U.S. Environmental Protection Agency Office of Solid Waste, OSW (renamed Office of Resource Conservation and Recovery, ORCR, on January 18, 2009) November 15, 1996 HAZARDOUS WASTE CHARACTERISTICS SCOPING STUDY TABLE OF CONTENTS Page EXECUTIVE SUMMARY. . . . . . . . . . . . . . . . . . . ES-1 CHAPTER 1. INTRODUCTION. . . . . . . . . . . . . . 1-1 1.1 Purpose and Requirements of the Hazardous Waste Characteristics Scoping Study . . 1-1 1.2 Regulatory Background . . . . . . . . . 1-2 1.3 Approach for Studying Potential Gaps in the Hazardous Waste Characteristics . . 1-4 Step 1: Characterize Releases from Non-Hazardous Industrial Waste Management. . . . . . . . . . 1-4 Step 2: Categorize Risks Associated with Non-Hazardous Industrial Waste Management. . . . . . . . . . 1-4 Step 3: Review the Existing Characteristics . . . . . . . 1-4 Step 4: Identify Gaps Associated with Non-TC Chemicals. . . . . . . 1-6 Step 5: Identify Potential Gaps Associated with Certain Natural Resource Damages and Large-Scale Environmental Problems. . . . . . . . . . . 1-6 Step 6: Review State Expansions of TC and State Listings . . . . 1-7 Step 7: Evaluate the Industries and Waste Management Practices Associated with Potential Gaps. . . . . . . . . . . . . 1-7 Step 8: Assess Regulatory Programs' Coverage of Potential Gaps. . 1-8 Step 9: Present Integrated Evaluation of Nature and Extent of Potential Gaps. . . . . . . . 1-8 1.4 Report Outline. . . . . . . . . . . . . 1-8 CHAPTER 2. RELEASES FROM NON-HAZARDOUS INDUSTRIAL WASTE MANAGEMENT UNITS. . . . . . . . . 2-1 2.1 Methodology . . . . . . . . . . . . . . 2-1 2.1.1 Criteria For Selecting Releases. . . . . . . . . . . 2-1 2.1.2 Approach For Identifying Releases. . . . . . . . . . . 2-3 2.1.2.1 State Industrial D Programs. . . . . . . . . 2-3 2.1.2.2 State Superfund Programs. 2-5 2.1.2.3 Federal Superfund Program 2-6 2.1.2.4 Construction and Demolition (C&D) Landfill Report. . . . . . . . . . 2-6 2.1.3 Release Profile Preparation . 2-7 2.2 Results . . . . . . . . . . . . . . . . 2-8 2.2.1 Number of Cases By State. . . 2-8 2.2.2 Number of Cases By Industry . 2-10 2.2.3 Number of Cases By Type of Waste Management Unit . . . . 2-11 2.2.4 Type of Media Affected. . . . 2-12 2.2.5 Types of Contaminants Released. . . . . . . . . . . 2-13 2.3 Major Limitations . . . . . . . . . . . 2-17 CHAPTER 3. POTENTIAL GAPS ASSOCIATED WITH HAZARDOUS WASTE CHARACTERISTICS DEFINITIONS . . . 3-1 3.1 Types of Risks Addressed by RCRA Hazardous Waste Characteristics . . . . 3-1 3.1.1 Statutory and Regulatory Framework . . . . . . . . . . 3-1 3.1.2 Risks Associated with Physical Hazards. . . . . . . 3-3 3.1.3 Acute Toxic Hazards to Humans 3-4 3.1.4 Chronic Toxicity Risks to Humans. . . . . . . . . . . . 3-4 3.1.5 Risks to Non-Human Receptors. 3-5 3.1.6 Other Risks Associated with Non-Hazardous Industrial Waste Management. . . . . . . . . . 3-6 3.2 Ignitability Characteristic . . . . . . 3-8 3.2.1 Definition of Ignitability. . 3-8 3.2.2 Potential Gaps Related to Definition of Ignitability. . 3-9 3.2.3 Potential Gaps Related to Ignitability Test Methods. . . . 3-12 3.3 Corrosivity . . . . . . . . . . . . . . 3-12 3.3.1 Definition of Corrosivity . . 3-12 3.3.2 Potential Gaps Related to Definition of Corrosivity . . 3-13 3.3.3 Potential Gaps Related to Corrosivity Test Methods. . . 3-16 3.4 Reactivity. . . . . . . . . . . . . . . 3-16 3.4.1 Definition of Reactivity. . . 3-16 3.4.2 Potential Gaps Related to Definition of Reactivity. . . 3-17 3.4.3 Potential Gaps Related to Reactivity Test Methods . . . 3-19 3.5 Potential Gaps Associated with the Toxicity Characteristic . . . . . . . . 3-19 3.5.1 Definition of Toxicity Characteristic. . . . . . . . 3-19 3.5.2 Changes in Groundwater Pathway Analysis. . . . . . . . . . . 3-21 3.5.3 Potential Inhalation Pathway Risks Associated with TC Analytes. . . . . . . . . . . 3-27 3.5.4 Potential Risks from Surface Water Exposures. . . . . . . . 3-33 3.5.5 Potential Indirect Pathway Risks from TC Analytes . . . . 3-35 3.5.6 Potential for Acute Adverse Effects of Exposures to TC Analytes. . . . . . . . . . . 3-38 3.5.7 Potential Risks to Ecological Receptors from TC Analytes. . 3-38 3.6 Potential Gaps Associated with TCLP . . 3-39 3.6.1 TCLP Background . . . . . . . 3-39 3.6.2 Limitations of the TCLP . . . 3-42 CHAPTER 4. POTENTIAL GAPS ASSOCIATED WITH NON-TC CHEMICALS . . . . . . . . . . . . . . . 4-1 4.1 Overview of Methodology . . . . . . . . 4-1 Step 1: Identify and Classify Known Non-Hazardous Industrial Waste Constituents . . . . . . . . . 4-1 Step 2: Identify and Screen Possible Non-Hazardous Industrial Waste Constituents . . . . . . . . . 4-1 Step 3: Apply Hazard-Based Screening Criteria . . . . . . . . . . . 4-3 Step 4: Review Relevant Multipathway Risk Modeling Results. . . . . 4-3 Step 5: Identify Potential Acute Hazards. . . . . . . . . . . . 4-3 Step 6: Summarize Findings . . . . . . 4-3 4.2 Identify and Classify Known Constituents of Non-Hazardous Industrial Wastes. . . 4-3 4.3 Identify Possible Non-Hazardous Industrial Waste Constituents of Potential Concern . . . . . . . . . . . 4-7 4.3.1 Approach to Identifying Potentially Hazardous Chemicals . . . . . . . . . . 4-7 4.3.2 Screening Approach. . . . . . 4-9 4.3.3 Toxicity, Fate, and Transport Screening for Possible Non-Hazardous Industrial Waste Constituents. . . . . . 4-12 4.3.4 Release Volume Screening of Possible Non-Hazardous Industrial Waste Constituents. . . . . . . . . 4-15 4.3.5 Summary of Possible Non-Hazardous Industrial Waste Constituents. . . . . . . . . 4-20 4.4 Combine and Screen Known and Possible Non-Hazardous Industrial Waste Constituents. . . . . . . . . . . . . . 4-20 4.4.1 Combine the Lists . . . . . . 4-20 4.4.2 Screen Combined List Against Single Criteria . . . . . . . 4-25 4.4.3 Screen Combined List Against Multiple Parameters . . . . . 4-33 4.5 Driving Risk Pathways for the Known and Possible Non-Hazardous Industrial Waste Constituents. . . . . . . . . . . . . . 4-34 4.6 Potential Acute Hazards Associated With Known and Possible Non-Hazardous Industrial Waste Constituents . . . . . 4-35 4.7 Identify Individual Chemicals and Classes of Chemicals Constituting Potential Gaps. . . . . . . . . . . . . 4-40 CHAPTER 5. POTENTIAL GAPS ASSOCIATED WITH NATURAL RESOURCE DAMAGES AND LARGE-SCALE ENVIRONMENTAL PROBLEMS. . . . . . . . . 5-1 5.1 Damage to Groundwater Resources . . . . 5-1 5.2 Damage to Local Air Quality from Odors. 5-2 5.3 Large-Scale Environmental Problems. . . 5-4 5.3.1 Air Deposition to the Great Waters. . . . . . . . . . . . 5-4 5.3.2 Airborne Particulates . . . . 5-6 5.3.3 Global Climate Change . . . . 5-7 5.3.4 Potential Damages from Endocrine Disruptors. . . . . 5-9 5.3.5 Red Tides . . . . . . . . . . 5-13 5.3.6 Stratospheric Ozone Depletion 5-14 5.3.7 Tropospheric Ozone and Photochemical Air Pollution . 5-15 5.3.8 Water Pollution . . . . . . . 5-15 CHAPTER 6. STATE EXPANSIONS OF THE TOXICITY CHARACTERISTIC AND LISTINGS . . . . . . 6-1 6.1 State Expanded Toxicity Characteristics 6-1 6.2 State Only Listings . . . . . . . . . . 6-2 6.3 State Restrictions on Exemptions. . . . 6-5 6.4 Summary . . . . . . . . . . . . . . . . 6-6 CHAPTER 7. SUMMARY OF POTENTIAL GAPS . . . . . . . 7-1 7.1 Organization of the Analysis of Potential Gaps. . . . . . . . . . . . . 7-1 7.2 Summary of Potential Gaps . . . . . . . 7-2 CHAPTER 8. POTENTIAL GAPS AS FUNCTION OF INDUSTRY AND WASTE MANAGEMENT METHODS. . . . . . 8-1 8.1 Data Sources and Major Limitations. . . 8-1 8.2 Potential Gaps as a Function of Industry/Waste Source . . . . . . . . . 8-3 8.2.1 Non-Hazardous Industrial Waste Generation by Industry. 8-3 8.2.2 Industries Responsible for Documented Non-Hazardous Industrial Waste Releases . . 8-5 8.2.3 Occurrence of High-Hazard Industrial Waste Constituents by Industry . . . . . . . . . 8-8 8.2.4 Industries Reporting Releases of TC Analytes or Known or Possible Non-Hazardous Industrial Waste Constituents 8-18 8.3 Potential Gaps as a Function of Management Practices. . . . . . . . . . 8-21 8.3.1 Waste Management Practices by Waste Type and Industry . . . 8-21 8.3.2 Management Practices Seen in the Release Descriptions. . . 8-24 8.3.3 Potential Hazards Associated with Use Constituting Disposal. . . . . . . . . . . 8-29 8.3.4 Potential Hazards Associated with Other Management Practices . . . . . . . . . . 8-31 CHAPTER 9. POTENTIAL FOR GAPS TO BE ADDRESSED BY EXISTING REGULATIONS. . . . . . . . . . 9-1 9.1 RCRA Programs . . . . . . . . . . . . . 9-1 9.1.1 Hazardous Waste Programs . . . . 9-1 9.1.2 Subtitle D . . . . . . . . . . . 9-3 9.2 Medium-Specific Regulations . . . . . . 9-3 9.2.1 Clean Water Act. . . . . . . . . 9-4 9.2.2 Safe Drinking Water Act. . . . . 9-7 9.2.3 Clean Air Act Amendments . . . . 9-10 9.3 Federal Insecticide, Fungicide, and Rodenticide Act . . . . . . . . . . . . 9-13 9.4 Toxic Substance Control Act . . . . . . 9-15 9.5 Pollution Prevention. . . . . . . . . . 9-15 9.6 Occupational Safety and Health Act. . . 9-16 9.7 Hazardous Materials Transportation Act. 9-16 9.8 Summary . . . . . . . . . . . . . . . . 9-19 CHAPTER 10. SUMMARY EVALUATION OF NATURE AND EXTENT OF POTENTIAL GAPS . . . . . . . . . . . 10-1 10.1 Overview of the Evaluation of Potential Gaps. . . . . . . . . . . . . . . . . . 10-1 10.1.1 Objectives of the Gaps Analysis. . . . . . . . . . . 10-1 10.1.2 Criteria Used for Evaluating Gaps. . . . . . . . . . . . . 10-1 10.2 Findings of the Evaluation. . . . . . . 10-3 10.2.1 Potential Gaps Associated with the ICR Characteristics. 10-3 10.2.2 Potential Gaps Associated with TC Analytes. . . . . . . 10-8 10.2.3 Potential Gaps Associated with Non-TC Waste Constituents. . . . . . . . . 10-14 10.2.4 Potential Gaps Associated With Resource Damage and Large-Scale Environmental Problems. . . . . . . . . . . 10-22 10.2.5 Gaps Associated with State TC Expansions and Listings . . . . 10-25 10.2.6 Major Data Gaps and Uncertainties . . . . . . . . . 10-26 10.3 Framework for Determining an Appropriate Course of Action. . . . . . . . . . . . 10-27 10.3.1 Step 1: Identify Critical Research Needs and Next Steps Necessary to Analyze Key Issues and Fill Major Data Deficiencies. . . . 10-27 10.3.2 Step 2: Identify and Evalu- ate Options to Address Any Clearly Identified Gaps. . . 10-27 LIST OF EXHIBITS Page Exhibit 1-1 Scoping Study Approach. . . . . . . . . 1-5 Exhibit 2-1 Number of Release Descriptions By State 2-9 Exhibit 2-2 Number of Management Units & Volume of Waste Managed On-Site, by State (1985). 2-9 Exhibit 2-3 Number of Case Studies by Industry (SIC) . . . . . . . . . . . . . . . . . 2-10 Exhibit 2-4 Number of Case Studies By Waste Management Unit . . . . . . . . . . . . 2-12 Exhibit 2-5 TC Contaminants Detected in Case Studies . . . . . . . . . . . . . . . . 2-14 Exhibit 2-6 Contaminants with SMCLs Detected in Case Studies. . . . . . . . . . . . . . 2-15 Exhibit 2-7 Other Contaminants Detected in At Least Three Case Studies. . . . . . . . . . . 2-16 Exhibit 2-8 Most Common Constituents By Industry. . 2-18 Exhibit 3-1 Risks Potentially Associated with Non-Hazardous Industrial Waste Management. . . . . . . . . . . . . . . 3-7 Exhibit 3-2 Materials Formerly Classified by DOT as Combustible Liquids (which generally are not RCRA ignitable) . . . . . . . . 3-11 Exhibit 3-3 Other Definitions of Reactivity . . . . 3-18 Exhibit 3-4 TC Constituents and Regulatory Levels (mg/l). . . . . . . . . . . . . . . . . 3-20 Exhibit 3-5 Comparison of TC Regulatory Concentra- tions and HWIR-Waste Proposed Exit/ Leech Levels. . . . . . . . . . . . . . 3-25 Exhibit 3-6 Summary of Inhalation Pathway Screening Methods, Input Data, and Models Used for Bounding Risk Analysis. . . . . . . 3-29 Exhibit 3-7 Emission Fraction for Air Releases of Volatile TC Analytes. . . . . . . . . . 3-30 Exhibit 3-8 Inhalation Pathway Risks for TC Analytes and Their Dependence on Fate and Transport Properties. . . . . . . . 3-32 Exhibit 3-9 Major Fate and Transport Parameters for TC Analytes . . . . . . . . . . . . . . 3-36 Exhibit 3-10 Ratios of TC Leachate Regulatory Levels to Ambient Water Quality Criteria for Aquatic Life. . . . . . . . . . . . . . 3-40 Exhibit 4-1 Flow Chart of Procedures Used to Identify Non-TC Chemicals Posing Potential Gaps in the TC Characteristics . . . . . . . . . . . . 4-2 Exhibit 4-2 Known Non-Hazardous Industrial Waste Constituents Found in Case Studies, ISDB, Listings Documents, and Effluent Guidelines by Chemical Class. . . . . . 4-6 Exhibit 4-3 Lists Used to Identify Possible Non-Hazardous Industrial Waste Constituents. . . . . . . . . . . . . . 4-8 Exhibit 4-4 Criteria Considered for Screening Non-Hazardous Industrial Waste Constituents. . . . . . . . . . . . . . 4-10 Exhibit 4-5 Toxicity Screening Results for Possible Non-Hazardous Industrial Waste Constituents. . . . . . . . . . . . . . 4-13 Exhibit 4-6 Persistence and Bioconcentration/ Bioaccumulation Screening Results for Possible Non-Hazardous Industrial Waste Constituents. . . . . . . . . . . . . . 4-16 Exhibit 4-7 Screening of High-Toxicity, Persistent, Bioaccumulative/Bioconcentrating Possible Non-Hazardous Industrial Waste Constituents Against TRI Release Volumes . . . . . . . . . . . . . . . . 4-17 Exhibit 4-8 Possible Non-Hazardous Industrial Waste Constituents by Chemical Class. . . . . 4-21 Exhibit 4-9 Screening of Known Non-Hazardous Industrial Waste Constituents Against TRI Release Volumes . . . . . . . . . . 4-22 Exhibit 4-10 Toxicity Summary of Known and Possible Non-Hazardous Industrial Waste Constituents. . . . . . . . . . . . . . 4-26 Exhibit 4-11 Potential Endocrine Disruptors. . . . . 4-27 Exhibit 4-12 TRI Releases and Non-Confidential TSCA Production Volume Data for the Known and Possible Non-Hazardous Industrial Waste Constituents. . . . . . . . . . . 4-28 Exhibit 4-13 Volatility, Persistence, and Bio- accumulation/Bioconcentration Summary Potential of Known and Possible Non-Hazardous Industrial Waste Constituents. . . . . . . . . . . . . . 4-29 Exhibit 4-14 LNAPL/DNAPL Formation Potential of Known and Possible Non-Hazardous Industrial Waste Constituents . . . . . 4-32 Exhibit 4-15 Multiple Screening Criteria Applied to Known and Possible Non-Hazardous Industrial Waste Constituents . . . . . 4-34 Exhibit 4-16 Lowest Proposed HWIR-Waste Exit Levels for Known and Possible Non-Hazardous Industrial Waste Constituents . . . . . 4-36 Exhibit 4-17 Potential Acute Hazards Associated with Known and Possible Non-Hazardous Industrial Waste Constituents . . . . . 4-39 Exhibit 4-18 Potential Gaps in the Hazardous Waste Characteristics Identified Based on the Hazardous Properties of Known and Possible Non-Hazardous Industrial Waste Constituents. . . . . . . . . . . . . . 4-41 Exhibit 5-1 Constituents/Properties with SMCLs Found in Release Descriptions. . . . . . . . . . 5-2 Exhibit 5-2 Chemicals from Release Descriptions with Low Odor Thresholds. . . . . . . . 5-3 Exhibit 5-3 Initial List of Large-Scale Environmental Problems. . . . . . . . . 5-4 Exhibit 5-4 U.S. Sources of Air Pollutants of Concern for Great Waters. . . . . . . . 5-5 Exhibit 6-1 State Toxicity Characteristics: Additional Constituents and More Stringent Regulatory Levels . . . . . . 6-3 Exhibit 6-2 State Toxicity Criteria Applied to Whole Waste (Representative Sample) . . 6-4 Exhibit 7-1 Summary of Potential Gaps in the Hazardous Waste Characteristics . . . . 7-2 Exhibit 8-1 Estimated Generation of Non-Hazardous Industrial Waste by Major Industry Group . . . . . . . . . . . . . . . . . 8-4 Exhibit 8-2 Chemicals Exceeding Health-Based and Non-Health-Based Regulatory Levels in the Release Descriptions for Non-Hazardous Waste Management. . . . . 8-6 Exhibit 8-3 Numbers of Chemical Detections and Frequencies of Regulatory Exceedences in Release Descriptions . . . . . . . . 8-7 Exhibit 8-4 Most Frequently Occurring Constituents in the Release Descriptions . . . . . . 8-9 Exhibit 8-5 Occurrence of Waste Constituents by Industry Group. . . . . . . . . . . . . 8-11 Exhibit 8-6 Non-Hazardous Industrial Waste Constituents Reported Released by Industry. . . . . . . . . . . . . . . . 8-19 Exhibit 8-7 Volume of Non-Hazardous Industrial Waste Managed in Land-Based Facilities in 1985 . . . . . . . . . . . . . . . . 8-22 Exhibit 8-8 Active Non-Hazardous Industrial Waste Management Units in 1985 by Major Industry Group. . . . . . . . . . . . . 8-23 Exhibit 8-9 Non-Hazardous Industrial Waste Management by Industry and Waste Type from TSDR and ISDB. . . . . . . . . . . 8-25 Exhibit 8-10 Waste Management Unit Types in the Release Descriptions . . . . . . . . . 8-30 Exhibit 9-1 TC Constituents with Effluent Limits Established under CWA . . . . . . . . . 9-5 Exhibit 9-2 CWA Effluent Limitations Relevant to Certain Known Non-Hazardous Industrial Waste Constituents. . . . . . . . . . . 9-6 Exhibit 9-3 CWA Coverage of Industries Represented in Release Descriptions . . . . . . . . 9-7 Exhibit 9-4 TC Constituents with SDWA MCL Levels. . 9-8 Exhibit 9-5 MCLs for Known Non-Hazardous Industrial Waste Constituents of Concern in Groundwater Pathways. . . . . . . . . . 9-9 Exhibit 9-6 TC Constituents Designated as HAPs under CAA . . . . . . . . . . . . . . . 9-11 Exhibit 9-7 CAA Hazardous Air Pollutants (HAPs) Specified for Potential Gap Constituents. . . . . . . . . . . . . . 9-12 Exhibit 9-8 CAA Coverage of Industries Represented in Release Descriptions . . . . . . . . 9-13 Exhibit 9-9 Status of Pesticides That are TC Analytes or Known Non-Hazardous Industrial Waste Constituents . . . . . 9-14 Exhibit 9-10 TC Constituents with Established OSHA PELs. . . . . . . . . . . . . . . . . . 9-17 Exhibit 9-11 OSHA PELs Specified for Known Non-Hazardous Industrial Waste Constituents. . . . . . . . . . . . . . 9-18 Exhibit 9-12 Potential Gaps and Potential Non-RCRA Regulatory Control. . . . . . . . . . . 9-19 Exhibit 10-1 Evaluation of Potential Gaps Associated With the Ignitability, Corrosivity, and Reactivity (ICR) Characteristics. . . . 10-4 Exhibit 10-2 Evaluation of Potential Gaps Associated with Toxicity Characteristic Analytes and TCLP. . . . . . . . . . . . . . . . 10-9 Exhibit 10-3 Evaluation of Potential Gaps Associated with Non-TC Chemicals . . . . . . . . . 10-15 Exhibit 10-4 Evaluation of Potential Gaps Associated With Certain Large-Scale Environmental Problems. . . . . . . . . . . . . . . . 10-24 HAZARDOUS WASTE CHARACTERISTICS SCOPING STUDY: EXECUTIVE SUMMARY U.S. Environmental Protection Agency Office of Solid Waste, OSW (renamed Office of Resource Conservation and Recovery, ORCR, on January 18, 2009) November 15, 1996 EXECUTIVE SUMMARY The U.S. Environmental Protection Agency (EPA), Office of Solid Waste, OSW (renamed Office of Resource Conservation and Recovery, ORCR, on January 18, 2009) has investigated potential gaps in the current hazardous waste characteristics promulgated under the federal Resource Conservation and Recovery Act (RCRA). This report, the Hazardous Waste Characteristics Scoping Study, presents the findings of that investigation. THE SCOPING STUDY: AN EARLY STEP This study is a first step for the Agency in fulfilling a long-standing goal to review the adequacy and appropriateness of the hazardous characteristics. The study also fulfills an obligation in a consent decree with the Environmental Defense Fund (EDF). The study is by design a scoping study and, therefore, does not conclusively identify particular chemical classes for regulation, or fundamental flaws in the overall regulatory framework requiring immediate regulatory action. However, the study does identify several key areas that merit further analysis due to the significant potential for improving hazardous waste management practices and protection to health and the environment. Thus, the scoping study provides a catalogue of potential gaps in the hazardous waste characteristics. The Agency considers that this study is one very critical component of a broader array of efforts underway to review and improve the RCRA program, to ensure that regulation is appropriate to the degree of risk posed by hazardous wastes and waste management practices. Efforts involve both regulatory and de-regulatory actions, as appropriate for specific wastes and waste management practices. STUDY PROCESS AND FINDINGS Review of Current Characteristics The review of the current characteristic regulations evaluated the protectiveness of the characteristics against the risks they were intended to address and also risks they were not specifically intended to address. For example, EPA evaluated risks that are now addressed by the Toxicity Characteristic (TC), e.g., direct ingestion of groundwater, by considering new groundwater modeling techniques that have been in use since the promulgation of the current TC levels, as well as any changes to the toxicity values on which the original levels were based. In addition, EPA evaluated risks from other exposure pathways and to ecological receptors, which are both risks not intended to be protected by the original TC. The review of the current TC regulatory levels suggests that: (1) further analysis of the current TC regulatory levels should be conducted using new groundwater modeling techniques, as well as considering changes to toxicity values for specific constituents; and (2) non-groundwater pathways and ecological receptors--not currently addressed by TC provisions--may be of potential concern. The study included some screening analyses of potential air releases from surface impoundments and land application units. The Agency found that inhalation risk levels for a significant number of current TC constituents at the fenceline (under certain exposure conditions) exceeded the allowable risk levels upon which the TC is based. Waste piles and land application units may be of special concern for ecological receptors due to surface runoff. Thirteen TC constituents have regulatory levels that are 10,000 or more times higher than Ambient Water Quality Criteria concentrations, with four of these being at least 100,000 times higher, suggesting that the level of protectiveness of the TC may not be very high for ecological receptors. The study also identifies the need to examine a broader array of leaching procedures, in addition to the Toxicity Characteristic Leaching Procedure (TCLP), to better predict environmental releases from various waste types and waste management conditions. Notable examples are the inability of the TCLP to predict significant releases under highly alkaline conditions or to media other than groundwater, or to serve as a leaching procedure for oily wastes. The most obvious potential gap identified for the ignitability and reactivity characteristics is the reference to outdated DOT regulations. Other potential gaps identified for these characteristics include the exclusion of combustible liquids and lack of specific test methods for non-liquids for ignitability; exclusion of corrosive solids, not addressing corrosion of non-steel materials and solubilization of non-metals, and whether pH limits are adequately protective for corrosivity; and, an overly-broad definition and lack of specific test methods for reactivity. Releases from Non-Hazardous Industrial Waste Facilities The Agency identified actual releases of non-hazardous waste constituents as one means of finding potential problem constituents and management activities. EPA reviewed data on non-hazardous industrial waste management activities that was readily available from state monitoring and compliance files. The Agency focused on wastes that are not currently regulated as hazardous (by virtue of being listed or exhibiting a characteristic) to identify releases potentially causing human health or environmental damages. The Agency considered three major factors in judging whether a release was an appropriate case study for this evaluation. A release had to meet all three of the following criteria to be included: (1) The source of contamination had to be a waste management unit or other intended final disposal area that received only non-hazardous industrial waste; (2) A release from a waste management unit must have caused contamination at levels of potential concern (constituent-specific concentrations that exceed federal standards or state guidelines or regulations); and, (3) Documented evidence must be available to support the exceedences referred to in (2). EPA found 112 environmental release case studies in 12 states with readily available (and not necessarily representative) data on non-hazardous waste management units. The releases were found from facilities in 15 (2-digit) Standard Industry Classification (SIC) industries. The top four categories were: SIC 49: Electric, Gas, and Sanitary Services (refuse-side only); SIC 26: Paper & Allied Products; SIC 28: Chemical & Allied Products; and, SIC 20: Food & Kindred Products. Over 90 percent of the releases were from landfills or surface impoundments and nearly all (98 percent) involved groundwater contamination. This is most likely because groundwater monitoring is the most common method for detecting releases from waste management units. Many of the chemical constituents most commonly detected above a regulatory level are already addressed in the current TC, even though the release occurred from non-hazardous waste management. The 20 constituents most commonly detected above a regulatory level are inorganics. The constituents that exceeded state groundwater protection standards or health-based federal drinking water standards most frequently were lead, chromium, cadmium, benzene, arsenic and nitrates. All of these, with the exception of nitrates, are current TC constituents. Organic constituents, both TC and non-TC, were also identified in the case studies, however, they were detected less frequently than the inorganic toxicity characteristic constituents. This collection of release descriptions is not statistically representative of problem industries nor intended to identify particular problem facilities. The Agency believes that the case studies are indicative of the type of releases associated with the management of non-hazardous wastes in the types of facilities identified. The Agency also believes that information on releases from past waste management practices is useful in demonstrating the potential for human health or environmental damage. Non-TC Chemical Constituents In reviewing chemicals and chemical classes not currently regulated by the TC, EPA found in excess of 100 constituents that potentially occur in waste and may pose significant risks. EPA reviewed 37 regulatory or advisory lists of chemicals to identify possible constituents of non-hazardous wastes. EPA also compiled a list of chemicals which are "known" to be constituents of non-hazardous wastes because they were identified in the environmental release case studies or other Agency data sources on non-hazardous industrial wastes. EPA screened these chemicals and narrowed the list to possible constituents of non-hazardous waste that, by virtue of their toxicity, fate and transport properties, or exposure potential, could pose significant risks to human health and/or the environment. These chemicals were both inorganics and organics, and include volatiles, non-volatile organics, PAHs and pesticides. Because of the large number of constituents identified as candidates and the limited time available for the scoping study, no risk analyses were conducted. However, it may be a reasonable next step to assess the potential risks for a subset of these constituents. Natural Resource Damages/Large-Scale Environmental Problems The Agency examined the potential for broad environmental impacts from non-hazardous waste management. These impacts may include damages to natural resources which diminish the value and usability of a resource without threatening human health, as well as possible contributions to regional and global environmental problems. With respect to groundwater contamination, over 80 percent of the facilities identified in the case studies discussed earlier had releases exceeding secondary drinking water standards (non-health based standards). These releases were identified because exceedence of secondary standards may reduce the useability and, therefore, the value of the groundwater. Iron, chloride, sulfate and manganese were among the most frequently detected constituents exceeding secondary standards. In reviewing air deposition of toxic constituents to great waters, the Agency found a number of TC constituents, as well as some other chemicals identified in the study. However, it was not possible to assess the importance of waste to air deposition of toxics to the great waters. State-Only Hazardous Waste Regulations Some states have adopted hazardous waste identification rules that are broader or more stringent than federal RCRA Subtitle C regulations. These expansions reflect state judgements about gaps in the federal program. Data on hazardous waste regulations from eight states, California, Michigan, New Hampshire, Oregon, Rhode Island, Texas, Washington, and New Jersey were considered. Several states regulate additional constituents beyond the TC list ( 25 for California, 9 for Michigan, and 1 for Washington). California also applies a more aggressive leaching test, the waste extraction test (WET) to wastes. California also has a test for combinations of hazardous constituents, in which a combined concentration of the listed constituents cannot exceed 0.001 percent as a total in the waste. Four states also apply acute toxicity values (LD50 or LC50) for human or ecological toxicity to the whole waste. NEXT STEPS The potential gaps and areas of health and environmental concern identified here will require further, more detailed examination before regulatory action can be undertaken. For example, the study highlights risks to ecological receptors and possible inhalation risks to humans as potential gaps, as well as further evaluation of the adequacy of the TCLP. These topics were found to be potential gaps in more than one area of the study and will likely be specific areas of further investigation. Following release of this report, the Agency will engage in a variety of outreach activities in identifying appropriate next steps. While the Agency considers this a final report, comments from interested members of the public are solicited and will be used to help identify and structure follow-on activities. As noted above, revisions to the characteristics program will likely, in the long run, involve both regulatory and de-regulatory activities. CHAPTER 1. INTRODUCTION The U.S. Environmental Protection Agency (EPA), Office of Solid Waste, OSW (renamed Office of Resource Conservation and Recovery, ORCR, on January 18, 2009) has investigated potential gaps in the current hazardous waste characteristics promulgated under the federal Resource Conservation and Recovery Act (RCRA). This report, the Hazardous Waste Characteristics Scoping Study, presents the findings of that investigation. Chapter 1 presents background information on the Scoping Study as follows: ! Section 1.1 describes the purpose and scope of the Scoping Study; ! Section 1.2 discusses relevant aspects of the RCRA hazardous waste and non-hazardous waste programs; ! Section 1.3 summarizes the methodology used to prepare the Scoping Study; and ! Section 1.4 outlines the remaining chapters and appendices of the Study. 1.1 Purpose and Requirements of the Hazardous Waste Characteristics Scoping Study Agreement for Hazardous Waste Characteristics Scoping Study The Administrator shall perform a study of potential gaps in the coverage of the existing hazardous waste characteristics. The purpose of the study is to investigate if there are gaps in coverage, and the nature and extent of the gaps identified. The potential gaps in coverage to be addressed in the study [shall] incorporate both waste management practices and possible impacts to human health and the environment. With respect to waste management practices, the study shall, at a minimum, address releases from non-hazardous waste surface impoundments; waste piles; land treatment units; landfills; and various forms of use constituting disposal such as road application, dust suppression or use in a product applied to the land. Human health and environmental impacts to be addressed by the study shall include, but not be limited to: (a) impacts via non-groundwater exposure pathways, both direct and indirect, to human and ecological receptors; (b) impacts via the groundwater pathway to ecological receptors; (c) the potential for formation of non-aqueous phase liquids in groundwater; and (d) impacts via the groundwater pathway to human receptors caused by releases of toxic constituents not included in the current toxicity characteristic, such as EPA-classified carcinogens, priority pollutants identified in the Clean Water Act, and solvents used for purposes other than degreasing. The Administrator shall complete the study by November 15, 1996, and shall provide the plaintiff with two copies of the study immediately upon completion. Environmental Defense Fund, Inc. v. Browner, Civ. No. 89-0598, order granting stipulated motion of EDF and EPA for amendment of consent decree. May 17, 1996, pp. 18-19. 1 As stipulated under an amended consent decree with the Environmental Defense Fund (EDF) (presented in the text box below), the Agency has investigated potential gaps in the coverage of the existing RCRA hazardous waste characteristics. The purpose of this Study is to identify potential gaps in coverage and to investigate the nature and extent of such gaps. Based on the results of the Study, EPA will seek input from interested parties and determine the appropriate course of action to further address any significant potential gaps identified in the Study. 1.2 Regulatory Background This report focuses on wastes that are not currently regulated as hazardous (by virtue of being listed or exhibiting a characteristic). Industrial wastes are classified either as "hazardous waste" and managed under Subtitle C of the Resource Conservation and Recovery Act (RCRA) or as "non-hazardous waste" and managed under Subtitle D of RCRA, primarily under state programs. In the context of this report, the term "non-hazardous industrial waste" broadly refers to waste that is neither municipal solid waste, special waste, nor considered a hazardous waste under Subtitle C of RCRA. A brief description of the Agency's hazardous and non-hazardous waste classification systems is provided below. Subtitle C of RCRA, as amended, establishes a federal program for the comprehensive regulation of hazardous waste. Section 1004(7) of RCRA defines hazardous waste as "a solid waste, or a combination of solid wastes, which because of its quantity, concentration, or physical, chemical, or infectious characteristics may: (a) cause, or significantly contribute to an increase in mortality or an increase in serious irreversible, or incapacitating reversible, illness; or (b) pose a substantial present or potential hazard to human health or the environment when improperly treated, stored, transported, disposed of, or otherwise managed." Under RCRA Section 3001, EPA is charged with defining which solid wastes are hazardous by identifying the characteristics of hazardous waste and listing particular hazardous wastes. Current hazardous waste characteristics are ignitability, corrosivity, reactivity, and toxicity. The Agency's definitions of ignitability and reactivity have not changed materially since their adoption in 1980.1 The Agency's definition for corrosivity was last revised in 1993.2 The Agency's current definition of toxicity was promulgated in 1990,3 replacing the Extraction Procedure (EP) leach test with the Toxicity Characteristic Leaching Procedure (TCLP) and adding 25 organic chemicals to the list of toxic constituents of concern and establishing their regulatory levels. The Agency's definition of toxicity was last revised in 1993;4 however, this revision did not alter the framework for defining this characteristic. A solid waste is classified as listed hazardous waste if it is named on one of the following four lists developed by EPA: ! Nonspecific source or F wastes (40 CFR 261.31). These are generic wastes, commonly produced by manufacturing and industrial processes. Examples include spent halogenated solvents used in degreasing and wastewater treatment sludge from electroplating processes as well as dioxin wastes, most of which are "acutely hazardous" wastes due to the danger they present to human health or the environment. ! Specific source or K wastes (40 CFR 261.32). This list consists of wastes from specifically identified industries such as wood preserving, petroleum refining, and organic chemical manufacturing. These wastes typically include sludges, still bottoms, wastewaters, spent catalysts, and residues. ! Discarded commercial chemical products or P and U wastes (40 CFR 261.33(e) and (f)). The third and fourth lists consist of specific commercial chemical products and manufacturing chemical intermediates. They include chemicals such as chloroform and creosote, acids such as sulfuric acid and hydrochloric acid, and pesticides such as DDT and kepone. Disposal of non-hazardous solid waste is regulated under Subtitle D of RCRA. Subtitle D wastes include municipal solid waste, special waste, and industrial waste. ! Municipal solid waste includes household and commercial solid waste. Household waste is defined as any solid waste (including garbage, trash, and sanitary waste in septic tanks) derived from households (including single and multiple residences, hotels and motels, bunkhouses, ranger stations, crew quarters, campgrounds, picnic grounds, and day-use recreation areas) (40 CFR 258.2). Commercial waste refers to all types of solid waste generated by stores, offices, restaurants, warehouses, and other non-manufacturing activities, excluding residential and industrial wastes (40 CFR 258.2). ! Special waste, as used in this document, refers to oil and gas exploration and production waste, fossil fuel combustion wastes, cement kiln dust, and solid waste from the extraction, beneficiation, and processing of ores and minerals (40 CFR 261.4). ! Non-hazardous industrial waste refers to solid waste generated by manufacturing or industrial processes that is not a hazardous waste regulated under Subtitle C of RCRA or a special waste (40 CFR 258.2). Under Subtitle D, the management of non-hazardous industrial waste in land-based units must comply with 40 CFR Part 257, which establishes minimum federal standards for the management and siting of land-based units. Individual states are responsible for implementing 40 CFR Part 257 under their own authority. They have adopted statutory and regulatory frameworks for management of non-hazardous industrial wastes. These requirements vary widely from one state to another in terms of their design and operating requirements, monitoring requirements, and other management requirements such as recordkeeping, closure, post-closure care, and financial responsibility. Even within a given state, the non-hazardous industrial waste requirements may vary from facility to facility depending on the characteristics of the wastes managed and the environmental setting of the waste management unit. The Agency is currently developing "voluntary guidelines" for non-hazardous industrial waste management to better ensure that this waste is managed in a manner that is protective of human health and the environment. 1.3 Approach for Studying Potential Gaps in the Hazardous Waste Characteristics As shown in Exhibit 1-1, the general approach EPA used to perform the Scoping Study has nine steps. Each of these steps is discussed below. Step 1: Characterize Releases from Non-Hazardous Industrial Waste Management The Agency conducted detailed investigations to identify specific instances of environmental contamination resulting from the management of non-hazardous industrial wastes. These case studies provide real-world information on releases of these wastes into the environment, the chemicals released and their concentrations, and the waste management practices and industries involved. The preliminary findings of such research were presented in a draft report entitled "Hazardous Waste Characteristics Scoping Study: Environmental Release Descriptions" (September 24, 1996). EPA held a public meeting on October 10, 1996 to explain and obtain comments on the draft report. EPA has considered and, where appropriate, incorporated these comments in preparing this Scoping Study. Chapter 2 summarizes these investigations and Appendix A presents the individual environmental release descriptions. Step 2: Categorize Risks Associated with Non-Hazardous Industrial Waste Management This step identifies categories of risks to human health and the environment that may result from non-hazardous industrial waste management. The underlying premise of this step is that a gap in the hazardous waste characteristics is any significant risk to human health or the environment associated with non-hazardous industrial waste management that could be, but is not, addressed by the current characteristics. Thus, this assessment deals with both: ! Hazards that the current hazardous waste characteristics were intended to address, namely physical hazards such as fire and explosion and toxic groundwater contamination near waste management facilities; and ! Hazards that the characteristics were not intended to address, such as non-groundwater pathway exposures to toxins, damages to ecological receptors, and natural resource damages. EPA identified risks by types of receptors, types of toxic effects and physical hazards, exposure pathways, and time and spatial scales, as described in Section 3.1. The search for potential risks used broad definitions of risk and adverse effects and addressed all aspects of non-hazardous industrial waste management, without any prejudgment as to the likelihood that a risk was significant, whether it could be best addressed by the characteristics, or whether it was already addressed by other regulations. The results of this risk classification step were used in identifying and evaluating potential gaps, as described below. Step 3: Review the Existing Characteristics The identification of potential gaps continues with a review of the existing definitions of the characteristics. This step is next for two reasons. First, limitations in the characteristics' effectiveness in reducing the risks they were intended to address may constitute important potential gaps. When the characteristics were promulgated, the Agency identified physical hazards and acute toxic hazards during transport and disposal activities and chronic exposure to groundwater contaminated with waste Insert Exhibit 1-1 Scoping Study Approach constituents as being among the most important waste management risks. Reducing these risks remains an important goal of the characteristics. Second, this analysis lays the groundwork for evaluating other potential gaps. Step 3 begins by examining the definitions and test methods of the ignitability, corrosivity, and reactivity (ICR) characteristics, which are essentially unchanged since they were promulgated in 1980. EPA reviewed the assumptions and approaches used to develop these characteristics and compared the characteristics to approaches taken to controlling similar hazards under other federal and state regulatory schemes. Step 3 also examines the definition of the toxicity characteristic (TC), which was designed to protect against human health risks from exposure to hazardous waste constituents released to groundwater. EPA reviewed new information on the toxicity, fate, and transport of the TC constituents and improvements in groundwater modeling since the TC was revised in 1990. The Agency also examined the potential risks from TC constituents through inhalation, surface water, and indirect pathways and to ecological receptors. Chapter 3 describes these analyses. Step 4: Identify Gaps Associated with Non-TC Chemicals Potential gaps in the hazardous characteristics from non-TC chemicals are identified by, first, identifying two groups of constituents: ! "Known" non-hazardous industrial waste constituents: constituents "known" to be present in non-hazardous industrial wastes, based on the data gathered in the environmental release descriptions in Step 2, EPA's 1987 Telephone Screening Survey of non-hazardous industrial waste management facilities, EPA effluent guideline development documents, and recent hazardous waste listing determinations. ! "Possible" non-hazardous industrial waste constituents: constituents on various regulatory or advisory lists, which were screened for their toxicity, fate, and transport properties and for a proxy of their occurrence in non-hazardous industrial waste, using available environmental release data from the 1994 Toxics Release Inventory. Then, these two lists of constituents are evaluated and compared and chemicals are classified by physical properties, chemical composition, use, and origin. Finally, potential gaps were identified by applying multiple hazard-based screening criteria to specific chemicals and chemical classes. Chapter 4 describes these analyses. Step 5: Identify Potential Gaps Associated with Certain Natural Resource Damages and Large-Scale Environmental Problems As discussed above, steps 3 and 4 respectively examine potential gaps inherent in the current hazardous waste characteristics and associated with adverse human health or localized ecological effects from constituents not addressed by the toxicity characteristic. Step 5 addresses a third set of risks associated with non-hazardous industrial waste management: damages to natural resources that may not have direct human health or ecological effects, and large-scale environmental problems. The specific risks addressed are: C Pollution of groundwater by constituents that diminish the value and usability of the resource without threatening human health; C Air pollution through odors that harm the quality of life but may not have severe health effects; and C Large-scale environmental problems, including air deposition to the Great Waters, damages from endocrine disruptors and airborne particulates, global climate change, red tides, stratospheric ozone depletion, tropospheric ozone and photochemical air pollution and water pollution. Chapter 5 presents these analyses. Step 6: Review State Expansions of TC and State Listings Several states have expanded their hazardous waste management programs to regulate as hazardous certain wastes or waste constituents that are not hazardous under the federal program. Step 6 examines how states have expanded their toxicity characteristics and have listed as hazardous certain wastes that are not listed under the federal program. (Step 3 examines how states have regulated additional wastes by expanding their ICR characteristics.) These expansions beyond the federal hazardous waste identification rules reflect state judgments about gaps in the federal hazardous waste program and thereby constitute potential gaps that may merit further investigation. Chapter 6 presents this analysis. (Chapter 7 summarizes the potential gaps identified in Chapters 3 through 6.) Step 7: Evaluate the Industries and Waste Management Practices Associated with Potential Gaps The evaluation of potential gaps asks two basic questions: (1) What do the qualitative and quantitative indicators of risk show about the potential gaps? and (2) To what extent are the risks associated with the potential gaps addressed by other regulations? Steps 7, 8, and 9 address these questions. Step 7 addresses aspects of the first question. Specifically, it assesses the following: ! The amount of non-hazardous industrial wastes generated by various industries; ! The frequency with which various chemicals were detected or reported in releases from various industries; ! The management methods associated with the major non-hazardous industrial waste generators; and ! The management practices associated with documented environmental releases of non-hazardous industrial wastes. Because of data limitations, EPA could not evaluate all potential gaps against all of these criteria. Instead, this step focuses principally on the potential gaps identified in Steps 3 and 4. Chapter 8 presents this analysis. Step 8: Assess Regulatory Programs' Coverage of Potential Gaps The second major issue in evaluating potential gaps is the extent to which the risks are controlled by existing regulatory or other environmental programs. As noted above, risk-related gaps were identified solely in terms of their relationship to non-hazardous industrial waste management, and not with regard to whether they might be controlled under regulatory or other programs. Chapter 9 discusses how major federal and state regulatory programs may address some of the risks represented by the potential gaps. To the extent that they are already addressed or could be addressed more effectively by programs other than the hazardous waste regulations, the potential gaps may not merit further attention by the RCRA Subtitle C program. Step 9: Present Integrated Evaluation of Nature and Extent of Potential Gaps In the final step of the methodology, which is presented in Chapter 10, EPA integrates and summarizes all of the lines of evidence relating to particular potential gaps in the hazardous waste characteristics. The summary is presented in the form of several tables. This section also reviews the major data gaps and uncertainties of the analysis. 1.4 Report Outline This Scoping Study is organized in the same order as the methodology outlined above. Chapter 2 characterizes releases from non-hazardous industrial waste management; Chapter 3 categorizes risks associated with potential gaps in the characteristics and reviews the existing characteristics to identify potential gaps; Chapter 4 identifies potential gaps associated with non-TC chemicals; Chapter 5 identifies potential gaps associated with certain natural resource damages and large-scale environmental problems; Chapter 6 identifies potential gaps in the characteristics by reviewing how selected states have expanded the TC and listed wastes that are not listed as hazardous under the federal program; Chapter 7 summarizes the potential gaps identified in Chapters 3 through 6; Chapter 8 evaluates the extent of the risks presented by potential gaps; Chapter 9 discusses how major federal and state regulatory programs address the risks represented by the potential gaps; and Chapter 10 presents an integrated summary evaluation of the nature and extent of potential gaps and the associated major analytical limitations and describes the framework that the Agency will apply in developing a plan for addressing potential gaps in the hazardous waste characteristics identified in this Study. The Study also includes several appendices. Appendix A describes the individual environmental releases summarized in Chapter 2. Appendix B discusses several data sources used to identify environmental releases that were not successful in finding releases meeting EPA's stringent selection criteria. Appendix C provides a detailed comparison of the ICR characteristics to related approaches under other federal and state programs. Finally, a separate background document contains detailed information and analysis that supplements the screening-level risk analysis presented in Chapter 3 and the identification of "possible" non-hazardous industrial waste constituents in Chapter 4. CHAPTER 1. INTRODUCTION 1-1 1.1 Purpose and Requirements of the Hazardous Waste Characteristics Scoping Study 1-1 1.2 Regulatory Background 1-2 1.3 Approach for Studying Potential Gaps in the Hazardous Waste Characteristics 1-4 Step 1: Characterize Releases from Non-Hazardous Industrial Waste Management 1-4 Step 2: Categorize Risks Associated with Non-Hazardous Industrial Waste Management 1-4 Step 3: Review the Existing Characteristics 1-4 Step 4: Identify Gaps Associated with Non-TC Chemicals 1-6 Step 5: Identify Potential Gaps Associated with Certain Natural Resource Damages and Large-Scale Environmental Problems 1-6 Step 6: Review State Expansions of TC and State Listings 1-7 Step 7: Evaluate the Industries and Waste Management Practices Associated with Potential Gaps 1-7 Step 8: Assess Regulatory Programs' Coverage of Potential Gaps 1-8 Step 9: Present Integrated Evaluation of Nature and Extent of Potential Gaps 1-8 1.4 Report Outline 1-8 Exhibit 1-1 Scoping Study Approach 1-5 1 45 Federal Register 33084, May 19, 1980. 2 58 Federal Register 46049, August 31, 1993. 3 55 Federal Register 26987, June 29, 1990. 4 58 Federal Register 46049, August 31, 1993. Page 1-1 CHAPTER 2. RELEASES FROM NON-HAZARDOUS INDUSTRIAL WASTE MANAGEMENT UNITS This chapter presents the methodology and results of the Agency's efforts to identify contamination resulting from the management of non-hazardous industrial wastes. The Agency prepared a draft report entitled "Hazardous Waste Characteristics Scoping Study: Environmental Release Descriptions" which was released for public comment on September 25, 1996 (see 61 Federal Register 50295). This chapter summarizes the revised report, incorporating relevant comments on the draft report. This chapter is composed of three sections: ! Section 2.1 discusses the criteria, information sources, and methodology used to select releases to include in the report; ! Section 2.2 summarizes the release descriptions and presents findings of the study; and ! Section 2.3 presents the major limitations of the study. The environmental release descriptions described in this chapter are presented in Appendix A of this Scoping Study. 2.1 Methodology Based on 1985 data, 7.6 billion tons of non-hazardous industrial waste are generated and managed on-site annually by 17 major industries in the United States. Despite this large volume of non-hazardous industrial waste, EPA has few data concerning the releases, human health impacts, or environmental damages caused by such wastes. To identify such releases for purposes of the Scoping Study, the Agency reviewed readily available information from a wide variety of data sources. The purpose of this review was not to estimate risks posed, but rather to characterize releases due to non-hazardous industrial waste management practices. This section discusses the criteria and methodology used to select releases. 2.1.1 Criteria For Selecting Releases The Agency considered three major factors in judging whether a release is an appropriate case study for this report. To be included, a release had to meet all three of the criteria described below: 1. Source of Release. The source of contamination had to be a waste management unit that received only non-hazardous industrial waste. Releases were excluded if: a. Evidence suggested that the management unit also received municipal solid waste, special waste, or RCRA hazardous waste. Many facilities manage municipal, hazardous, and special wastes in the same waste management units as non-hazardous industrial waste. Releases from such units were not included in this report. b. The source of contamination could not be attributable solely to a non-hazardous industrial waste management unit. Releases were excluded where contamination (1) was detected at or near the facility, but the source of contamination was unknown; (2) was not from a waste management unit (e.g., was a product spill); or (3) was from a combination of non-hazardous industrial waste unit(s) and municipal, special, or hazardous waste unit(s). c. The source of contamination was industrial wastewater discharges that are point source discharges regulated under Section 402 of the Clean Water Act, as amended. d. The management method employed would be illegal in most states today. (Facilities were included if management practices would be legal today, even if no longer employed at a particular facility.) 2. Evidence of Damage. For purposes of the study, "damage" is considered to be a release exceeding one of the levels described below. All exceedences were examined for purposes of this scoping study. Exceedences may not actually represent significant risks. To be included in the Study, a release from a waste management unit must have caused contamination at levels of potential concern for that contaminated medium. Levels of potential concern used for this criterion were often based on federal or state drinking water standards for groundwater contamination and exceedences of background concentrations for soil contamination. Federal drinking water standards include maximum contaminant levels (MCLs) and secondary maximum contaminant levels (SMCLs). State drinking water standards, which are often stricter than the federal standards, also were considered. Releases were not included if contaminant concentrations were above background concentrations but below levels of potential concern. If at least one contaminant was detected at concentrations above a federal or state standard, then data were collected and presented for all contaminants detected at that site. 3. Test of Proof. Documented evidence must prove that a damage or danger from a non-hazardous industrial waste management unit has occurred. Evidence was accepted if it met one or more of the following three tests: a. Scientific investigation. Damages were found to exist as part of the findings of a scientific study. Such studies include both extensive formal investigations (e.g., in support of litigation or a state enforcement action) and the results of technical tests (e.g., monitoring of wells); b. Administrative ruling. Damages were found to exist through a formal administrative ruling, such as the conclusions of a site report by a field inspector, or through existence of an enforcement action that cited specific health or environmental dangers; and/or c. Court decision. Damages were found to exist through a ruling of a court of law or through an out-of-court settlement. 2.1.2 Approach For Identifying Releases The Agency investigated eight major data sources to identify potential releases: ! State Industrial D programs; ! State Superfund programs; ! Federal Superfund program; ! Draft EPA report on construction and demolition waste landfills; ! Federal RCRA corrective action program; ! Other federal and state data sources; ! Newspapers; and ! Other literature searches. EPA identified 112 facilities with environmental releases from 4 of the 8 data sources. As a result, this section summarizes the methodologies used to investigate only the four sources that resulted in case studies. Detailed descriptions of the other four methodologies are presented in Appendix B. Draft release descriptions were sent to facility owners/managers for data verification before inclusion in this final report. 2.1.2.1 State Industrial D Programs As specified under RCRA Subtitle D, states are the primary regulators of non-hazardous solid waste, also known as Subtitle D waste. EPA's role is largely limited to establishing guidelines for the development and implementation of state plans, providing technical assistance, and approving plans that comply with these requirements. States are responsible for developing and implementing their own plans. EPA identified states with potential case studies, then reviewed the state files for those potential case studies. The Agency is currently preparing voluntary guidelines on management standards for non-hazardous industrial wastes. As part of this effort, in 1995, the Agency contacted representatives from every state in the continental United States and asked them to identify known or potential environmental damages caused by non-hazardous industrial waste management units. The Agency visited and reviewed state files at four of the five states that reported the largest number of potential case studies, California, Texas, North Carolina, New Mexico, and Wisconsin, and prepared a report summarizing the results of the visits. The Agency did not visit California because, at the time, California was preparing a comprehensive report on its Solid Waste Assessment Test (SWAT) program, which included detailed information on environmental releases at non-hazardous industrial waste disposal sites. For the Scoping Study, the Agency chose to investigate seven additional states based on the reported numbers of potential case studies for these States. Overall, the Agency focused its review of non-hazardous industrial waste data on 12 of the 16 states that indicated having at least 10 potential case studies. The Agency limited its review to these 12 states due to significant time constraints associated with the Scoping Study. As the first step in identifying relevant releases or case studies, the Agency contacted the states by telephone to discuss the requirements and purpose of the release descriptions. For states that housed their files regionally, the Agency contacted each regional office with potential case studies. After scheduling appointments to review the state files, the Agency visited states to review and collect information about the specific releases of non-hazardous industrial wastes into the environment at concentrations of concern. The Agency did not visit California. During these trips, the Agency reviewed readily available documentation on each potential case study and collected documentation for only those releases that appeared to meet all three of the criteria described in Section 3.1.1. Over 80 percent of the facilities identified as potential case studies were excluded from further review, primarily because the facilities co-disposed non-hazardous industrial waste with municipal, hazardous, or special waste, or because the environmental damages discovered at the facility could not be directly linked to a non-hazardous industrial waste management unit. On an as-needed basis, EPA also made follow-up contact with state personnel most knowledgeable about particular sites to obtain additional relevant information. To ensure that facility-specific information was accurately compiled and presented, the Agency contacted the states and facilities by telephone to ask them to review the draft release descriptions prepared for this report. The Agency sent each state and facility their release descriptions, asked for their written comments on the descriptions, and incorporated relevant comments. Review of California's Industrial D Data. In 1984, the California State legislature passed a law that required testing of water and air media at all solid waste disposal sites. The law also required California's State Water Resource Control Board to rank all solid waste disposal sites in groups of 150 each, according to the threat these facilities or sites may pose to water quality. California's legislation requires site operators to submit a water quality "solid waste assessment test" (SWAT) report presenting the following information: ! An analysis of the surface and groundwater on, under, and within one mile of the solid waste disposal site to provide a reliable indication of whether there is any leakage of hazardous waste constituents; and ! A chemical characterization of the soil-pore liquid in those areas that are likely to be affected if the solid waste disposal site is leaking, as compared to geologically similar areas near the solid waste disposal site that are known to not have been affected by leakage or waste discharge. To expedite the review of California's Industrial D data, the Agency obtained a copy of California's Solid Waste Assessment Test database. The Agency reviewed the database to identify those facilities believed to manage only non-hazardous industrial waste and found to have leaked waste constituents outside the limits of the waste management unit at levels above California or federal regulatory standards. California's waste classification system was used to identify facilities believed to manage only non-hazardous industrial waste. The review of Industrial D data from 12 states identified a total of 104 releases that met the Agency's selection criteria. Hundreds of potential cases were reviewed to identify these releases. 2.1.2.2 State Superfund Programs Abandoned or uncontrolled hazardous substance sites not addressed by the federal Comprehensive Environmental Response, Compensation, and Liability Act (CERCLA) program may be subject to remediation under the state Superfund programs. EPA believes that some of these sites may be contaminated with industrial wastes that would not be hazardous under the current RCRA Subtitle C requirements. To expedite the process of identifying relevant sites and to cover the largest possible percentage of state Superfund sites, the Agency focused on the states with the largest programs. These states were identified according to the Environmental Law Institute's 1993 Analysis of State Superfund Programs. In July 1996, the Agency identified and contacted 13 states listed as having at least 1,000 state Superfund sites. Personnel from each of the 13 states were asked whether they produce publicly available summaries of their state Superfund programs. The Agency obtained the most recent annual state Superfund reports for Missouri, New Jersey, New York, and Texas and obtained a printout of California's database for review. Due to the significant time constraints associated with its analysis, the Agency did not pursue information from other states, which lacked detailed, readily available information on their Superfund program. Short published site descriptions for nearly 1,000 state Superfund sites from 5 states, California, Missouri, New Jersey, New York, and Texas, were reviewed to identify potential case studies that meet the Agency's selection criteria. A total of 60 sites were identified as potential case studies. The Agency contacted the five states by telephone to discuss the availability of existing information on those 60 sites. Two states (New York and Texas) indicated that they had additional information readily available for review. The Agency visited these states' Superfund offices to review and the additional information. The Agency identified one case study from New York as meeting all of the selection criteria. 2.1.2.3 Federal Superfund Program The Agency investigated several CERCLA data sources to identify releases relevant to the Scoping Study. The vast majority of the CERCLA sites were not expected to meet the Agency's selection criteria for two reasons. First, the majority of the sites are contaminated with RCRA hazardous wastes or with releases or spills from products. These sites will not meet the Agency's selection criteria for source of release. Second, most of the CERCLA sites contaminated with non-hazardous industrial wastes are also expected to be contaminated with hazardous wastes. Therefore, it is unlikely that a non-hazardous industrial waste management unit will be identified as the source of the release at a CERCLA site. Due in part to the large number (over 1,300) of CERCLA National Priority List (NPL) sites and the relatively small number of sites likely to meet the Agency's three release selection criteria, the Agency attempted to identify potential case study sites through telephone discussions with Regional EPA Superfund personnel and Regional members of the National Association of Remedial Project Managers and the National On-Scene Coordinator Association. Although the Regional Contacts agreed that the Agency should be able to identify at least a few case studies from the CERCLA program, they often were unable to identify specific sites. EPA Superfund staff in Region 4, however, identified two sites apparently meeting the Agency's selection criteria. The Agency visited Region 4's Superfund office and reviewed and copied the relevant files for these two sites. One of the two sites met the Agency's selection criteria. The following federal Superfund data sources were also reviewed; however no releases meeting the Agency's selection criteria were identified: ! Record of Decision (ROD) database; ! CERCLA Natural Resource Damage Claims; ! CERCLA Characterization Database; and ! Exposure assessments performed by the Agency for Toxic Substances and Disease Registry (ATSDR). 2.1.2.4 Construction and Demolition (C&D) Landfill Report On May 18, 1995, EPA's Office of Solid Waste, OSW (renamed Office of Resource Conservation and Recovery, ORCR, on January 18, 2009) published a draft report entitled Damage Cases: Construction and Demolition Waste Landfills. The report, prepared in support of EPA's rulemaking (60 Federal Register 30963, June 12, 1995) on conditionally exempt small quantity generators (CESQG), presents information on environmental releases from construction and demolition (C&D) waste landfills, which receive materials generated from the construction or destruction of structures such as buildings, roads, and bridges. One purpose of the report was to determine whether the disposal of C&D waste in landfills has threatened or damaged human health or the environment. The May 1995 report used three criteria to select potential C&D waste landfill damage cases. ! The landfill received predominantly C&D waste, with or without CESQG waste mixed in. C&D landfills known to have received significant quantities of municipal, industrial, or hazardous wastes were excluded. ! The use of the site as a C&D landfill had to be the only potential source of the observed contamination. Sites located near other potential sources of the contamination such as underground storage tanks were excluded. ! There was documented evidence of groundwater contamination, surface water contamination, or ecological damage at the site. "Contamination" was defined as an increase in chemical constituent concentrations above background or an exceedence of an applicable regulatory standard or criterion attributable to releases from the site. In preparing the May 1995 report, the Agency searched for C&D landfills meeting these criteria using four information sources: existing studies of C&D landfills, materials available through the federal Superfund program, representatives of EPA Regions, and representatives of state and county environmental agencies. The Agency identified 11 environmental releases in the May 1995 report. Although one of the Agency's criteria, as listed above, was to eliminate C&D landfills that received significant quantities of municipal or hazardous wastes, 5 of the 11 landfills received municipal, special, or hazardous wastes. Therefore, for purposes of this report, the Agency eliminated these five C&D landfill cases. Eliminating the landfills that managed even small quantities of municipal, special, or hazardous waste, ensures that the reported damages were caused by the non-hazardous industrial wastes, thereby meeting the Agency's selection criteria for the source of the release. 2.1.3 Release Profile Preparation The release profiles presented in Appendix A to the Scoping Study were prepared using a standard format. This format is discussed below. Because the release profiles were prepared under significant time constraints using readily available data, detailed descriptions of the facility, wastes, and waste management practices could not be developed. The data often provided only a brief description of the facility and focused primarily on the results of the environmental sampling conducted at the facility. "Facility Overview" discusses the facility's operations, how long the facility was or has been in operation, the location of the facility, surrounding land uses, the geologic and hydrogeologic conditions at the facility, and other environmental characteristics, provided this information was available. "Media Affected" identifies whether the damages are associated with groundwater, surface water, soil, and/or ecological receptors. "Wastes and Waste Management Practices" discusses the type(s) of wastes generated at the facility and the practices employed to manage the wastes including descriptions of the individual waste management units and groundwater monitoring practices, provided this information was available. "Extent of Contamination" discusses the groundwater contamination, surface water contamination, and/or soil contamination at the site. Constituents detected in groundwater or surface water above background levels are identified and compared to applicable state and federal standards. The maximum detected concentration for all tested constituents are given. In reporting exceedences of state or federal standards, EPA attempted to exclude constituents whose upgradient or background concentrations were as high as those in downgradient wells. "Corrective Actions/Regulatory Actions" discusses any corrective or regulatory actions that have been recommended, planned, or taken at the site. "Source" simply identifies the information source(s) used to prepare the release profiles. The main source of information was the facility-specific files located in state offices. 2.2 Results This section discusses the findings of the review of release data. It begins by summarizing the 112 documented release descriptions using the following five categories: ! Number of cases by state; ! Number of cases by industry; ! Number of cases by type of waste management method; ! Type of media affected; and ! Type and level of contaminants. Later chapters of this report also present these and additional release description data. 2.2.1 Number of Cases By State The 112 releases described in this chapter were found in 12 states. Because this report is a Scoping Study, these case studies were not intended to be geographically or statistically representative of the number of known or potential releases of non-hazardous industrial wastes identified by the Agency. Although these case studies are not statistically or geographically representative, they do illustrate the type of releases that have occurred from non-hazardous industrial waste management units in various parts of the country, as shown in Exhibit 2-1. The case studies were selected based on the availability of data. Due to the limited time available to collect data, the Agency largely focused its efforts on the states with the most available data on releases from non-hazardous industrial waste management units. This process identified releases in most areas of the nation, except the northwest, northern mountain states, and midwest. The states in these regions either were unable to identify any or identified few potential case studies in the Agency's 1995 efforts to estimate the number of potential releases from non-hazardous industrial waste management units by state. Exhibit 2-1 Number of Release Descriptions By State The available data on facilities that manage non-hazardous industrial waste indicate that the states addressed in this report manage some of the largest volumes of non-hazardous industrial waste. Also, seven of the 12 states represented in this report are among the 10 states with the largest number of on-site non-hazardous industrial waste management units in 1985. Exhibit 2-2 identifies the number of on-site management units and the volume of waste managed on-site in states. (See Chapter 8 for further discussion of waste generation by industry.) 2.2.2 Number of Cases By Industry The releases documented in this report were from facilities in 15 2-digit Standard Industry Classification (SIC) codes. (Industry data are presented at the two-digit level because more specific classification were not readily available for many facilities.) Over 31 percent of the cases involve Electric, Gas, and Sanitary Services facilities (SIC 49). All of these facilities are in the refuse system sector (SIC 4953). The top four SIC codes are SIC 49: Electric, Gas, and Sanitary Services, SIC 26: Paper & Allied Products, SIC 28: Chemical & Allied Products, and SIC 20: Food & Kindred Products. These four industry groups represent nearly 75 percent of the releases studied or evaluated in this report. Exhibit 2-3 identifies the number of cases by industry. These findings are generally consistent with the Agency's previous finding that four industries, Paper and Allied Products (SIC 26), Chemicals and Allied Products (SIC 28), Petroleum Refining & Related Industries (SIC 29), and Primary Metal Industries (SIC 33), generated more than 68 percent of the 7.6 billion tons of Industrial D waste managed on-site in 1985. Although these case studies were identified based on available data and other selection criteria, the number of cases identified per industry and the volume of waste generated per industry appear to be positively correlated. 2.2.3 Number of Cases By Type of Waste Management Unit Four major types of land-based treatment and storage units were identified in the case studies: landfills, surface impoundments, land application units, and waste piles. Exhibit 2-4 presents the number of case studies by waste management unit. Several cases studies discuss more than one unit, therefore, the total number of units is higher than the total number of case studies. Approximately 93 percent of the case studies involved landfills and/or surface impoundments. This finding may partly reflect the greater regulatory attention these units receive from the states, rather than necessarily imply that these units have more frequent releases than other types of waste management units. Over 90 percent of the landfills and 80 percent of the surface impoundments included in the case studies are unlined and over 70 percent of the units are no longer being used to manage non-hazardous industrial wastes. All 50 states have developed regulations for surface impoundments. Approximately 90, 46, and 18 percent of the states have developed regulations specifically for landfills, land application units, and waste piles, respectively. The large number of surface impoundments identified in this report is consistent with a finding of EPA's 1987 Telephone Screening Survey that slightly more than half of the facilities that generate and manage on-site non-hazardous industrial waste managed their wastes in on-site surface impoundments. The 1987 survey also indicated that 35 percent of the facilities managed their wastes on-site in waste piles, 19 percent in landfills, and 18 percent in land application units. Many states apply their non-hazardous industrial waste regulations on a site-by-site basis and, therefore, not all facilities in a state are subject to the same data collection and recordkeeping requirements. One recent report indicates that even states with waste pile regulations do not appear to be actively enforcing control, monitoring, and closure requirements, which may partly explain the small number of release descriptions for waste piles. The large number of landfills and surface impoundments in the release descriptions appears consistent with the type of management units used by the primary industries included in this report. Reportedly, the food processing industry has the largest number of non-hazardous industrial waste surface impoundments and land application units., Other major industries identified in this report with a large number of surface impoundments and landfills include the paper, electric power, chemical, mining, and metal finishing industries. 2.2.4 Type of Media Affected Nearly 98 percent of the case studies involved groundwater contamination. Approximately 31 percent of the case studies involved contamination of surface water or soil. No case studies had documented damages from releases to the air and nearly 30 percent of the case studies affected multiple media. The predominance of groundwater contamination is consistent with the use of groundwater monitoring as the most common method of detecting releases from waste management units. Surface water is not as routinely monitored as groundwater. Surface water sampling is seldom conducted at a facility until a release is identified. Soil sampling is conducted much less frequently than groundwater monitoring, and like surface water sampling, is seldom conducted until a release has been identified. Few states regulate air emissions from non-hazardous industrial waste management units. Thus, it is not surprising that no cases of damage from releases to the air were documented in the case studies collected for this report. 2.2.5 Types of Contaminants Released The number of and types of contaminants routinely analyzed for in groundwater and other types of samples varies among states and facilities. Although most facilities included in the case studies were monitored for a wide range of constituents, the 20 constituents most commonly detected to exceed regulatory levels were inorganics. Approximately 50 constituents were detected three or more times, and 70 constituents were detected fewer than three times. Exhibit 2-5 identifies all of the TC constituents that were detected in the case studies, Exhibit 2-6 presents all of the constituents with SMCLs that were identified in the case studies, and Exhibit 2-7 identifies the other constituents that were detected in at least three case studies. The exhibits also identify the number of cases where each constituent was detected, the number of times the constituent was detected above at least one regulatory level, the regulatory levels, the average maximum and the highest maximum detected concentration identified in the case studies, and the range of the ratio of the highest detected constituent concentrations to regulatory standards. Note, only constituents with regulatory standards are included in Exhibits 2-5, 2-6, and 2-7. Many inorganic constituents were elevated in groundwater monitoring wells. Constituents that exceeded state groundwater protection standards or federal drinking water standards most frequently were: ! Iron (49 detections) ! Cadmium (17 detections) ! Chloride (32 detections) ! Benzene (16 detections) ! Manganese (34 detections) ! Arsenic (15 detections) ! Sulfate (29 detections) ! Zinc (13 detections) ! Lead (22 detections) ! Aluminum (12 detections) ! Chromium (21 detections) ! Nitrate (12 detections) Six of the constituents identified above (iron, chloride, manganese, sulfate, zinc, and aluminum) have drinking water standards that are based only on SMCLs. A total of 25 TC constituents have been detected in the release descriptions. Exhibit 2-5 identifies 20 of the 25 TC constituents detected. Five TC constituents (2,4,6-trichlorophenol, 2,4-dinitrotoluene, o-cresol, p-cresol, and methyl ethyl ketone) were not included in Exhibit 2-5 because there were no federal or state standards established for them. All but 2 of the 20 TC constituents identified in Exhibit 2-5 (carbon tetrachloride, 1,4-dichlorobenzene) were detected above a federal or state standard. The majority (85 percent) of the TC constituents detected above a federal or state standard exceeded the standards by at least 1 time, 60 percent exceeded by 10 times, 50 percent exceeded by 100 times, 20 percent exceeded by 1,000 times, 10 percent exceeded by 10,000 times, and none exceeded by at least 100,000 times. The average maximum detected concentrations for five of the TC constituents (arsenic, benzene, selenium, vinyl chloride, and lindane) exceeded the TC regulatory levels established for these constituents and the highest maximum detected concentrations for over half of the identified TC constituents exceed TC regulatory levels. All SMCLs or similar state standards, except those for foaming agents, color, odor, and corrosivity, were violated by one or more release descriptions. As shown in Exhibit 2-6, the majority (90 percent) of the SMCL constituents exceeded the standards by at least 1 time, 80 percent exceeded by 10 times, 40 percent exceeded by 100 times, 20 percent exceeded by 1,000 times, 10 percent exceeded by 10,000 times, and none exceeded by at least 100,000 times. (Because silver has both a TC level and an SMCL, it is included in Exhibit 2-5 with the other TC constituents.) SMCLs are based on aesthetic considerations (e.g., taste and odor) and are not federally enforceable. Therefore, exceedences of the SMCLs do not necessarily indicate a potential danger to human health or the environment. Sixteen of the case studies (14 percent) were identified based only on an exceedence of an SMCL. This type of contamination is discussed further in Chapter 5. Exhibit 2-7 identifies 24 other constituents that were detected in the release descriptions. All but four of the constituents in Exhibit 2-7 (1,1-dichloroethane, nitrogen, vanadium, and cobalt) were detected above a federal or state regulatory level. Half (50 percent) of these other constituents exceeded one of the standards by at least 10 times, 13 percent exceeded by 100 times, 4 percent exceeded by 1,000 times, and none exceeded by at least 10,000 times. Constituents managed in landfills were detected in samples nearly three times more frequently than constituents managed in surface impoundments. All of the constituents presented in Exhibits 2-5, 2-6, and 2-7 are associated with wastes managed in landfills. Approximately 81 percent of the constituents are associated with both landfills and surface impoundments, 33 percent are associated with landfills, surface impoundments, and land application units, 33 percent are associated with landfills, surface impoundments, and waste piles, and 12 percent are associated with all 4 waste management units. The constituents that are associated only with landfills are antimony, beryllium, boron, cobalt, cyanides, silver, and thallium. Exhibit 2-8 identifies the 10 constituents for each of the 6 industries that were identified most frequently in the case studies. As the exhibit illustrates, inorganics are the most commonly detected chemicals. The commonly detected constituents are chloride, pH, iron, lead, total dissolved solids, manganese, sulfate, magnesium, zinc, and arsenic. 2.3 Major Limitations The findings presented in this chapter must be interpreted with care for several reasons, including the limited time available to collect data, potentially unrepresentative data, and the Agency's stringent release selection criteria. Each of these major limitations is discussed in detail below. Data were collected under significant time constraints. The significant amount of data included in this chapter were collected and analyzed over a four-month period. During this time the Agency reviewed previously collected data, readily available databases, and reports; identified and contacted appropriate state and federal personnel; visited state and EPA Regional offices; reviewed facility files; prepared case study summaries; developed a database to analyze the data; performed QA/QC on the data; sent draft case studies to states and facilities for review; prepared a draft report for public review; and incorporated comments into the report, as appropriate. Due to the time constraints of the consent decree, the Agency had to carefully prioritize its efforts and, in doing so, may have eliminated or missed a number of potential case studies that could have been included in the report if additional data were available and/or additional time was spent collecting and reviewing data. Data may be unrepresentative and/or out-of-date. In this report, the Agency did not attempt to estimate the proportion of non-hazardous management facilities currently experiencing constituent releases. Due primarily to the limited time available for data collection and analysis, the Agency relied upon readily available data. The Agency did not perform any new sampling or collect new data from facilities managing non-hazardous industrial wastes. Nor did the Agency perform a comprehensive review of previously collected state and federal data for all non-hazardous industrial waste management facilities. State file reviews were conducted in one to three days per state and were limited to those states that indicated having files of release incidents that met the Agency's selection criteria. Although the collection of release descriptions is not statistically representative in any way, these cases are indicative of the type of releases associated with the management of non-hazardous industrial waste. Because only readily available data were analyzed, the data may not reflect current waste generation and management practices at the particular facility. Environmental contamination resulting from waste disposal practices may take many years to become evident; some releases described in this report occurred over 20 years ago. The documented releases may have resulted from particular waste generation and disposal practices or other conditions that no longer exist. Specifically, process feedstocks, processing operations, waste characteristics, and/or waste management practices may have changed. Facilities may no longer manage their wastes in unlined units or in environmentally sensitive areas. Therefore, releases associated with a waste do not necessarily demonstrate that current waste management practices or regulations need to change. Conversely, the failure of a site to exhibit documented damages at present does not necessarily suggest that waste management has not or will not cause damage. The Agency, however, believes that information on dangers posed by past waste management practices is useful in demonstrating the potential for human health or environmental damages. The extent to which the findings can be used to draw conclusions concerning the relative performance of waste management practices among states or across industry sectors is also severely limited by variations in recordkeeping, monitoring, and other state requirements. Recordkeeping and monitoring procedures vary significantly among the states. Several states have complete and up-to-date central enforcement or monitoring records on facilities that generate and manage non-hazardous industrial wastes. Where states have such records, information on releases may be readily available. Thus, states with the most complete and accessible monitoring information on non-hazardous industrial wastes may appear to have more releases than states with less centralized information management programs. Stringent selection criteria. The Agency developed stringent selection criteria to help focus its data collection efforts and to avoid any misrepresentation of release incidents. By focusing solely on releases clearly associated with non-hazardous industrial waste management units, the Agency excluded numerous release incidents caused by accidental releases and spills of products. Although these incidents may have been caused by hazardous constituents similar to those managed in non-hazardous industrial waste management units, and may pose similar hazards, the Agency did not analyze these cases, largely because of the inability of RCRA to prevent product releases. The Agency also excluded release incidents that could not be linked to specific facilities. Thus, cases of groundwater and surface water contamination caused by multiple facilities were excluded because the source of the releases could not be associated with specific facilities or waste management units. The Agency also excluded numerous release incidents associated with facilities that manage hazardous, municipal, or special wastes in addition to non-hazardous industrial waste. Facilities that manage hazardous, municipal, or special wastes frequently co-dispose of their non-hazardous industrial wastes in the same or adjacent waste management units. Due to the close proximity of these different units, sampling results generally cannot identify the specific unit associated with the release. Thus, the Agency excluded cases where non-hazardous industrial waste was managed in the same or adjacent waste management units as hazardous, municipal, or special wastes, because the source of the release could not clearly be associated with the non-hazardous industrial waste. CHAPTER 2. RELEASES FROM NON-HAZARDOUS INDUSTRIAL WASTE MANAGEMENT UNITS2-1 2.1 Methodology . . . . . . . . . . . . . . . . . .2-1 2.1.1 Criteria For Selecting Releases . . .2-1 2.1.2 Approach For Identifying Releases . .2-3 2.1.2.1 State Industrial D Programs . . .2-3 2.1.2.2 State Superfund Programs. . . . .2-5 2.1.2.3 Federal Superfund Program . . . .2-6 2.1.2.4 Construction and Demolition (C&D) Landfill Report 2-6 2.1.3 Release Profile Preparation . . . . .2-7 2.2 Results . . . . . . . . . . . . . . . . . . . .2-8 2.2.1 Number of Cases By State. . . . . . .2-8 2.2.2 Number of Cases By Industry . . . . 2-10 2.2.3 Number of Cases By Type of Waste Management Unit 2-11 2.2.4 Type of Media Affected. . . . . . . 2-12 2.2.5 Types of Contaminants Released. . . 2-13 2.3 Major Limitations . . . . . . . . . . . . . . 2-17 Exhibit 2-1 Number of Release Descriptions By State .2-9 Exhibit 2-2 Number of Management Units & Volume of Waste Managed On-Site, by State (1985) 2-9 Exhibit 2-3 Number of Case Studies by Industry (SIC). . . 2-10 Exhibit 2-4 Number of Case Studies By Waste Management Unit 2-12 Exhibit 2-5 TC Contaminants Detected in Case Studies.2-14 Exhibit 2-6 Contaminants with SMCLs Detected in Case Studies 2-15 Exhibit 2-7 Other Contaminants Detected in At Least Three Case Studies 2-16 Exhibit 2-8 Most Common Constituents By Industry. . 2-18 CHAPTER 3. POTENTIAL GAPS ASSOCIATED WITH HAZARDOUS WASTE CHARACTERISTICS DEFINITIONS This chapter examines how well the existing hazardous waste characteristics address the types of risk they were intended to address, that is, their target risks. It also addresses certain other or non-target risks that are closely associated with the definitions of the hazardous characteristics. This evaluation of potential gaps begins by examining the characteristics' definitions and test methods. This approach is used for two reasons. First, limitations in the characteristics' effectiveness in reducing their target risks may themselves constitute important potential gaps. When the characteristics were promulgated, the Agency identified physical hazards and acute toxic hazards during transport and disposal activities, along with chronic exposure to groundwater contaminated with specific waste constituents, as being among the most important waste management risks. Reducing these risks remains an important goal of the characteristics. Second, this analysis lays the groundwork for evaluating other potential gaps. Specifically, risk-based screening methods are used to evaluate non-target risks from non-ground-water pathways associated with the toxicity characteristic (TC) analytes. The findings of that analyses are used to identify potential gaps associated with a wider universe of known and possible non-hazardous industrial waste constituents, as discussed in Chapter 4. This chapter revisits many of the assumptions and approaches used to develop the existing hazardous waste characteristics. The ignitability, corrosivity, and reactivity (ICR) characteristics are essentially unchanged since their initial promulgation in 1980. The TC characteristic was revised in 1990, but has not changed materially since then. Potential gaps in these characteristics may be identified if the state of knowledge about risks addressed by the characteristics has improved since the characteristics were promulgated; risks that were not specifically addressed may now be identified as more important, such as risks from releases to surface water, inhalation, and indirect pathways and ecological risks. In addition, the tests used to identify wastes with hazardous characteristics do not reliably identify all of the risks the characteristics were intended to address. The following sections address these issues. Section 3.1 reviews the statutory and regulatory language related to the types of risks the hazardous waste characteristics were intended to address and discusses the major categories of waste management risks addressed and not addressed by the current characteristics. Sections 3.2 through 3.4 discuss potential gaps associated with the ignitability, corrosivity, and reactivity characteristics, respectively. In addition, a detailed comparison of the ICR characteristics can be found in Appendix C. Section 3.5 discusses the potential gaps associated with the toxicity characteristic, including updated risk information on the TC analytes. Section 3.6 evaluates the toxicity characteristic leaching procedure (TCLP) as a predictor of constituent releases and potential risk. 3.1 Types of Risks Addressed by RCRA Hazardous Waste Characteristics 3.1.1 Statutory and Regulatory Framework The RCRA hazardous waste characteristics are a vital part of the comprehensive program of hazardous waste management established by Subtitle C of RCRA, as amended. Three provisions of the RCRA statute are particularly relevant to identifying and expanding the hazardous waste characteristics (and listing hazardous wastes). ! First, Section 1004(7) defines hazardous waste as "a solid waste, or combination of solid wastes, which because of its quantity, concentration, or physical, chemical, or infectious characteristics may (A) cause, or significantly contribute to an increase in mortality or an increase in serious irreversible, or incapacitating reversible, illness; or (B) pose a substantial present or potential hazard to human health or the environment when improperly treated, stored, transported, or disposed of, or otherwise managed." This definition indicates the general types of risks that the hazardous waste management regulations are meant to address. ! Second, Section 3001(a) requires EPA to "develop and promulgate criteria for identifying the characteristics of hazardous waste, and for listing hazardous wastes, . . . taking into account toxicity, persistence, and degradability in nature, potential for accumulation in tissue, and other related factors such as flammability, corrosiveness, and other hazardous characteristics. Such criteria shall be revised from time to time as may be appropriate." ! Third, Section 3001(b) requires EPA to "promulgate regulations identifying the characteristics of hazardous waste, and listing particular hazardous wastes, . . . which shall be based on the criteria promulgated under [Section 3001(a)] and shall be revised from time to time thereafter as may be appropriate." The Section also requires EPA to "identify or list those hazardous wastes which shall be subject to the [hazardous waste regulations] solely because of the presence in such wastes of certain constituents (such as identified carcinogens, mutagens, or teratogens) at levels in excess of levels which endanger human health." In response to the mandate of Section 3001(a), EPA promulgated two sets of criteria for identifying the characteristics of hazardous waste in 40 CFR 261.10(a). The first set of criteria reflects the statutory definition of hazardous waste and the types of risks that the characteristics are intended to address: "(1) The solid waste may (i) cause, or significantly contribute to, an increase in mortality or an increase in serious irreversible, or incapacitating reversible, illness; or (ii) pose a substantial present or potential hazard to human health or the environment when it is improperly treated, stored, transported, disposed, or otherwise managed." The second set of criteria considers implementation factors: "(2) The characteristic can be (i) measured by an available standardized test method which is reasonably within the capability of generators of solid waste or private sector laboratories that are available to serve generators of solid waste; or (ii) reasonably detected by generators of solid waste through their knowledge of their waste." As stated in the May 19, 1980, final rule, EPA adopted the second set of criteria because the primary responsibility for determining whether wastes exhibit a characteristic rests with generators, for whom standard and available testing protocols are essential. This Scoping Study addresses these criteria for the current characteristics in only a general way. The Agency, however, will carefully consider these factors when deciding the appropriate course of action for addressing any potential gaps in coverage that are identified in this Study. The following sections review the nature of the risks to human health and environment potentially posed by non-hazardous industrial waste management. These risks are associated with physical hazards, acute toxic hazards to humans, chronic toxic hazards to humans, risk to non-human receptors, and other hazards. In the discussion below, risks addressed by the hazardous waste characteristics are distinguished from those risks not directly or adequately addressed. The purpose of this section is to develop a preliminary list of possible gaps in the characteristics. At this stage, few judgments are made as to the nature and severity of any potential gaps. Instead, the remainder of this Report investigates these potential gaps. 3.1.2 Risks Associated with Physical Hazards Physical hazards include agents that cause direct physical harm such as thermal burns, wounds, contusions, or eye injuries, in contrast to agents causing harm through chemical burns or toxic effects. These hazards are controlled primarily through the ignitability, corrosivity, and reactivity (ICR) characteristics. EPA patterned these characteristics after similar regulations promulgated by the U.S. Department of Transportation, the National Fire Protection Association, and other organizations. The ICR characteristics are intended primarily to protect waste management and transportation workers against hazards often associated with hazardous materials. These hazards include flammability, explosivity, and the propensity to react violently with other wastes, corrode containers, and directly injure skin and eyes during transport or management activities. In addition, these characteristics are intended to prevent the facilitated release and transport of hazardous waste constituents. For example, the corrosivity test is designed, in part, to identify wastes that, because of their acidity or basicity, may facilitate the solubilization of metals from wastes. This solubilization increases the potential impact of metals in groundwater, thereby increasing the likelihood of risks to human health via contaminated groundwater. For the purposes of this Scoping Study, the question is: What physical risks may arise from the management of non-hazardous industrial wastes that are currently not covered by the characteristics? Several potentially significant physical risks are not effectively addressed by the hazardous characteristics. Some of the potential gaps arise from specific definitions of the ICR characteristics. These potential gaps, which are discussed in more detail in Sections 3.2 through 3.4, include: ! The lack of coverage of corrosive solids; ! The decision not to address liquids with moderate flash points; ! Limitations in the test procedures prescribed for reactivity; and ! Potential limitations of pH as an adequate indicator of corrosivity. These issues relate to protecting waste management and transportation workers from physical injuries, except where explosions or fire might release toxic particulates that could harm nearby residents. Physical hazards to residents near management facilities are not considered, based on the assumption that the general public has limited access to non-hazardous industrial waste management facilities. Other physical concerns relate to facilitated pollutant transport. For example, the corrosivity characteristic was not intended to address corrosion to liners or any materials other than steel or to prevent facilitated transport of organic chemicals through solubilization in discarded solvents. EPA considered, and decided to omit, a "solvent override" provision in the 1990 TC rule that would have classified as hazardous wastes with more than a specified concentration of hazardous organic solvents. The Agency, however, left open the possibility that such a provision could be reconsidered if additional data warrant it. A related issue is the potential formation of dense and light non-aqueous phase liquids (DNAPLs and LNAPLs). They are a potential concern both because they may facilitate pollutant transport and they have the potential for damaging groundwater resources and generating high remediation costs. Section 4.4 discusses the issue of DNAPL and LNAPL formation. 3.1.3 Acute Toxic Hazards to Humans The hazardous waste characteristics address some potential health risks from acute exposures to toxic chemicals. They limit the potential for release of toxic chemicals during transportation, storage, treatment, and disposal and resulting from fires, explosions, or violent reactions. There are no specific quantitative benchmarks that define acceptable acute exposure limits, however. The main focus of the ICR characteristics is on protecting workers, although the general public is implicitly protected under the assumption that protecting on-site workers would protect more distant resident populations as well. Sections 3.2 through 3.4 discuss potential gaps in the ICR characteristics. The characteristics were not intended to protect against other acute systemic toxicity hazards. Direct contact with a waste, in theory, could result in the absorption of an acutely toxic dose of a waste constituent from a non-corrosive waste. The Agency, however, considered this scenario to be highly improbable for non-hazardous industrial waste mismanagement. Similarly, acute exposures via contaminated surface or groundwater are possible, but were considered much less likely to be important than chronic toxicity under most circumstances. Because the TC focuses on the groundwater pathway, with the attendant long-term transport and dilution of pollutants, EPA assumed that chronic exposures would be dominant in determining the potential for adverse health effects. Section 3.5.6 discusses the potential for acute adverse effects of exposure to the TC analytes and Section 4.6 addresses acute risks from non-TC constituents. 3.1.4 Chronic Toxicity Risks to Humans As noted above, EPA intended the TC to be the major vehicle for controlling chronic health risks, although the reactivity and corrosivity characteristics also were intended to prevent releases that facilitate exposure to waste constituents. Although RCRA Section 3001 identifies a range of types of toxic effects of concern (toxicity, carcinogenicity, mutagenicity, and teratogenicity), the implementation of the TC is limited to 40 chemicals for which toxicity and groundwater fate and transport data were available when the Agency revised the characteristic in 1990. In addition, the levels of protectiveness achieved by the TC leachate concentration standards were determined using fate and transport models and assumptions that were current at the time. To the extent that the toxicity data and groundwater fate and transport models have changed or improved in the six years since the TC was promulgated, its expected level of protectiveness may also have changed. Section 3.5 discusses in detail potential gaps associated with the level of protectiveness of the TC in light of recent advances in toxicology and groundwater modeling. The TC was not intended to address several potentially important risks. These risks have been identified as significant contributors to risks from some hazardous waste constituents and management technologies, and might apply to non-hazardous industrial waste management as well. Probably the most important risks potentially not directly addressed by the TC are associated with exposure pathways other than groundwater. The TC did not attempt to address these risks because groundwater was thought to be the dominant risk pathway for waste management. Upon re-examining potential non-hazardous industrial waste management and mismanagement scenarios, however, EPA recognizes that other pathways also may be important. One pathway not directly addressed by the TC is the direct inhalation of volatile or particulate-bound waste constituents to air from waste management units during normal operation or after closure. Such exposures to on-site workers and off-site receptors through direct inhalation may be significant for some constituents. Other potentially important pathways include the surface water pathway and "indirect" pathways arising from air releases (e.g., air deposition to crops), runoff, and the discharge of contaminated groundwater to surface water. Also, bioaccumulation of certain contaminants in aquatic and/or terrestrial food chains could result in human exposures through the consumption of contaminated fish, shellfish, livestock, and game animals. In Section 3.5, a screening-level risk assessment and other information clarify the significance of these pathways for the TC analytes. Chapter 4 extends the screening-level analysis to non-TC constituents. 3.1.5 Risks to Non-Human Receptors Neither the TC nor the ICR characteristics were established specifically to reduce risks to non-human receptors. Such risk reduction, to the extent that it occurs, is a byproduct of the control of human health risks. For example, by preventing pollutant releases from fires and explosions or reducing pollutant transport, the characteristics protect the environment as well as human health. The quantitatively-defined levels of protection incorporated into the TC leachate concentration limits were based on human toxicity considerations; they do not consider toxicity to non-human receptors. While the exposure levels accepted as protective of human health may be generally protective of wildlife populations, notable exceptions arise both from the ecotoxicological properties of some chemicals and from differences between human and non-human receptor exposure patterns. The question therefore can be asked: To what extent is the TC protective of ecological receptors? As in the case of human health risks, the TC does not directly protect against risks from chemicals not on the TC list. Similarly, it is not clear how protective the existing TC levels are for the various exposure pathways that are most important for aquatic and terrestrial receptors. In the case of ecological receptors, as is the case for human health, both direct and indirect exposure pathways may be significant. These issues are addressed in more detail in Section 3.5 and Chapter 4 of this report. 3.1.6 Other Risks Associated with Non-Hazardous Industrial Waste Management In establishing the existing hazardous waste characteristics, the Agency focused exclusively on human health risks directly associated with local effects of accidents and on chemical contamination of the environment in the near vicinity of the management units. In Chapter 5 of this study, EPA has taken a broader view, and has expanded the scope of the risk identification to include risks other than those originally considered, even indirectly, in establishing the hazardous waste characteristics. These additional categories of risks include damages to natural resources and contributions to large-scale environmental problems. Non-hazardous industrial waste management has the potential to adversely affect the value or utility of natural resources, such as wetlands, groundwater, and air, without posing human health risks. For example, releases from non-hazardous industrial waste management units have polluted previously usable groundwater with constituents generally not considered toxic, such as iron, manganese, chloride, and total dissolved solids. The regulatory criteria violated by these releases, such as Secondary Maximum Concentration Levels (SMCLs) developed under the Safe Drinking Water Act, are not directly health-related, but relate instead to the aesthetic properties or usability of the water. Therefore, even though no health risk is predicted, the water is rendered unusable and the environment is thereby damaged. Similarly, odor from non-hazardous industrial waste management may be seen as an air resource damage, reducing the quality of life for affected individuals, even in the absence of direct health effects. The last category of risks are associated with the possible contribution of non-hazardous industrial waste management to large-scale environmental problems, including: ! Air deposition to the Great Waters; ! Damages from airborne particulates; ! Global climate change; ! Potential damages from endocrine disruptors; ! Red tides; ! Stratospheric ozone depletion; Tropospheric ozone and photochemical air pollution; and ! Water pollution. The possible relationship between non-hazardous industrial waste management and these risks is less clear than for the previously identified risks. As summarized in Exhibit 3-1, Section 3.1 has presented an intentionally broad inventory of potential risks to human health and the environment associated with the management of non-hazardous industrial wastes not currently identified as hazardous. This list provides a catalogue of risks for evaluation against the existing characteristics in the rest of this chapter and the following chapters. 3.2 Ignitability Characteristic This section describes potential gaps related to the definition of the RCRA ignitability characteristic and its test methods. The basic approach taken in identifying potential gaps for ignitability as well as for corrosivity and reactivity was to review the original 1980 rulemaking record and to compare the characteristic to approaches taken to controlling similar hazards under other regulatory schemes, including the U.S. Department of Transportation's (DOT's) hazardous materials regulations, the U.S. Occupational Safety and Health Administration's (OSHA's) worker health hazards standards, and state hazardous waste management standards. 3.2.1 Definition of Ignitability The ignitability characteristic is intended to "identify wastes capable of causing fires during routine transportation, storage and disposal, and wastes capable of exacerbating a fire once started." These risks include generally recognized fire hazards to waste management and transportation workers, such as burns and inhalation smoke or fumes, and the potential generation and facilitated transport in air of toxic particulates and fumes that could harm the general public. According to 40 CFR 261.21, a solid waste exhibits the characteristic of ignitability if a representative sample of the waste has any of the following properties: ! Is a liquid, other than an aqueous solution containing less than 24 percent alcohol by volume and has flash point less than 60oC (140oF), as determined by: -- A Pensky-Martens Closed Cup Tester, using the test method specified in ASTM Standard D-93-79 or D-93-80 (incorporated by reference, see ' 260.11), -- A Setaflash Closed Cup Tester, using the test method specified in ASTM standard D-3278-78 (incorporated by reference, see ' 260.11), or -- An equivalent test method approved by the Administrator under procedures set forth in '' 260.20 and 260.21; ! Is not a liquid and is capable, under standard temperature and pressure, of causing fire through friction, absorption of moisture or spontaneous chemical changes and, when ignited, burns so vigorously and persistently that it creates a hazard; ! Is an ignitable compressed gas as defined in 49 CFR 173.300 and as determined by the test methods described in that regulation or equivalent test methods approved by the Administrator under '' 260.20 and 260.21; or ! Is an oxidizer as defined in 49 CFR 173.151. 3.2.2 Potential Gaps Related to Definition of Ignitability Liquids with flash point at or above 140EF not covered. The RCRA ignitability characteristic includes liquid wastes with flash point less than 60oC (140oF). When promulgating the original characteristic, EPA acknowledged choosing a definition for ignitable liquid wastes that excluded some potential wastes that would meet the definition of hazardous materials under DOT regulations. The DOT definition of flammable liquid includes liquids with flash point not more than 60.5EC (141EF), or any material in liquid phase with a flash point at or above 37.8EC (100EF) that is intentionally heated and offered for transportation or transported at or above its flash point in a bulk packaging. The DOT definition of combustible liquid includes liquids with flash point above 60.5EC (141EF) and below 93EC (200EF). Thus, the RCRA ignitability characteristic covers wastes that would be classified as DOT flammable liquids, but not DOT combustible liquids. Consistent with DOT regulations, OSHA includes such "combustible" liquids in its definition of health hazard, and Rhode Island regulates them as hazardous wastes. In a background document supporting the promulgation of the original characteristics, EPA stated that the RCRA ignitability flash point limit of 140oF reflects conditions likely to be encountered during routine waste management. In support of this conclusion, the Agency referenced seven studies documenting temperatures and conditions at waste management units. The information available to the Agency at the time was limited, however. Furthermore, two of these studies reported temperatures of greater than 140oF. One study reported temperatures of approximately 160oF near the surface of a landfill, noting that aerobic conditions near the surface of landfills often result in relatively high temperatures. Data are still limited regarding whether temperatures greater than 140oF are encountered during non-hazardous industrial waste management, in what situations and how frequently this occurs, and what maximum temperatures are reached (particularly in hotter regions of the nation). One relevant issue is whether temperatures exceeding 140oF may be encountered during mismanagement (as opposed to routine waste management). Examples of possible mismanagement scenarios for ignitable wastes include: ! Wastes stored in closed, heat-containing facilities (e.g., metal sheds, upper floor warehouse spaces, or metal trucks) in hot climates and/or sunlight; and ! Wastes mixed in waste management units in a manner that might generate heat through chemical reactions, especially in the presence of hot climate or sunlight. No information is readily available regarding the universe of "combustible" industrial wastes currently being managed as non-hazardous. Nevertheless, some liquid materials with flash points generally in this range can be identified, as shown in Exhibit 3-2. Examples include certain alcohols, low molecular weight esters, ethylene glycol ethers, kerosene, jet fuels, certain petroleum byproducts, many "tints and paints," and individual chemicals including benzaldehyde, benzonitrile, and bromobenzene. If these materials are disposed of or are present in wastes, the wastes may be combustible, in spite of not being hazardous by the ignitability characteristic. In addition, mixtures of materials of differing flash points may fall into this category. Exclusion for aqueous liquids containing less than 24 percent alcohol may warrant reexamination. At the time of the original rulemaking, some commenters argued that liquid wastes such as wine and some latex paints that exhibit low flash points because of their alcohol content do not sustain combustion because of the high percentage of water and therefore should not be designated as characteristically hazardous waste. EPA agreed and excluded from the ignitability characteristic aqueous solutions containing less than 24 percent of alcohol by volume. A similar exclusion is found in DOT regulations. EPA stated that it hoped "to undertake further study to determine whether another exclusion limit is more appropriate and to evaluate tests which might be capable of identifying wastes which exhibit this phenomenon." EPA also intended to evaluate possible supplemental test methods to evaluate flammability hazards for these types of wastes. The exclusion for aqueous liquids containing alcohol has caused confusion during implementation and enforcement concerning whether it applies only to ethanol or more broadly to all alcohols. The exclusion also focuses on aqueous alcohol solutions, rather than on the underlying target of liquids with low flash points that do not sustain combustion. (Tests for sustained combustion are now available: ASTM has methods D-4206 and D-4207.) In addition, the rationale that certain liquids should not be considered ignitable if they do not sustain combustion may not be valid where an excluded aqueous solution could flash and ignite a co-managed non-hazardous waste that would sustain combustion. References to DOT regulations are outdated. The ignitability characteristic refers to a DOT definition of ignitable compressed gas (49 CFR 173.300) that has been withdrawn. Current DOT regulations at 49 CFR 173.115 define flammable gas, which is any material that is a gas at 20EC (68EF) or less and 101.3 kPa (kilopascals equal to 14.7 pounds per square inch) of pressure. The complete definition includes any material that has a boiling point of 20EC (68EF) or less at 101.3 kPa (14.7 psi)) that (1) is ignitable at 101.3 kPa (14.7 psi) when in a mixture of 13 percent or less by volume with air; or (2) has a flammable range at 101.3 kPa (14.7 psi) with air of at least 12 percent regardless of the lower limit. Likewise, the term oxidizer is no longer defined at 49 CFR 173.151. It is now found at 49 CFR 173.127. These out-of-date citations constitute a potential gap because they may cause regulatory confusion and misinterpretation and thereby may impede efficient and effective compliance and enforcement. 3.2.3 Potential Gaps Related to Ignitability Test Methods No test method is specified for non-liquids. The ignitability characteristic does not specify a test method for non-liquid wastes. In a background document supporting the original rulemaking, EPA stated that non-liquid wastes may present a hazard by virtue of their capacity to ignite and burn as a result of friction, moisture absorption, or spontaneous reaction under the normal temperatures and pressures encountered in waste management. The Agency noted that such wastes are akin to reactive wastes and can directly injure workers or others as a result of fire, induced explosions, or induced generation of toxic gases at almost any point in the waste management process. Examples of potential ignitable non-liquid wastes include soils highly contaminated with gasoline or other ignitable substances and sorbents used to cleanup spills of ignitable substances. In explaining the final rulemaking, the Agency stated that, although "EPA would have preferred providing a test method for identifying ignitable solids, it has determined . . . that there are no test methods capable of accurately identifying the small class of ignitable solids to which its regulation is directed. EPA is presently working with the Department of Transportation and other organizations to correct this deficiency." Since then, EPA has identified a test method that may be suitable for identifying ignitable solids. Method 1030 ("Ignitability of Solids") has been proposed for inclusion in the Third Edition of the EPA test methods manual "Test Methods for Evaluating Solid Waste, Physical/Chemical Methods," EPA Publication SW-846. The method is appropriate for pastes, granular materials, solids that can be cut into strips, and powdery substances. 3.3 Corrosivity 3.3.1 Definition of Corrosivity According to 40 CFR 261.22, a solid waste exhibits the characteristic of corrosivity if a representative sample of the waste has either of the following properties: ! Is aqueous and has a pH less than or equal to 2 or greater than or equal to 12.5, as determined by a pH meter using Method 9040 in "Test Methods for the Evaluation of Solid Waste, Physical/Chemical Methods," incorporated by reference in ' 260.11; or ! Is a liquid and corrodes steel (SAE 1020) at a rate greater than 6.35 mm (0.250 inch) per year at a test temperature of 55oC (130oF) as determined by the test method specified in NACE (National Association of Corrosion Engineers) Standard TM-01-69 as standardized in "Test Methods for the Evaluation of Solid Waste, Physical/Chemical Methods," EPA Publication SW-846, as incorporated by reference in ' 260.11. The first part of this definition encompasses wastes exhibiting low or high pH, which "can cause harm to human tissue, promote the migration of toxic contaminants from other wastes, react dangerously with other wastes, and harm aquatic life." Specifically, the Agency identified skin and eye damage to transporters who are directly exposed to the waste as a primary focus of this characteristic. The pH limits also were intended to address the potential solubilization of heavy metals allowing migration to groundwater, reactions with incompatible wastes resulting in fires, explosions, generation of flammable or toxic gases, generation of pressure inside vessels, and the dispersal of toxic vapors, mists, and particulates. The other part of the corrosivity characteristic relates to the corrosivity of waste to steel containers. The Agency identified this aspect of corrosivity as a hazard because "wastes capable of corroding metal can escape from the containers in which they are segregated and liberate other wastes." The consequences of liberating wastes from containers during transportation or storage include harm from direct contact, violent reactions, and the release of waste components to the environment. 3.3.2 Potential Gaps Related to Definition of Corrosivity Non-liquids are not covered. The current RCRA corrosivity characteristic is limited to liquids. Other regulatory programs, however, also cover corrosive non-liquids. For example: ! DOT regulates corrosive liquids and solids as hazardous materials; ! The OSHA definition of health hazard includes all corrosives regardless of physical form; ! The Basel Convention definitions of hazardous materials are not limited to liquids; and ! At least four states (California, New Hampshire, Rhode Island, and Washington) include non-aqueous wastes in their definitions of corrosivity. New Hampshire and Rhode Island specifically include corrosive gases as well as corrosive solids. The states that include non-liquids in their corrosivity characteristics specify mixing the non-aqueous waste with water and then testing for pH. The rationale for this approach is that the waste is likely to come into contact with water during land-based management. In addition, EPA has developed Method 9045 (Soil and Waste pH), which can be used to test some corrosive solid wastes. Finally, Method 1120 (Dermal Corrosion) may be applied to solids, liquids, and emulsions (see additional discussion below under "potential gaps related to corrosivity test methods"). pH limits may not cover some hazards. EPA originally proposed pH limits of 12.0 or greater and 3.0 or less, and a majority of commenters argued that these limits were too stringent. The commenters argued that the limit of 12.0 or greater would regulate as hazardous many lime-stabilized wastes and sludges, thereby discouraging use of a valuable treatment technique, and that the pH limit of 3.0 or less would regulate a number of substances generally thought to be innocuous (e.g., cola drinks) and many industrial wastewaters prior to neutralization. EPA agreed with these commenters and promulgated pH limits of 12.5 or greater and 2.0 or less in the 1980 final rule. The more stringent proposed pH limits were based on studies of eye tissue damage. These studies indicated that pH extremes above 11.5 and below 2.5 generally are not tolerated by human corneal tissue. EPA decided that basing pH limits on eye tissue damage was unnecessarily conservative. Thus, eye damage is a hazard not fully addressed by the corrosivity characteristic. The corrosivity characteristic also was intended to prevent harm to ecological receptors caused by the release of hazardous wastes with high- or low-pH. In discussing aquatic life in the original background document, EPA noted that the optimum pH range for freshwater fish is 6.5 to 9.0 and that an increase or decrease of 2 pH units beyond the optimum range causes severe effects. Levels of pH of 11.0 or greater and 3.5 or less are fatal to all species of fish. EPA also noted that altering surface water pH can reduce the productivity of food organisms essential to fish and wildlife. The pH limits of the corrosivity characteristic (2.0 and 12.5) are well beyond the safe range for aquatic life, but wastes presumably would be significantly diluted before the point of exposure to aquatic life. EPA did not conduct a risk assessment of such potential hazards (e.g., modeling the pathway of waste released to surface water and exposure to aquatic life) and thus it is not known under what circumstances high- or low-pH wastes could cause harm to aquatic receptors. Corrosion of materials other than steel is not directly addressed. In the second part of the corrosivity characteristic, EPA uses steel corrosion as an indicator of corrosivity. EPA adopted this aspect of corrosivity because "wastes capable of corroding metal can escape from the containers in which they are segregated and liberate other wastes." EPA adopted DOT's corrosion standard, noting that the rate at which a waste corrodes a material commonly used in container construction (low carbon steel) is a suitable measure of its hazardousness. The reliance on the steel corrosion rate may create a potential gap if there are plausible mismanagement scenarios where wastes are stored, transported, or disposed in containers made from materials more easily corroded than low carbon steel (e.g., plastic by organic solvents) or are disposed in solid waste management units lined with materials such as clay or synthetics. Also, there may be a potential gap in the characteristic if waste management scenarios result in conditions where wastes are subject to higher temperatures than the 130oF test temperature. Solubilization of hazardous constituents. The corrosivity characteristic also was intended to address the potential for high- and low-pH materials to solubilize potentially toxic waste constituents. EPA offers the example that a drop in pH from 4.0 to 2.0 increases the solubility of red mercury oxide or chromium hydroxide in water approximately 100 times. The general concern is for inorganic ions that may be converted to more soluble species. This characteristic does not address the potential solubilization of organic constituents by organic liquids such as solvents, nor does it address the formation of non-aqueous phase liquids (NAPLs) by such materials. EPA considered including a solvents "override" in the TC characteristic, but did not do so. The solvents override would have caused wastes with high concentrations of solvents to be classified as hazardous on the basis of potential NAPL formation. The issue of NAPL formation is discussed in more detail in Chapter 5. Lack of coverage of sensitizers and irritants. At least two types of materials that may pose potential hazards to humans through direct contact are not included in the corrosivity characteristic or any other characteristic: irritants and sensitizers. OSHA includes irritants in its definition of health hazard and defines irritant as a material that is not corrosive, but which causes a reversible inflammatory effect on living tissue by chemical action at the site of contact. A chemical is a skin irritant if, when tested on the intact skin of albino rabbits by the methods of 16 CFR 1500.41 for four hours exposure or by other appropriate techniques, it results in an empirical score of five or more. A chemical is an eye irritant if so determined under the procedure listed in 16 CFR 1500.42 or other appropriate techniques. (See 29 CFR 1910.1200.) OSHA also includes sensitizers in its definition of health hazard. A sensitizer is defined as a material that causes a substantial proportion of exposed people or animals to develop an allergic reaction in normal tissue after repeated exposure to the chemical. (See 29 CFR 1910.1200.) This analysis did not identify any specific non-hazardous industrial wastes that are irritants or sensitizers. Irritants and sensitizers, however, are common categories of materials and these properties are often identified in laboratory testing of materials. A major issue regarding this potential gap is whether any irritants and/or sensitizers pose a hazard in wastes that reaches the statutory level of hazard intended to be covered by RCRA Subtitle C. 3.3.3 Potential Gaps Related to Corrosivity Test Methods Use of pH as an indicator has limitations. EPA chose pH as a measure of corrosivity because "wastes exhibiting low or high pH can cause harm to human tissue, promote the migration of toxic contaminants from other wastes, react dangerously with other wastes, and harm aquatic life." The ability of some substances to damage human tissue, however, may not be adequately indicated by a pH measurement. Other regulatory and advisory bodies (e.g., DOT, OSHA, Basel Convention) use criteria based on full thickness destruction of human skin. Since the original rulemaking in 1980, Method 1120 (Dermal Corrosion) has been developed commercially. The dermal corrosion assay system is an in vitro test method which determines the corrosive potential of a substance toward human skin. It can be used to test liquids (aqueous or non-aqueous), solids (water soluble or insoluble), and emulsions. Method 1120 is essentially the same method that DOT uses. It replaced previous tests (e.g., Draize test) that used live animals with a test that uses a proprietary synthetic pig collagen material. 3.4 Reactivity 3.4.1 Definition of Reactivity The reactivity characteristic is "intended to identify wastes, which because of their extreme instability and tendency to react violently or explode, pose a problem at all stages of the waste management process." This characteristic was intended to reduce physical risks to transportation and disposal workers and to avoid incidents that could result in the release of toxic constituents into the air consequent to an explosion or violent reaction. 40 CFR 261.23 states that a solid waste exhibits the characteristic of reactivity if a representative sample of the waste has any of the following properties: ! Is normally unstable and readily undergoes violent change without detonating; ! Reacts violently with water; ! Forms potentially explosive mixtures with water; ! When mixed with water, generates toxic gases, vapor, or fumes in a quantity sufficient to present a danger to human health or the environment; ! Is a cyanide or sulfide bearing waste which, when exposed to pH conditions between 2 and 12.5 can generate toxic gases, vapors or fumes in a quantity sufficient to present a danger to human health or the environment; ! Is capable of detonation or explosive reaction if it is subjected to a strong initiating source or if heated under confinement; ! Is readily capable of detonation or explosive decomposition or reaction at standard temperature and pressure; or ! Is a forbidden explosive as defined in 49 CFR 173.51, or a Class A explosive as defined in 49 CFR 173.53 or a Class B explosive as defined in 49 CFR 173.88. 3.4.2 Potential Gaps Related to Definition of Reactivity The Definition is broad and lacks specificity. In discussing the reactivity characteristic in the 1980 final rule, EPA stated that "the definition was intended to identify wastes which, because of their extreme instability and tendency to react violently or explode, pose a problem at all stages of the waste management process." EPA noted that the reactivity characteristic encompasses a diverse class of physical properties and effects and overlaps somewhat with the ignitability characteristic. Some commenters argued that the definition was vague. They advocated using a quantitative definition accompanied by testing protocol(s). EPA responded that "the prose definition should provide generators with sufficient guidance to enable them to determine whether their wastes are reactive." EPA argued that most generators whose wastes are dangerous because they are reactive are well aware of this property and such wastes usually are generated from reactive feedstocks and/or processes producing reactive products or intermediates. EPA further stated that problems posed by reactivity appeared to be confined to a fairly narrow category of wastes. Theoretically, the reactivity characteristic could be clarified and made consistent with other programs (especially DOT) by developing more specific definitions of general terms such as "normally unstable," "violent change," "potentially explosive," "reacts violently with water," "readily capable of detonation," and so forth. Other programs include more specific definitions. For example, as shown in Exhibit 3-3, DOT has adopted definitions of spontaneously combustible material and dangerous when wet material, which could be used to clarify the RCRA characteristic. Specifically, DOT identifies an ignition time and violent reaction rate. Likewise, OSHA defines pyrophoric, unstable reactive, and water reactive, specifying reactive conditions such as shocks, pressure, and temperature which define the characteristic. The Basel Convention also defines similar terms. References to DOT regulations are outdated. Forbidden explosive are no longer defined in 49 CFR 173.51. The current DOT regulations define forbidden explosives at 49 CFR 173.54. Other explosives are defined at 49 CFR 173.50. 49 CFR 173.88 no longer exists. 3.4.3 Potential Gaps Related to Reactivity Test Methods Reactivity characteristic lacks test method(s). When the Agency promulgated the reactivity characteristic in 1980, no available tests were identified for use in defining the reactivity characteristic because: ! They were too restrictive and were confined to measuring how one specific aspect of reactivity correlates with a specific initiating condition or stress. ! Testing the reactivity of a sample does not necessarily reflect reactivity of the waste, because reactivity varies with properties including mass and surface area. ! Most available tests required subjective interpretation of results. ! Existing methods were not developed for testing wastes. Although EPA has identified a test method (Method 9010) for reactive sulfide and/or cyanide bearing wastes, the Agency has not identified suitable test methods to fully define the reactivity characteristic. 3.5 Potential Gaps Associated with the Toxicity Characteristic 3.5.1 Definition of Toxicity Characteristic The toxicity characteristic was designed by EPA to reduce risks to public health from chronic exposures to groundwater contamination caused by releases of toxic waste constituents. The Agency found "persuasive evidence that the contamination of groundwater through the leaching of waste contaminants from land disposed wastes is one of the most prevalent pathways by which toxic waste constituents migrate to the environment." The legislative history of RCRA and EPA's case studies of damages from hazardous waste management were cited as support for focusing the toxicity characteristic exclusively on groundwater pathway risks. EPA originally listed 14 contaminants as part of the toxicity characteristic. Subsequently, EPA added another 26 substances to the list, as shown in Exhibit 3-4. These 40 TC chemicals were selected because: ! The chemicals were included on the 40 CFR Part 261 Appendix VIII list of hazardous waste constituents that have been "shown to have toxic, carcinogenic, mutagenic, or teratogenic effects," and ! Appropriate chronic toxicity information had been developed and adequate fate and transport data were available to allow the modeling of groundwater fate and transport for each constituent.Thus, EPA found these chemicals to be among those posing the greatest risk to humans from chronic groundwater exposure. The remainder of Section 3.5 evaluates the TC in five steps: ! Section 3.5.2 examines whether new data on the toxicity and persistence of TC analytes and updated groundwater transport modeling techniques would result in allowable TC leachate concentrations different from those established in 1990. ! Section 3.5.3 presents screening-level exposure and risk modeling methods and results that are used to evaluate whether the current TC chemicals could pose risks to human health and environmental receptors through the inhalation pathway. ! Sections 3.5.4 and 3.5.5 evaluate potential risks from TC chemicals to human health through surface water pathways and indirect pathways, respectively. These risks are evaluated by comparing toxicity and fate and transport values to defined risk-related criteria, both singly and in combination, and by reviewing the results of the Agency's multipathway risk modeling for the analytes that was performed as part of the proposed Hazardous Waste Identification Rule (HWIR-Waste) development. ! Sections 3.5.6 and 3.5.7 evaluate the potential for acute adverse health effects of exposures to TC analytes and potential risks to ecological receptors from TC analytes, respectively. 3.5.2 Changes in Groundwater Pathway Analysis This section of the Scoping Study explores two issues related to the current TC regulatory levels: (1) whether new toxicity data indicate a potential need to revise the regulatory levels; and (2) whether, in light of recent developments in groundwater modeling techniques, the current dilution and attenuation factor (DAF) value of 100 still provides a reliable basis for assuring that human health is protected against risks from groundwater exposures to TC chemicals. Revisions to MCLs and Toxicity Criteria The toxicological bases for the establishment of TC analyte regulatory levels were chronic toxicological and health-based regulatory criteria that were current at the time of promulgation. These included Safe Drinking Water Act Maximum Contaminant Levels (MCLs), Reference Doses (RfDs), and Risk-Specific Doses (RSDs) based on ingestion pathway Cancer Slope Factors (CSFs). For almost all of the TC analytes, these values have not changed since 1990. The few changes have included: ! A reduction in the RfD for p-cresol by a factor of ten and the withdrawal of the MCL of 50 ug/l for lead and its replacement with an Action Level of 15 ug/l. For cresol and lead, the reductions in RfDs and promulgation of Action Levels indicate that the toxicological evaluation of these chemicals has changed such that the TC regulatory levels may be less protective than was originally intended. The changes for both of these analytes were an order of magnitude or less. ! The withdrawal of the MCL for silver, with its replacement by an SMCL at the same value. This change simply means that the critical toxic effect for silver (argyria, which is the collection of dark pigment in the skin and mucous membranes) has been downgraded from a health effect to a cosmetic effect. ! In addition, the RfD for pentachlorophenol has been reduced from 2 mg/l to 3x10-2 mg/l. More importantly, since the TC was revised, the Agency has promulgated a cancer slope factor for this compound, which is a suspect human carcinogen. Thus, the critical toxic endpoint has been changed from non-cancer to cancer induction. The promulgation of the Cancer Slope Factor implies that a much lower TC regulatory level (about 1000 times lower) would be needed to achieve the same level of protection against cancer risks as originally intended when the TC was promulgated. Advances in Groundwater Modeling To develop the existing TC regulatory levels, the Agency used the EPAMCL model to estimate the likely extent of dilution after the release of waste constituents from waste management units during their transport to the nearest drinking water wells. These calculations were conducted for municipal solid waste landfills and Subtitle D surface impoundments, taking into account the geochemical properties of the constituents, the size and configuration of the units, the vadose zone and groundwater regimes beneath the units, and the distribution of distances in the downgradient direction to the nearest drinking water well. Groundwater regimes were defined using distributions of transport parameter values typical of conditions throughout the United States. Receptor wells were assumed to be in the groundwater plume at a distribution of distances derived from a Subtitle D facility survey. Simulation methods were used to derive estimates of dilution-attenuation factors (DAFs) for each constituent and each type of waste management unit. After reviewing the results, the Agency elected to calculate acceptable leachate concentrations (regulatory levels) for each TC analyte using a single DAF value of 100. In other words, the threshold leachate concentration of each analyte above which wastes would be identified as TC hazardous was set equal to allowable drinking water concentration or other benchmark (10-5 cancer risk or Hazard Quotient (HQ) = 1.0) for the analytes multiplied by 100. Since the TC was promulgated, the Agency has continued to use the same general approach to evaluate the groundwater transport of pollutants in developing RCRA regulations. The exact techniques used in this modeling, however, have changed significantly. In recent rulemakings, the Agency has used an updated version of the EPAMCL model, known as EPACMTP, to derive constituent-specific DAFs for a wide range of pollutant releases from hazardous and non-hazardous waste management units. This model employs many of the same basic transport algorithms as the EPAMCL, with several important differences, including the following: ! The EPACMTP model uses a detailed metals speciation model (MINTEQA) to estimate leachate concentrations from wastes of defined ionic composition, whereas the EPAMCL model did not employ such a model; ! The EPACMTP, unlike EPAMCL, can model the adsorbtion to soil and transformation of organic waste constituents by hydrolysis into more toxic (or less toxic) transformation products; ! The EPACMTP directly simulates the interface between the saturated and vadose zones; ! The EPACMTP model can simulate groundwater mounding under management units, whereas the EPAMCL could not; and ! The EPACMTP model provides more flexibility in modeling finite, versus infinite, source terms. Recent applications of the model also used somewhat different assumptions regarding waste and facility characteristics, hydrogeological regimes, climatology, and receptor locations than those used in the development of the TC. Therefore, it is not possible, except in a very general way, to simply compare the DAF value used in establishing the TC allowable leachate concentrations with the constituent-specific DAF values for the same constituents derived in the subsequent analyses. In addition, DAF values derived for metals using the EPACMTP vary with the initial concentration of the constituent in the waste, because the model incorporates saturable binding and transport phenomena. In contrast, the DAFs derived using the EPAMCL model are concentration-invariant under most conditions. Recently, EPA has employed the EPACMTP model in two major regulatory development efforts. ! EPA applied the model in its development of proposed risk-based exit levels for the Proposed Hazardous Waste Identification Rule for Process Waste (HWIR-Waste). In that analysis, EPACMTP was used to back-calculate concentrations of constituents in wastes and in waste leachate corresponding to specific risk levels through groundwater exposures. The Agency is currently revising the proposed HWIR-Waste exit level groundwater risk modeling methods in response to comments from the Science Advisory Board and from other technical reviewers. Thus, the results of this modeling presented in this Scoping Study should be regarded as preliminary. ! In the Phase IV LDR regulatory development effort for mineral processing wastes, the model was used to derive constituent-specific DAFs for mineral processing wastes disposed of in surface impoundments and waste piles. The DAFs were then used to derive groundwater pathway risk estimates for exposure to waste constituents. The results of these analyses have been used to evaluate the extent to which changes in modeling techniques may have affected the assessment of groundwater fate and transport relative to the assessment used to derive the TC regulatory concentrations. As noted previously, a simple comparison of DAF values and/or calculated risk levels from the different modeling efforts is not possible without further analysis since the more recent modeling employs different groundwater transport models and different assumptions regarding facility characteristics, groundwater regimes, and receptor locations. In the case of the mineral processing risk assessment, for example, DAF values were derived specifically for facility sizes representative of the mineral processing industry, rather than Subtitle D management units. In addition, groundwater modeling was performed using climatologic data primarily from drier regions where many mineral processing facilities are located. While Subtitle D facilities were used to calculate releases for the HWIR-Waste proposal, the receptor wells were assumed to be distributed uniformly in the downgradient direction, instead of being confined to the plume. More importantly, the proposed exit levels were derived using a carcinogenic risk target of 10-6, rather than 10-5, and the 90th percentile, rather than the 85th percentile, estimates of risk were used. Using the 90th instead of 85th percentile of the risk output results in estimating higher risks for a given receptor for a given constituent concentration and in more stringent (lower) exit levels. Thus, the proposed HWIR-Waste risk calculations, especially for carcinogens, are substantially more conservative in several important respects than those used to derive the TC regulatory levels. In the mineral processing risk assessment, DAF values were derived for eight of the TC analyte metals. For waste piles, the DAF values for the majority of the TC metals were considerably higher than 100, the highest value being 1x1030 for lead. Barium, with a DAF value of 54, was the only metal for which the mineral processing waste pile DAF was less than the value of 100 used in the derivation of the TC regulatory concentrations. These results imply that the DAF value of 100 used in the TC derivation remains conservative for most metals when compared to values derived for this population of facilities. The situation is different, however, if the DAF values derived for mineral processing surface impoundments are used as a basis for comparison. In this case, the majority of the DAF values for the TC metals were less than 100. This finding suggests that the DAF value of 100 used to derive the TC regulatory levels may not provide adequate protection against groundwater risks from surface impoundments, which are the most frequent management type employed for non-hazardous industrial wastes. The large difference in DAF values for the two types of management units can be explained partly in terms of the comparative aridity of the locations for which DAFs were calculated. Where little moisture was available to drive transport from the waste piles through the vadose zone, DAF values tended to be high. In contrast, the surface impoundments provided a water supply that drove transport through the vadose zone into groundwater. The extent to which this effect would be seen in moister regions of the country is not clear. The HWIR-Waste proposed groundwater exit level calculations for the TC analytes are summarized in Exhibit 3-5, and compared to the TC regulatory levels. The majority of the exit levels are considerably lower (more stringent) than the corresponding TC regulatory levels. In 4 cases, the TC levels are comparable to or less than the exit level. For 9 analytes, the ratio of the TC regulatory level to the exit level is between 1 and 10. For 12 analytes, this ratio is between 10 and 100; for 5 analytes, the ratio is between 100 and 1,000; and for 6 analytes, the ratio is greater than 1,000. This distribution confirms that, generally, the assumptions and modeling approaches used to derive the HWIR-Waste proposed exit levels lead to somewhat more conservative or more protective results than those used to derive the TC regulatory levels. This conclusion holds true, even taking into account that the cancer risk target is 10-fold lower for setting some of the proposed exit levels than was used for setting the TC levels. For all but a few of the carcinogens among the TC analytes, the proposed exit levels are far more than 10 times lower than the corresponding TC regulatory levels. Thus, some other factors account for a significant proportion of the conservatism in these calculations. Some of this conservatism may be due to differences in modeling assumptions, rather than modifications in modeling techniques. For example: ! The HWIR-Waste proposed exit levels were derived to be protective of 90th percentile receptors, while the TC levels were set to be protective of 85th percentile receptors. ! As shown in Exhibit 3-5, some HWIR-Waste proposed exit levels were driven by exposure pathways other than groundwater. ! The proposed HWIR-Waste exit levels and the TC regulatory levels were designed for different purposes. The TC levels are designed to provide a method for identifying wastes that would otherwise be non-hazardous, while the proposed HWIR-Waste exit levels would relieve wastes previously identified as hazardous from stringent regulatory control. These issues are discussed in more detail in Sections 3.5.3 and 3.5.4. Other differences in modeling assumptions, such as the retention of constituents in waste management (loss terms) in TC modeling only and the differences in the assumed location of wells relative to the contamination source, influence the results in the other direction. Summary. Based on the preceding analyses, only general conclusions can be drawn about whether there are any significant gaps in the TC associated with the specific regulatory levels set for individual constituents. The wide range in the mineral processing DAF values illustrates the high degree of variability associated with specific groundwater modeling assumptions, and does not necessarily indicate whether the DAFs should be considered less or more protective than when they were originally derived. The HWIR-Waste proposed exit level calculations, on the other hand, suggest that the application of more recent modeling techniques might result in more conservative transport calculations. Some, but not all, of this greater level of protectiveness reflects a policy decision by the Agency regarding what proportion of receptors should be protected to the target risk level. In addition, advances in modeling techniques and differences in specific input assumptions also affect the differences in the apparent levels of protectiveness. 3.5.3 Potential Inhalation Pathway Risks Associated with TC Analytes This section investigates the general level of protectiveness of the allowable TC concentrations against direct inhalation, a risk that the TC was not specifically intended to protect against. EPA analyzed this issue by performing screening-level risk calculations for long-term air releases of the TC constituents from Subtitle D facilities. EPA used the CHEMDAT8 model using facility characteristic parameters for surface impoundments and land application units (LAUs). Release estimates for all of the organic TC analytes were developed for two scenarios. ! In the first scenario, releases were estimated from the same "high-end" surface impoundments and LAUs that were modeled in the proposed HWIR-Waste exit level modeling. ! The second scenario modeled releases from the "central tendency" impoundments and LAUs, which were considerable smaller and shallower than the high-end units. In both release scenarios, the concentrations of the organic TC analytes were assumed to be at the TC regulatory level for liquid wastes in surface impoundments and at 20 times the TC levels for nonwastewaters in land application units. The latter assumption roughly estimates the maximum concentration of the TC analyte that could be present without the waste being hazardous, assuming efficient leaching using the TCLP. For analytes that do not leach well, this approach may underestimate exposure concentrations and risks associated with air releases from non-hazardous industrial wastes, since nonwastewaters with high concentrations of constituents would not be identified as hazardous by the TCLP. Average releases to air were calculated for an assumed 40-year facility life-span under both scenarios. The basic approach and input assumptions used in the modeling are summarized in Exhibit 3-6. The organic TC analytes for which releases were modeled vary widely in molecular weight, vapor pressure, Henry's Law constant, and other physical properties that affect releases to air. Thus, the extent of release of these chemicals to air from land disposal facilities might be expected to differ widely. This is true to some extent; but, as can be seen in Exhibit 3-7, the estimated release of these compounds from land application units and surface impoundments over the expected facility life-span varies only moderately. In the case of the high-end land application units, between approximately 7 percent and 100 percent of the chemicals entering the units are released to the air over the facility life. The average proportion of the analytes released from these units was 81.6 percent, and the calculated releases were greater than 95 percent for two-thirds of the organic TC analytes. The results were similar for the central tendency LAU. Releases ranged from 27 to 100 percent of the analytes, and the average proportion released was 96.3 percent. The explanation for the predicted higher proportional releases from the central tendency LAU is not clear, but may be related to the shallower tilling depth assumed for the central tendency unit (0.2 compared to 0.3 meters). The proportions of the TC analytes released from surface impoundments are shown in the final two columns of Exhibit 3-7. The releases ranged from 6 to 77 percent of the applied total per year for the high-end impoundment, with an average release of 55.5 percent per year. Proportionate releases were again higher from the central tendency unit, ranging from 15 percent to 88 percent, with an average of 71.2 percent released annually. Similar to the situation for the LAUs, the higher proportional releases from the central tendency unit may be due to its considerably shallower depth (2 meters) compared to the high-end unit (7 meters). The limited impact of a chemical's Henry's Law constant on air releases is somewhat surprising in light of the broad spectrum of solubility and vapor pressure reflected in the chemicals modeled. Perhaps it can best be understood as indicating that, in the long run (a year or more), a high proportion of any of these organic chemicals placed in uncovered land management units will be released to the air, provided other removal pathways are not important. In actual practice, some land application units are covered to some extent, and other removal processes, such as leaching, biological and chemical degradation, and binding to soil or sediment, compete to reduce air emissions significantly. EPA calculated chronic risks from inhalation pathway exposures for all organic TC analytes. To calculate exposure concentrations, EPA multiplied release estimates by the long-term fenceline dispersion coefficients used in the proposed HWIR-Waste exit level calculations for the high-end and central tendency surface impoundments and LAU releases. The fenceline dispersion coefficients are used to represent the nearest credible residential exposure locations, in keeping with the proposed HWIR-Waste risk methodology. Exposure durations are assumed to be 30 years in the high-end exposure release and exposure scenario, and 9 years in the central tendency scenario. Exhibit 3-8 summarizes the results of the screening-level risk estimation for the TC analytes having inhalation pathway toxicity values in IRIS or HEAST (as discussed below). The first eight columns of the results indicate whether the estimated lifetime cancer risk associated with managing the analytes at the TC (or the TC multiplied by 20) concentrations in the various management units would be greater than 10-5 or if the inhalation pathway hazard quotient (HQ) would exceed 1.0. These risk threshold values are the same as those used in developing the TC analyte concentrations for groundwater exposures. For the 16 TC analytes with IRIS Unit Risk values, inhalation pathway cancer risks greater than 10-5 are not predicted for any of the TC analytes released from the central tendency surface impoundment. In contrast, cancer risks above 10-5 are predicted for 12 of these analytes released from the high-end impoundment. None of these analytes released from the central tendency LAU would result in an inhalation pathway risk greater than 10-5. Releases of four analytes (chloroform, 1,4-dichlorobenzene, 1,1-dichloroethylene, hexachlorobenzene, and toxaphene) from the high-end LAU would result in risks above this level. Of the four TC organics with inhalation RfCs, hazard quotients greater than 1.0 were calculated for three analytes (chlorobenzene, methyl ethyl ketone, and nitrobenzene) released from the central tendency surface impoundment. When releases are modeled from high-end impoundments, the one additional chemical (1,4-dichlorobenzene) also has an HQ greater than 1.0. Exactly the same pattern is seen for LAUs. These results indicate that, under assumptions of no degradation or release to other pathways, the cancer risks and non-cancer hazard indices associated with management of some of the organic TC analytes may be above levels of concern previously used in amending the TC. These risks may be overestimated if significant amounts of pollutants in waste are released through other pathways or are degraded biologically or chemically. For this reason, EPA used the proposed HWIR-Waste database to identify the TC analytes that are persistent in soil or water. As shown in the last two columns of Exhibit 3-8, most of the organic analytes that exceed the air risk targets under the assumption of no degradation are, in fact, not very persistent in either soil or water. Using a cutoff value for degradation rate constants of 0.5 year-1, which corresponds to a half-life in soil or water of about 17 months, only 3 of the organic TC analytes are expected to be very persistent. The relatively short half-lives in water or soil may reduce the potential concern for inhalation pathway risks associated with the other TC analytes to the same extent. These results illustrate the need for more detailed, site-specific modeling of all of the transport and degradation processes. Risks were calculated in this analysis for only those TC analytes having inhalation pathway toxicity values (Reference Concentrations or Unit Risk values) in IRIS. If instead inhalation pathway toxicity values were derived for the rest of the TC analytes from ingestion pathway values and used in similar risk calculations, the number of chemicals for which cancer risks and particularly non-cancer risks would exceed levels of regulatory concern would be much higher. These results have not been included in Exhibit 3-8 because EPA considers the level of uncertainty associated with such procedures to be unacceptably high. Evaluation of the proposed HWIR-Waste exit level calculations for the TC analytes confirms the potential concern for nongroundwater pathways. For some of the TC analytes, the HWIR-Waste proposed exit level calculations indicated that non-groundwater pathways are significant. Findings include the following: ! For 9 of the TC analytes, pathways other than human groundwater exposure drove the establishment of proposed exit levels. ! For six of the analytes, ingestion of contaminated milk or vegetables was the highest-risk exposure pathways. ! For one of the pollutants, the driving non-groundwater exposure pathway was direct inhalation. ! For two analytes, ecological risks rather than health risks drove the derivation of proposed exit levels. In all of these cases, the initial release was to air through volatilization. These indirect pathway risks will be discussed in more detail in the following sections. 3.5.4 Potential Risks from Surface Water Exposures This section investigates the general level of protectiveness of the TC regulatory levels against surface water exposures, a risk that the TC was not specifically intended to address. Waste constituents could be released to surface water from land management units through several mechanisms: ! Discharge of groundwater contaminated by leachate from waste management units; ! Transport of waste constituents to surface water bodies by runoff and overland transport of wastes released from the management unit; ! Direct releases through overland runoff of liquid wastes from surface impoundments; and ! Volatilization of constituents from land-based units, followed by deposition onto surface water or onto soil that eventually finds its way into surface water bodies. The surface water exposure pathways of potential significance for humans include direct consumptive use (i.e., ingestion and dermal contact with domestic water) and dermal contact and incidental ingestion of the surface water associated with recreational exposures. If the contaminants are persistent in sediment, dermal contact and incidental ingestion of small amounts of sediment also are possible exposure pathways. All of these release and exposure pathways have been analyzed in the development of hazardous waste management regulations and in other contexts. The experience gained in these exercises has led the Agency to a number of general conclusions regarding the importance of surface water exposures for human health risks: ! For common waste management practices, surface water exposure cannot be automatically ruled out as insignificant in comparison to groundwater, inhalation, and other indirect pathways. ! The significance of surface water releases depends heavily on the management practices employed by a facility and the specific interactions between surface water and groundwater at the facility. ! Generally, groundwater discharge significantly affects surface water quality only where groundwater constitutes a significant proportion of the total surface water in a water body. This pathway may be important for very large management units that generate large amounts of leachate, but usually significant surface water quality impacts are limited to relatively small streams adjacent to management units and to on-site or adjacent ponds derived mainly from leachate. ! Exposure to volatile contaminants in surface water is generally limited because these contaminants are depleted rapidly from surface water through volatilization. Air releases from surface water may themselves be significant from a health standpoint. Usually, however, volatilization from the management unit itself dominates, unless the unit is covered. ! Incidental ingestion and dermal contact with contaminated sediment tend not to be significant exposure pathways for humans, because of their infrequency and the relatively small amounts of contaminated sediment contacted (but see below). ! Indirect pathway exposures may be of concern, however. The contaminants that persist in sediment and have a high capacity to bioaccumulate and bioconcentrate are often the most significant contributors to human health risks. This capacity may overcome the high dilution factors often associated with releases to surface water. These persistent pollutants most often reach human receptors through the consumption of contaminated fish or shellfish. In evaluating the potential risks associated with proposed HWIR-Waste chemicals, EPA identified contaminants for which surface water pathways were of potential concern. Whether or not the surface water pathway was a concern depended on the waste treatment scenario. For wastewaters managed in surface impoundments, surface water was not a human health risk for any of the TC analytes. All of the proposed exit levels driving non-groundwater pathways for humans were associated with volatilization followed by deposition on soil and did not involve surface water. For nonwastewaters disposed in land application units and waste piles, however, more than 50 percent of the proposed exit levels for the HWIR-Waste constituents are driven by pathways involving surface water exposures. The driving (highest-risk) pathways were approximately equally divided among the contaminants between overland runoff followed by fish uptake, and overland runoff followed by surface water ingestion. These results must be interpreted cautiously. The analysis of the proposed HWIR-Waste exit levels cited above gives only a comparative, not an absolute, indication of the importance of the surface water exposure pathways for waste piles and land application units. The proposed exit levels calculated for these types of units are generally higher than those associated with surface impoundments, for example, indicating that the magnitude of the risks from wastes piles and land application units are, in general, lower than those associated with surface impoundments. Summary. The preceding analysis has explored the possibility that significant risks to health or the environment may be associated with exposures through surface water pathways. While a number of theoretical arguments suggest that such releases might be important under only a relatively narrow range of conditions, the proposed HWIR-waste modeling results indicate that these pathways may well be significant for some TC analytes disposed as non-hazardous industrial wastes. The possibility that the surface water releases and exposures represent a potential gap in the TC, especially for persistent and bioaccumulative chemicals, cannot be ruled out. 3.5.5 Potential Indirect Pathway Risks from TC Analytes "Indirect" pathways are any pathways involving more than one environmental medium (e.g., groundwater, air, surface water, soil, sediment, and biota) between the release and the exposed receptor. The initial release may be to any medium. Indirect exposure pathways often, but not always, involve uptake of environmental contaminants by living organisms, which, in turn, are consumed by human or other receptors. Some of the pathways discussed in the previous sections, such as groundwater releases to surface water, are, strictly speaking, indirect. This section, however, emphasizes pathways involving potential long-range transport of persistent pollutants and pathways involving biota (crops, fish, or livestock) prior to human exposures. Persistence, properties facilitating physical transport, and the potential to bioaccumulate in the environment are critical in the indirect pathways, and the physical/chemical and environmental fate properties of constituents significantly determine their movement through such pathways. Exhibit 3-9 summarizes some important physical, chemical, and environmental fate properties of the TC analytes relating to persistence, partitioning behavior between environmental media, and bioaccumulation. For each parameter, the exhibit compares each constituent's value to a criterion or cutoff value that roughly indicates whether the parameter will strongly influence the transport and partitioning of the chemical in the environment in a multipathway analysis. The derivation of these criteria are discussed in more detail in Section 4.3.2. The first column identifies TC analytes with a high Koc (high Kd for metals), generally indicating a propensity to bind to soils. A high value means that chemicals will leach only slowly to soil, but would bind to particulates if they were released through runoff or into the air. Essentially all of the chemicals with Koc values above 10,000 are pesticides. In addition, the majority of the TC metals would be expected to bind to some extent to particulates. The next column on Exhibit 3-9 shows the Henry's Law constants (kH) for the TC analytes, with values above 10-5 generally indicating a moderate to high capacity to volatilize from soil-water systems, which may be the first step in an indirect exposure pathway. About half (19) of the TC analytes have kH values above 10-5. As discussed in Section 3.5.3, variations in Henry's Law constants did not strongly effect the predicted long-term release of the TC analytes from surface impoundments and waste piles. Short-term releases, however, may be much more dependent on this parameter. The next two columns address the persistence of TC analytes in soils, water, and air. Data in these two columns summarize information from the proposed HWIR-Waste database on the estimated half-life of chemicals in air and the overall degradation rate constants in soils and surface water. Four of the TC analytes are identified as having long half-lives in air and 12 are persistent (have low degradation rate constants) in soil and/or surface water. The air half-life values must be interpreted cautiously, as the proposed HWIR-Waste database contains this information on only about 20 chemicals. Metals and many high-Koc organics would also be expected to have long half-lives in air if they were bound to particulates. As discussed earlier, the TC analytes with long half-lives in soil/water systems include primarily the metals and chlorinated pesticides. The final three columns of Exhibit 3-9 consider the propensity of TC analytes to bioaccumulate in aquatic and terrestrial ecosystems. The plant-soil bioconcentration factor (BCF) is an estimate of the typical ratio of the concentration of a constituent in soil to the concentration in a particular kind of plant (in this case, forage plants consumed by beef and dairy cattle). Similarly, the beef biotransfer factor is an estimated typical ratio of the concentrations of pollutants in the diet of beef cattle to the resultant concentrations in edible tissue. Finally, the BCF and bioaccumulation factor (BAF) values for fish represent the typical ratios of pollutant concentrations in surface water to that in fish tissue, considering only water exposures or considering all pathways, respectively. (These value tend to be quite similar for most chemicals.) Although the exhibit indicates that several constituents may bioaccumulate from soil to forage plants, in reality, only 2,4,5-trichloropropionic acid (Silvex) has a very high bioconcentration potential. The value of the forage biotransfer factor for this pesticide is five orders of magnitude greater than for any other chemical (greater than 106). Generally the same chemicals have high beef biotransfer factors, fish BCFs, and BAFs, with barium, mercury, and lindane bioconcentrating only in aquatic systems, and arsenic, chromium, selenium, and silver being significant for the beef exposure pathways alone. Summary. These single comparisons indicate the significant potential for many TC analytes to be transported through multiple media to reach the ultimate receptors. The data in Exhibits 3-8 and 3-9 show that the chlorinated pesticides (i.e., chlordane, endrin, heptachlor, heptachlor epoxide, hexachlorobenzene, methoxychlor, pentachlorophenol, and toxaphene), chloroform, and hexachloro-1,3-butadiene have the potential to participate in indirect exposure pathways and have non-groundwater pathways as their driving pathways. In addition, several high-toxicity and persistent metals, such as mercury, arsenic, and lead, also are of potential concern. 3.5.6 Potential for Acute Adverse Effects of Exposures to TC Analytes The TC was originally established based on the need to protect individuals from adverse health effects due to chronic exposures to the TC constituents consumed in groundwater. This approach to protecting against groundwater exposure risks is conservative because the relatively long time scale involved in groundwater transport to receptors, under most reasonable assumptions, means that limiting concentrations in any time period to the low chronic risk-based levels also will protect against short-term adverse effects. Short transient exposures to high levels of groundwater contaminants are extremely uncommon. Before the concentration of a pollutant reaches the relatively high level required to cause acute effects, it generally will have exceeded the allowable chronic exposure level for a long period of time. This relationship may not apply to exposure through pathways not involving slow releases to groundwater. For example, the rapid evaporation of volatile chemicals from a ruptured container, the catastrophic release due to overtopping of a surface impoundment, or runoff erosion from an extreme storm event have the potential to result in short-term acute exposures to humans and environmental receptors. For this reason, EPA has evaluated the potential level of protectiveness of the TC against acute exposures. EPA evaluated the potential for adverse effects associated with acute volatilization of chemicals from land management units. The approach was analogous to the screening-level risk modeling for chronic exposure, except that the short-term releases were calculated and exposure concentrations were compared to short-term exposure standards. This analysis indicates that the short-term concentrations of all of the volatile TC analytes calculated at the fenceline were far below applicable short-term exposure standards (in this case, occupational exposure standards). This simple modeling does not unconditionally eliminate the possibility of adverse effects from acute exposures to the TC analytes. Unusual release events, such as fires, or explosions, could result in higher exposures than calculated assuming simple volatilization. In addition, high winds or other events could result in high concentrations of particle-bound metals and other non-volatile analytes. The potential for these kinds of release events strongly depends on specific waste characteristics, site conditions, and management practices. 3.5.7 Potential Risks to Ecological Receptors from TC Analytes Risks to non-human receptors are the final category of risks evaluated by EPA. Like the inhalation, surface water, and indirect pathway risks, they were not expressly factored into the derivation of the regulatory levels for the TC analytes. While a substantial number of the TC chemicals are toxic to ecological receptors, the protection of ecological receptors was not a specific concern in the rulemaking. This section discusses potential gaps in the TC characteristic associated with harm to ecological receptors. Many of the same factors that contribute to potential risks for human receptors also contribute to potential risks for ecological receptors. Generally, harm to environmental receptors requires release of chemicals from containment and transport to sensitive receptors without extensive degradation or extreme dilution, just as in the case of human health risks. Thus, the physical properties of chemicals that contribute to persistence and transport in the environment, as shown in Exhibit 3-9, are indicators of potentially significant risks for ecological receptors. The fact that most of the persistent chemicals with high bioconcentration potentials are also pesticides, which are toxic to certain plants, insects, or other animals, adds to the potential risks. The degree of protection of ecological receptors afforded by the TC leachate concentrations does not appear very high for many of the most toxic pesticides. Exhibit 3-10 compares the TC regulatory levels to two basic measures of potential aquatic toxicity, the acute and chronic Ambient Water Quality Criteria (AWQC) for the protection of aquatic life. It shows that, for many analytes, the allowable leachate concentrations are many orders of magnitude above the corresponding AWQC. The shaded boxes in the table identify TC analytes with regulatory levels greater than 1,000 times the AWQC. The chemicals falling into this category again include the chlorinated pesticides, chlorobenzene, lead, mercury, silver, and 2,4,5-trichlorophenol. This ratio indicates that if the TC analytes were released from wastes to groundwater and from there discharged to surface water, a dilution of at least 1,000-fold would be required to reduce the concentration to levels not harmful to aquatic biota. Such a scenario may be unlikely, however, because, as noted above, these chemicals tend to bind strongly to soil and do not move readily in groundwater. (As is discussed in more detail in Chapter 2, however, some of these chemicals were found in groundwater at concentrations above health-based levels in the descriptions of environmental releases from non-hazardous industrial waste management units.) In a more likely scenario, the high ecotoxicity of these chemicals means that runoff transport of particulate wastes at concentrations not considered hazardous under the TC could cause adverse effects in water bodies near management units. As noted above, the concern for runoff exposures is borne out to some extent by the proposed HWIR-Waste modeling, where proposed exit levels are driven by this pathway for disposal in waste piles and land application units. In the case of silver and endrin (two of the chemicals in shaded boxes in Exhibit 3-10), the proposed exit levels were driven by runoff releases to surface water. Summary. Based on these findings, it appears that the level of protectiveness of the TC is not very high for some non-human receptors. At a minimum, the ecotoxicity parameters suggest a potential concern associated with the aquatic toxicity of chlorinated pesticides, as well as a few other chemicals. The severity of these potential gaps is addressed in more detail in later chapters. 3.6 Potential Gaps Associated with TCLP This section reviews the technical basis for the Toxicity Characteristic Leaching Procedure (TCLP) and discusses potential problems associated with its use based on a brief review of available literature and data. Specifically, this section focuses on whether the TCLP fails to accurately predict releases from identified classes of wastes into groundwater and non-groundwater pathways. 3.6.1 TCLP Background In 1980, prior to development of the TCLP, the Agency adopted the Extraction Procedure (EP) to identify wastes likely to leach hazardous concentrations of particular toxic constituents into the groundwater under conditions of improper management. In 1986, the Agency proposed a modified leaching procedure, the TCLP, to replace the EP. The Agency promulgated the final rule on the application of the TCLP in 1990. In finalizing the TCLP, the Agency intended to improve the leachate test procedure and eliminate some of the analytical difficulties involved in the EP. The TCLP is used to quantify the extractability of certain hazardous constituents from solid waste under a defined set of laboratory conditions. This test is used to evaluate the leaching of TC metals, volatile and semivolatile organic compounds, and pesticides from wastes. In principle, this procedure simulates the leaching of constituents into groundwater under conditions found in a municipal solid waste (MSW) landfill. The TCLP, however, does not simulate the release of contaminants to non-groundwater pathways. The TCLP is most commonly used by EPA and state agencies to evaluate the leaching potential of wastes, and for determining toxicity. The TCLP is promulgated in Appendix II of 40 CFR Part 261.24(a) and has been designated as EPA Method 1311 in "Test Methods for Evaluating Solid Waste, Physical/Chemical Methods - SW-846." In the TCLP, liquid wastes (those containing less than 0.5 percent dry solid material) are "extracted" by filtering the wastes through a 0.6 to 0.8 : glass fiber filter. Non-liquid samples (those containing greater than or equal to 0.5 percent dry solid material) are: ! Reduced to a particle size of less than 9.5 mm (liquid, if any, is separated from the solid phase) and extracted with an acetate buffer solution with either a pH of 5 or an acetic acid solution with a pH of 3, depending on the alkalinity of the waste (wastes with a pH of 5 and above are extracted with the acetic acid solution); ! A liquid-to-solid ratio of 20:1 is used for an extraction period of 18 hours; and ! The leachate is filtered and combined with the liquid portion of the wastes, if necessary. Contaminant analyses then are conducted on the extracts of the liquids and non-liquids. 3.6.2 Limitations of the TCLP The Agency reviewed TC constituent and concentration data collected on releases from the non-hazardous industrial waste management units discussed in Chapter 2 (see Exhibit 2-5). These data show that, of the 15 TC constituents detected in at least three case studies, eight are present in groundwater at levels much higher than their TC levels. If the wastes passed the TCLP before being placed in the management units, this could indicate that the TCLP underestimated the long-term releases for certain classes of wastes. One of the major limitations of these data, however, is that they may not reflect current waste analysis or management practices. For example, some data represent releases from waste disposal that occurred prior to implementation of the TCLP, and thus some of the releases that exceed TC levels could be due to problems with other extraction procedures or to the lack of any testing procedure. Nevertheless, some site data (not reported in Chapter 2) exists that may represent problems with the TCLP. For example, the kiln residues from the treatment of spent aluminum potliners at one facility are disposed in a monofill as non-hazardous wastes. EPA approved a delisting petition for the kiln residue waste based on TCLP data that showed the target constituents in the TCLP extract to be below treatment standards (which, for the TC constituents, are lower than the TC regulatory levels). When the leachate from the monofill was analyzed, however, levels of arsenic were found to be higher than its TC level. Other hazardous constituents, including cyanide and fluoride, were also found at levels higher than those predicted by the TCLP. Several technical and practical issues have been raised by the regulated community and others regarding the applicability of the TCLP for identifying hazardous waste. A number of comments were submitted to the Agency in response to the June 13, 1986 proposal to replace the EP with the TCLP. The Agency responded to the comments in the final rule, but also decided to continue to address commenters concerns and further evaluate modifications to the TCLP. The Agency stated that further improvements in the TCLP will be proposed as they are developed. Subsequent to that rulemaking, additional concerns have been raised by commenters during later rulemakings (e.g., rules addressing newly listed or identified wastes). Some of the key issues regarding the TCLP identified from these comments on various rulemakings and from other sources are outlined below. TCLP underestimates leachate from some high alkaline wastes or environments. The high alkalinity of some wastes may make the TCLP an inappropriate predictor of leachate composition. For example, the addition of acid during the TCLP might not reduce the pH of high alkaline waste to the same level as would occur over time in the environment. Thus, long-term leachate concentrations of constituents that are insoluble at higher pH ranges may be underestimated in the TCLP leachate compared to the actual leachate from the industrial landfills where a long-term acid environment (e.g., from acidic rain water) is present. Some toxic metal constituents are more mobile at both the higher and the lower pH ranges. For example, studies show that leaching of metals such as cadmium, chromium, and lead typically is limited when the pH is in the range of about 8 or 9, but can increase significantly when the pH either increases or decreases. Thus, if a waste is highly alkaline (e.g., pH >11) and the TCLP acidic leaching medium lowers the pH to only about 8 or 9, then the concentrations of these metals in the TCLP leachate could be significantly lower than would occur from either a highly alkaline or a highly acidic environment (depending on a number of factors, such as characteristics of any co-disposed wastes, type of treatment, and characteristics of the soil and rain water). Several commenters to the June 13, 1986 TCLP proposal expressed concern regarding the application of the TCLP to alkaline wastes. They noted that no high alkaline wastes were included in the development of the TCLP and, therefore, no conclusions could be made concerning the actual behavior of these wastes. The MEP, described in the text box, is one test that the Agency and others use that may better simulate the long-term leaching behavior of such wastes. TCLP underestimates the leachate concentrations from oily wastes and some paint wastes. Several reports indicate that oily and some paint wastes tend to clog the filters used to separate the extract from the solids prior to analysis, resulting in under-reporting of the extractable constituent concentrations. Several commenters on the June 13, 1986 TCLP proposal noted that, in the development of the TCLP, the Agency tested only 11 wastes. These commenters argued that increasing the variety of wastes (to include oily wastes, organic chemical wastes, and municipal wastes) and the number of extractions performed could refine the TCLP and enhance its accuracy. TCLP may not accurately mimic conditions commonly found in non-hazardous industrial waste disposal. As discussed in the 1980 final EP rule, several commenters responding to the proposed use of the EP for evaluating the leaching of hazardous constituents argued that the co-disposal assumption is not applicable to wastes that are never co-disposed with municipal solid wastes and thus do not leach at the aggressive rates characteristic of co-disposal situations. Thus, the commenters stated, the leachate procedure does not simulate the conditions found in industrial waste monofills. In response, the Agency stated that most wastes, even those that are unlikely to be disposed in a municipal landfill, are likely to come into contact with some form of acidic leaching medium during their management histories or could otherwise encounter environments that could cause the wastes to leach comparable levels of toxic constituents. This same debate occurred during development of the TCLP, and it continues today. For example, the Lead Industries Association Inc., commenting on the Phase IV supplemental proposed rule, cited an EPA study that stated that acetic acid leaching fluid could selectively solubilize toxicants (specifically lead) and incorrectly classify the material as hazardous when, in fact, no mobilization (leaching) would be expected to occur in the landfill environment. Kennecott Corporation and National Mining Association, also in response to the Phase IV supplemental proposed rule, stated similar concerns. The SPLP (see text box at right) is one test that has been considered for addressing this issue. TCLP may underestimate the chelation-facilitated mobility of some waste constituents. A recent analysis of the TCLP and Cal WET (see text box at right) indicates that the low chelation activity of the acetate buffer used in the TCLP may underestimate the ability of leachate containing chelating agents to mobilize waste constituents. Cal WET uses a citrate buffer that approximates the chelation ability of many other compounds of landfill leachate and, thus, overcomes the constraints of the TCLP test. TCLP does not account for the oxidation/reduction reactions occurring in landfills. A recent study noted that the addition of iron filings to stabilize foundry sand wastes seems to mask the potential leachability of lead by interfering with the TCLP. If metallic iron (iron filings) are added to the waste, the lead concentration in the TCLP extract may be decreased by an oxidation/reduction reaction to levels below the lead TC level. If, however, the waste is placed in a landfill or surface impoundment, the iron oxidizes over time and loses its ability to further reduce the lead ions. This results in the leaching of lead to the environment. Another recent study reviewed the practice of using iron as an additive in stabilizing paint waste. The study notes that the iron reduces the lead ions in paint waste to the less soluble metallic lead, which is subsequently removed by filtration from the leachate being analyzed. This use of iron allows the lead-containing waste to pass the TCLP. The study notes, however, that repeated leaching of the same waste sample increases the leaching rate to a point where lead is sufficiently solubilized to exceed the TC regulatory level. Finally, another study showed that oxidation/reduction potential has a significant effect on leaching of metals from stabilized waste materials. This study showed that the leaching of chromium increases significantly under highly oxidizing conditions, and the leaching of arsenic, vanadium, lead, and iron increase significantly under reducing conditions. TCLP may not predict long-term mobility of organic contaminants in some treated wastes. A fairly recent Superfund Innovative Technology Evaluation (SITE) field evaluation examined the long-term performance of stabilization treatment of lead and other metals, oil and grease, and mixed volatile and semivolatile organic compounds. Portland cement and a proprietary additive were used as stabilizing agents. Durability was tested with weathering tests by wet-dry and freeze-thaw cycling and by sampling stabilized treated waste after 9 and 18 months of burial. The results showed that organic contaminants were not effectively immobilized (although the testing also showed that lead and other metals remained highly immobilized, the physical properties of the stabilized treated waste deteriorated only slightly, and the porosity decreased). Another study conducted on the long-term leaching performance of commercially stabilized waste demonstrated a highly waste-dependent effect of time on the TCLP results. In this study, TCLP extraction was performed on both the raw waste and the treated waste. The treated waste consisted of samples at 28, 90, 200, 470, and 650 days after treatment. The results showed that leachate values for some metallic wastes increased over time. TCLP may not be appropriate for some contaminated soil. The Michigan Department of Natural Resources (MDNR) believes that the TCLP is not appropriate for soils contaminated with cyanides, sulfides, and hexavalent chromium. Furthermore, MDNR reports that the SPLP (see previous text box) more accurately simulates the conditions of contaminated soil and therefore is an appropriate alternative test for soil contaminated with cyanides, sulfides, and hexavalent chromium. TCLP does not predict releases to non-groundwater pathways. As discussed in Section 3.4, the TCLP was designed to simulate the leaching of waste constituents to groundwater and not for releases to non-groundwater pathways. The TCLP does not simulate the release of volatile organic contaminants into air either directly or through entrained dust, nor does it simulate releases through surface runoff. CHAPTER 3. POTENTIAL GAPS ASSOCIATED WITH HAZARDOUS WASTE CHARACTERISTICS DEFINITIONS . . . .3-1 3.1 Types of Risks Addressed by RCRA Hazardous Waste Characteristics 3-1 3.1.1 Statutory and Regulatory Framework. . . . . .3-1 3.1.2 Risks Associated with Physical Hazards. . . .3-3 3.1.3 Acute Toxic Hazards to Humans . . . . . . . .3-4 3.1.4 Chronic Toxicity Risks to Humans. . . . . . .3-4 3.1.5 Risks to Non-Human Receptors. . . . . . . . .3-5 3.1.6 Other Risks Associated with Non-Hazardous Industrial Waste Management 3-6 3.2 Ignitability Characteristic . . . . . . . . . . . 3-8 3.2.1 Definition of Ignitability. . . . . . . . . .3-8 3.2.2 Potential Gaps Related to Definition of Ignitability 3-9 3.2.3 Potential Gaps Related to Ignitability Test Methods 3-12 3.3 Corrosivity . . . . . . . . . . . . . . . . . . . 3-12 3.3.1 Definition of Corrosivity . . . . . . . . . 3-12 3.3.2 Potential Gaps Related to Definition of Corrosivity 3-13 3.3.3 Potential Gaps Related to Corrosivity Test Methods 3-16 3.4 Reactivity. . . . . . . . . . . . . . . . . . . . 3-16 3.4.1 Definition of Reactivity. . . . . . . . . . 3-16 3.4.2 Potential Gaps Related to Definition of Reactivity 3-17 3.4.3 Potential Gaps Related to Reactivity Test Methods 3-19 3.5 Potential Gaps Associated with the Toxicity Characteristic 3-19 3.5.1 Definition of Toxicity Characteristic . . . 3-19 3.5.2 Changes in Groundwater Pathway Analysis . . 3-21 3.5.3 Potential Inhalation Pathway Risks Associated with TC Analytes 3-27 3.5.4 Potential Risks from Surface Water Exposures3-33 3.5.5 Potential Indirect Pathway Risks from TC Analytes 3-35 3.5.6 Potential for Acute Adverse Effects of Exposures to TC Analytes 3-38 3.5.7 Potential Risks to Ecological Receptors from TC Analytes 3-38 3.6 Potential Gaps Associated with TCLP . . . . . . .3-39 3.6.1 TCLP Background . . . . . . . . . . . . . . 3-39 3.6.2 Limitations of the TCLP . . . . . . . . . . 3-42 Exhibit 3-1. Risks Potentially Associated with Non-Hazardous Industrial Waste Management 3-7 Exhibit 3-2 Materials Formerly Classified by DOT as Combustible Liquids (which generally are not RCRA ignitable). . . . 3-11 Exhibit 3-3 Other Definitions of Reactivity . . . .3-18 Exhibit 3-4 TC Constituents and Regulatory Levels (mg/l) 3-20 Exhibit 3-5 Comparison of TC Regulatory Concentrations and HWIR-Waste Proposed Exit/Leach Levels 3-25 Exhibit 3-6 Summary of Inhalation Pathway Screening Methods, Input Data, and Models Used for Bounding Risk Analysis 3-29 Exhibit 3-7 Emission Fraction for Air Releases of Volatile TC Analytes 3-30 Exhibit 3-9 Major Fate and Transport Parameters for TC Analytes . 3-36 Exhibit 3-10 Ratios of TC Leachate Regulatory Levels to Ambient Water Quality Criteria for Aquatic Life . . 3-40 CHAPTER 4. POTENTIAL GAPS ASSOCIATED WITH NON-TC CHEMICALS This chapter identifies potential gaps in the hazardous waste characteristics associated with chemicals not on the toxicity characteristic list. Chemicals and chemical classes are identified as potential gaps based on their hazardous properties such as toxicity to humans and ecological receptors, their fate and transport properties such as persistence and bioconcentration potential, and their potential for occurrence in non-hazardous industrial wastes. This approach to identifying gaps is complemented by the approach discussed in Chapter 5, which identifies gaps in terms of the important environmental risks and their potential association with waste management, rather than focusing on specific chemicals. 4.1 Overview of Methodology EPA identified potential gaps in the characteristics associated with non-TC chemicals through a six-step process, as shown in Exhibit 4-1. Each of these steps is described below. Step 1: Identify and Classify Known Non-Hazardous Industrial Waste Constituents An essential task in this analysis is identifying a universe of chemicals that are either known or likely to be present in non-hazardous industrial wastes, excluding TC analytes (which are addressed in Chapter 3). In the analysis that follows, these two classes of chemicals are referred to as known non-hazardous industrial waste constituents and possible non-hazardous industrial waste constituents, respectively. As described in Section 4.2, the identification of the "known" non-hazardous constituents is relatively straightforward, although reliable data on the composition of non-hazardous industrial waste are limited. The data sources used to identify these constituents are shown in the top panels of Exhibit 4-1. They are the non-hazardous industrial waste release descriptions (discussed in Chapter 2), the Industrial Studies Data Base (ISDB), Effluent Guidelines Development Documents, and Listing Documents from recent rulemakings for dyes and pigments and solvent wastes. As discussed in Section 4.2, the distinguishing characteristic that makes a chemical a "known" non-hazardous industrial waste constituent is that it has been documented through direct chemical analysis to occur either in non-hazardous industrial waste or in environmental media contaminated by releases from non-hazardous industrial waste management units. Step 2: Identify and Screen Possible Non-Hazardous Industrial Waste Constituents In addition to the chemicals that are known to be present in non-hazardous industrial wastes, EPA identified other chemicals that have a high likelihood of being present in such wastes and could pose significant risks to human health or the environment. Unlike the known non-hazardous industrial waste constituents, however, the possible waste constituents have not been confirmed as non-hazardous industrial waste constituents through direct chemical analysis in any of the data sources used by the Agency. To identify non-hazardous industrial waste constituents that could pose risks to human health or ecological receptors, the Agency reviewed 36 lists of chemicals created for regulatory and advisory purposes by EPA, other federal agencies, states, other countries, and advisory and scientific bodies. These lists were originally created based on criteria such as toxicity, fate and transport characteristics, production volume, widespread use, and detection in environmental media. Rather than include all the chemicals on these lists as possible non-hazardous industrial waste constituents, EPA narrowed the list of chemicals to those most likely to pose significant risks to human health and the environment. The screening was performed in two steps, as shown in the upper right-hand panels of Exhibit 4-1. First, chemicals were screened with regard to individual toxicity and fate and transport properties. Then, the resulting high-hazard chemicals were screened against 1994 national Toxic Release Inventory (TRI) release data, serving as a proxy for potential occurrence in waste. Section 4.3 describes the process of compiling and screening possible non-hazardous industrial waste constituents. Step 3: Apply Hazard-Based Screening Criteria In this step, which is described in detail in Section 4.4, EPA compared the lists of known and possible non-hazardous industrial waste constituents and screened them against single and multiple hazard-based screening criteria. In Step 2, individual chemicals that are possible non-hazardous industrial waste constituents were screened on the basis of single indicators of hazard (e.g., a low reference dose or a high bioconcentration factor). This step refines this analysis by examining both the known and possible non-hazardous industrial waste constituents against single and multiple indicators of toxicity, fate, transport, and occurrence in waste, and by reviewing the implications of this screening for classes of chemicals. Step 4: Review Relevant Multipathway Risk Modeling Results Section 4.5 reviews the results of the multipathway risk modeling conducted as part of the proposed HWIR-Waste (Hazardous Waste Identification Rule for Process Wastes) determination of exit levels, where available for chemicals on the combined list of known or possible non-hazardous industrial waste constituents. The proposed exit levels and risk-driving pathways provide information on the relative risks posed by the various constituents and on the most important exposure pathways. Step 5: Identify Potential Acute Hazards In the prior steps, the evaluation of potential hazards associated with the possible and known non-hazardous industrial waste constituents has focused on chronic toxic effects. In Section 4.6, the possible and known constituents are compared to acutely hazardous chemical lists developed by EPA and other regulatory agencies. This analysis thus addresses risks from acute exposures and from physical hazards associated with reactivity, flammability, and corrosivity. Step 6: Summarize Findings Chapter 4 concludes by identifying non-TC chemicals and groups of chemicals that constitute potential gaps in the hazardous waste characteristics. Section 4.7 presents a table identifying these potential gaps, the rationale for their identification, and the major issues and data gaps remaining to be resolved to judge the severity of these potential gaps. 4.2 Identify and Classify Known Constituents of Non-Hazardous Industrial Wastes Chemicals present in non-hazardous wastes that have been released from non-hazardous industrial waste management units into the environment may constitute potential gaps in the hazardous waste characteristics. This section reviews the available evidence concerning such chemicals. Reliable data concerning the chemical composition of non-hazardous industrial wastes, however, are quite limited for two major reasons. First, such wastes may be generated by virtually any industrial facility or operation and are inherently heterogeneous. Second, state requirements to analyze non-hazardous industrial wastes and to report analytical results are quite limited. In the course of this Scoping Study, the Agency identified four sources of information regarding the composition of non-hazardous industrial wastes: ! The descriptions of environmental releases from non-hazardous industrial waste management facilities, compiled as part of this Scoping Study, which were summarized in Chapter 2; The Industrial Studies Data Base (ISDB), which includes information on point of generation constituent concentrations on various industries; ! Chemicals identified as being present in liquid non-hazardous wastes by EPA Effluent Guideline Development Documents, as summarized in the Capacity Analysis for the Phase III Land Disposal Restrictions (LDR) Rule; and ! Chemicals identified as being present in non-hazardous industrial waste that were not listed as hazardous wastes in background documents for recent Agency listing/no-listing proposals for pigments and dyes industries and for solvents. The first source provides information on chemicals detected in environmental media (primarily groundwater) that were released from non-hazardous industrial waste management facilities, while the other three sources provide information on the composition of non-hazardous industrial wastes. Although not reflected in this Study, in future investigations the Agency will consider examining the constituents present in remediation waste from non-hazardous industrial waste management units. The descriptions of environmental releases in Chapter 2 identify the constituents found in environmental media near non-hazardous industrial waste management units, their maximum detected concentrations, the types of units from which the releases occurred, and the industries responsible for the releases. The release descriptions provide direct evidence of potential environmental exposure to non-hazardous industrial waste constituents and damage to human health and the environment. They, however, do not encompass all instances where non-hazardous industrial waste management has resulted in releases to the environment or other potential risks. As noted in Chapter 2, the release descriptions come from only a small proportion of the states. However, they do represent a large proportion of the readily identifiable releases from facilities regulated by state non-hazardous industrial waste programs. In addition, some types of occurrences (e.g., fires and explosions) and units (e.g., waste piles) are generally not regulated by these state programs, and would not show up in the records EPA examined. The quantitative data from these descriptions generally were limited to groundwater monitoring results. Few releases to other media were identified. In addition, the chemicals identified tend to be those whose monitoring is required under existing regulatory programs. The potential for identifying chemicals not already recognized as hazardous is therefore limited. Finally, the data sources evaluated did not provide useful information on various types of uses constituting disposal, such as cement additives, soil amendments, or aggregate. The ISDB was the second source of data used to identify known waste constituents. EPA has maintained this data set since 1982. It contains information on point-of-generation constituent concentrations for 16 industries. The sources of information include RCRA Section 3007 questionnaires, plant visit reports, sampling and analysis reports, and engineering analysis. Its major limitations include data that are sometimes more than 15 years old and the coverage of only selected industries. The third data source was information gathered by EPA's Office of Water in preparing Effluent Guidelines Development Documents. These data are summarized in OSW's Capacity Analysis Background Document for the Phase III LDR. The data describe the composition of non-hazardous industrial wastewaters generated by major industry groups. These wastewater data are of varying age, and therefore their continued representativeness is unclear. Also, the number of analytes in the database is quite limited. As seen below, a very high proportion of the waste constituents identified in this source also are identified in one or both of the two data sources described above. Thus, the effluent guidelines data serve mainly to confirm data from the other sources. The Agency also reviewed two recent proposed listing decisions for hazardous wastes, those for solvent wastes and for wastes from the dyes and pigments industries. Several additional chemicals were identified as being constituents of unlisted (non-hazardous) solvent waste streams that were not found in any of the other data sources: 2-methoxyethanol, 2-ethoxyethanol acetate, cyclohexanol, isophorone, and diethylamine. No non-hazardous industrial waste constituents from the dyes and pigments industry were identified, because all of the data concerning the compositions and generation rates of these wastes were held as confidential by the industries that submitted data. Excluding TC analytes, which are addressed in Chapter 3, a total of 146 chemicals were identified in the release descriptions, 183 in the ISDB, and 19 in the effluent guidelines data. An additional five unique constituents were found in the listings background document. Overall, a total of 250 unique chemicals were identified. The chemicals and waste constituents identified in the three data sources are sorted into major chemical classes and shown in Exhibit 4-2. These constituents span a wide range of chemical classes. Even with a number of possibly redundant entries, the most common category of chemicals was metals and inorganics, with 48 chemicals. Other prominent families of chemicals included volatile chlorinated organics (38), other semivolatile organics (46), other volatile organics (45), and pesticides and related compounds (29). Included among the chlorinated organics are several trihalomethanes and two chlorofluorocarbons. The "other semivolatile" category contains a wide range of compounds, many of which are found only in the ISDB data. The pesticides category contains mostly chlorinated organic pesticides and intermediates, but also contains some nonchlorinated compounds. Less prominent categories of chemicals include the PAHs (18 compounds), volatile hydrocarbons (12), phenolic compounds (8), and phthalate esters (6). The PAHs range from low-molecular weight, noncarcinogenic compounds (such as naphthalene) to the higher molecular weight carcinogens and mutagens (such as benzo(a)pyrene). All but one of the volatile hydrocarbons (styrene) are commonly found as constituents in kerosene, gasoline, and related fuels. Styrene is a monomer used in plastics production. The phenolic compounds include creosote components (cresols) and two nitrophenols. Most of the phthalate esters are found in all the first three data sources, including the suspect carcinogen bis-(2-ethylhexyl)-phthalate. Polychlorinated biphenyls (PCBs) and chlorinated dioxins (represented by 2,3,7,8-TCDD) were found in the ISDB. The number of compounds in the various categories does not necessarily reflect the relative potential importance of the chemicals or categories. As noted above, some chemicals occur only in one database, while others occur in two, three, or all four. In addition, some chemicals occur in more than one release description, that is, at more than one facility, or are identified as waste constituents from more than one industry group. Except for the chemicals in the release descriptions, there is no indication of the relative concentrations of the chemicals in wastes. Given the wide range of chemical classes represented in the lists, and the relatively small total number of non-TC chemicals in the four datasets (250), the Agency found no convincing reason to eliminate any candidate chemicals from inclusion in the gaps analysis. Given that toxicological and fate and transport data are available for most of these chemicals, all the chemicals were carried forward for further analysis. 4.3 Identify Possible Non-Hazardous Industrial Waste Constituents of Potential Concern This section describes the approach to identifying additional chemicals that might constitute potential gaps in the hazardous waste characteristics. Unlike the previous analysis, which began with four relatively narrow and specific data sources, this analysis begins with a wide range of data sources, in order to avoid excluding chemicals of potential concern. Subsequently, a substantial proportion of the large universe of chemicals are screened out on the basis of toxicity, fate and transport characteristics, and potential for occurrence in waste. A large portion also could not be evaluated because of a lack of data. The result is a focused list of possible non-hazardous industrial waste constituents that could pose significant risks to human health or the environment. The list of possible non-hazardous industrial waste constituents supplements the list of known non-hazardous industrial waste constituents developed in the previous section. 4.3.1 Approach to Identifying Potentially Hazardous Chemicals Excluding TC analytes, EPA identified over 2,300 distinct chemicals from 36 regulatory and advisory lists originally created by EPA, other federal agencies, state and national regulatory agencies, and special environmental task forces and advisory bodies. Exhibit 4-3 identifies these lists. The RCRA regulatory lists included are the 40 CFR 261 Appendix VII and VIII lists of hazardous waste constituents, the proposed HWIR-Waste Chemicals, and the HWIR-Media "Bright Line" chemicals. Other major federal regulatory lists include the Clean Water Act Section 307 Toxic Pollutants and Section 311(b)(2) Hazardous Substances, the CERCLA list of hazardous substances with reportable quantities, the Emergency Planning and Community Right-to-Know (EPCRA) Toxic Chemicals and Extremely Hazardous Substances lists, the Clean Air Act Amendments Section 112(b) Hazardous Air Pollutants and Section 112(r) Regulated Toxic Substances, and chemicals for which OSHA has published Permissible Exposure Limits (PELs). The U.S. Department of Transportation (DOT)Hazardous Materials Transportation Act (HMTA) Hazardous Materials Registry (HMR) also was used to identify potential gap chemicals, but could not be directly included in the database in time because of format differences in the available machine-readable forms of the list. Some of the advisory lists that were included are the 1992 EPA Hazardous Substance Task Force's Level 1 and Level 2 hazardous chemicals that were identified as not being controlled under RCRA or DOT regulations, the Focus Chemicals for the Great Waters Study, chemicals identified by Environment Canada under the ARET Toxics Scoring Protocols, chemicals identified by the University of Tennessee Chemical Ranking System, and the Michigan Critical Materials Register. Some lists address specific types of hazards, such as potential endocrine disruptors, acutely toxic chemicals, highly flammable chemicals, and highly reactive chemicals. Brief descriptions of the lists and the selection criteria that were applied to derive them are provided in "Background Document: Identification of Chemicals from Regulatory and Advisory Lists Representing Potential Gaps in the Hazardous Waste Characteristics." Naturally, there is a high degree of overlap among the chemical lists. Some lists are subsets of, combinations of, or otherwise derived from other lists. Nonetheless, the chemicals identified represent a very broad spectrum of potential hazards. High-volume and highly toxic chemicals appear on many lists, as do acutely toxic, flammable, and reactive chemicals. Several lists specifically seek to include carcinogens, mutagens, and teratogens. Some lists are derived based on considerations of ecotoxicity, persistence, and bioaccumulation potential, or based on specific environmental media or geographical concerns. The overall goal in the Scoping Study was to identify the broadest possible set of chemicals of potential concern, and then to screen them down to the chemicals with the highest potential to pose risks to human health or the environment. 4.3.2 Screening Approach EPA performed the hazard-based screening of potentially hazardous constituents in two steps. First, the entire list of chemicals was screened against criteria related to toxicity to humans and aquatic organisms and separately against various fate and transport criteria. Chemicals for which data were not available for at least one of these criteria were not included in further analysis. In the second step, EPA took all of the chemicals identified as either highly toxic, mobile, persistent, or bioaccumulative and first screened them against the proxy for occurrence in waste, namely the TRI release data. Any chemical passing this screen has a high potential for occurrence in waste and was identified as a possible non-hazardous industrial waste constituent. Chemicals were also retained in the analysis if they were not on the TRI list. Only the chemicals confirmed as having low releases through the TRI data were eliminated from being possible constituents. The criteria considered for use in screening (both the possible constituents described in this section and the combined lists discussed in Section 4.4) are summarized in Exhibit 4-4. These criteria were derived using professional judgment to provide a reasonable level of discrimination between chemicals with relatively high-hazard potential and those with lower potential. For most toxicity parameters, which were available only for a relatively small number of more toxic chemicals, the cutoff values were set at the 50th percentile of the entire range of values. For many fate and transport parameters, the criteria were set at or around the 75th percentile (or 25th percentile, if a low value implied high hazard potential) of the entire range of the parameter values for all of the chemicals for which the parameter was available. In some cases, the screening criteria were set at levels generally recognized as indicative of hazard potential. In the course of the Scoping Study, many different criteria for and approaches to the screening process were evaluated; the background document to this Study provides further detail. The criteria and approach described in this section is a relatively simple one that evolved from those previous efforts. One of the major lessons learned in that work was that screening is inherently imprecise, and no single screen will catch or exclude all the chemicals desired. Another lesson learned is that screening large lists against complex criteria can quickly become very complicated, and the return on the complexity, in terms of useful information, can be quite low. Therefore, EPA has focused on a relatively small number of criteria that are important in determining risk potential and has critically interpreted the results of the screening. In the case of carcinogens, two sets of criteria were used. The first set indicates whether a cancer slope factor (CSF) had been promulgated for the chemical. The second indicates whether an inhalation unit risk (UR) had been developed. These criteria identified the bulk of human carcinogens. For noncarcinogenic effects, two sets of criteria again were used. The first indicates whether an ingestion reference dose (RfD) had been developed at a sufficiently toxic level for the purposes of this analysis (i.e., below the 50th percentile). The second indicates whether an inhalation Reference Concentration (RfC) had been developed below the 50th percentile. For aquatic effects, the 50th percentile of the Chronic Ambient Water Quality Criteria (AWQC) was used. EPA used several criteria to screen fate and transport properties. The screening criteria for the fish bioconcentration factor (BCF) and bioaccumulation factor (BAF) were both set at 1,000 l/kg, the beef biotransfer factor was set at 7.8x10-3 day/kg, and the octanol-water partition coefficient (Kow) was set at 105. These four values indicate the potential for the chemicals to be taken up and/or accumulated by organisms. The vapor pressure criterion, used as a proxy for volatilization release, was set at 1 mm Hg. A Henry's Law constant (kH) value of 10-5 atm-m3/mole was also used to identify chemicals with high volatilization potential. The criterion used to identify persistent chemicals in soil or water (degradation rate constant less than 0.5/year) was selected based on an analysis of the EPACMTP findings for organic pollutant transport in groundwater, which indicated that, at rate constants above this value, the calculated DAF values begin to differ substantially from those for non-degrading pollutants with similar properties. As noted in Section 3.5, the screening-level risk analysis also was used to identify screening criteria and their importance. For example, Henry's Law constants were found not to be a good indicator of the potential for long-term volatilization releases, so that the parameter is not used as a primary screening factor (although it is examined briefly in the next section). Instead, vapor pressure is used to screen chemicals for volatilization release. Even this screen must be interpreted cautiously, however, since chemicals with low vapor pressures can still volatilize from treatment units if no other processes are occurring to limit the releases. The primary data source that is used as a proxy for occurrence of hazardous chemicals in non-hazardous industrial wastes is the release data, reported under the Emergency Planning and Community Right-to-Know Act (EPCRA) Toxic Release Inventory (TRI) requirements. For purposes of the screening conducted for this study, EPA considered those chemicals with releases to air, land, water, and underground injection exceeding one million pounds in 1994. Under EPCRA Section 313, facilities with more than 10 full-time employees that are classified in SIC codes 20 through 39 (i.e., manufacturing) must submit reports if they manufacture or process more than 25,000 pounds of a TRI chemical or otherwise use more than 10,000 pounds of a TRI chemical in a given calendar year. There were a total of 73 unique chemicals and 10 classes of chemicals in this category, out of the 345 individual chemicals for which reports are required. These chemicals account for greater than 99.8 percent of the total TRI releases of all chemicals. As discussed in Section 4.4.2, the combined list of known and possible non-hazardous industrial waste constituents were also screened against non-CBI 1994 production data from the TSCA Inventory. A major limitation of this screening approach is that quantitative toxicity and fate and transport parameter values were available for only a fraction of the over 2,300 non-TC chemicals identified. Human toxicity parameters were available for just over 430 chemicals, ambient water quality data for 105 chemicals, and complete fate and transport data for 194 chemicals. For this reason, the screening approaches were supplemented by searching lists that identify chemicals presenting specific types of hazards, even if no quantitative parameter value was available, and by applying professional judgment to identify where potential risk findings for individual chemicals may be generalized to broader classes of chemicals. The results of this screening are described in a background report (see footnote 6). 4.3.3 Toxicity, Fate, and Transport Screening for Possible Non-Hazardous Industrial Waste Constituents Exhibit 4-5 summarizes the results of the screening for possible non-hazardous industrial waste chemicals against the toxicity criteria. The first two columns indicate the chemicals that are suspect or known human carcinogens having ingestion CSFs or inhalation URs. The last three columns identify the chemicals with oral RfDs, inhalation RfCs, and AWQCs below the 50th percentile of these parameter values (a low value indicates high toxicity) for all chemicals for which these values have been developed. Note that this table does not include TC analytes or chemicals previously identified as known non-hazardous industrial waste constituents. As noted previously, the number of chemicals identified on all 37 lists of chemicals is much greater than the numbers of chemicals for which toxicity parameters have been developed. Furthermore, the list of chemicals, which includes practically all of the known chemicals from Section 4.2 and all of the TC analytes, includes almost all chemicals for which these toxicity values have been derived. Thus, the toxicological screen has the potential to screen out most of the possible non-hazardous industrial waste constituents simply because most of the constituents do not have toxicity values, and therefore the effectiveness of the individual toxicity screening criterion is substantially limited for a large proportion of the chemicals identified on the 37 lists. Nevertheless, because all chemicals with cancer toxicity values are considered high hazard for this portion of the analysis, no chemicals would be screened out on the basis of carcinogenicity. The toxicity screening reduced the number of chemicals dramatically from the original universe of over 2300. As noted above, this reduction is primarily a function of the relatively small number of chemicals (about 400) for which human or ecotoxicity data are available. The screened list contains about one-third (25/74) of the chemicals for which CSFs were available, and about one-quarter (13/52) of those for which inhalation unit risks are available. The chemicals with low (<50th percentile) RfDs comprise by far the largest (107) set of all the chemicals identified by the toxicity screening, representing about one-third of the total number of chemicals for which RfDs have been derived. A large proportion of these chemicals are pesticides. Relatively few chemicals were identified having low inhalation RfCs and AWQCs for aquatic life. Exhibit 4-6 summarizes the results of the screening of chemicals with regard to fate and transport properties. The first two columns address the potential to volatilize for soil and water, as indicated by the vapor pressure and Henry's Law constant. Since these parameters are directly related, the chemicals in these two columns overlap substantially. The next column lists chemicals with soil or water column degradation constants less than 0.5/year. Since the values for these two media are close for most of the chemicals, separate columns are not provided for each medium. The final three columns identify the chemicals with relatively high aquatic BCFs, beef biotransfer factors, or Kows. Since all three of these values are related to partitioning between lipid and water phases, the chemicals in these three columns also overlap substantially. As was the case for the toxicity screens, consistently-derived fate and transport parameters are not available to screen the majority of the chemicals. Thus, the menu of chemicals that are identified by the screening criteria related to each individual parameter again is determined primarily by the availability of data. In the case of the fate and transport screening, fewer chemicals are identified as being potentially hazardous. In addition, the fate and transport screening identifies a smaller proportion of the chemicals for which data are available. In all cases, the chemicals exceeding the screening criteria represent less than 10 percent of the chemicals for which data are available. 4.3.4 Release Volume Screening of Possible Non-Hazardous Industrial Waste Constituents Chemicals not screened out by the toxicity or fate and transport criteria were screened against the 1994 TRI data (used as a proxy for occurrence in wastes). The results of this final screening are presented in Exhibit 4-7. Of the 151 unique chemicals or classes of chemicals that were identified in the toxicity or fate and transport screening, TRI release data were available for 24 of them. Five of these chemicals (Freon 113, 1,3-butadiene, chlorine dioxide, chloroprene, and propylene dioxide) had TRI releases above one million pounds in 1994. Nineteen of the chemicals had TRI releases less than a million pounds. This latter group of chemicals were eliminated from further analysis. As noted previously, the remaining 132 chemicals for which no TRI data were available were retained in the analysis. 4.3.5 Summary of Possible Non-Hazardous Industrial Waste Constituents Exhibit 4-8 summarizes the results of the TRI screening process. It places the possible non-hazardous waste constituents into the same chemical categories as were used to characterize the known non-hazardous industrial waste constituents in Exhibit 4-2. The largest number of possible waste constituents (74) are pesticides and related compounds. As discussed in Section 4.3.2, these chemicals are identified as being potentially hazardous primarily by virtue of low RfDs, although there are also some potent ecotoxins, as well as persistent and bioaccumulative chemicals, among this group. The next most numerous category among the possible constituents are the other semivolatile organic chemicals. This diverse group includes chemicals recognized both for their toxicity and their fate and transport properties. Twelve metals/inorganic elements or groups are identified including five different thallium salts. Similarly, the other volatile organics group includes 5 nitrosamines among a total of 13 compounds. Also included in this group are two very toxic organometallic compounds, methyl mercury and tetraethyllead. Among the seven chlorinated organics are two of the five chemicals with TRI releases greater than one million pounds (Freon 113 and chloroprene). No other chemical category is represented by more than five chemicals. 4.4 Combine and Screen Known and Possible Non-Hazardous Industrial Waste Constituents In this section, the known (from Section 4.2) and possible (from Section 4.3) non-hazardous industrial waste constituents are combined and screened against toxicity, fate, and transport criteria. Unlike the prior section, screening is oriented more toward groups of chemicals rather than toward individual chemicals, and toward comparing the properties of known versus possible non-hazardous industrial waste constituents. There is, in addition, another screening step related to potential for occurrence in wastes, namely, comparison to 1994 non-confidential TSCA production volume data. 4.4.1 Combine the Lists The lists of known and possible non-hazardous industrial waste constituents are shown in Exhibits 4-2 and 4-8. Exhibit 4-9 summarizes the screening of the known non-hazardous industrial waste constituents in the same way that Exhibit 4-7 provides these data for the possible constituents. As seen in these exhibits, the distribution of chemicals within chemical classes is somewhat different between the known and possible non-hazardous industrial waste constituents. These differences, however, are exaggerated by the removal of the known constituents from consideration as possible constituents. (Logically, a chemical cannot be both a "known" and "possible" waste constituent.) The known non-hazardous industrial waste constituents are distinguished by a relatively high proportion of metals and inorganics, chlorinated volatile organics, other volatile organics, and polycyclic aromatic hydrocarbons, compared to the possible non-hazardous waste constituents. In contrast, pesticides and related compounds constitute a much higher proportion of the possible non-hazardous industrial waste constituents than the known constituents. The pattern of differences in chemical category can be partially explained by the differences in the data sources. The relatively high prominence of volatile organics among the possible constituents probably reflects the difficulties in controlling fugitive releases of these high-volume chemicals during storage and processing. Such chemicals are somewhat less likely to turn up in groundwater samples (in the release descriptions or in aqueous effluents) because of their high volatility. The prominence of the less volatile organics in the known non-hazardous industrial waste constituents again reflects the greater stability of these chemicals in solid and liquid wastes. Exhibit 4-9 also shows that the known waste constituents include a much higher number of chemicals with TRI release values greater than one million pounds (45) than is found among the possible constituents (5). This is primarily due to the fact that the known waste constituents were identified first. Many of the high TRI release chemicals also would have been identified as possible non-hazardous industrial waste constituents if they had not been identified as known constituents. The implications of these findings for the potential severity of gaps in the hazardous characteristics are discussed in more detail in Chapter 10. In the analysis that follows, the known and possible non-hazardous industrial waste constituent lists are combined, and screened against single and multiple parameters related to toxicity, fate and transport, and release potential. 4.4.2 Screen Combined List Against Single Criteria Quantitative Human Toxicity Indicators. Exhibit 4-10 summarizes the toxicological properties of the combined known and possible non-hazardous industrial waste constituents. The chemicals are screened using the same criteria as described for the possible constituents alone in Section 4.3, with the exception that additional criteria related to carcinogenic potency are added (oral CSF and inhalation UR > 50th percentile). The list of suspect carcinogens (i.e., the first and third columns in Exhibit 4-10) contains a large proportion of all chemicals for which EPA has developed CSFs and URs. The proportion of the chemicals with high CSFs or URs (i.e., the second and fourth columns) is likewise very near to one-half of the total suspect carcinogens. This finding indicates that, as expected, the large universe of chemicals initially screened contains almost all of the chemicals that EPA has evaluated as potential human carcinogens. Many classes of chemicals (inorganics, volatile chlorinated organics, pesticides, other volatile chemicals) are represented among the suspect carcinogens. Ecotoxicity. As shown in the last column of Exhibit 4-10, 18 of the combined known and possible constituents have low AWQCs (below 50th percentile), indicating the potential for adverse effects on aquatic organisms. Many of these chemicals are pesticides, and most of the pesticides are persistent chlorinated pesticides. Although most of these chemicals are no longer produced, their presence among the known non-hazardous industrial waste constituents may give rise for some concern. Also included in this group are selenium, silver, and hydrogen sulfide. Potential Endocrine Disruptors. Because of the rapidly-evolving state of knowledge regarding chemicals that may act as endocrine disruptors, estrogen inhibitors, or have other hormone-like effects, it is difficult to estimate precisely how many of the combined known and possible non-hazardous industrial waste constituents fall into this category. Based on the rather broad list of potential endocrine disruptors, 23 of the combined constituents are implicated as being potential endocrine disruptors (Exhibit 4-11). (Nine of the TC analytes are also potential endocrine disruptors.) Because of the lack of knowledge concerning dose-response relationships for exposures to single and multiple endocrine disruptors, it is difficult to predict if these chemicals would present risk to humans and non-human receptors. Nevertheless, the fact that so many of these chemicals are present among the constituents may cause concern. Potential for Frequent Occurrence in Wastes. The combined list of known and possible non-hazardous industrial waste constituents were also searched to identify those chemicals with high potential for occurrence in wastes, as indicated by TRI releases and/or non-confidential TSCA Inventory production data. The results of this analysis are summarized in Exhibit 4-12. Constituents are included in the table only if either TRI release data or non-CBI TSCA inventory data are available for them. Volatility and Persistence. As discussed in Section 3.5, volatility and persistence appear to be key indicators of potential risks for the TC analytes. In the first four columns of Exhibit 4-13, the known and possible non-hazardous industrial waste constituents are screened against these properties. Vapor pressure of 1.3x10-3 atmosphere (which is approximately equivalent to 1 mm Hg) is used to identify volatile chemicals. This measure approximates the potential to volatilize; many chemicals with lower vapor pressure could volatilize readily under certain waste management conditions. Even so, 70 known or possible non-hazardous industrial waste constituents fall into this category. This finding suggests that, as for the volatile TC analytes, volatilization releases and inhalation exposures (and possibly indirect exposures) may be a concern for some of these chemicals. Two chemicals, both chlorinated organics, are identified as having long half-lives (greater than 0.15 year) in air. This finding does not mean that all of the other constituents are too short-lived to be of concern through air exposures. Half-lives on the order of a few hours or days also may be of concern in terms of direct inhalation exposures. This criterion is more indicative of the potential for long-range (e.g., regional or global-scale) transport of these chemicals in the vapor phase. Also, as noted in Section 3.5, the air half-lives of many of the inorganic waste constituents (especially the metals) bound to particulates would also be limited only by how long the particles remained suspended in the atmosphere. The third column of Exhibit 4-13 identifies the non-hazardous industrial waste constituents that are relatively persistent either in soils or in the water column. The metals all fall into this category, along with the PAHs, many chlorinated pesticides, and 2,3,7,8-TCDD. The only volatile organic chemical in this category is 1,2-dichloropropane. Appearance in this category arouses concern for potential inhalation and indirect pathway exposure risks, as discussed in Section 3.5. A high Kow, as indicated in the fourth column, indicates a high potential to bind to soil organic matter. It is highly correlated with the tendency to bioaccumulate. Thirty-one of the known and possible waste constituents including many persistent pesticides and PAHs, are in this category. Bioaccumulation Potential. The last three columns of Exhibit 4-13 indicate the potential for bioaccumulation by the known and possible non-hazardous industrial waste constituents in aquatic and terrestrial food chains. The constituents with aquatic BCFs or BAFs greater than 1,000 are limited to the chlorinated pesticides, several phthalate esters, and diethylstilbestial (DES). This finding does not imply that no other constituents present significant risks through indirect pathways; nevertheless, the identified chemicals are all clearly recognized as being problematic from the point of view of bioconcentration. If these chemicals were released in significant amounts from non-hazardous waste industrial management activities, they could present substantial risks through food-chain exposures. The last column of the table lists chemicals that are taken up from feed by beef cattle with above-average (greater than 75th percentile) efficiency. This list includes most chemicals that also are of potential concern for aquatic ecosystems. Also, several additional classes of chemicals are identified, including the metals and PAHs. Although the beef biotransfer factor is only one of many parameters determining the potential for risks to humans from beef consumption, it is a reasonable indicator of potential concern for this pathway and is a useful indicator of exposure potential in other terrestrial food chains. LNAPL and DNAPL Formation. The potential to form nonaqueous phase liquids (NAPLs) is of great concern from the point of view of waste management risks. Historically, NAPLs have been serious problems in the remediation of hazardous waste, because of their high potential risks and high remediation costs. Any chemical that is relatively insoluble in water and is a liquid at ambient temperature can be the principal component of a NAPL. If the chemical or chemical mixture is denser than water, then a dense nonaqueous phase liquid (DNAPL) is formed. If the liquid is less dense than water, a light nonaqueous phase liquid (LNAPL) may be formed. DNAPLs are of particular concern because, when they escape to groundwater, they will sink through the unsaturated zone or aquifer until they encounter bedrock or another barrier. They can remain at the bottom of the aquifer (for example, in bedrock fractures) where they are hard, or in some cases nearly impossible, to remediate. Most DNAPLs undergo only limited degradation in the subsurface, and persist for long periods while slowly releasing soluble organic constituents to groundwater. Even with a moderate DNAPL release, dissolution may continue for hundreds of years or longer under natural conditions before all the DNAPLs are dissipated and concentrations of soluble organics in groundwater return to background levels. When released into surface water, DNAPLs tend to sink to the bottom and contaminate sediments. LNAPLs, in contrast, will tend to float on the surface of an aquifer, where they are easier to remedy; yet, they also can contaminate large volumes of groundwater through slow dissolution. Both LNAPLs and DNAPLs also can facilitate the transport of toxic waste constituents by solubilizing chemicals that would otherwise be immobile in waste or soil matrices. It is difficult to predict the circumstances under which LNAPL and DNAPL formation will occur and pose a risk to human health or the environment. Whether significant amounts of NAPLs will form depends on the composition of the wastes and the management practices employed. Reports of nonaqueous phase liquids were not found among the release descriptions for non-hazardous industrial waste management summarized in Chapter 2, possibly due to limitations in monitoring requirements. EPA has recently conducted a study of the potential for DNAPL formation at hazardous waste (NPL) sites, and identified several industries where NAPL formation is particularly likely to occur. These industries include wood treating sites, general manufacturing, organic chemical production, and "industrial waste landfills". A wide variety of chemicals have been found in NAPLs, and it appears that if a chemical is to be the major constituent of a NAPL, the most important requirements are relative insolubility in water and liquidity at ambient temperatures. Exhibit 4-14 identifies a number of the known and possible non-hazardous industrial waste constituents with the requisite physical properties. Since there is no clear dividing line between chemicals likely and not likely to form NAPLs, this list was developed using a combination of professional judgment and information about the physical properties of the waste constituents. All of the chemicals listed are organics, have relatively low water solubilities, and are liquid at room temperature (melting points greater than 71C, boiling point greater than 301C). Those indicated as being potential DNAPL formers have bulk liquid densities greater than 1 gm/cc, while those with densities less than water are indicated as potential LNAPL formers. The distinction is not clear-cut however, as a mixture of light and heavy constituents at different relative concentrations might have widely varying densities. Exhibit 4-14 identifies more potential DNAPL formers than LNAPL formers found among the known and possible waste constituents. Based on density considerations, the LNAPL formers tend to be primarily the non-halogenated hydrocarbons, including "BTEX" and compounds with similar properties, whereas the DNAPL formers tend to be primarily chlorinated and brominated chemicals. Not included in the NAPL list are pesticides that also fulfill the physical criteria, but which are no longer produced (see Chapter 9) and thus are less likely to be present in significant amounts in pure form in non-hazardous industrial wastes. These findings suggest that, on physical bases alone, many of the known and possible non-hazardous industrial waste constituents could form LNAPLs or DNAPLs. As noted above, however, when this actually occurs depends to a large degree on the specific characteristics of the wastes and waste management practices. EPA's analysis of DNAPL formation at NPL sites found that the contaminants most directly associated with DNAPL presence include creosote compounds, coal tar compounds, polychlorinated biphenyls (PCBs), chlorinated solvents, and mixed solvents. 4.4.3 Screen Combined List Against Multiple Parameters This section discusses the results of one last round of screening conducted on the entire combined list of known and possible non-hazardous industrial waste constituents. This analysis combines toxicity, persistence, volatility, and bioaccumulation screens in various combinations in order to identify the chemicals most likely to pose risks by various exposure pathways. Only constituents in the intersections of the screens remain (e.g., only constituents that are persistent and highly toxic). For human toxicity, the criteria have been applied in the following order: ! Persistent and Highly Toxic to Humans. This combination is intended to identify highly toxic chemicals that could pose risks through any pathways involving long-term release and transport of contaminants, such as groundwater and indirect pathways involving air, surface water, or groundwater releases. ! Persistent, Highly Toxic to Humans and Bioaccumulative. This screen narrows the above waste constituents to those with potential for adverse effects through indirect food chain exposure. ! Persistent, Highly Toxic to Humans, Bioaccumulative, and Volatile. This combination further narrows the above chemicals to those with potential to cause indirect pathway risks through air releases. A fourth screen applied persistent, ecotoxic, and bioaccumulative criteria to the combined list of constituents. This combination of screening criteria is intended to identify chemicals for which potential harm to ecological receptors is a potential concern. The individual criteria used in combination are described in Section 4.3. The persistence screen consisted of a determination of whether the chemicals had soil or water column degradation rate constants of less than 0.5/year. "Highly toxic" indicates any chemical having a CSF or Unit Risk above the 50th percentile of all chemicals, or a chronic RfD below the 50th percentile. Volatility was screened against Henry's Law constant of 10-5 atm-m3/mole, and bioaccumulation potential determined by an aquatic BCF or BAF value of greater than 1,000 L/Kg. The results of the combined screening of known and possible non-hazardous industrial waste constituents are summarized in Exhibit 4-15. To a substantial degree, these results parallel the screening-level modeling results for the TC analytes discussed in Section 3.5. Four of the nine persistent and highly toxic chemicals are chlorinated pesticides or degradation products, along with three metals (antimony, beryllium, and molybdenum), benzo(a)pyrene, and 2,3,7,8-TCDD. The appearance of benzo(a)pyrene suggests that other high molecular weight PAHs (some of which are also carcinogens) might also pass this screen if CSF values were available for these compounds. In addition, several other chlorinated pesticides have properties that just miss the toxicity or persistence cutoff values. When bioaccumulation potential is added to the screening conditions (second column of Exhibit 4-15), no chemicals drop out. This finding shows the high correlation between persistence and bioaccumulative potential: if a chemical was not persistent, it would lack the opportunity to accumulate in environmental media or tissue. When the criterion of volatility is added to the preceding screens, three chemicals, all persistent pesticides remain. This result again parallels the results seen for the TC analytes in Section 3.5. If vapor pressure cutoff (1 mm Hg), rather than Henry's Law constant (10-5 atm.-M3/mole) is used to characterize the potential to volatilize, none of the chemicals qualify in this category. The last column of Exhibit 4-15 identifies persistent, bioaccumulative, and ecotoxic chemicals. As might be expected from the previous screening results, these chemicals include chlorinated pesticides and 2,3,7,8-TCDD. Because the AWQC screen is based only on harmful concentrations, it does not include any screening for the concentrations normally encountered in the environment. Thus, if a much less toxic chemical (for example zinc or copper) were released into the environment in much larger amounts than the pesticides, the exposure concentrations might be much greater and adverse effects on ecological receptors might occur. 4.5 Driving Risk Pathways for the Known and Possible Non-Hazardous Industrial Waste Constituents EPA has previously evaluated the potential risks associated with the management of many known and possible non-hazardous industrial waste constituents in the context of deriving proposed risk-based exit levels for the proposed HWIR-Waste rulemaking. As discussed in Section 3.5, these proposed exit levels were derived by back-calculating concentrations in wastewaters and nonwastewaters corresponding to acceptable risk levels. The magnitude of the modeled exit levels is inversely proportional to the magnitude of risk posed by the chemical when placed in the specified management units. Proposed exit levels are calculated for groundwater exposures and other pathways. Thus, the proposed exit levels also indicate the relative importance of the exposure pathways for each chemical. Exhibit 4-16 tabulates the exit levels for 128 of the known or possible non-hazardous industrial waste constituents (i.e., the entire combined list prior to any screens that were also addressed in the HWIR-waste proposed rulemaking), and the exposure pathways that were risk drivers for setting the exit levels. As in the case of the similar analysis for the TC analytes in Section 3.5, many of the known or possible non-hazardous industrial waste constituents have proposed exit levels that are quite low (68 are below 0.1 mg/l). Therefore, the Agency has determined that the presence of these constituents in wastes at even relatively low concentrations may pose significant risks to human health. Again it should be noted that the target cancer risk level used to derive the exit levels was 10-6, rather than the 10-5 level used in the derivation of TC regulatory levels. Even so, these levels indicate potential cause for concern for many of these chemicals at even low concentrations in wastes. As was also the case for the TC analytes, non-groundwater pathway risks drive the establishment of exit levels for about one-quarter of the known or possible non-hazardous industrial waste constituents. The driving pathways include direct inhalation and vegetable and milk ingestion. Pesticides make up a large proportion of the chemicals for which non-groundwater pathways drive the risks, but many volatile chlorinated and nonchlorinated organics also fall into this category. Ecological, rather than human health risks, drive the setting of proposed exit levels for two chemicals (copper and parathion). These findings confirm the indications from the toxicity and fate and transport screening presented in the previous sections that inhalation and indirect pathways could be of concern for many of the known or possible non-hazardous industrial waste constituents. 4.6 Potential Acute Hazards Associated With Known and Possible Non-Hazardous Industrial Waste Constituents To this point, the evaluation of the potential hazards associated with the possible and known non-hazardous industrial waste constituents has focused on chronic toxic effects. As discussed in Chapter 3, waste constituents may also pose risks from acute exposures, as well as from physical hazards associated with reactivity, flammability, or corrosivity. To investigate the possibility of acute adverse effects, the Agency has compared list of the known and possible waste constituents to lists developed by the EPA and other regulatory agencies that identify such hazardous properties. The results of this comparison are summarized in Exhibit 4-17. As shown in the exhibit, 38 of the known and possible non-hazardous industrial waste constituents have been identified in one or more regulatory contexts as being acutely toxic. Although most of these chemicals are volatile organics, several acid gases and other inorganic compounds also are included. Appearance on these lists does not automatically indicate that acute adverse effects will occur, only that such effects could potentially be associated with management of wastes containing these chemicals. Fifteen of the waste constituents are also identified as being highly flammable. These are mostly volatile organics, along with a few inorganic gases and liquids. They substantially overlap with the previous list. Only two of the known or possible non-hazardous industrial waste constituents are identified as being highly reactive. 4.7 Identify Individual Chemicals and Classes of Chemicals Constituting Potential Gaps The analyses in the previous sections help to clarify the nature of potential gaps in the hazardous waste characteristics associated with specific chemicals and chemical classes related to chronic human health risks and ecological risks. The analyses identified groups of chemicals most likely to be present in non-hazardous industrial waste, and screened them in terms of their toxicity, fate, and transport properties. The results of the proposed HWIR-waste modeling were reviewed, where available, to confirm and expand the findings of the screening results. Finally, the known and possible non-hazardous industrial waste constituents were reviewed with regard to their potential to cause acute adverse effects. As a result of these efforts, a number of potential gaps have been identified, as summarized in Exhibit 4-18. This listing of potential gaps for non-TC analytes should not be taken as being either exhaustive or definitive. These gaps are potential, not actual gaps. They have been identified for purposes of targeting further analysis, not for purposes of choosing what constituent or wastes to regulate. Other potential gaps related to natural resource damages and regional or global environmental problems are discussed in Chapter 5. Also, Chapter 6 describes how several states have expanded the TC, implicitly identifying gaps in the TC. In Chapter 10, some of the unresolved issues identified in Exhibit 4-18 are discussed and the available information about the potential significance of these impacts is reviewed in detail. EPA recognizes the limitations of this analysis. As noted previously, the data concerning the composition of non-hazardous industrial wastes are quite limited and generally quite old. This lack of data in large part explains the need for the elaborate screening procedures employed in this chapter. Few data are available on the current patterns of non-hazardous industrial waste generation, management, and disposal. In addition, the chemical-specific screening is further complicated by the lack of toxicity and fate and transport parameter data for a large proportion of the universe of possible waste constituents, which necessitated extensive use of professional judgment to supplement the screening process. CHAPTER 4. POTENTIAL GAPS ASSOCIATED WITH NON-TC CHEMICALS. . . . . . .4-1 4.1 Overview of Methodology. . . . . . . . . . . . . . .4-1 Step 1: Identify and Classify Known Non-Hazardous Industrial Waste Constituents 4-1 Step 2: Identify and Screen Possible Non-Hazardous Industrial Waste Constituents 4-1 Step 3: Apply Hazard-Based Screening Criteria . . .4-3 Step 4: Review Relevant Multipathway Risk Modeling Results 4-3 Step 5: Identify Potential Acute Hazards. . . . . .4-3 Step 6: Summarize Findings. . . . . . . . . . . . .4-3 4.2 Identify and Classify Known Constituents of Non-Hazardous Industrial Wastes 4-3 4.3 Identify Possible Non-Hazardous Industrial Waste Constituents of Potential Concern 4-7 4.3.1 Approach to Identifying Potentially Hazardous Chemicals 4-7 4.3.2 Screening Approach . . . . . . . . . . . .4-9 4.3.3 Toxicity, Fate, and Transport Screening for Possible Non-Hazardous Industrial Waste Constituents 4-12 4.3.4 Release Volume Screening of Possible Non-Hazardous Industrial Waste Constituents 4-15 4.3.5 Summary of Possible Non-Hazardous Industrial Waste Constituents 4-20 4.4 Combine and Screen Known and Possible Non-Hazardous Industrial Waste Constituents 4-20 4.4.1 Combine the Lists. . . . . . . 4-20 4.4.2 Screen Combined List Against Single Criteria4-26 4.4.3 Screen Combined List Against Multiple Parameters 4-34 4.5 Driving Risk Pathways for the Known and Possible Non-Hazardous Industrial Waste Constituents 4-35 4.6 Potential Acute Hazards Associated With Known and Possible Non-Hazardous Industrial Waste Constituents 4-36 4.7 Identify Individual Chemicals and Classes of Chemicals Constituting Potential Gaps 4-41 Exhibit 4-1 Flow Chart of Procedures Used to Identify Non-TC Chemicals Posing Potential Gaps in the TC Characteristics 4-2 Exhibit 4-2 Known Non-Hazardous Industrial Waste Constituents Chemicals Found in Case Studies, ISDB, Listings Documents, Effluent Guideline Documents by Chemical Class and Listing Documents 4-6 Exhibit 4-3 Lists Used to Identify Possible Non-Hazardous Industrial Waste Constituents 4-8 Exhibit 4-4 Criteria Considered for Screening Non-Hazardous Industrial Waste Constituents 4-10 Exhibit 4-5 Toxicity Screening Results for Possible Non-Hazardous Industrial Waste Constituents 4-13 Exhibit 4-6 Persistence and Bioconcentration/ Bioaccumulation Screening Results for Possible Non-Hazardous Industrial Waste Constituents 4-16 Exhibit 4-7 Screening of High-Toxicity, Persistent, Bioaccumulative/Bioconcentrating Possible Non-Hazardous Industrial Waste Chemicals Against TRI Release Volumes 4-17 Exhibit 4-8 List of Known and Possible Non-Hazardous Industrial Waste Constituents By Chemical Class 4-21 Exhibit 4-9 Screening of Known Non-Hazardous Industrial Waste Constituents Against TRI Release Volumes 4-22 Exhibit 4-10 Toxicity Summary of Known and Possible Non-Hazardous Industrial Waste Constituents 4-27 Exhibit 4-11 Potential Endocrine Disruptors. . . 4-28 Exhibit 4-12 TRI Releases and TSCA Production Volume Data for the Known and Possible Non-Hazardous Industrial Waste Constituents 4-29 Exhibit 4-13 Volatility, Persistence, and Bioaccumulation/ Bioconcentration Potential of Known and Possible Non-Hazardous Industrial Waste Constituents 4-30 Exhibit 4-14 LNAPL/DNAPL Formation Potential of Known and Possible Non-Hazardous Industrial Waste Constituents 4-33 Exhibit 4-15 Multiple Screening Criteria Applied to Known and Possible Non-Hazardous Industrial Waste Constituents 4-35 Exhibit 4-16 Lowest Proposed HWIR-Waste Exit Levels for Known and Possible Non-Hazardous Industrial Waste Constituents 4-37 Exhibit 4-17 Potential Acute Hazards Associated with Known and Possible Non-Hazardous Industrial Waste Constituents 4-40 Exhibit 4-18 Potential Gaps in the Hazardous Waste Characteristics Identified Based on the Hazardous Properties of Known and Possible Non-Hazardous Industrial Waste Constituents 4-42 CHAPTER 5. POTENTIAL GAPS ASSOCIATED WITH NATURAL RESOURCE DAMAGES AND LARGE-SCALE ENVIRONMENTAL PROBLEMS This chapter discusses risks associated with non-hazardous industrial waste management that are not addressed in Chapters 3 or 4. Chapter 3 examined potential gaps inherent in the current hazardous waste characteristics, thereby focusing on the adverse effects that the characteristics were meant to address, namely risks arising primarily from acute events such as fires, explosions, and acute exposures of waste management and transportation workers, and health risks caused by local environmental contamination near waste management units. Chapter 4 examined potential gaps associated with adverse human health or localized ecological effects from constituents not included in the toxicity characteristic. This chapter addresses a third set of risks associated with non-hazardous industrial waste management. ! Section 5.1 addresses the pollution of groundwater by constituents that diminish the value and usability of the resource without threatening human health; ! Section 5.2 addresses damage from non-hazardous industrial waste management to air quality through odors that harm the quality of life but may not have severe health effects; and ! Section 5.3 examines possible contributions to regional and global environmental problems from the management of non-hazardous industrial waste, including: air deposition to the Great Waters, damages from airborne particulates, global climate change, potential damage from endocrine disruptors, red tides, stratospheric ozone depletion, tropospheric ozone and photochemical air pollution, and water pollution. These environmental problems may or may not meet the RCRA statutory or regulatory definitions of the types of risks that the hazardous waste management program is meant to address. 5.1 Damage to Groundwater Resources As noted in Chapter 2, the most common and well-documented impact of releases from non-hazardous industrial waste management is groundwater contamination. If contamination is present at high enough concentrations, the use of the groundwater as a water supply for human consumption or other use may result in adverse effects on health. Human health risks associated with exposure to toxic pollutants are not the only concern associated with groundwater contamination, however. Non-toxic pollutants such as iron, chloride, or total dissolved solids may be present in concentrations that damage the aesthetic qualities and usability of the water without posing outright health hazards. In areas where groundwater is used as a drinking water supply, such water pollution must be remediated, limitations must be placed on its use, and/or alternative sources must be found. These actions may be expensive and strain existing water supplies. Where alternative supplies are not economically available, groundwater resources of marginal quality, which do not exceed health-based levels, may continue to be used. Even where the polluted groundwater is not used for drinking water, the value of the resource may decline because it is no longer available for future use as drinking water without remediation. This non-toxic pollution of groundwater from non-hazardous industrial waste management was found relatively often in the environmental release descriptions summarized in Chapter 2. Seventy-five (84 percent) of the 89 release descriptions with data on regulatory levels had constituents detected at levels exceeding non-health-based or non-ecologically-based standards, principally on aesthetic or usability criteria developed under the Safe Drinking Water Act as Secondary Maximum Contaminant Levels (SMCLs). Releases at 70 of these 75 sites also exceeded health and/or ecological-based standards. Of the 177 non-TC constituents identified in the release case studies, 9 constituents (plus pH and total dissolved solids) have SMCLs. (Some of these constituents also have health-based or ecologically-based levels.) Exhibit 5-1 lists all constituents with SMCLs and shows how frequently they were found among the 89 case studies where concentration and regulatory standards data were available. The most commonly detected constituents, iron, chloride, and manganese, all have SMCLs. Also, all SMCLs, except those for foaming agents, color, and corrosivity, were violated by at least several documented releases. (See Exhibit 2-6 for additional data on the concentrations at which these constituents were detected.) 5.2 Damage to Local Air Quality from Odors Noxious odors historically have been reported in the vicinity of waste management facilities. Odor problems have caused minor health problems, reduced the quality of life, and reduced property values near such facilities. Information on the extent of such problems from non-hazardous industrial waste management is very limited. Odor problems were reported in several of the release descriptions initially identified by EPA, but these cases were excluded because they did not meet the Agency's strict selection criteria. Only one release description included reports by residents of odor problems. Nevertheless, the case study development methodology may have missed many cases of odor problems from non-hazardous industrial waste management facilities because state regulatory programs largely focus on groundwater concerns. Also, odor problems are often handled at the local level and thus the states may not get involved. The potential for odor problems clearly exists at non-hazardous industrial waste facilities that manage certain types of wastes. For example, food processing facilities (e.g., slaughterhouses that must dispose of offal and alimentary contents from slaughtered animals) may have odor problems if their air releases are not carefully managed. In addition to food wastes, potential odor problems may arise from chemical wastes. Exhibit 5-2 lists a number of the chemicals identified in the release descriptions (although not necessarily for odor) that have extremely low odor thresholds in either air or water. Ten of these chemicals have threshold odor concentrations in air (the lowest concentrations at which odors can be detected or recognized) of 0.01 mg/m3 or less, and six of them can be detected by odor in water solutions at concentrations of 0.006 mg/l or less. Because odor problems typically are handled locally and these problems likely do not meet the RCRA definition of risks meant to be addressed by the hazardous waste management program, EPA does not plan to investigate this area further following the Scoping Study. 5.3 Large-Scale Environmental Problems EPA considered whether any major large-scale environmental problems (e.g., global climate change, potential damage from endocrine disruptors) might be caused, at least to some extent, by non-hazardous industrial wastes. Depending on the types of wastes and on the relative contributions of these wastes to the problem areas, changes in the hazardous waste characteristics might be one method to help reduce further damages. EPA began this phase of the Scoping Study by developing an initial list of major large-scale environmental problems (or possible problems) that have potential links to non-hazardous industrial wastes (see Exhibit 5-3). Several of these problems overlap considerably with each other and with exposure and other damage pathways discussed previously. Furthermore, EPA recognizes that other environmental problems have potential links to non-hazardous industrial waste; however, given the limited resources available for this Scoping Study, the Agency chose to limit this analysis to some of the more likely areas of concern. Following the development of this list, EPA conducted preliminary evaluations of the problem areas to try to characterize the contributions to the problems from non-hazardous industrial wastes. Because these problems are typically characterized by highly complex interactions of a large number of factors, determining the exact contribution of non-hazardous industrial wastes to each problem is difficult and beyond the scope of this study. Instead, EPA was able to conduct only initial evaluations to identify areas that may have a significant contribution from non-hazardous industrial wastes and thus may warrant further analysis following the Scoping Study. For environmental problems with a possible link to non-hazardous industrial wastes, EPA identified (where possible) the industries and waste streams that could be contributing to the problems and the relevant statutes and programs that are addressing the areas. The environmental problems evaluated for this Scoping Study are discussed below in the order (alphabetical) listed in Exhibit 5-3. 5.3.1 Air Deposition to the Great Waters Pollutants emitted into the atmosphere are transported various distances and can be deposited to aquatic ecosystems far removed from their original sources. Studies show that significant portions)often greater than 50 percent)of pollutant loadings to the Great Waters (i.e., Great Lakes, Lake Champlain, Chesapeake Bay, and coastal waters) are from atmospheric deposition. Thus, this pathway is an important factor in the degradation of water quality and the associated adverse health and ecological effects. Because of the mounting concern that air pollution contributes to water pollution, Congress included Section 112(m), Atmospheric Deposition to Great Lakes and Coastal Waters, in the Clean Air Act Amendments of 1990. Both local and distant air emission sources contribute to a pollutant load at a given location. The sources of concern for the Great Waters primarily include industrial activities and processes involving combustion. At present, however, a complete and comprehensive inventory of the locations of particular sources and the amount of individual toxic pollutants that each source emits to the air is lacking. Nevertheless, EPA has identified several known air pollutants of concern for Great Waters. Exhibit 5-4 lists these pollutants and selected U.S. sources. Most pollutants in this exhibit are TC analytes, while a smaller set are chemicals (or chemical groups) of concern discussed in Chapter 4. Thus, these pollutants are likely candidates for further analysis as potential gaps in the hazardous waste characteristics. 5.3.2 Airborne Particulates Airborne particulate matter (PM) is one of the six high-priority research topics identified for the next few years by the EPA Office of Research and Development (ORD). PM includes dust, dirt, soot, smoke, and liquid droplets directly emitted into the air by sources such as factories, power plants, transportation sources, construction activity, fires, and windblown dust. Concern regarding PM from non-hazardous industrial waste includes toxic constituents entrained on particulates. PM is also formed in the atmosphere by condensation or transformation of emitted gases such as sulfur dioxide, nitrogen oxides, and volatile organic compounds into small droplets. Based on studies of human populations exposed to high concentrations of particles (often in the presence of sulfur dioxide) and on laboratory studies of animals and humans, the major concerns for human health include effects on breathing and respiratory symptoms, aggravation of existing respiratory and cardiovascular disease, alterations in the body's defense systems against foreign materials, damage to lung tissue, carcinogenesis, and premature death. The major subgroups of the populations that appear likely to be most sensitive to the effects of particulate matter include individuals with chronic obstructive pulmonary cardiovascular disease, individuals with influenza, asthmatics, the elderly, and children. Particulate matter may injure crops, trees and shrubs, and may damage metal surfaces, fabrics, and other materials. Fine particulates also impair visibility by scattering light and reducing visibility. The haze caused by fine particles can diminish crop yields by reducing sunlight. PM is increasingly being identified as posing a high potential for health and environmental risk and other potential damages. Nevertheless, EPA does not believe that PM is a significant waste characterization issue but rather a waste management issue. Furthermore, other programs (e.g., CAA National Ambient Air Quality Standards) are designed to address this area. Therefore, airborne particulates are not planned for further study as a potential gap. 5.3.3 Global Climate Change Evidence is mounting that the increasing concentrations of greenhouse gases (GHGs) will ultimately raise (and some believe are currently raising) atmospheric and ocean temperatures significantly, which may in turn alter global weather patterns. Global climate already has changed over the past century, and the balance of evidence suggests a discernible human influence. Climate is expected to continue to change in the future. EPA conducted a brief review of the major anthropogenic sources of the two predominant GHGs, carbon dioxide (CO2) and methane (CH4), to determine the relative contributions of non-hazardous industrial wastes, including their co-disposal with municipal solid waste (MSW). Before describing the results of this review, it is essential to understand some of the international conventions used to evaluate GHG emissions, as these conventions have a strong bearing on the results. The United States and all other parties to the Framework Convention on Climate Change agreed to develop inventories of GHGs for purposes of developing mitigation strategies and monitoring the progress of those strategies. The Intergovernmental Panel on Climate Change (IPCC) developed a set of inventory methods to be used as the international standard. The screening methodology used in this section to evaluate emissions and sinks of GHGs attempts to be consistent with IPCC's guidance. One of the elements of the IPCC guidance that deserves special mention is the approach used to address CO2 emissions from biogenic sources. For many countries, the treatment of CO2 releases from biogenic sources is most important when addressing releases from energy derived from biomass (e.g., burning wood), but this element is also important when evaluating waste management emissions (for example, the decomposition or combustion of grass clippings or paper). The carbon in paper and grass trimmings was originally removed from the atmosphere by photosynthesis, and under natural conditions, it would eventually cycle back to the atmosphere as CO2 due to degradation processes. The quantity of carbon that these natural processes cycle through the earth's atmosphere, waters, soils, and biota is much greater than the quantity added by anthropogenic GHG sources. But the focus of the Framework Convention on Climate Change is on anthropogenic emissions emissions resulting from human activities and subject to human control because these emissions have the potential to alter the climate by disrupting the natural balances in carbon's biogeochemical cycle. Thus, for processes with CO2 emissions, if the emissions are from biogenic materials and the materials are grown on a sustainable basis, then those emissions are considered to simply close the loop in the natural carbon cycle; that is, they return CO2 to the atmosphere that was originally removed by photosynthesis. In such cases, the CO2 emissions are not counted (and thus most CO2 emissions from landfills are not counted). On the other hand, CO2 emissions from burning fossil fuels are counted because these emissions would not enter the cycle were it not for human activity. Likewise, CH4 emissions from landfills are counted, even though the source of carbon is primarily biogenic. CH4 would not be emitted but for the human activity of landfilling the waste, which creates anaerobic conditions conducive to CH4 formation. This approach does not distinguish between the timing of CO2 emissions, provided that they occur in a reasonably short time scale relative to the speed of the processes that affect global climate change. That is, as long as the biogenic carbon would eventually be released as CO2, it does not matter whether it is released virtually instantaneously (e.g., from combustion) or over a period of a few decades (e.g., decomposition on the forest floor). CO2 accounts for the largest share of U.S. GHG emissions, comprising 1,408 million metric tons of carbon equivalent (MMTCE) out of total 1994 U.S. emissions of 1,666 MMTCE. Combustion of fossil fuels results in the vast majority of the CO2 emissions (1,390 MMTCE), with the remainder from industrial processes such as cement production, lime production, limestone consumption (e.g., iron and steel production), soda ash production and use, and CO2 manufacture. CO2 emitted from landfills as a product of both aerobic and anaerobic decomposition of organic wastes is not counted, as described above. Methane is the second most important GHG; U.S. emissions in 1994 were 166 MMTCE. Of the anthropogenic CH4 sources, the largest is landfills (which contribute 36 percent of the total U.S. methane emissions), agricultural activities (32 percent), coal mining (15 percent), production and processing of natural gas and oil (11 percent), fossil fuel combustion (3 percent), and wastewater treatment (0.6 percent). As explained above, CH4 from landfills is counted as an anthropogenic GHG. The majority of landfill CH4 emissions result from MSW landfills (90 to 95 percent), with the remaining methane emitted from the disposal of industrial wastes. Methane emissions from large MSW landfills, however, are currently regulated under EPA's recent New Source Performance Standards and Emissions Guidelines, which require collection and control of landfill gas. Small MSW landfills and industrial waste monofills are not subject to these new regulations and thus may warrant further investigation. This is particularly true for small landfills or monofills managing non-hazardous industrial wastes that have a high biochemical oxygen demand (such as wastes from paper mills and food processing), which have a high potential for generating CH4. In conclusion, non-hazardous industrial wastes may contribute to GHG emissions to the extent that they are highly degradable and either are disposed in small landfills (which are not subject to the landfill gas rule) or are released directly to the atmosphere. The emissions attributable to these wastes are small compared to other sources of GHGs. Nevertheless, the same highly putrescible wastes that would be of concern when disposed in a landfill environment are likely to cause taste and odor problems that adversely affect local air and water quality. To a large degree, the climate change risk (and much of the potential groundwater resource damage) could be readily averted for highly putrescible wastes by biological pretreatment prior to land disposal to reduce the potential for (a) methane formation and (b) production of odiferous compounds generated in an anaerobic environment. Further research could be conducted in this area to determine whether the potential contribution of non-hazardous industrial wastes to GHG emissions could be significant. However, given the current coverage of this problem area by other programs besides Subtitle C of RCRA, EPA does not plan to pursue global climate change within the context of the hazardous characteristics at this time. 5.3.4 Potential Damages from Endocrine Disruptors Over the past decade, increased attention has been given to a class of chemicals with high persistence, bioaccummulation potential, and toxicity. These chemicals, often referred to as PBTs, include a wide range of substances, generally several metals and a variety of organic compounds. EPA's involvement in PBT research and regulation has encompassed many programs. One of these programs, waste minimization, developed the Waste Minimization National Plan. This plan established a national goal to reduce the most persistent, bioaccumulative, and toxic chemicals in hazardous wastes by 25 percent by the year 2000 and by 50 percent by the year 2005. Currently many international organizations, including the North American Commission for Environmental Cooperation and various United Nations groups, are debating PBT public policy and ultimately could generate binding commitments (e.g., treaties) that could affect U.S. national policy on PBTs. For example, an initial list of 12 PBTs is being considered for control under an international protocol. Recently, interest in PBTs has escalated due to the growing attention on a subgroup of these chemicals called "endocrine disruptors" (EDs). EDs are substances that have the potential to interfere with hormonal systems in ecological and human receptors. The results of such interference might include adverse reproductive or developmental effects, certain kinds of cancers, learning and behavioral problems, and immune system deficiencies. Recent concern has focused on the potential synergistic effects of EDs. Significant scientific debate still exists regarding which chemicals are EDs and the degree to which EDs have caused or have the potential to cause adverse human health and environmental effects. This debate has prompted great interest in researching the scope of ED impacts. For example, the study of EDs is one of the six high-priority research topics identified by EPA's Office of Research and Development (ORD) for the next few years. It has also been made a high priority by the U.S. chemical industry; the Chemical Industry Institute for Toxicology (CIIT) has reprogrammed much of its research budget into this area. To the extent that the impact of EDs on the environment are largely unknown, these chemicals may represent a substantial gap in the hazardous waste regulations. Notwithstanding the current debate, recent review articles summarize convincing evidence that a variety of chemical pollutants can act as endocrine disruptors in wildlife populations. Some specific examples include the following: ! Reptiles. Researchers found that the reproductive development of alligators from Lake Apopka, Florida was severely impaired, apparently due to DDE, a metabolite of DDT and dicofol. The lake is located adjacent to an EPA Superfund site where a dicofol spill had occurred. The specific adverse effects included decreased testosterone and abnormal testicular cells in males and increased estrogen and altered ovaries (increased polyovular follicles and polynuclear oocytes) in females. ! Birds. A number of researchers have documented severely impaired reproductive success in herring gulls from the Great Lakes. Some specific observations include large clutch sizes (attributed to nest sharing by two females), female-female pair bonds, embryonic and chick mortality, and altered nest defense and incubation behavior. These effects were associated with high levels of organochlorines (e.g., DDT, dioxins, and mirex) in the 1960s and early 1970s. Reproductive success increased as levels of these compounds declined in the late 1970s and 1980s. Organochlorines that have been identified as estrogenic to bird embryos in laboratory studies include DDT and methoxychlor. In these cases, some of the causative agents appear to be organochlorine pesticides that are no longer produced (e.g., DDT) yet persist in the environment due to the nature of their chemical/physical properties. Although these chemicals are not generally expected to be components of non-hazardous industrial wastes, a number of similar chemicals currently used in industry have demonstrated similar endocrine disrupting properties in laboratory studies. These EDs are often present in treated sewage effluent, and are likely to be components of non-hazardous industrial waste. A recent field study found that effluent from sewage treatment works induced vitellogenin synthesis in male fish, indicating that the effluent is estrogenic. The effects were pronounced and occurred at all sites tested. The identity of the chemical or chemicals in the sewage effluent causing the effects is not known, however. A number of chemicals known to be present in sewage effluent were tested for estrogenic effects in fish. These chemicals included nonylphenol, octylphenol, bisphenol-A, DDT, and PCBs. Furthermore, a mixture of different estrogenic chemicals was found to be considerably more potent than each of the chemicals when tested individually, a finding that recently was replicated. In addition to the effects described above, other documented endocrine disrupting effects in wildlife populations from industrial effluents have unknown causative agents. For example, kraft mill effluent caused a variety of effects in two fish species: white suckers and mosquitofish. Lake Superior white suckers collected from a site receiving primary-treated bleached kraft mill effluent exhibited increased age to maturity, smaller gonads, lower fecundity with age, and an absence of secondary sex characteristics. Masculinization of female mosquitofish was noted downstream from the discharge of kraft mill effluent in Elevenmile Creek in Florida. Several of the chemicals identified in this section are also identified in Chapter 4 as known or possible non-hazardous industrial waste constituents. Some of the relevant chemical groups are described in more detail below. ! Alkylphenol Compounds. Alkylphenol-polyethoxylates are non-ionic surfactants commonly used in industrial and domestic detergents as well as some shampoos and cosmetics. Alkylphenols are used as antioxidants in some clear plastics. Alkylphenol-polyethoxylates are biodegraded to alkylphenols during sewage treatment. These compounds persist in rivers and their sediments and can migrate to groundwater. These compounds also have the ability to bioconcentrate in animals. ! Bisphenol-A. This compound is used to manufacture polycarbonate, a component in a wide array of plastics and other polymer products. Bisphenol-A also is used to manufacture epoxy resins, which are components of a variety of lacquers and adhesives. ! Phthalates. Phthalates are one of the most abundant man-made chemicals in the environment. Phthalate esters are used in the production of various plastics. Butylbenzyl phthalate (BBP) also is used in the production of vinyl floor tiles, adhesives, and synthetic leather. Di-n-butylphthalate (DBP) is a common plasticizer in food-packaging materials and polyvinyl chloride. Thousands of tons of plastics are disposed of annually in landfills, thus possibly enabling phthalate esters to migrate into soil and groundwater. These compounds have the ability to bioconcentrate in animals. As seen in Chapter 4, other categories of chemicals with ED characteristics (e.g., heavy metals) are present in wastes generated by numerous industries. In conclusion, the evidence that alkylphenols, bisphenol-A, and phthalates are endocrine disruptors is based mainly on laboratory studies. The effects of these chemicals on wildlife populations is not known. Based on the endocrine disrupting effects of organochlorines on populations of fish, birds, reptiles, and mammals, however, it is possible that alkylphenols, bisphenol-A, phthalates, and other chemicals also could have endocrine disrupting effects in wildlife. Furthermore, as seen in Chapter 4, it is likely that some of these chemicals (e.g., the phthalates) are also components of several non-hazardous industrial wastes. 5.3.5 Red Tides Red tides are rapid increases in growth (i.e., blooms) of freshwater and marine plants called dinoflagellates, which typically are microscopic unicellular organisms that photosynthesize but also have tails for movement. A red tide occurs when dinoflagellates multiply rapidly due to optimal growth conditions such as abundant dissolved nutrients and sunlight. They produce toxins to defend themselves from zooplankton and other aquatic grazers. The term red tides includes orange, brown, red, and even green blooms. Shellfish, such as clams, mussels, oysters, or scallops, consume dinoflagellates and can accumulate the toxins in their flesh. Usually, the shellfish are not severely affected, but they can contain enough toxins to sicken and even kill humans. The recently discovered Pfiesteria piscida is one of many species of dinoflagellate that causes red tides. It produces potent toxins that cause bleeding sores in fish and can adversely affect humans via air releases. It recently has caused massive fish kills in the Neuse and Pamlico Rivers in North Carolina. Several case studies have shown the relationship between the levels of nutrients, such as phosphorus, nitrogen, silicon, and iron, in coastal and fresh waters, and the proliferation of red tides. Studies also have shown that the high levels of nutrients and eutrophication of the water (which favors the development of red tides) are often caused by surrounding human development and industrial and domestic wastewaters. Recent development of agribusiness and factory farms in coastal areas releases wastes with high levels of nutrients into the water that may favor red tides. Some researchers believe that the occurrence of red tides has been increasing over the years, although improvements in the monitoring and reporting of red tides could account for this. Even if such an increase were occurring, however, a commensurate increase in human poisoning from ingestion of shellfish contaminated with dinoflagellate toxins has not been seen, likely because of the improved monitoring and reporting of red tides. Notwithstanding the potential link between red tides and constituents that are often found in non-hazardous industrial waste, little if any evidence has been found during this review concerning the degree to which these wastes may be contributing to the problem. Therefore, for the purposes of this hazardous waste characteristic gaps study, EPA does not plan to conduct further research in this area at this time. 5.3.6 Stratospheric Ozone Depletion The stratospheric ozone layer protects living organisms from damaging solar ultraviolet radiation (UV-B). Depletion of the ozone layer means a greater amount of UV-B radiation is reaching the earth's surface, which increases human skin cancers and cataracts, impairs human immune systems, reduces crop yields, and damages plant and animal life. Several industrial chemicals, including chlorofluorcarbons (CFCs), halons, carbon tetrachloride, methyl chloroform, and methyl bromide, are known to be stratospheric ozone-depleting substances (ODSs). For many years, ODSs have been used in a variety of manufacturing and other activities. With the ratification of the Montreal Protocol and its subsequent amendments and adjustments, the United States agreed to eliminate the production of ODSs by January 1, 1996 (with a few exceptions). In addition, the disposal of ODSs is tightly controlled in order to prevent further ozone depletion. Thus, EPA believes that, for purposes of the hazardous waste characteristic gaps analysis, ozone-depleting and non-ozone-depleting risks (e.g., via inhalation during combustion or from groundwater during land disposal of residuals) do not need to be examined further at this time. In a related area (though not necessarily a large-scale environmental problem), the ultimate elimination of ODSs has spurred the development of a large number of alternative chemicals and technologies to replace ODSs. In the United States, the Significant New Alternatives Policy (SNAP) Program was put in place by EPA to ensure that alternatives implemented to replace ODSs are not themselves environmentally harmful or unsafe for workers and others who might be exposed to the new chemicals. As part of this program, EPA has developed a series of SNAP Technical Background Documents to address the ODS substitutes. Before a new alternative is developed and introduced into interstate commerce, EPA must review the alternative and categorize it as acceptable, acceptable with limitations, or unacceptable, based on a risk screen of the alternative's characteristics. This risk screen addresses global atmospheric effects of the alternative, as well as worker, consumer, and general population exposure. Thus, groundwater damage and other more local adverse effects of the alternative from solid waste generation and management are included in this screening process. Therefore, EPA does not intend to conduct further investigations into the solid waste and hazardous characteristics implications of the SNAP-approved alternatives at this time. 5.3.7 Tropospheric Ozone and Photochemical Air Pollution Photochemical reactions between organic chemicals, nitrogen oxides, and other oxidizing agents can produce ozone and photochemical oxidant pollution. Such pollution occurs in areas where sunlight is intense, emissions of nitrogen oxides and volatile organic compounds (VOCs) are high, and atmospheric conditions impede regional air circulation. Some chemicals emitted from non-hazardous industrial waste management units could contribute to the total emissions of volatile organics in some locations. As shown in Exhibit 4-2, many potentially reactive VOCs have been found as constituents of non-hazardous industrial wastes. This contribution, however, appears to be quite small. Recent emissions studies have shown that, in most municipal areas where photochemical pollution is a problem, mobile and utility sources contribute the largest single portion of these emissions, with emissions from other sources generally contributing a smaller amounts. Thus, the Agency did not pursue this issue further as a potential gap in the hazardous waste characteristics. 5.3.8 Water Pollution Based on information reported to EPA by States, Tribes, and other jurisdictions with water quality responsibilities, about 40 percent of the Nation's surveyed rivers, lakes, and estuaries are not clean enough for basic uses such as fishing or swimming. Polluted runoff from rainstorms and snowmelt is the leading cause of this impairment. As seen below, the causes of this damage are highly varied. ! Rivers. Runoff from agricultural lands is the largest source of pollution for rivers. Municipal sewage treatment plants, storm sewers/urban runoff, and resource extraction also are among the leading sources. Bacteria, which can cause illnesses in swimmers and others involved in water-contact sports, are the most common pollutants impacting rivers. Siltation, nutrients (such as phosphates and nitrates), oxygen-depleting substances, and metals are the other leading causes of river pollution. ! Lakes. As with rivers, runoff from agricultural lands is the largest source of pollution. Municipal sewage treatment plants, storm sewers/urban runoff, and unspecified nonpoint sources also lead the list. Leading causes of lake pollution are nutrients, siltation, oxygen-depleting substances, metals, and suspended solids. ! Estuaries. Storm sewers and urban runoff are the leading sources of pollution in estuaries. Municipal sewage treatment plants, agriculture, industrial point sources, and petroleum activities also lead the list. Nutrients, such as phosphates and nitrates, are the most often reported pollutant in estuaries. Other leading causes of pollution are bacteria, oxygen-depleting substances, and oil and grease. Although non-hazardous industrial wastes contribute to this pollution to some degree (e.g., via sewage treatment and industrial point and non-point sources), it is unclear whether this contribution constitutes an actual gap in the hazardous waste characteristics. For example, significant changes in EPA's definition of solid waste would be needed before the hazardous waste characteristics could be used to prevent some of these wastes from entering surface waters and resulting in risks or damage. Industrial wastewaters that are point source discharges subject to regulation under the Clean Water Act are exempt from the definition of solid waste. Many of the wastes from agriculture one of the largest contributers to water pollution from runoff are exempt from the definition of hazardous waste (although they are solid wastes). Alternatively, EPA could increase controls on point and non-point sources of water pollution via other programs. Thus, for purposes of the hazardous characteristic scoping study, EPA does not plan to research this area further at this time. CHAPTER 5. POTENTIAL GAPS ASSOCIATED WITH NATURAL RESOURCE DAMAGES AND LARGE-SCALE ENVIRONMENTAL PROBLEMS 5-1 5.1 Damage to Groundwater Resources. . . . . . . . . . .5-1 5.2 Damage to Local Air Quality from Odors . . . . . . .5-2 5.3 Large-Scale Environmental Problems . . . . . . . . .5-4 5.3.1 Air Deposition to the Great Waters . . . .5-4 5.3.2 Airborne Particulates. . . . . . . . . . .5-6 5.3.3 Global Climate Change. . . . . . . . . . .5-7 5.3.4 Potential Damages from Endocrine Disruptors 5-9 5.3.5 Red Tides. . . . . . . . . . . . . . . . 5-13 5.3.6 Stratospheric Ozone Depletion. . . . . . 5-14 5.3.7 Tropospheric Ozone and Photochemical Air Pollution 5-15 5.3.8 Water Pollution. . . . . . . . . . . . . 5-15 Exhibit 5-1 Constituents/Properties with SMCLs Found in Release Descriptions 5-2 Exhibit 5-2 Chemicals from Release Descriptions with Low Odor Thresholds 5-3 Exhibit 5-3 Initial List of Large-Scale Environmental Problems 5-4 Exhibit 5-4 U.S. Sources of Air Pollutants of Concern for Great Waters 5-5 CHAPTER 6. STATE EXPANSIONS OF THE TOXICITY CHARACTERISTIC AND LISTINGS States may adopt hazardous waste regulations that are broader or more stringent than federal RCRA Subtitle C regulations. A number of states have done so by regulating additional wastes as hazardous. For example, states have: ! Expanded the ignitability, corrosivity, or reactivity (ICR) characteristics; ! Expanded the toxicity characteristic (TC); ! Listed wastes as hazardous that are not hazardous under the federal rules; and ! Restricted exemptions from the federal program. These expansions beyond the federal hazardous waste identification rules reflect state judgments about gaps in the federal program and thereby constitute potential gaps that may merit further investigation. EPA has identified examples of such expansions by using readily available information on state hazardous waste identification rules. In 1992, the EPA Office of Solid Waste, OSW (renamed Office of Resource Conservation and Recovery, ORCR, on January 18, 2009) examined state hazardous and non-hazardous industrial waste programs in 32 states. The study identified "state only" hazardous wastes, as well as high-risk designations for non-hazardous wastes. For the purposes of this Scoping Study, EPA used data from this report and briefly reviewed current hazardous waste regulations of eight states: California, Michigan, New Hampshire, Oregon, Rhode Island, Texas, Washington, and New Jersey. The first three sections of this chapter address state expansion of the TC, state only hazardous waste listings, and state restrictions on exemptions from the federal regulations, respectively. (State expansions of the ICR characteristics are addressed in Chapter 3.) In addition, Section 6.4 summarizes the findings of the chapter. 6.1 State Expanded Toxicity Characteristics States have expanded the federal toxicity characteristic by: ! Adding constituents to the list of TC analytes; ! Establishing regulatory levels for TC analytes that are more stringent than federal levels; ! Specifying alternative tests for identifying toxic hazardous waste; and ! Using alternative approaches (other than listing constituents and regulatory levels) to identify toxic hazardous wastes. California, Michigan, and Washington have added constituents to the list of TC analytes, as shown in Exhibit 6-1. Both California and Michigan have added zinc, and both California and Washington have added PCBs. Other additional constituents include certain metals, pesticides, dioxins, and potential carcinogens. An example of a state regulatory level that is lower that the federal TC level is California's regulatory level of 1.7 mg/l for pentachlorophenol (versus 100 mg/l under the federal TC). As discussed in Section 3.6, California requires use of the Wet Extraction Test (WET) in addition to the TCLP. Use of the WET test identifies several metal-containing wastes as hazardous that are generally not identified as hazardous using the TCLP. These wastes include spent catalysts from the petroleum refining and food industries and metal dusts, metal sludges, and baghouse wastes from industries including fabricated metals, leather and apparel, electric and electronic products, primary metals, motor vehicles, transportation equipment, chemicals and allied products, and others. Both California and Washington have established toxicity criteria for wastes based on acute oral LD50, acute dermal LD50, acute inhalation LC50, and acute aquatic 96-hour LC50 (see Exhibit 6-2). A waste is designated hazardous if a representative sample of the waste meets any of the acute toxicity criteria. For example, Washington specifies rat and fish (for acute aquatic toxicity) bioassay tests in a State test methods manual. Generators must either test a representative sample of the waste or use their knowledge of waste constituents and the literature regarding toxicity of those constituents to determine if the waste meets any of the acute toxicity criteria. Finally, California's regulations state that a waste exhibits the characteristic of toxicity if the waste, based on representative samples, "has shown through experience or testing to pose a hazard to human health or environment because of its carcinogenicity, acute toxicity, chronic toxicity, bioaccumulative properties or persistence in the environment" (22 CCR 66261.24(a)(8)). This broad provision tends to shift the burden of identifying toxic wastes to the generator, because in the absence of specific state criteria (e.g., constituents and regulatory levels) the generator is responsible for being aware of experience or tests that show a waste poses a hazard. 6.2 State Only Listings In addition to expanded characteristics, some states have listed state only hazardous wastes. The most common state-only listed wastes are PCBs and waste oil. At least four states include additional "F" Wastes; three include additional "K" wastes; five include additional "P" wastes; and six include additional "U" wastes. Examples of state listed wastes include but are not limited to the following: ! In California, wastes containing any of almost 800 listed materials are presumed hazardous, unless proven through testing not to exhibit any of California's criteria for identifying hazardous waste. ! Maine has listed certain wastes from the production of linuron and bromacil, and has listed proposed additions to the federal list of hazardous wastes. ! Maryland has listed 9 specific chemical warfare agents. ! Michigan has added certain chemical production wastes to its "K" or specific source list, and has listed many state-only "U" wastes including organics, inorganics in particle form, pharmaceuticals (e.g., phenobarbital), chemical warfare agents, and herbicides. ! New Hampshire has added a number of wastes to its "F" or non-specific source list, including certain wastes from industrial painting operations and from metals recovery operations. ! Oregon has listed certain pesticide residues and certain blister agents and nerve gas. 6.3 State Restrictions on Exemptions Another way that states have expanded the universe of wastes they regulate as hazardous is by choosing not to adopt exemptions in the federal regulations. Examples include but are not limited to the following: ! Colorado does not recognize exemptions for certain injected groundwater that exhibits the TC and is reinjected pursuant to free phase hydrocarbon recovery operations at petroleum facilities (40 CFR 261.4(b)(11)), certain used chlorofluorocarbon (CFC) refrigerants that are reclaimed for further use (40 CFR 261.4(b)(12)), or non-terne plated used oil filters (40 CFR 261.4(b)(13)). ! Connecticut, New Hampshire, Oregon, and Washington do not include exemptions for certain chromium-bearing wastes from leather tanning and finishing (40 CFR 261.4(b)(6)(ii)). ! Maine does not recognize exemptions at 40 CFR 261.4(b)(6) through (13). These include: -- TC chromium wastes where chromium in the waste is nearly exclusively trivalent chromium; -- certain chromium-bearing wastes from leather tanning and finishing; -- specified mining and mineral processing wastes; -- cement kiln dust; -- certain arsenical-treated wood wastes; -- petroleum contaminated media and debris that fail the TC; -- certain injected groundwater; -- used CFC refrigerants; and -- non-terne plated used oil filters. ! Massachusetts, New York, and North Dakota do not recognize exemptions at 40 CFR 261.4(b)(10) through (13). (These wastes include the last four wastes named directly above.) 6.4 Summary Some states appear to be regulating a significant number of wastes as hazardous that are not covered under federal RCRA regulations. Moreover, a few states have taken different approaches to identifying characteristic hazardous wastes. In particular, California and Washington regulations go beyond constituent-by-constituent definitions and apply acute toxicity criteria to the whole waste. State expansions of hazardous waste identification regulations reflect state judgment about gaps in the federal program. State expansions have filled these gaps, but only in the specific states with such expansions. Such potential gaps apparently are not being filled in the remaining states that have not expanded the federal hazardous waste definitions. CHAPTER 6. STATE EXPANSIONS OF THE TOXICITY CHARACTERISTIC AND LISTINGS6-1 6.1 State Expanded Toxicity Characteristics. . . . . . .6-1 6.2 State Only Listings. . . . . . . . . . . . . . . . .6-2 6.3 State Restrictions on Exemptions . . . . . . . . . .6-5 6.4 Summary. . . . . . . . . . . . . . . . . . . . . . .6-6 Exhibit 6-1 State Toxicity Characteristics: Additional Constituents and More Stringent Regulatory Levels 6-3 Exhibit 6-2 State Toxicity Criteria Applied to Whole Waste (Representative Sample) 6-4 CHAPTER 7. SUMMARY OF POTENTIAL GAPS This chapter reviews the broad categories of potential gaps identified in the previous three chapters. Different ways of organizing the potential gaps are discussed, and a single comprehensive list of the potential gaps is presented. This review lays the groundwork for evaluating the significance of the potential gaps in the following three chapters. 7.1 Organization of the Analysis of Potential Gaps EPA has identified five categories of potential gaps in the hazardous waste characteristics using different approaches in each area: ! ICR Characteristics. EPA identified potential gaps associated with these characteristics by reviewing the original 1980 rulemaking record and comparing the ICR definitions and test methods to approaches taken to controlling similar hazards under other federal and state regulatory schemes. ! TC Characteristic. The Agency identified potential gaps associated with this characteristic by examining the properties of the TC analytes to determine how they could pose hazards to human health or the environment. ! Non-TC Chemicals. In contrast with the prior step, EPA began with a set of properties (including the potential to appear in non-hazardous industrial wastes) and then identified individual chemicals and groups of chemicals that could constitute potential gaps in the characteristics. ! Natural Resource Damages and Large-scale Environmental Problems. The Agency examined evidence of possible gaps using a hybrid approach that considered potential gap chemicals on the basis of their hazardous properties (e.g., endocrine disruption, stratospheric ozone depletion) and reviewed other potential gaps starting from possible risks to the environment, which, in turn, implied that certain waste constituents might be of concern. ! State Expansion of TC and State Listings. EPA reviewed how states have expanded their TC and listed as hazardous certain wastes that are not hazardous under the federal rules. These expansions reflect state judgments about gaps in the federal rules and thereby constitute potential gaps for this Scoping Study. The potential gaps presented in the following section are organized primarily by the major categories identified above. Where appropriate, these categories are subdivided into groups of chemicals posing similar types of hazards, and occasionally are subdivided even further by specific hazardous properties or exposure pathways of concern. Some of the potential gaps overlap. For example, endocrine disruptors appear among the concerns associated with the non-TC analytes as well as in a category by themselves under large-scale environmental risks. Although this overlap is inevitable, the potential gaps have been organized so as to minimize it, without omitting any potentially significant gaps. EPA considered other methods of classifying the potential gaps for purposes of further analysis. Gaps could be identified, for example, in terms of individual chemicals and their specific properties and hazards. Alternatively, the gaps could be organized around groups of chemicals with specific hazardous properties or types of risks. EPA rejected these approaches for purposes of this Scoping Study as impractical because too many individual chemicals or groups of chemicals, risks, and pathways are involved. In addition, defining potential gaps in categories that do not parallel the approaches used to identify such gaps would make it more difficult to appreciate the evidence and uncertainty associated with each potential gap. 7.2 Summary of Potential Gaps Exhibit 7-1 lists the potential gaps in the hazardous waste characteristics identified by EPA in the preceding chapters. The individual gaps are organized according to the section or chapter in which they are discussed, with reference to specific chemical classes, exposure pathways, or types of risks, as appropriate. Potent