United States Office of Water EPA-821 -R-15-006
Environmental Protection Washington, DC 20460 September 2015
Agency
&EPA Environmental Assessment
for the Effluent Limitations
Guidelines and Standards
for the Steam Electric Power
Generating Point Source
Category
-------
vvEPA
United States
Environmental Protection
Agency
Environmental Assessment for the Effluent
Limitations Guidelines and Standards for the
Steam Electric Power Generating Point
Source Category
EPA-821-R-15-006
September 2015
U.S. Environmental Protection Agency
Office of Water (43 03 T)
Engineering and Analysis Division
1200 Pennsylvania Avenue, NW
Washington, DC 20460
-------
Acknowledgements and Disclaimer
This report was prepared by the U.S. Environmental Protection Agency. Neither the United
States Government nor any of its employees, contractors, subcontractors, or their employees
make any warrant, expressed or implied, or assume any legal liability or responsibility for any
third party's use of or the results of such use of any information, apparatus, product, or process
discussed in this report, or represents that its use by such party would not infringe on privately
owned rights.
-------
Table of Contents
TABLE OF CONTENTS
Page
ACRONYMS vm
GLOSSARY xi
SECTION 1 INTRODUCTION 1-1
SECTION 2 BACKGROUND AND SCOPE 2-1
SECTION 3 ENVIRONMENTAL AND HUMAN HEALTH CONCERNS 3-1
3.1 Types of Pollutants Discharged in Steam Electric Power Plant Wastewater 3-2
3.1.1 Metals and Toxic Bioaccumulative Pollutants 3-2
3.1.2 Nutrients 3-9
3.1.3 TDS 3-9
3.2 Loadings Associated with Steam Electric Power Plant Wastewater 3-12
3.2.1 Annual Baseline Pollutant Loadings 3-13
3.2.2 Comparison of Steam Electric Power Plant Loadings to Other
Industries 3-15
3.2.3 Comparison of Steam Electric Power Plant Loadings to Publicly
Owned Treatment Works 3-16
3.3 Environmental Impacts from Steam Electric Power Plant Wastewater 3-20
3.3.1 Ecological Impacts 3-20
3.3.2 Human Health Effects 3-27
3.3.3 Damage Cases and Other Documented Surface Water Impacts 3-28
3.3.4 Damage Cases and Other Documented Ground Water Impacts 3-35
3.3.5 Potential for Impacts to Occur in Other Locations 3-37
3.4 Discharge to Sensitive Environments 3-38
3.4.1 Pollutant Loadings to the Great Lakes Watershed 3-38
3.4.2 Pollutant Loadings to the Chesapeake Bay Watershed 3-40
3.4.3 Proximity to Impaired Waters 3-42
3.4.4 Proximity to Fish Consumption Advisory Waters 3-44
3.4.5 Proximity to Threatened and Endangered Species Habitats 3-45
3.4.6 Proximity to Drinking Water Resources 3-46
3.5 Long Environmental Recovery Times Associated with Pollutants in Steam
Electric Power Plant Wastewater 3-47
SECTION 4 ASSESSMENT OF EXPOSURE PATHWAYS 4-1
4.1 Discharge and Leaching to Surface Waters 4-2
4.1.1 Factors Controlling Environmental Impacts in Surface Waters 4-2
4.1.2 Assessment of the Surface Water Exposure Pathway 4-4
4.2 Leaching to Ground Water 4-7
4.2.1 Factors Controlling Environmental Impacts to Ground Water 4-7
4.2.2 Assessment of the Ground Water Exposure Pathway 4-12
4.3 Combustion Residual Surface Impoundments as Attractive Nuisance 4-12
-------
Table of Contents
TABLE OF CONTENTS (Continued)
Page
SECTION 5 SURFACE WATER MODELING 5-1
5.1 Immediate Receiving Water (IRW) Model 5-1
5.1.1 Water Quality Module 5-3
5.1.2 Wildlife Module 5-8
5.1.3 Human Health Module 5-10
5.2 Ecological Risk Modeling 5-12
SECTION 6 CURRENT IMPACTS FROM STEAM ELECTRIC POWER GENERATING
INDUSTRY 6-1
6.1 Water Quality Impacts 6-1
6.2 Wildlife Impacts 6-3
6.2.1 Impacts to Wildlife Indicator Species 6-3
6.2.2 Impacts to Fish and Waterfowl due to Dietary Selenium Exposure 6-4
6.2.3 Impacts to Benthic Organisms 6-6
6.3 Human Health Impacts 6-7
6.3.1 National-Scale Cohort Analysis 6-8
6.3.2 Environmental Justice Analysis 6-12
SECTION 7 ENVIRONMENTAL IMPROVEMENTS UNDER THE FINAL RULE 7-1
7.1 Pollutant Removals Under the Regulatory Options 7-5
7.2 Key Environmental Improvements 7-9
7.2.1 Improvements in Water Quality Under the Final Rule 7-9
7.2.2 Reduced Threat to Wildlife Under the Final Rule 7-13
7.2.3 Reduced Human Health Cancer Risk Under the Final Rule 7-15
.2.4 Reduced Threat of Non-Cancer Human Health Effects Under the
Final Rule 7-15
7.2.5 Reduced Human Health Risk for Environmental Justice Analysis 7-15
7.3 Pollutant-Specific Improvements 7-16
7.3.1 Arsenic 7-16
7.3.2 Mercury 7-19
7.3.3 Selenium 7-21
7.3.4 Cadmium 7-25
7.3.5 Thallium 7-25
7.4 Improvements to Sensitive Environments 7-28
7.4.1 Impaired Waters 7-28
7.4.2 Threatened and Endangered Species 7-31
7.4.3 Fish Advisory Waters 7-31
7.5 Improvements to Watersheds 7-31
7.6 Environmental and Human Health Improvements in Downstream Surface
Water 7-34
7.7 Attractive Nuisances 7-37
7.8 Other Secondary Improvements 7-37
7.9 Unresolved Drinking Water Impacts Due to Bromide Discharges 7-38
SECTION 8 CASE STUDY MODELING 8-1
8.1 Case Study Modeling Methodology 8-2
-------
Table of Contents
TABLE OF CONTENTS (Continued)
Page
8.1.1 Selection of Case Study Locations for Modeling 8-2
8.1.2 Scope and Technical Approach for Case Study Modeling 8-7
8.1.3 Development and Execution of WASP Models 8-11
8.1.4 Use of WASP Water Quality Model Outputs 8-13
8.1.5 Limitations of Case Study Modeling 8-14
8.2 Quantified Environmental Impacts and Improvements from Case Study
Modeling 8-14
8.2.1 Black Creek Case Study 8-15
8.2.2 Etowah River Case Study 8-24
8.2.3 Lick Creek & White River Case Study 8-31
8.2.4 Ohio River Case Study 8-41
8.2.5 Mississippi River Case Study 8-47
8.2.6 Lake Sinclair Case Study 8-52
8.3 Comparison of Case Study and IRW Modeling Results 8-58
SECTION 9 CONCLUSIONS 9-1
SECTION 10 REFERENCES 10-1
Appendix A: Literature Review Methodology and Results
Appendix B: Proximity Analyses Supporting Tables
Appendix C: Water Quality Module Methodology
Appendix D: Wildlife Module Methodology
Appendix E: Human Health Module Methodology
Appendix F: Overview of Ecological Risk Modeling Setup and Outputs
Appendix G: Overview of Case Study Modeling Setup and Outputs
Appendix H: Additional Model Results
Appendix I: Analyses for Alternate Scenario with Clean Power Plan
Appendix J: EA Loadings and TDD Loadings: Sensitivity Analysis
in
-------
List of Tables
LIST OF TABLES
Page
Table 2-1. Steam Electric Power Plant Wastestreams Evaluated in the EA 2-2
Table 2-2. Number of Plants Evaluated in the EA 2-4
Table 3-1. Key Metals and Toxic Bioaccumulative Pollutants Found In Steam Electric
Power Plant Wastewater 3-3
Table 3-2. Annual Baseline Pollutant Discharges from Steam Electric Power Plants
(Evaluated Wastestreams) 3-14
Table 3-3. Pollutant Loadings for the Final 2010 Effluent Guidelines Planning Process:
Top 10 Point Source Categories 3-15
Table 3-4. Comparison of Average Pollutant Loadings in the Evaluated Wastestreams to
an Average POTW 3-18
Table 3-5. Estimated Number of POTW Equivalents for Total Pollutant Loadings from
the Evaluated Wastestreams 3-19
Table 3-6. Summary of Studies Evaluating Lethal Effects of Pollutants in Steam Electric
Power Plant Wastewater 3-25
Table 3-7. Median Lethal Concentrations (LCso) for Pollutants in Steam Electric Power
Plant Wastewater 3-26
Table 3-8. Summary of Select Sites with Documented Surface Water Impacts from Steam
Electric Power Plant Wastewater 3-30
Table 3-9. Number and Percentage of Immediate Receiving Waters Identified as
Sensitive Environments 3-38
Table 3-10. Pollutant Loadings to the Great Lakes Watershed from the Evaluated
Wastestreams 3-40
Table 3-11. Pollutant Loadings to the Chesapeake Bay Watershed from the Evaluated
Wastestreams 3-41
Table 3-12. Number and Percentage of Immediate Receiving Waters Classified as
Impaired for a Pollutant Associated with the Evaluated Wastestreams 3-42
Table 3-13. Comparison of Number and Percentage of Steam Electric Power Plants
Located within 5 Miles of a Drinking Water Resource 3-47
Table 4-1. Steam Electric Power Plant Wastewater Environmental Pathways and Routes
of Exposure Evaluated in the EA 4-2
Table 4-2. Receiving Water Types for Steam Electric Power Plants Evaluated in the EA 4-3
Table 4-3. Exceedances of MCLs in Leachate Under Acidic, Neutral, and Basic
Conditions 4-9
Table 4-4. Range of Fly Ash and FGD Gypsum Total Content and Combustion Residual
Leaching Test Results (Initial Screening Concentrations) for Trace Metals 4-11
Table 5-1. Pollutants Considered for Analysis in the Immediate Receiving Water Model 5-5
iv
-------
List of Tables
List of Tables {Continued)
Page
Table 6-1. Number and Percentage of Immediate Receiving Waters with Estimated
Water Concentrations that Exceed the Water Quality Criteria at Baseline 6-2
Table 6-2. Number and Percentage of Immediate Receiving Waters That Exceed Wildlife
Fish Consumption NEHCs for Minks and Eagles (by Waterbody Type) at
Baseline 6-4
Table 6-3. Number and Percentage of Immediate Receiving Waters That Exceed Wildlife
Fish Consumption NEHCs for Minks and Eagles (by Pollutant) at Baseline 6-5
Table 6-4. Number and Percentage of Immediate Receiving Waters with Sediment
Pollutant Concentrations Exceeding CSCLs for Sediment Biota at Baseline 6-7
Table 6-5. Number and Percentage of Immediate Receiving Waters That Exceed Human
Health Evaluation Criteria (Lifetime Excess Cancer Risk) for Inorganic Arsenic at
Baseline 6-9
Table 6-6. Number and Percentage of Immediate Receiving Waters That Exceed Non-
Cancer Oral Reference Dose Values at Baseline 6-10
Table 6-7. Number and Percentage of Immediate Receiving Waters That Exceed Non-
Cancer Oral Reference Dose Values at Baseline by Pollutant 6-11
Table 6-8. Comparison of T4 Fish Tissue Concentrations at Baseline to Fish Advisory
Screening Values 6-12
Table 6-9. Number and Percentage of Immediate Receiving Waters That Exceed Human
Health Evaluation Criteria (Lifetime Excess Cancer Risk) for Inorganic Arsenic at
Baseline, by Race or Hispanic Origin 6-13
Table 6-10. Number and Percentage of Immediate Receiving Waters That Exceed Non-
Cancer Oral Reference Dose Values at Baseline, by Race or Hispanic Origin 6-14
Table 7-1. Regulatory Options for the Wastestreams Evaluated in the EA 7-2
Table 7-2. Description of Environmental Improvements Associated with the Final Rule 7-3
Table 7-3. Steam Electric Power Generating Industry Pollutant Removals for Metals,
Bioaccumulative Pollutants, Nutrients, Chlorides, and TDS Under Regulatory
Options 7-7
Table 7-4. Steam Electric Power Generating Industry TWPE Removals for Metals,
Bioaccumulative Pollutants, Nutrients, Chlorides, and TDS Under Regulatory
Options 7-8
Table 7-5. Key Environmental Improvements Under the Regulatory Options 7-11
Table 7-6. Number of Immediate Receiving Waters with Sediment Pollutant
Concentrations Exceeding CSCLs for Sediment Biota Under the Regulatory
Options 7-14
Table 7-7. Key Environmental Improvements for Arsenic Under the Regulatory Options 7-17
Table 7-8. Key Environmental Improvements for Mercury Under the Regulatory Options 7-20
-------
List of Tables
List of Tables {Continued)
Page
Table 7-9. Key Environmental Improvements for Selenium Under the Regulatory
Options 7-23
Table 7-10. Key Environmental Improvements for Cadmium Under the Regulatory
Options 7-26
Table 7-11. Key Environmental Improvements for Thallium Under the Regulatory
Options 7-27
Table 7-12. Pollutant Removals to Impaired Waters by Impairment Type 7-29
Table 7-13. Pollutant Removals to the Great Lakes Watershed Under the Regulatory
Options 7-33
Table 7-14. Key Environmental Improvements for Downstream Waters Under the
Regulatory Options 7-35
Table 8-1. Locations Selected for Case Study Modeling 8-4
Table 8-2. Summary of Morrow Generating Site Operations 8-15
Table 8-3. Summary of Plant Bowen Operations 8-25
Table 8-4. Summary of Petersburg Generating Station Operations 8-32
Table 8-5. Summary of Bruce Mansfield Operations 8-42
Table 8-6. Summary of W.H. Sammis Operations 8-42
Table 8-7. Summary of Rush Island Operations 8-48
Table 8-8. Summary of Plant Harllee Branch Operations 8-53
VI
-------
List of Figures
List of Figures
Page
Figure 2-1. Locations and Counts of Immediate Receiving Waters in EA Scope and
Modeling Analyses 2-5
Figure 3-1. Location of Plants that Directly Discharge the Evaluated Wastestreams to a
Surface Water Impaired due to Mercury 3-43
Figure 3-2. Location of Plants that Directly Discharge the Evaluated Wastestreams to a
Surface Water Impaired due to Metals, Other than Mercury 3-43
Figure 3-3. Location of Plants that Directly Discharge the Evaluated Wastestreams to a
Surface Water Impaired due to Nutrients 3-44
Figure 3-4. Location of Plants that Directly Discharge to a Surface Water with a Fish
Consumption Advisory 3-45
Figure 5-1. Overview of IRW Model 5-3
Figure 5-2. Water Quality Module: Pollutant Fate in the Waterbody 5-7
Figure 5-3. Flowchart of Selenium Ecological Risk Model 5-15
Figure 8-1. Overview of Case Study Modeling Locations 8-3
Figure 8-2. Black Creek WASP Modeling Area 8-16
Figure 8-3. Etowah River WASP Modeling Area 8-26
Figure 8-4. Lick Creek and White River WASP Modeling Area 8-33
Figure 8-5. Ohio River WASP Modeling Area 8-43
Figure 8-6. Mississippi River WASP Modeling Area 8-49
Figure 8-7. Lake Sinclair WASP and EDFC Modeling Area 8-54
Vll
-------
Acronyms
ACRONYMS
ASTM American Society for Testing and Materials
ATSDR Agency for Toxic Substances and Disease Registry
BAF Bioaccumulation factor
BASINS Better Assessment Science Integrating Point and Nonpoint Sources
BAT Best Available Technology Economically Achievable
BCF Bioconcentration factor
BPT Best Practicable Control Technology Currently Available
CBI Confidential business information
CCR Coal combustion residuals
CFR Code of Federal Regulations
CSCL Chemical stressor concentration limit
CSF Cancer slope factor
CWA Clean Water Act
DBF Disinfection by-products
DCN Document control number
DMR Discharge monitoring report
DOE Department of Energy
EA Environmental assessment
EF Enrichment factors
EFDC Environmental Fluid Dynamics Code
ELGs Effluent Limitations Guidelines and Standards
EP Extraction procedure
EPA U.S. Environmental Protection Agency
ER Exposure-response
ESA Endangered Species Act
FGD Flue gas desulfurization
FGMC Flue gas mercury control
FR Federal Register
FWS U.S. Fish and Wildlife Service
IRIS Integrated Risk Information System
IRW Immediate receiving water
Kdsw Suspended sediment-surface water partition coefficient
LADD Lifetime average daily dose
Ibs/yr Pounds per year
LCso Median lethal concentration
LECR Lifetime excess cancer risk
Vlll
-------
Acronyms
MCL Maximum contaminant level
MRL Minimal risk level
MGD Million gallons per day
mg/day Milligrams per day
mg/kg Milligrams per kilogram
mg/L Milligrams per liter
MW Megawatt
MWh Megawatt-hour
NEHC No effect hazard concentration
NHDPlus National Hydrography Dataset Plus
NOAA National Oceanic and Atmospheric Administration
NOAEL No-observed-adverse-effect level
NPDES National Pollutant Discharge Elimination System
NRWQC National Recommended Water Quality Criteria
NSPS New Source Performance Standards
NWIS National Water Information System
ORCR Office of Resource Conservation and Recovery
OSWER Office of Solid Waste and Emergency Response
PCB Polychlorinated biphenyls
POC Pollutant of concern
POTW Publicly owned treatment works
ppm Parts per million
PSES Pretreatment Standards for Existing Sources
PSNS Pretreatment Standards for New Sources
RCRA Resource Conservation and Recovery Act
RfD Reference dose
RIA Regulatory impact analysis
RSEI Risk-Screening Environmental Indicators
SDWA Safe Drinking Water Act
SQuiRT Screening Quick Reference Table
STORE! EPA's STOrage and RETrieval Data Warehouse
T3 Trophic level 3
T4 Trophic level 4
TC Toxicity characteristic
TCLP Toxicity characteristic leaching procedure
TDD Technical Development Document
TDS Total dissolved solids
TEL Threshold effects level
IX
-------
Acronyms
TMDL Total maximum daily load
TOC Total organic carbon
TRI Toxics Release Inventory
TSS Total suspended solids
TTF Trophic transfer factor
TTHM Total trihalomethanes
TWF Toxic weighting factor
TWPE Toxic weighted pound equivalent
ug/g Micrograms per gram
ug/L Micrograms per liter
USGS United States Geological Survey
WASP Water Quality Analysis Simulation Program
WHO World Health Organization
WMA Wildlife Management Area
WQI Water quality index
-------
Glossary
GLOSSARY
Acute - having a sudden onset or lasting a short time. An acute stimulus is severe enough to
induce a response rapidly. The word acute can be used to define either the exposure or the
response to an exposure (effect). The duration of an acute aquatic toxicity test is generally 4 days
or less and mortality is the response usually measured.
Aquifer - an underground formation or group of formations in rocks and soils containing enough
ground water to supply wells and springs.
Benthic - pertaining to the bottom (bed) of a waterbody.
Bioaccumulation - general term describing a process by which chemicals are taken up by an
organism either directly from exposure to a contaminated medium or by consumption of food
containing the chemical, resulting in a net accumulation of the chemical by an organism due to
uptake from all routes of exposure.
Bioavailability - the ability of a particular contaminant to be assimilated into the tissues of
exposed organisms.
Biomagnification - result of the process of bioaccumulation and biotransfer by which tissue
concentrations of chemicals in organisms at one trophic level exceed tissue concentrations in
organisms at the next lower trophic level in a food chain.
Bottom ash - the ash, including boiler slag, which settles in the furnace or is dislodged from
furnace walls. Economizer ash is included when it is collected with bottom ash.
Chronic - involving a stimulus that is lingering or continues for a long time; often signifies
periods from several weeks to years, depending on the reproductive life cycle of the species. This
term can be used to define either the exposure or the response to an exposure (effect). Chronic
exposures typically induce a biological response of relatively slow progress and long duration.
Combustion residuals - solid wastes associated with combustion-related power plant processes,
including fly and bottom ash from coal-, petroleum coke-, or oil-fired units; flue gas
desulfurization (FGD) solids; flue gas mercury control wastes; and other wastewater treatment
solids associated with steam electric power plant wastewater. In addition to the residuals that are
associated with coal combustion, this also includes residuals associated with the combustion of
other fossil fuels.
Combustion residual leachate - leachate from landfills or surface impoundments containing
combustion residuals. Leachate is composed of liquid, including any suspended or dissolved
constituents in the liquid, that has percolated through waste or other materials emplaced in a
landfill, or that passes through the surface impoundment's containment structure (e.g., bottom,
dikes, berms). Combustion residual leachate includes seepage and/or leakage from a combustion
residual landfill or impoundment unit. Combustion residual leachate includes wastewater from
landfills and surface impoundments located on non-adjoining property when under the
operational control of the permitted facility.
XI
-------
Glossary
Criterion continuous concentration - an estimate of the highest concentration of a material in
surface water to which an aquatic community can be exposed indefinitely (chronic exposure)
without resulting in an unacceptable effect.
Criterion maximum concentration - an estimate of the highest concentration of a material in
surface water to which an aquatic community can be exposed briefly (acute exposure) without
resulting in an unacceptable effect.
Direct discharge - (a) Any addition of any "pollutant" or combination of pollutants to "waters of
the United States" from any "point source," or (b) any addition of any pollutant or combination
of pollutant to waters of the "contiguous zone" or the ocean from any point source other than a
vessel or other floating craft which is being used as a means of transportation. This definition
includes additions of pollutants into waters of the United States from: surface runoff which is
collected or channeled by man; discharges though pipes, sewers, or other conveyances owned by
a State, municipality, or other person which do not lead to a treatment works; and discharges
through pipes, sewers, or other conveyances, leading into privately owned treatment works. This
term does not include an addition of pollutants by any "indirect discharger."
Edema - swelling caused by fluid in body tissues.
Effluent limitation - under Clean Water Act (CWA) section 502(11), any restriction, including
schedules of compliance, established by a state or the Administrator on quantities, rates, and
concentrations of chemical, physical, biological, and other constituents which are discharged
from point sources into navigable waters, the waters of the contiguous zone, or the ocean,
including schedules of compliance.
Evaluated wastestreams - subset of steam electric power plant wastewaters evaluated in the
environmental assessment (EA) and Benefits and Cost Analysis that includes FGD wastewater,
fly ash transport water, bottom ash transport water, and combustion residual leachate collected
from landfills or surface impoundments.
Exposure - the contact or co-occurrence of a stressor with a receptor.
Flue gas desulfurization (FGD) wastewater - wastewater generated specifically from the wet
FGD scrubber system that comes into contact with the flue gas or the FGD solids, including but
not limited to, the blowdown or purge from the FGD scrubber system, overflow or underflow
from the solids separation process, FGD solids wash water, and the filtrate from the solids
dewatering process. Wastewater generated from cleaning the FGD scrubber, cleaning FGD
solids separation equipment, cleaning the FGD solids dewatering equipment, or that is collected
in floor drains in the FGD process area is not considered FGD wastewater.
Flue gas mercury control (FGMC) wastewater - wastewater generated from an air pollution
control system installed or operated for the purpose of removing mercury from flue gas. This
includes fly ash collection systems when the particulate control system follows sorbent injection
or other controls to remove mercury from flue gas. FGD wastewater generated at plants using
oxidizing agents to remove mercury in the FGD system and not in a separate FGMC system is
not included in this definition.
Xll
-------
Glossary
Fly ash - the ash that is carried out of the furnace by a gas stream and collected by a capture
device such as a mechanical precipitator, electrostatic precipitator, and/or fabric filter.
Economizer ash is included in this definition when it is collected with fly ash. Ash is not
included in this definition when it is collected in wet scrubber air pollution control systems
whose primary purpose is particulate removal.
Gasification wastewater - any wastewater generated at an integrated gasification combined cycle
operation from the gasifier or the syngas cleaning, combustion, and cooling processes.
Gasification wastewater includes, but is not limited to the following: sour/grey water; CO2/steam
stripper wastewater; sulfur recovery unit blowdown, and wastewater resulting from slag handling
or fly ash handling, particulate removal, halogen removal, or trace organic removal. Air
separation unit blowdown, noncontact cooling water, and runoff from fuel and/or byproduct piles
are not considered gasification wastewater. Wastewater that is collected intermittently in floor
drains in the gasification process areas from leaks, spills and cleaning occurring during normal
operation of the gasification operation is not considered gasification wastewater.
Ground water - water that is found in the saturated part of the ground underneath the land
surface.
Hematological - pertaining to or emanating from blood cells.
Histopathological - pertaining to tissue changes.
Immediate receiving water - the segment of a receiving water where discharges from a point
source enter the surface water. The segment is defined by the hydrographic dataset supporting
the analysis (e.g., National Hydrography Dataset Plus, Version 1).
Impaired waters - a surface water is classified as a 303(d) impaired water when pollutant
concentrations exceed water quality standards and the surface water can no longer meet its
designated uses (e.g., drinking, recreation, and aquatic habitat).
Indirect discharge - wastewater discharged or otherwise introduced to a publicly owned
treatment works (POTW).
Invertebrates - animals without a backbone or spinal column; macroinvertebrates are
invertebrates that can be seen without a microscope (macro), such as aquatic insects, worms,
clams, snails, and crustaceans.
Landfill - a disposal facility or part of a facility where solid waste, sludges, or other process
residuals are placed in or on any natural or manmade formation in the earth for disposal and
which is not a storage pile, a land treatment facility, a surface impoundment, an underground
injection well, a salt dome or salt bed formation, an underground mine, a cave, or a corrective
action management unit.
Leachate - see combustion residual leachate.
Lentic - pertaining to still or slow-moving water, such as lakes or ponds.
Xlll
-------
Glossary
Lethal - causing death by direct action.
Lotic - pertaining to flowing water, such as streams and rivers.
Median lethal concentration (LCso) - a statistically or graphically estimated concentration that is
expected to be lethal to 50 percent of a group of organisms under specified conditions.
Mortality - death rate or proportion of deaths in a population.
Partition coefficient - the ratio of a pollutant concentration in one medium compared to another
(e.g., dissolved in the water column, sorbed to suspended sediment, and sorbed to benthic
sediment in a receiving water).
Piscivorous - habitually feeds on fish.
Plant-receiving water - the combination of a steam electric power plant and the immediate
receiving water into which evaluated wastestreams are discharged from that plant.
Point source - any discernable, confined, and discrete conveyance, including but not limited to,
any pipe, ditch, channel, tunnel, conduit, well, discrete fissure, container, rolling stock,
concentrated animal feeding operation, or vessel or other floating craft from which pollutants are
or may be discharged. The term does not include agricultural stormwater discharges or return
flows from irrigated agriculture. See CWA section 502(14), 33 U.S.C. 1362(14); 40 CFR §122.2.
Population - an aggregate of individuals of a species within a specified location in space and
time.
Publicly owned treatment works (POTW) - any device or system, owned by a state or
municipality, used in the treatment (including recycling and reclamation) of municipal sewage or
industrial wastes of a liquid nature that is owned by a state or municipality. This includes sewers,
pipes, or other conveyances only if they convey wastewater to a POTW providing treatment. See
CWA section 212, 33 U.S.C. 1292; 40 CFR §§122.2, 403.3.
Receptor - the ecological or human entity exposed to a stressor.
Receiving water - surface waters into which treated waste or untreated waste are discharged,
including those portions of the surface water downstream from the point source.
Sediment - particulate material lying below water.
Sensitivity - in relation to toxic substances, organisms that are more sensitive exhibit adverse
(toxic) effects at lower exposure levels than organisms that are less sensitive.
Steam electric power plant wastewater - wastewaters associated with or resulting from the
combustion process, including ash transport water from coal-, petroleum coke-, or oil-fired units;
air pollution control wastewater (e.g., FGD wastewater, FGMC wastewater, carbon capture
wastewater); and leachate from landfills or surface impoundments containing combustion
residuals.
XIV
-------
Glossary
Stressor - any physical, chemical, or biological entity that can induce an adverse response.
Sublethal - below the concentration that directly causes death. Exposure to sublethal
concentrations of a substance can produce effects on behavior, biochemical, and/or physiological
functions, and the structure of cells and tissues in organisms.
Surface water - all waters of the United States, including rivers, streams, lakes, reservoirs, and
seas.
Teratogenic - able to disturb the growth and development of an embryo or fetus.
Transport water - any wastewater that is used to convey fly ash, bottom ash, or economizer ash
from the ash collection or storage equipment, or boiler, and has direct contact with the ash.
Transport water does not include low volume, short duration discharges of wastewater from
minor leaks (e.g., leaks from valve packing, pipe flanges, or piping) or minor maintenance events
(e.g., replacement of valves or pipe sections).
Trophic level - position of an organism in the food chain.
Toxic pollutants - as identified under the CWA, 65 pollutants and classes of pollutants, of which
126 specific substances have been designated priority toxic pollutants. See Appendix A to 40
CFR§423.
XV
-------
Section 1—Introduction
SECTION 1
INTRODUCTION
The U.S. Environmental Protection Agency (EPA) is promulgating revised effluent
limitations guidelines and standards (ELGs) for the Steam Electric Power Generating Point
Source Category (40 CFR 423). In support of the development of the final rule, EPA conducted
an environmental assessment (EA) to evaluate the environmental impact of pollutant loadings
released under current (i.e., baseline) discharge practices and assess the potential environmental
improvement from pollutant loading removals under the final rule.1
Based on evidence in the literature, documented damage cases, and modeled receiving
water pollutant concentrations, it is clear that current steam electric power plant wastewater
discharge practices impact the water quality in receiving waters, impact the wildlife in the
surrounding environments, and pose a human health threat to nearby communities. Substantial
evidence exists that metals (e.g., arsenic, cadmium, mercury, selenium) from steam electric
power plant wastewater discharges transfer from the aquatic environment to terrestrial food
webs, indicating a potential for broader impacts to ecological systems by altering population
diversity and community dynamics in the areas surrounding steam electric power plants.
Ecosystem recovery from exposure to pollutants in power plant wastewater discharges can be
extremely slow, and even short periods of exposure (e.g., less than a year) can cause observable
ecological impacts that last for years.
Steam electric power plants discharge wastewater, which contains numerous pollutants,2
into waterbodies used for recreation and can present a threat to human health. Due to steam
electric power plant wastewater discharges, fish advisories have been issued to protect the public
from exposure to fish with elevated pollutant concentrations. Leaching of pollutants from surface
impoundments and landfills containing combustion residuals is known to impact off-site ground
water and drinking water wells at concentrations above maximum contaminant level (MCL)
drinking water standards, posing a threat to human health.3
In this report, EPA uses the term "steam electric power plant wastewater" to represent all
combustion-related wastewaters that contain pollutants covered by the revised steam electric
ELGs. For the EA, EPA evaluated only a subset of the wastestreams: flue gas desulfurization
(FGD) wastewater, fly ash transport water, bottom ash transport water, and combustion residual
1 The Clean Water Act does not require that EPA assess the water-related environmental impacts, or the benefits, of
its ELGs, and EPA did not make its decision on the final steam electric ELGs based on the expected benefits of the
rule. EPA does, however, inform itself of the benefits of its rule, as required by Executive Order 12866. See the
Benefits and Cost Analysis for the Effluent Limitations Guidelines and Standards for the Steam Electric Power
Generation Point Source Category (EPA-821-R-15-005).
2 The steam electric ELGs control the discharge of pollutants to surface waters and do not specifically regulate
"wastewater." To allow for more concise discussion in this EA report, EPA occasionally refers to "wastewater"
discharges and impacts without specifically referencing the pollutants in the wastewater discharges.
3 In this EA, EPA evaluated the threats to human health and the environment associated with pollutants leaching into
ground water from surface impoundments and landfills containing combustion residuals. If these leached pollutants
do not constitute the discharge of a pollutant to surface waters, then they are not controlled under the steam electric
ELGs. While the Coal Combustion Residuals (CCR) rulemaking is the major controlling action for these pollutant
releases to ground water, the ELGs could indirectly reduce impacts to ground water. These secondary improvements
are discussed in Section 7.8.
1-1
-------
Section 1—Introduction
leachate collected from landfills or surface impoundments). The goal of the EA was to answer
the following five questions regarding pollutant loadings from the evaluated wastestreams:
• What are the environmental concerns under current (i.e., baseline) discharge
practices?
• What are the environmental and exposure pathways for steam electric power plant
wastewater discharges to impact water quality, wildlife, and human health?
• What are the baseline environmental impacts to water quality and wildlife?
• What are the impacts to human health from baseline discharges?
• What are the potential improvements to water quality, wildlife, and human health
under the final rule?
The EA evaluated environmental concerns and potential exposures (wildlife and humans)
to pollutants commonly found in wastewater discharges from steam electric power plants. EPA
completed both qualitative and quantitative analyses. Qualitative analyses included reviewing
documented site impacts in literature and damage cases; assessing the pollutant loadings to
receiving waters and sensitive environments; and reviewing the effects of pollutant exposure on
ecological and human receptors. To quantify baseline impacts and improvements under the final
rule, EPA developed computer models to determine pollutant concentrations in the immediate
and downstream receiving waters, pollutant concentrations in fish tissue, and exposure doses to
ecological and human receptors from fish consumption. EPA compared the values calculated by
the models to benchmarks to determine the extent of the environmental impacts nationwide. EPA
also developed a model to determine the risk of reproductive impacts among fish and waterfowl
that have been exposed, via their diet, to selenium from steam electric power plant wastewater
discharges.
This report presents the methodology and results of the qualitative and quantitative
analyses performed to evaluate baseline discharges from steam electric power plants and
improvements under the final rule. The analyses presented in this report incorporate some
adjustments to current conditions in the industry. For example, these analyses account for
publicly announced plans from the steam electric power generating industry to retire or modify
steam electric generating units at specific power plants. These analyses also account for changes
to the industry that are expected to occur as a result of the recent CCR rulemaking by EPA's
Office of Solid Waste and Emergency Response (OSWER). These analyses, however, do not
reflect changes in the industry that may occur as a result of the Clean Power Plan [Clean Air Act
Section lll(d)].4
In addition to the EA, the final steam electric ELGs are supported by a number of reports
including:
Regulatory Impact Analysis for Effluent Limitations Guidelines and Standards for the
Steam Electric Power Generation Point Source Category, Document No. EPA-821-R-15-004.
This report presents a profile of the steam electric power generating industry, a summary of the
4 EPA completed a parallel set of quantitative EA analyses that reflect changes in the industry that may occur as a
result of the Clean Power Plan. Appendix I provides the results of those analyses.
-------
Section 1—Introduction
costs and impacts associated with the regulatory options, and an assessment of the final rule's
impact on employment and small businesses.
Benefits and Cost Analysis for the Effluent Limitations Guidelines and Standards for the
Steam Electric Power Generation Point Source Category (Benefits and Cost Analysis).,
Document No. EPA-821-R-15-005. This report summarizes the monetary benefits and societal
costs that result from implementation of the final rule.
Technical Development Document for Effluent Limitations Guidelines and Standards for
the Steam Electric Power Generating Point Source Category (TDD)., Document No. EPA-821-
R-15-007. This report includes background on the final rule; applicability and summary of the
final rule; industry description; wastewater characterization and identification of pollutants of
concern; treatment technologies and pollution prevention techniques; and documentation of
EPA's engineering analyses to support the final rule including cost estimates, pollutant loadings,
and non-water-quality impact assessment.
These reports are available in the public record for the final rule and on EPA's website at
http://water.epa.gov/scitech/wastetech/guide/steam_index.cfm.
The ELGs for the Steam Electric Power Generating Point Source Category are based on
data generated or obtained in accordance with EPA's Quality Policy and Information Quality
Guidelines. EPA's quality assurance and quality control activities for this rulemaking include the
development, approval, and implementation of Quality Assurance Project Plans for using
environmental data generated or collected from all sampling and analyses, existing databases,
and literature searches, and for developing any models that used environmental data. Unless
otherwise stated within this document, EPA evaluated the data used and associated data analyses
as described in these quality assurance documents to ensure they are of known and documented
quality, meet EPA's requirements for objectivity, integrity, and utility, and are appropriate for the
intended use.
1-3
-------
Section 2—Background and Scope
SECTION 2
BACKGROUND AND SCOPE
The final steam electric effluent limitations guidelines and standards (ELGs) apply to
establishments whose generation of electricity is the predominant source of revenue or principal
reason for operation, and whose generation results primarily from a process utilizing fossil-type
fuels (coal, oil, or gas), fuel derived from fossil fuel (e.g., petroleum coke, synthesis gas), or
nuclear fuel in conjunction with a thermal cycle using the steam water system as the
thermodynamic medium. The final rule applies to discharges associated with both the
combustion turbine and steam turbine portions of a combined cycle generating unit (see 40 CFR
423.10). EPA is revising or establishing best available technology economically achievable
(BAT) limitations, new source performance standards (NSPS), pretreatment standards for
existing sources (PSES), and pretreatment standards for new sources (PSNS) that apply to certain
discharges of seven wastestreams: flue gas desulfurization (FGD) wastewater, fly ash transport
water, bottom ash transport water, combustion residual leachate, flue gas mercury control
(FGMC) wastewater, gasification wastewater, and nonchemical metal cleaning wastes. See the
Technical Development Document (TDD)
(EPA-821-R-15-007) for more information on
the rule applicability and definitions, industry
description, wastestreams and pollutants of
concern, treatment technologies, baseline and
regulatory option pollutant loadings, costs of
implementing treatment technologies, and
revised standards.
As discussed in Section 1, EPA uses
the term "steam electric power plant
wastewater" to represent all combustion-
related wastewaters covered by the revised
steam electric ELGs. For the environmental Many steam electric power plants use large
assessment (EA), EPA evaluated only a subset surface impoundments to store and treat
of the wastestreams (see Table 2-1 below). wastewaters_ These impoundments are
"Combustion residuals" are the solid wastes hydrologically connected to surface and
associated with combustion-related power m-0undwater
plant processes, including fly and bottom ash;
FGD solids; FGMC wastes; and other wastewater treatment solids associated with steam electric
power plant wastewater. Steam electric power plants generate solid residuals from fuel
combustion and from emission control technologies. These solid residuals include fly ash,
bottom ash, and FGD solids. Plants remove these solid materials through both wet and dry
handling methods. Dry handling typically involves transferring the solids to a storage silo or
outdoor storage pile, to be either disposed of in a landfill or, depending on the particular residual,
5 EPA evaluated technology options associated with FGMC wastewater, gasification wastewater, and nonchemical
metal cleaning wastes as part of the regulatory options. However, no plants currently discharge FGMC wastewater,
all existing gasification plants are operating the technology used as the basis for the regulatory option, and EPA will
continue to reserve BAT/NSPS/PSES/PSNS for nonchemical metal cleaning wastes, as previously established
regulations do. Therefore, EPA estimated zero compliance costs and zero pollutant reductions associated with these
wastestreams and did not include these three wastestreams in the EA.
2-1
-------
Section 2—Background and Scope
used to create beneficial by-products such as wallboard or cement. However, many plants use
wet handling systems, which transport the wastes to a surface impoundment (e.g., ash pond)
using large quantities of water. For example, in wet systems, bottom ash collects at the bottom of
the boiler in a water bath, and the water containing the bottom ash is then typically transported to
a surface impoundment for storage and/or disposal. Fly ash may be handled similarly after it is
collected from the particulate collection system. The slurry stream exiting wet FGD systems,
which contains 10 to 20 percent FGD solids, is typically treated either in a surface impoundment
or in an advanced wastewater treatment system, then discharged to a receiving stream or reused
in other plant processes. Section 6 of the TDD describes the industry wastestreams in detail.
Table 2-1 lists the specific wastestreams evaluated in the EA.
Table 2-1. Steam Electric Power Plant Wastestreams Evaluated in the EA
Evaluated Wastestream
Description
Fly ash transport water
Water used to convey the fly ash particles removed from the flue gas via a collection
system.
Untreated ash transport waters contain significant concentrations of total suspended
solids (TSS) and metals, including arsenic, calcium, and titanium (see Section 6 of
the TDD for further details). The effluent from surface impoundments generally
contains low concentrations of TSS; however, metals are still present in the
wastewater, predominantly in dissolved form.
Bottom ash transport water
Water used to convey the bottom ash particles collected at the bottom of the boiler.
As noted above, untreated ash transport waters contain significant concentrations of
TSS and metals.
FGD wastewater
Wastewater generated from a wet FGD scrubber system. Wet FGD systems are used
to control sulfur dioxide (SO2) emissions from the flue gas generated in the plant's
boiler.
The pollutant concentrations in FGD wastewater vary from plant to plant depending
on the coal type, the sorbent used, the materials of construction in the FGD system,
the FGD system operation, the level of recycle within the absorber, and the air
pollution control systems operated upstream of the FGD system. FGD wastewater
contains significant concentrations of chlorides, total dissolved solids (TDS),
nutrients, and metals, including bioaccumulative pollutants such as arsenic, mercury,
and selenium (see Section 6 of the TDD for further details).
Combustion residual
leachate
Collected liquid that has percolated through or drains from a landfill or a surface
impoundment, where the steam electric power plant disposes of or stores a variety of
wastes from the combustion process.
Leachate contains high concentration of metals, such as boron, calcium, chloride,
and sodium, similar to FGD wastewaters and ash transport water. The metal
concentrations in the leachate are generally lower than those in FGD wastewater and
ash transport water (see Section 6 of the TDD for further details).
2-2
-------
Section 2—Background and Scope
Surface impoundments accumulate high
concentrations of toxic pollutants from fly ash
transport water, bottom ash transport water,
andFGD wastewater.
Surface impoundments act as a physical
treatment process to remove particulate
material from wastewater through gravitational
settling. The wastewater in surface
impoundments can include one specific type of
wastewater (e.g., fly ash transport water) or a
combination of wastewaters (e.g., fly ash
transport water and FGD wastewater).
Additionally, plants may transfer wastewater
streams from other operations into their on-site
impoundments (e.g., cooling tower blowdown
or metal cleaning wastes). The wastestreams
sent to surface impoundments can also include
coal pile runoff. Although coal pile runoff is
not the result of a combustion process, it can
contain many of the pollutants present in steam
electric power plant wastewater. Leachate or
seepage may occur from surface impoundments or landfills containing combustion residuals.6
Regardless of whether they use surface impoundments or an advanced treatment system, steam
electric power plants typically discharge wastewater into the natural environment where
numerous studies have raised concern regarding the toxicity of these wastestreams [ERG, 2013a;
NRC, 2006; Rowe et al, 2002; U.S. EPA, 2014a through 2014e]. Previous regulations at 40 CFR
423 control pH and polychlorinated biphenyls (PCBs) discharge from all wastestreams and TSS
and oil and grease from ash transport waters and other "low volume wastes" that include air
pollution control wastewater (see Section 1 of the TDD). Section 6 of the TDD discusses
wastewater characterization and selection of pollutants of concern.
Based on data EPA obtained from the 2010 Questionnaire for the Steam Electric Power
Generating Effluent Guidelines (Steam Electric Survey), EPA estimates that 1,079 steam electric
power plants are subject to the final rule (see Section 4 of the TDD). EPA limited the scope of
the EA to those plants that both 1) discharge directly to surface waters and 2) will reduce their
pollutant loadings as a result of the regulatory options evaluated, based on EPA projections.
Therefore, the EA scope excludes steam electric power plants that meet any of the following
criteria:
• Plants that do not discharge any of the wastestreams that are included in the final rule
(even if the plant does generate and reuse the wastestream without discharging to
surface waters).
• Plants that already comply with final rule or have plans to comply with the final rule
prior to the date when the plants would have to meet the new limitations and
standards.
In this EA, EPA evaluated the threats to human health and the environment associated with pollutants leaching into
ground water from surface impoundments and landfills containing combustion residuals. If these leached pollutants
do not constitute the discharge of a pollutant to surface waters, then they are not controlled under the steam electric
ELGs. While the CCR rulemaking is the major controlling action for these pollutant releases to ground water, the
ELGs could indirectly reduce impacts to ground water. These secondary improvements are discussed in Section 7.8.
2-3
-------
Section 2—Background and Scope
• Plants that that have announced plans to retire steam generating units (that would
otherwise be subject to the final rule) prior to the date that the plants would have to
meet the new limitations and standards.
• Plants that, based on EPA projections, will either convert to dry ash handling or
install tank-based FGD wastewater treatment systems to comply with the CCR
rulemaking.
• Plants that discharge only to publicly owned treatment works (POTWs).
In the EA, EPA evaluated the current impact and potential improvement to the
environment and human health from 195 plants that discharge directly to surface waters and that
EPA projects will reduce pollutant loadings as a result of the regulatory options evaluated. Table
2-2 presents the number of plants by discharge type (direct or indirect) included in the cost and
loadings analysis presented in Sections 9 and 10 of the TDD.
Table 2-2. Number of Plants Evaluated in the EA
Plant Description
Number of
Plants
Number of Plants in Scope of Final Rule
Plants that fall under the applicability of the final rule (40 CFR 423)
1,079
Cost and Loadings Analysis
Plants for which EPA calculated loadings in the cost and loadings analyses
(see Sections 9 and 10 of the TDD)
Plants that discharge only to surface waters (direct discharger)
Plants that discharge only to a POTW (indirect discharger)
Plants that discharge to surface waters and to a POTW (direct and indirect discharger)
202
191
7
4
Environmental Assessment
Plants evaluated in the EA (includes all direct dischargers)3
195
a - For the pollutant loadings and removals presented in this report, EPA included indirect dischargers to protect
confidential business information.
These 195 steam electric power plants discharge to the 222 immediate receiving waters
illustrated in Figure 2-1 (some plants discharge to multiple receiving waters). The EA includes
qualitative analysis of the pollutant loadings in evaluated wastestreams discharged from these
plants and the associated potential for environmental and human health impacts. As discussed in
Section 5, EPA developed and executed a national-scale immediate receiving water (IRW)
model to perform further quantitative modeling of the water quality, wildlife, and human health
impacts associated with discharges from the majority of these plants. The IRW model, which
excludes discharges to the Great Lakes and estuaries, encompasses 188 steam electric power
plants that discharge to 209 immediate receiving waters. As discussed in Section 8, EPA also
performed more detailed case study modeling of discharges from six steam electric power plants.
Figure 2-1 indicates the immediate receiving waters included in the IRW modeling and case
study modeling scopes.
2-4
-------
Section 2—Background and Scope
LEGEND
In EA Scope but Not Modeled
(Great Lakes and Estuaries) (13)
In EA Scope and National-Scale
IRW Model (203)
In EA Scope, National-Scale IRW
Model, and Case Study Model (6)
Not in EA Scope but Included in
Case Study Model (1)
Figure 2-1. Locations and Counts of Immediate Receiving Waters in EA Scope and Modeling Analyses
2-5
-------
Section 2—Background and Scope
EPA used the results from quantitative and qualitative assessments combined with the
literature review to evaluate and describe the environmental impacts caused by the discharge of
the evaluated wastestreams. EPA organized the remainder of this report into the following
sections:
• Section 3 describes the environmental concerns associated with the evaluated
wastestreams, including a discussion of the pollutants of concern and a review of
damage cases and other documented site impacts showing negative impacts to surface
water and ground water.
• Section 4 outlines how ecological and human receptors may be exposed to pollutants
(i.e., environmental pathways), describes the factors that control environmental
impacts for each pathway, and gives an overview of the methodology used to
quantitatively evaluate the environmental and human health impacts.
• Section 5 presents the modeling performed to support the EA including an overview
of the national-scale IRW model and the ecological risk model.
• Section 6 presents the environmental and human health impacts based on qualitative
review and quantitative assessments (modeling of plant-specific discharges) of
current (baseline) discharges.
• Section 7 presents the improvements to the environment and human health estimated
from the implementation of the regulatory options.
• Section 8 describes EPA's case study modeling of discharges from six steam electric
power plants, presents the environmental and human health impacts under baseline
conditions, and discusses the modeled improvements under the final rule.
• Section 9 presents EPA's conclusions on the environmental and human health
improvements estimated under the final rule.
2-6
-------
Section 3—Environmental And Human Health Concerns
SECTION 3
ENVIRONMENTAL AND HUMAN HEALTH CONCERNS
Current scientific literature indicates that steam electric power plant wastewater is not a
benign waste [NRC, 2006; Rowe et al., 2002]. Many of the common pollutants (e.g., selenium,
mercury, and arsenic) found in the evaluated wastestreams (i.e., fly ash and bottom ash transport
water, flue gas desulfurization (FGD) wastewater, and combustion residual leachate) present an
increased ecological threat due to their tendency to persist in the environment and bioaccumulate
in organisms. This often results in slow ecological recovery times following exposure. The toxic
impacts of steam electric power plant wastewater discharges on surface waters have been well
documented in studies of over 30 aquatic ecosystems receiving discharges from steam electric
power plants.7
Documented exceedances of drinking water maximum contaminant levels (MCLs)
downstream of steam electric power plants and the issuance offish advisories in receiving waters
indicate an ongoing human health concern caused by steam electric power plant wastewater
discharges. EPA identified more than 30 documented cases where ground water contamination
from surface impoundments extended beyond the plant boundaries, illustrating the threat to
ground water drinking water sources [ERG, 2015m].8 In other damage cases, EPA documented
locations where selenium in power plant wastewater discharges resulted in fish consumption
advisories being issued for surface waters.
The pollutants commonly discharged in the evaluated wastestreams cause environmental
harm by contaminating surface water and ground water (e.g., selenium concentrations from
steam electric power plants have resulted in fish kills). After being released into the environment,
pollutants can reside for a long time in the receiving waters, bioaccumulating and binding with
the sediment. There is documented evidence of slow ecological recovery as a result of these
pollutant discharges. Steam electric power plants also discharge to sensitive environments (e.g.,
impaired waters, waters under a fish consumption advisory, Great Lakes, valuable estuaries, and
drinking water sources). Some impacts might not be realized for years due to the persistent and
bioaccumulative nature of the pollutants released. Based on EPA's calculated baseline pollutant
loadings, the total amount of toxic pollutants currently being released in wastewater discharges
from steam electric power plants is significant and raises concerns regarding the long-term
impacts to aquatic organisms, wildlife, and humans that are exposed to these pollutants. For
details on the pollutant loadings analysis, see Section 10 of the Technical Development
Document (TDD) (EPA-821-R-15-007).
This section details environmental concerns associated with wastewater discharges from
steam electric power plants including changes in surface water quality and sediment
contamination levels; changes in ground water quality and potential contamination of private
7 Sources include ATSDR, 1998a, 1998b and 1998c; Charlotte Observer, 2010; DOE, 1992; EIP, 2010a and 2010b;
Roeetal., 2005; Sorensene/a/., 1983; Sorensen, 1988; Spechte/a/., 1984; and Vengoshetal., 2009.
8 In this EA, EPA evaluated the threats to human health and the environment associated with pollutants leaching into
ground water from surface impoundments and landfills containing combustion residuals. If these leached pollutants
do not constitute the discharge of a pollutant to surface waters, then they are not controlled under the steam electric
ELGs. While the Coal Combustion Residuals (CCR) rulemaking is the major controlling action for these pollutant
releases to ground water, the ELGs could indirectly reduce impacts to ground water. These secondary improvements
are discussed in Section 7.8.
-------
Section 3—Environmental And Human Health Concerns
drinking water wells; bioaccumulation of contaminants in fish and aquatic life, fish eaten by
piscivorous wildlife (i.e., fish-eating wildlife), and fish eaten by humans; and toxic effects on
fish and aquatic life. The section is organized into the following subsections:
• Section 3.1: Types of pollutants discharged in steam electric power plant wastewater.
• Section 3.2: Pollutant loadings associated with steam electric power plant wastewater.
• Section 3.3: Environmental impacts from steam electric power plant wastewater,
including ecological impacts, human health effects, damage cases and other
documented site impacts, and potential for impacts to occur in other locations.
• Section 3.4: Sensitive environments, including pollutant loadings to the Great Lakes
and Chesapeake Bay watersheds, impaired waters, waters issued fish advisories,
threatened and endangered species habitats, and drinking water resources.
• Section 3.5: Long recovery times.
3.1 TYPES OF POLLUTANTS DISCHARGED IN STEAM ELECTRIC POWER PLANT
WASTEWATER
This section provides an overview of the pollutants in steam electric power plant
wastewater discharges that are frequently cited as affecting local wildlife or pose a threat to
human health. A number of variables can affect the composition of steam electric power plant
wastewater, including fuel composition, type of combustion process, air pollution control
technologies implemented, and management techniques used to dispose of the wastewater
[Carlson and Adriano, 1993]. In addition, commingling steam electric power plant wastewater
with other wastestreams from the plant in surface impoundments can result in a chemically
complex effluent that is released to the environment [Rowe et al., 2002]. To identify pollutants
of concern for the final rule, EPA used the following sources of wastewater characterization
data: EPA's field sampling program; data supplied by industry or members of the public (e.g., in
questionnaire responses and public comments on the proposed rule); and various literature
sources (see Section 6 of the TDD and the preamble to the final rule for further details on
pollutants of concern). Pollutants such as metals, nutrients, and total dissolved solids (TDS),
including chloride and bromides, are the common pollutants found in steam electric power plant
wastewater that have been associated with documented environmental impacts or could have the
potential to cause environmental impacts based on the loadings and concentrations present in the
evaluated wastestreams.
3.1.1 Metals and Toxic Bioaccumulative Pollutants
Studies commonly cite metals and toxic bioaccumulative pollutants (e.g., mercury and
selenium) as the primary cause of ecological damage following exposure to steam electric power
plant wastewater [Rowe et al, 1996; Lemly, 1997a; Hopkins et al, 2000; Rowe et al, 2002] (see
Section 3.3.1). An important consideration in evaluating these pollutants is their bioavailability-
the ability of a particular contaminant to be assimilated into the tissues of exposed organisms. A
pollutant's bioavailability is affected by the characteristics of both the pollutant and surrounding
environment (e.g., temperature, pH, salinity, oxidation-reduction (redox) potential, total organic
content, suspended particulate content, and water velocity). Environmental conditions influence
the tendency of a dissolved pollutant to remain in solution or precipitate out of solution, sorb to
either organic or inorganic suspended matter in the water column, or sorb to the mixture of
-------
Section 3—Environmental And Human Health Concerns
materials (e.g., clays and humic matter) found in sediments [U.S. EPA, 2007a]. Pollutants that
precipitate out of solution can become concentrated in the sediments of a waterbody. Regardless,
organisms will bioaccumulate pollutants either by consuming pollutant-enriched sediments and
suspended particles, and/or by filtering ambient water containing dissolved pollutants.
Table 3-1 lists some of the common metals and toxic bioaccumulative pollutants found in
steam electric power plant wastewater that have been associated with documented health and
environmental impacts or could potentially cause health and environmental impacts based on the
loadings and concentrations present in the wastewater. Table 3-1 is intended to highlight the
pollutants of concern in steam electric power plant wastewater that are associated with health and
environmental impacts; it does not include all pollutants that may cause adverse impacts. Metals
and toxic bioaccumulative pollutants in steam electric power plant wastewater are present in both
soluble (i.e., dissolved) and paniculate (i.e., suspended) form. For example, EPA sampling data
collected for FGD wastewater in support of the steam electric ELGs shows that some pollutants
such as arsenic are present mostly in particulate form while other pollutants such as selenium and
boron are present mostly in soluble form. The remainder of the section provides additional
details on several key metals included in the environmental assessment (EA).
Table 3-1. Key Metals and Toxic Bioaccumulative Pollutants Found In Steam Electric
Power Plant Wastewater
Pollutant
Examples of Potential Health and Environmental Concerns
Aluminum
Aluminum contamination can lead to the inability of fish to maintain the balance of their fluids and
is associated with damage to amphibian eggs and larvae, mostly in areas under acid stress. Human
exposure to high concentrations has been linked to Alzheimer's disease.
Arsenic a
Arsenic contamination causes liver poisoning, developmental abnormalities, behavioral
impairments, metabolic failure, reduced growth, and appetite loss in fish and is associated with an
increased risk of the liver and bladder cancer in humans. Arsenic is also a potent endocrine
disrupter at low, environmentally relevant levels. Non-cancer impacts to humans can include
dermal, cardiovascular, and respiratory effects. Negative impacts can occur both after high-dose
exposure and repeated lower-dose exposures. Chronic exposure via drinking water has been
associated with excess incidence of miscarriages, stillbirths, preterm births, and low-birth weights.
Boron
Boron can be toxic to vegetation and to wildlife at certain water concentrations and dietary levels.
Human exposure to high concentrations can cause nausea, vomiting, and diarrhea.
Cadmium
Cadmium contamination can lead to developmental impairments in wildlife and skeletal
malformations in fish. Human exposure to high concentrations in drinking water and food can
irritate the stomach, leading to vomiting and diarrhea, and sometimes death. Chronic oral exposure
via diet or drinking water to lower concentrations can lead to kidney damage and weakened bones.
Chromium b
Chromium is not known to bioaccumulate in fish; however, high concentrations of chromium can
damage gills, reduce growth, and alter metabolism in fish. Human exposure to high concentrations
can cause gastrointestinal bleeding and lung problems.
Copper
Copper contamination can lead to reproductive failure, gill damage, and reduced sense of smell in
fish. Human exposure to high concentrations can cause nausea, vomiting, diarrhea, and liver and
kidney damage.
Iron
Iron contamination can reduce growth, increase susceptibility to injury and disease, and decrease
egg hatchability in fish. Human exposure to high concentrations can cause metabolic changes and
damage to the pancreas, liver, spleen, and heart.
Lead
Lead contamination can delay embryonic development, suppress reproduction, and inhibit growth
in fish. Human exposure to high concentrations in drinking water can cause serious damage to the
brain, kidneys, nervous system, and red blood cells.
3-3
-------
Section 3—Environmental And Human Health Concerns
Table 3-1. Key Metals and Toxic Bioaccumulative Pollutants Found In Steam Electric
Power Plant Wastewater
Pollutant
Manganese
Mercury °
Nickel
Selenium d
Thallium
Vanadium
Zinc
Examples of Potential Health and Environmental Concerns
Manganese primarily accumulates in organisms lower in the food chain such as phytoplankton,
algae, mollusks, and some fish. Although high levels can be toxic to humans, manganese is not
generally considered toxic when ingested. The most common impacts due to human exposure to
high concentrations involve the nervous system.
Once in the environment, mercury can convert into methylmercury, increasing the potential for
bioaccumulation. Methylmercury contamination can reduce growth and reproductive success in fish
and invertebrates. Human exposure at levels above the MCL for relatively short periods can result
in kidney and brain damage. Fetuses, infants, and children are particularly susceptible to impaired
neurological development from methylmercury exposure.
At low concentrations, nickel can inhibit the growth of microorganisms and algae. Nickel toxicity
in fish and aquatic invertebrates varies among species and can damage the lungs, immune system,
liver, and kidneys. Human exposure to high concentrations can cause gastrointestinal and kidney
damage.
Selenium readily bioaccumulates. Elevated concentrations have caused fish kills and numerous
sublethal effects (e.g. , organ damage, decreased growth rates, reproductive failure) to aquatic and
terrestrial organisms. In humans, short-term exposure at levels above the MCL can cause hair and
fingernail changes, damage to the peripheral nervous system, and fatigue and irritability. Long-term
exposure can damage the kidney, liver, and nervous and circulatory systems.
In humans, short-term exposure to thallium can lead to neurological symptoms, alopecia,
gastrointestinal effects, and reproductive and developmental damage. Long-term exposures at levels
above the MCL change blood chemistry and damage liver, kidney, intestinal and testicular tissues
and cause hair loss.
Vanadium contamination can increase blood pressure and cause neurological effects in animals.
There are very few reported cases of oral exposure to vanadium in humans; however, a few
reported incidences documented diarrhea and stomach cramps. It also has been linked to the
development of some neurological disorders and cardiovascular diseases.
Zinc contamination changes behavior, reduces oxygen supply, and impairs reproduction in fish. In
humans, short-term exposure can cause nausea, vomiting, and stomach cramps. Long-term
exposure can cause anemia.
a - Arsenic exists in two primary forms: arsenic III (arsenite) and arsenic V (arsenate).
b - Chromium exists in two primary forms: chromium III oxide and chromium VI (hexavalent chromium).
c - The EA evaluated two forms of mercury: total mercury and methylmercury.
d - Selenium exists in two primary forms: selenium IV (selenite) and selenium VI (selenate).
Selenium
Selenium is the most frequently cited pollutant associated with documented
environmental impacts to ecological receptors following exposure to steam electric power plant
wastewater [NRC, 2006]. The toxic potential of selenium is related to its chemical form and
solubility. The predominant chemical forms of selenium in aquatic systems that receive steam
electric power plant wastewater discharges are selenite and selenate [Besser et a/., 1996]. The
uptake of selenium by aquatic organisms is controlled by dissolved oxygen levels, hardness, pH,
salinity, temperature, and the other chemical constituents present [NFS, 1997]. In alkaline
conditions, selenite [Se(IV)] will oxidize in the presence of oxygen to become selenate [Se(VI)];
selenate is both stable and soluble and is the commonly found form of the chemical in alkaline
soils and waters. In acidic conditions, selenite is insoluble due to its tendency to bind to iron and
aluminum oxides [WHO, 1987]. Organic forms of selenium are more bioavailable for uptake
than selenate and selenite and may play an important role determining selenium toxicity in
exposed aquatic organisms [Besser etal., 1993; Rosetta and Knight, 1995].
-------
Section 3—Environmental And Human Health Concerns
The extent to which selenium is found in ecological receptors is affected by
bioaccumulation, biomagnification, and maternal transfer. Bioaccumulation occurs when an
organism absorbs a toxic substance through food and exposure to the environment at a faster rate
than the body can remove the substance. The bioaccumulation of selenium is of particular
concern due to its potential to impact higher trophic levels through biomagnification [Coughlan
and Velte, 1989] and offspring through maternal transfer [Hopkins et a/., 2006; Nagle et a/.,
2001]. A laboratory study demonstrated that diet can be an important source of trace element
exposure in aquatic snakes and potentially other amphibians [Hopkins et a/., 2002]. Hopkins
reported that the snakes accumulated significant concentrations of the trace elements, most
notably selenium. This study also revealed that amphibian prey species are able to migrate
considerable distances and can therefore be exposed to toxic levels of selenium even if they do
not inhabit a contaminated site. Because of bioaccumulation and biomagnification, selenium-
related environmental impacts can linger for years even after exposure to steam electric power
plant wastewater has ceased [Rowe et a/., 2002].
Selenium-related impacts observed by
scientists include lethal effects such as fish kills,
sublethal effects such as histopathological
changes and damage to reproductive and
developmental success, and the impacts of these
effects on aquatic populations and communities.
In a 1991 study, Sorensen found that dissolved
selenium levels as low as 3 to 8 micrograms per
liter (ug/L) in aquatic environments can be life-
threatening to fish [NFS, 1997]. Section 3.3.1
presents further details regarding the lethal and
sublethal effects on aquatic organisms caused by
selenium from steam electric power plant
wastewater.
Elevated arsenic tissue concentrations
are associated with several biological
impacts such as liver tissue death,
developmental abnormalities, and
reduced growth.
In addition to ecological impacts, EPA
has documented numerous damage cases where
selenium in steam electric power plant
wastewater discharges resulted in fish
consumption advisories being issued for surface
waters and selenium MCLs being exceeded in
ground water, suggesting that selenium
concentrations in power plant wastewater have the potential to impact human health [NRC,
2006; U.S. EPA, 2014a through 2014e]. Short-term exposure at levels above the MCL, 0.05
mg/L [U.S. EPA, 2009e], can cause hair and fingernail changes, damage to the peripheral
nervous system, and fatigue and irritability in humans. Long-term exposure can damage the
kidney, liver, and nervous and circulatory systems.
Toxic Pollutant Impacts to Ecological
Receptors
Selenium discharges have caused
numerous cases of fish kills and
population decline due to reproductive
impacts. Bioaccumulation can cause
selenium-related environmental impacts
to linger for years even after exposure to
steam electric power plant wastewater
has ceased.
Fish and invertebrates exposed to steam
electric power plant wastewater have
exhibited elevated mercury levels in
their tissues and developed sublethal
effects such as reduced growth and
reproductive success.
3-5
-------
Section 3—Environmental And Human Health Concerns
Mercury
Mercury is a volatile metal and highly toxic compound that represents an environmental
and human health threat even in small concentrations. One of the primary environmental
concerns regarding mercury concentrations in steam electric power plant wastewater is the
potential for methylmercury to form in combustion residual surface impoundments and
constructed wetlands prior to discharge and in surface waters following discharge.
Methylmercury is an organic form of mercury that readily bioaccumulates in fish and other
organisms and is associated with high rates of reproductive failure [WHO, 1976]. Bacteria found
in anaerobic conditions, such as those that may be present in sediments found on the bottom of
combustion residual surface impoundments or in river sediments, convert mercury to
methylmercury through a process called methylation [WHO, 1976]. Microbial methylation rates
increase in acidic and anoxic environments with j^^i
high concentrations of organic matter. Sublethal
effects from mercury exposure include reduced
growth and reproductive success, metabolic
changes, and abnormalities of the liver and kidneys.
Human exposure at levels above the MCL, 0.002
mg/L [U.S. EPA, 2009e], for relatively short periods
of time can result in kidney and brain damage.
Pregnant women who are exposed to mercury can
pass the contaminant to their developing fetus,
leading to possible mental retardation and damage
to other parts of the nervous system [ATSDR,
1999]. Studies have documented fish and
invertebrates exposed to mercury from steam
electric power plant wastewater exhibiting elevated
levels of mercury in their tissues and developing
sublethal effects such as reduced growth and
reproductive success [Rowe etal., 2002].
Toxic Pollutant Impacts to Human
Receptors
Pregnant women exposed to mercury
can pass the contaminant to their
developing fetus, leading to possible
mental retardation and damage to
other parts of the nervous system.
Inorganic arsenic is a carcinogen
(i.e., causes cancer). Cadmium is a
probable carcinogen.
Human exposure to high
concentrations of lead in drinking
water can cause serious damage to
the brain, kidneys, nervous system,
and red blood cells, especially in
children.
Arsenic
Arsenic, like selenium, is of concern because it is soluble in near-neutral pH and in
alkaline conditions, which are commonly associated with steam electric power plant wastewater.
As a soluble pollutant, arsenic leaches into ground water and is highly mobile. Arsenic is
frequently observed at elevated concentrations at sites located downstream from combustion
residual surface impoundments [NRC, 2006]. Inorganic arsenic, a carcinogen, is found in natural
and drinking waters mainly as trivalent arsenite (As(III)) or pentavalent arsenate (As(V)) [WHO,
2001]. Both the arsenite and arsenate forms are highly soluble in water.
Arsenic is also of concern due to its tendency to bioaccumulate in aquatic communities
and potentially impact higher-trophic-level organisms in the area. For example, studies have
documented water snakes, which feed on fish and amphibians, with arsenic tissue concentrations
higher than their prey [Rowe et a/., 2002]. Elevated arsenic tissue concentrations are associated
with several biological impacts such as liver tissue death, developmental abnormalities,
behavioral impairments, metabolic failure, reduced growth, and appetite loss [NRC, 2006; Rowe
et al., 2002; U.S. EPA 201 If].
3-6
-------
Section 3—Environmental And Human Health Concerns
Humans are exposed to arsenic primarily by ingesting contaminated drinking water
[WHO, 2001]. Humans are also exposed to arsenic by consuming contaminated fish. Of greatest
concern is inorganic arsenic, which can cause cancer in humans. Several studies have shown that
most arsenic in fish is organic and not harmful to humans. Inorganic arsenic typically accounts
for 4 percent or less of the total arsenic that accumulates in fish.9 The highest potential exposure
is for individuals whose diet is high in fish and particularly shellfish [U.S. EPA, 1997b].
As discussed in Section 3.3.4, EPA has documented several damage cases where arsenic
levels exceeded drinking water standards in ground water near combustion residual surface
impoundments [U.S. EPA, 2014b through 2014e]. Arsenic contamination of ground water at the
levels documented represents a potential human health threat, if either the aquifer is used as a
drinking water source or the ground water contaminates a downstream drinking water source.
Cadmium
The speciation and toxicity of cadmium in water depends on the water's salinity,
hardness, temperature, and organic content [WHO, 1992]. Cadmium tends to bioaccumulate
readily in mollusks, soil invertebrates, and microorganisms. Due to its chemical similarity to
calcium, it can also interfere with calcium uptake in aquatic organisms, which can cause
sublethal effects in fish such as skeletal malformation. Divalent cadmium (Cd(II)) is the species
most commonly found in an aquatic environment, but depending on the quality of the water,
cadmium can also occur as cadmium carbonate, hydroxide, sulfite, sulfate, or chlorides.
EPA determined that cadmium is a probable human carcinogen. Studies found lung
cancer in humans and rats exposed to cadmium via inhalation. In humans, chronic low-level
exposure to cadmium from contaminated air, drinking water, or food can cause kidney failure.
Chronic low-level exposure from contaminated drinking water or food can also lead to fragile
bones. Exposure via inhalation at high levels can damage lungs and exposure via food and
drinking water can irritate the stomach, leading to vomiting and diarrhea [ATSDR, 2012].
Thallium
Thallium typically exists as the monovalent or trivalent thallium ion [WHO, 1996]. It is
soluble in most waters and is readily available to aquatic life. Thallium can bioaccumulate in fish
and vegetation in fresh and marine waters, as well as marine invertebrates, which suggests that
thallium may be a potential threat to higher order organisms in vulnerable ecosystems [U.S.
EPA, 2011 a]. Studies in humans and animals indicate that thallium compounds are readily
absorbed through ingestion of food and water and maternal transfer [WHO, 1996].
In humans, elevated thallium concentrations can lead to neurological symptoms (e.g.,
weakness, sleep disorders, muscular problems), alopecia (i.e., loss of hair from the head and
body), and gastrointestinal effects (e.g., diarrhea and vomiting). Long-term exposures at levels
above the MCL, 0.002 mg/L [U.S. EPA, 2009e], lead to changes in blood chemistry, damage to
liver, kidney, and intestinal and testicular tissues, and hair loss. Thallium exposure can also cause
reproductive and developmental damage [U.S. EPA, 2009a].
9 Based on a 1996 literature review of toxicity and exposure concerns related to arsenic in seafood prepared for U.S.
EPA Region 10, inorganic arsenic comprised higher than four percent total arsenic for three species (shark, sturgeon,
and sucker). Inorganic arsenic for all other species accounted for less than 4 percent of the total arsenic [U.S. EPA,
1997bj
3-7
-------
Section 3—Environmental And Human Health Concerns
Lead
Neither metallic lead nor many of its common mineral forms are soluble in water,
although it can be soluble in some acids or water with low pH; thus, lead is commonly present in
precipitate form in water. Therefore, steam electric power plant wastewater may initially have
high concentrations of lead, but later sampling of the wastewater can show decreased
concentrations because the lead settles out quickly. Lead will accumulate in aquatic organisms,
but depends on the species. Studies have shown lead to delay embryonic development, suppress
reproduction, and inhibit growth rate among fish, crab, and several other aquatic organisms [U.S.
EPA, 1984]. Human exposure to high concentrations of lead in drinking water can seriously
damage the brain, kidneys, nervous system, and red blood cells, especially in children.
Boron
Boron is primarily found in the environment combined with oxygen in compounds called
borates [ATSDR, 201 Ob]. Boron concentrations in North American waters are typically below
0.1 mg/L [WHO, 1998], although areas with natural boron-rich deposits may have ground water
levels as high as 300 mg/L [ATSDR, 2010b]. The World Health Organization (WHO) suggests
that the potential of adverse effects of boron on the aquatic ecosystem is low because the no-
effect concentration (1 mg/L) is much greater than levels found in the ambient environment.
Boron does not magnify through the food chain, but does accumulate in aquatic and terrestrial
plants. While it is an essential micronutrient for higher plants, there is a small range between
deficiency and toxicity in some plants. Studies of acute exposure in fish yielded toxicity values
ranging from approximately 10 to 300 mg/L with rainbow trout and zebra fish being the most
sensitive. Mallard duckling growth was impacted at dietary levels of 30 and 300 milligrams per
kilogram (mg/kg), while survival was reduced at 1,000 mg/kg [WHO, 1998].
EPA has not set a numerical criterion under the National Recommended Water Quality
Criteria (NRWQC) for aquatic life, but it has issued a narrative criterion of 0.75 mg/L for
sensitive crops that receive long-term irrigation.
EPA has not set a NRWQC for human health. Very few human studies have examined
health effects resulting from boron exposure through oral ingestion. However, one study
documents nausea, vomiting, and diarrhea in an adult male who ingested 85 mg/kg of boron (30
g as boric acid) [ATSDR, 201 Ob]. In addition, animal experiments indicate that boron in the form
of boric acid and borate affects reproductive and developmental processes at levels that are
approximately 100 to 1,000 times greater than normal exposure levels, approximately 1.2
milligrams per day (mg/day) [WHO, 1998].
Manganese
In water, manganese tends to attach to particles or settle into the sediment [ATSDR,
2008b]. It occurs in both dissolved and suspended forms, depending on the water chemistry (e.g.,
pH) [WHO, 2011]. Manganese can bioaccumulate in lower organisms, such as phytoplankton,
algae, mollusks, and some fish, but not in higher organisms. Studies suggest that
biomagnification up the food chain is not significant [ATSDR, 2008b].
Due to a high bioaccumulation factor and concentrations in mollusks, EPA established a
criterion to protect consumers of marine mollusks—100 micrograms per liter (|ig/L) for marine
-------
Section 3—Environmental And Human Health Concerns
waters [U.S. EPA, 1986]. Although high levels can be toxic to humans, manganese is an
essential nutrient required to maintain health and is generally not considered to be toxic when
ingested [WHO, 2011]. EPA did not set a primary MCL for manganese in drinking water;
however, EPA did set secondary (nonenforceable) standards at 50 |ig/L to minimize
objectionable qualities in the drinking water that cause laundry stains and objectionable tastes in
beverages [U.S. EPA, 2009e].
3.1.2 Nutrients
Nutrients (e.g., phosphorus and nitrogen) are essential components for plants and animals
to grow and develop; however, increased nutrient concentrations can upset the delicate balance
of nutrient supply and demand required to maintain aquatic life in surface waters. For example,
excess nutrients can cause low oxygen in surface waters (hypoxia) and harmful algal blooms.
These are primarily problems for estuaries, such as the Chesapeake Bay, and coastal waters, such
as the Gulf of Mexico. Nutrient concentrations present in steam electric power plant wastewater
are primarily attributed to the fuel composition and air pollution controls in the combustion
process.
Total nitrogen loadings from coal-fired power plants could potentially increase
significantly in the future as air pollution limits become stricter and air pollution control use
increases. While wastewater from an individual steam electric power plant can have a relatively
low nitrogen concentration the total nitrogen loadings from a single plant can be significant due
to high wastewater discharge flow rates. Total nutrient loadings from multiple power plants are
especially a concern for waterbodies that are nutrient-impaired or in watersheds that contribute to
downstream nutrient problems. High nutrient loadings to surface waters can affect the ecological
stability of freshwater and saltwater aquatic systems. For example, excessive levels of nutrients
can stimulate rapid growth of plants, algae, and cyanobacteria on or near the waterbody surface,
which in turn can obstruct sunlight penetration, increase turbidity, and decrease dissolved oxygen
levels [U.S. EPA, 2015a]. These aquatic changes can potentially kill bottom-dwelling aquatic
plants. Cyanobacterial blooms can also produce toxic secondary metabolites, known as
cyanotoxins, that can have negative impacts to humans and wildlife that consume water
contaminated with cyanobacteria. The presence of high levels of cyanotoxins in recreational and
drinking water may cause fever, headaches, abdominal pain, and other symptoms in humans.
Severe human impacts include seizures, liver failure, respiratory arrest, and (rarely) death [U.S.
EPA, 2012d].
3.1.3 TDS
TDS, a reflection of water's salinity level, is a measure of the amount of dissolved matter
in water. TDS comprises primarily inorganic salts and dissolved metals, as well as a small
amount of organic matter. Common inorganic salts found in TDS can include cations (positively
charged ions), such as calcium, magnesium, potassium, and sodium, and anions (negatively
charged ions) such as carbonates, nitrates, bicarbonates, chlorides, and sulfates. TDS
concentrations in steam electric power plants wastestreams include contributions from dissolved
metals, chlorides, and bromides. Dissolved metals and other TDS constituents are found in
wastewater particularly at acidic pH levels when they exhibit high solubilities. The specific
constituents in TDS in steam electric power plant wastewater cause the negative impacts.
3-9
-------
Section 3—Environmental And Human Health Concerns
Bromides
Bromide is the anion of bromine; it commonly exists as salts with potassium and other
cations, which are usually very soluble in water. In water, bromide reacts to form hydrobromic
acid (HBr) and hypobromous (HOBr), bromous (HBrO2), and bromic (HBrO3) oxyacids.
Bromide is commonly found in nature, with levels ranging from trace amounts to 0.5 mg/L in
fresh water and levels ranging from 65 to over 80 mg/L in seawater. The bromide ion has a low
degree of toxicity, and animal testing suggests very low acute toxicity upon oral administration
[WHO, 2009].
While bromide itself is not thought to be toxic at levels present in the environment, its
reaction with other constituents in water may be cause for concern now and into the future. The
bromide ion in water can form brominated disinfection by-products (DBFs) when drinking water
plants use certain processes including chlorination and ozonation to disinfect the incoming
source water. Bromide can react with the ozone, forming bromates, or with chlorine or chlorine-
based disinfectants used at drinking water treatment plants, to form brominated and mixed
chloro-bromo DBFs, such as trihalomethanes (THMs) or haloacetic acids (HAAs) [WHO, 2009].
EPA has set MCLs for the following DBFs in chlorinated water:
• 0.010 mg/L for bromate due to increased cancer risk from long-term exposure.
• 0.060 for HAAs due to increased cancer risk from long-term exposure HAAs include
dichloroacetic acid, trichloroacetic acid, chloracetic acid, bromoacetic acid, and
dibromoacetic acid.
• 0.080 mg/L for total trihalomethanes (TTHMs) due to increased cancer risk and liver,
kidney, or central nervous system problems from long-term exposure [U.S. EPA,
2009e]. TTHMs include the brominated trihalomethanes (bromodichloromethane,
bromoform, dibromochloromethane) and chloroform. MCL goals for the individual
trihalomethanes include 0 (zero) for bromodichloromethane and bromoform.
Studies indicate that exposure to THMs and other DBFs from chlorinated water are
associated with human bladder cancer [Villanueva et a/., 2004; Cantor et a/., 2010]. Bromine-
substituted DBFs are generally thought to have higher risks of cancer and other adverse human
health effects compared to DBFs containing chlorine instead of bromine [Cantor et a/., 2010].
EPA has determined that bromodichloromethane and bromoform are likely to be carcinogenic to
humans by all exposure routes and there is suggestive evidence of dibromocloromethane
carcinogencity. Excess cancer risk (based on increased risk to 1-in-a-million) occurs at
concentrations above 0.001 mg/L for bromodichloromethane, 0.008 mg/L for bromoform, and
0.0008 mg/L for dibromochlormethane [U.S. EPA, 2005c].
DBF formation and the individual form of the DBF are influenced by factors such as
bromide ion concentration, pH of the source water, the disinfectant dose (ozone or chlorine),
reaction or contact time, and organic matter concentration and reactivity [Liang and Singer,
2003; U.S. EPA, 2005c]. Studies have shown that higher bromide levels in source waters shift
the distribution of the TTHMs towards brominated species [Krasner etal., 1989] and the types of
HAAs from chlorinated to brominated and mixed chloro-bromo haloacetic acids [Heller-
Grossman, 1993; Cowman and Singer, 1996].
3-10
-------
Section 3—Environmental And Human Health Concerns
Under the Safe Drinking Water Act (SDWA), drinking water treatment plants must
reduce DBFs in their treated water and reduce exposure to customers. EPA conducted a
nationwide survey that showed that bromide levels in source water above 400 ug/L corresponded
with increased levels of DBFs in the treated water [Weinberg, 2002]. Due to increased bromide
concentrations in surface water, drinking water treatment plants have found increased difficulty
meeting regulatory limits on DBFs [U.S. EPA, 2012a; Handke, 2009; Fiske et a/., 2011; States et
a/., 2013; Wilson et a/., 2013]. In general, drinking water produced using surface water had
higher concentrations of the DBFs than drinking water produced using ground water [U.S. EPA,
2005c].
The city of Pittsburgh, in cooperation with the University of Pittsburgh, completed a
multiyear study on the Allegheny River to determine the major sources of bromide discharges,
including coal-fired power plants. Typically, bromide concentrations are very low in the river,
but there are increased levels near industrial sites. The bromide concentration in the source water
provided a linear correlation to bromination in the drinking water. At a concentration of 0.050
mg/L in the source water, 62 percent of the TTHMs were the three brominated trihalomethane
species. At a concentration of 0.150 mg/L, 83 percent of the TTHMs were the three brominated
trihalomethane species [States etal., 2013].
The California Urban Water Agencies (CUWA) evaluated costs associated with increased
bromide levels in the source water for baseline and potential future DBF controls. CUWA
developed virtual water treatment plants (WTPs) to represent their different source water areas
and treatment needs, with virtual WTP design capacities ranging from 40 to 800 million gallons
per day. To achieve potential future standards on currently regulated pollutants, including DBFs,
CUWA estimated costs for capital improvements and added annual operation and maintenance
costs. On the low end, CUWA anticipated spending between $46 million to $923 million in
capital improvements and $1 million to $59 million on annual operation and maintenance costs
to each virtual WTP (costs vary based on the characteristics of the virtual WTP). On the high
end, CUWA anticipated spending between $98 million and almost $2 billion in capital
improvements and between $2 million and $127 million in annual operation and maintenance
costs for each virtual WTP [CUWA, 2011].
Bromide is naturally present in coal at trace levels and becomes part of the flue gas air
emissions following combustion at steam electric power plants. Combusting coal with higher
levels of bromide is known to improve removal of mercury from air emissions at steam electric
power plants that operate wet FGD scrubbers. Accordingly, steam electric power plant operators
might add bromide-containing salts (e.g., calcium bromide) during coal combustion to improve
mercury removal efficiency. The bromide-containing salts convert the mercury Hg° form into the
more water soluble Hg2+ form. Bromide is not typically removed from steam electric power plant
wastewaters prior to discharge to surface waters. As discussed earlier, bromides in surface waters
can react with organic matter in the surface water to form DBFs at drinking water treatment
plants. A recent study identified four drinking water treatment plants that experienced increased
levels of bromide in their source water, and corresponding increases in the formation of
brominated DBFs, after upstream steam electric power plants installed wet FGD scrubbers
[McTigue et a/., 2014]. Bromide loadings into surface waters from coal-fired steam electric
power plants could potentially increase in the future as more plant operators add bromide to help
control mercury emissions.
3-11
-------
Section 3—Environmental And Human Health Concerns
Chlorides
Studies have found that combustion residual leachate reaching ground water has caused
chloride levels to exceed secondary MCLs [NRC, 2006]. Chlorides contribute to the high TDS
levels typical of steam electric power plant wastewater, as do calcium and magnesium. Both
chlorides and TDS levels affect the availability and toxicity of other steam electric power plant
wastewater constituents, including metals. As TDS and chlorides levels fluctuate, so do the
amounts of other metals that dissolve due to solubility characteristics.
EPA recommends the following for chlorides: criterion maximum concentration of
860 mg/L (acute effects) and criterion continuous concentration of 230 mg/L (chronic effects)
[U.S. EPA, 2009d]. Exceeding these chlorides levels in wastewater discharges can be harmful to
animals and plants in nonmarine surface waters and can disrupt ecosystem structure. It can also
adversely affect biological wastewater treatment processes. Furthermore, excessively high
chlorides concentrations in surface waters can impair their use as source waters for potable water
supplies. If sodium is the predominant cation present, the water will have an unpleasant taste due
to the corrosive action of chloride ions.
3.2 LOADINGS ASSOCIATED WITH STEAM ELECTRIC POWER PLANT WASTEWATER
As discussed above, the pollutants
commonly found in steam electric power plant
wastewater such as metals, nutrients, and TDS
(including bromides and chlorides) can cause
considerable harm to surface waters, aquatic life, „_,. . ,. ,
•\A\-C ju u uu T?T>A *• * j 11 * * EPA estimates that discharges from
wildlife, and human health. EPA estimated pollutant
Pollutant Loadings: How Does the
Steam Electric Power Generating
Industry Compare?
steam electric power plants alone
contribute approximately one-third of
the toxic weighted pound equivalent
(TWPE) pollutant loadings to the
environment among all industrial
categories that report discharges under
NPDES permits.
loadings for the steam electric power plant
wastestreams evaluated and considered as part of the
revision to the steam electric ELGs (i.e., FGD
wastewater, fly ash transport water, bottom ash
transport water, and combustion residual leachate).
The total pollutant loadings for the evaluated
wastestreams are significant, with these discharges
accounting for over one-third of the toxic pollutants reported to be discharged in industrial
National Pollutant Discharge Elimination System (NPDES) permits [ERG, 2015a]. EPA
estimated the amount of pollutants (i.e., loadings) discharged by steam electric power plants
throughout the United States for the evaluated wastestreams as almost 3 million toxic-weighted
pound equivalents (TWPE) annually.10 EPA uses TWFs as a way to better understand how
treatment technologies and industry discharges compare to one another [U.S. EPA, 2012b].
Although EPA uses TWFs and the estimated TWPE as an indicator of a pollutant's relative
potential to cause harm, EPA does not use TWPE to represent actual aquatic or human health
impacts that may have occurred at specific locations due to these pollutant loadings. To assess
10 To calculate the TWPE, EPA multiplies a mass loading of a pollutant in pounds per year (Ib/yr) by a pollutant-
specific weighting factor, called the toxic weighting factor (TWF), to derive a "toxic equivalent" loading (Ib-
equivalent/yr), or TWPE. TWFs account for differences in toxicity across pollutants and allow mass loadings of
different pollutants to be compared on the basis of their toxic potential. EPA has developed TWFs for more than
1,000 pollutants based on aquatic life and human health toxicity data, as well as physical/chemical property data
[U.S. EPA, 2012b].
-------
Section 3—Environmental And Human Health Concerns
impacts to aquatic life or human health, EPA uses the amount of pollutant loadings discharged to
the surface water and the resulting concentrations in the surface waters.
When coupled with the types of impacts associated with the pollutants, the magnitude of
the loadings raises concern about the risks that these discharges present to the aquatic
environment and the surrounding ecosystem. This section presents the annual baseline11 pollutant
loadings associated with the evaluated wastestreams and compares steam electric discharges to
those of other industries to provide perspective on the magnitude of the loadings and subsequent
potential impact these wastestreams pose to the environment.
3.2.1 Annual Baseline Pollutant Loadings
In support of the final rule, EPA estimated the pollutant loadings discharged from steam
electric power plants for the evaluated wastestreams, as described in Section 10 of the TDD.12
Table 3-2 presents the baseline annual pollutant loadings discharged for select pollutants
considered for analysis in the EA.13 EPA presents these loadings in terms of pounds and TWPE
and lists the TWF where applicable. The pollutants with the highest annual TWPE discharges are
manganese, cadmium, boron, thallium, mercury, selenium, and arsenic. Although the total
pounds discharged of arsenic, cadmium, mercury, and
thallium are lower than other pollutants, their relative
toxicity (as represented by the TWF) results in a large
TWPE. Other pollutants, such as boron and
manganese, are relatively low in toxicity but have a
high TWPE due to the fairly high amount of these
pollutants in steam electric power plant wastewater
discharges. The high TWPE for selenium results from
a combination of its quantity discharged in steam
electric power plant wastewaters and its TWF.
Pollutant Loadings from Steam
Electric Power Plants Evaluated
Wastestreams
2,210,000,000 pounds of pollutants
per year.
2,680,000 pounds of TWPE per
year.
11 The analyses presented in this report incorporate some adjustments to current conditions in the industry. See
Section 1 for further details.
12 Prior to finalizing the rulemaking, EPA revised the datasets used to calculate pollutant loadings for bottom ash
transport water and fly ash transport water. The final industry loadings calculated using these revised datasets are
presented in the TDD. The total industry loadings presented in Section 3.2 reflect the revised datasets. However,
EPA did not rerun the EA models and other analyses to reflect the final loadings dataset. EA analyses used
previously calculated version of the steam electric power plant pollutant loadings that were derived following the
same methodology. The EA pollutant loadings are included in DCN SE05620. Pollutant-specific loadings and
removals presented in this report are based on the previously calculated version. Appendix J presents the results of a
sensitivity analysis that evaluated the potential for these loadings revisions to affect the EA analyses.
13 EPA selected the pollutants listed in Table 3-2 (which represent a subset of all steam electric pollutants of
concern) for analysis in the EA based on the following factors for each pollutant: presence of the pollutant in the
evaluated wastestreams (see Table 2-1); documented elevated levels of the pollutant in surface waters or wildlife
from exposure to steam electric power plant wastewater; and magnitude of the pollutant loadings to receiving
waters.
3-13
-------
Section 3—Environmental And Human Health Concerns
Table 3-2. Annual Baseline Pollutant Discharges from Steam Electric Power Plants
(Evaluated Wastestreams)
Pollutant a
TWFb
Annual Discharge,
pounds (Ibs) c
Annual TWPE,
pound-equivalent
(Ib-eq) c
Metals and Toxic Bioaccumulative Pollutants
Manganese
Cadmium
Boron
Thallium
Mercury
Selenium
Arsenic
Aluminum
Lead
Copper
Vanadium
Iron
Nickel
Zinc
Chromium VI
0.103
22.8
0.00834
2.85
110.0
1.12
3.47
0.0647
2.24
0.623
0.280
0.00560
0.109
0.0469
0.517
7,530,000
13,300
31,300,000
63,700
1,490
140,000
29,600
1,410,000
19,700
31,200
66,000
2,740,000
120,000
174,000
156
773,000
303,000
261,000
182,000
164,000
157,000
103,000
91,500
44,100
19,500
18,500
15,400
13,100
8,160
80.5
Nutrients
Total Nitrogen d
Total Phosphorus
Not applicable
Not applicable
16,900,000
214,000
Not applicable
Not applicable
Other
Chlorides
Total dissolved solids
2.435 X 10"5
930,000,000
22,600
Not applicable
Total Pollutants e
2,210,000,000
2,680,000
Sources: Abt, 2008; ERG, 2015a; ERG, 2015b; ERG, 2015f; U.S. EPA, 2012c.
Note: Numbers are rounded to three significant figures.
a - The list of pollutants included in this table is only a subset of pollutants included in the loadings analysis (see
Section 10 of the TDD).
b - TWFs for the following metals apply to all metal compounds: arsenic, chromium, copper, lead, manganese,
mercury, nickel, selenium, thallium, vanadium, and zinc. EPA updated TWFs for arsenic, cadmium, copper,
manganese, mercury, thallium, and vanadium for the steam electric ELGs pollutant loadings analysis.
c - These loadings reflect adjustments to current conditions in the industry. See Section 1 for further details. Data
source for pollutant specific loadings is DCN SE05620.
d - Total nitrogen is the sum of total Kjeldahl nitrogen and nitrate/nitrite as N.
e - The totals represent the pollutant loadings in discharges of the evaluated wastestreams - specifically, FGD
wastewater, fly ash transport wastewater, bottom ash transport wastewater, and combustion residual leachate (see
Section 10 of the TDD). Loadings presented are based on the final loadings analysis presented in the TDD. The
totals exclude loadings for pollutants not identified as POCs and for biochemical oxygen demand (BOD), chemical
oxygen demand (COD), total organic carbon (TOC), total dissolved solids (TDS), and total suspended solids (TSS).
3-14
-------
Section 3—Environmental And Human Health Concerns
3.2.2 Comparison of Steam Electric Power Plant Loadings to Other Industries
The total TWPE discharges from the steam electric power generating industry are higher
than the TWPEs estimated for many other industries. As part of the Preliminary 2010 Effluent
Guidelines Program Plan published on October 30, 2009 (74 FR 68599), EPA identified 10 point
source categories, out of 56, that represented the bulk of the estimated toxic wastewater
discharges (as measured by TWPE) from existing industrial point source categories. EPA ranked
each point source category by the amount of toxic pollutants in its discharges and identified the
Steam Electric Power Generating Point Source Category (40 CFR 423) as the category with the
highest TWPE. Table 3-3 presents the total TWPE estimated as part of the 2010 Effluent
Guidelines Planning Process for the remaining nine point source categories with the highest
TWPE [U.S. EPA, 201 Id]. The TWPE estimated for the 2010 Effluent Guidelines Planning
Process includes pollutant loadings estimated from discharge monitoring reports (DMRs) and
Toxic Release Inventory (TRI) reporting. Therefore, the industry totals may include double-
counting of certain chemical discharges (i.e., a facility must report a chemical on both its DMR
and its TRI reporting form).
Table 3-3. Pollutant Loadings for the Final 2010 Effluent Guidelines Planning Process:
Top 10 Point Source Categories
40 CFR Part
423
430
419
421
418
414
440
415
444
410
Point Source Category
Steam Electric Power Generating
Pulp, Paper, And Paperboard
Petroleum Refining
Nonferrous Metals Manufacturing
Fertilizer Manufacturing
Organic Chemicals, Plastics, And Synthetic Fibers
Ore Mining And Dressing
Inorganic Chemicals Manufacturing
Waste Combustors
Textile Mills
Total TWPE3
(Ib-eq/yr)
2,680000 b
1,030,000
1,030,000
994,000
826,000
649,000
448,000
299,000
254,000
250,000
Source: U.S. EPA, 201 Id.
a - Only TWPE totals for the steam electric power generating industry include updates to TWFs for arsenic,
cadmium, copper, manganese, mercury, thallium, and vanadium. The TWPE for all other point source categories is
estimated from DMRs and TRI reporting and may include double-counting of certain pollutant discharges (/'. e., a
facility must report a pollutant on both its DMR and its TRI reporting form). Loadings are rounded to three
significant figures.
b -EPA calculated the steam electric power generating industry (40 CFR 423) discharges for the final rule as total
2,680,000 TWPE annually (see Section 10 of the TDD). These loadings reflect adjustments to current conditions in
the industry. See Section 1 for further details.
EPA estimated that the total baseline TWPE from steam electric power plant wastewater
is almost three times the amount estimated for the pulp, paper, and paperboard industry,
petroleum refining industry, and nonferrous metals manufacturing (second, third, and fourth
highest ranking), and it is over five times the TWPE for four of the six other industries identified
as the top TWPE dischargers in the Final 2010 Effluent Guidelines Program Plan [U.S. EPA,
3-15
-------
Section 3—Environmental And Human Health Concerns
20lid].14 This suggests that the loadings from the subset of evaluated wastestreams represent a
greater environmental concern within the context of all industrial dischargers across the United
States.
3.2.3 Comparison of Steam Electric Power Plant Loadings to Publicly Owned Treatment
Works
To provide additional perspective on the magnitude of the pollutant loadings from steam
electric power plants, EPA compared loadings for the evaluated wastestreams to those of an
average publicly owned treatment works (POTW). EPA selected POTWs for comparison
because, for point sources, POTWs and steam electric power plants dwarf all other point source
discharges in terms of total TWPE of metals discharged to waters in the United States [U.S.
EPA, 2010c].15 In addition, the more than 16,000 POTWs are located across the United States
and provide a common metric to use for point source evaluations.
EPA calculated the average pollutant loadings discharged from a typical POTW using
EPA's Effluent Guidelines Program Plan DMR database, DMRLoadsAnalysis2009_v02.mdb.
EPA assumed that a typical POTW discharges wastewater at a rate of 3 to 5 million gallons per
day (MOD)16 based on the number of facilities by discharge flow rate reported in Metcalf and
Eddy, 2003 [ERG, 2015a]. EPA developed queries in the DMRLoadsAnalysis2009_v02.mdb to
do the following: 1) select POTWs that discharge between 3 and 5 MGD, and 2) calculate the
average DMR loadings (in pounds and TWPE per year) for each pollutant [ERG, 2015a]. Table
3-4 compares the average steam electric pollutant loadings by wastestream17 to the pollutant
14 Data sources for the other industry discharges include DMRs and TRI reports. EPA recognizes that the DMR and
TRI data have limitations (e.g., only a subset of facilities and a subset of pollutants might be included in the
estimated loadings); however, these are the most readily available data sets that represent discharges across the
United States.
15 Based on metal loadings (total TWPE) calculated by EPA's DMR Pollutant Loading Tool, 2010 data, by Standard
Industrial Classification (SIC) code. The top two industries are SIC 4952 - Sewerage Systems (i.e., POTWs) and
SIC 4911 - Electrical Services. EPA's DMR Pollutant Loading Tool is an online tool (http://cfpub.epa.gov/dmrA
that calculates pollutant loadings from permit and DMR data from EPA's Permit Compliance System (PCS) and
Integrated Compliance Information System for the National Pollutant Discharge Elimination System (ICIS-
NPDES). The tool also ranks dischargers, industries, and watersheds based on pollutant mass and toxicity, and
presents "top 10" lists to help users determine which facilities and industries are producing these discharges and
which watersheds are impacted. Facilities report pollutant discharge monitoring data in their DMR as mass-based
quantities (e.g., pounds per day) and/or concentrations (e.g., mg/L). The DMR Pollutant Loading Tool allows users
to gather annual loadings data. For this EA, EPA reviewed the 2010 loadings reported in DMRs.
The use of the DMR data has its limitations. Only pollutants included in the facility's NPDES permit are included in
the PCS and ICIS-NPDES databases; therefore, if a facility does not have mercury limitations, mercury discharges
from that facility will not be included in the total for industrial discharges. States (or other permitting authority) have
some discretion as to which data they make available (or enter) to PCS and ICIS-NPDES. For example, permitting
authorities enter DMR and permit information for facilities that are considered major dischargers. However, they do
not necessarily enter DMR or permit information into PCS for minor dischargers or facilities covered by a general
permit.
16 For comparison, the average discharge flow rates for the evaluated wastestreams are 0.45 MGD for FGD
wastewater; 3.5 MGD for fly ash transport water; 2.1 MGD for bottom ash transport water; and 0.08-0.09 MGD for
leachate [see Section 6 of the TDD].
17 EPA calculated the average pollutant loadings for each wastestream by dividing the total pollutant loadings for the
wastestream by the number of steam electric power plants discharging the wastestream [ERG, 2015a].
-------
Section 3—Environmental And Human Health Concerns
loadings from an average POTW assumed to discharge 3 to 5 MOD. The results of the analysis
demonstrate the following:
• Average FGD wastewater discharges contain over 200 times more boron and
manganese, over 75 times more selenium, and approximately 20 times more cadmium
and nickel than average POTW discharges.
• Average fly ash transport water discharges contain over 10 times more boron,
cadmium and thallium and over five times more arsenic, nickel, and selenium than
average POTW discharges.
• Average bottom ash transport water discharges contain 30 times more thallium;
approximately 10 times more manganese and nickel; and five times more cadmium
than average POTW discharges.
• Average combustion residual leachate wastewater discharges contain more boron,
iron, manganese, and selenium than average POTW discharges.
Nutrient loadings (total nitrogen and
total phosphorus) from the average steam
electric wastestreams are generally lower than
the nutrient loadings from an average POTW.
Total nitrogen loadings from an average FGD
wastestream are approximately equal to those
of an average POTW. Nitrogen loadings from
average fly ash and bottom ash transport
waters are less than the total nitrogen
discharges from an average POTW
(approximately 20 percent). The amount of
total phosphorus discharged by an average
POTW is over 20 times higher than that in the
average fly ash transport water, bottom ash
transport water discharges, and FGD
wastewater. EPA did not calculate nutrient
loadings for combustion residual leachate.
Loadings of the Evaluated Wastestreams
Compared to POTWs
FGD wastewater discharges contain:
- 200 times more manganese
- 200 times more boron
- 75 times more selenium
- 20 times more nickel
- 20 times more cadmium
Bottom ash transport water discharges
contain 30 times more thallium and 10
times more manganese and nickel.
Fly ash transport water discharges contain
five times more arsenic, nickel, and
selenium and 10 times more boron,
cadmium, and thallium.
• Combustion residual leachate contains over
four times more boron and iron.
For chlorides, EPA found that average
FGD wastewater discharges contain
approximately six times greater chlorides
loadings than an average POTW discharge. The average discharges of fly ash transport water,
bottom ash transport water, and combustion residual leachate from a steam electric power plant
contain less chlorides than a typical POTW discharge (less than 10 percent). EPA's DMR data
did not include pollutant loadings for TDS from POTWs; therefore, EPA could not compare
these pollutant loadings between steam electric and POTW discharges.
3-17
-------
Section 3—Environmental And Human Health Concerns
Table 3-4. Comparison of Average Pollutant Loadings in the Evaluated Wastestreams to an Average POTW
Pollutant
Aluminum
Arsenic
Boron
Cadmium
Chromium VI
Copper
Iron
Lead
Manganese
Mercury
Nickel
Selenium
Thallium
Vanadium
Zinc
Total Nitrogen
Total
Phosphorus
Chlorides
TDS
Average Plant FGD
Wastewater Discharge a'b
Loadings
(Ibs/yr)
1,530
9.54
334,000
81.2
(g)
17.9
1,150
5.71
74,500
5.50
620
1,410
16.7
20.8
983
128,000
457
10,200,000
40,400,000
TWPE
(Ib-eq/yr)
99.1
33.1
2,790
1,850
(£)
11.1
6.42
12.8
7,650
605
67.6
1,580
47.7
5.82
46.1
—
_
248
-
Average Plant Fly Ash
Transport Water
Discharge a'c
Loadings
(Ibs/yr)
8,490
312
17,900
47.7
2.62
263
5,140
152
486
7.85
180
134
137
220
734
23,400
864
83,500
1,760,000
TWPE
(Ib-eq/yr)
549
1,080
149
1,090
1.35
164
28.8
340
49.9
864
19.6
150
392
61.7
34.4
—
_
2.03
-
Average Plant Bottom
Ash Transport Water
Discharge a'd
Loadings
(Ibs/yr)
4,240
66.5
2,190
19.1
0.136
89.0
7,610
63.4
4,770
3.19
301
32.4
302
11.4
247
24,600
715
96,700
2,560,000
TWPE
(Ib-eq/yr)
274
231
18.3
435
0.070
55.5
42.6
142
490
351
32.7
36.3
863
3.21
11.6
—
_
2.35
-
Average Plant
Combustion Residual
Leachate Discharge a'e
Loadings
(Ibs/yr)
837
10.8
6,530
2.87
(g)
2.16
10,400
(g)
790
0.298
13.1
31.2
0.338
538
59.1
(g)
(g)
120,000
1,020,000
TWPE
(Ib-eq/yr)
54.1
37.5
54.5
65.3
(g)
1.34
58.4
(g)
81.1
32.8
1.43
35.0
0.964
151
2.77
—
_
2.93
~
Average POTW
Discharge a'f
Loadings
(Ibs/yr)
3,590
45.9
1,540
3.54
17.7
154
2,530
48.5
354
3,180
30.6
18.5
9.94
No data
453
123,000
17,800
1,610,000
No data
TWPE
(Ib-eq/yr)
215
159
12.8
80.6
9.02
95.3
14.2
109
36.1
350,000
3.06
20.7
28.2
No data
18.1
—
_
39.3
~
Note: Numbers are rounded to three significant figures.
a - TWPE presented in the table include updates to TWFs for arsenic, cadmium, copper, manganese, mercury, thallium, and vanadium.
b - Average loadings based on 88 plants assumed to discharge FGD wastewater under baseline conditions [ERG, 2015a].
c - Average loadings based on 50 plants assumed to discharge fly ash transport water under baseline conditions [ERG, 2015a].
d-Average loadings based on 183 plants assumed to discharge bottom ash transport water under baseline conditions [ERG, 2015a].
e - Average loadings based on 95 plants assumed to discharge combustion residual leachate under baseline conditions [ERG, 2015a].
f - Average loadings based on average loadings calculated for POTWs discharging 3 to 5 MOD of wastewater (see DCN SE01961).
g - EPA did not calculate loadings for this pollutant and wastestream. See the Costs and Loads Report (DCN SE05831).
3-18
-------
Section 3—Environmental And Human Health Concerns
To provide additional perspective on the magnitude of the loadings, EPA calculated the
equivalent number of typical POTWs that would discharge loadings equal to the 202 steam
electric power plants18 included in the baseline loadings analysis. Table 3-5 presents total
pollutant loadings for the evaluated wastestreams (for the 202 plants) and the number of typical
POTWs that would discharge equivalent loadings. The results demonstrate that the magnitude of
the total loadings from 202 steam electric power plants is equivalent to a significantly larger
number of typical POTWs for many of the pollutants commonly known to cause environmental
harm. For example, EPA estimated that the total loadings in discharges of the evaluated
wastestreams from these 202 plants are equivalent to approximately 20,000 POTW discharges of
boron and manganese; over 7,500 POTW discharges of selenium; over 6,000 POTW discharges
of thallium; over 3,500 POTW discharges of cadmium and nickel; over 1,000 POTW discharges
of iron; and over 500 POTW discharges of arsenic and chlorides. This suggests that, for the
evaluated wastestreams, 202 steam electric power plants contribute substantial pollutant loadings
to the environment.
Table 3-5. Estimated Number of POTW Equivalents for Total Pollutant Loadings from the
Evaluated Wastestreams
Pollutant
Aluminum
Arsenic
Boron
Cadmium
Chromium VI
Copper
Iron
Lead
Manganese
Mercury
Nickel
Selenium
Thallium
Vanadium
Zinc
Total Nitrogen
Total Phosphorus
Chlorides
IDS
Annual Discharge
pounds (Ibs)
1,410,000
29,600
31,300,000
13,300
156
31,200
2,740,000
19,700
7,530,000
1,490
120,000
140,000
63,700
66,000
174,000
16,900,000
214,000
930,000,000
4,210,000,000
Equivalent Number of Average
POTWs a
394
646
20,300
3,760
8.81
203
1,080
406
21,300
<1
3,920
7,560
6,410
No values for comparison
384
138
12.0
578
No values for comparison
Source: ERG, 2015a.
Note: Numbers are rounded to three
a - Equivalent number of POTWs is
electric power plants by the average
significant figures.
estimated by dividing the total annual pollutant loadings from the 202 steam
POTW loadings presented in Table 3-4 for a 4-MGD POTW.
18 The count of 202 steam electric power plants includes seven indirect dischargers that discharge wastewater to a
POTW and do not discharge any of the evaluated wastestreams directly to surface waters. EPA included these
indirect dischargers to protect confidential business information.
3-19
-------
Section 3—Environmental And Human Health Concerns
3.3 ENVIRONMENTAL IMPACTS FROM STEAM ELECTRIC POWER PLANT WASTEWATER
EPA identified environmental impacts from EPA's assessment of damage cases and
literature sources ("other documented site impacts") caused by steam electric power plant
wastewater and combustion residuals. EPA found over 150 steam electric power plants causing
environmental impacts to surface water and ground water environments following exposure to
steam electric power plant wastewater. Impacts identified in the damage cases and other
documented site impacts include lethal and sublethal impacts on fish, impacts on the diversity
and size of populations in the ecosystem, and impacts on drinking water quality. While these
impacted sites are often assumed to be anomalies, mounting evidence indicates that the
characteristics contributing to the documented impact (e.g., magnitude of the pollutant loadings,
type of pollutant present, plant operations, and wastewater handling techniques) are common
among steam electric power plant receiving water locations [Cherry et al, 2000; NRC, 2006;
Rowe etal., 2002].
Section 3.3.1 presents a qualitative discussion of the lethal and sublethal ecological
effects of pollutants in steam electric power plant wastewater. Section 3.3.2 summarizes
documented instances where steam electric power plant wastewater discharges have caused fish
advisories or exceeded MCLs presenting a potential human health concern. Section 3.3.3 and
Section 3.3.4 summarize the damage cases and other documented site impacts to surface water
and ground water, respectively. Section 3.3.5 discusses the potential for these environmental
impacts to occur at other locations.
3.3.1 Ecological Impacts
Documented ecological impacts associated with exposure to steam electric power plant
wastewater include acute effects (e.g., fish kills) and chronic effects (e.g., malformations, and
metabolic, hormonal, and behavioral disorders) upon biota within the receiving water and
surrounding environment. Effects have included reduced growth and reduced survival of aquatic
organisms and changes to the local habitat [Carlson and Adriano, 1993; Rowe et al., 2002].
This section provides examples of the lethal and sublethal effects on organisms exposed
to steam electric power plant wastewater pollutants (e.g., arsenic, cadmium, chromium, copper,
mercury, and selenium) in surface waters and sediment. Scientific studies reported in the
literature included:
• Field studies in which organisms collected from known contaminated sites were
compared to those collected from uncontaminated sites.
• Laboratory experiments in which organisms intentionally exposed to steam electric
power plant wastewater were compared to those unexposed.
Many of the scientific studies documented in the literature focused on selenium as a key
pollutant of environmental concern within steam electric power plant wastewater. However, due
to the complex nature of the wastewater, many studies evaluated the environmental effects of
metals in steam electric power plant wastewater in aggregate.
3-20
-------
Section 3—Environmental And Human Health Concerns
Lethal and Sublethal Effects of Selenium
Selenium can bioaccumulate to toxic levels in organisms inhabiting environments with
low selenium concentrations. For example, Lemly conducted a field study that investigated the
patterns of selenium biomagnification and toxicity in aquatic organisms inhabiting a cooling
water reservoir that received effluent from a power plant's surface impoundment [Lemly,
1985a]. Throughout the study, selenium concentrations in the reservoir averaged 10 |ig/L;
however, Lemly reported that fish tissue concentrations reached levels ranging from 500 to 4,000
times the average reservoir water selenium concentration. The results of the study indicated that
the extent of selenium bioaccumulation depended on the trophic level of the fish present in the
reservoir. Lemly observed that the selenium accumulation increased as the trophic level
increased, which potentially correlated with the observed elimination of multiple higher-tropic-
level fish species. Therefore, these findings suggest that—even at low concentration within a
surface water—selenium can accumulate and biomagnify to toxic levels in aquatic organisms
and pose a lethal threat to fish at the top of the trophic structure [Lemly, 1985a]. Predicting the
impacts of selenium in aquatic ecosystems can be particularly challenging, because impacts to
the ecosystem cannot be determined solely on the selenium concentration in the receiving water
as demonstrated in this study.
Selenium discharges also impact species diversity in receiving waters. In 1977, two years
after the initial operation of the Belews Creek Steam Station in North Carolina, the fish
community inhabiting the plant's cooling water reservoir (a lake) underwent rapid decline, and
species diversity drastically altered [Lemly, 1985a]. Lemly observed that 17 of the 20 fish
species originally present in the lake were eliminated after the power plant began operation,
including all game species (temperate perch [Percichthyidae], true perch and pike perch
[Percidae], and sunfish [Centrarchidae}). Lemly reported significant levels of selenium
accumulation in the eliminated species and statistically unchanged levels of selenium
accumulation in the surviving species, relative to levels before the power plant began operation.
Only three species maintained reproducing populations in the reservoir: one native species
(mosquitofish) and two introduced non-native species of minnows (fathead minnows and red
shiners) [Lemly, 1985a].
A number of scientific studies express concern over selenium exposure within lakes and
reservoirs where longer residence times allow for further bioaccumulation and a greater potential
to reach lethal concentrations. This is demonstrated by a series of major fish kills that occurred in
1978 and 1979 at Martin Creek Lake (Texas) due to the elevated concentrations of selenium in
the water and fish tissue [U.S. EPA, 2014b]. In particular, studies concluded that elevated
selenium concentrations were likely the primary contributor to fish kills in lakes and reservoirs,
decreasing population density and community diversity [Coughlan and Velte, 1989; Crutchfield,
2000b; Crutchfield and Ferguson, 2000a; Cumbie and Van Horn, 1978].
The sublethal effects of selenium vary widely and can impact growth, reproduction, and
survival of susceptible organisms. Scientists have demonstrated that various fish and amphibian
species are sensitive to elevated selenium concentrations such as those found in steam electric
power plant wastewater. In addition to lethal effects described above, these fish and amphibian
species have developed sublethal symptoms such as accumulation of selenium in tissue
(histopathological effects) and in the blood (hematological effects), resulting in decreased
3-21
-------
Section 3—Environmental And Human Health Concerns
growth, changes in weight, abnormal morphology, and reduced hatching success [Coughlan and
Velte, 1989; Lemly, 1993; Sager and Colfield, 1984; Sorensen, 1988; Sorensen and Bauer,
1984a; Sorensen etal, 1982, 1983, 1984b].
The literature indicates that the extent of selenium accumulation in fish tissue varies by
species, and selenium accumulates most significantly in the liver and reproductive tissues in
most species [Baumann and Gillespie, 1986; Sager and Colfield, 1984; Sorensen, 1988]. Other
studies have reported accumulation in the skeletal muscle, kidneys, gills, and hearts of fish,
resulting in pathological lesions, morphological changes, increased organ weight, and decreased
growth [Coughlan and Velte, 1989; Lemly, 2002; Sorensen and Bauer, 1984b]. Aquatic
organisms exposed to steam electric power plant wastewater have exhibited elevated selenium
concentrations in organs such as kidneys, liver, and gonads, resulting in abnormalities that hinder
growth and survival [Rowe et al, 2002].
In addition, selenium is highly teratogenic (i.e.., able to disturb the growth and
development of an embryo or fetus) and readily transferable from mother to egg [Chapman etal.,
2009; Janz et al, 2010; Lemly, 1997b; Maier and Knight, 1994]. Selenium is known to
bioaccumulate in the reproductive organs of fish and amphibian species. In one study, ovarian
selenium concentrations in bluegill fish were observed at levels 1,000 times greater than the
surrounding surface water [Baumann and Gillespie, 1986]. Multiple studies have documented
reproductive failure or diminished reproductive success in both fish and amphibians inhabiting
ponds, lakes, and reservoirs contaminated with selenium from steam electric power plant
wastewater discharges [Baumann and Gillespie, 1986; Crutchfield, 2000b; Cumbie and Van
Horn, 1978; Gillepsie et al, 1986; Hopkins et al, 2002; Nagle et al, 2001]. For example,
Hopkins et al. [2006] observed reduced hatching success, abnormal swimming, and
abnormalities in the face and skull in the offspring of selenium-contaminated female toads. Field
and captive feeding studies also show reproductive impairment (reduced hatchability of eggs)
among waterfowl exposed to elevated levels of selenium [Adams et al., 2003; Ohlendorf, 2003
and 2007; Beckon etal, 2008; U.S. DOI, 1998; Smith etal, 1998].
Histopathological effects (i.e., observable changes in tissue), increased metabolic rate,
and decreased growth rates are effects typically caused by contamination from steam electric
power plant wastewater. Water and fish samples collected before and after the discharge of
power plant wastewater from the surface impoundment to the Texas Utilities Martin Creek Lake
found that selenium concentrations were significantly elevated in the reservoir and in fish livers,
kidneys, and gonads. In 1984, Garrett and Inman reported that elevated selenium concentrations
persisted in the livers and kidneys of several species of fish for up to 3 years after the power
plant wastewater discharges ceased. Additionally, a 1988 study by Sorensen found that red ear
sunfish native to the reservoir exhibited ovary abnormalities related to elevated selenium
concentrations up to 8 years following an 8-month exposure to power plant wastewater
discharges. Although the surface impoundment discharge was short-lived, many of the
histopathological effects persisted for years after the discharge had ceased [Rowe et al, 2002].
These sublethal effects of selenium, while not directly resulting in the mortality of
exposed aquatic wildlife, can ultimately cause the types of population-level impacts described
under lethal impacts above. The available scientific evidence indicates that reproductive
success—specifically, offspring mortality and severe development abnormalities that affect the
3-22
-------
Section 3—Environmental And Human Health Concerns
ability of fish to swim, feed, and successfully avoid predation—is the critical assessment
endpoint when evaluating the potential for selenium exposure to result in population-level
impacts to resident fish species.
For a summary of the impacts of selenium on surface water, refer to Table A-10 in
Appendix A.
Lethal Effects of Other Pollutants
Scientific studies have confirmed that both acute and chronic exposure to pollutants in
steam electric power plant wastewater can be lethal to a wide range of aquatic organisms. For
example, Guthrie and Cherry [1976] found that shrimp darters and salamanders were highly
sensitive to acute exposures of steam electric power plant wastewater and experienced nearly
100 percent mortality following a five-day exposure to power plant wastewater discharges.
Invertebrates and fish also evaluated in the study were less sensitive to the acute exposure to
power plant wastewater and reported lower rates of mortality [Guthrie and Cherry, 1976].
Chronic exposures to power plant wastewater are also of concern; however, studies show
extreme differences in species sensitivity [Rowe et al., 2002]. For example, juvenile chubsuckers
(a benthic fish) exposed for 45 days to sediments, water, and food contaminated with power
plant wastewater experienced a 75 percent mortality rate [Hopkins et al, 2001]. In another study,
bullfrogs exposed to sediment and water from a combustion residual surface impoundment for
34 days demonstrated an 87 percent mortality rate (which was 41 percent greater than the
mortality rate of bullfrogs included in control group) [Rowe et al, 2002]. A third study reported
no lethal effects for banded snakes exposed for 2 years to fish collected from combustion
residual surface impoundments [Hopkins et al, 2002].
Other studies examined lethal effects of sediments contaminated with combustion
residuals. For example, eggs and hatchlings of fish and reptiles raised in contaminated sediment
reported higher mortality rates (16 to 94 percent) than eggs and hatchlings from control groups
[Hopkins et al, 2000; Nagle et al, 2001; Roe et al, 2006; Rowe et al, 1998a, 1998b, 2001;
Snodgrass et al., 2004]. Each of the studies observed elevated mortality rates in conjunction with
higher concentrations of steam electric power plant wastewater pollutants (e.g., arsenic,
cadmium, chromium, copper, selenium) in the exposed sediment.
Three studies evaluated the lethal effects of specific pollutants in steam electric power
plant wastewater on a variety of organisms (i.e., insects, fish, and amphibians) and determined
the median lethal concentration (LCso) for each pollutant-organism combination. LCso is the
concentration expected to be lethal to 50 percent of a group of organisms exposed for a given
time duration. Table 3-6 summarizes the results from the three experiments and Table 3-7
presents the LCso concentrations reported in the studies. Overall, the LCso studies report species-
specific differences, particularly among species living downstream of fly ash surface
impoundment discharges. The downstream species developed resistance to pollutants compared
to those living in unpolluted ponds. Because the LCso concentrations were much higher than
actual aquatic concentrations, there was no evidence in these experiments of acute lethal effects,
though long-term (1 to 3 months) lethal effects could not be ruled out [Benson and Birge, 1985;
Birge, 1978; SpechtetaL, 1984].
3-23
-------
Section 3—Environmental And Human Health Concerns
Sublethal Effects of Other Pollutants
Although the majority of sublethal effects documented in the literature primarily focus on
selenium concentrations in steam electric power plant wastewater, several studies discussed the
sublethal effects of other pollutants, such as arsenic, cadmium, chromium, copper, and lead
[Rowe et a/., 2002]. Sublethal effects from exposure to pollutants other than selenium in power
plant wastewater can include changes to morphology (e.g., fin erosion, oral deformities),
behavior (e.g., swimming ability, ability to catch prey, ability to escape from predators), and
metabolism that can negatively affect long-term survival. For example, a study of larval
bullfrogs living in combustion residual surface impoundments found that more than 95 percent
of individuals had abnormal oral structures, such as the absence of grazing teeth or entire rows of
teeth, which altered feeding habits and subsequently reduced growth rates in the affected
bullfrogs [Rowe etal, 1996]. In another study, tail malformations in larval bullfrogs attributed to
power plant wastewater exposure caused abnormal swimming behavior, and the affected
bullfrogs were preyed upon more frequently than bullfrogs from unpolluted sites [Raimondo et
al, 1998].
Several studies have demonstrated increased metabolic rates and decreased growth rates
in aquatic organisms exposed to steam electric power plant wastewater. Increased metabolism
causes organisms to waste energy during normal metabolic processes, which can affect growth.
In a 1998 study by Rowe, grass shrimp caged in a surface impoundment for eight months
experienced a 51 percent increase in standard metabolic rate. Similarly, crayfish captured near
the impoundment experienced increased metabolic rates and decreased growth rates—effects that
were also observed in crayfish collected from unpolluted sites and exposed to contaminated
sediments from the combustion residual surface impoundment [Rowe et al., 2002].
3-24
-------
Section 3—Environmental And Human Health Concerns
Table 3-6. Summary of Studies Evaluating Lethal Effects of
Pollutants in Steam Electric Power Plant Wastewater
Citation
Birge,
1978
Benson
and Birge,
1985
Specht et
al., 1984
Studied Organism
Eggs from goldfish, trout,
and toads
Minnows (fish) living in fly
ash-polluted ponds in
Kentucky compared to those
living in uncontaminated
ponds
Insects (coleopterans,
mayflies, and other insects)
exposed to fly ash surface
impoundment effluent from
the Appalachian Power
Plant in Giles County,
Virginia, compared to those
living in an uncontaminated
pond
Test
Performed
7- to 28-day
lethal effects
Acute (96-
hour) toxicity
Acute (96-
hour) toxicity
Trace Elements
Studied
22 elements
Cadmium
Copper
Zinc
Cadmium
Copper
Zinc
Summary of Results
Among the 22 elements tested,
cadmium, chromium, mercury, nickel,
lead, and silver were the most toxic to
all three species, with most LC50 being
0.1 milligrams per liter (mg/L) or less.
The study found a higher tolerance to
cadmium and copper in the exposed
fish compared to the fish from
unpolluted ponds. However, both
exposed and unexposed populations
exhibited similar tolerance to zinc. See
Table 3-7 for LC50 values.
The study observed a higher tolerance
to pollutants in exposed insects
compared to those living in unpolluted
ponds. See Table 3-7 for LC50 values.
3-25
-------
Section 3—Environmental And Human Health Concerns
Table 3-7. Median Lethal Concentrations (LCso) for Pollutants in Steam Electric Power Plant Wastewater
Pollutant
Aluminum
Arsenic
Cadmium
Chromium
Cobalt
Copper
Lead
Mercury
Nickel
Selenium
Silver
Vanadium
Zinc
LCSO, mg/L
7- to 28-Day Exposure
Trout
[Birge, 1978]
0.56
0.54
0.13
0.18
0.47
0.09
0.18
0.005
0.05
4.18
0.01
0.16
1.06
Goldfish
[Birge, 1978]
0.15
0.49
0.17
0.66
0.81
5.2
1.66
0.12
2.14
8.78
0.03
4.6
2.54
Toad
[Birge, 1978]
0.05
0.04
0.04
0.03
0.05
0.04
0.04
0.001
0.05
0.09
0.01
0.25
0.01
96-Hour Exposure
Exposed
Minnows
[Benson and
Birge, 1985]
3.89 a
9.55 b
0.36 a
0.41 b
6.14 a
5.96 b
Control
Minnows
[Benson and
Birge, 1985]
3.06 a
7.16 b
0.21 a
0.39 b
6.09 a
7.45 b
Mayflies
[Specht et a/.,
1984]
0.27
0.18
18.44
Other Insects
[Specht et a/.,
1984]
1
1.2-250
1
0.03-8.3
1
1
1
1
1
18.2
Acronyms: mg/L - milligrams per liter.
Shaded cells indicate that the pollutant was not evaluated.
a - Nominal water hardness of 100 mg/L calcium carbonate (CaCO3).
b - Nominal water hardness of 250 mg/L calcium carbonate (CaCO3).
3-26
-------
Section 3—Environmental And Human Health Concerns
3.3.2 Human Health Effects
Exposure to pollutants can cause non-
cancer effects in humans, including damage to
the circulatory, respiratory, or digestive systems
and neurological and developmental effects.
Steam electric power plant wastewater includes
toxic pollutants and known or suspected
carcinogens (e.g., arsenic and cadmium). In the
literature review, EPA identified potential
human impacts from consuming fish in
contaminated waters and from ingesting
drinking water contaminated by pollutants from
combustion residuals.
19
During the late 1970s, three power plant Numerous damage cases show exceedances of
cooling water reservoirs in Texas received drinking water standards at ground water and
discharges from surface impoundments drinking water wells due to leachatefrom
containing elevated selenium levels, resulting in nearby impoundments and landfills.
a series of fish kills. The reservoirs included Brandy Branch Reservoir, located in Harrison
County; Welsh Reservoir, located in Titus County; and Martin Creek Lake, located in Rusk
County. Investigations at the reservoirs implicated elevated selenium levels in the fish tissue as
the cause. In 1992, the Texas Department of Health issued a fish consumption advisory for the
three reservoirs after determining that the level of selenium in fish could pose a potential health
risk to humans, especially children 6 years or younger and pregnant women.
Ground water and drinking water supplies can be degraded by pollutants in steam electric
power plant wastewater and combustion residual leachate [Cross, 1981]. Combustion residual
leachate can migrate from the site in the ground water at concentrations that could contaminate
public or private drinking water wells and surface waters, even years following disposal of
combustion residuals [NRC, 2006], as exemplified in the following example. The Wisconsin
Electric Power Company (WEPCO) plant in Port Washington, Wisconsin, had disposed of fly
ash in a quarry for over 20 years (1943-1971) at a depth of 40 to 60 feet, with some of the
disposed ash below the water table. The disposal site is located in an upland area where down-
gradient ground water is used as a source of drinking water. The Wisconsin Department of
Natural Resources was notified in January 1980 and November 1990 that elevated levels of
sulfates, selenium, and boron were found in a private drinking water well located 250 feet down-
gradient from the coal-fired power plant waste disposal site. The impacted private well was
replaced with a deeper well to avoid further contamination [U.S. EPA, 2014c].
In this EA, EPA evaluated the threats to human health and the environment associated with pollutants leaching
into ground water from surface impoundments and landfills containing combustion residuals. If these leached
pollutants do not constitute the discharge of a pollutant to surface waters, then they are not controlled under the
steam electric ELGs. While the CCR rulemaking is the major controlling action for these pollutant releases to
ground water, the ELGs could indirectly reduce impacts to ground water. These secondary improvements are
discussed in Section 7.8.
3-27
-------
Section 3—Environmental And Human Health Concerns
As discussed in Section 3.3.4 and Appendix A, there have been documented exceedances
of MCL drinking water standards at off-site ground water and drinking water wells. Exceedances
of MCLs in the ground water indicate potential human health impacts if the pollutants enter
private drinking water wells. Section 3.3.4 outlines three documented instances where
combustion residual leachate contamination caused impacts to private drinking water wells.
Drinking water standards can also be exceeded in surface waters. For example, Duke
Energy's Riverbend Plant discharges surface impoundment effluent into Mountain Island Lake,
which supplies drinking water to 700,000 people. The county detected arsenic and zinc
concentrations above state standards in an area near the surface impoundment discharge pipe
[Charlotte Observer, 2010]. While most of the pollutants in the surface water would likely be
reduced to safe levels during drinking water treatment, elevated levels of pollutants in source
water can impact the effectiveness of drinking water treatment processes and the ability of
drinking water treatment plants to meet MCLs. Section 3.4.6 presents further details on drinking
water resources near steam electric power plants.
3.3.3 Damage Cases and Other Documented Surface Water Impacts
Changes in surface water chemistry due to contamination from steam electric power plant
wastewater can negatively impact all levels of an ecosystem, including lower food chain
organisms, which affect the ecosystem's food web; fish inhabiting the surface water; and wildlife
and humans when they bathe in or drink the water. As described in earlier sections, pollutants in
surface water can accumulate in aquatic organisms such as fish. When wildlife or humans ingest
these aquatic organisms, they can be exposed to a higher dose of contamination than through
direct exposure to the surface water. Documented surface water impacts associated with
discharges of steam electric power plant wastewater include damage to fish populations (i.e.,
physiological and morphological abnormalities and various behavioral, reproductive, and
developmental effects), decreased diversity in insect populations, and decline of aquatic
macroinvertebrate population. Impacts that
affect humans include exceedances of
NRWQC, fish consumption advisories, and
designation of surface waters as impaired
(limiting recreational activities).
EPA's damage case assessment found
26 proven damage case sites and 31 potential
damage case sites with surface water impacts
[U.S. EPA, 2014a through 2014e]. Including
documented site impacts from the literature
review, EPA identified impacts to surface
waters at nearly 70 steam electric power plants
following exposure to wastewater (more than
140 documented site impacts) [ERG, 2015m].
Some of the documented impact sites are the
same locations identified by EPA as damage
case sites. Table 3-8 highlights several damage
case and other documented impact sites where
Some wastewater surface impoundments are
located in, or near, large river floodplains.
Failure of the embankments of surface
impoundments can release catastrophic
amounts of pollutants into surrounding
ecosystems.
3-28
-------
Section 3—Environmental And Human Health Concerns
negative surface water impacts from steam electric power plant wastewater discharges have been
studied. In most cases, negative impacts have been studied and documented in multiple articles
and reports. Tables A-6 and A-7 in Appendix A summarize the damage cases from combustion
residual surface impoundments and landfills, respectively.
3-29
-------
Section 3—Environmental And Human Health Concerns
Table 3-8. Summary of Select Sites with Documented Surface Water Impacts from Steam Electric Power Plant Wastewater
Site Name and
Location
Number of
Documents that
Discuss Surface
Water Impacts
at the Site
EPA
Damage
Case
Assessment
Summary of Surface Water Impacts
Belews Lake,
NC
13
Proven
damage case
[U.S. EPA,
2014b]
In 1970, Duke Power Company constructed Belews Lake as a cooling water reservoir to support the Belews Creek
Steam Station. Almost immediately after surface impoundment effluent began discharging into the lake, fish
populations experienced morphological changes, reproductive failure, and eventually death. In 1985, the Belews
Creek Steam Station converted to a dry-ash transport system, ending the surface impoundment discharges to the
lake. However, even 11 years after the discharges ceased, reproductive abnormalities persisted in the fish
populations. Due to selenium concentrations, 16 of the 20 populations originally present in the reservoir were
entirely eliminated, including all primary sport fish [Lemly, 1997a; U.S. EPA, 2014b].
Brandy Branch
Reservoir, TX
Proven
damage case
[U.S. EPA,
2014b]
Brandy Branch Reservoir serves as a cooling water reservoir for Pirkey Power Plant. From 1986 to 1989, the Texas
Parks and Wildlife Department's) reported increases in the selenium concentrations of the fish inhabiting the
receiving water. As a result, the Texas Department of Health issued a fish consumption advisory for the reservoir,
because of the potential health impact due to the levels of selenium in fish. Since the fish kills in the 1980s,
Southwestern Electric Power Company has worked cooperatively to monitor fish tissue selenium concentrations,
which have decreased since the late 1980s [ATSDR, 1998a].
Euharlee Creek,
GA
Proven
damage case
[U.S. EPA,
2014b]
On July 28, 2002, a sinkhole developed in the surface impoundment at the Georgia Power Company in Cartersville,
GA. The sinkhole expanded to 4 acres, and an estimated 2.25 million gallons of ash/water mixture was released to a
tributary of the Euharlee Creek. Approximately 80 tons of ash entered Euharlee Creek through a stormwater drainage
pipe. This discharge deposited an ash blanket in the creek up to 8 inches deep over 1,850 square feet of the stream
bottom. Sampling at the ash discharge site found that concentrations of certain metals (arsenic, cadmium, chromium,
copper, lead, mercury, and nickel) exceeded EPA Region IV ecological sediment screening values (ESV'S)
indicating a potential for adverse impacts to aquatic life. Sediment concentrations of arsenic measured 14 ppm dry
weight-over five times the toxic threshold. Biological sampling indicated that benthic organisms in the tributary and
ash deposition zone of Euharlee Creek were either killed by contaminants or physically smothered. The resident fish
community, which consisted of at least 25 species, was displaced due to the irritation of high turbidity in the ash
plume as it moved through during the spill. One month after the spill, concentrations of selenium and cadmium were
elevated in crayfish, clams, mollusks, and insects at a Euharlee Creek site downstream from the ash deposit.
3-30
-------
Section 3—Environmental And Human Health Concerns
Table 3-8. Summary of Select Sites with Documented Surface Water Impacts from Steam Electric Power Plant Wastewater
Site Name and
Location
Number of
Documents that
Discuss Surface
Water Impacts
at the Site
EPA
Damage
Case
Assessment
Summary of Surface Water Impacts
Gibson Lake,
IN
Proven
damage case
[U.S. EPA,
2014b]
Gibson Lake is a man-made, shallow impoundment that receives surface impoundment effluent from Gibson
Generating Station. Starting in 1986, least terns, an endangered species of migratory birds, began using the dike in
Gibson Lake as a nesting ground for breeding. To protect the birds from potential toxic exposure, the plant began a
cooperative program with the Indiana Department of Natural Resources to protect the nesting birds by creating a
nearby alternative habitat, known as Cane Ridge Wildlife Management Area (WMA), which received water pumped
from Gibson Lake. In April 2007, Duke Energy closed access to the lake for recreational fishing due to elevated
selenium levels. A year later, the U.S. Fish and Wildlife Service (USFWS) became concerned about selenium levels
in the water and fish in the Cane Ridge WMA. The USFWS stopped the flow of water from Gibson Lake into Cane
Ridge, discouraged least terns from using the refuge, removed the contaminated fish, and plowed Cane Ridge to
redistribute and bury the selenium in the soil. Subsequently, the USFWS stopped the flow of water from Gibson
Lake into Cane Ridge and piped water from Wabash River instead. Cane Ridge was restocked with fish to lure back
migratory birds. As of 2010, fish populations in Gibson Lake still had selenium levels above the toxic threshold
[U.S. EPA, 2014b].
GlenLyn,VA
Proven
damage case
[U.S. EPA,
2014b]
Glen Lyn Plant discharged fly ash transport water from a surface impoundment into Adair Run, a tributary of the
New River. A 1984 study reported that the local insect diversity and density remained essentially the same upstream
(reference site) and downstream of the surface impoundment when the impoundment was not close to capacity.
However, as the settling impoundment reached its capacity, the insect density and diversity declined downstream.
After closure of the surface impoundment, it took up to 10 months for the insect populations to recover [Specht et
al., 1984].
Hyco Lake, NC
Proven
damage case
[U.S. EPA,
2014b]
Hyco Lake is a large cooling water reservoir that received effluent from a power plant, including combustion
residual leachate and fly ash transport water discharges containing high levels of selenium. In 1981, a large-scale fish
kill occurred in the reservoir, prompting numerous scientific studies to examine the extent and cause of the
environmental damage. Multiple studies detected selenium concentrations in the water and tissue offish inhabiting
the reservoir, while other trace elements were within normal concentration ranges. The selenium accumulated in the
fish in the lake, impacting reproduction and causing declines in fish populations in the late 1970s and the 1980s. A
fish consumption advisory was issued in 1988 for this lake due to selenium contamination.
3-31
-------
Section 3—Environmental And Human Health Concerns
Table 3-8. Summary of Select Sites with Documented Surface Water Impacts from Steam Electric Power Plant Wastewater
Site Name and
Location
Number of
Documents that
Discuss Surface
Water Impacts
at the Site
EPA
Damage
Case
Assessment
Summary of Surface Water Impacts
Martin Creek
Lake, TX
Proven
damage case
[U.S. EPA,
2014b]
Martin Creek Lake is a cooling water reservoir that also receives steam electric power plant wastewater discharges.
In 1978 and 1979, a series of major fish kills occurred due to the elevated concentrations of selenium in the water
and fish tissue. Numerous studies conducted throughout the 1980s documented histopathological and reproductive
damage in the fish populations inhabiting the lake. In addition, the studies determined that, even 8 years after
discharge ceased, the overall health of the aquatic populations near the discharge site remained adversely affected by
the selenium pollution. In 1992, a fish consumption advisory was issued for the lake due to discharges from the
steam electric power plant [U.S. EPA, 2014b].
McCoy Branch,
TN
Proven
damage case
[U.S. EPA,
2014b]
In 1986, coal ash slurry discharges from the Department of Energy's (DOE's) Chestnut Ridge Y-12 power plant into
McCoy Branch were found to contain elevated concentrations of trace elements, which violated the Tennessee Water
Quality Act. A 1992 report written by DOE documented bioaccumulation of contaminants in fish tissues, decreased
diversity inbenthic macroinvertebrate communities, and increased fish mortality and abnormalities at the site [U.S.
DOE, 1992].
Mountain
Island Lake,
NC
Location not
assessed
Duke Energy's Riverbend Plant discharges surface impoundment effluent into Mountain Island Lake, which supplies
drinking water to 700,000 people. The county staff has detected arsenic and zinc concentrations above state
standards in an area near the surface impoundment discharge pipe [Charlotte Observer, 2010]. The plant continues
to extensively monitor metal concentrations in Mountain Island Lake surrounding the point of discharge [NCDENR,
2011].
North Carolina
(Multiple
Locations)
Not applicable,
multiple sites
Location not
assessed
A study of receiving waters (including lakes and rivers) for 10 steam electric power plants in North Carolina
evaluated the environmental and ecological impacts that wastewater discharges have on surface waters. The study
found that the receiving waters at the 10 plants contain high levels of contaminants as a result of wastewater
discharges. From the data collected between 2010 and 2012, contaminant levels at multiple surface waters exceeded
drinking water standards and/or NRWQC. For example, arsenic concentrations at two outfalls were as high as 45
ug/L and 92 ug/L, respectively (the drinking water MCL for arsenic is 10 ug/L). When compared to the upstream
pollutant concentrations at the 10 North Carolina locations, data showed elevated levels of contaminants such as
boron, chromium, selenium, bromine, arsenic, and thallium. Elevated pollutant concentrations were also found in
lake sediments (arsenic and selenium) and pore water near lake bottoms (including manganese, arsenic, nickel, and
bromine). The study found elevated levels of arsenic and selenium in fish tissues for two of the lakes (Hyco Lake
and Mayo Lake). A report on fish in Mayo Lake found deformities consistent with ingestion of high selenium levels
[Ruble/a/., 2012].
3-32
-------
Section 3—Environmental And Human Health Concerns
Table 3-8. Summary of Select Sites with Documented Surface Water Impacts from Steam Electric Power Plant Wastewater
Site Name and
Location
Number of
Documents that
Discuss Surface
Water Impacts
at the Site
EPA
Damage
Case
Assessment
Summary of Surface Water Impacts
Rocky Run
Creek, WI
Proven
damage case
[U.S. EPA,
2014b]
Rocky Run Creek, a tributary of the Wisconsin River, receives effluent from Columbia Power Station's surface
impoundments. After the power station began operation in 1975, the aquatic macroinvertebrate populations declined
in the area. Two studies conducted at this site concluded that population density decreased, not because of death due
to coal ash toxicity, but because the aquatic macroinvertebrate populations avoided the area due to sublethal
alterations in the creek. Studies found increased TDS and total suspended solids (TSS), as well as a number of heavy
metals, downstream from the discharge. Some species of macroinvertebrates were totally eliminated 4 months after
discharges began.
Savannah River
Site, SC
23
Proven
damage case
[U.S. EPA,
2014b]
The Savannah River Site, which is owned by DOE, is divided into several areas, based on production, land use, and
other related characteristics. The D-area, a site utilized by numerous ecologists to study the impacts of coal-fired
power plant waste, houses a coal-fired power plant that discharges ash into a series of surface impoundments and a
swamp that ultimately drains into the Savannah River. Numerous studies observed organisms within these habitats
accumulated high concentrations of trace elements in their tissues and exhibited various physiological, behavioral,
and developmental effects. Sediments, water, and biota in the disposal system have elevated concentrations of trace
elements and heavy metals derived from bottom ash and fly ash deposited in the basins. The studies documented
several impacts to amphibians, reptiles, and fish, including five species offish that have been eliminated.
TVA's
Kingston Fossil
Plant, TN
Proven
damage case
[U.S. EPA,
2014b]
On December 22, 2008, the retaining wall of a surface impoundment at TVA's Kingston Fossil Plant broke and
released billions of gallons of coal ash slurry into the Emory, Clinch, and Tennessee Rivers. Tennessee Department
of Environment and Conservation found exceedances of the more stringent criteria for chronic exposure offish and
aquatic life at least once in January 2009 for several metals (e.g., aluminum, cadmium, iron, and lead). Seven months
after the spill, all fish collected had concentrations of selenium above a toxic threshold, and most were still
contaminated at that level 14 months after the spill. Twenty-one months after the spill, a high percentage of fish were
found with lesions, deformities, and infections, all symptoms of extreme stress. In addition, studies have shown
elevated levels of arsenic and mercury in sediments near the ash spill, as well as selenium levels exceeding the MCL
in three wells underneath the Kingston's coal ash disposal area, ash processing area, and gypsum disposal facility
[U.S. EPA, 2014b].
3-33
-------
Section 3—Environmental And Human Health Concerns
Table 3-8. Summary of Select Sites with Documented Surface Water Impacts from Steam Electric Power Plant Wastewater
Site Name and
Location
Welsh
Reservoir, TX
Number of
Documents that
Discuss Surface
Water Impacts
at the Site
2
EPA
Damage
Case
Assessment
Proven
damage case
[U.S. EPA,
2014b]
Summary of Surface Water Impacts
Welsh Reservoir serves as a cooling water reservoir for Welsh Power Plant. From 1986 to 1989, the Texas Park and
Wildlife Department reported increases in the selenium concentrations of the fish inhabiting the receiving water. As
a result, the Texas Department of Health (TDH) issued a fish consumption advisory for the reservoir because of the
potential health impact due to the levels of selenium in fish. In 1998, TDH collected 20 fish for reevaluation and
observed an average selenium concentration in the fish above the reported national averages. Therefore, the Agency
for Toxic Substances and Disease Registry (ATSDR) concluded in a report that there was no clear indication of an
overall change in selenium fish tissue concentrations over the 12 years [ATSDR, 1998b].
Sources: ATSDR, 1998a; ATSDR, 1998b; Charlotte Observer, 2010; ERG, 2013b; Lemly, 1997a; NCDENR, 2011; Ruhl et al, 2012; Specht et al, 1984; U.S. DOE,
1992; U.S. EPA, 2014b.
3-34
-------
Section 3—Environmental And Human Health Concerns
3.3.4 Damage Cases and Other Documented Ground Water Impacts
Pollutants in combustion residuals can leach into ground water from surface
impoundments and landfills at the site. Older surface impoundments and landfills are of
particular concern because they were often built without liners and leachate collection systems.
Liners are typically made of synthetic material, asphalt, clay, or a composite of materials (e.g.,
synthetic and clay) and are designed to collect leachate and prevent ground water contamination.
Combustion residuals held in unlined surface impoundments can enter the subsurface and
contaminate ground water. Pollutants in unlined landfills, used for the dry disposal of
combustion residuals, can also leach as precipitation flows through the residuals pile and
dissolves pollutants; the combustion residual leachate can eventually migrate into ground water.
New plants are increasingly installing liners in surface impoundments and landfills, but
pollutants can also enter the ground water when liners fail or when a disposal site is situated such
that natural ground water fluctuations come into contact with the disposed waste. Furthermore,
state regulation on leachate collection systems and impermeable liners is not uniform [EPRI,
1997; 65 FR 32214-32237, 2000].
Numerous damage cases and other documented site impacts demonstrate the toxic effects
of steam electric power plant wastewater contamination to ground water and the potential to
impact off-site sources due to combustion residual leachate migrating from landfills and surface
impoundments (often unlined). EPA's damage case assessment found 24 proven damage case
sites and 110 potential damage case sites with ground water impacts [U.S. EPA, 2014a through
2014e]. EPA identified impacts to ground water quality caused by combustion residual leachate
from 140 steam electric power plants (more than 130 documented site impacts) [ERG, 2015m].
Some of these documented site impacts are caused by ash contributions from multiple plants
(e.g., a landfill that stores ash from multiple plants). EPA identified some of the documented
impact sites as also being damage case sites. The majority of the damage cases and documented
site impacts reported ground water pollutant levels in on-site wells above regulatory levels;
however, only a portion of the cases indicated off-site contamination. Documented impacts to
off-site ground water resources may be lower due to long migration times within the subsurface
until the combustion residual leachate reaches a known monitoring point [NRC, 2006]. Further,
the limited number of studies documenting off-site contamination might reflect less extensive
monitoring of off-site ground water wells for evidence of impacts from combustion residual
leachate, which suggests off-site impacts may be underrepresented in the documented ground
water impacts [Cherry, 2000].
In surface impoundments, combustion residuals are in constant contact with water,
allowing toxic pollutants to leach into and eventually contaminate ground water. From an
environmental impact perspective, combustion residual surface impoundments are generally
considered less desirable than landfills for disposal because they provide constant saturated or
nearly saturated conditions and a relatively large hydraulic driving force to move combustion
residual leachate into the subsurface [Theis and Gardner, 1990]. Table A-4 in Appendix A
summarizes documented ground water damage cases from combustion residual surface
impoundments [U.S. EPA, 2014a through 2014e].
Although more desirable than surface impoundments, landfills pose their own ground
water contamination risks. If the landfills are not properly lined, the pollutants in combustion
residuals can leach into the soil during precipitation. In areas with acid rain, the precipitation's
3-35
-------
Section 3—Environmental And Human Health Concerns
low pH can accelerate the leaching of contaminants into ground water. In addition, heavy
precipitation can not only accelerate leaching, but also carry pollutants in stormwater runoff,
potentially contaminating ground water or surface water resources [Andersen and Madsen,
1983]. Table A-5 in Appendix A summarizes documented ground water damage cases from
combustion residual landfills [MDNRE, 2010; U.S. EPA, 2014a through 2014e].
While many damage cases document elevated pollutant levels in ground water wells, it is
unclear how many of these are private drinking water wells (as opposed to monitoring wells).
However, the fact that many sites reported MCL exceedances in ground water testing suggests
that potential impacts to drinking water resources are a realistic concern. The following three
damage cases are documented instances where uncollected combustion residual leachate
contaminated ground water and resulted in impacts to private drinking water wells.
Constellation Ash Disposal at Waugh Chapel and Turner Pits - Anne Arundel County,
Maryland
For over a decade, Constellation Energy Group (Constellation) supplied fly ash for
structural fill at the B.B.S.S. Inc. (BBSS) sand and gravel mines in Anne Arundel County,
Maryland. Fly ash from Constellation's Brandon Shores and Wagner plants was used to reclaim
portions of BBSS's Turner Pit starting in 1995 and the Waugh Chapel Pit starting in 2000. In the
fall of 2006, Anne Arundel County Health Department officials documented concentrations of
sulfate and metals (i.e., antimony, beryllium, cadmium, manganese, and nickel) exceeding the
state's screening criteria for potable aquifers in residential wells located downgradient from
Waugh Chapel and Turner Pits [MDNR, 2007].
An independent study of the contamination confirmed that the elevated concentrations of
sulfate and metals observed in the wells directly resulted from precipitation infiltrating the fly
ash deposited in the BBSS sand and gravel mines [MDNR, 2007]. In October 2007, the
Maryland Department of the Environment (MDE) fined Constellation and BBSS $1 million for
the ground water contamination and required the companies to restore the local aquifer water
quality [MDE, 2008]. In addition, Anne Arundel homeowners impacted by the contamination
filed a class action lawsuit against Constellation and were awarded a $45 million settlement. The
settlement required Constellation to pay the costs for converting 84 homes from well water to
public water; cease future deliveries of new coal ash to the quarry; and to establish trust funds to
compensate impacted property owners, enhance the neighborhood, and remediate and restore a
former quarry site [Schultz, 2008].
Gibson Generating Station Plant - Gibson County, Indiana
The Gibson Generating Station Plant has six unlined surface impoundments (four surface
impoundments and two settling/decant basins) and a landfill for combustion residuals. The
landfill consists of a 94-acre older portion built in the late 1970s that is unlined and a 43-acre
portion built in 2002 with a composite liner and leachate collection system. Additionally, the
plant has a 400-acre landfill (South Landfill), permitted in 2005, which also has a composite
liner and leachate collection system.
Samples from monitoring wells downgradient from the older landfill show high levels of
arsenic, boron, iron, and manganese. Leaching from the landfill has contaminated 12 drinking
water wells in the hamlet of East Mount Carmel, Indiana, with boron, manganese, iron, sulfate,
3-36
-------
Section 3—Environmental And Human Health Concerns
sodium, and IDS. Sampling performed by Duke Energy in 2007 and by the Natural Resources
Defense Council in 2008 show drinking water contamination from boron, iron, and manganese in
at least nine off-site private residential wells [U.S. EPA, 2014b].
Ground Water Violations Near North Carolina Power Plants With Surface
Impoundments - North Carolina
The North Carolina Department of Environment and Natural Resources reported ground
water contamination near combustion residual surface impoundments at all 14 of the state's coal-
fired power plants. Duke Energy and Progress Energy each own seven of the plants and perform
ground water monitoring as required by the state. Manganese and lead concentrations exceeded
state ground water standards at all 14 locations and TDS and chromium concentrations exceeded
state standards at seven locations. Boron levels at six plants exceeded state ground water
standards, and some plants had elevated levels of arsenic, selenium, thallium, antimony,
chlorides, and nickel. The state and plants have not identified the source of the contamination but
noted that the exceedances occurred at newly located wells. Drilling the wells may have affected
the concentration of naturally occurring elements such as lead and manganese [Ballard, 2012].20
3.3.5 Potential for Impacts to Occur in Other Locations
Key environmental characteristics that contributed to the impacts documented in Sections
3.3.3 and 3.3.4, such as chronic exposure to large pollutant loadings, plants discharging to waters
with long residence times, and unlined surface impoundments or landfills, are common at steam
electric power plants. This suggests that the impacts documented above indicate the greater
potential threat that steam electric power plant wastewater discharges pose to the environment.
Although substantial events such as fish kills are well documented, the extent to which more
subtle damages, such as histopathological changes, morphological deformities, and damage to
reproductive success, occur elsewhere is not known due to the limited extent of monitoring
programs.
Some of the documented environmental impacts discussed above occurred following
discharges of steam electric power plant wastewater under normal operations. Although the
actual amounts of pollutant loadings discharged may vary among steam electric power plants,
documented site impacts under normal operations do not indicate that the pollutant loadings
associated with the impacts are unusual for steam electric power plants. This suggests that
chronic exposure to typical steam electric power plant wastewater pollutant loadings can impact
the environment at other sites not documented in the literature.
The residence time of steam electric power plant wastewater pollutants in surface water is
a major factor in determining the impact to the environment and the length of the recovery time.
Many documented impact sites are lentic waterbodies such as lakes (i.e., still waters) where
pollutants can reside for long periods of time. These types of surface waters are at particular risk
to impacts from steam electric power plant wastewater discharges. Steam electric power plants
that discharge to a pond, lake, or reservoir may experience similar environmental effects as those
observed in the documented impacts from analogous aquatic systems [ERG, 2015J].
20 EPA notes that the impacts reported at North Carolina plants have not been documented in a peer-reviewed
literature source; however, the information shows that elevated levels of metal contamination can occur near ash
ponds.
3-37
-------
Section 3—Environmental And Human Health Concerns
3.4 DISCHARGE TO SENSITIVE ENVIRONMENTS
The pollutant loadings, ecological impacts, and human health concerns discussed in
Section 3.2 and Section 3.3 are also of concern due to the proximity of many steam electric
power plants to sensitive environments where the characteristics of steam electric power plant
wastewater may impair water quality (e.g., 303(d)-listed waters and waters with fish advisories)
or pose a threat to threatened and endangered species.21 EPA identified the number of surface
waters that receive discharges of the evaluated wastestreams and are located in close proximity
to the following sensitive environments:
• Great Lakes watershed (Section 3.4.1).
• Chesapeake Bay watershed (Section 3.4.2).
• Impaired waters (Section 3.4.3).
• Fish consumption advisory waters (Section 3.4.4).
• Threatened and endangered species habitats (Section 3.4.5).
• Drinking water resources (Section 3.4.6).
Table 3-9 summarizes the number and percentage of immediate receiving waters located
in sensitive environments.
Table 3-9. Number and Percentage of Immediate
Receiving Waters Identified as Sensitive Environments
Sensitive Environment
Great Lakes watershed
Chesapeake Bay watershed
Impaired water
Surface water impaired for a subset of pollutants associated with the
evaluated wastestreams b
Fish consumption advisory water
Surface water with a fish consumption advisory for a subset of
pollutants associated with the evaluated wastestreams °
Drinking water resource within 5 miles
Number (Percentage) of Immediate
Receiving Waters Identified a
25(11%)
13 (6%)
111(50%)
59 (27%)
140 (63%)
93 (42%)
199 (90%)
a - For the sensitive environment proximity analysis, EPA evaluated 222 immediate receiving waters that receive
discharges of the evaluated wastestreams [ERG, 2015c; ERG, 2015d].
b - Table B-l in Appendix B contains a complete list of the impairment categories identified in EPA's 303(d)-
listed waters and designates the subset of pollutants evaluated.
c - Table B-2 in Appendix B contains a complete list of the types of advisories identified under the sensitive
environment proximity analysis, including pollutants that are not associated with the evaluated wastestreams.
3.4.1 Pollutant Loadings to the Great Lakes Watershed
The Great Lakes watershed includes hundreds of tributaries, thousands of smaller lakes,
and extensive mineral deposits. The watershed provides a unique habitat that supports a wide
range of flora and fauna, including over 200 globally rare plants and animals and more than 40
species found only in the Great Lakes watershed. Rare species include the white catspaw pearly
mussel, the copper redhorse fish, and the Kirtland's warbler. The watershed provides a habitat
21 See the ERG memorandum "Proximity Analysis Methodology" (DCN SE04448) for a description of the
methodology used to evaluate the proximity of steam electric power plants to sensitive environments.
3-38
-------
Section 3—Environmental And Human Health Concerns
and food web for an estimated 180 species of native fish, including small- and large-mouth bass,
muskellunge, northern pike, lake herring, whitefish, walleye, and lake trout [Great Lakes
Restoration Initiative, 2010].
The Great Lakes provide humans with transportation, power, and recreational
opportunities including fishing and boating. Between the United States and Canada, the Great
Lakes have more than 10,000 miles of coastline and 30,000 islands. The watershed is home to
more than 30 million people. Recreational spending directly supports 107,000 jobs and nearly
250,000 jobs when secondary impacts are taken into consideration [Great Lakes Restoration
Initiative, 2010].
Environmental impacts documented in the Great Lakes are associated with a range of
stressors, including toxic and nutrient pollutants, invasive species, and habitat degradation. EPA
and Environment Canada have focused their Great Lakes Binational Toxics Strategy on
persistent toxic substances such as mercury [U.S. EPA and Environment Canada, 1997; Great
Lakes Restoration Initiative, 2010]. Mercury is a concern in all of the Great Lakes due to its
bioaccumulation in fish and wildlife and potential impacts on humans. For example, in a study of
65 hair samples from fish-eating and non-fish-eating women, average mercury concentrations in
hair were significantly greater (i.e., 128 to 443
percent higher concentration) for women who ate
several meals of sport-caught fish from the Great Watershed from the Evaluated
Lakes. EPA and Environment Canada have Wastestreams
documented a range of wildlife impacts from , ,, .... , ., . .
• ^i /-> ^ T i i • r • 1.15 million pounds of total nitrogen
mercury in the Great Lakes such as an increase or „ „„ , ,, . ...
,.,.,, ,... • u • 11 FT TO * 9,570 pounds of thallium
physiological abnormalities in herring gulls [U.S. 0 __„ , ,, .
CT/A /-c • m A innm * 8,730 pounds ofzmc
EPA and Environment Canada, 2009].
poundg of selemum
A -w-fiUT-AT-nA 4. A 4. A 4. • 2,170 pounds of arsenic
As part of the EA, EPA wanted to determine ,nnn , r., ,
,, . . r- • * * *u /-> * T i *uj • 1,900 pounds of lead
the extent or impacts to the Great Lakes watershed
Annual Discharges to the Great Lakes
that might be caused by discharges of the evaluated
wastestreams. The primary source of mercury in the Great Lakes watershed is atmospheric
deposition from sources around the Great Lakes watershed (e.g., fuel combustion, incineration,
and manufacturing) emitting approximately 70,000 pounds of mercury annually [Evers et a/.,
2011]. When compared to atmospheric deposition, mercury contributions from point source
discharges are less of a concern. Due to the bioaccumulative nature of mercury, EPA has placed
strict controls (e.g., mixing zones are not allowed in permits) to limit the total amount of mercury
entering the Great Lakes watershed. Monitoring within the Great Lakes watershed has indicated
a decrease in mercury point source discharges, primarily because of implemented control
strategies. EPA identified 23 steam electric power plants discharging to the Great Lakes
watershed with the majority discharging to Lake Michigan (11 plants) and Lake Erie (6 plants)
[ERG, 2015a]. In the Lake Erie Management Plan, EPA identified steam electric discharges as
contributing 57 percent of the mercury to Lake Erie from wastewater sources [U.S. EPA, 2008b].
The potential for bioaccumulative pollutant retention in still or slow-moving water, such
as the Great Lakes, is a particular concern. Many pollutants in steam electric power plant
wastewater can bioaccumulate in fish and then affect higher trophic levels and terrestrial
environments. Table 3-10 presents total pollutant loadings for the evaluated wastestreams
discharging to the Great Lakes watershed.
3-39
-------
Section 3—Environmental And Human Health Concerns
Table 3-10. Pollutant Loadings to the Great Lakes Watershed from the Evaluated
Wastestreams
Pollutant
Arsenic
Boron
Cadmium
Chromium VI
Copper
Lead
Manganese
Mercury
Nickel
Selenium
Thallium
Zinc
Total Nitrogen
Total Phosphorus
Chlorides
Total Dissolved Solids
Annual Discharge to the Great Lakes
Watershed (Ibs)
2,170
997,000
648
0.548
2,550
1,900
242,000
82.8
9,840
5,020
9,570
8,730
1,150,000
23,100
31,900,000
186,000,000
Annual TWPE Discharge to the
Great Lakes Watershed (Ib-eq)
7,510
8,310
14,700
0.283
1,590
4,250
24,900
9,110
1,070
5,630
27,300
409
-
-
778
-
Source: ERG, 2015a.
Note: Numbers are rounded to three significant figures.
3.4.2 Pollutant Loadings to the Chesapeake Bay Watershed
The Chesapeake Bay is the largest estuary in the United States and is a complex
ecosystem that provides habitats and food webs for diverse groups of animals and plants. A
variety of fish either live in the Chesapeake Bay and its tributaries year-round or visit its waters
as they migrate along the East Coast. The Chesapeake Bay Watershed covers 64,000 square
miles, with 11,684 miles of shoreline, and includes areas in six states: Delaware, Maryland, New
York, Pennsylvania, Virginia, and West Virginia, plus Washington, DC. The watershed includes
approximately 284,000 acres of tidal wetlands that provide critical habitats for fish, birds, crabs,
and other species [Chesapeake Bay Program, 2015a and 2015b],
The Chesapeake Bay and its tributaries
provide recreational and commercial opportunities,
with more than 100,000 streams, creeks, and rivers
in the watershed. Fishers commonly catch striped
bass and white perch and seafood production from
the Bay totals approximately 500 million pounds per
year [Chesapeake Bay Program, 2015].
The Chesapeake Bay was the first estuary in
the nation to be selected for restoration as an
integrated watershed and ecosystem. The watershed supports over 2,700 species of plants and
animals, including 348 species of finfish and 173 species of shellfish. Other aquatic life includes
algae, bay grasses, and other invertebrates. The watershed provides habitats for at least 29
species of waterfowl, with a population of nearly one million during the winter (representing
Annual Discharges to the Chesapeake
Bay from the Evaluated Wastestreams
• 993,000 pounds of total nitrogen
• 6,560 pounds of selenium
• 5,830 pounds of zinc
• 5,280 pounds of thallium
2,510 pounds of arsenic
3-40
-------
Section 3—Environmental And Human Health Concerns
approximately one-third of the Atlantic Coast's migratory population) [Chesapeake Bay
Program, 2015].
Most of the Chesapeake Bay and its tidal waters are listed as impaired for excess
nitrogen, phosphorus, and sediment. These pollutants cause oxygen-consuming algae blooms and
create "dead zones" where fish and shellfish cannot survive, block sunlight that is needed for
underwater grasses, and smother aquatic life on the bottom of the Bay. To restore water quality
in the Bay, EPA established Total Maximum Daily Load (TMDL) limits for the Chesapeake Bay
watershed in December 2010. These limits are 186 million pounds of nitrogen, 12.5 million
pounds of phosphorus, and 6.45 billion pounds of sediment each year, reducing the discharges to
the watershed by 25 percent for nitrogen, 24 percent for phosphorus, and 20 percent for
sediment. Pollutant loadings to the Chesapeake Bay watershed come from both point sources and
nonpoint sources. Point sources include municipal wastewater treatment facilities, industrial
discharge facilities (e.g., steam electric power plants and concentrated animal feeding
operations), NPDES permitted stormwater (municipal separate storm sewer systems (MS4) and
construction and industrial sites), and other sources. Nonpoint sources include agricultural land
runoff, atmospheric deposition, forest land runoff, nonregulated stormwater runoff, stream banks
and tidal shorelines, tidal resuspension, the ocean, wildlife, and natural background [U.S. EPA,
2010d].
EPA identified nine steam electric power plants discharging to the Chesapeake Bay
watershed and estimated that these plants discharge almost one million pounds of nitrogen and
over 16,000 pounds of phosphorus to the Bay annually [ERG, 2015a]. Table 3-11 presents the
baseline pollutant loadings for the evaluated wastestreams.
Table 3-11. Pollutant Loadings to the Chesapeake Bay Watershed from the Evaluated
Wastestreams
Pollutant
Arsenic
Boron
Cadmium
Chromium VI
Copper
Lead
Manganese
Mercury
Nickel
Selenium
Thallium
Zinc
Total Nitrogen
Total Phosphorus
Chlorides
Total Dissolved Solids
Annual Discharge to the Chesapeake
Bay Watershed (Ibs)
2,510
1,390,000
513
16.7
2,210
1,560
148,000
88.8
5,280
6,560
5,280
5,830
993,000
16,800
43,000,000
186,000,000
Annual TWPE Discharge to the
Chesapeake Bay Watershed (Ib-eq)
8,720
11,600
11,700
8.62
1,380
3,490
15,200
9,770
575
7,360
15,100
273
-
~
1,050
-
Source: ERG, 2015a.
Note: Numbers are rounded to three significant figures.
3-41
-------
Section 3—Environmental And Human Health Concerns
3.4.3 Proximity to Impaired Waters
A surface water is classified as a 303(d) impaired water when pollutant concentrations
exceed water quality standards and the surface water can no longer meet its designated uses (e.g.,
drinking, recreation, and aquatic habitat). Based on that definition, half of the immediate
receiving waters included in the EA are impaired waters.22 EPA reviewed the identified 303(d)
impairment categories and determined that approximately 27 percent of the immediate receiving
waters are impaired for a pollutant associated with the evaluated wastestreams, as shown in
Table 3-12. Figure 3-1, Figure 3-2, and Figure 3-3 illustrate the geographical location of plants
that directly discharge wastewater to a water classified as impaired by high concentrations of
mercury, metals (other than mercury), and nutrients.
Table 3-12. Number and Percentage of Immediate Receiving Waters Classified as
Impaired for a Pollutant Associated with the Evaluated Wastestreams
Pollutant Causing Impairment
Mercury
Metals, other than mercury b
Nutrients
TDS, including chlorides
Total for Any Pollutant c
Number (Percentage) of Immediate
Receiving Waters Identified a
30 (14%)
28 (13%)
19 (9%)
4 (2%)
70 (32%)
a - For the impaired waters proximity analysis, EPA evaluated 222 immediate receiving waters that receive
discharges of the evaluated wastestreams [ERG, 2015c; ERG, 2015d].
b - The EPA impaired water database listed 28 immediate receiving waters as impaired based on the "metal, other
than mercury" impairment category. Of those 28 immediate receiving waters, 13 receiving waters are also listed as
impaired for one or more specific metals in the EA analysis (arsenic, cadmium, chromium, copper, lead, manganese,
selenium, and zinc). One additional immediate receiving water is impaired for boron (but not included in the
"metals, other than mercury" impairment category).
c - Total does not equal the sum of the immediate receiving waters listed in the table. Some immediate receiving
waters are impaired for multiple pollutants.
22 Table B-l in Appendix B lists the impairment categories identified under the sensitive environments proximity
analysis, including pollutants that are not associated with the evaluated wastestreams.
3-42
-------
Section 3—Environmental And Human Health Concerns
Legend
Plants directly discharging to a surface
water impaired for mercury
0 100 200 400 600
800
I Miles
Figure 3-1. Location of Plants that Directly Discharge the Evaluated Wastestreams
to a Surface Water Impaired due to Mercury
Legend
Plants directly discharging to a surface
water impaired for metals (other than mercury)
0 100 200 400 600
800
I Miles
Figure 3-2. Location of Plants that Directly Discharge the Evaluated Wastestreams
to a Surface Water Impaired due to Metals, Other than Mercury
3-43
-------
Section 3—Environmental And Human Health Concerns
Legend
_ Plants directly discharging to a surface
water impaired for nutrients
0 100 200
400
600
800
I Miles
Figure 3-3. Location of Plants that Directly Discharge the Evaluated Wastestreams
to a Surface Water Impaired due to Nutrients
3.4.4 Proximity to Fish Consumption Advisory Waters
States, territories, and authorized tribes issue fish consumption advisories when pollutant
concentrations in fish tissue are considered unsafe for consumption [U.S. EPA, 201 le]. EPA
determined that 140 of the immediate receiving waters included in the EA (63 percent) are under
fish consumption advisories; 93 of the immediate receiving waters (42 percent) are under an
advisory for a pollutant associated with the evaluated wastestreams.23 All of these 93 immediate
receiving waters are under a fish consumption advisory for mercury and one of the receiving
waters is also under a fish consumption advisory for lead. EPA also reviewed fish consumption
advisories for arsenic, cadmium, and selenium but did not identify any immediate receiving
waters under advisories for these pollutants. Figure 3-4 illustrates the geographical location of
plants that directly discharge steam electric power plant wastewater to surface waters with a fish
consumption advisory for lead or mercury.
23 Table B-2 in Appendix B lists the types of advisories identified under the sensitive environment proximity
analysis, including pollutants that are not associated with the evaluated wastestreams.
3-44
-------
Section 3—Environmental And Human Health Concerns
Plants directly discharging to a surface
water with an advisory for lead
Plants directly discharging to a surface
water with an advisory for mercury
0 100 200
400
600
800
I Miles
Figure 3-4. Location of Plants that Directly Discharge to a Surface Water with a Fish
Consumption Advisory
3.4.5 Proximity to Threatened and Endangered Species Habitats
Under the Endangered Species Act (ESA), endangered species are those in danger of
extinction throughout all or a significant portion of its range. Threatened species are those
species that are likely to become endangered within the foreseeable future. A species may be
listed solely on the basis of their biological status and threats to their existence. The USFWS
considers five factors for listing: 1) damage to, or destruction of, a species' habitat; 2)
overutilization of the species for commercial, recreational, scientific, or education purposes; 3)
disease or predation; 4) inadequacy of existing protection; and 5) other natural or man-made
factors that affect the continued existence of the species.
EPA evaluated the extent to which the estimated range and critical habitats of currently
listed threatened and endangered species, or those in consideration for listing under the ESA (as
of December 2014), overlap with surface waters that are potentially affected by the final rule. As
described in the Benefits and Cost Analysis (EPA-821-R-15-005), these "affected areas" are
receiving waters that do not meet water quality metrics recognized to cause harm in organisms
under baseline conditions, but which do meet these metrics under the most stringent regulatory
option EPA analyzed (Option E). EPA identified 138 threatened and endangered species whose
habitats overlap with, or are located within, an "affected" surface water under baseline
conditions.24
24 The habitat locations evaluated for this analysis include waters downstream from steam electric power plant
discharges and reflect changes in the industry as a result of the Clean Power Plan [Clean Air Act Section 11 l(d)1.
3-45
-------
Section 3—Environmental And Human Health Concerns
In addition, EPA assessed the vulnerability of each species identified to changes in water
quality and developed the following categories:
• High vulnerability: species living in aquatic habitats for several life history stages
and/or species that obtain a majority of their food from aquatic sources.
• Moderate vulnerability: species living in aquatic habitats for one life history stage
and/or species that obtain some of their food from aquatic sources.
• Low vulnerability: species whose habitats overlap bodies of water, but whose life
history traits and food sources are terrestrial.
EPA classified 54 percent of the species (75 of 138 species) with habitats located within
an "affected" surface water as highly vulnerable to changes in water quality. The habitats of
these highly vulnerable species overlap a total of 145 affected stream reaches. For further details
on the threatened and endangered species analysis and results, see the Benefits and Cost Analysis
(EPA-821-R-15-005).
3.4.6 Proximity to Drinking Water Resources
EPA also evaluated the potential for steam electric power plants to pose a threat to public
sources of drinking water. Although many of the pollutants (e.g., selenium, mercury, arsenic,
nitrates) in the evaluated wastestreams would likely be reduced to safe levels during drinking
water treatment, these pollutants could potentially impact the effectiveness of the treatment
processes, which could increase public drinking water treatment costs.25 EPA evaluated the
proximity of steam electric power plants to the following sensitive environments for drinking
water resources:
• Drinking water intakes - drinking water sources that collect surface water through a
public water system. Intakes are protected under the SDWA of 1974 and its 1986 and
1996 amendments, which require delegated states and tribes to perform routine
testing to ensure that they meet state drinking water standards.
• Public wells - drinking water sources that collect ground water through a public
water system. Public wells are protected under the SDWA, which requires delegated
states and tribes to perform routine testing to ensure that they meet state drinking
water standards.
• Sole-source aquifers - drinking water sources that supply at least 50 percent of the
drinking water consumed in the area overlying the aquifer. These areas can have no
reasonably available alternative drinking water source(s) if the aquifer were to
become contaminated.
Table 3-13 summarizes the number and percentages of plants included in the national-
scale proximity analysis that are located within five miles of the evaluated drinking water
resources. The table also presents the number of drinking water resources that are located within
this five-mile buffer zone. For example, 67 steam electric power plants are located within 5 miles
25 For more information on drinking water treatment processes used to reduce or eliminate metals commonly
detected in the evaluated wastestreams from steam electric power plants, see the ERG memorandum "Drinking
Water Treatment Technologies that Can Reduce Metal and Selenium Concentrations Associated with Discharges
from Steam Electric Power Plants" (DCN SE02154).
3-46
-------
Section 3—Environmental And Human Health Concerns
of a drinking water system intake or drinking water reservoir. Within 5 miles of these 67 plants
are 113 drinking water system intakes or reservoirs.
Table 3-13. Comparison of Number and Percentage of Steam Electric Power Plants
Located within 5 Miles of a Drinking Water Resource
Type of Drinking Water
Resource
Intakes and reservoirs
Public wells b
Sole-source aquifers
Number of Drinking Water
Resources within 5 Miles of a Steam
Electric Power Plant
113
2,057
8
Number (Percentage) of Steam
Electric Power Plants
Located within 5 Miles of a
Drinking Water Resource a
67 (33%)
157 (81%)
7 (4%)
Sources: ERG, 2015c; ERG, 2015d
a - For the drinking water resource proximity analysis, EPA evaluated 222 immediate receiving waters that receive
discharges of the evaluated wastestreams from 195 steam electric power plants.
b - Counts include two springs and 29 wellheads.
3.5 LONG ENVIRONMENTAL RECOVERY TIMES ASSOCIATED WITH POLLUTANTS IN STEAM
ELECTRIC POWER PLANT WASTEWATER
Recovery of the environment from exposure to steam electric power plant wastewater is
affected by continued cycling of contaminants within the ecosystem, bioaccumulation, and the
potential alterations to ecological processes, such as population and community dynamics in the
surrounding ecosystems. The ability of aquatic and adjacent terrestrial environments to recover
from even short periods of exposure to steam electric power plant wastewater depends on the
distance from discharge, the pollutant concentrations, pollutant residence time, and the time
elapsed since exposure. In particular, accumulation of metals and other bioacummulative
pollutants in sediments can slow recovery of aquatic systems following exposure to power plant
wastewater due to the potential for resuspension
in the water column and for benthic organisms to
provide a pathway for exposure long after power
plant wastewater discharges have ended. For
example, Lemly [1985a, 1997a, 1999]
documented that benthic pathways can continue to
provide toxic doses of selenium to wildlife even
10 years after water column selenium
concentrations are below levels of concern. Ruhl
et al. [2012] documented elevated levels of power
plant wastewater pollutants (including arsenic and
selenium) in pore water, even in cases where the
water column concentrations are not elevated.
This study found that arsenic is retained in lake
sediments and pore water through a cycle of
adsorption and desorption, likely in response to
seasonal changes in the lake water chemistry
[Ruhl etal, 2012].
As discussed in Section 3.1, many of the pollutants in steam electric power plant
wastewater (e.g., arsenic, mercury, selenium) readily bioaccumulate in exposed biota. The
3-47
Short Exposures to Steam Electric Power
Plant Wastewater Can Equate to Lasting
Ecological Effects
In Martin Creek Lake, ecological effects
persisted for at least 8 years following 8
months of fly ash discharges into the lake.
Ash pond discharges to Belews Lake in
North Carolina resulted in elevated levels of
arsenic, selenium, and zinc in the water and
impacts to fish populations. Even 11 years
after discharges ceased, selenium levels in
the sediments still posed a risk to wildlife
that feed on benthic organisms.
-------
Section 3—Environmental And Human Health Concerns
bioaccumulation of these pollutants is of particular concern due to their impact on higher trophic
levels, local terrestrial environments, and transient species, in addition to the aquatic organisms
directly exposed to the wastewater. Aquatic systems with long residence times and potential
contamination with bioaccumulative pollutants often experience persistent environmental effects
following exposure to steam electric power plant wastewater.
Population decline attributed to exposure to steam electric power plant wastewater can
alter the structure of aquatic communities and cause cascading effects within the food web that
result in long-term impacts to ecosystem dynamics [Rowe et a/., 2002]. Reductions in organism
survival rates from abnormalities caused by exposure to power plant wastewater and alterations
in interspecies relationships, such as declining abundance or quality of prey, can delay ecosystem
recovery until key organisms within the food web return to levels prior to power plant
wastewater exposure. In a 1980 study of a creek in Wisconsin, fungal decomposition of detritus
was limited due to the effects of power plant
wastewater. As a result, the benthic
invertebrate population, which graze on detrital
material, declined as did benthic fish that prey
upon small invertebrates because of the
reduced available resources [Magnuson et a/.,
1980].
Belews Lake, a 1,500-hectare cooling
reservoir constructed to support the Belews
Creek Steam Station in Stokes County, North
Carolina, is a well-documented site that
highlights the effects that steam electric power
plant wastewater can have on fish populations
and the subsequent long recovery time. In
1970, Duke Energy began monitoring the fish
populations in Belews Lake prior to any
discharges of steam electric power plant
Studies have linked historical discharges of
selenium from the Belews Creek Steam Station
with persistent ecological impacts in the plant's
cooling reservoir.
wastewater. From 1974 to 1985, Duke Energy discharged surface impoundment effluent into
Belews Lake. Almost immediately after these discharges began, rapid and dramatic changes in
the fish populations were observed [Lemly, 1993]. By 1975, morphological abnormalities (e.g.,
partial fin loss, head deformities, cataracts) were reported for all 19 fish species monitored in the
lake. Within 2 years after surface impoundment effluent was released into the lake, several
species stopped reproducing, leaving only four species by 1978 (i.e., 4 years after discharges
began). Water samples collected in the lake reported elevated levels of arsenic, selenium, and
zinc. Large predatory fish were some of the first species to die out completely, due to the lethal
and sublethal effects of exposure to surface impoundment effluent. Because a top predator was
gone, some fish that exhibited developmental abnormalities were able to survive, despite their
otherwise high susceptibility to predation [Lemly, 1993]. The study eventually correlated the
observed fish abnormalities with high selenium whole-body concentrations, and identified the
planktonic community as the key source of selenium to the impacted fish. In 1985, the Belews
Creek Steam Station switched to disposing of the coal ash in a dry landfill and ended the surface
impoundment discharges to the lake. In a 1997 study, Lemly determined that there was evidence
that the lake was recovering; however, even 11 years after the discharges ceased, selenium levels
in the sediments still posed a risk to wildlife that feed on benthic organisms. Lemly also
3-48
-------
Section 3—Environmental And Human Health Concerns
observed that despite the reduction in the selenium concentration in fish ovaries, reproductive
abnormalities remained persistent, highlighting the long ecological recovery time observed in
Belews Lake.
In addition to population density effects, the diversity of species in the communities in
both field and experimental studies exposed to steam electric power plant wastewater has altered,
which can further prolong ecosystem recovery [Benson and Birge, 1985; Guthrie and Cherry,
1976; Rowe et a/., 2001; Specht et a/., 1984]. In a study of fish populations in Martin Creek
Lake following a short 8-month period in which the lake received fly ash surface impoundment
discharges, both planktivorous (i.e.., diet primarily consists of plankton) and carnivorous (i.e..,
diet primarily consists of meat) fish populations were severely reduced [Garrett and Inman,
1984]. Three years after the effluent release was halted, planktivorous fish populations remained
extremely low, while carnivorous fish populations had nearly recovered. Carnivorous fish have a
more diverse diet than planktivorous fish and therefore benefited from an increase in food
availability as the aquatic system recovered; however, the size of carnivorous fish in the lake
suggested that surviving adults continued to have reproductive impairments [Garrett and Inman,
1984]. Sorensen (1988) documented that ecological impacts in the lake remained evident even up
to 8 years after the 8-month exposure to fly ash transport water discharges, with sunfish
populations continuing to exhibit tissue damage to the liver, kidneys, gills, and ovaries and
impaired overall reproductive health. Fish samples taken in 1996 and 1997 showed that the
selenium concentration (2.3 parts per million (ppm) average for all sample fish) remained well
above the national average range of between 0.1 and 1.5 ppm [ATSDR, 1998a].
3-49
-------
Section 4—Assessment of Exposure Pathways
SECTION 4
ASSESSMENT OF EXPOSURE PATHWAYS
An exposure pathway is defined as the route a pollutant takes from its source (e.g.,
combustion residual surface impoundments) to its endpoint (e.g., a surface water), and how
receptors (e.g., fish, wildlife, or people) can come into contact with it. Exposure pathways are
typically described in terms of five components:
• Source of contamination (e.g., steam electric power plant wastewater).
• Environmental pathway—the environmental medium or transport mechanism that
moves the pollutant away from the source through the environment (e.g., discharges
to surface waters).
• Point of exposure—the place (e.g., private drinking water well) where receptors (e.g.,
people) come into contact with a pollutant from the source of contamination.
• Route of exposure—the way (e.g., ingestion, skin contact) receptors come into
contact with the pollutant.
• Receptor population—the aquatic life, wildlife, or people exposed to the pollutant.
The exposure pathway plays an
important role in determining the potential
effects of steam electric power plant
wastewater on the environment. For example,
the physical and chemical characteristics of
receiving waters can affect the fate and
transport of pollutants from combustion
residual surface impoundments to the
environment and ultimately impact how the
pollutants interact with the biological
community.
EPA identified four primary exposure
pathways of concern for steam electric power
plant wastewater entering the environment: 1)
discharges entering surface waters, 2)
uncollected combustion residual leachate
infiltrating through soil to nearby surface
water, 3) uncollected combustion residual leachate entering ground water, and 4) direct contact
with steam electric power plant wastewater stored in surface impoundments. This section
describes the factors that control the magnitude of impacts to water quality, wildlife, and human
health associated with exposure to steam electric power plant discharges and presents an
overview of EPA's environmental assessment (EA) of the steam electric power generating
industry, in which EPA evaluated the national-scale effects of power plant wastewater pollutants
on the environment. Table 4-1 presents the environmental pathways, routes of exposure, and
environmental concerns identified during the literature review and the types of analyses
conducted to determine the impacts under baseline conditions and regulatory options.
Pollutants from steam electric power plant
wastewater stored in surface impoundments can
reach receptor populations (such as wildlife or
people) through various exposure pathways.
4-1
-------
Section 4—Assessment of Exposure Pathways
Table 4-1. Steam Electric Power Plant Wastewater Environmental Pathways and
Routes of Exposure Evaluated in the EA
Environmental Pathway
Steam electric power plant
wastewater discharges to surface
waters
Uncollected combustion residual
leachate infiltration to nearby
surface waters from combustion
residual surface impoundment or
landfill
Uncollected combustion residual
leachate entering ground water
from combustion residual
surface impoundment or landfill
Combustion residual surface
impoundment
Route of Exposure
Direct contact with
surface water
Ingestion of surface water
Direct contact with
sediment
Consumption of aquatic
organisms
Direct contact with
surface water or sediment
Ingestion of ground water
Direct contact with or
ingestion of surface water
Environmental Concern
Toxic effects on aquatic
organisms a
Degradation of surface
water quality used as intake
to drinking water plants
Toxic effects on benthic
organisms
Bioaccumulation of
contaminants and resulting
toxic effects on wildlife
Toxic effects on humans
consuming contaminated
fish
Toxic effects on humans
and aquatic wildlife
Changes in ground water
quality
Contaminated private
drinking water wells
Toxic effects on wildlife
Bioaccumulation of
contaminants in wildlife
Analysis to Determine
Environmental Impact
Water quality impacts
analysis (quantitative) -
see Section 4. 1.2
Wildlife impacts
analysis (quantitative) -
see Section 4. 1.2
Human health impacts
analysis (quantitative) -
see Section 4. 1.2
Ground water quality
impacts analysis
(qualitative) - see
Section 4.2.2
Attractive nuisances
analysis (qualitative) -
see Section 4.3
a - The term "toxic effects" refers to impacts upon exposure, ingestion, inhalation, or assimilation into any
organism, either directly from the environment or indirectly by ingestion through food chains. These effects can
include death, disease, behavioral abnormalities, cancer, genetic mutations, physiological malfunctions (including
malfunctions in reproduction), or physical deformations, in receptors (e.g., aquatic organisms, wildlife, humans) or
their offspring.
4.1 DISCHARGE AND LEACHING TO SURFACE WATERS
Steam electric power plants commonly discharge wastewater directly to surface waters
following storage and treatment (e.g., particulate settling) in surface impoundments. In addition
to effluent discharges, uncollected combustion residual leachate can migrate through the soil and
into the surface water. Section 4.2 further discusses the impacts of uncollected combustion
residual leachate.
4.1.1 Factors Controlling Environmental Impacts in Surface Waters
One of the primary factors controlling the environmental impact of steam electric power
plant wastewater on surface waters is the residence time of the pollutants once they enter an
4-2
-------
Section 4—Assessment of Exposure Pathways
aquatic system. Residence times are often determined by the flow rate of the receiving water and
type of ecosystem it supports. The potential for pollutant retention in lentic aquatic systems (i.e.,
still or slow-moving water, such as lakes or ponds) and the creation of hot spots in lotic aquatic
systems (i.e., flowing water, such as streams and rivers) are of particular concern when
bioaccumulative pollutants are present. Many of the pollutants in steam electric power plant
wastewater discharges bioaccumulate, complicating estimates of potential impacts in surface
waters because the pollutants can affect higher trophic levels, local terrestrial environments, and
transient species, in addition to the aquatic organisms directly exposed to the wastewater.
Based on industry responses to EPA's 2010 Questionnaire for the Steam Electric Power
Generating Effluent Guideline (Steam Electric Survey),26 EPA determined that 18 percent of the
222 receiving waters included in the scope of the EA, all of which receive steam electric power
plant wastewater discharges, are lentic systems such as lakes, ponds, reservoirs, and estuaries
(Table 4-2). The majority of ecological studies on the impact of power plant wastewater in
aquatic environments have focused on lentic systems [Rowe et a/., 2002]. In lentic aquatic
systems, the hydraulic residence time, or the amount of time it takes for the water in the aquatic
system to be replaced by inflowing streams or precipitation is relatively long, allowing pollutants
to build up over time and making these systems more vulnerable to impacts from power plant
wastewater. In addition, aquatic organisms are limited in their ability to avoid areas of high
pollutant concentrations and are restricted to the food supply available only within the
waterbody.
Table 4-2. Receiving Water Types for Steam Electric Power Plants Evaluated in the EA
Receiving Water Type
River/Stream
Lake/Pond/Reservoir
Great Lakes
Estuary and others (bay)
Total Receiving Waters
Number (Percentage) of Immediate
Receiving Waters a
183 (82%)
26 (12%)
11(5%)
2 (1%)
222 (100%)
Source: ERG, 2015d.
a - The EA encompasses a total of 222 immediate receiving waters and loadings from 195 steam electric power
plants (some of which discharge to multiple receiving waters). The immediate receiving water (IRW) model,
which excludes the Great Lakes and estuaries, encompasses a total of 209 immediate receiving waters and
loadings from 188 steam electric power plants.
Based on responses to EPA's Steam Electric Survey, EPA determined that 82 percent of
aquatic environments that receive discharges of the evaluated wastestreams are lotic systems
such as rivers and streams [ERG, 2015J]. Lotic systems dilute discharges more quickly than
lentic systems. The moving water in lotic systems also provides a transport mechanism to
disperse pollutants greater distances from the power plant, and enables aquatic organisms to
move away from the areas contaminated by steam electric power plant discharges [Rowe et a/.,
26 Results presented in this report are based on plant responses to the Steam Electric Survey, which represent 2009
data. However, the analyses presented in this report incorporate some adjustments to current conditions in the
industry. See Section 1 for further details.
4-3
-------
Section 4—Assessment of Exposure Pathways
2002]. Although power plant wastewater discharges into a lotic system can distribute pollutants
across a greater spatial area, changes in flow velocity may result in the concentration of
pollutants at a single location further downstream [Rowe et al., 2002]. For example, power plant
wastewater discharged to a river may encounter areas of slower moving water downstream
where pollutants would fall out of suspension and concentrate in a limited area. These pockets of
higher pollutant concentrations, or hot spots, could be vulnerable to continued resuspension as
stream velocities are affected by rainfall, resulting in the aquatic organisms being exposed to
pollutants over much longer periods of time [Lemly, 1997a; Rowe etal., 2002].
4.1.2 Assessment of the Surface Water Exposure Pathway
EPA developed and executed models to quantify the water quality, wildlife, and human
health impacts resulting from discharges of the evaluated wastestreams to surface waters. These
models consist of the following: 1) a national-scale IRW model that evaluates the discharges
from 186 steam electric power plants and focuses on impacts within the immediate surface
water27 where discharges occur, and 2) case study models that perform more sophisticated and
extensive modeling of selected waterbodies that receive, or are downstream from, steam electric
power plant wastewater discharges. Section 5 describes the IRW model and Section 8 describes
the case study models. In addition, as part of the benefits and cost analysis, EPA also evaluated
surface water concentrations downstream from steam electric discharges using EPA's Risk-
Screening Environmental Indicators (RSEI) model; see the Benefits and Cost Analysis (EPA-
821-R-15-005).
The remainder of this section discusses the scope of EPA's environmental assessment of
the steam electric power generating industry in terms of evaluated pollutants, evaluated
waterbody types, and evaluated environmental impacts.
Evaluated Pollutants
The EA quantitative analyses focused on the environmental impacts associated with
discharges of toxic, bioaccumulative pollutants to surface waters. A key factor in determining the
pollutants to include in the quantitative analyses was the potential for pollutant loadings to be
diluted in the receiving waters following discharge. For example, EPA determined that the rivers
and streams included in the IRW model had a median average annual flow of 2,808 cubic feet
per second (cfs) and that 57 percent had an average annual flow greater than 1,000 cfs. Due to
the potential for dilution, EPA focused the quantitative analyses on pollutants where the total
mass loadings and not the concentration are critical factors in determining the potential for
environmental impact. Section 5.1.2 lists the pollutants selected for quantitative analyses and
how they were selected.
27 The length of the immediate receiving water, as represented in the national-scale IRW model, ranges from
between 1 to 5 miles from the steam electric power plant outfall. See the ERG memorandum "Water Quality
Module: Plant and Receiving Water Characteristics" (DCN SE04513) for details on the immediate discharge zone
and length of stream reach represented.
4-4
-------
Section 4—Assessment of Exposure Pathways
The EA quantitative analyses did not focus on water quality impacts associated with
discharges of nutrients (total nitrogen and total phosphorus).28 While discharges of large amounts
of nutrients to surface waters can cause environmental problems (e.g., eutrophication), EPA
focused the EA quantitative analyses on 10 toxic pollutants that can bioaccumulate in fish and
impact wildlife and human receptors via fish consumption. Additionally, nutrient-related impacts
tend to be site-specific depending on environmental factors (e.g., water-body temperature, the
limiting nutrient in the system, algal species in the waterbody, and availability of oxygen in the
water).
While the EA quantitative analyses did not address nutrient-related impacts, EPA did
include nutrient loadings in the Benefits and Cost Analysis. EPA estimated total nitrogen and
total phosphorus concentrations in receiving waters using dilution equations as input values to
analyze benefits related to improvements in water quality. EPA used the SPARROW (SPAtially
Referenced Regressions On Watershed attributes) model to provide baseline concentrations, as
well as concentrations under each regulatory option. EPA used these concentrations to develop
subindices for a water quality index (WQI), a value that translates water quality measurements,
gathered for multiple parameters that represent various aspects of water quality, into a single
numerical indicator. Total nitrogen and total phosphorous are only two of the subindices
included in the WQI; the others are dissolved oxygen, biochemical oxygen demand, fecal
coliform, total suspended solids (TSS), and heavy metals. EPA then used the WQI as a basis for
calculating a willingness to pay for an increase in water quality as a result of the different
regulatory options. See the Benefits and Cost Analysis for further details on the analysis and the
results.
EPA identified total dissolved solids (TDS) and chlorides as the pollutants with the
largest loadings under baseline conditions (see Table 3-2); however, EPA did not perform
quantitative analyses of these pollutants for several reasons. TDS from the evaluated
wastestreams consists largely of dissolved metals that are already captured in the analysis.
Therefore, estimates of potential environmental impacts from TDS would double-count many of
the environmental impacts and potential improvements assessed. Chlorides lack partition
coefficient data (which are necessary for the water quality modeling performed in this EA) and
have limited numeric threshold criteria data for comparison.
Evaluated Waterbody Types
In selecting the appropriate methodologies for the quantitative analyses, EPA considered
the types of receiving waters commonly impacted by steam electric power plants and the
pollutants typically found in the evaluated wastestreams. The IRW model and the selected case
study models quantify the environmental risks within rivers/streams and lakes/ponds (including
reservoirs), based on the determination that 94 percent of the final outfall receiving water
designations fell within these two categories.
The EA quantitative analyses did not evaluate pollutant concentrations in the Great Lakes
and estuarine systems, which represented 6 percent of all final outfall receiving waters. The
28 EPA evaluated the nutrient impacts to the Great Lakes and Chesapeake Bay systems from a total mass loadings
perspective, discussed in Section 3.4.
4-5
-------
Section 4—Assessment of Exposure Pathways
specific hydrodynamics and scale of the analysis required to appropriately model and quantify
receiving water concentrations in the Great Lakes and estuarine systems are more complex than
the IRW model.29 In selecting the receiving waters to evaluate in the case study analyses, EPA
focused primarily on rivers and streams based on the following: 1) the determination that 82
percent of the final outfall receiving water designations fell within this category, and 2) the
relative simplicity of the hydrodynamics in river and stream case study models. This allowed
EPA to develop and execute a larger set of case studies. EPA also developed one case study to
represent the impacts of steam electric discharges to a lake. Refer to Section 8 for discussion of
the receiving waters selected for case study analyses.
Evaluated Environmental Impacts
EPA focused the evaluation of environmental impacts on four key areas resulting from
discharges of harmful pollutants to surface waters (rivers, streams, lakes, ponds, and reservoirs):
• Water Quality Impacts: Potential toxic effects to aquatic life based on changes in
surface water quality—specifically, exceedances of the acute and chronic National
Recommended Water Quality Criteria (NRWQC) for freshwater aquatic life.
• Wildlife Impacts: Potential toxic effects on benthic organisms based on changes in
sediment quality within surface waters—specifically, exceedances of chemical
stressor concentration limits (CSCL) for sediment biota.
• Wildlife Impacts: Bioaccumulaton of contaminants and potential toxic effects on
wildlife from consuming contaminated aquatic organisms, specifically:
- Risk of adverse reproductive impacts in fish and waterfowl that consume aquatic
organisms with elevated levels of selenium (as determined by the ecological risk
modeling methodology described in Section 5.2).
Potential risk of reduced reproduction rates in piscivorous wildlife, based on
exceedances of no effect hazard concentration (NEHC) benchmarks.
• Human Health Impacts: Potential toxic effects to human health from consuming
contaminated fish and water, specifically:
Exceedances of the human health NRWQC based on two standards: 1) standard
for the consumption of water and organisms and 2) standard for the consumption
of organisms.
Exceedances of drinking water maximum contaminant levels (MCLs). Although
MCLs apply to drinking water produced by public water systems and not surface
waters themselves, EPA identified immediate receiving waters that exceeded a
MCL as an indication of the degradation of the overall water quality following
exposure to the evaluated wastestreams.
29 EPA evaluated the impacts to the Great Lakes and Chesapeake Bay systems from a total mass loadings
perspective, discussed in Section 3.4. See the ERG memorandum "Site-Specific Estuary Dilution Analysis" (DCN
SE02152) for details on EPA's initial screening analysis of the modeled receiving water concentrations in the Great
Lakes and estuary systems compared to water quality benchmarks.
4-6
-------
Section 4—Assessment of Exposure Pathways
- Risk of cancer and non-cancer threats (e.g., reproductive or neurological impacts)
due to consuming fish caught from contaminated receiving waters.
4.2 LEACHING TO GROUND WATER
Combustion residual landfills and surface impoundments can impact local ground water
through leaching.30 Once in ground water, pollutants can migrate from the site and contaminate
public or private drinking water wells and surface waters [NRC, 2006]. Contamination of
drinking water wells is of particular concern because more than one-third of the U.S. population
relies on ground water for drinking water. According to the U.S. Geological Survey (USGS), one
in every five samples of ground water used as a source for drinking contains at least one
contaminant at a level of concern for human health [USGS, 2015].
The fate of pollutants that leach from combustion residuals to ground water is controlled
by many biological and geochemical (e.g., adsorption, desorption, and precipitation reactions
with aquifer materials) processes that can vary over large spatial and temporal scales [NRC,
2006]. This section describes the pollutant concentrations, chemical characteristics (e.g.,
solubility, teachability, persistence, and mobility), and fate and transport processes that influence
the potential environmental impact of uncollected combustion residual leachate.
4.2.1 Factors Controlling Environmental Impacts to Ground Water
Environmental impacts to ground water are determined by the pollutant concentrations in
the combustion residual leachate and the rate of pollutant transport in the ground water. The
pollutant concentrations in the combustion residual leachate depend on factors such as
characteristics of the combustion residuals, site conditions (e.g., rainfall amount and pH of the
pore water in the surface impoundment or landfill), and combustion residual residence time in
the surface impoundment or landfill.31 The rate of pollutant transport in ground water depends on
factors such as the biogeochemical characteristics of the subsurface (e.g., soil pH and oxidation-
reduction potentials), local rates of ground water recharge, and unsaturated and saturated ground
water flow velocities.
Pollutant Concentrations in Combustion Residual Leachate
Combustion residual characteristics include the mineralogy of the waste (e.g., lime,
gypsum, iron, and aluminum oxide content) and pollutant solubility in the pore water. The
mobility of pollutants may be altered due to changes in pH, carbon and chloride content, and
interaction with other wastes from steam electric power plants [Thorneloe et.al, 2010]. The
waste mineralogy can vary based on the chemical composition in the fuel source (e.g., the
30 In this EA, EPA evaluated the threats to human health and the environment associated with pollutants leaching
into ground water from surface impoundments and landfills containing combustion residuals. If these leached
pollutants do not constitute the discharge of a pollutant to surface waters, then they are not controlled under the
steam electric ELGs. While the Coal Combustion Residuals (CCR) rulemaking is the major controlling action for
these pollutant releases to ground water, the ELGs could indirectly reduce impacts to ground water. These
secondary improvements are discussed in Section 7.8.
31 Leaching experiments indicate that the chemistry of leachates is based on both the chemical composition of the
waste and other factors such as site conditions [Thorneloe et ai, 2010]. Thorneloe [2010] specifically looked at fly
ash and bottom ash waste from coal-fired power plants.
4-7
-------
Section 4—Assessment of Exposure Pathways
specific coal seam and geographic location of the mine) and operational characteristics at the
plant. Many laboratory investigations have examined the solubility characteristics of various
pollutants associated with fly ash [Prasad et a/., 1996; Thorneloe et.al., 2010]. The results of
these investigations largely depend on multiple factors, and they tend to be more applicable
qualitatively rather than quantitatively (e.g., results from investigations can be used to determine
the likelihood of a pollutant to dissolve in the combustion residual leachate, but not the amount).
Concentrations of inorganic pollutants derived from calcium, sodium, magnesium, potassium,
iron, sulfur, and carbon are relatively high in aqueous solution of fly ash because of their high
total concentrations in the ash [Prasad et a/., 1996].
The pH of the pore water is a dominant
factor in the leaching of pollutants from
unlined surface impoundments and landfills.
Because most pollutants in combustion
residuals exhibit weak acidic or weak basic
behavior in aqueous solution, the pore water
pH strongly influences the concentrations of
pollutants in the combustion residual leachate.
Steam electric power plants generate
combustion residuals in high-temperature
processes, and many acids and acidic
precursors (e.g., carbon dioxide, hydrogen
sulfide, hydrochloric acid) are volatilized prior
to waste collection. Therefore, combustion
residuals typically yield an alkaline reaction in
water, but acidic reactions have also been
observed [Theis and Gardner, 1990]. Acidic
pore water allows pollutants from the
combustion residuals to remain in solution, increasing their mobility and the potential for ground
water contamination. The results of a study of three power plants in Turkey indicated that
combustion residuals in the deeper layers of landfills and on the bottoms of the surface
impoundments may continue to leach if the pH value drops in the surrounding environment
[Baba and Kaya, 2004].32
Table 4-3 presents data collected by EPA's Steam Electric Survey regarding pollutant
concentrations in the combustion residual leachate under acidic, neutral, and basic (or alkaline)
conditions. Arsenic exceeded its MCL for more than 60 percent of the samples in both acidic and
basic combustion residual leachate. Similarly, the majority of manganese samples exceeded its
secondary MCL under all pH conditions, with 95 percent of the samples exceeding the MCL in
ThepH level of pore water in surface
impoundments can strongly influence the
concentration of pollutants in leachate from
impoundments to ground water.
32 This conclusion was based on a comparison of ash extraction procedures used. The study examined how the
concentration of trace elements in the ash can vary based on the procedure used, comparing the EPA-developed EP
(extraction procedure) and its replacement method, TCLP (toxicity characteristic leaching procedure), and the
ASTM (American Society for Testing and Materials) Method D-3987. A comparison of the results revealed that the
ASTM procedure indicated much lower dissolved metal concentrations than the EP and TCLP procedures. These
results indicate that pH is an important parameter affecting the leaching rate of metals from ash deposits. The lower
pH values in the EP and TCLP methods increase the leaching rate of inorganic constituents of fly ash and bottom
ash [Fleming et al, 1996].
4-8
-------
Section 4—Assessment of Exposure Pathways
acidic conditions. Selenium had varying concentrations under all pH conditions, but exceeded its
MCL more frequently under basic conditions. Overall, the results support the conclusion that pH
levels influence the concentrations of pollutants in the combustion residual leachate.
Table 4-3. Exceedances of MCLs in Leachate Under Acidic, Neutral, and Basic
Conditions
Pollutant
Arsenic
Boron
Cadmium
Chromium
Copper
Lead
Manganese
Mercury
Nickel
Selenium
Thallium
Zinc
MCL
(mg/L)
0.01
7a
0.005
0.1
1.3
0.015
0.05 b
0.002
No MCL
0.05
0.002
5b
Total Number of Samples
Acidic
21
21
21
21
21
21
21
21
21
21
21
21
Neutral
64
64
63
64
64
62
64
64
64
64
62
63
Basic
90
91
90
90
91
86
89
89
87
90
86
86
Percentage of Total Samples
Exceeding MCL
Acidic
62%
14%
29%
0%
0%
5%
95%
5%
NC
14%
52%
0%
Neutral
30%
31%
3%
0%
0%
0%
81%
16%
NC
17%
10%
0%
Basic
71%
31%
29%
18%
0%
2%
54%
8%
NC
31%
14%
0%
Source: ERG, 2015d.
Acronyms: mg/L (milligrams per liter); MCL (Maximum contaminant level); NC (not calculated; no MCL for
comparison).
Note: Data are for untreated leachate collected in leachate collection systems at steam electric landfills and surface
impoundments.
a - The drinking water equivalent level, used for noncarcinogenic endpoints, is listed rather than the MCL.
b - MCL is a secondary (nonenforceable) standard.
In addition to the pH of the pore water, amounts of precipitation can affect pollutant
concentrations in the combustion residual leachate. Although landfills are dry disposal sites,
rainfall and frozen precipitation infiltrate through the waste, dissolving pollutants that can then
leach from the landfill. Landfills in drier climates generate less combustion residual leachate than
landfills in wetter climates.
The last factor affecting pollutant concentrations in the combustion residual leachate is
the combustion residual residence time in the surface impoundment or landfill. In a study of
metals (calcium, copper, iron, lead, magnesium, manganese, potassium, sodium, and zinc)
leaching from fly ash and bottom ash, all pollutants decreased in concentration with time of
leaching, except for calcium, which released at a constant rate [Kopsick and Angino, 1981]. The
most commonly noted leachate release curve is an initial flush curve, where the highest
concentrations of pollutants are released as the leachate initially forms, with rapidly decreasing
concentrations over time. Therefore, active surface impoundments receiving fresh combustion
residuals will produce a leachate with elevated concentrations of pollutants that have a greater
potential to contaminate drinking water sources and surface waters. Most inactive surface
impoundments where pollutants have initially already leached from the combustion residuals
4-9
-------
Section 4—Assessment of Exposure Pathways
should produce a leachate with decreasing concentrations of pollutants [Kopsick and Angino,
1981].
Thorneloe et al. [2010] studied the leaching behavior of coal combustion residuals in
landfills, performing tests using a range of pH conditions and liquid-solid ratios expected during
management via landfills or beneficial use. Combustion residual leachate concentrations for most
pollutants were variable over a range of coal types, plant configurations, and combustion residual
types (i.e., fly ash or flue gas desulfurization (FGD) gypsum). The study showed significantly
different leaching results (liquid-solid partitioning [equilibrium] as a function of pH) for similar
combustion residual types and plants. The variability in pollutant leaching results was several
orders of magnitude higher than the variability in the pollutant concentrations in the combustion
residuals; this indicates that the pollutant
concentrations alone cannot predict the
leaching of metals, as noted above. Table 4-4
presents pollutant concentrations in
combustion residual samples across a pH
range of 5.4 to 12.4 and the range of pollutant
concentrations in the combustion residual
table also includes indicator
leachate. The
values for
characteristic
Conservation
each pollutant: toxicity
(TC) values for Resource
and Recovery Act (RCRA)
hazardous waste regulatory determination and
drinking water MCLs for combustion residual
leachate concentrations. As shown in the table,
the maximum combustion residual leachate
pollutant concentrations:
Most surface impoundments are unlined,
allow ing pollutants to infiltrate into ground
water and eventually into surface waters.
• Exceed the TC values for RCRA hazardous waste determinations for arsenic, barium,
chromium, and selenium (in fly ash).
• Exceed the TC values for RCRA hazardous waste determinations for selenium (in
FGD gypsum).
• Exceed the MCLs for nine metals (in fly ash and FGD gypsum): antimony, arsenic,
barium (fly ash only), boron, cadmium, chromium, molybdenum, selenium, and
thallium.
The higher pollutant concentrations in the combustion residual leachate indicate greater
mobility of the pollutant from the solid/slurry residual to the liquid phase. The concentration of
the pollutants in the combustion residual leachate can be hundreds to thousands of times greater
than the MCL.
4-10
-------
Section 4—Assessment of Exposure Pathways
Table 4-4. Range of Fly Ash and FGD Gypsum Total Content and Combustion Residual
Leaching Test Results (Initial Screening Concentrations) for Trace Metals
Pollutant
Antimony
Arsenic
Barium
Boron
Cadmium
Chromium
Mercury
Molybdenum
Selenium
Thallium
Range of Combustion
Residual Content
Fly Ash
(mg/kg)
3.0-14
17-510
50-7,000
NA
0.3-1.8
66-210
0.1-1.5
6.9-77
1.1-210
0.72-13
FGD
Gypsum
(mg/kg)
0.14-8.2
0.95-10
2.4-67
NA
0.11-0.61
1.2-20
0.01-3.1
1.1-12
2.3-46
0.24-2.3
Range of Leaching Test
Results: Concentration in the
Combustion Residual
Leachate
Fly Ash
(Mg/L)
0.3-11,000
0.32-18,000
50-670,000
210-270,000
0.1-320
0.3-7,300
0.01-0.50
0.5-130,000
5.7-29,000
0.3-790
FGD
Gypsum
(Mg/L)
O.3-330
0.32-1,200
30-560
12-270,000
0.2-240
O.3-240
0.01-0.66
0.36-1,900
3.6-16,000
0.3-1,100
Indicator Values
TC Value for
Hazardous Waste
Designation
(MS/L)
~
5,000
100,000
~
1,000
5,000
200
~
1,000
~
Drinking
Water
MCL
(Mg/L)
6
10
2,000
7,000 a
5
100
2
200 a
50
2
Source: Thorneloe etal, 2010.
Acronyms: Acronyms: MCL (maximum contaminant level); mg/kg (milligrams per kilogram); TC (Toxicity
Characteristics); ug/L (micrograms per liter); NA (Not Available).
a - The drinking water equivalent level, used for noncarcinogenic endpoints, is listed rather than the MCL.
Transporting Pollutants in the Ground Water
Predicting the movement of combustion residual pollutants in ground water can be
challenging due to the wide range of biogeochemical characteristics between sites and within a
given site. Pollutant transport times can vary, and combustion residual pollutants can take many
years to reach local drinking water wells and surface waters [NRC, 2006]. For example, in the
damage case at the Virginia Power Yorktown Power Station Chisman Creek Disposal Site in
Yorktown, Virginia, fly ash had been disposed of in abandoned, unlined sand and gravel pits at
the site for almost 20 years, from 1957 to 1974. However, ground water contamination was not
discovered until 1980, when nearby shallow residential wells became contaminated with nickel
and vanadium. Sampling also showed elevated levels of other heavy metals and toxic pollutants:
arsenic, beryllium, chromium, copper, molybdenum, and selenium [U.S. EPA, 2014b].
Natural mechanisms, such as soil buffering capacity, attenuation of trace pollutants in
certain soil types, amount of organic matter, and low soil permeability, can limit the transport of
combustion residual pollutants in the subsurface environment. The mobility of pollutants in the
subsurface strongly depends on soil-specific characteristics. Soil can have a buffering influence
over the leachate by raising or lowering the pH. As noted previously, the solubility of most trace
pollutants (the notable exceptions being arsenic and selenium) tends to decrease with increased
pH (i.e., alkaline conditions). In general, trace pollutants are less mobile in alkaline soils because
the pollutants will precipitate and/or adsorb onto hydrous iron and aluminum oxides. Theis and
Richter [1979] attempted to assess the factors influencing the attenuation of trace metals in
4-11
-------
Section 4—Assessment of Exposure Pathways
soil/ground water. Results show that the major solubility control for cadmium, nickel, and zinc is
adsorption by iron and manganese oxides while chromium, copper, and lead are controlled by
precipitation. In some cases, particles in leachate may seal a surface impoundment or landfill,
reducing the amount of leachate entering the ground water. Simsiman et al. [1987] and Kopsick
and Angino [1981] both reported evidence of some sealing and reduced permeability of
combustion residual surface impoundments, reducing seepage.
4.2.2 Assessment of the Ground Water Exposure Pathway
The EA focused on the discharges of toxic, bioaccumulative pollutants to surface waters
from the evaluated wastestreams. While Section 3.3 provides qualitative discussion of ground
water impacts based on a review of damage cases and other documented site impacts, the EA did
not quantify the environmental and human health impacts resulting from pollutants leaching into
the ground water from combustion residual surface impoundments and landfills. Additionally,
the models used for this EA did not consider pollutant loadings to surface waters caused by
combustion residual pollutants migrating through the soil and into surface waters, even though
this may be occurring at many of the plants. As shown in Tables A-4 and A-5 in Appendix A,
several damage cases have documented impacts to surface waters due to ground water
contamination from combustion residual surface impoundments and landfills. The EA may
therefore underestimate the number of cases where water quality standards are being exceeded in
immediate receiving waters (see Section 6).
On April 17, 2015, EPA published a RCRA rule that regulates the disposal of CCRs from
steam electric power plants (80 FR 21302). As part of the final CCR rulemaking, EPA's Office
of Solid Waste and Emergency Response (OSWER) evaluated ground water contamination
associated with combustion residuals in surface impoundments and landfills. The ground water
impact analysis for the CCR rule identified and quantified human health risks to private drinking
water wells due to potential ground water contamination from current CCR management
practices. The analysis determined that human health risks were primarily from exposures to
arsenic and molybdenum in ground water used as a source of drinking water. EPA identified
additional human health risks from exposures to boron, cadmium, cobalt, fluoride, mercury,
lithium, and thallium in ground water used as drinking water at certain sites based on the CCR
disposal practices. Refer to the Regulatory Impact Analysis: EPA's 2015 RCRA Final Rule
Regulating Coal Combustion Residual (CCR) Landfills and Surface Impoundments at Coal-Fired
Electric Utility Power Plants (EPA-HQ-RCRA-2009-0640-12034) for the results of the national-
scale analysis of ground water impacts.
4.3 COMBUSTION RESIDUAL SURFACE IMPOUNDMENTS AS ATTRACTIVE NUISANCE
An "attractive nuisance" is an area or habitat that attracts wildlife and is contaminated
with pollutants at concentrations high enough to potentially harm exposed organisms. Two
methods of handling steam electric power plant wastewater, surface impoundments and
constructed wetlands, are classified as lentic systems supporting aquatic vegetation and
organisms. These methods have been known to attract wildlife from other terrestrial habitats and
therefore can be considered attractive nuisances. As an attractive nuisance, a surface
impoundment can impact local wildlife as well as transient species that might rely on them
during critical reproduction periods such as seasonal breeding events [Rowe et a/., 2002].
4-12
-------
Section 4—Assessment of Exposure Pathways
Exposure to steam electric power plant wastewater during sensitive life cycle events is a concern
given that it has been associated with complete reproductive failure in various vertebrate species
[Cumbie and Van Horn, 1978; Gillespie and Baumann, 1986; Lemly, 1997a; Pruitt, 2000].
Organisms that frequent attractive nuisance sites at steam electric power plants, such as
surface impoundments, risk exposure to elevated pollutant concentrations. Several studies have
shown that terrestrial fauna nesting near combustion residual surface impoundments can have
higher levels of arsenic, cadmium, chromium, lead, mercury, selenium, strontium, and vanadium
than the same species at reference sites [Bryan et al., 2003; Burger et al., 2002; Hopkins et al.,
1997, 1998, 2000, 2006; Nagle et al, 2001; Rattner et al, 2006]. Table A-8 in Appendix A
summarizes documented examples of impacts to wildlife associated with attractive nuisances at
steam electric power plants.
In several of these instances, histopathological effects (i.e., changes in pollutant tissue
concentrations) were observed. For example, birds nesting near a combustion residual surface
impoundment produced eggs with higher selenium concentrations than eggs found at the
reference site. Although egg selenium concentrations near combustion residual surface
impoundments exceeded thresholds that
signify adverse effects on reproduction, the
study did not observe any reduction in
reproductive success [Bryan et al, 2003]. In a
study conducted by Hopkins et al. [1998],
sediment from a contaminated combustion
residual surface impoundment had arsenic
levels more than 100 times higher than the
levels found in reference site sediments. Adult
toads captured in the contaminated surface
impoundment reported a sevenfold difference
in arsenic levels between those from reference
sites [Hopkins et al, 1998]. Although the
study did not measure any indicators of Surface impoundments and constructed
reduced survival or reproductive success in the wetlands can act as attractive nuisances by
toads, the results indicate that exposure to attracting wildlife and exposing them to
combustion residual surface impoundments are elevated pollutant levels.
a potential threat [Hopkins et al., 1998].
Multiple studies have linked attractive nuisance areas at steam electric power plants to
diminished reproductive success. Field studies have documented adverse effects on reproduction
for turtles and toads living near selenium-laden combustion residual surface impoundments
[Hopkins et al., 2006; Nagle et al., 2001]. In another study, an interior least tern (Sternula
antillarum), an endangered migratory bird, began nesting at Gibson Lake, an artificial shallow
pond that receives combustion residual surface impoundment effluent from the Gibson
Generating Station in Indiana. Within several years, nearby combustion residual surface
impoundments at the Gibson Generating Station were also attracting nesting least terns, placing
these sensitive species in direct contact with steam electric power plant wastewater. To address
the attractive nuisance problem presented by the surface impoundments, the Gibson Generating
Station began a cooperative program with the Indiana Department of Natural Resources to
4-13
-------
Section 4—Assessment of Exposure Pathways
protect the nesting birds by creating a nearby alternative habitat known as the Cane Ridge
Wildlife Management Area (WMA) [Pruitt, 2000]. Cane Ridge WMA received water from
Gibson Lake and, in 2008, the U.S. Fish and Wildlife Service became concerned about selenium
levels in the water and fish present in the Cane Ridge WMA [USFWS, 2008]. Accordingly, the
bottom of Cane Ridge was plowed to redistribute and bury the selenium in the soil and the water
flowing from Gibson Lake into Cane Ridge was stopped and replaced with water piped from the
Wabash River. Duke Energy paid to stock the Cane Ridge WMA ponds with fathead minnows to
lure back migratory birds. As of June 2009, avocets, dunlins, black terns, Forster's terns, Caspian
terns, and 50 endangered least terns have returned to Cane Ridge [USFWS, 2012].
Other well-documented cases of attractive nuisance settings with characteristics (e.g.,
elevated concentrations of specific pollutants) similar to those associated with steam electric
power plants provide further support that combustion residual surface impoundments have the
potential to pose a threat to wildlife. For example, exposed organisms in attractive nuisance
settings affected by urban and agricultural wastes have exhibited elevated tissue concentrations
of pollutants, with some organisms experiencing a combination of reproductive or sublethal
effects that adversely impact their survival [Clark, 1987; Hofer et a/., 2010; King et a/., 1994;
Ohlendorf et a/., 1986, 1988a, 1988b, 1989, 1990; Tsipoura et a/., 2008]. Although these
examples do not directly relate to steam electric power plants, they highlight the potential
dangers of attractive nuisances and ability for pollutants to bioaccumulate in the surrounding
wildlife [Ohlendorf et a/., 1986, 1989, 1990]. Table A-9 in Appendix A summarizes documented
examples of impacts to wildlife associated with attractive nuisances that are not specific to steam
electric power plants.
4-14
-------
Section 5—Surface Water Modeling
SECTION 5
SURFACE WATER MODELING
Based on the documented environmental impacts discussed in the literature, EPA
identified several key environmental and human health concerns and pathways of exposure to
evaluate in the environmental assessment (EA). Environmental concerns include degradation of
surface water, sediment, and ground water quality; toxic effects on aquatic and benthic
organisms; bioaccumulation of contaminants and resultant toxic effects on wildlife; toxic effects
on humans consuming contaminated fish; and contamination of drinking water resources.
EPA focused its quantitative analyses on discharges of the evaluated wastestreams to
surface water - one of the primary exposure pathways of concern discussed in Section 4. To
quantify baseline impacts and improvements under the final steam electric effluent limitations
guidelines and standards (ELGs), EPA developed models to determine pollutant concentrations
in the immediate receiving waters, pollutant concentrations in fish tissue, and exposure doses to
ecological and human receptors from consuming aquatic organisms. This section describes the
immediate receiving water (IRW) model and the ecological risk model used in developing this
EA. Section 8 describes the development and execution of case study models using EPA's Water
Quality Analysis Simulation Program (WASP) to supplement the results of the IRW model.
5.1 IMMEDIATE RECEIVING WATER (IRW) MODEL
EPA developed the IRW model33 to quantify the environmental impacts to surface
waters, wildlife, and human health from the wastestreams evaluated for the regulatory options.
As part of this national assessment, EPA determined impacts in the immediate surface water
where steam electric power generating industry discharges occur, between 1 and 5 miles from
the outfall depending on the stream reach.34 As part of the benefits and cost analysis, EPA also
evaluated surface water concentrations downstream from steam electric discharges using EPA's
Risk-Screening Environmental Indicators (RSEI) model; see the Benefits and Cost Analysis
(EPA-821-R-15-005). The IRW model framework focused on four key areas of impacts:
• Impacts to aquatic life based on reduction in water quality from discharges of the
evaluated wastestreams.
• Impacts to aquatic life based on reduction in sediment quality from discharges of the
evaluated wastestreams.
• Impacts to wildlife from the bioaccumulation of contaminants in aquatic organisms
and fish, including piscivorous (fish-eating) wildlife.
• Impacts to human health from consuming contaminated fish.
33 The IRW model is the same model that EPA used for the national-scale analyses in support of the proposed ELGs.
EPA assigned the "IRW model" label to help distinguish the national-scale model from the case study models
developed in support of the final ELGs.
34 See the ERG memorandum "Water Quality Module: Plant and Receiving Water Characteristics" (DCN SE04513)
for details on the immediate discharge zone and length of stream reach represented.
5-1
-------
Section 5—Surface Water Modeling
As discussed in Section 4.1.2, EPA considered the type of receiving waters commonly
impacted by steam electric power plants and the pollutants typically found in the evaluated
wastestreams in selecting the appropriate methodologies for the quantitative analysis. The IRW
model quantified the environmental risks within rivers/streams and lakes/ponds/reservoirs, and
evaluated impacts from 10 toxic, bioaccumulative pollutants: arsenic, cadmium, copper,
hexavalent chromium (chromium VI), lead, mercury, nickel, selenium, thallium, and zinc. EPA's
IRW model includes three interrelated modules:
• Water quality module—calculates immediate-receiving-water-specific pollutant
concentrations in the water column and sediment and evaluates the impacts that
receiving water concentrations pose to aquatic life and human health.
• Wildlife module—evaluates the impact that sediment concentrations pose to aquatic
life, calculates the pollutant concentrations in exposed fish populations, and evaluates
the potential adverse effects to minks and eagles from consuming fish.
• Human health module—calculates non-cancer and cancer risks to human populations
from consuming fish.
Additionally, EPA used the selenium outputs from the IRW water quality module to
evaluate the risks to fish and waterfowl that consume aquatic organisms with elevated levels of
selenium (see Section 5.2). This ecological risk analysis expands on the results of the IRW
wildlife module described in this section.
The IRW water quality module uses plant-specific input data (plant-specific pollutant
loadings and cooling water flow rate),35 surface-water-specific characteristic data (e.g.., receiving
water flow rate, lake volume), and representative environmental parameters (e.g., partition
coefficients) to quantify the environmental impacts of the evaluated wastestreams to surface
waters. The module calculates pollutant concentrations in the surface water and sediment. These
concentrations are inputs to the IRW wildlife module, which calculates the bioaccumulation of
pollutants in fish tissue and determines impacts to wildlife. The fish tissue concentration
calculated in the IRW wildlife module becomes an input to the IRW human health module. This
section provides overviews of each module. Appendices C through E describe the IRW model
equations, input data, and assumed environmental parameters in further detail. The appendices
also describe the limitations and assumptions of the IRW model.
Figure 5-1 provides an overview of the IRW model inputs and the connections among the
three modules to support EPA's national-scale modeling framework.
35 EPA calculated annual pollutant loadings for the evaluated wastestreams and excluded any pollutants discharged
with other wastewaters (e.g., coal pile runoff). EPA incorporated cooling water flow rates into the IRW water
quality module on a site-by-site basis. EPA assumed no pollutant loadings were associated with cooling water
discharges to surface waters and used cooling water flow rates only to evaluate dilution effects.
5-2
-------
Section 5—Surface Water Modeling
Discharge
Loadings
(see TDD
Section 10)
Cooling Water
Flow Rate
(Steam Electric
Survey)
Default Values
Receiving Water
Characteristics
(GIS Data)
Risk to
Sediment
Biota
Water
Quality
Module
Wildlife
Module
Bioaccumulation in
Fish (Fish Tissue
Concentrations)
Risk to
Wildlife
Human
Health
Module
Risk to
» Human
Health
Risk to
Aquatic
Biota
Figure 5-1. Overview of IRW Model
5.1.1 Water Quality Module
EPA selected the steady-state equilibrium-partitioning model described in EPA's
Methodology for Assessing Health Risks Associated with Indirect Exposure to Combustor
Emissions (EPA 600-R-98-137) for the IRW water quality module. This selection was based on
three factors: 1) the model's ability to represent pollutants in the aquatic environment; 2) the
model's complexity, which EPA judged to be appropriate for a national-scale evaluation;36 and
3) the level of previous Agency and external peer reviews performed on the modeling
methodology. An equilibrium-partitioning model assumes that dissolved and sorbed pollutants in
a receiving water will quickly attain equilibrium in the immediate vicinity of the discharge point
because they dissolve or sorb in the surface water faster than they can be transported or dispersed
outside that area. The model also assumes that the equilibrium state for each pollutant can be
represented by a partition coefficient that divides the total mass of a pollutant in the waterbody
into four compartments:
• Constituents dissolved in the water column.
• Constituents sorbed onto suspended solids in the water column.
36 For a national-scale environmental assessment of over 200 receiving waters, data limitations inhibit the feasibility
of using more complex fate and transport receiving water models (dynamic or hydrodynamic) to estimate surface
water concentrations.
5-3
-------
Section 5—Surface Water Modeling
• Constituents sorbed onto sediments at the bottom of the waterbody.
• Constituents dissolved in pore water in the sediments at the bottom of the waterbody.
Table 5-1 lists the pollutants commonly found in the evaluated wastestreams with known
environmental impacts (see Section 3.1, Table 3-1). EPA selected a subset of these pollutants for
the water quality model based on the following criteria:
• The pollutant is known to be present in the evaluated wastestreams (i.e.., identified as
a pollutant of concern).
• Scientific literature documents elevated levels observed in surface waters or wildlife
from exposure to steam electric power plant wastewater.
• Partition coefficient data are available for the water quality model.
• Benchmarks are available to evaluate potential threats to wildlife or human health.
For the immediate receiving water quality analysis, EPA modeled 10 of the pollutants
shown in Table 5-1: arsenic, cadmium, chromium VI, copper, lead, mercury, nickel, selenium,
thallium, and zinc.
5-4
-------
Section 5—Surface Water Modeling
Table 5-1. Pollutants Considered for Analysis in the Immediate Receiving Water Model
Pollutant
Aluminum
Arsenic h
Boron
Cadmium
Chromium '
Copper
Iron
Lead
Manganese
Mercury J
Nickel
Selenium k
Thallium
Vanadium
Zinc
POCa
•/
•/
•/
•/
•/
•/
•/
•/
•/
•/
•/
•/
•/
•/
•/
Literature
Review b
•/
-------
Section 5—Surface Water Modeling
EPA developed the IRW water quality module in Microsoft Access™ using the
equilibrium-partition equations presented in Appendix C. The IRW water quality module is a
mathematical model used to represent the partitioning of pollutants through the surface water
after the wastestream has been discharged. The module output provides site-specific pollutant
concentrations in the water column (total, dissolved, and suspended) and sediment for 188 steam
electric power plants located across the United States that discharge to a river or stream or to a
lake, pond, or reservoir. Figure 5-2 depicts the pollutant concentrations calculated in the IRW
water quality module. EPA implemented this modeling approach through the following steps:
1. Characterize the immediate receiving water characteristics (e.g., depth of water
column, depth of waterbody, receiving water width, and flow independent mixing
value) using site-specific inputs. See the ERG memorandum "Water Quality Module:
Plant and Receiving Water Characteristics" (DCN SE04513).
2. Using the immediate receiving water characteristics, determine the fraction of
pollutant in the benthic sediment and in the water column and determine fraction of
pollutant in the water column that is dissolved.
3. Using the immediate receiving water characteristics and assumed input values,
calculate the water column volatilization rate constant, for volatile pollutants only
(i.e., mercury).
4. Calculate the water concentration dissipation rate (zero for nonvolatile pollutants).
5. Based on site-specific pollutant loadings (converting annual loadings to an average
daily loading), cooling water flow rates (for a subset of plants), and immediate
receiving water characteristics, calculate the total pollutant concentrations (e.g., total
arsenic) in the immediate receiving water, including the concentration in the water
column and in the benthic sediment.
6. Calculate the concentration of dissolved pollutant in the water column. Section 10 of
the TDD details the pollutant loadings methodology; the ERG memorandum "Water
Quality Module: Plant and Receiving Water Characteristics" (DCN SE004513)
describes the use of cooling water flow rates. Note that the pollutant loadings
included in the module do not represent the total pollutant loadings from steam
electric power plants; several wastestreams were not evaluated (e.g., stormwater
runoff, metal cleaning wastes, coal pile runoff). In addition, the module uses an
annual average discharge rate, assuming no seasonal or daily variation.
7. Quantify the number of sites that exceed the NRWQC and drinking water maximum
contaminant levels (MCLs) to evaluate the potential exposure of ecological receptors
(i.e., aquatic biota) and human receptors to toxic pollutants in the environment from
the evaluated wastestreams.
5-6
-------
Section 5—Surface Water Modeling
Mass Load Discharged
from Evaluated
Wastestreams
Cooling Water
from Plant (if
applicable)
Volatilization
*
CVKJ- Total concentration in
the water column
Cdm- Dissolved concentration
in the water column
C5W-Sorted (suspended)
concentration in the water
column
Cbs-Total concentration in
bottom (or benthic) sediment
CwTot-Total concentration in
thewaterbody (water column
and benthic sediment
combined)
3f
*
sw
A
v
Degradation
bs
WaterColumn
Bottom (Benthic)
Sediment
m
Equilibrium Reaction
Burial
where:
dbs- Depth of the benthic sediment
d^,- Depth of the water column
d - Depth of the waterbody
Pollutant concentration (C) is
calculated by the water quality
model using average annual
flow rate of the receiving water
(and cooling water flow rate
where applicable)
Source: Adapted from U.S. EPA, 1998b.
Figure 5-2. Water Quality Module: Pollutant Fate in the Waterbody
As an indicator of potential impacts, EPA compared the immediate receiving water
concentrations (under baseline and regulatory options) to the following NRWQCs:
• Freshwater acute and chronic aquatic life NRWQC.
• Human health NRWQC for the consumption of water and organisms.
• Human health NRWQC for the consumption of organisms.
EPA also compared immediate receiving water concentrations to drinking water MCLs.
EPA identified immediate receiving waters that exceeded a NRWQC or MCL as an indication of
the degradation of the overall water quality following exposure to the evaluated wastestreams.
Section 6.3 summarizes the NRWQC and MCL exceedances under baseline pollutant loadings.
Section 7.2 presents the percent reduction in number of immediate receiving waters that
potentially impact water quality under the final rule.
As with any modeling, EPA recognizes that model limitations exist and certain
assumptions need to be made. EPA used average annual pollutant loadings and normalized
effluent flow rates, which do not take into account temporal variability (e.g., variable plant
operating schedules, storm flows, low-flow events, catastrophic events). The IRW water quality
module does not account for ambient background pollutant concentrations or contributions from
other point and nonpoint sources, and assumes a constant flow rate in the receiving water based
on the annual average reported in National Hydrography Dataset Plus (NHDPlus). Appendix C
discusses these and additional module-specific limitations and assumptions and Section 6 and
5-7
-------
Section 5—Surface Water Modeling
Section 7 present the results of the IRW water quality module under baseline and regulatory
options.
5.1.2 Wildlife Module
As shown in Figure 5-1, the IRW wildlife module builds off the IRW water quality
module by using the calculated immediate receiving water and sediment concentrations to
calculate pollutant concentrations in fish populations exposed to the evaluated wastestreams and
to assess the potential to impact wildlife for the following categories:
• Impact to aquatic organisms from contact with sediment contaminated by the
evaluated wastestreams. To do this, the model quantifies the number of sites with
potential exposure of ecological receptors (i.e., sediment biota) to the pollutant in the
environment.
• Impact to piscivorous wildlife (i.e., wildlife that habitually feeds on fish) from
consuming fish impacted by the evaluated wastestreams. To do this, the model
quantifies the number of sites with potential exposure of ecological receptors (i.e..,
piscivorous wildlife) to the pollutant in the environment.
EPA developed the wildlife model in Microsoft Access™ to calculate pollutant
concentrations in fish populations exposed to the evaluated wastestreams and estimate daily
contaminant dose for wildlife receptors (i.e., minks and eagles) using equations presented in
Appendix D. EPA determined potential impacts to wildlife by comparing the concentration in the
contaminated media (i.e., water, sediment, or fish) to concentrations known to be protective of
negative impacts (i.e., benchmark). Benchmarks, which are pollutant- and endpoint-specific and
sometimes are species-specific, are an expression of the concentration level in contaminated
media that is protective against a specific endpoint (e.g., mortality). Endpoints frequently
reflected in benchmark values include sublethal effects (e.g., reduced reproduction, neurological
effects) and lethal effects. EPA implemented the wildlife modeling approach through the
following steps:
1. Compare the concentration of the contaminant in benthic sediment to the benchmark
for sediment biota.
2. Calculate the pollutant concentration in fish for trophic level three (T3) or trophic
level four (T4),37 using the calculated pollutant concentration in the water column and
the bioaccumulation factor (BAF) or bioconcentration factor (BCF).38 For mercury,
calculate the concentration of methylmercury in the fish. See Appendix D for details
on the IRW wildlife module and calculation of methylmercury concentration in fish.
3. Compare the concentration of the contaminant in the fish to the wildlife benchmarks
for ecological receptors (i.e., mink and eagle).
37 T3 fish (e.g., carp, smelt, perch, catfish, sucker, bullhead, sauger) are those that primarily consume invertebrates
and plankton, while T4 fish (e.g., salmon, trout, walleye, bass) are those that primarily consume other fish.
38 BCFs are more appropriate for use with pollutants where the primary pathway entering fish tissue is via the water,
whereas BAFs are more appropriate for pollutants where the primary pathway entering fish tissue is through a food
source (takes into account both water and diet). Where available, EPA used pollutant-specific BAFs.
-------
Section 5—Surface Water Modeling
4. Compare the baseline and regulatory option results (i.e., number of sites with
potential exposure of ecological receptors to concentrations above protective
benchmarks).
Adverse Effects to Aquatic Organisms from Contact with Sediment
EPA compared the concentration in the benthic sediment to benchmarks protective of
benthic organisms. EPA used threshold effects level (TEL) benchmarks provided in the National
Oceanic and Atmospheric Administration (NOAA) 2008 Screening Quick Reference Tables
(SQuiRTs), referred to as the chemical stressor concentration limit (CSCL), for the sediment
biota adverse impacts analysis. The CSCL is a chemical-specific media concentration that is
protective of ecological receptors of concern. The CSCL benchmark is species-specific, but can
be used to represent a community of organisms, such as amphibians or fish. Usually the most
sensitive (or lowest) CSCL for a species is used to represent the community. Table D-l in
Appendix D presents the benchmarks used for sediment exposure analysis. Section 6.2 discusses
the results of this analysis for baseline pollutant loadings.
Assessment of Pollutant Bioconcentration in Fish
EPA calculated fish tissue concentrations based on the following: 1) total water column
concentrations (i.e., dissolved plus sorbed) calculated in the IRW water quality module, and 2)
trophic-level-specific BAFs or BCFs. BAFs and BCFs are based on field and laboratory study
results compiled to develop a single factor or ratio for estimating the amount of pollutant
transferred into fish tissue at a given trophic level (i.e., rank in the food chain) based on the
pollutant concentration in the waterbody. EPA estimated fish tissue concentrations in milligrams
per kilogram (mg/kg) for T3 and T4 fish to account for the variability in fish likely consumed by
both wildlife and human receptors included in the IRW model.
Although using the total water column concentration in the bioaccumulation analysis may
overestimate the level of pollutants in the fish, it provides for a more environmentally protective
estimate of risk in the subsequent human health model because it assumes that all pollutants
within the waterbody (both dissolved and sorbed) are bioavailable to the exposed fish. The
exception to this methodology is mercury, where EPA based the fish tissue concentration
calculation on the dissolved concentration of methylmercury in the waterbody [U.S. EPA,
2005b]. Appendix D presents the BCFs and model equations for the analysis of pollutant
bioconcentration in fish tissue for T3 and T4 fish. EPA used the fish tissue concentrations to
evaluate impacts to piscivorous wildlife (see next section) and impacts to human health receptors
(see Section 5.1.3).
Impact to Piscivorous Wildlife
EPA based the piscivorous wildlife impact analysis on the methodology outlined in the
2008 U.S. Geological Survey (USGS) study Environmental Contaminants in Freshwater Fish
and Their Risk to Piscivorous Wildlife Based on a National Monitoring Program. The study
examined the impacts to minks and eagles from eating contaminated fish. Minks and eagles are
commonly used in ecological risk assessments as indicator species for potential impacts to fish-
eating mammals and birds in areas contaminated with bioaccumulative pollutants [USGS, 2008].
Minks and eagles are appropriate receptors for the steam electric power plant wildlife impact
5-9
-------
Section 5—Surface Water Modeling
analysis because their habitats span most of the country and their diet largely consists of adult
fish from the two trophic levels (i.e., T3 and T4 fish) included in the IRW wildlife module.
According to the literature [U.S. EPA, 1998a], minks consume mostly T3 fish, while eagles
consume mostly T4 fish. EPA evaluated the potential adverse effects to minks and eagles for
nine pollutants commonly found in the wastestreams of interest: arsenic, cadmium, chromium,
copper, mercury, nickel, lead, selenium, and zinc.39 The USGS method [USGS, 2008] is a
wildlife impact analysis using NOAELs (no-observed-adverse-effect levels), which were derived
from adult dietary exposure or tissue concentration studies based primarily on reproductive
endpoints. The study calculated a NEHC benchmark, which is based on the NOAEL, the food
consumption rate, and/or the biomagnification factor of each receptor. The report states that
piscivorous wildlife may be at an elevated risk for reduced reproduction rates if the measured
pollutant concentration in fish exceeds the NEHC. Therefore, EPA compared the mink-specific
and eagle-specific NEHC values from the USGS study with the T3 and T4 fish tissue
concentrations, respectively, to identify potential adverse impacts to the ecological receptors. In
the piscivorous wildlife analysis, a benchmark exceedance indicates that piscivorous mammals
or birds exposed to fish in the immediate receiving water of interest are at an elevated risk for
reduced reproduction rates or other health effects.
Table D-3 in Appendix D presents the NEHC values used to evaluate potential adverse
effects to wildlife. The text of Appendix D presents the equations used to compare model outputs
to benchmarks (NEHCs), along with model-specific limitations and assumptions. The results of
the IRW wildlife module under baseline conditions and the final rule are included in Section 6
and Section 7, respectively.
5.1.3 Human Health Module
As shown in Figure 5-1, the IRW human health module builds off the IRW wildlife
module, using the calculated T3 and T4 fish tissue concentrations. Its purpose is to evaluate the
cancer risk and potential to cause non-cancer health effects from consuming fish within the
following age and consumption categories:
• Child recreational fishers (six cohorts covering different age ranges).40
• Child subsistence fishers (six cohorts covering different age ranges).
• Adult recreational fishers.
• Adult subsistence fishers.
In addition, EPA evaluated potential impacts to different race populations using these
same cohorts as part of its environmental justice analysis. See the Regulatory Impact Analysis for
the Effluent Limitations Guidelines and Standards for the Steam Electric Power Generating
Point Source Category (RIA) (EPA-821-R-15-004).
40
39 Because there are no benchmarks for chromium VI or methylmercury, EPA used the total chromium and total
mercury benchmarks, respectively, which may underestimate the risk to wildlife.
The child cohort age ranges correspond to the ranges provided in the 2008 Child-Specific Exposure Factors
Handbook (EPA-600-R-06-096F) for body weights.
5-10
-------
Section 5—Surface Water Modeling
EPA developed the IRW human health module in Microsoft Access™ to estimate the
daily pollutant doses for human receptors as a result of eating T3 and T4 contaminated fish. EPA
used a mathematical model to estimate the potential threats to human receptors from pollutant
exposure. EPA estimated the average concentration of pollutants in a fish fillet consumed by
humans based on a consumption diet of 36 percent T3 and 64 percent T4 fish (see Appendix E).
The IRW human health module then calculates the daily dose of pollutants from fish
consumption for each cohort included in the analysis. EPA varied the fish consumption rate
based on the specific cohort using two factors: 1) type of fisher (recreational or subsistence) and
2) age (adult and six child cohorts). EPA first evaluated human health impacts based on type of
fisher and age of cohort using national-level consumption rates. For the environmental justice
analysis, EPA determined fish consumption rates using the race population in addition to the
other two factors. See Appendix E for further details. Using the fish consumption rate, EPA
determined an average daily pollutant dose for each human cohort evaluated. Table E-2 in
Appendix E presents the cohorts included in the IRW human health module and the
corresponding fish consumption rates used in the module. EPA implemented the human health
modeling approach through the following steps:
1. Calculate the pollutant concentration in a fish fillet.
2. Calculate the average daily dose of pollutant from fish consumption by each receptor
cohort (used for comparison to reference dose [RfD] values).
3. Calculate the lifetime average daily dose (LADD) for carcinogenic pollutants only, by
each receptor cohort (used to determine cancer risk).
4. Calculate the lifetime excess cancer risk (LECR) for carcinogenic pollutants only, by
each receptor cohort, using the LADD.
5. Compare the exposure doses of human receptor cohorts to appropriate benchmarks
(RfD and selected cancer benchmark: 1-in-a-million).
6. Compare the baseline and regulatory option results: reduction in the number of
immediate receiving waters with exposure doses from consuming fish that pose a
potential threat to human receptors.
Non-Cancer Threat to Human Receptors
EPA evaluated the non-cancer threat (e.g., reproductive or neurological impacts) to each
cohort by comparing the pollutant-specific average daily dose values for fish consumption to the
corresponding RfDs. EPA evaluated non-cancer risks for the following pollutants: inorganic
arsenic,41 cadmium, chromium VI, copper, methylmercury, nickel, selenium, thallium, and zinc.
Table E-3 in Appendix E presents the RfD values used in the non-cancer threat analysis. RfD
values are an expression of the consumption dose that is protective against a specific endpoint.
41 For this analysis, EPA used only the concentration of inorganic arsenic for the human health impact assessment.
Based on the literature review, arsenic in fish is mostly in the organic form and is not considered harmful. The
wildlife model calculates a total arsenic fish tissue concentration. To convert this number to inorganic arsenic, EPA
assumed that 4 percent of the total arsenic is inorganic based on EPA's 1997 document Arsenic and Fish
Consumption (EPA-822-R-97-003). The 1997 document reported that the inorganic arsenic concentration in fish is
between 0.4 and 4 percent of the total arsenic accumulating in fish [U.S. EPA, 1997b].
5-11
-------
Section 5—Surface Water Modeling
Endpoints frequently reflected in RfDs include various immunological, reproductive,
neurological, and other non-cancer effects. In the IRW human health module, when the RfD is
exceeded, it indicates a potential threat to humans for the endpoint associated with the RfD. For
example, exceeding the RfD for selenium indicates that the exposure dose from fish consumption
can cause non-cancer health effects, such as selenium-induced liver dysfunction or selenosis
(hair or nail loss, morphological changes of the nails, etc.) [U.S. EPA, 201 lc].
Cancer Risk to Human Receptors
Arsenic is the only pollutant included in the IRW model for which EPA has derived a
cancer slope factor for ingestion exposures.42 The IRW human health module calculates the
LADD for each receptor cohort based on an exposure duration (i.e., length of time a receptor is
in contact with the carcinogen) averaged over a lifetime (i.e., 70 years). For this analysis, EPA
assumed the exposure duration to be equal to the number of years represented by each cohort.
Using these exposure durations is appropriate for screening-level estimates of cancer risk and for
comparing changes between baseline and regulatory options.43 The model then multiplies the
LADD by the cancer slope factor to calculate the LECR from arsenic. LECR is an estimate of the
increase in cancer risk resulting from an exposure (i.e., consumption of contaminated fish). EPA
used the benchmark value for evaluating cancer risk of 1-in-a-million people. Therefore, a
calculated LECR greater than 1 x 10"6 indicates an increased cancer risk for humans that
consume fish exposed to discharges of evaluated wastestreams.
5.2 ECOLOGICAL RISK MODELING
Selenium bioaccumulation in aquatic organisms occurs primarily from ingesting food
rather than through direct exposure to dissolved selenium in the water column [Fan et al., 2002;
Ohlendorf etal, 1986; Saiki and Lowe, 1987; Presser and Ohlendorf, 1987; Luoma et al., 1992;
Presser et al., 1994; Chapman et al., 2009]. Unlike other bioaccumulative contaminants such as
mercury, the single largest step in selenium accumulation in aquatic environments occurs in
aquatic organisms at the base of the food web; algae, particulates, and microorganisms can
accumulate selenium to levels far greater than the concentration in the water column.
Bioaccumulation and transfer through aquatic food webs constitute the major selenium exposure
pathway in aquatic ecosystems.
Macrophytes, algae, phytoplankton, zooplankton, and macroinvertebrates at the base of
the food web easily bioaccumulate selenite and selenate and incorporate selenium in tissues as
selenomethionine, an organo-selenide. This selenomethionine is then released back to the water
42 Although EPA determined that lead and lead compounds can be "reasonably anticipated to be human
carcinogens," no numeric value has been determined to quantify the cancer risk. As stated on the IRIS website,
"quantifying lead's cancer risk involves many uncertainties, some of which may be unique to lead. Age, health,
nutritional state, body burden, and exposure duration influence the absorption, release, and excretion of lead. In
addition, current knowledge of lead pharmacokinetics indicates that an estimate derived by standard procedures
would not truly describe the potential risk. Thus, the Carcinogen Assessment Group recommends that a numerical
estimate not be used." (See http://www.epa.gov/iris/subst/0277.htm#reforal.)
43 To completely assess risk to an individual, EPA recommends that risks should be calculated by integrating
exposures throughout all life stages (i.e., adding multiple cohort risks from screening analysis). For example, the
exposure duration may be equal to the length of time a person lives in an area [U.S. EPA, 201 lb].
5-12
-------
Section 5—Surface Water Modeling
column as these plants and organisms die or are consumed [U.S. EPA, 2014f]. In general,
selenium concentrations in particulates (e.g., sediment, detritus, and primary producers such as
algae and biofilm) are 100 to 500 times higher than dissolved concentrations in selenate-
dominated environments such as streams and rivers. Where selenite or organo-selenide is
proportionately more abundant, such as in lakes, wetlands, some estuaries, and oceans, the ratio
can be much higher (1,000 to 10,000 times higher than dissolved concentrations). This variability
of particulate concentrations relative to dissolved concentrations across different aquatic
environments makes it difficult to develop a simple relationship between the concentration of
selenium in water and the concentration of selenium in organisms [Presser and Luoma, 2010].
The scientific community has devoted significant effort to understanding the mechanisms
of selenium bioaccumulation. The preferred approach, as described in Presser and Luoma
[2010], accounts for the variability in particulate concentrations described above by applying
site-specific enrichment factors (EFs) that represent the ratio of the concentration of selenium at
the base of the food web (i.e., parti culates) to the dissolved concentration in water. Subsequent
bioaccumulation by aquatic organisms is described through a series of empirically derived,
species-specific trophic transfer factors (TTFs) that link the selenium concentrations in
particulates and invertebrates to higher trophic-level organisms such as fish and birds. TTFs can
be derived from laboratory experiments or from field data. TTFs differ from traditional BCFs
(described in Section 5.1.2) in that they are the ratio of the selenium concentration in each animal
to the selenium concentration in its food, whereas BCFs represent the ratio of the selenium
concentration in an animal to the selenium concentration in the water of its environment. Using
TTFs therefore more accurately predicts selenium bioaccumulation in aquatic organisms because
it accounts for the significant role of dietary exposure.
Selenium toxicity among exposed fish and birds primarily is transferred to the eggs and
demonstrated via subsequent reproductive effects. Many studies and expert panels have shown
that reproductive effects, linked to egg-ovary selenium concentrations, are of greatest concern
and likely have led to observed reductions in sensitive fish species populations in waterbodies
having excessive selenium concentrations [Chapman et a/., 2009].
EPA developed and applied a probabilistic ecological risk model, based on the
bioaccumulation concepts described above, to assess the risk of adverse reproductive impacts
among fish and birds exposed to selenium in waterbodies that receive discharges of the evaluated
wastestreams. Figure 5-3 provides a general schematic of the approach, which follows these
general steps:
1. Apply a distribution of site-specific EFs (with separate distributions for lentic and
lotic systems) to the predicted dissolved selenium concentration from the IRW
water quality module, resulting in a distribution of predicted selenium
concentrations in particulates and primary producers for each receiving water.
2. Apply a TTF distribution for invertebrates (TTFinvert) to the outputs from Step 1,
resulting in a distribution of predicted selenium concentrations in invertebrates
that inhabit each receiving water.
3. To predict the bioaccumulation and reproductive risk among fish:
5-13
-------
Section 5—Surface Water Modeling
a. Apply a TTF distribution for fish (TTFflsh) to the outputs from Step 2,
resulting in a distribution of predicted selenium concentrations in the eggs
and ovaries of fish that inhabit each receiving water (some of the TTFs
incorporate tissue conversion factors to translate the outputs from whole
body or muscle concentrations into fish egg-ovary concentrations).
b. Apply an exposure-response function for fish (ERflsh) to the outputs from
Step 3a, resulting in a distribution showing the probability of a decline in
reproductive success across exposed fish populations.
4. To predict the bioaccumulation and reproductive risk among birds (specifically,
mallards):
a. Apply a TTF distribution for mallards (TTFmanard) to the outputs from Step
2, resulting in a distribution of predicted selenium concentrations in the
eggs of mallards that forage and/or breed in each receiving water.
b. Apply an exposure-response function for mallards (ERmaiiard) to the
outputs from Step 4a, resulting in a distribution showing the probability of
a decline in reproductive success across exposed mallard populations.
This modeling approach is consistent with the approach taken in developing the External
Peer Review Draft Aquatic Life Ambient Water Quality Criterion for Selenium - Freshwater
[U.S. EPA, 2014f] (referred to as the external peer review draft selenium criterion) and is based
on the same data sets and studies for EF, TTFinvert, TTFflsh, and ERflsh. For this EA, EPA
expanded the model to include data sets for TTFmanard and ERmanard and to include several
additional data sets and studies for EF, TTFinvert, TTFflsh, and ERflsh that were eventually
incorporated into the Draft Aquatic Life Ambient Water Quality Criterion for Selenium -
Freshwater [U.S. EPA, 2015b].
5-14
-------
Section 5—Surface Water Modeling
Fish
Reproductive
Risk
Mallard
Reproductive
Risk
Step 3b:
ER functii
fish
1
for *
t
Selenium
Concentration in
Fish Eggs/Ovaries
ST\
i
«
Selenium
Concentration in
Mallard Eggs
Apply
^Jiw
Step 4b:
ER functk
mallards
Step 3a:
Apply TTF distribution
for fish
(and tissue conversion
factors as necessary)
Step 2:
Apply TTF distribution
for invertebrates
Step 1:
Apply EF distribution
(separate distributions
for lentic and lotic
receiving waters)
Selenium
Concentration in
Invertebrates
Step4a:
Apply TTF distribution
for mallards
Selenium
Concentration in
Particulates and
Primary Producers
Dissolved Selenium
Concentration in
Water Column
Acronyms
EF - Enrichment factor
ER- Exposure-response
IRW - Immediate receiving water
TTF - Trophic transfer factor
IRW Model
Water
Quality
Module
indicates that output
is a distribution
Figure 5-3. Flowchart of Selenium Ecological Risk Model
5-15
-------
Section 5—Surface Water Modeling
Detailed information for some of the factors that influence selenium bioaccumulation at a
particular site, such as the form of selenium in the environment (e.g., selenate, selenite, and
organo-selenide) and the structure of the aquatic food web, is not available across the 209
immediate receiving waters modeled in this EA. The ecological risk model accounts for these
unknowns by applying distributions of EFs and TTFs based on data representing a wide variety
of lentic and lotic waterbodies and freshwater invertebrate and fish species, rather than relying
on a single statistical measure (e.g., mean or median) for those parameters. This approach
accounts for the variability across aquatic systems and captures the full range of food web
constructs that could occur in these receiving waters.
The remainder of this section further discusses EPA's development of the EFs, TTFs, and
ER functions in the ecological risk model and use of those functions to calculate risk of adverse
reproductive effects (performed using Oracle Crystal Ball software). Appendix F provides
additional details regarding data sources, data acceptance criteria, statistical methods, and
assumptions and limitations of the ecological risk model.
Enrichment Factors
EPA compiled a database of empirical measurements of selenium concentration (water,
sediment, biofilm, algae, phytoplankton, and detritus) from relevant field studies across a range
of aquatic systems. EPA then calculated EFs for a set of aquatic systems and applied statistical
methods to distinguish categories with similar bioaccumulation characteristics, consistent with
the approach followed in developing the external peer review draft selenium criterion [U.S. EPA,
2014f]. The key factor distinguishing EFs across systems is whether the data were collected from
lentic systems (e.g., lakes, reservoirs, and ponds) or lotic systems (e.g., rivers, creeks, and
streams). Therefore, the EPA developed EF distributions separately for lentic and lotic systems.
This effort produced EF distributions for both systems that are well described by
lognormal distributions with means (standard deviations) of 1,738 (2,499)44 for lentic systems
and 692 (787) for lotic systems.
Trophic Transfer Factors for Invertebrates and Fish
EPA compiled a database of empirical measurements of selenium concentration in
particulates, invertebrates, and fish from relevant field studies. EPA arranged the data by
developing data pairs representing the concentration in the consumer organism (invertebrate or
fish) and the concentration in the consumed material or lower-trophic-level organism (particulate
or invertebrate). The ratio between these two values defines the TTF for the consumer organism.
EPA limited these data pairs to measurements collected from the same aquatic site. EPA further
limited the data pairs by excluding measurements of material or lower-trophic-level organisms
deemed unlikely to be ingested by the higher-trophic-level organism. Many of the fish
concentration measurements required a further conversion to the concentration of selenium in
eggs, requiring a whole-body-to-egg/ovary conversion factor. This factor (egg/ovary
concentration = whole body concentration x 1.9) is based on paired measurements from
44 The EF incorporates a multiplier of 1,000. A mean EF of 1,738 for lentic systems indicates that, on average, the
concentration of selenium at the base of the food web is 1.738 times greater than the dissolved concentration in
water.
5-16
-------
Section 5—Surface Water Modeling
individual fish and is consistent with the value used to develop the external peer review draft
selenium criterion [U.S. EPA, 2014f].
This effort resulted in a TTFinvert distribution with a mean (standard deviation) of 2.84
(2.49) and a TTFfish distribution with a mean (standard deviation) of 1.6 (1.08).
Trophic Transfer Factors for Mallards
EPA selected the mallard (Anas platyrhynchos) as the representative bird species for the
ecological risk analysis. The mallard has been extensively evaluated in both field and laboratory
studies and has been shown to be relatively sensitive to selenium. Mallards are ubiquitous,
occurring in every state at specific times during the year, and are the species with the highest
probability of being found at any of the 209 modeled receiving waters. Dabbling ducks such as
mallards contribute important ecosystem services, such as transferring eggs and seeds of aquatic
organisms between isolated wetlands and maintaining the biodiversity of other organisms
[Bengtsson etal, 2014; Green and Elmberg, 2014].
Based on a review of Ohlendorf [2003], EPA developed a database of field measurements
of mallards and their likely food sources, expressed as a ratio of measured egg concentrations to
dietary concentrations. Many studies across a wide variety of species have shown that selenium
concentrations in bird eggs range from roughly equal to or three or four times the concentrations
in the diet of the female at the time of egg-laying [Ohlendorf and Heinz, 2011]. The resulting
TTFmaiiard distribution is best described by a triangular distribution, with a likeliest value of 2.5, a
minimum value of 0.4, and a maximum value of 4.1.
Exposure-Response Function for Fish
Larval mortality and reproductive teratogenesis (i.e., deformities in offspring) from
maternal transfer of selenium to eggs represent the most sensitive endpoints in fish. Deformities
in fish that affect feeding or respiration can be lethal shortly after hatching. Deformities that are not
directly lethal, but that distort the spine and fins, can affect larval survival by reducing swimming
ability and overall fitness. EPA therefore selected larval mortality and deformities as the target
endpoints for this analysis.
This approach is consistent with the approach taken to develop the external peer review
draft selenium criterion, and used the same extensively peer-reviewed exposure-response
function (i.e., curve) as was used in that analysis [U.S. EPA, 2014f]. Appendix F provides the
exposure-response function for fish, which translates the modeled egg-ovary concentration into
the probability of adverse reproductive effects.
Exposure-Response Function for Mallards
To derive the exposure-response function for mallards, EPA used the same set of six
progressive studies used to develop the TTFmanard distribution [Ohlendorf, 2003]. This approach
ensures consistency in the predicted bioaccumulation and reproductive response across different
selenium exposure levels.
5-17
-------
Section 5—Surface Water Modeling
The mallard exposure-response function in Ohlendorf [2003] is based on a regression
meta-analysis of six different laboratory studies that evaluated the effect of selenium on mallard
egg hatchability [Heinz etal, 1987, 1989; Heinz and Hoffman, 1996, 1998; Stanley etal, 1994,
1996]. This function formed the basis of the water quality criterion adopted by the Utah Water
Quality Board for Lake Gilbert, and underwent peer review by EPA Region 8. For this analysis,
EPA fit a logistic curve to the combined, control normalized data from the six mallard studies.
Appendix F provides the resulting exposure-response function for mallards.
Calculation of Reproductive Risk
In this analysis, risk is defined as the probability of a percentage reduction in
reproductive capacity based on larval mortality and deformity in fish and hatching success in
mallards. For any given exposure concentration to selenium predicted from the EF-TTF model,
the exposure-response function provides the probability of the effect occurring, termed a joint
probability model.
The EF-TTF models provide the predicted exposure distributions in fish and mallard
eggs. For each concentration, the probability of exposure occurring is compared to the
probability of effect at that exposure level. The resulting functions provide the probability of
larval mortality and deformities in fish and hatching failure in mallards.
5-18
-------
Section 6—Current Impacts from Steam Electric Power Generating Industry
SECTION 6
CURRENT IMPACTS FROM STEAM ELECTRIC POWER
GENERATING INDUSTRY
EPA developed the immediate receiving water (IRW) model and ecological risk model
described in Section 5 to quantify the current national-scale environmental impacts of direct
surface water discharges of the evaluated wastestreams (i.e., flue gas desulfurization (FGD)
wastewater, fly ash transport water, bottom ash transport water, and combustion residual
leachate) from steam electric power plants. This section presents the baseline results of the
modeled pollutant concentrations in surface waters and fish tissue and their potential impacts to
aquatic life, wildlife, and human health.
6.1 WATER QUALITY IMPACTS
The quality of a surface water is defined by its chemical, physical, and biological
characteristics and is measured to evaluate a water's potential to harm aquatic life and human
health. EPA assessed the quality of surface waters that receive discharges of the evaluated
wastestreams by comparing estimated pollutant concentrations in the water column to the
National Recommended Water Quality Criteria (NRWQC) and drinking water maximum
contaminant levels (MCLs). Based on the modeling results for surface water quality impacts,
approximately 62 percent of the lakes, ponds, and reservoirs (16 out of 26) and 43 percent of the
rivers and streams (78 out of 183) that receive discharges of the evaluated wastestreams have
estimated pollutant concentrations that exceed these water quality benchmarks and may have
quantifiably impaired water quality due to those discharges. Based on the modeling results,
human health criteria exceedances are more prevalent among the immediate receiving waters
than aquatic life criteria exceedances. Approximately 17 to 45 percent of the immediate
receiving waters had modeled pollutant concentrations that exceed a human health criterion,
while approximately 4 to 17 percent of the immediate receiving waters had modeled pollutant
concentrations that exceed an aquatic life criterion. The difference between exceedances for
human health and aquatic life criteria is due to the human health criteria for arsenic and thallium,
which are significantly lower than the aquatic life criteria for most of the modeled pollutants.
Due to data limitations at the national scale, EPA did not include other pollutant sources
(e.g., naturally -occurring pollutants, nonpoint source discharges, or other point source
discharges) in the IRW model. Quantified exceedances estimated by the IRW model represent
environmental impacts due entirely to the pollutant loadings in discharges of the evaluated
wastestreams from steam electric power plants. Table 6-1 presents the number and percentage of
immediate receiving waters with estimated pollutant concentrations that exceed each water
quality criterion under baseline conditions.
EPA identified arsenic, thallium, cadmium, and selenium as the primary pollutants
contributing to the water quality exceedances, as shown in Table 6-1. Humans are primarily at
risk for exposure to arsenic and thallium. Out of the 209 modeled immediate receiving waters:
• 94 exceed the human health NRWQC for the consumption of arsenic-contaminated
water and organisms (0.018 micrograms per liter (|ig/L)).
6-1
-------
Section 6—Current Impacts from Steam Electric Power Generating Industry
• 65 exceed the arsenic NRWQC for consumption of organisms only (0.14 |ig/L).
• 49 exceed the human health NRWQC for the consumption of thallium-contaminated
water and organisms (0.24 |ig/L).
• 45 exceed the thallium NRWQC for consumption of organisms only (0.47 |ig/L).
Therefore, humans consuming water and/or organisms inhabiting these waters are more
at risk of arsenic-related effects (skin damage, cardiovascular disease, and cancer in the skin,
lungs, bladder, and kidney) and thallium-related effects (changes in blood chemistry; damage to
liver, kidney, and intestinal and testicular tissues; hair loss; and reproductive and developmental
damage).
Aquatic organisms are primarily at risk due to exposure to cadmium and selenium.
Estimated pollutant concentrations in approximately 15 percent of the immediate receiving
waters (29 and 33 out of 209, respectively) exceed the aquatic life criterion for chronic exposure
to cadmium- and selenium-contaminated waters (0.25 and 5 |ig/L, respectively). Therefore,
aquatic organisms inhabiting these waters are under a greater threat for cadmium-related effects
(tissue damage and organ abnormalities) and selenium-related effects (reproductive failure,
deformities, reduced growth, increased metabolic rates, and death). Sublethal and lethal impacts
from chronic selenium exposure are frequently cited in literature. For more information on these
impacts, refer to Section 3.1.1.
Table 6-1. Number and Percentage of Immediate Receiving Waters with Estimated
Water Concentrations that Exceed the Water Quality Criteria at Baseline
Evaluation Criterion
Aquatic
Life
Criteria
Human
Health
Criteria
Freshwater Acute NRWQC
Freshwater Chronic NRWQC
Human Health Water and
Organism NRWQC
Human Health Organism Only
NRWQC
Drinking Water MCL
Total Number of Unique Immediate
Receiving Waters °
Number of Immediate Receiving Waters Exceeding a
Criterion a
Number of
Rivers and
Streams
9
30
78
55
31
78
Number of
Lakes, Ponds,
and
Reservoirs
0
5
16
11
5
16
Total Immediate Receiving
Waters b
Number
Exceeding
9
35
94
66
36
94
Percentage
Exceeding
4%
17%
45%
32%
17%
45%
Sources: ERG, 2015d; ERG, 2015h; ERG, 20151.
Acronyms: NRWQC (National Recommended Water Quality Criteria); MCL (maximum contaminant level).
a - The EA encompasses a total of 222 immediate receiving waters and loadings from 195 steam electric power
plants (some of which discharge to multiple receiving waters). The IRW model, which excludes the Great Lakes and
estuaries, encompasses a total of 209 immediate receiving waters (183 rivers and streams; 26 lakes, ponds, and
reservoirs) and loadings from 188 steam electric power plants.
b - These values are the sum and percentage of rivers, streams, lakes, ponds, and reservoirs impacted.
c - This represents the number of unique immediate receiving waters that exceeded at least one criterion.
6-2
-------
Section 6—Current Impacts from Steam Electric Power Generating Industry
Table H-l in Appendix H presents additional details on the number and percentage of
immediate receiving waters that are exceeding each water quality criterion by pollutant. For
more detailed information on the modeled immediate receiving water concentrations under
baseline conditions, see Figures H-l to H-l0 and Tables H-2 to H-l 1 in Appendix H.
6.2 WILDLIFE IMPACTS
As part of the national-scale wildlife impacts analysis, EPA assessed the impacts of the
evaluated wastestreams on the following categories:
• Impacts to wildlife indicator species (i.e., mink and eagle) due to consuming
contaminated fish (using the wildlife component of the IRW model).
• Impacts to fish and waterfowl due to dietary exposure and trophic transfer of
selenium (using the ecological risk model in combination with the water quality
component of the IRW model).
• Impacts to benthic organisms due to contact with contaminated sediment (using the
wildlife component of the IRW model).
The results of these analyses are described in the following sections.
6.2.1 Impacts to Wildlife Indicator Species
As described in Section 5.1.2, EPA assessed the potential impact to piscivorous wildlife
from the evaluated wastestreams by modeling fish tissue pollutant concentrations and comparing
these concentrations to no effect hazard concentrations (NEHC) for minks and eagles developed
by the U.S. Geological Survey (USGS). Based on the estimated fish tissue concentrations,
approximately 34 percent (71 out of 209) and 28 percent (58 out of 209) of the immediate
receiving waters pose a potential threat to eagles and minks, respectively, through the
consumption of contaminated fish. This result demonstrates that estimated pollutant
concentrations in fish that inhabit receiving waters immediately downstream from steam electric
power plant wastewater discharges pose a potential reproductive threat to surrounding minks and
eagles and indicates the potential broader impacts that steam electric power plant wastewater
discharges may pose to the greater environment as pollutants transfer from the aquatic
environment and begin to accumulate in terrestrial food webs.
As expected, based on documented environmental impacts, modeling results indicate that
pollutant concentrations in fish inhabiting lakes, ponds, and reservoirs are more likely to exceed
the NEHC benchmarks than pollutant concentrations in fish inhabiting rivers and streams. The
estimated fish tissue pollutant concentrations pose a potential reproductive threat to minks and
eagles in approximately 46 percent of modeled lakes, ponds, and reservoirs (12 out of 26) and in
32 percent of rivers and streams (59 out of 183) that were evaluated. These results are expected,
since fish populations inhabiting lake environments cannot travel to uncontaminated waters and
therefore continue to bioaccumulate pollutants.
Table 6-2 presents the number and percentage of immediate receiving waters that exceed
the USGS wildlife fish consumption NEHC for minks and eagles.
6-3
-------
Section 6—Current Impacts from Steam Electric Power Generating Industry
Table 6-2. Number and Percentage of Immediate Receiving Waters That Exceed Wildlife
Fish Consumption NEHCs for Minks and Eagles (by Waterbody Type) at Baseline
Evaluation Criterion
Mink fish consumption NEHC
Eagle fish consumption NEHC
Total Number of Unique
Immediate Receiving Waters °
Number of
Rivers and
Streams
47
59
59
Number of
Lakes, Ponds,
and Reservoirs
11
12
12
Total Receiving Waters a'b
Number
Exceeding
58
71
71
Percentage
Exceeding
28%
34%
34%
Sources: ERG, 2015d; ERG, 2015h; ERG, 20151.
Acronyms: NEHC (No Effect Hazard Concentration).
a - The EA encompasses a total of 222 immediate receiving waters and loadings from 195 steam electric power
plants (some of which discharge to multiple receiving waters). The IRW model, which excludes the Great Lakes and
estuaries, encompasses a total of 209 immediate receiving waters (183 rivers and streams; 26 lakes, ponds, and
reservoirs) and loadings from 188 steam electric power plants.
b - These values are the sum and percentage of rivers, streams, lakes, ponds, and reservoirs impacted.
c - This represents the number of unique immediate receiving waters that exceed a criterion.
The pollutants found to present the greatest threat to minks and eagles from fish
consumption were mercury and selenium. The modeled concentrations of mercury in fish tissue
exceeded the NEHC benchmarks for minks and eagles in 26 and 34 percent of the modeled
immediate receiving waters, respectively. Approximately 20 percent of the immediate receiving
waters contained fish with modeled selenium concentrations exceeding a fish consumption
NEHC benchmark for minks and eagles.
Table 6-3 presents the number and percentage of immediate receiving waters that exceed
a USGS wildlife fish consumption NEHC for minks and eagles by pollutant.
6.2.2 Impacts to Fish and Waterfowl due to Dietary Selenium Exposure
As discussed in Section 5.2, EPA expanded upon the piscivorous wildlife benchmark
analysis to include ecological risk modeling of the reproductive risks among fish and waterfowl
that consume aquatic organisms contaminated with elevated levels of selenium. Selenium is of
particular concern in aquatic environments because it can accumulate in sediment and
biomagnify to toxic levels in fish inhabiting selenium-contaminated waters (even at relatively
low concentrations), potentially eliminating piscivorous (fish-eating) wildlife higher in the food
chain [Ohlendorf et a/., 1988a]. Impacts to fish populations are well documented in the literature
[Garrett and Inman, 1984; Lemly, 1985a; Sorensen etal., 1982]. While exposed fish populations
may not experience lethal impacts, the sublethal damage to their reproductive systems can
eventually impact the survivability of fish populations near steam electric power plants. The
documented impacts at Belews Lake illustrate this is especially an issue in lakes, ponds, and
reservoirs, where healthy fish populations cannot migrate and seek out alternative food sources.
Decreased fish populations may cause cascading effects within the food web that can adversely
affect other organisms in the ecosystem.
6-4
-------
Section 6—Current Impacts from Steam Electric Power Generating Industry
Table 6-3. Number and Percentage of Immediate Receiving Waters That Exceed Wildlife
Fish Consumption NEHCs for Minks and Eagles (by Pollutant) at Baseline
Pollutant
Arsenic
Cadmium
Chromium VI
Copper
Lead
Mercury
Nickel
Selenium
Thallium
Zinc
Mink
Fish
Consumption
NEHC
(MS/g)a
7.65
5.66
17.7 c
41.2
34.6
0.37
12.5
1.13
ID
904
Immediate Receiving
Waters
Number
Exceeding b
0
6
0
1
1
55
0
42
NC
1
Percentage
Exceeding
0%
3%
0%
<1%
<1%
26%
0%
20%
NC
<1%
Eagle
Fish
Consumption
NEHC
(MS/g)a
22.4
14.7
26.6 c
40.5
16.3
0.5
67.1
4
ID
145
Immediate Receiving
Waters
Number
Exceeding b
0
4
0
1
2
71
0
42
NC
5
Percentage
Exceeding
0%
2%
0%
<1%
1%
34%
0%
20%
NC
2%
Sources: ERG, 2015d; ERG, 2015h; ERG, 20151.
Acronyms: ID (Insufficient data; no benchmarks were identified in the wildlife analysis for thallium); NC (Not
calculated); NEHC (No Effect Hazard Concentration); ug/g (micrograms/gram).
a - The wildlife fish consumption NEHC represents the maximum pollutant concentration in the fish that will result
in no observable adverse effects in wildlife (/'. e., minks or eagles) [USGS, 2008].
b - The EA encompasses a total of 222 immediate receiving waters and loadings from 195 steam electric power
plants (some of which discharge to multiple receiving waters). The IRW model, which excludes the Great Lakes and
estuaries, encompasses a total of 209 immediate receiving waters (183 rivers and streams; 26 lakes, ponds, and
reservoirs) and loadings from 188 steam electric power plants.
c - An NEHC benchmark is not available for chromium VI; therefore, EPA used the total chromium benchmark.
The results of the ecological risk model indicate that, under baseline conditions,
discharges of selenium from steam electric power plants elevate the risk of adverse reproductive
impacts among fish and mallards that inhabit, forage, or breed in the immediate receiving waters.
These reproductive impacts include larval mortality and deformities among fish and reduced egg
hatchability among mallards.
The ecological risk modeling results indicate that 15 percent of the lakes, ponds, and
reservoirs (four out of 26) and 11 percent of the rivers and streams (20 out of 183) that receive
discharges of the evaluated wastestreams present an elevated risk of negative reproductive
impacts to fish. For mallards, the counts are slightly higher, with 19 percent of the lakes, ponds,
and reservoirs (five out of 26) and 14 percent of the rivers and streams (26 out of 183) presenting
these risks. These results support the conclusion that lentic systems, which have higher potential
for pollutant retention due to longer residence times, are more likely to experience ecological
impacts due to discharges from steam electric power plants.
The results described above represent those immediate receiving waters whose median
modeled egg/ovary concentration is predicted to impact reproduction among at least 10 percent
of the exposed fish or mallard population. As described below, however, adjusting these criteria
reveals additional perspective regarding the prevalence of immediate receiving waters that may
be causing reproductive impacts due to selenium exposure.
6-5
-------
Section 6—Current Impacts from Steam Electric Power Generating Industry
Selecting the 90th percentile modeled egg/ovary concentration, meaning there is a 10
percent probability that the egg/ovary concentrations are greater than the selected concentration,
reveals that 20 percent of the immediate receiving waters (42 out of 209) present reproductive
risks to at least 10 percent of the exposed fish population. The results for mallards (21 percent)
are very similar. These counts are considerably higher than the results obtained using the median
modeled egg/ovary concentration, indicating the potential for more widespread ecological
impacts among those waterbodies and food webs that tend to experience higher bioaccumulation
of selenium.
The results of the ecological risk model indicate that sublethal effects from dietary
exposure to selenium (from discharges of the evaluated wastestreams) can lead to hidden
population-level effects among exposed fish and waterfowl by reducing reproductive success.
The results for mallards illustrate the broader effects throughout the food web that can result
from exposure to waterbodies contaminated with selenium. These results also indicate that
impacts to aquatic-dependent wildlife are not limited to piscivorous wildlife such as mink and
eagles.
The ecological risk model accounts only for those reproductive effects associated with
exposure to selenium. There might be more immediate receiving waters whose pollutant levels
result in elevated reproductive risk because they contain other pollutants at concentrations that
are harmful to wildlife.
For more information on the potential environmental impacts from selenium exposure,
refer to the selenium discussion in Section 3.1. For more detailed information on baseline
modeled fish tissue concentrations in the immediate receiving water for selenium and other
pollutants evaluated in the EA, see Figures H-ll to H-21 and Tables H-12 to H-22 in
Appendix H.
6.2.3 Impacts to Benthic Organisms
EPA also assessed the potential impact to wildlife exposed to sediments in surface waters
that receive discharges of the evaluated wastestreams by comparing estimated pollutant
concentrations in the sediment to chemical stressor concentration limit (CSCL) benchmarks for
sediment biota published by MacDonald, et. al. (2000) in Archives of Environmental
Contamination and Toxicology. Table 6-4 presents the number and percentage of immediate
receiving waters with sediment pollutant concentrations that exceed a CSCL. EPA calculated
that 22 percent of rivers and streams (40 out of 183) and 35 percent of lakes, ponds, and
reservoirs (9 out of 26) had estimated sediment pollutant concentrations that may be toxic to
wildlife.
Benthic organisms are at risk primarily due to exposure to mercury, nickel, and cadmium.
Estimated sediment pollutant concentrations in 13 to 23 percent of the immediate receiving
waters (27 to 49 out of 209) exceed the sediment biota CSCL benchmarks for exposure to
cadmium-contaminated, nickel-contaminated, and mercury-contaminated waters. Therefore,
benthic organisms inhabiting these waters are under a greater threat for sublethal effects such as
skeletal malformation and reduced growth and reproductive success. For more information on
these impacts, refer to Section 3.1.1.
6-6
-------
Section 6—Current Impacts from Steam Electric Power Generating Industry
As expected, based on documented environmental impacts, modeling results indicate that
pollutant concentrations in the benthic sediment in lakes, ponds and reservoirs are more likely to
exceed the sediment biota CSCL benchmarks than pollutant concentrations in the benthic
sediment of rivers and streams. Several publications in the literature confirm that sediment
impacts are more likely to occur in lakes where pollutants can accumulate in sediments over time
[Hopkins etal., 2000, 2003; Lemly, 1997a].
Table 6-4. Number and Percentage of Immediate Receiving Waters with Sediment
Pollutant Concentrations Exceeding CSCLs for Sediment Biota at Baseline
Pollutant
Arsenic
Cadmium
Chromium VI b
Copper
Lead
Mercury
Nickel
Selenium
Thallium
Zinc
Sediment
Benchmark
(mg/kg)
5.90
0.596
37.3
35.7
35
0.174
18.0
ID
ID
123
Total Number of Unique Immediate
Receiving Waters
Number of Immediate Receiving Waters Exceeding CSCLs for
Sediment Biota
Rivers and
Streams
7
22
0
6
5
40
29
NC
NC
14
40
Lakes, Ponds,
and Reservoirs
0
5
0
1
1
9
5
NC
NC
1
9
Total Immediate Receiving
Waters
Number a
7
27
0
7
6
49
34
NC
NC
15
49
Percent
3%
13%
0%
3%
3%
23%
16%
NC
NC
7%
23%
Sources: ERG, 2015d; ERG, 2015h; ERG, 20151.
Acronyms: CSCL (Chemical stressor concentration limit); ID (Insufficient data; no benchmarks were identified);
NC (Not calculated).a - The EA encompasses a total of 222 immediate receiving waters and loadings from 195
steam electric power plants (some of which discharge to multiple receiving waters). The IRW model, which
excludes the Great Lakes and estuaries, encompasses a total of 209 immediate receiving waters (183 rivers and
streams; 26 lakes, ponds, and reservoirs) and loadings from 188 steam electric power plants.
b - No benchmark for chromium VI. EPA used the total chromium benchmark, which may underestimate the impact
to wildlife.
6.3 HUMAN HEALTH IMPACTS
In addition to assessing water quality impacts on human health as discussed in Section
3.3.2, EPA expanded the analysis to evaluate human health impacts from consuming fish in
immediate receiving waters downstream from discharges of the evaluated wastestreams. The
purpose of this analysis was to evaluate the broader bioaccumulative effects of pollutants in
steam electric power plant discharges to see whether average daily doses of pollutants from fish
consumption could potentially exceed human health thresholds where water concentrations may
not indicate an issue. EPA evaluated multiple human cohorts (i.e., recreational and subsistence
fishers, children and adults) by calculating the average daily dose of pollutants from fish
consumption using the estimated fish tissue concentrations calculated in the model. EPA varied
the fish consumption rate of each cohort (based on age) to determine the average and long-term
daily doses for each pollutant. EPA calculated the lifetime excess cancer risk (LECR) based on
6-7
-------
Section 6—Current Impacts from Steam Electric Power Generating Industry
estimated fish tissue concentrations of inorganic arsenic and calculated non-cancer threats by
comparing the average daily doses to threshold values for all pollutants with published reference
doses. EPA first evaluated human health impacts based on type of fisher and age of cohort using
national-level consumption rates. For the environmental justice analysis, EPA determined fish
consumption rates using the race population in addition to the other two factors. For more
information on how EPA identified potential impacts to human receptors, see Section 5.1.3 and
Appendix E.
The human health module presents the risk results for each age group individually to
allow for further manipulation in the benefits analysis. The true cancer risk to a child would
depend on the amount of time the child consumed fish from locations downstream from steam
electric power plant discharges. For example, the cancer risk for a 6-year-old child who was born
and raised in the same place would be the sum of the LECRs from the 1 to <2 years, 2 to <3
years, and 3 to <6 years cohort groups.
A limitation of the national-scale IRW modeling that may underestimate the cancer risk
is the use of an average annual pollutant loading rate as the basis for the risk estimation; as
described earlier, the model does not consider the potential for pollutants to accumulate over
time in the environment. The model estimates a minimal cancer risk from consuming fish in
lakes, ponds, and reservoirs that receive discharges of the evaluated wastestreams. The cancer
risk is likely greater in a lake, where fish are limited in their food sources and can bioaccumulate
pollutants over a longer exposure period than is represented in the model.
6.3.1 National-Scale Cohort Analysis
Table 6-5 presents the number and percentage of immediate receiving waters where the
estimated LECR for the national-scale human receptor exceeds the selected threshold, 1-in-a-
million cancer risk for arsenic. Inorganic arsenic concentrations in fish result in an estimated
cancer risk greater than 1-in-a-million to adult subsistence fishers in approximately 12 percent of
the immediate receiving waters (25 out of 209) and to adult recreational fishers in approximately
6 percent of the immediate receiving waters (12 out of 209). Cancer risks for the child cohorts
are lower, with LECRs exceeding the cancer risk threshold in 2 to 4 percent of the immediate
receiving waters. Even given the limitations of the modeling framework discussed in Section 6.3,
the inorganic arsenic concentrations in fish can pose a cancer risk to adult subsistence fishers in
12 percent of the lakes and to adult recreational fishers in 8 percent of the lakes.
6-8
-------
Section 6—Current Impacts from Steam Electric Power Generating Industry
Table 6-5. Number and Percentage of Immediate Receiving Waters That Exceed Human
Health Evaluation Criteria (Lifetime Excess Cancer Risk) for Inorganic Arsenic at
Baseline
Receptor
Child
recreational
fisher
Cohort
1 to <2 years
2 to <3 years
3 to <6 years
6 to <11 years
11 to <16 years
16 to <21 years
Adult recreational fisher
Child
subsistence
fisher
1 to <2 years
2 to <3 years
3 to <6 years
6 to <11 years
11 to < 16 years
16 to <21 years
Adult subsistence fisher
Exposure
Duration
(Years)
1
1
3
5
5
5
49
1
1
3
5
5
5
49
Number of Immediate Receiving Waters Where
Lifetime Excess Cancer Risk Exceeds 1-in-a-Million a'b
Number of
Rivers and
Streams
4
4
6
6
6
6
10
6
6
7
8
6
6
22
Number of
Lakes, Ponds,
and Reservoirs
0
0
0
0
0
0
2
0
0
0
1
0
0
o
J
Total Receiving Waters c
Number
Exceeding
4
4
6
6
6
6
12
6
6
7
9
6
6
25
Percentage
Exceeding
2%
2%
3%
3%
3%
3%
6%
3%
3%
3%
4%
3%
3%
12%
Sources: ERG, 2015d; ERG, 2015h; ERG, 20151.
a - The EA encompasses a total of 222 immediate receiving waters and loadings from 195 steam electric power
plants (some of which discharge to multiple receiving waters). The IRW model, which excludes the Great Lakes and
estuaries, encompasses a total of 209 immediate receiving waters (183 rivers and streams; 26 lakes, ponds, and
reservoirs) and loadings from 188 steam electric power plants.
b - Inorganic arsenic cancer slope factor of 1.5 per milligrams per kilogram (mg/kg) per day.
c - These values are the sum and percentage of rivers, streams, lakes, ponds, and reservoirs impacted.
Based on the estimated fish tissue concentrations and average daily pollutant doses by
cohort, subsistence fishers (adults and children) have the greatest threat for non-cancer health
effects. This is because the average daily doses (for one or more pollutant) exceed the oral
reference dose values in 49 to 56 percent of the immediate receiving waters, depending on the
age group evaluated. Recreational fishers (adult or child) have less of a threat, with average daily
doses exceeding oral reference doses in 41 to 48 percent of the immediate receiving waters.
These results suggest that fish downstream from discharges of the evaluated wastestreams pose a
non-cancer health threat to surrounding fisher populations. Given the modeling limitations
described above, these results may underestimate these non-cancer health impacts.
Table 6-6 presents the number and percentage of immediate receiving waters where the
average daily dose of one or more pollutant exceeds an oral reference dose for non-carcinogens.
6-9
-------
Section 6—Current Impacts from Steam Electric Power Generating Industry
Table 6-6. Number and Percentage of Immediate Receiving Waters
That Exceed Non-Cancer Oral Reference Dose Values at Baseline
Receptor
Child
recreational
fisher
Cohort
1 to <2 years
2 to <3 years
3 to <6 years
6 to <11 years
11 to <16 years
16to<21years
Adult recreational fisher
Child
subsistence
fisher
1 to <2 years
2 to <3 years
3 to <6 years
6 to <11 years
11 to <16 years
16to<21years
Adult subsistence fisher
Exposure
Duration
(Years)
1
1
3
5
5
5
49
1
1
3
5
5
5
49
Number of Immediate Receiving Waters where Estimated
Exposure Doses Exceed Non-Cancer Reference Doses a
Number of
Rivers and
Streams
82
82
80
76
72
72
72
98
98
92
87
84
84
85
Number of
Lakes, Ponds,
and Reservoirs
18
18
18
16
14
14
14
20
20
19
19
18
18
18
Total Receiving Waters b
Number
Exceeding
100
100
98
92
86
86
86
118
118
111
106
102
102
103
Percentage
Exceeding
48%
48%
47%
44%
41%
41%
41%
56%
56%
53%
51%
49%
49%
49%
Sources: ERG, 2015d; ERG, 2015h; ERG, 20151.
a - The EA encompasses a total of 222 immediate receiving waters and loadings from 195 steam electric power
plants (some of which discharge to multiple receiving waters). The IRW model, which excludes the Great Lakes and
estuaries, encompasses a total of 209 immediate receiving waters (183 rivers and streams; 26 lakes, ponds, and
reservoirs) and loadings from 188 steam electric power plants.
b - These values are the sum and percentage of rivers, streams, lakes, ponds, and reservoirs impacted.
According to the exposure doses calculated from the estimated fish tissue concentrations,
methylmercury poses the greatest threat to cause non-cancer health effects in humans from fish
consumption. Mercury concentrations in fish pose a non-cancer threat to humans in
approximately 52 percent of the immediate receiving waters. Therefore, humans who consume
fish inhabiting these waters are at risk for developing mercury-related effects, which could
include neurological symptoms (e.g., affecting fine motor function, language skills, verbal
memory) and cardiovascular disease if exposed at high enough doses. In addition, thallium
concentrations in fish pose a non-cancer threat to humans in approximately 45 percent of
immediate receiving waters.45 Therefore, humans who consume thallium-contaminated fish
inhabiting these waters are more likely to develop neurological symptoms (e.g., weakness, sleep
disorders, muscular problems), alopecia (i.e., loss of hair from the head and body), and
gastrointestinal effects (e.g., diarrhea and vomiting).
Table 6-7 presents the number and percentage of immediate receiving waters where
average daily doses exceed an oral reference dose for non-carcinogens by pollutant.
' EPA used the chronic oral exposure value cited in U.S. EPA, 2010a for thallium chloride as the reference dose.
6-10
-------
Section 6—Current Impacts from Steam Electric Power Generating Industry
Table 6-7. Number and Percentage of Immediate Receiving Waters That Exceed Non-
Cancer Oral Reference Dose Values at Baseline by Pollutant
Pollutant
Inorganic arsenic
Cadmium
Chromium VI
Copper
Lead
Mercury (as methylmercury)
Nickel (soluble salts)
Selenium
Thallium (soluble salts)
Zinc
Oral
Reference Dose
(mg/kg/day)
0.0003 b
0.001 b
0.003 b
0.01 c
ID
0.0001 b
0.02 b
0.005 b
0.00001 d
0.3 b
Number of Immediate Receiving Waters where Estimated
Exposure Doses Exceed Non-Cancer Reference Doses a
Number Exceeding
o
J
32
0
6
NC
109
0
55
94
9
Percentage Exceeding
1%
15%
0%
3%
NC
52%
0%
26%
45%
4%
Sources: ERG, 2015d; ERG, 2015h; ERG, 20151.
Acronyms: NC (Not calculated); ID (Insufficient data; there is no current reference dose for lead).
a - The EA encompasses a total of 222 immediate receiving waters and loadings from 195 steam electric power
plants (some of which discharge to multiple receiving waters). The IRW model, which excludes the Great Lakes and
estuaries, encompasses a total of 209 immediate receiving waters (183 rivers and streams; 26 lakes, ponds, and
reservoirs) and loadings from 188 steam electric power plants.
b-U.S. EPA, 201 Ic.
c-ATSDR, 2010a.
d-U.S. EPA,2010a.
States, territories, and authorized tribes have the primary responsibility to protect
residents from the health risks of consuming contaminated noncommercially caught fish. They
inform the general population, including recreational and subsistence fishers, typically by issuing
advisories that notify the public that chemical contamination found in local fish may present a
public health hazard.
EPA modeled concentrations in T4 fish tissue and compared them to fish consumption
advisory screening values to assess the potential for discharges of the evaluated wastestreams to
cause or contribute to fish advisories and pose a human health hazard. Based on the modeling
results, up to 48 percent of the immediate receiving waters evaluated may contain fish with
contamination levels that could trigger advisories for recreational and subsistence fishers.
Mercury and selenium are the pollutants most likely to exceed screening values. This result
indicates that steam electric power plants are contributing to the already widespread
concentrations of mercury and selenium in fish throughout the country.
Table 6-8 presents the number and percentage of immediate receiving waters where the
modeled T4 fish tissue concentrations exceed screening values used for fish advisories.
6-11
-------
Section 6—Current Impacts from Steam Electric Power Generating Industry
Table 6-8. Comparison of T4 Fish Tissue Concentrations at Baseline to
Fish Advisory Screening Values
Pollutant
Inorganic arsenic
(noncarcinogen)
Inorganic arsenic
(carcinogen)
Cadmium
Mercury (as
methylmercury)
Selenium
Recreational Fishers
Screening
Value (ppm)a
1.2
0.026
4.0
0.4
20
Number
Exceeding b
0
4
8
76
22
Percentage
Exceeding
0%
2%
4%
36%
11%
Subsistence Fishers
Screening
Value (ppm) a
0.147
0.00327
0.491
0.049
2.457
Number
Exceeding b
o
J
9
22
101
46
Percentage
Exceeding
1%
4%
11%
48%
22%
Sources: ERG, 2015d; ERG, 2015h; ERG, 20151.
Acronyms: ppm (parts per million).
a - Screening values are defined as concentrations of target analytes in fish or shellfish tissue that are of potential
public health concern and that are used as threshold values against which levels of contamination in similar tissue
collected from the ambient environment can be compared. Exceedance of these screening values indicates that more
intensive site-specific monitoring and/or evaluation of human health risk should be conducted [U.S. EPA, 2000a,
Table 5-3].
b - The EA encompasses a total of 222 immediate receiving waters and loadings from 195 steam electric power
plants (some of which discharge to multiple receiving waters). The IRW model, which excludes the Great Lakes and
estuaries, encompasses a total of 209 immediate receiving waters (183 rivers and streams; 26 lakes, ponds, and
reservoirs) and loadings from 188 steam electric power plants.
6.3.2 Environmental Justice Analysis
As part of the EA, EPA evaluated whether the impacts from steam electric power plant
wastewater discharges disproportionately impact minority groups. This environmental justice
(EJ) analysis included looking at impacts based on race or Hispanic origin. Table 6-9 presents
the number and percentage of immediate receiving waters where the estimated LECR for the
human receptor exceeds the selected threshold, 1-in-a-million cancer risk for arsenic. Inorganic
arsenic concentrations in fish result in an estimated cancer risk greater than 1-in-a-million to
adult subsistence, minority fishers in approximately 12 to 15 percent of the immediate receiving
waters (26 to 32 out of 209) and to adult recreational fishers in approximately 7 to 9 percent of
the immediate receiving waters (14 to 19 out of 209). Cancer risks for the child cohorts are
lower. The estimated cancer risk among adult minority fishers is higher than the risk among adult
nonminority fishers (especially among the recreational fisher population).
6-12
-------
Section 6—Current Impacts from Steam Electric Power Generating Industry
Table 6-9. Number and Percentage of Immediate Receiving Waters That Exceed Human
Health Evaluation Criteria (Lifetime Excess Cancer Risk) for Inorganic Arsenic at
Baseline, by Race or Hispanic Origin
Receptor
Recreational
Subsistence
Race or Hispanic
Origin
Non-Hispanic White
Non-Hispanic Black
Mexican- American
Other Hispanic
Other, including
Multiple Races
Non-Hispanic White
Non-Hispanic Black
Mexican- American
Other Hispanic
Other, including
Multiple Races
Number of Immediate Receiving Waters Where
Lifetime Excess Cancer Risk Exceeds 1-in-a-Million a'b
lto<2
years
o
6
3
4
4
4
4
5
6
6
6
2to<3
years
3
3
4
4
4
4
5
6
6
6
3to<6
years
4
5
6
6
6
6
6
6
6
7
6 to <11
years
6
6
6
6
6
7
7
8
7
10
11 to <16
years
6
6
6
6
6
7
7
8
7
10
16 to <21
years
6
6
6
6
6
7
7
8
7
10
Adult
12
14
18
16
19
25
26
28
28
32
Sources: ERG, 2015d; ERG, 2015h; ERG, 20151.
a - The EA encompasses a total of 222 immediate receiving waters and loadings from 195 steam electric power
plants (some of which discharge to multiple receiving waters). The IRW model, which excludes the Great Lakes and
estuaries, encompasses a total of 209 immediate receiving waters (183 rivers and streams; 26 lakes, ponds, and
reservoirs) and loadings from 188 steam electric power plants.
b - Inorganic arsenic cancer slope factor of 1.5 per milligrams per kilogram (mg/kg) per day.
Based on the estimated fish tissue concentrations and average daily pollutant doses by
cohort, subsistence fishers (adults and children) have the greatest threat for non-cancer health
effects. This is because the average daily doses (for one or more pollutant) exceed the oral
reference dose values in 49 to 56 percent of the immediate receiving waters, depending on the
age group evaluated. Recreational fishers (adult or child) have less of a threat, with average daily
doses exceeding oral reference doses in 41 to 48 percent of the immediate receiving waters.
These results suggest that fish downstream from discharges of the evaluated wastestreams pose a
non-cancer health threat to surrounding fisher populations. Given the modeling limitations
described above, these results may underestimate these non-cancer health impacts.
Table 6-10 presents the number and percentage of immediate receiving waters where the
average daily dose of one or more pollutant exceeds an oral reference dose for non-carcinogens.
6-13
-------
Section 6—Current Impacts from Steam Electric Power Generating Industry
Table 6-10. Number and Percentage of Immediate Receiving Waters That Exceed Non-Cancer Oral Reference Dose Values at
Baseline, by Race or Hispanic Origin
Receptor
Recreational,
Child Fisher
Subsistence,
Child Fisher
Recreational,
Adult Fisher
Subsistence,
Adult Fisher
Race or Hispanic Origin
Non-Hispanic White
Non-Hispanic Black
Mexican-American
Other Hispanic
Other, including Multiple Races
Non-Hispanic White
Non-Hispanic Black
Mexican- American
Other Hispanic
Other, including Multiple Races
Non-Hispanic White
Non-Hispanic Black
Mexican- American
Other Hispanic
Other, including Multiple Races
Non-Hispanic White
Non-Hispanic Black
Mexican-American
Other Hispanic
Other, including Multiple Races
Number of Immediate Receiving Waters Where Pollutant Exceeds a Non-Cancer Reference Dose a
Inorganic
Arsenic
0 (0%)
0 (0%)
0 (0%)
0 (0%)
0 (0%)
3 (1%)
3 (1%)
3 (1%)
3 (1%)
3 (1%)
0 (0%)
0 (0%)
0 (0%)
0 (0%)
0 (0%)
3 (1%)
3 (1%)
3 (1%)
3 (1%)
3 (1%)
Cadmium
10 (5%)
12 (6%)
14 (7%)
13 (6%)
14 (7%)
21 (10%)
22(11%)
25 (12%)
25 (12%)
29 (14%)
10 (5%)
12 (6%)
14 (7%)
13 (6%)
14 (7%)
21 (10%)
22(11%)
25 (12%)
25 (12%)
29 (14%)
Copper
3 (1%)
4 (2%)
4 (2%)
4 (2%)
4 (2%)
5 (2%)
5 (2%)
6 (3%)
5 (2%)
6 (3%)
3 (1%)
4 (2%)
4 (2%)
4 (2%)
4 (2%)
5 (2%)
5 (2%)
6 (3%)
5 (2%)
6 (3%)
Mercury b
81 (39%)
84 (40%)
86 (41%)
84 (40%)
88 (42%)
98 (47%)
98 (47%)
100 (48%)
100 (48%)
104 (50%)
81 (39%)
84 (40%)
86 (41%)
84 (40%)
88 (42%)
98 (47%)
98 (47%)
100 (48%)
100 (48%)
104 (50%)
Selenium
32 (15%)
33 (16%)
33 (16%)
33 (16%)
34 (16%)
42 (20%)
43 (21%)
46 (22%)
46 (22%)
48 (23%)
32 (15%)
33 (16%)
33 (16%)
33 (16%)
34 (16%)
42 (20%)
43 (21%)
46 (22%)
46 (22%)
48 (23%)
Thallium c
55 (26%)
58 (28%)
63 (30%)
60 (29%)
63 (30%)
76 (36%)
78 (37%)
79 (38%)
79 (38%)
89 (43%)
55 (26%)
58 (28%)
63 (30%)
60 (29%)
63 (30%)
76 (36%)
78 (37%)
79 (38%)
79 (38%)
89 (43%)
Zinc
4 (2%)
4 (2%)
4 (2%)
4 (2%)
4 (2%)
5 (2%)
5 (2%)
6 (3%)
6 (3%)
6 (3%)
4 (2%)
4 (2%)
4 (2%)
4 (2%)
4 (2%)
5 (2%)
5 (2%)
6 (3%)
6 (3%)
6 (3%)
Sources: ERG, 2015d; ERG, 2015h; ERG, 20151.
a - The EA encompasses a total of 222 immediate receiving waters and loadings from 195 steam electric power plants (some of which discharge to multiple
receiving waters). The IRW model, which excludes the Great Lakes and estuaries, encompasses a total of 209 immediate receiving waters (183 rivers and
streams; 26 lakes, ponds, and reservoirs) and loadings from 188 steam electric power plants.
b - Mercury, as methylmercury.
c - Reference dose based on thallium (soluble salts).
6-14
-------
Section 7—Environmental Improvements Under the Final Rule
SECTION 7
ENVIRONMENTAL IMPROVEMENTS UNDER
THE FINAL RULE
In Section 6, EPA presented the environmental impacts to surface water quality, wildlife,
and human health estimated with EPA's immediate receiving water (IRW) model and ecological
risk model resulting from baseline discharges of the evaluated wastestreams. Under the final
steam electric effluent limitations guidelines and standards (ELGs), EPA evaluated six regulatory
options (Options A, B, C, D, E, and F). As part of this quantitative environmental assessment
(EA), EPA evaluated the environmental improvements associated with the reduction in pollutant
loadings from the evaluated wastestreams (i.e., flue gas desulfurization (FGD) wastewater, fly
ash transport water, bottom ash transport water, and combustion residual leachate) under Options
A, B, C, D, and E, described in Table 7-1.46
In the remainder of this document, EPA presents the results only for Options A through E
for existing sources. During development of the final rule, EPA decided not to base the final rule
on Option F for existing sources due primarily to the high cost of that Option, particularly in
light of the costs associated with other rulemakings expected to impact the steam electric
industry (see Section VIII.C.I of the preamble). As a result, EPA chose not to conduct particular
analyses for Option F to the same extent that it did for some of the other options considered.
Section 8 of the Technical Development Document (TDD) (EPA-821-R-15-007) details the
technology options for all wastestreams evaluated under each regulatory option for the final rule.
As described in Section 8 of the TDD, EPA selected Option D as the technology basis for the
best available technology economically achievable (BAT) and for pretreatment standards for
existing sources (PSES). See Section 12 of the TDD for further information on the limitations
and standards of the final rule. This section presents the improvements to surface water quality,
wildlife, and human health under the final rule as quantified by EPA's IRW model and
ecological risk model.
Based on the quantitative and qualitative analyses performed for the EA, EPA estimated
that a variety of environmental improvements would result from the pollutant loading removals
associated with the regulatory options. In particular, the EA evaluated the following: 1)
improvements in water quality, 2) reduction in threats to wildlife, 3) reduction in human health
cancer risks, 4) reduction in threats for non-cancer human health effects, and 5) other
unquantified environmental improvements. Table 7-2 lists the quantified and unqualified
environmental improvements estimated to result from the final rule's regulatory options and
designates which quantified improvements were monetized in the benefits analysis described in
the Benefits and Cost Analysis (EPA-821-R-15-005).
46 In addition to the wastestreams listed in Table 7-1, EPA evaluated technology options associated with flue gas
mercury control (FGMC) wastewater, gasification wastewater, and nonchemical metal cleaning wastes as part of
the regulatory options. However, no plants currently discharge FGMC wastewater, all existing gasification plants are
operating the technology used as the basis for the regulatory option, and EPA will continue to reserve
BAT/NSPS/PSES/PSNS for nonchemical metal cleaning wastes, as previously established regulations do.
Therefore, EPA estimated zero compliance costs and zero pollutant reductions associated with these wastestreams
and did not include these three wastestreams in the EA.
7-1
-------
Section 7—Environmental Improvements Under the Final Rule
Table 7-1. Regulatory Options for the Wastestreams Evaluated in the EA
Evaluated
Wastestream a
FGD wastewater
Fly ash transport
water
Bottom ash
transport water
Combustion
residual leachate
Option
A
Chemical
precipitation
Dry handling
Impoundment
(equal to BPT)
Impoundment
(equal to BPT)
Option
B
Chemical
precipitation +
biological
treatment
Dry handling
Impoundment
(equal to BPT)
Impoundment
(equal to BPT)
Option
C
Chemical
precipitation +
biological
treatment
Dry handling
Dry handling/
closed loop
(for units >400
MW);
impoundment
(equal to BPT) for
units <400 MW
Impoundment
(equal to BPT)
Option
D
Chemical
precipitation +
biological
treatment
Dry handling
Dry handling/
closed loop
Impoundment
(equal to BPT)
Option
E
Chemical
precipitation +
biological
treatment
Dry handling
Dry handling/
closed loop
Chemical
precipitation
Acronyms: BPT (Best practicable control technology currently available); MW (Megawatt).
a - The evaluated wastestreams and regulatory options listed in the table are a subset of regulatory options for the
steam electric ELGs. See Section 8 of the TDD for the full list of regulatory options.
7-2
-------
Section 7—Environmental Improvements Under the Final Rule
Table 7-2. Description of Environmental Improvements
Associated with the Final Rule
Assessment
Category
Water
Quality
Description of Environmental
Improvement
Reduced number of immediate
receiving waters exceeding an acute
or chronic aquatic life NRWQC
Reduced number of immediate
receiving waters exceeding a human
health NRWQC
Reduced number of immediate
receiving waters exceeding MCLs
Increased aesthetic benefits, such as
enhancement of adjoining site
amenities (e.g., residing, working,
traveling, and owning property near
water)
Improved water-based recreation,
including swimming, fishing,
boating, and near-water activities
from improved water quality
Improved quality of source water
used for drinking, irrigation, and
industrial use
Increased property values from water
quality improvements
Increased tourism and participation in
water-based recreation
Pollutant removals to impaired
waters
Pollutant removals to the Great Lakes
and Chesapeake Bay
Pollutant removals of toxic
contaminants, chlorides, and TDS to
receiving waters
Nutrient removals to receiving waters
Reduced risk of surface
impoundment failures
Reduced sediment contamination
Increased availability of ground
water resources
Improvement
Quantified
-------
Section 7—Environmental Improvements Under the Final Rule
Table 7-2. Description of Environmental Improvements
Associated with the Final Rule
Assessment
Category
Wildlife
Human
Health
Description of Environmental
Improvement
Reduced exposure among minks to
pollutants that bioaccumulate in fish
Reduced exposure among eagles to
pollutants that bioaccumulate in fish
Reduced selenium concentrations in
fish and waterfowl and associated
reduced reproductive risk
Improved aquatic and wildlife habitat
and improved protection of
threatened and endangered species
Improved commercial fisheries yield
due to aquatic habitat improvement
Enhanced existence, option, and
bequest values from improved
ecosystem health
Reduced risks to aquatic life from
exposure to steam electric pollutants
Reduced exposure to pollutants
associated with the wastestreams of
concern in surface impoundments
that serve as attractive nuisances
Reduced exposure to non-cancer
pollutants for recreational and
subsistence fishers
Reduced cancer risk in recreational
and subsistence fishers
Reduced incidences of cardiovascular
disease from reduced arsenic and lead
exposure
Reduced adverse health effects from
reduced in-utero mercury exposure
from maternal fish consumption
Reduced IQ loss and specialized
education from reduced childhood
exposure to lead from fish
consumption
Reduced adult mortality from air
pollutant emissions
Avoided climate change impacts
from carbon dioxide emissions
Reduced exposure to pollutants from
recreational water uses
Improvement
Quantified
-------
Section 7—Environmental Improvements Under the Final Rule
Table 7-2. Description of Environmental Improvements
Associated with the Final Rule
Assessment
Category
Description of Environmental
Improvement
Reduced injury associated with
impoundment failures
Reduced number of immediate
receiving waters exceeding fish
consumption advisory screening
values
Improvement
Quantified
-------
Section 7—Environmental Improvements Under the Final Rule
Excess chlorides levels in wastewater discharges can be harmful to animals and plants in
nonmarine surface waters and can disrupt ecosystem structure. Under the final rule, annual
chlorides loadings to surface waters will decrease by 21.8 million pounds (two percent).
The pollutant parameter, IDS, comprises dissolved solids such as chloride and metals.
Under the final rule, EPA calculated that annual IDS loadings to surface waters will decrease by
more than 1.32 billion pounds (31 percent). This decrease is at least partially due to the reduction
in total and dissolved metals discharged to receiving waters.47
47 EPA's estimated TDS removals do not account for additional removals that may be achieved as a result of steam
electric power plants opting to participate in the voluntary incentives program, in which they would be subject to
effluent limitations based on evaporation technology, including for TDS.
-------
Section 7—Environmental Improvements Under the Final Rule
Table 7-3. Steam Electric Power Generating Industry Pollutant Removals for Metals,
Bioaccumulative Pollutants, Nutrients, Chlorides, and TDS Under Regulatory Options
Pollutant Removals, Ibs/yr (Percent Reduction)'
Source: ERG, 2015a.
Acronyms: TDS (Total Dissolved Solids); Ibs/yr (pounds per year).
Note: Pollutant removals are rounded to three significant figures.
a - .>0 to 15 percent reduction; 16 to 30 percent reduction; 31 to 45 percent reduction;
>60 percent reduction
b - Total nitrogen loadings are the sum of total Kjeldahl nitrogen and nitrate/nitrite as N loadings.
46 to 60 percent reduction
7-7
-------
Section 7—Environmental Improvements Under the Final Rule
Table 7-4. Steam Electric Power Generating Industry TWPE Removals for Metals,
Bioaccumulative Pollutants, Nutrients, Chlorides, and TDS Under Regulatory Options
Pollutant Removals, TWPE/year (Percent Reduction)a
Source: ERG, 2015a.
Acronyms: TDS (Total Dissolved Solids); TWPE (Toxic Weighted Pound Equivalents).
Note: Pollutant removals are rounded to three significant figures.
N/A - The TWPE/year is not provided for total nitrogen, total phosphorus, and TDS because EPA has not
established a toxic weighting factor (TWF) for these pollutants.
a - >0 to 15 percent reduction; 16 to 30 percent reduction; 31 to 45 percent reduction; |
46 to 60 percent reduction
>60 percent reduction
7-8
-------
Section 7—Environmental Improvements Under the Final Rule
7.2 KEY ENVIRONMENTAL IMPROVEMENTS
As part of this EA, EPA conducted modeling of the expected environmental
improvements under Options A through E. EPA estimates the environmental improvements
under Option F, which were not modeled, to be incrementally greater than those under Option E
based on the pollutant reductions calculated.
Table 7-5 summarizes the key environmental improvements within the immediate
receiving waters due to the pollutant removals under the final rule (Option D) and other
evaluated regulatory options. The numbers of immediate receiving waters with water quality,
wildlife, and human health exceedances would:
• Decrease under Options A and B by no more than 33 percent, with most exceedances
being reduced by less than 15 percent.
• Decrease under Option C by 17 to 56 percent, with most exceedances being reduced
by less than 40 percent.
• Decrease under Option D by 45 to 83 percent, with most exceedances being reduced
by at least 56 percent.
• Decrease under Option E by 51 to 84 percent, with most exceedances being reduced
by at least 61 percent.
The final rule (Option D) will substantially improve water quality, wildlife, and human
health. Under the final rule, EPA estimates that:
• Receiving water exceedances of the NRWQC will decrease by 45 to 67 percent.
• Receiving water exceedances of the MCL benchmarks will decrease by 83 percent.
• The number of receiving waters with fish tissue concentrations exceeding the no
effect hazard concentration (NEHC) for selenium for eagles and minks will decrease
by 63 and 62 percent, respectively.
• Human exposures via fish consumption to pollutants with the potential to cause non-
cancer health effects will decrease by up to 56 percent.
• Human exposures to pollutants that present a cancer risk will decrease by up to 75
percent.
Results for the final rule are discussed in further detail in the sections following Table
7-5.
7.2.1 Improvements in Water Quality Under the Final Rule
EPA estimates that pollutant removals to surface waters associated with the final rule will
significantly improve water quality by reducing exceedances of the NRWQC and MCLs by up to
83 percent. The largest reductions in NRWQC exceedances are attributed to reduced loadings of
cadmium, selenium, arsenic, and thallium. Due to the substantial pollutant removals, EPA
projects that aquatic organisms will be less susceptible to chronic impacts such as:
7-9
-------
Section 7—Environmental Improvements Under the Final Rule
• Skeletal malformations;
• Organ damage;
• Developmental abnormalities;
• Behavioral impairments;
• Reproductive failure;
• Metabolic failure;
• Neurological effects;
• Gastrointestinal effects; and
• Fish kills.48
EPA estimates that up to 45 percent of the 209 evaluated immediate receiving waters
currently exceed NRWQC for the protection of human health, primarily due to arsenic and
thallium. EPA estimates that these arsenic and thallium removals will lower the number of
immediate receiving waters that exceed NRWQC designed to protect public health by 45 to 50
percent. By reducing MCL exceedances by 83 percent, the final rule will improve the quality of
source water available to drinking water treatment plants downstream from steam electric power
plants.
In addition to reducing NRWQC and MCL exceedances, the final rule will quantifiably
improve overall water quality - in the immediate receiving waters and downstream from steam
electric power plants. EPA calculates that, on average, receiving water concentrations of the 10
toxic, bioaccumulative pollutants evaluated in the EA will decrease by 57 percent.
48 Impacts documented in ATSDR, 2008a; Coughlan and Velte, 1989; Lemly, 1985b; Nagle et al, 2001; NRC,
2006; Rowe et al., 2002; U.S. EPA, 2009a; and U.S. EPA, 201 If.
-------
Section 7—Environmental Improvements Under the Final Rule
Table 7-5. Key Environmental Improvements Under the Regulatory Options
Evaluation Benchmark
Modeled Immediate Receiving
Waters Exceeding Benchmark
Under Baseline Conditionsa
Number
Percentage
Number of Immediate Receiving Waters Exceeding Benchmark
(Percent Reduction from Baseline Conditions) Under the Regulatory
Options b
Option A | Option B | Option C | Option D | Option E
Water Quality Results
Freshwater Acute NRWQC
Freshwater Chronic NRWQC
35
4%
17%
6
(33%)
34
(3%)
Human Health Water and Organism NRWQC
94
45%
90
(4%)
Human Health Organism Only NRWQC
Drinking Water MCL
Wildlife Results
Fish Ingestion NEHC for Minks
Fish Ingestion NEHC for Eagles
Human Health Results—Non-Cancer
Non-Cancer Reference Dose for Child
(recreational)
66
32%
62
(6%)
36
17%
34
(6%)
58
71
28%
34%
57
(2%)
65
61
(14%)
100
48
92
(8%)
90
(10%)
Non-Cancer Reference Dose for Adult
(recreational)
86
41%
77
(10%)
74
(14%)
Non-Cancer Reference Dose for Child
(subsistence)
118
56%
107
(9%)
104
(12%)
Non-Cancer Reference Dose for Adult
(subsistence)
103
49%
94
93
(10%)
7-11
-------
Section 7—Environmental Improvements Under the Final Rule
Table 7-5. Key Environmental Improvements Under the Regulatory Options
Evaluation Benchmark
Modeled Immediate Receiving
Waters Exceeding Benchmark
Under Baseline Conditionsa
Number
Percentage
Number of Immediate Receiving Waters Exceeding Benchmark
(Percent Reduction from Baseline Conditions) Under the Regulatory
Options b
Option A | Option B | Option C | Option D | Option E
Human Health Results—Cancer
Arsenic Cancer Risk for Child (recreational)
Arsenic Cancer Risk for Adult (recreational)
Arsenic Cancer Risk for Child (subsistence)
Arsenic Cancer Risk for Adult (subsistence)
12
25
3%
6%
4%
12%
5
(17%)
9
(25%)
7
(13%)
23
(8%)
23
(8%)
Source: ERG, 2015d; ERG, 2015h; ERG, 20151.
Acronyms: MCL (maximum contaminant level); NEHC (No Effect Hazard Concentration); NRWQC (National Recommended Water Quality Criteria).
a - The EA encompasses a total of 222 immediate receiving waters and loadings from 195 steam electric power plants (some of which discharge to multiple
receiving waters). The IRW model, which excludes the Great Lakes and estuaries, encompasses a total of 209 immediate receiving waters (183 rivers and
streams; 26 lakes, ponds, and reservoirs) and loadings from 188 steam electric power plants.
b - >0 to 15 percent reduction; 16 to 30 percent reduction; 31 to 45 percent reduction;
46 to 60 percent reduction|>60 percent reduction
7-12
-------
Section 7—Environmental Improvements Under the Final Rule
7.2.2 Reduced Threat to Wildlife Under the Final Rule
In the EA, EPA evaluated multiple threats to wildlife, including impacts to wildlife
indicator species by consuming contaminated fish; impacts to fish and waterfowl due to dietary
exposure to selenium; and exposure of benthic aquatic organisms to contaminated sediments.
The combination of lethal and sublethal effects (e.g., changes to morphology, behavior, and
metabolism) of exposure to steam electric power plant wastewater can cause cascading effects
through the food web.
As discussed in Section 7.2.1, the number of immediate receiving waters that can
potentially pose an acute or chronic threat to wildlife will decrease under the final rule,
improving wildlife populations and communities surrounding steam electric power plants (e.g.,
reduced impacts to population density and species diversity as discussed in Section 3). EPA
estimates that average fish tissue concentrations of the pollutants evaluated in the EA will
decrease by an average of 57 percent. EPA projects that these lower pollutant concentrations will
significantly improve the health of fish populations and the quality of fish available for
consumption by both humans and wildlife near steam electric power plants.
Based on the threats to minks and eagles from consuming fish contaminated by steam
electric power plant wastewater, pollutants can bioaccumulate and impact higher order species in
the food chain. Under the final rule, EPA estimates that exceedances of the NEHC for eagles and
minks will decrease by approximately 70 percent. See Section 7.3.3 for discussion of the reduced
risk of adverse reproductive effects among aquatic wildlife (fish and mallards) resulting from
dietary exposure to selenium.
EPA estimates that pollutant removals to surface waters associated with the final rule will
decrease the exposure of aquatic organisms to pollutants in the sediment, as shown in Table 7-6.
As discussed in Section 6.2.3, benthic organisms are at risk primarily due to exposure to
mercury, nickel, and cadmium. Under the final rule, the number of immediate receiving waters
with pollutant concentration in the sediment above chemical stressor concentration limits
(CSCL) will decrease by over 60 percent.
7-13
-------
Section 7—Environmental Improvements Under the Final Rule
Table 7-6. Number of Immediate Receiving Waters with Sediment Pollutant Concentrations Exceeding CSCLs for
Sediment Biota Under the Regulatory Options
Pollutant
Modeled Immediate
Receiving Waters
Exceeding CSCLs Under
Baseline Conditionsa
Number of Immediate Receiving Waters Exceeding Benchmark
(Percent Reduction from Baseline Conditions) Under the Regulatory Options b
Option A
Option B
Option C
Option E
Arsenic
Cadmium
Chromium VIc
7
(3%)
6
(14%)
6
(14%)
27
(13%)
21
(22%)
21
(22%)
0
(0%)
0
(N/A)
0
(N/A)
Copper
7
(3%)
5
(29%)
5
(29%)
Lead
6
(3%)
4
(33%)
4
(33%)
Mercury
49
(23%)
45
(8%)
44
(10%)
Nickel
Selenium
34
(16%)
28
(18%)
28
(18%)
NC
NC
NC
NC
Thallium
Zinc
NC
NC
NC
NC
15
(7%)
9
(40%)
9
(40%)
9
(40%)
Total
49
(23%)
45
(8%)
44
(10%)
27
(45%)
Source: ERG, 2015d;ERG, 2015h; ERG, 20151.
Acronyms: CSCL (Chemical stressor concentration limit); N/A (Not Applicable, no exceedances at baseline conditions to compare option results); NC (Not
calculated; no benchmark for comparison).
a - The EA encompasses a total of 222 immediate receiving waters and loadings from 195 steam electric power plants (some of which discharge to multiple
receiving waters). The IRW model, which excludes the Great Lakes and estuaries, encompasses a total of 209 immediate receiving waters (183 rivers and
streams; 26 lakes, ponds, and reservoirs) and loadings from 188 steam electric power plants.
b - >0 to 15 percent reduction; 16 to 30 percent reduction; 31 to 45 percent reduction;
c - EPA used the total chromium benchmark for this analysis.
46 to 60 percent reductionl>60 percent reduction
7-14
-------
Section 7—Environmental Improvements Under the Final Rule
7.2.3 Reduced Human Health Cancer Risk Under the Final Rule
Under baseline conditions, EPA estimates that 25 immediate receiving waters (12
percent) could contain fish contaminated with inorganic arsenic that present cancer risks above
the 1-in-a-million threshold for the most sensitive, national-scale cohort. EPA calculates that the
number of immediate receiving waters whose fish exceed this cancer risk threshold will decrease
by at least 56 percent for all national-scale cohorts under the final rule.
7.2.4 Reduced Threat of Non-Cancer Human Health Effects Under the Final Rule
Chronic exposure to toxic, bioaccumulative pollutants in steam electric power plant
wastewater can potentially compromise neurological and developmental functions and affect the
circulatory, respiratory, and digestive systems of exposed populations. EPA estimates that the
number of immediate receiving waters whose fish pose non-cancer health risks will decrease by
at least 52 percent for all national-scale cohorts under the final rule. As discussed in Section
7.2.2, EPA found that the pollutant concentrations in fish tissue will decrease, improving the
quality of fish available to recreational and subsistence fishers and subsequently lowering
exposures to toxic, bioaccumulative pollutants and the potential for humans to develop non-
cancer health effects (e.g., nausea, abdominal pain, sleep disorders, muscular problems, and
cardiovascular disease).
The pollutants that cause the potential for non-cancer health effects are selenium,
cadmium, mercury (as methylmercury), and, to a lesser degree, thallium. EPA calculates that the
final rule will decrease the number of immediate receiving waters with fish that, if consumed,
would exceed the reference doses for these pollutants, by the following amounts:
• Selenium: decrease by at least 51 percent for all national-scale cohorts.
• Cadmium: decrease by at least 53 percent for all national-scale cohorts.
• Methylmercury: decrease by at least 52 percent for all national-scale cohorts.
• Thallium: decrease by at least 62 percent for all national-scale cohorts.
Although the EA did not directly assess the potential non-cancer health effects posed by
lead,49 the final rule will lower the total annual loadings of lead to the environment by 19,000
pounds (98 percent), thus reducing the potential threat of hypertension, coronary heart disease,
and impaired cognitive function in exposed populations. For children in particular, lead exposure
can cause additional negative impacts, such as hyperactivity, behavioral and attention
difficulties, delayed mental development, and motor and perceptual skill deficits. The benefits to
adults and children from the reduced lead discharges are discussed in the Benefits and Cost
Analysis.
7.2.5 Reduced Human Health Risk for Environmental Justice Analysis
As discussed in Section 6.3.2, EPA evaluated the impacts that steam electric power plant
discharges have on environmental justice (EJ) cohorts in addition to the national-scale cohorts.
Under baseline conditions, EPA estimates that 32 immediate receiving waters (15 percent) could
49 Currently, there is no reference dose for lead—there is no safe level for ingestion of lead (see EPA's Integrated
Risk Information System (IRIS) website: http://www.epa.gov/IRIS/).
7-15
-------
Section 7—Environmental Improvements Under the Final Rule
contain fish contaminated with inorganic arsenic that present cancer risks above the 1-in-a-
million threshold for the most sensitive minority cohort. EPA estimates that the number of
immediate receiving waters whose fish exceed this cancer risk threshold will decrease by at least
46 percent for the average recreational fisher minority cohort and at least 51 percent for the
average subsistence fisher minority cohort under the final rule.50 These improvements are similar
to those for non-minority recreational and subsistence fisher cohorts (at least 33 and 50 percent,
respectively) under the final rule.
EPA estimates that the number of immediate receiving waters whose fish pose non-
cancer health risks will decrease by 56 percent for all recreational fisher minority cohorts and 53
percent for all subsistence fisher minority cohorts under the final rule. These improvements are
similar to those for non-minority recreational and subsistence fisher cohorts (56 and 52 percent,
respectively) under the final rule. The pollutants that cause the potential for non-cancer health
effects are selenium, cadmium, mercury (as methylmercury), and, to a lesser degree, thallium.
7.3 POLLUTANT-SPECIFIC IMPROVEMENTS
EPA identified several key pollutants (i.e., arsenic, mercury, selenium, cadmium, and
thallium) whose pollutant removals would primarily be responsible for the improvements in
water quality, wildlife, and human health attributed to the final rule. This section highlights the
environmental improvements associated with these five pollutants.
7.3.1 Arsenic
Under the final rule, EPA estimates 27,900 pounds per year of arsenic removals from
steam electric power plant discharges - a 94 percent reduction in annual loadings. The final rule
will decrease the number of immediate receiving waters exceeding human health NRWQC for
arsenic by up to 49 percent. The arsenic removals will reduce negative effects on aquatic
organisms, such as liver tissue death, developmental abnormalities, behavioral impairments,
metabolic failure, growth reduction, and appetite loss [NRC, 2006; Rowe etal., 2002; U.S. EPA,
201 If]. As a result, the final rule will decrease human exposure to arsenic through fish
consumption and thus lower the potential for exposed populations to develop arsenic-related
cancer and non-cancer health effects such as dermal, cardiovascular, and respiratory effects. The
final rule will decrease the number of immediate receiving waters exceeding the human health
cancer risk threshold for arsenic by up to 75 percent, depending on the evaluated cohort. Table
7-7 presents the key environmental improvements resulting from arsenic removals under the
regulatory options evaluated in the EA.
EPA did not see a reduction in the number of immediate receiving waters exceeding the
arsenic NEHCs for minks or eagles because there are no exceedances modeled at baseline. The
final rule, however, will still reduce the bioaccumulation of arsenic in the food web.
50 These values represent the average percentage improvements across the four race populations that comprise the
minority cohorts.
-------
Section 7—Environmental Improvements Under the Final Rule
Table 7-7. Key Environmental Improvements for Arsenic Under the Regulatory Options
Evaluation Benchmark
Modeled Immediate
Receiving Waters Exceeding
Benchmark Under Baseline
Conditionsa
Number
Percentage
Number of Immediate Receiving Waters Exceeding Benchmark
(Percent Reduction from Baseline Conditions) Under the Regulatory Options
Option A
Option B
Option C
Option D
Option E
Water Quality Results
Freshwater Acute NRWQC
1%
Freshwater Chronic NRWQC
2%
Human Health Water and Organism
NRWQC
94
45%
Human Health Organism Only NRWQC
65
31%
Drinking Water MCL
Wildlife Results
12
6%
Fish Ingestion NEHC for Minks
0%
Fish Ingestion NEHC for Eagles
0%
Human Health Results—Non-Cancer
Non-Cancer Reference Dose for Child
(recreational)
Non-Cancer Reference Dose for Adult
(recreational)
Non-Cancer Reference Dose for Child
(subsistence)
Non-Cancer Reference Dose for Adult
(subsistence)
1%
0%
1%
1%
7-17
-------
Section 7—Environmental Improvements Under the Final Rule
Table 7-7. Key Environmental Improvements for Arsenic Under the Regulatory Options
Evaluation Benchmark
Modeled Immediate
Receiving Waters Exceeding
Benchmark Under Baseline
Conditionsa
Number
Percentage
Number of Immediate Receiving Waters Exceeding Benchmark
(Percent Reduction from Baseline Conditions) Under the Regulatory Options
b
Option A
Option B
Option C
Option D
Option E
Human Health Results—Cancer
Arsenic Cancer Risk for Child
(recreational)
Arsenic Cancer Risk for Adult
(recreational)
Arsenic Cancer Risk for Child
(subsistence)
Arsenic Cancer Risk for Adult
(subsistence)
12
25
6%
4%
12%
5
(17%)
9
(25%)
7
(13%)
23
(8%)
23
(8%)
Source: ERG, 2015d; ERG, 2015h; ERG, 20151.
Acronyms: MCL (Maximum contaminant level); N/A (Not Applicable, no exceedances at baseline conditions to compare option results); NEHC (No Effect
Hazard Concentration); NRWQC (National Recommended Water Quality Criteria).
a - The EA encompasses a total of 222 immediate receiving waters and loadings from 195 steam electric power plants (some of which discharge to multiple
receiving waters). The IRW model, which excludes the Great Lakes and estuaries, encompasses a total of 209 immediate receiving waters (183 rivers and
streams; 26 lakes, ponds, and reservoirs) and loadings from 188 steam electric power plants.
b - >0 to 15 percent reduction; 16 to 30 percent reduction; 31 to 45 percent reduction;
46 to 60 percent reduction|>60 percent reduction
7-18
-------
Section 7—Environmental Improvements Under the Final Rule
7.3.2 Mercury
Under the final rule, EPA estimates 1,450 pounds per year of mercury removals from
steam electric power plant discharges - a 97 percent reduction in annual loadings. As discussed
in Section 6.2, estimated fish tissue concentrations for mercury (and selenium) exceed levels that
can affect reproduction in exposed mink and eagle populations. EPA estimates that the final rule
will decrease the number of immediate receiving waters with fish tissue concentrations that
exceed the mercury NEHC for eagles and minks by 62 and 64 percent, respectively. These
reductions also represent the potential improvement in exposure to mercury above effects
thresholds in other wildlife that consume fish from these receiving waters.
Under baseline pollutant loadings, EPA estimates that fish methylmercury concentrations
pose a non-cancer threat to subsistence fishers and recreational fishers in up to 52 and 46
percent, respectively, of immediate receiving waters. EPA calculates that fish tissue
concentrations of methylmercury will decrease under the final rule and, as a result, the number of
immediate receiving waters with exposure doses from fish consumption that exceed the
methylmercury reference dose will decrease by up to 57 percent. Because there are over 80
addressed by this final rule discharge to receiving waters that are under a fish advisory for
mercury (see Section 3.4.4), the final rule will reduce mercury loadings to those receiving waters
(see Section 7.4). Table 7-8 presents the key environmental improvements resulting from
mercury removals under the regulatory options.
7-19
-------
Section 7—Environmental Improvements Under the Final Rule
Table 7-8. Key Environmental Improvements for Mercury Under the Regulatory Options
Evaluation Benchmark
Modeled Immediate
Receiving Waters Exceeding
Benchmark Under Baseline
Conditionsa
Number
Percentage
Number of Immediate Receiving Waters Exceeding Benchmark
(Percent Reduction from Baseline Conditions) Under the Regulatory Options
b
Option A
Option B
Option C
Option D
Option E
Water Quality Results
Freshwater Acute NRWQC
Freshwater Chronic NRWQC
Human Health Water and Organism
NRWQC
0%
0%
No benchmark for comparison
Human Health Organism Only NRWQC
No benchmark for comparison
Drinking Water MCL
2%
Wildlife Results
Fish Ingestion NEHC for Minks
55
26%
Fish Ingestion NEHC for Eagles
71
34%
Human Health Results—Non-Cancer
Non-Cancer Reference Dose for Child
(recreational)
96
46%
Non-Cancer Reference Dose for Adult
(recreational)
82
39%
Non-Cancer Reference Dose for Child
(subsistence)
109
52%
Non-Cancer Reference Dose for Adult
(subsistence)
99
47%
Source: ERG, 2015d; ERG, 2015h; ERG, 20151.
Acronyms: MCL (Maximum contaminant level); N/A (Not Applicable, no exceedances at baseline conditions to compare option results); NEHC (No Effect
Hazard Concentration); NRWQC (National Recommended Water Quality Criteria).
a - The EA encompasses a total of 222 immediate receiving waters and loadings from 195 steam electric power plants (some of which discharge to multiple
receiving waters). The IRW model, which excludes the Great Lakes and estuaries, encompasses a total of 209 immediate receiving waters (183 rivers and
streams; 26 lakes, ponds, and reservoirs) and loadings from 188 steam electric power plants.
b - >0 to 15 percent reduction; 16 to 30 percent reduction; 31 to 45 percent reduction;
46 to 60 percent reduction|>60 percent reduction
7-20
-------
Section 7—Environmental Improvements Under the Final Rule
7.3.3 Selenium
Under the final rule, EPA estimates 136,000 pounds per year of selenium removals from
steam electric power plant discharges - a 97 percent reduction in annual loadings. Selenium is
one of the primary pollutants identified in the literature and by EPA as causing documented
environmental impacts to fish and wildlife from steam electric power plant discharges. EPA
estimates that immediate receiving water concentrations of total selenium will decrease under the
final rule by 71 percent on average, decreasing the amount of selenium that would bioaccumulate
or persist in the aquatic environment. Under the final rule, the number of immediate receiving
waters exceeding chronic aquatic life NRWQC will decrease by 55 percent and the number of
immediate receiving waters exceeding a drinking water MCL for selenium will decrease by 75
percent.
Reducing selenium loadings and
subsequent bioaccumulation will decrease by 52
percent the number of immediate receiving
waters with fish tissue concentrations exceeding
the NEHC for selenium for both eagles and
minks. These reductions also represent the
potential health improvements in other wildlife
that consume fish from these receiving waters, as
well as the potential decrease in bioaccumulation
of toxic pollutants in the broader food web near
steam electric power plants.
The results of the ecological risk model Selenium is known to cause fish deformities at
further support these predicted reductions in the high levels, such as these from Belews Lake,
bioaccumulative impact of selenium throughout NC.
the food web. Under the final rule, the ecological
risk modeling results indicate that:
• The risk of negative reproductive impacts among fish and/or mallards will be reduced
to less than one percent in each of the 26 modeled lentic immediate receiving waters.
• The number of immediate receiving waters that present a risk of reproductive impacts
among at least 10 percent of the exposed population will be reduced by 67 percent
(for fish) and 61 percent (for mallards).
• The number of immediate receiving waters that present a risk of reproductive impacts
among at least 50 percent of the exposed population will be reduced by 70 percent
(for fish) and 74 percent (for mallards).
These results are based on the median modeled egg/ovary selenium concentration in
exposed fish and mallards. Use of the 90th percentile modeled egg/ovary concentration, which
results in a higher predicted risk of reproductive impacts, shows similar improvements under the
final rule:
7-21
-------
Section 7—Environmental Improvements Under the Final Rule
• The risk of negative reproductive impacts among fish will be reduced to less than one
percent in all but one of the 26 modeled lentic immediate receiving waters.
• The number of immediate receiving waters that present a risk of reproductive impacts
among at least 10 percent of the exposed population will be reduced by 55 percent
(for fish) and 52 percent (for mallards). Under the final rule, none of the lentic
immediate receiving waters will pose this reproductive risk to fish or mallards.
• The number of immediate receiving waters that present a risk of reproductive impacts
among at least 50 percent of the exposed population will be reduced by 53 percent
(for fish) and 59 percent (for mallards).
Under the final rule, EPA estimates that fish selenium concentrations that pose a non-
cancer threat to subsistence fishers and recreational fishers will decrease in up to 53 and 56
percent of immediate receiving waters, respectively. This reduces the risk of developing non-
cancer health effects associated with selenium, such as pulmonary edema and lesions of the lung;
cardiovascular effects such as tachycardia; gastrointestinal effects including nausea, vomiting,
diarrhea, and abdominal pain; effects on the liver; and neurological effects such as aches,
irritability, chills, and tremors [U.S. EPA, 2000b]. Table 7-9 presents the key environmental
improvements resulting from selenium removals under the regulatory options.
7-22
-------
Section 7—Environmental Improvements Under the Final Rule
Table 7-9. Key Environmental Improvements for Selenium Under the Regulatory Options
Evaluation Benchmark
Modeled Immediate
Receiving Waters Exceeding
Benchmark Under Baseline
Conditionsa
Number
Percentage
Number of Immediate Receiving Waters Exceeding Benchmark
(Percent Reduction from Baseline Conditions) Under the Regulatory Options
Option A
Option B
Option C
Option D
Option E
Water Quality Results
Freshwater Acute NRWQC
No benchmark for comparison
N/A
N/A
Freshwater Chronic NRWQC d
33
Human Health Water and Organism
NRWQC
Human Health Organism Only NRWQC
Drinking Water MCL
Wildlife Results
Fish Ingestion NEHC for Minks
12
16%
30
(9%)
4%
7
(13%)
0%
1
(0%)
6%
10
(17%)
42
20%
40
(5%)
Fish Ingestion NEHC for Eagles
Negative Reproductive Effects in Fish'
42
20%
40
(5%)
24
11%
19
(21%)
Negative Reproductive Effects in
Mallards c
31
15%
26
(16%)
7-23
-------
Section 7—Environmental Improvements Under the Final Rule
Table 7-9. Key Environmental Improvements for Selenium Under the Regulatory Options
Evaluation Benchmark
Modeled Immediate
Receiving Waters Exceeding
Benchmark Under Baseline
Conditionsa
Number
Percentage
Number of Immediate Receiving Waters Exceeding Benchmark
(Percent Reduction from Baseline Conditions) Under the Regulatory Options
b
Option A
Option B
Option C
Option D
Option E
Human Health Results—Non-Cancer
Non-Cancer Reference Dose for Child
(recreational)
41
20%
39
(5%)
29
(29%)
Non-Cancer Reference Dose for Adult
(recreational)
32
15%
29
(9%)
Non-Cancer Reference Dose for Child
(subsistence)
55
26%
51
(7%)
39
(29%)
Non-Cancer Reference Dose for Adult
(subsistence)
43
21%
40
(7%)
30
(30%)
Source: ERG, 2015d; ERG, 2015h; ERG, 20151.
Acronyms: MCL (Maximum contaminant level); N/A (Not Applicable, no exceedances at baseline conditions to compare option results); NEHC (No Effect
Hazard Concentration); NRWQC (National Recommended Water Quality Criteria).
a - The EA encompasses a total of 222 immediate receiving waters and loadings from 195 steam electric power plants (some of which discharge to multiple
receiving waters). The IRW model, which excludes the Great Lakes and estuaries, encompasses a total of 209 immediate receiving waters (183 rivers and
streams; 26 lakes, ponds, and reservoirs) and loadings from 188 steam electric power plants.
46 to 60 percent reductionl>60 percent reduction
b - >0 to 15 percent reduction; 16 to 30 percent reduction; 31
c - These rows indicate the number of immediate receiving waters whose median modeled egg/ovary concentration is predicted to result in reproductive impacts
among at least 10 percent of the exposed fish or mallard population, as determined using the ecological risk model.
d - The EA analyses use the EPA recommended water quality criteria for selenium in the water column of 5 ug/L ~ in effect at the time of the modeling done,
both for the proposed rule in 2012, and the final rule in 2015. EPA used this criterion in its modeling for the final rule to allow for consistent comparisons
between the modeling done for the proposed rule and that done for the final rule. All modeling was done prior to EPA publishing new final draft criteria for
selenium on July 27, 2015. The new final draft criteria, which EPA now recommends, of 3.1 ug/L in freshwater flowing systems (rivers, streams) and 1.2 ug/L in
lakes and reservoirs, are lower than the criteria EPA used in these analyses. Had EPA conducted the modeling with these new recommended criteria, it would
have resulted in slightly greater estimated impacts (more exceedances of the new selenium criteria) than that revealed using the old criteria. As a result, this
would have led to slightly greater potential improvements due to control of selenium discharges under the final rule. Therefore, the estimates of the modeled
selenium impacts, and potential improvements of the final ELG, are conservative and tend, if anything, to underestimate both the impacts and the benefits.
7-24
-------
Section 7—Environmental Improvements Under the Final Rule
7.3.4 Cadmium
Under the final rule, EPA estimates 9,020 pounds per year of cadmium removals from
steam electric power plant discharges - a 68 percent reduction in annual loadings. At baseline
conditions, discharges of cadmium are the second largest toxic-weighted pollutant discharges
from the steam electric power generating industry among those pollutants evaluated in the EA
(see Section 3.2). The final rule will decrease the number of immediate receiving waters that
exceed acute and chronic NRWQC by up to 67 and 59 percent, respectively. The number of
immediate receiving waters with fish tissue concentrations that exceed NEHCs for minks and
eagles will decrease by 67 and 50 percent, respectively. Under the final rule, the number of
immediate receiving waters with fish containing cadmium concentrations that pose a risk of non-
cancer health effects will decrease by 53 to 70 percent, depending on the cohort. Table 7-10
presents the key environmental improvements resulting from cadmium removals under the
regulatory options.
7.3.5 Thallium
Under the final rule, EPA estimates 62,300 pounds per year of thallium removals from
steam electric power plant discharges - a 98 percent reduction in annual loadings. EPA estimates
that the final rule will decrease the number of immediate receiving waters exceeding human
health NRWQC and MCLs for thallium by up to 85 percent. Under the final rule, the number of
immediate receiving waters with fish containing thallium concentrations that can potentially
cause non-cancer health effects in humans (e.g., neurological symptoms, alopecia,
gastrointestinal effects, and reproductive and developmental damage) will decrease by up to 69
percent, depending on the cohort. Table 7-11 presents the key environmental improvements
resulting from thallium removals under the regulatory options.
7-25
-------
Section 7—Environmental Improvements Under the Final Rule
Table 7-10. Key Environmental Improvements for Cadmium Under the Regulatory Options
Evaluation Benchmark
Modeled Immediate
Receiving Waters Exceeding
Benchmark Under Baseline
Conditionsa
Number
Percentage
Number of Immediate Receiving Waters Exceeding Benchmark
(Percent Reduction from Baseline Conditions) Under the Regulatory Options b
Option A
Option B
Option C
Option E
Water Quality Results
Freshwater Acute NRWQC
Freshwater Chronic NRWQC
Human Health Water and Organism
NRWQC
29
4%
14%
No benchmark for comparison
6
(33%)
23
(21%)
N/A
6
(33%)
23
(21%)
N/A
6
(33%)
16
(45%)
N/A
Human Health Organism Only NRWQC
No benchmark for comparison
N/A
N/A
N/A
Drinking Water MCL
11
5%
7
(36%)
7
(36%)
6
(45%)
Wildlife Results
Fish Ingestion NEHC for Minks
Fish Ingestion NEHC for Eagles
2%
5
(17%)
3
(25%)
5
(17%)
3
(25%)
5
(17%)
3
(25%)
Human Health Results—Non-Cancer
Non-Cancer Reference Dose for Child
(recreational)
16
8%
12
(25%)
12
(25%)
9
(44%)
Non-Cancer Reference Dose for Adult
(recreational)
10
5%
7
(30%)
7
(30%)
6
(40%)
Non-Cancer Reference Dose for Child
(subsistence)
32
15%
26
(19%)
26
(19%)
19
(41%)
Non-Cancer Reference Dose for Adult
(subsistence)
22
11%
17
(23%)
17
(23%)
11
(50%)
Source: ERG, 2015d; ERG, 2015h; ERG, 20151.
Acronyms: MCL (Maximum contaminant level); N/A (Not Applicable, no exceedances at baseline conditions to compare option results); NEHC (No Effect
Hazard Concentration); NRWQC (National Recommended Water Quality Criteria).
a - The EA encompasses a total of 222 immediate receiving waters and loadings from 195 steam electric power plants (some of which discharge to multiple
receiving waters). The IRW model, which excludes the Great Lakes and estuaries, encompasses a total of 209 immediate receiving waters (183 rivers and
streams; 26 lakes, ponds, and reservoirs) and loadings from 188 steam electric power plants.
b - >0 to 15 percent reduction; 16 to 30 percent reduction; 31 to 45 percent reduction;
46 to 60 percent reduction|>60 percent reduction
7-26
-------
Section 7—Environmental Improvements Under the Final Rule
Table 7-11. Key Environmental Improvements for Thallium Under the Regulatory Options
Evaluation Benchmark
Modeled Immediate
Receiving Waters Exceeding
Benchmark Under Baseline
Conditionsa
Number | Percentage
Number of Immediate Receiving Waters Exceeding Benchmark
(Percent Reduction from Baseline Conditions) Under the Regulatory Options b
Option A | Option B | Option C | Option D | Option E
Water Quality Results
Freshwater Acute NRWQC
No benchmark for comparison
N/A
N/A
N/A
Freshwater Chronic NRWQC
No benchmark for comparison
N/A
N/A
Human Health Water and Organism
NRWQC
49
23%
46
(6%)
Human Health Organism Only NRWQC
45
22%
42
(7%)
Drinking Water MCL
34
16%
32
(6%)
Wildlife Results
Fish Ingestion NEHC for Minks
No benchmark for comparison
N/A
Fish Ingestion NEHC for Eagles
No benchmark for comparison
N/A
Human Health Results—Non-Cancer
Non-Cancer Reference Dose for Child
(recreational)
74
35%
73
(1%)
Non-Cancer Reference Dose for Adult
(recreational)
54
26%
51
(6%)
51
(6%)
Non-Cancer Reference Dose for Child
(subsistence)
94
45%
90
(4%)
90
(4%)
Non-Cancer Reference Dose for Adult
(subsistence)
77
37%
76
(1%)
76
(1%)
Source: ERG, 2015d; ERG, 2015h; ERG, 20151.
Acronyms: MCL (Maximum contaminant level); N/A (Not Applicable, no exceedances at baseline conditions to compare option results); NEHC (No Effect
Hazard Concentration); NRWQC (National Recommended Water Quality Criteria).
a - The EA encompasses a total of 222 immediate receiving waters and loadings from 195 steam electric power plants (some of which discharge to multiple
receiving waters). The IRW model, which excludes the Great Lakes and estuaries, encompasses a total of 209 immediate receiving waters (183 rivers and
streams; 26 lakes, ponds, and reservoirs) and loadings from 188 steam electric power plants.
b - >0 to 15 percent reduction; 16 to 30 percent reduction; 31 to 45 percent reduction;
46 to 60 percent reductioni>60 percent reduction
7-27
-------
Section 7—Environmental Improvements Under the Final Rule
7.4 IMPROVEMENTS TO SENSITIVE ENVIRONMENTS
As discussed in Section 3.4, EPA evaluated pollutant discharges to sensitive
environments (i.e., impaired waters, threatened and endangered species, and fish consumption
advisory waters) and sensitive watersheds (the Great Lakes and Chesapeake Bay). The purpose
was to assess if steam electric power plants discharge to receiving waters with existing
impairments or fish advisories and assess if discharges of the evaluated wastestreams increase
stress on threatened and endangered species. This section presents EPA's estimated pollutant
removals under five regulatory options to the evaluated sensitive environments.
The final rule will decrease pollutant loadings to sensitive environments, which will help
impaired waters to recover; decrease the bioaccumulation of toxic pollutants in fish, thereby
reducing the number of fish advisories; and reduce stress on threatened and endangered species
and sensitive watersheds such as Chesapeake Bay and the Great Lakes (see Section 7.5).
7.4.1 Impaired Waters
EPA determined that 59 of the immediate receiving waters are 303(d)-listed waterbodies,
designated as impaired for one or more pollutants found in the evaluated wastestreams.51
Mercury (30 immediate receiving waters), nutrients (19 immediate receiving waters), and
phosphorus (11 immediate receiving waters) are the most frequently identified impairment
categories among the surface waters that directly receive the evaluated wastestreams. Table 7-12
presents the pollutant removals to impaired waters (by impairment category) as a result of the
regulatory options.
Under the final rule, EPA estimates the following pollutant removals:
• Mercury removals of 168 pounds per year to mercury-impaired waters (decrease of
99 percent).
• Phosphorus removals of 4,100 pounds per year to nutrient-impaired waters (decrease
of 78 percent).
• Nitrogen removals of 471,000 pounds per year to nutrient-impaired waters (decrease
of 96 percent).
• Pollutant removals to receiving waters impaired for a metal (except mercury) include
4,100 pounds per year of arsenic (decrease of 95 percent); 1,770 pounds per year of
cadmium (decrease of 93 percent); 2,630 pounds per year of lead (decrease of 97
percent); 21,500 pounds per year of selenium (decrease of 97 percent); and 7,130
pounds per year of thallium (decrease of 97 percent).52
51 The count of impaired waters excludes the general impairment category "metals (not mercury)" and includes
receiving waters impaired for arsenic, boron, cadmium, chromium, copper, lead, manganese, mercury, selenium,
zinc, phosphorous, nutrients, TDS, or chlorides.
52 EPA presents pollutant loadings and removals for metals, other than mercury, for immediate receiving waters
designated as impaired for the general impairment category "metals (not mercury)" to protect confidential business
information. See all results in Table 7-12.
7-28
-------
Section 7—Environmental Improvements Under the Final Rule
Table 7-12. Pollutant Removals to Impaired Waters by Impairment Type
Impairment
Type/Number of
Receiving Waters b
Pollutant
Baseline
Loadings
(Ibs/yr)
Pollutant Removals (Ibs/yr) to Impaired Waters Under the Regulatory Options (Percent
Reduction)a
Option A
Option B
Option C
Option D
Option E
Mercury-Impaired Receiving Waters
30
Metals (Not Mercury)-Impaired Receiving Waters
7-29
-------
Section 7—Environmental Improvements Under the Final Rule
Table 7-12. Pollutant Removals to Impaired Waters by Impairment Type
Impairment
Type/Number of
Receiving Waters b
Pollutant
Baseline
Loadings
(Ibs/yr)
Pollutant Removals (Ibs/yr) to Impaired Waters Under the Regulatory Options (Percent
Reduction)a
Option A
Option B
Option C
Option D
Option E
Nutrient-Impaired Receiving Waters
TDS and Chlorides-Impaired Receiving Waters
Source: ERG, 2015c.
Acronyms: CBI (Confidential business information); Ibs/yr (pounds per year).
Note: Loadings and pollutant reductions are rounded to three significant figures.
46 to 60 percent reduction|>60 percent reduction
a - >0 to 15 percent reduction; 16 to 30 percent reduction; 31 to 45 percent reduction; |
b - For the impaired waters proximity analysis, EPA evaluated 222 immediate receiving waters that receive discharges of the evaluated wastestreams.
c - The EPA impaired water database listed 28 immediate receiving waters as impaired based on the "metal, other than mercury" impairment category. Of those
28 immediate receiving waters, 13 receiving waters are also listed as impaired for one or more specific metals (arsenic, cadmium, chromium, copper, lead,
manganese, selenium, and zinc). One additional immediate receiving water is impaired for boron (but not included in the "metals, other than mercury"
impairment category).
d - Total phosphorous and total nitrogen loadings are presented with this impairment category. Total nitrogen loadings are the sum of total Kjeldahl nitrogen and
nitrate/nitrite as N loadings.
7-30
-------
Section 7—Environmental Improvements Under the Final Rule
7.4.2 Threatened and Endangered Species
As discussed in Section 3.4.5, EPA identified 138 threatened and endangered species
whose habitats overlap with, or are located within, surface waters that exceeded NRWQC for the
protection of aquatic life under baseline conditions.53 To assess the potential improvements to
threatened and endangered species under the final rule, EPA initially selected only those species
identified as highly vulnerable to changes in water quality (75 of the 138 species) for evaluation.
EPA further excluded species from the analysis based on the following criteria: the species is
already presumed extinct, species habitat is unlikely to be affected by discharges of the evaluated
wastestreams (e.g., isolated headwaters), species listing status is due to habitat destruction
unrelated to steam electric power plant discharges (e.g., damming, stream channelization), and
other criteria. Based on the analysis, EPA identified 15 species out of the 75 that are highly
vulnerable to changes in water quality and whose recovery may be enhanced by the final rule.
Four of these 15 species inhabit waters that will no longer exceed NRWQC for the protection of
aquatic life following implementation of the final rule. The species may therefore experience
increases in population growth rates as a result of the final rule. See the Benefits and Cost
Analysis for further details on the methodology and results of EPA's threatened and endangered
species analysis.
7.4.3 Fish Advisory Waters
States, territories, and authorized tribes issue fish advisories to notify the public
(including recreational and subsistence fishers) of waterbodies containing fish with elevated and
potentially unhealthy contamination levels. Mercury is the most common pollutant found in
steam electric power plant wastewater for which fish advisories are issued to the surface waters
that receive the evaluated wastestreams (see Section 3.4.4). EPA determined that 88 of the 222
immediate receiving waters included in the EA are under a fish advisory for mercury. Under the
final rule, the number of immediate receiving waters with fish that exceed EPA's mercury
screening value for recreational fishers (based on steam electric power plant discharges only)
will decrease by 63 percent, thereby reducing the potential threat to human health from
consuming contaminated fish.
7.5 IMPROVEMENTS TO WATERSHEDS
As discussed in Section 3.4, both the Great Lakes and Chesapeake Bay watersheds have a
history of receiving pollutant discharges that negatively affect water quality, wildlife, and human
health. Both are well-studied, sensitive environments that are affected by pollutants commonly
found in steam electric power plant wastewater. Mercury is one of the primary pollutants of
concern in the Great Lakes,54 and nutrients are the primary pollutants of focus in the Chesapeake
Bay.
EPA identified 23 steam electric power plants that discharge into the Great Lakes
watershed. Table 7-13 presents the pollutant reductions to the Great Lakes watershed under the
53 The habitat locations evaluated for this analysis include waters downstream from steam electric power plant
discharges and reflect changes in the industry as a result of the Clean Power Plan [Clean Air Act Section 11 l(d)].
54 One of the main environmental pathways for mercury in the Great Lakes is from atmospheric deposition, which is
not in the scope of the final rule.
7-31
-------
Section 7—Environmental Improvements Under the Final Rule
regulatory options considered by EPA. Under the final rule, EPA estimates the following
pollutant removals to the Great Lakes watershed:
• 2,070 pounds of arsenic annually (96 percent reduction).
• 612 pounds of cadmium annually (95 percent reduction).
• 1,880 pounds of lead annually (99 percent reduction).
• 80.6 pounds of mercury annually (97 percent reduction).
• 4,800 pounds of selenium annually (96 percent reduction).
• 9,510 pounds of thallium annually (99 percent reduction).
• 1.15 million pounds of total nitrogen annually (>99 percent reduction).
• 21,800 pounds of total phosphorus annually (94 percent reduction).
EPA identified nine steam electric power plants that discharge to the Chesapeake Bay
watershed. Under the final rule, EPA estimates the following pollutant removals to the
Chesapeake Bay watershed:
• 2,430 pounds of arsenic annually (97 percent reduction).
• 476 pounds of cadmium annually (93 percent reduction).
• 1,540 pounds of lead annually (99 percent reduction).
• 87.1 pounds of mercury annually (98 percent reduction).
• 6,380 pounds of selenium annually (97 percent reduction).
• 5,220 pounds of thallium annually (99 percent reduction).
• 990,000 pounds of total nitrogen annually (>99 percent reduction).
• 14,900 pounds of total phosphorus annually (89 percent reduction).
7-32
-------
Section 7—Environmental Improvements Under the Final Rule
Table 7-13. Pollutant Removals to the Great Lakes Watershed Under the Regulatory Options
Pollutant
Arsenic
Boron
Cadmium
Chromium VI
Copper
Lead
Manganese
Mercury
Nickel
Selenium
Thallium
Zinc
Nitrogen, total
Phosphorus, total
Chlorides
Baseline Loadings
to the Great
Lakes Watershed
(Ibs/yr)
2,170
997,000
648
Pollutant Removals (Ibs/yr) to Great Lakes Watershed Under the Regulatory Options
(Percent Reduction)a
Option A
47.5 (2%)
9,190 (1%)
0.548
2,550
1,900
242,000
9,840
5,020
9,570
8,730
1,150,000
23,100
31,900,000
53.6 (8%)
0.471 (86%)
34.5 (1%)
Option B
47.5 (2%)
9,190 (1%)
19.4 (1%)
35,500 (15%)
4.56 (6%)
402 (4%)
53.6 (8%)
0.471 (86%)
34.5 (1%)
Option C
513 (24%)
22,600 (2%)
19.4 (1%)
35,500 (15%)
4.91 (6%)
183 (28%)
0.548 (>99%
608 (24%)
Option D
66,800 (7%)
612 (95'
449 (24%)
70,500 (29%)
22.6 (27%)
2,5100
1,880 (9<,
188,000 (7
80.6 (97
126 (3%)
780 (75%)
23.5
658 (8%)
2,420
135 (1%)
11,400
658 (8%)
380,000 (33%)
4,010 (80%)
2,200 (23%)
2,410 (28%)
556,000 (481!
135 (1%)
11,400
5,110(22%)
698,000 (2%)
9,510(9f
8,270 (9f
1,150,000 (>99
21,800 (94%)
3,000,000 (9%)
Option E
66,800 (7%)
3 (96%)
•8 (>99%)
20 (99%)
80 (99%)
000 (77%)
.2 (99%)
90 (99%)
00 (96%)
10 (99%)
00 (99%)
,000 (>99%)
21,800 (94%)
3,000,000 (9%)
TDS
186,000,000
3,890,000
3,890,000
22,300,000 (12%)
83,900,000 (45%)
83,900,000 (45%)
Source: ERG, 2015a; ERG, 2015c.
Acronyms: Ibs/yr (pounds per year).
Note: Loadings and pollutant removals are rounded to three significant figures.
a - >0 to 15 percent reduction; 16 to 30 percent reduction; 31 to 45 percent reduction; j
b - Total nitrogen loadings are the sum of total Kjeldahl nitrogen and nitrate/nitrite as N loadings.
7-33
-------
Section 7—Environmental Improvements Under the Final Rule
7.6 ENVIRONMENTAL AND HUMAN HEALTH IMPROVEMENTS IN DOWNSTREAM SURFACE
WATER
EPA estimates that the environmental and human health improvements in the immediate
receiving waters expected from the final rule will translate into considerable improvements in
water quality further downstream from steam electric power plant discharges. EPA calculated
downstream receiving water pollutant concentrations using EPA's Risk-Screening
Environmental Indicators (RSEI) model55 and compared these concentrations to the same
NRWQC and MCL water quality benchmarks used in the IRW model national-scale analysis.
EPA also evaluated the wildlife (mink and eagle NEHC benchmarks) and human health (cancer
and non-cancer) improvements in downstream surface waters using a simplified version of the
IRW model national-scale analysis. This approach involved calculating the water pollutant
concentrations that would result in exceedances if used as inputs to the wildlife and human
health modules in the IRW model; EPA then compared the downstream receiving water pollutant
concentrations in RSEI to these "threshold" concentrations to identify the downstream reaches
that would have at least one exceedance of a particular wildlife or human health benchmark.56
EPA used this approach to estimate the extent (in river miles) of environmental and human
health impacts in downstream surface waters under baseline conditions and the improvements
under the modeled regulatory options (Options A, B, C, D, and E). Table 7-14 presents the
results of this downstream analysis.
Based on the results of the downstream modeling, thousands of downstream river miles
are impacted by steam electric power plant discharges. Pollutant concentrations exceed NRWQC
for human health (water and organism) in 4,400 river miles downstream from immediate
receiving waters. However, under the final rule, this drops by 2,390 river miles (54 percent). The
final rule reduces the number of downstream exceedances for each of the NRWQCs and MCLs
evaluated. This reduction improves the water quality and aquatic habitats available to wildlife
and human populations located outside of the immediate vicinity of steam electric power plants.
In addition, pollutant removals under the final rule also reduce impacts to wildlife that rely on
downstream aquatic habitats as a food source. Up to 1,040 miles of surface waters downstream
from steam electric power plant discharges will no longer contain fish populations that exceed an
NEHC benchmark for minks or eagles. The final rule also decreases potential exposure of
humans to pollutants that can cause non-cancer health effects from consumption of contaminated
fish in up to 5,470 river miles. These results demonstrate that steam electric power plant
discharges are impacting surface waters beyond the immediate receiving waters. Pollutant
removals associated with the final rule will substantially improve the environmental and human
health for communities beyond the area immediately surrounding steam electric power plants.
55 EPA used pollutant loadings discharged to each receiving reach by steam electric power plants to estimate
concentrations in downstream reaches. The RSEI model uses a simple dilution and first-order decay equation to
calculate receiving water concentrations (metals are treated as conservative substances). The RSEI model assumes
that the plant's annual discharge is released at a constant rate throughout the year. In addition, EPA included
pollutant loadings from EPA's Toxics Release Inventory (TRI) database for other industries to represent background
pollutant concentrations in the downstream receiving waters. For further details on the RSEI model methodology
and assumptions, see the Benefits and Cost Analysis.
56 See the ERG memorandum "Downstream EA Modeling Methodology and Supporting Documentation" (DCN
SE04455) regarding the calculation of these water pollutant concentration thresholds.
7-34
-------
Section 7—Environmental Improvements Under the Final Rule
Table 7-14. Key Environmental Improvements for Downstream Waters Under the Regulatory Options
Evaluation Criteria
Number of River-
Miles Exceeding
Criteria Under
Baseline Conditions
Number of River-Miles Exceeding Criteria
(Percent Reduction from Baseline Conditions) Under the Regulatory Options a
Option A
Option B
Option C
Option D
Option E
Water Quality Results
Freshwater Acute NRWQC
417
396
(5%)
396
(5%)
390
(7%)
Freshwater Chronic
NRWQC
628
612
(3%)
569
(9%)
Human Health Water and
Organism NRWQC
4,400
3,670
(17%)
3,670
(17%)
Human Health Organism-only
NRWQC
1,560
1,300
(16%)
1,300
(16%)
Drinking Water
MCL
759
731
(4%)
726
(4%)
Wildlife Results
Fish Ingestion NEHC for Minks
1,180
917
(23%)
892
(25%)
Fish Ingestion NEHC for Eagles
2,000
Human Health Results—Non-Cancer
Non-cancer reference dose for
child (recreational)
6,350
1,730
(13%)
1,720
(14%)
4,900
(23%)
4,890
(23%)
Non-cancer reference dose for
adult (recreational)
3,760
2,960
(21%)
2,950
(21%)
Non-cancer reference dose for
child (subsistence)
10,100
Non-cancer reference dose for
adult (subsistence)
7,110
8,380
(17%)
8,350
(17%)
5,580
(22%)
5,570
(22%)
7-35
-------
Section 7—Environmental Improvements Under the Final Rule
Table 7-14. Key Environmental Improvements for Downstream Waters Under the Regulatory Options
Evaluation Criteria
Number of River-
Miles Exceeding
Criteria Under
Baseline Conditions
Number of River-Miles Exceeding Criteria
(Percent Reduction from Baseline Conditions) Under the Regulatory Options a
Option A
Option B
Option C
Option D
Option E
Human Health Results — Cancer
Cancer risk for child
(recreational)
Cancer risk for adult
(recreational)
Cancer risk for child
(subsistence)
Cancer risk for adult
(subsistence)
231
286
262
446
216
(7%)
263
(8%)
241
(8%)
383
(14%)
216
(7%)
263
(8%)
241
(8%)
383
(14%)
211
(9%)
251
(12%)
239
(9%)
358
(20%)
210
(9%)
246
(14%)
235
(10%)
328
(27%)
207
(10%)
245
(14%)
231
(12%)
304
(32%)
Source: ERG, 2015i; ERG, 20151.
Note: River miles are rounded to three significant figures.
a - >0 to 15 percent reduction; 16 to 30 percent reduction; 31 to 45 percent reduction; |
b - EPA evaluated a total of 73,000 river-miles in the downstream receiving water analysis for toxic, bioaccumulative pollutants. Downstream receiving water
concentrations are calculated until one of three conditions occurs: 1) the discharge travels 300 kilometers (km) downstream; 2) the discharge travels downstream
for a week; or 3) the concentration reaches 1 x 10"9 milligrams per liter (mg/L).
7-36
-------
Section 7—Environmental Improvements Under the Final Rule
7.7 ATTRACTIVE NUISANCES
EPA projects that the final rule will also decrease the environmental impact to wildlife
exposed to pollutants through direct contact with surface impoundments and constructed
wetlands at steam electric power plants. Multiple studies show that wildlife living near steam
electric surface impoundments exhibit elevated levels of arsenic, cadmium, chromium, lead,
mercury, selenium, strontium, and vanadium [Burger et al., 2002; Bryan et al., 2003; Hopkins et
al, 1997, 1998, 2000, 2002, 2006; Nagle etal, 2001; Rattner etal, 2006]. Multiple studies have
linked attractive nuisance areas at steam electric power plants to diminished reproduction
[Hopkins et al., 2002, 2006; Nagle et al, 2001]. While the final rule does not control pollutants
within surface impoundments or constructed wetlands prior to their discharge to surface waters,
EPA estimates that the final rule will decrease pollutant loadings to these waterbodies (e.g.,
through plants converting to dry handling their fly ash). These pollutant removals will decrease
the exposure of wildlife populations to toxic pollutants and decrease the threat that combustion
residual surface impoundments pose to surrounding wildlife.
7.8 OTHER SECONDARY IMPROVEMENTS
In addition to the improvements discussed above, other secondary, or ancillary, other
resources will see improvements that are associated directly or indirectly with the final rule.
Pollutant removals not only improve water quality in surface waters but enhances their aesthetic
(e.g., by improving clarity and decreasing odor and discoloration). Cleaner surface water
improves the source of drinking water for both surface water treatment plants and wells that are
influenced by surface water; water used for irrigation; and water used for industrial uses (less
contaminants). Recreational benefits from water quality improvements include more enjoyment
from swimming, fishing, and boating and potentially increased revenue from more people
partaking of recreational activities. The final rule may also reduce economic impacts such as
clean-up and treatment costs for contamination or impoundment failures, reduced injury
associated with surface impoundment failures, reduced water usage, reduced potential for algal
blooms, and decreased air emissions.
The Benefits and Cost Analysis monetizes benefits of implementing the final rule
(increased aesthetics, recreational improvements, increased availability of ground water
resources, reduced risk of surface impoundment failures, and air quality improvements). In
addition, the document also qualitatively discusses improvements to the quality of source water
for drinking, irrigation, and industrial use; quantity and quality of recreational opportunities;
improved commercial fisheries yields; increased property values; and reduced sediment
contamination within receiving waters.
While the final rule does not control pollutants leaching to ground water from surface
impoundments and landfills containing combustion residuals, EPA estimates that the final rule
will decrease pollutant loadings to surface impoundments (e.g., through plants converting to dry
handling their fly ash). These pollutant removals will decrease pollutants leaching from
combustion residual surface impoundments to ground water and decrease the potential human
health impacts associated with exposure to contaminated drinking water wells (see Section
3.3.4). EPA, however, did not quantify or monetize the benefits associated with this
improvement to ground water quality.
7-37
-------
Section 7—Environmental Improvements Under the Final Rule
7.9 UNRESOLVED DRINKING WATER IMPACTS DUE TO BROMIDE DISCHARGES
As discussed in Section 3.1.3, bromide in water can form brominated disinfection by-
products (DBFs), some potentially carcinogenic, when drinking water plants use certain
processes including chlorination and ozonation to disinfect the incoming source water. The
national effluent limitations guidelines and standards under the final rule (regulatory Option D)
do not directly control TDS levels (including bromides) in FGD wastewater discharges from all
steam electric power plants.57 Coal-fired steam electric power plants can discharge bromide due
to its natural presence in coal (which is released when burned and/or captured in particulates by
baghouses and FGD controls) or through bromide addition to flue gas control processes to reduce
mercury emissions. Steam electric power plant discharges occur close to more than 100 public
drinking water intakes on rivers and other waterbodies and there is evidence that bromide
discharges are already having adverse effects on the quality of drinking water sources.
While bromide itself is not thought to be toxic at levels present in the environment, its
reaction with other constituents in water may be of concern now and into the future. Drinking
water utilities should be concerned about bromides affecting drinking water sources, as bromide
loadings into surface waters could potentially increase in the future as more coal-fired steam
electric power plant operators add bromide to help control mercury emissions. Although EPA
decided not to finalize BAT requirements based on evaporation for treating FGD wastewater at
all steam electric power plants in the final rule, evaporation technology is potentially available
and may be appropriate for achieving water quality-based effluent limitations, depending on site-
specific conditions, where drinking water supplies need to be protected.
57 They do, however, directly control TDS in cases where steam electric power plants opt into the voluntary
incentives program, in which they would be subject to effluent limitations based on evaporation technology.
7^38
-------
Section 8—Case Study Modeling
SECTION 8
CASE STUDY MODELING
EPA developed dynamic water quality models of selected case study locations to
supplement the water quality component of the national-scale immediate receiving water (IRW)
model. EPA performed the case study modeling to provide additional resolution regarding the
baseline impacts and the expected environmental and human health improvements under the
final rule, while encompassing a broader temporal and spatial scope than what is included in the
IRW model. The case study models also validate and provide additional perspective on the
results of the IRW model for those waterbodies included in both models. The case study
modeling improves upon the IRW model in the following ways:
• Accounts for long-term pollutant loadings from steam electric power plants (under
both baseline conditions and the final rule) and estimates the resultant accumulation
of pollutants within the water column and sediments of the receiving water. These
models can more accurately assess baseline pollutant concentrations and the time
frame and magnitude of environmental improvements associated with the final rule.
• Accounts for fluctuations in receiving water flow rates by using daily stream flow
monitoring data instead of one annual average flow rate for the receiving water. This
approach better reflects the varying influence of dilution (or lack thereof) within the
receiving water during high-flow and low-flow conditions.
• Accounts for pollutant transport and accumulation within receiving water reaches that
are downstream from the discharge location. This approach can more accurately
estimate the river distance showing environmental impacts under baseline conditions
and improvements under the final rule.58
• Accounts for pollutant contributions from other point, nonpoint, and background
sources, to the extent practical, using available data sources. Incorporating non-
steam-electric pollutant sources and available water quality data provides a more
complete illustration of the compounding impacts of background pollutant
concentrations, steam electric power plant pollutant loadings, and other point source
dischargers.
This section describes EPA's methodology for developing and running the case study
models (Section 8.1); presents the results of the case study models for the selected case study
locations (Section 8.2); and compares the case study and IRW model results (Section 8.3).
The case study downstream modeling described in this section is separate from the downstream modeling EPA
performed using the Risk-Screening Environmental Indicators (RSEI) model and the SPARROW (SPAtially
Referenced Regressions On Watershed attributes) model. EPA used the national-scale RSEI and SPARROW models
to quantify changes in water quality in support of the benefits analysis for the final rule. See the Benefits and Cost
Analysis for the Effluent Limitations Guidelines and Standards for the Steam Electric Power Generating Point
Source Category (EPA-821-R-15-005).
8-1
-------
Section 8—Case Study Modeling
8.1 CASE STUDY MODELING METHODOLOGY
The case studies use EPA's Water Quality Analysis Simulation Program (WASP), a
dynamic compartment-modeling program for aquatic systems that simulates pollutant fate and
transport within both the water column and the benthic sediment. The WASP model helps users
interpret and predict water quality responses to natural phenomena and man-made pollution for
various pollutant management decisions. EPA's approach also relies on U.S. Geological Survey
(USGS) daily stream flow data downloaded through EPA's Better Assessment Science
Integrating Point and Nonpoint Sources (BASINS) interface to provide input time series flow
data for use in the WASP model.
This section is organized as follows:
• Section 8.1.1 discusses EPA's approach for selecting case study locations (i.e., steam
electric power plants and receiving waters) for case study modeling, including the
differences in selection criteria for lotic, lentic, and estuarine water systems.
• Section 8.1.2 summarizes the scope and general technical approach for the case study
modeling, including the selection of pollutants and wastestreams for modeling; the
data sources evaluated for non-steam-electric pollutant contributions; and approaches
for modeling pollutant levels before and after the assumed final rule compliance date.
• Section 8.1.3 explains the development and execution of the case study models using
WASP. Appendix G provides additional information regarding the specific input
parameters (e.g.., background pollutant concentrations, USGS time series flow data)
and model settings (e.g.., solids transport parameters) for each of the WASP models.
For additional documentation regarding the selection and calculation of the input
parameters and settings, refer to the ERG memorandum, "Technical Approach for
Case Study Water Quality Modeling of Aquatic Systems in Support of the Final
Steam Electric Power Generating Industry Environmental Assessment" (DCN
SE05570) (Case Study Water Quality Modeling Memorandum).
• Section 8.1.4 describes the use of the case study model outputs to determine impacts
to aquatic life based on changes in water quality; impacts to aquatic life based on
changes in sediment quality; impacts to wildlife from consuming contaminated
aquatic organisms; and impacts to human health from consuming contaminated fish.
• Section 8.1.5 lists some of the limitations and assumptions involved with EPA's case
study modeling.
8.1.1 Selection of Case Study Locations for Modeling
To select locations for detailed case study modeling, EPA developed site-selection
criteria to identify a collection of steam electric power plants and receiving waters that, when
evaluated as a group:
• Represent a reasonable cross-section of the range of receiving waters evaluated in the
environmental assessment (EA).
• Illustrate pollutant removals across the regulatory options evaluated by EPA.
8-2
-------
Section 8—Case Study Modeling
• Encompass discharges of all four wastestreams evaluated in the EA.
• Demonstrate pollutant loadings that are representative of those discharged by steam
electric power plants evaluated in the EA (i.e., discharges are typical of steam electric
power plants and not outlier values).
EPA evaluated 195 steam electric power plants that discharge directly to aquatic systems
with lotic characteristics (rivers and streams), lentic characteristics (lakes, ponds, and reservoirs),
or that are estuarine systems. Through the site-selection process described below, EPA identified
six representative case study locations (five lotic sites and one lentic site) that capture
improvements across multiple regulatory options, represent all four evaluated wastestreams (flue
gas desulfurization (FGD) wastewater, fly ash transport water, bottom ash transport water, and
combustion residual leachate), and represent both lentic and lotic aquatic environments. Figure
8-1 and Table 8-1 present the six receiving waters that EPA selected for case study modeling.
5) Mississippi River,
MO/IL
2) Etowah RivelvGA?
6) Lake Sinclair, GA
LEGEND
Receiving Waters Selected
for Case Study Modeling
Figure 8-1. Overview of Case Study Modeling Locations
8-3
-------
Section 8—Case Study Modeling
Table 8-1. Locations Selected for Case Study Modeling
Case Study
Location
Black Creek,
MS
Etowah River,
GA
Lick Creek &
White River, IN
Ohio River,
PA/WV/OH
Mississippi
River, MO/IL
Lake Sinclair,
GA
Water-
body
Type
Lotic
Lotic
Lotic
Lotic
Lotic
Lentic
Steam Electric
Power Plant(s)
Modeled
R.D. Morrow Sr.
Generating Site
Plant Bowen
Petersburg
Generating Station
Bruce Mansfield
Plant & W.H.
Sammis Plant
Rush Island b
Plant Harllee
Branch °
Evaluated Wastestreams Discharged
FGD
-------
Section 8—Case Study Modeling
Selection of Lotic Case Study Locations
To select lotic receiving waters to model using WASP, EPA reviewed all combinations of
steam electric power plants and their receiving waters evaluated in the EA for factors that would
negatively influence the ability to use WASP for case study water quality modeling or the ability
to discuss the case study modeling results in a public document. EPA completed an assessment
using industry responses to the 2010 Questionnaire for the Steam Electric Power Generating
Effluent Guidelines (the Steam Electric Survey), EPA's BASINS tool, National Hydrography
Dataset Plus (NHDPlus Version 1) hydrography layers, and USGS National Water Information
System (NWIS) data sources to identify and eliminate the lotic receiving waters that met one or
more of the following criteria from consideration for case study modeling:
• Confidential Business Information (CBI). EPA identified and eliminated steam
electric power plants with CBI claims on discharge flow rate data for any of the four
evaluated wastestreams. EPA eliminated these plants as potential case study locations
because CBI data, including modeled water concentrations based on CBI data, cannot
be discussed in a public document such as this EA report.
• Stream gage flow data. EPA identified and eliminated receiving waters that lack
sufficient stream gage flow data. Availability of a long-term, continuous stream flow
record for both the receiving water being modeled and any significant downstream
tributaries was a major factor in selecting case study locations because these data are
needed to construct the hydrodynamics in WASP. The primary considerations when
reviewing the sufficiency of stream gage flow data for use in WASP were the
following:
Location of USGS stream gage stations (the ideal location is within the vicinity of
the immediate receiving water being evaluated, plus additional locations within
the model area).
- A continuous stream flow record covering a time period that matches or exceeds
the length of the desired modeling period.
- Age of the stream gage flow data (data sets without data from within the previous
30 years were considered potentially unrepresentative of current flow conditions).
• Downstream waterbody characteristics. WASP's ability to accurately model water
quality using USGS stream gage flow data can be affected by flow control structures
such as dams that affect the linear flow and circulation of water, and thus influence
the transport of pollutants. EPA identified and eliminated receiving waters whose
downstream waterbodies exhibit these characteristics, unless the areas of concern
were sufficiently downstream to allow for modeling of a reasonable distance (i.e., at
least 25 miles) before encountering the area of concern.
• Influence by other point source dischargers that could not be modeled. EPA identified
receiving waters that could be significantly influenced by discharges from other point
sources (including other steam electric power plants) and evaluated whether those
point sources would meet the criteria listed above for case study modeling. If EPA
determined that a receiving water would be significantly influenced by other point
source discharges that could not be modeled (e.g., an upstream steam electric power
8-5
-------
Section 8—Case Study Modeling
plant exercising CBI claims) or represented in the model by STORET monitoring
data (see Section 8.1.3), EPA eliminated the receiving water from consideration. If
EPA deemed the pollutant loadings from the other point source discharges to be
insignificant compared to the steam electric power plant pollutant loadings being
evaluated, EPA included the receiving water in the analysis.59
Next, EPA assessed the representativeness of the steam electric power plants and
receiving waters that were not eliminated based on the criteria above. EPA selected the receiving
water flow rate, magnitude of pollutant loadings from the evaluated wastestreams, and water
column concentrations output calculated based on these values as the primary factors in
determining whether it considered a particular receiving water representative. EPA reviewed the
average annual flow rates (as defined in NHDPlus Version 1), baseline loadings of the modeled
pollutants, and water column concentrations output from the IRW model of each of the steam
electric power plants and receiving waters that were not eliminated after application of the
acceptance criteria. EPA assessed how each plant and receiving water compared to the general
population in the EA and eliminated plant and receiving water combinations that did not
reasonably represent typical conditions. From the population of lotic receiving waters that EPA
determined would be suitable for WASP modeling and representative of typical pollutant
loadings from discharges of the evaluated wastestreams, the Agency selected a collection that,
when evaluated as a group, demonstrated pollutant removals across all modeled regulatory
options and all four evaluated wastestreams. As a result, EPA identified five case study locations
as the best candidates for modeling as part of a representative set of steam electric power plants
that discharge to lotic systems. The selected case study locations are further described in Section
8.2.60 Additional information about EPA's methodology for selecting plants and receiving waters
that are representative and suitable for WASP modeling is further described in the Case Study
Water Quality Modeling Memorandum (DCN SE05570).
Selection of Lentic andEstuarine Case Study Locations
Water quality modeling of lentic systems (lakes, ponds, and reservoirs) or estuarine
systems involves more complex hydrodynamics that would not be adequately represented by
stream gage flow data. Modeling steam electric power plants that discharge to lentic or estuarine
systems requires using existing EPA-developed WASP models (or more specifically, the
underlying hydrodynamic data) for the specific waterbodies of interest. Accordingly, EPA
considered the availability of existing models a primary factor in selecting lentic and estuarine
systems for case study water quality modeling.
59 EPA considered receiving water flow rate, distance between outfalls, and relative magnitude of pollutant loadings
when assessing whether the discharges from upstream or downstream plants or point sources could significantly
affect the water quality modeling results for the selected case study location. EPA applied best professional
judgment using these criteria, but did not apply numeric thresholds.
60 Because of the level of effort required to design, execute, and evaluate the outputs for case study modeling, EPA
did not complete case study modeling for all candidates that met all acceptance criteria and were determined to be
representative. EPA used best professional judgment in determination of which five case study locations were the
best candidates for modeling and represent a reasonable cross-section of the range of receiving waters evaluated in
theEA.
8-6
-------
Section 8—Case Study Modeling
EPA identified one preexisting WASP model for a lake (Lake Sinclair, GA) that receives
steam electric power plant discharges from Georgia Power Company's Plant Harllee Branch. As
of April 16, 2015, this plant has decertified and retired all four of its coal-fired generating units.
Based on a review of the water concentration outputs generated by the IRW model in support of
the proposed ELGs (which were developed prior to the announcement of plans to retire Plant
Harllee Branch), EPA determined that Lake Sinclair remains a representative illustration of
lentic waterbodies that receive discharges of the evaluated wastestreams. As discussed in Section
3, pollutant loadings to lentic systems often more strongly affect water quality and ecosystem
health (compared to lotic systems) due to the longer residence times and associated long-term
accumulation of pollutants in these systems. Accordingly, and despite the retirement of Plant
Harllee Branch, EPA proceeded with case study modeling of Lake Sinclair to represent the
potential impacts of steam electric power plant discharges on lentic waterbodies (including the
26 lake, pond, and reservoir receiving waters evaluated in this EA) and the potential
environmental improvements under the final rule in other lentic waterbodies that receive
discharges of the evaluated wastestreams.
EPA also identified one preexisting water WASP model for an estuary (Hillsborough
Bay, FL) that receives steam electric power plant discharges. However, due to the hydrologic
complexity of the model, and because estuarine systems represent less than 2 percent of the
receiving waters evaluated in the EA, EPA elected to develop only freshwater river and lake
WASP models for this case study analysis. Additionally, the ecological risk modeling approach
described in Section 5.2 is based on selenium bioaccumulation within freshwater environments
and would not be appropriate to apply to estuarine or marine aquatic systems, which would limit
EPA's ability to analyze the ecological effects for the estuarine case study.
8.1.2 Scope and Technical Approach for Case Study Modeling
This section describes the scope and technical approach used for EPA's detailed case
study modeling, including the selection of pollutants and wastestreams evaluated, the inclusion
of other point and nonpoint sources, the development of a historical baseline for the case study
location, and the prediction of decreased water and sediment pollutant concentrations under the
regulatory options evaluated for the final rule.
Selection of Pollutants for Modeling
EPA approached the case study modeling with the goal of modeling the same 10
pollutants included in the IRW model, which are listed in Section 5.1. As described later in this
section, however, EPA was unable to perform case study modeling for chromium VI and
mercury. EPA performed case study water quality modeling for the following eight pollutants (or
"toxicants" as defined in the WASP model), which were also included in the IRW model:
• Arsenic (As).
• Cadmium (Cd).
• Copper (Cu).
• Lead (Pb).
• Nickel (Ni).
• Selenium (Se).
8-7
-------
Section 8—Case Study Modeling
• Thallium (Tl).
• Zinc (Zn).
These pollutants can be modeled using the Simple Toxicant module within WASP.
Similar to the water quality module of the IRW model, the Simple Toxicant module applies
pollutant-specific partition coefficients to estimate the degree to which pollutants in the water
column will adsorb to benthic sediments and suspended solids. Unlike the IRW model, the
Simple Toxicant module does not incorporate separate partition coefficients to define the benthic
sediment/pore water equilibrium and the suspended sediment/water column equilibrium.
Therefore, EPA selected only the suspended sediment-water (Kdsw) partition coefficient for each
pollutant (see Table C-4 in Appendix C).
EPA also considered using WASP to perform water quality modeling for chromium VI
and mercury. These pollutants, however, require using more data-intensive modules within
WASP. Accurately modeling chromium VI requires using the META4 module within WASP to
accurately predict pollutant speciation and depends on the availability of extensive site-specific
monitoring data. Modeling mercury (and methylmercury, a bioaccumulative organic form of
mercury) requires using the MERC7 module within WASP to account for transformation
processes such as methylation. Using the more data-intensive modules requires site-specific data
that were not available for all locations.
Evaluated Wastestreams
The case study models quantified the water quality impacts resulting from discharges of
the same four evaluated wastestreams included in the IRW model:
• Fly ash transport water.
• Bottom ash transport water.
• FGD wastewater.
• Combustion residual leachate.
As with the IRW model, EPA performed the WASP water quality modeling using
average daily pollutant loadings derived from average annual pollutant loadings and normalized
effluent flow rates. This assumption of a static loadings rate does not account for temporal
variability in the loadings to receiving waters due to factors such as variable plant operating
schedules, storm flows, low-flow events, and catastrophic events.
Inclusion of Other Point andNonpoint Sources
Accounting for pollutant contributions from non-steam-electric point sources and
nonpoint sources, to the extent practical using available data, can improve the accuracy of the
case study water quality models. EPA identified the following data sources that provide pollutant
loadings and/or concentration data for these other sources potentially affecting water quality in
the case study location:
• Discharge Monitoring Reports (DMR). Point source dischargers are required to report
certain wastewater monitoring data through the submittal of DMRs. However, they
are required to report only for the pollutants that are listed in the facility's National
-------
Section 8—Case Study Modeling
Pollutant Discharge Elimination System (NPDES) permit.61 EPA evaluated 2011
pollutant loadings data for direct dischargers including publicly owned treatment
works (POTWs) and industrial facilities.
• Toxics Release Inventory (TRI). TRI collects facility-reported estimates of
wastewater loadings data for both direct and indirect dischargers. The TRI database
does not include loadings from facilities with total annual chemical releases of less
than 500 pounds and incorporates assumptions regarding plants with annual releases
of less than 1,000 pounds. The point source loadings from smaller facilities, therefore,
may not be well represented in the TRI database.62 EPA evaluated 2011 pollutant
loadings data for industrial facilities with indirect discharges of a modeled pollutant.
EPA also evaluated TRI direct pollutant loadings data for these facilities and
pollutants if the facilities are not also required to report this pollutant in their DMRs
(to avoid double-counting direct discharges).
• STORET Monitoring Data. EPA's STORET database is a repository for water
quality, biological, and physical data compiled from many data sources and locations
throughout the country. The STORET database contains water quality and sediment
quality monitoring data for all eight modeled pollutants and other input parameters
for WASP including total organic carbon (TOC) and total suspended solids (TSS).
EPA reviewed these publicly available data sources to identify pollutant contributions
from non-steam-electric point sources and nonpoint sources that may impact the case study water
quality model. EPA also used available STORET monitoring data to help calibrate the modeled
outputs. For additional documentation regarding EPA's collection and use of these data, refer to
the Case Study Water Quality Modeling Memorandum (DCN SE05570).
Modeling of Pollutant Loadings Prior to the Final Rule
EPA developed and executed WASP models (as described in Section 8.1.3) for the
selected case study locations to predict the baseline accumulation of pollutants in the receiving
water and sediment leading up to implementation of the final rule.
The modeling periods start at 1982 (the year of the last revision to the steam electric
ELGs) or the date of installation of the most recent generating unit impacted by this rulemaking
(if after 1982), and extend to the assumed compliance date.63 If the available stream gage flow
61 In addition, states (or other permitting authorities) have some discretion as to which data they make available (or
enter) to the national database (i.e., Permit Compliance System (PCS) and Integrated Compliance Information
System for the National Pollutant Discharge Elimination System (ICIS-NPDES)). For example, permitting
authorities enter DMR and permit information for facilities that are considered major dischargers. However, they do
not necessarily enter DMR or permit information into PCS for minor dischargers or facilities covered by a general
permit.
62 Other limitations of the data collected in TRI include the following: small establishments are not required to
report, nor are facilities that do not meet reporting thresholds; releases reported are based on estimates, not
measurements; certain chemicals are reported as a class, not as individual compounds; facilities are identified by
North American Industrial Classification System (NAICS) code, not point source category; and TRI requires
facilities to only report certain chemicals and therefore all pollutants discharged from a facility may not be captured.
63 For each steam electric power plant in the case study modeling, EPA assumed a plant-specific date, derived from
the plant's permitting cycle, that the plant would achieve the limitation under the final rule.
-------
Section 8—Case Study Modeling
data did not cover the desired modeling period, EPA extrapolated the available data,
incorporating another partial cycle of the flow data to reach the total desired modeling period.
Historical pollutant loadings data for the evaluated wastestreams and non-steam-electric
point sources are very limited and difficult to obtain, so EPA used Steam Electric Survey data
(representing plant operations in 2009), STORET monitoring data, and 2011 TRI and DMR
loadings data as a representative set of discharge conditions. EPA acknowledges that these data
may not reflect the actual pollutant loadings over the entire modeling period; however, they
represent an appropriate estimation of annual pollutant loadings and how discharges may affect
individual aquatic systems over time.
For each case study location, EPA assumed that the annual, historical pollutant loadings
associated with fly ash transport water, bottom ash transport water, and combustion residual
leachate discharges were equal to the baseline pollutant loadings calculated for these
wastestreams (i.e.., the same annual pollutant loadings used to represent baseline conditions in the
national-scale IRW model). The impoundment and discharge of these wastestreams has been a
standard technique practiced since before 1982. EPA did not attempt to determine whether a
modeled plant had historical discharges of an evaluated wastestream that are not represented in
the baseline pollutant loadings. For example, for a plant that does not have fly ash transport
water pollutant loadings under baseline conditions, EPA did not attempt to determine whether
the plant had historical discharges of fly ash transport water.
In estimating the annual, historical pollutant loadings associated with FGD wastewater,
EPA accounted for the fact that steam electric power plants may have installed FGD systems
after the start of the modeling period. EPA used the FGD system installation dates, based on
industry responses to the Steam Electric Survey, to determine how to incorporate FGD
wastewater pollutant loadings into the case study model. If a plant installed multiple FGD
systems during the modeling period, EPA assumed that the annual, historical FGD wastewater
pollutant loadings associated with each individual system were proportional to that system's flow
rate contribution compared to the total FGD wastewater flow rate under baseline conditions. The
procedure for calculating and incorporating the proportional loadings for each FGD system is
further described in the Case Study Water Quality Modeling Memorandum (DCN SE05570).
EPA accounted for pollutant loadings from non-steam-electric point sources within the
modeling boundary by using 2011 TRI and DMR data. EPA assumed that the annual, historical
pollutant loadings for these point sources throughout the modeling period were equal to the
pollutant loadings reported in the 2011 TRI and DMR data sets. To account for contributions
from nonpoint sources, EPA evaluated STORET water quality monitoring data collected
upstream of the modeling boundary. The Agency used these monitoring data to represent the
pollutant contributions from all point, nonpoint, and background sources upstream of the
monitoring location, potentially avoiding the need to collect TRI and DMR pollutant loadings
data and perform WASP modeling of those upstream or tributary reaches. The Case Study Water
Quality Modeling Memorandum (DCN SE05570) further discusses how EPA incorporated DMR
pollutant loadings data, TRI pollutant loadings data, and STORET monitoring data into the
WASP water quality models.
8-10
-------
Section 8—Case Study Modeling
The results of this baseline modeling provided initial receiving water and sediment
concentrations for modeling discharges after the assumed compliance date, discussed in the
following section.
Modeling of Pollutant Loadings Under the Final Rule
EPA developed and executed WASP water quality models (as described in Section 8.1.3)
for the selected case study locations to predict the decreases of receiving water and sediment
pollutant concentrations (relative to baseline conditions) following implementation of the final
rule.
EPA executed separate models for continued baseline pollutant loadings and regulatory
option pollutant loadings (Options A through D)64. These modeling periods started at the
assumed compliance date, as determined by each steam electric power plant's permitting cycle,
and continued for at least 10 years after the assumed compliance date. EPA used the pollutant
loadings calculated under the regulatory options to represent the annual steam electric pollutant
loadings for each year of the period following implementation of the final rule. EPA assumed
that the pollutant contributions from non-steam-electric point sources (based on TRI and DMR
data) and from nonpoint sources (based on STORET monitoring data) would remain constant
and would be equal to those used to model the period leading up to implementation of the final
rule.
8.1.3 Development and Execution of WASP Models
EPA built each case study model using the BASINS setup tool for WASP, known as the
WASP Model Builder, which allows the user to open WASP directly from the BASINS
interface. As described in Section 8.1.2, EPA's approach used the Simple Toxicant module
within WASP for the eight modeled pollutants. The Simple Toxicant module puts stretches of
the modeled receiving water into segments based on the hydrologic characteristics. The WASP
model calculates the water column and benthic pollutant concentrations using user-defined
parameters and default assumption values. The process described in this section is based on
using WASP Version 7.52 and BASINS Version 4.1. Both represent the most current versions
available for EPA's analysis.
EPA followed the general approach described below in developing the WASP models for
each of the lotic case study locations:
• WASP calculates receiving water and sediment concentrations by dividing the
waterbody into segments and performing calculations for each segment. EPA used
NHDPlus Flowlines as the basis for defining waterbody segments. To maintain
reasonable model runtimes and reduce system instability, EPA further refined these
segments by combining short segments such that the flow time through each segment
is at least a tenth of a day. In some cases, segment travel times were shorter than the
64 Case study modeling omitted Option E because EPA determined that the additional pollutant removals for Option
E are only marginally better than Option D. Under Option E, only R.D. Morrow Generating Station and W.H.
Sammis plant would have additional removals.
-------
Section 8—Case Study Modeling
desired minimum because the segment was located between an upstream and
downstream tributary of some significance.
• EPA used USGS stream gage flow data to represent inflows at the upstream end of
the case study location, as well as any significant tributary with a USGS stream gage
station. In all cases, EPA scaled the stream gage flow data to account for the
difference in drainage area between the actual gage location and the point where the
contributing flow enters the model.
• For those tributaries without available USGS stream gage flow data for the simulation
period, EPA set the flow rate equal to the average annual flow rate as per NHDPlus
Version 1.
• To simplify the geographic extent of the modeling area, EPA did not model any
tributaries with mean annual flow rates of less than 5 cubic feet per second (cfs) as
per NHDPlus Version 1.
• EPA used stream gage flow data from the actual time period (e.g., 1982 - 2014) to
represent the baseline flow rate in the modeling area. EPA reused the historical flow
data to the extent necessary to complete the modeling period through the assumed
compliance date (e.g., 2015 - 2020), preferentially selecting flow data from periods
that excluded years of particularly high or low flow rates. Then EPA reused the
historical flow data to represent the period through the end of the model run (e.g.,
2020 - 2036). This approach ensured that the modeling periods before and after the
assumed compliance date were based on similar flow data.
• To represent non-steam-electric point sources within the modeling area, EPA
assigned the TRI and DMR pollutant loadings to the stream reach (as represented in
NHDPlus Version 1) that was closest to the location of the point source.
• EPA used STORET monitoring data, where available, to represent pollutant
contributions flowing into the modeling area from upstream point sources, nonpoint
sources, and background sources. Prior to incorporation into the WASP model, EPA
converted the pollutant concentrations to mass loadings (for all pollutants except
TOC and TSS) using the annual average flow rate for the stream segment where the
sample was collected (as represented in NHDPlus Version 1). This approach ensured
that the modeled pollutant concentrations flowing into the modeling area would vary
with changes in the stream flow rate.
• To define initial concentrations for the organic solids, sands, and silts/fines
parameters, EPA used TOC and TSS concentrations derived from STORET
monitoring data collected within the modeling area.
• EPA calibrated the WASP water quality models by modifying the solids transport
input parameters until the modeled pollutant concentrations in the benthic segments
closely matched the sediment concentrations derived from STORET monitoring data.
The existing WASP model used for Lake Sinclair already divides the waterbody into
segments and an existing Environmental Fluid Dynamics Code (EFDC) model provides
hydrodynamics for the lentic system. Using an existing model of a lentic system was a
reasonable approach to investigate the regulatory options without developing a detailed model
8-12
-------
Section 8—Case Study Modeling
from scratch. However, this approach does limit the modeling period to the period simulated in
the existing EFDC model. Other than these differences, the approach for developing the WASP
model for the lentic system was similar to the approach described above for lotic systems.
EPA developed the WASP water quality models (for both lotic and lentic systems) to
provide output data for pollutant concentration (total, dissolved, and sorbed) in the water column
and benthic segments on a daily output time step. The WASP models generate these outputs for
both the immediate receiving water and every downstream segment. As described in Section
8.1.2, EPA then executed the models to represent conditions before and after implementation of
the final rule.
Appendix G provides additional information regarding the specific input parameters (e.g.,
background pollutant concentrations, USGS time series flow data) and model settings (e.g.,
solids transport parameters) for each of the WASP water quality models. For additional
documentation regarding the use or bypassing of specific WASP model features, incorporating
stream gage flow and pollutant loadings data, and default settings and assumptions, refer to the
Case Study Water Quality Modeling Memorandum (DCN SE05570).
8.1.4 Use of WASP Water Quality Model Outputs
For each modeled segment, EPA used the water column and benthic sediment pollutant
concentration outputs (for baseline and Option D, both from the WASP model run representing
the time period after the assumed compliance date) to perform the following environmental and
human health analyses:
• EPA compared the modeled pollutant concentrations in the water column (daily
outputs) to the water quality benchmarks listed in Table C-7 of Appendix C and
calculated the frequency of exceedances over the entire modeling period (i.e., the
percentage of days that have a modeled exceedance).
• EPA compared the modeled pollutant concentrations in the benthic sediment (daily
outputs) to the sediment biota chemical stressor concentration limit (CSCL)
benchmarks listed in Table D-2 of Appendix D and calculated the frequency of
exceedances over the entire modeling period (i.e., the percentage of days with a
modeled exceedance).
• EPA compared the modeled pollutant concentrations in the water column (averaged
over the entire modeling period) to the water pollutant concentrations that would
result in exceedances if used as inputs to the wildlife and human health modules in
the IRW model (as described in Section 7.6).
For the Black Creek case study, which had relatively high concentrations of selenium
compared to the other selected case studies, EPA also performed ecological risk modeling
following the methodology described in Section 5.2.
Using the WASP water quality outputs in these analyses allowed EPA to evaluate, with
greater focus and accuracy, the potential for additional environmental and human health impacts
that were not reflected in the IRW model outputs. These included impacts associated with peak
pollutant concentrations during low-flow periods; long-term accumulation of pollutants in
-------
Section 8—Case Study Modeling
benthic sediment; impacts in downstream receiving waters; and pollutant contributions from
non-steam-electric sources.
8.1.5 Limitations of Case Study Modeling
The results of the case study models are intended to illustrate the types and magnitudes of
environmental impacts that are likely to have occurred, and which may continue to occur, in
surface waters that receive discharges of the evaluated wastestreams from steam electric power
plants. Similarly, the case study modeling results provide valuable information regarding the
relative magnitude of water quality improvements predicted for each of the regulatory options.
In developing the case study models, EPA found it necessary to incorporate several
assumptions that simplified the modeling approach while introducing uncertainty into the model
results. For example, due to a lack of data regarding temporal variability in point source
loadings, EPA assumed that the pollutant loadings from steam electric power plants and other
point sources are static loadings (i.e., a constant daily average loading rate). This approach does
not account for temporal variability in the loadings to receiving waters due to factors such as
variable plant operating schedules, storm flows, low-flow events, and catastrophic events. In
actuality, steam electric power plants and other point sources could adjust wastewater discharge
rates based on stream flow conditions or other considerations. For instance, a plant could reduce
discharges during periods of low flow in the receiving water and increase discharges during
periods of high flow, resulting in surface water concentrations that differ from what is predicted
by the case study model. These assumptions influence the relationship between modeled and
actual surface water concentrations at specific locations and times.
Appendix G further discusses the limitations and assumptions made in developing the
case study models and describes in more detail the development of each case study model,
including input parameters (e.g., pollutant loadings) and model settings. Refer to the Case Study
Water Quality Modeling Memorandum (DCN SE05570) for discussion of EPA's technical
approach and data acceptance criteria to incorporate DMR, TRI, and STORET monitoring data.
8.2 QUANTIFIED ENVIRONMENTAL IMPACTS AND IMPROVEMENTS FROM CASE STUDY
MODELING
As described in Section 8.1.1, EPA identified six representative case study locations that
would capture the types of impacts to surface waters associated with steam electric power plant
discharges, capture the improvements expected across the regulatory options, represent the four
wastestreams evaluated in the EA, and represent both lentic and lotic systems. Figure 8-1 and
Table 8-1 present the six receiving waters that EPA selected for case study modeling.
Section 8.2 introduces each of the six selected case study locations and presents the
scope, inputs, and modeling results. For each case study, EPA presents:
• Potential impacts to aquatic life, wildlife, and human health under baseline
conditions;
• Improvements to aquatic life, wildlife, and human health following compliance with
the final rule; and
8-14
-------
Section 8—Case Study Modeling
• Comparison of the case study and IRW model results for the case study location.
Although EPA modeled the expected environmental improvements under Options A
through D, this section primarily presents the water quality, wildlife, and human health
improvements under the final rule (Option D). Appendix G of this report includes figures
illustrating the water column concentrations output for the immediate receiving water both for
baseline conditions and following compliance with the final rule, for those modeled pollutants
that exceed one or more water quality benchmarks based on modeling results. These figures
present the National Recommended Water Quality Criteria (NRWQC) and Maximum
contaminant level (MCL) benchmarks for the modeled pollutant and the steady-state water
column concentration results from the IRW model. Appendix G also includes the average total
water column concentration for each of the modeled pollutants in WASP model segments
downstream of the modeled case study plants.
8.2.1 Black Creek Case Study
Black Creek flows south-southeast through southern Mississippi from Hattiesburg
through the De Soto National Forest until it converges with the Pascagoula River. Black Creek is
Mississippi's only designated National Wild and Scenic River (for 21 miles) under the National
Wild and Scenic Rivers System Act. South Mississippi Electric Power Association's R.D.
Morrow, Sr. (Morrow) Generating Site (Plant ID 1185) is a 400-megawatt (MW) coal-fired
power plant operating alongside Black Creek near Purvis, Mississippi. Morrow's two stand-
alone steam turbine generating units reported producing more than 2,000,000 megawatt-hours
(MWh) of electricity in 2009. Based on data obtained from the Steam Electric Survey, Morrow
Generating Site discharges FGD wastewater, bottom ash transport water, and combustion
residual leachate directly into Black Creek. Table 8-2 contains some general information on the
two steam electric generating units at Morrow Generating Site.
Table 8-2. Summary of Morrow Generating Site Operations
SE Unit
1
2
Fuel
Bituminous coal and
No. 2 fuel oil
Bituminous coal and
No. 2 fuel oil
Capacity
(MW)
200
200
Fly Ash
Dry conveyed
Dry conveyed
Bottom Ash
Wet handled to
impoundment
Wet handled to
impoundment
FGD
(Year Installed)
Wet system
(1978)
Wet system
(1978)
Source: ERG, 2015J.
Acronyms: FGD (Flue gas desulfurization); MW (Megawatt); SE (steam electric).
Modeling Area
The Black Creek WASP model encompasses a 95-mile reach of Black Creek, extending
from the Morrow Generating Site discharge outfall on Black Creek to the confluence of Black
Creek and Red Creek. The immediate receiving water that Morrow Generating Site discharges to
is approximately 1.6 miles long, as defined in the WASP model. This modeling area includes the
21-mile span of the waterway, from Moody's Landing to Fairley Bridge Landing, that is
8-15
-------
Section 8—Case Study Modeling
protected under the National Wild and Scenic River Systems Act. Figure 8-2 illustrates the
location and extent of the Black Creek WASP model.
Plant ID 1185
R.D. Morrow Sr.
LEGEND
Modeling Area
^^— Immediate Receiving Water
^^™ National Wild and Scenic River
^^— Other Modeled Reaches
Modeled Point Sources
• Case Study Steam Electric Power Plant
D Non-Steam-Electric Point Sources
Monitoring Data Stations
• Entering Modeling Area
o Within Modeling Area
024 8 12 16 20
Miles
Content may not reflect National Geographic'* current map policy Sources1 National GeographicTEsn.,
DeLorn-ie. HERE, UNEP-WQMC. USGS, NASA. ESA. METI. NRCAN...GEBCO, NOAA. increment P Co
Figure 8-2. Black Creek WASP Modeling Area
Identified Point Sources and Background Concentrations
As discussed below, EPA reviewed available pollutant loadings (DMR and TRI) and
monitoring data (STORET) for potential incorporation into the Black Creek WASP model to
represent pollutant contributions from background and non-steam-electric point sources, and for
use in calibrating the model results.
• Upstream pollutant contributions. EPA did not identify sufficient STORET
monitoring data to represent pollutant contributions from upstream of the Morrow
Generating Site immediate receiving water. EPA did not identify any upstream non-
steam-electric point sources with loadings for the eight modeled pollutants. EPA
therefore assumed pollutant concentrations of zero within the water column at the
upstream boundary of the modeling area.
• Downstream pollutant contributions. EPA incorporated STORET data from eight
monitoring stations to represent the pollutant contributions flowing into the modeling
8-16
-------
Section 8—Case Study Modeling
area downstream of the Morrow Generating Site immediate receiving water (i.e.,
tributaries flowing into Black Creek). EPA did not identify any non-steam-electric
point sources whose pollutant loadings would significantly influence the model
results in the downstream modeling area.
• Monitoring data within the modeling area. EPA compiled STORET data from two
monitoring stations located within the modeling area and used these data to calibrate
the WASP model.
Modeling Period
The modeling period starts in 1982 (the year of the last revision to the steam electric
ELGs) and extends through 2036, covering a period of 55 years. Based on Morrow Generating
Site's NPDES permitting cycle, EPA assumes that the plant will achieve the limitations under
the final rule by 2019.
Modeling Results - Water Quality
Under baseline conditions, the modeled pollutant concentrations in the immediate
receiving water and downstream reaches exceed the NRWQC water quality benchmarks for four
modeled pollutants, indicating that pollutant loadings from the Morrow Generating Site may
quantifiably reduce water quality in the modeled portions of Black Creek. The reduced water
quality is primarily attributed to arsenic, cadmium, selenium, and thallium. Intervals of higher
pollutant concentrations occur during periods of low flow in Black Creek for all eight modeled
pollutants.
The baseline modeled pollutant concentrations exceed human health criteria primarily for
arsenic, thallium, and selenium, as discussed below:
• Arsenic concentrations in the immediate receiving water exceed the water quality
benchmark for consumption of water and organisms (0.018 micrograms per liter
(|ig/L)) for 99 percent of the modeling period. These exceedances continue
downstream, generally at a reduced frequency, throughout the entire 95-mile-long
modeling area downstream of the plant.
• Arsenic concentrations in the immediate receiving water also exceed the higher water
quality benchmark for consumption of organisms only (0.14 |ig/L) for 16 percent of
the modeling period. These exceedances continue downstream, at a reduced
frequency, throughout the entire 95-mile-long modeling area downstream of the plant.
• Thallium concentrations in the immediate receiving water exceed the water quality
benchmark for consumption of water and organisms (0.24 |ig/L) for 17 percent of the
modeling period. These exceedances continue downstream, at a reduced frequency,
throughout the entire 95-mile-long modeling area downstream of the plant.
• Thallium concentrations in the immediate receiving water also exceed the higher
water quality benchmark for consumption of organisms only (0.47 |ig/L) for 1 percent
of the modeling period. These exceedances continue downstream throughout the
entire 95-mile-long modeling area downstream of the plant. The frequency of
8-17
-------
Section 8—Case Study Modeling
exceedances downstream ranges from less than 1 percent to 3 percent of the modeling
period.
• On rare occasions (less than 1 percent of the modeling period), selenium
concentrations in reaches downstream of the immediate receiving water exceed the
water quality benchmark for consumption of water and organisms (170 |ig/L). These
exceedances occur in 5.3 miles of the modeling area downstream of the plant and up
to 88 miles downstream of the plant.
These case study modeling results indicate that, under baseline conditions, humans
consuming water and/or organisms inhabiting these modeled portions of Black Creek could be at
an elevated risk of the negative effects associated with oral exposure to these pollutants (see
Section 3.1.1).
Aquatic organisms may be at risk for exposure to cadmium and selenium under baseline
conditions, as discussed below:
• Cadmium concentrations in the immediate receiving water exceed the freshwater
aquatic life criteria for chronic exposure (0.25 |ig/L) for 39 percent of the modeling
period. These exceedances continue downstream, at a reduced frequency, throughout
28 miles of the modeling area downstream of the plant.
• Selenium concentrations in the immediate receiving water exceed the freshwater
aquatic life criteria for chronic exposure (5.0 |ig/L) for 43 percent of the modeling
period. These exceedances continue downstream throughout the entire 95-mile-long
modeling area downstream of the plant. The frequency of exceedances downstream
ranges from 2 percent to 51 percent of the modeling period.
These case study modeling results indicate that, under baseline conditions, aquatic
organisms inhabiting these modeled portions of Black Creek could be at an elevated risk of the
negative effects associated with oral exposure to these pollutants (see Section 3.1.1).
Under baseline conditions, the modeled pollutant concentrations in the immediate
receiving water and downstream reaches occasionally exceed the MCL drinking water
benchmarks for three modeled pollutants. The baseline modeled pollutant concentrations exceed
drinking water criteria for cadmium, selenium, and thallium, as discussed below:
• On rare occasions (less than 1 percent of the modeling period), cadmium
concentrations in the immediate receiving water exceed the MCL benchmark (5
|ig/L). These exceedances continue downstream throughout the entire 95- mile-long
modeling area downstream of the plant. The frequency of exceedances downstream
ranges from less than 1 percent to 5 percent of the modeling period.
• On rare occasions (less than 1 percent of the modeling period), selenium
concentrations in the immediate receiving water exceed the MCL benchmark (50
|ig/L). These exceedances continue downstream, generally at a reduced frequency, in
93 miles of the modeling area downstream of the plant.
• On rare occasions (less than 1 percent of the modeling period), thallium
concentrations in downstream reaches of the modeling area exceed the MCL (2
-------
Section 8—Case Study Modeling
Hg/L). These exceedances occur in 8.9 miles of the modeling area downstream of the
plant and up to 92 miles downstream of the plant.
Modeling results do not indicate any exceedances of NRWQC or MCL criteria for the
other modeled pollutants (copper, nickel, lead, and zinc). Appendix G of this report includes
figures that illustrate the water column pollutant concentration output for the immediate
receiving water for arsenic, cadmium, selenium, and thallium. These figures also present the
NRWQC and MCL benchmarks for the pollutant and the steady-state water column pollutant
concentrations predicted by the IRW model.
The final rule modeling results show significantly decreased concentrations of all
modeled pollutants in the immediate receiving water, which will greatly improve water quality.
These pollutant removals result in fewer exceedances of NRWQC and MCL benchmarks
compared to those estimated in the baseline modeling. Case study modeling results for Black
Creek reveal the following water quality improvements under the final rule:
• For arsenic:
- Exceedances of the human health water quality benchmark for consumption of
water and organisms reduce in frequency from 99 percent to 94 percent of the
modeling period in the immediate receiving water. Additionally, the exceedances
of this benchmark reduce in frequency in all remaining sections of the
downstream modeling area following compliance with the final rule. Despite the
continued exceedances of this human health criteria, reducing the pollutant
concentrations in the water column may decrease the risk to humans consuming
contaminated water and organisms.
Exceedances of the human health water quality benchmark for consumption of
organisms reduce in frequency from 16 percent to 6 percent of the modeling
period in the immediate receiving water. Additionally, the exceedances of this
benchmark reduce in frequency in all remaining sections of the downstream
modeling area following compliance with the final rule. Despite the continued
exceedances of this human health criteria, reducing the pollutant concentrations in
the water column may decrease the risk to humans consuming contaminated
organisms.
• For cadmium:
- Exceedances of the aquatic life water quality criteria for chronic impacts are
eliminated throughout the entire modeling area.
Exceedances of the MCL benchmark are eliminated throughout the entire
modeling area.
• For selenium:
- Exceedances of the human health water quality benchmark for consumption of
water and organisms are eliminated throughout the entire modeling area.
Exceedances of the MCL benchmark are eliminated throughout the entire
modeling area.
8-19
-------
Section 8—Case Study Modeling
- Exceedances of the aquatic life water quality criteria for chronic impacts are
eliminated in 13 miles of the modeling area, including the immediate receiving
water. The exceedances of this benchmark reduce in frequency to less than 4
percent in all remaining sections of the downstream modeling area following
compliance with the final rule. Most of these exceedances occur within the first
year following compliance with the final rule (due to the gradual recovery of the
system following the pollutant loading removals). Despite the continued
exceedances of these human health criteria, reducing the pollutant concentrations
in the water column may decrease risk to humans consuming contaminated water
and/or organisms.
• For thallium:
Exceedances of the MCL benchmark are eliminated throughout the entire
modeling area.
- Exceedances of the human health water quality benchmark for consumption of
water and organisms reduce in frequency from 17 percent to less than 1 percent of
the modeling period in the immediate receiving water. Additionally, the
exceedances of this benchmark reduce in frequency in all remaining sections of
the downstream modeling area following compliance with the final rule. Despite
the continued exceedances of this human health criteria, reducing the pollutant
concentrations in the water column may decrease the risk to humans consuming
contaminated water and organisms.
Exceedances of the human health water quality benchmark for consumption of
organisms are eliminated in 6.2 miles of the modeling area, including the
immediate receiving water. Additionally, the exceedances of these benchmarks
reduce in frequency in all remaining sections of the downstream modeling area
following compliance with the final rule. Despite the continued exceedances of
this human health criteria, reducing the pollutant concentrations in the water
column may decrease risk to humans consuming contaminated organisms.
Modeling Results - Wildlife
EPA assessed the potential threat to piscivorous wildlife from the evaluated wastestreams
by modeling the average pollutant concentrations in the water column and comparing these to the
concentrations that would trigger exceedances of no effect hazard concentrations (NEHC) for
minks and eagles developed by the USGS. Under baseline conditions, Black Creek may pose a
risk to minks and eagles that consume fish contaminated with selenium. The average modeled
selenium concentrations in 90 miles of the Black Creek modeling area are greater than the
concentration that would translate to NEHC exceedances for minks and eagles, demonstrating
that the fish inhabiting these portions of Black Creek may pose a potential reproductive threat to
terrestrial food webs.
EPA also assessed the potential impact to wildlife exposed to sediments in surface waters
by comparing estimated pollutant concentrations in the sediment to sediment biota CSCL
benchmarks. Modeling results demonstrate that cadmium concentrations in the upper benthic
sediment of the immediate receiving water exceed the CSCL criteria (0.596 mg/kg) during 36
8-20
-------
Section 8—Case Study Modeling
percent of the modeling period. These exceedances continue downstream for 36 miles of the total
modeling area.
Ecological risk modeling results indicate that baseline selenium loadings also present an
elevated risk of widespread negative reproductive impacts (larval mortality and deformities)
among fish that inhabit the immediate receiving water of Black Creek. The results illustrate the
significant increase in risk that can result from minor variations in selenium bioaccumulation
patterns and toxicity responses within the organisms that inhabit a particular waterbody.
Specifically:
• The median (50th percentile) of the model outputs indicates that selenium
concentrations in the fish eggs and ovaries would cause reproductive impacts in less
than 1 percent of the exposed fish population.
• However, there is a 35 percent probability that these concentrations are high enough
to cause reproductive impacts in more than 30 percent of the exposed fish population.
• There is a 25 percent probability that these concentrations are high enough to cause
reproductive impacts in more than 80 percent of the exposed fish population.
Ecological risk modeling results also indicate an elevated risk of widespread negative
reproductive impacts (hatching failure) among mallards that forage or breed in the immediate
receiving water of Black Creek. Specifically:
• There is a 50 percent probability that selenium concentrations in the mallard eggs are
high enough to cause reproductive impacts in at least 9 percent of the exposed
mallard population.
• There is a 35 percent probability that these concentrations are high enough to cause
reproductive impacts in more than 20 percent of the exposed mallard population.
• There is a 10 percent probability that these concentrations are high enough to cause
reproductive impacts in more than 70 percent of the exposed mallard population.
Elevated risks of reproductive impacts to fish and mallards continue downstream from
the immediate receiving water. Ecological risk modeling results indicate that the entire 95-mile
modeled length of Black Creek has selenium concentrations that lead to a 10 percent or greater
probability of negative reproductive impacts among at least 17 percent of the exposed fish or
mallard populations. Additionally, several downstream segments of Black Creek (totaling 29
miles) have selenium concentrations that lead to a 25 percent or greater probability of negative
reproductive impacts among at least 10 percent of the exposed mallard population.
The case study modeling results demonstrate that the final rule will significantly reduce
pollutant concentrations and the associated impacts to wildlife that inhabit Black Creek. The
final rule will eliminate selenium exceedances of the NEHC benchmarks for minks and eagles in
all modeled reaches of Black Creek. The final rule will also eliminate CSCL benchmark
exceedances for cadmium in 27 miles of the modeling area, including the immediate receiving
water. The exceedances of this benchmark reduce in frequency to 3 percent or less in all
remaining sections of the downstream modeling area following compliance with the final rule.
Most of these remaining exceedances occur within the first year following compliance with the
-------
Section 8—Case Study Modeling
final rule. Ecological risk modeling results also indicate that the final rule will eliminate the risk
of selenium-related adverse reproductive impacts among exposed fish and mallards in all
modeled reaches of Black Creek (i.e., the risk to fish and mallards is less than 0.1 percent at the
95th percentile egg/ovary concentration).
Modeling Results - Human Health
EPA evaluated the potential threat to human receptors due to consumption of
contaminated fish from Black Creek. EPA modeled the average pollutant concentrations in the
water column and compared these to the concentrations that would trigger exceedances of either
the non-cancer reference dose or the 1-in-a-million lifetime excess cancer risk (LECR). Under
baseline conditions, the average water column concentration of arsenic throughout the modeling
area downstream of the plant does not result in an estimated cancer risk greater than 1-in-a-
million for any of the national-scale cohorts. See Appendix E for details on the human health
module of the IRW model and national-scale cohorts.
Based on the average pollutant concentrations in the water column under baseline
conditions, cadmium, selenium, and thallium pose the greatest threat to cause non-cancer health
effects in humans from fish consumption, as discussed below:
• Average thallium concentrations in the water column throughout the entire 95-mile-
long modeling area are greater than the concentration that would translate to
exceedance of the reference doses for at least one child subsistence fisher cohort
(with all child subsistence cohorts impacted by 59 or more miles of the modeling area
downstream of the plant), while the concentrations in 90 miles of the modeling area
are high enough to trigger exceedance of the reference dose for adult subsistence
fishers. Additionally, the average thallium concentrations in 59 miles of the modeling
area are high enough to trigger exceedance of the reference dose for at least one child
recreational fisher cohort.
• Average selenium concentrations in the water column throughout the entire 95-mile-
long modeling area are greater than the concentration that would translate to
exceedance of the reference dose for the adult subsistence fisher cohorts and at least
one child subsistence fisher cohort (with all child subsistence cohorts impacted by 90
or more miles). Additionally, the average selenium concentrations are high enough to
trigger exceedances of the reference doses for adult recreational fishers and at least
one child recreational fisher cohort in 13 miles and 90 miles of the modeling area,
respectively.
• Average cadmium concentrations in the water column in 38 miles of the modeling
area are greater than the concentration that would translate to exceedance of the
reference dose for at least one child subsistence fisher cohort.
Therefore, humans who consume cadmium-, selenium-, or thallium-contaminated fish
inhabiting these waters may be at greater risk for developing the negative health effects
associated with these pollutants, which are discussed in Section 3.1.1.
8-22
-------
Section 8—Case Study Modeling
The modeling results demonstrate significant reductions in average water column
concentrations of cadmium, selenium, and thallium under the final rule, which would reduce
average cadmium and selenium concentrations enough to eliminate the risk for non-cancer health
effects for all cohorts throughout the entire modeling area. These loadings reductions would also
reduce the thallium concentrations enough to eliminate the risk for non-cancer health effects for
adult subsistence and child recreational fishers. While the case study model continues to show
average thallium concentrations that may pose non-cancer health effects for at least one child
subsistence cohort, the total area of impact is reduced by up to 37 miles (with some child
subsistence cohort non-cancer risks being eliminated throughout the entire modeling period
downstream of the plant).
Interpretation of Black Creek Results
Case study modeling results for Black Creek indicate greater water quality, wildlife, and
human health impacts to the immediate receiving water under baseline conditions than predicted
by the IRW model. Case study modeling results for Black Creek also demonstrate water quality
benchmark exceedances and risks to wildlife and humans sustaining beyond Morrow Generating
Site's immediate receiving water. In some instances, the average water column concentrations
can increase in some portions of the downstream modeling area, posing a greater threat to
humans, aquatic organisms, and terrestrial ecosystems. This phenomenon is most pronounced for
modeled pollutants with the largest partition coefficients (i.e., lead, zinc, cadmium, and copper)
suggesting that sediment transport has significant influence in this small receiving water. Under
baseline conditions, significant water quality, wildlife, and human health impacts are identified
in the modeled area corresponding with 21-mile span of the waterway that is protected under the
National Wild and Scenic River Systems Act.
Ecological risk modeling results for the Black Creek case study indicate that the risk of
negative reproductive effects among fish and mallards exposed to selenium may be significantly
greater than predicted using water quality outputs from the IRW model. Use of the case study
water quality outputs, which include extended periods of elevated selenium concentrations that
are not reflected in the IRW model outputs, reveals the potential for widespread ecological
impacts among wildlife that inhabit, forage, or breed in the immediate receiving water of Black
Creek and its downstream waters.
The USGS stream gage flow data used in the case study model indicate that flow rates in
Black Creek are typically lower than the annual average flow rate used in the IRW model, while
greatly exceeding the annual average flow rate during occasional high-flow events. During the
frequent periods of below-average flow, the pollutant concentrations in the modeling area
quickly climb to levels associated with negative impacts to fish, wildlife, and humans.
The exceedances identified in the Black Creek WASP model are based solely on
discharges of the evaluated wastestreams from the steam electric power plant because EPA did
not identify any STORET monitoring data or point sources suggesting any other sources were
contributing pollutant discharges to the modeling area. The Black Creek WASP model may be
underestimating the pollutant concentrations actually present if there are other discharges that
were not captured in the DMR and TRI data sets. Under the final rule, case study modeling of
Black Creek indicates that the waterbody will exhibit fewer exceedances of water quality
8-23
-------
Section 8—Case Study Modeling
benchmarks; will no longer pose reproductive risks to higher trophic-level wildlife; will pose less
risk to benthic organisms; and will pose less risk to humans consuming fish. The extent of
improvements identified by the case study model is greater than what was projected by the IRW
model. The decrease of the average pollutant concentrations within the immediate receiving
water occurs very quickly after compliance with the final rule; however, some downstream
reaches of the modeling area take up to a year to reach equilibrium.
8.2.2 Etowah River Case Study
The Etowah River is a 164-mile-long waterway north of Atlanta, Georgia. The river
flows west-southwest from Amicalola Creek, the primary tributary, to Rome, Georgia, where it
meets the Oostanaula River and forms the Coosa River at their confluence. Once estimated to
have 91 native fish species, the Etowah watershed is biologically one of the richest river systems
in North America. Eight imperiled fish species,
three of which are federally listed as endangered
or threatened, are known to inhabit the Etowah
watershed, and five mollusk species are believed
to have been decimated [Etowah Aquatic
Habitat Conservation Plan, 2015].
The Etowah River serves as a source of
cooling water for, and receives steam electric
wastewater discharges from, Southern
Company's Plant Bowen (Plant ID 2244),
located in Cartersville, Georgia. In commercial
operation since 1975, Plant Bowen is bordered
on two sides by the Etowah River and Euharlee
Creek. Plant Bowen's four stand-alone steam
turbine generating units have a total nameplate
•^ £• i /inn A/m7 A ^i *i t • ^ Georgia Power Company's Plant Bowen
capacity of 3,499 MW. As the nation s ninth- F *
largest power plant in net generation of electricity, Plant Bowen reported producing almost
23,000,000 MWh of electricity in 2009 [Georgia Power, 2014]. Based on data EPA obtained in
responses to the Steam Electric Survey, Plant Bowen discharges two of the evaluated
wastestreams, FGD wastewater and bottom ash transport water, directly to the Etowah River.
Table 8-3 contains general information on the four steam electric generating units at Plant
Bowen.
In estimating the historical pollutant loadings associated with Plant Bowen's four FGD
systems, EPA incorporated the pollutant loadings from FGD wastewater as the systems were
installed, between 2008 and 2011. EPA did not model any FGD wastewater pollutant loadings
before the installation of Plant Bowen's first FGD system.
8-24
-------
Section 8—Case Study Modeling
Table 8-3. Summary of Plant Bowen Operations
SE Unit
1
2
o
6
4
Fuel
Bituminous coal and
No. 2 fuel oil
Bituminous coal and
No. 2 fuel oil
Bituminous coal and
No. 2 fuel oil
Bituminous coal and
No. 2 fuel oil
Capacity (MW)
806
789
952
952
Fly Ash
Dry conveyed
Dry conveyed
Dry conveyed
Dry conveyed
Bottom Ash
Wet handled to
impoundment
Wet handled to
impoundment
Wet handled to
impoundment
Wet handled to
impoundment
FGD
(Year Installed)
Wet system
(2010)
Wet system
(2009)
Wet system
(2008)
Wet system
(2008)
Source: ERG, 2015J.
Acronyms: FGD (Flue gas desulfurization); MW (Megawatt); SE (steam electric).
Modeling Area
The Etowah River WASP model encompasses a 35-mile segment of the Etowah River,
extending from the immediate receiving water to the confluence of the Etowah River and Silver
Creek. The immediate receiving water to which Plant Bowen discharges is approximately 3.6
miles long, as defined in the WASP model. Figure 8-3 illustrates the location and extent of the
Etowah River WASP model.
8-25
-------
Section 8—Case Study Modeling
Adairsville"
Shannon
•411
LEGEND
Modeling Area
^^— Immediate Receiving Water
^^— Other Modeled Reaches
Modeled Point Sources
I Case Study Steam Electric Power Plant
D Non-Steam-Electric Point Sources
Monitoring Data Stations
• Entering Modeling Area
o Within Modeling Area
01
5 Miles
O
O
Emerson*
Ail i
Plant ID 2244
Plant Bowen
Aragon
Content niay not reflect National Geograpnic's GUrnSK! -ra;: &•: K, Sauries National G
"'•peLorme HERE, UNEP-WCMC .USGS, NASA, ESA METI NRCAN GEBCO NOAA,
•ogrgphic Esr
increment P Corp
Figure 8-3. Etowah River WASP Modeling Area
Identified Point Sources and Background Concentrations
As discussed below, EPA reviewed available pollutant loadings (DMR and TRI) and
monitoring data (STORET) for potential incorporation into the Etowah River WASP model to
represent pollutant contributions from background and non-steam-electric point sources, and for
use in calibrating the model results.
• Upstream pollutant contributions. EPA incorporated STORET data from four
monitoring stations to represent the pollutant contributions from upstream of the Plant
Bowen immediate receiving water. EPA also identified two upstream non-steam-
electric point sources whose pollutant loadings (from DMR and TRI data sets) could
influence the model results; however, EPA assumed that the STORET data from the
four monitoring stations (which encompass all of the modeled pollutants except for
selenium) adequately reflect the pollutant contributions from upstream point sources.
Therefore, EPA did not incorporate pollutant loadings from the two identified
upstream non-steam-electric point sources.
8-26
-------
Section 8—Case Study Modeling
• Downstream pollutant contributions EPA incorporated STORET data from 10
monitoring stations to represent the pollutant concentrations flowing into the
modeling area downstream of the Plant Bowen immediate receiving water (i.e.,
tributaries flowing into the Etowah River). EPA did not identify any non-steam-
electric point sources whose pollutant loadings would significantly influence the
model results in the downstream modeling area.
• Monitoring data within the modeling area. EPA compiled STORET data from six
monitoring stations located within the modeling area and used these data to calibrate
the WASP model.
The contributions of arsenic, cadmium, copper, lead, and thallium from upstream sources
have a much greater influence on the modeled pollutant concentrations in the Etowah River than
the pollutant loadings from Plant Bowen. The contributions of nickel and zinc from upstream
sources also strongly influence the modeled pollutant concentrations in the Etowah River.
The Etowah River case study model did not account for the documented surface water
impacts from Plant Bowen that are discussed in Section 3.3.3. In 2002, a sinkhole developed in
the surface impoundment at Plant Bowen that released 2.25 million gallons of ash/water mixture,
estimated to contain 80 tons of ash, to Euharlee Creek, which immediately flows into the Etowah
River [U.S. EPA, 2014b]. Additionally, an extreme rainfall event in 2008 caused a dry ash
stockpile to collapse, depositing approximately two tons of ash in Euharlee Creek. The surface
water quality impacts resulting from these events are not reflected in this model; therefore, the
case study modeling could under-represent the actual baseline impacts of Plant Bowen on the
Etowah River.
Modeling Period
The modeling period starts in 1982 (the year of the last revision to the steam electric
ELGs) and extends through 2032, covering a period of 51 years. Based on Plant Bowen's
NPDES permitting cycle, EPA assumes that the plant will achieve the limitations under the final
rule by 2021.
Modeling Results - Water Quality
Under baseline conditions, the modeled pollutant concentrations in the immediate
receiving water and downstream reaches exceed the NRWQC water quality benchmarks for five
modeled pollutants, indicating that pollutant loadings from Plant Bowen may contribute to a
quantifiable reduction in water quality in the modeled portions of the Etowah River. The reduced
water quality is primarily attributed to arsenic, cadmium, selenium, thallium, and lead.
The baseline modeled water concentrations exceed human health criteria primarily for
arsenic and thallium, as discussed below:
• Arsenic concentrations in the immediate receiving water exceed the water quality
benchmark for consumption of water and organisms (0.018 |ig/L) for the entire
modeling period. These exceedances continue downstream, at the same frequency,
throughout the entire 35-mile-long modeling area downstream of the plant.
8-27
-------
Section 8—Case Study Modeling
• Arsenic concentrations in the immediate receiving water also exceed the higher water
quality benchmark for consumption of organisms only (0.14 |ig/L) for the entire
modeling period. These exceedances continue downstream, at the same frequency,
throughout the entire 35-mile-long modeling area downstream of the plant.
• Thallium concentrations in the immediate receiving water exceed the water quality
benchmarks for consumption of water and organisms (0.24 |ig/L) for more than 99
percent of the modeling period. These exceedances continue downstream, at an
increased frequency, throughout the entire 35-mile-long modeling area downstream
of the plant.
• Thallium concentrations in the immediate receiving water also exceed the higher
water quality benchmark for consumption of organisms only (0.47 |ig/L) for 90
percent of the modeling period. These exceedances continue downstream, at an
increased frequency, throughout the entire 35-mile-long modeling area downstream
of the plant.
These case study modeling results indicate that, under baseline conditions, humans
consuming water and/or organisms inhabiting these modeled portions of the Etowah River may
be more at risk of the negative effects associated with oral exposure to arsenic and thallium (see
Section 3.1.1).
Aquatic organisms may be at risk for exposure to cadmium and selenium under baseline
conditions, specifically:
• Cadmium concentrations in the immediate receiving water exceed the freshwater
aquatic life criteria for chronic exposure (0.25 |ig/L) for 52 percent of the modeling
period. These exceedances continue downstream throughout the 35-mile-long
modeling area downstream of the plant. The frequency of exceedances downstream
ranges from 33 percent to 55 percent of the modeling period.
• On rare occasions (less than 1 percent of the modeling period), selenium
concentrations in downstream reaches of the modeling area exceed the freshwater
aquatic life criteria for chronic exposure (5 |ig/L). These exceedances occur in 4.7
miles of the downstream modeling area downstream of the plant and up to 35 miles
downstream of the plant.
These modeling results indicate that, under baseline conditions, aquatic organisms
residing in the portions of the Etowah River with modeled exceedances may be more at risk to
negative impacts from chronic exposure to cadmium and selenium.
Under baseline conditions, the modeled pollutant concentrations in the immediate
receiving water and downstream reaches exceed the MCL drinking water benchmarks for four
modeled pollutants. The baseline modeled pollutant concentrations exceed drinking water
criteria for thallium, arsenic, cadmium and lead as discussed below:
• Thallium concentrations in the immediate receiving water exceed the MCL
benchmark (2 |ig/L) for 29 percent of the modeling period. These exceedances
8-28
-------
Section 8—Case Study Modeling
continue downstream, at a reduced frequency, throughout the entire 35-mile-long
modeling area downstream of the plant.
• On rare occasions (less than 1 percent of the modeling period), arsenic concentrations
in the immediate receiving water exceed the MCL benchmark (10 |ig/L). These
exceedances do not occur beyond the 3.6-mile-long immediate receiving water.
• On rare occasions (less than 1 percent of the modeling period), cadmium
concentrations in downstream reaches of the modeling area exceed the MCL
benchmark (5 |ig/L). These exceedances occur in 5.1 miles of the downstream
modeling area downstream of the plant and up to 35 miles downstream of the plant.
• On rare occasions (less than 1 percent of the modeling period), lead concentrations in
downstream reaches of the modeling area exceed the MCL benchmark (15 |ig/L).
These exceedances occur in 5.1 miles of the downstream modeling area downstream
of the plant and up to 35 miles downstream of the plant.
Modeling results do not indicate any exceedances of NRWQC or MCL criteria for the
other modeled pollutants (copper, nickel, and zinc). Appendix G of this report includes figures
that illustrate the water column pollutant concentration output for the immediate receiving water
for arsenic, cadmium, selenium, and thallium. These figures also present the NRWQC and MCL
benchmarks for the pollutant and the steady-state water column pollutant concentrations
predicted by the IRW model.
The final rule modeling results show a significant reduction in selenium concentrations
and moderately decreased concentrations of cadmium, nickel, and zinc within the Etowah River,
which will improve water quality. These pollutant removals result in fewer exceedances of
NRWQC and MCL benchmarks compared to those estimated in the baseline modeling. Case
study modeling results for the Etowah River reveal the following water quality improvements
under the final rule:
• Exceedances of the cadmium aquatic life water quality criteria for chronic impacts
reduce in frequency (by 13 percent) in the immediate receiving water. Additionally,
the exceedances of these benchmarks reduce in frequency in all remaining sections of
the downstream modeling area following compliance with the final rule. Despite
continued exceedances of these aquatic life criteria, reducing the pollutant
concentrations in the water column may decrease the risk to aquatic life in the Etowah
River.
• Exceedances of the selenium aquatic life water quality criteria for chronic impacts are
eliminated throughout the entire modeling area.
While case study modeling results continue to show exceedances for NRWQC
benchmark exceedances of arsenic and thallium and MCL benchmark exceedances of arsenic,
cadmium, lead, and thallium, the final rule will reduce loading contributions of these pollutants
from Plant Bowen.
8-29
-------
Section 8—Case Study Modeling
Modeling Results - Wildlife
Based on the average pollutant concentrations in the water column under baseline
conditions, the modeled portion of the Etowah River does not exceed the concentrations that
would translate to NEHC exceedances and does not pose a risk to minks and eagles that consume
contaminated fish. Despite the modeling not being able to quantify any improvements to minks
and eagles under the final rule, the pollutant loading removals will decrease bioaccumulation of
toxic pollutants in the terrestrial food chains.
Modeling results do not indicate that there are any pollutant concentrations in the upper
benthic sediment that exceed CSCL benchmarks of for any of the eight modeled pollutants;
therefore, the Etowah River does not pose a threat to benthic organisms in contact with
contaminated sediment. Despite the modeling not being able to quantify any improvements to
benthic organisms under the final rule, the pollutant loading removals will decrease the
concentrations of toxic pollutants in benthic sediment and decrease the exposure of organisms to
these pollutants.
Modeling Results - Human Health
EPA modeled the average pollutant concentrations in the water column and compared
these to the concentrations that would trigger exceedances of either the non-cancer reference
dose or the 1-in-a-million lifetime excess cancer risk (LECR). Under baseline conditions, the
average water column concentration of arsenic in the immediate receiving water over the
modeling period results in an estimated cancer risk greater than 1-in-a-million for adult
subsistence fishers. These exceedances do not occur beyond the 3.6-mile-long immediate
receiving water. Therefore, adults who frequently consume arsenic-contaminated fish inhabiting
the immediate receiving water may be at greater risks for development of cancer. Modeling
results demonstrate no reduction in the cancer risk from inorganic arsenic under the final rule.
Based on the average pollutant concentrations in the water column under baseline
conditions, selenium and thallium pose the greatest threat to cause non-cancer health effects in
humans from fish consumption, as discussed below:
• Average selenium concentrations in the immediate receiving water are greater than
the concentrations that would translate to exceedance of the reference doses for the
child (younger than 11 years old) subsistence fisher cohorts. The average selenium
concentrations throughout the entire 35-mile-long modeling area downstream of the
plant are greater than the concentration that would translate to an exceedance of the
reference dose for least one child subsistence cohort.
• Average thallium concentrations in the water column throughout the entire 35-mile-
long modeling area downstream of the plant are greater than the concentrations that
would translate to exceedance of the reference doses for adult and children
recreational and subsistence fishers (all national-scale cohorts evaluated).
Therefore, humans who consume selenium- or thallium-contaminated fish inhabiting the
modeled area of the Etowah River may be at greater risk for developing the negative health
effects associated with these pollutants, which are discussed in Section 3.1.1.
8-30
-------
Section 8—Case Study Modeling
The final rule modeling results demonstrate significant reductions in selenium
concentrations in the Etowah River, which will eliminate selenium exceedances of the non-
cancer health effects reference dose for all cohorts. While the modeling results continue to show
thallium water concentrations that would translate to exceedances of the non-cancer health
effects reference dose, the final rule will reduce thallium loading contributions from Plant
Bowen.
Interpretation of Etowah River Results
Case study modeling results for the Etowah River indicate greater water quality and
human health impacts than predicted by the IRW model (IRW modeling results did not indicate
any quantifiable impacts in the immediate receiving water of Plant Bowen). By accounting for
background pollutant contributions from upstream sources and other boundaries (for all modeled
pollutants except selenium), case study modeling predicts higher pollutant concentrations under
baseline conditions. For arsenic and thallium, and to a lesser extent cadmium, the projected
exceedances are driven by the background concentrations flowing into the Etowah River
modeling area. Plant Bowen's discharges of the evaluated wastestreams may be further
impairing the degraded waterway.
Case study modeling results for the Etowah River also demonstrate water quality
benchmark exceedances and risks to humans occur beyond Plant Bowen's immediate receiving
water. In some instances, the average water column concentrations can increase in some portions
of the downstream modeling area, posing a greater threat to humans and aquatic life. This
phenomenon is most pronounced for modeled pollutants with the largest partition coefficients
(i.e., lead, zinc, cadmium, and copper), suggesting that sediment transport has moderate
influence in the Etowah River.
Case study modeling of the Etowah River indicates that, under the final rule, the Etowah
River will exhibit fewer exceedances of water quality benchmarks and pose less risk to humans
consuming fish that inhabit these waters. The improvements identified by the case study model
are more extensive than what was projected by the IRW model. This is due in part to the greater
water quality and human health impacts under baseline conditions, which created additional
opportunities for modeled improvements, and in part to the identified improvements in
downstream reaches of the Etowah River that were not evaluated as part of the IRW model. The
average pollutant concentrations throughout the entire modeling area reduce promptly after
compliance with the final rule.
8.2.3 Lick Creek & White River Case Study
The White River is a two-forked river that primarily flows southwest through central and
southern Indiana. The two forks, the West Fork and the East Fork, are nearly equal in size when
they converge in Daviess Country, just north of Petersburg, Indiana. From this confluence, the
White River flows west-southwest for 50 river-miles until it joins the Wabash River at the
Illinois-Indiana state border. Located on the banks of the lower White River, Indianapolis Power
& Light's (IPL) Petersburg Generating Station (Plant ID 3997) has four stand-alone steam
turbine units with a nameplate capacity of 1,864 MW. The plant reported that these four coal-
fired generating units produced more than 12,000,000 MWh of electricity in 2009 in the Steam
8-31
-------
Section 8—Case Study Modeling
Electricity Survey. Petersburg Generating Station also operates three minor oil-burning internal
combustion units, which are exempt from the requirements of the final rule. Based on data
obtained in responses to the Steam Electric Survey, this power plant discharges FGD wastewater
and bottom ash transport water. Table 8-4
contains general information on the four coal-
fired generating units at Petersburg Generating
Station.
In estimating the historical pollutant
loadings associated with Petersburg Generating
Station's four FGD systems, EPA incorporated
the pollutant loadings from FGD wastewater as
the systems were installed, between 1977 and
1996. EPA included the pollutant loadings from
the FGD systems on units 3 and 4 at the start of
the historical modeling period (1986).
IPL 's Petersburg Generating Station
Table 8-4. Summary of Petersburg Generating Station Operations
SE Unit
1
2
3
4
Fuel
Subbituminous coal
and No. 2 fuel oil
Subbituminous coal
and No. 2 fuel oil
Subbituminous coal
and No. 2 fuel oil
Subbituminous coal
and No. 2 fuel oil
Capacity
(MW)
255
445
580
584
Fly Ash a
Dry conveyed
Dry conveyed
Dry conveyed
Dry conveyed
Bottom Ash
Wet handled to
impoundment
Wet handled to
impoundment
Wet handled to
impoundment
Wet handled to
impoundment
FGD
(Year Installed)
Wet system
(05/1996)
Wet system
(05/1996)
Wet system
(11/1977)
Wet system
(04/1986)
Source: ERG, 2015J.
Acronyms: FGD (Flue gas desulfurization); MW (Megawatt); SE (steam electric).
a - Based on EPA projections, Petersburg Generating Station will convert to dry ash handling to comply with the
CCR rulemaking.
Modeling Area
Based on data obtained in responses to the Steam Electric Survey, Petersburg Generating
Station discharges FGD wastewater and bottom ash transport water to Lick Creek, a 1.8-mile-
long tributary emptying into the White River. The White River WASP model encompasses Lick
Creek and a 52-mile reach of the White River, 49 miles of which is downstream of Lick Creek.
The immediate receiving water, Lick Creek, is the first of three upstream modeling boundaries
for this WASP model. The other upstream model boundaries are on the West Fork White River
and East Fork White River approximately one mile upstream of their confluence. EPA extended
the modeling area upstream of Lick Creek to capture and incorporate available STORET
monitoring data as further described below. The Lick Creek and White River WASP model ends
8-32
-------
Section 8—Case Study Modeling
at the confluence of the White River with the Wabash River. Figure 8-4 illustrates the location
and extent of the White River WASP model.
Plant ID 3997
Petersburg Generating
Station
Content may not reflect National G
DeLorme HERE UNEP-WGMi
LEGEND
Modeling Area
^^™ Immediate Receiving Water
^^— Other Modeled Reaches
Modeled Point Sources
• Case Study Steam Electric Power Plant
D Non-Steam-Electric Point Sources
Monitoring Data Stations
• Entering Modeling Area
o Within Modeling Area
012 4 6 8 10 12
Miles
Figure 8-4. Lick Creek and White River WASP Modeling Area
Identified Point Sources and Background Concentrations
As discussed below, EPA reviewed available pollutant loadings (DMR and TRI) and
monitoring data (STORET) for potential incorporation into the Lick Creek and White River
WASP model to represent pollutant contributions from background and non-steam-electric point
sources, and for use in calibrating the model results.
• Upstream pollutant contributions (Lick Creek). EPA did not identify sufficient
STORET monitoring data to represent pollutant contributions from upstream of the
Petersburg Generating Station immediate receiving water (Lick Creek). EPA did not
identify any upstream non-steam-electric point sources with loadings for the eight
modeled pollutants on Lick Creek. EPA therefore assumed pollutant concentrations
of zero within the water column at the upstream boundary of the modeling area.
• Upstream pollutant contributions (West Fork White River). EPA incorporated
STORET data from three monitoring stations to represent the pollutant contributions
8-33
-------
Section 8—Case Study Modeling
from upstream on the west fork of the White River. EPA also identified three
upstream non-steam-electric point sources whose pollutant loadings (from DMR and
TRI data sets) could influence the model results; however, EPA assumed that the
STORET monitoring data (which include all of the modeled pollutants except for
thallium) adequately reflect the pollutant contributions from upstream point sources.
Similarly, EPA identified that a steam electric power plant, Edwardsport Generating
Station (Plant ID 8544), has historically discharged to the west fork of the White
River 30 miles upstream of the start boundary. Edwardsport Generating Station
discontinued operation of all steam electric generating units in 2011 to construct a
new integrated gasification combined cycle power plant. EPA assumed that the
STORET monitoring data adequately reflect the pollutant contributions from this
point source. Therefore, EPA did not incorporate pollutant loadings from the three
identified upstream non-steam-electric point sources or Edwardsport Generating
Station into the WASP model.
• Upstream pollutant contributions (East Fork White River) EPA incorporated
STORET data from one monitoring station to represent the pollutant contributions
from upstream on the east fork of the White River. EPA also identified one upstream
non-steam-electric point source whose pollutant loadings (from DMR and TRI data
sets) could influence the model results; however, EPA assumed that the STORET
monitoring data (which include all of the modeled pollutants) adequately reflect the
pollutant contributions from upstream point sources. Therefore, EPA did not
incorporate pollutant loadings from this identified upstream non-steam-electric point
source in the WASP model.
• Downstream pollutant contributions. EPA incorporated STORET data from four
monitoring stations to represent the pollutant concentrations flowing into the
modeling area downstream of the Petersburg Generating Station immediate receiving
water, Lick Creek (i.e., tributaries flowing into the White River). EPA did identify
one non-steam-electric point source that discharges one or more of the modeled
pollutants within the modeling area. EPA incorporated the pollutant loadings from the
identified non-steam-electric point source into the model.
• Monitoring data within the modeling area. EPA compiled STORET data from 12
monitoring stations located within the modeling area and used these data to calibrate
the WASP model.
The contributions of arsenic, cadmium, copper, nickel, lead, and zinc from upstream
sources have a much greater influence on the modeled pollutant concentrations in White River
than the pollutant loadings from Petersburg Generating Station.
Due to the lack of pollutant loadings data, the White River case study model did not
account for the ground water impacts from Petersburg Generating Station associated with the
damage case listed in Appendix A. In 1997, the catastrophic release of coal combustion residuals
degraded the quality of ground water and surface water around the plant.
The White River case study model does not account for pollutant loadings from Hoosier
Energy's Frank E. Ratts (Ratts) Generating Station (Plant ID 2314), a 232-MW steam electric
power plant located less than a mile downstream of Petersburg Generating Station. Based on
8^34
-------
Section 8—Case Study Modeling
information obtained in responses to the Steam Electric Survey, Ratts Generating Station
discharged one or more of the evaluated wastestreams directly to the White River. This plant,
however, has publicly announced plans to retire all of its steam generating units prior to
implementation of the final rule. EPA therefore excluded pollutant loadings from the Ratts
Generating Station so that the changes in pollutant loadings during the modeling period, and the
associated environmental improvements, reflect only those attributable to the final rule.
Modeling Period
The modeling period starts in 1986 (the year the last generating unit at Petersburg
Generating Station began operating) and extends through 2034, covering a period of 49 years.
Based on Petersburg Generating Station's NPDES permitting cycle, EPA assumes that the plant
will achieve the limitations under the final rule by 2019.
Modeling Results - Water Quality
Under baseline conditions, the modeled pollutant concentrations in Lick Creek, the
immediate receiving water exceed NRWQC water quality benchmarks for five modeled
pollutants, indicating that pollutant loadings from the Petersburg Generating Station may
quantifiably reduce water quality in the modeled portions of Lick Creek. Additionally, the
modeled pollutant concentrations in portions of the White River downstream of Lick Creek
exceed NRWQC water quality benchmarks for four of the modeled pollutants, indicating that the
water quality downstream of Lick Creek may also be reduced by the pollutant loadings form
Petersburg Generating Station.
The baseline modeled pollutant concentrations exceed human health criteria primarily for
arsenic, thallium, and selenium, as discussed below:
• Arsenic concentrations in Lick Creek exceed the water quality benchmark for
consumption of water and organisms (0.018 |ig/L) for the entire modeling period.
These exceedances continue downstream in the White River, at the same frequency,
throughout the entire 50-mile-long modeling area downstream of the plant.
• Arsenic concentrations in Lick Creek also exceed the higher water quality benchmark
for consumption of organisms only (0.14 |ig/L) for the entire modeling period. These
exceedances continue downstream in the White River, generally at the same
frequency, throughout the entire 50-mile-long modeling area downstream of the plant.
• Thallium concentrations in Lick Creek exceed the water quality benchmarks for
consumption of water and organisms (0.24 |ig/L) for the entire modeling period.
These exceedances continue downstream in the White River, at a much lower
frequency (less than 2 percent of the modeling period), throughout the entire 50-mile-
long modeling area downstream of the plant.
• Thallium concentrations in Lick Creek also exceed the higher water quality
benchmark for consumption of organisms only (0.47 |ig/L) for the entire modeling
period. On rare occasions (less than 1 percent of the modeling period), thallium
concentrations in reaches downstream in the White River also exceed this benchmark.
8-35
-------
Section 8—Case Study Modeling
These downstream exceedances occur in 26 miles of the modeling area downstream
of the plant and up to 31 miles downstream of the plant.
• On rare occasions (less than 1 percent of the modeling period), selenium
concentrations in Lick Creek exceed the water quality benchmark for consumption of
water and organisms (170 |ig/L). These exceedances do not occur downstream after
the confluence of the Lick Creek and White River.
These case study modeling results indicate that, under baseline conditions, humans
consuming water and/or organisms inhabiting these modeled portions of Lick Creek and the
White River may be more at risk of the negative effects associated with oral exposure to these
pollutants (see Section 3.1.1).
Aquatic organisms may be at risk for exposure to copper, selenium, and cadmium under
baseline conditions, as discussed below:
• Copper concentrations in Lick Creek exceed the freshwater aquatic life criteria for
chronic exposure (9.0 |ig/L) for 45 percent of the modeling period. These
exceedances do not occur downstream after the confluence of the Lick Creek and
White River.
• Copper concentrations in Lick Creek also exceed the higher freshwater aquatic life
criteria for acute exposure (13 |ig/L) for 25 percent of the modeling period. These
exceedances do not occur downstream after the confluence of the Lick Creek and
White River.
• Selenium concentrations in Lick Creek exceed the freshwater aquatic life criteria for
chronic exposure (5.0 |ig/L) for 99 percent of the modeling period. On rare occasions
(less than 1 percent of the modeling period), selenium concentrations in reaches
downstream in the White River also exceed this benchmark. These downstream
exceedances occur in 21 miles of the modeling area downstream of the plant and up
to 32 miles downstream of the plant.
• Cadmium concentrations in Lick Creek exceed the freshwater aquatic life criteria for
chronic exposure (0.25 |ig/L) for 86 percent of the modeling period. On rare
occasions (less than 1 percent of the modeling period), cadmium concentrations in
reaches downstream in the White River also exceed this benchmark. These
downstream exceedances occur in 18 miles of the modeling area downstream of the
plant.
These modeling results indicate that, under baseline conditions, aquatic organisms
residing in the portions of Lick Creek and the White River with modeled exceedances may be
more at risk to negative impacts from chronic exposure to cadmium and selenium. Additionally,
the copper loadings from Petersburg Generating Station may pose a threat from chronic or acute
exposure.
Under baseline conditions, the modeled pollutant concentrations in Lick Creek and
downstream reaches in the White River exceed the MCL drinking water benchmarks for five
8-36
-------
Section 8—Case Study Modeling
modeled pollutants. The baseline modeled pollutant concentrations exceed drinking water
criteria for thallium, selenium, arsenic, lead, and cadmium as discussed below:
• Thallium concentrations in Lick Creek exceed the MCL benchmark (2 |ig/L) for 96
percent of the modeling period. These exceedances do not occur downstream after the
confluence of the Lick Creek and White River.
• Selenium concentrations in Lick Creek exceed the MCL benchmark (50 |ig/L) for 38
percent of the modeling period. These exceedances do not occur downstream after the
confluence of the Lick Creek and White River.
• Arsenic concentrations in Lick Creek exceed the MCL benchmark (10 |ig/L) for 34
percent of the modeling period. These exceedances occur in 8.0 miles of the modeling
area downstream of the plant and up to 35 miles downstream of the plant.
• On rare occasions (less than 1 percent of the modeling period), lead concentrations in
Lick Creek exceed the MCL benchmark (15 |ig/L). These exceedances continue to
occur downstream in 24 miles of the White River as far as the end of the model (50
miles downstream of the plant discharge).
• On rare occasions (less than 1 percent of the modeling period), cadmium
concentrations in Lick Creek exceed the MCL benchmark (0.25 |ig/L). These
exceedances do not occur downstream after the confluence of the Lick Creek and
White River.
Modeling results do not indicate any exceedances of NRWQC or MCL criteria for nickel
or zinc. Appendix G of this report includes figures that illustrate the water column pollutant
concentration output for the immediate receiving water for arsenic, cadmium, copper, lead,
selenium, and thallium. These figures also present the NRWQC and MCL benchmarks for the
pollutant and the steady-state water column pollutant concentrations predicted by the IRW
model.
The final rule modeling results show significantly decreased concentrations of all
modeled pollutants in the immediate receiving water (Lick Creek), which will greatly improve
water quality. The final modeling results also demonstrate that the reduction of pollutant
loadings from Petersburg Generating Station will significantly reduce the concentrations of
selenium and thallium in the White River, downstream of Lick Creek. These pollutant removals
result in fewer exceedances of NRWQC and MCL benchmarks compared to those estimated in
the baseline modeling. Case study modeling results for Lick Creek and the White River reveal
the following water quality improvements under the final rule:
• For arsenic:
Exceedances of the MCL benchmark are eliminated in Lick Creek. Despite the
continued exceedances of this benchmark, at the same frequency, downstream in
the White River, reducing the pollutant concentrations in the water column may
decrease the human health risk.
Exceedances of the human health water quality benchmark for consumption of
organisms reduce in frequency from 100 percent to 87 percent of the modeling
8-37
-------
Section 8—Case Study Modeling
period in Lick Creek. Despite the continued exceedances of this human health
criteria, at the same frequency, downstream in the White River, reducing the
pollutant concentrations in the water column may decrease the risk to humans
consuming contaminated organisms.
• For cadmium:
Exceedances of the aquatic life water quality criteria for chronic impacts are
eliminated throughout the entire modeling area.
Exceedances of the MCL benchmark (observed only in Lick Creek under baseline
conditions) are eliminated throughout the entire modeling area.
• For copper:
Exceedances of the aquatic life water quality criteria for chronic and acute
impacts (observed only in Lick Creek under baseline conditions) are eliminated
throughout the entire modeling area.
• For lead:
Exceedances of the MCL benchmark are eliminated in Lick Creek. Despite the
continued exceedances of this benchmark, at the same frequency, downstream in
the White River, reducing the pollutant concentrations in the water column may
decrease the human health risk.
• For selenium:
- Exceedances of the aquatic life water quality criteria for chronic impacts are
eliminated throughout the entire modeling area.
- Exceedances of the human health water quality benchmark for consumption of
water and organisms (observed only in Lick Creek under baseline conditions) are
eliminated throughout the entire modeling area.
Exceedances of the MCL benchmark (observed only in Lick Creek under baseline
conditions) are eliminated throughout the entire modeling area.
• For thallium:
Exceedances of the MCL benchmark reduce in frequency from 96 percent to less
than 1 percent of the modeling period in Lick Creek.
- Exceedances of the human health water quality benchmark for consumption of
water and organisms reduce in frequency from 100 percent to 84 percent of the
modeling period in Lick Creek. Exceedances of this benchmark are eliminated
through the modeling area downstream of the immediate receiving water (after
the confluence of the Lick Creek and White River).
- Exceedances of the human health water quality benchmark for consumption of
organisms reduce in frequency from 100 percent to 61 percent of the modeling
period in Lick Creek. Exceedances of this benchmark are eliminated through the
modeling area downstream of the immediate receiving water (after the confluence
of the Lick Creek and White River).
8-38
-------
Section 8—Case Study Modeling
The final rule modeling results demonstrate that, due to background concentrations of
arsenic from upstream sources, there will still be exceedances of the human health water quality
benchmark for consumption of water and organisms throughout the entire modeling area
downstream of the plant; however, the final rule will reduce the arsenic loadings that the
Petersburg Generating Station contributes to the White River.
Modeling Results - Wildlife
Under baseline conditions, Lick Creek may pose a risk to minks and eagles that consume
fish contaminated with selenium. The average modeled selenium concentration in Lick Creek is
more than 18 times greater than the concentration that would translate to NEHC exceedances for
minks and eagles, demonstrating that this portion of the immediate receiving water may pose a
potential reproductive threat to terrestrial food webs. The water concentrations downstream after
the confluence of the Lick Creek and White River do not pose a threat to these indicator species.
Modeling results indicate that on rare occasions (less than 1 percent of the modeling
period), nickel concentrations in benthic sediment downstream reaches exceed the CSCL
benchmark (18 mg/kg). These exceedances occur in 3.0 miles of the modeling area downstream
of the plant and up to 35 miles downstream of the plant.
The case study modeling results demonstrate that the final rule will significantly reduce
pollutant concentrations and the associated impacts to wildlife that inhabit Lick Creek. The final
rule will eliminate selenium exceedances of the NEHC benchmarks for minks and eagles in all
modeled reaches of Lick Creek. Despite the modeling not being able to quantify any
improvements to benthic organisms under the final rule, the pollutant loading removals will
decrease the concentrations of toxic pollutants in benthic sediment and decrease the exposure of
organisms to these pollutants.
Modeling Results - Human Health
EPA modeled the average pollutant concentrations in the water column and compared
these to the concentrations that would trigger exceedances of either the non-cancer reference
dose or the 1-in-a-million LECR. Under baseline conditions, the average water column
concentration of arsenic in the immediate receiving water over the modeling period results in an
estimated cancer risk of approximately 3-in-a-million for adult subsistence fishers. Therefore,
adults who frequently consume arsenic-contaminated fish inhabiting the immediate receiving
water may be at greater risks for development of cancer.
Based on the average pollutant concentrations in the water column under baseline
conditions, cadmium, selenium, and thallium pose the greatest threat to cause non-cancer health
effects in humans from fish consumption, as discussed below:
• Average thallium concentrations in Lick Creek are significantly greater than the
concentrations that would translate to exceedances of the reference doses for adult
and children recreational and subsistence fishers (all national-scale cohorts
evaluated), with some cohorts potentially being exposed to concentrations more than
200 times the reference dose. The water concentrations downstream after the
8-39
-------
Section 8—Case Study Modeling
confluence of the Lick Creek and White River do not pose a threat to any of the
evaluated cohorts.
• Average selenium concentrations in Lick Creek are greater than the concentration that
would translate to exceedances of the reference doses for adult and children
recreational and subsistence fishers (all national-scale cohorts evaluated). The water
concentrations downstream after the confluence of the Lick Creek and White River
do not pose a threat to any of the evaluated cohorts.
• Average cadmium concentrations in Lick Creek are greater than the concentration
that would translate to exceedances of the reference doses for the child (younger than
11 years old) subsistence fisher cohorts. The water concentrations downstream after
the confluence of the Lick Creek and White River do not pose a threat to any of the
evaluated cohorts.
Therefore, humans who consume thallium-, selenium-, or cadmium-contaminated fish
inhabiting Lick Creek may be at greater risk for developing the negative health effects associated
with these pollutants, which are discussed in Section 3.1.1.
The final rule modeling results demonstrate significant reductions in selenium and
cadmium concentrations in Lick Creek, which will eliminate exceedances of the non-cancer
health effects reference dose for all cohorts for these pollutants. While the modeling results
continue to show thallium water concentrations that would translate to exceedances of the non-
cancer health effects reference doses for all cohorts, the final rule will reduce the magnitude of
the human health impacts and reduce thallium loading contributions from Petersburg Generating
Station.
Interpretation of Lick Creek and White River Results
Case study modeling results for Lick Creek indicate that there are severe water quality,
wildlife, and human health impacts in Lick Creek. Case study modeling of Lick Creek reveals
more exceedances of water quality and human health benchmarks than the IRW model; however,
the IRW model predicts more impacts to benthic organisms than the case study modeling results.
The exceedances identified in Lick Creek are based solely on discharges of the evaluated
wastestreams from Petersburg Generating Station because EPA did not identify any STORET
monitoring data or point sources suggesting any other sources were contributing pollutant
discharges on this small tributary.
The pollutant loadings discharged by Petersburg Generating Station contribute to the
overall concentrations in the White River, along with other upstream sources. Case study
modeling indicates that some of the water quality impacts identified in Lick Creek for arsenic,
cadmium, selenium, thallium, and lead can occur in the White River, far downstream of where
Lick Creek flows into it. For thallium, these downstream impacts are solely caused by the
discharges of the evaluated wastestreams from the plant because EPA did not identify any other
sources of thallium within the modeling period. For arsenic and lead, the projected exceedances
are driven by the background concentrations flowing into the White River modeling area.
Pollutant loadings from Petersburg Generating Station may be further impairing the degraded
waterway for arsenic and lead. For lead and zinc, the average water column concentrations are
8-40
-------
Section 8—Case Study Modeling
highest downstream in the White River, indicating that pollutants with high partition coefficients
may pose a greater threat to humans and aquatic life in the White River than in Lick Creek. The
case study modeling results suggest that while high concentrations of toxic pollutants may dilute
once Lick Creek empties into the White River, there are still impacts downstream that are not
captured by the IRW model.
Under the final rule, case study modeling of Lick Creek and the White River indicate that
both these waterbodies will exhibit fewer exceedances of water quality benchmarks.
Additionally, Lick Creek will no longer pose reproductive risks to higher trophic-level wildlife
and will pose less risk to humans consuming fish for cancer and non-cancer impacts. Case study
modeling predicts more water quality improvements in the modeling area than the IRW model.
This is due in part to the greater water quality impacts under baseline conditions, which created
additional opportunities for modeled improvements, and in part to the identified improvements in
downstream reaches of the White River that were not evaluated as part of the IRW model. Case
study modeling predict fewer human health improvements than the IRW model. The average
pollutant concentrations throughout the entire modeling area reduce promptly after compliance
with the final rule.
8.2.4 Ohio River Case Study
The 948-mile Ohio River flows westward from Pittsburgh, Pennsylvania, to Cairo,
Illinois, where it meets the Mississippi River. According to 2013 TRI reporting, 23 million
pounds of chemicals were discharged into the Ohio River, more than any other surface water in
the TRI database [U.S. EPA, 2013a]. EPA identified that 24 steam electric power plants
evaluated in the EA discharge one or more of the evaluated wastestreams to the Ohio River or to
tributaries that flow into the Ohio River in under five miles. FirstEnergy Corp. (FirstEnergy)
owns and operates several of the coal-fired power plants that discharge to the Ohio River.
The Bruce Mansfield plant (Plant ID 2269) is FirstEnergy's largest coal-fired power plant
by nameplate capacity. The plant is located in Shippingport, Pennsylvania, along the Ohio River,
approximately 25 miles northwest of Pittsburgh. This plant operates three stand-alone steam
turbines, each with a nameplate capacity of 914 MW. These three generating units have a total
capacity of 2,741 MW and reported producing approximately 19,000,000 MWh of electricity in
2009 [ERG, 2015J]. The Bruce Mansfield plant discharges FGD wastewater and bottom ash
transport water directly to the Ohio River from the Little Blue Run surface impoundment, which
straddles the border of Pennsylvania and West Virginia. Table 8-5 contains general information
about the three coal-fired generating units at the Bruce Mansfield plant.
Located along the Ohio River in Stratton, Ohio, FirstEnergy's W.H. Sammis plant (Plant
ID 103) is the largest coal-fired power plant in Ohio. W.H. Sammis Plant's seven stand-alone
steam turbine generating units have a total nameplate capacity of 2,460 MW. Based on data EPA
obtained in responses to the Steam Electric Survey, the W.H. Sammis plant reported generating
more than 9,500,000 MWh of energy with these seven coal-fired generating units in 2009. The
W.H. Sammis plant discharges three of the evaluated wastestreams (FGD wastewater, bottom
ash transport water, and combustion residual leachate) directly to the Ohio River. Table 8-6
contains general information about each of the seven steam electric generating units at the W.H.
Sammis plant.
8-41
-------
Section 8—Case Study Modeling
Table 8-5. Summary of Bruce Mansfield Operations
SE Unit
1
2
3
Fuel
Bituminous coal and
No. 2 fuel oil
Bituminous coal and
No. 2 fuel oil
Bituminous coal and
No. 2 fuel oil
Capacity
(MW)
914
914
914
Fly Ash
Wet scrubber a
Wet scrubber a
Dry conveyed
Bottom Ash
Wet handled to
impoundment
Wet handled to
impoundment
Wet handled to
impoundment
FGD
(Year Installed)
Wet system
(1975)
Wet system
(1977)
Wet system
(1980)
Source: ERG, 2015J.
Acronyms: FGD (Flue gas desulfurization); MW (Megawatt); SE (steam electric).
a - EPA does not consider the ash collected by venturi-type wet scrubbers as fly ash, and therefore, the water
generated by these systems is not considered fly ash transport water.
Table 8-6. Summary of W.H. Sammis Operations
SE Unit
1
2
o
6
4
5
6
7
Fuel
Bituminous coal,
subbituminous coal,
and No. 2 fuel oil
Bituminous coal,
subbituminous coal,
and No. 2 fuel oil
Bituminous coal,
subbituminous coal,
and No. 2 fuel oil
Bituminous coal,
subbituminous coal,
and No. 2 fuel oil
Bituminous coal,
subbituminous coal,
and No. 2 fuel oil
Bituminous coal,
subbituminous coal,
and No. 2 fuel oil
Bituminous coal,
subbituminous coal,
and No. 2 fuel oil
Capacity (MW)
190
190
190
190
334
680
680
Fly Ash
Dry conveyed
Dry conveyed
Dry conveyed
Dry conveyed
Dry conveyed
Dry conveyed
Dry conveyed
Bottom Ash
Wet handled to
impoundment
Wet handled to
impoundment
Wet handled to
impoundment
Wet handled to
impoundment
Wet handled to
impoundment
Wet handled to
impoundment
Wet handled to
impoundment
FGD
(Year Installed)
Wet system
(2010)
Wet system
(2010)
Wet system
(2010)
Wet system
(2010)
Wet system
(2010)
Wet system
(2010)
Wet system
(2010)
Source: ERG, 2015J.
Acronyms: FGD (Flue gas desulfurization); MW (Megawatt); SE (steam electric).
In estimating the historical pollutant loadings associated with W.H. Sammis' three FGD
systems, EPA incorporated the pollutant loadings for FGD wastewater as the systems were
installed, between March and May 2010. EPA did not model any FGD wastewater pollutant
loadings in the model prior to the installation of W.H. Sammis plant's first FGD system.
8-42
-------
Section 8—Case Study Modeling
Modeling Area
The Ohio River WASP model encompasses a 49-mile-long reach of the Ohio River, 37
miles of which is downstream of one or both of the two modeled steam electric power plant
immediate receiving waters. Located furthest upstream, the Bruce Mansfield plant discharges
approximately 12 miles downstream of the start of the modeling area. The immediate receiving
water that the Bruce Mansfield plant discharges to is approximately 3.3 miles long, as defined in
the WASP model. W.H. Sammis plant discharges 13 miles downstream of the Bruce Mansfield
plant's immediate receiving water. The immediate receiving water that W.H. Sammis plant
discharges to is approximately 3.4 miles long, as defined in the WASP model. The modeling area
ends just upstream of the discharges from another steam electric power plant, the Cardinal plant
(Plant ID 3265). EPA did not model the pollutant loadings from the Cardinal plant because of
CBI claims on one or more of the evaluated wastestream flow rates. Figure 8-5 illustrates the
location and extent of the Ohio River WASP model.
PENNSYL-
VANIA
Plant ID 103
W, H. Sammis Plant
winuxsville.
Aliquippa.
Plant ID 2269
Bruce Mansfield Plant
•£. Ash Surface Impoundment
Q* for Bruce Mansfield Plant
(Little Blue Run)
TH too will*.
York villa
WestLibet ty
LEGEND
Modeling Area
^^— Immediate Receiving Water
^^" Other Modeled Reaches
Modeled Point Sources
• Case Study Steam Electric Power Plant
D Non-Steam-Electric Point Sources
Monitoring Data Stations
• Entering Modeling Area
o Within Modeling Area
024 8 12
) Miles
Figure 8-5. Ohio River WASP Modeling Area
8-43
-------
Section 8—Case Study Modeling
Identified Point Sources and Background Concentrations
As discussed below, EPA reviewed available pollutant loadings (DMR and TRI) and
monitoring data (STORET) for potential incorporation into the Ohio River WASP model to
represent pollutant contributions from background and non-steam-electric point sources, and for
use in calibrating the model results.
• Upstream pollutant contributions. EPA identified many upstream non-steam-
electric point sources whose pollutant loadings could influence the model results.
EPA identified STORET data from one monitoring station on the Ohio River
(approximately 28 river-miles upstream of Bruce Mansfield plant's immediate
receiving water). EPA incorporated the monitoring data (which encompass five of the
modeled pollutants) to represent the pollutant contributions flowing into the modeling
area. EPA identified additional STORET monitoring data from one station on a
tributary to the Ohio River; EPA incorporated these data to represent pollutant
contributions flowing in from that tributary. EPA also incorporated the pollutant
loadings, based on DMR and TRI data, from seven non-steam-electric point sources
upstream of the Bruce Mansfield plant's immediate receiving water to account for the
pollutant contributions not captured by the STORET monitoring data.
• Downstream pollutant contributions. EPA incorporated STORET data from eight
monitoring stations to represent TSS concentrations flowing into the modeling area
downstream of both steam electric power plant immediate receiving waters (i.e.,
tributaries flowing into the Ohio River). These monitoring stations all represent one
tributary that flows into the Ohio River near the downstream end of the modeling
area. EPA identified 29 non-steam-electric point sources whose pollutant loadings
could influence the model results downstream of the Bruce Mansfield plant
immediate receiving water and incorporated these pollutant loadings into the Ohio
River WASP model.
• Monitoring data within the modeling area. EPA compiled STORET data from
seven monitoring stations located within the modeling area and used these data to
calibrate the WASP model.
The contributions of copper, lead, nickel, and zinc from upstream sources are
significantly greater than the pollutant loadings from the Bruce Mansfield and W.H. Sammis
plants.
The Ohio River case study model did not account for the documented surface water and
ground water impacts from Bruce Mansfield or Little Blue Run that are listed in Appendix A. In
1993, a catastrophic release of steam electric power plant wastewater compromised the quality of
ground water and surface water around the Bruce Mansfield plant and Little Blue Run
impoundment. Due to the lack of pollutant loadings data, surface water quality impacts resulting
from this event are not reflected in this model; therefore, the case study modeling could
underrepresent the actual baseline impacts of the Bruce Mansfield plant on the Ohio River.
8-44
-------
Section 8—Case Study Modeling
Modeling Period
The modeling period starts in 1982 (year of the last revision to the steam electric ELGs)
and extends through 2036, covering a period of 55 years. Based on their NPDES permitting
cycles, EPA assumes that the Bruce Mansfield and W.H. Sammis plants will achieve the
limitations under the final rule by 2020 and 2021, respectively. EPA focused the assessment of
the improvements under the final rule on the period after the 2021 assumed compliance date.
Modeling Results - Water Quality
Under baseline conditions, the modeled pollutant concentrations in the modeled portion
of the Ohio River exceed a human health NRWQC water quality benchmark for one modeled
pollutant (arsenic), indicating that arsenic loadings from the two steam electric power plants may
contribute to a quantifiable reduction in water quality in the modeled portions of the Ohio River.
Arsenic concentrations in 33 miles of the modeling area downstream of the Bruce Mansfield
plant exceed the human health water quality benchmark for consumption of water and organisms
(0.018 |ig/L). These exceedances begin several miles downstream of the Bruce Mansfield plant
due to the pollutant loadings from a non-steam-electric point source. This area of exceedances
continues downstream of the W.H. Sammis plant for 24 miles (including the W.H. Sammis
plant's immediate receiving water) and exceeds the arsenic benchmark during 30 percent of the
modeling period. In some portions of the modeling area, the frequency of these exceedances
increases due to arsenic contributions from other non-steam-electric point sources. These case
study modeling results indicate that, under baseline conditions, humans consuming water and/or
organisms inhabiting these modeled portions of the Ohio River may be more at risk of the
negative effects associated with oral exposure to arsenic (see Section 3.1.1). On rare occasions
(less than 1 percent of the modeling period), the modeled pollutant concentrations exceed the
MCL drinking water benchmark for one pollutant (lead), indicating that lead loadings from the
two steam electric power plants may contribute to a quantifiable reduction in water quality in the
modeled portions of the Ohio River. These rare lead exceedances occur in 15 miles of the
modeling area downstream of the Bruce Mansfield plant, of which 13 miles are also downstream
of the W.H. Sammis plant (including the immediate receiving water).
Modeling results do not indicate any exceedances of human health NRWQC criteria for
the other modeled pollutants (cadmium, copper, nickel, selenium, thallium, and zinc) and do not
indicate any exceedances of aquatic life NRWQC or MCL criteria for any of the eight modeled
pollutants. Appendix G of this report includes figures that illustrate the water column pollutant
concentration output for the immediate receiving water for arsenic and lead. These figures also
present the NRWQC and MCL benchmarks for the pollutant and the steady-state water column
pollutant concentrations predicted by the IRW model.
The final rule modeling results show significantly decreased concentrations of four of the
modeled pollutants (arsenic, cadmium, selenium, and thallium) in the modeled portion of the
Ohio River, which will improve water quality. These pollutant removals result in less frequent
exceedances of human health NRWQC benchmarks compared to those estimated in the baseline
modeling. Arsenic exceedances of human health water quality benchmarks for consumption of
water and organisms reduce in frequency from 30 percent to 6 percent of the modeling period in
the W.H. Sammis plant's immediate receiving water. Additionally, the exceedances of these
8-45
-------
Section 8—Case Study Modeling
benchmarks reduce in frequency in all remaining sections of the downstream modeling area
following compliance with the final rule. Despite the continued exceedances of the arsenic
human health criteria and the lead MCL benchmark, reducing the pollutant concentrations in the
water column may decrease the risk to humans.
Modeling Results - Wildlife
Based on the average pollutant concentrations in the water column under baseline
conditions, the modeled portion of the Ohio River does not exceed the concentrations that would
translate to NEHC exceedances and does not pose a risk to minks and eagles that consume
contaminated fish. Despite the modeling not being able to quantify any improvements to minks
and eagles under the final rule, the pollutant loading removals will decrease bioaccumulation of
toxic pollutants in the terrestrial food chains.
Modeling results do not indicate that there are any pollutant concentrations in the upper
benthic sediment that exceed CSCL benchmarks for any of the eight modeled pollutants;
therefore, the modeled portion of the Ohio River does not pose a threat to benthic organisms in
contact with contaminated sediment. Despite the modeling not being able to quantify any
improvements to benthic organisms under the final rule, the pollutant loading removals will
decrease the concentrations of toxic pollutants in benthic sediment and decrease the exposure of
organisms to these pollutants.
Modeling Results - Human Health
Under baseline conditions, the average concentration of arsenic in fish over the modeling
period does not result in an estimated cancer risk greater than 1-in-a-million for any of the
national-scale cohorts.
Based on the average pollutant concentrations in the water column under baseline
conditions, thallium poses the greatest threat to cause non-cancer health effects in humans from
fish consumption. Average thallium concentrations in the W.H. Sammis plant's immediate
receiving water are greater than the concentration that would translate to exceedances of the
reference doses for the child (younger than 11 years old) subsistence fisher cohorts. Average
thallium concentrations in 24 miles of the modeling area downstream of the W.H. Sammis plant
are high enough to trigger exceedances of the reference dose for at least one subsistence cohort.
Therefore, humans who consume fish inhabiting these waters may be at greater risk for
developing the negative health effects associated with thallium, which are discussed in Section
3.1.1.
The final rule modeling results demonstrate significant reductions in thallium,
eliminating thallium exceedances of the non-cancer health effects reference dose throughout the
entire modeling area.
Interpretation of Ohio River Results
Case study modeling results for the Ohio River indicate greater water quality and human
health impacts under baseline conditions than predicted by the IRW model. The impacts
identified in the Ohio River by case study modeling are more extensive than the IRW model
8-46
-------
Section 8—Case Study Modeling
because EPA has accounted for pollutant contributions from upstream on the Ohio River, other
waterways flowing into the Ohio River, and non-steam electric point sources. Modeled alone, the
Bruce Mansfield plant and W.H. Sammis plant would not cause any quantifiable impacts over
the modeling period; however the modeled potion of the Ohio River is heavily industrialized.
EPA identified 34 non-steam electric point sources that discharge one or more of the modeled
pollutants and report to DMR or TRI. The pollutant contributions from the Bruce Mansfield
plant, W.H. Sammis plant, and these other non-steam electric point sources modeled accumulate
in the waterbody, increasing the overall water column concentrations to a degree that adversely
affects water quality and human health. EPA identified exceedances of human health
benchmarks that indicate that consuming water and/or organisms from the modeled portion of
the Ohio River, including the W.H. Sammis plant's immediate receiving water and areas
downstream, can cause health problems related to arsenic, lead, or thallium. The Ohio River case
study model results exemplify that, by not accounting for non-steam-electric point sources
discharging to the same waterbodies as steam electric power plants, the IRW model may be
under-representing the total number of receiving waters with impacts that are caused, in part, by
pollutant contributions from the steam electric power generating industry. The case modeling
results also suggest that the discharges of the evaluated wastestreams from Bruce Mansfield
plant and W.H. Sammis plant may be further impairing the degraded waterway.
Case study modeling of the Ohio River indicates that, under the final rule, the Ohio River
will exhibit less frequent exceedances of water quality benchmarks and will eliminate risk to
humans consuming fish that inhabit these waters. The human health non-cancer impacts and
improvements under the final rule are solely caused by the reduction in steam electric plant
pollutant loadings (there are no other input sources of thallium in the Ohio River WASP
model).The improvements identified by the case study model are more extensive than what was
projected by the IRW model for either of Bruce Mansfield plant or W.H. Sammis plant. This is
due in part to the greater water quality and human health impacts under baseline conditions,
which created additional opportunities for modeled improvements, and in part to the identified
improvements in downstream reaches of the Ohio River that were not evaluated as part of the
IRW model. The average pollutant concentrations throughout the entire modeling area reduce
within a year after compliance with the final rule.
8.2.5 Mississippi River Case Study
The Mississippi River watershed is the largest in North America, covering about 40
percent of the lower 48 states. The 190-mile stretch of the Mississippi River between the
confluence with the Missouri River at St. Louis, Missouri, and the confluence with the Ohio
River at Cairo, Illinois, is known as the Middle Mississippi River. South of St. Louis along this
stretch of the river, Ameren Corporation operates the Rush Island steam electric power plant
(Plant ID 5038) on the west bank of the Mississippi River. The Rush Island plant operates two
stand-alone steam turbine units with a nameplate capacity of 670 MW each. Together, these two
coal-fired generating units have a capacity of 1,340 MW and reported producing over 8,500,000
MWh of electricity in 2009 in the Steam Electric Survey. The Rush Island plant discharges fly
ash and bottom ash transport water directly to the Mississippi River. Table 8-7 contains general
information on the two coal-fired units at the Rush Island plant.
8-47
-------
Section 8—Case Study Modeling
Table 8-7. Summary of Rush Island Operations
SE Unit
1
2
Fuel
Subbituminous coal
and No. 2 fuel oil
Subbituminous coal
and No. 2 fuel oil
Capacity
(MW)
670
670
Fly Ash
Dry conveyance & wet
handled to impoundment
Dry conveyance & wet
handled to impoundment
Bottom Ash
Wet handled to
impoundment
Wet handled to
impoundment
FGD
(Year Installed)
No FGD system
No FGD system
Source: ERG, 2015J.
Acronyms: FGD (Flue gas desulfurization); MW (Megawatt); SE (steam electric).
Modeling Area
The Mississippi River WASP model encompasses a 46-mile-long reach of the
Mississippi River, 23 miles of which is downstream of the Rush Island plant immediate
receiving water. The model has two start boundaries that are on the Meramec River and
Mississippi River shortly upstream of their confluence. The immediate receiving water that the
Rush Island plant discharges to is approximately 1.5 miles long, as defined in the WASP model.
This model ends at the confluence of the Mississippi River and Kaskaskia River. Figure 8-6
illustrates the location and extent of the Mississippi River WASP model.
Identified Point Sources and Background Concentrations
As discussed below, EPA reviewed available pollutant loadings (DMR and TRI) and
monitoring data (STORET) for potential incorporation into the Mississippi River WASP model
to represent pollutant contributions from background and non-steam-electric point sources, and
for use in calibrating the model results.
• Upstream pollutant contributions from non-steam-electric point sources EPA
identified several upstream non-steam-electric point sources whose loadings could
influence the model results. EPA therefore extended the modeling area upstream to
model these point sources and incorporate upstream monitoring data. EPA identified
STORET data from four monitoring stations on the Mississippi River prior to the
confluence with the Meramec River (approximately 24 river-miles upstream of Rush
Island's immediate receiving water). EPA incorporated the monitoring data (which
encompass all of the modeled pollutants except for thallium) to represent the pollutant
contributions in the Mississippi River prior to where it converges with the Meramec
River. EPA assumed that the monitoring data adequately reflect the pollutant
contributions from upstream of this confluence. EPA incorporated the pollutant
loadings from three non-steam-electric point sources downstream of the convergence
to account for the pollutant contributions not captured by the STORET monitoring
data.
• Upstream pollutant contributions from steam electric sources EPA identified one
steam electric power plant, Ameren's Meramec plant (Plant ID 1435), whose loadings
could influence the model results at the Rush Island immediate receiving water and
other downstream locations. EPA incorporated the loadings from the Meramec plant
into the extended Mississippi River model, as discussed further below.
8-48
-------
Section 8—Case Study Modeling
Arnold*
LEGEND
Modeling Area
^^— Immediate Receiving Water
^^" Other Modeled Reaches
Modeled Point Sources
• Case Study Steam Electric Power Plant
• Other Steam Electric Power Plant
D Non-Steam-Electric Point Sources
Monitoring Data Stations
• Entering Modeling Area
o Within Modeling Area
012 4 6 8 10 12 14
J Miles
.Columb*.
.Red Bud
Plant ID 5038
Rush Island
Content may not reflect National Geographies currgni inap policy Sources .National Geographic. Esn.
DeLorme HERE, UNEP-WCMC USGS, NftSA'ESA MET! NRCAN. GEBCO t^JOAA increW.i' .. ;
Figure 8-6. Mississippi River WASP Modeling Area
• Downstream pollutant contributions. EPA incorporated STORET data from two
monitoring stations to represent pollutant concentrations flowing into the modeling
area downstream of the Rush Island immediate receiving water (i.e., tributaries
flowing into the Mississippi River). EPA did not identify any non-steam-electric point
sources whose pollutant loadings would significantly influence the model results in
the downstream modeling area.
• Monitoring data within the modeling area. EPA compiled STORET data from four
monitoring stations located within the modeling area and used these data to calibrate
the WASP model.
The Meramec plant discharges approximately 24 river miles upstream of the Rush Island
plant's immediate receiving water. EPA did not identify STORET monitoring data between the
two plants to represent the pollutant concentrations from the Meramec plant; therefore, EPA
incorporated the pollutant loadings from the Meramec plant (as calculated for this rulemaking)
into the Mississippi River model. The Meramec plant operates four coal-fired generating units
with a total nameplate capacity of 923 MW. All pollutant loadings from the evaluated
wastestreams are from bottom ash transport water. EPA assumed that the Meramec plant will
8-49
-------
Section 8—Case Study Modeling
comply with the standards of the final rule by 2019. EPA did not evaluate the water quality,
wildlife, or human health impacts associated with discharges from the Meramec plant because
this plant did not meet the case study location selection criteria described in Section 8.1.1. EPA
incorporated the loadings from Meramec plant solely to account for the upstream pollutant
contributions flowing into the Rush Island plant's immediate receiving water from upstream,
under baseline conditions and the final rule.
The contributions of arsenic, cadmium, copper, lead, nickel, and zinc from upstream
sources are significantly greater than the pollutant loadings from the Rush Island plant.
Modeling Period
The modeling period starts in 1982 (year of the last revision to the steam electric ELGs)
and extends through 2036, covering a period of 55 years. Based on their NPDES permitting
cycles, EPA assumes that the Meramec and Rush Island plants will achieve the limitations under
the final rule by 2019 and 2023, respectively. For the Rush Island plant's immediate receiving
water and downstream reaches, EPA focused the assessment of the baseline impacts and
improvements under the final rule on the period after the 2023 assumed compliance date.
Modeling Results - Water Quality
Under baseline conditions, the modeled pollutant concentrations in the Rush Island
plant's immediate receiving water and downstream reaches exceed human health NRWQC water
quality benchmarks for one modeled pollutant (arsenic), indicating that loadings from Rush
Island may contribute to a quantifiable reduction in water quality in the modeled portions of the
Mississippi River. Arsenic concentrations in the Rush Island plant's immediate receiving water
exceed the human health water quality benchmark for consumption of water and organisms
(0.018 |ig/L) and the human health water quality benchmark for consumption organisms (0.14
|ig/L) for the entire modeling period. These exceedances continue downstream, at the same
frequency, throughout the entire 23-mile-long modeling area downstream of the plant. The case
study modeling results indicate that, under baseline conditions, humans consuming water and/or
organisms that inhabit these modeled portions of the Mississippi River may be more at risk of the
negative effects associated with oral exposure to arsenic (see Section 3.1.1).
Modeling results do not indicate any exceedances of human health NRWQC benchmarks
for the other modeled pollutants (cadmium, copper, nickel, lead, selenium, thallium, and zinc). In
addition, modeling results do not indicate any exceedances of aquatic life NRWQC or MCL
criteria for any of the eight modeled pollutants. Appendix G of this report includes figures that
illustrate the water column pollutant concentration output for the immediate receiving water for
arsenic. This figure also presents the NRWQC and MCL benchmarks for the pollutant and the
steady-state water column pollutant concentrations predicted by the IRW model.
The final rule modeling continues to show human health NRWQC benchmark
exceedances for arsenic within the Mississippi River due to additional arsenic contributions from
other sources (i.e., Mississippi River background concentrations and non-steam electric point
sources). However, under the final rule, both the Meramec and Rush Island plants will no longer
8-50
-------
Section 8—Case Study Modeling
discharge any of the evaluated wastestreams and will therefore no longer contribute to the
arsenic or lead impairment of the Mississippi River.
Modeling Results - Wildlife
Based on the average pollutant concentrations in the water column under baseline
conditions, the modeled portion of the Mississippi River does not exceed the concentrations that
would translate to NEHC exceedances and does not pose a risk to minks and eagles that consume
contaminated fish. Despite the modeling not being able to quantify any improvements to minks
and eagles under the final rule, the pollutant loading removals will decrease bioaccumulation of
toxic pollutants in the terrestrial food chains.
Modeling results do not indicate that there are any pollutant concentrations in the upper
benthic sediment that exceed CSCL benchmarks of for any of the eight modeled pollutants;
therefore, the modeled portion of the Mississippi River does not pose a threat to benthic
organisms in contact with contaminated sediment. Despite the modeling not being able to
quantify any improvements to benthic organisms under the final rule, the pollutant loading
removals will decrease the concentrations of toxic pollutants in benthic sediment and decrease
the exposure of organisms to these pollutants.
Modeling Results - Human Health
EPA modeled the average pollutant concentrations in the water column and compared
these to the concentrations that would trigger exceedances of either the non-cancer reference
dose or the 1-in-a-million LECR. Under baseline conditions, the average water column
concentration of arsenic throughout the modeling area downstream of the plant results in an
estimated cancer risk greater than 1-in-a-million for adult subsistence fishers. Therefore, humans
who consume arsenic-contaminated fish inhabiting the immediate receiving water may be at
greater risks for development of cancer. Modeling results demonstrate no reduction in the cancer
risk from inorganic arsenic under the final rule.
Under baseline conditions, the average pollutant concentrations over the modeling period
does not pose the threat to cause non-cancer health effects for adult and children recreational and
subsistence fishers (all national-scale cohorts evaluated).
Interpretation of Mississippi River Results
Case study modeling results for the Mississippi River indicate greater water quality and
human health impacts under baseline conditions than predicted by the IRW model. By
accounting for pollutant contributions from background and upstream sources, the case study
model predicts higher pollutant concentrations under baseline conditions. For arsenic, the
projected exceedances are driven by the pollutant contributions entering the Mississippi River
upstream of the Rush Island plant. Alone, the steam electric discharges of the evaluated
wastestreams would not cause any quantifiable impacts, which is consistent with the IRW model
results; however, the pollutant loadings from the Rush Island plant may be further exacerbating
the impairment of the degraded waterway.
8-51
-------
Section 8—Case Study Modeling
The case study modeling of the Mississippi River indicates that, under the final rule, it
will continue to exceed all of the water quality and human health benchmarks observed at
baseline, with little to no reduction in frequency. Under the final rule, the Rush Island plant will
no longer discharge any fly ash or bottom ash transport water. After compliance with the final
rule, the modeled steam electric power plants will no longer contribute to the impairment of the
Mississippi River and the overall magnitude of the pollutant concentrations in the aquatic system
will decrease.
8.2.6 Lake Sinclair Case Study
Lake Sinclair is a reservoir located in central Georgia. The lake was created in 1953 when
the waters of the Oconee River were dammed by Georgia Power, a subsidiary of Southern
Company, to create a hydroelectric generating station. Georgia Power also owns and operates
Plant Harllee Branch (Plant ID 5762), a steam electric power plant situated on the northern shore
of Lake Sinclair. Based on 2009 data obtained in responses to the Steam Electric Survey, Plant
Harllee Branch operated four coal-fired generating units with a total nameplate capacity of 1,750
MW and produced more than 6,800,000 MWh of electricity in 2009. As of April 16, 2015 (the
date by which the plant would be required to comply with the U.S. EPA's Clean Power Plan
[Clean Air Act Section lll(d)]), this plant has decertified and retired all four of its coal-fired
generating units. Georgia Power cited several factors, including the cost to comply with existing
and future environmental regulations, recent and future economic conditions, and lower natural
gas prices, in the decision to close the plant. Plant Harllee Branch discharged FGD wastewater,
fly ash transport water, and bottom ash transport water directly to Lake Sinclair. Table 8-8
contains general information on the four coal-fired units at Rush Island Plant.
Despite the retirement of all coal-fired generating units at this plant, EPA proceeded with
case study modeling of Lake Sinclair to represent the potential impacts of steam electric
discharges on lentic waterbodies (including the 26 lake, pond, and reservoir receiving waters
evaluated in this EA) and the potential environmental improvements that could reasonably be
expected under the final rule in other lentic waterbodies that receive discharges of the evaluated
wastestreams. EPA did not include Plant Harllee Branch or Lake Sinclair in the other
quantitative and qualitative analyses in this EA for the final rule (e.g., the IRW model).
In estimating the historical pollutant loadings associated with Plant Harllee Branch, EPA
incorporated the loadings only from generating unit IDs 3 and 4 because generating unit IDs 1
and 2 were flagged for retirement at the time of the proposed revised ELGs. EPA incorporated
the loadings with the FGD wastewater as the systems were installed (starting in 2013). EPA did
not model any FGD wastestream loadings in the historical model prior to the installation of Plant
Harllee Branch's first FGD system.
8-52
-------
Section 8—Case Study Modeling
Table 8-8. Summary of Plant Harllee Branch Operations
SE Unit
la
2a
3
4
Fuel
Bituminous coal and
No. 2 fuel oil
Bituminous coal and
No. 2 fuel oil
Bituminous coal and
No. 2 fuel oil
Bituminous coal and
No. 2 fuel oil
Capacity
(MW)
299
359
544
544
Fly Ash
Wet handled to
impoundment
Wet handled to
impoundment
Wet handled to
impoundment
Wet handled to
impoundment
Bottom Ash
Wet handled to
impoundment
Wet handled to
impoundment
Wet handled to
impoundment
Wet handled to
impoundment
FGD
(Year Installed)
Wet system
(2014)
Wet system
(2014)
Wet system
(2013)
Wet system
(2013)
Source: ERG, 2015J.
Acronyms: FGD (Flue gas desulfurization); MW (Megawatt); SE (steam electric).
a - EPA did not model any pollutant loadings associated with these generating units.
Modeling Area
As discussed in Section 8.1.1, EPA relied upon the availability of existing models to
perform case study modeling of lentic systems: an existing WASP model that divided the
waterbody into segments and EFDC model that provided hydrodynamics and simulated the
aquatic system in three dimensions. The EFDC model uses stretch or sigma vertical coordinates
and Cartesian coordinates to represent the physical characteristics of Lake Sinclair.
The three-dimensional EFDC model, which provides the hydrodynamic foundation for
the WASP model, divides the waterbody into 1,235 segments; each segment represents a unique
location and stratum within Lake Sinclair. The model accounts for a total volume of
approximately 340 million cubic meters. In contrast to the WASP models that EPA developed to
model lotic systems, the Lake Sinclair model is not set up to quantify the pollutant
concentrations in the benthic sediment; therefore, EPA was unable to assess whether pollutant
accumulation in the sediment was occurring over prolonged discharge periods. Figure 8-7
illustrates the location and extent of the Lake Sinclair modeling area.
8-53
-------
Section 8—Case Study Modeling
\ GEORGIA \
J
Plant ID 5038
Plant Harllee Branch
LEGEND
Modeling Area
| Receiving Branch
I Other Modeled Reaches
Modeled Point Sources Monitoring Data Stations
I Case Study Steam Electric Power Plant • Entering Modeling Area
D Non-Steam-Electric Point Sources O Within Modeling Area
! Miles
Figure 8-7. Lake Sinclair WASP and EDFC Modeling Area
Identified Point Sources and Background Concentrations
As discussed below, EPA reviewed available pollutant loadings (DMR and TRI) and
monitoring data (STORET) for potential incorporation into the Lake Sinclair water quality model
to represent pollutant contributions from background and non-steam-electric point sources, and
for use in validating and calibrating the model results.
• Upstream pollutant contributions. EPA incorporated STORET data from three
monitoring stations to represent TOC and TSS contributions from upstream of Lake
Sinclair on the Oconee River. EPA did not identify sufficient STORET monitoring
data to represent the pollutant contributions of the eight modeled pollutants or any
upstream non-steam-electric point sources with loadings for the eight modeled
pollutants. EPA therefore assumed pollutant concentrations of zero within the water
column flowing into Lake Sinclair from the Oconee River.
• Other pollutant contributions. EPA incorporated STORET data from 15 monitoring
stations to represent the modeled pollutants, TOC, and TSS concentrations flowing
into Lake Sinclair from other streams. EPA did not identify any non-steam-electric
8-54
-------
Section 8—Case Study Modeling
point sources whose pollutant loadings would significantly influence the model
results.
• Monitoring data within the modeling area. EPA compiled STORET data from six
monitoring stations located within the modeling area and used these data to calibrate
the Lake Sinclair water quality model.
The pollutant concentrations entering the modeling area for arsenic, copper, lead, and
thallium which EPA calculated using monitoring data, are much greater than the pollutant
loadings from Lake Sinclair plant. The concentrations entering the modeling area for cadmium,
nickel, and zinc also strongly influence the model outputs.
Modeling Period
As discussed earlier in this section, EPA adopted the preexisting Lake Sinclair EFDC
model. The preexisting model was designed with seven years of hydrodynamic and flow input,
limiting the length of the period EPA could model. Based on Plant Harllee Branch's NPDES
permitting cycle, EPA assumed that the plant would have achieved the limitations under the final
rule by 2019 if it continued to operate. The modeling period begins in February 2012
(approximately seven years before the assumed compliance date) and extends through November
2025 (approximately seven years after the assumed compliance date).
Modeling Results - Water Quality
EPA selected three portions of Lake Sinclair to evaluate the modeled pollutant
concentrations: 1) the immediate receiving water (a 720,000-cubic-meter cell of the lake); 2) the
average of all segments in the reach of the lake where Plant Harllee Branch discharges, including
subsurface water segments (hereafter referred to as the "receiving branch"), and 3) the average of
all segments included in the Lake Sinclair model, including subsurface water segments (hereafter
referred to as the "entire modeling area").
Under baseline conditions, the modeled pollutant concentrations in Lake Sinclair,
including the immediate receiving water and the receiving reach, exceed NRWQC water quality
benchmarks for three modeled pollutants, indicating that pollutant loadings from Plant Harllee
Branch may quantifiably reduce water quality in the modeled portions of Lake Sinclair. The
reduced water quality is primarily attributed to arsenic, cadmium, and thallium.
The baseline modeled pollutant concentrations exceed human health criteria primarily for
arsenic and thallium, as discussed below:
• Arsenic concentrations exceed the water quality benchmark for consumption of water
and organisms (0.018 |ig/L):
In the immediate receiving water for the entire modeling period.
In all modeled segments of the receiving branch for more than 99 percent of the
modeling period.
In 97 percent of the entire modeling area for 10 percent or more of the modeling
period.
8-55
-------
Section 8—Case Study Modeling
• Arsenic concentrations also exceed the higher water quality benchmark for
consumption of organisms (0.14 |ig/L):
In five of the six modeled segments of the receiving branch for up to 19 percent of
the modeling period.
In 54 percent of the entire modeling area for 10 percent or more of the modeling
period.
• Thallium concentrations exceed the water quality benchmark for consumption of
water and organisms (0.24 |ig/L):
In three of the six modeled segments of the receiving branch for up to 6 percent of
the modeling period.
In 14 percent of the entire modeling area for 10 percent or more of the modeling
period.
• Thallium concentrations also exceed the higher water quality benchmark for
consumption of organisms (0.47 |ig/L):
In two of the six modeled segments of the receiving branch for less than 1 percent
of the modeling period.
In 11 percent of the entire modeling area for 10 percent or more of the modeling
period.
The case study modeling results indicate that, under baseline conditions, humans
consuming water and/or organisms that inhabit these modeled portions of Lake Sinclair may be
more at risk of the negative effects associated with oral exposure to arsenic and thallium (see
Section 3.1.1).
Aquatic organisms may be at risk for exposure to cadmium under baseline conditions.
Specifically, cadmium concentrations exceed the freshwater aquatic life criteria for chronic
exposure (0.25 |ig/L) in 4 percent of the entire modeling area for 10 percent or more of the
modeling period. These case study modeling results indicate that, under baseline conditions,
aquatic organisms inhabiting these modeled portions of Lake Sinclair could be at an elevated risk
of the negative effects associated with oral exposure to cadmium (see Section 3.1.1).
Under baseline conditions, the modeled pollutant concentrations in Lake Sinclair
occasionally exceed the MCL drinking water benchmarks for two of the modeled pollutants
(arsenic and thallium), as discussed below:
• Arsenic concentrations exceed the MCL drinking water criteria (10 |ig/L) in less than
1 percent of the segments for 10 percent or more of the modeling period.
• Thallium concentrations exceed the MCL drinking water criteria (2 |ig/L) in 5 percent
of the segments for 10 percent or more of the modeling period.
Modeling results do not indicate any exceedances of NRWQC or MCL criteria for the
other modeled pollutants (copper, lead, nickel, selenium, and zinc). Appendix G of this report
includes figures that illustrate the average water column pollutant concentration output for the
8^56
-------
Section 8—Case Study Modeling
entire lake for arsenic, cadmium, and thallium. These figures also present the NRWQC and MCL
benchmarks for the pollutant and the steady-state water column pollutant concentrations
predicted by the IRW model.
The final rule modeling results show significantly decreased average concentrations of
two of the modeled pollutants (nickel and selenium) in the modeled portion of Lake Sinclair.
Case study modeling results for Lake Sinclair reveal the water quality improvements for arsenic
under the final rule. Specifically, arsenic exceedances of the human health NRWQC benchmark
for consumption of water and organisms reduce in frequency from the entire modeling period to
23 percent of the modeling period in the immediate receiving water and reduce from above 99
percent of the modeling period to as low as 23 percent of the modeling period in the receiving
branch. Additionally, slightly less (2 percent of the modeling area) of Lake Sinclair will exceed
this benchmark under the final rule. Arsenic exceedances of the higher human health NRWQC
benchmark for consumption of organisms also reduce throughout the entire lake as 12 percent
less of the modeling area exceed this benchmark for more than 10 percent of the modeling
period.
While the modeling results demonstrate continuing arsenic, cadmium, and thallium
exceedances of NRWQC and MCL benchmarks in the receiving reach and the entire modeling
area, the pollutant loading contributions to the lake would be reduced under the final rule (if
Plant Harllee Branch did not retire all generating units).
Modeling Results - Wildlife
For the analysis of wildlife impacts and improvements, EPA assumed that aquatic life
travel freely throughout Lake Sinclair and do not confine themselves within particular segments
of the lake. EPA calculated the average fish tissue concentrations of all segments within the Lake
Sinclair model (i.e., entire modeling area) for purposes of the wildlife assessment.
Based on the average pollutant concentrations in the water column under baseline
conditions, the modeled portion of Lake Sinclair does not exceed the concentrations that would
translate to NEHC exceedances and does not pose a risk to minks and eagles that consume
contaminated fish. Despite the modeling not being able to quantify any improvements to minks
and eagles under the final rule, the pollutant loading removals will decrease bioaccumulation of
toxic pollutants in the terrestrial food chains (if Plant Harllee Branch did not retire all generating
units).
The Lake Sinclair EFDC model is not set up to quantify the pollutant concentrations in
the benthic sediment; therefore, EPA was unable to assess whether pollutant concentrations in
the sediment exceeded CSCL benchmarks and pose a threat to benthic organisms.
Modeling Results - Human Health
For the analysis of human health impacts and improvements, EPA also assumed that fish
travel freely throughout Lake Sinclair and do not confine themselves within particular segments
of the lake. EPA calculated the average fish tissue concentrations of all segments within the Lake
Sinclair model (i.e., entire modeling area) for purposes of the human health assessment.
8-57
-------
Section 8—Case Study Modeling
Under baseline conditions, the average water column concentration of arsenic in Lake
Sinclair over the modeling period does not result in an estimated cancer risk greater than 1-in-a-
million for any of the national-scale cohorts.
Based on the average pollutant concentrations in the water column under baseline
conditions, thallium poses the greatest threat to cause non-cancer health effects in humans from
fish consumption. Average thallium concentrations in the water column of the entire Lake
Sinclair modeling area are greater than the concentrations that would translate to exceedance of
the reference doses for adult and children recreational and subsistence fishers (all national-scale
cohorts evaluated). Therefore, humans who consume thallium-contaminated fish inhabiting the
modeled area of Lake Sinclair may be at greater risk for developing the negative health effects
associated with these pollutants, which are discussed in Section 3.1.1.
While the modeling results continue to show thallium water concentrations that would
translate to exceedances of the non-cancer health effects reference dose, the final rule will reduce
thallium loading contributions from Plant Harllee Branch (if Plant Harllee Branch did not retire
all generating units).
Interpretation of Lake Sinclair Results
The case study modeling results indicate that the water quality impacts are greater in the
receiving branch (closest portion of the lake to the Plant Harllee Branch discharge) of Lake
Sinclair compared to the rest of the lake. EPA identified that the receiving branch of Lake
Sinclair also exhibited more quantifiable improvements (i.e.., reduced NRWQC and MCL
benchmark exceedances) under the final rule than the average of all Lake Sinclair model
segments. Despite the model not indicating any wildlife or human health impacts in Lake
Sinclair, the reduction of pollutant loadings under the final rule would lessen the contribution of
steam electric power plant discharges on the entire aquatic and terrestrial ecosystems.
8.3 COMPARISON OF CASE STUDY AND IRW MODELING RESULTS
In general, the case study modeling results from the six case study models support the
overall conclusions of the IRW model.
Case study modeling of smaller receiving waters, such as Black Creek and Lick Creek,
indicate that more severe water quality, wildlife, and human health impacts are occurring at
baseline conditions than the IRW model predicted. Since flow rates in small receiving waters
fluctuate significantly, the case study modeling demonstrates impacts that can occur during
periods when the flow is lower than the annual average used in the IRW model. During the
frequent periods of low flow in smaller rivers and streams, the case study modeling shows that
pollutant concentrations quickly climb to levels that will negatively affect fish, wildlife, and
humans. The Black Creek and Lick Creek case study model also suggests the potential for
additional improvements under the final rule than the IRW model predicts. Case study modeling
therefore indicates that small receiving waters with highly variable flow rates may benefit from
the final rule more than the IRW model results suggest.
The case study modeling also demonstrates that the impacts from steam electric power
plant discharges can propagate much further downstream than the immediate receiving water
8^58
-------
Section 8—Case Study Modeling
used in the IRW modeling. In four of the six case study models, results illustrate that the
pollutant loadings from steam electric power plant discharges of the evaluated wastestreams may
contribute to water quality impacts up to 95 miles downstream of the plant discharge. These
additional impacts, as well as additional improvements under the final rule, are not represented in
the IRW modeling results.
Additionally, case study modeling of smaller water bodies revealed that downstream
reaches may be heavily influenced by the sediment transport and exhibit much higher water
column concentrations than the immediate receiving water. In the Black Creek, Etowah River,
and White River results, "hot spots" with higher pollutant concentrations were observed and
posed a greater risk to humans, aquatic life, and terrestrial food chains than reaches closer to the
steam electric power plants.
EPA performed one case study model of a representative lentic receiving water to assess
the potential impact on similar lakes or reservoirs that receive steam electric power plant
discharges of the evaluated wastestreams. Case study modeling of Lake Sinclair showed that
impacts are occurring in the lake, and these are more severe in the immediate area of the steam
electric discharge as compared to the lake average. The water quality improvements
demonstrated by the reduced exceedances of water quality benchmarks indicate that other lentic
receiving waters may also exhibit similar improvements. Although the case study modeling of
Lake Sinclair was unable to quantify the accumulation of pollutant concentrations in benthic
sediment, lower concentrations of pollutants under the final rule should reduce pollutant long-
term accumulation and consequential resuspension.
Each of the case study models demonstrated at least one exceedance of a water quality,
wildlife, or human health benchmark for a modeled pollutant discharged from stream electric
power plants. Under the final rule, the steam electric power plant(s) will contribute a reduced
loading of the pollutant(s), thereby improving water quality in these receiving waters. As
demonstrated by the Black Creek, Etowah River, Lick Creek and White River, Ohio River, and
Lake Sinclair case study modeling results, pollutant removals will result in quantifiable
improvements through reduced exceedances of environmental benchmarks.
8-59
-------
Section 9—Conclusions
SECTION 9
CONCLUSIONS
Based on evidence in the literature, damage cases, other documented impacts, and
modeled receiving water pollutant concentrations, it is clear that current wastewater discharge
practices at steam electric power plants are impacting the surrounding aquatic and terrestrial
environments and pose a human health threat to nearby communities. EPA estimates that
discharges from steam electric power plants contribute over one-third of the toxic-weighted
pollutant loadings of the combined discharges of all industrial categories currently required to
report discharges to U.S. waters. These discharges add large quantities of toxic bioaccumulative
pollutants (e.g., selenium, arsenic, and mercury) to the aquatic environment. Substantial evidence
exists that pollutants from steam electric power plant wastewater discharges are transferring from
the aquatic environment to terrestrial food webs; this indicates the potential for broader impacts
to ecological systems by altering population diversity and community dynamics in the areas
surrounding steam electric power plants. Ecosystem recovery from exposure to steam electric
power plant wastewater discharges can be extremely slow and even short periods of exposure
(e.g., less than a year) can cause observable ecological impacts that last for years. The strong
bioaccumulative properties and long residence times of pollutants in immediate receiving waters
reinforce the threat of these wastes to the local environment, and many of the impacts may not be
fully realized for years to come.
In addition, EPA's modeling demonstrates that pollutant loadings from discharges of the
evaluated wastestreams are impacting areas beyond the immediate receiving waters and pose a
threat to wildlife and human populations in thousands of river-miles downstream from steam
electric power plants under current discharge practices. Furthermore, EPA predicts that the
recently promulgated Clean Air Act requirements (i.e., Clean Power Plan) and other state and
local regulations may lead to additional air pollution controls (and resulting wastestreams) that
will increase the pollutant loadings to surface waters in the future. These additional pollutant
loadings above current baseline conditions will increase the number of immediate receiving
waters exceeding water quality, wildlife, and human health benchmarks in the future.65
Steam electric power plants discharge wastewater into waterbodies used for recreation,
and these discharges can present a potential threat to human health. Documented fish kills have
resulted in states issuing fish advisories to protect the public from exposure to fish with elevated
pollutant concentrations in recreational waters that receive these discharges. Combustion residual
leachate from surface impoundments and landfills is known to impact off-site ground water and
drinking water wells at concentrations above Maximum contaminant level (MCL) drinking water
standards and pose a potential threat to human health.
65 The analyses presented in this report incorporate some adjustments to current conditions in the industry. For
example, these analyses account for publicly announced plans from the steam electric power generating industry to
retire or modify steam electric generating units at specific power plants. These analyses also account for changes to
the industry that are expected to occur as a result of the recent Coal Combustion Residuals (CCR) rulemaking by
EPA's Office of Solid Waste and Emergency Response (OSWER). These analyses, however, do not reflect changes
in the industry that may occur as a result of the proposed Clean Power Plan [Clean Air Act section 111 (d)].
9-1
-------
Section 9—Conclusions
The final steam electric effluent limitations guidelines and standards (ELGs) will result in
quantifiable improvements in ecological and human health by reducing immediate receiving
water pollutant concentrations, on average, by 57 percent.66 The final rule will result in the
following environmental improvements as estimated by the national-scale immediate receiving
water (IRW) model:
• A 51 to 67 percent reduction in the number of immediate receiving waters exceeding
National Recommended Water Quality Criteria (NRWQC) for the protection of
aquatic life.
• A 45 to 50 percent reduction in the number of immediate receiving waters exceeding
an NRWQC for the protection of human health.
• A 63 to 64 percent reduction in the number of immediate receiving waters that
support fish whose tissue pollutant concentrations exceed benchmarks for the
protection of piscivorous wildlife (represented by minks and eagles).
• A 61 to 67 percent reduction in the number of immediate receiving waters where
selenium contamination in the food web presents reproductive risks67 to aquatic
wildlife (represented by fish and mallards).
• A 56 to 75 percent reduction in the number of immediate receiving waters that
support fish whose tissue pollutant concentrations pose a cancer risk to exposed
populations.
• A 52 to 56 percent reduction in the number of immediate receiving waters that
support fish whose tissue pollutant concentrations pose a risk of non-cancer health
effects in exposed populations.
The results of the case study modeling for selected plants and receiving waters indicate
that the environmental and human health impacts associated with steam electric power plant
discharges, and the corresponding improvements under the final rule, could be even more
extensive than those predicted by the IRW model. Case study modeling results demonstrate that
the impacts from steam electric power plant discharges of the evaluated wastestreams can
propagate much further downstream of the immediate receiving water. While the steam electric
power plant discharges may not cause these impacts in isolation, case study modeling reveals
that the discharges contribute to the further impairment of such waterways. Case study modeling
results identified a larger increase in baseline impacts and improvements under the final rule in
small receiving waters with variable flow than larger receiving waters. The analyses presented in
the environmental assessment (EA) focus on quantifying the environmental improvements within
rivers and lakes from post-compliance pollutant removals for metals, bioaccumulative pollutants,
and nutrients.
66 Reductions apply to the subset of pollutants evaluated in the environmental assessment (i.e., arsenic, cadmium,
chromium VI, copper, lead, mercury, nickel, selenium, thallium, and zinc).
67 For this statistic, reproductive risk is indicated by a 50-percent (or higher) probability that adverse reproductive
effects will occur in at least 10 percent of the exposed population offish and mallards.
-------
Section 9—Conclusions
While extensive, the environmental improvements quantified above do not encompass
the full range that will result from the final rule, such as the following improvements that are not
quantified (or have only limited analysis) in this EA:
• Reducing the loadings of bioaccumulative pollutants to the broader ecosystem,
decreasing long-term exposures and sublethal ecological effects.
• Reducing sublethal chronic effects of toxic pollutants on aquatic life not captured by
theNRWQC.
• Reducing loadings of pollutants for which EPA did not perform water quality
modeling in support of the EA (e.g., boron, manganese, aluminum, vanadium, and
iron).
• Mitigating impacts to aquatic and aquatic-dependent wildlife population diversity and
community structures.68
• Reducing wildlife exposure to pollutants through direct contact with combustion
residual impoundments and constructed wetlands built as treatment systems at steam
electric power plants.
• Reducing water withdrawals from surface waters and aquifers, leading to greater
availability of groundwater supplies for alternative uses and reducing fish
impingement and entrainment mortality due to surface water intake structures.
• Reducing the potential of harmful algal blooms to form.
Data limitations prevented EPA from
appropriately modeling the scale and
complexity of the ecosystem processes
potentially impacted by steam electric power
plant wastewater and therefore did not fully
quantify the improvements listed above.
However, damage cases and other documented
impacts in the literature reinforce that these
impacts are common in the environments
surrounding steam electric power plants and
fully support the conclusion that pollutant
removals will improve overall environmental
and wildlife health.
As surface impoundments accumulate fly ash,
bottom ash and flue gas de sulfur ization
sludges, they can begin to fill up and lose their
treatment capability.
Although the EA quantifies some
impacts to wildlife that consume fish
contaminated with pollutants from steam
electric power plant wastewater, it does not
capture the full range of exposure pathways through which bioaccumulative pollutants can enter
the surrounding food web. Wildlife can encounter bioaccumulative pollutants from steam electric
power plant wastewater discharges through direct exposure, drinking water, consuming
68 EPA did evaluate impacts to aquatic and aquatic-dependent wildlife from selenium contamination as part of the
ecological risk modeling. EPA did not quantify impacts that might occur due to other pollutant contamination.
9-3
-------
Section 9—Conclusions
contaminated vegetation, and consuming contaminated prey other than fish. Therefore, the
quantified improvements underestimate the complete loadings of bioaccumulative pollutants that
can impact wildlife in the ecosystem. EPA did quantify improvements to aquatic and aquatic-
dependent wildlife due to reduced selenium exposure via the food web. The reduced selenium
loadings under the final rule will significantly reduce the risk of negative reproductive effects to
wildlife in waterbodies that receive discharges from steam electric power plants. In addition to
the improvements resulting from reduced selenium loadings, EPA estimates that the post-
compliance pollutant removals under the final rule will lower the total amount of
bioaccumulative pollutants entering the food web in immediate receiving waters and downstream
waters.
EPA estimates that pollutant removals will also decrease sublethal effects associated with
many of the pollutants in steam electric power plant wastewater that may not be captured by
comparisons with NRWQC for aquatic life. Well-documented studies suggest that organisms in
aquatic environments near steam electric power plants exhibit chronic effects such as changes in
metabolic rates, decreased growth rates, changes in morphology (e.g., fin erosion, oral
deformities), and changes in behavior (e.g., decreased ability to swim, catch prey, or escape from
predators) that can negatively affect long-term survival [Raimondo et al, 1998; Rowe et al,
1996, 2002]. However, these effects are not fully quantified in the EA due to data limitations,
and therefore improvements to wildlife health and survival from the final rule may be
underestimated. Reduced organism survival rates from chronic effects such as abnormalities can
alter interspecies relationships (e.g., declines in the abundance or quality of prey) and prolong
ecosystem recovery. EPA was unable to quantify changes to aquatic and wildlife population
diversity and community dynamics; however, population effects (i.e., decline in number and type
of organisms present) attributed to exposure to steam electric power plant wastewater are well
documented in the literature [Lemly, 1985a; Garrett and Inman, 1984; Sorensen et al, 1982].
Changes in aquatic populations can alter the structure of aquatic communities and cause
cascading effects within the food web that have long-term impacts to ecosystem dynamics. EPA
estimates that post-compliance pollutant removals associated with the final rule will lower the
stressors that can alter population and community dynamics and will improve the overall
function of ecosystems surrounding steam electric power plants.
9-4
-------
Section 10—References
SECTION 10
REFERENCES
Abt. 2008. Abt Associates, Inc. "Revised Draft - Updating the Boron TWF." (December). DCN
SE04467.
Andersen LJ. and B. Madsen. 1983. Use of "The Capillary Barrier" as a Shield against
Groundwater Pollution from Fly-Ash Deposits. Water Science Technology, 15:207-212 (As
cited in Prasad et a/., 1996).
ATSDR. 1998a. Agency for Toxic Substances and Disease Registry. Health Consultations:
Martin Creek Lake. Department of Health and Human Services. Atlanta, GA. (September).
DCN SEO1877.
ATSDR. 1998b. Health Consultations: Brandy Branch Reservoir. Department of Health and
Human Services. Atlanta, GA. (September). DCN SE01878.
ATSDR. 1998c. Health Consultations: Welsh Reservoir. Department of Health and Human
Services. Atlanta, GA. (September). DCN SEO 1880.
ATSDR. 1999. Mercury Fact Sheet. Department of Health and Human Services. Atlanta, GA.
Available online at: http://www.atsdr.cdc.gov/toxfaqs/tf.asp?id=113&tid=24. DCN SE1990.
ATSDR. 2008a. Cadmium Fact Sheet. Department of Health and Human Services. Atlanta, GA
(September). DCN SE01898.
ATSDR. 2008b. Toxicological Profile for Manganese. Department of Health and Human
Services. Atlanta, GA. (September). DCN SEO 1899.
ATSDR. 2010a. Minimal Risk Levels (MRLs). Department of Health and Human Services.
Atlanta, GA. (December). Available online at:
http://www.atsdr.cdc.gov/mrls/pdfs/atsdr mrls december 2010.pdf. DCN SEO 1901.
ATSDR. 2010b. Toxicological Profile for Boron. Department of Health and Human Services.
Atlanta, GA. (November). DCN SEO 1902.
ATSDR. 2012. Toxicological Profile for Cadmium. Department of Health and Human Services
Atlanta, GA. (September). DCN SE04468.
Baba, A. and A. Kaya. 2004. Leaching characteristics of solid wastes from thermal power plants
of western Turkey and comparison of toxicity methodologies. Journal of Environmental
Management, 73:199-207. DCN SEO 1866.
Ballard, A.M. 2012. "Water Pollution Groundwater Violations Found Near All 14 North
Carolina Power Plants With Ash Ponds." BNA Today. (January 26). DCN SEO 1896.
Barrows, M.E., S.R. Petrocelli, K.J. Macek, and J.J. Carroll. 1980. Chapter 24: Bioconcentration
and Elimination of Selected Water Pollutants by Bluegill Sunfish (Lepomis macrochirus).
Dynamics, Exposure and Hazard Assessment of Toxic Chemicals. Edited by R. Haque. Ann
Arbor, MI: Ann Arbor Science. DCN SEO 1897.
Baumann, P.C. and R.B. Gillespie. 1986. Selenium bioaccumulation in gonads of largemouth
bass and bluegill from three power plant cooling reservoirs. Environmental Toxicology and
Chemistry, 5:695-701. DCN SE01846.
10-1
-------
Section 10—References
Bengtsson D., A. Avril, G. Gunnarsson, J. Elmberg, P. Soderquist, G. Norevik, C. Tolf, K. Safi,
W. Fiedler, M. Wikelski, B. Olsen, and J. Waldenstrom. 2014. Movements, home range size
and habitat selection of mallards during autumn migration. PLOS ONE, 9(6):el 00764.
doi:10.1371/journal.pone.0100764. DCN SE04616.
Benson, W.H., and WJ. Birge. 1985. Heavy metal tolerance in metallothionein induction in
fathead minnows: Results from field and laboratory investigations. Environmental
Toxicology and Chemistry, 4(2):209-217. DCN SE01849.
Besser, J.M., TJ. Canfield, and T.W. La Point. 1993. Bioaccumulation of organic and inorganic
selenium in a laboratory food chain. Environmental Toxicology and Chemistry. 12(l):57-72.
DCNSE06510.
Besser, J. M., J.P. Giesy, R.W. Brown, J.M. Buell, and G.A. Dawson, 1996. Selenium
bioaccumulation and hazards in a fish community affected by coal fly ash effluent.
Ecotoxicology and Environmental Safety, 35: 7-15. DCN SE02139.
Birge, WJ. 1978. Aquatic Toxicology of Trace Elements of Coal and Fly Ash. Energy and
Environmental Stress in Aquatic Systems, Selected Papers from a Symposium, held at
Augusta, Georgia November 2-4, 1977. DCN SE01854.
Bryan, A.L., Jr, W.A. Hopkins, J.A. Baionno, and B.P. Jackson. 2003. Maternal transfer of
contaminants to eggs in common grackles (Quiscalus quiscala) nesting on coal fly ash basins.
Archives of Environmental Contamination and Toxicology, 45(2):273-277. DCN SE01836.
Burger, J., K.F. Gaines, C.G. Lord, I.L. Brisbin, S. Shukla, and M. Gochfield. 2002. Metal levels
in raccoon tissues: differences on and off the Department of Energy's Savannah River site in
South Carolina. Environmental Monitoring and Assessment, 74:67-84. DCN SE01869.
Cantor, K.P., C.M. Villanueva, C.T. Silverman, J.D. Figueroa, F. X. Real, M. Garcia-Closas, N.
Malats, S. Chanock, M. Yeager, A. Tardon, R. Garcia-Closas, C. Serra., A. Carrato, G.
Castano-Vinyals, C. Samanic, N. Rothman, and M. Kogevinas. 2010. Polymorphisms in
GSTT1, GSTZ1, and CYP2E1, disinfection by-products, and risk of bladder cancer in Spain.
Environmental Health Perspectives, 118(11):1545-1550. DCN SE1981.
Carlson, C.L. and D.C. Adriano. 1993. Environmental impacts of coal combustion residues.
Journal of Environmental Quality, 22:227-247. DCN SE02144.
Chapman, P., W. Adams, M. Brooks, C. Delos, S. Luoma, W. Maher, H. Ohlendorf, T. Presser,
and D. Shaw, eds. 2009. Pellston Workshop on Ecological Assessment of Selenium in the
Aquatic Environment. PensacolaFL (USA): Society of Environmental Toxicology and
Chemistry (SETAC). DCN SE04519.
Charlotte Observer, 2010. "Coal Ash in Water Worries Officials." (October). DCN SE01879.
Cherry, D.S., R.J. Currie, and D.J. Soucek. 2000. Review of the Global Adverse Environmental
Impacts to Ground Water and Aquatic Ecosystem from Coal Combustion Wastes. Hoosier
Environmental Council and Citizens Coal Council. Indianapolis, Indiana. (March). DCN
SE01884.
Chesapeake Bay Program. 2015a. The Chesapeake Bay Watershed (website).
http://www.chesapeakebav.net/discover/baywatershed. DCN SE06506.
Chesapeake Bay Program. 2015b. Facts & Figures (website).
http://www.chesapeakebay.net/discover/bay 10I/facts. DCN SE06507.
10-2
-------
Section 10—References
Clark, D.R. 1987. Selenium accumulation in mammals exposed to contaminated California
irrigation drainwater. The Science of the Total Environment, 66:147-168. DCN SE01874.
Coughlan, DJ. and J.S. Velte. 1989. Dietary toxicity of selenium-contaminated red shiners to
striped bass. Transactions of the American Fisheries Society, 118:400-408. DCN SE01850.
Cowman, G. A. and P. C. Singer. 1996. Effect of bromide ion on haloacetic acid speciation
resulting from chlorination and chloramination of aquatic humic substances. Environmental
Science & Technology 30(1): 16-24. (As cited in VanBriesen, 2013).
Cross, F.L. 1981. Coal pile environmental impact problems. Pollution Engineering, 13(7):35-37.
DCN SEO1881.
Crutchfield, J. and S. Ferguson. 2000a. Predicting recovery of a fish population after heavy metal
impacts. Environmental Science & Policy, 3:183-189. DCN SE01838.
Crutchfield, J.U., Jr. 2000b. Recovery of a power plant cooling reservoir from selenium
bioaccumulation. Environmental Science & Policy, 3:145-163. DCN SE02138.
Cumbie, P.M. and S.L. Van Horn. 1978. Selenium Accumulation Associated with Fish Mortality
and Reproductive Failure. Proceedings of Annual Conference of the Southeastern
Association of Fish and Wildlife, 32:612-624. DCN SE01851.
CUWA. 2011. California Urban Water Agencies. Drinking Water Treatment Evaluation Project
Report. Malcolm Pirnie, Inc. (April). DCN SE06509.
Duke Energy. 2007. "Gibson Lake Will Not Reopen for Fishing April 1." Press Release. EPA-
HQ-OW-2008-0517-0427.
EPRI. 1997. Electric Power Research Institute. Coal Combustion By-products and Low-Volume
Wastes Comanagement Survey, EPRI TR-108369. Palo Alto, CA (as cited in Rowe et al.,
2002).
EIP (Environmental Integrity Project), Earthjustice, and Sierra Club. 2010a. Out of Control:
Mounting Damages from Coal Ash Waste Sites. (February). DCN SE00351.
EIP, Earthjustice, and Sierra Club. 2010b. In Harm's Way: Lack of Federal Coal Ash
Regulations Endangers Americans and their Environment. (August). DCN SEO 1365.
ERG. 2010. Eastern Research Group, Inc. "Drinking Water Treatment Technologies that Can
Reduce Metal and Selenium Concentrations Associated with Discharges from Steam Electric
Power Plants." DCN SE02154.
ERG. 2013a. "Literature Review - Impacts of Coal Pile Runoff and Leachate." DCN SE02153.
ERG. 2013b. Literature Search Database. DCN SE01962.
ERG. 2015a. "Loadings Analysis for the Steam Electric ELGs Environmental Assessment."
(September). DCN SE04446.
ERG. 2015b. "Review of Toxic Weighting Factors in Support of the Final Steam Electric
Effluent Limitations Guidelines and Standards." (September). DCN SE04479.
ERG. 2015c. "Proximity Analysis Methodology." (September). DCN SE04448.
ERG. 2015d. "Supporting Documentation for Summarization of Environmental Assessment
Analyses." (September). DCN SE04449.
10-3
-------
Section 10—References
ERG. 2015e. "Water Quality Module: Plant and Receiving Water Characteristics." (September).
DCNSE04513.
ERG. 2015f. "Pollutant Loadings Supporting Documentation." (September). DCN SE04515.
ERG. 2015g. "Fish Consumption Rates used in the Environmental Assessment Human Health
Module." (September). DCN SE04482.
ERG. 2015h. "Water Quality Module Supporting Documentation." (September). DCN SE04451.
ERG. 20151. "Wildlife and Human Health Modules Supporting Documentation." (September).
DCN SE04452.
ERG. 2015J. Steam Electric Technical Questionnaire Database ("Steam Electric Survey").
(September). DCN SE05903.
ERG. 2015k. "EA Model Validation and Calibration." (September). DCN SE04454.
ERG. 20151. "Downstream Modeling Methodology and Supporting Documentation."
(September). DCN SE04455.
ERG. 2015m. Damage Cases and Other Site Impacts. (September). DCN SE04518.
ERG. 2015n. "Technical Approach for Case Study Water Quality Modeling of Aquatic Systems
in Support of the Final Steam Electric Power Generating Industry Environmental
Assessment." (September). DCN SE05570.
ERG. 2015o. EA Loadings versus TDD Loadings Sensitivity Analysis Spreadsheet. (September).
DCNSE04539A1.
Etowah Aquatic Habitat Conservation Plan. 2015. Available online at:
http://www.etowahhcp.org. DCN SE05608.
Evers, D.C., J.G. Wiener, C.T. Driscoll, D.A., Gay, N. Basu, B.A. Monson, K.F. Lambert, H.A.
Morrison, J.T. Morgan, K.A. Williams, and A.G. Soehl. 2011. Great Lakes Mercury
Connections: The Extent and Effects of Mercury Pollution in the Great Lakes Region.
Biodiversity Research Institute. Gorham, Maine. Report BRI 2011-18. DCN SE01904.
Fan et al. 2002. T.W. Fan, SJ. The, D.E. Hinton, R.M. Higashi. Selenium biotransformations
into proteinaceous forms by foodweb organisms of selenium-laden drainage waters in
California. Aquat.Toxicol, 57(l-2):65-84. (April). DCN SE04618.
Fiske et al. 2011. P. S. Fiske, J. Oppenheimer, et al. In tank aeration predicts and reduces
THMs. Opflow: 22-24. (As cited in VanBriesen, 2013).
Fleming, L.N., H.N. Abineth, and H.I. Inyang, 1996. Leachate pH effects on the teachability of
metals from coal fly ash. Journal of Soil Contamination, 5(l):53-59 (as cited in Baba and
Kaya, 2004).
Garrett, G.P. and C.R. Inman. 1984. Selenium-induced changes in fish populations in a heated
reservoir. Proc. Ann. Conf. Southeast Assoc. Fish. Wildl. Agencies, 38:291-301. DCN
SE01885.
Georgia Power. 2014. Data sheet: Plant Bowen. Southern Company. Cartersville, GA. Available
online at: http://www.georgiapower.com/docs/about-us/1400756-.pdf DCN SE05609.
10-4
-------
Section 10—References
Gillespie, R.B. and P.C. Baumann. 1986. Effects of high tissue concentrations of selenium on
reproduction by bluegills. Transactions of the American Fisheries Society, 115:208-213.
DCNSE01852.
Great Lakes Restoration Initiative. 2010. Great Lakes Restoration Initiative Action Plan.
(February 21). DCN SE06427.
Green, AJ. and J. Elmberg. 2014. Ecosystem services provided by waterbirds. Biological
Reviews, 89(1):105-122. (February) DCN SE04619.
Guthrie, R.K. and Cherry, D.S. 1976. Pollutant removal from coal ash basin effluent. Journal of
the American Water Resources Association, 12(5):889-902. DCN SE01853.
Handke, P. 2009. Trihalomethane speciation and the relationship to elevated total dissolved solid
concentrations affecting drinking water quality at systems utilizing the Monongahela River
as a primary source during the 3rd and 4th quarters of 2008. Pennsylvania Department of
Environmental Protection Bureau of Water Standards and Facility Regulation. (As cited in
VanBriesen, 2013).
Heinz, G.H., DJ. Hoffman, AJ. Krynitsky, and D.M.G. Weller. 1987. Reproduction in mallards
fed selenium. Environmental Toxicology and Chemistry, 6:423-433. DCN SE04620.
Heinz, G.H., DJ. Hoffman, and L.G. Gold. 1989. Impaired reproduction of mallards fed an
organic form of selenium. Journal of Wildlife Management, 53:418-428. DCN SE04621.
Heinz, G.H. and DJ Hoffman. 1996. Comparison of the effects of seleno-L-methionine, seleno-
DLmethionine, and selenized yeast on reproduction of mallards. Environmental Pollution,
91:169-175. (As cited in Ohlendorf, 2003).
Heinz, G.H. andDJ. Hoffman. 1998. Methylmercury chloride and selenomethionine interactions
on health and reproduction in mallards. Environmental Toxicology and Chemistry, 17:139-
145.DCNSE04623.
Heller-Grossman. 1993. L. Heller-Grossman, L., J. Manka, et al. Formation and distribution of
haloacetic acids, THM, and TOX in chlorination of bromide-rich lake water. Water Research
27(8): 1323-1331. (As cited in VanBriesen, 2013).
Hofer, C., FJ. Gallagher, and C. Holzapfel. 2010. Metal accumulation and performance of
nestlings of passerine bird species at an urban brownfield site. Environmental Pollution,
158:1207-1213. DCN SE01867.
Hopkins, W.A., M.T. Mendon9a, and J.D. Congdon. 1997. Increased circulating levels of
testosterone and corticosterone in southern toads, Bufo terrestris, exposed to coal combustion
waste. General and Comparative Endocrinology, 108:237-246. DCN SE02146.
Hopkins, W.A., M.T. Mendon9a, C.L. Rowe, and J.D. Congdon. 1998. Elevated trace element
concentrations in southern toads, Bufo terrestris, exposed to coal combustion waste. Archives
of Environmental Contamination and Toxicology, 35:325-329. DCN SE02140.
Hopkins, W.A., J.W. Snodgrass, J.H. Roe, B.P. Jackson, J.C. Gariboldi, and J.D. Congdon. 2000.
Detrimental effects associated with trace element uptake in lake chubsuckers Erimyzon
sucetta exposed to polluted sediments. Archives of Environmental Contamination and
Toxicology, 39:193-199. DCN SE01841.
10-5
-------
Section 10—References
Hopkins, W.A., J.W. Snodgrass, J.H. Roe, B.P. Staub, B.P. Jackson, and J.D. Congdon. 2001.
Effects of food ration on survival and sublethal responses of lake chubsuckers (Erimyzon
sucetta) exposed to coal combustion wastes. Aquatic Toxicology, 21(5):906-913. DCN
SE02141.
Hopkins, W.A., J.H. Roe, J.W. Snodgrass, B.P. Staub, B.P. Jackson, and J.D. Congdon. 2002.
Effects of chronic dietary exposure to trace elements on banded water snakes (Nerodia
fasciata). Environmental Toxicology and Chemistry, 21(5):906-13. DCN SE01855.
Hopkins, W.A., J.W. Snodgrass, B.P. Staub, B.P. Jackson, and J.D. Congdon. 2003. Altered
swimming performance of benthic fish (Erimyzon sucetta) exposed to contaminated
sediments. Archives of Environmental Contamination and Toxicology., 44:383-389. DCN
SE01856.
Hopkins, W.A., S.E. Durant, B.P. Staub, C.L. Rowe, and B.P. Jackson. 2006. Reproduction,
embryonic development, and maternal transfer of contaminants in the amphibian
Gastrophryne carolinensis. Environmental Health Perspectives, 114(5):661-666. DCN
SE01839.
Janz, D.M., O.K. DeForest, M.L. Brooks, P.M. Chapman, G. Gilron, D. Hoff, W.A. Hopkins,
D.O. Mclntyre, C.A. Mebane, V.P. Palace, J.P. Skorupa and M.Wayland. 2010. Selenium
Toxicity to Aquatic Organisms (As cited in Chapman, et.al, 2009).
Jobson, H. E., 1996. Prediction of Traveltime and Longitudinal Dispersion in Rivers and
Streams., U.S. Geological Survey Water Resources Investigations Report 96-4013. (As cited
in the NHD Plus User's Guide).
King, K.A., T.W. Custer, and D.A. Weaver. 1994. Reproductive success of barn swallows
nesting near a selenium-contaminated lake in East Texas. U.S. Environmental Pollution,
84:53-58. DCN SE01870.
Kopsick, D.A. and E.E. Angino. 1981. Effect of leachate solutions from fly and bottom ash on
groundwater quality. Journal of Hydrology, 54:341-356. DCN SE01865.
Krasner et al., 1989. S. W. Krasner, M. J. McGuire, et al. The occurrence of disinfection by-
products in united-states drinking-water. Journal American Water Works Association 81(8):
41-53. (As cited in VanBriesen, 2013).
Kumada, H., S. Kimura, M. Yokote, and Y. Matida. 1972. Acute and chronic toxicity, uptake,
and retention of cadmium in freshwater organisms. Bulletin of Freshwater Fisheries Research
Laboratory, 22(2): 157-165. DCN SE01894.
Lemly, A.D. 1985a. Toxicology of selenium in a freshwater reservoir: implications for
environmental hazard evaluation and safety. Ecotoxicology and Environmental Safety,
10:314-338. DCN SE01842.
Lemly, A.D. 1985b. Ecological basis for regulating aquatic emissions from the power industry:
the case with selenium. Regulatory Toxicology and Pharmacology, 5:465-486. DCN
SE01887.
Lemly, A.D. 1993. Teratogenic effects of selenium in natural populations of freshwater fish.
Ecotoxicol. Environ. Safety, 26(2): 181-204. DCN SE04624.
Lemly, A.D., 1997a. Ecosystem recovery following selenium contamination in a freshwater
reservoir. Ecotoxicology and Environmental Safety, 36:275-281. DCN SE02138.
10-6
-------
Section 10—References
Lemly, A.D. 1997b. A teratogenic deformity index for evaluating impacts of selenium on fish
populations. Ecotoxicology and Environmental Safety, 37:259-266. DCN SE04625.
Lemly, A. D. 1999. Selenium transport and bioaccumulation in aquatic ecosystems: a proposal
for water quality criteria based on hydrological units. Ecotoxicology and Environmental
Safety. 42:150-156 (as cited in Rowe etal, 2002).
Lemly, A.D. 2002. Selenium Assessment in Aquatic Ecosystems. Springer-Verlag, New York.
(As cited in U.S. EPA, 2015b).
Liang, L. and P. Singer. 2003. Factors influencing the formation and relative distribution of
haloacetic acids and trihalomethanes in drinking water. Environmental & Science
Technology, 37:2920-2928. DCN SE01982.
Luoma, et.al. 1992. S.N. Luoma, C. Johns, N.S. Fisher, N.A. Steinberg, R.S. Oremland, J.R.
Reinfelder. Determination of Selenium Bioavailability to a Benthic Bivalve from Particulate
and Solute Pathways. Environ. Sci. Technol, 26:485-491. DCN SE04627.
MacDonald, D.D.; C. G. Ingersoll; and T. A. Berger. Development and evaluation of consensus-
based sediment quality guidelines for freshwater ecosystems. Archives of Environmental
Contamination and Toxicology 2000, 39(1)20 (as cited in NOAA, 2008).
Magnuson, J. J., A.M. Forbes, D.M. Harrell, J.D. Schwarzmeier. 1980. Responses of stream
invertebrates to an ashpit effluent Wisconsin power plant impact study. EPA/600/3-
80/081.U.S. EPA Office of Research and Development. August. DCN SE04628.
Maier, KJ. and A.W. Knight. 1994. Ecotoxicology of selenium in freshwater systems. Reviews
of Environmental Contamination and Toxicology, 134:31-48. (As cited in Chapman, et. al,
2009)
McTigue, N., K. Graf, and R. Brown. 2014. Occurrence and Consequences of Increased
Bromide in Drinking Water Sources. EE&T. DCN SE04503.
Metcalf and Eddy, Inc. 2003. Wastewater Engineering: Treatment and Reuse. McGraw-Hill, Inc.
4th edition. DCN SE01967.
MDE, 2008. Maryland Department of Environment. Constellation Reaches $54M Settlement on
Fly Ash Lawsuit. DCN SE02149.
MDNR. 2007. Maryland Department of Natural Resources. Assessment of Water Quality Impacts
Associated with Use of Coal Combustion Products as Structural Fill at the BBSS Site.
Annapolis, Maryland. DNR 12-6182007-221. (June). DCN SE02147.
MDNRE. 2010. Michigan Department of Natural Resources and Environment. Hazardous Waste
Management System, Identification and Listing of Special Wastes. Disposal of Coal
Combustion Residuals from Electric Utilities Docket Comments. EPA-HQ-RCRA-2009-
0640-6815. DCN SEO1895.
Murphy, B.R., G.J. Atchison, and A.W. Mclntosh. 1978. Cadmium and zinc in muscle of
bluegill (Lepomis macrochrius) and largemouth bass (Micropterus solmoides) from an
industrially contaminated lake. Environmental Pollution, 17(4):253-257 (as cited in U.S.
EPA, 201 Ob).
10-7
-------
Section 10—References
Nagle, R.D., C.L. Rowe, and J.D. Congdon. 2001. Accumulation of selective maternal transfer of
contaminants in the turtle Trachemys scripta associated with coal ash disposition. Archives
of Environmental Contamination and Toxicology, 40:531-536. DCN SE01843.
NCDENR, 2011. North Carolina Department of Environment and Natural Resources. Duke
Energy's Riverbend Steam Station - Groundwater Monitoring Results. (October) DCN
SE01271.
NOAA 2008. National Oceanic and Atmospheric Association. Screening Quick Reference
Tables (SQuiRTsj. DCN SE01905.
NFS. 1997. National Parks Service. Environmental Contaminants Encyclopedia Selenium Entry.
Fort Collins, Colorado. (July). DCN SE01906.
NRC. 2006. National Research Council of the National Academies. Managing Coal Combustion
Residues in Mines. National Academies Press, Washington, DC. DCN SE02136.
Ohlendorf, H.M., DJ. Hoffman, M.K. Saiki, and T.W. Aldrich. 1986. Embryonic mortality and
abnormalities of aquatic birds: apparent impacts of selenium from irrigation drainwater. The
Science of the Total Environment, 52:49-63. DCN SE01873.
Ohlendorf, H.M., A.W. Kilness, J.L. Simmons, R.K. Stroud, DJ. Hoffman, and J.F. Moore.
1988a. Selenium toxicosis in wild aquatic birds. Journal of Toxicology and Environmental
Health, 24:67-92. DCN SE01882.
Ohlendorf, H.M., R.L. Hothem, and T.W. Aldrich. 1988b. Bioaccumulation of selenium by
snakes and frogs in the San Joaquin Valley, California. Copeia, 3:704-710. DCN SE01883.
Ohlendorf, H.M., R.L. Hothem, and D. Welsh. 1989. Nest success, cause-specific nest failure,
and hatchability of aquatic birds at selenium-contaminated Kesterson Reservoir and a
reference site. The Condor, 91:787-796. DCN SE01872.
Ohlendorf, H.M., R.L. Hothem, C.M. Bunck, and K.C. Marois, 1990. Bioaccumulation of
selenium in birds at Kesterson Reservoir, California. Archives of Environmental
Contamination and Toxicology, 19:495-507. DCN SE01871.
Ohlendorf, H.M. 2003. Ecotoxicology of selenium. In Handbook ofEcotoxicology, 2nd., Eds.
DJ. Hoffman, B.A. Rattner, G.A. Burton Jr., and J.C. Cairns, Jr., pp.465-500. Lewis
Publishers, Boca Raton, FL. DCN SE04630.
Ohlendorf, H.M. and G.H. Heinz. 2011. Selenium in birds. In Environmental Contaminants in
Biota: Interpreting Tissue Concentrations. Eds. W.N. Beyer and J.P. Meador, pp. 669-701.
CRC Press. Boca Raton, FL: DCN SE04631.
Prasad B., N.N. Banjeree, and B.B. Dhar. 1996. Environmental assessment of coal ash disposal:
a review. Journal of Scientific & Industrial Research, 55:772-780. DCN SE01891.
Presser, T.S. and S.N. Luoma. 2010. A methodology for ecosystem-scale modeling of selenium.
Integrated Environmental Assessment and Management, 6(4):685-710. DCN SE04632.
Presser, T.S and H.M. Ohlendorf, 1987. Biogeochemical Cycling of Selenium in the San Joaquin
Valley, California, USA. Environmental Management. 11(6):805-821. (November). DCN
SE04633
10-8
-------
Section 10—References
Presser, et. al. 1994. T.S. Presser, M.A. Sylvester, and W.H. Low. Bioaccumulation of Selenium
from Natural Geologic Sources in Western States and its Potential Consequences.
Environmental Management. 18(3):423-436. DCN SE04634.
Pruitt, L., 2000. Indiana's First HCP Conserves Least Tern - Brief Article. Endangered Species
Bulletin. Accessed online at: FindArticles.com. DCN SE02148.
Raimondo, S.M., C.L. Rowe, and J.D. Congdon. 1998. Exposure to coal ash impacts swimming
performance and predator avoidance in larval bullfrogs. Journal of Herpetology, 32(2):289-
292. DCNSE01857.
Rattner, B.A., M.A. McKernan, K.M. Esereich, W.A. Link, G.H. Olsen, DJ. Hoffman, K.A.
Knowles, and P.C. McGowan. 2006. Toxicity and hazard of vanadium to mallard ducks
(Anas platyrhynchos) and Canada geese (Branta canadensis). Journal of Toxicology and
Environmental Health, Part A, 69:331-351. DCN SE01835.
Roe, J.H., W.A. Hopkins, and B.P. Jackson. 2005. Species- and stage-specific patterns of trace
element accumulation (tissue concentration) in amphibians exposed to coal-combustion
wastes. Environmental Pollution, 136:353-363. DCN SE01847.
Roe, J.H., W.A. Hopkins, S.E. Durant, and J.M. Unrine. 2006. Effects of competition and coal-
combustion wastes on recruitment and life history characteristics of salamanders in
temporary wetlands. Aquatic Toxicology, 79(2): 176-84. DCN SE01863.
Rosetta, T.N. and A.W. Knight. 1995. Bioaccumulation of selenate, selenite, and seleno-DL-
methionine by the brine fly larvae Ephydra cinerea Jones. Archives of Environmental
Contamination and Toxicology, 29(3):351-357. DCN SE04635.
Rowe, C.L., O.M. Kinney, and J.D. Congdon. 1996. Oral deformities in tadpoles (Rana
catesbeiana) associated with coal ash deposition: effects on grazing ability and growth.
Freshwater Biology, 36(3):723-730. DCN SE02142.
Rowe, C.L., O.M. Kinney, R.D. Nagle, and J.D. Congdon. 1998a. Elevated maintenance costs in
an anuran exposed to a mixture of trace elements during the embryonic and early larval
periods. Physiological Zoology, 71(l):27-35. DCN SE01845.
Rowe, C.L., O.M. Kinney, and J.D. Congdon. 1998b. Oral deformities in tadpoles of the bullfrog
(Rana catesbeiana) caused by conditions in a polluted habitat. Copeia, 1:244-246. DCN
SE01858.
Rowe, C.L., W.A. Hopkins, and V.R. Coffman. 2001. Failed recruitment of southern toads (Bufo
terrestris) in a trace element-contaminated breeding habitat. Archives of Environmental
Contamination and Toxicology, 40(3):399-405. DCN SE01859.
Rowe, C.L., W.A. Hopkins, and J.D. Congdon. 2002. Ecotoxicological implications of aquatic
disposal of coal combustion residues. Environmental Monitoring & Assessment, 80(3):207-
276. DCNSE02143.
Ruhl, L., A. Vengosh, G.S. Dwyer, H. Hsu-Kim, G. Schwartz, A. Romanski, and S.D. Smith.
2012. The impact of coal combustion residue effluent on water resources: a North Carolina
example. Environmental Science and Technology, 46:12226-12233. DCN SE01984.
Sager, D.R. and C.R. Colfield. 1984. Differential accumulation of selenium among axial muscle,
reproductive and liver tissues of four warmwater fish species. Journal of the American Water
Resources Association, 20(3):359-363. DCN SE01860.
10-9
-------
Section 10—References
Saiki and Lowe. 1987. M. K. Saiki and T. P. Lowe. Selenium in aquatic organisms from
subsurface agricultural drainage water, San Joaquin Valley, California. Archives of
Environmental Contamination and Toxicology. 16(6): 657 - 670. (November). DCN
SE04636.
Schultz, S. 2008. "Constellation Reaches $54M Settlement on Fly Ash Lawsuit." Baltimore
Business Journal. (December). DCN SE02149.
Simsiman, G.V., G. Chesters, and A.W. Andren. 1987. Effect of ash disposal ponds on
groundwater quality at a coal-fired power plant. Water Resources, 21(4):417-426. DCN
SE01868.
Snodgrass, J.W., W.A. Hopkins, J. Broughton, D. Gwinna, J.A. Baionno, and J. Burge. 2004.
Species-specific responses of developing anurans to coal combustion wastes. Aquatic
Toxicology, 66:171-182. DCN SE01844.
Sorensen, E.M.B., C.W. Harlan, and J.S. Bell. 1982. Renal changes in selenium-exposed fish.
Amer. J. Forensic Med. Pathol., 3:123-129. DCN SE01888.
Sorensen, E.M.B., J.S. Bell, and C.W. Harlan. 1983. Histopathological changes in selenium-
exposed fish. Amer. J. Forensic Med. Pathol., 4(2):111-123. DCN SE01862.
Sorensen, E.M.B. and T.L. Bauer. 1984a. A correlation between selenium accumulation in
sunfish and changes in condition factor and organ weight. Environmental Pollution Series A,
32:357-366. DCN SE01886.
Sorensen, E.M.B., P.M. Cumbie, T.L. Bauer, J.S. Bell, and C.W. Harlan. 1984b.
Histopathological, hematological, condition-factor, and organ weight changes associated
with selenium accumulation in fish from Belews Lake, North Carolina. Archives of
Environmental Contamination and Toxicology, 13:153-162. DCN SE01840.
Sorensen, E.M.B. 1988. Selenium accumulation, reproductive status, and histopathological
changes in environmentally exposed redear sunfish. Archives of Toxicology, 61:324-329.
DCN SEO1861.
Specht, W.L., D.L. Cherry, R.A. Lecleitner, and J. Cairns, Jr. 1984. Structural, functional, and
recovery responses of stream invertebrates to fly ash effluent. Canadian Journal of Fisheries
and Aquatic Sciences, 41(6):884-896. DCN SE01848.
Stanley, Jr., T.R., J.W. Spann, G.J. Smith, and R. Rosscoe. 1994. Main and interactive effects of
arsenic and selenium on mallard reproduction and duckling growth and survival. Archives of
Environmental Contamination and Toxicology, 26:444-451. DCN SE04637.
Stanley, Jr., T.R., G.J. Smith, D.J. Hoffman, G.H. Heinz, and R. Rosscoe. 1996. Effects of boron
and selenium on mallard reproduction and duckling growth and survival. Environmental
Toxicology and Chemistry, 15:1124-1132. DCN SE04638.
States, S. et al. 2013. Marcellus Shale drilling and brominated THMs in Pittsburgh, PA, drinking
water. Journal - American Water Works Association. DCN SE06508.
Stephan, C.E. 1993. Derivation of Proposed Human Health and Wildlife Bioaccumulation
Factors for the Great Lakes Initiative (Draft). PB93-154672. Environmental Research
Laboratory, Office of Research and Development, Duluth, MN. (March). DCN SEO 193 8.
10-10
-------
Section 10—References
Theis, T.L. and R.O. Richter. 1979. Chemical speciation of heavy metals in power plant ash
pond leachate. Environmental Science Technology, 13:219-224 (as cited in Simsiman et al.,
1987).
Theis, T.L. and K.H. Gardner. 1990. Environmental assessment of ash disposal. Critical Reviews
in Environmental Control, 20(l):21-24. (As cited in Prasad etal, 1996).
Thorneloe, S.A., D.S. Kosson, F. Sanchez, A.C. Garrabrants, and G. Helms. 2010. Evaluations of
the fate of metals in air pollution control residues from coal-fired power plants.
Environmental Science Technology, 44:7351-7356. DCN SE01889.
Tsipoura, N., J. Burger, R. Feltes, J. Yacabucci, D. Mizrahi, C. Jeitner, and M. Gochfeld. 2008.
Metal concentrations in three species of passerine birds breeding in the Hackensack
meadowlands of New Jersey. Environmental Research, 107:218-228. DCN SE01875.
U.S. DOE. 1992. U.S. Department of Energy. Ecological Effects of Contaminants inMcCoy
Branch, 1989-1990. Washington, DC. (January). DCN SE01837.
U.S. EPA. 1980. Ambient Water Quality Criteria for Copper (EPA-440-5-80-036). Washington,
DC. (October). DCN SE04487.
U.S. EPA. 1984. U.S. Environmental Protection Agency. Ambient Aquatic Life Water Quality
Criteria for Lead. (January). Available online at:
-------
Section 10—References
U.S. EPA. 2000b. Selenium Compounds (website). (January). Available online at:
http://www.epa.gov/ttn/atw/hlthef/selenium.html. DCN SE01923.
U.S. EPA. 2000c. Methodology for Deriving Ambient Water Quality Criteria for the Protection
of Human Health. EPA-822-B-00-004. Washington, DC. (October). DCN SE06429.
U.S. EPA. 2002. National Recommended Water Quality Criteria: 2002. EPA-822-R-02-047.
Washington DC. (November). DCN SE04484.
U.S. EPA. 2004. Draft Aquatic Life Water Quality Criteria for Selenium. EPA-822-D-04-001.
Washington, D.C. (November). DCN SE06511.
U.S. EPA. 2005a. Partition Coefficients for Metals in Surface Water, Soil, and Waste (Final).
EPA/600/R-05/074. Washington, DC. (July). DCN SE01892.
U.S. EPA. 2005b. Human Health Risk Assessment Protocol for Hazardous Waste Combustion
Facilities (HHRAP). EPA 520-R-05-006. Washington, DC. (September). DCN SE01926.
U.S. EPA. 2005c. Drinking Water Criteria Document for Brominated Trihalomethanes. EPA
822-R-05-011. Office of Water. Washington, DC. (November 15). DCN SE06443.
U.S. EPA. 2007a. Framework/or Metals Risk Assessment. EPA 120/R-07/001. Washington, DC.
(March). DCN SE01927.
U.S. EPA. 2008a. Child-Specific Exposure Factors Handbook. EPA/600/R-06/096F.
Washington, DC. (September). DCN SE01876.
U.S. EPA. 2008b. Lake Erie Lakewide Management Plan 2008. DCN SE01940.
U.S. EPA. 2009a. Toxicological Review of Thallium and Compounds. Integrated Risk
Information System. EPA/635/R-08/001F. Washington, DC. (September). DCN SE01907.
U.S. EPA. 2009b. Steam Electric Power Generating Point Source Category: Final Detailed
Study Report. EPA-821-R-09-008. Washington, DC. (October). DCN SE00003.
U. S. EPA. 2009c. Technical Support Document for the Preliminary 2010 Effluent Guidelines
Program Plan. EPA-821-R-09-006. Washington, DC. (October). DCN SE02134.
U.S. EPA. 2009d. National Recommended Water Quality Criteria. Washington, DC. DCN
SE01908.
U.S. EPA. 2009e. National Primary Drinking Water Regulations. EPA 816-F-09-004.
Washington, DC. (May). DCN SE01909.
U.S. EPA. 2010a. Provisional Peer-Review edToxicity Values for Thallium and Compounds.
National Center for Environmental Assessment. Superfund Health Risk Technical Support
Center, Cincinnati, OH. (September). DCN SE01941.
U.S. EPA. 2010b. Human and Ecological Risk Assessment of Coal Combustion Wastes (CCW)
(Draft). Washington, DC. (April). DCN SE01834.
U.S. EPA. 2010c. Discharge Monitoring Report (DMR) Pollutant Loading Tool: 2010 Data.
Available online at: http://cfpub.epa.gov/dmr/. DCN SE01959.
U.S. EPA. 2010d. Chesapeake Bay Total Maximum Daily Load for Nitrogen, Phosphorus and
Sediment. Washington, DC. (December). DCN SE01910.
10-12
-------
Section 10—References
U.S. EPA. 201 la. Technical Factsheet on: THALLIUM. National Primary Drinking Water
Regulations. Available online at:
http://water.epa.gov/drink/contaminants/basicinformation/historical/upload/Archived-
Technical-Fact-Sheet-on-Thallium.pdf. DCN SE01911.
U.S. EPA. 201 Ib. Exposure Factors Handbook: 2011 Edition. EPA/600/R-09/052F.
Washington, DC. (September). DCN SE01912.
U.S. EPA, 201 Ic. Integrated Risk Information System (IRIS). National Center for Environmental
Assessment. Washington, DC. Available online at: http://www.epa.gov/IRIS/. DCN
SE01960.
U.S. EPA. 201 Id. Technical Support Document for the 2010 Effluent Guidelines Program Plan.
EPA-821-R-09-006. Washington, DC. (October). DCN SE02135.
U.S. EPA. 20 lie. Office of Science & Technology Annual Report, Calendar Year 2011. EPA
820-R-12-006. Washington, DC. DCN SE01951.
U.S. EPA. 201 If. Ecological Toxicity Information (website). EPA Region 5. (December).
Available online at: http://www.epa.gov/R5Super/ecologv/toxprofiles.htm. DCN SE01913.
U.S. EPA. 2012a. "EPA Weighs Setting Possible First-Time Water Quality Criteria for
Bromide." Inside EPA. (January 4). DCN SE01949.
U.S. EPA. 2012b. Toxic Weighting Factors Methodology. EPA-820-R-12-005. Washington, DC.
(March). DCN SE04470.
U.S. EPA. 2012c. TWF Database (EPA-HQ-OW-2008-0517-0713). MS Excel™. Washington,
DC. DCN SE04465.
U.S. EPA. 2012d. Cyanobacteria and Cyanotoxins: Information for Drinking Water Systems.
EPA-810-F-11-001. Washington, DC. (July). DCN SE04466.
U.S. EPA. 2012e. Final Determination of Identified Proven Damage and Recently Alleged
Damage Cases. SE01966.
U.S. EPA. 2013a. DMR & TRI Loadings Database for Case Study Modeling. DCN SE05589.
U.S. EPA. 2013b. Impact Case Study Analysis Database. Washington, DC. DCN SE02156.
U.S. EPA. 2014a. CCR Damage Cases Database. Office of Solid Waste and Emergency
Response (OSWER)/Office of Resource Conservation and Recovery (ORCR). Washington,
DC. DCN SE04457.
U.S. EPA. 2014b. Damage Case Compendium Technical Support Document, Volume I: Proven
Damage Cases. OSWER/ORCR. Washington, DC. DCN SE04458.
U.S. EPA. 2014c. Damage Case Compendium Technical Support Document, Volume Ila:
Potential Damage Cases. OSWER/ORCR. Washington, DC. DCN SE04459.
U.S. EPA. 2014d. Damage Case Compendium Technical Support Document, Volume lib Part
One: Potential Damage Cases. OSWER/ORCR. Washington, DC. DCN SE04460.
U.S. EPA. 2014e. Damage Case Compendium Technical Support Document, Volume lib Part
Two: Potential Damage Cases. OSWER/ORCR. Washington, DC. DCN SE04461.
U.S. EPA. 2014f External Peer Review Draft Aquatic Life Ambient Water Quality Criterion for
Selenium - Freshwater. EPA 822-P-14-001. Washington, DC. (May). DCN SE06438.
10-13
-------
Section 10—References
U.S. EPA. 2014g. Human and Ecological Risk Assessment of Coal Combustion Residuals.
(December) DCN SE06442.
U.S. EPA. 2014h. National Fish Consumption Advisories NHD Indexed Dataset. Reach Address
Database (RAD). Extracted on July 7. Available online at: http://epamap32.epa.gov/radims/.
DCN SE04545.
U.S. EPA. 20141. National 303(d) Listed Impaired Waters National Hydrography Data (NHD)
Indexed Dataset. RAD. Extracted on August 4. Available online at:
http://www.epa.gov/waters/data/downloads.html. DCN SE04544.
U.S. EPA. 2015 a. Preventing Eutrophication: Scientific Support for Dual Nutrient Criteria. EPA
820-S-15-001. Washington, DC. (February). DCN SE04505.
U.S. EPA. 2015b. Draft Aquatic Life Ambient Water Quality Criterion for Selenium -
Freshwater. EPA 822-P-15-001. Washington, DC. (July). DCN SE06439.
U.S. EPA and Environment Canada. 1997. Great Lakes Binational Toxics Strategy. DCN
SE01942.
U.S. EPA and Environment Canada. 2009. State of the Great Lakes 2009. EPA 905-R-09-031.
DCN SE01939.
USFWS. 2008. U.S. Fish and Wildlife Service. "Steps Taken to Address Selenium Concerns at
Cane Ridge Wildlife Management Area." News Release. (June 13). DCN SE02150.
USFWS. 2012. "Selenium Threat Averted." News Release. (June 19). DCN SE01943.
USGS. 2008. U.S. Geological Survey. Environmental contaminants in freshwater fish and their
risk to piscivorous wildlife based on a national monitoring program. Environmental
Monitoring and Assessment, 152:469-494. DCN SE01893.
USGS. 2015. The Quality of the Nation's Groundwater. Barbara Mahler and Jon Campbell.
(January 21). DCN SE04639.
VanBriesen, Jeane M. 2013. Potential Drinking Water Effects of Bromide Discharges from Coal-
Fired Electric Power Plants. EPA-HQ-OW-2009-0819-4687-A10.
Vengosh, A. 2009. The Tennessee Valley Authority's Kingston Ash Slide: Potential Water
Quality Impacts of Coal Combustion Waste Storage. A Testimony to the Subcommittee on
Water Resources and Environment, U.S. House of Representatives. Available online at:
http://transportation.house.gov/Media/file/water/20090331/Vengosh%20Testimony.pdf.
DCN SEO1864.
Villanueva, C.M., K.P. Cantor, S. Cordier, J.J.K. Jaakkola, W.D. King, D.F. Lynch, S. Porru,
and M. Kogevinas. 2004. Disinfection byproducts and bladder cancer: a pooled analysis.
Epidemiology, 15(3):357-367. DCN SE01983.
Vogel, et. al, 1999. R.M. Vogel, I.W. Wilson, and C. Daly. Regional Regression Models of
Annual Streamflow for the United States. Journal of Irrigation and Drainage Engineering.
125 (3): 148-157. (As cited in the NHD Plus User's Guide).
Weinberg. 2002. H.S. Weinberg, S. W. Krasner, et al. The Occurrence of Disinfection By-
Products (DBFs) of Health Concern in Drinking Water: Results of a Nationwide DBF
Occurrence Study (EPA/600/R-02/068) Athens, GA. (As cited in VanBriesen, 2013).
10-14
-------
Section 10—References
WHO. 1976. World Health Organization. Environmental Health Criteria 1: Mercury.
International Programme on Chemical Safety. Geneva, Switzerland. DCN SE01914.
WHO. 1987. Environmental Health Criteria 58: Selenium. International Programme on Chemical
Safety. Geneva, Switzerland. DCN SE01915.
WHO. 1992. Environmental Health Criteria 135: Cadmium. International Programme on
Chemical Safety. Geneva, Switzerland. DCN SE01916.
WHO. 1996. Environmental Health Criteria 182: Thallium. International Programme on
Chemical Safety. Geneva, Switzerland. DCN SE01917.
WHO. 1998. Environmental Health Criteria 204: Boron. International Programme on Chemical
Safety. Geneva, Switzerland. DCN SE01918.
WHO. 2001. Arsenic in Drinking Water. Available online at:
http://www.who.int/mediacentre/factsheets/fs210/en/. (20 January). DCN SE01919.
WHO. 2009. Bromide in Drinking-Water. Background Document for Development of'WHO
Guidelines for Drinking-Water Quality. WHO/HSE/WSH/09.01/6. Geneva, Switzerland.
DCN SE01920.
WHO. 2011. Manganese in Drinking-Water. Background Document for Development of WHO
Guidelines for Drinking-Water Quality. WHO/HSE/WSH/03.04/104/Rev/l. Geneva,
Switzerland. DCN SE01921.
Wilson et al. 2013. J.M. Wilson, Y. Wang, et al. Sources of high total dissolved solids to
drinking water supply in Southwestern Pennsylvania. ASCE Journal of Environmental
Engineering. (As cited in VanBriesen, 2013).
10-15
-------
Appendix A— Literature Review Methodology and Results
APPENDIX A
LITERATURE REVIEW METHODOLOGY AND RESULTS
This appendix presents the methodology, resources, and summary results for the literature
review. The U.S. Environmental Protection Agency (EPA) used the keyword list in Table A-l to
identify peer-reviewed journal articles that document environmental and human health impacts
caused by steam electric power plant discharges of the evaluated wastestreams. The literature
search focused on information regarding impacts caused by pollutants of concern for the steam
electric power generating industry (e.g., toxic bioaccumulative pollutants such as mercury and
selenium, metals such as arsenic and lead, and nutrients) in the discharges. EPA also searched for
environmental assessments, impact studies, and related documents from state and federal
governments.
In addition, the literature search involved collecting information from newspapers,
environmental groups, industry organizations, and other non-peer-reviewed information sources.
These sources are considered to be "gray literature" and are not acceptable forms of formal
documentation of environmental impact events. However, these literature sources can provide
useful information for identifying potential areas of concern. Often, an environmental event is
reported in gray literature sources before it is well documented in peer-reviewed journals or
government reports. EPA used gray literature to help highlight areas of interest and facilitate
additional searches of peer-reviewed journals for more detailed information on the impacted area.
EPA used several different search engines to broaden the range of reference materials
represented in the results. The Agency searched the following search engines in the order
presented, using the keyword list in Table A-l:
• Scirus - A comprehensive science-specific search engine that provides access to a large
database of scientific, technical, and medical journals.
• Science Direct - An online library that features full text journals from Elsevier,
Academic Press, and other scholarly publishers.
• Ingenta - A scholarly research database that provides access to a large collection of
academic and professional research articles.
• Google Scholar - A search engine used to find other articles that cited previously
identified references as well as perform a general search of scholarly literature,
including peer-reviewed papers, theses, books, abstracts, and articles from academic
publishers, professional societies, preprint repositories, and universities and other
scholarly organizations.
• Google - A search engine used to perform a general search of information readily
available on the Internet.
A-l
-------
Appendix A— Literature Review Methodology and Results
Table A-l. Keyword Search Terms for Environmental Impacts from
Steam Electric Power Plants
Category
General Terms
Environmental Terms
Pollutants of Concern
Keyword
Ash pond
Discharge
Lake
Landfill
Leachate
Leaks
Lotic system
Plume
Pond
Power plant
Receiving water
River
Sediment
Steam electric
Stream
Surface waters
Water
Wastewater
Water pollution
Water quality
Waste management
Wastewater discharges
Algal blooms
Attractive nuisance
Background levels/concentrations
Bioaccumulation
Biomagnification
Biomagnify
Contamination
Environmental impact
Environmental assessment
Eutrophication
Fish
Fish consumption advisory
Fish kill
Fish mortality
Fish recovery
Hot Spot
Toxicity
Wildlife
Arsenic
Arsenate
Arsenite
Boron
Boric Acid
A-2
-------
Appendix A— Literature Review Methodology and Results
Table A-l. Keyword Search Terms for Environmental Impacts from
Steam Electric Power Plants
Category
Fuel Source Terms
Human Health Terms
Other Terms
Keyword
Chloride(s)
Chromium
Magnesium
Mercury
Metals
Methylmercury
Nitrate
Nitrogen
Selenium
Selenate
Selenite
Sulfate
Coal
Coal combustion by-products
Coal combustion residues
Oil
Cancer
Carcinogen
Carcinogenic
Drinking water
Health effects
Human health
Toxicity
Case study
Damage case assessment
Environmental impacts
Environmental aspects
To perform the literature search, EPA paired each fuel source term (see Table A-l) with at
least one keyword to focus the search results. Although EPA used multiple fuel source terms, the
environmental impacts from the steam electric power generating industry are documented most
commonly for coal-fired power plants. EPA used best professional judgment to create multiple
keyword combinations to further focus the literature search.
In addition to the key word combinations and search engines described above, EPA used
the following supplemental methods to identify more articles for the targeted topic areas:
• Reviewed references cited in previously identified published literature for additional
documented cases of environmental impact.
• Searched the Agency for Toxic Substances and Disease Registry's (ATSDR) website
for public health assessments and health consultations with information on the case
study sites referenced in Dr. Christopher Rowe's literature review paper published in
2002 [Rowe etal, 2002].
A-3
-------
Appendix A— Literature Review Methodology and Results
• Searched for case studies of attractive nuisances unrelated to the steam electric power
generating industry using the search engines described above.
• Reviewed EPA's December 2014 Coal Combustion Residuals (CCR) Damage Cases
Database and supporting compendiums [U.S. EPA, 2014a; U.S. EPA, 2014b; U.S.
EPA, 2014c; U.S. EPA, 2014d; U.S. EPA, 20146]1 and Michigan's Department of
Natural Resources and Environment (MDNRE's) Docket Comments (see Table A-3
for a full list of references).
• Searched magazines related to the steam electric industry and newspapers for articles
documenting additional environmental impacts.
EPA created a database for the literature review that documents the identified literature and
summarizes key information. EPA finalized the primary literature review on November 24, 2010;
however, the database also includes literature identified after the primary search efforts were
completed [ERG, 2013b]. EPA created a second database to summarize the damage cases and
other documented site impacts [ERG, 2015m].
The following tables in Appendix A summarize information EPA gathered from the
literature review:
• Table A-2. Summary of Literature Review Results by Information Source.
• Table A-3. Summary of Damage Cases and Other Documented Site Impacts to Surface
Water and Ground Water from Steam Electric Power Plant Discharges.
• Table A-4. Summary of Documented Ground Water Damage Cases from Surface
Impoundments.
• Table A-5. Summary of Documented Ground Water Damage Cases from Landfills.
• Table A-6. Summary of Documented Surface Water Damage Cases from Surface
Impoundments.
• Table A-7. Summary of Documented Surface Water Damage Cases from Landfills.
• Table A-8. Summary of Attractive Nuisances Related to Steam Electric Power Plants.
• Table A-9. Summary of Attractive Nuisances Unrelated to Steam Electric Power
Plants.
• Table A-10. Summary of Selenium Concentrations in the Environment and Organisms
Experiencing Adverse Effects.
Table A-2 highlights the results of the literature search, including documents identified by
keyword searches and relevant documents identified from supplemental methods. During the
period following completion of the literature review and the associated database, EPA obtained
additional documents (e.g., through public comments and informal searches) that supported
development of the final steam electric effluent limitations guidelines and standards (ELGs). EPA
1 These 2014 references are updates to EPA's September 18, 2012 review of damage cases which were primarily
identified in EPA's Damage COSQ Assessment Report, Environmental Integrity Project's (EIP's) Out of Control:
Mounting Damages From Coal Ash', and EIP's In Harm 's Way: Lack of Federal Coal Ash Regulations Endangers
Americans and Their Environment.
-------
Appendix A— Literature Review Methodology and Results
incorporated relevant information from the additional literature in the EA report and in the other
tables included in this Appendix.
Table A-3 summarizes the number of documented site impacts to surface water and ground
water identified during the literature search and organized by steam electric power plant. Table A-
4 and Table A-5 summarize the damage cases to ground water from combustion residuals surface
impoundments and landfills, respectively. Table A-6 and Table A-7 summarize the damage cases
to surface water from combustion residuals surface impoundments and landfills, respectively.
Table A-8 and Table A-9 summarize attractive nuisances identified during the literature search,
related and unrelated to steam electric power plants, respectively. Table A-10 presents selenium
concentrations in the environment that are documented in the literature as causing sublethal and
lethal effects to organisms.
A-5
-------
Appendix A— Literature Review Methodology and Results
Table A-2. Summary of Literature Review Results by Information Source
Source Type
Peer-Reviewed Literature a
Government Publication b
University Research °
Gray Literature d
Industry Publication e
Total
Number of Documents
Identified
151
53
13
18
4
239
Number of Documents
Reviewed f
128
47
12
16
3
206
Number of Documents that
Discussed Environmental
and Human Health Impacts
117
32
9
14
3
175
Source: ERG, 2013b.
a - Peer-reviewed literature consists of journal articles that undergo a formal review process prior to publishing.
b - Government publications are documents affiliated with state or federal government agencies.
c - University research includes finalized dissertations and theses, as well as papers published on behalf of a
university or presented at a conference.
d - Gray literature includes documents that are subjected to a less stringent review process (e.g., newspaper articles,
environmental group publications).
e - Industry publications include documents prepared by or for industry-affiliated entities.
f - EPA did not review several documents as part of the formal literature review either because EPA was unable to
acquire the full text of the document for review or because once the full text document was obtained a preliminary
review determined the document was not appropriate for inclusion in the literature review.
A-6
-------
Appendix A— Literature Review Methodology and Results
Table A-3. Summary of Surface Water and Ground Water Impacts Reported in Damage Cases and Other Documented Sites
from Steam Electric Power Plant Discharges
Plant Name
A.B. Brown Generating Station, Southern Indiana Gas and Electric Company (SIGECO) (IN)
Allen Fossil Plant Tennessee Valley Authority (TVA) (TN)
Allen Steam Generating Plant, Duke Power (NC)
Alma Station, Dairyland Power (WI)
Asheville Plant, Progress Energy (NC)
B.C. Cobb Power Plant, Consumers Energy (MI)
Bailly Generating Station, Northern Indiana Public Service Company (NIPSCO) (IN)
Belews Creek Steam Station, Duke Energy (NC)
Belle River Power Plant, Detroit Edison Company (MI)
Big Bend Station, Tampa Electric Company (FL)
Big Cajun 2 Power Plant, NRG Energy/Louisiana Generating, LLC (LA)
Brandon Shores, Constellation Energy (MD)
Brayton Point Station, Dominion (MA)
Bruce Mansfield Power Plant, First Energy (PA)
Buck Steam Station, Duke Energy (NC)
Bull Run Steam Plant, Tennessee Valley Authority (TVA) (TN)
C.D. Mclntosh, Jr. Power Plant, City of Lakeland (FL)
C.R. Huntley Generating Station, NRG Energy (NY)
Canadys Plant, South Carolina Electric & Gas (SCE&E) (SC)
Cape Fear Steam Plant, Progress Energy (NC)
Cardinal Plant, American Electric Power (AEP) (OH)
Cargill Salt Power Plant, Cargill (MI)
Cayuga Generating Station, Duke Energy (NY)
Chalk Point Generating Station, Mirant (MD)
Chesapeake Energy Facility, Dominion Power (VA)
Number of Damage Cases
and Other Literature that
Document
Surface Water Impacts a
0
0
1
0
2
0
0
14
1
1
0
0
0
1
1
1
0
0
0
0
Number of Damage Cases
and Other Literature that
Document
Ground Water Impacts a
1
1
1
2
1
2
2
1
1
1
1
1
1
1
0
1
1
1
1
1
1
1
1
1
2
A-7
-------
Appendix A— Literature Review Methodology and Results
Table A-3. Summary of Surface Water and Ground Water Impacts Reported in Damage Cases and Other Documented Sites
from Steam Electric Power Plant Discharges
Plant Name
Cholla Steam Electric Generating Station, Arizona Public Service Company (AZ)
Christ Power Plant, Gulf Power (Southern Company) (FL)
Clifty Creek Station, Indiana Kentucky Electric Company (IKEC) (IN)
Clinch River Plant, American Electric Power (AEP)/ Appalachian Power (VA)
Coal Creek Station, Cooperative Power Association/United Power (ND)
Coffeen Power Station, Ameren (IL)
Colbert Fossil Plant, Tennessee Valley Authority (TV A) (AL)
Coleto Creek Power Station, International Power (TX)
Colstrip Power Plant, PPL Montana (MT)
Columbia Electric Generating Station (WI)
Columbia Energy Center, Alliant Energy (WI)
Conesville Power Plant, American Electric Power (AEP) (OH)
Cross Generating Station, Santee Cooper/South Carolina Public Service Authority (SCPSA) (SC)
Cumberland Steam Plant, Tennessee Valley Authority (TVA) (TN)
Curtis Stanton Energy Center, Orlando Utility Commission (FL)
Dallman Station, City Water, Light and Power (IL)
Dan River Steam Station, Duke Energy (NC)
Danskammer Generating Station, Dynegy (NY)
D-Area Coal-Fired Power Plant, Savannah River Site (SRS) (SC)
Dave Johnston Power Plant (WY)
Dickerson Generating Station, Mirant (MD)
Dolet Hills Power Station, Central Louisiana Electric Co-Op (CLECO) Power, LLC (LA)
Duck Creek Station, Central Illinois Light Company (IL)
Dunkirk Generating Station, NRG Energy (NY)
EJ. Stoneman Generating Station, Dairyland Power Cooperative (WI)
Number of Damage Cases
and Other Literature that
Document
Surface Water Impacts a
0
0
0
1
0
0
0
0
0
5
1
0
0
1
1
0
2
0
24
1
1
0
0
0
0
Number of Damage Cases
and Other Literature that
Document
Ground Water Impacts a
1
1
1
0
1
1
1
1
1
0
0
1
1
1
1
1
1
1
0
1
1
1
1
1
1
A-8
-------
Appendix A— Literature Review Methodology and Results
Table A-3. Summary of Surface Water and Ground Water Impacts Reported in Damage Cases and Other Documented Sites
from Steam Electric Power Plant Discharges
Plant Name
East Bend Generating Station, Cinergy (KY)
Eckert Station, Lansing Board of Water & Light (MI)
Edgewater Generating Station, Alliant Energy (WI)
Elizabethtown Power Plant, North Carolina Power Holdings (NC)
Elrama Power Plant, Reliant Energy (PA)
Erickson Station, Lansing Board of Water & Light (MI)
Fair Station, Central Iowa Power Cooperative (IA)
Fayette Power Project, Lower Colorado River Authority (TX)
Flint Creek Power Plant, American Electric Power (AEP)/South West Electric Power Company
(SWEPCO) (AR)
Gallatin Fossil Plant, Tennessee Valley Authority (TVA) (TN)
General James M. Gavin Power Plant, American Electric Power/Ohio Power Company (OH)
George Neal Station North, Berkshire Hathaway /MidAmerican Energy Company (IA)
George Neal Station South, Berkshire Hathaway /MidAmerican Energy Company (IA)
Gibson Generating Station, Duke Energy (IN)
Glen Lyn Plant, American Electric Power (AEP)/ Appalachian Power (VA)
Grainger Generating Station, Santee Cooper/South Carolina Public Service Authority (SCPSA) (SC)
Greenidge Generation Plant, AES (NY)
Harbor Beach Power Plant, Detroit Edison Company (MI)
Hatfield's Ferry Power Station, Allegheny Energy (PA)
Havana Power Plant, Illinois Power Company (IL)
Hennepin Power Station, Illinois Power Company (IL)
Herbert A. Wagner, Constellation Energy (MD)
Hickling Generation Plant, AES (NY)
Hopewell Power Station, Dominion Power (VA)
Number of Damage Cases
and Other Literature that
Document
Surface Water Impacts a
0
0
0
0
1
0
0
0
1
0
1
0
0
5
6
1
0
1
1
0
0
0
0
0
Number of Damage Cases
and Other Literature that
Document
Ground Water Impacts a
1
1
1
1
1
1
2
1
1
1
1
1
1
1
0
1
1
1
1
1
1
1
1
1
A-9
-------
Appendix A— Literature Review Methodology and Results
Table A-3. Summary of Surface Water and Ground Water Impacts Reported in Damage Cases and Other Documented Sites
from Steam Electric Power Plant Discharges
Plant Name
Hunlock Power Station, UGI Development Company (PA)
Hutsonville Power Station, Central Illinois Public Service Company (IL)
Independence Steam Station, Entergy /Arkansas Power and Light (AR)
Indian River Generating Station, NRG Energy (DE)
J.H. Campbell Power Plant, Consumers Energy (MI)
J.R. Whiting Generating Plant, CMS/Consumers Energy (MI)
Jennison Generation Plant, AES (NY)
John Amos Plant, American Electric Power (AEP)/ Appalachian Power (WV)
John H. Warden Generating Station, Integrys (MI)
John Sevier Fossil Plant, Tennessee Valley Authority (TVA) (TN)
Johnsonville Fossil Plant, Tennessee Valley Authority (TVA) (TN)
Joliet Generating Station 9, Midwest Generation (IL)
Joppa Steam Plant, Ameren (Electric Energy) (IL)
Karn/Weadock Generating Facility, Consumer Energy (MI)
Kenansville Plant, Green Power Energy Holdings (NC)
Kingston Fossil Plant, Tennessee Valley Authority (TVA) (TN)
Lansing Smith Plant, Florida Power and Light (FL)
Lee Steam Plant, Progress Energy (NC)
Leland Olds Station, Basin Electric Power Cooperative (ND)
Lumberton Power Plant, North Carolina Power Holdings (NC)
Marion Plant, Southern Illinois Power Cooperative (IL)
Marshall Steam Station, Duke Energy (NC)
Martin Lake Steam Station, Texas Utilities Electric Service Company (TX)
Martin's Creek Power Plant, PPL (PA)
Marysville Power Plant, Detroit Edison Company (MI)
Number of Damage Cases
and Other Literature that
Document
Surface Water Impacts a
0
0
0
1
1
1
0
1
1
1
2
0
0
0
0
7
0
0
0
0
1
1
9
1
1
Number of Damage Cases
and Other Literature that
Document
Ground Water Impacts a
1
1
1
1
1
0
1
0
1
1
2
2
1
1
1
1
1
1
1
1
1
0
0
0
1
A-10
-------
Appendix A— Literature Review Methodology and Results
Table A-3. Summary of Surface Water and Ground Water Impacts Reported in Damage Cases and Other Documented Sites
from Steam Electric Power Plant Discharges
Plant Name
Mayo Steam Station, Progress Energy (NC)
McMeekin Station, SCANA/South Carolina Electric & Gas Company (SCE&G) (SC)
Mendosa Power Station, Ameren Energy Generating Company, (IL)
Merom Generating Station, Hoosier Energy (IN)
Miamiview Landfill, Cincinnati Gas & Electric Company (OH) b
Michigan City Generating Station, Northern Indiana Public Service Company (NIPSCO) (IN)
Mill Creek Plant, E ON U.S./Louisville Gas & Electric (LG&E) (KY)
Mitchell Power Station, Allegheny Energy (PA)
Montville Generating Station, NRG Energy /Montville Power, LLC (CT)
Morgantown Generating Station, Mirant (MD)
Muskingum River Plant, American Electric Power (AEP)/ Ohio Power Company (OH)
Nelson Dewey Generating Station, Alliant Energy (WI)
Northeastern Station, American Electric Power/Public Service Company Oklahoma (OK)
Oak Creek Power Plant, Wisconsin Energy (WE Energies (WE))/Wisconsin Electric Power
Company (WI)
Oak Ridge Y-12 Plant, Department of Energy (TN)
Paradise Fossil Plant, Tennessee Valley Authority (TVA) (KY)
Parish Generating Station, NRG Energy /Texas Genco II (TX)
Pearl Station, Prairie Power Inc. /Soy land Power Coop (IL)
Petersburg Generating Station, Indianapolis Power & Light (IN)
Phillips Power Plant, Duquesne Light Company (PA)
Pirkey Power Plant, Southwestern Electric Power Company (SWEPCO) (TX)
Plant Bowen, Georgia Power (GA)
Port Washington Facility, Wisconsin Electric Power Company (WEPCO) (WI)
Portland Generating Station, RRI Energy (PA)
Number of Damage Cases
and Other Literature that
Document
Surface Water Impacts a
1
0
0
1
0
0
0
0
1
2
0
0
0
1
4
0
0
0
0
1
2
1
0
1
Number of Damage Cases
and Other Literature that
Document
Ground Water Impacts a
0
1
1
1
1
1
1
1
1
2
1
1
1
0
0
0
2
1
A-ll
-------
Appendix A— Literature Review Methodology and Results
Table A-3. Summary of Surface Water and Ground Water Impacts Reported in Damage Cases and Other Documented Sites
from Steam Electric Power Plant Discharges
Plant Name
Powerton Plant, Commonwealth Edison (IL)
Prairie Creek Station, Interstate Power and Light (Alliant) (IA)
Presque Isle Power Plant, WE Energies (WE) (MI)
Pulliam Power Plant, Wisconsin Public Service Corp. (WI)
R.M. Heskett Station, Montana-Dakota Utilities (ND)
R.M. Schahfer Generating Station (IN)
Reid Gardner Generating Facility, Nevada Energy (NV)
Riverbend Steam Station, Duke Energy (NC)
Rock River Generating Station, Alliant Energy (WI)
Rocky Mount Power Plant (NC)
Rodemacher Power Station, Central Louisiana Electric Co-Op (CLECO) Power, LLC (LA)
Roxboro Plant, Progress Energy (NC)
Salem Harbor Station, Dominion (MA)
SCANA Williams Station (SC)
Seminole Generating Station, Seminole Electric Cooperative (FL)
Seward Generating Station, RRI Energy (PA)
Shawnee Fossil Plant, Tennessee Valley Authority (TVA) (KY)
Sheldon Station, Nebraska Public Power District (NE)
Sherburne County (Sherco) Generating Plant, Xcel Energy /Southern Minnesota Municipal Power
Agency (MM)
Shiras, Marquette Board of Light & Power (MI)
Spurlock Station, Eastern Kentucky Power Cooperative (KY)
Sutton Steam Plant, Progress Energy (NC)
Unnamed Plant 1°
Number of Damage Cases
and Other Literature that
Document
Surface Water Impacts a
1
0
0
0
0
0
1
4
0
0
0
8
0
1
1
1
1
0
0
0
0
1
1
Number of Damage Cases
and Other Literature that
Document
Ground Water Impacts a
1
1
1
1
1
1
1
0
1
1
1
0
1
0
1
1
1
1
1
1
1
1
0
A-12
-------
Appendix A— Literature Review Methodology and Results
Table A-3. Summary of Surface Water and Ground Water Impacts Reported in Damage Cases and Other Documented Sites
from Steam Electric Power Plant Discharges
Plant Name
Unnamed Plant 2 °
Unnamed Plant 3 °
Unnamed Plant 4 °
Urquhart Station, South Carolina Electric & Gas Company (SGE&E) (SC)
Valley Power Plant, Wisconsin Energy (WI)
Venice Power Station, Union Electric Company /Ameren Energy /AmerenUE (IL)
Vermillion Power Station, Illinois Power (IL)
W.C. Beckjord Station, Duke Energy (formerly Cinergy) (OH)
W.J. Neal Station, Basin Electric Power Cooperative (ND)
Wateree Station, SCE&G (SC)
Waukegan Generating Station, Midwest Generation (Edison International) (IL)
Welsh Power Plant, Southwestern Electric Power Company (SWEPCO) (TX)
Westover Generation Plant, AES (NY)
Widows Creek Fossil Plant, Tennessee Valley Authority (TVA) (AL)
Winyah Generating Station, Santee Cooper/South Carolina Public Service Authority (SCPSA) (SC)
Wood River Power Station, Illinois Power Company (IL)
Yorktown Power Station, Virginia Electric Power and Power Company (VEPCO) (VA)
Total
Number of Damage Cases
and Other Literature that
Document
Surface Water Impacts a
1
1
1
0
0
0
0
0
1
1
0
3
0
0
0
0
0
152
Number of Damage Cases
and Other Literature that
Document
Ground Water Impacts a
0
0
0
1
1
1
1
1
1
1
1
0
1
1
1
1
1
149
Source: ERG, 2015m; U.S. EPA, 2014a through 2014e.
a - One case study or damage case may document impacts to both ground water and surface water.
b - The damage case source did not specifically identify the plant name; therefore, EPA used the name of the damage case.
c - EPA was unable to identify the steam electric power plant associated with this documented impact. For the purpose of counting the unique number of plants,
these impacts were assumed to be associated with a plant not already identified elsewhere in this table.
A-13
-------
Appendix A— Literature Review Methodology and Results
Table A-4. Summary of Ground Water Impacts Reported in Damage Cases from Steam Electric Power Plant Surface
Impoundments
Damage Case Site
Allen Fossil Plant Tennessee
Valley Authority (TVA) (TN)
Allen Steam Generating Plant,
Duke Power (NC)
Alma Off-site Fly Ash Landfill,
Dairyland Power (WI)
Asheville Steam Electric Plant,
Progress Energy (NC)
Bailly Generating Station,
Northern Indiana Public Service
Company (NIPSCO) (IN)
Bangor Quarry Ash Disposal
Site, Portland Generating Station,
RRI Energy (PA)
BC Cobb, Consumers Energy
(MI)
Belews Creek Steam Station,
Duke Energy (NC)
Big Bend Station, Tampa Electric
Company (FL)
Big Cajun 2 Power Plant, NRG
Energy /Louisiana Generating,
LLC (LA)
Type of Waste in
Impoundment a
Pond/Impoundment
Pond/Impoundment
Pond/Impoundment
Pond/Impoundment
Pond/Impoundment,
Landfill
Pond/Impoundment,
Landfill
Pond/Impoundment
Pond/Impoundment,
Landfill
Pond/Impoundment,
Landfill
Pond/Impoundment
Pollutants of Concern
Arsenic, Manganese, TDS
Manganese, Iron, pH, Nitrate,
Nickel
Sulfate, Manganese, Boron,
Selenium, Cadmium
Boron, Chromium, Iron,
Manganese, Thallium, Nitrate,
Sulfate, pH, TDS, Cadmium,
Arsenic, Antimony
Arsenic, Cadmium
Selenium, Boron, Cadmium,
Hexavalent Chromium, Iron,
Manganese, Sulfate, TDS,
Aluminum, Fluoride
Boron, Lithium, Manganese,
Sulfate, Ammonia
Selenium, Arsenic, Boron,
Cadmium, Iron, Lead,
Manganese, Nitrate, Sulfate, pH,
Bromide
Arsenic, Aluminum, Boron,
Chloride, Fluoride, Iron,
Manganese, Molybdenum,
Sulfate, Sodium, Thallium, TDS
Selenium, TDS, Barium, Arsenic
Exceeded
MCLb
X
X
X
X
X
X
X
X
X
Exceeded Federal/
State WQC/
Standards b
X
X
X
Ground Water
Impacted Surface
Waters c
X
X
Impacted
Off-Site
Source d
X
X
X
A-14
-------
Appendix A— Literature Review Methodology and Results
Table A-4. Summary of Ground Water Impacts Reported in Damage Cases from Steam Electric Power Plant Surface
Impoundments
Damage Case Site
Brandywine Coal Ash Landfill,
Mirant Mid-Atlantic LLC (MD)
Bull Run Steam Plant, Tennessee
Valley Authority (TVA) (TN)
C.D. Mclntosh, Jr. Power Plant,
City of Lakeland (FL)
C.R. Huntley Flyash Landfill
(NY)
Canadys Plant, South Carolina
Electric & Gas (SCE&E) (SC)
Cape Fear Steam Plant, Progress
Energy (NC)
Cardinal Fly Ash Reservoir
(FAR) 1 and 2, American
Electric Power (AEP) (OH)
Cayuga Coal Ash Disposal
Landfill, AES (NY)
Cholla Steam Electric Generating
Station, Arizona Public Service
Company (AZ)
Type of Waste in
Impoundment a
Pond/Impoundment,
Landfill
Pond/Impoundment
Pond/Impoundment,
Landfill
Pond/Impoundment,
Landfill
Pond/Impoundment
Pond/Impoundment
Pond/Impoundment,
Landfill
Pond/Impoundment,
Landfill
Pond/Impoundment
Pollutants of Concern
Selenium, Cadmium, Lead,
Manganese, Iron, Aluminum,
Sulfate, TDS, Chloride
Aluminum, Cadmium, Iron,
Sulfate, Arsenic, Cobalt, Calcium,
Manganese, Molybdenum, Boron,
Nickel
Selenium, Arsenic, Cadmium,
Lead, Manganese, Vanadium,
Nitrate, Iron, Sulfate, TDS, pH
Arsenic, Iron, Manganese,
Sulfate, TDS, Cadmium, Barium,
Lead, TSS
Arsenic, Nickel, Selenium
Lead, Chromium, Boron, Iron,
Manganese, Sulfate, Selenium
Arsenic, Boron, Molybdenum
Selenium, Arsenic, Boron,
Cadmium, Lead, TDS,
Aluminum, Manganese, Sulfate,
Barium, Sodium, Iron,
Chromium, Zinc
Sulfate, TDS, Chloride, Fluoride
Exceeded
MCLb
X
X
X
X
X
X
X
X
X
Exceeded Federal/
State WQC/
Standards b
X
X
X
X
X
Ground Water
Impacted Surface
Waters c
X
Impacted
Off-Site
Source d
X
X
X
A-15
-------
Appendix A— Literature Review Methodology and Results
Table A-4. Summary of Ground Water Impacts Reported in Damage Cases from Steam Electric Power Plant Surface
Impoundments
Damage Case Site
Clifty Creek Station, Indiana
Kentucky Electric Company
(IKEC) (IN)
Coal Creek Station Surface
Impoundments, Cooperative
Power Association/United Power
(ND)
Colbert Fossil Plant, Tennessee
Valley Authority (TVA) (AL)
Coleto Creek Power Station,
International Power (TX)
Colstrip Power Plant, PPL
Montana (MT)
Cross Generating Station, Santee
Cooper/South Carolina Public
Service Authority (SCPSA) (SC)
Cumberland Steam Plant,
Tennessee Valley Authority
(TVA) (TN)
Curtis Stanton Energy Center,
Orlando Utility Commission (FL)
Dallman Station Ash and FGD
Ponds, City Water, Light and
Power (IL)
Type of Waste in
Impoundment a
Pond/Impoundment,
Landfill
Pond/Impoundment
Pond/Impoundment,
Landfill
Pond/Impoundment
Pond/Impoundment,
Landfill
Pond/Impoundment,
Landfill
Pond/Impoundment,
Landfill
Pond/Impoundment,
Landfill
Pond/Impoundment,
Landfill
Pollutants of Concern
Boron, Manganese, Iron, Sulfate,
Magnesium
Selenium, Arsenic, Sulfate,
Chloride, Boron, Chromium, Iron,
Sodium, TDS
Cadmium, Antimony, Arsenic,
Lead, Nitrate, Aluminum, Iron,
Manganese, Boron, Molybdenum,
Cobalt, Lithium, Sulfate,
Chromium
Arsenic, Lead, Boron, Cobalt,
Nickel, Vanadium
Selenium, Boron, Sulfate, TDS,
Molybdenum, Arsenic, Chloride
Arsenic, Cadmium, Chromium,
Sodium, Sulfate, Iron, Aluminum,
Chloride, TDS
Selenium, Arsenic, Aluminum,
Boron, Chloride, Iron,
Manganese, Sulfate, TDS,
Vanadium
Aluminum, Chloride, Iron,
Manganese, Sodium, Sulfate,
TDS, Vanadium, pH
Arsenic, Chromium, Sodium,
Boron, Manganese, Iron, Sulfate,
TDS
Exceeded
MCLb
X
X
X
X
X
X
X
X
X
Exceeded Federal/
State WQC/
Standards b
X
Ground Water
Impacted Surface
Waters c
Impacted
Off-Site
Source d
X
A-16
-------
Appendix A— Literature Review Methodology and Results
Table A-4. Summary of Ground Water Impacts Reported in Damage Cases from Steam Electric Power Plant Surface
Impoundments
Damage Case Site
Dan River Steam Station, Duke
Energy (NC)
Dave Johnston Power Plant
(WY)
Dolet Hills Power Station,
Central Louisiana Electric Co-Op
(CLECO) Power, LLC (LA)
Duck Creek Station, Central
Illinois Light Company (IL)
EJ. Stoneman Generating
Station, Dairy land Power
Cooperative (WI)
Edgewater 1-4 Ash Disposal Site,
Alliant (formerly Wisconsin
Power & Light) (WI)
Fayette Power Project (Sam
Seymour), Lower Colorado River
Authority (TX)
Flint Creek Power Plant,
American Electric Power
(AEP)/South West Electric
Power Company (SWEPCO)
(AR)
Fly Ash Landfill, Coffeen/White
& Brewer Trucking (IL)
Type of Waste in
Impoundment a
Pond/Impoundment,
Landfill
Pond/Impoundment,
Landfill
Pond/Impoundment,
Landfill
Pond/Impoundment
Pond/Impoundment
Pond/Impoundment,
Landfill
Pond/Impoundment,
Landfill
Pond/Impoundment,
Landfill
Pond/Impoundment,
Landfill
Pollutants of Concern
Chromium, Iron, Lead,
Manganese, Silver, Sulfate,
Arsenic, Antimony, Boron, TDS,
pH
Cadmium, Manganese, Sulfate,
Boron
Selenium, Arsenic, Lead,
Chloride, TDS, Sulfate, Iron, pH
Sulfate, TDS, Chloride,
Manganese, Iron, Boron
Cadmium, Chromium, Sulfate,
Manganese, Iron, Zinc, Boron,
Barium
Boron, Sulfate, Iron, Chloride,
TDS, Arsenic, Selenium
Selenium, Aluminum, Chloride,
Cobalt, Manganese,
Molybdenum, Sulfate, TDS,
Vanadium
Selenium, Barium, Cadmium,
Chromium, Iron, Lead,
Manganese, pH, Silver, Sulfate,
TDS
Sulfate, TDS, Manganese,
Cadmium, Chromium, Thallium,
Beryllium, Boron, Nickel,
Barium, Iron, Zinc, Aluminum,
Sodium
Exceeded
MCLb
X
X
X
X
X
X
X
X
Exceeded Federal/
State WQC/
Standards b
X
X
Ground Water
Impacted Surface
Waters c
Impacted
Off-Site
Source d
X
X
A-17
-------
Appendix A— Literature Review Methodology and Results
Table A-4. Summary of Ground Water Impacts Reported in Damage Cases from Steam Electric Power Plant Surface
Impoundments
Damage Case Site
Gallatin Fossil Plant, Tennessee
Valley Authority (TVA) (TN)
General James M. Gavin Power
Plant, American Electric
Power/Ohio Power Company
(OH)
George Neal Station North
Landfill, Berkshire
Hathaway /MidAmerican Energy
Company (IA)
Gibson Generating Station, Duke
Energy (IN)
Grainger Generating Station,
Santee Cooper/South Carolina
Public Service Authority
(SCPSA) (SC)
Havana Power Plant, Illinois
Power Company (IL)
Hennepin Power Station, Illinois
Power Company (IL)
Hunlock Power Station, UGI
Development Company (PA)
Hutsonville Power Station,
Central Illinois Public Service
Company (IL)
Type of Waste in
Impoundment a
Pond/Impoundment
Pond/Impoundment,
Landfill
Landfill,
Pond/Impoundment
Pond/Impoundment,
Landfill, Cooling
Reservoir
Pond/Impoundment
Pond/Impoundment
Pond/Impoundment,
Landfill
Pond/Impoundment
Pond/Impoundment
Pollutants of Concern
Boron, Beryllium, Cadmium,
Iron, Manganese, Nickel, Sulfate,
TDS, Arsenic, Mercury,
Vanadium, Cobalt
Arsenic, Barium, Cadmium,
Lead, Molybdenum, Sulfate,
TDS, Aluminum, Copper, Nickel,
Zinc, Manganese, Chloride
Iron, Manganese, Sulfate, Arsenic
Selenium, Arsenic, Boron,
Manganese, Iron, Sodium
Arsenic, pH
Manganese, Sulfate, Boron
Sulfate, TDS, Boron, Iron,
Manganese
Arsenic, Iron, Manganese
Sulfate, TDS, Manganese, Boron
Exceeded
MCLb
X
X
X
X
X
X
X
Exceeded Federal/
State WQC/
Standards b
X
X
Ground Water
Impacted Surface
Waters c
X
Impacted
Off-Site
Source d
X
X
X
A-18
-------
Appendix A— Literature Review Methodology and Results
Table A-4. Summary of Ground Water Impacts Reported in Damage Cases from Steam Electric Power Plant Surface
Impoundments
Damage Case Site
Independence Steam Station,
Entergy /Arkansas Power and
Light (AR)
J.H. Campbell, Consumers
Energy (MI)
John Sevier Fossil Plant,
Tennessee Valley Authority
(TVA) (TN)
Johnsonville Fossil Plant,
Tennessee Valley Authority
(TVA) (TN)
Joppa Steam Plant Ash Ponds,
Ameren (Electric Energy) (IL)
Karn/Weadock Generating
Facility, Consumer Energy (MI)
Kingston Fossil Plant, Tennessee
Valley Authority (TVA) (TN)
Lansing Smith Plant, Florida
Power and Light (FL)
Lee Steam Plant, Progress
Energy (NC)
Leland Olds Station, Basin
Electric Power Cooperative (ND)
Type of Waste in
Impoundment a
Pond/Impoundment,
Landfill
Pond/Impoundment
Pond/Impoundment
Pond/Impoundment,
Landfill
Pond/Impoundment
Pond/Impoundment,
Landfill
Pond/Impoundment
Pond/Impoundment
Pond/Impoundment
Pond/Impoundment
Pollutants of Concern
Cadmium, Iron, Lead,
Manganese, pH, Sulfate, TDS,
Arsenic, Chlorine
pH, Antimony, Boron, Cadmium,
Chromium, Iron, Lead, Selenium,
Vanadium, Aluminum, Nickel,
Thallium, Manganese, Zinc
Arsenic, Aluminum, Cadmium,
Manganese, Boron, Strontium,
Sulfate, Selenium, Hexavalent
Chromium
Arsenic, Aluminum, Boron,
Cadmium, Chromium, TDS, Iron,
Lead, Manganese, Molybdenum,
Sulfate, Cobalt
Lead, Chromium, Cobalt, Boron,
Manganese, Sulfate, Iron, TDS
Arsenic, Boron, Lithium,
Arsenic, Selenium, Manganese,
Cobalt, Aluminum, Ammonia,
Thallium, Iron
Aluminum, Cadmium, Chloride,
Chromium, Fluoride, Sulfate,
Manganese, Iron, Radium-226,
Radium-228, TDS, Sodium
Arsenic, Lead, Boron,
Manganese, Iron, Chromium, pH
Arsenic, Boron, Lead, Sulfate
Exceeded
MCLb
X
X
X
X
X
X
X
X
X
X
Exceeded Federal/
State WQC/
Standards b
X
X
X
X
Ground Water
Impacted Surface
Waters c
X
X
Impacted
Off-Site
Source d
X
X
X
X
X
X
A-19
-------
Appendix A— Literature Review Methodology and Results
Table A-4. Summary of Ground Water Impacts Reported in Damage Cases from Steam Electric Power Plant Surface
Impoundments
Damage Case Site
Lincoln Stone Quarry Landfill,
Joliet Generating Station 29,
Midwest Generation (IL)
Lincoln Stone Quarry Landfill,
Joliet Generating Station 9,
Midwest Generation (IL)
Little Blue Run Surface
Impoundment, Bruce Mansfield
Power Plant, First Energy (PA)
Mahoney Landfill, Powerton
Plant, Commonwealth Edison
(IL)
Marion Plant, Southern Illinois
Power Cooperative (IL)
McMeekin Station,
SCANA/South Carolina Electric
& Gas Company (SCE&G) (SC)
Mendosa Power Station Ash
Ponds, Ameren Energy
Generating Company, (IL)
Michigan City Site (IN)
Mill Creek Plant, EON
U.S./Louisville Gas & Electric
(LG&E) (KY)
Type of Waste in
Impoundment a
Pond/Impoundment
Pond/Impoundment,
Landfill
Pond/Impoundment
Pond/Impoundment,
Landfill
Pond/Impoundment,
Landfill
Pond/Impoundment,
Landfill
Pond/Impoundment
Pond/Impoundment
Pond/Impoundment,
Landfill
Pollutants of Concern
Antimony, Manganese, Sulfate,
Chloride, TDS
Arsenic, Ammonia, Boron,
Molybdenum, pH, Sulfate, TDS,
Barium, Copper, Selenium,
Cadmium
Selenium, Arsenic, Aluminum,
Antimony, Barium, Boron,
Cadmium, Calcium, Chloride,
Hexavalent Chromium, Fluoride,
Iron, Lead, Manganese, pH,
Sodium, Sulfate, TDS, TSS,
Thallium, Turbidity
Arsenic, Selenium, Chromium,
TDS, Cadmium, Lead, Nitrate,
Iron, Manganese, Sulfate, Boron,
Boron, Cadmium, Iron,
Aluminum, TDS, Sulfate
Chromium, Lead, Sulfate, Iron,
TDS
Arsenic, Boron, Manganese,
Chromium (?), Sulfate, TDS
Arsenic, Lead
Arsenic, Chloride, Sulfate, TDS
Exceeded
MCLb
X
X
X
X
X
X
X
X
X
Exceeded Federal/
State WQC/
Standards b
X
X
X
Ground Water
Impacted Surface
Waters c
X
Impacted
Off-Site
Source d
X
X
X
A-20
-------
Appendix A— Literature Review Methodology and Results
Table A-4. Summary of Ground Water Impacts Reported in Damage Cases from Steam Electric Power Plant Surface
Impoundments
Damage Case Site
Mitchell Power Station,
Allegheny Energy (PA)
Montville Generating Station,
NRG Energy /Montville Power,
LLC (CT)
Morgantown Generating Station,
Faulkner Off-site Disposal
Facility (MD)
Muskingum River Plant,
American Electric Power (AEP)/
Ohio Power Company (OH)
Nelson Dewey Ash Disposal
Facility, Alliant (formerly
Wisconsin Power & Light) (WI)
Northeastern Station Ash
Landfill, American Electric
Power/Public Service Company
Oklahoma (OK)
Oak Ridge Y-12 Plant, Chestnut
Ridge Operable Unit 2, Oak
Ridge Reservation, Department
of Energy (TN)
Paradise Fossil Plant, Tennessee
Valley Authority (TVA) (KY)
Parish Generating Station, NRG
Energy /Texas Genco II (TX)
Pearl Station, Prairie Power
Inc./Soyland Power Coop (IL)
Type of Waste in
Impoundment a
Pond/Impoundment,
Landfill
Pond/Impoundment
Pond/Impoundment,
Landfill
Pond/Impoundment
Pond/Impoundment
Landfill,
Pond/Impoundment
Pond/Impoundment
Pond/Impoundment
Pond/Impoundment,
Landfill
Pond/Impoundment
Pollutants of Concern
Arsenic, Boron, Iron,
Molybdenum, Manganese, Nickel
Arsenic, Beryllium, Cadmium,
Copper, Iron, Lead, Manganese,
Nickel, pH, Zinc
Iron, pH, Cadmium, Aluminum,
Chloride, Manganese, Sulfate,
TDS, Copper, Lead, Selenium
Barium, Iron, Sulfate
Selenium, Arsenic, Sulfate,
Boron, Fluoride, Cadmium (?),
Iron
Selenium, Arsenic, Barium,
Chromium, Lead, Vanadium,
Thallium, Sulfate, pH
Selenium, Arsenic, Aluminum,
Iron, Zinc, Manganese, Thallium
(?)
Arsenic, Boron, Chromium,
Copper, Manganese
Arsenic, Selenium, Barium,
Boron, Chromium, Cobalt,
Manganese, Molybdenum, Sulfate
Arsenic, Chromium, Boron,
Manganese, Sulfate, Chlorine,
Iron, TDS, Lead, Boron
Exceeded
MCLb
X
X
X
X
X
X
X
X
X
Exceeded Federal/
State WQC/
Standards b
X
X
Ground Water
Impacted Surface
Waters c
X
X
X
Impacted
Off-Site
Source d
X
X
X
X
X
A-21
-------
Appendix A— Literature Review Methodology and Results
Table A-4. Summary of Ground Water Impacts Reported in Damage Cases from Steam Electric Power Plant Surface
Impoundments
Damage Case Site
Phillips Power Plant Landfill,
Duquesne Light Company (PA)
Prairie Creek Generating Station
Ash Landfill, Interstate Power
and Light (Alliant) (IA)
R.M. Schahfer Generating
Station (IN)
Reid Gardner Generating
Facility, Nevada Energy (NV)
Rock River Ash Disposal
Facility, Alliant (formerly
Wisconsin Power & Light) (WI)
Rodemacher Power Station,
Central Louisiana Electric Co-Op
(CLECO) Power, LLC (LA)
Seminole Generating Station,
Seminole Electric Cooperative
(FL)
Seward Generating Station, RRI
Energy (PA)
Type of Waste in
Impoundment a
Pond/Impoundment,
Landfill
Pond/Impoundment,
Landfill
Landfill,
Pond/Impoundment
Pond/Impoundment,
Landfill
Pond/Impoundment
Pond/Impoundment,
Landfill
Pond/Impoundment,
Landfill
Pond/Impoundment,
Landfill
Pollutants of Concern
TDS, Chloride, Fluoride,
Manganese, Aluminum, Arsenic
Arsenic, Boron, Manganese,
Sulfate, Iron
Sulfate, Iron, Manganese,
Molybdenum, Chlorine, Sodium,
Boron
Selenium, Arsenic, Chloride,
Sulfate, TDS, Nitrate, Boron,
Chromium, Manganese,
Magnesium, Molybdenum,
Sodium, Vanadium, Titanium,
Barium, Iron, Aluminum
Mercury, Arsenic, Sulfate, Iron,
Selenium, Boron, TDS
Arsenic, Lead, pH, TDS,
Chloride, Sulfate
Arsenic, Chloride, Chlorine,
Sulfate, Iron, TDS, Boron,
Aluminum, Lead, Sodium
Selenium, Arsenic, Aluminum,
Antimony, Cadmium, Chloride,
Chromium, Iron, Lead,
Manganese, Nickel, pH, Sulfate,
TDS, Zinc,
Exceeded
MCLb
X
X
X
X
X
X
X
Exceeded Federal/
State WQC/
Standards b
X
X
X
X
X
Ground Water
Impacted Surface
Waters c
X
X
Impacted
Off-Site
Source d
X
X
X
X
A-22
-------
Appendix A— Literature Review Methodology and Results
Table A-4. Summary of Ground Water Impacts Reported in Damage Cases from Steam Electric Power Plant Surface
Impoundments
Damage Case Site
Shawnee Fossil Plant, Tennessee
Valley Authority (TVA) (KY)
Sherburne County (Sherco)
Generating Plant, Xcel
Energy/Southern Minnesota
Municipal Power Agency (MN)
Spurlock Station, Eastern
Kentucky Power Cooperative
(KY)
Sutton Steam Plant, Progress
Energy (NC)
Urquhart Station, South Carolina
Electric & Gas Company
(SGE&E) (SC)
Venice Power Station Ash Ponds,
Union Electric Company /Ameren
Energy/ AmerenUE (IL)
Vermillion Power Station,
Illinois Power (IL)
W.C. Beckjord Station, Duke
Energy (formerly Cinergy) (OH)
W.J. Neal Station Surface
Impoundment, Basin Electric
Power Cooperative (ND)
Wateree Station, SCE&G (SC)
Type of Waste in
Impoundment a
Pond/Impoundment,
Landfill
Pond/Impoundment,
Landfill
Pond/Impoundment,
Landfill
Pond/Impoundment
Pond/Impoundment,
Landfill
Pond/Impoundment
Pond/Impoundment
Pond/Impoundment
Pond/Impoundment
Pond/Impoundment,
Landfill
Pollutants of Concern
Selenium, Arsenic, Boron, pH,
Sulfate, TDS, Beryllium, Cobalt,
Nickel, Molybdenum,
Manganese, Vanadium
Arsenic, Cadmium, Lead, Sulfate,
Selenium, Boron
Arsenic, Sulfate, TDS
Arsenic, Boron, Manganese, Iron,
Thallium, Selenium, Antimony,
Lead, Sulfate, TDS
Arsenic, Nickel
Arsenic, Boron, Cadmium, Iron,
Manganese, TDS
Sulfate, TDS, Boron, Iron,
Manganese, Chloride
Selenium, Sulfate
Selenium, Arsenic, Chromium,
Cadmium, Lead, Zinc, Aluminum
Arsenic, Chromium, Cadmium,
Lead, Iron
Exceeded
MCLb
X
X
X
X
X
X
X
X
X
Exceeded Federal/
State WQC/
Standards b
X
X
X
Ground Water
Impacted Surface
Waters c
X
X
Impacted
Off-Site
Source d
X
X
X
X
X
X
A-23
-------
Appendix A— Literature Review Methodology and Results
Table A-4. Summary of Ground Water Impacts Reported in Damage Cases from Steam Electric Power Plant Surface
Impoundments
Damage Case Site
Waukegan Generating Station
Ash Ponds, Midwest Generation
(Edison International) (IL)
Weber Ash Disposal Site, AES
Creative Resources (NY)
Westland Disposal Site,
Dickerson Generating Station,
Mirant (MD)
Widows Creek Fossil Plant,
Tennessee Valley Authority
(TVA) (AL)
Winyah Generating Station,
Santee Cooper/South Carolina
Public Service Authority
(SCPSA) (SC)
Wood River Power Station,
Illinois Power Company (IL)
Yorktown Power Station,
Chisman Creek Disposal Site,
Virginia Electric Power and
Power Company (VEPCO) (VA)
Type of Waste in
Impoundment a
Pond/Impoundment,
Landfill
Pond/Impoundment,
Landfill
Pond/Impoundment,
Landfill
Pond/Impoundment
Pond/Impoundment
Pond/Impoundment
Pond/Impoundment,
Landfill
Pollutants of Concern
Arsenic, Antimony, Boron,
Manganese, Sulfate, TDS, Iron
Sulfate, TDS, Manganese, Iron,
Aluminum, pH
Selenium, Arsenic, Barium,
Chromium, Cobalt, Copper, Iron,
Zinc, Sulfate, Chlorine, Hardness,
TDS, Aluminum
Lead, Cobalt, Boron, Iron,
Manganese, Aluminum, Sulfate
Arsenic, Chromium, Sulfate, Iron,
Chloride
Sulfate, TDS, Chloride,
Manganese, Iron, Boron
Sulfate, Nickel, Vanadium,
Selenium
Exceeded
MCLb
X
X
X
X
X
X
Exceeded Federal/
State WQC/
Standards b
X
X
X
Ground Water
Impacted Surface
Waters c
X
Impacted
Off-Site
Source d
Sources: ERG, 2015m; U.S. EPA, 2012e (DCN SE01966); U.S. EPA, 2013b; U.S. EPA, 2014a through 2014e.
Acronyms: FGD (Flue Gas Desulfurization); MCL (Maximum Contaminant Level); TDS (Total Dissolved Solids); WQC (Water Quality Criteria).
a - The term "ash" was used when the impact case study source did not identify the type of ash present at the waste management unit.
b - An "X" indicates that one or more of the pollutants listed exceeded MCLs or federal/state WQC/standards.
c - An "X" indicates that the ground water contaminated the surface water with one or more of the pollutants listed.
d - An "X" indicates that the ground water contaminated a source outside the plant property boundaries.
A-24
-------
Appendix A— Literature Review Methodology and Results
Table A-5. Summary of Ground Water Impacts Reported in Damage Cases from Steam Electric Power Plant Landfills
Damage Case Site
A.B. Brown Generating Station,
Southern Indiana Gas and Electric
Company (SIGECO) (IN)
Alma On-site Fly Ash Landfill,
Dairyland Power (WI)
Bailly Generating Station, Northern
Indiana Public Service Company
(NIPSCO) (IN)
Bangor Quarry Ash Disposal Site,
Portland Generating Station, RRI
Energy (PA)
Battlefield Golf Club, Chesapeake
Energy Facility, Dominion Power
(VA)
BBSS Sand and Gravel Quarries,
Constellation Energy (MD)
Belews Creek Steam Station, Duke
Energy (NC)
Big Bend Station, Tampa Electric
Company (FL)
Brandywine Coal Ash Landfill,
Mirant Mid-Atlantic LLC (MD)
C.D. Mclntosh, Jr. Power Plant, City
of Lakeland (FL)
Type of
Waste in
Landfill a
FGD
Fly Ash
Ash
Bottom Ash,
Fly Ash,
Other
Fly Ash
Fly Ash,
Bottom Ash
Fly Ash,
FGD
Bottom Ash,
Fly Ash,
FGD, Other
Bottom Ash,
Fly Ash
Ash, FGD
Pollutants of Concern
Arsenic, Sodium, Boron, Sulfate, TDS,
Chloride, pH
Sulfate, Manganese
Arsenic, Cadmium
Selenium, Boron, Cadmium, Hexavalent
Chromium, Iron, Manganese, Sulfate,
TDS, Aluminum, Fluoride
Arsenic, Cadmium, Chromium, Copper,
Lead, Manganese, Thallium, Zinc,
Vanadium, Iron, Boron, Aluminum
Arsenic, Selenium, Aluminum, Cadmium,
Thallium, Manganese, Sulfate, Beryllium,
Lead, Nickel
Selenium, Arsenic, Boron, Cadmium,
Iron, Lead, Manganese, Nitrate, Sulfate,
pH, Bromide
Arsenic, Aluminum, Boron, Chloride,
Fluoride, Iron, Manganese, Molybdenum,
Sulfate, Sodium, Thallium, TDS
Selenium, Cadmium, Lead, Manganese,
Iron, Aluminum, Sulfate, TDS, Chloride
Selenium, Arsenic, Cadmium, Lead,
Manganese, Vanadium, Nitrate, Iron,
Sulfate, TDS, pH
Exceeded
MCLb
X
X
X
X
X
X
X
X
X
Exceeded
Federal/ State
WQC/
Standards b
X
X
X
Ground Water
Impacted
Surface Waters c
X
X
Impacted
Off-Site
Source d
X
X
X
X
A-25
-------
Appendix A— Literature Review Methodology and Results
Table A-5. Summary of Ground Water Impacts Reported in Damage Cases from Steam Electric Power Plant Landfills
Damage Case Site
C.R. Huntley Flyash Landfill (NY)
Cardinal Fly Ash Reservoir (FAR) 1
and 2, American Electric Power
(AEP) (OH)
Cayuga Coal Ash Disposal Landfill,
AES(NY)
CCW Landfill, Trans-Ash, Inc. (TN)
Cedar-Sauk Landfill, Wisconsin
Electric Power Company (WEPCO)
(WI)
Clifty Creek Station, Indiana
Kentucky Electric Company (IKEC)
(IN)
Coal Ash Pit #3, Sheldon Station,
Nebraska Public Power District (NE)
Coal Combustion Waste Landfill,
Merom Generating Station, Hoosier
Energy (IN)
Colbert Fossil Plant, Tennessee
Valley Authority (TVA) (AL)
Colstrip Power Plant, PPL Montana
(MT)
Type of
Waste in
Landfill a
Bottom Ash,
Fly Ash,
Other
Bottom Ash,
Fly Ash,
FGD
Bottom Ash,
Fly Ash,
Other
Bottom Ash,
Fly Ash
Fly Ash,
Bottom Ash
Fly Ash,
Other
Fly Ash
Fly Ash,
Bottom Ash
Bottom Ash,
Fly Ash,
Other
Bottom Ash,
Fly Ash,
FGD
Pollutants of Concern
Arsenic, Iron, Manganese, Sulfate, TDS,
Cadmium, Barium, Lead, TSS
Arsenic, Boron, Molybdenum
Selenium, Arsenic, Boron, Cadmium,
Lead, TDS, Aluminum, Manganese,
Sulfate, Barium, Sodium, Iron,
Chromium, Zinc
Mercury, Iron, Boron, Sulfate, Arsenic,
Chromium, Lead
Selenium, Sulfate, Boron
Boron, Manganese, Iron, Sulfate,
Magnesium
Selenium, Sulfate
Barium, Chromium, Cadmium, Lead,
Sulfate, Chloride, Sodium
Cadmium, Antimony, Arsenic, Lead,
Nitrate, Aluminum, Iron, Manganese,
Boron, Molybdenum, Cobalt, Lithium,
Sulfate, Chromium
Selenium, Boron, Sulfate, TDS,
Molybdenum, Arsenic, Chloride
Exceeded
MCLb
X
X
X
X
X
X
X
X
X
X
Exceeded
Federal/ State
WQC/
Standards b
X
X
X
Ground Water
Impacted
Surface Waters c
X
Impacted
Off-Site
Source d
X
X
X
X
A-26
-------
Appendix A— Literature Review Methodology and Results
Table A-5. Summary of Ground Water Impacts Reported in Damage Cases from Steam Electric Power Plant Landfills
Damage Case Site
Conesville Fixed FGD Sludge
Landfill, American Electric Power
(AEP) (OH)
Crist Plant Ash Landfill, Gulf Power
(Southern Company) (FL)
Cross Generating Station, Santee
Cooper/South Carolina Public
Service Authority (SCPSA) (SC)
Cumberland Steam Plant, Tennessee
Valley Authority (TVA) (TN)
Curtis Stanton Energy Center,
Orlando Utility Commission (FL)
Dallman Station Ash and FGD
Ponds, City Water, Light and Power
(IL)
Dan River Steam Station, Duke
Energy (NC)
Danskammer Waste Management
Facility, Central Hudson Gas and
Electric Corporation (NY)
Dave Johnston Power Plant (WY)
Dolet Hills Power Station, Central
Louisiana Electric Co-Op (CLECO)
Power, LLC (LA)
East Bend Scrubber Sludge Landfill,
Cinergy (KY)
Type of
Waste in
Landfill a
Fly Ash,
FGD
Fly ash,
Bottom Ash,
FGD
Bottom Ash,
FGD
Bottom Ash,
Fly Ash,
FGD
Bottom Ash,
Other
Ash, FGD
Bottom Ash,
Fly Ash,
Other
Ash
Fly Ash
Bottom Ash,
Fly Ash,
FGD, Other
FGD
Pollutants of Concern
Arsenic, Cadmium, Chromium, Calcium,
Magnesium, TDS, Sulfate, Iron, Selenium
Arsenic, Cadmium, Manganese,
Chromium, Sodium, Sulfate, Aluminum,
Chlorine, Iron, pH, TDS
Arsenic, Cadmium, Chromium, Sodium,
Sulfate, Iron, Aluminum, Chloride, TDS
Selenium, Arsenic, Aluminum, Boron,
Chloride, Iron, Manganese, Sulfate, TDS,
Vanadium
Aluminum, Chloride, Iron, Manganese,
Sodium, Sulfate, TDS, Vanadium, pH
Arsenic, Chromium, Sodium, Boron,
Manganese, Iron, Sulfate, TDS
Chromium, Iron, Lead, Manganese,
Silver, Sulfate, Arsenic, Antimony,
Boron, TDS, pH
Sulfate, Sulfide, TDS, Turbidity, Iron,
Magnesium, Manganese, Sodium, Boron,
pH
Cadmium, Manganese, Sulfate, Boron
Selenium, Arsenic, Lead, Chloride, TDS,
Sulfate, Iron, pH
TDS, Iron, Sulfate, Manganese, Chloride
Exceeded
MCLb
X
X
X
X
X
X
X
X
Exceeded
Federal/ State
WQC/
Standards b
X
X
X
Ground Water
Impacted
Surface Waters c
Impacted
Off-Site
Source d
A-27
-------
Appendix A— Literature Review Methodology and Results
Table A-5. Summary of Ground Water Impacts Reported in Damage Cases from Steam Electric Power Plant Landfills
Damage Case Site
Edgewater 1-4 Ash Disposal Site,
Alliant (formerly Wisconsin Power &
Light) (WI)
Fair Station Ash Landfill, Central
Iowa Power Cooperative (IA)
Fayette Power Project (Sam
Seymour), Lower Colorado River
Authority (TX)
Fern Valley Landfill, Orion Power
Holdings, Inc. (a subsidiary of RRI
Energy) (PA)
Flint Creek Power Plant, American
Electric Power (AEP)/South West
Electric Power Company (SWEPCO)
(AR)
Fly Ash Landfill, Coffeen/White &
Brewer Trucking (IL)
Fly Ash Landfill, Don Frame
Trucking, Inc. (NY)
General James M. Gavin Power
Plant, American Electric Power/Ohio
Power Company (OH)
George Neal Station North Landfill,
Berkshire Hathaway /MidAmerican
Energy Company (IA)
Type of
Waste in
Landfill a
Ash
Ash
Bottom Ash,
Fly Ash,
FGD, Other
Fly Ash
Bottom Ash,
Fly Ash,
Other
Fly Ash,
FGD, Bottom
Ash
Bottom Ash,
Fly Ash,
Other
Bottom Ash,
Fly Ash,
FGD, Other
Fly Ash
Pollutants of Concern
Boron, Sulfate, Iron, Chloride, TDS,
Arsenic, Selenium
Selenium, Manganese, Sulfate, Iron
Selenium, Aluminum, Chloride, Cobalt,
Manganese, Molybdenum, Sulfate, TDS,
Vanadium
Selenium, Aluminum, Boron, Chloride,
Sulfate, TDS
Selenium, Barium, Cadmium, Chromium,
Iron, Lead, Manganese, pH, Silver,
Sulfate, TDS
Sulfate, TDS, Manganese, Cadmium,
Chromium, Thallium, Beryllium, Boron,
Nickel, Barium, Iron, Zinc, Aluminum,
Sodium
Lead, Sulfate, TDS, Manganese, Iron
Arsenic, Barium, Cadmium, Lead,
Molybdenum, Sulfate, TDS, Aluminum,
Copper, Nickel, Zinc, Manganese,
Chloride
Iron, Manganese, Sulfate, Arsenic
Exceeded
MCLb
X
X
X
X
X
X
X
X
X
Exceeded
Federal/ State
WQC/
Standards b
X
X
X
Ground Water
Impacted
Surface Waters c
X
Impacted
Off-Site
Source d
X
X
X
X
A-28
-------
Appendix A— Literature Review Methodology and Results
Table A-5. Summary of Ground Water Impacts Reported in Damage Cases from Steam Electric Power Plant Landfills
Damage Case Site
George Neal Station South Ash
Monofill, Berkshire
Hathaway /MidAmerican Energy
Company (IA)
Gibson Generating Station, Duke
Energy (IN)
Hatfield's Ferry Power Station,
Allegheny Energy (PA)
Hennepin Power Station, Illinois
Power Company (IL)
Highway 59 Landfill, Wisconsin
Electric Power Company (WEPCO)
(WI)
Independence Steam Station,
Entergy /Arkansas Power and Light
(AR)
Indian River Generating Station,
NRG Energy (DE)
John Warden Ash Site (MI)
Johnsonville Fossil Plant, Tennessee
Valley Authority (TVA) (TN)
K.R. Rezendes South Main Street
Ash Landfill, Salem Harbor and
Brayton Point Plants, Pacific Gas and
Electric (PG&E) (MA)
Type of
Waste in
Landfill a
Bottom Ash,
Fly Ash
Bottom Ash,
Fly Ash
Bottom Ash,
Fly Ash,
FGD
Fly Ash
Bottom Ash,
Fly Ash
Bottom Ash,
Fly Ash,
Other
Ash
Ash, Other
Bottom Ash,
Fly Ash
Ash
Pollutants of Concern
Selenium, Arsenic, Barium, Zinc, Iron,
Manganese, Sulfate
Selenium, Arsenic, Boron, Manganese,
Iron, Sodium
Arsenic, Aluminum, Boron, Chromium,
Manganese, Molybdenum, Thallium,
TDS, Sulfate, Selenium
Sulfate, TDS, Boron, Iron, Manganese
Selenium, Sulfate, Boron, Manganese,
Chloride, Iron, Arsenic, Molybdenum,
TDS
Cadmium, Iron, Lead, Manganese, pH,
Sulfate, TDS, Arsenic, Chlorine
Selenium, Mercury, Arsenic, Aluminum,
Barium, Cadmium, Chromium, Copper,
Lead, Nickel, Thallium, Zinc, Iron,
Manganese
Boron, Lithium
Arsenic, Aluminum, Boron, Cadmium,
Chromium, TDS, Iron, Lead, Manganese,
Molybdenum, Sulfate, Cobalt
Selenium, Arsenic (?)
Exceeded
MCLb
X
X
X
X
X
X
X
X
Exceeded
Federal/ State
WQC/
Standards b
X
X
X
X
X
Ground Water
Impacted
Surface Waters c
X
Impacted
Off-Site
Source d
X
X
X
X
X
X
A-29
-------
Appendix A— Literature Review Methodology and Results
Table A-5. Summary of Ground Water Impacts Reported in Damage Cases from Steam Electric Power Plant Landfills
Damage Case Site
Karn/Weadock Generating Facility,
Consumer Energy (MI)
Lincoln Stone Quarry Landfill, Joliet
Generating Station 9, Midwest
Generation (IL)
Mahoney Landfill, Powerton Plant,
Commonwealth Edison (IL)
Marion Plant, Southern Illinois
Power Cooperative (IL)
McMeekin Station, SCANA/South
Carolina Electric & Gas Company
(SCE&G) (SC)
Miamiview Landfill, Cincinnati Gas
& Electric Company (OH)
Mill Creek Plant, EON
U.S. /Louisville Gas & Electric
(LG&E) (KY)
Mitchell Power Station, Allegheny
Energy (PA)
Morgantown Generating Station,
Faulkner Off-site Disposal Facility
(MD)
Muscatine County Landfill (IA)
Muskegon County Type III Landfill
(MI)
North Lansing Landfill, Lansing
Board of Water & Light (MI)
Type of
Waste in
Landfill a
Ash, Fly
Ash, Bottom
Ash
Ash
Bottom Ash,
Fly Ash,
Other
Bottom Ash,
Fly Ash,
FGD
Ash
FGD
Bottom Ash,
Fly Ash,
FGD, Other
Bottom Ash,
Fly Ash
Bottom Ash,
Fly Ash,
Other
Ash
Fly Ash
Ash, Other
Pollutants of Concern
Arsenic, Boron, Lithium,
Arsenic, Ammonia, Boron, Molybdenum,
pH, Sulfate, TDS, Barium, Copper,
Selenium, Cadmium
Arsenic, Selenium, Chromium, TDS,
Cadmium, Lead, Nitrate, Iron,
Manganese, Sulfate, Boron,
Boron, Cadmium, Iron, Aluminum, TDS,
Sulfate
Chromium, Lead, Sulfate, Iron, TDS
Sulfate, Manganese
Arsenic, Chloride, Sulfate, TDS
Arsenic, Boron, Iron, Molybdenum,
Manganese, Nickel
Iron, pH, Cadmium, Aluminum, Chloride,
Manganese, Sulfate, TDS, Copper, Lead,
Selenium
Selenium, Sulfate
Boron, Manganese
Selenium, Boron, Lithium, Manganese,
Sulfate, Lead
Exceeded
MCLb
X
X
X
X
X
X
X
X
X
X
Exceeded
Federal/ State
WQC/
Standards b
X
X
X
Ground Water
Impacted
Surface Waters c
X
X
Impacted
Off-Site
Source d
X
X
X
X
X
A-30
-------
Appendix A— Literature Review Methodology and Results
Table A-5. Summary of Ground Water Impacts Reported in Damage Cases from Steam Electric Power Plant Landfills
Damage Case Site
Northeastern Station Ash Landfill,
American Electric Power/Public
Service Company Oklahoma (OK)
Parish Generating Station, NRG
Energy /Texas Genco II (TX)
Petersburg Generating Station,
Indianapolis Power & Light (IN)
Phillips Power Plant Landfill,
Duquesne Light Company (PA)
Pine Hill Landfill, Marquette Board
of Light & Power (MI)
Port Washington Facility, Wisconsin
Electric Power Company (WEPCO)
(WI)
Prairie Creek Generating Station Ash
Landfill, Interstate Power and Light
(Alliant) (IA)
Presque Isle Power Plant, WE
Energies (WE) (MI)
Pulliam Ash Disposal Site,
Wisconsin Power Supply Company
(WPSC) (WI)
R.M. Heskett Station, Montana-
Dakota Utilities (ND)
R.M. Schahfer Generating Station
(IN)
Type of
Waste in
Landfill a
Bottom Ash,
Fly Ash
Fly Ash,
Bottom Ash,
FGD
(Emergency
Only)
Not Specified
Ash, FGD
Fly Ash
Bottom Ash,
Fly Ash
Ash
Bottom Ash,
Fly Ash
Bottom Ash,
Fly Ash
Ash
Ash, FGD
Pollutants of Concern
Selenium, Arsenic, Barium, Chromium,
Lead, Vanadium, Thallium, Sulfate, pH
Arsenic, Selenium, Barium, Boron,
Chromium, Cobalt, Manganese,
Molybdenum, Sulfate
Sulfate, TDS
TDS, Chloride, Fluoride, Manganese,
Aluminum, Arsenic
Boron, Lithium, Sodium
Selenium, Boron, Sulfate
Arsenic, Boron, Manganese, Sulfate, Iron
Boron, Molybdenum, Selenium, Sodium,
Sulfate, Lithium
Sulfate, Manganese, Iron, Boron, Zinc,
Aluminum, Chlorine, TDS, pH
Sulfate, Boron, Cadmium, Selenium,
Nitrate
Sulfate, Iron, Manganese, Molybdenum,
Chlorine, Sodium, Boron
Exceeded
MCLb
X
X
X
X
X
£
Exceeded
Federal/ State
WQC/
Standards b
X
X
X
Ground Water
Impacted
Surface Waters c
X
X
Impacted
Off-Site
Source d
X
X
X
A-31
-------
Appendix A— Literature Review Methodology and Results
Table A-5. Summary of Ground Water Impacts Reported in Damage Cases from Steam Electric Power Plant Landfills
Damage Case Site
Range Road Landfill, Detroit Edison
(MI)
Reid Gardner Generating Facility,
Nevada Energy (NV)
Rodemacher Power Station, Central
Louisiana Electric Co-Op (CLECO)
Power, LLC (LA)
Seminole Generating Station,
Seminole Electric Cooperative (FL)
Seward Generating Station, RRI
Energy (PA)
Shawnee Fossil Plant, Tennessee
Valley Authority (TVA) (KY)
Sherburne County (Sherco)
Generating Plant, Xcel
Energy/Southern Minnesota
Municipal Power Agency (MN)
Spurlock Station, Eastern Kentucky
Power Cooperative (KY)
Swift Creek Structural Fill, ReUse
Technology, Inc./ Full Circle
Solutions (NC)
Type of
Waste in
Landfill a
Ash
Fly Ash,
FGD
Bottom Ash,
Fly Ash,
Other
Fly Ash,
FGD, Other
Ash, Other
Bottom Ash,
Fly Ash
Bottom Ash,
Fly Ash,
FGD
Bottom Ash,
Fly Ash,
FGD
Fly Ash
Pollutants of Concern
Boron, Lithium, Manganese
Selenium, Arsenic, Chloride, Sulfate,
TDS, Nitrate, Boron, Chromium,
Manganese, Magnesium, Molybdenum,
Sodium, Vanadium, Titanium, Barium,
Iron, Aluminum
Arsenic, Lead, pH, TDS, Chloride, Sulfate
Arsenic, Chloride, Chlorine, Sulfate, Iron,
TDS, Boron, Aluminum, Lead, Sodium
Selenium, Arsenic, Aluminum, Antimony,
Cadmium, Chloride, Chromium, Iron,
Lead, Manganese, Nickel, pH, Sulfate,
TDS, Zinc,
Selenium, Arsenic, Boron, pH, Sulfate,
TDS, Beryllium, Cobalt, Nickel,
Molybdenum, Manganese, Vanadium
Arsenic, Cadmium, Lead, Sulfate,
Selenium, Boron
Arsenic, Sulfate, TDS
Arsenic, Lead, Sulfate
Exceeded
MCLb
X
X
X
X
X
X
X
X
Exceeded
Federal/ State
WQC/
Standards b
X
X
X
X
X
Ground Water
Impacted
Surface Waters c
X
X
Impacted
Off-Site
Source d
X
X
X
X
X
X
X
A-32
-------
Appendix A— Literature Review Methodology and Results
Table A-5. Summary of Ground Water Impacts Reported in Damage Cases from Steam Electric Power Plant Landfills
Damage Case Site
Urquhart Station, South Carolina
Electric & Gas Company (SGE&E)
(SC)
Wateree Station, SCE&G (SC)
Waukegan Generating Station Ash
Ponds, Midwest Generation (Edison
International) (IL)
Weber Ash Disposal Site, AES
Creative Resources (NY)
Westland Disposal Site, Dickerson
Generating Station, Mirant (MD)
Yard 520 Landfill Site (Brown's
Landfill),Northern Indiana Public
Service Company (NIPSCO) (IN)
Yorktown Power Station, Chisman
Creek Disposal Site, Virginia Electric
Power and Power Company
(VEPCO) (VA)
Type of
Waste in
Landfill a
Fly Ash,
Bottom Ash,
Other
Bottom Ash,
Fly Ash,
FGD
Ash
Ash
Fly Ash
Fly Ash,
Other
Fly Ash
Pollutants of Concern
Arsenic, Nickel
Arsenic, Chromium, Cadmium, Lead, Iron
Arsenic, Antimony, Boron, Manganese,
Sulfate, TDS, Iron
Sulfate, TDS, Manganese, Iron,
Aluminum, pH
Selenium, Arsenic, Barium, Chromium,
Cobalt, Copper, Iron, Zinc, Sulfate,
Chlorine, Hardness, TDS, Aluminum
Arsenic, Manganese, Boron,
Molybdenum, Lead, Selenium, Iron,
Sulfate, Ammonium
Sulfate, Nickel, Vanadium, Selenium
Exceeded
MCLb
X
X
X
X
X
X
X
Exceeded
Federal/ State
WQC/
Standards b
X
X
X
Ground Water
Impacted
Surface Waters c
X
Impacted
Off-Site
Source d
X
X
Sources: ERG, 2015m; U.S. EPA, 2012e (DCN SE01966); U.S. EPA, 2013b; U.S. EPA, 2014a through 2014e.
Acronyms: FGD (Flue Gas Desulfurization); MCL (Maximum Contaminant Level); TDS (Total Dissolved Solids); WQC (Water Quality Criteria).
a - The term "ash" was used when the impact case study source did not identify the type of ash present at the waste management unit.
b - An "X" indicates that one or more of the pollutants listed exceeded MCLs or federal/state WQC/standards.
c - An "X" indicates that the ground water contaminated the surface water with one or more of the pollutants listed.
d - An "X" indicates that the ground water contaminated a source outside the plant property boundaries.
A-33
-------
Appendix A— Literature Review Methodology and Results
Table A-6. Summary of Surface Water Impacts Reported in Damage Cases from Steam Electric Power Plant Surface
Impoundments
Damage Case Site
Asheville Steam Electric
Plant, Progress Energy
(NC)
Bangor Quarry Ash
Disposal Site, Portland
Generating Station, RPJ
Energy (PA)
Belews Creek Steam
Station, Duke Energy
(NC)
Big Bend Station, Tampa
Electric Company (FL)
Brandywine Coal Ash
Landfill, Mirant Mid-
Atlantic LLC (MD)
Bull Run Steam Plant,
Tennessee Valley
Authority (TVA) (TN)
Cardinal Fly Ash
Reservoir (FAR) 1 and 2,
American Electric Power
(AEP) (OH)
Cayuga Coal Ash
Disposal Landfill, AES
(NY)
Type of Waste in
Impoundment a
Pond/Impoundment
Pond/Impoundment,
Landfill
Pond/Impoundment,
Landfill
Pond/Impoundment,
Landfill
Pond/Impoundment,
Landfill
Pond/Impoundment
Pond/Impoundment,
Landfill
Pond/Impoundment,
Landfill
Pollutants of Concern
Boron, Chromium, Iron,
Manganese, Thallium, Nitrate,
Sulfate, pH, TDS, Cadmium,
Arsenic, Antimony
Selenium, Boron, Cadmium,
Hexavalent Chromium, Iron,
Manganese, Sulfate, TDS,
Aluminum, Fluoride
Selenium, Arsenic, Boron,
Cadmium, Iron, Lead, Manganese,
Nitrate, Sulfate, pH, Bromide
Arsenic, Aluminum, Boron,
Chloride, Fluoride, Iron,
Manganese, Molybdenum, Sulfate,
Sodium, Thallium, TDS
Selenium, Cadmium, Lead,
Manganese, Iron, Aluminum,
Sulfate, TDS, Chloride
Aluminum, Cadmium, Iron, Sulfate,
Arsenic, Cobalt, Calcium,
Manganese, Molybdenum, Boron,
Nickel
Arsenic, Boron, Molybdenum
Selenium, Arsenic, Boron,
Cadmium, Lead, TDS, Aluminum,
Manganese, Sulfate, Barium,
Sodium, Iron, Chromium, Zinc
Exceeded
Federal/State
WQC/Standards b
X
X
X
X
X
X
Issued a Fish
Consumption
Advisory c
Impact Resulted
from Ground Water
Contamination d
X
X
X
Impacted
Off-Site
Source e
X
X
X
X
X
A-34
-------
Appendix A— Literature Review Methodology and Results
Table A-6. Summary of Surface Water Impacts Reported in Damage Cases from Steam Electric Power Plant Surface
Impoundments
Damage Case Site
Clinch River Plant,
American Electric Power
(AEP)/Appalachian
Power (VA)
Columbia Energy Center,
Alliant Energy (WI)
Cumberland Steam Plant,
Tennessee Valley
Authority (TVA) (TN)
Curtis Stanton Energy
Center, Orlando Utility
Commission (FL)
Dan River Steam Station,
Duke Energy (NC)
Dave Johnston Power
Plant (WY)
Flint Creek Power Plant,
American Electric Power
(AEPySouth West
Electric Power Company
(SWEPCO) (AR)
General James M. Gavin
Power Plant, American
Electric Power/Ohio
Power Company (OH)
Gibson Generating
Station, Duke Energy
(IN)
Type of Waste in
Impoundment a
Pond/Impoundment
Pond/Impoundment,
Landfill
Pond/Impoundment,
Landfill
Pond/Impoundment,
Landfill
Pond/Impoundment
Pond/Impoundment,
Landfill
Pond/Impoundment,
Landfill
Pond/Impoundment,
Landfill
Pond/Impoundment,
Landfill, Cooling
Reservoir
Pollutants of Concern
Aluminum, pH, Copper
Cadmium, Copper, Barium,
Aluminum, Iron, Zinc, Arsenic,
Selenium, Lead, Manganese
Selenium, Arsenic, Aluminum,
Boron, Chloride, Iron, Manganese,
Sulfate, TDS, Vanadium
Aluminum, Chloride, Iron,
Manganese, Sodium, Sulfate, TDS,
Vanadium, pH
Arsenic, Copper, Iron, Aluminum
Cadmium, Manganese, Sulfate,
Boron
Selenium, Barium, Cadmium,
Chromium, Iron, Lead, Manganese,
pH, Silver, Sulfate, TDS
Arsenic, Barium, Cadmium, Lead,
Molybdenum, Sulfate, TDS,
Aluminum, Copper, Nickel, Zinc,
Manganese, Chloride
Selenium, Arsenic, Boron,
Manganese, Iron, Sodium
Exceeded
Federal/State
WQC/Standards b
X
X
X
X
X
X
X
X
Issued a Fish
Consumption
Advisory c
X
Impact Resulted
from Ground Water
Contamination d
Impacted
Off-Site
Source e
X
X
X
X
A-35
-------
Appendix A— Literature Review Methodology and Results
Table A-6. Summary of Surface Water Impacts Reported in Damage Cases from Steam Electric Power Plant Surface
Impoundments
Damage Case Site
Glen Lyn Plant,
American Electric Power
(AEP)/Appalachian
Power (VA)
Grainger Generating
Station, Santee
Cooper/South Carolina
Public Service Authority
(SCPSA) (SC)
J.H. Campbell,
Consumers Energy (MI)
J.R. Whiting Generating
Plant, CMS/Consumers
Energy (MI)
John Sevier Fossil Plant,
Tennessee Valley
Authority (TVA) (TN)
Johnsonville Fossil Plant,
Tennessee Valley
Authority (TVA) (TN)
Kingston Fossil Plant,
Tennessee Valley
Authority (TVA) (TN)
Type of Waste in
Impoundment a
Pond/Impoundment
Pond/Impoundment
Pond/Impoundment
Pond/Impoundment
Pond/Impoundment
Pond/Impoundment,
Landfill
Pond/Impoundment
Pollutants of Concern
Selenium, Cadmium, Copper,
Chromium, Zinc, pH, Nickel
Arsenic, pH
pH, Antimony, Boron, Cadmium,
Chromium, Iron, Lead, Selenium,
Vanadium, Aluminum, Nickel,
Thallium, Manganese, Zinc
Selenium, Arsenic, Cobalt, Nickel,
Bromine, Chromium
Arsenic, Aluminum, Cadmium,
Manganese, Boron, Strontium,
Sulfate, Selenium, Hexavalent
Chromium
Arsenic, Aluminum, Boron,
Cadmium, Chromium, TDS, Iron,
Lead, Manganese, Molybdenum,
Sulfate, Cobalt
Arsenic, Selenium, Manganese,
Cobalt, Aluminum, Ammonia,
Thallium, Iron
Exceeded
Federal/State
WQC/Standards b
X
X
X
X
X
Issued a Fish
Consumption
Advisory c
X
Impact Resulted
from Ground Water
Contamination d
X
X
Impacted
Off-Site
Source e
X
X
X
X
A-36
-------
Appendix A— Literature Review Methodology and Results
Table A-6. Summary of Surface Water Impacts Reported in Damage Cases from Steam Electric Power Plant Surface
Impoundments
Damage Case Site
Little Blue Run Surface
Impoundment, Bruce
Mansfield Power Plant,
First Energy (PA)
Little Scary Creek Fly
Ash Impoundment, John
Amos Plant, American
Electric Power
(AEP)/Appalachian
Power (WV)
Mahoney Landfill,
Powerton Plant,
Commonwealth Edison
(IL)
Marion Plant, Southern
Illinois Power
Cooperative (IL)
Martin's Creek Power
Plant, PPL (PA)
Montville Generating
Station, NRG
Energy /Montville Power,
LLC (CT)
Morgantown Generating
Station, Faulkner Off-site
Disposal Facility (MD)
Type of Waste in
Impoundment a
Pond/Impoundment
Pond/Impoundment
Pond/Impoundment,
Landfill
Pond/Impoundment,
Landfill
Pond/Impoundment
Pond/Impoundment
Pond/Impoundment,
Landfill
Pollutants of Concern
Selenium, Arsenic, Aluminum,
Antimony, Barium, Boron,
Cadmium, Calcium, Chloride,
Hexavalent Chromium, Fluoride,
Iron, Lead, Manganese, pH,
Sodium, Sulfate, TDS, TSS,
Thallium, Turbidity
Selenium, Mercury, Arsenic,
Copper
Arsenic, Selenium, Chromium,
TDS, Cadmium, Lead, Nitrate, Iron,
Manganese, Sulfate, Boron,
Boron, Cadmium, Iron, Aluminum,
TDS, Sulfate
Arsenic, Selenium, Lead,
Aluminum, Copper, Chromium,
Iron
Arsenic, Beryllium, Cadmium,
Copper, Iron, Lead, Manganese,
Nickel, pH, Zinc
Iron, pH, Cadmium, Aluminum,
Chloride, Manganese, Sulfate, TDS,
Copper, Lead, Selenium
Exceeded
Federal/State
WQC/Standards b
X
X
X
X
X
X
Issued a Fish
Consumption
Advisory c
Impact Resulted
from Ground Water
Contamination d
X
X
Impacted
Off-Site
Source e
X
X
X
X
X
A-37
-------
Appendix A— Literature Review Methodology and Results
Table A-6. Summary of Surface Water Impacts Reported in Damage Cases from Steam Electric Power Plant Surface
Impoundments
Damage Case Site
Oak Ridge Y-12 Plant,
Chestnut Ridge Operable
Unit 2, Oak Ridge
Reservation, Department
of Energy (TN)
Phillips Power Plant
Landfill, Duquesne Light
Company (PA)
Plant Bowen, Georgia
Power (GA)
Reid Gardner Generating
Facility, Nevada Energy
(NV)
Savannah River Site, D-
Area, Department of
Energy (SC)
Seminole Generating
Station, Seminole
Electric Cooperative (FL)
Seward Generating
Station, RRI Energy (PA)
Shawnee Fossil Plant,
Tennessee Valley
Authority (TVA) (KY)
Type of Waste in
Impoundment a
Pond/Impoundment
Pond/Impoundment,
Landfill
Pond/Impoundment
Pond/Impoundment,
Landfill
Pond/Impoundment
Pond/Impoundment,
Landfill
Pond/Impoundment,
Landfill
Pond/Impoundment,
Landfill
Pollutants of Concern
Selenium, Arsenic, Aluminum, Iron,
Zinc, Manganese, Thallium (?)
TDS, Chloride, Fluoride,
Manganese, Aluminum, Arsenic
Arsenic, Cadmium, Chromium,
Lead, Mercury, Nickel, Copper
Selenium, Arsenic, Chloride,
Sulfate, TDS, Nitrate, Boron,
Chromium, Manganese,
Magnesium, Molybdenum, Sodium,
Vanadium, Titanium, Barium, Iron,
Aluminum
Cadmium, Chromium, Copper,
Mercury, Selenium, Zinc, Iron,
Aluminum
Arsenic, Chloride, Chlorine, Sulfate,
Iron, TDS, Boron, Aluminum, Lead,
Sodium
Selenium, Arsenic, Aluminum,
Antimony, Cadmium, Chloride,
Chromium, Iron, Lead, Manganese,
Nickel, pH, Sulfate, TDS, Zinc,
Selenium, Arsenic, Boron, pH,
Sulfate, TDS, Beryllium, Cobalt,
Nickel, Molybdenum, Manganese,
Vanadium
Exceeded
Federal/State
WQC/Standards b
X
X
X
X
X
X
Issued a Fish
Consumption
Advisory c
Impact Resulted
from Ground Water
Contamination d
X
X
X
X
Impacted
Off-Site
Source e
X
X
X
X
X
X
A-38
-------
Appendix A— Literature Review Methodology and Results
Table A-6. Summary of Surface Water Impacts Reported in Damage Cases from Steam Electric Power Plant Surface
Impoundments
Damage Case Site
Sutton Steam Plant,
Progress Energy (NC)
W.J. Neal Station Surface
Impoundment, Basin
Electric Power
Cooperative (ND)
Wateree Station, SCE&G
(SC)
Westland Disposal Site,
Dickerson Generating
Station, Mirant (MD)
Type of Waste in
Impoundment a
Pond/Impoundment
Pond/Impoundment
Pond/Impoundment,
Landfill
Pond/Impoundment,
Landfill
Pollutants of Concern
Arsenic, Boron, Manganese, Iron,
Thallium, Selenium, Antimony,
Lead, Sulfate, TDS
Selenium, Arsenic, Chromium,
Cadmium, Lead, Zinc, Aluminum
Arsenic, Chromium, Cadmium,
Lead, Iron
Selenium, Arsenic, Barium,
Chromium, Cobalt, Copper, Iron,
Zinc, Sulfate, Chlorine, Hardness,
TDS, Aluminum
Exceeded
Federal/State
WQC/Standards b
X
X
X
Issued a Fish
Consumption
Advisory c
Impact Resulted
from Ground Water
Contamination d
X
X
Impacted
Off-Site
Source e
X
X
X
Sources: ERG, 2015m; U.S. EPA, 2012e (DCN SE01966); U.S. EPA, 2013b; U.S. EPA, 2014a through 2014e.
Acronyms: FGD (Flue Gas Desulfurization); TDS (Total Dissolved Solids); TOC (Total Organic Carbon); TOH (Total Organic Hydrocarbons); TSS (Total
Suspended Solids); WQC (Water Quality Criteria).
a - The term "ash" was used when the impact case study source did not identify the type of ash present at the waste management unit.
b - An "X" indicates that one or more of the pollutants listed exceeded federal/state WQC/standards.
c - An "X" indicates that the contaminated surface water was issued a fish consumption advisory.
d - An "X" indicates that the ground water contaminated the surface water with one or more of the pollutants listed.
e - An "X" indicates that the surface water contaminated a source outside the plant property boundaries.
A-39
-------
Appendix A— Literature Review Methodology and Results
Table A-7. Summary of Surface Water Impacts Reported in Damage Cases from Steam Electric Power Plant Landfills
Damage Case Site
Bangor Quarry Ash Disposal
Site, Portland Generating
Station, RRI Energy (PA)
Battlefield Golf Club,
Chesapeake Energy Facility,
Dominion Power (VA)
Belews Creek Steam Station,
Duke Energy (NC)
Big Bend Station, Tampa
Electric Company (FL)
Brandywine Coal Ash
Landfill, Mirant Mid-
Atlantic LLC (MD)
Cardinal Fly Ash Reservoir
(FAR) 1 and 2, American
Electric Power (AEP) (OH)
Cayuga Coal Ash Disposal
Landfill, AES (NY)
CCW Landfill, Trans-Ash,
Inc. (TN)
Coal Combustion Waste
Landfill, Merom Generating
Station, Hoosier Energy (IN)
Columbia Energy Center,
Alliant Energy (WI)
Type of
Waste in
Landfill a
Bottom Ash,
Fly Ash, Other
Fly Ash
Fly Ash, FGD
Bottom Ash,
Fly Ash, FGD,
Other
Bottom Ash,
Fly Ash
Bottom Ash,
Fly Ash, FGD
Bottom Ash,
Fly Ash, Other
Bottom Ash,
Fly Ash
Fly Ash,
Bottom Ash
Bottom Ash,
Fly Ash
Pollutants of Concern
Selenium, Boron, Cadmium, Hexavalent
Chromium, Iron, Manganese, Sulfate,
TDS, Aluminum, Fluoride
Arsenic, Cadmium, Chromium, Copper,
Lead, Manganese, Thallium, Zinc,
Vanadium, Iron, Boron, Aluminum
Selenium, Arsenic, Boron, Cadmium,
Iron, Lead, Manganese, Nitrate, Sulfate,
pH, Bromide
Arsenic, Aluminum, Boron, Chloride,
Fluoride, Iron, Manganese, Molybdenum,
Sulfate, Sodium, Thallium, TDS
Selenium, Cadmium, Lead, Manganese,
Iron, Aluminum, Sulfate, TDS, Chloride
Arsenic, Boron, Molybdenum
Selenium, Arsenic, Boron, Cadmium,
Lead, TDS, Aluminum, Manganese,
Sulfate, Barium, Sodium, Iron,
Chromium, Zinc
Mercury, Iron, Boron, Sulfate, Arsenic,
Chromium, Lead
Barium, Chromium, Cadmium, Lead,
Sulfate, Chloride, Sodium
Cadmium, Copper, Barium, Aluminum,
Iron, Zinc, Arsenic, Selenium, Lead,
Manganese
Exceeded
Federal/ State
WQC/
Standards b
X
X
X
X
X
X
Issued a Fish
Consumption
Advisory c
Impact Resulted
from Ground
Water
Contamination d
X
X
X
Impacted
Off-Site
Source e
X
X
X
X
X
X
A-40
-------
Appendix A— Literature Review Methodology and Results
Table A-7. Summary of Surface Water Impacts Reported in Damage Cases from Steam Electric Power Plant Landfills
Damage Case Site
Cumberland Steam Plant,
Tennessee Valley Authority
(TVA) (TN)
Curtis Stanton Energy
Center, Orlando Utility
Commission (FL)
Dave Johnston Power Plant
(WY)
Fern Valley Landfill, Orion
Power Holdings, Inc. (a
subsidiary of RPJ Energy)
(PA)
Flint Creek Power Plant,
American Electric Power
(AEP)/South West Electric
Power Company (SWEPCO)
(AR)
General James M. Gavin
Power Plant, American
Electric Power/Ohio Power
Company (OH)
Gibson Generating Station,
Duke Energy (IN)
Hatfield's Ferry Power
Station, Allegheny Energy
(PA)
Indian River Generating
Station, NRG Energy (DE)
Type of
Waste in
Landfill a
Bottom Ash,
Fly Ash, FGD
Bottom Ash,
Other
Fly Ash
Fly Ash
Bottom Ash,
Fly Ash, Other
Bottom Ash,
Fly Ash, FGD,
Other
Bottom Ash,
Fly Ash
Bottom Ash,
Fly Ash, FGD
Ash
Pollutants of Concern
Selenium, Arsenic, Aluminum, Boron,
Chloride, Iron, Manganese, Sulfate, TDS,
Vanadium
Aluminum, Chloride, Iron, Manganese,
Sodium, Sulfate, TDS, Vanadium, pH
Cadmium, Manganese, Sulfate, Boron
Selenium, Aluminum, Boron, Chloride,
Sulfate, TDS
Selenium, Barium, Cadmium, Chromium,
Iron, Lead, Manganese, pH, Silver,
Sulfate, TDS
Arsenic, Barium, Cadmium, Lead,
Molybdenum, Sulfate, TDS, Aluminum,
Copper, Nickel, Zinc, Manganese,
Chloride
Selenium, Arsenic, Boron, Manganese,
Iron, Sodium
Arsenic, Aluminum, Boron, Chromium,
Manganese, Molybdenum, Thallium,
TDS, Sulfate, Selenium
Selenium, Mercury, Arsenic, Aluminum,
Barium, Cadmium, Chromium, Copper,
Lead, Nickel, Thallium, Zinc, Iron,
Manganese
Exceeded
Federal/ State
WQC/
Standards b
X
X
X
X
X
X
X
X
Issued a Fish
Consumption
Advisory c
Impact Resulted
from Ground
Water
Contamination d
X
Impacted
Off-Site
Source e
X
X
X
X
X
A-41
-------
Appendix A— Literature Review Methodology and Results
Table A-7. Summary of Surface Water Impacts Reported in Damage Cases from Steam Electric Power Plant Landfills
Damage Case Site
John Warden Ash Site (MI)
Johnsonville Fossil Plant,
Tennessee Valley Authority
(TVA) (TN)
Mahoney Landfill, Powerton
Plant, Commonwealth
Edison (IL)
Marion Plant, Southern
Illinois Power Cooperative
(IL)
Morgantown Generating
Station, Faulkner Off-site
Disposal Facility (MD)
Oak Creek Power Plant,
Wisconsin Energy (WE
Energies (WE))/Wisconsin
Electric Power Company
(WI)
Phillips Power Plant
Landfill, Duquesne Light
Company (PA)
Range Road Landfill, Detroit
Edison (MI)
Reid Gardner Generating
Facility, Nevada Energy
(NV)
Type of
Waste in
Landfill a
Ash, Other
Bottom Ash,
Fly Ash
Bottom Ash,
Fly Ash, Other
Bottom Ash,
Fly Ash, FGD
Bottom Ash,
Fly Ash, Other
Bottom Ash,
Fly Ash, FGD,
Other
Ash, FGD
Ash
Fly Ash, FGD
Pollutants of Concern
Boron, Lithium
Arsenic, Aluminum, Boron, Cadmium,
Chromium, TDS, Iron, Lead, Manganese,
Molybdenum, Sulfate, Cobalt
Arsenic, Selenium, Chromium, TDS,
Cadmium, Lead, Nitrate, Iron,
Manganese, Sulfate, Boron,
Boron, Cadmium, Iron, Aluminum, TDS,
Sulfate
Iron, pH, Cadmium, Aluminum,
Chloride, Manganese, Sulfate, TDS,
Copper, Lead, Selenium
Arsenic, Chromium, TCE, Diesel Fuel
TDS, Chloride, Fluoride, Manganese,
Aluminum, Arsenic
Boron, Lithium, Manganese
Selenium, Arsenic, Chloride, Sulfate,
TDS, Nitrate, Boron, Chromium,
Manganese, Magnesium, Molybdenum,
Sodium, Vanadium, Titanium, Barium,
Iron, Aluminum
Exceeded
Federal/ State
WQC/
Standards b
X
X
X
X
X
X
X
Issued a Fish
Consumption
Advisory c
Impact Resulted
from Ground
Water
Contamination d
X
X
X
X
X
Impacted
Off-Site
Source e
X
X
X
X
X
X
X
A-42
-------
Appendix A— Literature Review Methodology and Results
Table A-7. Summary of Surface Water Impacts Reported in Damage Cases from Steam Electric Power Plant Landfills
Damage Case Site
Seminole Generating
Station, Seminole Electric
Cooperative (FL)
Seward Generating Station,
RRI Energy (PA)
Shawnee Fossil Plant,
Tennessee Valley Authority
(TVA) (KY)
Wateree Station, SCE&G
(SC)
Westland Disposal Site,
Dickerson Generating
Station, Mirant (MD)
Type of
Waste in
Landfill a
Fly Ash, FGD,
Other
Ash, Other
Bottom Ash,
Fly Ash
Bottom Ash,
Fly Ash, FGD
Fly Ash
Pollutants of Concern
Arsenic, Chloride, Chlorine, Sulfate, Iron,
TDS, Boron, Aluminum, Lead, Sodium
Selenium, Arsenic, Aluminum,
Antimony, Cadmium, Chloride,
Chromium, Iron, Lead, Manganese,
Nickel, pH, Sulfate, TDS, Zinc,
Selenium, Arsenic, Boron, pH, Sulfate,
TDS, Beryllium, Cobalt, Nickel,
Molybdenum, Manganese, Vanadium
Arsenic, Chromium, Cadmium, Lead,
Iron
Selenium, Arsenic, Barium, Chromium,
Cobalt, Copper, Iron, Zinc, Sulfate,
Chlorine, Hardness, TDS, Aluminum
Exceeded
Federal/ State
WQC/
Standards b
X
X
X
X
Issued a Fish
Consumption
Advisory c
Impact Resulted
from Ground
Water
Contamination d
X
X
Impacted
Off-Site
Source e
X
X
X
X
Sources: ERG, 2015m; U.S. EPA, 2012e (DCN SE01966); U.S. EPA, 2013b; U.S. EPA, 2014a through 2014e.
Acronyms: FGD (Flue Gas Desulfurization); TDS (Total Dissolved Solids); TOC (Total Organic Carbon); TOH (Total Organic Hydrocarbons); TSS (Total
Suspended Solids); WQC (Water Quality Criteria).
a - The term "ash" was used when the impact case study source did not identify the type of ash present at the waste management unit.
b - An "X" indicates that one or more of the pollutants listed exceeded federal/state WQC/standards.
c - An "X" indicates that the contaminated surface water was issued a fish consumption advisory.
d - An "X" indicates that the ground water contaminated the surface water with one or more of the pollutants listed.
e - An "X" indicates that the surface water contaminated a source outside the plant property boundaries.
A-43
-------
Appendix A— Literature Review Methodology and Results
Table A-8. Summary of Attractive Nuisances Related to Steam Electric Power Plants
Species
Common Crackles
(Quiscalus quiscala)
Raccoons
(Procyon lotor)
Interior Least Tern
(Sterna antillarum)
Southern Toads
(Bufo terrestris)
Southern Toads
(Bufo terrestris)
Attractive Nuisance
Site Description
Nested in close
proximity to a coal-fired
power plant's fly ash
pond.
Lived in close proximity
to a coal-fired power
plant' sash pond.
Nested on a dike in a
coal-fired power plant's
ash pond.
• Inhabited an ash
basin and nearby
swamp.
• Reference (control)
site organisms were
transferred to
contaminated
locations.
• Inhabited an ash pond
and nearby swamp.
• Reference site
organisms were
transferred to
contaminated
locations.
Pollutant Concentrations
in the Environment or
Diet
Not measured in study
Not measured in study
Not measured in study
Not measured in study
Pond sediment =39.64
ug/g arsenic, 4.38 ug/g
selenium
Pollutant Concentrations
in the Organism
(Mg/g)
Eggs = 5.9 selenium
• Heart = 2.8 arsenic
• Kidney = 3.2 cadmium,
0.43 strontium
• Muscle = 0.95 chromium
• Liver = 0.34 lead, 1.5
mercury
Not observed in study
Not measured in study
Adult males = 1.58 arsenic,
17.40 selenium
Observed Effects
Histopathological
Histopathological
Not observed in study
Elevated
corticosterone and
testosterone levels
Histopathological
Study Type
Field
Field
Field
Outdoor
mesocosm
Outdoor
mesocosm
Citation
Bryane/a/.,
2003
Burger et al,
2002
Pruitt, 2000
and Duke
Energy, 2007
Hopkins et al. ,
1997
Hopkins et al.,
1998
A-44
-------
Appendix A— Literature Review Methodology and Results
Table A-8. Summary of Attractive Nuisances Related to Steam Electric Power Plants
Species
Larval Bullfrogs
(Rana catesbeiana)
Eastern Narrow-
Mouth Toads
(Gastrophryne
carolinensis)
Barn Swallow
(Himndo mstica)
Slider Turtles
(Trachemys scripta)
Canada Geese
(Branta Canadensis)
Attractive Nuisance
Site Description
Inhabited bottom ash
ponds near a coal-fired
power plant.
Inhabited a selenium-
laden site located near a
coal-fired power plant.
Nested near a selenium-
laden pond associated
with a coal-fired power
plant.
• Inhabited a selenium-
laden basin that
receives fly ash
transport water near a
coal-fired power
plant.
• Eggs were incubated
in ash-contaminated
soil.
Inhabited pens near a
vanadium-laden ash
pond associated with an
oil-fired power plant
Pollutant Concentrations
in the Environment or
Diet
Pond sediment = 49.39
ug/g arsenic, 0.72 ug/g
cadmium, 23.85 ug/g
chromium, 84.72 ug/g
copper, 6.11 ug/g selenium,
106.39 ug/g strontium,
45.83 ug/g vanadium
• Site water =3. 93 ug/L
selenium
• Soil = 38.25 ug/L
selenium
• Lab water = 0.28 ug/L
selenium
Not provided in the
literature
Ash-contaminated soil =
2.56 ug/g selenium
Site water = 467,000 ug/L
vanadium
Pollutant Concentrations
in the Organism
(Mg/g)
Whole body concentration
= 33. 10 arsenic, 5.47
cadmium, 18.25 chromium,
116.72 copper, 20.25
selenium, 39.89 strontium,
17.32 vanadium
• Females = 42.40
selenium
• Eggs = 43.96 selenium
Eggs = 2.8 selenium
Adult Females = 37. 18
(mean concentration),
selenium
• Liver = 57.3 vanadium
• Kidney = 226 vanadium
Observed Effects
• Morphological
• Decreased
swimming speeds
• Reproductive
• Histopathological
Histopathological
Reproductive
• Lethal
• Histopathological
Study Type
Field
Outdoor
mescosm
Field
Outdoor
mescosm
Outdoor
mesocosm
Citation
Hopkins et al. ,
2000
Hopkins et al. ,
2006
King et al. ,
1994
Nagle et al.,
2001
Rattnere/a/.,
2006
Acronyms: ug/g (Micrograms per Grams); ug/L (Micrograms per Liters).
A-45
-------
Appendix A— Literature Review Methodology and Results
Table A-9. Summary of Attractive Nuisances Unrelated to Steam Electric Power Plants
Site Name, Location, and
Contamination Source
Kesterson Reservoir, CA
Agricultural Runoff
Kesterson Reservoir, CA
Agricultural Runoff
Kesterson Reservoir, CA
Agricultural Runoff
Kesterson Reservoir, CA
Agricultural Runoff
Kesterson Reservoir, CA
Agricultural Runoff
Kesterson Reservoir, CA
Agricultural Runoff
Liberty State Park, NJ
Industrial and Urban
Activities
Organism Affected
California Vole (Microtus
californicus)
American Coot (Fulica americana),
Mallard (Anas platyrhynchos)
Pied-Billed Grebes (Podilymbus
podiceps),
Common Moorhen (Gallinula
chloropus),
Black-Necked Stilts (Himantopus
mexicanus)
Gopher Snakes (Pituophis
melanoleucus), Bullfrogs (Rana
catesbeiana)
Eared Grebe (podiceps nigricollis),
Mallard (Anas platyrhynchos),
Cinnamon Teal (Anas cyanoptera),
Gadwall (Anas strepera),
American Coot (Fulica americana),
Killdeer (Charadrius vociferous),
Black-Necked Stilt (Himantopus
mexicanus),
American Avocet (Recurvirostra
americana)
Mosquitofish (Gambusia
affmis),
American Coot (Fulica americana),
Ducks (Anas spp.)
House Wren (troglodytes aedon),
American Robin (Turdus
migratorus)
Documented Effects
Mean selenium concentrations in livers
were significantly elevated.
Mean selenium concentrations in bird
eggs and livers were elevated;
organisms exhibited severe reproductive
failure and deformities.
Selenium concentrations in livers were
10 times those found in nearby control
areas; organisms exhibited severe
lesions and embryonic deformities.
Selenium concentrations in snake and
frog livers were significantly elevated.
Hatchlings exhibited mortality,
deformity, and lack of embryonic
development.
Selenium concentrations in livers,
kidneys, and muscles were elevated;
organisms exhibited reduced body
weight.
Lead, arsenic, chromium, copper, and
iron concentrations in bird feathers were
elevated.
Trace Pollutant Concentrations
(ppm)
Liver =119 selenium
• Eggs = 2.2-1 10 selenium
• Liver =19-130 selenium
• Water = 300,000 selenium
• Liver = 94.4 selenium
• Water = 300,000 selenium
• Snake liver =11.1 selenium
• Frog liver = 45.0 selenium
Water =300 selenium
• Fish = 120 - 140 selenium
• Coot liver = 76.7 selenium
• Duck liver = 25.2 selenium
Feather = 4,200 lead; 1,000
chromium; 6,200 copper; 600
arsenic
Citation
Clarke/ al., 1987
Ohlendorf et al. ,
1986
Ohlendorf et al. ,
1988a
Ohlendorf et al. ,
1988b
Ohlendorf et al. ,
1989
Ohlendorf et al. ,
1990
Hofere/a/.,2010
A-46
-------
Appendix A— Literature Review Methodology and Results
Table A-9. Summary of Attractive Nuisances Unrelated to Steam Electric Power Plants
Site Name, Location, and
Contamination Source
Organism Affected
Documented Effects
Trace Pollutant Concentrations
(ppm)
Citation
Meadowlands, NJ
Industrial and Urban
Activities
Red-winged blackbird (agelaius
phoeniceus), marsh wrens
(Cistothoms palustris), tree swallow
(Tachycineta bicolor)
Lead and chromium concentrations in
blood were elevated; mercury and
chromium concentrations in eggs were
elevated.
Swallow blood = 0.94 lead;
1.03 chromium
Wren eggs = 0.2 mercury
Blackbird eggs = 0.12
chromium
Tsipourae/a/.,2008
Acronym: ppm (parts per million).
A-47
-------
Appendix A— Literature Review Methodology and Results
Table A-10. Summary of Selenium Concentrations in the Environment and Organisms Experiencing Adverse Effects
Plant Name
Belews Creek
Steam Station,
Duke Energy
(NC)
D-Area Coal-
Fired Power
Plant,
Savannah
River Site
(SRS) (SC)
Species
Striped bass
(Morone
saxatilis)
Largemouth bass
(Microptems
salmoides) a
Pomoxis spp.
Lepomis spp. b
Lealums spp. °
Largemouth bass
(Microptems
salmoides)
Green sunfish
(Lepomis
cyanellus)
Banded water
snakes
(Nerodia
fasciata)
Route of Selenium
Exposure
Consumed a selenium-laden
diet by eating red shiners
collected from a site
receiving coal ash pond
sluice water.
Inhabited a selenium-laden
cooling water reservoir
receiving both fly ash and
bottom ash pond effluent
from a coal-fired power
plant.
Inhabited a selenium-laden
cooling water reservoir
receiving effluent from the
coal ash pond.
Inhabited a selenium-laden
lake receiving coal fly ash
sluice water.
Consumed a selenium-laden
diet by eating prey collected
from a contaminated site
located near a coal-fired
power plant.
Selenium
Concentrations in the
Environment (jig/L)
or Diet (jig/g)
Red Shiners = 9.6
ug/g (average whole-
body concentration),
wet
Site water d = 10 ug/L
Ash effluent = 100-
200 ug/L
Site water =10 ug/L
Site water = 13 ug/L
Sediment =5-14
M-g/g, dry
Prey items f = 22.7
ug/g (geometric least
squared mean), dry
Selenium
Concentrations in the
Organism (jig/g)
Skeletal muscle = 3.8
(higher average
concentration), wet
Biomasse = 0.1-1.0
(mean)
Body = 41.0 -77.1
(54.6 mean
concentration), wet
Body = 0.3 1-15.5
(6.32 mean
concentration), wet
Visceral tissue = 40+
(highest mean
concentration), wet
Liver = 21.4, wet
Skeletal muscle =
12.9, wet
Hematocrit =33, wet
Gonads = 17.64
(female), 19.06 (male)
Kidney = 25.38
(female), 32.04 (male)
Liver = 24.08
(female), 24.22 (male)
Observed Effects
Modified
behavior
Decreased growth
Histopathological
Lethal
Lethal
Reproductive
Lethal
Histopathological
Hematological
Reproductive
Histopathological
Study Type
(Surface
Water Type)
Laboratory
(reservoir)
Field
(reservoir)
Field
(reservoir)
Field
(lake)
Laboratory
(not
specified)
Citation
Coughlan
and Velte,
1989
Cumbie and
Van Horn,
1978
Lemly,
1985a
Sorensen et
al., 1984b
Hopkins et
al., 2002
A-48
-------
Appendix A— Literature Review Methodology and Results
Table A-10. Summary of Selenium Concentrations in the Environment and Organisms Experiencing Adverse Effects
Plant Name
Roxboro Plant,
Progress
Energy (NC)
Species
Eastern narrow-
mouth toads
(Gastrophryne
carolinensis)
Slider turtles
(Trachemys
scripta)
Largemouth bass
(Microptems
salmoides)
Bluegill
(Lepomis
macrochims)
Bluegill
(Lepomis
macrochims)
Green sunfish h
(Lepomis
cyanellus)
Route of Selenium
Exposure
Inhabited a selenium-laden
site located near a coal-fired
power plant.
Inhabited a selenium-laden
pond receiving sluiced fly
ash near a coal-fired power
plant.
Eggs were incubated in ash-
contaminated soil.
Inhabited a selenium-laden
cooling water reservoir
receiving ash pond effluent
from a coal-fired power
plant.
Inhabited a selenium-laden
reservoir receiving coal ash
pond effluent.
Inhabited a selenium-laden
reservoir receiving coal ash
pond effluent.
Selenium
Concentrations in the
Environment (jig/L)
or Diet (jig/g)
Site water =3. 93 ug/L
Soil = 8.25 ug/L
Lab water = 0.28 ug/L
Ash-contaminated soil
= 2.56 ug/g, dry
Not provided in the
literature.
Not provided in the
literature
Site water ' = 7 - 14
ug/L
Selenium
Concentrations in the
Organism (jig/g)
Females = 42.40
Eggs = 43. 96
Adult females = 37. 18
(mean concentration),
dry
Carcass = 2.86 (mean,
female), 2.63 (mean,
male)
Gonad = 4.40 (mean,
female), 2.38 (mean,
male)
Carcass = 2.74 (mean,
female), 4.64 (mean,
male)
Gonad = 4.63 (mean,
female), 3.35 (mean,
male)
Not provided in the
literature
Biomass J = 2,744 -
3,793 (mean)
Observed Effects
Reproductive
Histopathological
Reproductive
Reproductive
Histopathological
Lethal
Lethal
Reproductive
Study Type
(Surface
Water Type)
Outdoor
mescosm
(combustion
residuals
pond)
Outdoor
mescosm
(combustion
residuals
pond)
Field
(reservoir)
Field
(reservoir) g
Field
(reservoir)
Citation
Hopkins et
al, 2006
Nagle et al.,
2001
Baumann
and
Gillespie,
1986
Crutchfield
and
Ferguson,
2000a
Crutchfield,
2000b
A-49
-------
Appendix A— Literature Review Methodology and Results
Table A-10. Summary of Selenium Concentrations in the Environment and Organisms Experiencing Adverse Effects
Plant Name
Martin Lake
Steam Station,
Texas Utilities
Electric
Service
Company (TX)
Species
Bluegill
(Lepomis
macrochims)
Bluegill '
(Lepomis
macrochims)
Largemouth bass
m (Micropterus
salmoides)
Green sunfish
(Lepomis
cyanellus)
Redear sunfish
(Lepomis
microlophus)
Redear sunfish
(Lepomis
microlophus)
Route of Selenium
Exposure
Inhabited a selenium-laden
cooling water reservoir of a
coal-fired power plant.
Inhabited a selenium-laden
cooling water reservoir of a
coal-fired power plant.
Inhabited a selenium-laden
lake receiving coal fly ash,
scrubber sludge, and coal
bottom ash.
Inhabited a selenium-laden
lake receiving coal fly ash,
scrubber sludge, and coal
bottom ash.
Selenium
Concentrations in the
Environment (jig/L)
or Diet (jig/g)
Site water k = 9 - 12
ug/L
Site water = < 10-20
ug/L
Not provided in the
literature
Not provided in the
literature
Selenium
Concentrations in the
Organism (jig/g)
Testes = 4.37 (mean
concentration)
Ovaries = 6.96 (mean
concentration)
Liver = 34 (mean
concentration), wet
Gonad= 12.1 (mean,
female), 5.4 (mean,
male), wet
Muscle =13 (mean
concentration), wet
Liver = 10.2 (mean
concentration), wet
Gonad =10.3 (mean,
female), wet
Muscle = 6.7 (mean
concentration), wet
Hepatopancreas =1.31
- 9.30, wet
Hepatopancreas = 2.8
-11.03, wet
Liver = 20
Observed Effects
Reproductive
Histopathological
Histopathological
Histopathological
Study Type
(Surface
Water Type)
Laboratory
(reservoir)
Field
(reservoir)
Field
(lake)
Field
(lake)
Citation
Gillepsie et
al., 1986
Sager and
Colfield,
1984
Sorensen et
al., 1982
Sorensen et
al., 1983
A-50
-------
Appendix A— Literature Review Methodology and Results
Table A-10. Summary of Selenium Concentrations in the Environment and Organisms Experiencing Adverse Effects
Plant Name
Species
Redear sunfish
(Lepomis
microlophus)
Redear sunfish
(Lepomis
microlophus)
Route of Selenium
Exposure
Inhabited a selenium-laden
lake receiving coal ash pond
wastewater.
Inhabited a selenium-laden
lake located near a coal-
fired power plant.
Selenium
Concentrations in the
Environment (jig/L)
or Diet (jig/g)
Not provided in the
literature
Not provided in the
literature
Selenium
Concentrations in the
Organism (jig/g)
Hepatopancreas = 8.4
-27.2 ug/L
Kidney =11.4 -115.7
ug/L
Ovaries = 0-5.9 ug/L
Testes = 0 - 54.2 ug/L
Liver = 7.63 (mean
concentration)
Observed Effects
Increased weight
loss
Histopathological
Reproductive
Study Type
(Surface
Water Type)
Field
(lake)
Field
(lake)
Citation
Sorensen
and Bauer,
1984a
Sorensen,
1988
Acronyms: kg/ha (kilogram per hectacre); ug/L (micrograms per liter); ug/g (micrograms per gram).
a - Multiple fish species were studied; however, as presented by the report, the largemouth bass andpomoxis spp. had the lowest documented selenium biomass
concentrations.
b - Multiple fish species were studied; however, as presented by the report, the Lepomis spp. had the highest documented selenium skeletal muscle concentrations.
c - Multiple fish species were studied; however, as presented by the report, the Letalums spp. had the lowest documented selenium body concentrations.
d - This selenium concentration is dissolved.
e - This concentration is measured in the units kg/ha. The range of selenium concentrations was reported annually from 1982 to 1989, before the steam electric
power plant converted to dry ash handling. Both fish species had the same range of selenium concentrations.
f - The banded water snakes were fed weekly combinations of previously frozen prey items inhabiting the coal ash-contaminated site.
g - The data used in this study were census data collected from routine biological monitoring undertaken by the steam electric power plant.
h - Multiple fish species were studied; however, as presented by the report, the green sunfish had the highest documented selenium biomass concentrations.
i - These are the selenium water concentrations detected prior to the conversion to a dry fly ash handling system.
j - This concentration is measured in the units kg/ha. The range of selenium concentrations was reported annually from 1982 to 1989, before the steam electric
power plant converted to dry ash handling.
k - This concentration was not measured for this study but was reported in a previous study conducted at the same site.
1 - Multiple fish species were studied; however, as presented by the report, the bluegills had the highest documented selenium liver tissue concentration.
m - Multiple fish species were studied; however, as presented by the report, the largemouth bass had the lowest documented selenium liver tissue concentration.
A-51
-------
Appendix B—Proximity Analyses Supporting Tables
APPENDIX B
PROXIMITY ANALYSES SUPPORTING TABLES
Table B-l. Immediate Receiving Waters 303(d) Impairments Listing
Cause Group Name
Algal Growth
Algal Growth
Cause Unknown
Cause Unknown - Impaired Biota
Cause Unknown - Impaired Biota
Dioxins
Dioxins
Dioxins
Fish Consumption Advisory
Flow Alteration(s)
Habitat Alterations
Mercury
Mercury
Mercury
Metals (Other Than Mercury)
Metals (Other Than Mercury)
Metals (Other Than Mercury)
Metals (Other Than Mercury)
Metals (Other Than Mercury)
Metals (Other Than Mercury)
Metals (Other Than Mercury)
Metals (Other Than Mercury)
Metals (Other Than Mercury)
Metals (Other Than Mercury)
Metals (Other Than Mercury)
Metals (Other Than Mercury)
Noxious Aquatic Plants
Nutrients
Nutrients
Nutrients
Nutrients
Nutrients
Nutrients
Cause Name
Algal Growth
Chlorophyll-A
Cause Unknown
Benthic Macroinvertebrates
Bioassessments
Fish Bioassessments
2,3,7,8-Tetrachlorodibenzo-P-Dioxin
(Only)
Dioxin
Dioxins
Fish Consumption Advisory
Flow Alteration(s)
Habitat Alterations
Fish Consumption Advisory - Mercury
Mercury
Mercury In Fish Tissue
Aluminum
Arsenic
Cadmium
Chromium, Total
Copper
Iron
Lead
Manganese
Metals (Other Than Mercury)
Selenium
Silver
Zinc
Macrophytes
Eutrophication
Nitrogen, Total
Nutrient/Eutrophication Biological
Indicators
Nutrients
Phosphorus
Phosphorus, Total
Found in
Combustion
Wastewater
V
/•
V
v'
v'
^
^
^
^
^
V
^
-------
Appendix B—Proximity Analyses Supporting Tables
Table B-l. Immediate Receiving Waters 303(d) Impairments Listing
Cause Group Name
Oil And Grease
Oil And Grease
Organic Enrichment/Oxygen
Depletion
Organic Enrichment/Oxygen
Depletion
Pathogens
Pathogens
Pathogens
Pathogens
Pathogens
Pathogens
Pathogens
Pesticides
Pesticides
Pesticides
Pesticides
Pesticides
Pesticides
Pesticides
Pesticides
Pesticides
pH/Acidity/Caustic Conditions
pH/Acidity/Caustic Conditions
Polychlorinated Biphenyls (PCBs)
Polychlorinated Biphenyls (PCBs)
Polychlorinated Biphenyls (PCBs)
Salinity /Total Dissolved
Solids/Chlorides/Sulfates
Salinity /Total Dissolved
Solids/Chlorides/Sulfates
Sediment
Sediment
Sediment
Sediment
Taste, Color, And Odor
Temperature
Toxic Inorganics
Toxic Organics
Cause Name
Oil
Oil And Grease
Dissolved Oxygen
Dissolved Oxygen Saturation
Bacteria
Coliforms
Enterococcus Bacteria
Escherichia Coli (E. Coli)
Fecal Conform
Indicator Bacteria
Pathogens
Atrazine
Chlordane
Chlorpyrifos
ODD
DDE
DDT
Dieldrin
Mirex
Organochlorine Pesticides
pH
pH, Low
Fish Consumption Advisory - PCBs
PCBs In Fish Tissue
Polychlorinated Biphenyls (PCBs)
Salinity /Total Dissolved Solids/Chlorides
Total Dissolved Solids (TDS)
Sedimentation/Siltation
Siltation
Solids (Suspended/Bedload)
Suspended Sediment
Taste and Odor
Temperature
Boron
Polycyclic Aromatic Hydrocarbons (PAHs)
(Aquatic Ecosystems)
Found in
Combustion
Wastewater
^
v'
v'
v'
^
^
V
v'
^
^
^
^
^
Evaluated in
theEA
•/
^
^
B-2
-------
Appendix B—Proximity Analyses Supporting Tables
Table B-l. Immediate Receiving Waters 303(d) Impairments Listing
Cause Group Name
Turbidity
Turbidity
Cause Name
Total Suspended Solids (TSS)
Turbidity
Found in
Combustion
Wastewater
^
^
Evaluated in
theEA
Source: U.S. EPA, 20141. National 303(d) Listed Impaired Waters National Hydrography Data (NHD) Indexed
Dataset. Reach Address Database (RAD). Extracted on August 4. Available online at:
http://www.epa.gov/waters/data/downloads.html. DCN SE04544.
Note: A surface water is classified as a 303(d) impaired water when pollutant concentrations exceed water quality
standards and the surface water can no longer meet its designated uses (e.g., drinking, recreation, and aquatic
habitat). In even-numbered years, states submit their lists of impaired waters (known as the "303(d) list") to EPA.
These state-submitted, Geographic Information System (GIS) datasets are collected by EPA and indexed to the
National Hydrography Dataset (NHDPlus) at 1:100K resolution (i.e., 303(d) impaired waters proximity database).
For this EA, EPA reviewed the 303(d) impaired waters proximity database to identify steam electric power plant
immediate receiving waters identified as impaired for a pollutant associated with the evaluated wastestreams (i.e.,
FGD wastewater, fly ash transport water, bottom ash transport water, and combustion residual leachate).
B-3
-------
Appendix B—Proximity Analyses Supporting Tables
Table B-2. Immediate Receiving Waters Fish Consumption Advisory Listing
Pollutant
Chlordane
Chlorinated pesticides
DDT
Dieldrin
Dioxin
Lead
Mercury
Mirex
Not Specified
PCBs (Total)
Perfluorooctane sulfonate
Toxaphene
Found in Combustion Wastewater
v'
V
Evaluated in the EA
•/
V
Source: U.S. EPA, 2014h. National Fish Consumption Advisories NHD Indexed Dataset. RAD. Extracted on July 7.
Available online at: http://epamap32.epa.gov/radims/. DCN SE04545.
B-4
-------
Appendix C—Water Quality Module Methodology
APPENDIX C
WATER QUALITY MODULE METHODOLOGY
This appendix presents the model equations, input variables, pollutant benchmarks, and
methodology limitations/assumptions for the immediate receiving water (IRW) model water
quality module.
The IRW water quality module equations are organized by the methodology for
nonvolatile pollutants (i.e., arsenic, cadmium, chromium (VI), copper, lead, nickel, selenium,
thallium, and zinc) and volatile pollutants (i.e., mercury). EPA used the equations to calculate
total and dissolved pollutant concentrations in receiving waters and total pollutant concentrations
in sediment within the immediate discharge zone. Model input requirements for the equations
presented in Appendix C can be divided into four major categories: 1) input variable described
by another equation; 2) site-specific input variable; 3) model assumption variable; and 4) site-
specific assumption variable based on predetermined data. The following tables in Appendix C
describe the input requirements and data sources used in the water quality module:
• Table C-l. Site-Specific Model Input Variables.
• Table C-2. Model Assumption Input Variables.
• Table C-3. Site-Specific Assumption Input Variables.
• Table C-4. Surface Water Partition Coefficients.
• Table C-5. Total Suspended Solids (TSS) Concentrations in Surface Waters.
• Table C-6. Regional Surface Water Temperatures.
• Table C-7. National Recommended Water Quality Criteria (NRWQC) and Drinking
Water Maximum Contaminant Level (MCL) Benchmarks.
EPA calculated pollutant loadings from the evaluated wastestreams as part of its
engineering analysis (see Section 10 of the Technical Development Document for the Effluent
Limitations Guidelines and Standards for the Steam Electric Power Generating Point Source
Category (TDD) [EPA 821-R-15-007]). The IRW water quality module performs calculations on
a per immediate-receiving-water basis. For steam electric power plants that discharge to multiple
receiving waters, EPA divided the plant-specific pollutant loadings accordingly among the
receiving waters based on water diagrams provided in the Questionnaire for the Steam Electric
Power Generating Effluent Guidelines (Steam Electric Survey) responses. EPA used the IRW
model to evaluate the environmental impacts from 188 steam electric power plants in the
receiving water quantitative analysis (209 unique immediate receiving waters).
EPA modeled chromium (VI) in the water quality module, but did not take into
consideration arsenic or mercury speciation. EPA included assumptions of pollutant speciation
for arsenic and mercury as appropriate in the subsequent wildlife and human health modules (see
Appendix D and Appendix E, respectively). EPA used total selenium loadings in the water
quality module; however, due to the partition coefficients available, EPA assumed the dominant
form of selenium in the receiving water was selenate (i.e., selenium (VI)). Although selenium
speciation likely occurs within combustion residual surface impoundments prior to discharge,
CM
-------
Appendix C—Water Quality Module Methodology
EPA selected the selenate partition coefficient because it is expected to be the predominant form
present in well-oxygenated alkaline surface waters and the rate of conversion between selenate
and selenite (i.e., selenium (IV)) is reported to be slow in most natural waters [U.S. EPA, 2004].
IRW Model: Water Quality Module Equations
EPA calculated the nonvolatile pollutant concentrations for the following compartments
within the receiving water:
Total pollutant concentration in water column (Cwc);
Dissolved pollutant concentration in water column (Cdw); and
Total pollutant concentration in sediment (Cbs).
EPA used the equations presented below to calculate receiving water concentrations for
arsenic, cadmium, chromium (VI), copper, lead, mercury, nickel, selenium, thallium, and zinc.
EQUATION C-l
Cw
•'total
Tot, Rivers
f + K x V
Lwater rvwt v river
Where:
CWTot,Rivers
Ltotal
(^cool
(Driver
Iwater
Kwt
V river
—
=
=
Total pollutant concentration in the
waterbody (water and sediment) in rivers and
streams from pollutant loading (grams per
cubic meter [g/m3] or milligrams per liter
[mg/L])
Average pollutant loading from steam
effluent (grams per day [g/day])
Total cooling water effluent flow (cubic
meters per day [m3/day])
Receiving water average annual flow
(m3/day)
Fraction of total waterbody pollutant
concentration in water column (unitless)
Water concentration dissipation rate constant
(I/day)
Flow independent mixing volume for rivers
and streams (m3)
Output from Equation C-l
Site-specific value from
engineering analysis,
based on annual average
(see Table C-l)
Site-specific value from
engineering analysis
(see Table C-l)
Site-specific value from
NHDPlus
(see Table C-l)
Output from
Equation C-6
Output from Equation
C-10
Output from
Equation C-ll
C-2
-------
Appendix C—Water Quality Module Methodology
EQUATION C-2
Where:
Cw
L
'total
Tot, Lake
'lake
CWTot, Lake
Ltotal
Qcool
Qlake
Iwater
Kwt
Vlake
=
=
=
Total pollutant concentration in the waterbody
(water and sediment) in lakes, ponds, and
reservoirs from pollutant loading (g/m3 or
mg/L)
Average pollutant loading from steam effluent
(g/day)
Total cooling water effluent flow (m3/day)
Average annual flow exiting the lake, pond, or
reservoir (m3/day)
Fraction of total waterbody pollutant
concentration in water column (unitless)
Water concentration dissipation rate constant
(I/day)
Flow independent mixing volume for lakes,
ponds, and reservoirs (m3)
Output from Equation C-2
Site-specific value from
engineering analysis, based
on annual average
(see Table C-l)
Site-specific value from
engineering analysis
(see Table C-l)
Site-specific value from
NHDPlus
(see Table C-l)
Output from
Equation C-6
Output from Equation C-10
Output from Equation C-l 2
EQUATION C-3
Where:
C =f
^WC ^w:
water
Cw
tot (Rivers or Lakes)
^wc
Iwater
CWTot
(Rivers or
Lakes)
=
=
Total pollutant concentration in water column
(mg/L)
Fraction of total waterbody pollutant
concentration in water column (unitless)
Total pollutant concentration in the waterbody
(water and sediment) from pollutant loading
(g/m3 or mg/L)
Output from Equation C-3
Output from
Equation C-6
Output from Equation C-l
or Equation C-2
C-3
-------
Appendix C—Water Quality Module Methodology
dz
(Rivers or
Lakes)
dw
(Rivers or
Lakes)
=
Depth of the waterbody (meters [m])
Depth of water column (m)
River or stream: output
from Equation C-9
Lake, pond, or reservoir:
site-specific value (see
Table C-l)
River or stream: output
from
Equation C-7
Lake, pond, or reservoir:
site-specific value (see
Table C-l)
EQUATION C-4
Where:
= P
^
•(-
Kdsw x TSS x 0.000001,
Cdw
CwC
Kdsw
TSS
0.000001
=
=
=
Dissolved pollutant concentration in water
(mg/L)
Total pollutant concentration in water column
(mg/L)
Suspended sediment-surface water partition
coefficient (milliliters per gram [mL/g])
Total suspended solids (mg/L)
Conversion factor (L/mL)(g/mg)
Output from
Equation C-4
Output from Equation C-3
Model assumption value
(see Table C-2 and Table
C-4)
Site-specific assumption
value (see Table C-3 and
Table C-5)
Conversion factor
EQUATION C-5
Where:
Cw,
tot (Rivers or Lakes) j
Cbs
fBenth
CWTot
(Rivers or
Lakes)
=
=
Total pollutant concentration in sediment
(mg/L)
Fraction of total waterbody pollutant
concentration in benthic sediment (unitless)
Total pollutant concentration in the waterbody
(water and sediment) from pollutant loading
(g/m3 or mg/L)
Output from Equation C-5
Output from
Equation C-l 5
Output from Equation C-l
or
Equation C-2
C-4
-------
Appendix C—Water Quality Module Methodology
dz
(Rivers or
Lakes)
db
(Rivers or
Lakes)
Depth of the waterbody (m)
Depth of upper benthic sediment layer (m)
River or stream: output
from Equation C-9
Lake, pond, or reservoir:
site-specific value (see
Table C-l)
Model assumption value of
0.03m
(see Table C-2)
EQUATION C-6
1 water '
Where:
[1 + (Kdsw x TSS x 0.000001)] x $s
^z
[1 + (Kdsw x TSS x 0.000001)] x ^ + f(bsp + Kdbs x bsc)
twater
Kdsw
TSS
0.000001
dw
(Rivers or
Lakes)
dz
(Rivers or
Lakes)
bsp
Kdbs
=
=
=
=
=
Fraction of total waterbody pollutant
concentration in water column (unitless)
Suspended sediment-surface water
partition coefficient (mL/g)
Total suspended solids (mg/L)
Conversion factor (L/mL)(g/mg)
Depth of water column (m)
Depth of the waterbody (m)
Bed sediment porosity (cubic centimeter
per cubic centimeter [cmVcm3])
Bottom sediment-pore water partition
coefficient (mL/g)
Output from
Equation C-6
Model assumption value
(see Table C-2 and Table C-4)
Site-specific assumption value
(see Table C-3 and Table C-5)
Conversion factor
River or stream: output from
Equation C-7
Lake, pond, or reservoir:
site-specific value (see Table
C-l)
River or stream: output from
Equation C-9
Lake, pond, or reservoir:
site-specific value (see Table
C-l)
Model assumption value of 0.6
cm3/cm3
(see Table C-2)
Model assumption value
(see Table C-2 and Table C-4)
C-5
-------
Appendix C—Water Quality Module Methodology
bsc
db
Bed sediment particle concentration (gram
per cubic centimeter [g/cm3]) or (kilogram
per liter [kg/L])
Depth of upper benthic layer (m)
Model assumption value of 1
g/cm3
(see Table C-2)
Model assumption value of
0.03m
(see Table C-2)
EQUATION C-7
Where:
d
w-
v x Width
dw, river
Driver
V
Width river
=
=
Depth of water column (m)
Receiving water average annual flow
(m3/s)
Receiving water velocity (m/s)
Receiving water width (m)
Output from
Equation C-7
Site-specific value from NHD
Plus
(see Table C-l)
Site-specific value from NHD
Plus
(see Table C-l)
Output from Equation C-8
EQUATION C-8
Where:
Widthriver
Driver
=
Receiving water width (m)
Receiving water average annual flow
(m3/s)
Output from Equation C-8
Site-specific value from NHD
Plus
(see Table C-l)
EQUATION C-9
Where:
dz, river db ~r dw, rive
dz, river
db
=
Depth of the waterbody (m)
Depth of upper benthic sediment layer (m)
Output from Equation C-9
Model assumption value 0.03
m
(see Table C-2)
C-6
-------
Appendix C—Water Quality Module Methodology
dw, river
=
Depth
of water column (m)
Output from
Equation C-7
EQUATION C-10
= (fw
water
(f
water
(f
faenth
Where:
Kwt
Iwater
ksw
fbenth
ksed
kvol
Kb
—
=
Water concentration dissipation rate
constant (I/day) for nonvolatile pollutants
(see Equation C-16 for volatile pollutants)
Fraction of total waterbody pollutant
concentration in water column (unitless)
Degradation rate for water column (I/day)
Fraction of total waterbody pollutant
concentration in benthic sediment
(unitless)
Degradation rate for sediment (I/day)
Water column volatilization loss rate
constant (I/day)
Benthic burial rate (I/day)
Output from Equation C-10
Output from
Equation C-6
Model assumption value of
0/day
(see Table C-2)
Output from
EquationC-15
Model assumption value of
0/day
(see Table C-2)
Model assumption value of
0/day
(see Table C-2)
Output from
EquationC-14
EQUATION C-11
Where:
Vriver = Widthriver x
dz,rive
Vriver
Widthriver
Len
dz, river
=
=
=
Flow independent mixing volume for
rivers and streams (m3)
Receiving water width (m)
Length of stream reach (m)
Depth of the waterbody (m)
Output from
Equation C-ll
Output from Equation C-8
Site-specific value from NHD
Plus
(see Table C-l)
Output from Equation C-9
C-7
-------
Appendix C—Water Quality Module Methodology
EQUATION C-12
Where:
Vlake = Area * dz,lake
Vlake
Area
dz,lake
=
=
Flow independent mixing volume for
lakes, ponds, and reservoirs (m3)
Surface area of the lake (m)
Depth of the lake (m)
Output from
Equation C-12
Site-specific value from NHD
Plus
(see Table C-l)
Site-specific value
(see Table C-l)
EQUATION C-13
Where:
l+KdswxTSSx 0.000001
fd
Kdsw
TSS
0.000001
=
=
=
=
Dissolved fraction in water (unitless)
Suspended sediment-surface water
partition coefficient (mL/g)
Total suspended solids (mg/L)
Conversion factor (L/mL)(g/mg)
Output from
Equation C-13
Model assumption value
(see Table C-2 and Table C-4)
Site-specific assumption value
(see Table C-3 and Table C-5)
Conversion factor
EQUATION C-l4
Where:
WB
rbenth
Kb
fbenth
WB
=
=
Benthic burial rate (I/day)
Fraction of total waterbody pollutant
concentration in benthic sediment
(unitless)
Rate of burial (m/day)
Output from
Equation C-l 4
Output from
Equation C-l 5
Model assumption value of 0
m/day (see Table C-2)
-------
Appendix C—Water Quality Module Methodology
db
=
Depth of upper benthic sediment layer (m)
Model assumption value of
0.03 m (see Table C-2)
EQUATION C-15
rBenth :
Where:
(bsp + Kdbsxbsc)x-,b
d r dul
[1 + (Kdsw x TSS x 0.000001)] x ^ + (bsp + Kdbs x bsc) x ^
az L azJ
fbenth
bsp
Kdbs
bsc
db
dz
Kdsw
TSS
0.000001
dw
(Rivers or
Lakes)
=
=
=
=
=
Fraction of total waterbody pollutant
concentration in benthic sediment
(unitless)
Bed sediment porosity (cmVcm3)
Bottom sediment-pore water partition
coefficient (mL/g)
Bed sediment particle concentration
(g/cm3) or (kg/L)
Depth of upper benthic sediment layer (m)
Depth of the waterbody (m)
Suspended sediment-surface water
partition coefficient (mL/g)
Total suspended solids (mg/L)
Conversion factor (L/mL)(g/mg)
Depth of water column (m)
Output from
Equation C-15
Model assumption value of 0.6
cm3/cm3
(see Table C-2)
Model assumption value
(see Table C-2 and Table C-4)
Model assumption value of 1
g/cm3
(see Table C-2)
Model assumption value of
0.03m
(see Table C-2)
Output from Equation C-9
Model assumption value
(see Table C-2 and Table C-4)
Site-specific assumption value
(see Table C-3 and Table C-5)
Conversion factor
River or stream: output from
Equation C-7
Lake, pond, or reservoir:
site-specific value (see Table
C-l)
EPA calculated the volatile pollutant concentrations in each of the three compartments
within the receiving water by building off the equations used to calculate nonvolatile pollutant
concentrations. The water concentration dissipation rate constant, Kwt, in Equation C-10 was
replaced with a Kwt,volatile factor (see Equation C-l6) that takes into account volatilization loss
C-9
-------
Appendix C—Water Quality Module Methodology
(kvoi). EPA used the equations presented below in combination with the preceding equations to
calculate receiving water concentrations for mercury only.
EQUATION C-16
Kwt, volatile = (fwater X ksw) + (ffaenth X ksed) + (fwater X fd X kvol) + (ffaenth X Kb)
Where:
-K.wt, volatile
twater
ksw
fbenth
ksed
fd
kvoi
Kb
=
=
=
=
=
Water concentration dissipation rate
constant (I/day)
Fraction of total waterbody pollutant
concentration in water column (unitless)
Degradation rate for water column (I/day)
Fraction of total waterbody pollutant
concentration in benthic sediment
(unitless)
Degradation rate for sediment (I/day)
Dissolved fraction in water (unitless)
Water column volatilization loss rate
constant (I/day)
Benthic burial rate (I/day)
Output from Equation C-16
Output from
Equation C-6
Model assumption value of
0/day
(see Table C-2)
Output from
Equation C-15
Model assumption value of
0/day
(see Table C-2)
Output from
EquationC-13
Output from
Equation C-17
Output from
EquationC-14
EQUATION C-17
Where:
kvoi =
kvoi
Kv
fd
=
=
=
Water column volatilization loss rate
constant (I/day)
Diffusion transfer rate (m/day)
Dissolved fraction in water (unitless)
Output from
Equation C-17
Output from
EquationC-18
Output from
Equation C-13
C-10
-------
Appendix C—Water Quality Module Methodology
dw
(Rivers or
Lakes)
Depth of water column (m)
River or stream: output from
Equation C-7
Lake, pond, or reservoir:
site-specific value (see Table
C-l)
EQUATION C-l8
Where:
Kv =
Kv
Diffusion transfer rate (m/day)
Output from
Equation C-l8
'water
Temperature correction (unitless)
Model assumption value of
1.026
(see Table C-2)
Tw
Temperature of the waterbody (degrees
Kelvin [°K])
River or stream: site-specific
assumption value
(see Table C-3 and Table C-6)
Lake, pond, or reservoir:
model assumption value (see
Table C-3 and Table C-6)
Thic
Temperature of HLC (°K)
Default model value of 298°K
(see Table C-2)
KL
(Rivers or
Lakes)
Liquid-phase transfer coefficient (m/day)
River or stream: output from
Equation C-l9
Lake, pond, or reservoir:
output from Equation C-21
Kg
(Rivers or
Lakes)
Gas-phase transfer coefficient (m/day)
River or stream:
model assumption value of
100 m/day
(see Table C-2)
Lake, pond, or reservoir:
output from
Equation C-23
C-ll
-------
Appendix C—Water Quality Module Methodology
HLC
R
Henry's Law Constant (atm-nrVmole) *
Universal gas constant (atm-m3/°K-mole)
Known value of 0.01 13 atm-
m3/mol
(see Table C-2)
Known value of 0.00008205
atm-m3/°K-mole
(see Table C-2)
EQUATION C-19
Where:
K
L(Rivers) '
|10-4xDwxv
x 86,400
KL(Rivers)
Dw
V
dz,river
86,400
=
=
=
=
=
Liquid-phase transfer coefficient (m/day)
Diffusivity of the pollutant in water
(square centimeter per second [cm2/s])
Receiving water velocity (m/s)
Depth of waterbody (m)
Conversion factor (s/day)
Output from
Equation C-19
Output from
Equation C-20
Site-specific value from NHD
Plus
(see Table C-l)
Output from Equation C-9
Conversion factor
EQUATION C-20
Where:
-5
22x10
MW2/3
Dw
MW
=
Diffusivity of the pollutant in water
(cm2/s)
Molecular weight (grams per mole
[g/mol])
Output from
Equation C-20
Known value of 200.59 g/mol
for mercury
(see Table C-2)
Units for Henry's Law Constant are atmospheres of absolute pressure (atm) per cubic meter (m3) per mole (mol).
C-12
-------
Appendix C—Water Quality Module Methodology
EQUATION C-21
Where:
K
L(Lakes)
~"0-67
KL(Lakes)
Cd
Wio
Pa
PW
k
X2
SCw
86,400
=
=
=
=
=
=
Liquid-phase transfer coefficient (m/day)
Drag coefficient (unitless)
Wind velocity 10 meters above water
surface (m/s)
Density of air corresponding to water
temperature (g/cm3)
Density of water corresponding to water
temperature (g/cm3)
Von Karman's constant (unitless)
Dimensionless viscous sublayer thickness
(unitless)
Water Schmidt number (dimensionless)
Conversion factor (s/day)
Output from Equation C-21
Model assumption value of
0.0011
(see Table C-2)
Site-specific assumption value
(see Table C-3)
Model assumption value of
0.0012 g/cm3
(see Table C-2)
Model assumption value of 1
g/cm3
(see Table C-2)
Known value of 0.4
(see Table C-2)
Model assumption value of 4
(see Table C-2)
Output from
Equation C-22
Conversion factor
EQUATION C-22
Sc -
W
Where:
SCw
|Iw
PW
DW
=
=
Water Schmidt number (dimensionless)
Viscosity of water corresponding to water
temperature (g/cm-s)
Density of water corresponding to water
temperature (g/cm3)
Diffusivity of the pollutant in water
(cm2/s)
Output from
Equation C-22
Model assumption value of
0.0169 g/cm-s
(see Table C-2)
Model assumption value of 1
g/cm3
(see Table C-2)
Output from
Equation C-20
C-13
-------
Appendix C—Water Quality Module Methodology
EQUATION C-23
Where:
Kg(Lakes) ~ V ^d X W10 x
x 86,400
Kg(lakes)
Cd
WlO
k
X2
SCa
86,400
=
=
=
=
=
=
Gas-phase transfer coefficient (m/day)
Drag coefficient (unitless)
Wind velocity 10 meters above water
surface (m/s)
Von Karman's constant (unitless)
Dimensionless viscous sublayer thickness
(unitless)
Air Schmidt number (dimensionless)
Conversion factor (s/day)
Output from
Equation C-23
Model assumption value of
0.0011
(see Table C-2)
Site-specific assumption value
(see Table C-3)
Known value of 0.4
(see Table C-2)
Model assumption value of 4
(see Table C-2)
Output from
Equation C-24
Conversion factor
EQUATION C-24
Where:
Sca =
(1.32 + 0.009TJ x 105
L9
MW
2/3
SCa
Ta
MW
=
=
Air Schmidt number (dimensionless)
Air temperature °K
Molecular weight (g/mol)
Output from
Equation C-24
Site-specific assumption value
(see Table C-3)
Known value of 200.59 g/mol
for mercury
(see Table C-2)
EPA calculated the potential water quality impacts to aquatic life and humans by
comparing the pollutant concentration in the water column (Cwc or Cdw, depending on the
benchmark) to the water quality benchmarks presented in Table C-7.
C-14
-------
Appendix C—Water Quality Module Methodology
IRW Model: Water Quality Module Inputs
Table C-l. Site-Specific Input Variables
Input
Variable
Input Category and Description
Data Source
Ltotal
Plant-specific effluent characteristic
Total waterbody loading
EPA estimated the pollutant discharge loadings using the
methodology presented in Section 10 of the TDD.
Qcool
Plant-specific effluent characteristic
Total cooling water effluent flow
by receiving water
EPA determined the estimated cooling water flow for each plant
by outfall based an assessment of industry survey results using
the methodology outlined in Water Quality Module: Plant and
Receiving Water Characteristics [ERG, 2015e].
S^nver
Receiving water characteristic for
rivers and streams
Waterbody annual flow
EPA extracted average annual flow values from the NHD Plus
dataset using the methodology outlined in Water Quality
Module: Plant and Receiving Water Characteristics [ERG,
2015e]. The NHD Plus dataset includes estimated mean annual
flow values for each stream reach within the network using the
Vogel Method [Vogel et al., 1999] and the Unit Runoff Method.
Receiving water characteristic for
rivers and streams
Receiving water velocity
EPA extracted average annual velocity values from the NHD
Plus dataset using the methodology outlined in Water Quality
Module: Plant and Receiving Water Characteristics [ERG,
2015e]. The NHD Plus dataset includes estimated mean annual
velocity values for each stream reach within the network using
the Jobson Method [Jobson, 1996] and the estimated mean
annual flow values.
Len
Receiving water characteristic for
rivers and streams
Length of stream reach
EPA estimated the stream reach length based on outfall
locations using the methodology described in Water Quality
Module: Plant and Receiving Water Characteristics [ERG,
2015e].
Qlake
Receiving water characteristic for
lakes, ponds, and reservoirs
Average discharge flow exiting the
lake/pond system
EPA extracted average annual flow values from the NHD Plus
dataset using the methodology outlined in Water Quality
Module: Plant and Receiving Water Characteristics [ERG,
2015e]. The NHD Plus dataset includes estimated mean annual
flow values for the stream reach exiting the lake using the
Vogel Method [Vogel et al., 1999] and the Unit Runoff Method.
Area
Receiving water characteristic for
lakes, ponds, and reservoirs
Surface area of the lake, pond, or
reservoir
EPA estimated the lake surface area based on NHD Plus data or
site-specific sources as described in Water Quality Module:
Plant and Receiving Water Characteristics [ERG, 2015e].
dz,lake
Receiving water characteristic for
lakes, ponds, and reservoirs
Depth of the lake, pond, or
reservoir
EPA estimated the depth of the lake, pond, or reservoir based on
site-specific data as described in Water Quality Module: Plant
and Receiving Water Characteristics [ERG, 2015e].
dw.lake
Receiving water characteristic for
lakes, ponds, and reservoirs
Depth of the water column
EPA estimated the depth of the lake, pond, or reservoir based on
site-specific data as described in Water Quality Module: Plant
and Receiving Water Characteristics [ERG, 2015e].
C-15
-------
Appendix C—Water Quality Module Methodology
Table C-2. Model Assumption Input Variables and Known Variables
Input Variable
bsp
bsc
db
-Ksw
kvd
ksed
WB
Wwater
Kg(Rivers)
R
cd
pa
Pw
k
Description
Bed sediment porosity
Bed sediment particle
concentration
Depth of upper benthic layer
Degradation rate for water
column
Water column volatilization
loss rate constant
Degradation rate for
sediment
Rate of burial
Temperature correction
Gas phase transfer
coefficient for rivers or
streams
Ideal gas constant
Drag coefficient
Density of air corresponding
to water temperature
Density of water
corresponding to water
temperature
Von Karman's constant
Assumed/
Known
Value
0.6 cm3/cm3
1 g/cm3
0.03m
0/day
0/day
0/day
0/day
1.026
(unitless)
36,500 m/yr
(100 m/day)
0.00008205
atm-m3/
K-mole
0.0011
(unitless)
0.00 12 g/cm3
1 g/cm3
0.4
(unitless)
Assumption Rationale/Data Source
Bed sediment porosity is the volume of water per
volume of benthic space with typical values
ranging between 0.8 and 0.4 [U.S. EPA, 1998b].
EPA selected an average value to use for this input
variable.
Bed sediment particle concentrations typically
range between 0.5 to 1.5 g/cm3 [U.S. EPA, 1998d].
EPA selected an average value to use for this input
variable.
The upper benthic layer variable represents the
portion of the bed in equilibrium with the water
column. Typical values can range from 0.01 to
0.05 m [U.S. EPA, 1998b]. EPA selected an
average value to use for this input variable.
EPA assumed no loss from pollutant degradation
in the water column, as an environmentally
conservative assumption.
EPA selected a volatilization rate of 0 for
nonvolatile pollutants (i.e., all pollutants except
mercury).
EPA assumed no loss from pollutant degradation
in the sediment, as an environmentally
conservative assumption.
EPA assumed no pollutant loss from burial within
the waterbody sediments, as an environmentally
conservative assumption.
EPA selected the temperature correction factor
based on the value provided in U.S. EPA, 1998b.
EPA selected the gas phase transfer coefficient for
rivers and streams based on the value provided in
U.S. EPA, 1998b.
The ideal gas constant is a known chemical
constant.
EPA selected the drag coefficient based on the
value provided in U.S. EPA, 1998b.
EPA selected the density of air corresponding to
water temperature based on the value provided in
U.S. EPA, 2005b.
EPA selected the density of water corresponding to
water temperature based on the value provided in
U.S. EPA, 2005b.
The von Karman constant is a known
dimensionless constant used to describe the
velocity profile of a turbulent fluid flow near a
boundary.
C-16
-------
Appendix C—Water Quality Module Methodology
Table C-2. Model Assumption Input Variables and Known Variables
Input Variable
Description
Assumed/
Known
Value
Assumption Rationale/Data Source
Kds,
Suspended sediment- surface
water partition coefficient
Table C-4
The suspended sediment partition coefficient
describes the partitioning of a pollutant between
sorbing material, in this case suspended sediment
and surface water. EPA identified U.S. EPA,
2005a as the primary source for the pollutant-
specific suspended sediment partition coefficients.
Kdbs
Bottom sediment-pore water
partition coefficient
Table C-4
The bottom sediment partition coefficient
describes the partitioning of a pollutant between
sorbing material, in this case bottom sediment and
pore water. EPA identified U.S. EPA, 2005a as the
primary source for the pollutant-specific bed
sediment partition coefficients.
Dimensionless viscous
sublayer thickness
4
(unitless)
EPA selected the viscous sublayer thickness value
based on the value provided in U.S. EPA, 2005b.
Viscosity of water
corresponding to water
temperature
0.0169 g/cm-
s
EPA selected the viscosity of water value based on
the value provided in U.S. EPA, 2005b.
HLC
Henry's Law Constant
0.0113
atm-m3/mol
Henry's Law Constant is used in
Equation C-18 to estimate the receiving water
concentration for volatile pollutants. Mercury is
the only volatile pollutant included in the IRW
model. Therefore, the assumed model default value
is set to Henry's Law Constant for mercury at 298
Temperature of Henry's Law
Constant
298 °K
The value 298 °K is the standard temperature value
provided for Henry's Law Constant.
MW
Molecular weight
200.59 g/mol
Molecular weight is used in
Equation C-20 and
Equation C-24 to estimate the receiving water
concentration for volatile pollutants. Mercury is
the only volatile pollutant included in the IRW
model. Therefore, the assumed model default value
is set to the molecular weight for mercury.
Table C-3. Site-Specific Assumption Input Variables
Input Variable
TSS
Description
Total suspended solids
Assumed
Value
Table C-5
Data Source
EPA used the geometric mean of the regional and
national TSS concentrations determined as part of
the Human and Ecological Risk Assessment of
Coal Combustion Residuals [U.S. EPA, 2014g].
C-17
-------
Appendix C—Water Quality Module Methodology
Table C-3. Site-Specific Assumption Input Variables
Input Variable
Description
Assumed
Value
Data Source
Wio
Wind velocity 10 m above
the water surface
Table C-l
National Climatic Data Center national mean
annual wind speed GIS coverage (downloaded
05/12/2011 from
http://hurricane.ncdc.noaa.gov/cgi-
bin/climaps/climaps.pl?directive=quick search&s
ubrnum). EPA selected, as an environmentally
conservative estimate, the lower of the wind speed
range values for the analysis.
Ta
Air temperature
Table C-2
National Climatic Data Center national mean
annual temperature GIS coverage (downloaded
05/12/2011 from
http://hurricane.ncdc.noaa.gov/cgi-
bin/climaps/climaps.pl?directive=quick search&s
ubrnum). EPA selected, as an environmentally
conservative estimate, the lower of the air
temperature range values for the analysis.
Tw
Temperature of the surface
water
Table C-6
EPA used the regional surface temperatures
determined as part of the Human and Ecological
Risk Assessment of Coal Combustion Residuals
[U.S. EPA,2014g].
AMNUAL
MEAN WIND SPEED
I | STATES
13 MEAN SPEED OF WIND (MPH)
^H - ANNUAL -
A<6.0
^^ B 6.0 - 6.9
| C 7.0-7.9
^^ D 8.0-8.9
| E 9.0 - 9.9
^^ F 10.0- 10.9
^^ G 11.0- 11.9
| H > 11.9
TITLE
Figure C-l. National Climatic Data Center National Mean Annual Wind Speed
C-18
-------
Appendix C—Water Quality Module Methodology
ANNUAL
MEAN DAILY AVERAGE TEMPERATURE
| | STATES
13 MEAN DAILY AVG TEMP (DEG_F)
^^| . ANNUAL -
^^^ A < 32.0
B 32,0 -40.0
^^^ C 40 1 -45.0
| D 45.1 - 50.0
E 50.1 -550
F 55 1 - 60.0
G60.1 -65.0
H 65.1 - 70.0
I > 70 0
/"•/ TITLE
Figure C-2. National Climatic Data Center National Mean Annual Temperature
Table C-4. Partition Coefficients
Pollutant
Arsenic
Cadmium
Chromium (VI)
Copper
Lead
Mercury (II)
Nickel
Selenium (IV)
Thallium
Zinc
Suspended Sediment-
Water Partition
Coefficient (Kdsw)
(mL/g)
7,900
79,000
16,000
50,000
500,000
200,000
20,000
25,000
13,000
100,000
Bottom Sediment-Pore
Water Partition
Coefficient (Kdbs)
(mL/g)
250
2,000
50
3,200
40,000
79,000
7,900
4,000
20
13,000
Source: U.S. EPA, 2005a.
C-19
-------
Appendix C—Water Quality Module Methodology
Table C-5. TSS Concentrations in Surface Waters
Hydrologic
Region a
1
2
o
J
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
Lakes
(national)
Number of
Measurements
9,007
47,202
43,395
29,577
39,900
4,137
34,494
46,231
3,254
62,791
48,969
7,280
13,974
26,699
9,162
19,965
173,136
42,022
4,360
Number of
Annual Medians
33
38
36
37
38
28
37
38
35
38
38
35
37
38
37
33
37
37
99
Annual Median TSS (mg/L)
(log triangular distribution)
Min
3.2
10
6.3
6.3
4
5
32
50
13
10
25
40
32
16
20
4
2
13
1
Max
40
316
79
794
100
316
1,585
316
3,162
398
794
1,995
79,433
5,012
19,953
2,512
316
398
398
Geometric
Mean
8
32
25
25
25
16
63
158
32
126
200
79
200
158
200
16
6
63
25
Weighted
Geometric
Mean
6
40
25
25
25
20
100
126
63
126
126
126
398
251
398
25
10
50
25
Source: U.S. EPA, 2010b; Legacy STORET database.
a - For rivers and streams, EPA used the geometric mean TSS concentration for the corresponding hydrogeologic
region. For lakes, ponds, and reservoirs, EPA used a national geometric mean.
Table C-6. Regional Surface Water Temperatures
Hydrologic Region
1
2
3
4
5
6
7
8
9
10
Climate
North
North
South
North
North
South
North
South
North
North
Surface Water
Temperature (°C)
14 (Northern Median)
16
21
14
17
18
15
20
10
13
Surface Water
Temperature (°K)
287
289
294
287
290
291
288
293
283
286
C-20
-------
Appendix C—Water Quality Module Methodology
Table C-6. Regional Surface Water Temperatures
Hydrologic Region
11
12
13
14
15
16
17
18
Climate
South
South
South
South
South
South
North
South
Surface Water
Temperature (°C)
17
21
17 (Southern Median)
9
17
9
14 (Northern Median)
15
Surface Water
Temperature (°K)
290
294
290
282
290
282
287
288
Source: U.S. EPA, 2010b; Legacy STORET database.
Table C-7. NRWQC and MCL Benchmarks
Pollutant
Arsenic
Cadmium
Chromium (VI)
Copper
Lead
Mercury
Nickel
Selenium
Thallium
Zinc
FW Acute
NRWQC
Benchmark a'b
(mg/L)
0.34 (d)
0.002 (d)
0.016 (d)
0.013 (d,e)
0.065 (d)
0.0014 (d)
0.47 (d)
~
~
0.12(d)
FW Chronic
NRWQC
Benchmark a'b
(mg/L)
0.15(d)
0.00025 (d)
O.Oll(d)
0.009 (d,e)
0.0025 (d)
0.00077 (d)
0.052 (d)
0.005
~
0.12(d)
HHWO
NRWQC
Benchmark a'b
(mg/L)
0.000018 (f)
-
-
1.3
~
~
0.61
0.17
0.00024
7.4
HHO
NRWQC
Benchmark a'b
(mg/L)
0.00014 (f)
-
-
~
~
~
4.6
4.2
0.00047
26
MCL
Benchmark a'c
(mg/L)
0.01
0.005
o.i (g)
1.3 (Action
Level); 1.0 (h)
0.015 (Action
Level)
0.002 (f)
-
0.05
0.002
5(h)
Acronyms: MCL (Maximum Contaminant Level); NRWQC (National Recommended Water Quality Criteria).
a - "--" designates instances where a benchmark does not exist for the pollutant or the benchmark is a secondary
standard.
b - National Recommended Water Quality Criteria. Washington, D.C. [U.S. EPA, 2009d]. Pollutant concentrations
were compared to the freshwater (FW) acute and chronic NRWQC and the human health (HH) water and organisms
(WO) and organisms only (O) NRWQC.
c - National Primary Drinking Water Regulations. EPA 816-F-09-004. May. Washington, D.C. [U.S. EPA, 2009e].
d - Benchmark is expressed in terms of the dissolved pollutant in the water column.
e - The 2009 NRWQC for copper are calculated using the biotic ligand model; therefore, there is no national value.
For this analysis, EPA used the 2002 NRWQC values [U.S. EPA, 2002].
f - Benchmark is for inorganic form of pollutant.
g - MCL is for total chromium.
h - Secondary (nonenforceable) drinking water standard.
C-21
-------
Appendix C—Water Quality Module Methodology
IRW Model: Water Quality Module Methodology Limitations and Assumptions
The limitations and assumptions in the IRW water quality module are as follows:
• The module is based on annual-average pollutant loadings, normalized effluent flow
rates from the steam electric power plants, and annual-average flow rates within the
immediate receiving waters. The module does not consider temporal variability (e.g.,
seasonal differences, storm flows, low-flow events, catastrophic events). The result of
this limitation on the water quality module outputs is unknown.
• The module represents only the waterbody concentration within the immediate
discharge zone (i.e., approximately 1 to 10 kilometers [km] from the outfall) and does
not calculate pollutant concentrations in downstream waters. This limitation results in
a potential underestimation of the extent of surface waters with environmental and
human health impacts under baseline conditions and improvements under the
regulatory options.
• The module does not take into consideration pollutant speciation within the receiving
stream. This limitation is particularly relevant to the wildlife impact analysis as many
of the ecological impacts are tied to a specific pollutant species. For example,
inorganic arsenic is typically more toxic to aquatic life than organic arsenic. This
limitation results in a potential overestimation of the number of immediate receiving
waters with exceedances of water quality benchmarks for inorganic forms of the
pollutant (e.g., the human health NRWQCs for arsenic).
• The module assumes that equilibrium is quickly attained within the waterbody
following discharge and is consistently maintained between the water column and
surficial bed sediments. This assumption is especially significant regarding pollutant
equilibrium within lakes, ponds, and reservoirs. The module equations presented in
Appendix C do not take into consideration the effects of currents, inversion, or
temperature variations within the water column, but assume that the entire mass of the
lake, pond, or reservoir is at equilibrium. As a result, the module outputs do not
reflect the potential spatial and temporal variability of pollutant concentrations within
the immediate receiving water, and potentially underestimate the existence of isolated
"hot spots" of elevated pollutant concentrations. The module does not account for the
accumulation of pollutant concentrations in bottom sediments and pore water that
occur over prolonged discharge periods.
• The module assumes that pollutants dissolved or sorbed within the water column and
bottom sediments can be described by a partition coefficient. EPA used a single
partition coefficient to characterize the pollutant in the immediate receiving waters.
The partition coefficient in a specific waterbody will be influenced by geochemical
parameters (e.g., pH and presence of particulate organic matter and other sorbing
material). EPA used a mean or median value for the partition coefficients (central
tendency of Kd values) based on data gathered from published sources, statistical
analysis of retrieved data, geochemical modeling, and expert judgment [U.S. EPA,
2005a]. The result of this assumption on the water quality module outputs is unknown
because of unknown site-specific factors.
C-22
-------
Appendix C—Water Quality Module Methodology
The module assumes that pollutants sorbed to bottom sediments are considered a net
loss from the water column. This assumes that bottom sediments are not resuspended
and deposited further downstream, but remain within the immediate discharge zone
and do not further contribute to the dissolved or suspended sediment concentrations
within the water column. This assumption results in a potential overestimation of
pollutant concentrations within the benthic sediments and a potential underestimation
of pollutant concentrations within the water column and downstream reaches.
The module assumes a pollutant burial rate of zero within benthic sediment. This is
an environmentally protective assumption that might overestimate impacts to
sediment receptors to some degree. The burial rate constant is a function of the
deposition of sediments from the water column to the upper bed and accounts for the
soil eroding into a waterbody becoming bottom sediment rather than suspended
sediment. The rate of burial used for each segment of a waterbody may be difficult to
obtain [U.S. EPA, 1998b]. EPA had neither measured values nor the data to
determine burial rates for each immediate receiving water. The pollutants with more
than 10 percent immediate receiving waters showing impacts to sediment receptors
include cadmium, mercury, and nickel (see Table 6-4). This assumption results in a
potential overestimation of impacts in the benthic sediment.
The module does not take into account ambient background pollutant concentrations
or contributions from other point and nonpoint sources. Also, the pollutant loadings
included in the module are not representative of the total pollutant loadings from
steam electric power plants, as there are several waste streams that are not included in
the analysis (e.g., stormwater runoff, metal cleaning wastes, coal pile runoff).
Because of this approach, the module potentially underestimates the number and
magnitude of benchmark exceedances at baseline and under the regulatory options.
The module also potentially underestimates the number of environmental and human
health improvements under the regulatory options (i.e., a higher number of
exceedances under baseline conditions creates additional opportunities for
improvement under the regulatory options). The results of EPA's case study
modeling, which does take into account ambient background pollutant concentrations
and contributions from other point and nonpoint sources, support this assessment of
the water quality module's limitations (see Section 8).
C-23
-------
Appendix D—Wildlife Module Methodology
APPENDIX D
WILDLIFE MODULE METHODOLOGY
This appendix presents the model equations, input variables, pollutant benchmarks, and
methodology limitations/assumptions for the immediate receiving water (IRW) model wildlife
module. Wildlife impacts include the following ecological receptors:
• Aquatic and sediment organisms (amphibians, fish, invertebrates) in direct contact with
receiving water and/or sediment in the immediate discharge zone of steam electric
power plants.
• Wildlife (minks and eagles)1 that consume fish from receiving waters in the immediate
discharge zone of steam electric power plants.
EPA estimated pollutant concentrations in the immediate receiving water and sediment
using the IRW model water quality module (see Appendix C). The wildlife module uses these
concentrations as inputs.
Model input requirements for the equations presented in Appendix D can be divided into
four major categories: 1) input variable described by another equation; 2) site-specific input
variable; 3) model assumption variable; and 4) pollutant-specific variable. The following tables in
Appendix D describe the input requirements and data sources used in the wildlife module and
impacts analysis:
• Table D-l. Chemical Stressor Concentration Limits (CSCLs) for Sediment Biota.
• Table D-2. Bioconcentration Factors (BCFs) and Bioaccumulation Factors (BAFs) for
Trophic Level 3 (T3) and Trophic Level 4 (T4) Fish.
• Table D-3. No Effect Hazard Concentration (NEHC) Benchmarks for Minks and Bald
Eagles.
IRW Model: Wildlife Module Equations, Input Variables, and Impact Analysis
Impact to Aquatic Life Receptors from Direct Contact with Sediment. EPA determined the
potential negative impact to aquatic organisms from direct contact with the sediment in immediate
receiving waters by comparing the pollutant concentration in the sediment (Cbs from the water
quality module) to the CSCL benchmarks for sediment biota listed in Table D-l. The wildlife
module expresses this comparison as a hazard quotient (HQ). An HQ of higher than one (i.e.,
pollutant concentration exceeds benchmark) indicates a potential impact to the exposed organism.
EPA used Equation D-l to calculate the HQ for sediment biota.
1 EPA selected minks and eagles to represent national-scale impacts from steam electric power plants because their
habitats cover the entire United States (i.e., can be used for a national assessment).
EM
-------
Appendix D—Wildlife Module Methodology
EQUATION D-l
Where:
-bs
CSCLsed
HQsed
Cbs
CSCLsed
=
=
=
Hazard quotient for contact with sediment
Total pollutant concentration in sediment
(milligrams per liter [mg/L])
Ecological benchmark for sediment
(milligrams per kilograms [mg/kg])
Output from Equation D-l
Water quality module output
Equation C-5
Receptor-specific benchmark
(see Table D-l)
Adverse Effects to Piscivorous Wildlife. EPA determined the potential negative impact to
piscivorous wildlife (i.e., wildlife that consume fish) from the ingestion of contaminated fish by
calculating fish tissue concentrations and comparing these concentrations to ecological
benchmarks. Equation D-2 calculates pollutant concentrations in fish for the evaluated pollutants,
except for mercury. Because the more toxic form of mercury is methylmercury, EPA used
Equation D-3 for this pollutant [U.S. EPA, 2005b]. Equation D-3 estimates the concentration of
methylmercury in fish tissue, as opposed to total mercury.
EQUATION D-2
EQUATION D-3
Where:
CfishT ~
BCFn
CfishT = (0.15 xCdw)xBCFT
CfishT
CxWC
^dw
0.15
BCFx
=
=
=
Pollutant concentration in fish (wet weight),
where T represents trophic level T3 or T4
(mg/kg)
Total pollutant concentration in water (mg/L)
Dissolved pollutant concentration in water
(mg/L)
Fraction of dissolved total mercury as
dissolved methylmercury (unitless)
Bioconcentration factor or bioaccumulation
factor for specified trophic level (liters per
kilogram [L/kg])
Output from Equation D-2 or
Equation D-3
Water quality module output
Equation C-3
Water quality module output
Equation C-4
Model assumption value [U.S.
EPA, 2005b]
Pollutant-specific value
(see Table D-2)
EPA compared the calculated T3 fish tissue concentration to the ecological benchmark for
minks and the calculated T4 fish tissue concentration to the ecological benchmark for eagles. EPA
selected NEHC benchmarks for minks and eagles (Table D-3) as the ecological benchmarks for
piscivorous wildlife. The wildlife module expresses this comparison as an HQ. EPA used Equation
D-4 to calculate HQ values for arsenic, cadmium, chromium (VI), copper, lead, mercury (as
methylmercury), nickel, selenium, thallium, and zinc.
D-2
-------
Appendix D—Wildlife Module Methodology
EQUATION D-4
Where:
NEHC
HQi
CfishT
NEHC
=
=
Hazard quotient for ingestion offish
Pollutant concentration in fish (wet weight),
where T represents trophic level T3 or T4
(mg/kg)
No effect hazard concentration (|ig/g)
Output from Equation D-4
Output from Equation D-2 or
Equation D-3
Receptor- and pollutant-
specific (see Table D-3)
Table D-l. CSCL Benchmarks for Sediment Biota
Pollutant in Wildlife
Impact Assessment
Arsenic
Cadmium
Chromium (VI)
Copper
Lead
Mercury
Nickel
Selenium
Thallium
Zinc
CSCL Benchmark
Value (mg/kg)
5.90
0.596
37.3
35.7
35
0.174
18.0
None identified
None identified
123
Notes
No benchmark for chromium VI. EPA used the total chromium
benchmark, which may underestimate the impact to wildlife.
EPA compares the mercury, not methylmercury, concentration in
the sediment to the benchmark.
EPA could not complete the analysis for this pollutant - no
benchmark for comparison.
Source: MacDonald, D.D.; C. G. Ingersoll; and T. A. Berger. Development and Evaluation of Consensus-Based
Sediment Quality Guidelines for Freshwater Ecosystems. Archives of Environmental Contamination and
Toxicology 2000, 39(1)20 (as cited in NOAA, 2008).
a - The benchmarks used for the analysis are threshold effect levels (TELs).
D-3
-------
Appendix D—Wildlife Module Methodology
Table D-2. Bioconcentration Factors (BCFs) and Bioaccumulation Factors (BAFs) for
Trophic Level 3 (T3) and Trophic Level 4 (T4) Fish
Pollutant
Arsenic
Cadmium
Chromium (VI)
Copper a
Lead
Methylmercury
Nickel b
Selenium
Thallium
Zinc
BCF or BAF
BCF
BCF
BCF
BCF
BAF
BAF
BCF
BAF
BCF
BCF
Factor for Trophic
Level 3 (T3) Fish
(L/kg)
4.00E+00
2.70E+02
6.00E-01
3.60E+01
4.60E+01
1.60E+06
0.8
4.90E+02
3.40E+01
3.50E+02
Factor for Trophic
Level 4 (T4) Fish
(L/kg)
4.00E+00
2.70E+02
6.00E-01
3.60E+01
4.60E+01
6.80E+06
0.8
1.70E+03
1.30E+02
3.50E+02
Source
Barrows et al, 1980
Kumadae/a/., 1972
Stephan, 1993
U.S. EPA, 1980
Stephan, 1993
U.S. EPA, 1997a
Stephan, 1993
Lemly, 1985a
Barrows et al., 1980
and Stephan, 1993
Murphy et al., 1978
a - BCF not specific to a particular trophic level; applies to fish consumed by humans.
b - Nickel (soluble salts).
Table D-3. NEHC Benchmarks for Mink and Bald Eagles
Pollutant in
Wildlife Impact
Assessment
Arsenic
Cadmium
Chromium (VI)
Copper
Lead
Methylmercury
Nickel,
Selenium
Thallium
Zinc
NEHC Benchmark
Value for Mink
(T3 Fish) (jig/g)
7.65
5.66
17.7
41.2
34.6
0.37
12.5
1.13
None identified
904
NEHC Benchmark
Value for Eagle
(T4 Fish) (jig/g)
22.4
14.7
26.6
40.5
16.3
0.5
67.1
4
None identified
145
Notes
No benchmark for chromium VI. EPA used the
total chromium benchmark, which may
underestimate the impact to wildlife.
No benchmark for methylmercury. EPA used the
total mercury benchmark, which may
underestimate the impact to wildlife.
EPA could not complete the analysis for this
pollutant - no benchmark for comparison.
Source: USGS, 2008.
D-4
-------
Appendix D—Wildlife Module Methodology
IRW Model: Wildlife Module Methodology Limitations and Assumptions
EPA was required to make assumptions about various inputs, resulting in limitations with
respect to the wildlife module output and interpretation. Variability occurs from heterogeneous
characteristics, such as body weight differences within a population or the contaminant levels in
the environment. Uncertainty represents a lack of knowledge about factors such as the adverse
effects from exposure to pollutants. The assumptions and limitations of the wildlife module include
the following:
• Additive Risks Across Pathways. The wildlife module does not consider additive risks
across pathways. For example, the modeled impacts to wildlife from ingesting
contaminated fish do not consider the risk from direct contact with surface water. The
receptors chosen for the wildlife ingestion model, minks and eagles, do not spend large
amounts of time in contact with the surface water; therefore, not including the impact
of direct contact with surface water should only minimally underestimate the impacts.
In addition, the wildlife module does not consider the impact from water ingestion.
Because many of the pollutants considered in this analysis are bioaccumulative in
nature, the model considers only ingestion of the food source since it is likely the dose
from the food source dominates the dose from water ingestion.
• Use of BCFs and BAFs. Where available, EPA used BAFs to represent the
accumulation of pollutants in fish tissue (e.g., for selenium and methylmercury).
Otherwise, EPA used BCFs, which do not account for accumulation of pollutants via
the food web. For certain pollutants, exposure via the aquatic food web can be more
significant than exposure via ingestion of water.2 The result of this limitation on the
wildlife module output for those pollutants that use a BCF is an under-representation
of pollutant bioaccumulation in fish tissue where exposure via the aquatic food web is
significant. However, BCFs are useful in a screening-level assessment and appropriate
for a national-level environmental assessment (EA) where site-specific data are not
available and collection of site-specific data is not viable. The limitation of using a
single, national-level BAF/BCF is unknown due to site-specific considerations.
• Receptor Populations Evaluated. EPA considered the limitations and made multiple
assumptions in choosing receptor populations to evaluate. First, EPA assumed that,
because this is a national model, the receptor species and receiving water occur together
(i.e., all receiving waters evaluated in the wildlife module are habitat for the receptor
species even though that may not always be the case). In addition, due to the scope of
the project, EPA considered a limited number of species for use as receptors. For the
wildlife receptors, EPA chose minks and eagles due to their national distribution and
data available to conduct the analysis [USGS, 2008]. By choosing a limited number of
species, the wildlife module inherently excludes the impacts to critical assessment
endpoints such as threatened and endangered species. EPA attempts to address this
2 EPA Office of Water Health and Ecological Criteria Division agrees that all the routes (e.g., food, sediment, and
water) by which fish and shellfish are exposed to highly bioaccumulative pollutants may be important in
determining the accumulation in fish tissue and the subsequent transfer to human receptors. In addition, EPA agrees
that distributions of BAFs/BCFs may be better than single BAFs/BCFs because they account for changes in
bioaccumulation/bioconcentration rates at different water concentrations. EPA is working to develop BAF/BCF
distributions for several pollutants to better represent the bioaccumulation in aquatic organisms.
-------
Appendix D—Wildlife Module Methodology
limitation in the impact assessment by presenting a proximity analysis of steam electric
power plants to habitats of threatened and endangered species (see Section 3.4.5 of this
report) and an evaluation of the ecological risk to aquatic organisms and avian receptors
from selenium contamination (see Section 5.2 of this report).
Wildlife Receptor Diet. To provide an environmentally protective estimate of dietary
pollutant exposure, the wildlife module assumes that the diet of adult minks and bald
eagles consists entirely offish inhabiting the immediate receiving waters. EPA believes
this assumption is reasonable based on the following two factors: 1) It is possible that
in some habitats the dietary composition for both minks and eagles consists largely of
fish and EPA aims to be protective of wildlife across all habitats. For example, studies
have shown dietary composition as high as 75 and 85 percent fish for bald eagles and
minks, respectively [U.S. EPA, 1993]. In addition, it is likely that the other organisms
consumed by minks and eagles are also contaminated with the pollutants of concern
and are unaccounted for in the model; and 2) With respect to home ranges, the case
study water quality modeling results (see Section 8) demonstrate that pollutants
discharged from steam electric power plants can continue to occur at elevated levels
downstream from the immediate receiving waters, contaminating fish outside of
immediate receiving waters and resulting in additional potential for pollutant exposure
among piscivorous wildlife. Overall, however, this assumption likely results in a
potential overestimation of exposure to the modeled species.
Bioavailability and Speciation of Pollutants. The IRW model assumes that all forms
of a pollutant are equally bioavailable to ecological receptors. Therefore, data inputs
for the wildlife module include total pollutant concentration in the water column (i.e.,
dissolved plus particles sorbed to suspended sediment) or sediment concentration for
all pollutants analyzed, except where noted. In addition, some pollutant forms are more
toxic to organisms, such as various forms of arsenic. While different forms of arsenic
exist in the water column, it is not possible to determine the percentages of each due to
the complexities of the chemistry of a particular waterbody. Because of bioavailability
and pollutant speciation assumptions made for the wildlife impact assessment, the
impact to receptors may be over- or underestimated.
Indirect Ecological Effects. The wildlife module does not consider indirect ecological
effects, such as depletion of food sources. Such indirect effects are difficult to assess
and are thought to have minimal impact on some wildlife species because the impacted
receiving water is only a small portion of the species' habitat. In addition, many species
will move into other areas in search of prey if food sources in their current habitat
decline.
Full Mixing Effects for Receiving Water. The water quality module assumes that the
receiving waterbody is fully mixed. In reality, the water in lakes might stratify,
especially if they are deep enough. Chemical speciation, mostly based on pH, varies by
strata; for example, if the hypolimnion (i.e., lowest stratum of a lake) has a much lower
pH than the epilimnion (i.e., upper stratum), the concentration or speciation of many
pollutants may vary between the two layers. Therefore, bottom-dwelling organisms
would be exposed to different species and concentrations of pollutants. Due to the
complexity of these relationships and necessity for site-specific data, none of the impact
analyses considered stratification of receiving waters. The result of this limitation on
the wildlife module outputs is unknown.
-------
Appendix D—Wildlife Module Methodology
Multiple Pollutant Exposures. According to EPA's Steam Electric Power Generating
Point Source Category: Final Detailed Study Report [U.S. EPA, 2009b], receptors will
be exposed to multiple constituents simultaneously. However, the wildlife module
examines the impact of individual pollutants to receptors and does not take into account
how the interaction of multiple pollutants impacts the receptors. For example, EPA did
not consider the impact of mercury on the uptake or toxicity of selenium. There is
evidence in the literature that these two compounds interact with each other in the
environment and may decrease the level of impact of each pollutant on a receptor;
conversely, the interaction of other pollutants may increase the impact to a receptor.
However, because benchmarks are based on the toxicity of individual chemicals, and
the relationships between chemicals are complex, it is beyond the scope of this analysis
to include the effects of multiple pollutant interactions on receptors.
Ecological Benchmarks. EPA used ecological benchmarks as described above to
determine impacts to aquatic organisms from direct contact with contaminated
sediment. The benchmarks represent threshold effect levels TELs. If an organism
ingests chemical concentration above the TEL, some effect (or response) will be
produced. If the concentration ingested is below the TEL, no effect (or response) will
occur. The TEL represents the concentration of a chemical that would result in "no
effect," therefore the results presented in EA report are a more environmentally
protective impact estimate [USGS, 2008].
D-7
-------
Appendix E—Human Health Module Methodology
APPENDIX E
HUMAN HEALTH MODULE METHODOLOGY
This appendix presents the model equations, input variables, benchmarks, and
methodology limitations/assumptions for the immediate receiving water (IRW) model human
health module. Human health impacts include the following receptor groups:
• Child cohorts (recreational) that consume fish exposed to pollutants as a result of
discharges from steam electric power plants.
• Child cohorts (subsistence) that consume fish exposed to pollutants as a result of
discharges from steam electric power plants.
• Adult cohorts (recreational) that consume fish exposed to pollutants as a result of
discharges from steam electric power plants.
• Adult cohorts (subsistence) that consume fish exposed to pollutants as a result of
discharges from steam electric power plants.
In addition to the national-scale cohorts evaluated as part of the environmental assessment
(EA), EPA also estimated annual-average daily dose of pollutants for human receptors based on
race and Hispanic origin as an environmental justice analysis.
EPA estimated pollutant concentrations in fish tissue using the IRW model wildlife module
(see Appendix D). The human health module uses these concentrations as inputs.
Model input requirements for the equations presented in Appendix E can be divided into
five major categories: 1) input variable described by another equation; 2) site-specific input
variable; 3) model assumption variable; 4) receptor cohort-specific variable; and 5) pollutant-
specific variable. The following tables in Appendix E describe the input requirements and data
sources used in the human health module:
• Table E-l. Calculation of Consumption Ratio for Trophic Level 3 (Fxs) and Trophic
Level 4 (Fi4) Fish.
• Table E-2. Model Assumption Input Variables for the Human Health Module.
• Table E-3. Receptor Cohort-Specific Input Variables for the Human Health Module.
• Table E-4. Environmental Justice Analysis: Receptor Cohort-Specific Consumption
Rate by Race or Hispanic Origin for the Human Health Module.
• Table E-5. Pollutant-Specific Input Variables in the Human Health Module.
IRW Model: Human Health Module Equations
EPA estimated the pollutant concentrations in fish fillets consumed by humans (i.e., dose)
using an assumed consumption ratio of T3 and T4 fish and site-specific pollutant concentrations
in fish. For each cohort, EPA calculated the average daily dose (ADD) of the pollutant from eating
fish and compared this ADD to non-cancer human health benchmarks (i.e., reference doses
[RfDs]). The human health module expresses this comparison as a hazard quotient (HQ). An HQ
of higher than one (i.e., pollutant dosage exceeds benchmark) indicates a potential non-cancer
-------
Appendix E—Human Health Module Methodology
threat to the human cohort. EPA also calculated a lifetime average daily dose (LADD) and a
corresponding lifetime excess cancer risk (LECR) for each cohort. This study used the 1-in-a-
million cancer risk benchmark as an acceptable risk threshold when evaluating exposures
associated with fish consumption.
EPA used the equations presented below to calculate the pollutant concentration in the fish
fillet; the ADD for arsenic, cadmium, chromium (VI), copper, lead, mercury, nickel, selenium,
thallium, and zinc; the associated non-cancer threat HQ; and the LADD and LECR values for
arsenic.
EQUATION E-l
Where:
C
fish fillet ~rX3 A MIshXSF "rrX4 A *-fishX4F
Cfishjillet
CfishT3F
CfishT4F
Fl3
FT4
=
=
=
Average fish fillet concentration ingested by
humans (milligrams per kilograms [mg/kg])
Concentration of contaminant in fish at
trophic level 3 (mg/kg)
Concentration of contaminant in fish at
trophic level 4 (mg/kg)
Fraction of trophic level 3 fish intake
(unitless)
Fraction of trophic level 4 fish intake
(unitless)
Output from Equation E-l
Site-specific wildlife module
output Equation D-2 and
Equation D-3
Site-specific wildlife module
output Equation D-2 and
Equation D-3
Model assumption value of
0.36 (see calculation below)
Model assumption value of
0.64 (see calculation below)
To determine the fraction of T3 and T4 fish intake for human cohorts, EPA started with
the data presented in the 2011 Emissions Factor Handbook, Table 10-74 [U.S. EPA, 201 lb]. EPA
then completed the following analysis:
1. Assigned trophic levels to fish if not already listed in the table.
2. Totaled the quantities offish consumed by trophic level.
3. Determined fraction offish consumed at each trophic level.
Table E-l documents the data and analysis performed. EPA chose to use the factors for
fish intake that corresponded to rivers and streams; this is the most common receiving water source
in the IRW model.
E-2
-------
Appendix E—Human Health Module Methodology
Table E-l. Calculation of Consumption Ratio for Trophic Level 3 (Fxs) and Trophic Level 4
(FT4> Fish
Species
Landlocked salmon
Atlantic salmon
Togue (Lake trout)
Brook trout
Brown trout
Yellow perch
White perch
Bass (Smallmouth and
largemouth)
Pickerel
Lake whitefish
Hornpout (Catfish and
bullheads)
Bottom fish (Suckers,
carp and sturgeon)
Chub
Smelt
Other
TOTALS
Trophic
Level
4
4
4
4
4
o
J
o
J
4
3
o
J
o
J
3
o
J
o
J
4
Ice Fishing
Count of
Fish
Consumed
832
3
483
1,309
275
235
2,544
474
1,091
111
47
50
0
7,808
201
15,463
Mass
Consumed
(kg)
290
1.1
200
100
54
9.1
160
120
180
20
8.2
81
0
150
210
1,583
Lakes and Ponds
Count of
Fish
Consumed
928
33
459
3,294
375
1,649
6,540
73
553
558
1,291
62
252
428
90
16,587
Mass
Consumed
(kg)
340
9.9
160
210
56
52
380
5.9
91
13
100
22
35
4.9
110
1,590
Rivers and Streams
Count of
Fish
Consumed
305
17
33
10,185
338
188
3,013
787
303
55
180
100
219
4,269
54
20,046
Mass
Consumed
(kg)
120
11
2.7
420
23
7.4
180
130
45
2.7
7.8
6.7
130
37
45
1,168
Totals by Trophic Level
T3 Total
T4 Total
11,886
3376
608
765.1
11,333
5162
698
781.8
8,327
11665
417
751.7
Calculation of Factors by Trophic Level
T3 Factor
T4 Factor
0.77
0.22
0.38
0.48
0.68
0.31
0.44
0.49
0.42
0.58
0.36
0.64
Source: U.S. EPA, 201 Ib.
Bold indicates factors selected for the human health model.
Equation E-2 calculates the ADD, which is the daily intake of the contaminant from fish
ingestion. Based on a literature review (including EPA and Agency for Toxic Substances and
Disease Registry (ATSDR) references), arsenic in fish is mostly in the organic form and not
harmful to humans. The inorganic form of arsenic is harmful to humans; EPA's 1997 document,
Arsenic and Fish Consumption, reported the inorganic arsenic concentration in fish is between 0.4
- 4 percent of the total arsenic accumulating in fish. EPA estimated the inorganic arsenic
E-3
-------
Appendix E—Human Health Module Methodology
concentration in fish by assuming 4 percent of the total arsenic is inorganic. EPA used the
inorganic arsenic concentration in fish to determine human health impacts. The human health
model multiplies the Cfishjuiet concentration by 4 percent for arsenic (converting concentration
from total to inorganic).
Equation E-3 calculates the LADD, based on the ADD. Arsenic is the only carcinogenic
pollutant included in the EA. The model calculates the LADD of arsenic for each child cohort (six
recreational and six subsistence) and for each adult cohort (one recreational and one subsistence).
EPA assumed the exposure durations (ED) for use in the LADD calculation are equal to the length
of time in that cohort range. EPA selected an exposure frequency of 350 days per year, assuming
residents take an average of two weeks of vacation away from their homes each year.
Equation E-4 calculates the non-cancer HQ, based on the ADD.
Equation E-5 calculates the LECR for inorganic arsenic, based on the LADD.
EQUATION E-2
ADD
-fish fillet
x CR
-fish
1,000 x BW
Where:
ADD
Cfishjillet
CRfish
Ffish
1,000
BW
=
=
=
=
=
Daily dose of pollutant from fish ingestion
(mg/kg BW/day)
Average fish fillet concentration ingested by
humans (mg/kg)
Consumption rate offish (g ww/day)
Fraction offish intake from contaminated
source
Conversion factor (grams per kilograms
[g/kg])
Body weight (kg)
Output from Equation E-2
Output from Equation E-l
Receptor cohort-specific
value (see Table E-3 and
Table E-4)
Model assumption value of 1
Conversion factor
Receptor cohort-specific
value (see Table E-3)
E-4
-------
Appendix E—Human Health Module Methodology
EQUATION E-3
Where:
LADD
ADD x ED x EF
AT x 365
LADD
ADD
ED
EF
AT
365
=
=
=
=
=
Lifetime average daily dose (mg/kg BW/day)
Daily dose of pollutant from fish ingestion
(mg/kg BW/day)
Exposure duration for oral ingestion (yr)
Exposure frequency (days/yr)
Averaging time (yr)
Conversion factor (days/yr)
Output from
Equation E-3
Output from Equation E-2
Receptor cohort-specific
value (assumed value)
(see Table E-3)
Model assumption value of
350
Model assumption value of 70
[U.S. EPA, 201 Ib]
EQUATION E-4
Where:
HQ
ADD
RfD
HQ
ADD
RfD
=
=
=
Hazard quotient
Daily dose of pollutant from fish ingestion
(mg/kg BW/day)
Non-cancer reference dose (mg/kg BW/day)
Output from Equation E-4
Output from Equation E-2
Pollutant-specific value
(see Table E-5)
EQUATION E-5
Where:
LECR = LADD x CSF
LECR
LADD
CSF
=
=
=
Lifetime excess cancer risk
Lifetime average daily dose (mg/kg BW/d)
Cancer slope factor (mg/kg BW/day)"1
Output from Equation E-5
Output from
Equation E-3
Pollutant-specific value
(see Table E-5)
E-5
-------
Appendix E—Human Health Module Methodology
IRW Model: Human Health Module Inputs and Benchmarks
Table E-2. Model Assumption Input Variables for the Human Health Module
Input
Variable
FT3
FT4
Fflsh
EF
AT
Description
Fraction of trophic level 3 fish intake
Fraction of trophic level 4 fish intake
Fraction of fish intake from
contaminated source
Exposure frequency (days/yr)
Averaging time (yr)
Assumed
Value
0.36
0.64
1
350
70
Assumption Rationale/Data Source
U.S. EPA, 20 lib
U.S. EPA, 20 lib
EPA assumed that all fish consumed by the
receptor is from the contaminated surface water.
EPA assumed that the fisher travels away from
home for 15 days per year and does not eat fish
from contaminated surface water during that
period.
U.S. EPA, 20 lib
For the EA and benefits analyses,1 EPA focused on human exposure to contaminated fish
for recreational and subsistence fishers. Recreational fishers are non-commercial, non-subsistence
fishers and are more vulnerable to pollutant exposure by intake of contaminated fish from a
specific waterbody compared to the general population. Subsistence fishers are individuals who
consume fresh caught fish as a major food source. Intake rates for subsistence fishers are generally
higher than for the general population, and subsistence fishers are more vulnerable to pollutant
exposure by intake of contaminated fish from a specific waterbody compared to both recreational
fishers and the general population. Because of the focus of human exposure to a subset of the
general population that more frequently consume local fish, EPA selected fish consumption rates
from studies based on "consumer only" data. Consumer-only fish consumption rates are the
average intake rates across only those individuals that consumed fish and shellfish during the
survey time period. See the memorandum "Fish Consumption Rates Used in the Environmental
Assessment Human Health Module" for further details [ERG, 2015g].
The human health module calculates annual-average daily doses of pollutants for
recreational and subsistence fishers and does not calculate the annual-average daily doses of
pollutants for the general population. In its benefits analysis (see the Benefits and Cost Analysis),
EPA only evaluates impacts to a subset of the population living near the immediate and
downstream receiving waters.
The EPA document, Methodology for Deriving Ambient Water Quality Criteria for the
Protection of Human Health (Table 5-1) determined protective fish intake rates using the following
percentiles by fisher type: 1) general population and recreational fisher: 90th percentile of per capita
data and 2) subsistence fisher: 99th percentile of per capita data [U.S. EPA, 2000c]. The document
does not provide guidance on which percentiles to use for consumer-only fish intake rates.
Therefore, EPA used best professional judgment and using the following percentiles by fisher type:
1) recreational fisher: mean of consumer-only data and 2) subsistence fisher: 95th percentile of
consumer-only data.
1 See the Benefits and Cost Analysis for the Effluent Limitations Guidelines and Standards for the Steam Electric
Power Generation Point Source Category (EPA-821-R-15-005) (Benefits and Cost Analysis).
-------
Appendix E—Human Health Module Methodology
Table E-3. Receptor Cohort-Specific Input Variables for the Human Health Module
Receptor
Child
Recreational
Fisher
Child
Subsistence
Fisher
Cohort a
1 to <2 years
2 to <3 years
3 to <6 years
6 to <11 years
11 to < 16 years
16 to <21 years
1 to <2 years
2 to <3 years
3 to <6 years
6 to <11 years
11 to < 16 years
16 to <21 years
Adult Recreational Fisher °
Adult Subsistence Fisher °
Body Weight
(kg)a
11.4
13.8
18.6
31.8
56.8
71.6
11.4
13.8
18.6
31.8
56.8
71.6
80
80
Consumption
Rate (g/kg-day) b
1.60
1.60
1.30
1.10
0.660
0.660
4.90
4.90
3.60
2.90
1.70
1.70
0.665
2.05
Consumption
Rate (g/day) b
18.2
22.1
24.2
35.0
37.5
47.3
55.9
67.6
67.0
92.2
96.6
121.7
53.2
164
Exposure
Duration (years)
1
1
3
5
5
5
1
1
3
5
5
5
49
49
Sources: U.S. EPA, 2008a; U.S. EPA, 201 Ib.
Acronyms: g/day (grams per day); g/kg-day (grams per kilogram body weight per day); kg (kilograms).
a - The child cohort age ranges correspond to the ranges provided in the 2008 Child-Specific Exposure Factor
Handbook (EFH) for body weights [U.S. EPA, 2008a].
b - EPA determined consumption rates for child cohorts using data from Table 10-1 (Recommend Per Capita and
Consumer-Only Values for Fish Intake) for finfish consumption [U.S. EPA, 201 Ib]. EPA used consumer-only fish
consumption rates: mean values for recreational fishers and 95th percentile values for subsistence fishers. EPA
converted the listed consumption rate (g/kg-day) to g/day by multiplying by mean body weight for each cohort as
listed in U.S. EPA, 2008b [ERG, 2015g]. Fish intake rates provided in the reference [U.S. EPA, 201 Ib] are
recommended for the consumer-only population; the selection of consumption rates for exposure assessment
purposes may vary depending on the exposure scenarios being evaluated.
c - Table 10-1 [U.S. EPA, 201 Ib] presented multiple adult groups. EPA used the average fish consumption rate for
age groups "21 to <50 years" and "50+ years" to calculate a single adult cohort fish consumption rate.
E-7
-------
Appendix E—Human Health Module Methodology
Table E-4. Environmental Justice Analysis: Receptor Cohort-Specific Input Consumption Rate by Race or Hispanic Origin for the
Human Health Module
Receptor
Recreational
Subsistence
Race or Hispanic Origin
Non-Hispanic White
Non-Hispanic Black
Mexican-American
Other Hispanic
Other, including Multiple Races
Non-Hispanic White
Non-Hispanic Black
Mexican-American
Other Hispanic °
Other, including Multiple Races °
CRfish,
g/kg-day
(All ages) a
0.67
0.77
0.93
0.82
0.96
1.9
2.1
2.8
2.7
3.6
Consumption Rate (CRfish), g/day, by Cohort b
lto<2
years
7.64
8.78
10.6
9.35
10.9
21.7
23.9
31.9
30.8
41.0
2to<3
years
9.25
10.6
12.8
11.3
13.2
26.2
29.0
38.6
37.3
49.7
3to<6
years
12.5
14.3
17.3
15.3
17.9
35.3
39.1
52.1
50.2
67.0
6 to <11
years
21.3
24.5
29.6
26.1
30.5
60.4
66.8
89.0
85.9
114
11 to <16
years
38.1
43.7
52.8
46.6
54.5
108
119
159
153
204
16 to <21
years
48
55.1
66.6
58.7
68.7
136
150
200
193
258
Adult
53.6
61.6
74.4
65.6
76.8
152
168
224
216
288
Source: U.S. EPA, 201 Ib.
Acronyms: CR&h (consumption rate); g/day (grams per day); g/kg-day (grams per kilogram body weight per day)
a - For recreational fishers, EPA used the mean, consumer-only fish consumption rate for finfish (excludes shellfish). For subsistence fishers, EPA used the 95th
percentile, consumer-only fish consumption rate for finfish (excludes shellfish). See Table 10-8 of U.S. EPA, 201 Ib.
b - Consumption rates provided as single value by race and Hispanic origin (as g/kg-day). EPA multiplied these values by cohort-specific body weights, as listed in Table
E-3, to calculate a cohort-specific consumption rate in g/day. Numbers presented as three significant digits.
c - Consumption rates for this race or Hispanic origin are less statistically reliable due to the comparatively smaller data set.
E-8
-------
Appendix E—Human Health Module Methodology
Table E-5. Pollutant-Specific Benchmarks for the Human Health Module
Pollutant in Human Health
Impact Assessment
Arsenic, inorganic
Cadmium, total
Chromium (VI)
Copper
Lead, total
Methylmercury
Nickel, total
Selenium, total
Thallium, total
Zinc, total
RfD
(mg/kg-day)
3.00E-04
l.OOE-03
3.00E-03
l.OOE-02
None available
l.OOE-04
2.00E-02
5.00E-03
l.OOE-05
3.00E-01
CSF
(mg/kg-day) x
1.50E+00
Notes a
RfD and CSF for drinking water ingestion
RfD for food consumption
RfD for drinking water ingestion
Used the intermediate oral minimal risk
level (MRL) as the reference dose
[ATSDR, 2010a]
RfD for fish consumption only
RfD for soluble salts; used for food
consumption
RfD for food consumption
Used value cited in U.S. EPA, 2010a for
thallium chloride as the reference dose;
used for chronic oral exposure
RfD for food consumption
Acronyms: mg/kg-day
a - References include
pollutants.
(milligrams per kilogram body weight per day)
ATSDR, 2010a for copper; U.S. EPA, 2010a for thallium, and U.S. EPA, 201 Ic for all other
IRW Model: Human Health Module Limitations and Assumptions
The human health module limitations and assumptions include the following:
• Additive Risks Across Pathways. The human health module does not consider additive
risks across pathways. For example, the module assumes that the human population
consuming the fish is not also ingesting contaminated drinking water. Exposures from
fish consumption and drinking water are likely to occur over different time frames
(because of ground water travel) and may involve different receptors (e.g., a resident
near a receiving water exposed to ground water contamination may not be a recreational
fisher). Similarly, the module assumes that these populations are not coming in direct
contact with contaminated surface water or sediment through recreation. Based on
these assumptions, the model may underestimate total risk to human health from
combustion wastewater.
• Bioavailability and Speciation of Pollutants. The assumptions listed for the wildlife
module in Appendix D apply to pollutant concentrations modeled in fish and therefore
affect the human health impact assessment.
• Full Mixing Effects for Receiving Water. The assumptions listed for the wildlife
module in Appendix D apply to pollutant concentrations modeled in fish and therefore
affect the human health impact assessment.
• Multiple Pollutant Exposures. According to previous analyses and literature reviewed
[U.S. EPA, 2009b], people who ingest fish from impacted waters will be exposed to
E-9
-------
Appendix E—Human Health Module Methodology
multiple pollutants from the wastestreams evaluated. However, the module evaluates
each pollutant individually. Such an approach does not account for interactive effects
that might be associated with exposures to mixtures. For example, some pollutants may
have a higher risk when consumed together because of their interaction, whereas other
pollutants may have less impact on human health when consumed together. Due to the
complexity of these interactions and because benchmarks are based on the toxicity of
individual pollutants, it is not possible to examine these synergistic effects in this
analysis. Based on this limitation, risks of pollutants may be over- or underestimated.
Sources of Consumed Fish. The human health module assumes that all of the fish
consumed by recreational and subsistence fishers is caught from the immediate
receiving water, except during a two-week time period once per year. This assumption
potentially overestimates the annual-average daily dose of the pollutants for these
receptors, particularly for recreational fishers. The proportion of fish eaten by an
individual from local surface waters will vary (e.g., consumption rate estimates in
studies might include seafood purchased from a grocery store and not locally caught).2
Human Exposure Factors. Individual exposure factors, such as ingestion rate, body
weight, and exposure duration, are variable due to the physical characteristics,
activities, and behavior of the individual. EPA used the most current data regarding
exposure assumptions, and these values represent EPA's current guidance on exposure
data [U.S. EPA, 2008a; U.S. EPA, 201 lb].
Human Health Benchmarks. Uncertainties generally associated with human health
benchmarks are discussed in detail in EPA's Guidelines for Carcinogen Risk
Assessment [U.S. EPA, 2005c] and Integrated Risk Information System (IRIS) [U.S.
EPA, 201 lc]. IRIS defines the RfD as "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 threat of deleterious
effects during a lifetime." RfDs are typically based on an assumption of lifetime
exposure and may not be appropriate when applied to less-than-lifetime exposure
situations [U.S. EPA, 201 lc]. The cancer slope factor is an estimate of the human
cancer risk per milligram of chemical per kilogram body weight per day. To calculate
the LADD used for the cancer risk assessment, EPA used the time in the cohort group
(i.e., 1, 3, or 5 years depending on child cohort and 49 years for adult cohort) as the
ED. The ED is the length of time exposure occurs at the concentration. This analysis
may over- or under-estimate the cancer risk if exposure is shorter than or longer than
the ED, respectively. LADDs are appropriate when developing screening-level
estimates; however, EPA recommends calculating that risk by integrating exposures or
risks through all life stages (e.g., chronic exposure for a child may occur across cohorts)
[U.S. EPA, 201 lb].
2 For the benefits analysis, EPA further defined the affected population (i.e., individuals potentially exposed to
steam electric power plant pollutants via consumption of contaminated fish) as recreational and subsistence fishers
who fish reaches that are affected by steam electric power plant discharges (including immediate receiving waters
and downstream reaches), as well as their household members. EPA estimated the number of people who are likely
to fish affected reaches based on typical travel distances to a fishing site, presence of substitute fishing locations,
data on the locations and status offish consumption advisories for affected reaches, and information on anglers'
awareness and adherence to those advisories. See the Benefits and Cost Analysis.
-------
Appendix F—Overview of Ecological Risk Modeling Setup and Outputs
APPENDIX F
OVERVIEW OF ECOLOGICAL RISK MODELING SETUP
AND OUTPUTS
This appendix summarizes the inputs, outputs, and methodology limitations/assumptions
for the ecological risk modeling that EPA used to evaluate reproductive risks associated with
dietary exposure to selenium. EPA performed ecological risk modeling for two sets of water
quality outputs:
• Dissolved selenium concentrations in the immediate receiving waters of all modeled
steam electric power plants, based on the outputs from the water quality module of the
national-scale immediate receiving water (IRW) model (see Appendix C).
• Dissolved selenium concentrations in the immediate receiving water and downstream
reaches of Black Creek, Mississippi, based on the outputs from the Black Creek case
study water quality model (see Appendix G).
Model input requirements for the ecological risk model can be divided into four major
categories: 1) dissolved selenium concentrations; 2) site-specific enrichment factors (EFs), which
represent the ratio of the concentration of selenium at the base of the food web (i.e., particulates)
to the dissolved concentration in water; 3) species-specific trophic transfer factors (TTFs), which
describe subsequent bioaccumulation by higher trophic-level aquatic organisms such as fish and
birds; and 4) exposure-response (ER) functions, which translate the modeled selenium
concentrations in fish and birds into the associated reduction in reproductive success.
The ecological risk modeling methodology is described in Section 5.2 of the EA report.
This modeling approach is consistent with the approach taken in developing the Draft Aquatic Life
Ambient Water Quality Criterion for Selenium - Freshwater [U.S. EPA, 2014f] (referred to as the
draft selenium criterion) and is based on the same data sets and studies for EF, TTFinvert, TTFssh,
and ERfish. For this EA, EPA expanded the model to include data sets for TTFmaiiard and ERmaiiard.
The following sections describe these inputs and their sources; summarize the ecological
risk modeling results; and discuss the limitations and assumptions associated with this modeling.
Dissolved Selenium Concentrations
As described above, the dissolved selenium concentrations for the national-scale and case
study ecological risk models are derived from the IRW water quality module and the Black Creek
case study water quality model, respectively. Dissolved selenium concentrations used in the
national-scale ecological risk model are provided in DCN SE04612.1 Dissolved selenium
concentrations used in the case study ecological risk model are provided in DCN SE04615. Prior
to use as inputs for the Black Creek case study ecological risk model, EPA calculated three-month
rolling averages of the dissolved selenium concentration output from the Black Creek case study
water quality model. This resulted in one average concentration for each calendar month
1 EPA removed identifying information, such as the immediate receiving water name and the steam electric power
plant name, from this reference to prevent disclosure of confidential business information (CBI).
-------
Appendix F—Overview of Ecological Risk Modeling Setup and Outputs
throughout the entire modeling period after the assumed compliance date for the Morrow
Generating Site (2019-2036). Use of a three-month rolling average avoided the calculation of
significantly elevated reproductive risks in response to short-term (e.g., daily or weekly)
fluctuations in the dissolved selenium concentration.
Enrichment Factors
As discussed in Section 5.2 of the EA report, the EFs used in the ecological risk modeling
effort are consistent with those used in developing the draft selenium criterion [U.S. EPA, 2014f].
This effort produced EF distributions for lentic systems (e.g., lakes, reservoirs, and ponds) and
lotic systems (e.g., rivers, creeks, and streams). These distributions are well described by
lognormal distributions with means (standard deviations) of 1,738 (2,499)2 for lentic systems and
692 (787) for lotic systems. These EF distributions are illustrated in Figure F-l and Figure F-2.
Trophic Transfer Factors
As discussed in Section 5.2 of the EA report, the TTFs used to represent selenium
bioaccumulation in invertebrates and fish in the national-scale ecological risk model are also
consistent with those used in developing the draft selenium criterion [U.S. EPA, 2014f]. This
resulted in a TTFinvert distribution with a mean (standard deviation) of 2.84 (2.49)3 and a TTFfish
distribution with a mean (standard deviation) of 1.6 (1.08). These TTF distributions are illustrated
in Figure F-l.
Based on a review of Ohlendorf [2003], EPA developed a TTF distribution for mallards.
The resulting TTFmaiiard distribution is best described by a triangular distribution, with a likeliest
value of 2.5, a minimum value of 0.4, and a maximum value of 4.1. This TTF distribution is
illustrated in Figure F-l.
For the Black Creek case study ecological risk model, EPA refined the TTFinvert and TTFfish
datasets to include only invertebrate and fish species that are representative of those collected
during surveys of Black Creek and other nearby rivers and streams as part of EPA's National
Aquatic Resource Survey (NARS). This resulted in smaller distributions that are more likely to
reflect bioaccumulation patterns within the species that actually inhabit Black Creek. These TTF
distributions are illustrated in Figure F-2.
Exposure Response Functions
To estimate the risk of negative reproductive effects among fish, EPA used the same
extensively peer-reviewed ER function (i.e., curve) as was used in the draft selenium criterion
[U.S. EPA, 2014f]. This ER function is illustrated in Figure F-3.
2 The EF for a given waterbody is the ratio of the concentration of selenium at the base of the food web (/'. e.,
participates) to the dissolved concentration in water, multiplied by 1,000. A mean EF of 1,738 for lentic systems
indicates that, on average, the concentration of selenium at the base of the food web is 1.738 times greater than the
dissolved concentration in water.
3 The TTF for a given trophic level is the ratio of the concentration in the organism to the concentration in the
consumed material or lower-trophic-level organism. A mean TTF of 2.84 for invertebrates indicates that, on
average, the concentration of selenium in the tissues of invertebrates is 2.84 times greater than the concentration in
particulates consumed by invertebrates.
-------
Appendix F—Overview of Ecological Risk Modeling Setup and Outputs
To develop the ER function for mallards, EPA fit a logistic curve to the combined, control
normalized data from six different laboratory studies that evaluated the effect of selenium on
mallard egg hatchability [Hdnzetal., 1987, 1989; Heinz and Hoffman, 1996, 1998; Stanley ef or/.,
1994, 1996]. This ER function is illustrated in Figure F-4.
F-3
-------
Appendix F—Overview of Ecological Risk Modeling Setup and Outputs
Lotic EF Distribution
Lognormal distribution with parameters:
Location 33.00
Mean 692.00
Std. Dev. 787.00
Selected range is from 70.00 to 3,333.00
3333
Lentic EF Distribution
Lognormal distribution with parameters:
Location 21.00
Mean 1,738.00
Std. Dev. 2,499.00
Selected ranee is from 95.00 to 12,476.00
95.00
12,476.
00
Invertebrate TTF (TTFinverl) Distribution
Lognormal distribution with parameters:
Location 0.12
Mean 2.84
Std. Dev. 2.49
Selected range is from 0.13 to 20.47
Fish TTF (TTFfish) Distribution
Lognormal distribution with parameters:
Location 0.04
Mean 1.63
Std. Dev. 1.08
Selected range is from 0.08 to 58.00
0.28
15.57
20.47
0.21
5.46
8.86
Mallard TTF (TTF,,,^,^,,) Distribution
Triangular distribution with parameters:
Minimum 0.36
Likeliest 2.50
Maximum 4.10
Selected range is from 0.36 to 4.10
0.36
2.62666
6667
Figure F-l. Input EF and TTF Distributions for National-Scale Ecological Risk Model
Baseline and Final Rule (Option D)
F-4
-------
Appendix F—Overview of Ecological Risk Modeling Setup and Outputs
Lotic EF Distribution
Lognormal distribution with parameters:
Location 33.00
Mean 692.00
Std. Dev. 787.00
Selected range is from 70.00 to 3,333.00
70.00
Fish TTF (TTFnsh) Distribution
Lognormal distribution with parameters:
Location 0.04
Mean 1.76
Std. Dev. 1.25
Selected range is from 0.13 to 58.00
0.20
8.00
10.49
Dissolved Selenium (ug/L) Distribution - Baseline
Triangular distribution with parameters:
Minimum 0.97
Likeliest 1.82
Maximum 9.70
Selected range is from 0.97 to 9.70
0.97
6.26
9.70
Invertebrate TTF (TTFillvcrt) Distribution
Lognormal distribution with parameters:
Location 0.05
Mean 3.21
Std. Dev. 2.52
Selected range is from 0.24 to 19.33
0.29
14.71
19.33
Mallard TTF (TTFn,al]ard) Distribution
Triangular distribution with parameters:
Minimum 0.36
Likeliest 2.50
Maximum 4.10
Selected range is from 0.36 to 4.10
0.36
2.63
4.10
Dissolved Selenium (ug/L) Distribution - Final Rule
Triangular distribution with parameters:
Minimum 0.01
Likeliest 0.05
Maximum 0.28
Selected range is from 0.01 to 0.28
o.oi
0.18
0.28
Figure F-2. Input EF, TTF, and Dissolved Selenium Distributions for Morrow Generating
Site Immediate Receiving Water (Black Creek Case Study) Ecological Risk Model -
Baseline and Final Rule (Option D)
F-5
-------
Appendix F—Overview of Ecological Risk Modeling Setup and Outputs
ro
E
o
Q.
O
1.0
.9
.8
.7
.6
.5
.4
.3
.2
.1
0
i:
-.2
.2 .4 .6 .8 1.0 1.2 1.4
Log(mg Se/kg egg dw in fish)
1.6
1.8
Figure F-3. Exposure-Response Function for Fish Reproductive Success
O
"on
CO
CD
o
o
^
C/3
c
o
o
Q-
O
1.1
1.0
.9
.8
.7
.6
.5
.4
.3
.2
.1
0
.5 1.0
Log (mg Se/kg egg) in mallard
1.5
2.0
Figure F-4. Exposure-Response Function for Mallard Egg Hatchability
F-6
-------
Appendix F—Overview of Ecological Risk Modeling Setup and Outputs
Ecological Risk Model Outputs
Table F-l and Table F-2 summarize the results of the national-scale ecological risk model
for fish under baseline conditions and the final rule, respectively.
Table F-3 and Table F-4 summarize the results of the national-scale ecological risk model
for mallards under baseline conditions and the final rule, respectively.
Table F-5 and Table F-6 summarize the results of the case study ecological risk model for
birds and mallards, respectively, under baseline conditions. Under the final rule, none of the
modeled stream segments resulted in a modeled risk of greater than 0.1 percent for either fish or
mallards.
F-7
-------
Appendix F—Overview of Ecological Risk Modeling Setup and Outputs
Table F-l. Number (and Percentage) of Receiving Waters in National-Scale Ecological Risk
Model with Selenium-Driven Reproductive Effects in Fish - Baseline
Percentile a
Lakeb
River b
Total b
1 Percent of Fish Population Experiencing Negative Reproductive Effects
10th:
25th:
Median:
75th:
90th:
95th:
0 (0%)
2 (7.7%)
4 (15%)
6 (23%)
8(31%)
8(31%)
14 (7.7%)
17 (9.3%)
24 (13%)
32 (17%)
36 (20%)
42 (23%)
14 (6.7%)
19(9.1%)
28 (13%)
38 (18%)
44 (21%)
50 (24%)
10 Percent of Fish Population Experiencing Negative Reproductive Effects
10th:
25th:
Median:
75th:
90th:
95th:
0 (0%)
1 (3.8%)
4 (15%)
6 (23%)
7 (27%)
8(31%)
12 (6.6%)
14 (7.7%)
20(11%)
29 (16%)
35 (19%)
39 (21%)
12 (5.7%)
15 (7.2%)
24(11%)
35 (17%)
42 (20%)
47 (22%)
50 Percent of Fish Population Experiencing Negative Reproductive Effects
10th:
25th:
Median:
75th:
90th:
95th:
0 (0%)
0 (0%)
3 (12%)
5 (19%)
6 (23%)
8(31%)
10 (5.5%)
14 (7.7%)
17 (9.3%)
27 (15%)
34 (19%)
35 (19%)
10 (4.8%)
14 (6.7%)
20 (9.6%)
32 (15%)
40 (19%)
43 (21%)
75 Percent of Fish Population Experiencing Negative Reproductive Effects
10th:
25th:
Median:
75th:
90th:
95th:
0 (0%)
0 (0%)
3 (12%)
5 (19%)
6 (23%)
7 (27%)
10 (5.5%)
14 (7.7%)
17 (9.3%)
26 (14%)
31(17%)
34 (19%)
10 (4.8%)
14 (6.7%)
20 (9.6%)
31(15%)
37 (18%)
41 (20%)
90 Percent of Fish Population Experiencing Negative Reproductive Effects
10th:
25th:
Median:
75th:
90th:
95th:
0 (0%)
0 (0%)
2 (7.7%)
5 (19%)
6 (23%)
6 (23%)
9 (4.9%)
13 (7.1%)
17 (9.3%)
22 (12%)
29 (16%)
34 (19%)
9 (4.3%)
13 (6.2%)
19(9.1%)
27 (13%)
35 (17%)
40 (19%)
Notes:
a - Percentile refers to the risk percentile. For example, values in the 90th percentile row indicate the numbers of
receiving waters whose selenium concentrations are high enough to result in a 10 percent probability of the indicated
reproductive effect.
b - The national-scale ecological risk model encompasses a total of 209 immediate receiving waters (183 rivers and
streams; 26 lakes, ponds, and reservoirs) and loadings from 188 steam electric power plants.
F-8
-------
Appendix F—Overview of Ecological Risk Modeling Setup and Outputs
Table F-2. Number (and Percentage) of Receiving Waters in National-Scale Ecological Risk
Model with Selenium-Driven Reproductive Effects in Fish - Final Rule (Option D)
Percentile a
Lakeb
River b
Total b
1 Percent of Fish Population Experiencing Negative Reproductive Effects
10th:
25th:
Median:
75th:
90th:
95th:
0 (0%)
0 (0%)
0 (0%)
0 (0%)
1 (3.8%)
1 (3.8%)
3 (1.6%)
5 (2.7%)
11(6%)
16 (8.7%)
21 (11%)
25 (14%)
3 (1.4%)
5 (2.4%)
11(5.3%)
16 (7.7%)
22(11%)
26 (12%)
10 Percent of Fish Population Experiencing Negative Reproductive Effects
10th:
25th:
Median:
75th:
90th:
95th:
0 (0%)
0 (0%)
0 (0%)
0 (0%)
0 (0%)
1 (3.8%)
3 (1.6%)
3 (1.6%)
8 (4.4%)
15 (8.2%)
19 (10%)
23 (13%)
3 (1.4%)
3 (1.4%)
8 (3.8%)
15 (7.2%)
19(9.1%)
24(11%)
50 Percent of Fish Population Experiencing Negative Reproductive Effects
10th:
25th:
Median:
75th:
90th:
95th:
0 (0%)
0 (0%)
0 (0%)
0 (0%)
0 (0%)
0 (0%)
3 (1.6%)
3 (1.6%)
6 (3.3%)
12 (6.6%)
19 (10%)
20(11%)
3 (1.4%)
3 (1.4%)
6 (2.9%)
12 (5.7%)
19(9.1%)
20 (9.6%)
75 Percent of Fish Population Experiencing Negative Reproductive Effects
10th:
25th:
Median:
75th:
90th:
95th:
0 (0%)
0 (0%)
0 (0%)
0 (0%)
0 (0%)
0 (0%)
2(1.1%)
3 (1.6%)
5 (2.7%)
9 (4.9%)
15 (8.2%)
19 (10%)
2 (0.96%)
3 (1.4%)
5 (2.4%)
9 (4.3%)
15 (7.2%)
19(9.1%)
90 Percent of Fish Population Experiencing Negative Reproductive Effects
10th:
25th:
Median:
75th:
90th:
95th:
0 (0%)
0 (0%)
0 (0%)
0 (0%)
0 (0%)
0 (0%)
2(1.1%)
3 (1.6%)
4 (2.2%)
9 (4.9%)
15 (8.2%)
19 (10%)
2 (0.96%)
3 (1.4%)
4 (1.9%)
9 (4.3%)
15 (7.2%)
19(9.1%)
Notes:
a - Percentile refers to the risk percentile. For example, values in the 90th percentile row indicate the numbers of
receiving waters whose selenium concentrations are high enough to result in a 10 percent probability of the indicated
reproductive effect.
b - The national-scale ecological risk model encompasses a total of 209 immediate receiving waters (183 rivers and
streams; 26 lakes, ponds, and reservoirs) and loadings from 188 steam electric power plants.
F-9
-------
Appendix F—Overview of Ecological Risk Modeling Setup and Outputs
Table F-3. Number (and Percentage) of Receiving Waters in National-Scale Ecological Risk
Model with Selenium-Driven Reproductive Effects in Mallards - Baseline
Percentile a
Lakeb
River b
Total b
1 Percent of Mallard Population Experiencing Hatching Failure
10th:
25th:
Median:
75th:
90th:
95th:
3 (12%)
5 (19%)
6 (23%)
8(31%)
9 (35%)
13 (50%)
18 (9.8%)
26 (14%)
34 (19%)
38 (21%)
47 (26%)
52 (28%)
21 (10%)
31(15%)
40 (19%)
46 (22%)
56 (27%)
65(31%)
10 Percent of Mallard Population Experiencing Hatching Failure
10th:
25th:
Median:
75th:
90th:
95th:
0 (0%)
3 (12%)
5 (19%)
6 (23%)
8(31%)
8(31%)
14 (7.7%)
17 (9.3%)
26 (14%)
32 (17%)
36 (20%)
42 (23%)
14 (6.7%)
20 (9.6%)
31(15%)
38 (18%)
44 (21%)
50 (24%)
50 Percent of Mallard Population Experiencing Hatching Failure
10th:
25th:
Median:
75th:
90th:
95th:
0 (0%)
0 (0%)
2 (7.7%)
4 (15%)
6 (23%)
6 (23%)
10 (5.5%)
13 (7.1%)
17 (9.3%)
22 (12%)
28 (15%)
34 (19%)
10 (4.8%)
13 (6.2%)
19(9.1%)
26 (12%)
34 (16%)
40 (19%)
75 Percent of Mallard Population Experiencing Hatching Failure
10th:
25th:
Median:
75th:
90th:
95th:
0 (0%)
0 (0%)
0 (0%)
3 (12%)
5 (19%)
6 (23%)
7 (3.8%)
10 (5.5%)
14 (7.7%)
17 (9.3%)
22 (12%)
27 (15%)
7 (3.3%)
10 (4.8%)
14 (6.7%)
20 (9.6%)
27 (13%)
33 (16%)
90 Percent of Mallard Population Experiencing Hatching Failure
10th:
25th:
Median:
75th:
90th:
95th:
0 (0%)
0 (0%)
0 (0%)
1 (3.8%)
4 (15%)
5 (19%)
3 (1.6%)
8 (4.4%)
12 (6.6%)
14 (7.7%)
18 (9.8%)
22 (12%)
3 (1.4%)
8 (3.8%)
12 (5.7%)
15 (7.2%)
22(11%)
27 (13%)
Notes:
a - Percentile refers to the risk percentile. For example, values in the 90th percentile row indicate the numbers of
receiving waters whose selenium concentrations are high enough to result in a 10 percent probability of the indicated
reproductive effect.
b - The national-scale ecological risk model encompasses a total of 209 immediate receiving waters (183 rivers and
streams; 26 lakes, ponds, and reservoirs) and loadings from 188 steam electric power plants.
F-10
-------
Appendix F—Overview of Ecological Risk Modeling Setup and Outputs
Table F-4. Number (and Percentage) of Receiving Waters in National-Scale Ecological Risk
Model with Selenium-Driven Reproductive Effects in Mallards - Final Rule (Option D)
Percentile a
Lakeb
River b
Total b
1 Percent of Mallard Population Experiencing Hatching Failure
10th:
25th:
Median:
75th:
90th:
95th:
0 (0%)
0 (0%)
0 (0%)
0 (0%)
0 (0%)
2 (7.7%)
7 (3.8%)
12 (6.6%)
19 (10%)
23 (13%)
26 (14%)
26 (14%)
7 (3.3%)
12 (5.7%)
19(9.1%)
23(11%)
26 (12%)
28 (13%)
10 Percent of Mallard Population Experiencing Hatching Failure
10th:
25th:
Median:
75th:
90th:
95th:
0 (0%)
0 (0%)
0 (0%)
0 (0%)
0 (0%)
0 (0%)
3 (1.6%)
6 (3.3%)
12 (6.6%)
17 (9.3%)
21 (11%)
25 (14%)
3 (1.4%)
6 (2.9%)
12 (5.7%)
17(8.1%)
21 (10%)
25 (12%)
50 Percent of Mallard Population Experiencing Hatching Failure
10th:
25th:
Median:
75th:
90th:
95th:
0 (0%)
0 (0%)
0 (0%)
0 (0%)
0 (0%)
0 (0%)
3 (1.6%)
3 (1.6%)
5 (2.7%)
9 (4.9%)
14 (7.7%)
18 (9.8%)
3 (1.4%)
3 (1.4%)
5 (2.4%)
9 (4.3%)
14 (6.7%)
18 (8.6%)
75 Percent of Mallard Population Experiencing Hatching Failure
10th:
25th:
Median:
75th:
90th:
95th:
0 (0%)
0 (0%)
0 (0%)
0 (0%)
0 (0%)
0 (0%)
2(1.1%)
3 (1.6%)
3 (1.6%)
6 (3.3%)
9 (4.9%)
13(7.1%)
2 (0.96%)
3 (1.4%)
3 (1.4%)
6 (2.9%)
9 (4.3%)
13 (6.2%)
90 Percent of Mallard Population Experiencing Hatching Failure
10th:
25th:
Median:
75th:
90th:
95th:
0 (0%)
0 (0%)
0 (0%)
0 (0%)
0 (0%)
0 (0%)
1 (0.55%)
2(1.1%)
3 (1.6%)
3 (1.6%)
6 (3.3%)
9 (4.9%)
1 (0.48%)
2 (0.96%)
3 (1.4%)
3 (1.4%)
6 (2.9%)
9 (4.3%)
Notes:
a - Percentile refers to the risk percentile. For example, values in the 90th percentile row indicate the numbers of
receiving waters whose selenium concentrations are high enough to result in a 10 percent probability of the indicated
reproductive effect.
b - The national-scale ecological risk model encompasses a total of 209 immediate receiving waters (183 rivers and
streams; 26 lakes, ponds, and reservoirs) and loadings from 188 steam electric power plants.
F-ll
-------
Appendix F— Overview of Ecological Risk Modeling Setup and Outputs
Table F-5. Risk of Selenium-Driven Reproductive Effects in Fish Downstream from Morrow Generating Site Immediate
Receiving Water (Black Creek Case Study) - Baseline
Black Creek WASP Model Segment ID b'c
Percentilea
39
38
37
36
35
34
33
32
31
30
29
28
27
10th:
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%
25th:
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%
Median: 0.381% <0.1%
0.1%
0.1%
0.1%
0.1%
0.1%
0.1%
7541
83.0%
90a
95th:
>99.9% 98.7% 98.3%
>99.9% >99.9% >99.9%
26
25
24
94.6%
>99.9%
23
93.4%
>99.9% 99.8% >99.9% >99.9% 94.2%
92.8% 80.6%
82.6%
99.7% 99.6%
22
21
20
19
18
17
16
15
14
10th:
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%
25th:
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%
Median:
0.1%
0.1%
0.1%
0.1%
0.1%
0.1%
0.1%
75th: 1.11%
0.226%
2.42%
2.39%
2.14%
1.82%
1.81%
2.41%
0.723%
0.330%
0.345%
0.331%
90th:
83.9%
95th:
13
12
11
10
80.1%
99.5%
8
99.6%
66.5%
98.9%
64.6%
60.3%
98.7%
97.9%
0.323%
58.4%
10th:
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%
2541
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%
Median:
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%
75t
0.237%
0.273%
0.266%
0.993%
0.509%
0.303%
0.312%
0.273%
0.313%
0.375%
0.375%
0.292%
90th:
95th
57.9%
60.3%
59.2%
97.9%
72.3% 66.7%
59.1%
59.7%
56.3%
58.4%
63.1%
98.4%
63.1%
59.5%
97.9%
0.421%
59.5%
98.3%
Note: Percentages are rounded to three significant figures.
a - Percentile refers to the risk percentile. For example, based on the values in the 75th percentile row for Segment 39, there is a 25 percent probability that selenium
concentrations in fish eggs/ovaries are high enough to cause negative reproductive effects in 83 percent of the exposed fish population inhabiting that segment of
Black Creek.
b - Segment 39 is the immediate receiving water for Morrow Generating Site. Segment 1 is farthest downstream from the immediate receiving water. The 39
segments comprise a total of 95 miles of Black Creek.
c - >0 to 5 percent risk; 5 to 35 percent risk; 35 to 65 percent risk;
F-12
-------
Appendix F— Overview of Ecological Risk Modeling Setup and Outputs
Table F-6. Risk of Selenium-Driven Reproductive Effects in Mallards Downstream from Morrow Generating Site Immediate
Receiving Water (Black Creek Case Study) - Baseline
Percentile a
10th:
25th:
Median:
75th:
Black Creek WASP Model Segment ID b'c
39
0.1%
0.872%
9.18%
37.3%
^^BEffl 71.1%
^^B^ 86.1%
10th:
25th:
Median:
75th:
90th:
95th:
10th:
25th:
Median:
75th:
90th:
95th:
26
0.1%
0.1%
1.14%
9.46%
33.2%
53.1%
13
0.1%
0.1%
0.698%
7.20%
25.5%
44.3%
38
0.1%
0.268%
3.46%
19.4%
49.5%
68.0%
25
0.1%
0.11%
1.66%
11.5%
35.3%
53.7%
12
0.1%
0.1%
0.698%
7.12%
26.1%
44.4%
37
0.1%
0.253%
3.21%
18.6%
47.4%
66.7%
24
0.1%
0.109%
1.53%
11.2%
35.6%
54.7%
11
0.1%
0.1%
0.698%
6.63%
25.4%
44.2%
36
O.l%
0.153%
2.33%
15.0%
41.4%
60.5%
23
0.1%
0.1%
1.51%
11.5%
36.0%
55.3%
10
O.l%
O.l%
1.09%
8.59%
31.0%
48.6%
35
O.l%
0.155%
2.27%
14.8%
40.5%
58.6%
22
0.1%
0.1%
1.57%
10.9%
34.5%
53.5%
9
O.l%
O.l%
0.986%
7.89%
26.8%
45.7%
34
0.1%
0.139%
1.90%
12.6%
38.3%
57.2%
21
0.1%
0.1%
1.12%
10.0%
32.9%
52.5%
8
0.1%
0.1%
0.750%
7.35%
26.5%
44.3%
33
0.1%
0.117%
1.92%
13.7%
40.5%
59.7%
20
0.1%
0.1%
1.12%
10.7%
33.5%
51.2%
7
0.1%
0.1%
0.750%
7.42%
26.0%
43.4%
32
0.1%
0.167%
2.33%
14.6%
41.6%
60.6%
19
0.1%
0.1%
1.55%
10.9%
33.9%
52.4%
6
0.1%
0.1%
0.75%
7.17%
25.9%
43.6%
31
O.l%
O.l%
0.463%
5.33%
22.1%
38.4%
18
0.1%
0.1%
1.12%
9.28%
31.1%
50.2%
5
O.l%
O.l%
0.789%
7.21%
26.9%
44.9%
30
O.l%
O.l%
0.451%
4.81%
21.2%
37.2%
17
0.1%
0.1%
0.911%
7.76%
27.2%
44.9%
4
O.l%
O.l%
0.750%
7.03%
26.1%
43.6%
29
0.1%
0.1%
0.298%
3.57%
17.6%
33.2%
16
0.1%
0.1%
0.698%
7.53%
27.4%
44.9%
3
0.1%
0.1%
0.898%
7.65%
27.0%
44.8%
28
0.1%
0.1%
1.17%
9.98%
33.6%
52.5%
15
0.1%
0.1%
0.698%
7.06%
26.6%
44.9%
2
0.1%
0.1%
0.1%
6.75%
27.2%
45.6%
27
O.l%
O.l%
1.10%
9.13%
32.0%
51.7%
14
0.1%
0.1%
0.698%
7.32%
26.5%
44.4%
1
O.l%
O.l%
0.900%
7.20%
27.3%
45.5%
Note: Percentages are rounded to three significant figures.
a - Percentile refers to the risk percentile. For example, based on the values in the 75th percentile row for Segment 39, there is a 25 percent probability that selenium
concentrations in mallard eggs are high enough to cause negative reproductive effects in 37.3 percent of the exposed mallard population inhabiting that segment of
Black Creek.
b - Segment 39 is the immediate receiving water for Morrow Generating Site. Segment 1 is farthest downstream from the immediate receiving water. The 39
segments comprise a total of 95 miles of Black Creek.
c - >0 to 5 percent risk; 5 to 35 percent risk; 35 to 65 percent risk;
F-13
-------
Appendix F— Overview of Ecological Risk Modeling Setup and Outputs
Ecological Risk Model Methodology Limitations and Assumptions
The limitations and assumptions of the ecological risk modeling methodology include the
following:
• Water Quality Inputs. The assumptions listed for the IRW model water quality module
in Appendix C apply to the dissolved selenium concentrations that support the national-
scale ecological risk model. The assumptions listed for the case study water quality
model in Appendix G apply to the Black Creek case study ecological risk model. As
discussed in Section 8 of the EA report, the case study models do incorporate available
data regarding background pollutant concentrations and pollutant loading contributions
from non-steam-electric point sources. For the Black Creek case study, however, EPA
did not identify sufficient STORET monitoring data to represent upstream pollutant
contributions, and did not identify any upstream non-steam-electric point sources with
loadings for the modeled pollutants. EPA therefore assumed pollutant concentrations
of zero within the water column at the upstream boundary of the modeling area. This
results in a potential underestimation of dissolved selenium concentrations (and the
associated risk of negative reproductive effects among fish and mallards) within the
Black Creek modeling area.
• Receptor Populations Evaluated, EPA assumed that the receptor species and receiving
water occur together (i.e., all receiving waters evaluated in the national-scale and case
study ecological risk models are habitat for fish and mallards even though that may not
always be the case). This results in a potential overestimation of the number of
immediate receiving waters whose elevated selenium concentrations are causing
negative reproductive impacts among exposed fish and mallards.
• Species Represented by Exposure-Response Functions. EPA used exposure-response
functions that are based on vetted functions from the literature for brown trout
(representative offish) and mallard (representative of avian). Brown trout are amongst
the most sensitive fish species to selenium [U.S. EPA, 2014f]. EPA selected the mallard
as the representative avian species, which may not reflect potential impacts to other
species that consume primarily fish rather than invertebrates, and that may show
differential sensitivity. The literature suggests that mallards are among the most
sensitive bird species to selenium [Chapman et a/., 2009]. Therefore, use of these
exposure-response functions results in an environmentally protective estimate of
reproductive risk among the fish and avian species found at any given waterbody.
• Multiple Pollutant Exposures. According to EPA's Steam Electric Power Generating
Point Source Category: Final Detailed Study Report [U.S. EPA, 2009b], receptors will
be exposed to multiple constituents simultaneously. However, the ecological risk
model examines the impact of only selenium to receptors and does not take into account
how the interaction of multiple pollutants impacts the receptors. For example, EPA did
not consider the impact of mercury on the uptake or toxicity of selenium. There is
evidence in the literature that these two compounds interact with each other in the
environment and may decrease the level of impact of selenium on a receptor;4
conversely, the interaction of other pollutants may increase the impact to a receptor. It
4 In a notable but unexplained exception to this general rule, Heinz and Hoffman (1998) found that selenium and
mercury interact to create additive or synergistic toxic effects in mallard embryos.
-------
Appendix F— Overview of Ecological Risk Modeling Setup and Outputs
is beyond the scope of this analysis to include the effects of multiple pollutant
interactions on receptors; however, the consideration of only selenium-driven impacts
in this analysis likely results in an underestimation of the overall negative reproductive
impacts among fish and mallards resulting from exposure to the variety of pollutants in
steam electric power plant wastewater discharges.
Composition of Fish and Mallard Diet. In this analysis, EPA assumed that mallard
diets consisted entirely of invertebrates, which potentially overestimates the dietary
intake of selenium (because invertebrates tend to bioaccumulate selenium to a higher
degree than submerged aquatic vegetation, another component of mallard diets). EPA
also assumed that the diets offish and mallards consisted entirely of aquatic organisms
that inhabit the modeled waterbodies. These assumptions result in an environmentally
protective estimation of dietary selenium uptake if fish and mallards also consume
organisms from other waterbodies that are not contaminated with selenium.
F-15
-------
Appendix G—Overview of Case Study Modeling Setup and Outputs
APPENDIX G
OVERVIEW OF CASE STUDY MODELING
SETUP AND OUTPUTS
This appendix presents additional information about the model development, input
variables, pollutant benchmarks, and methodology limitations/assumptions applicable to case
study modeling performed using EPA's Water Quality Analysis Simulation Program (WASP).
This appendix also presents additional information regarding the site-specific design, site-specific
input parameters (e.g., background pollutant concentrations, U.S. Geological Survey (USGS) time
series flow data, steam electric power plant pollutant loadings), and model settings (e.g., solids
constants and sediment transport parameters) for each of the WASP models. For additional
documentation regarding the selection of case study locations, development of the case study
models, and outputs produced by the WASP models, refer to the ERG memorandum, "Technical
Approach for Case Study Water Quality Modeling of Aquatic Systems in Support of the Final
Steam Electric Power Generating Industry Environmental Assessment" (DCN SE05570) (Case
Study Water Quality Modeling Memorandum).
CASE STUDY MODEL SETUP - ALL MODELS
This section of the appendix focuses on the development of the case study models,
including the limitations/assumptions, input parameters, and methodologies that are applicable to
all of the case study models.
Model Development & Input Variables
WASP Model Default Parameters. The Simple Toxicant module within WASP groups
reaches of the modeled receiving water (i.e., the individual COMTDs as defined in NHDPlus
Version 1) into segments based on the hydrologic characteristics. The WASP model calculates the
water column and benthic pollutant concentrations of the eight modeled pollutants using user-
defined parameters and default assumption values. Table G-l presents the WASP default
parameters and values that EPA used for all the case study models.
Benthic Sediment Depth. All of the case study models are designed with two layers of
segments representing the upper and lower benthic sediment layer, except for the Lake Sinclair
model where benthic layers are not simulated. For each model, the depth of the upper and lower
benthic sediment layers are 0.03m and 0.25m, respectively.
Pollutant Partition Coefficients & Densities. The Simple Toxicant module within WASP
applies pollutant-specific partition coefficients to estimate the degree to which pollutants in the
water column will adsorb to benthic sediments and suspended solids. EPA selected the suspended
sediment-water (Kdsw) partition coefficient for each of the eight modeled pollutants. Refer to Table
C-4 in Appendix C of the Environmental Assessment for the Effluent Limitations Guidelines and
Standards for the Steam Electric Power Generating Point Source Category (EP A-861 -R-15-006),
hereafter referred to as the "EA Report," for the suspended sediment water partition coefficients
used for each modeled pollutant. Additionally, the Simple Toxicant module requires the user to
input a density for each modeled pollutant. Table G-2 presents the density values EPA used for
each pollutant, based on published values from literature.
CM
-------
Appendix G—Overview of Case Study Modeling Setup and Outputs
Table G-l. Solids Constants and Sediment Transport Parameters - All Models
Input Parameter
Silts and Fines Density
Sand Density
Organic Solids Density
fcritcoh
vRCohMult
vRCohExp
vRNonCohEx
D50 silt
D50 sand
D50 POM
vDexp silt
vD exp san
vD exp POM
TAUcritcoh a
TAU cDl sib
TAU cD2 sib
TAU cDl sa
TAU cD2 sa
TAU cDl POb
TAU cD2 POb
Description
WASP default density for silts/fines
WASP default density for sand
WASP default density for organic solids
Critical cohesive sediment fraction; above which
sediment bed acts cohesively
Shear stress multiplier for cohesive resuspension
Shear stress exponent for cohesive resuspension
Shear stress exponent for noncohesive resuspension
Particle diameter for silt
Particle diameter for sand
Particle diameter for organic solids
Shear stress exponent for silt deposition
Shear stress exponent for sand deposition
Shear stress exponent for organic solids deposition
Critical shear stress for erosion of cohesive bed
Lower critical shear stress for silt; below which
deposition is maximum
Upper critical shear stress for sand; above which
deposition is zero
Lower critical shear stress for sand; below which
deposition is maximum
Upper critical shear stress for sand; above which
deposition is zero
Lower critical shear stress for organic solids; below
which deposition is maximum
Upper critical shear stress for organic solids; above
which deposition is zero
Value Used
2.650
2.650
1.350
0.200
2.500
2.500
1.500
0.025
0.250
0.012
1.000
1.000
1.000
3.500 or 5.000
3.500 or 5.000
7.000 or 10.000
4.000
5.000
3.500 or 5.000
7.000 or 10.000
Units
g/cm3
g/cm3
g/cm3
(fraction)
g/m2/sec
(unitless)
(unitless)
mm
mm
mm
(unitless)
(unitless)
(unitless)
N/m2
N/m2
N/m2
N/m2
N/m2
N/m2
N/m2
Acronyms: g/cm3 (grams per cubic centimeter); g/m2/sec (grams per square meter per second); mm (millimeter);
N/m2 (newton per square meter)
a - The value of this input parameter varies the critical sheer stress values for sediment transport. The value
specified for this parameter, which can be set between 0.5 and 8.0 N/m2, was determined as a result of calibration
performed for each case study model. EPA determined that for all WASP models except for the Mississippi River
site, a value of 3.5 N/m2 was reasonable and resulted in modeled solids output comparable to the actual monitoring
data results. For the Mississippi River WASP model, a value of 5.0 N/m2 was deemed more appropriate based on
model calibration.
b - WASP uses default values for these input parameters based on the value specified for 'TAUcritcoh.'
G-2
-------
Appendix G—Overview of Case Study Modeling Setup and Outputs
Table G-2. Pollutant Densities - All Models
Pollutant
Arsenic
Cadmium
Copper
Lead
Nickel
Selenium
Thallium
Zinc
Density
(g/cm3)
5.75
8.70
8.96
11.34
8.91
4.80
11.85
7.14
Organic Solids, Sands, and Silts/Fines. To define initial concentrations for the organic
solids, sands, and silts/fines parameters, EPA used total organic carbon (TOC) and total suspended
solids (TSS) concentrations derived from STORET monitoring data collected within the WASP
modeling area. EPA calculated the concentrations of organic solids (OS), sands, and silts/fines
using Equation G-l, Error! Reference source not found. Equation G-2, and Equation G-3 below.
G-3
-------
Appendix G—Overview of Case Study Modeling Setup and Outputs
EQUATION G-l
EQUATION G-2
EQUATION G-3
Where:
Cos = TOG x f0
Csand = (TSS - Cos) x fsand
Csf=(TSS-Cos)xfsf
Cos
Csand
Csf
TOC
TSS
fos
Isand
fsf
=
=
=
=
=
=
=
=
Initial concentration of organic solids
(mg/L)
Initial concentration of sands (mg/L)
Initial concentration of silts/fines (mg/L)
Total organic carbon (mg/L)
Total suspended solids (mg/L)
Fraction of total organic carbon that is
organic solids (unitless)
Fraction of total suspended solids
composed of sands
Fraction of total suspended solids
composed of silts/fines
Output from Equation G-l
Output from Equation G-2
Output from Equation G-3
Site-specific value derived from
STORET monitoring data
Site-specific value derived from
STORET monitoring data
Model assumption value of 0.5
Model assumption value of 0.05
Model assumption value of 0.95
Calibration of Sediment Transport Parameters. The concentrations of the modeled
pollutants are influenced by sediment transport; therefore, EPA calibrated specific sediment
transport parameters where possible. EPA calibrated the model outputs by manipulating one
sediment transport parameter, 'Critical Shear Stress for Erosion of Cohesive Bed' (defined as
'TAUcritcoh' in WASP), until the modeled TSS concentrations in the water column segments
(represented by the sum of organic matter, sands, and silts/fines) closely matched the available
TSS STORET monitoring data. The 'Critical Shear Stress for Erosion of Cohesive Bed' value used
for each case study model is presented in the case study model-specific sections of this appendix.1
Calibration of Initial Concentration of Sediment in Benthic Segments. In some cases, the
initial concentration of sediment in the benthic segments was adjusted during the calibration
process, as very large spikes in total solids concentration were sometimes observed during high
1 If EPA observed a significant difference between the modeled TSS concentrations and actual observed TSS
concentrations, the sediment transport calibration values were given further review; however, those differences,
when they occurred, were often attributable to the pollutant contributions flowing in from the model boundaries.
G-4
-------
Appendix G—Overview of Case Study Modeling Setup and Outputs
flow events near the beginning of the simulation period. These large spikes were an indication that
too much sediment was present in the modeled benthic segments at the start of the simulation,
indicating that calibration of the sediment concentration was necessary. Where monitored pollutant
data were available, the total concentration of pollutant was plotted alongside the actual observed
results from STORET monitoring data as another check in the calibration process. The initial
concentrations of the organic solids, sands, and silts/fines in the benthic sediment used for each
case study model are presented in the case study model-specific sections of this appendix.
Steam Electric Power Plant Pollutant Loadings. EPA calculated pollutant loadings from
the evaluated wastestreams as part of its engineering analysis (see Section 10 of the Technical
Development Document for the Effluent Limitations Guidelines and Standards for the Steam
Electric Power Generating Point Source Category (TDD) [EPA 821-R-15-007]). The baseline and
regulatory option pollutant loadings used for each case study are presented in the case study model-
specific sections of this appendix. The Case Study Water Quality Modeling Memorandum further
describes the methodology for calculating and incorporating steam electric power plant loadings
data into the WASP models.
Non-Steam Electric Loadings. EPA incorporated pollutant loadings and/or concentrations
data from Discharge Monitoring Reports (DMR), the Toxics Release Inventory (TRI), and EPA's
STORET monitoring database to represent pollutant contributions from non-steam-electric point
sources and nonpoint sources that may impact the case study water quality model. EPA
incorporated pollutant loadings data from DMR and TRI data for each of the eight pollutants to
account for the pollutant contributions within the modeling area. STORET monitoring data were
incorporated to account for contributions upstream of the modeling boundaries and for use in
calibration. For the modeled pollutants (not including TOC and TSS), EPA converted the average
concentration or annual load to a daily mass loading.2 Each case study model-specific section of
this appendix presents the non-steam electric pollutant loadings incorporated into the model. The
Case Study Water Quality Modeling Memorandum further describes the methodology for
collecting, assessing, and incorporating DMR and TRI pollutant loadings data into the WASP
models.
WASP Output Analysis Methodology
The WASP models generate output data for pollutant concentration (total, dissolved, and
sorbed) in each water column and benthic segment on a daily output time step. For the purposes
of assessing the baseline impacts and the improvements under the final rule, EPA used the baseline
and regulatory option WASP model outputs from the period after the steam electric power plant's
assumed compliance date.3 Using this period of water quality output ensures that the baseline and
regulatory option analyses are both based on the same underlying flow data, meaning that the
differences in modeled pollutant concentrations are solely attributable to the pollutant loading
reductions under the final rule.
2 EPA converted the average concentration calculated from the STORET monitoring data to a mass loading using
the average annual flow rate for the stream reach represented by the monitoring station(s).
3 For case studies with pollutant loadings from multiple steam electric power plants (Ohio River and Mississippi
River), EPA used the later of the two assumed compliance dates.
-------
Appendix G—Overview of Case Study Modeling Setup and Outputs
Water Quality Assessment. The WASP models generate daily pollutant concentrations in
the water column of all water column segments within the models. EPA quantified the water
quality impacts as the percent of days where the water column concentration, total or dissolved,
exceed the National Recommended Water Quality Criteria (NRWQC) or Maximum Contaminant
Level (MCL) benchmarks listed in Table C-7 in Appendix C. EPA also quantified the total river
miles exhibiting exceedances and the distance downstream of the steam electric power plant(s)
that showed any exceedances of these benchmarks at any point during the modeling period.
Wildlife Assessment. The WASP models generate daily pollutant concentrations in the
upper and lower benthic sediment segments within the models. EPA quantified the impact to
benthic organisms as the percent of days where the total sediment concentration in the upper
benthic segments exceed the Chemical Stressor Concentration Limit (CSCL) benchmarks for
sediment biota listed in Table D-l in Appendix D. EPA also quantified the total number of river
miles exhibiting exceedances and the distance downstream of the steam electric power plant(s)
that showed any exceedances of these CSCLs at any point during the modeling period.
EPA calculated the annual average pollutant concentrations in the water column (averaged
over the entire modeling period) of all water column segments. To determine negative impacts to
piscivorous wildlife (i.e., wildlife that consume fish) from the ingestion of contaminated fish, EPA
compared the calculated annual average water column concentrations to "threshold" water
concentrations that would result in exceedances of no effect hazard concentrations (NEHCs) for
minks and eagles developed by the USGS.4 Since minks are estimated to have a four-year life
expectancy, EPA completed this analysis using four-year rolling average water concentration
values. EPA quantified the total river miles with NEHC exceedances and how far downstream of
the plant these impacts are observed.
Refer to Appendix F regarding the methodology for performing ecological risk modeling
using water quality outputs from the Black Creek WASP model.
Human Health Assessment. EPA calculated the annual average pollutant concentrations in
the water column (averaged over the entire modeling period) of all water column segments. To
determine negative impacts to human receptors from the ingestion of contaminated fish, EPA
compared the calculated annual average concentrations to "threshold" water concentrations that
would trigger exceedances of either the non-cancer reference dose or the 1-in-a-million lifetime
excess cancer risk (LECR) benchmark for selected cohorts.5 EPA quantified the total river miles
with LECR benchmark exceedances and how far downstream of the plant these impacts are
observed.
Case Study Modeling Methodology Limitations and Assumptions
The case study modeling methodology shares the following limitations and assumptions
with the IRW model water quality module (see Appendix C for further discussion):
4 Refer to the memorandum "Downstream EA Modeling Methodology and Supporting Documentation" (DCN
SE04455) for the water column concentrations that result in exceedances of the NEHC benchmarks.
5 Refer to the memorandum "Downstream EA Modeling Methodology and Supporting Documentation" (DCN
SE04455) for the water column concentrations that result in exceedances of the non-cancer reference doses or LECR
benchmark for selected cohorts.
-------
Appendix G—Overview of Case Study Modeling Setup and Outputs
• The models are based on annual-average pollutant loadings and normalized flow rates
from the steam electric power plants. Unlike the water quality module, however, the
case study models do account for temporal variability in the receiving water flow
rates.
• The models do not take into consideration pollutant speciation within the receiving
stream.
• The models assume that pollutants dissolved or sorbed within the water column and
bottom sediments can be described by a single partition coefficient.
• The pollutant loadings included in the models are not representative of the total
pollutant loadings from steam electric power plants, as there are several waste
streams that are not included in the analysis (e.g., stormwater runoff, metal cleaning
wastes, coal pile runoff). Unlike the water quality module, however, the case study
models do take into account ambient background pollutant concentrations and
contributions from other point and nonpoint sources.
In addition to the above, the case study modeling methodology incorporates the following
limitations and assumptions:
• The models assume that pollutant contributions from background sources and other
point and nonpoint sources are constant over the entire modeling period. This
assumption reduces the variability in modeled pollutant concentrations over time and
results in a potential underestimation of periods with elevated pollutant
concentrations above benchmark levels (under both baseline conditions and the
regulatory options).
• The models incorporate DMR and TRI loadings data to represent other point source
dischargers. In DMR, facilities are required to report loadings only for the pollutants
that are listed in the facility's National Pollutant Discharge Elimination System
(NPDES) permit. This limitation results in a potential underestimation of the pollutant
loadings from point sources that discharge a modeled pollutant but are not required to
report wastewater monitoring data as part of their NPDES permit. TRI collects
facility-reported estimates of wastewater loadings data for both direct and indirect
dischargers. The TRI releases database does not include loadings from facilities with
total annual chemical releases of less than 500 Ibs and incorporates assumptions
regarding plants with annual releases of less than 1,000 Ibs. This limitation results in
a potential underestimation of pollutant loadings from smaller point sources. Other
limitations of the data collected in TRI include the following: small establishments
are not required to report, nor are facilities that do not meet reporting thresholds;
releases reported are based on estimates, not measurements; certain chemicals are
reported as a class, not as individual compounds; facilities are identified by NAICS
code, not point source category; and TRI requires facilities to report only certain
chemicals, therefore all pollutants discharged from a facility may not be captured.
The effect of these limitations on the case study model outputs is unknown.
• In cases where STORET monitoring data results are reported as below the
quantitation limit, EPA assumed the result was equal to one-half the low-level
analytical method detection limit for purposes of averaging the monitoring data
results. The effect of this assumption on the case study model outputs is unknown and
-------
Appendix G—Overview of Case Study Modeling Setup and Outputs
depends on whether actual background concentrations at the time and location of
monitoring were higher or lower than the assumed concentration.
The models assume that stream flow conditions throughout the modeling period can
be represented by selected ranges of historical stream flow data. The effect of this
assumption on the case study model outputs is unknown and depends on whether
actual stream flow rates are higher or lower than those used in the models.
For each steam electric power plant, EPA assumed a plant-specific date (derived from
the plant's permitting cycle) upon which the plant would achieve compliance with the
final rule. The selection of the assumed compliance date influences the timing of
when the modeled baseline impacts and improvements under the final rule would
occur, but does not affect the magnitude of these impacts and improvements.
By incorporating wildlife, human health, and ecological risk analyses, the models
incorporate all of the limitations and assumptions described for those analyses (see
Appendices D, E, and F).
G-8
-------
Appendix G—Overview of Case Study Modeling Setup and Outputs
CASE STUDY MODEL SETUPS AND OUTPUTS - BLACK CREEK, MS
This section presents information regarding the site-specific design, site-specific input
parameters (e.g., background pollutant concentrations, USGS time series flow data), model
settings (e.g., sediment transport parameters), and case study modeling results for the Black Creek
case study model.
Model Development & Input Variables
WASP Model Design. The Black Creek WASP model starts at the R.D. Morrow, Sr.
(Morrow) Generating Site's immediate receiving water (COMID 18104316), as defined by the
IRW model, and extends approximately 95 miles downstream to just upstream of where Big Black
Creek converges with Red Creek (COMID 18106998).
The Black Creek WASP model consists of 174 modeled segments. Segment IDs 1-39
represent the surface water of Black Creek with Segment ID 1 being the most downstream segment
and Segment ID 39 being the most upstream segment and immediate receiving water. The
remaining model segments represent tributary surface waters (Segment IDs 40-58), the upper
benthic layers (Segment IDs 59-116), and the lower benthic layers (Segment IDs 117-174). Figure
G-l illustrates the segmentation of the Black Creek WASP model.
The modeling period starts in 1982 (the year of the last revision to the steam electric ELGs)
and extends through 2036, covering a period of 55 years. Based on Morrow Generating Site's
NPDES permitting cycle, EPA assumes that the plant will achieve the limitations under the final
rule by 2019.
Incorporation of Flow Data. EPA used USGS stream flow data from one USGS stream
gage to represent inflow at the upstream end of the modeling area of the Black Creek WASP
model. EPA scaled the Black Creek stream gage data from Gage ID 02479130 to account for the
difference in drainage area between the actual gage location and the point where the contributing
flows enter the modeling area.
EPA used USGS stream flow data from one USGS stream gage to represent inflow from
Cypress Creek, a significant tributary to the Black Creek WASP modeling area. EPA scaled the
Cypress Creek stream gage data from Gage ID 0247155 to account for the difference in drainage
area between the actual gage location and the point where the contributing flows enter the
modeling area.
Figure G-l illustrates the two stream flow gages from which EPA incorporated USGS
stream flow data. Table G-3 presents additional information about the two stream gages and the
time period covered in the stream flow data record at each. Table G-4 presents how EPA
incorporated the stream flow data from these stream gages into the model to complete a full record
of flow data for the entire modeling period. For all other local inflows, EPA used the mean annual
flow defined in NUDPlus Version 1.
G-9
-------
Appendix G—Overview of Case Study Modeling Setup and Outputs
Figure G-l. Geographic Extent and Segmentation - Black Creek WASP Model
Model Input Variables. Table G-5 presents the pollutant loadings modeled from Morrow
Generating Plant at the evaluated wastestream level, both at baseline and after the plant achieves
the limitations under the final rule. EPA did not identify any point sources with 2011 DMR or TRI
loadings which would impact the Black Creek case study model.
Table G-6 presents the pollutant contributions flowing into the Black Creek WASP model
boundaries calculated using available STORET monitoring data.
Table G-7 presents the initial concentrations for the organic solids, sands, and silts/fines
values derived from STORET monitoring data collected. For tributaries where STORET
monitoring data were not available, EPA assumed the average boundary concentration from all
tributaries entering the modeling area. Based on the average of STORET data available within the
model, EPA calculated the initial concentrations of organic solids, sands, and silts/fines in the
water column segments were 3.43 mg/L, 0.78 mg/L, and 14.74 mg/L, respectively.
EPA calibrated the model outputs by manipulating the sediment transport parameters until
the modeled concentrations in the benthic segments closely matched the available sediment
concentration monitoring data derived from STORET. Table G-8 presents the sediment transport
parameters resulting from EPA's calibration effort. EPA assumed the initial concentrations of
organic solids, sands, and silts/fines in the benthic segments were equal to 10,000 mg/L each.
G-10
-------
Appendix G—Overview of Case Study Modeling Setup and Outputs
Model Results
Case study modeling of Black Creek revealed water quality benchmark exceedances in the
immediate receiving water and/or in downstream segments for arsenic, cadmium, selenium, and
thallium. Figure G-2, Figure G-3, and Figure G-4 illustrate the water concentration outputs for
these pollutants in the immediate receiving water before and after the assumed compliance date
for the final rule.6
Case study modeling of Black Creek revealed that average water column concentrations of
three pollutants (cadmium, selenium, and thallium) in the immediate receiving water and/or
downstream segments would trigger exceedances of wildlife and/or human health benchmarks.
Table G-9 and Table G-10 illustrate the average modeled pollutant concentration in each water
column segment downstream of Morrow Generating Site (including the immediate receiving
water) for baseline and following compliance with the final rule, respectively. Table G-ll and
Table G-12 present the total miles with average water column concentrations translating to
exceedances of these benchmarks for baseline and under the final rule, respectively.
Refer to Appendix F regarding the results of ecological risk modeling using water quality
outputs from the Black Creek WASP model.
6 To improve clarity, Figure G-2, Figure G-3, and Figure G-4 present the baseline water column concentrations
leading up to the assumed compliance date of Morrow Generating Station. All analyses of the WASP model outputs
were performed on the baseline output after the assumed compliance date.
(Ml
-------
Appendix G—Overview of Case Study Modeling Setup and Outputs
Table G-3. USGS Stream Gages with Flow Data Used in Black Creek WASP Model
Gage ID
2479130
2479155
USGS Gage
Location
Black Creek near
Brooklyn, MS
Cypress Creek near
Janice, MS
Stream Flow Record Period
Full Record from 10/01/1970
-04/14/2014
Full Record from 10/01/1966
-04/15/2014
Cumulative Drainage
Area Represented by
Gage (sq km)
929
138
Model Boundary
Black Creek
Cypress Creek
Cumulative Drainage
Area at Model
Boundary (sq km)
379
158
Scale Factor
0.408
1.143
Acronyms: USGS (U.S. Geological Survey).
Table G-4. Stream Flow Data Periods - Black Creek WASP Model
Modeling Period
Corresponding Stream Flow Data Period
Black Creek (Gage ID 2479130)
01/01/1982
10/01/2013
01/01/1998
10/01/2029
Cypress Creek (Gage ID
01/01/1982
10/01/2013
01/01/1998
10/01/2029
-09/30/2013
-12/31/2020
- 09/30/2029
-12/31/2036
01/01/1982
10/01/2005
01/01/1982
10/01/2005
2479155)
-09/30/2013
-12/31/2020
- 09/30/2029
-12/31/2036
01/01/1982
10/01/2005
01/01/1982
10/01/2005
-09/30/2013
-12/31/2012
-09/30/2013
-12/31/2012
-09/30/2013
-12/31/2012
-09/30/2013
-12/31/2012
G-12
-------
Appendix G—Overview of Case Study Modeling Setup and Outputs
Table G-5. Pollutant Loadings - Morrow Generating Site
Wastestream
Pollutant Loadings (g/day)
As
Cd
Cu
Ni
Pb
Se
Tl
Zn
Baseline a
FGD Wastewater
Fly Ash Transport Water
Bottom Ash Transport Water
Combustion Residual Leachate
Total
6.87
-
3.68
6.29
16.84
101.88
-
1.02
1.66
104.56
19.68
4.50
1.24
25.42
794.50
16.93
7.61
819.03
4.22
-
3.39
~
7.61
1,057.22
1.87
18.19
1,077.27
12.43
17.26
0.19
29.88
1,259.97
-
13.83
34.52
1,308.32
Final Rule b
FGD Wastewater
Fly Ash Transport Water
Bottom Ash Transport Water
Combustion Residual Leachate
Total
5.28
-
-
6.29
11.57
3.81
-
-
1.66
5.47
3.42
1.24
4.66
5.70
7.61
13.32
3.07
-
-
~
3.07
5.18
18.19
23.37
8.87
0.19
9.06
18.07
-
-
34.52
52.59
Acronyms: FGD (flue gas desulfurization).
a - The baseline pollutant loadings are modeled throughout the entire modeling period (from 01/01/1982 through 12/31/2036).
b - The final rule pollutant loadings are modeled only after the assumed compliance date (from 01/01/2019 through 12/31/2036).
G-13
-------
Appendix G—Overview of Case Study Modeling Setup and Outputs
Table G-6. Pollutant Contributions from STORET Monitoring Data - Black Creek WASP Model
Model Boundary
Clear Creek
Little Black Creek
Big Creek c
Big Creek c
Cypress Creek
Hickory Creek
Model Boundary
COMID
18104458
18104706
18104940
18104992
18108034
18106316
Station ID(s) (lat, long)
NLA06608-2010 (31.20,-89.30)
PA361 (31.09,-89.49)
PA043(31.07,-89.27)
PA240(31.07,-89.17)
PA360(31.14,-89.24)
OWW04440-HBN8 (31.02,-89.01)
PA056(31.03,-89.02)
112D33 (30.97,-88.97)
Parameter
TOC
TOC
TSS
TOC
TSS
TOC
TSS
TSS
TOC
TOC
Average Concentration
(Mg/L)a
4,420.00
7,400.00
4,642.86
10,000.00
7,000.00
10,333.33
4,666.67
10,000.00
18,000.00
3,000.00
Mass Loading
(g/day)b
-
-
-
-
-
-
-
-
-
-
Acronyms: TOC (Total Organic Carbon); TSS (Total Suspended Solids).
a -Where more than one monitoring station located on the same tributary system reported acceptable results for the same pollutant, EPA calculated and incorporated
the weighted average concentration across the monitoring stations (weighted by number of samples at each station).
b - For the modeled pollutants (not including TOC and TSS), EPA converted the average concentration to a mass loading using the average annual flow rate for the
stream reach represented by the monitoring station(s).
c - There are two distinct tributary systems that are identified as "Big Creek" in the National Hydrography Dataset Plus (NHDPlus Version 1) database.
G-14
-------
Appendix G—Overview of Case Study Modeling Setup and Outputs
Table G-7. Organic Solids, Sands, and Silts/Fines Inputs - Black Creek WASP Model
Model Boundary
Black Creek d
Clear Creek
Little Black Creek
Big Creek e
Big Creek e
Cypress Creek
Hickory Creek
All Other Inflows f
Model Boundary
COMID
18104316
18104458
18104706
18104940
18104992
18108034
18106316
N/A
Organic Solids Concentration
(mg/L)a
3.43
2.21
3.70
5.00
5.17
5.00
1.50
3.76
Sands Concentration
(mg/L)b
0.78
*
0.23
0.35
0.23
0.90
*
0.43
Silts/Fines Concentration
(mg/L)c
14.74
*
4.41
6.65
4.43
17.10
*
8.14
Acronyms: N/A (Not Applicable).
* - No TSS results available. The 'All Other Inflows' concentration was used in this scenario.
a - The organic solids concentration was calculated using Equation G-land the STORET monitoring data presented in Table G-6.
b - The sands concentration was calculated using Equation G-2 and the STORET monitoring data presented in Table G-6.
c - The silts/fines concentration was calculated using Equation G-3 and the STORET monitoring data presented in Table G-6.
d - The organic solids, sands, and silts/fines concentrations presented for this segment were used as the initial surface water conditions.
e - There are two distinct tributary systems that are identified as "Big Creek" in the National Hydrography Dataset Plus (NHDPlus Version 1) database.
f - For tributaries where boundary concentrations from STORET monitoring data were not available, EPA assumed the average boundary concentration from all
tributaries entering the modeling area.
G-15
-------
Appendix G—Overview of Case Study Modeling Setup and Outputs
Table G-8. Sediment Transport Parameters - Black Creek WASP Model
Input Parameter
TAUcritcoh
TAU cDl sia
TAU cD2 sia
TAU cDl POa
TAU cD2 POa
Value Used
3.5
3.5
7.0
3.5
7.0
Units
N/m2
N/m2
N/m2
N/m2
N/m2
Note: Table G-l presents additional solids constants and sediment transport parameters that are used in
each of the case study models.
a - This parameter is a WASP model default based on the value of the 'TAUcritcoh' parameter.
G-16
-------
Appendix G—Overview of Case Study Modeling Setup and Outputs
o>
•o
5
•5
.0
0.01
2014
2014
2015
2016
2017
2018
2019
2020
2021
2022
2023
2015
2016 2017 2018 2019 2020
Assumed Compliance Date of Morrow Generating Site
2021
2022
2023
CASE STUDY OUTPUT
^^ Baseline
^^~ Final Rule
IRW MODEL OUTPUT AQUATIC LIFE NRWQC BENCHMARKS HUMAN HEALTH NRWQC BENCHMARKS
Baseline Freshwater Acute Consumptionof OrganismsOnly
FinalRuie Freshwater Chronic Consumptionof Water & Organisms
DRINKING WATER BENCHMARKS
DrinkingWaterMCL
Figure G-2. Modeled Concentrations in Black Creek Water Column at Morrow Generating Site Immediate Receiving Water
(Total Cadmium, Dissolved Cadmium)
G-17
-------
Appendix G—Overview of Case Study Modeling Setup and Outputs
2014
2015
2016 2017 2018 2019 2020
Assumed Compliance Date of Morrow Generating Site
2021
2022
2023
CASE STUDY OUTPUT
^^ Baseline
^^~ Final Rule
IRW MODEL OUTPUT AQUATIC LIFE NRWQC BENCHMARKS HUMAN HEALTH NRWQC BENCHMARKS DRINKING WATER BENCHMARKS
Baseline Freshwater Acute Consumption of Organisms Only DrinkingWater MCL
FinalRule Freshwater Chronic Consumption of Water & Organisms
Figure G-3. Modeled Concentrations in Black Creek Water Column at Morrow Generating Site Immediate Receiving Water
(Total Arsenic, Total Thallium)
G-18
-------
Appendix G—Overview of Case Study Modeling Setup and Outputs
10
E
3
01
W
0.1
iv
!l-L
i
2014 2015 2016 2017 2018 2019 2020
Assumed Compliance Date of Morrow Generating Site
2021
2022
2023
CASE STUDY OUTPUT
^^~ Baseline
— Final Rule
IRW MODELOUTPUT AQUATIC LIFE NRWQC BENCHMARKS HUMAN HEALTH NRWQC BENCHMARKS DRINKING WATER BENCHMARKS
Baseline Freshwater Acute Consumption of Organisms Only OrinkingWater MCL
FinalRule Freshwater Chronic Consumption of Water & Organisms
Figure G-4. Modeled Concentrations in Black Creek Water Column at Morrow Generating Site Immediate Receiving Water
(Total Selenium)
G-19
-------
Appendix G—Overview of Case Study Modeling Setup and Outputs
Table G-9. Average Water Column Concentrations Downstream of Morrow Generating Site at Baseline
Segment Data
Segment ID
39
38
37
36
35
34
33
32
31
30
29
28
27
26
25
24
23
22
21
20
19
18
17
16
15
Segment Name
Black Creek/ IRW
Black Creek
Black Creek
Black Creek
Black Creek
Black Creek
Black Creek
Black Creek
Black Creek
Black Creek
Black Creek
Black Creek
Black Creek
Black Creek
Black Creek
Black Creek
Black Creek
Black Creek
Black Creek
Black Creek
Black Creek
Black Creek
Black Creek
Black Creek
Black Creek
Segment Length
(mi)
1.64
1.44
2.23
2.68
0.93
2.10
1.89
1.68
1.84
1.48
1.44
2.64
2.09
2.66
1.31
1.07
2.86
3.02
1.59
2.50
1.98
4.21
2.62
2.75
2.09
Distance
Downstream
(mi)
1.64
3.08
5.31
7.99
8.92
11.01
12.90
14.58
16.43
17.90
19.35
21.99
24.08
26.74
28.05
29.12
31.98
35.00
36.59
39.09
41.07
45.29
47.91
50.66
52.75
Average Total Water Column Concentration over Modeling Period (jig/L)a
As
0.0833
0.0543
0.0521
0.0445
0.0450
0.0420
0.0483
0.0476
0.0313
0.0282
0.0241
0.0396
0.0364
0.0348
0.0398
0.0413
0.0425
0.0425
0.0382
0.0396
0.0399
0.0349
0.0315
0.0299
0.0309
Cd
0.6045
0.4095
0.3172
0.2883
0.3114
0.2960
0.6251
0.6712
0.5851
0.3999
0.3275
0.9409
0.7642
0.6855
1.1003
1.2014
1.3070
1.3056
1.1168
1.2133
1.2327
1.1106
0.9820
0.9354
1.0301
Cu
0.1407
0.0942
0.0774
0.0693
0.0735
0.0696
0.1220
0.1307
0.1074
0.0783
0.0650
0.1816
0.1489
0.1344
0.2131
0.2311
0.2498
0.2499
0.2147
0.2319
0.2352
0.2114
0.1872
0.1780
0.1945
Pb
0.0498
0.0346
0.0226
0.0218
0.0249
0.0240
0.0625
0.0694
0.0619
0.0400
0.0324
0.1095
0.0866
0.0764
0.1383
0.1532
0.1688
0.1690
0.1431
0.1569
0.1596
0.1451
0.1276
0.1218
0.1357
Ni
4.2330
2.7890
2.5298
2.2009
2.2628
2.1255
3.0284
3.1224
2.3412
1.8857
1.5902
3.4735
3.0067
2.7946
3.8927
4.1371
4.3820
4.3861
3.8483
4.0771
4.1222
3.6660
3.2730
3.1087
3.3126
Se
5.6497
3.7362
3.3195
2.9067
3.0074
2.8291
4.2797
4.4510
3.4341
2.6870
2.2546
5.2132
4.4546
4.1124
5.9734
6.3833
6.7989
6.8023
5.9276
6.3200
6.3956
5.7048
5.0823
4.8313
5.1842
Tl
0.1510
0.0989
0.0926
0.0798
0.0812
0.0760
0.0988
0.1000
0.0695
0.0597
0.0509
0.0969
0.0868
0.0821
0.0951
0.0999
0.1045
0.1048
0.0931
0.0977
0.0986
0.0873
0.0783
0.0743
0.0782
Zn
7.7217
5.2410
3.9426
3.6359
3.9689
3.7863
6.3651
7.4057
6.2251
4.5225
3.7426
12.6119
9.9344
8.7650
14.4045
15.7678
17.2212
17.2252
14.6726
15.9712
16.2267
14.5811
12.8610
12.2591
13.5792
G-20
-------
Appendix G—Overview of Case Study Modeling Setup and Outputs
Table G-9. Average Water Column Concentrations Downstream of Morrow Generating Site at Baseline
Segment Data
Segment ID
14
13
12
11
10
9
8
7
6
5
4
o
J
2
1
Segment Name
Black Creek
Black Creek
Black Creek
Black Creek
Black Creek
Black Creek
Black Creek
Black Creek
Black Creek
Black Creek
Black Creek
Black Creek
Black Creek
Black Creek/ End
Segment Length
(mi)
4.55
2.35
2.14
2.01
4.00
1.80
3.50
3.02
3.33
3.16
3.36
1.90
3.66
3.85
Distance
Downstream
(mi)
57.30
59.65
61.79
63.80
67.80
69.61
73.10
76.12
79.45
82.61
85.97
87.87
91.54
95.38
Average Total Water Column Concentration over Modeling Period (jig/L)a
As
0.0305
0.0300
0.0194
0.0192
0.0269
0.0282
0.0265
0.0261
0.0261
0.0260
0.0263
0.0248
0.0241
0.0247
Cd
1.0067
0.9822
0.2514
0.2467
0.5034
0.6248
0.5620
0.5480
0.5551
0.5475
0.5658
0.4646
0.4279
0.4799
Cu
0.1903
0.1860
0.0569
0.0558
0.1033
0.1242
0.1125
0.1099
0.1109
0.1096
0.1129
0.0947
0.0877
0.0943
Pb
0.1325
0.1290
0.0208
0.0211
0.0565
0.0747
0.0662
0.0642
0.0650
0.0639
0.0666
0.0517
0.0462
0.0492
Ni
3.2498
3.1903
1.4524
1.4283
2.2124
2.4762
2.2782
2.2346
2.2472
2.2301
2.2768
2.0354
1.9406
2.0362
Se
5.0822
4.9820
2.0605
2.0254
3.2481
3.6902
3.3875
3.3201
3.3481
3.3199
3.3970
2.9817
2.8222
2.9758
Tl
0.0769
0.0756
0.0409
0.0402
0.0604
0.0655
0.0610
0.0600
0.0603
0.0599
0.0609
0.0557
0.0536
0.0556
Zn
13.2571
12.9326
3.0507
2.9991
6.4548
8.1467
7.3174
7.1365
7.2115
7.1144
7.3715
5.9687
5.4496
5.7478
Acronyms: IRW (Immediate receiving water).
a - Concentrations represent the average daily total pollutant concentration in the water column. The averaging period is the entire modeling period after the assumed
compliance date.
G-21
-------
Appendix G—Overview of Case Study Modeling Setup and Outputs
Table G-10. Average Water Column Concentrations Downstream of Morrow Generating Site Under Final Rule
Segment Data
Segment ID
39
38
37
36
35
34
33
32
31
30
29
28
27
26
25
24
23
22
21
20
19
18
17
Segment Name
Black Creek/ IRW
Black Creek
Black Creek
Black Creek
Black Creek
Black Creek
Black Creek
Black Creek
Black Creek
Black Creek
Black Creek
Black Creek
Black Creek
Black Creek
Black Creek
Black Creek
Black Creek
Black Creek
Black Creek
Black Creek
Black Creek
Black Creek
Black Creek
Segment Length
(mi)
1.64
1.44
2.23
2.68
0.93
2.10
1.89
1.68
1.84
1.48
1.44
2.64
2.09
2.66
1.31
1.07
2.86
3.02
1.59
2.50
1.98
4.21
2.62
Distance
Downstream
(mi)
1.64
3.08
5.31
7.99
8.92
11.01
12.90
14.58
16.43
17.90
19.35
21.99
24.08
26.74
28.05
29.12
31.98
35.00
36.59
39.09
41.07
45.29
47.91
Average Total Water Column Concentration over Modeling Period (jig/L)a
As
0.0575
0.0375
0.0360
0.0308
0.0311
0.0290
0.0335
0.0330
0.0217
0.0196
0.0167
0.0272
0.0249
0.0239
0.0279
0.0289
0.0298
0.0298
0.0268
0.0278
0.0280
0.0245
0.0221
Cd
0.0322
0.0218
0.0169
0.0153
0.0166
0.0158
0.0532
0.0545
0.0469
0.0335
0.0274
0.0639
0.0536
0.0491
0.1061
0.1205
0.1339
0.1349
0.1214
0.1276
0.1291
0.1176
0.1029
Cu
0.0261
0.0175
0.0144
0.0129
0.0136
0.0129
0.0273
0.0286
0.0233
0.0171
0.0142
0.0358
0.0300
0.0274
0.0442
0.0486
0.0530
0.0534
0.0469
0.0500
0.0507
0.0458
0.0404
Pb
0.0204
0.0141
0.0092
0.0089
0.0102
0.0098
0.0274
0.0301
0.0269
0.0174
0.0140
0.0460
0.0366
0.0323
0.0607
0.0675
0.0746
0.0748
0.0639
0.0697
0.0709
0.0643
0.0565
Ni
0.0702
0.0464
0.0419
0.0366
0.0378
0.0355
0.1016
0.0972
0.0778
0.0635
0.0534
0.0916
0.0823
0.0784
0.1996
0.2233
0.2437
0.2467
0.2291
0.2351
0.2370
0.2143
0.1881
Se
0.1250
0.0828
0.0734
0.0644
0.0669
0.0629
0.1766
0.1717
0.1392
0.1107
0.0927
0.1718
0.1518
0.1434
0.3454
0.3880
0.4253
0.4305
0.3976
0.4099
0.4137
0.3747
0.3285
Tl
0.0460
0.0301
0.0282
0.0243
0.0247
0.0232
0.0309
0.0312
0.0217
0.0187
0.0159
0.0301
0.0270
0.0255
0.0313
0.0330
0.0347
0.0348
0.0311
0.0325
0.0329
0.0291
0.0261
Zn
0.3167
0.2153
0.1615
0.1485
0.1629
0.1555
0.6420
0.6460
0.5689
0.3883
0.3143
0.7238
0.6070
0.5566
1.3552
1.5491
1.7278
1.7427
1.5701
1.6503
1.6710
1.5246
1.3331
G-22
-------
Appendix G—Overview of Case Study Modeling Setup and Outputs
Table G-10. Average Water Column Concentrations Downstream of Morrow Generating Site Under Final Rule
Segment Data
Segment ID
16
15
14
13
12
11
10
9
8
7
6
5
4
o
6
2
1
Segment Name
Black Creek
Black Creek
Black Creek
Black Creek
Black Creek
Black Creek
Black Creek
Black Creek
Black Creek
Black Creek
Black Creek
Black Creek
Black Creek
Black Creek
Black Creek
Black Creek/ End
Segment Length
(mi)
2.75
2.09
4.55
2.35
2.14
2.01
4.00
1.80
3.50
3.02
3.33
3.16
3.36
1.90
3.66
3.85
Distance
Downstream
(mi)
50.66
52.75
57.30
59.65
61.79
63.80
67.80
69.61
73.10
76.12
79.45
82.61
85.97
87.87
91.54
95.38
Average Total Water Column Concentration over Modeling Period (jig/L)a
As
0.0209
0.0217
0.0214
0.0210
0.0136
0.0134
0.0187
0.0197
0.0185
0.0182
0.0183
0.0182
0.0184
0.0173
0.0169
0.0173
Cd
0.0988
0.1092
0.1085
0.1068
0.0236
0.0232
0.0514
0.0652
0.0585
0.0571
0.0580
0.0568
0.0582
0.0510
0.0469
0.0497
Cu
0.0386
0.0422
0.0416
0.0408
0.0118
0.0116
0.0223
0.0271
0.0245
0.0240
0.0242
0.0239
0.0245
0.0211
0.0195
0.0208
Pb
0.0540
0.0602
0.0590
0.0574
0.0090
0.0095
0.0249
0.0330
0.0291
0.0282
0.0286
0.0281
0.0292
0.0228
0.0204
0.0217
Ni
0.1821
0.1951
0.1937
0.1921
0.0729
0.0716
0.1231
0.1485
0.1352
0.1322
0.1333
0.1314
0.1329
0.1250
0.1196
0.1233
Se
0.3154
0.3423
0.3421
0.3385
0.1176
0.1157
0.2037
0.2420
0.2222
0.2181
0.2201
0.2174
0.2204
0.2041
0.1936
0.1998
Tl
0.0248
0.0261
0.0257
0.0253
0.0134
0.0132
0.0200
0.0218
0.0203
0.0200
0.0201
0.0200
0.0203
0.0186
0.0186
0.0186
Zn
1.2784
1.4175
1.4070
1.3864
0.2752
0.2737
0.6375
0.8157
0.7296
0.7113
0.7222
0.7066
0.7233
0.6233
0.5720
0.5997
Acronyms: IRW (Immediate receiving water).
a - Concentrations represent the average daily total pollutant concentration in the water column. The averaging period is the entire modeling period after the assumed
compliance date.
G-23
-------
Appendix G—Overview of Case Study Modeling Setup and Outputs
Table G-ll. Total Miles of Black Creek with Wildlife And Human Health Impacts at Baseline
Wildlife & Human Health Impact Thresholds
WL - NEHC, T3 (mink)
WL - NEHC, T4 (eagle)
HH - Non-Cancer Adult Subsistence
HH - Non-Cancer Adult Recreational
HH - Non-Cancer Child Subsistence (1 to <2 y.o.) a
HH - Non-Cancer Child Subsistence (16 to <21 y.o) b
HH - Non-Cancer Child Recreational (1 to <2 y.o.) a
HH - Non-Cancer Child Recreational (16 to <21 y.o.)b
HH - Cancer Adult Subsistence
HH - Cancer Adult Recreational
HH - Cancer Child Subsistence (6 to <11 y.o.) a
HH - Cancer Child Subsistence (1 to <2 y.o.) b
HH - Cancer Child Recreational (6 to 1 1 y.o.) a
HH - Cancer Child Recreational (1 to <2 y.o.)b
Total Miles with Average Water Column Concentration Translating to Wildlife or Human Health
Benchmark Exceedances (mi)
As
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
Cd
0.00
0.00
0.00
0.00
37.64
0.00
0.00
0.00
No LECR
No LECR
No LECR
No LECR
No LECR
No LECR
Cu
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
No LECR
No LECR
No LECR
No LECR
No LECR
No LECR
Pb
0.00
0.00
NoRfD
NoRfD
NoRfD
NoRfD
NoRfD
NoRfD
No LECR
No LECR
No LECR
No LECR
No LECR
No LECR
Ni
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
No LECR
No LECR
No LECR
No LECR
No LECR
No LECR
Se
89.79
89.79
95.38
12.75
95.38
89.79
89.79
11.43
No LECR
No LECR
No LECR
No LECR
No LECR
No LECR
Tl
No NEHC
No NEHC
89.79
0.00
95.38
58.53
58.53
0.00
No LECR
No LECR
No LECR
No LECR
No LECR
No LECR
Zn
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
No LECR
No LECR
No LECR
No LECR
No LECR
No LECR
Acronyms: WL (Wildlife); HH (Human health); NEHC (No effect hazard concentration); Rfd (Reference dose); LECR (Lifetime excess cancer risk); y.o. (year old).
a - This row represents the most sensitive child fisher cohort.
b - This row represents the least sensitive child fisher cohort.
G-24
-------
Appendix G—Overview of Case Study Modeling Setup and Outputs
Table G-12. Total Miles of Black Creek with Wildlife And Human Health Impacts Under Final Rule
Wildlife and Human Health Impact Thresholds
WL - NEHC, T3 (mink)
WL - NEHC, T4 (eagle)
HH - Non-Cancer Adult Subsistence
HH - Non-Cancer Adult Recreational
HH - Non-Cancer Child Subsistence (1 to <2 y.o.) a
HH - Non-Cancer Child Subsistence (16 to <21 y.o.) b
HH - Non-Cancer Child Recreational (1 to <2 y.o.) a
HH - Non-Cancer Child Recreational (16 to <21 y.o.)b
HH - Cancer Adult Subsistence
HH - Cancer Adult Recreational
HH - Cancer Child Subsistence (6 to <11 y.o.) a
HH - Cancer Child Subsistence (1 to <2 y.o.) b
HH - Cancer Child Recreational (6 to 1 1 y.o.) a
HH - Cancer Child Recreational (1 to <2 y.o.)b
Total Miles with Average Water Column Concentration Translating to Wildlife or Human Health
Benchmark Exceedances (mi)
As
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
Cd
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
No LECR
No LECR
No LECR
No LECR
No LECR
No LECR
Cu
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
No LECR
No LECR
No LECR
No LECR
No LECR
No LECR
Pb
0.00
0.00
NoRfD
NoRfD
NoRfD
NoRfD
NoRfD
NoRfD
No LECR
No LECR
No LECR
No LECR
No LECR
No LECR
Ni
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
No LECR
No LECR
No LECR
No LECR
No LECR
No LECR
Se
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
No LECR
No LECR
No LECR
No LECR
No LECR
No LECR
Tl
No NEHC
No NEHC
0.00
0.00
58.53
0.00
0.00
0.00
No LECR
No LECR
No LECR
No LECR
No LECR
No LECR
Zn
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
No LECR
No LECR
No LECR
No LECR
No LECR
No LECR
Acronyms: WL (Wildlife); HH (Human health); NEHC (No effect hazard concentration); Rfd (Reference dose); LECR (Lifetime excess cancer risk); y.o. (year old).
a - This row represents the most sensitive child fisher cohort.
b - This row represents the least sensitive child fisher cohort.
G-25
-------
Appendix G—Overview of Case Study Modeling Setup and Outputs
CASE STUDY MODEL SETUPS AND OUTPUTS - ETOWAH RIVER, GA
This section presents information regarding the site-specific design, site-specific input
parameters (e.g., background pollutant concentrations, USGS time series flow data), model
settings (e.g., sediment transport parameters), and case study modeling results for the Etowah
River case study model.
Model Development & Input Variables
WASP Model Design. The Etowah River WASP model starts at Plant Bowen's immediate
receiving water (COMID 6499098), as defined by the IRW model, and extends approximately 35
miles downstream to just upstream of where the Etowah River converges with Silver Creek
(COMID 6500350).
The Etowah River WASP model consists of 96 modeled segments. Segment IDs 1-18
represent the surface water of the Etowah River with Segment ID 1 being the most downstream
segment and Segment ID 18 being the most upstream segment and immediate receiving water. The
remaining model segments represent tributary surface waters (Segment IDs 19-32), the upper
benthic layers (Segment IDs 33-64), and the lower benthic layers (Segment IDs 65-96). Figure
G-5 illustrates the segmentation of the Etowah River WASP model.
The modeling period starts in 1982 (the year of the last revision to the steam electric ELGs)
and extends through 2032, covering a period of 51 years. Based on Plant Bowen's NPDES
permitting cycle, EPA assumes that the plant will achieve the limitations under the final rule by
2021.
Incorporation of Flow Data. EPA used USGS stream flow data from one USGS stream
gage to represent inflow at the upstream end of the modeling area of the Etowah River WASP
model. EPA scaled the Etowah River stream gage data from Gage ID 02395000 to account for the
difference in drainage area between the actual gage location and the point where the contributing
flows enter the modeling area.
EPA used USGS stream flow data from one USGS stream gage to represent inflow from
Two Run Creek, a significant tributary to the Etowah River WASP modeling area. EPA scaled the
Two Run Creek stream gage data from Gage ID 02395120 to account for the difference in drainage
area between the actual gage location and the point where the contributing flows enter the
modeling area.
Figure G-5 illustrates the two stream flow gages from which EPA incorporated USGS
stream flow data. Table G-13 presents additional information about the two stream gages and the
time period covered in the stream flow data record at each. Table G-14 presents how EPA
incorporated the stream flow data from these stream gages into the model to complete a full record
of flow data for the entire modeling period. For all other local inflows, EPA used the mean annual
flow defined in NUDPlus Version 1.
G-26
-------
Appendix G—Overview of Case Study Modeling Setup and Outputs
t ,
(
'
Figure G-5. Geographic Extent and Segmentation - Etowah River WASP Model
Model Input Variables. Table G-15 presents the pollutant loadings modeled from Plant
Bowen at the evaluated wastestream level, both at baseline and after the plant achieves the
limitations under the final rule. EPA did not identify any point sources with 2011 DMR or TRI
loadings which would impact the Etowah River case study model and could not be accounted for
using STORET monitoring data.
Table G-16 presents the pollutant contributions flowing into the Etowah River WASP
model boundaries calculated using available STORET monitoring data.
Table G-17 presents the initial concentrations for the organic solids, sands, and silts/fines
values derived from STORET monitoring data collected. For tributaries where STORET
monitoring data were not available, EPA assumed the average boundary concentration from all
tributaries entering the modeling area. Based on the average of STORET data available within the
model, EPA calculated the initial concentrations of organic solids, sands, and silts/fines in the
water column segments were 2.56 mg/L, 0.90 mg/L, and 17.19 mg/L, respectively.
EPA calibrated the model outputs by manipulating the sediment transport parameters until
the modeled concentrations in the benthic segments closely matched the available sediment
concentration monitoring data derived from STORET. Table G-18 presents the sediment transport
parameters resulting from EPA's calibration effort. EPA assumed the initial concentrations of
organic solids, sands, and silts/fines in the benthic segments were equal to 500 mg/L each.
G-27
-------
Appendix G—Overview of Case Study Modeling Setup and Outputs
Model Results
Case study modeling of the Etowah River revealed water quality benchmark exceedances
in the immediate receiving water and/or in downstream segments for arsenic, cadmium, selenium,
and thallium.7 Figure G-6 and Figure G-7 illustrate the water concentration outputs for these
pollutants in the immediate receiving water before and after the assumed compliance date for the
final rule.8
Case study modeling of the Etowah River revealed that average water column
concentrations of three pollutants (arsenic, selenium, and thallium) in the immediate receiving
water and/or downstream segments would trigger exceedances of human health benchmarks. Table
G-19 and Table G-20 illustrate the average modeled pollutant concentration in each water column
segment downstream of Plant Bowen (including the immediate receiving water) for baseline and
following compliance with the final rule, respectively. Table G-21 and Table G-22 present the total
miles with average water column concentrations translating to exceedances of these benchmarks
for baseline and under the final rule, respectively.
7 Case study modeling also revealed isolated downstream exceedances of water quality benchmarks for lead.
8 To improve clarity, Figure G-6 and Figure G-7 present the baseline water column concentrations leading up to the
assumed compliance date of Plant Bowen. All analyses of the WASP model outputs were performed on the baseline
output after the assumed compliance date.
-------
Appendix G—Overview of Case Study Modeling Setup and Outputs
Table G-13. USGS Stream Gages with Flow Data Used in Etowah River WASP Model
Gage ID
02395000
02395120
USGS Gage
Location
Etowah River near
Kingston, GA
Two Run Creek near
Kingston, GA
Stream Flow Record Period
Partial Record from
07/18/2928-09/30/2013
(Missing Data between
10/24/1995 - 10/01/2008)
Full Record from 05/02/1980
-09/30/2013
Cumulative Drainage
Area Represented by
Gage (sq km)
4,239
85
Model Boundary
Etowah River
Two Run Creek
Cumulative Drainage
Area at Model
Boundary (sq km)
3,683
130
Scale Factor
0.869
1.52
Acronyms: USGS (U.S. Geological Survey).
Table G-14. Stream Flow Data Periods - Etowah River WASP Model
Modeling Period
Corresponding Stream Flow Data Period
Etowah River (Gage ID 02395000)
01/01/1982 -
10/24/1995 -
10/01/2008 -
10/01/2013 -
01/01/1994 -
10/24/2007 -
10/01/2020 -
10/01/2025-
Two Run Creek (Gage ID
01/01/1982 -
10/01/2013 -
01/01/1994 -
10/01/2025-
10/23/1995
09/30/2008
9/30/2013
12/31/2020
10/23/2007
09/30/2020
9/30/2025
12/31/2032
01/01/1982 -
10/24/1967 -
10/01/2008 -
10/01/2005-
01/01/1982 -
10/24/1967 -
10/01/2008 -
10/01/2005-
02395120)
09/30/2013
12/31/2020
09/30/2025
12/31/2032
01/01/1982 -
10/01/2005-
01/01/1982 -
10/01/2005-
10/23/1995
09/30/1980
9/30/2013
12/31/2012
10/23/1995
09/30/1980
9/30/2013
12/31/2012
09/30/2013
12/31/2012
09/30/2013
12/31/2012
G-29
-------
Appendix G—Overview of Case Study Modeling Setup and Outputs
Table G-15. Pollutant Loadings - Plant Bowen
Wastestream
Pollutant Loadings (g/day)
As
Cd
Cu
Ni
Pb
Se
Tl
Zn
Baseline a
FGD Wastewater c
Fly Ash Transport Water
Bottom Ash Transport Water
Combustion Residual Leachate
Total
27.56
-
13.79
~
41.35
408.74
-
3.81
~
412.55
78.96
16.86
~
95.82
3187.42
63.46
~
3,250.88
16.93
-
12.69
-
29.62
4241.43
6.99
~
4,248.42
49.87
64.69
~
114.56
5054.84
-
51.86
~
5,106.71
Final Rule"0
FGD Wastewater
Fly Ash Transport Water
Bottom Ash Transport Water
Combustion Residual Leachate
Total
21.18
-
-
~
21.18
15.28
-
-
~
15.28
13.71
~
13.71
22.89
~
22.89
12.31
-
-
~
12.31
20.77
~
20.77
35.60
~
35.60
72.51
-
-
~
72.51
Acronyms: FGD (flue gas desulfurization).
a - The baseline pollutant loadings are modeled throughout the entire modeling period (from 01/01/1982 through 12/3 1/2032).
b - The final rule pollutant loadings are modeled only after the assumed compliance date (from 01/01/2021 through 12/3 1/2032).
c - In estimating the historical pollutant loadings associated with Plant Bowen' s four FGD systems, EPA incorporated the pollutant loadings from FGD wastewater
as the systems were installed, between 2008 and 20 1 1 . EPA did not model any FGD wastewater pollutant loadings before the installation of Plant Bowen' s first FGD
system.
G-30
-------
Appendix G—Overview of Case Study Modeling Setup and Outputs
Table G-16. Pollutant Contributions from STORET Monitoring Data - Etowah River WASP Model
Model Boundary
Etowah River
Euharlee Creek
Two Run Creek
Connesena Creek
Toms Creek
Model Boundary
COMID
6499098
6497752
6497374
6497306
6499778
Station ID(s) (lat, long)
14310011 (34.15,-84.77)
1404130102 (34.15,-84.77)
1404130103 (34.15,-84.77)
1404130105 (34.12,-84.82)
1404140704 (34.13,-84.94)
1404140701 (34.12,-84.95)
14340201 (34.22,-84.97)
1404150501 (34.24,-84.97)
1404160201 (34.26,-84.99)
Parameter
As
Cd
Cu
Ni
Pb
Tl
Zn
TOC
TSS
Pb
TOC
TSS
As
Cd
Cu
Ni
Pb
Tl
Zn
TOC
TSS
TOC
TSS
TOC
Average Concentration
Oig/L)a
-
-
~
-
-
-
-
3,531.41
8,775.41
~
6,734.53
16,323.08
~
-
-
-
~
-
-
7,996.03
12,847.83
4,191.06
4,640.00
9,465.83
Mass Loading (g/day)b
9,993.11
1,279.89
5,103.32
2,909.40
2,631.57
5,004.55
7,666.84
-
-
1,480.69
-
-
693.96
86.75
346.98
173.49
138.79
346.98
693.96
~
-
-
-
-
G-31
-------
Appendix G—Overview of Case Study Modeling Setup and Outputs
Table G-16. Pollutant Contributions from STORET Monitoring Data - Etowah River WASP Model
Model Boundary
Spring Creek
Dykes Creek
Model Boundary
COMID
6499820
6499782
Station ID(s) (lat, long)
1404160301 (34.21,-85.07)
14340991 (34.21,-85.07)
1404160401 (34.25,-85.08)
1404160402 (34.26,-85.09)
Parameter
As
Cd
Cu
Ni
Pb
Tl
Zn
TOC
TSS
TOC
TSS
Average Concentration
Oig/L)a
-
-
-
-
~
-
-
8,526.71
14,434.78
2,350.53
3,661.11
Mass Loading (g/day)b
541.04
67.63
270.52
202.89
54.10
270.52
270.52
~
-
-
-
Acronyms: TOC (Total Organic Carbon); TSS (Total Suspended Solids).
a -Where more than one monitoring station located on the same tributary system reported acceptable results for the same pollutant, EPA calculated and incorporated
the weighted average concentration across the monitoring stations (weighted by number of samples at each station).
b - For the modeled pollutants (not including TOC and TSS), EPA converted the average concentration to a mass loading using the average annual flow rate for the
stream reach represented by the monitoring station(s).
G-32
-------
Appendix G—Overview of Case Study Modeling Setup and Outputs
Table G-17. Organic Solids, Sands, and Silts/Fines Inputs - Etowah River WASP Model
Model Boundary
Etowah River
Euharlee Creek
Two Run Creek
Connesena Creek
Toms Creek
Spring Creek
Dykes Creek
All Other Inflows'1
Model Boundary
COMID
6499098
6497752
6497374
6497306
6499778
6499820
6499782
N/A
Organic Solids Concentration
(mg/L)a
1.77
3.37
4.00
2.10
4.73
4.26
1.18
3.06
Sands Concentration
(mg/L)b
0.44
0.82
0.64
0.23
*
0.72
0.18
0.51
Silts/Fines Concentration
(mg/L)c
8.33
15.50
12.20
4.41
*
13.71
3.48
9.61
Acronyms: N/A (Not Applicable).
* - No TSS results available. The 'All Other Inflows' concentration was used in this scenario.
a - The organic solids concentration was calculated using Equation G-l and the STORET monitoring data presented in Table G-16.
b - The sands concentration was calculated using Equation G-2 and the STORET monitoring data presented in Table G-16.
c - The silts/fines concentration was calculated using Equation G-3 and the STORET monitoring data presented in Table G-16.
d - For tributaries where boundary concentrations from STORET monitoring data were not available, EPA assumed the average boundary concentration from all
tributaries entering the modeling area.
Table G-18. Sediment Transport Parameters - Etowah River WASP Model
Input Parameter
TAUcritcoh
TAU cDl sia
TAU cD2 sia
TAU cDl POa
TAU cD2 POa
Value Used
3.5
3.5
7.0
3.5
7.0
Units
N/m2
N/m2
N/m2
N/m2
N/m2
Note: Table G-l presents additional solids constants and sediment transport parameters that are used in each
of the case study models.
a - This parameter is a WASP model default based on the value of the 'TAUcritcoh' parameter.
G-3 3
-------
Appendix G—Overview of Case Study Modeling Setup and Outputs
10
0.01
2016
2017
2018
2019
2020
2021
2022
2023
2016
2017
2018
2019 2020 2021 2022
Assumed Compliance Date of Plant Bowen
2023
2024
2025
2024
2025
CASE STUDY OUTPUT
^^ Baseline
^^~ Final Rule
IRW MODEL OUTPUT AQUATIC LIFE NRWQC BENCHMARKS HUMAN HEALTH NRWQC BENCHMARKS DRINKING WATER BENCHMARKS
Baseline Freshwater Acute Consumption of OrganismsOnly DrinkingWater MCL
FinalRule Freshwater Chronic Consumption of Water & Organisms
Figure G-6. Modeled Concentrations in Etowah River Water Column at Plant Bowen Immediate Receiving Water
(Total Arsenic, Total Thallium)
G-34
-------
ra
3.
E
73
ro
0
o
8
5
0.01
0.001
Appendix G—Overview of Case Study Modeling Setup and Outputs
2016
2017
2018
2019
2020
2021
2022
2023
2024
2025
2016
2017
2018
2019 2020 2021 2022
Assumed Compliance Date of Plant Bowen
2023
2024
2025
CASE STUDY OUTPUT
^^ Baseline
^^~ Final Rule
IRW MODEL OUTPUT AQUATIC LIFE NRWQC BENCHMARKS HUMAN HEALTH NRWQC BENCHMARKS DRINKING WATER BENCHMARKS
Baseline Freshwater Acute Consumption of OrganismsOnly DrinkingWater MCL
FinalRule Freshwater Chronic Consumption of Water & Organisms
Figure G-7. Modeled Concentrations in Etowah River Water Column at Plant Bowen Immediate Receiving Water
(Dissolved Cadmium, Total Selenium)
G-35
-------
Appendix G—Overview of Case Study Modeling Setup and Outputs
Table G-19. Average Water Column Concentrations Downstream of Plant Bowen at Baseline
Segment Data
Segment
ID
18
17
16
15
14
13
12
11
10
9
8
7
6
5
4
3
2
1
Segment Name
Etowah River / IRW
Etowah River
Etowah River
Etowah River
Etowah River
Etowah River
Etowah River
Etowah River
Etowah River
Etowah River
Etowah River
Etowah River
Etowah River
Etowah River
Etowah River
Etowah River
Etowah River
Etowah River / End
Segment
Length
(mi)
3.61
1.48
1.42
0.58
1.20
3.69
1.09
1.29
0.37
2.95
2.70
0.90
1.26
2.82
2.19
2.48
1.89
2.81
Distance
Downstrea
m (mi)
3.61
5.09
6.51
7.10
8.29
11.99
13.08
14.36
14.74
17.69
20.39
21.29
22.55
25.38
27.57
30.05
31.94
34.75
Average Total Water Column Concentration over Modeling Period (jig/L)a
As
3.5521
2.5373
2.4625
2.4351
2.3959
2.4026
2.3771
2.3582
2.4742
2.4181
2.7308
2.6890
2.6458
2.6189
2.7324
2.6886
2.6892
2.6554
Cd
0.5095
0.3532
0.3077
0.2988
0.2871
0.3190
0.3115
0.2976
0.3076
0.3033
0.5530
0.5256
0.4943
0.4847
0.6494
0.6536
0.6629
0.6282
Cu
1.6421
1.1484
1.0395
1.0163
0.9850
1.0550
1.0354
1.0034
1.0550
1.0363
1.6659
1.5999
1.5239
1.4972
1.8852
1.8873
1.9009
1.8203
Pb
0.6667
0.5990
0.4300
0.4017
0.3660
0.4924
0.4681
0.4155
0.4226
0.4246
1.3016
1.1982
1.0827
1.0559
1.7094
1.7431
1.7746
1.6351
Ni
2.0928
1.4836
1.4091
1.3887
1.3601
1.3918
1.3739
1.3538
1.3632
1.3339
1.7191
1.6785
1.6334
1.6113
1.7807
1.7639
1.7696
1.7279
Se
1.4225
1.0056
0.9470
0.9320
0.9111
0.9399
0.9269
0.9108
0.8887
0.8701
1.1694
1.1380
1.1032
1.0873
1.2069
1.1981
1.2032
1.1704
Tl
1.7789
1.2664
1.2178
1.2025
1.1809
1.1944
1.1805
1.1678
1.2246
1.1972
1.4387
1.4116
1.3821
1.3658
1.4685
1.4495
1.4526
1.4270
Zn
3.7456
2.5855
2.2000
2.1272
2.0316
2.3093
2.2502
2.1304
2.2114
2.1861
4.2600
4.0264
3.7597
3.6830
5.0578
5.1046
5.1547
4.8579
Acronyms: IRW (Immediate receiving water).
a - Concentrations represent the average daily total pollutant concentration in the water column. The averaging period is the entire modeling period after the assumed
compliance date.
G-36
-------
Appendix G—Overview of Case Study Modeling Setup and Outputs
Table G-20. Average Water Column Concentrations Downstream of Plant Bowen Under Final Rule
Segment Data
Segment
ID
18
17
16
15
14
13
12
11
10
9
8
7
6
5
4
3
2
1
Segment Name
Etowah River/IRW
Etowah River
Etowah River
Etowah River
Etowah River
Etowah River
Etowah River
Etowah River
Etowah River
Etowah River
Etowah River
Etowah River
Etowah River
Etowah River
Etowah River
Etowah River
Etowah River
Etowah River/End
Segment
Length
(mi)
3.61
1.48
1.42
0.58
1.20
3.69
1.09
1.29
0.37
2.95
2.70
0.90
1.26
2.82
2.19
2.48
1.89
2.81
Distance
Downstream
(mi)
3.61
5.09
6.51
7.10
8.29
11.99
13.08
14.36
14.74
17.69
20.39
21.29
22.55
25.38
27.57
30.05
31.94
34.75
Average Total Water Column Concentration over Modeling Period (jig/L)a
As
3.5450
2.5322
2.4576
2.4302
2.3911
2.3978
2.3723
2.3534
2.3036
2.2517
2.5377
2.4979
2.4579
2.4331
2.4324
2.3930
2.3926
2.3624
Cd
0.3900
0.2704
0.2355
0.2287
0.2197
0.2442
0.2385
0.2278
0.2401
0.2368
0.4328
0.4113
0.3866
0.3791
0.5100
0.5134
0.5197
0.4923
Cu
1.6162
1.1302
1.0231
1.0003
0.9695
1.0384
1.0191
0.9876
1.0396
1.0212
1.6418
1.5768
1.5019
1.4753
1.8580
1.8602
1.8754
1.7955
Pb
0.6624
0.5960
0.4278
0.3998
0.3642
0.4900
0.4657
0.4134
0.4206
0.4227
1.2965
1.1935
1.0785
1.0525
1.7033
1.7368
1.7578
1.6212
Ni
0.9963
0.7063
0.6709
0.6611
0.6475
0.6627
0.6542
0.6446
0.6706
0.6560
0.8479
0.8280
0.8056
0.7947
0.8992
0.8908
0.8939
0.8728
Se
0.0072
0.0051
0.0048
0.0047
0.0046
0.0049
0.0049
0.0048
0.0046
0.0045
0.0062
0.0060
0.0059
0.0059
0.0072
0.0072
0.0073
0.0072
Tl
1.7515
1.2469
1.1990
1.1840
1.1626
1.1760
1.1624
1.1499
1.2071
1.1801
1.4182
1.3915
1.3624
1.3462
1.4481
1.4295
1.4325
1.4072
Zn
2.2700
1.5668
1.3333
1.2891
1.2312
1.4006
1.3636
1.2910
1.4017
1.3855
2.7158
2.5654
2.3926
2.3441
3.2327
3.2636
3.2965
3.1060
Acronyms: IRW (Immediate receiving water).
a - Concentrations represent the average daily total pollutant concentration in the water column. The averaging period is the entire modeling period after the assumed
compliance date.
G-37
-------
Appendix G—Overview of Case Study Modeling Setup and Outputs
Table G-21. Total Miles of Etowah River with Wildlife And Human Health Impacts at Baseline
Wildlife and Human Health Impact Thresholds
WL - NEHC, T3 (mink)
WL - NEHC, T4 (eagle)
HH - Non-Cancer Adult Subsistence
HH - Non-Cancer Adult Recreational
HH - Non-Cancer Child Subsistence (1 to <2 y.o.) a
HH - Non-Cancer Child Subsistence (16 to <21 y.o.) b
HH - Non-Cancer Child Recreational (1 to <2 y.o.) a
HH - Non-Cancer Child Recreational (16 to <21 y.o.)b
HH - Cancer Adult Subsistence
HH - Cancer Adult Recreational
HH - Cancer Child Subsistence (6 to <11 y.o.) a
HH - Cancer Child Subsistence (1 to <2 y.o.) b
HH - Cancer Child Recreational (6 to 1 1 y.o.) a
HH - Cancer Child Recreational (1 to <2 y.o.)b
Total Miles with Average Water Column Concentration Translating to Wildlife or Human Health
Benchmark Exceedances (mi)
As
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
3.61
0.00
0.00
0.00
0.00
0.00
Cd
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
No LECR
No LECR
No LECR
No LECR
No LECR
No LECR
Cu
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
No LECR
No LECR
No LECR
No LECR
No LECR
No LECR
Pb
0.00
0.00
NoRfD
NoRfD
NoRfD
NoRfD
NoRfD
NoRfD
No LECR
No LECR
No LECR
No LECR
No LECR
No LECR
Ni
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
No LECR
No LECR
No LECR
No LECR
No LECR
No LECR
Se
0.00
0.00
0.00
0.00
34.75
0.00
0.00
0.00
No LECR
No LECR
No LECR
No LECR
No LECR
No LECR
Tl
No NEHC
No NEHC
34.75
34.75
34.75
34.75
34.75
34.75
No LECR
No LECR
No LECR
No LECR
No LECR
No LECR
Zn
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
No LECR
No LECR
No LECR
No LECR
No LECR
No LECR
Acronyms: WL (Wildlife); HH (Human health); NEHC (No effect hazard concentration); Rfd (Reference dose); LECR (Lifetime excess cancer risk); y.o. (year old).
a - This row represents the most sensitive child fisher cohort.
b - This row represents the least sensitive child fisher cohort.
G-38
-------
Appendix G—Overview of Case Study Modeling Setup and Outputs
Table G-22. Total Miles of Etowah River with Wildlife And Human Health Impacts Under Final Rule
Wildlife and Human Health Impact Thresholds
WL - NEHC, T3 (mink)
WL - NEHC, T4 (eagle)
HH - Non-Cancer Adult Subsistence
HH - Non-Cancer Adult Recreational
HH - Non-Cancer Child Subsistence (1 to <2 y.o.) a
HH - Non-Cancer Child Subsistence (16 to <21 y.o.) b
HH - Non-Cancer Child Recreational (1 to <2 y.o.) a
HH - Non-Cancer Child Recreational (16 to <21 y.o.)b
HH - Cancer Adult Subsistence
HH - Cancer Adult Recreational
HH - Cancer Child Subsistence (6 to <11 y.o.) a
HH - Cancer Child Subsistence (1 to <2 y.o.) b
HH - Cancer Child Recreational (6 to 1 1 y.o.) a
HH - Cancer Child Recreational (1 to <2 y.o.)b
Total Miles with Average Water Column Concentration Translating to Wildlife or Human Health
Benchmark Exceedances (mi)
As
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
3.61
0.00
0.00
0.00
0.00
0.00
Cd
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
No LECR
No LECR
No LECR
No LECR
No LECR
No LECR
Cu
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
No LECR
No LECR
No LECR
No LECR
No LECR
No LECR
Pb
0.00
0.00
NoRfD
NoRfD
NoRfD
NoRfD
NoRfD
NoRfD
No LECR
No LECR
No LECR
No LECR
No LECR
No LECR
Ni
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
No LECR
No LECR
No LECR
No LECR
No LECR
No LECR
Se
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
No LECR
No LECR
No LECR
No LECR
No LECR
No LECR
Tl
No NEHC
No NEHC
34.75
34.75
34.75
34.75
34.75
34.75
No LECR
No LECR
No LECR
No LECR
No LECR
No LECR
Zn
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
No LECR
No LECR
No LECR
No LECR
No LECR
No LECR
Acronyms: WL (Wildlife); HH (Human health); NEHC (No effect hazard concentration); Rfd (Reference dose); LECR (Lifetime excess cancer risk); y.o. (year old).
a - This row represents the most sensitive child fisher cohort.
b - This row represents the least sensitive child fisher cohort.
G-39
-------
Appendix G—Overview of Case Study Modeling Setup and Outputs
CASE STUDY MODEL SETUPS AND OUTPUTS - LICK CREEK & WHITE RIVER, IN
This section presents information regarding the site-specific design, site-specific input
parameters (e.g., background pollutant concentrations, USGS time series flow data), model
settings (e.g., sediment transport parameters), and case study modeling results for the Lick Creek
and White River case study model.
Model Development & Input Variables
WASP Model Design. The Lick Creek and White River WASP model starts at the
convergence of the West Fork White River (COMID 18471042) and the East Fork White River
(COMTD 18446060). The model extends approximately 52 miles downstream to just up stream of
where the White River converges with the Wabash River (COMID 18471318). Petersburg
Generating Station's immediate receiving water, Lick Creek (COMID 18471122) is approximately
3 miles downstream of the confluence of the East Fork and West Fork of the White River.
The Lick Creek and White River WASP model consists of 78 modeled segments. Segment
IDs 1-19 represent the surface water of the White River with Segment ID 1 being the most
downstream segment, Segment ID 19 being the West Fork White River, and Segment 18 being the
East Fork White River. Lick Creek, the immediate receiving water, is represented as Segment 76
and intersects the White River between Segment 16 and Segment 17. The remaining model
segments represent tributary surface waters (Segment IDs 20-25), the upper benthic layers
(Segment IDs 26-50 & 77), and the lower benthic layers (Segment IDs 51-75 & 78). Figure G-8
illustrates the segmentation of the Etowah River WASP model.
The modeling period starts in 1986 (the year the last generating unit at Petersburg
Generating Station began operating) and extends through 2034, covering a period of 49 years.
Based on Petersburg Generating Station's NPDES permitting cycle, EPA assumes that the plant
will achieve the limitations under the final rule by 2019.
Incorporation of Flow Data. EPA used USGS stream flow data from one USGS stream
gage to represent inflow at the upstream end of the modeling area of the Lick Creek and White
River WASP model. EPA scaled the White River stream gage data from Gage ID 033740000 to
account for the difference in drainage area between the actual gage location and the point where
the contributing flows enter the modeling area at the East Fork White River and West Fork White
River modeling boundaries.
No USGS stream flow data were available on Lick Creek; therefore, EPA used stream flow
data from one USGS stream gage on nearby Kessinger Ditch as a surrogate stream to represent
inflow from Lick Creek. EPA scaled the Kessinger Ditch stream gage data from Gage ID 03360895
to produce a dataset with an average annual flow rate that closely approximates that of Lick Creek,
as defined by NHDPlus Version 1.
Figure G-8 illustrates the two stream flow gages from which EPA incorporated USGS
stream flow data. Table G-23 presents additional information about the two stream gages and the
time period covered in the stream flow data record at each. Table G-24 presents how EPA
incorporated the stream flow data from these stream gages into the model to complete a full record
of flow data for the entire modeling period. For all other local inflows, EPA used the mean annual
flow defined in NHDPlus Version 1.
(MO
-------
Appendix G—Overview of Case Study Modeling Setup and Outputs
Figure G-8. Geographic Extent and Segmentation - Lick Creek & White River WASP
Model
Model Input Variables. Table G-25 presents the pollutant loadings modeled from
Petersburg Generating Station at the evaluated wastestream level, both at baseline and after the
plant achieves the limitations under the final rule.
Table G-26 presents the pollutant loadings modeled from non-steam electric point sources
with 2011 DMR or TRI loadings which would impact the Lick Creek and White River case study
model.
Table G-27 presents the pollutant contributions flowing into the Lick Creek and White
River WASP model boundaries calculated using available STORET monitoring data.
Table G-28 presents the initial concentrations for the organic solids, sands, and silts/fines
values derived from STORET monitoring data collected. For tributaries where STORET
monitoring data were not available, EPA assumed the average boundary concentration from all
tributaries entering the modeling area. Based on the average of STORET data available within the
model, EPA calculated the initial concentrations of organic solids, sands, and silts/fines in the
water column segments were 1.99 mg/L, 4.70 mg/L, and 89.24 mg/L, respectively.
EPA calibrated the model outputs by manipulating the sediment transport parameters until
the modeled concentrations in the benthic segments closely matched the available sediment
concentration monitoring data derived from STORET. Table G-29 presents the sediment transport
parameters resulting from EPA's calibration effort. EPA assumed the initial concentrations of
organic solids, sands, and silts/fines in the benthic segments were equal to 500 mg/L each.
G-41
-------
Appendix G—Overview of Case Study Modeling Setup and Outputs
Model Results
Case study modeling of Lick Creek and the White River revealed water quality benchmark
exceedances in the immediate receiving water and/or in downstream segments for arsenic,
cadmium, copper, lead, selenium, and thallium. Figure G-9, Figure G-10, Figure G-l 1, and Figure
G-12 illustrate the water concentration outputs for these pollutants in the immediate receiving
water before and after the assumed compliance date for the final rule.9
Case study modeling of Lick Creek and the White River revealed that average water
column concentrations of four pollutants (arsenic, cadmium, selenium, and thallium) in the
immediate receiving water and/or downstream segments would trigger exceedances of wildlife
and/or human health benchmarks. Table G-30 and Table G-31 illustrate the average modeled
pollutant concentration in each water column segment downstream of Petersburg Generating
Station (including the immediate receiving water) for baseline and following compliance with the
final rule, respectively. Table G-32 and Table G-33 present the total miles with average water
column concentrations translating to exceedances of these benchmarks for baseline and under the
final rule, respectively.
9 To improve clarity, Figure G-9, Figure G-10, Figure G-l 1, and Figure G-12 present the baseline water column
concentrations leading up to the assumed compliance date of Petersburg Generating Station. All analyses of the
WASP model outputs were performed on the baseline output after the assumed compliance date.
(M2
-------
Appendix G—Overview of Case Study Modeling Setup and Outputs
Table G-23. USGS Stream Gages with Flow Data Used in Lick Creek and White River WASP Model
Gage ID
3374000
3374000
3360895
USGS Gage
Location
White River near
Petersburg, IN
White River near
Petersburg, IN
Kessinger Ditch
near Monroe City,
IN
Stream Flow Record Period
Full Record from 04/01/1928
-12/11/2013
Full Record from 04/01/1928
-12/11/2013
Full Record from 10/01/1992
- 9/30/1998
Cumulative Drainage
Area Represented by
Gage (sq km)
28,825
28,825
64.27 a
Model Boundary
West Fork White
River
East Fork White
River
Lick Creek
Cumulative Drainage
Area at Model
Boundary (sq km)
13,923
14,880
4.46 b
Scale Factor
0.483
0.516
0.069 c
Acronyms: USGS (U.S. Geological Survey).
a - This value represents the mean annual flow (in cfs), as defined by NHDPlus Version 1, at gage ID 3360895.
b - This value represents the mean annual flow (in cfs), as defined by NHDPlus Version 1, of the Lick Creek immediate receiving water.
c - This value represents the scale factor determined by the dividend of the mean annual flow of at gage ID 3360895 and the Lick Creek immediate receiving water.
G-43
-------
Appendix G—Overview of Case Study Modeling Setup and Outputs
Table G-24. Stream Flow Data Periods - Lick Creek and White River WASP
Model
Modeling Period
Corresponding Stream Flow Data Period
White River (Gage ID 3374000)
01/01/1986-12/11/2013
12/12/2013-12/31/2018
01/01/2002-12/11/2029
12/12/2029-12/31/2034
01/01/1986-12/11/2013
12/12/2005-12/31/2010
01/01/1986-12/11/2013
12/12/2005-12/31/2010
Kessinger Ditch (Gage ID 3360895)
01/01/1986 - 9/30/1986
10/01/1986 - 9/30/1992
10/01/1992 - 9/30/1998
10/01/1998 - 9/30/2004
10/01/2004 - 9/30/2010
10/01/2010 - 9/30/2016
10/01/2016-12/31/2018
01/01/2002 - 9/30/2002
10/01/2002 - 9/30/2008
10/01/2008 - 9/30/2014
10/01/2014 - 9/30/2020
10/01/2020 - 9/30/2026
10/01/2026 - 9/30/2032
10/01/2032-12/31/2034
01/01/1998-09/30/1998
10/01/1992-09/30/1998
10/01/1992-09/30/1998
10/01/1992-09/30/1998
10/01/1992-09/30/1998
10/01/1992-09/30/1998
10/01/1992-12/31/1994
01/01/1998-09/30/1998
10/01/1992-09/30/1998
10/01/1992-09/30/1998
10/01/1992-09/30/1998
10/01/1992-09/30/1998
10/01/1992-09/30/1998
10/01/1992-12/31/1994
G-44
-------
Appendix G—Overview of Case Study Modeling Setup and Outputs
Table G-25. Pollutant Loadings - Petersburg Generating Station
Wastestream
Pollutant Loadings (g/day)
As
Cd
Cu
Ni
Pb
Se
Tl
Zn
Baseline a
FGD Wastewater c
Fly Ash Transport Water
Bottom Ash Transport Water
Combustion Residual Leachate
Total
2.86
-
49.78
~
52.64
2.07
-
25.34
~
27.40
1.85
174.33
~
176.18
4.47
150.96
~
155.43
1.66
-
79.01
~
80.67
455.14
5.40
~
460.54
4.81
67.21
~
96.27
9.80
-
152.59
~
162.39
Final Rule"0
FGD Wastewater
Fly Ash Transport Water
Bottom Ash Transport Water
Combustion Residual Leachate
Total
2.86
-
-
~
2.86
2.07
-
-
~
2.07
1.85
~
1.85
3.09
~
3.09
1.66
-
-
~
1.66
2.81
~
2.81
4.81
~
4.81
9.80
-
-
~
9.80
Acronyms: FGD (flue gas desulfurization).
a - The baseline pollutant loadings are modeled throughout the entire modeling period (from 01/01/1986 through 12/31/2034).
b - The final rule pollutant loadings are modeled only after the assumed compliance date (from 01/01/2019 through 12/31/2034).
c - In estimating the historical pollutant loadings associated with Petersburg Generating Station's four FGD systems, EPA incorporated the pollutant loadings from
FGD wastewater as the systems were installed, between 1977 and 1996. The pollutant loadings associated with FGD systems installed before the start of the
modeling period (01/01/1986) are incorporated at the beginning or the model.
G-45
-------
Appendix G—Overview of Case Study Modeling Setup and Outputs
Table G-26. Pollutant Contributions from Non-Steam Electric Point Sources - Lick Creek and White River WASP Model
Facility Name
Pride Mine S-321 a
Model COMID
18471050
(White River)
City
Monroe City
Location (lat, long)
(38.54,-87.27)
Parameter
Cu
Ni
Zn
Average Daily Pollutant Loadings
(g/day)
9.23
9.23
9.23
a - EPA identified that this industrial facility is a direct discharger with 2011 DMR loadings.
Table G-27. Pollutant Contributions from STORET Monitoring Data - Lick Creek and White River WASP Model
Model Boundary
West Fork White
River
East Fork White
River
Model Boundary
COMID
18471042
18446060
Station ID(s) (lat, long)
10947 (38.56,-87.24)
2719 (38.56,-87.24)
WWL090-0028 (35.55,-87.24)
2619 (38.54,-87.22)
Parameter
As
Cu
Ni
Pb
Zn
TOC
TSS
As
Cd
Cu
Ni
Pb
Zn
TOC
TSS
Average Concentration
(Mg/L)a
-
-
~
-
-
5,104.00
104,000.00
-
-
~
-
-
-
3,475.43
62,087.96
Mass Loading
(g/day)b
19,498.53
74,468.84
130,549.28
37,390.75
228,842.01
-
-
17,881.15
506.03
35,794.47
43,219.91
20,429.79
134,155.14
-
-
G-46
-------
Appendix G—Overview of Case Study Modeling Setup and Outputs
Table G-27. Pollutant Contributions from STORET Monitoring Data - Lick Creek and White River WASP Model
Model Boundary
Conger Creek
Upper River
Deshee
Model Boundary
COMID
18471078
18471082
Station ID(s) (lat, long)
2511(38.52,-87.45)
2513 (38.51,-87.45)
WWL100-0002 (38.51,-87.44)
2512(38.52,-87.53)
Parameter
Cu
Pb
Zn
TOC
TSS
Pb
Zn
TOC
TSS
Average Concentration
(Mg/L)a
-
-
-
5,700.00
95,200.00
-
~
3,120.00
18,600.00
Mass Loading
(g/day)b
1,045.39
269.15
2,736.70
~
-
362.50
1,100.85
-
-
Acronyms: TOC (Total Organic Carbon); TSS (Total Suspended Solids).
a -Where more than one monitoring station located on the same tributary system reported acceptable results for the same pollutant, EPA calculated and incorporated
the weighted average concentration across the monitoring stations (weighted by number of samples at each station).
b - For the modeled pollutants (not including TOC and TSS), EPA converted the average concentration to a mass loading using the average annual flow rate for the
stream reach represented by the monitoring station(s).
G-47
-------
Appendix G—Overview of Case Study Modeling Setup and Outputs
Table G-28. Organic Solids, Sands, and Silts/Fines Inputs - Lick Creek and White River WASP Model
Model Boundary
West Fork White River
East Fork White River
Conger Creek
Upper River Deshee
All Other Inflows d
Model Boundary
COMID
18471042
18446060
18471078
18471082
N/A
Organic Solids Concentration
(mg/L)a
2.55
1.74
2.85
1.56
2.17
Sands Concentration
(mg/L)b
5.20
3.10
4.76
0.93
3.50
Silts/Fines Concentration
(mg/L)c
98.80
58.98
90.44
17.67
66.47
Acronyms: N/A (Not Applicable).
a - The organic solids concentration was calculated using Equation G-l and the STORET monitoring data presented in Table G-27.
b - The sands concentration was calculated using Equation G-2 and the STORET monitoring data presented in Table G-27.
c - The silts/fines concentration was calculated using Equation G-3 and the STORET monitoring data presented in Table G-27.
d - For tributaries where boundary concentrations from STORET monitoring data were not available, EPA assumed the average boundary concentration from all
tributaries entering the modeling area.
Table G-29. Sediment Transport Parameters - Lick Creek and White River WASP Model
Input Parameter
TAUcritcoh
TAU cDl sia
TAU cD2 sia
TAU cDl POa
TAU cD2 POa
Value Used
3.5
3.5
7.0
3.5
7.0
Units
N/m2
N/m2
N/m2
N/m2
N/m2
Note: Table G-l presents additional solids constants and sediment transport parameters that are used in each of the
case study models.
a - This parameter is a WASP model default based on the value of the 'TAUcritcoh' parameter.
G-48
-------
Appendix G—Overview of Case Study Modeling Setup and Outputs
o>
0.001
2014
2015
2016 2017 2018 2019 2020
Assumed Compliance Date of Petersburg Generating Station
2021
2022
2023
CASE STUDY OUTPUT
^^ Baseline
^^~ Final Rule
IRW MODEL OUTPUT AQUATIC LIFE NRWQC BENCHMARKS HUMAN HEALTH NRWQC BENCHMARKS
Baseline Freshwater Acute Consumptionof OrganismsOnly
FinalRule Freshwater Chronic Consumptionof Water & Organisms
DRINKING WATER BENCHMARKS
DrinkingWaterMCL
Figure G-9. Modeled Concentrations in Lick Creek Water Column at Petersburg Generating Station Immediate Receiving
Water (Total Cadmium, Dissolved Cadmium)
G-49
-------
Appendix G—Overview of Case Study Modeling Setup and Outputs
0)
0.
0.
o
o
•o
i
o
in
in
0.01
0.01
2014
2015
2016
2017
2018
2019
2020
2021
2022
2023
2014
2015
2016 2017 2018 2019 2020
Assumed Compliance Date of Petersburg Generating Station
2021
2022
2023
CASE STUDY OUTPUT
^^ Baseline
^^~ Final Rule
IRW MODEL OUTPUT AQUATIC LIFE NRWQC BENCHMARKS HUMAN HEALTH NRWQC BENCHMARKS DRINKING WATER BENCHMARKS
Baseline Freshwater Acute Consumption of OrganismsOnly DrinkingWater MCL
FinalRule Freshwater Chronic Consumption of Water & Organisms
Figure G-10. Modeled Concentrations in Lick Creek Water Column at Petersburg Generating Station Immediate Receiving
Water (Total Selenium, Dissolved Copper)
G-50
-------
Appendix G—Overview of Case Study Modeling Setup and Outputs
0.01
01
_i
E
^3
"S
3
o
2014
2015
2016
2017
2018
2019
2020
2021
2022
2014
2015
2016 2017 2018 2019 2020
Assumed Compliance Date of Petersburg Generating Station
2021
2022
2023
2023
CASE STUDY OUTPUT
^^ Baseline
^^~ Final Rule
IRW MODEL OUTPUT AQUATIC LIFE NRWQC BENCHMARKS HUMAN HEALTH NRWQC BENCHMARKS DRINKING WATER BENCHMARKS
Baseline Freshwater Acute Consumption of OrganismsOnly DrinkingWater MCL
FinalRule Freshwater Chronic Consumption of Water & Organisms
Figure G-ll. Modeled Concentrations in Lick Creek Water Column at Petersburg Generating Station Immediate Receiving
Water (Total Arsenic, Total Thallium)
G-51
-------
Appendix G—Overview of Case Study Modeling Setup and Outputs
100
_ 10
-I
ID
.0
0.1
0.01
2014
2015
2016 2017 2018 2019 2020
Assumed Compliance Date of Petersburg Generating Station
2021
2022
2023
CASE STUDY OUTPUT
^^~ Baseline
— Final Rule
IRW MODELOUTPUT AQUATIC LIFE NRWQC BENCHMARKS HUMAN HEALTH NRWQC BENCHMARKS DRINKING WATER BENCHMARKS
Baseline Freshwater Acute ConsumptionofOrganismsOnly Drinking Water MCL
FinalRule Freshwater Chronic Consumption of Water & Organisms
Figure G-12. Modeled Concentrations in Lick Creek Water Column at Petersburg Generating Station Immediate Receiving
Water (Total Lead)
G-52
-------
Appendix G—Overview of Case Study Modeling Setup and Outputs
Table G-30. Average Water Column Concentrations Downstream of Petersburg Generating Station at Baseline
Segment Data
Segment ID
76
16
15
14
13
12
11
10
9
8
7
6
5
4
3
2
1
Segment Name
Lick Creek /IRW
White River
White River
White River
White River
White River
White River
White River
White River
White River
White River
White River
White River
White River
White River
White River
White River / End
Segment Length
(mi)
1.82
2.53
3.64
3.39
3.39
3.39
4.43
1.78
3.88
3.22
2.97
2.97
2.97
2.97
2.97
2.97
1.17
Distance
Downstream
(mi)
1.82
4.35
7.99
11.38
14.77
18.17
22.59
24.37
28.26
31.48
34.45
37.42
40.39
43.36
46.33
49.30
50.47
Average Total Water Column Concentration over Modeling Period (jig/L)a
As
7.8099
2.1741
1.9842
1.8988
1.8498
1.8294
1.9038
1.8990
1.9106
2.8165
2.9378
2.6471
2.5550
2.4986
2.4569
2.4265
2.4061
Cd
1.4260
0.0169
0.0130
0.0108
0.0096
0.0090
0.0165
0.0197
0.0187
0.0657
0.0572
0.0521
0.0494
0.0474
0.0457
0.0443
0.0455
Cu
12.1623
3.9202
3.1428
2.7187
2.4878
2.3642
3.4944
4.0254
3.8692
11.7204
10.4477
9.4987
9.0307
8.6786
8.3900
8.1520
8.3071
Pb
2.2256
1.4878
1.0360
0.7764
0.6399
0.5643
1.4571
1.8743
1.7404
7.8841
6.5913
6.0681
5.7329
5.4787
5.2645
5.0746
5.2341
Ni
17.0962
7.9745
6.8940
6.3429
6.0350
5.8819
6.9169
7.2995
7.2015
15.3397
14.7231
13.2868
12.7097
12.3055
11.9847
11.7264
11.7954
Se
43.0318
0.0217
0.0176
0.0156
0.0147
0.0143
0.0290
0.0333
0.0325
0.0872
0.0813
0.0724
0.0687
0.0661
0.0640
0.0623
0.0646
Tl
12.0267
0.0053
0.0046
0.0042
0.0041
0.0040
0.0056
0.0060
0.0059
0.0119
0.0119
0.0105
0.0101
0.0098
0.0095
0.0093
0.0095
Zn
7.9902
11.0843
8.3919
6.8805
6.0692
5.6257
10.2181
12.3741
11.7421
44.4205
38.2040
34.9036
33.1043
31.7293
30.5571
29.5784
30.2942
Acronyms: IRW (Immediate receiving water).
a - Concentrations represent the average daily total pollutant concentration in the water column. The averaging period is the entire modeling period after the assumed
compliance date.
G-53
-------
Appendix G—Overview of Case Study Modeling Setup and Outputs
Table G-31. Average Water Column Concentrations Downstream of Petersburg Generating Station Under Final Rule
Segment Data
Segment ID
76
16
15
14
13
12
11
10
9
8
7
6
5
4
3
2
1
Segment Name
Lick Creek /IRW
White River
White River
White River
White River
White River
White River
White River
White River
White River
White River
White River
White River
White River
White River
White River
White River / End
Segment
Length
(mi)
1.82
2.53
3.64
3.39
3.39
3.39
4.43
1.78
3.88
3.22
2.97
2.97
2.97
2.97
2.97
2.97
1.17
Distance
Downstream
(mi)
1.82
4.35
7.99
11.38
14.77
18.17
22.59
24.37
28.26
31.48
34.45
37.42
40.39
43.36
46.33
49.30
50.47
Average Total Water Column Concentration over Modeling Period (jig/L)a
As
0.4306
2.1711
1.9815
1.8962
1.8473
1.8269
1.9010
1.8961
1.9077
2.8120
2.9331
2.6430
2.5510
2.4948
2.4531
2.4228
2.4024
Cd
0.1093
0.0159
0.0123
0.0103
0.0092
0.0086
0.0156
0.0185
0.0176
0.0608
0.0530
0.0484
0.0459
0.0440
0.0425
0.0411
0.0421
Cu
0.7301
3.9132
3.1377
2.7144
2.4838
2.3604
3.4867
4.0157
3.8599
11.6855
10.4162
9.4716
9.0055
8.6545
8.3679
8.1243
8.2818
Pb
0.0469
1.4855
1.0346
0.7754
0.6392
0.5636
1.4556
1.8722
1.7386
7.8735
6.5828
6.0604
5.7261
5.4724
5.2585
5.0689
5.2278
Ni
1.2803
7.9669
6.8877
6.3372
6.0295
5.8767
6.9088
7.2905
7.1928
15.3177
14.7021
13.2683
12.6922
12.2886
11.9682
11.7104
11.7788
Se
0.2617
0.0001
0.0001
0.0001
0.0001
0.0001
0.0054
0.0064
0.0063
0.0137
0.0129
0.0117
0.0110
0.0106
0.0102
0.0099
0.0100
Tl
0.5995
0.0003
0.0002
0.0002
0.0002
0.0002
0.0006
0.0007
0.0007
0.0013
0.0013
0.0011
0.0011
0.0011
0.0010
0.0010
0.0010
Zn
0.8859
11.0790
8.3882
6.8771
6.0663
5.6229
10.2123
12.3665
11.7352
44.3905
38.1791
34.8812
33.0831
31.7090
30.5324
29.5539
30.2683
Acronyms: IRW (Immediate receiving water).
a - Concentrations represent the average daily total pollutant concentration in the water column. The averaging period is the entire modeling period after the assumed
compliance date.
G-54
-------
Appendix G—Overview of Case Study Modeling Setup and Outputs
Table G-32. Total Miles of Lick Creek and White River with Wildlife And Human Health Impacts at Baseline
Wildlife and Human Health Impact Thresholds
WL - NEHC, T3 (mink)
WL - NEHC, T4 (eagle)
HH - Non-Cancer Adult Subsistence
HH - Non-Cancer Adult Recreational
HH - Non-Cancer Child Subsistence (1 to <2 y.o.) a
HH - Non-Cancer Child Subsistence (16 to <21 y.o.) b
HH - Non-Cancer Child Recreational (1 to <2 y.o.) a
HH - Non-Cancer Child Recreational (16 to <21 y.o.)b
HH - Cancer Adult Subsistence
HH - Cancer Adult Recreational
HH - Cancer Child Subsistence (6 to <11 y.o.) a
HH - Cancer Child Subsistence (1 to <2 y.o.) b
HH - Cancer Child Recreational (6 to 1 1 y.o.) a
HH - Cancer Child Recreational (1 to <2 y.o.)b
Total Miles with Average Water Column Concentration Translating to Wildlife or Human Health
Benchmark Exceedances (mi)
As
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
1.82
0.00
0.00
0.00
0.00
0.00
Cd
0.00
0.00
0.00
0.00
1.82
0.00
0.00
0.00
No LECR
No LECR
No LECR
No LECR
No LECR
No LECR
Cu
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
No LECR
No LECR
No LECR
No LECR
No LECR
No LECR
Pb
0.00
0.00
NoRfD
NoRfD
NoRfD
NoRfD
NoRfD
NoRfD
No LECR
No LECR
No LECR
No LECR
No LECR
No LECR
Ni
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
No LECR
No LECR
No LECR
No LECR
No LECR
No LECR
Se
1.82
1.82
1.82
1.82
1.82
1.82
1.82
1.82
No LECR
No LECR
No LECR
No LECR
No LECR
No LECR
Tl
No NEHC
No NEHC
1.82
1.82
1.82
1.82
1.82
1.82
No LECR
No LECR
No LECR
No LECR
No LECR
No LECR
Zn
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
No LECR
No LECR
No LECR
No LECR
No LECR
No LECR
Acronyms: WL (Wildlife); HH (Human health); NEHC (No effect hazard concentration); Rfd (Reference dose); LECR (Lifetime excess cancer risk); y.o. (year old).
a - This row represents the most sensitive child fisher cohort.
b - This row represents the least sensitive child fisher cohort.
G-55
-------
Appendix G—Overview of Case Study Modeling Setup and Outputs
Table G-33. Total Miles of Lick Creek and White River with Wildlife And Human Health Impacts Under Final Rule
Wildlife and Human Health Impact Thresholds
WL - NEHC, T3 (mink)
WL - NEHC, T4 (eagle)
HH - Non-Cancer Adult Subsistence
HH - Non-Cancer Adult Recreational
HH - Non-Cancer Child Subsistence (1 to <2 y.o.) a
HH - Non-Cancer Child Subsistence (16 to <21 y.o.) b
HH - Non-Cancer Child Recreational (1 to <2 y.o.) a
HH - Non-Cancer Child Recreational (16 to <21 y.o.)b
HH - Cancer Adult Subsistence
HH - Cancer Adult Recreational
HH - Cancer Child Subsistence (6 to <11 y.o.) a
HH - Cancer Child Subsistence (1 to <2 y.o.) b
HH - Cancer Child Recreational (6 to 1 1 y.o.) a
HH - Cancer Child Recreational (1 to <2 y.o.)b
Total Miles with Average Water Column Concentration Translating to Wildlife or Human Health
Benchmark Exceedances (mi)
As
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
Cd
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
No LECR
No LECR
No LECR
No LECR
No LECR
No LECR
Cu
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
No LECR
No LECR
No LECR
No LECR
No LECR
No LECR
Pb
0.00
0.00
NoRfD
NoRfD
NoRfD
NoRfD
NoRfD
NoRfD
No LECR
No LECR
No LECR
No LECR
No LECR
No LECR
Ni
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
No LECR
No LECR
No LECR
No LECR
No LECR
No LECR
Se
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
No LECR
No LECR
No LECR
No LECR
No LECR
No LECR
Tl
No NEHC
No NEHC
1.82
1.82
1.82
1.82
1.82
1.82
No LECR
No LECR
No LECR
No LECR
No LECR
No LECR
Zn
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
No LECR
No LECR
No LECR
No LECR
No LECR
No LECR
Acronyms: WL (Wildlife); HH (Human health); NEHC (No effect hazard concentration); Rfd (Reference dose); LECR (Lifetime excess cancer risk); y.o. (year old).
a - This row represents the most sensitive child fisher cohort.
b - This row represents the least sensitive child fisher cohort.
G-56
-------
Appendix G—Overview of Case Study Modeling Setup and Outputs
CASE STUDY MODEL SETUPS AND OUTPUTS - OHIO RIVER, PA/WV/OH
This section presents information regarding the site-specific design, site-specific input
parameters (e.g., background pollutant concentrations, USGS time series flow data), model
settings (e.g., sediment transport parameters), and case study modeling results for the Ohio River
case study model.
Model Development & Input Variables
WASP Model Design. The Ohio River WASP model starts approximately 12 miles
upstream of the first steam electric power plant immediate receiving water at COMTD 3821033.
There are two coal-fired plants modeled in the Ohio River WASP simulation. The upstream plant,
Bruce Mansfield plant, discharges to the Ohio River (COMID 3821113) from a large surface
impoundment named Little Blue Run. Approximately 13 miles downstream of this immediate
receiving water is the W.H. Sammis plant immediate receiving water (COMID 3821343). Ending
just upstream of the Cardinal Plant immediate receiving water, the entire Ohio River WASP model
is 49 miles long.
The Ohio River WASP model consists of 84 modeled segments. Segment IDs 1-17
represent the surface water of the Ohio River with Segment ID 1 being the most downstream
segment and Segment ID 17 being the most upstream segment. The immediate receiving waters
of the Bruce Mansfield plant and the W.H. Sammis plant are located at Segment ID 13 and 9,
respectively. The remaining model segments represent tributary surface waters (Segment IDs 18-
28), the upper benthic layers (Segment IDs 29-56), and the lower benthic layers (Segment IDs 57-
84). Figure G-13 illustrates the segmentation of the Ohio River WASP model.
The modeling period starts in 1982 (year of the last revision to the steam electric ELGs)
and extends through 2036, covering a period of 55 years. Based on their NPDES permitting cycles,
EPA assumes that Bruce Mansfield and W.H. Sammis plants will achieve the limitations under the
final rule by 2020 and 2021, respectively. EPA focused the assessment of the improvements under
the final rule on the period after the 2021 assumed compliance date for W.H. Sammis Plant.
Incorporation of Flow Data. EPA used USGS stream flow data from one USGS stream
gage to represent inflow at the upstream end of the modeling area of the Ohio River WASP model.
EPA scaled the Ohio River stream gage data from Gage ID 03086000 to account for the difference
in drainage area between the actual gage location and the point where the contributing flows enter
the modeling area.
EPA used USGS stream flow data from three USGS stream gages to represent inflow from
three tributaries to the Ohio River WASP modeling area, as described below:
• EPA scaled the Little Beaver Creek stream gage data from Gage ID 03109500 to
account for the difference in drainage area between the actual gage location and the
point where the contributing flows enter the modeling area.
• EPA scaled the Yellow Creek stream gage data from Gage ID 03110000 to account
for the difference in drainage area between the actual gage location and the point
where the contributing flows enter the modeling area.
G-57
-------
Appendix G—Overview of Case Study Modeling Setup and Outputs
• EPA scaled the Raccoon Creek stream gage data from Gage ID 03108000 to account
for the difference in drainage area between the actual gage location and the point
where the contributing flows enter the modeling area.
Figure G-13 illustrates the two stream flow gages from which EPA incorporated USGS
stream flow data. Table G-34 presents additional information about the four stream gages and the
time period covered in the stream flow data record at each. Table G-35 presents how EPA
incorporated the stream flow data from these stream gages into the model to complete a full
record of flow data for the entire modeling period. For all other local inflows, EPA used the
mean annual flow defined in NHDPlus Version 1.
Figure G-13. Geographic Extent and Segmentation - Ohio River WASP Model
Model Input Variables. Table G-36 presents the pollutant loadings modeled from Bruce
Mansfield plant at the evaluated wastestream level, both at baseline and after the plant achieves
the limitations under the final rule. Table G-37 presents the pollutant loadings modeled from W.H.
Sammis plant at the evaluated wastestream level, both at baseline and after the plant achieves the
limitations under the final rule.
Table G-38 presents the pollutant loadings modeled from non-steam electric point sources
with 2011 DMR or TRI loadings which would impact the Ohio River case study model.
Table G-39 presents the pollutant contributions flowing into the Ohio River WASP model
boundaries calculated using available STORET monitoring data.
G-58
-------
Appendix G—Overview of Case Study Modeling Setup and Outputs
Table G-40 presents the initial concentrations for the organic solids, sands, and silts/fines
values derived from STORET monitoring data collected. For tributaries where STORET
monitoring data were not available, EPA assumed the average boundary concentration from all
tributaries entering the modeling area. Based on the average of STORET data available within the
model, EPA calculated the initial concentrations of organic solids, sands, and silts/fines in the
water column segments were 1.36 mg/L, 0.57 mg/L, and 10.85 mg/L, respectively.
EPA calibrated the model outputs by manipulating the sediment transport parameters until
the modeled concentrations in the benthic segments closely matched the available sediment
concentration monitoring data derived from STORET. Table G-41 presents the sediment transport
parameters resulting from EPA's calibration effort. EPA assumed the initial concentrations of
organic solids, sands, and silts/fines in the benthic segments were equal to 100 mg/L each.
Model Results
Case study modeling of the Ohio River revealed water quality benchmark exceedances in
the W.H. Sammis plant immediate receiving water and/or in downstream segments for arsenic and
lead. Figure G-14 illustrates the water concentration outputs for these pollutants in the immediate
receiving water before and after the assumed compliance date for the final rule.10
Case study modeling of the Ohio River revealed that average water column concentrations
of thallium in the W.H. Sammis plant immediate receiving water and/or downstream segments
would trigger exceedances of human health benchmarks. Figure G-15 illustrates the water
concentration outputs for thallium in the W.H. Sammis plant immediate receiving water before
and after the assumed compliance date for the final rule. Table G-42 and Table G-43 illustrate the
average modeled pollutant concentration in each water column segment downstream of Bruce
Mansfield plant (including the Bruce Mansfield plant immediate receiving water) for baseline and
following compliance with the final rule, respectively. Table G-44 and Table G-45 present the total
miles with average water column concentrations translating to exceedances of these benchmarks
for baseline and under the final rule, respectively.
10 To improve clarity, Figure G-14 and Figure G-15 present the baseline water column concentrations leading up to
the assumed compliance date of Bruce Mansfield plant and W.H. Sammis plant. All analyses of the WASP model
outputs were performed on the baseline output after the assumed compliance date.
-------
Appendix G—Overview of Case Study Modeling Setup and Outputs
Table G-34. USGS Stream Gages with Flow Data Used in Ohio River WASP Model
Gage ID
3086000
3109500
3110000
3108000
USGS Gage
Location
Ohio River near
Sewickley, PA
Little Beaver Creek
Yellow Creek
Raccoon Creek
Stream Flow Record Period
Full Record from 01/01/1982
-09/30/2013
Full Record from 01/01/1982
-09/30/2013
Full Record from 01/01/1982
-09/30/2013
Full Record from 01/01/1982
-09/30/2013
Cumulative Drainage
Area Represented by
Gage (sq km)
50,475
1,286
382
464
Model Boundary
Ohio River
Little Beaver Creek
Yellow Creek
Raccoon Creek
Cumulative Drainage
Area at Model
Boundary (sq km)
58,947
1,345
612
477
Scale Factor
1.170
1.046
1.600
1.028
Acronyms: USGS (U.S. Geological Survey).
G-60
-------
Appendix G—Overview of Case Study Modeling Setup and Outputs
Table G-35. Stream Flow Data Periods - Ohio River WASP Model
Modeling Period
Corresponding Stream Flow Data Period
Ohio River (Gage ID 3086000)
01/01/1982-09/30/2013
10/01/2013-12/31/2020
01/01/1998-09/30/2029
10/01/2029-12/31/2036
01/01/1982
10/01/2005
01/01/1982
10/01/2005
-09/30/2013
-12/31/2012
-09/30/2013
-12/31/2012
Little Beaver Creek (Gage ID 3109500)
01/01/1982-09/30/2013
10/01/2013-12/31/2020
01/01/1998-09/30/2029
10/01/2029-12/31/2036
01/01/1982
10/01/2005
01/01/1982
10/01/2005
-09/30/2013
-12/31/2012
-09/30/2013
-12/31/2012
Yellow Creek (Gage ID 3110000)
01/01/1982-09/30/2013
10/01/2013-12/31/2020
01/01/1998-09/30/2029
10/01/2029-12/31/2036
01/01/1982
10/01/2005
01/01/1982
10/01/2005
-09/30/2013
-12/31/2012
-09/30/2013
-12/31/2012
Raccoon Creek (Gage ID 3108000)
01/01/1982-09/30/2013
10/01/2013-12/31/2020
01/01/1998-09/30/2029
10/01/2029-12/31/2036
01/01/1982
10/01/2005
01/01/1982
10/01/2005
-09/30/2013
-12/31/2012
-09/30/2013
-12/31/2012
G-61
-------
Appendix G—Overview of Case Study Modeling Setup and Outputs
Table G-36. Pollutant Loadings - Bruce Mansfield Plant
Wastestream
Pollutant Loadings (g/day)
As
Cd
Cu
Ni
Pb
Se
Tl
Zn
Baseline a
FGD Wastewater
Fly Ash Transport Water
Bottom Ash Transport Water
Combustion Residual Leachate
Total
29.09
-
50.21
~
79.30
431.42
-
13.86
~
445.28
83.34
61.38
~
144.72
3,364.27
231.01
-
3,595.28
17.87
-
46.21
~
64.08
4,476.75
25.46
~
4,502.21
52.63
235.52
~
288.15
5,335.30
-
188.79
~
5,524.09
Final Rule b
FGD Wastewater
Fly Ash Transport Water
Bottom Ash Transport Water
Combustion Residual Leachate
Total
22.35
-
-
~
22.35
16.13
-
-
~
16.13
14.48
-
14.48
24.16
~
24.16
12.99
-
-
~
12.99
21.93
~
21.93
37.58
~
37.58
76.53
-
-
~
76.53
Acronyms: FGD (flue gas desulfurization).
a - The baseline pollutant loadings are modeled throughout the entire modeling period (from 01/01/1982 through 12/31/2036).
b - The final rule pollutant loadings are modeled only after the assumed compliance date (from 01/01/2021 through 12/31/2036).
G-62
-------
Appendix G—Overview of Case Study Modeling Setup and Outputs
Table G-37. Pollutant Loadings - W.H. Sammis Plant
Wastestream
Pollutant Loadings (g/day)
As
Cd
Cu
Ni
Pb
Se
Tl
Zn
Baseline a
FGD Wastewater c
Fly Ash Transport Water
Bottom Ash Transport Water
Combustion Residual Leachate
Total
5.82
-
353.61
2.34
361.77
4.20
-
97.59
0.62
102.41
3.77
432.31
0.46
436.54
9.09
1,626.99
2.83
1,638.91
3.38
-
325.44
-
328.82
925.46
179.30
6.75
1,111.51
9.78
1,658.76
0.07
1,668.61
19.92
-
1,329.69
12.82
1,362.43
Final Rule b
FGD Wastewater
Fly Ash Transport Water
Bottom Ash Transport Water
Combustion Residual Leachate
Total
5.82
-
-
2.34
8.16
4.20
-
-
0.62
4.82
3.77
0.46
4.23
6.29
2.83
9.12
3.38
-
-
-
3.38
5.71
6.75
12.46
9.78
0.07
9.85
19.92
-
-
12.82
32.74
Acronyms: FGD (flue gas desulfurization).
a - The baseline pollutant loadings are modeled from 01/01/1982 through 12/31/2036.
b - The final rule pollutant loadings are modeled from 01/01/2021 through 12/31/2036.
c - In estimating the historical pollutant loadings associated with W.H. Sammis plant's three FGD systems, EPA incorporated the pollutant loadings from FGD
wastewater as the systems were installed, between March and May 2010. EPA did not model any FGD wastewater pollutant loadings before the installation of W.H.
Sammis plant's first FGD system.
G-63
-------
Appendix G—Overview of Case Study Modeling Setup and Outputs
Table G-38. Pollutant Contributions from Non-Steam Electric Point Sources -Ohio River WASP Model
Facility Name
City of Chester3
East Liverpool WWTP a
Town of Newell3
Wellsville STP a
Hancock County PSD a
Hancock County PSD WWTP a
City of New Cumberland a
Toronto WWTP a
Model COMID
3821165
(Ohio River)
3821167
(Ohio River)
3821149
(Ohio River)
3821273
(Ohio River)
3821301
(Ohio River)
3821355
(Ohio River)
3824147
(Ohio River)
3824175
(Ohio River)
City, State
Chester, WV
East Liverpool, OH
Newell, WV
Wellsville, OH
New Cumberland, WV
New Cumberland, WV
New Cumberland, WV
Toronto, OH
Location (lat, long)
(40.61,-80.57)
(40.62,-80.58)
(40.62,-80.61)
(40.60,-80.66)
(40.58,-80.66)
(40.51,-80.62)
(40.49,-80.60)
(40.50,-80.61)
Parameter
Cu
Pb
Zn
Cu
Zn
Cu
Pb
Zn
Cu
Pb
Zn
Cu
Pb
Cu
Pb
Zn
Cu
Pb
Zn
Zn
Average Daily Pollutant
Loadings (g/day)
32.63
24.40
87.47
13.96
375.00
3.92
1.49
6.80
48.94
2.64
134.64
4.69
1.57
6.05
0.87
45.16
6.85
0.48
18.25
150.75
G-64
-------
Appendix G—Overview of Case Study Modeling Setup and Outputs
Table G-38. Pollutant Contributions from Non-Steam Electric Point Sources -Ohio River WASP Model
Facility Name
City of Weirton a
City of Steubenville, Wastewater
Treatment Plant a
City of Follansbee a
Mingo Junction WTP a
CityofWellsburg3
CBS Beaver Groundwater
Remediation b
Model COMID
3824185
(Ohio River)
3824195
(Ohio River)
3824211
(Ohio River)
19453097
(Cross Creek)
19453103
(Ohio River)
3821033
(Two Mile Run)
City, State
Weirton, WV
Steubenville, OH
Follansbee, WV
Mingo Junction, OH
Wellsburg, WV
Beaver, PA
Location (lat, long)
(40.38,-80.61)
(40.36,-80.61)
(40.32,-80.60)
(40.31,-80.61)
(40.27,80.62)
(40.69,-80.31)
Parameter
As
Cd
Cu
Ni
Pb
Zn
Cu
Zn
As
Cd
Cu
Ni
Pb
Zn
Cu
As
Cd
Cu
Ni
Pb
Zn
Zn
Average Daily Pollutant
Loadings (g/day)
12.03
15.49
149.90
101.78
60.88
1,040.57
116.49
560.18
1.47
14.91
183.20
24.38
14.08
392.83
35.80
436.14
31.90
1,159.12
2.26
0.24
20.64
1,772.26
G-65
-------
Appendix G—Overview of Case Study Modeling Setup and Outputs
Table G-38. Pollutant Contributions from Non-Steam Electric Point Sources -Ohio River WASP Model
Facility Name
Horsehead Corp Monaca Smelter b
BASF Monaca Plant b
Lyondell Chem Beaver Valley b
Allegheny Technologies Midland
Plant b
Heritage-WTI Inc. b
Homer Laughlin China Co b
Model COMID
3821033
(Ohio River)
3821039
(Ohio River)
3821057
(Ohio River)
3821109
(Ohio River)
3821157
(Ohio River)
3821149
(Ohio River)
City, State
Monaca, PA
Monaca, PA
Monaca, PA
Midland, PA
East Liverpool, OH
Newell, WV
Location (lat, long)
(40.67,-80.34)
(40.66,-80.35)
(40.66,-80.36)
(40.64,-80.47)
(40.63,-80.55)
(40.62,-80.61)
Parameter
As
Cd
Cu
Pb
Se
Zn
Zn
Cu
Ni
Pb
Zn
Ni
As
Cd
Cu
Ni
Pb
Zn
Cd
Ni
Pb
Se
Zn
Average Daily Pollutant
Loadings (g/day)
16.34
66.43
190.61
63.10
12.39
1,259.91
257.26
72.06
64.22
36.23
83.68
441.29
0.84
0.57
5.42
2.69
5.00
48.85
1.13
7.71
0.99
1.45
1,101.25
G-66
-------
Appendix G—Overview of Case Study Modeling Setup and Outputs
Table G-38. Pollutant Contributions from Non-Steam Electric Point Sources -Ohio River WASP Model
Facility Name
Ergon West Virginia Inc b
Marsh Bellofram Corporation b
Mountaineer Park Incorporated b
Titanium Metals Corp b
Mittal Steel USA Weirton Inc b
Severstal Wheeling Inc -
Steubenville Plant b
Severstal Wheeling Inc -
Follansbee b
Model COMID
3821189
(Ohio River)
3821301
(Ohio River)
3821301
(Ohio River)
3824175
(Ohio River)
3824175
(Ohio River)
3824211
(Ohio River)
3824211
(Ohio River)
City, State
Newell, WV
Newell, WV
Chester, WV
Toronto, OH
Weirton, WV
Steubenville, OH
Follansbee, WV
Location (lat, long)
(40.61,-80.63)
(40.58,-80.65)
(40.57,-80.65)
(40.45,-80.61)
(40.43,-80.60)
(40.35,-80.61)
(40.35,-80.61)
Parameter
As
Cu
Zn
Cu
Ni
Pb
Zn
Cu
Pb
Zn
Cu
Zn
Cu
Ni
Pb
Se
Zn
Zn
As
Cd
Cu
Ni
Pb
Se
Zn
Average Daily Pollutant
Loadings (g/day)
304.33
7.99
13.55
2.24
0.30
0.15
1.03
2,669.15
2,358.45
36,768.62
0.09
1.63
385.18
63.48
182.75
252.54
1,935.80
1,042.47
250.85
0.01
201.14
0.06
0.33
3,364.80
460.56
G-67
-------
Appendix G—Overview of Case Study Modeling Setup and Outputs
Table G-38. Pollutant Contributions from Non-Steam Electric Point Sources -Ohio River WASP Model
Facility Name
RG Steel Wheeling LLC Beech
Bottom Plant b
Koppers Follansbee Tar Plant b
Wheeling-Nisshin b
Wheeling Pittsburgh Steel
Steubenville South Mingo b
NGC Industries LLC A Subsidiary
c
Whemco-Steel Castings Inc °
Mittal Steel USA Weirton Inc °
Model COMID
3824211
(Ohio River)
3824211
(Ohio River)
3824211
(Ohio River)
3824211
(Ohio River)
3821097
(Ohio River)
3821109
(Ohio River)
3824175
(Ohio River)
City, State
Beech Bottom, WV
Follansbee, WV
Follansbee, WV
Mingo Junction, OH
Shippingport, PA
Midland, PA
Weirton, WV
Location (lat, long)
(40.35,-80.61)
(40.34,-80.61)
(40.33,-80.60)
(40.32,-80.60)
(40.63,-80.42)
(40.63,-80.45)
(40.42,-80.60)
Parameter
As
Cu
Ni
Pb
Se
Zn
As
Se
Zn
Pb
Zn
Cu
Zn
Pb
Ni
Cu
Ni
Pb
Zn
Average Daily Pollutant
Loadings (g/day)
2.22
3.57
68.46
2.86
5.92
229.79
11.94
2.33
15.42
5.06
55.43
0.73
9.33
0.62
0.76
518.22
134.22
334.29
1,923.75
a - EPA identified that this publicly operated treatment works (POTW) facility is a direct discharger with 2011 DMR loadings.
b - EPA identified that this industrial facility is a direct discharger with 2011 DMR loadings.
c - EPA identified that this facility is a direct discharger with 2011 TRI loadings.
G-68
-------
Appendix G—Overview of Case Study Modeling Setup and Outputs
Table G-39. Pollutant Contributions from STORET Monitoring Data - Ohio River WASP Model
Model Boundary
Ohio River
Raccoon Creek
Buffalo Creek
Model Boundary
COMID
3821033
3821043
19453099
Station ID(s) (lat, long)
WQN0902(40.53,-80.19)
WQN0903 (40.63,-80.34)
O-092-0004 (40.26,-80.55)
O-092-0003 (40.25,-80.59)
O-092-0001 (40.24,-80.59)
O-092-0012 (40.23,-80.52)
O-092-0006 (40.20,-80.60)
O-092-0002 (40.20,-80.56)
O-092-0007 (40.19,-80.55)
O-092-0008 (40.16,-80.53)
Parameter
Cu
Ni
Pb
Zn
TOC
TSS
Cu
Ni
Pb
Zn
TOC
TSS
TSS
Average Concentration
Oig/L)a
-
-
„
-
2,426.67
21,434.78
-
-
~
2,232.63
16,893.62
10,333.33
Mass Loading
(g/day)b
175,758.13
126,664.12
79,371.40
1,247,520.00
-
-
376.43
1,663.34
525.00
13,504.33
-
-
-
Acronyms: TOC (Total Organic Carbon); TSS (Total Suspended Solids).
a -Where more than one monitoring station located on the same tributary system reported acceptable results for the same pollutant, EPA calculated and incorporated
the weighted average concentration across the monitoring stations (weighted by number of samples at each station).
b - For the modeled pollutants (not including TOC and TSS), EPA converted the average concentration to a mass loading using the average annual flow rate for the
stream reach represented by the monitoring station(s).
G-69
-------
Appendix G—Overview of Case Study Modeling Setup and Outputs
Table G-40. Organic Solids, Sands, and Silts/Fines Inputs - Ohio River WASP Model
Model Boundary
Ohio River
Raccoon Creek
Buffalo Creek
All Other Inflows d
Model Boundary
COMID
3821033
3821043
3821043
N/A
Organic Solids Concentration
(mg/L)a
1.21
1.12
*
1.16
Sands Concentration
(mg/L)b
1.07
0.84
0.52 e
0.81
Silts/Fines Concentration
(mg/L)c
20.36
16.05
9.82 e
15.41
Acronyms: N/A (Not Applicable).
* - No TOC results available. The 'All Other Inflows' concentration was used in this scenario.
a - The organic solids concentration was calculated using Equation G-l and the STORET monitoring data presented in Table G-39.
b - The sands concentration was calculated using Equation G-2 and the STORET monitoring data presented in Table G-39.
c - The silts/fines concentration was calculated using Equation G-3 and the STORET monitoring data presented in Table G-39.
d - For tributaries where boundary concentrations from STORET monitoring data were not available, EPA assumed the average boundary concentration from all
tributaries entering the modeling area.
e - These concentrations were calculated using the 'All Other Inflows' concentration.
Table G-41. Sediment Transport Parameters - Ohio River WASP Model
Input Parameter
TAUcritcoh
TAU cDl sia
TAU cD2 sia
TAU cDl POa
TAU cD2 POa
Value Used
3.5
3.5
7.0
3.5
7.0
Units
N/m2
N/m2
N/m2
N/m2
N/m2
Note: Table G-l presents additional solids constants and sediment transport parameters that are used in each
of the case study models.
a - This parameter is a WASP model default based on the value of the 'TAUcritcoh' parameter.
G-70
-------
Appendix G—Overview of Case Study Modeling Setup and Outputs
10
1 ••'
0.001
2015
2015
2016
2017
2018
2019
2020
2021
2022
2023
2024
2016 2017 2018 2019 2020 2021 2022 2023
Assumed Compliance Date of Bruce Mansfield Assumed Compliance Date of W.H. Sammis
2024
CASE STUDY OUTPUT
^^ Baseline
^^~ Final Rule
IRW MODEL OUTPUT AQUATIC LIFE NRWQC BENCHMARKS HUMAN HEALTH NRWQC BENCHMARKS
Baseline Freshwater Acute Consumption of OrganismsOnly
FinalRule Freshwater Chronic Consumption of Water & Organisms
DRINKING WATER BENCHMARKS
DrinkingWaterMCL
Figure G-14. Modeled Concentrations in Ohio River Water Column at W.H. Sammis Plant Immediate Receiving Water
(Total Arsenic, Total Lead)
G-71
-------
Appendix G—Overview of Case Study Modeling Setup and Outputs
2016
2016 2017 2018 2019 2020 2021 2022 2023
Assumed Compliance Date of Bruce Mansfield Assumed Compliance Date of W.H. Sammis
2024
CASE STUDY OUTPUT
^^~ Baseline
IRW MODELOUTPUT AQUATIC LIFE NRWQC BENCHMARKS HUMAN HEALTH NRWQC BENCHMARKS DRINKING WATER BENCHMARKS
--- Baseline ........... Freshwater Acute --- ConsumptionofOrganismsOnly Drinking Water MCL
--- FinalRule ............ Freshwater Chronic Consumption of Water & Organisms
Figure G-15. Modeled Concentrations in Ohio River Water Column at W.H. Sammis Plant Immediate Receiving Water
(Total Thallium)
G-72
-------
Appendix G—Overview of Case Study Modeling Setup and Outputs
Table G-42. Average Water Column Concentrations Downstream of Bruce Mansfield Plant at Baseline
Segment Data
Segment
ID
13
12
11
10
9
8
7
6
5
4
o
J
2
1
Segment Name
Ohio River / Mansfield
IRW
Ohio River
Ohio River
Ohio River
Ohio River / Sammis IRW
Ohio River
Ohio River
Ohio River
Ohio River
Ohio River
Ohio River
Ohio River
Ohio River / End
Segment
Length
(mi)
3.31
3.71
3.26
2.40
3.43
3.88
3.45
1.76
1.33
2.02
3.08
3.06
1.85
Distance
Downstream
(mi)
3.31
7.02
10.29
12.69
16.12
20.00
23.45
25.21
26.54
28.56
31.64
34.70
36.55
Average Total Water Column Concentration over Modeling Period (jig/L)a
As
0.0020
0.0083
0.0083
0.0093
0.0158
0.0165
0.0157
0.0155
0.0157
0.0156
0.0209
0.0202
0.0285
Cd
0.0082
0.0094
0.0111
0.0145
0.0165
0.0147
0.0127
0.0121
0.0120
0.0117
0.0182
0.0183
0.0195
Cu
2.6416
2.8186
3.1227
3.9276
3.8615
3.6063
3.2123
3.0856
3.0046
2.9513
3.1309
3.0956
3.1928
Pb
0.6965
0.8586
0.9913
1.4032
1.4652
1.1281
0.9050
0.8119
0.7527
0.7228
0.8725
0.9017
0.9529
Ni
2.4564
2.4577
2.5719
3.0175
2.8946
2.9481
2.7435
2.6984
2.6689
2.6419
2.7427
2.6655
2.6998
Se
0.0867
0.0883
0.0939
0.1118
0.1285
0.1275
0.1225
0.1200
0.1183
0.1171
0.1877
0.1836
0.1859
Tl
0.0058
0.0057
0.0059
0.0067
0.0394
0.0401
0.0380
0.0375
0.0372
0.0369
0.0371
0.0360
0.0362
Zn
15.5174
17.6384
20.0130
26.6303
26.8808
23.2738
19.8669
18.6383
17.8563
17.3905
19.0188
19.1172
19.7848
Acronyms: IRW (Immediate receiving water).
a - Concentrations represent the average daily total pollutant concentration in the water column. The averaging period is the entire modeling period after the assumed
compliance date.
G-73
-------
Appendix G—Overview of Case Study Modeling Setup and Outputs
Table G-43. Average Water Column Concentrations Downstream of Bruce Mansfield Plant Under Final Rule
Segment Data
Segment
ID
13
12
11
10
9
8
7
6
5
4
3
2
1
Segment Name
Ohio River / Mansfield
IRW
Ohio River
Ohio River
Ohio River
Ohio River / Sammis IRW
Ohio River
Ohio River
Ohio River
Ohio River
Ohio River
Ohio River
Ohio River
Ohio River / End
Segment
Length
(mi)
3.31
3.71
3.26
2.40
3.43
3.88
3.45
1.76
1.33
2.02
3.08
3.06
1.85
Distance
Downstream
(mi)
3.31
7.02
10.29
12.69
16.12
20.00
23.45
25.21
26.54
28.56
31.64
34.70
36.55
Average Total Water Column Concentration over Modeling Period (jig/L)a
As
0.0008
0.0072
0.0071
0.0080
0.0076
0.0080
0.0076
0.0075
0.0077
0.0077
0.0130
0.0125
0.0208
Cd
0.0011
0.0013
0.0016
0.0020
0.0021
0.0019
0.0016
0.0016
0.0018
0.0018
0.0075
0.0076
0.0084
Cu
2.6393
2.8161
3.1199
3.9241
3.8493
3.5949
3.2019
3.0756
2.9948
2.9417
3.1209
3.0855
3.1826
Pb
0.6959
0.8577
0.9900
1.4013
1.4561
1.1213
0.8995
0.8069
0.7481
0.7183
0.8674
0.8961
0.9468
Ni
2.3863
2.3875
2.4986
2.9313
2.7809
2.8337
2.6370
2.5936
2.5653
2.5393
2.6388
2.5642
2.5974
Se
0.0006
0.0007
0.0012
0.0014
0.0016
0.0016
0.0063
0.0062
0.0061
0.0061
0.0746
0.0731
0.0738
Tl
0.0008
0.0007
0.0008
0.0009
0.0010
0.0011
0.0010
0.0010
0.0010
0.0010
0.0010
0.0010
0.0010
Zn
15.4307
17.5355
19.8906
26.4824
26.7054
23.1208
19.7342
18.5127
17.7351
17.2733
18.8868
18.9822
19.6433
Acronyms: IRW (Immediate receiving water).
a - Concentrations represent the average daily total pollutant concentration in the water column. The averaging period is the entire modeling period after the assumed
compliance date.
G-74
-------
Appendix G—Overview of Case Study Modeling Setup and Outputs
Table G-44. Total Miles of Ohio River with Wildlife And Human Health Impacts at Baseline
Wildlife and Human Health Impact Thresholds
WL - NEHC, T3 (mink)
WL - NEHC, T4 (eagle)
HH - Non-Cancer Adult Subsistence
HH - Non-Cancer Adult Recreational
HH - Non-Cancer Child Subsistence (1 to <2 y.o.) a
HH - Non-Cancer Child Subsistence (16 to <21 y.o.) b
HH - Non-Cancer Child Recreational (1 to <2 y.o.) a
HH - Non-Cancer Child Recreational (16 to <21 y.o.)b
HH - Cancer Adult Subsistence
HH - Cancer Adult Recreational
HH - Cancer Child Subsistence (6 to <11 y.o.) a
HH - Cancer Child Subsistence (1 to <2 y.o.) b
HH - Cancer Child Recreational (6 to 1 1 y.o.) a
HH - Cancer Child Recreational (1 to <2 y.o.)b
Total Miles with Average Water Column Concentration Translating to Wildlife or Human Health
Benchmark Exceedances (mi)
As
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
Cd
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
No LECR
No LECR
No LECR
No LECR
No LECR
No LECR
Cu
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
No LECR
No LECR
No LECR
No LECR
No LECR
No LECR
Pb
0.00
0.00
NoRfD
NoRfD
NoRfD
NoRfD
NoRfD
NoRfD
No LECR
No LECR
No LECR
No LECR
No LECR
No LECR
Ni
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
No LECR
No LECR
No LECR
No LECR
No LECR
No LECR
Se
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
No LECR
No LECR
No LECR
No LECR
No LECR
No LECR
Tl
No NEHC
No NEHC
0.00
0.00
23.86
0.00
0.00
0.00
No LECR
No LECR
No LECR
No LECR
No LECR
No LECR
Zn
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
No LECR
No LECR
No LECR
No LECR
No LECR
No LECR
Acronyms: WL (Wildlife); HH (Human health); NEHC (No effect hazard concentration); Rfd (Reference dose); LECR (Lifetime excess cancer risk); y.o. (year old).
a - This row represents the most sensitive child fisher cohort.
b - This row represents the least sensitive child fisher cohort.
G-75
-------
Appendix G—Overview of Case Study Modeling Setup and Outputs
Table G-45. Total Miles of Ohio River with Wildlife And Human Health Impacts Under Final Rule
Wildlife and Human Health Impact Thresholds
WL - NEHC, T3 (mink)
WL - NEHC, T4 (eagle)
HH - Non-Cancer Adult Subsistence
HH - Non-Cancer Adult Recreational
HH - Non-Cancer Child Subsistence (1 to <2 y.o.) a
HH - Non-Cancer Child Subsistence (16 to <21 y.o.) b
HH - Non-Cancer Child Recreational (1 to <2 y.o.) a
HH - Non-Cancer Child Recreational (16 to <21 y.o.)b
HH - Cancer Adult Subsistence
HH - Cancer Adult Recreational
HH - Cancer Child Subsistence (6 to <11 y.o.) a
HH - Cancer Child Subsistence (1 to <2 y.o.) b
HH - Cancer Child Recreational (6 to 1 1 y.o.) a
HH - Cancer Child Recreational (1 to <2 y.o.)b
Total Miles with Average Water Column Concentration Translating to Wildlife or Human Health
Benchmark Exceedances (mi)
As
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
Cd
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
No LECR
No LECR
No LECR
No LECR
No LECR
No LECR
Cu
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
No LECR
No LECR
No LECR
No LECR
No LECR
No LECR
Pb
0.00
0.00
NoRfD
NoRfD
NoRfD
NoRfD
NoRfD
NoRfD
No LECR
No LECR
No LECR
No LECR
No LECR
No LECR
Ni
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
No LECR
No LECR
No LECR
No LECR
No LECR
No LECR
Se
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
No LECR
No LECR
No LECR
No LECR
No LECR
No LECR
Tl
No NEHC
No NEHC
0.00
0.00
0.00
0.00
0.00
0.00
No LECR
No LECR
No LECR
No LECR
No LECR
No LECR
Zn
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
No LECR
No LECR
No LECR
No LECR
No LECR
No LECR
Acronyms: WL (Wildlife); HH (Human health); NEHC (No effect hazard concentration); Rfd (Reference dose); LECR (Lifetime excess cancer risk); y.o. (year old).
a - This row represents the most sensitive child fisher cohort.
b - This row represents the least sensitive child fisher cohort.
G-76
-------
Appendix G—Overview of Case Study Modeling Setup and Outputs
CASE STUDY MODEL SETUPS AND OUTPUTS - MISSISSIPPI RIVER, MO/IL
This section presents information regarding the site-specific design, site-specific input
parameters (e.g., background pollutant concentrations, USGS time series flow data), model
settings (e.g., sediment transport parameters), and case study modeling results for the Mississippi
River case study model.
Model Development & Input Variables
WASP Model Design. The Mississippi River WASP model encompasses a 46-mile-long
reach of the Mississippi River, 23 miles of which is downstream of the Rush Island plant
immediate receiving water (COMID 3629181). The model has two start boundaries that are on the
Meramec River (COMID 5052703) and Mississippi River (COMID 3629071) shortly upstream of
their confluence. This model ends at the confluence of the Mississippi River and Kaskaskia River
(COMID 5089872).
The Mississippi River WASP model consists of 90 modeled segments. Segment IDs 1-16
represent the surface water of the Ohio River with Segment ID 1 being the most downstream
segment and Segment ID 16 being the most upstream segment. The Meramec River start boundary,
which is also the Meramec plant's immediate receiving water (COMID 5052703), is represented
by Segment ID 17. The immediate receiving water of the Rush Island is located at Segment ID 9.
The remaining model segments represent tributary surface waters (Segment IDs 18-30), the upper
benthic layers (Segment IDs 31-60), and the lower benthic layers (Segment IDs 61-90). Figure
G-16 illustrates the segmentation of the Mississippi River WASP model.
The modeling period starts in 1982 (year of the last revision to the steam electric ELGs)
and extends through 2036, covering a period of 55 years. Based on their NPDES permitting cycles,
EPA assumes that the Meramec and Rush Island plants will achieve the limitations under the final
rule by 2019 and 2023, respectively. For the Rush Island plant's immediate receiving water and
downstream reaches, EPA focused the assessment of the baseline impacts and improvements under
the final rule on the period after the 2023 assumed compliance date for the Rush Island plant.
Incorporation of Flow Data. EPA used USGS stream flow data from one USGS stream
gage to represent inflow at the upstream end of the modeling area of the Mississippi River WASP
model. EPA scaled the Mississippi River stream gage data from Gage ID 07010000 to account for
the difference in drainage area between the actual gage location and the point where the
contributing flows enter the modeling area.
EPA used USGS stream flow data from one other USGS stream gages to represent inflow
from the Meramec River, a tributary to the Mississippi River WASP modeling area. EPA scaled
the Meramec River stream gage data from Gage ID 07019000 to account for the difference in
drainage area between the actual gage location and the point where the contributing flows enter
the modeling area.
Figure G-16 illustrates the two stream flow gages from which EPA incorporated USGS
stream flow data. Table G-46 presents additional information about the four stream gages and the
time period covered in the stream flow data record at each. Table G-47 presents how EPA
incorporated the stream flow data from these stream gages into the model to complete a full record
-------
Appendix G—Overview of Case Study Modeling Setup and Outputs
of flow data for the entire modeling period. For all other local inflows, EPA used the mean annual
flow defined in NHDPlus Version 1.
Figure G-16. Geographic Extent and Segmentation - Mississippi River WASP Model
Model Input Variables. Table G-48 presents the pollutant loadings modeled from Bruce
Meramec plant at the evaluated wastestream level, both at baseline and after the plant achieves the
limitations under the final rule. Table G-49 presents the pollutant loadings modeled from Rush
Island plant at the evaluated wastestream level, both at baseline and after the plant achieves the
limitations under the final rule.
Table G-50 presents the pollutant loadings modeled from non-steam electric point sources
with 2011 DMR or TRI loadings which would impact the Mississippi River case study model.
Table G-51 presents the pollutant contributions flowing into the Mississippi River WASP
model boundaries calculated using available STORET monitoring data.
Table G-52 presents the initial concentrations for the organic solids, sands, and silts/fines
values derived from STORET monitoring data collected. For tributaries where STORET
monitoring data were not available, EPA assumed the average boundary concentration from all
tributaries entering the modeling area. Based on the average of STORET data available within the
model, EPA calculated the initial concentrations of organic solids, sands, and silts/fines in the
water column segments were 2.74 mg/L, 2.73 mg/L, and 51.94 mg/L, respectively.
EPA calibrated the model outputs by manipulating the sediment transport parameters until
the modeled concentrations in the benthic segments closely matched the available sediment
G-78
-------
Appendix G—Overview of Case Study Modeling Setup and Outputs
concentration monitoring data derived from STORET. Table G-53 presents the sediment transport
parameters resulting from EPA's calibration effort. EPA assumed the initial concentrations of
organic solids, sands, and silts/fines in the benthic segments were equal to 100 mg/L each.
Model Results
Case study modeling of the Mississippi River revealed water quality benchmark
exceedances in the immediate receiving water and/or in downstream segments for arsenic. Figure
G-17 illustrates the water concentration outputs for arsenic in the Rush Island plant immediate
receiving water before and after the assumed compliance date for the final rule.11
Case study modeling of the Mississippi River revealed that average water column
concentrations of arsenic in the Rush Island plant's immediate receiving water and/or downstream
segments would trigger exceedances of human health benchmarks. Table G-54 and Table G-55
illustrate the average modeled pollutant concentration in each water column segment downstream
of the Rush Island plant (including the immediate receiving water) for baseline and following
compliance with the final rule, respectively. Table G-56 and Table G-57 present the total miles
with average water column concentrations translating to exceedances of these benchmarks for
baseline and under the final rule, respectively.
11 To improve clarity, Figure G-17 presents the baseline water column concentrations leading up to the assumed
compliance date of Rush Island plant. All analyses of the WASP model outputs were performed on the baseline
output after the assumed compliance date.
-------
Appendix G—Overview of Case Study Modeling Setup and Outputs
Table G-46. USGS Stream Gages with Flow Data Used in Mississippi River WASP Model
Gage ID
7010000
7019000
USGS Gage
Location
Mississippi River
near St. Louis, MO
Meramec River near
Eureka, MO
Stream Flow Record Period
Full Record from 01/01/1880
-11/19/2014
Full Record from 10/01/1903
-02/04/2015
Cumulative Drainage
Area Represented by
Gage (sq km)
1,668,452
9,811
Model Boundary
Mississippi River
Meramec River
Cumulative Drainage
Area at Model
Boundary (sq km)
1,667,867
10,264
Scale Factor
1.000
1.046
Acronyms: USGS (U.S. Geological Survey).
Table G-47. Stream Flow Data Periods - Mississippi River WASP Model
Modeling Period
Corresponding Stream Flow Data Period
Mississippi River (Gage ID 7010000)
01/01/1982 -
10/01/2014-
01/01/1998 -
10/01/2030-
Meramec River (Gage ID
01/01/1982 -
10/01/2014-
01/01/1998 -
10/01/2030-
09/30/2014
12/31/2020
09/30/2030
12/31/2036
01/01/1982
10/01/2002
01/01/1982
10/01/2002
7019000)
09/30/2014
12/31/2020
09/30/2030
12/31/2036
01/01/1982
10/01/2002
01/01/1982
10/01/2002
-09/30/2014
-12/31/2008
-09/30/2014
-12/31/2008
-09/30/2014
-12/31/2008
-09/30/2014
-12/31/2008
G-80
-------
Appendix G—Overview of Case Study Modeling Setup and Outputs
Table G-48. Pollutant Loadings - Meramec Plant
Wastestream
Pollutant Loadings (g/day)
As
Cd
Cu
Ni
Pb
Se
Tl
Zn
Baseline a
FGD Wastewater
Fly Ash Transport Water
Bottom Ash Transport Water
Combustion Residual Leachate
Total
-
-
425.25
~
425.25
-
-
117.36
~
117.36
519.89
~
519.89
1,956.61
~
1,956.61
-
-
391.37
-
391.37
215.63
~
215.63
1,994.81
~
1,994.81
-
-
1,599.08
~
1,599.08
Final Rule b
FGD Wastewater
Fly Ash Transport Water
Bottom Ash Transport Water
Combustion Residual Leachate
Total
~
-
-
~
-
~
-
-
~
-
~
~
~
-
-
~
-
~
~
~
-
-
~
-
Acronyms: FGD (flue gas desulfurization).
a - The baseline pollutant loadings are modeled throughout the entire modeling period (from 01/01/1982 through 12/31/2036).
b - The final rule pollutant loadings are modeled only after the assumed compliance date (from 01/01/2019 through 12/31/2036).
G-81
-------
Appendix G—Overview of Case Study Modeling Setup and Outputs
Table G-49. Pollutant Loadings - Rush Island Plant
Wastestream
Pollutant Loadings (g/day)
As
Cd
Cu
Ni
Pb
Se
Tl
Zn
Baseline a
FGD Wastewater
Fly Ash Transport Water
Bottom Ash Transport Water
Combustion Residual Leachate
Total
-
2,617.69
109.07
~
2,726.76
-
338.24
55.52
~
393.76
1,490.40
381.96
~
1,872.36
1,152.47
330.76
~
1,483.22
-
1,054.61
173.11
-
1,227.72
1,171.40
11.83
~
1,183.23
1,220.86
200.40
~
1,421.26
-
3,112.40
334.33
~
3,446.73
Final Rule b
FGD Wastewater
Fly Ash Transport Water
Bottom Ash Transport Water
Combustion Residual Leachate
Total
~
-
-
~
-
~
-
-
~
-
~
~
~
-
-
~
-
~
~
~
-
-
~
-
Acronyms: FGD (flue gas desulfurization).
a - The baseline pollutant loadings are modeled from 01/01/1982 through 12/31/2022.
b - The final rule pollutant loadings are modeled from 01/01/2023 through 12/31/2036.
G-82
-------
Appendix G—Overview of Case Study Modeling Setup and Outputs
Table G-50. Pollutant Contributions from Non-Steam Electric Point Sources - Mississippi River WASP Model
Facility Name
MSD Meramec Treatment
Plant3
Doe Run Co Herculaneum
Smelter b
Doe Run Co Herculaneum
Smelter c
Model COMID
3629071
(Mississippi River)
3629127
(Mississippi River)
3634867 d
(Joachim Creek)
City, State
St. Louis, MO
Herculaneum, MO
Herculaneum, MO
Location (lat, long)
(38.39,-90.34)
(38.26,-90.38)
(38.26,-90.38)
Parameter
As
Cd
Cu
Ni
Pb
Zn
Cd
Cu
Pb
Zn
As
Cd
Cu
Ni
Pb
Zn
Average Daily Pollutant Loadings
(g/day)
139.02
3.50
52.2
52.5
156.5
999.6
156.87
11.56
49.42
66.51
6.09
6.09
8.35
0.61
280.97
36.80
a - EPA identified that this publicly operated treatment works (POTW) facility is a direct discharger with 2011 DMR loadings.
b - EPA identified that this industrial facility is a direct discharger with 2011 DMR loadings.
c - EPA identified that this facility is also an indirect discharger with 2011 TRI loadings.
d - These pollutant loadings for Doe Run Co Herculaneum are indirectly discharged to Joachim Creek via the Herculaneum Sewer District POTW.
G-83
-------
Appendix G—Overview of Case Study Modeling Setup and Outputs
Table G-51. Pollutant Contributions from STORET Monitoring Data - Mississippi River WASP Model
Model Boundary
Mississippi River
Maeystown Creek
South Gabouri
Creek
Model Boundary
COMID
3629071
3629179
3630453
Station ID(s) (lat, long)
1707.02/3.7 (38.43,-90.29)
GRW04449-331 (38.41,-90.32)
J-36 (38.40,-90.32)
1707.03/41.0 (38.36,-90.36)
JD-02 (38.21,-90.26)
1707.02/121/0.9/1. 5 (37.97,-90.06)
Parameter
As
Cd
Cu
Ni
Pb
Zn
TSS
TOC
As
Cd
Cu
Ni
Pb
Zn
TOC
TSS
TSS
Average Concentration
ftig/L)a
-
-
-
-
~
-
220,098.26
5,298.95
-
-
-
-
-
-
3,928.00
43,000.00
5,000.00
Mass Loading
(g/day)b
1,533,384.42
63,000.95
1,772,153.59
4,216,002.40
1,764,990.67
6,485,964.73
-
-
49.83
1.21
38.90
11.55
29.09
152.55
-
-
-
Acronyms: TOC (Total Organic Carbon); TSS (Total Suspended Solids).
a -Where more than one monitoring station located on the same tributary system reported acceptable results for the same pollutant, EPA calculated and incorporated
the weighted average concentration across the monitoring stations (weighted by number of samples at each station).
b - For the modeled pollutants (not including TOC and TSS), EPA converted the average concentration to a mass loading using the average annual flow rate for the
stream reach represented by the monitoring station(s).
G-84
-------
Appendix G—Overview of Case Study Modeling Setup and Outputs
Table G-52. Organic Solids, Sands, and Silts/Fines Inputs - Mississippi River WASP Model
Model Boundary
Mississippi River
Maeystown Creek
South Gabouri Creek
All Other Inflows d
Model Boundary
COMID
3629071
3629179
3630453
N/A
Organic Solids Concentration
(mg/L)a
2.65
1.96
*
2.31
Sands Concentration
(mg/L)b
11.00
2.15
0.25 e
4.47
Silts/Fines Concentration
(mg/L)c
209.09
40.85
4.75 e
84.90
Acronyms: N/A (Not Applicable).
* - No TOC results available. The 'All Other Inflows' concentration was used in this scenario.
a - The organic solids concentration was calculated using Equation G-l and the STORET monitoring data presented in Table G-51.
b - The sands concentration was calculated using Equation G-2 and the STORET monitoring data presented in Table G-51.
c - The silts/fines concentration was calculated using Equation G-3 and the STORET monitoring data presented in Table G-51.
d - For tributaries where boundary concentrations from STORET monitoring data were not available, EPA assumed the average boundary concentration from all
tributaries entering the modeling area.
e - These concentrations were calculated using the 'All Other Inflows' concentration.
Table G-53. Sediment Transport Parameters - Mississippi River WASP Model
Input Parameter
TAUcritcoh
TAU cDl sia
TAU cD2 sia
TAU cDl POa
TAU cD2 POa
Value Used
5.0
5.0
10.0
5.0
10.0
Units
N/m2
N/m2
N/m2
N/m2
N/m2
Note: Table G-l presents additional solids constants and sediment transport parameters that are used in
each of the case study models.
a - This parameter is a WASP model default based on the value of the 'TAUcritcoh' parameter.
G-85
-------
Appendix G—Overview of Case Study Modeling Setup and Outputs
a
5
0.1
0.01
2014 2015 2016 2017 2018 2019 2020
Assumed Compliance Date of Meramec
2021 2022 2023 2024 202S
Assumed Compliance Date of Rush Island
2026
2027
CASE STUDY OUTPUT IRW MODELOUTPUT AQUATIC LIFE NRWQC BENCHMARKS HUMAN HEALTH NRWQC BENCHMARKS DRINKING WATER BENCHMARKS
^^~ Baseline Baseline Freshwater Acute ConsumptionofOrganismsOnly Drinking Water MCL
^^FinalRufe FinalRule Freshwater Chronic Consumption of Water & Organisms
Figure G-17. Modeled Concentrations in Mississippi River Water Column at Rush Island Plant Immediate Receiving Water
(Total Arsenic)
G-86
-------
Appendix G—Overview of Case Study Modeling Setup and Outputs
Table G-54. Average Water Column Concentrations Downstream of Rush Island Plant at Baseline
Segment Data
Segment
ID
9
8
7
6
5
4
3
2
1
Segment Name
Mississippi River /
Rush Island IRW
Mississippi River
Mississippi River
Mississippi River
Mississippi River
Mississippi River
Mississippi River
Mississippi River
Mississippi River / End
Segment
Length
(mi)
1.48
2.69
4.33
2.21
1.25
2.93
1.40
1.92
5.06
Distance
Downstream
(mi)
1.48
4.17
8.49
10.70
11.95
14.88
16.27
18.19
23.25
Average Total Water Column Concentration over Modeling Period (jig/L)a
As
3.2912
4.0944
3.0972
3.1050
3.1057
3.1055
3.1065
3.1078
3.1123
Cd
0.1287
0.1237
0.1171
0.1174
0.1174
0.1173
0.1173
0.1173
0.1173
Cu
3.5878
3.5546
3.2754
3.2833
3.2834
3.2816
3.2819
3.2820
3.2832
Pb
3.5149
3.2102
3.1789
3.1859
3.1858
3.1835
3.1834
3.1831
3.1830
Ni
8.7116
9.5052
8.0477
8.0684
8.0693
8.0667
8.0682
8.0699
8.0766
Se
0.0044
0.0046
0.0040
0.0040
0.0040
0.0040
0.0040
0.0040
0.0040
Tl
0.0086
0.0099
0.0080
0.0080
0.0080
0.0080
0.0080
0.0080
0.0080
Zn
13.0108
12.3554
11.8195
11.8468
11.8467
11.8392
11.8395
11.8393
11.8412
Acronyms: IRW (Immediate receiving water).
a - Concentrations represent the average daily total pollutant concentration in the water column. The averaging period is the entire modeling period after the assumed
compliance date.
G-87
-------
Appendix G—Overview of Case Study Modeling Setup and Outputs
Table G-55. Average Water Column Concentrations Downstream of Rush Island Plant Under Final Rule
Segment Data
Segment
ID
9
8
7
6
5
4
o
J
2
1
Segment Name
Mississippi River /
Rush Island IRW
Mississippi River
Mississippi River
Mississippi River
Mississippi River
Mississippi River
Mississippi River
Mississippi River
Mississippi River / End
Segment
Length
(mi)
1.48
2.69
4.33
2.21
1.25
2.93
1.40
1.92
5.06
Distance
Downstream
(mi)
1.48
4.17
8.49
10.70
11.95
14.88
16.27
18.19
23.25
Average Total Water Column Concentration over Modeling Period (jig/L)a
As
3.2833
4.0847
3.0898
3.0976
3.0984
3.0982
3.0992
3.1004
3.1049
Cd
0.1275
0.1225
0.1159
0.1162
0.1162
0.1161
0.1162
0.1162
0.1162
Cu
3.5819
3.5488
3.2700
3.2779
3.2780
3.2763
3.2765
3.2767
3.2779
Pb
3.5109
3.2066
3.1753
3.1823
3.1822
3.1799
3.1798
3.1795
3.1795
Ni
8.7034
9.4964
8.0402
8.0609
8.0618
8.0592
8.0607
8.0624
8.0691
Se
0.0008
0.0009
0.0008
0.0008
0.0008
0.0008
0.0008
0.0008
0.0008
Tl
0.0006
0.0006
0.0005
0.0005
0.0005
0.0005
0.0005
0.0005
Zn
12.9984
12.3438
11.8083
11.8357
11.8356
11.8281
11.8284
11.8282
0.0005 11.8300
Acronyms: IRW (Immediate receiving water).
a - Concentrations represent the average daily total pollutant concentration in the water column. The averaging period is the entire modeling period after the assumed
compliance date.
Goo
-OO
-------
Appendix G—Overview of Case Study Modeling Setup and Outputs
Table G-56. Total Miles of Mississippi River with Wildlife And Human Health Impacts at Baseline
Wildlife and Human Health Impact Thresholds
WL - NEHC, T3 (mink)
WL - NEHC, T4 (eagle)
HH - Non-Cancer Adult Subsistence
HH - Non-Cancer Adult Recreational
HH - Non-Cancer Child Subsistence (1 to <2 y.o.) a
HH - Non-Cancer Child Subsistence (16 to <21 y.o.) b
HH - Non-Cancer Child Recreational (1 to <2 y.o.) a
HH - Non-Cancer Child Recreational (16 to <21 y.o.)b
HH - Cancer Adult Subsistence
HH - Cancer Adult Recreational
HH - Cancer Child Subsistence (6 to <11 y.o.) a
HH - Cancer Child Subsistence (1 to <2 y.o.) b
HH - Cancer Child Recreational (6 to 1 1 y.o.) a
HH - Cancer Child Recreational (1 to <2 y.o.)b
Total Miles with Average Water Column Concentration Translating to Wildlife or Human Health
Benchmark Exceedances (mi)
As
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
23.25
0.00
0.00
0.00
0.00
0.00
Cd
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
No LECR
No LECR
No LECR
No LECR
No LECR
No LECR
Cu
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
No LECR
No LECR
No LECR
No LECR
No LECR
No LECR
Pb
0.00
0.00
NoRfD
NoRfD
NoRfD
NoRfD
NoRfD
NoRfD
No LECR
No LECR
No LECR
No LECR
No LECR
No LECR
Ni
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
No LECR
No LECR
No LECR
No LECR
No LECR
No LECR
Se
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
No LECR
No LECR
No LECR
No LECR
No LECR
No LECR
Tl
No NEHC
No NEHC
0.00
0.00
0.00
0.00
0.00
0.00
No LECR
No LECR
No LECR
No LECR
No LECR
No LECR
Zn
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
No LECR
No LECR
No LECR
No LECR
No LECR
No LECR
Acronyms: WL (Wildlife); HH (Human health); NEHC (No effect hazard concentration); Rfd (Reference dose); LECR (Lifetime excess cancer risk); y.o. (year old).
a - This row represents the most sensitive child fisher cohort.
b - This row represents the least sensitive child fisher cohort.
G-89
-------
Appendix G—Overview of Case Study Modeling Setup and Outputs
Table G-57. Total Miles of Mississippi River with Wildlife And Human Health Impacts Under Final Rule
Wildlife and Human Health Impact Thresholds
WL - NEHC, T3 (mink)
WL - NEHC, T4 (eagle)
HH - Non-Cancer Adult Subsistence
HH - Non-Cancer Adult Recreational
HH - Non-Cancer Child Subsistence (1 to <2 y.o.) a
HH - Non-Cancer Child Subsistence (16 to <21 y.o.) b
HH - Non-Cancer Child Recreational (1 to <2 y.o.) a
HH - Non-Cancer Child Recreational (16 to <21 y.o.)b
HH - Cancer Adult Subsistence
HH - Cancer Adult Recreational
HH - Cancer Child Subsistence (6 to <11 y.o.) a
HH - Cancer Child Subsistence (1 to <2 y.o.) b
HH - Cancer Child Recreational (6 to 1 1 y.o.) a
HH - Cancer Child Recreational (1 to <2 y.o.)b
Total Miles with Average Water Column Concentration Translating to Wildlife or Human Health
Benchmark Exceedances (mi)
As
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
23.25
0.00
0.00
0.00
0.00
0.00
Cd
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
No LECR
No LECR
No LECR
No LECR
No LECR
No LECR
Cu
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
No LECR
No LECR
No LECR
No LECR
No LECR
No LECR
Pb
0.00
0.00
NoRfD
NoRfD
NoRfD
NoRfD
NoRfD
NoRfD
No LECR
No LECR
No LECR
No LECR
No LECR
No LECR
Ni
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
No LECR
No LECR
No LECR
No LECR
No LECR
No LECR
Se
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
No LECR
No LECR
No LECR
No LECR
No LECR
No LECR
Tl
No NEHC
No NEHC
0.00
0.00
0.00
0.00
0.00
0.00
No LECR
No LECR
No LECR
No LECR
No LECR
No LECR
Zn
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
No LECR
No LECR
No LECR
No LECR
No LECR
No LECR
Acronyms: WL (Wildlife); HH (Human health); NEHC (No effect hazard concentration); Rfd (Reference dose); LECR (Lifetime excess cancer risk); y.o. (year old).
a - This row represents the most sensitive child fisher cohort.
b - This row represents the least sensitive child fisher cohort.
G-90
-------
Appendix G—Overview of Case Study Modeling Setup and Outputs
CASE STUDY MODEL SETUPS AND OUTPUTS - LAKE SINCLAIR, GA
This section presents information regarding the site-specific design, site-specific input
parameters (e.g., background pollutant concentrations, EFDC model flow data), model settings
(e.g., sediment transport parameters), and case study modeling results for the Lake Sinclair case
study model.
Model Development & Input Variables
WASP Model Design. As discussed in Section 8.1.1 of the EA Report, EPA relied on the
availability of an existing water quality model to perform case study modeling of Lake Sinclair.
In contrast to the lotic case study models, the Lake Sinclair WASP model relies on Environmental
Fluid Dynamics Code (EFDC) hydrodynamics to simulate the aquatic system in three
dimensions.12 The scope of the Lake Sinclair WASP model is limited by the boundaries of the pre-
existing EFDC hydrodynamics. The modeling area encompasses the main body of Lake Sinclair,
from Wallace Dam to Sinclair Dam, and the major tributaries feeding into the Lake.
The three-dimensional EFDC model, which provides the hydrodynamic foundation for the
WASP model, divides the waterbody into 1,235 segments. Each segment represents a unique
location and stratum within Lake Sinclair. The EFDC model uses stretch or sigma vertical
coordinates and Cartesian coordinates to represent the physical characteristics of Lake Sinclair.
Plant Harllee Branch's immediate receiving water is identified by the coordinate code 1=30 J=32
K=5, where each coordinate represents the position on x, y, and z axes, respectively. The Lake
Sinclair model does not have any segments representing benthic sediment. The model accounts for
a total volume of approximately 340 million cubic meters.
As discussed earlier in this section, EPA adopted the preexisting Lake Sinclair EFDC
model. The pre-existing model was designed with seven years of hydrodynamic and flow input,
limiting the length of the period EPA could model. Based on Plant Harllee Branch's NPDES
permitting cycle, EPA assumed that the plant would have achieved the limitations under the final
rule by 2019, if it continued to operate. The modeling period begins in February 2012
(approximately seven years before the assumed compliance date) and extends through November
2025 (approximately seven years after the assumed compliance date).
Incorporation of Flow Data. EPA did not incorporate any USGS flow data into the Lake
Sinclair WASP model. Instead, EPA used the seven years of hydrodynamic and flow input
integrated into the EFDC model. Table G-58 presents how the EFDC hydrodynamic data were
incorporated into the model to complete a full record of flow data for the entire modeling period.
Model Input Variables. As discussed in Section 8.2.6 of the EA Report, Plant Harllee
Branch retired all of coal-fired generating units in April 2015. Despite the retirement of this plant,
EPA proceeded with case study modeling of Lake Sinclair to represent the potential impacts of
steam electric discharges on lentic waterbodies. Table G-59 presents the pollutant loadings
modeled from Plant Harllee Branch at the evaluated wastestream level, both at baseline and after
12 The Black Creek, Etowah River, Lick Creek and White River, Ohio River, and Mississippi River case study
models relied on NHDPlus Version 1 hydrodynamics for simulating lotic aquatic systems.
-------
Appendix G—Overview of Case Study Modeling Setup and Outputs
the plant achieves the limitations under the final rule.13 EPA did not identify any point sources
with 2011 DMR or TRI loadings which would impact the Lake Sinclair case study model.
Table G-60 presents the pollutant contributions flowing into the Lake Sinclair WASP
model boundaries calculated using available STORET monitoring data.
Table G-61 presents the initial concentrations for the organic solids, sands, and silts/fines
values derived from STORET monitoring data collected. For tributaries where STORET
monitoring data were not available, EPA assumed the average boundary concentration from all
tributaries entering the modeling area. Based on the average of STORET data available within the
model, EPA calculated the initial concentrations of organic solids, sands, and silts/fines in the
water column segments were 1.91 mg/L, 0.20 mg/L, and 3.85 mg/L, respectively.
Model Results
Case study modeling of Lake Sinclair revealed water quality benchmark exceedances in
the immediate receiving water and neighboring segments for arsenic and thallium. Figure G-18
illustrates the water concentration outputs averaged for all model segments before and after the
assumed compliance date for the final rule.14 Case study modeling also revealed frequent (more
than 50 percent of the modeling period) water quality benchmark exceedances of three pollutants
(arsenic, cadmium, and thallium) in some segments of Lake Sinclair.
Case study modeling of the Lake Sinclair revealed that the average water column
concentrations of thallium of all segments in the WASP model would trigger exceedances of
human health benchmarks.
13 EPA calculated pollutant loadings at the wastestream level for Plant Harllee Branch using the same loadings
methodology that EPA used for other plants in the loadings analyses. EPA did not include Plant Harllee Branch or
Lake Sinclair in the other quantitative and qualitative analyses in this EA for the final rule (e.g., the IRW model).
14 To improve clarity, Figure G-18 presents the baseline water column concentrations leading up to the assumed
compliance date of Plant Harllee Branch. All analyses of the WASP model outputs were performed on the baseline
output after the assumed compliance date.
-------
Appendix G—Overview of Case Study Modeling Setup and Outputs
Table G-58. Stream Flow Data Periods - Lake Sinclair WASP Model
Modeling Period
Corresponding Stream Flow Data Period
Lake Sinclair (EFDC Hydrodynamic Model)
02/01/2012-12/31/2018
01/01/2019-11/30/2025
2/1/2001-12/31/2007
2/1/2001-12/31/2007
G-93
-------
Appendix G—Overview of Case Study Modeling Setup and Outputs
Table G-59. Pollutant Loadings - Plant Harllee Branch
Wastestream
Pollutant Loadings (g/day)
As
Cd
Cu
Ni
Pb
Se
Tl
Zn
Baseline a
FGD Wastewater c
Fly Ash Transport Water
Bottom Ash Transport Water
Combustion Residual Leachate
Total
35.18
44.28
22.29
~
101.75
521.69
12.01
6.15
~
539.85
100.78
97.91
27.25
~
225.94
4,068.20
55.28
102.56
~
4,226.04
21.61
39.77
20.52
~
81.90
5,413.46
14.80
11.30
~
5,439.56
63.65
13.57
104.56
~
181.78
6,451.65
360.25
83.82
~
6,895.72
Final Rule b
FGD Wastewater
Fly Ash Transport Water
Bottom Ash Transport Water
Combustion Residual Leachate
Total
27.03
-
-
~
27.03
19.50
-
-
~
19.50
17.50
-
17.50
29.21
~
29.21
15.71
-
-
~
15.71
26.51
~
26.51
45.44
~
45.44
92.54
-
-
~
92.54
Acronyms: FGD (flue gas desulfurization).
Note: Plant Harllee Branch has retired all coal-fired generating units. EPA calculated pollutant loadings at the wastestream level for Plant Harllee Branch using the
same loadings methodology that EPA used for other plants in the loadings analyses. EPA did not include Plant Harllee Branch in the other quantitative and
qualitative analyses in this EA for the final rule (e.g., the IRW model).
a - The baseline pollutant loadings are modeled throughout the entire modeling period (from 02/01/2012 through 11/30/2025).
b - The final rule pollutant loadings are modeled only after the assumed compliance date (from 01/01/2019 through 11/30/2025).
c - In estimating the historical pollutant loadings associated with Plant Harllee Branch's FGD systems, EPA incorporated the pollutant loadings from FGD
wastewater when the system was installed in 2013. EPA did not model any FGD wastewater pollutant loadings before the installation of Plant Harllee Branch's FGD
system.
G-94
-------
Appendix G—Overview of Case Study Modeling Setup and Outputs
Table G-60. Pollutant Contributions from STORET Monitoring Data - Lake Sinclair WASP Model
Model Boundary
Oconee River
Crooked Creek
Rooty Creek
Little River
Model Boundary
COMID
1057503
1056407
1057629
1057681
Station ID(s) (lat, long)
0301100602 (33.35,-83. 16)
3038901 (33.35,-83.16)
0301100603 (33.33,-83. 14)
0301 180202 (33.32,-83.28)
0301180301 (33.32,-83.27)
3040101 (33.32,-83.37)
0301180302 (33.29,-83.35)
3040501 (33.29,-83.25)
3042001 (33.30,-83.42)
0301150301 (33.29,-83.43)
0301 150302 (33.29,-83.43)
3041701 (33.31,-83.44)
0301150102 (33. 31,-83.44)
Parameter
TOC
TSS
TOC
TSS
As
Cd
Cu
Ni
Pb
Tl
Zn
TOC
TSS
As
Cd
Cu
Ni
Pb
Tl
Zn
TOC
TSS
Average Concentration
Oig/L)a
3,818.44
6,941.46
7,124.62
18,992.31
~
-
-
-
~
-
-
5,347.26
11,635.71
-
-
-
-
-
~
-
4,960.21
15,576.92
Mass Loading
(g/day)b
-
-
-
-
58.89
14.99
45.10
33.07
29.59
58.95
452.25
~
-
960.78
243.11
1,037.67
644.08
482.01
961.37
6,098.66
-
-
G-95
-------
Appendix G—Overview of Case Study Modeling Setup and Outputs
Table G-60. Pollutant Contributions from STORET Monitoring Data - Lake Sinclair WASP Model
Model Boundary
Murder Creek
Big Cedar Creek
Model Boundary
COMID
1057679
1056893
Station ID(s) (lat, long)
0301 160703 (33.27,-83.48)
3043401 (33.25,-83.48)
0301 160701 (33.25,-83.48)
3043801 (33.19,-83.44)
0301170401 (33. 19,-83.44)
Parameter
As
Cd
Cu
Ni
Pb
Tl
Zn
TOC
TSS
As
Cd
Cu
Ni
Pb
Tl
Zn
TOC
TSS
Average Concentration
Oig/L)a
-
-
-
-
~
-
-
2,773.47
21,383.33
-
-
-
-
-
~
-
3,407.30
20,223.08
Mass Loading
(g/day)b
642.79
162.65
328.26
347.78
322.48
643.18
1,654.57
~
-
450.16
113.90
229.89
243.56
225.84
450.44
345.37
-
-
Acronyms: TOC (Total Organic Carbon); TSS (Total Suspended Solids).
a -Where more than one monitoring station located on the same tributary system reported acceptable results for the same pollutant, EPA calculated and incorporated
the weighted average concentration across the monitoring stations (weighted by number of samples at each station).
b - For the modeled pollutants (not including TOC and TSS), EPA converted the average concentration to a mass loading using the average annual flow rate for the
stream reach represented by the monitoring station(s).
G-96
-------
Appendix G—Overview of Case Study Modeling Setup and Outputs
Table G-61. Organic Solids, Sands, and Silts/Fines Inputs - Lake Sinclair WASP Model
Model Boundary
Oconee River
Crooked Creek
Rooty Creek
Little River
Murder Creek
Big Cedar Creek
All Other Inflows d
Model Boundary
COMID
1057503
1056407
1057629
1057681
1057679
1056893
N/A
Organic Solids Concentration
(mg/L)a
1.91
3.56
2.67
2.48
1.39
1.70
2.29
Sands Concentration
(mg/L)b
0.35
0.95
0.58
0.78
1.07
1.01
0.79
Silts/Fines Concentration
(mg/L)c
6.59
18.04
11.05
14.80
20.31
19.21
15.00
Acronyms: N/A (Not Applicable).
a - The organic solids concentration was calculated using Equation G-l and the STORET monitoring data presented in Table G-60.
b - The sands concentration was calculated using Equation G-2 and the STORET monitoring data presented in Table G-60.
c - The silts/fines concentration was calculated using Equation G-3 and the STORET monitoring data presented in Table G-60.
d - For tributaries where boundary concentrations from STORET monitoring data were not available, EPA assumed the average boundary concentration from all
tributaries entering the modeling area.
Table G-62. Sediment Transport Parameters - Lake Sinclair WASP Model
Input Parameter
TAUcritcoh
TAU cDl sia
TAU cD2 sia
TAU cDl POa
TAU cD2 POa
Value Used
3.5
3.5
7.0
3.5
7.0
Units
N/m2
N/m2
N/m2
N/m2
N/m2
Note: Table G-l presents additional solids constants and sediment transport parameters that are used in each
of the case study models.
a - This parameter is a WASP model default based on the value of the 'TAUcritcoh' parameter.
G-97
-------
Appendix G—Overview of Case Study Modeling Setup and Outputs
S
o
0.01
2012 2013 2014 2015 2016 2017 2018 2019 2020 2021 2022 2023 2024 2025
to
3
o
0.01
2012 2013 2014 2015
2016 2017 2018 2019 2020 2021
Assumed Compliance Date of Plant Harllee Branch
2022 2023 2024 2025
CASE STUDY OUTPUT IRW MODEL OUTPUT AQUATIC LIFE NRWQC BENCHMARKS HUMAN HEALTH NRWQC BENCHMARKS DRINKING WATER BENCHMARKS
^^~ Baseline Baseline Freshwater Acute Consumption of OrganismsOnly DrinkingWater MCL
^^~ Final Rule RnalRule Freshwater Chronic Consumption of Water & Organisms
Figure G-18. Average Modeled Concentrations in All Segments in Lake Sinclair WASP Model (Total Arsenic, Total Thallium)
G-98
-------
Appendix H—Additional Model Results
APPENDIX H
ADDITIONAL MODEL RESULTS
Table H-l. Number and Percentage of Immediate Receiving Waters that Exceeded a
Criterion by Pollutant and Criteria Type at Baseline Pollutant Loadings
Pollutant
Arsenic
Cadmium
Chromium VI
Copper
Lead
Mercury
Nickel
Selenium
Thallium
Zinc
Number of Immediate Receiving Waters that Exceeded a Criterion a
Freshwater
Acute
NRWQC
3(c)
9(c)
0(c)
6(c)
0(c)
l(c)
2(c)
No criterion
No criterion
4(c)
Freshwater
Chronic
NRWQC
4(c)
29 (c)
0(c)
7(c)
5(c)
l(c)
8(c)
33
No criterion
4(c)
Human
Health
Water and
Organism
NRWQC
94 (d)
No
criterion
No
criterion
0
No
criterion
No
criterion
4
8
49
1
Human
Health
Organism
Only
NRWQC
65 (d)
No
criterion
No
criterion
No
criterion
No
criterion
No
criterion
0
1
45
0
Drinking
Water
MCL
12
11
0(e)
o(f);
ife)
7(f)
5(d)
No
criterion
12
34
Ife)
Total Receiving Waters b
Number
Exceeding
94
29
0
7
7
5
8
33
49
4
Percentage
Exceeding
45%
14%
0%
3%
3%
2%
4%
16%
23%
2%
Source: ERG, 2015d; ERG, 2015h.
Acronyms: MCL (Maximum Contaminant Level); NRWQC (National Recommended Water Quality Criteria).
a - A total of 209 immediate receiving waters (183 rivers and streams; 26 lakes, ponds, and reservoirs) were
included in the water quality model. Table C-7 presents the criteria used for the analysis.
b - These values are the sum and percentage of rivers, streams, lakes, ponds, and reservoirs impacted.
c - NRWQC is expressed in terms of the dissolved pollutant in the water column.
d - NRWQC or MCL is for inorganic form of metal. For the benchmark comparison, EPA used the total pollutant
concentration in the water column. This might overestimate the number of exceedances.
e - MCL is for total chromium.
f - MCL used for comparison is the drinking water action level.
g - MCL used for comparison is a secondary (nonenforceable) drinking water standard.
H-l
-------
Appendix H—Additional Model Results
Figure H-l. Baseline Total Arsenic Concentration in the
Immediate Receiving Water
Total Arsenic Concentration in Receiving Water (mg/L)
Source: ERG, 2015d; ERG, 2015h.
Table H-2. Total Arsenic Concentration (mg/L) in the Immediate Receiving Water by
Percentile
Percentile
5th
25th
50th
75th
95th
Max
Scenario
Baseline
3.45E-08
9.61E-07
7.88E-06
0.001
0.016
1.86
Option A
2.07E-08
6.28E-07
5.49E-06
4.40E-04
0.008
1.86
Option B
2.07E-08
6.28E-07
5.49E-06
4.40E-04
0.008
1.86
Option C
0
1.21E-07
2.82E-06
9.23E-05
0.006
1.86
Option D
0
0
3.62E-07
1.62E-05
0.003
1.86
Option E
0
0
1.93E-07
9.68E-06
9.76E-04
1.13
Source: ERG, 2015d; ERG, 2015h.
H-2
-------
Appendix H—Additional Model Results
Figure H-2. Baseline Total Cadmium Concentration in the
Immediate Receiving Water
Total Cadmium Concentration in Receiving Water (mg/L)
Source: ERG, 2015d; ERG, 2015h.
Table H-3. Total Cadmium Concentration (mg/L) in the Immediate Receiving Water by
Percentile
Percentile
5th
25th
50th
75th
95th
Max
Scenario
Baseline
1.43E-08
5.10E-07
5.15E-06
1.75E-04
0.005
0.490
Option A
1.04E-08
2.25E-07
2.10E-06
1.22E-04
0.003
0.490
Option B
1.04E-08
2.25E-07
2.10E-06
1.22E-04
0.003
0.490
Option C
0
5.15E-08
9.87E-07
3.66E-05
0.002
0.490
Option D
0
0
1.54E-07
8.42E-06
0.001
0.490
Option E
0
0
1.36E-07
6.99E-06
7.04E-04
0.204
Source: ERG, 2015d; ERG, 2015h.
H-3
-------
Appendix H—Additional Model Results
Figure H-3. Chromium VI Concentration in the Immediate
Receiving Water
v v
Chromium VI Concentration in Receiving Water (mg/L)
Source: ERG, 2015d; ERG, 2015h.
Table H-4. Chromium VI Concentration (mg/L) in the Immediate Receiving Water by
Percentile
Percentile
5th
25th
50th
75th
95th
Max
Scenario
Baseline
0
0
0
0
5.38E-06
0.019
Option A
0
0
0
0
1.33E-06
0.013
Option B
0
0
0
0
1.33E-06
0.013
Option C
0
0
0
0
7.87E-08
0.013
Option D
0
0
0
0
0
0.013
Option E
0
0
0
0
0
0.013
Source: ERG, 2015d; ERG, 2015h.
H-4
-------
Appendix H—Additional Model Results
Figure H-4. Baseline Total Copper Concentration in the
Immediate Receiving Water
Total Copper Concentration in Receiving Water (mg/L)
Source: ERG, 2015d; ERG, 2015h.
Table H-5. Total Copper Concentration (mg/L) in the Immediate Receiving Water by
Percentile
Percentile
5th
25th
50th
75th
95th
Max
Scenario
Baseline
1.64E-08
8.86E-07
8.30E-06
2.81E-04
0.015
1.15
Option A
1.01E-08
5.37E-07
6.27E-06
2.33E-04
0.009
0.778
Option B
1.01E-08
5.37E-07
6.27E-06
2.33E-04
0.009
0.778
Option C
0
7.86E-08
1.57E-06
4.21E-05
0.002
0.778
Option D
0
0
1.33E-07
7.10E-06
0.001
0.778
Option E
0
0
1.21E-07
6.27E-06
6.32E-04
0.778
Source: ERG, 2015d; ERG, 2015h.
H-5
-------
Appendix H—Additional Model Results
Figure H-5. Baseline Total Lead Concentration in the
Immediate Receiving Water
Total Lead Concentration in Receiving Water (mg/L)
Source: ERG, 2015d; ERG, 2015h.
Table H-6. Total Lead Concentration (mg/L) in the Immediate Receiving Water by
Percentile
Percentile
5th
25th
50th
75th
95th
Max
Scenario
Baseline
1.41E-09
4.47E-07
3.61E-06
7.65E-05
0.009
0.757
Option A
0
2.22E-07
2.91E-06
6.98E-05
0.007
0.510
Option B
0
2.22E-07
2.91E-06
6.98E-05
0.007
0.510
Option C
0
1.36E-09
3.65E-07
5.99E-06
0.001
0.510
Option D
0
0
2.65E-09
4.47E-07
7.22E-05
0.510
Option E
0
0
2.65E-09
4.47E-07
7.22E-05
0.510
Source: ERG, 2015d; ERG, 2015h.
H-6
-------
Appendix H—Additional Model Results
Figure H-6. Baseline Total Mercury Concentration in the
Immediate Receiving Water
\-
Total Mercury Concentration in Receiving Water (mg/L)
Source: ERG, 2015d; ERG, 2015h.
Table H-7. Total Mercury Concentration (mg/L) in the Immediate Receiving Water by
Percentile
Percentile
5th
25th
50th
75th
95th
Max
Scenario
Baseline
1.70E-09
4.50E-08
3.56E-07
1.68E-05
0.001
0.056
Option A
5.32E-10
2.29E-08
1.79E-07
1.34E-05
2.62E-04
0.020
Option B
3.94E-10
1.86E-08
1.77E-07
1.28E-05
2.58E-04
0.020
Option C
0
1.86E-09
6.24E-08
2.31E-06
1.15E-04
0.020
Option D
0
0
4.20E-09
2.14E-07
4.17E-05
0.020
Option E
0
0
2.32E-09
1.05E-07
8.96E-06
0.020
Source: ERG, 2015d; ERG, 2015h.
H-7
-------
Appendix H—Additional Model Results
Figure H-7. Baseline Total Nickel Concentration in the
Immediate Receiving Water
Total Nickel Concentration in Receiving Water (mg/L)
Source: ERG, 2015d; ERG, 2015h.
Table H-8. Total Nickel Concentration (mg/L) in the Immediate Receiving Water by
Percentile
Percentile
5th
25th
50th
75th
95th
Max
Scenario
Baseline
7.14E-08
3.31E-06
3.34E-05
0.001
0.049
2.25
Option A
4.16E-08
1.31E-06
1.81E-05
0.001
0.034
2.25
Option B
3.00E-08
1.11E-06
1.81E-05
0.001
0.033
2.25
Option C
0
1.86E-07
4.58E-06
1.37E-04
0.008
2.25
Option D
0
0
4.17E-07
1.62E-05
0.004
2.25
Option E
0
0
2.47E-07
1.05E-05
0.002
0.616
Source: ERG, 2015d; ERG, 2015h.
H-8
-------
Appendix H—Additional Model Results
Figure H-8. Baseline Total Selenium Concentration in the
Immediate Receiving Water
Total Selenium Concentration in Receiving Water (mg/L)
Source: ERG, 2015d; ERG, 2015h.
Table H-9. Total Selenium Concentration (mg/L) in the Immediate Receiving Water by
Percentile
Percentile
5th
25th
50th
75th
95th
Max
Scenario
Baseline
9.12E-08
2.74E-06
5.46E-05
0.001
0.064
5.38
Option A
3.84E-08
2.46E-06
3.67E-05
0.001
0.040
5.38
Option B
2.05E-08
5.01E-07
5.30E-06
3.08E-04
0.017
5.38
Option C
0
1.19E-07
2.35E-06
9.68E-05
0.013
5.38
Option D
0
0
3.82E-07
2.61E-05
0.010
5.38
Option E
0
0
3.82E-07
2.61E-05
0.010
5.38
Source: ERG, 2015d; ERG, 2015h.
H-9
-------
Appendix H—Additional Model Results
Figure H-9. Baseline Total Thallium Concentration in the
Immediate Receiving Water
Total Thallium Concentration in Receiving Water (mg/L)
Source: ERG, 2015d; ERG, 2015h.
Table H-10. Total Thallium Concentration (mg/L) in the Immediate Receiving Water by
Percentile
Percentile
5th
25th
50th
75th
95th
Max
Scenario
Baseline
1.09E-08
1.31E-06
1.49E-05
1.91E-04
0.035
1.75
Option A
5.95E-09
7.82E-07
1.20E-05
1.54E-04
0.033
1.75
Option B
5.95E-09
7.82E-07
1.20E-05
1.54E-04
0.033
1.75
Option C
0
6.08E-08
2.33E-06
3.71E-05
0.004
1.75
Option D
0
0
1.89E-07
5.87E-06
3.42E-04
0.591
Option E
0
0
1.89E-07
5.87E-06
3.42E-04
0.591
Source: ERG, 2015d; ERG, 2015h.
H-10
-------
Appendix H—Additional Model Results
Figure H-10. Baseline Total Zinc Concentration in the
Immediate Receiving Water
Total Zinc Concentration in Receiving Water (mg/L)
Source: ERG, 2015d; ERG, 2015h.
Table H-ll. Total Zinc Concentration (mg/L) in the Immediate Receiving Water by
Percentile
Percentile
5th
25th
50th
75th
95th
Max
Scenario
Baseline
2.07E-07
5.40E-06
6.37E-05
0.002
0.081
10.2
Option A
9.14E-08
2.43E-06
2.12E-05
0.002
0.039
10.2
Option B
9.14E-08
2.43E-06
2.12E-05
0.002
0.039
10.2
Option C
0
4.67E-07
1.10E-05
4.11E-04
0.032
10.2
Option D
0
0
1.44E-06
7.72E-05
0.019
10.2
Option E
0
0
7.84E-07
3.54E-05
0.003
1.43
Source: ERG, 2015d; ERG, 2015h.
H-ll
-------
Appendix H—Additional Model Results
Figure H-ll. Baseline Total Arsenic Concentration in
Trophic Level 3 and Trophic Level 4 Fish Tissue in the
Immediate Receiving Watera
Total Arsenic Concentration in Fish Tissue (mg/kg)
Source: ERG, 2015d; ERG, 20151.
a - The wildlife module applies the same total arsenic bioconcentration factors (BCFs) for both trophic level 3 (T3)
and trophic level 4 (T4) fish (see Appendix D). Therefore, the estimated concentrations presented here are identical
for both trophic levels.
Table H-12. Total Arsenic Concentration (mg/kg) in Fish Tissue (Trophic Level 3 &
Trophic Level 4) by Percentile a
Percentile
5th
25th
50th
75th
95th
Max
Scenario
Baseline
1.38E-07
3.85E-06
3.15E-05
0.002
0.062
7.45
Option A
8.28E-08
2.51E-06
2.20E-05
0.002
0.032
7.45
Option B
8.28E-08
2.51E-06
2.20E-05
0.002
0.032
7.45
Option C
0
4.86E-07
1.13E-05
3.69E-04
0.024
7.45
Option D
0
0
1.45E-06
6.49E-05
0.014
7.45
Option E
0
0
7.71E-07
3.87E-05
0.004
4.53
Source: ERG, 2015d; ERG, 20151.
a - The wildlife module applies the same total BCFs for both trophic level 3 (T3) and trophic level 4 (T4) fish (see
Appendix D). Therefore, the estimated concentrations presented here are identical for both trophic levels.
H-12
-------
Appendix H—Additional Model Results
Figure H-12. Baseline Total Cadmium Concentration in
Trophic Level 3 and Trophic Level 4 Fish Tissue in the
Immediate Receiving Watera
Total Cadmium Concentration in Fish Tissue (mg/kg)
Source: ERG, 2015d; ERG, 20151.
a - The wildlife module applies the same total cadmium BCFs for both T3 and T4 fish (see Appendix DError!
Reference source not found.). Therefore, the estimated concentrations presented here are identical for both
trophic levels.
Table H-13. Total Cadmium Concentration (mg/kg) in Fish Tissue (Trophic Level 3 &
Trophic Level 4) by Percentile a
Percentile
5th
25th
50th
75th
95th
Max
Scenario
Baseline
3.85E-06
1.38E-04
0.001
0.047
1.40
132
Option A
2.81E-06
6.08E-05
5.67E-04
0.033
0.738
132
Option B
2.81E-06
6.08E-05
5.67E-04
0.033
0.738
132
Option C
0
1.39E-05
2.66E-04
0.010
0.505
132
Option D
0
0
4.17E-05
0.002
0.332
132
Option E
0
0
3.67E-05
0.002
0.190
55.1
Source: ERG, 2015d; ERG, 20151.
a - The wildlife module applies the same total cadmium BCFs for both trophic level 3 (T3) and trophic level 4 (T4)
fish (see Appendix D). Therefore, the estimated concentrations presented here are identical for both trophic levels.
H-13
-------
Appendix H—Additional Model Results
Figure H-13. Baseline Chromium VI Concentration in
Trophic Level 3 and Trophic Level 4 Fish Tissue in the
Immediate Receiving Watera
250 -i
200 -
M)
.> 150
100 -
s
s
Z
50 -
Chromium VI Concentration in Fish Tissue (mg/kg)
Source: ERG, 2015d; ERG, 20151.
a - BCFs for chromium VI are not available; EPA used the total chromium BCF values. The wildlife module applies
the same total chromium BCFs for both T3 and T4 fish (see Appendix D). Therefore, the estimated concentrations
presented here are identical for both trophic levels.
Table H-14. Chromium VI Concentration (mg/kg) in Fish Tissue (Trophic Level 3 &
Trophic Level 4) by Percentile a
Percentile
5th
25th
50th
75th
95th
Max
Scenario
Baseline
0
0
0
0
3.67E-07
0.011
Option A
0
0
0
0
5.18E-08
0.008
Option B
0
0
0
0
5.18E-08
0.008
Option C
0
0
0
0
3.91E-09
0.008
Option D
0
0
0
0
0
0.008
Option E
0
0
0
0
0
0.008
Source: ERG, 2015d; ERG, 20151.
a - The wildlife module applies the same total chromium BCFs for both trophic level 3 (T3) and trophic level 4 (T4)
fish (see Appendix D). Therefore, the estimated concentrations presented here are identical for both trophic levels.
H-14
-------
Appendix H—Additional Model Results
K
u
Figure H-14. Total Copper Concentration in Trophic Level 3
and Trophic Level 4 Fish Tissue in the Immediate Receiving
Watera
«
£
M
=
T
"53
u
u
rt
O)
is
•3
u
s
I
(M
O
b.
a
50 -
45 -
40 -
35 -
30 -
25 -
20 -
15 -
10 -
5 -
0 -
s
Z
Total Copper Concentration in Fish Tissue (mg/kg)
Source: ERG, 2015d; ERG, 20151.
a - The wildlife module applies the same total copper BCFs for both T3 and T4 fish (see Appendix D).
Therefore, the estimated concentrations presented here are identical for both trophic levels.
Table H-15. Total Copper Concentration (mg/kg) in Fish Tissue (Trophic Level 3 &
Trophic Level 4) by Percentile a
Percentile
5th
25th
50th
75th
95th
Max
Scenario
Baseline
5.89E-07
3.19E-05
2.99E-04
0.010
0.540
41.5
Option A
3.65E-07
1.93E-05
2.26E-04
0.008
0.340
28.0
Option B
3.65E-07
1.93E-05
2.26E-04
0.008
0.340
28.0
Option C
0
2.83E-06
5.66E-05
0.002
0.072
28.0
Option D
0
0
4.78E-06
2.56E-04
0.036
28.0
Option E
0
0
4.36E-06
2.26E-04
0.023
28.0
Source: ERG, 2015d; ERG, 20151.
a - The wildlife module applies the same total copper BCFs for both trophic level 3 (T3) and trophic level 4 (T4)
fish (see Appendix D). Therefore, the estimated concentrations presented here are identical for both trophic levels.
H-15
-------
Appendix H—Additional Model Results
Figure H-15. Total Lead Concentration in Trophic Level 3
and Trophic Level 4 Fish Tissue in the Immediate Receiving
Watera
Total Lead Concentration in Fish Tissue (nig/kg)
Source: ERG, 2015d; ERG, 20151.
a - The wildlife module applies the same total lead BCFs for both T3 and T4 fish (see Appendix D). Therefore, the
estimated concentrations presented here are identical for both trophic levels.
Table H-16. Total Lead Concentration (mg/kg) in Fish Tissue (Trophic Level 3 &
Trophic Level 4) by Percentile a
Percentile
5th
25th
50th
75th
95th
Max
Scenario
Baseline
0
2.12E-06
7.01E-05
0.001
0.343
34.8
Option A
0
7.94E-07
4.95E-05
0.001
0.319
23.5
Option B
0
7.94E-07
4.95E-05
0.001
0.319
23.5
Option C
0
0
5.57E-06
1.83E-04
0.047
23.5
Option D
0
0
0
1.03E-05
0.002
23.5
Option E
0
0
0
1.03E-05
0.002
23.5
Source: ERG, 2015d; ERG, 20151.
a - The wildlife module applies the same total lead BCFs for both trophic level 3 (T3) and trophic level 4 (T4) fish
(see Appendix D). Therefore, the estimated concentrations presented here are identical for both trophic levels.
H-16
-------
Appendix H—Additional Model Results
K
Figure H-16. Methylmercury Concentration in Trophic Level
3 Fish Tissue in the Immediate Receiving Water
*4*
•*J
«
£
M
=
T
"53
u
u
rt
O)
•^j
2
•3
u
S
S
HH
(M
O
i.
U
A
S
s
Z
60 -i
50 -
40 -
30 -
20 -
10 -
0 -
Methylmercury Concentration in Fish Tissue (mg/kg)
Source: ERG, 2015d; ERG, 20151.
Table H-17. Methylmercury Concentration (mg/kg) in Fish Tissue (Trophic Level 3) by
Percentile a
Percentile
5th
25th
50th
75th
95th
Max
Scenario
Baseline
2.86E-05
0.001
0.010
0.455
16.826
414.6
Option A
1.63E-05
8.10E-04
0.005
0.314
9.42
183
Option B
9.58E-06
4.69E-04
0.005
0.279
9.42
183
Option C
0
5.71E-05
0.001
0.045
2.66
183
Option D
0
0
1.76E-04
0.006
1.43
183
Option E
0
0
9.28E-05
0.004
0.230
183
Source: ERG, 2015d; ERG, 20151.
a - EPA calculated methylmercury fish tissue concentrations using bioaccumulation factors which do not fully
account for the complexity of biogeochemical reactions that can occur within an aquatic environment and result in
lower bioaccumulation rates of mercury in fish. For example, fish are known to bioaccumulate mercury at lower
rates when exposed to surface waters with high selenium concentrations. In addition, bioaccumulation factors do not
account for a maximum limit a fish could accumulate before a lethal concentration is reached. To address the
outliers in mercury fish tissue concentrations, EPA compared fish tissue concentrations to site-specific data available
in the national fish advisory database and established calibration factors to lower the outlier values. Fish tissue
concentrations presented in the figure and table above represent the uncalibrated values calculated by the wildlife
model. For further details on the methodology for selecting calibration factors see ERG memorandum "EA Model
Validation and Calibration" (DCN SE04454).
H-17
-------
Appendix H—Additional Model Results
Figure H-17. Methylmercury Concentration in Trophic Level
4 Fish Tissue in the Immediate Receiving Water
Methylmercury Concentration in Fish Tissue (mg/kg)
Source: ERG, 2015d; ERG, 20151.
Table H-18. Methylmercury Concentration (mg/kg) in Fish Tissue (Trophic Level 4) by
Percentile a
Percentile
5th
25th
50th
75th
95th
Max
Scenario
Baseline
1.21E-04
0.005
0.044
1.93
71.5
1,762
Option A
6.91E-05
0.003
0.021
1.33
40.1
779
Option B
4.07E-05
0.002
0.020
1.19
40.1
779
Option C
0
2.43E-04
0.006
0.190
11.3
779
Option D
0
0
7.48E-04
0.027
6.07
779
Option E
0
0
3.94E-04
0.017
0.976
779
Source: ERG, 2015d; ERG, 20151.
a - EPA calculated methylmercury fish tissue concentrations using bioaccumulation factors which do not fully
account for the complexity of biogeochemical reactions that can occur within an aquatic environment and result in
lower bioaccumulation rates of mercury in fish. For example, fish are known to bioaccumulate mercury at lower
rates when exposed to surface waters with high selenium concentrations. In addition, bioaccumulation factors do not
account for a maximum limit a fish could accumulate before a lethal concentration is reached. To address the
outliers in mercury fish tissue concentrations, EPA compared fish tissue concentrations to site-specific data available
in the national fish advisory database and established calibration factors to lower the outlier values. Fish tissue
concentrations presented in the figure and table above represent the uncalibrated values calculated by the wildlife
model. For further details on the methodology for selecting calibration factors see ERG memorandum "EA Model
Validation and Calibration" (DCN SE04454).
H-18
-------
Appendix H—Additional Model Results
Figure H-18. Total Nickel Concentration in Trophic Level 3
and Trophic Level 4 Fish Tissue in the Immediate Receiving
Watera
.S*5
V V V
Total Nickel Concentration in Fish Tissue (mg/kg)
Source: ERG, 2015d; ERG, 20151.
a - The wildlife module applies the same total nickel BCFs for both T3 and T4 fish (see Appendix D). Therefore,
the estimated concentrations presented here are identical for both trophic levels.
Table H-19. Total Nickel Concentration (mg/kg) in Fish Tissue (Trophic Level 3 &
Trophic Level 4) by Percentile a
Percentile
5th
25th
50th
75th
95th
Max
Scenario
Baseline
5.71E-08
2.65E-06
2.67E-05
0.001
0.040
1.80
Option A
3.33E-08
1.05E-06
1.44E-05
0.001
0.027
1.80
Option B
2.40E-08
8.88E-07
1.44E-05
0.001
0.027
1.80
Option C
0
1.49E-07
3.66E-06
1.09E-04
0.007
1.80
Option D
0
0
3.34E-07
1.30E-05
0.003
1.80
Option E
0
0
1.98E-07
8.37E-06
0.001
0.493
Source: ERG, 2015d; ERG, 20151.
a - The wildlife module applies the same total nickel BCFs for both trophic level 3 (T3) and trophic level 4 (T4) fish
(see Appendix D). Therefore, the estimated concentrations presented here are identical for both trophic levels.
H-19
-------
Appendix H—Additional Model Results
Figure H-19. Total Selenium Concentration in Trophic Level
3 Fish Tissue in the Immediate Receiving Water
Total Selenium Concentration in Fish Tissue (mg/kg)
Source: ERG, 2015d; ERG, 20151.
Table H-20. Total Selenium Concentration (mg/kg) in Fish Tissue (Trophic Level 3) by
Percentile
Percentile
5th
25th
50th
75th
95th
Max
Scenario
Baseline
4.47E-05
0.001
0.027
0.428
31.6
2,638
Option A
1.88E-05
0.001
0.018
0.374
19.5
2,638
Option B
1.01E-05
2.45E-04
0.003
0.151
8.12
2,638
Option C
0
5.83E-05
0.001
0.047
6.55
2,638
Option D
0
0
1.87E-04
0.013
4.86
2,638
Option E
0
0
1.87E-04
0.013
4.86
2,638
Source: ERG, 2015d; ERG, 20151.
H-20
-------
Appendix H—Additional Model Results
Figure H-20. Total Selenium Concentration in Trophic Level
4 Fish Tissue in the Immediate Receiving Water
K
-------
Appendix H—Additional Model Results
Figure H-21. Total Thallium Concentration in Trophic Level
3 Fish Tissue in the Immediate Receiving Water
V
V
Total Thallium Concentration in Fish Tissue (mg/kg)
Source: ERG, 2015d; ERG, 20151.
Table H-22. Total Thallium Concentration (mg/kg) in Fish Tissue (Trophic Level 3) by
Percentile
Percentile
5th
25th
50th
75th
95th
Max
Scenario
Baseline
3.70E-07
4.46E-05
5.05E-04
0.006
1.20
59.6
Option A
2.02E-07
2.66E-05
4.07E-04
0.005
1.13
59.6
Option B
2.02E-07
2.66E-05
4.07E-04
0.005
1.13
59.6
Option C
0
2.07E-06
7.91E-05
0.001
0.131
59.6
Option D
0
0
6.43E-06
2.00E-04
0.012
20.1
Option E
0
0
6.43E-06
2.00E-04
0.012
20.1
Source: ERG, 2015d; ERG, 20151.
H-22
-------
Appendix H—Additional Model Results
M
=
.2
•3
u
s
S
HH
5t-
o
u
A
s
s
Z
Figure H-22. Total Thallium Concentration in Trophic Level
4 Fish Tissue in the Immediate Receiving Water
Total Thallium Concentration in Fish Tissue (mg/kg)
Source: ERG, 2015d; ERG, 20151.
Table H-23. Total Thallium Concentration (mg/kg) in Fish Tissue (Trophic Level 4) by
Percentile
Percentile
5th
25th
50th
75th
95th
Max
Scenario
Baseline
1.41E-06
1.70E-04
0.002
0.025
4.58
228
Option A
7.74E-07
1.02E-04
0.002
0.020
4.31
228
Option B
7.74E-07
1.02E-04
0.002
0.020
4.31
228
Option C
0
7.90E-06
3.02E-04
0.005
0.500
228
Option D
0
0
2.46E-05
7.63E-04
0.044
76.8
Option E
0
0
2.46E-05
7.63E-04
0.044
76.8
Source: ERG, 2015d; ERG, 20151.
H-23
-------
Appendix H—Additional Model Results
Figure H-23. Total Zinc Concentration in Trophic Level 3
and Trophic Level 4 Fish Tissue in the Immediate Receiving
Watera
Total Zinc Concentration in Fish Tissue (mg/kg)
Source: ERG, 2015d; ERG, 20151.
a - The wildlife module applies the same total zinc BCFs for both T3 and T4 fish (see Appendix D). Therefore, the
estimated concentrations presented here are identical for both trophic levels.
Table H-24. Total Zinc Concentration (mg/kg) in Fish Tissue (Trophic Level 3 &
Trophic Level 4) by Percentile a
Percentile
5th
25th
50th
75th
95th
Max
Scenario
Baseline
7.25E-05
0.002
0.022
0.809
28.4
3,576
Option A
3.20E-05
8.50E-04
0.007
0.687
13.6
3,576
Option B
3.20E-05
8.50E-04
0.007
0.687
13.6
3,576
Option C
0
1.63E-04
0.004
0.144
11.0
3,576
Option D
0
0
5.04E-04
0.027
6.59
3,576
Option E
0
0
2.74E-04
0.012
1.17
501
Source: ERG, 2015d; ERG, 20151.
a - The wildlife module applies the same total zinc BCFs for both trophic level 3 (T3) and trophic level 4 (T4) fish
(see Appendix D). Therefore, the estimated concentrations presented here are identical for both trophic levels.
H-24
-------
Appendix I—Analysis for Alternate Scenario with Clean Power Plan
APPENDIX I
ANALYSIS FOR ALTERNATE SCENARIO WITH CLEAN
POWER PLAN
As discussed in Section 1, the environmental assessment (EA) report presents the
methodology and results of the qualitative and quantitative analyses performed to evaluate
baseline discharges from steam electric power plants and improvements under the final steam
electric effluent limitations guidelines and standards (ELGs). The analyses presented in the
report incorporate some adjustments to current conditions in the industry. The analyses in the
report, however, do not reflect changes in the industry that may occur as a result of the Clean
Power Plan [Clean Air Act Section lll(d)] (CPP). This appendix presents the results of EPA's
quantitative EA analysis that does reflect changes in the industry that may occur as a result of the
CPP. Table 1-1 presents the number of plants included in this alternate scenario analysis
compared to those in the EA report.
Table 1-1. Number of Plants Evaluated in the EA Alternate Scenario Analysis Compared to
the EA Report
Plant Description
Number of Plants
in EA Report
Number of Plants
in Alternate
Scenario Analysis
Number of Plants in Scope of Final Rule
Plants that fall under the applicability of the final rule (40 CFR 423)
1,079
1,079
Cost and Loadings Analysis
Plants for which EPA calculated loadings in the cost and loadings
analyses (see Sections 9 and 10 of the TDD)
Plants that discharge only to surface waters (direct discharger)
Plants that discharge only to a POTW (indirect discharger)
Plants that discharge to surface waters and to a POTW (direct and
indirect discharger)
202
191
7
4
151
145
o
J
o
J
Environmental Assessment
Plants evaluated in the EA (includes all direct dischargers)3
195
148
Acronyms: CFR (Code of Federal Regulations); POTW (publicly owned treatment works); TDD (Technical
Development Document for Effluent Limitations Guidelines and Standards for the Steam Electric Power Generating
Point Source Category (TDD), Document No. EPA-821-R-15-007)
a - For the pollutant loadings and removals presented in this appendix, EPA included indirect dischargers to protect
confidential business information.
The 148 steam electric power plants in the EA alternate scenario analysis discharge to the
172 immediate receiving waters illustrated in Figure 1-1 (some plants discharge to multiple
receiving waters). Table 1-2 presents the count of receiving water types for the 172 immediate
receiving waters.
1-1
-------
Appendix I—Analysis for Alternate Scenario with Clean Power Plan
LEGEND
In EA Scope but Not Modeled
(Great Lakes and Estuaries) (9)
In EA Scope and National-Scale
IRW Model (163)
Figure 1-1. Locations and Counts of Immediate Receiving Waters in EA Scope and Modeling Analyses
1-2
-------
Appendix I—Analysis for Alternate Scenario with Clean Power Plan
Table 1-2. Receiving Water Types for Steam Electric Power Plants
Evaluated in the EA
Receiving Water Type
River/Stream
Lake/Pond/Reservoir
Great Lakes
Estuary
Total Receiving Waters
Number (Percentage) of Immediate
Receiving Waters in the Alternate Scenario
Analysis a
144 (84%)
19(11%)
8 (5%)
1 (<1%)
172 (100%)
Source: ERG, 2015d.
a - The alternate scenario analysis encompasses a total of 172 immediate receiving waters and loadings from 148
steam electric power plants (some of which discharge to multiple receiving waters). The immediate receiving
water (IRW) model, which excludes the Great Lakes and estuaries, encompasses a total of 163 immediate
receiving waters and loadings from 143 steam electric power plants.
EPA evaluated the annual baseline pollutant discharges of the evaluated wastestreams
from steam electric power plants reflecting changes in the industry that may occur as a result of
the CPP. Table 1-3 presents the annual pollutant loadings in pounds and toxic-weighted pound
equivalents (TWPE).1'2 Table 1-4 compares pollutant discharges, as TWPE, from the steam
electric power generating industry to discharges from the other top ten discharging point source
categories, as estimated by EPA for the 2010 Effluent Guidelines Planning Process [U.S. EPA,
2011d].
1 To calculate the TWPE, EPA multiplies a mass loading of a pollutant in pounds per year (Ib/yr) by a pollutant-
specific weighting factor, called the toxic weighting factor (TWF), to derive a "toxic equivalent" loading (Ib-
equivalent/yr), or TWPE. TWFs account for differences in toxicity across pollutants and allow mass loadings of
different pollutants to be compared on the basis of their toxic potential. EPA has developed TWFs for more than
1,000 pollutants based on aquatic life and human health toxicity data, as well as physical/chemical property data
[U.S. EPA, 2012b].
2 Prior to finalizing the rulemaking, EPA revised the datasets used to calculate pollutant loadings for bottom ash
transport water and fly ash transport water. The final industry loadings calculated using these revised datasets are
presented in the TDD. The total industry loadings presented in Appendix I reflect the revised datasets. However,
EPA did not rerun the EA models and other analyses to reflect the final loadings dataset. EA analyses used
previously calculated version of the steam electric power plant pollutant loadings that were derived following the
same methodology. The EA pollutant loadings are included in DCN SE05622. Pollutant-specific loadings and
removals presented in this report are based on the previously calculated version. Appendix J presents the results of a
sensitivity analysis that evaluated the potential for these loadings revisions to affect the EA analyses.
1-3
-------
Appendix I—Analysis for Alternate Scenario with Clean Power Plan
Table 1-3. Annual Baseline Pollutant Discharges from Steam Electric Power Plants
(Evaluated Wastestreams)
Pollutant a
TWFb
Annual Discharge,
pounds (Ibs) c
Annual TWPE,
pound-equivalent
(Ib-eq) c
Metals and Toxic Bioaccumulative Pollutants
Manganese
Cadmium
Boron
Mercury
Selenium
Thallium
Arsenic
Aluminum
Lead
Vanadium
Copper
Iron
Nickel
Zinc
Chromium VI
0.103
22.8
0.00834
110.0
1.12
2.85
3.47
0.0647
2.24
0.280
0.623
0.00560
0.109
0.0469
0.517
6,320,000
10,900
24,600,000
1,180
113,000
43,900
22,200
1,070,000
14,600
55,600
24,000
2,110,000
94,200
145,000
119
649,000
249,000
205,000
129,000
127,000
125,000
77,100
69,400
32,700
15,600
15,000
11,800
10,300
6,800
61.4
Nutrients
Total Nitrogen d
Total Phosphorus
Not applicable
Not applicable
13,100,000
154,000
Not applicable
Not applicable
Other
Chlorides
Total dissolved solids
2.435 X 10-5
Not applicable
722,000,000
3,290,000,000
17,600
Not applicable
Total Pollutants6
1,700,000,000
2,140,000
Sources: Abt, 2008; ERG, 2015a; ERG, 2015b; ERG, 2015f; U.S. EPA, 2012c.
Note: Numbers are rounded to three significant figures.
a - The list of pollutants included in this table is only a subset of pollutants included in the loadings analysis (see
Section 10 of the Technical Development Document (TDD) (EPA-821-R-15-007).
b - TWFs for the following metals apply to all metal compounds: arsenic, chromium, copper, lead, manganese,
mercury, nickel, selenium, thallium, vanadium, and zinc. EPA updated TWFs for arsenic, cadmium, copper,
manganese, mercury, thallium, and vanadium for the steam electric ELGs pollutant loadings analysis.
c - These loadings reflect adjustments to current conditions in the industry to account for publicly announced plans
from the steam electric power generating industry to retire or modify steam electric generating units at specific
power plants; changes to the industry that are expected to occur as a result of the recent Coal Combustion Residuals
(CCR) rulemaking by EPA's Office of Solid Waste and Emergency Response (OSWER); and changes to the
industry that are expected to occur as a result of the CPP. Data source for pollutant specific loadings is DCN
SE05622.
d - Total nitrogen is the sum of total Kjeldahl nitrogen and nitrate/nitrite as N.
e - The totals represent the pollutant loadings in discharges of the evaluated wastestreams - specifically, flue gas
desulfurization (FGD) wastewater, fly ash transport wastewater, bottom ash transport wastewater, and combustion
residual leachate (see Section 10 of the TDD). Loadings presented are based on the final loadings analysis presented
in the TDD. The totals exclude loadings for pollutants not identified as pollutants of concern (POCs) and for
biochemical oxygen demand (BOD), chemical oxygen demand (COD), total organic carbon (TOC), total dissolved
solids (TDS), and total suspended solids (TSS).
1-4
-------
Appendix I—Analysis for Alternate Scenario with Clean Power Plan
Table 1-4. Pollutant Loadings for the Final 2010 Effluent Guidelines Planning Process:
Top 10 Point Source Categories
40CFRPart
423
430
419
421
418
414
440
415
444
410
Point Source Category
Steam Electric Power Generating
Pulp, Paper, And Paperboard
Petroleum Refining
Nonferrous Metals Manufacturing
Fertilizer Manufacturing
Organic Chemicals, Plastics, And Synthetic Fibers
Ore Mining And Dressing
Inorganic Chemicals Manufacturing
Waste Combustors
Textile Mills
Total TWPEa
(Ib-eq/yr)
2, 140,000 b
1,030,000
1,030,000
994,000
826,000
649,000
448,000
299,000
254,000
250,000
Source: U.S. EPA, 201 Id.
Note: Numbers are rounded to three significant figures.
a - Only TWPE totals for the steam electric power generating industry include updates to TWFs for arsenic,
cadmium, copper, manganese, mercury, thallium, and vanadium. The TWPE for all other point source categories is
estimated from discharge monitoring reports (DMRs) and Toxic Release Inventory (TRI) reporting and may include
double-counting of certain pollutant discharges (/'. e., a facility must report a pollutant on both its DMR and its TRI
reporting form).
b -EPA calculated the steam electric power generating industry (40 CFR 423) discharges for the alternate scenario
analysis as total of 2,140,000 TWPE annually (see Section 10 of the TDD).
EPA estimated that the total alternate scenario analysis TWPE from steam electric power
plant wastewater (see Table 1-4) is over two times the amount estimated for the pulp, paper, and
paperboard industry; petroleum refining industry; and nonferrous metals manufacturing (second,
third, and fourth highest ranking), and it is over five times the TWPE for four of the six other
industries identified as the top TWPE dischargers in the Final 2010 Effluent Guidelines Program
Plan [U.S. EPA, 201 Id].3
To provide additional perspective on the magnitude of the pollutant loadings from steam
electric power plants in the alternate scenario analysis, EPA compared loadings for the evaluated
wastestreams to those of an average publicly owned treatment works (POTW). Table 1-5
compares the average steam electric pollutant loadings by wastestream4 to the pollutant loadings
from an average POTW assumed to discharge 3 to 5 MGD. EPA also calculated the equivalent
number of typical POTWs that would discharge loadings equal to the 151 steam electric power
plants5 included in the alternate scenario analysis. Table 1-6 presents total pollutant loadings for
3 Data sources for the other industry discharges include DMRs and TRI reports. EPA recognizes that the DMR and
TRI data have limitations (e.g., only a subset of facilities and a subset of pollutants might be included in the
estimated loadings); however, these are the most readily available data sets that represent discharges across the
United States.
4 EPA calculated the average pollutant loadings for each wastestream by dividing the total pollutant loadings for the
wastestream by the number of steam electric power plants discharging the wastestream [ERG, 2015a].
5 The count of 151 steam electric power plants includes three indirect dischargers that discharge wastewater to a
POTW and do not discharge any of the evaluated wastestreams directly to surface waters. EPA included these
indirect dischargers to protect confidential business information.
1-5
-------
Appendix I—Analysis for Alternate Scenario with Clean Power Plan
the evaluated wastestreams (for the 151 plants) and the number of typical POTWs that would
discharge equivalent loadings.
1-6
-------
Appendix I—Analysis for Alternate Scenario with Clean Power Plan
Table 1-5. Comparison of Average Pollutant Loadings in the Evaluated Wastestreams to an Average POTW
Pollutant
Aluminum
Arsenic
Boron
Cadmium
Chromium VI
Copper
Iron
Lead
Manganese
Mercury
Nickel
Selenium
Thallium
Vanadium
Zinc
Total Nitrogen
Total
Phosphorus
Chlorides
TDS
Average Plant FGD
Wastewater Discharge a'b
Loadings
(Ibs/yr)
1,720
9.68
333,000
91.7
(g)
19.6
1,270
5.82
81,800
6.24
701
1,470
17.0
21.0
1,110
132,000
453
10,100,000
40,800,000
TWPE
(Ib-eq/yr)
111
33.6
2,780
2,090
(g)
12.2
7.10
13.0
8,400
687
76.4
1,640
48.6
5.87
52.3
—
_
246
-
Average Plant Fly Ash
Transport Water
Discharge a'c
Loadings
(Ibs/yr)
9,010
310
19,800
49.2
2.48
282
5,740
157
522
7.76
188
132
134
209
814
25,000
849
84,600
1,870,000
TWPE
(Ib-eq/yr)
583
1,080
166
1,120
1.28
176
32.1
351
53.6
854
20.5
148
384
58.5
38.2
—
_
2.06
-
Average Plant Bottom
Ash Transport Water
Discharge a'd
Loadings
(Ibs/yr)
3,880
61.1
2,060
17.7
0.145
83.0
6,960
58.6
4,340
3.04
275
29.5
276
12.2
227
22,500
657
88,500
2,340,000
TWPE
(Ib-eq/yr)
251
212
17.2
403
0.0750
51.7
39.0
131
446
334
30.0
33.1
789
3.42
10.6
—
_
2.16
-
Average Plant
Combustion Residual
Leachate Discharge a'e
Loadings
(Ibs/yr)
988
12.7
7,700
3.39
(£)
2.55
12,200
(g)
933
0.351
15.4
36.7
0.399
631
69.8
(g)
(g)
142,000
1,200,000
TWPE
(Ib-eq/yr)
63.9
44.2
64.2
77.2
(g)
1.59
68.5
(g)
95.8
38.7
1.68
41.2
1.14
177
3.27
—
_
3.45
~
Average POTW
Discharge a'f
Loadings
(Ibs/yr)
3,590
45.9
1,540
3.54
17.7
154
2,530
48.5
354
3,180
30.6
18.5
9.94
No data
453
123,000
17,800
1,610,000
No data
TWPE
(Ib-eq/yr)
215
159
12.8
80.6
9.02
95.3
14.2
109
36.1
350,000
3.06
20.7
28.2
No data
18.1
—
_
39.3
~
Note: Numbers are rounded to three significant figures.
a - TWPE presented in the table include updates to TWFs for arsenic, cadmium, copper, manganese, mercury, thallium, and vanadium.
b - Average loadings based on 69 plants assumed to discharge FGD wastewater under baseline conditions [ERG, 2015a].
c - Average loadings based on 40 plants assumed to discharge fly ash transport water under baseline conditions [ERG, 2015a].
d-Average loadings based on 135 plants assumed to discharge bottom ash transport water under baseline conditions [ERG, 2015a].
e - Average loadings based on 70 plants assumed to discharge combustion residual leachate under baseline conditions [ERG, 2015a].
f - Average loadings based on average loadings calculated for POTWs discharging 3 to 5 MOD of wastewater (see DCN SE01961).
g - EPA did not calculate loadings for this pollutant and wastestream. See the Costs and Loads Report (DCN SE05831).
1-7
-------
Appendix I—Analysis for Alternate Scenario with Clean Power Plan
Table 1-6. Estimated Number of POTW Equivalents for Total Pollutant Loadings from the
Evaluated Wastestreams
Pollutant
Aluminum
Arsenic
Boron
Cadmium
Chromium VI
Copper
Iron
Lead
Manganese
Mercury
Nickel
Selenium
Thallium
Vanadium
Zinc
Total Nitrogen
Total Phosphorus
Chlorides
TDS
Annual Discharge
pounds (Ibs) a
1,070,000
22,200
24,600,000
10,900
119
24,000
2,110,000
14,600
6,320,000
1,180
94,200
113,000
43,900
55,600
145,000
13,100,000
154,000
722,000,000
3,290,000,000
Equivalent Number of Average
POTWs b
299
484
16,000
3,090
6.72
156
835
301
17,800
0.370
3,080
6,110
4,410
No values for comparison
320
107
8.65
448
No values for comparison
Source: ERG, 2015a.
Note: Numbers are rounded to three significant figures.
a - Annual discharge based on pollutant discharges from 151 steam electric power plants, including three indirect
dischargers.
b - Equivalent number of POTWs is estimated by dividing the total annual pollutant loadings from the 151 steam
electric power plants by the average POTW loadings presented in Table 1-5 for a 4-MGD POTW.
1-8
-------
Appendix I—Analysis for Alternate Scenario with Clean Power Plan
EPA identified the number of surface waters that receive discharges of the evaluated
wastestreams and are located in close proximity to sensitive environments. Table 1-7 summarizes
the number and percentage of immediate receiving waters in the alternate scenario analysis that
are located in sensitive environments.
Table 1-7. Number and Percentage of Immediate
Receiving Waters Identified as Sensitive Environments
Sensitive Environment
Great Lakes watershed
Chesapeake Bay watershed
Impaired water
Surface water impaired for a subset of pollutants associated with the
evaluated wastestreams b
Fish consumption advisory water
Surface water with a fish consumption advisory for a subset of
pollutants associated with the evaluated wastestreams °
Drinking water resource within 5 miles
Number (Percentage) of Immediate
Receiving Waters Identified a
15 (9%)
11(6%)
91 (53%)
45 (26%)
116(67%)
79 (46%)
152 (88%)
a - For the sensitive environment proximity analysis, EPA evaluated 172 immediate receiving waters that receive
discharges of the evaluated wastestreams [ERG, 2015c; ERG, 2015d].
b - Table B-l in Appendix B contains a complete list of the impairment categories identified in EPA's 303(d)-
listed waters and designates the subset of pollutants evaluated.
c - Table B-2 in Appendix B contains a complete list of the types of advisories identified under the sensitive
environment proximity analysis, including pollutants that are not associated with the evaluated wastestreams.
d - The values presented in Section 3.4.5 of the report are based on an analysis of habitat locations that reflect
changes in the industry as a result of the CPP.
Table 1-8 and Table 1-9 present the pollutant loadings to the Great Lakes watershed and
the Chesapeake Bay watershed, respectively, accounting for changes in the industry baseline as a
result of the CPP. Table 1-10 presents the number of immediate receiving waters classified as
impaired in the alternate scenario analysis.
Based on a review of immediate receiving waters that reflect changes in the industry as a
result of the CPP, EPA determined that 116 immediate receiving waters (67 percent) are under
fish consumption advisories; 79 of the immediate receiving waters (46 percent) are under an
advisory for a pollutant associated with the evaluated wastestreams.6 All of these 79 immediate
receiving waters are under a fish consumption advisory for mercury and one of the receiving
waters is also under a fish consumption advisory for lead.
The results of the threatened and endangered species analysis presented in Section 3.4.5
already account for changes in the industry as a result of the CPP. Table I-11 presents the
number of steam electric power plants located within five miles of a drinking water resource and
the number of drinking water resources located within five miles of a steam electric power plant.
6 Table B-2 in Appendix B lists the types of advisories identified under the sensitive environment proximity
analysis, including advisories for pollutants that are not associated with the evaluated wastestreams.
1-9
-------
Appendix I—Analysis for Alternate Scenario with Clean Power Plan
Table 1-8. Pollutant Loadings to the Great Lakes Watershed from the Evaluated
Wastestreams a
Pollutant
Arsenic
Boron
Cadmium
Chromium VI
Copper
Lead
Manganese
Mercury
Nickel
Selenium
Thallium
Zinc
Total Nitrogen
Total Phosphorus
Chlorides
Total Dissolved Solids
Annual Discharge to the Great Lakes
Watershed (Ibs)
1,030
760,000
286
0.548
1,170
869
112,000
37.5
4,310
3,540
4,320
3,860
646,000
10,900
24,100,000
116,000,000
Annual TWPE Discharge to the
Great Lakes Watershed (Ib-eq)
3,590
6,340
6,520
0.283
728
1,950
11,500
4,130
470
3,960
12,300
181
-
-
587
-
Source: ERG, 2015a.
Note: Numbers are rounded to three significant figures.
a - Pollutant loadings based on 14 steam electric power plants discharging to 15 immediate receiving waters in the
Great Lakes watershed.
1-10
-------
Appendix I—Analysis for Alternate Scenario with Clean Power Plan
Table 1-9. Pollutant Loadings to the Chesapeake Bay Watershed from the Evaluated
Wastestreams a
Pollutant
Arsenic
Boron
Cadmium
Chromium VI
Copper
Lead
Manganese
Mercury
Nickel
Selenium
Thallium
Zinc
Total Nitrogen
Total Phosphorus
Chlorides
Total Dissolved Solids
Annual Discharge to the Chesapeake
Bay Watershed (Ibs)
680
1,080,000
199
0
765
571
106,000
24.4
2,880
4,710
2,880
2,630
670,000
7,920
34,200,000
139,000,000
Annual TWPE Discharge to the
Chesapeake Bay Watershed (Ib-eq)
2,360
9,000
4,530
0
477
1,280
10,900
2,690
313
5,290
8,210
123
-
-
832
-
Source: ERG, 2015a.
Note: Numbers are rounded to three significant figures.
a - Pollutant loadings based on seven steam electric power plants discharging to 11 immediate receiving waters in
the Chesapeake Bay watershed.
Table 1-10. Number and Percentage of Immediate Receiving Waters Classified as
Impaired for a Pollutant Associated with the Evaluated Wastestreams
Pollutant Causing Impairment
Mercury
Metals, other than mercury b
Nutrients
TDS, including chlorides
Total for Any Pollutant c
Number (Percentage) of Immediate
Receiving Waters Identified a
21 (12%)
24 (14%)
15 (9%)
2 (1%)
56 (33%)
a - For the impaired waters proximity analysis, EPA evaluated 172 immediate receiving waters that receive
discharges of the evaluated wastestreams [ERG, 2015c; ERG, 2015d].
b - The EPA impaired water database listed 24 immediate receiving waters as impaired based on the "metal, other
than mercury" impairment category. Of those 24 immediate receiving waters, 13 receiving waters are also listed as
impaired for one or more specific metals in the EA analysis (arsenic, cadmium, manganese, selenium, and zinc).
One additional immediate receiving water is impaired for boron (but not included in the "metals, other than
mercury" impairment category).
c - Total does not equal the sum of the immediate receiving waters listed in the table. Some immediate receiving
waters are impaired for multiple pollutants.
1-11
-------
Appendix I—Analysis for Alternate Scenario with Clean Power Plan
Table 1-11. Comparison of Number and Percentage of Steam Electric Power Plants
Located within 5 Miles of a Drinking Water Resource
Type of Drinking Water
Resource
Intakes and reservoirs
Public wells b
Sole-source aquifers
Number of Drinking Water
Resources within 5 Miles of a Steam
Electric Power Plant
87
1,530
5
Number (Percentage) of Steam
Electric Power Plants
Located within 5 Miles of a
Drinking Water Resource a
52 (35%)
116(78%)
5 (3%)
Sources: ERG, 2015c; ERG, 2015d.
a - For the drinking water resource proximity analysis, EPA evaluated 172 immediate receiving waters that receive
discharges of the evaluated wastestreams from 148 steam electric power plants.
b - Counts include two springs and 29 wellheads.
Current impacts from the steam electric power generating industry under the alternate
scenario analysis include water quality impacts (Table 1-12); wildlife impacts (Table 1-13 and
Table 1-14); impacts to benthic organisms (Table 1-15); human health impacts to national-scale
cohorts representing recreational and subsistence fishers (Table 1-16 through Table 1-19); and
human health impacts to cohorts representing recreational and subsistence fishers by race or
Hispanic origin (Table 1-20 and Table 1-21, respectively).
The ecological risk modeling results under the alternate scenario analysis indicate that 16
percent of the lakes, ponds, and reservoirs (3 out of 19) and 13 percent of the rivers and streams
(18 out of 144) that receive discharges of the evaluated wastestreams present an elevated risk of
negative reproductive impacts to fish. For mallards, the counts are slightly higher, with the same
number of lakes, ponds, and reservoirs and 15 percent of the rivers and streams (22 out of 144)
presenting these risks.
Selecting the 90th percentile modeled egg/ovary concentration, meaning there is a 10
percent probability that the egg/ovary concentrations are greater than the selected concentration,
reveals that 19 percent of the immediate receiving waters (31 out of 163) present reproductive
risks to at least 10 percent of the exposed fish population. The results for mallards (20 percent)
are very similar. These counts are considerably higher than the results obtained using the median
modeled egg/ovary concentration, indicating the potential for more widespread ecological
impacts among those waterbodies and food webs that tend to experience higher bioaccumulation
of selenium.
1-12
-------
Appendix I—Analysis for Alternate Scenario with Clean Power Plan
Table 1-12. Number and Percentage of Immediate Receiving Waters with Estimated
Water Concentrations that Exceed the Water Quality Criteria
Evaluation Criterion
Aquatic
Life
Criteria
Human
Health
Criteria
Freshwater Acute NRWQC
Freshwater Chronic NRWQC
Human Health Water and
Organism NRWQC
Human Health Organism Only
NRWQC
Drinking Water MCL
Total Number of Unique Immediate
Receiving Waters °
Number of Immediate Receiving Waters Exceeding a Criterion a
Number of
Rivers and
Streams
7
25
61
44
25
61
Number of
Lakes, Ponds,
and
Reservoirs
0
o
6
12
7
4
12
Total Immediate Receiving
Waters b
Number
Exceeding
7
28
73
51
29
73
Percentage
Exceeding
4%
17%
45%
31%
18%
45%
Sources: ERG, 2015d; ERG, 2015h; ERG, 20151.
Acronyms: NRWQC (National Recommended Water Quality Criteria); MCL (maximum contaminant level).
a - The alternate scenario analysis encompasses a total of 172 immediate receiving waters and loadings from 148
steam electric power plants (some of which discharge to multiple receiving waters). The IRW model, which
excludes the Great Lakes and estuaries, encompasses a total of 163 immediate receiving waters (144 rivers and
streams; 19 lakes, ponds, and reservoirs) and loadings from 143 steam electric power plants.
b - These values are the sum and percentage of rivers, streams, lakes, ponds, and reservoirs impacted.
c - This represents the number of unique immediate receiving waters that exceeded at least one criterion.
Table 1-13. Number and Percentage of Immediate Receiving Waters That Exceed Wildlife
Fish Consumption NEHCs for Minks and Eagles (by Waterbody Type)
Evaluation Criterion
Mink fish consumption NEHC
Eagle fish consumption NEHC
Total Number of Unique
Immediate Receiving Waters °
Number of
Rivers and
Streams
38
48
48
Number of
Lakes, Ponds,
and Reservoirs
8
8
8
Total Receiving Waters a'b
Number
Exceeding
46
56
56
Percentage
Exceeding
28%
34%
34%
Sources: ERG, 2015d; ERG, 2015h; ERG, 2015i
Acronyms: NEHC (No Effect Hazard Concentration).
a - The alternate scenario analysis encompasses a total of 172 immediate receiving waters and loadings from 148
steam electric power plants (some of which discharge to multiple receiving waters). The IRW model, which
excludes the Great Lakes and estuaries, encompasses a total of 163 immediate receiving waters (144 rivers and
streams; 19 lakes, ponds, and reservoirs) and loadings from 143 steam electric power plants.
b - These values are the sum and percentage of rivers, streams, lakes, ponds, and reservoirs impacted.
c - This represents the number of unique immediate receiving waters that exceed a criterion.
1-13
-------
Appendix I—Analysis for Alternate Scenario with Clean Power Plan
Table 1-14. Number and Percentage of Immediate Receiving Waters That Exceed
Wildlife Fish Consumption NEHCs for Minks and Eagles (by Pollutant)
Pollutant
Arsenic
Cadmium
Chromium VI
Copper
Lead
Mercury
Nickel
Selenium
Thallium
Zinc
Mink
Fish
Consumption
NEHC
(ug/g) a
7.65
5.66
17.7 c
41.2
34.6
0.37
12.5
1.13
ID
904
Immediate Receiving
Waters
Number
Exceeding b
0
5
0
0
0
43
0
33
NC
1
Percentage
Exceeding
0%
3%
0%
0%
0%
26%
0%
20%
NC
1%
Eagle
Fish
Consumption
NEHC
(ug/g) a
22.4
14.7
26.6 c
40.5
16.3
0.5
67.1
4
ID
145
Immediate Receiving
Waters
Number
Exceeding b
0
4
0
0
2
55
0
33
NC
4
Percentage
Exceeding
0%
2%
0%
0%
1%
34%
0%
20%
NC
2%
Sources: ERG, 2015d; ERG, 2015h; ERG, 20151.
Acronyms: ID (Insufficient data; no benchmarks were identified in the wildlife analysis for thallium); NC (Not
calculated); NEHC (No Effect Hazard Concentration); ug/g (micrograms/gram).
a - The wildlife fish consumption NEHC represents the maximum pollutant concentration in the fish that will result
in no observable adverse effects in wildlife (i.e., minks or eagles) [USGS, 2008].
b - The alternate scenario analysis encompasses a total of 172 immediate receiving waters and loadings from 148
steam electric power plants (some of which discharge to multiple receiving waters). The IRW model, which
excludes the Great Lakes and estuaries, encompasses a total of 163 immediate receiving waters and loadings from
143 steam electric power plants.
c - An NEHC benchmark is not available for chromium VI; therefore, EPA used the total chromium benchmark.
1-14
-------
Appendix I—Analysis for Alternate Scenario with Clean Power Plan
Table 1-15. Number and Percentage of Immediate Receiving Waters with Sediment
Pollutant Concentrations Exceeding TELs for Sediment Biota
Pollutant
Arsenic
Cadmium
Chromium VI b
Copper
Lead
Mercury
Nickel
Selenium
Thallium
Zinc
Sediment
Benchmark
(mg/kg)
5.90
0.596
37.3
35.7
35
0.174
18.0
ID
ID
123
Total Number of Unique
Immediate Receiving Waters
Number of Immediate Receiving Waters Exceeding TELs for Sediment
Biota
Rivers and
Streams
5
19
0
4
3
33
24
NC
NC
12
33
Lakes, Ponds,
and Reservoirs
0
3
0
1
1
7
3
NC
NC
1
7
Total Immediate Receiving
Waters
Number a
5
22
0
5
4
40
27
NC
NC
13
40
Percent
3%
13%
0%
3%
2%
25%
17%
NC
NC
8%
25%
Sources: ERG, 2015d; ERG, 2015h; ERG, 20151.
Acronyms: ID (Insufficient data; no benchmarks were identified); NC (Not calculated).
a - The alternate scenario analysis encompasses a total of 172 immediate receiving waters and loadings from 148
steam electric power plants (some of which discharge to multiple receiving waters). The IRW model, which
excludes the Great Lakes and estuaries, encompasses a total of 163 immediate receiving waters (144 rivers and
streams; 19 lakes, ponds, and reservoirs) and loadings from 143 steam electric power plants.
b - No benchmark for chromium VI. EPA used the total chromium benchmark, which may underestimate the impact
to wildlife.
1-15
-------
Appendix I—Analysis for Alternate Scenario with Clean Power Plan
Table 1-16. Number and Percentage of Immediate Receiving Waters That Exceed Human
Health Evaluation Criteria (Lifetime Excess Cancer Risk) for Inorganic Arsenic
Receptor
Child
recreational
fisher
Cohort
1 to <2 years
2 to <3 years
3 to <6 years
6 to <11 years
11 to <16 years
16 to <21 years
Adult recreational fisher
Child
subsistence
fisher
1 to <2 years
2 to <3 years
3 to <6 years
6 to <11 years
11 to <16 years
16to<21years
Adult subsistence fisher
Exposure
Duration
(Years)
1
1
3
5
5
5
49
1
1
3
5
5
5
49
Number of Immediate Receiving Waters Where
Lifetime Excess Cancer Risk Exceeds 1-in-a-Million a'b
Number of
Rivers and
Streams
4
4
4
4
4
4
7
4
4
5
6
4
4
19
Number of
Lakes, Ponds,
and Reservoirs
0
0
0
0
0
0
2
0
0
0
0
0
0
2
Total Receiving Waters c
Number
Exceeding
4
4
4
4
4
4
9
4
4
5
6
4
4
21
Percentage
Exceeding
2%
2%
2%
2%
2%
2%
6%
2%
2%
3%
4%
2%
2%
13%
Sources: ERG, 2015d; ERG, 2015h; ERG, 20151.
a - The alternate scenario analysis encompasses a total of 172 immediate receiving waters and loadings from 148
steam electric power plants (some of which discharge to multiple receiving waters). The IRW model, which
excludes the Great Lakes and estuaries, encompasses a total of 163 immediate receiving waters (144 rivers and
streams; 19 lakes, ponds, and reservoirs) and loadings from 143 steam electric power plants.
b - Inorganic arsenic cancer slope factor of 1.5 per milligrams per kilogram (mg/kg) per day.
c - These values are the sum and percentage of rivers, streams, lakes, ponds, and reservoirs impacted.
1-16
-------
Appendix I—Analysis for Alternate Scenario with Clean Power Plan
Table 1-17. Number and Percentage of Immediate Receiving Waters
That Exceed Non-Cancer Oral Reference Dose Values
Receptor
Child
recreational
fisher
Cohort
1 to <2 years
2 to <3 years
3 to <6 years
6 to <11 years
11 to < 16 years
16 to <21 years
Adult recreational fisher
Child
subsistence
fisher
1 to <2 years
2 to <3 years
3 to <6 years
6 to <11 years
11 to < 16 years
16 to <21 years
Adult subsistence fisher
Exposure
Duration
(Years)
1
1
3
5
5
5
49
1
1
3
5
5
5
49
Number of Immediate Receiving Waters where Estimated
Exposure Doses Exceed Non-Cancer Reference Doses a
Number of
Rivers and
Streams
62
62
61
60
57
57
57
76
76
70
67
63
63
65
Number of
Lakes, Ponds,
and Reservoirs
13
13
13
12
10
10
10
14
14
14
14
13
13
13
Total Receiving Waters b
Number
Exceeding
75
75
74
72
67
67
67
90
90
84
81
76
76
78
Percentage
Exceeding
46%
46%
45%
44%
41%
41%
41%
55%
55%
52%
50%
47%
47%
48%
Sources: ERG, 2015d; ERG, 2015h; ERG, 20151.
a - The alternate scenario analysis encompasses a total of 172 immediate receiving waters and loadings from 148
steam electric power plants (some of which discharge to multiple receiving waters). The IRW model, which
excludes the Great Lakes and estuaries, encompasses a total of 163 immediate receiving waters (144 rivers and
streams; 19 lakes, ponds, and reservoirs) and loadings from 143 steam electric power plants.
b - These values are the sum and percentage of rivers, streams, lakes, ponds, and reservoirs impacted.
1-17
-------
Appendix I—Analysis for Alternate Scenario with Clean Power Plan
Table 1-18. Number and Percentage of Immediate Receiving Waters That Exceed Non-
Cancer Oral Reference Dose Values at Baseline by Pollutant
Pollutant
Inorganic arsenic
Cadmium
Chromium VI
Copper
Lead
Mercury (as methylmercury)
Nickel (soluble salts)
Selenium
Thallium (soluble salts)
Zinc
Oral
Reference Dose
(mg/kg/day)
0.0003 b
0.001 b
0.003 b
0.01 c
ID
0.0001 b
0.02 b
0.005 b
0.00001 d
0.3 b
Number of Immediate Receiving Waters where Estimated
Exposure Doses Exceed Non-Cancer Reference Doses a
Number Exceeding
o
J
27
0
4
NC
84
0
41
72
7
Percentage Exceeding
2%
17%
0%
2%
NC
52%
0%
25%
44%
4%
Sources: ERG, 2015d; ERG, 2015h; ERG, 20151.
Acronyms: NC (Not calculated); ID (Insufficient data; there is no current reference dose for lead).
a - The alternate scenario analysis encompasses a total of 172 immediate receiving waters and loadings from 148
steam electric power plants (some of which discharge to multiple receiving waters). The IRW model, which
excludes the Great Lakes and estuaries, encompasses a total of 163 immediate receiving waters and loadings from
143 steam electric power plants.
b-U.S. EPA, 201 Ic.
c-ATSDR, 2010a.
d-U.S. EPA,2010a.
1-18
-------
Appendix I—Analysis for Alternate Scenario with Clean Power Plan
Table 1-19. Comparison of T4 Fish Tissue Concentrations to Fish Advisory Screening
Values
Pollutant
Inorganic arsenic
(noncarcinogen)
Inorganic arsenic
(carcinogen)
Cadmium
Mercury (as
methylmercury)
Selenium
Recreational Fishers
Screening
Value (ppm)a
1.2
0.026
4.0
0.4
20
Number
Exceeding b
0
4
6
58
19
Percentage
Exceeding
0%
2%
4%
36%
12%
Subsistence Fishers
Screening
Value (ppm) a
0.147
0.00327
0.491
0.049
2.457
Number
Exceeding b
o
J
1
18
77
36
Percentage
Exceeding
2%
4%
11%
47%
22%
Sources: ERG, 2015d; ERG, 2015h; ERG, 20151.
Acronyms: ppm (parts per million).
a - Screening values are defined as concentrations of target analytes in fish or shellfish tissue that are of potential
public health concern and that are used as threshold values against which levels of contamination in similar tissue
collected from the ambient environment can be compared. Exceedance of these screening values indicates that more
intensive site-specific monitoring and/or evaluation of human health risk should be conducted [U.S. EPA, 2000a,
Table 5-3].
b - The alternate scenario analysis encompasses a total of 172 immediate receiving waters and loadings from 148
steam electric power plants (some of which discharge to multiple receiving waters). The IRW model, which
excludes the Great Lakes and estuaries, encompasses a total of 163 immediate receiving waters and loadings from
143 steam electric power plants.
1-19
-------
Appendix I—Analysis for Alternate Scenario with Clean Power Plan
Table 1-20. Number and Percentage of Immediate Receiving Waters That Exceed Human
Health Evaluation Criteria (Lifetime Excess Cancer Risk) for Inorganic Arsenic, by Race
or Hispanic Origin
Receptor
Recreational
Subsistence
Race or Hispanic
Origin
Non-Hispanic White
Non-Hispanic Black
Mexican- American
Other Hispanic
Other, including
Multiple Races
Non-Hispanic White
Non-Hispanic Black
Mexican-American
Other Hispanic
Other, including
Multiple Races
Number of Immediate Receiving Waters Where
Lifetime Excess Cancer Risk Exceeds 1-in-a-Million a'b
lto<2
years
o
6
o
6
4
4
4
4
4
4
4
4
2to<3
years
3
3
4
4
4
4
4
4
4
4
3to<6
years
4
4
4
4
4
4
4
4
4
5
6 to <11
years
4
4
4
4
4
5
5
6
5
7
11 to <16
years
4
4
4
4
4
5
5
6
5
7
16 to <21
years
4
4
4
4
4
5
5
6
5
7
Adult
9
11
14
13
15
21
22
23
23
26
Sources: ERG, 2015d; ERG, 2015h; ERG, 20151.
a - The alternate scenario analysis encompasses a total of 172 immediate receiving waters and loadings from 148
steam electric power plants (some of which discharge to multiple receiving waters). The IRW model, which
excludes the Great Lakes and estuaries, encompasses a total of 163 immediate receiving waters and loadings from
143 steam electric power plants.
b - Inorganic arsenic cancer slope factor of 1.5 per milligrams per kilogram (mg/kg) per day.
1-20
-------
Appendix I—Analysis for Alternate Scenario with Clean Power Plan
Table 1-21. Number and Percentage of Immediate Receiving Waters That Exceed Non-Cancer Oral Reference Dose Values, by
Race or Hispanic Origin
Receptor
Recreational,
Child Fisher
Subsistence,
Child Fisher
Recreational,
Adult Fisher
Subsistence,
Adult Fisher
Race or Hispanic Origin
Non-Hispanic White
Non-Hispanic Black
Mexican- American
Other Hispanic
Other, including Multiple Races
Non-Hispanic White
Non-Hispanic Black
Mexican-American
Other Hispanic
Other, including Multiple Races
Non-Hispanic White
Non-Hispanic Black
Mexican- American
Other Hispanic
Other, including Multiple Races
Non-Hispanic White
Non-Hispanic Black
Mexican-American
Other Hispanic
Other, including Multiple Races
Number of Immediate Receiving Waters Where Pollutant Exceeds a Non-Cancer Reference Dose a
Inorganic
Arsenic
0 (0%)
0 (0%)
0 (0%)
0 (0%)
0 (0%)
0 (0%)
0 (0%)
0 (0%)
0 (0%)
0 (0%)
3 (2%)
3 (2%)
3 (2%)
3 (2%)
3 (2%)
3 (2%)
3 (2%)
3 (2%)
3 (2%)
3 (2%)
Cadmium
8 (5%)
9 (6%)
11(7%)
10 (6%)
11(7%)
8 (5%)
9 (6%)
11(7%)
10 (6%)
11(7%)
17(10%)
18(11%)
20 (12%)
20 (12%)
24 (15%)
17(10%)
18(11%)
20 (12%)
20 (12%)
24 (15%)
Copper
3 (2%)
4 (2%)
4 (2%)
4 (2%)
4 (2%)
3 (2%)
4 (2%)
4 (2%)
4 (2%)
4 (2%)
4 (2%)
4 (2%)
4 (2%)
4 (2%)
4 (2%)
4 (2%)
4 (2%)
4 (2%)
4 (2%)
4 (2%)
Mercury b
63 (39%)
64 (39%)
66 (40%)
64 (39%)
68 (42%)
63 (39%)
64 (39%)
66 (40%)
64 (39%)
68 (42%)
74 (45%)
74 (45%)
76 (47%)
76 (47%)
79 (48%)
74 (45%)
74 (45%)
76 (47%)
76 (47%)
79 (48%)
Selenium
26 (16%)
27 (17%)
27 (17%)
27 (17%)
28 (17%)
26 (16%)
27 (17%)
27 (17%)
27 (17%)
28 (17%)
33 (20%)
34 (21%)
36 (22%)
36 (22%)
38 (23%)
33 (20%)
34 (21%)
36 (22%)
36 (22%)
38 (23%)
Thallium c
44 (27%)
45 (28%)
48 (29%)
47 (29%)
48 (29%)
44 (27%)
45 (28%)
48 (29%)
47 (29%)
48 (29%)
58 (36%)
58 (36%)
60 (37%)
60 (37%)
67 (41%)
58 (36%)
58 (36%)
60 (37%)
60 (37%)
67 (41%)
Zinc
4 (2%)
4 (2%)
4 (2%)
4 (2%)
4 (2%)
4 (2%)
4 (2%)
4 (2%)
4 (2%)
4 (2%)
4 (2%)
4 (2%)
5 (3%)
5 (3%)
5 (3%)
4 (2%)
4 (2%)
5 (3%)
5 (3%)
5 (3%)
Sources: ERG, 2015d; ERG, 2015h; ERG, 20151.
a - The alternate scenario analysis encompasses a total of 172 immediate
discharge to multiple receiving waters). The IRW model, which excludes
and loadings from 143 steam electric power plants.
b - Mercury, as methylmercury.
c - Reference dose based on thallium (soluble salts).
receiving waters and loadings from 148 steam electric power plants (some of which
the Great Lakes and estuaries, encompasses a total of 163 immediate receiving waters
1-21
-------
Appendix I—Analysis for Alternate Scenario with Clean Power Plan
EPA evaluated environmental improvements as a result of the regulatory options,
reflecting changes in the industry as a result of the CPP. Table 1-22 and Table 1-23 present
pollutant removals under the regulatory options.
Table 1-22. Steam Electric Power Generating Industry Pollutant Removals for Metals,
Bioaccumulative Pollutants, Nutrients, Chlorides, and TDS Under Regulatory Options
Thallium
Zinc
Nitrogen, total1
Phosphorus, total
Chlorides
Pollutant Removals, Ibs/yr (Percent Reduction)a
107,000
(74%)
1,590,000
(12%)
33,900
(22%)
3,380,000
107,000
(74%)
10,000,00(
(76%)
33,900
(22%)
3,380,000
130,000
(89%)
.200.0'
98,300
(64%)
12,000,000
(2%)
138,000
(95%)
.100.0(
122,000
(79%)
15,300,000
(2%)
.41,000
(97%)
100.000
.22,000
(79%)
15,300,000
(2%)
TDS
684,000,000
(21%)
684,000,000
(21%)
913,000,000
(28%)
999,000,000
(30%)
999,000,000
(30%)
Source: ERG, 2015a.
Acronyms: TDS (Total Dissolved Solids); Ibs/yr (pounds per year).
Note: Pollutant removals are rounded to three significant figures.
a - >0 to 15 percent reduction; 16 to 30 percent reduction; 31 to 45 percent reduction;
>60 percent reduction
b - Total nitrogen loadings are the sum of total Kjeldahl nitrogen and nitrate/nitrite as N loadings.
46 to 60 percent reduction
1-22
-------
Appendix I—Analysis for Alternate Scenario with Clean Power Plan
Table 1-23. Steam Electric Power Generating Industry TWPE Removals for Metals,
Bioaccumulative Pollutants, Nutrients, Chlorides, and TDS Under Regulatory Options
Pollutant Removals, TWPE/year (Percent Reduction)a
Source: ERG, 2015a.
Acronyms: TDS (Total Dissolved Solids); TWPE (Toxic Weighted Pound Equivalents).
Note: Pollutant removals are rounded to three significant figures.
N/A - The TWPE/year is not provided for total nitrogen, total phosphorus, and TDS because EPA has not
established a toxic weighting factor (TWF) for these pollutants.
a - >0 to 15 percent reduction; 16 to 30 percent reduction; 31 to 45 percent reduction;
>60 percent reduction
46 to 60 percent reduction
Table 1-24 presents key environmental improvements as a result of the regulatory options
and reflecting changes in the industry as a result of the CPP. Table 1-25 shows environmental
improvements for benthic organisms. Key environmental improvements based on reduced
discharges of arsenic, mercury, selenium, cadmium, and thallium are included in Table 1-26
through Table 1-30.
1-23
-------
Appendix I—Analysis for Alternate Scenario with Clean Power Plan
Table 1-24. Key Environmental Improvements Under the Regulatory Options
Evaluation Benchmark
Modeled Immediate Receiving
Waters Exceeding Benchmark
Under Baseline Conditionsa
Number
Percentage
Number of Immediate Receiving Waters Exceeding Benchmark
(Percent Reduction from Baseline Conditions) Under the Regulatory
Options b
Option A | Option B | Option C | Option D | Option E
Water Quality Results
Freshwater Acute NRWQC
Freshwater Chronic NRWQC
28
4%
17%
5
(29%)
27
(4%)
Human Health Water and Organism NRWQC
73
45%
Human Health Organism Only NRWQC
Drinking Water MCL
Wildlife Results
Fish Ingestion NEHC for Minks
70
(4%)
51
31%
48
(6%)
29
18%
27
(7%)
46
28%
46
(0%)
Fish Ingestion NEHC for Eagles
Human Health Results—Non-Cancer
Non-Cancer Reference Dose for Child
(recreational)
Non-Cancer Reference Dose for Adult
(recreational)
Non-Cancer Reference Dose for Child
(subsistence)
56
34%
52
(7%)
75
67
90
46%
41%
55%
69
(8%)
60
(10%)
81
(10%)
Non-Cancer Reference Dose for Adult
(subsistence)
78
48%
72
(8%)
1-24
-------
Appendix I—Analysis for Alternate Scenario with Clean Power Plan
Table 1-24. Key Environmental Improvements Under the Regulatory Options
Evaluation Benchmark
Modeled Immediate Receiving
Waters Exceeding Benchmark
Under Baseline Conditionsa
Number
Percentage
Number of Immediate Receiving Waters Exceeding Benchmark
(Percent Reduction from Baseline Conditions) Under the Regulatory
Options b
Option A | Option B | Option C | Option D | Option E
Human Health Results—Cancer
Arsenic Cancer Risk for Child (recreational)
Arsenic Cancer Risk for Adult (recreational)
Arsenic Cancer Risk for Child (subsistence)
2%
6%
4%
3
(25%)
7
(22%)
6
(0%)
(25%)
7
(22%)
6
(0%)
3
(25%)
5
(44%)
5
(17%)
Arsenic Cancer Risk for Adult (subsistence)
21
13%
19
(10%)
19
(10%)
13
(38%)
Source: ERG, 2015d; ERG, 2015h; ERG, 20151.
Acronyms: MCL (maximum contaminant level); NEHC (No Effect Hazard Concentration); NRWQC (National Recommended Water Quality Criteria).
a - The alternate scenario analysis encompasses a total of 172 immediate receiving waters and loadings from 148 steam electric power plants (some of which
discharge to multiple receiving waters). The IRW model, which excludes the Great Lakes and estuaries, encompasses a total of 163 immediate receiving waters
and loadings from 143 steam electric power plants.
b - >0 to 15 percent reduction; 16 to 30 percent reduction; 31 to 45 percent reduction;
46 to 60 percent reduction|>60 percent reduction
1-25
-------
Appendix I—Analysis for Alternate Scenario with Clean Power Plan
Table 1-25. Number of Immediate Receiving Waters with Sediment Pollutant Concentrations Exceeding TELs for Sediment
Biota Under the Regulatory Options
Pollutant
Arsenic
Modeled Immediate
Receiving Waters
Exceeding CSCLs Under
Baseline Conditionsa
Number of Immediate Receiving Waters Exceeding Benchmark
(Percent Reduction from Baseline Conditions) Under the Regulatory Options b
Source: ERG, 2015d; ERG, 2015h; ERG, 20151.
Acronyms: CSCL (Chemical stressor concentration limit); N/A (Not Applicable, no exceedances at baseline conditions to compare option results); NC (Not
calculated; no benchmark for comparison).
a - The alternate scenario analysis encompasses a total of 172 immediate receiving waters and loadings from 148 steam electric power plants (some of which
discharge to multiple receiving waters). The IRW model, which excludes the Great Lakes and estuaries, encompasses a total of 163 immediate receiving waters
and loadings from 143 steam electric power plants.
b - >0 to 15 percent reduction; 16 to 30 percent reduction; 31 to 45 percent reduction;
c - EPA used the total chromium benchmark for this analysis.
46 to 60 percent reductionl>60 percent reduction
1-26
-------
Appendix I—Analysis for Alternate Scenario with Clean Power Plan
Table 1-26. Key Environmental Improvements for Arsenic Under the Regulatory Options
Evaluation Benchmark
Modeled Immediate
Receiving Waters Exceeding
Benchmark Under Baseline
Conditionsa
Number | Percentage
Number of Immediate Receiving Waters Exceeding Benchmark
(Percent Reduction from Baseline Conditions) Under the Regulatory
Options b
Option A
Option B
Option C
Option D
Option E
Water Quality Results
Freshwater Acute NRWQC
Freshwater Chronic NRWQC
Human Health Water and Organism
NRWQC
73
2%
2%
45%
Human Health Organism Only NRWQC
51
31%
Drinking Water MCL
Wildlife Results
6%
Fish Ingestion NEHC for Minks
0%
Fish Ingestion NEHC for Eagles
0%
Human Health Results—Non-Cancer
Non-Cancer Reference Dose for Child
(recreational)
Non-Cancer Reference Dose for Adult
(recreational)
Non-Cancer Reference Dose for Child
(subsistence)
Non-Cancer Reference Dose for Adult
(subsistence)
1%
0%
2%
2%
1-27
-------
Appendix I—Analysis for Alternate Scenario with Clean Power Plan
Table 1-26. Key Environmental Improvements for Arsenic Under the Regulatory Options
Evaluation Benchmark
Modeled Immediate
Receiving Waters Exceeding
Benchmark Under Baseline
Conditionsa
Number
Percentage
Number of Immediate Receiving Waters Exceeding Benchmark
(Percent Reduction from Baseline Conditions) Under the Regulatory
Options b
Option A
Option B
Option C
Option D
Option E
Human Health Results—Cancer
Arsenic Cancer Risk for Child
(recreational)
Arsenic Cancer Risk for Adult
(recreational)
Arsenic Cancer Risk for Child
(subsistence)
2%
6%
4%
3
(25%)
7
(22%)
6
(0%)
(25%)
7
(22%)
6
(0%)
3
(25%)
5
(44%)
5
(17%)
Arsenic Cancer Risk for Adult
(subsistence)
21
13%
19
(10%)
19
(10%)
13
(38%)
Source: ERG, 2015d; ERG, 2015h; ERG, 20151.
Acronyms: MCL (Maximum contaminant level); N/A (Not Applicable, no exceedances at baseline conditions to compare option results); NEHC (No Effect
Hazard Concentration); NRWQC (National Recommended Water Quality Criteria).
a - The alternate scenario analysis encompasses a total of 172 immediate receiving waters and loadings from 148 steam electric power plants (some of which
discharge to multiple receiving waters). The IRW model, which excludes the Great Lakes and estuaries, encompasses a total of 163 immediate receiving waters
and loadings from 143 steam electric power plants.
b - >0 to 15 percent reduction; 16 to 30 percent reduction; 31 to 45 percent reduction;
46 to 60 percent reduction|>60 percent reduction
1-28
-------
Appendix I—Analysis for Alternate Scenario with Clean Power Plan
Table 1-27. Key Environmental Improvements for Mercury Under the Regulatory Options
Evaluation Benchmark
Modeled Immediate Receiving Waters
Exceeding Benchmark Under Baseline
Conditions a
Number
Percentage
Number of Immediate Receiving Waters Exceeding
Benchmark
(Percent Reduction from Baseline Conditions) Under the
Regulatory Options b
Option A
Option B
Option C
Option D
Water Quality Results
Freshwater Acute NRWQC
Freshwater Chronic NRWQC
Human Health Water and Organism NRWQC
Human Health Organism Only NRWQC
Drinking Water MCL
0
1
0%
0
(N/A)
1% I
No benchmark for comparison 1 N/A
No benchmark for comparison
4
2%
N/A
4
(0%)
0
(N/A)
N/A
N/A
4
(0%)
Wildlife Results
rish Ingestion JNEHC lor Minks
Fish Ingestion NEHC for Eagles
43
55
26%
34%
40
(7%)
48
(13%)
39
(9%)
48
(13%)
0
(N/A)
N/A
N/A
4
0
(N/A)
mill
Option E
0
(N/A)
•»
N/A
N/A
N/A 1
N/A
1
•(SftM (50%)
34
(38%)
Human Health Results — Non-Cancer
Non-Cancer Reference Dose for Child
(recreational)
Non-Cancer Reference Dose for Adult
(recreational)
Non-Cancer Reference Dose for Child
(subsistence)
Non-Cancer Reference Dose for Adult
(subsistence)
72
64
84
75
44%
39%
52%
46%
65
(10%)
55
(14%)
74
(12%)
68
(9%)
62
(14%)
54
(16%)
73
(13%)
66
(12%)
46
(36%)
41
(36%)
55
(35%)
49
(35%)
(75%)
\
^^fs^^m
(58%)
(81%)
17
(69%)
HIM
Source: ERG, 2015d; ERG, 2015h; ERG, 20151.
Acronyms: MCL (Maximum contaminant level); N/A (Not Applicable, no exceedances at baseline conditions to compare option results); NEHC (No Effect
Hazard Concentration); NRWQC (National Recommended Water Quality Criteria).
a - The alternate scenario analysis encompasses a total of 172 immediate receiving waters and loadings from 148 steam electric power plants (some of which
discharge to multiple receiving waters). The IRW model, which excludes the Great Lakes and estuaries, encompasses a total of 163 immediate receiving waters
and loadings from 143 steam electric power plants.
b - >0 to 15 percent reduction; 16 to 30 percent reduction; 31 to 45 percent reduction;
46 to 60 percent reduction|>60 percent reduction
1-29
-------
Appendix I—Analysis for Alternate Scenario with Clean Power Plan
Table 1-28. Key Environmental Improvements for Selenium Under the Regulatory Options
Evaluation Benchmark
Modeled Immediate
Receiving Waters Exceeding
Benchmark Under Baseline
Conditionsa
Number
Percentage
Number of Immediate Receiving Waters Exceeding Benchmark
(Percent Reduction from Baseline Conditions) Under the Regulatory
Options b
Option A
Option B
Option C
Option D
Option E
Water Quality Results
Freshwater Acute NRWQC
Freshwater Chronic NRWQC
Human Health Water and Organism
NRWQC
Human Health Organism Only NRWQC
Drinking Water MCL
No benchmark for comparison
27
Wildlife Results
10
17%
5%
1%
6%
N/A
25
(7%)
7
(13%)
1
(0%)
9
(10%)
N/A
Fish Ingestion NEHC for Minks
33
20%
32
(3%)
Fish Ingestion NEHC for Eagles
Negative Reproductive Effects in Fish'
Negative Reproductive Effects in
Mallards °
33
20%
32
(3%)
21
25
13%
15%
17
(19%)
21
(16%)
1-30
-------
Appendix I—Analysis for Alternate Scenario with Clean Power Plan
Table 1-28. Key Environmental Improvements for Selenium Under the Regulatory Options
Evaluation Benchmark
Modeled Immediate
Receiving Waters Exceeding
Benchmark Under Baseline
Conditionsa
Number
Percentage
Number of Immediate Receiving Waters Exceeding Benchmark
(Percent Reduction from Baseline Conditions) Under the Regulatory
Options b
Option A
Option B
Option C
Option D
Option E
Human Health Results—Non-Cancer
Non-Cancer Reference Dose for Child
(recreational)
33
20%
o *•>
32
(3%)
Non-Cancer Reference Dose for Adult
(recreational)
26
16%
23
(12%)
Non-Cancer Reference Dose for Child
(subsistence)
41
25%
39
(5%)
Non-Cancer Reference Dose for Adult
(subsistence)
34
21%
32
(6%)
Source: ERG, 2015d; ERG, 2015h; ERG, 20151.
Acronyms: MCL (Maximum contaminant level); N/A (Not Applicable, no exceedances at baseline conditions to compare option results); NEHC (No Effect
Hazard Concentration); NRWQC (National Recommended Water Quality Criteria).
a - The alternate scenario analysis encompasses a total of 172 immediate receiving waters and loadings from 148 steam electric power plants (some of which
discharge to multiple receiving waters). The IRW model, which excludes the Great Lakes and estuaries, encompasses a total of 163 immediate receiving waters
and loadings from 143 steam electric power plants.
c - These rows indicate the number of immediate receiving waters whose median modeled egg/ovary concentration is predicted to result in reproductive impacts
among at least 10 percent of the exposed fish or mallard population, as determined using the ecological risk model.
1-31
-------
Appendix I—Analysis for Alternate Scenario with Clean Power Plan
Table 1-29. Key Environmental Improvements for Cadmium Under the Regulatory Options
Evaluation Benchmark
Modeled Immediate Receiving
Waters Exceeding Benchmark
Under Baseline Conditionsa
Number
Percentage
Number of Immediate Receiving Waters Exceeding Benchmark
(Percent Reduction from Baseline Conditions) Under the Regulatory
Options b
Option A | Option B | Option C | Option D | Option E
Water Quality Results
Freshwater Acute NRWQC
4%
4
(43%)
4
(43%)
(71%)
9
(61%)
Freshwater Chronic NRWQC
23
14%
18
(22%)
18
(22%)
Human Health Water and Organism NRWQC
No benchmark for comparison
N/A
N/A
Human Health Organism Only NRWQC
No benchmark for comparison
N/A
N/A
Drinking Water MCL
5%
6
(25%)
6
(25%)
Wildlife Results
Fish Ingestion NEHC for Minks
3%
4
(20%)
4
(20%)
Fish Ingestion NEHC for Eagles
2%
3
(25%)
3
(25%)
Human Health Results—Non-Cancer
Non-Cancer Reference Dose for Child
(recreational)
Non-Cancer Reference Dose for Adult
(recreational)
Non-Cancer Reference Dose for Child
(subsistence)
Non-Cancer Reference Dose for Adult
(subsistence)
13
27
18
8%
5%
17%
11%
9
(31%)
6
(25%)
22
(19%)
13
(28%)
9
(31%)
6
(25%)
22
(19%)
13
(28%)
Source: ERG, 2015d; ERG, 2015h; ERG, 20151.
Acronyms: MCL (Maximum contaminant level); N/A (Not Applicable, no exceedances at baseline conditions to compare option results); NEHC (No Effect
Hazard Concentration); NRWQC (National Recommended Water Quality Criteria).
a - The alternate scenario analysis encompasses a total of 172 immediate receiving waters and loadings from 148 steam electric power plants (some of which
discharge to multiple receiving waters). The IRW model, which excludes the Great Lakes and estuaries, encompasses a total of 163 immediate receiving waters
and loadings from 143 steam electric power plants.
b - >0 to 15 percent reduction; 16 to 30 percent reduction; 31 to 45 percent reduction;
46 to 60 percent reduction|>60 percent reduction
1-32
-------
Appendix I—Analysis for Alternate Scenario with Clean Power Plan
Table 1-30. Key Environmental Improvements for Thallium Under the Regulatory Options
Evaluation Benchmark
Modeled Immediate
Receiving Waters Exceeding
Benchmark Under Baseline
Conditionsa
Number
Percentage
Number of Immediate Receiving Waters Exceeding Benchmark
(Percent Reduction from Baseline Conditions) Under the Regulatory Options b
Option A
Option B
Option C
Option D
Option E
Water Quality Results
Freshwater Acute NRWQC
No benchmark for comparison
N/A
N/A
N/A
Freshwater Chronic NRWQC
No benchmark for comparison
N/A
N/A
Human Health Water and Organism
NRWQC
39
24%
36
(8%)
36
(8%)
Human Health Organism Only NRWQC
35
21%
32
(9%)
Drinking Water MCL
27
17%
25
(7%)
Wildlife Results
Fish Ingestion NEHC for Minks
No benchmark for comparison
N/A
N/A
Fish Ingestion NEHC for Eagles
No benchmark for comparison
N/A
N/A
Human Health Results—Non-Cancer
Non-Cancer Reference Dose for Child
(recreational)
Non-Cancer Reference Dose for Adult
(recreational)
55
43
34%
26%
54
(2%)
41
(5%)
54
(2%)
41
(5%)
Non-Cancer Reference Dose for Child
(subsistence)
72
44%
69
(4%)
69
(4%)
Non-Cancer Reference Dose for Adult
(subsistence)
58
36%
58
(0%)
58
(0%)
Source: ERG, 2015d; ERG, 2015h; ERG, 20151.
Acronyms: MCL (Maximum contaminant level); N/A (Not Applicable, no exceedances at baseline conditions to compare option results); NEHC (No Effect
Hazard Concentration); NRWQC (National Recommended Water Quality Criteria).
a - The alternate scenario analysis encompasses a total of 172 immediate receiving waters and loadings from 148 steam electric power plants (some of which
discharge to multiple receiving waters). The IRW model, which excludes the Great Lakes and estuaries, encompasses a total of 163 immediate receiving waters
and loadings from 143 steam electric power plants.
b - >0 to 15 percent reduction; 16 to 30 percent reduction; 31 to 45 percent reduction;
46 to 60 percent reduction|>60 percent reduction
1-33
-------
Appendix I—Analysis for Alternate Scenario with Clean Power Plan
Under the alternate scenario analysis, EPA evaluated environmental improvements to
sensitive waters as a result of the regulatory options and reflecting changes in the industry as a
result of the CPP. EPA determined that 91 of the immediate receiving waters are 303(d)-listed
waterbodies, designated as impaired for one or more pollutants found in the evaluated
wastestreams.7 Table 1-31 presents the pollutant removals to impaired waters under the
regulatory options.
EPA determined that 79 of the 172 immediate receiving waters included in the alternate
scenario analysis are under a fish advisory for mercury. Under the final rule, the number of
immediate receiving waters with fish that exceed EPA's mercury screening value for recreational
fishers (based on steam electric power plant discharges only) will decrease by 59 percent,
thereby reducing the potential threat to human health from consuming contaminated fish.
Under the alternate scenario analysis, EPA identified 14 steam electric power plants that
discharge into the Great Lakes watershed. Table 1-32 presents the pollutant removals to the Great
Lakes watershed under the regulatory options considered by EPA.
Under the alternate scenario analysis, EPA identified seven steam electric power plants
that discharge to the Chesapeake Bay watershed. Under the final rule, EPA estimates the
following pollutant removals to the Chesapeake Bay watershed:
• 603 pounds of arsenic annually (89 percent reduction).
• 167 pounds of cadmium annually (84 percent reduction).
• 555 pounds of lead annually (97 percent reduction).
• 22.8 pounds of mercury annually (93 percent reduction).
• 4,550 pounds of selenium annually (96 percent reduction).
• 2,830 pounds of thallium annually (98 percent reduction).
• 667,000 pounds of total nitrogen annually (>99 percent reduction).
• 6,450 pounds of total phosphorus annually (81 percent reduction).
Finally, EPA evaluated the improvements to downstream receiving waters. Table 1-33
presents the number of river miles impacted by steam electric power plant discharges at baseline
and under the regulatory options for the alternate scenario analysis. The table also presents the
percent reduction in number of impacted river miles.
7 The count of impaired waters excludes the general impairment category "metals (not mercury)" and includes receiving waters
impaired for arsenic, boron, cadmium, chromium, copper, lead, manganese, mercury, selenium, zinc, phosphorous, nutrients,
IDS, or chlorides.
1-34
-------
Appendix I—Analysis for Alternate Scenario with Clean Power Plan
Table 1-31. Pollutant Removals to Impaired Waters by Impairment Type
Impairment
Type/Number of
Receiving Waters b
Pollutant
Baseline
Loadings
(Ibs/yr)
Pollutant Removals (Ibs/yr) to Impaired Waters Under the Regulatory Options (Percent
Reduction)a
Option A
Option B
Option C
Option D
Option E
Mercury-Impaired Receiving Waters
21
Mercury
123
52.3
(42%)
52.6
(43%)
Metals (Not Mercury)-Impaired Receiving Waters
1-35
-------
Appendix I—Analysis for Alternate Scenario with Clean Power Plan
Table 1-31. Pollutant Removals to Impaired Waters by Impairment Type
Impairment
Type/Number of
Receiving Waters b
Pollutant
Baseline
Loadings
(Ibs/yr)
Pollutant Removals (Ibs/yr) to Impaired Waters Under the Regulatory Options (Percent
Reduction)a
Option A
Option B
Option C
Option D
Option E
Nutrient-Impaired Receiving Waters
TDS and Chlorides-Impaired Receiving Waters
Source: ERG, 2015c.
Acronyms: CBI (Confidential business information); Ibs/yr (pounds per year).
Note: Loadings and pollutant removals are rounded to three significant figures.
46 to 60 percent reduction|>60 percent reduction
a - >0 to 15 percent reduction; 16 to 30 percent reduction; 31 to 45 percent reduction; |
b - For the impaired waters proximity analysis, EPA evaluated 172 immediate receiving waters that receive discharges of the evaluated wastestreams.
c - The EPA impaired water database listed 24 immediate receiving waters as impaired based on the "metal, other than mercury" impairment category. Of those
24 immediate receiving waters, 13 receiving waters are also listed as impaired for one or more specific metals (arsenic, cadmium, manganese, selenium, and
zinc). One additional immediate receiving water is impaired for boron (but not included in the "metals, other than mercury" impairment category).
d - Total phosphorous and total nitrogen loadings are presented with this impairment category. Total nitrogen loadings are the sum of total Kjeldahl nitrogen and
nitrate/nitrite as N loadings.
1-36
-------
Appendix I—Analysis for Alternate Scenario with Clean Power Plan
Table 1-32. Pollutant Removals to the Great Lakes Watershed Under the Regulatory
Options
Pollutant
Arsenic
Baseline
Loadings to
the Great
Lakes
Watershed
(Ibs/yr)
1,030
Pollutant Removals (Ibs/yr) to Great Lakes Watershed Under the Regulatory
Options (Percent Reduction)a
Option A
46.7
Option B
Option C
Option D
Option E
46.7 (5%)
509 (49%) 955 (92%) 1,000 (97%)
Boron
Cadmium
Chromium
VI
Copper
Lead
Manganese
Mercury
Nickel
Selenium
Thallium
Zinc
Nitrogen,
totalb
Phosphorus,
total
Chlorides
760,000
1,380
1,380
14,700 (2%) 27,300 (4%)
286
6.03 (2%)
6.03 (2%)
257 (909
27,300 (4%)
(93%)
0.548
0.471 (86%) 0.548 (>99%) 0.548 (>99°.
1,170
869
112,000
37.5
4,310
3,540
4,320
3,860
646,000
10,900
24,100,000
18.8 (2%)
1.20 (3%)
20.6
26.6 (2%)
18.8 (2%)
) (98%)
(99%)
68,300 (61%) 68.300 (61%)
1.48 (4%)
29.3 (1%)
20.9 (1%)
21.8 (1%)
55.5 (1%)
4,210 (98%) 4,260 (99%)
2,890(82%) 3,120(88%) 3,350(95%) 3,350(95%)
2,190(51%) 4,280(99%) 4,280(99%)
1,790 (46%) 3,470 (90%) 3,760 (97%)
21.8 (1%)
2,420
135 (1%)
11,400
55.5 (1%)
299,000
135(1°
474,000
(73%)
643,000
(>99%)
643,000
11,400
9,850 (91%) 9,850 (91%)
693,000 (3%)
1,350,000
(6%)
1,350,000
(6%)
TDS
116,000,000
187,000
187,000
18,400,000
(16%)
36,100,000
(31%)
36,100,000
(31%)
Source: ERG, 2015a; ERG, 2015c.
Acronyms: Ibs/yr (pounds per year); TDS (total dissolved solids).
Note: Loadings and pollutant removals are rounded to three significant figures.
a - >0 to 15 percent reduction; 16 to 30 percent reduction; 31 to 45 percent reduction;
>60 percent reduction
b - Total nitrogen loadings are the sum of total Kjeldahl nitrogen and nitrate/nitrite as N loadings.
46 to 60 percent reduction
1-37
-------
Appendix I—Analysis for Alternate Scenario with Clean Power Plan
Table 1-33. Key Environmental Improvements for Downstream Waters Under the Regulatory Options
Evaluation Criteria
Number of River-
Miles Exceeding
Criteria Under
Baseline Conditions
Number of River-Miles Exceeding Criteria
(Percent Reduction from Baseline Conditions) Under the Regulatory Options a
Option A
Option B
Option C
Option D
Option E
Water Quality Results
Freshwater Acute NRWQC
412
395
(4%)
395
(4%)
388
(6%)
Freshwater Chronic
NRWQC
605
592
(2%)
560
(8%)
Human Health Water and
Organism NRWQC
4,050
3,390
(16%)
3,390
(16%)
Human Health Organism-only
NRWQC
1,500
Drinking Water
MCL
751
1,230
(18%)
1,230
(18%)
725
(3%)
720
(4%)
Wildlife Results
Fish Ingestion NEHC for Minks
1,070
893
(17%)
862
(19%)
Fish Ingestion NEHC for Eagles
1,870
Human Health Results—Non-Cancer
Non-cancer reference dose for
child (recreational)
Non-cancer reference dose for
adult (recreational)
5,800
3,420
1,580
(15%)
1,560
(16%)
4,380
(24%)
2,830
(17%)
4,380
(25%)
2,820
(17%)
Non-cancer reference dose for
child (subsistence)
9,240
Non-cancer reference dose for
adult (subsistence)
6,540
7,790
(16%)
7,760
(16%)
5,050
(23%)
5,050
(23%)
1-38
-------
Appendix I—Analysis for Alternate Scenario with Clean Power Plan
Table 1-33. Key Environmental Improvements for Downstream Waters Under the Regulatory Options
Evaluation Criteria
Number of River-
Miles Exceeding
Criteria Under
Baseline Conditions
Number of River-Miles Exceeding Criteria
(Percent Reduction from Baseline Conditions) Under the Regulatory Options a
Option A
Option B
Option C
Option D
Option E
Human Health Results — Cancer
Cancer risk for child
(recreational)
Cancer risk for adult
(recreational)
Cancer risk for child
(subsistence)
Cancer risk for adult
(subsistence)
227
286
262
414
216
(5%)
263
(8%)
241
(8%)
375
(9%)
216
(5%)
263
(8%)
241
(8%)
375
(9%)
211
(7%)
251
(12%)
239
(9%)
355
(14%)
210
(8%)
246
(14%)
235
(10%)
328
(21%)
207
(9%)
245
(14%)
231
(12%)
304
(26%)
Source: ERG, 2015i; ERG, 20151.
Note: River miles are rounded to three significant figures.
a - >0 to 15 percent reduction; 16 to 30 percent reduction; 31 to 45 percent reduction; |
b - EPA evaluated a total of 72,100 river-miles in the downstream receiving water analysis for toxic, bioaccumulative pollutants. Downstream receiving water
concentrations are calculated until one of three conditions occurs: 1) the discharge travels 300 kilometers (km) downstream; 2) the discharge travels downstream
for a week; or 3) the concentration reaches 1 x 10~9 milligrams per liter (mg/L).
1-39
-------
Appendix J—EA Loadings and TDD Loadings: Sensitivity Analysis
APPENDIX J
EA LOADINGS AND TDD LOADINGS: SENSITIVITY
ANALYSIS
As discussed in Section 3, the analyses presented in the environmental assessment (EA)
report are based on loadings datasets that differ from those that are summarized in the Technical
Development Document for Effluent Limitations Guidelines and Standards for the Steam Electric
Power Generating Point Source Category (TDD)., Document No. EPA-821-R-15-007. This
appendix presents a sensitivity analysis that evaluates the difference between the two pollutant
loadings datasets (the "EA loadings" and the "TDD loadings") and estimates the change in counts
of environmental exceedances that would have resulted from use of the TDD loadings dataset. The
analyses in this section reflect changes in the industry that may occur as a result of the Clean Power
Plan [Clean Air Act Section 11 l(d)] (CPP).
Table J-l quantifies the difference in baseline loadings between the EA loadings and TDD
loadings for each of the ten pollutants that are modeled in the EA analyses.
Impacts to Exceedances across All Pollutants
To estimate the influence that using the TDD loadings would have on the overall counts of
exceedances identified in the EA Report, EPA took the following steps:
1. EPA determined how many immediate receiving waters had exceedances that were
due, in part or in whole, to selenium, thallium, or chromium VI. Because the EA
loadings for these pollutants are equal to (or, in the case of selenium, slightly greater
than) the corresponding TDD loadings, each immediate receiving water in this group
would have had exceedances if EPA had used the TDD loadings.
2. Of the remaining receiving waters with exceedances, EPA determined how many had
exceedances that were due, in part or in whole, to arsenic (whose loadings are 9.4
percent lower using the TDD loadings). By assuming that the difference in loadings
would result in an equal change in the count of exceedances, EPA assumed that use of
the TDD loadings would have resulted in 9.4 percent fewer exceedances among this
group of immediate receiving waters.
3. Of the remaining receiving waters with exceedances, EPA determined how many had
exceedances that were due, in part or in whole, to zinc (whose loadings are 14 percent
lower in the TDD loadings). By assuming that the difference in loadings would result
in an equal change in the count of exceedances, EPA assumed that use of the TDD
loadings would have resulted in 14 percent fewer exceedances among this group of
immediate receiving waters.
4. EPA repeated this process for the remaining modeled pollutants (in order of increasing
change between the EA loadings and TDD loadings) until all immediate receiving
waters with exceedances were taken into account.
Table J-2 presents the results of this analysis, which demonstrates that use of the TDD
loadings in place of the EA loadings would have only minimal effect on the overall counts of
J-l
-------
Appendix J—EA Loadings and TDD Loadings: Sensitivity Analysis
exceedances identified by the immediate receiving water (IRW) model. The benchmark
exceedances that would be most affected by use of the TDD loadings are exceedances of chemical
stressor concentration limits (CSCLs) for sediment biota. Exceedances of this benchmark under
baseline conditions would be approximately 4 percentage points lower (41 percent versus 45
percent) based on use of the TDD loadings instead of the EA loadings. All other benchmark
exceedances change by 2 percentage points or less.
This analysis assumes a linear relationship between a loadings reduction and a change in
exceedances for that pollutant. As discussed below, however, this assumption likely overestimates
the effect of a loadings change on the count of exceedances.
Impacts to Individual Pollutant Exceedances
Table 1-22 in Appendix I presents the industry-wide pollutant-specific removals under the
regulatory options (reflecting changes in the industry as a result of the CPP). Table 1-25 through
Table 1-30 present the pollutant-specific environmental improvements under the regulatory
options. A comparison of the values in these tables indicates that an industry-wide pollutant
loading reduction of x under the regulatory options usually results in a reduction in benchmark
exceedances of less than x. For example, looking at Option A:
• Cadmium: Loadings reduced by 72 percent; exceedances reduced by approximately 19
to 43 percent.
• Mercury: Loadings reduced by 62 percent; exceedances reduced by approximately 7 to
14 percent.
• Arsenic: Loadings reduced by 56 percent; exceedances reduced by approximately 4 to
33 percent.
• Selenium: Loadings reduced by 21 percent; exceedances reduced by approximately 3
to 19 percent.
• Thallium: Loadings reduced by 13 percent; exceedances reduced by approximately 0
to 9 percent.
This suggests that the use of the TDD loadings instead of the EA loadings would have a
less-than-linear effect on the number of exceedances in the EA for each pollutant. Based on this
observation, EPA estimates that use of the TDD loadings would result in the following
approximate effects in the baseline counts of pollutant-specific exceedances identified using the
EA loadings:
• Selenium, thallium, and chromium VI: No decrease in exceedances.
• Arsenic, zinc, mercury: Approximately 10 percent fewer exceedances.
• Cadmium, copper, and nickel: Approximately 20 percent fewer exceedances.
• Lead: Approximately 25 percent fewer exceedances.
J-2
-------
Appendix J—EA Loadings and TDD Loadings: Sensitivity Analysis
Table J-l. Comparison of Annual Baseline Pollutant Discharges from Steam Electric Power Plants (Evaluated
Wastestreams), EA Loadings versus TDD Loadings
Pollutant
Arsenic
Cadmium
Chromium (VI)
Copper
Lead
Mercury
Nickel
Selenium
Thallium
Zinc
Baseline Loadings
EA Version
(Ibs/yr)
22,200
10,900
119
24,000
14,600
1,180
94,200
113,000
43,900
145,000
TDD
Version
(Ibs/yr)
20,100
8,290
119
16,400
7,670
992
61,900
115,000
43,900
124,000
Percent
Change
-9.4%
-24%
0%
-32%
-47%
-16%
-34%
1.4%
0%
-14%
Option D Removals
EA Version
(Ibs/yr)
20,700
10,300
119
23,400
14,200
1,150
92,400
110,000
42,800
138,000
TDD
Version
(Ibs/yr)
18,700
7,660
119
15,800
7,340
961
60,200
111,000
42,800
117,000
Percent
Change
-10%
-26%
0%
-33%
-48%
-16%
-35%
1.4%
0.0%
-15%
Option D Removals
EA Version
(%)
93%
94%
100%
98%
98%
97%
98%
97%
98%
95%
TDD
Version
(%)
93%
92%
100%
97%
96%
97%
97%
97%
98%
95%
Percent
Change
-0.73%
-1.9%
0%
-1.1%
-2.0%
-0.47%
-0.87%
0.032%
-0.020%
-0.79%
Source: ERG, 2015o.
Note: Loadings and pollutant removals are rounded to three significant figures. Percentages are rounded to two significant figures.
J-3
-------
Appendix J—EA Loadings and TDD Loadings: Sensitivity Analysis
Table J-2. Comparison of Modeled Baseline Exceedances (Using EA Loadings) and
Approximated Baseline Exceedances (Using TDD Loadings)
Evaluation Benchmark
Freshwater Acute NRWQC
Freshwater Chronic NRWQC
Human Health Water and Organism
NRWQC
Human Health Organism Only NRWQC
Drinking Water MCL
Fish Ingestion NEHC for Minks
Fish Ingestion NEHC for Eagles
CSCLs for Sediment Biota
Negative Reproductive Effects in Fish
from Selenium b
Negative Reproductive Effects in
Mallards from Selenium b
Non-Cancer Reference Dose for Child
(recreational)
Non-Cancer Reference Dose for Adult
(recreational)
Non-Cancer Reference Dose for Child
(subsistence)
Non-Cancer Reference Dose for Adult
(subsistence)
Baseline Exceedances in
Appendix I
(EA Loadings Version)
Number a
7
28
73
51
29
46
56
40
21
25
75
67
90
78
Percentage
4%
17%
45%
31%
18%
28%
34%
25%
13%
15%
46%
41%
55%
48%
Baseline Approximated
Exceedances
(TDD Loadings Version)
Number a
5.85
27.8
69.8
49.5
29.0
44.0
52.4
34.2
21.0
25.0
72.7
64.2
87.8
75.7
Percentage
4%
17%
43%
30%
18%
27%
32%
21%
13%
15%
45%
39%
54%
46%
Source: ERG, 2015o.
Acronyms: CSCL (Chemical stressor concentration limit); MCL (Maximum contaminant level); NEHC (No Effect
Hazard Concentration); NRWQC (National Recommended Water Quality Criteria).
a - The alternate scenario analysis encompasses a total of 172 immediate receiving waters and loadings from 148
steam electric power plants (some of which discharge to multiple receiving waters). The IRW model, which
excludes the Great Lakes and estuaries, encompasses a total of 163 immediate receiving waters and loadings from
143 steam electric power plants.
b - These rows indicate the number of immediate receiving waters whose median modeled egg/ovary concentration
is predicted to result in reproductive impacts among at least 10 percent of the exposed fish or mallard population, as
determined using the ecological risk model.
J-4
------- |