United States
Environmental Protection
Agency
Technical Development Document for
Proposed Effluent Limitation
Guidelines and Standards for
the Airport Deicing Category
July 2009
-------
Technical Development Document
for Proposed Effluent Limitation Guidelines and Standards for the
Airport Deicing Category
Engineering and Analysis Division
Office of Water / Office of Science and Technology
U.S. Environmental Protection Agency
Washington, D.C. 20460
July 2009
EPA 821-R-09-004
-------
Technical Development Document for Proposed Effluent
Limitation Guidelines and Standards for the Airport Deicing Category
ACKNOWLEDGMENTS AND DISCLAIMER
This document was prepared by Environmental Protection Agency Office of Water staff. The
following contractors provided assistance in performing the analyses supporting the conclusions
detailed in this document.
Eastern Research Group, Inc.
Office of Water staff have reviewed and approved this document for publication. Neither the
United States Government nor any of its employees, contractors, subcontractors, or their
employees makes any warranty, expressed or implied, or assumes 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 document, or represents that its use by such a party would
not infringe on privately owned rights.
July 2009
-------
Technical Development Document for Proposed Effluent
Limitation Guidelines and Standards for the Airport Deicing Category
CONTENTS
Page
1. LEGAL AUTHORITY 1-1
1.1 Clean Water Act (CWA) 1-1
1.1.1 Best Practicable Control Technology Currently Available (BPT) 1-1
1.1.2 Best Conventional Pollutant Control Technology (BCT) 1-1
1.1.3 Best Available Technology Economically Achievable (BAT) 1-1
1.1.4 New Source Performance Standards (NSPS) 1-2
1.1.5 Pretreatment Standards for Existing Sources (PSES) 1-2
1.1.6 Pretreatment Standards for New Sources (PSNS) 1-2
1.2 Effluent Guidelines Plan Requirements 1-3
2. APPLICABILITY AND SUBCATEGORIZATION 2-1
2.1 Applicability of the Regulation 2-1
2.2 Subcategorization 2-1
2.2.1 ADF Usage 2-1
2.2.2 FAA Classification 2-2
2.2.3 Airport Departures 2-2
2.2.4 Land Availability 2-2
2.2.5 Conclusions 2-2
3. SUMMARY AND SCOPE OF REGULATION 3-1
3.1 Summary of Rule 3-1
3.1.1 Best Available Technology Economically Achievable (BAT) 3-1
3.1.2 New Source Performance Standards (NSPS) 3-1
3.1.3 Pretreatment Standards for Existing Sources (PSES) 3-1
3.1.4 Pretreatment Standards for New Sources (PSNS) 3-2
4. DATA COLLECTION ACTIVITIES 4-1
4.1 Preliminary Data Summary 4-1
4.2 Site Visits 4-2
4.3 Industry Questionnaires (Surveys) 4-5
4.3.1 Recipient Selection and Questionnaire Distribution 4-5
4.3.2 Questionnaire Information Collected 4-6
4.3.3 Questionnaire Technical Review, Coding, and Data Entry 4-9
4.4 Field Sampling 4-10
4.5 Permit Review 4-10
4.5.1 Airport Selection for Permit Review 4-12
4.5.2 Obtaining NPDES Permits 4-12
4.5.3 Permit Review Process 4-12
4.6 Industry-Submitted Data 4-16
4.7 Literature Reviews 4-17
4.7.1 Current Deicing Practices and Treatment Technologies 4-18
4.7.2 Current Airport Deicing Runoff Data 4-19
4.7.3 Chemical Information and Environmental Impact Studies 4-19
4.7.4 Current Deicing Runoff Regulations 4-19
4.8 References 4-20
July 2009 ii
-------
Technical Development Document for Proposed Effluent
Limitation Guidelines and Standards for the Airport Deicing Category
CONTENTS (Continued)
Page
5. OVERVIEW OF THE INDUSTRY 5-1
5.1 Industry Statistics 5-1
5.1.1 Airports 5-1
5.1.2 Airlines 5-5
5.2 Industry Practices 5-7
5.2.1 Airfield Deicing Practices 5-7
5.2.2 Aircraft Deicing Practices 5-9
5.2.3 Airport Deicing Stormwater Collection and Control 5-13
5.3 References 5-14
6. DEICING CHEMICAL USE AND DEICING STORMWATER CHARACTERIZATION 6-1
6.1 Deicing Chemical Usage 6-1
6.1.1 Airfield Chemical Use 6-1
6.1.2 Aircraft Chemical Use and Purchasing Patterns 6-2
6.2 Deicing Stormwater Characterization 6-3
6.2.1 Airfield Deicing Chemicals and Associated Deicing Stormwater 6-3
6.2.2 Aircraft Deicing Chemicals and Associated Deicing Stormwater 6-4
6.3 References 6-13
7. POLLUTANTS OF CONCERN 7-1
7.1 Identification of Airport Deicing/Anti-icing Stormwater Pollutants 7-1
7.2 Pollutants of Concern Selection Criteria 7-8
7.3 Identification of Potential Pollutants of Concern 7-9
7.4 Selection of Regulated Pollutants for Proposal 7-10
7.5 References 7-13
8. POLLUTION PREVENTION AND TREATMENT TECHNOLOGIES APPLICABLE TO AIRPORT
DEICING OPERATIONS 8-1
8.1 Deicing Stormwater Collection 8-1
8.1.1 Deicing Stormwater Collection and Conveyance 8-1
8.1.2 Deicing Stormwater Storage 8-5
8.2 Treatment and Recycling 8-7
8.2.1 Oil/Water Separators and DAF 8-8
8.2.2 Biological Treatment 8-8
8.2.3 Membrane Separation 8-12
8.2.4 Filtration 8-13
8.2.5 Mechanical Vapor Recompression 8-13
8.2.6 Distillation 8-13
8.3 Pollution Prevention and Product Substitution Practices 8-14
8.3.1 Infrared Deicing 8-14
8.3.2 Forced Air Deicing 8-16
8.3.3 Aircraft Deicing/Anti-Icing Product Substitution Practices 8-16
8.3.4 Airfield Deicing Product Substitution Practices 8-17
8.3.5 Best Management Practices (BMPs) 8-17
8.4 References 8-20
July 2009 iii
-------
Technical Development Document for Proposed Effluent
Limitation Guidelines and Standards for the Airport Deicing Category
CONTENTS (Continued)
Page
9. CONTROL AND TREATMENT PERFORMANCE 9-1
9.1 Development of Proposed Control and Treatment Scenarios 9-1
9.1.1 GRVs 9-1
9.1.2 GRVs in Combination with Plug and Pump Systems 9-2
9.1.3 Deicing Pads 9-2
9.1.4 AFB Biological Treatment 9-2
9.1.5 Recycle/Recovery Operations 9-3
9.2 Performance of Control and Treatment Scenarios 9-3
9.2.1 GRV Collection (20 Percent Efficiency Scenario) 9-4
9.2.2 Plug and pump Collection (40 Percent Efficiency Scenario) 9-4
9.2.3 Deicing Pad Collection (60 Percent Efficiency Scenario) 9-4
9.2.4 AFB Treatment Performance 9-5
9.3 References 9-5
10. POLLUTANT LOADINGS AND POLLUTANT LOAD REDUCTION ESTIMATES 10-1
10.1 Data Sources 10-1
10.2 Pollutant Loadings Methodology Overview 10-2
10.3 Step 1: Estimate the Amount of Applied ADF and Pavement Deicing
Chemicals 10-2
10.3.1 Pavement Deicing and Pavement Anti-Icing Chemical Usage EstimatelO-3
10.3.2 ADF Usage Estimate 10-8
10.4 Step 2: Calculate the Amount of Pollutant Load Associated with the Applied
Chemicals 10-12
10.4.1 Calculate the Total Mass of Each Pollutant 10-18
10.4.2 Determine the Theoretical Oxygen Demand of Each Chemical 10-18
10.4.3 Determine the COD of Each Chemical 10-18
10.4.4 Determine the BOD5 of Each Chemical 10-19
10.5 Step 3: Estimate the Amount of Baseline Pollutant Load that is Discharged
Directly 10-19
10.5.1 Direct Discharge of Pavement Deicers 10-19
10.5.2 Direct Discharge of ADF 10-19
10.6 Step 4: Estimate Pollutant Loading Removals for Each EPA
Collection/Control Scenario 10-20
10.7 Approach for Calculating Urea-Related Reductions 10-26
10.8 References 10-28
11. TECHNOLOGY COSTS 11-1
11.1 Summary of Costs 11-1
11.2 Development of CostModel Inputs 11-8
11.2.1 Model Site Development 11-8
11.2.2 Airport Operations Data 11-9
11.2.3 Precipitation Data and Site Characteristics 11-9
11.3 General Methodology for Estimating Collection and Treatment Technology
Costs 11-11
July 2009 iv
-------
Technical Development Document for Proposed Effluent
Limitation Guidelines and Standards for the Airport Deicing Category
CONTENTS (Continued)
Page
11.3.1 Overview of the ADF Collection and Treatment Cost Model 11-11
11.3.2 Cost Model Equation Development 11-13
11.3.3 Cost Model Design 11-29
11.3.4 Annualized Costs for ADF Collection and Treatment Alternatives.... 11-38
11.4 Airfield Deicing Costs 11-39
11.4.1 Urea and Potassium Acetate Chemical Costs and Application Rates.. 11-39
11.4.2 Cost Impact of Discontinuing Urea Airfield Deicing 11-40
11.4.3 Urea Monitoring Costs 11-40
11.5 References 11-44
12. NON-WATER QUALITY IMPACTS 12-1
12.1 Energy Requirements 12-1
12.2 Air Emissions 12-3
12.3 Solid Waste Generation 12-9
12.4 References 12-9
13. REGULATORY OPTION SELECTION 13-1
13.1 Regulatory Options Evaluated 13-1
13.2 Option Selection 13-4
13.2.1 BAT 13-5
13.2.2 PSES/PSNS 13-6
13.2.3 NSPS 13-7
13.3 References 13-8
14. LIMITATIONS AND STANDARDS: DATA SELECTION AND CALCULATION 14-1
14.1 Selected Pollutant Parameters 14-1
14.1.1 Chemical Oxygen Demand (COD) 14-1
14.1.2 Ammonia as Nitrogen (Ammonia) 14-1
14.2 Overview of Data Review and Selection 14-2
14.2.1 Data Selection Criteria 14-2
14.2.2 Other Considerations in Data Selection 14-4
14.2.3 Importance of Comments for Data Evaluations for Final Limitations.. 14-5
14.3 Conventions for Modeling Multiple Data Sets from the Same Facility 14-6
14.4 COD: Data Selected as Basis of Proposed Limitations 14-6
14.4.1 Albany Treatment System 14-7
14.5 Ammonia: Data Selected as Basis of Proposed Limitation 14-11
14.6 Limitations: Basis and Calculations 14-12
14.6.1 Definitions 14-13
14.6.2 Percentile Basis of the Limitations 14-13
14.6.3 Estimation Procedures for Percentiles 14-14
14.7 Achievability of Limitations 14-18
14.7.1 Statistical and Engineering Review of Limitations 14-18
14.7.2 Compliance with Limitations 14-21
14.8 References 14-23
July 2009 v
-------
Technical Development Document for Proposed Effluent
Limitation Guidelines and Standards for the Airport Deicing Category
LIST OF TABLES
Page
3-1 BAT Limitations 3-1
3-2 New Source Performance Standards 3-1
3-3 Summary of Proposed Airport Deicing Effluent Limitation Guidelines and
Standards 3-2
4-1 Airports Visited and Reasons for Site Visits 4-3
4-2 Deicing Questionnaire Response Rates 4-5
4-3 Airports Selected for Sampling and the Reason for Their Selection 4-11
4-4 Top 50 Airports in the United States with the Highest ADF Usage, Estimated
Based on Snow and Freezing Precipitation Days and Total Airport Departures 4-13
4-5 Permit Review General Information Table 4-15
4-6 Permit Review Pollutant-Specific Information Table 4-16
4-7 Summary of Costing Data Provided by Industry 4-17
4-8 Summary of Long-Term Analytical Data Provided by Industry 4-17
5-1 Number of U.S. Airports by Airport Type in 2004 5-2
5-2 Deicing Airports by FAA Region for the Three Winter Seasons 5-3
5-3 Airline Classifications 5-6
5-4 National Estimate of Airports Using Deicing Chemicals or Materials 5-8
5-5 Summary of Airfield Pollution Prevention Practices 5-9
5-6 Deicing/Anti-Icing Chemicals Purchased by Airlines that Deiced Their Own
Aircraft 5-11
5-7 Deicing/Anti-Icing Chemicals Purchased for Aircraft Deiced by an FBO 5-11
5-8 Summary of Mechanical and Nonchemical Aircraft Deicing Methods Used by an
Airline 5-12
5-9 Summary of Mechanical and Nonchemical Aircraft Deicing Methods Used By
FBOs 5-12
July 2009 vi
-------
Technical Development Document for Proposed Effluent
Limitation Guidelines and Standards for the Airport Deicing Category
LIST OF TABLES (Continued)
Page
5-10 Summary of Aircraft Pollution Prevention Practices 5-13
5-11 Summary of Airport Collection, Containment, and Conveyance Methods 5-14
6-1 Surveyed U.S. Commercial Airports - Airfield Chemical Usage 6-2
6-2 U.S. Commercial Airports - National Estimate of Airfield Chemical Usage 6-2
6-3 U.S. Commercial Airports - National Estimate of Aircraft Chemical
Usage/Purchase 6-3
6-4 EPA's Analytical Results for Pond 3E Effluent and Pond 6 Effluent, DTW 6-5
6-5 MSP and DTW Grab Sample Data Summary for Collected Deicing Stormwater 6-7
6-6 DEN, PIT, and ALB - 5-Day Average Data Summary for Untreated Deicing
Stormwater 6-9
6-7 RFD - 1-Day Data Summary for Untreated Deicing Stormwater 6-12
7-1 Pollutants Under Consideration as Potential Pollutants of Concern 7-3
7-2 Potential Pollutants of Concern Selected for Proposed Regulation 7-12
8-1 ADF Alternatives 8-16
10-1 Three-Year Average Amount of Pavement Deicing Chemical Usage, in Pounds 10-4
10-2 ADF Estimates Based on Airline Detailed Questionnaire Responses 10-9
10-3 ADF Data Reported in the Airport Questionnaire 10-11
10-4 ADF Annual Usage Estimates for All Airports that Received the Airport
Questionnaire 10-13
10-5 Theoretical Oxygen Demand Calculations for Deicing Chemicals 10-18
10-6 ADF COD Baseline Loads and Loading Reductions for Each Control and
Treatment Scenario, by Airport 10-21
10-7 Baseline COD Load and Potential Load Reduction Associated with the
Discontinued Use of Urea as an Airfield Deicing Chemical 10-27
11-1 Annualized Costs by Surveyed Airport for Each Collection and Control Scenario
Evaluated by EPA 11-3
July 2009 vii
-------
Technical Development Document for Proposed Effluent
Limitation Guidelines and Standards for the Airport Deicing Category
LIST OF TABLES (Continued)
Page
11-2 Airport Questionnaire Data to Estimate Spent ADF Collection and Treatment
Costs 11-10
11-3 Reported Costs for GRVs and Outfall-Normalized Capital and Annual Costs 11-15
11-4 Normalized Capital and Operating Costs for the Plug-and-Pump Collection
System 11-16
11-5 Normalized Installed Capital Costs for Centralized Deicing Pads 11-17
11 -6 Flow-Normalized Installed Capital Costs for the UF/RO ADF Treatment System .... 11 -20
11-7 Expected ADF Percentage Collected and ADF Concentration in Collected
Stormwater 11-20
11-8 Flow-Normalized Annual O&M Costs for the UF/RO Recycle and Recovery
System 11-21
11-9 Flow-Normalized Installed Capital Costs for the MVR/Distillation ADF
Treatment System 11-22
11-10 Load-Normalized Installed Capital Costs for the Anaerobic Fluid Bed Reactors 11 -23
11-11 Load Normalized Annual O&M Costs for the Anaerobic Fluid Bed Reactors 11 -23
11-12 Installed Capital Costs for Various Retention Ponds 11-25
11-13 Estimated Cost for 1,000 Linear Feet of Stormwater Piping 11-27
11-14 Storage Tank Volumes and Installed Capital Cost for Various Airports 11-28
11-15 Airport Deicing Cost Model Equations, Input Variables and Assumptions 11-32
11-16 Average Cost for Urea and Potassium Acetate, 2002-2005 11-39
11-17 Typical Application Rates for Potassium Acetate 11-39
11-18 Application Rates for Sodium Acetate and Urea 11-40
11-19 Cost for Application of Urea and Potassium Acetate, per 1000 Square Feet 11 -40
11-20 Incremental Costs for Airports to Change from Urea to Potassium Acetate for
Airfield Deicing 11-42
11-21 Estimated Annual Costs for Airports to Conduct Effluent Monitoring Program For
Urea Airfield Deicing 11-43
July 2009 viii
-------
Technical Development Document for Proposed Effluent
Limitation Guidelines and Standards for the Airport Deicing Category
LIST OF TABLES (Continued)
Page
12-1 Estimated Electrical Requirements for All AFB Treatment Systems 12-2
12-2 Potential Electricity Generation from AFB Biogas Generation 12-2
12-3 Estimated Incremental Criteria Pollutant Emissions from GRVs 12-4
12-4 BTS to EDMS Matching Based on Aircraft 12-5
12-5 BTS to EDMS Matching Based on LTOs 12-5
12-6 Incremental Criteria Pollutant Emissions Associated with Aircraft Deicing
Operations 12-7
12-7 Comparison of Total Annual LTO Aircraft Emissions to Emissions Resulting in
Deicing Operations 12-8
12-8 Potential Air Emissions from AFB Treatment Systems 12-8
12-9 Estimated Sludge Generation from AFB Bioreactors Treating ADF Contaminated
Stormwater 12-9
13-1 Regulatory Options Evaluated for the Airport Deicing Category 13-3
13-2 Factors Evaluated by EPA in Option Selection 13-5
13-3 Urea Replacement Load Removals and Costs 13-5
14-1 COD: EPA and Airport Self-Monitoring Effluent Data Collected During EPA's
Sampling Episode 14-8
14-2 COD: Dates Excluded Because Units Operated in Series 14-9
14-3 COD: Dates Excluded Because Influent Concentration Reported as Zero 14-10
14-4 COD: Dates Excluded Because of Performance Excursions 14-10
14-5 COD: Summary of Albany Self-Monitoring Effluent Data After Exclusions 14-11
14-6 Ammonia: Data from Albany Airport Used to Develop Limitations 14-12
14-7 COD and Ammonia: Proposed Limitations with Long-Term Averages and
Variability Factors 14-15
14-8 COD: 99th Percentile Estimates from Each Treatment Unit 14-15
14-9 COD: Effect of Number of Daily Values in Weekly Averages 14-16
July 2009 ix
-------
Technical Development Document for Proposed Effluent
Limitation Guidelines and Standards for the Airport Deicing Category
LIST OF TABLES (Continued)
Page
14-10 Ammonia: Consideration of Autocorrelation for Proposed Limitations, Long-
Term Averages, and Variability Factors 14-18
14-11 COD: Dates and Values Greater than Proposed Limitation of 271 mg/L 14-20
14-12 COD: Summary Statistics of Influent Concentrations 14-21
July 2009
-------
Technical Development Document for Proposed Effluent
Limitation Guidelines and Standards for the Airport Deicing Category
LIST OF FIGURES
Page
5-1 Percentage of Airports Deicing Airfield Pavement Each Month 5-4
5-2 Discharge Status of Airports 5-5
8-1 Deicing Pad Equipped with Fixed Deicing Booms at Pittsburgh Airport 8-2
8-2 Mobile Deicer Truck Performing Gate Deicing at Chicago O'Hare Airport 8-3
8-3 Glycol Recovery Vehicle 8-5
8-4 Pond for Deicing Stormwater Storage at Denver International Airport 8-6
8-5 Frac Tanks 8-7
8-6 Aerated Pond Installation atPortland Airport 8-10
8-7 Typical Anaerobic Fluid Bed Treatment System for Treatment of Stormwater
Contaminated by ADF 8-11
8-8 Engineered Wetlands Installation at Buffalo Niagara Airport 8-12
8-9 Infrared Hangar at JFK Airport 8-15
10-1 ADF Factor vs. PG/EG Gallons for U.S. Airports (excluding Alaska) 10-11
10-2 ADF Factor vs. PG/EG Gallons for Alaskan Airports 10-12
11-1 Diagram Showing the Steps for Developing Collection System Costs by the
Airport Deicing Cost Model 11-30
11-2 Diagram Showing the Alternatives for Developing Treatment System Costs by the
Airport Deicing Cost Model 11-31
12-1 Landing and Take Off Cycle 12-6
14-1 Simplified Drawing of Albany Treatment System and Sample Points 14-8
July 2009 xi
-------
Technical Development Document for Proposed Effluent 1. Legal Authority
Limitation Guidelines and Standards for the Airport Deicing Category
1. LEGAL AUTHORITY
Effluent limitation guidelines and standards for the Airport Deicing Category would be
promulgated under the authority of Sections 301, 304, 306, 307, 308, and 501 of the Clean Water
Act, 33 U.S.C. 1311, 1314, 1316, 1317, 1318, and 1361.
1.1 Clean Water Act (CWA)
Congress passed the Federal Water Pollution Control Act Amendments of 1972, also
known as the Clean Water Act (CWA), to "restore and maintain the chemical, physical, and
biological integrity of the Nation's waters" (Section 101(a)). To implement the Act, the United
States Environmental Protection Agency (EPA) is to issue effluent limitation guidelines,
pretreatment standards, and new source performance standards for industrial dischargers. These
guidelines and standards are summarized briefly in the following sections.
1.1.1 Best Practicable Control Technology Currently Available (BPT)
Traditionally, EPA establishes BPT effluent limitations based on the average of the best
performances of facilities within the industry, grouped to reflect various ages, sizes, processes, or
other common characteristics. EPA may promulgate BPT effluent limits for conventional, toxic,
and non-conventional pollutants. In specifying BPT, EPA looks at a number of factors. EPA first
considers the cost of achieving effluent reductions in relation to the effluent reduction benefits.
The Agency also considers the age of the equipment and facilities, the processes employed,
engineering aspects of the control technologies, and required process changes, non-water quality
environmental impacts (including energy requirements), and such other factors as the
Administrator deems appropriate. See CWA sec. 304(b)(l)(B)). If, however, existing
performance is uniformly inadequate, EPA may establish limitations based on higher levels of
control than currently in place in an industrial category when based on an Agency determination
that the technology is available in another category or subcategory, and can be practically
applied.
1.1.2 Best Conventional Pollutant Control Technology (BCT)
The 1977 amendments to the CWA required EPA to identify additional levels of effluent
reduction for conventional pollutants associated with BCT technology for discharges from
existing industrial point sources. In addition to other factors specified in section 304(b)(4)(B),
the CWA requires that EPA establish BCT limitations after consideration of a two part "cost-
reasonableness" test. EPA explained its methodology for the development of BCT limitations in
July 1986 (51 FR 24974). Section 304(a)(4) designates the following as conventional pollutants:
biochemical oxygen demand over 5 days (BODS), total suspended solids (TSS), fecal coliform,
pH, and any additional pollutants defined by the Administrator as conventional. The
Administrator designated oil andgrease as an additional conventional pollutant on July 30, 1979
(44FR44501;40CFR401.16).
1.1.3 Best A vailable Technology Economically A chievable (BA T)
BAT represents the second level of stringency for controlling direct discharge of toxic
and nonconventional pollutants. In general, BAT effluent limitation guidelines represent the best
economically achievable performance of facilities in the industrial subcategory or category. The
July 2009 1-1
-------
Technical Development Document for Proposed Effluent 1. Legal Authority
Limitation Guidelines and Standards for the Airport Deicing Category
factors considered in assessing BAT include the cost of achieving BAT effluent reductions, the
age of equipment and facilities involved, the process employed, potential process changes, and
non-water quality environmental impacts including energy requirements, and such other factors
as the Administrator deems appropriate. The Agency retains considerable discretion in assigning
the weight to be accorded these factors. An additional statutory factor considered in setting BAT
is economic achievability. Generally, EPA determines economic achievability based on total
costs to the industry and the effect of compliance with BAT limitations on overall industry and
subcategory financial conditions. As with BPT, where existing performance is uniformly
inadequate, BAT may reflect a higher level of performance than is currently being achieved
based on technology transferred from a different subcategory or category. BAT may be based
upon process changes or internal controls, even when these technologies are not common
industry practice.
1.1.4 New Source Performance Standards (NSPS)
NSPS reflect effluent reductions that are achievable based on the best available
demonstrated control technology. New facilities have the opportunity to install the best and most
efficient production processes and wastewater treatment technologies. As a result, NSPS should
represent the most stringent controls attainable through the application of the best available
demonstrated control technology for all pollutants (that is, conventional, nonconventional, and
priority pollutants). In establishing NSPS, EPA is directed to take into consideration the cost of
achieving the effluent reduction and any non-water quality environmental impacts and energy
requirements.
1.1.5 Pretreatment Standards for Existing Sources (PSES)
Pretreatment standards apply to discharges of pollutants to publicly owned treatment
works (POTWs) rather than to discharges to waters of the United States. PSES are designed to
prevent the discharge of pollutants that pass through, interfere with, or are otherwise
incompatible with the operation of POTWs. Categorical pretreatment standards are technology-
based and analogous to BAT effluent limitation guidelines. The General Pretreatment
Regulations, which set forth the framework for the implementation of categorical pretreatment
standards, are found at 40 CFR Part 403. These regulations establish pretreatment standards that
apply to all non-domestic dischargers. See 52 FR 1586 (January 14, 1987).
1.1.6 Pretreatment Standards for New Sources (PSNS)
Section 307(c) of the Act calls for EPA to promulgate PSNS at the same time it
promulgates NSPS. Such pretreatment standards must prevent the discharge of any pollutant into
a POTW that may interfere with, pass through, or may otherwise be incompatible with the
POTW. EPA promulgates categorical PSES based principally on BAT technology for existing
sources. EPA promulgates PSNS based on best available demonstrated technology for new
sources. New indirect discharges have the opportunity to incorporate into their facilities the best
available demonstrated technologies. The Agency considers the same factors in promulgating
PSNS as it considers in promulgating NSPS.
July 2009 1-2
-------
Technical Development Document for Proposed Effluent 1. Legal Authority
Limitation Guidelines and Standards for the Airport Deicing Category
1.2 Effluent Guidelines Plan Requirements
Section 304(m) of the CWA, added by the Water Quality Act of 1987, requires EPA to
establish schedules for (1) reviewing and revising existing effluent limitation guidelines and
standards ("effluent guidelines") and (2) promulgating new effluent guidelines. On September 2,
2004, EPA published an Effluent Guidelines Plan (69 FR 53705) that established schedules for
developing new and revised effluent guidelines for several industry categories. One of the
industries for which the Agency established a schedule was the Airport Deicing Category.
July 2009 1-3
-------
Technical Development Document for Proposed Effluent 2. Applicability and Subcategorization
Limitation Guidelines and Standards for the Airport Deicing Category
2. APPLICABILITY AND SUBCATEGORIZATION
The proposed regulations for the Airport Deicing Category include effluent limitation
guidelines and standards for the control of pollutants in wastewater. This document presents the
information and rationale supporting the proposed effluent limitation guidelines and standards.
Section 2.0 highlights the applicability and Subcategorization basis of this proposed regulation.
2.1 Applicability of the Regulation
Airports in the scope of this proposed regulation are defined as Primary Commercial
Airports with greater than 1,000 annual jet departures. The wastewater flows covered by the
proposed rule include all stormwater contaminated with spent aircraft deicing fluid (ADF) as
well as stormwater contaminated with airfield deicing chemicals. EPA has estimated that 218
airports would be covered by this proposed regulation.
2.2 Subcategorization
EPA may divide a point source category into groupings called "subcategories" to provide
a method for addressing variations between products, processes, and other factors, which result
in distinctly different effluent characteristics. EPA used published literature, site visit interviews
and data, industry questionnaire responses, and EPA sampling data for the Subcategorization
analysis. Various Subcategorization criteria were analyzed for trends in discharge flow rates,
pollutant concentrations, and treatability to determine if/where Subcategorization was warranted.
EPA analyzed several factors to determine whether subcategorizing an industrial category and
considering different technology options for those subcategories would be appropriate. For this
analysis, EPA evaluated the characteristics of the industrial category to determine their potential
to provide the Agency with a means to differentiate effluent quantity and quality among
facilities. EPA also evaluated the design, process, and operational characteristics of the different
industry segments to determine technology control options that might be applied to reduce
effluent quantity and improve effluent quality. The factors associated with the Airport Deicing
category are as follows:
• ADF usage;
• Federal Aviation Administration (FAA) classifications;
• Airport departures; and
• Land availability.
2.2.1 ADF Usage
Ethylene and propylene glycols are the main ingredients in aircraft deicing fluid. Through
EPA's research, it became apparent that the volume of glycol required to deice a single aircraft
varied greatly depending on a plethora of variables including weather conditions, aircraft size
and operator training. EPA reviewed industry questionnaire responses and determined that ADF
usage is the best indicator for the volume of deicing operations that occur at an airport. ADF
usage can range from zero to hundreds of thousands of gallons per year at airports across the
United States.
July 2009 2-1
-------
Technical Development Document for Proposed Effluent 2. Applicability and Subcategorization
Limitation Guidelines and Standards for the Airport Deicing Category
2.2.2 FAA Classification
The Airport and Airway Improvement Act (AAIA) and the FAA classify airports by size
based on the volume of commercial traffic. (Non-commercial airports, commonly known as
"General Aviation" airports, are not specifically defined by the AAIA.) EPA utilized this
classification system to organize its data collection and analysis of the aviation industry. The
AAIA defines airports by categories of airport activities, including Commercial Service (Primary
and Non-Primary), Cargo Service, and Reliever. Commercial Service Airports are publicly
owned airports that have at least 2,500 passenger boardings each calendar year and received
scheduled passenger service. The definition also includes passengers who continue on an aircraft
in international flight that stops at an airport in any of the 50 states for a non-traffic purpose,
such as refueling or aircraft maintenance rather than passenger activity. Primary Commercial
Service airports have more than 10,000 passenger boardings each year. Primary airports are
further subdivided into Large Hub, Medium Hub, Small Hub and Non-Hub classifications, based
on the percentage of total passenger boardings within the U.S. in the most current calendar year
ending before the start of the current fiscal year.
Early on in the regulatory process, EPA made the assumption that the majority of the
deicing in the U.S. would occur at Primary Commercial airports and particularly those with jet
departures. General aviation aircraft, as well as smaller commercial non-jet aircraft are expected
to suspend flights during inclement weather, whereas commercial aircraft with scheduled service
are much more likely to deice in order to meet customer demands.
2.2.3 Airport Departures
While ADF usage and FAA classification are important criteria for determining where
most of the aircraft deicing would occur, EPA considered an additional criterion in order to
reflect the economic achievability of potential regulatory requirements Within the Primary
Commercial airport stratum, the size of an airport, based on departures, was used by EPA to
classify affordability of this proposed regulation with respect to a specific airport. This size
threshold includes a minimum number of annual jet departures as well as the minimum number
of total annual departures.
2.2.4 Land Availability
EPA is aware that airports across the country have different amounts of land that may be
available for facility modifications, such as for installation of environmental controls. EPA
collected some basic information from airports on their current configurations. However, neither
the aviation industry nor the FAA has developed a standard definition of land availability, and
EPA did not formulate a definition but has requested public comment on this topic in the
proposed rule.
2.2.5 Conclusions
Establishing formal subcategories is not necessary for the Airport Deicing category
because the proposed rule is structured to address the relevant factors (i.e., amount of ADF
applied and number of departures) and establish a set of requirements that encompasses the range
of situations that an airport may encounter during deicing operations. Both the aircraft deicing
and pavement deicing requirements use airport size thresholds, which exclude smaller facilities.
July 2009 2-2
-------
Technical Development Document for Proposed Effluent 2. Applicability and Subcategorization
Limitation Guidelines and Standards for the Airport Deicing Category
The use of a performance standard, as compared to a technology specification, provides
flexibility for airports in meeting the requirements.
July 2009 2-3
-------
Technical Development Document for Proposed Effluent
Limitation Guidelines and Standards for the Airport Deicing Category
3. Summary and Scope of Regulation
3.
3.1
SUMMARY AND SCOPE OF REGULATION
Summary of Rule
The components of the proposed rules to the Airport Deicing Category are described in
the following subsections.
3.1.1
Best Available Technology Economically Achievable (BAT)
EPA is proposing BAT for the Airport Deicing Category to control priority and
nonconventional pollutants in deicing stormwaters from direct dischargers through the regulation
of Chemical Oxygen Demand (COD) and ammonia as nitrogen. Table 3-1 presents the BAT
Limitations proposed.
Table 3-1. BAT Limitations
Wastestream
Aircraft Deicing
Airfield Pavement Deicing
Pollutant or Pollutant
Property
COD
Ammonia as Nitrogen
Daily Maximum
271 mg/L
14.7 mg/L
Weekly Average
154 mg/L
Table 3-3 presents the technology basis for the effluent limitation guidelines.
3.1.2 New Source Performance Standards (NSPS)
EPA is proposing NSPS for the Airport Deicing Category to control priority,
nonconventional, and conventional pollutants in wastewater from new direct dischargers through
the regulation of COD and ammonia as nitrogen. Table 3-2 presents the NSPS proposed.
Table 3-2. New Source Performance Standards
Wastestream
Aircraft Deicing
Airfield Pavement Deicing
Pollutant or Pollutant Property
COD
Ammonia as Nitrogen
Daily Maximum
271 mg/L
14.7 mg/L
Weekly Average
154 mg/L
Table 3-3 summarizes the technology basis for NSPS.
3.1.3
Pretreatment Standards for Existing Sources (PSES)
EPA is not proposing PSES for the Airport Deicing Category, based on information
collected by the Agency. POTWs across the country are accepting wastewater associated with
deicing runoff and EPA is not aware of any specific pass-through, interference or sludge
contamination concerns for these POTWs. EPA is aware that high concentration and/or volume
"slug" discharges of deicing stormwater can create POTW upset, and many of the airports that
discharge to POTWs have airport-specific requirements on allowable chemical oxygen demand
(COD) or BOD discharge loading per day. They may also have requirements for discharging at
various concentration levels over time.
July 2009
3-1
-------
Technical Development Document for Proposed Effluent
Limitation Guidelines and Standards for the Airport Deicing Category
3. Summary and Scope of Regulation
3.1.4
Pretreatment Standards for New Sources (PSNS)
As with existing sources, EPA is not proposing PSNS for the Airport Deicing Category,
based on information collected during the data collection phase of this project. EPA does not
expect any pass-through concerns with the discharge of deicing stormwaters to a POTW.
Table 3-3. Summary of Proposed Airport Deicing Effluent Limitation Guidelines and
Standards
Regulatory
Level
BAT
NSPS
Technology Basis
1.60%or20%ADF
capture
2. Biological treatment
3. Pavement deicer product
substitution
1. 60% ADF Capture
2. Biological treatment
3. Pavement deicer product
substitution
Technical Components
Airports > 1,000 Annual Jet
Departures and >10,000 Annual
Departures
1. Capture 60% of available ADF (for
airports having >460,000 gal. ADF
usage) or capture 20% (for airports
<460,000 gal. ADF usage)
2. Treat wastewater to meet effluent
limit for COD
3. Certify use of non-urea-based
pavement deicers
or
Meet effluent limit for ammonia
1. Capture 60% of available ADF
2. Treat wastewater to meet effluent
limit for COD
3. Certify use of non-urea-based
pavement deicers
or
Meet effluent limit for ammonia
Airports > 1,000 Annual Jet
Departures and <10,000
Annual Departures
1. Certify use of non-urea-
based pavement deicers
or
Meet effluent limit for
ammonia
1. Certify use of non-urea-
based pavement deicers
or
Meet effluent limit for
ammonia
Note: All references to ADF are for normalized ADF, which is ADF less any water added by the manufacturer or
customer before ADF application.
July 2009
3-2
-------
Technical Development Document for Proposed Effluent 4. Data Collection Activities
Limitation Guidelines and Standards for the Airport Deicing Category
4. DATA COLLECTION ACTIVITIES
To characterize airport deicing operations and to develop the proposed effluent limitation
guidelines and standards, EPA collected and evaluated technical and economic data from a
variety of sources. This section details the following data sources used for the Airport Deicing
Category rulemaking effort:
Section 4.1 - Preliminary Data Summary
Section 4.2 - Site Visits
Section 4.3 - Industry Questionnaires (Surveys)
Section 4.4 - Field Sampling
Section 4.5 - Permit Review
Section 4.6 - Industry-Supplied Data
Section 4.7 - Literature Reviews
For the development of the 2004 Effluent Guidelines Plan, the Agency reviewed
available information on aircraft deicing/anti-icing fluid (ADF) use at airports. EPA found that
the potential existed for airports to discharge nontrivial amounts of nonconventional and toxic
pollutants and that ADF is not properly recaptured and reused or properly treated before
discharge. However, due to the variety of ADFs in use and the limited information on the
chemical composition of these ADFs, EPA was unable to estimate the toxic-weighted pollutant
discharges associated with these discharges and the potential effluent reductions that could be
achieved through application of more stringent control mechanisms. Therefore, the Agency
initiated the data collection activities discussed in this section.
4.1 Preliminary Data Summary
EPA's initial source of wastewater discharge information for the aviation industry was
the Preliminary Data Summary (PDS): Airport Deicing Operations, which was published in
August 2000 (USEPA, 2000). This study focused on approximately 200 U.S. airports with
potentially significant deicing/anti-icing operations. For the study, EPA collected information
from industry questionnaires, engineering site visits, wastewater sampling activities, meetings
with industry and regulatory agencies, and technical and scientific literature. See Section 3.0 of
the PDS for detailed information on the study's data collection activities.
The questionnaires that were reviewed included the 1993 Screener Questionnaire for the
Transportation Equipment Cleaning Industry, and a set of questionnaires distributed during the
study to major and regional airports and airlines, technology vendors, and POTWs.
From September 1997 through March 1999, the Agency conducted 16 airport site visits
and 6 sampling episodes to collect information about deicing processes, deicing equipment, and
deicing wastewater generation, collection, handling, and treatment technologies.
EPA met with the Federal Aviation Administration (FAA), deicing fluid manufacturers
and formulators, airlines, industry associations, technology vendors, and other interested parties
to discuss environmental and operational issues related to aircraft deicing and anti-icing
operations.
July 2009 4-1
-------
Technical Development Document for Proposed Effluent 4. Data Collection Activities
Limitation Guidelines and Standards for the Airport Deicing Category
Literature searches provided information on the toxicity, industry usage, and mitigation
techniques for ADFs. The literature also covered topics such as alternative fluid types, pollution
prevention practices, economic and financial data, and environmental impacts.
4.2 Site Visits
Between December 2004 and November 2005, EPA conducted 20 airport site visits to
collect current information about aircraft and airfield deicing practices, deicing equipment,
deicing stormwater generation, collection, handling, and control. During these site visits, EPA
also evaluated potential sampling locations for the sampling program described in Section 4.4.
EPA used information collected from the PDS, updated airport literature searches, and
other Agency-supplied data to assess potential airports for site visits. EPA also solicited
recommendations from industry trade associations, including the Air Transport Association
(ATA), the American Association of Airline Executives (AAAE) and Airports Council
International-North America (ACI-NA). EPA considered the following criteria in evaluating
which airports to visit:
• Hub size and location of the airport;
• Aircraft deicing fluid handling practices;
• Deicing stormwater collection and control practices; and
• ADF-contaminated stormwater discharge practices.
In general, EPA visited medium and large hub airports operating in northern climates that
conduct aircraft and airfield deicing operations each winter. EPA also visited some small hub
airports to evaluate potential issues related to an airport's size. The Agency visited airports that
use a variety of deicing practices (such as gate deicing, centralized deicing pads, deicing trucks
and stationary booms, infrared deicing hangars) and various deicing stormwater collection and
control technologies (such as dedicated deicing stormwater collection systems, stormwater
treatment through biological systems, and glycol recovery systems). Table 4-1 lists the 20
airports visited, the visit date, and EPA's criteria for visiting each airport.
During the site visits, EPA collected the following information:
• General airport and deicing operations information, including size and age of the
airport, permit status, information on the entities that perform deicing operations
(both aircraft and airfield), and current airline tenant information;
• Description of the deicing/anti-icing operations conducted at the airport, including
the types of equipment used, locations of deicing operations, and information on
any pollution prevention or "state-of-the-art" systems in use at the airport that
improved their deicing operations;
• Deicing chemicals used, including ADF type (e.g., Type I-IV) and any chemical
usage information available;
• Description of the deicing stormwater collection and control systems used at the
airport, including any glycol recovery or stormwater treatment systems and their
effectiveness and any available cost information for these systems; and
• Airport monitoring and discharge of deicing stormwater, including pollutants
monitored and frequency of monitoring.
July 2009 4-2
-------
Technical Development Document for Proposed Effluent
Limitation Guidelines and Standards for the Airport Deicing Category
4. Data Collection Activities
Table 4-1. Airports Visited and Reasons for Site Visits
Airport Name
Washington Dulles
International
Baltimore-Washington
International
Chicago O'Hare
International
General Mitchell
International
(Milwaukee)
Detroit Metropolitan
Wayne County
Ronald Reagan
Washington National
Syracuse Hancock
International
Albany International
Pittsburgh International
Cincinnati/Northern
Kentucky International
Richmond International
Minneapolis-St. Paul
International/World-
Chamberlain
James M Cox Dayton
International
Portland International
(Oregon)
Seattle-Tacoma
International
LaGuardia (New York)
John F. Kennedy
International (New
York)
Airport
Code
IAD
BWI
ORD
MKE
DTW
DCA
SYR
ALB
PIT
CVG
RIC
MSP
DAY
PDX
SEA
LGA
JFK
Date of Visit
12/1/2004
12/15/2004
1/26/2005
1/27/2005
1/28/2005
2/1/2005
2/9/2005
2/10/2005
2/10/2005
2/11/2005
2/16/2005
2/18/2005
2/25/2006
7/26/2005
7/27/2006
10/11/2005
10/11/2005
Criteria for Site Visit
Local large hub airport, ADF-contaminated stormwater
collection and glycol recovery, indirect discharger
Local large hub airport, deicing pads, ADF-
contaminated stormwater collection, indirect
discharger
Large hub airport, ADF-contaminated stormwater
collection with indirect discharge, upgrades to system
since the PDS site visit
Medium hub airport, ADF-contaminated stormwater
collection and indirect discharge, extensive monitoring
data in collaboration with U.S. Geological Survey
(USGS)
Large hub airport, deicing pads, ADF-contaminated
stormwater collection with glycol recovery, both direct
and indirect discharger
Local large hub airport, changes in ADF practices
since PDS site visit, ADF-contaminated stormwater
collection
Small hub airport, deicing pads, aerated stormwater
lagoons, indirect discharger
Small hub airport, ADF-contaminated stormwater
collection with anaerobic and aerobic treatment, direct
discharger, upgrades to system since the PDS site visit
Large hub airport, deicing pads, glycol recovery and
treatment (ultra filtration and reverse osmosis) of ADF-
contaminated stormwater, direct discharger
Large hub airport, variety of ADF-contaminated
stormwater-related activities, recommended by FAA as
a site visit location, on-site aerobic treatment, direct
and indirect discharger
Local small hub airport
Large hub airport, ADF collection with glycol
recovery, direct and indirect discharger
Small hub airport, centralized deicing with ADF-
contaminated stormwater collection
Medium hub northwestern airport, indirectly
discharges high strength deicing stormwater, sends low
strength deicing stormwater to detention pond and then
to direct discharge
Large hub northwestern airport, industrial stormwater
treatment on site
Large hub airport, direct discharger, part of New York
City area visits
Large hub airport with a high percentage of
international flights, direct discharger, part of New
York City area visits, future plans for infrared deicing
July 2009
4-3
-------
Technical Development Document for Proposed Effluent
Limitation Guidelines and Standards for the Airport Deicing Category
4. Data Collection Activities
Table 4-1 (Continued)
Airport Name
Newark Liberty
International
Salt Lake City
International
Denver International
Airport
Code
EWR
SLC
DIA
Date of Visit
10/12/2005
11/8/2005
11/9/2005
Criteria for Site Visit
Large hub airport, infrared technology since 1999
Large hub western airport, ADF -contaminated
stormwater collection with glycol recovery
Large hub western airport, deicing pads, ADF-
contaminated stormwater collection with glycol
recovery
July 2009
4-4
-------
Technical Development Document for Proposed Effluent
Limitation Guidelines and Standards for the Airport Deicing Category
4. Data Collection Activities
This information is documented in the Site Visit Report (SVR) for each airport visited. The
SVRs are located in the administrative record for this rulemaking.
4.3
Industry Questionnaires (Surveys)
EPA distributed three questionnaires to directly support the Airport Deicing rulemaking.
Section 4.3.1 discusses the recipient selection process, distribution, and mail-out results for the
three airport deicing questionnaires. Section 4.3.2 discusses the organization of and type of
technical information requested in each questionnaire.
4.3.1
Recipient Selection and Questionnaire Distribution
EPA distributed an airline screener questionnaire followed by a detailed airline
questionnaire, and an airport questionnaire. The overall focus of the questionnaires was on
airports and airlines that perform deicing and anti-icing on aircraft and/or airfield pavement.
Airports were selected for the detailed airport questionnaire by airport type (i.e., large hub,
medium hub, small hub, and non-hub), days and amount of snow or freezing precipitation, and
the number of departures. EPA performed a census design for large and medium hub airports and
a stratified random sample design for small and non-hub airports (see the statistical support
memorandum DCN ADO 1208).
FAA classifies large commercial airports into size categories of "hubs," based on the
number of annual enplanements that occur at the airport. Enplanements represent the number of
passengers boarding the plane for departure. Large hubs are airports that represent more than 1
percent of total U.S. passenger enplanements. Medium hubs are defined as airports that account
for more than 0.25 percent but less than 1 percent of total passenger enplanements. Small hubs
enplane 0.05 to 0.25 percent of the total passenger enplanements. Airports with less than 0.05
percent of the total passenger enplanements but more than 10,000 annual enplanements are
considered non-hub primary airports.
EPA selected airlines for the airline screener questionnaire by the number of departures at
the selected airports and then selected those airlines for the airline detailed questionnaire if
deicing was performed on their aircraft at these airports. EPA used weighting factors to scale up
the airport survey data to represent national estimates. Table 4-2 summarizes the response rates
for the questionnaires.
Table 4-2. Deicing Questionnaire Response Rates
Questionnaire Type
Airport Deicing Detailed Questionnaire
Airline Screener Questionnaire
Airline Deicing Detailed Questionnaire
Distributed
153
72
58
Returned
Undelivered
0 (0%)
1 (1%)
0 (0%)
Returned
Completed
150 (98%)
70 (97%)
49 (84%)
Not Returned
3 (2%) 1
1 (1%)
9 (16%)
EPA determined that one airport recipient was out of scope and removed it from the sample frame.
July 2009
4-5
-------
Technical Development Document for Proposed Effluent 4. Data Collection Activities
Limitation Guidelines and Standards for the Airport Deicing Category
4.3.1.1 Airport Questionnaire
EPA selected 153 airports to receive the airport questionnaire and distributed the
questionnaire to these airports in April 2006. Of the 153 airport questionnaires distributed, 150
were returned. EPA removed one of the three non-respondent airports from the sample frame
because it was a city airport operating as a tenant at a military airport. EPA determined that its
selection was based on data for the military airport operations, not the city airport, and military
airports were not included in the sample frame.
4.3.1.2 Airline Screener Questionnaire
EPA selected 94 airlines as recipients of the screener questionnaire. The recipient group
was comprised of a random sample of airlines with greater than 1,000 departures per year
operating at the airports that were sent the airport questionnaire, and a judgment sample of small
and foreign airlines for which additional information would be useful in developing effluent
guidelines, but that were not captured into the random sample. In April 2006, the Agency
distributed the airline screener questionnaire to 72 of the 94 selected airlines. The 22 remaining
airlines were foreign carriers operating at the selected airports. EPA collected aircraft deicing
and anti-icing information for these 22 foreign carriers through contacts with airport managers
where the carriers operated.
Of the 72 screener questionnaires distributed, 70 were returned. Of the two not returned,
one questionnaire was returned undelivered, as the airline had ceased operations.
4.3.1.3 Airline Detailed Questionnaire
Using the responses from the airline screener questionnaire, EPA selected 58 airlines that
responded that they deice planes directly to receive a more detailed questionnaire. This
questionnaire was distributed in March 2007. The selection included 448 airline/airport
combinations. The Agency categorized the airline/airport combinations according to the entity
that performed most of the deicing for the 2002/2003, 2003/2004, and 2004/2005 winter seasons
as listed below:
• Airline combinations that deice their own aircraft;
• Airline combinations that contract to fixed-base operators (FBOs) for deicing
services; or
• Airline combinations that contract to other airlines for deicing services.
Of the 58 airline detailed questionnaires sent, 49 were returned and 9 were not returned.
4.3.2 Questionnaire Information Collected
EPA designed the questionnaires to collect current information with sufficient detail to
support development of effluent guidelines. The questionnaires collected information on deicing
operations performed on aircraft and airfield pavement, including deicing stormwater generation,
collection, characterization, management, and treatment. The airline screener supported the
selection of recipients for the airline detailed questionnaire. This section describes the technical
information collected and the purpose of each of the three questionnaires.
July 2009 4-6
-------
Technical Development Document for Proposed Effluent 4. Data Collection Activities
Limitation Guidelines and Standards for the Airport Deicing Category
4.3.2.1 Airport Questionnaire
EPA divided the airport deicing questionnaire into the following parts and sections:
PART A: TECHNICAL INFORMATION
Section 1: General Airport Information
Section 2: Airport Deicing and Anti-Icing Operations
Section 3: Deicing Stormwater Containment and/or Collection
Section 4: Deicing Stormwater Treatment/Recovery
Section 5: Analytical Data
Section 6: Pollution Prevention Practices
PART B: FINANCIAL AND ECONOMIC INFORMATION
Section 1: Ownership and Management Structure
Section 2: Airport Finances
Section 3: Capital Expenditures
Section 4: Airport Operations
Section 1 (Questions 1 through 24) requested information to identify the airport and
primary contacts, to confirm that aircraft deicing/anti-icing was performed during the three
designated winter seasons (2002/2003, 2003/2004, and 2004/2005), and to characterize deicing
operations. This information included the destination of deicing/anti-icing Stormwater, receiving
surface waters, the entity that performed aircraft and/or airfield pavement deicing, and the
number of deicing/anti-icing days per winter season. This information helped EPA update the
industry profile, characterize deicing/anti-icing operations, and determine the proximity and
types of ecosystems within and beyond airport boundaries
Section 2 (Questions 25 through 31) requested detailed information about airport
deicing/anti-icing Stormwater sources, flows, and destinations as well as deicing/anti-icing
chemicals, materials, and practices. EPA used this information to develop an industry profile of
deicing Stormwater generation and collection, to determine baseline loadings using airfield
deicing chemical usage, and to develop and evaluate possible regulatory technology options and
compliance costs estimates.
Section 3 (Questions 32 through 39) requested information on the collection,
containment, conveyance, discharge and/or disposal methods for deicing Stormwater, and
pollution prevention and best management practices. EPA used this information to develop an
industry profile of deicing Stormwater collection/containment/conveyance methods and to
evaluate pollution prevention and best management practices.
Section 4 (Questions 40 through 51) requested information on deicing Stormwater
treatment technologies and units operated by the airport and included deicing Stormwater
treatment diagrams, design and operating specifications, sources of wastewater influent,
treatment chemical additions, treatment operations and maintenance costs, and discharge
practices. EPA used this information to develop control technology options, regulatory options,
and compliance cost estimates.
July 2009 4-7
-------
Technical Development Document for Proposed Effluent 4. Data Collection Activities
Limitation Guidelines and Standards for the Airport Deicing Category
Section 5 (Questions 52 through 55) requested information concerning the availability of
deicing stormwater characterization data, receiving water in-stream monitoring data, and/or data
characterizing the effectiveness of treatment of deicing stormwater. EPA used this information to
follow up with selected airports to request long-term monitoring data, to estimate pollutant
discharge loadings, to characterize behavior of the discharge in the receiving water, to assess
deicing stormwater treatment technologies, and to help assess environmental impacts.
Section 6 (Questions 56 through 68) requested information to evaluate the status of
pollution prevention practices at each airport and to identify pollution prevention technologies.
EPA used this information to identify appropriate practices as regulatory options and to prepare
an industry profile of pollution prevention practices.
Part B of the questionnaire requested airport financial and economic information. Section
1 requested information on the ownership and management structure that EPA used to develop
the industry profile and to estimate economic impacts of an effluent guideline. Section 2
requested information on operation finances that EPA used to project the potential impacts of the
rule. Section 3 requested information on current capital airport expenditures that EPA used to
assess the capability of airports to pay for deicing-related capital improvements. Section 4
requested information on the finances for airport operations including the airport's financial
statement that EPA used to determine the airport's cost of capital.
4.3.2.2 Airline Screener Questionnaire
The airline screener questionnaire requested information on which entity performed most
of the deicing/anti-icing operations on an airline's aircraft, including the name of another airline,
fixed base operator (FBO), or private contractor that performed the service. EPA used this
information to identify potential airline detailed questionnaire recipients and to indicate the
potential contribution of FBOs to deicing operations and to the discharge of ADF-contaminated
stormwater.
The screener included three questions. Question 1 requested the contact information of
the airline contact that could verify or clarify the screener information. Question 2 asked who
performed most of the deicing/anti-icing on the respondent's aircraft at specific airports and, if
applicable, requested the identity of the other airline, FBO, or private contractor that performed
the deicing. Question 3 provided an opportunity for the respondent to provide additional
information or comment on the screener responses.
4.3.2.3 Airline Detailed Questionnaire
EPA divided the airline detailed questionnaire into the following parts and sections:
PART A: TECHNICAL INFORMATION
Section 1: General Airline Information
Section 2: Airline Deicing and Anti-Icing Operations (at each airport specified in
Section 1)
Section 3: Pollution Prevention Practices (at each airport specified in Section 1)
July 2009 4-8
-------
Technical Development Document for Proposed Effluent 4. Data Collection Activities
Limitation Guidelines and Standards for the Airport Deicing Category
PART B: DEICING COSTS AND OPERATIONS
Section 1: Airline Deicing Costs and Operations
Section 2: Airport-Specific Deicing Costs and Operations
Section 1 (Questions 1 through 4) requested verification of the airline name and address,
and identification of the primary and secondary contacts to clarify or verify the technical
questionnaire responses.
Section 2 (Questions 5 through 18) requested information on deicing/anti-icing
operations performed on the airline's aircraft and/or by the airline for another airline's aircraft at
each specified airport. The Agency used this information to:
• Develop an industry profile of ADF usage and deicing stormwater generation;
• Estimate pollutant loadings;
• Characterize deicing stormwater;
• Evaluate differences in airport deicing stormwater generation and characteristics;
• Identify pollutants of concern; and
• Identify opportunities for pollution prevention through chemical substitution and
best management practices.
Section 3 (Questions 19 through 31) requested information on the airline's pollution
prevention practices including a description of each practice and any costs and/or savings from
its implementation. The Agency evaluated this information to identify appropriate practices that
could become part of regulatory options and to develop an industry profile.
PartB of the questionnaire requested airline financial and economic information.
Section 1 requested information on ownership, aircraft deicing costs and operations, and the
airline's financial statement that EPA used to develop the industry economic profile and to
conduct the economic analysis. Section 2 requested detailed information regarding airline-
specific deicing costs and operations at specific airport locations. The Agency also used this
information for the industry economic profile and to determine the economic impacts of the rule.
4.3.3 Questionnaire Technical Review, Coding, and Data Entry
EPA completed detailed technical reviews of the screener and the two detailed
questionnaires for completeness, accuracy, and consistency of the responses. In some cases, the
Agency followed up with the airport or airline by email or telephone to clarify responses or to
obtain missing or incomplete technical information. During the technical review, EPA coded
responses to facilitate entry of technical data into the airline screener and the airline and airport
questionnaire databases.
The Agency developed databases containing the technical information provided by
questionnaire respondents of each questionnaire. After detailed technical review and coding,
EPA entered data from the questionnaires into the appropriate database using a double-key entry
and verification procedure to identify and resolve differences between the two data entry tasks.
The database dictionary for each questionnaire presents the database structure and codes and
July 2009 4-9
-------
Technical Development Document for Proposed Effluent 4. Data Collection Activities
Limitation Guidelines and Standards for the Airport Deicing Category
defines each field in the database files. These dictionaries are located in the administrative record
for the Airport Deicing rulemaking.
4.4 Field Sampling
EPA conducted sampling episodes at six airports from March 2005 through August 2006
to characterize ADF and ADF-contaminated stormwater discharges and to evaluate treatment
technologies for stormwater affected by aircraft and airfield deicing practices. EPA used existing
industry profile information and information collected during airport site visits to determine the
most appropriate locations for sampling. The Agency evaluated the following criteria for
selecting sampling sites:
• Size and location of the airport;
• Deicing stormwater collection and control practices; and
• ADF-contaminated stormwater discharge practices.
EPA conducted the episodes to characterize deicing stormwater and assess the
capabilities and effectiveness of several different treatment technologies such as anaerobic
treatment, aerobic treatment, distillation, reverse osmosis, mechanical vapor recompression,
aeration, and chemical addition. Table 4-3 lists the airports selected for EPA sampling, the
reason for selection, and the points that were sampled.
4.5 Permit Review
During the regulatory development process, EPA conducted a National Pollutant
Discharge Elimination System (NPDES) permit review to understand what permit authorities are
currently requiring of airports with respect to deicing stormwater control. The permit review
performed three functions:
1. Assessed the current state of deicing stormwater control;
2. Helped evaluate the effectiveness of various deicing stormwater control measures;
and
3. Identified potential measures that EPA could use to further control deicing
stormwater.
The following discussion describes the process EPA used to select airports for permit review and
to obtain and review NPDES permits for this analysis.
July 2009 4-10
-------
Technical Development Document for Proposed Effluent
Limitation Guidelines and Standards for the Airport Deicing Category
4. Data Collection Activities
Table 4-3. Airports Selected for Sampling and the Reason for Their Selection
Airport Name
Detroit Metropolitan
Wayne County
Detroit, MI
Episode 6508
Minneapolis/St. Paul
International
Minneapolis/St. Paul,
MI
Episode 6509
Albany International
Albany, NY
Episode 6523
Pittsburgh
International
Pittsburgh, PA
Episode 6528
Denver International
Denver, CO
Episode 6522
Greater Rockford
Rockford, IL
Episode 6529 and 6530
Airport
Code
DTW
MSP
ALB
PIT
DIA
RFD
Dates of
Sampling
3/31/05
4/28/05
2/5/06-
2/10/06
2/26/06-
3/3/06
3/26/06-
3/31/06
4/20/06
and
8/29/06
Reason for Sampling
Collects highly
concentrated ADF for
recycling, significant
stormwater volumes,
direct and indirect
discharger
On-site collection and
recycling facility, direct
and indirect discharger
Reported recovery
efficiency of 72% of
applied ADF through
ADF -contaminated
stormwater collection
with anaerobic and
aerobic treatment
Reported recovery
efficiency of 60-66% of
applied ADF through
collection and treatment
(ultrafiltration and
reverse osmosis) of
ADF -contaminated
stormwater
ADF -contaminated
stormwater collection
with glycol recovery
On-site aerated lagoon
treatment system (run in
batch mode) for its
deicing-contaminated
stormwater
Sample Points
• Untreated deicing stormwater
• Effluent from ADF contaminated
stormwater collection pond
• Effluent from pavement deicer
stormwater collection pond
• ADF, as applied
• QC samples :
• High concentration ADF-
contaminated stormwater storage
tank
• Low concentration ADF-
contaminated stormwater storage
tank
• Influent to pavement deicer
stormwater collection pond
• Effluent from pavement deicer
stormwater collection pond
• ADF, as applied
• QC samples :
• Influent to anaerobic treatment
• Effluent from anaerobic treatment
• Effluent from aerobic treatment
• QC samples :
• Influent to RO treatment
• Effluent from RO treatment
• QC samples :
• Influent to MVRs
• Influent to distillation column
• Distillate from MVRs
• Overhead from distillation column
• Effluent from treatment
• QC samples :
Spring Sampling
• Influent to aerobic pond treatment
• QC samples :
Summer Sampling
• Effluent from aerobic pond
treatment
• QC samples :
1 QC samples may include source water, duplicate samples, trip blanks, equipment blanks, bottle blanks, field
blanks.
July 2009
4-11
-------
Technical Development Document for Proposed Effluent 4. Data Collection Activities
Limitation Guidelines and Standards for the Airport Deicing Category
4.5.1 Airport Selection for Permit Review
For this review, EPA selected the top 50 U.S. airports based on ADF usage. At the time
of review, usage estimates from the airline survey were not available; therefore, EPA estimated
the airports with the highest usage using a "weighting factor" based on the number of snow or
freezing precipitation (SOFP) days and commercial departures as a measure of ADF usage.
Table 4-4 displays the results of the weighting factor analysis and lists the 50 airports for which
EPA reviewed permits.
4.5.2 Obtaining NPDES Permits
From the list of selected airports for permit review, EPA identified those for which it
already had permits. While airports were not required to submit permits as part of their survey
response, some airports did so. Furthermore, there were some permits already available in the
airport deicing record files. Therefore, as a first step, EPA reviewed the administrative record
index and survey responses to identify in-house permit availability.
As a next step, EPA identified NPDES permit numbers for those permits not available in-
house. Some airports reported permit numbers in the airport questionnaire, so in these cases,
EPA obtained the data from the questionnaire database. For the remaining airports, EPA
determined permit numbers by searching for facilities on EPA's Envirofacts search tool. The
facilities were searched by facility name, location, SIC code (4581), or a combination of any of
the three. In a few cases, there were airports for which EPA could not identify permit numbers
from either the questionnaire or Envirofacts. For these airports, EPA searched the Internet or
contacted permitting authorities or the airport directly to obtain a copy of the permit.
After identifying the permit numbers, EPA obtained a copy of the permit from a state or regional
permit database. If a permit was not available online, EPA contacted the appropriate regional,
state, or local permitting authority to obtain a copy. If still unsuccessful in obtaining the permit,
EPA contacted the airport directly to request a copy.
4.5.3 Permit Review Process
The objectives of the permit review were to answer the following questions:
• What are the monitoring requirements for deicing area outfalls?
• What pollutants are monitored?
• Are there numeric limits listed in the permit for deicing area outfalls?
• What parameters are limited?
• What are the limits for each parameter?
• How were the limits developed?
• Are there deicing operation best management practices (BMPs) required by the
permit?
• What BMPs are required?
• When does the permit expire?
• If it is a general permit, are there differences between the permit and the EPA
Multi-Sector General Permit (MSGP)?
July 2009 4-12
-------
Technical Development Document for Proposed Effluent
Limitation Guidelines and Standards for the Airport Deicing Category
4. Data Collection Activities
Table 4-4. Top 50 Airports in the United States with the Highest ADF Usage, Estimated Based on Snow and Freezing
Precipitation Days and Total Airport Departures
Rank
1
2
o
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
Airport
ID
1006
1126
1138
1028
1012
1053
1113
1069
1142
1107
1145
1095
1139
1136
1029
1066
1010
1011
1148
1021
1089
1024
1141
1129
1059
1111
Airport Code
ORD
MSP
DTW
DEN
ANC
BOS
CVG
CLE
IAD
PIT
EWR
MOW
PHL
MKE
LGA
SLC
FAI
STL
MCI
BUF
JFK
IND
DCA
BDL
ROC
CMH
Airport Name
Chicago O'Hare International
Minneapolis - St Paul International - Wold-
Chamberlain
Detroit Metropolitan Wayne County
Denver International
Ted Stevens Anchorage International
General Edward Lawrence Logan International
(Boston)
Cincinnati/Northern Kentucky International
Cleveland - Hopkins International
Washington Dulles International
Pittsburgh International
Newark Liberty International
Chicago Midway International
Philadelphia International
General Mitchell International (Milwaukee)
La Guardia (New York City)
Salt Lake City International
Fairbanks International
Lambert - St Louis International
Kansas City International
Buffalo Niagara International
John F Kennedy International (New York City)
Indianapolis International
Ronald Reagan Washington National
Bradley International (Windsor Locks)
Greater Rochester International
Port Columbus International
State
IL
MN
MI
CO
AK
MA
KY
OH
DC
PA
NJ
IL
PA
WI
NY
UT
AK
MO
MO
NY
NY
IN
DC
CT
NY
OH
Snow or
Freezing
Precipitation
Days
26
41
31
26
55
26
17
36
17
31
16
26
12
31
12
14
89
17
27
48
12
21
12
31
44
26
Total Airport
Departures
467,721
246,286
250,629
264,051
88,126
186,253
247,165
116,569
238,635
125,143
203,082
108,385
227,749
85,128
192,127
160,472
24,919
129,414
76,016
41,916
154,606
83,769
134,346
51,389
35,726
59,938
Weighting Factor
SOFP Days x
Departures •*•
100,000
121.6
101.0
77.7
68.7
48.5
48.4
42.0
42.0
40.6
38.8
32.5
28.2
27.3
26.4
23.1
22.5
22.2
22.0
20.5
20.1
18.6
17.6
16.1
15.9
15.7
15.6
July 2009
-------
Technical Development Document for Proposed Effluent
Limitation Guidelines and Standards for the Airport Deicing Category
4. Data Collection Activities
Table 4-4 (Continued)
Rank
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
Airport
ID
1036
1026
1065
1080
1140
1128
1079
1058
1037
1123
1020
1121
1147
1068
1105
1108
1124
1074
1153
1109
1018
1100
1022
1051
Airport Code
BWI
DFW
ALB
SYR
MEM
CLT
MHT
GRR
IAH
DAY
ATL
PVD
RDU
OMA
GEG
SDF
DSM
SEN
CAK
ILN
GSO
TOL
FWA
HYA
Airport Name
Baltimore - Washington International
Dallas/Fort Worth International
Albany International
Syracuse Hancock International
Memphis International
Charlotte/Douglas International
Manchester
Gerald R Ford International (Grand Rapids)
George Bush Intercontinental (Houston)
James M Cox Dayton International
Hartsfield - Jackson Atlanta International
Theodore Francis Green State (Providence)
Raleigh - Durham International
Eppley Airfield (Omaha)
Spokane International
Louisville International - Standiford Field
Des Moines International
South Bend Regional
Akron - Canton Regional
Airborne Airpark (Wilmington)
Piedmont Triad International (Greensboro)
Toledo Express
Fort Wayne International
Barnstable Municipal - Boardman/Polando Field
(Hyannis)
State
MD
TX
NY
NY
TN
NC
NH
MI
TX
OH
GA
RI
NC
NE
WA
KY
IA
IN
OH
OH
NC
OH
IN
MA
Snow or
Freezing
Precipitation
Days
12
4
36
44
8
6
36
48
4
26
2
21
10
26
31
12
31
48
41
21
14
36
31
26
Total Airport
Departures
124,033
360,933
39,324
30,840
166,910
214,396
34,860
25,015
248,339
35,709
459,765
43,671
86,302
33,022
27,269
65,586
23,951
13,722
14,911
25,508
38,257
14,385
16,247
18,782
Weighting Factor
SOFP Days x
Departures •*•
100,000
14.9
14.4
14.2
13.6
13.4
12.9
12.5
12.0
9.9
9.3
9.2
9.2
8.6
8.6
8.5
7.9
7.4
6.6
6.1
5.4
5.4
5.2
5.0
4.9
July 2009
-------
Technical Development Document for Proposed Effluent
Limitation Guidelines and Standards for the Airport Deicing Category
4. Data Collection Activities
EPA also consulted Envirofacts as necessary to fill in any numeric limit data gaps from
the permits and to cross-check for accuracy with the permitted limits.
To facilitate interpreting the results, EPA created a Microsoft® Access database to store
the data obtained from the reviews. The database consists of two tables: a General Information
table, and a Pollutant-Specific Information table. Tables 4-5 and 4-6 detail the database table
descriptions. A summary of the permit review is presented in the memorandum Airport Deicing
Operations NPDES Permit Review Summary (ERG, 2007).
Table 4-5. Permit Review General Information Table
Data Element
AirporuD
PermitJD
Permit_Expiration
Permit_BMPs
Permit BMPs Description
General_Permit
General Permit Difference
from MSGP
Permit_Monitoring
Permit Limits
Permit_Limit_Rationale
Data Element Description
The airport identification number used for the Airport Questionnaire.
The airport NPDES identification number.
The permit expiration date.
A checkbox that identifies the presence of BMPs in the permit
A field that allows the BMPs in the permit to be listed.
A checkbox that identifies general permits.
A field that provides a description of any differences that exist between general
permits and the MSGP.
A checkbox that indicates if the permit requires monitoring.
A checkbox that indicates if the permit has numeric limits.
A field that provides a description of what rationale was used to determine the limits
in the permit.
July 2009
4-15
-------
Technical Development Document for Proposed Effluent
Limitation Guidelines and Standards for the Airport Deicing Category
4. Data Collection Activities
Table 4-6. Permit Review Pollutant-Specific Information Table
Data Element
AirportID
Permit_SamPoint
Permit_Stream_Description
Permit_Pollutant
OtherDesc
PermitTimes
PermitFreq
PermitLimitNumeric
PermitLimitUnit
LimitType
Season
Data Element Description
The airport identification number used for the detailed airport questionnaire.
The outfall/sampling area identifying number.
Description of the outfall/sampling area. EPA tried to determine which outfalls
receive deicing stormwater and to include only information from those outfalls. If
the deicing outfalls cannot be determined, EPA included all outfalls.
Pollutants that are monitored and/or limited at each outfall, using the following
same codes that are used in the questionnaire database:
• BOD • Metal • ORG (organic • TSS (total
• COD • N pollutants) suspended
• Fecal coliform • OG (oil & • pH solids)
grease) • TOC (total • Other
organic carbon)
If a pollutant is monitored/limited that does not have a code, "Other" was selected
as the Permit_Pollutant and the pollutant name was entered in this field. This
method of tracking pollutants was used to be consistent with the questionnaire
database.
To be consistent with the questionnaire database, the frequency of monitoring for
each pollutant and outfall was recorded in the PermitTimes and PermitFreq fields.
These fields allow the frequency to be reported in a number/unit manner. For
example, a yearly report is entered as PermitTimes = 1 and PermitFreq = Year.
Frequency Codes were also used in the PermitFreq field for daily (D), monthly
(M), and quarterly (Q) reports.
Numeric value of the permit limit for each pollutant and outfall.
Unit of the permit limit for each pollutant and outfall.
Indicates whether the limit is a minimum value, maximum value, average, or
simply a reporting requirement. This also incorporates the timespan of the limit
using the frequency codes as above (e.g., daily maximum = DMAX; weekly
average = WAVG).
For deicing outfalls, the limits may vary by season for various parameters. Usually,
this field is populated with Summer, Winter or All (as in year-round).
4.6
Industry-Submitted Data
Based on airport site visits, EPA sampling episodes, and responses to the airport
questionnaire, EPA requested costing and long-term analytical data for managing deicing
stormwater from specific airports. The Agency used this information to develop control
technology options, compliance cost estimates, and to evaluate pollution prevention and best
management practices. Tables 4-7 and 4-8 list those airports providing data, the type of system
used to manage or treat the airport's deicing stormwater, and the costing and/or analytical data
submitted by the airport.
July 2009
4-16
-------
Technical Development Document for Proposed Effluent
Limitation Guidelines and Standards for the Airport Deicing Category
4. Data Collection Activities
Table 4-7. Summary of Costing Data Provided by Industry
Airport
Akron-Canton Regional
Albany International
Cincinnati/Northern
Kentucky International
Denver International
General Mitchell
International
Minneapolis-St. Paul
International -Wold
Chamberlain
Pittsburgh International
Seattle-Tacoma International
Type of Deicing Stormwater
Management
Anaerobic ADF-contaminated
Stormwater treatment system
AFB/aerobic ADF-contaminated
Stormwater treatment system
Glycol recovery and recycling
system
Storage, recovery, and recycling;
mechanical vapor recompression
(MVR) and distillation system
Recovery and recycling; anaerobic
digester
ADF collection (deicing pads, plug
and pump system)
ADF collection at deicing pads;
ADF-contaminated Stormwater
recovery and recycling
ADF to industrial waste treatment
plant
Costing Data Provided
Capital and operating and maintenance costs
for the airport's new anaerobic fluidizedbed
(AFB) treatment system
Capital and operating and maintenance costs
for the airport's AFB/aerobic treatment system
Capital and operating and maintenance costs
for glycol collection and treatment
Capital costs for storage and the
recycle/recovery system
Engineering and monitoring-related costs
Capital costs for deicing pads and operating
and maintenance costs for plug and pump
system
Operating and maintenance costs for deicing
pads
Study costs for determining all known and
reasonable technology (AKART) for handling
aircraft deicing fluids
Table 4-8. Summary of Long-Term Analytical Data Provided by Industry
Airport
Albany International
Denver International
Detroit Metropolitan -
Wayne County
Pittsburgh International
Salt Lake City International
Type of Deicing Stormwater
Management
Anaerobic/aerobic ADF-contaminated
Stormwater treatment
Storage, recovery, and recycling;
MVR/Distillation
Recycling; distillation and recovery
Ultrafiltration/Reverse Osmosis; ADF-
contaminated Stormwater recovery and
recycling
ADF recovery and recycling
Long-Term Analytical Data
Provided
Ammonia, COD
COD
Ammonia
Ammonia, urea
COD
4.7
Literature Reviews
EPA conducted preliminary literature searches during the effluent guideline development
process to supplement information acquired from site visits, sampling, and questionnaires. The
purpose for the literature searches was three-fold:
• To collect information on current airport deicing practices and trends, and gather
information on state-of-the-art deicing Stormwater treatment and/or glycol
recovery technologies;
July 2009
4-17
-------
Technical Development Document for Proposed Effluent 4. Data Collection Activities
Limitation Guidelines and Standards for the Airport Deicing Category
• To collect available data from airports currently monitoring wastewater
discharges; and
• To obtain studies on the toxicity and environmental impact of current deicing
fluids and deicing stormwater runoff.
The following subsections list the data sources used for each literature search.
4.7.1 Current Deicing Practices and Treatment Technologies
EPA performed keyword searches on three on-line search engines: 1) Cambridge
Scientific Abstracts (CSA); 2) Dialog Version 5.0; and 3) Google™. CSA provides access to
over 50 databases published by CSA and its publishing partners, such as Aqualine,
Environmental Sciences & Pollution Management Database, and Water Resources Abstracts.
Dialog provides access to over 900 databases and handles more than 700,000 searches. The
databases in Dialog that contain articles pertaining to airport deicing are BIOSIS Toxicology,
Life Sciences Abstracts, Institute for Science Information, ProQuest Info & Learning, Ei
Compendex, Enviroline, TGG National Newspaper Index, GEOBASE, NTIS, and Wilson
Applied Science & Technology Abs.
The keywords for the literature searches included: airport deicing, aircraft, airfield,
runway, aircraft deicing, aircraft deicing fluid (ADF), runway deicing, anti-icing, anti-icing fluid,
airport stormwater, snow melt, centralized deicing pads, environmental assessment,
environmental impact study (EIS), fish mortality, fish kill, and publicly owned treatment works
(POTW).
EPA also used other on-line journal databases, such as Science Direct, Scirus, and
Infotrak, for subject-specific articles. The treatment technologies featured in the articles found
included:
• Aerobic fluidized bed reactor/ biological treatment;
• Aerated storage tanks;
• Anaerobic co-digestion of aircraft deicing fluid and municipal wastewater sludge;
• Batch-loaded anaerobic fluidized bed reactor;
• Glycol reclamation/recycling and concentration;
• Infrared technology;
• Phytoremediation;
• Plant-enhanced remediation;
• Spray irrigation;
• Subsurface-flow constructed wetlands; and
• Surface detention ponds.
July 2009 4-18
-------
Technical Development Document for Proposed Effluent 4. Data Collection Activities
Limitation Guidelines and Standards for the Airport Deicing Category
4.7.2 Current Airport Deicing Runoff Data
In addition to sampling and airport questionnaire data, EPA procured airport deicing
runoff information from its Permit Compliance System (PCS) database and on-line journals.
EPA downloaded all data reports from PCS for SIC code 4581: Airports, flying fields, and
services. Not all airports report to their permitting authority, so the scope of runoff data is
limited. The pollutant parameters include temperature, dissolved oxygen, biochemical oxygen
demand (BOD), total suspended solids (TSS), metals, fecal coliform, aromatic hydrocarbon, pH,
and oil and grease. For on-line searches, EPA procured journals that discussed deicing runoff
containing ADF chemicals such as benzotriazole, propylene/ethylene glycol, and alklyphenol
ethoxylates. EPA also collected Minneapolis/ St. Paul International Airport monitoring data
during the site visit to that airport.
4.7.3 Chemical Information and Environmental Impact Studies
The methodology and databases used for chemical information and environmental impact
study findings are similar to those used for the deicing practices and treatment technology
search. EPA conducted searches for the following categories:
• Chemical Properties of ADF Ingredients: physical appearance, structure,
solubility, reactivity;
• Human Toxicity: Inhalation, ingestion, dermal effect, oral rat lethal dose (LD50)
values;
• Aquatic Toxicity: Aquatic life lethal concentration (LC50) values; and
• Chemical Fate and Transport: Soil sorption, fate in river, streams, and
estuaries, breakdown pathways in anaerobic and aerobic conditions, and
biodegradability.
In addition to journal articles, EPA gathered chemical information from Material Safety
Data Sheets (MSDSs), Chemfmder.com, Wikipedia, the Pesticides Action Network (PAN)
Pesticides Database, and the U.S. Patents Database.
The keywords for the pollutant term search included: propylene glycol, propylene glycol-
based fluids, ethylene glycol, ethylene glycol-based fluids, urea, potassium acetate, calcium
magnesium acetate (CMA), sodium acetate, sodium formate, dissolved oxygen, biodegradation,
BOD, and ADF additives (e.g., tolytriazole, benzotriazole, nonylphenols, nonylphenol
ethoxylate, etc.).
4.7.4 Current Deicing Runoff Regulations
In addition to the searches described above, EPA searched the Internet using Google™ to
review regulatory documents that contain guidelines, operation controls, management programs,
laws, statues, and certification requirements related to airport deicing from the United States,
Canada, Germany, Norway, and other European countries.
July 2009 4-19
-------
Technical Development Document for Proposed Effluent 4. Data Collection Activities
Limitation Guidelines and Standards for the Airport Deicing Category
4.8 References
ERG. 2007. Memorandum from Jason Huckaby (ERG) to Brian D'Amico and Eric Strassler
(U.S. EPA): Airport Deicing Operations NPDES Permit Review Summary. (April 16). DCN
AD00611.
USEPA.2008a. Airline Screener Questionnaire Database. U.S. Environmental Protection
Agency/Office of Water. Washington, D.C. DCN AD00937.
USEPA. 2008b. Airline Detailed Questionnaire Database. U.S. Environmental Protection
Agency/Office of Water. Washington, D.C. DCN AD00938.
USEPA. 2008c. Airport Questionnaire Database. U.S. Environmental Protection Agency/Office
of Water. Washington, D.C. DCN AD00927.
USEPA. 2000. Preliminary Data Summary: Airport Deicing Operations. U.S. Environmental
Protection Agency/Office of Water. Washington, D.C. EPA-821-R-00-016. Available online at
http://www.epa.gov/guide/airport. DCN AD00005.
USEPA. 2005. Supporting Statement: Survey of Airport Deicing Operations. U.S. Environmental
Protection Agency/Office of Water. Washington, D.C. DCN AD00447.
July 2009 4-20
-------
Technical Development Document for Proposed Effluent 5. Overview of the Industry
Limitation Guidelines and Standards for the Airport Deicing Category
5. OVERVIEW OF THE INDUSTRY
The Airport Deicing Category includes the deicing and/or anti-icing of airfield pavement
and aircraft. This section provides an overview of the airport deicing/anti-icing performed by
selected airports and airlines. The overview includes statistics on the number and location of
airports and airlines that perform deicing/anti-icing (Section 5.1), deicing and anti-icing practices
performed on airfields and aircraft and methods used to collect and control deicing stormwater
(Section 5.2), and the references used in this section (Section 5.3).
5.1 Industry Statistics
Data sources for statistics on the number and types of airports and airlines include
responses to EPA's airport and airline questionnaires, the Bureau of Transportation Statistics,
Federal Aviation Administration (FAA), and EPA''s Preliminary Data Summary: Airport
Deicing Operations (PDS) (USEPA, 2000). Data provided in responses to EPA questionnaires
are based on deicing/anti-icing operations performed during the winter seasons of 2002/2003,
2003/2004, and 2004/2005, hereafter referred to as the three winter seasons.
5.1.1 Airports
The North American Industry Classification System (NAICS) identification number
applicable to airport deicing is 488119: Other Airport Operations. The U.S. Census Bureau
describes this industry as establishments primarily engaged in the following: (1) operating
international, national, or civil airports or public flying fields, or (2) supporting airport
operations, such as runway maintenance services, hangar rental, and/or cargo handling services.
The airport questionnaire data presented in this section are based on the 150 respondents
to EPA's airport questionnaire. In some cases, EPA applied weighting factors to the information
provided by selected airport questionnaire recipients to scale up the questionnaire data to
represent national estimates.
5.1.1.1 Number and Types of Airports
FAA's general categories of airports include commercial, general aviation, and relievers.
Commercial airports are public airports receiving scheduled passenger service and having more
than 2,500 enplaned passengers (number of passengers boarding a plane for departure) each year.
General aviation airports have less than 2,500 enplanements per year or do not receive scheduled
commercial service. Relievers are high-capacity general aviation airports in major metropolitan
areas, and provide an alternative for small aircraft using busy commercial airports.
Airports may be further classified into several different categories, depending on the size
and activity level of the airport. Often both of these factors can be determined by the number of
enplanements or operations (number of arrivals and departures) at the airport in a given year.
FAA classifies large commercial airports into "hubs," based on the number of annual
enplanements that occur at the airport. Large hubs are defined as airports with more than one
percent of total U.S. passenger enplanements. Medium hubs are defined as airports with more
than 0.25 percent but less than 1 percent of total passenger enplanements. Small hubs account for
0.05 to 0.25 percent of the total passenger enplanements. Airports with less than 0.05 percent of
the total passenger enplanements but more than 10,000 annual enplanements are considered non-
July 2009 5-1
-------
Technical Development Document for Proposed Effluent
Limitation Guidelines and Standards for the Airport Deicing Category
5. Overview of the Industry
hub primary airports. Nonprimary commercial services are those airports that have 2,500 to
10,000 enplanements a year.
According to FAA 2004 data and the FAA's National Plan of Integrated Airport Systems
(NPIAS) Report to Congress, about 3,344 airports operated in the United States in 2004. Table 5-
1 identifies the number of airports by type as defined by number of enplanements. For all airport
types, excluding general aviation airports, the totals in Table 5-1 represent counts for January
through December 2004. FAA's designation of hub status depends on the percent of total
passenger boardings occurring at each airport, resulting in variation in the number of airports in
each hub category from year to year.
Table 5-1. Number of U.S. Airports by Airport Type in 2004
Airport Type
Large Hub
Medium Hub
Small Hub
Non-hub
Other Nonprimary
General Aviation :
General Aviation Relievers :
TOTAL
Number of Airports
33
36
67
231
130
2,573
274
3,344
1 General aviation and general aviation reliever airports (open to the public) from the NPIAS Report to
Congress. (USDOT, 2008a)
Note: Airport counts will differ depending on the source and year of data represented.
EPA distributed the airport questionnaire to 153 airports that included, based on 2004
information, all large hub, all medium hub, and a statistical sampling of small hub and non-hub
airports, as well as some general aviation/cargo, and other nonprimary airports. EPA determined
that one airport recipient was out of scope and removed it from the sample frame. EPA received
responses from 150 of these airports. The estimated total number of airports nationally that
perform deicing and/or anti-icing of airfield pavement and/or aircraft during the three winter
seasons surveyed by EPA is 334. By airport type, this includes 28 large hubs, 36 medium hubs,
40 small hubs, 226 non-hubs, and 4 general aviation/cargo airports.
5.1.1.2 Geographic Location of Deicing Airports
The location of the airport and its climate have a direct impact on deicing operations.
Airport deicing/anti-icing operations occurred in 44 states in the three winter seasons. As shown
below, the FAA divides the United States into the following nine regions:
July 2009
5-2
-------
Technical Development Document for Proposed Effluent
Limitation Guidelines and Standards for the Airport Deicing Category
5. Overview of the Industry
Region
Alaskan
Central
Eastern
Great Lakes
New England
Northwest Mountain
Southern
Southwest
Western-Pacific
State
AK
IA, KS, MO
DE, MD, NJ, NY, PA, VA, WV
IL, IN, MI, MN, ND, OH, SD, WI
CT, ME, MA, NH, RI, VT
CO, ID, MT, OR, UT, WA, WY
AL, FL, GA, KY, MS, NC, PR, TN, VI
AR, LA, MM, OK, TX
AZ, CA, HI, NV, GU, AS, MH
Source: FAA, "FAA Regional Offices,"
http://www.faa.gov/about/office org/headquarters offices/arp/regional offices/. (U.S. DOT, 2008b).
Table 5-2 summarizes the regions for the airports that reported deicing in the EPA airport
questionnaire for the three winter seasons surveyed by EPA. (Note: these are not national
estimates.) The Great Lakes and Eastern regions reported the highest number of deicing airports.
Table 5-2. Deicing Airports by FAA Region for the Three Winter Seasons
Region
Great Lakes
Eastern
Southern
Northwest Mountain
Western-Pacific
Southwest
Alaskan
New England
Central
Airports Reporting Deicing and/or Anti-Icing in EPA
Airport Questionnaire
31
22
20
17
15
14
10
6
5
Source: EPA airport questionnaire database (USEPA, 2008c).
5.1.1.3 Weather Impacts on Airport Deicing/Anti-Icing
Airports conduct deicing/anti-icing operations when weather conditions, such as
precipitation and/or temperature, have the potential to cause icing. Precipitation includes
snowfall, rainfall, sleet (including freezing rain), and ice. The type of precipitation affects the
volume and type of deicing/anti-icing chemicals used on aircraft and airfield pavement. For
example, freezing rain requires the most deicing/anti-icing agent usage because the rain freezes
on contact and coats the aircraft or airfield pavement to form a solid layer of ice. Dry-weather
deicing is performed when the ambient temperature is cold enough to form ice on aircraft wings
and surfaces (below 55° F), and generally requires a small volume of aircraft deicing/anti-icing
fluids (ADF).
The duration of the deicing/anti-icing season is also determined by the climate at an
airport location. Airfield pavement deicing can begin as early as September and continue through
July 2009
5-3
-------
Technical Development Document for Proposed Effluent
Limitation Guidelines and Standards for the Airport Deicing Category
5. Overview of the Industry
May in colder climates and/or areas with high numbers of snow or freezing precipitation days.
The national estimate of airports performing airfield pavement deicing is 215, which is lower
than the national estimate of airports performing deicing operations overall. The difference is
that there are airports that have some aircraft deicing (usually defrost deicing) but no airfield
pavement deicing. In general, these airports are located in warm and/or dry weather climates
with minimal winter storm events. For airfield pavement deicing, December, January, and
February are the peak deicing months, and September and May have the lowest occurrences of
airfield pavement deicing. Figure 5-2 presents the percentage of these 215 airports deicing
airfield pavement for each month. The time frame during which an airport conducts deicing
during a typical winter season ranges from two to nine months, and a majority of airports
typically conduct deicing/anti-icing operations for five months a season. For the three winter
seasons surveyed by EPA, the average reported number of airfield pavement deicing days among
these 215 airports ranges from 0.3 to 240 days.
70%
60%
0)
S*
a
0)
Q_
ono/
*
71%
A
^
69%
e>
/
61%
s>
^
5%
^
5%
0% 0% 0% | |
>/ ^ ./ / ^
^ / o*
Month
24%
^
55%
<
*cP
71%
*
Figure 5-1. Percentage of Airports Deicing Airfield Pavement Each Month
5.1.1.4 Destination of Airport Deicing Stormwater
Airport questionnaire respondents reported direct, indirect, and zero discharge of deicing
stormwater. Direct dischargers discharge deicing stormwater directly to U.S. surface waters,
such as creeks, rivers, ponds, lakes, or oceans. Indirect dischargers convey deicing stormwater by
pipe, conduit, or hauling to a publicly owned or other treatment works. A zero discharger
disposes of deicing stormwater using methods other than direct or indirect discharge. Figure 5-3
presents the discharge status of airports by destination. Nationally, EPA estimates that 176
July 2009
5-4
-------
Technical Development Document for Proposed Effluent
Limitation Guidelines and Standards for the Airport Deicing Category
5. Overview of the Industry
airports discharge to surface water only, 52 airports discharge both directly to surface water and
indirectly to a publicly owned treatment works (POTW), 10 airports discharge to a POTW only,
and 96 airports have zero discharge.
A majority of the zero dischargers reported conducting aircraft deicing only (i.e., no
deicing of airfield pavement). These airports generally are in warm and/or dry weather climates
and have minimal dry weather (defrost) deicing. For airports with minimal deicing, it is assumed
that some portion of the ADF applied is lost as fugitive emissions during aircraft taxiing and
take-off and EPA relied on the airport's questionnaire response in assuming that no residual
ADF from those fugitive emissions was likely to result in a direct discharge. Airports reported
various methods for maintaining zero discharge that included evaporation, storage in surface
impoundments, contract hauling, and recycle/recovery of deicing stormwater. The most common
zero discharge method reported was evaporation.
It should be noted, that even though these airports are considered to be zero discharge for
the purpose of this regulation due to the nature of the limited aircraft deicing that occurs, there
may still be direct discharges at these facilities of stormwater that is not associated with aircraft
deicing materials.
Zero Discharge
29%
Indirect
Discharge Only
3%
Direct Discharge
Only
52%
Direct and
Indirect
Discharge
16%
Figure 5-2. Discharge Status of Airports
5.1.2
Airlines
The NAICS code for airlines is 481: Air Transportation. Specific NAICS codes for
respondents to the airline deicing questionnaires are: (1) 481111: Scheduled Passenger Air
Transportation described by the U. S. Census Bureau as establishments primarily engaged in
providing air transportation of passengers or passengers and freight over regular routes and on
regular schedules; and (2) 481112: Scheduled Freight Air Transportation described by the U.S.
Census Bureau as establishments primarily engaged in providing air transportation of cargo
without transporting passengers over regular routes and on regular schedules.
July 2009 5-5
-------
Technical Development Document for Proposed Effluent
Limitation Guidelines and Standards for the Airport Deicing Category
5. Overview of the Industry
The airline data presented in this section are based on the 49 respondents to EPA's airline
detailed questionnaire and additional information from the 70 respondents to EPA's airline
screener questionnaire. Statistics for airlines do not have weighting factors and are based on the
actual number of respondents.
5.1.2.1 Types of Airlines
The four classifications of airlines are major, national, regional, and cargo and are based
on the type of service they offer and their annual revenues. Classification is based on the
economic and financial aspects of their aircraft fleet. Table 5-3 lists the criteria for the four
classifications.
Table 5-3. Airline Classifications
Airline Type
Major
National
Regional:
Large
Medium
Small
(commuters)
Cargo
Annual Revenues
>$100 million
$100 million to $1 billion
$20 million to $100 million
<$20 million
No revenue cut-off
No revenue cut-off
Type of Service
Regular schedules
Regular schedules
Limited to single U.S. region
Scheduled
Scheduled
Scheduled
Scheduled
Aircraft Fleet
Large jets: >60 seats
Payload >18,000 Ibs
Medium and large jets
>60 seats
Lesser or greater than 60 seats
<30 seats
Passenger aircraft with seats
removed
Source: Preliminary Data Summary: Airport Deicing Operations (Revised) (USEPA, 2000).
There were 20 major airlines in the United States in 2006. Many national airlines
typically serve multiple U.S. regions whereas regional airlines are generally limited to a single
region of the country.
Small regional airlines are the largest segment of the regional airline business. Regional
airlines may be private business carriers, commercial airlines, charter airlines, or provide a
combination of these services. Private business carriers represent about 60 percent of regional
airline flights. Regional airlines serve all airports served by major airlines as well as smaller
airports that are not served by any major airline. They typically operate out of one gate area
unless they are affiliates of major airlines and operate at the gates of their affiliate. Regional
airlines conduct a disproportionately large number of flight operations per passenger because
their aircraft are smaller and carry fewer passengers per operation.
EPA administered the airline screener and detailed questionnaires to airlines with greater
than 1,000 departures at specific airports that were selected to receive the airport questionnaire.
The airline screener was distributed to 72 airlines, and 58 airlines received the detailed
questionnaire. The airline detailed questionnaire recipients included major (16), national (17),
regional (6), small or commuter (2), small certified (6), and cargo (1) airlines and requested
information for one or more airport locations. EPA also requested information from 10 foreign
July 2009
5-6
-------
Technical Development Document for Proposed Effluent 5. Overview of the Industry
Limitation Guidelines and Standards for the Airport Deicing Category
airlines that operated at U.S. airports. EPA received responses from 70 screener recipients and 49
detailed questionnaire recipients.
All respondents to the airline detailed questionnaire reported conducting deicing/anti-
icing operations on their aircraft at a total of 57 airport locations.
5.1.2.2 Types of Airline/Airport Relationships
The relationship between airports and airlines regarding deicing operations is one of
dependency and cooperation. Airports and airlines conduct deicing using chemical and
nonchemical methods in the same airfield locations and both contribute pollutants to deicing
stormwater. Airlines may conduct deicing on their own aircraft, deicing for other airlines, and/or
may also use fixed-base operators (FBOs). An airline's deicing methods must be approved by the
FAA for air safety. Airports provide the collection and control systems to contain and/or treat
deicing stormwater generated as a result of aircraft and airfield deicing. However, both airports
and airlines implement pollution prevention practices such as evaluating application rates, using
alternate chemicals, pretreating pavement and aircraft, and manually removing snow and ice to
reduce the quantity of pollutants discharged and the amount of deicing stormwater generated.
5.2 Industry Practices
Airport deicing and anti-icing operations involve chemical and mechanical methods and
are conducted at varied locations and by different entities. This subsection discusses these
practices and pollution prevention methods used by airports and airlines as reported by
respondents to the airport and airline deicing questionnaires.
5.2.1 Airfield Deicing Practices
Airfield pavement deicing/anti-icing removes or prevents the accumulation of frost,
snow, or ice on runways, taxiways, aprons, gates, and ramps. A combination of mechanical
methods and chemical deicing/anti-icing agents are used, and these methods are typically
conducted by airport personnel, FBOs, or private contractors. To reduce the quantity of
pollutants or the amount of deicing stormwater, airports also use various pollution prevention
measures.
Responses to EPA's airport questionnaire indicated that airport personnel (67 percent of
respondents) typically have primary responsibility for airfield pavement deicing/anti-icing. The
remainder of airports reported that a combination of FBO/private contractor (13 percent),
airlines/tenants (9 percent), military (11 percent), and other entities (4 percent) also conduct
deicing. Airlines and tenants may be responsible for deicing/anti-icing their respective gates or
leased areas. Responses to the questionnaire also indicated that most airports use a combination
of mechanical and chemical deicing/anti-icing methods to remove snow or freezing precipitation
(ice or sleet) from their airfield pavements.
5.2.1.1 Chemical Deicing/Anti-Icing
The type of precipitation and temperature affect the volume and type of deicing agents
required for deicing/anti-icing. Common pavement deicing/anti-icing agents used at airports
include potassium acetate, sand, airside urea, sodium acetate, glycol-based fluids, and sodium
July 2009 5-7
-------
Technical Development Document for Proposed Effluent
Limitation Guidelines and Standards for the Airport Deicing Category
5. Overview of the Industry
formate, as reported by respondents to EPA's airport questionnaire. Potassium acetate and
propylene glycol-based fluids were reported as the top deicing/anti-icing chemicals (by weight)
used on airfield pavement during the three winter seasons surveyed by EPA; some respondents
also reported using a mixture of these agents as well as heated sand. Table 5-4 provides national
estimates of the number of airports using these agents. Airports purchase primarily ready-to-
apply rather than concentrated formulations of these chemicals. See Section 6.1 for a detailed
discussion of deicing chemical usage.
Table 5-4. National Estimate of Airports Using Deicing Chemicals or Materials
Deicing/ Anti-Icing
Chemical or
Material
Potassium Acetate
Sand
Airside Urea
Sodium Acetate
Propylene Glycol-
Based Fluids
Sodium Formate
Ethylene Glycol-
Based Fluids
Number of Airports Using Deicing Chemical/Material
2002/2003
94
103
58
39
16
22
6
2003/2004
104
104
60
34
16
1
6
2004/2005
111
98
59
33
16
23
6
Average
103
102
59
35
16
15
6
Percentage of
Airports Using
Deicing Chemical/
Material
31
30
17
10
5
5
2
Source: EPA airport questionnaire responses (scaled to national estimates) (USEPA, 2008c).
5.2.1.2 Mechanical and Nonchemical Deicing/Anti-Icing
Mechanical methods, such as plows, brushes, blowers, and shovels for snow removal, are
the primary forms of airfield pavement deicing and may be used in combination with chemical
methods. One facility uses heated pavement through pavement temperature sensors to prevent
airfield pavement from icing. Of the estimated 215 airports that conduct pavement deicing, EPA
estimates that 212 (99 percent) use mechanical methods on airfield pavement.
5.2.1.3 Pollution Prevention Practices
To reduce the quantity of pollutants discharged and the amount of deicing stormwater
generated, airports implement various pollution prevention practices that control pollution from
airport deicing chemicals (e.g., glycol) thus minimizing pollutant loads through reductions in
chemical usage. Physical snow removal, specialized employee training, and pretreatment of
airfields in advance of precipitation are the most common practices used by airports. The
national estimate of airports implementing one or more pollution prevention practices is 244.
Table 5-5 summarizes EPA's national estimates of the number and percentage of airports that
used airfield pollution prevention practices. See Section 9.1 for detailed descriptions of the
pollution prevention practices used by airports.
July 2009
5-8
-------
Technical Development Document for Proposed Effluent
Limitation Guidelines and Standards for the Airport Deicing Category
5. Overview of the Industry
Table 5-5. Summary of Airfield Pollution Prevention Practices
Pollution Prevention Practice
Physical removal of snow or freezing precipitation
Specialized employee training
Pretreatment of airfield in advance of precipitation
Runway ice detection system
Enhanced weather forecasting
Heated sand
Evaluation of application rates of deicing fluids
Use of alternative chemicals
Use of prewet dry chemical constituents
Other
Estimated Number of
Airports Using Practice
232
153
101
95
77
74
56
40
32
88
Percentage of Airports
Using Practice
69
46
30
28
23
22
17
12
10
26
Source: EPA airport questionnaire database (scaled to national estimates) (USEPA, 2000c).
5.2.2
Aircraft Deicing Practices
Aircraft deicing involves removing frost, snow, or ice from aircraft. Aircraft anti-icing
entails preventing frost, snow, or ice from accumulating on surfaces. Both chemical and
nonchemical- deicing/anti-icing methods are conducted on aircraft at varied airport locations and
by different entities. The FAA also influences aircraft deicing, as it has approval authority for the
deicing/anti-icing practices and procedures selected by the airlines. Airlines also perform
pollution prevention practices similar to airports to reduce the quantity of pollutants discharged
and/or reduce the amount of deicing stormwater generated.
Aircraft deicing may be conducted by an airline, FBO, or private contractor. Often, larger
airline carriers deice their own aircraft and possibly the aircraft of other airlines. In addition, the
entity conducting aircraft deicing for an airline may vary depending on the airport location. All
of the airline questionnaire respondents reported deicing their own aircraft at one or more of their
airport locations. Airline respondents also reported FBOs (84 percent of the airline respondents)
and/or another airline (56 percent of the airline respondents) deiced their aircraft at some of their
airport locations.
Aircraft deicing is conducted at a variety of airport locations and most commonly at
deicing pads and terminal gates and apron areas. Airline respondents reported aircraft deicing at
the following locations including the percent of the airline respondents reporting using a
location:
Deicing pad (80 percent);
Passenger terminal gates/apron areas (78 percent);
Aircraft parking aprons (46 percent);
Airfield ramps (42 percent);
Taxiways (24 percent);
Cargo apron areas (16 percent); and
Other locations (12 percent) (e.g., hangar, etc.).
July 2009
5-9
-------
Technical Development Document for Proposed Effluent 5. Overview of the Industry
Limitation Guidelines and Standards for the Airport Deicing Category
5.2.2.1 Chemical Deicing/Anti-Icing
The type of precipitation and temperature determines the volume and type of deicing
chemicals required to deice/anti-ice aircraft. Two types of aircraft deicing are conducted: wet-
weather and dry-weather. Wet-weather deicing is conducted when snow, sleet, or freezing rain
accumulates on the aircraft. Dry-weather deicing is conducted when frost or ice forms on the
aircraft due to changes in the ambient temperature or when fuel tanks become cooled during
high-altitude flight, and forming ice at lower altitudes and after landing. Dry-weather deicing
requires significantly smaller volumes of ADFs than wet-weather deicing.
Aircraft deicing/anti-icing chemicals are categorized into four classes: Type I, Type II,
Type III, and Type IV. Not all types are currently used. Airlines surveyed by EPA reported
consistently using only Type I and Type IV fluids. ADFs vary by composition and allowable
holdover times (i.e., the amount of time the residual fluid protects aircraft from ice formation).
They generally contain either ethylene glycol or propylene glycol, water, and additives to
remove or prevent ice and snow. Type I ADF is used to remove ice and snow that has
accumulated on aircraft, and Type IV fluids are used for anti-icing to increase holdover times for
an aircraft prior to takeoff. Deicing fluids are usually heated prior to application, while anti-icing
fluids are typically applied at ambient temperatures.
All fluids are usually applied under pressure using a nozzle, often from mobile deicing
trucks (reported by 31 airlines). Below are additional types of ADF application equipment used,
as reported in responses to the airline questionnaire:
• Other equipment (5 respondents) (e.g., brooms, ground sprayer and ladder,
palletized equipment and fork lift, towed tower, small portable unit, self-
contained mobile unit);
• Fixed boom (3 respondents); and
• Handheld bottle/container (1 respondent).
Table 5-6 identifies the type of ADF fluids purchased by airlines that deiced their own
aircraft during the three winter seasons, as well as the average across those seasons, based on the
49 airline respondents to the airline detailed questionnaire. Table 5-7 lists the ADF fluids
purchased by an FBO during the three winter seasons and the average across those seasons, as
reported by 42 airline respondents. As shown in the tables, Type I and Type IV propylene glycol
are the most commonly purchased ADF fluids for deicing aircraft, both by airlines that deice
their own aircraft and airlines that use FBOs.
July 2009 5-10
-------
Technical Development Document for Proposed Effluent
Limitation Guidelines and Standards for the Airport Deicing Category
5. Overview of the Industry
Table 5-6. Deicing/Anti-Icing Chemicals Purchased by Airlines that Deiced Their Own
Aircraft
Deicing/Anti-Icing
Chemical
Type I Propylene Glycol
Type IV Propylene Glycol
Type I Ethylene Glycol
Type IV Ethylene Glycol
Type II Propylene Glycol
Number of Airlines Purchasing Chemicals
2002/2003
29
22
8
6
0
2003/2004
29
22
8
5
0
2004/2005
28
23
8
4
1
Average Number of
Airlines Purchasing
Chemicals
29
22
8
5
1
Source: EPA airline detailed questionnaire database (USEPA, 2008b).
Table 5-7. Deicing/Anti-Icing Chemicals Purchased for Aircraft Deiced by an FBO
Deicing/Anti-Icing
Chemical
Type I Propylene Glycol
Type IV Propylene Glycol
Type I Ethylene Glycol
Type IV Ethylene Glycol
Type II Propylene Glycol
Type II Ethylene Glycol
Number of Airlines for Chemicals Purchased by an
FBO
2002/2003
35
31
17
13
0
0
2003/2004
36
35
13
9
2
0
2004/2005
39
37
13
9
1
1
Average Number of
Airlines for Chemicals
Purchased by an FBO
37
34
14
10
1
1
Source: EPA airline detailed questionnaire database (USEPA, 2008b).
5.2.2.2 Mechanical and Nonchemical Deicing/Anti-Icing
Mechanical and other nonchemical methods used to deice aircraft include brooms, ropes,
hot water, infrared heating, and forced air. Brooms and ropes are not the primary method of
aircraft deicing, especially wet-weather deicing, because they are so time- and labor-intensive,
but rather used in combination with chemical deicing. Forced air/hot air systems are used to blow
or melt snow and ice from aircraft surfaces. Infrared heating deicing systems consist of an open
hangar-type structure with infrared generators suspended from the ceiling. The infrared
wavelengths are targeted to heat ice and snow, and minimize heating of aircraft components.
This system reduces the volume of ADF fluid required, but cannot provide anti-icing protection.
Aircraft may also be stored in a hangar to prevent snow or ice from accumulating if a storm
event is predicted. The most common methods of deicing/anti-icing used by airline questionnaire
respondents are mechanical methods and hangar storage. Table 5-8 and Table 5-9 summarize the
use of these methods for deicing/anti-icing aircraft by airlines and by FBOs, respectively.
July 2009
5-11
-------
Technical Development Document for Proposed Effluent
Limitation Guidelines and Standards for the Airport Deicing Category
5. Overview of the Industry
Table 5-8. Summary of Mechanical and Nonchemical Aircraft Deicing Methods Used by an
Airline
Mechanical/Nonchemical
Method
Mechanical (e.g., brooms, ropes)
Hangar storage
Forced air
Hot water
Infrared heating
Number of Airlines
2003/2004
22
15
9
5
1
2003/2004
21
16
7
4
1
2004/2005
21
16
7
4
1
Average Number
of Airlines
21
16
8
4
1
Source: EPA airline detailed questionnaire database (USEPA, 2008b).
Table 5-9. Summary of Mechanical and Nonchemical Aircraft Deicing Methods Used By
FBOs
Mechanical/Nonchemical
Method
Mechanical (e.g., brooms, ropes)
Hangar storage
Forced air
Hot water
Infrared heating
Number of Airlines
2003/2004
10
5
4
3
0
2003/2004
10
5
5
4
0
2004/2005
10
5
5
4
0
Average Number
of Airlines
10
5
5
4
0
Source: EPA airline detailed questionnaire database (USEPA, 2008b).
5.2.2.3 Pollution Prevention Practices
To reduce the quantity of pollutants discharged and the amount of deicing stormwater
generated, airlines implement pollution prevention practices at various airports. These practices
control pollution from aircraft deicing chemicals and minimize pollutant loads. Specialized
training, implementation of a pollution prevention policy, and physical snow removal are the
most common pollution prevention practices used by airlines. Table 5-10 summarizes airline
pollution prevention practices reported in response to the airline detailed questionnaire. See
Section 9 for detailed descriptions of the pollution prevention practices used by airlines.
July 2009
5-12
-------
Technical Development Document for Proposed Effluent
Limitation Guidelines and Standards for the Airport Deicing Category
5. Overview of the Industry
Table 5-10. Summary of Aircraft Pollution Prevention Practices
Pollution Prevention Practice
Specialized employee training
Instituting pollution prevention policy
Physical removal of snow or freezing precipitation
Overnight pretreatment/storage of aircraft
Custom fluid blending
Enhanced weather forecasting
Evaluation of application rates of deicing fluids
Pretreating aircraft with hot water
Use of alternative chemicals
Other
Number of Airlines
Reporting Practice
43
43
31
30
27
25
24
9
2
30
5.2.3
Source: EPA airline detailed questionnaire database (USEPA, 2008b).
Airport Deicing Stormwater Collection and Control
Deicing and anti-icing operations are conducted at multiple locations of the airfield, and
the fluids are widely dispersed during and after application via ramp runoff, taxiway drippage,
and residual on aircraft. Deicing stormwater is contained and collected by implementing
designated deicing areas, stormwater drainage systems, glycol recovery, storage tanks,
containment ponds, and plug and pump systems. Typical sources of deicing stormwater are:
Terminal gates and aprons/areas;
Aircraft deicing pads;
Taxiways;
Airfield ramps;
Runways;
Cargo apron areas;
Maintenance hangar ramps;
Aircraft parking areas;
Military bases; and
ADF-contaminated snow dumps.
Table 5-11 summarizes the collection and control methods used by airports. Based on
responses to the airport questionnaire, an estimated 246 U.S. airports use containment, collection
and/or conveyance measures to control the discharge of deicing stormwaters to surface waters
and/or POTWs. Stormwater drainage systems and containment ponds and basins are used by
most of the airports. See Section 9.0 for detailed discussions of deicing stormwater collection
and control methods used by the airport deicing category.
July 2009
5-13
-------
Technical Development Document for Proposed Effluent
Limitation Guidelines and Standards for the Airport Deicing Category
5. Overview of the Industry
Table 5-11. Summary of Airport Collection, Containment, and Conveyance Methods
Collection/Containment/Conveyance Method
Stormwater drainage system
Containment pond/basin
Aboveground/underground Tank
Glycol recovery vehicles/sweepers
Other (vegetated swales, snow melters, absorbant)
Plug and pump
Estimated Number of
Airports
211
121
57
54
34
29
Percentage of
Airports
63
36
17
16
10
9
Source: EPA airport questionnaire database (Scaled to national estimate) (USEPA, 2008c).
EPA estimates that approximately half (46 percent) of the U.S. airports with deicing
operations also operate systems to treat or recover their deicing stormwater. The treatment and
recovery technologies reported by airports include equalization (46 percent), oil/water separation
(5 percent), sand or other media filtration (4 percent), membrane separation (1 percent), and
biological treatment (1 percent). Airports reporting other types of treatment technologies
comprise 8 percent of the population, including mechanical vapor recompression (MVR),
aeration, and distillation. These technologies are described in detail in Section 9.
5.3
References
USDOT. 2008a. Report to Congress: National Plan of Integrated Airport Systems (NPIAS)
2009-2013. U.S. Department of Transportation, Federal Aviation Administration. Washington,
D.C. http://www.faa.gov/airports_airtraffic/airports/planning_capacity/npias/reports/
USDOT. 2008b. FAA Regional Offices. U.S. Department of Transportation, Federal Aviation
Administration. Washington, D.C.
www.faa.gov/about/office org/headquarters offices/arp/regional offices/
USEPA. 2008a. Airline Screener Deicing Questionnaire Database. U.S. Environmental
Protection Agency/Office of Water. Washington, D.C. DCN AD00937.
USEPA. 2008b. Airline Detailed Deicing Questionnaire Database. U.S. Environmental
Protection Agency/Office of Water. Washington, D.C. DCN AD00938.
USEPA. 2008c. Airport Deicing Questionnaire Database. U.S. Environmental Protection
Agency/Office of Water. Washington, D.C. DCN AD00927.
USEPA. 2000. Preliminary Data Summary: Airport Deicing Operations. EPA-821-R-00-016,
U.S. Environmental Protection Agency/Office of Water. Washington, D.C.
http://www.epa.gov/guide/airport. DCN AD00005
USEPA. 2005. Supporting Statement: Survey of Airport Deicing Operations. U.S. Environmental
Protection Agency/Office of Water. Washington, D.C. DCN AD00447.
July 2009
5-14
-------
Technical Development Document for Proposed Effluent 6. Deicing Chemical Use and
Limitation Guidelines and Standards for the Airport Deicing Category Deicing Stormwater Characterization
6. DEICING CHEMICAL USE AND DEICING STORMWATER CHARACTERIZATION
This section summarizes EPA's estimate of the amount of airfield and aircraft
deicing/anti-icing chemicals currently in use by U.S. commercial airports and provides
information on deicing stormwater pollutant characteristics to the extent possible. Deicing
stormwater discharges are "weather-dependent" and they are, by nature, highly variable. In
addition, deicing chemical disposition after its intended use is not fully understood. These
chemicals may be lost to evaporation, dispersion, or soil absorption; collected; or released into
the environment and discharged from the airport. Limited information is currently available on
each of these disposition methods. This section presents the information EPA has collected on
the types of pollutants present in deicing stormwater and their ranges of concentrations.
6.1 Deicing Chemical Usage
As discussed in Section 5, several deicing chemicals are commonly used at U.S.
commercial airports. These chemicals are used for either airfield or aircraft deicing/anti-icing
and their usages are described below.
6.1.1 Airfield Chemical Use
Pavement deicing/anti-icing removes or prevents frost, snow, or ice from accumulating
on runways, taxiways, aprons, gates, and ramps. Airports use mechanical and chemical methods
for this purpose, more often using mechanical methods. Because ice, sleet, and snow may be
difficult to remove by mechanical methods alone, many airports also use sand and/or chemical
deicing agents such as potassium acetate, sodium acetate, sodium formate, glycol-based
products, or urea. Based on the data collected by EPA in the 2006 Airport Deicing Questionnaire
(airport questionnaire), the most common airfield deicing chemical currently used by U.S.
airports is potassium acetate (approximately 64 percent of airfield chemical usage by weight). In
addition, there is a trend by U.S. airports to cease or decrease their use of urea for airfield deicing
due to concerns with water quality impacts from its discharge.
Table 6-1 lists total airfield chemical usage as reported by the 150 U.S. airports that
responded to the airport questionnaire for the 2002/2003, 2003/2004, and 2004/2005 deicing
seasons. Ninety airports reported use of airfield deicing chemicals, approximately thirty eight
airports reported no airfield deicing (including no sand use), while the remainder reported only
sand use or unknown chemical use.
July 2009 6-1
-------
Technical Development Document for Proposed Effluent
Limitation Guidelines and Standards for the Airport Deicing Category
6. Deicing Chemical Use and
Deicing Stormwater Characterization
Table 6-1. Surveyed U.S. Commercial Airports - Airfield Chemical Usage
Chemical
Potassium acetate
Propylene glycol-
based fluids
Airside urea
Sodium acetate
Sodium formate
Ethylene glycol-
based fluids
2002/2003 Total
Airport Usage
(tons/year)
22,804
3,317
3,015
2,815
1,663
1,038
2003/2004 Total
Airport Usage
(tons/year)
20,267
4,147
3,804
3,195
696
465
2004/2005 Total
Airport Usage
(tons/year)
20,029
2,884
4,031
2,663
1,359
691
Average Total
Airport Usage
(tons/year)
21,292
3,870
3,553
3,072
1,064
656
Percentage of
Chemical
Usage
64
12
11
9
3
2
Source: EPA Airport Questionnaire, 2002-2005 deicing seasons (USEPA, 2008a).
Note: This table is based on data from 90 airports that reported airfield deicing chemical usage. The three-year
average is not a straight average of the total annual amounts; the average for each airport was evaluated and
calculated separately.
Using EPA's survey statistical weighting factors, the total usage by chemical can be
scaled up to estimate a national airfield chemical usage for all commercial U.S. airports (see
Table 6-2).
Table 6-2. U.S. Commercial Airports - National Estimate of Airfield Chemical Usage
Chemical
Potassium acetate
Propylene glycol-based fluids
Airside urea
Sodium acetate
Sodium formate
Ethylene glycol-based fluids
Estimated Total Airport Usage
(tons/year)
22,538
3,883
4,127
3,100
1,117
774
Percentage of
Chemical Usage
63
11
12
9
3
2
6.1.2
Aircraft Chemical Use and Purchasing Patterns
There are four types of aircraft deicing fluids (ADFs) manufactured around the world,
referred to by the aviation and chemical industries as Types I through IV. Of these, Type I and
Type IV are commonly used at U.S. commercial airports. Type I ADF is used to defrost and
deice aircraft and Type IV ADF is used to prevent icing from reoccurring (anti-icing) after initial
deicing with a Type I ADF. ADFs contain a primary freezing point depressant (typically
propylene glycol (PG) or ethylene glycol (EG)) and other additives. ADFs work by adhering to
aircraft surfaces to remove and/or prevent snow and ice accumulation by virtue of their
depressed freezing points. Airports conduct two types of deicing: dry weather deicing to remove
frost and wet weather deicing and anti-icing during precipitation such as snow, sleet (ice pellets),
or freezing rain. Airports may also perform dry weather deicing on some types of aircraft whose
fuel tanks become super-cooled during high-altitude flight, resulting in formation of frost/ice on
aircraft wings at lower altitudes and after landing. Based on the data collected by EPA in the
2006 Airline Deicing Questionnaire (airline questionnaire), the most common ADF is Type I
July 2009
6-2
-------
Technical Development Document for Proposed Effluent 6. Deicing Chemical Use and
Limitation Guidelines and Standards for the Airport Deicing Category Deicing Stormwater Characterization
PG-based fluids (approximately 77 percent of ADF usage/purchase). In addition, U.S. airports
have been trending towards greater use of PG-based fluids and less use of EG-based fluids.
Table 6-3 presents a national estimate of the average aircraft chemical usage/purchase by
U.S. commercial airports by type of fluid.
Table 6-3. U.S. Commercial Airports - National Estimate of Aircraft Chemical
Usage/Purchase l
Chemical
Type I PG ADF
Type IV PG ADF
Type I EG ADF
Type IV EG ADF
Average Total Airport
Usage/Purchase
(million gallons/year) 2
19.305
2.856
2.575
0.306
Percentage of Chemical Usage
77.1
11.4
10.3
1.2
Sources: EPA Airline Questionnaire (USEPA, 2008b); Airport Deicing Loadings Database (USEPA, 2008c).
1 EPA used the ADF purchase information to represent usage, per airline industry recommendations.
2 Total gallons normalized to 100% PG/EG.
6.2 Deicing Stormwater Characterization
EPA evaluated data from a variety of sources to better understand the components of
deicing chemicals and ADFs that end up in deicing Stormwater. These data include information
on the additives included in ADFs, data collected during sampling of concentrated and diluted
ADFs used at the Detroit Metropolitan Wayne County (DTW) and Minneapolis/St. Paul
International (MSP) airports during the 2003/2004 deicing season, data collected during
sampling of deicing Stormwater at the Albany International (ALB), Pittsburgh International
(PIT), Denver International (DIA), and Greater Rockford (RFD) airports during the 2004/2005
deicing season, current data for airports included in the PCS database, and deicing Stormwater
data collected by EPA during the Preliminary Data Study, through site visits, or through industry
or permit authority submissions. Section 8 summarizes the types of pollutants found in deicing
Stormwater based on these sources.
6.2.1 Airfield Deicing Chemicals and Associated Deicing Stormwater
Most solid airfield deicing chemical products are composed of a freezing point depressant
(e.g., potassium acetate, sodium acetate) and minimal additives (e.g., corrosion inhibitors).
Liquid airfield deicing chemical products are composed of a freezing point depressant (e.g.,
potassium acetate, propylene glycol), water, and minimal additives. The airfield deicing products
that include salts (i.e., potassium acetate, sodium acetate, and sodium formate) will all ionize in
water, creating positive salt ions (K+, Na+) and BOD load as the acetate or formate ion degrades
into carbon dioxide (CC^) and water.
Urea is typically applied to pavement and runway areas in granular form. Urea degrades
by hydrolysis to CC>2 and ammonia, which can be toxic to aquatic organisms even at very low
concentrations. Once ammonia is formed, it either remains in solution as ammonia or its ionized
form (NH4+), biologically converts to other nitrogen forms (e.g., NOs or N2), or volatilizes to the
July 2009 6-3
-------
Technical Development Document for Proposed Effluent 6. Deicing Chemical Use and
Limitation Guidelines and Standards for the Airport Deicing Category Deicing Stormwater Characterization
air. The formation of ammonia is dependent on the pH and temperature of the receiving water.
The higher the pH and temperature, the more ammonia is formed. Another potentially toxic
byproduct of urea degradation is nitrous acid, which reacts with secondary amines to form
nitrosamines, many of which are known carcinogens.
EPA has limited data on airfield deicing stormwater that do not also contain aircraft
deicing area stormwater. Most of EPA's stormwater data include both airfield and aircraft
deicing components. However, during sampling at Detroit Metro, EPA collected samples from
the airport's runway and open area ponds (Pond 3 East and Pond 6). These ponds do not contain
aircraft deicing area stormwater since a separate pond collects runoff from the gate and deicing
pad areas where the stormwater is expected to contain ADF. Detroit sometimes uses sand and
potassium acetate for runway traction and deicing in addition to their usual mechanical snow
removal equipment. It does not use urea-based deicers. Table 6-4 presents the sampling data
from the two airfield/open area runoff ponds.
6.2.2 Aircraft Deicing Chemicals and Associated Deicing Stormwater
ADFs contain a primary freezing point depressant (typically PG or EG) and additives.
Typical additives are thickening agents, wetting agents, corrosion inhibitors, buffer, and dye,
which make up 1 to 4 percent of the fluid mass. Type IV fluids have higher concentrations of the
freezing point depressant and greater viscosity so that the fluid stays on the aircraft until take-off.
The actual composition of ADFs varies and information on specific additive compounds
is usually considered proprietary by ADF manufacturers. EPA believes that typical ADFs most
likely include the following components:
ADF Component
Propylene glycol or ethylene glycol
Surfactant/wetting agent
Corrosion inhibitor/flame retardant
pH buffer
Dyes
Water
Composition (%)
50-88
About 0.5
About 0.5
About 0.25
<1
Remainder
Source: Environmental Impact and Benefit Assessment for Proposed Effluent Limitation Guidelines and Standards
for the Airport Deicing Category (USEPA. 2009)
July 2009 6-4
-------
Technical Development Document for Proposed Effluent
Limitation Guidelines and Standards for the Airport Deicing Category
6. Deicing Chemical Use and
Deicing Stormwater Characterization
Table 6-4. EPA's Analytical Results for Pond 3E Effluent and Pond 6 Effluent, DTW
Analyte
5-Day Biochemical Oxygen Demand (BOD5)
Chloride
Chemical Oxygen Demand (COD)
Nitrate/Nitrite (NO2 + NO3-N)
Sulfate
Total Dissolved Solids (TDS)
Total Kjeldahl Nitrogen (TKN)
Total Organic Carbon
Total Phosphorus
Total Suspended Solids (TSS)
Aluminum
Aluminum, Dissolved
Barium
Barium, Dissolved
Calcium
Calcium, Dissolved
Chromium :
Copper :
Iron
Iron, Dissolved
Magnesium
Magnesium, Dissolved
Manganese
Manganese, Dissolved
Molybdenum
Molybdenum, Dissolved
Sodium
Sodium, Dissolved
Titanium
Zinc1
Acetone
Propylene Glycol - 1671 2
Propylene Glycol - 8015D 2
Unit
mg/L
mg/L
mg/L
mg/L
mg/L
mg/L
mg/L
mg/L
mg/L
mg/L
ug/L
ug/L
ug/L
ug/L
ug/L
ug/L
Ug/L
ug/L
Ug/L
ug/L
ug/L
ug/L
Ug/L
Ug/L
ug/L
Ug/L
ug/L
Ug/L
ug/L
ug/L
ug/L
mg/L
mg/L
Pond 3 East Effluent
146
855
273
0.0400
50.1
1,790
1.51
813
0.340
150
1,660
ND (50.0)
92.3
77.5
96,600
91,000
ND (10.0)
ND (10.0)
3,630
407
20,300
18,200
508
462
ND (10.0)
ND (10.0)
547,000
522,000
26.2
41.4
ND (50.0)
ND (10.0)
ND (10.0)
Pond 6 Effluent
43.0
315
111
0.110
51.0
833
0.990
314
0.280
149
2,110
64.5
81.8
66.5
69,400
63,500
ND (10.0)
ND (10.0)
4,390
600
17,400
15,700
411
335
ND (10.0)
ND (10.0)
191,000
189,000
34.8
43.8
ND (50.0)
ND (10.0)
ND (10.0)
Source: Final Sampling Episode Report Detroit Metropolitan Wayne County International Airport (DTW) (USEPA,
2006a).
1 Pollutant listed by EPA as a priority pollutant. See 40 CFR Part 423, Appendix A.
2Number following analyte name refers to analytical method. 1671 is a Clean Water Act method (USEPA, 1998)
and 8015D is a hazardous waste method promulgated under the Resource Conservation and Recovery Act (RCRA)
(USEPA, 2003).
ND - Not detected (number in parentheses is reporting limit).
July 2009 6-5
-------
Technical Development Document for Proposed Effluent 6. Deicing Chemical Use and
Limitation Guidelines and Standards for the Airport Deicing Category Deicing Stormwater Characterization
Despite limited public information, EPA has identified three main classes of additives
widely used among ADF manufacturers. Benzotriazole (BT) and methyl-substituted
benzotriazole (MeBT) are corrosion inhibitor/flame retardants that reduce flammability from
corrosion of metal components carrying a direct current. Alkylphenol/alkylphenol ethoxylates
(AP/APEO) are nonionic surfactants widely used to reduce surface tension in aircraft deicers.
Finally, triethanolamine is a pH buffer. (See Aircraft Deicing and Anti-icing Fluids Fate,
Transport and Environmental Impacts (ERG, 2007). EPA also has information indicating that
high molecular weight, nonlinear polymers may be used as thickening agents in ADFs (see
Aircraft Deicing Fluids (ERG, 2007b) and various classes of dyes can be used to color the ADF.
The classes of dyes identified as potentially used in ADFs include azo, xanthene, triphenyl
methane, and anthroquinone dyes (see Questions Regarding Pylam Dye Use in ADF (ERG,
2007a).
Analyses conducted by U.S. Geological Survey (USGS) at General Mitchell International
airport in Milwaukee, WI, and EPA's sampling programs, have confirmed the presence of
glycols, triazole compounds, and alkylphenol compounds in deicing stormwater. EPA collected
deicing stormwater samples at MSP and DTW during the 2004/2005 winter season. At MSP,
EPA collected samples of deicing stormwater from segregated high concentration and low
concentration storage tanks. At DTW, Northwest collected its deicing stormwater from a
March 24, 2005 deicing event into a portable "firac" tank, which was then sampled by EPA.
Table 6-5 lists the constituents detected in these deicing stormwaters and their concentrations.
During the 2005/2006 deicing season, EPA collected five consecutive days of samples of
influent to and effluent from deicing stormwater treatment at ALB, PIT, DEN, and RFD. The
sampled deicing stormwater at these airports, prior to treatment, shows a wide range of
constituents and constituent concentrations between the airports, as shown in Tables 6-6 and 6-7.
Where, the five day average for each pollutant at DEN, PIT, and ALB are used for comparison
between the airports.
July 2009 6-6
-------
Technical Development Document for Proposed Effluent
Limitation Guidelines and Standards for the Airport Deicing Category
6. Deicing Chemical Use and
Deicing Stormwater Characterization
Table 6-5. MSP and DTW Grab Sample Data Summary for Collected Deicing Stormwater
Analyte
Classical Pollutants
Ammonia as Nitrogen (NH3-N)
BOD5
Chloride
COD
Hexane Extractable Material (HEM)
Nitrate/Nitrite (NO2 + NO3-N)
Silica Gel Treated HEM (SGT-HEM)
Sulfate
TDS
TKN
Total Organic Carbon
Total Phosphorus
Total Recoverable Phenolics
TSS
Unit
mg/L
mg/L
mg/L
mg/L
mg/L
mg/L
mg/L
mg/L
mg/L
mg/L
mg/L
mg/L
mg/L
mg/L
MSP
High
Concentration
Storage Tank
ND (0.05)
115,000
45.0
358,000
50.0
0.0950
17.0
21.2
1,370
13.5
96,100
6.49
0.150
89.0
MSP
Low
Concentration
Storage Tank
ND (0.05)
8,000
27.0
16,000
ND (5.00)
O.0600
ND (5.00)
13.6
559
5.61
5,660
<2.10
0.0375
19.5
DTW
Northwest Frac
Tank
0.790
140,000
25.0
332,000
22.0
0.240
ND (6.00)
20.3
1,440
71.1
93,100
0.320
< 0.007
11.5
Total and Dissolved Metals
Aluminum
Aluminum, Dissolved
Antimony, Dissolved :
Barium
Barium, Dissolved
Calcium
Calcium, Dissolved
Copper :
Copper, Dissolved :
Iron
Iron, Dissolved
Magnesium
Magnesium, Dissolved
Manganese
Manganese, Dissolved
Mercury :
Mercury, Dissolved :
ug/L
ug/L
Ug/L
ug/L
Ug/L
ug/L
Ug/L
ug/L
Ug/L
ug/L
ug/L
Ug/L
ug/L
Ug/L
ug/L
Ug/L
ug/L
525
ND (500)
201
114
36.4
68,200
59,600
ND (100)
ND (100)
11,000
4,960
9,230
8,490
887
756
ND(40)
ND(40)
508
136
ND (20.0)
67.1
61.9
35,200
34,500
37.6
16.4
7,470
6,030
4,250
4,080
317
308
ND(2)
ND(2)
ND (500)
ND (500)
ND (200)
52.4
46.9
127,000
125,000
ND (100)
ND (100)
1,410
1,370
12,900
13,000
433
423
45.1
68.7
Sources: Final Sampling Episode Report Minneapolis/St. Paul International Airport (MSP) (USEPA, 2006b); Final
Sampling Episode Report Detroit Metropolitan Wayne County International Airport (DTW) (USEPA, 2006a).
1 Pollutant listed by EPA as a priority pollutant. See 40 CFR Part 423, Appendix A.
EXCLUDE - Data excluded from the data set (see data review narratives in Appendix C for details).
< - Average result includes at least one nondetect value.
ND - Not detected (number in parenthesis is reporting limit).
July 2009 6-7
-------
Technical Development Document for Proposed Effluent
Limitation Guidelines and Standards for the Airport Deicing Category
6. Deicing Chemical Use and
Deicing Stormwater Characterization
Table 6-5 (Continued)
Analyte
Molybdenum
Molybdenum, Dissolved
Sodium
Sodium, Dissolved
Tin
Tin, Dissolved
Titanium
Zinc1
Zinc, Dissolved :
Unit
ug/L
ug/L
ug/L
ug/L
Ug/L
ug/L
Ug/L
ug/L
Ug/L
MSP
High
Concentration
Storage Tank
19,100
19,000
48,700
48,100
611
616
ND (100)
492
444
MSP
Low
Concentration
Storage Tank
794
771
18,700
18,600
32.1
32.5
13.5
291
277
DTW
Northwest Frac
Tank
15,900
16,000
22,800
19,200
673
646
ND (100)
119
119
Volatile and Semivolatile Organics
Acetone
Propylene Glycol -167 '1 2
Propylene Glycol - 8015D 2
Trichloroethene :
ug/L
mg/L
mg/L
ug/L
1,440
—
193,000
ND(10)
23,700
—
8,600
ND(10)
3,340
192,000
170,000
14.5
Sources: Final Sampling Episode Report Minneapolis/St. Paul International Airport (MSP) (USEPA, 2006b); Final
Sampling Episode Report Detroit Metropolitan Wayne County International Airport (DTW) (USEPA, 2006a).
1 Pollutant designated by EPA as a priority pollutant in 40 CFR Part 423.
2Number following analyte name refers to analytical method. 1671 is a Clean Water Act method (USEPA, 1998)
and 8015D is a hazardous waste method promulgated under the Resource Conservation and Recovery Act (RCRA)
(USEPA, 2003).
EXCLUDE - Data excluded from the data set (see data review narratives in Appendix C for details).
< - Average result includes at least one nondetect value.
ND - Not detected (number in parenthesis is reporting limit).
July 2009 6-8
-------
Technical Development Document for Proposed Effluent
Limitation Guidelines and Standards for the Airport Deicing Category
6. Deicing Chemical Use and
Deicing Stormwater Characterization
Table 6-6. DEN, PIT, and ALB - 5-Day Average Data Summary for Untreated Deicing
Stormwater
Analyte
Alkalinity
Ammonia As Nitrogen (NH3-N)
BOD5
COD
Chloride
Hardness
HEM
Nitrate/Nitrite (NO3-N + NO2-N)
Sulfate
TDS
TKN
Total Organic Carbon (TOC)
Total Orthophosphate
Total Phosphorus
Total Recoverable Phenolics
TSS
Arsenic
Barium
Boron
Calcium
Copper
Iron
Magnesium
Manganese
Molybdenum
Selenium
Sodium
Tin
Zinc
Acetone
Benzoic Acid
Methyl Ethyl Ketone
Phenol
EthyleneGlycol-16711
Units
mg/L
mg/L
mg/L
mg/L
mg/L
mg/L
mg/L
mg/L
mg/L
mg/L
mg/L
mg/L
mg/L
mg/L
mg/L
mg/L
ug/L
ug/L
ug/L
Ug/L
Mg/L
Ug/L
Ug/L
ug/L
Mg/L
ug/L
ug/L
ug/L
ug/L
ug/L
ug/L
ug/L
ug/L
mg/L
DEN Airport Effluent
from Equalization
Feed Tank
5-day Average
706
0.448
149,000
247,000
120
362
9.20
0.0266
60.0
NC
6.41
89,000
<1.03
<2.76
0.0608
<17.8
<81.8
<13.2
<723
103,000
305
1,210
5,360
156
11,900
172
254,000
<258
<81.1
4,100
716
ND(50)
ND (100)
<167
PIT Airport
Influent to RO
Unit
5-day Average
481
ND (0.05)
16,600
28,300
11.6
542
ND (6.0)
0.0204
48.1
1,670
9.04
7,720
O.0196
0.0778
0.0187
<8.40
12.7
103
532
155,000
ND(10)
5,870
6,260
532
ND(10)
31.8
54,300
41.0
71.8
10,900
ND(50)
ND(50)
ND (100)
<65.6
ALB Airport Influent to
Anaerobic Treatment
System
5-day Average
159
O.262
3,400
5,350
90.0
248
ND (5.0)
0.0284
26.4
650
1.61
1,570
0.115
0.946
ND (0.005)
16.6
ND (10)
42.5
ND (100)
48,300
ND (10)
6,270
9,990
736
ND (10)
<5.38
89,600
ND(30)
48.3
15,400
278
<58.5
24.5
ND (10)
Sources: Draft Sampling Episode Report Denver International Airport (USEPA, 2006c); Draft Sampling Episode
Report Pittsburgh International Airport (USEPA, 2006d); Draft Sampling Episode Report Albany International
Airport (USEPA, 2006e).
ND - Not detected (number in parentheses is reporting limit).
NC - Not collected.
July 2009
6-9
-------
Technical Development Document for Proposed Effluent
Limitation Guidelines and Standards for the Airport Deicing Category
6. Deicing Chemical Use and
Deicing Stormwater Characterization
Table 6-6 (Continued)
Analyte
Ethylene Glycol - 8015D 1
Propylene Glycol - 1671 l
Propylene Glycol - 8015D J
Tolyltriazole
Nonylphenol, total
Nony Iphenol- 1 -Ethoxylate
Nonylphenol-2-Ethoxylate
Nony lphenol-3 -Ethoxylate
Nonylphenol-4-Ethoxylate
Nony lphenol-5 -Ethoxylate
Nonylphenol-6-Ethoxylate
Nonylphenol-7 -Ethoxylate
Nonylphenol-8-Ethoxylate
Nonylphenol-9-Ethoxylate
Nonylphenol-10-Ethoxylate
Nony Iphenol- 1 1 -Ethoxylate
Nonylphenol-12-Ethoxylate
Nonylphenol- 1 3 -Ethoxylate
Nonylphenol-14-Ethoxylate
Nonylphenol- 1 5 -Ethoxylate
Nonylphenol-16-Ethoxylate
Nonylphenol-17-Ethoxylate
Nonylphenol- 1 8-Ethoxylate
Octy Iphenol
Octylphenol-2 -Ethoxylate
Octy lphenol-3 -Ethoxylate
Octy lphenol-4-Ethoxy late
Octy lphenol-5-Ethoxy late
Octy lphenol-6-Ethoxy late
Octy lphenol-7-Ethoxy late
Octy lphenol-8-Ethoxy late
Octy lphenol-9-Ethoxy late
Octy Iphenol- 1 0-Ethoxy late
Units
mg/L
mg/L
mg/L
ug/L
ug/L
ug/L
Ug/L
ug/L
Ug/L
Ug/L
ug/L
Ug/L
ug/L
Ug/L
ug/L
Ug/L
ug/L
ug/L
ug/L
ug/L
ug/L
ug/L
ug/L
ug/L
ug/L
ug/L
Mg/L
Mg/L
Mg/L
Mg/L
Mg/L
Mg/L
Mg/L
DEN Airport Effluent
from Equalization
Feed Tank
5-day Average
<172
174,000
173,000
10,100
ND (5.0)
ND (7.4)
ND (21.0)
17.8
16.4
21.5
50.4
60.7
86.2
79.1
92.5
100
216
167
116
69.9
43.4
23.3
12.4
<8.80
71.8
1,460
1,260
891
441
198
116
44.6
22.5
PIT Airport
Influent to RO
Unit
5-day Average
<73.6
15,700
15,900
7,860
22.2
130
190
59.9
15.4
213
403
619
841
942
1,050
1,040
833
589
386
222
107
53.5
23.3
ND (0.01)
ND (0.144)
4.38
ND (2.26)
ND (2.93)
ND (2.69)
ND (2.58)
ND(1.85)
ND (0.636)
ND (0.636)
ALB Airport Influent to
Anaerobic Treatment
System
5-day Average
ND (10)
2,570
2,630
325
ND (12.0)
ND (19.0)
ND (53.0)
3.90
3.01
5.70
12.5
15.4
24.7
24.7
38.1
40.4
33.7
25.2
17.8
8.80
4.69
2.27
1.09
ND (2.00)
0.159
2.66
ND (2.26)
ND (2.93)
ND (2.69)
ND (2.58)
ND (1.85)
ND (0.636)
ND (0.636)
Sources: Draft Sampling Episode Report Denver International Airport (USEPA, 2006c); Draft Sampling Episode
Report Pittsburgh International Airport (USEPA, 2006d); Draft Sampling Episode Report Albany International
Airport (USEPA, 2006e).
ND - Not detected (number in parentheses is reporting limit).
1 Number following analyte name refers to analytical method. 1671 is a Clean Water Act method (USEPA, 1998)
and 8015D is a hazardous waste method promulgated under the Resource Conservation and Recovery Act (RCRA)
(USEPA, 2003).
July 2009 6-10
-------
Technical Development Document for Proposed Effluent
Limitation Guidelines and Standards for the Airport Deicing Category
6. Deicing Chemical Use and
Deicing Stormwater Characterization
Table 6-6 (Continued)
Analyte
Octylphenol-1 1-Ethoxylate
Octy Iphenol- 1 2-Ethoxy late
Total Nonylphenol-3-Ethoxylate-
Nonlyphenol- 18-
Total Octylphenol-2-Ethoxylate-
Octy Iphenol- 1 2-Ethoxy late
Units
Hg/L
Hg/L
ug/L
Hg/L
DEN Airport Effluent
from Equalization
Feed Tank
5-day Average
12.0
8.08
1,170
4,530
PIT Airport
Influent to RO
Unit
5-day Average
ND (0.267)
ND(0.113)
7,400
ND (16.0)
ALB Airport Influent to
Anaerobic Treatment
System
5-day Average
ND (0.267)
ND (0.113)
260
ND (16.0)
Sources: Draft Sampling Episode Report Denver International Airport (USEPA, 2006c); Draft Sampling Episode
Report Pittsburgh International Airport (USEPA, 2006d); Draft Sampling Episode Report Albany International
Airport (USEPA, 2006e).
ND - Not detected (number in parentheses is reporting limit).
1 Number following analyte name refers to analytical method. 1671 is a Clean Water Act method (USEPA, 1998)
and 8015D is a hazardous waste method promulgated under the Resource Conservation and Recovery Act (RCRA)
(USEPA, 2003).
July 2009 6-11
-------
Technical Development Document for Proposed Effluent
Limitation Guidelines and Standards for the Airport Deicing Category
6. Deicing Chemical Use and
Deicing Stormwater Characterization
Table 6-7. RFD - 1-Day Data Summary for Untreated Deicing Stormwater
Analyte
Alkalinity
Ammonia As Nitrogen (NH3-N)
BOD5
COD
Chloride
Hardness
Nitrate/Nitrite (NO3-N + NO2-N)
Sulfate
TDS
TKN
TOC
Total Phosphorus
TSS
Barium
Calcium
Iron
Magnesium
Manganese
Sodium
Acetone
Methyl Ethyl Ketone
Propylene Glycol - 8015D 1
Tolyltriazole
Bisphenol A
N-Nonylphenol-2-Ethoxylate
N-Nonylphenoxyl-2-Carboxylic Acid
Octylphenol-9-Ethoxylate
Units
mg/L
mg/L
mg/L
mg/L
mg/L
mg/L
mg/L
mg/L
mg/L
mg/L
mg/L
mg/L
mg/L
ug/L
ug/L
Ug/L
Ug/L
ug/L
Ug/L
ug/L
Ug/L
mg/L
Ug/L
ng/L
NC
NC
ug/L
Influent to Aerobic Treatment
System, Spring
1,030
59.6
603
646
14.0
112
0.0190
5.65
384
82.8
137
0.330
85.0
20.3
6,600
108
14,700
164
4,790
86.6
136
31.0
45.3
ND (12,000)
50.0
41.0
ND(3.18)
Source: Draft Sampling Episode Report Greater Rockford Airport, Sampling Episode 6529 April 20, 2006 and
Sampling Episode 6530 August 29, 2006 (USEPA, 2006f).
ND - Not detected (number in parentheses is reporting limit).
NC - Not collected.
1 Number following analyte name refers to analytical method. 1671 is a Clean Water Act method (USEPA, 1998)
and 8015D is a hazardous waste method promulgated under the Resource Conservation and Recovery Act (RCRA)
(USEPA, 2003).
July 2009
6-12
-------
Technical Development Document for Proposed Effluent 6. Deicing Chemical Use and
Limitation Guidelines and Standards for the Airport Deicing Category Deicing Stormwater Characterization
6.3 References
USEPA. 1998. Method 1671: Volatile Organic Compounds Specific to the Pharmaceutical
Manufacturing Industry by Gas Chromatography/Flame lonization Detector. U.S.
Environmental Protection Agency/Office of Water. Washington, D.C.
http://www.epa.gov/waterscience/methods/method/files/1671.pdf
USEPA. 2003. Method 8015D: Nonhalogenated Organics Using Gas Chromatography/Flame
lonization Detector. (From "Test Methods for Evaluating Solid Waste, Physical/Chemical
Methods," publication no. SW-846.) U.S. Environmental Protection Agency/Office of Solid
Waste. Washington, D.C. http://www.epa.gov/epawaste/hazard/testmethods/pdfs/8015d_r4.pdf
USEPA. 2008a. Airport Questionnaire Database. U.S. Environmental Protection Agency/Office
of Water. Washington, D.C. DCN AD00927.
USEPA. 2008b. Airline Questionnaire Database. U.S. Environmental Protection Agency/Office
of Water. Washington, D.C. DCN AD00938.
USEPA. 2008c. Airport Deicing Loadings Database. U.S. Environmental Protection
Agency/Office of Water. Washington, D.C. DCN AD00857.
USEPA. 2009. Environmental Impact and Benefit Assessment for Proposed Effluent Limitation
Guidelines and Standards for the Airport Deicing Category. EPA 821-R-09-003. (July). DCN
AD01197.
USEPA. 2006a. Final Sampling Episode Report Detroit Metropolitan Wayne County
International Airport (DTW), Episode 6508. U.S. Environmental Protection Agency/Office of
Water. Washington, D.C. (July 12). DCN AD00620.
USEPA. 2006b. Final Sampling Episode Report Minneapolis/St. Paul International Airport
(MSP), Episode 6509. U.S. Environmental Protection Agency/Office of Water. Washington,
D.C. (July 12). DCN AD00622.
USEPA. 2006c. Final Sampling Episode Report Denver International Airport, Episode 6522.
U.S. Environmental Protection Agency/Office of Water. Washington, D.C. (August 2). DCN
AD00840.
USEPA. 2006d. Final Sampling Episode Report Pittsburgh International Airport, Episode 6528.
U.S. Environmental Protection Agency/Office of Water. Washington, D.C. (November 1). DCN
AD00841.
USEPA. 2006e. Final Sampling Episode Report Albany International Airport, Episode 6523.
U.S. Environmental Protection Agency/Office of Water. Washington, D.C. (July 19). DCN
AD00842.
USEPA. 2006f. Final Sampling Episode Report Greater Rockford Airport, Sampling Episode
6529 (April 20, 2006) and Sampling Episode 6530 (August 29, 2006, January 18, 2008). U.S.
Environmental Protection Agency/Office of Water. Washington, D.C. DCN AD00839.
July 2009 6-13
-------
Technical Development Document for Proposed Effluent 6. Deicing Chemical Use and
Limitation Guidelines and Standards for the Airport Deicing Category Deicing Stormwater Characterization
ERG. 2007a. Personal communication (teleconj between Mary Willett (ERG) and Bob Reynolds
(Pylam Dyes). Questions Regarding Pylam Dye Use inADF. (April 16). DCN AD00686.
ERG. 2007b. Personal communication (telecon) between Mary Willett (ERG) and Jeffery Carey
(Naveon). Aircraft Deicing Fluids. (April 16). DCN AD00688.
Corsi, SR et al. 2003. Nonylphenol Ethoxylates and Other Additives in Aircraft Deicers, Anti-
icers, and Waters Receiving Airport Runoff. (January 1). DCN AD00083.
July 2009 6-14
-------
Technical Development Document for Proposed Effluent 7. Pollutants of Concern
Limitation Guidelines and Standards for the Airport Deicing Category
7. POLLUTANTS OF CONCERN
EPA identified pollutants in stormwater associated with deicing activities that should
potentially be controlled. Pollutants of concern may include pollutants directly associated with
deicing chemicals, by-products from deicing activities (e.g. metals), and/or pollutant parameters
that are influenced by deicing chemicals (e.g., BOD and COD).
EPA reviewed its deicing stormwater sampling data as well as the information available
through NPDES permits to identify conventional, nonconventional, and priority pollutants
present in airport deicing stormwater and evaluate whether they are causing environmental
impacts from their discharge. This section presents the results of EPA's evaluation and identifies
potential pollutants of concern and those proposed for regulation.
7.1 Identification of Airport Deicing/Anti-icing Stormwater Pollutants
Airport deicing stormwater is generated when airfield chemicals and aircraft deicing/anti-
icing fluids (ADFs) mix with stormwater (either directly or as a result of snowmelt). Because
deicing stormwater is weather-dependent, it is highly variable in nature and pollutant
concentrations may vary greatly. In addition, other airport-related activities, including aircraft
fueling and maintenance activities, may contribute pollutants to stormwater that is also
contaminated with deicing chemicals. Because of the inherent difficulties in characterizing
deicing stormwater, EPA evaluated pollutants detected in deicing stormwater, pollutants present
in source water, and pollutants that are present in ADFs prior to use to determine the pollutants
likely to be present in deicing stormwater.
EPA considered multiple sources of information to identify potential pollutants of
concern in deicing stormwater including the following:
• EPA sampling data from the Preliminary Data Summary;
• NPDES permits for airports to determine pollutants that are currently monitored
or limited at airports;
• EPA sampling data collected during the 2004/2005 and 2005/2006 deicing
seasons to identify pollutants present in untreated deicing stormwater;
• EPA sampling data collected during the 2004/2005 and 2005/2006 deicing
seasons to determine pollutants present in source water; and
• EPA sampling data collected during the 2004/2005 deicing season, current
research, and expert sources to determine ADF constituents.
EPA also looked at toxic weighting factors to assess the comparative toxicity of different
pollutants. Each of these data sources is discussed below.
Airport Deicing Operations PDS
For the PDS, EPA sampled:
• Type I ADFs;
• Lagoon stormwater from Albany International Airport;
• Untreated deicing stormwater from Kansas City International Airport;
• Untreated deicing stormwater from Bradley International Airport;
July 2009 7-1
-------
Technical Development Document for Proposed Effluent 7. Pollutants of Concern
Limitation Guidelines and Standards for the Airport Deicing Category
• Untreated deicing stormwater from Greater Rockford Airport; and
• Stormwater outfalls from Bradley International Airport.
The pollutants detected in one or more of these samples were summarized in Table 8-4 of the
Preliminary Data Summary: Airport Deicing Operations report (USEPA, 2000) and are
presented in Table 7-1 of this report.
NPDES Permits
EPA reviewed NPDES individual and general stormwater permits for airports that are
estimated to have the most deicing operations in the United States. The permit review is
summarized in the Airport Deicing Operations NPDES Permit Review Summary memorandum
(ERG, 2007a). Table 7-1 lists pollutants that have monitoring and limit requirements in current
airport NPDES permits.
Pollutants Present in Untreated Deicing Stormwater
Under the current rulemaking effort, EPA collected samples at the following six airports
during the 2004/2005 and 2005/2006 deicing seasons:
• Detroit Metropolitan Wayne County (DTW) airport;
• Minneapolis-St. Paul International (MSP) airport;
• Albany International (ALB) airport;
• Greater Rockford (RFD) airport;
• Pittsburgh International (PIT) airport; and
• Denver International (DEN) airport.
Table 7-1 lists pollutants detected in untreated deicing stormwater from these locations.
Pollutants Present in Source Water
Table 7-1 also lists pollutants detected in source water samples at the airports EPA
sampled at during the 2004/2005 and 2005/2006 deicing seasons.
Deicing/Anti-Icing Fluid Constituents
EPA does not have sufficient information on all of the constituents of airfield chemicals
and ADFs to fully characterize them for Table 7-1. However, EPA's airport questionnaire does
identify which chemicals and brand-name products are commonly used, which can help to define
the pollutants expected to be in deicing stormwater.
July 2009 7-2
-------
Technical Development Document for Proposed Effluent
Limitation Guidelines and Standards for the Airport Deicing Category
7. Pollutants of Concern
Table 7-1. Pollutants Under Consideration as Potential Pollutants of Concern
Analyte
Pollutants
Identified in
the PDS
Sampling
Pollutants
Monitored in
NPDES
Permits
Pollutants Identified
in Raw ADF in
Research or 2004-
2006 EPA Sampling
Pollutants Identified
in Untreated
Stormwater in 2004-
2006 EPA Sampling
Pollutants Identified
in Source Water in
2004-2006 EPA
Sampling
Toxic Weighting
Factor
Classical*
Alkalinity
Ammonia As Nitrogen (NH3-N)
5 -Day Biochemical Oxygen
Demand (BOD5)
Chemical Oxygen Demand
(COD)
Chloride
Dissolved Oxygen
Hardness
Oil & Grease
Silica-Gel Treated
Hexane Extractable Material
(SGT-HEM)
Hexane Extractable Material
(HEM)
Nitrate/Nitrite (NO3-N + NO2-N)
Sulfate
Total Dissolved Solids (TDS)
Total Kjeldahl Nitrogen (TKN)
Total Organic Carbon (TOC)
Total Orthophosphate
Total Phosphorus
Total Petroleum Hydrocarbons
(TPH)
Total Recoverable Phenolics
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
0.0011
0.000024
0.000006
1TWF assumed to be equal to 2.8 for any of the alkylphenol ethoxylates.
Note: Octylphenol and nonylphenol should have a higher toxicity than the alkylphenol ethoxylates.
July 2009
-------
Technical Development Document for Proposed Effluent
Limitation Guidelines and Standards for the Airport Deicing Category
7. Pollutants of Concern
Table 7-1 (Continued)
Analyte
Total Suspended Solids (TSS)
Pollutants
Identified in
the PDS
Sampling
Pollutants
Monitored in
NPDES
Permits
X
Pollutants Identified
in Raw ADF in
Research or 2004-
2006 EPA Sampling
Pollutants Identified
in Untreated
Stormwater in 2004-
2006 EPA Sampling
X
Pollutants Identified
in Source Water in
2004-2006 EPA
Sampling
Toxic Weighting
Factor
Metals
Aluminum
Antimony
Arsenic
Barium
Boron
Cadmium
Calcium
Chromium
Copper
Iron
Lead
Magnesium
Manganese
Mercury
Molybdenum
Potassium
Selenium
Silver
Sodium
Thallium
Tin
Titanium
Vanadium
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
0.065
0.012
4.04
0.0020
0.18
23.12
0.000028
0.076
0.63
0.0056
2.24
0.00087
0.070
117.12
0.20
1.12
0.000005
1.03
0.30
0.029
0.035
1TWF assumed to be equal to 2.8 for any of the alkylphenol ethoxylates.
Note: Octylphenol and nonylphenol should have a higher toxicity than the alkylphenol ethoxylates.
July 2009
-------
Technical Development Document for Proposed Effluent
Limitation Guidelines and Standards for the Airport Deicing Category
7. Pollutants of Concern
Table 7-1 (Continued)
Analyte
Zinc
Pollutants
Identified in
the PDS
Sampling
X
Pollutants
Monitored in
NPDES
Permits
X
Pollutants Identified
in Raw ADF in
Research or 2004-
2006 EPA Sampling
X
Pollutants Identified
in Untreated
Stormwater in 2004-
2006 EPA Sampling
X
Pollutants Identified
in Source Water in
2004-2006 EPA
Sampling
X
Toxic Weighting
Factor
0.047
Organics
Acetone
Benzene, toluene, ethylbenzene,
xylene (BTEX)
Benzole Acid
Bis(2-Ethylhexyl) Phthalate
Di-n-butyl Phthalate
Diethylene Glycol
N-Dodecane
Ethylene Glycol
N-Hexadecane
Methyl Ethyl Ketone
Naphthalene
Phenol
Propylene Glycol
N-Tetradecane
1,2,4- Trimethylbenzene
Trichloroethene
Tolyltriazole
Benzotriazole
5-Methyl-lH-benzotriazole
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
0.000008
0.032 benzene
0.0056 toluene
0.0014ethylene
benzene
0.0043 xylene
0.00033
0.25
0.012
0.00000074
0.0043
0.0013
0.0043
0.000026
0.016
0.028
0.000057
0.0043
0.019
0.0018
1TWF assumed to be equal to 2.8 for any of the alkylphenol ethoxylates.
Note: Octylphenol and nonylphenol should have a higher toxicity than the alkylphenol ethoxylates.
July 2009
-------
Technical Development Document for Proposed Effluent
Limitation Guidelines and Standards for the Airport Deicing Category
7. Pollutants of Concern
Table 7-1 (Continued)
Analyte
Pollutants
Identified in
the PDS
Sampling
Pollutants
Monitored in
NPDES
Permits
Pollutants Identified
in Raw ADF in
Research or 2004-
2006 EPA Sampling
Pollutants Identified
in Untreated
Stormwater in 2004-
2006 EPA Sampling
Pollutants Identified
in Source Water in
2004-2006 EPA
Sampling
Toxic Weighting
Factor
Alkylphenols
Nonylphenol, total
Nony Iphenol- 1 -Ethoxy late
Nonylphenol-2-Ethoxylate
Nony lphenol-3 -Ethoxy late
Nony lphenol-4-Ethoxy late
Nony lphenol-5 -Ethoxy late
Nony lphenol-6-Ethoxy late
Nony lphenol-7 -Ethoxy late
Nony lphenol-8-Ethoxy late
Nony lphenol-9-Ethoxy late
Nonylphenol- 10-Ethoxy late
Nonylphenol- 1 1 -Ethoxy late
Nonylphenol- 12-Ethoxy late
Nonylphenol- 1 3 -Ethoxy late
Nonylphenol- 14-Ethoxy late
Nonylphenol- 1 5 -Ethoxy late
Nonylphenol- 16-Ethoxy late
Nonylphenol- 17-Ethoxy late
Nony lphenol-18-Ethoxy late
Octylphenol
Octylphenol-2 -Ethoxy late
Octy lphenol-3 -Ethoxy late
Octylphenol-4-Ethoxylate
Octy lphenol-5 -Ethoxy late
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
0.85
i
i
i
i
i
i
i
i
i
i
i
i
i
i
i
i
i
i
0.30
i
i
i
i
1TWF assumed to be equal to 2.8 for any of the alkylphenol ethoxylates.
Note: Octylphenol and nonylphenol should have a higher toxicity than the alky Iphenol ethoxylates.
July 2009
-------
Technical Development Document for Proposed Effluent
Limitation Guidelines and Standards for the Airport Deicing Category
7. Pollutants of Concern
Table 7-1 (Continued)
Analyte
Octylphenol-6-Ethoxylate
Octylphenol-7-Ethoxylate
Octylphenol-8-Ethoxylate
Octylphenol-9-Ethoxylate
Octylphenol-10-Ethoxylate
Octylphenol-1 1-Ethoxylate
Octylphenol-12-Ethoxylate
Total Nonyrphenol-3-Ethoxylate-
Nonlyphenol-18-Ethoxylate
Total Octyrphenol-2-Ethoxylate-
Octylphenol-12-Ethoxylate
Pollutants
Identified in
the PDS
Sampling
Pollutants
Monitored in
NPDES
Permits
Pollutants Identified
in Raw ADF in
Research or 2004-
2006 EPA Sampling
X
X
X
X
X
X
X
X
X
Pollutants Identified
in Untreated
Stormwater in 2004-
2006 EPA Sampling
X
X
X
X
X
X
X
X
X
Pollutants Identified
in Source Water in
2004-2006 EPA
Sampling
X
Toxic Weighting
Factor
i
i
i
i
i
i
i
2.80
2.80
1TWF assumed to be equal to 2.8 for any of the alkylphenol ethoxylates.
Note: Octylphenol and nonylphenol should have a higher toxicity than the alkylphenol ethoxylates.
July 2009
-------
Technical Development Document for Proposed Effluent 7. Pollutants of Concern
Limitation Guidelines and Standards for the Airport Deicing Category
The commonly used airfield deicing and anti-icing chemicals are listed below, along with
the approximate percentage of total pavement chemical usage they comprise:
• Potassium acetate (64 percent);
• Propylene glycol-based fluids (12 percent);
• Urea (11 percent);
• Sodium acetate (9 percent);
• Sodium formate (3 percent); and
• Ethylene glycol-based fluids (2 percent).
EPA collected and analyzed samples of unused, or "raw," ADF during the sampling
episodes at DTW and MSP. EPA did not analyze samples for alkylphenols during these sampling
episodes; therefore, no alkylphenol data are available for the raw ADF.
Finally, EPA reviewed research conducted by USGS to determine potential ADF
constituents. Steven Corsi of USGS conducts research and sampling on ADF and deicing
stormwater at General Mitchell International Airport in Milwaukee, WI. He has published
several papers presenting his research and sampling results and has identified the following
pollutants in ADF stormwater and snowmelt:
BODS;
COD;
• Propylene and ethylene glycol;
• Alkylphenols (AP), and alkylphenol ethoxylates (APEO); and
• Benzotriazole (BT) and its methylated derivatives (MeBT).
Toxic Weighting Factors
To assess relative toxicity of the pollutants present in deicing stormwater, EPA
considered the toxic weighting factors (TWFs) of the pollutants. EPA develops TWFs based on
toxicity data contained in the EPA Toxics Database.(U.S. EPA, 2007) This database includes
data from over 100 references and contains aquatic life and human health toxicity data, as well as
physical/chemical property data, for more than 1,900 pollutants. EPA developed a TWF for
alkylphenols based on available toxicity information. (ERG, 2007b) Pollutant TWFs, where
available, are included in Table 7-1.
7.2 Pollutants of Concern Selection Criteria
Having identified pollutants that are likely to be present in airport deicing stormwater,
EPA then considered which pollutants should potentially be controlled. EPA considered the
following criteria in assessing potential pollutants of concern:
• Whether the pollutant is present in deicing stormwater from a source other than
deicing/anti-icing chemical use;
• Whether the pollutant is discharged in relatively small amounts and/or is likely to
cause toxic effects;
• Whether the pollutant is detected in the effluent from a small number of airports
and is uniquely related to those facilities; or
July 2009 7-8
-------
Technical Development Document for Proposed Effluent 7. Pollutants of Concern
Limitation Guidelines and Standards for the Airport Deicing Category
• Whether the pollutant can be analyzed by using EPA-approved or other
established methods.
After consideration of the criteria listed above, EPA developed a list of those pollutants that are
considered potential pollutants of concern.
7.3 Identification of Potential Pollutants of Concern
EPA compared the pollutants detected in deicing stormwater to ADF constituents and
determined that many pollutants present in the stormwater are not ADF constituents. Stormwater
contains pollutants from sources other than ADF; these sources may include, but are not limited
to, the following:
• Source water pollutants (present in the water used at the airport facility);
• Pollutants from aircraft and vehicle fueling;
• Pollutants from maintenance-related operations; or
• Pollutants from roof runoff.
EPA also considered the other criteria listed in Section 7.2 above to assess potential pollutants of
concern. Below is a summary of EPA's evaluation of potential pollutants of concern by the
following analytical categories: classical parameters, metals, and organic pollutants.
Classical Parameters
The major components of both airfield deicing chemicals and aircraft deicing fluids are
organic and degrade in the environment after their release. Because of this, COD and BOD5
concentrations are generally high in deicing stormwater. Both of these pollutant parameters are
also good indicators of the amount of acetates, urea, glycols, and formates in deicing stormwater
and the environmental impacts of the deicing stormwater on oxygen demand in receiving waters.
EPA believes that those airports requiring monitoring and control of ammonia, TKN, and
nitrate/nitrite are likely doing so to monitor discharges of urea. Information collected during
EPA's airport site visits seemed to indicate that airports have been phasing out the use of urea for
airfield deicing. However, EPA's analysis of urea use from the airport questionnaire has actually
shown an increase in the amount of urea use during the 2002/2003 through 2004/2005 deicing
seasons, with approximately the same number of airports using urea in each of those seasons.
Several of the classical parameters detected in deicing stormwater are from stormwater,
dilution water, or other airport operations; these pollutants are not present in ADF but are present
in deicing stormwater. Pollutants from other airport sources aside from ADF include alkalinity,
hardness, oil and grease, TPH, and TSS.
Metals
Multiple metals have been detected in samples of airport deicing stormwater. Some of
these metals were also detected in the ADF samples collected by EPA. Many of these metals are
not original components of deicing products (e.g., aluminum and chromium); they are present as
background concentrations from the stormwater or source water used for ADF dilution or they
July 2009 7-9
-------
Technical Development Document for Proposed Effluent 7. Pollutants of Concern
Limitation Guidelines and Standards for the Airport Deicing Category
are metals picked up by stormwater runoff from aircraft maintenance/operation areas or building
roofs.
Organic Pollutants
Organic pollutants present in deicing stormwater include propylene glycol, ethylene
glycol, triazole compounds, alkylphenols, and alkylphenol ethoxylates. Other organics may also
be present from the breakdown of glycols, urea, acetates, and formates.
Toxicity studies conducted by Corsi and others (Corsi et al., 2006) indicate that deicing
stormwater may exhibit toxicity from the additive compounds included in the aircraft (and in
some cases) airfield deicing products.
Based on these findings, EPA identified the following as potential pollutants of concern
for the Airport Deicing Category:
COD;
BODS;
• Ethylene glycol;
• Propylene glycol;
• Benzotriazole;
• 5-Methyl-lH-benzotriazole;
• Nonylphenol, total;
• Octylphenol, total;
• Total nonylphenol-3-ethoxylate-nonlyphenol-18-ethoxylate;
• Total octylphenol-2-ethoxylate-octylphenol-12-ethoxylate;
• Ammonia as nitrogen;
• Nitrate/Nitrite;
TKN;
• Antimony;
• Copper;
• Iron;
• Mercury;
• Molybdenum;
• Tin;
• Zinc; and
• Acetone.
7.4 Selection of Regulated Pollutants for Proposal
Table 7-2 lists the potential pollutants of concern identified in Section 7.3, along with an
explanation of whether EPA selected the pollutant for regulation. Based on the documented
environmental impacts from airport deicing runoff, EPA focused on regulating those pollutants
exerting oxygen demand and contributing toxicity to receiving water bodies. EPA found that the
impacts of slug loads of ADF stormwater on the dissolved oxygen of receiving streams, as well
as color and odor issues associated with high ADF concentrations in stormwater discharge, are
well documented. The main component of ADF is glycol, which exhibits significant oxygen
demand. Research by Corsi (Corsi et al., 2006) also identified potential toxicity concerns that
July 2009 7-10
-------
Technical Development Document for Proposed Effluent 7. Pollutants of Concern
Limitation Guidelines and Standards for the Airport Deicing Category
may be linked to ADF additives, specifically triazoles and alkylphenols. In conversations with
ADF manufacturers, EPA has been told that the use of triazole compounds in ADF is being
discontinued and that triazole use in European ADFs has been phased out. Alkylphenols and
their ethoxylates have also been identified as potential toxic components of ADF. EPA's
sampling data have confirmed the presence of these compounds; however, EPA believes that
insufficient information is currently available to fully characterize the extent to which these
compounds are present in deicing stormwater and their impact.
For stormwater discharges from airfield deicing operations, EPA focused on the
continued use of urea. Urea breaks down into ammonia, and the resulting ammonia toxicity in
receiving streams has helped to discourage urea use as an airfield deicing chemical in the past.
Alternative airfield deicing chemicals that are less toxic than ammonia are available and are
predominantly composed of a salt ion (potassium or sodium) and either acetate or formate. When
inadequately treated, urea-contaminated wastewater also may contribute to nitrogen enrichment
and eutrophication of receiving waters. EPA evaluated whether regulation of airfield deicing
stormwater was practical or cost-effective. Since airfield deicing stormwater losses tend to occur
over large areas and the volumes of dilute stormwater may be very high, at this time, EPA could
not identify an "economically achievable" means to regulate airfield deicing stormwater other
then to encourage a complete transition away from urea use.
Based on the known environmental impacts from deicing stormwater discharges, EPA
has selected COD and ammonia as N for proposed regulation. COD is a good indicator
parameter to monitor the overall oxygen demand resulting from the discharge of gly col-based
ADFs and any other organic constituents present in the stormwater. Ammonia as N is proposed
for regulation to ensure that airports cease using urea as an airfield deicer, since other less toxic
products are available.
EPA evaluated the impacts of the airport deicing collection and treatment scenarios on
both BOD5 and COD discharges. EPA selected COD for regulation and not BOD5 for the
following reasons:
• While both of these parameters are good indicators of the gly col-based oxygen
demand component of deicing stormwater, COD will also capture the oxygen
demand from nitrogen and other organic components of the stormwater that may
not be represented in a BODs result.
• COD analyses are simple to conduct and can be measured in real time compared
to a 5-day test for BOD.
• COD eliminates the need to consider receiving water temperature when
evaluating water quality concerns.
• Toxic ADF additive compounds in deicing stormwater may have a negative and
variable affect on the acclimation of the active cultures used in BOD analysis,
making the method less robust than COD analysis for these wastewaters.
July 2009 7-11
-------
Technical Development Document for Proposed Effluent
Limitation Guidelines and Standards for the Airport Deicing Category
7. Pollutants of Concern
Table 7-2. Potential Pollutants of Concern Selected for Proposed Regulation
Potential Pollutant of Concern
BODS
COD
Ethylene glycol
Propylene glycol
Benzotriazole
5-Methyl-lH-benzotriazole
Nonylphenol, Total
Octylphenol, Total
Total nonylphenol-3-ethoxylate-
nonlyphenol-18-ethoxylate
Total octyrphenol-2-ethoxylate-
octylphenol- 12-ethoxylate
Ammonia as nitrogen
Nitrate/Nitrite
TKN
Antimony
Copper
Iron
Mercury
Molybdenum
Tin
Zinc
Acetone
Selected for
Regulation
X
X
Explanation of Selection or Non-selection for Proposal
COD as surrogate
Proposed for regulation
COD as surrogate
COD as surrogate
Limited data available to support selection and potential
discontinued use
Limited data available to support selection and potential
discontinued use
Limited data available to support selection
Limited data available to support selection
Limited data available to support selection
Limited data available to support selection
Proposed for regulation to monitor urea use
Ammonia as nitrogen as surrogate for urea use
Ammonia as nitrogen as surrogate for urea use
Limited impact data for metals and metal treatment not
included in BAT
Limited impact data for metals and metal treatment not
included in BAT
Limited impact data for metals and metal treatment not
included in BAT
Limited impact data for metals and metal treatment not
included in BAT
Limited impact data for metals and metal treatment not
included in BAT
Limited impact data for metals and metal treatment not
included in BAT
Limited impact data for metals and metal treatment not
included in BAT
COD as surrogate
July 2009
7-12
-------
Technical Development Document for Proposed Effluent 7. Pollutants of Concern
Limitation Guidelines and Standards for the Airport Deicing Category
EPA is not proposing to regulate metals, even though in some cases they may be present
from deicing operations and not just maintenance or fueling operations. EPA's sampled
concentration levels for those metals identified as potential pollutants of concern are, in general,
low (usually below EPA water quality standard levels). As such, EPA does not have sufficient
data to support proposing a regulation at the national level. EPA believes that any site-specific
concerns with detected metals can be addressed through facility-specific effluent limitations in
NPDES permits.
7.5 References
USEPA. 2000. Preliminary Data Summary: Airport Deicing Operations. U.S. Environmental
Protection Agency/Office of Water. Washington, D.C. EPA-821-R-00-016. (August).
http://www.epa.gov/guide/airport
ERG. 2006. Memorandum to Brian D'Amico and Eric Strassler (U.S. EPA) from Steve
Strackbein and Cortney Itle (ERG): Airport Deicing Operations SWPPP and Permit Review.
(November 8). DCN AD00859.
ERG. 2007a. Memorandum to Brian D'Amico and Eric Strassler (U.S. EPA) from Jason
Huckaby (ERG): Airport Deicing Operations NPDES Permit Review Summary. (April 16) DCN
AD00611.
Corsi, SR, et al. 2003. "Nonylphenol Ethoxylates and Other additives in Aircraft Deicers, Anti-
icers, and Waters Receiving Airport Runoff." Environmental Science & Technology, 37:4031-
4037. DCN AD00083.
USGS. 2004. General Mitchell International Airport and U.S. Geological Survey. Data summary
for monitoring and assessment of changes in water quality due to deicer management Water
Years 1997-2004. (September). DCN AD00085.
Corsi, SR, et al. 2000. "Aircraft and Runway Deicers at General Mitchell International Airport,
Milwaukee, Wisconsin, USA. Biochemical Oxygen Demand and Dissolved Oxygen in
Receiving Streams." (November 22). DCN AD00081.
Corsi, SR, et al. 2006. "Characterization of Aircraft Deicer and Anti-icer Components and
Toxicity in Airport Snowbanks and Snowmelt Runoff." Environmental Science & Technology.
DCN AD00326.
US EPA. 2006. Toxic Weighting Factor Development in Support of CWA 304(m) Planning
Process. U.S. Environmental Protection Agency/Office of Water. Washington, D.C. (June). DCN
AD00861.
ERG. 2007b. Memorandum from Maureen Kaplan (ERG) to Mary Willett (ERG): Toxic
Weighting Factors (TWFs) for Nonylphenol, Octylphenol, and Alkyl Phenol Ethoxylate. (March
30). DCN AD00862.
USEPA. 2007. Ecotoxicology ("ECOTOX") Database. U.S. Environmental Protection Agency.
Washington, D.C. www.epa.gov/ecotox/
July 2009 7-13
-------
Technical Development Document for Proposed Effluent 8. Pollution Prevention and Treatment Technologies
Limitation Guidelines and Standards for the Airport Deicing Category Applicable to Airport Deicing Operations
8. POLLUTION PREVENTION AND TREATMENT TECHNOLOGIES APPLICABLE TO
AIRPORT DEICING OPERATIONS
The NPDES permit program, along with the emergence of problems such as fish kills and
odor reports, have prompted airports and airlines to investigate a wide range of pollution
prevention and treatment practices. These practices are designed to eliminate or minimize the
environmental impact of ADF and airfield pavement chemicals without compromising safety.
This section summarizes the common techniques used to collect deicing stormwater and the
treatment or recycling steps implemented prior to discharge. This section also discusses pollution
prevention practices used by U.S. airports and airlines.
Each method of collection, treatment or recycling, or pollution prevention selected by an
airport or airline often depends on a variety of airport-specific or airline-specific factors,
including climate, amount of chemical deicing and anti-icing agents applied, number of airlines
operating at a particular airport, aircraft fleet mix, number of aircraft operations, costs, presence
of existing infrastructure, availability of land, and affect on aircraft departures. EPA recognizes
that some of the practices discussed in this section may not be practical or economically feasible
for all U.S. airports.
Section 8.1 discusses deicing stormwater collection, Section 8.2 describes deicing
stormwater treatment and recycling, and Section 8.3 presents pollution prevention (e.g. product
substitution) practices.
8.1 Deicing Stormwater Collection
The collection of deicing stormwater from aircraft deicing/anti-icing operations helps
prevent or minimize discharges at stormwater outfalls. Airports currently use a variety of
collection and conveyance methods, including designated aircraft deicing pads, gate and ramp
area drainage collection systems, grassed swales, storm sewer plugs, and specially designed
glycol recovery vehicles (GRVs), each of which is discussed in Section 8.1.1. Individual airports
often rely on a combination of these collection strategies to effectively collect ADF-
contaminated stormwater.
Section 8.1.2 presents common methods for storing and discharging deicing stormwater,
including detention ponds or constructed wetlands, retention ponds, permanent storage tanks or
frac tanks, discharge to a publicly owned treatment works (POTW), trucking waste off site, or
any combination of these methods. The following subsections describe in detail the various
wastewater collection methods used by the industry.
8.1.1 Deicing Stormwater Collection and Conveyance
This subsection describes the various wastewater collection and conveyance methods
commonly used by airports. Airport stormwater collection systems are designed to collect
deicing stormwater from several different locations at which deicing operations are performed,
including aircraft deicing at centralized deicing pads, at the gates, or at parking or cargo aprons.
Collection systems may collect stormwater from airfield deicing locations including; ramps,
taxiways, and runways. Common methods of collecting and conveying deicing stormwater
include deicing pad collection systems, gate and ramp area drainage collection systems, grassed
swales, storm sewer plugs, and GRVs.
July 2009 8-1
-------
Technical Development Document for Proposed Effluent 8. Pollution Prevention and Treatment Technologies
Limitation Guidelines and Standards for the Airport Deicing Category Applicable to Airport Deicing Operations
Deicing Pads
A centralized deicing pad is a facility on an airfield built specifically for aircraft deicing
operations. It is typically a paved area adjacent to a gate area, taxiway, or runway, and
constructed with a drainage system separate from the airport's main storm drain system. It is
usually constructed of concrete with sealed joints to prevent the loss of sprayed ADF through the
joints. The pad's collection system is typically connected to a stormwater storage facility, which
then may send the stormwater to an on-site or off-site treatment facility.
Deicing pads restrict aircraft deicing to a confined area, and allow for the capture of
deicing stormwater at the point of generation, thereby minimizing the volume of spent deicing
fluid discharged in an uncontrolled manner. Aircraft deicing pads also centralize deicing
activities, which allow airports to more easily collect high concentration ADF-contaminated
stormwater, creating a better opportunity for recycle and recovery of glycol. Transporting spent
ADF off site to wastewater treatment plants, POTWs, or recycling facilities is also more
economical when the amount of deicing stormwater is minimized.
One benefit of deicing aircraft on deicing pads instead of at the gates is that it moves
aircraft from the gate areas, thereby opening gates for arriving aircraft. Another benefit is that
pads are commonly located near the heads of runways, where planes can be deiced just prior to
takeoff; as a result, less Type IV anti-icing fluid may be necessary due to shorter holdover times,
reducing the amount of glycols transferred off of the deicing pad or released into the air.
Figure 8-1 shows an example of a centralized deicing pad with fixed boom sprayers.
Figure 8-1. Deicing Pad Equipped with Fixed Deicing Booms at Pittsburgh Airport
July 2009
8-2
-------
Technical Development Document for Proposed Effluent 8. Pollution Prevention and Treatment Technologies
Limitation Guidelines and Standards for the Airport Deicing Category Applicable to Airport Deicing Operations
Gate and Ramp Area Drainage Collection Systems
Other than deicing pads, the most common areas for deicing operations are passenger
terminal gates and aircraft parking or cargo ramps/aprons. To collect wastewater generated at
these locations, some airports have installed collection systems or modified existing stormwater
drainage systems. The typical collection system consists of graded concrete pavement with
trench or square drains that convey wastewater to a storage facility or discharge point through a
diversion box. Figure 8-2 shows an example of a gate deicing area. Gate and ramp collection
systems generate low concentration, ADF-contaminated stormwater because more stormwater
gets mixed in with the spent ADF and because there are increased fugitive losses due to vehicle
traffic around the planes. For some stormwater drainage systems, a diversion box allows
uncontaminated stormwater to be diverted to stormwater outfalls.
Figure 8-2. Mobile Deicer Truck Performing Gate Deicing at Chicago O'Hare Airport
Grassed Swales
Grassed swales are shallow, open-channel engineered landscape features that are covered
with grass and other erosion-resistant vegetation. Grass swales slow and control stormwater flow
rates and act as a filter to lower pollution loads through the removal of solids. Swales are
designed to treat runoff by filtering the stormwater through vegetation and subsoil, and allow
stormwater to infiltrate underlying soils. To work effectively, stormwater discharged into a swale
should occur in a thin, sheet flow pattern to maximize infiltration. Grassed swales are popular
because they are low cost compared to other control measures (CASQA, 2003)
Plug and Pump Systems
Some airports use storm drain inserts or plugs to close the drains and allow the collection
of ADF-contaminated stormwater within the existing airport stormwater drainage system. When
July 2009
8-3
-------
Technical Development Document for Proposed Effluent 8. Pollution Prevention and Treatment Technologies
Limitation Guidelines and Standards for the Airport Deicing Category Applicable to Airport Deicing Operations
aircraft are undergoing deicing/anti-icing, the inserts are installed to force contaminated
stormwater to pool in drainage piping until the stormwater can be vacuumed or pumped out. This
practice prevents manholes and stormwater piping from overflowing. Once deicing/anti-icing
activity ends and the contaminated stormwater is removed, the storm drain inserts can be
removed, or deactivated allowing uncontaminated stormwater to pass through the drain. Plug-
and-pump collection systems are applicable to airports that deice at the gate. One benefit of
deicing at the gate is that it allows the components of the existing collection system
infrastructure (i.e. existing storm sewers) to be incorporated into the plug-and-pump collection
system and can reduce the costs associated with deicing control.
Minneapolis-St. Paul International (MSP) airport is one example of an airport using a
plug and pump collection system. At MSP airport, 30 percent of deicing operations take place in
the airport's sixteen plug and pump containment areas. During the deicing season, the plug and
pump areas are fitted with compression plugs in the storm sewers to prevent contaminated or
potentially contaminated stormwater from being discharged. The stormwater plugs convert the
stormwater sewer pipes and manholes into individual stormwater retention systems that can each
retain between 5,000 to 42,000 gallons of stormwater. These individual stormwater retention
systems are monitored during the day to determine how full they are (to prevent overflow) and to
determine how to manage the ADF-contaminated stormwater based on its composition and
strength. Contaminated stormwater is pumped or vacuumed from the sewer pipes and tested to
determine the glycol concentration. Based on the glycol concentration of the wastewater, it is
stored in either a low concentration storage tank or a high concentration storage tank prior to
being shipped offsite for recycling (USEPA, 2008a).
Glycol Recovery Vehicles (GRVs)
GRVs are specially-designed vehicles that remove glycol and other deicing fluid runoff
from airport deicing pads and gate locations by vacuuming liquid from pavement surfaces.
Figure 8-3 shows an example of a GRV. GRVs help prevent stormwater from infiltrating and
contaminating surrounding waterways. Once ADF-contaminated wastewater is vacuumed from
airport surfaces, it is typically transported to an on-site storage facility where the airport can treat
and discharge, recycle or ship the waste offsite for treatment or recycling.
July 2009 8-4
-------
Technical Development Document for Proposed Effluent 8. Pollution Prevention and Treatment Technologies
Limitation Guidelines and Standards for the Airport Deicing Category Applicable to Airport Deicing Operations
8.1.2
Figure 8-3. Glycol Recovery Vehicle
Deicing Stormwater Storage
This subsection describes the various Stormwater storage methods commonly used by
airports. Airport Stormwater storage systems are designed to collect deicing Stormwater from
several different locations around an airport and to accommodate highly variable flows and
volumes. Common methods of Stormwater storage at airports include; detention ponds,
equalization ponds, retentions ponds, and storage or frac tanks.
Detention Ponds and Equalization Ponds
Detention ponds are open-water ponds that collect deicing Stormwater runoff from
runways and other airport property. Detention ponds are designed to temporarily hold deicing
Stormwater anywhere from one day to two months and allow solids to settle while reducing
oxygen demand through surface oxygenation and volatilization prior to discharge to receiving
waters. Detention ponds can be lined or gravel-filled and may contain microscopic bacteria that
biodegrade deicing and anti-icing materials. Pump stations are commonly implemented to pump
metered runoff to discharge or further treatment. Detention basins often use aeration to increase
dissolved oxygen levels.
Equalization ponds are detention ponds designed to thoroughly mix ADF-contaminated
Stormwater so that consistent concentrations of pollutants can be pumped from the pond to
treatment/recycling operations or to other disposal. Equalization ponds may contain moving
parts, such as mixers, to ensure the liquid in the pond is completely mixed.
Retention Ponds
Retention ponds are designed to hold collected deicing Stormwater indefinitely. Usually
the pond is designed to allow overflow to drain to another location (e.g., a second retention pond
July 2009 8-5
-------
Technical Development Document for Proposed Effluent 8. Pollution Prevention and Treatment Technologies
Limitation Guidelines and Standards for the Airport Deicing Category Applicable to Airport Deicing Operations
or other overflow structure) when the water level gets above the pond capacity. Retention ponds
can also be used to treat deicing stormwater as part of a batch process (using chemical addition
and/or aeration) prior to discharge. With retention ponds, airports can collect ADF-contaminated
runoff throughout a deicing season and have the option of trucking it off site for treatment or
treating it on site in the retention pond.
Advantages and Disadvantages of Ponds in Treating ADF-Contaminated
Stormwater
Ponds require large areas for installation, and the normal operations of these systems
require treatment for many months after the end of the annual deicing season, before the
wastewater can be discharged. FAA discourages the installation of new stormwater retention
ponds at airports, as they can be a lure for migratory birds, which are a safety hazard for aircraft.
(FAA, 2007). For airports with existing retention ponds, however, where adequate storage
capacity is available, aerated pond systems may be able to provide efficient treatment. See
section 8.2.2 below for further discussion of aerated pond systems. Figure 8-4 shows an example
of an airport pond used for deicing stormwater storage.
Figure 8-4. Pond for Deicing Stormwater Storage at Denver International Airport
Storage Tanks and Frac Tanks
Airports that treat ADF-contaminated stormwater often use storage tanks to store the
stormwater prior to on-site treatment or transfer off site. These types of tanks can be constructed
as aboveground tanks or underground tanks. Collecting and storing deicing stormwater in storage
tanks allows an airport to equalize pollutant concentrations and can allow for a consistent flow
rate into an on-site treatment system, which is important to ensuring consistent treatment results.
July 2009
8-6
-------
Technical Development Document for Proposed Effluent 8. Pollution Prevention and Treatment Technologies
Limitation Guidelines and Standards for the Airport Deicing Category Applicable to Airport Deicing Operations
Frac tanks are mobile storage tanks that can provide temporary storage of collected
deicing stormwater or can be connected by a hose or pipeline to an alternative area. Frac tanks
can be easily removed from airport grounds and replaced with empty tanks when existing tanks
fill up. Figure 8-5 shows an example of two firac tanks.
Figure 8-5. Frac Tanks
8.2
Treatment and Recycling
Because of the high oxygen demand of ADF-contaminated stormwater, some airports
must treat their deicing stormwater prior to discharge. This section describes the use of oil/water
separators and dissolved air flotation (DAF) units that are used predominantly to address
stormwater contaminants that are not specifically from deicing operations, e.g., oil and grease
and solids(discussed in Section 8.2.1) and biological treatment through either on-site or off-site
treatment systems (discussed in Section 8.2.2).
Recycling of glycol from spent ADF decreases the amount ADF-contaminated
stormwater that reaches and potentially impairs surface and ground waters. The process to
recover glycol from spent ADF may take several steps. The recycle and recovery technologies
currently in use by U.S. airports include membrane separation (discussed in Section 8.2.3),
filtration (discussed in Section 8.2.4), mechanical vapor recompression (discussed in Section
8.2.5), and distillation (discussed in Section 8.2.6).
Recovered glycol is generally sold to help recover expenses associated with ADF
application, collection, and control. EPA has observed that on-site recycling can be successful
and economically viable at airports that collect large enough volumes of high-concentration
ADF-contaminated stormwater. Key criteria for designing a recovery system include the type of
ADF being collected, glycol concentration of the stormwater, total consumption of ADF per
season, and peak ADF volume application rates. Other factors to consider are the number of
deicing days per season at an airport and future air traffic plans.
July 2009 8-7
-------
Technical Development Document for Proposed Effluent 8. Pollution Prevention and Treatment Technologies
Limitation Guidelines and Standards for the Airport Deicing Category Applicable to Airport Deicing Operations
8.2.1 Oil/Water Separators and DAF
Oil/water separators increase the glycol concentration of ADF-contaminated stormwater
streams by removing oily constituents from the waste stream. Aboveground skimmer and
underflow weir oil/water separators are used at airports. These types of separators typically
diffuse influent wastewater across the entire face of the separator prior to entering the separation
chamber. The separation chamber consists of perpendicular media designed to separate small oil
droplets from solution. Once the oil floats to the top of the separation chamber a skimmer is
utilized to remove the oil from the surface and store in a separate location (e.g., storage tank).
Static inclined plate oil/water separators are also used at airports and pass floatable liquids over a
single inclined plate (perpendicular). Passing fluid over the inclined plate reduces the velocity of
the influent stream and prevents channeling by spreading out the flow over the entire separator.
An additional benefit provided by plate separators is that they allow sludge or heavy solids to
break away from oil droplets and to settle to the bottom. Oil/water separation is not useful in
removing glycol and other dissolved pollutants in ADF-contaminated stormwater.
Dissolved air flotation (DAF) clarifies wastewaters by removing suspended matter such
as oil or solids. The system dissolves air in wastewater under pressure and then releases the air at
atmospheric pressure in a flotation tank, allowing the released air to form tiny bubbles that
adhere to suspended matter in the wastewater. These bubbles float to the surface where the
suspended matter can be removed by a skimming device. Like oil/water separators, DAF units
are not useful in removing glycols and other dissolved pollutants in ADF-contaminated
stormwater.
EPA conducted a site visit at Seattle-Tacoma International (SEA) airport, which operates
a DAF unit. The airport's DAF treatment process consists of adding coagulation chemicals to the
influent wastewater in a rapid mix chamber, gently mixing the chemicals in a flocculation tank to
encapsulate suspended solids and oil droplets, and removing the floe and other oil particles. Air
bubbles released into the wastewater attach to the suspended solids and colloidal oil particles,
forming a floating material that is removed and pumped to a sludge sump. The treated water in
the DAF units flows over an outlet weir where it is combined with other waters prior to
discharge to Puget Sound. (USEPA, 2007a).
8.2.2 Biological Treatment
This subsection describes the treatment of ADF-contaminated stormwater through
biological processes. Biological treatment consists of two types of processes, aerobic or
anaerobic. The treatment can occur onsite at an airport or off site at POTWs or other treatment
facilities.
POTW Treatment of ADF-contaminated Stormwater
Where practical, airports discharge their deicing stormwater to a POTW for biological
treatment. POTW systems generally operate using activated sludge in an aerobic biological
treatment system and may also incorporate anaerobic digestion of the sludge generated. Airports
may be prevented from discharging to a local POTW due to one or more or the following
reasons: (1) limited hydraulic or loading capacity at the POTW, (2) high POTW wastewater
July 2009 8-8
-------
Technical Development Document for Proposed Effluent 8. Pollution Prevention and Treatment Technologies
Limitation Guidelines and Standards for the Airport Deicing Category Applicable to Airport Deicing Operations
treatment and/or conveyance fees, (3) inability of the local POTW to handle highly variable
pollutant loadings, and (4) airport infrastructure constraints.
Aerated Lagoons or Detention Ponds
Aerated biological treatment systems used on site at airports are effective for treating
low-concentration ADF-contaminated stormwater. These systems commonly consist of aerated
treatment lagoons or ponds that are open to the atmosphere, though open-topped tank systems
may also be used. Treatment lagoons and ponds work well with large volumes (millions of
gallons) of wastewater and are usually operated as a batch process. Aeration in lagoons and
ponds increases the level of dissolved oxygen in the water, which is needed to decompose
organic matter such as glycols. In addition, the presence of oxygen helps to oxidize certain
elements that are suspended in the water and oxidation causes some materials to become heavier
so that they will settle out of the water column quicker. Without proper aeration, bacteria will not
be able to decompose the organic matter in a pond quickly or efficiently. Aeration devices are
used to agitate the lagoon or pond surface, which helps to transfer atmospheric oxygen into the
wastewater to promote biological treatment processes, to vent carbon dioxide and other gaseous
elements from the water and also to increase the amount of wastewater exposed to ambient air,
allowing other volatile organics to oxygenate and evaporate.
The Greater Rockford airport is an example of an airport operating an aerated pond
treatment system for deicing stormwater. Greater Rockford airport collects ADF-contaminated
stormwater throughout the deicing season into a 16-million gallon aerated detention pond. Their
aerobic digestion system consists of the aerated detention pond, a settling pond, a recycling
pump, and a chemical addition building. The biodegradation of glycol is temperature-dependant
and predominantly occurs during the spring and early summer months when ambient
temperatures are higher. Airport personnel monitor the process, adding nutrients, antifoaming
agents, and pH adjustment chemicals as needed. When the BODs concentration of the pond has
been reduced to less than 30 mg/L, airport personnel discharge the treated stormwater from the
settling pond to Rock River (USEPA, 1999).
Figure 8-6 shows an aerated pond installation at Portland Airport.
July 2009 8-9
-------
Technical Development Document for Proposed Effluent 8. Pollution Prevention and Treatment Technologies
Limitation Guidelines and Standards for the Airport Deicing Category Applicable to Airport Deicing Operations
Figure 8-6. Aerated Pond Installation at Portland Airport
Anaerobic Treatment
Anaerobic treatment systems can effectively treat ADF-contaminated stormwater with a
range of glycol concentrations. This type of treatment usually occurs in a closed tank in which
microscopic bacteria in an oxygen-deficient environment biodegrade deicing and anti-icing
materials. In anaerobic treatment systems, microorganisms consume the organic matter and
convert it to methane and carbon dioxide in the absence of oxygen, which creates much less
sludge than an aerobic system.
Anaerobic Fluidized Bed (AFB) treatment is a demonstrated technology for addressing
ADF-contaminated stormwater at both the Albany, New York (ALB) and Akron/Canton, Ohio
(CAK) airports. The AFB treatment system uses a vertical, cylindrical tank in which the ADF-
contaminated stormwater is pumped upwards through a bed of granular activated carbon at a
velocity sufficient to fluidize, or suspend, the media. A thin film of microorganisms grows and
coats each granular activated carbon particle, providing a vast surface area for biological growth.
The anaerobic microorganisms that develop occur naturally in sediment, peat bogs, cattle
intestines, and even brewer's yeast. Breakdown products from the AFB treatment system include
methane, carbon dioxide and new biomass. Effluent from the AFB can be discharged to a local
POTW or, in most cases, directly to surface water. Figure 8-7 presents a diagram showing the
major components of a typical AFB treatment system (ERG, 2007).
July 2009
8-10
-------
Technical Development Document for Proposed Effluent 8. Pollution Prevention and Treatment Technologies
Limitation Guidelines and Standards for the Airport Deicing Category Applicable to Airport Deicing Operations
Blogas to Boiler
andtor Flare
Blofllm
Growth
Control ^
Biofilm Coated
Media
Fluidized
Bod Reactor
Fluid izjtul
Pump ^_
1' Wastewater
Influent
Figure 8-7. Typical Anaerobic Fluid Bed Treatment System for Treatment of Stormwater
Contaminated by ADF
Land Application and Constructed Wetlands
Low-concentration deicing stormwater may also be treated on-site at an airport using
either land application or discharge through constructed wetlands. Land application involves
spraying deicing stormwater onto a land surface for infiltration and biological degradation within
the soil. As an example, Salt Lake City International (SLC) airport, uses land application to
dispose of batch volumes of low-concentration deicing stormwater on a periodic basis. The
system involves spraying approximately 300,000 to 400,000 gallons of stormwater over nutrient-
enriched land using agricultural wheels. The application occurs over a two day period at a rate of
about one gallon per square foot. The sprayed glycol then degrades in the soil over a week to
month long period (USEPA, 2008b). Since glycol-based ADFs readily degrade in both high clay
and sandy soil systems, this type of system can be effective for low concentrations and limited
volumes. Biodegradation occurs in the soil through carbon respiration of soil microbes, which
consume oxygen and release carbon dioxide. Zurich International Airport, Switzerland, also uses
a spray irrigation system for ADF treatment. At that airport, a heated sprinkler system applies
spent ADF to a 20 ha (49.4 acre) area. (Jungo, E. and P. Schob, 2005, Jungo, E. 2005)
Constructed wetlands are artificial marshes or swamps placed inline with stormwater
drainage at airports. The wetlands act as biofilters and help remove sediments and pollutants
such as heavy metals from the wastewater. Physical, chemical, and biological processes combine
in wetlands to remove contaminants from the wastewater. Discharge from constructed wetlands
can also be collected and treated if treatment from the wetlands is insufficient to meet discharge
permit limits or it can be discharged to a POTW or surface water. Airports that use constructed
wetlands for treatment include London Heathrow and Toronto, Canada and in the United States,
EPA is aware of two recently constructed systems at Buffalo Niagara International Airport and
Washington Dulles International Airport.
July 2009
8-11
-------
Technical Development Document for Proposed Effluent 8. Pollution Prevention and Treatment Technologies
Limitation Guidelines and Standards for the Airport Deicing Category Applicable to Airport Deicing Operations
Buffalo Niagara International Airport, Buffalo, NY, installed an engineered wetland
system in 2009. The system consists of a subsurface flow wetland with forced bed aeration.
Figure 8-8 presents a photo of the wetland systems at Buffalo Niagara. (Jacques Whitford
NAWE, 2008, Minkel, K. et. al., 2009) Washington Dulles International Airport has also
recently installed constructed wetland biological treatment units adjacent to the airport's new
runway to provide treatment to runway and fugitive ADF stormwater runoff in the area. The
wetland units utilize a complex system of collection and distribution pipes and the design allows
for even distribution of runoff to maximize effective treatment across each unit. Each wetland
unit incorporates multiple layers of sand, stone and a special soil mixture, topped with specific
surface vegetation including cattails and other plant species native to the mid-Atlantic region.
The vegetation was selected for its tolerance to flooding, extensive root depth, oxygen-carrying
capacity of the root system, and poor wildlife habitat. Contaminated stormwater is piped to the
wetland units where contaminants are absorbed by the plants' root system, broken down by soil
microorganisms, or physically filtered by the media comprising each layer. (ERG, 2009)
8.2.3
Figure 8-8. Engineered Wetlands Installation at Buffalo Niagara Airport
Membrane Separation
Membrane separation is an efficient one- or two-step process utilized in recycling spent
ADF that incorporates ultrafiltration (UF) and/or reverse osmosis (RO). In an UF/RO process,
fluid is filtered at a high temperature (75° C) using an ultrafiltration membrane as stage one.
Next, the deicing fluid (ultrafiltration filtrate) can be dewatered using a reverse osmosis
membrane as stage two. Since the ultrafiltration membrane is effective at removing contaminants
such as turbidity, color and odor, reverse osmosis stage two is used for dewatering and glycol
separation. The combined UF/RO process will produce a final glycol concentration of
approximately 10 percent from an original concentration ranging between 0.5 and 4 percent.
Pittsburgh International airport (PIT) is an example of an airport using this type of system. The
PIT system first treats ADF-contaminated stormwater through an ultrafiltration unit to remove
suspended solids. At this point, the stormwater is 0.5 to 4 percent propylene glycol. The
July 2009
8-12
-------
Technical Development Document for Proposed Effluent 8. Pollution Prevention and Treatment Technologies
Limitation Guidelines and Standards for the Airport Deicing Category Applicable to Airport Deicing Operations
stormwater is next treated through a reverse osmosis unit. Following reverse osmosis, the
stormwater is split into two outputs: 1) concentrate that is approximately 10 percent propylene
glycol, and 2) permeate that has a very low glycol concentration. The concentrated propylene
glycol is transported off site for further processing and the permeate that contains small amounts
of glycol, cBOD and COD can either be discharged to surface water, or to a POTW for further
processing (USEPA, 2006).
8.2.4 Filtration
Primary filtration, which removes solids greater than 10 microns, is commonly the first
step in glycol recycling systems because it removes suspended solids and prevents subsequent
processing units from plugging. Popular primary filters used in glycol recycling are made of
either polypropylene cartridges or bag filters.
8.2.5 Mechanical Vapor Recompression
Mechanical vapor recompression (MVR) is an evaporation method that uses
mechanically driven compressors or blowers to increase the pressure of the vapor produced. The
increase in pressure causes the vapor's temperature to increase, which allows it to heat the liquid
being concentrated. Benefits of using MVR in glycol recycling include:
• Low specific operating costs;
• Low specific energy consumption;
• Short residence times of the product; and
• Simplicity of the process.
EPA conducted a site visit and subsequent sampling episode at Denver International
(DEN) airport, which operates MVRs in the following manner. MVRs concentrate the deicing
pad storage tank influent, which ranges from 1 to 12 percent propylene glycol, to a final
concentration of 50 to 55 percent glycol. Each MVR has a capacity of 25,000 gallons per day.
The effluent from the MVRs goes to a storage tank where it is stockpiled prior to being sent to a
distillation system (USEPA, 2007b).
8.2.6 Distillation
Distillation can be an effective way to recycle glycol by separating the water from the
glycol in ADF-contaminated stormwater. A drawback of distillation is that it creates
contaminated washdown water that cannot be discharged and must be treated further. Distillation
columns are also very large and expensive. Because distillation is energy-intensive, it is
generally not cost-effective to distill and recycle waste glycol solutions at low concentrations
(less than 15 percent). Design variables include temperature, distillation column height, and
reflux ratio. This process is commonly done in batches to ensure proper distillation and desired
results.
EPA conducted site visits at SLC and DEN airports, both of which operate distillation
columns.
July 2009 8-13
-------
Technical Development Document for Proposed Effluent 8. Pollution Prevention and Treatment Technologies
Limitation Guidelines and Standards for the Airport Deicing Category Applicable to Airport Deicing Operations
• At DEN airport, the distillation system runs 24 hours a day for a three-week cycle
and processes about 225,000 gallons of propylene glycol before the system is
halted for cleaning and maintenance. The distillate from the distillation column is
discharged to the airport's storage ponds. The distillation column bottoms
compile sludge that is classified as specialized nonhazardous waste. The 98 to 99
percent propylene glycol from the distillation column is pumped to a polisher
(USEPA, 2007b).
• At SLC airport, 40 to 45 percent glycol-concentrated stormwater is passed
through a finisher to increase the concentration to approximately 70 to 80 percent
glycol and then discharged to a storage tank. The stored concentrate is sent to a
distillation column where it is heated to 250 to 260° F to produce a final product
of 100 percent glycol. Distillate from the distillation column is discharged to a
storage tank and the column bottoms are disposed of in an incinerator (USEPA,
2008b).
8.3 Pollution Prevention and Product Substitution Practices
Pollution prevention practices reduce the generation or discharge of pollutants produced
during aircraft/airfield deicing operations. Pollution prevention practices implemented
throughout the aviation industry include infrared deicing (discussed in Section 8.3.1), forced-air
deicing (discussed in Section 8.3.2), product substitution practices (discussed in Sections 8.3.3
and 8.3.4), and best management practices (BMPs) (discussed in Section 8.3.5).
8.3.1 Infrared Deicing
Infrared heating is the transmission of energy by means of electromagnetic waves or rays.
Infrared energy is invisible and travels at the speed of light in straight lines from the heat source
(the emitter) to all surfaces and objects (the receivers) without significantly heating the space
(air) through which it passes. This heating process is much faster than conventional heating
mechanisms used by conventional deicing (convection and conduction), where the deicing fluid
spray is cooled by ambient air.
Infrared (IR) heating systems have been used at a few U.S. airports for several years and
have been demonstrated to effectively deice aircraft while substantially reducing ADF usage.
Currently, infrared deicing systems are used at two large hub airports, John F. Kennedy and
Newark Liberty, and one non-hub airport, Rhinelander-Oneida County, Wisconsin. Figure 8-9
shows a picture of the infrared hangar at John F. Kennedy (JFK) airport.
July 2009 8-14
-------
Technical Development Document for Proposed Effluent 8. Pollution Prevention and Treatment Technologies
Limitation Guidelines and Standards for the Airport Deicing Category Applicable to Airport Deicing Operations
Figure 8-9. Infrared Hangar at JFK Airport
(Photo courtesy of Radiant Aviation Services, Inc.)
Infrared-based aircraft deicing systems offer two advantages over traditional glycol-based
deicing methods. From an environmental standpoint, they can greatly reduce the amount of
glycol-based fluids used for aircraft deicing, while from an operational standpoint, they are
relatively inexpensive to operate, as they use natural gas or propane as fuel.
Any infrared deicing facility design must take into account the physical characteristics of
all aircraft that will use the system. Design factors include the maximum tail height, the shape of
tails, maximum wingspans, and differences in the length and width of the fuselage. The site
selected for an infrared deicing system must comply with the same FAA regulations that apply to
glycol-based aircraft deicing facilities, including aircraft separation rules, air traffic control tower
line-of-sight criteria, and requirements to not interfere with radar signals, navigational aides, and
airport lighting. FAA issued a new Advisory Circular in 2005 specifically for infrared deicing
facilities (FAA, 2005). As with traditional aircraft deicing facilities, an infrared deicing facility
must provide taxiways that allow aircraft to bypass the deicing facility.
While EPA encourages the use of this technology, industry practice thus far has
suggested that it may not be applicable at all airports. Because IR systems are not widely
available or used, EPA does not propose to identify IR as an available technology for the
purposes of establishing effluent guidelines. However, the Agency may reconsider this
technology, if sufficient data support a conclusion that the technology is available. Documents
provided by a vendor claim that use of an IR system can reduce the amount of Type I ADF
required by up to 90 percent per aircraft (Radiant Aviation, 2008; ERG, 2004; Belcher-Hoppe
Associates, Inc., 2004).
July 2009
8-15
-------
Technical Development Document for Proposed Effluent 8. Pollution Prevention and Treatment Technologies
Limitation Guidelines and Standards for the Airport Deicing Category Applicable to Airport Deicing Operations
8.3.2 Forced Air Deicing
Forced air deicing uses large volumes of air at low pressure to remove loose
accumulations of snow and ice from an aircraft prior to chemical deicing. Aircraft deicing trucks
or fixed booms equipped with forced air nozzles help reduce the amount of glycol that is used
during deicing and defrosting operations. In light snow conditions, forced air may completely
replace the need for deicing fluid. For light frost and light deicing events, fluid-injected forced
air can be used to reduce the amount of deicing fluid by up to 75 percent. For heavy deicing
events, a fluid forced air technique is used. This technique requires the ADF application rate
(gallons per minute or gpm) sprayed from the boom to be reduced from 60 gpm to 40 gpm to
achieve a 25 percent reduction in deicing fluid per aircraft. Forced air deicing can reduce
operational expenses, environmental impacts, and subsequent environmental monitoring or
remediation expenses (Icewolf product flyer; IDS, 2006).
8.3.3 Aircraft Deicing/Anti-Icing Product Substitution Practices
One solution to the environmental problems associated with glycol-based
ADF is replacing such fluids with more environmentally friendly products. ADFs must comply
with Aerospace Material Specifications (AMS) published by SAE International, an independent
standards development organization that works cooperatively with the FAA. These standards—
AMS 1424 for Type I fluids and AMS 1428 standards for Type IV fluids-require a specified
level of product performance and compatibility and any alternative product must meet the same
standards. To be economically viable, alternative products must also be of comparable price and
be at least as effective in maintaining air safety as the glycol-based fluids they replace.
EPA is aware of one non-glycol, plant-derived product currently being marketed by
Cryotech Deicing Technology. A new Type I ADF product called "r>pSustam" uses 1,3-
propanediol rather than propylene glycol, and is manufactured by a fermentation process using
cornstarch. The manufacturer claims performance equal or better than PG or EG-based deicers.
Table 8-1 lists other aircraft deicing alternatives that EPA is aware of based on literature
reviews, industry meetings, and site visits. In addition, the airport and airline industry
associations as well as U.S. Air Force have conducted and are continuing to conduct research
into other potential substitutes for propylene glycol and ethylene glycol-based fluids.
Table 8-1. ADF Alternatives
Alternative
Propylene Glycol
Hot Air, Forced Air, and
Tempered Steam Deicing
Infrared Deicing
Cryotech Bio-PDO™
Warm Fuel for Wing Deicing
Comments
ADF usage data indicates a trend towards greater propylene glycol use as an
alternative to ethylene glycol use.
The use of hot air, forced air, or tempered steam when deicing aircraft provides an
alternative to typical deicing fluid application techniques using deicing trucks with
conventional spray nozzles alone. These alternatives can provide more effective
deicing than conventional spraying technologies and result in lower ADF usage.
Infrared deicing provides an alternative to conventional ADF usage and can greatly
reduce (though not eliminate) the use of ADFs for deicing and anti-icing.
This bio-based product is currently being marketed as an alternative to
conventional propylene glycol-based Type I fluids.
Alternative to defrost deicing with ADF.
July 2009 8-16
-------
Technical Development Document for Proposed Effluent 8. Pollution Prevention and Treatment Technologies
Limitation Guidelines and Standards for the Airport Deicing Category Applicable to Airport Deicing Operations
8.3.4 Airfield Deicing Product Substitution Practices
Environmental problems associated with past airfield deicing products like urea led to the
development of the alternative airfield deicing chemicals used today. Potassium acetate has
replaced urea as the primary airfield deicer at many U.S. airports. The U.S. Armed Forces no
longer purchases airfield deicers that contain urea and instead use potassium acetate, sodium
acetate, and sodium formate. These airfield deicers are highly effective at low temperatures,
exert much lower oxygen demand than urea, and offer less environmental impact.
The aviation industry is currently evaluating the impact of common airfield deicing
chemicals on the environment, runway infrastructure, and chemical corrosion of aircraft carbon
brakes. Based on the results of this work, EPA anticipates that alternative airfield deicing
chemicals will be identified and ultimately incorporated into practice. EPA is aware of one such
product, Cryotech's Bio-PDO™, which is currently being used as an additive for their potassium
acetate runway product. The Cyrotech BX36 runway product has similar performance to their
widely-used E36 product, but with reductions in electrical conductivity, reduction in potassium
content (reducing carbon brake issues), and a bio-based material composition of 75 percent,
allowing for easy degradation.
8.3.5 Best Management Practices (BMPs)
BMPs are techniques used to control stormwater runoff, sediment control, and soil
stabilization, as well as management decisions to prevent or reduce airport pollution. EPA
defines a BMP as a "technique, measure or structural control that is used for a given set of
conditions to manage the quantity and improve the quality of stormwater runoff in the most cost-
effective manner." This subsection describes the following aircraft/airfield deicing-related
BMPs:
• Application rates and deicing fluid dilution;
• Airfield prewetting;
• Ice detection systems;
• Enhanced weather forecasting;
• Heated sand;
• Separation of contaminated snow;
• Annual employee training;
• Mechanical deicing and snow removal;
• Yearly inspections of deicing equipment and infrastructure; and
• Type IV ADF anti-icing.
Application Rates and Deicing Fluid Dilution
Deicing personnel can minimize the amount of deicing fluid used at an airport by varying
deicing fluid application rates and the ADF dilution mix to best match each deicing event
condition. Application rates are commonly evaluated every time deicing is required. However,
ADF is usually pre-mixed to a set 55 percent or 45 percent glycol concentration. Systems that
allow for ADF dilution adjustment based on the weather conditions can result in less ADF usage.
Ice thickness, ambient temperatures, and plane size all determine application rates and fluid
dilution requirements. For small planes with small amounts of frost, as little as 50 gallons of
July 2009 8-17
-------
Technical Development Document for Proposed Effluent 8. Pollution Prevention and Treatment Technologies
Limitation Guidelines and Standards for the Airport Deicing Category Applicable to Airport Deicing Operations
ADF can be used while large planes with thick ice accumulations can take up to 2000 gallons to
deice. Application rates metered by chemical metering systems installed to deicing trucks or
booms allow the applicator to control the distribution of chemicals and maintain a consistent
application rate at all times. Using a chemical metering system also allows the operator to change
the application rate in the midst of application based on changing weather conditions.
Airfield Prewetting
Airfield prewetting involves applying a dry chemical followed by a light coating of liquid
deicer to airfield pavement. During icy conditions, a granular deicer is generally used to
penetrate ice and increase surface area prior to using a liquid deicer (generally propylene glycol)
to help the solid deicer stick to the pavement and prevent dry pellets from blowing off. EPA's
Airport Deicing Questionnaire indicated that prewetting is common and this practice helps to
minimize the cost of materials by increasing the rate at which the liquid deicer contacts icy
surfaces and by minimizing wind losses of solid deicer. Following prewetting, snow and ice are
generally removed from airfield areas using mechanical equipment (such as plows and brooms).
Ice Detection Systems
Ice detection systems include sensors installed on runways and taxiways that transmit
constant surface and subsurface temperature readings to deicing control personnel. These sensors
indicate whether there is a potential for ice to form on the paved surfaces. Airports can then use
the information provided by individual deicer manufacturers to determine whether deicing/anti-
icing chemicals should be applied.
Enhanced Weather Forecasting
Many of the larger U.S. airports use enhanced weather forecasting systems to help
determine when they will require deicing activities. A popular weather forecasting product on the
market is the Weather Support to Deicing Decision Making (WSDDM) software program. This
system is a "nowcasting" system that is used to confirm National Weather Service data and
forecasts. WSDDM provides forecasts, monitors storms and provides real-time storm
information, and estimates and detects precipitation. WSDDM can provide the following
services to the user:
• Real-time snow gauge data (updated every minute) of the liquid equivalent
snowfall rate at the airport and two to three sites 10 to 20 km away from the
airport;
• Real-time radar reflectivity from radars depicting current locations of
precipitation and snow;
• Meteorological data at the airport and two to three sites 10 to 20 km away from
the airport updated every minute and displayed in text and time line form, with
the time line going back to two hours;
• Thirty-minute nowcast of radar reflectivity based on the use of a cross-correlation
technique on the radar reflectivity data updated every six minutes; and
• Thirty-minute nowcast of liquid equivalent snowfall rate at the airport and the off-
site snow gauge locations by applying a real-time snow gauge-radar reflectivity
calibration algorithm at each of the snow gauge sites, updated every six minutes.
July 2009 8-18
-------
Technical Development Document for Proposed Effluent 8. Pollution Prevention and Treatment Technologies
Limitation Guidelines and Standards for the Airport Deicing Category Applicable to Airport Deicing Operations
WSDDM can improve pollution prevention by pinpointing when deicing operations are
actually needed by lowering the amount of ADFs used to keep departing aircraft free of ice and
snow contamination at takeoff, and by lowering the amount of anti-icing chemicals used to
prevent ice and snow from bonding to aircraft and taxiways (ERG, 2005).
Heated Sand
Sand trucks can be parked in a heated garage with heat piped to the truck. Heating sand
trucks accomplishes two things. First, heating the truck prevents the moisture in the sand from
freezing and clumping, which would prevent the sand from being efficiently disbursed when
applied. Second, applying heated sand to icy surfaces can melt ice, which can then refreeze with
the sand providing needed traction and minimizing the need to use ADFs.
Separation of Contaminated Snow
Airports often segregate glycol-contaminated snow (commonly called "pink snow") and
haul it to a designated area where the snow melt can be collected. Deicing pads often contain
designated areas to store contaminated snow so that the snow melt can be commingled with other
deicing stormwater.
Annual Employee Training
An important factor affecting the efficiency of aircraft deicing/anti-icing operations is the
training and experience of personnel performing these operations. Well-trained and experienced
deicing/anti-icing personnel improve the efficiency of aircraft deicing/anti-icing operations and
minimize the volume of ADF used. The training and experience of airport personnel may also
affect the efficiency of airfield deicing operations. Airport personnel are typically responsible for
clearing taxiways, gate areas, ramps, aprons, and deicing pads. When these areas are not
adequately cleared, snow and ice accumulate on the undercarriage and the underside of aircraft
during taxing and must be removed prior to takeoff.
Many airports conduct annual operations and maintenance training, with specialized
winter operations training usually conducted prior to the deicing season. This specialized training
can include but is not limited to proper use of equipment, application of chemicals, and location
of snow melt areas. Staff may also be trained in environmental awareness detailing the
requirements of airport permits and BMPs associated with airport deicing/anti-icing operations.
Mechanical Deicing and Snow Removal
The amount of ADF required to deice aircraft can be minimized by mechanically deicing
the aircraft prior to chemical deicing. Mechanical deicing is generally economical only for small
aircraft since mechanical deicing of large aircraft is labor-intensive and time consuming. A
drawback of mechanical deicing is that aircraft (e.g., aircraft antennae and sensors) risk being
damaged by incorrect mechanical deicing methods. Despite the risk of aircraft damage, many
airlines use brooms, squeegees, and ropes, among other items, to remove ice and snow from
aircraft surfaces. These methods are more effective at removing snow rather than ice.
Snow is commonly mechanically removed on airfield pavement, including passenger
ramps, gate positions, taxiways, and runways, prior to ADF being applied, to prevent
July 2009 8-19
-------
Technical Development Document for Proposed Effluent 8. Pollution Prevention and Treatment Technologies
Limitation Guidelines and Standards for the Airport Deicing Category Applicable to Airport Deicing Operations
contamination of the snow. The following types of equipment are used to remove snow from
airfield surfaces: self-propelled snow brooms, high speed snow blowers and snow plows, and
utility trucks or tractors fitted with snow brooms or plows. Physical removal is generally used for
snow rather than ice.
Yearly Inspections of Deicing Equipment and Infrastructure
Inspections are conducted to ensure that equipment used for deicing operations is
working properly and to determine where maintenance may be needed. Storage tanks are
inspected to ensure there are no leaks. ADF application equipment is inspected to determine if
gauges are working properly and that fluid is not being spilled. Trench or square drains are
inspected to ensure there is no clogging and that water conveyed through the drain does not
escape. Other equipment and airport infrastructure may require yearly inspections to make
certain that deicing chemicals are applied with properly functioning equipment and collected
with suitable infrastructure.
Type IV ADF Anti-icing
Type IV ADF protects aircraft from ice, snow, or slush accumulations on cleaned aircraft
surfaces. Type IV fluids form a protective film on treated surfaces, protecting against ice
formation and/or snow accumulation. Pretreating aircraft with Type IV fluid is an anti-icing
technique sometimes used when ice is in the forecast and an aircraft is expected to remain on the
ground for an extended period of time. Applying Type IV ADF anti-icing fluid can reduce the
amount of deicing fluid needed for an aircraft by reducing the amount of ice that forms on the
aircraft.
8.4 References
CASQA. 2003. Vegetated Swale, from "California Stormwater BMP Handbook: Industrial and
Commercial" (January 1). California Stormwater Quality Association.
http://www.cabmphandbooks.com/Industrial.asp .
USEPA. 2008a. Final Engineering Site Visit Report for Minneapolis-St. Paul International
Airport. U.S. Environmental Protection Agency/Office of Water. Washington, D.C. (January 24).
DCNAD00793.
FAA. 2007. Federal Aviation Administration Advisory Circular No. 150/5200-33B. "Hazardous
Wildlife Attractants on or Near Airports." (August 28).
http://www.faa.gov/airports airtraffic/airports/resources/advisory circulars/.
USEPA. 2007a. Final Engineering Site Visit Report for Seattle - Tacoma International Airport.
U.S. Environmental Protection Agency/Office of Water. Washington, D.C. (July 1). DCN
AD00778.
USEPA. 1999. Final Engineering Site Visit Report for Greater Rockford International Airport.
U.S. Environmental Protection Agency/Office of Water. Washington, D.C. DCN T10402.
July 2009 8-20
-------
Technical Development Document for Proposed Effluent 8. Pollution Prevention and Treatment Technologies
Limitation Guidelines and Standards for the Airport Deicing Category Applicable to Airport Deicing Operations
ERG. 2007. Memorandum from Juliana Stroup and Mary Willett (ERG) to Brian D'Amico
(EPA): Aircraft Deicing Stormwater Control Technologies and Their Removal Efficiencies.
(December 17). DCN AD00855.
USEPA. 2008b. Final Engineering Site Visit Report for Salt Lake City International Airport.
U.S. Environmental Protection Agency/Office of Water. Washington, D.C. (January 24). DCN
AD00792.
Jungo, E and P. Schob, 2005. "Disposal of De-Icing Effluents by Irrigation" Paper presented at
International Water Association Leading Edge Conference on Water and Wastewater Treatment
Technologies, Sapparo, Japan. (6-8 June). DCN AD01200.
Jungo, E, 2005. "Disposal of De-Icing Effluents by Irrigation; Zurich Airport" PowerPoint
presentation. (November). DCN AD01199.
Jacques Whitford NAWE, 2008. "Project: Buffalo Airport" Project Description Sheet. DCN
AD01186.
Minkel, K. et. al., 2009. "Innovative Approaches to Managing Stormwater Runoff: Constructed
Wetlands." Presentation at FAA 32nd Annual Airport Conference, Hershey PA. (March 4). DCN
AD01184.
ERG, 2009. Memorandum from Steve Strackbein and Mary Willett (ERG) to Brian D'Amico
and Eric Strassler (EPA): Washington Dulles Runway 1L-19R Construction Project Details and
Update. (June 29). DCN ADO 1193.
USEPA. 2006. Final Engineering Site Visit Report for Pittsburgh International Airport. U.S.
Environmental Protection Agency/Office of Water. Washington, D.C. (November 1). DCN
AD00774.
USEPA. 2007b. Final Engineering Site Visit Report for Denver International Airport. U.S.
Environmental Protection Agency/Office of Water. Washington, D.C. (November 4). DCN
AD00779.
FAA. 2005. Federal Aviation Administration Advisory Circular 120-89. "Ground De-Icing using
Infrared Energy." (December 13).
Radiant Aviation. 2008. Radiant Infrared Aircraft Deicing Systems. Product Flyer. DCN
AD01198.
ERG. 2004. Aircraft Deicing using an Infrared System called Ice-Cat. (December 13). DCN
AD00095.
Becher-Hoppe Associates, Inc. 2004. Airport Facilities - Rhinelande -Oneida County Airport.
(December 6). DCN AD00097.
Icewolf. The Evolution of Aircraft Deicing Systems, Icewolf, Product Flyer. DCN AD00096.
July 2009 8-21
-------
Technical Development Document for Proposed Effluent 8. Pollution Prevention and Treatment Technologies
Limitation Guidelines and Standards for the Airport Deicing Category Applicable to Airport Deicing Operations
IDS. 2006. Deicing the IDS Way, Product Flyer (January 4). Integrated Deicing Services. DCN
D00077.
ERG. 2005. Memorandum from Steve Strackbein (ERG) to Project File: Summary of Weather
Support to Deicing Decision Making (WSDDM): A Winter Weather Now casting System.
(January 3). DCN AD00185.
July 2009 8-22
-------
Technical Development Document for Proposed Effluent 9. Control and Treatment Performance
Limitation Guidelines and Standards for the Airport Deicing Category
9. CONTROL AND TREATMENT PERFORMANCE
This section describes the control and treatment scenarios that EPA evaluated for the
proposed airport deicing rulemaking. Each scenario is comprised of groups of ADF collection
activities and a wastewater treatment technology identified to reduce or eliminate the discharge
of pollutants from airports. EPA identified these technologies from technical literature (including
case studies and previous rulemaking efforts), responses to EPA airport and airline surveys, and
EPA site visits and sampling episodes.
The Agency does not require sites to implement these specific activities and technologies
to comply with the proposed rulemaking; sites can implement any technology (or completely
eliminate their discharge through contract hauling or recycling) as long as they achieve the
proposed effluent limitations.
Section 9.1 summarizes the criteria EPA used in selecting the proposed control and
treatment scenarios and Section 9.2 describes the performance efficiencies achieved by those
scenarios. Sections 9.3 and 9.4 describe the development of the long-term mean concentrations
for chemical oxygen demand (COD) and ammonia, respectively, using long-term performance
data from Albany International (ALB) airport.
9.1 Development of Proposed Control and Treatment Scenarios
EPA reviewed several different ADF-contaminated stormwater control and treatment
technologies, as described in Section 8. These technologies were assessed against the following
criteria to determine those technologies to be considered for proposal:
• Their ability to collect and contain ADF-contaminated stormwater;
• The prevalence of their use by the aviation industry;
• Engineering aspects of the technology affecting its ease of use; and
• Their ability to reduce pollutant loadings.
Based on its evaluation, EPA identified three ADF-contaminated stormwater collection
technologies and one stormwater treatment technology for further evaluation and incorporation
into the proposal options. The collection technologies identified included glycol recovery
vehicles (GRVs), GRVs in combination with plug and pump systems, and deicing pads. The
treatment technology identified for evaluation was anaerobic fluidized bed (AFB) biological
treatment. There are also several recycle and recovery alternatives capable of achieving effective
pollutant reductions. These technologies are described below along with the reasons for their
selection.
9.1.1 GRVs
As described in Section 8.1.1, GRVs collect spent aircraft and airfield deicing fluids as
well as any slush or snow from gate areas, ramps, aircraft parking areas, taxiways, and aircraft
holding pads. The GRV is a high-airflow, wide-sweep-path vacuum used to efficiently collect
and separate glycol and ice. This technology works most efficiently when it is used to collect
runoff soon after deicing is completed and with as little dilution as possible from direct
precipitation and nondeicing runoff. EPA selected GRVs for further evaluation since they are
commonly used by airports now for deicing stormwater collection, can be used at both the gate
July 2009 9-1
-------
Technical Development Document for Proposed Effluent 9. Control and Treatment Performance
Limitation Guidelines and Standards for the Airport Deicing Category
and centralized deicing locations, can collect deicing stormwater at the point of generation, and
represent a low-cost collection technology that does not require infrastructure changes.
Limitations noted in using GRVs include the following: potential problems with ponding
of deicing stormwater during heavy precipitation, added vehicular traffic in congested gate areas,
irregular ground surfaces undermining collection effectiveness, and slush clogging the collection
systems under heavy snow conditions.
9.1.2 GRVs in Combination with Plug and Pump Systems
Plug and pump systems intercept and temporarily hold deicing-contaminated stormwater
close to the source using existing storm sewers. Plug and pump systems are often supplemented
with GRVs. Drainage blocks, consisting of valves or inflatable plugs, are installed within the
existing drains to prevent concentrated deicing storm water from entering the drainage system
and discharging to surface water. Deicing stormwater is then collected using pumps or GRVs,
and transported elsewhere for treatment or processing. Blocking mechanisms can be opened or
removed during non-deicing periods to allow uncontaminated stormwater to drain normally.
EPA selected GRVs in combination with plug and pump systems for further evaluation since
they are well documented as a collection technology at many airports, allow deicing stormwater
to be collected at the gate, and involve minimal infrastructure changes compared to other types
of collection technologies.
An alternative to the plug and pump system is installing diversion equipment (e.g.
diversion valves) within the airport's existing stormwater drainage system. In a stormwater
diversion system, the stormwater is routed to storage prior to treatment or discharge, instead of
being blocked from drainage areas. A well-operated diversion system, operating in combination
with GRVs, can be considered an equivalent technology to a plug and pump system with GRVs.
9.1.3 Deicing Pads
Centralized deicing pads restrict aircraft deicing to a controlled, confined area,
minimizing the volume of fugitive spent deicing fluid and allowing the deicing waste to be
captured. A deicing pad is specially graded to capture and route contaminated runoff to tanks. If
the pads are located near gate areas or at the heads of runways, planes can be deiced just prior to
takeoff; as a result, less Type IV anti-icing fluid may be necessary for shorter holdover times,
reducing the amount of glycols released onto the runway or into the air. The goal is to
concentrate aircraft deicing activity in a centralized location to minimize containment areas and
runoff volumes and maximize concentrations of deicers in that runoff. EPA selected deicing pads
for further evaluation since they represent state-of-the-art for spent deicing fluid collection and
can significantly reduce the amount of deicing stormwater generated, increase the amount of
concentrated deicing stormwater collected, and therefore create additional opportunities for cost-
effective recycle and recovery operations.
9.1.4 AFB Biological Treatment
AFB biological treatment is a demonstrated technology for treating ADF-contaminated
stormwater. This treatment technology is currently used at two airports, the ALB airport and
Akron-Canton Regional (CAK) airport, and is planned for at least one additional airport in the
near future. The AFB treatment system uses a vertical, cylindrical tank in which the ADF-
July 2009 9-2
-------
Technical Development Document for Proposed Effluent 9. Control and Treatment Performance
Limitation Guidelines and Standards for the Airport Deicing Category
contaminated stormwater is pumped upwards through a bed of granular activated carbon at a
velocity sufficient to fluidize, or suspend, the media. A thin film of microorganisms grows and
coats each granular activated carbon particle, providing a vast surface area for biological growth.
The anaerobic microorganisms that develop occur naturally in sediment, peat bogs, cattle
intestines, and even brewer's yeast. These microorganisms provide treatment of the ADF-
contaminated stormwater. AFB treatment system by-products include methane, carbon dioxide,
and new biomass. Effluent from the AFB can be discharged to a local POTW or, in most cases,
directly to surface water.
Treating wastes using an anaerobic biological system as compared to an aerobic system
offers several advantages. The anaerobic system requires less energy since aeration is not
required and the anaerobic system produces less than 10 percent of the sludge of an aerobic
process. In addition, because the biological process is contained in a sealed reactor, odors are
eliminated. EPA selected AFB biological treatment for further evaluation since it represents the
best technology currently in use by airports to treat deicing stormwater prior to direct discharge.
9.1.5 Recycle/Recovery Operations
Recycling and recovery operations for spent ADF may, in some cases, be able to achieve
pollutant reductions equivalent to AFB treatment and could perform as alternative technologies
to AFB. Two of these technologies, which EPA evaluated in its sampling program, include
mechanical vapor recompression (MVR) followed by distillation used at the Denver
International (DEN) airport and ultrafiltration (UF)/ reverse osmosis (RO) used at the Pittsburgh
International (PIT) airport. The systems EPA evaluated are both used prior to indirect discharge
and have not been demonstrated to sufficiently treat ADF-contaminated stormwater prior to
direct discharge. Both of these systems are discussed below.
MVR followed by distillation is a demonstrated system used to recycle and recover spent
ADF. The system is typically used when glycol concentrations in ADF-contaminated stormwater
are greater than 5 percent; this type of a system is not generally practical for lower concentration
glycol-contaminated stormwater, which would typically be discharged to a POTW for treatment.
The MVR/distillation technology generates a concentrated glycol stream (containing greater than
99 percent glycol) that can be sold as a chemical feedstock. The effluents from the
MVR/distillation system contain propylene glycol, carbonaceous BOD (cBOD), and COD, and
must be discharged to a POTW for further treatment.
UF/RO technology is also a demonstrated recycle and recovery system for spent ADF.
The technology generates a concentrated glycol stream that can be recovered and contract hauled
off site for resale as a chemical feedstock or recycled for possible use in toilets onboard
commercial aircraft (i.e., as lavatory fluid). The effluent from the UF/RO system contains small
amounts of glycol, cBOD, and COD and can either be discharged to surface water or to a POTW
for further processing.
9.2 Performance of Control and Treatment Scenarios
EPA evaluated the performance of the selected control and treatment scenarios based on
their ability to collect and contain ADF-contaminated stormwater and/or their ability to reduce
July 2009 9-3
-------
Technical Development Document for Proposed Effluent 9. Control and Treatment Performance
Limitation Guidelines and Standards for the Airport Deicing Category
pollutant loadings. The performance data used to assess the technology effectiveness include
both EPA-collected data and industry-submitted data.
9.2.1 GR V Collection (20 Percent Efficiency Scenario)
EPA identified performance data on GRVs from several airports. The Gerald R. Ford
International (GRR) airport (Grand Rapids, MI) system, which is based on using two tow-behind
glycol collection units in conjunction with catch basin inserts to collect aircraft deicing runoff
around the terminal gates and apron, reported collecting 29 percent of all applied glycol during
the 2005/2006 deicing season. Mass balance data on glycol usage and collection at Theodore
Francis Green State (PVD) airport (Providence, RI) between 2002 and 2006 indicate that its
glycol-collection-vehicle-based system annually collects between 26 and 48 percent of all
applied glycol. At Milwaukee's General Mitchell International (MKE) airport, a system that
combines mobile glycol collection with a plug and pump approach using in-line sewer valves
and balloons around the terminal apron collected between 22.5 and 33 percent of all applied
glycol in the deicing seasons between 2002 and 2006. At Buffalo Niagara International (BUF)
airport, a system consisting of two glycol collection vehicles operating at the gates combined
with an apron collection system and effective snow management allowed the airport to collect
approximately 53 percent of applied glycol. (ERG, 2007) Overall, collection efficiencies of
applied glycol from these airports ranged from 22.5 to 53 percent, although these systems also
used some combination of catch basin inserts, plug and pump technology, and/or apron systems.
Data summarized in the Transportation Research Board (TRB) Airport Cooperative
Research Program Report No. 14, Deicing Planning Guidelines and Practices for Stormwater
Management Systems reported collection efficiencies between 23 - 48 percent for glycol
collection vehicles. (TRB Airport Cooperative Research Program, 2009) Based on these data,
EPA estimates that GRVs alone recover 20 percent of applied glycol.
9.2.2 Plug and pump Collection (40 Percent Efficiency Scenario)
For many airports, a plug and pump system can be easily deployed and operational in a
short timeframe. Data summarized in the Transportation Research Board (TRB) Airport
Cooperative Research Program Report No. 14, Deicing Planning Guidelines and Practices for
Stormwater Management Systems reported collection efficiencies between 20-35 percent for
plug and pump systems. (TRB Air Cooperative Research Program, 2009) Plug and pump
systems typically enhance the effectiveness of a GRV collection system; therefore, EPA assumed
that facilities effectively operating a plug and pump system in combination with GRV collection
systems recover 40 percent of applied glycol.
9.2.3 Deicing Pad Collection (60 Percent Efficiency Scenario)
EPA reviewed data on the performance of centralized deicing facilities from a number of
larger airports across the United States and Europe. PIT airport observed collection efficiencies
at the centralized deicing pads of between 60 and 66 percent over the varying winter seasons
(CDM, 2006). Analysis of 10 years of records of ADF usage, glycol collection, glycol recycling,
and discharges to the POTW for the DEN airport's deicing runoff control program indicates an
average annual collection efficiency of 64 percent of applied glycols, ranging between 44 and 70
percent annually (Denver International Airport, 2006). Analysis of daily mass balance
monitoring data collected between 1999 and 2006 at the Detroit Metropolitan Wayne County
July 2009 9-4
-------
Technical Development Document for Proposed Effluent 9. Control and Treatment Performance
Limitation Guidelines and Standards for the Airport Deicing Category
(DTW) airport indicates that the central deicing facility-based program recycled 45 to 51 percent
of applied glycol. Additional glycol at lower concentrations is captured and sent to POTW
treatment, but the monitoring program design does not support estimation of this fraction. The
collection efficiencies reviewed from U.S. airports was consistent with information on European
airports. Aha et al., reported collection of applied glycol at European airports ranging from 80
percent at Oslo Airport to 51 percent at Munich. (Aha, et. al, 2005a, Aha, et. al., 2005b)
Additionally, Baltimore-Washington International (BWI) airport operates a deicing runoff
control system that uses central deicing pads coupled with limited at-gate deicing, isolation and
containment using trench drains and glycol recovery vehicles, and management of contaminated
snow from the designated deicing areas. Between 1998 and 2006, glycol recovery ranged from
59 to 86 percent of total glycol applied, and averaged 69 percent (Williams, 2006).
Overall, collection efficiencies at centralized deicing pads from these airports and as
reported in the Transportation Research Board (TRB) Airport Cooperative Research Program
Report No. 14, ranged from 44 to 86 percent of applied glycol. (TRB Air Cooperative Research
Program, 2009) Several airports reported a relatively consistent efficiency of around 60 to 66
percent over the varying winter seasons. Based on this information, EPA estimates that facilities
effectively operating a centralized deicing pad recover 60 percent of applied glycol.
9.2.4 AFB Treatment Performance
EPA collected data on the effectiveness of AFB treatment systems through literature
review, its own sampling efforts, and industry-supplied data. These systems have demonstrated
effective treatment for targeted pollutants, including COD, biochemical oxygen demand (BODs),
and glycol. Based on EPA's sampling data from ALB airport, COD was reduced by greater than
97 percent, BOD5 was reduced by greater than 98 percent, and glycol was reduced by greater
than 99 percent (ERG, 2007).
9.3 References
ERG. 2007. Memorandum from Juliana Stroup and Mary Willett (ERG) to Brian D'Amico
(EPA): Aircraft Deicing Stormwater Control Technologies and Their Removal Efficiencies.
(December 17). DCN AD00855.
TRB Airport Cooperative Research Program, 2009. Deicing Planning Guidelines and Practices
for Stormwater Management Systems ACRP Report No. 14. DCN AD01191.
CDM. 2006. 2006 Deicing Action Plan Update - Pittsburgh International Airport (Excerpts).
Denver International Airport 2006. Design Standards Manual, Aircraft Deicing - DRAFT.
Aha, S. et al. 2005a. Munich Airport Trip Report: USA.
Aha, S., et al. 2005b. Oslo Airport Trip Report.
Williams, M. 2006. Baltimore-Washington International Airport Deicing Mass Balance.
July 2009 9-5
-------
Technical Development Document for Proposed Effluent 10. Pollutant Loadings and
Limitation Guidelines and Standards for the Airport Deicing Category Pollutant Load Reduction Estimates
10. POLLUTANT LOADINGS AND POLLUTANT LOAD REDUCTION ESTIMATES
Pollutant loadings from airport deicing operations are highly variable and airport-
specific. Because the use of deicing and anti-icing chemicals is weather-dependent, the pollutant
loadings at each airport vary from year to year based on weather conditions; the pollutant
loadings also vary from airport to airport based on each airport's climate. In addition, the amount
of applied chemical that is discharged to surface water is airport-specific, based on the existing
stormwater separation, collection, and/or containment system present at each airport.
Due to the variations of the pollutant discharges, EPA determined that it would not be
appropriate to develop baseline pollutant loadings using end-of-pipe monitoring data. Monitoring
data on an airport's deicing stormwater outfall(s) provide only a "snap shot" of a single point in
time, during a monthly monitoring event or, in some cases, a single storm event. In addition,
these data are available only for a select number of airports and are airport-specific. Although
these data provide information on the types of pollutants present in airport deicing stormwater
and the range of concentrations that may reach outfall points, there is insufficient basis for
extrapolating or transferring the data across large timeframes (e.g., an entire winter season) or to
other airports.
Therefore, EPA developed a pollutant loading estimation methodology based on the use
of aircraft deicing/anti-icing fluid (ADF) and airfield chemicals at the airports surveyed by EPA.
The methodology takes into account EPA's existing data sources and will provide a better
estimate of the loadings than those based on sporadic monitoring data alone. This section
discusses the data sources available to EPA to support its pollutant loadings and loading
reduction estimates (Section 10.1), an overview of EPA's pollutant loading methodology
(Section 10.2), and a summary of the calculation steps that make up EPA's loading methodology
(Sections 10.3 through 10.6). Section 10.7 summarizes EPA's approach for estimating loading
reductions associated with the discontinued use of urea as an airfield deicing chemical.
10.1 Data Sources
EPA considered the following available data when developing a pollutant loadings
estimation methodology for airport deicing operations (see Section 4 for more information about
the data sources):
• Pavement deicing chemical usage/purchase information for the 2002/2003,
2003/2004, and 2004/2005 deicing seasons, as reported by airport personnel in the
airport questionnaire;
• ADF purchase information for the 2002/2003, 2003/2004, and 2004/2005 deicing
seasons, as reported by airline personnel in the airline detailed questionnaire;
• Standard airport information available from the Federal Aviation Administration
(FAA), including the number of operations and departures by airport;
• Weather information for each airport from National Oceanic and Atmospheric
Administration (NOAA), including temperature, freezing precipitation, and
snowfall data;
• Existing airport stormwater collection and containment systems, as reported by
airport personnel in the airport questionnaire or during EPA site visits;
July 2009 10-1
-------
Technical Development Document for Proposed Effluent 10. Pollutant Loadings and
Limitation Guidelines and Standards for the Airport Deicing Category Pollutant Load Reduction Estimates
• Standard chemical information about ADF and pavement deicing chemicals,
including molecular formulas and densities; and
• Analytical data from EPA sampling episodes of airport deicing operations.
10.2 Pollutant Loadings Methodology Overview
EPA developed the following methodology to estimate pollutant loadings based on the
available data listed above:
• Step 1: Estimate the amount of ADF and pavement deicing chemicals applied at
each airport during a typical winter season;
• Step 2: Calculate the amount of pollutant load associated with the applied
chemicals, based on the chemical properties of the chemicals;
• Step 3: Estimate the amount of the applied chemical's pollutant load that is
discharged directly, based on the airport's existing stormwater collection,
containment and/or treatment system; and
• Step 4: Estimate pollutant loading removals for each EPA control/treatment
scenario.
The subsections below describe each step in detail.
10.3 Step 1: Estimate the Amount of Applied ADF and Pavement Deicing
Chemicals
To develop pollutant loadings estimates associated with airport deicing operations, EPA
first estimated the amount of applied pavement deicing chemicals and ADF based on data
collected in EPA's airport and airline detailed questionnaires. EPA requested data on three
winter seasons (2002/2003, 2003/2004, and 2004/2005); these data were averaged for each
airport to account for any variability in the severity of the winter weather over those three years.
In the airport questionnaire, EPA requested usage data for pavement deicing chemicals. In the
airline screener and detailed questionnaires, EPA requested ADF purchase data instead of usage
data. During the airline questionnaire development process, airport and airline personnel reported
to EPA that purchase data are documented and tracked more frequently than usage data and are a
good indicator of the amount of chemical used because operations do not stockpile the chemicals
from year to year. The reported amount of chemical purchased was used as a surrogate for the
amount of chemical applied.
EPA sent the airport questionnaire to 153 airports, which were selected using a stratified
random sample of airports based on airport type (i.e., hub, non-hub), the number of snow or
freezing precipitation (SOFP) days, and the number of departures. For more information about
the airport questionnaire, see Section 4.3.
The airline questionnaire collected information from 58 airlines that indicated in their
screener questionnaire that plane deicing was conducted on their aircraft. These 58 airlines
July 2009 10-2
-------
Technical Development Document for Proposed Effluent 10. Pollutant Loadings and
Limitation Guidelines and Standards for the Airport Deicing Category Pollutant Load Reduction Estimates
represent 448 airline/airport combinations. For more information about the airline questionnaire,
see Section 4.3.
10.3.1 Pavement Deicing and Pavement Anti-Icing Chemical Usage Estimate
In the airport questionnaire, EPA requested that airport personnel report the
purchase/usage amount, concentration, and brand name of the following pavement deicing
materials for the 2002/2003, 2003/2004, and 2004/2005 deicing seasons:
• Urea;
• Potassium Acetate;
• Calcium Magnesium Acetate (CMA);
• Sodium Acetate;
• Sand;
• Sodium Formate;
• Ethylene Glycol-Based Fluids;
• Propylene Glycol-Based Fluids; and
Other: (Specify).
EPA evaluated the data provided for each reported chemical to determine the most
appropriate way to estimate the average amount used over the three winter seasons reported. In
addition, EPA read the comments provided by the airport personnel to determine any extenuating
circumstances that affect chemical use. For example, airport personnel may have reported that
urea was replaced with potassium acetate at the airport during the three years reported. In this
case, EPA used the potassium acetate average in the loadings estimate and did not use any of the
urea data to better represent the airport's current practices. Ninety airports reported pavement
deicing chemical usage values to EPA in their questionnaire responses.
The three-year average pavement deicing chemical usage EPA calculated from the
reported data (normalized to pure chemical) is shown by airport and chemical in Table 10-1. In
the questionnaire, airports reported pounds or gallons of pavement deicing chemical used along
with the concentration of the chemical. EPA calculated the amount of applied chemical by
multiplying the reported mass of each chemical by the reported concentration. If airport
personnel reported a volume of chemical in the airport or airline detailed questionnaire, EPA
multiplied the reported volume by the reported concentration and the chemical density.
No airports reported the use of CMA. Multiple airports reported the use of sand, but these
data are not included in Table 10-1, because sand was not included in the loads analysis. Only
one airport reported the use of granular potassium acetate. Potassium acetate (in the liquid form)
is the most commonly used pavement deicing chemical.
July 2009 10-3
-------
Technical Development Document for Proposed Effluent
Limitation Guidelines and Standards for the Airport Deicing Category
10. Pollutant Loadings and
Pollutant Load Reduction Estimates
Table 10-1. Three-Year Average Amount of Pavement Deicing Chemical Usage, in Pounds
Airport
ID
1006
1007
1010
1011
1012
1013
1014
1016
1017
1018
1020
1021
1022
1023
1024
1026
1028
1029
1031
1032
1036
1041
1043
1044
1050
1052
Airport Name
Chicago O'Hare International
Yeager
Fairbanks International
Lambert - St Louis International
Ted Stevens Anchorage International
Wiley Post- Will Rogers Mem
Albuquerque International Sunport
Tri - State/Milton J Ferguson Field
Austin Straubel International
Piedmont Triad International
Hartsfield - Jackson Atlanta
International
Buffalo Niagara International
Fort Wayne International
Seattle - Tacoma International
Indianapolis International
Dallas/Fort Worth International
Denver International
La Guardia
Richmond International
Austin - Bergstrom International
Baltimore - Washington International
Glacier Park International
Ralph Wien Memorial
Roanoke Regional/Woodrum Field
Aspen - Pitkin County /Sardy Field
Wilmington International
Ethylene
Glycol-
Based Fluids
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
Granular
Potassium
Acetate
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
20,333
0
Potassium
Acetate
272,987
0
218,896
5,822,755
1,015,403
0
4,586
27,089
38,121
57,869
314,456
0
248,180
97,914
839,922
18,179
3,350,089
1,062,816
245,463
17,483
1,151,158
0
39,525
131,536
46,732
0
Sodium
Acetate
0
2,560
0
0
0
0
0
0
0
0
0
0
0
696
506,000
0
0
1,747
17,000
0
306,560
0
0
0
0
0
Sodium
Formate
0
0
0
0
0
0
0
0
0
0
0
7,760
0
0
0
0
0
0
0
0
156,800
0
0
0
0
0
Propylene
Glycol-
Based Fluids
6,069,745
10,298
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
420,794
0
0
0
0
0
0
0
0
Urea
0
26,650
383,333
0
1,670,733
20,000
0
62,667
41,540
98,667
0
0
267,963
0
0
0
0
0
0
0
0
333
10,000
0
0
0
Source: Airport Deicing Operations Loads Database (USEPA, 2008a).
July 2009
-------
Technical Development Document for Proposed Effluent
Limitation Guidelines and Standards for the Airport Deicing Category
10. Pollutant Loadings and
Pollutant Load Reduction Estimates
Table 10-1 (Continued)
Airport
ID
1053
1057
1058
1059
1063
1065
1066
1068
1069
1070
1071
1074
1078
1079
1080
1082
1084
1085
1088
1089
1090
1094
1095
1098
1100
1101
Airport Name
General Edward Lawrence Logan
International
Will Rogers World
Gerald R Ford International
Greater Rochester International
Evansville Regional
Albany International
Salt Lake City International
Eppley Airfield
Cleveland - Hopkins International
City of Colorado Springs Municipal
Tweed - New Haven
South Bend Regional
Nashville International
Manchester
Syracuse Hancock International
Trenton Mercer
Bismarck Municipal
Waterloo Municipal
Outagamie County Regional
John F Kennedy International
Boise Air Terminal/Gowen Field
Boeing Field/King County International
Chicago Midway International
Aberdeen Regional
Toledo Express
Portland International
Ethylene
Glycol-
Based Fluids
1,045,791
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
Granular
Potassium
Acetate
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
Potassium
Acetate
0
42,583
62,433
298,335
28,825
196,535
255,058
250,525
2,947,968
198,504
0
0
163,779
318,109
11,792
12,393
16,114
0
229,290
167,054
127,865
10,255
1,259,838
21,857
196,535
256,478
Sodium
Acetate
0
2,910
0
0
0
150,000
0
4,527
196,438
0
291
0
0
0
0
0
0
0
0
3,812,682
0
0
0
0
0
10,292
Sodium
Formate
0
0
4,975
0
0
0
0
0
6,247
0
0
0
0
0
0
4,704
0
0
0
0
0
0
0
0
0
211,183
Propylene
Glycol-
Based Fluids
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
220,469
0
0
0
0
0
0
Urea
5,700
0
0
0
0
0
1,467,340
0
0
0
0
32,440
0
22,500
0
0
0
6,000
0
0
405,677
0
0
0
0
0
Source: Airport Deicing Operations Loads Database (USEPA, 2008a).
July 2009
-------
Technical Development Document for Proposed Effluent
Limitation Guidelines and Standards for the Airport Deicing Category
10. Pollutant Loadings and
Pollutant Load Reduction Estimates
Table 10-1 (Continued)
Airport
ID
1103
1104
1105
1107
1108
1109
1110
1111
1112
1113
1114
1116
1117
1118
1119
1120
1121
1123
1124
1126
1128
1129
1136
1137
Airport Name
Juneau International
Nome
Spokane International
Pittsburgh International
Louisville International - Standiford
Field
Airborne Airpark
Aniak
Port Columbus International
Deadhorse
Cincinnati/Northern Kentucky
International
Stewart International
Reno/Tahoe International
Cherry Capital
Bethel
Rickenbacker International
Rapid City Regional
Theodore Francis Green State
James M Cox Dayton International
Des Moines International
Minneapolis - St Paul
International/Wold - Chamberlain
Charlotte/Douglas International
Bradley International
General Mitchell International
Dallas Love Field
Ethylene
Glycol-
Based Fluids
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
Granular
Potassium
Acetate
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
Potassium
Acetate
0
39,307
240
1,550,440
747,299
1,841,386
548,790
14,413
2,655,839
78,614
26,205
63,474
0
72,849
0
141,287
151,917
211,821
1,117,715
109,356
481,947
1,199,246
218
Sodium
Acetate
0
0
0
13,333
0
0
0
0
0
0
2,520
0
0
0
0
0
0
0
81,157
64,667
0
218,817
0
0
Sodium
Formate
0
0
0
111,067
109,760
1,354,767
0
6,226
0
4,000
0
0
0
0
23,520
0
0
0
0
0
0
0
127,400
0
Propylene
Glycol-
Based Fluids
0
0
951
0
0
0
0
0
0
0
0
0
0
0
0
6,484
0
0
0
0
0
0
0
0
Urea
508,000
0
638,667
0
0
0
2,400
0
20,000
0
151,800
7,238
0
66,000
0
0
0
0
0
0
232,950
16,748
0
0
Source: Airport Deicing Operations Loads Database (USEPA, 2008a).
July 2009
-------
Technical Development Document for Proposed Effluent
Limitation Guidelines and Standards for the Airport Deicing Category
10. Pollutant Loadings and
Pollutant Load Reduction Estimates
Table 10-1 (Continued)
Airport
ID
1138
1139
1140
1141
1142
1144
1145
1146
1147
1148
1149
1150
1151
1153
Airport Name
Detroit Metropolitan Wayne County
Philadelphia International
Memphis International
Ronald Reagan Washington National
Washington Dulles International
Central Wisconsin
Newark Liberty International
Northwest Arkansas Regional
Raleigh - Durham International
Kansas City International
Fort Worth Alliance
Greater Rockford
Kalamazoo/Battle Creek International
Akron - Canton Regional
Ethylene
Glycol-
Based Fluids
0
0
0
0
0
1,858
0
264,162
0
0
0
0
0
0
Granular
Potassium
Acetate
0
0
0
0
0
0
0
0
0
0
0
0
0
0
Potassium
Acetate
1,771,433
0
496,699
353,544
2,823,474
58,674
2,945,968
0
0
604,017
5,241
339,913
0
19,494
Sodium
Acetate
0
0
74,325
45,667
617,580
0
7,760
0
0
3,267
0
0
4,000
162
Sodium
Formate
0
0
0
0
0
0
0
0
0
0
0
0
0
0
Propylene
Glycol-
Based Fluids
0
1,011,565
0
0
0
0
0
0
0
0
0
0
0
0
Urea
0
0
0
62,667
0
84,393
0
27,000
89,333
0
0
655,067
0
21,333
Source: Airport Deicing Operations Loads Database (USEPA, 2008a).
July 2009
-------
Technical Development Document for Proposed Effluent 10. Pollutant Loadings and
Limitation Guidelines and Standards for the Airport Deicing Category Pollutant Load Reduction Estimates
10.3.2 ADF Usage Estimate
EPA requested the purchase amount, concentration, and brand name in the airline
detailed questionnaire for the 2002/2003, 2003/2004, and 2004/2005 deicing seasons for the
following ADF chemicals:
• Type I Ethylene Glycol-Based Fluid (EG Type I);
• Type II Ethylene Glycol-Based Fluid (EG Type II);
• Type IV Ethylene Glycol-Based Fluid (EG Type IV);
• Type I Propylene Glycol-Based Fluid (PG Type I);
• Type II Propylene Glycol-Based Fluid (PG Type II);
• Type III Propylene Glycol-Based Fluid (PG Type III);
• Type IV Propylene Glycol-Based Fluid (PG Type IV); and
• Isopropyl Alcohol-Based Fluid.
Questionnaire responses provided sufficient data to estimate ADF usage at 56 airports. In
some cases, data were not available for every airline operating at an airport. In these instances,
EPA extrapolated the amount of ADF used at the reporting airlines to estimate the total amount
of ADF used by the entire airport, based on the number of airport operations (departures) at the
reporting airlines and the total amount of airport operations. Table 10-2 presents the ADF
estimates based on airline questionnaire responses. No airports reported the purchase of EG
Type II, EG Type III, PG Type II, PG Type III, or Isopropyl Alcohol-Based Fluid.
In addition to the ADF data reported in the airline detailed questionnaire, 10 airports
reported their estimate of total annual ADF usage to EPA in the comment section of the airport
questionnaire (see Table 10-3). These ADF data were combined with the ADF data reported in
the airline detailed questionnaires, resulting in 66 airports with total PG/EG (gallons) usage
estimates.
July 2009 10-8
-------
Technical Development Document for Proposed Effluent
Limitation Guidelines and Standards for the Airport Deicing Category
10. Pollutant Loadings and
Pollutant Load Reduction Estimates
Table 10-2. ADF Estimates Based on Airline Detailed Questionnaire Responses
Airport
ID
1003
1006
1010
1011
1012
1013
1021
1022
1024
1026
1028
1029
1036
1037
1043
1047
1053
1058
1059
1065
1066
1069
1074
1079
1080
1089
1095
1100
1103
1104
1105
1107
1109
1110
Airport Name
Ketchikan International
Chicago O'Hare International
Fairbanks International
Lambert - St Louis International
Ted Stevens Anchorage International
Wiley Post- Will Rogers Mem
Buffalo Niagara International
Fort Wayne International
Indianapolis International
Dallas/Fort Worth International
Denver International
La Guardia
Baltimore - Washington
International
George Bush Intercontinental
Airport/Houston
Ralph Wien Memorial
Sacramento Mather
General Edward Lawrence Logan
International
Gerald R Ford International
Greater Rochester International
Albany International
Salt Lake City International
Cleveland - Hopkins International
South Bend Regional
Manchester
Syracuse Hancock International
John F Kennedy International
Chicago Midway International
Toledo Express
Juneau International
Nome
Spokane International
Pittsburgh International
Airborne Airpark
Aniak
Estimated
PG/EG
(GPY)
18,182
1,516,626
83,335
325,122
420,735
3,056
281,836
50,412
452,155
166,790
1,043,138
485,157
323,623
10,242
2,500
1,282
995,249
98,156
229,158
125,775
570,540
582,321
29,586
177,307
186,351
560,031
293,834
46,449
48,014
3,047
67,984
943,982
432,416
476
PG
Type I
(%)
0
80
0
23
0
9
92
92
91
43
87
75
90
82
27
100
82
86
91
93
22
90
75
87
97
82
88
64
0
15
92
88
74
100
PG
Type IV
(%)
0
20
0
1
0
0
9
8
9
12
10
22
10
18
0
0
17
13
9
7
6
10
25
13
o
3
18
12
5
0
0
8
12
26
0
EG
Typel
(%)
100
0
100
70
100
91
0
0
0
38
4
2
0
0
73
0
0
0
0
0
52
0
0
0
0
0
0
29
100
85
0
0
0
0
EG
Type IV
(%)
0
0
0
6
0
0
0
0
0
7
0
1
0
0
0
0
0
0
0
0
20
0
0
0
0
0
0
2
0
0
0
0
0
0
Source: Airport Deicing Operations ADF Usage Database (USEPA, 2008b).
Note: PG/EG gallons represent total usage normalized to 100 percent glycol. Values may not sum to 100 due to
rounding.
GPY - Gallons per year.
July 2009 10-9
-------
Technical Development Document for Proposed Effluent
Limitation Guidelines and Standards for the Airport Deicing Category
10. Pollutant Loadings and
Pollutant Load Reduction Estimates
Table 10-2 (Continued)
Airport
ID
1111
1113
1117
1118
1123
1124
1126
1128
1129
1136
1138
1139
1140
1141
1142
1145
1148
1149
1150
1151
1152
1153
Airport Name
Port Columbus International
Cincinnati/Northern Kentucky
International
Cherry Capital
Bethel
James M Cox Dayton International
Des Moines International
Minneapolis - St Paul
International/Wold - Chamberlain
Charlotte/Douglas International
Bradley International
General Mitchell International
Detroit Metropolitan Wayne County
Philadelphia International
Memphis International
Ronald Reagan Washington National
Washington Dulles International
Newark Liberty International
Kansas City International
Fort Worth Alliance
Greater Rockford
Kalamazoo/Battle Creek
International
Duluth International
Akron - Canton Regional
Estimated
PG/EG
(GPY)
288,374
715,836
11,524
4,897
90,580
79,658
1,456,537
143,572
427,068
152,944
2,152,292
979,983
199,174
219,533
1,076,083
1,123,057
203,726
1,522
146,856
22,002
68,168
60,246
PG
Type I
(%)
92
24
75
40
89
84
93
81
88
90
93
88
88
81
77
86
75
97
79
84
96
90
PG
Type IV
(%)
8
5
0
0
11
14
7
19
12
9
7
12
12
16
22
14
8
3
21
16
4
10
EG
Typel
(%)
0
61
0
60
0
3
0
0
0
0
0
0
0
o
3
i
0
17
0
0
0
0
0
EG
Type IV
(%)
0
11
25
0
0
0
0
0
0
1
0
0
0
0
0
0
0
0
0
0
0
0
Source: Airport Deicing Operations ADF Usage Database (USEPA, 2008b).
Note: PG/EG gallons represent total usage normalized to 100 percent glycol. Values may not sum to 100 due to
rounding.
GPY - Gallons per year.
July 2009 10-10
-------
Technical Development Document for Proposed Effluent
Limitation Guidelines and Standards for the Airport Deicing Category
10. Pollutant Loadings and
Pollutant Load Reduction Estimates
Table 10-3. ADF Data Reported in the Airport Questionnaire
Airport ID
1115
1062
1072
1060
1096
1097
1025
1143
1001
1019
Airport Name
Jacksonville International
Birmingham International
Gillette-Campbell County
Williamson County Regional
Santa Fe Municipal
Lovell Field
Tupelo Regional
San Francisco International
Montgomery Regional (Dannelly Field)
Ontario International
PG/EG (GPY)
1,000
5,000
880
150
1,108
4,148
820
105
232
35
Source: Airport Deicing Operations ADF Usage Database (USEPA, 2008b).
GPY - Gallons per year.
Using the airline and airport questionnaire data on ADF purchases, airport departure data,
and climate data, EPA correlated the estimate of the amount of ADF used to the climate and size
of each airport. EPA created an "ADF Factor" to estimate the relative amount of deicing
occurring at each airport based on the airport's climate and number of departures. EPA
calculated the ADF Factor by multiplying the 30-year annual average number of SOFP days by
the average number of annual departures at each airport during 2004-2006 (USEPA, 2008b).
EPA graphed the total gallons of PG/EG purchased with the ADF factor and determined the
equation of the line. During this analysis, EPA noted a difference in the relationship of ADF
Factor and ADF usage for Alaskan airports compared to other airports. Due to this difference,
EPA developed a separate graph and equation for Alaskan airports. The graph for non-Alaskan
airports is presented in Figure 10-1; the graph for Alaskan airports is presented in Figure 10-2.
2,500,000 -i
2 000 000
PG/EG
Gallons 1 500 000
per Year
1,000,000 -
500 000
0 4
T^osoTSx I
R2 = 0.7622 •
^^ |
.^^
^^^ . I
^^^
• ^^^^
^^^
* _^^*»
{Jj&*
0 2,000,000 4,000,000 6,000,000 8,000,000
ADF Factor (SOFP Days x Average Annual Departures)
Figure 10-1. ADF Factor vs. PG/EG Gallons for U.S. Airports (excluding Alaska)
July 2009
10-11
-------
Technical Development Document for Proposed Effluent
Limitation Guidelines and Standards for the Airport Deicing Category
10. Pollutant Loadings and
Pollutant Load Reduction Estimates
500,000 -,
400 000
PG/EG 300000 -
Gallons
Per Year
200 000
100 000
y = 0.0632X
R2 = 0.7031 »
^^
^^
AT*. A A A
0 2,000,000 4,000,000 6,000,000
ADF Factor (SOFP Days x Average Annual Departures)
Figure 10-2. ADF Factor vs. PG/EG Gallons for Alaskan Airports
EPA used the line equations to estimate the total gallons of ADF used at airports that did
not have available ADF data in the airport or airline detailed questionnaires. Based on the
estimated total gallons of ADF used at an airport, EPA calculated the distribution of different
types of ADF (PG/EG, Type I/Type IV) based on the average percent distribution of the reported
ADF amounts. See the Airport Deicing Loadings Calculations memorandum (ERG, 2008d) for
more detail. Table 10-4 presents the final estimates of ADF usage for all airports that received
the airport questionnaire.
10.4
Step 2:Calculate the Amount of Pollutant Load Associated with the Applied
Chemicals
As deicing chemicals break down in the environment, they increase chemical oxygen
demand (COD) and biochemical oxygen demand (BOD). EPA calculated the amount of COD
and BOD (presented as 5 day BOD, or BOD5) associated with the degradation of the applied
deicing/anti-icing chemicals.
EPA considered two approaches to estimate the amount of COD and BOD5 associated
with deicing chemicals. The first approach involved using laboratory empirical COD and BOD5
data for deicing chemicals. The second approach involved using standard chemical information
and stoichiometric equations to estimate COD and BODs for each chemical.
EPA determined it would not be suitable to use empirical data to estimate loadings for
three main reasons. First, empirical COD and BOD5 data were not readily available for all
deicing/anti-icing chemicals. Second, the available empirical data were outdated and brand-
specific. Finally, chemical formulations vary significantly over time and by brand, so it is
inappropriate to apply any given set of empirical data to all airports and chemicals.
July 2009
10-12
-------
Technical Development Document for Proposed Effluent
Limitation Guidelines and Standards for the Airport Deicing Category
10. Pollutant Loadings and
Pollutant Load Reduction Estimates
Table 10-4. ADF Annual Usage Estimates for All Airports that Received the Airport
Questionnaire
Airport
ID
1001
1002
1003
1004
1005
1006
1007
1008
1009
1010
1011
1012
1013
1014
1015
1016
1017
1018
1019
1020
1021
1022
1023
1024
1025
1026
1027
1028
1029
1030
1031
1032
1033
1034
1035
1036
Airport Name
Montgomery Regional (Dannelly Field)
Bert Mooney
Ketchikan International
Norfolk International
Roberts Field
Chicago O'Hare International
Yeager
Tucson International
Cold Bay
Fairbanks International
Lambert-St Louis International
Ted Stevens Anchorage International
Wiley Post- Will Rogers Mem
Albuquerque International Sunport
Gulfport-Biloxi International
Tri-State/Milton J. Ferguson Field
Austin Straubel International
Piedmont Triad International
Ontario International
Hartsfield - Jackson Atlanta
International
Buffalo Niagara International
Fort Wayne International
Seattle-Tacoma International
Indianapolis International
Tupelo Regional
Dallas/Fort Worth International
Craven County Regional
Denver International
La Guardia
Williamsport Regional
Richmond International
Austin-Bergstrom International
Me Carran International
Metropolitan Oakland International
San Diego International
Baltimore-Washington International
PG Type I
(gallons)
166
9,722
0
22,084
10,720
1,213,301
35,450
1,675
43,759
0
74,778
0
275
46,107
1,109
9,259
44,442
49,169
25
259,100
259,289
46,379
112,631
411,461
587
71,720
628
907,530
363,868
7,204
42,442
17,198
7,613
0
0
291,261
PG Type IV
(gallons)
22
1,293
0
2,938
1,426
303,325
4,715
223
5,821
0
3,251
0
0
6,133
148
1,232
5,912
6,540
3
34,465
25,365
4,033
14,982
40,694
78
20,015
84
104,314
106,735
958
5,646
2,288
1,013
0
0
32,362
EG Type I
(gallons)
41
2,400
18,182
5,451
2,646
0
8,750
413
10,800
83,335
227,586
420,735
2,781
11,380
274
2,285
10,969
12,136
6
63,950
0
0
27,799
0
145
63,380
155
41,726
9,703
1,778
10,476
4,245
1,879
0
0
0
EG Type IV
(gallons)
3
177
0
402
195
0
646
31
797
0
19,507
0
0
840
20
169
810
896
0
4,720
0
0
2,052
0
11
11,675
11
0
4,852
131
773
313
139
0
0
0
Source: Airport Deicing Operations ADF Usage Database (USEPA, 2008b).
Note: Values may not sum to total usage amounts due to rounding.
July 2009
10-13
-------
Technical Development Document for Proposed Effluent
Limitation Guidelines and Standards for the Airport Deicing Category
10. Pollutant Loadings and
Pollutant Load Reduction Estimates
Table 10-4 (Continued)
Airport
ID
1037
1038
1039
1040
1041
1042
1043
1044
1045
1046
1047
1048
1049
1050
1051
1052
1053
1054
1055
1056
1057
1058
1059
1060
1061
1062
1063
1064
1065
1066
1067
1068
1069
1070
Airport Name
George Bush Intercontinental
Airport/Houston
Luis Munoz Marin International
Kahului
Louis Armstrong New Orleans
International
Glacier Park International
Orlando International
Ralph Wien Memorial
Roanoke Regional/Woodrum Field
Norman Y. Mineta San Jose
International
Long Island Mac Arthur
Sacramento Mather
Redding Municipal
Lanai
Aspen-Pitkin Co/Sardy Field
Barnstable Muni-Boardman/Polando
Field
Wilmington International
General Edward Lawrence Logan
International
Jackson Hole
Miami International
Santa Maria Pub/Capt G Allan Hancock
Field
Will Rogers World
Gerald R. Ford International
Greater Rochester International
Williamson County Regional
William P Hobby
Birmingham International
Evansville Regional
Falls International
Albany International
Salt Lake City International
Helena Regional
Eppley Airfield
Cleveland-Hopkins International
City of Colorado Springs Municipal
PG Type I
(gallons)
8,399
0
0
0
27,578
0
675
23,552
0
31,135
1,282
495
0
10,742
33,008
1,556
816,104
24,413
0
0
35,409
84,414
208,534
107
10,134
3,578
14,412
8,137
116,971
125,519
13,147
79,386
524,089
54,230
PG Type IV
(gallons)
1,844
0
0
0
3,668
0
0
3,133
0
4,141
0
66
0
1,429
4,391
207
169,192
3,247
0
0
4,710
12,760
20,624
14
1,348
476
1,917
1,082
8,804
34,232
1,749
10,560
58,232
7,214
EG Type I
(gallons)
0
0
0
0
6,807
0
1,825
5,813
0
7,685
0
122
0
2,651
8,147
384
0
6,026
0
0
8,740
0
0
26
2,501
883
3,557
2,008
0
296,681
3,245
19,594
0
13,385
EG Type IV
(gallons)
0
0
0
0
502
0
0
429
0
567
0
9
0
196
601
28
0
445
0
0
645
0
0
2
185
65
263
148
0
114,108
240
1,446
0
988
Source: Airport Deicing Operations ADF Usage Database (USEPA, 2008b).
Note: Values may not sum to total usage amounts due to rounding.
July 2009
10-14
-------
Technical Development Document for Proposed Effluent
Limitation Guidelines and Standards for the Airport Deicing Category
10. Pollutant Loadings and
Pollutant Load Reduction Estimates
Table 10-4 (Continued)
Airport
ID
1071
1072
1073
1074
1075
1077
1078
1079
1080
1081
1082
1083
1084
1085
1086
1087
1088
1089
1090
1091
1092
1093
1094
1095
1096
1097
1098
1099
1100
1101
1102
1103
1104
1105
1106
1107
1108
1109
Airport Name
Tweed-New Haven
Gillette-Campbell County
Honolulu International
South Bend Regional
Pensacola Regional
Kona International at Keahole
Nashville International
Manchester
Syracuse Hancock International
Bob Hope
Trenton Mercer
Tampa International
Bismarck Municipal
Waterloo Municipal
Palm Beach International
El Paso International
Outagamie County Regional
John F Kennedy International
Boise Air Terminal/Gowen Fid
Rochester International
Lewiston-Nez Perce County
Los Angeles International
Boeing Field/King County International
Chicago Midway International
Santa Fe Municipal
Lovell Field
Aberdeen Regional
Sacramento International
Toledo Express
Portland International
John Wayne Airport-Orange County
Juneau International
Nome
Spokane International
Fort Lauderdale/Hollywood
International
Pittsburgh International
Louisville International-Standiford Field
Airborne Airpark
PG Type I
(gallons)
3,523
630
0
22,189
592
0
65,479
154,257
180,760
0
3,858
0
15,018
5,594
732
11,608
41,375
459,225
51,086
24,717
17,781
0
3,688
258,574
793
2,968
9,997
0
29,728
80,173
0
0
457
62,545
0
830,704
91,849
319,988
PG Type IV
(gallons)
469
84
0
7,396
79
0
8,710
23,050
5,591
0
513
0
1,998
744
97
1,544
5,504
100,806
6,795
3,288
2,365
0
491
35,260
105
395
1,330
0
2,322
10,664
0
0
0
5,439
0
113,278
12,217
112,428
EG Type I
(gallons)
870
155
0
0
146
0
16,161
0
0
0
952
0
3,707
1,381
181
2,865
10,212
0
12,609
6,101
4,389
0
910
0
196
733
2,467
0
13,470
19,788
0
48,014
2,590
0
0
0
22,670
0
EG Type IV
(gallons)
64
11
0
0
11
0
1,193
0
0
0
70
0
274
102
13
211
754
0
931
450
324
0
67
0
14
54
182
0
929
1,461
0
0
0
0
0
0
1,673
0
Source: Airport Deicing Operations ADF Usage Database (USEPA, 2008b).
Note: Values may not sum to total usage amounts due to rounding.
July 2009
10-15
-------
Technical Development Document for Proposed Effluent
Limitation Guidelines and Standards for the Airport Deicing Category
10. Pollutant Loadings and
Pollutant Load Reduction Estimates
Table 10-4 (Continued)
Airport
ID
1110
1111
1112
1113
1114
1115
1116
1117
1118
1119
1120
1121
1122
1123
1124
1125
1126
1127
1128
1129
1130
1131
1132
1133
1134
1135
1136
1137
1138
1139
1140
1141
1142
1143
1144
1145
1146
Airport Name
Aniak
Port Columbus International
Deadhorse
Cincinnati/Northern Kentucky
International
Stewart International
Jacksonville International
Reno/Tahoe International
Cherry Capital
Bethel
Rickenbacker International
Rapid City Regional
Theodore Francis Green State
Southwest Florida International
James M Cox Dayton International
Des Moines International
Sarasota/Bradenton International
Minneapolis-St Paul International/Wold-
Chamberlain
Willow Run
Charlotte/Douglas International
Bradley International
San Antonio International
Wilkes-Barre/Scranton International
Chippewa Valley Regional
Phoenix Sky Harbor International
St George Municipal
Lafayette Regional
General Mitchell International
Dallas Love Field
Detroit Metropolitan Wayne County
Philadelphia International
Memphis International
Ronald Reagan Washington National
Washington Dulles International
San Francisco International
Central Wisconsin
Newark Liberty International
Northwest Arkansas Regional
PG Type I
(gallons)
476
265,304
56,478
171,801
23,086
716
53,382
8,643
1,959
7,661
18,185
107,383
680
80,616
66,913
0
1,354,580
7,313
116,293
375,820
9,119
30,426
9,486
0
9,978
1,065
137,650
26,622
2,001,632
862,385
175,273
177,822
828,584
75
31,203
965,829
22,013
PG Type IV
(gallons)
0
23,070
7,513
35,792
3,071
95
7,101
0
0
1,019
2,419
14,284
90
9,964
11,152
0
101,958
973
27,279
51,248
1,213
4,047
1,262
0
1,327
142
13,765
3,541
150,660
117,598
23,901
35,125
236,738
10
4,151
157,228
2,928
EG Type I
(gallons)
0
0
13,940
436,660
5,698
177
13,176
0
2,938
1,891
4,488
26,504
168
0
2,390
0
0
1,805
0
0
2,251
7,510
2,341
0
2,463
263
0
6,571
0
0
0
6,586
10,761
19
7,702
0
5,433
EG Type IV
(gallons)
0
0
1,029
78,742
421
13
973
2,881
0
140
331
1,956
12
0
0
0
0
133
0
0
166
554
173
0
182
19
1,529
485
0
0
0
0
0
1
568
0
401
Source: Airport Deicing Operations ADF Usage Database (USEPA, 2008b).
Note: Values may not sum to total usage amounts due to rounding.
July 2009
10-16
-------
Technical Development Document for Proposed Effluent
Limitation Guidelines and Standards for the Airport Deicing Category
10. Pollutant Loadings and
Pollutant Load Reduction Estimates
Table 10-4 (Continued)
Airport
ID
1147
1148
1149
1150
1151
1152
1153
Airport Name
Raleigh-Durham International
Kansas City International
Fort Worth Alliance
Greater Rockford
Kalamazoo/Battle Creek International
Duluth International
Akron - Canton Regional
PG Type I
(gallons)
73,281
152,794
1,477
116,016
18,482
65,442
54,221
PG Type IV
(gallons)
9,748
16,298
46
30,840
3,520
2,727
6,025
EG Type I
(gallons)
18,087
34,633
0
0
0
0
0
EG Type IV
(gallons)
1,335
0
0
0
0
0
0
Source: Airport Deicing Operations ADF Usage Database (USEPA, 2008b).
Note: Values may not sum to total usage amounts due to rounding.
July 2009
10-17
-------
Technical Development Document for Proposed Effluent
Limitation Guidelines and Standards for the Airport Deicing Category
10. Pollutant Loadings and
Pollutant Load Reduction Estimates
EPA selected the second approach, calculating loadings based on standard chemical
information and stoichiometric equations. The advantage of this methodology over using
empirical data is that it can be used for all deicing chemicals. In addition, this methodology
allows the calculations and assumptions used to be clearly presented. EPA checked the validity
of the COD and BOD5 concentrations for propylene glycol and ethylene glycol calculated using
this methodology against the available empirical data and found a good match.
10.4.1
Calculate the Total Mass of Each Pollutant
First, EPA estimated the total mass of each chemical based on the airline and airport
questionnaire responses (which specified varying formulations of ADF and deicing products). To
calculate the total mass of applied chemical, EPA multiplied the reported mass of each chemical
by the reported concentration of the chemical. Alternatively, if airport personnel reported a
volume of chemical in the airport or airline detailed questionnaire, EPA multiplied the reported
volume by the reported concentration and the chemical density.
10.4.2
Determine the Theoretical Oxygen Demand of Each Chemical
Next, EPA determined the theoretical oxygen demand (ThOD) associated with the
degradation of each of the deicing chemicals. The ThOD estimate was based on the molecular
formula of the chemical and the stoichiometric equation of the breakdown of the chemical to the
end products of carbon dioxide and water. Table 10-5 lists the calculated ThOD for each
chemical.
Table 10-5. Theoretical Oxygen Demand Calculations for Deicing Chemicals
Deicing Compound
Propylene glycol
Ethylene glycol
Urea
Potassium acetate
Sodium acetate
Calcium magnesium acetate
Sodium formate
Molecular
Formula
C3H802
C2H602
N2H4CO
KC2H302
NaC2H3O2
C8H12CaMg08
NaHCO2
Stoichiometric Formula
C3H8O2 + 4 O2 -> 3 CO2 + 4 H2O
2[C2H6O2] + 5 O2 -> 4 CO2 + 6 H2O
N2H4CO + 4 O2 -> 2 HNO3 + CO2 + H2O
[C2H3O2]~ + 1.75 O2 -> 1 CO2 + 1.5 H2O
[C2H3O2]~ + 1.75 O2 -> 2 CO2 + 1.5 H2O
[C8H12O8]4~ + 7 O2 -> 8 CO2 + 6 H2O
2[HCO2]~ + 0.5 O2 -> 2 CO2 + H2O
ThOD (Moles of O2
per Mole of Deicing
Compound)
4.0
2.5
4.0
1.75
1.75
7.0
0.25
10.4.3
Determine the COD of Each Chemical
EPA next determined the COD loading associated with the chemical's degradation. EPA
assumed that the chemical would completely degrade in the environment over time and therefore
the calculated ThOD load would be equivalent to the COD load. EPA estimated the COD load
associated with each reported chemical based on the calculated mass of the chemical purchased,
the molecular weight of the chemical, the ThOD, and the molecular weight of oxygen, using the
equation below:
July 2009
10-18
-------
Technical Development Document for Proposed Effluent 10. Pollutant Loadings and
Limitation Guidelines and Standards for the Airport Deicing Category Pollutant Load Reduction Estimates
/-^>T->T Af A \ /^u • i/ ,1 \ 434grams . . f moles of chemical
COD Load (pounds) = Chemical (pounds) x - x Chemical Molecular Weight -
pound ^ grams of chemical y
„, _^f moles of O2 ^ /grams of O2^| pound
x ThOD - x O2 Molecular Weight
moles of chemical ) ^ moles of Ch ) 434 grams
10.4.4 Determine the BOD5 of Each Chemical
EPA calculated the BOD5 loading based on the estimated COD loading. EPA developed
an industry-specific relationship between COD and BODs using analytical data for untreated
deicing stormwater from the EPA sampling episodes at Albany International airport, Pittsburgh
International airport, and Denver International airport. The average COD/BOD5 ratio was 1.67.
This relationship was used to calculate the BOD5 associated with the degradation of the deicing
chemical. See the Airport Deicing Loading Calculations memorandum (ERG, 2008d) for more
information.
10.5 Step 3: Estimate the Amount of Baseline Pollutant Load that is Discharged
Directly
The amount of applied chemical that is discharged directly is airport-specific and
dependant upon the existing stormwater collection/treatment system present at each airport.
Typically, ADF is applied at a number of specific locations at the airport, including gates,
deicing pads, and/or aprons. Pavement deicing chemicals are applied on a larger area and variety
of locations, including runways, taxiways, aprons, and gates. Based on EPA site visits and
questionnaire responses, EPA assumed that pavement deicing chemicals could be present in
almost every airport outfall, whereas ADF is usually present at a smaller number of outfalls that
drain only aircraft deicing areas.
10.5.1 Direct Discharge of Pavement Deicers
EPA estimated the amount of pavement deicers that are directly discharged. Because
pavement deicing chemicals are applied at a large variety of areas at an airport, the amount of
pavement deicers being directly discharged could range from close to 100 percent on pavement
areas near outfall drains, to nearly 0 percent for chemicals that may fall onto grassy areas and
infiltrate into the ground during a thaw. Estimating a percentage of direct discharge release of
pavement deicers at a particular airport is difficult without performing a detailed study of each
airport. Therefore, EPA assumed 100 percent direct discharge of pavement deicers to represent
the maximum possible amount of discharge.
10.5.2 Direct Discharge of ADF
EPA estimated the direct discharge amount of ADF by first estimating the amount of
applied ADF that would be available for discharge. EPA assumed that 80 percent of applied
Type I ADF falls onto the pavement at the deicing area and is available for discharge; the
remaining 20 percent is lost to evaporation, wind, or tire tracking, or adheres to the plane and is
later sheared off during taxiing and takeoff (Switzenbaum, et al., 1999). EPA assumed that 10
percent of Type IV ADF falls to the pavement in the deicing area and is available for discharge;
the remaining 90 percent adheres to the plane. The Agency multiplied the total amount of applied
July 2009 10-19
-------
Technical Development Document for Proposed Effluent 10. Pollutant Loadings and
Limitation Guidelines and Standards for the Airport Deicing Category Pollutant Load Reduction Estimates
ADF by the appropriate percentage available for discharge to determine the amount of ADF that
is available for discharge.
Next, EPA determined the percentage of available ADF that would be directly discharged
at each airport, depending on the airport's current control and treatment systems (ERG, 2008c).
EPA estimated collection and control percentages of spent ADF for each airport based on
information provided during EPA site visits and in the airport questionnaire. If the airport did not
provide an ADF collection and control percentage estimate, EPA personnel reviewed the
airport's questionnaire responses and the reported collection and control percentage of similar
systems to determine an estimate for the airport. Table 10-6 in Section 10.6 presents the
collection percentages for each airport.
The COD load associated with each ADF chemical applied at an airport was reduced by
EPA's estimate of the airport's current collection and control percentage to estimate the amount
of ADF directly discharged. These estimates represent the baseline amount of ADF discharged to
the environment. Table 10-6 in Section 10.6 presents EPA's estimate of the amount of ADF load
directly discharged by each airport, as measured in pounds of COD.
10.6 Step 4: Estimate Pollutant Loading Removals for Each EPA
Collection/Control Scenario
EPA's regulatory options require a specific collection and treatment percentage of
available (spent) ADF. EPA evaluated three control and treatment scenarios as discussed in
Section 10:
• 20% Efficiency Scenario: collection and treatment of 20 percent of spent ADF;
• 40% Efficiency Scenario: collection and treatment of 40 percent of spent ADF;
and
• 60% Efficiency Scenario: collection and treatment of 60 percent of spent ADF.
EPA estimated the direct discharge COD load of each collection and control scenario
accounting for the following two components:
• The COD load associated with the applied ADF, minus any reductions achieved
by the collection and control practices implemented; and
• The COD load that would be discharged from anaerobic fluidized bed (AFB)
biological treatment.
EPA estimated the amount of COD load that would be discharged from treatment for
each airport that had load reductions in a scenario. Using analytical data from its sampling
episodes, EPA determined that AFB systems remove 97.5 percent of COD. Therefore, EPA
assumed that 97.5 percent of the COD load going to treatment would be removed and 2.5 percent
would be discharged.
After estimating the loads for each scenario, EPA estimated the loading reductions as
compared to baseline. Table 10-6 lists the ADF COD baseline loads and reductions for each
control and treatment scenario evaluated for proposal.
July 2009 10-20
-------
Technical Development Document for Proposed Effluent
Limitation Guidelines and Standards for the Airport Deicing Category
10. Pollutant Loadings and
Pollutant Load Reduction Estimates
Table 10-6. ADF COD Baseline Loads and Loading Reductions for Each Control and
Treatment Scenario, by Airport
Airport
Montgomery Regional (Dannelly Field)
Bert Mooney
Ketchikan International :
Norfolk International
Roberts Field
Chicago O'Hare International
Yeager
Tucson International
Cold Bay
Fairbanks International
Lambert-St Louis International
Ted Stevens Anchorage International
Wiley Post- Will Rogers Mem
Albuquerque International Sunport
Gulfport-Biloxi International
Tri-State/Milton J. Ferguson Field
Austin Straubel International
Piedmont Triad International
Ontario International
Hartsfield - Jackson Atlanta International
Buffalo Niagara International
Fort Wayne International
Seattle-Tacoma International
Indianapolis International
Tupelo Regional
Dallas/Fort Worth International
Craven County Regional
Denver International
La Guardia
Williamsport Regional
Richmond International
Current
ADF
Collection
0
100
NA
20
100
40
40
20
0
60
60
40
0
20
100
0
40
0
0
60
40
0
0
40
0
60
0
93
0
60
40
Baseline
COD Load
(pounds)
2,359
0
0
251,092
0
8,733,878
302,292
19,045
621,909
319,152
1,230,807
2,416,962
29,821
524,227
0
131,595
378,973
698,801
356
1,472,952
1,832,048
545,428
1,600,739
2,907,633
8,339
593,712
8,928
777,648
4,487,109
40,953
361,921
COD Load
Reduction
for 20%
Collection
and Control
Scenario
(pounds)
460
0
0
0
0
0
0
0
121,272
0
0
0
5,815
0
0
25,661
0
136,266
214
0
0
109,086
312,144
0
1,626
0
5,357
0
874,986
0
0
COD Load
Reduction
for 40%
Collection
and Control
Scenario
(pounds)
920
0
0
61,204
0
0
0
4,642
242,545
0
0
0
11,630
127,780
0
51,322
0
272,532
214
0
0
218,171
624,288
0
3,252
0
5,357
0
1,749,972
0
0
COD Load
Reduction
for 60%
Collection
and Control
Scenario
(pounds)
1,380
0
0
122,408
0
2,838,510
98,245
9,285
363,817
0
0
785,513
17,445
255,561
0
76,983
123,166
408,799
214
0
595,416
327,257
936,432
944,981
4,878
0
5,357
0
2,624,959
0
117,624
1 Ketchikan was sent an airport questionnaire but did not respond. EPA developed an estimate of annualized costs
for this airport using existing Alaskan airport information.
2 Falls International was sent an airport questionnaire but did not respond.
NA - Not applicable; the airport reported no deicing operations in the airport questionnaire.
NE - No percent capture was estimated; the airport did not respond to the airport questionnaire.
July 2009 10-21
-------
Technical Development Document for Proposed Effluent
Limitation Guidelines and Standards for the Airport Deicing Category
10. Pollutant Loadings and
Pollutant Load Reduction Estimates
Table 10-6 (Continued)
Airport
Austin-Bergstrom International
Me Carran International
Metropolitan Oakland International
San Diego International
Baltimore-Washington International
George Bush Intercontinental
Airport/Houston
Luis Munoz Marin International
Kahului
Louis Armstrong New Orleans
International
Glacier Park International
Orlando International
Ralph Wien Memorial
Roanoke Regional/Woodrum Field
Norman Y. Mineta San Jose International
Long Island Mac Arthur
Sacramento Mather
Redding Municipal
Lanai
Aspen-Pitkin Co/Sardy Field
Barnstable Muni-Boardman/Polando
Field
Wilmington International
General Edward Lawrence Logan
International
Jackson Hole
Miami International
Santa Maria Pub/Capt G Allan Hancock
Fid
Will Rogers World
Gerald R. Ford International
Greater Rochester International
Williamson County Regional
Current
ADF
Collection
40
40
100
100
60
40
NA
NA
100
0
NA
0
0
10
60
20
0
NA
40
0
0
0
100
100
0
40
50
100
Baseline
COD Load
(pounds)
146,656
64,919
0
0
1,374,218
60,235
0
0
0
391,940
0
25,326
334,727
0
176,997
11,932
7,035
0
91,601
469,111
22,118
9,740,474
0
0
0
503,236
600,371
1,228,022
0
COD Load
Reduction
for 20%
Collection
and Control
Scenario
(pounds)
0
0
0
0
0
0
0
0
0
76,428
0
4,939
65,272
0
0
0
1,372
0
0
91,477
4,313
1,899,392
0
0
0
98,131
0
0
0
COD Load
Reduction
for 40%
Collection
and Control
Scenario
(pounds)
0
0
0
0
0
0
0
0
0
152,857
0
9,877
130,544
0
0
5,966
2,744
0
0
182,953
8,626
3,798,785
0
0
0
196,262
0
0
0
COD Load
Reduction
for 60%
Collection
and Control
Scenario
(pounds)
47,663
21,099
0
0
0
20,078
0
0
0
229,285
0
14,816
195,815
0
0
5,966
4,116
0
30,534
274,430
12,939
5,698,177
0
0
0
294,393
200,124
239,464
0
1 Ketchikan was sent an airport questionnaire but did not respond. EPA developed an estimate of annualized costs
for this airport using existing Alaskan airport information.
2 Falls International was sent an airport questionnaire but did not respond.
NA - Not applicable; the airport reported no deicing operations in the airport questionnaire.
NE - No percent capture was estimated; the airport did not respond to the airport questionnaire.
July 2009 10-22
-------
Technical Development Document for Proposed Effluent
Limitation Guidelines and Standards for the Airport Deicing Category
10. Pollutant Loadings and
Pollutant Load Reduction Estimates
Table 10-6 (Continued)
Airport
William P Hobby
Birmingham International
Evansville Regional
Falls International 2
Albany International
Salt Lake City International
Helena Regional
Eppley Airfield
Cleveland-Hopkins International
City of Colorado Springs Municipal
Tweed-New Haven
Gillette-Campbell County
Honolulu International
South Bend Regional
Pensacola Regional
Kona International at Keahole
Nashville International
Manchester
Syracuse Hancock International
Bob Hope
Trenton Mercer
Tampa International
Bismarck Municipal
Waterloo Municipal
Palm Beach International
El Paso International
Outagamie County Regional
John F Kennedy International
Boise Air Terminal/Gowen Fid
Rochester International
Lewiston-Nez Perce County
Los Angeles International
Current
ADF
Collection
100
0
0
NE
92
60
100
0
40
40
0
100
NA
100
100
NA
60
0
60
100
0
100
0
0
100
100
0
0
60
40
100
NA
Baseline
COD Load
(pounds)
0
50,847
204,825
0
109,890
1,794,858
0
1,128,249
3,709,111
462,436
50,074
0
0
0
0
0
372,242
1,828,125
844,427
0
54,836
0
213,443
79,504
0
0
588,037
5,489,149
290,420
210,773
0
0
COD Load
Reduction
for 20%
Collection
and Control
Scenario
(pounds)
0
9,915
39,941
0
0
0
0
220,009
0
0
9,764
0
0
0
0
0
0
356,484
0
0
10,693
0
41,621
15,503
0
0
114,667
1,070,384
0
0
0
0
COD Load
Reduction
for 40%
Collection
and Control
Scenario
(pounds)
0
19,830
79,882
0
0
0
0
440,017
0
0
19,529
0
0
0
0
0
0
712,969
0
0
21,386
0
83,243
31,006
0
0
229,335
2,140,768
0
0
0
0
COD Load
Reduction
for 60%
Collection
and Control
Scenario
(pounds)
0
29,745
119,823
0
0
0
0
660,026
1,236,370
150,292
29,293
0
0
0
0
0
0
1,069,453
0
0
32,079
0
124,864
46,510
0
0
344,002
3,211,152
0
68,501
0
0
1 Ketchikan was sent an airport questionnaire but did not respond. EPA developed an estimate of annualized costs
for this airport using existing Alaskan airport information.
2 Falls International was sent an airport questionnaire but did not respond.
NA - Not applicable; the airport reported no deicing operations in the airport questionnaire.
NE - No percent capture was estimated; the airport did not respond to the airport questionnaire.
July 2009 10-23
-------
Technical Development Document for Proposed Effluent
Limitation Guidelines and Standards for the Airport Deicing Category
10. Pollutant Loadings and
Pollutant Load Reduction Estimates
Table 10-6 (Continued)
Airport
Boeing Field/King County International
Chicago Midway International
Santa Fe Municipal
Lovell Field
Aberdeen Regional
Sacramento International
Toledo Express
Portland International
John Wayne Airport-Orange County
Juneau International
Nome
Spokane International
Fort Lauderdale/Hollywood International
Pittsburgh International
Louisville International-Standiford Field
Airborne Airpark
Aniak
Port Columbus International
Deadhorse
Cincinnati/Northern Kentucky
International
Stewart International
Jacksonville International
Reno/Tahoe International
Cherry Capital
Bethel
Rickenbacker International
Rapid City Regional
Theodore Francis Green State
Southwest Florida International
James M Cox Dayton International
Des Moines International
Sarasota/Bradenton International
Current
ADF
Collection
40
100
100
0
100
20
20
20
100
0
0
100
NA
60
60
40
0
0
100
87
40
100
20
100
0
0
0
60
100
60
40
NA
Baseline
COD Load
(pounds)
31,446
0
0
42,182
0
0
383,445
911,546
0
459,700
30,113
0
0
3,931,605
522,149
2,331,713
5,540
3,120,055
0
822,345
196,863
0
606,939
0
50,915
108,882
258,448
610,460
0
380,946
490,533
0
COD Load
Reduction
for 20%
Collection
and Control
Scenario
(pounds)
0
0
0
25,309
0
0
0
0
0
89,641
5,872
0
0
0
0
0
1,080
608,411
0
0
0
0
0
0
9,928
21,232
50,397
0
0
0
0
0
COD Load
Reduction
for 40%
Collection
and Control
Scenario
(pounds)
0
0
0
25,309
0
0
93,465
222,189
0
179,283
11,744
0
0
0
0
0
2,161
1,216,822
0
0
0
0
147,941
0
19,857
42,464
100,795
0
0
0
0
0
COD Load
Reduction
for 60%
Collection
and Control
Scenario
(pounds)
10,220
0
0
25,309
0
0
186,929
444,379
0
268,924
17,616
0
0
0
0
757,807
3,241
1,825,232
0
0
63,980
0
295,883
0
29,785
63,696
151,192
0
0
0
159,423
0
1 Ketchikan was sent an airport questionnaire but did not respond. EPA developed an estimate of annualized costs
for this airport using existing Alaskan airport information.
2 Falls International was sent an airport questionnaire but did not respond.
NA - Not applicable; the airport reported no deicing operations in the airport questionnaire.
NE - No percent capture was estimated; the airport did not respond to the airport questionnaire.
July 2009 10-24
-------
Technical Development Document for Proposed Effluent
Limitation Guidelines and Standards for the Airport Deicing Category
10. Pollutant Loadings and
Pollutant Load Reduction Estimates
Table 10-6 (Continued)
Airport
Minneapolis-St Paul International/Wold-
Chamberlain
Willow Run
Charlotte/Douglas International
Bradley International
San Antonio International
Wilkes-Barre/Scranton International
Chippewa Valley Regional
Phoenix Sky Harbor International
St George Municipal
Lafayette Regional
General Mitchell International
Dallas Love Field
Detroit Metropolitan Wayne County
Philadelphia International
Memphis International
Ronald Reagan Washington National
Washington Dulles International
San Francisco International
Central Wisconsin
Newark Liberty Intl
Northwest Arkansas Regional
Raleigh-Durham Intl
Kansas City Intl
Fort Worth Alliance
Greater Rockford
Kalamazoo/Battle Creek Intl
Duluth Intl
Akron - Canton Regional
Current
ADF
Collection
60
40
0
60
0
0
100
20
100
0
41
40
100
85
0
40
40
100
0
0
0
0
40
60
60
60
0
60
Baseline
COD Load
(pounds)
6,362,897
62,356
1,392,605
1,778,704
129,604
432,418
0
0
0
15,133
957,716
227,012
0
1,530,580
2,073,854
1,309,731
6,052,151
0
443,467
11,464,956
312,852
1,041,489
1,279,726
6,898
557,825
88,053
765,304
255,826
COD Load
Reduction
for 20%
Collection
and Control
Scenario
(pounds)
0
0
271,558
0
25,273
84,322
0
0
0
9,080
0
0
0
0
404,401
0
0
0
86,476
2,235,666
61,006
203,090
0
0
0
0
149,234
0
COD Load
Reduction
for 40%
Collection
and Control
Scenario
(pounds)
0
0
543,116
0
50,545
168,643
0
0
0
9,080
0
0
0
0
808,803
0
0
0
172,952
4,471,333
122,012
406,181
0
0
0
0
298,468
0
COD Load
Reduction
for 60%
Collection
and Control
Scenario
(pounds)
0
20,785
814,674
0
75,818
252,965
0
0
0
9,080
308,417
73,779
0
0
1,213,204
436,577
1,966,949
0
259,428
6,706,999
183,018
609,271
415,911
0
0
0
447,703
0
1 Ketchikan was sent an airport questionnaire but did not respond. EPA developed an estimate of annualized costs
for this airport using existing Alaskan airport information.
2 Falls International was sent an airport questionnaire but did not respond.
NA - Not applicable; the airport reported no deicing operations in the airport questionnaire.
NE - No percent capture was estimated; the airport did not respond to the airport questionnaire.
July 2009 10-25
-------
Technical Development Document for Proposed Effluent 10. Pollutant Loadings and
Limitation Guidelines and Standards for the Airport Deicing Category Pollutant Load Reduction Estimates
For each scenario, EPA estimated no load reductions for the airport if the airport collects
and controls less than the required percentage of spent ADF (e.g., for the 20 percent efficiency
scenario, if an airport currently collects and controls 20 percent or more of spent ADF, no load
reductions were estimated for the airport).
For airports that used small quantities of ADF, EPA assumed that the airport would
collect and haul away the ADF contaminated stormwater instead of collecting it for onsite
treatment. This assumption was made for all scenarios, and EPA assumed that the load
reductions for these airports would be the same across all scenarios. For more information,
please refer to the Airport Deicing Loadings Calculations memorandum (ERG, 2008d).
10.7 Approach for Calculating Urea-Related Reductions
For this proposal, EPA evaluated the impact of restricting urea as an airfield deicing
chemical and replacing its use with potassium acetate. Both urea and potassium acetate produce
COD as they degrade; however, the COD effect from potassium acetate use is significantly less.
Discontinuing the use of urea as an airfield deicing chemical will also help to eliminate receiving
water toxicity impacts including ammonia formation as urea breaks down in water and a
potential contribution to nutrient enrichment.
As described above, EPA calculated the COD load associated with urea use at the
surveyed airports. EPA then evaluated the amount of potassium acetate that would be required to
replace the current average urea use using a comparison of application rates under varying winter
conditions (see EPA's Urea/Potassium Acetate memorandum (ERG, 2008b) for the details of
this analysis). Based on the COD load associated with the equivalent potassium acetate use, EPA
determined the potential reductions in COD load. Table 10-7 presents the baseline COD load
associated with urea, the estimated equivalent COD load if industry converted from urea to
potassium acetate, and the load reduction.
July 2009 10-26
-------
Technical Development Document for Proposed Effluent
Limitation Guidelines and Standards for the Airport Deicing Category
10. Pollutant Loadings and
Pollutant Load Reduction Estimates
Table 10-7. Baseline COD Load and Potential Load Reduction Associated with the
Discontinued Use of Urea as an Airfield Deicing Chemical
Airport ID
1007
1010
1012
1013
1016
1017
1018
1022
1041
1043
1053
1066
1074
1079
1090
1103
1105
1110
1112
1114
1116
1118
1128
1129
1141
1144
1146
1147
1150
1153
Airport
Yeager
Fairbanks International
Ted Stevens Anchorage
International
Wiley Post- Will Rogers Mem
Tri - State/Milton J Ferguson Field
Austin Straubel International
Piedmont Triad International
Fort Wayne International
Glacier Park International
Ralph Wien Memorial
General Edward Lawrence Logan
International
Salt Lake City International
South Bend Regional
Manchester
Boise Air Terminal/Gowen Field
Juneau International
Spokane International
Aniak
Deadhorse
Stewart International
Reno/Tahoe International
Bethel
Charlotte/Douglas International
Bradley International
Ronald Reagan Washington
National
Central Wisconsin
Northwest Arkansas Regional
Raleigh - Durham International
Greater Rockford
Akron - Canton Regional
Urea Load
(IbsofCOD)
56,797
816,961
3,560,670
42,624
133,555
88,530
210,279
571,082
710
21,312
12,148
3,127,198
69,136
47,952
864,579
1,082,651
1,361,128
5,115
42,624
323,516
15,426
140,659
496,464
35,692
133,555
179,859
57,542
190,387
1,396,079
45,466
Equivalent Potassium
Acetate Load
(IbsofCOD)
15,663
225,297
981,943
11,755
36,831
24,414
57,990
157,490
196
5,877
3,350
862,402
19,066
13,224
238,429
298,568
375,365
1,411
11,755
89,218
4,254
38,790
136,912
9,843
36,831
49,601
15,869
52,504
385,004
12,538
Potential Load
Reduction
(IbsofCOD)
41,133
591,664
2,578,727
30,869
96,724
64,116
152,289
413,592
514
15,435
8,798
2,264,796
50,070
34,728
626,150
784,083
985,763
3,704
30,869
234,299
11,172
101,869
359,551
25,849
96,724
130,259
41,674
137,883
1,011,076
32,927
July 2009
10-27
-------
Technical Development Document for Proposed Effluent 10. Pollutant Loadings and
Limitation Guidelines and Standards for the Airport Deicing Category Pollutant Load Reduction Estimates
10.8 References
Switzenbaum, et al. 1999. Workshop: Best Practices for Airport Deicing Stormwater. DCN
AD00893.
ERG. 2008a. Memorandum from Steve Strackbein (ERG) to Brian D'Amico and Eric Strassler
(EPA): The Development of Snow and Freezing Precipitation (SOFP) Days and the Aircraft
Deicing and Anti-icing Fluid (ADAF) Factor. (January 16). DCN AD00856.
ERG. 2008b. Memorandum from Steve Strackbein and Mary Willett (ERG) to Brian D'Amico
and Eric Strassler (EPA): Cost Comparison of Potassium Acetate and Urea Airfield Deicers.
(March 17). DCN AD00843.
ERG. 2008c. Memorandum from Juliana Stroup and Mary Willett (ERG) to Brian D'Amico
(EPA): ADF Capture and Control Efficiency Review. (July 7). DCN AD00854.
ERG. 2008d. Memorandum from Cortney Itle (ERG) to Brian D'Amico (EPA) : Airport Deicing
Loadings Calculations. (April 17). DCN AD001140.
USEPA, 2008a. Airport Deicing Loadings Database. U.S. Environmental Protection
Agency/Office of Water. Washington, D.C. DCN AD00857.
EPA. 2008b. Airport Deicing Operations ADF Usage Database. U.S. Environmental Protection
Agency/Office of Water. Washington, D.C. DCN AD001141.
July 2009 10-28
-------
Technical Development Document for Proposed Effluent 11. Technology Costs
Limitation Guidelines and Standards for the Airport Deicing Category
11. TECHNOLOGY COSTS
This section presents EPA's estimates of costs for the Airport Deicing Category to
implement the collection and treatment technologies described in Section 9. EPA estimated the
compliance costs for each collection and treatment scenario to determine potential economic
impacts on the industry. EPA also weighed these costs against the pollutant load reduction
benefits. This section includes cost estimates for the collection and treatment scenarios that EPA
evaluated for the proposed regulation including those that EPA ultimately rejected. Also
included in this section are estimated costs for airports to implement new airfield deicing
alternatives to replace urea. The Agency is reporting estimates of potential economic impacts
associated with the total estimated annualized costs of the proposed regulation separately, in the
Economic Analysis document.
Section 11.1 summarizes the annualized costs for each airport to collect and control
(through treatment) differing amounts of spent ethylene- and propylene-based aircraft deicing
fluid. The remainder of this section discusses the following information:
• Section 11.2: Selection and development of cost model inputs;
• Section 11.3: The methodology for estimating collection and treatment
technology costs, including an overview of the cost model and example
calculations showing how the model estimates costs and cost annualization;
• Section 11.4 Airfield deicing costs; and
• Section 11.5: References used in this section.
Tables are presented within the text and figures are presented at the end of the subsections.
11.1 Summary of Costs
This subsection summarizes EPA's annualized costs for each airport to collect and treat
differing amounts of spent aircraft deicing fluid (ADF). EPA estimated annualized costs based
on the current level of spent ADF collected and controlled at each airport and the additional
capital costs and annual operating costs needed to achieve the target collection and control
percentage. For those airports that EPA estimated collect and control spent ADF at levels greater
than 60 percent, no incremental annualized costs are required. (Section 10.5.2 discusses the
current collection and control percentages at each airport.) Table 11-1 presents annualized costs
for each airport included in EPA's airport questionnaire by the following scenarios:
• 20% Efficiency Scenario: includes glycol recovery vehicles (GRVs) and
anaerobic fluidized bed (AFB) treatment;
• 40% Efficiency Scenario: includes plug and pump with GRVs and AFB
treatment; and
• 60% Efficiency Scenario: includes deicing pads with GRVs and AFB treatment.
Section 11.3 provides details on the methodology used to estimate annualized costs from
capital and annual operating costs. Specific cost model outputs containing both capital and
July 2009 ll-l
-------
Technical Development Document for Proposed Effluent 11. Technology Costs
Limitation Guidelines and Standards for the Airport Deicing Category
annual costs for individual airports are included in the Airport Deicing Category administrative
record.
EPA selected AFB treatment as the best available technology for ADF-contaminated
stormwater, based on its ability to produce an effluent stream clean enough for direct discharge.
Other technologies evaluated for the Airport Deicing Category can recover and recycle glycol
from ADF-contaminated stormwater. However, the residual waste streams from these
technologies may require additional treatment or must be discharged to a Public Owned
Treatment Works (POTW). In addition, the anaerobic fluidized bed reactor has the flexibility to
accept ADF-contaminated stormwater with a wide range of glycol concentrations. The recycle
and recovery technologies evaluated by EPA typically require higher glycol concentrations in
their feed streams to become economical to operate (Switzenbaum, 1999).
For airports that occasionally deice aircraft primarily to remove frost, installing
permanent collection and treatment equipment for spent ADF would not be practical. Instead,
EPA believes these airports would contract out the deicing stormwater collection and removal
tasks, and their costs would not vary between the 20 percent and 60 percent collection/control
scenarios. This costing approach affects the following airports listed in Table 11-1:
• Ontario International, California;
• Craven County Regional, North Carolina;
• Sacramento Mather, California;
• Lovell Field, Tennessee; and
• Lafayette Regional, Louisiana.
Specific details regarding the costs for occasional removal of spent ADF by a local contractor is
included in a memorandum entitled Estimated Annual Costs for Airports with Limited ADF Use
(ERG, 2008a).
Annualized costs shown in Table 11-1 do not consider airfield deicing. EPA recognizes
that airports will incur additional costs to change from urea to potassium acetate airfield deicing;
however, the Agency decided to not include these costs in the annualized costs for managing
spent ADF. The Agency has prepared a detailed memorandum entitled Cost Comparison of
Potassium Acetate and Urea Airfield Deicers (ERG, 2008b), which outlines the possible
incremental costs to the industry to replace urea. Section 11.4 summarizes the estimated cost for
airports to change from urea airfield deicing to potassium acetate deicing.
July 2009 11-2
-------
Technical Development Document for Proposed Effluent
Limitation Guidelines and Standards for the Airport Deicing Category
11. Technology Costs
Table 11-1. Annualized Costs by Surveyed Airport for Each Collection and Control
Scenario Evaluated by EPA l
Airport
Montgomery Regional
(Dannelly Field)
Bert Mooney
Ketchikan International
Norfolk International
Roberts Field
Chicago O'Hare International
Yeager
Tucson International
Cold Bay
Fairbanks International
Lambert-St Louis
International
Ted Stevens Anchorage
International
Wiley Post- Will Rogers
Memorial
Albuquerque International
Gulfport-Biloxi International
Tri-State/Milton J. FEPAuson
Field
Austin Straubel International
Piedmont Triad International
Ontario International 3
Hartsfield - Jackson Atlanta
International
Buffalo Niagara International
Fort Wayne International
Seattle-Tacoma International
Indianapolis International
Tupelo Regional
Dallas/Fort Worth
International
Current ADF
Collection
0
100
2
20
100
40
40
20
0
60
60
40
0
20
100
0
40
0
0
60
40
0
60
40
0
60
Annualized Cost
for 20% Collection
and Control
Scenario
(2006 $)
$92,700
$0
2
$0
$0
$0
$0
$0
$354,700
$0
$0
$0
$69,300
$0
$0
$143,900
$0
$1,099,600
$1,100
$0
$0
$571,000
$0
$0
$83,200
$0
Annualized Cost
for 40% Collection
and Control
Scenario
(2006 $)
$352,800
$0
2
$1,799,200
$0
$0
$0
$985,300
$721,200
$0
$0
$0
$65,600
$5,763,600
$0
$431,800
$0
$7,639,600
$1,100
$0
$0
$2,423,100
$0
$0
$354,000
$0
Annualized Cost
for 60% Collection
and Control
Scenario
(2006 $)
$97,200
$0
2
$770,300
$0
$16,875,404
$1,002,671
$75,900
$552,700
$0
$0
$4,269,900
$174,300
$1,374,600
$0
$208,000
$703,300
$1,126,200
$1,100
$0
$2,396,500
$293,600
$0
$7,679,900
$71,100
$0
1 Treatment includes installation and operation of an anaerobic fluidized bed biological treatment system unless
otherwise specified.
2 Ketchikan was sent an airport questionnaire but did not respond. An estimate of annualized costs was developed
for this airport using existing Alaska airport information.
3 Airport uses small amounts of ADF and assumes will contract all operations for ADF removal and disposal.
4 Cost assumes additional contract hauling of collected ADF contaminated stormwater.
5 International Falls was sent an airport questionnaire but did not respond.
July 2009 11-3
-------
Technical Development Document for Proposed Effluent
Limitation Guidelines and Standards for the Airport Deicing Category
11. Technology Costs
Table 11-1 (Continued)
Airport
Craven County Regional 3
Denver International
La Guardia
Williamsport Regional
Richmond International
Austin-Bergstrom
International
McCarran International
Baltimore-Washington
International
George Bush Intercontinental
Airport/Houston
Glacier Park International
Ralph Wien Memorial
Roanoke Regional/Woodrum
Field
Long Island MacArthur
Sacramento Mather 3
Redding Municipal
Aspen-Pitkin Co/Sardy
Field 4
Barnstable Muni-
Boardman/Polando Field
Wilmington International
General Edward Lawrence
Logan International
Jackson Hole
Will Rogers World
Gerald R. Ford International 4
Greater Rochester
International
Williamson County Regional
William P Hobby
Birmingham International
Current ADF
Collection
0
93
0
60
40
40
40
60
40
0
0
0
60
20
0
40
0
0
0
100
0
40
50
100
100
0
Annualized Cost
for 20% Collection
and Control
Scenario
(2006 $)
$2,000
$0
$2,881,300
$0
$0
$0
$0
$0
$0
$483,500
$266,200
$621,400
$0
$0
$111,200
$0
$409,600
$311,200
$4,289,000
$0
$521,200
$0
$0
$0
$0
$221,100
Annualized Cost
for 40% Collection
and Control
Scenario
(2006 $)
$2,000
$0
$6,440,400
$0
$0
$0
$0
$0
$0
$972,700
$511,200
$3,680,400
$0
$2,200
$646,400
$0
$1,566,300
$2,726,800
$7,116,000
$0
$2,201,600
$0
$0
$0
$0
$466,700
Annualized Cost
for 60% Collection
and Control
Scenario
(2006 $)
$2,000
$0
$6,238,900
$0
$1,063,300
$927,400
$294,200
$0
$10,100
$843,500
$546,800
$639,100
$0
$2,200
$78,800
$156,800
$652,200
$139,000
$9,544,000
$0
$856,900
$546,600
$2,134,500
$0
$0
$355,000
1 Treatment includes installation and operation of an anaerobic fluidized bed biological treatment system unless
otherwise specified.
2 Ketchikan was sent an airport questionnaire but did not respond. An estimate of annualized costs was developed
for this airport using existing Alaska airport information.
3 Airport uses small amounts of ADF and assumes will contract all operations for ADF removal and disposal.
4 Cost assumes additional contract hauling of collected ADF contaminated stormwater.
5 International Falls was sent an airport questionnaire but did not respond.
July 2009 11-4
-------
Technical Development Document for Proposed Effluent
Limitation Guidelines and Standards for the Airport Deicing Category
11. Technology Costs
Table 11-1 (Continued)
Airport
Evansville Regional
Falls International
Albany International
Salt Lake City International
Helena Regional
Eppley Airfield
Cleveland-Hopkins
International 4
City of Colorado Springs
Municipal
Tweed-New Haven
Gillette-Campbell County
South Bend Regional
Pensacola Regional
Nashville International
Manchester
Syracuse Hancock
International
Trenton Mercer
Bismarck Municipal
Waterloo Municipal
El Paso International
Outagamie County Regional
John F Kennedy International
Boise Air Terminal/Gowen
Field
Rochester International
Lewiston-Nez Perce County
Boeing Field/King County
International
Chicago Midway
International
Santa Fe Municipal
Lovell Field 3
Current ADF
Collection
0
5
92
60
100
0
40
40
0
100
100
100
60
0
60
0
0
0
100
0
0
60
40
100
40
100
100
0
Annualized Cost
for 20% Collection
and Control
Scenario
(2006 $)
$297,500
5
$0
$0
$0
$531,600
$0
$0
$115,200
$0
$0
$0
$0
$839,900
$0
$476,300
$199,600
$130,200
$0
$461,500
$2,989,500
$0
$0.
$0
$0
$0
$0
$6,200
Annualized Cost
for 40% Collection
and Control
Scenario
(2006 $)
$1,166,900
5
$0
$0
$0
$1,466,400
$0
$0
$395,400
$0
$0
$0
$0
$1,474,900
$0
$3,978,300
$760,800
$414,000
$0
$1,902,000
$9,774,000
$0
$0
$0
$0
$0
$0
$6,200
Annualized Cost
for 60% Collection
and Control
Scenario
(2006 $)
$401,200
5
$0
$0
$0
$1,176,500
$1,838,200
$906,700
$127,400
$0
$0
$0
$0
$1,757,600
$0
$223,100
$271,000
$155,000
$0
$643,000
$6,029,600
$0
$397,100
$0
$250,900
$0
$0
$6,200
1 Treatment includes installation and operation of an anaerobic fluidized bed biological treatment system unless
otherwise specified.
2 Ketchikan was sent an airport questionnaire but did not respond. An estimate of annualized costs was developed
for this airport using existing Alaska airport information.
3 Airport uses small amounts of ADF and assumes will contract all operations for ADF removal and disposal.
4 Cost assumes additional contract hauling of collected ADF contaminated stormwater.
5 International Falls was sent an airport questionnaire but did not respond.
July 2009 11-5
-------
Technical Development Document for Proposed Effluent
Limitation Guidelines and Standards for the Airport Deicing Category
11. Technology Costs
Table 11-1 (Continued)
Airport
Aberdeen Regional
Toledo Express
Portland International
Juneau International
Nome
Spokane International
Pittsburgh International
Louisville International-
Standiford Field
Airborne Airpark
Aniak
Port Columbus International
Deadhorse
Cincinnati/Northern
Kentucky International
Stewart International
Jacksonville International
Reno/Tahoe International
Cherry Capital
Bethel
Rickenbacker International
Rapid City Regional
Theodore Francis Green State
James M Cox Dayton
International
Des Moines International
Minneapolis-St Paul
International/Wold-
Chamberlain
Willow Run 4
Charlotte/Douglas
International
Bradley International
San Antonio International
Current ADF
Collection
100
20
20
0
0
100
60
60
40
0
0
100
87
40
100
20
100
0
0
0
60
60
40
60
40
0
60
0
Annualized Cost
for 20% Collection
and Control
Scenario
(2006 $)
$0
$0
$0
$496,700
$462,100
$0
$0
$0
$0
$334,200
$1,407,400
$0
$0
$0
$0
$0
$0
$589,200
$197,100
$462,400
$0
$0
$0
$0
$0
$1,664,000
$0
$692,700
Annualized Cost
for 40% Collection
and Control
Scenario
(2006 $)
$0
$1,314,800
$2,467,400
$1,896,500
$1,786,500
$0
$0
$0
$0
$1,123,500
$3,772,600
$0
$0
$0
$0
$1,054,800
$0
$1,875,100
$1,018,000
$2,925,300
$0
$0
$0
$0
$0
$2,241,900
$0
$4,692,700
Annualized Cost
for 60% Collection
and Control
Scenario
(2006 $)
$0
$684,200
$2,229,800
$813,500
$552,400
$0
$0
$0
$4,478,100
$412,700
$2,873,300
$0
$0
$0
$0
$1,047,400
$0
$1,079,900
$196,500
$412,500
$0
$0
$928,300
$0
$123,900
$4,312,700
$0
$638,900
1 Treatment includes installation and operation of an anaerobic fluidized bed biological treatment system unless
otherwise specified.
2 Ketchikan was sent an airport questionnaire but did not respond. An estimate of annualized costs was developed
for this airport using existing Alaska airport information.
3 Airport uses small amounts of ADF and assumes will contract all operations for ADF removal and disposal.
4 Cost assumes additional contract hauling of collected ADF contaminated stormwater.
5 International Falls was sent an airport questionnaire but did not respond.
July 2009 11-6
-------
Technical Development Document for Proposed Effluent
Limitation Guidelines and Standards for the Airport Deicing Category
11. Technology Costs
Table 11-1 (Continued)
Airport
Wilkes-Barre/Scranton
International
Chippewa Valley Regional
St George Municipal
Lafayette Regional 3
General Mitchell
International 4
Dallas Love Field
Detroit Metropolitan Wayne
County
Philadelphia International
Memphis International
Ronald Reagan Washington
National 4
Washington Dulles
International
San Francisco International
Central Wisconsin
Newark Liberty International
Northwest Arkansas Regional
Raleigh-Durham
International
Kansas City International
Fort Worth Alliance
Greater Rockford
Kalamazoo/Battle Creek
International
Duluth International
Akron - Canton Regional
Current ADF
Collection
0
100
100
0
41
40
100
85
0
40
40
100
0
0
0
0
40
60
60
60
0
60
Annualized Cost
for 20% Collection
and Control
Scenario
(2006 $)
$361,300
$0
$0
$2,400
$0
$0
$0
$0
$1,817,200
$0
$0
$0
$331,400
$5,295,600
$314,000
$1,096,400
$0
$0
$0
$0
$377,700
$0
Annualized Cost
for 40% Collection
and Control
Scenario
(2006 $)
$1,012,200
$0
$0
$2,400
$0
$0
$0
$0
$3,771,500
$0
$0
$0
$961,000
$10,611,400
$1,167,100
$3,963,300
$0
$0
$0
$0
$1,043,200
$0
Annualized Cost
for 60% Collection
and Control
Scenario
(2006 $)
$574,400
$0
$0
$2,400
$1,706,300
$912,000
$0
$0
$3,769,600
$1,987,300
$18,648,600
$0
$517,000
$11,277,300
$514,400
$1,880,300
$2,661,400
$0
$0
$0
$595,700
$0
1 Treatment includes installation and operation of an anaerobic fluidized bed biological treatment system unless
otherwise specified.
2 Ketchikan was sent an airport questionnaire but did not respond. An estimate of annualized costs was developed
for this airport using existing Alaska airport information.
3 Airport uses small amounts of ADF and assumes will contract all operations for ADF removal and disposal.
4 Cost assumes additional contract hauling of collected ADF contaminated stormwater.
5 International Falls was sent an airport questionnaire but did not respond.
July 2009 11-7
-------
Technical Development Document for Proposed Effluent 11. Technology Costs
Limitation Guidelines and Standards for the Airport Deicing Category
Table 11-1 shows that for some airports annualized GRV related costs under the 20%
collection/control scenario can meet or exceed the costs estimated under the deicing pad 60%
collection/control scenario. In addition, there are also cases where the 40% collection/control
scenario amortized costs meet or exceed the costs estimated under the deicing pad 60%
collection/control scenario. These occurrences are a function of the airport factors used to scale
costs by scenario and whether the predominant costs for an airport are annual costs or capital
costs. In the case of GRVs and plug and pump systems, the model airport costs are scaled based
on an airport's number of deicing outfalls. For deicing pad systems, the model airport costs are
scaled based on an airport's number of annual departures. Cases where the GRV costs meet or
exceed the deicing pad costs are predominantly smaller airports with a low number of departures
and a high number of deicing stormwater outfalls. In these cases, the high number of deicing
outfalls results in higher GRV costs (compared to the model airports) and lower deicing pad
costs (compared to the model airports) due to the low number of departures per year. Cases
where the plug and pump costs meet or exceed the deicing pad costs are predominantly airports
with a high number of deicing outfalls compared to their departures. In addition, plug and pump
systems tend to be labor intensive resulting in high annual O&M costs compared to deicing pad
systems, which have higher capital costs but lower annual O&M costs.
11.2 Development of Cost Model Inputs
This subsection describes the key inputs to EPA's cost model for the Airport Deicing
Category: model sites, ADF usage, deicing days per season, annual airport departures, current
collection and control technologies in place, estimated percentage of ADF in collected
stormwater and the number of deicing outfalls at each airport, precipitation data, and physical
features of each airport. Also discussed are the data sources used to determine these parameters
and how the cost model uses the input data.
11.2.1 Model Site Development
The Agency used a model-site approach to estimate costs for the Airport Deicing
Category. A model airport is an operating airport whose regulatory status and unit operation and
treatment information were used as parameters for the cost model. EPA selected an airport-by-
airport approach to estimate compliance costs based on a comparison of information from the
model airports, as opposed to a more generalized approach, to better characterize the current
collection and control systems in place for spent ADF and to account for current site conditions
and airport operations.
To select model airports on which to base the costs, EPA reviewed information collected
during site visits and sampling episodes and compared the information to the control objectives
to determine which airports could be considered models for the remainder of the industry. Using
industry-supplied cost data collected from the model airports, the Agency was able to predict
costs for other airports to achieve similar results.
An analysis of the design, operation, and applicability of collection and control
technologies evaluated for proposal was discussed in Section 9. Based on the analysis of
collection and control technologies, EPA decided costs should be developed for three collection
technologies and AFB biological treatment since it provides the best achievable treatment for
ADF-contaminated stormwater. The remainder of this subsection will focus on model airports
July 2009 11-8
-------
Technical Development Document for Proposed Effluent 11. Technology Costs
Limitation Guidelines and Standards for the Airport Deicing Category
having the collection technologies listed in Section 11-1 above and AFB treatment for ADF-
contaminated stormwater.
The Agency made costing assumptions based on information from a limited number of
model airports having the selected ADF collection and treatment technology in place. Thus, for
any given airport, the estimated costs may deviate from those that the airport would actually
incur. However, EPA considers the compliance costs to be accurate when aggregated on an
industry-wide basis.
11.2.2 Airport Operations Data
The primary source of airport operations data used to calculate collection and treatment
costs were responses to the Agency's 2006 airport questionnaire (USEPA, 2006a). EPA entered
data from all questionnaires into an electronic database that the cost model then accessed to
determine if any spent ADF collection and treatment technologies were currently being used; the
operations data needed to estimate costs for additional collection and treatment was also entered
into the database. Table 11-2 lists the airport operating data accessed by the cost model to
estimate both capital and annual operating costs for spent ADF collection and treatment.
Two additional pieces of airport operating data not requested in the airport questionnaire
that were required to estimate costs included annual ADF use and the number of aircraft
departures. EPA collected data on annual ADF usage in its 2006 airline detailed questionnaire
(USEPA, 2006b) and information on aircraft departures from the Bureau of Transportation
Statistics, T-100 Segment Database (USDOT, 2006). ADF use, combined with the expected
percentage efficiency for a specific collection technology, determined both pollutant reductions
to surface water and pollutant loadings to on-site treatment. EPA used data on the number of
aircraft departures as a metric to relate model site collection system costs to other airports that
were not achieving the collection percentage objective.
11.2.3 Precipitation Data and Site Characteristics
Estimating costs to collect and treat ADF-contaminated stormwater requires knowing the
expected volume of the stormwater. To predict the annual volume of ADF-contaminated
stormwater generated at an airport, EPA used precipitation data along with airport site
characteristics and assumed ADF-contaminated stormwater would be collected from areas where
ADF is applied and possibly from areas where ADF may drip from the aircraft during taxi and
takeoff. EPA obtained precipitation data from 1976 to 2006 (30 years) from the National Climate
Data Center (NOAA) for each airport questionnaire respondent and then averaged the data to
estimate a monthly average. These data were used by the cost model on an airport-specific basis.
EPA uses these data, combined with the number of deicing months taken from the questionnaire
responses, to estimate precipitation that may be contaminated by ADF.
July 2009 11-9
-------
Technical Development Document for Proposed Effluent
Limitation Guidelines and Standards for the Airport Deicing Category
11. Technology Costs
Table 11-2. Airport Questionnaire Data to Estimate Spent ADF Collection and Treatment
Costs
Airport Questionnaire
Number
Q3
Q7
Q10
Qllb
Qllf
Q20
Q21
Q25
Q33
Q35
Q36
Q37
Q43 _ Q48
Description of Question
Location of airport
Use of aircraft deicing chemicals
Discharge of deicing stormwater to surface
water
Number of deicing stormwater outfalls
SWPPP Information
Number of deicing days per year
Months deicing is typically performed
Location where aircraft deicing is
performed at the airport
Current collection and containment
methods for deicing stormwater
Segregation of high and low concentration
stormwater and percentage of glycol in
collected stormwater
Destination or disposal method for
collected stormwater
On-site or off-site glycol recovery
Types of on-site treatment for collected
stormwater
Cost Model Application
Determine geographical location for the
airport
Determine if airport should be included in
collection and treatment cost estimates
Determine if on-site treatment may be
required
Estimate the cost for annual monitoring
Estimate cost to either create/update
existing SWPPP with monitoring data
Estimate the collection system operating
time
Estimate the treatment system operating
period
Estimate collection area for ADF-
contaminated stormwater
Determine technology in place and
estimate current collection percentage
Determine concentration and volume of
collected stormwater that may require on-
site treatment
Determine volume of stormwater that may
require further on-site treatment
Determine volume of stormwater that may
require further on-site treatment
Determine if treatment technology is in
place at airport
July 2009
11-10
-------
Technical Development Document for Proposed Effluent 11. Technology Costs
Limitation Guidelines and Standards for the Airport Deicing Category
EPA estimated the airport site characteristics, including paved and grass areas where
ADF may have dripped from aircraft during taxi or takeoff, based on a relationship to total
airport runway area (FAA). The method used to estimate paved and grass areas relative to total
airport runway area is described in a memorandum entitled Methodology to Estimate ADF
Contaminated Stormwater Flows (DCN AD00908), which can be found in the Airport Deicing
Category administrative record. Combining the monthly precipitation data with the paved and
grass area data allows the cost model to estimate total volume of ADF impacted stormwater that
may require collection and treatment and to predict the capital and annual cost for the collection
and treatment equipment.
11.3 General Methodology for Estimating Collection and Treatment Technology
Costs
This subsection describes the methodology for estimating costs, including the
components of cost, development of cost equations that use airport-specific model inputs to
estimate capital and annual operating costs, the cost model, and any assumptions made to
develop costs.
11.3.1 Overview of the ADF Collection and Treatment Cost Model
Managing ADF-contaminated stormwater is a multi-step process. EPA developed a cost
model to estimate deicing stormwater control costs for each of these steps and the various
alternatives within each step. Costs for each selected alternative are combined for each airport
included in the costing effort to develop cost estimates for the different EPA collection and
control scenarios. For example, regulatory costs at an airport may include a combination of
alternatives from the collection, containment and storage, and treatment categories.
The proposed EPA regulatory requirements may require an airport to collect and control
deicing stormwater through a variety of mechanisms. As discussed in Section 10, based on
information provided in the airport questionnaire and data gathered during EPA's engineering
site visits to various airports, EPA decided to estimate costs for three collection options. Those
collection technologies include GRVs alone, a combination of GRVs and a plug and pump
system in the existing stormwater drainage system, and centralized deicing pads in combination
with GRVs. Each collection alternatives is expected to provide a different level of collection
efficiency for ADF-contaminated stormwater. Using only GRVs to mop up ADF-contaminated
stormwater from areas within the airport where aircraft are deiced is expected to collect only 20
percent of the applied ADF. Adding a plug and pump system to the existing stormwater drainage
system to collect contaminated stormwater before it leaves the airport in combination with GRVs
is expected to collect up to 40 percent of the applied ADF. Changing from gate and apron
deicing to centralized deicing pads along with GRVs is expected to increase collection to more
than 60 percent of the applied ADF. Sections 9 and 10 provide detailed information on each of
these collection alternatives.
Once ADF-contaminated stormwater has been collected, the airport has a variety of
alternatives for control, ranging from disposal at a POTW, to off-site recycle and recovery, on-
site recycle and recovery, or on-site treatment and disposal. Again, using information gathered
from the airport questionnaire and EPA's engineering site visits, EPA designed the cost model to
July 2009 11-11
-------
Technical Development Document for Proposed Effluent 11. Technology Costs
Limitation Guidelines and Standards for the Airport Deicing Category
estimate costs for airports using one of the following alternatives to manage ADF-contaminated
storm water:
• Discharge to a POTW through existing municipal sewers;
• Contract haul to a POTW;
• Contract haul to an off-site glycol recovery and recycling facility;
• Treat on site at a glycol recovery and recycling facility using either ultrafiltration
and reverse osmosis or mechanical vapor recompression and distillation; or
• Treat on site to meet specific criteria and discharge to surface water using either
an AFB treatment system or aerated ponds and lagoons.
For any of the above selections, the airport may need on-site containment or storage for
the collected deicing stormwater before it reaches its final destination. In these cases,
containment and storage selections can include ponds, underground storage tanks, above-ground
storage tanks, or temporary storage tanks (e.g., frac tanks). EPA decided to estimate costs for
each storage option and, based on each airport's physical features such as available space, select
the best storage alternative. The collection and control scenarios, for which EPA estimated costs
did not include on-site treatment through a recovery and recycling facility or aerated ponds and
lagoons for costing. As discussed previously, however, these technologies may be viable
alternatives to AFB treatment at specific airports.
Airports that currently do not have any collection and containment for ADF-
contaminated stormwater may also need to determine their best alternatives to achieve EPA's
proposed rule. For these cases, airports may need to conduct an extensive monitoring program at
each of their outfalls to determine how ADF-contaminated stormwater is leaving the airport and
to develop design parameters for a new collection and containment system. For large airports
with many outfalls, the costs to conduct a continual wet-weather monitoring program during the
deicing season can be extensive. In addition, airports that elect to install on-site treatment and
discharge to surface waters will be faced with monthly monitoring requirements to verify
performance. Therefore, EPA decided to include costs for airports to conduct an initial
monitoring program as well as monthly COD effluent monitoring from on-site treatment
systems.
The airport deicing cost model considers each of these alternatives to develop a costing
scheme for collecting and containing ADF-contaminated stormwater at each airport. The cost
model also takes into account the effectiveness of each airport's current collection and control
program for ADF-contaminated stormwater to determine what incremental costs should be
applied to improve collection efficiency to comply with the proposed rule.
In general, EPA's approach to develop costs for the surveyed airports consists of the
following steps:
• Step 1: Develop cost equations for each collection, storage, and treatment
alternative evaluated for proposal using the model airport data;
• Step 2: Estimate an airport's current level of spent ADF collection (i.e., 20
percent, 40 percent or greater than 60 percent) based on information provided in
the airport questionnaire;
July 2009 11-12
-------
Technical Development Document for Proposed Effluent 11. Technology Costs
Limitation Guidelines and Standards for the Airport Deicing Category
• Step 3: Apply the collection and treatment cost equations to those airports that
currently collect and manage less than the collection and control scenario
percentage being evaluated to determine airport-specific capital and annual costs
for that scenario; and
• Step 4: Estimate annualized costs by airport for each collection and control
scenario from the airport-specific capital and annual cost components.
The airport-specific component costs are combined as needed based on the regulatory
option and then scaled up to represent national numbers based on EPA's weighting factors. A
description of the weighting factors for the Airport Deicing Category is provided in the
administrative record (USEPA, 2008a). The subsections below describe each of the steps for
developing annualized cost estimates by airport.
11.3.2 Cost Model Equation Development
Based on the available data, EPA developed cost equations for each collection, storage,
and treatment alternative that could be applied to those airports not currently achieving the target
collection percentage in the proposed rule. In general, the Agency developed cost equations from
the model airports using empirical data rather than attempting to estimate costs for individual
components within a particular collection or treatment alternative. EPA believes using empirical
data is a better way to estimate costs because all installed capital and annual operating costs are
rolled under a single value, eliminating the concern that specific components may not have been
included. EPA used the empirical cost data from the model airports and information supplied by
equipment vendors along with airport-specific information to develop normalized cost equations
that could then be projected to other airports based on common variables. Section 12.4 provides
specific examples of how the empirical costs from the model airports were projected to other
airports based using a common variable. The subsections below describe the development of the
normalized cost equations for each collection, storage, and treatment alternative for ADF-
contaminated stormwater.
11.3.2.1 Collection and Control Alternatives for Spent ADF
To select appropriate collection and control technologies, airports must first evaluate the
amount of ADF-contaminated stormwater leaving the site through a monitoring or surveillance
program. The monitoring program can involve seasonal wet-weather sampling at stormwater
outfalls suspected of ADF contamination. To develop costs for an airport-wide sampling and
monitoring program, EPA estimated labor hours for both sample plan development and sample
collection, determined vendor costs for equipment to collect samples and measure flows, and
obtained unit costs from environmental laboratories for sample analysis. Specifics for each of
these costs are provided in a memorandum entitled Estimated Costs for Initial Monitoring,
Engineering Assessment and SWPPP Updates for Airports (ERG, 2008c).
To establish an estimated monitoring cost for a typical airport, EPA used a surrogate
airport having five deicing stormwater outfalls and 21 deicing days per year. Estimated annual
monitoring costs for this airport would be approximately $410,700. Using this cost as a basis,
EPA developed an equation that the cost model used to estimate annual monitoring costs based
on the number of deicing outfalls and deicing days per year at other airports.
July 2009 11-13
-------
Technical Development Document for Proposed Effluent 11. Technology Costs
Limitation Guidelines and Standards for the Airport Deicing Category
Using the flow and concentration data from the monitoring program, airports will
perform various engineering assessments to evaluate the percentage of spent deicing fluid that is
reaching surface water and will develop options to decrease losses to the environment. The
engineering assessment will likely be conducted by a consulting firm on a time-and-materials
basis and will prepare a report that can be used to either prepare a new SWPPP or update an
existing SWPPP with new best management practices (BMPs) for ADF-contaminated
stormwater. EPA estimated costs for an outside consultant to prepare an engineering assessment
using the data from the surrogate airport having five deicing outfalls and 21 deicing days per
year. EPA estimated the cost for this assessment to be $28,400, or $5,680 per stormwater outfall.
Using the engineering unit cost, the Agency developed an equation that the cost model used to
estimate costs for an engineering assessment based on the number of deicing stormwater outfalls.
Once airports have determined the amount of COD and glycol entering surface waters
and performed an engineering assessment, they can design a collection and control strategy to
decrease the amount of ADF leaving the airport through stormwater outfalls. Airports typically
use one of three collection technologies for ADF-contaminated stormwater: GRVs, plug and
pump, and deicing pads. The normalized capital and annual costs for each collection and control
alternative are described below.
Glycol Recovery Vehicles
Airports use GRVs to collect ADF-contaminated stormwater from various locations
including gate and apron areas, taxi areas and centralized deicing areas. GRVs can be either
truck-mounted systems or tow-behind units. GRVs are expected to collect approximately 20
percent of the ADF applied to the aircraft. Because GRVs collect ADF-contaminated stormwater
from a variety of locations, the glycol concentration in stormwater collected by GRVs is
expected to range between 0.5 and 10 percent (Switzenbaum, et al., 1999).
Using information collected by EPA during the rulemaking development, deicing outfall-
normalized GRV costs were developed based on model site data reported in the airport
questionnaire. EPA assumed that the number of deicing outfalls would be an approximation of
the size of the deicing area; more deicing outfalls would indicate a larger deicing area and
increased costs associated with removing ADF from that area. Therefore, EPA related GRV
capital and annual cost data reported by six model airports to an airport's reported number of
deicing outfalls to develop an average capital and annual cost per deicing outfall. Table 11-3
shows the cost of GRVs reported in the airport questionnaire, the number of outfalls at each
reporting airport, and the normalized GRV capital cost based on the number of outfalls.
July 2009 11-14
-------
Technical Development Document for Proposed Effluent
Limitation Guidelines and Standards for the Airport Deicing Category
11. Technology Costs
Table 11-3. Reported Costs for GRVs and Outfall-Normalized Capital and Annual Costs
Airport
1
2
o
3
4
5
6
Number of
GRVs
1
1
1
1
1
1
Number of
Deicing
Stormwater
Outfalls
3
5
2
5
5
5
GRV Capital
Cost
(2006 $)
$416,000
$315,000
$318,000
$374,000
$910,000
$610,000
GRV Annual
Operating and
Maintenance
Cost
(2006 $/yr)
$16,200
$82,400
$5,200
NA
$45,800
NA
Averages
GRV
Normalized
Capital Cost
($/outfall)
$138,500
$63,000
$158,800
$74,800
$91,000
$121,900
$108,000
GRV
Normalized
Annual Cost
($/yr/outfall)
$5,400
$16,500
$2,600
NA
$4,600
NA
$7,300
Source: EPA Airport Questionnaire Database, 2006 (USEPA, 2006a).
NA - Data not available.
By multiplying the average GRV normalized capital and annual cost factors in Table 11-3 with
the number of deicing outfalls reported in the airport questionnaire, the cost model could predict
capital and annual costs for those airports currently not using GRVs to collect at least 20 percent
of applied ADF.
Plug and Pump with GRVs
The plug and pump collection system with GRVs is applicable to airports that deice at the
gate rather than at centralized areas. One benefit of deicing at the gate is that it allows the
components of the existing collection system infrastructure (i.e. existing storm sewers) to be
incorporated into the plug-and-pump collection system. This can reduce the costs associated with
designing and constructing a centralized deicing system (e.g., deicing pads). Another benefit of
at-gate deicing is that airline employees who conduct deicing can conduct other tasks such as
baggage handling and aircraft departure duties. The primary drawback to at-gate deicing is that a
much more dilute ADF-contaminated stormwater is collected relative to centralized deicing
systems, which reduces the feasibility of ADF recycling (Switzenbaum, et al., 1999).
The plug and pump collection system utilizes the airport's existing stormwater collection
system infrastructure in combination with GRVs to contain and collect ADF-contaminated
stormwater. The plug-and-pump system operates by placing either temporary inflatable balloons
or storm sewer shutoff valves in the existing storm sewer system. During deicing events, the
balloons are inflated and storm sewer shutoff valves are closed, trapping the ADF-contaminated
stormwater in the collection system. GRVs pump the trapped contaminated stormwater from the
storm sewer system and transport it to on-site storage while also mopping up ADF-contaminated
stormwater from the gate, ramp, and apron area surfaces following application.
Two airports provided sufficient information to estimate costs for a plug and pump type
collection system. During the 2005 deicing season, the first airport used sewer balloons at eight
locations, storm sewer shutoff valves at four locations, and two catch basin inserts. In addition,
this airport utilized two GRVs; one is a traditional truck-based GRV and the other is a GRV unit
(a V-Quip Ramp Ranger) that is towed behind a tractor. The estimated capital cost for this
July 2009
11-15
-------
Technical Development Document for Proposed Effluent
Limitation Guidelines and Standards for the Airport Deicing Category
11. Technology Costs
airport's block-and-pump system is approximately $790,000. The plug and pump system
operated at this airport prevents ADF-contaminated stormwater from discharging through three
deicing stormwater outfalls (USEPA, 2006c).
The second airport, in a small area where ADF-contaminated stormwater is generated,
uses the plug and pump system to prevent ADF-contaminated stormwater from discharging to
surface waters. This airport operates 22 block-and-pump locations that prevent ADF-
contaminated stormwater from leaving the airport through a maximum of five outfalls.
According to the airport, the annual budget to operate the plug and pump system is
approximately $1,300,000. This cost includes permit monitoring, glycol management, the
pumping contractor, plus other miscellaneous costs (ERG, 2007b).
To estimate both capital and annual operating and maintenance (O&M) costs for other
airports where plug and pump collection may be applicable, EPA normalized the costs based on
the number of deicing outfalls. EPA assumed that the number of deicing outfalls would reflect
the size of the deicing area (i.e., more deicing outfalls would indicate a larger deicing area and
increased costs associated with removing ADF from that area). Table 11-4 presents the
normalized capital and annual costs for the plug and pump collection system at these airports.
Table 11-4. Normalized Capital and Operating Costs for the Plug-and-Pump Collection
System
Airport
Airport 1
Airport 2
Deicing
Outfalls
o
J
5
Total Capital
Cost
(2006 $) l
$790,000
NA
Annual O&M Cost
(2006 $)
NA
$1,300,000
Normalized
Capital Cost
($/outfall)
$263,400
NA
Normalized Annual
O&M Cost
($/outfall)
NA
$260,000
Includes both wastewater storage and treatment equipment.
NA - Data not available.
Deicing Pads with GRVs
Central deicing pads for deicing waste management minimize the volume of deicing
waste by restricting deicing to confined and controlled areas. Data collected by EPA indicate that
deicing pads allow airports to collect nearly 68 percent of the ADF (USEPA, 2006e) applied to
aircraft as compared to block-and-pump collection systems, which can capture up to 42.5 percent
of the applied ADF (USEPA, 2006c). Central deicing pads are generally constructed of concrete
with sealed joints to prevent losing glycol through the joints. A number of airports also use
GRVs in combination with centralized deicing pads to maximize collecting and containing ADF-
contaminated stormwater. Pads are typically located near the gate areas or near the ends of the
runways so that planes can be deiced just prior to takeoff. Deicing at central pads is
environmentally preferable to deicing at the gate because it minimizes the size of the collection
area and the amount of wastewater that must be collected. However, central deicing pads are
sometimes difficult to manage because of scheduling aircraft in and out of the pads during storm
events. In addition, although the name implies a small collection area, central pads designed to
accommodate more than one commercial aircraft generally encompass several acres.
July 2009
11-16
-------
Technical Development Document for Proposed Effluent
Limitation Guidelines and Standards for the Airport Deicing Category
11. Technology Costs
Three airports have provided information on their costs to install centralized deicing
pads, show in Table 11-5. To estimate installed capital costs for deicing pads, EPA normalized
the costs based on the number of aircraft take-offs during the deicing season. EPA decided to use
the number of aircraft take-offs as the normalizing factor since the number of deicing pads and
their size is directly related to expected airport ground traffic during the deicing season. The
normalized capital costs for centralized deicing pads at the three airports, which provided cost
data is provided in Table 11-5.
Table 11-5. Normalized Installed Capital Costs for Centralized Deicing Pads
Airport
Airport 1
Airport 2
Airport 3
Deicing Season
Aircraft Take-offs
14,911
125,143
246,286
Total Installed Capital Cost
for all Deicing Pads
(2006 $)
$5,100,000
$35,000,000
$79,300,000
Average
Normalized Capital Cost
($/annual takeoffs)
$342
$280
$322
$314.56
Deicing pads are expected to require less operation and maintenance than plug-and-pump
systems. Although deicing pads typically require GRVs to mop up additional ADF-contaminated
stormwater that does not reach the underground collection system, they require less manual labor
to install than to install storm sewer plugs and catch basin inserts. In addition, most plug-and-
pump systems use the existing storm sewer system, so gates and valves typically must be opened
and closed manually instead of having programmable logic controller (PLC) controlled gates and
valves that an operator can manage from a remote location. In addition, if the underground
storage system associated with the deicing pads has sufficient capacity to contain major storm
events, then temporary tanks typically are not required. According to information EPA obtained
during the site visit to the General Mitchell International Airport in Milwaukee (USEPA, 2006c),
the airport rents five temporary tanks to temporarily store spent ADF-contaminated stormwater
before it is sent to the POTW for anaerobic treatment. According to General Mitchell
International airport personnel available during the site visit, the total rental cost for temporary
tanks is approximately $5,000/month.
Because no annual O&M cost data for deicing pads were provided in responses to the
airport questionnaire or available at the time of cost model development, EPA used annual
block-and-pump cost data, but subtracted the rental cost for temporary storage tanks. According
to the questionnaire data, the deicing season in the Milwaukee area is approximately five months,
so total rental of frac tanks during the deicing season is approximately $25,000. The number of
annual departures in Milwaukee is reported to be 85,128 (USDOT, 2006). Normalizing frac tank
rental to the number of departures gives $0.29/departure for rental of frac tanks. If O&M costs
for the block-and-pump system in Milwaukee were normalized to the number of departures,
costs would be $9.16/depature ($260,000/outfall x 3 outfalls/85,128 departures). Subtracting the
departure-normalized frac tank costs from the normalized block-and-pump O&M cost gives an
estimated cost of $8.87/departure ($9.16 - $0.29).
July 2009
11-17
-------
Technical Development Document for Proposed Effluent 11. Technology Costs
Limitation Guidelines and Standards for the Airport Deicing Category
11.3.2.2 Off-Site Treatment Alternatives
After collecting deicing stormwater, the airport can choose to either treat the stormwater
or send it away to be treated before its ultimate discharge. Off-site treatment alternatives include
contract hauling to a POTW for anaerobic treatment, discharging to a POTW for aerobic
biological treatment, or contract hauling to an off-site glycol recovery/recycling facility.
Discharge to a POTW
Airports have two options for discharging to POTWs: discharge to the POTW through
municipal sanitary sewers or contract haul to the POTW. From information provided in the
airport questionnaire, EPA found that 62 airports currently discharge either a portion or all of
their ADF-contaminated stormwater to a POTW through a municipal sanitary sewer system for
aerobic treatment. To estimate costs for discharging to the POTW via the sanitary sewer, EPA
averaged unit-cost data provided by five airports in the airport questionnaire. The average cost
for aerobic POTW treatment of wastewater from these five airports is $0.0I/gallon (USEPA,
2006a).
Since most airports that have can discharge ADF-contaminated stormwater to POTWs
are likely already doing so, EPA believes these airports will not install new on-site treatment but
will instead discharge additional stormwater to the POTW to comply with the proposed rule. For
these airports, EPA developed the following equation to estimate incremental POTW charges.
ADF Use x Target % Capture x • 1 $°'°1 '
o,
o glycol gal
- ADF Use x Current % Capture x x —:— (11-1)
^ F % glycol gal J
= Incremental POTW Costs
For this equation, the estimated percent glycol in the collected stormwater is expected to
vary depending on the collection technology used. For example, if GRVs are used to collect 2
percent of applied ADF, the percent glycol in the collected stormwater is expected to be
approximately 1.5 percent (1). If plug and pump with GRVs is used to collect ADF-contaminated
stormwater, the glycol percentage is expected to be 3 percent (Switzenbaum, et al., 1999), and if
deicing pads are used to collect ADF, the percentage is expected to be approximately 5 percent.
Another alternative is contract hauling ADF-contaminated stormwater to a POTW having
anaerobic digesters. EPA obtained contract hauling and POTW anaerobic treatment costs from
one airport that contract hauls to a POTW for anaerobic treatment for approximately $0.16 per
gallon(USEPA, 2006c). EPA believes that, since this airport is currently contract hauling to a
POTW with anaerobic digestion, it will continue to do so even if it collects additional ADF-
contaminated stormwater to comply with the proposed rule. Using the unit-cost data for both
hauling and treatment, EPA developed an equation to estimate incremental costs for this airport
to continue contract hauling to a POTW for anaerobic treatment.
July 2009 11-18
-------
Technical Development Document for Proposed Effluent 11. Technology Costs
Limitation Guidelines and Standards for the Airport Deicing Category
Off-Site Glycol Recovery and Recycle
Airports within close proximity to a recovery and recycle facility, such as those operated
by the Environmental Quality Company or Inland Technologies, have an option to contract-haul
their collected ADF-contaminated stormwater. The airport questionnaire database shows that an
estimated 13 airports currently ship their collected ADF-contaminated stormwater to an off-site
facility for glycol recovery. EPA believes airports will continue to contract haul additional ADF-
contaminated stormwater to comply with the proposed rule. To estimate incremental costs to
contract haul additional ADF-contaminated stormwater to an off-site facility for glycol recovery,
EPA obtained cost data from one airport (USEPA, 2006a). This airport currently collects 40
percent of its applied ADF and spends $4.88 per gallon on average to contract haul and treat its
ADF-contaminated stormwater for off-site glycol recovery and recycle. Based on the contract-
haul costs from this airport, EPA developed an equation to estimate incremental contract-haul
and off-site glycol recovery and recycle costs.
Of the 13 airports that currently ship their collected ADF-contaminated stormwater
offsite, EPA estimated that five airports currently collect less than 60 percent of their applied
glycol for off-site glycol recovery. These five airports will likely continue to contract-haul any
additional collected ADF to comply with the proposed rule, and therefore the cost model applied
the incremental off-site glycol recovery costs to only these five airports.
11.3.2.3 On-Site Treatment Alternatives
On-site treatment alternatives evaluated by EPA during the data collection phase of this
rulemaking included ultrafiltration with reverse osmosis (USEPA, 2006f), mechanical vapor
recompression (MVR) with distillation (USEPA, 2006g), anaerobic fluid bed biological
treatment (AFB) (USEPA, 2006h), and aerobic biological treatment ponds (USEPA, 2008b). For
these on-site treatment alternatives, only the anaerobic fluid bed and aerobic biological treatment
ponds effectively treated the wastewater for discharge to surface water through a permitted
outfall. Although ultrafiltration with reverse osmosis and MVR both can recover and potentially
recycle glycol, neither generated effluents that could be directly discharged to surface water.
Residual streams from each of these technologies contained significant levels of COD that
required discharge to a POTW for further processing. The following subsections provide EPA's
method for estimating costs for each of these technologies along with the cost equations used by
the cost model to predict costs for all applicable airports.
Ultrafiltration with Reverse Osmosis
Ultrafiltration with reverse osmosis (UF/RO) treatment is considered a recycle and
recovery system for spent ADF. The technology generates a concentrated glycol-containing
stream that can be recycled for possible use in toilets onboard commercial aircraft (i.e., lavatory
fluid) or recovered and contract hauled off site for resale as a chemical feedstock. The effluent
from the UF/RO system contains lesser amounts of glycol, cBOD, and COD that is typically
discharged to a POTW for further processing. The combined UF/RO process will increase the
glycol concentration in collected ADF-contaminated stormwater to approximately 10 percent
from its original concentration of between 0.5 and 5 percent depending on the type of collection
technology used by the airport.
July 2009 11-19
-------
Technical Development Document for Proposed Effluent
Limitation Guidelines and Standards for the Airport Deicing Category
11. Technology Costs
Installed capital and O&M costs for a UF/RO treatment system to recycle and recover
spent ADF-contaminated stormwater were provided by New Logic Research, Inc. (ERG, 2007c).
According to New Logic, capital costs for a 3 million gal/yr treatment system is approximately
$962,000 (New Logic Research, 2001). This capital cost does not include storage tanks needed
for flow equalization prior to treatment; including storage tanks would increase the total installed
cost to approximately $3.85 million.
To estimate both capital and annual O&M costs for airports to install and operate UF/RO
treatment systems, EPA normalized the costs based on flow (gal/day). Table 10-8 presents the
flow-normalized installed capital costs for the UF/RO treatment system. The costs in Table 10-8
assume the effluent from the UF/RO treatment system can be discharged to a POTW for further
glycol, cBOD, and COD reductions. Costs in Table 11-6 also include costs for storage tanks and
transfer equipment. The storage tanks provide sufficient equalization to dampen flow and
concentration changes indicative of deicing events.
Table 11-6. Flow-Normalized Installed Capital Costs for the UF/RO ADF Treatment
System
Location
New Logic Research Inc.,
Emeryville, CA
Design Flow
(gal/day)
18,750
Installed Capital Cost
(2006 $) 1
$3,850,000
Design Flow-Normalized
Capital Cost
($/gal/day)
$205.33
Includes both wastewater storage and treatment equipment.
EPA calculated estimated installed capital costs for a UF/RO treatment system using the
cost model by multiplying the normalized capital cost by the estimated daily flow for each
airport. Estimated daily flows to treatment were calculated from ADF purchase information for
each airport, the expected spent ADF collection efficiency for the airport (e.g., 40 percent if plug
and pump with GRVs are used), the assumed glycol concentration in the collected stormwater,
and the estimated number of days per year the UF/RO system is operated. For the glycol
concentration in stormwater, EPA assumed a range of concentrations depending on the type of
collection method used by the airport to collect spent ADF. Table 11-7 lists the concentration
ranges, along with the expected ADF percentage collected for each major type of collection
method.
Table 11-7. Expected ADF Percentage Collected and ADF Concentration in Collected
Stormwater
ADF and Stormwater Collection
Method
GRVs Only
Plug and pump with GRVs
Centralized Deicing Pads with GRVs
Expected ADF Percentage
Collected
20
40
60
Expected ADF Concentration in
Stormwater
1.5
o
J
5
EPA selected the range of ADF concentrations in collected stormwater based in
information provided in the literature. According to one source (19), the minimum concentration
of glycol in stormwater should range between 3 and 5 percent to be practical for recovery.
July 2009 11-20
-------
Technical Development Document for Proposed Effluent 11. Technology Costs
Limitation Guidelines and Standards for the Airport Deicing Category
Another source indicated typical values for UF/RO treatment can be approximately 1.5 percent
(20). Since the concentration of glycol in stormwater will likely increase as more centralized
locations are used at an airport for deicing, EPA decided to pair the range of concentrations with
the increasing performance of the collection system.
UF/RO treatment systems for ADF-contaminated stormwater will require flow
equalization tanks prior to treatment to provide a more consistent flow to the system, even
though aircraft deicing operations may not be occurring. Since operating days per year for the
UF/RO systems will vary between airports and is based on annual precipitation and the number
of deicing events per year, EPA decided to arbitrarily select five months of operating time
(approximately 150 days/yr). Therefore, total annual volumes of collected stormwater are
divided by 150 days/yr to determine an average daily flow to the UF/RO treatment system.
Annual O&M costs for a UF/RO treatment system are calculated using the same
methodology described above. Table 11-8 shows the flow-normalized costs for operation and
maintenance of the UF/RO recycle and recovery system. EPA obtained the costs from New
Logic (20) and escalated them to 2006 costs assuming an inflation rate of 3 percent per year.
According to New Logic, annual costs to operate a 3 million gallon per year (MGY) treatment
system are approximately $320,700. For O&M costs, the flow to treatment will be multiplied
$0.107/gallon.
Table 11-8. Flow-Normalized Annual O&M Costs for the UF/RO Recycle and Recovery
System
New Logic Research Inc., Emeryville, CA
Design Flow
(gal/yr)
3,000,000
Annual O&M Cost
(2006 $) 1
$320,700
Flow-Normalized
Annual O&M Cost
($/gal)
$0.107
Includes both wastewater storage and treatment equipment annual costs.
MVR with Distillation
The MVR and distillation units are considered a recovery and recycle system for spent
ADF. The system is typically used when glycol concentrations in the stormwater are greater than
5 percent and present an opportunity to generate a recyclable product that could result in a
deicing cost credit in some cases. When glycol concentrations are less than 5 percent, recovery
and recycle of glycol is not practical, and therefore glycol-contaminated stormwater is typically
discharged directly to the POTW for treatment (Switzenbaum, et al., 1999). The
MVR/distillation technology generates a concentrated glycol-containing stream (>99 percent)
that can be sold as a chemical feedstock or possibly be recycled as ADF with additional
processing. The effluents from the MVR and distillation units contain glycol, cBOD, and COD
that must be discharged a POTW for further treatment.
Two airports provided installed capital and O&M costs for a MVR/distillation system for
recovery and recycle of propylene glycol from spent ADF-contaminated stormwater (ERG,
1007d; ERG, 2007e). According to the first airport, installed capital costs for their 10-million
gallon/season treatment system was approximately $19,421,000. The second airport reported the
installed capital cost of their system to be $20,315,000 and it treats approximately 3.3 million
July 2009 11-21
-------
Technical Development Document for Proposed Effluent
Limitation Guidelines and Standards for the Airport Deicing Category
11. Technology Costs
gallons/season. This cost also includes costs for design, storage tanks, collection piping and
structures, pumps, electrical and controls, plus the ADF processing equipment.
To estimate both installed capital and annual O&M costs for other airports to install and
operate MVR/distillation to recover and recycle propylene glycol from ADF-contaminated
stormwater, EPA normalized the costs based on flow (gal/day). Table 11-11 lists the flow-
normalized installed capital costs for the MVR/distillation recovery and recycle system. The
costs assume the effluent from the overheads from the MVR/distillation treatment system can be
discharged to a POTW for further glycol, cBOD, and COD reductions. The costs shown in Table
11-9 do not include costs to treat the overheads by the POTW.
Table 11-9. Flow-Normalized Installed Capital Costs for the MVR/Distillation ADF
Treatment System
Airport
1
2
Average
Design Flow
(gal/season)
3,265,000
10,000,000
Installed Capital Cost
(2006 $) 1
$20,315,000
$19,421,000
Flow Normalized Capital
Cost
($/gal)
$6.22
$1.94
$4.08
Includes both wastewater storage and treatment equipment.
To estimate the installed capital cost for a MVR/distillation glycol recovery and recycle
treatment system, the cost model multiplied by the average flow normalized capital cost in Table
11-11 by the estimated annual flow for each airport.
Annual O&M costs are calculated using the same methodology described above.
According to Inland Technologies (ERG, 2007f), costs are approximately $0.05/gallon to treat
ADF-contaminated stormwater using the MVR/distillation system. This cost includes costs for
electricity for the evaporators, labor, and general maintenance equipment. O&M costs for an
MVR/distillation recovery and recycle system is calculated by multiplying the annual flow of
ADF-contaminated stormwater to the MVR system by $0.05/gallon.
AFB Biological Treatment
The AFB biological treatment system uses a vertical, cylindrical tank in which the ADF-
contaminated stormwater is pumped upwards through a bed of granular activated carbon at a
velocity sufficient to fluidize, or suspend, the media. The organic carbon in the ADF-
contaminated stormwater is treated by a thin film of microorganisms that grows and coats each
granular activated carbon particle, providing a vast surface area for biological growth. The
anaerobic microorganisms occur naturally in sediment, peat bogs, cattle intestines, and even
brewer's yeast. Breakdown products from the AFB treatment system include methane, carbon
dioxide, and new biomass. Effluent from the AFB system can be discharged to a local POTW or,
in most cases, directly to surface water.
Treating wastes using an AFB biological treatment system has several advantages over an
aerobic system. The AFB system requires much less energy since aeration is not required and
produces significantly less sludge than an aerobic process. In addition, because the biological
July 2009
11-22
-------
Technical Development Document for Proposed Effluent
Limitation Guidelines and Standards for the Airport Deicing Category
11. Technology Costs
process is contained in a sealed reactor, odors are eliminated. Methane generated by the AFB can
be used to heat facility boilers, and if economically feasible, to generate electricity (USDoD,
2003) creating a potential source of revenue.
Two airports have provided installed capital and O&M costs for AFB biological
treatment systems to treat ADF-contaminated stormwater (ERG, 2007g; ERG, 2007h). Table 11-
10 lists the load and flow-normalized installed capital costs for the both AFB treatment systems.
The costs shown in Table 11-10 include costs for storage equipment such as tanks and ponds
prior to anaerobic treatment. The storage equipment provides sufficient equalization to dampen
flow and concentration changes indicative of deicing events.
Table 11-10. Load-Normalized Installed Capital Costs for the Anaerobic Fluid Bed
Reactors
Airport
1
2
Average
COD Loading
(Ibs/day)
5,200
3,400
Installed Capital Cost
(2006 $) 1
$8,100,000
$5,990,000
COD Load-Normalized
Capital Cost
($/lbs COD/day)
$1,558
$1,761
$1,659
Includes both wastewater storage and treatment equipment.
Because of the large difference between the design flow-normalized capital costs shown
in Table 11-10, EPA decided to estimate installed capital costs for AFB treatment systems based
on COD loading. COD loading is calculated by converting the annual applied ADF (gal/yr) at
each airport to average daily COD (Ibs/day) throughout the entire deicing season and assuming
an ADF collection efficiency based on the selected collection and control technology. COD
loading (Ibs/day) is then be multiplied by $l,659/lbs COD/day to determine the installed capital
cost.
Annual O&M costs for the AFB biological treatment system are calculated using the
same methodology described above. Table 11-11 lists the COD loading and flow-normalized
costs for O&M of the AFB treatment system. For O&M costs, the COD loading to treatment
(Ibs/day) is multiplied by $77.72/yr/lbs/day.
Table 11-11. Load Normalized Annual O&M Costs for the Anaerobic Fluid Bed Reactors
Airport
1
2
Average
COD Loading
(Ibs/day)
5,200
3,400
Annual O&M Cost
(2006 $) 1
$510,000
$195,000
COD Load-Normalized
Annual O&M Cost
($/yr/lbs/day)
$98.08
$57.35
$77.72
Includes both wastewater storage and treatment equipment annual costs.
July 2009
11-23
-------
Technical Development Document for Proposed Effluent 11. Technology Costs
Limitation Guidelines and Standards for the Airport Deicing Category
Aerobic Biological Treatment Ponds and Piping Costs
Airports use retention ponds to equalize both the flow and concentration of aircraft
deicing fluid (ADF)-contaminated stormwater prior to POTW discharge, or to reduce BODs
prior to discharging the wastewater to surface water discharge. The actual size of the retention
ponds depends primarily on expected precipitation and drainage area. The ability to remove
BOD5 from retention pond effluent depends on oxygen availability for the natural
microorganisms present in the retention pond.
To estimate sizes and costs for ponds at airports, EPA used information provided by a
large Midwest airport to estimate total pond volume relative to airport runway area. Next, EPA
used information from an airport in the central United States to develop a unit cost for retention
ponds (2006 $/gal). For airports that include surface aerators to enhance biological treatment,
EPA used data provided by a mid-size airport located in the upper Midwest to estimate the
number and size of surface aerators required for biological treatment.
One concern raised by EPA is the space requirements for retention ponds. Unlike other
technologies, ponds require a large footprint and some airports might not have sufficient space
for expand and construct a pond (e.g., airports within major cities). To determine if space was
available for a retention pond system at an airport, EPA examined the total current land use at
representative airports located in urban areas. Based on the analysis, EPA decided that if an
airport currently utilizes more then 35 percent of its current area for active airport operations,
then it was not a candidate for a retention pond system. These airports would need to use above-
ground or below-ground tanks to store ADF-contaminated stormwater prior to treatment or off-
site discharge to a POTW.
EPA previously developed a relationship between gate/apron areas and runway area
using runway data (FAA) and gate and apron information developed from Google Earth (Google,
2007), a global mapping system, for specific airports. The relationship of paved gate/apron areas
to airport runway area is estimated to be 1.8 acres gate/apron per acre of runway. EPA used
information from the large Midwest airport's pond storage volume develop the following cost
equation to estimate retention pond volume based on the airport's published runway area.
43,560ft2 1.8 gate and apron acre 169,082 gal
Runway Area, x x
acre run way acre gate and apron acre (H-2)
= Estimated Retention Pond Volume (gal)
To estimate the cost for the retention pond, EPA used costing data provided by a large
central U.S. airport. Table 11-12 lists the costs (2006 $) for various ponds that are currently
being utilized to collect and control ADF at this airport (ERG, 2007e).
July 2009 11-24
-------
Technical Development Document for Proposed Effluent
Limitation Guidelines and Standards for the Airport Deicing Category
11. Technology Costs
Table 11-12. Installed Capital Costs for Various Retention Ponds
Pond
1
2
3
4
5
Volume
(gal)
6,000,000
12,500,000
4,200,000
3,200,000
8,800,000
Installed Capital Cost
(2006 $)
$4,683,000
$2,617,000
$4,562,000
$1,055,000
$2,110,000
Average
Unit Capital Cost
($/gal)
$0.78
$0.21
$1.09
$0.33
$0.24
$0.53
Using the data from Table 11-12, EPA calculated an average unit capital cost of $0.53 per gallon
of pond storage. EPA then combined the equation for estimating the retention pond volume with
this average unit capital cost to develop an equation for estimating installed costs for retention
ponds at airports when the total runway area is known.
To determine annual O&M costs for retention ponds at airports, EPA obtained
information provided in the airport questionnaire. According to one response to the
questionnaire, annual costs to provide weed and sediment control at an airport's three ponds is
approximately $6,000/yr. The total volume of all three ponds is 20,800,000 gallons. Based on an
annual cost of $6,000/yr and a total volume of 20,800,000 gallons, the unit O&M costs for
retention ponds is approximately $0.0003/gallon. EPA combined the pond volume equation
above with the unit cost for O&M to generate an equation for estimating annual O&M costs for
retention pond at airports from total runway area.
To enhance biological treatment within a retention pond, airports have retrofit ponds with
surface aerators to increase BOD5 removal. The amount of aeration needed is directly related to
the amount of ADF used at the airport, since ADF is the primary source of BOD5 in airport
stormwater. To estimate the number of surface aerators needed for retention ponds based on total
annual ADF usage, EPA used information provided by the large Midwest airport. EPA
conducted both a site visit and sampling episode at the airport (USEPA, 2008b). This airport uses
16 25-horsepower surface aerators in its 16-million gallon treatment pond to remove BOD5 prior
to direct discharge. The aerators reduce the BODs concentration in the retention pond from
approximately 2,000 mg/1 in the winter to less than 30 mg/L when the pond is discharged in late
spring. According to estimates based on airline detailed questionnaire data, this airport uses
approximately 146,800 gallons of ADF (both Type I and Type IV) annually.
To estimate the cost to place surface aerators in retention ponds at other airports, EPA
contacted Aqua Aerobics, Inc (ERG, 2007i), a major supplier of surface aerators to the biological
treatment industry. According to Aqua Aerobics, the book price for a 25-horsepower surface
aerator is approximately $11,200 and the total installed costs for the aerators can be
approximated by assuming 1.5 times the purchased equipment cost. Using that estimate, the total
estimated installed capital cost for all aerators in the retention pond for the Midwest airport is
approximately $269,000 (2006 $).
Based on the airport's annual ADF estimated usage of 146,800 gallons/year and the total
estimated installed cost for aerators, EPA estimated the unit cost for surface aerators to be
July 2009
11-25
-------
Technical Development Document for Proposed Effluent 11. Technology Costs
Limitation Guidelines and Standards for the Airport Deicing Category
$1.83/gallon ADF used per year. Combining the installed capital cost for aerators with the cost
for the retention pond, EPA developed an equation that can be used to estimate the installed
capital cost for retention ponds with surface aerators at airports.
To determine annual O&M costs for retention ponds with aerators, EPA accounted for
the annual cost for weed and sediment control, plus the costs for labor, electrical, materials,
laboratory analysis, and additional operating chemicals. The large Midwest airport stated in its
questionnaire that it spends approximately $83,000 per year to operate the aeration pond, with
electrical cost accounting for nearly 70 percent of the total annual cost. Based on 146,800 gallons
per year of ADF use, the unit annual cost to operate the aerated retention ponds is approximately
$0.56/gallon ADF used.
11.3.2.3 Holding Tanks and Transfer Piping
Piping costs to transfer collected stormwater to holding tanks prior to either on-site or
off-site ADF management will vary for each airport. Variables that could affect transfer piping
costs include existing subsurface utilities, soil types, elevation changes, and anticipated peak
precipitation events. Because each of these variables is airport-specific and the details are not
available, EPA assumed costs would need to be estimated for each airport to construct 1,000
liner feet of new subsurface stormwater conveyance piping from various areas around the airport.
EPA believes 1,000 linear feet of piping may be overly conservative at some airports, but less
than required at others.
Elements of a stormwater piping system include subsurface concrete piping, manholes
and catch basins, a lift station and associated pumps, and knife gates through out the system to
control the direction of flow. EPA obtained 2002 costs for individual elements within the system
from RSMeans Heavy Construction Cost Data (RS, 2002) and escalated the costs to 2006
dollars. The Agency added cost factors for plumbing, electrical, mechanical, and site work to the
direct costs based on the Department of Defense MILCON estimating procedures (USDoD,
2001) to obtain total installed direct costs. Indirect costs for engineering, permits, scheduling,
performance bonds, and contractor markups were added to the total installed direct costs. Table
11-13 shows the estimated installed capital cost for 1,000 linear feet of a new stormwater
conveyance piping system at an airport is approximately $502,000. Details regarding capital
costs plus indirect cost factors are provided in a memorandum entitled Estimated Capital and
O&M Costs for Collection Ponds and Stormwater Piping (ERG, 2008d) available in the Airport
Deicing Category administrative record.
July 2009 11-26
-------
Technical Development Document for Proposed Effluent
Limitation Guidelines and Standards for the Airport Deicing Category
11. Technology Costs
Table 11-13. Estimated Cost for 1,000 Linear Feet of Stormwater Piping
Description
Trenching for stormwater piping
Backfill and compact trench after piping
Concrete stormwater piping
Manholes/catch basins
Manhole frames and covers
Knife gates including handwheel operator
Pump station to transfer collected stormwater to a holding tank
Plumbing (connectors, extra labor, etc.)
Mechanical systems (valves)
Electrical systems (conduit, motor starters, switches, wiring, etc.)
Instrumentation (PLCs, sensors, hardware, software, etc.)
Site work (clearing, grading, surveying)
Engineering
Permits
Scheduling
Performance bonds
Insurance (risk, equipment floater, public liability)
Contractor markup (handling, procuring, subcontracting, change orders, etc.)
Overhead and profit
Total Installed Capital Cost for 1,000' of Stormwater Piping
Total Cost
$3,800
$2,500
$13,100
$11,200
$3,600
$7,400
$130,500
$56,300
$60,300
$27,200
$30,500
$24,100
$29,600
$7,400
$3,000
$9,300
$8,500
$37,000
$37,000
$502,000
Source: RSMeans Heavy Construction Cost Data, 2002. (RS, 2002)
Annual O&M costs for the piping and conveyance system include labor, electrical and
spare parts for pumps and gates. To estimate labor costs, EPA assumed one operator would
monitor the system during each deicing day. The average hourly rate for an operator to monitor
the system is $32/hr based on data hourly cost data provided in the airport questionnaire. Based
on a 24-hour event, the labor cost for the stormwater conveyance piping system can be estimated
from the number of deicing days per year.
To estimate electrical costs for the collection system, EPA assumed that one 40-
horsepower pump would operate in the lift station during each storm event and would transfer
stormwater from the lift station to a retention pond. Based on a typical electrical cost of
$0.07/kWh, total cost for electricity to operate the transfer pumps can be estimated based on the
number of deicing days per year.
To estimate maintenance equipment costs for the lift station pumps and knife gates, EPA
used information from Perry's Chemical Engineers Handbook (Perry, 1984). According to
Perry's, annual replacement parts can equal approximately 6 percent of the capital cost for
equipment. Using the cost data provided in Table 11-13, the annual cost for replacement parts for
the lift station and knife gate assuming 6 percent replacement would be approximately
$8,300/year. Combining the labor, electrical, and maintenance equipment costs gives the total
annual O&M cost for the transfer and piping system.
July 2009
11-27
-------
Technical Development Document for Proposed Effluent
Limitation Guidelines and Standards for the Airport Deicing Category
11. Technology Costs
Holding Tanks
Airports use storage tanks to equalize either flow and/or concentration of ADF-
contaminated stormwater prior to POTW discharge or to an on-site treatment system. The actual
size of the storage tanks depends primarily on the amount of ADF-contaminated stormwater
generated during precipitation events and the rate at which the wastewater in the tanks can be
discharged to either the POTW or the on-site treatment system.
For airport deicing operations, ERG obtained storage tank volumes from five model
airports (USEPA, 2006e; ERG, 2007a; McQueen, R., teal.; USEPA, 2006i; USEPA, 2006J). EPA
also obtained installed capital costs for the storage tanks at three additional model airports (ERG,
2007d; ERG, 2007e; ERG, 2007h). Using departure data for each of these airports, EPA
normalized tank volumes and costs to the number of airport departures per year to develop an
equation that could be used to estimate storage tank sizes and costs at other airports. EPA chose
airport departures as a normalizing factor since the amount of tank storage needed is directly
related to the number of aircraft being deiced. In addition, the storage tanks at these airports are
used to contain ADF-contaminated stormwater prior to discharge to a POTW and/or on-site
treatment, and therefore the hydraulic capacity has likely been designed with specific discharge
requirements (e.g., maximum flow and equalized pollutant concentrations).
Table 11-14 lists departure information, storage tanks volumes, costs, and normalized
tank volumes and costs. The data in Table 11-14 indicate the volume of storage tanks at the five
airports ranges between 0.4 million gallons and 8 million gallons, with the average being
approximately 2.9 million gallons. The average unit cost for storage tanks calculated from the
data in Table 11-14 is $1.67/gallon. The departure-normalized storage tank volume, calculated
from the data provided in Table 10-16, is 24 gallons/departure/year.
Table 11-14. Storage Tank Volumes and Installed Capital Cost for Various Airports
Airport
1
2
3
4
5
Average
Number of Airport
Departures per
Year
264,051
14,911
96,475
125,143
247,165
Total Storage
Tank Volume
(gal)
420,000
1,500,000
2,000,000
2,500,050
8,000,000
Storage Tank Volume
per Departure
(gal/departure/yr)
2
101
21
20
32
24 1
Installed
Capital Cost
(2006 $)
$795,000
$2,890,000
NA
NA
$9,440,000
Storage Tank
Capital Cost
($/gal)
$1.89
$1.93
NA
NA
$1.18
$1.67
Average does not include either Airports 1 or 2.
NA - Not available.
Using both the average volume and cost data in Table 11-14, EPA developed an equation to
estimate costs for storage tanks at all airports based on the number of airport departures per year.
Only limited data were available in responses to the airport questionnaire to determine
the annual O&M cost for storage tanks. One airport reported that annual maintenance costs for
its storage tanks ranged between $50,000 and $100,000. Using the number of departures per year
from this airport (41,916/yr) and the average annual O&M cost ($75,000/yr), EPA calculated the
July 2009
11-28
-------
Technical Development Document for Proposed Effluent 11. Technology Costs
Limitation Guidelines and Standards for the Airport Deicing Category
normalized annual O&M cost for storage tanks to be $1.79/departure/year. Using the normalized
cost factor, EPA developed an equation to estimate annual O&M costs for storage tanks at
airports.
11.3.3 Cost Model Design
This subsection describes how the Airport Deicing Cost Model uses the capital and
annual cost equations, in combination with the variables included in the model input data, to
predict costs for each airport.
11.3.3.1 Airport Deicing Cost Model Description
EPA developed the Airport Deicing Cost Model (hereinafter referred to as Cost Model)
using Microsoft Access and the model uses various tables structured from the airport
questionnaire data to provide input to the design equations. EPA designed the model to first
evaluate an airport's status with the proposed collection efficiency and then build costs based on
the appropriate types of collection and treatment alternatives needed to achieve the target
collection efficiency. Figure 11-1 is a diagram showing how the cost model selects and costs the
various collection alternatives based on input information from the airport questionnaire.
The Cost Model uses a series of "Yes -No" statements and the airport's current
collection percentage for applied ADF to determine which collection technology costs to apply.
For example, if EPA estimates an airport is currently collecting 42 percent of its applied ADF,
the model calculates costs for centralized deicing pads using the appropriate cost equation and
the input variables for the specific airport. If data from the airport questionnaire indicates an
airport is currently collecting less than 20 percent of all applied ADF, the cost model estimates
costs for each collection option: GRVs, plug and pump with GRVs, and centralized deicing pads
using GRVs.
Figure 11-2 shows how the Cost Model selects the appropriate options for the collected
ADF-contaminated stormwater. As Figure 11-2 indicates, only the AFB treatment system has the
capability of treating either high or low glycol concentrations in the collected stormwater and
provides an effluent of sufficient quality for direct discharge. Technologies such as UF/RO and
MVR/distillation are more economical at higher glycol concentrations (TRB Airport Cooperative
Research Program, 2009), but still have a residual stream that contains COD concentrations
above acceptable levels, forcing airports to discharge these residual streams to POTWs for
further treatment.
The Cost Model determines which treatment technologies can be paired with the
appropriate collection technology. As indicated previously, recycle and recovery technologies
such as UF/RO and MVR with distillation require higher percentages of glycol in collected
stormwater to become economical. Therefore, the cost model is designed to provide costs for
recycle and recovery technologies only when the collection technology will provide the
appropriate glycol concentration in the collected stormwater. Likewise, technologies such as the
AFB can operate over a wide range of glycol concentrations and therefore can be applied to all
types of collection alternatives.
July 2009 11-29
-------
Technical Development Document for Proposed Effluent
Limitation Guidelines and Standards for the Airport Deicing Category
11. Technology Costs
Does Airport collect more than
60% of applied ADF?
Yes
No costs
No
Is Airport located in a warm
climate and have small annual
ADF use?
Yes
Apply costs for contractor
recovery and disposal
No
Does Airport collect more than
40% but less than 60% of applied
ADF?
Yes
Cost for centralized deicing
pads
No
Does Airport collect more than
20% but less than 40% of applied
ADF?
Yes
Cost for centralized deicing
pads and block and pump
with GRVs
No
Does Airport collect less than 20%
of applied ADF?
Yes
Cost for GRVs, block and pump
with GRVs and centralized deicing
pads
Figure 11-1. Diagram Showing the Steps for Developing Collection System Costs by the
Airport Deicing Cost Model
July 2009
11-30
-------
Technical Development Document for Proposed Effluent
Limitation Guidelines and Standards for the Airport Deicing Category
11. Technology Costs
Aircraft Deicing
Fluid Application
0.5-3% Glycol
1.5 to 10% Glycol
GRVs Only (20% Capture)
or
Block and Pump with GRVs
(40% Capture)
Install Centralized
Deicing Pads
AFB
Treatment and
Direct Discharge
Aerated
Ponds and Direct
Discharge
MVE/Distillation
Glycol Recovery
Figure 11-2. Diagram Showing the Alternatives for Developing Treatment System Costs by
the Airport Deicing Cost Model
The Cost Model provides output costs in Microsoft Excel for each selected collection and
treatment alternative. The outputs include both installed capital cost and annual operating and
maintenance costs. Cost model outputs for each airport that is currently not achieving 60 percent
collection and control of spent ADF are then used to calculate annualized costs by airport.
Section 11.3.4 provides more detail regarding cost annualization. For those airports, which EPA
estimates are achieving greater than 60 percent collection and control of applied ADF, the cost
model output is $0, indicating the airport will require no incremental cost to comply with the
proposed rule.
11.3.3.2 Airport Deicing Cost Model Equations
As indicated previously, the Cost Model uses capital and annual cost equations in
combination with various input parameters to estimate costs for collection and treatment
alternatives at each airport. Table 11-15 describes each cost equation, the equation used by the
model, the input variable and its source, and any assumptions used by the model to estimate
costs.
July 2009
11-31
-------
Technical Development Document for Proposed Effluent
Limitation Guidelines and Standards for the Airport Deicing Category
11. Technology Costs
Table 11-15. Airport Deicing Cost Model Equations, Input Variables and Assumptions
Calculation Description
Equation
Variables and Sources
Assumptions
Estimates costs for airports to
conduct a one-time monitoring
program at each stormwater outfall
to determine COD and glycol
levels reaching surface water
based on current ADF collection
and control practices.
Monitoring Cost ($): Number of
Deicing Outfalls x Number Deicing
Days/yrx $3,911
Number of outfalls from the airport
questionnaire.
Number of deicing days per year for
each airport from airport questionnaire.
Labor costs assume airport personnel
will conduct wet-weather sampling. If an
outside consultant conducts sampling,
costs may be higher.
Costs assume samples will be analyzed
for BOD, COD, ammonia, and glycol. If
other pollutants are included, costs will
be higher.
Estimates costs for an airport to
contract a one-time engineering
assessment to determine applicable
collection and control
technologies. Assessment is based
on data collected during the
monitoring program.
Engineering Assessment Cost ($):
Number of Outfalls x $5,680/outfall
Number of outfalls from the airport
questionnaire.
Costs are based on a consultant labor rate
of$85/hr.
Labor hours for various tasks are
estimates based on professional
judgment.
to
Estimates the incremental cost for
airports to discharge additional
ADF-contaminated stormwater to
a POTW for aerobic biological
treatment. This equation is
applicable to those airports that
already discharge their ADF-
contaminated stormwater to a
POTW.
Incremental POTW Costs ($/yr):
(ADF Use x Target % Collect x l/%
glycol x $0.01/gal) - (ADF Use x
Current % Collect x i/% glycol x
$0.01/gal)
ADF use from airline detailed
questionnaire
Collection % from selected collection
technology
Glycol percentage from collection
technology (1.5% for GRVs only, 3%
for plug and pump with GRVs, and 5%
for deicing pads)
The concentration of glycol in collected
stormwater is expected to be 1.5% from
GRVs, 3% from plug and pump, and 5%
from deicing pads.
Assumes the POTW will accept
additional ADF-contaminated
stormwater.
Estimates the incremental cost for
airports to contract haul additional
ADF-contaminated stormwater to
a POTW for anaerobic digestion.
This equation is applicable to those
airports that currently contract haul
to a POTW for anaerobic
treatment.
Incremental POTW Contract Haul
Cost ($/yr): (ADF Use x Target %
Collect x l/% glycol x $0.16/gal) -
(ADF Use x Current % Collect x l/%
glycol x$0.16/gal)
ADF use from airline detailed
questionnaire
Collection % from selected collection
technology
Glycol percentage from collection
technology (1.5% for GRVs only, 3%
for plug and pump with GRVs, and 5%
for deicing pads)
The concentration of glycol in collected
stormwater is expected to be 1.5% from
GRVs, 3% from plug and pump, and 5%
from deicing pads.
Assumes the POTW will accept
additional ADF-contaminated
stormwater for anaerobic treatment.
July 2009
-------
Technical Development Document for Proposed Effluent
Limitation Guidelines and Standards for the Airport Deicing Category
11. Technology Costs
Table 11-15 (Continued)
Calculation Description
Equation
Variables and Sources
Assumptions
Estimates the incremental cost for
airports to contract haul additional
ADF-contaminated stormwater to
an off-site glycol recovery/recycle
facility. This equation is applicable
to those airports that currently
contract haul to an off-site glycol
recovery /recycle facility.
Incremental Contract Haul and
Recycle Cost ($/yr): (ADF Use x
Target % Collect x $4.88) - (ADF Use
x Current % Collect x $4.88)
ADF use from airline detailed
questionnaire
Collection % from selected collection
technology
Assumes airport has the ability to
contract haul additional ADF-
contaminated stormwater and that the
recovery/recycle facility can accept
additional stormwater for glycol
recovery.
Estimates costs to purchase and
operate GRVs to collect 20% of
applied ADF.
GRV Capital Cost ($) = $108,000
outfall number
GRV Annual Cost($/yr) = $7,300 :
outfall number
Number of stormwater deicing outfalls
from the airport questionnaire
The number of deicing outfalls would be
an approximation of the size of the
deicing area that relates to the number of
GRVs required.
Estimates costs to install and
operate a block-and pump
collection system with GRVs to
collect 40% of applied ADF.
Plug and pump Capital Cost ($) =
$263,400 x outfall number
Plug and pump Annual Cost ($/yr):
$260,000 x outfall number
Number of stormwater deicing outfalls
from the airport questionnaire
The number of deicing outfalls would be
an approximation of the size of the
deicing area that relates to the number of
GRVs required.
Estimates costs to install and
operate a deicing pads collection
system with GRVs.
Deicing Pad Capital Cost ($) =
$314.56 x departures
Deicing Pad Annual Cost ($/yr) =
$8.87 x departures
Annual aircraft departures from Bureau
of Transportation Statistics
The number of deicing pads and their
size is directly related to expected airport
ground traffic during the deicing season.
Estimated cost to install and
operate a UF/RO glycol recovery
system.
UF/RO Capital Cost ($) = ADF Use
Collect % x l/% glycol x 1/150 x
$205.33
UF/RO Annual Cost ($/yr) = ADF
Use x Collect % x l/% glycol x
$0.107
ADF use from airline detailed
questionnaire
Collection % from selected collection
technology
Glycol percentage from collection
technology (3% for plug and pump and
5% for deicing pads)
The concentration of glycol in collected
stormwater is expected to be 3% from
plug and pump and 5% from deicing
pads.
The system is expected to operate for
150 days per year according to response
to the airport questionnaire
July 2009
-------
Technical Development Document for Proposed Effluent
Limitation Guidelines and Standards for the Airport Deicing Category
11. Technology Costs
Table 11-15 (Continued)
Calculation Description
Equation
Variables and Sources
Assumptions
Estimates cost to install and
operate a MVR/distillation glycol
recovery system.
MVR/Distillation Capital Cost ($) =
ADF Use x Collect % x l/ % glycol x
$4.08
MVR/Distillation Annual Cost ($/yr)
= ADF Use x Collect % x l/ % glycol
x $0.05
ADF use from airline detailed
questionnaire
Collection % from selected collection
technology
Glycol percentage from collection
technology (3% for plug and pump and
5% for deicing pads)
The concentration of glycol in collected
stormwater is expected to be 3% from
plug and pump and 5% from deicing
padsl.
Estimates costs to install an AFB
bioreactor treatment system. AFB
system costs include a holding
tank to equalize both flow and
concentration.
AFB Capital Cost ($) = ADF Use x
14.38 x Collect % x 1/194 x $1,659
AFB Annual Cost ($/yr) = ADF Use
14.38 x Collect % x 1/194 x $77.72
ADF use from airline detailed
questionnaire
Collection % from selected collection
technology
Ultimate COD is 14.38 Ibs COD/gal
Type I ADF.
Operating period is 194 days per year
from airport questionnaire.
Estimated costs to install a
nonaerated biological treatment
pond.
Pond volume = Runway Area/43,560
x 1.8 x 169,082
Pond Capital Cost ($) = Pond volume
x $0.53
Pond Annual Cost ($/yr) = Pond
volume x $0.0003
Runway area from FAA National Flight
Data Center
Conversion factor from square feet to
acres is 43,560.
Gate/apron collection area is 1.8 x
runway area (acres).
Volume of pond is 169,082 gallons per
acre of collection area.
Estimates costs to install an
aerated biological treatment pond.
Pond volume = Runway Area/43,560
x 1.8 x 169,082
Aerated Pond Capital Cost ($) = Pond
volume x $0.53 + ADF Use x $1.83
Aerated Pond Annual Cost ($/yr) =
ADF Use x $0.56/gal
Runway area from FAA National Flight
Data Center
ADF use from airline detailed
questionnaire
Conversion factor from square feet to
acres is 43,560.
Gate/apron collection area is 1.8 x
runway area (acres).
Volume of pond is 169,082 gallons per
acre of collection area.
Estimated costs to install storage
tanks to contain collected ADF-
contaminated stormwater
Storage Tank Capital Cost ($) =
Departures x 24 x $1.67
Storage Tank Annual Cost ($/yr)
Departures x $1.79
Annual aircraft departures from Bureau
of Transportation Statistics
Storage tank volume requirement is 24
gal/departure/yr.
July 2009
-------
Technical Development Document for Proposed Effluent
Limitation Guidelines and Standards for the Airport Deicing Category
11. Technology Costs
Table 11-15 (Continued)
Calculation Description
Equation
Variables and Sources
Assumptions
Estimated costs to install
additional stormwater piping to
convey ADF-contaminated water
from collection to treatment.
Piping Capital Cost ($) = $502,000
Piping System Labor Cost ($/yr) =
Number deicing days/yr x 24 x $32
Piping System Electrical Cost ($/yr) =
40 x 0.7456 x 24 x Number deicing
days/yr x $0.07
Piping System Annual Cost ($/yr) =
Labor Cost + Electrical Cost
None.
Deicing days/yr from airport
questionnaire.
Based on 1,000 linear feet of 12"
diameter subsurface piping.
Motor size for pumps in collection
system assumed to be 40 HP.
Conversion factor for horsepower to kW
is 0.7456.
Operators maintain collection systems 24
hrs per day during deicing event.
Labor rate of $32/hr from airport
questionnaire.
Electrical rate of $0.07 kWh from airport
questionnaire.
July 2009
-------
Technical Development Document for Proposed Effluent 11. Technology Costs
Limitation Guidelines and Standards for the Airport Deicing Category
The equations in Table 11-15 are based on an airport collecting ADF-contaminated
stormwater from areas within the airport expected to have high concentrations (e.g., greater than
0.5 percent) glycol. However, EPA also considered the cost impact of collecting stormwater
from all areas within an airport that could potentially have airfield and ADF contamination.
Those areas include runways and taxiways, grass areas in and around taxiways and runways, and
areas at the ends of runways where ADF may be lost from the aircraft during take-off. To
estimate one-time costs for an "all-flow scenario," EPA developed equations that would be used
by the Cost Model to calculate total precipitation volume and estimate collection and treatment
costs.
To estimate the total airport areas where potential airfield and ADF contamination may
occur, EPA developed a relationship between total airport runway area relative to grass areas,
apron and gate areas, and taxiway areas. See the memorandum entitled Methodology to Estimate
the Total Volume of Airport Stormwater Impacted by ADF (DCN AD00914) available in the
Airport Deicing Category administrative record. Airport runway area is published for all airports
and therefore this variable could be used to estimate all other areas at airports. First, EPA
developed a relationship between taxiway length and width to associated runway length and
width, since aircraft use taxiways to either enter or exit a runway, and taxiways typically parallel
each runway. EPA's analysis found that for the six major airports studied, taxiway areas are
approximately 1.4 times larger than the area of the associated runway. The taxiways' additional
area is likely associated with the number of turn-offs and connectors that allow planes to access
runways at numerous points throughout their entire length.
To estimate the airport gate/apron areas covered by concrete relative to the runway area,
EPA conducted a Google Earth (Google, 2007) analysis of the same six major airports. The
results showed that, on average, the paved gate/apron areas are approximately 1.8 times larger
than the total runway area. To estimate the grass areas surrounding gate/apron areas, taxi ways,
and runways where airfield deicers may be lost and aircraft deicing fluid may fall, EPA
performed a gross estimation again using Google Earth. Because of the uncertainty of the
amount of ADF that may drip or drain to grass areas during aircraft movement, EPA decided to
include all grass areas between taxiways and runways plus a 20-foot grass strip running along
side each runway. Based on this assumption, EPA estimated that the ratio of grass area to runway
area is approximately 2. Combining precipitation data for each airport (NOAA) with the areas
where airfield deicers and ADF may have been lost to either paved or grass areas allowed the
Cost Model to calculate the total annual volume of all stormwater that may have been
contaminated. Applying the total volume of potentially contaminated stormwater to the cost
equations in Table 11-15 calculated industry-wide capital costs approaching 1 trillion dollars and
annual O&M costs of over 1 billion dollars. Based on this one-time analysis, EPA decided not to
include collection and treatment of airfield-related stormwater in the proposed regulatory
options.
11.3.3.3 Example Cost Calculations
The following examples show how the Cost Model used the equations in Table 11-15
along with the airport questionnaire data to estimate costs for various airports. The examples
below do not represent actual airports, but are instead designed to show the capabilities of the
Cost Model.
July 2009 11-36
-------
Technical Development Document for Proposed Effluent 11. Technology Costs
Limitation Guidelines and Standards for the Airport Deicing Category
Example 1
Airport A has six deicing outfalls and currently has no collection or control equipment in
place for ADF-contaminated stormwater. The airport has 85,000 jet aircraft departures per year
and uses approximately 490,000 gallons per year of Type 1 and Type IV ADF. The airport has
23 deicing days per year spanning a five-month period. It does not discharge to a POTW or
contract haul ADF to an off-site recovery/recycle facility. Target collection of applied ADF is 60
percent and the airport decides to install and operate an anaerobic fluid bed treatment system.
One-Time Monitoring and Engineering Cost
Monitoring Cost = 6 Outfalls x 23 Days/yr x $3,911 = $540,000
Engineering Cost = 6 Outfalls x $5,680/outfall = $34,000
Deicing Pad Cost to Achieve 60% Collection Efficiency
Deicing Pad Capital Cost = 85,000 departures/yr x $314.56 = $26.7 million
Deicing Pad Annual Cost ($/yr) = 85,000 departures/yr x $8.87 = $754,000
Anaerobic Fluid Bed Biological Treatment System
AFB Capital Cost ($) = 490,000 gal/yr x 14.38 Ibs COD/gal ADF x 0.6 x 1/194 days/yr x $1,659
= $36.1 million
AFB Annual Cost ($/yr) = 490,000 gal/yr x 14.38 Ib COD/gal ADF x 0.6 x 1/194 x $77.72
= $1.7 million
Total Collection and Treatment System Capital Cost: $63.3 million
Total Collection and Treatment System Annual Cost: $2.4 million/yr
Example 2
Airport B has 3 deicing outfalls and currently uses GRVs to collect approximately 20
percent of its applied ADF. The concentration of glycol in the collected stormwater is
approximately 1.5 percent. The airport has conducted a previous monitoring program at each
outfall and developed a Storm Water Pollution Prevention Plan (SWPPP). The airport has
117,400 jet aircraft departures per year and uses approximately 417,000 gallons per year of Type
1 and Type IV ADF. It has 14 deicing days per year spanning a four-month period. The airport
currently discharges collected deicing stormwater to a POTW. Target collection of applied ADF
is 60 percent using deicing pads. The expected concentration of glycol in the ADF-contaminated
stormwater collected from the pads is 5 percent. The airport has decided to install and operate a
UF/RO system for on-site glycol recovery. Residual effluent from the RO will contain COD that
will continue to be discharged to the POTW at no additional incremental cost.
July 2009 11-37
-------
Technical Development Document for Proposed Effluent 11. Technology Costs
Limitation Guidelines and Standards for the Airport Deicing Category
Deicing Pad Cost to Achieve 60% Collection Efficiency
Deicing Pad Capital Cost = 117,400 departures/yr x $314.56 = $36.9 million
Deicing Pad Annual Cost ($/yr) = 117,400 departures/yr x $8.87 = $1 million/yr
Collection Tank and Piping for ADF Contaminated Stormwater
Storage Tank Capital Cost ($) = 117,400 x 24 x $1.67 = $4.7 million
Storage Tank Annual Cost ($/yr) = 117,400 x $1.79 = $210,000/yr
Piping and Lift Station Capital Cost ($) = $502,000
Piping and Lift Station Annual Cost ($/yr)
Labor Cost ($/yr) = 14 days/yr x 24 x $32 = $10,800/yr
Electrical Cost ($/yr) = 40 x 0.7456 x 24 x 14 days/yr x $0.07 = $700/yr
UF/RO Recovery and Recycle Treatment System
UF/RO Capital Cost ($) = 417,000 gal/yr x 0.6 x 1/0.05 x l/(4x30) x $205.33 = $8.6 million
UF/RO Annual Cost ($/yr) = 417,000 gal/yr x 0.6 x 1/0.05 x $0.107/gal = $535,000/yr
Total Collection and Treatment System Capital Cost: $50.7 million
Total Collection and Treatment System Annual Cost: $1.8 million/yr
11.3.4 Annualized Costs for ADF Collection and Treatment Alternatives
The first step in projecting the economic and financial impacts of proposed effluent
limitation guidelines and standards (ELGs) on airports is cost annualization. For each airport,
EPA used the capital and operating and maintenance costs from the Cost Model for each ADF
target removal percentage over 20 years, discounted future costs using an airport-specific
opportunity cost of capital, and annualized those costs to represent 20 equal annual cost
payments incurred by the airport. Because the expected service life of each technology basis
differs, the capital cost estimates incorporate costs to replace GRVs and block-and-pump
technologies; for the purposes of projecting capital costs, EPA expects both these technologies
will require replacement after 10 years, while a deicing pad is expected to last 20 years before
needing to be replaced.
EPA assumed airports will issue tax-exempt, fixed coupon rate serial General Airport
Revenue Bonds (GARBs) to fund capital expenditures. The Agency also assumed airports will
issue bonds equivalent to the net present value of capital costs plus 3 percent to account for bond
issuance costs. The Cost Model annualized capital costs using each airport's nominal bond rate
for its most recent GARB issue. This was converted to a real rate using an average annual
inflation rate of 2.3 percent over the last five years. The average nominal discount rate for costed
airports was 5.25 percent, which is equivalent to 2.87 percent after accounting for inflation.
Costs were then annualized over 20 years.
July 2009 11-38
-------
Technical Development Document for Proposed Effluent
Limitation Guidelines and Standards for the Airport Deicing Category
11. Technology Costs
11.4
Airfield Deicing Costs
While the proposed regulation requirement is a numeric limit for ammonia, it is
anticipated that the means of compliance will be product substitution. This subsection includes
EPA's cost evaluation for changing from urea to potassium acetate for pavement deicing.
Information collected by EPA as part of this proposed rulemaking effort indicates that using urea
as an airfield deicing chemical is being phased out due to concerns with its environmental
impacts and the availability of less harmful alternatives. Responses to EPA's airport
questionnaire indicated that potassium acetate was by far the predominant airfield deicing
chemical in use from 2002 to 2005, representing about 80 percent of all airfield deicing chemical
use; however, approximately 35 of the surveyed airports continue to use urea for airfield deicing.
11.4.1
Urea and Potassium Acetate Chemical Costs and Application Rates
To determine cost differences between urea and potassium acetate, EPA utilized data
from the airport questionnaire. EPA also obtained unit costs for potassium acetate and urea from
eight airports that use both for airfield deicing. Table 11-16 shows the average costs for urea and
potassium acetate based on the information provided by the eight airports from 2002 to 2005.
Table 11-16. Average Cost for Urea and Potassium Acetate, 2002-2005
Year
2002
2003
2004
2005
Average Urea Cost
$268.17/ton
$280.57/ton
$297.90/ton
$300.2 I/ton
Average Liquid Potassium Acetate Cost
$2.8 I/gallon
$2.86/gallon
$2.86/gallon
$2.92/gallon
Potassium acetate is applied at different rates depending on the weather conditions and the
thickness of the ice layer at the time of application. Table 11-17 shows the typical deicing, anti-
icing and prewetting application rates for four commercial potassium acetate runway deicers.
Table 11-17. Typical Application Rates for Potassium Acetate
Brand Name
Safeway® KA Runway
Deicing Fluid :
CryotechE-36®LRD2'3
IceClearRDF4
PEAK® PA 5
Deicing Application Rates
1 gal/1000 ft2
1 gal/1000 ft2 (thin ice) and
3 gal/1000 ft2 (2.5cm thick
ice)
1 gal/1000 ft2 (thin ice) and
3 gal/1000 ft2 (lin. thick ice)
1 gal/1000 ft2
Anti-Icing
Application Rates
0.4gal/l,000 ft2
0.5 gal/1,000 ft2
0.5 gal/1,000 ft2
0.4 gal/1,000 ft2
Prewetting Application Rates
70% solid and 30% liquid
85-95% solid and 5-15% liquid,
or 130g/kg of solid deicer,
1.25gal/1001bs. solid deicer
70% solid and 30% liquid
1 http://www.theblackfootcompany.com/new/de-ice/ka.htm.
2http://www.proviron.com/Deice/E36/e36_tech.html.
3 http://www.p2pays.org/ref/19/18054.htm.
4 http://www.orisonmarketing.com/deicers/RDFl/RDFl.html.
5http://www.oldworldind.com/chemicals/specs/pk_pa_spec.pdf.
July 2009
11-39
-------
Technical Development Document for Proposed Effluent
Limitation Guidelines and Standards for the Airport Deicing Category
11. Technology Costs
Although EPA could not obtain actual application rates for potassium acetate at
individual airports, EPA did obtain airfield application rates for sodium acetate; therefore, the
Agency used sodium acetate rates as a surrogate to estimate potassium acetate application rates.
The amount of sodium acetate required to provide the same protection time as urea is between 66
and 70 percent (Transport Canada, 1998). Table 11-18 lists typical application rates for Cryotech
NAAC®, a commercial sodium acetate deicer, as well as the calculated application rates for urea.
Table 11-18. Application Rates for Sodium Acetate and Urea
Sodium Acetate, Cryotech NAAC®, Application Rate 1
Near 32° F on thin ice = 3 - 5 lbs/1000 ft2
Less than 10° F on 1 inch ice =10-25 lbs/1000 ft2
Calculated Urea Application Rate
Near 32° F on thin ice = 4.3 - 7.1 lbs/1000 ft2
Less than 10° F 2 on 1 inch ice = 14.3 - 35.7 lbs/1000
ft2
1 http://www.peterschemical.com/sodium-acetate/.
2Urea loses its effect at temperatures below 20°F.
Using the information in Table 11-18, EPA estimated the application costs for urea and
potassium acetate based on the 2005 average unit costs ($/l,000 ft2), as shown in Table 11-19.
Table 11-19. Cost for Application of Urea and Potassium Acetate, per 1000 Square Feet
Chemical
Urea
Potassium Acetate
Deicing Application Cost
(per 1000 ft2)
$0.65 - $1.07, Near 32° F on thin ice
$2.92 (thin ice) and $8.76 (thick ice)
Anti-Icing Application Cost
(per 1000 ft2)
$1.17 -$1.46
11.4.2
Cost Impact of Discontinuing Urea Airfield Deicing
The national average use of urea, based on responses to the airport questionnaire for the
three deicing seasons between 2002 and 2005, was 7,075,865 pounds per year, costing an
estimated $l,064,998/yr. Using the available range of application rates (small coverage area and
large coverage area for the same amount of urea) and statistical airport weighting values, EPA
estimated the chemical costs for all airports currently using urea to change to potassium acetate.
Table 11-20 presents the data to determine the additional chemical cost associated with a
switch to potassium acetate. The estimated increase in cost for these airports to change from urea
to potassium acetate for airfield deicing is $3.8 million per year (2006 $). The costs in
Table 11-20 are not weighted to the entire airport industry and do not reflect capital costs for
new equipment that may be required for airports to change from a solid form of urea application
to a liquid form of potassium acetate.
11.4.3
Urea Monitoring Costs
Although EPA anticipates that airports subject to the proposed rule will use alternatives
to urea for pavement deicing, the Agency has developed numerical effluent limitations for
ammonia as a compliance alternative. Demonstrating compliance with this limitation would
require total nitrogen monitoring for those airports still using urea for airfield deicing. EPA
July 2009
11-40
-------
Technical Development Document for Proposed Effluent 11. Technology Costs
Limitation Guidelines and Standards for the Airport Deicing Category
calculated yearly monitoring costs for the 35 airports that reported urea use during the 2005
deicing season. The Agency calculated total monitoring costs by combining the costs associated
with analytical sample testing and sample collection labor costs. In calculating analytical testing
costs, EPA assumed that every airport would need to test one sample per day, per airport outfall
that receives deicing stormwater. EPA also assumed that testing would occur five days a week
for 26 weeks, totaling 130 samples per outfall per year at an approximate cost of $24.00 per
sample. In calculating labor costs, EPA assumed that each sample collected would take one hour
to retrieve and process at a cost of $33.00 per hour. Table 11-21 presents the total costs
associated with monitoring urea at each airport, broken down by analytical testing and labor
costs for each airport. EPA predicts total annual urea monitoring costs for these airports could
reach approximately $1.1 million per year (2006 $).
July 2009 11-41
-------
Technical Development Document for Proposed Effluent
Limitation Guidelines and Standards for the Airport Deicing Category
11. Technology Costs
Table 11-20. Incremental Costs for Airports to Change from Urea to Potassium Acetate for
Airfield Deicing
Airport Name
Yeager
Fairbanks International
Ted Stevens Anchorage International
Wiley Post- Will Rogers Mem
Tri - State/Milton J FEPAuson Field
Austin Straubel International
Piedmont Triad International
Fort Wayne International
Glacier Park International
Ralph Wien Memorial
General Edward Lawrence Logan
International
Salt Lake City International
South Bend Regional
Manchester
Waterloo Municipal
Boise Air Terminal/Gowen Field
Juneau International
Spokane International
Aniak
Deadhorse
Stewart International
Reno/Tahoe International
Bethel
Charlotte/Douglas International
Bradley International
Ronald Reagan Washington National
Central Wisconsin
Northwest Arkansas Regional
Raleigh - Durham International
Greater Rockford
Akron - Canton Regional
Estimated Actual
Annual Urea Cost
($/yr 2006)
$4,000
$57,700
$251,500
$3,000
$9,400
$6,300
$14,800
$40,300
$50
$1,500
$860
$220,900
$4,900
$3,400
$900
$61,100
$76,500
$96,100
$400
$3,000
$22,900
$1,100
$9,900
$35,100
$2,500
$9,400
$12,700
$4,100
$13,400
$98,600
$3,200
Predicted Potassium
Acetate Cost
($/yr 2006)
$18,100
$260,300
$1,134,500
$13,600
$42,600
$28,200
$67,000
$182,000
$230
$6,800
$3,900
$996,400
$22,000
$15,300
$4,100
$275,500
$345,000
$433,700
$1,600
$13,600
$103,100
$4,900
$44,800
$158,200
$11,400
$42,600
$57,300
$18,300
$60,700
$444,900
$14,500
Total Incremental Cost ($/yr)
Incremental Cost to
Change from Urea to
Potassium Acetate
($/yr 2006)
$14,100
$202,600
$883,100
$10,600
$33,100
$22,000
$52,200
$141,600
$180
$5,300
$3,000
$775,600
$17,100
$11,900
$3,200
$214,400
$268,500
$337,600
$1,300
$10,600
$80,200
$3,800
$34,900
$123,100
$8,900
$33,100
$44,600
$14,300
$47,200
$346,200
$11,300
$3,755,600
July 2009
11-42
-------
Technical Development Document for Proposed Effluent
Limitation Guidelines and Standards for the Airport Deicing Category
11. Technology Costs
Table 11-21. Estimated Annual Costs for Airports to Conduct Effluent Monitoring
Program For Urea Airfield Deicing
Airport Name
Yeager
Fairbanks International
Ted Stevens Anchorage International
Wiley Post- Will Rogers Mem
Tri - State/Milton J FEPAuson Field
Austin Straubel International
Piedmont Triad International
Fort Wayne International
Glacier Park International
Ralph Wien Memorial
General Edward Lawrence Logan
International
Salt Lake City International
City of Colorado Springs Municipal
South Bend Regional
Manchester
Waterloo Municipal
Boise Air Terminal/Gowen Field
Rochester International
Juneau International
Spokane International
Aniak
Deadhorse
Stewart International
Reno/Tahoe International
Bethel
Charlotte/Douglas International
Bradley International
San Antonio International
Wilkes - Barre/Scranton International
Ronald Reagan Washington National
Central Wisconsin
Northwest Arkansas Regional
Raleigh - Durham International
Greater Rockford
Akron - Canton Regional
Monitoring
Analytical Costs
($/yr)
$37,440
$21,840
$15,600
NA
$3,120
$6,240
$74,880
$24,960
$3,120
$3,120
$9,360
$15,600
$3,120
NA
$3,120
$3,120
NA
$9,360
$15,600
NA
$9,360
NA
NA
$6,240
$15,600
$6,240
$28,080
$46,800
$6,240
$12,480
$6,240
$9,360
$31,200
$3,120
$28,080
Monitoring Labor
Costs
($/yr)
$51,480
$30,030
$21,450
NA
$4,290
$8,580
$102,960
$34,320
$4,290
$4,290
$12,870
$21,450
$4,290
NA
$4,290
$4,290
NA
$12,870
$21,450
NA
$12,870
NA
NA
$8,580
$21,450
$8,580
$38,610
$64,350
$8,580
$17,160
$8,580
$12,870
$42,900
$4,290
$38,610
Total Annual Monitoring Cost ($/yr)
Total Urea
Monitoring Costs
($/yr)
$88,920
$51,870
$37,050
NA
$7,410
$14,820
$177,840
$59,280
$7,410
$7,410
$22,230
$37,050
$7,410
NA
$7,410
$7,410
NA
$22,230
$37,050
NA
$22,230
NA
NA
$14,820
$37,050
$14,820
$66,690
$111,150
$14,820
$29,640
$14,820
$22,230
$74,100
$7,410
$66,690
$1,089,300
NA - Airports with "NA" reported urea use but reported no deicing impacted outfalls.
July 2009 11-43
-------
Technical Development Document for Proposed Effluent 11. Technology Costs
Limitation Guidelines and Standards for the Airport Deicing Category
11.5 References
ERG. 2008a. Memorandum from Cortney Itle and Robyn Reid (ERG) to Airport Deicing
Administrative Record: Estimated Annual Costs for Airports with Limited ADF Use. (February).
DCNAD00913.
ERG. 2008b. Memorandum from Steve Strackbein and Mary Willett (ERG) to Brian D'Amico
and Eric Strassler (EPA): Cost Comparison of Potassium Acetate and Urea Airfield Deicers.
(March 17). DCN AD00843.
ERG. 2008c. Memorandum from Cortney Itle and Robyn Reid (ERG) to Airport Deicing
Administrative Record: Estimated Costs for Initial Monitoring, Engineering Assessment and
SWPPP Updates for Airports. (February). DCN AD00910.
ERG. 2008d. Memorandum from Cortney Itle and Robyn Reid (ERG) to Airport Deicing
Administrative Record: Estimated Capital and O&M Costs for Collection Ponds and Stormwater
Piping. (February). DCN AD00912.
ERG. 2007. Memorandum from Juliana Stroup and Mary Willett (ERG) to Brian D'Amico
(EPA): Aircraft Deicing Stormwater Control Technologies and Their Removal Efficiencies.
(December 17). DCN AD00855.
Switzenbaum, M., et al. 1999. Workshop: Best Management Practices for Airport Deicing
Stormwater. (July 28). DCN AD00893.
USEPA. 2006a. Airport Deicing Questionnaire Database. U.S. Environmental Protection
Agency/Office of Water. Washington, D.C. DCN AD00927.
USEPA. 2006b. Airline Deicing Questionnaire Database. U.S. Environmental Protection
Agency/Office of Water. Washington, D.C. DCN AD00938.
USDOT. 2006. U.S. Department of Transportation, Bureau of Transportation Statistics, T-100
Segment Database. DCN AD00928.
NOAA. U.S. Department of Commerce, National Oceanic and Atmospheric Administration
(NOAA) National Climate Data Center. DCN AD00929.
FAA. U.S. Federal Aviation Administration (FAA) National Flight Data Center. DCN AD00930.
USEPA. 2008a. Development of Statistical Weighting Factors for Airport Deicing Category.
U.S. Environmental Protection Agency/Office of Water. Washington, D.C. DCN AD00931.
USEPA. 2006c. Engineering Site Visit Report for General Mitchell International Airport,
Milwaukee, WI. U.S. Environmental Protection Agency/Office of Water. Washington, D.C. DCN
AD00772.
July 2009 11-44
-------
Technical Development Document for Proposed Effluent 11. Technology Costs
Limitation Guidelines and Standards for the Airport Deicing Category
USEPA. 2006d. Engineering Site Visit Report for Minneapolis-St. Paul International Airport,
Minneapolis, MN. U.S. Environmental Protection Agency/Office of Water. Washington, D.C.
DCNAD00793.
USEPA. 2006e. Engineering Site Visit Report for Denver International Airport, Denver, CO.
U.S. Environmental Protection Agency/Office of Water. Washington, D.C. DCN AD00779.
ERG. 2007a. Personal communication (email) between Mary Willett (ERG) and Kevin Gurchak
(Pittsburgh International Airport). (December 6). DCN AD00873.
McQueen, R., et al. Development of a Deicer Management System for Akron-Canton Airport.
DCN AD00894.
ERG. 2007b. Personal communication (email) between Mary Willett (ERG) and Roy Fuhrmann,
Minneapolis-St. Paul International Airport. (May 3). DCN AD00872.
USEPA. 2006f. Sampling Episode Report for Pittsburgh International Airport. U.S.
Environmental Protection Agency/Office of Water. Washington, D.C. DCN AD00841.
USEPA. 2006g. Sampling Episode Report for Denver International Airport. U.S. Environmental
Protection Agency/Office of Water. Washington, D.C. DCN AD00840.
USEPA. 2006h. Sampling Episode Report for Albany International Airport. U.S. Environmental
Protection Agency/Office of Water. Washington, D.C. DCN AD00842.
USEPA. 2008b. Sampling Episode Report for Rockford International Airport. U.S.
Environmental Protection Agency/Office of Water. Washington, D.C. DCN AD00839.
ERG. 2007c. Personal communication between Mark Briggs (ERG) and New Logic Research,
Inc., Emmeryville, CA. (June 12). DCN AD00895.
TRB Airport Cooperative Research Program, 2009. Deicing Planning Guidelines and Practices
for Stormwater Management Systems ACRP Report No. 14. DCN AD01191.
New Logic Research. 2001. VSEP Filtration for Glycol Recovery, Application Note. DCN
AD00897.
ERG. 2007d. Personal communication (email) between Mary Willett (ERG) and Donald
Chapman (Cincinnati/Northern Kentucky International Airport). (May 4). DCN AD00867.
ERG. 2007e. Personal communication (email) between Mary Willett (ERG) and Keith Pass
(Denver International Airport). (July 12). DCN AD00868.
ERG. 2007f. Personal communication between Mark Briggs (ERG) and Roger Lamdola (Inland
Technologies - Canada). (March 19). DCN AD00881.
July 2009 11-45
-------
Technical Development Document for Proposed Effluent 11. Technology Costs
Limitation Guidelines and Standards for the Airport Deicing Category
USDoD. 2003. U.S. Department of Defense. ESTCP, Mineralization of TNT, RDX, and By-
Products in an Anaerobic Granular Activated Carbon Fluidized Bed Reactor, (April). DCN
AD00898.
ERG. 2007g. Personal communication (email) between Mary Willett (ERG) and Mark Sober
(Albany International Airport). (December 12). DCN AD00865 and AD00866.
ERG. 2007h. Personal communication (email) between Mary Willett (ERG) and R. McQueen
(Akron-Canton International Airport). (May 3). DCN AD00863 and AD00864.
ERG. 2007L Personal communication between Mark Briggs (ERG) and Aqua-Aerobic Systems,
Inc., Rockford, IL. (August 31). DCN AD00932.
RS. 2002. RSMeans. Means Heavy Construction Cost Data, 16th Annual Edition.
http://www.rsmeans.com DCN AD00899.
USDoD. 2001. U.S. Department of Defense. Military Construction Program 1391. DCN
AD00900.
Perry. 1984. Perry's Chemical Engineers Handbook, Sixth Edition, http://www.knovel.com
DCNAD00901.
USEPA. 2006L Engineering Site Visit Report for Greater Cincinnati International Airport. U.S.
Environmental Protection Agency/Office of Water. Washington, D.C. DCN AD00767.
USEPA. 2006J. Engineering Site Visit Report for Portland International Airport. DCN AD00775.
Google. 2007. Google Earth Geographical Mapping System.
Transport Canada. 1998. Transportation Development Center, "Laboratory Testing of Tire
Friction Under Winter Conditions (TP 13392E)." (May). DCN AD00902.
July 2009 11-46
-------
Technical Development Document for Proposed Effluent 12. Non-Water Quality Impacts
Limitation Guidelines and Standards for the Airport Deicing Category
12. NON- WATER QUALITY IMPACTS
Sections 304(b) and 306 of the Clean Water Act require EPA to consider non-water-
quality environmental impacts (including energy requirements) associated with effluent
limitation guidelines and standards. To comply with these requirements, EPA considered the
potential impact of the collection and treatment technologies on energy consumption, air
emissions, and solid waste generation.
Section 12.1 discusses the energy requirements for implementing the proposed collection
and treatment technologies at in-scope airports (defined as commercial airports with 1,000 or
more jet departures per year). Sections 12.2 and 12.3 discuss the impact of these technologies on
air emissions and on wastewater treatment sludge generation, respectively. Section 12.4 presents
the references used in this section.
12.1 Energy Requirements
This subsection discusses the net energy requirements for the selected collection and
treatment scenarios evaluated by EPA for proposal. Net energy consumption considers electrical
requirements for pumping collected fluid from centralized deicing pads, and electrical
requirements for operating the anaerobic fluidized bed (AFB) bioreactors and the aerated ponds
and fuel requirements for glycol recovery vehicles (GRVs). EPA did not consider electrical
requirements for the ultrafiltration/reverse osmosis (UF/RO) and mechanical vapor
recompression (MVR)/distillation recovery and recycle technologies in this analysis since these
technologies were ultimately not selected as part of the proposed regulatory options. Detailed
calculations regarding net energy consumption for the collection and treatment technologies are
provided in a separate memorandum entitled Energy Requirements for ADF Contaminated
Stormwater Collection and Treatment Alternatives (DCN ADO 1 166).
To estimate incremental electrical requirements associated with pumping collected ADF
to either tanks or ponds, EPA assumed airports would continuously operate three 40-horsepower
(hp) electric motors during each deicing day. EPA also conservatively assumed that all airports
would use pumps rather than allow ADF-impacted Stormwater to flow by gravity to holding
tanks or ponds. To estimate electrical use by airport based on the number of deicing days per
year, EPA developed the following equation:
Pumping Electrical = 3 pumps x 40 -— x 0.7456 — x 24 — x SOFP (12-1)
pump hp day yr
where:
kW = Kilowatts; and
SOFP = Snow or other freezing precipitation.
Using the equation above and the SOFP days for those airports that EPA assumed would
install additional collection equipment, total incremental electrical usage associated with the
proposed rule would be approximately 1.2 million kilowatt hours per year (kWh/yr).
EPA developed another relationship between electrical use and chemical oxygen demand
(COD) removal by the AFB bioreactors based on information provided by Albany International
(ALB) airport. Using the information from ALB, EPA estimated the electrical requirement for
July 2009 12-1
-------
Technical Development Document for Proposed Effluent
Limitation Guidelines and Standards for the Airport Deicing Category
12. Non-Water Quality Impacts
COD removal as approximately 1.3 kWh/lb COD removed. Using this unit rate, EPA estimated
total electrical requirements to remove COD based on each of the ADF collection scenarios.
Table 12-1 lists the electrical requirements for the AFBs.
Table 12-1. Estimated Electrical Requirements for All AFB Treatment Systems
Regulatory Scenario
20% ADF Capture
40% ADF Capture
60% ADF Capture
Total COD Removal
(pounds/yr)
7,329,700
15,282,600
34,545,100
AFB Electrical
Requirements
(kWh/lb COD Removed)
1.3
1.3
1.3
Total Electrical
Requirement
(million kWh/yr)
9.5
19.9
44.9
The AFB treatment systems also generate biogas that can be used as a source of heat
when burned in facility boilers or when converted to electricity using technologies such as
microturbines or fuel cells. To estimate the potential electricity that could be generated if all
AFB treatment systems installed microturbines to generate electricity, EPA developed a
relationship between biogas generation and COD removal based on data provided by ALB
airport. EPA estimated the AFB reactors will generate approximately 8 cubic feet of biogas per
pound of COD removed, and the biogas is approximately 60 percent methane. Because one cubic
foot of methane provides 0.01 therms (1) and 1 therm is equivalent to 29.3 kWh when converted
to electrical energy (USC, 2007), EPA was able to predict the potential electrical energy
available from biogas generated by the AFBs treating ADF-contaminated stormwater. Table 12-2
presents electricity generation from biogas generated by the AFBs.
Table 12-2. Potential Electricity Generation from AFB Biogas Generation
Regulatory
Scenario
20% ADF Capture
40% ADF Capture
60% ADF Capture
Total COD
Removal
(pounds/yr)
7,329,700
15,282,600
34,545,100
Potential Biogas
Generation
(million ft3/yr) 1
57
119
161
Potential Methane
Generation
(million ft3/yr) 2
34
71
97
Potential Electrical
Generation
(million kWh/yr) 3
10
21
47
Calculation based on 8 cubic feet of biogas per Ib COD removed.
2 Assumes biogas is approximately 60% methane per Metcalf and Eddy Wastewater Engineering and Design (9).
3 Calculation based on 100 therms per cubic foot of methane and 29.3 kWh per therm.
The comparison of the potential electrical generation from converting biogas to
electricity to the electrical requirements shown in Table 12-1 indicates that AFB treatment of
ADF-contaminated stormwater could generate nearly the same amount of electricity that is
needed to operate the treatment systems.
Fuel use by GRVs collecting ADF-contaminated stormwater is another incremental
energy requirement for compliance with the proposed rule. To estimate incremental diesel fuel
use by GRVs, EPA obtained annual diesel fuel costs for GRVs from the Airport Deicing
Questionnaire (USEPA, 2006a) and then estimated diesel fuel use based on the unit cost for
diesel fuel (e.g., $/gal). According to questionnaire data, one airport spent $17,600 on diesel fuel
to operate GRVs for recovery of spent ADF and stormwater during the 2004-2005 deicing
July 2009 12-2
-------
Technical Development Document for Proposed Effluent 12. Non-Water Quality Impacts
Limitation Guidelines and Standards for the Airport Deicing Category
season. This airport collected approximately 20 percent of their applied ADF in stormwater in
the 2004/2005 deicing season. Based on an average diesel fuel cost of $2.07 per gallon during
the 2004/2005 deicing season (USDOE, 2006), EPA estimates this airport burned approximately
8,500 gallons of diesel fuel in GRVs. Based on annual ADF use and size, EPA estimated diesel
fuel use in GRVs to be 0.08 gal/gal ADF applied. Using this relationship, EPA estimated total
incremental No. 2 diesel fuel consumption at all in-scope airports installing additional collection
equipment to be 604,000 gallons per year. This volume is a conservative estimate since it is
based on an airport that currently use only GRVs to collect ADF-contaminated stormwater (e.g.,
20% ADF collection). If airports installs deicing pads or utilizes plug and pump to collect ADF-
contaminated stormwater, GRV usage is expected to be less, and therefore diesel fuel use is also
expected to be less.
EPA compared incremental diesel fuel use by GRVs at all airports to diesel fuel use on a
national basis. According to the Energy Information Administration, approximately 25.4 million
gallons per day of No. 2 diesel fuel was consumed in the United States in 2005 (USDOE, 2006).
Total annual diesel fuel use by GRVs to collect ADF-contaminated stormwater at airports would
account for less than 0.005 percent of the daily diesel fuel use on a national level.
12.2 Air Emissions
Additional air emissions as a result of the proposed rule can be attributed to added diesel
fuel combustion by GRVs collecting ADF-contaminated stormwater, from additional jet engine
taxi time related to deicing pads, and from anaerobic treatment of ADF. Emissions from these
sources are discussed below.
Emissions From GRV Collection
As discussed in Section 12.1, EPA conservatively estimated that GRVs collecting ADF-
contaminated stormwater at airports will consume an additional 604,000 gallons per year of
No. 2 diesel fuel.. To estimate air emissions related to combustion of No. 2 diesel fuel in the
internal combustion engines on GRVs, EPA used published emission factors for internal
combustion engines (USEPA, 2006b). The Agency selected emission factors for gasoline and
diesel industrial engines rather than on-road mobile sources because the emission factors for the
industrial engines include equipment such as fork lifts and industrial sweepers and scrubbers
(USEPA, 2006b). To estimate emissions from the GRVs, EPA first converted the additional
604,000 gallons of diesel fuel to million British Thermal Units (MMBtu) and then applied the
appropriate emission factors (USEPA, 2006b. Table 12-3 shows the estimated increase in criteria
pollutant emissions associated with the use of GRVs. Additional details regarding emissions
from GRVs are contained in a memorandum titled "Air Emissions from Airport Deicing
Collection and Treatment Technologies" (DCN ADO 1165).
July 2009 12-3
-------
Technical Development Document for Proposed Effluent
Limitation Guidelines and Standards for the Airport Deicing Category
12. Non-Water Quality Impacts
Table 12-3. Estimated Incremental Criteria Pollutant Emissions from GRVs
Criteria Pollutant
Carbon Monoxide
Carbon Dioxide
Nitrogen Oxides
Sulfur Dioxide
PM10
Diesel Fuel
Consumption in
GRV Internal
Combustion Engine
(gal/yr)
604,000
604,000
604,000
604,000
604,000
Diesel Fuel
Consumption in GRV
Internal Combustion
Engine
(MMBtu/yr) 1
84,500
84,500
84,500
84,500
84,500
Emission Factor for
Diesel Fuel
Combustion in the
Internal Combustion
Engine 2
0.95
164
4.41
0.29
0.31
Estimated Annual
Emissions from
GRVs Burning Diesel
Fuel
(tons/yr)
40
6,900
186
12
13
Heat content of diesel fuel is approximately 140,000 Btu/gal per Perry's Chemical Engineers Handbook, 6th
Edition, Figure 9-4.
2 Emission factors from EPA Compilation of Emission Factors AP-42.
PM10 - Paniculate matter less than 10 um.
The annual emissions provided in Table 12-3 indicate that an additional 4,781 tons per
year of carbon dioxide will be emitted from GRVs combusting additional diesel fuel to comply
with the proposed rule. Carbon dioxide is the primary greenhouse gas attributed to climate
change; although 6,900 additional tons per year appears to be considerable, the amount is very
small relative to other sources. For example, in 2006, industrial facilities combusting fossil fuels
emitted 948 million tons of CC>2 equivalents. An additional 6,900 tons per year from GRVs is
less than a 0.001 percentage increase in the overall CC>2 emissions from all industrial sources
(USEPA, 2008).
Emissions From Transportation to Aircraft Deicing Pads
To estimate aircraft emissions associated with the additional time spent taxiing to and
from newly installed deicing pad and idling during deicing, EPA used the seven busiest airports
where deicing pads would likely be installed to comply with the proposed rule. Those airports
include Boston Logan, Cleveland-Hopkins, Newark Liberty, Washington Dulles, New York's
LaGuardia and Kennedy airports, and Chicago O'Hare. To estimate aircraft emissions for each
airport from transportation to newly installed deicing pads, input files such as departure
information, types of aircraft being deiced, and deicing days were compiled and applied to the
Emissions and Dispersion Modeling System (EDMS). EDMS is an emission estimating tool
developed by the U.S. Department of Transportation's Federal Aviation Administration (FAA)
(USDOT, 2008). This computer model integrates all airport emission sources, mobile and
stationary, into a single model. EDMS requires aircraft-specific activity data — specifically, the
make and model number of the aircraft used by the seven airports included in this assessment.
For this assessment, EPA obtained 2007 aircraft-specific activity data from the U.S. Department
of Transportation's Bureau of Transportation Statistics (BTS) T-100 data set for each of the
seven airports included in this study (USDOT, 2008).
However, the aircraft make and model data from the BTS do not match exactly with the
aircraft make and model available in the EDMS model, so EPA developed an aircraft crosswalk
to match the BTS data to the aircraft in the EDMS model. As shown in Table 12-4, the crosswalk
was able to match 89 percent of the aircraft to the EDMS model, Table 12-5 shows that the
matched aircraft accounted for 97 percent of the activity based on landing and take-off cycles
July 2009 12-4
-------
Technical Development Document for Proposed Effluent
Limitation Guidelines and Standards for the Airport Deicing Category
12. Non-Water Quality Impacts
(LTOs). Many of the aircraft that were not matched were helicopters and smaller planes that
probably do not account for a significant portion of deicing activities, and excluding them from
this assessment is probably warranted. After matching either the aircraft make or model, EPA
used either the associated default engine in EDMS or the most common engine in the model
runs.
Table 12-4. BTS to EDMS Matching Based on Aircraft
Airport
Boston Logan
Cleveland-Hopkins
Newark Liberty
Washington Dulles
New York Kennedy
New York LaGuardia
Chicago O'Hare
Totals
Total Aircraft
71
50
67
65
62
40
57
412
Matched Aircraft
64
46
55
58
53
39
53
368
Percent Match
90%
92%
82%
89%
85%
98%
93%
89%
Table 12-5. BTS to EDMS Matching Based on LTOs
Airport
Boston Logan
Cleveland-Hopkins
Newark Liberty
Washington Dulles
New York Kennedy
New York LaGuardia
Chicago O'Hare
Totals
Total LTO
185,524
114,457
214,716
153,842
184,576
192,837
462,930
1,508,882
Matched LTO
162,737
114,438
212,650
149,753
168,438
192,193
460,577
1,460,786
Percent Match
88%
100%
99%
97%
91%
100%
99%
97%
Typically, the EDMS input file quantifies aircraft activity relative to an aircraft's LTO
cycle. The cycle begins when the aircraft approaches the airport on its descent from cruising
altitude, then lands and taxis to the gate, where it idles during passenger deplaning. The cycle
continues as the aircraft idles during passenger boarding, taxis back out onto the runway, takes
off, and ascends (climbout) to cruising altitude. Thus, the six specific operating modes in an LTO
cycle noted in Figure 12-1 are the following:
• Approach;
• Taxi/idle-in;
• Taxi/idle-out;
• Idling;
• Takeoff; and
• Climbout.
The LTO cycle provides a basis for calculating aircraft emissions. During each mode of
operation, an aircraft engine operates at a specific power setting and fuel consumption rate for a
July 2009 12-5
-------
Technical Development Document for Proposed Effluent
Limitation Guidelines and Standards for the Airport Deicing Category
12. Non-Water Quality Impacts
given aircraft make and model. Emissions for one complete cycle are calculated using emission
factors for each operating mode for each specific aircraft engine combined with the typical
period of time the aircraft is in the operating mode.
Figure 12-1. Landing and Take Off Cycle
For this assessment, EPA ran the EDMS model using default time-in-mode values for
each component of the LTO cycle. Next, the Agency adjusted the time-in-mode values in the
model to account for additional time spent traveling to the deicing pad (15 minutes), engine
idling while deicing (30 minutes), and taxing away from the deicing pad (15 minutes) and reran
the model with these adjusted time-in-mode values. Then, EPA subtracted the baseline model
run from the second model run to estimate the additional emissions associated with deicing as
noted in the following equation:
AEX1 = ABX1 - BX1
(12-2)
where:
AEX1 =
ABX1 =
Annual additional emission associated aircraft deicing for airport /' and
pollutant x (tons per year);
Annual emission using time in mode values increased to account for time
spent taxiing to and from the deicing pad and idling at the deicing pad for
airport / and pollutant x (tons per year); and
Baseline annual emissions assuming no deicing activities for airport / and
pollutant x (tons per year).
Because the BTS data are in terms of annual LTOs, EPA adjusted these values to reflect
the SOFP days for each airport by multiplying the annual values by the SOFP days divided by
365 days per year, as noted in the following equation:
AExlxSP.
365
(12-3)
where:
AEX1 =
Additional emission associated aircraft deicing for airport /' and pollutant*
for SOFP period (tons/SOFP period);
Annual additional emission associated aircraft deicing for airport / and
pollutant x (tons per year);
July 2009
12-6
-------
Technical Development Document for Proposed Effluent
Limitation Guidelines and Standards for the Airport Deicing Category
12. Non-Water Quality Impacts
SP; = SOFP period for airport /' (days); and
365 = Days per year.
Table 12-6 presents these incremental emission estimates for criteria pollutants for each of the
seven airports used in the modeling program.
Table 12-6. Incremental Criteria Pollutant Emissions Associated with Aircraft Deicing
Operations
Airport
Boston Logan
Cleveland-Hopkins
Newark Liberty
Washington Dulles
New York Kennedy
New York LaGuardia
Chicago O'Hare
CO
(tons/yr)
118
93
111
67
94
62
357
THC
(tons/yr)
13
12
14
8
12
6
35
voc
(tons/yr)
14
13
16
8
13
6
38
NOX
(tons/yr)
21
15
21
11
17
11
61
SOX
(tons/yr)
5
4
5
3
4
3
15
PM10
(tons/yr)
0.4
0.3
0.5
0.2
0.4
0.2
1
CO - Carbon monoxide.
THC - Total hydrocarbons.
VOC - Volatile Organic Compounds.
NOX - Nitrogen Oxides.
SOX-Sulfur Dioxide.
PM10 - Paniculate Matter less than 10 um.
EPA also estimated total annual LTO aircraft emissions for the seven airports to compare
aircraft emissions associated only with deicing. Table 12-7 compares the total estimated LTO
emissions and the percentage increase in emissions from aircraft to comply with the proposed
rule. The data indicate that the proposed rule could increase carbon monoxide emissions from
aircraft at the impacted airports by as much as 6.9 percent due to additional ground-time needed
for pad deicing. Although the annual percentage increase in criteria pollutant emissions from the
seven airports included in this analysis is a concern, the actual increase in emissions (e.g., 903
tons per year of carbon monoxide) is insignificant when compared to total criteria pollutant
emissions for the aircraft sector. For example, in 2002, EPA estimated total carbon monoxide
emissions from the aircraft sector at approximately 260,000 tons (USEPA, National Emissions
Inventory website). Because increased criteria pollutant emissions resulting from additional
aircraft deicing time account for less than a 0.3 percentage increase in the aircraft sector annual
criteria pollutant emissions, EPA believes the very small increase is justifiable given the benefits
from the proposed rule.
July 2009
12-7
-------
Technical Development Document for Proposed Effluent
Limitation Guidelines and Standards for the Airport Deicing Category
12. Non-Water Quality Impacts
Table 12-7. Comparison of Total Annual LTO Aircraft Emissions to Emissions Resulting in
Deicing Operations
Emissions
Estimated Total Annual
Aircraft LTO Emissions for
Seven Airports
Estimated Total Emissions
from Deicing Operations for
Seven Airports
Estimated Percent Increase
in LTO Aircraft Emissions
Due to Proposed Rule
CO
(tons/yr)
13,174
903
6.9%
THC
(tons/yr)
1,464
99
6.8%
voc
(tons/yr)
1,591
108
6.8%
NOX
(tons/yr)
13,376
158
1.2%
SOX
(tons/yr)
1,135
38
3.4%
PM10
(tons/yr)
107
3
2.8%
CO - Carbon monoxide.
THC - Total hydrocarbons.
VOC - Volatile Organic Compounds.
NOX - Nitrogen Oxides.
SOX-Sulfur Dioxide.
PM10 - Paniculate Matter less than 10 um.
Emissions from AFB Treatment Systems
Anaerobic digestion of glycols found in ADF contaminated stormwater generate biogas
containing approximately 60 percent methane and 40 percent carbon dioxide. Airports installing
AFBs for treatment of ADF contaminated stormwater are expected to burn a portion of the gas in
on-site boilers in order to maintain reactor temperature. The remainder of gas can be either
combusted in a microturbine for electricity generation or flared. Regardless of the combustion
technology, nearly all biogas generated by AFBs is converted to carbon dioxide, the primary
green house gas. Table 12-8 shows biogas generation and potential carbon dioxide emissions
from AFB treatment systems for the proposed collection scenarios.
Table 12-8. Potential Air Emissions from AFB Treatment Systems
Regulatory Scenario
20% ADF Capture
40% ADF Capture
60% ADF Capture
Total COD Removal
(pounds/yr)
7,329,700
15,282,600
34,545,100
Potential Biogas
Generation
(million ft3/yr) 1
57
119
161
Potential Carbon Dioxide
Generation
(tons/yr) 2
3,700
7,600
17,300
Calculation based on 8 cubic feet of biogas per Ib COD removed. Biogas is 60% methane and 40% CO2 (Metcalf
& Eddy, 1979)
2 Assumes 99.9 percent of biogas is converted to CO2 during combustion.
Carbon dioxide is the primary greenhouse gas attributed to climate change; although
17,300 additional tons per year for 60% ADF capture appears to be considerable, the amount is
very small relative to other sources. For example, in 2006, industrial facilities combusting fossil
fuels emitted 948 million tons of CO2 equivalents. An additional 17,300 tons per year of carbon
dioxide from AFB treatment is less than 0.002 percent of the annual industrial carbon dioxide
emissions nationwide.
July 2009
-------
Technical Development Document for Proposed Effluent
Limitation Guidelines and Standards for the Airport Deicing Category
12. Non-Water Quality Impacts
12.3
Solid Waste Generation
AFB bioreactors will generate sludge that will require disposal, likely in an off-site
landfill. To estimate annual sludge generation by the AFB bioreactors that may be installed at
airports to treat ADF-contaminated stormwater, EPA first estimated the potential COD removal
for the proposed collection and treatment scenarios and then applied published anaerobic
biomass yield information (Metcalf & Eddy, 1979) to estimate total sludge generation on a
national basis. The biomass yield calculation, which simply multiplies the COD removal by the
yield, is a rough method of estimating sludge generation and does not account for other factors
such as degradation or inorganic material (e.g., AFB media) that may be entrained into the
sludge. However, this method does provide an order of magnitude estimate of sludge generation
that can be compared to other types of common biological treatment systems to determine if
AFB sludge generation would be unusually high at airports treating ADF-contaminated
stormwater.
Table 12-9 shows the total COD removal from each collection and treatment scenario and
the estimated sludge that would likely require disposal. This sludge is expected to be non-
hazardous and can be disposed in a municipal landfill. Detailed calculations showing how EPA
estimated biomass amounts are provided in a memorandum entitled Estimated Sludge
Generation from AFBs Treating ADF-Contaminated Stormwater (DCN ADO 1164).
Table 12-9. Estimated Sludge Generation from AFB Bioreactors Treating ADF
Contaminated Stormwater
Regulatory Scenario
20% ADF Capture
40% ADF Capture
60% ADF Capture
Total COD Removal
(pounds/yr) 1
7,329,700
15,282,600
34,545,100
Anaerobic Biomass Yield
(Ibs biomass/lb COD
removed) 2
0.03
0.03
0.03
Total Sludge Generation
(tons/yr)
110
229
518
Total COD removal from all AFB bioreactors which may be installed at airports.
2 Biomass yield from Metcalf and Eddy.
To provide some perspective on the potential total amount of biomass produced annually
by the AFB biological reactors treating ADF-contaminated stormwater, EPA compared the total
biomass generation data in Table 12-9 with the national biosolids estimates for all domestic
wastewater treatment plants throughout the United States. According to EPA's Municipal and
Solid Waste Division, approximately 8.2 million dry tons of biosolids will be produced in 2010
(USEPA, 1999). Using the highest biosolids generation amount shown in Table 12-9 (518
tons/yr), EPA estimates that AFB bioreactors treating ADF-contaminated stormwater will
increase biosolids generation in the United States by less than 0.01 percent. EPA believes this
very small percentage increase in biosolids generation is justifiable based on the benefits of the
Aircraft Deicing Operations proposed rule.
12.4
References
USC. 2007. U.S. Code Title 15, Commerce and Trade, Chapter 6. (January). Weights and
Measures and Standard Time. DCN AD00918.
July 2009
12-9
-------
Technical Development Document for Proposed Effluent 12. Non-Water Quality Impacts
Limitation Guidelines and Standards for the Airport Deicing Category
USEPA. 2006a. Airport Deicing Questionnaire Database. U.S. Environmental Protection
Agency/Office of Water. Washington, D.C. DCN AD00927.
USDOE. 2006. U.S. Department of Energy, Energy Information Administration. No. 2 Fuel Oil
All Sales/Deliveries by Prime Supplier. DCN AD00919.
USEPA. 2006b. Compilation of Air Pollutant Emission Factors (AP-42), Fifth Edition, Section
3.3.DCNAD00920.
USEPA. 2008. Inventory of U.S. Greenhouse gas Emissions and Sinks: 1990 - 2006. U.S.
Environmental Protection Agency/Office of Water. Washington, D.C. EPA 430-R-08-05 (April).
DCNAD00921.
USDOT. 2006. U.S. Department of Transportation, Federal Aviation Administration, Emissions
and Dispersion Modeling System, Version 4.5. Washington, D.C. (June). DCN AD00922.
USDOT. 2008. Bureau of Transportation Statistics, T-100 Segment, 2007. Washington, D.C.
DCNAD00923.
USEPA. National Emissions Inventory Air Pollutant Emission Trends Web Site
www. epa. gov/ttn/cheif/trends.
Metcalf and Eddy. 1979. Wastewater Engineering, Treatment/Disposal/Reuse, Second Edition.
McGraw Hill Press. DCN AD00924.
USEPA. 1999. Biosolids Generation, Use and Disposal in the United States. U.S. Environmental
Protection Agency/Office of Water. Washington, D.C. EPA 530D-R-99-009. DCN AD00925.
July 2009 12-10
-------
Technical Development Document for Proposed Effluent 13. Regulatory Option Selection
Limitation Guidelines and Standards for the Airport Deicing Category
13. REGULATORY OPTION SELECTION
This section presents the regulatory options evaluated by EPA for the Airport Deicing
Category and discusses the factors considered in determining the selected option for Best
Available Technology Economically Achievable (BAT) and New Source Performance Standards
(NSPS). Factors considered include reductions in pollutant discharges to the environment, costs
to the industry, size of airports involved, deicing practices used by the airports, changes to
deicing practices required, and non-water-quality environmental impacts. EPA is not setting Best
Practicable Control Technology Currently Available (BPT), Best Control Technology (BCT),
Pretreatment Standards for Existing Sources (PSES) or Pretreatment Standards for New Sources
(PSNS) for this point source category at this time.
The regulatory option selected provides the technology basis of the effluent limitation
guidelines and standards (ELGs) presented in this section. Owners or operators of airports
subject to these regulations would not be required to use the specific stormwater collection and
treatment technologies selected by EPA to establish the ELGs. Rather, an airport could choose to
use any combination of operational changes, stormwater collection technologies, and stormwater
treatment or control technologies to comply with the limitations and standards, provided the
limitations and standards are not achieved through prohibited dilution.
Section 13.1 discusses EPA's approach in developing the regulatory options considered
and Section 13.2 presents the rationale for the options selected under BAT and NSPS.
13.1 Regulatory Options Evaluated
Section 6 of this document summarizes EPA's estimates of the amount of airfield and
aircraft deicing chemicals currently used by U.S. commercial airports. Based on these usage
estimates, and as described in Section 10, EPA assessed the potential current direct discharge
loadings of five-day biochemical oxygen demand (BODs) and chemical oxygen demand (COD)
from both airfield and aircraft operations (referred to as baseline loadings and presented in
EPA's Airport Deicing Loadings Calculation memorandum (ERG, 2008a). EPA found that only
34 percent of the baseline COD load is estimated to come from airfield operations. In addition,
airfield deicing operations are likely to occur over a greater drainage area than aircraft deicing
operations alone. As such, the volume of deicing stormwater potentially generated from airfield
operations can be large, and collecting and controlling that stormwater may not be feasible or
cost-effective. EPA evaluated estimated costs for collecting and treating all deicing stormwater
at surveyed airports, rather than stormwater from aircraft deicing operation areas only, and found
that the potential costs were prohibitive (see Section 11.3.3.2). Because the majority of COD
discharge from airport deicing operations is believed to originate from aircraft deicing/anti-icing
operations, and collecting and controlling airfield deicing stormwater appears to be cost-
prohibitive, EPA's regulatory options focused on aircraft deicing operations.
As described in Section 9, EPA evaluated three different ADF collection and treatment
scenarios for aircraft deicing operations:
• 20% collection and treatment scenario uses glycol recovery vehicles (GRVs)
for deicing stormwater collection and anaerobic fluidized bed (AFB) treatment for
deicing stormwater control;
July 2009 13-1
-------
Technical Development Document for Proposed Effluent 13. Regulatory Option Selection
Limitation Guidelines and Standards for the Airport Deicing Category
• 40% collection and treatment scenario uses GRVs in combination with block-
and-pump technology for deicing stormwater collection and AFB treatment for
deicing stormwater control; and
• 60% collection and treatment scenario uses centralized deicing pads for deicing
stormwater collection and AFB treatment for deicing stormwater control.
EPA selected AFB treatment as the basis for BAT. The other three wastewater treatment
technologies that EPA considered (UF/RO, MVR/distillation, and aerobic biological treatment
ponds) were less effective at pollutant removal compared to AFB systems and are described in
Section 8.2. In addition, treating spent ADF with the mechanical methods, UF/RO and
MVR/distillation results in a concentrated waste stream that also must be disposed of. While
these technologies have potential as a part of an airport's pollutant control strategy, they are not
as effective as AFB when used as stand-alone treatment options, i.e. the pollutant removals they
achieve are not as great as the removals achieved by AFB systems.
Aerobic biological treatment ponds were not selected as the technology basis for BAT for
mainly logistical reasons. The ponds require large areas for installation, and the normal
operations of these systems require treatment for many months after the end of the annual
deicing season, before the wastewater can be discharged. Additionally, FAA discourages the
installation of new stormwater detention ponds at airports, as they can be a lure for migratory
birds. In those situations, birds and aircraft are safety hazards to each other. For airports with
existing detention ponds, however, where adequate storage capacity is available, aerated pond
systems may be able to provide efficient treatment that meets the standard.
Since collection and treatment of airfield deicing stormwater may not be practical, EPA
evaluated whether there were other regulatory options that could control the discharge of airfield
load and/or toxicity. Information collected by EPA during the airport site visits indicated options
might be to limit the amount of urea or eliminate the use of urea as an airfield deicing chemical
and replace it with other, less toxic products. Because other less toxic products (e.g., potassium
acetate) are available for substitution, most airports have already made the substitution, and the
total cost of substitution is not prohibitive, EPA is including as part of each regulatory option a
numeric effluent limit on ammonia in stormwater associated with deicing activities. Ammonia
was chosen to be a surrogate for urea.
EPA evaluated the costs, loading reductions, and economic impacts associated with each
ADF collection and treatment scenario for those airports included in EPA's airport
questionnaires. As discussed above, a BAT determination must be "economically achievable."
In order to meet this criteria EPA was required to look at a subset of the U.S. airport population,
a total of over 3,300 public airports.
Early in the regulatory development process, EPA focused on deicing activities at
primary airports, particularly those with extensive jet traffic. Operators of general aviation
aircraft, as well as smaller commercial non-jet aircraft, typically suspend flights during icing
conditions, whereas commercial airlines are much more likely to deice their jets in order to meet
customer demands.
July 2009 13-2
-------
Technical Development Document for Proposed Effluent
Limitation Guidelines and Standards for the Airport Deicing Category
13. Regulatory Option Selection
Based on the survey results, EPA estimated that 320 airports conduct deicing operations.
The Agency analyzed various industry characteristics that would be an indicator of affordability
for the candidate control and treatment technologies. This included a review of the relative sizes
of various airports (based on annual departures), the levels of deicing activity, traffic
characteristics (i.e. passenger vs. cargo operations), the extent of pollution controls and treatment
in place, and the costs of various technologies. EPA further classified airports based on the
number of annual jet departures. EPA found that there were some primary airports with high
percentages of non-jet traffic, and so it excluded airports with less than 1,000 annual jet
departures from the scope of the proposed rule. These airports have a higher proportion of
propeller-aircraft flights, which are typically delayed or cancelled during icing conditions. The
Agency estimated that the 218 larger primary airports perform most of the deicing operations,
and applying the regulation to this group would result in a substantial level of pollutant reduction
while being economically achievable for the industry. Table 13-1 presents the regulatory options
EPA evaluated for the Airport Deicing Category.
Table 13-1. Regulatory Options Evaluated for the Airport Deicing Category
Option
Option Description
Airports Subject to
Option
1
All primary commercial airports with over 1,000 jet departures that conduct
deicing per year and are not General Aviation/Cargo (GA/C) are in scope;
Airports with > 10,000 departures per year required to collect and control
20% of spent ADF and treat to numeric limit; and
All in-scope airports must meet effluent limit for ammonia or certify use of
non-urea-based pavement deicers.
110
All primary commercial airports with over 1,000 jet departures that conduct
deicing per year and are not GA/C are in scope;
Airports with > 10,000 departures per year required to collect and control
40% of spent ADF and treat to numeric limit; and
All in-scope airports must meet effluent limit for ammonia or certify use of
non-urea-based pavement deicers.
110
All primary commercial airports with over 1,000 jet departures that conduct
deicing per year and are not GA/C are in scope;
Airports with > 10,000 departures per year required to collect and control
20% of spent ADF and treat to numeric limit;
Airports with >460,000 gallons of ADF usage per year are required to collect
and control 60% of spent ADF and treat to numeric limit;
All in-scope airports must meet effluent limit for ammonia or certify use of
non-urea-based pavement deicers.
110
(14 airports subject to
60% requirement, 96
airports subject to
20% requirement)
All primary commercial airports with over 1,000 jet departures that conduct
deicing per year and are not GA/C are in scope;
Airports with >1,000 jet departures per year required to collect and control
20% of spent ADF and treat to numeric limit;
Airports with >460,000 gallons of ADF usage per year are required to collect
and control 60% of spent ADF and treat to numeric limit;
All in-scope airports must meet effluent limit for ammonia or certify use of
non-urea-based pavement deicers.
218
(14 airports subject to
60% requirement, 204
airports subject to
20% requirement)
Note: All references to ADF are for normalized ADF, which is ADF less any water added by the manufacturer or
customer before ADF application.
EPA used the following criteria to frame the regulatory options and the rationale for their
use:
July 2009
-------
Technical Development Document for Proposed Effluent 13. Regulatory Option Selection
Limitation Guidelines and Standards for the Airport Deicing Category
Primary commercial airport criterion. EPA focused on this category of airports
because they conduct the most operations, are expected to operate throughout the winter, and can
apply for funding. General aviation airports generally do not operate as many flights and will
often suspend operations during inclement weather.
Annual total aircraft departures criterion. EPA evaluated different annual departure
levels in developing the regulatory options to assess the benefit and impact of encompassing
more or fewer small/non-hub airports.
Annual jet departures criterion. Based on airport contact information, most aircraft
deicing operations occur on jet aircraft. Smaller non-jet aircraft are generally either stored in
hangers before winter flights or are grounded during inclement weather. EPA selected a
minimum number of jet departures to ensure that the airports considered for regulation would
conduct deicing regularly through out the winter season and thereby reflect airports with the
majority of COD load.
Annual ADF usage criterion. EPA based this criterion on the relationship between ADF
usage and the economic impact of the costs to comply with the 60 percent capture and treatment
scenario. Many of the large ADF usage airports that have evaluated their deicing operations and
have implemented best demonstrated technologies are achieving 60 percent capture and
treatment. EPA selected this criterion to include the large ADF usage airports that are not
currently addressing their deicing operations.
13.2 Option Selection
EPA evaluated the following factors in selecting options:
Reductions in pollutant discharges to the environment. EPA evaluated current direct
discharge of COD load (at baseline) and under the various regulatory options (Section 10
presents capture and treatment scenario load removals by airport and Table 13-2 summarizes the
regulatory option reductions);
Costs to the industry. EPA's cost analysis assessed the capital, annual operating and
maintenance (O&M), and annualized costs associated with each regulatory option to assess the
impact of those costs compared to total annual revenue by airport (Section 11 presents airport's
annualized costs by capture and treatment scenario and Table 13-2 summarizes the regulatory
option costs);
Size of the airport. EPA used departure data, Bureau of Transportation Statistics (BTS)
hub size designations, and small business definitions to assess the size of the airports considered
under the various regulatory options (airport size designations are discussed in the Economic
Analysis document (USEPA, 2009));
Deicing practices used by the industry. EPA assessed current airport deicing operations
and their impact on direct discharges to determine an existing collection and treatment level for
each airport evaluated (see Section 10.5.2 discussion); and
July 2009 13-4
-------
Technical Development Document for Proposed Effluent
Limitation Guidelines and Standards for the Airport Deicing Category
13. Regulatory Option Selection
Changes to deicing practices required. EPA used all of the airport operations data
collected to date to profile the best demonstrated practices and technologies currently in use for
airport deicing operations and to compare to the existing practices in use.
Table 13-2 summarizes several of the factors EPA evaluated for option selection, and
Sections 13.2.1 through 13.2.3 present EPA's proposed regulatory option for BAT, PSES/PSNS,
andNSPS.
Table 13-2. Factors Evaluated by EPA in Option Selection
Option
1
2
3
4
Option Removals
(million Ibs)
26.4
36.2
44.6
46.7
Option Annualized
Costs
(2006 million $)
36.4
110.1
91.3
105.0
Number of Airports with
Revenue Impact >3% of
Annual Revenue
9
20
11
57
Small Airports
with Revenue
Impact >3%
o
J
3
o
5
11
Each of the regulatory options evaluated include replacement of urea with potassium acetate as
an airfield deicing chemical. A breakout of the load removals and costs associated with urea
replacement are shown below in Table 13-3.
Table 13-3. Urea Replacement Load Removals and Costs
Replacement of Urea for Airfield Pavement Deicing
Description
No use of urea
(product sub.)
Airports Subject
to Option
218
COD Removal
(million Ibs)
12.7
Ammonia
Removal
(million Ibs)
4.7
Annual Removal
Cost
(2006 Million $)
$5.7
Removal $/lb
$0.26
13.2.1
BAT
Effluent limitation guidelines based on BAT represent the best available treatment
performance for deicing operations that is economically achievable. The Clean Water Act
(CWA) establishes BAT as the principal national means of controlling the direct discharge of
priority pollutants and nonconventional pollutants to waters of the United States. Based on
section 304(b)(2)(B) of the CWA, the factors considered in assessing BAT include:
The age of equipment and facilities involved;
The process used;
Process changes required;
Engineering aspects of control technologies;
The cost of achieving effluent reduction;
Non-water quality environmental impacts (including energy requirements); and
Other factors the Administrator deems appropriate.
July 2009
13-5
-------
Technical Development Document for Proposed Effluent 13. Regulatory Option Selection
Limitation Guidelines and Standards for the Airport Deicing Category
The Agency retains considerable discretion in assigning the weight to be accorded these
factors. BAT may include process changes or internal controls, even when these technologies are
not common industry practice.
Table 13-1 lists the BAT regulatory options considered by the Agency. Analysis of the
benefits of the collection and treatment scenarios evaluated by EPA in reducing pollutant
discharges to the environment, the cost to the industry, and the non-water quality environmental
impacts are described in Sections 10, 11, and 12, respectively. The Economic Analysis document
(USEPA, 2009) describes the economic impact of these scenarios on airports and airlines,
including impacts to small airports.
EPA evaluated the costs and economic impacts associated with each option and
determined that three of the options were economically achievable. After considering the
pollutant load removals, non-water quality impacts, and potential impact on small airports, EPA
selected Option 3. This option requires 60 percent collection and treatment for those airports
with the largest ADF usage (460,000 gallons or more annually) and 20 percent collection and
treatment for those airports with greater than 10,000 departures per year. EPA selected Option 3
because it provides the highest level of ADF- and urea-related COD removal, while also being
economically achievable, of the four options crafted by EPA. This option will require that high
ADF-usage airports that have not instituted deicing operation controls do so, leveling the playing
field with those large ADF usage airports that have already invested in collection and control
technologies.
13.2.2 PSES/PSNS
PSES are designed to prevent the discharge of pollutants that pass through, interfere with,
or are otherwise incompatible with the operation of publicly owned treatment works (POTWs).
The CWA required pretreatment for pollutants that pass through POTWs in amounts that would
exceed direct discharge of effluent limitations or limit POTW sludge management alternatives,
including the beneficial use of sludges on agricultural lands. Pretreatment standards are to be
technology-based and analogous to BAT for removal of priority and nonconventional pollutants.
Section 307(c) of the CWA requires EPA to promulgate PSNS at the same time that it
promulgates NSPS. New indirect discharging facilities, like new direct discharging facilities,
have the opportunity to incorporate the best available demonstrated technologies, including
process changes and in-plant treatment technologies that reduce pollution to the maximum extent
feasible. Pretreatment standards for new sources are to be technology-based and analogous to
NSPS for the removal of priority and nonconventional pollutants.
EPA is not proposing to set PSES or PSNS at this time for the Airport Deicing Category.
The main pollutants of concern for this category are BOD5 and COD. EPA is proposing a
biological treatment process (AFB) as BAT for direct discharges. POTWs are specifically
designed to treat BODs and COD using a biological treatment process (either aerobic or
anaerobic) and thus both the BAT and POTW technologies are equivalent. Therefore, EPA does
not believe that regulation is warranted for indirect discharges. As discussed in Section 5, over
62 commercial airports nationwide are currently indirectly discharging their ADF stormwater to
a POTW either in place of or in addition to direct discharge and EPA anticipates that this
practice will continue where practical.
July 2009 13-6
-------
Technical Development Document for Proposed Effluent 13. Regulatory Option Selection
Limitation Guidelines and Standards for the Airport Deicing Category
EPA is not aware of specific pass-through concerns for POTWs accepting airport deicing
stormwater. EPA is aware that slug loading of deicing stormwater can create POTW upset, and
many of the airports that discharge indirectly to POTWs have airport-specific requirements on
allowable BOD5 or COD discharge loading per day and may also have requirements for ramping
the load up or down over time. This is usually accomplished by storing deicing stormwater in
retention ponds, detention ponds, lagoons, or tanks and metering the discharge to meet permit
requirements.
13.2.3 NSPS
The basis for NSPS under Section 306 of the CWA is the "best available demonstrated
control technology." New sources have the opportunity to design and install the best and most
efficient process operations and wastewater treatment systems. Accordingly, Congress directed
EPA to consider the best demonstrated alternative processes, process changes, in-plant control
measures, and end-of-pipe wastewater treatment technologies that reduce pollution to the
maximum extent feasible.
NSPS establish quantitative limits on the direct discharge of conventional, priority, and
nonconventional pollutants to U.S. waters. These standards are based upon the performance of
specific advanced technologies, but do not specify which technologies must be used to achieve
compliance. NSPS are applied to individual facilities through NPDES permits issued by EPA or
authorized states under Section 402 of the CWA. Each facility then chooses its own approach to
complying with its permits limitations.
New sources for airport facilities will include the following: 1) new stand-alone airports
and 2) new substantially independent airport runways and the departures from those runways.
Much of the current air traffic growth has occurred as existing airport expansion
(including addition of new runways and concourses) at U.S. small, medium, and large hub
airports. There is also documented expansion at reliever airports located close to major hubs.
EPA does not anticipate significant new stand-alone airport construction in the near future.
EPA evaluated the best demonstrated practices and technologies used by the aviation
industry and found that the most effective systems include consolidating deicing operations using
deicing pad facilities and treating collected deicing stormwater through biological treatment prior
to direct discharge. The Denver International Airport is an excellent example of best practices for
deicing stormwater collection; the airport designed each of its runways with its own deicing pad
system and currently collects approximately 68 percent of their applied ADF for on-site glycol
recycle and recovery prior to indirect discharge. The Albany Airport is the basis for best
treatment practices prior to direct discharge using an AFB biological treatment system. EPA has
therefore selected the 60 percent collection and treatment scenario as best demonstrated practice
for the industry and is setting NSPS as 60 percent capture of available ADF with direct discharge
effluent limitations based on AFB treatment. The performance level of NSPS is equivalent to
BAT requirements for large ADF usage airports. Standards for COD are being established for
new sources consistent with the BAT performance level. In addition, EPA is proposing stringent
discharge limits on ammonia, based on product substitution for urea-based airfield pavement
July 2009 13-7
-------
Technical Development Document for Proposed Effluent 13. Regulatory Option Selection
Limitation Guidelines and Standards for the Airport Deicing Category
deicers, consistent with the proposed BAT requirements. Alternatively, facilities may certify use
of non-urea-based pavement deicers.
13.3 References
ERG. 2008a. Memorandum from Cortney Itle (ERG) to Brian D'Amico (EPA): Airport Deicing
Loading Calculations. (April 17). DCN AD01140.
USEPA. 2009. Economic Analysis for Proposed Effluent Limitation Guidelines and Standards
for the Airport Deicing Category. U.S. Environmental Protection Agency/Office of Water.
Washington, D.C. EPA 821-R-09-005. DCN AD01196.
July 2009 13-8
-------
Technical Development Document for Proposed Effluent 14. Limitation and Standards
Limitation Guidelines and Standards for the Airport Deicing Category Data Selection and Calculation
14. LIMITATIONS AND STANDARDS: DATA SELECTION AND CALCULATION
This section describes the data selection and statistical methodology that EPA used to
calculate the proposed limitations for the airport deicing point source category. As described in
this section, the proposed effluent limitations and standards account for variation in treatment
performance of the model technology. For simplicity, the following discussion refers only to
effluent limitations guidelines; however, the discussion also applies to new source standards.
EPA is proposing limitations for chemical oxygen demand (COD) and ammonia as
nitrogen (the latter as a compliance alternative), and Section 14.1 briefly describes the pollutant
parameters. Section 14.2 provides an overview of EPA's data review and selection process.
Section 14.3 describes EPA's data conventions. Sections 14.4 and 14.5 describe the COD and
ammonia as nitrogen data selected as the basis of the proposed limitations. Section 14.6
describes the percentile basis and calculations used for the limitations. Section 14.7 describes
achievability and compliance related to the limitations. Section 14.8 provides references.
14.1 Selected Pollutant Parameters
As described in Section 7, there are a number of pollutants associated with the discharge
from airport deicing operations. EPA is proposing effluent limitations for two pollutant
parameters, chemical oxygen demand (COD) and ammonia. This section briefly describes the
pollutant parameters and the chemical analytical methods used to measure their concentrations.
14.1.1 Chemical Oxygen Demand (COD)
COD is a measure of the total organic matter content of both wastewaters and natural
waters. Measurement of COD can be used to rapidly recognize deterioration in wastewater
treatment plant performance and the need for corrective action. EPA evaluated data collected by
the Albany International Airport in New York, and by EPA. EPA evaluated data for chemical
oxygen demand (COD) that was measured using EPA Method 410.4 and Hach 8000, both of
which are listed as approved for compliance monitoring in 40 CFR Part 136. EPA determined
COD using Method 410.4, and Albany Airport used Hach 8000. Data from the two methods are
directly comparable.
Method 410.4 is a colorimetric procedure with a measurement range of 3 to 900 mg/L for
automated procedures and a measurement range of 20 to 900 mg/L for manual procedures. The
Hach 8000 method is a colorimetric procedure that utilizes a preliminary digestion procedure and
can be used for various concentration ranges. A user has the option of purchasing three different
sets of reagents and standards. The first has a measurement range of 0 to 40 mg/L; the second: 0
to 150 mg/L; and the third: 0 to 1500 mg/L. The industry data had a lower measurement limit of
2.0 mg/L.
14.1.2 Ammonia as Nitrogen (Ammonia)
Ammonia as nitrogen (ammonia) is generated as a by-product of the use of urea-based
products for deicing operations. Ammonia can be directly toxic to fish and other aquatic
organisms and can reduce ambient dissolved oxygen concentrations in receiving surface waters.
In the data evaluated for the proposal, ammonia was measured using Methods 350.1 and 350.2,
both of which are listed as approved for compliance monitoring in 40 CFR Part 136. Albany
July 2009 14-1
-------
Technical Development Document for Proposed Effluent 14. Limitation and Standards
Limitation Guidelines and Standards for the Airport Deicing Category Data Selection and Calculation
Airport supplied data that was generated using Method 350.1 (DCN AD00824), and EPA used
Method 350.2. Both methods produce comparable results.
Method 350.1 is an automated colorimetric method that uses a continuous flow analytical
system and has a detection range of 0.01 to 2.0 mg/L. Method 350.2 utilizes either colorimetric,
titrimetric, or electrode procedures to measure ammonia. Method 350.2 has a lower measurement
range limit of 0.20 milligrams per liter (mg/L) for the colorimetric and electrode procedures and
a lower measurement range limit of 1.0 mg/L for the titrimetric procedure.
14.2 Overview of Data Review and Selection
As described in Sections 14.4 and 14.5, EPA qualitatively reviewed all the available
influent and effluent data for COD and ammonia. For purposes of limitations development, data
are defined to be numerical values resulting from laboratory determination of pollutants in
physical samples collected from influent and effluent wastestreams. A laboratory expresses the
results of its analysis either numerically or as "not quantitated" for a pollutant in a sample. When
the result is expressed numerically, then the pollutant was quantitated, or more commonly
referred to as "detected," in the sample. The definition includes measured values (e.g., 10 mg/L)
and values reported as being below some level of quantification (e.g., <10mg/L). The latter are
often referred to as "non-detected" and are usually reported with a "detection limit." (EPA also
uses terms such as "quantitation limit" in other documentation.) The definition of "data"
excludes estimated values and statistical summaries, such as averaged values.
This section describes EPA's review of the available data. Section 14.2.1 describes the
criteria that EPA applied in selecting data for the development of the proposed limitations.
Section 14.2.2 describes other considerations that were evaluated as part of the data review.
Section 14.2.3 discusses the importance of comments in EPA's evaluations of the data for the
final limitations.
14.2.1 Data Selection Criteria
This section describes the criteria that EPA applied in selecting data to use as the basis
for the proposed effluent limitations. EPA has used these or similar criteria in developing
limitations and standards for other industries. EPA uses these criteria to select data that reflect
consistently good performance of the model technology in treating the industrial wastes under
normal operating conditions. Model technology is technology that is carefully designed and
diligently operated. The following paragraphs describe the criteria and modifications specific to
the airport deicing category.
One criterion generally requires that the influents and effluents from the treatment
components represent typical wastewater from the industry, with no incompatible wastewater
from other sources (e.g., sanitary wastes). Application of this criterion results in EPA selecting
only those facilities where the commingled wastewaters did not result in substantial dilution,
unequalized slug loads that result in frequent upsets and/or overloads, more concentrated
wastewaters, or wastewaters with different types of pollutants than those generated by the
categorical (i.e., airport deicing) wastewater.
July 2009 14-2
-------
Technical Development Document for Proposed Effluent 14. Limitation and Standards
Limitation Guidelines and Standards for the Airport Deicing Category Data Selection and Calculation
A second criterion typically ensures that the pollutants were present in the influent at
sufficient concentrations to evaluate treatment effectiveness. To evaluate whether the data meet
this criterion for the final rule, EPA often uses a "long-term average test" for sites where EPA
possesses paired influent and effluent data. EPA has used such comparisons in developing
regulations for other industries, e.g., the Iron and Steel Category (EPA 2002). The test looks at
the influent concentrations to ensure a pollutant is present at sufficient concentration to evaluate
treatment effectiveness. If a pollutant fails the test (i.e., not present at a treatable concentration),
EPA excludes the data for that pollutant at that facility from its long-term average and variability
calculations. In this manner, EPA would ensure that the limitations resulted from treatment and
not simply the absence or substantial dilution of that pollutant in the wastestream. If industry
supplies EPA with effluent data, but not the corresponding influent data, EPA may choose to use
the effluent data without performing a long-term average test provided EPA determines that the
pollutant would have been present at consistently treatable concentrations at the facility. This
approach would satisfy EPA's objective to include as much data from as many facilities as
possible in the calculation of limits.
A third criterion requires that the facility must have the model treatment technology and
demonstrate consistently diligent operation. Application of this criterion typically eliminates any
facility with treatment other than the model technology. Exceptions are generally rare, but may
include facilities with treatment, or performance, that is equivalent to the model technology. EPA
generally determines whether a facility meets this criterion based upon personal visits, its ability
to comply with its existing discharge requirements, discussions with facility management, and/or
comparison to the performance of treatment systems at other facilities. EPA often contacts
facilities to determine whether data submitted were representative of normal operating conditions
for the facility and equipment. As a result of this review, EPA typically eliminates facilities that
experience repeated operating problems with their treatment systems. In addition, EPA typically
excludes data when the facility has not optimized its treatment. For example, facilities may use
the model technology as a pretreatment step before sending the wastewater to a publicly owned
treatment works (POTW), and consequently, might not fully optimize its system.
A fourth criterion typically requires that the data cannot represent periods of frequent
unequalized slug loading treatment upsets or shut-down periods1 because these excursion data do
not reflect performance that would be expected from well-designed and operated treatment
systems. Furthermore, it would not be appropriate for the limitations to be based, in part, upon
data reflecting extreme events that are beyond the control of the facility, because regulatory
provisions at 40 CFR 122.41(n) would apply to such circumstances. More specifically, after the
final limitations are incorporated into permits, EPA expects that when such events occur that the
facility would abide by the procedural requirements in §122.41(n) to obtain an affirmative
defense to any potential enforcement action.
In applying the fourth criterion, EPA evaluates the pollutant concentrations, flow values,
mass loadings, plant logs, and other available information. As part of this evaluation, EPA
1 EPA applies this criterion to data from two types of shut-downs. First, treatment systems are sometimes halted to
control upset conditions. As part of the recovery activities, the facility may pump out wastewater from the
equipment (e.g., tanks) which contains highly concentrated wastes associated with the upset. Second, the facility
may shut down its operations for maintenance and other atypical operations. During these periods, the facility may
still operate its treatment system, but typically discharges effluent associated with atypical influent. For example, the
influent might include cleaning solvents instead of process wastewater.
July 2009 14-3
-------
Technical Development Document for Proposed Effluent 14. Limitation and Standards
Limitation Guidelines and Standards for the Airport Deicing Category Data Selection and Calculation
typically asks the facility about process or treatment conditions that may have resulted in
extreme values (high and low). As a consequence of this review, EPA may identify certain time
periods and other outliers in the data that reflect poor performance by an otherwise well-operated
site.
The fourth criterion also is applied in EPA's review of data corresponding to "start-up"
periods. Most industries incur start-up conditions only during the adjustment period associated
with installing new treatment systems. During this acclimation and optimization process, the
concentration values tend to be highly variable with occasional extreme values (high and low).
After this initial adjustment period, the systems should operate at steady state for years with
relatively low variability around a long-term average. Because start-up conditions reflect one-
time operating conditions, EPA generally excludes such data in developing the limitations. In
contrast, EPA expects airports to encounter start-up operations at the start of every deicing
season because they probably will cease treatment operations during warmer months. Because
this adjustment period will occur every year for the Airport Deicing Category, EPA is proposing
to include start-up data in the data set used as the basis of the limitations. However, through its
application of the other three criteria, EPA would exclude extreme conditions that do not
demonstrate the level of control possible with proper operation and control even during start-up
periods.
In part, by retaining start-up data for limitations development, the limitations will be
achievable because EPA based these limits on typical treatment during the entire season. Once
the treatment system reaches steady state, EPA expects a typically well-designed and operated
system to run continuously until the end of the deicing season. Conversely, EPA might determine
that systems that operated only during relatively short periods, such as during each winter storm
event (i.e., of only several days duration), might be poorly operated because the model
technology requires more time to reach steady state. In other words, it would be ineffective and
disruptive to turn the system on and off throughout the deicing season, particularly for biological
systems, such as the model technology, and EPA may reject data if it determines that it reflects
this type of operation.
14.2.2 Other Considerations in Data Selection
In comments on proposed regulations across a range of industry categories and
subcategories, industries often suggest that EPA consider additional criteria in selecting data as
the basis of the limitations. Because EPA is aware of the issues behind these suggested criteria, it
routinely considers whether they are relevant and should be considered as it develops new
proposed regulations. As explained below, EPA also considered these criteria for the airport
deicing rulemaking, but determined that they were not relevant in selecting data as the basis of
the proposed limitations. EPA's rationale is consistent with its findings for other industries.
Commenters often suggest a criterion related to the size of facilities because of concerns
about a perceived impact of volume, or flow, of wastewater on treatment performance. In
considering this issue for the airport deicing industry, EPA concluded that the size of the airport,
deicing pads, and other features, by themselves, would not affect the performance of the
treatment system. Instead, the airport size and water flow would determine the size of the
treatment system, rather than its performance. EPA expects that each airport would build and
operate a system that is sized appropriately for its volume of wastewater. Because the method of
July 2009 14-4
-------
Technical Development Document for Proposed Effluent 14. Limitation and Standards
Limitation Guidelines and Standards for the Airport Deicing Category Data Selection and Calculation
treatment is the same regardless of the flow, properly-sized systems should all perform in the
same manner, and thus, achieve the same effluent concentrations. Before reaching this
conclusion for airport deicing and other industries, EPA reviewed treatment technologies, such
as biological treatment, oil-water separators, dissolved air floatation, and settling tanks. EPA's
record supports the finding that for a variety of industrial sectors, well-operated and designed
treatment systems treat different wastewaters with varying flows to a narrow range of effluent
concentrations (EPA 2006).
In addition, commenters typically suggest a criterion that would require a minimum
number of facilities be used as the basis of the limitations. Such suggestions are based upon two
main concerns. First, commenters are concerned that the limitations do not reflect treatment from
a range of typical influents, because the concentration levels vary from facility to facility.
Second, commenters are concerned that not all facilities could achieve the same high level of
performance from the model technology. For the first concern, as part of its evaluation of the
effect of flow described above, EPA also considered the impact of influent pollutant
concentrations. EPA found that well-operated and designed treatment systems are capable of
treating the wastewaters to a narrow range of effluent pollutant concentrations. For the second
concern, EPA only needs to demonstrate that the model technology can be operated at the level
of performance on which the limitations are based. EPA's selection of the model technology
used at the Albany, New York airport as the basis of the limitations is appropriate because that
facility demonstrates that the technology can achieve the levels reflected in the proposed
limitations.
The Clean Water Act specifically authorizes EPA to base BAT/NSPS limitations and
standards on the performance of a single facility. It is well established that BAT represents the
best performance in the industrial category. Thus, it is not unusual for EPA to base effluent
limitations on data from a single facility. For example, in the Organic Chemicals, Plastics and
Synthetic Fibers (OCPSF) effluent guideline promulgated in 1987, EPA based 38 percent of the
limitations on the performance of a single facility (EPA, 1987). Courts have recognized that EPA
must act on the information it has, and need not wait for perfect information. See e.g., BASF
Wyandotte Corp. v. Costle, 598 F.2d 637, 652-653 (1st Circuit, 1979.)
14.2.3 Importance of Comments for Data Evaluations for Final Limitations
EPA has provided data in the proposed rulemaking record and explained the criteria used
to review and evaluate this data. EPA encourages airport operators to submit comments about the
proposed limitations in any case. For example, if an airport has concerns that the final rule may
include more stringent limitations than those proposed, the airport may wish to provide
comments and data that support the proposed limitations.
EPA will consider the comments and other information in performing its data review for
the final rule. As a result of considering new information when applying the data selection
criteria, EPA may reach new conclusions about whether certain proposal data should be included
or excluded as the basis of the final limitations. It also is possible, as a consequence of new data,
that EPA would revise its approach and/or calculate different values for the performance
limitations and standards. As a consequence, the final limitations and standards could be more or
less stringent than those proposed.
July 2009 14-5
-------
Technical Development Document for Proposed Effluent 14. Limitation and Standards
Limitation Guidelines and Standards for the Airport Deicing Category Data Selection and Calculation
14.3 Conventions for Modeling Multiple Data Sets from the Same Facility
This section describes EPA's conventions for modeling multiple data sets from the same
facility. Data from a particular facility are sometimes collected at different times, from different
treatment units, or by different organizations. In such cases where multiple data sets exist for the
same facility, EPA often statistically models the data as if each data set represents a different
facility. This section describes conventions applied to the data from the airport deicing industry.
EPA generally considers data from different time periods to characterize different
operating conditions due to changes such as management, personnel, and procedures. Because
EPA expects airports to operate treatment systems only for a limited time each year, it
considered whether the conditions and performance for each deicing season tend to vary in a
manner that more appropriately reflects different treatment systems. Because it may better
capture the variability of airport deicing operations under a range of conditions, EPA has
calculated the proposed limitations using all of the data, without regard to season. For
informational purposes, the data and summaries are presented by season.
EPA generally uses data from separate treatment units (depending on the rulemaking,
also called "trains" or "systems") as if they characterized separate facilities. EPA has determined
this is appropriate because the units were operated separately. Even if the wastes were generated
by the same processes or drawn from the same storage pond, EPA considers that the performance
of each system can vary due to slightly different influents, equipment, and other factors.
EPA generally considers data from different organizations to characterize different
collection methods and analytical methods. The different organizations typically are EPA
sampling teams and the facility's monitoring crew at the treatment system. EPA often separates
such data into multiple data sets, to better model the variability consistent with the use of a single
analytical method and the same collection procedures. Consistent use of a single method and
procedure is often required by permits and is typical of monitoring for compliance. Therefore
this convention generally is used to model typical variability that each facility would experience
in compliance monitoring activities. EPA then determines which, or all, multiple data sources are
appropriate choices as the basis of the limitations.
14.4 COD: Data Selected as Basis of Proposed Limitations
In establishing the proposed limitations, EPA reviewed COD data from a treatment plant
at the Albany International Airport, which used the model BAT. (Selection of the model
technology is described in Section 13, and in the preamble to today's rule.) EPA collected COD
data during an EPA sampling episode at Albany Airport, and obtained several years of
monitoring data and other information from Albany. After evaluating data from EPA's sampling
episode and the data and information supplied by Albany, EPA determined that the Albany data
were the only available performance data from the model technology.2 Thus, all other data sets
were excluded by applying the third criteria in Section 14.2.1, because they did not demonstrate
the performance of the model technology. The following sections describe the Albany airport
2 Akron Canton Airport in Ohio started operating the model technology in mid-November in 2007. When EPA was
completing its technical analysis of the industry in late spring 2008, the airport's treatment unit was not operating at
full capacity.
July 2009 14-6
-------
Technical Development Document for Proposed Effluent 14. Limitation and Standards
Limitation Guidelines and Standards for the Airport Deicing Category Data Selection and Calculation
and apply the criteria in identifying the specific data points used as the basis of the proposed
limitations.
14.4.1 Albany Treatment System
EPA based the proposed limitations for COD upon data from Albany Airport's treatment
system. This system consists of two identical units that are consistent with EPA's model
technology described in Section 13. The airport diverts runoff from deicing operations into a
lagoon. Personnel at the facility then pump water from the lagoon to one anaerobic unit or the
other for treatment. The airport generally operates the two treatment systems in parallel, but
sometimes runs them in series. At the end of each unit, regardless of whether the system is in
parallel or series, the airport monitors COD concentrations each morning by collecting grab
samples to evaluate the treatment performance. These locations are labeled as ArprtR-101 and
ArprtR-102 in Figure 14-1. During its five-day sampling episode conducted from February 5th to
9th in 2006, EPA measured COD and ammonia concentrations in composite samples collected at
a point (labeled EPA_SP-2) where the two units combine into a single flow before entering an
aerobic polishing pond for more treatment. After this step, the airport typically directly
discharges waste into Shaker Creek, a tributary of the Mohawk River, which has been classified
as a New York State Class A drinking water stream. As a consequence, the airport is required to
meet stringent limitations when it discharges directly to the creek. In warmer weather (i.e., when
the soil is 50 degrees or warmer), the airport sometimes uses the treated wastewater for
irrigation. In addition, the airport has the capability of discharging to a POTW, although it
seldom uses this discharge mechanism.
As the basis for the proposed limitations for COD, EPA selected the data at sample points
ArprtR-101 and ArprtR-102 because each unit is the same as the model treatment system that
EPA identified in Section 13. The airport has monitored their performance for a relatively long
period of time, and provided EPA with data from December 1, 1999 through April 10, 2009 (ten
deicing seasons). Because the influent was highly concentrated, it was not necessary to perform
the long-term average test described in Section 14.2.3
DCN AD01181 provides the influent and effluent COD data as originally submitted by
the airport and the data are graphically displayed in the statistical support memo
(DCN#AD01208). The following sections describe the exclusion of data collected from the EPA
sampling episode and the airport's self-monitoring data.
3 EPA typically compares average influent levels to a multiplier of 5 to 10 times the quantitation limit (or reporting
limit). As explained in Section 14.1.1, the airport data had a lower measurement limit of 2.0 mg/L. Thus, in the long-
term average test, EPA would probably have compared the influent data to a reference level of 10 to 20 mg/L.
However, because the minimum influent value of 100 mg/L exceeded this range of potential reference values, the
average values also will meet the requirements of the LTA test.
July 2009 14-7
-------
Technical Development Document for Proposed Effluent
Limitation Guidelines and Standards for the Airport Deicing Category
14. Limitation and Standards
Data Selection and Calculation
EPA_SP-2
(COD, Ammonia)
Figure 14-1. Simplified Drawing of Albany Treatment System and Sample Points
14.4.1.1 COD Data from EPA Sampling Episode at Albany
During EPA's sampling episode, EPA and the airport collected separate sets of samples
and their concentrations are provided in Table 14-1. At sample point EPA_SP-2, EPA collected
samples of the combined flow from the two units. In contrast, the airport monitored the effluent
directly from each unit at sample points ArptR-101 and ArptR-102. Both sets of samples
demonstrate the performance of the model technology because no additional treatment steps exist
between the airport sample points and EPA's sample point. Rather than include data from two
different sources (i.e., EPA and the airport) for the same dates, EPA preferentially selected the
airport data because they were part of longer-term monitoring. Although the EPA data were
therefore excluded as the basis of the limitations, EPA notes that all of the values are less than
the value of 271 mg/L for the proposed daily maximum limitation.
Table 14-1. COD: EPA and Airport Self-Monitoring Effluent Data Collected During EPA's
Sampling Episode
Sample Date
2/5/06
2/6/06
2/7/06
2/8/06
2/9/06
COD Concentrations (mg/L)
EPA Sampling Episode Data
Original Sample
72
228
92
81
193
Field Duplicate
(where collected)
208
177
Airport Self-Monitoring Data
AprtR-101
29
53
56
31
48
AprtR-102
74
108
94
90
96
July 2009
14-8
-------
Technical Development Document for Proposed Effluent
Limitation Guidelines and Standards for the Airport Deicing Category
14. Limitation and Standards
Data Selection and Calculation
14.4.1.2 COD Self-Monitoring Data from Albany Airport
The airport typically runs the two units in parallel. When operated in this manner, EPA
considers each unit's performance to be consistent with the model technology. In its evaluation
of the data from each unit, EPA applied the criteria and other considerations described in Section
14.2. As a result of this evaluation, EPA excluded data associated with atypical operations,
influent concentrations reported as zero, estimated values, and poor performance. The exclusions
are described below.
By applying the third criterion in Section 14.2.1, EPA excluded all periods when the units
were operated in series because the data did not reflect treatment by the model technology. Table
14-2 identifies these time periods by deicing season (e.g., Season99 started in 1999 and
continued into 2000). During these periods, one unit provided initial treatment and the second
provided additional treatment. Although the facility reported effluent values from each unit
during this period of operating in series, it only discharged the effluent from the second unit in
the series. EPA considers the effluent data from the first unit to reflect less than optimal
performance, because the operators presumably would have not optimized treatment because
they intended to treat the wastes a second time (criterion 3). EPA considers the effluent data
from the second unit to characterize treatment of atypical influent (i.e., effluent that had been
treated by the first unit).
Table 14-2. COD: Dates Excluded Because Units Operated in Series
Season
Season99
SeasonOO
SeasonOl
Season02
SeasonOS
Season06
SeasonO?
Number Days Excluded
32
31
o
6
8
3
25
79
31
Beginning Date
2/9/2000
12/12/2000
2/7/2002
11/17/2002
4/5/2003
1/10/2006
1/15/2007
2/2/2008
End Date
3/11/2000
1/11/2001
2/9/2002
11/24/2002
4/7/2003
2/3/2006
4/3/2007
3/3/2008
EPA excluded effluent values for both units when the influent concentration was reported
as "zero" for two reasons. First, if COD could not be detected in the influent (criterion 2), then
the effluent concentrations would not reflect any additional treatment. Second, it appears that the
plant used this convention to indicate shut-downs of the system (criterion 4). Table 14-3
identifies the dates when the influent concentration was reported to be zero. For the final rule,
EPA also may exclude effluent values when the influent concentration was greater than zero, but
the flow (or feed) rate was reported as zero, because these data also may indicate that the system
was not operating.
July 2009
14-9
-------
Technical Development Document for Proposed Effluent
Limitation Guidelines and Standards for the Airport Deicing Category
14. Limitation and Standards
Data Selection and Calculation
Table 14-3. COD: Dates Excluded Because Influent Concentration Reported as Zero
Season
SeasonOl
Season02
Season06
Date
6/14/2002
11/16/2002
4/10/2007
EPA also excluded any data values that were estimates because they did not meet EPA's
definition for "data" described in Section 14.2. There were two types of estimated values. One
type was indicated by italicized font in the facility's spreadsheets as a convention to indicate that
the operator's log indicated issues with the sample or its analysis, and thus, the reported value
was an estimate. The second type was a series of identical values reported over consecutive
monitoring days. Although it is possible to find exactly the same pollutant concentrations on
consecutive days, variations from day to day are more typical. In response to EPA's questions
about the strings of identical values, the facility stated that it sometimes carried down the last
known number when they did not monitor. (DCN ADO 1206) To model the variability likely to
be present in the effluent, EPA assumed that the first non-zero value was the only day when
COD was monitored when three or more days had identical values for the same unit. In other
words, EPA retained the value for the first day and deleted the (identical) values on subsequent
days for that unit. If the values were zero, EPA assumed that the unit was not operating and
excluded the entire time period, including the first reported zero value. The statistical support
memo (DCN#ADO 1208) identifies these exclusions.
In addition, by applying the fourth criterion in Section 14.2, EPA excluded periods that
did not reflect the typical performance of the technology. As shown in Table 14-4, these
exclusions included treatment system upsets and method error. For example, EPA excluded the
maximum value of 1283 mg/L recorded at ArprtR-101 on 3/21/2001 because it was inconsistent
with the other data values during that time period. The plant management agrees that the value
does not reflect normal operations, and suspects that it was likely a sample with high solids
content. (DCN AD00825)
Table 14-4. COD: Dates Excluded Because of Performance Excursions
Season
Season99
SeasonOO
Number Days Excluded
11
42
1
1
Beginning Date
12/1/2000
1/12/2001
3/20/01
3/21/01
End Date
12/11/2000
2/22/2001
3/20/01
3/21/01
Reason
System Upset
System Upset
Method Error
System Upset
After incorporating the exclusions, more than 2500 measurements of COD remained and were
used as the basis of the proposed limitations. Table 14-5 summarizes the data. The statistical
support memo (DCN AD01208) provides a listing and plots of the data.
July 2009
14-10
-------
Technical Development Document for Proposed Effluent
Limitation Guidelines and Standards for the Airport Deicing Category
14. Limitation and Standards
Data Selection and Calculation
Table 14-5. COD: Summary of Albany Self-Monitoring Effluent Data After Exclusions
Unit
ArprtR_101
ArprtR_102
Season
Season99
SeasonOO
SeasonOl
Season02
SeasonOS
Season04
SeasonOS
Season06
SeasonO?
SeasonOS
ALL
Season99
SeasonOO
SeasonOl
Season02
SeasonOS
Season04
SeasonOS
Season06
SeasonO?
SeasonOS
ALL
#of
Days
147
112
168
180
146
140
90
62
124
120
1289
141
117
165
183
147
145
95
62
98
120
1273
Standard
Deviation
50.31
71.93
20.97
64.41
103.19
30.76
18.08
24.08
98.07
59.85
66.67
17.41
51.50
20.23
51.73
30.03
73.96
33.21
41.68
10.01
24.92
45.65
COD Concentrations (mg/L) l
Minimum
1
2
9
20
2
2
8
1
9
15
1
1
11
10
25
1
2
22
2
12
12
1
Maximum
326
575
157
655
699
275
162
136
1042
674
1042
93
393
168
685
210
725
275
148
58
282
725
Median
14.0
64.0
37.0
73.0
50.0
25.0
29.5
17.0
41.5
35.0
37.0
11.0
55.0
35.0
51.0
60.0
76.0
72.0
46.5
34.0
37.0
45.0
Mean2
28.60
75.46
44.02
86.80
76.32
31.51
31.56
23.10
54.48
42.74
52.28
16.85
63.09
40.01
57.61
63.12
87.42
77.19
61.34
33.70
39.07
53.40
In this summary, non-detected values are set equal to the detection limit.
The mean is calculated as the arithmetic average.
14.5
Ammonia: Data Selected as Basis of Proposed Limitation
For ammonia, EPA is proposing a compliance alternative with a daily maximum
limitation for airports that use urea deicers on the runways. This section describes the data
selected as the basis of the proposed limitation for ammonia.
After evaluating the available data, EPA transferred the ammonia data from the anaerobic
fluidized bed technology which is the model technology for COD. This transfer was necessary
because an AFB system by design creates ammonia as a by-product of wastewater treatment.
Consequently, AFB discharges could have higher ammonia concentrations than typically found
in airfield runoff when urea is not present. If the treated aircraft discharges then were discharged
to the same pipe as the runway runoff, the airport might have difficulties complying with the
ammonia limitation. EPA also confirmed that ADF treatment provided less concentrated
discharges than observed from the application of urea products (see DCN AD01194). For these
reasons, EPA determined that it was appropriate to use the ADF data as a basis of limitations that
would apply to runway runoff.
July 2009
14-11
-------
Technical Development Document for Proposed Effluent
Limitation Guidelines and Standards for the Airport Deicing Category
14. Limitation and Standards
Data Selection and Calculation
As it had for COD, EPA initially evaluated the Albany data for setting proposed
limitations for ammonia, because EPA had data on the performance of its model technology. In
contrast to its practice of monitoring COD at the end of each treatment unit, the airport
monitored ammonia at its permit compliance point after additional treatment provided by an
aerobic polishing step. The anaerobic polishing step would result in decreased ammonia
concentrations and would not represent ammonia discharges from use of urea on airfields. Thus,
to propose an effluent limit consistent with the model technology EPA used the EPA data and
excluded the Albany ammonia data (criterion 3).
Instead, because they reflect the capability of the model treatment system, EPA used its
sampling data from EPA_SP-2, shown in Table 14-6, as the basis of the proposed ammonia
limitations. (Section 14.6 describes field duplicates and the importance of daily values for the
limitation calculations.) In analyzing samples from this episode, the laboratory achieved a
detection limit of 0.05 mg/L. During the laboratory's quality assurance step of the chemical
analysis, it detected ammonia in the laboratory preparation blank at a concentration of 0.069
mg/L. Other quality assurance parameters, initial calibration blanks and continuing calibration
blanks, were between 0.052 mg/L and 0.054 mg/L. The ammonia results for all samples are
greater than ten times the blank result, with the exception of four influent and one source water
samples that were not used as the basis of the proposed limitation. As a consequence, EPA
determined that the data were of acceptable quality to use as the basis of the proposed
limitations.
Table 14-6. Ammonia: Data from Albany Airport Used to Develop Limitations
Sample Day
1
2
3
4
5
Ammonia Concentrations (mg/L)
Influent
ND(O.l)
ND(O.l)
ND(O.l)
ND(O.l)
0.91
Effluent
Original Sample
2.58
4.14
4.45
6.12
6.65
Field Duplicate
(where collected)
3.95
5.54
Daily Value Used in
Limitations Calculations
2.58
4.05
5.00
6.12
6.65
14.6
Limitations: Basis and Calculations
The proposed limitations, as presented in today's notice, are provided as the daily
maximum limitation for COD and ammonia. In addition, the notice includes a weekly average
limitation for COD. This section defines the limitations (Section 14.6.1); describes the statistical
percentile basis of the limitations (Section 14.6.2); and the estimation of the percentiles for COD
and ammonia (Sections 14.6.3). The statistical support memo (DCN AD01208) describes the
calculations used to model the ammonia data, as well as additional statistical models that EPA
may consider in developing the final limitations.
July 2009
14-12
-------
Technical Development Document for Proposed Effluent 14. Limitation and Standards
Limitation Guidelines and Standards for the Airport Deicing Category Data Selection and Calculation
14.6.1 Definitions
Definitions provided in 40 CFR 122.2 describe the limitations in terms of "daily
discharge" which it defines as "the 'discharge of a pollutant' measured during a calendar day or
any 24-hour period that reasonably represents the calendar day for purposes of sampling." As a
consequence, EPA generally arithmetically averages all measurements recorded for each
uniquely reported time period (e.g., 12/21/2004) before calculating limitations. EPA refers to this
averaged value as the "daily value."
First, in calculating the limitations, EPA ensures that it has only one value for each day.
Field duplicates are one example of multiple measurements, and were included in the ammonia
data used to develop the proposed limitations. Field duplicates are two samples collected for the
same sample point at approximately the same time, flagged as duplicates for a single sample
point, and measured separately. Because the analytical data from each duplicate pair characterize
the same conditions at that time at a single sample point, EPA typically averages the data to
obtain one data value for those conditions on that day. For example, Table 14-7 shows the field
duplicates and daily, averaged, values for ammonia.
Second, EPA uses the daily values in calculating the limitations. Definitions provided in
40 CFR 122.2 further describe the "maximum daily discharge limitation" as the "highest
allowable 'daily discharge.'" The "average weekly discharge limitation" is the "highest
allowable average of 'daily discharges' over a calendar week, calculated as the sum of all 'daily
discharges' measured during a calendar week divided by the number of 'daily discharges'
measured during that week."
Although EPA has not proposed a monthly average limitation, the following sections will
describe EPA's deliberations and evaluations of the monthly average limitation. The maximum
for monthly average limitation (also referred to as the "average monthly discharge limitation"
and "monthly average limitation") is the "highest allowable average of 'daily discharges' over a
calendar month, calculated as the sum of all 'daily discharges' measured during a calendar
month divided by the number of 'daily discharges' measured during that month."
14.6.2 Percentile Basis of the Limitations
EPA uses a statistical framework to establish limitations that facilities are capable of
complying with at all times. Statistical methods are appropriate for dealing with effluent data
because the quality of effluent, even in well-operated systems, is subject to a certain amount of
fluctuation or uncertainty. Statistics is the science of dealing with uncertainty in a logical and
consistent manner. Statistical methods together with engineering analysis of operating
conditions, therefore, provide a logical and consistent framework for analyzing a set of effluent
data and determining values from the data that form a reasonable basis for effluent limitations.
Using statistical methods, EPA has derived numerical values for its proposed daily maximum
limitations and weekly average limitations.
The statistical percentiles are intended, on one hand, to be high enough to accommodate
reasonably anticipated variability within control of the facility. The limitations also reflect a level
of performance consistent with the CWA requirement that these limitations be based on the best
technologies that are properly operated and maintained.
July 2009 14-13
-------
Technical Development Document for Proposed Effluent 14. Limitation and Standards
Limitation Guidelines and Standards for the Airport Deicing Category Data Selection and Calculation
In establishing daily maximum limitations, EPA's objective is to restrict the discharges
on a daily basis at a level that is achievable for an airport that targets its treatment system design
and operation at the long-term average while allowing for the variability around the long-term
average that results from normal operations. This variability means that at certain times airports
may discharge at a level that is greater than the long-term average. This variability also means
that airports may at other times discharge at a level that is considerably lower than the long-term
average. To allow for possibly higher daily discharges, EPA has established the daily maximum
limitation at a relatively high level (i.e., the 99th percentile). Due to routine variability in treated
effluent, an airport that discharges consistently at a level near the daily maximum limitation,
instead of the long-term average, may experience frequent values exceeding the limitations. For
this reason, EPA recommends that airports target the treatment system at the long-term average
that it derived for the model technology.
In its derivation of the weekly average limitation for COD, EPA used an estimate of the
97th percentile of the weekly averages of the daily measurements. This percentile basis is the
midpoint of the percentiles used for the daily maximum limitation (i.e., 99th percentile of the
distribution of daily values) and the monthly average limitation (i.e., 95th percentile of the
distribution of monthly average values). Courts have upheld EPA's use of these percentiles, and
the selection of the 97th percentile is a logical extension of this practice. Compliance with the
daily maximum limitation is determined by a single daily value; therefore, EPA considers the
99th percentile to provide a reasonable basis for the daily maximum limitation by providing an
allowance for an occasional extreme discharge. Because compliance with the monthly average
limitation is based upon more than one daily measurement and averages are less variable than
daily discharges, EPA has determined that facilities should be capable of controlling the average
of daily discharges to avoid extreme monthly averages above the 95th percentile. In a similar
manner to the monthly average limitation, compliance with the weekly average limitation also
would be based upon more than one daily measurement. However, the airport would monitor for
a shorter time and thus would have fewer opportunities to counterbalance highly concentrated
daily discharges with lower ones. For this reason, EPA is proposing a larger percentile for the
weekly average limitation than the one used for the monthly average limitation. Consequently,
EPA is proposing the 97th percentile as an appropriate basis for limiting average discharges on a
weekly basis. EPA also considers this level of control in avoiding extreme weekly average
discharges to be possible for airports using the model technology.
14.6.3 Estimation Procedures for Percentiles
This section describes the estimation procedures that EPA used to calculate the
limitations considered for the proposed rulemaking. Table 14-7 provides a summary of the
limitations that EPA proposes for COD and ammonia. Sections 14.6.3.1, 14.6.3.2, and 14.6.3.3
describe the estimation procedures used to model the COD data and the June 8, 2009
memorandum on calculation of percentiles (DCN AD01213) describes the calculations used by
the statistical software. Section 14.6.3.4 describes the procedures for ammonia. Also, as
described in Section 14.6.4.3 and 14.6.3.4, EPA considered a weekly average limitation for
ammonia and monthly average limitations for both pollutants.
July 2009 14-14
-------
Technical Development Document for Proposed Effluent
Limitation Guidelines and Standards for the Airport Deicing Category
14. Limitation and Standards
Data Selection and Calculation
Table 14-7. COD and Ammonia: Proposed Limitations with Long-Term Averages and
Variability Factors
Parameter
Limitations (mg/L)
Long-Term Average (mg/L)
Variability Factors
Time Period
Daily Maximum
Weekly Average
All
Daily
Weekly
COD
271
154
41.0
6.61
3.8
Ammonia
14.7
N/A
5.24
2.81
N/A
14.6.3.1 COD: Daily Maximum Limitation and the 99th Percentile
For COD, EPA based the proposed daily maximum limitation on an estimate of the 99th
percentile of the distribution of the daily values. This section describes the percentile estimates
and the long-term average.
First, EPA used nonparametric methods to estimate the 99th percentile of the daily values
from each unit. A simple nonparametric estimate of the 99th percentile of an effluent
concentration data set is the observed value that exceeds 99 percent of the observed data points.
Because EPA had more than 1200 data points for each unit, it determined that the empirical
approach would provide reasonable estimates of the 99th percentiles.
Second, EPA set the proposed limitation equal to the median of the two 99th percentile
estimates, or 271 mg/L. The median is, by definition, the midpoint of all available data values
ordered (i.e., ranked) from smallest to largest. As result, half of the unit 99th percentiles are
higher than the median, and half are lower. (In this particular case, because there are two units,
the median is equal to the arithmetic average (or mean).)
Table 14-8 summarizes the percentile estimates for the two units, the minimum and
maximum values observed in the data, and the 50th percentiles. Because of the importance of
targeting treatment to the long-term average, EPA recommends that facilities design, maintain,
and operate the treatment system to achieve a long-term average of 41 mg/L which is the median
of the 50th percentiles, of 37 and 45 mg/L, from the two units. The allowance for variability, or
the ratio of the limitation to the long-term average, is 6.6. (EPA usually refers to this allowance
as the "variability factor.") In other words, the daily maximum limitation is 6.6 times greater
than the long-term average achievable by the model technology. By targeting the system to the
long-term average and controlling its variability within this range, the facility will be capable of
complying with the limitation.
Table 14-8. COD: 99th Percentile Estimates from Each Treatment Unit
Treatment Unit
ArprtR-101
ArprtR-102
Median Values
Number of Daily
Values
1289
1273
Concentrations (mg/L)
Minimum
1
1
50th Percentile
37
45
41
Maximum
1042
725
P99
326
216
271
July 2009
14-15
-------
Technical Development Document for Proposed Effluent
Limitation Guidelines and Standards for the Airport Deicing Category
14. Limitation and Standards
Data Selection and Calculation
14.6.3.2 COD: Weekly Average Limitation and the 97th Percentile
For the weekly average limitation of COD, EPA first calculated, for each unit, the
arithmetic average of the measurements observed during each week, excluding weekends (to be
consistent with the assumed monitoring costs, although permit authorities may specify different
monitoring requirements). EPA then used the nonparametric method to derive a 97th percentile of
the more than 200 weekly averages for each unit, and set the proposed limitation equal to the
median of the two 97th percentile estimates, or 154 mg/L. The statistical support memo (DCN
AD01208) lists the weekly averages.
Because data was not always available for every weekday during a week, EPA examined
whether the weekly averages were affected by the number of weekdays included in the average.
As shown in Table 14-9, the value of the limitation varied only slightly if the weeks were
required to have data for all five days. The June 23, 2009 memorandum on percentiles for
weekly averages (DCN ADO 1214) provides a more detailed evaluation.
Table 14-9. COD: Effect of Number of Daily Values in Weekly Averages
Number of Daily
Values in Average
5
4 or 5
ItoS1
Unit ArprtR-101
Number of
Weekly
Averages
155
181
209
97th
Percentile
176.8
176.8
162.4
Unit AirprtR-102
Number of
Weekly
Averages
157
181
203
97th
Percentile
133.6
133.6
145.5
Median of 97th
Percentiles
155.2
155.2
153.95
Averages in this row were used as the basis of the proposed weekly average limitation.
14.6.3.3 COD: Monthly Average Limitation and the 95th Percentile (NOT
Proposed)
For COD, EPA is proposing and soliciting comment on use of a weekly average instead
of a monthly average limitation because it appears to be a better fit for this industry from a
monitoring perspective. However, two factors may warrant another approach in the final rule.
First, a week may be too short a period to ensure that airports will optimize their systems
appropriately over a longer period to achieve the long-term average. Second, the industry and
permit writers may prefer other alternatives. Another approach may include the monthly average
limitation. For comparison purposes, EPA tentatively estimated 112 mg/L as the 95th percentile
of the monthly averages using a statistical model based upon the lognormal distribution. The July
20, 2009 memorandum on the COD monthly average limitation (DCN ADO 1212) describes these
calculations. If EPA were to establish a monthly average limitation, it would examine the
statistical properties of the data to determine the appropriate model and statistical assumptions.
The August 2009 memorandum on time series modeling of effluent data (DCN AD01209)
describes this evaluation for several years of the monitoring data.
14.6.3.4 Ammonia: Percentile Estimates Based Upon the Lognormal Distribution
Because the ammonia data set had fewer than 100 observations, EPA used a parametric
approach to model the data. If a data set consists of fewer than 100 observations the best that can
July 2009
14-16
-------
Technical Development Document for Proposed Effluent 14. Limitation and Standards
Limitation Guidelines and Standards for the Airport Deicing Category Data Selection and Calculation
be done, using nonparametric methods, is to use the maximum value as an approximate
nonparametric estimate of the 99th percentile, but this can underestimate the true value.
Parametric methods require that a probability distribution be specified and this allows estimation
of unknown parameters from the available data. The estimated parameters are a function of the
defined distribution and the data, and thus the parametric method enables the percentiles of
effluent concentrations to be computed analytically. EPA's selection of parametric methods in
developing limitations for other industries is well documented (e.g., Iron and Steel; Pulp, Paper
and Paperboard; Metal Products and Machinery categories). EPA considers the lognormal
distribution to be appropriate for the ammonia data, and this section describes its application in
estimating the proposed daily maximum limitation. The daily maximum limitation of 14.7 mg/L
is based upon an estimate of the 99th percentile of the lognormal distribution of the daily values.
The calculations include an adjustment for possible bias due to statistical autocorrelation.
The adjusted variance then better reflects the underlying variability that would be present if the
data were collected over a longer period. When data are said to be positively autocorrelated, it
means that measurements taken at specific time intervals (such as 1 day or 2 days apart) are
related. For example, positive autocorrelation would be present in the data if the final effluent
concentration was relatively high one day and was likely to remain at similar high values the
next and possibly succeeding days. EPA sampling data, used as the basis of the limitations, were
collected on five consecutive days, and thus, the data may be autocorrelated, but the length of
time was not sufficient for autocorrelation evaluations. Albany Airport's self-monitoring data
also were not suitable for the evaluation because they were collected at three-week intervals
rather than consecutive days. In contrast, the Iron and Steel (I&S) rule had 244 data points for
ammonia that generally were collected on consecutive days, so it was possible to evaluate
autocorrelation in the data. Because the model technologies for both industries are biological
systems, EPA concludes that the I&S autocorrelation adjustment is a reasonable transfer that can
be used to calculate the airport deicing limitations.4 Table 14-10 summarizes the proposed long-
term average and daily maximum limitation, with and without the adjustment for autocorrelation.
The proposed daily maximum limitation of 14.7 mg/L is 2.8 times greater than the long-term
average, of 5.24 mg/L, achievable by the model technology. By targeting the system to the long-
term average and controlling its variability within this range, the facility will be capable of
complying with the limitation. However, ammonia is generated as a by-product of the model
technology, and EPA expects the concentrations of ammonia to have similar variability to what
is being treated (i.e., COD). In contrast to the COD limitations, which are based on a mixture of
start-up and steady state periods, the ammonia limitation is based upon data collected only
during steady state operations. In the preamble to the proposed rulemaking, EPA requests
additional data that reflect ammonia discharges during start-up operations.
4 EPA has not incorporated a similar adjustment for autocorrelation into the data for the COD limitations because
the limitation is based upon more than 2500 measurements collected over 10 years, which presumably would show a
full range of variability expected by the model technology. (DCNs AD01210 and AD01214)
July 2009 14-17
-------
Technical Development Document for Proposed Effluent
Limitation Guidelines and Standards for the Airport Deicing Category
14. Limitation and Standards
Data Selection and Calculation
Table 14-10. Ammonia: Consideration of Autocorrelation for Proposed Limitations, Long-
Term Averages, and Variability Factors
Statistical Parameter
Long-Term Average (mg/L)
Variability Factor
Daily Maximum Limitation (mg/L)
Adjusted for Autocorrelation?
No
4.97
2.25
11.2
Yes (Proposed)
5.24
2.81
14.7
Percent
Difference
5%
25%
31%
Unlike COD, EPA is not proposing a weekly ammonia effluent limitation. The
technology basis for the COD effluent limitations would operate throughout the deicing season
with continuous discharges allowing for weekly monitoring. In contrast, urea is applied to
airfield pavement as needed, and discharges would occur for a short time after the initial
application, as the urea works its way through the stormwater collection and any associated
treatment system that may be present. Most airports would have non-continuous and somewhat
infrequent urea discharges. Consequently, it would be difficult to assume a single value for the
monitoring frequency that could reasonably be applied to all airports, regardless of climatic
conditions. In developing the average limitations, this assumed monitoring frequency is used in
the statistical calculations. After reviewing any supplementary information and comments on
EPA's proposed limits, EPA may reevaluate whether weekly and/or monthly average limitations
are necessary for proper control of ammonia. After modeling the data using the lognormal
distribution as shown in the statistical support memo (DCN AD01208), EPA estimated values of
9.75 and 6.98 mg/L for the weekly average limitation and monthly average limitation.
14.6.3.5 Significant Digits for Proposed Limitations
In presenting the values of the proposed limitations, EPA rounded the values to three
significant digits. EPA used a rounding procedure where values of five and above are rounded up
and values of four and below are rounded down. For example, a value of 5.235 would be
rounded to 5.24, while a value of 5.234 would be rounded to 5.23.
14.7
Achievability of Limitations
EPA promulgates limitations that sites are capable of complying with at all times by
properly operating and maintaining their processes and treatment technologies. As a consequence
of using the percentile basis for each proposed limitation, treatment systems that are designed
and operated to achieve long-term average levels should be capable of compliance with the
limitations, which incorporate variability, at all times. As verification that the limitations are
achievable, EPA performs additional statistical and engineering reviews, as described in Section
14.7.1. As a result of these reviews, EPA has concluded that these limitations are achievable, and
thus, EPA expects facilities to comply with the limitations as explained in Section 14.7.2.
14.7.1
Statistical and Engineering Review of Limitations
In conjunction with the statistical methods, EPA performs an engineering review to
verify that the limitations are reasonable based upon the design and expected operation of the
control technologies and the airport conditions. The following sections describe two types of
July 2009
14-18
-------
Technical Development Document for Proposed Effluent 14. Limitation and Standards
Limitation Guidelines and Standards for the Airport Deicing Category Data Selection and Calculation
comparisons. First, EPA compares the proposed limitations to the data used to develop the
limitations. Second, EPA compares the limitations to the influent data.
14.7.1.1 Comparison to Data Used As Basis for the Limitations
As part of its data evaluations, EPA compared the value of the proposed limitations to the
values used to calculate the limitations. None of the data selected for ammonia were greater than
its proposed daily maximum limitation which supports the engineering and statistical
conclusions that the limitation values are appropriate. Because of the statistical methodology
used for the COD limitations some values were greater than the proposed limitations.
For the COD, appropriately one percent of the values were greater than the proposed
daily maximum limitation, which is consistent with the statistical basis (i.e., use of the 99th
percentile) of the limitation. Table 14-11 lists the data from both units and the influent, when
one, or both effluent values were greater than the limitation. Of the 27 values greater than the
proposed limitation, 20 were from the ArprtR-101 unit, and 7 from ArprtR-102 unit. Both units
had values greater than the proposed limitation on three dates: 3/31/2001, 1/4/2005, and
12/25/2008. For the final rule, EPA may contact Albany Airport to better understand these 27
values, determine whether they should be considered upsets of the treatment units, and evaluate
controls that will protect against these more concentrated discharges.
July 2009 14-19
-------
Technical Development Document for Proposed Effluent
Limitation Guidelines and Standards for the Airport Deicing Category
14. Limitation and Standards
Data Selection and Calculation
Table 14-11. COD: Dates and Values Greater than Proposed Limitation of 271 mg/L
Season
Season99
SeasonOO
Season02
SeasonOS
Season04
SeasonOS
SeasonO?
SeasonOS
Date
16MAR2000
23MAR2000
01MAR2001
11MAR2001
12MAR2001
22MAR2001
31MAR2001
18MAY2003
19MAY2003
20MAY2003
22MAY2003
20DEC2003
03JAN2004
08JAN2004
08FEB2004
09FEB2004
16MAR2004
17MAR2004
04JAN2005
04FEB2005
09DEC2005
09JAN2008
10JAN2008
25DEC2008
COD Concentrations (mg/L) l
Influent
6,170
6,560
6,240
5,430
6,520
6,040
5,460
5,870
6,020
5,920
6,645
2,955
4,885
7,085
6,770
6,980
8,280
8,300
8,670
5,845
1,100
8,630
8,630
6,280
ArprtR 101
326
315
276
288
575
129
357
(Estimated to be 800)
460
655
290
278
690
387
435
453
316
699
275
38
162
1,042
433
674
ArprtR 102
33
93
232
64
92
393
288
685
95
86
101
0
36
37
74
49
124
118
725
360
275
Out of service
Out of service
282
Bold text indicates effluent values greater than the limitations.
Of the 460 weekly averages of the COD concentrations, 14 averages had values that were
greater than the proposed weekly average limitation of 154 mg/L. Of those 14 averages, 11 were
during weeks when the unit also had one or more daily values that were greater than the daily
maximum limitation. The statistical support memo (DCN AD01208) identifies the weeks and the
corresponding daily values.
14.7.1.2 Comparison to Influent
In addition to evaluating the data used as the basis of the limitations, EPA often compares
the value of the proposed limitations to influent concentration levels. In these comparisons, EPA
determines if the limitations perform as expected.
As part of its evaluation to determine if the COD limitation was sufficiently stringent to
require that the influent be treated, EPA evaluated the COD influent discharges from Albany
Airport. As shown in the summary statistics in Table 14-12, all influent values were greater than
the proposed limitation during seven deicing seasons. For the other three seasons, only three
July 2009 14-20
-------
Technical Development Document for Proposed Effluent
Limitation Guidelines and Standards for the Airport Deicing Category
14. Limitation and Standards
Data Selection and Calculation
values (11/30/2005, 1/1/2007, and 1/2/2007) were less than the proposed limitation.5 This
finding confirmed that the proposed limitation can only be met through treatment.
Table 14-12. COD: Summary Statistics of Influent Concentrations
Season
Season99
SeasonOO
SeasonOl
Season02
SeasonOS
Season04
SeasonOS
Season06
SeasonO?
SeasonOS
ALL
#ofDays
149
117
169
183
147
145
98
64
128
120
1,320
COD: Influent Concentration (mg/L)
Standard
Deviation
1,571
1,254
1,511
1,807
2,437
1,777
1,768
1,677
2,150
4,409
2,933
Minimum
1,000
1,797
342
2,915
655
1,848
100
100
485
3,550
100
Maximum
6,560
7,950
7,105
10,470
10,060
8,870
7,410
5,760
10,000
18,300
18,300
Median
2,703
5,505
3,975
7,260
6,460
5,430
4,475
1,903
7,525
8,875
5,400
Arithmetic
Average
3,138
5,204
3,949
7,107
5,951
5,243
4,166
2,567
6,684
10,022
5,538
14.7.2 Compliance with Limitations
EPA promulgates limitations that sites are capable of complying with at all times by
properly operating and maintaining their processes and treatment technologies. However, the
issue of exceedances or excursions (values that exceed the limitations) is often raised. In other
rules, including EPA's final OCPSF rule, commenters suggested that EPA include a provision
that a facility is in compliance with permit limitations provided its discharge does not exceed the
specified limitations, with the exception that the discharge may exceed the daily maximum
limitation 1 day out of 100. EPA's general approach in that case for developing limitations based
on percentiles was the same as this rule and was upheld in Chemical Manufacturers Association
v. U.S. Environmental Protection Agency, 870 F.2d 177, 230 (5th Cir. 1989). The Court
determined the following:
EPA reasonably concluded that the data points exceeding the 99th and 95th
percentiles represent either quality-control problems or upsets because there can
be no other explanation for these isolated and extremely high discharges. If these
data points result from quality-control problems, the exceedances they represent
are within the control of the plant. If, however, the data points represent
exceedances beyond the control of the industry, the upset defense is available.
Id. at 230.
This issue also was raised in EPA's Phase I rule for the Pulp, Paper and Paperboard
Category (EPA 1998). In that rulemaking, EPA used the same general percentile approach for
5 For all three dates, the facility reported the same values for influent (100 mg/L) and ArprtR-101 (30 mg/L). The
facility reported a value of 30 mg/L for ArprtR-102 on the first date, and estimated values of 30 mg/L for the two
dates in 2007.
July 2009 14-21
-------
Technical Development Document for Proposed Effluent 14. Limitation and Standards
Limitation Guidelines and Standards for the Airport Deicing Category Data Selection and Calculation
developing monthly average limitations that it used for daily maximum limitation for the
proposed airport deicing rule. The percentile approach for the monthly average limitation was
upheld in National Wildlife Federation et al. v. Environmental Protection Agency, 286 F.3d 554,
573 (D.C. Cir. 2002). The Court determined that:
EPA's approach to developing monthly limitations was reasonable. It established
limitations based on percentiles achieved by facilities using well-operated and
controlled processes and treatment systems. It is therefore reasonable for EPA to
conclude that measurements above the limitations are due to either upset
conditions or deficiencies in process and treatment system maintenance and
operation. EPA has included an affirmative defense that is available to mills that
exceed limitations due to an unforeseen event. EPA reasonably concluded that
other exceedances would be the result of design or operational deficiencies. EPA
rejected Industry Petitioners' claim that facilities are expected to operate
processes and treatment systems so as to violate the limitations at some pre-set
rate. EPA explained that the statistical methodology was used as a framework to
establish the limitations based on percentiles. These limitations were never
intended to have the rigid probabilistic interpretation that Industry Petitioners
have adopted. Therefore, we reject Industry Petitioners' challenge to the effluent
limitations.
As that Court recognized, EPA's allowance for reasonably anticipated variability in its
effluent limitations, coupled with the availability of the upset defense, reasonably accommodates
acceptable excursions. Any further excursion allowances would go beyond the reasonable
accommodation of variability and would jeopardize the effective control of pollutant discharges
on a consistent basis and/or bog down administrative and enforcement proceedings in detailed
fact-finding exercises, contrary to Congressional intent. See, for example, Rep. No. 92-414, 92d
Congress, 2d Sess. 64, reprinted in^4 Legislative History of the Water Pollution Control Act
Amendments of 1972 (at 1482); Legislative History of the Clean Water Act of 1977 (at 464-65).
More recently, for EPA's rule for the iron and steel industry, EPA's selection of
percentiles was upheld in American Coke and Coal Chemicals Institute v. Environmental
Protection Agency, 452 F.3d 930, 945 (D.C. Cir. 2006). The Court determined that:
The court will not second-guess EPA's expertise with regard to what the
maximum effluent limits represent. See Nat'l Wildlife, 286 F.3d at 571-73. As
EPA explains in the Final Development Document, the daily and monthly average
effluent limitations are not promulgated with the expectation that a plant will
operate with an eye toward barely achieving the limitations. Final Development
Document at § 14.6.2. Should a plant do so, it could be expected to exceed these
limits frequently because of the foreseeable variation in treatment effectiveness.
Rather, the effluent limitations are promulgated with the expectation that plants
will be operated with an eye towards achieving the equivalent of the LTA for the
BAT-1 model technology. Id. However, even operated with the goal of achieving
the BAT-1 LTA, a plant's actual results will vary. EPA's maximum daily
limitations are designed to be forgiving enough to cover the operations of a well-
operated model facility 99% of the time, while its maximum monthly average
limitations are designed to be forgiving enough to accommodate the operations of
July 2009 14-22
-------
Technical Development Document for Proposed Effluent 14. Limitation and Standards
Limitation Guidelines and Standards for the Airport Deicing Category Data Selection and Calculation
a well-operated model facility 95% of the time. See id. EPA's choice of percentile
distribution represented by its maximum effluent limitation under the CWA
represents an expert policy judgment that is not arbitrary or capricious.
EPA expects that airports will comply with promulgated limitations at all times. If the
exceedance is caused by an upset condition, the airport would have an affirmative defense to an
enforcement action if the requirements of 40 CFR 122.41(n) are met. If the exceedance is caused
by a design or operational deficiency, EPA has determined that the airport's performance does
not represent the appropriate level of control (best available technology for existing sources; best
available demonstrated technology for new sources). For promulgated limitations and standards,
EPA has determined that such exceedances can be controlled by diligent process and wastewater
treatment system operational practices such as frequent inspection and repair of equipment, use
of backup systems, and operator training and performance evaluations.
14.8 References
EPA. 1987. Organic Chemicals and Plastics and Synthetic Fibers Category Effluent Limitations
Guidelines, Pretreatment Standards and New Source Performance Standards; Final Rule. 40
CFR Part 414, 52 FR 42522, November 5, 1987.
EPA. 1998. National Emission Standards for Hazardous Air Pollutants for Source Category:
Pulp and Paper Production; Effluent Limitations Guidelines, Pretreatment Standards, and New
Source Performance Standards: Pulp, Paper, and Paperboard Category; Final Rule. 40 CFR
Part 430, 63 FR 18503, April 15, 1998.
EPA. 2002. Development Document for Effluent Limitations Guidelines and Standards for the
Iron and Steel Manufacturing Point Source Category. EPA-821-R-02-004. Section 14.1.
www.epa.gov/guide/ironsteel/
EPA. 2006. Notice of Availability of Final 2006 Effluent Guidelines Program Plan.
December 21, 2006; 71 FR 76655
July 2009 14-23
------- |