that have or may have resulted in impacts to drinking
3 2    water resources. In one example, an inner string of casing burst during hydraulic fracturing, which
3 3    resulted in a release of fluids on the land surface and possibly into the aquifer near Killdeer, North
34    Dakota. The EPA found that, based on the data analysis performed for the study, the only potential
3 5    source consistent with conditions observed in two impacted monitoring wells was the blowout that
                   This document is a draft for review purposes only and does not constitute Agency policy.
      June 2015                                    10-9                  DRAFT—DO NOT CITE OR QUOTE

-------
      Hydraulic Fracturing Drinking Water Assessment                                      Chapter 10 - Synthesis


 1    occurred during hydraulic fracturing [U.S. EPA, 2015)]. In other examples, inadequately cemented
 2    casing has contributed to impacts to drinking water resources. In Bainbridge, Ohio, inadequately
 3    cemented casing in a hydraulically fractured well contributed to the buildup of natural gas and high
 4    pressures along the outside of a production well. This ultimately resulted in movement of natural
 5    gas into local drinking water aquifers [Bair etal., 2010: ODNR, 2008]. In the Mamm Creek gas field
 6    in Colorado, inadequate cement placement in a production well allowed methane and benzene to
 7    migrate along the production well and through natural faults and fractures  to drinking water
 8    resources [Science Based Solutions LLC. 2014: Crescent. 2011: COGCC. 2004]. These cases illustrate
 9    how construction issues, sustained casing pressure, and the presence of natural faults and fractures
10    can work together to create pathways for fluids to migrate toward drinking water resources.

11    Fracturing older wells may also increase the potential for impacts to drinking water resources via
12    movement of liquids and gases from the inside of the production well or along the outside of the
13    production well to ground water resources. The EPA estimated that 6% of 23,000 oil and gas
14    production wells were drilled more than 10 years before being hydraulically fractured in 2009 or
15    2010. Although new wells can be designed to withstand the stresses associated with hydraulic
16    fracturing operations, older wells may not have been built or tested to the same specifications and
17    their reuse for this purpose could be of concern. Moreover, aging and use of the well can contribute
18    to casing degradation, which can be accelerated by exposure to corrosive chemicals such as
19    hydrogen sulfide, carbonic acid, and brines.

         •   Can subsurface migration of fluids- both liquids and gases- to drinking water resources occur,
             and what local geologic or artificial features might allow this?
20    Physical separation between the production zone and drinking water resources can help protect
21    drinking water. Many hydraulic fracturing operations target deep formations such as the Marcellus
22    Shale or the Haynesville Shale (Louisiana/Texas], where the vertical distance between the base of
23    drinking water resources and the top of the shale formation may be a mile or greater. Numerical
24    modeling and microseismic studies based on a Marcellus Shale-like environment suggest that
25    fractures created during hydraulic fracturing are unlikely to extend upward from these deep
26    formations into shallow drinking water aquifers.

27    Not all hydraulic fracturing is performed in zones that are deep below drinking water resources.
28    For example, operations in the Antrim Shale (Michigan] and the  New Albany Shale
29    (Illinois/Indiana/Kentucky] take place at shallower depths (100 to 1,900 ft or 30 to 579 m], with
30    less vertical separation between the formation and drinking water resources (NETL. 2013: GWPC
31    and ALL Consulting. 2009]. The EPA's survey of oil and gas production wells hydraulically fractured
32    by nine service companies in  2009 and 2010 estimated that 20% of 23,000  wells had less than
33    2,000 ft (610 m] of measured distance between the point of shallowest hydraulic fracturing and the
34    base of the protected ground water resources reported by well operators.

35    There are also places in the subsurface where oil and gas resources and drinking water resources
36    co-exist in the same formation. Evidence indicates that hydraulic fracturing occurs within these
37    formations. This results in the introduction of fracturing fluids into formations that may currently
38    serve, or in the future could serve, as a source of drinking water  for public or private use. According

                   This document is a draft for review purposes only and does not constitute Agency policy.
      June 2015                                   10-10                 DRAFT—DO NOT CITE OR QUOTE

-------
      Hydraulic Fracturing Drinking Water Assessment                                       Chapter 10 - Synthesis


 1    to the data examined, the overall frequency of occurrence of this practice appears to be low, with
 2    the activity generally concentrated in some areas in the western United States. The practice of
 3    injecting fracturing fluids into a formation that also contains a drinking water resource directly
 4    affects the quality of that water, since some of the fluid likely remains in the formation following
 5    hydraulic fracturing. Hydraulic fracturing in a drinking water resource is a concern in the short-
 6    term (should there be people currently using these zones as a drinking water supply) and the long-
 7    term (if drought or other conditions necessitate the future use of these zones for drinking water).

 8    Liquid and gas movement from the production zone to underground drinking water resources may
 9    also occur via other production wells or injection wells near hydraulic fracturing operations.
10    Fractures created during hydraulic fracturing can intersect nearby wells or their fracture networks,
11    resulting in the flow of fluids into those wells. These well communications, or "frac hits," are more
12    likely to occur if wells are close to each other or on the same well pad. In the Woodford Shale in
13    Oklahoma, the likelihood of well communication was less than 10% between wells more than 4,000
14    ft (1,219 m) apart, but rose to nearly 50% between wells less  than 1,000 ft (305 m) apart (Ajani and
15    Kelkar. 2012). If an offset well is not able to withstand the stresses applied during the hydraulic
16    fracturing of a neighboring well, well components may fail,  which  could result in a release of fluids
17    at the surface from the offset well. The EPA identified incidents in which surface spills of hydraulic
18    fracturing-related fluids were attributed to well communication events.

19    Older or inactive wells—including oil and gas wells, injection  wells, or drinking water wells—near
20    the hydraulic  fracturing operation may pose an even greater potential for impacts. A study in
21    Oklahoma found that older wells were more likely to be negatively affected by the stresses applied
22    by hydraulic fracturing in neighboring wells (Ajani and Kelkar. 2012). In some cases, inactive wells
23    in the vicinity of hydraulic fracturing activities may not have been plugged properly—many wells
24    plugged before the 1950s were done so with little or no cement. The Interstate Oil and Gas Compact
25    Commission estimates that over one million wells may have been  drilled in the United States prior
26    to a formal regulatory system being in place, and the status and location of many of these wells are
27    unknown (IOGCC, 2008). State programs exist to plug identified inactive wells, and work is on-
28    going to identify and address such wells.

      10.1.4. Flowback and Produced Water (Chapter 7)
29    Water, of variable quality, is a byproduct of oil and gas production. After hydraulic fracturing, the
30    injection pressure is released and water flows back from the well. Initially this water is similar to
31    the hydraulic  fracturing fluid, but as time goes on the composition is affected by the characteristics
32    of the formation and possible reactions between the formation and the fracturing fluid. Water
33    initially produced from the well after hydraulic fracturing is sometimes called flowback in the
34    literature, and the term appears in this assessment However, hydraulic fracturing fluids and any
35    formation water returning to the surface are often referred to collectively as produced water. This
36    definition of produced water is used in this assessment

37    The amount of produced water varies, but typically averages 10% to 25% of injected volumes,
38    depending upon the amount of time since fracturing and the particular well (see Figure 10-la).
39    However, there are exceptions to this, such as in the Barnett Shale in Texas where the total volume

                   This document is a draft for review purposes only and does not constitute Agency policy.
      June 2015                                    10-11                  DRAFT—DO NOT CITE OR QUOTE

-------
      Hydraulic Fracturing Drinking Water Assessment                                      Chapter 10 - Synthesis


 1    of produced water can equal or exceed the injected volume of hydraulic fracturing fluid (see Figure
 2    10-Ib). Flow rates are generally high initially, and then decrease over time throughout oil or gas
 3    production.

 4    Impacts on drinking water resources have the potential to occur if produced water is spilled and
 5    enters surface water or ground water. Environmental transport of chemical constituents in
 6    produced water depends on the characteristics of the spill (e.g., volume and duration), the
 7    composition of spilled fluids, and the characteristics of the surrounding environment.

         Research Questions: Flow/back and Produced Water

         •   What is currently known about the frequency, severity, and causes of spills of flow/back and
             produced water?
 8    Surface spills of produced water from hydraulically fractured wells have occurred. As noted in the
 9    Chemical Mixing section above, the frequency of on-site spills from hydraulic fracturing activities
10    could be estimated for two states, but not nationally. Estimates of spill frequencies at hydraulic
11    fracturing sites in Colorado and Pennsylvania, including spills of produced water, ranged from
12    approximately 0.4 to 12.2 spills per 100 wells. Available data generally precluded estimates of
13    produced water spill rates separately from estimates of overall spill frequency. Away from the well,
14    produced water spills from pipelines and truck transport also have the potential to impact drinking
15    water resources.

16    The EPA characterized spill volumes and causes for 225 cases in which produced water spilled on
17    or near a well pad. These spills occurred between January 2006 and April 2012 in 11 states. The
18    median reported volume per produced water spill was 990 gallons (3,750 L), more than double that
19    for spills of hydraulic fracturing fluids and chemicals. The causes of produced water  spills were
20    reported as human error, equipment failure, container integrity failure, miscellaneous causes (e.g.,
21    well communication), and unknown causes. Most of the total volume spilled  (74%) for all 225 cases
2 2    combined was caused by a failure of container integrity.

         •   What is the composition of hydraulic fracturing flowback and produced water,  and what
             factors might influence this composition?
23    A combination of factors influence the composition of produced water, including: the composition
24    of injected hydraulic fracturing fluids, the type of formation fractured, subsurface processes, and
25    residence time. The initial chemical composition of produced water primarily reflects the chemistry
26    of the injected fluids. At later times, the chemical composition of produced water reflects the
2 7    geochemistry of the fractured formation.

28    Produced water varies in quality from fresh to highly saline, and can contain high levels of major
29    anions and cations, metals, organics, and naturally occurring radionuclides. Produced water from
30    shale and tight gas formations typically contains high levels of total dissolved solids  (TDS) and ionic
31    constituents (e.g., bromide, calcium, chloride, iron, potassium, manganese, magnesium, and
32    sodium). Produced water also may contain metals (e.g., barium, cadmium, chromium, lead, and
                   This document is a draft for review purposes only and does not constitute Agency policy.
      June 2015                                    10-12                 DRAFT—DO NOT CITE OR QUOTE

-------
      Hydraulic Fracturing Drinking Water Assessment                                      Chapter 10 - Synthesis


 1    mercury), and organic compounds such as benzene. Produced water from coalbed methane
 2    typically has much lower TDS levels compared to other produced water types, particularly if the
 3    coalbed was deposited under fresh water conditions.

 4    We identified 134 chemicals that have been detected in hydraulic fracturing produced water. These
 5    include chemicals added during the chemical mixing stage, as well as naturally occurring organic
 6    chemicals and radionuclides, metals, and other constituents of subsurface rock formations
 7    mobilized by the hydraulic fracturing process. Data on measured chemical concentrations in
 8    produced water were available for 75 of these  134 chemicals.

 9    Most of the available data on produced water content are for shale and coalbed methane
10    formations, while less data are available for tight formations, such as sandstones. The composition
11    of produced water must be determined through sampling and analysis, both of which have
12    limitations—the former due to challenges in accessing production equipment, and the latter due to
13    difficulties identifying target analytes before analysis and the lack of appropriate analytical
14    methods. Most current data are for inorganic chemicals, while less data exist for organic chemicals.
15    Many more  organic chemicals were reported as used in hydraulic fracturing fluid than have been
16    identified in produced water. The difference may be due  to analytical limitations, limited study
17    scopes, and undocumented subsurface reactions.

         •   What are the chemical, physical, and toxicological properties of hydraulic fracturing flowback
             and produced water constituents?
18    The identified constituents of produced water include inorganic chemicals (cations and anions, i.e.,
19    metals, metalloids, non-metals, and radioactive materials), organic chemicals and compounds, and
20    unidentified materials measured as total organic carbon  and dissolved  organic carbon. Some
21    constituents are readily transported with water (i.e., chloride and bromide), while others depend
22    strongly on the geochemical conditions in the receiving water  body (i.e., radium and barium), and
23    assessment of their transport is based on site-specific factors.  We were able to obtain actual or
24    estimated physicochemical properties for 86 (64%) of the 134 chemicals identified in produced
2 5    water.

26    As in the case of chemicals in hydraulic fracturing fluid, chemical properties that affect the
2 7    likelihood of an organic chemical in produced water reaching and impacting drinking water
28    resources include: mobility, solubility, and volatility. In general, physicochemical properties suggest
29    that organic chemicals in produced water tend to be less mobile in the environment. Consequently,
30    if spilled, these chemicals may remain in soils or sediments near spill sites. Low mobility may result
31    in smaller dissolved contaminant plumes in ground water, although these chemicals can be
32    transported with sediments  in surface water or small particles in ground water.  Organic chemical
33    properties vary with salinity, and effects depend on the nature of the chemical.

34    Oral reference values and/or oral slope factors from selected federal, state, and international
35    sources were available for 83 (62%) of the  134 chemicals detected in produced water. From U.S.
36    federal sources alone, chronic oral reference values were available for 70 (52%) of the 134
37    chemicals, and oral slope factors were available for 20 chemicals (15%). Of the chemicals that had

                   This document is a draft for review purposes only and does not constitute Agency policy.
      June 2015                                   10-13                 DRAFT—DO NOT CITE OR QUOTE

-------
      Hydraulic Fracturing Drinking Water Assessment                                      Chapter 10 - Synthesis


 1    values available, noted health effects include the potential for carcinogenesis, immune system
 2    effects, changes in body weight, changes in blood chemistry, pulmonary toxicity, neurotoxicity, liver
 3    and kidney toxicity, and reproductive and developmental toxicity. As noted above, evaluating any
 4    potential risk to human populations would require knowledge of the specific chemicals that are
 5    present at a particular site, whether or not humans are exposed to those chemicals and, if so, at
 6    what levels and for what duration, and the toxicity of the chemicals. The chemicals present in
 7    produced water can vary based on the formation and specific well, due to differences in fracturing
 8    fluid formulation and formation geology. Accordingly, consideration of hazards and risks associated
 9    with these chemicals would be most useful on a site-specific basis and is beyond the scope of this
10    assessment  .

         •   // sp;7/s occur, how might hydraulic fracturing flowback and produced water contaminate
             drinking water resources?
11    Impacts to drinking water resources from spills or releases of produced water depend on the
12    volume, timing, and composition of the produced water. Impacts are more likely the greater the
13    volume of the spill, the longer the duration of the release, and the higher the concentration of
14    produced water constituents (i.e., salts, naturally occurring radioactive material, and metals).

15    The EPA characterization of hydraulic fracturing-related spills found that 8% of the 225 produced
16    water spills included in the study reached surface water or ground water. These spills tended to be
17    of greater volume than spills that did not reach a water body. A well blowout in Bradford County,
18    Pennsylvania spilled an estimated 10,000  gal (38,000 L) of produced water into a tributary of
19    Towanda Creek, a state-designated trout fishery. The largest volume spill identified in this
20    assessment occurred in North Dakota, where approximately 2.9 million gal (11 million  L) of
21    produced water spilled from a broken pipeline and impacted surface and ground water.

22    Chronic releases can and do occur from produced water disposed in unlined pits or impoundments,
23    and can have long-term impacts. Ground water impacts may persist longer than surface water
24    impacts because of lower flow rates and decreased mixing. Plumes from unlined pits used for
25    produced water have been shown to persist for long periods and extend to nearby surface water
26    bodies.

      10.1.5. Wastewater Management and Waste Disposal (Chapter 8)
27    Hydraulic fracturing generates large volumes of produced water that require management. In this
2 8    section we refer to produced water and any other waters generated onsite by the single term
29    "wastewater." Clark and Veil (2009) estimated that in 2007 approximately one million active oil
30    and gas wells in the United States generated 2.4 billion gal per day (9.1  billion L per day) of
31    wastewater. There is currently no reliable way to estimate what fraction of this total volume can be
32    attributed to hydraulically fractured wells. Wastewater volumes in a region can increase sharply as
33    hydraulic fracturing activity increases.

34    Wastewater management and disposal could affect drinking water resources through multiple
3 5    mechanisms including: inadequate treatment of wastewater prior to discharge to a receiving water,
3 6    accidental releases during transport or leakage from wastewater storage pits, unpermitted

                   This document is a draft for review purposes only and does not constitute Agency policy.
      June 2015                                   10-14                  DRAFT—DO NOT CITE OR QUOTE

-------
      Hydraulic Fracturing Drinking Water Assessment                                      Chapter 10 - Synthesis


 1    discharges, migration of constituents in wastewaters following land application, inappropriate
 2    management of residual materials from treatment, or accumulation of wastewater constituents in
 3    sediments near outfalls of centralized waste treatment facilities (CWTs) or publicly owned
 4    treatment works (POTWs) that have treated hydraulic fracturing wastewater. The scope of this
 5    assessment excludes potential impacts to drinking water from the disposal of hydraulic fracturing
 6    wastewater in underground injection control (UIC) wells.

        Research Questions: Wastewater Management and Waste Disposal

         •   What are the common treatment and disposal methods for hydraulic fracturing wastewater,
             and where are these methods practiced?
 7    Hydraulic fracturing wastewater is managed using several options including disposal in UIC wells
 8    (also called disposal wells); through evaporation ponds; treatment at CWTs, followed by reuse or
 9    by discharge to either surface waters or POTWs; reuse with minimal or no treatment; and land
10    application or road spreading. Treatment of hydraulic fracturing wastewater by POTWs was used in
11    the past in Pennsylvania. This decreased sharply following new state-level requirements and a
12    request by the Pennsylvania Department of Environmental Protection (PA DEP) for well operators
13    to stop sending Marcellus Shale wastewater to POTWs (and 15 CWTs) discharging to surface
14    waters.

15    Wastewater management decisions are generally based on the availability and associated costs
16    (including transportation) of disposal or treatment facilities. A survey of state agencies found that,
17    in 2007, more than 98% of produced water from the oil and gas industry was managed via
18    underground injection (Clark and Veil. 2009). Available information suggests that disposal wells are
19    also the primary management practice for hydraulic fracturing wastewater in most regions in the
20    United States (e.g., the Barnett Shale; see Figure 10-lb). The Marcellus Shale region is a notable
21    exception, where most wastewater is reused because  of the small number of disposal wells in
22    Pennsylvania (see Figure 10-la). Although this assessment does not address potential effects on
2 3    drinking water resources from the use of disposal wells, any changes in cost of disposal or
24    availability of disposal wells would likely influence wastewater management decisions.

2 5    Wastewater from some hydraulic fracturing operations is sent to CWTs, which may discharge
26    treated wastewater to surface waters, POTWs, or back to well operators for reuse in other
2 7    hydraulic fracturing operations. Available  data indicate that the use of CWTs for treating hydraulic
28    fracturing wastewater is greater in the Marcellus Shale region than other parts of the country. Most
29    of the CWTs accepting hydraulic fracturing wastewater in Pennsylvania cannot significantly reduce
30    TDS, and many of these facilities provide treated wastewater to well operators for reuse and do not
31    currently discharge treated wastewater to surface water.

3 2    Reuse of wastewater for subsequent hydraulic fracturing operations may require no treatment,
33    minimal treatment, or more extensive treatment. Operators reuse a substantial amount (ca. 70-
34    90%) of Marcellus Shale wastewater in Pennsylvania  (see Figure 10-la). Lesser amounts of reuse
                   This document is a draft for review purposes only and does not constitute Agency policy.
      June 2015                                   10-15                  DRAFT—DO NOT CITE OR QUOTE

-------
      Hydraulic Fracturing Drinking Water Assessment                                      Chapter 10 - Synthesis


 1    occur in other areas (e.g., the Barnett Shale; see Figure 10-lb). In certain formations, such as the
 2    Bakken Shale in North Dakota, there is currently no indication of appreciable reuse.

 3    In some cases, wastewater is used for land applications such as irrigation or road spreading for
 4    deicing or dust suppression. Land application has the potential to introduce wastewater
 5    constituents to surface water and ground water due to runoff and migration of brines. Studies of
 6    road spreading of conventional oil and gas brines have found elevated levels of metals in soils and
 7    chloride in ground water.

         •   How effective are conventional POTWs and commercial treatment systems in removing
             organic and inorganic contaminants of concern in hydraulic fracturing wastewater?
 8    Publicly owned treatment works using basic treatment processes are not designed to effectively
 9    reduce TDS concentrations in highly saline hydraulic  fracturing wastewater—although specific
10    constituents or constituents groups can be removed (e.g., metals, oil, and grease by chemical
11    precipitation or other processes). In some cases, wastewater treated at CWTs may be sent to a
12    POTW for additional treatment and discharge. It is blended with POTW influent to prevent
13    detrimental effects on biological processes in the POTW that aid  in the treatment of wastewater.

14    Centralized waste treatment facilities with advanced wastewater treatment options,  such as
15    reverse osmosis, thermal distillation, or mechanical vapor recompression, reduce TDS
16    concentrations and can treat contaminants currently  known to be in hydraulic fracturing
17    wastewater. However, there are limited data on the composition of hydraulic fracturing
18    wastewater, particularly for organic constituents. It is unknown  whether advanced treatment
19    systems are effective at removing constituents that are generally not tested for.

         •   What are the potential impacts from surface water disposal of treated hydraulic fracturing
             wastewater on drinking water treatment facilities?
2 0    Potential  impacts to drinking water resources may occur if hydraulic fracturing wastewater is
21    inadequately treated and discharged to surface water. Inadequately treated hydraulic fracturing
22    wastewater may increase concentrations of TDS, bromide, chloride, and iodide in receiving waters.
23    In particular, bromide and iodide are precursors of disinfection byproducts (DBFs) that can form in
24    the presence of organic carbon in drinking water treatment plants or wastewater treatment plants.
25    Drinking water treatment plants are required to monitor for certain types of DBFs, because some
2 6    are toxic and can cause cancer.

2 7    Radionuclides can also be found in inadequately treated hydraulic fracturing wastewater from
28    certain shales, such as the Marcellus. A recent study by the PADEP (2015b] found elevated radium
29    concentrations in the tens to thousands of picocuries  per liter and gross alpha and gross beta in the
30    hundreds to thousands of picocuries per liter in effluent samples from some CWTs receiving oil and
31    gas wastewater. Radium, gross alpha, and gross beta were also detected in effluents from POTWs
32    receiving oil and gas wastewater (mainly as effluent from CWTs), though at lower concentrations
33    than from the CWTs. Research in Pennsylvania also indicates the accumulation of radium in
34    sediments and soils affected by the outfalls of some treatment plants that have handled oil and gas
35    wastewater, including Marcellus Shale wastewater, and other wastewaters (PADEP.  2015b:

                   This document is a draft for review purposes only and does not constitute Agency policy.
      June 2015                                   10-16                  DRAFT—DO NOT CITE OR QUOTE

-------
      Hydraulic Fracturing Drinking Water Assessment                                       Chapter 10 - Synthesis


 1    Warner etal., 2013a]. Mobilization of radium from sediments and potential impacts on downstream
 2    water quality depend upon how strongly the radium has sorbed to sediments. Impacts may also
 3    occur if sediment is resuspended (e.g., following storm events). There is no evidence of radionuclide
 4    contamination in drinking water intakes due to inadequately treated hydraulic fracturing
 5    wastewater.

 6    Hydraulic fracturing wastewaters contain other constituents such as barium, boron, and heavy
 7    metals. Barium in particular has been documented in some shale gas produced waters. Little data
 8    exist on metal and organic compound concentrations in untreated and treated wastewaters in
 9    order to evaluate whether treatment is effective, and whether there  are potential downstream
10    effects on drinking water resources when wastewater is treated and discharged.

      10.2.Key Data Limitations and Uncertainties
11    This assessment used available data and literature to examine the potential impacts of hydraulic
12    fracturing for oil and gas on drinking water resources nationally. As  part of this effort, we identified
13    data limitations and uncertainties associated with current information on hydraulic fracturing and
14    its potential to affect drinking water resources. In particular, data limitations preclude a
15    determination of the frequency of impacts with any certainty. There is a high degree of uncertainty
16    about whether the relatively few instances of impacts noted  in this report are the result of a rarity
17    of effects or a lack of data. These limitations and uncertainties are discussed in brief below.

      10.2.1. Limitations in monitoring data  and chemical information
18    While many activities conducted as part of the hydraulic fracturing water cycle take place above
19    ground, hydraulic fracturing itself occurs below ground and is not directly observable. Additionally,
20    potential mechanisms identified in this assessment may result in impacts to drinking water
21    resources that are below ground (e.g., spilled fluids leaching into ground water). Because of this,
22    monitoring data are needed before, during, and after hydraulic fracturing to characterize the status
23    of the well being fractured and the presence, migration, or transformation of chemicals in the
24    subsurface. These data can include results from mechanical integrity tests performed on
25    hydraulically fractured oil and gas production wells and data on local water quality collected pre-
26    and post-hydraulic fracturing. In particular, baseline data on local water quality is needed to
27    quantify changes to drinking water resources and to provide insights into whether nearby
28    hydraulic fracturing activities may have caused any detected changes. The limited amount of data
29    collected before and during hydraulic fracturing activities reduces the ability to determine whether
30    hydraulic fracturing affected drinking water resources in cases of alleged contamination.

31    Water quality testing for hydraulic fracturing-related chemicals is routinely conducted for a small
32    subset of chemicals reportedly used in hydraulic fracturing fluids or detected in produced water.
33    Public water systems regularly test for selected contaminants under the National Primary Drinking
34    Water Regulations. Approximately 6% of the 1,173 chemicals in Table A-2 and Table A-4 are
                   This document is a draft for review purposes only and does not constitute Agency policy.
      June 2015                                    10-17                  DRAFT—DO NOT CITE OR QUOTE

-------
      Hydraulic Fracturing Drinking Water Assessment                                        Chapter 10 - Synthesis


 1    routinely tested for under these regulations.: Private water wells are usually tested less often and
 2    for fewer potential contaminants than public water supplies [USGS. 2014c]. Since chemical use
 3    varies widely across the country, testing for any particular chemical may or may not be appropriate
 4    for detecting potential impacts on a drinking water resource from a nearby hydraulic fracturing
 5    operation. Furthermore, the concentration, mobility, and detectability (as determined by the lowest
 6    concentration that an analytical method is able to determine a chemical's presence) of chemicals
 7    used in or produced by hydraulic fracturing operations will affect whether or not it would be
 8    identified in a drinking water resource in the event of its release into the environment.

 9    Information (identity, frequency of use, physicochemical and toxicological properties, etc.) on the
10    chemicals associated with the hydraulic fracturing water cycle is not complete and limits
11    understanding of potential impacts on drinking water resources. Well operators identified one or
12    more chemicals as confidential in approximately 70% of wells reported to FracFocus 1.0 and
13    analyzed by the EPA (U.S. EPA.  2015a). Additionally, chemicals found in flowback and produced
14    water (see Table A-4) were identified for a limited number of geographic locations and formations.
15    These characterization studies are constrained by available methods for detecting organic and
16    inorganic compounds in flowback and produced water. The identity of hydraulic fracturing-related
17    chemicals is necessary to understand their chemical, physical, and toxicological properties, which
18    determine how they might move through the environment to  drinking water resources and any
19    resulting effects. Knowing their identities would also help inform what chemicals to test for in the
2 0    event of suspected drinking water impacts and, in the case of wastewater, may help predict
21    whether current treatment systems are effective at removing  them.

22    Peer reviewed toxicity data for known hydraulic fracturing-related chemicals is very limited. Of the
23    1,173 hydraulic fracturing-related chemicals identified in Appendix A, 147 have chronic oral
24    reference values and/or oral slope factors from the sources that met the selection criteria  for
25    inclusion in this assessment  Because the majority of chemicals identified in this report do not have
26    chronic oral reference values and/or oral slope factors, risk assessors  at the local and regional level
27    may need to use alternative sources of toxicity information that could  introduce greater
28    uncertainties. It also makes an assessment of potential cumulative effects of exposure to chemical
29    mixtures in hydraulic fracturing fluid, flowback, or produced water difficult
      1 We identified 73 chemicals that are reported to be used in hydraulic fracturing fluids (see Table A-2) or that have been
      detected in produced water (see Table A-4) that are tested for as part of the contaminant monitoring conducted for 40
      different drinking water standards under the National Primary Drinking Water Regulations (NPDWR). For inorganic
      chemicals regulated under the NPDWR, we identified the chemical or element itself, its regulated ion (as applicable), or
      other more complex forms on the list of hydraulic fracturing-related chemicals. For regulated organic chemicals, we
      identified only the chemical itself on the list of hydraulic fracturing-related chemicals with three exceptions: (1) we
      identified all four trihalomethanes that comprise total trihalomethanes, (2) we identified two of the five regulated
      chlorinated/brominated haloacetic acids as their sodium salts, and (3) we identified a subset of polychlorinated biphenyls
      (PCBs) as Aroclor 1248. Although various forms of petroleum distillates are used in hydraulic fracturing fluids and may
      contain BTEX or benzo(a)pyrene (the regulated entities that can occur naturally in petroleum distillates), we did not
      include them in our count of 73 chemicals.
                    This document is a draft for review purposes only and does not constitute Agency policy.
      June 2015                                     10-18                   DRAFT—DO NOT CITE OR QUOTE

-------
      Hydraulic Fracturing Drinking Water Assessment                                      Chapter 10 - Synthesis


      10.2.2. Other Contributing Limitations
 1    We found other limitations that hamper our ability to assess the potential impacts of hydraulic
 2    fracturing on drinking water resources nationally. These include the number and location of
 3    hydraulically fractured wells, the location of drinking water resources, and information on industry
 4    practices and any changes that may take place in practices in the coming years. Our estimates of the
 5    number of fractured wells are based on an evaluation of several commercial and public sources and
 6    a number of assumptions. This lack of a definitive well count particularly contributes to
 7    uncertainties regarding total water use or total wastewater volume estimates, and would limit any
 8    kind of cumulative impact assessment

 9    There are also some fundamental gaps in our understanding of drinking water resources, including
10    where they are located in relation to hydraulic fracturing operations and which might be most
11    vulnerable to impacts from hydraulic fracturing activities. Improving our assessment of potential
12    drinking water impacts requires better information, particularly about private drinking water well
13    locations and the depths of drinking water resources in relation to the hydraulically fractured
14    formations and well construction features (e.g., casing and cement). This information would allow
15    us to better assess whether subsurface drinking water resources are isolated from hydraulically
16    fractured oil and gas production wells.

17    Finally, this assessment summarizes available information on industry practices with respect to the
18    hydraulic fracturing water cycle. While some information on hydraulic fracturing activities is
19    available for many areas of the United States, specific data on water withdrawals for hydraulic
2 0    fracturing, volumes of flowback and produced water generated, and the disposal or reuse of
21    wastewaters is needed to better characterize potential impacts of hydraulic fracturing on drinking
22    water resources. Additionally, industry practices are rapidly changing (e.g., the number of wells
23    fractured, the location of activities, and the chemicals used), and it is unclear how changes in
24    industry practices could affect potential drinking water impacts in the future. Consideration of
2 5    future development scenarios was not a part of this assessment, but such an evaluation could help
2 6    establish potential short- and long-term impacts to drinking water resources and how to assess
27    them.

      10.3.Conclusions
28    Through this national-level assessment, we have identified potential mechanisms by which
29    hydraulic fracturing could affect drinking water resources. Above ground mechanisms can affect
30    surface and ground water resources and include water withdrawals at times or in locations of low
31    water availability, spills of hydraulic fracturing fluid and chemicals or produced water, and
3 2    inadequate treatment and discharge of hydraulic fracturing wastewater. Below ground mechanisms
3 3    include movement of liquids and gases via the production well into underground drinking water
34    resources and movement of liquids and gases from the fracture zone to these resources via
3 5    pathways in subsurface rock formations.

36    We did not find evidence that these mechanisms have led to widespread, systemic impacts on
37    drinking water resources in the United States. Of the potential mechanisms identified in this report,
                   This document is a draft for review purposes only and does not constitute Agency policy.
      June 2015                                   10-19                 DRAFT—DO NOT CITE OR QUOTE

-------
      Hydraulic Fracturing Drinking Water Assessment                                       Chapter 10 - Synthesis


 1    we found specific instances where one or more of these mechanisms led to impacts on drinking
 2    water resources, including contamination of drinking water wells. The cases occurred during both
 3    routine activities and accidents and have resulted in impacts to surface or ground water. Spills of
 4    hydraulic fracturing fluid and produced water in certain cases have reached drinking water
 5    resources, both surface and ground water. Discharge of treated hydraulic fracturing waste water has
 6    increased contaminant concentrations in receiving surface waters. Below ground movement of
 7    fluids, including gas, most likely via the production well, have contaminated drinking water
 8    resources. In some cases, hydraulic fracturing fluids have also been directly injected into drinking
 9    water resources, as defined in this assessment, to produce oil or gas that co-exists in those
10    formations.

11    The number of identified cases where drinking water resources were impacted are small relative to
12    the number of hydraulically fractured wells. This could reflect a rarity of effects on drinking water
13    resources, or may be an underestimate as a result of several factors. There is insufficient pre- and
14    post-hydraulic fracturing data on the quality of drinking water resources. This inhibits a
15    determination of the frequency of impacts. Other limiting factors include the presence of other
16    causes of contamination, the  short duration of existing studies, and inaccessible information related
17    to hydraulic fracturing activities.

      10.4.Use of the Assessment
18    The practice of hydraulic fracturing is simultaneously expanding and changing rapidly. Over 60% of
19    new oil and gas wells are likely to be hydraulically fractured, and this percentage may be over 90%
20    in some locations. Economic  forces are likely to cause short term volatility in the number of wells
21    drilled and fractured, yet hydraulic fracturing is expected to continue to expand and drive an
22    increase in domestic oil and gas production in coming decades [EIA. 2014a].1 As a result, hydraulic
23    fracturing will likely increase in existing locations, while also  potentially expanding to new areas.

24    This state-of-the-science assessment contributes to the understanding of the potential impacts of
2 5    hydraulic fracturing on drinking water resources and the factors that may influence those impacts.
26    The findings in this assessment can be used by federal, state, tribal, and local officials; industry; and
2 7    the public to better understand and address any vulnerabilities of drinking water resources to
28    hydraulic fracturing activities. This assessment can also be used to help facilitate and inform
29    dialogue among interested stakeholders, and support future efforts, including: providing context to
30    site-specific exposure or risk assessments, local and regional public health assessments, and to
31    assessments of cumulative impacts of hydraulic fracturing on drinking water resources over time
32    or over defined geographic areas of interest

33    We hope the identification of limitations and uncertainties will promote greater attention to these
34    areas through pre- and post- hydraulic fracturing monitoring programs and by researchers. We also
      1 In their reference case projections, the U.S. Energy Information Administration (EIA) forecasts that U.S. gas production
      by 2035 will have increased 50% over 2012 levels. Crude oil production is projected to increase almost 40% above
      current levels by 2025, before declining in subsequent decades [EIA. 2014a).
                    This document is a draft for review purposes only and does not constitute Agency policy.
      June 2015                                     10-20                  DRAFT—DO NOT CITE OR QUOTE

-------
     Hydraulic Fracturing Drinking Water Assessment                                        Chapter 10 - Synthesis


1    hope it will lead to greater dissemination of data in forms accessible by a wide-range of researchers
2    and audiences.

3    Finally, and most importantly, this assessment advances the scientific basis for decisions by federal,
4    state, tribal, and local officials; industry; and the public, on how best to protect drinking water
5    resources now and in the future.

     10.5.References for Chapter 10

     Ajani. A: Kelkar. M. (2012). Interference study in shale plays. Paper presented at SPE Hydraulic Fracturing
        Technology Conference, February 6-8, 2012, The Woodlands, TX.

     Bair. ES: Freeman. DC: Senko. IM. (2010). Subsurface gas invasion Bainbridge Township, Geauga County, Ohio.
        (Expert Panel Technical Report). Columbus, OH: Ohio Department of Natural Resources.
        http://oilandgas.ohiodnr.gov/resources/investigations-reports-violations-reformstfTHR

     Clark. CE: Veil. IA. (2009). Produced water volumes and management practices in the United States (pp. 64).
        (ANL/EVS/R-09/1). Argonne, IL: Argonne National Laboratory.
        http://www.circleofblue.org/waternews/wp-
        content/uploads/2010/09/ANL EVS  R09 produced water volume report 2437.pdf

     Crescent (Crescent Consulting, LLC). (2011). East Mamm creek project drilling and cementing study.
        Oklahoma City, OK. http://cogcc.state.co.us/Library/PiceanceBasin/EastMammCreek/ReportFinal.pdf

     EIA (Energy Information Administration). (2014a). Annual energy outlook 2014 with projections to 2040.
        (DOE/EIA-0383(2014)). Washington, D.C.: U.S. Energy Information Administration.
        http://www.eia.gov/forecasts/aeo/pdf/0383(2014).pdf

     GWPC and ALL Consulting (Ground Water Protection Council (GWPC) and ALL Consulting). (2009). Modern
        shale gas development in the United States: A primer. (DE-FG26-04NT15455). Washington, DC: U.S.
        Department of Energy, Office of Fossil  Energy and National Energy Technology Laboratory.
        http://www.gwpc.org/sites/default/files/Shale%20Gas%20Primer%202009.pdf

     Hansen. E: Mulvaney. D: Betcher. M. (2013). Water resource reporting and water footprint from Marcellus
        Shale development in West Virginia and Pennsylvania. Durango, CO: Earthworks Oil & Gas Accountability
        Project, http://www.downstreamstrategies.com/documents/reports publication/marcellus wv pa.pdf

     IOGCC (Interstate Oil and Gas Compact Commission). (2008). Protecting our country's resources: The states'
        case, orphaned well plugging initiative. Oklahoma City, OK: Interstate Oil and Gas Compact Commission
        (IOGCC). http://iogcc.myshopify.com/products/protecting-our-countrys-resources-the-states-case-
        orphaned-well-plugging-initiative-2008

     Ma. G: Geza. M: Xu. P. (2014). Review of flowback and produced water management, treatment, and beneficial
        use for major shale gas development basins. Shale Energy Engineering Conference 2014, Pittsburgh,
        Pennsylvania, United States.

     Michie. TW: Koch. CA. (1991). Evaluation of injection-well risk management in the Williston Basin. J Pet Tech
        43: 737-741. http://dx.doi.org/10.2118/20693-PA

     NETL (National Energy Technology Laboratory). (2013). Modern shale gas development in the United States:
        An update. Pittsburgh, PA: U.S. Department of Energy. National Energy Technology Laboratory.
        http://www.netl.doe.gov/File%20Librarv/Research/Oil-Gas/shale-gas-primer-update-2013.pdf
                   This document is a draft for review purposes only and does not constitute Agency policy.

     June 2015                                      10-21                   DRAFT—DO NOT CITE OR QUOTE

-------
Hydraulic Fracturing Drinking Water Assessment                                         Chapter 10 - Synthesis
Nicot. IP: Hebel. AK: Ritter. SM: Walden. S: Baier. R: Galusky. P: Beach. I: Kyle. R: Symank. L: Breton. C. (2011).
   Current and projected water use in the Texas mining and oil and gas industry - Final Report. (TWDB
   Contract No. 0904830939). Nicot, JP; Hebel, AK; Ritter, SM; Walden, S; Baier, R; Galusky, P; Beach, J; Kyle,
   R; Symank, L; Breton, C.
   http://www.twdb.texas.gov/publications/reports/contracted reports/doc/0904830939 MiningWaterUs
   e.pdf

Nicot. IP: Reedy. RC: Costley. RA: Huang. Y. (2012). Oil & gas water use in Texas: Update to the 2011 mining
   water use report. Nicot, JP; Reedy, RC; Costley, RA; Huang, Y.
   http://www.twdb.state.tx.us/publications/reports/contracted reports/doc/0904830939 2012Update M
   iningWaterUse.pdf

Nicot. IP: Scanlon. BR: Reedy. RC: Costley. RA. (2014). Source and fate of hydraulic fracturing water in the
   Barnett Shale: a historical perspective. Environ Sci Technol 48: 2464-2471.
   http://dx.doi.org/10.1021/es404050r

ODNR. DMRM. (Ohio Department of Natural Resources, Division of Mineral Resources Management). (2008).
   Report on the investigation of the natural gas invasion of aquifers in Bainbridge Township of Geauga
   County, Ohio. Columbus, OH: ODNR.
   http://oilandgas.ohiodnr.gov/portals/oilgas/pdf/bainbridge/report.pdf

PA PEP (Pennsylvania Department of Environmental Protection). (2015a). PA DEP oil & gas reporting
   website, statewide data downloads by reporting period, waste and production files downloaded for
   Marcellus/unconventional wells, July 2009 December 2014. Harrisburg, PA. Retrieved from
   https://www.paoilandgasreporting.state.pa.us/publicreports/Modules/DataExports/DataExports.aspx

PA DEP (Pennsylvania Department of Environmental Protection). (2015b). Technologically enhanced
   naturally occurring radioactive materials (TENORM) study report. Harrisburg, PA.
   http://www.elibrary.dep.state.pa.us/dsweb/Get/Document-105822/PA-DEP-TENORM-
   Study Report Rev. 0  01-15-2015.pdf

Papoulias. DM: Velasco. AL. (2013). Histopathological analysis offish from Acorn Fork Creek, Kentucky,
   exposed to hydraulic fracturing fluid releases. Southeastern Naturalist 12: 92-111.

Scanlon. BR: Reedy. RC: Nicot. IP. (2014). Will water scarcity in semiarid regions limit hydraulic fracturing of
   shale plays? Environmental Research Letters 9. http://dx.doi.Org/10.1088/1748-9326/9/12/124011

Science Based Solutions LLC.  (2014). Summary of hydrogeology investigations in the Mamm Creek field area,
   Garfield County. Laramie, Wyoming. http://www.garfield-county.com/oil-gas/documents/Summary-
   Hydrogeologic-Studies-Mamm%2 OCreek-Area-Feb-10-2014.pdf

Shaffer. PL: Arias Chavez. LH: Ben-Sasson. M: Romero-Vargas Castrillon. S: Yip. NY: Elimelech. M.  (2013).
   Desalination and reuse of high-salinity shale gas produced water: drivers, technologies, and future
   directions. Environ Sci Technol 47: 9569-9583.

U.S. EPA (U.S. Environmental  Protection Agency). (2014c). Drinking water contaminants. Available online at
   http://water.epa.gov/drink/contaminants/

U.S. EPA (U.S. Environmental  Protection Agency). (2015a). Analysis of hydraulic fracturing fluid data from the
   FracFocus chemical disclosure registry 1.0 [EPA Report]. (EPA/601/R-14/003). Washington, D.C.: Office
   of Research and Development, U.S. Environmental Protection Agency.
   http://www2.epa.gov/hfstudy/analysis-hydraulic-fracturing-fluid-data-fracfocus-chemical-disclosure-
   registry-1-pdf

U.S. EPA (U.S. Environmental  Protection Agency). (2015b). Analysis of hydraulic fracturing fluid data from the
   FracFocus chemical disclosure registry 1.0: Project database  [EPA Report]. (EPA/601/R-14/003).
   Washington, D.C.: U.S. Environmental Protection Agency, Office of Research and Development.
   http://www2.epa.gov/hfstudv/epa-proiect-database-developed-fracfocus-l-disclosures
               This document is a draft for review purposes only and does not constitute Agency policy.

June 2015                                       10-22                   DRAFT—DO NOT CITE OR QUOTE

-------
Hydraulic Fracturing Drinking Water Assessment                                          Chapter 10 - Synthesis
U.S. EPA (U.S. Environmental Protection Agency). (2015c). Case study analysis of the impacts of water
   acquisition for hydraulic fracturing on local water availability [EPA Report]. (EPA/600/R-14/179).
   Washington, D.C.

U.S. EPA (U.S. Environmental Protection Agency). (2015J). Retrospective case study in Killdeer, North Dakota:
   study of the potential impacts of hydraulic fracturing on drinking water resources [EPA Report]. (EPA
   600/R-14/103). Washington, D.C.

USGS (U.S. Geological Survey). (2014c). The quality of our nations waters water quality in principal aquifers
   of the United States, 19912010. (Circular 1360). Reston, VA. http://dx.doi.org/10.3133/cirl360

Vengosh, A: lackson, RB: Warner, N: Darrah, TH: Kondash, A. (2014). A critical review of the risks to water
   resources from unconventional shale gas development and hydraulic fracturing in the United States.
   Environ Sci Technol 48: 36-52. http://dx.doi.org/10.1021/es405118y

Warner. NR: Christie. CA: lackson. RB: Vengosh. A. (2013a). Impacts of shale gas wastewater disposal on
   water quality in western Pennsylvania. Environ Sci Technol 47:11849-11857.
   http://dx.doi.org/10.1021/es402165b

Ziemkiewicz. P:  Ouaranta. ID: Mccawley. M. (2014). Practical measures for reducing the risk of environmental
   contamination in shale energy production. Environ Sci Process Impacts 16:1692-1699.
   http://dx.doi.org/10.1039/c3em00510k
               This document is a draft for review purposes only and does not constitute Agency policy.

June 2015                                       10-23                    DRAFT—DO NOT CITE OR QUOTE

-------
Hydraulic Fracturing Drinking Water Assessment                                               All References
References for All Chapters
Abou-Sayed. IS: Sorrell. MA: Foster. RA: Atwood. EL: Youngblood. DR. (2011). Haynesville shale development
   program: From vertical to horizontal. Paper presented at North American Unconventional Gas Conference
   and Exhibition, June 14-16, 2011, The Woodlands, TX.

Abriola. LM: Finder. GF. (1985a). A multiphase approach to the modeling of porous-media contamination by
   organic-compounds .2. Numerical-simulation. Water Resour Res 21:19-26.

Abriola. LM: Finder. GF. (1985b). A multiphase approach to the modeling of porous media contamination by
   organic compounds: 1. Equation development. Water Resour Res 21:11-18.
   http://dx.doi.org/10.1029/WR021i001p00011

Adachi. I: Siebrits. E: Peirce. A: Desroches. I. (2007). Computer simulation of hydraulic fractures. International
   Journal of Rock Mechanics and Mining Sciences 44: 739-757.
   http://dx.doi.0rg/10.1016/j.ijrmms.2006.ll.006

Ajani. A: Kelkar. M. (2012). Interference study in shale plays. Paper presented at SPE Hydraulic Fracturing
   Technology Conference, February 6-8, 2012, The Woodlands, TX.

Al-Ghazal. M: Al-Driweesh. S: Al-Shammari. F. (2013). First successful application of an environment friendly
   fracturing fluid during on-the-fly proppant fracturing. Paper presented at International Petroleum
   Technology Conference, March, 26-28, 2013, Beijing, China.

Alanco. (2012). New subsidiary Alanco Energy Services, Inc. to provide produced water disposal services to
   natural gas industry. Alanco. http://www.alanco.com/news 040912.asp

Ali. M: Taoutaou. S: Shafqat. AU: Salehapour. A:  Noor. S. (2009). The use of self healing cement to ensure long
   term zonal isolation for HPHT wells subject to hydraulic fracturing operations in Pakistan. Paper
   presented at International Petroleum Technology Conference, December 7-9, 2009, Doha, Qatar.

Ali. SA: Clark. Wl: Moore. WR: Dribus. IR. (2010). Diagenesis and reservoir quality. Oilfield Rev 22:14-27.

ALL Consulting (ALL Consulting, LLC). (2004). Coal bed methane primer: New source of natural gas and
   environmental implications. Tulsa, OK: U.S. Department of Energy, National Petroleum Technology Center.
   http://bogc.dnrc.mt.gov/PDF/Web%20Version.pdf

ALL Consulting (ALL Consulting, LLC). (2012). The modern practices of hydraulic fracturing: A focus on
   Canadian resources. Tulsa, Oklahoma: ALL Consulting LLC.

Allen. TE. (2013). Pregel blender prototype designed to reduce cost and environmental problems. (SPE-
   27708-MS). Allen, TE.  http://dx.doi.org/10.2118/27708-MS

Alley. B: Beebe. A: Rodgers. I: Castle. IW. (2011). Chemical and physical characterization of produced waters
   from conventional and unconventional fossil fuel resources. Chemosphere 85: 74-82.
   http://dx.doi.0rg/10.1016/j.chemosphere.2011.05.043

Allison. D: Folds. PS: Harless. PI: Howell. M: Vargus. GW: Stipetich. A. (2009).  Optimizing openhole completion
   techniques for horizontal foam-drilled wells. Paper presented at SPE Eastern Regional Meeting, September
   23-25,2009, Charleston, WV.

Alvarez-Cohen. L: Speitel. GE. (2001). Kinetics of aerobic cometabolism of chlorinated solvents.
   Biodegradation 12:105-126. http://dx.doi.Org/10.1023/A:1012075322466

Alzahrani. S: Mohammad. AW: Hilal. N: Abdullah. P: laafar. 0. (2013). Comparative study of NF and RO
   membranes in the treatment of produced water-Part I: Assessing water quality. Desalination 315: 18-26.
   http://dx.doi.0rg/10.1016/j.desal.2012.12.004


              This document is a draft for review purposes only and does not constitute Agency policy.
June 2015                                        1                     DRAFT—DO NOT CITE OR QUOTE

-------
Hydraulic Fracturing Drinking Water Assessment                                                All References
AMEC. Hinckley.. HDR. (AMEC Environment & Infrastructure, Inc, Hinckley Consulting, HDR Engineering, Inc).
   (2014). Hydrogeologic study of the Laramie County control area. Prepared for the Wyoming State
   Engineers Office. Cheyenne, WY: Wyoming State Engineer's Office, http://seo.wyo.gov/seo-
   files/Final%20Draft%20Corrected%20Stamped.pdf?attredirects=0&d=l

Aminto. A: Olson. MS.  (2012). Four-compartment partition model of hazardous components in hydraulic
   fracturing fluid additives. Journal of Natural Gas Science & Engineering 7: 16-21. http://dx.doi.org/DOI:
   10.1016/j.jngse.2012.03.006

ANRC (Arkansas Natural Resources Commission). (2014). Non-riparian water use certification. Available
   online at http://anrc.ark.org/divisions/water-resources-management/non-riparian-water-use-
   certification-program/

API (American Petroleum Institute). (2000). Overview of exploration and production waste volumes and
   waste management practices in the United States, http://www.api.org/environment-health-and-
   safety/environmental-performance/~/media/Files/EHS/Environmental  Performance/ICF-Waste-
   Survey-of-EandP-Wastes-2000.ashx

API (American Petroleum Institute). (2005). Modeling study of produced water release scenarios.
   (Publication Number 4734). Washington, D.C.: API Publishing Services.
   http://www.api.Org/~/media/files/ehs/environmental performance/4734.pdf?la=en

API (American Petroleum Institute). (2010a). Isolating potential flow zones during well construction
   [Standard] (1st ed.). (RP 65-2). Washington, DC: API Publishing Services.
   http://www.techstreet.com/products/preview/1695866

API (American Petroleum Institute). (2010b). Water management associated with hydraulic fracturing.
   Washington, D.C.: API Publishing Services.
   http://www.api.Org/~/media/Files/Policy/Exploration/HF2 el.pdf

API (American Petroleum Institute). (2011). Practices for mitigating surface impacts associated with
   hydraulic fracturing. Washington, DC: API Publishing Services.
   http://www.shalegas.energy.gov/resources/HF3 e7.pdf

Argonne National Laboratory. (2014). Water use and management in the Bakken shale  oil play. (DOE Award
   No.: FWP 49462). Pittsburgh, PA: National Energy Technology Laboratory.
   http://www.ipd.anl.gov/anlpubs/2014/05/104645.pdf

Arkadakskiy. S..: Rostron.  Bl. (2012a). Stable isotope geochemistry helps in reducing out-of-zone hydraulic
   fracturing and unwanted brine production from the Bakken Reservoir. Available online at
   http://isobrine.com/wp-content/uploads/2013/04/Arkadakskiy Rostron 20121.pdf

Arkadakskiy, S: Rostron, B. (2012b). Stable isotope geochemistry helps in reducing out-of-zone hydraulic
   fracturing and unwanted brine production from the Bakken Reservoir. Available online at
   http://isobrine.com/resources/

Arkadakskiy. S: Rostron. B. (2013a). Tracking out-of-zone hydraulic fracturing in the Bakken with naturally
   occuring tracers. Paper presented at GeoConvention 2013: Integration Canadian Society of Petroleum
   Geologists, May 6-10, 2013, Calgary, AB, Canada.

Arkadakskiy. S: Rostron. B. (2013b). Tracking out-of-zone hydraulic fracturing in the Bakken with naturally
   occuring tracers. Paper presented at GeoConvention 2013: Integration, May 6-10, 2013, Calgary, Alberta.

Arthur. ID: Bohm. B: Cornue. D. (2009a). Environmental considerations of modern shale gas development.
   Paper presented at SPE Annual Technical Conference and Exhibition, October 4-7, 2009, New Orleans, LA.

Arthur. ID: Bohm. B: Coughlin. Bl: Layne. M. (2009b). Evaluating the environmental implications of hydraulic
   fracturing in shale gas reservoirs. Paper presented at SPE Americas E&P Environmental and Safety
   Conference, March 23-25,2009, San Antonio, TX.
               This document is a draft for review purposes only and does not constitute Agency policy.

June 2015                                        2                     DRAFT—DO NOT CITE OR QUOTE

-------
Hydraulic Fracturing Drinking Water Assessment                                                All References
Arthur. ID: Bohm. B: Layne. M. (2008). Hydraulic fracturing considerations for natural gas wells of the
   Marcellus shale. Paper presented at The Ground Water Protection Council Annual Forum, September 21-
   24, 2008, Cincinnati, OH.

Arthur. ID: Layne. MA: Hochheiser. HW: Arthur. R. (2014). Spatial and statistical analysis of hydraulic
   fracturing activities in U.S. shale plays and the effectiveness of the FracFocus chemical disclosure system.
   In 2014 SPE hydraulic fracturing technology conference. Richardson, TX: Society of Petroleum Engineers.
   http://dx.doi.org/10.2118/168640-MS

ATSDR (Agency for Toxic Substances and Disease Registry). (2005). Public health assessment guidance
   manual (Update). Atlanta, GA: U.S. Department of Health and Human Services, Public Health Service.
   http://www.atsdr.cdc.gov/hac/PHAManual/toc.html

ATSDR (Agency for Toxic Substances and Disease Registry). (2013). Health consultation: Chesapeake ATGAS
   2H well site, Leroy Hill Road, Leroy, Leroy Township, Bradford County, PA. Atlanta, GA: ATSDR, Division of
   Cummunity Health Investigations.
   http://www.atsdr.cdc.gov/HAC/pha/ChesapeakeATGASWellSite/ChesapeakeATGASWellSiteHC1028201
   3 508.pdf

AWWA (American Water Works Association). (2013). Water and hydraulic fracturing: A white paper from the
   American Water Works Association. Denver, CO.
   http://www.awwa.Org/Portals/0/files/legreg/documents/AWWAFrackingReport.pdf

AWWA (American Water Works Association). (1999). Residential end uses of water. In PW Mayer; WB
   DeOreo (Eds.). Denver, CO: AWWA Research Foundation and American Water Works Association.
   http://www.waterrf.org/PublicReportLibrary/RFR90781 1999 241A.pdf

Ayala. S: Barber. T: Dessinges. MN: Frey. M: Horkowitz. I: Leugemors. E: Pessin. I: Way. CH: Badry. R: Kholy. IE:
   Gait. A: Hjellesnet. M: Sock. D. (2006). Improving oilfield service efficiency. Houston, TX: Schlumberger
   Limited.
   http://www.slb.eom/~/media/Files/resources/oilfield review/ors06/aut06/improving oilfield service
   efficiency.pdf

Bacher. D. (2013). Oil company fined $60,000 for illegally discharging tracking fluid. Available online at
   https://www.indybay.org/newsitems/2013/ll/17/18746493.php?show comments=l (accessed March
   6,2015).

Bachu. S: Bennion. DB. (2009). Experimental assessment of brine and/or C02 leakage through well cements
   at reservoir conditions. Int J Greenhouse Gas Control 3: 494-501.
   http://dx.doi.0rg/10.1016/j.ijggc.2008.ll.002

Bair. ES: Digel.  RK. (1990). Subsurface transport of inorganic and organic solutes from experimental road
   spreading of oilfield brine. Ground Water Monitoring and Remediation 10: 94-105.

Bair. ES: Freeman. DC: Senko. IM. (2010). Subsurface gas invasion Bainbridge Township, Geauga County, Ohio.
   (Expert Panel Technical Report). Columbus, OH: Ohio Department of Natural Resources.
   http://oilandgas.ohiodnr.gov/resources/investigations-reports-violations-reformstfTHR

Baker Hughes.  (2014a).  Baker Hughes green chemicals program overview. Available online

Baker Hughes.  (2014b). Well count. U.S. onshore well count [Database]. Houston, TX: Baker Hughes, Inc.
   Retrieved from http://phx.corporate-ir.net/phoenix.zhtml?c=79687&p=irol-wellcountus

Baldassare. F. (2011). The origin of some natural gases in Permian through Devonian Age systems in the
   Appalachian Basin and the relationship to incidents of stray gas migration. Presentation presented at
   Technical workshop for hydraulic fracturing study, chemical and analytical methods, February2425,2011,
   Arlington, VA.
               This document is a draft for review purposes only and does not constitute Agency policy.

June 2015                                        3                     DRAFT—DO NOT CITE OR QUOTE

-------
Hydraulic Fracturing Drinking Water Assessment                                                All References
Baldassare. Fl: McCaffrey. MA: Harper. IA. (2014). A geochemical context for stray gas investigations in the
   northern Appalachian Basin: Implications of analyses of natural gases from Neogene-through Devonian-
   age strata. AAPG Bulletin 98: 341-372. http://dx.doi.org/10.1306/06111312178

Banasiak. LI: Schafer. AI. (2009). Removal of boron, fluoride and nitrate by electrodialysis in the presence of
   organic matter. J Memb Sci 334:101-109. http://dx.doi.Org/10.1016/j.memsci.2009.02.020

Barati. R: Liang. IT.  (2014). A review of fracturing fluid systems used for hydraulic fracturing of oil and gas
   wells. J Appl  Polymer Sci Online pub. http://dx.doi.org/10.1002/app.40735

Barbot. E: Vidic. NS: Gregory. KB: Vidic. RD. (2013). Spatial and temporal correlation of water quality
   parameters of produced waters from Devonian-age shale following hydraulic fracturing. Environ Sci
   Technol 47: 2562-2569.

Barker. IF: Fritz. P. (1981). Carbon isotope fractionation during microbial methane oxidation. Nature 293:
   289-291. http://dx.doi.org/10.1038/293289aO

Barnes. T. (2010). 2 drillers fined for Pennsylvania gas well blowout. Available online at http://www.post-
   gazette.com/news/environment/2010/07/14/2-drillers-fined-for-Pennsylvania-gas-well-
   blowout/stories/201007140241 (accessed March 3,2015).

Bartos. TT: Hallberg. LL. (2011). Generalized potentiometric surface, estimated depth to water, and estimated
   saturated thickness of the high plains aquifer system, Marchjune 2009, Laramie County, Wyoming.
   Available online at http://pubs.usgs.gov/sim/3180/

Beckwith. R. (2011). Proppants: Where in the world. J Pet Tech 63: 36-41.

Bene. PG: Harden. B: Griffin. SW: Nicot. IP. (2007). Northern Trinity/Woodbine aquifer groundwater
   availability model: Assessment of groundwater use in the Northern Trinity aquifer due to urban growth
   and Barnett shale development. (TWDB Contract Number: 0604830613). Austin, TX: R. W. Harden &
   Associates, Inc.
   http://www.twdb.state.tx.us/groundwater/models/gam/trnt n/TRNT N Barnett Shale Reportpdf

Benko. KL: Drewes. IE. (2008). Produced water in the Western United States: Geographical distribution,
   occurrence, and composition. Environ Eng Sci 25: 239-246.

Benotti. Ml: Stanford. BD: Snyder. SA. (2010). Impact of drought on wastewater contaminants in an urban
   water supply. J Environ Qual 39:1196-1200.

Bertoncello. A: Wallace. I: Honarpour. MM: Kabir. C: Blyton. CA. (2014). Imbibition and water blockage in
   unconventional  reservoirs: Well management implications during flowback and early production. SPE
   Journal 17.

Bethke. CM.: Yeakel. S. (2014). The geochemists workbench. Release 10.0. GWB essentials guide (Version
   Release 10.0). Champaign, II: Aqueous Solutions, LLC. Retrieved from
   http://www.gwb.com/pdf/GWB10/GWBessentials.pdf

Bibby. Kl: Brantley. SL: Reible. DP: Linden. KG: Mouser. PI: Gregory. KB: Ellis. BR: Vidic. RD. (2013). Suggested
   reporting parameters for investigations of wastewater from unconventional shale gas  extraction. Environ
   Sci Technol 47:13220-13221. http://dx.doi.org/10.1021/es404960z

Bishop. RE. (2013). Historical analysis of oil and gas well plugging in New York: Is the regulatory system
   working? New Solutions: A Journal of Environmental and Occupational Health Policy 23:103-116.
   http://dx.doi.Org/10.2190/NS.23.l.g

Bl Services Company.  (2009). BJ fracturing manual 2.0. service company confidential business information
   document (Revision No. 1 ed. ed.). Houston, TX.

Blauch. ME. (2010). Developing effective and environmentally suitable fracturing fluids using hydraulic
   fracturing flowback waters. Paper presented at SPE Unconventional Gas Conference, February 23-25,
   2010, Pittsburgh, PA.


               This document is a draft for review purposes only and does not constitute Agency policy.
June 2015                                        4                     DRAFT—DO NOT CITE OR QUOTE

-------
Hydraulic Fracturing Drinking Water Assessment                                               All References
Blauch. ME: Myers. RR: Moore. TR: Lipinski. BA. (2009). Marcellus shale post-frac flowback waters - where is
   all the salt coming from and what are the implications? In Proceedings of the SPE Eastern Regional
   Meeting. Richardson, TX: Society of Petroleum Engineers.

BLM (Bureau of Land Management). (2013a). Abandoned mine lands: A new legacy. Washington, DC: U.S.
   Department of the Interior, Bureau of Land Management.
   http://www.blm.gov/pgdata/etc/medialib/blm/wo/MINERALS  REALTY  AND RESOURCE PROTECTIO
   N /ami/ami documents.Par.81686.File.dat/AML NewLegacy.pdf

BLM (Bureau of Land Management). (2013b).  Hydraulic fracturing white paper, appendix e. Casper, WY:
   Bureau of Land Management, Wyoming State Office.
   http://www.blm.gOV/pgdata/etc/medialib/blm/wy/information/NEPA/og/2014/02feb.Par.49324.File.d
   at/vlAppE.pdf

Blondes. MS: Cans. KD: Thordsen. II: Reidy. ME: Thomas. B: Engle. MA: Kharaka. YK: Rowan. EL. (2014). Data:
   U.S. Geological Survey National Produced Waters Geochemical Database v2.0 (Provisional) [Database]:
   U.S. Geological Survey:: USGS. Retrieved from
   http://energy.usgs.gov/EnvironmentalAspects/EnvironmentalAspectsofEnergyProductionandUse/Produ
   cedWaters.aspx#3822349-data

Boghici. R. (2009). Water quality in the Carrizo-Wilcox aquifer, 19902006. (Report 372). Austin, TX: Texas
   Water Development Board.
   http://www.twdb.texas.gov/publications/reports/numbered reports/doc/R372Carrizo-Wilcox.pdf

Borrirukwisitsak. S: Keenan. HE: Gauchotte-Lindsay. C. (2012). Effects of salinity, pH and temperature on the
   octanol-water partition coefficient of bisphenol A. IJESD 3: 460-464.
   http://dx.doi.org/10.7763/IIESD.2012.V3.267

Boschee. P. (2012). Handling produced water from hydraulic fracturing. Oil and Gas Facilities 1: 23-26.

Boschee. P. (2014). Produced and flowback water recycling and reuse: Economics, limitations, and
   technology. Oil and Gas Facilities 3:16-22.

Bosworth. S: El-Saved. HS: Ismail. G: Ohmer. H: Stracke. M: West. C: Retnanto. A. (1998). Key issues in
   multilateral technology. Oilfield Rev 10:14-28.

Bouchard. D: Cornaton. F: Hohener. P: Hunkeler. D. (2011). Analytical modelling of stable isotope
   fractionation of volatile organic compounds in the unsaturated zone. J Contam Hydrol 119: 44-54.
   http://dx.doi.0rg/10.1016/j.jconhyd.2010.09.006

Bouska. V. (1981). Geochemistry of coal. New York, NY: Elsevier.

Bowen. ZH: Oelsner. GP: Cade. BS: Gallegos. TI: Farag. AM: Mott. DN: Potter. Cl: Cinotto. PI: Clark. ML: Kappel.
   WM: Kresse. TM: Melcher. CP: Paschke. SS: Susong. DP: Varela. BA. (2015). Assessment of surface water
   chloride and conductivity trends in areas of unconventional oil and gas development-Why existing
   national data sets cannot tell us what we would like to know. Water Resour  Res 51: 704-715.
   http://dx.doi.org/10.1002/2014WR016382

Boyd. D: Al-Kubti. S: Khedr. OH: Khan. N: Al-Nayadi. K: Degouy. D: Elkadi. A: Kindi. ZA. (2006). Reliability of
   cement bond log interpretations compared to physical communication tests between formations. Paper
   presented at Abu Dhabi International Petroleum Exhibition and Conference, November 5-8, 2006, Abu
   Dhabi, UAE.

Boyer. CM: Glenn. SA: Claypool. BR: Weida. SD: Adams. ID: Huck. PR: Stidham. IE. (2005). Application of
   viscoelastic fracturing fluids in Appalachian Basin reservoirs (SPE 98068 ed.). Richardson, TX: Society of
   Petroleum Engineers. http://dx.doi.org/10.2118/98068-MS

Boyer. EW: Swistock. BR: Clark. I: Madden. M:  Rizzo. DE. (2011). The impact of Marcellus gas drilling on rural
   drinking water supplies. Harrisburg, PA: The Center for Rural Pennsylvania.
   http://www.iogawv.com/Resources/Docs/Marcellus-drinking-water-2011.pdf
               This document is a draft for review purposes only and does not constitute Agency policy.

June 2015                                        5                    DRAFT—DO NOT CITE OR QUOTE

-------
Hydraulic Fracturing Drinking Water Assessment                                                All References
Boysen. DB: Boysen. IA: Boysen. IE. (2002). Creative Strategies for Produced Water Disposal in the Rocky
   Mountain Region. Paper presented at 9th Annual International Petroleum Environmental Conference,
   October 2002, Albuquerque, NM.

Boysen. IE: Harju. IA: Shaw. B: Fosdick. M: Grisanti. A: Sorensen. IA. (1999). The current status of commercial
   deployment of the freeze thaw evaporation treatment of produced water. Richardson, TX: Society of
   Petroleum Engineers. http://dx.doi.org/10.2118/52700-MS

Brannon, HP: Daulton, PI: Hudson, HG: lordan, AK. (2011). Progression toward implementation of
   environmentally responsible fracturing processes (SPE-147534-MS ed.). Denver, CO: Society of Petroleum
   Engineers. http://dx.doi.org/10.2118/147534-MS

Brannon. HP: Daulton. PI: Post. MA: Hudson. HG: lordan. AK. (2012). The quest to exclusive use of
   environmentally responsible fracturing products and systems. Richardson, TX: Society of Petroleum
   Engineers. http://www.onepetro.org/mslib/app/Preview.do?paperNumber=SPE-152068-
   MS&societyCode=SPE

Brannon. HP: Kendrick. DE: Luckey. E: Stipetich. A. (2009). Multistage Fracturing of Horizontal Shale Gas
   Wells Using >90% Foam Provides Improved Production. In 2009 SPE Eastern regional meeting: limitless
   potential/formidable challenges (SPE 124767 ed.). Richardson, TX: Society of Petroleum Engineers.
   http://dx.doi.org/10.2118/124767-MS

Brannon. HP: Pearson. CM. (2007). Proppants and fracture conductivity. In Modern fracturing - enhancing
   natural gas production (1st ed.). Houston, TX: Energy Tribune Publishing, Inc.

Brantley, SL: Yoxtheimer, D: Arjmand, S: Grieve, P: Vidic, R: Pollak, I: Llewellyn, GT: Abad, I: Simon, C. (2014).
   Water resource impacts during unconventional shale gas development: The Pennsylvania experience. Int J
   Coal Geol 126:140-156. http://dx.doi.Org/10.1016/i.coal.2013.12.017

Breit. GN. (2002). USGS Produced waters database. Available online at
   http://energy.cr.usgs.gov/prov/prodwat/

Brinck. EL: Frost. CD. (2007). Detecting infiltration and impacts of introduced water using strontium isotopes.
   Ground Water 45: 554-568. http://dx.doi.Org/10.llll/i.1745-6584.2007.00345.x

Brown, DG: laffe, PR. (2001). Effects of nonionic surfactants on bacterial transport through porous media.
   Environ Sci Technol 35: 3877-3883. http://dx.doi.org/10.1021/es010577w

Brown, HP: Grijalva, VE: Raymer, LL. (1970). New developments in sonic wave train display and analysis in
   cased holes. (SPWLA-1970-F). Brown, HD; Grijalva, VE; Raymer, LL.
   https://www.onepetro.org/conference-paper/SPWLA-1970-F

Browne. PI: BD. L. (1999). The development of a rapid hydration on-the-fly crosslinked water fracturing fluid.
   Paper presented at CSPG and Petroleum Society Joint Convention, Digging Deeper, Finding a Better
   Bottom Line, June 14 18,1999, Calgary, Canada.

Brufatto. C: Cochran. I: Conn. L: El-Zeghaty. SZA. A: Fraboulet. B: Griffin. T: lames. S: Munk. T: lustus.  F:  Levine.
   IR: Montgomery. C: Murphy. D: Pfeiffer. I: Pornpoch. T: Rishmani. L. (2003). From mud to cement -
   Building gas wells. Oilfield Rev 15: 62-76.

Bruff. M:  likich. SA. (2011). Field demonstration of an integrated water treatment technology solution in
   Marcellus shale. Paper presented at SPE Eastern Regional  Meeting, August 17-19,2011, Columbus,  OH.

Byrnes. AP. (2011). Role of induced and natural imbibition in frac fluid transport and fate in gas shales.
   Presentation presented at Technical Workshops for Hydraulic Fracturing Study: Fate & Transport, March
   28-29,2011, Arlington, VA.
               This document is a draft for review purposes only and does not constitute Agency policy.

June 2015                                        6                     DRAFT—DO NOT CITE OR QUOTE

-------
Hydraulic Fracturing Drinking Water Assessment                                               All References
California Department of Conservation. (2015). Monthly production and injection databases, statewide
   production and injection data [Database]. Sacramento, CA: California Department of Conservation,
   Division of Oil, Gas & Geothermal Resources. Retrieved from
   http://www.conservation.ca.gov/dog/prod injection db/Pages/Index.aspx

California Department of Water Resources. (2015). California state water project overview. Available online
   athttp://www.water.ca.gov/swp/ (accessed February 20,2015).

Calvert, DG: Smith, DK. (1994). Issues and techniques of plugging and abandonment of oil and gas wells.
   Paper presented at SPE Annual Technical Conference and Exhibition, September 25-28,1994, New
   Orleans, LA.

Camacho. LM. ar: Dumee. L: Zhang. I: Li. I:  Duke. M: Gomez. I: Gray. S. (2013). Advances in membrane
   distillation for water desalination and purification applications. Water 5: 94-196.
   http://dx.doi.org/10.3390/w5010094

Caniglia. I. (2014). Youngstown contractor sentenced to 28 months for dumping fracking waste. Available
   online at http://www.cleveland.com/court-
   justice/index.ssf/2014/08/youngstown contractor sentence.html (accessed March 12, 2015).

Carter. KE: Hammack. RW: Hakala. IA. (2013). Hydraulic Fracturing and Organic Compounds - Uses, Disposal
   and Challenges. SPE Eastern Regional Meeting, Pittsburgh, Pennsylvania, USA.

Castle. SL: Thomas. BF: Reager. IT: Rodell. M: Swenson. SC: Famiglietti. IS. (2014). Groundwater depletion
   during drought threatens future water security of the Colorado River Basin. Geophys Res Lett 41: 5904-
   5911. http://dx.doi.org/10.1002/2014GL061055

CCST (California Council on Science and Technology). (2014). Advanced well stimulation technologies in
   California: An independent review of scientific and technical information. Sacramento, CA.
   http://ccst.us/publications/2014/2014wst.pdf

CCST (California Council on Science and Technology). (2015). An independent scientifc assessment of well
   stimulation in California, Volume 1: Well stimulation technologies and their past, present, and potential
   future use in California. Sacramento, CA. http://www.ccst.us/publications/2015/2015SB4-vl.pdf

Ceres (Coalition for Environmentally Responsible Economies). (2014). Hydraulic fracturing & water stress:
   water demand by the numbers. Boston, Massachusetts. https://www.ceres.org/issues/water/shale-
   energy/shale-and-water-maps/hydraulic-fracturing-water-stress-water-demand-by-the-numbers

Chapman. EC: Capo. RC: Stewart. BW: Kirby. CS: Hammack. RW: Schroeder. KT: Edenborn. HM. (2012).
   Geochemical and strontium isotope characterization of produced waters from Marcellus Shale natural gas
   extraction. Environ Sci Technol 46: 3545-3553.

Chaudhuri. S: Ale. S. (2013). Characterization of groundwater resources in the Trinity and Woodbine aquifers
   in Texas. Sci Total Environ 452: 333-348. http://dx.doi.Org/10.1016/j.scitotenv.2013.02.081

Chaudhuri. S: Ale. S. (2014a). Long term (1960-2010) trends in groundwater contamination and salinization
   in the Ogallala aquifer in Texas. J Hydrol 513: 376-390. http://dx.doi.Org/10.1016/j.jhydrol.2014.03.033

Chaudhuri. S: Ale. S. (2014b). Temporal evolution of depth-stratified groundwater salinity in municipal wells
   in the major aquifers in Texas, USA. Sci Total Environ 472: 370-380.
   http://dx.doi.0rg/10.1016/j.scitotenv.2013.10.120

ChemicalBook (ChemicalBook Inc.). (2010). Sorbitan trioleate. Available online at
   http://www.chemicalbook.com/chemicalproductproperty en cb4677178.htm (accessed April 6,2015).

Chester Engineers. (2012). The Pittsburgh water and sewer authority 40 year plan. (PWSA Project No. R-
   Dl.10015-11). Pittsburgh, PA: The Pittsburgh Water and Sewer Authority.
   http://apps.pittsburghpa.gov/pwsa/PWSA 40-vear Plan.pdf
               This document is a draft for review purposes only and does not constitute Agency policy.

June 2015                                        7                     DRAFT—DO NOT CITE OR QUOTE

-------
Hydraulic Fracturing Drinking Water Assessment                                                All References
Chiado. ED. (2014). The impact of shale gas/oil waste on MSW landfill composition and operations. In CL
   Meehan; JM VanBriesen; F Vahedifard; X Yu; C Quiroga (Eds.), Shale energy engineering 2014 technical
   challenges, environmental issues, and public policy (pp. 412-420). Reston, VA: American Society of Civil
   Engineers, http://dx.doi.org/10.1061/9780784413654.044

Ciezobka. I: Salehi. I. (2013). Controlled hydraulic fracturing of naturally fractured shales: A case study in the
   Marcellus Shale examining how to identify and exploit natural fractures. (SPE-164524-MS). Ciezobka, J;
   Salehi, I. http://dx.doi.org/10.2118/164524-MS

Cipolla. C: Weng. X: Mack. M: Ganguly. U: Gu. H: Kresse. 0: Cohen. C. (2011). Integrating microseismic mapping
   and complex fracture modeling to characterize hydraulic fracture complexity. Paper presented at SPE
   Hydraulic Fracturing Technology Conference, January 24-26,2011, The Woodlands, TX.

Clark. CE: Horner. RM: Harto. CB. (2013). Life Cycle Water Consumption for Shale Gas and Conventional
   Natural Gas. Environ Sci Technol 47:11829-11836. http://dx.doi.org/10.1021/es4013855

Clark. CE: Veil. IA. (2009). Produced water volumes and management practices in the United States (pp. 64).
   (ANL/EVS/R-09/1). Argonne, IL: Argonne National Laboratory.
   http://www.circleofblue.org/waternews/wp-
   content/uploads/2010/09/ANL EVS  R09  produced water volume report 2437.pdf

Cluff. M: Hartsock. A: Macrae. I: Carter. K: Mouser. PI. (2014). Temporal changes in microbial ecology and
   geochemistry in produced water from hydraulically fractured Marcellus Shale Gas Wells. Environ Sci
   Technol 48: 6508-6517. http://dx.doi.org/10.1021/es501173p

COGCC. Colorado Oil and Gas Conservation Commission Order No. 1V-276, (2004).
   https://cogcc.state.co.us/orders/orders/lv/276.html

COGCC (Colorado Oil and Gas Conservation Commission). (2014). Risk-based inspections: Strategies to
   address environmental risk associated with oil and gas operations. (OGCC2014PROJECT #7948). Denver,
   CO.

COGCC (Colorado Oil and Gas Conservation Commission). (2015). COGIS - all production reports to date.
   Denver, CO. Retrieved from http://cogcc.state.co.us/

Cohen, HA: Parratt, T: Andrews, CB. (2013). Comments on 'Potential contaminant pathways from
   hydraulically fractured shale to aquifers' [Comment]. Ground Water 51: 317-319; discussion 319-321.
   http://dx.doi.org/10.llll/gwat.12015

Colborn. T: Kwiatkowski. C: Schultz. K: Bachran. M. (2011). Natural gas operations from a public health
   perspective. Hum Ecol Risk Assess 17:1039-1056. http://dx.doi.org/10.1080/10807039.2011.605662

Coleman. N. (2011). Produced formation water sample results from shale plays. Presentation presented at
   Technical Workshops for the Hydraulic Fracturing Study: Chemical & Analytical Methods, February 2425,
   2011, Arlington, VA.

Colorado Division of Water Resources: Colorado Water Conservation Board: Colorado Oil and Gas
   Conservation Commission. (2014). Water sources and demand for the hydraulic fracturing of oil and gas
   wells in Colorado from 2010 through 2015 [Fact Sheet], http://cewc.colostate.edu/2012/02/water-
   sources-and-demand-for-the-hydraulic-fracturing-of-oil-and-gas-wells-in-colorado-from-2010-through-
   20157

Considine, T: Watson, R: Considine, N: and Martin, I. (2012). Environmental impacts during Marcellus shale
   gas drilling: Causes, impacts, and remedies. (Report 2012-1). Buffalo, NY: Shale Resources and Society
   Institute. http://cce.cornell.edu/EnergyClimateChange/NaturalGasDev/Documents/UBSRSI-
   Environmental%20Impact%20Report%202012.pdf

Cooley. H: Gleick. PH: Wolff. G. (2006). Desalination, with a grain of salt: A California perspective. Oakland, CA:
   Pacific Institute for Studies in Development, Environment, and Security, http://www.pacinst.org/wp-
   content/uploads/2013/02/desalination report3.pdf
               This document is a draft for review purposes only and does not constitute Agency policy.

June 2015                                        8                     DRAFT—DO NOT CITE OR QUOTE

-------
Hydraulic Fracturing Drinking Water Assessment                                                All References
Corseuil. HX: Kaipper. BI: Fernandes. M. (2004). Cosolvency effect in subsurface systems contaminated with
   petroleum hydrocarbons and ethanol. Water Res 38:1449-1456.
   http://dx.doi.0rg/10.1016/j.watres.2003.12.015

Corseuil. HX: Monier. AL: Fernandes. M: Schneider. MR: Nunes. CC: Do Rosario. M: Alvarez. PI. (2011). BTEX
   plume dynamics following an ethanol blend release: geochemical footprint and thermodynamic
   constraints on natural attenuation. Environ Sci Technol 45: 3422-3429.
   http://dx.doi.org/10.1021/esl04055q

Coulter. GR: Gross. BC: Benton. EG: Thomson. CL. (2006). Barnett shale hybrid fracs - One operator's design,
   application, and results. Richardson, TX: Society of Petroleum Engineers.
   http://dx.doi.org/10.2118/102063-MS

Council of Canadian Academies. (2014). Environmental impacts of shale gas extraction in Canada. Ottawa,
   Ontario.
   http://www.scienceadvice.ca/uploads/eng/assessments%20and%20publications%20and%20news%20
   releases/Shale%20gas/ShaleGas fullreportEN.pdf

Countess, S: Boardman, G: Hammack, R: Hakala, A: Sharma, S: Parks, I. (2014). Evaluating leachability of
   residual solids from hydraulic fracturing in the Marcellus shale. In Shale energy engineering 2014:
   Technical challenges, environmental issues, and public policy. Reston, VA: American Society of Civil
   Engineers, http://dx.doi.org/10.1061/9780784413654.012

Craig. MS: Wendte. SS: Buchwalter. IL. (2012). Barnett shale horizontal restimulations: A case study of 13
   wells. SPE Americas unconventional resources conference, June 5-7, 2012, Pittsburgh, PA.

Crescent (Crescent Consulting, LLC). (2011). East Mamm creek project drilling and cementing study.
   Oklahoma City, OK. http://cogcc.state.co.us/Library/PiceanceBasin/EastMammCreek/ReportFinal.pdf

Crook. R. (2008). Cementing: Cementing horizontal wells. Halliburton.

Curtice. Rl: Salas. WDI: Paterniti. ML. (2009). To gel or not to gel? In 2009 SPE annual technical conference &
   exhibition. Richardson, TX: Society of Petroleum Engineers. http://dx.doi.org/10.2118/124125-MS

Cusick. M. (2013). EPA fines western PA treatment plants for Marcellus wastewater violations. Available
   online at http://stateimpact.npr.org/pennsylvania/2013/05/24/epa-fines-western-pa-treatment-plants-
   for-marcellus-wastewater-violations/ (accessed March 6, 2015).

Dahi Taleghani, A: Ahmadi, M: Olson, IE. (2013). Secondary fractures and their potential impacts on hydraulic
   fractures efficiency. In AP In Bunger; J McLennan; R Jeffrey (Eds.), Effective and sustainable hydraulic
   fracturing. Croatia: InTech. http://dx.doi.org/10.5772/56360

Dahi Taleghani. A: Olson. IE. (2009). Numerical modeling of multi-stranded hydraulic fracture propagation:
   Accounting for the interaction between induced and natural fractures. In 2009 SPE Annual Technical
   Conference and Exhibition. Richardson, TX: Society of Petroleum Engineers.
   http://dx.doi.org/10.2118/124884-MS

Dahm. K: Chapman. M. (2014). Produced water treatment primer: Case studies of treatment applications.
   (S&T Research Project #1617). Denver CO: U.S. Department of the Interior.
   http://www.usbr.gov/research/projects/download  product.cfm?id=1214.

Dahm. KG: Guerra. KL: Xu. P: Drewes. IE. (2011). Composite geochemical database for coalbed methane
   produced water quality in the Rocky Mountain region. Environ Sci Technol 45: 7655-7663.
   http://dx.doi.org/10.1021/es201021n

Pake, LP. (1978). Fundamentals of reservoir engineering. Boston, MA: Elsevier.
   http://www.ing.unp.edu.ar/asignaturas/reservorios/Fundamentals%20of%20Reservoir%20Engineering
   %20%28LP.%20Dake%29.pdf
               This document is a draft for review purposes only and does not constitute Agency policy.

June 2015                                        9                     DRAFT—DO NOT CITE OR QUOTE

-------
Hydraulic Fracturing Drinking Water Assessment                                                All References
Daneshy. AA. (2009). Factors controlling the vertical growth of hydraulic fractures. (SPE-118789-MS).
   Daneshy, AA. http://dx.doi.org/10.2118/118789-MS

Darrah. TH: Vengosh. A: lackson. RB: Warner. NR: Poreda. Rl. (2014). Noble gases identify the mechanisms of
   fugitive gas contamination in drinking-water wells overlying the Marcellus and Barnett Shales. PNAS 111:
   14076-14081. http://dx.doi.org/10.1073/pnas.1322107111

Daulton. D: Post. M: McMahon. I: Kuc. B: Ake. C: Hughes. B: Hill. D. (2012). Global chemical evaluation process
   review to qualify regulatory  and environmental characteristics for oilfield chemical products. Paper
   presented at SPE Annual Technical Conference and Exhibition, October 8-10, 2012, San Antonio, TX.

Davies, Rl: Almond, S: Ward, RS: lackson, RB: Adams, C: Worrall, F: Herringshaw, LG: Gluyas, IG: Whitehead,
   MA. (2014). Oil and gas wells and their integrity: Implications for shale and unconventional resource
   exploitation. Marine and Petroleum Geology 56: 239-254.
   http://dx.doi.0rg/10.1016/j.marpetgeo.2014.03.001

Davies. Rl: Foulger. GR: Mathias. S: Moss. I: Hustoft. S: Newport. L. (2013). Reply: Davies etal. (2012),
   Hydraulic fractures: How far can they go? Marine and Petroleum Geology 43: 519-521.
   http://dx.doi.0rg/10.1016/j.marpetgeo.2013.02.001

Davies. Rl: Mathias. SA: Moss. I: Hustoft. S: Newport. L.  (2012). Hydraulic fractures: How far can they go?
   Marine and Petroleum Geology 37:1-6. http://dx.doi.Org/10.1016/j.marpetgeo.2012.04.001

Davis. IP: Struchtemeyer. CG: Elshahed. MS. (2012). Bacterial communities associated with production
   facilities of two newly drilled thermogenic natural gas wells in the Barnett Shale (Texas, USA). Microb Ecol
   64: 942-954. http://dx.doi.org/10.1007/s00248-012-0073-3

De Pater. Cl: Baisch. S. (2011). Geomechanical study of the Bowland shale seismicity: Synthesis report.
   Nottingham, England: British Geological Survey.
   https://www.bucknell.edu/script/environmentalcenter/marcellus/default.aspx?articleid=MF08SXMW82
   CV1MQAXK7ZINIPP

DeArmond. PD: DiGoregorio. AL. (2013a). Characterization of liquid chromatography-tandem mass
   spectrometry method for the determination of acrylamide in complex environmental samples. Anal
   Bioanal Chem 405: 4159-4166. http://dx.doi.org/10.1007/s00216-013-6822-4

DeArmond. PD: DiGoregorio. AL. (2013b).  Rapid liquid chromatography-tandem mass spectrometry-based
   method for the analysis of alcohol ethoxylates and alkylphenol ethoxylates in environmental samples. J
   Chromatogr A1305:154-163. http://dx.doi.Org/10.1016/j.chroma.2013.07.017

Dehghanpour. H: Lan. 0: Saeed. Y: Fei. H: Oi. Z. (2013). Spontaneous imbibition of brine and oil in gas shales:
   Effect of water adsorption and resulting microfractures. Energy Fuels 27: 3039-3049.
   http://dx.doi.org/10.1021/ef4002814

Dehghanpour. H: Zubair. HA: Chhabra. A: Ullah. A. (2012). Liquid intake of organic shales. Energy Fuels 26:
   5750-5758. http://dx.doi.org/10.1021/ef3009794

Demorest. PL: Wallace. ES. (1992). Radiochemical determination of norm in produced water utilizing wet
   chemistry separation followed by radiochemical analysis. In JP Ray; Engelhardt; Fr (Eds.), Produced
   water: Technological/environmental issues and solutions (pp. 257-266). New York, NY: Plenum Press.
   http://dx.doi.org/10.1007/978-l-4615-2902-6 21

DOE (U.S. Department of Energy). (2002). Handbook on best management practices and mitigation strategies
   for coalbed methane in the Montana portion of the Powder River Basin. Tulsa, OK: U.S. Department of
   Energy, National Energy Technology Laboratory. http://bogc.dnrc.mt.gov/PDF/BMPHandbookFinal.pdf

DOE (U.S. Department of Energy). (2003). Handbook on coalbed methane produced water: Management and
   beneficial use alternatives. Tulsa, OK: ALL Consulting, http://www.all-
   llc.com/publicdownloads/CBM BU  Screen.pdf
               This document is a draft for review purposes only and does not constitute Agency policy.

June 2015                                        10                     DRAFT—DO NOT CITE OR QUOTE

-------
Hydraulic Fracturing Drinking Water Assessment                                               All References
DOE (U.S. Department of Energy). (2004). A white paper describing produced water from production of crude
   oil, natural gas, and coal bed methane. Lemont, IL: Argonne National Laboratory.
   http://seca.doe.gov/technologies/oil-gas/publications/oil pubs/prodwaterpaper.pdf

DOE (U.S. Department of Energy). (2006). A guide to practical management of produced water from onshore
   oil and gas operations in the United States. Washington, DC: U.S. Department of Energy, National
   Petroleum Technology Office.
   http://fracfocus.0rg/sites/default/files/publications/a guide to practical management of produced wat
   er from onshore oil and gas operations in the united states.pdf

DOE (U.S. Department of Energy). (2011a). A comparative study of the Mississippian Barnett shale, Fort
   Worth basin, and Devonian Marcellus shale, Appalachian basin. (DOE/NETL-2011/1478).
   http://www.netl.doe.gov/technologies/oil-gas/publications/brochures/DOE-NETL-2011-
   1478%20Marcellus-Barnett.pdf

DOE (U.S. Department of Energy). (2011b). Shale gas Applying technology to solve americas energy
   challenges. Available online at http://groundwork.iogcc.org/topics-index/shale-gas/topic-
   resources/shale-gas-applying-technology-to-solve-americas-energy-challe

Domenico. PA: Schwartz. FW. (1997). Physical and chemical hydrology. In Physical and chemical
   hydrogeology (2nd ed.). Hoboken, NJ: Wiley.

Dresel. PE: Rose. AW. (2010). Chemistry and origin of oil and gas well brines in western Pennsylvania (pp.
   48). (Open-File Report OFOG 1001.0). Harrisburg, PA: Pennsylvania Geological Survey, 4th ser.
   http://www.marcellus.psu.edu/resources/PDFs/brines.pdf

Drewes. I: Cath. T: Debroux. I: Veil. I. (2009). An integrated framework for treatment and management of
   produced water - Technical assessment of produced water treatment technologies (1st ed.). (RPSEA
   Project 07122-12). Golden, CO: Colorado School of Mines.
   http://aqwatec.mines.edu/research/projects/Tech Assessment PW  Treatment Tech.pdf

Drillinglnfo. Inc.. (2012). DI Desktop August 2012 download [Database]. Austin, TX. Retrieved from
   http: //info, drillinginfo. com/

Drillinglnfo. Inc.. (2014a). DI Desktop June 2014 download [Database]. Austin, TX: Drillinglnfo. Retrieved
   from http://info.drillinginfo.com/

Drillinglnfo. Inc.. (2014b). Drillinglnfo Inc. DI Desktop raw data feed [Database].

Drohan. PI: Brittingham. M. (2012). Topographic and soil constraints to shale-gas development in the
   northcentral Appalachians. Soil Sci Soc Am J 76: 1696-1706. http://dx.doi.org/10.2136/sssaj2012.0087

Duraisamy. RT: Beni. AH: Henni. A. (2013). State of the art treatment of produced water. In W Elshorbagy; RK
   Chowdhury (Eds.), Water treatment (pp. 199-222). Rijeka, Croatia: InTech.
   http://dx.doi.org/10.5772/53478

Dusseault. MB: Gray. MN: Nawrocki. PA. (2000). Why oilwells leak: Cement behavior and long-term
   consequences. Paper presented at SPE International Oil and Gas Conference and Exhibition in China,
   November 7-10, 2000,  Beijing, China.

Dutta. R: Lee.  C. -H: Odumabo. S: Ye. P: Walker. SC: Karpyn. ZT: Ayala. LF. (2014). Experimental investigation
   of fracturing-fluid migration caused by spontaneous imbibition in fractured low-permeability sands. SPE
   Reserv Eval Engin 17: 74-81.

Easton. I. (2014). Optimizing fracking wastewater management. Pollution Engineering January 13.

Eberhard. M. (2011). Fracture design and stimulation - monitoring. Presentation presented at Technical
   Workshops for the Hydraulic Fracturing Study: Well Construction & Operations, March 10-11, 2011,
   Arlington,  VA.
               This document is a draft for review purposes only and does not constitute Agency policy.

June 2015                                        11                    DRAFT—DO NOT CITE OR QUOTE

-------
Hydraulic Fracturing Drinking Water Assessment                                                All References
Economides. Ml: Hill. A. d: Ehlig-Economides. C: Zhu. D. (2013). Petroleum production systems. In Petroleum
   production systems (2nd ed.). Englewood Cliffs, NJ: Prentice Hall.

Economides. Ml: Mikhailov. DN: Nikolaevskiy. VN. (2007). On the problem of fluid leakoff during hydraulic
   fracturing. Transport in Porous Media 67: 487-499. http://dx.doi.org/10.1007/sll242-006-9038-7

EERC (Energy and Environmental Research Center, University of North Dakota). (2010). Bakken water
   opportunities assessment phase 1. (2010-EERC-04-03). Grand Forks, ND: Energy and Environmental
   Research Center. http://www.undeerc.org/bakken/pdfs/FracWaterPhaseIreport.pdf

EERC (Energy and Environmental Research Center, University of North Dakota). (2011). Bakken water
   opportunities assessment phase 2: evaluation of brackish groundwater treatment for use in hydraulic
   fracturing of the Bakken Play, North Dakota. (2011-EERC-12-05). Grand Forks, ND: Energy and
   Environmental Research Center. http://www.undeerc.org/Water/pdf/BakkenWaterOppPhase2.pdf

EERC (Energy and Environmental Research Center, University of North Dakota). (2013). BakkenSmart: water
   [Fact Sheet]. Grand Forks, ND: Energy and Environmental Research Center.
   http://www.undeerc.org/bakken/pdfs/NDIC-NDPC-Water-Fact-Sheet.pdf

EIA (Energy Information Administration). (2011a). Review of emerging resources: U.S. shale gas and shale oil
   plays. United States Department of Energy, http://www.eia.gov/analysis/studies/usshalegas/

EIA (Energy Information Administration). (2011b). Shale gas and oil plays, lower 48 States [Map]. Available
   online at http://www.eia.gov/pub/oil gas/natural gas/analysis publications/maps/maps.htm

EIA (Energy Information Administration). (2012a). Formation crosswalk. Washington, DC: U.S. Energy
   Information Administration.

EIA (Energy Information Administration). (2012b). Today in energy: Geology and technology drive estimates
   of technically recoverable resources. Washington, DC: U.S. Energy Information Administration.
   http://www.eia.gov/todayinenergy/detail.cfm?id=7190

EIA (Energy Information Administration). (2012c). What is shale gas and why is it important? [December 5].
   Washington, DC: U.S. Energy Information Administration.
   http://www.eia.gov/energy in brief/article/about shale gas.cfm

EIA (Energy Information Administration). (2013a). Analysis & projections: AE02014 early release overview.
   Release date: December 16, 2013. Washington, DC: U.S. Energy Information Administration.
   http://www.eia.gov/forecasts/aeo/er/executive summary.cfm

EIA (Energy Information Administration). (2013b). Annual energy outlook 2013 with projections to 2040.
   (DOE/EIA-0383). Washington, DC: U.S. Energy Information Administration.
   http://www.eia.gov/forecasts/archive/aeol3/pdf/0383(2013).pdf

EIA (Energy Information Administration). (2013c). Technically recoverable shale oil and shale gas resources:
   an assessment of 137 shale formations in 41 countries outside the United States (pp. 730). Washington,
   D.C.: Energy Information Administration, U.S. Department of Energy.
   http://www.eia.gov/analysis/studies/worldshalegas/

EIA (Energy Information Administration). (2013d). U.S. field production of crude oil. Release date: September
   27, 2014. Washington, DC: U.S. Energy Information Administration.
   http://www.eia.gov/dnav/pet/hist/LeafHandler.ashx?n=pet&s=mcrfpusl&f=a

EIA (Energy Information Administration). (2014a). Annual energy outlook 2014 with projections to 2040.
   (DOE/EIA-0383(2014)). Washington, D.C.: U.S. Energy Information Administration.
   http://www.eia.gov/forecasts/aeo/rjdf/0383r20141.rjdf
               This document is a draft for review purposes only and does not constitute Agency policy.

June 2015                                        12                     DRAFT—DO NOT CITE OR QUOTE

-------
Hydraulic Fracturing Drinking Water Assessment                                                All References
El A (Energy Information Administration). (2014b). Natural gas. U.S. crude oil and natural gas proved
   reserves. With data for 2012. Table 2. Principal tight oil plays: oil production and proved reserves, 2011-
   12. Release date: April 10, 2014. Washington, DC: U.S. Energy Information Administration.
   http://www.eia.gov/naturalgas/crudeoilreserves/

EIA (Energy Information Administration). (2014c). Natural gas. U.S. Energy Information Administration:
   Independent statistics and analysis. Available online at http://www.eia.gov/naturalgas/

EIA (Energy Information Administration). (2014d). Natural gas. U.S. natural gas gross withdrawals.
   Washington, DC: U.S. Energy Information Administration.
   http://www.eia.gov/dnav/ng/ng prod sum a  EPGO FGW mmcf a.htm

EIA (Energy Information Administration). (2014e). Natural gas: Natural gas prices [Database]. Washington,
   DC: U.S. Energy Information Administration. Retrieved from
   http://www.eia.gov/dnav/ng/ng pri sum dcu nus a.htm

EIA (Energy Information Administration). (2014f). October 2014 month energy review. (DOE/EIA-
   0035(2014/10)). Washington, D.C.: U.S. Energy Information Administration.
   http://www.eia.gov/totalenergy/data/monthly/archive/00351410.pdf

EIA (Energy Information Administration). (2014g). Petroleum & other liquids. Crude oil and natural gas
   drilling activity. Washington, DC: U.S. Energy Information Administration.
   http://www.eia.gov/dnav/pet/pet crd drill si a.htm

EIA (Energy Information Administration). (2014h). Today in energy: Tight oil production pushes U.S. crude
   supply to over 10% of world total. Washington, DC: U.S.  Energy Information Administration.
   http://www.eia.gov/todayinenergy/detail.cfm?id=15571.

EIA (Energy Information Administration). (2015a). Glossary. Available online at
   http://www.eia.gov/tools/glossary/

EIA (Energy Information Administration). (2015b). Lower 48 states shale plays. Available online at
   http://www.eia.gov/oil gas/rpd/shale gas.pdf

Eiceman. GA. (1986). Hazardous organic wastes from natural gas production, processing and distribution:
   Environmental fates. (WRRI report, no. 227). New Mexico: Water Resources Research Institute.
   http://wrri.nmsu.edu/publish/techrpt/abstracts/abs227.html

Eisner, L: Fischer, T: Le Calvez, IH. (2006). Detection of repeated hydraulic fracturing (out-of-zone growth) by
   microseismic monitoring. The Leading Edge (Tulsa) 25:  548-554. http://dx.doi.Org/10.1190/l.2202655

Elbel. I: Britt. L. (2000). Fracture treatment design. In MJ Economides; KG Nolle (Eds.), Reservoir stimulation
   (3rd ed.). New York, NY: John Wiley & Sons.

Ely. IW. (1989). Chapter 7: Fracturing fluids and additives. In JL Gidley; SA Holditch; DE Nierode; RW Veatch
   Jr (Eds.), Recent advances in hydraulic fracturing (pp. 131-146). Richardson, TX: Society of Petroleum
   Engineers.

Enform. (2013). Interim industry recommended practice 24: fracture stimulation: Interwellbore
   communication 3/27/2013 (1.0 ed.). (IRP 24). Calgary, Alberta:  Enform Canada.
   http://www.enform.ca/safety resources/publications/PublicationDetails.aspx?a=29&type=irp

Engelder, T. (2012). Capillary tension and imbibition sequester frack fluid in Marcellus gas  shale [Letter].
   PNAS  109: E3625; author reply E3626. http://dx.doi.org/10.1073/pnas.1216133110

Engle, MA: Bern, CR: Healy, RW: Sams, II: Zupancic, IW: Schroeder, KT. (2011). Tracking solutes and water
   from subsurface drip irrigation application of coalbed methaneproduced waters, Powder River Basin,
   Wyoming. Environmental Geosciences 18:169-187.
               This document is a draft for review purposes only and does not constitute Agency policy.

June 2015                                        13                     DRAFT—DO NOT CITE OR QUOTE

-------
Hydraulic Fracturing Drinking Water Assessment                                                All References
Engle. MA: Rowan. EL. (2014). Geochemical evolution of produced waters from hydraulic fracturing of the
   Marcellus Shale, northern Appalachian Basin: A multivariate compositional data analysis approach. Int J
   Coal Geol 126: 45-56. http://dx.doi.Org/10.1016/i.coal.2013.ll.010

Entrekin. S: Evans-White. M: lohnson. B: Hagenbuch. E. (2011). Rapid expansion of natural gas development
   poses a threat to surface waters. Front Ecol Environ 9: 503-511. http://dx.doi.org/10.1890/110053

Environment Canada. (2004). Threats to Water Availability in Canada, http://www.ec.gc.ca/inre-
   nwri/default.asp?lang=En&n=OCD66675-l

EPA (Environmental Protection Agency). (2000). Development document for effluent limitations guidelines
   and standards for the centralized waste treatment industry. (821R00020). Washington, DC: U.S.
   Environmental Protection Agency.

ERCB (Energy Resource Conservation Board). (2012). Midway Energy Ltd. hydraulic Fracturing incident:
   Interwellbore communication January 13,2012. (ERCB Investigation Report, Red Deer Field Centre).
   Calgary, Alberta: Energy Resources Conservation Board.

Ertel, D: McManus, K: Bogdan, I. (2013). Marcellus wastewater treatment: Case study. In Summary of the
   technical workshop on wastewater treatment and related modeling (pp. A56-A66). Williamsport, PA:
   Eureka Resources, LLC. http://www2.epa.gov/hfstudy/summary-technical-workshop-wastewater-
   treatment-and-related-modeling

ESN Rocky Mountain (Environmental Services Network Rocky Mountain). (2003). Produced gas and water
   testing of CBM gas wells in the Raton Basin. Golden, CO.
   https://cogcc.state.co.us/Library/RatonBasin/ESN%20Final Reportpdf

ESRI (Environmental Systems Research Institute Inc.). (2010). US states shapefile. Redlands, California.
   Retrieved from http://www.arcgis.com/home/item.html?id=la6cae723afl4f9cae228bl33aebc620

Fakhru'1-Razi. A: Pendashteh. A: Abdullah. LC: Biak. PR: Madaeni. SS: Abidin. ZZ. (2009). Review of
   technologies for oil and gas produced water treatment [Review]. J Hazard Mater 170: 530-551.

Famiglietti. IS: Lo. M: Ho. SL: Bethune. I: Anderson. Kl: Syed. TH: Swenson. SC: de Linage. CR: Rodell. M.
   (2011). Satellites measure recent rates of groundwater depletion in California's Central Valley. Geophys
   Res Lett 38: L03403. http://dx.doi.org/10.1029/2010GL046442

Farag. AM: Harper. DP. (2014). A review of environmental impacts of salts from produced waters on aquatic
   resources. Int J Coal Geol 126:157-161. http://dx.doi.Org/10.1016/j.coal.2013.12.006

Ferrar. Kl: Michanowicz. PR: Christen. CL: Mulcahy. N: Malone. SL: Sharma. RK. (2013). Assessment of effluent
   contaminants from three facilities discharging  Marcellus Shale wastewater to surface waters in
   Pennsylvania. Environ  Sci Technol 47: 3472-3481.

Fertl. WH: Chilingar. GV. (1988). Total organic carbon content determined from well logs. SPE Formation
   Evaluation 3: 407-419. http://dx.doi.org/10.2118/15612-PA

Fink. IK. (2003). Oil field chemicals. In Oil field chemicals. Boston, MA: Gulf Professional Publishing.

Finkel. M: Hays. I: Law. A. (2013). The shale gas boom and the need for rational policy. Am J Public Health
   103: 1161-1163. http://dx.doi.org/10.2105/AIPH.2013.301285

Fisher. IB: Sublette. KL. (2005). Environmental releases from exploration and production operations in
   Oklahoma: Type, volume, causes, and prevention. Environmental Geosciences 12: 89-99.
   http://dx.doi.org/10.1306/eg.11160404039

Fisher. K. (2012). Green frac fluid chemistry optimizes well productivity, environmental performance
   [Magazine]. The American Oil and Gas Reporter, March 2012,4.

Fisher. M: Warpinski. N. (2012). Hydraulic fracture height growth: Real data. SPE Prod Oper 27: 8-19.
   http://dx.doi.org/10.2118/145949-PA
               This document is a draft for review purposes only and does not constitute Agency policy.

June 2015                                        14                     DRAFT—DO NOT CITE OR QUOTE

-------
Hydraulic Fracturing Drinking Water Assessment                                               All References
Fisher. RS. (1998). Geologic and geochemical controls on naturally occurring radioactive materials (NORM) in
   produced water from oil, gas, and geothermal operations. Environmental Geosciences 5:139-150.
   http://dx.doi.0rg/10.1046/i.1526-0984.1998.08018.x

Fitzgerald. DP: McGhee. BF: McGuire. IA. (1985). Guidelines for 90 % accuracy in zone-isolation decisions. J
   Pet Tech 37: 2013-2022. http://dx.doi.org/10.2118/12141-PA

Fiaer. E: Holt. RM: Horsrud. P: Raaen. AM: Risnes. R. (2008). Petroleum related rock mechanics (2nd edition
   ed.). Amsterdam, The Netherlands:  Elsevier.

Flewelling. SA: Sharma. M. (2014). Constraints on upward migration of hydraulic fracturing fluid and brine.
   Ground Water 52: 9-19. http://dx.doi.org/10.llll/gwat.12095

Flewelling. SA: Tymchak. MP: Warpinski. N. (2013). Hydraulic fracture height limits and fault interactions in
   tight oil and gas formations. Geophys Res Lett 40: 3602-3606. http://dx.doi.org/10.1002/grl.50707

Flournoy. RM: Feaster. IH. (1963). Field observations on the use of the cement bond log and its application to
   the evaluation of cementing problems. Richardson, TX: Society of Petroleum Engineers.
   http://dx.doi.org/10.2118/632-MS

Fountain. 1C: lacobi. RD. (2000). Detection of buried faults and fractures using soil gas analysis.
   Environmental and Engineering Geoscience 6: 201-208. http://dx.doi.Org/10.2113/gseegeosci.6.3.201

Fredd. CN: Olsen. TN: Brenize. G: Quintero. BW: Bui. T: Glenn. S: Boney. CL. (2004). Polymer-free fracturing
   fluid exhibits improved cleanup for unconventional natural gas well applications.  Richardson, TX: Society
   of Petroleum Engineers. http://dx.doi.org/10.2118/91433-MS

Freeze. RA: Cherry. IA. (1979). Groundwater. In Groundwater. Upper Saddle River, NJ: Prentice Hall.

Gassiat. C: Gleeson. T: Lefebvre. R: Mckenzie. I. (2013). Numerical simulation of potential contamination of
   shallow aquifers over long time scales. Water Resour Res 49: 8310-8327.
   http://dx.doi.org/10.1002/2013WR014287

Geiver. L. (2013). Frac water treatment yields positive results for Houston Co. Retrieved from
   http://www.thebakken.com/articles/20/frac-water-treatment-yields-positive-results-for-houston-co

Georgakakos. A: Fleming. P: Dettinger.  M: Peters-Lidard. C: Richmond. TC: Reckhow. K: White. K: Yates. D.
   (2014). Water resources. In JM Melillo; TC Richmond; GW Yohe (Eds.), Climate change impacts in the
   United States (pp. 69-112). Washington, D.C.: U.S. Global Change  Research Program.
   http://www.globalchange.gov/ncadac

George. PG: Mace. RE: Petrossian. R. (2011). Aquifers of Texas. (Report 380). Austin, TX: Texas Water
   Development Board.
   http://www.twdb.state.tx.us/publications/reports/numbered reports/doc/R380 AquifersofTexas.pdf

Gidley, IL: Holditch, SA: Nierode, DE: Veatch  Ir., RW. (1989). Recent advances in hydraulic fracturing.
   Richardson, TX: Society of Petroleum Engineers.

Gilmore, K: Hupp, R: Glathar, I. (2013). Transport of Hydraulic Fracturing Water and Wastes in the
   Susquehanna River Basin, Pennsylvania. J Environ Eng 140: B4013002.
   http://dx.doi.org/10.1061/(ASCE)EE.1943-7870.0000810

GNB (Government of New Brunswick). (2013). Responsible environmental management of oil and natural gas
   activities in New Brunswick - rules  for industry. New Brunswick, Canada.
   http://www2.gnb.ca/content/dam/gnb/Corporate/pdf/ShaleGas/en/RulesforIndustry.pdf

GNB (Government of New Brunswick). (2015). FAQs hydraulic fracturing (fraccing). New Brunswick, Canada.
   http://www2.gnb.ca/content/dam/gnb/Corporate/pdf/ShaleGas/en/FAO HvdraulicFracturing.pdf
               This document is a draft for review purposes only and does not constitute Agency policy.

June 2015                                        15                    DRAFT—DO NOT CITE OR QUOTE

-------
Hydraulic Fracturing Drinking Water Assessment                                               All References
Goldstein. BD: Brooks. BW: Cohen. SD: Gates. AE: Honeycutt. ME: Morris. IB: Orme-Zavaleta. I: Penning. TM:
   Snawder, I. (2014). The role of lexicological science in meeting the challenges and opportunities of
   hydraulic fracturing. Toxicol Sci 139: 271-283. http://dx.doi.org/10.1093/toxsci/kfu061

Gomes. I: Cocke. D: Das. K: Guttula. M: Tran. D: Beckman: I. (2009). Treatment of produced water by
   electrocoagulation. Shiner, TX: KASELCO, LLC. http://www.kaselco.com/index.php/library/industry-
   white-papers

Gonneea, ME: Morris, PI: Dulaiova, H: Charette, MA. (2008). New perspectives on radium behavior within a
   subterranean estuary. Mar Chem 109: 250-267. http://dx.doi.Org/10.1016/j.marchem.2007.12.002

Goodwin, Kl: Crook, Rl. (1992). Cement sheath stress failure. S P E Drilling & Completion 7: 291-296.
   http://dx.doi.org/10.2118/20453-PA

Goodwin. S: Carlson. K: Knox. K: Douglas. C: Rein. L. (2014). Water intensity assessment of shale gas resources
   in the Wallenberg field in northeastern Colorado. Environ Sci Technol 48: 5991-5995.
   hllp://dx.doi.org/10.1021/es404675h

Gorody, AW. (2012). Factors affecting Ihe variability of slray gas concenlralion and composition in
   groundwaler. Environmental Geosciences 19:17-31. hllp://dx.doi.org/10.1306/eg.l2081111013

Gradient (2013). National human heallh risk evaluation for hydraulic fracluring fluid additives. Gradient
   hllp://www.energy.senale.gov/public/index.cfm/files/serve?File Jd=53a41a78-cQ6c-4695-a7be-
   84225aa7230f

Greenhunler (Greenhunler Resources). (2014). Oillfield water managemenl solutions. Available online al
   hllp://www.greenhunlerenergy.com/operalions/owms.him

Gregory, KB: Vidic, RD: Dzombak, DA. (2011). Water managemenl challenges associated wilh Ihe production
   of shale gas by hydraulic fracluring. Elemenls 7:181-186.

Gross. SA: Avens. HI: Banducci. AM: Sahmel. I: Panko. IM: Tvermoes. BE. (2013). Analysis of BTEX
   groundwaler concenlralions from surface spills associated wilh hydraulic fracluring operations. J Air
   Waste Manag Assoc 63: 424-432. hllp://dx.doi.org/10.1080/10962247.2012.759166

GTI (Gas Technology Inslilule). (2012). Barnell and Appalachian shale water managemenl and resuse
   technologies. (Reporlno. 08122-05.FINAL.l). Sugar Land, TX: Research Parlnership to Secure Energy for
   America, RPSEA. hllps://www.nell.doe.gov/file%201ibrary/research/oil-
   gas/Nalural%20Gas/shale%20gas/08122-05-final-reportpdf

Gu. H: Siebrils. E. (2008).  Effecl of formation modulus conlrasl on hydraulic fraclure heighl containment S P
   E Prod Oper 23: 170-176. hllp://dx.doi.org/10.2118/103822-PA

Gu. M: Mohanly. KK. (2014). Effecl of foam quality on effectiveness of hydraulic fracluring in shales.
   International Journal of Rock Mechanics and Mining Sciences 70: 273-285.
   hllp://dx.doi.org/10.1016/j.ijrmms.2014.05.013

Guerra. K: Dahm. K: Dundorf. S. (2011). Oil and gas produced water managemenl and beneficial use in Ihe
   western United Stales. (Science and Technology Program Reporl No. 157). Denver, CO: U.S. Deparlmenl of
   Ihe Interior Bureau of Reclamation.

Gupta. DVS: Hlidek. BT. (2009). Frac fluid recycling and water conservation: A case history. In 2009 Hydraulic
   fracluring technology  conference. Woodlands, Texas:  Society of Pelroleum Engineers.
   hllp://dx.doi.org/10.2118/119478-MS

Gupta, DVS: Valko, P. (2007). Fracluring fluids and formation damage. In M Economides; T Martin (Eds.),
   Modern fracluring: enhancing nalural gas production (pp. 227-279). Houston, TX: Energy Tribune
   Publishing Inc.
               This document is a draft for review purposes only and does not constitute Agency policy.

June 2015                                       16                    DRAFT—DO NOT CITE OR QUOTE

-------
Hydraulic Fracturing Drinking Water Assessment                                                All References
GWPC (Groundwater Protection Council). (2009). State oil and natural gas regulations designed to protect
   water resources. Morgantown, WV: U.S. Department of Energy, National Energy Technology Laboratory.
   http://www.gwpc.org/sites/default/files/state oil and gas regulations designed to protect water resou
   rces  O.pdf

GWPC (Groundwater Protection Council). (2014). State oil and natural gas regulations designed to protect
   water resources. Morgantown, WV: U.S. Department of Energy, National Energy Technology Laboratory.
   http://www.gwpc.org/sites/default/files/files/Oil%20and%20Gas%20Regulation%20Report%20Hyperl
   inked%20Version%2 OFinal-rfs.pdf

GWPC (Groundwater Protection Council). (2015). FracFocus - chemical disclosure registry. Available online at
   http://fracfocus.org/

GWPC and ALL Consulting (Ground Water Protection Council (GWPC) and ALL Consulting). (2009). Modern
   shale gas development in the United States: A primer. (DE-FG26-04NT15455). Washington, DC: U.S.
   Department of Energy, Office of Fossil Energy and National Energy Technology Laboratory.
   http://www.gwpc.org/sites/default/files/Shale%20Gas%20Primer%202009.pdf

GWPC and IOGCC (Ground Water Protection Council and Interstate Oil and Gas Compact Commission).
   (2014). Fracturing fluid management. Available online at http://fracfocus.org/hydraulic-fracturing-how-
   it-works/drilling-risks-safeguards (accessed December 12,2014).

H.R. Rep. 111-316. Department of the Interior. Environment, and Related Agencies Appropriation Act. 2010:
   Conference report (to accompany H.R. 2996), (2009). http://www.gpo.gov/fdsys/pkg/CRPT-
   Hlhrpt316/pdf/CRPT-lllhrpt316.pdf

Halliburton. (1988). Primer on Hydraulic Fracturing. Provided to EPA on March 2, 2011. Available at Docket
   ID: EPA-HQ-ORD-2010-0674-1634. (HESI-3031). Halliburton.
   http://www.regulations.gov/#!documentDetail:D=EPA-HQ-ORD-2010-0674-1634

Halliburton. (2013). Hydraulic fracturing 101. Available online at
   http://www.halliburton.com/public/projects/pubsdata/hydraulic fracturing/index.html

Halliburton. (2014). Hydraulic fracturing 101. Available online at
   http://www.halliburton.com/public/proiects/pubsdata/hydraulic fracturing/fracturing 101.html

Haluszczak. LO: Rose. AW: Kump. LR. (2013). Geochemical evaluation of flowbackbrine from Marcellus gas
   wells in Pennsylvania, USA. Appl Geochem 28: 55-61.
   http://dx.doi.0rg/10.1016/j.apgeochem.2012.10.002

Hamieh. BM: Beckman. IR. (2006). Seawater desalination using Dewvaporation technique: theoretical
   development and design evolution. Desalination 195: 1-13.
   http://dx.doi.0rg/10.1016/j.desal.2005.09.034

Hammack. R: Harbert. W: Sharma. S: Stewart. B: Capo. R: Wall. A: Wells. A: Diehl. R: Blaushild. D: Sams. I:
   Veloski, G. (2014). An evaluation of fracture growth and gas/fluid migration as horizontal Marcellus Shale
   gas wells are hydraulically fractured in Greene County, Pennsylvania. (NETL-TRS-3-2014). Pittsburgh, PA:
   U.S. Department of Energy, National Energy Technology Laboratory.
   http://www.netl.doe.gov/File%20Library/Research/onsite%20research/publications/NETL-TRS-3-
   2014 Greene-County-Site 20140915 1 l.pdf

Hammer. R: VanBriesen. I. (2 012). In frackings wake: New rules are needed to protect our health and
   environment from contaminated wastewater. New York, NY: Natural Resources Defense Council.
   http://www.nrdc.org/energy/files/fracking-wastewater-fullreport.pdf

Hansen. E: Mulvaney. D: Betcher. M. (2013). Water resource reporting and water footprint from Marcellus
   Shale development in West Virginia and Pennsylvania. Durango, CO: Earthworks Oil & Gas Accountability
   Project, http://www.downstreamstrategies.com/documents/reports publication/marcellus wv pa.pdf
               This document is a draft for review purposes only and does not constitute Agency policy.

June 2015                                        17                     DRAFT—DO NOT CITE OR QUOTE

-------
Hydraulic Fracturing Drinking Water Assessment                                                All References
Harms. WM: Yeager. R. (1987). Diesel-based gel concentrate reduces stimulation costs. Oil and Gas Journal 85:
   37-39.

Harrison. SS. (1983). Evaluating system for ground-water contamination hazards due to gas-well drilling on
   the Glaciated Appalachian Plateau. Ground Water 21: 689-700. http://dx.doi.Org/10.llll/j.1745-
   6584.1983.tb01940.x

Harrison. SS. (1985). Contamination of aquifers by overpressurizing the annulus of oil and gas wells. Ground
   Water 23: 317-324.

Hayes. T. (2009). Sampling and analysis of water streams associated with the development of Marcellus shale
   gas. Des Plaines, IL: Marcellus Shale Coalition, http://eidmarcellus.org/wp-
   content/ uploads/2 012/11/MSCommission-Report.pdf

Hayes. T. (2011). Characterization of Marcellus and Barnett shale flowback waters and technology
   development for water reuse. Paper presented at Hydraulic Fracturing Study: March 2011 Technical
   Workshop on Water Resource Management, March 29-30,2011, Arlington, VA.

Hayes, T: Severin, B. (2012a). Characterization of flowback water from the the Marcellus and the Barnett
   shale regions. Barnett and Appalachian shale water management and reuse technologies. (08122-05.09;
   Contract 08122-05). Hayes, T; Severin, B. http://www.rpsea.org/media/files/project/2146b3aO/08122-
   05-RT-Characterization Flowback Waters Marcellus Barnett Shale  Regions-03-20-12.pdf

Hayes. T: Severin. BF. (2012b). Evaluation of the aqua-pure mechanical vapor recompression system in the
   treatment of shale gas flowback water - Barnett and Appalachian shale water management and reuse
   technologies. (08122-05.11). Hayes, T; Severin, BF. http://barnettshalewater.org/documents/08122-
   05.11-EvaluationofMVR-3-12-2012.pdf

Hayes, TD: Halldorson, B: Horner, P: Ewing, I: Werline, IR: Severin, BF. (2014). Mechanical vapor
   recompression for the treatment of shale-gas flowback water. Oil and Gas Facilities  3: 54-62.

Haymond. D. (1991). The Austin Chalk - An overview. HGS Bulletin 33: 27-30, 32, 34.

He. Z. (2011). Flow of gas and water in hydraulically fractured shale gas reservoirs. EPA HF Workshop, March
   28-29,2011, Arlington, VA.

Healy. RW: Bartos. TT: Rice. CA: Mckinley. MP: Smith. BD. (2011). Groundwater chemistry near an
   impoundment for produced water, Powder River Basin, Wyoming, USA. J Hydrol 403: 37-48.
   http://dx.doi.0rg/10.1016/j.jhydrol.2011.03.042

Healy. RW: Rice. CA: Bartos. TT: Mckinley. MP. (2008). Infiltration from an impoundment for coal-bed natural
   gas, Powder River Basin, Wyoming: Evolution of water and sediment chemistry. Water Resour Res 44:
   n/a-n/a. http://dx.doi.org/10.1029/2007wr006396

Heermann. SE: Powers. SE. (1998). Modeling the partitioning of BTEX in water-reformulated gasoline systems
   containing ethanol. J Contam Hydrol 34: 315-341. http://dx.doi.org/10.1016/S0169-7722(98)00099-0

Heilweil. VM: Stolp. Bl: Kimball. BA: Susong. DP: Marston. TM: Gardner. PM. (2013). A stream-based methane
   monitoring approach for evaluating groundwater impacts associated with unconventional gas
   development. Ground Water 51: 511-524. http://dx.doi.org/10.llll/gwat.12079

Hladik. ML: Focazio. Ml: Engle.  M. (2014). Discharges of produced waters from oil and gas extraction via
   wastewater treatment plants are sources of disinfection by-products to receiving streams. Sci Total
   Environ 466. http://dx.doi.Org/10.1016/j.scitotenv.2013.08.008

Hoi ditch, SA. (2007). Chapter 8: Hydraulic fracturing. In JD Clegg (Ed.), Petroleum engineering handbook (pp.
   IV-323 - IV-366). Richardson, TX: Society of Petroleum Engineers. http://store.spe.org/Petroleum-
   Engineering-Handbook-Volume-IV-Production-Operations-Engineering-P61.aspx
               This document is a draft for review purposes only and does not constitute Agency policy.

June 2015                                        18                     DRAFT—DO NOT CITE OR QUOTE

-------
Hydraulic Fracturing Drinking Water Assessment                                                All References
Holtsclaw. I: Loveless. D: Saini. R: Fleming. I. (20111. SPE 146832: Environmentally-focused crosslinked gel
   system results in high retained proppant-pack conductivity. Presentation presented at Society of
   Petroleum Engineers Annual Conference, November 2,2011, Denver, CO.

Horn. A: Hu. I: Patton. M. (2013). QA/QC of water blending enhances crosslinked gel completions. Available
   online at http://content.stockpr.com/hydrozonix/files/downloads/1013HEP-hydrozonix.pdf
Horsey. CA. (1981). Depositional environments of the Pennsylvanian Pottsville Formation in the Black
   Warrior Basin of Alabama. Journal of Sedimentary Research 51: 799-806.
   http://dx.doi.org/10.1306/212F7DB5-2B24-llD7-8648000102C1865D

House of Representatives (U.S. House of Representatives). (2011). Chemicals used in hydraulic fracturing.
   Washington, D.C.: U.S. House of Representatives, Committee on Energy and Commerce, Minority Staff.
   http://democrats.energycommerce.house.gov/sites/default/files/documents/Hydraulic-Fracturing-
   Chemicals-2011-4-18.pdf
Houston. N: Blauch. M: Weaver. D: Miller. PS: O'Hara. D. (2009). Fracture-stimulation in the Marcellus shale:
   Lessons learned in fluid selection and execution. In 2009 SPE eastern regional meeting: limitless
   potential/formidable challenges. Richardson, TX: Society of Petroleum Engineers.
   http://dx.doi.org/10.2118/125987-MS

Hristozov. PR: Zabeo. A: Foran. C: Isigonis. P: Critto. A: Marcomini. A: Linkov. I. (2014). A weight of evidence
   approach for hazard screening of engineered nanomaterials. Nanotoxicology 8: 72-87.
   http://dx.doi.org/10.3109/17435390.2012.750695
Hulme, D. (2005). CBM co-produced water management, disposal, treatment and use. Hulme, D.
   http://www.uwyo.edu/haub/ruckelshaus-institute/ files/docs/publications/2005-cbm-water-
   management-hulme.pdf

Hunter. IA: Moser. PH. (1990). Ground water availability in Jefferson County, Alabama: geological survey
   special map 224. Tuscaloosa, AL: Geological Survey of Alabama.
   http://www.ogb.state.al.us/documents/pubs/onlinepubs/Special Maps/SM224 .PDF
Hyne. Ml. (2012). Nontechnical guide to petroleum geology, exploration, drilling and production. In
   Nontechnical guide to petroleum geology, exploration,  drilling and production (3 ed.). Tulsa, OK: PennWell
   Corporation.
IAEA (International Atomic Energy Agency). (2014). The environmental behaviour of radium: revised edition.
   Vienna, Austria. http://www-pub.iaea.org/MTCD/Publications/PDF/trs476 web.pdf

Igunnu. ET: Chen. GZ. (2014). Produced water treatment technologies. International Journal of Low-Carbon
   Technologies 9:157-177. http://dx.doi.org/10.1093/ijlct/cts049
IHS (Global Insight). (2009).  Measuring the economic and  energy impacts of proposals to regulate hydraulic
   fracturing. Prepared for American Petroleum Institute. Lexington, MA:  IHS Global Insight.
   http://s3.amazonaws.com/propublica/assets/natural  gas/ihs gi hydraulic fracturing taskl.pdf

IHS (Global Insight). (2013).  Americas new energy future:  The unconventional oil and gas revolution and the
   US economy. Douglas County, Colorado.
   http://www.energyxxi.org/sites/default/files/pdf/Americas  New Energy Future Phase3.pdf
Ingraffea. AR: Wells. MT: Santoro. RL: Shonkoff. SB. (2014). Assessment and risk analysis of casing and
   cement impairment in oil and gas wells in Pennsylvania, 2000-2012. PNAS 111: 1095510960.
   http://dx.doi.org/10.1073/pnas.1323422111

IOGCC (Interstate Oil and Gas Compact Commission). (2002). States experience with hydraulic fracturing. A
   survey of the interstate oil and gas compact commission. IOGCC (Interstate Oil and Gas Compact
   Commission). http://energyindepth.org/docs/pdf/IOGCC%20Hydraulic%20Fracturing%20Study%2007-
   2002.pdf
               This document is a draft for review purposes only and does not constitute Agency policy.
June 2015                                        19                     DRAFT—DO NOT CITE OR QUOTE

-------
Hydraulic Fracturing Drinking Water Assessment                                                All References
IOGCC (Interstate Oil and Gas Compact Commission). (2008). Protecting our country's resources: The states'
   case, orphaned well plugging initiative. Oklahoma City, OK: Interstate Oil and Gas Compact Commission
   (IOGCC). http://iogcc.myshopify.com/products/protecting-our-countrys-resources-the-states-case-
   orphaned-well-plugging-initiative-2008

IOGCC (Interstate Oil and Gas Compact Commission). (2015). Groundwork: hydraulic fracturing - state
   progress. Available online at http://groundwork.iogcc.ok.gov/topics-index/hydraulic-fracturing/state-
   progress

IPCC (Intergovernmental Panel on Climate Change). (2007). Climate change 2007: Impacts, adaptation and
   vulnerability. Cambridge, UK: Cambridge University Press, http://www.ipcc.ch/ipccreports/ar4-wg2.htm

Irwin. C. (2013). Hydraulic fracturing: A way to go greener? Available online at
   http://breakingenergy.com/2013/04/23/hydraulic-fracturing-a-way-to-go-greener/

IUPAC (International Union of Pure and Applied Chemistry). (2014). Gold Book: matrix effect. Available
   online at http://goldbook.iupac.org/M03759.html (accessed April 8, 2015).

lackson, G: Flores, C: Abolo, N: Lawal, H. (2013a). A novel approach to modeling and forecasting frac hits in
   shale gas wells. Paper presented at EAGE Annual Conference & Exhibition incorporating SPE Europec,
   June 10-13, 2013, London, UK.

lackson. RB: Carpenter. SR: Dahm. CN: Mcknight. DM: Naiman. Rl: Postel. SL: Running. SW. (2001). Water in a
   changing world. Ecol Appl 11:1027-1045. http://dx.doi.org/10.1890/1051-
   076ir2001)011[1027:WIACW]2.0.CO:2

lackson. RB: Vengosh. A: Darrah. TH: Warner. NR: Down. A: Poreda. Rl: Osborn. SG: Zhao. K: Karr. ID. (2013b).
   Increased stray gas abundance in a subset of drinking water wells near Marcellus  shale gas extraction.
   PNAS 110:11250-11255. http://dx.doi.org/10.1073/pnas.1221635110

lackson. RE: Dussealt. MB. (2014). Gas release mechanisms from energy wellbores. Presentation presented at
   48th US Rock Mechanics/Geomechanics Symposium, June 1-4, 2014, Minneapolis, Minnesota.

lackson. RE: Gorody. AW: Mayer. B: Roy. IW: Ryan. MC: Van Stempvoort. DR. (2013c). Groundwater protection
   and unconventional gas extraction: the critical need for field-based hydrogeological research. Ground
   Water 51: 488-510. http://dx.doi.org/10.llll/gwat.12074

lacob. R. (2011). Incident action plan, Franchuk 44-20 SWH incident. Piano, Texas: Denbury Onshore, LLC.

liang. M: Hendrickson. CT: Vanbriesen. IM. (2014). Life Cycle Water Consumption and Wastewater Generation
   Impacts of a Marcellus Shale Gas Well. Environ Sci Technol 48:1911-1920.
   http://dx.doi.org/10.1021/es4047654

Jones. IR: Britt. LK. (2009). Design and appraisal of hydraulic fractures. In Design and appraisal of hydraulic
   fractures. Richardson, TX: Society of Petroleum Engineers.

ludson. R: Richard. A:  Dix. PI: Houck. K: Martin. M: Kavlock. R: Dellarco.  V: Henry. T: Holderman. T: Sayre. P:
   Tan. S: Carpenter. T: Smith.  E. (2009). The toxicity data landscape for environmental chemicals [Review].
   Environ Health Perspect 117: 685-695. http://dx.doi.org/10.1289/ehp.0800168

Kahrilas. GA: Blotevogel. I: Stewart. PS: Borch. T. (2015). Biocides in hydraulic fracturing fluids: a critical
   review of their usage, mobility, degradation, and toxicity. Environ Sci Technol 49:16-32.
   http://dx.doi.org/10.1021/es503724k

Kansas Water Office. (2014). How is water used in oil and gas exploration in Kansas?  Topeka, KA.
   http://www.kwo.org/about us/BACs/KWIF/rpt Hydraulic%20Fracturing KS Water FAQ 03082012  fina
   1 ki.pdf

Kappel. WM. (2013). Dissolved methane in groundwater, Upper Delaware River Basin, Pennsylvania and New
   York (pp. 1-6). (2013-1167). U. S. Geological Survey, http://pubs.usgs.gov/of/2013/1167/pdf/ofr2013-
   1167.pdf
               This document is a draft for review purposes only and does not constitute Agency policy.

June 2015                                        20                     DRAFT—DO NOT CITE OR QUOTE

-------
Hydraulic Fracturing Drinking Water Assessment                                                All References
Kappel. WM: Nystrom. EA. (2012). Dissolved methane in New York groundwater, 19992011. (Open-File
   Report 20121162). Washington, DC: U.S. Geological Survey, http://pubs.usgs.gov/of/2012/1162/

Kappel. WM: Williams. IH: Szabo. Z. (2013). Water resources and shale gas/oil production in the Appalachian
   basin critical issues and evolving developments. (Open-File Report 20131137). Troy, NY: U.S. Geological
   Survey, http://pubs.usgs.gov/of/2013/1137/pdf/ofr2013-1137.pdf

Kargbo. DM: Wilhelm. RG: Campbell. PI. (2010). Natural gas plays in the Marcellus Shale: Challenges and
   potential opportunities. Environ Sci Technol 44: 5679-5684. http://dx.doi.org/10.1021/es903811p

Kassotis. CD: Tillitt. DE: Wade Davis. I: Hormann. AM: Nagel. SC. (2014). Estrogen and androgen receptor
   activities of hydraulic fracturing chemicals and surface and ground water in a drilling-dense region.
   Endocrinology 155: 897-907. http://dx.doi.org/10.1210/en.2013-1697

Kaushal. SS: Groffman. PM: Likens. GE: Belt. KT: Stack. WP: Kelly. VR: Band. LE: Fisher. GT. (2005). Increased
   salinization of fresh water in the northeastern United States. PNAS 102: 13517-13520.
   http://dx.doi.org/10.1073/pnas.0506414102

Kell. S. (2011). State oil and gas agency groundwater investigations and their role in advancing regulatory
   reforms, a two-state review: Ohio and Texas. Ground Water Protection Council.
   http://fracfocus.org/sites/default/files/publications/state oil  gas  agency groundwater investigations o
   ptimized.pdf

Kelly. WR. (2008). Long-term trends in chloride concentrations in shallow aquifers near Chicago. Ground
   Water 46:  772-781. http://dx.doi.Org/10.llll/i.1745-6584.2008.00466.x

Kennedy/lenks Consultants. (2002). Evaluation of technical and economic feasibility of treating oilfield
   produced water to create a new water resource, http://www.gwpc.org/sites/default/files/event-
   sessions/Roger Funston PWC2002 O.pdf

Kenny. IF: Barber. NL: Hutson. SS: Linsey. KS: Lovelace. IK: Maupin. MA. (2009). Estimated use of water in the
   United States in 2005. (Circular 1344). Reston, VA: U.S. Geological Survey.
   http://pubs.usgs.gov/circ/1344/

Kharaka. YK: Kakouros. E: Abbott. MM. (2002). Environmental impacts of petroleum production: 1- The fate
   of inorganic and organic chemicals in produced water from the Osage-Skiatook Petroleum Environmental
   Research B site, Osage County, OK. 9th International Petroleum Environmental Conference, October 22-25,
   2002, Albuquerque, NM.

Kharaka. YK: Kakouros. E: Thordsen. II: Ambats. G: Abbott. MM. (2007). Fate and groundwater impacts of
   produced water releases at OSPER B site, Osage County, Oklahoma. Appl Geochem 22: 2164-2176.
   http://dx.doi.0rg/10.1016/j.apgeochem.2007.04.005

Kim. GH: Wang. IY. (2014). Interpretation of hydraulic fracturing pressure in tight gas formations. Journal of
   Energy Resources Technology 136: 032903. http://dx.doi.Org/10.1115/l.4026460

Kim. I: Moridis. Gl. (2013). Development of the T+M coupled flowgeomechanical simulator to describe
   fracture propagation and coupled flowthermalgeomechanical processes in tight/shale gas systems.
   Computers and Geosciences 60:184-198. http://dx.doi.Org/10.1016/j.cageo.2013.04.023

Kim. I: Moridis. Gl. (2015). Numerical analysis of fracture propagation during hydraulic fracturing operations
   in  shale gas systems. International Journal of Rock Mechanics and Mining Sciences 76:127-137.

Kim. I: Urn. ES: Moridis. Gl. (2014). Fracture propagation, fluid flow, and geomechanics of water-based
   hydraulic fracturing in shale gas systems and electromagnetic geophysical monitoring of fluid migration.
   SPE Hydraulic Fracturing Technology Conference, February 4-6, 2014, The Woodlands, Texas.

King. G: King.  D. (2013). Environmental risk arising from well-construction failure: Differences between
   barrier and well failure, and estimates of failure frequency across common well types, locations, and well
   age. SPE Prod Oper 28. http://dx.doi.org/10.2118/166142-PA
               This document is a draft for review purposes only and does not constitute Agency policy.

June 2015                                        21                    DRAFT—DO NOT CITE OR QUOTE

-------
Hydraulic Fracturing Drinking Water Assessment                                                All References
King. GE. (2010). Thirty years of gas shale fracturing: what have we learned? Society of Petroleum Engineers.
   http://dx.doi.org/10.2118/133456-MS

King. GE. (2012). Hydraulic fracturing 101: What every representative, environmentalist, regulator, reporter,
   investor, university researcher, neighbor and engineer should know about estimating frac risk and
   improving frac performance in unconventional gas and oil wells. SPE Hydraulic Fracturing Technology
   Conference, February 6-8, 2012, The Woodlands, TX.

Kirksey, I. (2013).  Optimizing wellbore integrity in well construction. Presentation presented at North
   American Wellbore Integrity Workshop, Octoberl6-17,2013, Denver, CO.

Klein, M: Kenealey, G: Makowecki, B. (2012). Comparison of hydraulic fracture fluids in multi-stage fracture
   stimulated horizontal wells in the Pembina Cardium formation. In 2012 SPE hydrocarbon economics and
   evaluation symposium. Richardson, TX: Society of Petroleum Engineers.
   http://dx.doi.org/10.2118/162916-MS

Knappe. D: Fireline. IL. (2012). Fracking 101: Shale gas extraction using horizontal drilling and hydraulic
   fracturing. Presentation presented at NCAWWA-WEA Annual Conference, November 14, 2012, Raleigh,
   NC.

Konikow. LF. (2013a). Groundwater depletion in the United States  (1900-2008): U.S. Geological Survey
   Scientific Investigations Report 2013-5079. Reston, VA: U.S. Geological Survey.
   http://pubs.usgs.gov/sir/2013/5079

Konikow. LF. (2013b). Groundwater depletion in the United States  (19002008). (USGS Scientific
   Investigations Report 2013). Reston, VA: U.S. Geological Survey, http://pubs.usgs.gov/sir/2013/5079

Konikow. LF: Kendy. E. (2005). Groundwater depletion: A global problem. Hydrogeo J13: 317-320.
   http://dx.doi.org/10.1007/sl0040-004-0411-8

Korfmacher. KS: Jones. WA: Malone. SL: Vinci. LF. (2013). Public health and high volume hydraulic fracturing.
   New Solutions: A Journal of Environmental and Occupational Health Policy 23:13-31.
   http://dx.doi.Org/10.2190/NS.23.l.c

Krasner. SW. (2009). The formation and control of emerging disinfection by-products of health concern
   [Review]. Philos Transact A Math Phys Eng Sci 367: 4077-4095. http://dx.doi.org/10.1098/rsta.2009.010

Kundert. D: Mullen. M. (2009). Proper evaluation of shale gas reservoirs  leads to a more effective hydraulic-
   fracture stimulation. Paper presented at SPE Rocky Mountain Petroleum Technology Conference, April 14-
   16, 2009, Denver, CO.

Kuthnert. N: Werline. R: Nichols. K. (2012). Water reuse and recycling in the oil and gas industry: Devons
   water management success. Presentation presented at 2nd Annual Texas Water Reuse Conference, July
   20, 2012, Forth Worth, TX.

Kuwayama, Y: Olmstead, S: Krupnick, A. (2015). Water quality and  quantity impacts of hydraulic fracturing.
   Current Sustainable/Renewable Energy Reports 2:17-24. http://dx.doi.org/10.1007/s40518-014-0023-4

LA Ground Water Resources Commission (Louisiana Ground Water Resources Commission). (2012).
   Managing Louisianas Groundwater Resources: An interim report to the Louisiana Legislature. Baton
   Rouge, LA: Louisiana Department of Natural Resources.
   http:// dnr.louisiana.gov/index. cfm?md=pagebuilder&tmp=home&pid=907

Lacazette. A: Geiser. P. (2013). Comment on Davies etal.,  2012 Hydraulic fractures: How far can they go?
   Marine and Petroleum Geology 43: 516-518. http://dx.doi.Org/10.1016/j.marpetgeo.2012.12.008

Lange. T: Sauter. M: Heitfeld. M: Schetelig. K: Brosig. K: lahnke. W: Kissinger. A: Helmig. R: Ebigbo. A: Class. H.
   (2013). Hydraulic fracturing in unconventional gas reservoirs: risks in the geological system part 1.
   Environmental Earth Sciences 70: 3839-3853. http://dx.doi.org/10.1007/sl2665-013-2803-3
               This document is a draft for review purposes only and does not constitute Agency policy.

June 2015                                        22                    DRAFT—DO NOT CITE OR QUOTE

-------
Hydraulic Fracturing Drinking Water Assessment                                                All References
Langmuir. D:Riese.AC. (1985). THE THERMODYNAMIC PROPERTIES OF RADIUM. Geochim Cosmo Act 49:
   1593-1601.

Laurenzi. II: Jersey. GR. (2013). Life cycle greenhouse gas emissions and freshwater consumption of Marcellus
   shale gas. Environ Sci Technol 47: 4896-4903. http://dx.doi.org/10.1021/es305162w

Lawal. H: Abolo. NU: lackson. G: Sahai. V: Flores. C. (2014). A quantitative approach to analyze fracture area
   loss in shale gas reservoirs. SPE Latin America and Caribbean Petroleum Engineering Conference, May 21-
   23, 2014, Maracaibo, Venezuela.

LDEO (Louisiana Department of Environmental Quality). (2008). Ground water use advisory: Commissioner
   of conservation recommends wise water use planning in the Haynesville Shale.
   http:// dnr.louisiana.gov/index. cfm?md=newsroom&tmp=detail&aid=509

LEau LLC. (2008).  Dew vaporation desalination 5,000-gallon-per-day pilot plant. (Desalination and Water
   Purification Research and Development Program Report No. 120). Denver, CO: Bureau of Reclamation, U.S.
   Department of the Interior. http://www.usbr.gov/research/AWT/reportpdfs/reportl20.pdf

LeBas, R: Lord, P: Luna, D: Shahan, T. (2013). Development and use of high-TDS recycled produced water for
   crosslinked-gel-based hydraulic fracturing. In 2013 SPE hydraulic fracturing technology conference.
   Richardson, TX: Society of Petroleum Engineers. http://dx.doi.org/10.2118/163824-MS

Lecampion. B: leffrey. R: Detournay. E. (2005). Resolving the geometry of hydraulic fractures from tilt
   measurements. Pure Appl Geophys 162: 2433-2452. http://dx.doi.org/10.1007/s00024-005-2786-4

Lee. PS: Herman. ID: Elsworth. D: Kim. HT: Lee. HS. (2011). A critical evaluation of unconventional gas
   recovery from the marcellus shale, northeastern United States. K S C E Journal of Civil Engineering 15:
   679-687. http://dx.doi.org/10.1007/sl2205-011-0008-4

Lee. K: Neff. I. (2011). Produced water: Environmental risks and advances in mitigation technologies. New
   York, NY: Springer, http://dx.doi.org/10.1007/978-l-4614-0046-2

Lefebvre. 0: Moletta. R. (2006). Treatment of organic pollution in industrial saline wastewater: a literature
   review [Review]. Water Res 40: 3671-3682. http://dx.doi.Org/10.1016/j.watres.2006.08.027

LePage, I: De Wolf, C: Bemelaar, I: Nasr-El-din, HA. (2013). An environmentally friendly stimulation fluid for
   high-temperature applications. SPE Journal 16:104-110. http://dx.doi.org/10.2118/121709-PA

Levis, E. (2011). Texas company pays $93,710 settlement for polluting Clearfield County Creek. Pennsylvania
   Fish and Boat Commission. https://www.fish.state.pa.us/newsreleases/2011press/eog-settlement.htm

Linaric.  M: Markic. M: Sipos. L. (2013). High salinity wastewater treatment. Water Sci Technol 68:1400-1405.
   http://dx.doi.org/10.2166/wst.2013.376

Linkov. I: Welle. P: Loney. D: Tkachuk. A: Canis. L: Kim. IB: Bridges. T. (2011). Use of multicriteria decision
   analysis to support weight of evidence evaluation. RiskAnal 31:1211-1225.
   http://dx.doi.0rg/10.llll/i.1539-6924.2011.01585.x

Llewellyn, GT. (2014). Evidence and mechanisms for Appalachian Basin brine migration to shallow aquifers in
   NE Pennsylvania, USA.  Hydrogeo J 22:1055-1066. http://dx.doi.org/10.1007/sl0040-014-1125-l

Lowe. T: Potts. M: Wood. D. (2013). A case history of comprehensive hydraulic fracturing monitoring in the
   Cana Woodford. In 2013 SPE annual technical conference and exhibition. Richardson, TX: Society of
   Petroleum Engineers. http://dx.doi.org/10.2118/166295-MS

Lustgarten, A. (2009). Frack fluid spill in Dimock contaminates stream, killing fish. Available online at
   http://www.propublica.org/article/frack-fluid-spill-in-dimock-contaminates-stream-killing-fish-921

Lutz, BD: Lewis, AN:  Doyle, MW. (2013). Generation, transport, and disposal of wastewater associated with
   Marcellus Shale gas development. Water Resour Res 49: 647-656.
               This document is a draft for review purposes only and does not constitute Agency policy.

June 2015                                        23                    DRAFT—DO NOT CITE OR QUOTE

-------
Hydraulic Fracturing Drinking Water Assessment                                                All References
Ma. G: Geza. M: Xu. P. (2014). Review of flowback and produced water management, treatment, and beneficial
   use for major shale gas development basins. Shale Energy Engineering Conference 2014, Pittsburgh,
   Pennsylvania, United States.

MacDonald. Rl: Frantz. IH: Schlotterbeck. ST: Adams. B: Sikorski. D. (2003). An update of recent production
   responses obtained from Devonian shale and Berea wells stimulated with nitrogen foam (with proppant)
   vs. nitrogen-only, Pike Co., KY. Richardson, TX: Society of Petroleum Engineers.
   http://dx.doi.org/10.2118/84834-MS

Malone. M: Ely. IW. (2007). Execution of hydraulic fracturing treatments. In M Economides; T Martin (Eds.),
   Modern fracturing: enhancing natural gas production (pp. 323-360). Houston, TX: ET Publishing.

Malone. S: Kelso. M: Auch. T: Edelstein. K: Ferrar. K: lalbert. K. (2015). Data inconsistencies from states with
   unconventional oil and gas activity. J Environ Sci Health A Tox Hazard Subst Environ Eng 50: 501-510.
   http://dx.doi.org/10.1080/10934529.2015.992678

Maloney. KO: Yoxtheimer. DA. (2012). Production and disposal of waste materials from gas and oil extraction
   from the Marcellus shale play in Pennsylvania. Environmental Practice 14: 278-287.
   http://dx.doi.org/10.10170S146604661200035X

Mantell. ME.  (2011). Produced water reuse and recycling challenges and opportunities across major shale
   plays. Presentation presented at Hydraulic Fracturing Study: March 2011 Technical Workshop #4 on
   Water Resources Management, March 29-30, 2011, Arlington, VA.

Mantell. ME.  (2013a).  Recycling andreuse of produced water to reduce freshwater use in hydraulic fracturing
   operations. Presentation presented at Water acquisition  modeling: Assessing impacts through modeling
   and other means Technical workshop on wastewater treatment and related modeling, June 4, 2013,
   Arlington, VA.

Mantell. ME.  (2013b). Recycling and reuse of produced water to reduce freshwater use in hydraulic fracturing
   operations. In Summary of the technical workshop on water acquisition modeling: Assessing impacts
   through modeling and other means (pp. A20-A27). Washington, D.C.: U.S. Environmental Protection
   Agency, http://www2.epa.gov/hfstudy/summary-technical-workshop-water-acquisition-modeling-
   assessing-impacts-through-modeling-and

Marshak, S. (2004). Essentials of geology (1st ed.). New York, NY: W.W.  Norton and Company.

Martin. T: Valko. P. (2007). Hydraulic fracture design for production enhancement. In M Economides; T
   Martin (Eds.), Modern fracturing enhancing natural gas production.  Houston, TX: ET Publishing.

Matthews. 1C: Li. S: Swann. CT: Ericksen. RL. (2006). Incubation with moist top soils enhances solubilization of
   radium and other components from oilfield scale and sludge: Environmental concerns from Mississippi.
   Environmental Geosciences 13: 43-53.

Maule. AL: Makey. CM: Benson. EB: Burrows. II:  Scammell. MK. (2013). Disclosure of hydraulic fracturing fluid
   chemical  additives: analysis of regulations. New Solutions: A Journal of Environmental and Occupational
   Health Policy 23:167-187. http://dx.doi.Org/10.2190/NS.23.l.j

Maupin. MA: Kenny. IF: Hutson. SS: Lovelace. IK: Barber. NL: Linsey. KS. (2014). Estimated use of water in the
   United States in 2010. (USGS Circular 1405). Reston, VA:  U.S. Geological Survey.
   http://dx.doi.org/10.3133/cirl405

McDaniel, I: Watters, L: Shadravan, A. (2014). Cement sheath durability: Increasing cement sheath integrity to
   reduce gas migration in the Marcellus Shale  Play. In SPE hydraulic fracturing technology conference
   proceedings. Richardson, TX: Society of Petroleum  Engineers. http://dx.doi.org/10.2118/168650-MS

McElreath. D. (2011). Comparison of hydraulic fracturing fluids composition with produced formation water
   following fracturing Implications for fate and transport. Presentation presented at Hydraulic Fracturing
   Study: March 2011 Technical Workshop on Fate and Transport, March 2829,2011, Arlington, VA.
               This document is a draft for review purposes only and does not constitute Agency policy.

June 2015                                        24                    DRAFT—DO NOT CITE OR QUOTE

-------
Hydraulic Fracturing Drinking Water Assessment                                               All References
Mclntosh. 1C: Walter. LM. (2005). Volumetrically significant recharge of Pleistocene glacial meltwaters into
   epicratonic basins: Constraints imposed by solute mass balances. Chem Geol 222: 292-309.
   http://dx.doi.0rg/10.1016/j.chemgeo.2005.07.010

Mclntosh. 1C: Walter. LM: Martini. AM. (2002). Pleistocene recharge to midcontinent basins: effects on salinity
   structure and microbial gas generation. Geochim Cosmo Act 66:1681-1700.
   http://dx.doi.org/10.1016/S0016-7037(01)00885-7

McKay, SF: King, Al. (2006). Potential ecological effects of water extraction in small, unregulated streams.
   River Research and Applications 22:1023-1037. http://dx.doi.org/10.1002/rra.958

McKenzie, LM: Guo, R: Witter, RZ: Savitz, DA: Newman, L: Adgate, IL. (2014). Birth outcomes and maternal
   residential proximity to natural gas development in rural Colorado. Environ Health Perspect 122: 412-
   417. http://dx.doi.org/10.1289/ehp.1306722

MCOR (Marcellus Center for Outreach and Research). (2012). Extent and thickness of Marcellus Shale.
   University Park, PA: Pennsylvania State University. Retrieved from
   http://www.marcellus.psu.edu/images/Marcellus thickness.gif

Methanol Institute. (2013). Methanol safe handling manual. Alexandria, VA.
   http://www.methanol.org/Health-And-Safety/Safe-Handling/Methanol-Safe-Hanlding-Manual.aspx

Michie. TW: Koch. CA. (1991). Evaluation of injection-well risk management in the Williston Basin. J Pet Tech
   43: 737-741. http://dx.doi.org/10.2118/20693-PA

Minnich. K. (2011). A water chemistry perspective on flowback reuse with several case studies. Minnich, K.
   http://www2.epa.gov/sites/production/files/documents/10 Minnich - Chemistry 508.pdf

Miskimins, IL. (2008). Design and life cycle considerations for unconventional reservoir wells. In 2008 SPE
   Unconventional Reservoirs Conference. Richardson, TX: Society of Petroleum Engineers.
   http://dx.doi.org/10.2118/114170-MS

Mitchell. AL: Small. M: Gasman. EA. (2013a). Surface water withdrawals for Marcellus Shale gas development:
   performance of alternative regulatory approaches in the Upper Ohio River Basin.  Environ Sci Technol 47:
   12669-12678. http://dx.doi.org/10.1021/es403537z

Mitchell. Bl. (1970) Viscosity of foam. (Doctoral Dissertation). The University of Oklahoma,

Mitchell, I: Pabon, P: Collier, ZA: Egeghy, PP: Cohen-Hubal, E: Linkov, I: Vallero, DA. (2013b). A decision
   analytic approach to exposure-based chemical prioritization. PLoS ONE 8: e70911.
   http://dx.doi.org/1371/journal.pone.0070911

Molofsky. LI: Connor. IA: Wylie. AS: Wagner. T: Farhat. SK. (2013). Evaluation of methane sources in
   groundwater  in northeastern Pennsylvania. Ground Water 51: 333-349.
   http://dx.doi.org/10.llll/gwat.12056

Montana GWIC (Montana Ground Water Information Center). (2009). Coal bed methane powder river basin:
   Montana Bureau of Mines and Geology Ground Water Information Center. Butte, MT.

Montgomery. C. (2013). Fracturing fluid components. In A Bunder; J McLennon; R Jeffrey (Eds.), Effective and
   Sustainable Hydraulic Fracturing. Croatia: InTech. http://dx.doi.org/10.5772/56422

Montgomery. CT: Smith. MB. (2010). Hydraulic fracturing - History of an enduring technology. J Pet Tech 62:
   26-32.

Morillon, A: Vidalie, IF: Syahnudi, U: Suripno, S: Hadinoto, EK. (2002). Drilling and waste management; SPE
   73931. Presentation presented at The SPE International Conference on Health, Safety, and Environment in
   Oil and Gas Exploration and Production, March 20-22, 2002, Kuala Lumpur,  Malaysia.

Moudgal. Cl: Venkatapathy. R: Choudhury. H: Bruce. RM: Lipscomb. 1C. (2003). Application of QSTRs in the
   selection of a  surrogate toxicity value for chemical of concern. Environ Sci Technol 37: 5228-5235.
               This document is a draft for review purposes only and does not constitute Agency policy.

June 2015                                        25                    DRAFT—DO NOT CITE OR QUOTE

-------
Hydraulic Fracturing Drinking Water Assessment                                                All References
Mouser. P: Liu. S: Cluff. M: McHugh. M: Lenhart. I: MacRae. I. (In Press) Biodegradation of hydraulic fracturing
   fluid organic additives in sediment-groundwater microcosms.

Muehlenbachs. L: Spiller. E: Timmins. C. (2012). Shale gas development and property values: Differences
   across drinking water sources. (NBER Working Paper No. 18390). Cambridge, MA: National Bureau of
   Economic Research, http://www.nber.org/papers/wl8390

Mukherjee. H: Poe jr.. B: Heidt. I: Watson. T: Barree. R. (2000). Effect of pressure depletion on fracture-
   geometry evolution and production performance. SPE Prod Facil 15:144-150.
   http://dx.doi.org/10.2118/65064-PA

Murali Mohan, A: Hartsock, A: Hammack, RW: Vidic, RD: Gregory, KB. (2013). Microbial communities in
   flowback water impoundments from hydraulic fracturing for recovery of shale gas. FEMS Microbiol Ecol.
   http://dx.doi.org/10.llll/1574-6941.12183

Murdoch. PS: Baron. IS: Miller. TL. (2000). Potential effects of climate chance on surface-water quality in
   North America. J Am Water Resour Assoc 36: 347-366.

Murray, KE. (2013). State-scale perspective on water use and production associated with oil and gas
   operations, Oklahoma, U.S. Environ Sci Technol 47: 4918-4925. http://dx.doi.org/10.1021/es4000593

Myers. T. (2012a). Author's reply. Ground Water 50: 828-830. http://dx.doi.Org/10.llll/j.1745-
   6584.2012.00991.x

Myers. T. (2012b). Potential contaminant pathways from hydraulically fractured  shale to aquifers. Ground
   Water 50: 872-882. http://dx.doi.Org/10.llll/j.1745-6584.2012.00933.x

Myers. T. (2013). Author's reply for comments on potential contaminant pathways from hydraulically
   fractured shale to aquifers' [Comment]. Ground Water 51: 319321.
   http://dx.doi.org/10.llll/gwat.12016

National Drought Mitigation Center. (2015). U.S. drought monitor. Available online at
   http://droughtmonitor.unl.edu/Home.aspx (accessed February 27, 2015).

Neff. IM. (2002). Bioaccumulation in marine organisms:  Effect of contaminants from oil well produced water.
   Amsterdam: Elsevier.

Nelson. AW: May. D:  Knight. AW: Eitrheim. ES: Mehrhoff. M: Shannon. R: Litman. R: Schultz. MK. (2014).
   Matrix complications in the determination of radium levels in hydraulic fracturing flowback water from
   Marcellus Shale. 1: 204-208. http://dx.doi.org/10.1021/ez5000379

NETL (National Energy Technology Laboratory). (2013). Modern shale gas  development in the United States:
   An update. Pittsburgh, PA: U.S. Department of Energy. National Energy Technology Laboratory.
   http://www.netl.doe.gov/File%20Library/Research/Oil-Gas/shale-gas-primer-update-2013.pdf

NETL (National Energy Technology Laboratory). (2014). Evaporation [Fact Sheet]. Pittsburgh, PA: US
   Department of Energy, http://www.netl.doe.gov/research/coal/crosscutting/pwmis/tech-desc/evap

New Mexico Bureau  of Mines and Mineral Resources. (1994). Coalbed methane in the upper cretaceous
   fruitland formation, San Juan Basin, New Mexico and Colorado. Socorro, NM: New Mexico Bureau of Mines
   and Mineral Resources,  New Mexico Institute of Mining and Technology.
   https://geoinfo.nmt.edu/publications/monographs/bulletins/146/

Newell. R. (2011). Shale gas and the outlook for U.S. natural gas markets and global gas resources.
   Presentation presented at US EIA presentation at OECD Meetings, June 21, 2011, Paris, France.
               This document is a draft for review purposes only and does not constitute Agency policy.

June 2015                                        26                     DRAFT—DO NOT CITE OR QUOTE

-------
Hydraulic Fracturing Drinking Water Assessment                                               All References
Nicot. IP: Hebel. AK: Ritter. SM: Walden. S: Baier. R: Galusky. P: Beach. I: Kyle. R: Symank. L: Breton. C. (2011).
   Current and projected water use in the Texas mining and oil and gas industry - Final Report. (TWDB
   Contract No. 0904830939). Nicot, JP; Hebel, AK; Ritter, SM; Walden, S; Baier, R; Galusky, P; Beach, J; Kyle,
   R; Symank, L; Breton, C.
   http://www.twdb.texas.gov/publications/reports/contracted reports/doc/0904830939 MiningWaterUs
   e.pdf

Nicot. IP: Reedy. RC: Costley. RA: Huang. Y. (2012). Oil & gas water use in Texas: Update to the 2011 mining
   water use report. Nicot, JP; Reedy, RC; Costley, RA; Huang, Y.
   http://www.twdb.state.tx.us/publications/reports/contracted reports/doc/0904830939 2012Update M
   iningWaterUse.pdf

Nicot. IP: Scanlon. BR. (2012). Water use for shale-gas production in Texas, U.S. Environ Sci Technol 46: 3580-
   3586. http://dx.doi.org/10.1021/es204602t

Nicot, IP: Scanlon, BR: Reedy, RC: Costley, RA. (2014). Source and fate of hydraulic fracturing water in the
   Barnett Shale: a historical perspective. Environ Sci Technol 48: 2464-2471.
   http://dx.doi.org/10.1021/es404050r

NM OSE (New Mexico Office of the State Engineer). (2013). New Mexico water use by categories 2010.
   (Technical Report 54). Santa Fe, NM: New Mexico Office of the State Engineer, Water Use and
   Conservation Bureau.
   http://www.ose.state.nm.us/Pub/TechnicalReports/TechReport%2054NM%20Water%20Use%20by%2
   OCategories%20.pdf

North Dakota Department of Health. (2015). Oil field environmental incident summary, incident
   20150107160242. Available online at
   http://www.ndhealth.gov/EHS/FOIA/Spills/Summary  Reports/20150107160242 Summary Reportpdf

North Dakota Department of Mineral Resources. (2013). North Dakota Department of Mineral Resources:
   Government Finance Interim Committee 12/12/2013. Presentation presented at Department of Mineral
   Resources: Update on the Status of Oil and Gas Development in the State, 12/12/2013, Bismarck, ND.

North Dakota State Water Commission. (2010). Water appropriation requirements, current water use,  &
   water availability for energy industries in North Dakota: a 2010 summary. Bismarck, ND.
   http://www.swc.nd.gov/4dlink9/4dcgi/GetContentPDF/PB-1800/W&E%20RPT%20FinalR.pdf

North Dakota State Water Commission. (2014). Facts about North Dakota tracking and water use. Bismarck,
   ND. http://www.swc.nd.gov/4dlink9/4dcgi/GetContentPDF/PB-2419/Fact%20Sheet.pdf

Nowak. N:  Bradish. I. (2010). High density polyethylene (HOPE) lined produced water evaporation ponds.
   Presentation presented at 17th International Petroleum and Biofuels Environmental Conference, August
   31-September 2, 2010, San Antonio, TX.

NPC (National Petroleum Council). (2011a). Management of produced water from oil and gas wells. (Paper
   #2-17). Washington, DC. http://www.npc.org/Prudent  Development-Topic Papers/2-
   17 Management of Produced Water Paper.pdf

NPC (National Petroleum Council). (2011b). Plugging and abandonment of oil and gas wells. (Paper #2-25).
   Washington, DC: National Petroleum Council (NPC).

NRC (National Research Council). (2010). Management and effects of coalbed methane produced water in the
   western United States. Washington, DC: National Academies Press.
   http://www.nap.edu/catalog.php7record id=12915

NRC (National Research Council). (2012). Water reuse: Potential for expanding the nations water supply
   through reuse of municipal wastewater. Committee on the Assessment of Water Reuse as an Approach for
   Meeting Future Water Supply Need. Washington, DC: The National Academies Press.
   http://www.nap.edu/openbook.php7record id=13303
               This document is a draft for review purposes only and does not constitute Agency policy.

June 2015                                       27                    DRAFT—DO NOT CITE OR QUOTE

-------
Hydraulic Fracturing Drinking Water Assessment                                               All References
NRG (National Research Council). (2014). A framework to guide selection of chemical alternatives.
   Washington, D.C.: The National Academies Press, http://www.nap.edu/catalog/18872/a-framework-to-
   guide-selection-of-chemical-alternatives

NSTC (National Science and Technology Council). (2000). Integrated assessment of hypoxia in the Northern
   Gulf of Mexico. Washington, DC: U.S. National Science and Technology Council, Committee on Environment
   and Natural Resources, http://oceanservice.noaa.gov/products/hypox final.pdf

NYSDEC (New York State Department of Environmental Conservation). (2011). Revised draft supplemental
   generic environmental impact statement (SGEIS) on the oil, gas and solution mining regulatory program:
   Well permit issuance for horizontal drilling and high-volume hydraulic fracturing to develop the Marcellus
   shale and other low-permeability gas reservoirs. Albany, NY: NY SDEC.
   http://www.dec.ny.gov/energy/75370.html

ODNR. DMRM. (Ohio Department of Natural Resources, Division of Mineral Resources Management). (2008).
   Report on the investigation of the natural  gas invasion of aquifers in Bainbridge Township of Geauga
   County, Ohio. Columbus, OH: ODNR.
   http://oilandgas.ohiodnr.gov/portals/oilgas/pdf/bainbridge/report.pdf

Ohio EPA (Ohio Environmental Protection Agency). (2012a). Considerations for public water systems prior to
   providing raw or treated water to oil and natural gas companies.
   http://www.epa.state.oh.us/Portals/0/general%20pdfs/Considerations%20for%20Public%20Water%2
   OSystems%2 OPrior%2 Oto%2 OProviding%2 ORaw%2 Oor%2 OTreated%2 OWater%2 Oto%2 00il%2 Oand%2
   ONatural%20Gas%2 OCompanies.pdf

Ohio EPA (Ohio Environmental Protection Agency). (2012b). Ohios regulations:  a guide for operators drilling
   in the Marcellus and Utica Shales. Columbus, OH.
   http://www.epa.state.oh.us/Portals/0/general%20pdfs/Ohio%20Regulations%20-
   %2 OA%2 OGuide%2 Ofor%2 00perators%2 ODrilling%2 Oin%2 Othe%2 OMarcellus%2 Oand%2 OUtica%2 OSh
   ales.pdf

Oil and Gas Mineral Services. (2010). MineralWise: Oil and gas terminology. Available online at
   http://www.mineralweb.com/library/oil-and-gas-terms/

Olawoyin. R: Wang. IY: Oyewole. SA. (2013). Environmental safety assessment of drilling operations in the
   Marcellus-shale gas development. S P E Drilling & Completion 28: 212-220.
   http://dx.doi.org/10.2118/163095-PA

Olmstead. SM: Muehlenbachs.  LA: Shih. IS: Chu. Z: Krupnick. Al. (2013). Shale gas development impacts on
   surface water quality in Pennsylvania. PNAS 110: 4962-4967.
   http://dx.doi.org/10.1073/pnas.1213871110

Olson, IE. (2011). Hydraulic fracturing overview. Presentation presented at Summer Institute B: Energy,
   Climate and Water in the 21st Century, TXESS Revolution, Texas Earth and Space Science Revolution
   Professional Development  for Educators, June, 2011, Austin, TX.

Olsson. 0: Weichgrebe. D: Rosenwinkel. KH. (2013). Hydraulic fracturing wastewater in Germany:
   composition, treatment, concerns. Environmental Earth Sciences 70: 3895-3906.
   http://dx.doi.org/10.1007/sl2665-013-2535-4

OMB (U.S. Office of Management and Budget). (2004). Final information quality bulletin for peer review.
   Washington, DC: US Office of Management and Budget (OMB).
   http://www.whitehouse.gov/sites/default/files/omb/assets/omb/memoranda/fy2005/m05-03.pdf

Orem. W: Tatu. C: Varonka. M:  Lerch. H: Bates. A: Engle. M: Crosby. L: Mcintosh. I. (2014). Organic substances
   in produced and formation water from unconventional natural gas  extraction in coal and shale. Int J Coal
   Geol 126: 20-31. htto://dx.doi.org/10.1016/i.coal.2014.01.003
              This document is a draft for review purposes only and does not constitute Agency policy.

June 2015                                       28                    DRAFT—DO NOT CITE OR QUOTE

-------
Hydraulic Fracturing Drinking Water Assessment                                               All References
Orem. WH: Tatu. CA: Lerch. HE: Rice. CA: Bartos. TT: Bates. AL: Tewalt. S: Corum. MD. (2007). Organic
   compounds in produced waters from coalbed natural gas wells in the Powder River Basin, Wyoming, USA.
   Appl Geochem 22: 2240-2256. http://dx.doi.Org/10.1016/j.apgeochem.2007.04.010

Osborn. SG: Vengosh. A: Warner. NR: lackson. RB. (2011). Methane contamination of drinking water
   accompanying gas-well drilling and hydraulic fracturing. PNAS 108: 8172-8176.
   http://dx.doi.org/10.1073/pnas.1100682108

OSHA (Occupational Safety & Health Administration). (2014a). Personal communication: email exchanges
   between Tandy Zitkus, OSHA and Rebecca Daiss, U.S. EPA. Available online

OSHA (Occupational Safety & Health Administration). (2014b). Personal communication: phone conversation
   between Tandy Zitkus, OSHA and Rebecca Daiss, U.S. EPA. Available online

OSHA (Occupational Safety & Health Administration). (2015). Oil and gas well drilling and servicing etool:
   Well completion. Available online at
   https://www.osha.gov/SLTC/etools/oilandgas/well completion/well completion.html

Otton, IK: Zielinski, RA: Smith, BD: Abbott, MM. (2007). Geologic controls on movement of produced-water
   releases at US geological survey research Site A, Skiatook lake, Osage county, Oklahoma. Appl Geochem 22:
   2138-2154. http://dx.doi.0rg/10.1016/j.apgeochem.2007.04.015

OWRB (Oklahoma Water Resources Board). (2014). The Oklahoma comprehensive water plan. Available
   online at http://www.owrb.ok.gov/supply/ocwp/ocwp.php

PA DCNR (Pennsylvania Department of Conservation and Natural Resources). (2015). Thermal maturation
   and petroleum generation. Available online at
   http://www.dcnr.state.pa.us/topogeo/econresource/oilandgas/marcellus/sourcerock index/sourcerock
   maturation/index.htm (accessed April 9, 2015).

PA PEP (Pennsylvania Department of Environmental Protection). (2000). Pennsylvanias plan for addressing
   problem abandoned wells and orphaned wells. Harrisburg, PA: PADEP.

PA PEP (Pennsylvania Department of Environmental Protection). (2009a). Inspection Report, inspection
   record #1835041, enforcement record #251134. Harrisburg, PA: Commonwealth of Pennsylvania
   Department of Environmental Protection, Oil and Gas Management Program.

PA PEP (Pennsylvania Department of Environmental Protection). (2009b). Stray natural gas migration
   associated with oil and gas wells [draft report]. Harrisburg, PA.
   http://www.dep.state.pa.us/dep/subject/advcoun/oil gas/2009/Stray%20Gas%20Migration%20Cases.p
   df
PADEP (Pennsylvania Department of Environmental Protection). (2010). DEP Fines Atlas Resources for
   drilling wastewater spill in Washington County. Available online at
   http://www.portal.state.pa.us/portal/server.pt/community/newsroom/14287?id=13595&typeid=l
   (accessed February 13, 2014).

PADEP (Pennsylvania Department of Environmental Protection). (2 01 la). Letter from Pennsylvania
   Department of the Environment to US EPA Region 3 Administrator Shawn Garvin. Available online at
   http://www.epa.gov/region3/marcellus shale?#inforeqsbypadep

PA DEP. Road-spreading of brine for dust control and road stabilization. 43 Pa.B. 7377 § 78.70 (2011b).
   http://www.pabulletin.com/secure/data/vol43/43-50/2362b.html

PADEP (Pennsylvania Department of Environmental Protection). (2011c). Surface water sample analytical
   results from XTO 308 response data from XTO February 3, 2011 CAWP addendum. Indiana, PA: XTO
   Energy.
               This document is a draft for review purposes only and does not constitute Agency policy.

June 2015                                       29                    DRAFT—DO NOT CITE OR QUOTE

-------
Hydraulic Fracturing Drinking Water Assessment                                               All References
PA PEP (Pennsylvania Department of Environmental Protection). (2015a). PA DEP oil & gas reporting
   website, statewide data downloads by reporting period, waste and production files downloaded for
   Marcellus/unconventional wells, July 2009 December 2014. Harrisburg, PA. Retrieved from
   https://www.paoilandgasreporting.state.pa.us/publicreports/Modules/DataExports/DataExports.aspx

PA DEP (Pennsylvania Department of Environmental Protection). (2015b). Technologically enhanced
   naturally occurring radioactive materials (TENORM) study report. Harrisburg, PA.
   http://www.elibrary.dep.state.pa.us/dsweb/Get/Document-105822/PA-DEP-TENORM-
   Study Report Rev. 0 01-15-2015.pdf

Palmer. ID: Moschovidis. ZA: Cameron. IR. (2005). Coal failure and consequences for coalbed methane wells.
   Paper presented at SPE annual technical conference and exhibition, October 9-12, 2005, Dallas, TX.

Papoulias. DM: Velasco. AL. (2013). Histopathological analysis offish from Acorn Fork Creek, Kentucky,
   exposed to hydraulic fracturing fluid releases. Southeastern Naturalist 12: 92-111.

Parker. KM: Zeng. T: Harkness. I: Vengosh. A: Mitch. WA. (2014). Enhanced formation of disinfection
   byproducts in shale gas wastewater-impacted drinking water supplies. Environ Sci Technol 48:11161-
   11169. http://dx.doi.org/10.1021/es5028184

Pashin. 1C: Mcintyre-Redden. MR: Mann. SD: Kopaska-Merkel. DC: Varonka. M: Orem. W. (2014). Relationships
   between water and gas chemistry in mature coalbed methane reservoirs of the Black Warrior Basin. Int J
   Coal Geol 126: 92-105. http://dx.doi.Org/10.1016/j.coal.2013.10.002

Patel. PS: Robart. Cl: Ruegamer. M: Yang. A. (2014). Analysis of US hydraulic fracturing fluid system and
   proppant trends. In 2014 SPE hydraulic fracturing technology conference. Richardson, TX:  Society of
   Petroleum Engineers. http://dx.doi.org/10.2118/168645-MS

Paterra, P. (2011). DEP shuts down Tri-County Waste Water over illegal dumping. Available online at
   http://triblive.eom/x/pittsburghtrib/news/regional/s 728516.html#axzz3UCvkvM7t (accessed March
   12,2015).

Patzek. TW: Male. F: Marder. M. (2013). Gas production in the Barnett Shale obeys a simple scaling theory.
   PNAS 110: 19731-19736. http://dx.doi.org/10.1073/pnas.1313380110

Pearson, CM: Griffin, L: Wright, CA: Weijers, L.  (2013). Breaking up is hard to do: creating hydraulic fracture
   complexity in the Bakken central basin. In 2013 SPE hydraulic fracturing technology conference.
   Richardson, TX: Society of Petroleum Engineers. http://dx.doi.org/10.2118/163827-MS

Pearson. K. (2011). Geologic models and  evaluation of undiscovered conventional and continuous oil and gas
   resourcesUpper Cretaceous Austin Chalk, U.S. Gulf Coast. In US Geological Survey Scientific Investigations
   Report. (20125159). U.S. Geological Survey, http://pubs.usgs.gov/sir/2012/5159/

Penttila. B: Heine. L: Craft. E. (2013). Manuscript in preparation assessing the hazard data gap for hydraulic
   fracturing chemicals. Penttila, B; Heine, L; Craft, E.

Peraki. M: Ghazanfari. E. (2014). Electrodialysis treatment of flow-back water for environmental protection in
   shale gas development. In Shale gas development Shale energy engineering 2014. Reston, VA: American
   Society of Civil Engineers, http://dx.doi.org/10.1061/9780784413654.008

Peterman. ZE: Thamke. I: Futa. K: Oliver. T. (2012). Strontium isotope evolution of produced water in the East
   Poplar Oil Field, Montana. Presentation presented at US Geological Survey AAPG annual convention and
   exhibition, April 23, 2012, Long Beach, California.

Phillips. A. (2014). Frackers spill Olympic pools worth of hydrochloric acid in Oklahoma. Available online at
   http://thinkprogress.org/climate/2014/07/31/3466283/olympic-pool-sized-hydrochloric-acid-spill-
   Oklahoma/

Finder. GF: Celia. MA.  (2006). Subsurface hydrology. Hoboken, NJ: John Wiley & Sons, Inc.
   http://dx.doi.org/10.1002/0470044209
               This document is a draft for review purposes only and does not constitute Agency policy.

June 2015                                        30                    DRAFT—DO NOT CITE OR QUOTE

-------
Hydraulic Fracturing Drinking Water Assessment                                                All References
Finder. GF: Gray. WG. (2008). Essentials of multiphase flow and transport in porous media. Hoboken, NJ: John
   Wiley & Sons.

Plumlee. MH: Debroux. IF: Taffler. D: Graydon. IW: Mayer. X: Dahm. KG: Hancock. NT: Guerra. KL: Xu. P:
   Drewes. IE: Cath. TY. (2014). Coalbed methane produced water screening tool for treatment technology
   and beneficial use. 5: 22-34. http://dx.doi.Org/10.1016/j.juogr.2013.12.002

Plummer. M: Wood. T: Huang. H: Guo. L: Reiten. I: Chandler. K: Metesh. I. (2013). Water needs and availability
   for hydraulic fracturing in the Bakken formation, eastern Montana. Presentation presented at 2013
   Technical Workshop, Water Acquisition Modeling: Assessing Impacts Through Modeling and Other Means,
   June 4, 2013, Arlington, VA.

Porcelli. D: Kim. CK: Martin. P: Moore. WS: Phaneuf. M. (2014). Properties of radium. In The environmental
   behaviour of radium: revised edition. Vienna, Austria: International Atomic Energy Agency. http://www-
   pub.iaea.org/MTCD/Publications/PDF/trs476 web.pdf

Powell. B. (2013). Secondary containment: regulations and best management practices in the Appalachian
   Basin. (AADE-13-FTCE-18). Houston, Texas: American Association of Drilling Engineers.

Powers. SE: Hunt. CS: Heermann. SE: Corseuil. HX: Rice. D: Alvarez. Pll. (2001). The transport and fate of
   ethanol BTEX in groundwater contaminated by gasohol. Environ Sci Technol 31: 79-123.
   http://dx.doi.org/10.1080/20016491089181

Purestream (Purestream Services). (2013). Purestream services will begin commercial operations to treat
   Eagle  Ford Shale produced and frac flow-back water in Gonzalez County, Texas. Retrieved from
   http://purestreamtechnology.com/index.php/component/content/article/72-press-releases/206-
   purestream-services-will-begin-commercial-operations-to-treat-eagle-ford-shale-produced-and-frac-
   flow-back-water-in-gonzalez-county-texas

Rahm. BG: Bates. IT: Bertoia. LR: Galford. AE: Yoxtheimer. DA: Riha. SI. (2013). Wastewater management and
   Marcellus Shale gas development: trends, drivers, and planning implications. J Environ Manage 120:  105-
   113. http://dx.doi.0rg/10.1016/j.jenvman.2013.02.029

Rahm. BG: Riha. SI. (2012). Toward strategic management of shale gas development: Regional, collective
   impacts on water resources. Environ Sci Pol 17:12-23. http://dx.doi.Org/10.1016/j.envsci.2011.12.004

Rahm. BG: Riha. SI. (2014). Evolving shale gas management: water resource risks, impacts, and lessons
   learned [Review]. Environ Sci Process Impacts 16:1400-1412. http://dx.doi.org/10.1039/c4em00018h

Rahm. BG: Vedachalam. S: Bertoia. LR: Mehta. D: Vanka. VS: Riha. SI. (2015). Shale gas operator violations in
   the Marcellus and what they tell us about water resource risks. Energy Policy 82:1-11.
   http://dx.doi.0rg/10.1016/j.enpol.2015.02.033

Ramirez.  P. (2002). Oil field produced water discharges into wetlands in Wyoming. (97-6-  6F34). Cheyenne,
   WY: U.S. Fish and Wildlife Service, http://www.fws.gov/mountain-
   prairie/contaminants/papers/r6718c02.pdf

Rasa. E: Bekins.  BA: Mackay. DM: de Sieves. NR: Wilson. IT: Feris. KP: Wood. IA: Scow. KM.  (2013). Impacts of
   an ethanol-blended fuel release on groundwater and fate of produced methane: Simulation of field
   observations. Water Resour Res 49: 4907-4926. http://dx.doi.org/10.1002/wrcr.20382

Rassenfoss. S. (2011). From flowback to fracturing: Water recycling grows in the Marcellus shale. J Pet Tech
   63: 48-51.

Reagan. MT:  Moridis. Gl: lohnson. IN: Keen. ND. (2015). Numerical simulation of the environmental impact of
   hydraulic fracturing of tight/shale gas reservoirs on near-surface groundwater: background, base cases,
   shallow reservoirs, short-term gas and water transport. Water Resour Res 51:1-31.
   http://dx.doi.org/10.1002/2014WR016086
               This document is a draft for review purposes only and does not constitute Agency policy.

June 2015                                        31                     DRAFT—DO NOT CITE OR QUOTE

-------
Hydraulic Fracturing Drinking Water Assessment                                               All References
Renpu. W. (2011). Advanced well completion engineering (Third ed.). Houston, TX: Gulf Professional
   Publishing.

Repetski. IE: Ryder. RT: Weary. PI: Harris. AG: Trippie. MH. (2008). Thermal maturity patterns (CAI and %Ro)
   in upper ordovician and devonian rocks of the Appalachian Basin: A major revision of USGS map I917E
   using new subsurface collections. U.S. Geological Survey. http://pubs.usgs.gov/sim/3006/SIM3006.pdf

Resnikoff. M: Alexandrova. E: Travers. I. (2010). Radioactivity in Marcellus shale: Report prepared for
   Residents for the Preservation of Lowman and Chemung (RFPLC). New York, NY:  Radioactive Waste
   Management Associates.

Reuters. (2014). UPDATE 2-oil well in North Dakota out of control, leaking. Available online at
   http://www.reuters.com/article/2014/02/14/energy-crude-blowout-idUSL2NOLI15820140214
   (accessed March 2,2015).

Revesz. KM: Breen. Kl: Baldassare. Al: Burruss. RC. (2012). Carbon and hydrogen isotopic evidence for the
   origin of combustible gases in water-supply wells in north-central Pennsylvania. Appl Geochem 27:  361-
   375. http://dx.doi.0rg/10.1016/j.apgeochem.2011.12.002

Reynolds. RR: Kiker. RD. (2003). Produced water and associated issues  a manual for the independent
   operator. (Oklahoma Geological Survey Open-File Report 6-2003). Tulsa, OK: Oklahoma Geological Survey.
   http://karl.nrcce.wvu.edu/regional/pww/produced water.pdf

Rice. CA. (1999). Waters co-produced with coal-bed methane from the FerronSandstone in east central Utah:
   chemical and isotopic composition, volumes.and impacts of disposal [Abstract]. Geological Society of
   America Abstracts with Programs 31: A385.

Rice. CA: Ellis. MS: Bullock. IH. (2000). Water co-produced with coalbed methane in the Powder River basin,
   Wyoming: Preliminary compositional data. (Open File-Report 00-372). Denver, CO: U.S. Department of the
   Interior, U.S. Geological Survey.

Rich. AL: Crosby. EC. (2013). Analysis of reserve pit sludge from unconventional natural gas hydraulic
   fracturing and drilling operations for the presence of technologically enhanced naturally occurring
   radioactive material (TENORM). New Solutions: A Journal of Environmental and Occupational Health
   Policy 23:117-135. http://dx.doi.Org/10.2190/NS.23.l.h

Richardson. SD: Plewa. Ml: Wagner. ED: Schoeny. R: Demarini. DM. (2007). Occurrence, genotoxicity, and
   carcinogenicity of regulated and emerging disinfection by-products in drinking water: A review and
   roadmap for research [Review]. Mutat Res 636:178-242. http://dx.doi.Org/10.1016/j.mrrev.2007.09.00

Rickman. R: Mullen. Ml: Petre. IE: Grieser. WV: Kundert. D. (2008). A practical use of shale petrophysics for
   stimulation design optimization: all shale plays are not clones of the Barnett shale. In 2008 SPE annual
   technical conference & exhibition. Richardson, TX: Society of Petroleum Engineers.
   http://dx.doi.org/10.2118/115258-MS

Ricther, BC: Kreitler, CW. (1993). Geochemical techniques for identifying sources of ground-water
   salinization. Boca Raton, FL: CRC Press, http://www.crcpress.com/product/isbn/9781566700009

Rivett. MO: Wealthall. GP: Dearden. RA:  McAlary. TA. (2011). Review of unsaturated-zone transport and
   attenuation of volatile organic compound (VOC) plumes leached from shallow source zones [Review]. J
   Contam Hydrol 123:130-156. http://dx.doi.Org/10.1016/i.iconhyd.2010.12.013

Robertson, 10: Chilingar, GV: Khilyuk, LF: Endres, B. (2012). Migration of gas from oil/gas fields. Energ Source
   Part A 34: 1436-1447. http://dx.doi.org/10.1080/15567030903077899

Rodvelt, GD: Yeager, VI: Hyatt, MA. (2013). Case history: challenges using ultraviolet light to control bacteria
   in Marcellus completions. In 2011 SPE eastern regional meeting. Richardson, TX: Society of Petroleum
   Engineers. http://dx.doi.org/10.2118/149445-MS
               This document is a draft for review purposes only and does not constitute Agency policy.

June 2015                                        32                    DRAFT—DO NOT CITE OR QUOTE

-------
Hydraulic Fracturing Drinking Water Assessment                                                All References
Rolls. Rl: Leigh. C: Sheldon. F. (2012). Mechanistic effects of low-flow hydrology on riverine ecosystems:
   ecological principles and consequences of alteration. Freshwater Science 31:1163-1186.
   http://dx.doi.0rg/10.1899/12-002.l

Ross. D: King. G. (2007). Well completions. In MJ Economides; T Martin (Eds.), Modern fracturing: Enhancing
   natural gas production (1 ed., pp. 169-198). Houston, Texas: ET Publishing.

Rowan. EL: Engle. MA: Kirby. CS: Kraemer. TF. (2011). Radium content of oil- and gas-field produced waters
   in the northern Appalachian Basin (USA): Summary and discussion of data. (Scientific Investigations
   Report 20115135). Reston, VA: U.S. Geological Survey, http://pubs.usgs.gov/sir/2011/5135/

Rowan, EL: Engle, MA: Kraemer, TF: Schroeder, KT: Hammack, RW: Doughten, MW. (2015). Geochemical and
   isotopic evolution of water produced from Middle Devonian Marcellus shale gas wells, Appalachian basin,
   Pennsylvania. AAPG Bulletin 99:181-206. http://dx.doi.org/10.1306/07071413146

Rowe. D: Muehlenbachs. K. (1999). Isotopic fingerprints of shallow gases in the Western Canadian
   sedimentary basin: tools for remediation of leaking heavy oil wells. Organic Geochemistry 30: 861-871.
   http://dx.doi.org/10.1016/S0146-6380(99)00068-6

Roy. SB: Ricci. PF: Summers. KV: Chung. CF: Goldstein. RA. (2005). Evaluation of the sustainability of water
   withdrawals in the United States, 1995 to 2025. J Am Water Resour Assoc 41: 1091-1108.

Roychaudhuri. B: Tsotsis. TT: lessen. K (2011). An experimental and numerical investigation of spontaneous
   imbibition in gas shales. Paper presented at SPE Annual Technical Conference and Exhibition, October 30 -
   November 2, 2011, Denver, Colorado.

Rozell. PI: Reaven. SI. (2012). Water pollution risk associated with natural gas extraction from the Marcellus
   Shale. Risk Anal 32:13821393. http://dx.doi.Org/10.llll/i.1539-6924.2011.01757.x

Rupp. B: Appel. KE: Gundert-Remy. U. (2010). Chronic oral LOAEL prediction by using a commercially
   available computational QSAR tool. Arch Toxicol 84: 681-688. http://dx.doi.org/10.1007/s00204-010-


Rushing. IA: Newsham. KE: Blasingame. TA. (2013). Rock typing: Keys to understanding productivity in tight
   gas sands. SPE Unconventional Reservoirs Conference, February 1012, 2008, Keystone, Colorado, USA.

Rushton. L: Castaneda. C. (2014). Drilling into hydraulic fracturing and the associated wastewater
   management issues. Washington, WD:  Paul Hastings, LLP. http://www.paulhastings.com/docs/default-
   source/PDFs/stay-current-hydraulic-fracturing-wastewater-management.pdf

Rutledge. IT: Phillips. WS. (2003). Hydraulic stimulation of natural fractures as revealed by induced
   microearthquakes, Carthage Cotton Valley gas field, east Texas. Geophysics 68: 441-452.
   http://dx.doi.0rg/10.1190/l.1567214

Rutqvist. I: Rinaldi. AP: Cappa. F: Moridis.  Gl. (2013). Modeling of fault reactivation and induced seismicity
   during hydraulic fracturing of shale-gas reservoirs. Journal of Petroleum Science and Engineering 107: 31-
   44. http://dx.doi.0rg/10.1016/i.petrol.2013.04.023

Rutqvist. I: Rinaldi. AP: Cappa. F: Moridis.  Gl. (2015). Modeling of fault activation and seismicity by injection
   directly into  a fault zone associated with hydraulic fracturing of shale-gas reservoirs. Journal of Petroleum
   Science and Engineering 127: 377-386. http://dx.doi.Org/10.1016/j.petrol.2015.01.019

Saba, T: Mohsen, F: Garry, M: Murphy, B: Hilbert, B. (2012). White paper: Methanol use in hydraulic fracturing
   fluids. (1103844.000 0101 0711 TS26). Maynard, MA:  Exponent.

Sabins, F. (1990). Problems in cementing horizontal wells. J Pet Tech 42: 398-400.
   http://dx.doi.org/10.2118/20005-PA
               This document is a draft for review purposes only and does not constitute Agency policy.

June 2015                                        33                    DRAFT—DO NOT CITE OR QUOTE

-------
Hydraulic Fracturing Drinking Water Assessment                                                All References
SAIC and GES (SAIC Energy, Environment & Infrastructure, LLC and Groundwater & Environmental Services,
   Inc). (2011). ATGAS investigation initial site characterization and response, April 19, 2011 to May 2, 2011,
   ATGAS2H Well Pad, permit no. 37-015-21237, Leroy Township, Bradford County, PA. Harrisburg,
   Pennsylvania: Chesapeake Appalachia, LLC.
   http://www.chk.com/news/articles/documents/atgas initial site characterization report final 0829201
   l.pdf

Saiers. IE: Earth. E. (2012). Comment on 'Potential contaminant pathways from hydraulically fractured shale
   aquifers' [Comment]. Ground Water 50: 826-828; discussion 828-830. http://dx.doi.Org/10.llll/j.1745-
   6584.2012.00990.x

Sang. W: Stoof. CR: Zhang. W: Morales. VL: Gao. B: Kay. RW: Liu. L: Zhang. Y: Steenhuis. TS. (2014). Effect of
   hydrofracking fluid on colloid transport in the unsaturated zone. Environ Sci Technol 48: 8266-8274.
   http://dx.doi.org/10.1021/es501441e

Santa Cruz Biotechnology. (2015). Sorbitane trioleate (CAS 26266-58-0). Available online at
   http://www.scbt.com/datasheet-281154-Sorbitane-Trioleate.html (accessed April 6, 2015).

Scanlon, BR: Reedy, RC: Nicot, IP. (2014). Will water scarcity in semiarid regions limit hydraulic fracturing of
   shale plays? Environmental Research Letters 9. http://dx.doi.Org/10.1088/1748-9326/9/12/124011

Schindler. DW. (1997). Widespread effects of climatic warming on freshwater ecosystems in North America.
   Hydrolog Process 11:1043-1067.

Schlegel. ME: Mclntosh. 1C: Petsch. ST: Orem. WH: lones. EIP: Martini. AM. (2013). Extent and limits of
   biodegradation by in situ methanogenic consortia in shale and formation fluids. Appl Geochem 28:172-
   184. http://dx.doi.0rg/10.1016/i.apgeochem.2012.10.008

Schlumberger (Schlumberger Limited). (2006). Fundamentals of formation testing. Sugar Land, Texas.
   http://www.slb.eom/~/media/Files/evaluation/books/fundamentals formation testing overview.pdf

Schlumberger (Schlumberger Limited). (2014). Schlumberger oilfield glossary. Available online at
   http://www.glossary.oilfield.slb.com/

Schlumberger (Schlumberger Limited). (2015). Stimulation. Available online at
   http://www.slb.com/services/completions/stimulation.aspx

Schmidley. EB: Smith. Bl. (2011). Personal communication from Schmidley and Smith to DiCello: CAWP
   Addendum EM Survey & Well Location; XTO Energy, Inc. Marquardt Release. Available online

Schmidt. CW. (2013). Estimating wastewater impacts from (racking. Environ Health Perspect 121: A117.
   http://dx.doi.org/10.1289/ehp.121-all7

Schmidt. V: McDonald. DA. (1979). The role of secondary porosity in the course of sandstone diagenesis. In PA
   Schole; PR Schluger (Eds.), Aspects of diagenesis :  based on symposia sponsored by the Eastern and by the
   Rocky Mountain Sections, The Society of Economic Paleontologists and Mineralogists (pp. 175-207). Tulsa,
   OK: The Society of Economic Paleontologists and Mineralogists (SEPM).
   http://archives.datapages.com/data/sepm sp/SP26/The Role of Secondary Porosity.html

Schnoor. IL. (1996). Environmental modeling: Fate and transport of pollutants in water, air, and soil. In
   Environmental modeling: Fate and transport of pollutants in water, air, and soil (1 ed.). Hoboken, NJ:
   Wiley-Interscience.

Schubert. I: Rosenmeier. I: Zatezalo. M. (2014). A review of NORM/TENORM in wastes and waters associated
   with Marcellus shale gas development and production. In CL Meehan; JM Vanbriesen; F Vahedifard; X Yu;
   C Quiroga (Eds.), Shale energy engineering 2014: technical challenges, environmental issues, and public
   policy (pp. 492-501). Reston, VA: American Society of Civil Engineers.
   http://dx.doi.org/10.1061/9780784413654.052
               This document is a draft for review purposes only and does not constitute Agency policy.

June 2015                                        34                     DRAFT—DO NOT CITE OR QUOTE

-------
Hydraulic Fracturing Drinking Water Assessment                                                All References
Schwarzenbach. RP: Gschwend. PM: Imboden. DM. (2002). Environmental Organic Chemistry. In
   Environmental organic chemistry (2 ed.). Hoboken, NJ: John Wiley & Sons, Inc.

Science Applications International Corporation. (2010). XTO - Marquardt pad soil and water corrective action
   workplan. (XTO-EPA0001074). Indiana, PA: XTO Energy.

Science Based Solutions LLC. (2014). Summary of hydrogeology investigations in the Mamm Creek field area,
   Garfield County. Laramie, Wyoming. http://www.garfield-county.com/oil-gas/documents/Summary-
   Hydrogeologic-Studies-Mamm%2 OCreek-Area-Feb-10-2014.pdf

Senior. LA. (2014). A reconnaissance spatial and temporal baseline assessment of methane and inorganic
   constituents in groundwater in bedrock aquifers, pike county, Pennsylvania, 201213 (pp. i-106). (2014-
   5117). Senior, LA. http://pubs.usgs.gov/sir/2014/5117/support/sir2014-5117.pdf

Shafer. L. (2011). Water recycling and purification in the Pinedale anticline field: results from the anticline
   disposal project. In 2011 SPE Americas E&P health, safety, security & environmental conference.
   Richardson, TX: Society of Petroleum Engineers. http://dx.doi.org/10.2118/141448-MS

Shaffer, PL: Arias Chavez, LH: Ben-Sasson, M: Romero-Vargas Castrillon, S: Yip, NY: Elimelech, M. (2013).
   Desalination and reuse of high-salinity shale gas produced water: drivers, technologies, and future
   directions. Environ Sci Technol 47: 9569-9583.

Shapiro. SA: Kriiger. OS: Dinske. C: Langenbruch. C. (2011). Magnitudes of induced earthquakes and
   geometric scales of fluid-stimulated rock volumes. Geophysics 76: WC55-WC63.
   http://dx.doi.0rg/10.1190/geo2010-0349.l

Sharma. S: Bowman. L: Schroeder. K: Hammack. R. (2014a). Assessing changes in gas migration pathways at a
   hydraulic fracturing site: Example from Greene County, Pennsylvania, USA. Appl Geochem.
   http://dx.doi.0rg/10.1016/j.apgeochem.2014.07.018

Sharma. S: Mulder. ML: Sack. A: Schroeder. K: Hammack. R. (2014b). Isotope approach to assess hydrologic
   connections during Marcellus Shale drilling. Ground Water 52: 424433.
   http://dx.doi.org/10.llll/gwat.12083

Sheng. Z. (2005). An aquifer storage and recovery system with reclaimed wastewater to preserve native
   groundwater resources in El Paso, Texas. J Environ Manage 75: 367-377.
   http://dx.doi.0rg/10.1016/j.jenvman.2004.10.007

Shires, T: Lev-On, M. (2012). Characterizing pivotal sources of methane emissions from unconventional
   natural gas production - summary and analysis of API and ANGA survey responses. Washington, DC:
   American Petroleum Institute. American Natural Gas Alliance.

Siegel. PI: Azzolina. NA: Smith. Bl: Perry. AE: Bothun. RL. (In Press) Methane concentrations in water wells
   unrelated to proximity to existing oil and gas wells in northeastern Pennsylvania. Environ Sci Technol.
   http://dx.doi.org/10.1021/es505775c

Silva. IM: Matis. H: Kostedt. WL: Watkins. V. (2012). Produced water pretreatment for water recovery and salt
   production. (08122-36). Niskayuna,  NY: Research Partnership to Secure Energy for America.
   http://www.rpsea.org/media/files/project/18621900/08122-36-FR-
   Pretreatment Water Mgt Frac Water Reuse Salt-01-26-12.pdf

Simunek. I: Seina. M: van Genuchten. MT. (1998). The HYDRUS-1D software package for simulating the one-
   dimensional movement of water, heat, and multiple solutes in variably-saturated media, Version 2.0,
   IGWMC-TPS-70. Available online

Sionix (Sionix Corporation). (2011). Sionix to build Bakken water treatment plant. Retrieved from
   http://www.rigzone.com/news/article pf.asp?a id=110613

Sirivedhin. T:  Dallbauman. L. (2004). Organic matrix in produced water from the Osage-Skiatook petroleum
   environmental research site, Osage county, Oklahoma. Chemosphere 57:  463-469.
               This document is a draft for review purposes only and does not constitute Agency policy.

June 2015                                        35                     DRAFT—DO NOT CITE OR QUOTE

-------
Hydraulic Fracturing Drinking Water Assessment                                                All References
Siolander. SA: Clark. I: Rizzo. D: Turack. I. (2011). Water facts #31: Introduction to hydrofracturing. University
   Park, PA: Penn State College of Agricultural Sciences - Cooperative Extension. http://www.shale-gas-
   information-platform.org/fileadmin/ship/dokumente/introduction to hydrofracturing-2.pdf

Skalak. Kl: Engle. MA: Rowan. EL: lolly. GD: Conko. KM: Benthem. Al: Kraemer. TF. (2014). Surface disposal of
   produced waters in western and southwestern Pennsylvania: Potential for accumulation of alkali-earth
   elements in sediments. Int J Coal Geol 126:162-170. http://dx.doi.Org/10.1016/i.coal.2013.12.001

Skjerven, T: Lunde, 0: Perander, M: Williams, B: Farquhar, R: Sinet, I: Sasby, I: Haga, HB: Finnseth, 0: lohnsen,
   S. (2011). Norwegian Oil and Gas Association recommended guidelines for well integrity. (117, Revision
   4). Norway: Norwegian Oil and Gas Association.
   http://www.norskoljeoggass.no/Global/Retningslinjer/Boring/117%20-
   %20Recommended%20guidelines%20Well%20integrity%20rev4%2006.06.%2011.pdf

Skoumal. Rl: Brudzinski. MR: Currie. BS. (2015). Earthquakes induced by hydraulic fracturing in Poland
   Township, Ohio. Seismological Society of America Bulletin 105:  189-197.
   http://dx.doi.org/10.1785/0120140168

Slutz, I: Anderson, I: Broderick, R: Horner, P. (2012). Key shale gas water management strategies: An
   economic assessment tool. Paper presented at International Conference on Health, Safety and
   Environment in Oil and Gas Exploration and Production, September 11-13,2012, Perth, Australia.

Smolen. II. (2006). Cased hole and production log evaluation. Tulsa, OK: PennWell Books.

Soeder. PI: Kappel. WM. (2009). Water resources and natural gas production from the Marcellus shale [Fact
   Sheet] (pp. 6). (U.S. Geological Survey, Fact Sheet 20093032). Soeder, DJ; Kappel, WM.
   http://pubs.usgs.gov/fs/2009/3032/pdf/FS2009-3032.pdf

Solley, WB: Pierce, RR: Perlman, HA. (1998). Estimated use of water in the United States in 1995. (USGS
   Circular:  1200). U.S. Geological Survey, http://pubs.er.usgs.gov/publication/cirl200

Spellman. FR. (2012). Environmental impacts of hydraulic fracturing. In Environmental impacts of hydraulic
   fracturing. Boca Raton, Florida: CRC Press.

SRBC (Susquehanna River Basin Commission). (2013). Comprehensive plan for the water resources of the
   Susquehanna River basin. Harrisburg, PA. http://www.srbc.net/planning/comprehensiveplan.htm

States. S: Cyprych. G: Stoner. M: Wydra. F: Kuchta. I: Monnell. I: Casson. L. (2013). Marcellus Shale drilling and
   brominated THMs in Pittsburgh, Pa., drinking water. J Am Water Works Assoc 105: E432-E448.
   http://dx.doi.org/10.5942/jawwa.2013.105.0093

Stewart. DR. (2013a). Analytical testing for hydraulic fracturing fluid water recovery and reuse. In Summary
   of the technical workshop on analytical chemical methods (pp. B6-B10). Stewart, DR.
   http://www2.epa.gov/sites/production/files/documents/analytical-chemical-methods-technical-
   workshop-summary.pdf

Stewart. DR. (2013b). Treatment for beneficial use of produced water and hydraulic fracturing flowback
   water. Presentation presented at US EPA Technical Workshop on Wastewater Treatment and Related
   Modeling For Hydraulic Fracturing, April 18, 2013, Research Triangle Park, NC.

Stinger Wellhead Protection. I. nc. (2010). Stinger Wellhead Protection. Houston, TX: Stinger Wellhead
   Protection, Inc. http://etdevelopers.com/design-preview/STS/product-
   catalog/STS  Product Catalog  2010-SWP.pdf

STO (Statoil). (2013). Shale facts: drilling and hydraulic fracturing, how it's done, responsibly.  (Global
   Version, April 2013). Stavanger, Norway.
   http://www.statoil.com/no/OurOperations/ExplorationProd/ShaleGas/FactSheets/Downloads/Shale Dr
   illingHvdraulicFacturing.pdf
               This document is a draft for review purposes only and does not constitute Agency policy.

June 2015                                        36                    DRAFT—DO NOT CITE OR QUOTE

-------
Hydraulic Fracturing Drinking Water Assessment                                                All References
Stringfellow. WT: Domen. IK: Camarillo. MK: Sandelin. WL: Borglin. S. (2014). Physical, chemical, and
   biological characteristics of compounds used in hydraulic fracturing. J Hazard Mater 275: 37-54.
   http://dx.doi.0rg/10.1016/j.jhazmat.2014.04.040

Strong. L: Gould. T: Kasinkas. L: Sadowsky. M: Aksan. A: Wackett. L. (2013). Biodegradation in waters from
   hydraulic fracturing: chemistry, microbiology, and engineering. J  Environ Eng 140: B4013001.
   http://dx.doi.org/10.1061/fASCEjEE.1943-7870.0000792

STRONGER (State Review of Oil and Natural Gas Environmental Regulations). (2011a). Louisiana hydraulic
   fracturing state review. Oklahoma City, OK.
   http://www.strongerinc.org/sites/all/themes/stronger02/downloads/Final%20Louisiana%20HF%20Re
   view%203-2011.pdf

STRONGER (State Review of Oil and Natural Gas Environmental Regulations). (2011b). Ohio hydraulic
   fracturing state review. Oklahoma City, OK.
   http://www.strongerinc.org/sites/all/themes/stronger02/downloads/Final%20Report%20of%202011
   %200H%20HF%20Review.pdf

STRONGER (State Review of Oil and Natural Gas Environmental Regulations). (2012). Arkansas hydraulic
   fracturing state review. Oklahoma City, OK. http://www.aogc.state.ar.us/notices/AR HFR FINAL.pdf

Sturchio. NC: Banner. IL: Binz. CM: Heraty. LB: Musgrove. M. (2001).  Radium geochemistry of ground waters
   in Paleozoic carbonate aquifers, midcontinent, USA. Appl Geochem 16:109-122.

Sumi. L. (2004). Pit pollution: Backgrounder on the issues, with a New Mexico case study. Washington, DC:
   Earthworks: Oil and Gas Accountability Project.
   http://www.earthworksaction.org/files/publications/PitReport.pdf

Sun, M: Lowry, GV: Gregory, KB. (2013). Selective oxidation of bromide in wastewater brines from hydraulic
   fracturing. Water Res 47: 3723-3731. http://dx.doi.Org/10.1016/j.watres.2013.04.041

Swann. C: Matthews. I:  Ericksen. R: Kuszaul. I. (2004). Evaluations  of radionuclides of uranium, thorium, and
   radium with produced fluids, precipitates, and sludges from oil, gas, and oilfield brine injections wells.
   (DE-FG26-02NT15227). Washington, D.C.: U.S. Department of Energy.
   http://www.olemiss.edu/depts/mmri/programs/norm final.pdf

Swanson. VE. (1955). Uranium in marine black shales of the United States. In Contributions to the geology of
   uranium and thorium by the United States Geological Survey and Atomic Energy Commission for the
   United Nations International Conference on Peaceful Uses of Atomic Energy, Geneva, Switzerland, 1955
   (pp. 451-456). Reston, VA: U.S. Geological Survey, http://pubs.usgs.gov/pp/0300/report.pdf

SWN (Southwestern Energy). (2011). Frac fluid  whats in it? Houston, TX.
   http://www.swn.com/operations/documents/frac fluid fact sheetpdf

SWN (Southwestern Energy). (2014). Field Site Visit at Southwestern Energy. Available online

Syed. T: Cutler. T. (2010). Well integrity technical and regulatory considerations for C02 injection wells. In
   2010 SPE international conference on health, safety & environment in oil and gas exploration and
   production. Richardson, TX: Society of Petroleum Engineers. http://dx.doi.org/10.2118/125839-MS

Taylor. A. (2012). Watering the boom in Oklahoma: supplies, demands, and neighbors. Presentation
   presented at 2012 Kansas Water Issues  Forums, February 29-March 1,2012, Wichita and Hays, Kansas.

Taylor. RS: Lestz. RS: Loree. D: Funkhouser. GP:  Fyten. G: Attaway. D: Watkins. H. (2006). Optimized C02
   miscible hydrocarbon fracturing fluids. Calgary, Alberta, Canada:  Petroleum  Society of Canada.
   http://dx.doi.org/10.2118/2006-168

The Royal Society and the Royal Academy of Engineering. (2012). Shale gas extraction in the UK: A review of
   hydraulic fracturing. London. http://www.raeng.org.uk/news/publications/list/reports/Shale Gas.pdf
               This document is a draft for review purposes only and does not constitute Agency policy.

June 2015                                        37                     DRAFT—DO NOT CITE OR QUOTE

-------
Hydraulic Fracturing Drinking Water Assessment                                                All References
Thompson. AM. (2010) Induced fracture detection in the Barnett Shale, Ft. Worth Basin, Texas. (Master's
   Thesis). University of Oklahoma, Norman, OK.

Thordsen. II:. Kharaka. YK: Ambats. G: Kakouros. E: Abbott. MM. (2007). Geochemical data from produced
   water contamination investigations: Osage-Skiatook Petroleum Environmental Research (OSPER) sites,
   Osage County, Oklahoma. (Open-File Report 2007-1055). Reston, VA: United States Geological Survey.

Tidwell. VC: Kobos. PH: Malczynski. L. enA: Klise. G: Castillo. CR. (2012). Exploring the water-thermoelectric
   power nexus. J Water Resour Plann Manag 138: 491-501. http://dx.doi.org/10.1061/rASCE) WR.1943-
   5452.0000222

Tidwell, VC: Zemlick, K: Klise, G. (2013). Nationwide water availability data for energy-water modeling.
   Albuquerque, New Mexico: Sandia National Laboratories, http://prod.sandia.gov/techlib/access-
   control.cgi/2013/139968.pdf

Tiemann. M: Folger. P: Carter. NT. (2014). Shale energy technology assessment: Current and emerging water
   practices. Washington, DC: Congressional Research Service, http://nationalaglawcenter.org/wp-
   content/uploads//assets/crs/R43635.pdf

Tilley. Bl: Muehlenbachs. K (2012). Fingerprinting of gas contaminating groundwater and soil in a
   petroliferous region, Alberta, Canada. In RD Morrison; G O'Sullivan (Eds.), Environmental forensics:
   Proceedings of the 211INEF Conference (pp. 115-125). London: RSC Publishing.
   http://dx.doi.org/10.1039/9781849734967-00115

TIPRO (Texas Independent Producers and Royalty Owners Association). (2012). Bradenhead pressure
   management. Austin, TX. http://www.tipro.org/UserFiles/BHP Guidance Final 071812.pdf

Titler. RV: Curry. P. (2011). Chemical analysis of major constituents and trace contaminants of rock salt.
   Harrisburg, PA: Pennsylvania Department of Environmental Protection.
   http://files.dep.state.pa. us/water/Wastewater%20Management/WastewaterPortalFiles/Rock%20Salt%
   20Paper%20final%20052711.pdf

Tudor. EH: Nevison. GW: Allen. S: Pike. B. (2009). Case study of a novel hydraulic fracturing method that
   maximizes effective hydraulic fracture length. In 2009 SPE annual technical conference & exhibition.
   Richardson, TX: Society of Petroleum Engineers. http://dx.doi.org/10.2118/124480-MS

TWDB (Texas Water Development Board). (2012). Water for Texas 2012 state water plan. Austin, TX.
   http://www.twdb.state.tx.us/waterplanning/swp/2012/index.asp

Tyrrell. P. (2012). Water needs for oil & gas well drilling and fracturing. Presentation presented at 85th
   Annual AWSE Fall Conference, Septermber 23-26,2012, Omaha, Nebraska.

Tyrrell. P. (2013). Wyoming update: water rights for hydraulic fracturing. Presentation presented at Summer
   172nd Western States Water Council Meeting, June 24-26,2013, Casper, Wyoming.

U.S. Army Corps of Engineers. (2011). Final Garrison Dam/Lake Sakakawea project, North Dakota surplus
   water report. Volume 1. Omaha, NE: The U.S. Army Corps of Engineers, Omaha District
   http://www.swc.nd.gov/4dlink9/4dcgi/GetSubContentPDF/PB-
   2811/Garrison%20Dam%20Lake%20Sakakawea%20Surplus%20Water%20Report.pdf

U.S. Census Bureau. (2010). Special release - census blocks with population and housing unit counts, 2010
   TIGER/Line shapefiles  [Computer Program]. Suitland, MD: U.S. Census Bureau, Geography Division.
   Retrieved from https://www.census.gov/geo/maps-data/data/tiger-line.html

U.S. Census Bureau. (2013a). Annual estimates of the resident population: April 1, 2010 to July 1, 2013.
   Suitland, MD: U.S. Census Bureau, Population Division.
   http://factfinder2.census.gov/faces/tableservices/isf/pages/productview.xhtml?src=bkmk
               This document is a draft for review purposes only and does not constitute Agency policy.

June 2015                                       38                     DRAFT—DO NOT CITE OR QUOTE

-------
Hydraulic Fracturing Drinking Water Assessment                                                All References
U.S. Census Bureau. (2013b). Cartographic boundary shapefiles metropolitan and micropolitan statistical
   areas and related statistical areas (Combined statistical areas, 500k). Suitland, MD. Retrieved from
   https://www.census.gov/geo/maps-data/data/cbf/cbf msa.html

U.S. Census Bureau. (2013c). Metropolitan and micropolitan statistical areas main. Available online at
   http://www.census.gov/population/metro/ (accessed January 12, 2015).

U.S. Census Bureau. (2014). American FactFinder. Available online at
   http://factfinder.census.gov/faces/nav/jsf/pages/index.xhtml

U.S. Department of lustice. (2014). Company owner sentenced to more than two years in prison for dumping
   fracking waste in Mahoning River tributary. Available online at
   http://www.justice.gov/usao/ohn/news/2014/05auglupo.html (accessed March 4, 2015).

U.S. EPA (U.S. Environmental Protection Agency). (1991). Manual of individual and non-public water supply
   systems [EPA Report]. (EPA 570/9-91-004). Washington, D.C.

U.S. EPA (U.S. Environmental Protection Agency). (1992). Guidance for data useability in risk assessment
   (part A) - final. (Publication 9285.7-09A). Washington, D.C.
   http://www.epa.gov/oswer/riskassessment/datause/parta.htm

U.S. EPA (U.S. Environmental Protection Agency). (1996). Soil screening guidance: technical background
   document, part 2 [EPA Report] (2nd ed.). (EPA/540/R-95/128). Washington, DC: U.S. Environmental
   Protection Agency, Office of Solid Waste and Emergency Response.
   http://www.epa.gov/superfund/health/conmedia/soil/pdfs/part 2.pdf

U.S. EPA (U.S. Environmental Protection Agency). (1999). Understanding oil spills and oil spill response [EPA
   Report]. (EPA 540-K-99-007). Washington, D.C.: U.S. Environmental Protection Agency, Office of
   Emergency and Remedial Response.
   http://www4.nau.edu/itep/waste/hazsubmap/docs/OilSpill/EPAUnderstandingOilSpillsAndOilSpillResp
   onsel999.pdf

U.S. EPA (U.S. Environmental Protection Agency). (2002a). A review of the reference dose and reference
   concentration processes. (EPA/630/P-02/002F). Washington,  DC: U.S. Environmental Protection Agency,
   Risk Assessment Forum. http://cfpub.epa.gov/ncea/cfm/recordisplay.cfm?deid=51717

U.S. EPA (U.S. Environmental Protection Agency). (2002b). Toxicological review of benzene (noncancerous
   effects) [EPA Report]. (EPA/635/R-02/001F). Washington, DC.

U.S. EPA (U.S. Environmental Protection Agency). (2003). A summary of general assessment factors for
   evaluating the quality of scientific and technical information [EPA Report]. (EPA/100/B-03/001).
   Washington, DC: U.S. Environmental Protection Agency, Office of Research and Development.
   http://www.epa.gov/spc/assess.htm

U.S. EPA (U.S. Environmental Protection Agency). (2004). Evaluation of impacts to underground sources of
   drinking water by hydraulic fracturing of coalbed methane reservoirs. (EPA/816/R-04/003). Washington,
   DC.: U.S. Environmental Protection Agency, Office of Solid Waste.

U.S. EPA (U.S. Environmental Protection Agency). (2005). Pollution prevention (P2) framework [EPA Report].
   (EPA-748-B-04-001). Washington, DC: Office of Pollution Prevention and Toxics.
   http://www.epa.gov/oppt/sf/pubs/p2frame-june05a2.pdf

U.S. EPA (U.S. Environmental Protection Agency). (2006). National Primary Drinking Water Regulations:
   Stage 2 Disinfectants and Disinfection Byproducts Rule.
   http://water.epa.gov/lawsregs/rulesregs/sdwa/stage2/

U.S. EPA (U.S. Environmental Protection Agency). (2007). Monitored natural attenuation of inorganic
   contaminants in ground water: volume Itechnical basis for assessment [EPA Report]. (EPA/600/R-
   07/139). Washington, D.C. http://nepis.epa.gov/Adobe/PDF/60000N4K.pdf
               This document is a draft for review purposes only and does not constitute Agency policy.

June 2015                                        39                     DRAFT—DO NOT CITE OR QUOTE

-------
Hydraulic Fracturing Drinking Water Assessment                                                All References
U.S. EPA (U.S. Environmental Protection Agency). (2010). Toxicological review of acrylamide (CAS No. 79-06-
   1) in support of summary information on the Integrated Risk Information System (IRIS) [EPA Report].
   (EPA/635/R-07/008F). Washington, DC.

U.S. EPA (U.S. Environmental Protection Agency). (2011a). Design for the Environment program alternatives
   assessment criteria for hazard evaluation (version 2.0). Washington, D.C.
   http://www2.epa.gov/saferchoice/alternatives-assessment-criteria-hazard-evaluation

U.S. EPA (U.S. Environmental Protection Agency). (2011b). Ground water cleanup at Superfund Sites [EPA
   Report]. (EPA 540-K-96 008). Washington, DC: U. S. Environmental Protection Agency, Office Water.
   http://www.epa.gov/superfund/health/conmedia/gwdocs/brochure.htm

U.S. EPA (U.S. Environmental Protection Agency). (2011c). Plan to study the potential impacts of hydraulic
   fracturing on drinking water resources [EPA Report]. (EPA/600/R-11/122). Washington, DC: U.S.
   Environmental Protection Agency, Office of Research and Development
   http://www2.epa.gov/hfstudy/plan-study-potential-impacts-hydraulic-fracturing-drinking-water-
   resources-epa600r-11122

U.S. EPA (U.S. Environmental Protection Agency). (2011d). Terminology services (TS): Vocabulary catalog -
   IRIS glossary. Available online at
   http://ofmpub.epa.gov/sor internet/registry/termreg/searchandretrieve/glossariesandkeywordlists/se
   arch.do?details=&glossaryName=IRIS%20Glossary (accessed May 21, 2015).

U.S. EPA (U.S. Environmental Protection Agency). (2012a). 5.2 Dissolved oxygen and biochemical oxygen
   demand. In Water Monitoring and Assessment, http://water.epa.gov/type/rsi/monitoring/vms52.cfm

U.S. EPA (U.S. Environmental Protection Agency). (2012b). Estimation Programs Interface Suite for Microsoft
   Windows (EPI Suite) [Computer Program]. Washington DC: US Environmental Protection Agency.
   Retrieved from http://www.epa.gov/oppt/exposure/pubs/episuitedl.htm

U.S. EPA (U.S. Environmental Protection Agency). (2012c). Geologic sequestration of carbon dioxide:
   underground injection control (UIC) program class VI well construction guidance [EPA Report]. (EPA 816-
   R-ll-020). Washington, D.C.
   http://water.epa.gov/type/groundwater/uic/class6/upload/epa816rll020.pdf

U.S. EPA (U.S. Environmental Protection Agency). (2012d). Oil and natural gas sector: standards of
   performance for crude oil and natural gas production, transmission, and distribution. Background
   supplemental technical support document for the final new source performance standards. Washington,
   D.C. http://www.epa.gov/airquality/oilandgas/pdfs/20120418tsd.pdf

U.S. EPA (U.S. Environmental Protection Agency). (2012e). Public drinking water systems: facts and figures.
   Washington, DC: U.S. Environmental Protection Agency, Office of Water.
   http://water.epa.gov/infrastructure/drinkingwater/pws/factoids.cfm

U.S. EPA (U.S. Environmental Protection Agency). (2012f). Study of the potential impacts of hydraulic
   fracturing on drinking water resources: Progress report. (EPA/601/R-12/011). Washington, DC: U.S.
   Environmental Protection Agency, Office of Research and Development.
   http://nepis.epa.gov/exe/ZyPURL.cgi?Dockey=P100FH8M.txt

U.S. EPA (U.S. Environmental Protection Agency). (2013a). Data received from oil and gas exploration and
   production companies, including hydraulic fracturing service companies 2011 to 2013. Non-confidential
   business information source documents are located in Federal Docket ID:  EPA-HQ-ORD2010-0674.
   Available at http://www.regulations.gov.

U.S. EPA (U.S. Environmental Protection Agency). (2013b). Drinking water and ground water statistics, fiscal
   year 2011. Washington, DC: U.S. Environmental Protection Agency, Office of Water.
   http://water.epa.gov/scitech/datait/databases/drink/sdwisfed/upload/epa816rl3003.pdf
               This document is a draft for review purposes only and does not constitute Agency policy.

June 2015                                        40                     DRAFT—DO NOT CITE OR QUOTE

-------
Hydraulic Fracturing Drinking Water Assessment                                                All References
U.S. EPA (U.S. Environmental Protection Agency). (2013c). Inventory of U.S. greenhouse gas emissions and
   sinks: 1990-2011. Washington, DC: U.S. Environmental Protection Agency, Office of Atmospheric
   Programs. http://www.epa.gov/climatechange/Downloads/ghgemissions/US-GHG-Inventory-2013-
   Main-Textpdf

U.S. EPA (U.S. Environmental Protection Agency). (2013d). Supplemental programmatic quality assurance
   project plan for work assignment 5-83 technical support for the hydraulic fracturing drinking water
   assessment. Washington, D.C. http://www2.epa.gov/sites/production/files/documents/literature-
   review-qappl.pdf

U.S. EPA (U.S. Environmental Protection Agency). (2013e). SW-846 on-line. Available online at
   http://www.epa.gov/epawaste/hazard/testmethods/sw846/online/index.htm (accessed April 8,2015).

U.S. EPA (U.S. Environmental Protection Agency). (2013f). Toxicological review of 1,4-Dioxane (with
   inhalation update) (CAS No. 123-91-1) in support of summary information on the Integrated Risk
   Information System (IRIS) [EPA Report]. (EPA-635/R-11/003-F). Washington, DC.

U.S. EPA (U.S. Environmental Protection Agency). (2013g). XTO Energy, Inc. Settlement. Available online at
   http://www2.epa.gov/enforcement/xto-energy-inc-settlement

U.S. EPA (U.S. Environmental Protection Agency). (2014a). Alternatives assessment for the flame retardant
   decabromodiphenyl ether (DecaBDE). Washington, D.C. http://www2.epa.gov/saferchoice/partnership-
   evaluate-flame-retardant-alternatives-decabde-publications

U.S. EPA (U.S. Environmental Protection Agency). (2014b). Development of rapid radiochemical method for
   gross alpha and gross beta activity concentration in flowback and produced waters from hydraulic
   fracturing operations [EPA Report]. (EPA/600/R-14/107). Washington, D.C.
   http://www2.epa.gov/hfstudy/development-rapid-radiochemical-method-gross-alpha-and-gross-beta-
   activity-concentration

U.S. EPA (U.S. Environmental Protection Agency). (2014c). Drinking water contaminants. Available online at
   http://water.epa.gov/drink/contaminants/

U.S. EPA (U.S. Environmental Protection Agency). (2014d). Flame retardant alternatives for
   hexabromocyclododecane (HBCD) [EPA Report]. (EPA/740/R-14/001). Washington, D.C.
   http://www2.epa.gov/saferchoice/partnership-evaluate-flame-retardant-alternatives-hbcd-publications

U.S. EPA (U.S. Environmental Protection Agency). (2014e). Greenhouse gas reporting program, Subpart W -
   Petroleum and natural gas systems. Washington, D.C.

U.S. EPA (U.S. Environmental Protection Agency). (2014f). Minimizing and managing potential impacts of
   injection-induced seismicity from class II disposal wells: Practical approaches [EPA Report]. Washington,
   D.C. http://www.epa.gov/r5water/uic/ntwg/pdfs/induced-seismicity-201502.pdf

U.S. EPA (U.S. Environmental Protection Agency). (2014g). Quality assurance project plan - Revision no. 2:
   Data and literature evaluation for the EPA's study of the potential impacts of hydraulic fracturing (HF) on
   drinking water resources [EPA Report]. Washington, D.C.

U.S. EPA (U.S. Environmental Protection Agency). (2014h). Quality management plan- Revision no. 2: Study of
   the potential impacts of hydraulic fracturing for oil and gas on drinking water resources [EPA Report].
   Washington, D.C. http://www2.epa.gov/hfstudy/quality-management-plan-revision-no-2-study-
   potential-impacts-hydraulic-fracturing-oil-and

U.S. EPA (U.S. Environmental Protection Agency). (20141). Retrospective case study in northeastern
   Pennsylvania: study of the potential impacts of hydraulic fracturing on drinking water resources [EPA
   Report]. (EPA 600/R-14/088). Washington, D.C.

U.S. EPA (U.S. Environmental Protection Agency). (2014J). Safe drinking water information system (SDWIS).
   Data obtained from the Office of Water [Database]. Washington, D.C.: Office of Water. Retrieved from
   http://water.epa.gov/scitech/datait/databases/drink/sdwisfed/index.cfm
               This document is a draft for review purposes only and does not constitute Agency policy.

June 2015                                        41                    DRAFT—DO NOT CITE OR QUOTE

-------
Hydraulic Fracturing Drinking Water Assessment                                                All References
U.S. EPA (U.S. Environmental Protection Agency). (2014k). The verification of a method for detecting and
   quantifying diethylene glycol, triethylene glycol, tetraethylene glycol, 2-butoxyethanol and 2-
   methoxyethanol in ground and surface waters [EPA Report]. (EPA/600/R-14/008). Washington, D.C.
   http://www2.epa.gov/hfstudy/verification-method-detecting-and-quantifying-diethylene-glycol-
   triethylene-glycol

U.S. EPA (U.S. Environmental Protection Agency). (2015a). Analysis of hydraulic fracturing fluid data from the
   FracFocus chemical disclosure registry 1.0 [EPA Report]. (EPA/601/R-14/003). Washington, D.C.: Office
   of Research and Development, U.S. Environmental Protection Agency.
   http://www2.epa.gov/hfstudy/analysis-hydraulic-fracturing-fluid-data-fracfocus-chemical-disclosure-
   registry-1-pdf

U.S. EPA (U.S. Environmental Protection Agency). (2015b). Analysis of hydraulic fracturing fluid data from the
   FracFocus chemical disclosure registry 1.0: Project database [EPA Report]. (EPA/601/R-14/003).
   Washington, D.C.: U.S. Environmental Protection Agency, Office of Research and Development.
   http://www2.epa.gov/hfstudy/epa-project-database-developed-fracfocus-l-disclosures

U.S. EPA (U.S. Environmental Protection Agency). (2015c). Case study analysis of the impacts of water
   acquisition for hydraulic fracturing on local water availability [EPA Report]. (EPA/600/R-14/179).
   Washington, D.C.

U.S. EPA (U.S. Environmental Protection Agency). (2015d). DMR spreadsheet Pennsylvania wastewater
   treatment plants per Region 3 Information Request. Data provided by request. Washington, D.C.: Region
   3, U.S. Environmental Protection Agency.

U.S. EPA (U.S. Environmental Protection Agency). (2015e). Effluent data from Pennsylvania wastewater
   treatment plants per Region 3 Information Request. Data provided by request. Washington, D.C.: Region 3,
   U.S. Environmental Protection Agency.

U.S. EPA (U.S. Environmental Protection Agency). (2015f). EPA Enforcement and Compliance History. Online:
   Effluent Charts: SEECO-Judsonia Water Reuse Recycling Facility. Available online at
   http://echo.epa.gov/effluent-chartstfAR0052051

U.S. EPA (U.S. Environmental Protection Agency). (2015g). Inventory of U.S. greenhouse gas emissions and
   sinks: 1990-2013. (EPA 430-R-15-004). Washington, D.C.
   http://www.epa.gov/climatechange/Downloads/ghgemissions/US-GHG-Inventory-2015-Main-Text.pdf

U.S. EPA (U.S. Environmental Protection Agency). (2015h). Key documents about mid-Atlantic oil and gas
   extraction. Available online at http://www.epa.gov/region3/marcellus shale/#aoinfoww (accessed May
   7,2015).

U.S. EPA. National primary drinking water regulations public notification rule and consumer confidence
   report rule health effects language, (parts 141.201, and 141.151), (U.S. Government Publishing
   Office2015i). http://www.ecfr.gov/cgi-bin/text-
   idx?SID=4d25ec04bc44e54blefdf307855f3185&node=pt40.23.141&rgn=div5

U.S. EPA (U.S. Environmental Protection Agency). (2015J). Retrospective case study in Killdeer, North Dakota:
   study of the potential impacts of hydraulic fracturing on drinking water resources [EPA Report]. (EPA
   600/R-14/103). Washington, D.C.

U.S. EPA (U.S. Environmental Protection Agency). (2015k). Retrospective case study in southwestern
   Pennsylvania: study of the potential impacts of hydraulic fracturing on drinking water resources [EPA
   Report]. (EPA 600/R-14/084). Washington, D.C.

U.S. EPA (U.S. Environmental Protection Agency). (20151). Retrospective case study in the Raton Basin,
   Colorado: study of the potential impacts of hydraulic fracturing on drinking water resources [EPA Report].
   (EPA 600/R-14/091). Washington, D.C.
               This document is a draft for review purposes only and does not constitute Agency policy.

June 2015                                        42                     DRAFT—DO NOT CITE OR QUOTE

-------
Hydraulic Fracturing Drinking Water Assessment                                                All References
U.S. EPA (U.S. Environmental Protection Agency). (2015m). Retrospective case study in Wise County, Texas:
   study of the potential impacts of hydraulic fracturing on drinking water resources [EPA Report]. (EPA
   600/R-14/090). Washington, D.C.

U.S. EPA (U.S. Environmental Protection Agency). (2015n). Review of state and industry spill data:
   characterization of hydraulic fracturing-related spills [EPA Report]. (EPA/601/R-14/001). Washington,
   D.C.: Office of Research and Development, U.S. Environmental Protection Agency.

U.S. EPA (U.S. Environmental Protection Agency). (2015o). Review of well operator files for hydraulically
   fractured oil and gas production wells: Well design and construction [EPA Report]. (EPA/601/R-14/002).
   Washington, D.C.: Office of Research and Development, U.S. Environmental Protection Agency.

U.S. EPA (U.S. Environmental Protection Agency). (2015p). Sources contributing bromide and inorganic
   species to drinking water intakes on the Allegheny river in western Pennsylvania [EPA Report].
   (EPA/600/R-14/430). Washington, D.C.

U.S. EPA (U.S. Environmental Protection Agency). (2015q). Technical development document for proposed
   effluent limitation guidelines and standards for oil and gas extraction. (EPA-821-R-15-003). Washington,
   D.C. http://water.epa.gov/scitech/wastetech/guide/oilandgas/unconv.cfm

U.S. EPA (U.S. Environmental Protection Agency). (2015r). WaterSense: water supply in the U.S. Available
   online at http://www.epa.gov/WaterSense/pubs/supply.html (accessed January 12,2015).

U.S. GAP (U.S. Government Accountability Office). (2012). Energy-water nexus: Information on the quantity,
   quality, and management of water produced during oil and gas production. (GAO-12-156). Washington,
   D.C. http://www.gao.gov/products/GAO-12-156

U.S. GAP (U.S. Government Accountability Office). (2014). Freshwater: Supply concerns continue, and
   uncertainties complicate planning. Report to Congressional requesters. (GAO-14-430). Washington, DC:
   U.S. Government Accountability Office (GAO). http://www.gao.gov/assets/670/663343.pdf

U.S. Global Change Research Program. (2009). Global climate change impacts in the United States. New York,
   NY: Cambridge University Press, http://downloads.globalchange.gov/usimpacts/pdfs/climate-impacts-
   reportpdf

Upstream Pumping. (2015). Upstream pumping: Wellhead technology and services. Available online at
   http://upstreampumping.com/

URS Corporation. (2011). Water-related issues associated with gas production in the Marcellus shale:
   Additives use flowback quality and quantities regulations on-site treatment green technologies alternate
   water sources water well-testing. (NYSERDA Contract PO Number 10666).

USGS (U.S. Geological Survey). (2000). Coal-bed methane: Potential and concerns [Fact Sheet]. (Fact Sheet
   123-00). http://pubs.usgs.gov/fs/fsl23-00/fsl23-00.pdf

USGS (U.S. Geological Survey). (2002). Natural gas production in the United States [Fact Sheet]. (USGS Fact
   Sheet FS-113-01). Denver, CO.

USGS (U.S. Geological Survey). (2003). Ground-Water depletion across the nation, http://pubs.usgs.gov/fs/fs-
USGS (U.S. Geological Survey). (2006). Produced Water Database [Database]: U.S. Geological Survey :: USGS.
   Retrieved from
   http://energy.usgs.gov/EnvironmentalAspects/EnvironmentalAspectsofEnergyProductionandUse/Produ
   cedWaters.aspx#3822110-overview

USGS (U.S. Geological Survey). (2007). Water-quality assessment of the high plains aquifer, 19992004.
   (Professional Paper 1749). Reston, VA. http://pubs.usgs.gov/pp/1749/downloads/pdf/P1749front.pdf
               This document is a draft for review purposes only and does not constitute Agency policy.

June 2015                                        43                    DRAFT—DO NOT CITE OR QUOTE

-------
Hydraulic Fracturing Drinking Water Assessment                                                All References
USGS (U.S. Geological Survey). (2009). Water quality in the high plains aquifer, Colorado, Kansas, Nebraska,
   New Mexico, Oklahoma, South Dakota, Texas, and Wyoming, 19992004. Reston, VA.
   http://pubs.usgs.gov/circ/1337/

USGS (U.S. Geological Survey). (2010). Volatile organic compounds in the nation's ground water and drinking-
   water supply wells: Supporting information. Glossary. Available online at
   http://water.usgs.gov/nawqa/vocs/national assessment/report/glossary.html

USGS (U.S. Geological Survey). (2013a). Map of assessed shale gas in the United States, 2012.
   http://pubs.usgs.gov/dds/dds-069/dds-069-z/

USGS (U.S. Geological Survey). (2013b). National hydrography dataset: High-resolution flowline data: The
   national map. Retrieved from http://viewer.nationalmap.gov/viewer/

USGS (U.S. Geological Survey). (2013c). National Water Information System (NWIS) [Database]. Retrieved
   from http://waterdata.usgs.gov/nwis

USGS (U.S. Geological Survey). (2014a). Energy glossary and acronym list. Available online at
   http://energy.usgs.gov/GeneralInfo/HelpfulResources/EnergyGlossary.aspxtft

USGS (U.S. Geological Survey). (2014b). Estimated use of water in the United States, county-level data for
   2010. Reston, VA. http://water.usgs.gov/watuse/data/2010/

USGS (U.S. Geological Survey). (2014c). The quality of our nations waters water quality in principal aquifers
   of the United States, 19912010. (Circular 1360). Reston, VA. http://dx.doi.org/10.3133/cirl360

USGS (U.S. Geological Survey). (2014d). Trends in major-ion constituents and properties for selected
   sampling sites in the tongue and powder river watersheds, Montana and Wyoming, based on data
   collected during water years 19802010. (Scientific Investigations Report 20135179). Reston, VA.
   http://pubs.usgs.gov/sir/2013/5179/

USGS (U.S. Geological Survey). (2014e). U.S. Geological Survey national produced waters geochemical
   database v2.0 (PROVISIONAL). Available online at
   http://energy.usgs.gov/EnvironmentalAspects/EnvironmentalAspectsofEnergyProductionandUse/Produ
   cedWaters.aspx#3822349-data

USGS (U.S. Geological Survey). (2014f). USGS investigations of water produced during hydrocarbon reservoir
   development [Fact Sheet]. Reston, VA.  http://dx.doi.org/10.3133/fs20143104

USGS (U.S. Geological Survey). (2014g). WaterWatch. Available online at http://waterwatch.usgs.gov/

USGS (U.S. Geological Survey). (2014h). Withdrawal and consumption of water by thermoelectric power
   plants in the United States, 2010. (Scientific Investigations Report 20145184). Reston, VA.
   http://dx.doi.org/10.3133/sir20145184

USGS (U.S. Geological Survey). (2015). Trends in hydraulic  fracturing distributions and treatment fluids,
   additives, proppants, and water volumes applied to wells drilled in the United States from 1947 through
   2010data analysis and comparison to the literature. (U.S. Geological Survey Scientific Investigations
   Report 20145131). Reston, VA. http://dx.doi.org/10.3133/sir20145131

UWS (Universal Well Services). (2008). Environmental response plan for field operations. (PATT-EPA-
   0001060). Meadville, PA: Universal Well Services, Inc.

Vaidyanathan. G. (2013a). Hydraulic fracturing: when 2 wells meet, spills can often follow. Available online at
   http://www.eenews.net/energywire/stories/1059985587 (accessed September 2, 2014).

Vaidyanathan. G. (2013b). XTO comes out swinging against 'unwarranted' criminal charges in Pa. E&E News
   0.

Vaidyanathan. G. (2014). Email communications between Gayathri Vaidyanathan and Ken Klewicki regarding
   the New Mexico Oil Conservation Division District 3 Well Communication Data. Available online
               This document is a draft for review purposes only and does not constitute Agency policy.

June 2015                                        44                    DRAFT—DO NOT CITE OR QUOTE

-------
Hydraulic Fracturing Drinking Water Assessment                                                All References
Valko. PP. (2009). Assigning value to stimulation in the Barnett Shale: A simultaneous analysis of 7000 plus
   production hystories and well completion records. Paper presented at SPE Hydraulic Fracturing
   Technology Conference, January 19-21, 2009, The Woodlands, TX.

van Vliet. MTH: Zwolsman. IIG. (2008). Impact of summer droughts on the water quality of the Meuse river. J
   Hydrol 353: 1-17. http://dx.doi.Org/10.1016/j.jhydrol.2008.01.001

Van Voast. WA. (2003). Geochemical signature of formation waters associated with coalbed methane. AAPG
   Bulletin 87: 667-676.

Veil. I A. (2011). Water management practices used by Fayetteville shale gas producers. (ANL/EVS/R-11/5).
   Washington, DC: U.S. Department of Energy, National Energy Technology Laboratory.
   http://www.ipd.anl.gov/anlpubs/2011/06/70192.pdf

Veil. IA: Puder. MG: Elcock. D: Redweik. Rl. (2004). A white paper describing produced water from production
   of crude oil, natural gas, and coalbed methane. Lemont, IL: Argonne National Laboratory.

Vengosh. A: lackson. RB: Warner. N: Darrah. TH: Kondash. A. (2014). A critical review of the risks to water
   resources from unconventional shale gas development and hydraulic fracturing in the United States.
   Environ Sci Technol 48: 36-52. http://dx.doi.org/10.1021/es405118y

Venkatapathy. R: Moudgal. Cl: Bruce. RM. (2004). Assessment of the oral rat chronic lowest observed adverse
   effect level model in TOPKAT, a QSAR software package for toxicity prediction. J chem inf comput sci 44:
   1623-1629. http://dx.doi.org/10.1021/ci049903s

Verdegem. MCI: Bosnia. RH. (2009). Water withdrawal for brackish and inland aquaculture, and options to
   produce more fish in ponds with present water use. Water Policy 11:  52-68.
   http://dx.doi.org/10.2166/wp.2009.003

Vidas. H:  Hugman. B. (2008). Availability, economics, and production potential of North American
   unconventional natural gas supplies. (F-2008-03). Washington, DC: The INGAA Foundation, Inc.
   http://www.ingaa.org/File.aspx?id=7878

Vidic. RD: Brantley. SL: Vandenbossche. IM: Yoxtheimer. D: Abad. ID. (2013). Impact of shale gas development
   on regional water quality [Review]. Science 340:1235009. http://dx.doi.org/10.1126/science.1235009

Vincent. M. (2011). Restimulation of unconventional reservoirs: when are refracs beneficial? Journal of
   Canadian Petroleum Technology 50: 36-52. http://dx.doi.org/10.2118/136757-PA

Vine. ID: Tourtelot. EB. (1970). Geochemistry of black shale deposits; A summary report. Econ Geol 65: 253-
   272. http://dx.doi.org/10.2113/gsecongeo.65.3.253

Vittitow. IG. Sr. (2010). Well control incident analysis, EOG Resources Inc., Punxautawney Hunting Club 36H,
   Clearfield County, Pennsylvania. Bedrock Engineering.
   http://www.pahouse.com/EnvResources/documents/BEDROCK ENGINEERING PHC 36H  Incident Repo
   rt Final.pdf

Volz. CD:  Ferrar. K: Michanowicz. D: Christen. C: Kearney. S: Kelso. M: Malone. S. (2011). Contaminant
   Characterization of Effluent from Pennsylvania Brine Treatment Inc., Josephine Facility Being Released
   into Blacklick Creek, Indiana County, Pennsylvania: Implications for Disposal of Oil and Gas Flowback
   Fluids from Brine Treatment Plants [Standard].  Volz, CD; Ferrar, K; Michanowicz, D; Christen, C; Kearney,
   S; Kelso, M; Malone, S. http://www2.epa.gov/hfstudy/contaminant-characterization-effluent-
   pennsylvania-brine-treatment-inc-josephine-facility

Vulgamore. TB: Clawson. TD: Pope. CD: Wolhart. SL: Mayerhofer. Ml: Machovoe. SR: Waltman. CK. (2007).
   Applying hydraulic fracture diagnostics to optimize stimulations in the Woodford Shale. Richardson, TX:
   Society of Petroleum Engineers. http://dx.doi.org/10.2118/110029-MS

Waldron. P. (2014). In search of greener fracking for natural gas. Available online at
   http://www.insidescience.org/content/search-greener-fracking-natural-gas/1791
               This document is a draft for review purposes only and does not constitute Agency policy.

June 2015                                        45                    DRAFT—DO NOT CITE OR QUOTE

-------
Hydraulic Fracturing Drinking Water Assessment                                                All References
Walsh. IM. (2013). Water management for hydraulic fracturing in unconventional resourcesPart 1. Oil and
   Gas Facilities 2.

Walter. GR: Benke. RR: Pickett. DA. (2012). Effect of biogas generation on radon emissions from landfills
   receiving radium-bearing waste from shale gas development. J Air Waste Manag Assoc 62:1040-1049.
   http://dx.doi.org/10.1080/10962247.2012.696084

Wang. W: Dahi Taleghani. A. (2014). Cement sheath integrity during hydraulic fracturing; an integrated
   modeling approach. In 2014 SPE hydraulic fracturing technology conference. Richardson, TX: Society of
   Petroleum Engineers. http://dx.doi.org/10.2118/168642-MS

Wang, Z: Krupnick, A. (2013). A retrospective review of shale gas development in the United States. What let
   to the boom? (RFF DP 13-12). Washington, DC: Resources for the Future.
   http://www.rff.org/RFF/documents/RFF-DP-13-12.pdf

Warner. NR: Christie. CA: lackson. RB: Vengosh. A. (2013a). Impacts of shale gas wastewater disposal on
   water quality in western Pennsylvania. Environ Sci Technol 47:11849-11857.
   http://dx.doi.org/10.1021/es402165b

Warner. NR: lackson. RB: Darrah. TH: Osborn. SG: Down. A: Zhao. K: White. A: Vengosh. A. (2012).
   Geochemical evidence for possible natural migration of Marcellus Formation brine to shallow aquifers in
   Pennsylvania. PNAS 109:11961-11966. http://dx.doi.org/10.1073/pnas.1121181109

Warner. NR: Kresse.  TM: Hays. PD: Down. A: Karr. ID: lackson. RB: Vengosh. A. (2013b). Geochemical and
   isotopic variations in shallow groundwater in areas of the Fayetteville Shale development, north-central
   Arkansas. Appl Geochem 35: 207-220.

Warpinski. N. (2009). Microseismic monitoring: Inside and out. J Pet Tech 61: 80-85.
   http://dx.doi.org/10.2118/118537-MS

Water Research Foundation. (2010). Assessment of inorganics accumulation in drinking water system scales
   and sediments. Denver, CO. http://www.waterrf.org/PublicReportLibrary/3118.pdf

Water Research Foundation. (2014). Water quality impacts of extreme weather-related events.
   http://www.waterrf.org/Pages/Projects.aspx?PID=4324

Watson. TL: Bachu. S. (2009). Evaluation of the potential for gas and C02 leakage along wellbores. SPE
   Drilling & Completion 24:115-126. http://dx.doi.org/10.2118/106817-PA

Wattenberg. EV: Bielicki. IM: Suchomel. AE: Sweet. IT: Void. EM: Ramachandran. G. (In Press) Assessment of
   the acute and chronic health hazards of hydraulic fracturing fluids. J Occup Environ Hyg.
   http://dx.doi.org/10.1080/15459624.2015.1029612

Watts.  KR. (2006). A Preliminary Evaluation of Vertical Separation between Production Intervals of Coalbed-
   Methane Wells and Water-Supply Wells in the Raton Basin, Huerfano and Las Animas Counties, Colorado,
   1999-2004.15.

WAWSA (Western Area Water Supply Authority). (2011). Project progress report: western area water supply
   project: appendix N. Williston, ND. http://www.legis.nd.gov/assembly/62-
   2011/docs/pdf/wrlOlOllappendixn.pdf

Weaver. IW: Xu. I: Mravik. SC. (In Press) Scenario analysis of the impact on drinking water intakes from
   bromide in the discharge of treated oil  and gas waste water. J Environ Eng.

Webb.  CH: Nagghappan. L: Smart. G: Hoblitzell. I: Franks. R. (2009). Desalination of oilfield-produced water at
   the San Ardo water reclamation facility, Ca. In SPE Western regional meeting 2009. Richardson, TX:
   Society of Petroleum Engineers. http://dx.doi.org/10.2118/121520-MS

Weijermars. R. (2014). US shale gas production outlook based on well roll-out rate scenarios. Appl Energ 124:
   283-297. http://dx.doi.org/10.1016/i.apenergv.2014.02.058
               This document is a draft for review purposes only and does not constitute Agency policy.

June 2015                                       46                     DRAFT—DO NOT CITE OR QUOTE

-------
Hydraulic Fracturing Drinking Water Assessment                                                All References
Weinhold. B. (2012). The future of tracking: new rules target air emissions for cleaner natural gas production.
   Environ Health Perspect 120: a272-a279. http://dx.doi.org/10.1289/ehp.120-a272

Weinstein. I: Phillippi. M: Walters. HG. (2009). Dry-polymer blending eliminates need for hydrocarbon carrier
   fluids. In 2009 SPE/EPA/DOE E&P Environmental & Safety Conference. Richardson, TX: Society of
   Petroleum Engineers. http://dx.doi.org/10.2118/121002-MS

Wendel. K. (2011). Wastewater technologies critical for continued growth of Marcellus. Available online at
   http://www.ogfi.com/articles/print/volume-8/issue-ll/features/wastewater-technologies-critical-
   for.html (accessed March 9, 2015).

Weng, X: Kresse, 0: Cohen, C: Wu, R: Gu, H. (2011). Modeling of hydraulic fracture network propagation in a
   naturally fractured formation. Paper presented at SPE Hydraulic Fracturing Technology Conference,
   January 24-26,2011, The Woodlands, TX.

Wertz. I. (2014). Fracking site operator faces contempt complaint after acid spill. Available online at
   http://stateimpact.npr.org/oklahoma/2014/08/14/fracking-site-operator-faces-contempt-complaint-
   after-acid-spill/

Wess. I: Ahlers. H: Dobson. S. (1998). Concise International Chemical Assessment Document 10: 2-
   Butoxyethanol. World Health Organization.
   http://www.who.int/ipcs/publications/cicad/cicad 10 revised.pdf

West Virginia PEP (West Virginia Department of Environmental Protection). (2011). Memorandum of
   agreement from the Division of Water and Waste Management to the Division of Highway:
   WVDOH/WVDEP Salt brine from gas wells agreement. Available online at
   http://www.dep.wv.gov/WWE/Documents/WVDOHWVDEP%20Salt%20Brine%20Agreement.pdf

West Virginia PEP (West Virginia Department of Environmental Protection). (2013). West Virginia water
   resources management plan. (Article 22-26). Charleston, WV.
   http://www.dep.wv.gov/WWE/wateruse/WVWaterPlan/Documents/WVWRMP.pdf

West Virginia PEP (West Virginia Department of Environmental Protection). (2014). Personal
   communication: email from Jason Harmon, West Virginia DEP to Megan Fleming, U.S. EPA with attachment
   of WV DEP fracturing water database. Available online

Whitehead. PG: Wade. Al: Butterfield. D. (2009). Potential impacts of climate change on water quality and
   ecology in six UK rivers. 40: 113-122. http://dx.doi.org/10.2166/nh.2009.078

Whittemore. DO. (2007). Fate and identification of oil-brine contamination in different hydrogeologic
   settings. Appl Geochem 22: 2099-2114. http://dx.doi.Org/10.1016/j.apgeochem.2007.04.002

Williams. IE: Taylor. LE: Low. PI. (1998). Hydrogeology and Groundwater Quality of the Glaciated Valleys of
   Bradford, Tioga, and Potter Counties, Pennsylvania. 98.

Wilson, B. (2014). Geologic and baseline groundwater evidence for naturally occurring, shallowly sourced,
   thermogenic gas in northeastern Pennsylvania. AAPG Bulletin 98: 373-394.
   http://dx.doi.org/10.1306/08061312218

Wilson. IM: Van Briesen. IM. (2013). Source water changes and energy extraction activities in the
   Monongahela River, 2009-2012. Environ Sci Technol 47: 1257512582.
   http://dx.doi.org/10.1021/es402437n

Wilson. IM: Vanbriesen. IM. (2012). Oil and gas produced water management and surface drinking water
   sources in Pennsylvania. Environmental Practice 14: 288-300.

Woitanowicz. AK. (2008). Environmental control of well integrity. In ST Orszulik (Ed.), Environmental
   technology in the oil industry (pp. 53-75). Houten, Netherlands: Springer Netherlands.
               This document is a draft for review purposes only and does not constitute Agency policy.

June 2015                                       47                     DRAFT—DO NOT CITE OR QUOTE

-------
Hydraulic Fracturing Drinking Water Assessment                                                All References
Wolfe. D: Graham. G. (2002). Water rights and beneficial use of produced water in Colorado. Denver, CO:
   American Water Resources Association, http://www.gwpc.org/sites/default/files/event-
   sessions/Dick Wolfe PWC02  O.pdf

Wright. PR: McMahon. PB: Mueller. DK: Clark. ML. (2012). Groundwater-quality and quality-control data for
   two monitoring wells near Pavillion, Wyoming, April and May 2012. (USGS Data Series 718). Reston,
   Virginia: U.S. Geological Survey. http://pubs.usgs.gov/ds/718/DS718 508.pdf

Wuchter, C: Banning, E: Mincer, Tl: Drenzek, Ml: Coolen, Ml. (2013). Microbial diversity and methanogenic
   activity of Antrim Shale formation waters from recently fractured wells. FMICB 4:1-14.
   http://dx.doi.org/10.3389/fmicb.2013.00367

WYOGCC (Wyoming Oil and Gas Conservation Commission). (2014). Pavillion Field Well Integrity Review.
   Casper, Wyoming.
   http://wogcc.state.wy.us/pavillionworkinggrp/PAVILLION REPORT 1082014 Final Reportpdf

Wyoming State Engineer's Office. (2014). Groundwater control areas and advisory boards. Available online at
   http://seo.wyo.gov/ground-water/groundwater-control-areas-advisory-boards

Xu. B: Hill. AD: Zhu. D: Wang. L. (2011). Experimental evaluation of guar fracture fluid filter cake behavior.
   Paper presented at SPE Hydraulic Fracturing Technology Conference, January 24-26, 2011, The
   Woodlands, TX.

Xu. P: Drewes. IE: Heil. D. (2008). Beneficial use of co-produced water through membrane treatment:
   Technical-economic assessment. Desalination 225:139-155.
   http://dx.doi.0rg/10.1016/i.desal.2007.04.093

Yang. Y: Robart. Cl: Ruegamer. M. (2013). Analysis of U.S. Hydraulic Fracturing Design Trends. SPE Hydraulic
   Fracturing Technology Conference, The Woodlands, Texas, USA.

Yeager. RR: Bailey. DE. (2013). Diesel-based gel concentrate improves Rocky Mountain region fracture
   treatments. Richardson, TX: Society of Petroleum Engineers. http://dx.doi.org/10.2118/17535-MS

Yergin. D. (2011). The quest: energy, security and the remaking of the modern world. In The quest: energy,
   security and the remaking of the modern world. New York, NY: Penquin Press.

Younos. T: Tulou. KE. (2005). Overview of desalination techniques. Journal of Contemporary Water Research
   & Education 132: 3-10. http://dx.doi.Org/10.llll/i.1936-704X.2005.mpl32001002.x

Zhang. L: Anderson. N: Dilmore. R: Soeder. PI: Bromhal. G. (2014a). Leakage detection of Marcellus Shale
   natural gas at an Upper Devonian gas monitoring well: a 3-d numerical modeling approach. Environ Sci
   Technol 48:10795-10803. http://dx.doi.org/10.1021/es501997p

Zhang. T: Gregory. K: Hammack. RW: Vidic. RD. (2014b). Co-precipitation of radium with barium and
   strontium sulfate and its impact on the fate of radium during treatment of produced water from
   unconventional gas extraction. Environ Sci Technol 48: 4596-4603.
   http://dx.doi.org/10.1021/es405168b

Ziemkiewicz. P: Donovan. I: Hause. I: Gutta. B: Fillhart. I: Mack. B: O'Neal. M. (2013). Water quality literature
   review and field monitoring of active shale gas wells: Phase II for Assessing Environmental Impacts of
   Horizontal Gas Well Drilling Operations. Charleston, WV: West Virginia Department of Environmental
   Protection.

Ziemkiewicz. P: Ouaranta. ID: Mccawley. M. (2014). Practical measures for reducing the risk of environmental
   contamination in shale energy production. Environ Sci Process Impacts 16:1692-1699.
   http://dx.doi.org/10.1039/c3em00510k

Zoback. MD. (2010).  Reservoir geomechanics. Cambridge, UK: Cambridge University Press.
               This document is a draft for review purposes only and does not constitute Agency policy.

June 2015                                       48                     DRAFT—DO NOT CITE OR QUOTE

-------
DRAFT- DO NOT CITE OR QUOTE                              EPA/600/R-15/047b

                                                  External Review Draft

           United States
           Environmental Protection
           Agency
 Assessment of the Potential Impacts of Hydraulic
    Fracturing for Oil and Gas on Drinking Water
                          Resources

                     (Appendices A - J)
                              NOTICE


     THIS DOCUMENT IS AN EXTERNAL REVIEW DRAFT, for review purposes only. It has not

     been formally disseminated by EPA. It does not represent and should not be construed to

     represent any Agency determination or policy.
                     Office of Research and Development
                    U.S. Environmental Protection Agency
                         Washington, DC 20460

-------
Hydraulic Fracturing Drinking Water Assessment                                            Appendices A-J
                                       DISCLAIMER

This document is an external review draft. This information is distributed solely for the purpose of
pre-dissemination peer review under applicable information quality guidelines. It has not been
formally disseminated by EPA. It does not represent and should not be construed to represent any
Agency determination or policy. Mention of trade names or commercial products does not
constitute endorsement or recommendation for use.
              This document is a draft for review purposes only and does not constitute Agency policy.
June 2015                                     i                    DRAFT—DO NOT CITE OR QUOTE

-------
Hydraulic Fracturing Drinking Water Assessment                                               Appendices A-J


Contents: Appendices
Appendix A.    Chemicals Identified in Hydraulic Fracturing Fluids and/or Flowback and

Produced Water
  A.I.   Supplemental Tables and Information	A-l

         Table A-l. Description of sources used to create lists of chemicals used in fracturing fluids or
                    detected in flowback or produced water	A-l
         Table A-2. Chemicals reported to be used in hydraulic fracturing fluids	A-4
         Table A-3. List of generic names of chemicals reportedly used in hydraulic fracturing fluids	A-46
         Table A-4. Chemicals detected in flowback or produced water	A-58
  A.2.   References for Appendix A	A-63

Appendix B.    Water Acquisition Tables	B-l
  B.I.   Supplemental Tables	B-l

         Table B-l. Annual average hydraulic fracturing water use  and consumption in  2011 and 2012
                    compared  to total annual water use and consumption in 2010 by state	B-l
         Table B-2. Annual average hydraulic fracturing water use  and consumption in  2011 and 2012
                    compared  to total annual water use and consumption in 2010 by county	B-3
         Table B-3. Comparison of water use  per well estimates from the EPA's project database of
                    disclosures to FracFocus 1.0 (U.S. EPA, 2015c) and literature sources	B-20
         Table B-4. Comparison of well counts from the EPA's project database of disclosures to FracFocus
                    1.0 (U.S. EPA, 2015c) and state databases for North Dakota, Pennsylvania, and West
                    Virginia	B-21
         Table B-5. Water use per hydraulically fractured well as reported in the EPA's project database of
                    disclosures to FracFocus 1.0 (U.S. EPA, 2015c) by state and basin	B-22
         Table B-6. Estimated percent domestic use water from ground water and self-supplied by county.
                    	B-26
         Table B-7.  Projected hydraulic fracturing water use by Texas counties between 2015 and  2060,
                    expressed  as a percentage of 2010 total county water use	B-40
  B.2.   References for Appendix B	B-52

Appendix C.    Chemical Mixing Supplemental Tables and Information	C-l
  C.I.   Supplemental Tables and Information	C-l

         Table C-l. Chemicals reported to FracFocus in 10% or more of disclosures for gas-producing wells,
                    with the number of disclosures where chemical is reported, percentage of disclosures,
                    and the median maximum concentration (% by mass)  of that chemical in hydraulic
                    fracturing fluid	C-l
         Table C-2. Chemicals reported to FracFocus  in 10% or more of disclosures for oil-producing wells,
                    with the number of disclosures where chemical is reported, percentage of disclosures,
                    and the median maximum concentration (% by mass)  of that chemical in hydraulic
                    fracturing fluid	C-3
         Table C-3a. Top chemicals reported to  FracFocus for each state and number (and percentage) of
                    disclosures where a chemical is reported for that  state, Alabama to Montana (U.S.
                    EPA, 2015c)	C-5
              This document is a draft for review purposes only and does not constitute Agency policy.
June 2015                                           i                    DRAFT—DO NOT CITE OR QUOTE

-------
Hydraulic Fracturing Drinking Water Assessment                                               Appendices A-J

          Table C-3b. Top chemicals reported to FracFocus for each state and number (and percentage) of
                    disclosures where a chemical is reported for that state, New Mexico to Wyoming (U.S.
                    EPA, 2015c)	C-12
          Table C-4. Estimated  mean, median, 5th percentile, and 95th  percentile volumes in gallons for
                    chemicals  reported to  FracFocus  in  100  or more disclosures,  where  density
                    information was available	C-20
          Table C-5. Estimated  mean, median, 5th  percentile, and  95th percentile volumes in liters for
                    chemicals  reported to  FracFocus  in  100  or more disclosures,  where  density
                    information was available	C-23
          Table C-6. Calculated mean, median, 5th percentile, and 95th percentile chemical masses reported
                    to FracFocus in 100 or more disclosures, where density information was available... C-26
          Table C-7. Associated chemical densities and references used to calculate chemical mass and
                    estimate chemical volume	C-29
          Table C-8. Selected  physicochemical properties of chemicals reported as used  in hydraulic
                    fracturing fluids	C-32
   C.2.   References for Appendix C	C-76

Appendix D.    Designing, Constructing, and Testing Wells for Integrity	D-l
   D.I.   Design Goals for Well Construction	D-l
   D.2.   Well Components	D-l
          Text Box D-l. Selected Industry-Developed Specifications and Recommended Practices for Well
                    Construction in North America	D-2
     D.2.1.   Casing	D-2
     D.2.2.   Cement	D-3
          Figure D-l. Atypical staged cementing process	D-8
   D.3.   Well Completions	D-9
          Figure D-2. Examples of well completion types	D-9
   D.4.   Mechanical Integrity Testing	D-10
     D.4.1.   Internal Mechanical Integrity	D-ll
     D.4.2.   External Mechanical Integrity	D-12
   D.5.   References for Appendix D	D-13

Appendix E.    Flowback and Produced Water Supplemental Tables and Information	E-l
   E.I.   Flowback and Long-Term Produced Water Volumes	E-l
          Table E-l. Flowback and long-term produced water characteristics for wells in unconventional
                    formations, formation-level data	E-2
   E.2.   Produced Water Content	E-6
     E.2.1.   Introduction	E-6
     E.2.2.   General Water Quality Parameters	E-6
          Table E-2. Reported concentrations of general water quality parameters in produced water for
                    unconventional    shale   and   tight    formations,   presented    as:   average
                    (minimum-maximum) or median (minimum-maximum)	E-7
              This document is a draft for review purposes only and does not constitute Agency policy.
June 2015                                           ii                    DRAFT—DO NOT CITE OR QUOTE

-------
Hydraulic Fracturing Drinking Water Assessment                                             Appendices A-J

         Table E-3. Reported concentrations of general water quality parameters in produced water for
                    unconventional coalbed basins, presented as: average (minimum-maximum)	E-10
     E.2.3.    Salinity and Inorganics	E-ll
         Table E-4. Reported concentrations (mg/L) of inorganic constituents contributing to salinity in
                    unconventional shale and tight formations  produced water, presented as: average
                    (minimum-maximum) or median (minimum-maximum)	E-12
         Table E-5. Reported concentrations (mg/L) of inorganic constituents contributing to salinity in
                    produced  water  for  unconventional  CBM  basins,  presented  as:  average
                    (minimum-maximum)	E-14
     E.2.4.    Metals and Metalloids	E-14
         Table E-6. Reported concentrations (mg/L) of metals and metalloids from unconventional shale
                    and tight formation produced water, presented as: average (minimum-maximum) or
                    median (minimum-maximum)	E-15
         Table E-7.  Reported concentrations (mg/L) of metals and metalloids from unconventional coalbed
                    produced water, presented as: average (minimum-maximum)	E-18
     E.2.5.    Naturally  Occurring  Radioactive Material  (NORM) and  Technically  Enhanced Naturally
     Occurring Radioactive Material (TENORM)	E-20
         Table E-8.  Reported concentrations (in pCi/L) of radioactive constituents in unconventional shale
                    and  sandstone produced water, presented as: average (minimum-maximum)  or
                    median (minimum-maximum)	E-22
     E.2.6.    Organics	E-24
         Table E-9. Concentrations of select organic parameters from  unconventional  shale, a tight
                    formation, and coalbed produced water, presented as: average (minimum-maximum)
                    or median (minimum-maximum)	E-25
         Table E-10. Reported concentrations (u.g/L)  of organic  constituents in produced water for two
                    unconventional shale  formations, presented as: average (minimum-maximum)  or
                    median (minimum-maximum)	E-28
         Table E-ll. Reported concentrations of organic constituents in 65 samples of produced water
                    from the Black Warrior CBM Basin, presented as average (minimum-maximum)	E-30
     E.2.7.    Chemical Reactions	E-31
     E.2.8.    Microbial Community Processes and Content	E-32
   E.3.    Produced Water Content Spatial Trends	E-34
     E.3.1.    Variability between Plays of the Same Rock Type	E-34
     E.3.2.    Local Variability	E-36
   E.4.    Example Calculation for Roadway Transport	E-36
     E.4.1.    Estimation of Transport Distance	E-36
     E.4.2.    Estimation of Wastewater Volumes	E-37
     E.4.3.    Estimation of Roadway Accidents	E-37
         Table E-12. Combination truck crashes in 2012 for the 2,469,094 registered combination trucks,
                    which traveled 163,458 million miles (U.S. Department of Transportation, 2012).a... E-37
         Table E-13. Large truck crashes in 2012 (U.S. Department of Transportation, 2012).a	E-38
     E.4.4.    Estimation of Material Release Rates in Crashes	E-38
     E.4.5.    Estimation of Volume Released in Accidents	E-38
              This document is a draft for review purposes only and does not constitute Agency policy.
June 2015                                          iii                  DRAFT—DO NOT CITE OR QUOTE

-------
Hydraulic Fracturing Drinking Water Assessment                                               Appendices A-J

          Table E-14. Estimate of total truck-travel miles per well in the Susquehanna River Basin based on
                    the transport analysis performed by Gilmore et al. (2013)	E-39
   E.5.   References for Appendix E	E-39

Appendix F.    Wastewater Treatment and Waste Disposal Supplemental Information	F-l
   F.I.   Estimates of Wastewater Production in Regions where Hydraulic Fracturing is Occurring	F-l
          Table F-l. Estimated volumes (millions of gallons) of wastewater based on state data for selected
                    years and numbers of wells producing fluid	F-2
   F.2.   Overview of Treatment Processes for Treating Hydraulic Fracturing Wastewater	F-6
     F.2.1.   Basic Treatment	F-6
          Figure F-l. Electrocoagulation unit	F-7
     F.2.2.   Advanced Treatment	F-8
          Figure F-2. Photograph of reverse osmosis system	F-9
          Figure F-3. Picture of mobile electrodialysis units in Wyoming	F-10
          Figure F-4. Picture of a mechanical vapor recompression unit near Decatur, Texas	F-ll
          Figure F-5. Mechanical vapor recompression process design - Maggie Spain Facility	F-12
          Figure F-6. Picture of a compressed bed ion exchange unit	F-13
          Figure F-7. Discharge water process used in the Pinedale Anticline field	F-14
   F.3.   Treatment Technology Removal Capabilities	F-14
          Table F-2.  Removal  efficiency of  different hydraulic fracturing wastewater constituents using
                    various wastewater treatment technologies.3	F-15
          Table F-3. Treatment processes for hydraulic fracturing wastewater organic constituents	F-18
          Table F-4. Estimated effluent concentrations  for  example constituents  based  on  treatment
                    process removal efficiencies	F-20
   F.4.   Centralized Waste Treatment Facilities and Waste Management Options	F-23
     F.4.1.   Discharge Options for CWTs	F-23
   F.5.   Water Quality for Reuse	F-24
          Table F-5. Water quality requirements for reuse	F-24
          Figure F-8. Diagram of treatment for reuse of flowbackand produced water	F-26
   F.6.   Hydraulic Fracturing Impacts on POTWs	F-27
     F.6.1.   Potential Impacts on Treatment Processes	F-27
   F.7.   Hydraulic Fracturing and DBPs	F-27
   F.8.   References for Appendix F	F-28

Appendix G.    Identification and Hazard Evaluation of Chemicals across the Hydraulic Fracturing
Water Cycle Supplemental Tables and Information	G-l
   G.I.   Criteria for Selection and Inclusion  of Reference Value (RfV) and Oral Slope Factor (OSF) Data Sources
         G-l
     G.I.I.   Included Sources	G-3
     G.1.2.   Excluded Sources	G-3
   G.2.   Glossary of Toxicity Value Terminology	G-4

              This document is a draft for review purposes only and does not constitute Agency policy.
June 2015                                          iv                   DRAFT—DO NOT CITE OR QUOTE

-------
Hydraulic Fracturing Drinking Water Assessment                                                Appendices A-J
   G.3.   Tables	G-9
          Table G-la. Chemicals reported to be used in hydraulic fracturing fluids, with available federal
                    chronic RfVsand OSFs	G-9
          Table G-lb. Chemicals reported to be used  in hydraulic fracturing fluids, with available state
                    chronic RfVsand OSFs	G-18
          Table G-lc.  Chemicals  reported  to be  used in hydraulic fracturing fluids,  with available
                    international chronic RfVs and OSFs	G-19
          Table G-ld. Chemicals reported to be used in hydraulic fracturing fluids, with available less-than-
                    chronic RfVsand OSFs	G-20
          Table G-2a. Chemicals reported to be detected in flowback or produced water, with available
                    federal chronic RfVsand OSFs	G-23
          Table G-2b. Chemicals reported to be detected in flowback or produced water, with available
                    state chronic RfVsand OSFs	G-31
          Table G-2c. Chemicals reported to be detected in flowback or produced water, with available
                    international chronic RfVs and OSFs	G-33
          Table G-2d. Chemicals reported to be detected in flowback or produced water, with available less-
                    than-chronic RfVsand OSFs	G-34
   G.4.   References for Appendix G	G-36

Appendix H.     Description  of EPA Hydraulic Fracturing Study Publications Cited in This Assessment	H-l
          Table H-l. Titles, descriptions, and citations for EPA hydraulic fracturing study publications cited
                    in this  assessment	H-l

Appendix I.      Unit Conversions	1-1

Appendix J.      Glossary	J-l
   J.I.    Glossary Terms and Definitions	J-l
   J.2.    References for Appendix J	J-17
              This document is a draft for review purposes only and does not constitute Agency policy.
June 2015                                           v                   DRAFT—DO NOT CITE OR QUOTE

-------
Hydraulic Fracturing Drinking Water Assessment                            Appendix A
                     Appendix A
Chemicals Identified in Hydraulic Fracturing
Fluids and/or Flowback and Produced Water
        This document is a draft for review purposes only and does not constitute Agency policy.

June 2015                                   DRAFT—DO NOT CITE OR QUOTE

-------
     Hydraulic Fracturing Drinking Water Assessment                                            Appendix A


     Appendix A. Chemicals  Identified in Hydraulic
         Fracturing Fluids and/or Flowback and Produced

         Water

     A.I.   Supplemental Tables and Information
 1   The EPA identified authoritative sources for information on hydraulic fracturing chemicals and, to
 2   the extent possible, verified the chemicals used in hydraulic fracturing fluids and detected in
 3   flowback and produced water of hydraulically fractured wells. The EPA used 10 sources to identify
 4   the chemicals used in hydraulic fracturing fluids or detected in flowback or produced water. Seven
 5   sources are government entities (Congressional, federal, or state) that obtained the data directly
 6   from industry. The remaining three represent collaborations between state, non-profit, academic,
 7   and industry groups. FracFocus is the result of a collaboration between the Ground Water
 8   Protection Council (a non-profit coalition of state ground water protection agencies) and
 9   the Interstate Oil and Gas Compact Commission (a multi-state government agency). The Marcellus
10   Shale Coalition is a drilling industry trade group. &)]jbojnry^^        is a peer-reviewed journal
11   article. Most of the listed chemicals were cited by multiple sources.

12   Seven of the ten sources obtained information about the  chemicals used in hydraulic fracturing
13   fluids from material safety data sheets (MSDSs) provided by chemical manufacturers for the
14   products they sell, as required by the Occupational Safety and Health Administration (OSHA). The
15   MSDSs must list all hazardous ingredients if they comprise atleast 1% of the product; for
16   carcinogens, the reporting threshold is 0.1%. However, chemical manufacturers may withhold
17   information (e.g., chemical name, concentration of the substance in a mixture) about a hazardous
18   substance from MSDSs if it is claimed as confidential business information (CBI), provided that
19   certain conditions are met (OSHA1_20_13,).
     Table A-l. Description of sources used to create lists of chemicals used in fracturing fluids or
              detected in flowback or produced water.
              The number next to each citation in the reference column corresponds to numbers in the reference
              columns found in Table A-2, Table A-3, and Table A-4.
      Description / Content
Reference
      Chemicals and other components used by 14 hydraulic fracturing service
      companies from 2005 to 2009 as reported to the House Committee on Energy
        , 7,        ,_     ..... 7    .     ,          ,       .         HouseofRepresentatives
      and Commerce. For each hydraulic fracturing product reported, companies       -       -
      also provided an MSDS with information about the product's chemical
      components.
                  This document is a draft for review purposes only and does not constitute Agency policy.

     June 2015                                  A-l                  DRAFT—DO NOT CITE OR QUOTE

-------
     Hydraulic Fracturing Drinking Water Assessment
Appendix A
Description / Content
Chemicals used during natural gas operations with some potential health
effects. The list of chemicals was compiled from MSDSs from several sources,
including the Bureau of Land Management, U.S. Forest Service, state agencies,
and industry.
Chemicals used or proposed for use in hydraulic fracturing in the Marcellus
Shale in New York based on product composition disclosures and MSDSs
submitted to the New York State Department of Environmental Conservation
(NYSDEC). Also includes data provided separately to NYSDEC by well operators
on analytical results of flowback water samples from Marcellus Shale
operations in Pennsylvania and West Virginia.
Chemicals reported to be used by nine hydraulic fracturing service companies
from 2005 to 2010. Companies provided the chemical names in MSDSs,
product bulletins, and formulation sheets.
MSDSs provided to the EPA during on-site visits to hydraulically fractured oil
and gas wells in Oklahoma and Colorado.
Characteristics of undiluted chemicals found in hydraulic fracturing fluids
associated with coalbed methane production, based on MSDSs, literature
searches, reviews of relevant MSDSs provided by service companies, and
discussions with field engineers, service company chemists, and state and
federal employees.
Chemicals used in Pennsylvania for hydraulic fracturing activities based on
MSDSs provided by industry.
Chemical records entered in FracFocus by oil and gas operators for individual
wells from January 1, 2011, through February 28, 2013. FracFocus is a publicly
accessible hydraulic fracturing chemical registry developed by the Ground
Water Protection Council and the Interstate Oil and Gas Compact Commission.
Chemicals claimed as confidential business information (CBI) do not have to be
reported in FracFocus.
Chemicals detected in flowback from 19 hydraulically fractured shale gas wells
in Pennsylvania and West Virginia, based on analyses conducted by 17
Marcellus Shale Coalition member companies.
Chemicals reportedly detected in flowback and produced water from 81 wells
provided to the EPA by nine well operating companies.
Reference
(2)

(2011Ja-b (3)

(4)

Sheets
(6)

(7)

U.S. EPA(2015c)a(8)

Haies (9)

(10)

     a Sources used to identify chemicals used in hydraulic fracturing fluids.
     bSources used to identify chemicals detected in flowback and produced water.
1    Once it had identified chemicals used in hydraulic fracturing fluids and chemicals detected in
2    flowback/preduced water, the EPA conducted an initial review of the chemicals for preliminary
3    validation of provided chemical name and Chemical Abstracts Service Registry Number (CASRN)
4    combinations. A CASRN is a unique numeric identifier assigned by the Chemical Abstracts Service
5    (CAS) to a chemical substance when it enters the CAS Registry Database. The EPA Office of Research
                   This document is a draft for review purposes only and does not constitute Agency policy.
     June 2015                                      A-2                   DRAFT—DO NOT CITE OR QUOTE

-------
      Hydraulic Fracturing Drinking Water Assessment                                            Appendix A


 1    and Development's National Center for Computational Toxicology (NCCT) provided the final formal
 2    validation and verification of the listed chemicals.

 3    The EPA first compared the hydraulic fracturing chemical CASRNs and names with chemicals listed
 4    in NCCT's Distributed Structure-Searchable Toxicity Database network (DSSTox) database (IIS,
 5    EPA,_2fll3b). For the CASRNs and chemical names that did not appear in the DSSTox database, the
 6    EPA's Substance Registry Services database and the U.S. National Library of Medicine ChemID
 7    database were used to verify accurate chemical name and CASRN pairing (NL!MJ_2Iil4j UJLEJPAi
 8    2014c). The EPA also identified cases where CASRN/name combinations could not be verified by
 9    use of selected public sources and flagged those cases for resolution by NCCT.

10    NCCT then verified all of the CASRN and chemical names for the chemical lists generated by the EPA
11    in accordance with NCCT DSSTox Chemical Information Quality Review Procedures
12    (httpi//wwwLepa^
13    on the identification and validation of CASRN/chemical name combinations and resolution of
14    inconsistencies  and problems including duplications, CASRN errors, and CASRN/chemical name
15    mismatches.

16    The general methodology for resolving conflicts between CASRN/chemical name combinations and
17    other chemical identification issues differed slightly depending on the data provided by each
18    source. To resolve chemical/CASRN conflict in data provided by the nine service companies, the
19    EPA worked with each company to verify the CASRN/chemical combinations proposed by NCCT. In
20    cases of CASRN/chemical name mismatches in data provided by FracFocus, chemical names were
21    considered primary to the CASRN (i.e., the name overrode the CASRN). When the chemical name
22    was non-specific and the CASRN was valid, then the CASRN was considered primary to the chemical
23    name, and the correct specific chemical name from DSSTox was assigned to the CASRN. For  all other
24    sources, the CASRN was considered primary unless it was invalid or missing. In such cases, the
25    chemical name was primary. All Toxic Substance Control Act (TSCA) CBI chemical lists were
26    managed in accordance with TSCA CBI procedures.

27    Chemicals with verified CASRNs that are used in hydraulic fracturing fluids are presented in Table
28    A-2. Generic chemicals used in hydraulic fracturing fluids are presented in Table A-3. Chemicals
29    with verified CASRNs that have been detected in flowback or produced water are presented in
30    Table A-4. Chemicals found in both fracturing fluids (see Table A-2) and flowback and  produced
31    water (see Table A-4) are italicized in each table.
                   This document is a draft for review purposes only and does not constitute Agency policy.
     June 2015                                   A-3                  DRAFT—DO NOT CITE OR QUOTE

-------
Hydraulic Fracturing Drinking Water Assessment
Appendix A
Table A-2. Chemicals reported to be used in hydraulic fracturing fluids.

           An "X" indicates the availability of physicochemical properties from EPI Suite™ (see Appendix C) and
           selected toxicity reference values (see Appendix G). An empty cell indicates no information was
           available from the sources we consulted. Reference number corresponds to the citations in Table A-l.
           Italicized chemicals are found in both fracturing fluids and flowback/produced water.
Chemical name
(13Z)-N,N-bis(2-hydroxyethyl)-N-methyldocos-
13-en-l-aminium chloride
(2,3-dihydroxypropyl)trimethylammonium
chloride
(E)-Crotonaldehyde
[Nitrilotris(methylene)]tris-phosphonic acid
pentasodium salt
l-(l-Naphthylmethyl)quinolinium chloride
l-(Alkyl* amino)-3-aminopropane *(42%C12,
26%C18, 15%C14, 8%C16, 5%C10, 4%C8)
l-(Phenylmethyl)pyridinium Et Me derivs.,
chlorides
1,2,3-Trimethylbenzene
1, 2, 4- Trimethylbenzen e
l,2-Benzisothiazolin-3-one
l,2-Dibromo-2,4-dicyanobutane
1,2-Dimethylbenzene
1,2-Ethanediamine, polymer with 2-
methyloxirane
1,2-Ethanediaminium, N,N'-bis[2-[bis(2-
hydroxyethyl)methylammonio]ethyl]-N,N'-
bis(2-hydroxyethyl)-N,N'-dimethyl-,
tetrachloride
1,2-Propylene glycol
1,2-Propylene oxide
1,3,5-Triazine
1,3,5-Triazine- l,3,5(2H,4H,6H)-triethanol
1, 3, 5- Trimethylbenzen e
1,3- Butadiene
CASRN
120086-58-0
34004-36-9
123-73-9
2235-43-0
65322-65-8
68155-37-3
68909-18-2
526-73-8
95-63-6
2634-33-5
35691-65-7
95-47-6
25214-63-5
138879-94-4
57-55-6
75-56-9
290-87-9
4719-04-4
108-67-8
106-99-0
Physico-
chemical
properties
X
X
X
X
X
X
X
X
X
X
X
X

X
X
X
X
X
X
X
Selected
toxicity
reference
value


X











X
X



X
Reference
1
8
1,4
1
1
8
1, 2, 3, 4, 6, 8
1,4
1, 2, 3, 4, 5
1,3,4
1,4
4
8
1,4
1, 2, 3, 4, 8
1,4
8
1,4
1,4
8
               This document is a draft for review purposes only and does not constitute Agency policy.

June 2015                                         A-4                     DRAFT—DO NOT CITE OR QUOTE

-------
Hydraulic Fracturing Drinking Water Assessment
Appendix A
Chemical name
1,3-Dichloropropene
1,4-Dioxane
l,4-Dioxane-2,5-dione, 3,6-dimethyl-, (3R,6R)-,
polymer with (3S,6S)-3,6-dimethyl-l,4-dioxane-
2,5-dione and (3R,6S)-rel-3,6-dimethyl-l,4-
dioxane-2,5-dione
1,6-Hexanediamine
1,6-Hexanediamine dihydrochloride
l-[2-(2-Methoxy-l-methylethoxy)-l-
methylethoxy]-2-propanol
l-Amino-2-propanol
1-Benzylquinolinium chloride
1-Butanol
l-Butoxy-2-propanol
1-Decanol
l-Dodecyl-2-pyrrolidinone
1-Eicosene
l-Ethyl-2-methylbenzene
1-Hexadecene
1-Hexanol
1-Hexanol, 2-ethyl-, manuf. of, by products
from, distn. residues
IH-lmidazole-l-ethanamine, 4,5-dihydro-, 2-
nortall-oil alkyl derivs.
l-Methoxy-2-propanol
1-Octadecanamine, acetate (1:1)
1-Octadecanamine, N,N-dimethyl-
1-Octadecene
1-Octanol
1-Pentanol
1-Propanaminium, 3-amino-N-(carboxymethyl)-
N,N-dimethyl-, N-coco acyl derivs., chlorides,
sodium salts
CASRN
542-75-6
123-91-1
9051-89-2
124-09-4
6055-52-3
20324-33-8
78-96-6
15619-48-4
71-36-3
5131-66-8
112-30-1
2687-96-9
3452-07-1
611-14-3
629-73-2
111-27-3
68909-68-7
68442-97-7
107-98-2
2190-04-7
124-28-7
112-88-9
111-87-5
71-41-0
61789-39-7
Physico-
chemical
properties
X
X

X
X
X
X
X
X
X
X
X
X
X
X
X


X
X
X
X
X
X

Selected
toxic ity
reference
value
X
X






X
















Reference
8
2,3,4
1,4,8
1,2
1
4
8
1,3,4
1, 2, 3, 4, 7
8
1,4
1,4
3
4
3
1,4,8
4
2,4
1, 2, 3, 4
8
1,3,4
3
1,4
8
1
                This document is a draft for review purposes only and does not constitute Agency policy.
June 2015                                          A-5                      DRAFT—DO NOT CITE OR QUOTE

-------
Hydraulic Fracturing Drinking Water Assessment
Appendix A
Chemical name
1-Propanaminium, 3-amino-N-(carboxymethyl)-
N,N-dimethyl-, N-coco acyl derivs., inner salts
1-Propanaminium, 3-chloro-2-hydroxy-N,N,N-
trimethyl-, chloride
1-Propanaminium, N-(3-aminopropyl)-2-
hydroxy-N,N-dimethyl-3-sulfo-, N-coco acyl
derivs., inner salts
1-Propanaminium, N-(carboxymethyl)-N,N-
dimethyl-3-[(l-oxooctyl)amino]-, inner salt
1-Propanesulfonic acid
1-Propanol
1-Propanol, zirconium(4+) salt
1-Propene
l-tert-Butoxy-2-propanol
1-Tetradecene
1-Tridecanol
1-Undecanol
2-(2-Butoxyethoxy)ethanol
2-(2-Ethoxyethoxy)ethanol
2-(2-Ethoxyethoxy)ethyl acetate
2-(Dibutylamino)ethanol
2-(Hydroxymethylamino)ethanol
2-(Thiocyanomethylthio)benzothiazole
2,2'-(diazene-l,2-diyldiethane-l,l-diyl)bis-4,5-
dihydro-lH-imidazole dihydrochloride
2,2'-(Octadecylimino)diethanol
2,2'-[Ethane-l,2-diylbis(oxy)]diethanamine
2,2'-Azobis(2-amidinopropane) dihydrochloride

2,2-Dibromo-3-nitrilopropionamide
2,2-Dibromopropanediamide
2,4-Hexadienoic acid, potassium salt, (2E,4E)-
CASRN
61789-40-0
3327-22-8
68139-30-0
73772-46-0
5284-66-2
71-23-8
23519-77-9
115-07-1
57018-52-7
1120-36-1
112-70-9
112-42-5
112-34-5
111-90-0
112-15-2
102-81-8
34375-28-5
21564-17-0
27776-21-2
10213-78-2
929-59-9
2997-92-4

10222-01-2
73003-80-2
24634-61-5
Physico-
chemical
properties

X


X
X

X
X
X
X
X
X
X
X
X
X
X
X
X
X
X

X
X
X
Selected
toxic ity
reference
value












X
X



X








Reference
1, 2, 3, 4
8
1,3,4
8
3
1, 2, 4, 5
1,4,8
2
8
3
1,4
2
2,4
1,4
1,4
1,4
1,4
2
3
1
1,4
1,4
1, 2, 3, 4, 6,
7,8
3
3
                This document is a draft for review purposes only and does not constitute Agency policy.
June 2015                                          A-6                      DRAFT—DO NOT CITE OR QUOTE

-------
Hydraulic Fracturing Drinking Water Assessment
Appendix A
Chemical name
2,6,8-Trimethyl-4-nonanol
2-Acrylamide - 2-propanesulfonic acid and N,N-
dimethylacrylamide copolymer
2-Acrylamido -2-methylpropanesulfonic acid
copolymer
2-Acrylamido-2-methyl-l-propanesulfonic acid
2-Amino-2-methylpropan-l-ol
2-Aminoethanol ester with boric acid (H3BO3)
(1:1)
2-Aminoethanol hydrochloride
2-Bromo-3-nitrilopropionamide
2-Butanone oxime
2-Butenediamide, (2E)-, N,N'-bis[2-(4,5-dihydro-
2-nortall-oil alkyl-lH-imidazol-l-yl)ethyl] derivs.
2-Butoxy-l-propanol
2-Butoxyethanol
2-Dodecylbenzenesulfonic acid- n-(2-
aminoethyl)ethane-l,2-diamine(l:l)
2-Ethoxyethanol
2-Ethoxynaphthalene
2-Ethyl-l-hexanol
2-Ethyl-2-hexenal
2-Ethylhexyl benzoate
2-Hydroxyethyl acrylate
2-Hydroxyethylammonium hydrogen sulphite
2-Hydroxy-N,N-bis(2-hydroxyethyl)-N-
methylethanaminium chloride
2-Mercaptoethanol
2-Methoxyethanol
2-Methyl-l-propanol
2-Methyl-2,4-pentanediol
CASRN
123-17-1
NOCAS_51252
NOCAS_51255
15214-89-8
124-68-5
10377-81-8
2002-24-6
1113-55-9
96-29-7
68442-77-3
15821-83-7
111-76-2
40139-72-8
110-80-5
93-18-5
104-76-7
645-62-5
5444-75-7
818-61-1
13427-63-9
7006-59-9
60-24-2
109-86-4
78-83-1
107-41-5
Physico-
chemical
properties
X


X
X

X
X
X

X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
Selected
toxic ity
reference
value











X

X








X
X

Reference
8
2
8
1,3
8
8
4,8
1, 2, 3, 4, 5
1
3,8
8
1, 2, 3, 4, 6,
7,8
8
6
3
1, 2, 3, 4, 5
2
4
1,4
1
8
1,4
4
1,2,4
1,2,4
                This document is a draft for review purposes only and does not constitute Agency policy.
June 2015                                          A-7                      DRAFT—DO NOT CITE OR QUOTE

-------
Hydraulic Fracturing Drinking Water Assessment
Appendix A
Chemical name
2-Methyl-3(2H)-isothiazolone
2-Methyl-3-butyn-2-ol
2-Methylbutane
2-Methylquinoline hydrochloride
2-Phosphono-l,2,4-butanetricarboxylic acid
2-Phosphonobutane-l,2,4-tricarboxylic acid,
potassium salt (l:x)
2-Propanol, aluminum salt
2-Propen-l-aminium, N,N-dimethyl-N-2-
propenyl-, chloride, homopolymer
2-Propenamide, homopolymer
2-Propenoic acid, 2-(2-hydroxyethoxy)ethyl
ester
2-Propenoic acid, 2-ethylhexyl ester, polymer
with 2-hydroxyethyl 2-propenoate
2-Propenoic acid, 2-methyl-, polymer with 2-
propenoic acid, sodium salt
2-Propenoic acid, 2-methyl-, polymer with
sodium 2-methyl-2-[(l-oxo-2-propen-l-
yl)amino]-l-propanesulfonate (1:1)
2-Propenoic acid, ethyl ester, polymer with
ethenyl acetate and 2,5-furandione, hydrolyzed
2-Propenoic acid, ethyl ester, polymer with
ethenyl acetate and 2,5-furandione,
hydrolyzed, sodium salt
2-Propenoic acid, polymer with 2-propenamide,
sodium salt
2-Propenoic acid, polymer with ethene, zinc salt
2-Propenoic acid, polymer with ethenylbenzene
2-Propenoic acid, polymer with sodium
ethanesulfonate, peroxydisulfuric acid,
disodium salt- initiated, reaction products with
tetrasodium ethenylidenebis (phosphonata)
2-Propenoic acid, polymer with sodium
phosphinate (1:1), sodium salt
CASRN
2682-20-4
115-19-5
78-78-4
62763-89-7
37971-36-1
93858-78-7
555-31-7
26062-79-3
25038-45-3
13533-05-6
36089-45-9
28205-96-1
136793-29-8
113221-69-5
111560-38-4
25987-30-8
28208-80-2
25085-34-1
397256-50-7
129898-01-7
Physico-
chemical
properties
X
X
X
X
X
X



X










Selected
toxic ity
reference
value




















Reference
1,2,4
3
2
3
1,4
1
1
3
8
4
8
8
8
4,8
8
3,4,8
8
8
8
8
                This document is a draft for review purposes only and does not constitute Agency policy.
June 2015                                          A-8                      DRAFT—DO NOT CITE OR QUOTE

-------
Hydraulic Fracturing Drinking Water Assessment
Appendix A
Chemical name
2-Propenoic acid, sodium salt (1:1), polymer
with sodium 2-methyl-2-((l-oxo-2-propen-l-
yl)amino)-l-propanesulfonate (1:1)
2-Propenoic acid, telomer with sodium 4-
ethenylbenzenesulfonate (1:1), sodium 2-
methyl-2-[(l-oxo-2-propen-l-yl)amino]-l-
propanesulfonate (1:1) and sodium sulfite (1:1),
sodium salt
2-Propenoic, polymer with sodium phosphinate
3-(Dimethylamino)propylamine
3,4,4-Trimethyloxazolidine
3,5,7-Triazatricyclo(3.3.1.13,7)decane, l-(3-
chloro-2-propenyl)-, chloride, (Z)-
3,7-Dimethyl-2,6-octadienal
3-Hydroxybutanal
3-Methoxypropylamine
3-Phenylprop-2-enal
4,4-Dimethyloxazolidine
4,6-Dimethyl-2-heptanone
4-[Abieta-8,ll,13-trien-18-yl(3-oxo-3-
phenylpropyl)amino]butan-2-one hydrochloride
4-Ethyloct-l-yn-3-ol
4-Hydroxy-3-methoxybenzaldehyde
4-Methoxybenzyl formate
4-Methoxyphenol
4-Methyl-2-pentanol
4-Methyl-2-pentanone
4-Nonylphenol
4-Nonylphenol polyethoxylate
5-Chloro-2-methyl-3(2H)-isothiazolone
Acetaldehyde

Acetic acid
CASRN
37350-42-8
151006-66-5
71050-62-9
109-55-7
75673-43-7
51229-78-8
5392-40-5
107-89-1
5332-73-0
104-55-2
51200-87-4
19549-80-5
143106-84-7
5877-42-9
121-33-5
122-91-8
150-76-5
108-11-2
108-10-1
104-40-5
68412-54-4
26172-55-4
75-07-0

64-19-7
Physico-
chemical
properties



X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X

X
X

X
Selected
toxic ity
reference
value

























Reference
1
4
3,4
8
8
3
3
1,2,4
8
1, 2, 3, 4, 7
8
8
1,4
1, 2, 3, 4
3
3
4
1,4
5
8
2,3,4
1,2,4
1,4
1, 2, 3, 4, 5,
6,7,8
                This document is a draft for review purposes only and does not constitute Agency policy.
June 2015                                          A-9                      DRAFT—DO NOT CITE OR QUOTE

-------
Hydraulic Fracturing Drinking Water Assessment
Appendix A
Chemical name
Acetic acid ethenyl ester, polymer with ethenol
Acetic acid, C6-8-branched alkyl esters
Acetic acid, hydroxy-, reaction products with
triethanolamine
Acetic acid, mercapto-, monoammonium salt
Acetic anhydride
Acetone
Acetonitrile, 2,2',2"-nitrilotris-
Acetophenone
Acetyltriethyl citrate
Acrolein
Acrylamide
Acrylamide/ sodium acrylate copolymer
Acrylamide-sodium-2-acrylamido-2-
methlypropane sulfonate copolymer
Acrylic acid
Acrylic acid, with sodium-2-acrylamido-2-
methyl-1-propanesulfonate and sodium
phosphinate
Alcohols (C13-C15), ethoxylated
Alcohols, C10-12, ethoxylated
Alcohols, C10-14, ethoxylated
Alcohols, Cll-14-iso-, CIS-rich
Alcohols, Cll-14-iso-, CIS-rich, butoxylated
ethoxylated
Alcohols, Cll-14-iso-, CIS-rich, ethoxylated
Alcohols, C12-13, ethoxylated
Alcohols, C12-14, ethoxylated
Alcohols, C12-14, ethoxylated propoxylated
Alcohols, C12-14-secondary
Alcohols, C12-14-secondary, ethoxylated
Alcohols, C12-15, ethoxylated
CASRN
25213-24-5
90438-79-2
68442-62-6
5421-46-5
108-24-7
67-64-1
7327-60-8
98-86-2
77-89-4
107-02-8
79-06-1
25085-02-3
38193-60-1
79-10-7
110224-99-2
64425-86-1
67254-71-1
66455-15-0
68526-86-3
228414-35-5
78330-21-9
66455-14-9
68439-50-9
68439-51-0
126950-60-5
84133-50-6
68131-39-5
Physico-
chemical
properties

X
X
X
X
X
X
X
X
X
X


X
X

X

X

X
X

X
X


Selected
toxic ity
reference
value





X

X

X
X


X













Reference
1,4
4
3
2,8
1, 2, 3, 4, 7
1, 3, 4, 6
1,4
1
1,4
2
1, 2, 3, 4
1, 2, 3, 4, 8
1, 2, 3, 4
2,4
8
8
3
3
3
1
3,4,8
4
2, 3, 4, 8
1, 3, 4, 8
1,3,4
3,4,8
3,4
                This document is a draft for review purposes only and does not constitute Agency policy.
June 2015                                         A-10                     DRAFT—DO NOT CITE OR QUOTE

-------
Hydraulic Fracturing Drinking Water Assessment
Appendix A
Chemical name
Alcohols, C12-16, ethoxylated
Alcohols, C14-15, ethoxylated
Alcohols, C6-12, ethoxylated
Alcohols, C7-9-JSO-, C8-rich, ethoxylated
Alcohols, C8-10, ethoxylated propoxylated
Alcohols, C9-11, ethoxylated
Alcohols, C9-ll-iso-, ClO-rich, ethoxylated
Alkanes C10-16-branched and linear
Alkanes, CIO- 14
Alkanes, C12-14-iso-
Alkanes, C13-16-iso-
Alkenes, C>10 .alpha. -
Alkenes, C>8
Alkenes, C24-25 alpha-, polymers with maleic
anhydride, docosyl esters
Alkyl quaternary ammonium with bentonite
Alkyl* dimethyl ethylbenzyl ammonium
chloride *(50%C12, 30%C14, 17%C16, 3%C18)
Alkyl* dimethyl ethylbenzyl ammonium
chloride *(60%C14, 30%C16, 5%C12, 5%C18)
Alkylbenzenesulfonate, linear
Almandite and pyrope garnet
alpha-[3.5-dimethyl-l-(2-methylpropyl)hexyl]-
omega-hydroxy-poly(oxy-l,2-ethandiyl)
alpha-Amylase
alpha-Lactose monohydrate
alpha-Terpineol
Alumina
Aluminatesilicate
Aluminum
Aluminum calcium oxide (AI2CaO4)
CASRN
68551-12-2
68951-67-7
68439-45-2
78330-19-5
68603-25-8
68439-46-3
78330-20-8
90622-52-9
93924-07-3
68551-19-9
68551-20-2
64743-02-8
68411-00-7
68607-07-8
71011-24-0
85409-23-0_l
68956-79-6
42615-29-2
1302-62-1
60828-78-6
9000-90-2
5989-81-1
98-55-5
1344-28-1
1327-36-2
7429-90-5
12042-68-1
Physico-
chemical
properties
X
X
X
X

X
X


X
X
X



X
X
X



X
X




Selected
toxic ity
reference
value

























X

Reference
3,4,8
3,4,8
3,4,8
2,4,8
3
3,4
1, 2, 4, 8
4
1
2,4,8
1,4
1, 3, 4, 8
1
8
4
8
8
1,4,6
1,4
3
4
8
3
1,2,4
8
1,4,6
2
                This document is a draft for review purposes only and does not constitute Agency policy.
June 2015                                         A-ll                      DRAFT—DO NOT CITE OR QUOTE

-------
Hydraulic Fracturing Drinking Water Assessment
Appendix A
Chemical name
Aluminum chloride
Aluminum chloride hydroxide sulfate
Aluminum chloride, basic
Aluminum oxide (AI2O3)
Aluminum oxide silicate
Aluminum silicate
Aluminum sulfate
Amaranth
Amides, C8-18 and C18-unsatd., N,N-
bis(hydroxyethyl)
Amides, coco, N-[3-(dimethylamino)propyl]
Amides, coco, N-[3-(dimethylamino)propyl],
alkylation products with chloroaceticacid,
sodium salts
Amides, coco, N-[3-(dimethylamino)propyl],
alkylation products with sodium 3-chloro-2-
hydroxypropanesulfonate
Amides, coco, N-[3-(dimethylamino)propyl], N-
oxides
Amides, from C16-22 fatty acids and
diethylenetriamine
Amides, tall-oil fatty, N,N-bis(hydroxyethyl)
Amides, tallow, N-[3-(dimethylamino)propyl],N-
oxides
Amine oxides, cocoalkyldimethyl
Amines, C14-18; C16-18-unsaturated, alkyl,
ethoxylated
Amines, C8-18 and C18-unsatd. alkyl
Amines, coco alkyl
Amines, coco alkyl, acetates
Amines, coco alkyl, ethoxylated
Amines, coco alkyldimethyl
Amines, dicoco alkyl
CASRN
7446-70-0
39290-78-3
1327-41-9
90669-62-8
12068-56-3
12141-46-7
10043-01-3
915-67-3
68155-07-7
68140-01-2
70851-07-9
70851-08-0
68155-09-9
68876-82-4
68155-20-4
68647-77-8
61788-90-7
68155-39-5
68037-94-5
61788-46-3
61790-57-6
61791-14-8
61788-93-0
61789-76-2
Physico-
chemical
properties







X
















Selected
toxic ity
reference
value
























Reference
1,4
8
3,4
8
1,2,4
1,2,4
1,4
4
3
1,4
1,4
8
1,3,4
3
3,4
1,4
8
1
5
4
1,4
8
8
8
                This document is a draft for review purposes only and does not constitute Agency policy.
June 2015                                         A-12                      DRAFT—DO NOT CITE OR QUOTE

-------
Hydraulic Fracturing Drinking Water Assessment
Appendix A
Chemical name
Amines, dicoco alkylmethyl
Amines, ditallow alkyl, acetates
Amines, hydrogenated tallow alkyl, acetates
Amines, N-tallow alkyltrimethylenedi-,
ethoxylated
Amines, polyethylenepoly-, ethoxylated,
phosphonomethylated
Amines, polyethylenepoly-, reaction products
with benzyl chloride
Amines, tallow alkyl
Amines, tallow alkyl, ethoxylated, acetates
(salts)
Amines, tallow alkyl, ethoxylated, phosphates
Aminotrimethylene phosphonic acid
Ammonia
Ammonium (lauryloxypolyethoxy)ethyl sulfate
Ammonium acetate
Ammonium acrylate
Ammonium acrylate-acrylamide polymer
Ammonium bisulfate
Ammonium bisulfite
Ammonium chloride
Ammonium citrate (1:1)
Ammonium citrate (2:1)
Ammonium dodecyl sulfate
Ammonium fluoride
Ammonium hydrogen carbonate
Ammonium hydrogen difluoride
Ammonium hydrogen phosphonate
Ammonium hydroxide
Ammonium lactate
CASRN
61788-62-3
71011-03-5
61790-59-8
61790-85-0
68966-36-9
68603-67-8
61790-33-8
68551-33-7
68308-48-5
6419-19-8
7664-41-7
32612-48-9
631-61-8
10604-69-0
26100-47-0
7803-63-6
10192-30-0
12125-02-9
7632-50-0
3012-65-5
2235-54-3
12125-01-8
1066-33-7
1341-49-7
13446-12-3
1336-21-6
515-98-0
Physico-
chemical
properties









X


X
X




X
X
X

X



X
Selected
toxic ity
reference
value



























Reference
8
8
4
8
1,4
1
8
1,3,4
4
1,4,8
1, 2, 3, 4, 7
4
1, 3, 4, 5, 8
8
2,4,8
2
1, 2, 3, 4, 7
1, 2, 3, 4, 5,
6,8
3
8
1
1,4
1,4
1, 3, 4, 7
4
1,3,4
8
                This document is a draft for review purposes only and does not constitute Agency policy.
June 2015                                         A-13                      DRAFT—DO NOT CITE OR QUOTE

-------
Hydraulic Fracturing Drinking Water Assessment
Appendix A
Chemical name
Ammonium ligninsulfonate
Ammonium nitrate
Ammonium phosphate
Ammonium sulfate
Ammonium thiosulfate
Amorphous silica
Anethole
Aniline
Antimony pentoxide
Antimony trichloride
Antimony trioxide
Arsenic
Ashes, residues
Asphalt, sulfonated, sodium salt
Attapulgite
Aziridine, polymer with 2-methyloxirane
Barium sulfate
Bauxite
Benactyzine hydrochloride
Bentonite
Bentonite, benzyl(hydrogenated tallow alkyl)
dimethylammonium stearate complex
Benzamorf
Benzene
Benzene, l,l'-oxybis-, sec-hexyl derivs.,
sulfonated, sodium salts
Benzene, l,l'-oxybis-, tetrapropylene derivs.,
sulfonated
Benzene, l,l'-oxybis-, tetrapropylene derivs.,
sulfonated, sodium salts
Benzene, C10-16-alkyl derivs.
CASRN
8061-53-8
6484-52-2
7722-76-1
7783-20-2
7783-18-8
99439-28-8
104-46-1
62-53-3
1314-60-9
10025-91-9
1309-64-4
7440-38-2
68131-74-8
68201-32-1
12174-11-7
31974-35-3
7727-43-7
1318-16-7
57-37-4
1302-78-9
121888-68-4
12068-08-5
71-43-2
147732-60-3
119345-03-8
119345-04-9
68648-87-3
Physico-
chemical
properties






X
X










X


X
X



X
Selected
toxic ity
reference
value


X




X

X
X
X










X




Reference
2
1,2,3
1,4
1, 2, 3, 4, 6
8
1,7
3
2,4
1,4
1,4
8
4
4
2
2,3
4,8
1,2,4
1,2,4
8
1, 2, 4, 6
3,4
1,4
1,3,4
8
8
3,4,8
1
                This document is a draft for review purposes only and does not constitute Agency policy.
June 2015                                         A-14                     DRAFT—DO NOT CITE OR QUOTE

-------
Hydraulic Fracturing Drinking Water Assessment
Appendix A
Chemical name
Benzene, ethenyl-, polymer with 2-methyl-l,3-
butadiene, hydrogenated
Benzenemethanaminium, N,N-dimethyl-N-(2-
((l-oxo-2-propen-l-yl)oxy)ethyl)-, chloride
(1:1), polymer with 2-propenamide
Benzenesulfonicacid
Benzenesulfonic acid, (1-methylethyl)-,
Benzenesulfonic acid, (1-methylethyl)-,
ammonium salt
Benzenesulfonic acid, (1-methylethyl)-, sodium
salt
Benzenesulfonic acid, C10-16-alkyl derivs.
Benzenesulfonic acid, C10-16-alkyl derivs.,
compds. with cyclohexylamine
Benzenesulfonic acid, C10-16-alkyl derivs.,
compds. with triethanolamine
Benzenesulfonic acid, C10-16-alkyl derivs.,
potassium salts
Benzenesulfonic acid, dodecyl-, branched,
compds. with 2-propanamine
Benzenesulfonic acid, mono-C10-16 alkyl
derivs., compds. with 2-propanamine
Benzenesulfonic acid, mono-C10-16-alkyl
derivs., sodium salts
Benzoicacid
Benzyl chloride
Benzyldimethyldodecylammonium chloride
Benzylhexadecyldimethylammonium chloride
Benzyltrimethylammonium chloride
Bicine
Bio-Perge
Bis(l-methylethyl)naphthalenesulfonicacid,
cyclohexylamine salt
Bis(2-chloroethyl) ether
CASRN
68648-89-5
74153-51-8
98-11-3
37953-05-2
37475-88-0
28348-53-0
68584-22-5
255043-08-4
68584-25-8
68584-27-0
90218-35-2
68648-81-7
68081-81-2
65-85-0
100-44-7
139-07-1
122-18-9
56-93-9
150-25-4
55965-84-9
68425-61-6
111-44-4
Physico-
chemical
properties


X
X
X
X

X
X
X
X

X
X
X
X
X
X
X

X
X
Selected
toxic ity
reference
value






X






X
X






X
Reference
8
3
2
4
3,4
8
1,4
1
8
1,4,8
4
1,4
8
1,4,7
1, 2, 4, 8
2,8
8
8
1,4
8
1
8
                This document is a draft for review purposes only and does not constitute Agency policy.
June 2015                                         A-15                     DRAFT—DO NOT CITE OR QUOTE

-------
Hydraulic Fracturing Drinking Water Assessment
Appendix A
Chemical name
Bisphenol A
Bisphenol A/ Epichlorohydrin resin
Bisphenol A/ Novolac epoxy resin
Blast furnace slag
Borax
Boric acid

Boric acid (H3BO3), compd. with 2-
aminoethanol (l:x)
Boric oxide
Boron potassium oxide (B4K2O7)
Boron potassium oxide (B4K2O7), tetrahydrate
Boron potassium oxide (B5KO8)
Boron sodium oxide
Boron sodium oxide pentahydrate
Bronopol
Butane
Butanedioic acid, sulfo-, l,4-bis(l,3-
dimethylbutyl) ester, sodium salt
Butene
Butyl glycidyl ether
Butyl lactate
Butyryl trihexyl citrate
C.I. Acid Red 1
C.I. Acid violet 12, disodium salt
C.I. Pigment Red 5
C.I. Solvent Red 26
C10-16-Alkyldimethylamines oxides
C10-C16 ethoxylated alcohol
Cll-15-Secondary alcohols ethoxylated
C12-14tert-alkyl ethoxylated amines
C8-10 Alcohols
CASRN
80-05-7
25068-38-6
28906-96-9
65996-69-2
1303-96-4
10043-35-3

26038-87-9
1303-86-2
1332-77-0
12045-78-2
11128-29-3
1330-43-4
12179-04-3
52-51-7
106-97-8
2373-38-8
25167-67-3
2426-08-6
138-22-7
82469-79-2
3734-67-6
6625-46-3
6410-41-9
4477-79-6
70592-80-2
68002-97-1
68131-40-8
73138-27-9
85566-12-7
Physico-
chemical
properties
X













X
X
X
X
X
X
X
X
X
X
X
X
X

X

Selected
toxic ity
reference
value
X





























Reference
4
1,2,4
1,4
2,3
1, 2, 3, 4, 6
1, 2, 3, 4, 6, 7

8
1, 2, 3, 4
8
8
1
1,2,4
8
1, 2, 3, 4, 6
2,5
1
8
1,4
1,4
8
4
4
4
4
4
1, 2, 3, 4, 8
1,2,8
3
8
                This document is a draft for review purposes only and does not constitute Agency policy.
June 2015                                         A-16                     DRAFT—DO NOT CITE OR QUOTE

-------
Hydraulic Fracturing Drinking Water Assessment
Appendix A
Chemical name
Calcined bauxite
Calcium aluminate
Calcium bromide
Calcium carbide (CaC2)
Calcium chloride
Calcium dichloride dihydrate
Calcium dodecylbenzene sulfonate
Calcium fluoride
Calcium hydroxide
Calcium hypochlorite
Calcium magnesium hydroxide oxide
Calcium oxide
Calcium peroxide
Calcium sulfate
Calcium sulfate dihydrate
Camphor
Canola oil
Carbon black
Carbon dioxide
Carbonic acid calcium salt (1:1)
Carbonic acid, dipotassium salt
Carboxymethyl guar gum, sodium salt
Castor oil
Cedarwood oil
Cellophane
Cellulose
Chloride
Chlorine
Chlorine dioxide
Choline bicarbonate
CASRN
66402-68-4
12042-78-3
7789-41-5
75-20-7
10043-52-4
10035-04-8
26264-06-2
7789-75-5
1305-62-0
7778-54-3
58398-71-3
1305-78-8
1305-79-9
7778-18-9
10101-41-4
76-22-2
120962-03-0
1333-86-4
124-38-9
471-34-1
584-08-7
39346-76-4
8001-79-4
8000-27-9
9005-81-6
9004-34-6
16887-00-6
7782-50-5
10049-04-4
78-73-9
Physico-
chemical
properties






X








X


X

X








X
Selected
toxic ity
reference
value



























X
X

Reference
2,8
2
4
8
1, 2, 3, 4, 7
1,4
4
1,4
1, 2, 3, 4
1,2,4
4
1, 2, 4, 7
1, 3, 4, 8
1,2,4
2
3
8
1,2,4
1, 3, 4, 6
1,2,4
1, 2, 3, 4, 8
1,2,4
8
3
1,4
1, 2, 3, 4
4,8
2
1, 2, 3, 4, 8
3,8
                This document is a draft for review purposes only and does not constitute Agency policy.
June 2015                                         A-17                     DRAFT—DO NOT CITE OR QUOTE

-------
Hydraulic Fracturing Drinking Water Assessment
Appendix A
Chemical name
Choline chloride
Chromium (III)
Chromium (VI)
Chromium acetate, basic
Chromium(lll) acetate
Citric acid
Citronella oil
Citronellol
Citrus extract
Coal, granular
Cobalt(ll) acetate
Coco-betaine
Coconut oil
Coconut oil acid/Diethanolamine condensate
(2:1)
Coconut trimethylammonium chloride
Copper
Copper sulfate
Copper(l) chloride
Copper(l) iodide
Copper(ll) chloride
Copper(ll) sulfate, pentahydrate
Corn flour
Corn sugar gum
Corundum (Aluminum oxide)
Cottonseed, flour
Coumarin
Cremophor(R) EL
Cristobalite
Crystalline silica, tridymite
CASRN
67-48-1
16065-83-1
18540-29-9
39430-51-8
1066-30-4
77-92-9
8000-29-1
106-22-9
94266-47-4
50815-10-6
71-48-7
68424-94-2
8001-31-8
68603-42-9
61789-18-2
7440-50-8
7758-98-7
7758-89-6
7681-65-4
7447-39-4
7758-99-8
68525-86-0
11138-66-2
1302-74-5
68308-87-2
91-64-5
61791-12-6
14464-46-1
15468-32-3
Physico-
chemical
properties
X




X

X






X










X



Selected
toxic ity
reference
value

X
X












X


X










Reference
1, 3, 4, 7, 8
2,6
6
2
1,2
1, 2, 3, 4, 7
3
3
1, 3, 4, 8
1,2,4
1,4
3
8
1
1,8
1,4
1,4,8
1,4
1, 2, 4, 6
1,3,4
8
4
1,2,4
4,8
2,4
3
1,3
1,2,4
1,2,4
                This document is a draft for review purposes only and does not constitute Agency policy.
June 2015                                         A-18                     DRAFT—DO NOT CITE OR QUOTE

-------
Hydraulic Fracturing Drinking Water Assessment
Appendix A
Chemical name
Cum en e
Cupric chloride dihydrate
Cyclohexane
Cyclohexanol
Cyclohexanone
Cyclohexylamine sulfate
D&C Red 28
D&C Red No. 33
Daidzein
Dapsone
Dazomet

Decamethylcyclopentasiloxane
Decyldimethylamine
Deuterium oxide
D-Glucitol
D-Gluconic acid
D-Glucopyranoside, methyl
D-Glucose
Di(2-ethylhexyl) phthalate
Diammonium peroxydisulfate
Diatomaceous earth
Diatomaceous earth, calcined
Dibromoacetonitrile
Dicalcium silicate
Dichloromethane
Didecyldimethylammonium chloride
Diethanolamine
Diethylbenzene
Diethylene glycol
CASRN
98-82-8
10125-13-0
110-82-7
108-93-0
108-94-1
19834-02-7
18472-87-2
3567-66-6
486-66-8
80-08-0
533-74-4

541-02-6
1120-24-7
7789-20-0
50-70-4
526-95-4
3149-68-6
50-99-7
117-81-7
7727-54-0
68855-54-9
91053-39-3
3252-43-5
10034-77-2
75-09-2
7173-51-5
111-42-2
25340-17-4
111-46-6
Physico-
chemical
properties
X

X
X
X
X
X
X
X
X
X


X

X
X
X
X
X



X

X
X
X
X
X
Selected
toxic ity
reference
value
X



X














X





X
X



Reference
1, 2, 3, 4
1,4,7
1,7
8
1,4
8
4
8
8
1,4
1, 2, 3, 4, 7, 8

8
3,4
8
1,3,4
1,4
2
1,4
1,4
1, 2, 3, 4, 6,
7,8
2,4
1,2,4
1, 2, 3, 4, 8
1,2,4
8
1, 2, 4, 8
1, 2, 3, 4, 6
1,3,4
1, 2, 3, 4, 7
                This document is a draft for review purposes only and does not constitute Agency policy.
June 2015                                         A-19                     DRAFT—DO NOT CITE OR QUOTE

-------
Hydraulic Fracturing Drinking Water Assessment
Appendix A
Chemical name
Diethylene glycol monomethyl ether
Diethylenetriamine
Diethylenetriamine reaction product with fatty
acid dimers
Diisobutyl ketone
Diisopropanolamine
Diisopropylnaphthalene
Dimethyl adipate
Dimethyl glutarate
Dimethyl polysiloxane
Dimethyl succinate
Dimethylaminoethanol
Dimethyldiallylammonium chloride
Diphenyl oxide
Dipotassium monohydrogen phosphate
Dipropylene glycol
Di-sec-butylphenol
Disodium dodecyl(sulphonatophenoxy)
benzenesulphonate
Disodium ethylenediaminediacetate
Disodium ethylenediaminetetraacetate
dihydrate
Disodium octaborate
Disodium octaborate tetrahydrate
Disodium sulfide
Distillates, petroleum, catalytic reformer
fractionator residue, low-boiling
Distillates, petroleum, heavy arom.
Distillates, petroleum, hydrodesulfurized light
catalytic cracked
Distillates, petroleum, hydrodesulfurized
middle
CASRN
111-77-3
111-40-0
68647-57-4
108-83-8
110-97-4
38640-62-9
627-93-0
1119-40-0
63148-62-9
106-65-0
108-01-0
7398-69-8
101-84-8
7758-11-4
25265-71-8
31291-60-8
28519-02-0
38011-25-5
6381-92-6
12008-41-2
12280-03-4
1313-82-2
68477-31-6
67891-79-6
68333-25-5
64742-80-9
Physico-
chemical
properties
X
X

X
X
X
X
X

X
X
X
X

X
X
X
X
X







Selected
toxic ity
reference
value


























Reference
1,2,4
1, 2, 4, 5
2
8
8
3,4
8
1,4
1,2,4
8
2,4
3,4
3
5
1,3,4
1
1
1,4
1
4,8
1,4
8
1,4
1,4
1
1
                This document is a draft for review purposes only and does not constitute Agency policy.
June 2015                                         A-20                     DRAFT—DO NOT CITE OR QUOTE

-------
Hydraulic Fracturing Drinking Water Assessment
Appendix A
Chemical name
Distillates, petroleum, hydrotreated heavy
naphthenic
Distillates, petroleum, hydrotreated heavy
paraffinic

Distillates, petroleum, hydrotreated light
Distillates, petroleum, hydrotreated light
naphthenic
Distillates, petroleum, hydrotreated light
paraffinic
Distillates, petroleum, hydrotreated middle
Distillates, petroleum, light catalytic cracked
Distillates, petroleum, light hydrocracked
Distillates, petroleum, solvent-dewaxed heavy
paraffinic
Distillates, petroleum, solvent-refined heavy
naphthenic
Distillates, petroleum, steam-cracked
Distillates, petroleum, straight-run middle
Distillates, petroleum, sweetened middle
Ditallow alkyl ethoxylated amines
D-Lactic acid
D-Limonene

Docusate sodium
Dodecamethylcyclohexasiloxane
Dodecane
Dodecylbenzene
Dodecylbenzenesulfonic acid
Dodecylbenzenesulfonic acid,
monoethanolamine salt
Edifas B
EDTA, copper salt
Endo- 1,4-. beta. -mannanase
CASRN
64742-52-5
64742-54-7

64742-47-8
64742-53-6
64742-55-8
64742-46-7
64741-59-9
64741-77-1
64742-65-0
64741-96-4
64742-91-2
64741-44-2
64741-86-2
71011-04-6
10326-41-7
5989-27-5

577-11-7
540-97-6
112-40-3
123-01-3
27176-87-0
26836-07-7
9004-32-4
12276-01-6
37288-54-3
Physico-
chemical
properties















X
X

X

X
X
X
X



Selected
toxic ity
reference
value
















X





X




Reference
1, 2, 3, 4
1,2,4
1, 2, 3, 4, 5,
7,8
1,2,8
8
1, 2, 3, 4, 8
1,4
3
1
1,4
1,4
1,2,4
1,4
3
1,4
1, 3, 4, 5, 7, 8

1
8
8
3,4
2, 3, 4, 8
1,4
2,3,4
1,5,6
3,8
                This document is a draft for review purposes only and does not constitute Agency policy.
June 2015                                         A-21                      DRAFT—DO NOT CITE OR QUOTE

-------
Hydraulic Fracturing Drinking Water Assessment
Appendix A
Chemical name
Epichlorohydrin
Epoxy resin
Erucic amidopropyl dimethyl betaine
Ethanaminium, N,N,N-trimethyl-2-[(l-oxo-2-
propenyl)oxy]-, chloride
Ethanaminium, N,N,N-trimethyl-2-[(l-oxo-2-
propenyl)oxy]-,chloride, polymer with 2-
propenamide
Ethanaminium, N,N,N-trimethyl-2-[(2-methyl-l-
oxo-2-propen-l-yl)oxy]-, chloride (1:1), polymer
with 2-propenamide
Ethanaminium, N,N,N-trimethyl-2-[(2-methyl-l-
oxo-2-propenyl)oxy]-, methyl sulfate,
homopolymer
Ethane

Ethanol
Ethanol, 2,2',2"-nitrilotris-, tris(dihydrogen
phosphate) (ester), sodium salt
Ethanol, 2,2'-iminobis-, N-coco alkyl derivs., N-
oxides
Ethanol, 2,2'-iminobis-, N-tallow alkyl derivs.
Ethanol, 2,2'-oxybis-, reaction products with
ammonia, morpholine derivs. residues
Ethanol, 2,2-oxybis-, reaction products with
ammonia, morpholine derivs. residues, acetates
(salts)
Ethanol, 2,2-oxybis-, reaction products with
ammonia, morpholine derivs. residues, reaction
products with sulfur dioxide
Ethanol, 2-[2-[2-(tridecyloxy)ethoxy]ethoxy]-,
hydrogen sulfate, sodium salt
Ethanol, 2-amino-, polymer with formaldehyde
Ethanol, 2-amino-, reaction products with
ammonia, by-products from,
phosphonomethylated
CASRN
106-89-8
25085-99-8
149879-98-1
44992-01-0
69418-26-4
35429-19-7
27103-90-8
74-84-0

64-17-5
68171-29-9
61791-47-7
61791-44-4
68909-77-3
68877-16-7
102424-23-7
25446-78-0
34411-42-2
68649-44-5
Physico-
chemical
properties
X


X



X

X
X





X


Selected
toxic ity
reference
value
X


















Reference
1,4,8
1,4,8
1,3
3
1,3,4
8
8
2,5
1, 2, 3, 4, 5,
6,8
4
1
1
4,8
4
4
1,4
4
4
                This document is a draft for review purposes only and does not constitute Agency policy.
June 2015                                         A-22                      DRAFT—DO NOT CITE OR QUOTE

-------
Hydraulic Fracturing Drinking Water Assessment
Appendix A
Chemical name
Ethanolamine
Ethoxylated dodecyl alcohol
Ethoxylated hydrogenated tallow alkylamines
Ethoxylated, propoxylated trimethylolpropane
Ethyl acetate
Ethyl acetoacetate
Ethyl benzoate
Ethyl lactate
Ethyl salicylate
E thylbenzene
Ethylcellulose
Ethylene

Ethylene glycol
Ethylene oxide
Ethylenediamine
Ethylenediaminetetraacetic acid
Ethylenediaminetetraacetic acid tetrasodium
salt
Ethylenediaminetetraacetic acid, diammonium
copper salt
Ethylenediaminetetraacetic acid, disodium salt
Ethyne
Fats and Glyceridic oils, vegetable,
hydrogenated
Fatty acid, tall oil, hexa esters with sorbitol,
ethoxylated
Fatty acids, C 8-18 and C18-unsaturated
compounds with diethanolamine
Fatty acids, C14-18 and C16-18-unsatd., distn.
residues
Fatty acids, C18-unsatd., dimers
CASRN
141-43-5
9002-92-0
61790-82-7
52624-57-4
141-78-6
141-97-9
93-89-0
97-64-3
118-61-6
100-41-4
9004-57-3
74-85-1

107-21-1
75-21-8
107-15-3
60-00-4
64-02-8
67989-88-2
139-33-3
74-86-2
68334-28-1
61790-90-7
68604-35-3
70321-73-2
61788-89-4
Physico-
chemical
properties
X
X


X
X
X
X
X
X

X

X
X
X
X
X

X
X




X
Selected
toxic ity
reference
value




X




X



X
X
X










Reference
1, 2, 3, 4, 6
4
4
3
1,4,7
1,4
3
3
3
1, 2, 3, 4, 7
2
8
1, 2, 3, 4, 6,
7,8
1, 2, 3, 4
2,4
1,2,4
1, 2, 3, 4
4
1, 3, 4, 8
7
8
1,4
3
2
2
                This document is a draft for review purposes only and does not constitute Agency policy.
June 2015                                         A-23                      DRAFT—DO NOT CITE OR QUOTE

-------
Hydraulic Fracturing Drinking Water Assessment
Appendix A
Chemical name
Fatty acids, C18-unsatd., dimers, compds. with
ethoxylated tall-oil fatty acid-
polyethylenepolyamine reaction products
Fatty acids, C18-unsatd., dimers, ethoxylated
propoxylated
Fatty acids, coco, ethoxylated
Fatty acids, coco, reaction products with
diethylenetriamine and soya fatty acids,
ethoxylated, chloromethane-quaternized
Fatty acids, coco, reaction products with
ethanolamine, ethoxylated
Fatty acids, tall oil, reaction products with
acetophenone, formaldehyde and thiourea
Fatty acids, tall-oil
Fatty acids, tall-oil, reaction products with
diethylenetriamine
Fatty acids, tall-oil, reaction products with
diethylenetriamine, maleic anhydride,
tetraethylenepentamine and
triethylenetetramine
Fatty acids, tallow, sodium salts
Fatty acids, vegetable-oil, reaction products
with diethylenetriamine
Fatty quaternary ammonium chloride
FD&C Blue no. 1
FD&C Yellow 5
FD&C Yellow 6
Ferric chloride
Ferric sulfate
Ferrous sulfate monohydrate
Ferumoxytol
Fiberglass
Formaldehyde
CASRN
68132-59-2
68308-89-4
61791-29-5
68604-75-1
61791-08-0
68188-40-9
61790-12-3
61790-69-0
68990-47-6
8052-48-0
68153-72-0
61789-68-2
3844-45-9
1934-21-0
2783-94-0
7705-08-0
10028-22-5
17375-41-6
1309-38-2
65997-17-3
50-00-0
Physico-
chemical
properties












X
X
X





X
Selected
toxic ity
reference
value




















X
Reference
8
8
3
8
3
3
1, 2, 3, 4
1,4
8
1,3
3
1,4
1,4
8
8
1,3,4
1,4
2
8
2,3,4
1, 2, 3, 4
                This document is a draft for review purposes only and does not constitute Agency policy.
June 2015                                         A-24                     DRAFT—DO NOT CITE OR QUOTE

-------
Hydraulic Fracturing Drinking Water Assessment
Appendix A
Chemical name
Formaldehyde polymer with 4,1,1-
(dimethylethyl)phenol and methyloxirane
Formaldehyde polymer with methyl oxirane, 4-
nonylphenol and oxirane
Formaldehyde, polymer with 4-(l,l-
dimethylethyl)phenol, 2-methyloxirane and
oxirane
Formaldehyde, polymer with 4-(l,l-
dimethylethyl)phenol, 2-methyloxirane, 4-
nonylphenol and oxirane
Formaldehyde, polymer with 4-nonylphenol
and oxirane
Formaldehyde, polymer with 4-nonylphenol
and phenol
Formaldehyde, polymer with ammonia and
phenol
Formaldehyde, polymer with bisphenol A
Formaldehyde, polymer with Nl-(2-
aminoethyl)-l,2-ethanediamine, benzylated
Formaldehyde, polymer with nonylphenol and
oxirane
Formaldehyde, polymers with branched 4-
nonylphenol, oxirane and 2-methyloxirane
Formaldehyde/ amine
Formamide
Formic acid

Formic acid, potassium salt
Frits, chemicals
Fuel oil, no. 2
Fuels, diesel
Fuels, diesel, no. 2
Fuller's earth
Fumaric acid
Fumes, silica
CASRN
29316-47-0
63428-92-2
30704-64-4
68188-99-8
30846-35-6
40404-63-5
35297-54-2
25085-75-0
70750-07-1
55845-06-2
153795-76-7
50-00-0_3
75-12-7
64-18-6

590-29-4
65997-18-4
68476-30-2
68334-30-5
68476-34-6
8031-18-3
110-17-8
69012-64-2
Physico-
chemical
properties












X
X

X





X

Selected
toxic ity
reference
value













X









Reference
3
4,8
1, 2, 4, 8
8
1,4
8
1,4
4
8
4
13
1, 2, 3, 4
1, 2, 3, 4
1, 2, 3, 4, 6, 7

1,3,4
8
1,2
2
2,4,8
2
1, 2, 3, 4, 6
8
                This document is a draft for review purposes only and does not constitute Agency policy.
June 2015                                         A-25                      DRAFT—DO NOT CITE OR QUOTE

-------
Hydraulic Fracturing Drinking Water Assessment
Appendix A
Chemical name
Furfural
Furfuryl alcohol
Galantamine hydrobromide
Gas oils, petroleum, straight-run
Gelatin
Gilsonite
Gluconic acid
Glutaraldehyde
Glycerides, C14-18 and C16-18-unsatd. mono-
and di-
Glycerol
Glycine, N-(carboxymethyl)-N-(2-hydroxyethyl)-
, disodium salt
Glycine, N-(hydroxymethyl)-, monosodium salt
Glycine, N,N-bis(carboxymethyl)-, trisodium salt
Glycine, N-[2-[bis(carboxymethyl)amino]ethyl]-
N-(2-hydroxyethyl)-, trisodium salt
Glycolic acid
Glycolic acid sodium salt
Glyoxal
Glyoxylic acid
Goethite (Fe(OH)O)
Guar gum

Guargum, carboxymethyl 2-hydroxypropyl
ether, sodium salt
Gypsum (Ca(SO4).2H2O)
Hematite
Hemicellulase
Hemicellulase enzyme concentrate
Heptane
Heptene, hydroformylation products, high-
boiling
CASRN
98-01-1
98-00-0
69353-21-5
64741-43-1
9000-70-8
12002-43-6
133-42-6
111-30-8
67701-32-0
56-81-5
135-37-5
70161-44-3
5064-31-3
139-89-9
79-14-1
2836-32-0
107-22-2
298-12-4
1310-14-1
9000-30-0

68130-15-4
13397-24-5
1317-60-8
9012-54-8
9025-56-3
142-82-5
68526-88-5
Physico-
chemical
properties
X
X
X



X
X

X
X
X
X
X
X
X
X
X








X

Selected
toxic ity
reference
value
X















X











Reference
1,4
1,4
8
1,4
1,4
1,2,4
7
1, 2, 3, 4, 7
8
1, 2, 3, 4, 5
1
8
1, 2, 3, 4
1
1,3,4
1,3,4
1,2,4
1
8
1, 2, 3, 4, 7, 8

1, 2, 3, 4, 7
2,4
1,2,4
1, 2, 3, 4, 5
3,4
1,2
1,4
                This document is a draft for review purposes only and does not constitute Agency policy.
June 2015                                         A-26                     DRAFT—DO NOT CITE OR QUOTE

-------
Hydraulic Fracturing Drinking Water Assessment
Appendix A
Chemical name
Hexadecyltrimethylammonium bromide
Hexane
Hexanedioic acid
Humic acids, commercial grade
Hydrazine
Hydrocarbons, terpene processing by-products

Hydrochloric acid
Hydrogen fluoride
Hydrogen peroxide
Hydrogen sulfide
Hydroxyethylcellulose
Hydroxylamine hydrochloride
Hydroxylamine sulfate (2:1)
Hydroxypropyl cellulose
Hydroxypropyl guar gum

Hydroxyvalerenic acid
Hypochlorous acid
Illite
llmenite (FeTiCh), cone.
Indole
Inulin, carboxymethyl ether, sodium salt
Iridium oxide
Iron
Iron oxide
Iron oxide (FesCU)
Iron(ll) sulfate
Iron(ll) sulfate heptahydrate
Iron(lll) oxide
Isoascorbic acid
CASRN
57-09-0
110-54-3
124-04-9
1415-93-6
302-01-2
68956-56-9

7647-01-0
7664-39-3
7722-84-1
7783-06-4
9004-62-0
5470-11-1
10039-54-0
9004-64-2
39421-75-5

1619-16-5
7790-92-3
12173-60-3
98072-94-7
120-72-9
430439-54-6
12030-49-8
7439-89-6
1332-37-2
1317-61-9
7720-78-7
7782-63-0
1309-37-1
89-65-6
Physico-
chemical
properties
X
X
X














X



X








X
Selected
toxic ity
reference
value

X
X

X



















X






Reference
1
5
1, 2, 4, 6
2
8
1,3,4
1, 2, 3, 4, 5,
6,7,8
1,2,4
1,3,4
1,2
1, 2, 3, 4
1,3,4
4
2,4
1, 3, 4, 5, 6, 8

8
8
8
8
2
1,4
8
2,4
1,4
4
2
1, 2, 3, 4
1,2,4
1,3,4
                This document is a draft for review purposes only and does not constitute Agency policy.
June 2015                                         A-27                      DRAFT—DO NOT CITE OR QUOTE

-------
Hydraulic Fracturing Drinking Water Assessment
Appendix A
Chemical name
Isobutane
Isobutene
Isooctanol
Isopentyl alcohol
Isopropanol

Isopropanolamine dodecylbenzene
Isopropylamine
Isoquinoline
Isoquinoline, reaction products with benzyl
chloride and quinoline
Isoquinolinium, 2-(phenylmethyl)-, chloride
Isotridecanol, ethoxylated
Kaolin
Kerosine, petroleum, hydrodesulfurized
Kieselguhr
Kyanite
Lactic acid
Lactose
Latex 2000 TM
Lauryl hydroxysultaine
Lavandula hybrida abrial herb oil
L-Dilactide
Lead
Lecithin
L-Glutamic acid
Lignite
Lignosulfuric acid
Ligroine
Limestone
Linseed oil
CASRN
75-28-5
115-11-7
26952-21-6
123-51-3
67-63-0

42504-46-1
75-31-0
119-65-3
68909-80-8
35674-56-7
9043-30-5
1332-58-7
64742-81-0
61790-53-2
1302-76-7
50-21-5
63-42-3
9003-55-8
13197-76-7
8022-15-9
4511-42-6
7439-92-1
8002-43-5
56-86-0
129521-66-0
8062-15-5
8032-32-4
1317-65-3
8001-26-1
Physico-
chemical
properties
X
X
X
X
X

X
X
X
X
X





X
X

X

X


X





Selected
toxic ity
reference
value






















X







Reference
2
8
1,4,5
1,4
1, 2, 3, 4, 6, 7

1,3,4
1,4
8
3
3
1, 3, 4, 8
1,2,4
1,2,4
1,2,4
1,2,4
1,4,8
3
2,4
1
3
1,4
1,4
4
8
2
2
8
1, 2, 3, 4
8
                This document is a draft for review purposes only and does not constitute Agency policy.
June 2015                                         A-28                     DRAFT—DO NOT CITE OR QUOTE

-------
Hydraulic Fracturing Drinking Water Assessment
Appendix A
Chemical name
L-Lactic acid
Magnesium carbonate (1:1)
Magnesium carbonate (l:x)
Magnesium chloride
Magnesium chloride hexahydrate
Magnesium hydroxide
Magnesium iron silicate
Magnesium nitrate
Magnesium oxide
Magnesium peroxide
Magnesium phosphide
Magnesium silicate
Magnesium sulfate
Maleicacid homopolymer
Methanamine-N-methyl polymer with
chloromethyl oxirane
Methane

Methanol
Methenamine
Methoxyacetic acid
Methyl cellulose
Methyl salicylate
Methyl vinyl ketone
Methylcyclohexane
Methylene bis(thiocyanate)
Methylenebis(5-methyloxazolidine)
Methyloxirane polymer with oxirane, mono
(nonylphenol) ether, branched
Mica
Mineral oil - includes paraffin oil
CASRN
79-33-4
7757-69-9
546-93-0
7786-30-3
7791-18-6
1309-42-8
19086-72-7
10377-60-3
1309-48-4
14452-57-4
12057-74-8
1343-88-0
7487-88-9
26099-09-2
25988-97-0
74-82-8

67-56-1
100-97-0
625-45-6
9004-67-5
119-36-8
78-94-4
108-87-2
6317-18-6
66204-44-2
68891-11-2
12001-26-2
8012-95-1
Physico-
chemical
properties
X














X

X
X
X

X
X
X
X
X



Selected
toxic ity
reference
value

















X










X
Reference
1,4,8
8
1,3,4
1,2,4
4
1,4
1,4
1,2,4
1, 2, 3, 4
1,4
1
1,4
8
8
4
2,5
1, 2, 3, 4, 5,
6,7,8
1,2,4
8
8
1, 2, 3, 4, 7
1,4
1
2
2
3
1, 2, 4, 6
4,8
                This document is a draft for review purposes only and does not constitute Agency policy.
June 2015                                         A-29                     DRAFT—DO NOT CITE OR QUOTE

-------
Hydraulic Fracturing Drinking Water Assessment
Appendix A
Chemical name
Mineral spirits
Mono- and di- potassium salts of phosphorous
acid
Montmorillonite
Morpholine
Morpholinium, 4-ethyl-4-hexadecyl-, ethyl
sulfate
MT6
Mullite
N-(2-Acryloyloxyethyl)-N-benzyl-N,N-
dimethylammonium chloride
N-(3-Chloroallyl)hexaminium chloride
N,N,N-Trimethyl-2[l-oxo-2-propenyl]oxy
ethanaminimum chloride, homopolymer
N,N,N-Trimethyl-3-((l-oxooctadecyl)amino)-l-
propanaminium methyl sulfate
N,N,N-Trimethyloctadecan-l-aminium chloride
N,N'-Dibutylthiourea
N,N-Dimethyldecylamine oxide
N,N-Dimethylformamide
N,N-Dimethylmethanamine hydrochloride
N,N-Dimethyl-methanamine-N-oxide
N,N-dimethyloctadecylamine hydrochloride
N,N'-Methylenebisacrylamide
Naphtha, petroleum, heavy catalytic reformed
Naphtha, petroleum, hydrotreated heavy
Naphthalene

Naphthalenesulfonic acid, bis(l-methylethyl)-
Naphthalenesulfonic acid, polymer with
formaldehyde, sodium salt
Naphthalenesulphonic acid, bis (1-methylethyl)-
methyl derivatives
CASRN
64475-85-0
13492-26-7
1318-93-0
110-91-8
78-21-7
76-31-3
1302-93-8
46830-22-2
4080-31-3
54076-97-0
19277-88-4
112-03-8
109-46-6
2605-79-0
68-12-2
593-81-7
1184-78-7
1613-17-8
110-26-9
64741-68-0
64742-48-9
91-20-3

28757-00-8
9084-06-4
99811-86-6
Physico-
chemical
properties



X
X


X
X

X
X
X
X
X
X
X
X
X


X

X

X
Selected
toxic ity
reference
value














X






X




Reference
2
8
2
1,2,4
8
8
1,2, 4, 8
3
8
3
1
1,3,4
1,4
1,3,4
1, 2, 4, 5, 8
1, 4, 5, 7
3
1,4
1,4
1, 2, 3, 4
1, 2, 3, 4, 8
1, 2, 3, 4, 5, 7

1,3,4
2
1
                This document is a draft for review purposes only and does not constitute Agency policy.
June 2015                                         A-30                     DRAFT—DO NOT CITE OR QUOTE

-------
Hydraulic Fracturing Drinking Water Assessment
Appendix A
Chemical name
Naphthenic acid ethoxylate
Navy fuels JP-5
Nickel sulfate
Nickel(ll) sulfate hexahydrate
Nitriles, tallow, hydrogenated
Nitrilotriacetamide
Nitrilotriacetic acid
Nitrilotriacetic acid trisodium monohydrate
Nitrogen
N-Methyl-2-pyrrolidone
N-Methyldiethanolamine
N-Methylethanolamine
N-Methyl-N-hydroxyethyl-N-
hydroxyethoxyethylamine
N-Oleyl diethanolamide
Nonyl nonoxynol-10
Nonylphenol (mixed)
Octamethylcyclotetrasiloxane
Octoxynol-9
Oil of eucalyptus
Oil of lemongrass
Oil of rosemary
Oleic acid
Olivine-group minerals
Orange terpenes
Oxirane, 2-methyl-, polymer with oxirane, ether
with (chloromethyl) oxirane polymer with 4,4'-
(1-methylidene) bis[phenol]
Oxirane, 2-methyl-, polymer with oxirane,
mono(2-ethylhexyl) ether
Oxirane, 2-methyl-, polymer with oxirane,
monodecyl ether
CASRN
68410-62-8
8008-20-6_2
7786-81-4
10101-97-0
61790-29-2
4862-18-4
139-13-9
18662-53-8
7727-37-9
872-50-4
105-59-9
109-83-1
68213-98-9
13127-82-7
9014-93-1
25154-52-3
556-67-2
9036-19-5
8000-48-4
8007-02-1
8000-25-7
112-80-1
1317-71-1
8028-48-6
68036-95-3
64366-70-7
37251-67-5
Physico-
chemical
properties
X




X
X
X

X
X
X
X
X







X





Selected
toxic ity
reference
value






X
X

X

















Reference
4
1, 2, 3, 4, 8
2
1,4
4
1,4,7
1,4
1,4
1, 2, 3, 4, 6
1,4
2,4,8
4
4
1,4
4
1,4
8
1, 2, 3, 4, 8
3
3
3
2,4
4
4
8
8
8
                This document is a draft for review purposes only and does not constitute Agency policy.
June 2015                                         A-31                      DRAFT—DO NOT CITE OR QUOTE

-------
Hydraulic Fracturing Drinking Water Assessment
Appendix A
Chemical name
Oxirane, methyl-, polymer with oxirane, mono-
C10-16-alkyl ethers, phosphates
Oxygen
Ozone
Paraffin waxes and Hydrocarbon waxes
Paraformaldehyde
PEG- 10 Hydrogenated tallow amine
Pentaethylenehexamine
Pentane
Pentyl acetate
Pentyl butyrate
Peracetic acid
Perboric acid, sodium salt, monohydrate
Perlite
Petrolatum, petroleum, oxidized
Petroleum
Petroleum distillate hydrotreated light
Phenanthrene
Phenol
Phenol, 4,4'-(l-methylethylidene)bis-, polymer
with 2-(chloromethyl)oxirane, 2-methyloxirane
and oxirane
Phenol-formaldehyde resin
Phosphine
Phosphonic acid
Phosphonic acid (dimethylamino(methylene))
Phosphonic acid, (((2-[(2-
hydroxyethyl)(phosphonomethyl)amino)ethyl)i
mino]bis(methylene))bis-, compd. with 2-
aminoethanol
Phosphonic acid, (l-hydroxyethylidene)bis-,
potassium salt
CASRN
68649-29-6
7782-44-7
10028-15-6
8002-74-2
30525-89-4
61791-26-2
4067-16-7
109-66-0
628-63-7
540-18-1
79-21-0
10332-33-9
93763-70-3
64743-01-7
8002-05-9
6742-47-8
85-01-8
108-95-2
68123-18-2
9003-35-4
7803-51-2
13598-36-2
29712-30-9
129828-36-0
67953-76-8
Physico-
chemical
properties






X
X
X
X
X





X
X




X
X
X
Selected
toxic ity
reference
value

















X


X




Reference
1,4
4
8
1
2
1,3
4
2,5
3
3
8
1,8
4
3
1,2
8
6
1,2,4
8
1, 2, 4, 7
1,4
1,4
1
1
4
                This document is a draft for review purposes only and does not constitute Agency policy.
June 2015                                         A-32                      DRAFT—DO NOT CITE OR QUOTE

-------
Hydraulic Fracturing Drinking Water Assessment
Appendix A
Chemical name
Phosphonic acid, (l-hydroxyethylidene)bis-,
tetrasodium salt
Phosphonic acid,
[[(phosphonomethyl)imino]bis[2,l-
ethanediylnitrilobis(methylene)]]tetrakis-
Phosphonicacid,
[[(phosphonomethyl)imino]bis[2,l-
ethanediylnitrilobis(methylene)]]tetrakis-,
ammonium salt (l:x)
Phosphonic acid,
[[(phosphonomethyl)imino]bis[2,l-
ethanediylnitrilobis(methylene)]]tetrakis-,
sodium salt
Phosphonic acid,
[[(phosphonomethyl)imino]bis[6,l-
hexanediylnitrilobis(methylene)]]tetrakis-
Phosphoric acid
Phosphoric acid, aluminium sodium salt
Phosphoric acid, ammonium salt (1:3)
Phosphoric acid, diammonium salt
Phosphoric acid, mixed decyl and Et and octyl
esters
Phosphorous acid
Phthalic anhydride
Pine oils
Pluronic F-127
Policapram (Nylon 6)
Poly (acrylamide-co-acrylic acid), partial sodium
salt
Poly(acrylamide-co-acrylic acid)
Poly(lactide)
Poly(oxy-l,2-ethanediyl), .alpha.-(nonylphenyl)-
. omega. -hydroxy-, phosphate
Poly(oxy-l,2-ethanediyl), .alpha.-(octylphenyl)-
. omega. -hydroxy-, branched
CASRN
3794-83-0
15827-60-8
70714-66-8
22042-96-2
34690-00-1
7664-38-2
7785-88-8
10361-65-6
7783-28-0
68412-60-2
10294-56-1
85-44-9
8002-09-3
9003-11-6
25038-54-4
62649-23-4
9003-06-9
26680-10-4
51811-79-1
68987-90-6
Physico-
chemical
properties
X
X
X
X
X






X







X
Selected
toxic ity
reference
value





X
X

X


X








Reference
1,4
1,2,4
3
3
1,4,8
1,2,4
1,2
8
2
1
1
1,4
1,2,4
1, 3, 4, 8
1,4
3,4
4,8
1
1,4
1,4
                This document is a draft for review purposes only and does not constitute Agency policy.
June 2015                                         A-33                      DRAFT—DO NOT CITE OR QUOTE

-------
Hydraulic Fracturing Drinking Water Assessment
Appendix A
Chemical name
Poly(oxy-l,2-ethanediyl), .alpha.,. alpha. l-[[(9Z)-
9-octadecenylimino]di-2,l-
ethanediyl]bis[.omega.-hydroxy-
Poly(oxy-l,2-ethanediyl), .alpha.-[(9Z)-l-oxo-9-
octadecenyl]-.omega.-hydroxy-
Poly(oxy-l,2-ethanediyl), .alpha.-hydro-
.omega. -hydroxy-, mono-C10-14-alkyl ethers,
phosphates
Poly(oxy-l,2-ethanediyl), .alpha.-hydro-
. omega. -hydroxy-, mono-C8-10-alkyl ethers,
phosphates
Poly(oxy-l,2-ethanediyl), . alpha. -isodecyl-
. omega. -hydroxy-
Poly(oxy-l,2-ethanediyl), .alpha.-sulfo-.omega.-
hydroxy-, C10-16-alkyl ethers, sodium salts
Poly(oxy-l,2-ethanediyl), . alpha. -sulfo-. omega. -
hydroxy-, C12-14-alkyl ethers, sodium salts
Poly(oxy-l,2-ethanediyl), alpha-(2,3,4,5-
tetramethylnonyl)-omega-hydroxy
Poly(oxy-l,2-ethanediyl), alpha-(nonylphenyl)-
omega-hydroxy-,branched, phosphates
Poly(oxy-l,2-ethanediyl), alpha-hexyl-omega-
hydroxy
Poly(oxy-l,2-ethanediyl), alpha-hydro-omega-
hydroxy-, (9Z)-9-octadecenoate
Poly(oxy-l,2-ethanediyl), alpha-hydro-omega-
hydroxy-, ether with alpha-fluoro-omega-(2-
hydroxyethyl)poly(difluoromethylene) (1:1)
Poly(oxy-l,2-ethanediyl), alpha-hydro-omega-
hydroxy-, ether with D-glucitol (2:1), tetra-(9Z)-
9-octadecenoate
Poly(oxy-l,2-ethanediyl), alpha-sulfo-omega-
(decyloxy)-, ammonium salt (1:1)
Poly(oxy-l,2-ethanediyl), alpha-sulfo-omega-
(hexyloxy)-, ammonium salt (1:1)
Poly(oxy-l,2-ethanediyl), alpha-sulfo-omega-
(hexyloxy)-, C6-10-alkyl ethers, ammonium salts
CASRN
26635-93-8
9004-96-0
68585-36-4
68130-47-2
61827-42-7
68585-34-2
68891-38-3
68015-67-8
68412-53-3
31726-34-8
56449-46-8
65545-80-4
61723-83-9
52286-19-8
63428-86-4
68037-05-8
Physico-
chemical
properties
















Selected
toxic ity
reference
value
















Reference
1,4
8
8
8
8
8
1,4
1
1
3,8
3
1
8
4
1,3,4
3,4
                This document is a draft for review purposes only and does not constitute Agency policy.
June 2015                                         A-34                     DRAFT—DO NOT CITE OR QUOTE

-------
Hydraulic Fracturing Drinking Water Assessment
Appendix A
Chemical name
Poly(oxy-l,2-ethanediyl), alpha-sulfo-omega—
(nonylphenoxy)-
Poly(oxy-l,2-ethanediyl), alpha-sulfo-omega-
(octyloxy)-, ammonium salt (1:1)
Poly(oxy-l,2-ethanediyl), alpha-sulfo-omega-
hydroxy-, C10-12-alkyl ethers, ammonium salts
Poly(oxy-l,2-ethanediyl), alpha-tridecyl-omega-
hydroxy-
Poly(oxy-l,2-ethanediyl), alpha-undecyl-omega-
hydroxy-, branched and linear
Poly-(oxy-l,2-ethanediyl)-alpha-undecyl-
omega-hydroxy
Poly(oxy-l,2-ethanediyl)-nonylphenyl-hydroxy
branched
Poly(sodium-p-styrenesulfonate)
Poly(tetrafluoroethylene)
Poly [imino(l,6-dioxo-l,6-hexanediyl)imino- 1,6-
hexanediyl]
Polyacrylamide
Polyacrylate/ polyacrylamide blend
Polyacrylic acid, sodium bisulfite terminated
Polyethylene glycol
Polyethylene glycol (9Z)-9-octadecenyl ether
Polyethylene glycol ester with tall oil fatty acid
Polyethylene glycol monobutyl ether
Polyethylene glycol mono-C8-10-alkyl ether
sulfate ammonium
Polyethylene glycol nonylphenyl ether
Polyethylene glycol tridecyl ether phosphate
Polyethyleneimine
Polyglycerol
Poly-L-aspartic acid sodium salt
Polyoxyethylene sorbitan trioleate
CASRN
9081-17-8
52286-18-7
68890-88-0
24938-91-8
127036-24-2
34398-01-1
127087-87-0
25704-18-1
9002-84-0
32131-17-2
9003-05-8
NOCAS_51256
66019-18-9
25322-68-3
9004-98-2
68187-85-9
9004-77-7
68891-29-2
9016-45-9
9046-01-9
9002-98-6
25618-55-7
34345-47-6
9005-70-3
Physico-
chemical
properties
























Selected
toxic ity
reference
value
























Reference
4
4
8
1,3,4
1
1, 3, 4, 8
1, 2, 3, 4
1,4
8
2
1, 2, 4, 6
2
3
1, 2, 3, 4, 7, 8
8
1
1,4
1,3,4
1, 2, 3, 4, 8
1,3,4
4
2
8
3
                This document is a draft for review purposes only and does not constitute Agency policy.
June 2015                                         A-35                      DRAFT—DO NOT CITE OR QUOTE

-------
Hydraulic Fracturing Drinking Water Assessment
Appendix A
Chemical name
Polyoxyethylene(10)nonylphenyl ether
Polyoxyl 15 hydroxystearate
Polyoxypropylenediamine
Polyphosphoric acids, esters with
triethanolamine, sodium salts
Polyphosphoric acids, sodium salts
Polypropylene glycol
Polypropylene glycol glycerol triether,
epichlorohydrin, bisphenol A polymer
Polyquaternium 5
Polysorbate 20
Polysorbate 60
Polysorbate 80
Polyvinyl acetate copolymer
Polyvinyl acetate, partially hydrolyzed
Polyvinyl alcohol
Polyvinyl alcohol/polyvinyl acetate copolymer
Polyvinylidene chloride
Polyvinylpyrrolidone
Portland cement
Potassium acetate
Potassium aluminum silicate
Potassium antimonate
Potassium bisulfate
Potassium borate
Potassium borate (l:x)
Potassium carbonate sesquihydrate
Potassium chloride
Potassium dichromate
Potassium hydroxide
CASRN
26027-38-3
70142-34-6
9046-10-0
68131-72-6
68915-31-1
25322-69-4
68683-13-6
26006-22-4
9005-64-5
9005-67-8
9005-65-6
9003-20-7
304443-60-5
9002-89-5
NOCAS_50147
9002-85-1
9003-39-8
65997-15-1
127-08-2
1327-44-2
29638-69-5
7646-93-7
12712-38-8
20786-60-1
6381-79-9
7447-40-7
7778-50-9
1310-58-3
Physico-
chemical
properties


















X









Selected
toxic ity
reference
value




X























Reference
1, 2, 3, 4, 8
8
1
1
1,4
1,2,4
1
1,4
8
3,4
3,4
2
8
1,2,4
2
8
8
2,4
1,3,4
5
1,4
8
3
1,3
5
1, 2, 3, 4, 5,
6,7
4
1, 2, 3, 4, 6
                This document is a draft for review purposes only and does not constitute Agency policy.
June 2015                                         A-36                     DRAFT—DO NOT CITE OR QUOTE

-------
Hydraulic Fracturing Drinking Water Assessment
Appendix A
Chemical name
Potassium iodide
Potassium metaborate
Potassium oleate
Potassium oxide
Potassium persulfate
Potassium phosphate, tribasic
Potassium sulfate
Propane
Propanol, l(or 2)-(2-methoxymethylethoxy)-

Propargyl alcohol
Propylene carbonate
Propylene pentamer
p-Xylene
Pyridine, alkyl derivs.
Pyridinium, l-(phenylmethyl)-, alkyl derivs.,
chlorides
Pyridinium, l-(phenylmethyl)-, C7-8-alkyl
derivs., chlorides
Pyrimidine
Pyrrole

Quartz-alpha (SiO2)
Quaternary ammonium compounds (2-
ethylhexyl) hydrogenated tallow alkyl)dimethyl,
methyl sulfates
Quaternary ammonium compounds, (oxydi-2,1-
ethanediyl)bis[coco alkyldimethyl, dichlorides
Quaternary ammonium compounds,
benzyl(hydrogenated tallow alkyl)dimethyl,
bis(hydrogenated tallow
alkyl)dimethylammonium salt with bentonite
CASRN
7681-11-0
13709-94-9
143-18-0
12136-45-7
7727-21-1
7778-53-2
7778-80-5
74-98-6
34590-94-8

107-19-7
108-32-7
15220-87-8
106-42-3
68391-11-7
100765-57-9
70914-44-2
289-95-2
109-97-7

14808-60-7
308074-31-9
68607-28-3
71011-25-1
Physico-
chemical
properties


X




X
X

X
X
X
X



X
X





Selected
toxic ity
reference
value
X




X




X













Reference
1,4
1, 2, 3, 4, 8
4
1,4
1,2,4
8
2
2,5
1, 2, 3, 4
1, 2, 3, 4, 5,
6,7,8
1,4
1
1,4
1,4
4,8
6
2
2
1, 2, 3, 4, 5,
6,8
8
2, 3, 4, 8
8
                This document is a draft for review purposes only and does not constitute Agency policy.
June 2015                                         A-37                      DRAFT—DO NOT CITE OR QUOTE

-------
Hydraulic Fracturing Drinking Water Assessment
Appendix A
Chemical name
Quaternary ammonium compounds,
benzylbis(hydrogenated tallow alkyl)methyl,
salts with bentonite
Quaternary ammonium compounds, benzyl-
C10-16-alkyldimethyl, chlorides
Quaternary ammonium compounds, benzyl-
C12-16-alkyldimethyl, chlorides
Quaternary ammonium compounds, benzyl-
C12-18-alkyldimethyl, chlorides
Quaternary ammonium compounds,
bis(hydrogenated tallow alkyl)dimethyl, salts
with bentonite
Quaternary ammonium compounds,
bis(hydrogenated tallow alkyl)dimethyl, salts
with hectorite
Quaternary ammonium compounds, di-C8-10-
alkyldimethyl, chlorides
Quaternary ammonium compounds, dicoco
alkyldimethyl, chlorides
Quaternary ammonium compounds,
pentamethyltallowalkyltrimethylenedi-,
dichlorides
Quaternary ammonium compounds,
trimethyltallow alkyl, chlorides
Quinaldine
Quinoline
Raffinates (petroleum)
Raffinates, petroleum, sorption process
Residual oils, petroleum, solvent-refined
Residues, petroleum, catalytic reformer
fractionator
Rhodamine B
Rosin
Rutile titanium dioxide
Sand
CASRN
68153-30-0
68989-00-4
68424-85-1
68391-01-5
68953-58-2
71011-27-3
68424-95-3
61789-77-3
68607-29-4
8030-78-2
91-63-4
91-22-5
68514-29-4
64741-85-1
64742-01-4
64741-67-9
81-88-9
8050-09-7
1317-80-2
308075-07-2
Physico-
chemical
properties






X



X
X




X



Selected
toxic ity
reference
value


X








X








Reference
2,5,6
1,4
1, 2, 4, 8
8
2, 3, 4, 8
2
2
1
4
1,4
8
2,4
5
1, 2, 4, 8
5
1,4,8
4
1,4
8
8
                This document is a draft for review purposes only and does not constitute Agency policy.
June 2015                                         A-38                     DRAFT—DO NOT CITE OR QUOTE

-------
Hydraulic Fracturing Drinking Water Assessment
Appendix A
Chemical name
Scandium oxide
Sepiolite
Silane, dichlorodimethyl-, reaction products
with silica
Silica
silica gel, cryst. -free
Silica, amorphous, fumed, cryst.-free
Silica, vitreous
Silicic acid, aluminum potassium sodium salt
Siloxanes (Polysiloxane)
Siloxanes and Silicones, di-Me, 3-hydroxypropyl
Me, ethoxylated propoxylated
Siloxanes and Silicones, di-Me, Me hydrogen
Siloxanes and silicones, di-Me, polymers with
Me silsesquioxanes
Siloxanes and Silicones, di-Me, reaction
products with silica
Siloxanes and silicones, dimethyl,
Silwet 177
Sodium 1-octanesulfonate
Sodium 2-mercaptobenzothiolate
Sodium acetate
Sodium aluminate
Sodium benzoate
Sodium bicarbonate
Sodium bis(tridecyl) sulfobutanedioate
Sodium bisulfite
Sodium borate
Sodium bromate
Sodium bromide
Sodium bromosulfamate
CASRN
12060-08-1
63800-37-3
68611-44-9
7631-86-9
112926-00-8
112945-52-5
60676-86-0
12736-96-8
9011-19-2
68937-55-3
68037-59-2
68037-74-1
67762-90-7
63148-52-7
27306-78-1
5324-84-5
2492-26-4
127-09-3
1302-42-7
532-32-1
144-55-8
2673-22-5
7631-90-5
1333-73-9
7789-38-0
7647-15-6
1004542-84-0
Physico-
chemical
properties















X
X
X

X
X
X





Selected
toxic ity
reference
value



























Reference
8
2
2,4
1, 2, 3, 4, 8
3,4
1,3,4
1,4,8
4
4
8
8
4
4
4
1
3
2
1,3,4
2,4
3
1, 2, 3, 4, 7
4
1,3,4
1, 4, 6, 7
1,2,4
1, 2, 3, 4, 7
8
                This document is a draft for review purposes only and does not constitute Agency policy.
June 2015                                         A-39                     DRAFT—DO NOT CITE OR QUOTE

-------
Hydraulic Fracturing Drinking Water Assessment
Appendix A
Chemical name
Sodium C14-16 alpha-olefin sulfonate
Sodium caprylamphopropionate
Sodium carbonate
Sodium chlorate
Sodium chloride

Sodium chlorite

Sodium chloroacetate
Sodium cocaminopropionate
Sodium decyl sulfate
Sodium D-gluconate
Sodium diacetate
Sodium dichloroisocyanurate
Sodium dl-lactate
Sodium dodecyl sulfate
Sodium erythorbate (1:1)
Sodium ethasulfate
Sodium formate
Sodium hydrogen sulfate
Sodium hydroxide
Sodium hydroxymethanesulfonate
Sodium hypochlorite
Sodium iodide
Sodium ligninsulfonate
Sodium l-lactate
Sodium maleate (l:x)
Sodium metabisulfite
Sodium metaborate
Sodium metaborate dihydrate
Sodium metaborate tetrahydrate
Sodium metasilicate
CASRN
68439-57-6
68610-44-6
497-19-8
7775-09-9
7647-14-5

7758-19-2

3926-62-3
68608-68-4
142-87-0
527-07-1
126-96-5
2893-78-9
72-17-3
151-21-3
6381-77-7
126-92-1
141-53-7
7681-38-1
1310-73-2
870-72-4
7681-52-9
7681-82-5
8061-51-6
867-56-1
18016-19-8
7681-57-4
7775-19-1
16800-11-6
10555-76-7
6834-92-0
Physico-
chemical
properties
X
X
X





X

X
X
X
X
X
X
X
X
X


X



X
X





Selected
toxic ity
reference
value



X


X
















X








Reference
1,3,4
4
1, 2, 3, 4, 8
1,4
1, 2, 3, 4, 5, 8

1, 2, 3, 4, 5, 8

3
1
1
4
1,4
2
8
8
1, 3, 4, 8
1
2,8
4
1, 2, 3, 4, 7, 8
8
1, 2, 3, 4, 8
4
2
8
8
1
3,4
1,4
1,4,8
1,2,4
                This document is a draft for review purposes only and does not constitute Agency policy.
June 2015                                         A-40                     DRAFT—DO NOT CITE OR QUOTE

-------
Hydraulic Fracturing Drinking Water Assessment
Appendix A
Chemical name
Sodium molybdate(VI)
Sodium nitrate
Sodium nitrite
Sodium N-methyl-N-oleoyltaurate
Sodium octyl sulfate
Sodium oxide
Sodium perborate
Sodium perborate tetrahydrate
Sodium peroxoborate
Sodium persulfate

Sodium phosphate
Sodium polyacrylate
Sodium pyrophosphate
Sodium salicylate
Sodium sesquicarbonate
Sodium silicate
Sodium starch glycolate
Sodium sulfate
Sodium sulfite
Sodium thiocyanate
Sodium thiosulfate
Sodium thiosulfate, pentahydrate
Sodium trichloroacetate
Sodium trimetaphosphate
Sodium xylenesulfonate
Sodium zirconium lactate
Sodium zirconium lactic acid (4:4:1)
Solvent naphtha, petroleum, heavy aliph.
Solvent naphtha, petroleum, heavy arom.
Solvent naphtha, petroleum, light aliph.
CASRN
7631-95-0
7631-99-4
7632-00-0
137-20-2
142-31-4
1313-59-3
11138-47-9
10486-00-7
7632-04-4
7775-27-1

7632-05-5
9003-04-7
7758-16-9
54-21-7
533-96-0
1344-09-8
9063-38-1
7757-82-6
7757-83-7
540-72-7
7772-98-7
10102-17-7
650-51-1
7785-84-4
1300-72-7
15529-67-6
10377-98-7
64742-96-7
64742-94-5
64742-89-8
Physico-
chemical
properties



X
X









X
X




X


X

X





Selected
toxic ity
reference
value













X










X






Reference
8
2
1,2,4
4
1
1
4
1, 4, 5, 8
1
1, 2, 3, 4, 7, 8

1,4
1, 2, 3, 4
1,2,4
1,4
1,2
1,2,4
2
1, 2, 3, 4
2,4,8
1,4
1, 2, 3, 4
1,4
1,4
8
1,3,4
8
1,4
2,4,8
1, 2, 4, 5, 8
8
                This document is a draft for review purposes only and does not constitute Agency policy.
June 2015                                         A-41                      DRAFT—DO NOT CITE OR QUOTE

-------
Hydraulic Fracturing Drinking Water Assessment
Appendix A
Chemical name
Solvent naphtha, petroleum, light arom.
Sorbic acid
Sorbitan sesquioleate
Sorbitan, mono-(9Z)-9-octadecenoate
Sorbitan, monooctadecanoate
Sorbitan, tri-(9Z)-9-octadecenoate
Spirit of ammonia, aromatic
Stannous chloride dihydrate
Starch
Steam cracked distillate, cyclodiene dimer,
dicyclopentadiene polymer
Stoddard solvent
Stoddard solvent IIC
Strontium chloride
Styrene
Subtilisin
Sucrose
Sulfamic acid
Sulfan blue
Sulfate
Sulfo NHS Biotin
Sulfomethylated quebracho
Sulfonic acids, C10-16-alkane, sodium salts
Sulfonic acids, petroleum
Sulfonic acids, petroleum, sodium salts
Sulfur dioxide
Sulfuric acid
Sulfuric acid, mono-C12-18-alkyl esters, sodium
salts
Sulfuric acid, mono-C6-10-alkyl esters,
ammonium salts
CASRN
64742-95-6
110-44-1
8007-43-0
1338-43-8
1338-41-6
26266-58-0
8013-59-0
10025-69-1
9005-25-8
68131-87-3
8052-41-3
64742-88-7
10476-85-4
100-42-5
9014-01-1
57-50-1
5329-14-6
129-17-9
14808-79-8
119616-38-5
68201-64-9
68608-21-9
61789-85-3
68608-26-4
7446-09-5
7664-93-9
68955-19-1
68187-17-7
Physico-
chemical
properties

X
X
X
X
X







X

X

X








X
X
Selected
toxic ity
reference
value












X
X














Reference
1,2,4
8
4
1, 2, 3, 4
8
8
8
1,4
1,2,4
1
1,3,4
1,2,4
4
2
8
1, 2, 3, 4
1,4
8
1,4
8
2
6
1
3
2,4,8
1, 2, 4, 7
4
1,4,8
                This document is a draft for review purposes only and does not constitute Agency policy.
June 2015                                         A-42                      DRAFT—DO NOT CITE OR QUOTE

-------
Hydraulic Fracturing Drinking Water Assessment
Appendix A
Chemical name
Symclosene
Talc
Tall oil
Tall oil imidazoline
Tall oil, compound with diethanolamine
Tall oil, ethoxylated
Tall-oil pitch
Tallow alkyl amines acetate
Tar bases, quinoline derivatives, benzyl
chloride-quaternized
Tegin M
Terpenes and Terpenoids, sweet orange-oil
Terpineol
tert-Butyl hydroperoxide
tert-Butyl perbenzoate
Tetra-calcium-alumino-ferrite
Tetradecane
Tetradecyldimethylbenzylammonium chloride
Tetraethylene glycol
Tetraethylenepentamine
Tetrakis(hydroxymethyl)phosphonium sulfate
Tetramethyl orthosilicate
Tetramethylammonium chloride

Tetrasodium pyrophosphate
Thiamine hydrochloride
Thiocyanic acid, ammonium salt
Thioglycolic acid
Thiourea
Thiourea, polymer with formaldehyde and 1-
phenylethanone
Thuja plicata donn ex. D. don leaf oil
CASRN
87-90-1
14807-96-6
8002-26-4
61791-36-4
68092-28-4
65071-95-6
8016-81-7
61790-60-1
72480-70-7
8043-29-6
68647-72-3
8000-41-7
75-91-2
614-45-9
12068-35-8
629-59-4
139-08-2
112-60-7
112-57-2
55566-30-8
681-84-5
75-57-0

7722-88-5
67-03-8
1762-95-4
68-11-1
62-56-6
68527-49-1
68917-35-1
Physico-
chemical
properties
X











X
X

X
X
X
X
X

X


X
X
X
X


Selected
toxic ity
reference
value























X



X


Reference
2
1, 3, 4, 6, 7
4,8
4
1
4,8
4
8
1,3,4
8
1, 3, 4, 8
1,3
1,4
1
1,2,4
8
1,4,8
1,4
1,4
1, 2, 3, 4, 7
1
1, 2, 3, 4, 7, 8

8
8
2,3,4
1, 2, 3, 4
1, 2, 3, 4, 6
1,4,8
3
                This document is a draft for review purposes only and does not constitute Agency policy.
June 2015                                         A-43                      DRAFT—DO NOT CITE OR QUOTE

-------
Hydraulic Fracturing Drinking Water Assessment
Appendix A
Chemical name
Tin(ll) chloride
Titanium dioxide
Titanium(4+) 2-[bis(2-
hydroxyethyl)amino]ethanolate propan-2-olate
(1:2:2)
Titanium, isopropoxy (triethanolaminate)
Toluene
Tributyl phosphate
Tributyltetradecylphosphonium chloride
Tricalcium phosphate
Tricalcium silicate
Tridecane
Triethanolamine
Triethanolamine hydrochloride
Triethanolamine hydroxyacetate
Triethanolamine polyphosphate ester
Triethyl citrate
Triethyl phosphate
Triethylene glycol
Triethylenetetramine
Triisopropanolamine
Trimethanolamine
Trimethyl borate
Trimethylamine
Trimethylamine quaternized
polyepichlorohydrin
Trimethylbenzene
Triphosphoric acid, pentasodium salt
Tripoli
Tripotassium citrate monohydrate
Tripropylene glycol monomethyl ether
CASRN
7772-99-8
13463-67-7
36673-16-2
74665-17-1
108-88-3
126-73-8
81741-28-8
7758-87-4
12168-85-3
629-50-5
102-71-6
637-39-8
68299-02-5
68131-71-5
77-93-0
78-40-0
112-27-6
112-24-3
122-20-3
14002-32-5
121-43-7
75-50-3
51838-31-4

25551-13-7
7758-29-4
1317-95-9
6100-05-6
25498-49-1
Physico-
chemical
properties




X
X
X


X
X
X
X

X
X
X
X
X
X

X





X
X
Selected
toxic ity
reference
value




X
X

X

















X



Reference
1
1,2,4
1
1,4
1,3,4
1,2,4
1,3,4
1,4
1,2,4
8
1,2,4
8
3
1,4,8
1,4
1,4
1,2,3
4
1,4
3
8
8
1, 2, 3, 4, 5, 8

1,2,4
1,4
4
4
2
                This document is a draft for review purposes only and does not constitute Agency policy.
June 2015                                         A-44                     DRAFT—DO NOT CITE OR QUOTE

-------
Hydraulic Fracturing Drinking Water Assessment
Appendix A
Chemical name
Trisodium citrate
Trisodium citrate dihydrate
Trisodium ethylenediaminetetraacetate
Trisodium ethylenediaminetriacetate
Trisodium phosphate
Trisodium phosphate dodecahydrate
Tritan R (X-100)
Triton X-100
Tromethamine
Tryptone
Ulexite
Undecane
Undecanol, branched and linear
Urea
Vermiculite
Vinyl acetate ethylene copolymer
Vinylidene chloride/methylacrylate copolymer
Water
White mineral oil, petroleum
Xylenes
Yeast extract
Zeolites
Zinc
Zinc carbonate
Zinc chloride
Zinc oxide
Zinc sulfate monohydrate
Zirconium nitrate
Zirconium oxide sulfate
Zirconium oxychloride
CASRN
68-04-2
6132-04-3
150-38-9
19019-43-3
7601-54-9
10101-89-0
92046-34-9
9002-93-1
77-86-1
73049-73-7
1319-33-1
1120-21-4
128973-77-3
57-13-6
1318-00-9
24937-78-8
25038-72-6
7732-18-5
8042-47-5
1330-20-7
8013-01-2
1318-02-1
7440-66-6
3486-35-9
7646-85-7
1314-13-2
7446-19-7
13746-89-9
62010-10-0
7699-43-6
Physico-
chemical
properties
X
X
X
X




X


X

X





X










Selected
toxic ity
reference
value




X














X


X







Reference
3
1,4
1,3
1,4,8
1,2,4
1
8
1,3,4
3,4
8
1, 2, 3, 8
3,8
8
1, 2, 4, 8
2
1,4
4
2,4,8
1,2,4
1,2,4
8
8
2
2
1,2
1,4
8
2,6
1,4
1,2,4
                This document is a draft for review purposes only and does not constitute Agency policy.
June 2015                                         A-45                     DRAFT—DO NOT CITE OR QUOTE

-------
Hydraulic Fracturing Drinking Water Assessment
Appendix A
Chemical name
Zirconium(IV) chloride tetrahydrofuran complex
Zirconium(IV) sulfate
Zirconium, l,l'-((2-((2-hydroxyethyl)(2-
hydroxypropyl)amino)ethyl)imino)bis(2-
propanol) complexes
Zirconium, acetate lactate oxo ammonium
complexes
Zirconium, chloro hydroxy lactate oxo sodium
complexes
Zirconium, hydroxylactate sodium complexes
Zirconium,tetrakis[2-[bis(2-
hydroxyethyl)amino-kN]ethanolato-kO]-
CASRN
21959-01-3
14644-61-2
197980-53-3
68909-34-2
174206-15-6
113184-20-6
101033-44-7
Physico-
chemical
properties







Selected
toxic ity
reference
value







Reference
5
2,6
4
4,8
4
1,4
1, 2, 4, 5
 Table A-3. List of generic names of chemicals reportedly used in hydraulic fracturing fluids.

           In some cases, the generic chemical name masks a specific chemical name and CASRN provided to the
           EPA and claimed as CBI by one or more of the nine hydraulic fracturing service companies.
Generic chemical name
2-Substituted aromatic amine salt
Acetylenic alcohol
Acrylamide acrylate copolymer
Acrylamide copolymer
Acrylamide modified polymer
Acrylamide-sodium acrylate copolymer
Acrylate copolymer
Acrylic copolymer
Acrylic polymer
Acrylic resin
Acyclic hydrocarbon blend
Acylbenzylpyridinium choride
Alcohol alkoxylate
Alcohol and fatty acid blend
Reference
1,4
1
4
1,4
4
4
1
1
1,4
4
1,4
8
1,4
2
               This document is a draft for review purposes only and does not constitute Agency policy.

June 2015                                       A-46                   DRAFT—DO NOT CITE OR QUOTE

-------
Hydraulic Fracturing Drinking Water Assessment
Appendix A
Generic chemical name
Alcohol ethoxylates
Alcohols
Alcohols, C9-C22
Aldehydes
Alfa-alumina
Aliphatic acids
Aliphatic alcohol
Aliphatic alcohol glycol ether
Aliphatic alcohols, ethoxylated
Aliphatic amine derivative
Aliphatic carboxylic acid
Alkaline bromide salts
Alkaline metal oxide
Alkanes/alkenes
Alkanolamine derivative
Alkanolamine/aldehyde condensate
Alkenes
Alklaryl sulfonic acid
Alkoxylated alcohols
Alkoxylated amines
Alkyaryl sulfonate
Alkyl alkoxylate
Alkyl amide
Alkyl amine
Alkyl amine blend in a metal salt solution
Alkyl aryl amine sulfonate
Alkyl aryl polyethoxy ethanol
Alkyl dimethyl benzyl ammonium chloride
Alkyl esters
Alkyl ether phosphate
Alkyl hexanol
Alkyl ortho phosphate ester
Reference
4
1,4
1,4
1,4,5
1,4
1, 2, 3, 4
2
3,4
2
1
4
1,4
4
4
2
1,2,4
1,4
1,4
1
1,4
1, 2, 3, 4
1,4
4
1,4
1,4
4
3,4
4
1,4
4
1,4
1,4
                This document is a draft for review purposes only and does not constitute Agency policy.
June 2015                                         A-47                     DRAFT—DO NOT CITE OR QUOTE

-------
Hydraulic Fracturing Drinking Water Assessment
Appendix A
Generic chemical name
Alkyl phosphate ester
Alkyl phosphonate
Alkyl pyridines
Alkyl quaternary ammonium chlorides
Alkyl quaternary ammonium salt
Alkylamine alkylaryl sulfonate
Alkylamine salts
Alkylaryl sulfonate
Alkylated quaternary chloride
Alkylated sodium naphthalenesulphonate
Alkylbenzenesulfonate
Alkylbenzenesulfonic acid
Alkylethoammonium sulfates
Alkylphenol ethoxylates
Alkylpyridinium quaternary
Alphatic alcohol polyglycol ether
Aluminum oxide
Amide
Amidoamine
Amine
Amine compound
Amine oxides
Amine phosphonate
Amine salt
Amino compounds
Amino methylene phosphonic acid salt
Ammonium alcohol ether sulfate
Ammonium salt
Ammonium salt of ethoxylated alcohol sulfate
Amorphous silica
Amphoteric surfactant
Anionic acrylic polymer
Reference
1,4
4
2
1,4
4
4
2
1,4
1,2,4
2
2
1,4,5
1
1,4
4
2
1,4
4
1,4
1,4
4
1,4
1,4
1
1,4
1,4
1,4
1,4
1,4
4
2
2
                This document is a draft for review purposes only and does not constitute Agency policy.
June 2015                                         A-48                     DRAFT—DO NOT CITE OR QUOTE

-------
Hydraulic Fracturing Drinking Water Assessment
Appendix A
Generic chemical name
Anionic copolymer
Anionic polyacrylamide
Anionic polyacrylamide copolymer
Anionic polymer
Anionic surfactants
Antifoulant
Antimonate salt
Aqueous emulsion of diethylpolysiloxane
Aromatic alcohol glycol ether
Aromatic aldehyde
Aromatic hydrocarbons
Aromatic ketones
Aromatic polyglycol ether
Arsenic compounds
Ashes, residues
Bentone clay
Biocide
Biocide component
Bis-quaternary methacrylamide monomer
Blastfurnace slag
Borate salts
Cadmium compounds
Carbohydrates
Carboxylmethyl hydroxypropyl guar
Cationic polyacrylamide
Cationic polymer
Cedar fiber, processed
Cellulase enzyme
Cellulose derivative
Cellulose ether
Cellulosic polymer
Ceramic
Reference
1,4
1,2,4
1,4,6
1,3,4
2,4,6
1,4
1,4
2
1
1,4
3,4
1, 2, 3, 4
1
4
4
4
4
1,4
4
4
1,2,4
4
1,2,4
4
4
2,4
2
1
1,2,4
2
2
4
                This document is a draft for review purposes only and does not constitute Agency policy.
June 2015                                         A-49                     DRAFT—DO NOT CITE OR QUOTE

-------
Hydraulic Fracturing Drinking Water Assessment
Appendix A
Generic chemical name
Chlorous ion solution
Chromates
Chrome-free lignosulfonate compound
Citrus rutaceae extract
Common white
Complex alkylaryl polyo-ester
Complex aluminum salt
Complex carbohydrate
Complex organometallic salt
Complex polyamine salt
Complex substituted keto-amine
Complex substituted keto-amine hydrochloride
Copper compounds
Coric oxide
Cotton dust (raw)
Cottonseed hulls
Cured acrylic resin
Cured resin
Cured urethane resin
Cyclic alkanes
Defoamer
Dibasic ester
Dicarboxylic acid
Diesel
Dimethyl silicone
Dispersing agent
Emulsifier
Enzyme
Epoxy
Epoxy resin
Essential oils
Ester Salt
Reference
1
1,4
2
4
4
1
1,4
2
1
7
1
1
6
4
2
2
1,4
1,4,5
1,4
1,4
4
4
1,4
1,4,6
1,4
1
4
4
4
1,4
1,4
2,4
                This document is a draft for review purposes only and does not constitute Agency policy.
June 2015                                         A-50                     DRAFT—DO NOT CITE OR QUOTE

-------
Hydraulic Fracturing Drinking Water Assessment
Appendix A
Generic chemical name
Esters
Ether compound
Ether salt
Ethoxylated alcohol blend
Ethoxylated alcohol/ester mixture
Ethoxylated alcohols
Ethoxylated alkyl amines
Ethoxylated amine blend
Ethoxylated amines
Ethoxylated fatty acid
Ethoxylated fatty acid ester
Ethoxylated nonionic surfactant
Ethoxylated nonylphenol
Ethoxylated sorbitol esters
Ethylene oxide-nonylphenol polymer
Fatty acid amine salt mixture
Fatty acid ester
Fatty acid tall oil
Fatty acid, ethoxylate
Fatty acids
Fatty alcohol alkoxylate
Fatty alkyl amine salt
Fatty amine carboxylates
Fatty imidazoline
Fluoroaliphatic polymeric esters
Formaldehyde polymer
Glass fiber
Glyceride esters
Glycol
Glycol blend
Glycol ethers
Ground cedar
Reference
2,4
4
4
4
4
1, 2, 4, 5, 7
1,4
4
1,4
4
1,4
1,4
1,2,4
1,4
4
4
1,2,4
1,4
4
1
1,4
1,4
1,4
4
1,4
1
1,4
2
4
2
1,4,7
2
                This document is a draft for review purposes only and does not constitute Agency policy.
June 2015                                         A-51                      DRAFT—DO NOT CITE OR QUOTE

-------
Hydraulic Fracturing Drinking Water Assessment
Appendix A
Generic chemical name
Ground paper
Guar derivative
Guar gum
Haloalkyl heteropolycycle salt
Hexanes
High molecular weight polymer
High pH conventional enzymes
Hydrocarbons
Hydrogen solvent
Hydrotreated and hydrocracked base oil
Hydrotreated distillate, light C9-16
Hydrotreated heavy naphthalene
Hydrotreated light distillate
Hydrotreated light petroleum distillate
Hydroxyalkyl imino carboxylic sodium salt
Hydroxycellulose
Hydroxyethyl cellulose
Imidazolium compound
Inner salt of alkyl amines
Inorganic borate
Inorganic chemical
Inorganic particulate
Inorganic salt
Iso-alkanes/n-alkanes
Isomeric aromatic ammonium salt
Latex
Lead compounds
Low toxicity base oils
Lubra-Beads course
Maghemite
Magnetite
Metal salt
Reference
2
1,4
4
1,4
1
2
2
1
4
1,4
4
5
2,4
4
2
6
1,2,4
4
1,4
1,4
4
1,4
2,4
1,4
1,4
2,4
4
1,4
4
1,4
1,4
1
                This document is a draft for review purposes only and does not constitute Agency policy.
June 2015                                         A-52                      DRAFT—DO NOT CITE OR QUOTE

-------
Hydraulic Fracturing Drinking Water Assessment
Appendix A
Generic chemical name
Metal salt solution
Mineral
Mineral fiber
Mineral filler
Mineral oil
Mixed titanium ortho ester complexes
Modified acrylamide copolymer
Modified acrylate polymer
Modified alkane
Modified bentonite
Modified cycloaliphatic amine adduct
Modified lignosulfonate
Naphthalene derivatives
Neutralized alkylated napthalene sulfonate
Nickel chelate catalyst
Nonionic surfactant
N-tallowalkyltrimethylenediamines
Nuisance particulates
Nylon
Olefinic sulfonate
Olefins
Organic acid salt
Organic acids
Organic alkyl amines
Organic chloride
Organic modified bentonite clay
Organic phosphonate
Organic phosphonate salts
Organic phosphonic acid salts
Organic polymer
Organic polyol
Organic salt
Reference
1
1,4
2
1
4
1,4
2,4
4
1,4
4
1,4
2,4
1,4
4
4
1
4
1,2,4
4
1,4
1,4
1,4
1,4
4
4
4
1,4
1,4
1,4
4
4
1,4
                This document is a draft for review purposes only and does not constitute Agency policy.
June 2015                                         A-53                      DRAFT—DO NOT CITE OR QUOTE

-------
Hydraulic Fracturing Drinking Water Assessment
Appendix A
Generic chemical name
Organic sulfur compound
Organic surfactants
Organic titanate
Organo amino silane
Organo phosphonic acid
Organo phosphonic acid salt
Organometallic ammonium complex
Organophilicclay
Oxidized tall oil
Oxoaliphaticacid
Oxyalkylated alcohol
Oxyalkylated alkyl alcohol
Oxyalkylated alkylphenol
Oxyalkylated fatty acid
Oxyalkylated fatty alcohol salt
Oxyalkylated phenol
Oxyalkylated phenolic resin
Oxyalkylated polyamine
Oxyalkylated tallow diamine
Oxyethylated alcohol
Oxylated alcohol
P/F resin
Paraffin inhibitor
Paraffinic naphthenic solvent
Paraffinic solvent
Paraffins
Pecan shell
Petroleum distallate blend
Petroleum gas oils
Petroleum hydrocarbons
Petroleum solvent
Phosphate ester
Reference
1,4
1
1,4
4
4
4
1
4
2
2
1,4
2,4
1, 2, 3, 4
1,4
2
1,4
4
1
2
2
1,4
4
4
1
1,4
1
2
2,3,4
1
4
2
1,4
                77]/s document is a draft for review purposes only and does not constitute Agency policy.
June 2015                                          A-54                     DRAFT—DO NOT CITE OR QUOTE

-------
Hydraulic Fracturing Drinking Water Assessment
Appendix A
Generic chemical name
Phosphonate
Phosphonic acid
Phosphoric acid, mixed polyoxyalkylene aryl and alkyl esters
Plasticizer
Polyacrylamide copolymer
Polyacrylamides
Polyacrylate
Polyactide resin
Polyalkylene esters
Polyaminated fatty acid
Polyaminated fatty acid surfactants
Polyamine
Polyamine polymer
Polyanionic cellulose
Polyaromatic hydrocarbons
Polycyclic organic matter
Polyelectrolyte
Polyether polyol
Polyethoxylated alkanol
Polyethylene copolymer
Polyethylene glycols
Polyethylene wax
Polyglycerols
Polyglycol
Polyglycol ether
Polylactide resin
Polymer
Polymeric hydrocarbons
Polymerized alcohol
Polymethacrylate polymer
Polyol phosphate ester
Polyoxyalkylene phosphate
Reference
2
1,4
4
1,2
4
1
1,4
4
4
2
2
1,4
4
1
6
6
4
2
2,3,4
4
4
4
2
2
6
4
2,4
3,4
4
4
2
2
                This document is a draft for review purposes only and does not constitute Agency policy.
June 2015                                         A-55                     DRAFT—DO NOT CITE OR QUOTE

-------
Hydraulic Fracturing Drinking Water Assessment
Appendix A
Generic chemical name
Polyoxyalkylene sulfate
Polyoxyalkylenes
Polyphenylene ether
Polyphosphate
Polypropylene glycols
Polyquaternary amine
Polysaccaride polymers in suspension
Polysaccharide
Polysaccharide blend
Polyvinylalcohol/polyvinylactetate copolymer
Potassium chloride substitute
Quarternized heterocyclic amines
Quaternary amine
Quaternary amine salt
Quaternary ammonium chloride
Quaternary ammonium compound
Quaternary ammonium salts
Quaternary compound
Quaternary salt
Quaternized alkyl nitrogenated compd
Red dye
Refined mineral oil
Resin
Salt of amine-carbonyl condensate
Salt of fatty acid/polyamine reaction product
Salt of phosphate ester
Salt of phosphono-methylated diamine
Salts
Salts of oxyalkylated fatty amines
Sand
Sand, AZ silica
Sand, brown
Reference
2
1,4,7
4
4
2
4
2
4
4
4
4
4
2,4
4
4
1,2,4
1,2,4
1,4
1,4
4
4
2
4
3,4
3,4
1
1,4
4
4
4
4
4
                77]/s document is a draft for review purposes only and does not constitute Agency policy.
June 2015                                          A-56                     DRAFT—DO NOT CITE OR QUOTE

-------
Hydraulic Fracturing Drinking Water Assessment
Appendix A
Generic chemical name
Sand, sacked
Sand, white
Secondary alcohol
Silica sand, 100 mesh, sacked
Silicone emulsion
Silicone ester
Sodium acid pyrophosphate
Sodium calcium magnesium polyphosphate
Sodium phosphate
Sodium salt of aliphatic amine acid
Sodium xylene sulfonate
Softwood dust
Starch blends
Substituted alcohol
Substituted alkene
Substituted alklyamine
Substituted alkyne
Sulfate
Sulfomethylated tannin
Sulfonate
Sulfonate acids
Sulfonate surfactants
Sulfonated asphalt
Sulfonicacid salts
Sulfur compound
Sulphonic amphoterics
Sulphonic amphoterics blend
Surfactant blend
Surfactants
Synthetic copolymer
Synthetic polymer
Tallow soap
Reference
4
4
1,4
4
1
4
4
4
4
2
4
2
6
1,2,4
1
1,4
4
4
2,5
4
1
1
2
1,4
1,4
4
4
3,4
1,2,4
2
4
4
                77]/s document is a draft for review purposes only and does not constitute Agency policy.
June 2015                                          A-57                     DRAFT—DO NOT CITE OR QUOTE

-------
Hydraulic Fracturing Drinking Water Assessment
Appendix A
Generic chemical name
Telomer
Terpenes
Titanium complex
Triethanolamine zirconium chelate
Triterpanes
Vanadium compounds
Wall material
Walnut hulls
Zirconium complex
Zirconium salt
Reference
4
1,4
4
14
4
4
1
1,2,4
2,4
4
 Table A-4. Chemicals detected in flowback or produced water.

           An "X" indicates the availability of physicochemical properties from EPI Suite™ and selected toxicity
           reference values (see Appendix G). An empty cell indicates no information was available from the
           sources we consulted. Reference number corresponds to the citations in Table A-l. Italicized
           chemicals are found in both fracturing fluids and flowback/produced water.
Chemical name
1,2,3-Trichlorobenzene
1,2,4-Trichlorobenzene
1, 2, 4- Trimethylbenzen e
1,2-Propylene glycol
1, 3, 5- Trimethylbenzen e
1,4-Dioxane
2,4-Dimethylphenol
2,6-Dichlorophenol
2-Methylnaphthalene
2-Methylpropanoic acid
2-Methylpyridine
7,12-Dimethylbenz(a)anthracene
Acetic acid
CASRN
87-61-6
120-82-1
95-63-6
57-55-6
108-67-8
123-91-1
105-67-9
87-65-0
91-57-6
79-31-2
109-06-8
57-97-6
64-19-7
Physico-
chemical
properties
X
X
X
X
X
X
X
X
X
X
X
X
X
Selected
toxicity
reference
value
X
X

X

X
X

X


X

Reference
3,9
9
3, 9, 10
3,9
3, 9, 10
9,10
3, 9, 10
3,9
3, 9, 10
10
3,9
3,9
3, 9, 10
               This document is a draft for review purposes only and does not constitute Agency policy.

June 2015                                        A-58                    DRAFT—DO NOT CITE OR QUOTE

-------
Hydraulic Fracturing Drinking Water Assessment
Appendix A
Chemical name
Acetone
Acetophenone
Acrolein
Acrylonitrile
Aldrin
Aluminum
Ammonia
Antimony
Aroclor 1248
Arsenic
Barium
Benzene
Benzo(a)pyrene
Benzo(b)fluoranthene
Benzo(g,h,i)perylene
Benzo(k)fluoranthene
Benzyl alcohol
Beryllium
beta-Hexachlorocyclohexane
Bis(2-chloroethyl) ether
Boron
Bromide
Bromodichloromethane
Bromoform
Butanoicacid
Butylbenzene
Cadmium
Caesium-137
Calcium
Carbon dioxide
CASRN
67-64-1
98-86-2
107-02-8
107-13-1
309-00-2
7429-90-5
7664-41-7
7440-36-0
12672-29-6
7440-38-2
7440-39-3
71-43-2
50-32-8
205-99-2
191-24-2
207-08-9
100-51-6
7440-41-7
319-85-7
111-44-4
7440-42-8
24959-67-9
75-27-4
75-25-2
107-92-6
104-51-8
7440-43-9
10045-97-3
7440-70-2
124-38-9
Physico-
chemical
properties
X
X
X
X
X



X


X
X
X
X
X
X

X
X


X
X
X
X



X
Selected
toxic ity
reference
value
X
X
X
X
X
X

X

X
X
X
X
X

X
X
X
X
X
X

X
X

X
X



Reference
3, 9, 10
3,9
9
3,9
3,9
3, 9, 10
3, 9, 10
3, 9, 10
3,9
3, 9, 10
3, 9, 10
3, 9, 10
3,9
3,9
3, 9, 10
3,9
3, 9, 10
3, 9, 10
3,9
3,9
3, 9, 10
3, 9, 10
3
3, 9, 10
9,10
9,10
3, 9, 10
3
3, 9, 10
3, 9, 10
                This document is a draft for review purposes only and does not constitute Agency policy.
June 2015                                         A-59                     DRAFT—DO NOT CITE OR QUOTE

-------
Hydraulic Fracturing Drinking Water Assessment
Appendix A
Chemical name
Carbon disulfide
Chloride
Chlorine
Chlorodibromomethane
Chloroform
Chloromethane
Chromium
Chromium (III)
Chromium (VI)
Cobalt
Copper
Cumene
Cyanide
delta-Hexachlorocyclohexane
Di(2-ethylhexyl) phthalate
Dibenz(a,h)anthracene
Dibutyl phthalate
Dichloromethane
Dieldrin
Diethyl phthalate
Dioctyl phthalate
Diphenylamine
Endosulfan 1
Endosulfan II
Endrin aldehyde
E thy 1 benzene
Ethylene glycol
Fluoranthene
Fluorene
Fluoride
CASRN
75-15-0
16887-00-6
7782-50-5
124-48-1
67-66-3
74-87-3
7440-47-3
16065-83-1
18540-29-9
7440-48-4
7440-50-8
98-82-8
57-12-5
319-86-8
117-81-7
53-70-3
84-74-2
75-09-2
60-57-1
84-66-2
117-84-0
122-39-4
959-98-8
33213-65-9
7421-93-4
100-41-4
107-21-1
206-44-0
86-73-7
16984-48-8
Physico-
chemical
properties
X


X
X
X





X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X

Selected
toxic ity
reference
value
X

X
X
X


X
X
X
X
X
X

X
X
X
X
X
X
X
X



X
X
X
X

Reference
3,9
3, 9, 10
3,10
3
3, 9, 10
3,10
3, 9, 10
3
3,10
3, 9, 10
3, 9, 10
3,9
3, 9, 10
9
3, 9, 10
3,9
3, 9, 10
9,10
9
9
9,10
3,9
3,9
3,9
3,9
3, 9, 10
3,9
3,9
3, 9, 10
3, 9, 10
                This document is a draft for review purposes only and does not constitute Agency policy.
June 2015                                         A-60                     DRAFT—DO NOT CITE OR QUOTE

-------
Hydraulic Fracturing Drinking Water Assessment
Appendix A
Chemical name
Formic acid
Heptachlor
Heptachlor epoxide
Heptanoic acid
Hexanoic acid
lndeno(l,2,3-cd)pyrene
Iron
Isopropanol
Isovaleric acid
Lead
Lindane
Lithium
Magnesium
Manganese
m-Cresol
Mercury
Methanol
Methyl bromide
Methyl ethyl ketone
Molybdenum
Naphthalene
Nickel
Nitrate
Nitrite
N-Nitrosodiphenylamine
o-Cresol
p,p'-DDE
p-Cresol
p-Cymene
Pentanoicacid
CASRN
64-18-6
76-44-8
1024-57-3
111-14-8
142-62-1
193-39-5
7439-89-6
67-63-0
503-74-2
7439-92-1
58-89-9
7439-93-2
7439-95-4
7439-96-5
108-39-4
7439-97-6
67-56-1
74-83-9
78-93-3
7439-98-7
91-20-3
7440-02-0
14797-55-8
14797-65-0
86-30-6
95-48-7
72-55-9
106-44-5
99-87-6
109-52-4
Physico-
chemical
properties
X
X
X
X
X
X

X
X

X



X

X
X
X

X



X
X
X
X
X
X
Selected
toxic ity
reference
value
X
X
X


X
X


X
X
X

X
X
X
X
X
X
X
X

X
X
X
X
X
X


Reference
10
3,9
3,9
10
10
3,9
3, 9, 10
3,9
10
3, 9, 10
3,9
3, 9, 10
3, 9, 10
3, 9, 10
3, 9, 10
3, 9, 10
3,9
3,9
3, 9, 10
3, 9, 10
3, 9, 10
3, 9, 10
3, 9, 10
3, 9, 10
3,9
3, 9, 10
3,9
3, 9, 10
9,10
10
                This document is a draft for review purposes only and does not constitute Agency policy.
June 2015                                         A-61                      DRAFT—DO NOT CITE OR QUOTE

-------
Hydraulic Fracturing Drinking Water Assessment
Appendix A
Chemical name
Phenanthrene
Phenol
Phorate
Phosphorus
Potassium
Propionic acid
Propylbenzene
Pyrene
Pyridine
Radium
Radium-226
Radium-228
Safrole
sec-Butylbenzene
Selenium
Silica
Silicon
Silver
Sodium
Strontium
Sulfate
Sulfite
Tetrachloroethylene
Thallium
Tin
Titanium
Toluene
Vanadium
CASRN
85-01-8
108-95-2
298-02-2
7723-14-0
7440-09-7
79-09-4
103-65-1
129-00-0
110-86-1
7440-14-4
13982-63-3
15262-20-1
94-59-7
135-98-8
7782-49-2
7631-86-9
7440-21-3
7440-22-4
7440-23-5
7440-24-6
14808-79-8
14265-45-3
127-18-4
7440-28-0
7440-31-5
7440-32-6
108-88-3
7440-62-2
Physico-
chemical
properties
X
X
X


X
X
X
X



X
X








X



X

Selected
toxic ity
reference
value

X
X
X



X
X



X

X


X

X


X

X

X
X
Reference
3, 9, 10
3, 9, 10
9
3,9
3, 9, 10
10
9
9,10
3, 9, 10
3
3,10
3,10
3,9
9
3, 9, 10
10
10
3, 9, 10
3, 9, 10
3, 9, 10
3, 9, 10
3
3,9
3, 9, 10
9,10
3, 9, 10
3, 9, 10
3,10
                This document is a draft for review purposes only and does not constitute Agency policy.
June 2015                                         A-62                      DRAFT—DO NOT CITE OR QUOTE

-------
Hydraulic Fracturing Drinking Water Assessment
Appendix A



Chemical name
Xylenes
Zinc
Zirconium



CASRN
1330-20-7
7440-66-6
7440-67-7

Physico-
chemical
properties
X


Selected
toxic ity
reference
value
X
X




Reference
3, 9, 10
3, 9, 10
3
A.2.   References for Appendix A

Colborn. T: Kwiatkowski. C: Schultz. K: Bachran. M. (2011). Natural gas operations from a public health
   perspective. Hum Ecol Risk Assess 17:1039-1056. http://dx.doi.org/10.1080/10807039.2011.605662

Hayes. T. (2009). Sampling and analysis of water streams associated with the development of Marcellus shale
   gas. Des Plaines, IL: Marcellus Shale Coalition, http://eidmarcellus.org/wp-
   content/uploads/2012/ll/MSCommission-Report.pdf

House of Representatives (U.S. House of Representatives). (2011). Chemicals used in hydraulic fracturing.
   Washington, D.C.: U.S. House of Representatives, Committee on Energy and Commerce, Minority Staff.
   http://democrats.energycommerce.house.gov/sites/default/files/documents/Hydraulic-Fracturing-
   Chemicals-2011-4-18.pdf

NLM (National Institutes  of Health, National Library of Medicine). (2014). ChemID plus advanced. Available
   online at http://chem.sis.nlm.nih.gov/chemidplus/

NYSDEC (New York State Department of Environmental Conservation). (2011). Revised draft supplemental
   generic environmental impact statement (SGEIS) on the oil, gas and solution mining regulatory program:
   Well permit issuance for horizontal drilling and high-volume hydraulic fracturing to develop the Marcellus
   shale and other low-permeability gas reservoirs. Albany, NY: NY SDEC.
   http://www.dec.ny.gov/energy/75370.html

OSHA. Title 29 - Department of Labor. Subpart z Toxic and hazardous substances, hazard communication. §
   1910.1200 (2013). http://www.gpo.gov/fdsys/pkg/CFR-2013-title29-vol6/xml/CFR-2013-title29-vol6-
   secl910-1200.xml

PA PEP  (Pennsylvania Department of Environmental Protection). (2010). Chemicals used by hydraulic
   fracturing companies  in Pennsylvania for surface and hydraulic fracturing activities. Harrisburg, PA:
   Pennsylvania Department of Environmental Protection (PADEP).
   http://files.dep.state.pa.us/OilGas/BOGM/BOGMPortalFiles/MarcellusShale/Frac%201ist%206-30-
   2010.pdf.

Sheets. MSP,  (a) Encana/Halliburton Energy Services, Inc.: Duncan, Oklahoma. Provided by Halliburton
   Energy Services during an onsite visit by the EPA on May 10, 2010; (b) Encana Oil and Gas (USA), Inc.:
   Denver, Colorado.  Provided to US EPA Region 8. Material Safety Data Sheets.

U.S. EPA (U.S. Environmental Protection Agency). (2004). Evaluation of impacts to underground sources of
   drinking water by  hydraulic fracturing of coalbed methane reservoirs. (EPA/816/R-04/003). Washington,
   DC.: U.S. Environmental  Protection Agency, Office of Solid Waste.
               This document is a draft for review purposes only and does not constitute Agency policy.

June 2015                                      A-63                    DRAFT—DO NOT CITE OR QUOTE

-------
Hydraulic Fracturing Drinking Water Assessment                                                  Appendix A
U.S. EPA (U.S. Environmental Protection Agency). (2011b). Sampling data for flowback and produced water
   provided to EPA by nine oil and gas well operators (non-confidential business information). US
   Environmental Protection Agency.
   http://www.regulations.gov/#!docketDetail:rpp=100:so=DESC:sb=docId:po=0:D=EPA-HQ-ORD-2010-
   0674

U.S. EPA (U.S. Environmental Protection Agency). (2013a). Data received from oil and gas exploration and
   production companies, including hydraulic fracturing service companies 2011 to 2013. Non-confidential
   business information source documents are located in Federal Docket ID: EPA-HQ-ORD2010-0674.
   Available at http://www.regulations.gov.

       , (U.S. Environmental Protection Agency). (2013b). Distributed structure-searchable toxicity
   (DSSTOX) database network. Available online at http://www.epa.gov/ncct/dsstox/index.html
               This document is a draft for review purposes only and does not constitute Agency policy.

June 2015                                       A-64                   DRAFT—DO NOT CITE OR QUOTE

-------
Hydraulic Fracturing Drinking Water Assessment                                   Appendix B
                          Appendix  B
Water Acquisition Tables
           This document is a draft for review purposes only and does not constitute Agency policy.
June 2015                                            DRAFT—DO NOT CITE OR QUOTE

-------
Hydraulic Fracturing Drinking Water Assessment
Appendix B
Appendix B.  Water Acquisition Tables
B.I.   Supplemental Tables
Table B-l. Annual average hydraulic fracturing water use and consumption in 2011 and 2012
          compared to total annual water use and consumption in 2010 by state.

          Hydraulic fracturing water use data from the EPA's project database of disclosures to FracFocus 1.0
                    Sc)- Annual total water use data from the U.S. Geological Survey (USGS) Water Census
                        :)- Estimates of consumptions derived from hydraulic fracturing water use and total
         water use data. States listed in descending order by the volume of hydraulic fracturing water use.
State
Texas
Pennsylvania
Arkansas
Colorado
Oklahoma
Louisiana
North Dakota
West Virginia
Wyoming
New Mexico
Ohio
Utah
Montana
Kansas
California
Michigan
Total annual water
use in 2010
(millions of gal)a'b
9,052,000
2,967,450
4,124,500
4,015,000
1,157,050
3,117,100
419,750
1,288,450
1,715,500
1,153,400
3,445,600
1,627,900
2,792,250
1,460,000
13,870,000
3,942,000
Annual average
hydraulic fracturing
water use in 2011
and 2012
(millions of gal)c
19,942
5,105
3,676
3,277
2,949
2,462
2,181
657
538
371
273
251
155
66
44
28
Hydraulic fracturing
water use compared
to total water use
(%)d
0.2
0.2
0.1
0.1
0.3
0.1
0.5
0.1
<0.1
<0.1
<0.1
<0.1
<0.1
<0.1
<0.1
<0.1
Hydraulic fracturing
water consumption
com pa red to total
water consumption
(%)d'e
0.7
1.4
0.1
0.1
0.8
0.4
2.9
0.5
<0.1
<0.1
0.1
<0.1
<0.1
<0.1
<0.1
<0.1
             This document is a draft for review purposes only and does not constitute Agency policy.

June 2015                                   B-l                  DRAFT—DO NOT CITE OR QUOTE

-------
Hydraulic Fracturing Drinking Water Assessment
Appendix B
State
Mississippi
Alaska'
Virginia
Alabama
TOTAL for all 20 states
Total annual water
use in 2010
(millions of gal)a'b
1,434,450
397,850
2,792,250
3,635,400
64,407,900
Annual average
hydraulic fracturing
water use in 2011
and 2012
(millions of gal)c
18
7
1
1
42,001
Hydraulic fracturing
water use compared
to total water use
(%)d
<0.1
<0.1
<0.1
<0.1
0.1
Hydraulic fracturing
water consumption
compared to total
water consumption
(%)d'e
<0.1
<0.1
<0.1
<0.1
0.2
 a Texas, Colorado, Pennsylvania, North Dakota, Oklahoma, and Utah all made some degree of reporting to FracFocus
 mandatory rather than voluntary during this time period analyzed, January 1, 2011, to February 28, 2013. Three other states
 started requiring disclosure to either FracFocus or the state (Louisiana, Montana, and Ohio), and five states required or began
 requiring disclosure to the state (Arkansas, Michigan, New Mexico, West Virginia, and Wyoming). Alabama, Alaska, California,
 Kansas, Mississippi, and Virginia did not have reporting requirements during the period of time studied (U.S. EPA, 2015a).
 b State-level data accessed from the USGS website (http://water.usgs.gov/watuse/data/2010/) on January 27, 2015. Total
 water withdrawals per day (located in downloaded Table 1) were multiplied by 365 days to estimate total water use for the
 year (Maupin et al., 2014).
 c Average of water used for hydraulic fracturing in 2011 and 2012 as reported to FracFocus (U.S. EPA, 2015c).
 d Percentages were calculated by averaging annual water use for hydraulic fracturing reported in FracFocus in 2011 and 2012
 for a given state (U.S. EPA, 2015c), and then dividing by 2010 USGS hydraulic fracturing water use (Maupin et al., 2014) and
 multiplying by 100. Note that the annual hydraulic fracturing water use reported in FracFocus (the numerator) was not added
 to the 2010 total USGS water use value in the denominator, and the percentage is simply calculated as by dividing annual
 hydraulic fracturing use by 2010 total water use or consumption. This was done because of the difference in years between
 the two datasets, and because the USGS 2010 Census (Maupin et al., 2014) already included an estimate of hydraulic
 fracturing water use in its mining category. This approach is also consistent with that of other literature on this topic; see
 Nicot and Scanlon (2012).
 e Consumption values were calculated with use-specific consumption rates predominantly from the USGS, including 19.2% for
 public supply,  19.2% for domestic use, 60.7% for irrigation, 60.7% for livestock, 14.8% for industrial uses, 14.8% for mining
 (Sollevetal.. 1998). and 2.7% for thermoelectric power (USGS. 2014). We used a rate of 71.6% for aquaculture (from
 Verdegem and Bosma. 2009) (evaporation per  kg fish + infiltration per kg)/(total water use per kg) *100. These rates were
 multiplied by each USGS water use value (Maupin et al., 2014) to yield a total water consumption estimate. To calculate a
 consumption amount for hydraulic fracturing, we used a consumption rate of 82.5%. This was calculated by taking the median
 value for all reported produced water/injected  water percentages in Tables 7-1 and 7-2 of this assessment and then
 subtracting from 100%. If a range of values was given, the midpoint was used. Note that this is likely a  low estimate of
 consumption since much of this return water is not subsequently treated and reused, but rather disposed of in underground
 injection wells—see Chapter 8.
 f All  reported hydraulic fracturing disclosures for Alaska passed state locational quality assurance methods, but not county
 methods (U.S. EPA, 2015c). Thus, only state-level cumulative values were reported here, and no county-level data are
 provided in subsequent tables.
                 This document is a draft for review purposes only and does not constitute Agency policy.

June 2015                                              B-2                        DRAFT—DO NOT CITE OR QUOTE

-------
Hydraulic Fracturing Drinking Water Assessment
Appendix B
Table B-2. Annual average hydraulic fracturing water use and consumption in 2011 and 2012
           compared to total annual water use and consumption in 2010 by county.

           Counties listed contained wells used for hydraulic fracturing according to the EPA's project database of
           disclosures to FracFocus 1.0 (UiSi_EPAt2015c). Annual total water use data from the USGS Water
           Census (Mau|)in_et_ajil_2014). Estimates of consumption derived from hydraulic fracturing water use
           and total water use data.
State
Alabama
Arkansas
California
Colorado
Cou nty
Jefferson
Tuscaloosa
Cleburne
Conway
Faulkner
Independence
Logan
Sebastian
Van Buren
White
Yell
Colusa
Glenn
Kern
Los Angeles
Sutter
Ventura
Adams
Arapahoe
Boulder
Total annual
water use in 2010
(millions of gal)a
29,685.5
14,319.0
9,471.8
10,643.4
3,204.7
57,195.5
1,525.7
1,365.1
1,587.8
32,131.0
1,507.5
304,782.3
221,420.0
788,359.9
1,118,363.7
263,511.8
262,610.2
84,285.8
68,255.0
84,537.7
Annual average
hydraulic
fracturing water
use in 2011 and
2012 (millions of
gal)b
0.6
0.5
740.9
798.1
284.0
80.3
2.4
0.6
899.6
869.8
<0.1
<0.1
<0.1
41.7
0.2
0.2
1.8
3.2
4.0
4.1
Hydraulic
fracturing water
use compared to
total water use
(%)C
<0.1
<0.1
7.8
7.5
8.9
0.1
0.2
<0.1
56.7
2.7
<0.1
<0.1
<0.1
<0.1
<0.1
<0.1
<0.1
<0.1
<0.1
<0.1
Hydraulic fracturing
water consumption
compared to total
water consumption
(%)c'd
<0.1
<0.1
32.9
21.2
13.7
0.3
0.3
<0.1
168.8
4.7
<0.1
<0.1
<0.1
<0.1
<0.1
<0.1
<0.1
<0.1
<0.1
<0.1
               This document is a draft for review purposes only and does not constitute Agency policy.

June 2015                                       B-3                    DRAFT—DO NOT CITE OR QUOTE

-------
Hydraulic Fracturing Drinking Water Assessment
Appendix B
State
Colorado,
cont.
Kansas
Cou nty
Broomfield
Delta
Dolores
El Paso
Elbert
Fremont
Garfield
Jackson
La Plata
Larimer
Las Animas
Mesa
Moffat
Morgan
Phillips
Rio Blanco
Routt
San Miguel
Weld
Yuma
Barber
Clark
Comanche
Finney
Grant
Total annual
water use in 2010
(millions of gal)a
2,336.0
131,221.2
2,040.4
42,380.2
5,040.7
53,366.7
95,436.6
126,968.9
122,873.6
150,690.3
26,911.5
275,476.5
62,093.8
67,901.0
21,509.5
97,513.4
74,460.0
13,848.1
168,677.5
80,595.7
2,164.5
1,898.0
3,011.3
102,685.5
47,128.8
Annual average
hydraulic
fracturing water
use in 2011 and
2012 (millions of
gal)b
4.5
0.5
0.1
<0.1
<0.1
0.6
1,804.2
1.0
3.5
5.4
7.9
122.1
14.5
3.9
0.2
147.3
0.1
0.3
1,149.4
0.4
9.9
0.8
25.6
2.4
0.2
Hydraulic
fracturing water
use compared to
total water use
(%)C
0.2
<0.1
<0.1
<0.1
<0.1
<0.1
1.9
<0.1
<0.1
<0.1
<0.1
<0.1
<0.1
<0.1
<0.1
0.2
<0.1
<0.1
0.7
<0.1
0.5
<0.1
0.9
<0.1
<0.1
Hydraulic fracturing
water consumption
compared to total
water consumption
(%)c'd
0.4
<0.1
<0.1
<0.1
<0.1
<0.1
2.7
<0.1
<0.1
<0.1
<0.1
0.1
<0.1
<0.1
<0.1
0.2
<0.1
<0.1
1.0
<0.1
0.7
0.1
1.2
<0.1
<0.1
                This document is a draft for review purposes only and does not constitute Agency policy.
June 2015                                          B-4                      DRAFT—DO NOT CITE OR QUOTE

-------
Hydraulic Fracturing Drinking Water Assessment
Appendix B
State
Kansas, cont.
Louisiana
Cou nty
Gray
Harper
Haskell
Hodgeman
Kearny
Lane
Meade
Morton
Ness
Seward
Sheridan
Stanton
Stevens
Sumner
Allen
Beau regard
Bienville
Bossier
Caddo
Calcasieu
Caldwell
Claiborne
De Soto
East Feliciana
Jackson
Total annual
water use in 2010
(millions of gal)a
69,379.2
1,357.8
72,496.3
8,460.7
64,134.2
5,628.3
55,958.2
17,403.2
1,478.3
57,443.7
26,393.2
41,420.2
72,124.0
3,442.0
8,942.5
10,161.6
4,810.7
5,599.1
53,644.1
81,621.3
1,398.0
952.7
13,373.6
1,350.5
1,456.4
Annual average
hydraulic
fracturing water
use in 2011 and
2012 (millions of
gal)b
3.3
17.3
0.1
2.7
<0.1
0.8
<0.1
<0.1
1.6
<0.1
0.7
<0.1
0.1
0.2
0.1
2.3
108.9
110.1
153.6
0.1
<0.1
3.8
1,085.9
3.7
<0.1
Hydraulic
fracturing water
use compared to
total water use
(%)C
<0.1
1.3
<0.1
<0.1
<0.1
<0.1
<0.1
<0.1
0.1
<0.1
<0.1
<0.1
<0.1
<0.1
<0.1
<0.1
2.3
2.0
0.3
<0.1
<0.1
0.4
8.1
0.3
<0.1
Hydraulic fracturing
water consumption
compared to total
water consumption
(%)c'd
<0.1
2.0
<0.1
<0.1
<0.1
<0.1
<0.1
<0.1
0.2
<0.1
<0.1
<0.1
<0.1
<0.1
<0.1
0.1
10.0
4.9
1.7
<0.1
<0.1
1.1
47.4
0.7
<0.1
                This document is a draft for review purposes only and does not constitute Agency policy.
June 2015                                          B-5                      DRAFT—DO NOT CITE OR QUOTE

-------
Hydraulic Fracturing Drinking Water Assessment
Appendix B
State
Louisiana,
cont.
Michigan
Mississippi
Montana
Cou nty
Lincoln
Natchitoches
Rapides
Red River
Sabine
Tangipahoa
Union
Webster
West Feliciana
Winn
Cheboygan
Gladwin
Kalkaska
Missaukee
Ogemaw
Roscommon
Amite
Wilkinson
Daniels
Garfield
Glacier
Musselshell
Richland
Total annual
water use in 2010
(millions of gal)a
3,000.3
12,530.5
199,976.2
1,606.0
1,522.1
7,329.2
1,481.9
2,664.5
15,191.3
846.8
2,777.7
850.5
1,233.7
1,423.5
1,179.0
1,000.1
792.1
1,270.2
1,408.9
1,631.6
46,760.2
26,827.5
94,797.8
Annual average
hydraulic
fracturing water
use in 2011 and
2012 (millions of
gal)b
3.3
12.7
1.7
569.6
395.2
1.9
4.9
1.2
2.3
1.1
<0.1
1.1
24.0
<0.1
<0.1
2.4
14.4
3.2
0.6
0.5
5.1
0.4
83.5
Hydraulic
fracturing water
use compared to
total water use
(%)C
0.1
0.1
<0.1
35.5
26.0
<0.1
0.3
<0.1
<0.1
0.1
<0.1
0.1
1.9
<0.1
<0.1
0.2
1.8
0.3
<0.1
<0.1
<0.1
<0.1
0.1
Hydraulic fracturing
water consumption
compared to total
water consumption
(%)c'd
0.3
0.2
<0.1
83.2
76.6
0.1
1.0
0.1
0.1
0.4
<0.1
0.4
3.7
<0.1
<0.1
0.9
3.8
0.4
0.1
<0.1
<0.1
<0.1
0.1
                This document is a draft for review purposes only and does not constitute Agency policy.
June 2015                                          B-6                      DRAFT—DO NOT CITE OR QUOTE

-------
Hydraulic Fracturing Drinking Water Assessment
Appendix B
State
Montana,
cont.
New Mexico
North Dakota
Cou nty
Roosevelt
Rosebud
Sheridan
Chaves
Colfax
Eddy
Harding
Lea
Rio Arriba
Roosevelt
San Juan
Sandoval
Billings
Bottineau
Burke
Divide
Dunn
Golden Valley
Mckenzie
Mclean
Mountrail
Stark
Williams
Total annual
water use in 2010
(millions of gal)a
31,539.7
71,412.3
7,354.8
88,078.2
17,450.7
70,612.9
1,168.0
64,057.5
39,080.6
63,367.7
125,432.3
23,922.1
762.9
1,164.4
394.2
806.7
1,076.8
208.1
13,753.2
7,873.1
1,248.3
1,168.0
7,705.2
Annual average
hydraulic
fracturing water
use in 2011 and
2012 (millions of
gal)b
52.1
3.5
9.7
2.8
0.7
225.6
0.1
113.7
16.5
<0.1
11.6
0.4
44.4
0.1
63.6
102.2
309.5
4.6
588.4
12.2
449.4
48.0
558.5
Hydraulic
fracturing water
use compared to
total water use
(%)C
0.2
<0.1
0.1
<0.1
<0.1
0.3
<0.1
0.2
<0.1
<0.1
<0.1
<0.1
5.8
<0.1
16.1
12.7
28.7
2.2
4.3
0.2
36.0
4.1
7.2
Hydraulic fracturing
water consumption
compared to total
water consumption
(%)c'd
0.2
<0.1
0.2
<0.1
<0.1
0.5
<0.1
0.3
0.1
<0.1
<0.1
<0.1
16.2
<0.1
40.8
18.6
43.1
3.8
6.2
0.4
98.3
8.5
11.3
                This document is a draft for review purposes only and does not constitute Agency policy.
June 2015                                          B-7                      DRAFT—DO NOT CITE OR QUOTE

-------
Hydraulic Fracturing Drinking Water Assessment
Appendix B
State
Ohio
Oklahoma
Cou nty
Ashland
Belmont
Carroll
Columbiana
Coshocton
Guernsey
Harrison
Jefferson
Knox
Medina
Muskingum
Noble
Portage
Stark
Tuscarawas
Wayne
Alfalfa
Beaver
Beckham
Blaine
Bryan
Caddo
Canadian
Carter
Coal
Total annual
water use in 2010
(millions of gal)a
2,033.1
65,528.5
1,127.9
3,763.2
53,775.5
2,379.8
481.8
632,917.3
3,270.4
3,540.5
6,018.9
478.2
18,414.3
16,479.8
14,165.7
6,051.7
2,996.7
15,341.0
4,099.0
3,763.2
5,062.6
24,064.5
5,584.5
159,906.5
1,193.6
Annual average
hydraulic
fracturing water
use in 2011 and
2012 (millions of
gal)b
1.5
1.9
152.7
30.7
5.4
8.4
16.5
26.2
1.1
1.3
5.1
8.3
3.2
2.4
6.7
1.7
182.7
23.1
108.0
203.3
10.3
25.4
441.9
161.9
85.9
Hydraulic
fracturing water
use compared to
total water use
(%)C
0.1
<0.1
13.5
0.8
<0.1
0.4
3.4
<0.1
<0.1
<0.1
0.1
1.7
<0.1
<0.1
<0.1
<0.1
6.1
0.2
2.6
5.4
0.2
0.1
7.9
0.1
7.2
Hydraulic fracturing
water consumption
compared to total
water consumption
(%)c'd
0.2
0.1
37.3
2.2
0.1
0.7
7.3
0.1
0.1
0.1
0.3
3.4
0.1
<0.1
0.2
0.1
12.0
0.3
4.7
9.3
0.4
0.3
15.6
0.5
21.5
                This document is a draft for review purposes only and does not constitute Agency policy.
June 2015                                          B-8                      DRAFT—DO NOT CITE OR QUOTE

-------
Hydraulic Fracturing Drinking Water Assessment
Appendix B
State
Oklahoma,
cont.
Cou nty
Custer
Dewey
Ellis
Garvin
Grady
Grant
Harper
Hughes
Jefferson
Johnston
Kay
Kingfisher
Kiowa
Latimer
Le Flore
Logan
Love
Major
Marshall
McClain
Noble
Oklahoma
Osage
Pawnee
Payne
Total annual
water use in 2010
(millions of gal)a
3,281.4
10,953.7
8,486.3
16,279.0
13,537.9
5,569.9
3,266.8
3,394.5
4,496.8
1,671.7
16,957.9
3,744.9
5,022.4
1,062.2
8,635.9
4,077.1
2,011.2
6,321.8
2,613.4
2,952.9
12,990.4
47,836.9
6,971.5
4,839.9
4,332.6
Annual average
hydraulic
fracturing water
use in 2011 and
2012 (millions of
gal)b
19.0
162.6
184.3
15.0
111.5
77.8
8.8
30.5
<0.1
32.9
17.3
10.2
0.1
0.6
0.3
4.2
4.4
1.2
98.4
2.1
25.3
1.2
3.8
15.7
9.9
Hydraulic
fracturing water
use compared to
total water use
(%)C
0.6
1.5
2.2
0.1
0.8
1.4
0.3
0.9
<0.1
2.0
0.1
0.3
<0.1
0.1
<0.1
0.1
0.2
<0.1
3.8
0.1
0.2
<0.1
0.1
0.3
0.2
Hydraulic fracturing
water consumption
compared to total
water consumption
(%)c'd
1.2
6.2
3.2
0.4
2.3
5.2
0.4
2.2
<0.1
4.7
0.4
0.5
<0.1
0.1
<0.1
0.3
0.5
<0.1
7.2
0.2
1.8
<0.1
0.2
1.4
0.6
                This document is a draft for review purposes only and does not constitute Agency policy.
June 2015                                          B-9                      DRAFT—DO NOT CITE OR QUOTE

-------
Hydraulic Fracturing Drinking Water Assessment
Appendix B
State
Oklahoma,
cont.
Pennsylvania
Cou nty
Pittsburg
Roger Mills
Seminole
Stephens
Texas
Washita
Woods
Allegheny
Armstrong
Beaver
Blair
Bradford
Butler
Cameron
Centre
Clarion
Clearfield
Clinton
Columbia
Crawford
Elk
Fayette
Forest
Greene
Huntingdon
Total annual
water use in 2010
(millions of gal)a
6,314.5
2,847.0
124,837.3
49,990.4
110,208.1
3,310.6
4,139.1
234,140.2
65,853.3
157,793.2
8,303.8
4,354.5
5,730.5
292.0
16,560.1
1,843.3
111,051.3
6,161.2
3,810.6
5,091.8
7,876.7
16,465.2
744.6
13,023.2
5,121.0
Annual average
hydraulic
fracturing water
use in 2011 and
2012 (millions of
gal)b
349.0
235.5
0.1
27.7
0.1
102.1
155.1
13.6
55.7
30.5
5.9
1,059.4
121.8
6.6
38.5
8.1
111.5
94.4
5.6
2.4
37.5
120.2
7.7
359.0
2.7
Hydraulic
fracturing water
use compared to
total water use
(%)C
5.5
8.3
<0.1
0.1
<0.1
3.1
3.7
<0.1
0.1
<0.1
0.1
24.3
2.1
2.3
0.2
0.4
0.1
1.5
0.1
<0.1
0.5
0.7
1.0
2.8
0.1
Hydraulic fracturing
water consumption
compared to total
water consumption
(%)c'd
16.0
12.6
<0.1
0.3
<0.1
5.4
10.9
<0.1
1.8
0.2
0.2
78.2
6.0
4.1
0.5
1.4
2.3
3.0
0.4
0.1
1.9
2.7
1.6
24.7
0.2
                This document is a draft for review purposes only and does not constitute Agency policy.
June 2015                                          B-10                      DRAFT—DO NOT CITE OR QUOTE

-------
Hydraulic Fracturing Drinking Water Assessment
Appendix B
State
Pennsylvania,
cont.
Texas
Cou nty
Indiana
Jefferson
Lawrence
Lycoming
McKean
Potter
Somerset
Sullivan
Susquehanna
Tioga
Venango
Warren
Washington
Westmoreland
Wyoming
Andrews
Angelina
Archer
Atascosa
Austin
Bee
Borden
Bosque
Brazos
Brooks
Total annual
water use in 2010
(millions of gal)a
21,819.7
1,730.1
36,598.6
5,854.6
4,723.1
2,281.3
10,833.2
222.7
1,617.0
2,909.1
2,989.4
5,099.1
130,535.0
14,607.3
4,788.8
23,363.7
5,540.7
2,536.8
15,038.0
2,555.0
3,087.9
2,427.3
3,544.2
24,790.8
1,204.5
Annual average
hydraulic
fracturing water
use in 2011 and
2012 (millions of
gal)b
16.2
13.8
27.0
704.6
60.5
16.5
5.8
66.5
751.3
566.3
2.4
2.3
433.7
207.0
150.0
236.2
0.8
0.1
327.3
2.1
20.0
8.0
0.7
7.7
1.5
Hydraulic
fracturing water
use compared to
total water use
(%)C
0.1
0.8
0.1
12.0
1.3
0.7
0.1
29.9
46.5
19.5
0.1
<0.1
0.3
1.4
3.1
1.0
<0.1
<0.1
2.2
0.1
0.6
0.3
<0.1
<0.1
0.1
Hydraulic fracturing
water consumption
compared to total
water consumption
(%)c'd
0.7
1.7
1.0
33.8
4.9
1.0
0.2
79.8
123.4
47.3
0.3
0.2
4.6
3.8
15.2
2.7
<0.1
<0.1
4.0
0.1
1.1
1.0
<0.1
0.1
0.3
                This document is a draft for review purposes only and does not constitute Agency policy.
June 2015                                          B-ll                      DRAFT—DO NOT CITE OR QUOTE

-------
Hydraulic Fracturing Drinking Water Assessment
Appendix B
State
Texas, cont.
Cou nty
Burleson
Cherokee
Clay
Cochran
Coke
Colorado
Concho
Cooke
Cottle
Crane
Crockett
Crosby
Culberson
Dallas
Dawson
DeWitt
Denton
Dimmit
Ector
Edwards
Ellis
Erath
Fayette
Fisher
Franklin
Total annual
water use in 2010
(millions of gal)a
10,694.5
24,845.6
1,963.7
24,035.3
12,713.0
52,465.1
2,832.4
4,533.3
733.7
8,566.6
4,281.5
27,261.9
14,311.7
112,204.7
28,842.3
2,394.4
60,684.9
4,073.4
21,958.4
332.2
8,530.1
5,876.5
9,008.2
2,854.3
1,956.4
Annual average
hydraulic
fracturing water
use in 2011 and
2012 (millions of
gal)b
3.0
0.5
<0.1
3.0
0.3
0.1
<0.1
454.3
0.3
92.3
279.0
1.3
37.7
5.6
17.5
546.6
455.0
1,794.2
226.5
<0.1
4.2
0.8
13.7
1.8
<0.1
Hydraulic
fracturing water
use compared to
total water use
(%)C
<0.1
<0.1
<0.1
<0.1
<0.1
<0.1
<0.1
10.0
<0.1
1.1
6.5
<0.1
0.3
<0.1
0.1
22.8
0.7
44.0
1.0
<0.1
<0.1
<0.1
0.2
0.1
<0.1
Hydraulic fracturing
water consumption
compared to total
water consumption
(%)c'd
<0.1
<0.1
<0.1
<0.1
<0.1
<0.1
<0.1
29.9
0.1
5.7
29.5
<0.1
0.4
<0.1
0.1
48.6
2.3
81.3
4.6
<0.1
0.1
<0.1
1.2
0.1
<0.1
                This document is a draft for review purposes only and does not constitute Agency policy.
June 2015                                          B-12                      DRAFT—DO NOT CITE OR QUOTE

-------
Hydraulic Fracturing Drinking Water Assessment
Appendix B
State
Texas, cont.
Cou nty
Freestone
Frio
Gaines
Garza
Glasscock
Goliad
Gonzales
Grayson
Gregg
Grimes
Hansford
Hardeman
Hardin
Harrison
Hartley
Haskell
Hemphill
Hidalgo
Hockley
Hood
Houston
Howard
Hutchinson
Irion
Jack
Total annual
water use in 2010
(millions of gal)a
297,861.9
20,589.7
121,778.6
5,234.1
20,680.9
142,963.2
7,121.2
8,143.2
33,010.6
112,500.3
43,643.1
2,230.2
2,376.2
11,869.8
113,555.2
12,143.6
3,150.0
171,630.3
46,314.9
9,351.3
3,686.5
10,811.3
34,437.8
1,335.9
2,241.1
Annual average
hydraulic
fracturing water
use in 2011 and
2012 (millions of
gal)b
53.9
127.5
21.6
0.6
598.1
<0.1
577.9
9.3
9.4
15.5
2.9
0.4
0.1
141.6
1.9
0.1
263.9
8.0
3.0
76.0
8.6
97.6
0.3
411.4
14.0
Hydraulic
fracturing water
use compared to
total water use
(%)C
<0.1
0.6
<0.1
<0.1
2.9
<0.1
8.1
0.1
<0.1
<0.1
<0.1
<0.1
<0.1
1.2
<0.1
<0.1
8.4
<0.1
<0.1
0.8
0.2
0.9
<0.1
30.8
0.6
Hydraulic fracturing
water consumption
compared to total
water consumption
(%)c'd
0.5
0.9
<0.1
<0.1
4.2
<0.1
17.6
0.3
0.2
0.3
<0.1
<0.1
<0.1
6.0
<0.1
<0.1
16.3
<0.1
<0.1
2.2
0.6
2.7
<0.1
74.5
2.2
                This document is a draft for review purposes only and does not constitute Agency policy.
June 2015                                          B-13                      DRAFT—DO NOT CITE OR QUOTE

-------
Hydraulic Fracturing Drinking Water Assessment
Appendix B
State
Texas, cont.
Cou nty
Jefferson
Jim Hogg
Johnson
Jones
Karnes
Kenedy
Kent
King
Kleberg
Knox
La Salle
Lavaca
Lee
Leon
Liberty
Limestone
Lipscomb
Live Oak
Loving
Lynn
Madison
Marion
Martin
Maverick
McMullen
Total annual
water use in 2010
(millions of gal)a
88,585.5
306.6
9,241.8
5,679.4
1,861.5
456.3
6,132.0
1,485.6
1,171.7
9,800.3
2,474.7
3,763.2
3,120.8
2,171.8
20,662.7
11,158.1
11,015.7
1,916.3
781.1
19,892.5
1,554.9
3,606.2
14,063.5
20,498.4
657.0
Annual average
hydraulic
fracturing water
use in 2011 and
2012 (millions of
gal)b
<0.1
0.1
582.0
<0.1
1,055.2
0.2
0.4
<0.1
3.4
<0.1
1,288.7
45.0
1.2
56.2
<0.1
10.7
89.0
294.0
138.4
1.1
45.3
5.9
432.0
52.4
745.9
Hydraulic
fracturing water
use compared to
total water use
(%)C
<0.1
<0.1
6.3
<0.1
56.7
0.1
<0.1
<0.1
0.3
<0.1
52.1
1.2
<0.1
2.6
<0.1
0.1
0.8
15.3
17.7
<0.1
2.9
0.2
3.1
0.3
113.5
Hydraulic fracturing
water consumption
compared to total
water consumption
(%)c'd
<0.1
0.1
18.5
<0.1
120.1
0.1
<0.1
<0.1
0.5
<0.1
93.7
2.0
0.1
6.6
<0.1
0.9
1.1
40.1
94.1
<0.1
8.2
0.9
4.7
0.4
350.4
                This document is a draft for review purposes only and does not constitute Agency policy.
June 2015                                          B-14                     DRAFT—DO NOT CITE OR QUOTE

-------
Hydraulic Fracturing Drinking Water Assessment
Appendix B
State
Texas, cont.
Cou nty
Medina
Menard
Midland
Milam
Mitchell
Montague
Montgomery
Moore
Nacogdoches
Navarro
Newton
Nolan
Nueces
Ochiltree
Oldham
Orange
Palo Pinto
Panola
Parker
Pecos
Polk
Potter
Reagan
Reeves
Roberts
Total annual
water use in 2010
(millions of gal)a
19,228.2
1,014.7
12,891.8
16,665.9
6,559.1
3,989.5
32,565.3
57,075.1
5,891.1
18,699.0
2,263.0
4,124.5
85,767.7
21,348.9
2,124.3
150,128.2
18,403.3
6,365.6
8,241.7
52,954.2
204,009.5
2,029.4
9,333.1
20,772.2
7,690.6
Annual average
hydraulic
fracturing water
use in 2011 and
2012 (millions of
gal)b
0.2
<0.1
307.4
4.9
11.0
925.3
0.2
<0.1
271.7
4.8
0.2
4.5
1.0
33.3
1.3
0.3
9.6
346.5
261.7
8.2
0.2
0.4
410.5
164.2
38.2
Hydraulic
fracturing water
use compared to
total water use
(%)C
<0.1
<0.1
2.4
<0.1
0.2
23.2
<0.1
<0.1
4.6
<0.1
<0.1
0.1
<0.1
0.2
0.1
<0.1
0.1
5.4
3.2
<0.1
<0.1
<0.1
4.4
0.8
0.5
Hydraulic fracturing
water consumption
compared to total
water consumption
(%)c'd
<0.1
<0.1
3.7
0.1
0.3
77.8
<0.1
<0.1
12.5
0.1
<0.1
0.2
<0.1
0.2
0.1
<0.1
0.3
20.7
9.8
<0.1
<0.1
<0.1
7.8
1.1
1.2
                This document is a draft for review purposes only and does not constitute Agency policy.
June 2015                                          B-15                      DRAFT—DO NOT CITE OR QUOTE

-------
Hydraulic Fracturing Drinking Water Assessment
Appendix B
State
Texas, cont.
Cou nty
Robertson
Runnels
Rusk
Sabine
San Augustine
San Patricio
Schleicher
Scurry
Shelby
Sherman
Smith
Somervell
Starr
Stephens
Sterling
Stonewall
Sutton
Tarrant
Terrell
Terry
Tyler
Upshur
Upton
Van Zandt
Walker
Total annual
water use in 2010
(millions of gal)a
158,344.3
2,847.0
582,134.9
799.4
1,131.5
4,172.0
967.3
14,187.6
4,920.2
78,073.5
11,231.1
746,005.3
9,552.1
13,446.6
719.1
923.5
1,153.4
104,430.2
543.9
48,362.5
1,872.5
8,610.4
7,975.3
4,139.1
4,478.6
Annual average
hydraulic
fracturing water
use in 2011 and
2012 (millions of
gal)b
45.4
<0.1
65.8
31.1
182.1
1.1
27.0
1.1
133.6
<0.1
0.2
4.8
5.0
2.6
36.6
0.9
1.6
1,443.0
0.1
7.5
0.1
0.2
462.6
0.1
3.4
Hydraulic
fracturing water
use compared to
total water use
(%)C
<0.1
<0.1
<0.1
3.9
16.1
<0.1
2.8
<0.1
2.7
<0.1
<0.1
<0.1
0.1
<0.1
5.1
0.1
0.1
1.4
<0.1
<0.1
<0.1
<0.1
5.8
<0.1
0.1
Hydraulic fracturing
water consumption
compared to total
water consumption
(%)c'd
0.2
<0.1
0.3
13.9
50.8
<0.1
5.0
<0.1
8.2
<0.1
<0.1
<0.1
0.1
0.1
11.9
0.3
0.3
3.9
<0.1
<0.1
<0.1
<0.1
14.2
<0.1
0.2
                This document is a draft for review purposes only and does not constitute Agency policy.
June 2015                                          B-16                     DRAFT—DO NOT CITE OR QUOTE

-------
Hydraulic Fracturing Drinking Water Assessment
Appendix B
State
Texas, cont.
Utah
Virginia
Cou nty
Waller
Ward
Washington
Webb
Wharton
Wheeler
Wichita
Wilbarger
Willacy
Wilson
Winkler
Wise
Wood
Yoakum
Young
Zapata
Zavala
Carbon
Duchesne
San Juan
Sevier
Uintah
Buchanan
Dickenson
Wise
Total annual
water use in 2010
(millions of gal)a
9,829.5
6,909.5
2,430.9
15,862.9
81,606.7
6,522.6
25,936.9
12,683.8
15,209.6
7,843.9
5,274.3
24,966.0
19,334.1
77,325.3
21,162.7
2,697.4
14,410.2
15,067.2
119,811.3
10,632.5
52,512.6
100,229.0
313.9
1,741.1
1,927.2
Annual average
hydraulic
fracturing water
use in 2011 and
2012 (millions of
gal)b
0.1
107.3
2.2
1,117.8
<0.1
858.0
0.1
0.2
0.1
84.5
7.7
529.7
0.2
7.5
0.1
1.1
130.0
7.3
85.5
0.3
<0.1
157.5
0.6
0.8
0.1
Hydraulic
fracturing water
use compared to
total water use
(%)C
<0.1
1.6
0.1
7.0
<0.1
13.2
<0.1
<0.1
<0.1
1.1
0.1
2.1
<0.1
<0.1
<0.1
<0.1
0.9
<0.1
0.1
<0.1
<0.1
0.2
0.2
<0.1
<0.1
Hydraulic fracturing
water consumption
compared to total
water consumption
(%)c'd
<0.1
4.6
0.2
18.2
<0.1
21.5
<0.1
<0.1
<0.1
1.7
0.5
8.9
<0.1
<0.1
<0.1
0.1
1.3
0.1
0.1
<0.1
<0.1
0.2
0.3
0.2
<0.1
                This document is a draft for review purposes only and does not constitute Agency policy.
June 2015                                          B-17                      DRAFT—DO NOT CITE OR QUOTE

-------
Hydraulic Fracturing Drinking Water Assessment
Appendix B
State
West Virginia
Wyoming
Cou nty
Barbour
Brooke
Doddridge
Hancock
Harrison
Lewis
Marion
Marshall
Monongalia
Ohio
Pleasants
Preston
Ritchie
Taylor
Tyler
Upshur
Webster
Wetzel
Big Horn
Campbell
Carbon
Converse
Fremont
Goshen
Hot Springs
Total annual
water use in 2010
(millions of gal)a
773.8
4,551.6
405.2
28,718.2
20,232.0
901.6
5,982.4
158,358.9
42,102.8
3,825.2
24,703.2
2,890.8
587.7
824.9
4,934.8
1,814.1
1,292.1
1,467.3
143,368.4
44,318.3
137,130.5
56,972.9
186,150.0
144,248.0
28,572.2
Annual average
hydraulic
fracturing water
use in 2011 and
2012 (millions of
gal)b
19.9
54.8
78.5
1.2
40.2
2.4
70.1
84.5
6.8
116.5
<0.1
8.4
2.8
52.9
2.1
34.9
2.3
78.2
2.9
11.7
4.5
106.8
28.2
5.8
0.3
Hydraulic
fracturing water
use compared to
total water use
(%)C
2.6
1.2
19.4
<0.1
0.2
0.3
1.2
0.1
<0.1
3.0
<0.1
0.3
0.5
6.4
<0.1
1.9
0.2
5.3
<0.1
<0.1
<0.1
0.2
<0.1
<0.1
<0.1
Hydraulic fracturing
water consumption
compared to total
water consumption
(%)c'd
6.9
5.1
69.4
<0.1
1.9
0.8
4.9
0.7
0.1
10.4
<0.1
1.4
1.7
17.6
0.2
6.8
0.3
11.9
<0.1
0.1
<0.1
0.3
<0.1
<0.1
<0.1
                This document is a draft for review purposes only and does not constitute Agency policy.
June 2015                                          B-18                      DRAFT—DO NOT CITE OR QUOTE

-------
Hydraulic Fracturing Drinking Water Assessment
Appendix B
State
Wyoming,
cont.
Cou nty
Johnson
Laramie
Lincoln
Natrona
Niobrara
Park
Sublette
Sweetwater
Uinta
Washakie
Total annual
water use in 2010
(millions of gal)a
43,205.1
86,297.0
74,562.2
62,885.9
25,148.5
111,317.7
61,006.1
61,699.6
79,518.9
60,400.2
Annual average
hydraulic
fracturing water
use in 2011 and
2012 (millions of
gal)b
<0.1
18.3
0.8
1.8
0.1
0.9
314.8
39.4
0.6
1.1
Hydraulic
fracturing water
use compared to
total water use
(%)C
<0.1
<0.1
<0.1
<0.1
<0.1
<0.1
0.5
0.1
<0.1
<0.1
Hydraulic fracturing
water consumption
compared to total
water consumption
(%)c'd
<0.1
<0.1
<0.1
<0.1
<0.1
<0.1
0.7
0.1
<0.1
<0.1
 a County-level data accessed from the USGS website (http://water.usgs.gov/watuse/data/2010/) on November 11, 2014. Total
 daily water withdrawals were multiplied by 365 days to estimate total water use for the year (Maupin et al., 2014).
 b Average of water used for hydraulic fracturing in 2011 and 2012, as reported to FracFocus (U.S. EPA, 2015c).
 c Percentages were calculated by averaging annual water use for hydraulic fracturing reported in FracFocus in 2011 and 2012
 for a given county (U.S. EPA. 2015c). and then dividing by 2010 USGS total water use for that county (Maupin etal.. 2014) and
 multiplying by 100.
 d Consumption values were calculated with use-specific consumption rates predominantly from the USGS, including 19.2% for
 public supply, 19.2% for domestic use, 60.7% for irrigation, 60.7% for livestock, 14.8% for industrial uses, 14.8% for mining
 (Sollevet al.. 1998). and 2.7% for thermoelectric power (USGS. 2014). We used a rate of 71.6% for aquaculture (from
 Verdegem and Bosma, 2009) (evaporation per kg fish + infiltration per kg)/(total water use per kg)*100. These rates were
 multiplied by each USGS water use value (Maupin etal., 2014) to yield a total water consumption estimate. To calculate a
 consumption amount for hydraulic fracturing, we used a consumption rate of 82.5%. This was calculated by taking the median
 value for all reported produced water/injected water percentages in Tables 7-1 and 7-2 of this assessment and then
 subtracting from 100%. If a range of values was given, the midpoint was used. Note that this is likely a low estimate of
 consumption since much of this return water is not subsequently treated and reused, but rather disposed of in underground
 injection wells—see Chapter 8.
                 This document is a draft for review purposes only and does not constitute Agency policy.

June 2015                                             B-19                       DRAFT—DO NOT CITE OR QUOTE

-------
Hydraulic Fracturing Drinking Water Assessment
Appendix B
Table B-3. Comparison of water use per well estimates from the EPA's project database of
            disclosures to FracFocus 1.0 (U.S.  EPA, 2015c) and literature sources.
            Source: (U.S. EPA, 2015c)
State
Colorado
North Dakota
Oklahoma
Pennsylvaniad
Texas
Texas
Texas
Average6
Median6
Basin3
Denver



Fort Worth
Salt
Western Gulf


Water use per well (gal) -
FracFocus estimate13
403,686
2,140,842
2,591,778
4,301,701
3,881,220
3,139,980
3,777,648


Water use per well (gal) -
Literature estimateb'c
2,900,000
2,200,000
3,000,000
4,450,000
4,500,000
4,000,000
4,600,000


FracFocus estimate as
a percentage of
literature estimate (%)
14
97
86
97
86
78
82
77
86
 a In cases where a basin is not specified, estimates were for the entire state and not specific to a particular basin. Basin
 boundaries for the FracFocus estimates were determined from data from the U.S. EIA (see U.S. EPA, 2015b).
 bThe type of literature estimate determined the specific comparison with FracFocus. If averages were given in the literature
 (as for North Dakota and Pennsylvania), those values were compared with FracFocus averages; where medians were given in
 the literature (as for Colorado, Oklahoma, and Texas), they were compared with FracFocus medians.
 c Literature estimates were from the following sources: Colorado (Goodwin et al.. 2014). North Dakota (North Dakota State
 Water Commission. 2014). Pennsylvania (Mitchell et al.. 2013). and Texas (Nicot et al.. 2012)—see far right-column and
 footnotes in Table B-5 for details on literature estimates. Where the literature provided a range, the mid-point was used. Only
 literature estimates that were not directly derived from FracFocus were included.
 d The results from Mitchell et al. (2013) were used for Pennsylvania since they were derived from Pennsylvania Department of
 Environment Protection records. Estimates from Hansen et al. (2013) were not included here because they were based on
 FracFocus.
 e Average and median percentage calculations were not weighted by the number of wells for a given estimate.
                This document is a draft for review purposes only and does not constitute Agency policy.

June 2015                                           B-20                      DRAFT—DO NOT CITE OR QUOTE

-------
Hydraulic Fracturing Drinking Water Assessment
                                                                                                      Appendix B
Table B-4. Comparison of well counts from the EPA's project database of disclosures to FracFocus 1.0 (U.S. EPA, 2015c) and state
            databases for North Dakota, Pennsylvania, and West Virginia.
State
North Dakota"
Pennsylvania0
WestVirginiad
Average
FracFocus well counts3
2011
613
1,137
93

2012
1,458
1,257
176

Total
2,071
2,394
269

State database well counts
2011
1,225
1,963
214

2012
1,740
1,347
251

Total
2,965
3,310
465

FracFocus counts as a percentage
of state database counts
2011
50%
58%
43%
50%
2012
84%
93%
70%
82%
Total
70%
72%
58%
67%
 a FracFocus disclosures from U.S. EPA (2015c).
 b For North Dakota state well counts, we used a North Dakota Department of Mineral Resources online database containing a list of horizontal wells completed in the Bakken
 Formation. Data for North Dakota were accessed on July 9, 2014 at https://www.dmr.nd.gov/oilgas/bakkenwells.asp.
 c For Pennsylvania state well counts, we used completed horizontal wells as a proxy for hydraulically fractured wells in the state. The Pennsylvania Department of
 Environmental Protection has online databases of permitted and spudded wells, which differentiate between conventional and unconventional wells and can generate
 summary statistics at both the county and state scale. The number of spudded wells (i.e., wells drilled) provided a better comparison with the number of hydraulically
 fractured wells in FracFocus than that of permitted wells. The number of permitted wells was nearly double that of spudded in 2011 and 2012, indicating that  almost half of
 the wells permitted were not drilled in that same year. Therefore,  we used spudded wells here. Data for Pennsylvania were accessed on February 11, 2014 from
 http://www.depreportingservices.state.pa.us/ReportServer/Pages/ReportViewer.aspx7/Oil Gas/Spud External Data.
 d For West Virginia state well counts, data on the number of hydraulically fractured wells per year were received from the West Virginia Department of Environmental
 Protection on February 25, 2014.
                     June 2015
This document is a draft for review purposes only and does not constitute Agency policy.

                                   B-21                      DRAFT—DO NOT CITE OR QUOTE

-------
Hydraulic Fracturing Drinking Water Assessment
                                                                                              Appendix B
Table B-5. Water use per hydraulically fractured well as reported in the EPA's project database of disclosures to FracFocus 1.0
           (U.S. EPA, 2015c) by state and basin.

           Souce:
           Other literature estimates are also included where available. NA indicates other literature estimates were not available. All FracFocus estimates
           were limited to disclosures with valid state, county, and volume information. States listed in order addressed in Chapter 4.
State
Texas
Colorado
Basin/total3
Permian
Western Gulf
Fort Worth
TX-LA-MS Salt
Anadarko
Other
Total
Denver
Uinta-Piceance
Raton
Other
Total
Number of
disclosures
8,419
4,549
2,564
626
604
120
16,882
3,166
1,520
146
66
4,898
Mean
(gal)
1,068,511
3,915,540
3,880,724
4,261,363
4,128,702
1,601,897
2,494,452
753,887
2,739,523
108,003
605,740
1,348,842
Median
(gal)
841,134
3,777,648
3,881,220
3,139,980
3,341,310
184,239
1,420,613
403,686
1,798,414
95,974
183,408
463,462
10th percentile
(gal)
40,090
173,832
923,381
193,768
492,421
21,470
58,709
143,715
840,778
24,917
34,412
147,353
90th percentile
(gal)
1,814,633
6,786,052
6,649,406
10,010,707
8,292,996
5,678,588
6,115,195
2,588,946
5,066,380
211,526
601,816
3,092,024
Literature estimates
Many formations reported15
4.5-4.7 million gal (median, Eagle
Ford play)b
4.5 million gal (median, Barnett play)b
6-7.5 million gal (median, Texas-
Haynesville play) and 0.5-1 million
gallons (median, Cotton Valley play)b
Many formations reported15
NA
Not reported by stateb
2.9 million gal (median, Wattenberg
field of Niobrara play)0
NA
NA
NA
NA
                   June 2015
This document is a draft for review purposes only and does not constitute Agency policy.

                                 B-22                    DRAFT—DO NOT CITE OR QUOTE

-------
Hydraulic Fracturing Drinking Water Assessment
                                                                                                    Appendix B
State
Wyoming
Pennsylvania
West Virginia
Ohio
North Dakota
Montana
Oklahoma
Basin/total3
Greater Green River
Powder River
Other
Total
Appalachian
Total
Appalachian
Total
Appalachian
Total
Williston
Total
Williston
Other
Total
Anadarko
Arkoma
Ardmore
Other
Total
Number of
disclosures
861
351
193
1,405
2,445
2,445
273
273
146
146
2,109
2,109
187
20
207
935
158
98
592
1,783
Mean
(gal)
841,702
739,129
613,618
784,746
4,301,701
4,301,701
5,034,217
5,034,217
4,206,955
4,206,955
2,140,842
2,140,842
1,640,085
945,541
1,572,979
3,742,703
6,323,750
6,637,332
1,963,480
3,539,775
Median
(gal)
752,979
5,927
41,664
322,793
4,184,936
4,184,936
5,012,238
5,012,238
3,887,499
3,887,499
2,022,380
2,022,380
1,552,596
1,017,701
1,455,757
3,259,774
6,655,929
8,021,559
1,866,144
2,591,778
10th percentile
(gal)
147,020
5,353
22,105
5,727
2,313,649
2,313,649
3,170,210
3,170,210
2,885,568
2,885,568
969,380
969,380
375,864
157,639
367,326
1,211,700
172,375
81,894
1,319,247
1,260,906
90th percentile
(gal)
1,493,266
2,863,182
1,818,606
1,837,602
6,615,981
6,615,981
7,297,080
7,297,080
5,571,027
5,571,027
3,313,482
3,313,482
3,037,398
1,575,197
2,997,552
6,972,652
9,589,554
8,835,842
2,785,352
7,402,230
Literature estimates
NA
NA
NA
NA
4.2-4.6 million gal (average, Marcellus
play, Susquehanna River Basin)d
4.1-4.5d and 4.3-4.6e million gal
(average)
NA
4.7-6 million gal (average)d
NA
NA
NA
2.2 million gal (average)'
NA
NA
NA
Many formations reported8
Many formations reported8
Many formations reported8
NA
3 million gal (median)8
                    June 2015
This document is a draft for review purposes only and does not constitute Agency policy.
                                   B-23                      DRAFT—DO NOT CITE OR QUOTE

-------
Hydraulic Fracturing Drinking Water Assessment
                                                                                                    Appendix B
State
Kansas
Arkansas
Louisiana
Utah
New Mexico
Basin/total3
Total
Arkoma
Total
TX-LA-MS Salt
Other
Total
Uinta-Piceance
Other
Total
Permian
San Juan
Other
Total
Number of
disclosures
121
1,423
1,423
939
27
966
1,396
10
1,406
732
363
50
1,145
Mean
(gal)
1,135,973
5,190,254
5,190,254
5,289,100
896,899
5,166,337
375,852
58,874
373,597
991,369
159,680
33,787
685,882
Median
(gal)
1,453,788
5,259,965
5,259,965
5,116,650
232,464
5,077,863
304,105
56,245
302,075
426,258
97,734
8,358
175,241
10th percentile
(gal)
10,836
3,234,963
3,234,963
2,851,654
87,003
1,812,099
77,166
28,745
76,286
89,895
27,217
1,100
35,638
90th percentile
(gal)
2,227,926
7,121,249
7,121,249
7,984,838
3,562,400
7,945,630
770,699
97,871
769,360
2,502,923
313,919
98,841
1,871,666
Literature estimates
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
                    June 2015
This document is a draft for review purposes only and does not constitute Agency policy.
                                   B-24                     DRAFT—DO NOT CITE OR QUOTE

-------
Hydraulic Fracturing Drinking Water Assessment
                                                                                                           Appendix B
State
California
Basin/total3
San Joaquin
Other
Total
Number of
disclosures
677
34
711
Mean
(gal)
131,653
132,391
131,689
Median
(gal)
77,238
36,099
76,818
10th percentile
(gal)
22,100
13,768
21,462
90th percentile
(gal)
285,029
361,192
285,306
Literature estimates
NA
NA
130,000 gallon
(average) h
 a Basin boundaries for the FracFocus estimates were determined from data from the U.S. EIA (see U.S. EPA, 2015b).
 b Literature estimates for Texas were from Nicot et al. (2012), using proprietary data from IMS. In most cases, Nicot et al. reported at the play scale or smaller, rather than the
 EIA basin scale used for FracFocus. We reference 2011 and 2012 (partial year) for Nicot et al. where possible to overlap with the period of study for FracFocus, though more
 years were available for most formations. A range is reported for some medians because median water use was different for the two years. There were five formations
 reported for the Permian Basin (Wolfberry, Wolfcamp, Canyon, Clearfork, and San Andres-Greyburg). The most active area in the Permian Basin in 2011-2012 was the
 Wolfberry, which reported a median of 1 to 1.1 million gallons per well—these were mostly vertical wells. For the TX-LA-MS Salt Basin, Nicot et al. reported two formations
 (TX-Haynesville and Cotton Valley), with similar levels of activity in 2011-2012. Wells in TX-Haynesville were predominantly  horizontal, while those in Cotton Valley were
 predominantly vertical (though horizontal wells in Cotton Valley were also reported). There were three fields reported in the Anadarko Basin (Granite Wash, Cleveland, and
 Marmaton). The most active area in the Anadarko Basin  in 2011-2012 was the Granite Wash, which reported a median of 3.3 to 5.2 million gallons per well and where wells
 were mostly horizontal.
 c Literature estimates for the Denver Basin were from Goodwin et al. (2014). Goodwin et al. assessed 200 randomly sampled wells in the Wattenberg Field of the Denver Basin
 (Niobrara Play), using industry data for wells operated by Noble Energy, drilled between January 1, 2010, and July 1, 2013. Water consumption is reported rather than water
 use, but Goodwin et al. assume, based on Noble Energy practices, that water use  and water consumption were identical because none of the flowback or produced water is
 reused for hydraulic fracturing. Goodwin et al. reported drilling water consumed,  hydraulic fracturing water consumed, and total water consumed. We present hydraulic
 fracturing water consumption here (hydraulic fracturing water consumption was approximately 95% of the total).
 d Hansen et al. (2013), using data from FracFocus via Skytruth. For the Susquehanna River Basin portion of the Marcellus play, and for Pennsylvania as a whole, the range of
 annual averages is reported for 2011 and 2012. Similarly, for West Virginia, the range of annual averages is  reported for 2011 and 2012 (partial year).
 e Mitchell et al. (2013), using data reported to the Pennsylvania Department of Environmental Protection. Mitchell et al. reported water use in the Ohio River Basin for 2011
 and 2012 (partial year) for horizontal and vertical wells. Here we report results for horizontal wells, which made up the majority of wells over the two-year period (i.e., 93%,
 1,191 horizontal wells versus 96 vertical wells). A range is reported as before because the average water use differed between the two years.
 f Literature estimates for North Dakota were from an informational bulletin from the North Dakota State Water Commission (2014). No further information was available.
 g Murray (2013), who assessed water use for oil and gas operations from 2000-2010 for eight formations in Oklahoma using data from the Oklahoma Corporation Commission.
 It is not possible to extract an estimate corresponding to 2011-2012 from  Murray without the raw data, because medians were presented for the 10-year period rather than
 separated by year.
 h Literature estimates for California were from a  California Council on Science and Technology report using data from FracFocus (CCST, 2014).
                      June 2015
This document is a draft for review purposes only and does not constitute Agency policy.

                                     B-25                       DRAFT—DO NOT CITE OR QUOTE

-------
Hydraulic Fracturing Drinking Water Assessment
Appendix B
Table B-6. Estimated percent domestic use water from ground water and self-supplied by
          county.

          Counties listed contained hyd rau I ica My fractured wells with valid state, county, and volume
          information (lI
           Data estimated from the USGS Water Census (MaufilrLetj|^2014).
State
Alabama
Arkansas
California
Colorado
County
Jefferson
Tuscaloosa
Cleburne
Conway
Faulkner
Independence
Logan
Sebastian
Van Buren
White
Yell
Colusa
Glenn
Kern
Los Angeles
Sutter
Ventura
Adams
Arapahoe
Boulder
Broomfield
Delta
Dolores
El Paso
Elbert
Fremont
Percent domestic use water
from ground watera'b
11.9
10.7
0.0
8.6
48.0
20.5
0.0
0.0
6.4
0.4
1.8
97.9
96.5
74.5
45.0
19.4
30.9
18.1
19.3
1.7
0.0
59.6
55.2
19.6
100.0
15.6
Percent domestic use
water self supplieda'c
0.8
6.1
0.0
8.6
3.5
9.4
0.0
0.0
6.4
0.0
1.8
10.3
21.6
1.7
4.2
4.6
3.9
2.8
1.3
1.5
0.0
28.4
51.4
5.1
75.2
15.6
               This document is a draft for review purposes only and does not constitute Agency policy.

June 2015                                      B-26                   DRAFT—DO NOT CITE OR QUOTE

-------
Hydraulic Fracturing Drinking Water Assessment
Appendix B
State
Colorado, cont.
Kansas
Cou nty
Garfield
Jackson
La Plata
Larimer
Las Animas
Mesa
Moffat
Morgan
Phillips
Rio Blanco
Routt
San Miguel
Weld
Yuma
Barber
Clark
Comanche
Finney
Grant
Gray
Harper
Haskell
Hodgeman
Kearny
Lane
Meade
Morton
Ness
Seward
Percent domestic use water
from ground watera'b
36.7
84.4
24.4
2.3
26.3
7.3
36.4
57.9
100.0
60.2
22.6
71.4
4.7
100.0
100.0
100.0
100.0
100.0
100.0
100.0
100.0
100.0
100.0
100.0
100.0
100.0
100.0
100.0
100.0
Percent domestic use
water self supplied3'0
28.5
40.7
11.3
0.8
16.0
6.2
25.8
4.9
25.3
32.5
5.9
32.5
0.7
38.1
19.0
24.2
19.2
2.1
23.8
36.4
10.3
35.2
42.3
14.6
24.1
25.4
21.7
24.2
15.7
                This document is a draft for review purposes only and does not constitute Agency policy.
June 2015                                          B-27                      DRAFT—DO NOT CITE OR QUOTE

-------
Hydraulic Fracturing Drinking Water Assessment
Appendix B
State
Kansas, cont.
Louisiana
Michigan
Cou nty
Sheridan
Stanton
Stevens
Sumner
Allen
Beau regard
Bienville
Bossier
Caddo
Calcasieu
Caldwell
Claiborne
De Soto
East Feliciana
Jackson
Lincoln
Natchitoches
Rapides
Red River
Sabine
Tangipahoa
Union
Webster
West Feliciana
Winn
Cheboygan
Gladwin
Kalkaska
Missaukee
Ogemaw
Roscommon
Percent domestic use water
from ground watera'b
100.0
100.0
100.0
51.3
100.0
100.0
100.0
29.4
12.2
98.3
100.0
100.0
55.8
100.0
100.0
100.0
23.2
100.0
83.2
67.5
100.0
100.0
100.0
100.0
100.0
100.0
100.0
100.0
100.0
100.0
100.0
Percent domestic use
water self supplied3'0
44.9
29.8
25.9
0.0
7.5
20.6
16.8
14.6
8.8
12.7
6.5
10.4
21.8
11.8
13.8
4.2
11.4
3.3
27.6
36.2
26.9
11.2
11.3
2.4
16.4
76.4
84.5
89.0
90.6
90.8
91.9
                This document is a draft for review purposes only and does not constitute Agency policy.
June 2015                                          B-28                      DRAFT—DO NOT CITE OR QUOTE

-------
Hydraulic Fracturing Drinking Water Assessment
Appendix B
State
Mississippi
Montana
New Mexico
North Dakota
Cou nty
Amite
Wilkinson
Daniels
Garfield
Glacier
Musselshell
Richland
Roosevelt
Rosebud
Sheridan
Chaves
Colfax
Eddy
Harding
Lea
Rio Arriba
Roosevelt
San Juan
Sandoval
Billings
Bottineau
Burke
Divide
Dunn
Golden Valley
Mckenzie
Mclean
Mountrail
Stark
Williams
Percent domestic use water
from ground watera'b
100.0
100.0
100.0
100.0
62.1
89.9
100.0
84.2
51.3
100.0
100.0
30.7
100.0
100.0
100.0
84.0
100.0
14.6
98.9
NA
100.0
100.0
100.0
100.0
100.0
75.8
12.5
65.7
NA
27.4
Percent domestic use
water self supplied3'0
26.0
11.1
29.4
70.0
17.7
54.5
30.8
20.9
10.3
31.0
11.8
2.6
2.2
25.0
17.4
42.3
8.9
12.9
23.2
33.3
13.7
12.5
12.5
21.4
7.7
15.7
9.9
11.5
5.7
7.3
                This document is a draft for review purposes only and does not constitute Agency policy.
June 2015                                          B-29                      DRAFT—DO NOT CITE OR QUOTE

-------
Hydraulic Fracturing Drinking Water Assessment
Appendix B
State
Ohio
Oklahoma
Cou nty
Ashland
Belmont
Carroll
Columbiana
Coshocton
Guernsey
Harrison
Jefferson
Knox
Medina
Muskingum
Noble
Portage
Stark
Tuscarawas
Wayne
Alfalfa
Beaver
Beckham
Blaine
Bryan
Caddo
Canadian
Carter
Coal
Custer
Dewey
Ellis
Garvin
Grady
Grant
Harper
Percent domestic use water
from ground watera'b
98.8
76.4
96.4
63.2
99.3
37.6
65.6
33.1
99.2
98.4
93.4
8.0
32.6
91.2
94.0
99.1
100.0
100.0
100.0
100.0
26.0
45.4
100.0
17.5
31.5
70.8
100.0
100.0
41.3
100.0
100.0
100.0
Percent domestic use
water self supplied3'0
57.4
8.9
76.4
43.2
34.9
9.5
45.9
10.2
41.1
83.1
17.0
8.0
18.3
30.9
23.5
49.0
14.6
47.9
10.6
8.8
7.8
35.1
0.0
0.5
27.5
13.2
22.5
31.4
15.8
34.2
13.2
22.6
                This document is a draft for review purposes only and does not constitute Agency policy.
June 2015                                          B-30                      DRAFT—DO NOT CITE OR QUOTE

-------
Hydraulic Fracturing Drinking Water Assessment
Appendix B
State
Oklahoma, cont.
Pennsylvania
Cou nty
Hughes
Jefferson
Johnston
Kay
Kingfisher
Kiowa
Latimer
Le Flore
Logan
Love
Major
Marshall
Mcclain
Noble
Oklahoma
Osage
Pawnee
Payne
Pittsburg
Roger Mills
Seminole
Stephens
Texas
Washita
Woods
Allegheny
Armstrong
Beaver
Blair
Bradford
Butler
Cameron
Percent domestic use water
from ground watera'b
23.6
13.5
53.4
39.2
100.0
10.3
12.6
14.3
61.1
100.0
100.0
20.1
95.9
23.3
22.0
18.0
38.2
47.9
0.6
80.1
78.8
99.2
100.0
53.9
100.0
15.7
45.3
54.7
34.9
100.0
51.8
29.0
Percent domestic use
water self supplied3'0
6.7
1.8
1.1
4.6
28.3
0.0
12.6
13.1
34.6
3.8
28.1
4.4
23.9
14.3
2.5
14.9
27.7
12.6
0.0
19.4
16.1
14.9
10.9
18.2
14.7
15.3
36.8
26.8
24.0
65.2
42.8
29.0
                This document is a draft for review purposes only and does not constitute Agency policy.
June 2015                                          B-31                      DRAFT—DO NOT CITE OR QUOTE

-------
Hydraulic Fracturing Drinking Water Assessment
Appendix B
State
Pennsylvania, cont.
Texas
Cou nty
Centre
Clarion
Clearfield
Clinton
Columbia
Crawford
Elk
Fayette
Forest
Greene
Huntingdon
Indiana
Jefferson
Lawrence
Lycoming
McKean
Potter
Somerset
Sullivan
Susquehanna
Tioga
Venango
Warren
Washington
Westmoreland
Wyoming
Andrews
Angelina
Archer
Atascosa
Austin
Bee
Percent domestic use water
from ground watera'b
93.1
61.5
38.4
48.4
77.5
97.7
25.3
19.2
100.0
31.9
73.2
52.2
60.7
40.5
60.0
56.6
93.7
42.6
100.0
79.9
81.3
95.9
96.9
21.6
21.3
100.0
100.0
100.0
16.9
100.0
100.0
100.0
Percent domestic use
water self supplied3'0
21.3
55.8
22.7
38.1
56.7
66.0
15.6
16.1
78.3
31.9
57.8
49.1
46.1
38.8
29.3
33.3
58.1
33.5
76.9
74.7
58.3
32.7
49.4
21.5
19.8
70.6
23.4
9.8
16.9
16.3
55.6
52.5
                This document is a draft for review purposes only and does not constitute Agency policy.
June 2015                                          B-32                      DRAFT—DO NOT CITE OR QUOTE

-------
Hydraulic Fracturing Drinking Water Assessment
Appendix B
State
Texas, cont.
Cou nty
Borden
Bosque
Brazos
Brooks
Burleson
Cherokee
Clay
Cochran
Coke
Colorado
Concho
Cooke
Cottle
Crane
Crockett
Crosby
Culberson
Dallas
Dawson
DeWitt
Denton
Dimmit
Ector
Edwards
Ellis
Erath
Fayette
Fisher
Franklin
Freestone
Frio
Gaines
Percent domestic use water
from ground watera'b
100.0
88.7
100.0
100.0
100.0
87.5
44.6
100.0
29.0
100.0
96.8
75.5
100.0
100.0
100.0
35.6
100.0
1.0
100.0
100.0
9.0
100.0
100.0
100.0
32.2
100.0
100.0
NA
0.9
100.0
100.0
100.0
Percent domestic use
water self supplied3'0
71.4
30.3
2.1
35.3
42.9
26.1
36.7
23.3
28.9
45.4
5.0
8.9
21.4
14.3
42.5
19.0
13.8
0.7
33.8
42.3
3.6
30.5
28.3
42.1
7.9
43.3
27.6
36.8
0.0
31.2
20.4
45.5
                This document is a draft for review purposes only and does not constitute Agency policy.
June 2015                                          B-33                      DRAFT—DO NOT CITE OR QUOTE

-------
Hydraulic Fracturing Drinking Water Assessment
Appendix B
State
Texas, cont.
Cou nty
Garza
Glasscock
Goliad
Gonzales
Grayson
Gregg
Grimes
Hansford
Hardeman
Hardin
Harrison
Hartley
Haskell
Hemphill
Hidalgo
Hockley
Hood
Houston
Howard
Hutchinson
Irion
Jack
Jefferson
Jim Hogg
Johnson
Jones
Karnes
Kenedy
Kent
King
Kleberg
Knox
Percent domestic use water
from ground watera'b
20.1
NA
NA
96.8
56.0
20.8
100.0
100.0
87.6
100.0
43.8
100.0
100.0
100.0
9.2
100.0
70.8
79.7
100.0
27.3
100.0
46.7
25.0
NA
34.9
60.5
100.0
100.0
100.0
100.0
100.0
86.2
Percent domestic use
water self supplied3'0
17.2
100.0
66.7
15.9
4.2
14.1
26.0
16.4
13.3
29.5
24.8
39.7
15.7
27.5
1.6
27.4
39.8
36.6
19.8
14.9
50.0
43.8
5.8
25.0
6.8
60.5
17.6
25.0
37.5
33.3
1.9
24.2
                This document is a draft for review purposes only and does not constitute Agency policy.
June 2015                                          B-34                     DRAFT—DO NOT CITE OR QUOTE

-------
Hydraulic Fracturing Drinking Water Assessment
Appendix B
State
Texas, cont.
Cou nty
La Salle
Lavaca
Lee
Leon
Liberty
Limestone
Lipscomb
Live Oak
Loving
Lynn
Madison
Marion
Martin
Maverick
McMullen
Medina
Menard
Midland
Milam
Mitchell
Montague
Montgomery
Moore
Nacogdoches
Navarro
Newton
Nolan
Nueces
Ochiltree
Oldham
Orange
Palo Pinto
Percent domestic use water
from ground watera'b
100.0
100.0
100.0
100.0
98.5
46.5
100.0
32.8
NA
64.1
100.0
13.7
100.0
27.6
100.0
98.0
36.4
100.0
82.5
100.0
57.1
100.0
100.0
55.6
22.0
100.0
100.0
5.6
100.0
100.0
99.1
11.7
Percent domestic use
water self supplied3'0
43.3
56.0
15.9
41.4
42.5
32.5
23.5
32.1
0.0
32.2
66.9
8.4
48.9
27.6
40.0
23.6
36.4
22.1
41.1
14.7
49.7
26.6
8.1
21.6
22.0
63.7
17.6
5.6
16.8
58.8
41.2
11.7
                This document is a draft for review purposes only and does not constitute Agency policy.
June 2015                                          B-35                      DRAFT—DO NOT CITE OR QUOTE

-------
Hydraulic Fracturing Drinking Water Assessment
Appendix B
State
Texas, cont.
Cou nty
Panola
Parker
Pecos
Polk
Potter
Reagan
Reeves
Roberts
Robertson
Runnels
Rusk
Sabine
San Augustine
San Patricio
Schleicher
Scurry
Shelby
Sherman
Smith
Somervell
Starr
Stephens
Sterling
Stonewall
Sutton
Tarrant
Terrell
Terry
Tyler
Upshur
Upton
Van Zandt
Percent domestic use water
from ground watera'b
96.6
63.5
100.0
41.9
100.0
100.0
100.0
100.0
97.1
13.5
90.7
76.2
78.0
88.8
100.0
32.5
66.2
100.0
48.0
87.7
23.2
13.5
NA
NA
100.0
3.7
100.0
100.0
100.0
54.1
100.0
65.7
Percent domestic use
water self supplied3'0
58.7
41.1
31.3
41.7
12.6
16.2
31.1
33.3
22.5
13.5
41.8
69.0
74.4
21.8
40.0
27.7
58.2
33.3
13.7
69.3
23.2
13.5
18.8
40.0
26.7
1.3
25.0
16.7
73.6
23.2
15.2
39.0
                This document is a draft for review purposes only and does not constitute Agency policy.
June 2015                                          B-36                     DRAFT—DO NOT CITE OR QUOTE

-------
Hydraulic Fracturing Drinking Water Assessment
Appendix B
State
Texas, cont.
Utah
Virginia
West Virginia
Cou nty
Walker
Waller
Ward
Washington
Webb
Wharton
Wheeler
Wichita
Wilbarger
Willacy
Wilson
Winkler
Wise
Wood
Yoakum
Young
Zapata
Zavala
Carbon
Duchesne
San Juan
Sevier
Uintah
Buchanan
Dickenson
Wise
Barbour
Brooke
Doddridge
Hancock
Harrison
Lewis
Percent domestic use water
from ground watera'b
57.7
100.0
100.0
48.2
99.4
100.0
100.0
8.8
100.0
28.4
100.0
100.0
51.3
21.3
100.0
19.3
13.9
100.0
50.0
57.1
68.3
100.0
87.7
NA
2.5
5.9
24.1
33.4
60.6
67.7
8.8
29.5
Percent domestic use
water self supplied3'0
30.6
37.2
4.5
36.0
0.5
45.9
31.3
2.9
11.5
28.4
6.9
3.8
50.4
12.9
36.0
18.9
13.9
15.2
1.2
10.4
47.5
10.0
3.1
27.6
2.5
2.3
24.8
6.8
62.1
6.9
8.9
30.3
                This document is a draft for review purposes only and does not constitute Agency policy.
June 2015                                          B-37                      DRAFT—DO NOT CITE OR QUOTE

-------
Hydraulic Fracturing Drinking Water Assessment
Appendix B
State
West Virginia, cont.
Wyoming
Cou nty
Marion
Marshall
Monongalia
Ohio
Pleasants
Preston
Ritchie
Taylor
Tyler
Upshur
Webster
Wetzel
Big Horn
Campbell
Carbon
Converse
Fremont
Goshen
Hot Springs
Johnson
Laramie
Lincoln
Natrona
Niobrara
Park
Percent domestic use water
from ground watera'b
5.8
96.5
5.3
5.4
100.0
66.1
45.2
14.9
44.4
27.3
41.9
96.3
79.4
100.0
63.8
96.5
49.3
100.0
31.9
40.8
38.1
82.4
69.0
100.0
18.9
Percent domestic use
water self supplied3'0
4.9
12.0
5.5
3.4
27.9
41.0
46.4
14.9
39.2
27.8
43.2
28.6
11.3
0.6
6.7
17.0
23.7
21.1
8.2
35.4
13.0
9.0
6.6
16.3
13.7
                This document is a draft for review purposes only and does not constitute Agency policy.
June 2015                                          B-38                      DRAFT—DO NOT CITE OR QUOTE

-------
Hydraulic Fracturing Drinking Water Assessment
Appendix B
State
Wyoming, cont.
Cou nty
Sublette
Sweetwater
Uinta
Washakie
Percent domestic use water
from ground watera'b
54.6
3.5
19.5
100.0
Percent domestic use
water self supplied3'0
22.1
0.4
11.5
16.0
a Data accessed from the USGS website (httg^/waterusgsj5oy/watuse/data/2010/) on November 11, 2014. Domestic water
use is water used for indoor household purposes such as drinking, food preparation, bathing, washing clothes and dishes,
flushing toilets, and outdoor purposes such as watering lawns and gardens (
b Percent domestic water use from ground water estimated with the following equation: (Domestic public supply volume from
ground water + Domestic self-supplied volume from ground water)/ Domestic total water use volume * 100. Domestic public
supply volume from ground water was estimated by multiplying the volume of domestic water from public supply by the ratio
of public supply volume from ground water to total public supply volume.
c Percent domestic water use self-supplied estimated by dividing the volume of domestic water self-supplied by total domestic
water use volume.
                This document is a draft for review purposes only and does not constitute Agency policy.

June 2015                                           B-39                      DRAFT—DO NOT CITE OR QUOTE

-------
Hydraulic Fracturing Drinking Water Assessment
                                                                                           Appendix B
Table B-7. Projected hydraulic fracturing water use by Texas counties between 2015 and 2060, expressed as a percentage of 2010
          total county water use.

          Hydraulic fracturing water use data from Nicot_et_aL_(20121. Total water use data from 2010 from the USGS Water Census (Mauein_et_aL-2Q14). All
          254 Texas counties are listed by descending order of percentages in 2030.
Texas county
McMullen
Irion
LaSalle
San Augustine
Sterling
Dimmit
Sabine
Leon
Karnes
Loving
Shackelford
Madison
Schleicher
Sutton
Shelby
DeWitt
Hemphill
Projected hydraulic fracturing water use as a percentage of 2010 total water usea'b
2015
126.2
36.1
58.4
60.2
12.0
38.2
9.6
9.9
48.1
13.1
0.0
5.5
10.5
0.0
11.0
26.9
25.7
2020
137.0
59.2
58.3
56.2
32.0
38.1
19.2
19.3
43.0
17.4
7.9
11.8
15.8
11.0
20.4
24.1
23.1
2025
152.1
70.5
59.7
52.2
39.9
38.9
28.7
27.0
37.9
23.4
15.7
15.7
19.1
15.1
19.4
21.4
20.5
2030
165.1
63.7
60.8
48.2
40.5
39.0
38.3
34.6
32.6
29.4
23.6
19.7
19.7
19.1
18.4
18.4
17.8
2035
176.7
53.4
61.9
44.2
41.0
38.7
35.1
32.9
27.2
28.8
21.2
17.4
17.1
23.2
17.4
15.4
15.2
2040
164.0
43.1
54.6
40.2
34.7
33.9
31.9
29.0
21.8
26.2
18.9
15.2
14.5
20.6
15.7
12.3
12.6
2045
145.3
32.8
45.3
36.2
28.3
27.9
28.7
25.1
16.4
23.6
16.5
13.0
11.9
18.1
14.1
9.3
10.0
2050
126.6
22.4
36.0
32.1
21.9
22.0
25.6
21.2
11.0
20.9
14.1
10.9
9.3
15.5
12.5
6.3
7.3
2055
108.0
12.1
26.7
28.1
15.6
16.0
22.3
17.3
5.6
18.3
11.8
8.7
6.7
12.9
10.9
3.2
4.7
2060
89.3
5.4
17.4
24.1
10.7
10.1
19.2
13.5
0.2
15.7
9.4
6.5
4.7
10.3
9.3
0.2
2.1
                   June 2015
This document is a draft for review purposes only and does not constitute Agency policy.

                                B-40                    DRAFT—DO NOT CITE OR QUOTE

-------
Hydraulic Fracturing Drinking Water Assessment
                                                                                                    Appendix B
Texas county
Terrell
Coryell
Montague
Crockett
Upton
Borden
Live Oak
Reagan
Clay
Wheeler
Lavaca
Washington
Nacogdoches
Hill
Jack
Panola
Jim Hogg
Howard
Parker
Hamilton
Johnson
Midland
Projected hydraulic fracturing water use as a percentage of 2010 total water usea'b
2015
0.0
7.0
28.6
7.6
12.1
3.1
13.3
11.2
3.2
17.6
7.9
0.0
7.9
17.1
3.5
7.2
4.8
4.4
3.7
8.8
14.2
6.7
2020
9.7
24.4
24.5
12.5
15.2
8.6
12.4
14.0
5.9
15.3
13.2
6.7
11.4
14.7
5.3
10.2
6.4
7.1
5.0
10.7
11.9
8.3
2025
13.2
22.8
20.4
14.8
14.1
12.0
11.5
12.7
8.6
13.1
12.0
11.8
10.7
12.2
7.1
9.2
8.0
8.5
6.3
8.9
9.5
7.7
2030
16.8
16.5
16.3
13.4
12.9
12.1
11.8
11.3
11.3
10.8
10.7
10.7
10.0
9.8
8.8
8.5
8.0
8.0
7.6
7.1
7.1
7.1
2035
20.4
10.1
12.2
11.2
11.7
12.2
12.2
9.9
10.3
8.6
9.4
9.6
9.2
7.3
7.9
7.7
6.9
6.8
6.8
5.3
4.7
6.2
2040
18.2
3.8
8.2
9.1
9.8
10.3
12.7
8.1
9.4
6.3
8.1
8.6
8.3
4.9
7.1
7.0
6.0
5.6
6.1
3.5
2.4
5.2
2045
15.9
0.0
4.1
6.9
7.9
8.4
13.2
6.4
8.4
4.1
6.7
7.5
7.5
2.4
6.2
6.3
4.9
4.4
5.3
1.8
0.0
4.1
2050
13.6
0.0
0.0
4.7
5.9
6.4
11.7
4.6
7.5
1.8
5.4
6.4
6.6
0.0
5.3
5.5
3.9
3.2
4.5
0.0
0.0
3.0
2055
11.3
0.0
0.0
2.5
4.0
4.5
9.8
2.8
6.6
0.0
4.0
5.3
5.7
0.0
4.4
4.8
2.9
2.1
3.8
0.0
0.0
2.0
2060
9.0
0.0
0.0
1.1
2.7
3.1
7.8
1.6
5.6
0.0
2.7
4.3
4.9
0.0
3.5
4.0
1.8
1.3
3.0
0.0
0.0
1.2
                    June 2015
This document is a draft for review purposes only and does not constitute Agency policy.
                                   B-41                     DRAFT—DO NOT CITE OR QUOTE

-------
Hydraulic Fracturing Drinking Water Assessment
                                                                                                    Appendix B
Texas county
Kenedy
Fayette
Lee
Winkler
Wilson
Martin
Burleson
Atascosa
Bosque
Webb
Gonzales
Marion
Harrison
Eastland
Archer
Zavala
Roberts
Maverick
Cooke
Ward
Austin
Reeves
Projected hydraulic fracturing water use as a percentage of 2010 total water usea'b
2015
4.1
3.9
2.1
2.9
6.7
5.7
1.0
6.3
1.8
7.5
8.0
1.1
4.3
0.0
1.0
4.7
6.9
2.5
11.9
2.7
0.0
1.4
2020
5.4
8.4
4.1
3.8
7.7
7.1
2.9
5.7
3.0
7.1
7.1
2.4
6.1
3.9
2.4
5.5
6.0
3.0
9.3
3.2
1.2
1.8
2025
6.8
7.6
5.3
5.1
7.0
6.5
4.3
5.6
4.3
6.3
6.2
3.8
5.5
5.9
3.6
5.2
5.1
3.6
6.7
4.2
2.5
2.7
2030
6.8
6.6
6.5
6.3
6.2
6.0
5.7
5.6
5.5
5.4
5.3
5.1
5.1
5.0
4.9
4.9
4.2
4.2
4.1
4.1
3.7
3.7
2035
5.9
5.5
5.8
6.0
5.4
5.3
5.1
5.6
5.1
4.6
4.4
5.2
4.6
4.2
4.5
4.6
3.4
4.8
1.5
4.0
3.4
3.9
2040
5.1
4.4
5.1
5.4
4.6
4.4
4.5
5.6
4.6
3.8
3.6
4.7
4.2
3.3
4.1
4.3
2.5
4.5
0.0
3.6
3.0
3.6
2045
4.1
3.4
4.3
4.7
3.9
3.5
3.9
5.0
4.2
3.1
2.7
4.2
3.7
2.5
3.7
4.0
1.6
4.0
0.0
3.2
2.6
3.3
2050
3.3
2.3
3.6
4.1
3.1
2.6
3.3
4.2
3.7
2.3
1.8
3.7
3.3
1.7
3.3
3.4
0.7
3.6
0.0
2.7
2.2
3.0
2055
2.4
1.2
2.9
3.4
2.3
1.8
2.6
3.4
3.2
1.4
0.9
3.2
2.9
0.8
2.9
2.7
0.0
3.1
0.0
2.3
1.9
2.6
2060
1.6
0.2
2.1
2.8
1.5
1.2
2.0
2.7
2.8
0.5
0.0
2.7
2.4
0.0
2.5
2.0
0.0
2.6
0.0
1.9
1.5
2.3
                    June 2015
This document is a draft for review purposes only and does not constitute Agency policy.
                                   B-42                     DRAFT—DO NOT CITE OR QUOTE

-------
Hydraulic Fracturing Drinking Water Assessment
                                                                                                    Appendix B
Texas county
Glasscock
Tyler
Hood
Garza
Andrews
Crane
Erath
Wise
Upshur
Mitchell
Ector
Culberson
Lipscomb
Angelina
Houston
Frio
Newton
Kleberg
Brooks
Brazos
Comanche
Ochiltree
Projected hydraulic fracturing water use as a percentage of 2010 total water usea'b
2015
3.1
1.9
1.4
1.5
2.3
1.3
0.9
3.6
0.2
1.2
1.5
0.3
1.7
0.4
2.1
1.8
1.8
1.0
1.0
0.4
0.4
0.6
2020
4.1
2.6
2.0
2.0
3.0
1.7
1.4
3.2
0.9
1.6
2.0
0.4
3.0
0.9
2.7
1.8
2.3
1.4
1.3
0.9
0.7
1.1
2025
3.9
3.2
2.6
2.5
2.9
2.1
1.9
2.8
1.7
2.0
2.1
1.3
2.6
1.5
2.4
1.9
2.1
1.7
1.7
1.2
1.0
1.5
2030
3.6
3.2
3.2
2.9
2.7
2.6
2.4
2.4
2.4
2.4
2.3
2.2
2.1
2.1
2.1
1.9
1.8
1.7
1.7
1.5
1.4
1.2
2035
3.1
2.8
2.9
2.7
2.6
3.1
2.2
2.0
2.9
2.1
2.2
2.9
1.7
2.2
1.8
1.8
1.6
1.5
1.5
1.4
1.2
1.0
2040
2.6
2.4
2.6
2.4
2.3
2.8
2.0
1.6
2.6
1.9
1.9
2.6
1.3
2.0
1.5
1.8
1.3
1.3
1.2
1.2
1.1
0.7
2045
2.1
2.0
2.2
2.1
2.0
2.5
1.8
1.2
2.3
1.7
1.7
2.4
0.8
1.8
1.2
1.7
1.0
1.1
1.0
1.0
1.0
0.5
2050
1.5
1.6
1.9
1.8
1.7
2.2
1.6
0.8
2.1
1.4
1.4
2.1
0.4
1.6
0.9
1.5
0.8
0.8
0.8
0.8
0.8
0.2
2055
1.0
1.1
1.6
1.5
1.4
1.9
1.4
0.4
1.8
1.2
1.2
1.9
0.0
1.4
0.6
1.2
0.5
0.6
0.6
0.7
0.7
0.0
2060
0.7
0.7
1.3
1.2
1.1
1.7
1.2
0.0
1.5
0.9
1.0
1.6
0.0
1.2
0.3
0.9
0.3
0.4
0.4
0.5
0.5
0.0
                    June 2015
This document is a draft for review purposes only and does not constitute Agency policy.
                                   B-43                      DRAFT—DO NOT CITE OR QUOTE

-------
Hydraulic Fracturing Drinking Water Assessment
                                                                                                    Appendix B
Texas county
Palo Pinto
Limestone
Duval
Stephens
Dawson
Scurry
Bee
Val Verde
Colorado
Tarrant
Zapata
Ellis
Jim Wells
Lynn
Henderson
Hansford
Gaines
Gregg
Refugio
Caldwell
Pecos
Anderson
Projected hydraulic fracturing water use as a percentage of 2010 total water usea'b
2015
0.3
0.9
0.7
0.1
0.5
0.0
0.8
0.0
<0.1
2.1
0.5
0.3
0.4
0.0
0.1
0.0
0.2
0.1
0.2
0.4
0.1
0.1
2020
0.6
1.0
0.9
0.4
0.8
0.6
1.1
0.5
0.3
1.7
0.7
0.5
0.6
0.4
0.3
0.4
0.3
0.2
0.3
0.5
0.1
0.2
2025
0.9
1.1
1.1
0.8
1.0
0.8
1.1
0.8
0.6
1.3
0.8
0.6
0.7
0.6
0.5
0.8
0.5
0.3
0.4
0.4
0.2
0.3
2030
1.2
1.2
1.1
1.1
1.1
1.0
1.0
0.9
0.9
0.9
0.8
0.8
0.7
0.7
0.7
0.7
0.5
0.4
0.4
0.4
0.4
0.4
2035
1.1
1.1
1.0
1.0
1.1
1.2
0.9
1.1
0.8
0.4
0.7
0.7
0.6
0.8
0.8
0.5
0.5
0.4
0.3
0.3
0.5
0.4
2040
1.0
1.0
0.8
0.9
1.0
1.1
0.7
1.0
0.7
0.0
0.6
0.6
0.5
0.8
0.7
0.4
0.4
0.4
0.3
0.3
0.4
0.4
2045
0.8
0.8
0.7
0.8
0.8
0.9
0.6
0.9
0.6
0.0
0.5
0.6
0.4
0.7
0.6
0.3
0.4
0.4
0.2
0.2
0.4
0.4
2050
0.7
0.7
0.5
0.6
0.6
0.8
0.4
0.8
0.5
0.0
0.4
0.5
0.4
0.6
0.5
0.2
0.3
0.3
0.2
0.2
0.3
0.3
2055
0.6
0.6
0.4
0.5
0.5
0.7
0.3
0.6
0.4
0.0
0.3
0.4
0.3
0.5
0.4
0.1
0.2
0.3
0.1
0.1
0.3
0.3
2060
0.5
0.4
0.3
0.4
0.3
0.5
0.1
0.5
0.4
0.0
0.2
0.3
0.2
0.4
0.4
0
0.2
0.2
0.1
0.1
0.2
0.2
                    June 2015
This document is a draft for review purposes only and does not constitute Agency policy.
                                   B-44                     DRAFT—DO NOT CITE OR QUOTE

-------
Hydraulic Fracturing Drinking Water Assessment
                                                                                                    Appendix B
Texas county
Young
San Patricio
Smith
Cherokee
McLennan
Terry
Starr
Cochran
Jasper
Dallas
Robertson
Grimes
Yoakum
Freestone
Cass
Hutchinson
Rusk
Willacy
Victoria
Sherman
Calhoun
Lubbock
Projected hydraulic fracturing water use as a percentage of 2010 total water usea'b
2015
0.0
0.2
0.1
0.1
0.1
0.0
0.2
0.1
0.2
0.2
0.1
<0.1
0.1
0.1
<0.1
0.0
<0.1
<0.1
<0.1
0.0
<0.1
0.0
2020
0.1
0.3
0.1
0.2
0.1
0.2
0.2
0.2
0.3
0.3
0.2
0.1
0.1
0.1
0.1
0.1
0.1
0.1
0.1
0.0
0.1
0.0
2025
0.2
0.4
0.2
0.2
0.2
0.2
0.3
0.2
0.2
0.2
0.2
0.1
0.2
0.1
0.1
0.2
0.1
0.1
0.1
<0.1
0.1
<0.1
2030
0.4
0.4
0.3
0.3
0.3
0.3
0.3
0.2
0.2
0.2
0.2
0.2
0.2
0.2
0.2
0.1
0.1
0.1
0.1
0.1
0.1
0.1
2035
0.3
0.3
0.4
0.4
0.3
0.3
0.2
0.3
0.2
0.1
0.2
0.1
0.1
0.2
0.2
0.1
0.1
0.1
0.1
0.1
0.1
0.1
2040
0.3
0.3
0.3
0.3
0.2
0.3
0.2
0.2
0.1
0.1
0.1
0.1
0.1
0.1
0.2
0.1
0.1
0.1
0.1
0.1
0.1
0.1
2045
0.3
0.2
0.3
0.3
0.2
0.3
0.2
0.2
0.1
<0.1
0.1
0.1
0.1
0.1
0.1
0.1
0.1
0.1
<0.1
<0.1
<0.1
0.1
2050
0.2
0.2
0.3
0.2
0.2
0.2
0.1
0.2
0.1
0.0
0.1
0.1
0.1
0.1
0.1
<0.1
0.1
<0.1
<0.1
<0.1
<0.1
0.1
2055
0.2
0.1
0.2
0.2
0.2
0.2
0.1
0.2
0.1
0.0
0.1
0.1
0.1
0.1
0.1
<0.1
0.1
<0.1
<0.1
<0.1
<0.1
0.1
2060
0.1
0.1
0.2
0.2
0.1
0.2
0.1
0.1
<0.1
0.0
0.1
0.1
0.1
0.1
0.1
0.0
<0.1
<0.1
<0.1
<0.1
<0.1
0.1
                    June 2015
This document is a draft for review purposes only and does not constitute Agency policy.
                                   B-45                     DRAFT—DO NOT CITE OR QUOTE

-------
Hydraulic Fracturing Drinking Water Assessment
                                                                                                    Appendix B
Texas county
Jackson
Matagorda
Polk
Wharton
Nueces
Hidalgo
Cameron
Somervell
Goliad
Brazoria
Fort Bend
Aransas
Armstrong
Bailey
Bandera
Bastrop
Baylor
Bell
Bexar
Blanco
Bowie
Brewster
Projected hydraulic fracturing water use as a percentage of 2010 total water usea'b
2015
<0.1
<0.1
<0.1
<0.1
<0.1
<0.1
<0.1
<0.1
<0.1
<0.1
<0.1
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
2020
<0.1
<0.1
<0.1
<0.1
<0.1
<0.1
<0.1
<0.1
<0.1
<0.1
<0.1
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
2025
0.1
<0.1
<0.1
<0.1
<0.1
<0.1
<0.1
<0.1
<0.1
<0.1
<0.1
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
2030
0.1
<0.1
<0.1
<0.1
<0.1
<0.1
<0.1
<0.1
<0.1
<0.1
<0.1
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
2035
<0.1
<0.1
<0.1
<0.1
<0.1
<0.1
<0.1
<0.1
<0.1
<0.1
<0.1
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
2040
<0.1
<0.1
<0.1
<0.1
<0.1
<0.1
<0.1
<0.1
<0.1
<0.1
<0.1
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
2045
<0.1
<0.1
<0.1
<0.1
<0.1
<0.1
<0.1
<0.1
<0.1
<0.1
<0.1
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
2050
<0.1
<0.1
<0.1
<0.1
<0.1
<0.1
<0.1
<0.1
<0.1
<0.1
<0.1
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
2055
<0.1
<0.1
<0.1
<0.1
<0.1
<0.1
<0.1
<0.1
<0.1
<0.1
<0.1
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
2060
<0.1
<0.1
<0.1
<0.1
<0.1
<0.1
<0.1
<0.1
<0.1
<0.1
<0.1
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
                    June 2015
This document is a draft for review purposes only and does not constitute Agency policy.
                                   B-46                     DRAFT—DO NOT CITE OR QUOTE

-------
Hydraulic Fracturing Drinking Water Assessment
                                                                                                    Appendix B
Texas county
Briscoe
Brown
Burnet
Callahan
Camp
Carson
Castro
Chambers
Childress
Coke
Coleman
Collin
Collingsworth
Comal
Concho
Cottle
Crosby
Dallam
Deaf Smith
Delta
Denton
Dickens
Projected hydraulic fracturing water use as a percentage of 2010 total water usea'b
2015
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
1.7
0.0
2020
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
1.1
0.0
2025
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.6
0.0
2030
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
2035
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
2040
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
2045
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
2050
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
2055
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
2060
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
                    June 2015
This document is a draft for review purposes only and does not constitute Agency policy.
                                   B-47                     DRAFT—DO NOT CITE OR QUOTE

-------
Hydraulic Fracturing Drinking Water Assessment
                                                                                                    Appendix B
Texas county
Donley
Edwards
El Paso
Falls
Fannin
Fisher
Floyd
Foard
Franklin
Galveston
Gillespie
Gray
Grayson
Guadalupe
Hale
Hall
Hardeman
Hardin
Harris
Hartley
Haskell
Hays
Projected hydraulic fracturing water use as a percentage of 2010 total water usea'b
2015
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
2020
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
2025
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
2030
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
2035
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
2040
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
2045
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
2050
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
2055
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
2060
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
                    June 2015
This document is a draft for review purposes only and does not constitute Agency policy.
                                   B-48                     DRAFT—DO NOT CITE OR QUOTE

-------
Hydraulic Fracturing Drinking Water Assessment
                                                                                                    Appendix B
Texas county
Hockley
Hopkins
Hudspeth
Hunt
Jeff Davis
Jefferson
Jones
Kaufman
Kendall
Kent
Kerr
Kimble
King
Kinney
Knox
Lamar
Lamb
Lampasas
Liberty
Llano
McCulloch
Mason
Projected hydraulic fracturing water use as a percentage of 2010 total water usea'b
2015
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
2020
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
2025
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
2030
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
2035
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
2040
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
2045
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
2050
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
2055
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
2060
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
                    June 2015
This document is a draft for review purposes only and does not constitute Agency policy.
                                   B-49                     DRAFT—DO NOT CITE OR QUOTE

-------
Hydraulic Fracturing Drinking Water Assessment
                                                                                                    Appendix B
Texas county
Medina
Menard
Milam
Mills
Montgomery
Moore
Morris
Motley
Navarro
Nolan
Oldham
Orange
Parmer
Potter
Presidio
Rains
Randall
Real
Red River
Rockwall
Runnels
San Jacinto
Projected hydraulic fracturing water use as a percentage of 2010 total water usea'b
2015
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
2020
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
2025
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
2030
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
2035
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
2040
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
2045
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
2050
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
2055
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
2060
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
                    June 2015
This document is a draft for review purposes only and does not constitute Agency policy.
                                   B-50                     DRAFT—DO NOT CITE OR QUOTE

-------
Hydraulic Fracturing Drinking Water Assessment
                                                                                                          Appendix B
Texas county
San Saba
Stonewall
Swisher
Taylor
Throckmorton
Titus
Tom Green
Travis
Trinity
Uvalde
Van Zandt
Walker
Waller
Wichita
Wilbarger
Williamson
Wood
Projected hydraulic fracturing water use as a percentage of 2010 total water usea'b
2015
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
2020
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
2025
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
2030
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
2035
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
2040
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
2045
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
2050
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
2055
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
2060
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
 a Total water use data accessed from the USGS website (http://water.usgs.gov/watuse/data/2010/) on April 21, 2015. Data from Nicotetal. 12012) transcribed.
 b Percentages calculated by dividing projected hydraulic fracturing water use volumes from Nicot et al. 12012) by 2010 total water use from the USGS and multiplying by 100.
 Percentages less than 0.1 were not rounded and simply noted as "<0.1", but where the percentage was actually zero because there was no projected hydraulic fracturing
 water use we noted that as "0.0".
                     June 2015
This document is a draft for review purposes only and does not constitute Agency policy.

                                     B-51                      DRAFT—DO NOT CITE OR QUOTE

-------
Hydraulic Fracturing Drinking Water Assessment                                                 Appendix B
B.2.   References for Appendix B

CCST (California Council on Science and Technology). (2014). Advanced well stimulation technologies in
   California: An independent review of scientific and technical information. Sacramento, CA.
   http://ccst.us/publications/2014/2014wst.pdf

Goodwin, S: Carlson, K: Knox, K: Douglas, C: Rein, L. (2014). Water intensity assessment of shale gas resources
   in the Wallenberg field in norlheaslern Colorado. Environ Sci Technol 48: 5991-5995.
   hllp://dx.doi.org/10.1021/es404675h

Hansen. E: Mulvaney. D: Belcher. M. (2013). Waler resource reporting and water foolprinl from Marcellus
   Shale developmenl in Wesl Virginia and Pennsylvania. Durango, CO: Earlhworks Oil & Gas Accountability
   Project hllp://www.downslreamslralegies.com/documenls/reporls  publicalion/marcellus  wv pa.pdf

Maupin. MA: Kenny. IF: Hulson. SS: Lovelace. IK: Barber. NL: Linsey. KS. (2014). Estimated use of water in Ihe
   United Stales in 2010. (USGS Circular 1405). Reslon, VA: U.S. Geological Survey.
   hllp://dx.doi.org/10.3133/cirl405

Milchell. AL: Small. M: Gasman. EA. (2013). Surface water wilhdrawals for Marcellus Shale gas developmenl:
   performance of alternative regulatory approaches in Ihe Upper Ohio River Basin. Environ Sci Technol 47:
   12669-12678. hllp://dx.doi.org/10.1021/es403537z

Murray. KE. (2013). Stale-scale perspective on water use and production associated wilh oil and gas
   operations, Oklahoma, U.S. Environ Sci Technol 47: 4918-4925. hllp://dx.doi.org/10.1021/es4000593

Nicol. IP: Reedy. RC: Coslley. RA: Huang. Y. (2012). Oil & gas water use in Texas: Update to Ihe 2011 mining
   water use report Nicol, JP; Reedy, RC; Coslley, RA; Huang, Y.
   hllp://www.lwdb.slale.lx.us/publicalions/reporls/conlracled reporls/doc/0904830939 2012Updale M
   iningWalerUse.pdf

Nicol, IP: Scanlon, BR. (2012). Waler use for shale-gas production in Texas, U.S. Environ Sci Technol 46:  3580-
   3586. hllp://dx.doi.org/10.1021/es204602l

Norlh Dakota Slate Waler Commission. (2014). Facls aboul Norlh Dakota (racking and water use. Bismarck,
   ND. hllp://www.swc.nd.gov/4dlink9/4dcgi/GelConlenlPDF/PB-2419/Facl%20Sheetpdf

Solley. WB: Pierce. RR: Perlman. HA. (1998). Estimated use of water in Ihe United Slates in 1995. (USGS
   Circular: 1200). U.S. Geological Survey, http://pubs.er.usgs.gov/publicalion/cirl200

U.S. EPA (U.S. Environmental Protection Agency). (2015a). Analysis of hydraulic fracluring fluid data from Ihe
   FracFocus chemical disclosure regislry 1.0 [EPA Report]. (EPA/601/R-14/003). Washington, D.C.: Office
   of Research and Developmenl, U.S. Environmental Protection Agency.
   http://www2.epa.gov/hfsludy/analysis-hydraulic-fracluring-fluid-dala-fracfocus-chemical-disclosure-
   regislry-1-pdf

U.S. EPA (U.S. Environmental Protection Agency). (2015b). Analysis of hydraulic fracluring fluid data from Ihe
   FracFocus chemical disclosure regislry 1.0: Data managemenl and quality assessmenl report [EPA
   Report]. (EPA/601/R-14/006). Washington, D.C.: U.S. Environmental Protection Agency, Office of
   Research and Developmenl. http://www2.epa.gov/siles/produclion/files/2015-
   03/documenls/fracfocus date managemenl report final 032015 508.pdf

U.S. EPA (U.S. Environmental Protection Agency). (2015c). Analysis of hydraulic fracluring fluid data from Ihe
   FracFocus chemical disclosure regislry 1.0: Projecl database [EPA Report]. (EPA/601/R-14/003).
   Washington, D.C.: U.S. Environmental Protection Agency, Office of Research and Developmenl.
   http://www2.epa.gov/hfsludy/epa-projecl-dalabase-developed-fracfocus-l-disclosures

USGS (U.S. Geological Survey). (2014). Wilhdrawal and consumption of water by Ihermoeleclric power planls
   in Ihe United Stales, 2010. (Scientific Investigations Report 20145184). Reslon, VA.
   http://dx.doi.org/10.3133/sir20145184
              This document is a draft for review purposes only and does not constitute Agency policy.

June 2015                                      B-52                   DRAFT—DO NOT CITE OR QUOTE

-------
Hydraulic Fracturing Drinking Water Assessment                                                    Appendix B
                        HL (2009). Water withdrawal for brackish and inland aquaculture, and options to
   produce more fish in ponds with present water use. Water Policy 11: 52-68.
               This document is a draft for review purposes only and does not constitute Agency policy.

June 2015                                         B-53                     DRAFT—DO NOT CITE OR QUOTE

-------
Hydraulic Fracturing Drinking Water Assessment                              Appendix C
                      Appendix C
Chemical Mixing Supplemental Tables and
Information
         This document is a draft for review purposes only and does not constitute Agency policy.

June 2015                                     DRAFT—DO NOT CITE OR QUOTE

-------
Hydraulic Fracturing Drinking Water Assessment
Appendix C
Appendix C.  Chemical Mixing Supplemental Tables and
   Information
C.I.   Supplemental Tables and Information
Table C-l. Chemicals reported to FracFocus in 10% or more of disclosures for gas-producing
        wells, with the number of disclosures where chemical is reported, percentage of
        disclosures, and the median maximum concentration (% by mass) of that chemical
        in hydraulic fracturing fluid.
        Chemicals ranked by frequency of occurrence (y^
Chemical name
Hydrochloric acid
Methanol
Distillates, petroleum, hydrotreated light
Isopropanol
Water
Ethanol
Propargyl alcohol
Glutaraldehyde
Ethylene glycol
Citric acid
Sodium hydroxide
Peroxydisulfuric acid, diammonium salt
Quartz
2,2-Dibromo-3-nitrilopropionamide
Sodium chloride
Guargum
Acetic acid
2-Butoxyethanol
Naphthalene
Solvent naphtha, petroleum, heavy arom.
Quaternary ammonium compounds,
benzyl-C12-16-alkyldimethyl, chlorides
Potassium hydroxide
Ammonium chloride
Choline chloride
CASRN
7647-01-0
67-56-1
64742-47-8
67-63-0
7732-18-5
64-17-5
107-19-7
111-30-8
107-21-1
77-92-9
1310-73-2
7727-54-0
14808-60-7
10222-01-2
7647-14-5
9000-30-0
64-19-7
111-76-2
91-20-3
64742-94-5
68424-85-1
1310-58-3
12125-02-9
67-48-1
Number of
disclosures
12,351
12,269
11,897
8,008
7,998
6,325
5,811
5,635
5,493
4,832
4,656
4,618
3,758
3,668
3,608
3,586
3,563
3,325
3,294
3,287
3,259
2,843
2,483
2,477
Percentage of
disclosures
72.8%
72.3%
70.1%
47.2%
47.1%
37.3%
34.2%
33.2%
32.4%
28.5%
27.4%
27.2%
22.1%
21.6%
21.3%
21.1%
21.0%
19.6%
19.4%
19.4%
19.2%
16.8%
14.6%
14.6%
Median maximum
concentration in
hydraulic fracturing
fluid (% by mass)
15%
30%
30%
30%
63%
5%
10%
30%
35%
60%
5%
100%
10%
100%
30%
60%
50%
10%
5%
30%
7%
15%
10%
75%
            This document is a draft for review purposes only and does not constitute Agency policy.
June 2015                               C-l                DRAFT—DO NOT CITE OR QUOTE

-------
Hydraulic Fracturing Drinking Water Assessment
Appendix C
Chemical name
Poly(oxy-l,2-ethanediyl)-nonylphenyl-
hydroxy (mixture)
Sodium chlorite
1,2,4-Trimethylbenzene
Carbonic acid, dipotassium salt
Methenamine
Formic acid
Didecyl dimethyl ammonium chloride
N,N-Dimethylformamide
Phenolic resin
Thiourea polymer
Polyethylene glycol
CASRN
127087-87-0
7758-19-2
95-63-6
584-08-7
100-97-0
64-18-6
7173-51-5
68-12-2
9003-35-4
68527-49-1
25322-68-3
Number of
disclosures
2,455
2,372
2,229
2,154
2,134
2,118
2,063
1,892
1,852
1,702
1,696
Percentage of
disclosures
14.5%
14.0%
13.1%
12.7%
12.6%
12.5%
12.2%
11.2%
10.9%
10.0%
10.0%
Median maximum
concentration in
hydraulic fracturing
fluid (% by mass)
5%
10%
1%
60%
1%
60%
10%
13%
5%
30%
60%
 Note: Analysis considered 17,035 disclosures and 291,363 ingredient records that met selected quality assurance criteria,
 including: completely parsed; unique combination of fracture date and API well number; fracture date between January 1,
 2011, and February 28, 2013; valid CASRN; and valid concentrations. Disclosures that did not meet quality assurance criteria
 (1,587) or other, query-specific criteria were excluded from analysis.
                 This document is a draft for review purposes only and does not constitute Agency policy.
June 2015                                              C-2                        DRAFT—DO NOT CITE OR QUOTE

-------
Hydraulic Fracturing Drinking Water Assessment
Appendix C
Table C-2. Chemicals reported to FracFocus in 10% or more of disclosures for oil-producing
          wells, with the number of disclosures where chemical is reported, percentage of
          disclosures, and the median maximum concentration (% by mass) of that chemical
          in hydraulic fracturing fluid.
          Chemicals ranked by frequency of occurrence (LLS._EPAI_2015c).
Chemical name
Methanol
Distillates, petroleum, hydrotreated light
Peroxydisulfuric acid, diammonium salt
Ethylene glycol
Hydrochloric acid
Guar gum
Sodium hydroxide
Quartz
Water
Isopropanol
Potassium hydroxide
Glutaraldehyde
Propargyl alcohol
Acetic acid
2-Butoxyethanol
Solvent naphtha, petroleum, heavy arom.
Sodium chloride
Ethanol
Citric acid
Phenolic resin
Naphthalene
Nonyl phenol ethoxylate
Diatomaceous earth, calcined
Methenamine
Tetramethylammonium chloride
Carbonic acid, dipotassium salt
Ethoxylated propoxylated C12-14 alcohols
Choline chloride
Boron sodium oxide
Tetrakis(hydroxymethyl)phosphonium sulfate
1,2,4-Trimethylbenzene
CASRN
67-56-1
64742-47-8
7727-54-0
107-21-1
7647-01-0
9000-30-0
1310-73-2
14808-60-7
7732-18-5
67-63-0
1310-58-3
111-30-8
107-19-7
64-19-7
111-76-2
64742-94-5
7647-14-5
64-17-5
77-92-9
9003-35-4
91-20-3
9016-45-9
91053-39-3
100-97-0
75-57-0
584-08-7
68439-51-0
67-48-1
1330-43-4
55566-30-8
95-63-6
Number of
disclosures
12,484
10,566
10,350
10,307
10,029
9,110
8,609
8,577
8,538
8,031
7,206
5,927
5,599
4,623
4,022
3,821
3,692
3,536
3,310
3,109
3,060
2,829
2,655
2,559
2,428
2,402
2,342
2,264
2,228
2,130
2,118
Percentage of
disclosures
71.8%
60.8%
59.6%
59.3%
57.7%
52.4%
49.5%
49.4%
49.1%
46.2%
41.5%
34.1%
32.2%
26.6%
23.1%
22.0%
21.2%
20.3%
19.0%
17.9%
17.6%
16.3%
15.3%
14.7%
14.0%
13.8%
13.5%
13.0%
12.8%
12.3%
12.2%
Median maximum
concentration in
hydraulic fracturing
fluid (% by mass)
30%
40%
100%
30%
15%
50%
10%
2%
67%
15%
15%
15%
5%
30%
10%
5%
25%
45%
60%
5%
5%
20%
100%
1%
1%
60%
2%
75%
30%
50%
1%
              This document is a draft for review purposes only and does not constitute Agency policy.
June 2015                                     C-3                   DRAFT—DO NOT CITE OR QUOTE

-------
Hydraulic Fracturing Drinking Water Assessment
Appendix C
Chemical name
Boric acid
Polyethylene glycol
2-Mercaptoethanol
2,2-Dibromo-3-nitrilopropionamide
Formic acid
Sodium persulfate
Phosphonicacid
Sodium tetraborate decahydrate
Potassium metaborate
Ethylenediaminetetraacetic acid tetrasodium
salt hydrate
Poly(oxy-l,2-ethanediyl)-nonylphenyl-hydroxy
(mixture)
CASRN
10043-35-3
25322-68-3
60-24-2
10222-01-2
64-18-6
7775-27-1
13598-36-2
1303-96-4
13709-94-9
64-02-8
127087-87-0
Number of
disclosures
2,070
2,025
2,012
1,988
1,948
1,914
1,865
1,862
1,682
1,676
1,668
Percentage of
disclosures
11.9%
11.7%
11.6%
11.4%
11.2%
11.0%
10.7%
10.7%
9.7%
9.6%
9.6%
Median maximum
concentration in
hydraulic fracturing
fluid (% by mass)
25%
5%
100%
98%
60%
100%
1%
30%
60%
0%
5%
 Note: Analysis considered 17,640 disclosures and 385,013 ingredient records that met selected quality assurance criteria,
 including: completely parsed; unique combination of fracture date and API well number; fracture date between January 1,
 2011, and February 28, 2013; valid CASRN; and valid concentrations. Disclosures that did not meet quality assurance criteria
 (2,268) or other, query-specific criteria were excluded from analysis.
                 This document is a draft for review purposes only and does not constitute Agency policy.

June 2015                                              C-4                       DRAFT—DO NOT CITE OR QUOTE

-------
Hydraulic Fracturing Drinking Water Assessment
                                                                                           Appendix C
Table C-3a. Top chemicals reported to FracFocus for each state and number (and percentage) of disclosures where a chemical is
          reported for that state, Alabama to Montana (U.S. EPA, 2015c).

          Source: (U1S._EPAt_2015c). The top 20 most frequent chemicals were identified for the 20 states that reported to FracFocus, resulting in a total of
          93 chemicals. The chemicals were ranked by counting the number of states where that chemical was in the top 20; chemicals used most widely
          among the most states come first. For example, methanol is reported in 19 of 20 states, so methanol is ranked first.
Chemical name
Methanol
Distillates, petroleum,
hydrotreated light
Ethylene glycol
Isopropanol
Quartz
Sodium hydroxide
Ethanol
Guargum
Hydrochloric acid
Peroxydisulfuric acid,
diammonium salt
CASRN
67-56-1
64742-47-8
107-21-1
67-63-0
14808-60-7
1310-73-2
64-17-5
9000-30-0
7647-01-0
7727-54-0
Alabama
55
(100%)

55
(100%)
55
(100%)




55
(100%)

Alaska

9
(45%)
20
(100%)
13
(65%)
20
(100%)
20
(100%)

10
(50%)

10
(50%)
Arkansas
1333
(99.7%)
743
(55.6%)
291
(21.8%)
586
(43.9%)

285
(21.3%)
603
(45.1%)

1330
(99.5%)

California
228
(39.0%)
322
(55.0%)
350
(59.8%)

519
(88.7%)
403
(68.9%)

545
(93.2%)

484
(82.7%)
Colorado
2883
(63.3%)
3358
(73.7%)

2586
(56.8%)
1048
(23.0%)
996
(21.9%)
2258
(49.6%)

2408
(52.9%)

Kansas
77
(79.4%)
87
(89.7%)
61
(62.9%)
24
(24.7%)
22
(22.7%)
27
(27.8%)
78
(80.4%)

82
(84.5%)
21
(21.6%)
Louisiana
596
(59.2%)
844
(83.9%)
341
(33.9%)
515
(51.2%)
377
(37.5%)
535
(53.2%)
420
(41.7%)
494
(49.1%)
569
(56.6%)
273
(27.2%)
Michigan
13
(92.9%)
14
(100%)
10
(71.4%)
11
(78.6%)





8
(57.1%)
Mississippi Montana
3
(75%)
4
(100%)
3
(75%)

2
(50%)
2
(50%)
4
(100%)
2
(50%)


121
(62.7%)
115
(59.6%)
95
(49.2%)
123
(63.7%)
124
(64.2%)
105
(54.4%)

83
(43.0%)
45
(23.3%)
119
(61.7%)
                  June 2015
This document is a draft for review purposes only and does not constitute Agency policy.

                                C-5                    DRAFT—DO NOT CITE OR QUOTE

-------
Hydraulic Fracturing Drinking Water Assessment
Appendix C
Chemical name
Propargyl alcohol
Glutaraldehyde
Naphthalene
2-Butoxyethanol
Citric acid
Saline
Solvent naphtha,
petroleum, heavy arom.
Quaternary ammonium
compounds, benzyl-C12-
16-alkyldimethyl, chlorides
2,2-Dibromo-3-
nitrilopropionamide
Potassium hydroxide
Choline chloride
Polyethylene glycol
1,2,4-Trimethylbenzene
CASRN
107-19-7
111-30-8
91-20-3
111-76-2
77-92-9
7647-14-5
64742-94-5
68424-85-1
10222-01-2
1310-58-3
67-48-1
25322-68-3
95-63-6
Alabama


55
(100%)
55
(100%)




55
(100%)


55
(100%)

Alaska



20
(100%)









Arkansas
813
(60.8%)
737
(55.1%)





375
(28.0%)





California













Colorado


1363
(29.9%)


1574
(34.5%)
1507
(33.1%)

2215
(48.6%)

1235
(27.1%)

1211
(26.63%)
Kansas
69
(71.1%)
73
(75.3%)
41
(42.3%)

45
(46.4%)

42
(43.3%)
52
(53.6%)




39
(40.2%)
Louisiana
299
(29.7%)
364
(36.3%)
293
(29.2%)


408
(40.6%)



340
(33.8%)



Michigan
5
(35.7%)

12
(85.7%)
11
(78.6%)




10
(71.4%)


7
(50%)

Mississippi Montana

2
(50%)



2
(50%)

2
(50%)

4
(100%)





95
(49.2%)



135
(70.0%)

70
(36.3%)
115
(59.6%)

69
(35.8%)

                                    This document is a draft for review purposes only and does not constitute Agency policy.
                    June 2015                                          C-6                      DRAFT—DO NOT CITE OR QUOTE

-------
Hydraulic Fracturing Drinking Water Assessment
Appendix C
Chemical name
Ammonium chloride
Diatomaceous earth,
calcined
Didecyl dimethyl
ammonium chloride
Sodium chlorite
Sodium erythorbate
N,N-Dimethylformamide
Nonyl phenol ethoxylate
Poly(oxy-l,2-
ethanediyl)-
nonylphenyl-hydroxy
(mixture)
Sodium persulfate
Tetramethylammonium
chloride
1,2-Propylene glycol
5-Chloro-2-methyl-3(2H)-
isothiazolone
Acetic acid
Ammonium acetate
CASRN
12125-02-9
91053-39-3
7173-51-5
7758-19-2
6381-77-7
68-12-2
9016-45-9
127087-87-
0
7775-27-1
75-57-0
57-55-6
26172-55-4
64-19-7
631-61-8
Alabama














Alaska

20
(100%)









20
(100%)


Arkansas
277
(20.7%)

317
(23.7%)

435
(32.5%)









California

417
(71.3%)





1150
(25.2%)



389
(66.5%)


Colorado
1280
(28.0%)






39
(40.2%)




959
(21.0%)

Kansas




29
(29.9%)









Louisiana



352
(35.0%)








284
(28.2%)

Michigan










10
(71.4%)



Mississippi Montana


2
(50%)
4
(100%)




4
(100%)




2
(50%)









85
(44.0%)




                                    This document is a draft for review purposes only and does not constitute Agency policy.
                    June 2015                                          C-7                      DRAFT—DO NOT CITE OR QUOTE

-------
Hydraulic Fracturing Drinking Water Assessment
Appendix C
Chemical name
Boric acid
Carbonic acid, dipotassium
salt
Cristobalite
Formic acid
Hemicellulase enzyme
Hemicellulase enzyme
concentrate
Iron(ll) sulfate
heptahydrate
Magnesium chloride
Magnesium nitrate
Phenolic resin
Sodium hypochlorite
Sodium tetraborate
decahydrate
Solvent naphtha,
petroleum, heavy aliph.
l-Butoxy-2-propanol
CASRN
10043-35-3
584-08-7
14464-46-1
64-18-6
9012-54-8
9025-56-3
7782-63-0
7786-30-3
10377-60-3
9003-35-4
7681-52-9
1303-96-4
64742-96-7
5131-66-8
Alabama



55
(100%)










Alaska
3
(15%)

20
(100%)




20
(100%)
20
(100%)


14
(70%)


Arkansas














California


389
(66.5%)


395
(67.5%)

389
(66.5%)
389
(66.5%)




315
(53.8%)
Colorado

1159
(25.4%)








1046
(23.0%)



Kansas














Louisiana



293
(29.1%)










Michigan






7 (50%)





7
(50%)

Mississippi Montana












2
(50%)















                                    This document is a draft for review purposes only and does not constitute Agency policy.
                    June 2015                                          C-8                      DRAFT—DO NOT CITE OR QUOTE

-------
Hydraulic Fracturing Drinking Water Assessment
                                                                                                    Appendix C
Chemical name
1-Propanol
1,2-Ethanediaminium, N,
N'-bis[2-[bis(2-hydroxyeth
yl)methylammonio]ethyl]-
N,N'bis(2-hydroxyethyl)-
N,N'-dimethyl-,tetrachl
oride
2-bromo-3-
nitrilopropionamide
2-Ethylhexanol
2-Methyl-3(2H)-
isothiazolone
2-Propenoic acid, polymer
with 2-propenamide
Alkenes, C>10 .alpha. -
Benzene, l,l'-oxybis-,
tetrapropylene derivs.,
sulfonated
Benzenesulfonic acid,
dodecyl-, compd. with Nl-
(2-aminoethyl)-l,2-
ethanediamine (1:?)
Benzyldimethyldodecylam
monium chloride
CASRN
71-23-8
138879-94-4
1113-55-9
104-76-7
2682-20-4
9003-06-9
64743-02-8
119345-03-8
40139-72-8
139-07-1
Alabama










Alaska




20
(100%)





Arkansas

343
(58.6%)




241
(18.0%)


268
(20.0%)
California




389
(66.5%)





Colorado
1232
(27.0%)









Kansas










Louisiana










Michigan










Mississippi Montana













83
(43.0052%
)



50
(25.9%)
48
(24.9%)

                    June 2015
This document is a draft for review purposes only and does not constitute Agency policy.
                                   C-9                      DRAFT—DO NOT CITE OR QUOTE

-------
Hydraulic Fracturing Drinking Water Assessment
Appendix C
Chemical name
Benzylhexadecyldimethyla
mmonium chloride
Boron sodium oxide
C10-C16 ethoxylated
alcohol
Calcium chloride
Carbon dioxide
Cinnamaldehyde (3-
phenyl-2-propenal)
Diethylene glycol
Diethylene glycol
monobutyl ether
Diethylenetriamine
Distillates, petroleum,
hydrotreated light
paraffinic
Distillates, petroleum,
hydrotreated middle
Ethoxylated C12-16
alcohols
Ethoxylated C14-15
alcohols
Formic acid, potassium
salt
CASRN
122-18-9
1330-43-4
68002-97-1
10043-52-4
124-38-9
104-55-2
111-46-6
112-34-5
111-40-0
64742-55-8
64742-46-7
68551-12-2
68951-67-7
590-29-4
Alabama





55
(100%)








Alaska


3
(15%)
20
(100%)






3
(15%)



Arkansas
268
(20.0%)











241
(18.0%)

California

361
(61.7%)







314
(53.7%)




Colorado














Kansas














Louisiana














Michigan




7
(50%)


7
(50%)






Mississippi Montana






















55
(28.5%)





                                    This document is a draft for review purposes only and does not constitute Agency policy.
                    June 2015                                          C-10                     DRAFT—DO NOT CITE OR QUOTE

-------
Hydraulic Fracturing Drinking Water Assessment
Appendix C
Chemical name
Glycerin, natural
Isotridecanol, ethoxylated
Methenamine
Naphtha, petroleum,
hydrotreated heavy
Poly(oxy-l,2-ethanediyl),
.alpha.,.alpha.'-[[(9Z)-9-
octadecenylimino]di-2,l-
ethanediyl]bis[. omega. -
hydroxy-
Potassium chloride
Sodium bromate
Sodium perborate
tetrahydrate
Sulfamicacid
Terpenes and Terpenoids,
sweet orange-oil
Tetradecyl dimethyl
benzyl ammonium
chloride
Tetrakis(hydroxymethyl)p
hosphonium sulfate
CASRN
56-81-5
9043-30-5
100-97-0
64742-48-9
26635-93-8
7447-40-7
7789-38-0
10486-00-7
5329-14-6
68647-72-3
139-08-2
55566-30-8
Alabama












Alaska












Arkansas










268
(20.0%)

California

312
(53.3%)










Colorado












Kansas












Louisiana


298
(29.6%)









Michigan
7
(50%)



9
(64.3%)
7
(50%)
7
(50%)





Mississippi Montana








2
(50%)
2
(50%)














                                    This document is a draft for review purposes only and does not constitute Agency policy.
                    June 2015                                          C-ll                     DRAFT—DO NOT CITE OR QUOTE

-------
Hydraulic Fracturing Drinking Water Assessment
                                                                                                Appendix C
Chemical name
Thiourea polymer
Tri-n-butyl tetradecyl
phosphonium chloride
Trisodium phosphate
CASRN
68527-49-1
81741-28-8
7601-54-9
Alabama



Alaska



Arkansas
384
(28.7%)


California



Colorado



Kansas


19
(19.6%)
Louisiana



Michigan



Mississippi Montana






Note for Table C-3a and C-3b: Analysis considered 34,675 disclosures and 676,376 ingredient records that met selected quality assurance criteria, including: completely parsed;
unique combination of fracture date and API well number; fracture date between January 1, 2011, and February 28, 2013; valid CASRN; and valid concentrations. Disclosures
that did not meet quality assurance criteria (3,855) or other, query-specific criteria were excluded from analysis.
Table C-3b. Top chemicals reported to FracFocus for each state and number (and percentage) of disclosures where a chemical is
           reported for that state, New Mexico to Wyoming (U.S. EPA, 2015c).

           Source:                 The top 20 most frequent chemicals were identified for the 20 states that reported to FracFocus, resulting in a total of
           93 chemicals. The chemicals were ranked by counting the number of states where that chemical was in the top 20; chemicals used most widely
           among the most states come first. For example, methanol is reported in 19 of 20 states,  so methanol is ranked first.
Chemical name
Methanol
Distillates, petroleum,
hydrotreated light
Ethylene glycol
Isopropanol
CASRN
67-56-1
64742-47-8
107-21-1
67-63-0
New
Mexico
1012
(90.8%)
699
(62.7%)
503
(45.1%)
695
(62.3%)
North
Dakota
1059
(53.3%)
943
(47.5%)
724
(36.4%)
739
(37.2%)
Ohio
76
(52.1%)
122
(83.6%)
83
(56.8%)
71
(48.6%)
Oklahoma
1270
(70.3%)
1270
(70.3%)
843
(46.7%)
764
(42.28%)
Pennsylvania
1633
(68.6%)
1434
(60.2%)
807
(33.9%)
735
(30.9%)
Texas
12664
(78.5%)
10677
(66.1%)
9591
(59.4%)
7731
(47.9%)
Utah
984
(78.5%)
934
(74.5%)
1065
(85.0%)
661
(52.8%)
Virginia
48
(60.8%)

22
(27.8%)
43 (54.4%)
West
Virginia
153
(64.0%)
196
(82.0%)
141
(59.0%)
74
(31.0%)
Wyoming
460
(38.4%)
612
(51.1%)

516
(43.1%)
                   June 2015
This document is a draft for review purposes only and does not constitute Agency policy.

                                 C-12                     DRAFT—DO NOT CITE OR QUOTE

-------
Hydraulic Fracturing Drinking Water Assessment
                                                                                                    Appendix C
Chemical name
Quartz
Sodium hydroxide
Ethanol
Guargum
Hydrochloric acid
Peroxydisulfuric acid,
diammonium salt
Propargyl alcohol
Glutaraldehyde
Naphthalene
2-Butoxyethanol
Citric acid
Saline
Solvent naphtha,
petroleum, heavy arom.
CASRN
14808-60-7
1310-73-2
64-17-5
9000-30-0
7647-01-0
7727-54-0
107-19-7
111-30-8
91-20-3
111-76-2
77-92-9
7647-14-5
64742-94-5
New
Mexico
762
(68.3%)
329
(29.5%)
529
(47.4%)
702
(63.0%)
880
(78.9%)
836
(75.0%)
760
(68.2%)
632
(56.7%)

412
(37.0%)
447
(40.1%)


North
Dakota
920
(46.3%)
1028
(51.7%)
545
(27.4%)
1094
(55.1%)

1089
(54.8%)


864
(43.5%)


491
(24.7%)
981
(49.4%)
Ohio
66
(45.2%)

87
(59.6%)
74
(50.7%)
145
(99.3%)
93
(63.7%)
72
(49.3%)
105
(71.9%)


96
(65.8%)


Oklahoma
491
(27.2%)
490
(27.1%)
838
(46.4%)
457
(25.3%)
1372
(75.9%)
713
(39.5%)
732
(40.5%)
989
(54.7%)
448
(24.8%)

644
(35.6%)

557
(30.8%)
Pennsylvania

406
(17.0%)
388
(16.3%)
538
(22.6%)
2279
(95.7%)

1371
(57.6%)
819
(34.4%)

498
(20.9%)
701
(29.4%)


Texas
6869
(42.6%)
7371
(45.7%)
3439
(21.3%)
6863
(42.5%)
11424
(70.8%)
8666
(53.7%)
6269
(38.8%)
6470
(40.1%)

3898
(24.1%)
3820
(23.7%)
3462
(21.4%)
2751
(17.0%)
Utah
503
(40.1%)
466
(37.2%)

538
(42.9%)
1064
(84.9%)
483
(38.5%)
456
(36.4%)

478
(38.1%)
663
(52.9%)
992
(79.2%)


Virginia


50
(63.3%)

68
(86.1%)

22
(27.8%)

7
(8.9%)
70
(88.6%)
63
(79.8%)
7
(8.9%)
7
(8.9%)
West
Virginia
53
(22.2%)

130
(54.3%)
55
(23.0%)
229
(95.8%)
128
(53.6%)
138
(57.7%)
169
(70.7%)

62
(25.9%)
98
(41.0%)
53
(22.2%)

Wyoming
356
(29.7%)
688
(57.4%)
298
(24.9%)
823
(68.7%)

771
(64.4%)

260
(21.7%)



274
(22.9%)
415
(34.6%)
                    June 2015
This document is a draft for review purposes only and does not constitute Agency policy.
                                   C-13                      DRAFT—DO NOT CITE OR QUOTE

-------
Hydraulic Fracturing Drinking Water Assessment
                                                                                                    Appendix C
Chemical name
Quaternary ammonium
compounds, benzyl-C12-
16-alkyldimethyl, chlorides
2,2-Dibromo-3-
nitrilopropionamide
Potassium hydroxide
Choline chloride
Polyethylene glycol
1,2,4-Trimethylbenzene
Ammonium chloride
Diatomaceous earth,
calcined
Didecyl dimethyl
ammonium chloride
Sodium chlorite
Sodium erythorbate
N,N-Dimethylformamide
Nonyl phenol ethoxylate
CASRN
68424-85-1
10222-01-2
1310-58-3
67-48-1
25322-68-3
95-63-6
12125-02-9
91053-39-3
7173-51-5
7758-19-2
6381-77-7
68-12-2
9016-45-9
New
Mexico



384
(34.4%)



419
(37.6%)




333
(29.9%)
North
Dakota


1176
(59.2%)

567
(28.5%)
496
(25.0%)



482
(24.3%)



Ohio
54
(37.0%)

106
(72.6%)
55
(37.7%)




46
(31.6%)


68
(46.6%)

Oklahoma
597
(33.0%)










355
(19.6%)

Pennsylvania
373
(15.7%)
804
(33.8%)


688
(28.9%)

732
(30.7%)






Texas


6369
(39.5%)










Utah



649
(51.8%)



435
(34.7%)



410
(32.7%)
447
(35.7%)
Virginia

22
(27.8%)

45
(57.0%)

7
(8.9%)




10
(12.7%)

25
(31.6%)
West
Virginia
53
(22.2%)





50
(20.9%)

49
(20.5%)

76
(31.8%)


Wyoming









271
(22.6%)



                    June 2015
This document is a draft for review purposes only and does not constitute Agency policy.
                                   C-14                     DRAFT—DO NOT CITE OR QUOTE

-------
Hydraulic Fracturing Drinking Water Assessment
Appendix C
Chemical name
Poly(oxy-l,2-ethanediyl)-
nonylphenyl-hydroxy
(mixture)
Sodium persulfate
Tetramethylammonium
chloride
1,2-Propylene glycol
5-Chloro-2-methyl-3(2H)-
isothiazolone
Acetic acid
Ammonium acetate
Boric acid
Carbonic acid, dipotassium
salt
Cristobalite
Formic acid
Hemicellulase enzyme
Hemicellulase enzyme
concentrate
Iron(ll) sulfate
heptahydrate
Magnesium chloride
CASRN
127087-87-0
7775-27-1
75-57-0
57-55-6
26172-55-4
64-19-7
631-61-8
10043-35-3
584-08-7
14464-46-1
64-18-6
9012-54-8
9025-56-3
7782-63-0
7786-30-3
New
Mexico












331
(29.7%)


North
Dakota


579
(29.1%)





482
(24.2%)






Ohio







82
(56.2%)







Oklahoma















Pennsylvania

373
(15.7%)









367
(15.4%)



Texas















Utah















Virginia
7
(8.9%)


22
(27.8%)







11
(13.9%)

22
(27.8%)

West
Virginia















Wyoming

308
(25.7%)
315
(26.3%)



323
(27.0%)








                                    This document is a draft for review purposes only and does not constitute Agency policy.
                    June 2015                                          C-15                     DRAFT—DO NOT CITE OR QUOTE

-------
Hydraulic Fracturing Drinking Water Assessment
                                                                                                    Appendix C
Chemical name
Magnesium nitrate
Phenolic resin
Sodium hypochlorite
Sodium tetraborate
decahydrate
Solvent naphtha,
petroleum, heavy aliph.
l-Butoxy-2-propanol
1-Propanol
1,2-Ethanediaminium, N,
N'-bis[2-[bis(2-hydroxy
ethyl) methylammonio]
ethyl]-N,N'bis(2-
hydroxyethyl)-N,N'-
dimethyl-, tetrachloride
2-Bromo-3-
nitrilopropionamide
2-Ethylhexanol
2-Methyl-3(2H)-
isothiazolone
2-Propenoic acid, polymer
with 2-propenamide
Alkenes, C>10 .alpha. -
CASRN
10377-60-3
9003-35-4
7681-52-9
1303-96-4
64742-96-7
5131-66-8
71-23-8
138879-94-4
1113-55-9
104-76-7
2682-20-4
9003-06-9
64743-02-8
New
Mexico

419
(37.6%)











North
Dakota













Ohio













Oklahoma













Pennsylvania













Texas

2903
(18.0%)











Utah











486
(38.8%)

Virginia








11
(13.9%)




West
Virginia













Wyoming


282
(23.5%)
265
(22.1%)









                    June 2015
This document is a draft for review purposes only and does not constitute Agency policy.
                                   C-16                     DRAFT—DO NOT CITE OR QUOTE

-------
Hydraulic Fracturing Drinking Water Assessment
                                                                                                    Appendix C
Chemical name
Benzene, l,l'-oxybis-,
tetrapropylene derivs.,
sulfonated
Benzenesulfonic acid,
dodecyl-, compd. with Nl-
(2-aminoethyl)-l,2-
ethanediamine (1:?)
Benzyldimethyldodecylam
monium chloride
Benzylhexadecyldimethyla
mmonium chloride
Boron sodium oxide
C10-C16 ethoxylated
alcohol
Calcium chloride
Carbon dioxide
Cinnamaldehyde (3-
phenyl-2-propenal)
Diethylene glycol
Diethylene glycol
monobutyl ether
Diethylenetriamine
Distillates, petroleum,
hydrotreated light
paraffinic
CASRN
119345-03-8
40139-72-8
139-07-1
122-18-9
1330-43-4
68002-97-1
10043-52-4
124-38-9
104-55-2
111-46-6
112-34-5
111-40-0
64742-55-8
New
Mexico













North
Dakota













Ohio









45
(30.8%)



Oklahoma













Pennsylvania













Texas













Utah













Virginia













West
Virginia













Wyoming













                    June 2015
This document is a draft for review purposes only and does not constitute Agency policy.
                                   C-17                     DRAFT—DO NOT CITE OR QUOTE

-------
Hydraulic Fracturing Drinking Water Assessment
                                                                                                    Appendix C
Chemical name
Distillates, petroleum,
hydrotreated middle
Ethoxylated C12-16
alcohols
Ethoxylated C14-15
alcohols
Formic acid, potassium
salt
Glycerin, natural
Isotridecanol, ethoxylated
Methenamine
Naphtha, petroleum,
hydrotreated heavy
Poly(oxy-l,2-ethanediyl),
.alpha.,.alpha.'-[[(9Z)-9-
octadecenylimino]di-2,l-
ethanediyl]bis[. omega. -
hydroxy-
Potassium chloride
Sodium bromate
Sodium perborate
tetrahydrate
Sulfamicacid
Terpenes and terpenoids,
sweet orange-oil
CASRN
64742-46-7
68551-12-2
68951-67-7
590-29-4
56-81-5
9043-30-5
100-97-0
64742-48-9
26635-93-8
7447-40-7
7789-38-0
10486-00-7
5329-14-6
68647-72-3
New
Mexico














North
Dakota














Ohio














Oklahoma











351
(19.4%)


Pennsylvania














Texas














Utah














Virginia














West
Virginia

57
(23.8%)












Wyoming



361
(30.1%)



384
(32.1%)






                    June 2015
This document is a draft for review purposes only and does not constitute Agency policy.
                                   C-18                     DRAFT—DO NOT CITE OR QUOTE

-------
Hydraulic Fracturing Drinking Water Assessment
                                                                                                           Appendix C
Chemical name
Tetradecyl dimethyl
benzyl ammonium
chloride
Tetrakis(hydroxymethyl)p
hosphonium sulfate
Thiourea polymer
Tri-n-butyl tetradecyl
phosphonium chloride
Trisodium phosphate
CASRN
139-08-2
55566-30-8
68527-49-1
81741-28-8
7601-54-9
New
Mexico





North
Dakota





Ohio





Oklahoma





Pennsylvania



350
(14.7%)

Texas





Utah

945
(75.4%)



Virginia





West
Virginia





Wyoming





Note for Table C-3a and C-3b: Analysis considered 34,675 disclosures and 676,376 ingredient records that met selected quality assurance criteria, including: completely parsed;
unique combination of fracture date and API well number; fracture date between January 1, 2011, and February 28, 2013; valid CASRN; and valid concentrations. Disclosures
that did not meet quality assurance criteria (3,855) or other, query-specific criteria were excluded from analysis.
                      June 2015
This document is a draft for review purposes only and does not constitute Agency policy.

                                     C-19                       DRAFT—DO NOT CITE OR QUOTE

-------
Hydraulic Fracturing Drinking Water Assessment
Appendix C
Table C-4. Estimated mean, median, 5th percentile, and 95th percentile volumes in gallons for
          chemicals reported to FracFocus in 100 or more disclosures, where density
          information was available.
          Chemicals are listed in alphabetical order. Density information came from Reaxys® and other sources.
          All density sources are referenced in Table C-7.
Name
(4R)-l-methyl-4-(prop-l-en-2-
yl)cyclohexene
l-Butoxy-2-propanol
1-Decanol
1-Octanol
1-Propanol
1,2-Propylene glycol
1,2,4-Trimethylbenzene
2-Butoxyethanol
2-Ethylhexanol
2-Mercaptoethanol
2,2-Dibromo-3-nitrilopropionamide
Acetic acid
Acetic anhydride
Acrylamide
Adipicacid
Aluminum chloride
Ammonia
Ammonium acetate
Ammonium chloride
Ammonium hydroxide
Benzyl chloride
Carbonic acid, dipotassium salt
Chlorine dioxide
Choline chloride
Cinnamaldehyde (3-phenyl-2-propenal)
CASRN
5989-27-5
5131-66-8
112-30-1
111-87-5
71-23-8
57-55-6
95-63-6
111-76-2
104-76-7
60-24-2
10222-01-2
64-19-7
108-24-7
79-06-1
124-04-9
7446-70-0
7664-41-7
631-61-8
12125-02-9
1336-21-6
100-44-7
584-08-7
10049-04-4
67-48-1
104-55-2
Volume (gallons)
Mean
2,702
167
28
5
128
13,105
38
385
100
1,175
183
646
239
95
153
2
44
839
440
7
52
467
31
2,131
68
Median
406
21
4
4
55
72
6
26
11
445
5
47
50
3
0
0
35
117
48
2
0
113
11
290
3
5th
Percentile
0
5
0
0
6
4
0
0
0
0
0
0
3
0
0
0
2
0
3
0
0
0
0
28
0
95th
Percentile
19,741
654
33
10
367
61,071
43
1,811
292
4,194
341
1,042
722
57
109
0
138
1,384
458
14
40
1,729
28
4,364
697
              This document is a draft for review purposes only and does not constitute Agency policy.
June 2015                                      C-20                   DRAFT—DO NOT CITE OR QUOTE

-------
Hydraulic Fracturing Drinking Water Assessment
Appendix C
Name
Citric acid
Dibromoacetonitrile
Diethylene glycol
Diethylenetriamine
Dodecane
Ethanol
Ethanolamine
Ethyl acetate
Ethylene glycol
Ferric chloride
Formalin
Formic acid
Fumaric acid
Glutaraldehyde
Glycerin, natural
Glycolicacid
Hydrochloric acid
Isopropanol
Isopropylamine
Magnesium chloride
Methanol
Methenamine
Methoxyacetic acid
N,N-Dimethylformamide
Naphthalene
Nitrogen, liquid
Ozone
Peracetic acid
Phosphonic acid
Phosphoric acid Divosan X-Tend
formulation
Potassium acetate
CASRN
77-92-9
3252-43-5
111-46-6
111-40-0
112-40-3
64-17-5
141-43-5
141-78-6
107-21-1
7705-08-0
50-00-0
64-18-6
110-17-8
111-30-8
56-81-5
79-14-1
7647-01-0
67-63-0
75-31-0
7786-30-3
67-56-1
100-97-0
625-45-6
68-12-2
91-20-3
7727-37-9
10028-15-6
79-21-0
13598-36-2
7664-38-2
127-08-2
Volume (gallons)
Mean
163
22
168
92
190
831
70
0
614
0
200
501
2
1,313
413
38
28,320
2,095
83
14
1,218
3,386
36
119
72
41,841
15,844
300
1,201
13
204
Median
20
13
16
21
31
121
30
0
184
0
0
38
0
122
109
10
3,110
55
121
0
110
100
4
10
12
26,610
15,473
268
0
4
1
5th
Percentile
1
1
0
0
0
1
0
0
4
0
0
1
0
2
10
4
96
0
0
0
2
0
2
0
0
3,091
8,785
50
0
0
0
95th
Percentile
269
45
102
207
151
2,645
283
0
2,470
0
8
1,229
12
1,165
911
94
26,877
1,264
172
2
3,731
3,648
115
216
204
108,200
26,063
663
3
15
974
                This document is a draft for review purposes only and does not constitute Agency policy.
June 2015                                          C-21                      DRAFT—DO NOT CITE OR QUOTE

-------
Hydraulic Fracturing Drinking Water Assessment
Appendix C
Name
Propargyl alcohol
Saline
Saturated sucrose
Silica, amorphous
Sodium carbonate
Sodium formate
Sodium hydroxide
Sulfur dioxide
Sulfuricacid
tert-Butyl hydroperoxide (70% solution in
Water)
Tetramethylammonium chloride
Thioglycolic acid
Toluene
Tridecane
Triethanolamine
Triethyl phosphate
Triethylene glycol
Triisopropanolamine
Trimethyl borate
Undecane
CASRN
107-19-7
7647-14-5
57-50-1
7631-86-9
497-19-8
141-53-7
1310-73-2
7446-09-5
7664-93-9
75-91-2
75-57-0
68-11-1
108-88-3
629-50-5
102-71-6
78-40-0
112-27-6
122-20-3
121-43-7
1120-21-4
Volume (gallons)
Mean
183
876
1
6,877
228
0
551
0
3
156
970
55
18
190
846
55
5,198
46
83
273
Median
2
85
1
8
16
0
38
0
0
64
483
7
0
31
60
1
116
4
40
29
5th
Percentile
0
0
0
0
0
0
0
0
0
0
2
2
0
0
0
0
28
1
4
0
95th
Percentile
51
1,544
2
38,371
1,319
0
1,327
0
3
557
3,508
229
11
190
2,264
533
945
330
283
1,641
 Note: Analysis considered 34,495 disclosures and 672,358 ingredient records that met selected quality assurance criteria,
 including: completely parsed; unique combination of fracture date and API well number; fracture date between January 1,
 2011, and February 28, 2013; criteria for water volumes; valid CASRN; and valid concentrations. Disclosures that did not meet
 quality assurance criteria (4,035) or other, query-specific criteria were excluded from analysis.
                 This document is a draft for review purposes only and does not constitute Agency policy.
June 2015                                             C-22                       DRAFT—DO NOT CITE OR QUOTE

-------
Hydraulic Fracturing Drinking Water Assessment
Appendix C
Table C-5. Estimated mean, median, 5th percentile, and 95th percentile volumes in liters for
          chemicals reported to FracFocus in 100 or more disclosures, where density
          information was available.
          Chemicals are listed in alphabetical order. Density information came from Reaxys® and other sources.
          All density sources are referenced in Table C-7.
Name
(4R)-l-methyl-4-(prop-l-en-2-
yl)cyclohexene
l-Butoxy-2-propanol
1-Decanol
1-Octanol
1-Propanol
1,2-Propylene glycol
1,2,4-Trimethylbenzene
2-Butoxyethanol
2-Ethylhexanol
2-Mercaptoethanol
2,2-Dibromo-3-nitrilopropionamide
Acetic acid
Acetic anhydride
Acrylamide
Adipicacid
Aluminum chloride
Ammonia
Ammonium acetate
Ammonium chloride
Ammonium hydroxide
Benzyl chloride
Carbonic acid, dipotassium salt
Chlorine dioxide
Choline chloride
Cinnamaldehyde (3-phenyl-2-propenal)
Citric acid
CASRN
5989-27-5
5131-66-8
112-30-1
111-87-5
71-23-8
57-55-6
95-63-6
111-76-2
104-76-7
60-24-2
10222-01-2
64-19-7
108-24-7
79-06-1
124-04-9
7446-70-0
7664-41-7
631-61-8
12125-02-9
1336-21-6
100-44-7
584-08-7
10049-04-4
67-48-1
104-55-2
77-92-9
Volume (L)
Mean
10,229
631
107
21
483
49,607
145
1,459
377
4,449
692
2,446
906
361
578
6
166
3,177
1,666
27
196
1,769
117
8,068
258
618
Median
1,536
80
14
14
208
274
24
98
40
1,685
18
176
189
10
0
0
134
444
182
6
1
429
43
1,096
12
77
5th
Percentile
0
18
1
1
22
15
0
0
1
0
0
0
12
0
0
0
7
0
11
1
0
0
1
107
0
5
95th
Percentile
74,729
2,475
123
39
1,391
231,179
165
6,856
1,106
15,878
1,292
3,945
2,734
216
414
0
523
5,238
1,733
52
151
6,544
106
16,521
2,638
1,019
              This document is a draft for review purposes only and does not constitute Agency policy.
June 2015                                      C-23                   DRAFT—DO NOT CITE OR QUOTE

-------
Hydraulic Fracturing Drinking Water Assessment
Appendix C
Name
Dibromoacetonitrile
Diethylene glycol
Diethylenetriamine
Dodecane
Ethanol
Ethanolamine
Ethyl acetate
Ethylene glycol
Ferric chloride
Formalin
Formic acid
Fumaricacid
Glutaraldehyde
Glycerin, natural
Glycolic acid
Hydrochloric acid
Isopropanol
Isopropylamine
Magnesium chloride
Methanol
Methenamine
Methoxyacetic acid
N,N-Dimethylformamide
Naphthalene
Nitrogen, liquid
Ozone
Peracetic acid
Phosphonicacid
Phosphoric acid Divosan X-Tend
formulation
Potassium acetate
Propargyl alcohol
CASRN
3252-43-5
111-46-6
111-40-0
112-40-3
64-17-5
141-43-5
141-78-6
107-21-1
7705-08-0
50-00-0
64-18-6
110-17-8
111-30-8
56-81-5
79-14-1
7647-01-0
67-63-0
75-31-0
7786-30-3
67-56-1
100-97-0
625-45-6
68-12-2
91-20-3
7727-37-9
10028-15-6
79-21-0
13598-36-2
7664-38-2
127-08-2
107-19-7
Volume (L)
Mean
82
636
347
719
3,144
264
0
2,324
0
756
1,896
9
4,972
1,565
146
107,204
7,932
314
52
4,609
12,817
136
449
271
158,384
59,976
1,137
4,547
51
775
693
Median
50
61
80
117
458
112
0
697
0
2
144
0
462
412
39
11,772
210
458
0
416
378
17
38
44
100,731
58,570
1,016
2
15
3
9
5th
Percentile
4
1
0
0
6
0
0
14
0
0
2
0
6
38
14
362
1
0
0
6
0
8
2
0
11,700
33,254
190
0
0
0
0
95th
Percentile
170
384
785
572
10,011
1,070
0
9,349
0
31
4,653
46
4,409
3,447
356
101,741
4,786
652
8
14,125
13,810
436
819
774
409,583
98,658
2,511
11
57
3,690
193
                This document is a draft for review purposes only and does not constitute Agency policy.
June 2015                                          C-24                     DRAFT—DO NOT CITE OR QUOTE

-------
Hydraulic Fracturing Drinking Water Assessment
Appendix C
Name
Saline
Saturated sucrose
Silica, amorphous
Sodium carbonate
Sodium formate
Sodium hydroxide
Sulfur dioxide
Sulfuric acid
tert-Butyl hydroperoxide (70% solution in
Water)
Tetramethylammonium chloride
Thioglycolicacid
Toluene
Tridecane
Triethanolamine
Triethyl phosphate
Triethylene glycol
Triisopropanolamine
Trimethyl borate
Undecane
CASRN
7647-14-5
57-50-1
7631-86-9
497-19-8
141-53-7
1310-73-2
7446-09-5
7664-93-9
75-91-2
75-57-0
68-11-1
108-88-3
629-50-5
102-71-6
78-40-0
112-27-6
122-20-3
121-43-7
1120-21-4
Volume (L)
Mean
3,317
5
26,031
862
1
2,087
2
10
591
3,672
208
69
721
3,203
209
19,676
174
314
1,035
Median
321
2
32
62
1
144
0
0
242
1,830
28
0
118
228
6
439
16
152
111
5th
Percentile
0
0
0
0
0
1
0
0
0
8
6
0
0
0
0
106
4
16
0
95th
Percentile
5,844
6
145,251
4,991
1
5,024
0
12
2,109
13,279
868
41
721
8,570
2,019
3,579
1,249
1,072
6,212
 Note: Analysis considered 34,495 disclosures and 672,358 ingredient records that met selected quality assurance criteria,
 including: completely parsed; unique combination of fracture date and API well number; fracture date between January 1,
 2011, and February 28, 2013; criteria for water volumes; valid CASRN; and valid concentrations. Disclosures that did not meet
 quality assurance criteria (4,035) or other, query-specific criteria were excluded from analysis.
                 This document is a draft for review purposes only and does not constitute Agency policy.

June 2015                                             C-25                       DRAFT—DO NOT CITE OR QUOTE

-------
Hydraulic Fracturing Drinking Water Assessment
Appendix C
Table C-6. Calculated mean, median, 5th percentile, and 95th percentile chemical masses
           reported to FracFocus in 100 or more disclosures, where density information was
           available.
           Density information came from Reaxys® and other sources. All density sources are referenced in Table
           C-7. Number of disclosures reported for each chemical is also included.
Name
(4R)-l-methyl-4-(prop-l-en-2-
yl)cyclohexene
l-Butoxy-2-propanol
1-Decanol
1-Octanol
1-Propanol
1,2-Propylene glycol
1,2,4-Trimethylbenzene
2-Butoxyethanol
2-Ethylhexanol
2-Mercaptoethanol
2,2-Dibromo-3-
nitrilopropionamide
Acetic acid
Acetic anhydride
Acrylamide
Adipic acid
Aluminum chloride
Ammonia
Ammonium acetate
Ammonium chloride
Ammonium hydroxide
Benzyl chloride
Carbonic acid, dipotassium salt
Chlorine dioxide
Choline chloride
Cinnamaldehyde (3-phenyl-2-
propenal)
CASRN
5989-27-5
5131-66-8
112-30-1
111-87-5
71-23-8
57-55-6
95-63-6
111-76-2
104-76-7
60-24-2
10222-01-2
64-19-7
108-24-7
79-06-1
124-04-9
7446-70-0
7664-41-7
631-61-8
12125-02-9
1336-21-6
100-44-7
584-08-7
10049-04-4
67-48-1
104-55-2
Mass (kg)
Mean
8,593
555
89
17
386
51,095
126
1,313
313
489
1,660
2,544
969
408
785
15
111
3,718
2,530
48
214
4,298
321
9,440
284
Median
1,290
71
12
12
167
282
21
88
34
185
44
183
203
11
0
0
90
520
277
11
1
1,042
117
1,282
13
5th
Percentile
0
16
1
1
18
15
0
0
0
0
0
0
12
0
0
0
4
0
16
2
0
0
3
125
0
95th
Percentile
62,772
2,178
102
32
1,113
238,114
143
6,170
918
1,747
3,102
4,103
2,925
244
564
0
351
6,129
2,633
94
165
15,902
291
19,329
2,902
Disclosures
578
773
434
434
1,481
1,023
3,976
6,778
1,291
2,051
4,927
7,643
1,377
251
233
122
398
1,504
3,288
1,173
1,833
4,093
331
4,241
1,377
              This document is a draft for review purposes only and does not constitute Agency policy.
June 2015                                      C-26                   DRAFT—DO NOT CITE OR QUOTE

-------
Hydraulic Fracturing Drinking Water Assessment
Appendix C
Name
Citric acid
Dibromoacetonitrile
Diethylene glycol
Diethylenetriamine
Dodecane
Ethanol
Ethanolamine
Ethyl acetate
Ethylene glycol
Ferric chloride
Formalin
Formic acid
Fumaricacid
Glutaraldehyde
Glycerin, natural
Glycolic acid
Hydrochloric acid
Isopropanol
Isopropylamine
Magnesium chloride
Methanol
Methenamine
Methoxyacetic acid
N,N-Dimethylformamide
Naphthalene
Nitrogen, liquid
Ozone
Peracetic acid
Phosphonicacid
Phosphoric acid Divosan X-Tend
formulation
Potassium acetate
CASRN
77-92-9
3252-43-5
111-46-6
111-40-0
112-40-3
64-17-5
141-43-5
141-78-6
107-21-1
7705-08-0
50-00-0
64-18-6
110-17-8
111-30-8
56-81-5
79-14-1
7647-01-0
67-63-0
75-31-0
7786-30-3
67-56-1
100-97-0
625-45-6
68-12-2
91-20-3
7727-37-9
10028-15-6
79-21-0
13598-36-2
7664-38-2
127-08-2
Mass (kg)
Mean
989
193
712
330
539
2,484
267
0
2,557
0
816
2,313
15
4,972
1,972
217
107,204
6,187
213
120
3,641
15,380
161
422
220
129,875
129
1,251
7,730
48
1,216
Median
123
118
68
76
88
361
113
0
767
0
2
176
0
462
519
58
11,772
163
311
1
329
454
20
36
35
82,599
126
1,117
3
14
5
5th
Percentile
8
11
1
0
0
4
0
0
15
0
0
2
0
6
47
21
362
1
0
0
5
0
9
2
0
9,594
71
209
0
0
0
95th
Percentile
1,630
403
430
746
429
7,908
1,081
0
10,283
0
34
5,677
75
4,409
4,343
530
101,741
3,733
444
18
11,159
16,572
514
770
627
335,858
212
2,762
18
54
5,793
Disclosures
7,503
272
1,732
784
131
9,233
585
110
14,767
118
456
3,781
224
10,963
1,829
595
20,996
15,058
255
1,113
23,225
4,412
584
2,972
5,945
713
209
221
2,216
315
325
                This document is a draft for review purposes only and does not constitute Agency policy.
June 2015                                          C-27                      DRAFT—DO NOT CITE OR QUOTE

-------
Hydraulic Fracturing Drinking Water Assessment
Appendix C
Name
Propargyl alcohol
Saline
Saturated sucrose
Silica, amorphous
Sodium carbonate
Sodium formate
Sodium hydroxide
Sulfur dioxide
Sulfuric acid
tert-Butyl hydroperoxide
(70% solution in water)
Tetramethylammonium chloride
Thioglycolicacid
Toluene
Tridecane
Triethanolamine
Triethyl phosphate
Triethylene glycol
Triisopropanolamine
Trimethyl borate
Undecane
CASRN
107-19-7
7647-14-5
57-50-1
7631-86-9
497-19-8
141-53-7
1310-73-2
7446-09-5
7664-93-9
75-91-2
75-57-0
68-11-1
108-88-3
629-50-5
102-71-6
78-40-0
112-27-6
122-20-3
121-43-7
1120-21-4
Mass (kg)
Mean
658
7,197
6
57,267
2,191
2
4,445
2
18
532
4,296
277
59
541
3,588
222
22,038
177
292
766
Median
9
696
2
71
158
1
306
0
0
218
2,141
37
0
88
255
6
491
17
141
82
5th
Percentile
0
0
0
0
0
1
2
0
0
0
10
8
0
0
0
0
119
4
14
0
95th
Percentile
183
12,682
7
319,553
12,678
2
10,701
0
22
1,898
15,537
1,155
35
541
9,599
2,140
4,008
1,274
997
4,597
Disclosures
10,771
6,673
125
2,423
396
204
12,585
224
402
814
3,162
156
214
132
1,498
991
528
251
294
241
 Note: Analysis considered 34,495 disclosures and 672,358 ingredient records that met selected quality assurance criteria,
 including: completely parsed; unique combination of fracture date and API well number; fracture date between January 1,
 2011, and February 28, 2013; criteria for water volumes; valid CASRN; and valid concentrations. Disclosures that did not meet
 quality assurance criteria (4,035) or other, query-specific criteria were excluded from analysis.
                 This document is a draft for review purposes only and does not constitute Agency policy.

June 2015                                             C-28                       DRAFT—DO NOT CITE OR QUOTE

-------
Hydraulic Fracturing Drinking Water Assessment
Appendix C
Table C-7. Associated chemical densities and references used to calculate chemical mass and
          estimate chemical volume.
Name
(4R)-l-methyl-4-(prop-l-en-2-yl)cyclohexene
l-Butoxy-2-propanol
1-Decanol
1-Octanol
1-Propanol
1,2-Propylene glycol
1,2,4-Trimethylbenzene
2-Butoxyethanol
2-Ethylhexanol
2-Mercaptoethanol
2,2-Dibromo-3-nitrilopropionamide
Acetic acid
Acetic anhydride
Acrylamide
Adipicacid
Aluminum chloride
Ammonia
Ammonium acetate
Ammonium chloride
Ammonium hydroxide
Benzyl chloride
Carbonic acid, dipotassium salt
Chlorine dioxide
Choline chloride
Cinnamaldehyde (3-phenyl-2-propenal)
Citric acid
Dibromoacetonitrile
Diethylene glycol
Diethylenetriamine
Dodecane
CASRN
5989-27-5
5131-66-8
112-30-1
111-87-5
71-23-8
57-55-6
95-63-6
111-76-2
104-76-7
60-24-2
10222-01-2
64-19-7
108-24-7
79-06-1
124-04-9
7446-70-0
7664-41-7
631-61-8
12125-02-9
1336-21-6
100-44-7
584-08-7
10049-04-4
67-48-1
104-55-2
77-92-9
3252-43-5
111-46-6
111-40-0
112-40-3
Density
(g/mL)
0.84
0.88
0.83
0.82
0.8
1.03
0.87
0.9
0.83
0.11
2.4
1.04
1.07
1.13
1.36
2.44
0.67
1.17
1.519
1.8
1.09
2.43
2.757
1.17
1.1
1.6
2.37
1.12
0.95
0.75
Reference
DejoyeTanzi et al. (2012)
Pal etal. (2013)

Faria et al. (2013)

Dubey and Kumar (2013)

Rani and Maken (2013)
Moosavi etal. (2013)
He etal. (2008)

Dhondgeetal. (2010)
Laavi et al. (2012)

Rawatetal. (1976)
Pels (1900)

Chafer et al. (2010)
Radwan and Hanna (1976)

Carpenter and Davis (1957)
Thalladi etal. (2000)
Sigma-Aldrich (2015a)

Harlow etal. (1997)
Biltz and Balz (1928)

Haynes (2014)
Xiao et al. (2013)

Sarkar etal. (2012)
Sigma-Aldrich (2014b)

Haynes (2014)
Shanley and Collin (1961)
Masood etal. (1976)

Bennett and Yuill (1935)
Wilt (1956)

Chasib (2013)
Dubey and Kumar (2011)

Baragietal. (2013)
              This document is a draft for review purposes only and does not constitute Agency policy.

June 2015                                     C-29                   DRAFT—DO NOT CITE OR QUOTE

-------
Hydraulic Fracturing Drinking Water Assessment
Appendix C
Name
Ethanol
Ethanolamine
Ethyl acetate
Ethylene glycol
Ferric chloride
Formalin
Formic acid
Fumaric acid
Glutaraldehyde
Glycerin, natural
Glycolicacid
Hydrochloric acid
Isopropanol
Isopropylamine
Magnesium chloride
Methanol
Methenamine
Methoxyacetic acid
N,N-Dimethylformamide
Naphthalene
Nitrogen, liquid
Ozone
Peracetic acid
Phosphonic acid
Phosphoric acid Divosan X-Tend formulation
Potassium acetate
Propargyl alcohol
Saline
Saturated sucrose
Silica, amorphous
Sodium carbonate
Sodium formate
CASRN
64-17-5
141-43-5
141-78-6
107-21-1
7705-08-0
50-00-0
64-18-6
110-17-8
111-30-8
56-81-5
79-14-1
7647-01-0
67-63-0
75-31-0
7786-30-3
67-56-1
100-97-0
625-45-6
68-12-2
91-20-3
7727-37-9
10028-15-6
79-21-0
13598-36-2
7664-38-2
127-08-2
107-19-7
7647-14-5
57-50-1
7631-86-9
497-19-8
141-53-7
Density
(g/mL)
0.79
1.01
0.89
1.1
2.9
1.08
1.22
1.64
1
1.26
1.49
1
0.78
0.68
2.32
0.79
1.2
1.18
0.94
0.81
0.8
0.002144
1.1
1.7
0.94
1.57
0.95
2.17
1.13
2.2
2.54
1.97
Reference
Kiselevetal. (2012)

Blanco et al. (2013)
Laavi et al. (2013)

Rodnikova et al. (2012)

Haynes (2014)
Alfa Aesar (2015)

Casanova etal. (1981)
Huffman and Fox (1938)

Oka (1962)
Egorovetal. (2013)

Pijper(1971)
Steinhauseretal. (1990)

Zhang etal. (2013)

Sarkar and Roy (2009)
Haynes (2014)
Kiselevetal. (2012)
Mak(1965)

Haynes (2014)
Smirnovand Badelin (2013)

Dyshin et al. (2008)
finemech (2012)

Haynes (2014)
Sigma-Aldrich (2015b)
Sigma-Aldrich (2014a)

Fadeeva et al. (2004)
Haynes (2014)
Vijaya Kumar etal. (1996)
Sigma-Aldrich (2010)

Hagen and Kaatze (2004)
Fuiinoetal. (2004)

Haynes (2014)
Fuess etal. (1982)
                This document is a draft for review purposes only and does not constitute Agency policy.
June 2015                                          C-30                      DRAFT—DO NOT CITE OR QUOTE

-------
Hydraulic Fracturing Drinking Water Assessment
Appendix C
Name
Sodium hydroxide
Sulfur dioxide
Sulfuric acid
tert-Butyl hydroperoxide
(70% solution in water)
Tetramethylammonium chloride
Thioglycolicacid
Toluene
Tridecane
Triethanolamine
Triethyl phosphate
Triethylene glycol
Triisopropanolamine
Trimethyl borate
Undecane
CASRN
1310-73-2
7446-09-5
7664-93-9
75-91-2
75-57-0
68-11-1
108-88-3
629-50-5
102-71-6
78-40-0
112-27-6
122-20-3
121-43-7
1120-21-4
Density
(g/mL)
2.13
1.3
1.83
0.9
1.17
1.33
0.86
0.75
1.12
1.06
1.12
1.02
0.93
0.74
Reference
Haynes (2014)
Sigma-Aldrich (2015c)
Sigma-Aldrich (2015d)

Sigma-Aldrich (2007)
Haynes (2014)
Biilmann (1906)
Martinez-Reina et al. (2012)

Zhang etal. (2011)
Blanco et al. (2013)

Krakowiak et al. (2001)
Afzal et al. (2009)

IUPAC(2014)
Sigma-Aldrich (2015e)

deOliveiraetal. (2011)
                This document is a draft for review purposes only and does not constitute Agency policy.
June 2015                                          C-31                      DRAFT—DO NOT CITE OR QUOTE

-------
Hydraulic Fracturing Drinking Water Assessment
Appendix C
Table C-8. Selected physicochemical properties of chemicals reported as used in hydraulic fracturing fluids.
           Properties are provided for chemicals, where available from EPI Suite™ version 4.1 (UA
Chemical name
(13Z)-N,N-bis(2-hydroxyethyl)-N-
methyldocos-13-en-l-a minium chloride
(2,3-Dihydroxypropyl)trimethyl
ammonium chloride
(E)-Crotonaldehyde
[Nitrilotris(methylene)]tris-phosphonic
acid pentasodium salt
l-(l-Naphthylmethyl)quinolinium
chloride
l-(Alkyl* amino)-3-aminopropane
*(42%C12, 26%C18, 15%C14, 8%C16,
5%C10, 4%C8)
l-(Phenylmethyl)pyridinium Et Me
derivatives, chlorides
CASRN
120086-58-0
34004-36-9
123-73-9
2235-43-0
65322-65-8
68155-37-3
68909-18-2
Log Kow
Estimated
4.38
-5.8
0.6
-5.45
5.57
4.74
4.1
Measured
-
-
-
-3.53
-
-
-
Water solubility
Estimate from log Kow
(mg/Lat25°C)
0.3827
1.00 x 10s
4.15xl04
1.00 x 10s
0.02454
23.71
14.13
Henry's law constant
(atm-m3/mol at 25°C)
Bond
method
3.32 x ID"15
9.84 x 1Q-18
5.61 x 1Q-5
1.65 x 1Q-34
1.16xlQ-7
6.81 x 1Q-8
1.78 x 1Q-5
Group
method 25
-
-
1.90 x ID"5
-
-
2.39 x ID"8
-
Measured
-
-
1.94 x ID"5
-
-
-
-
                                  This document is a draft for review purposes only and does not constitute Agency policy.
                   June 2015                                       C-32                   DRAFT—DO NOT CITE OR QUOTE

-------
Hydraulic Fracturing Drinking Water Assessment
Appendix C
Chemical name
1,2,3-Trimethylbenzene
1,2,4-Trimethylbenzene
l,2-Benzisothiazolin-3-one
l,2-Dibromo-2,4-dicyanobutane
1,2-Dimethylbenzene
1,2-Ethanediaminium, N,N'-bis[2-[bis(2-
hydroxyethyl)methylammonio]ethyl]-
N,N'-bis(2-hydroxyethyl)-N,N'-dimethyl-,
tetrachloride
1,2-Propylene glycol
1,2-Propylene oxide
1,3,5-Triazine
l,3,5-Triazine-l,3,5(2H,4H,6H)-triethanol
CASRN
526-73-8
95-63-6
2634-33-5
35691-65-7
95-47-6
138879-94-4
57-55-6
75-56-9
290-87-9
4719-04-4
Log Kow
Estimated
3.63
3.63
0.64
1.63
3.09
-23.19
-0.78
0.37
-0.2
-4.67
Measured
3.66
3.63
-
-
3.12
-
-0.92
0.03
0.12
-
Water solubility
Estimate from log Kow
(mg/Lat25°C)
75.03
79.59
2.14xl04
424
224.1
1.00 x 10s
S.llxlO5
1.29 x 105
1.03 x 105
1.00 x 10s
Henry's law constant
(atm-m3/mol at 25°C)
Bond
method
7.24 x ID"3
7.24 x 1Q-3
6.92 x 1Q-9
3.94 x ID"10
6.56 x 1Q-3
2.33 x 1Q-35
1.74 x ID"7
1.60 x ID"4
1.21 x 1Q-6
1.08 x 1Q-11
Group
method 25
6.58 x ID"3
6.58 x ID"3
-
-
6.14xlQ-3
-
1.31 x ID"10
1.23 x ID"4
-
-
Measured
4.36 x ID"3
6.16xlQ-3
-
-
S.lSxlQ-3
-
1.29 x ID"8
6.96 x ID"5
-
-
                                    This document is a draft for review purposes only and does not constitute Agency policy.
                    June 2015                                          C-33                     DRAFT—DO NOT CITE OR QUOTE

-------
Hydraulic Fracturing Drinking Water Assessment
Appendix C
Chemical name
1,3,5-Trimethylbenzene
1,3-Butadiene
1,3-Dichloropropene
1,4-Dioxane
1,6-Hexanediamine
1,6-Hexanediamine dihydrochloride
l-[2-(2-Methoxy-l-methylethoxy)-l-
methylethoxy]-2-propanol
l-Amino-2-propanol
1-Benzylquinolinium chloride
1-Butanol
l-Butoxy-2-propanol
CASRN
108-67-8
106-99-0
542-75-6
123-91-1
124-09-4
6055-52-3
20324-33-8
78-96-6
15619-48-4
71-36-3
5131-66-8
Log Kow
Estimated
3.63
2.03
2.29
-0.32
0.35
0.35
-0.2
-1.19
4.4
0.84
0.98
Measured
3.42
1.99
2.04
-0.27
-
-
-
-0.96
-
0.88
-
Water solubility
Estimate from log Kow
(mg/Lat25°C)
120.3
792.3
1,994
2.14xl05
5.34 x 105
5.34 x 105
1.96 x 105
1.00 x 10s
6.02
7.67 x 104
4.21 xlO4
Henry's law constant
(atm-m3/mol at 25°C)
Bond
method
7.24 x ID"3
7.79 x ID"2
2.45 x ID"2
5.91 x 1Q-6
3.21 x 1Q-9
3.21 x 1Q-9
2.36 x 1Q-11
4.88 x ID"10
1.19 x 1Q-6
9.99 x 1Q-6
1.30 x ID"7
Group
method 25
6.58 x ID"3
7.05 x ID"2
3.22 x 1Q-3
1.12 x ID"7
7.05 x ID"10
7.05 x ID"10
4.55 x 1Q-13
2.34 x ID"10
-
9.74 x ID"6
4.88 x ID"8
Measured
8.77 x 1Q-3
7.36 x ID"2
3.55 x 1Q-3
4.80 x 1Q-6
-
-
-
-
-
8.81 x ID"6
-
                                    This document is a draft for review purposes only and does not constitute Agency policy.
                    June 2015                                          C-34                     DRAFT—DO NOT CITE OR QUOTE

-------
Hydraulic Fracturing Drinking Water Assessment
Appendix C
Chemical name
1-Decanol
l-Dodecyl-2-pyrrolidinone
1-Eicosene
l-Ethyl-2-methylbenzene
1-Hexadecene
1-Hexanol
l-Methoxy-2-propanol
1-Octadecanamine, acetate (1:1)
1-Octadecanamine, N,N-dimethyl-
1-Octadecene
1-Octanol
1-Pentanol
CASRN
112-30-1
2687-96-9
3452-07-1
611-14-3
629-73-2
111-27-3
107-98-2
2190-04-7
124-28-7
112-88-9
111-87-5
71-41-0
Log Kow
Estimated
3.79
5.3
10.03
3.58
8.06
1.82
-0.49
7.71
8.39
9.04
2.81
1.33
Measured
4.57
4.2
-
3.53
-
2.03
-
-
-
-
3
1.51
Water solubility
Estimate from log Kow
(mg/Lat25°C)
28.21
5.862
1.26 x ID"5
96.88
0.001232
6,885
1.00 x 10s
0.04875
0.008882
1.256x 10'4
814
2.09 x 104
Henry's law constant
(atm-m3/mol at 25°C)
Bond
method
5.47 x ID"5
7.12 x ID"7
1.89 x 101
8.71 x 1Q-3
6.10
1.76 x 1Q-5
5.56 x 1Q-8
9.36 x ID"4
4.51 x 1Q-3
10.7
S.lOxlQ-5
1.33 x 1Q-5
Group
method 25
7.73 x ID"5
-
6.74 x 101
9.52 x ID"3
1.69 x 101
1.94 x 1Q-5
1.81 x 1Q-8
2.18xlQ-3
3.88 x ID"2
3.38 xlO1
3.88 x 1Q-5
1.38 x 1Q-5
Measured
3.20 x 1Q-5
-
-
5.53 x ID"3
-
1.71 x ID"5
9.20 x ID"7
-
-
-
2.45 x ID"5
1.30 x ID"5
                                    This document is a draft for review purposes only and does not constitute Agency policy.
                    June 2015                                          C-35                     DRAFT—DO NOT CITE OR QUOTE

-------
Hydraulic Fracturing Drinking Water Assessment
Appendix C
Chemical name
1-Propanaminium, 3-chloro-2-hydroxy-
N,N,N-trimethyl-, chloride
1-Propanesulfonic acid
1-Propanol
1-Propene
l-tert-Butoxy-2-propanol
1-Tetradecene
1-Tridecanol
1-Undecanol
2-(2-Butoxyethoxy)ethanol
2-(2-Ethoxyethoxy)ethanol
2-(2-Ethoxyethoxy)ethyl acetate
CASRN
3327-22-8
5284-66-2
71-23-8
115-07-1
57018-52-7
1120-36-1
112-70-9
112-42-5
112-34-5
111-90-0
112-15-2
Log Kow
Estimated
-4.48
-1.4
0.35
1.68
0.87
7.08
5.26
4.28
0.29
-0.69
0.32
Measured
-
-
0.25
1.77
-
-
-
-
0.56
-0.54
-
Water solubility
Estimate from log Kow
(mg/Lat25°C)
1.00 x 10s
1.00 x 10s
2.72 x 105
1,162
5.24 x 104
0.01191
4.533
43.04
7.19xl04
8.28 x 105
3.09 x 104
Henry's law constant
(atm-m3/mol at 25°C)
Bond
method
9.48 x ID"17
2.22 x 1Q-8
7.52 x 1Q-6
1.53 x ID"1
1.30 x ID"7
3.46
1.28 x ID"4
7.26 x 1Q-5
1.52 x 1Q-9
8.63 x ID"10
5.62 x 1Q-8
Group
method 25
-
-
6.89 x ID"6
1.58 x ID"1
5.23 x 1Q-8
8.48
2.18xlQ-4
1.09 x ID"4
4.45 x 1Q-11
2.23 x 1Q-11
7.22 x ID"10
Measured
-
-
7.41 x ID"6
1.96 x ID"1
-
-
-
-
7.20 x ID"9
2.23 x ID"8
2.29 x 1Q-8
                                    This document is a draft for review purposes only and does not constitute Agency policy.
                    June 2015                                          C-36                     DRAFT—DO NOT CITE OR QUOTE

-------
Hydraulic Fracturing Drinking Water Assessment
Appendix C
Chemical name
2-(Dibutylamino)ethanol
2-(Hydroxymethylamino)ethanol
2-(Thiocyanomethylthio)benzothiazole
2,2'-(Diazene-l,2-diyldiethane-l,l-
diyl)bis-4,5-dihydro-lH-imidazole
dihydrochloride
2,2'-(Octadecylimino)diethanol
2,2'-[Ethane-l,2-
diylbis(oxy)]diethanamine
2,2'-Azobis(2-amidinopropane)
dihydrochloride
2,2-Dibromo-3-nitrilopropionamide
2,2-Dibromopropanediamide
2,4-Hexadienoicacid, potassium salt,
(2E,4E)-
CASRN
102-81-8
34375-28-5
21564-17-0
27776-21-2
10213-78-2
929-59-9
2997-92-4
10222-01-2
73003-80-2
24634-61-5
Log Kow
Estimated
2.01
-1.53
3.12
2.12
6.85
-2.17
-3.28
1.01
0.37
1.62
Measured
2.65
-
3.3
-
-
-
-
0.82
-
1.33
Water solubility
Estimate from log Kow
(mg/Lat25°C)
3,297
1.00 x 10s
41.67
193.3
0.08076
1.00 x 10s
1.00 x 10s
2,841
1.00 x 104
1.94 x 104
Henry's law constant
(atm-m3/mol at 25°C)
Bond
method
9.70 x ID"9
1.62 x ID"12
6.49 x ID"12
S.llxlQ-14
1.06 x 1Q-8
2.50 x 1Q-13
1.21 x 1Q-14
6.16 x 1Q-14
3.58 x ID"14
5.72 x ID"7
Group
method 25
1.02 x ID"8
-
-
-
7.39 x ID"12
S.lOxlQ-16
-
-
-
4.99 x ID"8
Measured
-
-
-
-
-
-
-
1.91 x ID"8
-
-
                                    This document is a draft for review purposes only and does not constitute Agency policy.
                    June 2015                                          C-37                     DRAFT—DO NOT CITE OR QUOTE

-------
Hydraulic Fracturing Drinking Water Assessment
Appendix C
Chemical name
2,6,8-Trimethyl-4-nonanol
2-Acrylamido-2-methyl-l-propanesulfonic
acid
2-Amino-2-methylpropan-l-ol
2-Aminoethanol hydrochloride
2-Bromo-3-nitrilopropionamide
2-Butanone oxime
2-Butoxy-l-propanol
2-Butoxyethanol
2-Dodecylbenzenesulfonic acid- N-(2-
aminoethyl)ethane-l,2-diamine(l:l)
2-Ethoxyethanol
2-Ethoxynaphthalene
CASRN
123-17-1
15214-89-8
124-68-5
2002-24-6
1113-55-9
96-29-7
15821-83-7
111-76-2
40139-72-8
110-80-5
93-18-5
Log Kow
Estimated
4.48
-2.19
-0.74
-1.61
-0.31
1.69
0.98
0.57
4.78
-0.42
3.74
Measured
-
-
-
-1.31
-
0.63
-
0.83
-
-0.32
-
Water solubility
Estimate from log Kow
(mg/Lat25°C)
24.97
1.00 x 10s
1.00 x 10s
1.00 x 10s
3,274
3.66 x 104
4.21 xlO4
6.45 x 104
0.7032
7.55 x 105
38.32
Henry's law constant
(atm-m3/mol at 25°C)
Bond
method
9.63 x ID"5
S.lSxlQ-15
6.48 x ID"10
3.68 x ID"10
5.35 x 1Q-13
1.04 x 1Q-5
1.30 x ID"7
9.79 x 1Q-8
6.27 x 1Q-8
5.56 x 1Q-8
4.13 x 1Q-5
Group
method 25
4.45 x ID"4
-
-
9.96 x ID"11
-
-
4.88 x ID"8
2.08 x ID"8
-
1.04 x ID"8
4.06 x ID"4
Measured
-
-
-
-
-
-
-
1.60 x ID"6
-
4.70 x ID"7
-
                                    This document is a draft for review purposes only and does not constitute Agency policy.
                    June 2015                                          C-38                     DRAFT—DO NOT CITE OR QUOTE

-------
Hydraulic Fracturing Drinking Water Assessment
Appendix C
Chemical name
2-Ethyl-l-hexanol
2-Ethyl-2-hexenal
2-Ethylhexyl benzoate
2-Hydroxyethyl acrylate
2-Hydroxyethylammonium hydrogen
sulphite
2-Hydroxy-N,N-bis(2-hydroxyethyl)-N-
methylethanaminium chloride
2-Mercaptoethanol
2-Methoxyethanol
2-Methyl-l-propanol
2-Methyl-2,4-pentanediol
2-Methyl-3(2H)-isothiazolone
CASRN
104-76-7
645-62-5
5444-75-7
818-61-1
13427-63-9
7006-59-9
60-24-2
109-86-4
78-83-1
107-41-5
2682-20-4
Log Kow
Estimated
2.73
2.62
5.19
-0.25
-1.61
-6.7
-0.2
-0.91
0.77
0.58
-0.83
Measured
-
-
-
-0.21
-1.31
-
-
-0.77
0.76
-
-
Water solubility
Estimate from log Kow
(mg/Lat25°C)
1,379
548.6
1.061
5.07 x 105
1.00 x 10s
1.00 x 10s
1.94 x 105
1.00 x 10s
9.71 xlO4
3.26 x 104
5.37 x 105
Henry's law constant
(atm-m3/mol at 25°C)
Bond
method
S.lOxlQ-5
2.06 x ID"4
2.52 x ID"4
4.49 x 1Q-9
3.68 x ID"10
4.78 x 1Q-19
1.27 x ID"7
4.19 x 1Q-8
9.99 x 1Q-6
4.06 x ID"7
4.96 x 1Q-8
Group
method 25
4.66 x ID"5
4.88 x ID"4
2.34 x ID"4
7.22 x ID"10
9.96 x 1Q-11
-
3.38 x ID"8
7.73 x ID"9
1.17 x 1Q-5
3.97 x ID"10
-
Measured
2.65 x ID"5
-
-
-
-
-
1.80 x ID"7
3.30 x ID"7
9.78 x 1Q-6
-
-
                                    This document is a draft for review purposes only and does not constitute Agency policy.
                    June 2015                                          C-39                     DRAFT—DO NOT CITE OR QUOTE

-------
Hydraulic Fracturing Drinking Water Assessment
Appendix C
Chemical name
2-Methyl-3-butyn-2-ol
2-Methylbutane
2-Methylquinoline hydrochloride
2-Phosphono-l,2,4-butanetricarboxylic
acid
2-Phosphonobutane-l,2,4-tricarboxylic
acid, potassium salt (l:x)
2-Propenoic acid, 2-(2-
hydroxyethoxy)ethyl ester
3-(Dimethylamino)propylamine
3,4,4-Trimethyloxazolidine
3,5,7-Triazatricyclo(3.3.1.13,7))decane, 1-
(3-chloro-2-propenyl)-, chloride, (Z)-
3,7-Dimethyl-2,6-octadienal
CASRN
115-19-5
78-78-4
62763-89-7
37971-36-1
93858-78-7
13533-05-6
109-55-7
75673-43-7
51229-78-8
5392-40-5
Log Kow
Estimated
0.45
2.72
2.69
-1.66
-1.66
-0.52
-0.45
0.13
-5.92
3.45
Measured
0.28
-
2.59
-
-
-0.3
-
-
-
-
Water solubility
Estimate from log Kow
(mg/Lat25°C)
2.40 x 105
184.6
498.5
1.00 x 10s
1.00 x 10s
3.99 x 105
1.00 x 10s
8.22 x 105
1.00 x 10s
84.71
Henry's law constant
(atm-m3/mol at 25°C)
Bond
method
1.04 x ID"6
1.29
7.60 x ID"7
1.17xlQ-26
1.17xlQ-26
6.98 x ID"11
6.62 x ID"9
6.63 x ID"6
1.76 x ID"8
3.76 x ID"4
Group
method 25
-
1.44
2.13 x ID"6
-
-
1.54 x ID"12
4.45 x ID"9
-
-
4.35 x ID"5
Measured
3.91 x 1Q-6
1.40
-
-
-
-
-
-
-
-
                                    This document is a draft for review purposes only and does not constitute Agency policy.
                    June 2015                                          C-40                     DRAFT—DO NOT CITE OR QUOTE

-------
Hydraulic Fracturing Drinking Water Assessment
Appendix C
Chemical name
3-Hydroxybutanal
3-Methoxypropylamine
3-Phenylprop-2-enal
4,4-Dimethyloxazolidine
4,6-Dimethyl-2-heptanone
4-[Abieta-8,ll,13-trien-18-yl(3-oxo-3-
phenylpropyl)amino]butan-2-one
hydrochloride
4-Ethyloct-l-yn-3-ol
4-Hydroxy-3-methoxybenzaldehyde
4-Methoxybenzyl formate
4-Methoxyphenol
4-Methyl-2-pentanol
CASRN
107-89-1
5332-73-0
104-55-2
51200-87-4
19549-80-5
143106-84-7
5877-42-9
121-33-5
122-91-8
150-76-5
108-11-2
Log Kow
Estimated
-0.72
-0.42
1.82
-0.08
2.56
7.72
2.87
1.05
1.61
1.59
1.68
Measured
-
-
1.9
-
-
-
-
1.21
-
1.58
-
Water solubility
Estimate from log Kow
(mg/Lat25°C)
1.00 x 10s
1.00 x 10s
2,150
1.00 x 10s
528.8
0.002229
833.9
6,875
2,679
1.65 x 104
1.38 x 104
Henry's law constant
(atm-m3/mol at 25°C)
Bond
method
4.37 x ID"9
1.56 x ID"7
1.60 x 1Q-6
3.02 x 1Q-6
2.71 xlQ-4
2.49 x ID"12
4.27 x 1Q-6
8.27 x 1Q-11
1.15 x 1Q-6
3.32 x 1Q-8
1.76 x 1Q-5
Group
method 25
2.28 x ID"9
1.94 x ID"8
3.38 x ID"7
-
4.55 x ID"4
1.20 x ID"14
-
2.81 x ID"9
2.13 x ID"6
5.35 x ID"7
3.88 x 1Q-5
Measured
-
-
-
-
-
-
-
2.15xlQ-9
-
-
4.45 x ID"5
                                    This document is a draft for review purposes only and does not constitute Agency policy.
                    June 2015                                          C-41                     DRAFT—DO NOT CITE OR QUOTE

-------
Hydraulic Fracturing Drinking Water Assessment
Appendix C
Chemical name
4-Methyl-2-pentanone
4-Nonylphenol
5-Chloro-2-methyl-3(2H)-isothiazolone
Acetaldehyde
Acetic acid
Acetic acid, C6-8-branched alkyl esters
Acetic acid, hydroxy-, reaction products
with triethanolamine
Acetic acid, mercapto-, monoammonium
salt
Acetic anhydride
Acetone
Acetonitrile, 2,2',2"-nitrilotris-
CASRN
108-10-1
104-40-5
26172-55-4
75-07-0
64-19-7
90438-79-2
68442-62-6
5421-46-5
108-24-7
67-64-1
7327-60-8
Log Kow
Estimated
1.16
5.99
-0.34
-0.17
0.09
3.25
-2.48
0.03
-0.58
-0.24
-1.39
Measured
1.31
5.76
-
-0.34
-0.17
-
-1
0.09
-
-0.24
-
Water solubility
Estimate from log Kow
(mg/Lat25°C)
8,888
1.57
1.49 x 105
2.57 x 105
4.76 x 105
117.8
1.00 x 10s
2.56 x 105
3.59 x 105
2.20 x 105
1.00 x 10s
Henry's law constant
(atm-m3/mol at 25°C)
Bond
method
1.16 x ID"4
5.97 x 1Q-6
3.57 x 1Q-8
6.78 x 1Q-5
5.48 x ID"7
9.60 x ID"4
4.18xlO-12
1.94 x 1Q-8
3.57 x 1Q-5
4.96 x 1Q-5
2.61 x 1Q-15
Group
method 25
1.34 x ID"4
1.23 x ID"5
-
6.00 x ID"5
2.94 x ID"7
1.07 x 1Q-3
3.38 x 1Q-19
-
-
3.97 x ID"5
-
Measured
1.38 x ID"4
3.40 x 1Q-5
-
6.67 x ID"5
1.00 x ID"7
-
7.05 x ID"13
-
5.71 x ID"6
3.50 x ID"5
-
                                    This document is a draft for review purposes only and does not constitute Agency policy.
                    June 2015                                          C-42                     DRAFT—DO NOT CITE OR QUOTE

-------
Hydraulic Fracturing Drinking Water Assessment
Appendix C
Chemical name
Acetophenone
Acetyltriethyl citrate
Acrolein
Acrylamide
Acrylic acid
Acrylic acid, with sodium-2-acrylamido-2-
methyl-1-propanesulfonate and sodium
phosphinate
Alcohols, CIO- 12, ethoxylated
Alcohols, Cll-14-iso-, CIS-rich
Alcohols, Cll-14-iso-, CIS-rich,
ethoxylated
Alcohols, C12-13, ethoxylated
CASRN
98-86-2
77-89-4
107-02-8
79-06-1
79-10-7
110224-99-2
67254-71-1
68526-86-3
78330-21-9
66455-14-9
Log Kow
Estimated
1.67
1.34
0.19
-0.81
0.44
-2.19
5.47
5.19
4.91
5.96
Measured
1.58
-
-0.01
-0.67
0.35
-
-
-
-
-
Water solubility
Estimate from log Kow
(mg/Lat25°C)
4,484
688.2
1.40 x 105
5.04 x 105
1.68 x 105
1.00 x 10s
0.9301
5.237
5.237
0.2995
Henry's law constant
(atm-m3/mol at 25°C)
Bond
method
9.81 x ID"6
6.91 x 1Q-11
3.58 x 1Q-5
5.90 x 1Q-9
2.89 x ID"7
S.lSxlQ-15
1.95 x ID"2
1.28 x ID"4
1.25 x 1Q-6
2.58 x ID"2
Group
method 25
1.09 x ID"5
-
1.94 x ID"5
-
1.17 x ID"7
-
2.03 x ID"2
2.62 x ID"4
7.73 x ID"7
2.87 x ID"2
Measured
1.04 x 1Q-5
-
1.22 x ID"4
1.70 x ID"9
3.70 x ID"7
-
-
-
-
-
                                    This document is a draft for review purposes only and does not constitute Agency policy.
                    June 2015                                          C-43                     DRAFT—DO NOT CITE OR QUOTE

-------
Hydraulic Fracturing Drinking Water Assessment
Appendix C
Chemical name
Alcohols, C12-14, ethoxylated
propoxylated
Alcohols, C12-14-secondary
Alcohols, C12-16, ethoxylated
Alcohols, C14-15, ethoxylated
Alcohols, C6-12, ethoxylated
Alcohols, C7-9-iso-, C8-rich, ethoxylated
Alcohols, C9-11, ethoxylated
Alcohols, C9-11-ISO-, ClO-rich,
ethoxylated
Alkanes, C12-14-iso-
Alkanes, C13-16-iso-
Alkenes, OlOalpha-
CASRN
68439-51-0
126950-60-5
68551-12-2
68951-67-7
68439-45-2
78330-19-5
68439-46-3
78330-20-8
68551-19-9
68551-20-2
64743-02-8
Log Kow
Estimated
6.67
5.19
6.45
7.43
4.49
2.46
4.98
4.9
6.65
7.63
8.55
Measured
-
-
-
-
-
-
-
-
-
-
-
Water solubility
Estimate from log Kow
(mg/Lat25°C)
0.02971
5.237
0.09603
0.009765
8.832
1,513
2.874
3.321
0.03173
0.003311
0.0003941
Henry's law constant
(atm-m3/mol at 25°C)
Bond
method
7.08 x ID"4
1.28 x ID"4
3.43 x ID"2
6.04 x ID"2
l.lOxlQ-2
3.04 x ID"7
1.47 x ID"2
1.47 x ID"2
1.24 x 101
2.19X101
8.09
Group
method 25
1.23 x ID"4
3.62 x ID"4
4.06 x ID"2
S.lOxlQ-2
1.02 x ID"2
1.38 x ID"7
1.44 x ID"2
2.39 x ID"2
2.28 xlO1
4.55 x 101
2.39 x 101
Measured
-
-
-
-
-
-
-
-
-
-
-
                                    This document is a draft for review purposes only and does not constitute Agency policy.
                    June 2015                                          C-44                     DRAFT—DO NOT CITE OR QUOTE

-------
Hydraulic Fracturing Drinking Water Assessment
Appendix C
Chemical name
Alkyl* dimethyl ethylbenzyl ammonium
chloride *(50%C12, 30%C14, 17%C16,
3%C18)
Alkyl* dimethyl ethylbenzyl ammonium
chloride *(60%C14, 30%C16, 5%C12,
5%C18)
Alkylbenzenesulfonate, linear
alpha-Lactose monohydrate
alpha-Terpineol
Amaranth
Aminotrimethylene phosphonic acid
Ammonium acetate
Ammonium acrylate
Ammonium citrate (1:1)
CASRN
85409-23-0_l
68956-79-6
42615-29-2
5989-81-1
98-55-5
915-67-3
6419-19-8
631-61-8
10604-69-0
7632-50-0
Log Kow
Estimated
3.97
4.95
4.71
-5.12
3.33
1.63
-5.45
0.09
0.44
-1.67
Measured
-
-
-
-
2.98
-
-3.53
-0.17
0.35
-1.64
Water solubility
Estimate from log Kow
(mg/Lat25°C)
3.23
0.3172
0.8126
1.00 x 10s
371.7
1.789
1.00 x 10s
4.76 x 105
1.68 x 105
1.00 x 10s
Henry's law constant
(atm-m3/mol at 25°C)
Bond
method
l.llxlQ-11
1.96 x ID"11
6.27 x ID"8
4.47 x ID"22
1.58 x 1Q-5
1.49 x 1Q-30
1.65 x 1Q-34
5.48 x ID"7
2.89 x ID"7
8.33 x 1Q-18
Group
method 25
-
-
-
9.81 x ID"45
3.15 x ID"6
-
-
2.94 x ID"7
1.17 x ID"7
-
Measured
-
-
-
-
1.22 x ID"5
-
-
1.00 x ID"7
3.70 x ID"7
4.33 x 1Q-14
                                    This document is a draft for review purposes only and does not constitute Agency policy.
                    June 2015                                          C-45                     DRAFT—DO NOT CITE OR QUOTE

-------
Hydraulic Fracturing Drinking Water Assessment
Appendix C
Chemical name
Ammonium citrate (2:1)
Ammonium dodecyl sulfate
Ammonium hydrogen carbonate
Ammonium lactate
Anethole
Aniline
Benactyzine hydrochloride
Benzamorf
Benzene
Benzene, C10-16-alkyl derivatives
Benzenesulfonic acid
Benzenesulfonic acid, (1-methylethyl)-,
CASRN
3012-65-5
2235-54-3
1066-33-7
515-98-0
104-46-1
62-53-3
57-37-4
12068-08-5
71-43-2
68648-87-3
98-11-3
37953-05-2
Log Kow
Estimated
-1.67
2.42
-0.46
-0.65
3.39
1.08
2.89
4.71
1.99
8.43
-1.17
0.29
Measured
-1.64
-
-
-0.72
-
0.9
-
-
2.13
9.36
-
-
Water solubility
Estimate from log Kow
(mg/Lat25°C)
1.00 x 10s
163.7
8.42 x 105
1.00 x 10s
98.68
2.08 x 104
292.1
0.8126
2,000
0.0002099
6.90 x 105
2.46 x 104
Henry's law constant
(atm-m3/mol at 25°C)
Bond
method
8.33 x ID"18
1.84 x ID"7
6.05 x 1Q-9
1.13 x ID"7
2.56 x ID"4
1.90 x 1Q-6
2.07 x ID"10
6.27 x 1Q-8
5.39 x 1Q-3
1.78 x ID"1
2.52 x 1Q-9
4.89 x 1Q-9
Group
method 25
-
-
-
-
2.23 x ID"3
2.18xlQ-6
-
-
5.35 x ID"3
3.97 x ID"1
-
-
Measured
4.33 x ID"14
-
-
8.13xlQ-8
-
2.02 x ID"6
-
-
5.55 x ID"3
-
-
-
                                    This document is a draft for review purposes only and does not constitute Agency policy.
                    June 2015                                          C-46                     DRAFT—DO NOT CITE OR QUOTE

-------
Hydraulic Fracturing Drinking Water Assessment
Appendix C
Chemical name
Benzenesulfonic acid, (1-methylethyl)-,
ammonium salt
Benzenesulfonic acid, (1-methylethyl)-,
sodium salt
Benzenesulfonic acid, C10-16-alkyl
derivatives, compounds with
cyclohexylamine
Benzenesulfonic acid, C10-16-alkyl
derivatives, compounds with
triethanolamine
Benzenesulfonic acid, C10-16-alkyl
derivatives, potassium salts
Benzenesulfonic acid, dodecyl-, branched,
compounds with 2-propanamine
Benzenesulfonic acid, mono-C10-16-alkyl
derivatives, sodium salts
Benzoicacid
CASRN
37475-88-0
28348-53-0
255043-08-4
68584-25-8
68584-27-0
90218-35-2
68081-81-2
65-85-0
Log Kow
Estimated
0.29
0.29
4.71
5.2
5.2
4.49
4.22
1.87
Measured
-
-
-
-
-
-
-
1.87
Water solubility
Estimate from log Kow
(mg/Lat25°C)
2.46 x 104
2.46 x 104
0.8126
0.255
0.255
1.254
2.584
2,493
Henry's law constant
(atm-m3/mol at 25°C)
Bond
method
4.89 x ID"9
4.89 x 1Q-9
6.27 x 1Q-8
8.32 x 1Q-8
8.32 x 1Q-8
6.27 x 1Q-8
4.72 x 1Q-8
1.08 x ID"7
Group
method 25
-
-
-
-
-
-
-
4.55 x ID"8
Measured
-
-
-
-
-
-
-
3.81 x ID"8
                                    This document is a draft for review purposes only and does not constitute Agency policy.
                    June 2015                                          C-47                     DRAFT—DO NOT CITE OR QUOTE

-------
Hydraulic Fracturing Drinking Water Assessment
Appendix C
Chemical name
Benzyl chloride
Benzyldimethyldodecylammonium
chloride
Benzylhexadecyldimethylammonium
chloride
Benzyltrimethylammonium chloride
Bicine
Bis(l-methylethyl)naphthalenesulfonic
acid, cyclohexylamine salt
Bis(2-chloroethyl) ether
Bisphenol A
Bronopol
Butane
CASRN
100-44-7
139-07-1
122-18-9
56-93-9
150-25-4
68425-61-6
111-44-4
80-05-7
52-51-7
106-97-8
Log Kow
Estimated
2.79
2.93
4.89
-2.47
-3.27
2.92
1.56
3.64
-1.51
2.31
Measured
2.3
-
-
-
-
-
1.29
3.32
-
2.89
Water solubility
Estimate from log Kow
(mg/Lat25°C)
1,030
36.47
0.3543
1.00 x 10s
3.52 x 105
43.36
6,435
172.7
8.37 x 105
135.6
Henry's law constant
(atm-m3/mol at 25°C)
Bond
method
2.09 x ID"3
7.61 xlQ-12
2.36 x 1Q-11
3.37 x 1Q-13
1.28 x 1Q-14
9.29 x ID"10
1.89 x ID"4
9.16 x ID"12
6.35 x ID"21
9.69 x ID"1
Group
method 25
3.97 x ID"4
-
-
-
-
-
4.15 x ID"7
-
-
8.48 x ID"1
Measured
4.12xlQ-4
-
-
-
-
-
1.70 x ID"5
-
-
9.50 x ID"1
                                    This document is a draft for review purposes only and does not constitute Agency policy.
                    June 2015                                          C-48                     DRAFT—DO NOT CITE OR QUOTE

-------
Hydraulic Fracturing Drinking Water Assessment
Appendix C
Chemical name
Butanedioic acid, sulfo-, l,4-bis(l,3-
dimethylbutyl) ester, sodium salt
Butene
Butyl glycidyl ether
Butyl lactate
Butyryl trihexyl citrate
C.I. Acid Red 1
C.I. Acid Violet 12, disodium salt
C.I. Pigment Red 5
C.I. Solvent Red 26
CIO- 16-Alkyldimethyla mines oxides
C10-C16 Ethoxylated alcohol
CASRN
2373-38-8
25167-67-3
2426-08-6
138-22-7
82469-79-2
3734-67-6
6625-46-3
6410-41-9
4477-79-6
70592-80-2
68002-97-1
Log Kow
Estimated
3.98
2.17
1.08
0.8
8.21
0.51
0.59
7.65
9.27
2.87
4.99
Measured
-
2.4
0.63
-
-
-
-
-
-
-
-
Water solubility
Estimate from log Kow
(mg/Lat25°C)
0.1733
354.8
2.66 x 104
5.30 x 104
5.56 x ID"5
6.157
3.379
4.38 x 1Q-5
5.68 x 1Q-5
89.63
4.532
Henry's law constant
(atm-m3/mol at 25°C)
Bond
method
1.61 x ID"12
2.03 x ID"1
4.37 x 1Q-6
8.49 x 1Q-5
3.65 x 1Q-9
3.73 x 1Q-29
2.21 x 1Q-30
4.36 x ID"21
5.48 x 1Q-13
1.14xlO-13
1.25 x 1Q-6
Group
method 25
-
2.68 x ID"1
5.23 x ID"7
-
-
-
-
-
4.66 x ID"13
-
4.66 x ID"7
Measured
-
2.33 x ID"1
2.47 x ID"5
1.92 x 1Q-6
-
-
-
-
-
-
-
                                    This document is a draft for review purposes only and does not constitute Agency policy.
                    June 2015                                          C-49                     DRAFT—DO NOT CITE OR QUOTE

-------
Hydraulic Fracturing Drinking Water Assessment
Appendix C
Chemical name
C12-14 tert-Alkyl ethoxylated amines
Calcium dodecylbenzene sulfonate
Camphor
Carbon dioxide
Carbonic acid, dipotassium salt
Choline bicarbonate
Choline chloride
Citric acid
Citronellol
Coconut trimethylammonium chloride
Coumarin
Cumene
CASRN
73138-27-9
26264-06-2
76-22-2
124-38-9
584-08-7
78-73-9
67-48-1
77-92-9
106-22-9
61789-18-2
91-64-5
98-82-8
Log Kow
Estimated
3.4
4.71
3.04
0.83
-0.46
-5.16
-5.16
-1.67
3.56
1.22
1.51
3.45
Measured
-
-
2.38
0.83
-
-
-
-1.64
3.91
-
1.39
3.66
Water solubility
Estimate from log Kow
(mg/Lat25°C)
264.2
0.8126
339.1
2.57 x 104
8.42 x 105
1.00 x 10s
1.00 x 10s
1.00 x 10s
105.5
2,816
5,126
75.03
Henry's law constant
(atm-m3/mol at 25°C)
Bond
method
1.29 x ID"10
6.27 x 1Q-8
7.00 x 1Q-5
1.52 x ID"2
6.05 x 1Q-9
2.03 x 1Q-16
2.03 x 1Q-16
8.33 x 1Q-18
5.68 x 1Q-5
9.42 x 1Q-11
6.95 x 1Q-6
1.05 x ID"2
Group
method 25
-
-
-
-
-
-
-
-
2.13 x ID"5
-
-
1.23 x ID"2
Measured
-
-
S.lOxlQ-5
1.52 x ID"2
-
-
-
4.33 x ID"14
-
-
9.92 x ID"8
l.lSxlQ-2
                                    This document is a draft for review purposes only and does not constitute Agency policy.
                    June 2015                                          C-50                     DRAFT—DO NOT CITE OR QUOTE

-------
Hydraulic Fracturing Drinking Water Assessment
Appendix C
Chemical name
Cyclohexane
Cyclohexanol
Cyclohexanone
Cyclohexylamine sulfate
D&C Red no. 28
D&C Red no. 33
Daidzein
Dapsone
Dazomet
Decyldimethylamine
D-Glucitol
D-Gluconic acid
CASRN
110-82-7
108-93-0
108-94-1
19834-02-7
18472-87-2
3567-66-6
486-66-8
80-08-0
533-74-4
1120-24-7
50-70-4
526-95-4
Log Kow
Estimated
3.18
1.64
1.13
1.63
9.62
0.48
2.55
0.77
0.94
4.46
-3.01
-1.87
Measured
3.44
1.23
0.81
1.49
-
-
-
0.97
0.63
-
-2.2
-
Water solubility
Estimate from log Kow
(mg/Lat25°C)
43.02
3.37 x 104
2.41 x 104
6.40 x 104
1.64 x ID"8
11.87
568.4
3,589
1.94 x 104
82.23
1.00 x 10s
1.00 x 10s
Henry's law constant
(atm-m3/mol at 25°C)
Bond
method
2.55 x ID"1
4.90 x 1Q-6
S.llxlQ-5
1.38 x 1Q-5
6.37 x ID"21
1.15 x 1Q-26
3.91 x 1Q-16
S.llxlQ-14
2.84 x 1Q-3
4.68 x ID"4
7.26 x 1Q-13
4.74 x 1Q-13
Group
method 25
1.94 x ID"1
3.70 x ID"6
1.28 x 1Q-5
-
-
-
-
-
-
2.45 x ID"3
2.94 x ID"29
-
Measured
1.50 x ID"1
4.40 x 1Q-6
9.00 x 1Q-6
4.16xlQ-6
-
-
-
-
4.98 x ID"10
-
-
-
                                    This document is a draft for review purposes only and does not constitute Agency policy.
                    June 2015                                          C-51                     DRAFT—DO NOT CITE OR QUOTE

-------
Hydraulic Fracturing Drinking Water Assessment
Appendix C
Chemical name
D-Glucopyranoside, methyl
D-Glucose
Di(2-ethylhexyl) phthalate
Dibromoacetonitrile
Dichloromethane
Didecyldimethylammonium chloride
Diethanolamine
Diethylbenzene
Diethylene glycol
Diethylene glycol monomethyl ether
Diethylenetriamine
Diisobutyl ketone
CASRN
3149-68-6
50-99-7
117-81-7
3252-43-5
75-09-2
7173-51-5
111-42-2
25340-17-4
111-46-6
111-77-3
111-40-0
108-83-8
Log Kow
Estimated
-2.5
-2.89
8.39
0.47
1.34
4.66
-1.71
4.07
-1.47
-1.18
-2.13
2.56
Measured
-
-3.24
7.6
-
1.25
-
-1.43
3.72
-
-
-
-
Water solubility
Estimate from log Kow
(mg/Lat25°C)
1.00 x 10s
1.00 x 10s
0.001132
9,600
l.lOxlO4
0.9
1.00 x 10s
58.86
1.00 x 10s
1.00 x 10s
1.00 x 10s
528.8
Henry's law constant
(atm-m3/mol at 25°C)
Bond
method
1.56 x ID"14
9.72 x 1Q-15
l.lSxlQ-5
4.06 x ID"7
9.14xlQ-3
6.85 x ID"10
3.92 x 1Q-11
1.16 x ID"2
2.03 x 1Q-9
6.50 x ID"10
S.lOxlQ-13
2.71 x ID"4
Group
method 25
2.23 x ID"24
1.62 x ID"26
1.02 x 1Q-5
-
3.01 x ID"3
-
3.46 x ID"15
1.47 x ID"2
1.20 x 1Q-13
1.65 x 1Q-11
1.09 x ID"14
4.55 x ID"4
Measured
-
-
2.70 x ID"7
-
3.25 x ID"3
-
3.87 x ID"11
2.61 x ID"3
-
-
-
1.17xlQ-4
                                    This document is a draft for review purposes only and does not constitute Agency policy.
                    June 2015                                          C-52                     DRAFT—DO NOT CITE OR QUOTE

-------
Hydraulic Fracturing Drinking Water Assessment
Appendix C
Chemical name
Diisopropanolamine
Diisopropylnaphthalene
Dimethyl adipate
Dimethyl glutarate
Dimethyl succinate
Dimethylaminoethanol
Dimethyldiallylammonium chloride
Diphenyl oxide
Dipropyleneglycol
Di-sec-butylphenol
Disodium
dodecyl(sulphonatophenoxy)benzenesulp
honate
CASRN
110-97-4
38640-62-9
627-93-0
1119-40-0
106-65-0
108-01-0
7398-69-8
101-84-8
25265-71-8
31291-60-8
28519-02-0
Log Kow
Estimated
-0.88
6.08
1.39
0.9
0.4
-0.94
-2.49
4.05
-0.64
5.41
5.05
Measured
-0.82
-
1.03
0.62
0.35
-
-
4.21
-
-
-
Water solubility
Estimate from log Kow
(mg/Lat25°C)
1.00 x 10s
0.2421
7,749
2.02 x 104
3.96 x 104
1.00 x 10s
1.00 x 10s
15.58
S.llxlO5
3.723
0.0353
Henry's law constant
(atm-m3/mol at 25°C)
Bond
method
6.91 x ID"11
1.99 x 1Q-3
9.77 x ID"7
7.36 x ID"7
5.54 x ID"7
1.77 x 1Q-9
7.20 x ID"12
l.lSxlQ-4
3.58 x 1Q-9
3.74 x 1Q-6
6.40 x 1Q-16
Group
method 25
1.90 x ID"14
1.94 x ID"3
1.28 x ID"7
9.09 x 1Q-8
6.43 x 1Q-8
1.77 x 1Q-9
-
2.81 x ID"4
6.29 x ID"10
6.89 x 1Q-6
-
Measured
-
-
2.31 x ID"6
6.43 x ID"7
-
3.73 x ID"7
-
2.79 x ID"4
-
-
-
                                    This document is a draft for review purposes only and does not constitute Agency policy.
                    June 2015                                          C-53                     DRAFT—DO NOT CITE OR QUOTE

-------
Hydraulic Fracturing Drinking Water Assessment
Appendix C
Chemical name
Disodium ethylenediaminediacetate
Disodium ethylenediaminetetraacetate
dihydrate
D-Lactic acid
D-Limonene
Docusate sodium
Dodecane
Dodecylbenzene
Dodecylbenzenesulfonic acid
Dodecylbenzenesulfonic acid,
monoethanolamine salt
Epichlorohydrin
CASRN
38011-25-5
6381-92-6
10326-41-7
5989-27-5
577-11-7
112-40-3
123-01-3
27176-87-0
26836-07-7
106-89-8
Log Kow
Estimated
-4.79
-3.86
-0.65
4.83
6.1
6.23
7.94
4.71
4.71
0.63
Measured
-
-
-0.72
4.57
-
6.1
8.65
-
-
0.45
Water solubility
Estimate from log Kow
(mg/Lat25°C)
1.00 x 10s
2.28 x 105
1.00 x 10s
4.581
0.001227
0.1099
0.001015
0.8126
0.8126
5.06 x 104
Henry's law constant
(atm-m3/mol at 25°C)
Bond
method
l.lOxlQ-16
1.17xlO-23
1.13 x ID"7
3.80 x ID"1
5.00 x ID"12
9.35
1.34 x ID"1
6.27 x 1Q-8
6.27 x 1Q-8
5.62 x 1Q-5
Group
method 25
-
-
-
-
-
1.34 x 101
2.81 x ID"1
-
-
2.62 x ID"6
Measured
-
5.77 x ID"16
8.13xlQ-8
3.19X1Q-2
-
8.18
-
-
-
3.04 x ID"5
                                    This document is a draft for review purposes only and does not constitute Agency policy.
                    June 2015                                          C-54                     DRAFT—DO NOT CITE OR QUOTE

-------
Hydraulic Fracturing Drinking Water Assessment
Appendix C
Chemical name
Ethanaminium, N,N,N-trimethyl-2-[(l-
oxo-2-propenyl)oxy]-, chloride
Ethane
Ethanol
Ethanol, 2,2',2"-nitrilotris-,
tris(dihydrogen phosphate) (ester),
sodium salt
Ethanol, 2-[2-[2-(tridecyloxy)
ethoxy]ethoxy]-, hydrogen sulfate,
sodium salt
Ethanolamine
Ethoxylated dodecyl alcohol
Ethyl acetate
Ethyl acetoacetate
Ethyl benzoate
CASRN
44992-01-0
74-84-0
64-17-5
68171-29-9
25446-78-0
141-43-5
9002-92-0
141-78-6
141-97-9
93-89-0
Log Kow
Estimated
-3.1
1.32
-0.14
-3.13
2.09
-1.61
4.5
0.86
-0.2
2.32
Measured
-
1.81
-0.31
-
-
-1.31
-
0.73
0.25
2.64
Water solubility
Estimate from log Kow
(mg/Lat25°C)
1.00 x 10s
938.6
7.92 x 105
1.00 x 10s
42
1.00 x 10s
14.19
2.99 x 104
5.62 x 104
421.5
Henry's law constant
(atm-m3/mol at 25°C)
Bond
method
6.96 x ID"15
5.50 x ID"1
5.67 x 1Q-6
3.08 x 1Q-36
9.15xlO-13
3.68 x ID"10
9.45 x ID"7
2.33 x ID"4
1.57 x ID"7
4.61 x 1Q-5
Group
method 25
-
4.25 x ID"1
4.88 x ID"6
-
-
9.96 x ID"11
3.30 x ID"7
1.58 x ID"4
-
2.45 x ID"5
Measured
-
5.00 x ID"1
5.00 x ID"6
-
-
-
-
1.34 x ID"4
1.20 x ID"6
7.33 x 1Q-5
                                    This document is a draft for review purposes only and does not constitute Agency policy.
                    June 2015                                          C-55                     DRAFT—DO NOT CITE OR QUOTE

-------
Hydraulic Fracturing Drinking Water Assessment
Appendix C
Chemical name
Ethyl lactate
Ethyl salicylate
Ethylbenzene
Ethylene
Ethylene glycol
Ethylene oxide
Ethylenediamine
Ethylenediaminetetraacetic acid
Ethylenediaminetetraacetic acid
tetrasodium salt
Ethylenediaminetetraacetic acid,
disodium salt
Ethyne
CASRN
97-64-3
118-61-6
100-41-4
74-85-1
107-21-1
75-21-8
107-15-3
60-00-4
64-02-8
139-33-3
74-86-2
Log Kow
Estimated
-0.18
3.09
3.03
1.27
-1.2
-0.05
-1.62
-3.86
-3.86
-3.86
0.5
Measured
-
2.95
3.15
1.13
-1.36
-0.3
-2.04
-
-
-
0.37
Water solubility
Estimate from log Kow
(mg/Lat25°C)
4.73 x 105
737.1
228.6
3,449
1.00 x 10s
2.37 x 105
1.00 x 10s
2.28 x 105
2.28 x 105
2.28 x 105
1.48 x 104
Henry's law constant
(atm-m3/mol at 25°C)
Bond
method
4.82 x ID"5
6.04 x 1Q-6
7.89 x 1Q-3
9.78 x ID"2
1.31 x ID"7
1.20 x ID"4
1.03 x 1Q-9
1.17 x 10-23
1.17xlQ-23
1.17xlQ-23
2.40 x ID"2
Group
method 25
-
3.01 x ID"9
8.88 x ID"3
1.62 x ID"1
5.60 x 1Q-11
5.23 x 1Q-5
1.77 x ID"10
-
-
-
2.45 x ID"2
Measured
5.83 x ID"7
-
7.88 x ID"3
2.28 x ID"1
6.00 x 1Q-8
1.48 x ID"4
1.73 x 1Q-9
5.77 x 1Q-16
5.77 x 1Q-16
5.77 x 1Q-16
2.17xlO-2
                                    This document is a draft for review purposes only and does not constitute Agency policy.
                    June 2015                                          C-56                     DRAFT—DO NOT CITE OR QUOTE

-------
Hydraulic Fracturing Drinking Water Assessment
Appendix C
Chemical name
Fatty acids, C18-unsaturated, dimers
FD&C Blue no. 1
FD&C Yellow no. 5
FD&C Yellow no. 6
Formaldehyde
Formamide
Formic acid
Formic acid, potassium salt
Fumaricacid
Furfural
Furfuryl alcohol
Galantamine hydrobromide
CASRN
61788-89-4
3844-45-9
1934-21-0
2783-94-0
50-00-0
75-12-7
64-18-6
590-29-4
110-17-8
98-01-1
98-00-0
69353-21-5
Log Kow
Estimated
14.6
-0.15
-1.82
1.4
0.35
-1.61
-0.46
-0.46
0.05
0.83
0.45
2.29
Measured
-
-
-
-
0.35
-1.51
-0.54
-0.54
-0.48
0.41
0.28
-
Water solubility
Estimate from log Kow
(mg/Lat25°C)
2.31 x ID"10
0.2205
7.388
242.7
5.70 x 104
1.00 x 10s
9.55 x 105
9.55 x 105
1.04 x 105
5.36 x 104
2.21 xlO5
1,606
Henry's law constant
(atm-m3/mol at 25°C)
Bond
method
4.12 x ID"8
2.25 x 1Q-35
1.31 x 1Q-28
3.26 x 1Q-23
9.29 x 1Q-5
1.53 x 1Q-8
7.50 x ID"7
7.50 x ID"7
1.35 x ID"12
1.34 x 1Q-5
2.17 x ID"7
1.70 x 1Q-13
Group
method 25
9.74 x ID"9
-
-
-
6.14xlQ-5
-
S.llxlQ-7
S.llxlQ-7
8.48 x ID"14
-
-
-
Measured
-
-
-
-
3.37 x ID"7
1.39 x ID"9
1.67 x ID"7
1.67 x ID"7
-
3.77 x ID"6
7.86 x ID"8
-
                                    This document is a draft for review purposes only and does not constitute Agency policy.
                    June 2015                                          C-57                     DRAFT—DO NOT CITE OR QUOTE

-------
Hydraulic Fracturing Drinking Water Assessment
                                                                                                    Appendix C
Chemical name
Gluconic acid
Glutaraldehyde
Glycerol
Glycine, N-(carboxymethyl)-N-(2-
hydroxy ethyl)-, disodium salt
Glycine, N-(hydroxymethyl)-,
monosodium salt
Glycine, N,N-bis(carboxymethyl)-,
trisodium salt
Glycine, N-[2-
[bis(carboxymethyl)amino]ethyl]-N-(2-
hydroxyethyl)-, trisodium salt
Glycolicacid
Glycolic acid sodium salt
Glyoxal
CASRN
133-42-6
111-30-8
56-81-5
135-37-5
70161-44-3
5064-31-3
139-89-9
79-14-1
2836-32-0
107-22-2
Log Kow
Estimated
-1.87
-0.18
-1.65
-3.04
-3.41
-3.81
-4.09
-1.07
-1.07
-1.66
Measured
-
-
-1.76
-
-
-
-
-1.11
-1.11
-
Water solubility
Estimate from log Kow
(mg/Lat25°C)
1.00 x 10s
1.67 x 105
1.00 x 10s
1.90 x 105
7.82 x 105
7.39 x 105
4.31 x 105
1.00 x 10s
1.00 x 10s
1.00 x 10s
Henry's law constant
(atm-m3/mol at 25°C)
Bond
method
4.74 x ID"13
l.lOxlQ-7
6.35 x 1Q-9
3.90 x 1Q-17
1.80 x ID"12
1.19xlO-16
3.81 x ID"24
8.54 x ID"8
8.54 x ID"8
3.70 x ID"7
Group
method 25
-
2.39 x ID"8
1.51 x ID"15
-
-
-
-
6.29 x ID"11
6.29 x ID"11
-
Measured
-
-
1.73 x ID"8
-
-
-
-
-
-
3.33 x ID"9
                    June 2015
This document is a draft for review purposes only and does not constitute Agency policy.
                                   C-58                     DRAFT—DO NOT CITE OR QUOTE

-------
Hydraulic Fracturing Drinking Water Assessment
Appendix C
Chemical name
Glyoxylic acid
Heptane
Hexadecyltrimethylammonium bromide
Hexane
Hexanedioicacid
Hydroxyvalerenic acid
Indole
Isoascorbic acid
Isobutane
Isobutene
Isooctanol
Isopentyl alcohol
CASRN
298-12-4
142-82-5
57-09-0
110-54-3
124-04-9
1619-16-5
120-72-9
89-65-6
75-28-5
115-11-7
26952-21-6
123-51-3
Log Kow
Estimated
-1.4
3.78
3.18
3.29
0.23
3.31
2.05
-1.88
2.23
2.23
2.73
1.26
Measured
-
4.66
-
3.9
0.08
-
2.14
-1.85
2.76
2.34
-
1.16
Water solubility
Estimate from log Kow
(mg/Lat25°C)
1.00 x 10s
3.554
28.77
17.24
1.67 x 105
282.1
1,529
1.00 x 10s
175.1
399.2
1,379
4.16xl04
Henry's law constant
(atm-m3/mol at 25°C)
Bond
method
2.98 x ID"9
2.27
2.93 x ID"10
1.71
9.53 x ID"12
-
8.86 x ID"7
4.07 x ID"8
9.69 x ID"1
2.40 x ID"1
S.lOxlQ-5
1.33 x 1Q-5
Group
method 25
-
2.39
-
1.69
S.lOxlQ-13
-
1.99 x ID"6
-
1.02
2.34 x ID"1
4.66 x ID"5
1.65 x 1Q-5
Measured
-
2.00
-
1.80
4.71 x ID"12
-
5.28 x ID"7
-
1.19
2.18X1Q-1
9.21 x ID"5
1.41 x 1Q-5
                                    This document is a draft for review purposes only and does not constitute Agency policy.
                    June 2015                                          C-59                     DRAFT—DO NOT CITE OR QUOTE

-------
Hydraulic Fracturing Drinking Water Assessment
Appendix C
Chemical name
Isopropanol
Isopropanolamine dodecylbenzene
Isopropylamine
Isoquinoline
Isoquinoline, reaction products with
benzyl chloride and quinoline
Isoquinolinium, 2-(phenylmethyl)-,
chloride
Lactic acid
Lactose
Lauryl hydroxysultaine
L-Dilactide
L-Glutamic acid
CASRN
67-63-0
42504-46-1
75-31-0
119-65-3
68909-80-8
35674-56-7
50-21-5
63-42-3
13197-76-7
4511-42-6
56-86-0
Log Kow
Estimated
0.28
7.94
0.27
2.14
2.14
4.4
-0.65
-5.12
-1.3
1.65
-3.83
Measured
0.05
8.65
0.26
2.08
2.08
-
-0.72
-
-
-
-3.69
Water solubility
Estimate from log Kow
(mg/Lat25°C)
4.02 x 105
0.001015
8.38 x 105
1,551
1,551
6.02
1.00 x 10s
1.00 x 10s
7.71 xlO4
3,165
9.42 x 105
Henry's law constant
(atm-m3/mol at 25°C)
Bond
method
7.52 x ID"6
1.34 x ID"1
1.34 x 1Q-5
6.88 x ID"7
6.88 x ID"7
1.19xlQ-6
1.13 x ID"7
4.47 x ID"22
1.04 x ID"21
1.22 x ID"5
1.47 x ID"14
Group
method 25
1.14xlQ-5
2.81 x ID"1
-
4.15 x ID"7
4.15 x ID"7
-
-
9.81 x ID"45
-
-
-
Measured
S.lOxlQ-6
-
4.51 x ID"5
-
-
-
8.13xlQ-8
-
-
-
-
                                    This document is a draft for review purposes only and does not constitute Agency policy.
                    June 2015                                          C-60                     DRAFT—DO NOT CITE OR QUOTE

-------
Hydraulic Fracturing Drinking Water Assessment
Appendix C
Chemical name
L-Lactic acid
Methane
Methanol
Methenamine
Methoxyacetic acid
Methyl salicylate
Methyl vinyl ketone
Methylcyclohexane
Methylene bis(thiocyanate)
Methylenebis(5-methyloxazolidine)
Morpholine
CASRN
79-33-4
74-82-8
67-56-1
100-97-0
625-45-6
119-36-8
78-94-4
108-87-2
6317-18-6
66204-44-2
110-91-8
Log Kow
Estimated
-0.65
0.78
-0.63
-4.15
-0.68
2.6
0.41
3.59
0.62
-0.58
-0.56
Measured
-0.72
1.09
-0.77
-
-
2.55
-
3.61
-
-
-0.86
Water solubility
Estimate from log Kow
(mg/Lat25°C)
1.00 x 10s
2,610
1.00 x 10s
1.00 x 10s
1.00 x 10s
1,875
6.06 x 104
28.4
2.72 x 104
1.00 x 10s
1.00 x 10s
Henry's law constant
(atm-m3/mol at 25°C)
Bond
method
1.13 x ID"7
4.14X1Q-1
4.27 x 1Q-6
1.63 x ID"1
4.54 x 1Q-8
4.55 x 1Q-6
2.61 x 1Q-5
3.39 x ID"1
2.61 x 1Q-8
1.07 x ID"7
1.14xlQ-7
Group
method 25
-
6.58 x ID"1
3.62 x ID"6
-
8.68 x ID"9
2.23 x ID"9
1.38 x 1Q-5
3.30 x ID"1
-
-
3.22 x ID"9
Measured
8.13xlQ-8
6.58 x ID"1
4.55 x 1Q-6
1.64 x 1Q-9
6.42 x 1Q-9
9.81 x 1Q-5
4.65 x 1Q-5
4.30 x ID"1
-
-
i.iexio-6
                                    This document is a draft for review purposes only and does not constitute Agency policy.
                    June 2015                                          C-61                     DRAFT—DO NOT CITE OR QUOTE

-------
Hydraulic Fracturing Drinking Water Assessment
Appendix C
Chemical name
Morpholinium, 4-ethyl-4-hexadecyl-,
ethyl sulfate
N-(2-Acryloyloxyethyl)-N-benzyl-N,N-
dimethylammonium chloride
N-(3-Chloroallyl)hexaminium chloride
N,N,N-Trimethyl-3-((l-
oxooctadecyl)amino)-l-propanaminium
methyl sulfate
N,N,N-Trimethyloctadecan-l-aminium
chloride
N,N'-Dibutylthiourea
N,N-Dimethyldecylamine oxide
N,N-Dimethylformamide
N,N-Dimethylmethanamine
hydrochloride
CASRN
78-21-7
46830-22-2
4080-31-3
19277-88-4
112-03-8
109-46-6
2605-79-0
68-12-2
593-81-7
Log Kow
Estimated
4.54
-1.39
-5.92
4.38
4.17
2.57
1.4
-0.93
0.04
Measured
-
-
-
-
-
2.75
-
-1.01
0.16
Water solubility
Estimate from log Kow
(mg/Lat25°C)
0.9381
4.42 x 105
1.00 x 10s
0.7028
2.862
2,287
2,722
9.78 x 105
1.00 x 10s
Henry's law constant
(atm-m3/mol at 25°C)
Bond
method
2.66 x ID"12
5.62 x 1Q-16
1.76 x 1Q-8
2.28 x 1Q-16
5.16xlO-10
4.17 x 1Q-6
4.88 x 1Q-14
7.38 x 1Q-8
3.65 x 1Q-5
Group
method 25
-
-
-
-
-
-
-
-
1.28 x ID"4
Measured
-
-
-
-
-
-
-
7.39 x ID"8
1.04 x ID"4
                                    This document is a draft for review purposes only and does not constitute Agency policy.
                    June 2015                                          C-62                     DRAFT—DO NOT CITE OR QUOTE

-------
Hydraulic Fracturing Drinking Water Assessment
Appendix C
Chemical name
N,N-Dimethyl-methanamine-N-oxide
N,N-dimethyloctadecylamine
hydrochloride
N,N'-Methylenebisacrylamide
Naphthalene
Naphthalenesulfonicacid, bis(l-
methylethyl)-
Naphthalenesulphonic acid, bis (1-
methylethyl)-methyl derivatives
Naphthenic acid ethoxylate
Nitrilotriacetamide
Nitrilotriacetic acid
Nitrilotriacetic acid trisodium
monohydrate
CASRN
1184-78-7
1613-17-8
110-26-9
91-20-3
28757-00-8
99811-86-6
68410-62-8
4862-18-4
139-13-9
18662-53-8
Log Kow
Estimated
-3.02
8.39
-1.52
3.17
2.92
4.02
3.41
-4.75
-3.81
-3.81
Measured
-
-
-
3.3
-
-
-
-
-
-
Water solubility
Estimate from log Kow
(mg/Lat25°C)
1.00 x 10s
0.008882
7.01 x 104
142.1
43.36
3.45
112.5
1.00 x 10s
7.39 x 105
7.39 x 105
Henry's law constant
(atm-m3/mol at 25°C)
Bond
method
3.81 x ID"15
4.51 x 1Q-3
1.14xlQ-9
5.26 x ID"4
9.29 x ID"10
l.lSxlQ-9
3.62 x ID"8
1.61 x ID"18
1.19 x ID"16
1.19xlO-16
Group
method 25
-
3.88 x ID"2
-
3.70 x ID"4
-
-
2.74 x ID"9
-
-
-
Measured
-
-
-
4.40 x ID"4
-
-
-
-
-
-
                                    This document is a draft for review purposes only and does not constitute Agency policy.
                    June 2015                                          C-63                     DRAFT—DO NOT CITE OR QUOTE

-------
Hydraulic Fracturing Drinking Water Assessment
Appendix C
Chemical name
N-Methyl-2-pyrrolidone
N-Methyldiethanolamine
N-Methylethanolamine
N-Methyl-N-hydroxyethyl-N-
hydroxyethoxyethylamine
N-Oleyl diethanolamide
Oleic acid
Pentaethylenehexamine
Pentane
Pentyl acetate
Pentyl butyrate
Peracetic acid
CASRN
872-50-4
105-59-9
109-83-1
68213-98-9
13127-82-7
112-80-1
4067-16-7
109-66-0
628-63-7
540-18-1
79-21-0
Log Kow
Estimated
-0.11
-1.5
-1.15
-1.78
6.63
7.73
-3.67
2.8
2.34
3.32
-1.07
Measured
-0.38
-
-0.94
-
-
7.64
-
3.39
2.3
-
-
Water solubility
Estimate from log Kow
(mg/Lat25°C)
2.48 x 105
1.00 x 10s
1.00 x 10s
1.00 x 10s
0.1268
0.01151
1.00 x 10s
49.76
996.8
101.9
1.00 x 10s
Henry's law constant
(atm-m3/mol at 25°C)
Bond
method
3.16 x ID"8
8.61 x 1Q-11
8.07 x ID"10
1.34 x ID"12
9.35 x 1Q-9
4.48 x 1Q-5
8.36 x ID"24
1.29
5.45 x ID"4
9.60 x ID"4
1.39 x 1Q-6
Group
method 25
-
2.45 x ID"14
2.50 x ID"10
5.23 x 1Q-17
1.94 x ID"12
1.94 x 1Q-5
2.56 x ID"27
1.20
4.45 x ID"4
8.88 x ID"4
-
Measured
3.20 x ID"9
3. 14x10-"
-
-
-
-
-
1.25
3.88 x ID"4
-
2.14xlQ-6
                                    This document is a draft for review purposes only and does not constitute Agency policy.
                    June 2015                                          C-64                     DRAFT—DO NOT CITE OR QUOTE

-------
Hydraulic Fracturing Drinking Water Assessment
                                                                                                    Appendix C
Chemical name
Phenanthrene
Phenol
Phosphonic acid
(dimethylamino(methylene))
Phosphonic acid, (((2-[(2-hydroxyethyl)
(phosphonomethyl)amino)ethyl)imino]bis
(methylene))bis-, compd. with 2-
aminoethanol
Phosphonic acid, (1-hydroxyethylidene)
bis-, potassium salt
Phosphonic acid, (1-hydroxyethylidene)
bis-, tetrasodium salt
Phosphonic acid, [[(phosphonomethyl)
imino]bis[2,l-ethanediylnitrilobis
(methylene)]]tetrakis-
CASRN
85-01-8
108-95-2
29712-30-9
129828-36-0
67953-76-8
3794-83-0
15827-60-8
Log Kow
Estimated
4.35
1.51
-1.9
-6.73
-0.01
-0.01
-9.72
Measured
4.46
1.46
-
-
-
-
-
Water solubility
Estimate from log Kow
(mg/Lat25°C)
0.677
2.62 x 104
1.00 x 10s
1.00 x 10s
1.34 x 105
1.34 x 105
1.00 x 10s
Henry's law constant
(atm-m3/mol at 25°C)
Bond
method
5.13 x ID"5
5.61 x ID"7
1.00 x ID"24
5.29 x ID"42
9.79 x 1Q-26
9.79 x 1Q-26
-
Group
method 25
2.56 x ID"5
6.58 x ID"7
-
-
-
-
-
Measured
4.23 x ID"5
3.33 x ID"7
-
-
-
-
-
                    June 2015
This document is a draft for review purposes only and does not constitute Agency policy.
                                   C-65                     DRAFT—DO NOT CITE OR QUOTE

-------
Hydraulic Fracturing Drinking Water Assessment
                                                                                                    Appendix C
Chemical name
Phosphonic acid, [[(phosphonomethyl)
imino]bis[2,l-ethanediylnitrilobis
(methylene)]]tetrakis-, ammonium salt
(l:x)
Phosphonic acid, [[(phosphonomethyl)
imino]bis[2,l-ethanediylnitrilobis
(methylene)]]tetrakis-, sodium salt
Phosphonic acid, [[(phosphonomethyl)
imino]bis[6,l-hexanediylnitrilobis
(methylene)]]tetrakis-
Phthalic anhydride
Poly(oxy-l,2-ethanediyl),
.alpha.-(octylphenyl)-.omega.-hydroxy-,
branched
Potassium acetate
Potassium oleate
Propane
CASRN
70714-66-8
22042-96-2
34690-00-1
85-44-9
68987-90-6
127-08-2
143-18-0
74-98-6
Log Kow
Estimated
-9.72
-9.72
-5.79
2.07
5.01
0.09
7.73
1.81
Measured
-
-
-
1.6
-
-0.17
7.64
2.36
Water solubility
Estimate from log Kow
(mg/Lat25°C)
1.00 x 10s
1.00 x 10s
1.00 x 10s
3,326
3.998
4.76 x 105
0.01151
368.9
Henry's law constant
(atm-m3/mol at 25°C)
Bond
method
-
-
-
6.35 x ID"6
1.24 x ID"7
5.48 x ID"7
4.48 x 1Q-5
7.30 x ID"1
Group
method 25
-
-
-
-
1.07 x ID"6
2.94 x ID"7
1.94 x 1Q-5
6.00 x ID"1
Measured
-
-
-
1.63 x ID"8
-
1.00 x ID"7
-
7.07 x ID"1
                    June 2015
This document is a draft for review purposes only and does not constitute Agency policy.
                                   C-66                     DRAFT—DO NOT CITE OR QUOTE

-------
Hydraulic Fracturing Drinking Water Assessment
Appendix C
Chemical name
Propanol, l(or2)-(2-
methoxymethylethoxy)-
Propargyl alcohol
Propylene carbonate
Propylene pentamer
p-Xylene
Pyrimidine
Pyrrole
Quaternary ammonium compounds, di-
C8-10-alkyldimethyl, chlorides
Quinaldine
Quinoline
Rhodamine B
CASRN
34590-94-8
107-19-7
108-32-7
15220-87-8
106-42-3
289-95-2
109-97-7
68424-95-3
91-63-4
91-22-5
81-88-9
Log Kow
Estimated
-0.27
-0.42
0.08
6.28
3.09
-0.06
0.88
2.69
2.69
2.14
6.03
Measured
-
-0.38
-0.41
-
3.15
-0.4
0.75
-
2.59
2.03
-
Water solubility
Estimate from log Kow
(mg/Lat25°C)
4.27 x 105
9.36 x 105
2.58 x 105
0.05601
228.6
2.87 x 105
3.12xl04
90.87
498.5
1,711
0.0116
Henry's law constant
(atm-m3/mol at 25°C)
Bond
method
1.15 x ID"9
5.88 x ID"7
3.63 x ID"4
3.92 x ID"1
6.56 x 1Q-3
2.92 x 1Q-6
9.07 x 1Q-6
2.20 x ID"10
7.60 x ID"7
6.88 x ID"7
-
Group
method 25
1.69 x ID"9
-
-
1.09 x ID"3
6.14xlQ-3
-
7.73 x ID"6
-
2.13 x ID"6
1.54 x ID"6
-
Measured
-
l.lSxlQ-6
3.45 x ID"8
-
6.90 x ID"3
-
1.80 x ID"5
-
-
1.67 x ID"6
-
                                    This document is a draft for review purposes only and does not constitute Agency policy.
                    June 2015                                          C-67                     DRAFT—DO NOT CITE OR QUOTE

-------
Hydraulic Fracturing Drinking Water Assessment
Appendix C
Chemical name
Sodium 1-octanesulfonate
Sodium 2-mercaptobenzothiolate
Sodium acetate
Sodium benzoate
Sodium bicarbonate
Sodium bis(tridecyl) sulfobutanedioate
Sodium C14-16 alpha-olefin sulfonate
Sodium caprylamphopropionate
Sodium carbonate
Sodium chloroacetate
Sodium decyl sulfate
Sodium D-gluconate
CASRN
5324-84-5
2492-26-4
127-09-3
532-32-1
144-55-8
2673-22-5
68439-57-6
68610-44-6
497-19-8
3926-62-3
142-87-0
527-07-1
Log Kow
Estimated
1.06
2.86
0.09
1.87
-0.46
11.15
4.36
-0.26
-0.46
0.34
1.44
-1.87
Measured
-
2.42
-0.17
1.87
-
-
-
-
-
0.22
-
-
Water solubility
Estimate from log Kow
(mg/Lat25°C)
5,864
543.4
4.76 x 105
2,493
8.42 x 105
7.46 x ID"9
2.651
615.1
8.42 x 105
1.95 x 105
1,617
1.00 x 10s
Henry's law constant
(atm-m3/mol at 25°C)
Bond
method
9.15 x ID"8
3.63 x 1Q-8
5.48 x ID"7
1.08 x ID"7
6.05 x 1Q-9
8.51 x 1Q-11
4.95 x ID"7
1.19 x 1Q-9
6.05 x 1Q-9
1.93 x ID"7
1.04 x ID"7
4.74 x 1Q-13
Group
method 25
-
-
2.94 x ID"7
4.55 x ID"8
-
-
-
2.45 x ID"10
-
8.88 x ID"8
-
-
Measured
-
-
1.00 x ID"7
3.81 x ID"8
-
-
-
-
-
9.26 x ID"9
-
-
                                    This document is a draft for review purposes only and does not constitute Agency policy.
                    June 2015                                          C-68                     DRAFT—DO NOT CITE OR QUOTE

-------
Hydraulic Fracturing Drinking Water Assessment
Appendix C
Chemical name
Sodium diacetate
Sodium dichloroisocyanurate
Sodium dl-lactate
Sodium dodecyl sulfate
Sodium erythorbate (1:1)
Sodium ethasulfate
Sodium formate
Sodium hydroxymethanesulfonate
Sodium l-lactate
Sodium maleate (l:x)
Sodium N-methyl-N-oleoyltaurate
Sodium octyl sulfate
CASRN
126-96-5
2893-78-9
72-17-3
151-21-3
6381-77-7
126-92-1
141-53-7
870-72-4
867-56-1
18016-19-8
137-20-2
142-31-4
Log Kow
Estimated
0.09
1.28
-0.65
2.42
-1.88
0.38
-0.46
-3.85
-0.65
0.05
4.43
0.46
Measured
-0.17
-
-0.72
-
-1.85
-
-0.54
-
-0.72
-0.48
-
-
Water solubility
Estimate from log Kow
(mg/Lat25°C)
4.76 x 105
3,613
1.00 x 10s
163.7
1.00 x 10s
1.82 x 104
9.55 x 105
1.00 x 10s
1.00 x 10s
1.04 x 105
0.4748
1.58 x 104
Henry's law constant
(atm-m3/mol at 25°C)
Bond
method
5.48 x ID"7
3.22 x ID"12
1.13 x ID"7
1.84 x ID"7
4.07 x 1Q-8
5.91 x 1Q-8
7.50 x ID"7
4.60 x 1Q-13
1.13 x ID"7
1.35 x ID"12
1.00 x ID"12
5.91 x 1Q-8
Group
method 25
2.94 x ID"7
-
-
-
-
-
S.llxlQ-7
-
-
8.48 x ID"14
-
-
Measured
1.00 x ID"7
-
S.lSxlQ-8
-
-
-
1.67 x ID"7
-
8.13xlQ-8
-
-
-
                                    This document is a draft for review purposes only and does not constitute Agency policy.
                    June 2015                                          C-69                     DRAFT—DO NOT CITE OR QUOTE

-------
Hydraulic Fracturing Drinking Water Assessment
Appendix C
Chemical name
Sodium salicylate
Sodium sesquicarbonate
Sodium thiocyanate
Sodium trichloroacetate
Sodium xylenesulfonate
Sorbic acid
Sorbitan sesquioleate
Sorbitan, mono-(9Z)-9-octadecenoate
Sorbitan, monooctadecanoate
Sorbitan, tri-(9Z)-9-octadecenoate
Styrene
Sucrose
CASRN
54-21-7
533-96-0
540-72-7
650-51-1
1300-72-7
110-44-1
8007-43-0
1338-43-8
1338-41-6
26266-58-0
100-42-5
57-50-1
Log Kow
Estimated
2.24
-0.46
0.58
1.44
-0.07
1.62
14.32
5.89
6.1
22.56
2.89
-4.27
Measured
2.26
-
-
1.33
-
1.33
-
-
-
-
2.95
-3.7
Water solubility
Estimate from log Kow
(mg/Lat25°C)
3,808
8.42 x 105
4.36 x 104
1.20 x 104
5.89 x 104
1.94 x 104
2.31 x 1Q-11
0.01914
0.01218
1.12 x 1Q-19
343.7
1.00 x 10s
Henry's law constant
(atm-m3/mol at 25°C)
Bond
method
1.42 x ID"8
6.05 x 1Q-9
1.46 x ID"4
2.39 x 1Q-8
3.06 x 1Q-9
5.72 x ID"7
7.55 x ID"12
1.42 x ID"12
1.61 x ID"12
4.02 x 1Q-11
2.76 x 1Q-3
4.47 x ID"22
Group
method 25
5.60 x ID"12
-
-
-
-
4.99 x ID"8
1.25 x ID"16
5.87 x ID"20
2.23 x 1Q-19
2.68 x 1Q-13
2.81 x 1Q-3
-
Measured
7.34 x ID"9
-
-
1.35 x ID"8
-
-
-
-
-
-
2.75 x ID"3
-
                                    This document is a draft for review purposes only and does not constitute Agency policy.
                    June 2015                                          C-70                     DRAFT—DO NOT CITE OR QUOTE

-------
Hydraulic Fracturing Drinking Water Assessment
Appendix C
Chemical name
Sulfan blue
Sulfuric acid, mono-C12-18-alkyl esters,
sodium salts
Sulfuric acid, mono-C6-10-alkyl esters,
ammonium salts
Symclosene
tert-Butyl hydroperoxide
tert-Butyl perbenzoate
Tetradecane
Tetradecyldimethylbenzylammonium
chloride
Tetraethylene glycol
Tetraethylenepentamine
CASRN
129-17-9
68955-19-1
68187-17-7
87-90-1
75-91-2
614-45-9
629-59-4
139-08-2
112-60-7
112-57-2
Log Kow
Estimated
-1.34
3.9
0.46
0.94
0.94
2.89
7.22
3.91
-2.02
-3.16
Measured
-
-
-
-
-
-
7.2
-
-
-
Water solubility
Estimate from log Kow
(mg/Lat25°C)
50.67
5.165
1.58 x 104
4,610
1.97 x 104
159.2
0.009192
3.608
1.00 x 10s
1.00 x 10s
Henry's law constant
(atm-m3/mol at 25°C)
Bond
method
1.31 x ID"26
4.29 x ID"7
5.91 x 1Q-8
6.19 x 1Q-11
1.60 x 1Q-5
2.06 x ID"4
1.65 x 101
1.34 x 1Q-11
4.91 x 1Q-13
2.79 x ID"20
Group
method 25
-
-
-
-
-
-
2.68 xlO1
-
5.48 x ID"19
4.15 x ID"23
Measured
-
-
-
-
-
-
9.20
-
-
-
                                    This document is a draft for review purposes only and does not constitute Agency policy.
                    June 2015                                          C-71                     DRAFT—DO NOT CITE OR QUOTE

-------
Hydraulic Fracturing Drinking Water Assessment
Appendix C
Chemical name
Tetrakis(hydroxymethyl)phosphonium
sulfate
Tetramethylammonium chloride
Thiamine hydrochloride
Thiocyanic acid, ammonium salt
Thioglycolic acid
Thiourea
Toluene
Tributyl phosphate
Tributyltetradecylphosphonium chloride
Tridecane
Triethanolamine
CASRN
55566-30-8
75-57-0
67-03-8
1762-95-4
68-11-1
62-56-6
108-88-3
126-73-8
81741-28-8
629-50-5
102-71-6
Log Kow
Estimated
-5.03
-4.18
0.95
0.58
0.03
-1.31
2.54
3.82
11.22
6.73
-2.48
Measured
-
-
-
-
0.09
-1.08
2.73
4
-
-
-1
Water solubility
Estimate from log Kow
(mg/Lat25°C)
1.00 x 10s
1.00 x 10s
3,018
4.36 x 104
2.56 x 105
5.54 x 105
573.1
7.355
7.90 x ID"7
0.02746
1.00 x 10s
Henry's law constant
(atm-m3/mol at 25°C)
Bond
method
9.17xlO-13
4.17 x ID"12
8.24 x 1Q-17
1.46 x ID"4
1.94 x 1Q-8
1.58 x ID"7
5.95 x 1Q-3
3.19 x 1Q-6
2.61 x ID"1
1.24 x 101
4.18xlO-12
Group
method 25
-
-
-
-
-
-
5.73 x ID"3
-
-
1.90 x 101
3.38 x ID"19
Measured
-
-
-
-
-
1.98 x ID"9
6.64 x ID"3
1.41 x 1Q-6
-
2.88
7.05 x ID"13
                                    This document is a draft for review purposes only and does not constitute Agency policy.
                    June 2015                                          C-72                     DRAFT—DO NOT CITE OR QUOTE

-------
Hydraulic Fracturing Drinking Water Assessment
Appendix C
Chemical name
Triethanolamine hydrochloride
Triethanolamine hydroxyacetate
Triethyl citrate
Triethyl phosphate
Triethylene glycol
Triethylenetetramine
Triisopropanolamine
Trimethanolamine
Trimethylamine
Tripotassium citrate monohydrate
Tripropylene glycol monomethyl ether
Trisodium citrate
CASRN
637-39-8
68299-02-5
77-93-0
78-40-0
112-27-6
112-24-3
122-20-3
14002-32-5
75-50-3
6100-05-6
25498-49-1
68-04-2
Log Kow
Estimated
-2.48
-2.97
0.33
0.87
-1.75
-2.65
-1.22
-3.95
0.04
-1.67
-0.2
-1.67
Measured
-1
-
-
0.8
-1.75
-
-
-
0.16
-1.64
-
-1.64
Water solubility
Estimate from log Kow
(mg/Lat25°C)
1.00 x 10s
1.00 x 10s
2.82 x 104
1.12xl04
1.00 x 10s
1.00 x 10s
1.00 x 10s
1.00 x 10s
1.00 x 10s
1.00 x 10s
1.96 x 105
1.00 x 10s
Henry's law constant
(atm-m3/mol at 25°C)
Bond
method
4.18xlO-12
6.28 x 1Q-11
6.39 x ID"10
5.83 x ID"7
3.16 x 1Q-11
9.30 x 1Q-17
9.77 x ID"12
1.42 x 1Q-8
3.65 x 1Q-5
8.33 x 1Q-18
2.36 x 1Q-11
8.33 x 1Q-18
Group
method 25
3.38 x ID"19
-
-
-
2.56 x ID"16
6.74 x ID"19
4.35 x 1Q-18
-
1.28 x ID"4
-
4.55 x ID"13
-
Measured
7.05 x ID"13
-
3.84 x ID"9
3.60 x ID"8
-
-
-
-
1.04 x ID"4
4.33 x ID"14
-
4.33 x ID"14
                                    This document is a draft for review purposes only and does not constitute Agency policy.
                    June 2015                                          C-73                     DRAFT—DO NOT CITE OR QUOTE

-------
Hydraulic Fracturing Drinking Water Assessment
Appendix C
Chemical name
Trisodium citrate dihydrate
Trisodium ethylenediaminetetraacetate
Trisodium ethylenediaminetriacetate
Tromethamine
Undecane
Urea
Xylenes
CASRN
6132-04-3
150-38-9
19019-43-3
77-86-1
1120-21-4
57-13-6
1330-20-7
Log Kow
Estimated
-1.67
-3.86
-4.32
-1.56
5.74
-1.56
3.09
Measured
-1.64
-
-
-
-
-2.11
3.2
Water solubility
Estimate from log Kow
(mg/Lat25°C)
1.00 x 10s
2.28 x 105
1.00 x 10s
1.00 x 10s
0.2571
4.26 x 105
207.2
Henry's law constant
(atm-m3/mol at 25°C)
Bond
method
8.33 x ID"18
1.17 x 1Q-23
3.58 x ID"20
8.67 x 1Q-13
7.04
3.65 x ID"10
6.56 x 1Q-3
Group
method 25
-
-
-
-
9.52
-
6.14xlQ-3
Measured
4.33 x ID"14
5.77 x 1Q-16
-
-
1.93
1.74 x ID"12
7.18xlQ-3
 "--" indicates no information available.
                                     This document is a draft for review purposes only and does not constitute Agency policy.
                     June 2015                                          C-74                      DRAFT—DO NOT CITE OR QUOTE

-------
      Hydraulic Fracturing Drinking Water Assessment                                             Appendix C


 1    The EPI (Estimation Programs Interface) Suite™ (USJiPJLZOlla) is an open-source, Windows®-
 2    based suite of physicochemical property and environmental fate estimation programs developed by
 3    the EPA's Office of Pollution Prevention Toxics and Syracuse Research Corporation. More
 4    information on EPI  Suite™ is available at hJlBi66vww^^
 5    Although only physicochemical properties from EPI Suite™ are provided here, other sources of
 6    information were also consulted. QikProp (Sc^inidin^ej^ZJllZ,) and LeadScope® (Inc11_2012.) are
 7    commercial products designed primarily as drug development and screening tools. QikProp is
 8    specifically focused on drug discovery and provides predictions for physically significant
 9    descriptors and pharmaceutically (and toxicologically) relevant properties useful in predicting
10    ADME (adsorption, distribution, metabolism, and excretion) characteristics of drug candidates.
11    QikProp's use of whole-molecule descriptors that have a straightforward physical interpretation (as
1 2    opposed to fragment-based descriptors).

13    LeadScope® is a program designed for interpreting chemical and biological screening data that can
14    assist pharmaceutical scientists in finding promising drug candidates. The software organizes the
15    chemical data by structural features familiar to medicinal chemists. Graphs are used to summarize
16    the data, and structural classes are highlighted that are statistically correlated with biological
17    activity. It incorporates chemically-based data mining, visualization, and advanced informatics
18    techniques (e.g., prediction tools, scaffold generators). Note that properties generated by QikProp
19    and LeadScope® are generally more relevant to drug development than to environmental
20    assessment

21    Physicochemical properties of chemicals were generated from the two-dimensional (2-D) chemical
22    structures  from the EPA National Center for Computational Toxicology's Distributed Structure-
23    Searchable Toxicity (NCCT DSSTox) Database Network in structure- data file  (SDF) format. For EPI
24    Suite™ properties, both the desalted and non-desalted 2-D files were run using the program's batch
25    mode (i.e.,  processing many molecules at once) to calculate environmentally-relevant, chemical
26    property descriptors. The chemical descriptors in QikProp require 3-D chemical structures. For
27    these calculations, the 2-D desalted chemical structures were converted to 3-D using the RebuildSD
28    function in the Molecular Operating Environment software (CCiLlQll). All computed
29    physicochemical properties are added into the structure-data file prior to assigning toxicological
30    properties.

3 1    Both LeadScope® and Qikprop software require input of desalted structures. Therefore, the
32    structures  were desalted, a process where salts and complexes are simplified to the neutral,
33    uncomplexed form of the chemical, using "Desalt Batch" option in ACD Labs ChemFolder. All
34    LeadScope® general chemical descriptors (Parent Molecular Weight, AlogP, Hydrogen Bond
35    Acceptors, Hydrogen Bond Donors, Lipinski Score, Molecular Weight, Parent Atom Count, Polar
36    Surface Area, and Rotatable Bonds) were calculated by default

37    All physicochemical properties generated from EPI Suite™, QikProp, and LeadScope® will be made
38    available to the public in an electronic format in 2015.
                   This document is a draft for review purposes only and does not constitute Agency policy.
      June 2015                                    C-75                  DRAFT—DO NOT CITE OR QUOTE

-------
 Hydraulic Fracturing Drinking Water Assessment                                                 Appendix C
C.2.    References for Appendix C

Afzal. W: Mohammadi. AH: Richon. D. (2009). Volumetric properties of mono-, di-, tri-, and polyethylene
   glycol aqueous solutions from (273.15 to 363.15) K: experimental measurements and correlations. Journal
   of Chemical and Engineering Data 54:1254-1261. http://dx.doi.org/10.1021/je800694a

Alfa Aesar. (2015). A16163: Formaldehyde, 37% w/w aq. soln., stab, with 7-8% methanol. Available online at
   https://www.alfa.com/en/catalog/A16163 (accessed May 4,2015).

Baragi. IG: Maganur. S: Malode. V: Baragi. SI. (2013). Excess molar volumes and refractive indices of binary
   liquid mixtures of acetyl acetone with n-Nonane, n-Decane and n-Dodecane at (298.15, 303.15, and
   308.15) K. Journal of Molecular Liquids 178:175-177. http://dx.doi.Org/10.1016/i.molliq.2012.ll.022

Bennett, GM: Yuill, IL. (1935). The crystal form of anhydrous citric acid. J Chem Soc 1935:130.
   http://dx.doi.org/10.1039/IR9350000130

Biilmann. E. (1906). [Studien iiber organische Thiosauren III]. Justus Liebigs Annalen der Chemie 348:133-
   143. http://dx.doi.org/10.1002/ilac.19063480110

Biltz. W: Balz. G. (1928). [Uber molekular- und  atomvolumina. XVIII. Das volumen des ammoniaks in
   kristallisierten ammoniumsalzen]. Zeitschrift fur Anorganische und Allgemeine Chemie 170: 327-341.
   http://dx.doi.org/10.1002/zaac.19281700141

Blanco, A: Garcia-Abuin, A: Gomez-Diaz, D: Navaza, IM: Villaverde, OL. (2013). Density, speed of sound,
   viscosity, surface tension, and excess volume of n-ethyl-2-pyrrolidone plus ethanolamine (or
   diethanolamine or triethanolamine) from T = (293.15 to 323.15) K. Journal of Chemical and Engineering
   Data 58: 653-659. http://dx.doi.org/10.1021/je301123j

Carpenter. EL: Davis. HS. (1957). Acrylamide. Its preparation and properties. Journal of Applied Chemistry 7:
   671-676. http://dx.doi.org/10.1002/ictb.5010071206

Casanova. C: Wilhelm. E: Grolier. IPE: Kehiaian. HV. (1981). Excess volumes and excess heat-capacities of
   (water + alkanoic acid). The Journal of Chemical Thermodynamics 13: 241-248.
   http://dx.doi.org/10.1016/0021-9614r81)90123-3

CCG (Chemical Computing Group). (2011). Molecular Operating Environment (MOE) Linux (Version 2011.10)
   [Computer Program]. Montreal,  Quebec. Retrieved from http://www.chemcomp.com/software.htm

Chafer. A: Lladosa. E: Monton. IB: Cruz Burguet. M. a. (2010). Liquid-liquid equilibria for the system 1-methyl
   propyl ethanoate (1) + acetic acid (2) + water (3) at (283.15 and 323.15) K. Journal of Chemical and
   Engineering Data 55: 523-525. http://dx.doi.org/10.1021/je900332x

Chasib. KF. (2013). Extraction of phenolic pollutants (phenol and p-chlorophenol) from industrial
   wastewater. Journal of Chemical and Engineering Data 58:1549-1564.
   http://dx.doi.org/10.1021/je4001284

de Oliveira. LH: da Silva. IL. Ir: Aznar. M. (2011). Apparent and partial molar volumes at infinite dilution and
   solid-liquid equilibria of dibenzothiophene plus alkane systems. Journal of Chemical and Engineering Data
   56: 3955-3962. http://dx.doi.org/10.1021/ie200327s

Dejoye Tanzi, C: Abert Vian, M: Ginies, C: Elmaataoui, M: Chemat, F. (2012). Terpenes as green solvents for
   extraction of oil from microalgae. Molecules 17: 8196-8205.
   http://dx.doi.org/10.3390/moleculesl7078196

Dhondge. SS: Pandhurnekar. CP: Parwate. DV. (2010). Density, speed of sound, and refractive index of
   aqueous binary mixtures of some glycol ethers atT=298.15 K. Journal of Chemical and Engineering Data
   55: 3962-3968. http://dx.doi.org/10.1021/ie901072c

Dubey. GP: Kumar. K. (2011). Thermodynamic properties of binary liquid mixtures of diethylenetriamine
   with alcohols at different temperatures. Thermochim Acta 524: 7-17.
   http://dx.doi.0rg/10.1016/i.tca.2011.06.003
               This document is a draft for review purposes only and does not constitute Agency policy.

June 2015                                       C-76                    DRAFT—DO NOT CITE OR QUOTE

-------
 Hydraulic Fracturing Drinking Water Assessment                                                 Appendix C
Dubey. GP: Kumar. K. (2013). Studies of thermodynamic, thermophysical and partial molar properties of
   liquid mixtures of diethylenetriamine with alcohols at293.15 to 313.15 K. Journal of Molecular Liquids
   180:164-171. http://dx.doi.Org/10.1016/i.molliq.2013.01.011

Dyshin. AA: Eliseeva. 0V: Kiselev. MG: Al'per. GA. (2008). The volume characteristics of solution of
   naphthalene in heptane-ethanol mixtures at 298.15 K. Russian Journal of Physical Chemistry A, Focus on
   Chemistry 82: 1258-1261. http://dx.doi.org/10.1134/S0036024408080037

Egorov, GI: Makarov, DM: Kolker, AM. (2013). Volume properties of liquid mixture of water plus glycerol over
   the temperature range from 278.15 to 348.15 K at atmospheric pressure. ThermochimActa570: 16-26.
   http://dx.doi.0rg/10.1016/i.tca.2013.07.012

Fadeeva. YA: Shmukler. LE: Safonova. LP. (2004). Physicochemical properties of the H3P04-
   dimethylformamide system. Russian Journal of General Chemistry 74: 174-178.
   http://dx.doi.Org/10.1023/B:RUGC.0000025496.07304.66

Faria. MAP: Martins. Rl: Cardoso. MIE. M: Barcia. OE. (2013). Density and viscosity of the binary systems
   ethanol + butan-1-ol, + pentan-1-ol, + heptan-1-ol, + octan-1-ol, nonan-1-ol, + decan-1-ol at 0.1 mpa and
   temperatures from 283.15 Kto 313.15 K. Journal of Chemical and Engineering Data 58: 3405-3419.
   http://dx.doi.org/10.1021/je400630f

Pels. G. (1900). Ueber die Frage der isomorphen vertretung von halogen und hydroxyl. In Zeitschrift fur
   Kristallographie, Kristallgeometrie, Kristallphysik, Kristallchemie. Frankfurt: Leipzig.
   http://babel.hathitrust.org/cgi/pt?id=ucl.b3327977:view=lup:seq=5

finemech (finemech Precision Mechanical Components). (2012). Technical resources: Liquid nitrogen, LN2.
   Available online at http://www.finemech.com/tech resources/liquid nitrogen.html

Fuess, H: Bats, IW: Dannohl, H: Meyer, H: Schweig, A. (1982). Comparison of observed and calculated
   densities. XII. Deformation density in complex anions. II. Experimental and theoretical densities in sodium
   formate. Acta Crystallogr B  B38: 736-743. http://dx.doi.org/10.1107/S0567740882003999

Fujino. S: Hwang. C: Morinaga.  K. (2004). Density, surface tension, and viscosity of PbO-B203-Si02 glass
   melts. Journal of the American Ceramic Society 87:10-16. http://dx.doi.Org/10.llll/j.1151-
   2916.2004.tbl9937.x

Hagen. R: Kaatze. U. (2004). Conformational kinetics of disaccharides in aqueous solutions. J Chem Phys 120:
   9656-9664. http://dx.doi.Org/10.1063/l.1701835

Harlow. A: Wiegand. G: Franck. EU. (1997). The Density of Ammonia at High Pressures to 723 K and 950 MPa.
   101: 1461-1465. http://dx.doi.org/10.1002/bbpc.199700007

Haynes. WM. (2014). CRC handbook of chemistry and physics. In WM Haynes (Ed.), (95 ed.). Boca Raton, FL:
   CRC Press, http://www.hbcpnetbase.com/

He, YM: liang, RF: Zhu,  F: Luan, TG: Huang, ZQ: Ouyang, GF. (2008). Excess molar volumes and surface
   tensions of 1,2,4-trimethylbenzene and 1,3,5-trimethylbenzene with isopropyl acetate and isobutyl
   acetate at (298.15, 308.15, and 313.15)K. Journal of Chemical and Engineering Data 53:1186-1191.
   http://dx.doi.org/10.1021/je800046k

Huffman. HM: Fox. SW. (1938). Thermal data. X. The heats of combustion and free energies, at 25, of some
   organic compounds concerned in carbohydrate metabolism. J Am Chem Soc 60:1400-1403.
   http://dx.doi.org/10.1021/ja01273a036

Inc.. L. (2012). Leadscope [Computer Program]. Columbus, Ohio. Retrieved from http://www.leadscope.com

IUPAC (International Union of Pure and Applied Chemistry). (2014). Global availability of information on
   agrochemicals: Triisopropanolamine. Available online at
   http://sitem.herts.ac.uk/aeru/iupac/Reports/1338.htm
               This document is a draft for review purposes only and does not constitute Agency policy.

June 2015                                       C-77                    DRAFT—DO NOT CITE OR QUOTE

-------
 Hydraulic Fracturing Drinking Water Assessment                                                 Appendix C
Kiselev. VD: Kashaeva. HA: Shakirova. II: Potapova. LN: Konovalov. AI. (2012). Solvent effect on the enthalpy
   of solution and partial molar volume of the ionic liquid l-butyl-3-methylimidazolium tetrafluoroborate.
   Journal of Solution Chemistry 41:1375-1387. http://dx.doi.org/10.1007/sl0953-012-9881-9

Krakowiak. I: Bobicz. D: Grzybkowski. W. (2001). Limiting partial molar volumes of tetra-n-alkylammonium
   perchlorates in N,N-dimethylacetamide, triethylphosphate and dimethyl sulfoxide atT=298.15 K. The
   Journal of Chemical Thermodynamics 33: 121-133. http://dx.doi.org/10.1006/icht.2000.0725

Laavi, H: Pokki, IP: Uusi-Kyyny, P: Massimi, A: Kim, Y: Sapei, E: Alopaeus, V. (2013). Vapor-liquid equilibrium
   at 350 k, excess molar enthalpies at 298 K, and excess molar volumes at 298 K of binary mixtures
   containing ethyl acetate, butyl acetate, and 2-butanol. Journal of Chemical and Engineering Data 58: 1011-
   1019. http://dx.doi.org/10.1021/ie400036b

Laavi. H: Zaitseva. A: Pokki. IP: Uusi-Kyyny. P: Kim. Y: Alopaeus. V. (2012). Vapor-liquid equilibrium, excess
   molar enthalpies, and excess molar volumes of binary mixtures containing methyl isobutyl ketone (MIBK)
   and 2-butanol, tert-pentanol, or 2-ethyl-l-hexanol. Journal of Chemical and Engineering Data  57: 3092-
   3101. http://dx.doi.org/10.1021/ie300678r

Mak, TCW. (1965). Hexamethylenetetramine hexahydrate: A new type of clathrate hydrate. J Chem Phys 43:
   2799. http://dx.doi.0rg/10.1063/l.1697212

Martinez-Reina. M: Amado-Gonzalez. E: Mauricio Munoz-Munoz. Y. (2012). Study of liquid-liquid equilibria of
   toluene plus (hexane, heptane, or cyclohexane) with l-ethyl-3-methylimidazolium ethylsulfate at 308.15
   K. Bull Chem Soc Jpn 85: 1138-1144. http://dx.doi.org/10.1246/bcsj.20120112

Masood, ARM: Pethrick, RA: Swinton, FL. (1976). Physicochemical studies of super-cooled liquids - cyclic
   carbonates and alpha.beta-unsaturated aldehydes. Faraday Trans 1 72: 20-28.
   http://dx.doi.org/10.1039/fl9767200020

Moosavi. M: Motahari. A: Omrani. A: Rostami. AA. (2013). Thermodynamic study on some alkanediol
   solutions: Measurement and modeling. Thermochim Acta 561:1-13.
   http://dx.doi.0rg/10.1016/i.tca.2013.03.010

Oka. S. (1962). Studies on lactone formation in vapor phase. III. Mechanism of lactone formation from diols.
   Bull Chem Soc Jpn 35: 986-989. http://dx.doi.org/10.1246/bcsi.35.986

Pal. A: Kumar. H: Maan. R: Sharma. HK. (2013). Densities and speeds of sound of binary liquid mixtures of
   some n-alkoxypropanols with methyl acetate, ethyl acetate, and n-butyl acetate at T = (288.15, 293.15,
   298.15, 303.15, and 308.15)  K. Journal of Chemical and Engineering Data 58: 225-239.
   http://dx.doi.org/10.1021/je300789a

Pijper. WP. (1971). Molecular and crystal structure of glycollic acid. Acta Crystallogr B B27: 344-348.
   http://dx.doi.org/10.1107/S056774087100219X

Radwan. MHS: Hanna. AA. (1976). Binary azeotropes containing butyric acids. Journal of Chemical and
   Engineering Data 21: 285-289. http://dx.doi.org/10.1021/je60070a032

Rani. M: Maken. S. (2013). Excess molar enthalpies and excess molar volumes of formamide+1-propanol or 2-
   propanol and thermodynamic modeling by Prigogine-Flory-Patterson theory and Treszczanowicz-Benson
   association model. Thermochim Acta 559: 98-106. http://dx.doi.Org/10.1016/j.tca.2013.02.010

Rawat. BS: Gulati. IB: Mallik. KL. (1976). Study of some sulphur-group solvents for aromatics extraction by gas
   chromatography. Journal of Applied Chemistry and Biotechnology 26: 247-252.
   http://dx.doi.org/10.1002/ictb.5020260504

Rodnikova, MN: Solonina, IA: Egorov, GI: Makarov, DM: Gunina, MA. (2012). The bulk properties of dioxane
   solutions in ethylene glycol at 2575C. Russian Journal of Physical Chemistry A, Focus on Chemistry 86:
   330-332. http://dx.doi.org/10.1134/S0036024412020239
               This document is a draft for review purposes only and does not constitute Agency policy.

June 2015                                       C-78                    DRAFT—DO NOT CITE OR QUOTE

-------
 Hydraulic Fracturing Drinking Water Assessment                                                  Appendix C
Sarkar. BK: Choudhury. A: Sinha. B. (2012). Excess molar volumes, excess viscosities and ultrasonic speeds of
   sound of binary mixtures of 1,2-dimethoxyethane with some aromatic liquids at 298.15 K. Journal of
   Solution Chemistry 41: 53-74. http://dx.doi.org/10.1007/sl0953-011-9780-5

Sarkar. L: Roy. MN. (2009). Density, viscosity, refractive index, and ultrasonic speed of binary mixtures of 1,3-
   dioxolane with 2-methoxyethanol, 2-ethoxyethanol, 2-butoxyethanol, 2-propylamine, and
   cyclohexylamine. Journal of Chemical and Engineering Data 54: 3307-3312.
   http://dx.doi.org/10.1021/ie900240s

Schrodinger. (2012). Qikprop [Computer Program]. New York, New York: Schrodinger, LLC. Retrieved from
   http://www.schrodinger.com/products/14/17

Shanley. P: Collin. RL. (1961). The crystal structure of the high temperature form of choline chloride. Acta
   Cryst 14: 79-80. http://dx.doi.org/10.1107/S0365110X61000292

Sigma-Aldrich. (2007). Material safety data sheet: Tert-butyl hydroperoxide (70% solution in water).
   Available online at http://www.orcbs.msu.edu/msds/111607 DLI 027 TERT-BUTYL.PDF

Sigma-Aldrich. (2010). Product information: Sodium chloride. Available online at
   https:// www.sigmaaldrich.com/content/dam/sigma-aldrich/docs/Sigma-
   Aldrich/Product Information Sheet/s7653pis.pdf

Sigma-Aldrich. (2014a). Material safety data sheet: Phosphorus acid. Available online at
   http://www.sigmaaldrich.eom/catalog/product/sial/215112?lang=en®ion=US

Sigma-Aldrich. (2014b). Material safety data sheet: Potassium carbonate. Available online at
   http://www.sigmaaldrich.com/catalog/product/aldrich/367877?lang=en®ion=US

Sigma-Aldrich. (2015a). Material safety data sheet: Aluminum chloride [Fact Sheet]. St. Louis, MO.
   http://www.sigmaaldrich.com/catalog/product/aldrich/563919?lang=en®ion=US

Sigma-Aldrich. (2015b). Material safety data sheet: Peracetic acid solution. Available online at
   http://www.sigmaaldrich.eom/catalog/product/sial/2 69336?lang=en®ion=US

Sigma-Aldrich. (2015c). Material safety data sheet: Sulfur dioxide. Available online at
   http://www.sigmaaldrich.com/catalog/product/aldrich/295698?lang=en®ion=US

Sigma-Aldrich. (2015d). Material safety data sheet: Sulfuric acid. Available online at
   http://www.sigmaaldrich.com/catalog/product/aldrich/339741?lang=en®ion=US

Sigma-Aldrich. (2015e). Material safety data sheet: Trimethyl borate. Available online at
   http://www.sigmaaldrich.com/catalog/product/aldrich/447218?lang=en®ion=US

Smirnov. VI: Badelin. VG. (2013). Enthalpy characteristics of dissolution of L-tryptophan in water plus
   formamides binary solvents at 298.15 K. Russian Journal of Physical Chemistry A, Focus on Chemistry 87:
   1165-1169. http://dx.doi.org/10.1134/S0036024413070285

Steinhauser. 0: Boresch. S: Bertagnolli. H. (1990). The effect of density variation on the structure of liquid
   hydrogen chloride. A Monte Carlo study. J Chem Phys 93: 2357-2363.
   http://dx.doi.0rg/10.1063/l.459015

Thalladi. VR: Nusse. M: Boese. R. (2000). The melting point alternation in alpha.omega-alkanedicarboxylic
   acids. J Am Chem Soc 122: 9227-9236. http://dx.doi.org/10.1021/ja0011459

U.S. EPA (U.S. Environmental Protection Agency). (2012a). Estimation Programs Interface Suite for Microsoft
   Windows (EPI Suite) [Computer Program]. Washington DC:  US Environmental Protection Agency.
   Retrieved from http://www.epa.gov/oppt/exposure/pubs/episuitedl.htm

U.S. EPA (U.S. Environmental Protection Agency). (2015c). Analysis of hydraulic fracturing fluid data from the
   FracFocus chemical disclosure registry 1.0: Project database [EPA Report]. (EPA/601/R-14/003).
   Washington, D.C.: U.S. Environmental Protection Agency, Office of Research and Development.
   http://www2.epa.gov/hfstudv/epa-proiect-database-developed-fracfocus-l-disclosures
               This document is a draft for review purposes only and does not constitute Agency policy.

June 2015                                       C-79                   DRAFT—DO NOT CITE OR QUOTE

-------
 Hydraulic Fracturing Drinking Water Assessment                                                  Appendix C
Vijaya Kumar. R: Anand Rao. M: Venkateshwara Rao. M: Ravi Kumar. YVL: Prasad. DHL. (1996). Bubble
   temperature measurements on 2-propyn-l-ol with 1,2-dichloroethane, 1,1,1-trichloroethane, and 1,1,2,2-
   tetrachloroethane. Journal of Chemical and Engineering Data 41:1020-1023.
   http://dx.doi.org/10.1021/je9600156

Wilt. IW. (1956). Notes - the halodecarboxylation of cyanoacetic acid. J Org Chem 21: 920-921.
   http://dx.doi.org/10.1021/io01114a607

Xiao, LN: Xu, IN: Hu, YY: Wang, LM: Wang, Y: Ding, H: Cui, XB: Xu, 10. (2013). Synthesis and characterizations
   of the first [V16039C1J6- (V16039) polyanion. Dalton Transactions (Online) 42: 5247-5251.
   http://dx.doi.org/10.1039/c3dt33081h

Zhang. L: Guo. Y: Xiao. I: Gong. X: Fang. W. (2011).  Density, refractive index, viscosity, and surface tension of
   binary mixtures of exo-tetrahydrodicyclopentadiene with some n-alkanes from (293.15 to 313.15) K.
   Journal of Chemical and Engineering Data 56: 4268-4273. http://dx.doi.org/10.1021/je200757a

Zhang. Z: Yang. L: Xing. Y: Li. W. (2013). Vapor-liquid equilibrium for ternary and binary mixtures of 2-
   isopropoxypropane, 2-propanol, and n,n-dimethylacetamide at 101.3 kPa. Journal of Chemical and
   Engineering Data 58: 357-363. http://dx.doi.org/10.1021/je300994y
               This document is a draft for review purposes only and does not constitute Agency policy.

June 2015                                       C-80                    DRAFT—DO NOT CITE OR QUOTE

-------
 Hydraulic Fracturing Drinking Water Assessment                                Appendix D
                        Appendix D
Designing, Constructing, and Testing Wells for
Integrity
         This document is a draft for review purposes only and does not constitute Agency policy.

June 2015                                        DRAFT—DO NOT CITE OR QUOTE

-------
      Hydraulic Fracturing Drinking Water Assessment                                           Appendix D


     Appendix D.  Designing,  Constructing, and Testing Wells

         for Integrity
 1   This appendix presents the goals for the design and construction of oil and gas production wells,
 2   the well components used to achieve those goals, and methods for testing well integrity to help
 3   verify that the goals for well performance are achieved. This information provides additional
 4   background for the well component discussions presented in Chapter 6. Information on the
 5   pathways associated with the well that can cause fluid movement into drinking water resources is
 6   presented in Chapter 6.

     D.I.  Design Goals for Well Construction
 7   Simply stated, production wells are designed to move oil and gas from the production zone (within
 8   the oil and gas reservoir) into the well and then through the well to the surface. There are typically
 9   a variety of goals for well design (RenpMJLQJLl), but the main purposes are facilitating the flow of
10   oil and gas from the hydrocarbon reservoirs to the well (production management) while isolating
11   that oil and gas and the hydrocarbon reservoirs from nearby ground water resources (zonal
12   isolation).

13   To achieve these goals, operators design and construct wells to have and maintain mechanical
14   integrity throughout the life of the well. A properly designed and constructed well has two types of
15   mechanical integrity: internal and external. Internal mechanical integrity refers to the absence of
16   significant leakage within the production tubing, casing, or packer. External mechanical integrity
17   refers to the absence of significant leakage along the well outside of the casing.

18   Achieving mechanical integrity involves designing the well components to resist the stresses they
19   will encounter. Each well component must be designed to withstand all of the stresses to which the
20   well will be subjected,  including burst pressure, collapse, tensile, compression (or bending), and
21   cyclical stresses (see Section 6.2.1 for additional information on these stresses). Well materials
22   should also be compatible with the fluids (including liquids or gases) with which they come into
23   contact to prevent leaks caused by corrosion.

24   These goals are accomplished by the use of one or more layers of casing, cement, and mechanical
25   devices (such as packers), which provide the main barrier preventing migration of fluids from the
26   well into drinking water sources.

     D.2.  Well Components
27   Casing and cement are used in the design and construction of wells to achieve the goals of
28   mechanical integrity and zonal isolation. Several industry-developed specifications and best
29   practices for well construction have been established to guide well operators in the construction
30   process; see Text Box D-l. (Information is not available to determine how often these practices are
31   used or how well they  prevent the development of pathways for fluid movement to drinking water
32   resources.) The sections below describe options available for casing, cement, and other well
33   components.


                  This document is a draft for review purposes only and does not constitute Agency policy.
     June 2015                                    D-l                  DRAFT—DO NOT CITE OR QUOTE

-------
       Hydraulic Fracturing Drinking Water Assessment                                              Appendix D
      Text Box D-l. Selected Industry-Developed Specifications and Recommended
      Practices for Well Construction in North America.

 1    American Petroleum Institute (API)

 2    •   API Guidance Document HF1—Hydraulic Fracturing Operations—Well Construction and Integrity
 3       Guidelines [AP!.2009a]
 4    •   API RP 10B-2-Recommended Practice for Testing Well Cements [API, 2013]
 5    •   API RP 10D-2-Recommended Practice for Centralizer Placement and Stop Collar Testing [API. 2004]
 6    •   API RP 5Cl-Recommended Practices for Care and Use of Casing and Tubing [API, 19991
 7    •   API RP 65-2-Isolating Potential Flow Zones during Well Construction [API, 2010a]
 8    •   API Specification 10A—Specification on Cements and Materials for Well Cementing [API, 2010b]
 9    •   API Specification HDl-Packers and Bridge Plugs [API. 2009b1
10    •   API Specification 5CT-Specification for Casing and Tubing [API, 20111

11    Canadian Association of Petroleum Producers (CAPP) and Enform

12    •   Hydraulic Fracturing Operating Practices: Wellbore Construction and Quality Assurance [CAPP, 20131
13    •   Interim Industry Recommended Practice Volume #24—Fracture Stimulation: Inter-wellbore
14       Communication [Enform, 2013]

15    Marcellus Shale Coalition (MSC)

16    •   Recommended Practices—Drilling and Completions [MSC. 2013]


      D.2.1.  Casing
17    Casing is steel pipe that is placed into the wellbore (the cylindrical hole drilled through the
18    subsurface rock formation) to maintain the stability of the wellbore, to transport the hydrocarbons
19    from the subsurface to the surface, and to prevent intrusion of other fluids into the well and
2 0    wellbore. Up to four types of casing may be present in a well, including (from largest to smallest-
21    diameter): conductor casing, surface casing, intermediate casing, and production casing. Each is
22    described below.

2 3    The conductor casing is the largest diameter string of casing. It is typically in the range of 3 0 in.
24    (76 cm) to 42 in. (107 cm) in diameter (Hyne. 2012). Its main purpose is to prevent unconsolidated
25    material, such as sand, gravel, and soil, from collapsing into the wellbore. Therefore, the casing is
26    typically installed from the surface to the top of the bedrock or other consolidated formations. The
27    conductor casing may or may not be cemented in place.

28    The next string of casing is the surface casing. Atypical surface casing diameter is 13.75 in. (34.93
29    cm), but diameter can vary (Hyne. 2012). The surface casing's main purposes are to isolate any
30    ground water resources that are to be protected by preventing fluid migration along the wellbore


                   This document is a draft for review purposes only and does not constitute Agency policy.
      June 2015                                     D-2                   DRAFT—DO NOT CITE OR QUOTE

-------
      Hydraulic Fracturing Drinking Water Assessment                                            Appendix D


 1    once the casing is cemented and to provide a sturdy structure to which blow-out prevention
 2    equipment can be attached. For these reasons, the surface casing most commonly extends from the
 3    surface to some distance beneath the lowermost geologic formation containing ground water
 4    resources to be protected. The specific depth to which the surface casing is set is often governed by
 5    the depth of the ground water resource as defined and identified for protection in state regulations.

 6    Intermediate casing is typically used in wells to control pressure in an intermediate-depth
 7    formation. It may be used to reduce or prevent exposure of weak formations to pressure from the
 8    weight of the drilling fluid or cement or to allow better control of over-pressured formations. The
 9    intermediate casing extends from the surface through the formation of concern. There may be more
10    than one string of concentric intermediate casing present or none at all, depending on the
11    subsurface geology. Intermediate casing may be cemented, especially through over-pressured
12    zones; however, it is not always cemented to the surface. Intermediate casing, when present, is
13    often 8.625  in. (21.908 cm) in diameter but can vary (Hyjie1_2012.).

14    Production casing extends from the surface into the production zone. The main purposes of the
15    production casing are to isolate the hydrocarbon product from fluids in surrounding formations
16    and to transport the product to the surface. It can also be used to inject fracturing fluids, receive
17    flowback during hydraulic fracturing operations (e.g., if tubing or a temporary fracturing string is
18    not present), and prevent other fluids from mixing with and diluting the produced hydrocarbons.
19    The production casing is generally cemented to some point above the production zone. Production
20    casing is often 5.5 in. (14.0 cm) in diameter but can vary (Hyne1_2()12.).

21    Liners are another type of metal tubular (casing-like) well component that can be used to fulfill the
22    same purposes as intermediate and production casing in the production zone. Like casing, they are
2 3    steel pipe, but differ in that they do not extend from the production zone to the surface. Rather, they
24    are connected to the next largest string of casing by a hanger that is attached to the casing. A frac
2 5    sleeve is a specialized type of liner that is used during fracturing. It has plugs that can be opened
26    and closed by dropping balls from the surface (see the discussion of well completions below for
2 7    additional information on the use of frac sleeves).

28    Production tubing is the smallest, innermost steel pipe in the well and is distinguished from casing
29    by not being cemented in place. It is used to transport the hydrocarbons to the surface. Fracturing
30    may be done through the tubing if present, or through the production casing. Because casing cannot
31    be replaced, tubing is often used, especially if the hydrocarbons contain corrosive substances such
32    as hydrogen sulfide or carbon dioxide. Tubing may not be used in high-volume production wells.
33    Typical tubing diameter is between 1.25 in. (3.18 cm) and 4.5 in. (11.4 cm) (HyjieJ_2fil2.).

      D.2.2.  Cement
34    Cement is the main barrier preventing fluid movement along the wellbore outside the casing. It also
3 5    lends mechanical strength to the well and protects the casing from corrosion by naturally occurring
36    formation fluids. Cement is placed in the annulus, which is the space between two adjacent casings
37    or the space between the outermost casing and the rock formation through which the wellbore was
                   This document is a draft for review purposes only and does not constitute Agency policy.
      June 2015                                    D-3                   DRAFT—DO NOT CITE OR QUOTE

-------
      Hydraulic Fracturing Drinking Water Assessment                                             Appendix D


 1    drilled. The sections below describe considerations for selecting cement and additives, as well as
 2    cementing procedures and techniques.

      D.2.2.1.   Considerations for Cementing
 3    The length and location of the casing section to be cemented and the composition of the cement can
 4    vary based on numerous factors, including the presence and locations of weak formations, over- or
 5    under-pressured formations, or formations containing fluids; formation permeability; and
 6    temperature. State requirements for oil and gas production well construction and the relative costs
 7    of well construction options are also factors.

 8    Improper cementing can lead to the formation of channels (small connected voids) in the cement,
 9    which can—if they extend across multiple formations or connect to other existing channels or
10    fractures—present path ways for fluid migration. This section describes some of the considerations
11    and concerns for proper cement placement and techniques and materials that are available to
12    address these concerns. Careful selection of cements (and additives) and design of the cementing
13    job can avoid integrity problems related to cement

14    To select the appropriate cement type, properties, and additives, operators consider the required
15    strength needed to withstand downhole conditions and compatibility with subsurface chemistry, as
16    described below:

17        •   The cement design needs to achieve the strength required under the measured or
18            anticipated downhole conditions. Factors that are taken into account to achieve proper
19            strength can include density, thickening time, the presence of free water, compressive
20            strength, and formation permeability (Kenj3uJ2j311). Commonly, cement properties are
21            varied during the process, with a "weaker" (i.e., less dense) lead cement, followed by a
22            "stronger" (denser) tail cement. The lead cement is designed with a lower density to
23            reduce pressure on the formation and better displace drilling fluid without a large concern
24            for strength. The stronger tail cement provides greater strength for the deeper portions of
25            the well the operator considers as requiring greater strength.
26        •   The compatibility of the cement with the chemistry of formation fluids, hydrocarbons,
2 7            and hydraulic fracturing fluids is important for maintaining well integrity through the life
28            of the well. Most oil and gas wells are constructed using some form of Portland cement.
29            Portland cement is a specific type of cement consisting primarily of calcium silicates with
30            additional iron and aluminum. Industry specifications for recommended cements are
31            determined by the downhole pressure, temperature, and chemical compatibility required.
32    There are a number of considerations in the design and execution of a cement job. Proper
33    centralization of the casing within the wellbore is one of the more important considerations. Others
34    include the potential for lost cement, gas invasion, cement shrinkage, incomplete removal of drilling
35    mud, settling of solids in the wellbore, and water loss into the formation while curing. These
36    concerns, and techniques available to address them, include the following:
                   This document is a draft for review purposes only and does not constitute Agency policy.
      June 2015                                    D-4                  DRAFT—DO NOT CITE OR QUOTE

-------
      Hydraulic Fracturing Drinking Water Assessment                                            Appendix D


 1        •   Improper centralization of the casing within the wellbore can lead to preferential flow
 2            of cement on the side of the casing with the larger space and little to no cement on the side
 3            closest to the formation. If the casing is not centered in the wellbore, cement will flow
 4            unevenly during the cement job, leading to the formation of cement channels. Kirksey
 5            [2013] notes that, if the casing is off-center by just 25%, the cement job is almost always
 6            inadequate. Centralizers are used to keep the casing in the center of the hole and allow an
 7            even cement job. To ensure proper centralization, centralizers are placed at regular
 8            intervals along the casing [API, 2010a]. Centralizer use is especially key in horizontal
 9            wells, as the casing will tend to settle (due to gravity) to the bottom of the wellbore if the
10            casing is not centered [Sabins. 1990). leading to inadequate cement on the lower side.
11        •   Lost cement (sometimes referred to as lost returns) refers to cement that moves out of
12            the wellbore and into the formation instead of filling up the annulus between the casing
13            and the formation. Lost cement can occur in weak formations that fail (fracture) under
14            pressure of the cement or in particularly porous, permeable, or naturally fractured
15            formations. Lost cement can result in lack of adequate cement across a water- or brine-
16            bearing zone. To avoid inadequate placement of cement due to lost cement, records of
17            nearby wells can be examined to determine zones where lost cement returns occur (API.
18            2009a). If records from nearby wells are not available, cores and logs may be used to
19            identify any high-permeability or mechanically weak formations that might lead to lost
20            cement Steps can then be taken to eliminate or reduce loss of cement to the formation.
21            Staged cementing (see below) can reduce the hydrostatic pressure on the formation and
22            may avoid fracturing weak formations (Lyons and Pligsa. 2004). Additives are also
23            available that will lessen the flow of cement into highly porous formations (API. 2010a: Ali
24            etal.. 2009).
25        •   Gas invasion and cement shrinkage during cement setting can also cause channels and
26            poor bonding. During the cementing process, the hydrostatic pressure from the cement
27            column keeps formation gas from entering the cement. As the cement sets (hardens), the
2 8            hydrostatic pressure decreases; if it becomes less than the formation pressure, gas can
29            enter the cement, leading to channels.  Cement also shrinks as it sets, which can lead to
30            poor bonding and formation of microannuli. These problems can be avoided by using
31            cement additives that increase setting time or expand to offset shrinkage (McDanieletal.,
32            2014: Wojtanowicz. 2008: Dusseaultetal.. 2000). Foamed cement can help alleviate
33            problems with shrinkage, although care needs to be taken in cement design to ensure the
34            proper balance of pressure between the cement column and formation (API, 2010a).
35            Cement additives are also available that will expand upon contact with certain fluids such
36            as hydrocarbons. These cements, termed self-healing cements, are relatively new but have
37            shown early promise in some fields (Ali etal.. 2009). Rotating the casing during cementing
38            will also delay cement setting. Another technique called pulsation, where pressure pulses
39            are applied to the cement while it is setting, also can delay cement setting and loss of
40            hydrostatic pressure until the cement is strong enough to resist gas penetration (Stein et
41            al.. 2003).
                   This document is a draft for review purposes only and does not constitute Agency policy.
      June 2015                                   D-5                  DRAFT—DO NOT CITE OR QUOTE

-------
      Hydraulic Fracturing Drinking Water Assessment                                             Appendix D


 1        •   Another important issue is removal of drilling mud. If drilling mud is not completely
 2            removed, it can gather on one side of the wellbore and prevent that portion of the
 3            wellbore from being adequately cemented. The drilling mud can then be eroded away after
 4            the cement sets, leaving a channel. Drilling mud can be removed by circulating a denser
 5            fluid (spacer fluid) to flush the drilling mud out fKirksey. 2013: Brufatto etal.. 20031
 6            Mechanical devices called scratchers can also be attached to the casing and the casing
 7            rotated or reciprocated to scrape drilling mud from the wellbore [Hyne. 2012: Crook.
 8            2008]. The spacer fluid, which is circulated prior to the cement to wash the drilling fluid
 9            out of the wellbore, must be designed with the appropriate properties and pumped in such
10            a way that it displaces the drilling fluid without mixing with the cement [Kirksey. 2013:
11            API. 2010a: Brufatto etal.. 20031.
12        •   Also of concern in horizontal wells is the possibility of solids settling at the bottom of the
13            wellbore and free water collecting at the top of the wellbore. This can lead to channels and
14            poor cement bonding. The cement slurry must be properly designed for horizontal wells to
15            minimize free water and solids settling.
16        •   If there is free water in the cement, pressure can cause water loss into the formation,
17            leaving behind poor cement or channels [Jiang etal.. 2012]. In horizontal wells, free water
18            can also accumulate at the top of the wellbore, forming a channel [Sabins. 1990].
19            Minimizing free water in the cement design and using fluid loss control additives  can help
20            control loss of water [Ross and King, 2007].

      D.2.2.2.  Cement Placement Techniques
21    The primary cement job is  most commonly conducted by pumping the cement down the inside of
22    the casing, then out the bottom of the casing where it is then forced up the space between the
23    outside of the casing and the formation. (The cement can also be placed in the space between two
24    casings.] If continuous cement (i.e., a sheath of cement placed along the entire wellbore] is
25    desired, cement is circulated through the annulus until cement that is pumped down the central
26    casing flows out of the annulus at the surface. A spacer fluid is often pumped ahead of cement to
27    remove any excess drilling fluid left in the wellbore; even if the operator does not plan to circulate
28    cement to the surface, the spacer fluid will still return to the surface, as this is necessary to remove
29    the drilling mud from the annulus. If neither the spacer fluid nor the cement returns to the surface,
30    this indicates that fluids are being lost into the formation.

31    Staged cementing is a technique that reduces pressure on the formation by decreasing the height
32    (and therefore the weight] of the cement column. This may be necessary if the estimated weight
33    and pressure associated with standard cement emplacement could damage zones where the
34    formation intersected is weak. The reduced hydrostatic pressure at the bottom of the cement
35    column can also reduce the loss of water to permeable formations, improving the quality of the
36    cement job. In multiple-stage cementing, cement is circulated to just below a cement collar placed
37    between two sections of casing. A cement collar will have been placed between two sections of
38    casing, just above, with ports that can be opened by dropping a weighted tool. Two plugs—which
39    are often referred to as bombs or darts because of their shape—are then dropped. The first plug is
                   This document is a draft for review purposes only and does not constitute Agency policy.
      June 2015                                   D-6                  DRAFT—DO NOT CITE OR QUOTE

-------
      Hydraulic Fracturing Drinking Water Assessment                                            Appendix D


 1    dropped, once the desired cement for the first stage has been pushed out of the casing by a spacer
 2    fluid. It closes the section of the well below the cement collar and stops cement from flowing into
 3    the lower portion of the well. The second plug (or opening bomb) opens the cement ports in the
 4    collar, allowing cement to flow into the annulus between the casing and formation. Cement is then
 5    circulated down the wellbore, out the cement ports, into the annulus, and up to the surface. Once
 6    cementing is complete, a third plug is dropped to close the cement ports, preventing the newly
 7    pumped cement from flowing back into the well (J^ronjLiniiJB^^         see Figure D-l.
 8   Another less commonly used primary cementing technique is reverse circulation cementing. This
 9   technique has been developed to decrease the force exerted on weak formations. In reverse
10   circulation cementing, the cement is pumped down the annulus directly between the outside of the
1 1   outermost casing and the formation. This essentially allows use of lower density cement and lower
1 2   pumping pressures. With reverse circulation cementing, greater care must be taken in calculating
1 3   the required cement, ensuring proper cement circulation, and locating the beginning and end of the
14   cemented portion.

15   Another method used to cement specific portions of the well without circulating cement along the
1 6   entire wellbore length is to use a cement basket. A cement basket is a device that attaches to the
1 7   well casing. It is made of flexible material such as canvas or rubber that can conform to the shape of
1 8   the wellbore. The cement basket acts as a one-way barrier to cement flow. Cement can be circulated
19   up the wellbore past the cement basket, but when circulation stops the basket prevents the cement
2 0   from falling back down the wellbore. Cement baskets can be used to isolate weak formations or
2 1   formations with voids. They can also be placed above large voids such as mines or caverns with
2 2   staged cementing used to cement the casing above the void.

23   If any deficiencies are identified, remedial cementing may be performed. The techniques available
24   to address deficiencies in the primary cement job including cement squeezes or top-job cementing.
25   A cement squeeze injects cement under high pressure to fill in voids or spaces in the primary
26   cement job caused by high pressure, failed formations, or improper removal of drilling mud.
27   Although cement squeezes can be used to fix deficiencies in the primary cement job, they require
28   the well to be perforated, which can weaken the well and make it susceptible to degradation by
29   pressure and temperature cycling as would occur during fracturing (Crescent,_2011). Another
30   method of secondary cementing is the top job. In a top job, cement is pumped down the annulus
31   directly to fill the remaining uncemented space when cement fails to circulate to the surface.
                   This document is a draft for review purposes only and does not constitute Agency policy.
     June 2015                                   D-7                  DRAFT—DO NOT CITE OR QUOTE

-------
 Hydraulic Fracturing Drinking Water Assessment
                               Appendix D
                     Second-stage slurry
                  Second-stage flow path
                        Circulating ports

                          Cement collar
                          Opening bomb
                        First-stage slurry
                     First-stage flow path
          Note: Figure not to scale


Two-Stage Cementing Process
Figure D-l. A typical staged cementing process.
               This document is a draft for review purposes only and does not constitute Agency policy.
June 2015                                        D-8                    DRAFT—DO NOT CITE OR QUOTE

-------
      Hydraulic Fracturing Drinking Water Assessment
                                       Appendix D
      D.3.   Well Completions
 1    Completion refers to how the well is prepared for production and how flow is established between
 2    the formation and the surface. Figure D-2 presents examples of well completion types, including
 3    cased, formation packer, and open hole completion.

                            Legend
                          Cement
                          Casing
                          Wellbore
                          Induced fracture
                   Cemented Casing
                     Completion
                         Fracture stage
                               Perforations
                     1111
                Vd
 Formation Packer
    Completion
        Fracture stage


^^~ EJ CT  B8  E8 E8^
Open Hole
Completion
Fracture stage
  HfTT
                                                             \
                                                             Packer
       Note: Nottoscale. Conductor casing not shown.
      Figure D-2. Examples of well completion types.
               Configurations shown include cased, formation packer, and open hole completion. From U.S.
               EPA(2015f).
 4    A cased completion, where the casing extends to the end of the wellbore and is cemented in place,
 5    is the most common configuration of the well in the production zone [U.S. EPA. 2015Q. Perforations
 6    are made through the casing and cement and into the formation using small explosive charges
 7    called "perf guns" or other devices, such as sand jets. Hydraulic fracturing then is conducted
 8    through the perforations. This is a common technique in wells that produce from several different
 9    depths and in low-permeability formations that are fractured [Renpu. 2011). While perforations do
10    control the initiation point of the fracture, this can be a disadvantage if the perforations are not
11    properly aligned with the local stress field. If the perforations are not aligned, the fractures will
12    twist to align with the stress field, leading to tortuosity in the fractures and making fluid movement
13    through them more difficult [Cramer, 2008]. Fracturing stages can be isolated from each  other
14    using various mechanisms such as plugs or baffle rings, which close off a section of the well when a
15    ball of the correct size is dropped down the well.
                   This document is a draft for review purposes only and does not constitute Agency policy.
      June 2015                                    D-9                  DRAFT—DO NOT CITE OR QUOTE

-------
      Hydraulic Fracturing Drinking Water Assessment                                            Appendix D


 1    A packer is a mechanical device used to selectively seal off certain s ections of the wellbore.
 2    Packers can be used to seal the space between the tubing and casing, between two casings, or
 3    between the production casing and formation. The packer has one or more rubber elements that
 4    can be manipulated downhole to increase in diameter and make contact with the inner wall of the
 5    next-largest casing or the formation, effectively sealing the annulus created between the outside of
 6    the tubing and the inside of the casing. Packers vary in how they are constructed and how they are
 7    set, based on the downhole conditions in which they are used. There are two types of packers:
 8    internal packers and formation packers. Internal packers are used to seal the space between the
 9    casing and tubing or between two different casings. They isolate the outer casing layers from
10    produced fluids and prevent fluid movement into the annulus. Formation packers seal the space
11    between the casing and the formation and are often used to isolate fracture stages; they can be used
12    to separate an open hole completion into separate fracture stages. Packers can seal an annulus by
13    several different mechanisms. Mechanical packers expand mechanically against the  formation and
14    can exert a significant force on the formation. Swellable packers have elastomer sealing elements
15    that swell when they come into contact with a triggering fluid such as water or hydrocarbons. They
16    exert less force on the formation and can seal larger spaces but take some time to fully swell
17                               Internal mechanical integrity tests such as pressure tests can verify
18    that the packer is functioning as designed and has not corroded or deteriorated.

19    In an open hole completion, the production casing extends just into the production zone and the
20    entire length of the wellbore through the production zone is left uncased. This is only an option in
21    formations where the wellbore is stable enough to not collapse into the wellbore. In formations that
22    are unstable, a slotted liner may be used in open hole completions to control sand production
23    (Rejipu^JLQll). Perforations are not needed in an open hole completion, since the production zone
24    is not cased. An open hole completion can be fractured in a single  stage  or in multiple stages.

25    If formations are to be fractured in stages, additional completion methods are needed to separate
26    the stages from each other and control the location of the fractures. One possibility is use of a liner
27    with formation packers to isolate each stage. The liner is equipped with sliding sleeves that can be
28    opened by dropping balls down the casing to open each stage. Fracturing typically occurs from the
29    end of the well and continues toward the beginning of the production zone.

      D.4.   Mechanical Integrity Testing
30    While proper design and construction of the well's casing and cement are important, it is also
31    important to verify the well was constructed and is performing as designed. Mechanical integrity
32    tests (MITs) can verify that the well was constructed as planned and can detect damage to the
33    production well that occurs during operations, including hydraulic fracturing activities. Verifying
34    that a well has mechanical integrity can prevent potential impacts to drinking water resources by
35    providing early warning of a problem with the well or cement and allowing repairs.

36    It is important to note that if a well fails an MIT, this does not mean the well has failed  or that an
37    impact on drinking water resources has occurred. An MIT failure is a warning that one or more
38    components of the well are not performing as designed and is an indication that corrective actions
                   This document is a draft for review purposes only and does not constitute Agency policy.
      June 2015                                   D-10                  DRAFT—DO NOT CITE OR QUOTE

-------
       Hydraulic Fracturing Drinking Water Assessment                                              Appendix D


 1    are necessary. If well remediation is not performed, a loss of well integrity could occur, which could
 2    result in fluid movement from the well.

      D.4.1.  Internal Mechanical Integrity
 3    Internal mechanical integrity is an absence of significant leakage in the tubing, casing, or packers
 4    within the well system. Loss of internal mechanical integrity is usually due to corrosion or
 5    mechanical failure of the well's tubular and mechanical components.

 6    Internal mechanical integrity can be tested by the use of pressure testing, annulus pressure
 7    monitoring, ultrasonic monitoring, and casing inspection logs or caliper logs:
 8         •   Pressure testing involves raising the pressure in the wellbore to a set level and shutting
 9            in the well. If the well has internal mechanical integrity, the pressure should remain
10            constant with only small changes due to temperature fluctuation. Typically, the well is
1 1            shut in (i.e., production is stopped and the wellhead valves closed) for half an hour, and if
12            the pressure remains within 5% of the original reading, the well is considered to have
13            passed the test Usually, the well is pressure tested to the maximum expected pressure; for
14            a well to be used for hydraulic fracturing this would be the pressure applied during
15            hydraulic fracturing. Pressure tests, however, can cause debonding of the cement from the
16            casing, so test length is often limited to reduce  this effect (APL_2010a).
17         •   If the annulus between the tubing and casing is sealed by a packer, annulus pressure
1 8            monitoring can give an indication of the integrity of the tubing and casing. If the tubing,
19            casing, and packer all have mechanical integrity, the pressure in the annulus should not
20            change except for small changes in response to temperature fluctuations. The annulus can
21            be filled with a non-corrosive liquid and the level of the liquid can be used as another
22            indication of the integrity of the casing, tubing, and packer. The advantage of monitoring
23            the tubing/production casing annulus is it can give a continuous, real-time indication of
24            the internal integrity of the well. Even if the annulus is not filled with a fluid, monitoring its
25            pressure can indicate leaks. If pressure builds up in the annulus and then recovers quickly
2 6            after having bled off, that condition is referred  to as sustained casing pressure or surface
27            casing vent flow and is a sign of a leak in the tubing or casing
28            Monitoring of annuli between other sets of casings can also provide information on the
29            integrity of those casings. It can also provide information on external mechanical integrity
30            for annuli open to the formation (see Section D.4.2 for additional information on external
3 1            MITs). Ja£ksOTLeLiljllll3J also note that monitoring annular pressure allows the
32            operator to vent gas before it accumulates enough pressure to cause migration into
33            drinking water resources. Measuring annulus flow rate also allows detection of gas
34            flowing into the annulus (
35         •   A newer tool uses ultrasonic monitors to detect leaks in casing and other equipment It
3 6            measures the attenuation of an ultrasonic signal as it is transmitted through the wellbore.
37            The tool measures transmitted ultrasonic signals as it is lowered down the wellbore. The
38            tool can pick up ultrasonic signals created by the leak, similar to noise logs. The tool only
                   This document is a draft for review purposes only and does not constitute Agency policy.
      June 2015                                    D-ll                  DRAFT—DO NOT CITE OR QUOTE

-------
      Hydraulic Fracturing Drinking Water Assessment                                             Appendix D


 1            has a range of a few feet but is claimed to detect leaks as small as half a cup per minute
 2            (JuiiajlJ^^
 3        •   Caliper logs have mechanical fingers that extend from a central tool and measure the
 4            distance from the center of the wellbore to the side of the casing. Running a caliper log can
 5            identify areas where corrosion has altered the diameter of the casing or where holes have
 6            formed in the casing. Caliper logs may also detect debris or obstructions in the well. Casing
 7            inspection and caliper logs are primarily used to determine the condition of the casing.
 8            Regular use of them may identify problems such as corrosion and allow mitigation before
 9            they cause of loss of integrity to the casing. To run these logs in a producing well, the
10            tubing must first be pulled.
11        •   Casing inspection logs are instruments lowered into the casing to inspect the casing for
12            signs of wear or corrosion. One type of casing log uses video equipment to detect
13            corrosion or holes. Another type uses electromagnetic pulses to detect variations in metal
14            thickness. Running these logs in a producing well requires the tubing to be  pulled.
15    If an internal mechanical integrity problem is detected, first, the location of the problem must be
16    found. Caliper or casing inspection logs can detect locations of holes in casing. Locations of leaks
17    can also be detected by sealing off different sections of the well using packers and performing
18    pressure tests on each section until the faulty section is located. If the leaks are in the tubing or a
19    packer, the problem may be remedied by replacing the well component Casing leaks may be
20    remedied by performing a cement squeeze (see the section on cementing).

      D.4.2.  External Mechanical Integrity
21    External well mechanical integrity is demonstrated by establishing the absence of significant fluid
22    movement along the outside of the casing, either between the outer casing and cement or between
23    the cement and the wellbore. Failure of an external MIT can indicate improper cementing or
24    degradation of the cement emplaced in the annular space between the outside of the casing and the
25    wellbore. This type of failure can lead  to movement of fluids out of intended production zones and
26    toward drinking water resources.

27    Several types of logs are available to evaluate external mechanical integrity, including temperature
2 8    logs, noise logs, oxygen activation logs, radioactive tracer logs, and cement evaluation logs.

29        •   Temperature logs measure  the temperature in the wellbore. They are capable of
3 0            measuring small changes in temperature. They can be performed using instruments that
31            are lowered down the well on a wireline or they can be done using fiber optic sensors
3 2            permanently installed in the  well. When performed immediately after cementing, they can
3 3            detect the heat from the cement setting and determine the location of the top of cement.
34            After the cement has set, temperature logs can sense the difference in temperatures
35            between formation fluids and injected or produced fluids. They may also detect
36            temperature changes due to  cooling or warming caused by flow. In this way temperature
37            logs may detect movement of fluid outside the casing in the wellbore (ArthyrJMll).
                   This document is a draft for review purposes only and does not constitute Agency policy.
      June 2015                                   D-12                 DRAFT—DO NOT CITE OR QUOTE

-------
      Hydraulic Fracturing Drinking Water Assessment                                             Appendix D


 1            Temperature logs require interpretation of the causes of temperature changes and are
 2            therefore subject to varying results among different users.
 3        •   Noise logs are sensitive microphones that are lowered down the well on a wireline. They
 4            are capable of detecting small noises caused by flowing fluids, such as fluids flowing
 5            through channels in the cement [Arthur. 2012]. They are most effective at detecting fast-
 6            moving gas leaks and less successful with more slowly moving liquid migration.
 7        •   Oxygen activation logs consist of a neutron source and one or more detectors that are
 8            lowered on a wireline. The neutron source bombards oxygen molecules surrounding the
 9            wellbore and converts them into unstable nitrogen molecules that rapidly decay back to
10            oxygen, emitting gamma radiation in the process. Gamma radiation detectors above or
11            below the neutron source measure how quickly the oxygen molecules are moving away
12            from the source, thereby determining flow associated with water.
13        •   Radioactive tracer logs involve release of a radioactive tracer and then passing a
14            detector up or down the wellbore to measure the path the tracers have taken. They can be
15            used to determine if fluid is flowing up the wellbore. Tracer logs can be very sensitive but
16            may be limited in the range over which leaks can be detected.
17        •   Cement evaluation logs (also known as cement bond logs) are acoustic logs consisting of
18            an instrument that sends out acoustic signals along with receivers, separated by some
19            distance, that record the acoustic signals. As the acoustic signals pass through the casing
20            they will be attenuated to an extent, depending on whether the pipe is free or is bonded to
21            cement. By analyzing the return acoustic signal, the degree of cement bonding with the
2 2            casing can be determined. The cement evaluation log measures the sound attenuation as
2 3            sound waves passing through the cement and casing. There are different types of cement
24            evaluation logs available. Some instruments can only return an average value over the
2 5            entire wellbore.  Other instruments are capable of measuring the cement bond radially.
2 6            Cement logs do not actually determine whether fluid movement through the annulus is
2 7            occurring. They only can determine whether cement is present in the annulus and in some
2 8            cases can give a qualitative assessment of the quality of the cement in the annulus. Cement
29            evaluation logs are used to calculate a bond index which varies between 0 and 1, with 1
3 0            representing the strongest bond and 0 representing the weakest bond.
31    If the well fails an external MIT, damaged or missing cement may be repaired using a cement
32    squeeze (Wojtanowicz. 2008]. A cement squeeze involves injection of cement slurry into voids
33    behind the casing or into permeable formations. Different types of cement squeezes are available
34    depending on  the location of the void needing to be filled and well conditions [Kirksey. 2013].
35    Cement squeezes are not always successful, however, and may need to be repeated to successfully
36    seal off flow [Wojtanowicz. 2008].

      D.5.   References for Appendix D
      AM, M: Taoutaou, S: Shafqat, AD: Salehapour, A: Noor, S. (2009]. The use of self healing cement to ensure long
        term zonal isolation for HPHT wells subject to hydraulic fracturing operations in Pakistan. Paper
        presented at International Petroleum Technology Conference, December 7-9,2009, Doha, Qatar.


                   This document is a draft for review purposes only and does not constitute Agency policy.
      June 2015                                   D-13                 DRAFT—DO NOT CITE OR QUOTE

-------
 Hydraulic Fracturing Drinking Water Assessment                                                 Appendix D
API (American Petroleum Institute). (1999). Recommended practice for care and use of casing and tubing
   [Standard] (18th ed.). (API RP 5C1). Washington, DC: API Publishing Services.

API (American Petroleum Institute). (2004). Recommended practice for centralizer placement and stop collar
   testing (First ed.). (API RP 10D-2 (R2010)).

API (American Petroleum Institute). (2009a). Hydraulic fracturing operations - Well construction and
   integrity guidelines [Standard] (First ed.). Washington, DC: API Publishing Services.

API (American Petroleum Institute). (2009b). Packers and bridge plugs (Second ed.). (API Spec 11D1).

API (American Petroleum Institute). (2010a). Isolating potential flow zones during well construction
   [Standard] (1st ed.). (RP 65-2). Washington, DC: API Publishing Services.
   http://www.techstreet.com/products/preview/1695866

API (American Petroleum Institute). (2010b). Specification for cements and materials for well cementing
   [Standard] (24th ed.). (ANSI/API SPECIFICATION 10A). Washington, DC: API Publishing Services.
   http://www.techstreet.com/products/1757666

API (American Petroleum Institute). (2011). Specification for casing and tubing - Ninth edition [Standard]
   (9th ed.). (API SPEC 5CT). Washington, DC: API Publishing Services.
   http://www.techstreet.com/products/1802047

API (American Petroleum Institute). (2013). Recommended practice for testing well cements [Standard] (2nd
   ed.). (RP 10B-2). Washington, DC: API Publishing Services.
   http://www.techstreet.com/products/1855370

Arthur. ID. (2012). Understanding and assessing well integrity relative to wellbore stray gas intrusion issues.
   Presentation presented at Ground Water Protection Council Stray Gas - Incidence & Response Forum, July
   24-26,2012, Cleveland, OH.

Brufatto. C: Cochran. I: Conn. L: El-Zeghaty. SZA. A: Fraboulet. B: Griffin. T: lames. S: Munk. T: lustus. F: Levine.
   IR: Montgomery. C: Murphy. D: Pfeiffer. I: Pornpoch. T: Rishmani. L. (2003). From mud to cement -
   Building gas wells. Oilfield Rev 15: 62-76.

CAPP (Canadian Association of Petroleum Producers). (2013). CAPP hydraulic fracturing operating practice:
   Wellbore construction and quality assurance. (2012-0034).
   http://www.capp.ca/getdoc.aspx?DocId=218137&DT=NTV

Cramer. DP. (2008). Stimulating unconventional reservoirs: Lessons learned, successful practices, areas for
   improvement. SPE Unconventional Reservoirs Conference, February 10-12, 2008, Keystone, CO.

Crescent (Crescent Consulting, LLC). (2011). East Mamm creek project drilling and cementing study.
   Oklahoma City, OK. http://cogcc.state.co.us/Library/PiceanceBasin/EastMammCreek/ReportFinal.pdf

Crook, R. (2008). Cementing: Cementing horizontal wells. Halliburton.

Dusseault. MB: Gray. MN: Nawrocki. PA. (2000). Why oilwells leak: Cement behavior and long-term
   consequences. Paper presented at SPE International Oil and Gas Conference and Exhibition in China,
   November 7-10, 2000, Beijing, China.

Enform. (2013). Interim industry recommended practice 24: fracture stimulation: Interwellbore
   communication 3/27/2013 (1.0 ed.). (IRP 24). Calgary, Alberta: Enform Canada.
   http://www.enform.ca/safety resources/publications/PublicationDetails.aspx?a=29&type=irp

Hyne, Nl. (2012). Nontechnical guide to petroleum geology, exploration, drilling and production. In
   Nontechnical guide to petroleum geology, exploration, drilling and production (3 ed.). Tulsa, OK: PennWell
   Corporation.

Jackson. RE: Gorody. AW: Mayer. B: Roy. IW: Ryan. MC: Van Stempvoort. DR. (2013). Groundwater protection
   and unconventional gas extraction: the critical need for field-based hydrogeological research. Ground
   Water 51: 488-510. http://dx.doi.org/10.llll/gwat.12074
               This document is a draft for review purposes only and does not constitute Agency policy.

June 2015                                       D-14                   DRAFT—DO NOT CITE OR QUOTE

-------
 Hydraulic Fracturing Drinking Water Assessment                                                 Appendix D
liang. L: Guillot. D: Meraji. M: Kumari. P: Vidick. B: Duncan. B: Gaafar. GR: Sansudin. SB. (2012). Measuring
   isolation integrity in depleted reservoirs. SPWLA 53rd Annual Logging Symposium, June 16-20, 2012,
   Cartagena, Colombia.

lulian. IY: King. GE: lohns. IE: Sack. IK: Robertson. DB. (2007). Detecting ultrasmall leaks with ultrasonic leak
   detection, case histories from the North Slope, Alaska. Paper presented at International Oil Conference and
   Exhibition in Mexico, June 27-30, 2007, Veracruz, Mexico.

Kirksey, I. (2013). Optimizing wellbore integrity in well construction. Presentation presented at North
   American Wellbore Integrity Workshop, Octoberl6-17,2013, Denver, CO.

Lyons, WC: Pligsa, Gl. (2004). Standard handbook of petroleum and natural gas engineering (2nd ed.).
   Houston, TX: Gulf Professional Publishing, http://www.elsevier.com/books/standard-handbook-of-
   petroleum-and-natural-gas-engineering/lyons-phd-pe/978-0-7506-7785-l

McDaniel. BW: Rispler. KA. (2009). Horizontal wells with multistage fracs prove to be best economic
   completion for many low permeability reservoirs. Paper presented at SPE Eastern Regional Meeting,
   September 23-15, 2009, Charleston, WV.

McDaniel. I: Watters. L: Shadravan. A. (2014). Cement sheath durability: Increasing cement sheath integrity to
   reduce gas migration in the Marcellus Shale Play. In SPE hydraulic fracturing technology conference
   proceedings. Richardson, TX: Society of Petroleum Engineers. http://dx.doi.org/10.2118/168650-MS

MSC (Marcellus Shale Coalition). (2013). Recommended practices: Drilling and completions. (MSC RP 2013-
   3). Pittsburgh, Pennsylvania.

Renpu. W. (2011). Advanced well completion engineering (Third ed.). Houston, TX: Gulf Professional
   Publishing.

Ross. D: King. G. (2007). Well completions. In MJ Economides; T Martin (Eds.), Modern fracturing: Enhancing
   natural gas production (1 ed., pp. 169-198). Houston, Texas: ET Publishing.

Sabins. F. (1990). Problems in cementing horizontal wells. J Pet Tech 42: 398-400.
   http://dx.doi.org/10.2118/20005-PA

Stein, D: Griffin Jr. Tl: Dusterhoft, D.  (2003). Cement pulsation reduces remedial cementing costs. GasTIPS 9:
   22-24.

U.S. EPA (U.S. Environmental Protection Agency). (2015f). Review of well operator files for hydraulically
   fractured oil and gas production  wells: Well design and construction [EPA Report]. (EPA/601/R-14/002).
   Washington, D.C.: Office of Research and Development, U.S. Environmental Protection Agency.

Watson. TL: Bachu. S. (2009). Evaluation of the potential for gas and C02 leakage along wellbores. SPE
   Drilling & Completion 24:115-126. http://dx.doi.org/10.2118/106817-PA

Wojtanowicz, AK. (2008). Environmental control of well integrity. In ST Orszulik (Ed.), Environmental
   technology in the oil industry (pp. 53-75). Houten, Netherlands: Springer Netherlands.
               This document is a draft for review purposes only and does not constitute Agency policy.

June 2015                                       D-15                    DRAFT—DO NOT CITE OR QUOTE

-------
 Hydraulic Fracturing Drinking Water Assessment                            Appendix E
                     Appendix E
Flowback and Produced Water Supplemental
Tables and Information
        This document is a draft for review purposes only and does not constitute Agency policy.

June 2015                                   DRAFT—DO NOT CITE OR QUOTE

-------
    Hydraulic Fracturing Drinking Water Assessment                                    Appendix E
    Appendix E.  Flowback and Produced Water
      Supplemental Tables and Information
    E.I.   Flowback and Long-Term Produced Water Volumes
1   The EPA       estimates of flowback volumes and long-term produced water volumes used to
2   generate the summaries appearing in Table 7-3 of Chapter 7 appear below in Table E-l.
              This document is a draft for review purposes only and does not constitute Agency policy.
    June 2015                             E-l               DRAFT—DO NOT CITE OR QUOTE

-------
Hydraulic Fracturing Drinking Water Assessment
                                                                                         Appendix E
Table E-l. Flowback and long-term produced water characteristics for wells in unconventional formations, formation-level data.
          Source:
Basin
Anadarko
Appalachian
Arkoma
Denver-
Julesburg
Resource
Type
Shale
Tight
Shale
Shale
Shale
Tight
Unconventional
Formation
Woodford
Cleveland
Granite Wash

Mississippi Lime
Marcellus
Utica
Fayetteville
Niobrara
Codell
Codell-Niobrara
Drill
Type
H
H
V
H
V
H
H
V
H
H
H
V
D
V
H
D
V
Fracturing Fluid
(Mgal)
Median
4.7
0.81
0.69
6.2
0.56
1.8
4.4
2.6
4.0
5.1
2.6
0.32
0.28
0.27
2.6
0.45
0.30
Range
1.0-12
0.2-4.0
0.11-3
0.2-9.4
0.05-3
0.82-2.4
0.9-11
0.53-6.6
1.0-11
1.7-11
0.73-3.4
0.27-3.3
0.21-0.46
0.13-0.46
0.15-2.7
0.21-0.47
0.13-0.46
Number of
Data Points
2,239
144
4
77
26
428
14,010
66
150
1,668
69
367
78
185
62
116
592
Flowback
(% of Fracturing Fluid Returned)
Median
34
-
-
-
-
-
7
40
4
-
13
11
-
-
7
-
-
Range3
20-50
12-40
-
7-22
-
50
4-47
21-60
2-27
10-20
6-25
7-35
-
-
-
-
-
Number of
Data Points
3
2
2
2
2
1
4,374
7
73
2
16
9
0
0
32
0
0
Long-Term Produced Water Rates
(gpd)
Median
5,500
82
32
1,300
500
-
860
230
510
430
680
340
-
-
34
-
29
Rangeb
3,200-6,400
20-300
6.6-170
0-2,200
170-1,300
37,000-120,000
54-13,000
100-1,200
210-1,200
150-2,300
260-810
240-600
-
-
19-140
-
13-65
Number of
Data Points
198
571
390
273
2,413
4
4,984
714
82
2,305
250
5,474
0
0
32
0
1,677
June 2015
This document is a draft for review purposes only and does not constitute Agency policy.
                               E-2
DRAFT—DO NOT CITE OR QUOTE

-------
Hydraulic Fracturing Drinking Water Assessment
                                                                                                    Appendix E
Basin
Denver-
Julesburg,
cont.
Fort Worth
Green River
Green River,
cont.
Illinois
Michigan
Permian
Resource
Type
Tight
cont.
Shale
Shale
Tight

Shale
Shale
Shale
Unconventional
Formation
Muddy J
Barnett
Hilliard-Baxter-
Mancos
Lance
Mesaverde
New Albany
Antrim
Avalon & Bone
Spring
Barnett-
Woodford
Drill
Type
D
V
H
V
H
V
D
D
V
H
V
D
H
H
Fracturing Fluid
(Mgal)
Median
0.59
0.28
3.6
1.3
1.7
1.3
1.2
0.23
0.17
--
--
2.2
1.1
2.1
Range
.025-0.62
0.16-0.62
1-7.3
0.4-1.9
1.0-5.6
0.81-3.5
0.76-1.9
0.16-0.31
0.081-
0.29
-
0.05
0.94-4.5
0.73-2.8
0.5-4.5
Number of
Data Points
162
292
23,917
3,589
2
29
180
73
14
0
1
20
17
2
Flowback
(% of Fracturing Fluid Returned)
Median
-
-
30
-
-
3
6
8
21
-
-
13
-
-
Range3
-
-
21-40
-
-
1-50
1-17
0-37
6-83
-
25-75
5-31
-
-
Number of
Data Points
0
0
11
0
0
31
170
61
11
0
2
16
0
0
Long-Term Produced Water Rates
(gpd)
Median
230
55
920
250
37
410
860
190
290
-
-
950
0
-
Rangeb
64-390
9.3-500
160-4,200
170-580
15-58
250-580
360-1,200
150-440
140-610
2,900
4,600
220-2,400
0-2,300
-
Number of
Data Points
3
129
10,349
3,318
7
1,050
1,140
445
1,081
2
1
183
37
0
June 2015
This document is a draft for review purposes only and does not constitute Agency policy.
                                   E-3
DRAFT—DO NOT CITE OR QUOTE

-------
Hydraulic Fracturing Drinking Water Assessment
                                                                                                    Appendix E
Basin
Permian,
cont.
San Juan
TX-LA-MS
Resource
Type
Shale,
cont.
Tight
Tight
Shale
Tight
Unconventional
Formation
Devonian (TX)
Wo If camp
Spraberry
Mesaverde (San
Juan)
Dakota
Bossier
Haynesville
Cotton Valley
Drill
Type
H
V
H
D
V
V
D
V
D
H
V
D
H
V
H
D
V
Fracturing Fluid
(Mgal)
Median
0.32
0.27
1.4
1.3
0.81
--
--
0.2
0.12
2.7
0.4
0.28
5.3
0.61
4.2
.48
.28
Range
0.13-0.89
0.12-1.0
1.1-3.9
0.26-1.7
0.078-1.7
1.0
-
0.063-
0.22
0.07-0.3
1.7-3.6
0.19-1.7
0.13-0.8
0.95-15
.14-3.5
.25-6.0
.084-4.0
.019-.94
Number of
Data Points
10
16
55
12
60
1
0
19
52
2
16
21
3,222
9
30
24
76
Flowback
(% of Fracturing Fluid Returned)
Median
-
-
-
16
-
-
-
-
4
-
-
-
5
-
-
-
-
Range3
-
-
-
15-20
-
-
-
-
1-40
-
-
-
5-30
-
<60
<60
<60
Number of
Data Points
0
0
0
3
0
0
0
0
30
0
0
0
3
0
2
2
2
Long-Term Produced Water Rates
(gpd)
Median
880
400
3,000
310
910
870
18
65
160
750
470
320
1,700
210
770
950
640
Rangeb
310-1,800
150-3,000
210-19,000
22-8,700
130-1,700
100-4,000
12-260
29-120
41-370
610-1,200
180-1,100
130-1,300
84-1,800
56-850
130-2,700
630-1,800
370-1,800
Number of
Data Points
381
162
104
259
926
66
48
6
379
25
1,203
253
1,249
263
335
1801
10,717
June 2015
This document is a draft for review purposes only and does not constitute Agency policy.
                                   E-4
DRAFT—DO NOT CITE OR QUOTE

-------
Hydraulic Fracturing Drinking Water Assessment
                                                                                                        Appendix E
Basin
TX-LA-MS,
cont.
Western Gulf
Williston
Resource
Type
Tight,
cont.
Shale
Tight

Unconventional
Formation
Travis Peak
Eagle Ford
Pearsall
Austin Chalk
Vicksburg
Wilcox Lobo
Olmos
Bakken
Drill
Type
H
V
H
V
H
H
V
D
H
V
D
V
H
V
Fracturing Fluid
(Mgal)
Median
3.0
0.9
5.0
2.9
3.7
0.94
.016
0.11
2.1
0.21
.058
--
2.0
1.1
Range
0.25-6
0.2-4
1.0-14
2.0-4.1
3.3-4.1
0.58-1.3
0.084-0.6
0.1-0.13
0.66-2.6
0.06-0.6
.056-.076
0.15
0.35-10
.35-2.9
Number of
Data Points
2
2
2,485
9
2
15
20
4
4
14
3
2
2,203
12
Flowback
(% of Fracturing Fluid Returned)
Median
-
-
4
-
-
-
-
-
-
-
-
-
19
-
Range3
-
-
2-8
-
-
-
-
-
-
-
-
-
5-47
-
Number of
Data Points
0
0
1,800
0
0
0
0
0
0
0
0
0
206
0
Long-Term Produced Water Rates
(gpd)
Median
200
980
110
-
200
720
1,000
-
330
620
-
-
680
1,000
Rangeb
39-1,700
330-1,800
9.1-250
-
54-370
290-2,400
650-1,900
-
62-740
330-1,400
-
-
380-1,500
340-3,100
Number of
Data Points
5
1,380
498
0
12
1,097
937
0
77
1,514
0
0
1,739
222
 "--" indicates no data; H, horizontal well; D, directional well; V, vertical well.
 a For some formations, if only one data point was reported, the EPA reported it in the range column and did not report a median value.
 b For some formations, the number of data points was not reported in the data source. In these instances, the EPA reported the number of data points as equal to one, even if
 the source reported a range and median value.
June 2015
This document is a draft for review purposes only and does not constitute Agency policy.
                                     E-5
DRAFT—DO NOT CITE OR QUOTE

-------
      Hydraulic Fracturing Drinking Water Assessment                                              Appendix E


      E.2.    Produced Water Content

      E.2.1.  Introduction
 1    In the main text of Chapter 7, we describe aspects of flowback and produced water composition,
 2    including temporal changes in water quality parameters of flowback (Section 7.5) and major classes
 3    of compounds in produced water (Section 7.6). In section 7.7 we describe variability as occurring
 4    on three levels: between different rock types (e.g., coal vs. sandstone), between formations
 5    composed of the same rock types (e.g., Barnett Shale vs. Bakken Shale), and within formations of
 6    the same rock type (e.g., northeastern vs. southwestern Marcellus Shale). In this appendix we
 7    present data from the literature which illustrates the differences among these three variability
 8    levels.

      E.2.2.  General Water Quality Parameters
 9    As noted in Chapter 7, the EPA identified data characterizing the content of unconventional
10    flowback and produced water in a total of 12 shale and tight formations and coalbed methane
11    (CBM) basins. These formations and basins span 18 states. Note that in this subsection we treat all
12    fluids as produced water. As a consequence, the variability of reported concentrations is likely
13    higher than if the data could be standardized to a specific point on the flowback-to-produced water
14    continuum. Table E-2 and Table E-3 provide supporting data on general water quality parameters
15    of produced water for 12  formations.
                   This document is a draft for review purposes only and does not constitute Agency policy.
      June 2015                                    E-6                   DRAFT—DO NOT CITE OR QUOTE

-------
Hydraulic Fracturing Drinking Water Assessment
                                                                                   Appendix E
Table E-2. Reported concentrations of general water quality parameters in produced water for unconventional shale and tight
         formations, presented as: average (minimum-maximum) or median (minimum-maximum).
Parameter
States
Acidity
Alkalinity
Ammonium
Bicarbonate
Biochemical
oxygen
demand
(BOD)
Carbonate
Chloride
Chemical
oxygen
demand
Units
n/a
mg/L
mg/L
mg/L
mg/L
mg/L
mg/L
mg/L
mg/L
Shales
Bakken3
MT, ND
-
-
-
291
(122-610)
-
-
119,000
(90,000-
133,000)
-
Barnettb
TX
NC
(ND-ND)
725
(215-1,240)
-
-
582
(101-2,120)
-
34,700
(9,600-
60,800)
2,945
(927-3,150)
Fayettevillec
AR
-
1,347
(811-1,896)
-
-
-
-
9,156
(5,507-
12,287)
-
Marcellus
PAd
NC
(<5-473)
165
(8-577)
-
-
-
-
57,447
(64-
196,000)
15,358
(195-
36,600)
PA, WVe
162
(5-925)
99.8
(7.5-577)
-
-
141
(2.8-12,400)
-
49,000
(64.2-
196,000)
4,670
(195-
36,600)
Tight formations
Cotton Valley
Group'
LA, TX
-
-
89
(40-131)
-
-
-
101,332
(3,167-
221,498.7)
-
Devonian
Sandstone6
PA
-
99
(43-194)
-
524
(ND-8,440)
-
-
132,567
(58,900-
207,000)
-
Mesaverde'
CO, NM, UT,
WY
-
-
-
2,230
(1,281-13,650)
-
227
(ND-1,680)
4,260
(8-
75,000)
-
Oswego'
OK
-
582
(207-1,220)
-
-
-
-
44,567
(23,000-
75,000)
-
June 2015
This document is a draft for review purposes only and does not constitute Agency policy.
                             E-7
DRAFT—DO NOT CITE OR QUOTE

-------
Hydraulic Fracturing Drinking Water Assessment
                                                                                                    Appendix E
Parameter
States
DO
DOC
Hardness as
CaCO3
Oil and
grease
pH
Specific
conductivity
Specific
gravity
IDS
Total
Kjeldahl
nitrogen
Units
n/a
mg/L
mg/L
mg/L
mg/L
SU
|oS/cm
--
mg/L
mg/L
Shales
Bakken3
MT, ND
-
-
-
-
5.87
(5.47-6.53)
213,000
(205,000-
220,800)
1.13
(1.0961-
1.155)
196,000
(150,000-
219,000)
-
Barnettb
TX
-
11.2
(5.5-65.3)
5,800
(3,500-21,0
00)
163.5
(88.2-
1,430)
7.05
(6.5-7.2)
111,500
(34,800-
179,000)
-
50,550
(16,400-
97,800)
171
(26-298)
Fayettevillec
AR
-
-
-
-
-
-
-
13,290
(9,972-
15,721)
-
Marcellus
PAd
-
-
34,000
(630-
95,000)
74
(5-802)
6.6
(5.1-8.4)
-
-
106,390
(680-
345,000)
-
PA, WVe
-
117
(3.3-5,960)
25,000
(156-
106,000)
16.85
(4.7-802)
6.5
(4.9-7.9)
183,000
(479-
763,000)
-
87,800
(680-
345,000)
94.9
(5.6-312)
Tight formations
Cotton Valley
Group'
LA, TX
-
-
-
-
-
-
-
164,683
(5,241-
356,666)
-
Devonian
Sandstone8
PA
0.8
(0.2-2.5)
-
-
-
6.3
(5.5-6.8)
184,800
(118,000-
211,000)
-
235,125
(106,000-
354,000)
-
Mesaverde'
CO, NM, UT,
WY
-
-
-
-
8
(5.8-11.62)
-
-
15,802
(1,032-
125,304)
-
Oswego'
OK
-
-
-
-
6.3
(6.1-6.4)
-
-
73,082
(56,541-
108,813)
-
June 2015
This document is a draft for review purposes only and does not constitute Agency policy.
                                   E-8
DRAFT—DO NOT CITE OR QUOTE

-------
Hydraulic Fracturing Drinking Water Assessment
                                                                                                           Appendix E
Parameter
States
TOC
Total
suspended
solids
Turbidity
Units
n/a
mg/L
mg/L
NTU
Shales
Bakken3
MT, ND
-
-
-
Barnettb
TX
9.75
(6.2-36.2)
242
(120-535)
239
(144-314)
Fayettevillec
AR
-
-
-
Marcellus
PAd
160
(1.2-1,530)
352
(4-7,600)
-
PA, WVe
89.2
(1.2-5,680)
127
(6.8-3,220)
126
(2.3-1,540)
Tight formations
Cotton Valley
Group'
LA, TX
198
(184-212)
-
-
Devonian
Sandstone8
PA
-
-
-
Mesaverde'
CO, NM, UT,
WY
-
-
-
Oswego'
OK
-
-
-
 n/a, not applicable; -, no value available; NC, not calculated; ND, not detected., SU= standard units, bolded italic numbers are medians
 a SteQ§n_et^lI_[2010}. n = 3. Concentrations were calculated based on Stepan et al.'s raw data. Samples had charge balance errors of 1.74, -0.752, and -0.220%
                         ^ n = 16. This data source reported concentrations without direct presentation of raw data.
 c .Warner_et,a,LJ_2013}. n = 6. Concentrations were calculated based on Warner et al.'s raw data. Both flowback and produced water included.
 d Barbot et al. (2013). n = 134-159. This data source reported concentrations without direct presentation of raw data.
 eMiM§§120091. n = 31-67. Concentrations were calculated based on Hayes's raw data. Both flowback and produced water included. Non-detects and contaminated blanks
  omitted.
 f B!ondes_et_iL_{20141. Cotton Valley Group, n=2; Mesa Verde, n = 1-407; Oswego, n = 4-30. Concentrations were calculated based on raw data presented in the U.S.
  Geological Survey (USGS) National Produced Water Database v2.0.
 5 Drese|_and_Rose_{2010}. n = 3-15. Concentrations were calculated based on Dresel and Rose's raw data.
June 2015
This document is a draft for review purposes only and does not constitute Agency policy.
                                      E-9
DRAFT—DO NOT CITE OR QUOTE

-------
Hydraulic Fracturing Drinking Water Assessment
Appendix E
Table E-3. Reported concentrations of general water quality parameters in produced water
           for unconventional coalbed basins, presented as: average (minimum-maximum).
Parameter
States
Alkalinity
Ammonium
Bicarbonate
Carbonate
Chloride
Chemical oxygen
demand
Dissolved oxygen
DOC
Hardness as CaCO3
Hydrogen sulfide
Oil and grease
pH
Phosphate
Specific
conductivity
IDS
Total Kjeldahl
nitrogen
TOC
Total suspended
solids
Turbidity
Units
n/a
mg/L
mg/L
mg/L
mg/L
mg/L
mg/L
mg/L
mg/L
mg/L
mg/L
mg/L
SU
mg/L
US/cm
mg/L
mg/L
mg/L
mg/L
NTU
Black Warrior3
AL, MS
355 (3-1,600)
3.60(0.16-8.91)
427 (2-1,922)
3 (0-64)
9,078 (11-42,800)
830 (0-10,500)
-
3.37 (0.53-61.41)
871 (3-6,150)
-
-
7.5 (5.3-9.0)
0.435
(0.026-3.570)
20,631
(718-97,700)
14,319
(589-61,733)
6.08(0.15-38.40)
6.03 (0.00-103.00)
78 (0-2,290)
74 (0-539)
Powder Riverb
MT,WY
1,384 (653-2,672)
-
1,080 (236-3,080)
2.17(0.00-139.0)
21 (BDL-282)
-
1.07 (0.11-3.48)
3.18(1.09-8.04)
-
-
-
7.71(6.86-9.16)
BDL(BDL-BDL)
1,598
(413-4,420)
997 (252-2,768)
0.48 (BDL-4.70)
3.52 (2.07-6.57)
11.0 (1.4-72.7)
8.2 (0.7-57.0)
Ratonb
CO, NM
1,107 (130-2,160)
-
1,124 (127-2,640)
51.30
(1.30-316.33)
787 (4.8-8,310)
-
0.39 (0.01-3.52)
1.26 (0.30-8.54)
-
4.41 (BDL-190.0)
9.10(0.60-17.6)
8.19(6.90-9.31)
0.04 (BDL-1.00)
3,199 (742-11,550)
2,512 (244-14,800)
2.61(BDL-26.10)
1.74 (0.25-13.00)
32.3 (1.0-580.0)
4.5 (0.3-25.0)
San Juanb
AZ, CO, NM, UT
3,181 (51-11,400)
-
3,380 (117-13,900)
40.17 (0.00-1,178)
624(BDL-20,100)
-
0.51 (0.04-1.69)
3.21 (0.89-11.41)
-
23.00
(23.00-23.00)
-
7.82 (5.40-9.26)
1.89 (BDL-9.42)
5,308
(232-18,066)
4,693 (150-39,260)
0.46 (BDL-3.76)
2.91(0.95-9.36)
47.2 (1.4-236.0)
61.6 (0.8-810.0)
 n/a, not applicable; -, no value available; BDL, below detection limit.
 a          n = 206. Concentrations were calculated based on raw data presented in the reference.
 b Dahm_et_aL_{2011}. Powder River, n = 31; Raton, n = 40; San Juan, n = 20. This data source reported concentrations without
 presentation of raw data.
               This document is a draft for review purposes only and does not constitute Agency policy.

June 2015                                       E-10                    DRAFT—DO NOT CITE OR QUOTE

-------
      Hydraulic Fracturing Drinking Water Assessment                                              Appendix E


      E.2.3.  Salinity and Inorganics
 1    Table E-4 and Table E-5 provide supporting data on salinity and inorganic constituents of produced
 2    water for 12 formations.

      £.2.3.1.  Processes Controlling Salinity and Inorganics Concentrations
 3    Multiple mechanisms likely control elevated salt concentrations in flowback and produced water
 4    and are largely dependent upon post-injection fluid interactions and the formation's stratigraphic
 5    and hydrogeologic environment [Barbotetal.. 2013]. High inorganic ionic loads observed in
 6    flowback and produced water are expressed as TDS.

 7    Subsurface brines or formation waters are saline fluids associated with the targeted formation.
 8    Shale and sandstone brines are typically much more saline than coalbed waters. After hydraulic
 9    fracturing fluids are injected into the subsurface, the injected fluids (which are typically not sources
10    of high TDS) mix with in situ brines, which typically contain high ionic loads [Haluszczak et al..
11    20131.

12    Deep brines, present in over- or underlying strata, may naturally migrate into targeted formations
13    over geologic time or artificially intrude if a saline aquifer is breached during hydraulic fracturing
14    (Chapman etal.. 2012: Maxwell. 2011: Blauchetal.. 20091 Whether it is through natural or induced
15    intrusion, saline fluids may contact the producing formation and introduce novel salinity sources to
16    the produced water [Chapman etal., 2012].

17    The dissolution salts associated with formation solids both increases TDS concentrations and alters
18    formation porosity and permeability [Blauchetal., 2009]. Additionally, the mobilization of connate
19    fluids (deposition-associated pore fluids] and formation fluids during hydraulic fracturing likely
20    contributes to increased TDS levels (Dresel and Rose, 2010: Blauchetal., 2009]. Despite the general
21    use of fresh water for hydraulic fracturing fluid,  some elevated salts in produced water may result
22    from the use of reused saline flowback or produced water as a hydraulic fracturing base fluid
23    (Hayes. 2009].
                   This document is a draft for review purposes only and does not constitute Agency policy.
      June 2015                                    E-ll                  DRAFT—DO NOT CITE OR QUOTE

-------
Hydraulic Fracturing Drinking Water Assessment
                                                                                   Appendix E
Table E-4. Reported concentrations (mg/L) of inorganic constituents contributing to salinity in unconventional shale and tight
          formations produced water, presented as: average (minimum-maximum) or median (minimum-maximum).
Parameter
States
Bromide
Calcium
Chloride
Fluoride
Iodine
Nitrate as N
Nitrite as N
Phosphorus
Potassium
Shale
Bakken3
MT, ND
-
9,680
(7,540-
13,500)
119,000
(90,000-
133,000)
-
-
-
-
NC
(ND-0.03)
2,970
(0-5,770)
Barnettb
TX
589
(117-798)
1,600
(1,110-6,730)
34,700
(9,600-60,800)
3.8
(3.5-12.8)
-
-
4.7
(3.5-38.1)
0.395
(0.19-0.7)
316
(80-750)
Fayettevillec
AR
111
(96-144)
317
(221-386)
9,156
(5,507-12,287)
-
-
NC
(ND-ND)
-
-
-
Marcellus
PAd
511
(0.2-1,990)
7,220
(38-41,000)
57,447
(64-
196,000)
-
-
-
-
-
-
PA,WVe
512
(15.8-1,990)
7,465
(173-33,000)
49,000
(64.2-196,000)
0.975
(0.077-32.9)
-
1.7
(0.65-15.9)
11.8
(1.1-146)
0.3 (0,08-21.8)
337
(38-3,950)
Tight Formations
Cotton Valley
Group'
LA, TX
498
(32-1,338)
19,998
(181-51,400)
101,332
(3,167-
221,498.7)
-
20
(1-36)
-
-
-
1,975
(8-7,099)
Devonian
Sandstone6
PA
1,048
(349-1,350)
20,262
(8,930-
34,400)
132,567
(58,900-
207,000)
-
39
(11-56)
-
-
-
858
(126-3,890)
Mesaverde'
CO, NM, UT,
WY
-
212
(1.01-4,580)
4,260
(8-75,000)
-
1.01
(1.01-1.01)
0.6
(0.6-0.6)
-
-
160
(4-2,621)
Oswego'
OK
-
5,903
(3,609-8,662)
44,567
(23,000-75,000)
-
-
-
-
-
-
June 2015
This document is a draft for review purposes only and does not constitute Agency policy.
                             E-12
DRAFT—DO NOT CITE OR QUOTE

-------
Hydraulic Fracturing Drinking Water Assessment
                                                                                                          Appendix E
Parameter
States
Silica
Sodium
Sulfate
Sulfide
Sulfite
IDS
Shale
Bakken3
MT, ND
7
(6.41-7)
61,500
(47,100-
74,600)
660
(300-1,000)
-
-
196,000
(150,000-
219,000)
Barnettb
TX
-
18,850
(4,370-28,200)
709
(120-1,260)
NC
(ND-ND)
-
50,550
(16,400-
97,800)
Fayettevillec
AR
52
(13-160)
3,758
(3,152-4,607)
NC
(ND-3)
-
-
13,290
(9,972-
15,721)
Marcellus
PAd
-
21,123
(69-
117,000)
71
(0-763)
-
-
106,390
(680-
345,000)
PA,WVe
-
21,650
(63.8-95,500)
58.9
(2.4-348)
3.2
(1.6-5.6)
12.4
(5.2-73.6)
87,800
(680-345,000)
Tight Formations
Cotton Valley
Group'
LA, TX
4
(4-4)
39,836
(1,320-
85,623.24)
407
(ND-
2,200.46)
-
-
164,683
(5,241-
356,666)
Devonian
Sandstone8
PA
-
58,160
(24,400-
83,300)
20
(1-140)
0.7
(0.1-2.5)
-
235,125
(106,000-
354,000)
Mesaverde'
CO, NM, UT,
WY
-
5,828
(132-48,817)
837
(ND-14,612)
-
-
15,802
(1,032-
125,304)
Oswego'
OK
-
19,460
(13,484-31,328)
183
(120-271)
-
-
73,082
(56,541-
108,813)
 -, no value available; NC, not calculated; ND, not detected. Boldeditalic numbers are medians.
 a St§Ban_et_aLJ20101- n = 3. Concentrations were calculated based on Stepan et al.'s raw data. Samples had charge balance errors of 1.74, -0.752, and -0.220%
 b Mil§i^DiSevennJ^Oj,2bi. n = 16. This data source reported concentrations without presentation of raw data.
 c W§£D§L§ljL-i2013). n = 6. Concentrations were calculated based on Warner et al.'s raw data. Both flowback and produced water included.
 d Si£bot_et_aL_(2013|. n = 134-159. This data source reported concentrations without presentation of raw data.
 e             " = S'65- Concentrations were calculated based on Hayes's raw data. Both flowback and produced water included. Non-detects and contaminated blanks
  omitted.
 f BjMHtes_etjLJ20141 Cotton Valley Group, n = 2; Mesa Verde, n = 1-407; Oswego, n = 4-30. Concentrations were calculated based on raw data presented in the USGS
  National Produced Water Database v2.0.
                       n = 3-15. Concentrations were calculated based on Dresel and Rose's raw data.
June 2015
This document is a draft for review purposes only and does not constitute Agency policy.
                                     E-13
DRAFT—DO NOT CITE OR QUOTE

-------
     Hydraulic Fracturing Drinking Water Assessment
Appendix E
     Table E-5. Reported concentrations (mg/L) of inorganic constituents contributing to salinity in
               produced water for unconventional CBM basins, presented as: average
               (minimum-maximum).
Parameter
State
Barium
Boron
Bromide
Calcium
Chloride
Fluoride
Magnesium
Nitrate
Nitrite
Phosphorus
Potassium
Silica
Sodium
Strontium
Sulfate
IDS
Black Warrior3
AL, MS
45.540 (0.136-352)
0.185(0-0.541)
-
218 (0-1,640)
9,078 (11-42,800)
6.13 (0.00-22.60)
68.12(0.18-414.00)
8.70 (0.00-127.50)
0.03 (0.00-2.08)
0.32 (0.00-5.76)
12.02 (0.46-74.00)
8.66(1.04-18.10)
4,353 (126-16,700)
11.354 (0.015-142.000)
5.83 (0.00-302.00)
14,319 (589-61,733)
Powder Riverb
MT,WY
0.61(0.14-2.47)
0.17 (BDL-0.39)
0.09 (BDL-0.26)
32.09 (2.00-154.0)
21 (BDL-282)
1.57 (0.40-4.00)
14.66 (BDL-95.00)
-
-
-
11.95 (BDL-44.00)
6.46 (4.40-12.79)
356 (12-1,170)
0.60(0.10-1.83)
5.64 (BDL-300.0)
997 (252-2,768)
Ratonb
CO, NM
1.67 (BDL-27.40)
0.36 (BDL-4.70)
4.86 (0.04-69.60)
14.47 (0.81-269.0)
787 (4.8-8,310)
4.27 (0.59-20.00)
3.31(0.10-56.10)
-
-
-
6.37 (BDL-29.40)
7.05 (4.86-10.56)
989 (95-5,260)
5.87 (BDL-47.90)
14.75 (BDL-253.00)
2,512 (244-14,800)
San Juanb
AZ, CO, NM, UT
10.80 (BDL-74.0)
1.30 (0.21-3.45)
9.77 (BDL-43.48)
53.29 (1.00-5,530)
624(BDL-20,100)
1.76 (0.58-10.00)
15.45 (BDL-511.0)
-
-
-
26.99 (BDL-970.0)
12.37 (3.62-37.75)
1,610 (36-7,834)
5.36 (BDL-27.00)
25.73 (BDL-1,800)
4,693 (150-39,260)
     -, no value available; BDL, below detection limit.
     a          n = 206. Concentrations were calculated based on the authors' raw data.
     b               Powder River, n = 31; Raton, n = 40; San Juan, n = 20. This data source reported concentrations without
     presentation of raw data.
     E.2.4.  Metals and Metalloids
1    Table E-6 and Table E-7 provide supporting data on metal constituents of produced water for 12
2    formations.
                   This document is a draft for review purposes only and does not constitute Agency policy.

     June 2015                                     E-14                   DRAFT—DO NOT CITE OR QUOTE

-------
Hydraulic Fracturing Drinking Water Assessment
                                                                                    Appendix E
Table E-6. Reported concentrations (mg/L) of metals and metalloids from unconventional shale and tight formation produced
          water, presented as: average (minimum-maximum) or median (minimum-maximum).
          Note that calcium, potassium, and sodium appear in Table E-4.
Parameter
States
Aluminum
Antimony
Arsenic
Barium
Beryllium
Boron
Cadmium
Chromium
Cobalt
Shale
Bakken3
MT, ND
-
-
-
10
(0-24.6)
-
116
(39.9-192)
-
-
-
Barnettb
TX
0.43
(0.37-2.21)
NC
(ND-ND)
NC
(ND-ND)
3.6
(0.93-17.9)
NC (ND-ND)
30.3
(7.0-31.9)
NC
(ND-ND)
0.03
(0.01-0.12)
0.01
(0.01-0.01)
Fayettevillec
AR
-
-
-
4
(3-5)
-
4.800
(2.395-
21.102)
-
-
-
Marcellus
PAd
-
-
-
2,224
(0.24-13,80
0)
-
-
-
-
-
PA, WVe
2.57
(0.22-47.2)
0.028
(0.018-0.038)
0.101
(0.013-0.124)
542.5
(2.590-
13,900)
-
12.2
(0.808-145)
-
0.079
(0.011-0.567)
-
Tight Formation
Cotton Valley
Group'
LA, TX
-
-
-
160
(ND-400.52)
-
37
(2-100)
-
-
-
Devonian
Sandstone6
PA
-
-
-
1,488
(7-4,370)
-
-
-
-
-
Mesaverde'
CO, NM, UT,
WY
-
-
-
139
(4-257)
-
10
(1-14.2)
-
-
-
Oswego'
OK
-
-
-

-
-
-
-
-
June 2015
This document is a draft for review purposes only and does not constitute Agency policy.
                             E-15
DRAFT—DO NOT CITE OR QUOTE

-------
Hydraulic Fracturing Drinking Water Assessment
                                                                                                    Appendix E
Parameter
States
Copper
Iron
Lead
Lithium
Magnesium
Manganese
Mercury
Molybdenum
Nickel
Selenium
Silver
Shale
Bakken3
MT, ND
NC
(ND-0.21)
96
(ND-120)
-
-
1,270
(630-1,750)
7
(4-10.2)
-
NC
(ND-<0.2)
-
-
-
Barnettb
TX
0.29
(0.06-0.52)
24.9
(12.1-93.8)
0.02
(0.01-0.02)
19.0
(2.56-37.4)
255
(149-755)
0.86
(0.25-2.20)
NC
(ND-ND)
0.02
(0.02-0.03)
0.04
(0.03-0.05)
0.03
(0.03-0.04)
-
Fayettevillec
AR
-
7
(1-13)
-
9.825
(2.777-
28.145)
61
(47-75)
2
(2-3)
-
-
-
-
-
Marcellus
PAd
-
-
-
-
632
(17-2,550)
-
-
-
0.1815
(0.007-
0.137)
-
-
PA, WVe
0.506
(0.253-4.150)
53.65
(2.68-574)
0.066
(0.003-0.970)
53.85
(3.410-323)
678
(40.8-2,020)
2.825
(0.369-
18.600)
0.00024
-
0.419
(0.068-0.769)
0.004
4
(3-6)
Tight Formation
Cotton Valley
Group'
LA, TX
0.7
(0.48-1)
-
-
23
(1-53)
1,363
(27-3,712.98)
30.33
(30.33-30.33)
-
-
-
-
-
Devonian
Sandstone8
PA
0.04
(0.01-0.13)
188
(90-458)
0.02
(0.01-0.04)
97.8
(20.2-315)
2,334
(797-3,140)
19
(5.6-68)
-
-
-
-
-
Mesaverde'
CO, NM, UT,
WY
-
9
(1-29)
-
3
(1-33)
74
(1-2,394)
-
-
-
-
-
-
Oswego'
OK
-
61
(41-78)
-
-
753
(486-1,264)
-
-
-
-
-
-
June 2015
This document is a draft for review purposes only and does not constitute Agency policy.
                                   E-16
DRAFT—DO NOT CITE OR QUOTE

-------
Hydraulic Fracturing Drinking Water Assessment
                                                                                                           Appendix E
Parameter
States
Strontium
Thallium
Tin
Titanium
Zinc
Shale
Bakken3
MT, ND
764
(518-1,010)
-
-
-
7
(2-11.3)
Barnettb
TX
529
(48-1,550)
NC
(ND-0.14)
NC
(ND-ND)
0.02
(0.02-0.03)
0.15
(0.10-0.36)
Fayettevillec
AR
27
(14-49)
-
-
-
-
Marcellus
PAd
1,695
(0.6-8,460)
-
-
-
-
PA, WVe
1,240
(0.580-8,020)
0.168
-
-
0.391
(0.087-247)
Tight Formation
Cotton Valley
Group'
LA, TX
2,312
(39-9,770)
-
-
-
-
Devonian
Sandstone8
PA
3,890
(404-13,100)
-
-
-
0.20
(0.03-1.26)
Mesaverde'
CO, NM, UT,
WY
-
-
-
-
-
Oswego'
OK
-
-
-
-
-
 -, no value available; NC, not calculated; ND, not detected; BDL, below detection limit. Bolded italic numbers are medians.
 a Stej>an_et_al._{2010}. n = 3. Concentrations were calculated based on Stepan et al.'s raw data.
 b JHa^es^nd_Seyerin_t2012bl. n = 16. This data source reported concentrations without presentation of raw data.
 c Warner_et_aL_(2013). n = 6. Concentrations were calculated based on Warner et al.'s raw data. Both flowback and produced water included.
 d Barbo£e££JI_[2013}. n = 134-159. This data source reported concentrations without presentation of data.
 e Ha¥§s_{2009}. n = 48. Concentrations were calculated based on Hayes's raw data. Both flowback and produced water included. Non-detects and contaminated blanks
 omitted.
 f BJ°!ld§s_et_aLJ_2014). Cotton Valley Group, n = 2; Mesa Verde, n = 1-407; Oswego, n = 4-30. Concentrations were calculated based on raw data presented in the USGS
 National Produced Water Database v2.0.
 g|)reseJ_ajTdJRoseJ2010}. n = 3-15. Concentrations were calculated based on Dresel and  Rose's raw data.
June 2015
This document is a draft for review purposes only and does not constitute Agency policy.
                                     E-17
DRAFT—DO NOT CITE OR QUOTE

-------
Hydraulic Fracturing Drinking Water Assessment
Appendix E
Table E-7. Reported concentrations (mg/L) of metals and metalloids from unconventional
          coalbed produced water, presented as: average (minimum-maximum).
Parameter
States
Aluminum
Antimony
Arsenic
Barium
Beryllium
Boron
Cadmium
Calcium
Cesium
Chromium
Cobalt
Copper
Iron
Lead
Lithium
Magnesium
Manganese
Mercury
Molybdenum
Nickel
Potassium
Rubidium
Selenium
Silver
Sodium
Strontium
Thallium
Tin
Titanium
Black Warrior3
AL, MS
0.037 (0-0.099)
0.006 (0.00-0.022)
0.002 (0.0-0.085)
45.540 (0.136-352)
0.0 (0.0-0.008)
0.185 (0-0.541)
0.001 (0.00-0.015)
218 (0-1,640)
0.011 (0.0-0.072)
0.002 (0.0-0.351)
0.023(0.00-0.162)
0.001 (0.0-0.098)
8.956 (0.045-93.100)
0.008 (0.00-0.250)
1.157 (0-8.940)
68.12(0.18-414.00)
0.245 (0.006-4.840)
0.000 (0.000-0.000)
0.002 (0-0.083)
0.015 (0.0-0.358)
12.02 (0.46-74.00)
0.013(0.0-0.114)
0.002 (0.00-0.063)
0.015 (0.0-0.565)
4,353 (126-16,700)
11.354 (0.015-142.000)
-
0.00 (0.00-0.009)
0.003 (0.0-0.045)
Powder Riverb
MT,WY
0.018 (BDL-0. 124)
BDL(BDL-BDL)
0.001 (BDL-0.004)
0.61(0.14-2.47)
BDL(BDL-BDL)
0.17 (BDL-0.39)
BDL (BDL-0.002)
32.09 (2.00-154.0)
-
0.012 (BDL-0.250)
BDL(BDL-BDL)
0.078 (BDL-1.505)
1.55 (BDL-190.0)
BDL(BDL-BDL)
0.13 (BDL-0.34)
14.66 (BDL-95.00)
0.02 (BDL-0. 16)
-
0.005 (BDL-0.029)
0.141 (BDL-2.61)
11.95 (BDL-44.00)
-
0.006 (BDL-0.046)
0.003 (0.003-0.003)
356 (12-1,170)
0.60(0.10-1.83)
-
0.006 (BDL-0.028)
BDL (BDL-0.002)
Raton b
CO, NM
0.193 (BDL-2,900)
BDL(BDL-BDL)
0.010 (BDL-0.060)
1.67 (BDL-27.40)
BDL(BDL-BDL)
0.36 (BDL-4.70)
0.002 (BDL-0.003)
14.47 (0.81-269.0)
-
0.105 (BDL-3.710)
0.001 (BDL-0.018)
0.091 (BDL-4.600)
7.18(0.09-95.90)
0.023 (BDL-0.233)
0.32 (0.01-1.00)
3.31 (0.10-56.10)
0.11(0.01-2.00)
-
0.002 (BDL-0.035)
0.015 (0.004-0.11)
6.37 (BDL-29.40)
-
0.017 (BDL-0.100)
0.015 (BDL-0.140)
989 (95-5,260)
5.87 (BDL-47.90)
-
0.008 (BDL-0.021)
BDL (BDL-0.002)
San Juanb
AZ, CO, NM, UT
0.069 (BDL-0.546)
BDL(BDL-BDL)
0.001 (BDL-0.020)
10.80 (BDL-74.0)
BDL(BDL-BDL)
1.30 (0.21-3.45)
0.002 (BDL-.006)
53.29 (1.00-5,530)
-
0.002 (BDL-0.023)
0.001 (BDL-0.017)
0.058 (BDL-0.706)
6.20 (BDL-258.0)
0.023 (BDL-0.390)
1.61 (0.21-4.73)
15.45 (BDL-511.0)
0.19(BDL-1.34)
-
0.020 (BDL-0.040)
0.020 (BDL-0.13)
26.99 (BDL-970.0)
-
0.018 (BDL-0.067)
BDL(BDL-BDL)
1,610 (36-7,834)
5.36 (BDL-27.00)
-
0.017 (BDL-0.039)
0.004 (BDL-0.020)
              This document is a draft for review purposes only and does not constitute Agency policy.

June 2015                                    E-18                  DRAFT—DO NOT CITE OR QUOTE

-------
      Hydraulic Fracturing Drinking Water Assessment
Appendix E
Parameter
States
Vanadium
Zinc
Black Warrior3
AL, MS
0.001 (0.0-0.039)
0.024 (0.0-0.278)
Powder Riverb
MT,WY
BDL(BDL-BDL)
0.063 (BDL-0.390)
Raton b
CO, NM
0.001 (BDL-0.013)
0.083 (0.010-3.900)
San Juanb
AZ, CO, NM, UT
BDL(BDL-BDL)
0.047 (0.005-0.263)
      -, no value available; BDL, below detection limit.
      a DOE (2014). n = 206. Concentrations were calculated based on the authors' raw data.
      b Dahm et al. (2011). Powder River, n = 31; Raton, n = 40; San Juan, n = 20. This data source reported concentrations without
      presentation of raw data.
      E.2.4.1.  Processes Controlling Mineral Precipitation and Dissolution
 1    Hydraulic fracturing treatments introduce fluids into the subsurface that are not in equilibrium
 2    with respect to formation mineralogy. Subsurface geochemical equilibrium modeling and
 3    saturation indices are therefore used to assess the solution chemistry of unconventional produced
 4    water and the subsequent likelihood of precipitation and dissolution reactions [Engle and Rowan,
 5    2014: Barbotetal.. 2013). Dissolution and precipitation reactions between fracturing fluids,
 6    formation solids, and formation water contribute to the chemistry of flowback and produced water.

 7    For example, early flowback fluids may be under-saturated with respect to certain constituents or
 8    minerals associated with formation solids. Through time, as fluid-rock geochemistry returns to
 9    equilibrium, formation minerals will dissolve into solution and return in flowback.

10    Depending upon the formation chemistry and composition of the hydraulic fracturing fluid, the
11    hydraulic fracturing fluid may initially have a lower ionic strength than existing formation fluids.
12    Consequently, salts, carbonate, sulfate, and silicate minerals may undergo dissolution or
13    precipitation. Proppants may also undergo dissolution or serve as nucleation sites for precipitation
14    [McLinetaL 20111.

15    Currently, relatively little literature quantitatively explores subsurface dissolution and
16    precipitation reactions between hydraulic fracturing fluids and formation solids and water.
17    However, the processes that take place will likely be a function of the solubilities of the minerals,
18    the chemistry of the fluid, pH, redox conditions, and temperature.

19    Documented dissolution processes in unconventional resources include the dissolution of feldspar
20    followed by sodium enrichment in coalbed produced water [Rice etal.. 2008]. Dissolution of
21    barium-rich minerals (barite (BaSCU) and witherite  (BaCOs)), and strontium-rich minerals (celestite
22    (SrSCU) and strontianite (SrCOs)) are known to enrich shale produced waters in barium and
23    strontium [Chapman etal.. 2012).

24    Known precipitation processes in unconventional resources include the precipitation of carbonate
2 5    and subsequent reduction of calcium and magnesium concentrations in coalbed produced water
26    [Rice etal.. 2008). Additionally, calcium carbonate precipitation is suspected to cause declines in pH
27    and alkalinity levels in shale produced water [Barbotetal., 2013].
                    This document is a draft for review purposes only and does not constitute Agency policy.
      June 2015                                     E-19                   DRAFT—DO NOT CITE OR QUOTE

-------
      Hydraulic Fracturing Drinking Water Assessment                                              Appendix E


 1    The subsurface processes associated with fluid-rock interactions take place over a scale of weeks to
 2    months through the generation of flowback and produced water. Note that the types and extent of
 3    subsurface dissolution and precipitation reactions change with time, from injection through
 4    flowback and production. For instance, Engle and Rowan [2014] found that early Marcellus Shale
 5    flowback was under-saturated with respect to gypsum (CaS04-2H20), halite (NaCl), celestite,
 6    strontianite, and witherite, indicating that these minerals would dissolve in the subsurface. Fluids
 7    were oversaturated with respect to barite. Saturation indices for gypsum, halite, celestite, and
 8    barite all increased during production. Knowing when dissolution and precipitation will likely
 9    occur is important, because dissolution and precipitation of minerals change formation
10    permeability and porosity, which can affect production [Andre etal., 2006].

11    Additionally, pyrite (FeS2] is an important minor mineral in reduced sedimentary rocks. Pyrite is
12    the primary form of sulfur and iron occurrence in shales [Leventhal and Hosterman. 1982] and is
13    also a common mineral phase generated in coals in which organic matter is closely associated
14    [Ward, 2002]. Pyrite content in shales can vary from less than 1% to several percent [Chermak and
15    Schreiber. 2014: Vulgamore etal.. 2007]. Researchers have found a strong association of trace
16    metals (i.e., nickel, copper, cadmium, chromium, cobalt, lead, selenium, vanadium, and zinc] with
17    pyrite in shales [Chermak and Schreiber. 2014: Tuttle etal.. 2009: Leventhal and Hosterman. 1982].

18    Although studies considering pyrite oxidation within the context of hydraulic fracturing are
19    currently lacking, it is likely that the introduction of oxygenated fluids to freshly exposed surfaces
20    in the subsurface during hydraulic fracturing can initiate limited, short-term pyrite oxidation or
21    dissolution. Pyrite dissolution may increase iron and trace element concentrations and acidity  in
22    produced waters [Nordstrom and Alpers. 1999: Moses and Herman. 1991].

23    The extent to which the oxidative dissolution of pyrite would exert a control on post-injection
24    subsurface fluid chemistry is unknown, although an ongoing U.S. Geological Survey (USGS] study
25    anticipates it may be more significant than previously hypothesized [Li and Brantley. 2011].
26    Regardless, relative to other reactions contributing to the  composition of flowback and produced
27    water (i.e., dissolution of salts], pyrite oxidation appears to be less significant. Ultimately, reactions
28    resulting from temporary changes in subsurface redox conditions will be less important relative to
29    other reactions that are less redox-dependent

      E.2.5.  Naturally Occurring Radioactive Material (NORM) and Technically Enhanced Naturally
             Occurring  Radioactive Material (TENORM)

      £.2.5.1. Formation Solids Levels of NORM
3 0    Elevated uranium levels in formation solids have been used to identify potential areas of natural
31    gas production for decades (Fertl and Chilingar. 1988]. Marine black shales are estimated to contain
32    an average of 5-20 ppm uranium depending on depositional conditions, compared to an average of
33    less than 5 ppm among all shales (USGS, 1961]. Shales that bear significant levels of uranium
34    include the Barnett in Texas, the Woodford in Oklahoma, the New Albany in the Illinois Basin, the
35    Chattanooga Shale in the southeastern United States, and a group of black shales in Kansas and
36    Oklahoma fSwanson. 19551

                   This document is a draft for review purposes only and does not constitute Agency policy.
      June 2015                                    E-20                  DRAFT—DO NOT CITE OR QUOTE

-------
      Hydraulic Fracturing Drinking Water Assessment                                             Appendix E
                     l identified Marcellus samples with uranium ranging from 4-72 ppm, with an
 2    average of 30 ppm. Additionally, shale samples taken from three counties within the Marcellus
 3    Shale had uranium concentrations ranging from 8 to 84 ppm (BTGS1_201Jj
 4    1981)- QlMBHlLiMJifihra^         compiled mineralogy and trace element data available in the
 5    literature for nine U.S. hydrocarbon-producing shales. In this combined data set, uranium levels
 6    among different shale plays were found to vary over three orders of magnitude, with samples of the
 7    Utica Shale containing approximately 0-5 ppm uranium and samples of the Woodford Shale
 8    containing uranium in the several-hundred-ppm range.
 9    VineXlSMl reported that the principal uranium-bearing coal deposits of the United States are
10    found in Cretaceous and Tertiary formations in the northern Great Plains and Rocky Mountains; in
11    some areas of the West, coal deposits have been found with uranium concentrations in the range of
12    thousands of ppm or greater. In contrast, most Mississippian, Pennsylvanian, and Permian coals in
13    the north-central and eastern United States contain less than 10 ppm uranium, rarely containing
14    50 ppm or more.

      £.2.5.2.   Produced Water Levels of TENORM
15    Background data on NORM in the Marcellus Shale and Devonian sandstones are given in Table E-8.
                   This document is a draft for review purposes only and does not constitute Agency policy.
      June 2015                                   E-21                  DRAFT—DO NOT CITE OR QUOTE

-------
Hydraulic Fracturing Drinking Water Assessment
                                                                                               Appendix E
Table E-8. Reported concentrations (in pCi/L) of radioactive constituents in unconventional shale and sandstone produced water,
           presented as: average (minimum-maximum) or median (minimum-maximum).
Parameter
States
Gross alpha
Gross beta
Radium-226
Radium-228
Total Radium
Uranium235
Uranium238
Marcellus
NY, PAb
6,845 (ND-123,000)
1,170 (ND-12,000)
1,869(1X10-16,920)
557(ND-2,589)
2,530(0.192-18,045)
1 (ND-20)
42 (ND-497)

Flowbackc
10,700 (288-71,000)
2,400 (742-21,300)
4,500 (551-25,500)
633 (248-1,740)
-
-
-
PA NORM STUDY

Conventional
Produced Waterd
1,835 (465-2,570)
909(402-1,140)
243 (81 - 819)
128 (26 - 896)
371 (107 - 1,715)
-
-
Unconventional
Produced Water0
11,300 (2,240-41,700)
3.445 (1.5-7,600)
6,300 (1,700-26,600)
941 (366-1,900)
7,180 (2,336-28,500)
-
-
Devonian Sandstone3
PA
-
-
2,367 (200-5,000)
-
-
-
-
 n/a, not applicable; -, no value available; BDL, below detection limit. Bolded italic numbers are medians.
 a £resej_and_Rose_[20101. n = 3. Concentrations presented were calculated based on Dresel and Rose's raw data.
 b                 " = 51. Concentrations presented were calculated based on Rowan et al.'s raw data for Marcellus samples. Uranium data from Barbot et al. (2013) n = 14.
              n = 9. Data reported in Table 3-14.
 d£^DE£120151. n = 9. Values calculated from Table 3-15 for unfiltered samples.
 e PA_DEP_{20151. n = 4. Values calculated from Table 3-15 for unfiltered samples.
June 2015
This document is a draft for review purposes only and does not constitute Agency policy.
                                 E-22
DRAFT—DO NOT CITE OR QUOTE

-------
      Hydraulic Fracturing Drinking Water Assessment                                             Appendix E
      £.2.5.3.  Mobilization of Naturally Occurring Radioactive Material
 1    Similar to conventional oil and gas production, in unconventional oil and gas production,
 2    radionuclides native to the targeted formation return to the surface with produced water. The
 3    principal radionuclides found in oil and gas produced waters include radium-226 of the uranium-
 4    238 decay series and radium-228 of the thorium-232 decay series [White. 1992]. Levels of
 5    TENORM in produced water are controlled by geologic and geochemical interactions between
 6    injected and formation fluids, and the targeted formation [Bank, 2011]. Mechanisms controlling
 7    NORM mobilization into produced water include  (1) the TENORM content of the targeted
 8    formation; (2) factors governing the release of radionuclides, particularly radium, from the
 9    reservoir matrix; and (3) the geochemistry of the produced water [Choppin. 2007. 2006: Fisher.
10    19981.

11    Organic-rich shales and coals are enriched in uranium, thorium, and other trace metals in
12    concentrations several times above those seen in typical shales or sedimentary rocks [Diehl etal.,
13    2004: USGS. 1997: Wignall and Myers. 1988: Tourtelot. 1979: Vine and Tourtelot. 1970). Unlike
14    shales and coals, sandstones are generally not organic-rich source rocks themselves. Instead,
15    hydrocarbons migrate into these formations over long periods of time [Clark and Veil, 2009]. Since
16    TENORM and organic contents are typically positively correlated due to the original, reduced
17    depositional environment [Fertl and Chilingar, 1988], it is unlikely that sandstones would be
18    enriched in TENORM to the same extent as oil- and gas-bearing shales and coals. Therefore, concern
19    related to TENORM within produced water is focused on operations targeting shales and coalbeds.

20    Radium is most soluble and mobile in chloride-rich, high-TDS, reducing environments [Sturchio et
21    al.. 2001: Zapecza and Szabo. 1988: Langmuir and Riese. 1985]. In formation fluids with high TDS,
22    calcium,  potassium, magnesium, and sodium compete with dissolved radium for sorption sites,
23    limiting radium sorption onto solids and allowing it to accumulate in solution at higher
24    concentrations [Fisher, 1998: Webster et al., 1995]. The positive correlation between TDS and
25    radium is well established and TDS is a useful indicator of radium and TENORM activity within
26    produced water, especially in lithologically homogenous reservoirs [Rowan etal.. 2011: Sturchio et
27    al.. 2001: Fisher. 1998: Kraemer and Reid. 1984].

28    Uranium and thorium are poorly soluble under reducing conditions and are therefore more
29    concentrated in formation solids than in solution [Fisher, 1998: Kraemer and Reid, 1984: Langmuir
30    and Herman. 1980]. However, because uranium becomes more soluble in oxidizing environments,
31    the introduction of relatively oxygen-rich fracturing fluids may promote the temporary
32    mobilization of uranium during hydraulic fracturing and early flowback. In addition, the physical
33    act of hydraulic fracturing creates fresh fractures and exposes organic-rich and highly reduced
34    surfaces  from which radionuclides could be released from the rock into formation fluids.

35    Produced water geochemistry determines, in part, the fate of subsurface radionuclides, particularly
36    radium. Radium may remain in the host mineral or it may be released into formation fluids, where
37    it can remain in solution as the dissolved Ra2+ ion, be adsorbed onto  oxide grain coatings or clay
38    particles by ion exchange, substitute for other cations during the precipitation of minerals, or form


                   This document is a draft for review purposes only and does not constitute Agency policy.
      June 2015                                    E-23                  DRAFT—DO NOT CITE OR QUOTE

-------
      Hydraulic Fracturing Drinking Water Assessment                                              Appendix E


 1    complexes with chloride, sulfate, and carbonate ions (Rowan_et_aL_2011.: Sturchio_etal.1_2001;
 2    LangmuiLandJlieseJ_1985.]. Uranium- and thorium-containing materials with a small grain size, a
 3    large surface-to-volume ratio, and the presence of uranium and thorium near grain surfaces
 4    promote the escape of radium into formation fluids. Vinsan_et_aL[2009} point to alpha decay along
 5    fracture surfaces as a primary control on radium mobilization in crystalline bedrock aquifers.
 6    Radium may also occur in formation fluids due to other processes, such as the decay of dissolved
 7    parent isotopes and adsorption-desorption reactions on formation surfaces (Sturchio_etaL_2001]-

 8    Preliminary results from fluid-rock interaction studies [Bank. 2011] indicate that a significant
 9    percentage of uranium in the Marcellus Shale may be subject to mobilization by hydrochloric acid,
10    which is used as a fracturing fluid additive. Understanding these processes will determine the
11    extent to which such processes might influence the TENORM content of flowback and produced
12    water.

      E.2.6.  Organics
13    Background data on organics in seven formations is given in Table E-9.
                   This document is a draft for review purposes only and does not constitute Agency policy.
      June 2015                                    E-24                  DRAFT—DO NOT CITE OR QUOTE

-------
Hydraulic Fracturing Drinking Water Assessment
                                                                                   Appendix E
Table E-9. Concentrations of select organic parameters from unconventional shale, a tight formation, and coalbed produced
          water, presented as: average (minimum-maximum) or median (minimum-maximum).
Parameter
States
TOC
DOC
BOD
Oil and grease
Benzene
Toluene
Ethylbenzene
Unit
n/a
mg/L
mg/L
mg/L
mg/L
Hg/L
Hg/L
Hg/L
Shale
Barnett3
TX
9.75
(6.2-36.2)
11.2
(5.5-65.3)
582
(101-2,120)
163.5
(88.2-1,430)
680
(49-5,300)
760
(79-8,100)
29
(2.2-670)
Marcellus
PAb
160
(1.2-
1,530)
43
(5-695)
-
74
(5-802)
-
-
-
PA, WVC
89.2
(1.2-5680)
117
(3.3-5,960)
141
(2.8-
12,400)
16.9
(4.7-802)
220
(5.8-2,000)
540
(5.1-6,200)
42
(7.6-650)
Tight
Formation
Cotton Valley
Groupd
LA, TX
198
(184-212)
-
-
-
-
-
-
Coal
Powder Rivere
MT,WY
3.52
(2.07-6.57)
3.18
(1.09-8.04)
-
-
-
-
-
Ratone
CO, NM
1.74
(0.25-13.00)
1.26
(0.30-8.54)
-
9.10
(0.60-17.6)
4.7
(BDL-220.0)
4.7 (BDL-78.0)
0.8 (BDL-18.0)
San Juan8
AZ, CO, NM, UT
2.91
(0.95-9.36)
3.21
(0.89-11.41)
-
-
149.7
(BDL-500.0)
1.7
(BDL-6.2)
10.5 (BDL-24.0)
Black Warrior'
AL, MS
6.03
(0.00-103.00)
3.37
(0.53-61.41)
-
-
-
-
-
June 2015
This document is a draft for review purposes only and does not constitute Agency policy.
                             E-25
DRAFT—DO NOT CITE OR QUOTE

-------
Hydraulic Fracturing Drinking Water Assessment
                                                                                                          Appendix E
Parameter
States
Xylenes
Average total BTEXB
Unit
n/a
Hg/L
Hg/L
Shale
Barnett3
TX
360
(43-1,400)
1,829
Marcellus
PAb
-
2,910
PA, WVC
300
(15-6,500)
1,102
Tight
Formation
Cotton Valley
Groupd
LA, TX
-
-
Coal
Powder Rivere
MT,WY
-
-
Raton"
CO, NM
9.9
(BDL-190.0)
20.1
San Juan8
AZ, CO, NM, UT
121.2
(BDL-327.0)
283.1
Black Warrior'
AL, MS
-
-
 n/a, not applicable; -, no value available; BDL, below detection limit. Bolded italic numbers are medians.
 a jHayes^nd_Severin_[2012b). n = 16. This data source reported concentrations without presentation of raw data.
 b Barbot^t^L_{2013}. n = 55; no presentation of raw data.
 c Hayes_{2009} n = 13-67. Concentrations were calculated based on Hayes' raw data. Both flowback and produced water included. Non-detects and contaminated blanks
 omitted.
 d Blondes_et^lI_[2014}. n = 2. Concentrations were calculated based on raw data presented in the USGS National Produced Water Database v2.0.
 e Dahm_et_aL_(2011). Powder River, n = 31; Raton, n = 40; San Juan, n = 20. This data source reported concentrations without presentation of raw data.
 f DOE_{2014}. n = 206. Concentrations were calculated based on the authors' raw data.
 g Average total BTEX was calculated by summing the average/median concentrations of benzene, toluene, ethylbenzene, and xylenes for a unique formation or basin. Minimum
 to maximum ranges were not calculated due to inaccessible raw data.
June 2015
This document is a draft for review purposes only and does not constitute Agency policy.
                                     E-26
DRAFT—DO NOT CITE OR QUOTE

-------
      Hydraulic Fracturing Drinking Water Assessment                                             Appendix E


 1    Several classes of naturally occurring organic chemicals are present in conventional and
 2    unconventional produced waters, with large concentration ranges (Lee_MMlNML20il)- These
 3    organic classes include total organic carbon (TOG); saturated hydrocarbons; BTEX (benzene,
 4    toluene, ethylbenzene, and xylenes); and polyaromatic hydrocarbons (PAHs) (see Table E-9). While
 5    TOG concentrations in produced water are detected at the milligrams to grams per liter level,
 6    concentrations of individual organic compounds are typically detected at the micrograms to
 7    milligrams per liter level.

 8    TOG indicates the level of dissolved and undissolved organics in produced water, including non-
 9    volatile and volatile organics (AchM^aMMJMll)- TOG concentrations in conventional produced
10    water vary widely from less than 0.1 mg/L to more than 11,000 mg/L. Average TOG concentrations
11    in unconventional produced water range from less than 2.00  mg/L in the Raton CBM basin to
12    approximately 200 mg/L in the Cotton Valley Group sandstones, although individual measurements
13    have exceeded 5,000 mg/L in the Marcellus Shale (see Table E-9).

14    Dissolved organic carbon (DOC) is a general indicator of organic loading and is the fraction of
15    organic carbon available for complexing with metals and supporting microbial growth. DOC  values
16    in unconventional produced water range from less than 1.50  mg/L (average) in the Raton Basin to
17    more than 115 mg/L (median) in the Marcellus Shale (see Table E-9). Individual DOC
18    concentrations in the Marcellus Shale produced water approach 6,000 mg/L. For comparison, DOC
19    levels in fresh water systems are typically below 5 mg/L, while raw wastewater can exceed
20    50 mg/L

2 1    Biochemical oxygen demand (BOD) is a conventional pollutant under the U.S. Clean Water Act It is
22    an indirect measure of biodegradable organics in produced water and an estimate of the oxygen
23    demand on a receiving water. Median BOD levels for Barnett and Marcellus Shales produced water
24    exceed 30 mg/L, and both reported maximum concentrations exceeding 12,000 mg/L (Table E-9).
25    In some circumstances wide variation in produced water median BOD levels may be reflective of
26    flowback reuse in fracturing fluids (
27    Lastly, BTEX is associated with petroleum. Benzene was found in produced water from several
28    basins: average produced water benzene concentration from the Barnett Shale was 680 [ig/L, from
29    the Marcellus Shale was 220 [ig/L (median), and from the San Juan Basin was 150 [ig/L (see Table
30    E-9). Total BTEX concentrations for conventional produced water vary widely from less than
31    100 [ig/L to nearly 580,000 [J.g/L. For comparison, average total BTEX concentrations in
32    unconventional produced water range from 20 [J.g/L in the Raton Basin to nearly 3,000 [J.g/L in the
33    Marcellus play (see Table E-9). From these data, average total BTEX levels in shale produced water
34    are one to two orders of magnitude higher than those in CBM produced water.

35    In addition to abundant BTEX, a variety of volatile and semi-volatile organic compounds VOCs and
36    SVOCs have been detected in shale and coalbed produced water. Shale produced water contains
37    naphthalene, alkylated toluenes, and methylated aromatics in the form of several benzene and
38    phenol compounds, as shown in Table E-10. Like BTEX, naphthalene, methylated phenols, and
39    acetophenone are associated with petroleum. Detected shale produced water organics such as

                   This document is a draft for review purposes only and does not constitute Agency policy.
      June 2015                                  E-27                 DRAFT— DO NOT CITE OR QUOTE

-------
    Hydraulic Fracturing Drinking Water Assessment
Appendix E
1   acetone, 2-butanone, carbon disulfide, and pyridine are potential remnants of chemical additives
2   used as friction reducers or industrial solvents (JHayesJ_2009).
    Table E-10. Reported concentrations (u.g/L) of organic constituents in produced water for two
              unconventional shale formations, presented as: average (minimum-maximum) or
              median (minimum-maximum).
Parameter
States
Acetone
Carbon disulfide
Chloroform
Isopropylbenzene
Naphthalene
Phenolic compounds
1,2,4-Trimethylbenzene
1,3,5-Trimethylbenzene
1,2-Diphenylhydrazine
1,4-Dioxane
2-Methylnaphthalene
2-Methylphenol
2,4-Dichlorophenol
2,4-Dimethylphenol
3-Methylphenol and
4-Methylphenol
Acetophenone
Benzidine
Benzo(a)anthracene
Benzo(a)pyrene
Benzo(b)fluoranthene
Benzo(g,h,i)perylene
Benzo(k)fluoranthene
Benzyl alcohol
Bis(2-Ethylhexyl) phthalate
Butyl benzyl phthalate
Barnett3
TX
145 (27-540)
-
-
35 (0.8-69)
238 (4.8-3,100)
119.65 (9.3-230)
173 (6.9-1,200)
59 (6.4-300)
4.2 (0.5-7.8)
6.5 (3.1-12)
1,362 (5.4-20,000)
28.3 (5.8-76)
(ND-15)
14.5 (8.3-21)
41 (7.8-100)
(ND-4.6)
(ND-35)
(ND-17.0)
(ND-130.0)
42.2 (0.5-84.0)
42.3 (0.7-84.0)
32.8 (0.6-65.0)
81.5 (14.0-200)
210 (4.8-490)
34.3 (1.9-110)
Marcellusb
MD, NY, OH, PA, VA, WY
83 (14-5,800)
400 (19-7,300)
28
120 (86-160)
195 (14-1,400)
-
66.5 (7.7-4,000)
33 (5.2-1,900)
-
-
3.4 (2-120)
13 (11-15)
-
12
11.5 (0.35-16)
13 (10-22)
-
-
6.7
10
6.9
5.9
41 (17-750)
20 (9.6-870)
-
                  This document is a draft for review purposes only and does not constitute Agency policy.

    June 2015                                   E-28                  DRAFT—DO NOT CITE OR QUOTE

-------
     Hydraulic Fracturing Drinking Water Assessment
Appendix E
Parameter
States
Chrysene
Di-n-octyl phthalate
Di-n-butyl phthalate
Dibenz(a,h)anthracene
Diphenylamine
Fluoranthene
Fluorene
lndeno(l,2,3-cd)pyrene
N-Nitrosodiphenylamine
N-Nitrosomethylethylamine
Phenanthrene
Phenol
Pyrene
Pyridine
Barnett3
TX
120 (0.57-240)
(ND-270)
41(1.5-120)
77 (3.2-150)
5.3 (0.6-10.0)
(ND-0.18)
0.8 (0.46-1.3)
71 (2.9-140)
8.9 (7.8-10)
(ND-410)
107 (0.52-1,400)
63 (17-93)
0.2 (ND-0.18)
413 (100-670)
Marcellusb
MD, NY, OH, PA, VA, WY
-
15
14 (11-130)
3.2 (2.3-11)
-
6.1
8.4
3.1 (2.4-9.5)
2.7
-
9.75 (3-22)
10 (2.4-21)
13
250 (10-2,600)
     -, no value available; ND, not detected.
     a Haves and Severin (2012b). n = 16. Data from days 1-23 of flowback. This data source reported concentrations without
     presentation of raw data.
     b Hayes (2009). n = 1-35. Data from days 1-90 of flowback. Concentrations were calculated from Hayes' raw data. Non-detects
     and contaminated blanks omitted.
1    The organic profile of CBM produced water is characterized by high levels of aromatic and
2    halogenated compounds compared to other unconventional produced waters [Sirivedhin and
3    Dallbauman, 2004]. PAHs and phenols are the most common organic compounds found in coalbed
4    produced water. Produced water from coalbeds in the Black Warrior Basin mainly contains
5    phenols, multiple naphthalic PAHs, and various decanoic and decenoic fatty acids (see Table E-ll).
6    CBM-associated organics are also known to include biphenyls, alkyl aromatics, hydroxypyridines,
7    aromatic amines, and nitrogen-, oxygen-, and sulfur-bearing heterocyclics [Orem etal.. 2014:
8    Pashin etal.. 2014: Benko and Drewes. 2008: Orem etal.. 2007: Fisher and Santamaria. 2002].
                   This document is a draft for review purposes only and does not constitute Agency policy.
     June 2015                                      E-29                   DRAFT—DO NOT CITE OR QUOTE

-------
     Hydraulic Fracturing Drinking Water Assessment
Appendix E
    Table E-ll. Reported concentrations of organic constituents in 65 samples of produced water
              from the Black Warrior CBM Basin, presented as average (minimum-maximum).
Parameter
States
Benzothiazole
Caprolactam
Cyclic octaatomic sulfur
Dimethyl-naphthalene
Dioctyl phthalate
Dodecanoic acid
Hexadecanoic acid
Hexadecenoic acid
Methyl-biphenyl
Methyl-naphthalene
Methyl-quinoline
Naphthalene
Octadecanoic acid
Octadecenoic acid
Phenol, 2,4-bis(l,l-dimethyl)
Phenol, 4-(l,l,3,3-tetramethyl)
Phenolic compounds
Tetradecanoic acid
Tributyl phosphate
Trimethyl-naphthalene
Triphenyl phosphate
Number of observations
-
45
10
29
39
57
30
50
25
18
52
31
49
32
29
21
17
-
53
23
23
6
Concentration (|ig/L)a
AL, MS
0.25 (0.01-3.04)
0.75 (0.02-2.39)
1.06(0.10-9.63)
0.79 (0.01-9.51)
0.21 (0.01-2.30)
1.13 (0.67-2.52)
1.58(1.17-3.02)
1.69(1.13-8.37)
0.25(0.01-2.13)
0.77 (0.01-15.55)
0.96 (0.03-3.75)
0.41 (0.01-6.57)
1.95 (1.62-3.73)
1.87 (1.60-3.47)
0.45 (0.01-4.94)
1.65 (0.01-18.34)
19.06 (ND-192.00)
1.51(0.94-5.32)
0.26 (0.01-2.66)
0.65 (0.01-4.49)
1.18(0.01-6.77)
     -, no value available.
     a          Concentrations were calculated based on the authors' raw data.
1                characterized the content of Marcellus Shale produced water including organics (see
2    Table E-10). The author tested for the majority of VOCs and SVOCs, pesticides and PCBs, based on
3    the recommendation of the Pennsylvania and West Virginia Departments of Environmental
4    Protection. Only 0.5% of VOCs and 0.03% of SVOCs in the produced water were detected above
5    1 mg/L. Approximately 96% of VOCs, 98% of SVOCs, and virtually all pesticides and PCBs were at
6    nondetectable levels.
                  This document is a draft for review purposes only and does not constitute Agency policy.
    June 2015                                    E-30                  DRAFT—DO NOT CITE OR QUOTE

-------
      Hydraulic Fracturing Drinking Water Assessment                                              Appendix E


      E.2.7.  Chemical Reactions
 1    Section E.2.7.1 describes general aspects of subsurface chemical reactions that might occur during
 2    hydraulic fracturing operations. Here we augment the discussion by describing subsurface chemical
 3    processes.

      E.2.7.1.  Injected Chemical Processes
 4    Hydraulic fracturing injects relatively oxygenated fluids into a reducing environment, which may
 5    mobilize trace or major constituents into solution. Injection of oxygenated fluids may lead to
 6    short-term changes in the subsurface redox state, as conditions may shift from reducing to
 7    oxidizing. The chemical environment in hydrocarbon-rich unconventional reservoirs, such as black
 8    shales, is generally reducing, as evidenced by the presence of pyrite and methane [Engle and
 9    Rowan, 2014: Dresel and Rose, 2010]. For black shales, reducing conditions are a product of
10    original accumulations of organic matter whose decay depleted oxygen to create rich organic
11    sediments within oil- and gas-producing formations [Tourtelot, 1979: Vine and Tourtelot, 1970].
12    Yet reactions resulting from temporary redox shifts are likely to be less important than those
13    resulting from other longer-term physical and geochemical processes. Temporary subsurface redox
14    shifts may be due to the short timeframe for fluid injection (a few days to a few weeks] and the use
15    of oxygen scavengers to prevent downhole equipment corrosion.

16    Hydraulic fracturing fluid injection introduces novel chemicals into the subsurface.* As such, the
17    geochemistry of injected and native fluids will not be in equilibrium. Over the course of days to
18    months, a complex series of reactions will equilibrate disparate fluid chemistries. The evolution of
19    flowback  and produced water geochemistry are dependent upon the exposure of formation solids
20    and fluids to novel chemicals within hydraulic fracturing fluid. Chemical additives interact with
21    reservoir solids and either mobilize constituents or themselves become adsorbed to solids. Such
22    additives  include metallic salts, elemental complexes, salts of organic acids, organometallics, and
23    other metal compounds [Montgomery. 2013: House of Representatives. 2011].

24    The salts, elemental complexes, organic acids, organometallics, and other metal-containing
25    compounds may interact with metals and metalloids in the target formation through processes
26    such  as ion exchange, adsorption, desorption, chelation, and complexation. For instance, natural
27    organic ligands (e.g., citrate] are molecules that can form coordination compounds with heavy
28    metals such as cadmium, copper, and lead [Martinez and McBride, 2001: Stumm and Morgan, 1981:
29    Bloomfieldetal.. 1976]. Citrate-bearing compounds are used in hydraulic fracturing fluids as
30    surfactants, iron control agents, and biocides. Studies of the additives' interactions with formation
31    solids at concentrations representative of hydraulic fracturing fluids are lacking.

3 2    Furthermore, pH will likely play a role in the nature and extent of these processes, as the low pH of
33    hydraulic fracturing fluids may mobilize trace constituents. The pH of injected fluids may differ
34    from existing subsurface conditions due to the use of dilute acids (e.g., hydrochloric or acetic] used
35    for cleaning perforations and  fractures during hydraulic fracturing treatments (Montgomery, 2013:
      1 For more information on chemical additive usage, refer to Chapter 5 (Chemical Mixing).

                   This document is a draft for review purposes only and does not constitute Agency policy.
      June 2015                                    E-31                  DRAFT—DO NOT CITE OR QUOTE

-------
      Hydraulic Fracturing Drinking Water Assessment                                              Appendix E


 1    GWPC and ALL Consulting, 2009]. Metals within formation solids may be released through the
 2    dissolution of acid-soluble phases such as iron and manganese oxides or hydroxides [Yangetal.,
 3    2009: Kashemetal..20Q7: Filgueiras etal.. 20021 Thus, the pH of hydraulic fracturing fluids, or
 4    changes in system pH that may occur as fluid recovery begins, may influence which metals and
 5    metalloids are likely to be retained within the formation and which may be recovered in flowback.
 6    Ultimately, more research is needed to fully understand how the injection of hydraulic fracturing
 7    fluids affects subsurface geochemistry and resultant flowback and produced water chemistry.

      E.2.8.  Microbial  Community Processes and Content
 8    By design, hydraulic fracturing releases hydrocarbons and other reduced mineral species from
 9    freshly fractured shale, sandstone, and coal, resulting in saltier in situ fluids, the release of
10    formation solids, and increased interconnected fracture networks with rich colonization surfaces
11    that are ideal for microbial growth [Wuchter et al.. 2013: Curtis. 2002]. Depending upon the
12    formation, microorganisms may be native to the subsurface and/or introduced from non-sterile
13    equipment and fracturing fluids. Additionally, microorganisms compete for novel organics in the
14    form of chemical additives [Wuchter et al., 2 013: Arthur etal., 2009]. Since large portions of
15    hydraulic fracturing fluid can remain emplaced in the targeted formation, long-term microbial
16    activity is supported through these novel carbon and energy resources [Oremetal., 2014: Murali
17    Mohan etal.. 2013a: Struchtemeyer and Elshahed. 2012: Bottero etal.. 2010]. Such physical and
18    chemical changes to the environment at depth stimulate  microbial activity and influence flowback
19    and produced water content in important ways.

20    Several studies characterizing produced water from unconventional formations (i.e., the Barnett,
21    Marcellus, Utica, and Antrim Shales] indicate that taxa with recurring physiologies compose shale
22    flowback and produced water microbial communities [Murali Mohan etal.. 2013b: Wuchter etal.,
23    2013].  Such physiologies include sulfur cyclers (e.g., sulfidogens: sulfur, sulfate, and thiosulfate
24    reducers]; fermenters; acetogens; hydrocarbon oxidizers; methanogens; and iron, manganese, and
25    nitrate reducers (Davis etal.. 2012].

26    Based on their physiologies, microorganisms cycle substrates  at depth by mobilizing or
2 7    sequestering constituents in and out of solution. Mobilization can occur through biomethylation,
28    complexation, and leaching. Sequestration can occur through intracellular sequestration,
29    precipitation, and sorption to biomass.

30    The extent to which constituents are mobilized or sequestered depends upon the prevailing
31    geochemical environment after hydraulic fracturing and through production. Significant
3 2    environmental factors that influence the extent of microbially mediated reactions are increases in
33    ionic content (i.e., salinity, conductivity, total nitrogen, bromide, iron, and potassium]; decreases in
34    acidity, and organic and inorganic carbon; the availability of diverse electron acceptors and donors;
35    and the availability of sulfur-containing compounds (Cluff etal.. 2014: Murali Mohan etal.. 2013b:
36    Davis etal.. 2012]. Examples follow that illustrate how subsurface microbial activity influences the
37    content of produced water.
                   This document is a draft for review purposes only and does not constitute Agency policy.
      June 2015                                    E-32                  DRAFT—DO NOT CITE OR QUOTE

-------
      Hydraulic Fracturing Drinking Water Assessment                                              Appendix E


 1    Under prevailing anaerobic and reducing conditions, microorganisms can mobilize or sequester
 2    metals found in unconventional produced water [Gadd. 2004]. Microbial enzymatic reduction
 3    carried out by chromium-, iron-, manganese-, and uranium-reducing bacteria can both mobilize and
 4    sequester metals (Vanengelenetal.. 2008: Garcia etal.. 2004: Mataetal.. 2002: Gauthier etal..
 5    1992: Myers and Nealson, 1988: Lovley and Phillips, 1986]. For instance, iron and manganese
 6    species go into solution when reduced, while chromium and uranium species precipitate when
 7    reduced [Gadd. 2004: Newman. 2001: Ahmannetal.. 1994].

 8    Metals can also be microbially solubilized by complexing with extracellular metabolites,
 9    siderophores (metal-chelating compounds], and microbially generated bioligands (e.g., organic
10    acids] [Glorius etal.. 2008: Francis. 2007: Gadd. 2004: Hernlemetal.. 1999].  For example,
11    Pseudomonas spp. secrete acids that act as bioligands to form complexes with uranium(VI] [Glorius
12    etal.. 20081

13    Many sulfur-cycling taxa have been found in hydraulic fracturing flowback and produced water
14    communities [Murali Mohan etal.. 2013b: Mohan etal.. 2011]. Immediately following injection,
15    microbial sulfate reduction is stimulated by diluting high-salinity formation waters with fresh
16    water (high salinities inhibit sulfate reduction]. Microbial sulfate reduction oxidizes organic matter
17    and decreases aqueous sulfate concentrations, thereby increasing the solubility of barium [Cheung
18    etal.. 2010: Lovley and Chapelle.  1995].

19    Sulfidogens also reduce sulfate, as well as elemental sulfur  and other sulfur species (e.g.,
20    thiosulfate] prevalent in the subsurface, contributing to biogenic sulfide or hydrogen sulfide gas in
21    produced water [Alain etal.. 2002: Ravotetal.. 1997]. Sulfide can also sequester metals in sulfide
22    phases [Ravotetal.. 1997: Lovley and Chapelle. 1995]. Sources of sulfide also include formation
23    solids (e.g., pyrite in shale] and remnants of drilling muds (e.g., barite and sulfonates], or other
24    electron donor sources (Davis etal.. 2012: Kim etal.. 2010: Collado etal.. 2009: Grabowskietal..
25    2005].

2 6    Additionally, anaerobic hydrocarbon oxidizers associated with shale produced water can readily
27    degrade simple and complex carbon compounds across a considerable salinity and redox range
28    (Murali Mohan et al.. 2013b: Fichter etal.. 2012: Timmis. 2010: Lalucat etal.. 2006: Yakimovetal..
29    2005: McGowan etal.. 2004: Hedlund etal.. 2001: Cayol etal.. 1994: Gauthier etal.. 1992: Zeikus et
30    al.. 1983].

31    Lastly, microbial fermentation produces organic acids, alcohols, and gases under anaerobic
32    conditions, as is the case during methanogenesis. Some nitrogen-cycling genera have been
33    identified in unconventional shale gas systems. These include genera involved in nitrate reduction
34    and denitrification fKim etal.. 2010: Yoshizawa etal.. 2010: Yoshizawa etal.. 2009: Lalucat etal..
35    2006]. These genera likely couple sugar, organic carbon, and sulfur species oxidation to nitrate
36    reduction and denitrification processes.
                   This document is a draft for review purposes only and does not constitute Agency policy.
      June 2015                                    E-33                  DRAFT—DO NOT CITE OR QUOTE

-------
      Hydraulic Fracturing Drinking Water Assessment                                             Appendix E


 1    Consequently, using a variety of recurring physiologies, microorganisms mobilize and sequester
 2    constituents in and out of solution to influence the content of flowback and produced water in
 3    important ways.

      E.3.   Produced Water Content Spatial Trends

      E.3.1. Variability between Plays of the Same Rock Type

      E.3.1.1.   Shale Formation Variability
 4    The content of shale produced water varies geographically, as shown by data from four formations
 5    (the Bakken, Barnett, Fayetteville, and Marcellus Shales; see Table E-2, Table E-4, Table E-6, Table
 6    E-9, Table E-10). For several constituents, variability between shale formations is common. The
 7    average/median TDS concentrations in the Marcellus (87,800 to 106,390 mg/L) and Bakken
 8    (196,000  mg/L) Shales are one order of magnitude greater than the average TDS concentrations
 9    reported for the Barnett and Fayetteville Shales (see Table E-2). As Fayetteville produced water
10    contains the lowest reported average TDS concentration (13,290 mg/L), average concentrations for
11    many inorganics (i.e., bromide, calcium, chloride, magnesium, sodium, and strontium) that
12    contribute to dissolved solids loads are the lowest compared to average concentrations for the
13    same inorganics in Bakken, Barnett, and Marcellus produced water (see Table E-4 and Table E-6).
14    Average concentrations for metals reported within Bakken and Marcellus produced water are also
15    higher than those within the Barnett or Fayetteville formations (see Table E-6).

16    Additionally, Marcellus produced water is enriched in barium (average concentration of 2,224 mg/1
17    in Barbotetal. (2013) or median calculated from Hayes (2009) of 542.5 mg/L) and strontium
18    (average concentration of 1,695 mg/L (Barbotetal.. 2013) or median calculated from Hayes
19    f20091of 1,240 mg/L) by one to three orders of magnitude compared to Bakken, Barnett, and
20    Fayetteville produced water (see Table E-6). Subsequently, radionuclide variability expressed as
21    isotopic ratios (e.g., radium-228/radium-226, strontium-87/strontium-86) are being used to
22    determine the reservoir source for produced water (Chapman etal... 2012: Rowan etal... 2011:
23    Blauch et al.. 2009). Lastly, Barnett and Bakken produced waters are enriched in sulfate.

24    Although  organic data are limited, average BTEX concentrations are higher in Marcellus compared
25    to Barnett produced water by one order of magnitude, whereas concentrations of benzene alone
26    are marginally higher in  Barnett compared to Marcellus produced water (see Table E-9 and Table
27    E-10).

      E.3.1.2.   Tight Formation Variability
2 8    The average concentrations for various constituents in tight formation produced water vary
29    geographically between sandstone formations (the Cotton Valley Group, Devonian sandstone, and
30    the Mesaverde and Oswego), as shown in Table E-2, Table E-4, and Table E-6. The average TDS
31    concentrations in the Devonian sandstone (235,125 mg/L) and Cotton Valley Group
32    (164,683  mg/L) are one to two orders of magnitude greater than the  average TDS concentrations
33    reported for the Mesaverde (15,802 mg/L) and Oswego Formations (73,082 mg/L) (see Table E-2).


                  This document is a draft for review purposes only and does not constitute Agency policy.
      June 2015                                  E-34                  DRAFT—DO NOT CITE OR QUOTE

-------
      Hydraulic Fracturing Drinking Water Assessment                                                Appendix E


 1    Mesaverde produced water also contained the lowest average concentrations for many of the
 2    inorganic components of TDS (i.e., calcium, chloride, iron, magnesium, and sodium; see Table E-4
 3    and Table E-6).

 4    Little variability was reported in pH between these four tight formations (see Table E-2).
 5    Mesaverde produced water was enriched in sulfate, with an average concentration of 83 7 mg/L
 6    (see Table E-4), whereas Devonian produced water was enriched in barium, which had an average
 7    concentration of 1,488 mg/L (see Table E-6).

      E.3.1.3.  Coalbed Variability
 8    Geochemical analysis showed that the Powder River Basin is predominately characterized by
 9    bicarbonate water types with a large intrusion of sodium-type waters across a large range of
10    magnesium and calcium concentrations (Dahmetal.. 2011).* In contrast, the Raton Basin is typified
11    by sodium-type waters with low calcium and magnesium concentrations. A combination of Powder
12    River and Raton produced water compositional characteristics typifies the San Juan Basin (Dahm et
13    al.. 2011). Lastly, Black Warrior Basin produced water is differentiated based upon its sodium
14    bicarbonate- or sodium chloride-type waters (DOE, 2014: Pashinetal., 2014).

15    Regional variability is observed in average produced water concentrations for various constituents
16    of four CBM basins (Powder River, Raton, San Juan, and Black Warrior; see Table E-3, Table E-5,
17    Table E-7, Table E-9, and Table E-ll), but particularly between produced water of the Black
18    Warrior Basin and the others. As the average TDS concentration in Black Warrior Basin produced
19    water (14,319 mg/L) is one to two orders of magnitude higher than that of the other three
20    presented in Table E-3, average concentrations for TDS contributing ions (i.e., calcium, chloride,
21    and sodium) were also higher than in the Powder River, Raton, and San Juan Basins. These high
22    levels follow from the marine depositional environment of the Black Warrior Basin (Horsey. 1981).

23    Powder River Basin produced water has the lowest average TDS concentration (997 mg/L), which
24    is consistent with Dahmetal. (2011) reporting that nearly a quarter of all the produced water
25    sampled  from the Powder River Basin meets the U.S. drinking water secondary standard for TDS
26    (less than 500 mg/L).2 In addition, the Black Warrior Basin appears to be slightly enriched in
27    barium, compared to the other three CBM basins  (see Table E-5). Lastly, the three western CBM
28    basins  (Powder River, Raton, and San Juan) are much more alkaline and enriched in bicarbonate
29    than their eastern counterpart (the Black Warrior Basin; see Table E-3).
      1 Water is classified as a "type" if the dominant dissolved ion is greater than 50% of the total. A sodium-type water
      contains more that 50% of the cation milliequivalents (mEq) as sodium. Similarly, a sodium-bicarbonate water contains
      50% of the cation mEq as sodium, and 50% of the anion mEq as bicarbonate (USGS. 2002).
      2 MCL refers to the highest level of a contaminant that is allowed in drinking water. MCLs are enforceable standards.
      These include primary MCLs for barium, cadmium, chromium, lead, mercury, and selenium. National Secondary Drinking
      Water Regulations (NSDWRs or secondary standards) are non-enforceable guidelines regulating contaminants that may
      cause cosmetic effects (such as skin or tooth discoloration) or aesthetic effects (such as taste, odor, or color) in drinking
      water. Secondary MCLs are recommended for aluminum, chloride, copper, iron, manganese, pH, silver, sulfate, TDS, and
      others. See http://water.epa.gov/drink/contaminants/index.cfmtfPrimary for more information.

                    This document is a draft for review purposes only and does not constitute Agency policy.
      June 2015                                    E-35                  DRAFT—DO NOT CITE OR QUOTE

-------
      Hydraulic Fracturing Drinking Water Assessment                                             Appendix E


 1    Average concentrations of benzene, ethylbenzene, and xylenes are higher in San Juan compared to
 2    Raton produced water by two orders of magnitude, whereas concentrations of toluene are
 3    marginally higher in Raton compared to San Juan produced water (see Table E-9).

      E.3.2.  Local Variability
 4    Spatial variability of produced water content frequently exists within a single producing formation.
 5    For instance, Marcellus Shale barium levels increase along a southwest to northeast transect
 6    [Barbotetal.. 2013]. Additionally, produced water from the northern and southern portions of the
 7    San Juan Basin differ in TDS, due to ground water recharge in the northern basin leading to higher
 8    chloride concentrations than in the southern portion [Dahm etal.. 2011: Van Voast. 2003).

 9    Spatial variability of produced water content also exists at a local level due to the stratigraphy
10    surrounding the producing formation. For example, deep saline aquifers, if present in the over- or
11    underlying strata, may over geologic time encroach upon shales, coals, and sandstones via fluid
12    intrusion processes [Blauch etal.. 2009). Evidence of deep brine migration from adjacent strata into
13    shallow aquifers via natural faults and fractures has been noted previously in the Michigan Basin
14    and the Marcellus Shale [Vengosh etal.. 2014: Warner etal.. 2012: Weaver etal.. 1995]. By
15    extension, in situ hydraulic connectivity, which is stimulated by design during hydraulic fracturing,
16    may lead to the migration of brine-associated constituents in under- and overlying strata into
17    producing formations, as discussed in Chapter 6.

18    As hydrocarbon source rocks often form repeating sedimentary sequences, contact between these
19    layers presents opportunities for an exchange of organics and inorganics [Fredrickson and Balkwill.
20    2006: U.S. EPA. 2004). For instance, diffusion of carbon sources and electron donors occurs at
21    subsurface shale-sandstone interfaces,  suggesting a stratigraphic role in the exchange of
22    constituents between formations [Fredrickson and Balkwill, 2006].

      E.4.    Example Calculation for Roadway Transport
2 3    This section provides background information for the roadway transport calculation appearing in
24    Chapter 7.

      E.4.1.  Estimation of Transport Distance
25    In a study of wastewater management for the Marcellus Shale, Rahm etal. [2013] used data
26    reported to the Pennsylvania Department of Environmental Protection (PA DEP] to estimate the
27    average distance wastewater was transported. For the period from 2008 to 2010, the distance
28    transported was approximately 100 km, but it was reduced by 30% for 2011. The reduction was
29    attributed to increased treatment infrastructure in Lycoming County, an area of intensive  hydraulic
30    fracturing operations in northeastern Pennsylvania. For the part of Pennsylvania within the
31    Susquehanna River Basin, Gilmore etal. [2013] estimated the likely transport distances for drilling
32    waste to landfills (256 km or 159 mi]; produced water to disposal wells (388 km or 241 mi]; and
33    commercial wastewater treatment plants (CWTPs] (158 km or 98 mi]. These distances are longer
                   This document is a draft for review purposes only and does not constitute Agency policy.
      June 2015                                   E-36                 DRAFT—DO NOT CITE OR QUOTE

-------
      Hydraulic Fracturing Drinking Water Assessment
Appendix E
 1    than the values from Rahm_et_aL_[2M3}.. in part, because wells in the Susquehanna Basin are
 2    further to the east of Ohio disposal wells and some CWTPs.

      E.4.2.  Estimation of Wastewater Volumes
 3    In an example water balance calculation, Gilrnore_etaLX2013} used 380,000 gal of flowback as the
 4    volume transported to CWTPs, 450,000 gal of flowback transported to injection wells, and 130,000
 5    gal of un-reusable treated water also transported to injection wells for a total estimated wastewater
 6    volume of 960,000 gal per well.

      E.4.3.  Estimation of Roadway Accidents
 7    The U.S. Department of Transportation (DOT) published statistics on roadway accidents (UJL
 8    De^artmejit_of_Transportation1_2^12.] which indicate that the combined total of combination truck
 9    crashes in 2012 was 179,736, or 110 per 100 million vehicle miles (1.77 million km) (see Table
10    E-12). As an indicator of the uncertainty of these data, DOT reported 122,240 large truck crashes
11    from a differing set of databases (see Table E-13), with a rate of 75 per 100 million vehicle miles,
12    which is 68% of the number of combination truck crashes.
      Table E-12. Combination truck crashes in 2012 for the 2,469,094 registered combination
               trucks, which traveled 163,458 million miles (U.S. Department of Transportation,
               2012).a
Type of crash
Property damage only
Injury
Fatal
Total
Combination trucks
involved in crashes
135,000
42,000
2,736
179,736
Rates per 100 million vehicle miles
traveled by combination trucks
82.8
25.5
1.74
110
      3 A combination truck is defined as a truck tractor pulling any number of trailers (Uj

                   This document is a draft for review purposes only and does not constitute Agency policy.
      June 2015                                    E-37                  DRAFT—DO NOT CITE OR QUOTE

-------
      Hydraulic Fracturing Drinking Water Assessment
                                                                      Appendix E
      Table E-13. Large truck crashes in 2012 (U.S. Department of Transportation, 2012).a
Type of crash
Towaway crashes
Injury
Fatal
Totals
Total crashes
72,644
45,794
3,802
122,240
Large trucks with cargo tanks
Number
4,364
3,245
360
7,969
Percentage
6.0%
7.1%
9.5%
6.5%
      a A large truck is defined as a truck with a gross vehicle weight rating greater than 10,000 pounds (U.S. Department of
      Transportation, 2012).
      E.4.4.  Estimation of Material Release Rates in Crashes
 1    Estimates ranging from 5.6% to 36% have been made for the probability of material releases from
 2    crashed trucks. Craft [2004] used data from three databases to estimate the probability of spills in
 3    fatality accidents at 3 6%, which may overestimate the probability for all types of accidents [Rozell
 4    and Reaven, 2012].l The U.S. Department of Transportation [2012] provides estimates of
 5    hazardous materials releases from large truck crashes. For all types of hazardous materials carried,
 6    408 of 2,903 crashes, or 14%, were known to have hazardous materials releases. The occurrence of
 7    a release was unknown for 18% of the crashes. These crashes were not distinguished by truck type,
 8    so they likely overestimated the number of tanker crashes. Harwoodetal. [1993] used accident
 9    data from three states (California, Illinois, and Michigan] to develop hazardous materials release
10    rate estimates for different types of roadways, accidents, and settings (urban or rural]. For
11    roadways in rural settings the probability of release ranged from 8.1% to 9.0%, while in urban
12    settings the probability ranged from 5.6%  to 6.9%.

      E.4.5.  Estimation of Volume Released in Accidents
13    Based on the estimated volume (960,000 gal (3.63 million L] per well] and disposal distances used
14    byRahmetal. [2013] and Gilmore etal. [2013], and an assumed 20,000 L (5,300 gal]-containing
15    truck [Gilmore etal.. 2013]. the total travel distance by trucks ranges from 9,620 miles (14,900 km]
16    to 17,760 miles (28,570 km] per well (see Table E-14].
      1 The three databases were the Trucks Involved in Fatal Accidents developed by the Center for National Truck Statistics at
      the University of Michigan, the National Automotive Sampling System's General Estimates System (GES) produced by the
      National Highway Transportation Safety Agency, and the Motor Carrier Management Information System (MCMIS) Crash
      File produced by the Federal Motor Carrier Safety Administration.
      June 2015
This document is a draft for review purposes only and does not constitute Agency policy.
                               E-38                   DRAFT—DO NOT CITE OR QUOTE

-------
      Hydraulic Fracturing Drinking Water Assessment
Appendix E
      Table E-14. Estimate of total truck-travel miles per well in the Susquehanna River Basin based
               on the transport analysis performed by Gilmore et al. (2013)1
Action
Waste per well
(million gal)
Gilmore et al. (2013) distance estim

Produced water
to CWTP
Produced water
to disposal well
CWTP effluent to
disposal well
Total
Rahm et al. (2013)

Transport 100 km
Transport 70 km
0.38
0.45
0.13
0.96
Trucks
(20 mVtruck)
Miles
traveled
per truck
Total miles
traveled
(per well)
Material release rate bounds
5.6%
36%
Crashes per 100 million miles
75
110
75
110
ates
72
85
25
182
26.9
147
133

1,937
12,495
3,325
17,757



3



4



18



27
distance estimates
0.96
0.96
182
182
62.1
43.5
11,300
9,620
2
1
3
2
12
8
17
12
 1    The Susquehanna River Basin Commission reported 1,928 well pads permitted within the basin
 2    [SRBC. 2012]. Assuming two wells per pad, the total distance traveled to haul hydraulic fracturing
 3    wastewater is 68.4 million miles (110 million km).

 4    Combining these data with the DOT crash data gives an estimated 76 crashes per year using the
 5    combination truck crash rate or 52 per year using the DOT large truck crash rate. Based on the
 6    various assumptions of travel distances, crash rates, and estimated minimum and maximum
 7    material release rates, the number of crashes with releases ranges from 1 to 27 (see Table E-14).

 8    Several limitations are inherent in this analysis, including differing rural road accident rates and
 9    highway rates, differing wastewater endpoints, and differing amounts of produced water transport
10    Further, the estimates present an upper bound on impacts, because not all releases of wastewater
11    would reach or impact drinking water resources.

      E.5.   References for Appendix E
      Acharya, HR: Henderson, C: Matis, H: Kommepalli, H: Moore, B: Wang, H. (2011). Cost effective recovery of
        low-TDS frac flowback water for reuse. (Department of Energy: DE-FE0000784). Niskayuna, NY: GE Global
        Research. http://www.netl.doe.gov/file%201ibrarv/Research/oil-gas/FE0000784 FinalReportpdf
                   This document is a draft for review purposes only and does not constitute Agency policy.
      June 2015                                    E-39                  DRAFT—DO NOT CITE OR QUOTE

-------
Hydraulic Fracturing Drinking Water Assessment                                                  Appendix E
Ahmann. D: Roberts. AL: Krumholz. LR: Morel. FM. (1994). Microbe grows by reducing arsenic [Letter].
   Nature 371: 750. http://dx.doi.org/10.1038/371750aO

Alain. K: Pignet. P: Zbinden. M: Quillevere. M: Duchiron. F: Donval. IP: Lesongeur. F: Raguenes. G: Crassous. P:
   Querellou. I: Cambon-Bonavita. MA. (2002). Caminicella sporogenes gen. nov., sp. nov., a novel
   thermophilic spore-forming bacterium isolated from an East-Pacific Rise hydrothermal vent. Int J Syst
   Evol Microbiol 52:1621-1628.

Andre, L: Rabemanana, V: Vuataz, FD. (2006). Influence of water-rock interactions on fracture permeability of
   the deep reservoir at Soultz-sous-Forets, France. Geothermics 35: 507-531.
   http://dx.doi.0rg/10.1016/j.geothermics.2006.09.006

Arthur. ID: Bohm. B: Cornue. D. (2009). Environmental considerations of modern shale gas development.
   Paper presented at SPE Annual Technical Conference and Exhibition, October 4-7, 2009, New Orleans, LA.

Bank. T. (2011). Trace metal geochemistry and mobility in the Marcellus shale. In Proceedings of the
   Technical Workshops for the Hydraulic Fracturing Study: Chemical & Analytical Methods. Bank, T.
   http://www2.epa.gov/sites/production/files/documents/tracemetalgeochemistryandmobilityinthemarc
   ellusformationl.pdf

Bank. T: Fortson. LA: Malizia. TR: Benelli. P. (2012). Trace metal occurrences in the Marcellus Shale
   [Abstract]. Geological Society of America Abstracts with Programs 44: 313.

Barbot. E: Vidic. NS: Gregory. KB: Vidic. RD.  (2013). Spatial and temporal correlation of water quality
   parameters of produced waters from Devonian-age shale following hydraulic fracturing. Environ Sci
   Technol 47: 2562-2569.

Benko. KL: Drewes. IE. (2008). Produced water in the Western United States: Geographical distribution,
   occurrence, and composition. Environ Eng Sci 25: 239-246.

Blauch. ME: Myers. RR: Moore. TR: Lipinski. BA. (2009). Marcellus shale post-frac flowback waters - where is
   all the salt coming from and what are the implications? In Proceedings of the SPE Eastern Regional
   Meeting. Richardson, TX: Society of Petroleum Engineers.

Blondes. MS: Cans. KD: Thordsen. II: Reidy. ME: Thomas. B: Engle.  MA: Kharaka. YK: Rowan. EL. (2014). Data:
   U.S. Geological Survey National Produced Waters Geochemical  Database v2.0  (Provisional) [Database]:
   U.S. Geological Survey:: USGS. Retrieved from
   http://energy.usgs.gov/EnvironmentalAspects/EnvironmentalAspectsofEnergyProductionandUse/Produ
   cedWaters.aspx#3822349-data

Bloomfield. C: Kelson. W: Pruden. G.  (1976). Reactions between metals and humidified organic matter.
   Journal of Soil Science 27:16-31. http://dx.doi.Org/10.llll/j.1365-2389.1976.tb01971.x

Bottero. S: Picioreanu. C: Delft. TU: Enzien. M: van Loosdrecht. MCM: Bruining. H: Heimovaara. T. (2010).
   Formation damage and impact on gas flow caused by biofilms growing within proppant packing used in
   hydraulic fracturing. Paper presented at SPE International Symposium and Exhibiton on Formation
   Damage Control, February 10-12, 2010,  Lafayette, Louisiana.

BTGS (Bureau of Topographic and Geologic Survey). (2011). Geochemical analyses of selected lithologies
   from geologic units in central, north-central, and southeastern  Pennsylvania. (OFMI 1101.0). Middletown,
   PA: Bureau of Topographic and Geologic Survey, Pennsylvania  Geological Survey.

Cayol, IL: Ollivier, B: Lawson anani soh, A: Fardeau, ML: Ageron, E: Grimont, PAD: Prensier, G: Guezennec, I:
   Magot. M: Garcia. IL. (1994). Haloincola saccharolytica subsp. senegalensis subsp. nov., Isolated from the
   sediments of a Hypersaline lake, and emended description of Haloincola saccharolytica. International
   Journal of Systematic Bacteriology 44: 805-811. http://dx.doi.org/10.1099/00207713-44-4-805

Chapman. EC: Capo. RC: Stewart. BW: Kirby. CS: Hammack. RW: Schroeder. KT: Edenborn. HM. (2012).
   Geochemical and strontium isotope characterization of produced waters from Marcellus Shale natural gas
   extraction. Environ Sci Technol 46: 3545-3553.
               This document is a draft for review purposes only and does not constitute Agency policy.

June 2015                                       E-40                    DRAFT—DO NOT CITE OR QUOTE

-------
Hydraulic Fracturing Drinking Water Assessment                                                  Appendix E
Chermak. IA: Schreiber. ME. (2014). Mineralogy and trace element geochemistry of gas shales in the United
   States: Environmental implications. Int J Coal Geol 126: 32-44.
   http://dx.doi.0rg/10.1016/i.coal.2013.12.005

Cheung. K: Klassen. P: Mayer. B: Goodarzi. F: Aravena. R. (2010). Major ion and isotope geochemistry of fluids
   and gases from coalbed methane and shallow groundwater wells in Alberta, Canada. Appl Geochem 25:
   1307-1329. http://dx.doi.0rg/10.1016/i.apgeochem.2010.06.002

Choppin, GR. (2006). Actinide speciation in aquatic systems. Mar Chem 99: 83-92.
   http://dx.doi.0rg/10.1016/j.marchem.2005.003.011

Choppin, GR. (2007). Actinide speciation in the environment. Journal of Radioanal Chem 273: 695-703.
   http://dx.doi.org/10.1007/sl0967-007-0933-3

Clark. CE: Veil. IA. (2009). Produced water volumes and management practices in the United States (pp. 64).
   (ANL/EVS/R-09/1). Argonne, IL: Argonne National Laboratory.
   http://www.circleofblue.org/waternews/wp-
   content/uploads/2010/09/ANL EVS R09 produced water volume report 2437.pdf

Cluff. M: Hartsock. A: Macrae. I: Carter. K: Mouser. PI. (2014). Temporal changes in microbial ecology and
   geochemistry in produced water from hydraulically fractured Marcellus Shale Gas Wells. Environ Sci
   Technol 48: 6508-6517. http://dx.doi.org/10.1021/es501173p

Collado. L: Cleenwerck. I: Van Trappen. S: De Vos. P: Figueras. Ml. (2009). Arcobacter mytili sp. nov., an
   indoxyl acetate-hydrolysis-negative bacterium isolated from mussels. Int J Syst Evol Microbiol 59:1391-
   1396. http://dx.doi.0rg/10.1099/ijs.0.003749-0

Craft. R. (2004). Crashes involving trucks carrying hazardous materials. (FMCSA-RI-04-024). Washington,
   D.C.: U.S. Department of Transportation, http://ntl.bts.gov/lib/51000/51300/51302/fmcsa-ri-04-024.pdf

Curtis. IB. (2002). Fractured shale-gas systems. AAPG Bulletin 86:1921-1938.
   http://dx.doi.org/10.1306/61EEDDBE-173E-llD7-8645000102C1865D

Dahm. KG: Guerra. KL: Xu.  P:  Drewes. IE. (2011). Composite geochemical database for coalbed methane
   produced water quality in the Rocky Mountain region. Environ Sci Technol 45: 7655-7663.
   http://dx.doi.org/10.1021/es201021n

Davis. IP: Struchtemeyer. CG: Elshahed. MS. (2012). Bacterial communities associated with production
   facilities of two newly drilled thermogenic natural gas wells in the Barnett Shale (Texas, USA). Microb Ecol
   64: 942-954. http://dx.doi.org/10.1007/s00248-012-0073-3

Diehl. SF: Goldhaber. MB: Hatch. JR. (2004). Modes of occurrence of mercury and other trace elements in coals
   from the warrior field, Black Warrior Basin, Northwestern Alabama. Int] Coal Geol 59:193-208.
   http://dx.doi.0rg/10.1016/i.coal.2004.02.003

DOE (U.S. Department of Energy). (2014). Water management strategies for improved coalbed methane
   production in the Black Warrior Basin. Available online at http://www.netl.doe.gov/research/oil-and-
   gas/project-summaries/natural-gas-resources/de-fe0000888

Dresel. PE: Rose. AW.  (2010). Chemistry and origin of oil  and gas well brines in western Pennsylvania (pp.
   48). (Open-File Report  OFOG 1001.0). Harrisburg, PA: Pennsylvania Geological Survey, 4th ser.
   http://www.marcellus.psu.edu/resources/PDFs/brines.pdf

Engle. MA: Rowan. EL. (2014). Geochemical evolution of produced waters from hydraulic fracturing of the
   Marcellus Shale, northern Appalachian Basin: A multivariate compositional data analysis approach. Int J
   Coal Geol 126: 45-56. http://dx.doi.Org/10.1016/j.coal.2013.ll.010

Fertl. WH: Chilingar. GV. (1988). Total organic carbon content determined from well logs. SPE Formation
   Evaluation 3: 407-419.  http://dx.doi.org/10.2118/15612-PA
               This document is a draft for review purposes only and does not constitute Agency policy.

June 2015                                       E-41                    DRAFT—DO NOT CITE OR QUOTE

-------
Hydraulic Fracturing Drinking Water Assessment                                                  Appendix E
Fichter. I: Moore. R: Braman. S: Wunch. K: Summer. E: Holmes. P. (2012). How hot is too hot for bacteria? A
   technical study assessing bacterial establishment in downhole drilling, fracturing, and stimulation
   operations. Paper presented at NACE International Corrosion Conference & Expo, March 11-15,2012, Salt
   Lake City, UT.

Filgueiras. AV: Lavilla. I: Bendicho. C. (2002). Chemical sequential extraction for metal partitioning in
   environmental solid samples. J Environ Monit 4: 823-857. http://dx.doi.org/10.1039/b207574c

Fisher, IG: Santamaria, A. (2002). Dissolved organic constituents in coal-associated waters and implications
   for human and ecosystem health. Paper presented at 9th Annual International Petroleum Environmental
   Conference, October 22-25, 2002, Albuquerque, NM.

Fisher. RS. (1998). Geologic and geochemical controls on naturally occurring radioactive materials (NORM) in
   produced water from oil, gas, and geothermal operations. Environmental Geosciences 5:139-150.
   http://dx.doi.0rg/10.1046/j.1526-0984.1998.08018.x

Francis. Al. (2007).  Microbial mobilization and immobilization of plutonium. J Alloy Comp 444: 500-505.
   http://dx.doi.0rg/10.1016/i.iallcom.2007.01.132

Fredrickson. IK: Balkwill. PL. (2006). Geomicrobial processes and biodiversity in the deep terrestrial
   subsurface. Geomicrobiologyjournal23: 345-356. http://dx.doi.org/10.1080/01490450600875571

Gadd. GM. (2004). Microbial influence on metal mobility and application for bioremediation. Geoderma 122:
   109-119. http://dx.doi.0rg/10.1016/j.geoderma.2004.01.002

Garcia. MT: Mellado. E: Ostos. 1C: Ventosa. A. (2004). Halomonas organivorans sp. nov., a moderate halophile
   able to degrade aromatic compounds. Int J Syst Evol Microbiol 54:1723-1728.
   http://dx.doi.0rg/10.1099/iis.0.63114-0

Gauthier. Ml: Lafay. B: Christen. R: Fernandez. L: Acquaviva. M: Bonin. P: Bertrand. 1C. (1992). Marinobacter
   hydrocarbonoclasticus gen. nov., sp. nov., a New, Extremely Halotolerant, Hydrocarbon-Degrading Marine
   Bacterium. International Journal of Systematic Bacteriology 42: 568-576.
   http://dx.doi.org/10.1099/00207713-42-4-568

Gilmore. K: Hupp. R: Glathar. I.  (2013). Transport of Hydraulic Fracturing Water and Wastes in the
   Susquehanna River Basin, Pennsylvania. J Environ Eng 140: B4013002.
   http://dx.doi.org/10.1061/rASCE1EE.1943-7870.0000810

Glorius, M: Moll, H: Geipel, G: Bernhard, G. (2008). Complexation of uranium(VI) with aromatic acids such as
   hydroxamic and benzoic acid investigated by TRLFS. Journal of Radioanal Chem 277: 371-377.
   http://dx.doi.org/10.1007/sl0967-007-7082-6

Grabowski. A: Nercessian. 0: Fayolle. F: Blanchet. D: leanthon. C. (2005). Microbial diversity in production
   waters of a low-temperature biodegraded oil reservoir. FEMS Microbiol Ecol 54: 427-443.
   http://dx.doi.0rg/10.1016/j.femsec.2005.05.007

GWPC and ALL Consulting (Ground Water Protection Council (GWPC) and ALL Consulting). (2009). Modern
   shale gas development in the United States: A primer. (DE-FG26-04NT15455). Washington, DC: U.S.
   Department of Energy, Office of Fossil Energy and National Energy Technology Laboratory.
   http://www.gwpc.org/sites/default/files/Shale%20Gas%20Primer%202009.pdf

Haluszczak. LO: Rose. AW: Kump. LR. (2013). Geochemical evaluation of flowbackbrine from Marcellus gas
   wells in Pennsylvania, USA. Appl Geochem 28: 55-61.
   http://dx.doi.0rg/10.1016/j.apgeochem.2012.10.002

Harwood, DW: Viner, IG: Russell, ER. (1993). Procedure for developing truck accident and release rates for
   hazmat routing. Journal of Transportation Engineering 119:189-199.
   http://dx.doi.org/10.1061/(ASCE)0733-947X(1993)119:2(189)
               This document is a draft for review purposes only and does not constitute Agency policy.

June 2015                                       E-42                    DRAFT—DO NOT CITE OR QUOTE

-------
Hydraulic Fracturing Drinking Water Assessment                                                 Appendix E
Hatch. IR: Leventhal. IS. (1981). Geochemistry of organic-rich shales and coals from middle Pennsylvanian
   Cherokee group and lower part of Marmaton group, Oklahoma, Kansas, Missouri, and Iowa [Abstract].
   AAPG Bulletin 65: 936.

Hayes. T. (2009). Sampling and analysis of water streams associated with the development of Marcellus shale
   gas. Des Plaines, IL: Marcellus Shale Coalition, http://eidmarcellus.org/wp-
   content/ uploads/2 012/11/MSCommission-Report.pdf

Hayes, T: Severin, BF. (2012b). Evaluation of the aqua-pure mechanical vapor recompression system in the
   treatment of shale gas flowback water - Barnett and Appalachian shale water management and reuse
   technologies. (08122-05.11). Hayes, T; Severin, BF. http://barnettshalewater.org/documents/08122-
   05.11-EvaluationofMVR-3-12-2012.pdf

Hedlund. BP: Geiselbrecht. AD: Staley. IT.  (2001). Marinobacter strain NCE312 has a Pseudomonas-like
   naphthalene dioxygenase. FEMS Microbiol Lett 201: 47-51.

Hernlem. Bl: Vane. LM: Sayles. GD. (1999). The application of siderophores for metal recovery and waste
   remediation: Examination of correlations for prediction of metal affinities. Water Res 33: 951-960.

Horsey. CA. (1981). Depositional environments of the Pennsylvanian Pottsville Formation in the Black
   Warrior Basin of Alabama. Journal of Sedimentary Research 51: 799-806.
   http://dx.doi.org/10.1306/212F7DB5-2B24-llD7-8648000102C1865D

House of Representatives (U.S. House of Representatives).  (2011). Chemicals used in hydraulic fracturing.
   Washington, D.C.: U.S. House  of Representatives, Committee on Energy and Commerce, Minority Staff.
   http://democrats.energycommerce.house.gov/sites/default/files/documents/Hydraulic-Fracturing-
   Chemicals-2011-4-18.pdf

Kashem, MA: Singh, BR: Kondo, T: Huq, SMI: Kawai, S. (2007). Comparison of extractability of Cd, Cu, Pb and
   Zn with sequential extraction in contaminated and non-contaminated soils. Int J Environ Sci Tech 4: 169-
   176. http://dx.doi.org/10.1007/BF03326270

Katsoyiannis. A: Samara. C. (2007). The fate of dissolved organic carbon (DOC) in the wastewater treatment
   process and its importance in the removal of wastewater contaminants. Environ Sci Pollut Res Int 14: 284-
   292. http://dx.doi.org/10.1065/espr2006.05.302

Kim. HM: Hwang. CY: Cho. BC. (2010). Arcobacter marinus sp. nov. Int J Syst Evol Microbiol 60: 531-536.
   http://dx.doi.0rg/10.1099/iis.0.007740-0

Kraemer. TF: Reid. DF. (1984). The occurrence and behavior of radium in saline formation water of the U.S.
   Gulf Coast region. Isotope Geoscience 2:153-174.

Lalucat. I: Bennasar. A: Bosch. R: Garcia-Valdes. E: Palleroni. Ml. (2006). Biology of Pseudomonas stutzeri
   [Review]. Microbiol Mol Biol  Rev 70: 510-547. http://dx.doi.org/10.1128/MMBR.00047-05

Langmuir, D: Herman, IS. (1980). The mobility of thorium in natural waters at low temperatures. Geochim
   Cosmo Act 44: 1753-1766. http://dx.doi.org/10.1016/0016-7037(80)90226-4

Langmuir. D: Riese. AC. (1985). THE THERMODYNAMIC PROPERTIES OF RADIUM. Geochim Cosmo Act 49:
   1593-1601.

Lee. K: Neff. I. (2011). Produced water: Environmental risks and advances in mitigation technologies. New
   York, NY: Springer, http://dx.doi.org/10.1007/978-l-4614-0046-2

Leventhal. IS: Hosterman. IW. (1982). Chemical and mineralogical analysis Of Devonian black-shale samples
   from Martin County, Kentucky - Caroll and Washington counties, Ohio - Wise County, Virginia - and
   Overton County, Tennessee, USA. Chem Geol 37: 239-264.

Li. L: Brantley. SL. (2011). Development of a subsurface reactive transport model for predicting potential
   water quality problems at Marcellus shale. (USGS Project 2011PA159B). Li, L; Brantley, SL.
   http://water.usgs.gov/wrri/2011grants/progress/2011PA159B.pdf
               This document is a draft for review purposes only and does not constitute Agency policy.

June 2015                                      E-43                   DRAFT—DO NOT CITE OR QUOTE

-------
Hydraulic Fracturing Drinking Water Assessment                                                  Appendix E
Lovley. PR: Chapelle. FH. (1995). Deep subsurface microbial processes. Rev Geophys 33: 365-381.
   http://dx.doi.org/10.1029/95RG01305

Lovley. PR: Phillips. El. (1986). Organic matter mineralization with reduction of ferric iron in anaerobic
   sediments. Appl Environ Microbiol 51: 683-689.

Martinez. CE: McBride. MB. (2001). Cd, Cu, Pb, and Zn coprecipitates in Fe oxide formed at different pH: Aging
   effects on metal solubility and extractability by citrate. Environ Toxicol Chem 20:122-126.
   http://dx.doi.org/10.1002/etc.5620200112

Mata. I A: Martfnez-Canovas. I: Ouesada. E: Bejar. V. (2002). A detailed phenotypic characterisation of the type
   strains of Halomonas species. Syst Appl Microbiol 25: 360-375. http://dx.doi.org/10.1078/0723-2020-
   00122

Maxwell. SC. (2011). Hydraulic fracture height growth. Recorder 36: 18-22.

McGowan. L: Herbert. R: Muyzer. G. (2004). A comparative study of hydrocarbon degradation by
   Marinobacter sp., Rhodococcus sp. and Corynebacterium sp. isolated from different mat systems. Ophelia
   58: 271-281. http://dx.doi.org/10.1080/00785236.2004.10410235

McLin. K: Brinton. D: Moore. I. (2011). Geochemical modeling of water-rock-proppant interactions. Thirty-
   Sixth Workshop on Geothermal Reservoir Engineering, January 31 - February 2,2011, Stanford University,
   Stanford, California.

Mohan. AM: Gregory. KB: Vidic. RD: Miller. P: Hammack. RW. (2011). Characterization of microbial diversity
   in treated and untreated flowback water impoundments from gas fracturing operations. Paper presented
   atSPE Annual Technical Conference and Exhibition, October 30 - November 2, 2011, Denver, CO.

Montgomery, C. (2013). Fracturing fluid components. In A Bunder; J McLennon; R Jeffrey (Eds.), Effective and
   Sustainable Hydraulic Fracturing. Croatia: InTech. http://dx.doi.org/10.5772/56422

Moses. CO: Herman. IS. (1991). Pyrite oxidation at circumneutral pH. Geochim Cosmo Act 55: 471-482.

Murali Mohan. A: Hartsock. A: Bibby. Kl: Hammack. RW: Vidic. RD: Gregory. KB. (2013a). Microbial
   community changes in hydraulic fracturing fluids and produced water from shale gas extraction. Environ
   Sci Technol 47:13141-13150. http://dx.doi.org/10.1021/es402928b

Murali Mohan. A: Hartsock. A: Hammack. RW: Vidic. RD: Gregory. KB. (2013b). Microbial communities in
   flowback water impoundments from hydraulic fracturing for recovery of shale gas. FEMS  Microbiol Ecol.
   http://dx.doi.org/10.llll/1574-6941.12183

Muylaert. K:  Dasseville. R: De Brabandere. L: Dehairs. F: Vyverman. W. (2005). Dissolved organic carbon in
   the freshwater tidal reaches of the Schelde estuary. Estuar Coast Shelf Sci 64: 591-600.
   http://dx.doi.0rg/10.1016/j.ecss.2005.04.010

Myers, CR: Nealson, KH. (1988). Bacterial manganese reduction and growth with manganese oxide as the sole
   electron acceptor. Science 240:1319-1321. http://dx.doi.org/10.1126/science.240.4857.1319

Newman, DK (2001). Microbiology- How bacteria respire minerals. Science 292:1312-1313.
   http://dx.doi.org/10.1126/science.1060572

Nordstrom. DK: Alpers.  CN. (1999). Geochemistry  of acid mine waters. In GS Plumlee; MJ Logsdon (Eds.), The
   Environmental Geochemistry of Mineral Deposits Part A: Processes, Techniques, and Health Issues Society
   of Economic Geologists, Denver (pp. 133-160). Littleton, CO: Society of Economic Geologists.

Orem, W: Tatu, C: Varonka, M: Lerch, H: Bates, A: Engle, M: Crosby, L: Mcintosh, I. (2014). Organic substances
   in produced and formation water from unconventional natural gas extraction in coal and shale. Int J Coal
   Geol 126: 20-31. http://dx.doi.Org/10.1016/i.coal.2014.01.003
               This document is a draft for review purposes only and does not constitute Agency policy.

June 2015                                       E-44                    DRAFT—DO NOT CITE OR QUOTE

-------
Hydraulic Fracturing Drinking Water Assessment                                                  Appendix E
Orem. WH: Tatu. CA: Lerch. HE: Rice. CA: Bartos. TT: Bates. AL: Tewalt. S: Corum. MD. (2007). Organic
   compounds in produced waters from coalbed natural gas wells in the Powder River Basin, Wyoming, USA.
   Appl Geochem 22: 2240-2256. http://dx.doi.Org/10.1016/j.apgeochem.2007.04.010

PA PEP (Pennsylvania Department of Environmental Protection). (2015). Technologically enhanced naturally
   occurring radioactive materials (TENORM) study report. Harrisburg, PA.
   http://www.elibrary.dep.state.pa.us/dsweb/Get/Document-105822/PA-DEP-TENORM-
   Study Report Rev. 0 01-15-2015.pdf

Pashin. 1C: Mcintyre-Redden. MR: Mann. SD: Kopaska-Merkel. DC: Varonka. M: Orem. W. (2014). Relationships
   between water and gas chemistry in mature coalbed methane reservoirs of the Black Warrior Basin. Int J
   Coal Geol 126: 92-105. http://dx.doi.Org/10.1016/j.coal.2013.10.002

Rahm. BG: Bates. IT: Bertoia. LR: Galford. AE: Yoxtheimer. DA: Riha. SI. (2013). Wastewater management and
   Marcellus Shale gas development: trends, drivers, and planning implications. J Environ Manage 120: 105-
   113. http://dx.doi.0rg/10.1016/j.jenvman.2013.02.029

Ravot. G: Magot. M: Ollivier. B: Patel. BK: Ageron. E: Grimont. PA: Thomas. P: Garcia. IL. (1997).
   Haloanaerobium congolense sp. nov., an anaerobic, moderately halophilic, thiosulfate- and sulfur-reducing
   bacterium from an African oil field. FEMS Microbiol Lett 147: 81-88.

Rice. CA: Flores. RM: Strieker. GD: Ellis. MS. (2008). Chemical and stable isotopic evidence for water/rock
   interaction and biogenic origin of coalbed methane, Fort Union Formation, Powder River Basin, Wyoming
   and Montana USA. Int J Coal Geol 76: 76-85. http://dx.doi.Org/10.1016/j.coal.2008.05.002

Rowan, EL: Engle, MA: Kirby, CS: Kraemer, TF. (2011). Radium content of oil- and gas-field produced waters
   in the northern Appalachian Basin (USA): Summary and discussion of data. (Scientific Investigations
   Report 20115135). Reston, VA: U.S. Geological Survey, http://pubs.usgs.gov/sir/2011/5135/

Rozell. PI: Reaven. SI. (2012). Water pollution risk associated with natural gas extraction from the Marcellus
   Shale. Risk Anal 32:13821393. http://dx.doi.Org/10.llll/j.1539-6924.2011.01757.x

Sirivedhin. T: Dallbauman. L. (2004). Organic matrix in produced water from the Osage-Skiatook petroleum
   environmental research site, Osage county, Oklahoma. Chemosphere 57: 463-469.

SRBC. (2012). Flowback and Produced Water Volume.

Stepan. PI: Shockey. RE: Kurz. BA: Kalenze. NS: Cowan. RM:  Ziman. II: Harm. IA. (2010). Bakken water
   opportunities assessment: Phase I. (2010-EERC-04-03).  Bismarck, ND:  North Dakota Industrial
   Commission, http://www.nd.gov/ndic/ogrp/info/g-018-036-fi.pdf

Struchtemeyer. CG: Elshahed. MS. (2012). Bacterial communities associated with hydraulic fracturing fluids in
   thermogenic natural gas wells in North Central Texas, USA. FEMS Microbiol Ecol 81: 13-25.
   http://dx.doi.0rg/10.llll/i.1574-6941.2011.01196.x

Stumm, W: Morgan, II. (1981). Aquatic chemistry: An introduction emphasizing chemical equilibria in natural
   waters (2nd ed.). New York, NY: Wiley.

Sturchio. NC: Banner. IL: Binz. CM: Heraty. LB: Musgrove. M. (2001). Radium geochemistry of ground waters
   in Paleozoic carbonate aquifers, midcontinent, USA. Appl Geochem 16:109-122.

Swanson. VE. (1955). Uranium in marine black shales of the United States. In Contributions to the geology of
   uranium and thorium by the United States Geological Survey and Atomic Energy Commission for the
   United Nations International Conference on Peaceful Uses of Atomic Energy, Geneva, Switzerland, 1955
   (pp. 451-456). Reston, VA: U.S. Geological Survey, http://pubs.usgs.gov/pp/0300/report.pdf

Timmis. KN. (2010). Handbook of hydrocarbon and lipid microbiology. Berlin, Germany: Springer-Verlag.
   http://www.springer.com/life+sciences/microbiology/book/978-3-540-77584-3

Tourtelot. HA. (1979). Black shale - its deposition and diagenesis. Clays and Clay Minerals 27: 313-321.
   http://dx.doi.org/10.1346/CCMN.1979.0270501
               This document is a draft for review purposes only and does not constitute Agency policy.

June 2015                                       E-45                    DRAFT—DO NOT CITE OR QUOTE

-------
Hydraulic Fracturing Drinking Water Assessment                                                  Appendix E
Tuttle. MLW: Breit. GN: Goldhaber. MB. (2009). Weathering of the New Albany Shale, Kentucky: II.
   Redistribution of minor and trace elements. Appl Geochem 24:1565-1578.
   http://dx.doi.0rg/10.1016/j.apgeochem.2009.04.034

U.S. Department of Transportation. (2012). Large truck and bus crash facts 2012. Washington, D.C.: Federal
   Motor Carrier Safety Administration, U.S. Department of Transportation.
   http://ai.fmcsa.dot.gov/CarrierResearchResults/PDFs/LargeTruckandBusCrashFacts2012.pdf

U.S. EPA (U.S. Environmental Protection Agency). (2004). Evaluation of impacts to underground sources of
   drinking water by hydraulic fracturing of coalbed methane reservoirs. (EPA/816/R-04/003). Washington,
   DC.: U.S. Environmental Protection Agency, Office of Solid Waste.

U.S. EPA (U.S. Environmental Protection Agency). (2015g). Technical development document for proposed
   effluent limitation guidelines and standards for oil and gas extraction. (EPA-821-R-15-003). Washington,
   D.C. http://water.epa.gov/scitech/wastetech/guide/oilandgas/unconv.cfm

USGS (U.S. Geological Survey). (1961). Geology and geochemistry of uranium in marine black shales: a review.
   (U.S. Geological Survey Professional Paper 356-C). Reston, VA. http://pubs.usgs.gov/pp/0356c/report.pdf

USGS (U.S. Geological Survey). (1997). Radioactive elements in coal and fly ash: Abundance, forms, and
   environmental significance [Fact Sheet]. (U.S. Geological Survey Fact Sheet FS-163-97).
   http://pubs.usgs.gov/fs/1997/fsl63-97/FS-163-97.pdf

USGS (U.S. Geological Survey). (2002). Water quality and environmental isotopic analyses of ground-water
   samples collected from the Wasatch and Fort Union formations in areas of coalbed methane
   developmentimplications to recharge and groundwater flow, eastern Powder river basin, Wyoming.
   (Report 02-4045). Reston, VA. http://pubs.usgs.gov/wri/wri024045/

Van Voast, WA. (2003). Geochemical signature of formation waters associated with coalbed methane. AAPG
   Bulletin 87: 667-676.

Vanengelen. MR: Peyton. BM: Mormile. MR: Pinkart. HC. (2008). Fe(III), Cr(VI), and Fe(III) mediated Cr(VI)
   reduction in alkaline media using a Halomonas isolate from Soap Lake, Washington. Biodegradation 19:
   841-850. http://dx.doi.org/10.1007/sl0532-008-9187-l

Vengosh, A: lackson, RB: Warner, N: Darrah, TH: Kondash, A. (2014). A critical review of the risks to water
   resources from unconventional shale gas development and hydraulic fracturing in the United States.
   Environ Sci Technol 48: 36-52. http://dx.doi.org/10.1021/es405118y

Vine. ID. (1956). Uranium-bearing coal in the United States. In Contributions to the geology of uranium and
   thorium by the United States Geological Survey and Atomic Energy Commission for the United Nations
   International Conference on Peaceful Uses of Atomic Energy, Geneva, Switzerland, 1955. Reston, VA: U.S.
   Geological Survey, http://pubs.er.usgs.gov/publication/pp300

Vine. ID: Tourtelot. EB. (1970). Geochemistry of black shale deposits; A summary report. Econ Geol 65: 253-
   272. http://dx.doi.org/10.2113/gsecongeo.65.3.253

Vinson. PS: Vengosh. A:  Hirschfeld. D: Dwyer. GS. (2009). Relationships between radium and radon
   occurrence and hydrochemistry in fresh groundwater from fractured crystalline rocks, North Carolina
   (USA). Chem Geol 260:159-171. http://dx.doi.Org/10.1016/i.chemgeo.2008.10.022

Vulgamore. TB: Clawson. TD: Pope. CD: Wolhart.  SL:  Mayerhofer. Ml: Machovoe. SR: Waltman. CK. (2007).
   Applying hydraulic fracture diagnostics to optimize stimulations in the Woodford Shale. Richardson, TX:
   Society of Petroleum Engineers. http://dx.doi.org/10.2118/110029-MS

Ward, CR. (2002). Analysis and significance of mineral matter in coal seams. Int J Coal Geol 50:135-168.
   http://dx.doi.org/10.1016/S0166-5162r02100117-9
               This document is a draft for review purposes only and does not constitute Agency policy.

June 2015                                       E-46                    DRAFT—DO NOT CITE OR QUOTE

-------
Hydraulic Fracturing Drinking Water Assessment                                                  Appendix E
Warner. NR: Christie. CA: lackson. RB: Vengosh. A. (2013). Impacts of shale gas wastewater disposal on water
   quality in western Pennsylvania. Environ Sci Technol 47:11849-11857.
   http://dx.doi.org/10.1021/es402165b

Warner. NR: lackson. RB: Darrah. TH: Osborn. SG: Down. A: Zhao. K: White. A: Vengosh. A. (2012). Reply to
   Engelder: Potential for fluid migration from the Marcellus Formation remains possible. PNAS 109: E3626-
   E3626. http://dx.doi.org/10.1073/pnas.1217974110

Weaver, TR: Frape, SK: Cherry, IA. (1995). Recent cross-formational fluid flow and mixing in the shallow
   Michigan Basin. Geol Soc Am Bulletin 107: 697-707. http://dx.doi.org/10.1130/0016-
   7606(1995)107<0697:RCFFFA>2.3.CO:2

Webster. IT: Hancock. Gl: Murray. AS. (1995). Modelling the effect of salinity on radium desorption from
   sediments. Geochim Cosmo Act 59: 2469-2476. http://dx.doi.org/10.1016/0016-7037(95)00141-7

White. Gl. (1992). Naturally occurring radioactive materials (NORM) in oil and gas industry equipment and
   wastes: A literature review. (DOE/ID/01570-T158). Bartlesville, OK: U.S. Department of Energy.

Wignall, PG: Myers, Kl. (1988). Interpreting benthic oxygen levels in mudrocks: A new approach. Geology 16:
   452-455. http://dx.doi.org/10.1130/0091-7613(1988)016<0452:IBOLIM>2.3.CO:2

Wuchter. C: Banning. E: Mincer. Tl: Drenzek. Nl: Coolen. Ml. (2013). Microbial diversity and methanogenic
   activity of Antrim Shale formation waters from recently fractured wells. FMICB 4:1-14.
   http://dx.doi.org/10.3389/fmicb.2013.00367

Yakimov. MM: Denaro. R: Genovese. M: Cappello. S: D'Auria. G: Chernikova. TN: Timmis. KN: Golyshin. PN:
   Giluliano. L. (2005). Natural microbial diversity in superficial sediments of Milazzo Harbor (Sicily) and
   community successions during microcosm enrichment with various hydrocarbons. Environ Microbiol 7:
   1426-1441. http://dx.doi.0rg/10.llll/j.1462-5822.2005.00829.x

Yang. IS: Lee. IY: Baek. K: Kwon. TS: Choi. I. (2009). Extraction behavior of As, Pb, and Zn from mine tailings
   with acid and base solutions. J Hazard Mater 171:1-3. http://dx.doi.Org/10.1016/j.jhazmat.2009.06.021

Yoshizawa. S: Wada.  M: Kita-Tsukamoto. K: Ikemoto. E: Yokota. A:  Kogure. K. (2009). Vibrio azureus sp. nov., a
   luminous marine bacterium isolated from seawater. Int J Syst Evol Microbiol 59:1645-1649.
   http://dx.doi.0rg/10.1099/ijs.0.004283-0

Yoshizawa. S: Wada.  M: Yokota. A: Kogure. K. (2010). Vibrio sagamiensis sp. nov., luminous marine bacteria
   isolated from sea water. J Gen Appl Microbiol 56: 499-507.

Zapecza. OS: Szabo. Z. (1988). Natural radioactivity in ground watera review. In National Water Summary
   1986Hydrologic Events and Ground-Water Quality, Water-Supply Paper 2325. Reston, VA: U.S. Geological
   Survey, http://pubs.er.usgs.gov/publication/wsp2325

Zeikus. IG: Hegge. PW: Thompson. TE: Phelps. Tl: Langworthy. TA. (1983). Isolation and description of
   Haloanaerobium  praevalens gen. nov. and sp. nov., an obligatory anaerobic halophile common to Great Salt
   Lake sediments. Curr Microbiol 9: 225-233. http://dx.doi.org/10.1007/BF01567586
               This document is a draft for review purposes only and does not constitute Agency policy.

June 2015                                       E-47                    DRAFT—DO NOT CITE OR QUOTE

-------
Hydraulic Fracturing Drinking Water Assessment                            Appendix F
                     Appendix F
Wastewater Treatment and Waste Disposal
Supplemental Information
        This document is a draft for review purposes only and does not constitute Agency policy.

June 2015                                   DRAFT—DO NOT CITE OR QUOTE

-------
     Hydraulic Fracturing Drinking Water Assessment                                             Appendix F


     Appendix F.  Wastewater Treatment and Waste

         Disposal Supplemental  Information
 1   This appendix provides additional information for context and background to support the
 2   discussions of hydraulic fracturing waste water management and treatment in Chapter 8 of the
 3   Hydraulic Fracturing Drinking Water Assessment Information in this appendix includes: estimates
 4   compiled for several states for volumes of wastewater generated in regions where hydraulic
 5   fracturing is occurring; an overview of the technologies that can be used to treat hydraulic
 6   fracturing wastewater; calculations of estimated treatment process effluent concentrations for
 7   example constituents; a description of the different discharge options for centralized waste
 8   treatment plants; and the water quality needed for wastewater to be reused for hydraulic
 9   fracturing. Discussion is also provided on difficulties that can arise during treatment of hydraulic
10   fracturing wastewaters: the potential impacts of hydraulic fracturing wastewater on biological
11   treatment processes; and an overview of the formation of disinfection byproducts.

     F.I.   Estimates of Wastewater Production in Regions where Hydraulic
            Fracturing is Occurring
12   Table F-l presents estimated wastewater volumes for several states in areas with hydraulic
13   fracturing activity. These data were compiled from production data available on state databases
14   and were tabulated by year. For California, data were compiled for Kern County, where about 95%
15   of hydraulic fracturing is taking place (CCST,_2015.). Production records from Colorado, Utah, and
16   Wyoming include the producing formation for each well reported; data from these states were
17   filtered to select data from formations indicated in the literature as targets for hydraulic fracturing.
18   Data presented for these three states include statewide estimates as well as estimates for selected
19   basins. Data from New Mexico are available from the states in files for three basins as well as for the
2 0   state; these data were not filtered further.

21   Results in Table F-l illustrate some of the challenges associated with obtaining estimates of
2 2   hydraulic fracturing wastewater volumes, especially using publicly available data. Some of the
23   values likely include reported values from conventional wells (wells that may not be hydraulically
24   fractured, and are typically not subject to modern, high volume hydraulic fracturing). For example,
25   the well counts for California, Colorado, Utah, and Wyoming were in the thousands or tens of
26   thousands at least as early as 2000, several years before the surge of modern hydraulic fracturing
27   began in the mid-2000s. The data used for California were from Kern County but are not specific to
2 8   hydraulic fracturing activity. Where producing formations are provided, the accuracy of the
29   estimates will depend upon correct selection of hydraulically fractured formations. Thus, both
3 0   underestimation and overestimation may be possible because of a lack of clear indication of which
31   wells were hydraulically fractured.
                  This document is a draft for review purposes only and does not constitute Agency policy.
     June 2015                                  F-l                  DRAFT—DO NOT CITE OR QUOTE

-------
Hydraulic Fracturing Drinking Water Assessment
                                                                                         Appendix F
Table F-l. Estimated volumes (millions of gallons) of wastewater based on state data for selected years and numbers of wells
          producing fluid.

State
California







Colorado


















Basin
San Joaquin3







All basins
with hy-
draulically
fractured
formations







Denver


Piceance


Principal
lithologies
Shale,
unconsoli-
dated sands





-











Sandstone,
shale

Sandstone



Data type
Produced
water





Wells
Produced
water









Wells
Produced
water
Wells
Produced
water
Wells

2000
46,000






33,695
7,300










11,264
140

1,829
3,500

1,134

2004
48,000






39,033
11,000










14,934
160

1,511
5,800

2,473

2008
58,000






46,519
21,000










28,282
170

1,277
9,300

6,486

2010
65,000






49,201
14,000










33,929
160

1,204
6,900

9,105

2011
71,000






51,031
12,000










35,999
160

1,193
6,500

10,057

2012
75,000






51,567
12,000










33,371
150

1,131
6,800

10,868

2013
74,000






52,763
7,700










37,613
110

1,072
4,300

10,954

2014
-






-
-










-
-

-
-

-

Comments
Data from CA Department of
Conservation, Oil and Gas
Division.3 Produced water data
compiled for Kern County.
Data may also represent
contributions from production
without hydraulic fracturing.

Data from CO Oil and Gas
Conservation Commission. b
Produced water includes
flowback. Data filtered for
formations indicated in
literature as undergoing
hydraulic fracturing and
matched to corresponding
basins. Example counties
selected for presentation as
well as estimated state total.







June 2015
This document is a draft for review purposes only and does not constitute Agency policy.
                               F-2
DRAFT—DO NOT CITE OR QUOTE

-------
Hydraulic Fracturing Drinking Water Assessment
                                                                                                    Appendix F

State
Colorado,
cont.




New Mexico



















Basin
Raton


San Juan


Permian









Raton


San Juan


Total


Principal
lithologies
Coalbed
methane

Coalbed
methane

Shale,
sandstone








Coalbed
methane

Coalbed
methane

-



Data type
Produced
water
Wells
Produced
water
Wells
Produced
water







Wells
Produced
water
Wells
Produced
water
Wells
Produced
water
Wells

2000
2,400

681
1,000

1,183
-








-
-

-
-

-
-

-

2004
4,100

1,634
1,100

1,605
-








-
-

-
-

-
-

-

2008
8,900

2,795
1,300

1,975
-








-
-

-
-

-
-

-

2010
4,300

2,734
2,000

2,220
-








-
-

-
-

-
-

-

2011
3,200

2,778
1,200

2,303
-








-
-

-
-

-
-

-

2012
2,700

2,710
1,100

2,328
31,000








29,839
510

1,495
1,700

22,492
33,000

53,826

2013
2,100

2,545
650

2,333
31,000








30,336
540

1,502
2,000

22,349
34,000

54,237

2014
-

-
-

-
20,000








30,237
310

1,526
1,100

22,076
22,000

53,889

Comments






Data from New Mexico Oil
Conservation Division.0 Data
provided by the state by basin
and for the entire state.
Unclear how much
contribution from production
without hydraulic fracturing.
Produced water includes
flowback.










June 2015
This document is a draft for review purposes only and does not constitute Agency policy.
                                   F-3
DRAFT—DO NOT CITE OR QUOTE

-------
Hydraulic Fracturing Drinking Water Assessment
                                                                                                    Appendix F

State
Utah



















Wyoming












Basin
All basins
with hy-
draulically
fractured
formations






Kaiparowits/
Uinta

San Juan/
Uinta

Uinta


All basins
with hy-
draulically
fractured
formations







Principal
lithologies
-










Coalbed
methane

Coalbed
methane

Shale/sand-
stone

-












Data type
Produced
water








Wells
Produced
water
Wells
Produced
water
Wells
Produced
water
Wells
Produced
water











2000
1,200









3,080
860

1,718
2

62
350

1,067
1,300












2004
1,200









4,377
740

2,517
49

223
420

1,396
1,400












2008
2,300









7,409
1,300

3,761
350

910
560

2,282
1,300












2010
2,400









S,432
1,400

4,329
270

933
680

2,745
1,500












2011
2,700









9,101
1,800

4,838
240

959
700

2,888
1,600












2012
2,900









10,075
2,000

5,538
230

951
640

3,115
1,700












2013
3,400









10,661
2,400

6,046
190

867
830

3,257
1,600












2014
2,800









10,900
1,900

6,334
120

870
790

3,223
1,800












Comments
Data from State of Utah Oil
and Gas Program. d Produced
water includes flowback. Data
filtered by formation indicated
in the literature as hydraul-
ically fractured and matched to
basins. Data presented for
selected basins as well as for
all formations likely to be
hydraulically fractured.










Data from Wyoming Oil and
Gas Conservation
Commission.6 Produced water
may include flowback. Data
filtered by formation indicated
in the literature as
hydraulically fractured and
matched to basins. Data
presented for selected basins
as well as for all formations
likely to be hydraulically
fractured.
June 2015
This document is a draft for review purposes only and does not constitute Agency policy.
                                   F-4
DRAFT—DO NOT CITE OR QUOTE

-------
Hydraulic Fracturing Drinking Water Assessment
                                                                                                           Appendix F

State
Wyoming,
cont.

















Basin


Big Horn


Denver


Green River


Powder
River

Wind River/
Powder
River

Principal
lithologies


Sandstone


Sandstone


Sandstone/
shale

Coalbed
methane

Sandstone/
shale



Data type
Wells

Produced
water
Wells
Produced
water
Wells
Produced
water
Wells
Produced
water
Wells
Produced
water

Wells

2000
3,470

380

365
54

142
0

44
690

1,953
130


966

2004
3,373

350

359
44

118
1

44
630

1,900
330


957

2008
3,535

350

387
49

124
2

60
620

2,001
330


1,013

2010
3,620

380

397
59

140
8

67
660

2,028
400


988

2011
3,728

430

412
76

167
5

67
700

2,119
420


963

2012
3,843

440

414
90

204
5

59
840

2,207
290


959

2013
4,030

420

407
97

230
9

64
970

2,352
110


977

2014
4,213

440

403
170

278
15

67
1,100

2,565
41


900

Comments


















 a California Department of Conservation, Oil and Gas Division. Oil & Gas - Online Data. Monthly Production and Injection Databases:
 ftp://ftp.consrv.ca.gov/pub/oil/new database format/.
 b Colorado Oil and Gas Conservation Commission. Data: Downloads: Production Data: http://cogcc.state.co.us/data2.htmlff/downloads.
 c New Mexico Oil Conservation Division. Production Data. Production Summaries: All Wells Data: http://gotech.nmt.edu/gotech/Petroleum  Data/allwells.aspx.
 d Utah Department of Natural Resources. Division of Oil, Gas, and Mining. Data Research Center. Database Download Files:
 http://oilgas.ogm.utah.gov/Data Center/DataCenter.cfmffproduction.
 e Wyoming Oil and Gas Conservation Commission. Production files by county and year:
 http://wogcc.state.wy. us/prod uctioncountvvear.cfm?Oops=#oops#&RequestTimeOut=6500.
June 2015
This document is a draft for review purposes only and does not constitute Agency policy.
                                      F-5
DRAFT—DO NOT CITE OR QUOTE

-------
      Hydraulic Fracturing Drinking Water Assessment                                              Appendix F


      F.2.    Overview of Treatment Processes for Treating Hydraulic Fracturing
             Wastewater
 1    Treatment technologies discussed in this appendix are classified as basic or advanced. Basic
 2    treatment technologies are ineffective for reducing total dissolved solids (TDS) and are typically not
 3    labor intensive. Advanced treatment technologies can remove TDS and/or are complex in nature
 4    (e.g., energy- and labor-intensive).

      F.2.1.  Basic Treatment
 5    Basic treatment technologies include physical separation, coagulation/oxidation,
 6    electrocoagulation, sedimentation, and disinfection. These technologies are effective at removing
 7    total suspended solids (TSS), oil and grease, scale-forming compounds, and metals, and they can
 8    minimize microbial activity. Basic treatment is typically incorporated in a permanent treatment
 9    facility (i.e., fixed location) but can also be part of a mobile unit for onsite treatment applications.

      F.2.1.1.  Physical Separation
10    The most basic treatment need for oil and gas wastewaters, including those from hydraulic
11    fracturing operations, is separation to remove suspended solids, and oil and grease. The separation
12    method largely depends on the type of resource (s) targeted by the hydraulic fracturing operation.
13    Down-hole separation techniques, including mechanical blocking devices and water shut-off
14    chemicals to prevent or minimize water flow to the well, may be used during production in shale
15    plays containing greater amounts of liquid hydrocarbons. To treat water at the surface, separation
16    technologies such as hydrocyclones, dissolved air or induced gas flotation systems, media (sand)
17    filtration, and biological aerated filters can remove suspended solids and some organics from
18    hydraulic fracturing wastewater.

19    Media filtration can also remove hardness and some metals if chemical precipitation (i.e.,
20    coagulation, lime softening) is also employed (Boschee.JJllJ:). An example of a centralized waste
21    treatment facility (CWT) that uses chemical precipitation and media filtration to treat hydraulic
22    fracturing waste is the Water Tower Square Gas Well Wastewater Processing Facility in
23    Pennsylvania (see Table 8-7). One or more of these technologies is typically used prior to advanced
24    treatment such as reverse osmosis (RO) because advanced treatment processes foul, scale, or
25    otherwise do not operate effectively in the presence  of TSS, certain organics, and/or some metals
26    and metalloid compounds (BosA^S±MHA', Drewes^aij^iKW). The biggest challenge associated
27    with use of these separation technologies is solids disposal from the resulting sludge (IgunmLind
28    OieiLlQll).

      F.2.1.2.  Coagulation/Oxidation
29    Coagulation is the process of agglomerating small, unsettleable particles into larger particles to
30    promote settling. Chemical coagulants such as alum,  iron chloride, and polymers can be used to
31    precipitate TSS, some dissolved solids (except monovalent ions such as sodium and chloride), and
32    metals from hydraulic fracturing wastewater. Adjusting the pH using chemicals such as lime or
33    caustic soda can increase the potential for some constituents, including dissolved metals, to form
34    precipitates. Chemical precipitation is often used in industrial wastewater treatment as a

                   This document is a draft for review purposes only and does not constitute Agency policy.
      June 2015                                    F-6                   DRAFT—DO NOT CITE OR QUOTE

-------
      Hydraulic Fracturing Drinking Water Assessment                                               Appendix F


 1    pretreatment step to decrease the pollutant loading on subsequent advanced treatment
 2    technologies; this strategy can save time, money, energy consumption and the lifetime of the
 3    infrastructure.

 4    Processes using advanced oxidation and precipitation have been applied to hydraulic fracturing
 5    wastewaters in on-site and mobile systems. Hydroxyl radicals generated by cavitation processes
 6    and the addition of ozone can degrade organic compounds and inactivate micro-organisms. The
 7    process can also aid in the precipitation of elements, which cause hardness and scaling in the
 8    treated water (e.g. calcium, magnesium). The process can also reduce sulfate and carbonate
 9    concentrations in the treated water. This type of treatment can be very effective for on-site reuse of
10    waste water [Ely etal.. 2011].

11    The produced solid residuals from coagulation/oxidation processes typically require further
12    treatment, such as de-watering [Duraisamy etal., 2013: Hammer and VanBriesen, 2012].

      F.2.1.3.   Electrocoagulation
13    Electrocoagulation (EC] (Figure F-l] combines the principles of coagulation and electrochemistry
14    into one process (Gomes etal.. 2009]. An electrical current added to the wastewater produces
15    coagulants that then neutralize the charged particles, causing them to destabilize, precipitate, and
16    settle. EC may be used in place of, or in addition to, chemical coagulation. EC can be effective for
17    removal of organics, TSS, and metals, but it is less effective for removing TDS and sulfate. Although
18    it is still considered an emerging technology for unconventional oil and gas wastewater treatment,
19    EC has been used in mobile treatment systems to treat hydraulic fracturing wastewaters
20    (Halliburton. 2014: Igunnu and Chen. 2014]. Limitations with this technology are the potential for
21    scaling, corrosion, and bacterial growth (Gomes etal.. 2009].
      Figure F-l. Electrocoagulation unit.
                Source: Dunkel (2013).
                   This document is a draft for review purposes only and does not constitute Agency policy.
      June 2015                                     F-7                   DRAFT—DO NOT CITE OR QUOTE

-------
      Hydraulic Fracturing Drinking Water Assessment                                              Appendix F


      F.2.1.4.  Sedimentation
 1    Treatment plants may include sedimentation tanks, clarifiers, or some other form of settling basin
 2    to allow larger particles to settle out of the water where they can eventually be collected,
 3    dewatered, and disposed of. These types of tanks/basins all serve the same purpose - to reduce the
 4    amount of solids going to subsequent processes (i.e., overload the media filters).

      F.2.1.5.  Disinfection
 5    Some hydraulic fracturing applications may require disinfection to kill bacteria after treatment and
 6    prior to reuse. Chlorine is a common disinfectant. Chlorine dioxide, ozone, or ultraviolet light can
 7    also be used. This is an important step for reused water because bacteria can cause problems for
 8    further hydraulic fracturing operations by multiplying rapidly and causing build-up in the well
 9    bore, which decreases gas  extraction efficiency.

      F.2.2.  Advanced Treatment
10    Advanced treatment technologies consist of membranes  (reverse osmosis (RO), nanofiltration,
11    ultrafiltration, microfiltration, electrodialysis, forward osmosis, and membrane distillation),
12    thermal distillation technologies, crystallizers, ion exchange, and adsorption. These technologies
13    are effective for  removing TDS and/or targeted compounds. They typically require pretreatment to
14    remove solids and other constituents that may damage or otherwise impede the technology from
15    operating as designed. Advanced treatment technologies can be energy intensive and are typically
16    employed when a purified water effluent is necessary for direct discharge, indirect discharge, or
17    reuse. In some instances, these water treatment technologies can make use of methane generated
18    by the gas well as an energy source. Some advanced treatment technologies can be made mobile for
19    on-site treatment.

      F.2.2.1.  Membranes
20    Pressure-driven membrane processes including microfiltration, ultrafiltration, nano filtration, and
21    RO (Figure F-2) are being used in some settings to treat oil and gas wastewater. These processes
22    use hydraulic pressure to overcome the osmotic pressure of the influent waste stream, forcing  clean
23    water through the membrane (Drewes^taL_2009_). Microfiltration and ultrafiltration processes do
24    not reduce TDS but can remove TSS and some metals and organics (Drewes^taL.2009.). RO and
25    nano filtration are capable of removing TDS, including anions and radionuclides. RO, however, may
26    be limited to treating TDS levels of approximately 40,000 mg/L TDS (Shaffer_etaL_2013j Younos
27
                   This document is a draft for review purposes only and does not constitute Agency policy.
      June 2015                                    F-8                   DRAFT—DO NOT CITE OR QUOTE

-------
     Hydraulic Fracturing Drinking Water Assessment
Appendix F
     Figure F-2. Photograph of reverse osmosis system.
               Source: Thinkstock.

     f.2.2.2.   Electrodialysis
1    Electrodialysis relies on positively and negatively charged particles and coated membranes to
2    separate contaminants from the water (Figure F-3). Electrodialysis has been considered for use by
3    the shale gas industry, but it is not currently widely utilized [ALL Consulting. 2013). IDS
4    concentrations above 15,000 mg/L are difficult to treat by electrodialysis [ALL Consulting, 2013],
5    and oil and divalent cations (e.g. Ca, Fe, Mg) can foul the membranes (Hayes and Severin, 2012b:
6    Guolinetal.,2008].
                   This document is a draft for review purposes only and does not constitute Agency policy.
     June 2015                                      F-9                   DRAFT—DO NOT CITE OR QUOTE

-------
      Hydraulic Fracturing Drinking Water Assessment
Appendix F
      Figure F-3. Picture of mobile electrodialysis units in Wyoming.
               Source: DOE (2006). Permission: ALL Consulting.

      F.2.2.3.  Forward Osmosis/Membrane Distillation
 1    Forward osmosis, an emerging technology for treating hydraulic fracturing wastewater, uses an
 2    osmotic pressure gradient across a membrane to draw the contaminants from a low osmotic
 3    solution (the feed water) to a high osmotic solution [Drewes etal.. 2009]. The selection of the
 4    constituents for the draw solution is very important as the constituents should be more easily
 5    removed from solution than the compounds (e.g. salts) in the feed. Alternatively, draw solutions can
 6    contain components that are more easily reused or recycled. Another emerging technology,
 7    membrane distillation, relies on a thermal gradient across a membrane surface to volatilize pure
 8    water and capture it in the distillate (Drewes et al., 2009).

      F.2.2.4.  Thermal Distillation
 9    Thermal distillation technologies, such as mechanical vapor recompression (MVR) (Figure  F-4) and
10    dewvaporation, use liquid-vapor separation by applying heat to the waste stream, vaporizing the
11    water to separate out impurities, and condensing the vapor into distilled water (Drewes etal.,
12    2009: LEauLLC. 2008: Hamieh and Beckman. 20061 MVR and dewvaporation can treat high-TDS
13    waters and have been proven in the field as effective for treating oil and gas wastewater (Hayes and
14    Severin. 2012b: Drewes  etal.. 2009). Like RO, these processes are energy intensive and are used
15    when the objective is very clean water (i.e., TDS less than 500 mg/L) for direct/indirect discharge
16    or if clean water is needed for reuse. As with membrane processes, scaling is an issue with these
17    technologies, and scale inhibitors may be needed for them to operate effectively (Igunnu and Chen.
18    2014).
                   This document is a draft for review purposes only and does not constitute Agency policy.
      June 2015                                   F-10                 DRAFT—DO NOT CITE OR QUOTE

-------
     Hydraulic Fracturing Drinking Water Assessment
Appendix F
     Figure F-4. Picture of a mechanical vapor recompression unit near Decatur, Texas.
               Source: Drewes et al. (2009). Permission provided.

1    CWTs such as the Judsonia Central Water Treatment Facility in Arkansas, and the Casella-Altela
2    Regional Environmental Services and Clarion Altela Environmental Services, both in Pennsylvania,
3    have NPDES permits and use MVR or thermal distillation for IDS removal. Figure F-5 shows a
4    diagram of the treatment train at another facility, the Maggie Spain facility in Texas, which uses
5    MVR in its treatment of Barnett Shale wastewater [Hayes and Severin. 2012a).
                  This document is a draft for review purposes only and does not constitute Agency policy.
     June 2015                                    F-ll                  DRAFT—DO NOT CITE OR QUOTE

-------
      Hydraulic Fracturing Drinking Water Assessment
    Appendix F
         Flowback
         Delivery
Flash Mixer
Lime and
Polymer,
pHIO
      Concentrated
      Brine to
      Deep Well
      or Reuse
                                                      Surge Tank
                                                      Acid to pH 4
                         Product Water Storage
                                                                                        Filtrate
                                           Product Water
                                           to Reuse or Discharge
                                                                                     Sludge Cake
                                                                                      to Landfill
      Figure F-5. Mechanical vapor recompression process design - Maggie Spain Facility.
               Adapted from: Hayes and Severin (2012a).

 1    Crystallizers can be employed at CWTs to treat high-TDS waters or to further concentrate the waste
 2    stream from a distillation process, reducing residual waste disposal volumes. The crystallized salt
 3    can be landfilled, deep-well injected, or used to produce pure salt products that may be salable
 4    [ErteletaL 20131.

 5    Another thermal method, freeze-thaw evaporation, involves spraying wastewater onto a freezing
 6    pad, allowing ice crystals to form, and the brine mixture that remains in solution to drain from the
 7    ice [Drewes etal.. 2009]. In warmer weather, the ice thaws and the purified water is collected. This
 8    technology cannot treat waters with high methanol concentrations and is only suitable for areas
 9    where the temperature is below freezing in the winter months [Igunnu and Chen. 2014). In
10    addition, freeze-thaw evaporation can only reduce TDS concentrations to approximately 1,000
11    mg/L, which is higher than the 500 mg/L TDS surface water discharge  limit required by most
12    permits  [Igunnu and Chen, 2014].

      F.2.2.5.   Ion Exchange and Adsorption
13    Ion exchange (Figure F-6)  is the process of exchanging ions on a media referred to as resin for
14    unwanted ions in the water. Ion exchange is used to treat for target ions that may be difficult to
15    remove by other treatment technologies or that may interfere with the effectiveness of advanced
16    treatment processes.
                   This document is a draft for review purposes only and does not constitute Agency policy.
      June 2015                                    F-12                  DRAFT—DO NOT CITE OR QUOTE

-------
      Hydraulic Fracturing Drinking Water Assessment
Appendix F
      Figure F-6. Picture of a compressed bed ion exchange unit.
               Source: Drewes et al. (2009). Permission provided.

 1    Adsorption is the process of adsorbing contaminants onto a charged granular media surface.
 2    Adsorption technologies can effectively remove organics, heavy metals, and some anions [Igunnu
 3    and Chen. 2014]. With ion exchange and adsorption processes, the type of resin or adsorptive
 4    media used (e.g., activated carbon, organoclay, zeolites) dictates the specific contaminants that will
 5    be removed from the water [Drewes etal.. 2009: Fakhru'1-Razi et al.. 2009).

 6    Because they can be easily overloaded by contaminants, ion exchange and adsorption treatment
 7    processes are generally used as a polishing step following other treatment processes or as a unit
 8    process in a treatment train rather than as stand-alone treatment [Drewes etal., 2009]. Stand-alone
 9    units require more frequent regeneration and/or replacement of the spent media making these
10    technologies more costly to operate [Igunnu and Chen, 2014]. Figure F-7 shows a schematic of the
11    Pinedale Anticline Water Reclamation Facility located in Wyoming, which uses an ion exchange unit
12    with boron-selective resin as a polishing step to treat hydraulic fracturing wastewater specifically
13    for boron [Boschee. 2012].
                   This document is a draft for review purposes only and does not constitute Agency policy.
      June 2015                                    F-13                  DRAFT—DO NOT CITE OR QUOTE

-------
      Hydraulic Fracturing Drinking Water Assessment
Appendix F
      Figure F-7. Discharge water process used in the Pinedale Anticline field.
               Source: Boschee (2012).

      F.3.   Treatment Technology Removal Capabilities
 1    Table F-2 provides removal efficiencies for common hydraulic fracturing wastewater constituents
 2    by treatment technology. With the exception of TSS and TDS, the studies cited demonstrate removal
 3    for a subset of constituents in a category [e.g.. Gomes et al.. 2009] reported that electrodialysis was
 4    an effective treatment for oil and grease, not all organics). The removal efficiencies include ranges
 5    of 1 to 33% (denoted by +), 34% to 66% (denoted by ++), and greater than 66% removal (denoted
 6    by +++). Cells denoted with "--" indicate that the treatment technology is not suitable for removal of
 7    that constituent or group of constituents. If a particular treatment technology only lists removal
 8    efficiencies for TDS, it can be assumed that in some cases, cations and anions would also be
 9    removed by that technology; therefore, where specific results were not provided in literature, cells
10    denoted with "Assumed" refer to cations and anions that comprise TDS.
                   This document is a draft for review purposes only and does not constitute Agency policy.
      June 2015                                   F-14                  DRAFT—DO NOT CITE OR QUOTE

-------
Hydraulic Fracturing Drinking Water Assessment
Appendix F
Table F-2. Removal efficiency of different hydraulic fracturing wastewater constituents using
          various wastewater treatment technologies.3
Treatment
Technology
Hydrocyclones
Evaporation
(freeze-thaw
evaporation)
Filtration
(granular media)
Chemical
precipitation
Sedimentation
(clarifier)
Dissolved air
flotation
Electro-
coagulation
Advanced
oxidation and
precipitation
Hydraulic Fracturing Wastewater Constituents
TSS
+++
(Duraisamy et
al.,2013)
+++
(Igunnu and
Chen, 2014;
Drewes et al..
2009)
+++
(Barrett, 2010)
+++
(Fakhru'l-Razi
etal., 2009)

++
(NMSU DACC
WUTAP, 2007)

+++
(Shammas,
2010)
+++
(Igunnu and
Chen, 2014;
Bukhari, 2008)


TDS
"
+++
(Igunnu and
Chen, 2014;
Drewes et al..
2009; Arthur
etal., 2005)






+
(Abrams,
2013)
Anions

Assumed






Metals

+++
(Igunnu and
Chen, 2014;
Drewes et al..
2009; Arthur
et al., 2005)

+++b
(Duraisamv et
al.,2013)
+++
(Fakhru'l-Razi
et al., 2009;
AWWA, 1999)


+
(Igunnu and
Chen, 2014)
+/+++
(Abrams,
2013)
Radio-
nuclides
"


+++c
(Zhang et al..
2014)




Organics
++
(Duraisamv et al.,
2013)
+++
(Igunnu and
Chen, 2014;
Duraisamv et al.,
2013; Drewes et
al., 2009)
+++
(Shafer, 2011;
Drewes et al.,
2009)
+++
(Fakhru'l-Razi et
al., 2009)

++/+++
(Duraisamv et al..
2013; Fakhru'l-
Razi et al., 2009)

+++
(Igunnu and
Chen, 2014;
Duraisamv et al..
2013; Fakhru'l-
Razi et al., 2009)
+++d
(Duraisamv etal..
2013)
(Fakhru'l-Razi et
al., 2009)
              This document is a draft for review purposes only and does not constitute Agency policy.

June 2015                                     F-15                  DRAFT—DO NOT CITE OR QUOTE

-------
Hydraulic Fracturing Drinking Water Assessment
Appendix F
Treatment
Technology
Reverse osmosis
Membrane
filtration (UF/MF)
Forward osmosis
Distillation,
including thermal
distillation (e.g.,
mechanical vapor
recompression
(MVR))
Ion exchange
Crystallization
Electrodialysis
Capacitive
deionization
(emerging
technology)
Hydraulic Fracturing Wastewater Constituents
TSS

+++
(Arthur et al.,
2QQ5)



-


TDS
++/+++e
(Alzahrani et
al., 2013;
Drewes et al..
2009)

+++
(Drewes et al..
2009)
+++'
(Haves et al.,
2014; Bruff
and Jikich,
2011; Drewes
etal., 2009)


+++
(ER, 2014)
+++k
(Drewes et al..
2009; Gomes
et al., 2009;
Arthur et al.,
2005)
+++'
(Drewes etal..
2009)
Anions
+++
(Alzahrani et
al.,2013)
(Arthur etal..
2QQ5)

Assumed
+++
(Bruff and
Jikich, 2011;
Drewes et al..
2009)
+++
(Drewes et
al., 2009)
Assumed
++/+++
(Banasiak and
Schafer,
2009)

Metals
++/+++f
(Alzahrani et
al.,2013)
(Drewes et al..
2009; AWWA,
1999)
+++
(Fakhru'l-Razi
et al., 2009)

Assumed
+++
(Haves et al.,
2014; Bruff
and Jikich,
2011; Drewes
et al., 2009)

+++
(Drewes et al..
2009; Arthur
et al., 2005)

Assumed
+/++/+++
(Banasiak and
Schafer, 2009)


Radio-
nuclides
+++
(Drewes et
al., 2009)


+++
(Bruff and
Jikich, 2011;
Drewes et
al., 2009)
+++
(Drewes et
al., 2009)
-


Organics
+/++/+++S
(Drewes et al.,
2009; Munter,
2000)
++/+++
(Duraisamy et al.,
2013; Fakhru'l-
Razi et al., 2009;
Haves and
Arthur, 2004;
AWWA, 1999)h


+/++/+++
(Haves et al..
2014; Duraisamy
et al., 2013;
Drewes et al.,
2009; Fakhru'l-
Razi et al., 2009)

+/++/+++
(Fakhru'l-Razi et
al., 2009;
Munter, 2000 )J

-
+++
(Gomes et al..
2009)

                This document is a draft for review purposes only and does not constitute Agency policy.
June 2015                                          F-16                     DRAFT—DO NOT CITE OR QUOTE

-------
     Hydraulic Fracturing Drinking Water Assessment
Appendix F
Treatment
Technology
Adsorption"1
Biological
treatment
Constructed
wetland/reed
beds
Hydraulic Fracturing Wastewater Constituents
TSS

+++
(Igunnu and
Chen, 2014;
Drewes et al..
2009)
++/+++
(Manios et al..
2QQ3)
TDS


+
(Arthur et al..
2005)
Anions
+/++/+++"
(Habuda-
Stanic et al..
2014)


Metals
+++
(Igunnu and
Chen, 2014;
Drewes et al..
2009)

++/+++
(Fakhru'l-Razi
et al., 2009)

Radio-
nuclides



Organics
+/++/+++
(Arthur et al.,
2005; Haves and
Arthur, 2004;
Munter, 2000)
+/++/+++
(Igunnu and
Chen, 2014;
Drewes et al.,
2009; Fakhru'l-
Razi et al., 2009)

+/+++
(Fakhru'l-Razi et
al., 2009; Arthur
et al., 2005)
      a To the extent possible, removal efficiencies are based on an individual treatment technology that does not assume extensive
      pretreatment or combined treatment processes. However, it should be noted that some processes cannot effectively operate
      without pretreatment (e.g., RO, media filtration, sedimentation).
      b Pretreatment (pH adjustment, aeration, solids separation) required.
      c Radium co-precipitation with barium sulfate.
      d The Fenton process.
      e Typically requires pretreatment. Not a viable technology if TDS influent >50,000 mg/L.
      f Iron and manganese oxides will foul the membranes.
      g Some organics will foul the membranes (e.g., organic acids).
      h Ultrafiltration membrane was modified with nanoparticles.
      ' Can typically handle high TDS concentrations.
      J Resin consisted of modified zeolites that targeted removal of BTEX.
      k Influent TDS for this technology should be <8,000 mg/L.
      1 Specific technology was an electronic water purifier which is a hybrid of capacitive deionization. Influent TDS for this
      technology should be <3,000 mg/L.
      m Typically polishing step, otherwise can overload bed quickly with organics.
      " Removal efficiency is dependent on the type of adsorbent used and the water quality characteristics (e.g., pH).
1    Given the variety of properties among classes of organic constituents, different treatment processes
2    may be required depending upon the types of organic compounds needing removal. Table F-3 lists
3    treatment processes and the classes of organic compounds they can treat
                     This document is a draft for review purposes only and does not constitute Agency policy.
     June 2015                                           F-17                     DRAFT—DO NOT CITE OR QUOTE

-------
     Hydraulic Fracturing Drinking Water Assessment
Appendix F
     Table F-3. Treatment processes for hydraulic fracturing wastewater organic constituents.
Treatment processes
Adsorption with activated carbon
Adsorption with organoclay media
Aeration
Dissolved air flotation
Freeze/thaw evaporation3
Ion exchange (with modified
zeolites)
Distillation
Chemical precipitation
Chemical Oxidation
Media filtration (walnut shell
media or sand)
Microfiltration
Ultrafiltration
Reverse osmosisb
Electrocoagulation
Biologically aerated filters
Reed bed technologies
Hydrocyclone separators
Organic compounds removed
Soluble organic compounds
Insoluble organic compounds
Volatile organic compounds
Volatile organic compounds, dispersed oil
TPH, volatile organic compounds, semi-
volatile organic compounds
BTEX, chemical oxygen demand,
biochemical oxygen demand
BTEX, polycyclic aromatic hydrocarbons
(PAHs)
Oil & grease
Oil & grease
Oil & grease
Oil & grease
Oil & grease, BTEX
Dissolved organics
Chemical oxygen demand, Biochemical
oxygen demand
Oil & grease, TPH, BTEX
Oil & grease, TPH, BTEX
Dispersed oil
References
Fakhru'l-Razietal. (2009)

Fakhru'l-Razietal. (2009)

Tchobanoglous et al. (2013)

Drewes et al. (2009)

Duraisamy et al. (2013); Drewes
etal. (2009)

Haves et al. (2014); Duraisamy et
al. (2013); Drewes et al. (2009);
Fakhru'l-Razietal. (2009);
Munter (2000)

Hayes et al. (2014); Duraisamy et
al. (2013); Drewes et al. (2009);
Fakhru'l-Razietal. (2009).

Drewes et al. (2009); Fakhru'l-
Razi et al. (2009)

Drewes et al. (2009); Fakhru'l-
Razi et al. (2009)

Drewes et al. (2009); Fakhru'l-
Razi et al. (2009)

Drewes et al. (2009); Fakhru'l-
Razi et al. (2009)

Drewes et al. (2009); Fakhru'l-
Razi et al. (2009)

Drewes etal. (2009); U.S. EPA
(2005)
Fakhru'l-Razietal. (2009)

Fakhru'l-Razietal. (2009)

Fakhru'l-Razietal. (2009)

Drewes et al. (2009)
      a Technology cannot be used if the methanol concentration in the hydraulic fracturing wastewater exceeds 5%.
      b RO will remove specific classes of organic compounds with removal efficiencies dependent on the compound's structure and
      the physical and chemical properties of the hydraulically fractured wastewater. Organoacids will foul membranes.
1    Table F-4 presents estimated effluent concentrations that could be produced by a variety of unit
2    treatment processes for several example constituents and for various influent concentrations. This
                    This document is a draft for review purposes only and does not constitute Agency policy.

     June 2015                                        F-18                    DRAFT—DO NOT CITE OR QUOTE

-------
     Hydraulic Fracturing Drinking Water Assessment                                               Appendix F


1    analysis uses treatment process removal efficiencies from literature used to develop Table F-2 and
2    average wastewater concentrations of several constituents presented in Chapter 7 and Appendix E.
3    These estimates were done to illustrate the combined effects of influent wastewater composition
4    and treatment process choice on achievable effluent concentrations. The removal efficiencies
5    represent a variety of studies, primarily at bench and pilot scale, and done with either conventional
6    or hydraulic fracturing wastewater. Removal efficiency for a given treatment process can vary due
7    to a number of factors, and constituent removal may be different in a full-scale facility that uses
8    several processes. Thus, the calculations shown in Table F-4 are intended to be rough
9    approximations for illustrative purposes.
                   This document is a draft for review purposes only and does not constitute Agency policy.
     June 2015                                     F-19                  DRAFT—DO NOT CITE OR QUOTE

-------
Hydraulic Fracturing Drinking Water Assessment
                                                                                         Appendix F
Table F-4. Estimated effluent concentrations for example constituents based on treatment process removal efficiencies.



Shale/
Sandstone
Play


Bakken
Barnett
Fayetteville
Marcellus
Cotton
Valley
Mesaverde
Marcellus
Bakken
Barnett
Fayetteville
Marcellus
Cotton
Valley
Devonian
Sandstone
Marcellus



Contaminant


Barium
Barium
Barium
Barium

Barium
Barium
Cadmium
Strontium
Strontium
Strontium
Strontium

Strontium
Strontium
Radium 226



MCL


2
2
2
2

2
2
5
-
-
-
-

—
-
-



Avg.
Influent
Cone.


10
3.6
4
2200

160
140
25
760
530
27
1700

2300
3900
620



Units


mg/L
mg/L
mg/L
mg/L

mg/L
mg/L
Mg/L
mg/L
mg/L
mg/L
mg/L

mg/L
mg/L
pCi/L
0
+j
E
8
e-Thaw Eva
N
01
£
LL.







2.5









E
O
E
iZ
re
0)
S

i
0.4
0.4
220

16
14
2.5
76
53
2.7
170

230
390

E
O
s
Q.
£
CL
8
E
O)
u















32















-















440


LL.

-------
Hydraulic Fracturing Drinking Water Assessment
                                                                                                    Appendix F
Shale/
Sandstone
Play
Devonian
Sandstone
Marcellus
Marcellus
Barnett
Marcellus
Cotton
Valley
Barnett
Marcellus
Barnett
Marcellus
Barnett
Marcellus
Barnett
Marcellus
Barnett
Marcellus
Barnett
Marcellus
Barnett
Contaminant
Radium 226
Radium 228
Total Radium
TOC
TOC
TOC
BOD
BOD
O&G
O&G
Benzene
Benzene
Toluene
Toluene
Ethyl benzene
Ethyl benzene
Xylenes
Xylenes
BTEX
MCL
~
-
5
-
-
-
-
-
-
-
5
5
1,000
1,000
700
700
10,000
10,000
-
Avg.
Influent
Cone.
2400
120
2500
9.8
160
200
580
40
160
74
680
360
760
1100
29
150
360
1300
1800
Units
pCi/L
pCi/L
pCi/L
mg/L
mg/L
mg/L
mg/L
mg/L
mg/L
mg/L
Mg/L
Mg/L
Mg/L
Mg/L
Mg/L
Mg/L
Mg/L
Mg/L
Mg/L
Freeze-Thaw Evaporation










68
36
76
110
2.9
15
36
130
180
Media Filtration








16
7.4









0
«
4-»
'5.
•o
01
^
Q.
8
E
01
.c
u
120
6.2
130
















-
-
-
















1700
85
1800
















Flotation (DAF)














17
90



Electro-coagulation






58
4











Advanced Oxidation and
precipitation


















7.3
Reverse osmosis
24
1.2
25
















Membrane Filtration
(UF/MF)



0.2
3.2
4


16
7.4
310
170
350
510


170
600

Distillation
24
1.2
25
















-
-
-







6.8
3.6






91
71
3.6
76
















Ion exchange
170
8.4
180



290
20










270






-
-










-






440
30










550
Electrodialysis








8
3.7









Adsorption



0.98
16
20


1.6
0.74
110
58
84
120
3.2
17
14
52
3.7



-
-
-












-



2.9
48
59












91
Biological Treatment
(biodisks, BAFs)



2.1
35
44
29
2
43
19












-
-
-
-
-














4
58
71
87
6











Constructed Wetland



1
16
20
47
3.2
9.8
4.4
ND
ND
ND
ND
ND
ND
ND
ND

June 2015
This document is a draft for review purposes only and does not constitute Agency policy.
                                   F-21
DRAFT—DO NOT CITE OR QUOTE

-------
Hydraulic Fracturing Drinking Water Assessment
                                                                                                    Appendix F




Shale/
Sandstone
Play


Marcellus
Barnett
Marcellus

Barnett


Marcellus


Barnett


Marcellus





Contaminant


BTEX
Naphthalene
Naphthalene
1,2,4-
Trimethyl-
benzene
1,2,4-
Trimethyl-
benzene
1,2,4-
Trimethyl-
benzene
1,2,4-
Trimethyl-
benzene




MCL


-
-
-

-


-


-


-





Avg.
Influent
Cone.


2900
240
360

170


430


59


310





Units


Mg/L
Mg/L
Mg/L

Mg/L


Mg/L


Mg/L


Mg/L

c
o

5.
ze-Thaw Evapo

£
LL.
290

















c
Media Filtratio



















0
*-
arnica! Precipite

U
















































1-_s
Flotation (DAF



















c
o
ectro-coagulat



















IS
c
anced Oxidatio
precipitation

•a
12
0.95
1.4

0.69


1.7


0.24


1.2




(A
Reverse osmos



















c

embrane Filtrat
(UF/MF)





















c
o
5

















150

































Ion exchange


440














-














870


















Electrodialysis





















Adsorption


5.8














-














150
















•£
01 . — .
ological Treatm
(biodisks, BAFs


















































IS
nstructed Wetl

u















ND = Non-detect
June 2015
This document is a draft for review purposes only and does not constitute Agency policy.
                                   F-22
DRAFT—DO NOT CITE OR QUOTE

-------
      Hydraulic Fracturing Drinking Water Assessment                                              Appendix F
      F.4.    Centralized Waste Treatment Facilities and Waste Management Options
 1    CWTs are designed to treat for site-specific wastewater constituents so that the effluent meets the
 2    requirements of the designated disposal option(s) (i.e., reuse, direct/indirect discharge). The most
 3    basic treatment processes that a CWT might use include (Eastolk2ill4j DuhorL_2012.):

 4        •   Physical treatment technologies such as dissolved air or gas flotation technologies, media
 5            filtration, hydrocyclones, and clarification;
 6        •   Chemical treatment technologies such as chemical precipitation and chemical oxidation;
 7            and
 8        •   Biological treatment technologies such as biological aerated filter systems and reed beds.
 9    While these technologies are effective at removing oil and grease, suspended solids, scale-forming
10    compounds, and some heavy metals, if IDS should be reduced as required by the intended disposal
11    option, advanced processes such as RO, thermal distillation, or evaporation are necessary.

      F.4.1.  Discharge Options for CWTs
12    Direct discharge CWTs are allowed to discharge treated wastewater directly to surface waters
13    under the NPDES permit program. Discharge limitations may be based on water quality standards
14    in the NPDES and technology-based effluent limitation guidelines under 40 CFR Part 437. In
15    addition, permitting authorities have permitted facilities for discharge under 40 CFR 435, Subpart
16    E. Judsonia Central Water Treatment Facility in Sunnydale, Arkansas is permitted to directly
17    discharge treated effluent from produced and flowback waters from the Fayetteville Shale play to
18    Byrd pond located on the property. Pinedale Anticline Field Wastewater Treatment Facility in
19    Wyoming, WY, originally designed to treat produced water from tight gas  plays in the Pinedale
20    Anticline Field to levels suitable for reuse, was upgraded to include RO treatment for discharge to a
21    local river. CWTs with NPDES discharge permits may also opt to treat oil and gas wastewater for
22    reuse. Some facilities have the ability to treat wastewater to different qualities (e.g., with or without
23    TDS removal), which they might do to target various reuse water quality criteria. Both the Judsonia
24    facility and Pinedale facility discussed above have the ability to employ either TDS- or non-TDS-
25    removal treatment depending on the customers' needs.

26    Indirect discharge CWTs may treat hydraulic fracturing wastewater and then discharge the treated
27    wastewater effluent to a POTW. Discharge to the POTW is controlled by an Industrial User
28    mechanism, which incorporates pretreatment standards established in 40 CFR Part 437. Two
29    facilities located in Pennsylvania (Eureka Resources) and Ohio (Patriot Water Treatment) include
30    indirect discharge as an option in wastewater treatment. The Eureka-Williamsport facility accepts
31    wastewater (primarily from the Marcellus Shale play) and either treats it for reuse or discharges it
32    to the local POTW. The Patriot facility offers services to hydraulic fracturing operators in the
33    Marcellus and Utica Shale plays for removal of solids and metals using chemical treatment. As of
34    March 2015, however, the Patriot facility is limited by the  Ohio Environmental Protection Agency in
35    accepting only "low salinity" (<50,000 mg/L TDS) produced water and may only discharge 100,000
36    gallons (380,000 L) per day to the Warren Ohio POTW.
                   This document is a draft for review purposes only and does not constitute Agency policy.
      June 2015                                   F-23                  DRAFT—DO NOT CITE OR QUOTE

-------
Hydraulic Fracturing Drinking Water Assessment
                                                                                        Appendix F
 1
 2
 3
 4
 5
 6
 7
 8
 9
10
11
12
13
14
15
16
Zero- discharge CWTs do not discharge treated wastewater; instead, the wastewater is treated and
reused in subsequent hydraulic fracturing operations. WVWEJJ2Q12J state that this practice
reduces potential effects on surface drinking water sources by reducing both direct and indirect
discharges. Zero-discharge facilities may offer different levels of treatment including minimal
treatment (for example, filtration), low-level treatment (chemical precipitation), and/or advanced
treatment (evaporation, crystallization). Reserved Environmental Services (RES) Mt Pleasant,
Pennsylvania, is a zero liquid discharge facility permitted by PA DEP to treat wastewater from the
Marcellus Shale play for reuse. Residual solids are dewatered and sent to a landfill. Treated
wastewater effluent is stored, monitored, and chlorinated for reuse (QWlSiyre^^
F.5.   Water Quality for Reuse
As of 2015, there is no consensus on the water quality requirements for reuse of wastewater for
hydraulic fracturing, and operator opinions vary on the minimum standards for the water quality
needed for fracturing fluids (ffidic^eLaL.JOllj MhmjsMM^Mi]. Table F-5 provides a list of
constituents and the recommended or observed target concentrations for reuse applications. The
wide concentration ranges for many constituents (e.g., TDS ranges from 500 to 70,000 mg/L),
suggest that water quality requirements for reuse are dictated by operation-specific requirements,
including operator preference and selection of fracturing fluid chemistry.
Table F-5. Water quality requirements for reuse.
          Source: U.S.EPA(2015g).
Constituent
TDS
Chloride
Sodium
Reasons for Limiting
Concentrations
Fluid stability
Fluid stability
Fluid stability
Recommended or observed base fluid target
concentrations (mg/L, after blending)b
500 - 70,000
2,000 - 90,000
2,000 - 5,000
Metals
Iron
Strontium
Barium
Silica
Calcium
Magnesium
Sulfate
Potassium
Scale formers3
Scaling
Scaling
Scaling
Scaling
Scaling
Scaling
Scaling
Scaling
Scaling
1-15
1
2-38
20
50 - 4,200
10 - 1,000
124 - 1,000
100 - 500
2,500
              This document is a draft for review purposes only and does not constitute Agency policy.
June 2015                                    F-24                  DRAFT—DO NOT CITE OR QUOTE

-------
     Hydraulic Fracturing Drinking Water Assessment
Appendix F
Constituent
Reasons for Limiting
Concentrations
Recommended or observed base fluid target
concentrations (mg/L, after blending)b
Other
Phosphate
TSS
Oil
Boron
pH(S.U.)
Bacteria (count s/mL)
Not Reported
Plugging
Fluid stability
Fluid stability
Fluid stability
Bacterial growth
10
50 - 1,500
5-25
0-10
6.5-8.1
0 - 10,000
     a Includes total of barium, calcium, manganese, and strontium.
     b Unless otherwise noted.
1    Wastewater quality can be managed for reuse by either blending it with freshwater and allowing
2    dilution to bring the concentrations of problematic constituents to an acceptable range or through
3    treatment [VeiL_201fl]. Treatment, if needed, can be conducted at facilities that are mobile, semi-
4    permanent modular systems, or fully permanent CWTs (NIicot_etaL_2012.). At a minimum, hydraulic
5    fracturing service providers generally prefer that the wastewater be treated to remove TSS,
6    microorganisms, and constituents that form scale or inhibit crosslinking in gelled fluid systems
7    (Eoscheei2QJA}- Figure F-8 shows a schematic of a treatment system to treat wastewater for reuse
8    that can remove suspended solids, hardness, and organic constituents.
                   This document is a draft for review purposes only and does not constitute Agency policy.
     June 2015                                     F-25                   DRAFT—DO NOT CITE OR QUOTE

-------
      Hydraulic Fracturing Drinking Water Assessment
                                                      Appendix F
                        Oil
                      Byproduct
            Lime or
            Caustic
   Sodium
   Sulfate
Soda
Ash
                                                                         Acid
            Frac
        Flowback
           Water
                                                      Treated
                                                      Water
                       am*
                        Air
    GAC:
Organics Polish
                         P ..'!».IJ*.  .-
                         .;;-,..'  f.
 Precip/CIarifier:
Hardness Removal
         Sand Filter:
         TSS Removal
              Oxidation:
            Chlorine Dioxide
      Figure F-8. Diagram of treatment for reuse of flowback and produced water.
                Source: KimbalL(2010l-

 1    In the Marcellus, the wastewater to be reused is first generally treated with oil/gas-water
 2    separation, filtration, and dilution (MjLgtiiLJJ0,1,4). Although many Marcellus treatment facilities
 3    only supply basic reuse treatment that removes oil and solids, advanced treatment facilities that
 4    use techniques such as RO or distillation methods are also in operation (VeiL_2010.).

 5    Reuse concerns can vary with the type of hydraulic fracturing fluid used (e.g., slickwater, linear gel,
 6    crosslinked gel, foam) (Wasylishen and Fulton,  2012) and the anticipated changes in water
 7    chemistry over time (transition from flowback to produced water) (Hammer_and_VanBriesen1
 8    2iH2). Elevated TDS is a concern, but residual constituents from previous fluid mixtures (e.g.,
 9    breakers) may also cause difficulties when reusing water for subsequent fracturing operations
10    (Montgomery. 2013: Walsh. 2013).

11    On-Site Treatment for Reuse

12    On-site systems that treat produced water for reuse can reduce potential impacts to drinking water
13    resources associated with transportation and disposal and facilitate the logistics of reuse by
14    preparing the water close to well sites. These systems sometimes consist of mobile units containing
15    one or more treatment processes that can be moved from site to site to treat waters in newly
16    developed sites that are not yet producing at full-scale.  Semi-permanent facilities that serve a
17    specific area also exist (HalJdorson^jZillSj Bosc!ieeJM12,).

18    Treatment systems are typically tailored for site-specific produced water chemical concentrations
19    and desired water quality treatment goals, including whether significant TDS removal is needed. If
                    This document is a draft for review purposes only and does not constitute Agency policy.
      June 2015                                     F-26                   DRAFT—DO NOT CITE OR QUOTE

-------
      Hydraulic Fracturing Drinking Water Assessment                                             Appendix F


 1    low IDS water is needed, more advanced treatment will be required (see Section 8.5 of Chapter 8),
 2    which can increase the treatment costs to three to four times higher than for treatment systems
 3    that do  not remove IDS [Halldorson. 2013). On-site facilities may be warranted where truck
 4    hauling or seasonal accessibility to and from a central facility is an issue [Boschee, 2014: Tiemann
 5    etal., 2014]. Operators may also consider on-site facilities if they have not fully committed to an
 6    area and the well counts are initially low. In those instances, they can later decide to add or remove
 7    units based on changing production volumes [Boschee, 2014].

      F.6.    Hydraulic Fracturing Impacts on POTWs

      F.6.1.  Potential Impacts on Treatment Processes
 8    Wastewater treatment processes used by POTWs are generally not designed or operated for
 9    wastewater containing high salt concentrations (>0.1-5% salt]. Four basic problems for biological
10    treatment of saline water have been described [Woolard and Irvine. 1995]: 1] microbes in
11    conventional treatment systems tend to be sensitive to changes in ionic strength, 2] microbial
12    metabolic functions are disrupted leading to decreased degradation of carbon compounds, 3]
13    effluent suspended solids are increased due to cell lysis and/or a reduction in organisms that
14    promote flocculation, and 4] the extent of salt acclimation is limited in conventional systems.

15    Biological pre-treatment may be beneficial as an added process in pre-treatment (e.g. prior to
16    indirect discharge from a CWT to a POTW] for removal of organic contaminants. Specialized
17    treatment systems using salt-tolerant bacteria may be beneficial as an additional level of treatment
18    for pre-treating (or polishing] waste waters in centralized treatment systems. (These processes
19    differ from conventional biological processes in standard wastewater treatment, which are not
20    suitable for large volumes of UOG wastewater.] In particular, membrane bioreactors (MBRs] have
21    been examined for the treatment of oil and gas wastewater (Dao etal.. 2013: Kose etal.. 2012:
22    Miller, 2011]. MBRs provide advantages over conventional aeration basin processes as they can be
23    implemented into existing treatment trains more easily and have a much smaller footprint than
24    aeration basins.

25    Because sudden increases in chloride concentration, above 5-8 g/L, may cause problems for
26    wastewater treatment (Ludzack and Noran. 1965]. POTWs planning to accept indirect discharge in
27    the future may find it valuable to restrict influent salt concentrations to a level that will not disturb
28    existing biological treatment processes.

      F.7.    Hydraulic Fracturing and DBFs

      F.7.1.1. Disinfection By-Products
29    This section provides background information on disinfection by-products  (DBFs] and their
30    formation to support the discussion in Section 8.6.1 of Chapter 8 regarding impacts on surface
31    waters  and downstream drinking water utilities due to elevated bromide and iodide in hydraulic
3 2    fracturing waste waters.
                   This document is a draft for review purposes only and does not constitute Agency policy.
      June 2015                                   F-27                  DRAFT—DO NOT CITE OR QUOTE

-------
      Hydraulic Fracturing Drinking Water Assessment                                              Appendix F


 1    Regulated DBFs are a small subset of the full spectrum of DBFs that include other chlorinated and
 2    brominated DBFs as well as nitrogenous and iodated DBFs. Some of the emerging unregulated
 3    DBFs may be more toxic than their regulated counterparts [Harkness etal.. 2015: McGuireetal..
 4    2014: Parker etal., 2014]. Of the many types of DBFs that can form when drinking water is
 5    disinfected, SDWA's Stage 1 and Stage 2 DBF Rules regulate four total trihalomethanes (TTHMs),
 6    five haloacetic acids (HAASs], bromate, and chlorite [U.S. EPA, 2006].

 7    Most brominated DBFs form when water containing organic material and bromide reacts with a
 8    disinfectant such as chlorine during drinking water treatment Parameters that affect DBF
 9    formation include concentration and type of organic material, disinfectant concentration, pH, water
10    temperature, and disinfectant contact time. In addition, many studies have found that elevated
11    bromide levels correlate with increased DBF formation [Singer, 2010: Obolensky and Singer, 2008:
12    Matamoros etal.. 2007: Huaetal.. 2006: Yang and Shang. 2004]. Some studies found similar results
13    for iodide as well [McGuire  etal.. 2014: Parker etal.. 2014]. Pope etal. [2007] reported that
14    increased bromide levels are the second best indicator of DBF formation, with pH being the first.

15    In addition, research finds that higher levels of bromide and iodide contribute to increased
16    concentrations of the brominated and iodated forms of DBFs (both regulated and unregulated],
17    which tend to be more cytotoxic, genotoxic, and carcinogenic than chlorinated species [McGuire et
18    al.. 2014: Parker etal.. 2014: States etal.. 2013: Krasner. 2009: Richardson etal.. 2007]. Studies
19    generally report that the ratios of halogen incorporation into DBFs reflect the ratio of halogen
20    concentrations in the source water [Criquetetal.. 2012: Tones etal.. 2012: Obolensky and Singer.
21    2008].

22    From a regulatory perspective, elevated bromide levels create difficulties in meeting drinking water
23    MCLs. When the TTHMs are predominately in the form of brominated DBFs, the higher molecular
24    weight of bromide (79.9 g/mol] relative to chloride (35.5 g/mol] causes the overall mass of the
25    TTHM sum to increase. This can lead to elevated concentrations of TTHM, in turn potentially
26    leading to violations of the TTHM MCL for the drinking water utility (Francis etal.. 2009].

27    High bromide levels are also cited as causing formation of nitrogenous DBF N-
28    nitrosodimethylamine (NDMA] in water disinfected with chloramines (Luh and Marinas. 2012].
29    Although NDMA is not regulated by the EPA as of early 2015, it is listed as a priority toxic pollutant,
30    and the EPA is planning to evaluate NDMA and other nitrosamines as candidates for regulation
31    during  the six-year review of the Microbial and Disinfection Byproducts (MDBP] rules (U.S. EPA.
32    2014a].

      F.8.    References for Appendix F
      Abrams. R. (2013]. Advanced oxidation frac water recycling system. Presentation presented at 20th
         International Petroleum Environmental Conference, November 12-14, 2013, San Antonio, TX.
      Acharya. HR: Henderson. C: Matis. H: Kommepalli. H: Moore. B: Wang. H. (2011]. Cost effective recovery of
         low-TDS frac flowback water for reuse. (Department of Energy: DE-FE0000784]. Niskayuna, NY: GE Global
         Research. http://www.netl.doe.gov/file%201ibrarv/Research/oil-gas/FE0000784 FinalReportpdf
                   This document is a draft for review purposes only and does not constitute Agency policy.
      June 2015                                    F-28                  DRAFT—DO NOT CITE OR QUOTE

-------
Hydraulic Fracturing Drinking Water Assessment                                                  Appendix F
ALL Consulting (ALL Consulting, LLC). (2013). Water treatment technology fact sheet: Electrodialysis [Fact
   Sheet]. Tulsa, OK. http://www.all-llc.com/publicdownloads/ED-EDRFactSheet.pdf

Alzahrani. S: Mohammad. AW: Hilal. N: Abdullah. P: laafar. 0. (2013). Comparative study of NF and RO
   membranes in the treatment of produced water-Part I: Assessing water quality. Desalination 315: 18-26.
   http://dx.doi.0rg/10.1016/j.desal.2012.12.004

Arthur. ID: Langhus. BG: Patel. C. (2005). Technical summary of oil and gas produced water treatment
   technologies. Tulsa, OK: ALL Consulting, LLC.
   http://www.odinoilandgas.eom/Portals/0/TreatmentOptionsReport.pdf

AWWA (American Water Works Association). (1999). Residential end uses of water. In PW Mayer; WB
   DeOreo (Eds.). Denver, CO: AWWA Research Foundation and American Water Works Association.
   http://www.waterrf.org/PublicReportLibrary/RFR90781 1999 241A.pdf

Banasiak. LI: Schafer. AI. (2009). Removal of boron, fluoride and nitrate by electrodialysis in the presence of
   organic matter. J Memb Sci 334:101-109. http://dx.doi.Org/10.1016/j.memsci.2009.02.020

Barrett, ME. (2010). Evaluation of sand filter performance. (CRWR Online Report 10-7). Austin, TX: Center for
   Research in Water Resources, University of Texas at Austin.
   http://www.crwr.utexas.edu/reports/pdf/2010/rptlO-07.pdf

Boschee. P. (2012). Handling produced water from hydraulic fracturing. Oil and Gas Facilities 1: 23-26.

Boschee. P. (2014). Produced and flowback water recycling and reuse: Economics, limitations, and
   technology. Oil and Gas Facilities 3:16-22.

Bruff. M: likich. SA. (2011). Field demonstration of an integrated water treatment technology solution in
   Marcellus shale. Paper  presented at SPE Eastern Regional Meeting, August 17-19,2011, Columbus, OH.

Bukhari. AA. (2008). Investigation of the electro-coagulation treatment process for the removal of total
   suspended solids and turbidity from municipal wastewater. Bioresour Technol 99: 914-921.
   http://dx.doi.0rg/10.1016/j.biortech.2007.03.015

CCST (California Council on Science and Technology). (2015). An independent scientifc assessment of well
   stimulation in California, Volume 1: Well stimulation technologies and their past, present, and potential
   future use in California. Sacramento, CA. http://www.ccst.us/publications/2015/2015SB4-vl.pdf

Criquet, I: Allard, S: Salhi, E: loll, CA: Heitz, A: von Gunten, U, rs. (2012). lodate and lodo-Trihalomethane
   Formation during Chlorination of Iodide-Containing Waters: Role of Bromide. Environ Sci Technol 46:
   7350-7357. http://dx.doi.org/10.1021/es301301g

Dao. TD: Mericq. IP: Laborie. S: Cabassud. C. (2013). A new method for permeability measurement of
   hydrophobic membranes in Vacuum Membrane Distillation process. Water Res 47: 20962104.

DOE (U.S. Department of Energy). (2006). A guide to practical management of produced water from onshore
   oil and gas operations in the  United States. Washington, DC: U.S. Department of Energy, National
   Petroleum Technology Office.
   http://fracfocus.Org/sites/default/files/publications/a guide to practical management of produced wat
   er from onshore oil and gas operations in  the united states.pdf

Drewes. I: Cath. T:  Debroux. I: Veil. I. (2009). An integrated framework for treatment and management of
   produced water - Technical assessment of produced water treatment technologies (1st ed.). (RPSEA
   Project 07122-12). Golden, CO: Colorado School of Mines.
   http://aqwatec.mines.edu/research/projects/Tech  Assessment PW  Treatment Tech.pdf

Duhon. H. (2012). Produced water treatment: Yesterday, today, and tomorrow. Oil and Gas Facilities 3: 29-31.

Dunkel. M. (2013). Reducing fresh water use in upstream oil and gas hydraulic fracturing. In Summary of the
   technical workshop on wastewater treatment and related modeling (pp. A37-A43). Irving, TX: Pioneer
   Natural Resources USA, Inc. http://www2.epa.gov/hfstudy/summary-technical-workshop-wastewater-
   treatment-and-related-modeling


               This document is a draft for review purposes only and does not constitute Agency policy.
June 2015                                       F-29                    DRAFT—DO NOT CITE OR QUOTE

-------
Hydraulic Fracturing Drinking Water Assessment                                                  Appendix F
Duraisamy. RT: Beni. AH: Henni. A. (2013). State of the art treatment of produced water. In W Elshorbagy; RK
   Chowdhury (Eds.), Water treatment (pp. 199-222). Rijeka, Croatia: InTech.
   http://dx.doi.org/10.5772/53478

Easton. I. (2014). Optimizing fracking wastewater management. Pollution Engineering January 13.

Ely. IW: Horn. A: Cathey. R: Fraim. M: lakhete. S. (2011). Game changing technology for treating and recycling
   frac water. Paper presented at SPE Annual Technical Conference and Exhibition, October 30 - November 2,
   2011, Denver, CO.

ER (Eureka Resources, LLC). (2014). Crystallization technology. Available online at http://www.eureka-
   resources.com/wp-content/uploads/2013/07/EURE-022 Crystallization 53013.pdf (accessed March 4,
   2015).

Ertel. D: McManus. K: Bogdan. I. (2013). Marcellus wastewater treatment: Case study. In Summary of the
   technical workshop on wastewater treatment and related modeling (pp. A56-A66). Williamsport, PA:
   Eureka Resources, LLC. http://www2.epa.gov/hfstudy/summary-technical-workshop-wastewater-
   treatment-and-related-modeling

Fakhru'1-Razi. A: Pendashteh. A: Abdullah. LC: Biak. PR:  Madaeni. SS: Abidin. ZZ. (2009). Review of
   technologies for oil and gas produced water treatment [Review]. J Hazard Mater 170: 530-551.

Francis. RA: Small. Ml: Vanbriesen. IM.  (2009). Multivariate distributions of disinfection by-products in
   chlorinated drinking water. Water Res 43: 3453-3468. http://dx.doi.Org/10.1016/j.watres.2009.05.008

Gomes. I: Cocke. D: Das. K: Guttula. M: Tran. D: Beckman: I. (2009). Treatment of produced water by
   electrocoagulation. Shiner, TX: KASELCO, LLC. http://www.kaselco.com/index.php/library/industry-
   white-papers

Guolin. I: Xiaoyu. W: Chunjie. H. (2008). The effect of oilfield polymer-flooding wastewater on anion exchange
   membrane performance. Desalination 220: 386-393.

Habuda-Stanic. M: Ravancic. ME: Flanagan. A. (2014).  A Review on Adsorption of Fluoride from Aqueous
   Solution. Materials 7: 6317-6366. http://dx.doi.org/10.3390/ma7096317

Halldorson, B. (2013). Successful oilfield water management: Five unique case studies. Presentation
   presented at EPA Technical Workshop - Wastewater Treatment and Related Modeling Research, April 18,
   2013, Triangle Park, NC.

Halliburton. (2014). Hydraulic fracturing 101. Available online at
   http://www.halliburton.com/public/projects/pubsdata/hydraulic fracturing/fracturing 101.html

Hamieh. BM: Beckman. IR. (2006). Seawater desalination using Dewvaporation technique: theoretical
   development and design evolution. Desalination 195: 1-13.
   http://dx.doi.0rg/10.1016/i.desal.2005.09.034

Hammer. R: VanBriesen. I. (2 012). In frackings wake:  New rules are needed to protect our health and
   environment from contaminated wastewater. New York, NY: Natural Resources Defense Council.
   http://www.nrdc.org/energy/files/fracking-wastewater-fullreport.pdf

Harkness. IS: Dwyer. GS: Warner. NR: Parker. KM: Mitch. WA: Vengosh. A. (2015). Iodide, Bromide, and
   Ammonium in Hydraulic Fracturing and Oil and Gas Wastewaters: Environmental Implications. Environ
   Sci Technol 49:1955-1963. http://dx.doi.org/10.1021/es504654n

Hayes. T: Severin. B. (2012a). Characterization of flowback water from the the Marcellus and the Barnett
   shale regions. Barnett and Appalachian shale water management and reuse technologies. (08122-05.09;
   Contract 08122-05). Hayes, T; Severin, B. http://www.rpsea.org/media/files/project/2146b3aO/08122-
   05-RT-Characterization Flowback Waters Marcellus Barnett Shale Regions-03-20-12.pdf
               This document is a draft for review purposes only and does not constitute Agency policy.

June 2015                                       F-30                    DRAFT—DO NOT CITE OR QUOTE

-------
Hydraulic Fracturing Drinking Water Assessment                                                  Appendix F
Hayes. T: Severin. BF. (2012b). Evaluation of the aqua-pure mechanical vapor recompression system in the
   treatment of shale gas flowback water - Barnett and Appalachian shale water management and reuse
   technologies. (08122-05.11). Hayes, T; Severin, BF. http://barnettshalewater.org/documents/08122-
   05.11-EvaluationofMVR-3-12-2012.pdf

Hayes. TD: Arthur. D. (2004). Overview of emerging produced water treatment technologies. Paper presented
   at llth Annual International Petroleum Environmental Conference, October 12-15, 2004, Albuquerque,
   NM.

Hayes. TD: Halldorson. B: Horner. P: Ewing. I: Werline. IR: Severin. BF. (2014). Mechanical vapor
   recompression for the treatment of shale-gas flowback water. Oil and Gas Facilities 3: 54-62.

Hua. GH: Reckhow. DA: Kim. I. (2006). Effect of bromide and iodide ions on the formation and speciation of
   disinfection byproducts during chlorination. Environ Sci Technol 40: 3050-3056.
   http://dx.doi.org/10.1021/es0519278

Igunnu.  ET: Chen. GZ. (2014). Produced water treatment technologies. International Journal of Low-Carbon
   Technologies 9:157-177. http://dx.doi.org/10.1093/iilct/cts049

lones. DB: Saglam. A: Song. H: Karanfil. T. (2012). The impact of bromide/iodide concentration and ratio on
   iodinated trihalomethane formation and speciation. Water Res 46: 11-20.
   http://dx.doi.0rg/10.1016/j.watres.2011.10.005

Kimball. B. (2010). Water treatment technologies for global unconventional gas plays. Presentation presented
   at US - China Industry Oil and Gas Forum, September 16, 2010, Fort Worth, TX.

Rose. B:  Ozgun. H: Ersahin. ME: Dizge. N:  Koseoglulmer. DY: Atay. B: Kaya. R: Altinbas. M: Sayili. S: Hoshan. P:
   Atay. D: Eren. E: Kinaci. C: Koyuncu. I. (2012). Performance evaluation of a submerged membrane
   bioreactor for the treatment of brackish oil and natural gas field produced water. Desalination 285: 295-
   300.

Krasner. SW. (2009). The formation and control of emerging disinfection by-products of health concern
   [Review]. Philos Transact A Math Phys Eng Sci 367: 4077-4095. http://dx.doi.org/10.1098/rsta.2009.010

LEau LLC. (2008). Dew vaporation desalination 5,000-gallon-per-day pilot plant. (Desalination and Water
   Purification Research and Development Program Report No. 120). Denver, CO: Bureau of Reclamation, U.S.
   Department of the Interior. http://www.usbr.gov/research/AWT/reportpdfs/reportl20.pdf

Ludzack, Fl: Noran, DK. (1965). Tolerance of high salinities by conventional wastewater treatment processes.
   J Water Pollut Control Fed 37:1404-1416.

Luh. I: Marinas. Bl. (2012). Bromide ion effect on N-nitrosodimethylamine formation by monochloramine.
   Environ Sci Technol 46: 5085-5092. http://dx.doi.org/10.1021/es300077x

Ma. G: Geza. M: Xu. P. (2014). Review of flowback and produced water management, treatment, and beneficial
   use for major shale gas development basins. Shale Energy Engineering Conference 2014,  Pittsburgh,
   Pennsylvania, United States.

Manios.  T: Stentiford. El: Millner. P. (2003). Removal of total suspended solids from wastewater in
   constructed horizontal flow subsurface wetlands. J Environ Sci Health A Tox Hazard Subst Environ Eng 38:
   1073-1085. http://dx.doi.org/10.1081/ESE-120019865

Matamoros, V: Mujeriego, R: Bayona, IM. (2007). Trihalomethane occurrence in chlorinated reclaimed water
   at full-scale wastewater treatment plants in NE Spain. Water Res 41: 3337-3344.
   http://dx.doi.0rg/10.1016/j.watres.2007.04.021

McGuire. Ml: Karanfil. T: Krasner. SW: Reckhow. DA: Roberson. I A: Summers. RS: Westerhoff. P: Xie. Y. (2014).
   Not your granddad's disinfection by-product problems and solutions. JAWWA 106: 54-73.
   http://dx.doi.org/10.5942/jawwa.2014.106.0128

Miller. P. (2011). Future of hydraulic fracturing depends on effective water treatment. Hydrocarbon Process
   90:13-13.


              This document is a draft for review purposes only and does not constitute Agency policy.
June 2015                                       F-31                    DRAFT—DO NOT CITE OR QUOTE

-------
Hydraulic Fracturing Drinking Water Assessment                                                  Appendix F
Montgomery. C. (2013). Fracturing fluid components. In A Bunder; J McLennon; R Jeffrey (Eds.), Effective and
   Sustainable Hydraulic Fracturing. Croatia: InTech. http://dx.doi.org/10.5772/56422

Munter. R. (2000). Industrial wastewater treatment. In LC Lundin (Ed.), Sustainable water management in the
   Baltic Sea Basin book II: Water use and management (pp. 195-210). Sida, Sweden: Baltic University
   Programme Publication, http://www.balticuniv.uu.se/index.php/boll-online-library/831-swm-2-water-
   use-and-management

Nicot, IP: Reedy, RC: Costley, RA: Huang, Y. (2012). Oil & gas water use in Texas: Update to the 2011 mining
   water use report Nicot, JP; Reedy, RC; Costley, RA; Huang, Y.
   http://www.twdb.state.tx.us/publications/reports/contracted reports/doc/0904830939 2012Update M
   iningWaterUse.pdf

NMSU DACC WUTAP (New Mexico State University, Dona Ana Community College, Water Utilities Technical
   Assistance Program). (2007). New Mexico wastewater systems operator certification study manual -
   Version 1.1. Santa Fe, NM: New Mexico Environment Department.
   http://www.nmrwa.org/sites/nmrwa.org/files/WastewaterOperatorStudyManual.pdf

Obolensky, A: Singer, PC. (2008). Development and interpretation of disinfection  byproduct formation models
   using the Information Collection Rule database. Environ Sci Technol 42: 5654-5660.
   http://dx.doi.org/10.1021/es702974f

ONG Services. (2015). ONGList: Reserved Environmental Services. Available online at
   http://www.onglist.com/Home/Search?SearchString=Reserved+environmental+services&Distance=&sea
   rchAddress=&CategoryTypeID=l&SubCategoryID

Parker. KM: Zeng. T: Harkness. I: Vengosh. A: Mitch. WA. (2014). Enhanced formation of disinfection
   byproducts in shale gas wastewater-impacted drinking water supplies. Environ Sci Technol 48:11161-
   11169. http://dx.doi.org/10.1021/es5028184

Pope. PG: Martin-Doole. M: Speitel. GE: Collins. MR. (2007). Relative significance of factors influencing DXAA
   formation during chloramination. JAWWA 99:144-156.

Richardson. SD: Plewa. Ml: Wagner. ED: Schoeny. R: Demarini. DM. (2007). Occurrence, genotoxicity, and
   carcinogenicity of regulated and emerging disinfection by-products in drinking water: A review and
   roadmap for research [Review]. Mutat Res 636:178-242. http://dx.doi.Org/10.1016/j.mrrev.2007.09.00

Shafer. L. (2011). Water recycling and purification in the Pinedale anticline field: results from the anticline
   disposal project. In 2011 SPE Americas E&P health, safety, security & environmental conference.
   Richardson, TX: Society of Petroleum Engineers. http://dx.doi.org/10.2118/141448-MS

Shaffer. PL: Arias Chavez. LH: Ben-Sasson. M: Romero-Vargas Castrillon. S: Yip. NY: Elimelech. M. (2013).
   Desalination and reuse of high-salinity shale gas produced water: drivers, technologies, and future
   directions. Environ Sci  Technol 47: 9569-9583.

Shammas, NK. (2010). Wastewater renovation by flotation. In LK Wang; NK Shammas; WA Selke; DB
   Aulenbach (Eds.), Flotation technology (pp. 327-345). New York, NY: Humana Press.
   http://dx.doi.org/10.1007/978-l-60327-133-2 9

Singer. P. (2010). Anomalous DBF  speciation patterns:  Examples and explanations. Water Quality Technology
   Conference and Exposition 2010, November, 14-18, 2010, Savannah, GA.

States, S: Cyprych, G: Stoner, M: Wydra, F: Kuchta, I: Monnell, I: Casson, L. (2013). Marcellus Shale drilling and
   brominated THMs in Pittsburgh, Pa., drinking water. J Am Water Works Assoc 105: E432-E448.
   http://dx.doi.org/10.5942/jawwa.2013.105.0093

Tchobanoglous. G: Burton. FL: Stensel. HP. (2013). Wastewater engineering: Treatment and reuse. In th (Ed.),
   (9780070418783 ed.).  Boston, MA: McGraw-Hill.
               This document is a draft for review purposes only and does not constitute Agency policy.

June 2015                                       F-32                    DRAFT—DO NOT CITE OR QUOTE

-------
Hydraulic Fracturing Drinking Water Assessment                                                  Appendix F
Tiemann. M: Folger. P: Carter. NT. (2014). Shale energy technology assessment: Current and emerging water
   practices. Washington, DC: Congressional Research Service, http://nationalaglawcenter.org/wp-
   content/uploads//assets/crs/R43635.pdf

U.S. EPA (U.S. Environmental Protection Agency). (2005). Membrane filtration guidance manual. (EPA 815-R-
   06-009). Washington, D.C.
   http://www.epa.gov/ogwdw/disinfection/lt2/pdfs/guide It2  membranefiltration final.pdf

U.S. EPA (U.S. Environmental Protection Agency). (2006). National Primary Drinking Water Regulations:
   Stage 2 Disinfectants and Disinfection Byproducts Rule.
   http://water.epa.gov/lawsregs/rulesregs/sdwa/stage2/

U.S. EPA. Announcement of preliminary regulatory determinations for contaminants on the third drinking
   water contaminant candidate list. EPA-HQ-OW-2012-0155 62715  -62750 (62736 pages) (2014a).
   https://www.federalregister.gov/articles/2014/10/20/2014-24582/announcement-of-preliminary-
   regulatory-determinations-for-contaminants-on-the-third-drinking-water#page-62715

U.S. EPA (U.S. Environmental Protection Agency). (2015g). Technical development document for proposed
   effluent limitation guidelines and standards for oil and gas extraction. (EPA-821-R-15-003). Washington,
   D.C. http://water.epa.gov/scitech/wastetech/guide/oilandgas/unconv.cfm

Veil. I A. (2010). Water management technologies used by Marcellus shale gas producers - Final Report.  (DOE
   Award No.:  FWP 49462). Veil, JA.
   http://fracfocus.org/sites/default/files/publications/water management in the marcellus.pdf

Vidic, RD: Brantley, SL: Vandenbossche, IM: Yoxtheimer, D: Abad, ID. (2013). Impact of shale gas development
   on regional water quality [Review]. Science 340:1235009. http://dx.doi.org/10.1126/science.1235009

Walsh, IM. (2013). Water management for hydraulic fracturing in unconventional resourcesPart 1. Oil and
   Gas Facilities 2.

Wasylishen. R:  Fulton. S. (2012). Reuse of flowback & produced water for hydraulic fracturing in tight oil.
   Calgary, Alberta, Canada: The Petroleum Technology Alliance Canada (PTAC).
   http://www.ptac.org/projects/151

Woolard, CR: Irvine, RL. (1995). Treatment of of hypersaline wastewater in the sequencing batch reactor.
   Water Res 29:1159-1168.

WVWRI (West Virginia Water Research Institute, West Virginia University). (2012). Zero discharge water
   management for horizontal shale gas well development. (DE-FE0001466). https://www.netl.doe.gov/File
   Library/Research/Oil-Gas/Natural Gas/shale gas/fe0001466-final-report.pdf

Yang. X: Shang. C. (2004). Chlorination byproduct formation in the presence of humic acid, model nitrogenous
   organic compounds, ammonia, and bromide. Environ Sci Technol 38: 4995-5001.
   http://dx.doi.org/10.1021/es049580g

Younos. T: Tulou. KE. (2005). Overview of desalination techniques. Journal of Contemporary Water Research
   & Education 132: 3-10. http://dx.doi.Org/10.llll/j.1936-704X.2005.mpl32001002.x

Zhang. T: Gregory. K: Hammack. RW: Vidic.  RD. (2014). Co-precipitation of radium with barium and strontium
   sulfate and its impact on the fate of radium during treatment of produced water from unconventional gas
   extraction. Environ Sci Technol 48: 4596-4603. http://dx.doi.org/10.1021/es405168b
               This document is a draft for review purposes only and does not constitute Agency policy.

June 2015                                       F-33                    DRAFT—DO NOT CITE OR QUOTE

-------
Hydraulic Fracturing Drinking Water Assessment                           Appendix G
                    Appendix G
Identification and Hazard Evaluation of
Chemicals across the Hydraulic Fracturing
Water Cycle Supplemental Tables and
Information
        This document is a draft for review purposes only and does not constitute Agency policy.

June 2015                                  DRAFT—DO NOT CITE OR QUOTE

-------
     Hydraulic Fracturing Drinking Water Assessment                                            Appendix G


     Appendix G.  Identification  and Hazard Evaluation of

         Chemicals  across the  Hydraulic Fracturing Water  Cycle

         Supplemental Tables  and Information
 1   Appendix G provides detail and supporting information on the oral reference values (RfVs) and oral
 2   slope factors (OSFs) that were identified in Chapter 9 of this assessment1 Section G.I provides
 3   detail on the criteria used to select sources of RfVs and OSFs for chemicals used or detected in
 4   hydraulic fracturing processes, and lists all sources of RfVs and OSFs that were considered for this
 5   study. Section G.2 provides a glossary of the toxicity value terminology that is used by these various
 6   sources. Lastly, all of the RfVs and OSFs collected from these sources are provided in Table G-l and
 7   Table G-2. Tables G-la through G-ld show the available RfVs and OSFs for chemicals used in
 8   hydraulic fracturing fluids, and Tables G-2 a through G-2d show the available RfVs and OSFs for
 9   chemicals detected in hydraulic fracturing flowback and wastewater. These tables provide cancer
10   weight-of-evidence (WOE) characterizations for these chemicals where available, and indicate
11   whether each chemical has available data on physicochemical properties or occurrence.

     G.I.   Criteria for Selection and Inclusion of Reference Value (RfV)  and Oral
            Slope Factor (OSF)  Data Sources
12   The criteria listed below were used to evaluate the quality of RfVs and OSFs considered for use in
13   the hazard analyses conducted in Chapter 9. These criteria were originally outlined in the hydraulic
14   fracturing research plan (U '•> U'/-,, ,:dl 1.1) and interim progress report (U1S._EPA1_2012c). Only data
15   sources that met these criteria were considered of sufficient quality to be included in the analyses.

16   The following criteria had to be met for a source to be deemed of sufficient quality:

17   1)  The body or organization generating or producing the peer-reviewed RfVs, peer-reviewed OSFs,
18       or peer reviewed qualitative assessment must be  a governmental or intergovernmental body.
19          a.  Governmental bodies include sovereign states, and federated states/units.
20          b.  Intergovernmental bodies are those whose members are sovereign states, and the
21             subdivisions or agencies of such intergovernmental bodies. The United Nations is an
2 2             example of an intergovernmental body. The International Agency for Research on
23             Cancer (IARC) is an agency of the World Health Organization (WHO), which is itself an
24             agency of the United Nations. Thus, IARC is considered a subdivision of the United
25             Nations.
     1 As defined in Chapter 9, the term RfV refers to reference values for noncancer effects occurring via the oral route of
     exposure and for chronic durations, except where noted.


                  This document is a draft for review purposes only and does not constitute Agency policy.

     June 2015                                  G-l                  DRAFT—DO NOT CITE OR QUOTE

-------
      Hydraulic Fracturing Drinking Water Assessment                                               Appendix G


 1    2)  The data source must include peer-reviewed RfVs, peer-reviewed OSFs, or peer reviewed
 2       qualitative assessments.
 3           a.  A committee that is established to derive the RfVs, OSFs, or qualitative assessments can
 4              have members of that same committee provide the peer review, so long as either the
 5              entire committee, or members of the committee who did not participate in the
 6              derivation of a specific section of a work product, conduct the review.
 7           b.  Peer reviewers who work for grantees of the organization deriving the RfVs, OSFs, or
 8              qualitative assessments are generally allowed, and this will not be considered to
 9              constitute a conflict/duality of interest.
10           c.  Peer reviewers may work in the same or different office, so long as they did not
11              participate in any way in the development of the product, and these individuals must be
12              free of conflicts/duality of interest with respect to the chemical(s) assigned.
13                  i.  For instance, peer reviewers for Program X, conducted by Office A, may also be
14                     employed by Office A so long as they did not participate in the creation of the
15                     Program X product they are reviewing.

16    3)  The RfVs, OSFs, or qualitative assessments must be based on peer-reviewed scientific data.
17           a.  There are cases where industry reports that were not published in a peer-reviewed,
18              scholarly journal maybe used, if the industry report has been adequately peer-reviewed
19              by an external body (external to the group generating the report, and external to  the
20              group generating the peer-reviewed RfVs, peer-reviewed OSFs, or peer-reviewed
21              qualitative assessment) that is free of conflicts/dualities of interest

22    4)  The RfVs, OSFs, or qualitative assessments must be focused on protection of the general public.
23           a.  Sources that are focused on workers are not appropriate as workers are assumed to
24              accommodate additional risk than the general public due to their status as workers.

25    5)  The body generating the values or qualitative assessments must be free of conflicts of interest
26       with respect to the chemicals for which it derives RfVs, OSFs, or qualitative assessments.
27           a.  If a body generating the RfVs, OSFs, or qualitative assessments accepts funding from an
2 8              interested party (i.e., a company or organization that may be impacted by past, present,
29              or future values or qualitative assessments), then the body has a conflict of interest
30           b.  For instance, if a non-profit organization is funded by an industry trade group, and the
31              non-profit generates RfVs, OSFs, or qualitative assessments for chemicals that trade
3 2              group is interested in, then the non-profit is considered to have a conflict of interest
3 3              with respect to those chemicals.

34    It is important to note that having a conflict/duality of interest for one chemical is sufficient to
3 5    disqualify the entire database, as it is assumed that conflicts/dualities of interest may exist for
36    other chemicals as well.
                   This document is a draft for review purposes only and does not constitute Agency policy.
      June 2015                                     G-2                   DRAFT—DO NOT CITE OR QUOTE

-------
      Hydraulic Fracturing Drinking Water Assessment                                             Appendix G


      G.I.I. Included Sources
 1    We applied our criteria to 16 different sources of RfVs and/or OSFs. After application of our criteria,
 2    we were left with eight sources. For those sources which did not meet our criteria, we provide an
 3    explanation of why they were excluded.

 4    The following sources were evaluated, met our criteria, and were selected as sources of reference
 5    doses or cancer slope factors for this analysis:

 6        •  U.S. EPA Integrated Risk Information System (IRIS)
 7        •  U.S. EPA Human Health Benchmarks for Pesticides (HHBP)
 8        •  U.S. EPA Provisional Peer-Reviewed Toxicity Values (PPRTVs)
 9        •  U.S. Agency for Toxic Substances and Disease Registry (ATSDR) Minimum Risk
10           Levels (MRLs)
11        •  California EPA Toxicity Criteria Database
12        •  International Programme On Chemical Safety (IPCS) Concise International Chemical
13           Assessment Documents (CICAD)
14    The following sources were evaluated, met our criteria, and were selected as sources of qualitative
15    cancer classifications:

16        •  International Agency for Research  on Cancer (IARC)
17        •  US National Toxicology Program Report on Carcinogens (RoC)
18    RfVs and/or OSFs from these data sources are  listed in Tables G-la through G-ld for chemicals used
19    in hydraulic fracturing fluid formulation, and Tables G-2a through G-2d for chemicals reported in
2 0    hydraulic fracturing flowback and produced water.

21    In addition, Table G-l and Table G-2 also list the EPA's drinking water maximum contaminant levels
22    (MCLs) and maximum contaminant goal levels (MCLG) when available. These values are generally
23    based on IRIS values,  and are treatment-based. MCL and MCLG values are listed for reference only,
24    and were not considered in the hazard analysis presented in Chapter 9.

      G.I.2. Excluded Sources
25        •  American Conference of Governmental Industrial Hygienists: The assessments
2 6           derived by this body are specific  to workers and are not generalizable to the general
27           public. In addition, this body is not a governmental or intergovernmental body. Thus, these
2 8           values were  excluded based on criteria 1 and 4.
29        •  European Chemicals  Bureau, Classification and Labeling Annex I  of Directive
30           67/548/EEC: These assessments are not based on peer-reviewed values, but are based on
31           data supplied by manufacturers.  Further, the enabling legislation states that
32           "Manufacturers, importers, and downstream users shall examine the information...to
3 3           ascertain whether it is adequate, reliable and scientifically valid for the purpose of the
                   This document is a draft for review purposes only and does not constitute Agency policy.
     June 2015                                   G-3                  DRAFT—DO NOT CITE OR QUOTE

-------
      Hydraulic Fracturing Drinking Water Assessment                                              Appendix G


 1            evaluation..." This clearly demonstrates that the data and the evaluation are not required
 2            to be peer-reviewed. Thus, these values were excluded based on criterion 2.
 3        •   Toxicology Excellence for Risk Assessment's (TERA's) International Toxicity
 4            Estimates for Risk Assessment (ITER): The ITER database is developed by TERA a
 5            501(c)(3) non-profit TERA accepts funding from various sources, including interested
 6            parties that may be impacted by their assessment work. Thus, ITER is excluded based on
 7            criteria 1 and 5.
 8        •   Other U.S. states: The EPA evaluated values from all states that had values reported on
 9            their websites. If a state's values were determined to be  largely duplicative of the EPA's
10            values (e.g.,  the state adopts EPA values, such as the regional screening levels, and does
1 1            not typically generate its own peer-reviewed values), that state's values were no longer
12            considered.  The EPA contacted those states whose values were determined to not be
1 3            duplicative of EPA's values, and confirmed whether or not a peer review process was used
14            to develop the state's values. The EPA determined that of the states with values not
1 5            duplicative of the EPA's values, only California's values met all of the EPA's criteria for this
16            report Other states with publicly accessible RfVs and/or OSFs include: Alabama, Florida,
17            Hawaii, and Texas.
18        •   WHO Guidelines for Drinking- Water Quality: The WHO Guidelines' values are not RfVs,
19            but rather drinking water values.

      G.2.   Glossary of Toxicity Value Terminology
20    This section defines the toxicity values and qualitative cancer classifications that are frequently
2 1    found in the sources identified above.

22    Lowest-observed-adverse-effect level (LOAEL): The lowest exposure level at which there are
23    biologically significant increases in frequency or severity of adverse effects between the exposed
24    population and its appropriate control group. Source: IlJS
25    Maximum allowable daily level (MADL): The maximum allowable daily level of a reproductive
26    toxicant at which the chemical would have no observable adverse reproductive effect, assuming
27    exposure at 1,000 times that level. Source: OEHHA|2012}.

28    Maximum contaminant level (MCL): The highest level of a contaminant that is allowed in
29    drinking water. MCLs are set as close to MCLGs as feasible using the best available treatment
30    technology and taking cost into consideration. MCLs are enforceable standards. Source: lISJiPA
31    ClMbi

32    Maximum contaminant level goal (MCLG): The level of a contaminant in drinking water below
3 3    which there is no known or expected risk to health. MCLGs allow for a margin of safety and are
34    nonenforceable public health goals. Source: HJ
35    Minimum risk level (MRL): An ATSDR estimate of daily human exposure to a hazardous substance
36    at or below which the substance is unlikely to pose a measurable risk of harmful (adverse),
                   This document is a draft for review purposes only and does not constitute Agency policy.
      June 2015                                    G-4                  DRAFT—DO NOT CITE OR QUOTE

-------
      Hydraulic Fracturing Drinking Water Assessment                                              Appendix G


 1    noncancerous effects. MRLs are calculated for a route of exposure (inhalation or oral) over a
 2    specified time period (acute, intermediate, or chronic). MRLs should not be used as predictors of
 3    harmful (adverse) health effects.

 4        •   Chronic MRL: Duration of exposure is 365 days or longer.
 5        •   Intermediate MRL: Duration of exposure is >14 to 364 days.
 6        •   Acute MRL: Duration of exposure is 1 to 14 days.
 7    Source :ATSDR_C2009J.

 8    No-observed-adverse-effect level (NOAEL): The highest exposure level at which there are no
 9    biologically significant increases in the frequency or severity of adverse effect between the exposed
10    population and its appropriate  control; some effects may be produced at this level, but they are not
11    considered adverse or precursors of adverse effects. Source: lLS._EPA_£2t)l tc]L

12    Oral slope factor (OSF): An upper-bound, approximating a 95% confidence limit, on the increased
13    cancer risk from a lifetime oral exposure to an agent. This estimate, usually expressed in units of
14    proportion (of a population) affected per mg/kg-day, is generally reserved for use in the low-dose
15    region of the dose-response relationship, that is, for exposures corresponding to risks less than 1 in
16    100. Source:  O
17    Reference dose (RfD) (U.S. EPA IRIS and PPRTV definition): An estimate (with uncertainty
1 8    spanning perhaps an order of magnitude) of a daily oral exposure to the human population
19    (including sensitive subgroups) that is likely to be without an appreciable risk of deleterious effects
20    during a lifetime. It can be derived from a NOAEL, LOAEL, or benchmark dose, with uncertainty
2 1    factors generally applied to reflect limitations of the data used. The RfD is generally used in the
22    EPA's noncancer health assessments.

23         •   Chronic RfD: Duration of exposure is up to a lifetime.
24         •   Subchronic RfD  (sRFD): Duration of exposure is up to 10% of an average lifespan.
2 5    Source : IL
26    Reference dose (RfD) (U.S. EPA HHBP definition): The particular concentration of a chemical
2 7    that is known not to cause health problems. A standard that also may be referred to as the
28    acceptable daily intake. Derived using the same EPA guidance for IRIS and PPRTV RfD
29    determination. Source: yjLEPA_£2015eJ.

30    Tolerable daily intake (TDI): An estimate of the intake of a substance, expressed on a body mass
3 1    basis, to which an individual in a (sub) population may be exposed daily over its lifetime without
3 2    appreciable health risk. Source : WHQX2Q15J.

33    Weight-of-evidence (WOE) characterization for carcinogenicity: A system used for
34    characterizing the extent to which the available data support the hypothesis that an agent causes
3 5    cancer in humans.
                   This document is a draft for review purposes only and does not constitute Agency policy.
      June 2015                                    G-5                   DRAFT—DO NOT CITE OR QUOTE

-------
      Hydraulic Fracturing Drinking Water Assessment                                              Appendix G

 1       •   EPA 1986 guidelines: Under the EPA's 1986 risk assessment guidelines, the WOE was
 2           described by categories "A through E," with Group A for known human carcinogens through
 3           Group E for agents with evidence of noncarcinogenicity. Five standard WOE descriptors
 4           were used:
 5               o  A: Human carcinogen
 6               o  Bl: Probable human carcinogen—based on limited evidence of carcinogenicity in
 7                  humans and sufficient evidence of carcinogenicity in animals
 8               o  B2: Probable human carcinogen—based on sufficient evidence of carcinogenicity in
 9                  animals
10               o  C: Possible human carcinogen
11               o  D: Not classifiable as to human carcinogenicity
12               o  E: Evidence of noncarcinogenicity for humans
13           Source: HSJPAI201M,
14       •   EPA 1996 proposed guidelines: The EPA's 1996 proposed guidelines outlined a major
15           change in the way hazard evidence was weighted in reaching conclusions about the human
16           carcinogenic potential of agents. These guidelines replaced the WOE letter categories with
17           the use of standard descriptors of conclusions incorporated into a brief narrative. Three
18           categories of descriptors with the narrative were used:
19               o  Known/likely
20               o  Cannot be determined
21               o  Not likely
22           Source: USS!_EPAX1996J.
23       •   EPA 1999 guidelines: The 1999 guidelines adopted a framework incorporating hazard
24           identification, dose-response assessment, exposure assessment, and risk characterization
2 5           with an emphasis on characterization of evidence and conclusions in each part of the
26           assessment. Five descriptors summarizing the WOE in the narrative were used:
27               o  Carcinogenic to humans
28               o  Likely to be carcinogenic to humans
29               o  Suggestive evidence of carcinogenicity, but not sufficient to assess human
30                  carcinogenic potential
31               o  Data are inadequate for an assessment of human carcinogenic potential
32               o  Not likely to be carcinogenic to humans
33           Source: US.EPAXI999J.
34       •   EPA 2005 guidelines: The approach outlined in the EPA's 2005 guidelines for carcinogen
3 5           risk assessment considers all scientific information in determining whether and under what
36           conditions an agent may cause cancer in humans and provides a narrative approach to

                   This document is a draft for review purposes only and does not constitute Agency policy.
      June 2015                                   G-6                  DRAFT—DO NOT CITE OR QUOTE

-------
      Hydraulic Fracturing Drinking Water Assessment                                              Appendix G

 1           characterize carcinogenicity rather than categories. Five standard WOE descriptors are
 2           used as part of the narrative:
 3               o   Carcinogenic to humans
 4               o   Likely to be carcinogenic to humans
 5               o   Suggestive evidence of carcinogenic potential
 6               o   Inadequate information to assess carcinogenic potential
 7               o   Not likely to be carcinogenic to humans
 8           Source: Hi£PA£20He],
 9       •   IARC Monographs on the evaluation of carcinogenic risks to humans: The IARC
10           classifies carcinogen risk as a matter of scientific judgement that reflects the strength of the
11           evidence derived from studies in humans, in experimental animals, from mechanistic data,
12           and from other relevant data. Five WOE classifications are used:
13               o   Group 1: Carcinogenic to humans
14               o   Group 2A: Probably carcinogenic to humans
15               o   Group 2B: Possibly carcinogenic to humans
16               o   Group 3: Not classifiable as to its carcinogenicity to humans
17               o   Group 4: Probably not carcinogenic to humans
18           Source: IAKCJ2015J.
19       •   NTP: The NTP describes the results of individual experiments on a chemical agent and
20           notes the strength of the evidence for conclusions regarding each study. Negative results, in
21           which the study animals do not have a greater incidence of neoplasia than control animals,
22           do not necessarily mean that a chemical is not a carcinogen, inasmuch as the experiments
23           are conducted under a limited set of conditions. Positive results demonstrate that a
24           chemical is carcinogenic for laboratory animals under the conditions of the study and
25           indicate that exposure to the chemical has the potential for hazard to humans. For each
26           separate experiment, one of the following five categories is selected to describe the findings.
27           These categories refer to the strength of the experimental evidence and not to potency or
28           mechanism.
29               o   Clear evidence of carcinogenic activity
30               o   Some evidence of carcinogenic activity
31               o   Equivocal evidence of carcinogenic activity
32               o   No evidence of carcinogenic activity
33               o   Inadequate study of carcinogenic activity
34           Source: NTP_C2014a}.
35       •   The RoC is a congressionally mandated, science-based, public health report that identifies
36           agents, substances, mixtures, or exposures (collectively called "substances") in our

                   This document is a draft for review purposes only and does not constitute Agency policy.
      June 2015                                    G-7                  DRAFT—DO NOT CITE OR QUOTE

-------
     Hydraulic Fracturing Drinking Water Assessment                                                Appendix G

1           environment that may potentially put people in the United States at increased risk for
2           cancer. NTP prepares the RoC on behalf of the Secretary of the Health and Human Services.
3           The listing criteria in the RoC Document are:
4                o  Known to be a human carcinogen
5                o  Reasonably anticipated to be a human carcinogen
6
                   This document is a draft for review purposes only and does not constitute Agency policy.
     June 2015                                      G-8                   DRAFT—DO NOT CITE OR QUOTE

-------
Hydraulic Fracturing Drinking Water Assessment
                                                                                               Appendix G
G.3.   Tables
Table G-la. Chemicals reported to be used in hydraulic fracturing fluids, with available federal chronic RfVs and OSFs.

           Chemicals from the FracFocus database are listed first, ranked by IRIS reference dose (RfD). The "--" symbol indicates that no value was available
           from the sources consulted. Additionally, an "x" indicates the availability of usage data from FracFocus               and physicochemical
           properties data from EPI Suite™ (see Appendix C). Italicized chemicals are found in both fracturing fluids and flowback/produced water.
Chemical Name
Acrylamide
Propargyl alcohol
Furfural
Benzene
Dichloromethane
Naphthalene
CASRN
79-06-1
107-19-7
98-01-1
71-43-2
75-09-2
91-20-3
Frac-
Focus
data
available
X
X
X
X
X
X
Physico-
chemical
data
available
X
X
X
X
X
X
IRIS
Chronic
RfDa
(mg/
kg-day)
0.002
0.002
0.003
0.004
0.006
0.02
OSFb
(per
mg/
kg-day)
0.5
-
-
0.015-
0.055
0.002
-
Cancer WOE
character-
ization
"Likely to be
carcinogenic
to humans"
-
-
A
"Likely to be
carcinogenic
in humans"
"Data are
inadequate to
assess human
carcinogenic
potential"
PPRTV
Chronic
RfDa
(mg/
kg-day)
-
-
-
-
-
-
OSFb
(per
mg/
kg-day)
-
-
-
-
-
-
Cancer
WOE
character-
ization
-
-
-
-
-
-
ATSDR
Chronic
oral MRLd
(mg/
kg-day)
0.001
-
-
0.0005
0.06
-
HHBP
Chronic
RfDe
(mg/kg-
day)
-
-
0.01
-
-
-
National Primary
Drinking Water
Regulations
Public
health goal'
(MCLG)
(mg/L)
0
-
-
0
0
-
MCI"
(mg/L)
-
-
-
0.005
0.005
-
June 2015
This document is a draft for review purposes only and does not constitute Agency policy.

                                 G-9
DRAFT—DO NOT CITE OR QUOTE

-------
Hydraulic Fracturing Drinking Water Assessment
                                                                                                    Appendix G
Chemical Name
1,4-D/oxane

Sodium chlorite


Chlorine dioxide

1,3-
Dichloropropene
Bisphenol A
Toluene

Ethylbenzene
CASRN
123-91-1

7758-19-2


10049-04-4

542-75-6
80-05-7
108-88-3

100-41-4
Frac-
Focus
data
available
X

X


X

X
X
X

X
Physico-
chemical
data
available
X






X
X
X

X
IRIS
Chronic
RfDa
(mg/
kg-day)
0.03

0.03


0.03

0.03
0.05
0.08

0.1
OSFb
(per
mg/
kg-day)
0.1

__


__

0.05
-


-
Cancer WOE
character-
izationc
"Likely to be
carcinogenic
to humans"
"Data are
inadequate to
assess human
carcinogen-
icity"
"Data are
inadequate to
assess human
carcinogen-
icity"
"Likely to be a
human
carcinogen"
-
"Inadequate
information to
assess the
carcinogenic
potential"
D
PPRTV
Chronic
RfDa
(mg/
kg-day)
-

..


..

-
-


-
OSFb
(per
mg/
kg-day)
-

..


..

-
-


-
Cancer
WOE
character-
izationc
-

..


..

-
-


-
ATSDR
Chronic
oral MRLd
(mg/
kg-day)
0.1

__


__

0.03
-


-
HHBP
Chronic
RfDe
(mg/kg-
day)
-

..


..

-
-


-
National Primary
Drinking Water
Regulations
Public
health goal'
(MCLG)
(mg/L)
-

1


..

-
-
1

0.7
MCI"
(mg/L)
-

0.8


..

-
-
1

0.7
June 2015
This document is a draft for review purposes only and does not constitute Agency policy.
                                   G-10
DRAFT—DO NOT CITE OR QUOTE

-------
Hydraulic Fracturing Drinking Water Assessment
                                                                                                    Appendix G
Chemical Name
1-Butanol
Cumene
Acetophenone
2-Butoxyethanol

Xylenes

Formaldehyde

Phenol

2-Methyl-l-
propanol
CASRN
71-36-3
98-82-8
98-86-2
111-76-2

1330-20-7

50-00-0

108-95-2

78-83-1
Frac-
Focus
data
available
X
X
X
X

X

X

X

X
Physico-
chemical
data
available
X
X
X
X

X

X

X

X
IRIS
Chronic
RfDa
(mg/
kg-day)
0.1
0.1
0.1
0.1

0.2

0.2

0.3

0.3
OSFb
(per
mg/
kg-day)
-
-
-
-

—

-

-

-
Cancer WOE
character-
izationc
D
D
D
"Not likely to
be carcino-
genic to
humans"
"Data are
inadequate to
assess the
carcinogenic
potential"
Bl
"Data are
inadequate
for an
assessment of
human
carcinogenic
potential"
-
PPRTV
Chronic
RfDa
(mg/
kg-day)
-
-
-
-

—

-

-

-
OSFb
(per
mg/
kg-day)
-
-
-
-

—

-

-

-
Cancer
WOE
character-
izationc
-
-
-
-

—

-

-

-
ATSDR
Chronic
oral MRLd
(mg/
kg-day)
-
-
-
-

0.2

0.2

-

-
HHBP
Chronic
RfDe
(mg/kg-
day)
-
-
-
-

—

-

-

-
National Primary
Drinking Water
Regulations
Public
health goal'
(MCLG)
(mg/L)
-
-
-
-

10

-

-

-
MCI"
(mg/L)
-
-
-
-

10

-

-

-
June 2015
This document is a draft for review purposes only and does not constitute Agency policy.
                                   G-ll
DRAFT—DO NOT CITE OR QUOTE

-------
Hydraulic Fracturing Drinking Water Assessment
                                                                                                    Appendix G
Chemical Name
Acetone
Ethyl acetate
Ethylene glycol
Methanol
Benzole acid
Aniline
Benzyl chloride
(E)-Crotonaldehyde
N,N-Dimethylform
amide
Epichlorohydrin
1,2-Propylene
glycol
CASRN
67-64-1
141-78-6
107-21-1
67-56-1
65-85-0
62-53-3
100-44-7
123-73-9
68-12-2
106-89-8
57-55-6
Frac-
Focus
data
available
X
X
X
X
X
X
X
X
X
X
X
Physico-
chemical
data
available
X
X
X
X
X
X
X
X
X
X
X
IRIS
Chronic
RfDa
(mg/
kg-day)
0.9
0.9
2
2
4
-
-
-
-
-
-
OSFb
(per
mg/
kg-day)
-
-
-
-
-
0.0057
0.17
-
-
0.0099
-
Cancer WOE
character-
izationc
"Data are
inadequate
for an
assessment of
human
carcinogenic
potential"
-
-
-
D
B2
B2
C
-
B2
-
PPRTV
Chronic
RfDa
(mg/
kg-day)
-
-
-
-
-
0.007
0.002
0.001
0.1
0.006
20
OSFb
(per
mg/
kg-day)
-
-
-
-
-
-
-
-
-
-
-
Cancer
WOE
character-
izationc
-
IN
-
-
-
-
-
-
IN
-
NL
ATSDR
Chronic
oral MRLd
(mg/
kg-day)
-
-
-
-
-
-
-
-
-
-
-
HHBP
Chronic
RfDe
(mg/kg-
day)
-
-
-
-
-
-
-
-
-
-
-
National Primary
Drinking Water
Regulations
Public
health goal'
(MCLG)
(mg/L)
-
-
-
-
-
-
-
-
-
0
-
MCI"
(mg/L)
-
-
-
-
-
-
-
-
-
-
-
June 2015
This document is a draft for review purposes only and does not constitute Agency policy.
                                   G-12
DRAFT—DO NOT CITE OR QUOTE

-------
Hydraulic Fracturing Drinking Water Assessment
                                                                                                    Appendix G
Chemical Name
2-(2-Butoxyethoxy)
ethanol
Hexanedioicacid
Quinoline
Ethylenediamine
Formic acid
Sodium chlorate
Quaternary
ammonium
compounds,
benzyl-C12-16-
alkyldimethyl,
chlorides
Benzenesulfonic
acid, C10-16-alkyl
derivs.
Ammonium
phosphate
Didecyldimethylam
monium chloride
2-(Thiocyano
methylthio)benzot
hiazole
CASRN
112-34-5
124-04-9
91-22-5
107-15-3
64-18-6
7775-09-9
68424-85-1
68584-22-5
7722-76-1
7173-51-5
21564-17-0
Frac-
Focus
data
available
X
X
X
X
X
X
X
X
X
X
X
Physico-
chemical
data
available
X
X
X
X
X




X
X
IRIS
Chronic
RfDa
(mg/
kg-day)
-
-
-
-
-
-
-
-
-
-
-
OSFb
(per
mg/
kg-day)
-
-
3
-
-
-
-
-
-
-
-
Cancer WOE
character-
izationc
-
-
"Likely to be
carcinogenic
in humans"
D
-
-
-
-
-
-
-
PPRTV
Chronic
RfDa
(mg/
kg-day)
0.03
2
-
0.09
0.9
-
-
-
49
-
-
OSFb
(per
mg/
kg-day)
-
-
-
-
-
-
-
-
-
-
-
Cancer
WOE
character-
izationc
IN
-
-
IN
IN
-
-
-
IN
-
-
ATSDR
Chronic
oral MRLd
(mg/
kg-day)
-
-
-
-
-
-
-
-
-
-
-
HHBP
Chronic
RfDe
(mg/kg-
day)
-
-
-
-
-
0.03
0.44
0.5
-
0.1
0.01
National Primary
Drinking Water
Regulations
Public
health goal'
(MCLG)
(mg/L)
-
-
-
-
-
-
-
-
-
-
-
MCI"
(mg/L)
-
-
-
-
-
-
-
-
-
-
-
June 2015
This document is a draft for review purposes only and does not constitute Agency policy.
                                   G-13
DRAFT—DO NOT CITE OR QUOTE

-------
Hydraulic Fracturing Drinking Water Assessment
                                                                                                    Appendix G
Chemical Name
Mineral oil -
includes paraffin
oil
Trisodium
phosphate
Triphosphoric acid,
pentasodium salt
Aluminum
Phosphoric acid
Iron
Tricalcium
phosphate
Bis(2-chloroethyl)
ether
Dodecylbenzenesul
fonic acid
Hydrazine
Tetrasodium
pyrophosphate
CASRN
8012-95-1
7601-54-9
7758-29-4
7429-90-5
7664-38-2
7439-89-6
7758-87-4
111-44-4
27176-87-0
302-01-2
7722-88-5
Frac-
Focus
data
available
X
X
X
X
X
X
X
X
X
X
X
Physico-
chemical
data
available







X
X


IRIS
Chronic
RfDa
(mg/
kg-day)
-
-
-
-
-
-
-
-
-
-
-
OSFb
(per
mg/
kg-day)
-
-
-
-
-
-
-
1.1
-
3
-
Cancer WOE
character-
izationc
-
-
-
-
-
-
-
B2
-
B2
-
PPRTV
Chronic
RfDa
(mg/
kg-day)
3
49
49
1
48.6
0.7
49
-
-
-
49
OSFb
(per
mg/
kg-day)
-
-
-
-
-
-
-
-
-
-
-
Cancer
WOE
character-
izationc
IN
IN
IN
IN
IN
IN
IN
-
-
-
IN
ATSDR
Chronic
oral MRLd
(mg/
kg-day)
-
-
-
1
-
-
-
-
-
-
-
HHBP
Chronic
RfDe
(mg/kg-
day)
-
-
-
-
-
-
-
-
0.5
-
-
National Primary
Drinking Water
Regulations
Public
health goal'
(MCLG)
(mg/L)
-
-
-
-
-
-
-
-
-
-
-
MCI"
(mg/L)
-
-
-
-
-
-
-
-
-
-
-
June 2015
This document is a draft for review purposes only and does not constitute Agency policy.
                                   G-14
DRAFT—DO NOT CITE OR QUOTE

-------
Hydraulic Fracturing Drinking Water Assessment
                                                                                                    Appendix G
Chemical Name
Potassium
phosphate, tribasic
Sodium
trimetaphosphate
Arsenic
Phosphine
Acrolein
Chromium (VI)
Di(2-ethylhexyl)
phthalate
Chlorine
Styrene
CASRN
7778-53-2
7785-84-4
7440-38-2
7803-51-2
107-02-8
18540-29-9
117-81-7
7782-50-5
100-42-5
Frac-
Focus
data
available
X
X







Physico-
chemical
data
available




X

X

X
IRIS
Chronic
RfDa
(mg/
kg-day)
-
-
0.0003
0.0003
0.0005
0.003
0.02
0.1
0.2
OSFb
(per
mg/
kg-day)
-
-
1.5
-
-
-
0.014
-
-
Cancer WOE
character-
izationc
-
-
A
D
"Data are
inadequate
for an
assessment of
human
carcinogenic
potential"
A (inhaled);
D(oral)
B2
-
-
PPRTV
Chronic
RfDa
(mg/
kg-day)
49
49
-
-
-
-
-
-
-
OSFb
(per
mg/
kg-day)
-
-
-
-
-
-
-
-
-
Cancer
WOE
character-
izationc
IN
IN
-
-
-
-
-
-
-
ATSDR
Chronic
oral MRLd
(mg/
kg-day)
-
-
0.0003
-
-
0.0009
0.06
-
-
HHBP
Chronic
RfDe
(mg/kg-
day)
-
-
-
-
-
-
-
-
-
National Primary
Drinking Water
Regulations
Public
health goal'
(MCLG)
(mg/L)
-
-
0
-
-
-
0
-
0.1
MCI"
(mg/L)
-
-
0.010
-
-
-
0.006
-
0.1
June 2015
This document is a draft for review purposes only and does not constitute Agency policy.
                                   G-15
DRAFT—DO NOT CITE OR QUOTE

-------
Hydraulic Fracturing Drinking Water Assessment
                                                                                                    Appendix G
Chemical Name
Zinc
Acrylic acid
Chromium (III)
Phthalic anhydride
Cyclohexanone
1,2-Propylene
oxide
2-(2-Ethoxyethoxy)
ethanol
Tributyl phosphate
2-Methoxyethanol
Polyphosphoric
acids, sodium salts
Phosphoric acid,
diammonium salt
CASRN
7440-66-6
79-10-7
16065-83-1
85-44-9
108-94-1
75-56-9
111-90-0
126-73-8
109-86-4
68915-31-1
7783-28-0
Frac-
Focus
data
available











Physico-
chemical
data
available

X

X
X
X
X
X
X


IRIS
Chronic
RfDa
(mg/
kg-day)
0.3
0.5
1.5
2
5
-
-
-
-
-
-
OSFb
(per
mg/
kg-day)
-
-
-
-
-
0.24
-
-
-
-
-
Cancer WOE
character-
izationc
"Inadequate
information to
assess
carcinogenic
potential"
-
"Data are
inadequate
for an
assessment of
human
carcinogenic
potential"
-
-
B2
-
-
-
-
-
PPRTV
Chronic
RfDa
(mg/
kg-day)
-
-
-
-
-
-
0.06
0.01
0.005
49
49
OSFb
(per
mg/
kg-day)
-
-
-
-
-
-
-
0.009
-
-
-
Cancer
WOE
character-
izationc
-
IN
-
-
IN
-
IN
LI
IN
IN
IN
ATSDR
Chronic
oral MRLd
(mg/
kg-day)
0.3
-
-
-
-
-
-
0.08
-
-
-
HHBP
Chronic
RfDe
(mg/kg-
day)
-
-
-
-
-
0.001
-
-
-
-
-
National Primary
Drinking Water
Regulations
Public
health goal'
(MCLG)
(mg/L)
-
-
-
-
-
-
-
-
-
-
-
MCI"
(mg/L)
-
-
-
-
-
-
-
-
-
-
-
June 2015
This document is a draft for review purposes only and does not constitute Agency policy.
                                   G-16
DRAFT—DO NOT CITE OR QUOTE

-------
Hydraulic Fracturing Drinking Water Assessment
Appendix G
Chemical Name
Sodium
pyrophosphate
Phosphoric acid,
aluminium sodium
salt
CASRN
7758-16-9
7785-88-8

Frac-
Focus
data
available



Physico-
chemical
data
available



IRIS
Chronic
RfDa
(mg/
kg-day)
-


OSFb
(per
mg/
kg-day)
-


Cancer WOE
character-
izationc
-


PPRTV
Chronic
RfDa
(mg/
kg-day)
49
49

OSFb
(per
mg/
kg-day)
-


Cancer
WOE
character-
izationc
IN
IN

ATSDR
Chronic
oral MRLd
(mg/
kg-day)
-


HHBP
Chronic
RfDe
(mg/kg-
day)
-


National Primary
Drinking Water
Regulations
Public
health goal'
(MCLG)
(mg/L)
-


MCI*
(mg/L)
-


ATSDR = Agency for Toxic Substances and Disease Registry; CASRN = Chemical Abstract Service Registry Number; IRIS = Integrated Risk Information System; PPRTV = Provisional
Peer Reviewed Toxicity Values; HHBP = Human Health Benchmarks for Pesticides
a Reference dose (RfD) (IRIS and PPRTV definition): An estimate (with uncertainty spanning perhaps an order of magnitude) of a daily oral exposure to the human population
(including sensitive subgroups) that is likely to be without an appreciable risk of deleterious effects during a lifetime. It can be derived from a no observed-adverse-effect level
(NOAEL), lowest observed-adverse-effect level (LOAEL), or benchmark dose (BMD), with uncertainty factors generally applied to reflect limitations of the data used. The RfD is
generally used in the EPA's noncancer health assessments. Chronic RfD:  Duration of exposure is up to a lifetime.
b Oral slope factor (OSF): An upper-bound, approximating a 95% confidence limit, on the increased cancer risk from a lifetime oral exposure to an agent. This estimate, usually
expressed in units of proportion (of a population) affected per mg/kg-day,  is generally reserved for use in the low dose region of the dose response relationship, that is, for
exposures corresponding to risks less than 1 in 100.
c Weight of evidence (WOE) characterization for carcinogenicity: A system used for characterizing the  extent to which the available data support the hypothesis that an agent
causes cancer in humans. See glossary for details.
d Minimum risk level (MRL): An ATSDR estimate of daily human  exposure to a hazardous substance at  or below which the substance  is unlikely to pose a measurable risk of
harmful (adverse), noncancerous effects. MRLs are calculated for a route of exposure (inhalation or oral) over a specified time  period (acute, intermediate, or chronic). MRLs
should not be used as predictors of harmful (adverse) health effects. Chronic MRL: Duration of exposure is 365 days or longer.
e Reference dose (RfD) (HHBP definition): The particular concentration of a chemical that is known not to cause health problems. A standard that also may be referred
to as the acceptable daily intake. Derived using the same EPA guidance for RfD determination.
f Maximum contaminant level goal  (MCLG): The level of a contaminant in drinking water below which  there is no  known or expected risk to health. MCLGs allow for a margin of
safety and are nonenforceable public health goals.
g Maximum contaminant level  (MCL): The highest level of a contaminant that is allowed in drinking water. MCLs are set as close to MCLGs as feasible using the best available
treatment technology and taking cost into consideration. MCLs are enforceable standards.
                                       This document is a draft for review purposes only and does not constitute Agency policy.
June 2015                                                                   G-17                                            DRAFT—DO NOT CITE OR QUOTE

-------
Hydraulic Fracturing Drinking Water Assessment
Appendix G
Table G-lb. Chemicals reported to be used in hydraulic fracturing fluids, with available state
           chronic RfVs and OSFs.

           Chemicals from the FracFocus database are listed first, ranked by California EPA maximum allowable
           daily level (MADL). The "--" symbol indicates that no value was available from the sources consulted.
           Additionally, an "x" indicates the availability of usage data from FracFocus                and
           physicochemical properties data from EPI Suite™ (see Appendix C). Italicized chemicals are found in
           both fracturing fluids and flowback/produced water.
Chemical name
Ethylene oxide
Benzene
N-Methyl-2-pyrrolidone
Acrylamide
Aniline
Benzyl chloride
1,4-Dioxane
Epichlorohydrin
Ethylbenzene
Nitrilotriacetic acid
Nitrilotriacetic acid trisodium
monohydrate
Thiourea
Bis(2-chloroethyl) ether
1,3-Butadiene
Hydrazine
1,3-Dichloropropene
Dichloromethane
Lead
Chromium (VI)
2-Methoxyethanol
2-Ethoxyethanol
CASRN
75-21-8
71-43-2
872-50-4
79-06-1
62-53-3
100-44-7
123-91-1
106-89-8
100-41-4
139-13-9
18662-53-8
62-56-6
111-44-4
106-99-0
302-01-2
542-75-6
75-09-2
7439-92-1
18540-29-9
109-86-4
110-80-5
FracFocus
data
available
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X




Physico-
chemical
data
available
X
X
X
X
X
X
X
X
X
X
X
X
X
X

X
X


X
X
California
Oral MADLa
(u,g/day)
20
24
17000
140
-
-
-
-
-
-
-
-
-
-
-
-
-
0.5
8.2
63
750
OSFb (per
mg/kg-day)
0.31
0.1
-
4.5
0.0057
0.17
0.027
0.08
0.011
0.0053
0.01
0.072
2.5
0.6
3
0.091
0.014
0.0085
0.5
-
-
               This document is a draft for review purposes only and does not constitute Agency policy.

June 2015                                        G-18                    DRAFT—DO NOT CITE OR QUOTE

-------
Hydraulic Fracturing Drinking Water Assessment
Appendix G
Chemical name
Di(2-ethylhexyl) phthalate
1,2-Propylene oxide
Arsenic
CASRN
117-81-7
75-56-9
7440-38-2
FracFocus
data
available



Physico-
chemical
data
available
X
X

California
Oral MADLa
(Hg/day)
20 (neonate male)
58 (infant male)
410 (adult)
-
-
OSFb (per
mg/kg-day)
0.003
0.24
9.5
 a Maximum allowable daily level (MADL): The maximum allowable daily level of a reproductive toxicant at which the chemical
 would have no observable adverse reproductive effect, assuming exposure at 1,000 times that level.
 b Oral slope factor (OSF): An upper-bound, approximating a 95% confidence limit, on the increased cancer risk from a lifetime
 oral exposure to an agent. This estimate, usually expressed in units of proportion (of a population) affected per mg/kg day, is
 generally reserved for use in the low-dose region of the dose-response relationship, that is, for exposures corresponding to
 risks less than 1 in 100.
Table G-lc. Chemicals reported to be used in hydraulic fracturing fluids, with available
            international chronic RfVs and OSFs.
            Chemicals from the FracFocus database are listed first, ranked by CICAD reference dose (TDI, or
            tolerable daily intake). An "x" indicates the availability of usage data from FracFocus (U^
            and physicochemical properties data from EPI Suite™ (see Appendix C). Italicized chemicals are found
            in both fracturing fluids and flowback/produced water.
Chemical name
D-Limonene
Potassium iodide
Sodium iodide
Copper(l) iodide
Glyoxal
Ethylene glycol
N-Methyl-2-pyrrolidone
Strontium chloride
Chromium (VI)
CASRN
5989-27-5
7681-11-0
7681-82-5
7681-65-4
107-22-2
107-21-1
872-50-4
10476-85-4
18540-29-9
FracFocus data
available
X
X
X
X
X
X
X


Physicochemical
data available
X



X
X
X


IPCS Chronic TDIa
(mg/kg-day)
0.1
0.01
0.01
0.01
0.2
0.05
0.6
0.13
0.0009
 IPCS = International Programme on Chemical Safety; CICAD = Concise International Chemical Assessment Documents
 a Tolerable daily intake (TDI): An estimate of the intake of a substance, expressed on a body mass basis, to which an individual
 in a (sub) population may be exposed daily over its lifetime without appreciable health risk.
                This document is a draft for review purposes only and does not constitute Agency policy.
June 2015                                           G-19                     DRAFT—DO NOT CITE OR QUOTE

-------
Hydraulic Fracturing Drinking Water Assessment
Appendix G
Table G-ld. Chemicals reported to be used in hydraulic fracturing fluids, with available less-
           than-chronic RfVs and OSFs.

           Chemicals from the FracFocus database are listed first, ranked by PPRTV subchronic reference dose
           (sRfD). The "--" symbol indicates that no value was available from the sources consulted. Additionally,
           an "x" indicates the availability of usage data from FracFocus                and physicochemical
           properties data from EPI Suite™ (see Appendix C). Italicized chemicals are found in both fracturing
           fluids and flowback/produced water.
Chemical name
Benzyl chloride
Epichlorohydrin
(E)-Crotonaldehyde
Benzene
Ethylbenzene
Ethylenediamine
N,N-
Dimethylformamide
2-(2-
Butoxyethoxy)ethanol
Hexane
Xylenes
Antimony trioxide
Iron
Toluene
Formic acid
Hexanedioicacid
Benzoicacid
1,2-Propylene glycol
Mineral oil - includes
paraffin oil
Phosphoric acid
Ammonium phosphate
Trisodium phosphate
Triphosphoric acid,
pentasodium salt
CASRN
100-44-7
106-89-8
123-73-9
71-43-2
100-41-4
107-15-3
68-12-2
112-34-5
110-54-3
1330-20-7
1309-64-4
7439-89-6
108-88-3
64-18-6
124-04-9
65-85-0
57-55-6
8012-95-1
7664-38-2
7722-76-1
7601-54-9
7758-29-4
FracFocus
data
available
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
Physico-
chemical data
available
X
X
X
X
X
X
X
X
X
X


X
X
X
X
X





PPRTV
sRfDa
(mg/kg-day)
0.002
0.006
0.01
0.01
0.05
0.2
0.3
0.3
0.3
0.4
0.5
0.7
0.8
0.9
2
4
20
30
48.6
49
49
49
ATSDR
Acute oral
MRLb
(mg/kg-day)
-
-
-
-
-
-
-
-
-
1
-
-
0.8
-
-
-
-
-
-
-
-
-
Intermediate
oral MRLC
(mg/kg-day)
-
-
-
-
0.4
-
-
-
-
0.4
-
-
0.02
-
-
-
-
-
-
-
-
-
               This document is a draft for review purposes only and does not constitute Agency policy.

June 2015                                        G-20                    DRAFT—DO NOT CITE OR QUOTE

-------
Hydraulic Fracturing Drinking Water Assessment
Appendix G
Chemical name
Tricalcium phosphate
Tetrasodium
pyrophosphate
Potassium phosphate,
tribasic
Sodium
trimetaphosphate
Acrylamide
1,4-Dioxane
Ethylene glycol
Naphthalene
Phenol
Sodium chlorite
Acetone
2-Butoxyethanol
Aluminum
Formaldehyde
1,3-Dichloropropene
Dichloromethane
Antimony trichloride
2-Methoxyethanol
Tributyl phosphate
Acrylic acid
2-(2-Ethoxyethoxy)
ethanol
Cyclohexanone
Polyphosphoric acids,
sodium salts
Phosphoric acid,
diammonium salt
Sodium
pyrophosphate
CASRN
7758-87-4
7722-88-5
7778-53-2
7785-84-4
79-06-1
123-91-1
107-21-1
91-20-3
108-95-2
7758-19-2
67-64-1
111-76-2
7429-90-5
50-00-0
542-75-6
75-09-2
10025-91-9
109-86-4
126-73-8
79-10-7
111-90-0
108-94-1
68915-31-1
7783-28-0
7758-16-9
FracFocus
data
available
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X









Physico-
chemical data
available




X
X
X
X
X

X
X

X
X
X

X
X
X
X
X



PPRTV
sRfDa
(mg/kg-day)
49
49
49
49
-
-
-
-
-
-
-
-
-
-
-
-
0.0004
0.02
0.03
0.2
0.6
2
49
49
49
ATSDR
Acute oral
MRLb
(mg/kg-day)
-
-
-
-
0.01
5
0.8
0.6
1
-
-
0.4
-
-
-
0.2
-
-
1.1
-
-
-
-
-
-
Intermediate
oral MRLC
(mg/kg-day)
-
-
-
-
0.001
0.5
0.8
0.6
-
0.1
2
0.07
1
0.3
0.04
-
-
-
0.08
-
-
-
-
-
-
                This document is a draft for review purposes only and does not constitute Agency policy.
June 2015                                          G-21                      DRAFT—DO NOT CITE OR QUOTE

-------
Hydraulic Fracturing Drinking Water Assessment
Appendix G
Chemical name
Phosphoric acid,
aluminium sodium salt
Acrolein
Di(2-ethylhexyl)
phthalate
Styrene
Arsenic
Chromium (VI)
Copper
Zinc
CASRN
7785-88-8
107-02-8
117-81-7
100-42-5
7440-38-2
18540-29-9
7440-50-8
7440-66-6
FracFocus
data
available








Physico-
chemical data
available

X
X
X




PPRTV
sRfDa
(mg/kg-day)
49
-
-
-
-
-
-
-
ATSDR
Acute oral
MRLb
(mg/kg-day)
-
-
-
0.1
0.005
-
0.01
-
Intermediate
oral MRLC
(mg/kg-day)
-
0.004
0.1
-
-
0.005
0.01
0.3
 a Reference dose (RfD): An estimate (with uncertainty spanning perhaps an order of magnitude) of a daily oral exposure to the
 human population (including sensitive subgroups) that is likely to be without an appreciable risk of deleterious effects during
 a lifetime. It can be derived from a no observed-adverse-effect level (NOAEL), lowest observed-adverse-effect level (LOAEL),
 or benchmark dose (BMD), with uncertainty factors generally applied to reflect limitations of the data used. The RfD is
 generally used in the EPA's noncancer health assessments. Subchronic RfD (sRFD): Duration of exposure is up to 10% of an
 average lifespan.
 b Minimum risk level (MRL): An ATSDR estimate of daily human exposure to a hazardous substance at or below which the
 substance is unlikely to pose a measurable risk of harmful (adverse), noncancerous effects. MRLs are calculated for a route of
 exposure (inhalation or oral) over a specified time period (acute, intermediate, or chronic). MRLs should not be used as
 predictors of harmful (adverse) health effects. Acute MRL: Duration of exposure is 1 to 14 days.
 c Minimum risk level (MRL): An ATSDR estimate of daily human exposure to a hazardous substance at or below which the
 substance is unlikely to pose a measurable risk of harmful (adverse), noncancerous effects. MRLs are calculated for a route of
 exposure (inhalation or oral) over a specified time period (acute, intermediate, or chronic). MRLs should not be used as
 predictors of harmful (adverse) health effects. Intermediate MRL: Duration of exposure is >14 to 364 days.
                 This document is a draft for review purposes only and does not constitute Agency policy.

June 2015                                              G-22                       DRAFT—DO NOT CITE OR QUOTE

-------
Hydraulic Fracturing Drinking Water Assessment
                                                                                              Appendix G
Table G-2a. Chemicals reported to be detected in flowback or produced water, with available federal chronic RfVs and OSFs.
           Chemicals are ranked by IRIS reference dose (RfD). The "--" symbol indicates that no value was available from the sources consulted. Additionally,
           an "x" indicates the availability of measured concentration data in flowback or produced water (see Appendix E) and physicochemical properties
           data from EPI Suite™ (see Appendix C). Italicized chemicals are found in both fracturing fluids and flowback/produced water.
Chemical Name
Heptachlor
epoxide
Phosphorus
Aldrin
Dieldrin
Arsenic
Lindane
Antimony
Acrolein
Cadmium
Heptachlor
CASRN
1024-57-3
7723-14-0
309-00-2
60-57-1
7440-38-2
58-89-9
7440-36-0
107-02-8
7440-43-9
76-44-8
Concen-
tration
data
available

X


X

X

X

Physico-
chemical
data
available
X

X
X

X

X

X
IRIS
Chronic
RfDa (mg/
kg-day)
0.000013
0.00002
0.00003
0.00005
0.0003
0.0003
0.0004
0.0005
0.0005
(water)
0.0005
OSFb
(per mg/
kg-day)
9.1
-
17
16
1.5
-
-
-
-
4.5
Cancer WOE
character-
ization"
B2
D
B2
B2
A
-
-
"Data are
inadequate for
an assessment
of human
carcinogenic
potential"
Bl
B2
PPRTV
Chronic
RfDa (mg/
kg-day)
-
-
-
-
-
-
-
-
-
-
OSFb
(per mg/
kg-day)
-
-
-
-
-
-
-
-
-
-
Cancer
WOE
character-
ization"
-
-
-
-
-
-
IN
-
-
-
ATSDR
Chronic
oral MRLd
(mg/
kg-day)
-
-
0.00003
0.00005
0.0003
-
-
-
0.0001
-
HHBP
Chronic
RfDe
(mg/kg-
day)
-
-
-
-
-
-
-
-
-
-
National Primary
Drinking Water
Regulations
Public
health
goal'
(MCLG)
(mg/L)
0
-
-
-
0
0.0002
0.006
-
0.005
0
MCLs
(mg/L)
0.0002
-
-
-
0.010
0.0002
0.006
-
0.005
0.0004
June 2015
This document is a draft for review purposes only and does not constitute Agency policy.
                                 G-23
DRAFT—DO NOT CITE OR QUOTE

-------
Hydraulic Fracturing Drinking Water Assessment
                                                                                                    Appendix G







Chemical Name


Cyanide


Pyridine
Methyl bromide
Beryllium
Chromium (VI)

Benzene


2-Methylnaphth
alene


Molybdenum
Silver
Selenium

Dichloromethane








CASRN


57-12-5


110-86-1
74-83-9
7440-41-7
18540-29-9

71-43-2


91-57-6



7439-98-7
7440-22-4
7782-49-2

75-09-2





Concen-
tration
data
available





X

X


X


X



X
X
X







Physico-
chemical
data
available


X


X
X



X


X







X



IRIS


Chronic
RfDa (mg/
kg-day)


0.0006


0.001
0.0014
0.002
0.003

0.004


0.004



0.005
0.005
0.005

0.006



OSFb
(per mg/
kg-day)


-


-
-
-
-
0 015-

0.055





-
-
-

0.002



Cancer WOE
character-
ization
"Inadequate
information to
assess the
carcinogenic
potential"
-
D
Bl
A (inhaled);
D(oral)

A

"Data are
inadequate to
assess human

carcinogenic
potential"
-
D
D
"Likely to be
carcinogenic in
humans"


PPRTV


Chronic
RfDa (mg/
kg-day)


-


-
-
-
-

	






-
-
-

-



OSFb
(per mg/
kg-day)


-


-
-
-
-

	






-
-
-

-


Cancer
WOE
character-
ization


-


-
-
-
-

	






-
-
-

-



ATSDR

Chronic
oral MRLd
(mg/
kg-day)


-


-
-
0.002
0.0009

0.0005


0.04



-
-
0.005

0.06



HHBP

Chronic
RfDe
(mg/kg-
day)


-


-
0.02
-
-

	






-
-
-

-

National Primary
Drinking Water
Regulations
Public
health
goal'
(MCLG)
(mg/L)


0.2


-
-
0.004
-

0






-
-
0.05

0




MCLs
(mg/L)


0.2


-
-
0.004
-

0.005






-
-
0.05

0.005

June 2015
This document is a draft for review purposes only and does not constitute Agency policy.
                                   G-24
DRAFT—DO NOT CITE OR QUOTE

-------
Hydraulic Fracturing Drinking Water Assessment
                                                                                                    Appendix G
Chemical Name
1,2,4-
Trichlorobenzene
Tetrachloroethyl
ene
Chloroform
Di(2-ethylhexyl)
phthalate
Naphthalene
2,4-
Dimethylphenol
Chlorodibromom
ethane
Bromoform
Bromodichlorom
ethane
Diphenylamine
1,4-Dioxane
CASRN
120-82-1
127-18-4
67-66-3
117-81-7
91-20-3
105-67-9
124-48-1
75-25-2
75-27-4
122-39-4
123-91-1
Concen-
tration
data
available


X
X
X
X



X
X
Physico-
chemical
data
available
X
X
X
X
X
X
X
X
X
X
X
IRIS
Chronic
RfDa (mg/
kg-day)
0.01
0.006
0.01
0.02
0.02
0.02
0.02
0.02
0.02
0.025
0.03
OSFb
(per mg/
kg-day)
-
0.0021
-
0.014
-
-
0.084
0.0079
0.062
-
0.1
Cancer WOE
character-
ization
D
"Likely to be
carcinogenic in
humans"
B2
B2
"Data are
inadequate to
assess human
carcinogenic
potential"
-
C
B2
B2
-
"Likely to be
carcinogenic to
humans"
PPRTV
Chronic
RfDa (mg/
kg-day)
-
-
-
-
-
-
-
-
-
-
-
OSFb
(per mg/
kg-day)
0.029
-
-
-
-
-
-
-
-
-
-
Cancer
WOE
character-
ization
LI
-
-
-
-
IN
-
-
-
IN
-
ATSDR
Chronic
oral MRLd
(mg/
kg-day)
0.1
0.008
0.01
0.06
-
-
0.09
0.02
0.02
-
0.1
HHBP
Chronic
RfDe
(mg/kg-
day)
-
-
-
-
-
-
-
-
-
0.1
-
National Primary
Drinking Water
Regulations
Public
health
goal'
(MCLG)
(mg/L)
0.07
0
-
0
-
-
-
-
-
-
-
MCLs
(mg/L)
0.07
0.005
-
0.006
-
-
-
-
-
-
-
June 2015
This document is a draft for review purposes only and does not constitute Agency policy.
                                   G-25
DRAFT—DO NOT CITE OR QUOTE

-------
Hydraulic Fracturing Drinking Water Assessment
                                                                                                    Appendix G
Chemical Name
Pyrene
Fluoranthene
Fluorene
m-Cresol
o-Cresol
Toluene
Chlorine
Ethylbenzene
Cumene
Acetophenone
Carbon disulfide
Dibutyl phthalate
CASRN
129-00-0
206-44-0
86-73-7
108-39-4
95-48-7
108-88-3
7782-50-5
100-41-4
98-82-8
98-86-2
75-15-0
84-74-2
Concen-
tration
data
available
X
X
X
X
X
X

X
X
X
X
X
Physico-
chemical
data
available
X
X
X
X
X
X

X
X
X
X
X
IRIS
Chronic
RfDa (mg/
kg-day)
0.03
0.04
0.04
0.05
0.05
0.08
0.1
0.1
0.1
0.1
0.1
0.1
OSFb
(per mg/
kg-day)
-
-
-
-
-
-
-
-
-
-
-
-
Cancer WOE
character-
ization
D
D
D
C
C
"Inadequate
information to
assess the
carcinogenic
potential"
-
D
D
D
-
D
PPRTV
Chronic
RfDa (mg/
kg-day)
-
-
-
-
-
-
-
-
-
-
-
-
OSFb
(per mg/
kg-day)
-
-
-
-
-
-
-
-
-
-
-
-
Cancer
WOE
character-
ization
-
IN
-
-
IN
-
-
-
-
-
-
-
ATSDR
Chronic
oral MRLd
(mg/
kg-day)
-
-
-
-
-
-
-
-
-
-
-
-
HHBP
Chronic
RfDe
(mg/kg-
day)
-
-
-
-
-
-
-
-
-
-
-
-
National Primary
Drinking Water
Regulations
Public
health
goal'
(MCLG)
(mg/L)
-
-
-
-
-
1
-
0.7
-
-
-
-
MCLs
(mg/L)
-
-
-
-
-
1
-
0.7
-
-
-
-
June 2015
This document is a draft for review purposes only and does not constitute Agency policy.
                                   G-26
DRAFT—DO NOT CITE OR QUOTE

-------
Hydraulic Fracturing Drinking Water Assessment
                                                                                                    Appendix G







Chemical Name
Nitrite
Manganese


Xylenes



Barium



Boron




Zinc




Phenol








CASRN
14797-65-0
7439-96-5


1330-20-7



7440-39-3



7440-42-8




7440-66-6




108-95-2





Concen-
tration
data
available
X
X


X



X



X




X




X





Physico-
chemical
data
available




X

















X



IRIS


Chronic
RfDa (mg/
kg-day)
0.1
0.14


0.2



0.2



0.2




0.3




0.3



OSFb
(per mg/
kg-day)
-
-


-



-



-




-




—



Cancer WOE
character-
ization
-
D
"Data are
inadequate to
assess the
carcinogenic
potential"
"Not likely to
be carcinogenic
to humans"
"Data are
inadequate to
assess the
carcinogenic
potential"
"Inadequate
information to
assess
carcinogenic
potential"
"Data are
inadequate to
assess human
carcinogenicity
rr


PPRTV


Chronic
RfDa (mg/
kg-day)
-
-


-



-



-




-




-



OSFb
(per mg/
kg-day)
-
-


-



-



-




-




-


Cancer
WOE
character-
ization
-
-


-



-



-




-




—



ATSDR

Chronic
oral MRLd
(mg/
kg-day)
-
-


0.2



0.2



-




0.3




—



HHBP

Chronic
RfDe
(mg/kg-
day)
-
-


-



-



-




-




-

National Primary
Drinking Water
Regulations
Public
health
goal'
(MCLG)
(mg/L)
1
-


10



2



-




-




—




MCLs
(mg/L)
1
-


10



2



-




-




—

June 2015
This document is a draft for review purposes only and does not constitute Agency policy.
                                   G-27
DRAFT—DO NOT CITE OR QUOTE

-------
Hydraulic Fracturing Drinking Water Assessment
                                                                                                    Appendix G
Chemical Name
Strontium
Methyl ethyl
ketone
Diethyl phthalate
Acetone
Chromium (III)
Nitrate
Ethylene glycol
Methanol
1,2-Propylene
glycol
Formic acid
Aluminum
CASRN
7440-24-6
78-93-3
84-66-2
67-64-1
16065-83-1
14797-55-8
107-21-1
67-56-1
57-55-6
64-18-6
7429-90-5
Concen-
tration
data
available
X


X

X




X
Physico-
chemical
data
available

X
X
X


X
X
X
X

IRIS
Chronic
RfDa (mg/
kg-day)
0.6
0.6
0.8
0.9
1.5
1.6
2
2
-
-
-
OSFb
(per mg/
kg-day)
-
-
-
-
-
-
-
-
-
-
-
Cancer WOE
character-
ization
-
"Data are
inadequate to
assess
carcinogenic
potential"
D
"Data are
inadequate to
assess human
carcinogen/city
rr
"Data are
inadequate to
assess human
carcinogenicity
rr
-
-
-
-
-
-
PPRTV
Chronic
RfDa (mg/
kg-day)
-
-
-
-
-
-
-
-
20
0.9
1
OSFb
(per mg/
kg-day)
-
-
-
-
-
-
-
-
-
-
-
Cancer
WOE
character-
ization
-
-
-
-
-
-
-
-
NL
IN
IN
ATSDR
Chronic
oral MRLd
(mg/
kg-day)
-
-
-
-
-
-
-
-
-
-
1
HHBP
Chronic
RfDe
(mg/kg-
day)
-
-
-
-
-
-
-
-
-
-
-
National Primary
Drinking Water
Regulations
Public
health
goal'
(MCLG)
(mg/L)
-
-
-
-
-
10
-
-
-
-
-
MCLs
(mg/L)
-
-
-
-
-
10
-
-
-
-
-
June 2015
This document is a draft for review purposes only and does not constitute Agency policy.
                                   G-28
DRAFT—DO NOT CITE OR QUOTE

-------
Hydraulic Fracturing Drinking Water Assessment
                                                                                                    Appendix G
Chemical Name
Iron
Bis(2-chloroethyl)
ether
Benzyl alcohol
Butylbenzene
Acrylonitrile
Phorate
beta-Hexachloro
cyclohexane
Benzo(a)pyrene
p,p'-DDE
Lithium
CASRN
7439-89-6
111-44-4
100-51-6
104-51-8
107-13-1
298-02-2
319-85-7
50-32-8
72-55-9
7439-93-2
Concen-
tration
data
available
X

X




X

X
Physico-
chemical
data
available

X
X
X
X
X
X
X
X

IRIS
Chronic
RfDa (mg/
kg-day)
-
-
-
-
-
-
-
-
-
-
OSFb
(per mg/
kg-day)
-
1.1
-
-
0.54
-
1.8
7.3
0.34
-
Cancer WOE
character-
ization
-
B2
-
-
Bl
-
C
B2
B2
-
PPRTV
Chronic
RfDa (mg/
kg-day)
0.7
-
0.1
0.05
-
-
-
-
-
0.002
OSFb
(per mg/
kg-day)
-
-
-
-
-
-
-
-
-
-
Cancer
WOE
character-
ization
IN
-
IN
IN
-
-
-
-
-
IN
ATSDR
Chronic
oral MRLd
(mg/
kg-day)
-
-
-
-
0.04
-
-
-
-
-
HHBP
Chronic
RfDe
(mg/kg-
day)
-
-
-
-
-
0.0005
-
-
-
-
National Primary
Drinking Water
Regulations
Public
health
goal'
(MCLG)
(mg/L)
-
-
-
-
-
-
-
0
-
-
MCLs
(mg/L)
-
-
-
-
-
-
-
0.0002
-
-
June 2015
This document is a draft for review purposes only and does not constitute Agency policy.
                                   G-29
DRAFT—DO NOT CITE OR QUOTE

-------
Hydraulic Fracturing Drinking Water Assessment
                                                                                                            Appendix G
Chemical Name
Cobalt
Vanadium
N-Nitrosodiphen
ylamine
CASRN
7440-48-4
7440-62-2
86-30-6
Concen-
tration
data
available
X
X
X
Physico-
chemical
data
available


X
IRIS
Chronic
RfDa (mg/
kg-day)
-
-
-
OSFb
(per mg/
kg-day)
-
-
0.0049
Cancer WOE
character-
ization
-
-
B2
PPRTV
Chronic
RfDa (mg/
kg-day)
0.0003
0.00007
-
OSFb
(per mg/
kg-day)
-
-
-
Cancer
WOE
character-
ization
LI
IN
-
ATSDR
Chronic
oral MRLd
(mg/
kg-day)
-
-
-
HHBP
Chronic
RfDe
(mg/kg-
day)
-
-
-
National Primary
Drinking Water
Regulations
Public
health
goal'
(MCLG)
(mg/L)
-
-
-
MCLs
(mg/L)
-
-
-
 ATSDR = Agency for Toxic Substances and Disease Registry; CASRN = Chemical Abstract Service Registry Number; IRIS = Integrated Risk Information System; PPRTV = Provisional
 Peer Reviewed Toxicity Values; HHBP = Human Health Benchmarks for Pesticides
 a Reference dose (RfD) (IRIS and PPRTV definition): An estimate (with uncertainty spanning perhaps an order of magnitude) of a daily oral exposure to the human population
 (including sensitive subgroups) that is likely to be without an appreciable risk of deleterious effects during a lifetime. It can be derived from a no observed-adverse-effect level
 (NOAEL), lowest observed-adverse-effect level (LOAEL), or benchmark dose (BMD), with uncertainty factors generally applied to reflect limitations of the data used. The RfD is
 generally used in the EPA's noncancer health assessments. Chronic RfD: Duration of exposure is up to a lifetime.
 b Oral slope factor (OSF): An upper-bound, approximating a 95% confidence limit, on the increased cancer risk from a lifetime oral exposure to an agent. This estimate, usually
 expressed in units of proportion (of a population) affected per mg/kg day, is generally reserved for use in the low dose region of the dose response relationship, that is, for
 exposures corresponding to risks less than 1 in 100.
 c Weight of evidence (WOE) characterization for carcinogenicity: A system used for characterizing the extent to which the available data support the hypothesis that an agent
 causes cancer in humans. See glossary for details.
 d Minimum risk level (MRL): An ATSDR estimate of daily human exposure to a hazardous substance at or below which the substance is unlikely to pose a measurable risk of
 harmful  (adverse), noncancerous effects. MRLs are calculated for a route of exposure (inhalation or oral) over a specified time period (acute, intermediate, or chronic). MRLs
 should not be used as predictors of harmful (adverse) health effects. Chronic MRL: Duration of exposure is 365 days or longer.
 e Reference dose (RfD) (HHBP definition): The particular concentration of a chemical that is known not to cause health problems. A standard that also may be referred to as the
 acceptable daily intake. Derived using the same  EPA guidance for RfD determination.
 f Maximum contaminant level goal (MCLG): The  level of a contaminant in drinking water below which there is no known or expected risk to health. MCLGs allow for a margin of
 safety and are nonenforceable public health goals.
 g Maximum contaminant level (MCL): The highest level of a contaminant that is allowed in drinking water. MCLs are set as close to MCLGs as feasible using the best available
 treatment technology and taking cost into consideration. MCLs are enforceable standards.
June 2015
This document is a draft for review purposes only and does not constitute Agency policy.

                                     G-30
DRAFT—DO NOT CITE OR QUOTE

-------
Hydraulic Fracturing Drinking Water Assessment
Appendix G
Table G-2b. Chemicals reported to be detected in flowback or produced water, with available
           state chronic RfVs and OSFs.

           Chemicals are ranked by California EPA maximum allowable daily level (MADL). The "--" symbol
           indicates that no value was available from the sources consulted. Additionally, an "x" indicates the
           availability of measured concentration data in flowback or produced water (see Appendix E) and
           physicochemical properties data from EPI Suite™ (see Appendix C). Italicized chemicals are found in
           both fracturing fluids and flowback/produced water.
Chemical name
Lead
Cadmium
Chromium (VI)
Dibutyl phthalate
Benzene
Acrylonitrile
1,4-Dioxane
Ethylbenzene
Di(2-ethylhexyl) phthalate
Arsenic
Bis(2-chloroethyl) ether
Heptachlor epoxide
1,2,4-Trichlorobenzene
Tetrachloroethylene
lndeno(l,2,3-cd)pyrene
Benzo(b)fluoranthene
Benzo(k)fluoranthene
Aldrin
beta-Hexachlorocyclohexane
Benzo(a)pyrene
Dibenz(a,h)anthracene
7,12-Dimethylbenz(a)anthracene
Lindane
Dieldrin
Chloroform
CASRN
7439-92-1
7440-43-9
18540-29-9
84-74-2
71-43-2
107-13-1
123-91-1
100-41-4
117-81-7
7440-38-2
111-44-4
1024-57-3
120-82-1
127-18-4
193-39-5
205-99-2
207-08-9
309-00-2
319-85-7
50-32-8
53-70-3
57-97-6
58-89-9
60-57-1
67-66-3
Concen-
tration
data
available
X
X

X
X

X
X
X
X




X
X
X


X
X



X
Physico-
chemical
data
available



X
X
X
X
X
X

X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
California
Oral MADLa
(u,g/day)
0.5
4.1
8.2
8.7
24
-
-
-
20 (neonate male)
58 (infant male)
410 (adult)
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
OSFb (per
mg/kg-day)
0.0085
15
0.5
-
0.1
1
0.027
0.011
0.003
9.5
2.5
5.5
0.0036
0.051
1.2
1.2
1.2
17
1.5
2.9
4.1
250
1.1
16
0.019
               This document is a draft for review purposes only and does not constitute Agency policy.

June 2015                                       G-31                    DRAFT—DO NOT CITE OR QUOTE

-------
Hydraulic Fracturing Drinking Water Assessment
Appendix G
Chemical name
p,p'-DDE
Bromoform
Bromodichloromethane
Heptachlor
N-Nitrosodiphenylamine
Safrole
Dichloromethane
CASRN
72-55-9
75-25-2
75-27-4
76-44-8
86-30-6
94-59-7
75-09-2
Concen-
tration
data
available




X


Physico-
chemical
data
available
X
X
X
X
X
X
X
California
Oral MADLa
(Hg/day)
-
-
-
-
-
-
-
OSFb (per
mg/kg-day)
0.34
0.011
0.13
4.1
0.009
0.22
0.014
 3 Maximum allowable daily level (MADL): The maximum allowable daily level of a reproductive toxicant at which the chemical
 would have no observable adverse reproductive effect, assuming exposure at 1,000 times that level.
 b Oral slope factor (OSF): An upper-bound, approximating a 95% confidence limit, on the increased cancer risk from a lifetime
 oral exposure to an agent. This estimate, usually expressed in units of proportion (of a population) affected per mg/kg day, is
 generally reserved for use in the low-dose region of the dose-response relationship, that is, for exposures corresponding to
 risks less than 1 in 100.
                 This document is a draft for review purposes only and does not constitute Agency policy.
June 2015                                             G-32                      DRAFT—DO NOT CITE OR QUOTE

-------
Hydraulic Fracturing Drinking Water Assessment
Appendix G
Table G-2c. Chemicals reported to be detected in flowback or produced water, with available
           international chronic RfVs and OSFs.

           Chemicals are ranked by CICAD reference dose (TDI -Tolerable Daily Intake). An "x" indicates the
           availability of measured concentration data in flowback or produced water (see Appendix E) and
           physicochemical properties data from EPI Suite™ (see Appendix C). Italicized chemicals are found in
           both fracturing fluids and flowback/produced water.
Chemical name
Heptachlor
Strontium
Chloroform
Mercury
Barium
Beryllium
Ethylene glycol
Tetrachloroethene
Chromium (VI)
Diethyl phthalate
CASRN
76-44-8
7440-24-6
67-66-3
7439-97-6
7440-39-3
7440-41-7
107-21-1
127-18-4
18540-29-9
84-66-2
Concentration
data available

X
X
X
X
X




Physicochemical
data available
X

X



X
X

X
IPCS Chronic TDIa
(mg/kg-day)
0.0001
0.13
0.015
0.002
0.02
0.002
0.05
0.05
0.0009
5
 IPCS = International Programme on Chemical Safety; CICAD = Concise International Chemical Assessment Documents
 a Tolerable Daily Intake (TDI): An estimate of the intake of a substance, expressed on a body mass basis, to which an individual
 in a (sub) population may be exposed daily over its lifetime without appreciable health risk.
                This document is a draft for review purposes only and does not constitute Agency policy.

June 2015                                         G-33                     DRAFT—DO NOT CITE OR QUOTE

-------
Hydraulic Fracturing Drinking Water Assessment
Appendix G
Table G-2d. Chemicals reported to be detected in flowback or produced water, with available
           less-than-chronic RfVs and OSFs.

           Chemicals are ranked by PPRTV subchronic reference dose (sRfD). The "--" symbol indicates that no
           value was available from the sources consulted. Additionally, an "x" indicates the availability of
           measured concentration data in flowback or produced water (see Appendix E) and physicochemical
           properties data from EPI Suite™ (see Appendix C). Italicized chemicals are found in both fracturing
           fluids and flowback/produced water.
Chemical name
Aldrin
Antimony
Vanadium
Lithium
Cobalt
2-Methylnaphthalene
Methyl bromide
Bromodichloromethane
1,2,3-Trichlorobenzene
Benzene
p-Cresol
Bromoform
Ethylbenzene
2,4-Dimethylphenol
Chlorodibromomethane
1,2,4-Trichlorobenzene
Butylbenzene
Benzyl alcohol
Pyrene
Xylenes
Iron
Toluene
Formic acid
1,2-Propylene glycol
CASRN
309-00-2
7440-36-0
7440-62-2
7439-93-2
7440-48-4
91-57-6
74-83-9
75-27-4
87-61-6
71-43-2
106-44-5
75-25-2
100-41-4
105-67-9
124-48-1
120-82-1
104-51-8
100-51-6
129-00-0
1330-20-7
7439-89-6
108-88-3
64-18-6
57-55-6
Concen-
tration
data
available

X
X
X
X
X



X
X

X
X



X
X
X
X
X


Physico-
chemical data
available
X




X
X
X
X
X
X
X
X
X
X
X
X
X
X
X

X
X
X
PPRTV
sRfDa
(mg/kg-day)
0.00004
0.0004
0.0007
0.002
0.003
0.004
0.005
0.008
0.008
0.01
0.02
0.03
0.05
0.05
0.07
0.09
0.1
0.3
0.3
0.4
0.7
0.8
0.9
20
ATSDR
Acute oral
MRLb
(mg/kg-day)
0.002
-
-
-
-
-
-
0.04
-
-
-
0.7
-
-
0.1
-
-
-
-
1
-
0.8
-
-
Intermediate
oral MRLC
(mg/kg-day)
-
-
0.01
-
0.01
-
0.003
-
-
-
-
0.2
0.4
-
-
0.1
-
-
-
0.4
-
0.02
-
-
               This document is a draft for review purposes only and does not constitute Agency policy.

June 2015                                        G-34                    DRAFT—DO NOT CITE OR QUOTE

-------
Hydraulic Fracturing Drinking Water Assessment
Appendix G
Chemical name
Acrolein
1,4-Dioxane
Ethylene glycol
Di(2-ethylhexyl)
phthalate
Naphthalene
Phenol
Acetone
Arsenic
Chromium (VI)
Copper
Zinc
Aluminum
Acrylonitrile
Dioctyl phthalate
Tetrachloroethylene
Fluoranthene
beta-
Hexachlorocyclohexane
Lindane
Dieldrin
Chloroform
Strontium
Tin
Barium
Boron
Cadmium
Carbon disulfide
Heptachlor
Phosphorus
CASRN
107-02-8
123-91-1
107-21-1
117-81-7
91-20-3
108-95-2
67-64-1
7440-38-2
18540-29-9
7440-50-8
7440-66-6
7429-90-5
107-13-1
117-84-0
127-18-4
206-44-0
319-85-7
58-89-9
60-57-1
67-66-3
7440-24-6
7440-31-5
7440-39-3
7440-42-8
7440-43-9
75-15-0
76-44-8
7723-14-0
Concen-
tration
data
available

X

X
X
X
X
X

X
X
X

X

X



X
X
X
X
X
X
X

X
Physico-
chemical data
available
X
X
X
X
X
X
X





X
X
X
X
X
X
X
X





X
X

PPRTV
sRfDa
(mg/kg-day)
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
0.1
-
-
-
-
-
-
-
-
-
-
-
-
ATSDR
Acute oral
MRLb
(mg/kg-day)
-
5
0.8
-
0.6
1
-
0.005
-
0.01
-
-
0.1
3
0.008
-
0.05
0.003
-
0.3
-
-
-
0.2
-
0.01
0.0006
-
Intermediate
oral MRLC
(mg/kg-day)
0.004
0.5
0.8
0.1
0.6
-
2
-
0.005
0.01
0.3
1
0.01
0.4
0.008
0.4
0.0006
0.00001
0.0001
0.1
2
0.3
0.2
0.2
0.0005
-
0.0001
0.0002
                This document is a draft for review purposes only and does not constitute Agency policy.
June 2015                                          G-35                     DRAFT—DO NOT CITE OR QUOTE

-------
Hydraulic Fracturing Drinking Water Assessment
                                                                                               Appendix G
Chemical name
Diethyl phthalate
Dibutyl phthalate
Fluorene
Dichloromethane
CASRN
84-66-2
84-74-2
86-73-7
75-09-2
Concen-
tration
data
available

X
X

Physico-
chemical data
available
X
X
X
X
PPRTV
sRfDa
(mg/kg-day)
-
-
-
-
ATSDR
Acute oral
MRLb
(mg/kg-day)
7
0.5
-
0.2
Intermediate
oral MRLC
(mg/kg-day)
6
-
0.4
-
 a Reference dose (RfD): An estimate (with uncertainty spanning perhaps an order of magnitude) of a daily oral exposure to the
 human population (including sensitive subgroups) that is likely to be without an appreciable risk of deleterious effects during
 a lifetime. It can be derived from a no observed-adverse-effect  level (NOAEL), lowest observed-adverse-effect level (LOAEL),
 or benchmark dose (BMD), with uncertainty factors generally applied to reflect limitations of the data used. The RfD is
 generally used in the EPA's noncancer health assessments. Subchronic RfD (sRFD): Duration of exposure is up to 10% of an
 average lifespan.
 b Minimum risk level (MRL): An ATSDR estimate of daily human  exposure to a hazardous substance at or below which the
 substance is unlikely to pose a measurable risk of harmful (adverse), noncancerous effects. MRLs are calculated for a route of
 exposure (inhalation or oral) over a specified time period (acute, intermediate, or chronic). MRLs should not be used as
 predictors of harmful (adverse) health effects. Acute MRL: Duration of exposure is 1 to 14 days.
 c Minimum risk level (MRL): An ATSDR estimate of daily human  exposure to a hazardous substance at or below which the
 substance is unlikely to pose a measurable risk of harmful (adverse), noncancerous effects. MRLs are calculated for a route of
 exposure (inhalation or oral) over a specified time period (acute, intermediate, or chronic). MRLs should not be used as
 predictors of harmful (adverse) health effects. Intermediate MRL: Duration of exposure is >14 to 364 days.
G.4.   References for Appendix G
ATSDR (Agency for Toxic Substances and Disease Registry). (2009). Glossary of terms. Available online at
   http://www.atsdr.cdc.gov/glossary.html
IARC (International Agency for Research on Cancer). (2015). IARC monographs - Classifications. Available
   online at http://monographs.iarc.fr/ENG/Classification/index.php
NTP (National Toxicology Program). (2014a). Definition of carcinogenicity results. Available online at
   http://ntp.niehs.nih.gov/results/pubs/longterm/defs/index.html
NTP (National Toxicology Program). (2014b). Report on carcinogens. Thirteenth edition. Research Triangle
   Park, NC: U.S. Department of Health and Human Services, Public Health Service.
   http://ntp.niehs.nih.gov/pubhealth/roc/rocl3/index.html
                                                                                    25701 (2012).
OEHHA. Title 27. California Code of Regulations Article 8. No Observable Effect Levels.
   http://www.oehha.ca.gov/prop65/law/pdf zip/RegsArt8.pdf
U.S. EPA (U.S. Environmental Protection Agency). (1996). Proposed guidelines for carcinogen risk assessment
   [EPA Report]. (EPA/600/P-92/003C). Washington, DC: U.S. Environmental Protection Agency, Risk
   Assessment Forum.
U.S. EPA (U.S. Environmental Protection Agency). (1999). Guidelines for carcinogen risk assessment [review
   draft] [EPA Report]. (NCEA-F-0644). Washington, DC.
   http://www.epa.gov/raf/publications/pdfs/CANCER GLS.PDF
                This document is a draft for review purposes only and does not constitute Agency policy.

June 2015                                         G-36                     DRAFT—DO NOT CITE OR QUOTE

-------
Hydraulic Fracturing Drinking Water Assessment                                                  Appendix C
U.S. EPA (U.S. Environmental Protection Agency). (2011a). Plan to study the potential impacts of hydraulic
   fracturing on drinking water resources [EPA Report]. (EPA/600/R-11/122). Washington, DC: U.S.
   Environmental Protection Agency, Office of Research and Development.
   http://www2.epa.gov/hfstudy/plan-study-potential-impacts-hydraulic-fracturing-drinking-water-
   resources-epa600r-11122

U.S. EPA (U.S. Environmental Protection Agency). (2011c). Terminology services (TS): Vocabulary catalog -
   IRIS glossary. Available online at
   http://ofmpub.epa.gov/sor internet/registry/termreg/searchandretrieve/glossariesandkeywordlists/se
   arch.do?details=&glossaryName=IRIS%20Glossary (accessed May 21, 2015).

U.S. EPA (U.S. Environmental Protection Agency). (2012c). Study of the potential impacts of hydraulic
   fracturing on drinking water resources: Progress report. (EPA/601/R-12/011). Washington, DC: U.S.
   Environmental Protection Agency, Office of Research and Development.
   http://nepis.epa.gov/exe/ZyPURL.cgi?Dockey=P100FH8M.txt

U.S. EPA (U.S. Environmental Protection Agency). (2014b). Drinking water contaminants. Available online at
   http://water.epa.gov/drink/contaminants/

U.S. EPA (U.S. Environmental Protection Agency). (2015a). Analysis of hydraulic fracturing fluid data from the
   FracFocus chemical disclosure registry 1.0 [EPA Report]. (EPA/601/R-14/003). Washington, D.C.: Office
   of Research and Development, U.S. Environmental Protection Agency.
   http://www2.epa.gov/hfstudy/analysis-hydraulic-fracturing-fluid-data-fracfocus-chemical-disclosure-
   registry-1-pdf

U.S. EPA (U.S. Environmental Protection Agency). (2015e). Human health benchmarks for pesticides.
   Available online at http://iaspub.epa.gov/apex/pesticides/f?p=HHBP:HOME

WHO (World Health Organization). (2015). Concise international chemical assessment documents. Available
   online at http://www.who.int/ipcs/publications/cicad/en/
               This document is a draft for review purposes only and does not constitute Agency policy.

June 2015                                       G-37                    DRAFT—DO NOT CITE OR QUOTE

-------
Hydraulic Fracturing Drinking Water Assessment                             Appendix H
                     Appendix H
Description of EPA Hydraulic Fracturing Study
Publications Cited in This Assessment
         This document is a draft for review purposes only and does not constitute Agency policy.

June 2015                                    DRAFT—DO NOT CITE OR QUOTE

-------
Hydraulic Fracturing Drinking Water Assessment
                                                                     Appendix H
Appendix  H.  Description of  EPA Hydraulic Fracturing
    Study  Publications Cited in This Assessment
Table H-l. Titles, descriptions, and citations for EPA hydraulic fracturing study publications
          cited in this assessment.
 Research project
Description
Citations
Analysis of existing data
 Literature Review
Review and assessment of existing
papers and reports, focusing on
peer-reviewed literature
Literature review is incorporated into this document.
 Spills Database
 Analysis
Characterization of hydraulic
fracturing-related spills using
information obtained from
selected state and industry data
sources
U.S. EPA (U.S. Environmental Protection Agency).
(2015). Review of state and industry spill data:
characterization of hydraulic fracturing-related spills
[EPA Report]. (EPA/601/R-14/001). Washington, D.C.:
Office of Research and Development, U.S.
Environmental Protection Agency.
 Service Company
 Analysis
Analysis of information provided
by nine hydraulic fracturing service
companies in response to a
September 2010 information
request on hydraulic fracturing
operations
Analysis of data received is incorporated into this
document.1
 Well File Review
Analysis of information provided
by nine oil and gas operators in
response to an August 2011
information request for 350 well
files
U.S. EPA (U.S. Environmental Protection Agency).
(2015). Review of well operator files for hydraulically
fractured oil and gas production wells: Well design and
construction [EPA Report]. (EPA/601/R-14/002).
Washington, D.C.: Office of Research and Development,
U.S. Environmental Protection Agency.
Analysis of data received is also incorporated into this
document.2
1 Data received and incorporated into this document is cited as: U.S. EPA (U.S. Environmental Protection Agency). (2013).
Data received from oil and gas exploration and production companies, including hydraulic fracturing service companies
2011 to 2013. Non-confidential business information source documents are located in Federal Docket ID: EPA-HQ-
ORD2010-0674. Available at MsiUMMMim^M^SS&gSI.
2 Data received and incorporated into this document is cited as: U.S. EPA (U.S. Environmental Protection Agency). (2011).
Sampling data for flowback and produced water provided to EPA by nine oil and gas well operators (non-confidential
business information). US Environmental Protection Agency.
               This document is a draft for review purposes only and does not constitute Agency policy.
June 2015                                       H-l                   DRAFT—DO NOT CITE OR QUOTE

-------
Hydraulic Fracturing Drinking Water Assessment
Appendix H
Research project
FracFocus Analysis
Description
Analysis of data compiled from
FracFocus 1.0, the national
hydraulic fracturing chemical
registry operated by the Ground
Water Protection Council and the
Interstate Oil and Gas Compact
Commission
Citations
U.S. EPA (U.S. Environmental Protection Agency).
(2015). Analysis of hydraulic fracturing fluid data from
the FracFocus chemical disclosure registry 1.0 [EPA
Report]. (EPA/601/R- 14/003). Washington, D.C.: Office
of Research and Development, U.S. Environmental
Protection Agency, http://www2.epa.gov/hfstudy/
analysis-hydraulic-fracturing-fluid-data-fracfocus-
chemical-disclosure-registry-1-pdf
U.S. EPA (U.S. Environmental Protection Agency).
(2015). Analysis of hydraulic fracturing fluid data from
the FracFocus chemical disclosure registry 1.0: project
database. Washington, D.C.: U.S. Environmental
Protection Agency, Office of Research and
Development.
U.S. EPA (U.S. Environmental Protection Agency).
(2015). Analysis of hydraulic fracturing fluid data from
the FracFocus chemical disclosure registry 1.0: Data
management and quality assessment report [EPA
Report]. (EPA/601/R- 14/006). Washington, D.C.: U.S.
Environmental Protection Agency, Office of Research
and Development, http://www2.epa.gov/sites/
production/files/2015-03/documents/fracfocus data
management report final 032015 508.pdf

                This document is a draft for review purposes only and does not constitute Agency policy.
June 2015                                          H-2                      DRAFT—DO NOT CITE OR QUOTE

-------
Hydraulic Fracturing Drinking Water Assessment
                                                                           Appendix H
 Research project
Description
                                                    Citations
 Scenario evaluations
 Subsurface
 Migration
 Modeling
Numerical modeling of subsurface
fluid migration scenarios that
explore the potential for fluids,
including liquids and gases to move
from the fractured zone to drinking
water aquifers
                                                    Kim, J; Moridis, GJ. (2013). Development of the T+M
                                                    coupled flow-geomechanical simulator to describe
                                                    fracture propagation and coupled flow-thermal-
                                                    geomechanical processes in tight/shale gas systems.
                                                    Computers and Geosciences 60: 184-198.
                                                    http://dx.doi.0rg/10.1016/i.cageo.2013.04.023
                                                     Kim, J; Moridis, GJ. (In Press). Numerical analysis of
                                                     fracture propagation during hydraulic fracturing
                                                     operations in shale gas systems. International Journal of
                                                     Rock Mechanics and Mining Sciences.
                                                     Kim, J; Urn, ES; Moridis, GJ. (2014). Fracture
                                                     Propagation, Fluid Flow, and Geomechanics of Water-
                                                     Based Hydraulic Fracturing in Shale Gas Systems and
                                                     Electromagnetic Geophysical  Monitoring of Fluid
                                                     Migration. SPE Hydraulic Fracturing Technology
                                                     Conference, The Woodlands, Texas, USA.
                                                     http://dx.doi.org/10.2118/168578-MS
                                                     Reagan, MT; Moridis, GJ; Johnson, JN; Keen, ND. (2015).
                                                     Numerical simulation of the environmental impact of
                                                     hydraulic fracturing of tight/shale gas reservoirs on
                                                     near-surface groundwater: background, base cases,
                                                     shallow reservoirs, short-term gas and water transport.
                                                     Water Resour Res 51: 1-31. http://dx.doi.org/10.1002/
                                                     2014WR016086
                                                     Rutqvist, J; Rinaldi, AP; Cappa, F; Moridis, GJ. (2013).
                                                     Modeling of fault reactivation and induced seismicity
                                                     during hydraulic fracturing of shale-gas reservoirs.
                                                     Journal of Petroleum Science and Engineering 107: 31-
                                                     44. http://dx.doi.0rg/10.1016/i.petrol.2013.04.023
                                                     Rutqvist, J; Rinaldi, AP; Cappa, F; Moridis, GJ. (2015).
                                                     Modeling of fault activation and seismicity by injection
                                                     directly into a fault zone associated with hydraulic
                                                     fracturing of shale-gas reservoirs. Journal of Petroleum
                                                     Science and Engineering 127: 377-386.
                                                     http://dx.doi.0rg/10.1016/i.petrol.2015.01.019
Surface Water
Modeling
                    Modeling of concentrations of
                    selected chemicals at public water
                    supplies downstream from
                    wastewater treatment facilities
                    that discharge treated hydraulic
                    fracturing wastewater to surface
                    waters
                                 Weaver, JW; Xu, J; Mravik, SC. (In Press) Scenario
                                 analysis of the impact on drinking water intakes from
                                 bromide in the discharge of treated oil and gas waste
                                 water. J Environ Eng.
                This document is a draft for review purposes only and does not constitute Agency policy.

June 2015                                          H-3                      DRAFT—DO NOT CITE OR QUOTE

-------
Hydraulic Fracturing Drinking Water Assessment
                                                                          Appendix H
 Research project
Description
                                                   Citations
Water Availability
Modeling
                   Assessment and modeling of
                   current and future scenarios
                   exploring the impact of water
                   usage for hydraulic fracturing on
                   drinking water availability in the
                   Upper Colorado River Basin and
                   the Susquehanna River Basin
                                 U.S. EPA (U.S. Environmental Protection Agency).
                                 (2015). Case study analysis of the impacts of water
                                 acquisition for hydraulic fracturing on local water
                                 availability [EPA Report]. (EPA/600/R-14/179).
                                 Washington,  D.C.
 Laboratory studies
 Source
 Apportionment
 Studies
Identification and quantification of
the source(s) of high bromide and
chloride concentrations at public
water supply intakes downstream
from wastewater treatment plants
discharging treated hydraulic
fracturing wastewater to surface
waters
                                                   U.S. EPA (U.S. Environmental Protection Agency).
                                                   (2015). Sources contributing bromide and inorganic
                                                   species to drinking water intakes on the Allegheny river
                                                   in western Pennsylvania [EPA Report]. (EPA/600/R-
                                                   14/430). Washington, D.C.
 Analytical Method
 Development
Development of analytical
methods for selected chemicals
found in hydraulic fracturing fluids
or wastewater
                                                   DeArmond, PD; DiGoregorio, AL (2013).
                                                   Characterization of liquid chromatography-tandem
                                                   mass spectrometry method for the determination of
                                                   acrylamide in complex environmental samples. Anal
                                                   Bioanal Chem 405: 4159-4166. '
                                                    DeArmond, PD; DiGoregorio, AL (2013). Rapid liquid
                                                    chromatography-tandem mass spectrometry-based
                                                    method for the analysis of alcohol ethoxylates and
                                                    alkylphenol ethoxylates in environmental samples. J
                                                    Chromatogr A 1305:154-163. httfiI//d)tdoLo£g/
                This document is a draft for review purposes only and does not constitute Agency policy.

June 2015                                          H-4                     DRAFT—DO NOT CITE OR QUOTE

-------
Hydraulic Fracturing Drinking Water Assessment
                                                                           Appendix H
 Research project
Description
Citations
 Analytical Method
 Development
 (cont.)
Development of analytical
methods for selected chemicals
found in hydraulic fracturing fluids
or wastewater (cont.)
U.S. EPA (U.S. Environmental Protection Agency).
(2014). Development of rapid radiochemical method for
gross alpha and gross beta activity concentration in
flowback and produced waters from hydraulic
fracturing operations [EPA Report].  (EPA/600/R-
14/107). Washington, D.C. http://www2.epa.gov/
hfstudy/development-rapid-radiochemical-method-
                                                     gross-alpha-and-gross-beta-activity-concentration
                                                     U.S. EPA (U.S. Environmental Protection Agency).
                                                     (2014). The verification of a method for detecting and
                                                     quantifying diethylene glycol, triethylene glycol,
                                                     tetraethylene glycol, 2-butoxyethanol and 2-
                                                     methoxyethanol in ground and surface waters [EPA
                                                     Report].  (EPA/600/R-14/008). Washington, D.C.
                                                     http://www2.epa.gov/hfstudv/verification-method-
                                                     detecting-and-quantifving-diethvlene-glycol-
                                                     triethylene-glycol
 Retrospective case studies
 Investigations of whether reported drinking water impacts may be associated with or caused by hydraulic
 fracturing activities
 Las Animas and
 Huerfano
 Counties, Colorado
Investigation of potential drinking
water impacts from coalbed
methane extraction in the Raton
Basin
U.S. EPA (U.S. Environmental Protection Agency).
(2015). Retrospective case study in the Raton Basin,
Colorado: study of the potential impacts of hydraulic
fracturing on drinking water resources [EPA Report].
(EPA 600/R-14/091). Washington, D.C.
 Dunn County,
 North Dakota
Investigation of potential drinking
water impacts from a well blowout
during hydraulic fracturing for oil
in the Bakken Shale
U.S. EPA (U.S. Environmental Protection Agency).
(2015). Retrospective case study in Killdeer, North
Dakota: study of the potential impacts of hydraulic
fracturing on drinking water resources [EPA Report].
(EPA 600/R-14/103). Washington, D.C.
 Bradford County,
 Pennsylvania
Investigation of potential drinking
water impacts from shale gas
development in the Marcellus
Shale
U.S. EPA (U.S. Environmental Protection Agency).
(2014). Retrospective case study in northeastern
Pennsylvania: study of the potential impacts of
hydraulic fracturing on drinking water resources [EPA
Report]. (EPA 600/R-14/088). Washington, D.C.
 Washington
 County,
 Pennsylvania
Investigation of potential drinking
water impacts from shale gas
development in the Marcellus
Shale
U.S. EPA (U.S. Environmental Protection Agency).
(2015). Retrospective case study in southwestern
Pennsylvania: study of the potential impacts of
hydraulic fracturing on drinking water resources [EPA
Report]. (EPA 600/R-14/084). Washington, D.C.
                This document is a draft for review purposes only and does not constitute Agency policy.
June 2015                                          H-5                      DRAFT—DO NOT CITE OR QUOTE

-------
Hydraulic Fracturing Drinking Water Assessment
                                                                           Appendix H
 Research project
Description
Citations
 Wise County,
 Texas
Investigation of potential drinking
water impacts from shale gas
development in the Barnett Shale
U.S. EPA (U.S. Environmental Protection Agency).
(2015). Retrospective case study in Wise County, Texas:
study of the potential impacts of hydraulic fracturing on
drinking water resources [EPA Report]. (EPA 600/R-
14/090). Washington, D.C.
                This document is a draft for review purposes only and does not constitute Agency policy.

June 2015                                          H-6                     DRAFT—DO NOT CITE OR QUOTE

-------
Hydraulic Fracturing Drinking Water Assessment                                      Appendix I
                            Appendix I
Unit Conversions
           This document is a draft for review purposes only and does not constitute Agency policy.
June 2015                                              DRAFT—DO NOT CITE OR QUOTE

-------
     Hydraulic Fracturing Drinking Water Assessment
                                                                                  Appendix I
     Appendix I.   Unit Conversions
     LENGTH
 2
 3
 4

 5
 6

 7
 8
     1 in (inch)



     1 ft (foot)


     1 mi (mile)
2.54 cm (centimeters)
25.4 mm (millimeters)
25,400 |im (microns)

0.3048m (meters)
30.48cm

5,280 ft
1,609.344m
1.6093 km (kilometers)
10   AREA
11   1 ft2 (square foot)
     1 acre
12
13
14
15
16   1 mi2
17
18
0.0929 m2 (square meters)

43,560 ft2
0.0016 mi2 (square miles)
0.4047 ha (hectares)
4,046.825 m2

639.9974 ac
258.9988 ha
2.5899 km2 (square kilometers)
19   MASS
20   lib (pound)
21

22   1 ton (short ton, U.S.)
23
24
                                            453.5924 g (grams)
                                            0.4536 kg (kilograms)

                                            2,000 Ibs
                                            907.185kg
                                            0.9072 metric tons
                  This document is a draft for review purposes only and does not constitute Agency policy.
     June 2015                                   1-1                  DRAFT—DO NOT CITE OR QUOTE

-------
      Hydraulic Fracturing Drinking Water Assessment
                                                                                 Appendix I
      VOLUME OR CAPACITY (LIQUID MEASURE)
 2
 3

 4
 5
 6
 7
10
11
12
13

14
15
1 bbl (barrel)


Igal
1 Mgal (million gallons)

1ft3
1 mi3 (cubic mile)
42 gal (gallons, U.S.)
158.9873 L (liters)

231 in3 (cubic inches)
0.1337 ft3 (cubic feet)
3.7854 L
0.0039 m3 (cubic meters)
3.7854 x 10-9 Mm3 (million cubic meters)

1.3368x105 ft3

1,728 in3
7.4805 gal
28.3169 L
0.0283m3

4.1682 km3 (cubic kilometers)
16    CONCENTRATION
17
18
19
20
21
22
1 mg/L (milligram per liter)
1.0 x IQ-6 kg/L (kilograms per liter)
1.0 x IQ-3 g/L (grams per liter)
1,000 |ig/L (micrograms per liter)
1.001 ppm (parts per million)
8.3454 x IQ-6 Ib/gal (pounds per gallon)
6.2428 x lO-5 lb/ft3 (pounds per cubic foot)
23

24
25
SPEED
1 mi/hr (mile per hour)
1.4666 ft/s (feet per second)
0.4470 m/s (meters per second)
26

27
28
DENSITY
Ig/mL
1,000 g/L
1.0 x 106mg/L
                   This document is a draft for review purposes only and does not constitute Agency policy.
      June 2015                                    1-2                   DRAFT—DO NOT CITE OR QUOTE

-------
      Hydraulic Fracturing Drinking Water Assessment
                                                                                     Appendix I
      VOLUME PER UNIT TIME
      1 ft3/s (cubic foot per second)
2
3
4
5
 6    1 ft3/day (cubic feet per day)
 7
10    1 bbl/day (barrel per day)
11
448.8312 gpm (gallons per minute)
 0.6163 Mgpd (million gallons per day)
28.3169 L/s (liters per second)
0.0283 m3/s (cubic meters per second)

0.0052 gpm
7.4805 gpd
0.0283 m3/d (cubic meters per day)

42 gpd
158.9873 L/d (liters per day)
12
     PRESSURE
13    1 psi (pound per square inch)
14
                                             6,894.7573 Pa (pascals)
                                             0.068 atm (standard atmospheres)
15    RADIATION
16   Activity
17    1 Ci (curie)

18    1 Bq (becquerel)
19

20    IpCi
21
22
                                             3.7 x 1010 decays per second

                                             2.703 x ID-" Ci
                                             27.027 pCi (picocuries)

                                             0.037 Bq
                                             0.037 decays per second
                                             2.22 decays per minute
2 3    Exposure
24    1 rem (roentgen equivalent in man)

25    ISv
                                             O.OlSv(sieverts)

                                             1 J/kg (joule per kilogram)
26

27
28
29
30
31
     ELECTRIC CONDUCTANCE
     1 S (siemen)
1 fl-1 (reciprocal of resistance)
1 A/V (ampere per volt)
1 kg-1 • nr2 • s3 • A2 (second cubed- ampere squared
per kilogram-square meter)
1.0 x 106 [j.S (microsiemens)
                   This document is a draft for review purposes only and does not constitute Agency policy.

      June 2015                                    1-3                  DRAFT—DO NOT CITE OR QUOTE

-------
     Hydraulic Fracturing Drinking Water Assessment                                                 Appendix I


1    TEMPERATURE	

2    [°F (degrees, Fahrenheit) - 32] x 5/9  =      °C (degrees, Celsius)

3    PERMEABILITY	

4    1cm2                               =      1.Ox 10-4 m2
5                                        «      1.0 x 108D (darcys)

6    ID                                 *      1.0xlO-i2m2
7                                        =      1,000 mD (millidarcys)
8                                               1.0 x 106 |iD (microdarcys)
                   This document is a draft for review purposes only and does not constitute Agency policy.
     June 2015                                      1-4                    DRAFT—DO NOT CITE OR QUOTE

-------
Hydraulic Fracturing Drinking Water Assessment                                         Appendix J
                              Appendix J
Glossary
            This document is a draft for review purposes only and does not constitute Agency policy.
June 2015                                                 DRAFT—DO NOT CITE OR QUOTE

-------
      Hydraulic Fracturing Drinking Water Assessment                                              Appendix J
      Appendix J.   Glossary
      J.I.   Glossary Terms and Definitions
 1    Acid mine drainage: Flow of water from areas that have been mined for coal or other mineral ores.
 2    The water has a low pH because of its contact with sulfur-bearing material and is harmful to
 3    aquatic organisms. (U.S. EPA, 2013d]

 4    Additive: A single chemical or chemical mixture designed to serve a specific purpose in the
 5    hydraulic fracturing fluid.1

 6    Adsorption: Adhesion of molecules of gas, liquid, or dissolved solids to a surface. (U.S. EPA. 2013d]

 7    Advection: A mechanism for moving chemicals in flowing water, where a chemical moves along
 8    with the flow of the water itself.

 9    Aeration: A process that promotes biological degradation of organic matter in water. The process
10    may be passive (as when waste is exposed to air] or active (as when a mixing or bubbling device
11    introduces the air]. (U.S. EPA. 2013d]

12    Aerobic mesophiles: Microorganisms that use oxygen for energy production and are tolerant of
13    moderate temperatures.

14    Analyte: The element, ion, or compound that an analysis seeks to identify; the compound of
15    interest fU.S. EPA. 2013dl

16    Annulus: Refers to either the space between the casing of a well and the wellbore or the space
17    between any two strings of tubing or casing. (U.S. EPA. 2013d]

18    API number: A unique identifying number for all oil and gas wells drilled in the United States. The
19    system was developed by the American Petroleum Institute. (Oil and Gas Mineral Services, 2010]

2 0    Aquifer: An underground geological formation, or group of formations, containing water. A source
21    of ground water for wells and springs. (U.S. EPA. 2013d]

22    Base fluid: The fluid into which additives and proppants are mixed to formulate a hydraulic
23    fracturing fluid.

24    Basin: A depression in the crust of the earth, caused by plate tectonic activity and subsidence, in
25    which sediments accumulate. Sedimentary basins vary from bowl-shaped to elongated troughs.
26    Basins can be bounded by faults. Rift basins are commonly symmetrical; basins along continental
2 7    margins tend to be asymmetrical. If rich hydrocarbon source rocks occur in combination with
2 8    appropriate depth and duration of burial, then a petroleum system can develop within the basin.
      1 Definitions that have no associated citation in this glossary were developed for this assessment.

                   This document is a draft for review purposes only and does not constitute Agency policy.
      June 2015                                    J-l                   DRAFT—DO NOT CITE OR QUOTE

-------
      Hydraulic Fracturing Drinking Water Assessment                                              Appendix J


 1    Most basins contain some amount of shale, thus providing opportunities for shale gas exploration
 2    and production. [Schlumberger. 2014]

 3    Biogenic: Methane that is produced in shallower formations by bacterial activity in anaerobic
 4    conditions. It is the ultimate dissimilation product of microbially mediated reactions of organic
 5    molecules.

 6    Blowout preventer (BOP): Casinghead equipment that prevents the uncontrolled flow of oil, gas,
 7    and mud from the well by closing around the drill pipe or sealing the hole. [Oil and Gas Mineral
 8    Services. 2010]

 9    Brackish water: Mixed fresh and salt waters. Used here to qualitatively refer to water that contains
10    higher total dissolved solids (TDS] than that typically used for fresh drinking water.

11    BTEX: An acronym for benzene, toluene, ethylbenzene, and xylenes. These chemicals are a group of
12    single ringed aromatic hydrocarbon based on the benzene structure. These compounds are found in
13    petroleum and are of specific importance because of their health effects.

14    Caliper log: A log that is used to check for any wellbore irregularities. It is run prior to primary
15    cementing as a means of calculating the amount of cement needed. Also run in conjunction with
16    other open hole logs for log corrections or run on cased holes to evaluate metal loss. [NYSDEC,
17    2011]

18    Capillarity: The action by which the surface of a liquid where it is in contact with a solid is elevated
19    or depressed depending on the relative attraction of the molecules of the liquid for each other and
20    for those of the solid. Capillary forces arise from the differential attraction between immiscible
21    fluids and solid surfaces; these are the forces responsible for capillary rise in small-diameter tubes
22    and porous materials. [Adapted from Pake, 1978]

23    Casing: Steel pipe that is lowered into a wellbore. Casing extends from the bottom of the hole to the
24    surface. [Schlumberger. 2014]

2 5    Casing inspection logs: An in situ record of casing thickness and integrity, to determine whether
26    and to what extent the casing has undergone corrosion. The term refers to an individual
27    measurement, or a combination of measurements using acoustic, electrical, and mechanical
28    techniques, to evaluate the casing thickness and other parameters. The log is usually presented
29    with the basic measurements and an estimate of metal loss. It was first introduced in the early
30    1960s. Today the terms casing-evaluation log and pipe-inspection log are used synonymously.
31    fSchlumberger. 20141

3 2    Cation exchange capacity: The total amount of cations (positively charged ions] that a soil can
33    hold.
                   This document is a draft for review purposes only and does not constitute Agency policy.
      June 2015                                    J-2                   DRAFT—DO NOT CITE OR QUOTE

-------
      Hydraulic Fracturing Drinking Water Assessment                                              Appendix J


 1    Cement: Material used to support and seal the well casing to the rock formations exposed in the
 2    borehole. Cement also protects the casing from corrosion and prevents movement of injectate up
 3    the borehole. fU.S. EPA. 2013dl

 4    Cement squeeze: A remedial cementing operation designed to force cement into leak paths in
 5    wellbore tubulars. The required squeeze pressure is achieved by carefully controlling pump
 6    pressure. Squeeze cementing operations may be performed to repair poor primary cement jobs,
 7    isolate perforations, or repair damaged casing or liner. [Schlumberger. 2014]

 8    Centralized waste treatment facility (CWT): any facility that treats (for disposal, recycling or
 9    recovery of material) any hazardous or non-hazardous industrial wastes, hazardous or non-
10    hazardous industrial waste water, and/or used material received from off-site. [U.S. EPA. 2012b]

11    Coalbed methane: Methane contained in coal seams. A coal seam is a layer or stratum of coal
12    parallel to the rock stratification. [U.S. EPA. 2013d]

13    Collapse pressure: The pressure at which a tube, or vessel, will catastrophically deform as a result
14    of differential pressure acting from outside to inside of the vessel or tube. [Schlumberger. 2014]

15    Collar: A threaded coupling used to join two lengths of pipe such as production tubing, casing, or
16    liner. The type of thread and style of collar varies with the specifications and manufacturer of the
17    tubing. [Schlumberger. 2014]

18    Combination truck: A truck tractor or a truck tractor pulling any number of trailers. [U.S.
19    Department of Transportation. 2012]

2 0    Community water systems: Public water systems that supply water to the  same population year-
21    round. fU.S. EPA. 2013d

2 2    Completion: A term used to describe the assembly of equipment at the bottom of the well that is
23    needed to enable production from an oil or gas well. It can also refer to the activities and methods
24    (including hydraulic fracturing] used to prepare a well for production following drilling.

25    Complexation: A reaction between two chemicals that form a new complex, either through
26    covalent bonding or ionic forces. This often results in  one chemical solubilizing the other.

27    Compressive strength: Measure of the ability of a substance to withstand compression. [NYSDEC.
28    2011]

29    Conductor casing: This large diameter casing is usually the first string of casing in a well. It is set
30    or driven into the unconsolidated material where the  well will be drilled to keep the loose material
31    from caving in. [NYSDEC. 2011]

32    Confidential business information (CBI): Information that contains trade  secrets, commercial or
33    financial information, or other information that has been claimed as confidential by the submitter.
34    The EPA has special procedures for handling such information. [U.S. EPA. 2013d]

                   This document is a draft for review purposes only and does not constitute Agency policy.
      June 2015                                    J-3                  DRAFT—DO NOT CITE OR QUOTE

-------
      Hydraulic Fracturing Drinking Water Assessment                                              Appendix J


 1    Contaminant: A substance that is either present in an environment where it does not belong or is
 2    present at levels that might cause harmful (adverse) health effects. [U.S. EPA. 2013d]

 3    Conventional reservoir: A reservoir in which buoyant forces keep hydrocarbons in place below a
 4    sealing caprock. Reservoir and fluid characteristics of conventional reservoirs typically permit oil
 5    or natural gas to flow readily into wellbores. The term is used to make a distinction from shale and
 6    other unconventional reservoirs, in which gas might be distributed throughout the reservoir at the
 7    basin scale, and in which buoyant forces or the influence of a water column on the location of
 8    hydrocarbons within the reservoir are not significant. [Schlumberger. 2014]

 9    Crosslinked gels: linear gels that are linked together by chemicals called crosslinkers, which may
10    link two or more chains together.

11    Crude oil: A general term for unrefined petroleum or liquid petroleum. [Schlumberger. 2014]

12    Cumulative effects: Refers to combined changes in the environment that can take place as a result
13    of multiple activities over time and/or space.

14    Cumulative water use/cumulative water: Refers to the amount of water used or consumed by all
15    hydraulic fracturing wells in a given area per year.

16    Cyclical stress: Refers to stress caused by frequent or rapid changes in temperature or pressure.

17    Deviated well: Any non-horizontal well in which the well bottom is intentionally located at a
18    distance (e.g., hundreds of feet] laterally from the wellhead.

19    Discharge: Any emission (other than natural seepage], intentional or unintentional. Includes, but is
20    not limited to, spilling, leaking, pumping, pouring, emitting, emptying, or dumping. (U.S. EPA.
21    2013d]

22    Disinfection byproduct (DBF):  A compound formed by the reaction of a disinfectant such as
23    chlorine with organic material  in the water supply. (U.S. EPA. 2013d]

24    Domestic water use: Includes indoor and outdoor water uses at residences, and includes uses such
25    as drinking, food preparation, bathing, washing clothes and dishes, flushing toilets, watering lawns
26    and gardens, and maintaining pools. (USGS, 2015]

27    Drill bit: The tool used to crush or cut rock. Most bits work by scraping or crushing the rock as part
28    of a rotational motion, while some bits work by pounding the rock vertically. (Schlumberger, 2014]

29    Drill collar: A component of a  drill string that provides weight on the bit for drilling. Drill collars
30    are thick-walled tubular pieces machined from solid bars of steel, usually plain carbon steel but
31    sometimes of nonmagnetic nickel-copper alloy or other nonmagnetic premium alloys.
32    (Schlumberger. 2014]

3 3    Drill cuttings: Ground rock produced by the drilling process.


                   This document is a draft for review purposes only and does not constitute Agency policy.
      June 2015                                    J-4                   DRAFT—DO NOT CITE OR QUOTE

-------
      Hydraulic Fracturing Drinking Water Assessment                                               Appendix J


 1    Drill string: The combination of the drillpipe, the bottomhole assembly, and any other tools used to
 2    make the drill bit turn at the bottom of the wellbore. [Schlumberger. 2014]

 3    Drilling fluid: Any of a number of liquid and gaseous fluids and mixtures of fluids and solids used
 4    when drilling boreholes. [Adapted from Schlumberger, 2014]

 5    Drinking water resource: Any body of ground water or surface water that now serves, or in the
 6    future could serve, as a source of drinking water for public or private use [U.S. EPA, 2013d]

 7    Dry gas: Refers to natural gas that occurs in the absence of liquid hydrocarbons. [Adapted from
 8    Schlumberger. 2014]

 9    Effluent: Waste material being discharged into the environment, either treated or untreated. [U.S.
10    EPA. 2013d]

11    Facultative anaerobes: Microorganisms that can use oxygen for energy production if it is present
12    in their environment, but can also use alternatives for energy production if no oxygen is present

13    Fault: A fracture or fracture zone along which there has been displacement of the sides relative to
14    each other. fNYSDEC. 20111

15    Field: Area of oil and gas production with at least one common reservoir for the entire area. [Oil
16    and Gas Mineral Services. 2010]

17    Flowback: The term is defined multiple ways in the literature. In general, it is either fluids
18    predominantly containing hydraulic fracturing fluid that return from a well to the surface or a
19    process used to prepare the well for production.

20    Fluid: A substance that flows when exposed to an external pressure; fluids include both liquids and
21    gases.

22    Fluid formulation: The entire suite of chemicals, proppant, and base fluid injected into a well
23    during hydraulic fracturing. [U.S. EPA. 2013d]

24    Formation: A body of earth material with distinctive and characteristic properties and a degree of
25    homogeneity in its physical properties. [U.S. EPA. 2013d]

2 6    Formation packer: A specialized casing part that has the same inner diameter as the casing but
27    whose outer diameter expands to make contact with the formation and seal the annulus between
28    the casing and formation, preventing migration of fluids.

29    Formation fluid: Fluid that occurs naturally within the pores of rock. These fluids consist primarily
30    of water, with varying concentrations of total dissolved solids, but may also contain oil or gas.
31    Sometimes referred to as native fluids, native brines, or reservoir fluids.

32    FracFocus Registry: A registry for oil and gas well operators to disclose information about
33    hydraulic fracturing well locations, and water and chemical use during hydraulic fracturing

                   This document is a draft for review purposes only and does not constitute Agency policy.
      June 2015                                    J-5                   DRAFT—DO NOT CITE OR QUOTE

-------
      Hydraulic Fracturing Drinking Water Assessment                                              Appendix J


 1    operations developed by the Ground Water Protection Council and the Interstate Oil and Gas
 2    Compact Commission.

 3    Fracture: A crack or breakage surface within a rock.

 4    Fracture geometry: Refers to characteristics of the fracture such as height and aperture (width).

 5    Fresh water: Qualitatively refers to water with relatively low IDS that is most readily available for
 6    drinking water currently.

 7    Gelled fluids: Fracturing fluids that are usually water-based with added gels to increase the fluid
 8    viscosity to aid in the transport of proppants. [Spellman. 2012: Gupta and Valko. 2007]

 9    Ground water: In the broadest sense, all subsurface water; more commonly that part of the
10    subsurface water in the saturated zone. [Solley etal., 1998]

11    Halite: A soft, soluble evaporate mineral commonly known as salt or rock salt Can be critical in
12    forming hydrocarbon traps and seals because it tends to flow rather than fracture during
13    deformation, thus preventing hydrocarbons from leaking out of a trap even during and after some
14    types of deformation. [Schlumberger. 2014]

15    Hazard evaluation: A component of risk assessment that involves gathering and evaluating data
16    on the types of health injuries or diseases  (e.g., cancer] that may be produced by a chemical and on
17    the conditions of exposure under which such health effects are produced.

18    Hazard identification: A process for determining if a chemical or a microbe can cause adverse
19    health effects in humans and what those effects might be. (U.S. EPA. 2013d]

20    Henry's law constant: Ratio of a chemical's vapor pressure in the atmosphere to its solubility in
21    water. The higher the Henry's law constant, the more volatile the compound will be from water.
22    fNYSDEC. 20111

2 3    Horizontal drilling: Drilling a portion of a well horizontally to expose more of the formation
24    surface area to the wellbore. (Oil and Gas Mineral Services. 2010]

25    Horizontal well: A well that is drilled vertically up to a point known as the kickoff point, where the
26    well turns toward the horizontal, extending into and parallel with the approximately horizontal
27    targeted producing formation.

28    Hydraulic fracturing: A stimulation technique used to increase production of oil and gas.
29    Hydraulic fracturing involves the injection of fluids under pressures great enough to fracture the
30    oil- and gas-production formations. (U.S. EPA. 2011a]

31    Hydraulic fracturing fluids: Engineered  fluids, typically consisting of a base fluid, additives, and
32    proppant, that are pumped under high pressure into the well to create and hold open fractures in
33    the formation.
                   This document is a draft for review purposes only and does not constitute Agency policy.
      June 2015                                    J-6                   DRAFT—DO NOT CITE OR QUOTE

-------
      Hydraulic Fracturing Drinking Water Assessment                                              Appendix J


 1    Hydraulic fracturing wastewater: Flowback and produced water that is managed using practices
 2    that include but are not limited to reuse in subsequent hydraulic fracturing operations, treatment
 3    and discharge, and injection into disposal wells.

 4    Hydraulic fracturing water cycle: The cycle of water in the hydraulic fracturing process,
 5    encompassing the acquisition of water, chemical mixing of the fracturing fluid, injection of the fluid
 6    into the formation, the production and management of flowback and produced water, and the
 7    ultimate treatment and disposal of hydraulic fracturing waste waters.

 8    Hydraulic gradient: Slope of a water table or potentiometric surface. More specifically, change in
 9    the hydraulic head per unit of distance in the direction of the maximum rate of decrease.  [U.S. EPA.
10    2013dl

11    Hydrocarbon: An organic compound containing only hydrogen and carbon, often occurring in
12    petroleum, natural gas, and coal. [U.S. EPA. 2013d]

13    Hydrostatic pressure: The pressure exerted by a column of fluid at a given depth.

14    Imbibition: The displacement of a non-wet fluid (i.e., gas) by a wet fluid (typically water). [Adapted
15    from Pake. 1978)

16    Immiscible: The chemical property in which two or more liquids or phases are incapable of
17    attaining homogeneity. [U.S. EPA, 2013d]

18    Impact: Any observed change in the quality or quantity of drinking water resources, regardless of
19    severity, that results from a mechanism.

2 0    Impact, potential: Any change in the quality or quantity of drinking water resources that could
21    logically occur, but has not yet been observed, as the result of a mechanism or potential mechanism.

22    Induced fracture: A fracture created during hydraulic fracturing.

23    Injection well: A well into which fluids are being injected (40 CFR 144.3).

24    Integrated risk information system (IRIS): An electronic database that contains the EPA's latest
25    descriptive and quantitative regulatory information about chemical constituents. Files on chemicals
26    maintained in IRIS contain information related to both noncarcinogenic and carcinogenic health
27    effects. (U.S. EPA. 2013d]

28    Intermediate casing: Casing that seals off intermediate depths and geologic formations that may
29    have considerably different reservoir pressures than deeper zones to be drilled. (Devereux. 1998;
30    Baker. 1979]

31    Karst: A type of topography that results from dissolution and collapse of carbonate rocks, such as
3 2    limestone, dolomite, and gypsum, and that is characterized by closed depressions or sinkholes,
33    caves, and underground drainage. (Soiiey etal, 1998]
                   This document is a draft for review purposes only and does not constitute Agency policy.
      June 2015                                    J-7                   DRAFT—DO NOT CITE OR QUOTE

-------
      Hydraulic Fracturing Drinking Water Assessment                                               Appendix J


 1    Kill fluid: A weighted fluid with a density that is sufficient to overcome the formation pressure and
 2    prevent fluids from flowing up the wellbore.

 3    Large truck: A truck with a gross vehicle weight rating greater than 10,000 pounds.
 4
 5    Lateral: A horizontal section of a well.

 6    Leakoff: The fraction of the injected fluid that infiltrates into the formation (e.g., through an
 7    existing natural fissure) and is not recovered during production.

 8    Linear gel: a series of chemicals linked together so that they form a chain.

 9    Liner: A casing string that does not extend to the top of the wellbore, but instead is anchored or
10    suspended from inside the bottom of the previous casing string. (SchJunibergerI_2fll4)

1 1    Lost cement: Refers to a failure of the cement to be circulated back to the surface, indicating that
1 2    the cement has escaped into the formation.

13    Lowest-observable-adverse effect level (LOAEL): The lowest exposure level at which there are
14    biologically significant increases in frequency or severity of adverse effects between the exposed
15    population and its appropriate control group.

16    Maximum allowable daily level (MADL): The maximum allowable daily level of a reproductive
17    toxicant at which the chemical would have no observable adverse reproductive effect, assuming
18    exposure at 1,000 times that level.

19    Maximum contaminant level (MCL): The highest level of a contaminant that is allowed in
20    drinking water. MCLs are enforceable standards. (UJ
21    Mechanical integrity: The absence of significant leakage within the injection tubing, casing, or
22    packer (known as internal mechanical integrity), or outside of the casing (known as external
23    mechanical integrity). (UJJiPAi1013d)

24    Mechanism: A means or series of events by which an activity within the hydraulic fracturing water
25    cycle has been observed to change the quality or quantity of drinking water resources.

26    Mechanism, potential: A means or series of events by which hydraulic fracturing activities could
27    logically or theoretically (for instance, based on modeling) change the quality or quantity of
28    drinking water resources but one that has not yet been observed.

29    Mechanism, suspected: A means or series of events by which hydraulic fracturing activities could
30    logically have resulted in an observed change in the quality or quantity of drinking water resources.
31    Available evidence may or may not be sufficient to determine if it is the only mechanism that caused
32    the observed change.
                   This document is a draft for review purposes only and does not constitute Agency policy.
      June 2015                                    J-8                   DRAFT—DO NOT CITE OR QUOTE

-------
      Hydraulic Fracturing Drinking Water Assessment                                              Appendix J


 1    Metropolitan combined statistical area: A core urban area of 50,000 or more people. [U.S.
 2    Census Bureau. 2013]

 3    Microaerophiles: Microorganisms that require small amounts of oxygen for energy production.

 4    Microannuli: Very small channels that form in the cement and that may serve as pathways for fluid
 5    migration to drinking water resources.

 6    Micropolitan combined statistical area: An urban core of at least 10,000, but less than 50,000,
 7    people. [U.S. Census Bureau. 2013]

 8    Microseismic monitoring: A technique to track the propagation of a hydraulic fracture as it
 9    advances through a formation. [Schlumberger. 2014]

10    Minimum risk level (MRL): An estimate of daily human exposure to a hazardous substance at or
11    below which the substance is unlikely to pose a measurable risk of harmful (adverse],
12    noncancerous effects. MRLs are calculated for a route of exposure (inhalation or oral] over a
13    specified time period (acute, intermediate, or chronic].

14    Mobility: The ratio of effective permeability to phase viscosity. The overall mobility is a sum of the
15    individual phase viscosities. Well productivity is directly proportional to the product of the mobility
16    and the layer thickness product (Schlumberger. 2014]

17    National Pollution Discharge Elimination System (NPDES): A national program under
18    Section 402 of the Clean Water Act for regulation of discharges of pollutants from point sources to
19    waters of the United States. Discharges are illegal unless authorized by an NPDES permit (U.S. EPA,
20    2013dl

21    National Secondary Drinking Water Regulations (NSDWR): Non-enforceable guidelines
22    regulating contaminants that may cause cosmetic effects (such as skin or tooth discoloration] or
23    aesthetic effects (such as taste, odor, or color] in drinking water (also referred to as secondary
24    standards].  (U.S. EPA. 2014b]

2 5    Natural gas: A naturally occurring mixture of hydrocarbon and nonhydrocarbon gases  in porous
26    formations beneath the earth's surface, often in association with petroleum. The principal
27    constituent of natural gas is methane. (Schlumberger. 2014]

28    Natural organic matter (NOM): Complex organic compounds that are formed from decomposing
29    plant animal and microbial material in soil and water. (U.S. EPA. 2013d]

3 0    Non-community water systems:  Water systems that supply water to at least 2 5 of the same
31    people at least six months per year, but not year-round. (U.S. EPA. 2013c]

32    Octanol-water partition coefficient (Kow): A coefficient representing the ratio of the solubility of a
33    compound in octanol (a nonpolar solvent] to its solubility in water (a polar solvent]. The higher the
34    Kow, the more nonpolar the compound. Log Kow is generally used as a relative indicator of the

                   This document is a draft for review purposes only and does not constitute Agency policy.
      June 2015                                    J-9                   DRAFT—DO NOT CITE OR QUOTE

-------
      Hydraulic Fracturing Drinking Water Assessment                                              Appendix J


 1    tendency of an organic compound to adsorb to soil. Log Kow values are generally inversely related to
 2    aqueous solubility and directly proportional to molecular weight. [U.S. EPA. 2013d]

 3    Offset well: An existing wellbore close to a proposed well that provides information for planning
 4    the proposed well. [Schlumberger. 2014]

 5    Open hole completion: A well completion that has no casing or liner set across the reservoir
 6    formation, allowing the produced fluids to flow directly into the wellbore. [Schlumberger. 2014]

 7    Oral slope factor (OSF): An upper-bound, approximating a 95% confidence limit, on the increased
 8    cancer risk from a lifetime oral exposure to an agent. This estimate, usually expressed in units of
 9    proportion (of a population] affected per mg/kg day, is generally reserved for use in the low dose
10    region of the dose response relationship, that is, for exposures corresponding to risks less than 1 in
11    100.

12    Organic carbon-water partition coefficient (Koc)'- A coefficient representing the amount of a
13    compound that is adsorbed to soil to the amount of a compound that is dissolved in water,
14    normalized to the total organic carbon content of the soil. The higher the Koc, the more likely a
15    compound is to adsorb to soils and sediments, and the less likely it is to migrate with water. Along
16    with log Kow, log Koc is used as a relative indicator of the tendency of an organic compound to adsorb
17    to soil.

18    Orphaned well: An inactive oil or gas well with no known (or financially solvent] owner.

19    Overburden:  Material of any nature, consolidated or unconsolidated, that overlies a deposit of
20    useful minerals or ores. (U.S. EPA. 2013d]

21    Packer: A device that can be run into a wellbore with a smaller initial outside diameter that then
22    expands externally to seal the wellbore. (Schlumberger. 2014]

23    Pad fluid: a mixture of base fluid, typically water and additives designed to create, elongate, and
24    enlarge fractures along the natural channels of the formation when injected under high pressure.

2 5    Partial cementing: Cementing a casing string along only a portion of its length.

2 6    Passby flow: A prescribed, low-streamflow threshold below which withdrawals are not allowed.
27    fU.S. EPA. 2015dl

28    Peer review:  A documented critical review of a specific major scientific and/or technical work
29    product Peer  review is intended to uncover any technical problems or unresolved issues in a
3 0    preliminary or draft work product through the use of independent experts. This information is then
31    used to revise the draft so that the final work product will reflect sound technical information and
32    analyses. The process of peer review enhances the scientific or technical work product so that the
33    decision or position taken by the EPA, based on that product, has a sound and credible basis. (U.S.
34    EPA. 2013dl
                   This document is a draft for review purposes only and does not constitute Agency policy.
      June 2015                                    J-10                  DRAFT—DO NOT CITE OR QUOTE

-------
      Hydraulic Fracturing Drinking Water Assessment                                               Appendix J


 1    Perforation: The communication tunnel created from the casing or liner into the reservoir
 2    formation through which injected fluids and oil or gas flows. Also refers to the process of creating
 3    communication channels, e.g., via the use of a jet perforating gun.

 4    Permeability: The ability of a material (e.g., rock or soil) to transmit fluid to move through pore
 5    spaces.

 6    Persistence: The length of time a compound stays in the environment, once introduced. A
 7    compound may persist for less than a second or indefinitely.

 8    Physicochemical properties: The inherent physical and chemical properties of a molecule such as
 9    boiling point, density, physical state, molecular weight, vapor pressure, etc. These properties define
10    how a chemical interacts with its environment. (UiS._EPA1_2()13d)

1 1    Play: A set of oil or gas accumulations sharing similar geologic, geographic properties, such as
12    source rock, hydrocarbon type, and migration pathways.
13    Poisson's ratio: A ratio of transverse-to-axial (or latitudinal- to-longitudinal) strain; characterizes
14    how a material is deformed under pressure.

15    Polar molecule: A molecule with a slightly positive charge at one part of the molecule and a
16    slightly negative charge on another. The water molecule, H20, is an example of a polar molecule,
17    where the molecule is slightly positive around the hydrogen atoms and negative around the oxygen
18    atom.

19    Porosity: A measure of pore space, or the percentage of the material (e.g., rock or soil) volume that
20    can be occupied by oil, gas, or water.

2 1    Produced water: Water that flows from oil and gas wells.

22    Production casing: The deepest casing set and serves primarily as the conduit for producing fluids,
23    although when cemented to the wellbore, this casing can also serve to seal off other subsurface
24    zones including ground water resources. (
25    Production well: A well that is used to bring fluids (such as oil or gas) to the surface.

26    Production zone: Refers to the portion of a subsurface rock zone that contains oil or gas to be
27    extracted (sometimes using hydraulic fracturing). The production zone is sometimes referred to as
2 8    the target zone.

29    Proppant/propping agent: A granular substance (sand grains, aluminum pellets, or other
30    material) that is carried in suspension by the fracturing fluid and that serves to keep the cracks
31    open when fracturing fluid is withdrawn after a fracture treatment (U1S._EPAJ_2013d)

3 2    Protected ground water resource: The deepest aquifer that the state or other regulatory agency
3 3    requires to be protected from fluid migration through or along wellbores.
                   This document is a draft for review purposes only and does not constitute Agency policy.
      June 2015                                    J-ll                  DRAFT—DO NOT CITE OR QUOTE

-------
      Hydraulic Fracturing Drinking Water Assessment                                              Appendix J


 1    Public water system source: The source of the surface or ground water used by a public water
 2    system, including source wells, intakes, reservoirs, infiltration galleries, and springs.

 3    Public water systems: Water systems that provide water for human consumption from surface or
 4    ground water through pipes or other infrastructure to at least 15 service connections or serve an
 5    average of at least 25 people for at least 60 days a year. [Safe Drinking Water Act, 2002]

 6    Publicly owned treatment works (POTW): Any device or system used in the treatment (including
 7    recycling and reclamation) of municipal sewage or industrial wastes of a liquid nature that is
 8    owned by a state or municipality. This definition includes sewers, pipes, or other conveyances only
 9    if they convey wastewater to a POTW providing treatment. fU.S. EPA. 2013dl

10    Quality assurance (QA): An integrated system of management activities involving planning,
11    implementation, documentation, assessment, reporting, and quality improvement to ensure that a
12    process, item, or service is of the type and quality  needed and expected by the customer. [U.S. EPA.
13    2013d]

14    Quality assurance project plan (QAPP): A formal document describing in comprehensive detail
15    the necessary quality assurance procedures, quality control activities, and other technical activities
16    that need to be implemented to ensure that the results of the work performed will satisfy the stated
17    performance or acceptance criteria. [U.S.  EPA. 2013d]

18    Quality management plan: A document that describes a quality system in terms of the
19    organizational structure, policy and procedures, functional responsibilities of management and
20    staff, lines of authority, and required interfaces for those planning, implementing, documenting, and
21    assessing all activities conducted. [U.S.  EPA. 2013d]

2 2    Radioactive tracer log: A record of the presence  of tracer material placed in or  around the
23    borehole to measure fluid movement in injection wells. [Schiumberger. 2014]

24    Radionuclide: Radioactive particle, man-made or natural, with a distinct atomic weight number.
2 5    Emits radiation in the form of alpha or  beta particles, or as gamma rays. Can have a long life as soil
26    or water pollutant Prolonged exposure to radionuclides increases the risk of cancer. [U.S. EPA,


28    Reference dose (RfD): An estimate (with uncertainty spanning perhaps an order of magnitude] of
29    a daily oral exposure to the human population (including sensitive subgroups] that is likely to be
30    without an appreciable risk of deleterious effects during a lifetime.

31    Reference value (RfV): An estimate of an exposure for a given duration to the human population
32    (including susceptible subgroups] that is likely to  be without an appreciable risk of adverse health
33    effects over a lifetime. Reference value  is  a generic term not specific to a given route of exposure.

34    Relative permeability: A dimensionless property allowing for comparison of the different abilities
35    of fluids to flow in multiphase settings. If a single fluid is present, its relative permeability is equal


                   This document is a draft for review purposes only and does not constitute Agency policy.
      June 2015                                    J-12                 DRAFT—DO NOT CITE OR QUOTE

-------
      Hydraulic Fracturing Drinking Water Assessment                                              Appendix J


 1    to 1, but the presence of multiple fluids generally inhibits flow and decreases the relative
 2    permeability.

 3    Reservoir: A porous and permeable geologic formation where hydrocarbons collect under
 4    pressure over geological time.

 5    Residuals: The solids generated or retained during the treatment of waste water. [U.S. EPA. 2013d]

 6    Safe Drinking Water Act (SDWA): The act designed to protect the nation's drinking water supply
 7    by establishing national drinking water standards (maximum contaminant levels or specific
 8    treatment techniques) and by regulating underground injection control wells. [U.S. EPA, 2013d]

 9    Sandstone: A clastic sedimentary rock whose grains are predominantly sand sized. The term is
10    commonly used to imply consolidated sand or a rock made of predominantly quartz sand, although
11    sandstones often contain feldspar, rock fragments, mica, and numerous additional mineral grains
12    held together with silica or another type of cement The relatively high porosity and permeability of
13    sandstones make them good reservoir rocks. [Schlumberger. 2014]

14    Science Advisory Board (SAB): A federal advisory committee that provides a balanced, expert
15    assessment of scientific matters relevant to the EPA. An important function of the Science Advisory
16    Board is to review EPA's technical programs and research plans. [U.S. EPA. 2013d]

17    Service company: A company that assists well operators by providing specialty services, including
18    hydraulic fracturing. [U.S. EPA. 2013d]

19    Shale: A fine-grained, fissile, detrital sedimentary rock formed by consolidation of clay- and silt-
20    sized particles into thin, relatively impermeable layers. [Schlumberger. 2014]

21    Shale gas: Natural gas generated and stored in shale.

22    Shale oil: Oil present in unconventional oil reservoirs that are made up of shale.

23    Shut-in: The process of sealing off a well by either closing the valves at the wellhead, adownhole
24    safety valve, or a blowout preventer.

25    Slickwater: A type of fracturing fluid that consists mainly of water with a very low portion of
26    additives like polymers that serve as friction reducers to reduce friction loss when pumping the
27    fracturing fluid downhole. [Barati and Liang. 2014]

28    Solubility: The amount of mass of a compound that will dissolve in a unit volume of solution. [U.S.
29    EPA. 2013d]

3 0    Sorption: The general term used to describe the partitioning of a chemical between soil and water
31    and depends on the nature of the solids and the properties of the chemical.

32    Source water: Surface or ground water, or reused wastewater, acquired for use in hydraulic
33    fracturing. fU.S. EPA. 2013dl

                   This document is a draft for review purposes only and does not constitute Agency policy.
      June 2015                                    J-13                   DRAFT—DO NOT CITE OR QUOTE

-------
      Hydraulic Fracturing Drinking Water Assessment                                               Appendix J


 1    Spacer fluid: A fluid pumped before the cement to clean drilling mud out of the wellbore.

 2    Spud (spud a well): To start the well drilling process by removing rock, dirt, and other
 3    sedimentary material with the drill bit [U.S. EPA. 2013d]

 4    Stages (frac stages): A single reservoir interval that is hydraulically stimulated in succession with
 5    other intervals.

 6    Stimulation: Refers to (1) injecting fluids to clear the well or pore spaces near the well of drilling
 7    mud or other materials that create blockage and inhibit optimal production (i.e., matrix treatment)
 8    and (2) injecting fluid to fracture the rock to optimize the production of oil or gas.

 9    Stray gas: Refers to the phenomenon of natural gas (primarily methane) migrating into shallow
10    drinking water resources or to the surface.

11    Strings: An assembled length of steel pipe configured to suit a specific wellbore.

12    Subsurface formation: A mappable body of rock of distinctive rock type (s), including the rock's
13    pore volume (i.e., the void space within a formation that fluid flow can occur, as opposed  to the bulk
14    volume which includes both pore and solid phase volume), with a unique stratigraphic position.

15    Surface casing: The shallowest cemented casing, with the widest diameter. Cemented surface
16    casing generally serves as an anchor for blowout protection equipment and to seal off drinking
17    water resources. (Baker, 1979]

18    Surface water: All water naturally open to the atmosphere (rivers, lakes, reservoirs, ponds,
19    streams, impoundments, seas, estuaries, etc.). (U.S. EPA. 2013d]

20    Surfactant: Used during the hydraulic fracturing process to decrease liquid surface tension and
21    improve fluid passage through the pipes. (U.S. EPA, 2013d]

22    Sustained casing pressure: Refers to cases when the pressure in any well annulus that is
2 3    measurable at the wellhead rebuilds after it is bled down, not caused solely by temperature
24    fluctuations or imposed by the operator. If the pressure is relieved by venting natural gas from the
25    annulus to the atmosphere, it will build up again once the annulus is closed (i.e., the pressure is
26    sustained). (Skjervenetal.  2011]

27    Technically recoverable resources: The volumes of oil and natural gas that could be produced
28    with current technology, regardless of oil  and natural gas prices and production costs. (EIA, 2013]

29    Temperature log: A log of the temperature of the fluids in the borehole; a differential temperature
30    log records the rate of change in temperature with depth and is sensitive to very small changes.
31    (U.S. EPA. 2013d]

32    Tensile strength: The force per unit cross-sectional area required to pull a substance apart.
33    (Schlumberger, 2014]
                   This document is a draft for review purposes only and does not constitute Agency policy.
      June 2015                                    J-14                  DRAFT—DO NOT CITE OR QUOTE

-------
      Hydraulic Fracturing Drinking Water Assessment                                              Appendix J


 1    Thermogenic: Methane that is produced by high temperatures and pressures in deep formations
 2    over geologic timescales. Thermogenic methane is formed by the thermal breakdown, or cracking,
 3    of organic material that occurs during deep burial of sediment.

 4    Tight oil: Oil found in relatively impermeable reservoir rock. (SchlumbergerJ_2fll4)

 5    Total dissolved solids (TDS): The quantity of dissolved material in a given volume of water. Total
 6    dissolved solids can include salts (e.g., sodium chloride), dissolved metals, radionuclides, and
 7    dissolved organics. (li£^
 8    Toxicity: The degree to which a substance or mixture of substances can harm humans or animals.
 9    Acute toxicity involves harmful effects in an organism through a single or short-term exposure.
1 0    Chronic toxicity is the ability of a substance or mixture of substances to cause harmful effects over
11    an extended period, usually upon repeated or continuous exposure, sometimes lasting for the entire
12    life of the exposed organism. Subchronic toxicity is the ability of the substance to cause effects for
13    more than 1 year but less than the lifetime of the exposed organism. (UiS._EPA1_2()13d)

14    Tubing: The narrowest casing set within a completed well, either hung directly from the wellhead
15    or secured at its bottom using a packer. Tubing is not typically cemented in the well.

1 6    Unconventional reservoir: A reservoir characterized by lower permeability than conventional
17    reservoirs. It can be the same formation where hydrocarbons are formed and also serve as the
18    source for hydrocarbons that migrate and accumulate in conventional reservoirs. Unconventional
19    reservoirs can include methane-rich coalbeds and oil- and/or gas-bearing shales and tight sands.

20    Unconventional resource: An umbrella term for oil and natural gas that is produced by means
2 1    that do not meet the criteria for conventional production. What has qualified as unconventional at
22    any particular time is  a complex function of resource characteristics, the available exploration and
23    production technologies, the economic environment, and the scale, frequency, and duration of
24    production from the resource. Perceptions of these factors inevitably change over time and often
25    differ among users of the term. At present, the term is used in reference to oil and gas resources
26    whose porosity, permeability, fluid trapping mechanism, or other characteristics differ from
27    conventional sandstone and carbonate reservoirs. Coalbed methane, gas hydrates, shale gas,
28    fractured reservoirs, and tight gas sands are considered unconventional resources. (MiljjmbergiL
29    2014)

30    Underground Injection Control (UIC): The program under the Safe Drinking Water Act that
3 1    regulates the use of wells to pump fluids into the ground. (UJLER
32    Unsaturated zone: The soil zone above the water table that is only partially filled by water; also
33    referred to as the "vadose zone."

34    Vapor pressure: The force per unit area exerted by a vapor in an equilibrium state with its pure
35    solid, liquid, or solution at a given temperature. Vapor pressure is a measure of a substance's
                   This document is a draft for review purposes only and does not constitute Agency policy.
      June 2015                                    J-15                  DRAFT—DO NOT CITE OR QUOTE

-------
      Hydraulic Fracturing Drinking Water Assessment                                               Appendix J


 1    propensity to evaporate. Vapor pressure increases exponentially with an increase in temperature.
 2    [U.S. EPA. 2013d]

 3    Vertical well: A well in which the wellbore is vertical throughout its entire length, from the
 4    wellhead at the surface to the production zone.

 5    Viscosity: A measure of the internal friction of a fluid that provides resistance to shear within the
 6    fluid, informally referred to as how "thick" a fluid is.

 7    Volatile: Readily vaporizable at a relatively low temperature. [U.S. EPA. 2013d]

 8    Volatilization: The process in which a chemical leaves the liquid phase and enters the gas phase.

 9    Wastewater treatment: Chemical, biological, and mechanical procedures applied to an industrial
10    or municipal discharge or to any other sources of contaminated water in order to remove, reduce,
11    or neutralize contaminants. [U.S. EPA. 2013d]

12    Water availability: There is no standard definition for water availability, and it has not been
13    assessed recently at the national scale [U.S. GAP. 2014]. Instead, a number of water availability
14    indicators have been suggested [e.g.. Roy etal.. 2005]. Here, availability is most often used to
15    qualitatively refer to the amount of a location's water that could, currently or in the future, serve as
16    a source of drinking water [U.S. GAP, 2014], which is a function of water inputs to ahydrologic
17    system (e.g., rain, snowmelt, groundwater recharge] and water outputs from that system  occurring
18    either naturally or through competing demands of users.

19    Water consumption: Water that is removed from the local hydrologic cycle following its use (e.g.,
20    via evaporation, transpiration, incorporation into products or crops, consumption by humans or
21    livestock], and is therefore  unavailable to other water users (Maupinetal.. 2014].

22    Water intensity: The amount of water used per unit of energy obtained. (Nicotetal., 2014:
23    Laurenzi and Jersey, 2013]

24    Water reuse: Any hydraulic fracturing wastewater that is used to offset total fresh water
2 5    withdrawals for hydraulic fracturing, regardless of the level of treatment required.

26    Water use: Water withdrawn for a specific purpose, part or all of which may be returned to the
27    local hydrologic cycle.

28    Water withdrawal: Water removed from the ground or diverted from a surface-water source for
29    use. (Nicotetal.. 2014: Laurenzi and Jersey. 2013]

3 0    Well blowout: The uncontrolled flow of fluids out of a well.

31    Well communication: Refers to fractures intersecting abandoned or active (producing] offset
32    wells near the well that is being stimulated.
                   This document is a draft for review purposes only and does not constitute Agency policy.
      June 2015                                    J-16                   DRAFT—DO NOT CITE OR QUOTE

-------
      Hydraulic Fracturing Drinking Water Assessment                                                Appendix J


 1    Well logging: A continuous measurement of physical properties in or around the well with
 2    electrically powered instruments to infer formation properties. Measurements may include
 3    electrical properties (resistivity and conductivity), sonic properties, active and passive nuclear
 4    measurements, measurements of the wellbore, pressure measurement, formation fluid sampling,
 5    sidewall coring tools, and others. Measurements may be taken via a wireline, which is a wire or
 6    cable that is used to deploy tools and instruments downhole and that transmits data to the surface.
 7    [Adapted from Schlumberger. 2014]

 8    Well operator: A company that controls and operates oil and gas wells. [U.S. EPA, 2013d]

 9    Well pad: A temporary drilling site, usually constructed of local materials such as sand and gravel.
10    After the drilling operation is over, most of the pad is usually removed or plowed back into the
11    ground. fNYSDEC. 2011]

12    Wellbore: The drilled hole or borehole, including the open hole or uncased portion of the well.

13    Wet gas: Refers to natural gas that typically contains less than 85% methane along with ethane and
14    more complex hydrocarbons.

15    Wetting/nonwetting: The preferential attraction of a fluid to the surface. In typical reservoirs,
16    water preferentially wets the surface, and gas is nonwetting.  [Adapted from Pake. 1978]

17    Workover: Refers to any maintenance activity performed on a well that involves ceasing
18    operations and removing the wellhead.

19    Young's modulus: A ratio of stress to strain that is a measure of the rigidity of a material.

      J.2.    References for Appendix J
      Baker. R. (1979). A primer of oilwell drilling (4th ed). Austin, TX: Petroleum Extension Service (PETEX).
      Barati. R: Liang. IT. (2014). A review of fracturing fluid systems used for hydraulic fracturing of oil and gas
         wells. J Appl Polymer Sci Online pub. http://dx.doi.org/10.1002/app.40735
      Pake. LP. (1978). Fundamentals of reservoir engineering. Boston, MA: Elsevier.
         http://www.ing.unp.edu.ar/asignaturas/reservorios/Fundamentals%20of%20Reservoir%20Engineering
         %20%28LP.%20Dake%29.pdf
      Devereux. S. (1998). Practical well planning and drilling manual. Tulsa, OK: PennWell Publishing Company.
         http://www.pennwellbooks.com/practical-well-planning-and-drilling-manual/
      EIA (Energy Information Administration). (2013). Technically recoverable shale oil and shale gas resources:
         an assessment of 137 shale formations in 41 countries outside the United States (pp. 730). Washington,
         D.C.: Energy Information Administration, U.S. Department of Energy.
         http://www.eia.gov/analysis/studies/worldshalegas/
      Gupta. DVS: Valko. P. (2007). Fracturing fluids and formation damage. In M Economides; T Martin (Eds.),
         Modern fracturing: enhancing natural gas production (pp. 227-279). Houston, TX: Energy Tribune
         Publishing Inc.
      Laurenzi. II: Jersey. GR. (2013).  Life cycle greenhouse gas emissions and freshwater consumption of Marcellus
         shale gas. Environ Sci Technol 47:  4896-4903. http://dx.doi.org/10.1021/es305162w
                    This document is a draft for review purposes only and does not constitute Agency policy.
      June 2015                                     J-17                  DRAFT—DO NOT CITE OR QUOTE

-------
Hydraulic Fracturing Drinking Water Assessment                                                  Appendix J
Maupin. MA: Kenny. IF: Hutson. SS: Lovelace. IK: Barber. NL: Linsey. KS. (2014). Estimated use of water in the
   United States in 2010. (USGS Circular 1405). Reston, VA: U.S. Geological Survey.
   http://dx.doi.org/10.3133/cirl405

Nicot. IP: Scanlon. BR: Reedy. RC: Costley. RA. (2014). Source and fate of hydraulic fracturing water in the
   Barnett Shale: a historical perspective. Environ Sci Technol 48: 2464-2471.
   http://dx.doi.org/10.1021/es404050r

NYSDEC (New York State Department of Environmental Conservation). (2011). Revised draft supplemental
   generic environmental impact statement (SGEIS) on the oil, gas and solution mining regulatory program:
   Well permit issuance for horizontal drilling and high-volume hydraulic fracturing to develop the Marcellus
   shale and other  low-permeability gas reservoirs. Albany, NY: NY SDEC.
   http://www.dec.ny.gov/energy/75370.html

Oil and Gas Mineral Services. (2010). MineralWise: Oil and gas terminology. Available online at
   http://www.mineralweb.com/library/oil-and-gas-terms/

Roy. SB: Ricci. PF: Summers. KV: Chung. CF: Goldstein. RA. (2005). Evaluation of the sustainability of water
   withdrawals in the United States, 1995 to 2025. J Am Water Resour Assoc 41:1091-1108.

Safe Drinking Water Act. Title XIV of the Public Health Service Act Safety of Public Water Systems (Safe
   Drinking Water  Act) as amended through P.L. 107-377, (2002). http://www.epw.senate.gov/sdwa.pdf

Schlumberger (Schlumberger Limited). (2014). Schlumberger oilfield glossary. Available online at
   http://www.glossary.oilfield.slb.com/

Skierven. T: Lunde. 0: Perander. M: Williams. B: Farquhar. R: Sinet. I: Sasby. I: Haga. HB: Finnseth. 0: lohnsen.
   S. (2011). Norwegian Oil and Gas Association recommended guidelines for well integrity. (117, Revision
   4). Norway: Norwegian Oil and Gas Association.
   http://www.norskoljeoggass.no/Global/Retningslinjer/Boring/117%20-
   %20Recommended%20guidelines%20Well%20integrity%20rev4%2006.06.%2011.pdf

Solley. WB: Pierce. RR: Perlman. HA. (1998). Estimated use of water in the United States in 1995. (USGS
   Circular:  1200). U.S. Geological Survey, http://pubs.er.usgs.gov/publication/cirl200

Spellman, FR. (2012). Environmental impacts of hydraulic fracturing. In Environmental impacts of hydraulic
   fracturing. Boca Raton, Florida: CRC Press.

U.S. Census Bureau. (2013). Metropolitan and micropolitan statistical areas main. Available  online at
   http://www.census.gov/population/metro/ (accessed January 12, 2015).

U.S. Department of  Transportation. (2012). Large truck and bus crash facts 2012. Washington, D.C.: Federal
   Motor Carrier Safety Administration, U.S. Department of Transportation.
   http://ai.fmcsa.dot.gov/CarrierResearchResults/PDFs/LargeTruckandBusCrashFacts2012.pdf

U.S. EPA (U.S. Environmental Protection Agency). (2011a).  Plan to study the potential impacts of hydraulic
   fracturing on drinking water resources [EPA Report]. (EPA/600/R-11/122). Washington, DC: U.S.
   Environmental Protection Agency, Office of Research and Development.
   http://www2.epa.gov/hfstudy/plan-study-potential-impacts-hydraulic-fracturing-drinking-water-
   resources-epa600r-11122

U.S. EPA. General definitions. 40 CFR § 437.2 (2012b). http://www.gpo.gov/fdsys/pkg/CFR-2012-title40-
   vol31/pdf/CFR-2012-title40-vol31-sec437-2.pdf

U.S. EPA (U.S. Environmental Protection Agency). (2013c).  Drinking water and ground water statistics, fiscal
   year 2011. Washington, DC: U.S. Environmental Protection Agency, Office of Water.
   http://water.epa.gov/scitech/datait/databases/drink/sdwisfed/upload/epa816rl3003.pdf
               This document is a draft for review purposes only and does not constitute Agency policy.

June 2015                                       J-18                    DRAFT—DO NOT CITE OR QUOTE

-------
Hydraulic Fracturing Drinking Water Assessment                                                   Appendix J
U.S. EPA (U.S. Environmental Protection Agency). (2013d). Terminology services (TS): Terms and acronyms.
   Available online at
   http://iaspub.epa.gov/sor internet/registry/termreg/searchandretrieve/termsandacronyms/search.do

U.S. EPA (U.S. Environmental Protection Agency). (2014b). Drinking water contaminants. Available online at
   http://water.epa.gov/drink/contaminants/

U.S. EPA (U.S. Environmental Protection Agency). (2015d). Case study analysis of the impacts of water
   acquisition for hydraulic fracturing on local water availability [EPA Report]. (EPA/600/R-14/179).
   Washington, D.C.

U.S. GAP (U.S. Government Accountability Office). (2014). Freshwater: Supply concerns continue, and
   uncertainties complicate planning. Report to Congressional requesters. (GAO-14-430). Washington, DC:
   U.S. Government Accountability Office (GAO). http://www.gao.gov/assets/670/663343.pdf

USGS (U.S. Geological Survey). (2015). Water use in the United States. Available online at
   http://water.usgs.gov/watuse/
               This document is a draft for review purposes only and does not constitute Agency policy.

June 2015                                        J-19                    DRAFT—DO NOT CITE OR QUOTE

-------
Hydraulic Fracturing Drinking Water Assessment                                     All Appendices' References
                    References:  All Appendices
Abrams. R. (2013). Advanced oxidation frac water recycling system. Presentation presented at 20th
   International Petroleum Environmental Conference, November 12-14, 2013, San Antonio, TX.

Acharya. HR: Henderson. C: Matis. H: Kommepalli. H: Moore. B: Wang. H. (2011). Cost effective recovery of
   low-TDS frac flowback water for reuse. (Department of Energy: DE-FE0000784). Niskayuna, NY: GE Global
   Research. http://www.netl.doe.gov/file%201ibrary/Research/oil-gas/FE0000784 FinalReportpdf

Afzal. W: Mohammadi. AH: Richon. D. (2009). Volumetric properties of mono-, di-, tri-, and polyethylene
   glycol aqueous solutions from (273.15 to 363.15) K: experimental measurements and correlations. Journal
   of Chemical and Engineering Data 54:1254-1261. http://dx.doi.org/10.1021/ie800694a

Ahmann. D: Roberts. AL: Krumholz. LR: Morel. FM. (1994). Microbe grows by reducing arsenic [Letter].
   Nature 371: 750. http://dx.doi.org/10.1038/371750aO

Alain. K: Pignet. P: Zbinden. M: Quillevere. M: Duchiron. F: Donval. IP: Lesongeur. F: Raguenes. G: Crassous. P:
   Querellou, I: Cambon-Bonavita, MA. (2002). Caminicella sporogenes gen. nov., sp. nov., a novel
   thermophilic spore-forming bacterium isolated from an East-Pacific Rise hydrothermal vent. Int J Syst
   Evol Microbiol 52:1621-1628.

Alfa Aesar. (2015). A16163: Formaldehyde, 37% w/w aq. soln., stab, with 7-8% methanol. Available online at
   https://www.alfa.com/en/catalog/A16163 (accessed May 4,2015).

Ali. M: Taoutaou. S: Shafqat. AU: Salehapour. A: Noor. S. (2009). The use of self healing cement to ensure long
   term zonal isolation for HPHT wells subject to hydraulic fracturing operations in Pakistan. Paper
   presented at International Petroleum Technology Conference, December 7-9, 2009, Doha, Qatar.

ALL Consulting (ALL Consulting, LLC). (2013). Water treatment technology fact sheet: Electrodialysis [Fact
   Sheet]. Tulsa, OK. http://www.all-llc.com/publicdownloads/ED-EDRFactSheet.pdf

Alzahrani. S: Mohammad. AW: Hilal. N: Abdullah. P: laafar. 0. (2013). Comparative study of NF and RO
   membranes in the treatment of produced water-Part I: Assessing water quality. Desalination 315: 18-26.
   http://dx.doi.0rg/10.1016/j.desal.2012.12.004

Andre. L: Rabemanana. V: Vuataz. FD. (2006). Influence of water-rock interactions on fracture permeability of
   the deep reservoir at Soultz-sous-Forets, France. Geothermics 35: 507-531.
   http://dx.doi.0rg/10.1016/j.geothermics.2006.09.006

API (American Petroleum Institute). (1999). Recommended practice for care and use of casing and tubing
   [Standard] (18th ed.). (API RP 5C1). Washington, DC: API Publishing Services.
              This document is a draft for review purposes only and does not constitute Agency policy.

June 2015                                       1                    DRAFT—DO NOT CITE OR QUOTE

-------
Hydraulic Fracturing Drinking Water Assessment                                      All Appendices' References
API (American Petroleum Institute). (2004). Recommended practice for centralizer placement and stop collar
   testing (First ed.). (API RP 10D-2 (R2010)).

API (American Petroleum Institute). (2009a). Hydraulic fracturing operations - Well construction and
   integrity guidelines [Standard] (First ed.). Washington, DC: API Publishing Services.

API (American Petroleum Institute). (2009b). Packers and bridge plugs (Second ed.). (API Spec 11D1).

API (American Petroleum Institute). (2010a). Isolating potential flow zones during well construction
   [Standard] (1st ed.). (RP 65-2). Washington, DC: API Publishing Services.
   http://www.techstreet.com/products/preview/1695866

API (American Petroleum Institute). (2010b). Specification for cements and materials for well cementing
   [Standard] (24th ed.). (ANSI/API SPECIFICATION 10A). Washington, DC: API Publishing Services.
   http://www.techstreet.com/products/1757666

API (American Petroleum Institute). (2011). Specification for casing and tubing - Ninth edition [Standard]
   (9th ed.). (API SPEC 5CT). Washington, DC: API Publishing Services.
   http://www.techstreet.com/products/1802047

API (American Petroleum Institute). (2013). Recommended practice for testing well cements [Standard] (2nd
   ed.). (RP 10B-2). Washington, DC: API Publishing Services.
   http://www.techstreet.com/products/1855370

Arthur. ID. (2012). Understanding and assessing well integrity relative to wellbore stray gas intrusion issues.
   Presentation presented at Ground Water Protection Council Stray Gas - Incidence & Response Forum, July
   24-26,2012, Cleveland, OH.

Arthur, ID: Bohm, B:  Cornue, D. (2009). Environmental considerations of modern shale gas development.
   Paper presented at SPE Annual Technical Conference and Exhibition, October 4-7, 2009, New Orleans, LA.

Arthur. ID: Langhus.  BG: Patel. C. (2005). Technical summary of oil and gas produced water treatment
   technologies. Tulsa, OK: ALL Consulting, LLC.
   http://www.odinoilandgas.eom/Portals/0/TreatmentOptionsReport.pdf

ATS PR (Agency for Toxic Substances and Disease Registry). (2009). Glossary of terms. Available online at
   http://www.atsdr.cdc.gov/glossary.html

AWWA (American Water Works Association). (1999). Residential end uses of water. In PW Mayer; WB
   DeOreo (Eds.). Denver, CO: AWWA Research Foundation and American Water Works Association.
   http://www.waterrf.org/PublicReportLibrary/RFR90781 1999 241A.pdf

Baker. R. (1979). A primer of oilwell drilling (4th ed.). Austin, TX: Petroleum Extension Service (PETEX).

Banasiak. LI: Schafer. AI. (2009). Removal of boron, fluoride and nitrate by electrodialysis in the presence of
   organic matter. J Memb Sci 334:101-109. http://dx.doi.Org/10.1016/j.memsci.2009.02.020

Bank. T. (2011). Trace metal geochemistry and mobility in the Marcellus shale. In Proceedings of the
   Technical Workshops for the Hydraulic Fracturing Study: Chemical & Analytical Methods. Bank, T.
   http://www2.epa.gov/sites/production/files/documents/tracemetalgeochemistryandmobilityinthemarc
   ellusformationl.pdf

Bank. T: Fortson. LA:  Malizia. TR: Benelli. P. (2012). Trace metal  occurrences in the Marcellus Shale
   [Abstract]. Geological Society of America Abstracts with Programs  44: 313.

Baragi, IG: Maganur,  S: Malode, V: Baragi, SI. (2013). Excess molar volumes and refractive indices of binary
   liquid mixtures of acetyl acetone with n-Nonane, n-Decane and n-Dodecane at (298.15, 303.15, and
   308.15) K. Journal of Molecular Liquids 178:175-177. http://dx.doi.Org/10.1016/i.molliq.2012.ll.022

Barati. R: Liang. IT. (2014). A review of fracturing fluid systems used for hydraulic fracturing of oil and gas
   wells. J Appl Polymer Sci Online pub. http://dx.doi.org/10.1002/app.40735
               This document is a draft for review purposes only and does not constitute Agency policy.

June 2015                                         2                     DRAFT—DO NOT CITE OR QUOTE

-------
Hydraulic Fracturing Drinking Water Assessment                                       All Appendices' References
Barbot. E: Vidic. NS: Gregory. KB: Vidic. RD. (2013). Spatial and temporal correlation of water quality
   parameters of produced waters from Devonian-age shale following hydraulic fracturing. Environ Sci
   Technol 47: 2562-2569.

Barrett. ME. (2010). Evaluation of sand filter performance. (CRWR Online Report 10-7). Austin, TX: Center for
   Research in Water Resources, University of Texas at Austin.
   http://www.crwr.utexas.edu/reports/pdf/2010/rptlO-07.pdf

Benko, KL: Drewes, IE. (2008). Produced water in the Western United States: Geographical distribution,
   occurrence, and composition. Environ Eng Sci 25: 239-246.

Bennett, GM: Yuill, IL. (1935). The crystal form of anhydrous citric acid. J Chem Soc 1935:130.
   http://dx.doi.org/10.1039/IR9350000130

Biilmann. E. (1906). [Studien iiber organische Thiosauren III]. Justus Liebigs Annalen der Chemie 348:133-
   143. http://dx.doi.org/10.1002/ilac.19063480110

Biltz. W: Balz. G. (1928). [Uber molekular- und atomvolumina. XVIII. Das volumen des ammoniaks in
   kristallisierten ammoniumsalzen]. Zeitschrift fur Anorganische und Allgemeine Chemie 170: 327-341.
   http://dx.doi.org/10.1002/zaac.19281700141

Blanco. A: Garcia-Abuin. A: Gomez-Diaz. D: Navaza. IM: Villaverde. OL. (2013). Density, speed of sound,
   viscosity, surface tension, and excess volume of n-ethyl-2-pyrrolidone plus ethanolamine (or
   diethanolamine or triethanolamine) from T = (293.15 to 323.15) K. Journal of Chemical and Engineering
   Data 58: 653-659. http://dx.doi.org/10.1021/je301123j

Blauch. ME: Myers. RR: Moore. TR: Lipinski. BA. (2009). Marcellus shale post-frac flowback waters - where is
   all the salt coming from and what are the implications? In Proceedings of the SPE Eastern Regional
   Meeting. Richardson, TX: Society of Petroleum Engineers.

Blondes. MS: Cans. KD: Thordsen. II: Reidy. ME: Thomas. B: Engle. MA: Kharaka. YK: Rowan. EL. (2014). Data:
   U.S. Geological Survey National Produced Waters Geochemical Database v2.0 (Provisional) [Database]:
   U.S. Geological Survey:: USGS. Retrieved from
   http://energy.usgs.gov/EnvironmentalAspects/EnvironmentalAspectsofEnergyProductionandUse/Produ
   cedWaters.aspx#3822349-data

Bloomfield. C: Kelson. W: Pruden. G.  (1976). Reactions between metals and humidified organic matter.
   Journal of Soil Science 27:16-31. http://dx.doi.Org/10.llll/i.1365-2389.1976.tb01971.x

Boschee.  P. (2012). Handling produced water from hydraulic fracturing. Oil and Gas Facilities 1: 23-26.

Boschee.  P. (2014). Produced and flowback water recycling and reuse: Economics, limitations, and
   technology. Oil and Gas Facilities 3:16-22.

Bottero. S: Picioreanu. C: Delft. TU: Enzien. M: van Loosdrecht. MCM: Bruining. H: Heimovaara. T. (2010).
   Formation damage and impact on gas flow caused by biofilms growing within proppant packing used in
   hydraulic fracturing. Paper presented at SPE International Symposium and Exhibiton on Formation
   Damage Control, February 10-12, 2010, Lafayette, Louisiana.

Brufatto.  C: Cochran. I: Conn. L: El-Zeghaty. SZA. A: Fraboulet. B: Griffin. T: lames. S: Munk. T: lustus. F: Levine.
   IR: Montgomery. C: Murphy. D: Pfeiffer. I: Pornpoch. T: Rishmani. L. (2003). From mud to cement -
   Building gas wells. Oilfield Rev 15: 62-76.

Bruff. M: likich. SA. (2011). Field demonstration of an integrated water treatment technology solution in
   Marcellus shale. Paper presented at SPE Eastern Regional Meeting, August 17-19,2011, Columbus, OH.

BTGS (Bureau of Topographic and Geologic Survey). (2011). Geochemical analyses of selected lithologies
   from geologic units in central, north-central, and southeastern Pennsylvania. (OFMI 1101.0). Middletown,
   PA: Bureau of Topographic and Geologic Survey, Pennsylvania Geological Survey.
               This document is a draft for review purposes only and does not constitute Agency policy.

June 2015                                        3                     DRAFT—DO NOT CITE OR QUOTE

-------
Hydraulic Fracturing Drinking Water Assessment                                       All Appendices' References
Bukhari. AA. (2008). Investigation of the electro-coagulation treatment process for the removal of total
   suspended solids and turbidity from municipal wastewater. Bioresour Technol 99: 914-921.
   http://dx.doi.0rg/10.1016/j.biortech.2007.03.015

CAPP (Canadian Association of Petroleum Producers). (2013). CAPP hydraulic fracturing operating practice:
   Wellbore construction and quality assurance. (2012-0034).
   http://www.capp.ca/getdoc.aspx?DocId=218137&DT=NTV

Carpenter, EL: Davis, HS. (1957). Acrylamide. Its preparation and properties. Journal of Applied Chemistry 7:
   671-676. http://dx.doi.org/10.1002/ictb.5010071206

Casanova, C: Wilhelm, E: Grolier, IPE: Kehiaian, HV. (1981). Excess volumes and excess heat-capacities of
   (water + alkanoic acid). The Journal of Chemical Thermodynamics 13: 241-248.
   http://dx.doi.org/10.1016/0021-9614f81)90123-3

Cayol. IL: Ollivier. B: Lawson anani soh. A: Fardeau. ML: Ageron. E: Grimont. PAD: Prensier. G: Guezennec. I:
   Magot. M: Garcia. IL. (1994). Haloincola saccharolytica subsp. senegalensis subsp. nov., Isolated from the
   sediments of a Hypersaline lake, and emended description of Haloincola saccharolytica. International
   Journal of Systematic Bacteriology 44: 805-811. http://dx.doi.org/10.1099/00207713-44-4-805

CCG (Chemical Computing Group). (2011). Molecular Operating Environment (MOE) Linux (Version 2011.10)
   [Computer Program]. Montreal, Quebec. Retrieved from http://www.chemcomp.com/software.htm

CCST (California Council on Science and Technology). (2014). Advanced well stimulation technologies in
   California: An independent review of scientific and technical information. Sacramento, CA.
   http://ccst.us/publications/2014/2014wst.pdf

CCST (California Council on Science and Technology). (2015). An independent scientifc assessment of well
   stimulation in California, Volume 1: Well stimulation technologies and their past, present, and potential
   future use in California. Sacramento, CA. http://www.ccst.us/publications/2015/2015SB4-vl.pdf

Chafer. A: Lladosa. E: Monton. IB: Cruz Burguet. M. a. (2010). Liquid-liquid equilibria for the system 1-methyl
   propyl ethanoate (1) + acetic acid (2) + water (3) at (283.15 and 323.15) K. Journal of Chemical and
   Engineering Data 55: 523-525. http://dx.doi.org/10.1021/ie900332x

Chapman, EC: Capo, RC: Stewart, BW: Kirby, CS: Hammack, RW: Schroeder, KT: Edenborn, HM. (2012).
   Geochemical and strontium isotope characterization of produced waters from Marcellus Shale natural gas
   extraction. Environ Sci Technol 46: 3545-3553.

Chasib. KF.  (2013).  Extraction of phenolic pollutants (phenol and p-chlorophenol) from industrial
   wastewater. Journal of Chemical and Engineering Data 58:1549-1564.
   http://dx.doi.org/10.1021/je4001284

Chermak. IA: Schreiber. ME.  (2014). Mineralogy and trace element geochemistry of gas shales in the United
   States: Environmental implications. Int J Coal Geol 126: 32-44.
   http://dx.doi.0rg/10.1016/i.coal.2013.12.005

Cheung. K: Klassen. P: Mayer. B: Goodarzi. F: Aravena. R. (2010). Major ion and isotope geochemistry of fluids
   and gases from coalbed methane and shallow groundwater wells in Alberta, Canada. Appl Geochem 25:
   1307-1329. http://dx.doi.0rg/10.1016/i.apgeochem.2010.06.002

Choppin.  GR. (2006). Actinide speciation in aquatic systems. Mar Chem 99: 83-92.
   http://dx.doi.0rg/10.1016/j.marchem.2005.003.011

Choppin.  GR. (2007). Actinide speciation in the environment. Journal of Radioanal Chem 273: 695-703.
   http://dx.doi.org/10.1007/sl0967-007-0933-3
               This document is a draft for review purposes only and does not constitute Agency policy.

June 2015                                        4                     DRAFT—DO NOT CITE OR QUOTE

-------
Hydraulic Fracturing Drinking Water Assessment                                      All Appendices' References
Clark. CE: Veil. IA. (2009). Produced water volumes and management practices in the United States (pp. 64).
   (ANL/EVS/R-09/1). Argonne, IL: Argonne National Laboratory.
   http://www.circleofblue.org/waternews/wp-
   content/uploads/2010/09/ANL EVS R09 produced water volume report 2437.pdf

Cluff. M: Hartsock. A: Macrae. I: Carter. K: Mouser. PI. (2014). Temporal changes in microbial ecology and
   geochemistry in produced water from hydraulically fractured Marcellus Shale Gas Wells. Environ Sci
   Technol 48: 6508-6517. http://dx.doi.org/10.1021/es501173p

Colborn. T: Kwiatkowski. C: Schultz. K: Bachran. M. (2011). Natural gas operations from a public health
   perspective. Hum Ecol Risk Assess 17:1039-1056. http://dx.doi.org/10.1080/10807039.2011.605662

Collado. L: Cleenwerck. I: Van Trappen. S: De Vos. P: Figueras. Ml. (2009). Arcobacter mytili sp. nov., an
   indoxyl acetate-hydrolysis-negative bacterium isolated from mussels. Int J Syst Evol Microbiol 59:1391-
   1396. http://dx.doi.0rg/10.1099/ijs.0.003749-0

Craft. R. (2004). Crashes involving trucks carrying hazardous materials. (FMCSA-RI-04-024). Washington,
   D.C.: U.S. Department of Transportation, http://ntl.bts.gov/lib/51000/51300/51302/fmcsa-ri-04-024.pdf

Cramer. DP. (2008). Stimulating unconventional reservoirs: Lessons learned, successful practices, areas for
   improvement. SPE Unconventional Reservoirs Conference, February 10-12, 2008, Keystone, CO.

Crescent (Crescent Consulting, LLC). (2011). East Mamm creek project drilling and cementing study.
   Oklahoma City, OK. http://cogcc.state.co.us/Library/PiceanceBasin/EastMammCreek/ReportFinal.pdf

Criquet. I: Allard. S:  Salhi. E: loll. CA: Heitz. A: von Gunten. U. rs. (2012). lodate and lodo-Trihalomethane
   Formation during Chlorination of Iodide-Containing Waters: Role of Bromide. Environ Sci Technol 46:
   7350-7357. http://dx.doi.org/10.1021/es301301g

Crook. R. (2008). Cementing: Cementing horizontal wells. Halliburton.

Curtis. IB. (2002). Fractured shale-gas systems. AAPG Bulletin 86:1921-1938.
   http://dx.doi.org/10.1306/61EEDDBE-173E-llD7-8645000102C1865D

Dahm. KG: Guerra. KL: Xu. P: Drewes. IE. (2011). Composite geochemical database for coalbed methane
   produced water quality in the Rocky Mountain region. Environ Sci Technol 45: 7655-7663.
   http://dx.doi.org/10.1021/es201021n

Pake, LP. (1978). Fundamentals of reservoir engineering. Boston, MA: Elsevier.
   http://www.ing.unp.edu.ar/asignaturas/reservorios/Fundamentals%20of%20Reservoir%20Engineering
   %20%28LP.%20Dake%29.pdf

Dao. TD: Mericq. IP: Laborie. S: Cabassud. C. (2013). A new method for permeability measurement of
   hydrophobic membranes in Vacuum Membrane Distillation process. Water Res 47:  20962104.

Davis, IP: Struchtemeyer, CG: Elshahed, MS. (2012). Bacterial communities associated with production
   facilities of two newly drilled thermogenic natural gas wells in the Barnett Shale (Texas, USA). Microb Ecol
   64: 942-954. http://dx.doi.org/10.1007/s00248-012-0073-3

de Oliveira. LH: da Silva. IL.  Ir: Aznar. M. (2011). Apparent and partial molar volumes at infinite dilution and
   solid-liquid equilibria of dibenzothiophene plus alkane systems. Journal of Chemical and Engineering Data
   56: 3955-3962. http://dx.doi.org/10.1021/ie200327s

Dejoye Tanzi. C: Abert Vian. M: Ginies. C: Elmaataoui. M: Chemat. F. (2012). Terpenes as green solvents for
   extraction of oil  from microalgae. Molecules 17: 8196-8205.
   http://dx.doi.org/10.3390/moleculesl7078196

Devereux.  S. (1998). Practical well planning and drilling manual. Tulsa, OK:  PennWell Publishing Company.
   http://www.pennwellbooks.com/practical-well-planning-and-drilling-manual/
               This document is a draft for review purposes only and does not constitute Agency policy.

June 2015                                        5                     DRAFT—DO NOT CITE OR QUOTE

-------
Hydraulic Fracturing Drinking Water Assessment                                      All Appendices' References
Dhondge. SS: Pandhurnekar. CP: Parwate. DV. (2010). Density, speed of sound, and refractive index of
   aqueous binary mixtures of some glycol ethers at T=298.15 K. Journal of Chemical and Engineering Data
   55: 3962-3968. http://dx.doi.org/10.1021/ie901072c

Diehl. SF: Goldhaber. MB: Hatch. IR. (2004). Modes of occurrence of mercury and other trace elements in coals
   from the warrior field, Black Warrior Basin, Northwestern Alabama. Int J Coal Geol 59:193-208.
   http://dx.doi.0rg/10.1016/i.coal.2004.02.003

DOE (U.S. Department of Energy). (2006). A guide to practical management of produced water from onshore
   oil and gas operations in the United States. Washington, DC: U.S. Department of Energy, National
   Petroleum Technology Office.
   http://fracfocus.0rg/sites/default/files/publications/a guide to practical management of produced wat
   er from onshore oil and gas operations in the united states.pdf

DOE (U.S. Department of Energy). (2014). Water management strategies for improved coalbed methane
   production in the Black Warrior Basin. Available online at http://www.netl.doe.gov/research/oil-and-
   gas/project-summaries/natural-gas-resources/de-fe0000888

Dresel, PE: Rose, AW.  (2010). Chemistry and origin of oil and gas well brines in western Pennsylvania (pp.
   48). (Open-File Report OFOG 1001.0). Harrisburg, PA: Pennsylvania Geological Survey, 4th ser.
   http://www.marcellus.psu.edu/resources/PDFs/brines.pdf

Drewes. I: Cath. T:  Debroux. I: Veil. I. (2009). An integrated framework for treatment and management of
   produced water - Technical assessment of produced water treatment technologies (1st ed.). (RPSEA
   Project 07122-12). Golden, CO: Colorado School of Mines.
   http://aqwatec.mines.edu/research/projects/Tech Assessment PW Treatment Tech.pdf

Dubey. GP: Kumar. K.  (2011). Thermodynamic properties of binary liquid mixtures of diethylenetriamine
   with alcohols at different temperatures. Thermochim Acta 524: 7-17.
   http://dx.doi.0rg/10.1016/j.tca.2011.06.003

Dubey. GP: Kumar. K.  (2013). Studies of thermodynamic, thermophysical and partial molar properties of
   liquid mixtures of diethylenetriamine with alcohols at 293.15 to 313.15 K. Journal of Molecular Liquids
   180:164-171. http://dx.doi.Org/10.1016/i.molliq.2013.01.011

Duhon, H. (2012).  Produced water treatment: Yesterday, today, and tomorrow. Oil and Gas Facilities 3: 29-31.

Dunkel. M. (2013). Reducing fresh water use in upstream oil and gas hydraulic fracturing. In Summary of the
   technical workshop on wastewater treatment and related modeling (pp. A37-A43). Irving, TX: Pioneer
   Natural Resources USA,  Inc. http://www2.epa.gov/hfstudy/summary-technical-workshop-wastewater-
   treatment-and-related-modeling

Duraisamy, RT: Beni, AH: Henni, A. (2013). State of the art treatment of produced water. In W Elshorbagy; RK
   Chowdhury (Eds.), Water treatment (pp. 199-222). Rijeka, Croatia: InTech.
   http://dx.doi.org/10.5772/53478

Dusseault. MB: Gray. MN: Nawrocki. PA. (2000). Why oilwells leak: Cement behavior and long-term
   consequences.  Paper presented at SPE International Oil and Gas Conference and Exhibition in China,
   November 7-10, 2000, Beijing, China.

Dyshin. AA: Eliseeva. 0V: Kiselev. MG: Al'per. GA. (2008). The volume characteristics of solution of
   naphthalene in heptane-ethanol mixtures at 298.15 K. Russian Journal of Physical Chemistry A, Focus on
   Chemistry 82: 1258-1261. http://dx.doi.org/10.1134/S0036024408080037

Easton. I. (2014). Optimizing fracking wastewater management. Pollution Engineering January 13.

Egorov. GI: Makarov. DM: Kolker. AM. (2013). Volume properties of liquid mixture of water plus glycerol over
   the temperature range from 278.15 to 348.15 K at atmospheric pressure. Thermochim Acta 570: 16-26.
   http://dx.doi.org/10.1016/i.tca.2013.07.012
               This document is a draft for review purposes only and does not constitute Agency policy.

June 2015                                        6                     DRAFT—DO NOT CITE OR QUOTE

-------
Hydraulic Fracturing Drinking Water Assessment                                      All Appendices' References
El A (Energy Information Administration). (2013). Technically recoverable shale oil and shale gas resources:
   an assessment of 137 shale formations in 41 countries outside the United States (pp. 730). Washington,
   D.C.: Energy Information Administration, U.S. Department of Energy.
   http://www.eia.gov/analysis/studies/worldshalegas/

Ely. IW: Horn. A: Cathey. R: Fraim. M: lakhete. S. (2011). Game changing technology for treating and recycling
   frac water. Paper presented at SPE Annual Technical Conference and Exhibition, October 30 - November 2,
   2011, Denver, CO.

Enform. (2013). Interim industry recommended practice 24: fracture stimulation: Interwellbore
   communication 3/27/2013 (1.0 ed.). (IRP 24). Calgary, Alberta: Enform Canada.
   http://www.enform.ca/safety resources/publications/PublicationDetails.aspx?a=29&type=irp

Engle. MA: Rowan. EL. (2014). Geochemical evolution of produced waters from hydraulic fracturing of the
   Marcellus Shale, northern Appalachian Basin: A multivariate compositional data analysis approach. Int J
   Coal Geol 126: 45-56. http://dx.doi.Org/10.1016/j.coal.2013.ll.010

ER (Eureka Resources, LLC). (2014). Crystallization technology. Available online at http://www.eureka-
   resources.com/wp-content/uploads/2013/07/EURE-022 Crystallization 53013.pdf (accessed March 4,
   2015).

Ertel. D: McManus. K: Bogdan. I. (2013). Marcellus wastewater treatment: Case study. In Summary of the
   technical workshop on wastewater treatment and related modeling (pp. A56-A66). Williamsport, PA:
   Eureka Resources, LLC. http://www2.epa.gov/hfstudy/summary-technical-workshop-wastewater-
   treatment-and-related-modeling

Fadeeva. YA: Shmukler. LE: Safonova. LP. (2004). Physicochemical properties of the H3P04-
   dimethylformamide system. Russian Journal of General Chemistry 74:174-178.
   http://dx.doi.Org/10.1023/B:RUGC.0000025496.07304.66

Fakhru'1-Razi. A: Pendashteh. A: Abdullah. LC: Biak. PR: Madaeni. SS: Abidin. ZZ. (2009). Review of
   technologies for oil and gas produced water treatment [Review]. J Hazard Mater 170: 530-551.

Faria. MAP: Martins. Rl: Cardoso. MIE. M: Barcia. OE. (2013). Density and viscosity of the binary systems
   ethanol + butan-1-ol, + pentan-1-ol, + heptan-1-ol, + octan-1-ol, nonan-1-ol, + decan-1-ol at 0.1 mpa and
   temperatures from 283.15 Kto 313.15 K. Journal of Chemical and Engineering Data 58: 3405-3419.
   http://dx.doi.org/10.1021/ie400630f

Pels. G. (1900). Ueber die Frage der isomorphen vertretung von halogen und hydroxyl. In Zeitschrift fur
   Kristallographie, Kristallgeometrie, Kristallphysik, Kristallchemie. Frankfurt: Leipzig.
   http://babel.hathitrust.org/cgi/pt?id=ucl.b3327977:view=lup:seq=5

Fertl, WH: Chilingar, GV. (1988). Total organic carbon content determined from well logs. SPE Formation
   Evaluation 3: 407-419. http://dx.doi.org/10.2118/15612-PA

Fichter, I: Moore, R: Braman, S: Wunch, K: Summer, E: Holmes, P. (2012). How hot is too hot for bacteria? A
   technical study assessing bacterial establishment in downhole drilling, fracturing, and stimulation
   operations. Paper presented at NACE International Corrosion Conference & Expo, March 11-15,2012, Salt
   Lake City, UT.

Filgueiras. AV: Lavilla. I: Bendicho. C. (2002). Chemical sequential extraction for metal partitioning in
   environmental solid samples. J Environ Monit 4: 823-857. http://dx.doi.org/10.1039/b207574c

finemech (finemech Precision Mechanical Components). (2012). Technical resources: Liquid nitrogen, LN2.
   Available online at http://www.finemech.com/tech resources/liquid nitrogen.html

Fisher. IG: Santamaria. A. (2002). Dissolved organic constituents in coal-associated waters and implications
   for human and ecosystem health. Paper presented at 9th Annual International Petroleum Environmental
   Conference, October 22-25, 2002, Albuquerque, NM.
               This document is a draft for review purposes only and does not constitute Agency policy.

June 2015                                        7                     DRAFT—DO NOT CITE OR QUOTE

-------
Hydraulic Fracturing Drinking Water Assessment                                       All Appendices' References
Fisher. RS. (1998). Geologic and geochemical controls on naturally occurring radioactive materials (NORM) in
   produced water from oil, gas, and geothermal operations. Environmental Geosciences 5:139-150.
   http://dx.doi.0rg/10.1046/i.1526-0984.1998.08018.x

Francis. Al. (2007). Microbial mobilization and immobilization of plutonium. J Alloy Comp 444: 500-505.
   http://dx.doi.0rg/10.1016/j.jallcom.2007.01.132

Francis. RA: Small. Ml: Vanbriesen. IM. (2009). Multivariate distributions of disinfection by-products in
   chlorinated drinking water. Water Res 43: 3453-3468. http://dx.doi.Org/10.1016/j.watres.2009.05.008

Fredrickson. IK: Balkwill. PL. (2006). Geomicrobial processes and biodiversity in the deep terrestrial
   subsurface. Geomicrobiologyjournal23: 345-356. http://dx.doi.org/10.1080/01490450600875571

Fuess. H: Bats. IW: Dannohl. H: Meyer. H: Schweig. A. (1982). Comparison of observed and calculated
   densities. XII. Deformation density in complex anions. II.  Experimental and theoretical densities in sodium
   formate. Acta Crystallogr B B38: 736-743. http://dx.doi.org/10.1107/S0567740882003999

Fujino. S: Hwang. C: Morinaga. K. (2004). Density, surface tension, and viscosity of PbO-B203-Si02 glass
   melts. Journal of the American  Ceramic Society 87:10-16. http://dx.doi.Org/10.llll/j.1151-
   2916.2004.tbl9937.x

Gadd. GM.  (2004). Microbial influence on metal mobility and application for bioremediation. Geoderma 122:
   109-119. http://dx.doi.0rg/10.1016/j.geoderma.2004.01.002

Garcia. MT: Mellado. E: Ostos. 1C: Ventosa. A. (2004). Halomonas organivorans sp. nov., a moderate halophile
   able to  degrade aromatic compounds. Int J Syst Evol Microbiol 54:1723-1728.
   http://dx.doi.0rg/10.1099/iis.0.63114-0

Gauthier, Ml: Lafay, B: Christen, R: Fernandez, L: Acquaviva,  M: Bonin, P: Bertrand, 1C. (1992). Marinobacter
   hydrocarbonoclasticus gen. nov., sp. nov., a New, Extremely Halotolerant, Hydrocarbon-Degrading Marine
   Bacterium. International Journal of Systematic Bacteriology 42: 568-576.
   http://dx.doi.org/10.1099/00207713-42-4-568

Gilmore. K: Hupp. R: Glathar. I. (2013). Transport of Hydraulic Fracturing Water and Wastes in the
   Susquehanna River Basin, Pennsylvania. J Environ Eng 140: B4013002.
   http://dx.doi.org/10.1061/(ASCE)EE.1943-7870.0000810

Glorius. M: Moll. H: Geipel. G: Bernhard. G. (2008). Complexation of uranium(VI) with aromatic acids such as
   hydroxamic and benzoic acid investigated by TRLFS. Journal of Radioanal Chem 277: 371-377.
   http://dx.doi.org/10.1007/sl0967-007-7082-6

Gomes. I: Cocke. D: Das. K: Guttula. M: Tran. D:  Beckman: I. (2009). Treatment of produced water by
   electrocoagulation. Shiner, TX:  KASELCO, LLC. http://www.kaselco.com/index.php/library/industry-
   white-papers

Goodwin, S: Carlson, K: Knox, K: Douglas, C: Rein, L. (2014). Water intensity assessment of shale gas resources
   in the Wallenberg field in norlheaslern  Colorado. Environ Sci Technol 48: 5991-5995.
   hllp://dx.doi.org/10.1021/es404675h

Grabowski. A: Nercessian. 0: Fayolle. F: Blanchel. D: leanlhon. C. (2005). Microbial diversity in production
   waters of a low-lemperalure biodegraded oil reservoir. FEMS Microbiol Ecol 54: 427-443.
   hllp://dx.doi.org/10.1016/i.femsec.2005.05.007

Guolin. I: Xiaoyu. W: Chunjie. H. (2008). The effecl of oilfield polymer-flooding waslewaler on anion exchange
   membrane performance. Desalination 220: 386-393.

Gupla. DVS: Valko. P. (2007). Fracluring fluids and formation damage. In M Economides; T Martin (Eds.),
   Modern fracluring: enhancing nalural gas production (pp. 227-279). Houston, TX: Energy Tribune
   Publishing Inc.
               This document is a draft for review purposes only and does not constitute Agency policy.

June 2015                                        8                     DRAFT—DO NOT CITE OR QUOTE

-------
Hydraulic Fracturing Drinking Water Assessment                                      All Appendices' References
GWPC and ALL Consulting (Ground Water Protection Council (GWPC) and ALL Consulting). (2009). Modern
   shale gas development in the United States: A primer. (DE-FG26-04NT15455). Washington, DC: U.S.
   Department of Energy, Office of Fossil Energy and National Energy Technology Laboratory.
   http://www.gwpc.org/sites/default/files/Shale%20Gas%20Primer%202009.pdf

Habuda-Stanic. M: Ravancic. ME: Flanagan. A. (2014). A Review on Adsorption of Fluoride from Aqueous
   Solution. Materials 7: 6317-6366. http://dx.doi.org/10.3390/ma7096317

Hagen, R: Kaatze, U. (2004). Conformational kinetics of disaccharides in aqueous solutions. J Chem Phys 120:
   9656-9664. http://dx.doi.Org/10.1063/l.1701835

Halldorson, B. (2013). Successful oilfield water management: Five unique case studies. Presentation
   presented at EPA Technical Workshop - Wastewater Treatment and Related Modeling Research, April 18,
   2013, Triangle Park, NC.

Halliburton. (2014). Hydraulic fracturing 101. Available online at
   http://www.halliburton.com/public/projects/pubsdata/hydraulic fracturing/fracturing 101.html

Haluszczak, LO: Rose, AW: Kump, LR. (2013). Geochemical evaluation of flowbackbrine from Marcellus gas
   wells in Pennsylvania, USA. Appl Geochem 28: 55-61.
   http://dx.doi.0rg/10.1016/j.apgeochem.2012.10.002

Hamieh. BM: Beckman. IR. (2006). Seawater desalination using Dewvaporation technique: theoretical
   development and design evolution. Desalination 195:1-13.
   http://dx.doi.0rg/10.1016/j.desal.2005.09.034

Hammer. R: VanBriesen. I. (2 012). In frackings wake: New rules are needed to protect our health and
   environment from contaminated wastewater. New York, NY: Natural Resources Defense Council.
   http://www.nrdc.org/energy/files/fracking-wastewater-fullreport.pdf

Hansen. E: Mulvaney. D: Betcher. M. (2013). Water resource reporting and water footprint from  Marcellus
   Shale development in West Virginia and Pennsylvania. Durango, CO: Earthworks Oil & Gas Accountability
   Project, http://www.downstreamstrategies.com/documents/reports  publication/marcellus wv pa.pdf

Harkness. IS: Dwyer. GS: Warner. NR: Parker. KM: Mitch. WA: Vengosh. A. (2015). Iodide, Bromide, and
   Ammonium in Hydraulic Fracturing and Oil and Gas Wastewaters: Environmental Implications. Environ
   Sci Technol 49:1955-1963. http://dx.doi.org/10.1021/es504654n

Harlow, A: Wiegand, G: Franck, EU. (1997). The Density of Ammonia at High Pressures to 723  K and 950 MPa.
   101: 1461-1465. http://dx.doi.org/10.1002/bbpc.199700007

Harwood. DW: Viner. IG: Russell. ER. (1993). Procedure for developing truck accident and release rates for
   hazmat routing. Journal of Transportation Engineering 119:189-199.
   http://dx.doi.org/10.1061/rASCE10733-947Xri9931119:2ri891

Hatch, IR: Leventhal, IS. (1981). Geochemistry of organic-rich shales and coals from middle Pennsylvanian
   Cherokee group and lower part of Marmaton group, Oklahoma, Kansas, Missouri, and Iowa [Abstract].
   AAPG Bulletin 65: 936.

Hayes. T. (2009). Sampling and analysis of water streams associated with the development of Marcellus shale
   gas. Des Plaines, IL: Marcellus Shale Coalition, http://eidmarcellus.org/wp-
   content/ uploads/2 012/11/MSCommission-Report.pdf

Hayes. T: Severin. B. (2012a). Characterization of flowback water from the the Marcellus and the Barnett
   shale regions. Barnett and Appalachian shale water management and reuse technologies. (08122-05.09;
   Contract 08122-05). Hayes, T; Severin, B. http://www.rpsea.org/media/files/project/2146b3aO/08122-
   05-RT-Characterization Flowback Waters Marcellus Barnett Shale Regions-03-20-12.pdf
               This document is a draft for review purposes only and does not constitute Agency policy.

June 2015                                        9                    DRAFT—DO NOT CITE OR QUOTE

-------
Hydraulic Fracturing Drinking Water Assessment                                      All Appendices' References
Hayes. T: Severin. BF. (2012b). Evaluation of the aqua-pure mechanical vapor recompression system in the
   treatment of shale gas flowback water - Barnett and Appalachian shale water management and reuse
   technologies. (08122-05.11). Hayes, T; Severin, BF. http://barnettshalewater.org/documents/08122-
   05.11-EvaluationofMVR-3-12-2012.pdf

Hayes. TD: Arthur. D. (2004). Overview of emerging produced water treatment technologies. Paper presented
   at llth Annual International Petroleum Environmental Conference, October 12-15, 2004, Albuquerque,
   NM.

Hayes. TD: Halldorson. B: Horner. P: Ewing. I: Werline. IR: Severin. BF. (2014). Mechanical vapor
   recompression for the treatment of shale-gas flowback water. Oil and Gas Facilities 3: 54-62.

Haynes. WM. (2014). CRC handbook of chemistry and physics. In WM Haynes (Ed.), (95 ed.). Boca Raton, FL:
   CRC Press, http://www.hbcpnetbase.com/

He. YM: liang. RF: Zhu. F: Luan. TG: Huang. ZO: Ouyang. GF. (2008). Excess  molar volumes and surface
   tensions of 1,2,4-trimethylbenzene and 1,3,5-trimethylbenzene with isopropyl acetate and isobutyl
   acetate at (298.15, 308.15, and 313.15)K. Journal of Chemical and Engineering Data 53:1186-1191.
   http://dx.doi.org/10.1021/je800046k

Hedlund. BP: Geiselbrecht. AD: Staley. IT. (2001). Marinobacter strain NCE312 has a Pseudomonas-like
   naphthalene dioxygenase. FEMS Microbiol Lett 201: 47-51.

Hernlem. Bl: Vane. LM: Sayles. GD. (1999). The application of siderophores for metal recovery and waste
   remediation: Examination of correlations for prediction of metal affinities. Water Res 33: 951-960.

Horsey. CA. (1981). Depositional environments of the Pennsylvanian Pottsville Formation in the Black
   Warrior Basin of Alabama. Journal of Sedimentary Research 51: 799-806.
   http://dx.doi.org/10.1306/212F7DB5-2B24-llD7-8648000102C1865D

House of Representatives (U.S. House of Representatives). (2011). Chemicals used in hydraulic fracturing.
   Washington, D.C.: U.S. House of Representatives, Committee on Energy and Commerce, Minority Staff.
   http://democrats.energycommerce.house.gov/sites/default/files/documents/Hydraulic-Fracturing-
   Chemicals-2011-4-18.pdf

Hua, GH: Reckhow, DA: Kim, I. (2006). Effect of bromide and iodide ions on the formation and speciation of
   disinfection byproducts during chlorination. Environ Sci Technol 40: 3050-3056.
   http://dx.doi.org/10.1021/es0519278

Huffman. HM: Fox. SW. (1938). Thermal data. X. The heats of combustion and free energies, at 25, of some
   organic compounds concerned in carbohydrate metabolism. J Am Chem Soc 60:1400-1403.
   http://dx.doi.org/10.1021/ja01273a036

Hyne. Ml. (2012). Nontechnical guide to petroleum geology, exploration, drilling and production. In
   Nontechnical guide to petroleum geology, exploration, drilling and production (3 ed.). Tulsa, OK: PennWell
   Corporation.

IARC (International Agency for Research on Cancer). (2015). IARC monographs - Classifications. Available
   online at http://monographs.iarc.fr/ENG/Classification/index.php

Igunnu. ET: Chen. GZ. (2014). Produced water treatment technologies. International Journal of Low-Carbon
   Technologies 9:157-177. http://dx.doi.org/10.1093/ijlct/cts049

Inc.. L. (2012). Leadscope [Computer Program]. Columbus, Ohio. Retrieved from http://www.leadscope.com

IUPAC (International Union of Pure and Applied Chemistry). (2014). Global availability of information on
   agrochemicals: Triisopropanolamine. Available online at
   http://sitem.herts.ac.uk/aeru/iupac/Reports/1338.htm
               This document is a draft for review purposes only and does not constitute Agency policy.

June 2015                                        10                     DRAFT—DO NOT CITE OR QUOTE

-------
Hydraulic Fracturing Drinking Water Assessment                                       All Appendices' References
lackson. RE: Gorody. AW: Mayer. B: Roy. IW: Ryan. MC: Van Stempvoort. DR. (2013). Groundwater protection
   and unconventional gas extraction: the critical need for field-based hydrogeological research. Ground
   Water 51: 488-510. http://dx.doi.org/10.llll/gwat.12074

liang. L: Guillot. D: Meraji. M: Kumari. P: Vidick. B: Duncan. B: Gaafar. GR: Sansudin. SB. (2012). Measuring
   isolation integrity in depleted reservoirs. SPWLA 53rd Annual Logging Symposium, June 16-20, 2012,
   Cartagena, Colombia.

lones, DB: Saglam, A: Song,  H: Karanfil, T. (2012). The impact of bromide/iodide concentration and ratio on
   iodinated trihalomethane formation and speciation. Water Res 46: 11-20.
   http://dx.doi.0rg/10.1016/j.watres.2011.10.005

Julian. IY: King. GE: Johns. IE: Sack. IK: Robertson. DB. (2007). Detecting ultrasmall leaks with ultrasonic leak
   detection, case histories from the North Slope, Alaska. Paper presented at International Oil Conference and
   Exhibition in Mexico, June 27-30, 2007, Veracruz, Mexico.

Kashem. MA: Singh. BR: Kondo. T: Huq. SMI: Kawai. S. (2007). Comparison of extractability of Cd, Cu, Pb and
   Zn with sequential extraction in contaminated and non-contaminated soils. Int J Environ Sci Tech 4: 169-
   176. http://dx.doi.org/10.1007/BF03326270

Katsoyiannis. A: Samara. C.  (2007). The fate of dissolved organic carbon (DOC) in the wastewater treatment
   process and its importance in the removal of wastewater contaminants. Environ Sci Pollut Res Int 14: 284-
   292. http://dx.doi.org/10.1065/espr2006.05.302

Kim. HM: Hwang. CY: Cho. BC. (2010). Arcobacter marinus sp. nov. Int J Syst Evol Microbiol 60: 531-536.
   http://dx.doi.0rg/10.1099/ijs.0.007740-0

Kimball. B. (2010). Water treatment technologies for global unconventional gas plays. Presentation presented
   at US - China Industry Oil and Gas Forum, September 16, 2010, Fort Worth, TX.

Kirksey. I. (2013). Optimizing wellbore integrity in well construction. Presentation presented at North
   American Wellbore Integrity Workshop, Octoberl6-17,2013, Denver, CO.

Kiselev. VD: Kashaeva. HA: Shakirova. II: Potapova. LN: Konovalov. AI. (2012). Solvent effect on the enthalpy
   of solution and partial molar volume of the ionic liquid l-butyl-3-methylimidazolium tetrafluoroborate.
   Journal of Solution Chemistry 41:1375-1387. http://dx.doi.org/10.1007/sl0953-012-9881-9

Rose. B: Ozgun. H: Ersahin.  ME: Dizge. N: Koseoglulmer. DY: Atay. B: Kaya. R: Altinbas. M: Sayili. S: Hoshan.  P:
   Atay, D: Eren, E: Kinaci, C:  Koyuncu, I. (2012). Performance evaluation of a submerged membrane
   bioreactor for the treatment of brackish oil and natural gas field produced water. Desalination 285:  295-
   300.

Kraemer. TF: Reid. DF. (1984). The occurrence and behavior of radium in saline formation water of the U.S.
   Gulf Coast region. Isotope Geoscience 2:153-174.

Krakowiak, I: Bobicz, D: Grzybkowski, W. (2001). Limiting partial molar volumes of tetra-n-alkylammonium
   perchlorates in N,N-dimethylacetamide, triethylphosphate and dimethyl sulfoxide atT=298.15 K. The
   Journal of Chemical Thermodynamics 33: 121-133. http://dx.doi.org/10.1006/jcht.2000.0725

Krasner. SW. (2009). The formation and control of emerging disinfection by-products of health concern
   [Review]. Philos Transact A Math Phys Eng Sci 367: 4077-4095. http://dx.doi.org/10.1098/rsta.2009.010

Laavi, H: Pokki, IP: Uusi-Kyyny, P: Massimi, A: Kim, Y: Sapei, E: Alopaeus,  V. (2013). Vapor-liquid equilibrium
   at 350 k, excess molar enthalpies at 298 K, and excess molar volumes at 298 K of binary mixtures
   containing ethyl acetate, butyl acetate, and 2-butanol. Journal of Chemical and Engineering Data 58: 1011-
   1019. http://dx.doi.org/10.1021/ie400036b
               This document is a draft for review purposes only and does not constitute Agency policy.

June 2015                                        11                    DRAFT—DO NOT CITE OR QUOTE

-------
Hydraulic Fracturing Drinking Water Assessment                                      All Appendices' References
Laavi. H: Zaitseva. A: Pokki. IP: Uusi-Kyyny. P: Kim. Y: Alopaeus. V. (2012). Vapor-liquid equilibrium, excess
   molar enthalpies, and excess molar volumes of binary mixtures containing methyl isobutyl ketone (MIBK)
   and 2-butanol, tert-pentanol, or 2-ethyl-l-hexanol. Journal of Chemical and Engineering Data 57: 3092-
   3101. http://dx.doi.org/10.1021/je300678r

Lalucat. I: Bennasar. A: Bosch. R: Garcia-Valdes. E: Palleroni. Ml. (2006). Biology of Pseudomonas stutzeri
   [Review]. Microbiol Mol Biol Rev 70: 510-547. http://dx.doi.org/10.1128/MMBR.00047-05

Langmuir, D: Herman, IS. (1980). The mobility of thorium in natural waters at low temperatures. Geochim
   Cosmo Act 44: 1753-1766. http://dx.doi.org/10.1016/0016-7037r80190226-4

Langmuir. D: Riese. AC. (1985). THE THERMODYNAMIC PROPERTIES OF RADIUM. Geochim Cosmo Act 49:
   1593-1601.

Laurenzi. II: Jersey. GR. (2013). Life cycle greenhouse gas emissions and freshwater consumption of Marcellus
   shale gas. Environ Sci Technol 47: 4896-4903. http://dx.doi.org/10.1021/es305162w

LEau LLC. (2008). Dew vaporation desalination 5,000-gallon-per-day pilot plant. (Desalination and Water
   Purification Research and Development Program Report No. 120). Denver, CO: Bureau of Reclamation, U.S.
   Department of the Interior. http://www.usbr.gov/research/AWT/reportpdfs/reportl20.pdf

Lee. K: Neff. I. (2011). Produced water: Environmental risks and advances in mitigation technologies. New
   York, NY: Springer, http://dx.doi.org/10.1007/978-l-4614-0046-2

Leventhal. IS: Hosterman. IW. (1982). Chemical and mineralogical analysis Of Devonian black-shale samples
   from Martin County, Kentucky - Caroll and Washington counties, Ohio - Wise County, Virginia - and
   Overton County, Tennessee, USA. Chem Geol 37: 239-264.

Li, L: Brantley, SL. (2011). Development of a subsurface reactive transport model for predicting potential
   water quality problems at Marcellus shale. (USGS Project 2011PA159B). Li, L; Brantley, SL.
   http://water.usgs.gov/wrri/2011grants/progress/2011PA159B.pdf

Lovley. PR: Chapelle. FH. (1995). Deep subsurface microbial processes. Rev Geophys 33: 365-381.
   http://dx.doi.org/10.1029/95RG01305

Lovley, PR: Phillips, El. (1986). Organic matter mineralization with reduction of ferric iron in anaerobic
   sediments. Appl Environ Microbiol 51:  683-689.

Ludzack,  Fl: Noran, DK. (1965). Tolerance  of high salinities by conventional wastewater treatment processes.
   J Water Pollut Control Fed 37:1404-1416.

Luh. I: Marinas. Bl. (2012). Bromide ion effect on  N-nitrosodimethylamine formation by monochloramine.
   Environ Sci Technol 46:  5085-5092. http://dx.doi.org/10.1021/es300077x

Lyons. WC: Pligsa. Gl. (2004). Standard handbook of petroleum and natural gas engineering (2nd ed.).
   Houston, TX: Gulf Professional Publishing, http://www.elsevier.com/books/standard-handbook-of-
   petroleum-and-natural-gas-engineering/lyons-phd-pe/978-0-7506-7785-l

Ma, G: Geza, M: Xu, P. (2014). Review of flowback and produced water management, treatment, and beneficial
   use for major shale gas development basins. Shale Energy Engineering Conference 2014, Pittsburgh,
   Pennsylvania, United States.

Mak. TCW. (1965). Hexamethylenetetramine hexahydrate: A new type of clathrate hydrate. J Chem Phys 43:
   2799. http://dx.doi.0rg/10.1063/l.1697212

Manios, T: Stentiford, El: Millner, P. (2003). Removal of total suspended solids from wastewater in
   constructed horizontal flow subsurface wetlands. J Environ Sci Health A Tox Hazard Subst Environ Eng 38:
   1073-1085. http://dx.doi.org/10.1081/ESE-120019865
               This document is a draft for review purposes only and does not constitute Agency policy.

June 2015                                        12                     DRAFT—DO NOT CITE OR QUOTE

-------
Hydraulic Fracturing Drinking Water Assessment                                      All Appendices' References
Martinez-Reina. M: Amado-Gonzalez. E: Mauricio Munoz-Munoz. Y. (2012). Study of liquid-liquid equilibria of
   toluene plus (hexane, heptane, or cyclohexane) with l-ethyl-3-methylimidazolium ethylsulfate at 308.15
   K. Bull Chem Soc Jpn 85: 1138-1144. http://dx.doi.org/10.1246/bcsi.20120112

Martinez. CE: McBride. MB. (2001). Cd, Cu, Pb, and Zn coprecipitates in Fe oxide formed at different pH: Aging
   effects on metal solubility and extractability by citrate. Environ Toxicol Chem 20:122-126.
   http://dx.doi.org/10.1002/etc.5620200112

Masood, ARM: Pethrick, RA: Swinton, FL. (1976). Physicochemical studies of super-cooled liquids - cyclic
   carbonates and alpha.beta-unsaturated aldehydes. Faraday Trans 1 72: 20-28.
   http://dx.doi.org/10.1039/fl9767200020

Mata. IA: Martinez-Canovas. I: Quesada. E: Bejar. V. (2002). A detailed phenotypic characterisation of the type
   strains of Halomonas species. Syst Appl Microbiol 25: 360-375. http://dx.doi.org/10.1078/0723-2020-
   00122

Matamoros. V: Mujeriego. R: Bayona. IM. (2007). Trihalomethane occurrence in chlorinated reclaimed water
   at full-scale wastewater treatment plants in NE Spain. Water Res 41: 3337-3344.
   http://dx.doi.0rg/10.1016/j.watres.2007.04.021

Maupin. MA: Kenny. IF: Hutson. SS: Lovelace. IK: Barber. NL: Linsey. KS. (2014). Estimated use of water in the
   United States in 2010. (USGS Circular 1405). Reston, VA: U.S. Geological Survey.
   http://dx.doi.org/10.3133/cirl405

Maxwell. SC. (2011). Hydraulic fracture height growth. Recorder 36: 18-22.

McDaniel. BW: Rispler. KA. (2009). Horizontal wells with multistage fracs prove to be best economic
   completion for many low permeability reservoirs. Paper presented at SPE Eastern Regional Meeting,
   September 23-15, 2009, Charleston, WV.

McDaniel. I: Watters. L: Shadravan. A. (2014). Cement sheath durability: Increasing cement sheath integrity to
   reduce gas migration in the Marcellus Shale Play. In SPE hydraulic fracturing technology conference
   proceedings. Richardson, TX: Society of Petroleum Engineers. http://dx.doi.org/10.2118/168650-MS

McGowan. L: Herbert. R: Muyzer. G. (2004). A comparative study of hydrocarbon degradation by
   Marinobacter sp., Rhodococcus sp. and Corynebacterium sp. isolated from different mat systems. Ophelia
   58: 271-281. http://dx.doi.org/10.1080/00785236.2004.10410235

McGuire, Ml: Karanfil, T: Krasner, SW: Reckhow, DA: Roberson, I A; Summers, RS: Westerhoff, P: Xie, Y. (2014).
   Not your granddad's disinfection by-product problems and solutions. JAWWA 106: 54-73.
   http://dx.doi.org/10.5942/jawwa.2014.106.0128

McLin. K: Brinton. D: Moore. I. (2011). Geochemical modeling of water-rock-proppant interactions. Thirty-
   Sixth Workshop on Geothermal Reservoir Engineering, January 31 - February 2,2011, Stanford University,
   Stanford, California.

Miller. P. (2011). Future of hydraulic fracturing depends on effective water treatment. Hydrocarbon Process
   90:13-13.

Mitchell. AL: Small. M: Gasman. EA. (2013). Surface water withdrawals for Marcellus Shale gas development:
   performance of alternative regulatory approaches in the Upper Ohio River Basin. Environ Sci Technol 47:
   12669-12678. http://dx.doi.org/10.1021/es403537z

Mohan. AM: Gregory. KB: Vidic. RD: Miller. P: Hammack. RW. (2011). Characterization of microbial diversity
   in treated and untreated flowback water impoundments from gas fracturing operations. Paper presented
   at SPE Annual Technical Conference and Exhibition,  October 30 - November 2, 2011, Denver, CO.

Montgomery. C. (2013). Fracturing fluid components. In A Bunder; J McLennon; R Jeffrey (Eds.), Effective and
   Sustainable Hydraulic Fracturing. Croatia: InTech. http://dx.doi.org/10.5772/56422
               This document is a draft for review purposes only and does not constitute Agency policy.

June 2015                                        13                    DRAFT—DO NOT CITE OR QUOTE

-------
Hydraulic Fracturing Drinking Water Assessment                                      All Appendices' References
Moosavi. M: Motahari. A: Omrani. A: Rostami. AA. (2013). Thermodynamic study on some alkanediol
   solutions: Measurement and modeling. Thermochim Acta 561:1-13.
   http://dx.doi.0rg/10.1016/i.tca.2013.03.010

Moses. CO: Herman. IS. (1991). Pyrite oxidation at circumneutral pH. Geochim Cosmo Act 55: 471-482.

MSC (Marcellus Shale Coalition). (2013). Recommended practices: Drilling and completions. (MSC RP 2013-
   3). Pittsburgh, Pennsylvania.

Munter. R. (2000). Industrial wastewater treatment. In LC Lundin (Ed.), Sustainable water management in the
   Baltic Sea Basin book II: Water use and management (pp. 195-210). Sida, Sweden: Baltic University
   Programme Publication, http://www.balticuniv.uu.se/index.php/boll-online-library/831-swm-2-water-
   use-and-management

Murali Mohan. A: Hartsock. A: Bibby. Kl: Hammack. RW: Vidic. RD: Gregory. KB. (2013a). Microbial
   community changes in hydraulic fracturing fluids and produced water from shale gas extraction. Environ
   Sci Technol 47:13141-13150. http://dx.doi.org/10.1021/es402928b

Murali Mohan, A: Hartsock, A: Hammack, RW: Vidic, RD: Gregory, KB. (2013b).  Microbial communities in
   flowback water impoundments from hydraulic fracturing for recovery of shale gas. FEMS Microbiol Ecol.
   http://dx.doi.org/10.llll/1574-6941.12183

Murray. KE. (2013). State-scale perspective on water use  and production associated with oil and gas
   operations, Oklahoma, U.S. Environ Sci Technol 47: 4918-4925. http://dx.doi.org/10.1021/es4000593

Muylaert. K: Dasseville. R: De Brabandere. L: Dehairs. F: Vyverman. W. (2005).  Dissolved organic carbon in
   the freshwater tidal reaches of the Schelde estuary. Estuar Coast Shelf Sci 64: 591-600.
   http://dx.doi.0rg/10.1016/j.ecss.2005.04.010

Myers. CR: Nealson. KH. (1988). Bacterial manganese reduction and growth with manganese oxide as the sole
   electron acceptor. Science 240:1319-1321. http://dx.doi.org/10.1126/science.240.4857.1319

Newman. DK  (2001). Microbiology- How bacteria respire minerals. Science 292:1312-1313.
   http://dx.doi.org/10.1126/science.1060572

Nicot, IP: Reedy, RC: Costley, RA: Huang, Y. (2012). Oil & gas water use in Texas: Update to the 2011 mining
   water use report Nicot, JP; Reedy, RC; Costley, RA; Huang, Y.
   http://www.twdb.state.tx.us/publications/reports/contracted reports/doc/0904830939 2012Update M
   iningWaterUse.pdf

Nicot. IP: Scanlon. BR. (2012). Water use for shale-gas production in Texas, U.S. Environ Sci Technol 46: 3580-
   3586. http://dx.doi.org/10.1021/es204602t

Nicot. IP: Scanlon. BR: Reedy. RC: Costley. RA. (2014). Source and fate of hydraulic fracturing water in the
   Barnett Shale: a historical perspective. Environ Sci Technol 48: 2464-2471.
   http://dx.doi.org/10.1021/es404050r

NLM (National Institutes of Health, National Library of Medicine). (2014). ChemID plus advanced. Available
   online at http://chem.sis.nlm.nih.gov/chemidplus/

NMSU DACC WUTAP (New Mexico State University, Dona Ana Community College, Water Utilities Technical
   Assistance Program). (2007). New Mexico wastewater systems operator certification study manual -
   Version 1.1. Santa Fe, NM: New Mexico Environment Department.
   http://www.nmrwa.org/sites/nmrwa.org/files/WastewaterOperatorStudyManual.pdf

Nordstrom, DK: Alpers, CN. (1999). Geochemistry of acid  mine waters. In GS Plumlee; MJ Logsdon (Eds.), The
   Environmental Geochemistry of Mineral Deposits Part A: Processes, Techniques, and Health Issues Society
   of Economic Geologists, Denver (pp. 133-160). Littleton, CO: Society of Economic Geologists.

North Dakota  State Water Commission. (2014). Facts about North Dakota tracking and water use. Bismarck,
   ND. http://www.swc.nd.gov/4dlink9/4dcgi/GetContentPDF/PB-2419/Fact%20Sheet.pdf
               This document is a draft for review purposes only and does not constitute Agency policy.

June 2015                                        14                     DRAFT—DO NOT CITE OR QUOTE

-------
Hydraulic Fracturing Drinking Water Assessment                                      All Appendices' References
NTP (National Toxicology Program). (2014a). Definition of carcinogenicity results. Available online at
   http://ntp.niehs.nih.gov/results/pubs/longterm/defs/index.html

NTP (National Toxicology Program). (2014b). Report on carcinogens. Thirteenth edition. Research Triangle
   Park, NC: U.S. Department of Health and Human Services, Public Health Service.
   http://ntp.niehs.nih.gov/pubhealth/roc/rocl3/index.html

NYSDEC (New York State Department of Environmental Conservation). (2011). Revised draft supplemental
   generic environmental impact statement (SGEIS) on the oil, gas and solution mining regulatory program:
   Well permit issuance for horizontal drilling and high-volume hydraulic fracturing to develop the Marcellus
   shale and other low-permeability gas reservoirs. Albany, NY: NY SDEC.
   http://www.dec.ny.gov/energy/75370.html

Obolensky. A: Singer. PC.  (2008). Development and interpretation of disinfection byproduct formation models
   using the Information Collection Rule database. Environ Sci Technol 42: 5654-5660.
   http://dx.doi.org/10.1021/es702974f

OEHHA. Title 27. California Code of Regulations Article 8. No Observable Effect Levels. § 25701 (2012).
   http://www.oehha.ca.gov/prop65/law/pdf zip/RegsArt8.pdf

Oil and Gas Mineral Services. (2010). MineralWise: Oil and gas terminology. Available online at
   http://www.mineralweb.com/library/oil-and-gas-terms/

Oka. S. (1962). Studies on lactone formation in vapor phase. III. Mechanism of lactone formation from diols.
   Bull Chem Soc Jpn 35: 986-989. http://dx.doi.org/10.1246/bcsi.35.986

ONG Services. (2015). ONGList: Reserved Environmental Services. Available online at
   http://www.onglist.com/Home/Search?SearchString=Reserved+environmental+services&Distance=&sea
   rchAddress=&CategoryTypeID=l&SubCategoryID

Orem. W: Tatu. C: Varonka. M: Lerch. H: Bates. A: Engle. M: Crosby. L: Mcintosh. I. (2014). Organic substances
   in produced and formation water from unconventional natural gas extraction in coal and shale. Int J Coal
   Geol 126: 20-31. http://dx.doi.Org/10.1016/j.coal.2014.01.003

Orem. WH: Tatu. CA: Lerch. HE: Rice. CA: Bartos. TT: Bates. AL: Tewalt. S: Corum. MD. (2007). Organic
   compounds in produced waters from coalbed natural gas wells in the Powder River Basin, Wyoming, USA.
   Appl Geochem 22: 2240-2256. http://dx.doi.Org/10.1016/i.apgeochem.2007.04.010

OSHA. Title 29 - Department of Labor. Subpart z Toxic and hazardous substances, hazard communication, §
   1910.1200 (2013). http://www.gpo.gov/fdsys/pkg/CFR-2013-title29-vol6/xml/CFR-2013-title29-vol6-
   secl910-1200.xml

PA PEP (Pennsylvania Department of Environmental Protection). (2010). Chemicals used by hydraulic
   fracturing companies  in Pennsylvania for surface and hydraulic fracturing activities. Harrisburg, PA:
   Pennsylvania Department of Environmental Protection (PADEP).
   http://files.dep.state.pa.us/OilGas/BOGM/BOGMPortalFiles/MarcellusShale/Frac%201ist%206-30-
   2010.pdf.

PA PEP (Pennsylvania Department of Environmental Protection). (2015). Technologically enhanced naturally
   occurring radioactive  materials (TENORM) study report. Harrisburg, PA.
   http://www.elibrary.dep.state.pa.us/dsweb/Get/Document-105822/PA-DEP-TENORM-
   Study Report Rev. 0  01-15-2015.pdf

Pal. A: Kumar. H: Maan. R: Sharma. HK. (2013). Densities and speeds of sound of binary liquid mixtures of
   some n-alkoxypropanols with methyl acetate, ethyl acetate, and n-butyl acetate at T = (288.15, 293.15,
   298.15, 303.15, and 308.15) K. Journal of Chemical and Engineering Data 58: 225-239.
   http://dx.doi.org/10.1021/ie300789a
               This document is a draft for review purposes only and does not constitute Agency policy.

June 2015                                        15                     DRAFT—DO NOT CITE OR QUOTE

-------
Hydraulic Fracturing Drinking Water Assessment                                       All Appendices' References
Parker. KM: Zeng. T: Harkness. I: Vengosh. A: Mitch. WA. (2014). Enhanced formation of disinfection
   byproducts in shale gas wastewater-impacted drinking water supplies. Environ Sci Technol 48:11161-
   11169. http://dx.doi.org/10.1021/es5028184

Pashin. 1C: Mcintyre-Redden. MR: Mann. SD: Kopaska-Merkel. DC: Varonka. M: Orem. W. (2014). Relationships
   between water and gas chemistry in mature coalbed methane reservoirs of the Black Warrior Basin. Int J
   Coal Geol 126: 92-105. http://dx.doi.Org/10.1016/i.coal.2013.10.002

Pijper, WP. (1971). Molecular and crystal structure of glycollie acid. Acta Crystallogr B B27: 344-348.
   http://dx.doi.org/10.1107/S056774087100219X

Pope, PG: Martin-Doole, M: Speitel, GE: Collins, MR. (2007). Relative significance of factors influencing DXAA
   formation during chloramination. JAWWA 99:144-156.

Radwan. MHS:  Hanna. AA. (1976). Binary azeotropes containing butyric acids. Journal of Chemical and
   Engineering Data 21: 285-289. http://dx.doi.org/10.1021/je60070a032

Rahm. BG: Bates. IT: Bertoia. LR: Galford. AE: Yoxtheimer. DA: Riha. SI. (2013). Wastewater management and
   Marcellus Shale gas development: trends, drivers, and planning implications. J Environ Manage 120:105-
   113. http://dx.doi.0rg/10.1016/j.ienvman.2013.02.029

Rani. M: Maken. S. (2013). Excess molar enthalpies and excess molar volumes of formamide+1-propanol or 2-
   propanol and thermodynamic modeling by Prigogine-Flory-Patterson theory and Treszczanowicz-Benson
   association model. Thermochim Acta 559: 98-106. http://dx.doi.Org/10.1016/j.tca.2013.02.010

Ravot. G: Magot. M: Ollivier. B: Patel. BK: Ageron. E: Grimont. PA: Thomas. P: Garcia. IL. (1997).
   Haloanaerobium congolense sp. nov., an anaerobic, moderately halophilic, thiosulfate- and sulfur-reducing
   bacterium from an African oil field. FEMS Microbiol Lett 147: 81-88.

Rawat. BS: Gulati. IB: Mallik. KL. (1976). Study of some sulphur-group solvents for aromatics extraction by gas
   chromatography. Journal of Applied Chemistry and Biotechnology 26: 247-252.
   http://dx.doi.org/10.1002/ictb.5020260504

Renpu. W. (2011). Advanced well completion engineering (Third ed.). Houston, TX: Gulf Professional
   Publishing.

Rice. CA: Flores. RM: Strieker. GD: Ellis. MS. (2008). Chemical and stable isotopic evidence for water/rock
   interaction  and biogenic origin of coalbed methane, Fort Union Formation, Powder River Basin, Wyoming
   and Montana USA. Int J Coal Geol 76: 76-85. http://dx.doi.Org/10.1016/j.coal.2008.05.002

Richardson. SD: Plewa. Ml: Wagner. ED: Schoeny. R: Demarini. DM. (2007). Occurrence, genotoxicity, and
   carcinogenicity of regulated and emerging disinfection by-products in drinking water: A review and
   roadmap for research [Review]. Mutat Res 636:178-242. http://dx.doi.Org/10.1016/j.mrrev.2007.09.00

Rodnikova. MN: Solonina. IA: Egorov. GI: Makarov. DM: Gunina. MA. (2012). The bulk properties of dioxane
   solutions in ethylene glycol at 2575C. Russian Journal of Physical Chemistry A, Focus on Chemistry 86:
   330-332. http://dx.doi.org/10.1134/S0036024412020239

Ross. D: King. G. (2007). Well completions. In MJ Economides; T Martin (Eds.), Modern fracturing: Enhancing
   natural gas  production (1 ed., pp. 169-198). Houston, Texas: ET Publishing.

Rowan. EL: Engle. MA: Kirby. CS: Kraemer. TF. (2011). Radium content of oil- and gas-field produced waters
   in the northern Appalachian Basin (USA): Summary and discussion of data. (Scientific Investigations
   Report 20115135). Reston, VA: U.S. Geological Survey, http://pubs.usgs.gov/sir/2011/5135/

Roy, SB: Ricci, PF: Summers, KV: Chung, CF: Goldstein, RA. (2005). Evaluation of the sustainability of water
   withdrawals in the United States, 1995 to 2025. J Am Water Resour Assoc 41: 1091-1108.

Rozell. PI: Reaven. SI. (2012). Water pollution risk associated with natural gas extraction from the Marcellus
   Shale. Risk Anal 32:13821393. http://dx.doi.Org/10.llll/i.1539-6924.2011.01757.x
               This document is a draft for review purposes only and does not constitute Agency policy.

June 2015                                        16                     DRAFT—DO NOT CITE OR QUOTE

-------
Hydraulic Fracturing Drinking Water Assessment                                      All Appendices' References
Sabins. F. (19901. Problems in cementing horizontal wells. J Pet Tech 42: 398-400.
   http://dx.doi.org/10.2118/20005-PA

Safe Drinking Water Act. Title XIV of the Public Health Service Act Safety of Public Water Systems (Safe
   Drinking Water Act) as amended through P.L. 107-377, (2002). http://www.epw.senate.gov/sdwa.pdf

Sarkar. BK: Choudhury. A: Sinha. B. (2012). Excess molar volumes, excess viscosities and ultrasonic speeds of
   sound of binary mixtures of 1,2-dimethoxyethane with some aromatic liquids at 298.15 K. Journal of
   Solution Chemistry 41: 53-74. http://dx.doi.org/10.1007/sl0953-011-9780-5

Sarkar. L: Roy. MN. (2009). Density, viscosity, refractive index, and ultrasonic speed of binary mixtures of 1,3-
   dioxolane with 2-methoxyethanol, 2-ethoxyethanol, 2-butoxyethanol, 2-propylamine, and
   cyclohexylamine. Journal of Chemical and Engineering Data 54: 3307-3312.
   http://dx.doi.org/10.1021/je900240s

Schlumberger (Schlumberger Limited). (2014). Schlumberger oilfield glossary. Available online at
   http://www.glossary.oilfield.slb.com/

Schrodinger. (2012). Qikprop [Computer Program]. New York, New York: Schrodinger, LLC. Retrieved from
   http://www.schrodinger.com/products/14/17

Shafer. L. (2011). Water recycling and purification in the Pinedale anticline field: results from the anticline
   disposal project. In 2011 SPE Americas E&P health, safety, security & environmental conference.
   Richardson, TX: Society of Petroleum Engineers. http://dx.doi.org/10.2118/141448-MS

Shaffer. PL: Arias Chavez. LH: Ben-Sasson. M: Romero-Vargas Castrillon. S: Yip. NY: Elimelech. M.  (2013).
   Desalination and reuse of high-salinity shale gas produced water: drivers, technologies, and future
   directions. Environ Sci Technol 47:  9569-9583.

Shammas. NK. (2010). Wastewater renovation by flotation. In LK Wang; NK Shammas; WA Selke;  DB
   Aulenbach (Eds.), Flotation technology (pp. 327-345). New York, NY: Humana Press.
   http://dx.doi.org/10.1007/978-l-60327-133-2  9

Shanley. P: Collin. RL. (1961). The crystal structure of the high temperature form of choline chloride. Acta
   Cryst 14: 79-80. http://dx.doi.org/10.1107/S0365110X61000292

Sheets. MSP, (a) Encana/Halliburton Energy Services, Inc.: Duncan, Oklahoma. Provided by Halliburton
   Energy Services during an onsite visit by the EPA on May 10, 2010; (b) Encana Oil and Gas (USA), Inc.:
   Denver, Colorado. Provided to US EPA Region 8. Material Safety Data Sheets.

Sigma-Aldrich. (2007). Material safety  data sheet: Tert-butyl hydroperoxide (70% solution in water).
   Available online at http://www.orcbs.msu.edu/msds/111607 DLI 027 TERT-BUTYL.PDF

Sigma-Aldrich. (2010). Product information: Sodium chloride. Available online at
   https:// www.sigmaaldrich.com/content/dam/sigma-aldrich/docs/Sigma-
   Aldrich/Product Information Sheet/s7653pis.pdf

Sigma-Aldrich. (2014a). Material safety data sheet:  Phosphorus acid. Available online at
   http://www.sigmaaldrich.eom/catalog/product/sial/215112?lang=en®ion=US

Sigma-Aldrich. (2014b). Material safety data sheet: Potassium carbonate. Available online at
   http://www.sigmaaldrich.com/catalog/product/aldrich/367877?lang=en®ion=US

Sigma-Aldrich. (2015a). Material safety data sheet:  Aluminum chloride [Fact Sheet]. St. Louis, MO.
   http://www.sigmaaldrich.com/catalog/product/aldrich/563919?lang=en®ion=US

Sigma-Aldrich. (2015b). Material safety data sheet: Peracetic acid solution. Available online at
   http://www.sigmaaldrich.eom/catalog/product/sial/2 69336?lang=en®ion=US

Sigma-Aldrich. (2015c). Material safety data sheet:  Sulfur dioxide. Available online at
   http://www.sigmaaldrich.com/catalog/product/aldrich/295698?lang=en®ion=US
               This document is a draft for review purposes only and does not constitute Agency policy.

June 2015                                        17                    DRAFT—DO NOT CITE OR QUOTE

-------
Hydraulic Fracturing Drinking Water Assessment                                      All Appendices' References
Sigma-Aldrich. (2015d). Material safety data sheet: Sulfuric acid. Available online at
   http://www.sigmaaldrich.com/catalog/product/aldrich/339741?lang=en®ion=US

Sigma-Aldrich. (2015e). Material safety data sheet: Trimethyl borate. Available online at
   http://www.sigmaaldrich.com/catalog/product/aldrich/447218?lang=en®ion=US

Singer. P. (2010). Anomalous DBF speciation patterns: Examples and explanations. Water Quality Technology
   Conference and Exposition 2010, November, 14-18, 2010, Savannah, GA.

Sirivedhin. T: Dallbauman. L. (2004). Organic matrix in produced water from the Osage-Skiatook petroleum
   environmental research site, Osage county, Oklahoma. Chemosphere 57: 463-469.

Skierven. T: Lunde. 0: Perander. M: Williams. B: Farquhar. R: Sinet. I: Sasby. I: Haga. HB: Finnseth. 0: lohnsen.
   S. (2011). Norwegian Oil and Gas Association recommended guidelines for well integrity. (117, Revision
   4). Norway: Norwegian Oil and Gas Association.
   http://www.norskoljeoggass.no/Global/Retningslinjer/Boring/117%20-
   %20Recommended%20guidelines%20Well%20integrity%20rev4%2006.06.%2011.pdf

Smirnov, VI: Badelin, VG. (2013). Enthalpy characteristics of dissolution of L-tryptophan in water plus
   formamides binary solvents at 298.15 K. Russian Journal of Physical Chemistry A, Focus on Chemistry 87:
   1165-1169. http://dx.doi.org/10.1134/S0036024413070285

Solley. WB: Pierce. RR: Perlman. HA.  (1998). Estimated use of water in the United States in 1995. (USGS
   Circular: 1200). U.S. Geological Survey, http://pubs.er.usgs.gov/publication/cirl200

Spellman. FR. (2012). Environmental impacts of hydraulic fracturing. In Environmental impacts of hydraulic
   fracturing. Boca Raton, Florida: CRC Press.

SRBC. (2012). Flowback and Produced Water Volume.

States. S: Cyprych. G: Stoner. M: Wydra. F: Kuchta. I: Monnell. I: Casson. L. (2013). Marcellus Shale drilling and
   brominated THMs in Pittsburgh, Pa., drinking water. J Am Water Works  Assoc 105: E432-E448.
   http://dx.doi.org/10.5942/jawwa.2013.105.0093

Stein. D: Griffin Jr. Tl: Dusterhoft. D. (2003). Cement pulsation reduces remedial cementing costs. GasTIPS 9:
   22-24.

Steinhauser. 0: Boresch. S: Bertagnolli. H. (1990). The effect of density variation on the structure of liquid
   hydrogen chloride. A Monte Carlo study. J Chem Phys 93: 2357-2363.
   http://dx.doi.0rg/10.1063/l.459015

Stepan. PI: Shockey. RE: Kurz. BA: Kalenze. NS: Cowan. RM: Ziman. II: Harju. IA. (2010). Bakken water
   opportunities assessment: Phase I. (2010-EERC-04-03). Bismarck, ND: North Dakota Industrial
   Commission, http://www.nd.gov/ndic/ogrp/info/g-018-036-fi.pdf

Struchtemeyer, CG: Elshahed, MS. (2012). Bacterial communities associated with hydraulic fracturing fluids in
   thermogenic natural gas wells in North Central Texas, USA. FEMS Microbiol Ecol 81: 13-25.
   http://dx.doi.0rg/10.llll/i.1574-6941.2011.01196.x

Stumm. W: Morgan. II. (1981). Aquatic chemistry: An introduction emphasizing chemical equilibria in natural
   waters (2nd ed.). New York, NY: Wiley.

Sturchio. NC: Banner.  IL: Binz. CM: Heraty. LB: Musgrove. M. (2001). Radium geochemistry of ground waters
   in Paleozoic carbonate aquifers, midcontinent, USA. Appl Geochem 16:109-122.

Swanson, VE. (1955). Uranium in marine black shales of the United States. In Contributions to the geology of
   uranium and thorium by the United States Geological Survey and Atomic Energy Commission for the
   United Nations International Conference on Peaceful Uses of Atomic Energy, Geneva, Switzerland, 1955
   (pp. 451-456). Reston, VA: U.S. Geological Survey, http://pubs.usgs.gov/pp/0300/report.pdf
               This document is a draft for review purposes only and does not constitute Agency policy.

June 2015                                        18                     DRAFT—DO NOT CITE OR QUOTE

-------
Hydraulic Fracturing Drinking Water Assessment                                       All Appendices' References
Tchobanoglous. G: Burton. FL: Stensel. HP. (2013). Wastewater engineering: Treatment and reuse. In th (Ed),
   (9780070418783 ed). Boston, MA: McGraw-Hill.

Thalladi. VR: Nusse. M: Boese. R. (2000). The melting point alternation in alpha.omega-alkanedicarboxylic
   acids. J Am Chem Soc 122: 9227-9236. http://dx.doi.org/10.1021/ja0011459

Tiemann. M: Folger. P: Carter. NT. (2014). Shale energy technology assessment: Current and emerging water
   practices. Washington, DC: Congressional Research Service, http://nationalaglawcenter.org/wp-
   content/uploads//assets/crs/R43635.pdf

Timmis. KN. (2010).  Handbook of hydrocarbon and lipid microbiology. Berlin, Germany: Springer-Verlag.
   http://www.springer.com/life+sciences/microbiology/book/978-3-540-77584-3

Tourtelot. HA. (1979). Black shale - its deposition and diagenesis. Clays and Clay Minerals 27: 313-321.
   http://dx.doi.org/10.1346/CCMN.1979.0270501

Tuttle. MLW: Breit. GN: Goldhaber. MB. (2009). Weathering of the New Albany Shale, Kentucky: II.
   Redistribution of minor and trace elements. Appl Geochem 24:1565-1578.
   http://dx.doi.0rg/10.1016/j.apgeochem.2009.04.034

U.S. Census Bureau. (2013). Metropolitan and micropolitan statistical areas main. Available online at
   http://www.census.gov/population/metro/ (accessed January 12, 2015).

U.S. Department of Transportation. (2012). Large truck and bus crash facts 2012. Washington, D.C.: Federal
   Motor Carrier Safety Administration, U.S. Department of Transportation.
   http://ai.fmcsa.dot.gov/CarrierResearchResults/PDFs/LargeTruckandBusCrashFacts2012.pdf

U.S. EPA (U.S. Environmental Protection Agency). (1996). Proposed guidelines for carcinogen risk assessment
   [EPA Report]. (EPA/600/P-92/003C). Washington, DC: U.S. Environmental Protection Agency, Risk
   Assessment Forum.

U.S. EPA (U.S. Environmental Protection Agency). (1999). Guidelines for carcinogen risk assessment [review
   draft] [EPA Report]. (NCEA-F-0644). Washington, DC.
   http://www.epa.gov/raf/publications/pdfs/CANCER GLS.PDF

U.S. EPA (U.S. Environmental Protection Agency). (2004). Evaluation of impacts to underground sources of
   drinking water by hydraulic fracturing of coalbed methane reservoirs. (EPA/816/R-04/003). Washington,
   DC.: U.S. Environmental Protection Agency, Office of Solid Waste.

U.S. EPA (U.S. Environmental Protection Agency). (2005). Membrane filtration guidance manual. (EPA 815-R-
   06-009). Washington, D.C.
   http://www.epa.gov/ogwdw/disinfection/lt2/pdfs/guide It2  membranefiltration final.pdf

U.S. EPA (U.S. Environmental Protection Agency). (2006). National Primary Drinking Water Regulations:
   Stage 2 Disinfectants and Disinfection Byproducts Rule.
   http://water.epa.gov/lawsregs/rulesregs/sdwa/stage2/

U.S. EPA (U.S. Environmental Protection Agency). (2011a). Plan to study the potential impacts of hydraulic
   fracturing on drinking water resources [EPA Report]. (EPA/600/R-11/122). Washington, DC: U.S.
   Environmental Protection Agency, Office of Research and Development.
   http://www2.epa.gov/hfstudy/plan-study-potential-impacts-hydraulic-fracturing-drinking-water-
   resources-epa600r-11122

U.S. EPA (U.S. Environmental Protection Agency). (2011b). Sampling data for flowback and produced water
   provided to EPA by nine oil and gas well operators (non-confidential business information). US
   Environmental Protection Agency.
   http://www.regulations.gov/#!docketDetail:rpp=100:so=DESC:sb=docId:po=0:D=EPA-HQ-ORD-2010-
   0674
               This document is a draft for review purposes only and does not constitute Agency policy.

June 2015                                        19                     DRAFT—DO NOT CITE OR QUOTE

-------
Hydraulic Fracturing Drinking Water Assessment                                      All Appendices' References
U.S. EPA (U.S. Environmental Protection Agency). (2011c). Terminology services (TS): Vocabulary catalog -
   IRIS glossary. Available online at
   http://ofmpub.epa.gov/sor internet/registry/termreg/searchandretrieve/glossariesandkeywordlists/se
   arch.do?details=&glossaryName=IRIS%20Glossary (accessed May 21, 2015).

U.S. EPA (U.S. Environmental Protection Agency). (2012a). Estimation Programs Interface Suite for Microsoft
   Windows (EPI Suite) [Computer Program]. Washington DC: US Environmental Protection Agency.
   Retrieved from http://www.epa.gov/oppt/exposure/pubs/episuitedl.htm

U.S. EPA. General definitions. 40 CFR § 437.2 (2012b). http://www.gpo.gov/fdsys/pkg/CFR-2012-title40-
   vol31/pdf/CFR-2012-title40-vol31-sec437-2.pdf

U.S. EPA (U.S. Environmental Protection Agency). (2012c). Study of the potential impacts of hydraulic
   fracturing on drinking water resources: Progress report. (EPA/601/R-12/011). Washington, DC: U.S.
   Environmental Protection Agency, Office of Research and Development.
   http://nepis.epa.gov/exe/ZyPURL.cgi?Dockey=P100FH8M.txt

U.S. EPA (U.S. Environmental Protection Agency). (2013a). Data received from oil and gas exploration and
   production companies, including hydraulic fracturing service companies 2011 to 2013. Non-confidential
   business information source documents are located in Federal Docket ID: EPA-HQ-ORD2010-0674.
   Available at http://www.regulations.gov.

U.S. EPA (U.S. Environmental Protection Agency). (2013b). Distributed structure-searchable toxicity
   (DSSTOX) database network. Available online at http://www.epa.gov/ncct/dsstox/index.html

U.S. EPA (U.S. Environmental Protection Agency). (2013c). Drinking water and ground water statistics, fiscal
   year 2011. Washington, DC: U.S. Environmental Protection Agency, Office of Water.
   http://water.epa.gov/scitech/datait/databases/drink/sdwisfed/upload/epa816rl3003.pdf

U.S. EPA (U.S. Environmental Protection Agency). (2013d). Terminology services (TS): Terms and acronyms.
   Available online at
   http://iaspub.epa.gov/sor internet/registry/termreg/searchandretrieve/termsandacronyms/search.do

U.S. EPA. Announcement of preliminary regulatory determinations for contaminants on the third drinking
   water contaminant candidate list. EPA-HO-OW-2012-0155 62715 -62750 (62736 pages) (2014a).
   https://www.federalregister.gov/articles/2014/10/20/2014-24582/announcement-of-preliminary-
   regulatory-determinations-for-contaminants-on-the-third-drinking-water#page-62715

U.S. EPA (U.S. Environmental Protection Agency). (2014b). Drinking water contaminants. Available online at
   http://water.epa.gov/drink/contaminants/

U.S. EPA (U.S. Environmental Protection Agency). (2014c). Substance registry services. Available online at
   http://ofmpub.epa.gov/sor internet/registry/substreg/home/overview/home.do

U.S. EPA (U.S. Environmental Protection Agency). (2015a). Analysis of hydraulic fracturing fluid data from the
   FracFocus chemical disclosure registry 1.0 [EPA Report]. (EPA/601/R-14/003). Washington, D.C.: Office
   of Research and Development, U.S. Environmental Protection Agency.
   http://www2.epa.gov/hfstudy/analysis-hydraulic-fracturing-fluid-data-fracfocus-chemical-disclosure-
   registry-1-pdf

U.S. EPA (U.S. Environmental Protection Agency). (2015b). Analysis of hydraulic fracturing fluid data from the
   FracFocus chemical disclosure registry 1.0: Data management and quality assessment report [EPA
   Report]. (EPA/601/R-14/006). Washington, D.C.: U.S. Environmental Protection Agency, Office of
   Research and Development, http://www2.epa.gov/sites/production/files/2015-
   03/documents/fracfocus data management report final 032015 508.pdf
               This document is a draft for review purposes only and does not constitute Agency policy.

June 2015                                        20                    DRAFT—DO NOT CITE OR QUOTE

-------
Hydraulic Fracturing Drinking Water Assessment                                      All Appendices' References
U.S. EPA (U.S. Environmental Protection Agency). (2015c). Analysis of hydraulic fracturing fluid data from the
   FracFocus chemical disclosure registry 1.0: Project database [EPA Report]. (EPA/601/R-14/003).
   Washington, D.C.: U.S. Environmental Protection Agency, Office of Research and Development.
   http://www2.epa.gov/hfstudy/epa-project-database-developed-fracfocus-l-disclosures

U.S. EPA (U.S. Environmental Protection Agency). (2015d). Case study analysis of the impacts of water
   acquisition for hydraulic fracturing on local water availability [EPA Report]. (EPA/600/R-14/179).
   Washington, D.C.

U.S. EPA (U.S. Environmental Protection Agency). (2015e). Human health benchmarks for pesticides.
   Available online at http://iaspub.epa.gov/apex/pesticides/f?p=HHBP:HOME

U.S. EPA (U.S. Environmental Protection Agency). (2015f). Review of well operator files for hydraulically
   fractured oil and gas production wells: Well design and construction [EPA Report]. (EPA/601/R-14/002).
   Washington, D.C.: Office of Research and Development, U.S. Environmental Protection Agency.

U.S. EPA (U.S. Environmental Protection Agency). (2015g). Technical development document for proposed
   effluent limitation guidelines and standards for oil and gas extraction. (EPA-821-R-15-003). Washington,
   D.C. http://water.epa.gov/scitech/wastetech/guide/oilandgas/unconv.cfm

U.S. GAP (U.S. Government Accountability Office). (2014). Freshwater: Supply concerns continue, and
   uncertainties complicate planning. Report to Congressional requesters. (GAO-14-430). Washington, DC:
   U.S. Government Accountability Office (GAO). http://www.gao.gov/assets/670/663343.pdf

USGS (U.S. Geological Survey). (1961). Geology and geochemistry of uranium in marine black shales: a review.
   (U.S. Geological Survey Professional Paper 356-C). Reston, VA. http://pubs.usgs.gov/pp/0356c/report.pdf

USGS (U.S. Geological Survey). (1997). Radioactive  elements in coal and fly ash: Abundance, forms, and
   environmental significance [Fact Sheet]. (U.S. Geological Survey Fact Sheet FS-163-97).
   http://pubs.usgs.gov/fs/1997/fsl63-97/FS-163-97.pdf

USGS (U.S. Geological Survey). (2002). Water quality and environmental isotopic analyses of ground-water
   samples collected from the Wasatch and Fort Union formations in areas of coalbed methane
   developmentimplications to recharge and groundwater flow, eastern Powder river basin, Wyoming.
   (Report 02-4045). Reston, VA. http://pubs.usgs.gov/wri/wri024045/

USGS (U.S. Geological Survey). (2014). Withdrawal and consumption of water by thermoelectric power plants
   in the United States, 2010. (Scientific Investigations Report 20145184). Reston, VA.
   http://dx.doi.org/10.3133/sir20145184

USGS (U.S. Geological Survey). (2015). Water use in the United States. Available online at
   http://water.usgs.gov/watuse/

Van Voast. WA. (2003). Geochemical signature of formation waters associated with coalbed methane. AAPG
   Bulletin 87: 667-676.

Vanengelen. MR: Peyton. BM: Mormile. MR: Pinkart. HC. (2008). Fe(III), Cr(VI), and Fe(III) mediated Cr(VI)
   reduction in alkaline media using a Halomonas isolate from Soap Lake, Washington. Biodegradation 19:
   841-850. http://dx.doi.org/10.1007/sl0532-008-9187-l

Veil. I A. (2010). Water management technologies used by Marcellus shale gas producers - Final Report. (DOE
   Award No.: FWP 49462). Veil, JA.
   http://fracfocus.org/sites/default/files/publications/water management in the marcellus.pdf

Vengosh. A: lackson. RB: Warner. N: Darrah. TH: Kondash. A. (2014). A critical review of the risks to water
   resources from unconventional shale gas development and hydraulic fracturing in the United States.
   Environ Sci Technol 48: 36-52. http://dx.doi.org/10.1021/es405118v
               This document is a draft for review purposes only and does not constitute Agency policy.

June 2015                                        21                    DRAFT—DO NOT CITE OR QUOTE

-------
Hydraulic Fracturing Drinking Water Assessment                                      All Appendices' References
Verdegem. MCI: Bosma. RH. (2009). Water withdrawal for brackish and inland aquaculture, and options to
   produce more fish in ponds with present water use. Water Policy 11: 52-68.
   http://dx.doi.org/10.2166/wp.2009.003

Vidic. RD: Brantley. SL: Vandenbossche. IM: Yoxtheimer. D: Abad. ID. (2013). Impact of shale gas development
   on regional water quality [Review]. Science 340:1235009. http://dx.doi.org/10.1126/science.1235009

Vijaya Kumar. R: Anand Rao. M: Venkateshwara Rao. M: Ravi Kumar. YVL: Prasad. DHL. (1996). Bubble
   temperature measurements on 2-propyn-l-ol with 1,2-dichloroethane, 1,1,1-trichloroethane, and 1,1,2,2-
   tetrachloroethane. Journal of Chemical and Engineering Data 41:1020-1023.
   http://dx.doi.org/10.1021/ie9600156

Vine. ID. (1956). Uranium-bearing coal in the United States. In Contributions to the geology of uranium and
   thorium by the United States Geological Survey and Atomic Energy Commission for the United Nations
   International Conference on Peaceful Uses of Atomic Energy, Geneva, Switzerland, 1955. Reston, VA: U.S.
   Geological Survey, http://pubs.er.usgs.gov/publication/pp300

Vine. ID: Tourtelot. EB. (1970). Geochemistry of black shale deposits; A summary report. Econ Geol 65: 253-
   272. http://dx.doi.org/10.2113/gsecongeo.65.3.253

Vinson. PS: Vengosh. A: Hirschfeld. D: Dwyer. GS. (2009). Relationships between radium and radon
   occurrence and hydrochemistry in fresh groundwater from fractured crystalline rocks, North Carolina
   (USA). Chem Geol 260:159-171. http://dx.doi.Org/10.1016/i.chemgeo.2008.10.022

Vulgamore. TB: Clawson. TD: Pope. CD: Wolhart. SL: Mayerhofer. Ml: Machovoe. SR: Waltman. CK. (2007).
   Applying hydraulic fracture diagnostics  to optimize stimulations in the Woodford Shale. Richardson, TX:
   Society of Petroleum Engineers. http://dx.doi.org/10.2118/110029-MS

Walsh,  IM. (2013). Water management for hydraulic fracturing in unconventional resourcesPart 1. Oil and
   Gas Facilities 2.

Ward. CR. (2002). Analysis and significance of mineral matter in coal seams. Int J  Coal Geol 50:135-168.
   http://dx.doi.org/10.1016/S0166-5162(02)00117-9

Warner. NR: Christie. CA: lackson. RB: Vengosh. A. (2013). Impacts of shale gas wastewater disposal on water
   quality in western Pennsylvania. Environ Sci Technol 47:11849-11857.
   http://dx.doi.org/10.1021/es402165b

Warner, NR: lackson, RB: Darrah, TH: Osborn, SG: Down, A: Zhao, K: White, A: Vengosh, A. (2012). Reply to
   Engelder: Potential for fluid migration from the Marcellus Formation remains possible. PNAS 109: E3626-
   E3626. http://dx.doi.org/10.1073/pnas.1217974110

Wasylishen. R: Fulton. S. (2012). Reuse of flowback & produced water for hydraulic fracturing in tight oil.
   Calgary, Alberta, Canada: The Petroleum Technology Alliance Canada (PTAC).
   http://www.ptac.org/projects/151

Watson. TL: Bachu. S.  (2009). Evaluation of the potential for gas and C02  leakage along wellbores. S P E
   Drilling & Completion 24:115-126. http://dx.doi.org/10.2118/106817-PA

Weaver. TR: Frape. SK: Cherry. IA. (1995). Recent cross-formational fluid flow and mixing in the shallow
   Michigan Basin. Geol Soc Am Bulletin 107: 697-707. http://dx.doi.org/10.1130/0016-
   7606(1995)107<0697:RCFFFA>2.3.CO:2

Webster. IT: Hancock. Gl: Murray. AS. (1995). Modelling the effect of salinity on radium desorption from
   sediments. Geochim Cosmo Act 59: 2469-2476. http://dx.doi.org/10.1016/0016-7037(95)00141-7

White.  Gl. (1992). Naturally occurring radioactive materials (NORM) in oil and gas industry equipment and
   wastes: A literature review. (DOE/ID/01570-T158). Bartlesville, OK: U.S. Department of Energy.

WHO (World Health Organization). (2015). Concise international chemical assessment documents. Available
   online at http://www.who.int/ipcs/publications/cicad/en/
               This document is a draft for review purposes only and does not constitute Agency policy.

June 2015                                       22                     DRAFT—DO NOT CITE OR QUOTE

-------
Hydraulic Fracturing Drinking Water Assessment                                       All Appendices' References
Wignall. PG: Myers. Kl. (1988). Interpreting benthic oxygen levels in mudrocks: A new approach. Geology 16:
   452-455. http://dx.doi.org/10.1130/0091-7613(1988)016<0452:IBOLIM>2.3.CO:2

Wilt. IW. (1956). Notes - the halodecarboxylation of cyanoacetic acid. J Org Chem 21: 920-921.
   http://dx.doi.org/10.1021/jo01114a607

Wojtanowicz. AK. (2008). Environmental control of well integrity. In ST Orszulik (Ed.), Environmental
   technology in the oil industry (pp. 53-75). Houten, Netherlands: Springer Netherlands.

Woolard. CR: Irvine. RL. (1995). Treatment of of hypersaline wastewater in the sequencing batch reactor.
   Water Res 29:1159-1168.

Wuchter. C: Banning. E: Mincer. Tl: Drenzek. Nl: Coolen. Ml. (2013). Microbial diversity and methanogenic
   activity of Antrim Shale formation waters from recently fractured wells. FMICB 4:1-14.
   http://dx.doi.org/10.3389/fmicb.2013.00367

WVWRI (West Virginia Water Research Institute, West Virginia University). (2012). Zero discharge water
   management for horizontal shale gas well development. (DE-FE0001466). https://www.netl.doe.gov/File
   Library/Research/Oil-Gas/Natural Gas/shale gas/fe0001466-final-report.pdf

Xiao. LN: Xu. IN: Hu. YY: Wang. LM: Wang. Y: Ding. H: Cui. XB: Xu. 10. (2013). Synthesis and characterizations
   of the first [V16039C1]6- (V16039) polyanion. Dalton Transactions (Online) 42: 5247-5251.
   http://dx.doi.org/10.1039/c3dt33081h

Yakimov. MM: Denaro. R: Genovese. M: Cappello. S: D'Auria. G: Chernikova. TN: Timmis. KN: Golyshin. PN:
   Giluliano. L. (2005). Natural microbial diversity in superficial sediments of Milazzo Harbor (Sicily) and
   community successions during microcosm enrichment with various hydrocarbons. Environ Microbiol 7:
   1426-1441. http://dx.doi.0rg/10.llll/i.1462-5822.2005.00829.x

Yang. IS: Lee. IY: Baek. K: Kwon. TS: Choi. I. (2009). Extraction behavior of As, Pb, and Zn from mine tailings
   with acid and base solutions. J Hazard Mater 171:1-3. http://dx.doi.Org/10.1016/j.jhazmat.2009.06.021

Yang. X: Shang. C. (2004). Chlorination byproduct formation in the presence of humic acid, model nitrogenous
   organic compounds, ammonia, and bromide. Environ Sci Technol 38: 4995-5001.
   http://dx.doi.org/10.1021/es049580g

Yoshizawa. S: Wada. M: Kita-Tsukamoto.  K: Ikemoto. E: Yokota. A: Kogure. K. (2009). Vibrio azureus sp. nov., a
   luminous marine bacterium isolated from seawater. Int J Syst Evol Microbiol 59:1645-1649.
   http://dx.doi.0rg/10.1099/ijs.0.004283-0

Yoshizawa. S: Wada. M: Yokota. A: Kogure. K. (2010). Vibrio sagamiensis sp. nov., luminous marine bacteria
   isolated from sea water. J Gen Appl Microbiol 56: 499-507.

Younos. T: Tulou. KE. (2005). Overview of desalination techniques. Journal of Contemporary Water Research
   & Education 132: 3-10. http://dx.doi.Org/10.llll/i.1936-704X.2005.mpl32001002.x

Zapecza. OS: Szabo. Z. (1988). Natural radioactivity in ground watera review. In National Water Summary
   1986Hydrologic Events and Ground-Water Quality, Water-Supply Paper 2325. Reston, VA: U.S. Geological
   Survey, http://pubs.er.usgs.gov/publication/wsp2325

Zeikus. IG: Hegge. PW: Thompson. TE: Phelps. Tl: Langworthy. TA. (1983). Isolation and description of
   Haloanaerobium praevalens gen. nov. and sp. nov., an obligatory anaerobic halophile common to Great Salt
   Lake sediments. Curr Microbiol 9: 225-233. http://dx.doi.org/10.1007/BF01567586

Zhang. L: Guo. Y: Xiao. I: Gong. X: Fang. W. (2011). Density, refractive index, viscosity, and surface tension of
   binary mixtures of exo-tetrahydrodicyclopentadiene with some n-alkanes from (293.15 to 313.15) K.
   Journal of Chemical and Engineering Data 56: 4268-4273. http://dx.doi.org/10.1021/je200757a

Zhang. T: Gregory. K: Hammack. RW: Vidic. RD. (2014). Co-precipitation of radium with barium and strontium
   sulfate and its impact on the fate of radium during treatment of produced water from unconventional gas
   extraction. Environ Sci Technol 48: 4596-4603. http://dx.doi.org/10.1021/es405168b
               This document is a draft for review purposes only and does not constitute Agency policy.

June 2015                                        23                     DRAFT—DO NOT CITE OR QUOTE

-------
Hydraulic Fracturing Drinking Water Assessment                                       All Appendices' References
                             [2013). Vapor-liquid equilibrium for ternary and binary mixtures of 2-
   isopropoxypropane, 2-propanol, and n,n-dimethylacetamide at 101.3 kPa. Journal of Chemical and
   Engineering Data 58: 357-363.
               This document is a draft for review purposes only and does not constitute Agency policy.

June 2015                                         24                      DRAFT—DO NOT CITE OR QUOTE

-------
United States
Environmental Protection
Agency
Office of Research and Development
Washington, DC 20460

Official Business
Penalty for Private Use
$300

-------