United States
Environmental Protection
Agency
vxEPA Contaminant Occurrence Support
Document for Category 1
Contaminants for the Second Six-
Year Review of National Primary
Drinking Water Regulations
-------
Office of Water (4607M)
EPA-815-B-09-010
October 2009
www.epa.gov/safewater
-------
EPA-OGWDW Contaminant Occurrence Support Document for 2" 6-Year Review Category 1 Contaminants October 2009
Disclaimer
This report is in support of the preliminary revise/take no action decisions for EPA's
second Six-Year Review of Existing Drinking Water Standards Federal Register Notice. This
report is intended to provide technical background for the second Six-Year Review.
This document is not a regulation itself, and it does not substitute for the Safe Drinking
Water Act (SDWA) or EPA's regulations. Mention of trade names or commercial products does
not constitute endorsement or recommendation for use.
in
-------
-------
EPA-OGWDW Contaminant Occurrence Support Document for 2" 6-Year Review Category 1 Contaminants October 2009
Acknowledgements
EPA's Office of Ground Water and Drinking Water (OGWDW) conducted the national
assessments of regulated contaminant occurrence in public drinking water presented in this
report in support of the second Six-Year Review of National Primary Drinking Water
Regulations. This effort was directed by Ms. Karen Wirth and Mr. Rajiv Khera.
EPA would like to thank the many states, territories, and tribes and their staff who
contributed data sets and valuable advice during EPA's compilation and review of the
contaminant occurrence (compliance monitoring) data. Thank you also to the many public water
systems and their staff who conducted the monitoring that provided the extensive amount of
regulated contaminant occurrence data used in this report.
-------
-------
EPA-OGWDW Contaminant Occurrence Support Document for 2" 6-Year Review Category 1 Contaminants October 2009
Executive Summary
This document provides occurrence-related background information on 22 regulated
drinking water contaminants in support of the Environmental Protection Agency's (EPA's)
second Six-Year Review of National Primary Drinking Water Regulations (the "second Six-Year
Review"). These 22 contaminants are referred to as "Category 1 contaminants" based on their:
relatively high rates of occurrence at or near concentration thresholds of concern; new or
ongoing health effects assessments; and potential new laboratory analytical capabilities. The 22
Category 1 contaminants are: benzene, carbofuran, carbon tetrachloride, chlordane, 1,2-dibromo-
3-chloropropane, 1,2-dichloroethane, dichloromethane, 2,4-dichlorophenoxyacetic acid, 1,2-
dichloropropane, endothall, heptachlor, heptachlor epoxide, hexachlorobenzene,
hexachlorocyclopentadiene, oxamyl, tetrachloroethylene, toluene, toxaphene, 1,1,2-
trichloroethane, trichloroethylene, vinyl chloride, and xylenes (total).
This support document is divided into contaminant-specific chapters. The individual
contaminant chapters include background information (regulatory history, general chemical
information, and environmental fate and behavior), use and environmental release information,
an overview of occurrence in ambient water, and national estimates of contaminant occurrence in
public drinking water systems.
The national drinking water occurrence estimates presented here are based on the data
contained in the National Compliance Monitoring Information Collection Request Dataset for
the Second Six-Year Review (the "Six Year Review-ICR Dataset"), which includes data
collected from 1998 to 2005. This is the largest and most comprehensive drinking water
contaminant dataset ever compiled by EPA's Drinking Water Program and includes 15 million
records for 69 regulated contaminants submitted by 47 states/primacy agencies (45 states plus
Region 8 and Region 9 tribes). Using this dataset, a variety of occurrence estimates were
generated to characterize the national occurrence of regulated contaminants in public water
systems (PWSs).
National occurrence estimates using the Six Year Review-ICR Dataset were generated
using a two-stage analytical approach. In the first stage (Stage 1 analysis) the data were reviewed
and quality-checked to ensure high quality and were then analyzed to generate simple, non-
parametric estimates of contaminant occurrence. The Stage 1 analysis, based on maximum
sample concentration values, is inherently conservative; it is designed not to underestimate
occurrence and is therefore protective of public health. Simple counts are made of the number of
systems, and populations served by those systems, with at least one sample analytical result
(measured contaminant concentration) above a specified concentration threshold. The second
stage (Stage 2 analysis) is based on estimated system mean concentrations of contaminants.
Because it uses estimated mean concentrations rather than maximum sample concentrations,
Stage 2 analysis is less conservative than the Stage 1 analysis. The Stage 2 analysis also provides
occurrence assessments that may be more reflective of potential chronic exposure.
For the Stage 1 analyses, contaminant occurrence data were analyzed in several different
ways to more fully characterize national occurrence. One measure is the percentage of systems
with at least one sample detection of a contaminant. Using this measure, a wide range of
-------
EPA-OGWDW Contaminant Occurrence Support Document for 2" 6-Year Review Category 1 Contaminants October 2009
occurrence was found among the 22 Category 1 contaminants assessed. Percentages ranged from
a high of 7.592% of systems with at least one detection of xylenes to a low of 0.132% of systems
with at least one toxaphene detection. The five Category 1 contaminants with the highest rates of
occurrence nationally are xylenes (found in 7.592% of systems), toluene (5.764%),
dichloromethane (3.903%), tetrachloroethylene (2.502%), and trichloroethylene (2.009%). (Both
percentage and absolute system and populations-served numbers are presented in the report.)
These detections represent single detections and are not representative of Maximum
Contaminant Level (MCL) violations.
A related but different Stage 1 measure (a preliminary estimate of potential drinking
water exposure to a contaminant), is estimated as the percent of the population-served by
systems with at least one sample detection of a particular contaminant. Based on this measure,
the five contaminants with the highest rates of occurrence or potential exposure are xylenes
(17.480% of the population-served by systems had at least one detection), dichloromethane
(16.542%), tetrachloroethylene (14.076%), toluene (13.215%), and trichloroethylene (13.018%).
The size distribution of systems with detections influences the size of the potentially exposed
population. (For example, a small number of very large systems with contaminant detections will
result in a higher potentially exposed population than will a larger number of small systems with
detections). These detections represent single detections and are not representative of chronic or
long-term exposure.
In addition to characterizing the rates of contaminant occurrence, the Stage 1 analysis can
provide information regarding the degree of contaminant occurrence by comparing data to an
important health concentration threshold such as a contaminant's MCL. The percentage of
systems with at least one sample detection greater than a contaminant's MCL concentration
ranged from a high of 0.481% for tetrachloroethylene to a low of 0.000% for carbofuran,
hexachlorocyclopentadiene, and xylenes. The five Category 1 contaminants with the highest
Stage 1 occurrence relative to the MCL concentration are tetrachloroethylene (0.478%), 1,2-
dibromo-3-chloropropane (0.388%), trichloroethylene (0.387%), dichloromethane (0.293%), and
benzene (0.125%). These detections represent single detections greater than the MCL
concentration and do not always represent a MCL violation.
A preliminary estimate of drinking water exposure is given by the percentage of the
population-served by systems with at least one sample detection greater than an MCL
concentration. Using this measure, the five Category 1 contaminants with the highest
occurrence/potential exposure are tetrachloroethylene (6.448% of population), trichloroethylene
(5.264%), carbon tetrachloride (2.189%), l,2-dibromo-3-chloropropane (1.627%), and
dichloromethane (1.298%).
The Stage 2 analysis, based on estimated contaminant mean concentrations (rather than
the Stage 1 sample maximum concentration analyses), provides similar though not identical lists
of more frequently occurring contaminants. The percentages of systems with estimated
contaminant mean concentrations greater than the MCL concentrations ranged from a high of
0.112% for l,2-dibromo-3-chloropropane to a low of 0.000% for carbofuran, 2,4-
dichlorophenoxyacetic acid, endothall, hexachlorocyclopentadiene, 1,2-dichloropropane,
oxamyl, 1,1,2-trichloroethane, and xylenes. Based on this measure, the five Category 1
contaminants with the highest occurrence are l,2-dibromo-3-chloropropane (0.112%),
-------
EPA-OGWDW Contaminant Occurrence Support Document for 2" 6-Year Review Category 1 Contaminants October 2009
trichloroethylene (0.050%), tetrachloroethylene (0.046%), dichloromethane (0.026%), and
benzene (0.020%).
Similar to Stage 1 analysis, Stage 2 analysis allows for preliminary exposure assessments
based on the percentages of the population-served by systems with estimated contaminant mean
concentrations greater than the MCL concentration. Using this measure, the five Category 1
contaminants with the highest occurrence/exposure are tetrachloroethylene (0.277% of the
population), trichloroethylene (0.181%), toxaphene (0.014%), l,2-dibromo-3-chloropropane
(0.013%), and benzene (0.006%).
-------
-------
EPA-OGWDW Contaminant Occurrence Support Document for 2" 6-Year Review Category 1 Contaminants October 2009
Contents
Disclaimer iii
Acknowledgements v
Executive Summary vii
Contents xi
Abbreviations xiii
1 Introduction 1-1
1.1 Purpose and Scope 1-1
1.2 The Six-Year Review Process 1-2
1.3 Data Sources 1-3
1.3.1 Agency for Toxic Substances and Disease Registry 1-3
1.3.2 Extension Toxicology Network 1-4
1.3.3 United States Geological Survey Pesticide Use Maps 1-4
1.3.4 Toxic Substances Control Act Chemical Substance Inventory 1-5
1.3.5 Toxics Release Inventory 1-5
1.3.6 National Water Quality Assessment 1-6
1.3.7 National Compliance Monitoring Information Collection Request (ICR) Dataset
for the Second Six-Year Review (Six-Year Review-ICR Dataset) 1-8
1.3.8 Additional Data 1-11
1.4 Occurrence Analysis 1-15
1.5 References 1-17
2 Benzene 2-1
3 Carbofuran 3-1
4 Carbon Tetrachloride 4-1
5 Chlordane 5-1
6 l,2-Dibromo-3-chloropropane 6-1
7 1,2-Dichloroethane 7-1
8 Dichloromethane 8-1
9 2,4-Dichlorophenoxyacetic Acid 9-1
10 1,2-Dichloropropane 10-1
11 Endothall 11-1
12 Heptachlor 12-1
13 Heptachlor Epoxide 13-1
14 Hexachlorobenzene 14-1
15 Hexachlorocyclopentadiene 15-1
16 Oxamyl 16-1
17 Tetrachloroethylene 17-1
18 Toluene 18-1
19 Toxaphene 19-1
20 1,1,2-Trichloroethane 20-1
21 Trichloroethylene 21-1
22 Vinyl Chloride 22-1
23 Xylenes 23-1
XI
-------
-------
EPA-OGWDW Contaminant Occurrence Support Document for 2" 6-Year Review Category 1 Contaminants October 2009
Abbreviations
ATSDR Agency for Toxic Substances and Disease Registry
CDC Centers for Disease Control
CERCLA Comprehensive Environmental Response, Compensation, and Liability Act
CWS Community Water System
DBCP l,2-Dibromo-3-chloropropane
EPA Environmental Protection Agency (United States)
EPCRA Emergency Planning and Community Right-to-Know Act
EPHT Environmental Public Health Tracking Program
EXTOXNET Extension TOXicology NETwork
FDA Food and Drug Administration
IARC International Agency for Research on Cancer
ICR Information Collection Request
IUR Inventory Update Reporting
MCL Maximum Contaminant Level
MCLG Maximum Contaminant Level Goal
Mg milligram
mg/kg/day milligrams per kilogram per day
mg/L milligrams per liter
mg/m3 milligrams per cubic meter
MRL Minimum Reporting Level
NAWQA National Water Quality Assessment
NLM National Library of Medicine
NPDWR National Primary Drinking Water Regulation(s)
NTNCWS Non-Transient Non-Community Water System
NWIS National Water Information System
OGWDW Office of Ground Water and Drinking Water
PDF Pesticide Data Program
Ppb parts per billion
Xlll
-------
EPA-OGWDW Contaminant Occurrence Support Document for 2" 6-Year Review Category 1 Contaminants October 2009
Ppm parts per million
PVC Polyvinyl chloride
PWS Public Water System
RCRA Resource Conservation and Recovery Act
RfD Reference Dose
RSC Relative Source Contribution
SARA Superfund Amendments and Reauthorization Act
SOC Synthetic Organic Chemical
SRP Scientific Review Panel
TOXNET TOXicology Data NETwork
TRI Toxics Release Inventory
TSCA Toxic Substances Control Act
USEPA United States Environmental Protection Agency
USGS United States Geological Survey
VOC Volatile Organic Compound
jig micrograms
|ig/kg/day micrograms per kilogram per day
|ig/kg/year micrograms per kilogram per year
|ig/L microgram per liter
|ig/m3 micrograms per cubic meter
XIV
-------
EPA-OGWDW Contaminant Occurrence Support Document for 2" 6-Year Review Category 1 Contaminants October 2009
1 Introduction
1.1 Purpose and Scope
The Environmental Protection Agency (EPA) performed a detailed review of compliance
monitoring data of 69 National Primary Drinking Water Regulations (NPDWR) in support of the
second Six-Year Review. This document summarizes contaminant occurrence information for 22
of those regulated contaminants. These 22 contaminants represent the "Category 1"
contaminants which are those contaminants with relatively high levels of occurrence at or near
concentration thresholds of concern, new or ongoing health effects assessments, and new
potential laboratory methods capabilities. The remaining 47 were classified as "Category 2"
contaminants (contaminants that meet some, but not all, of the occurrence, health effects, or
laboratory methods criteria defining the Category 1 contaminants). The occurrence estimates
reported are based on the National Compliance Monitoring Information Collection Request
Dataset for the Second Six-Year Review (the "Six Year Review-ICR Dataset"). The Six Year
Review-ICR Dataset contains compliance monitoring data from 1998 to 2005. This dataset is the
largest and most comprehensive compliance monitoring dataset ever compiled by EPA's
Drinking Water Program, and includes 15 million records for 69 regulated contaminants
submitted by 47 states/primacy agencies (45 states plus Region 8 and Region 9 tribes). Using
this dataset, a two-stage analytical approach was employed to estimate a variety of occurrence
measures to characterize the national occurrence of regulated contaminants in public water
systems (PWSs) to support the second Six-Year Review process.
This support document is organized so that information on one contaminant is presented
in each chapter. This introductory Chapter 1 presents general background information applicable
to all contaminants. This chapter includes a description of the information and data sources used
for this assessment as well as the analytical approach used to estimate national contaminant
occurrence from the compliance monitoring data. To provide regulatory context, the Six-Year
Review process is also briefly described.
Chapters 2 through 23 present information specific to each of the Category 1 regulated
drinking water contaminants that EPA is reviewing under this second Six-Year Review. The
Table of Contents provides a list of the Category 1 contaminants included in the chapters in this
report. (The contaminant chapters are arranged alphabetically.)
Each chapter includes a brief presentation of the contaminant's uses, production,
environmental releases, and occurrence in ambient water. Detailed information is then provided
for a contaminant's occurrence in drinking water and related information is given regarding
preliminary exposure estimates (estimates of the populations-served by PWSs found to have
contaminants occurring in their drinking water samples). USEPA (2009a) provides similar
information for the Category 2 contaminants. For a complete description and presentation of all
estimates of contaminant occurrence generated in support of the second Six-Year Review (for
Category 1 and 2 contaminants), please refer to the accompanying report The Analysis of the
l-l
-------
EPA-OGWDW Contaminant Occurrence Support Document for 2" 6-Year Review Category 1 Contaminants October 2009
Regulated Contaminant Occurrence Data from Public Water Systems in Support of the Second
Six-Year Review of National Primary Drinking Water Regulations (USEPA, 2009b).
1.2 The Six-Year Review Process
EPA's Office of Ground Water and Drinking Water (OGWDW) is responsible for
implementing the provisions of the Safe Drinking Water Act (SOWA). The 1996 SDWA
amendments require EPA to review existing NPDWRs no less often than every six years and, if
appropriate, revise them. As long as an NPDWR revision maintains or provides for greater
protection of public health, the SDWA 1996 amendments give the EPA Administrator discretion
to determine if revision is appropriate. EPA believes any revision must continue to meet the
basic statutory requirements of the SDWA (e.g., generally setting the maximum contaminant
level (MCL) as close to the maximum contaminant level goal (MCLG) as is feasible). EPA also
believes any revision must present significant opportunities to improve the level of public health
protection and/or to achieve cost savings while maintaining or improving the level of public
health protection. The Six-Year Review process evaluates the contaminant occurrence, health
effects, analytical methodology, and treatment technology factors relevant to this formal re-
assessment of each NPDWR. This support document presents information relevant to
contaminant occurrence.
During the first Six-Year Review, EPA developed a protocol document — EPA Protocol
for the Review of Existing National Primary Drinking Water Regulations (USEPA, 2003 a) and
updated USEPA (2009c) for the current review cycle — to describe the process and strategy for
review that EPA will use to meet its statutory requirement. To most efficiently utilize limited
resources, EPA conducts a series of analyses at the beginning of each Six-Year Review cycle,
intended to target those NPDWRs that are the most appropriate candidates for revision. EPA
uses available, scientifically sound data to make decisions regarding whether or not to revise a
regulation. EPA reviews the following key information to make decisions regarding regulatory
changes: current health risk assessments, technology assessments (including reviews of
laboratory analytical methods and treatment techniques), and occurrence and exposure
assessments.
EPA will consider regulatory revisions based on the various components of each primary
drinking water regulation, including possible changes to MCLs, MCLGs, treatment techniques,
analytical method and treatment capabilities. In some cases, EPA may also consider revisions to
monitoring or system reporting requirements as part of the Six-Year Review; however, in most
cases, these types of revisions will be considered through other vehicles. For any NPDWR that is
a potential candidate for revision based on its review, EPA will also take economic
considerations into account before making its "revise/no action" decision. Moreover, EPA will
apply basic risk management principles to determine whether these candidate regulations
warrant regulatory revision to ensure that any revision will present a meaningful opportunity to
improve the level of public health protection and/or present a meaningful opportunity for cost
savings while maintaining, or improving, the level of public health protection.
1-2
-------
EPA-OGWDW Contaminant Occurrence Support Document for 2" 6-Year Review Category 1 Contaminants October 2009
1.3 Data Sources
Several data sources were used in this report to provide information on contaminant use,
production, occurrence, and potential exposure. The primary data source used for contaminant
background and use information was the Agency for Toxic Substances and Disease Registry
(ATSDR). The Toxics Release Inventory (TRI) provided data on contaminant releases.
Production data were obtained from the Toxic Substances Control Act (TSCA) Chemical
Substance Inventory. Information on contaminant occurrence in ambient water was primarily
derived from the United States Geological Survey (USGS) National Water Quality Assessment
(NAWQA) Program. Drinking water contaminant occurrence analyses are generated using state
drinking water compliance monitoring contaminant occurrence data that EPA obtained from 45
states and 2 Primacy Entities (the "Six-Year Review-ICR Dataset"). This dataset is the largest
compliance monitoring data set compiled and analyzed to date by EPA's Drinking Water
Program. These primary data sources are described in detail in the subsequent sections of this
chapter. The information from and the analyses generated using these data sources are
summarized in the contaminant-specific chapters of this report.
1.3.1 Agency for Toxic Substances and Disease Registry
In 1980, Congress created the ATSDR to implement the health-related sections of laws
that protect the public from hazardous wastes and environmental spills of hazardous substances
(ATSDR, 2001). The Comprehensive Environmental Response, Compensation, and Liability Act
of 1980 (CERCLA), commonly known as the "Superfund" Act, provided the Congressional
mandate to remove or clean up abandoned and inactive hazardous waste sites and to provide
federal assistance in toxic emergencies. As the lead agency within the Public Health Service for
implementing the health-related provisions of CERCLA, ATSDR is charged under the
Superfund Act to assess the presence and nature of health hazards at specific Superfund sites, to
help prevent or reduce further exposure and the illnesses that result from such exposures, and to
expand the knowledge base about health effects from exposure to hazardous substances
(ATSDR, 2001).
In the 1984 amendments to the Resource Conservation and Recovery Act of 1976
(RCRA), which provides for the management of legitimate hazardous waste storage or
destruction facilities, ATSDR was authorized to conduct public health assessments at these sites,
when requested by EPA, states, or individuals. ATSDR was also authorized to assist EPA in
determining which substances should be regulated and the levels at which substances may pose a
threat to human health (ATSDR, 2001).
With the passage of the Superfund Amendments and Reauthorization Act of 1986
(SARA), ATSDR received additional responsibilities in environmental public health. This act
broadened ATSDR's responsibilities in the areas of public health assessments, establishment and
maintenance of toxicological databases, information dissemination, and medical education
(ATSDR, 2001).
1-3
-------
EPA-OGWDW Contaminant Occurrence Support Document for 2" 6-Year Review Category 1 Contaminants October 2009
ATSDR issues Toxicological Profiles for over 250 substances, including 18 of the 22
contaminants discussed in this report. These profiles contain exhaustive reports on the
substances' health effects, chemical and physical properties, use and production, potential for
human exposure, and analytical methods. Whenever available, ATSDR was used as a primary
source in this report for contaminant background and use information.
1.3.2 Extension Toxicology Network
The Extension TOXicology NETwork (EXTOXNET) is a cooperative effort of
University of California-Davis, Oregon State University, Michigan State University, Cornell
University, and the University of Idaho (EXTOXNET, 2008). Primary files are maintained and
archived at Oregon State University. The EXTOXNET InfoBase provides a variety of
information about pesticide toxicology and environmental chemistry. These include: discussions
of toxicological issues of concern; toxicology newsletters; other resources for toxicology
information; toxicology fact sheets; Pesticide Information Profiles; and Toxicology Information
Briefs. Pesticide Information Profiles were used to provide background information for some of
the Category 1 contaminants. Pesticide Information Profiles are not based on an exhaustive
literature search. The information does not in any way replace or supersede the information on
the pesticide product labeling or other regulatory requirements.
1.3.3 United States Geological Survey Pesticide Use Maps
The USGS has produced maps of pesticide use for 244 compounds used in United States
crop production, 4 of which are Category 1 contaminants discussed in this report.. The pesticide
use maps show the average annual pesticide use intensity expressed as average weight (in
pounds) of a pesticide applied to each square mile of agricultural land in a county. The maps
presented in this report were created by the USGS using state-level estimates of pesticide use
rates for individual crops that were compiled by the CropLife Foundation, Crop Protection
Research Institute based on information collected during 1999 through 2004, combined with
county-level data on harvested crop acreage obtained from the 2002 Census of Agriculture. The
maps do not represent a specific year, but rather show typical use patterns (i.e., average annual
pesticide use) over the five year period of 1999 through 2004. Use intensity rates are expressed
as the pounds applied per square mile of mapped agricultural land in a county (USGS, 2007).
USGS (2007) notes several key limitations of the data used to produce these maps. For
instance, the estimates of applied pesticides are averaged at the state-level, while the maps
extrapolate to the county-level by using county crop acreages from the Census of Agriculture.
Consequently, the maps do not truly represent the local variability of cropping and management
practices found within many states. Furthermore, the 2002 Census of Agriculture may not have
represented all crop acreage, nor included all types of pasture. The maps are not intended for
making local-scale estimates of pesticide use, such as estimates at the county-level (USGS,
2007).
1-4
-------
EPA-OGWDW Contaminant Occurrence Support Document for 2" 6-Year Review Category 1 Contaminants October 2009
1.3.4 Toxic Substances Control Act Chemical Substance Inventory
Information on chemicals manufactured in or imported into the United States and listed
on the TSCA Chemical Substance Inventory is collected periodically by EPA. This collection
commenced following the promulgation of the Inventory Update Reporting (IUR) regulation in
1986. The 1986 regulation required chemical manufacturers and importers to report to EPA
every four years the identity of, and basic manufacturing information for, chemical substances
manufactured and imported annually in quantities of 10,000 pounds or more (USEPA, 2003b).
Under the TSCA, companies that manufacture or import chemicals may be required to
report information about these chemicals including, the identity of the chemicals, the amounts
manufactured or processed, certain details about their manufacture, and other data. This
information is stored in EPA's TSCA Chemical Substance Inventory Update System and is used
by EPA and other government agencies to identify potential use and exposure scenarios so that
they may protect human health and the environment (USEPA, 2003b).
EPA requires that chemical manufacturers periodically update their information in the
TSCA Chemical Substance Inventory because the chemicals used in the United States may
change and the quantity manufactured and industrial, commercial, and consumer uses of a
chemical may change. In order to accomplish this update, EPA periodically holds an Inventory
Update Reporting cycle. Inventory data are currently available for years 1986, 1990, 1994, 1998,
and 2002. TSCA Inventory data are available for 15 of the 22 contaminants discussed in this
report (USEPA, 2003b).
1.3.5 Toxics Release Inventory
EPA established the TRI in 1987 in response to Section 313 of the Emergency Planning
and Community Right-to-Know Act (EPCRA). EPCRA section 313 requires facilities to report
to both EPA and the states annual information on toxic chemical releases from facilities that
meet reporting criteria. This reported information is maintained in a database accessible through
TRI Explorer (http://www.epa.gov/triexplorer). In 1990 Congress passed the Pollution
Prevention Act, which required that additional data on waste management and source reduction
activities be reported under TRI. The TRI database details not only the types and quantities of
toxic chemicals released to the air, water, and land by facilities, but also provides information on
the quantities of chemicals sent to other facilities for further management (USEPA, 2002 and
2003c). Today, TRI includes information on releases of nearly 650 chemicals (including 19 of
the 22 contaminants discussed in this report).
Although TRI can provide a general idea of release trends, it has limitations. For
example, not all facilities are required to report all releases. Facilities are required to report
releases if they manufacture or process more than 25,000 pounds or use more than 10,000
pounds of a substance. Reporting requirements have changed overtime (e.g., reporting
thresholds have decreased); this creates the potential for misleading data trends over time
(USEPA, 1996). TRI data are meant to reflect releases and should not be used to estimate
general public exposure to a chemical (USEPA, 2002).
1-5
-------
EPA-OGWDW Contaminant Occurrence Support Document for 2" 6-Year Review Category 1 Contaminants October 2009
1.3.6 National Water Quality Assessment
Lakes, rivers, and aquifers are the sources of most drinking water. The USGS NAWQA
program is a national-scale assessment of the occurrence of contaminants in surface and ground
water. Monitoring contaminant occurrence in these ambient waters provides information for the
potential for contaminants to adversely affect drinking water supplies and drinking water.
The USGS instituted the NAWQA program in 1992 to examine ambient water quality
status and trends in the United States. The NAWQA program is designed to apply nationally
consistent methods to provide a consistent basis for comparisons among study basins across the
country and over time. These occurrence assessments serve to facilitate interpretation of natural
and anthropogenic factors affecting national water quality. For more detailed information on the
NAWQA program design and implementation, please refer to Leahy and Thompson (1994) and
Hamilton et al. (2004).
The NAWQA program conducts monitoring and water quality assessments in significant
watersheds and aquifers referred to as "study units." The program's sampling approach is not
"statistically" designed (i.e., it does not involve random sampling), but it provides a
representative view of the nation's waters in its coverage and scope. Together, the 51 study units
monitored between 1992 and 2001 include the aquifers and watersheds that supply more than
60% of the nation's drinking water and water used for agriculture and industry. The NAWQA
program monitors the occurrence of chemicals such as pesticides, nutrients, volatile organic
compounds (VOCs), trace elements, and radionuclides, and the condition of aquatic habitats and
fish, insects, and algal communities (NRC, 2002; Hamilton etal, 2004). NAWQA has collected
data from over 6,400 surface water and 7,000 ground water sampling points.
Monitoring of study units occurs in stages. Between 1992 and 2001, approximately one-
third of the study units at a time were studied intensively for a period of three to five years,
alternating with a period of less intensive research and monitoring that lasted between five and
seven years. Thus, all participating study units rotated through intensive assessment in a ten-year
cycle (Leahy and Thompson, 1994). The first ten-year cycle was designated Cycle 1. Summary
reports are available for the 51 study units that underwent intensive monitoring in Cycle 1
(USGS, 2001). Cycle 2 monitoring is scheduled to proceed in 42 study units from 2002 to 2012
(Hamilton etal., 2004).
Pesticide National Synthesis Project
The NAWQA Pesticide National Synthesis Project is a national-scale assessment of the
occurrence and behavior of pesticides in streams and ground water of the United States and the
potential for pesticides to adversely affect drinking water supplies or aquatic ecosystems.
Results from the most recent Pesticide National Synthesis analysis, based on complete
Cycle 1 (1992-2001) data from NAWQA study units, are posted on the NAWQA Pesticide
National Synthesis website (Gilliom et al., 2007). Data for surface water and ground water are
1-6
-------
EPA-OGWDW Contaminant Occurrence Support Document for 2" 6-Year Review Category 1 Contaminants October 2009
presented separately, and results in each category are subdivided by land use category. Land use
categories include agricultural, urban, mixed (deeper aquifers of regional extent in the case of
ground water), and undeveloped. The National Synthesis analysis for pesticides is a first step
toward the USGS goals of describing the occurrence of pesticides in relation to different land use
and land management patterns, and developing a deeper understanding of the relationship
between spatial occurrence of contaminants and their fate, transport, persistence, and mobility
characteristics.
The surface water summary data presented in the Pesticide National Synthesis (Gilliom et
a/., 2007) only includes stream data. Sampling data from a single one-year period, generally the
year with the most complete data, were used to represent each stream site. Sites with fewer data
or significant gaps were excluded from the analysis. NAWQA stream sites were sampled
repeatedly throughout the year to capture and characterize seasonal and hydrologic variability. In
the National Synthesis analysis, the data were time-weighted to provide an estimate of the annual
frequency of detection and occurrence at a given concentration (Gilliom et a/., 2007). For
instance, the detection frequency can be thought of as the percent of the year in which detections
are found at a typical site in this land use category and the 95th percentile concentration can be
thought of as a concentration that is not exceeded for 95% of the year at a typical site in this
land-use category.
The Pesticide National Synthesis only analyzed ground water data from wells; data from
springs and agricultural tile drains were not included. The sampling regimen used for wells was
different than that for surface water. In the National Synthesis analysis (Gilliom et a/., 2007),
USGS uses a single sample to represent each well, generally the earliest sample with complete
data for the full suite of analytes.
EPA Summary Analysis of NAWQA Study Unit Monitoring Results
Whereas the NAWQA program often uses the most representative data for a site to
calculate summary statistics, EPA, with the cooperation of USGS, has performed a summary
analysis of all Cycle 1 water monitoring data from all study units (1992 to 2001) for many of the
regulated contaminants. EPA's analysis of the NAWQA data is analogous to the simple, straight-
forward Stage 1 analysis the agency performed on drinking water data from PWS compliance
monitoring. That is, all the occurrence data for a particular contaminant were compiled and
analyzed using non-parametric methods to yield simple summary statistics to characterize
contaminant occurrence.
All surface water data were included in the EPA summary analysis. For ground water, all
well data were used; however, as with the USGS National Pesticide Synthesis, data from springs
and drains were excluded. For each contaminant, EPA calculated detection frequencies simply as
the percentage of samples and the percentage of sites with at least one detection. (A detection is
an analytical result equal to or greater than the reporting limit.) EPA used USGS data without
any censoring or weighting. From samples with detections a number of descriptive statistics
were also calculated, including the minimum, median, 95th percentile, 99th percentile, and
maximum concentrations. Reporting limits varied over time during the NAWQA program. The
1-7
-------
EPA-OGWDW Contaminant Occurrence Support Document for 2" 6-Year Review Category 1 Contaminants October 2009
highest reporting limit used for each contaminant is presented with the results of the analysis.
Note that because reporting limits varied, the minimum concentration reported as a detection can
be lower than the highest reporting limit.
1.3.7 National Compliance Monitoring Information Collection Request (ICR) Datasetfor
the Second Six-Year Review (Six-Year Review-ICR Dataset)
Because there existed no national database of complete public drinking water compliance
monitoring data (all public drinking water contaminant sample results for analytical detections
and non-detections), EPA conducted a voluntary data call-in for states' compliance monitoring
records. These compliance monitoring data are the contaminant occurrence data collected under
the SDWA as part of the program to maximize compliance with federal public drinking water
standards. These data are collected by PWSs and reported to the states. EPA received
compliance monitoring data from a total of 51 states/primacy agencies: 46 states, American
Samoa, Washington, D.C., Region 8 tribes, Region 9 tribes, and the Navajo Nation.1 Through
extensive data management efforts, quality assurance evaluations, and communications with
state/primacy agency data management staff, EPA established the Six-Year Review-ICR
Dataset. This high quality, national contaminant occurrence database consists of data from 47
states/primacy agencies (see Exhibit 1-1 below). The data received from American Samoa,
Washington, D.C., and the Navajo Nation could not be included in the second Six-Year Review
data analyses because those data sets were in significantly different formats. The data from
Kansas were received after the data assessment and quality assurance measures were completed
and were not able to be included in this occurrence analysis.
1 Upon request, Regions 8 and 9 provided a crosswalk table that listed the state in which each tribal water system
was located. All analyses conducted at the state-level (e.g., maps illustrating the geographic distribution of
occurrence) present the tribal systems' data along with the state in which they are located.
1-8
-------
EPA-OGWDW Contaminant Occurrence Support Document for 2" 6-Year Review Category 1 Contaminants October 2009
Exhibit 1-1. States with Compliance Monitoring Data Included in the Six-Year
Review-ICR Dataset
D American Samoa
D District of Columbia
D Guam
D Northern Mariana Islands
D Puerto Rico
• Region S Tribes
D Region 9 Navajo Nation
• Region 9 Tribes
D Virgin Islands
^| Data used
J Data available but not used
| Data not available
The contaminant occurrence data from these 47 states/primacy agencies (45 states plus
Region 8 Tribes and Region 9 Tribes) comprise more than 15 million analytical records from
approximately 132,000 PWSs. Approximately 254 million people are served by these PWSs
nationally. The Six-Year Review-ICR Dataset is the largest, most comprehensive set of
compliance monitoring data ever used by EPA's Drinking Water Program.
Data quality, completeness, and representativeness are key considerations for the dataset.
Given the intended use of this national dataset as well as the size, scope, and variety of formats
of the constituent data sets received from the states, EPA conducted extensive data management
and quality assurance/quality control (QA/QC) assessments on the data to be included in the Six-
Year Review-ICR Dataset. For a review of the completeness and representativeness of the data
set, refer to USEPA (2009b). USEPA (2009b) also presents details of the tasks conducted for
data management (establishing consistent formats across the numerous state data sets, setting
uniform analytical results units, addressing non-numerical non-detection values, etc.) and data
quality checks (addressing incorrect analytical units, outlier values, etc.). Exhibit 1-2
summarizes the Six-Year Review-ICR data for each contaminant used to generate the occurrence
estimates presented in this report.
1-9
-------
EPA-OGWDW Contaminant Occurrence Support Document for 2" 6-Year Review Category 1 Contaminants October 2009
Exhibit 1-2. Overview of the Contaminant Occurrence Data in the Six-Year
Review-ICR Data Set for the Category 1 Contaminants
Contaminant
(MCL)
Number
of States
with Data
Total
Number
of
Records
Total
Number of
Systems
Total
Population-
Served by
Systems
MRL
Synthetic Organic Chemicals (SOCs)1
Carbofuran
(40 Mg/L)
Chlordane
(2 Mg/L)
2,4-Dichlorophenoxyacetic acid (2,4-D)
(70 Mg/L)
1 ,2-Dibromo-3-chloropropane
(0.2 Mg/L)
Endothall *
(100|jg/L)
Heptachlor
(0.4 Mg/L)
Heptachlor Epoxide
(0.2 Mg/L)
Hexachlorobenzene
(1 Mg/L)
Hexachlorocyclopentadiene
(50 Mg/L)
Oxamyl
(200 |jg/L)
Toxaphene
(3 Mg/L)
45
44
43
44
38
44
44
44
44
45
44
114,142
128,453
124,285
231,169
49,820
133,726
134,708
135,512
135,290
113,380
115,403
31,097
31,841
33,187
37,618
14,156
33,020
33,015
32,826
32,801
30,876
30,387
167,973,342
182,036,975
187,451,183
193,749,339
118,536,801
184,443,784
184,477,775
184,124,823
184,738,430
167,378,431
160,012,134
0.9 |jg/L
0.2 |jg/L
0.1 |jg/L
0.6 |jg/L
9|jg/L
0.04 |jg/L
0.02 |jg/L
0.1 |jg/L
0.1 |jg/L
2 Mg/L
1 Mg/L
Volatile Organic Compounds (VOCs)
Benzene
(5 Mg/L)
Carbon Tetrachloride
(5 Mg/L)
1,2-Dichloroethane
(5 Mg/L)
Dichloromethane
(5 Mg/L)
1 ,2-Dichloropropane
(5 Mg/L)
Tetrachloroethylene
(5 Mg/L)
Toluene
(1,000|jg/L)
1 , 1 ,2-Trichloroethane
(5 Mg/L)
Trichloroethylene
(5 Mg/L)
45
45
45
45
45
45
45
45
45
374,503
385,864
375,636
372,009
374,409
410,042
374,250
372,386
403,609
50,435
50,446
50,442
50,169
50,437
50,436
50,451
50,195
50,432
226,947,293
226,934,787
226,933,991
226,844,060
226,912,134
227,008,555
226,955,301
226,851,469
226,907,686
0.5 Mg/L
0.5 Mg/L
0.5 Mg/L
0.5 Mg/L
0.5 Mg/L
0.5 Mg/L
0.5 Mg/L
0.5 Mg/L
0.5 Mg/L
1-10
-------
EPA-OGWDW Contaminant Occurrence Support Document for 2" 6-Year Review Category 1 Contaminants October 2009
Contaminant
(MCL)
Vinyl Chloride
(2 ug/L)
Xylenes (Total)
(1 0,000 |jg/L)
Number
of States
with Data
45
44
Total
Number
of
Records
373,161
363,466
Total
Number of
Systems
50,411
47,698
Total
Population-
Served by
Systems
226,464,423
218,071,535
MRL
0.5 |jg/L
0.5 |jg/L
1. The lower number of systems sampling for SOCs, as compared to VOCs, relates to waivers in some states that
remove monitoring requirements for certain pesticides and herbicides at some water systems based on vulnerability
assessments or non-detection of these contaminants in three consecutive annual samples.
2. Statewide waivers are frequently granted for endothall.
1.3.8 Additional Data
In addition to the primary data sources described above, supplemental contaminant
occurrence information is also presented whenever available. This information generally
includes reviews of occurrence surveys and studies of contaminant occurrence in both ground
water and surface water drinking and non-drinking water sources. Furthermore, some of the
information is national and some regional in nature, providing additional contaminant
information that supplements the other contaminant occurrence, use, and production information
presented in this report. However, the detailed national estimates of contaminant occurrence in
drinking water, described in the following Section 1.4, were conducted only using the Six-Year
Review-ICR contaminant occurrence data.
Summary of Selected United States Geological Survey Data on Domestic Well Water Quality for
the Centers for Disease Control's National Environmental Public Health Tracking Program
USGS, in collaboration with the Centers for Disease Control and Prevention (CDC), has
released an online report on the occurrence of 11 priority water-quality constituents of possible
health concern in domestic wells located in 16 states across the United States. The states include
California, Connecticut, Florida, Massachusetts, Maryland, Maine, Missouri, New Hampshire,
New Jersey, New Mexico, New York, Oregon, Pennsylvania, Utah, Washington, and Wisconsin.
The overall purpose of the study is to demonstrate through a pilot effort how USGS water-
quality, water-use, and associated geospatial data can be integrated in the CDC Environmental
Public Health Tracking Program (EPHT) network.
The water quality of domestic (private) supplies is inconsistently regulated and generally
not well characterized. The USGS has two water-quality data sets in the National Water
Information System (NWIS) database that can be used to help define the water quality of
domestic-water supplies: (1) data from the NAWQA Program, and (2) USGS state data. Data
from domestic wells from the NAWQA Program were collected to meet one of the program's
objectives, which was to define the water quality of major aquifers in the United States. These
domestic wells were located primarily in rural areas. The NWIS database is a repository of
USGS water data collected for a variety of projects; consequently, project objectives and
analytical methods vary. This variability can bias statistical summaries of contaminant
l-ll
-------
EPA-OGWDW Contaminant Occurrence Support Document for 2" 6-Year Review Category 1 Contaminants October 2009
occurrence and concentrations; nevertheless, these data can be used to define the geographic
distribution of contaminants.
Eleven water-quality constituents were selected for the pilot study (primarily on the basis
of expected occurrence and potential human health impacts), including three Category 1
contaminants: carbon tetrachloride, tetrachloroethylene, and trichloroethylene. USGS samples
were collected using nationally consistent field and analytical methodologies.
Bartholomay et al. (2007) notes that major-aquifer studies have three important
characteristics: (1) samples were collected before any treatment to define the quality of water in
the aquifer; (2) sampled wells were spatially distributed and randomly selected among existing
wells within the aquifer so these should be representative of overall water quality; and (3) most
of the samples were collected from domestic wells (one well that provides water to one home) in
rural areas of the United States.
United States Department of Agriculture Pesticides Data Program
The USDA Pesticide Data Program (PDF) maintains a national pesticide residue
database. PDF was initiated in 1991 to collect data on pesticide residues in food with sampling
conducted on a statistically defensible representation of pesticide residuals in the United States
food supply (USDA, 2006a). Sampling and testing are conducted on fruits and vegetables, select
grains, milk, and (as of 2001) drinking water.
In 2001, the PDF water sampling pilot program was initiated at community water
systems (CWSs) in New York and California. These sites were selected to reflect two highly
populated regions with divergent climates and hydrogeological conditions. Sites reflected the
diversity of land uses within California and New York and included major metropolitan areas,
agricultural regions, and highly protected watersheds. Source waters for water systems
participating in PDF finished drinking water sampling were primarily surface waters (USDA,
2003).
For the 2002 and 2003 data collection efforts, PDF continued sampling CWSs in New
York and California but also added five new sites in Kansas, Colorado, and Texas. The sites in
Colorado, Kansas, and Texas were small, rural community water systems in regions where EPA
needed monitoring data. These sites served populations of fewer than 50,000 people, used
surface water, and were regions for which EPA had ancillary data (such as agricultural pesticide
usage) (USDA, 2004; USDA, 2005).
In 2004, PDF collected drinking water samples from CWSs in Michigan, North Carolina,
Ohio, Oregon, Pennsylvania, and Washington. Water samples were collected from both the raw
water intake and the finished drinking water after treatment. The untreated intake water and
treated water were collected as paired samples, taking into account each individual plant's
processing time (USDA, 2006a).
1-12
-------
EPA-OGWDW Contaminant Occurrence Support Document for 2" 6-Year Review Category 1 Contaminants October 2009
In 2005, drinking water samples were collected from CWSs in California, Florida,
Louisiana, Michigan, North Carolina, North Dakota, Ohio, Pennsylvania, and Washington. As in
2004, water samples were collected from both the raw water intake and the finished drinking
water after treatment. The untreated intake water and treated water were collected as paired
samples, taking into account each individual plant's processing time (USDA, 2006b).
For the 2004 and 2005 data collection efforts, site selection was conducted in
collaboration with EPA's Office of Pesticide Programs Environmental Fate and Effects Division.
All selected drinking water sites met the following criteria: (1) service to populations under
50,000; (2) use of surface water as the primary source of water; and (3) location in regions of
heavy agriculture where known amounts of pesticides were applied (USDA, 2006b).
Volatile Organic Compounds in the Nation's Ground Water and Drinking Water Supply Wells
The USGS conducted a national assessment of 55 VOCs in ground water in regionally
extensive aquifers or aquifer systems (Zogorski etal., 2006). The assessment provides a
foundation for subsequent decadal assessments of the USGS NAWQA Program to ascertain
long-term trends of VOC occurrence in these aquifers. The assessment of ground water included
analyses of about 3,500 water samples collected during 1985 to 2001 from various types of
wells, representing almost 100 different aquifer studies. Also included (and discussed separately
from aquifer studies) was an assessment of the occurrence of VOCs in samples collected from
drinking water supply wells, specifically domestic and public wells. The additional data
described in this report is for the VOC occurrence in samples from domestic and public wells
only.
Occurrence findings are compared between domestic and public wells to distinguish the
separate issues for these well types related to supply, environmental setting, and sources of
VOCs. For this purpose, the occurrence of 55 VOCs is based on analyses of samples collected at
the well head, and before any treatment or blending, from about 2,400 domestic wells and about
1,100 public wells. Findings from domestic well samples update earlier USGS studies and
provide improved national coverage of sampled wells. Findings for public well samples
constitute the most current understanding of the occurrence of a large number of VOCs in
untreated ground water used by systems across the United States. This assessment of public well
water complements compliance monitoring by water utilities that typically focus on drinking
water delivered to the public. For more details on this study, refer to Zogorski et al. (2006).
The VOC occurrence information used in this study came from three sources: (1) the
NAWQA Program that sampled aquifers throughout the United States from 1993 to 2002; (2)
local, state, and federal agencies that sampled aquifers in various parts of the United States from
1985 to 1997 (referred to as retrospective data); and (3) a collaborative study by several entities
including the USGS, that sampled ground waters used by community water systems (CWSs)
throughout the United States from 1999 to 2000. Various criteria were used to select wells for
inclusion in this assessment including: study type, number of wells in the study, intended
purpose of the sample, minimum number of compounds analyzed in each well sample, analytical
method, laboratory reporting levels, and minimum distance between wells. Additional
1-13
-------
EPA-OGWDW Contaminant Occurrence Support Document for 2" 6-Year Review Category 1 Contaminants October 2009
information on the sources of the data and the procedures used to select wells can be found in
Moran et al. (2006).
United States Geological Survey Pilot Monitoring Program
In 1999, a Pilot Monitoring Program (PMP) was initiated by USGS and EPA to provide
information on pesticide concentrations in drinking water. This study focused on small drinking
water supply reservoirs in areas with high pesticide use in order to test the sampling approach in
areas where pesticides are probably present (Blomquist etal., 2001). Sampling sites represent a
variety of geographic regions, as well as different cropping patterns. The ideal site candidates
were mostly small reservoirs located in high pesticide-use areas, with a high runoff potential.
Twelve water-supply reservoirs, considered vulnerable to pesticide contamination, were selected
from the list of candidates. These 12 sites were located in: California, Indiana, Louisiana,
Missouri, Ohio, Oklahoma, South Carolina, South Dakota, New York, North Carolina,
Pennsylvania, and Texas.
Samples were collected quarterly throughout the year and at weekly or biweekly intervals
following the primary pesticide-application periods. Water samples were collected from the raw-
water intake and from the finished drinking water tap prior to entering the distribution system. At
some sites, samples were also collected at the reservoir outflow.
Three different laboratory methods were used in the PMP - a USGS approved method,
and two provisional methods. (Note: Since the completion of the study, both provisional methods
have been approved by USGS without any changes to the procedures.)
First Six-Year Review of National Primary Drinking Water Regulations
Occurrence estimates for regulated contaminants were developed in support of EPA's
first Six-Year Review of NPDWRs. The first Six-Year Review data set consisted of data from 16
states that were assembled into a "national cross-section" that was qualitatively indicative
(though not statistically representative) of national occurrence. The occurrence estimates are
based on compliance monitoring data from 1993 to 1997 (USEPA, 2003d).
The state compliance monitoring occurrence data used in the first Six-Year Review
analysis were submitted by states for EPA review and study of the occurrence of regulated
contaminants in PWSs (USEPA, 1999). In the USEPA (1999) review, all 50 states were
evaluated through a methodology that included ranking of states' pollution potential, dividing
states into quartiles based on these rankings, and then selecting states that equally represent the
four pollution-potential quartiles. Another factor considered (when selecting states equally
across the four pollution-potential quartiles) was selection of states distributed geographically to
include a broad representation of climatic and hydrologic variability across the United States. In
this way, a subset, or cross-section, of states could be selected to reflect a national representation
of pollution potentials and climatic/hydrologic difference.
1-14
-------
EPA-OGWDW Contaminant Occurrence Support Document for 2" 6-Year Review Category 1 Contaminants October 2009
The 16-state national cross-section was selected because it provided a balanced national
cross-section of state occurrence data, was based only on high quality and adequately complete
state occurrence data sets, and that, in aggregate, is indicative of national contaminant
occurrence (USEPA, 2003d). The analyses are based on data from the group of 16 states,
including Alabama, California, Florida, Illinois, Indiana, Kentucky, Michigan, Montana,
Nebraska, New Jersey, New Mexico, Oregon, South Carolina, South Dakota, Texas, and
Vermont. All of these states are also in the Six-Year Review-ICR dataset.
1.4 Occurrence Analysis
A two-stage analytical approach was used to evaluate the Six-Year Review-ICR
contaminant occurrence data. The first stage of analysis provides a straightforward evaluation of
occurrence of all contaminants under consideration. The Stage 1 analysis consists of simple
counts and descriptive statistics of analytical occurrence data for each of the contaminants. These
conservative assessments provide a broad evaluation of contaminant occurrence. Occurrence
analyses for each contaminant are assessed for systems and population-served by systems. A
typical Stage 1 analysis is a simple count of the number (or percentage) of systems with at least
one analytical detection (at least one sample analytical result greater than or equal to the
minimum reporting level, MRL) of a specific contaminant, or at least one sample analytical
detection with a concentration greater than the MCL concentration. For a complete description of
the Stage 1 methodology, refer to USEPA (2009b). These findings for either the Stage 1 or Stage
2 analyses are not representative of MCL violations.
The basic Stage 1 analytical methodology is a conservative approach: occurrence
measures are based on simple counts of whether or not a system reported at least one sample
analytical detection greater than a specified concentration threshold.2 The approach implicitly
incorporates another conservative assumption that if a detection is found in a single entry
point/sample point in a system, then the entire population-served by the system is assumed to
potentially be exposed to the detected contaminant (even though other entry points/sampling
points at the system, and the populations-served by those entry/sample points, may exhibit no
contaminant occurrence). In reality, many systems obtain water from multiple water sources
(such as a mix of purchased and non-purchased water, ground water wells and surface water
intakes, etc.). In systems with multiple water sources or water intakes, contaminant occurrence in
one source or sample point does not necessarily mean occurrence in all sources or sample points
that distribute water to consumers. Because of this identified potential over-estimation of
populations-served by systems with contaminant detections using the system-level analysis,
additional Stage 1 analyses were conducted for sample points to provide added details of
contaminant occurrence.
2 These Stage 1 analyses are conservative in the sense that they are protective of human health (i.e., because they are
based on a single, maximum sample detection value rather than an average value for each system. The Stage 1
analyses are more likely to overestimate occurrence and potential risks to human health than underestimate them).
1-15
-------
EPA-OGWDW Contaminant Occurrence Support Document for 2" 6-Year Review Category 1 Contaminants October 2009
Partly based on the findings of the Stage 1 analysis, EPA selected a set of contaminants
for which additional parametric statistical estimations, the Stage 2 analyses, were conducted. The
Stage 2 analysis estimates national contaminant occurrence by generating estimated long-term
mean concentrations of a specific contaminant at systems nationally. These system means were
calculated using a simple arithmetic average of all detection and non-detection sample analytical
results for each public water system.
The analyses conducted here, though, are not equivalent to the analyses conducted to
estimate a potential MCL violation. Both these Stage 1 and Stage 2 analyses are based on every
analytical sample result from all years with data in the Six Year Review-ICR Dataset. The Stage
2 analysis (based on an estimated long-term average concentration) is similar to the calculation
conducted to determine an MCL violation, but the two calculations are different. For most
regulated drinking water contaminants, an MCL violation occurs when the concentration equal
to a contaminant's MCL is exceeded by the estimated annual average concentration based on a
limited number of consecutive quarterly compliance monitoring samples (typically four for
surface water systems and two for ground water systems). In contrast, a "Stage 2 MCL
exceedance" occurs when the concentration threshold equal to a particular contaminant's MCL is
exceeded by the estimated system average concentration based on multiple years (not two or four
consecutive quarters) of compliance monitoring samples. Calculations of MCL violations are not
conducted in this report.
A "detection" is when the presence of a contaminant is identified and quantified in a
water sample at or above the analytical method MRL.3 A "non-detection" is when the
contaminant has not been identified in a sample at or above the MRL. (Note that a "non-
detection" does not indicate zero concentration of the contaminant in the sample; it only
indicates that the contaminant is not present at a concentration at or above the MRL, but may be
present in the sample at any concentration ranging from slightly less than the MRL to zero.)
Sample non-detection records in the state data sets tend to be less uniform and less
complete than detections records. Non-detection records are typically reported with non-numeric
values (such as "ND"), are blank (missing), do not include the numeric value of the relevant
MRL, or report MRL values that are obviously wrong (e.g., higher than the MCL concentration).
In these cases, data management work was required to identify and replace any non-numeric,
missing, or incorrect non-detection data.
A convention was developed to replace any of these problematic non-detection results in
the records with the most common (modal) MRL value for the state in which the system was
located. A non-detection is indicated as "
-------
EPA-OGWDW Contaminant Occurrence Support Document for 2" 6-Year Review Category 1 Contaminants October 2009
The Stage 1 analyses are not affected by how non-detections are recorded. However, to
estimate mean contaminant concentration values for each system in the Stage 2 analyses, it was
necessary to substitute a numeric value for each reported non-detection result (each "
-------
EPA-OGWDW Contaminant Occurrence Support Document for 2" 6-Year Review Category 1 Contaminants October 2009
Bartholomay, R.C., J.M. Carter, S.L. Qi, P.J. Squillace, and G.L. Rowe. 2007. Summary of
selected U.S. Geological Survey data on domestic well water quality for the Centers for Disease
Control's National Environmental Public Health Tracking Program: U.S. Geological Survey
Scientific Investigations Report 2007-5213, 57 p. Available on the Internet at:
http://pubs.usgs.gov/sir/2007/5213/
Blomquist, J.D., J.M. Denis, J.L. Cowles, J.A. Hetrick, R.D. Jones, andN.B. Birchfield. 2001.
Pesticides in Selected Water-Supply Reservoirs and Finished Drinking Water, 1999-2000:
Summary of Results from a Pilot Monitoring Survey. U.S. Geological Survey Open-File Report
01-456. 65 pp. Available on the Internet at: http://md.water.usgs.gov/nawqa/OFR 01-456.pdf
Extension Toxicology Network (EXTOXNET). 2008. "EXTOXNET - Extension TOXicology
NETwork." Available on the Internet at: http://extoxnet.orst.edu/etn.txt.html.
Gilliom, R.J., J.E. Barbash, C.G. Crawford, P.A. Hamilton, J.D. Martin, N. Nakagaki, L.H.
Nowell, J.C. Scott, P.E. Stackelberg, G.P. Thelin, and D.M. Wolock. 2007. The Quality of Our
Nation's Waters - Pesticides in the Nation's Streams and Ground Water, 1992-2001. Appendix
7. Statistical summaries of water-quality data. U.S. Geological Survey Circular 1291, 172 p.
Available on the Internet at: http://water.usgs.gov/nawqa/pnsp/pubs/circl291/appendix7/.
Hamilton, P. A., T.L. Miller, and D.N. Myers. 2004. Water Quality in the Nation's Streams and
Aquifers: Overview of Selected Findings, 1991-2001. USGS Circular 1265. Available on the
Internet at: http://water.usgs.gov/pubs/circ/2004/1265/pdf/circularl265.pdf
Leahy, P.P. and T.H. Thompson. 1994. The National Water-Quality Assessment Program. U.S.
Geological Survey Open-File Report 94-70, 4 p. Available on the Internet at:
http://water.usgs.gov/nawqa/NAWQA.OFR94-70.html. Accessed October 25, 2004.
Moran, M.J., J.S. Zogorski, and B.L. Rowe. 2006. Approach to an Assessment of Volatile
Organic Compounds in the Nation's Ground Water and Drinking-Water Supply Wells: U.S.
Geological Survey Open-File Report OFR 2005-1452, 36 p. Available on the Internet at:
http://pubs.usgs.gov/of/2005/1452/
National Research Council (NRC). 2002. Opportunities to Improve the U.S. Geological Survey
National Water Quality Assessment Program. National Academy Press. 238 p. Available on the
Internet at: http://www.nap.edu/catalog/10267.html.
United States Department of Agriculture (USDA). 2003. Pesticide Data Program, Annual
Summary, Calendar Year 2001. Available on the Internet at:
http://www.ams.usda.gov/AMSvl .0/getfile?dDocName=PDP2001 Summary.
USDA. 2004. Pesticide Data Program, Annual Summary, Calendar Year 2002. Available on the
Internet at: http://www.ams.usda.gov/AMSvl.O/getfile?dDocName=PDP2002Summary.
1-18
-------
EPA-OGWDW Contaminant Occurrence Support Document for 2" 6-Year Review Category 1 Contaminants October 2009
USD A. 2005. Pesticide Data Program, Annual Summary, Calendar Year 2003. Available on the
Internet at: http://www.ams.usda.gov/AMSvl.0/getfile?dDocName=PDP2003Summary.
USDA. 2006a. Pesticide Data Program, Annual Summary, Calendar Year 2004. Available on
the Internet at: http://www.ams.usda.gov/AMSvl.O/getfile?dDocName=PDP2004Summary.
USDA. 2006b. Pesticide Data Program, Annual Summary, Calendar Year 2005. Available on
the Internet at: http://www.ams.usda.gov/AMSvl.O/getfile?dDocName=STELPRDC5049946.
United States Environmental Protection Agency (USEPA). 1996. 1996 Toxics Release
Inventory: Public Data Release Report. Chapter 3: Year-to-Year Comparison of Toxics Release
Inventory Data. Available on the Internet at: http://www.epa.gov/tri/tridata/tri96/pdr/chap3.pdf.
USEPA. 1999. A Review of Contaminant Occurrence in Public Water Systems. Office of Water.
EPA-816-R-99-006. 78pp.
USEPA. 2002. The Toxics Release Inventory (TRI) and Factors to Consider When Using TRI
Data. EPA-260-F-02-017. November 2002. Available on the Internet at:
http://www.epa.gov/triinter/2002 tri brochure.pdf
USEPA. 2003 a. EPA Protocol for the Review of Existing National Primary Drinking Water
Regulations. Office of Water. EPA-815-R-03-002. June 2003. Available on the Internet at:
http://www.epa.gov/safewater/standard/review/pdfs/support 6yr protocal final.pdf
USEPA. 2003b. Non-confidential Inventory Update Reporting (IUR) Production Volume
Information. Available on the Internet at: http://www.epa.gov/oppt/iur/tools/data/2002-vol.htm.
USEPA. 2003 c. How are the Toxics Release Inventory Data Used? —government, business,
academic and citizen uses. EPA-260-R-002-004. May 2003. Available on the Internet at:
http://www.epa.gov/tri/guide docs/pdf/2003/2003 datausepaper.pdf
USEPA. 2003d. Occurrence Estimation Methodology and Occurrence Findings for Six-Year
Review of National Primary Drinking Water Regulations. Office of Water. EPA-815-R-03-006.
June 2003.
USEPA. 2009a. Contaminant Occurrence Support Document for Category 2 Contaminants for
the Second Six-Year Review of National Primary Drinking Water Regulations. Office of Water.
EPA-815-B-09-011. October 2009.
USEPA. 2009b. The Analysis of Regulated Contaminant Occurrence Data from Public Water
Systems in Support of the Second Six-Year Review of National Primary Drinking Water
Regulations. Office of Water. EPA-815-B-09-006. October 2009.
USEPA. 2009c. EPA Protocol for the Second Review of Existing National Primary Drinking
Water Regulations (Updated). Office of Water. EPA-815-B-09-002. October 2009.
1-19
-------
EPA-OGWDW Contaminant Occurrence Support Document for 2" 6-Year Review Category 1 Contaminants October 2009
United States Geological Survey (USGS). 2001. "Summary publications from 51 NAWQA study
units sampled in 1991-2001." Available on the Internet at:
http://water.usgs.gov/nawqa/nawqasum/. Last updated May 19, 2004.
USGS. 2007. Pesticide National Synthesis Project, 2002 Pesticide Use Maps. Available on the
Internet at: http://water.usgs.gov/nawqa/pnsp/usage/maps/compound_listing.php?year=02.
Zogorski, J.S., J.M. Carter, T. Ivahnenko, W.W. Lapham, MJ. Moran, B.L. Rowe, P.J.
Squillace, and P.L. Toccalino. 2006. Volatile organic compounds in the nation's ground water
and drinking-water supply wells. U.S. Geological Survey Circular 1292, 101 p. Available on the
Internet at: http://pubs.er.usgs.gov/pubs/cir/cirl292.
1-20
-------
EPA-OGWDW Contaminant Occurrence Support Document for 2" 6-Year Review Category 1 Contaminants October 2009
2 Benzene
This chapter on benzene is part of a report that is organized so that each chapter presents
information on one specific contaminant. Chapter 1 of this report, Introduction, presents
background information on all sources of data used as well as the analytical approach used to
estimate contaminant occurrence in drinking water. This contaminant chapter includes
background information (such as regulatory history, general chemical information, and
environmental fate and behavior), use and environmental release information, and benzene
occurrence estimates in ambient water and drinking water. All drinking water occurrence
estimates were based on data from the National Compliance Monitoring Information Collection
Request (ICR) Dataset for the Second Six-Year Review (the "Six Year Review-ICR Dataset").
For detailed information regarding this dataset, data management, data quality assessments, and
the analytical approach used to generate all contaminant occurrence estimates presented in this
chapter, please refer to USEPA (2009).
2.1 Background
The United States Environmental Protection Agency (EPA) published the current
National Primary Drinking Water Regulations (NPDWR) for benzene on July 8, 1987 (52 FR
25690 (USEPA, 1987)). The NPDWR established a maximum contaminant level goal (MCLG)
of zero based on a cancer classification of A, known human carcinogen. The NPDWR also
established a maximum contaminant level (MCL) of 5 |ig/L based on analytical feasibility.
According to the Agency for Toxic Substances and Disease Registry (ATSDR, 2007),
benzene (CeHe) is a colorless sweet-smelling liquid that is highly flammable and slightly water
soluble. It can be smelled at 60 ppm in the air and tasted at 0.5 to 4.5 ppm in water. Benzene is
produced industrially from petroleum and ranks in the top 20 chemicals in terms of United States
production volume (ATSDR, 2007).
The main sources of benzene to the environment are industrial. Benzene is released to the
air through the burning of oil, coal, and motor gasoline. It may also leak from waste storage
operations and gasoline service stations. Gas emissions from forest fires and volcanoes are the
primary natural sources of benzene. Tobacco smoke is a primary indoor source (ATSDR, 2007).
Once in the air, benzene reacts with other chemicals and breaks down within a few days.
Releases of benzene to the soil and water may result from industrial discharges, disposal of
benzene-containing products, and leakage from underground gasoline storage tanks. Benzene's
solubility in water allows it to infiltrate into ground water. Once in the ground or water, benzene
breaks down slowly. Benzene does not build up in plants or animals (ATSDR, 2007).
Levels of benzene in outdoor air range from 0.00002 to 0.034 ppm. As a result, most
humans are exposed to at least a small amount of benzene daily. Inhalation of auto exhaust is a
typical mode of outdoor exposure. Indoors, the likelihood of exposure is also high because
benzene tends to volatilize from glues, paints, furniture, wax, and detergents. It is estimated that
smoking increases ones exposure to benzene by about tenfold, although those exposed to the
2-1
-------
EPA-OGWDW Contaminant Occurrence Support Document for 2" 6-Year Review Category 1 Contaminants October 2009
highest levels are typically industrial workers. It is possible for drinking water to become
contaminated with benzene, but for most of the population, airborne benzene presents a more
significant level of exposure (ATSDR, 2007).
2.2 Use, Production, and Releases
Benzene has a wide variety of uses. It is integral in the synthesis of other chemicals, such
as styrene, cumene, and cyclohexane. It is also used in the manufacture of rubbers, lubricants,
dyes, detergents, drugs, and pesticides (ATSDR, 2007).
The Toxic Substances Control Act (TSCA) Inventory contains confidential data reported
by industry as a partial update of the Chemical Substance Inventory Update System. The data are
reported under the Inventory Update Reporting regulation. Manufacturers and importers are
required to report company information and chemical information, including production volumes
for chemicals they manufactured or imported in excess of 10,000 pounds in the immediately
preceding fiscal year.
The TSCA Chemical Substance Inventory does not provide actual production volumes;
instead, it indicates at which of five production volume ranges a chemical was manufactured or
imported. The TSCA Inventory is updated every four years. Exhibit 2-1 presents the production
of benzene in the United States from 1986 to 2002. Note that the production of benzene in the
United States has consistently been above 1 billion pounds since 1986.
Exhibit 2-1: Production of Benzene in the United States
Inventory Update Year
1986 Range
1990 Range
1994 Range
1998 Range
2002 Range
Production Volume
> 1
> 1
> 1
> 1
> 1
billion pounds
billion pounds
billion pounds
billion pounds
billion pounds
Source: USEPA, 2003a
Benzene is listed as a Toxics Release Inventory (TRI) chemical. Exhibit 2-2a lists the
environmental releases of benzene from 1988 to 2006. (Exhibit 2-2b provides a graphical
presentation of the environmental releases.) Air emissions constituted most of the on- and off-site
releases (greater than 80%). Underground injection was the second largest pathway of
environmental release; underground injection releases fluctuated over the period of record,
falling to a minimum in 2001. Releases to land (e.g., spills or leaks within the boundaries of the
reporting facility) have decreased since 1998. Over the eight year period, releases of benzene
were reported by 49 states (all states but VT), as well as American Samoa, Guam, Puerto Rico,
2-2
-------
EPA-OGWDW Contaminant Occurrence Support Document for 2" 6-Year Review Category 1 Contaminants October 2009
the Northern Mariana Islands, and the Virgin Islands. All states/entities, except for American
Samoa, Nevada, and New Hampshire, reported benzene releases in all nine years.
Exhibit 2-2a: Table of Toxics Release Inventory (TRI) Data for Benzene
Year
1998
1999
2000
2001
2002
2003
2004
2005
2006
On-Site Releases (pounds)
Total Air
Emissions
7,878,844
7,901,923
7,180,688
5,885,935
6,047,311
6,322,790
5,966,169
5,609,884
5,279,355
Surface Water
Discharges
18,878
27,829
13,852
20,867
19,560
18,496
16,051
11,426
7,384
Underground
Injection
570,781
840,628
620,190
217,828
544,053
357,108
386,819
294,416
262,374
Releases to
Land
240,289
144,675
51,034
77,841
44,909
33,339
7,525
17,005
18,521
Off -site
Releases
(pounds)
255,906
275,305
65,788
150,813
115,542
115,583
140,329
162,553
452,719
Total On- &
Off-site
Releases
(pounds)
8,964,698
9,190,360
7,931,552
6,353,284
6,771,374
6,847,316
6,516,892
6,095,284
6,020,353
Source: USEPA, 2008
Exhibit 2-2b: Graph of Toxics Release Inventory (TRI) Data for Benzene
Pounds Released
Environmental Releases of Benzene
3 nnn nnn
7 nnn nnn
; nnn nnn
; nnn nnn
i nnn nnn
3 nnn nnn
•> nnn nnn
nnn nnn
JTi-i
PL
n_
_n
i — i
n .-,
n 1—1
n_T
1998 1999 2000 2001 2002 2003 2004 2005 2006
Year
•Air Emissions D Surface Water Discharges D Underground Injection DReleases to Land • Off-site Releases
Source: USEPA, 2008
2-3
-------
EPA-OGWDW Contaminant Occurrence Support Document for 2" 6-Year Review Category 1 Contaminants October 2009
2.3 Occurrence in Ambient Water
Lakes, rivers, and aquifers are the sources of most drinking water. The United States
Geological Survey (USGS) National Water Quality Assessment (NAWQA) program is a
national-scale assessment of the occurrence of contaminants in ambient surface and ground
water. Contaminant occurrence in ambient water provides information on the potential for
contaminants to adversely affect drinking water supplies. Data on the occurrence of benzene in
ambient water are available from the NAWQA program.
2.3.1 EPA Summary Analysis ofNA WQA Data
EPA, with the cooperation of the USGS, performed a summary analysis of all Cycle 1
(1992-2001) water monitoring data from all study units for benzene. Detection frequencies were
computed as the percentage of samples and sites with detections (i.e., with at least one result
equal to or greater than the reporting limit). Sample detections can be biased by frequent
sampling in areas with high (or low) occurrence. Calculating the percentage of sites with
detections can reduce this bias.
The results of the EPA analysis are presented in Exhibit 2-3. Overall, benzene was
detected in 11.8% of samples and at 6.4% of sites. Benzene was detected more frequently in
surface water than in ground water. The median concentration, based on detections from all sites,
was 0.0235 |ig/L. The 99th percentile concentration, based on detections from all sites, was 89
Exhibit 2-3: EPA Summary Analysis of Benzene Data from NAWQA Study Units,
1992-2001
ground
water
surface
water
all
sites
Detection Frequency
(detections are results > RL1)
Number
of
Samples
4,652
1,402
6,054
% Samples
with
Detections
5.1%
34.3%
11.8%
Number
of Sites
4,152
182
4,334
% Sites
with
Detections
4.8%
42.9%
6.4%
Concentration Values
(of detections, in ug/L)
Minimum
0.004
0.0038
0.0038
Median
0.04
0.0211
0.0235
95th
Percen-
tile
42
0.714
13.6
99th
Percen-
tile
290
16.5
89
Maximum
1,900
24
1,900
1. RLs (Reporting Limits) for benzene varied, but did not exceed 0.004 ug/L.
2.4 Occurrence in Drinking Water
Benzene is regulated as a volatile organic compound (VOC) in drinking water. All non-
purchased community water systems (CWSs) and non-transient non-community water systems
(NTNCWSs) are required to sample for VOCs. The maximum waiver period for VOCs is six
years for ground water systems and three years for surface water systems.
2-4
-------
EPA-OGWDW Contaminant Occurrence Support Document for 2" 6-Year Review Category 1 Contaminants October 2009
All CWSs and NTNCWSs must collect four consecutive quarterly samples during the
initial three-year compliance period.l If all four samples are non-detections, then the system may
reduce to annual sampling. After three annual samples without a detection, a system may be
granted a waiver. If a compound is detected (but at less than the MCL), the system must take one
sample per quarter until results are below the MCL (minimum of two quarterly samples for
ground water systems and four quarterly samples for surface water systems). If all quarterly
samples are below the MCL, the system may return to annual sampling. If a compound is
detected at a level greater than the MCL, the system (whether ground water or surface water)
must take four consecutive quarterly samples until all are below the MCL. If all quarterly
samples are below the MCL, the system may return to annual sampling.
The analysis of benzene occurrence presented in the following section is based on state
compliance monitoring data from the Six-Year Review-ICR Dataset. This is the largest and most
comprehensive compliance monitoring data set compiled by EPA's Drinking Water program to
date.
All of the Six-Year Review-ICR states provided occurrence data for benzene. These data
consist of 374,503 analytical results from 50,435 public water systems (PWSs) during the period
from 1998 to 2005. The number of sample results and systems vary by state, although the state
data sets have been reviewed and checked to ensure adequacy of coverage and completeness. The
national modal minimum reporting level (MRL) for benzene in the dataset is 0.5 |ig/L.
2.4.1 Stage 1A nalysis of Occurrence Data
Stage 1 analyses have been performed at the sample, sample point, and system levels.
Note that some systems have multiple sample points, such as source water sample points or entry
points to the distribution system. Exhibit 2-4 presents the number of samples, systems, and
sample points in the data set, as well as the population-served by the systems and sample points,
and the percentage of detections and the percentage of the population-served by the systems and
sample points with detections. Detections in systems provide a fundamental indication of
national occurrence, but are not representative of MCL violations. For benzene, 0.884% of
50,435 systems reported detections. Exhibit 2-5 lists the minimum, median, 90th percentile, and
maximum benzene concentrations based on detections from all systems in the Six-Year Review-
ICR Dataset. The median concentration of benzene, based on detections from all systems, was
1.1
1 All new systems or systems using a new water source that began operation after January 22, 2004 must demonstrate
compliance with the MCL within a period of time specified by the State. The system must also comply with the
initial sampling frequencies specified by the State to ensure that a system can demonstrate compliance with the MCL.
2-5
-------
EPA-OGWDW Contaminant Occurrence Support Document for 2" 6-Year Review Category 1 Contaminants October 2009
Exhibit 2-4: Benzene Occurrence Data from the Six-Year Review-ICR Dataset
Source Water Type
Number of
Samples
Number of
Systems
Number of
Sample Points
Pop. Served
by Systems
Pop. Served
by Sample
Points
Total Number
Ground Water
289,855
46,177
80,534
86,395,538
86,395,538
Surface Water
84,648
4,258
10,868
140,551,755
140,551,755
Combined Ground &
Surface Water
374,503
50,435
91,402
226,947,293
226,947,293
Percent with Detections
Ground Water
0.523%
0.836%
0.577%
2.382%
0.901%
Surface Water
0.146%
1 .409%
0.690%
7.210%
1.598%
Combined Ground &
Surface Water
0.438%
0.884%
0.591%
5.372%
1.333%
Exhibit 2-5: Benzene Summary Statistics
Source Water Type
Ground Water
Detection Value (ug/L)
Minimum
0.03
Median
1.2
90tn
Percentile
8
Maximum
93
Surface Water
0.05
0.7
3.19
22
Combined Ground &
Surface Water
0.03
1.1
7.8
93
The following maps illustrate the geographic distributions of benzene occurrence in
drinking water. Exhibit 2-6 shows the states with systems with at least one benzene detection and
Exhibit 2-7 shows the states with systems with at least one detection greater than the MCL
concentration. Note that five states did not submit data for use in the Six-Year Review (Kansas2,
Louisiana, Mississippi, Pennsylvania, and Washington).
The distribution of systems with detections of benzene is geographically dispersed.
Detection rates were generally low, although only one state reported no systems with detections.
The remaining 44 states reported low benzene detection rates. Twenty-eight states reported
2 Kansas submitted data after the data management and analyses had been conducted; data from Kansas are not
included in the analyses.
2-6
-------
EPA-OGWDW Contaminant Occurrence Support Document for 2" 6-Year Review Category 1 Contaminants October 2009
detections greater than the MCL concentration of 5 |ig/L, although only one state reported more
than 0.5% of systems with detections greater than 5 |ig/L.
Exhibit 2-6: Percentage of Systems with at Least One Benzene Detection, by State
States Not Submitting Any Data
States with No Detections
Less than 0.5% of Systems Detecting
| Between 0.5% and 1 % of Systems Detecting
I More than 1% of Systems Detecting
Exhibit 2-7: Percentage of Systems with at Least One Benzene Detection Greater
than the MCL Concentration (> 5 |jg/L): by State
States Not Submitting Any Data
State* with No Detections > MCL
Le*i than 0.1% of System* with Detections > MCL
Between 0.1% and 0.5% of Systems with Detections > MCL
More than O.S% of Systems with Detections > MCL
2-7
-------
EPA-OGWDW Contaminant Occurrence Support Document for 2" 6-Year Review Category 1 Contaminants October 2009
Exhibit 2-8 presents the Stage 1 analysis of benzene occurrence in drinking water from
systems and the population-served by those systems in the Six-Year Review-ICR Dataset relative
to several thresholds. The thresholds are: 5 |ig/L, 2.5 |ig/L, 1 |ig/L, and 0.5 |ig/L. About 110
systems (0.218% of all systems), serving about 884,000 persons, reported at least one detection
greater than or equal to 2.5 |ig/L; and 63 systems (approximately 0.125% of all systems), serving
nearly 487,000 persons, reported at least one detection greater than 5 |ig/L.
Exhibit 2-8: Benzene Stage 1 Analysis - Systems and Population with at Least
One Threshold Exceedance
Source
Water Type
Ground
Water
Threshold
> 5 |jg/L
>2.5pg/L
> 1 pg/L
> 0.5 |jg/L
Number of
Systems
Exceeding
Threshold
57
100
206
283
Percent of
Systems
Exceeding
Threshold
0.123%
0.217%
0.446%
0.613%
Total Population-
Served by
Systems
Exceeding
Threshold
338,086
700,772
1,079,744
1,486,410
Percent of
Population -Served
by Systems
Exceeding
Threshold
0.391%
0.811%
1 .250%
1.720%
Surface
Water
> 5 |jg/L
>2.5pg/L
> 1 pg/L
> 0.5 |jg/L
6
10
22
39
0.141%
0.235%
0.517%
0.916%
148,852
183,093
4,038,337
7,433,993
0.106%
0.130%
2.873%
5.289%
Combined
Ground &
Surface
Water
> 5 |jg/L
>2.5pg/L
> 1 pg/L
> 0.5 pg/L
63
110
228
322
0.125%
0.218%
0.452%
0.638%
486,938
883,865
5,118,081
8,920,403
0.215%
0.389%
2.255%
3.931%
The occurrence estimates presented above (in Exhibit 2-8) are based on the conservative
assumption that if a benzene detection is found in a single entry point/sample point in a system,
then the entire population-served by the system is assumed to potentially be exposed to benzene
(even though other entry points/sampling points at the system, and the population-served by
those entry/sample points, may exhibit no occurrence of benzene). The sample point analysis is a
less conservative estimate of the population-served by a system with a detection of benzene
because it estimates the potentially exposed population-served based only on the proportion of
the population-served by entry points/sample points with detections of benzene (rather than the
entire population-served by the system).
Exhibit 2-9 presents the Stage 1 analysis of benzene occurrence in drinking water from
sample points and population-served by those sample points in the Six-Year Review-ICR
Dataset. A total of 131 (0.143% of) sample points, serving an estimated 221,000 persons,
reported at least one detection greater than or equal to 2.5 |ig/L; and 72 (0.079% of) sample
points, serving an estimated 121,000 persons, reported at least one detection greater than 5 |ig/L.
2-8
-------
EPA-OGWDW Contaminant Occurrence Support Document for 2" 6-Year Review Category 1 Contaminants October 2009
Exhibit 2-9: Benzene Stage 1 Analysis - Sample Points and Population with at
Least One Threshold Exceedance
Source
Water Type
Ground
Water
Threshold
> 5 pg/L
>2.5|jg/L
> 1 |jg/L
> 0.5 |jg/L
Number of
Sample Points
Exceeding
Threshold
66
118
249
347
Percent of
Sample Points
Exceeding
Threshold
0.082%
0.147%
0.309%
0.431%
Total Population-
Served by
Sample Points
Exceeding
Threshold
92,882
164,756
361,001
544,841
Percent of
Population -Served
by Sample Points
Exceeding
Threshold
0.108%
0.191%
0.418%
0.631%
Surface
Water
> 5 pg/L
>2.5|jg/L
> 1 |jg/L
> 0.5 |jg/L
6
13
28
48
0.055%
0.120%
0.258%
0.442%
27,765
56,532
687,435
913,601
0.020%
0.040%
0.489%
0.650%
Combined
Ground &
Surface
Water
> 5 pg/L
>2.5|jg/L
> 1 |jg/L
> 0.5 |jg/L
72
131
277
395
0.079%
0.143%
0.303%
0.432%
120,647
221,288
1,048,436
1,458,442
0.053%
0.098%
0.462%
0.643%
2.4.2 Stage 2 A nalysis of Occurrence Data
For the Stage 2 analyses, three different substitution values -zero, one-half the MRL
value, and the full MRL value- were used to replace each non-detection record. Three arithmetic
mean benzene concentrations were calculated at each system using the zero, one-half MRL and
full MRL substitution values. These mean calculations were done for all systems with data in the
Six-Year Review-ICR dataset. Then the percentages of all systems with a mean concentration
greater than the each threshold were calculated.
Exhibit 2-10 presents the Stage 2 analysis of estimated system mean concentrations for
benzene occurrence in drinking water relative to several thresholds. The thresholds are: 0.5 |ig/L,
1 |ig/L, 2.5 |ig/L, and 5 |ig/L. Exhibit 2-11 presents similar information based on population-
served by the systems. Based on the Stage 2 analyses, no more than 16 systems (approximately
0.032% of all systems) had an estimated system mean greater than or equal to 2.5 |ig/L. These
systems serve about 111,000 persons. Ten systems (0.020% of all systems), serving about 14,000
persons, had an estimated system mean greater than the MCL concentration of 5 |ig/L.
2-9
-------
EPA-OGWDW Contaminant Occurrence Support Document for 2" 6-Year Review Category 1 Contaminants October 2009
Exhibit 2-10: Benzene Stage 2 Analysis - Systems with a System Mean Threshold
Exceedance
Source
Water Type
Ground
Water
Threshold
> 5 |jg/L
>2.5|jg/L
> 1 |jg/L
> 0.5 |jg/L
Number of Systems with Mean
Exceeding Threshold1
Full MRL
9
15
67
264
1/2 MRL
9
13
55
114
Zero
9
13
49
90
Percent of Systems with Mean
Exceeding Threshold1
Full MRL
0.019%
0.032%
0.145%
0.572%
1/2 MRL
0.019%
0.028%
0.119%
0.247%
Zero
0.019%
0.028%
0.106%
0.195%
Surface
Water
> 5 |jg/L
>2.5|jg/L
> 1 |jg/L
> 0.5 |jg/L
1
1
3
37
1
1
3
9
1
1
3
5
0.023%
0.023%
0.070%
0.869%
0.023%
0.023%
0.070%
0.211%
0.023%
0.023%
0.070%
0.117%
Combined
Ground &
Surface
Water
> 5 |jg/L
>2.5|jg/L
> 1 |jg/L
> 0.5 |jg/L
10
16
70
301
10
14
58
123
10
14
52
95
0.020%
0.032%
0.139%
0.597%
0.020%
0.028%
0.115%
0.244%
0.020%
0.028%
0.103%
0.188%
1 The results shown in this table are based on system arithmetic means that were calculated using all sample
detection records and all non-detection records, with the non-detection records replaced (substituted) with
concentration values equal to the full MRL, 14 MRL, and zero.
Exhibit 2-11: Benzene Stage 2 Analysis - Population-Served by Systems with a
System Mean Threshold Exceedance
Source
Water Type
Ground
Water
Threshold
> 5 pg/L
>2.5|jg/L
> 1 H9/L
> 0.5 |jg/L
Population -Served by Systems
with Mean Exceeding Threshold1
Full MRL
14,152
110,807
169,570
1,364,591
1/2 MRL
14,152
109,663
148,699
398,783
Zero
14,152
109,663
147,729
282,886
Percent of Population-Served by
Systems with Mean Exceeding
Threshold1
Full MRL
0.016%
0.128%
0.196%
1.579%
1/2 MRL
0.016%
0.127%
0.172%
0.462%
Zero
0.016%
0.127%
0.171%
0.327%
Surface
Water
> 5 pg/L
>2.5|jg/L
> 1 H9/L
> 0.5 |jg/L
50
50
10,625
7,318,292
50
50
10,625
86,052
50
50
10,625
21,533
0.00004%
0.00004%
0.008%
5.207%
0.00004%
0.00004%
0.008%
0.061%
0.00004%
0.00004%
0.008%
0.015%
2-10
-------
EPA-OGWDW Contaminant Occurrence Support Document for 2" 6-Year Review Category 1 Contaminants October 2009
Source
Water Type
Combined
Ground &
Surface
Water
Threshold
> 5 |jg/L
>2.5|jg/L
> 1 H9/L
> 0.5 |jg/L
Population -Served by Systems
with Mean Exceeding Threshold1
Full MRL
14,202
110,857
180,195
8,682,883
1/2 MRL
14,202
109,713
159,324
484,835
Zero
14,202
109,713
158,354
304,419
Percent of Population-Served by
Systems with Mean Exceeding
Threshold1
Full MRL
0.006%
0.049%
0.079%
3.826%
1/2 MRL
0.006%
0.048%
0.070%
0.214%
Zero
0.006%
0.048%
0.070%
0.134%
1. The results shown in this table are based on system arithmetic means that were calculated using all sample
detection records and all non-detection records, with the non-detection records replaced (substituted) with
concentration values equal to the full MRL, 1A MRL, and zero.
2.5 Additional Occurrence Data
Volatile Organic Compounds in the Nation's Ground Water and Drinking water Supply Wells
The USGS conducted a national assessment of 55 VOCs, including benzene, in samples
collected from drinking water supply wells, specifically domestic and public wells (Zogorski et
a/., 2006). This occurrence assessment was based on analyses of samples collected at the well
head, and before any treatment or blending, from about 2,400 domestic wells and about 1,100
public wells.
In domestic wells, benzene had a detection frequency of 0.21% at an assessment level of
0.2 |ig/L (based on 2,401 samples). The median concentration of detections was 0.015 |ig/L. Two
detections were within one order of magnitude of the MCL concentration (of 5 |ig/L). No
detections were greater than the MCL concentration.
In public wells, benzene had a detection frequency of 0.46% at an assessment level of 0.2
|ig/L (based on 1,095 samples). The median concentration of detections was 0.13 |ig/L. Five
detections were within one order of magnitude of the MCL concentration. No detections were
greater than the MCL concentration.
Occurrence of Water-Quality Constituents in Domestic Wells across the United States
The USGS, in collaboration with the Centers for Disease Control and Prevention (CDC),
released an online report on the occurrence of 11 priority water-quality constituents of possible
health concern in domestic wells located in 16 states across the United States (Bartholomay et
a/., 2007). Two water-quality data sets were used to define the water quality of domestic-water
supplies: (1) data from the NAWQA Program, and (2) USGS state data. The common assessment
level used to quantify the detection frequency of benzene was 0.2 |ig/L.
A total of 1,036 wells were sampled for benzene. The detection frequency for benzene
was 0.1%. All of the detections of benzene had concentrations less than, but within one order of
magnitude, of the MCL concentration (5 |ig/L).
2-11
-------
EPA-OGWDW Contaminant Occurrence Support Document for 2" 6-Year Review Category 1 Contaminants October 2009
First Six-Year Review ofNPDWRs
Similar types of Stage 1 contaminant occurrence analyses were conducted for the first
Six-Year Review. The first Six-Year Review assessed regulated contaminant occurrence data
collected from 1993 to 1997. The first Six-Year Review data set consisted of data from 16 states
that were assembled into a "national cross-section" that was qualitatively indicative (though not
statistically representative) of national occurrence. For the first Six-Year Review (USEPA,
2003b), 305 systems (approximately 1.31% of about 23,000 systems), reported at least one
detection of benzene. Forty-four systems (approximately 0.19% of all systems), reported at least
one detection greater than the MCL concentration.
2.6 Summary of Data
TRI data indicate that total off-site and on-site releases of benzene for the period 1998-
2006 ranged from approximately 6 million pounds to approximately 9 million pounds, with the
greatest releases in 1998 and 1999. Air emissions constituted at least 80% of the total amount
released. Off-site releases were the second largest category.
Ambient water data from the NAWQA program indicate that benzene was detected in
11.8% of 6,054 ground and surface water samples and at 6.4% of 4,334 sites. The median
concentration, based on detections from all sites, was 0.024 |ig/L.
The Stage 1 analysis of occurrence in drinking water indicates that benzene was detected
in 0.884% of 50,435 surface water and ground water systems. The median concentration, based
on detections from all systems, was 1.1 |ig/L. In total, 0.125% of systems (serving 0.215% of the
population) reported benzene detections greater than the MCL concentration of 5 |ig/L.
Detections of benzene were generally more common in ground water systems than surface water
systems, and a greater population was served by ground water systems reporting at least one
detection greater than the MCL concentration. A similar pattern was seen for ground water and
surface water sample points.
In the Stage 2 analysis, approximately 0.020% of surface and ground water systems
(serving 0.006% of the population) had an estimated system mean concentration of benzene
greater than the MCL concentration. The number of systems with threshold exceedances was
greater for ground water system means and a larger population was served by ground water
systems with means exceeding all thresholds.
2.7 References
Agency for Toxic Substances and Disease Registry (ATSDR). 2007. Toxicological Profile for
Benzene. August 2007. Available on the Internet at:
http ://www. atsdr.cdc. gov/toxprofiles/tp3 .html.
2-12
-------
EPA-OGWDW Contaminant Occurrence Support Document for 2" 6-Year Review Category 1 Contaminants October 2009
Bartholomay, R.C., J.M. Carter, S.L. Qi, P.J. Squillace, and G.L. Rowe. 2007. Summary of
selected U.S. Geological Survey data on domestic well water quality for the Centers for Disease
Control's National Environmental Public Health Tracking Program: U.S. Geological Survey
Scientific Investigations Report 2007-5213, 57 p. Available on the Internet at:
http://pubs.usgs.gov/sir/2007/5213/.
United States Environmental Protection Agency (USEPA). 1987. National Primary Drinking
Water Regulations—Synthetic Organic Chemicals; Monitoring for Unregulated Contaminants;
Final Rule. Federal Register. Vol. 52, No. 130. p. 25690, July 8, 1987.
USEPA. 2003a. Non-confidential Inventory Update Reporting (IUR) Production Volume
Information - Benzene. Last modified September 10, 2007. Available on the Internet at:
http://www.epa.gov/oppt/iur/tools/data/2002-vol.htm. [Search for benzene.] Accessed February
21,2008.
USEPA. 2003b. Occurrence Estimation Methodology and Occurrence Findings for Six-Year
Review of National Primary Drinking Water Regulations. Office of Water. EPA-815-R-03-006.
June 2003.
USEPA. 2008. TRI Explorer: Trends - Benzene. Released February 21, 2008. Available on the
Internet at: http://www.epa.gov/triexplorer/trends.htm. [Search for benzene.] Accessed February
29, 2008.
USEPA. 2009. The Analysis of Regulated Contaminant Occurrence Data from Public Water
Systems in Support of the Second Six-Year Review of National Primary Drinking Water
Regulations. Office of Water. EPA-815-B-09-006. October 2009.
Zogorski, J.S., J.M. Carter, T. Ivahnenko, W.W. Lapham, M.J. Moran, B.L. Rowe, P.J. Squillace,
and P.L. Toccalino. 2006. Volatile organic compounds in the nation's ground water and drinking
water supply wells. U.S. Geological Survey Circular 1292, 101 p. Available on the Internet at:
http://pubs.er.usgs.gov/pubs/cir/cirl292.
2-13
-------
EPA-OGWDW Contaminant Occurrence Support Document for 2" 6-Year Review Category 1 Contaminants October 2009
3 Carbofuran
This chapter on carbofuran is part of a report that is organized so that each chapter
presents information on one specific contaminant. Chapter 1 of this report, Introduction, presents
background information on all sources of data used as well as the analytical approach used to
estimate contaminant occurrence in drinking water. This contaminant chapter includes
background information (such as regulatory history, general chemical information, and
environmental fate and behavior), use and environmental release information, and carbofuran
occurrence estimates in ambient water and drinking water. All drinking water occurrence
estimates were based on data from the National Compliance Monitoring Information Collection
Request (ICR) Dataset for the Second Six-Year Review (the "Six Year Review-ICR Dataset").
For detailed information regarding this dataset, data management, data quality assessments, and
the analytical approach used to generate all contaminant occurrence estimates presented in this
chapter, please refer to USEPA (2009a).
3.1 Background
The United States Environmental Protection Agency (EPA) published the current
National Primary Drinking Water Regulations (NPDWR) for carbofuran on January 30, 1991 (56
FR 3526 (USEPA, 1991)). The NPDWR established a maximum contaminant level goal
(MCLG) and a maximum contaminant level (MCL) of 40 |ig/L. EPA based the MCLG on a
reference dose (RfD) of 5 jig/kg-day (0.005 mg/kg-day) and a cancer classification of E, evidence
of non-carcinogenicity for humans.
According to USEPA (2006), carbofuran (C^HisNOs), classified as a restricted use
pesticide, was first registered for use in the United States in 1969. Carbofuran is available in
liquid and granular formulations. Sales of granular carbofuran have been limited to 2,500 pounds
per year in the United States since 1994, and it can only be used on spinach grown for seed, pine
seedlings, cucurbits, and bananas (USEPA, 2006). (Before 1991, 80% of the total usage of
carbofuran was in granular formulations. The restrictions on use of granular carbofuran were
established to protect birds and were not related to human health concerns (EXTOXNET, 1996).)
Carbofuran is highly mobile in soils and enters surface water via runoff from treated
fields and enters ground water through the leaching of treated crops (USEPA, 2006). Carbofuran
is stable under neutral and acidic conditions, but degrades in alkaline conditions. If carbofuran is
released to soil or water, it will be broken down by hydrolysis and photodegradation. Bacterial
degradation rates vary depending on environmental conditions. Carbofuran is not expected to
bioaccumulate (USEPA, 2006).
3.2 Use, Production, and Releases
Carbofuran is a broad spectrum insecticide used primarily to control soil and foliar pests
on a variety of field, vegetable, and fruit crops. It is used to control weevils, aphids, corn borers,
beetles, nematodes, and rootworm. There are no residential uses of carbofuran (USEPA, 2006).
3-1
-------
EPA-OGWDW Contaminant Occurrence Support Document for 2" 6-Year Review Category 1 Contaminants October 2009
The process to cancel the remaining carbofuran registrations is currently underway by EPA.
Under the Agency's May 15, 2009 final rule, EPA's revocation of carbofuran tolerances became
effective on December 31, 2009, due to considerable risks associated with this pesticide in food
and drinking water (USEPA, 2009b).
Exhibit 3-1, compiled by the United States Geological Survey (USGS, 2007), shows the
geographic distribution of estimated average annual carbofuran use in the United States from
1999 through 2004. A breakdown of use by crop is also included. The map was created by the
USGS using state-level data sets on pesticide use rates from 1999-2004 compiled by the
CropLife Foundation, at the Crop Protection Research Institute, combined with county-level data
on harvested crop acreage obtained from the 2002 Census of Agriculture. Due to the nature of the
data sources, non-agricultural uses are not reflected here, and variations in use at the county-level
are also not well represented (Thelin and Gianessi, 2000). The USGS (2007) estimates that
almost 1 million pounds of carbofuran active ingredient were used on average each year between
1999 and 2004. The greatest use of carbofuran is in the Midwest.
Exhibit 3-1: Estimated Annual Agricultural Use for Carbofuran (c. 2002)
CARBOFURAN - insecticide
2002 estimated annual agricultural use
Aueragg annual use of
acws ngredenl
{pounds p«f square mil* of agrteuhural
tend In con rty)
I no estimated use
1 i '.lid ID 2.009
D OJ01 to 0.034
D 0.03S to 0.133
D 0.134 to 0.48
• >= 0.481
Crops
com
potaJoes
COttOn
aargtun
grapes
chls peppera
cucuniara find pi dklB&
sweet com
sunflower seed
Total
ptxinc>fi applied
356471
245S52
20582Q
101711
35SK
14632
9077
6941
SOT3
4595
Percent
nathmBluea
35.70
24.6S
20.51
10.1S
3J56
1^8
Oil
0.70
0.50
OJ6
Source: USGS, 2007
The Toxic Substances Control Act (TSCA) Inventory contains confidential data reported
by industry as a partial update of the Chemical Substance Inventory Update System. The data are
reported under the Inventory Update Reporting regulation. Manufacturers and importers are
3-2
-------
EPA-OGWDW Contaminant Occurrence Support Document for 2" 6-Year Review Category 1 Contaminants October 2009
required to report company information and chemical information, including production volumes
for chemicals they manufactured or imported in excess of 10,000 pounds in the immediately
preceding fiscal year.
The TSCA Chemical Substance Inventory does not provide actual production volumes;
instead, it indicates at which of five production volume ranges a chemical was manufactured or
imported. The TSCA Inventory is updated every four years. Exhibit 3-2 presents the production
of carbofuran in the United States from 1986 to 2002. Note that the production of carbofuran in
the United States increased after 1986.
Exhibit 3-2: Production of Carbofuran in the United States
Inventory Update Year
1986 Range
1990 Range
1994 Range
1998 Range
2002 Range
Production Volume
10,000 - 500,000 pounds
> 1 million
> 1 million
> 1 million
> 1 million
-10 million pounds
-10 million pounds
-10 million pounds
-10 million pounds
Source: USEPA, 2003a
Carbofuran is listed as a Toxics Release Inventory (TRI) chemical. Exhibit 3-3a lists the
environmental releases of carbofuran in the United States from 1998 to 2006. (Exhibit 3-3b
provides a graphical presentation of the environmental releases.) Carbofuran air emissions
generally decreased over the years, but constituted nearly all of the releases, both on- and off-site.
Surface water discharges peaked in 1999, but were otherwise negligible. There were no
underground injections or releases to land for the period of record. Off-site releases of carbofuran
were also minimal. The TRI data for carbofuran were reported from 10 states (Arkansas, Florida,
Iowa, Illinois, Indiana, Kentucky, Louisiana, New York, Texas, and West Virgina), with only
Arkansas, Louisiana, New York, and West Virginia reporting every year.
Exhibit 3-3a: Table of Toxics Release Inventory (TRI) Data for Carbofuran
Year
1998
1999
2000
2001
On-Site Releases (pounds)
Total Air
Emissions
2,937
834
1,628
1,435
Surface Water
Discharges
1
112
5
5
Underground
Injection
0
0
0
0
Releases
to Land
0
0
0
0
Off-Site
Releases
(pounds)
2
23
0
0
Total On- &
Off-site
Releases
(pounds)
2,940
969
1,633
1,440
3-3
-------
EPA-OGWDW Contaminant Occurrence Support Document for 2" 6-Year Review Category 1 Contaminants October 2009
Year
2002
2003
2004
2005
2006
On-Site Releases (pounds)
Total Air
Emissions
703
729
506
550
455
Surface Water
Discharges
5
6
1
6
5
Underground
Injection
0
0
0
0
0
Releases
to Land
0
0
0
0
0
Off-Site
Releases
(pounds)
0
0
0
0
16
Total On- &
Off-site
Releases
(pounds)
708
735
507
556
476
Source: USEPA, 2008
Exhibit 3-3b: Graph of Toxics Release Inventory (TRI) Data for Carbofuran
Environmental Releases of Carbofuran
3 nnn
~% o nnn
(0
03
-------
EPA-OGWDW Contaminant Occurrence Support Document for 2" 6-Year Review Category 1 Contaminants October 2009
3.3.1 NA WQA Pesticide National Synthesis Project
The NAWQA Pesticide National Synthesis Project is a national-scale assessment of the
occurrence and behavior of pesticides in streams and ground water of the United States and the
potential for pesticides to adversely affect drinking water supplies or aquatic ecosystems. Under
the National Synthesis Project, the USGS monitored carbofuran between 1992 and 2001 in
streams and wells across the country (Gilliom etal., 2007). The maximum long-term method
detection limit for carbofuran was 0.010 |ig/L.
In NAWQA stream samples (Exhibit 3-4), carbofuran was found at frequencies ranging
from 1.59% of samples in urban areas to 2.07% in undeveloped areas, 3.99% in mixed land use
settings, and 10.37% of samples in agricultural settings. The 95th percentile concentrations were
less than the method detection limit in mixed, undeveloped, and urban settings, and 0.044 jig/L
in agricultural settings. The highest concentration, 7.00 |ig/L, was found at an agricultural site.
Exhibit 3-4: USGS National Synthesis Summary of NAWQA Monitoring of
Carbofuran in Streams, 1992-2001
Land Use
Type
Agricultural
Mixed
Undeveloped
Urban
No. of Samples
(No. of Sites)
2,003(83)
1,371 (65)
144(8)
811 (30)
Detection
Frequency
10.37%
3.99%
2.07%
1.59%
50tn Percentile
(Median)
Concentration
(in ug/L)
ND1
ND1
ND1
ND1
95th Percentile
Concentration
(in |jg/L)
0.044
ND1
ND1
ND1
Maximum
Concentration
(in ug/L)
7.000
1.690
0.220
0.055
Source: Gilliom et a/., 2007
1. ND = not detected (Concentration is less than the maximum long term method detection limit and is expected to
be less than any higher percentile concentration shown in the table)
In ground water NAWQA samples (Exhibit 3-5), carbofuran was found at frequencies
ranging from 0% of samples in undeveloped areas to 0.40% in mixed land use settings, 0.70% of
samples in urban areas, and 1.71% of samples in agricultural settings. The 95th percentile
concentrations were less than the method detection limit in all settings. The highest
concentration, 1.3 |ig/L, was found at an agricultural site.
Exhibit 3-5: USGS National Synthesis Summary of NAWQA Monitoring of
Carbofuran in Ground Water, 1992-2001
Land Use Type
Agricultural
Mixed
Number
of Wells
1,404
2,727
Detection
Frequency
1.71%
0.40%
50tn Percentile
(Median)
Concentration
(in ug/L)
ND1
ND1
95th Percentile
Concentration
(in ug/L)
ND1
ND1
Maximum
Concentration
(in ug/L)
1.3
0.218
3-5
-------
EPA-OGWDW Contaminant Occurrence Support Document for 2" 6-Year Review Category 1 Contaminants October 2009
Land Use Type
Undeveloped
Urban
Number
of Wells
34
853
Detection
Frequency
0%
0.70%
50tn Percentile
(Median)
Concentration
(in ug/L)
ND1
ND1
95th Percentile
Concentration
(in ug/L)
ND1
ND1
Maximum
Concentration
(in ug/L)
ND1
0.093
Source: Gilliom et a/., 2007
1. ND = not detected (Concentration is less than the maximum long term method detection limit and is expected to
be less than any higher percentile concentration shown in the table)
3.3.2 EPA Summary Analysis ofNA WQA Data
Whereas the NAWQA program often uses the most representative data for a site to
calculate summary statistics, EPA, with the cooperation of the USGS, performed a summary
analysis of all Cycle 1 (1992-2001) water monitoring data from all study units for carbofuran.
Detection frequencies were computed as the percentage of samples and sites with detections (i.e.,
with at least one result equal to or greater than the reporting limit). Sample detections can be
biased by frequent sampling in areas with high (or low) occurrence. Calculating the percentage of
sites with detections can reduce this bias.
The results of the EPA analysis are presented in Exhibit 3-6. Overall, carbofuran was
detected in 3.8% of samples and at 4.4% of sites. It was detected more frequently in surface
water than in ground water. The median concentration, based on detections from all sites, was
0.035 |ig/L. The 99th percentile concentration, based on detections from all sites, was 2.6 |ig/L.
Exhibit 3-6: EPA Summary Analysis of Carbofuran Data from NAWQA Study Units,
1992-2001
ground
water
surface
water
all
sites
Detection Frequency
(detections are results > RL1)
Number
of
Samples
10,113
20,417
30,530
% Samples
with
Detections
0.6%
5.3%
3.8%
Number
of Sites
5,234
1,908
7,142
% Sites
with
Detections
0.9%
14.2%
4.4%
Concentration Values
(of detections, in ug/L)
Minimum
0.0018
0.002
0.0018
Median
0.0351
0.0352
0.0352
95th
Percen-
tile
0.563
0.784
0.782
99th
Percen-
tile
1.3
2.63
2.6
Maximum
1.3
9.87
9.87
1. RLs (Reporting Limits) for carbofuran varied, but did not exceed 0.002 pg/L Note that because this EPA analysis
involves more data points than the USGS analyses presented above, a direct comparison is not possible.
3.4 Occurrence in Drinking Water
Carbofuran is regulated as a synthetic organic chemical (SOC) in drinking water. All non-
purchased community water systems (CWSs) and non-transient non-community water systems
3-6
-------
EPA-OGWDW Contaminant Occurrence Support Document for 2" 6-Year Review Category 1 Contaminants October 2009
(NTNCWSs) are required to sample for SOCs. The maximum waiver period for SOCs is three
years, but this waiver can be renewed indefinitely. (Systems may receive a waiver at any time
without having collected an SOC sample. With a waiver, systems are not required to collect an
SOC sample.)
All CWSs and NTNCWSs without an SOC waiver must collect four consecutive
quarterly samples during the initial three-year compliance period.1 If all four samples are non-
detections, then a system serving less than 3,300 persons may reduce to one sample during each
consecutive compliance period; a system serving more than 3,300 persons may reduce to two
quarterly samples within a 12-month period during a 3-year compliance period. If a chemical is
detected (but is less than the MCL), the system must take one sample per quarter until results are
below the MCL (minimum of two quarterly samples for ground water systems and four quarterly
samples for surface water systems). If all quarterly samples are below the MCL, the system may
return to annual sampling. If a chemical is detected at a level greater than the MCL, the system
(whether ground water or surface water) must take quarterly samples until four consecutive
quarters are below the MCL. If all quarterly samples are below the MCL, the system may return
to annual sampling.
The analysis of carbofuran occurrence presented in the following section is based on state
compliance monitoring data from the Six-Year Review-ICR Dataset. This is the largest and most
comprehensive compliance monitoring data set compiled by EPA's Drinking Water program to
date.
All of the Six-Year Review-ICR states provided occurrence data for carbofuran. These
data consist of 114,142 analytical results from 31,097 public water systems (PWSs) during the
period from 1998 to 2005. The number of sample results and systems vary by state, although the
state data sets have been reviewed and checked to ensure adequacy of coverage and
completeness. The national modal minimum reporting level (MRL) for carbofuran in the dataset
is 0.9 |ig/L.
3.4.1 Stage 1A nalysis of Occurrence Data
Stage 1 analyses have been performed at the sample, sample point, and system levels.
Note that some systems have multiple sample points, such as source water sample points or entry
points to the distribution system. Exhibit 3-7 presents the number of samples, systems, and
sample points in the data set, as well as the population-served by the systems and sample points,
and the percentage of detections and the percentage of the population-served by the systems and
sample points with detections. Detections in systems provide a fundamental indication of
national occurrence, but are not representative of MCL violations. For carbofuran, 0.145% of
31,097 systems reported detections. Exhibit 3-8 lists the minimum, median, 90th percentile, and
maximum carbofuran concentrations based on detections from all systems in the Six-Year
Review-ICR Dataset. The median concentration of carbofuran, based on detections from all
systems, was 0.54 |ig/L.
1 All new systems or systems using a new water source that began operation after January 22, 2004 must demonstrate
compliance with the MCL within a period of time specified by the State. The system must also comply with the
initial sampling frequencies specified by the State to ensure that a system can demonstrate compliance with the MCL.
3-7
-------
EPA-OGWDW Contaminant Occurrence Support Document for 2" 6-Year Review Category 1 Contaminants October 2009
Exhibit 3-7: Carbofuran Occurrence Data from the Six-Year Review-ICR Dataset
Source Water Type
Number of
Samples
Number of
Systems
Number of
Sample Points
Pop. Served
by Systems
Pop. Served
by Sample
Points
Total Number
Ground Water
94,395
28,289
45,392
61,793,025
61,793,025
Surface Water
19,747
2,808
6,020
106,180,317
106,180,317
Combined Ground &
Surface Water
114,142
31,097
51,412
167,973,342
167,973,342
Percent with Detections
Ground Water
0.044%
0.120%
0.082%
5.154%
5.074%
Surface Water
0.096%
0.392%
0.249%
2.246%
1.968%
Combined Ground &
Surface Water
0.053%
0.145%
0.101%
3.315%
3.111%
Exhibit 3-8: Carbofuran Summary Statistics
Source Water Type
Ground Water
Detection Value (ug/L)
Minimum
0.23
Median
0.54
90tn
Percentile
1
Maximum
3
Surface Water
0.5
0.9
2.5
8
Combined Ground &
Surface Water
0.23
0.54
1.05
8
The following maps illustrate the geographic distribution of Carbofuran occurrence in
drinking water. Exhibit 3-9 shows the states with systems with at least one detection and Exhibit
3-10 shows the states with systems with at least one detection greater than the MCL
concentration. Note that five states did not submit data for use in the Six-Year Review (Kansas2,
Louisiana, Mississippi, Pennsylvania, and Washington).
Kansas submitted data after the data management and analyses had been conducted; data from Kansas are not
included in the analyses.
3-8
-------
EPA-OGWDW Contaminant Occurrence Support Document for 2" 6-Year Review Category 1 Contaminants October 2009
The distribution of systems with detections of carbofuran is geographically dispersed.
Detection rates were generally low, with only three states having greater than 1% of their systems
detecting carbofuran. No states reported detections greater than the MCL concentration of 40
Exhibit 3-9: Percentage of Systems with at Least One Carbofuran Detection, by
State
States Not Submitting Any Data
States with No Detections
Loss lhan 0.5% of Sysloms Detecting
| Between 0.5% and 1% of Systems Detecting
j More than 1% of Systems Detecting
Exhibit 3-11 presents the Stage 1 analysis of carbofuran occurrence in drinking water
from systems and the population-served by those systems in the Six-Year Review-ICR Dataset
relative to several thresholds. The thresholds are: 0.9 |ig/L, 7 |ig/L, 20 |ig/L, and 40 |ig/L. No
systems reported detections greater than 7 |ig/L, 20 |ig/L, or 40 |ig/L.
Exhibit 3-11: Carbofuran Stage 1 Analysis - Systems and Population with at Least
One Threshold Exceedance
Source
Water Type
Ground
Water
Threshold
> 40 |jg/L
> 20 |jg/L
> 1 H9/L
> 0.9 |jg/L
Number of
Systems
Exceeding
Threshold
0
0
0
7
Percent of
Systems
Exceeding
Threshold
0.000%
0.000%
0.000%
0.025%
Total Population -
Served by
Systems
Exceeding
Threshold
0
0
0
71,762
Percent of
Population -Served
by Systems
Exceeding
Threshold
0.000%
0.000%
0.000%
0.116%
3-9
-------
EPA-OGWDW Contaminant Occurrence Support Document for 2" 6-Year Review Category 1 Contaminants October 2009
Source
Water Type
Surface
Water
Threshold
> 40 |jg/L
> 20 |jg/L
> 7 |jg/L
> 0.9 |jg/L
Number of
Systems
Exceeding
Threshold
0
0
1
2
Percent of
Systems
Exceeding
Threshold
0.000%
0.000%
0.036%
0.071%
Total Population -
Served by
Systems
Exceeding
Threshold
0
0
3,186
257,648
Percent of
Population -Served
by Systems
Exceeding
Threshold
0.000%
0.000%
0.003%
0.243%
Combined
Ground &
Surface
Water
> 40 |jg/L
> 20 |jg/L
> 7 |jg/L
> 0.9 |jg/L
0
0
1
9
0.000%
0.000%
0.003%
0.029%
0
0
3,186
329,410
0.000%
0.000%
0.002%
0.196%
The occurrence estimates presented above (in Exhibit 3-11) are based on the conservative
assumption that if a carbofuran detection is found in a single entry point/sample point in a
system, then the entire population-served by the system is assumed to potentially be exposed to
carbofuran (even though other entry points/sampling points at the system, and the population-
served by those entry/sample points, may exhibit no occurrence of carbofuran). The sample point
analysis is a less conservative estimate of the population-served by a system with a detection of
carbofuran because it estimates the potentially exposed population-served based only on the
proportion of the population-served by entry points/sample points with detections of carbofuran
(rather than the entire population-served by the system).
Exhibit 3-12 presents the Stage 1 analysis of carbofuran occurrence in drinking water
from sample points and the population-served by those sample points in the Six-Year Review-
ICR Dataset. No sample points reported detections greater than 20 jig/L or 40 |ig/L.
Exhibit 3-12: Carbofuran Stage 1 Analysis - Sample Points and Population with at
Least One Threshold Exceedance
Source
Water Type
Ground
Water
Threshold
> 40 |jg/L
> 20 |jg/L
> 7 |jg/L
> 0.9 |jg/L
Number of
Sample Points
Exceeding
Threshold
0
0
0
7
Percent of
Sample Points
Exceeding
Threshold
0.000%
0.000%
0.000%
0.015%
Total Population-
Served by Sample
Points Exceeding
Threshold
0
0
0
70,287
Percent of
Population-
Served by
Sample Points
Exceeding
Threshold
0.000%
0.000%
0.000%
0.114%
3-10
-------
EPA-OGWDW Contaminant Occurrence Support Document for 2" 6-Year Review Category 1 Contaminants October 2009
Source
Water Type
Surface
Water
Threshold
> 40 |jg/L
> 20 |jg/L
> 1 pg/L
> 0.9 |jg/L
Number of
Sample Points
Exceeding
Threshold
0
0
1
4
Percent of
Sample Points
Exceeding
Threshold
0.000%
0.000%
0.017%
0.066%
Total Population-
Served by Sample
Points Exceeding
Threshold
0
0
3,186
257,648
Percent of
Population-
Served by
Sample Points
Exceeding
Threshold
0.000%
0.000%
0.003%
0.243%
Combined
Ground &
Surface
Water
> 40 |jg/L
> 20 |jg/L
> 1 pg/L
> 0.9 |jg/L
0
0
1
11
0.000%
0.000%
0.002%
0.021%
0
0
3,186
327,935
0.000%
0.000%
0.002%
0.195%
3.4.2 Stage 2 A nalysis of Occurrence Data
For the Stage 2 analyses, three different substitution values -zero, one-half the MRL
value, and the full MRL value- were used to replace each non-detection record. Three arithmetic
mean carbofuran concentrations were calculated at each system using the zero, one-half MRL
and full MRL substitution values. These mean calculations were done for all systems with data in
the Six-Year Review-ICR dataset. Then the percentages of all systems with a mean concentration
greater than each threshold were calculated.
Exhibit 3-13 presents the Stage 2 analysis of estimated system mean concentrations for
carbofuran occurrence in drinking water relative to several thresholds. The thresholds are: 0.9
|ig/L, 7 |ig/L, 20 |ig/L, and 40 |ig/L. Exhibit 3-14 presents similar information based on
population-served by the systems. No systems had an estimated system mean exceeding 20 jig/L
or 40 |ig/L.
Exhibit 3-13: Carbofuran Stage 2 Analysis - Systems with a System Mean
Threshold Exceedance
Source
Water Type
Ground
Water
Threshold
> 40 |jg/L
> 20 |jg/L
> 1 pg/L
> 0.9 pg/L
Number of Systems with Mean
Exceeding Threshold1
Full MRL
0
0
0
30
1/2 MRL
0
0
0
1
Zero
0
0
0
0
Percent of Systems with Mean
Exceeding Threshold1
Full MRL
0.000%
0.000%
0.000%
0.106%
1/2 MRL
0.000%
0.000%
0.000%
0.004%
Zero
0.000%
0.000%
0.000%
0.000%
Surface
Water
> 40 pg/L
> 20 pg/L
> 1 pg/L
0
0
1
0
0
1
0
0
1
0.000%
0.000%
0.036%
0.000%
0.000%
0.036%
0.000%
0.000%
0.036%
3-11
-------
EPA-OGWDW Contaminant Occurrence Support Document for 2" 6-Year Review Category 1 Contaminants October 2009
Source
Water Type
Threshold
> 0.9 |jg/L
Number of Systems with Mean
Exceeding Threshold1
Full MRL
25
1/2 MRL
1
Zero
1
Percent of Systems with Mean
Exceeding Threshold1
Full MRL
0.890%
1/2 MRL
0.036%
Zero
0.036%
Combined
Ground &
Surface
Water
> 40 |jg/L
> 20 |jg/L
> 1 H9/L
> 0.9 |jg/L
0
0
1
55
0
0
1
2
0
0
1
1
0.000%
0.000%
0.003%
0.177%
0.000%
0.000%
0.003%
0.006%
0.000%
0.000%
0.003%
0.003%
1. The results shown in this table are based on system arithmetic means that were calculated using all sample
detection records and all non-detection records, with the non-detection records replaced (substituted) with
concentration values equal to the full MRL, 1A MRL, and zero.
Exhibit 3-14: Carbofuran Stage 2 Analysis - Population-Served by Systems with a
System Mean Threshold Exceedance
Source
Water Type
Ground
Water
Threshold
> 40 |jg/L
> 20 |jg/L
> 7 |jg/L
> 0.9 |jg/L
Population-Served by Systems
with Mean Exceeding Threshold1
Full MRL
0
0
0
1,138,494
1/2 MRL
0
0
0
220
Zero
0
0
0
0
Percent of Population -Served by
Systems with Mean Exceeding
Threshold1
Full MRL
0.000%
0.000%
0.000%
1.842%
1/2 MRL
0.000%
0.000%
0.000%
0.0004%
Zero
0.000%
0.000%
0.000%
0.000%
Surface
Water
> 40 |jg/L
> 20 |jg/L
> 7 |jg/L
> 0.9 |jg/L
0
0
3,186
7,701,631
0
0
3,186
3,186
0
0
3,186
3,186
0.000%
0.000%
0.003%
7.253%
0.000%
0.000%
0.003%
0.003%
0.000%
0.000%
0.003%
0.003%
Combined
Ground &
Surface
Water
> 40 |jg/L
> 20 |jg/L
> 7 |jg/L
> 0.9 |jg/L
0
0
3,186
8,840,125
0
0
3,186
3,406
0
0
3,186
3,186
0.000%
0.000%
0.002%
5.263%
0.000%
0.000%
0.002%
0.002%
0.000%
0.000%
0.002%
0.002%
1. The results shown in this table are based on system arithmetic means that were calculated using all sample
detection records and all non-detection records, with the non-detection records replaced (substituted) with
concentration values equal to the full MRL, 14 MRL, and zero.
3-12
-------
EPA-OGWDW Contaminant Occurrence Support Document for 2" 6-Year Review Category 1 Contaminants October 2009
3.5 Additional Occurrence Data
United States Department of Agriculture (USDA) Pesticide Data Program
Carbofuran was included in the USDA Pesticide Data Program (PDF; USDA, 2007).
Note that temporal trends cannot be evaluated based on these data since, with the exception of
2002 and 2003, samples were not collected from the same sites and states in consecutive years.
Exhibit 3-15: PDF Monitoring of Carbofuran in Drinking Water, 2001-2005
Year
2001
2002
2003
2004
2005
Treated/
Untreated
Treated
Treated
Treated
Treated
Untreated
Treated
Untreated
#of
Sites
21
27
27
16
16
States
CA, NY
CA, CO,
KS, NY, TX
CA, CO,
KS, NY, TX
Ml, NC, OH,
OR, PA,
WA
CA, FL, LA,
Ml, NC, ND,
OH, PA,
WA
#of
Samples
296
550
583
379
380
374
375
Samples
with
Detects
0
6
8
0
0
0
0
%of
Samples
with
Detects
0%
1.1%
1 .4%
0%
0%
0%
0%
Range of
Values
Detected
(M9/L)
-
0.001-0.079
0.005-0.020
-
-
-
-
Range of
LCDs (|jg/L)
0.020-0.0225
0.0006-0.0225
0.0006-0.016
0.0006-0.017
0.0006-0.017
0.0006-0.017
0.0006-0.017
USGS 1999 Pilot Monitoring Program
Carbofuran was an analyte in the 1999 Pilot Monitoring Program (Blomquist et al.,
2001). Carbofuran was detected in less than 1% of the 323 raw water samples and approximately
1.3% of the 228 finished water samples. The 95th percentile value was reported as less than the
MRL for both raw and finished water samples. The maximum concentration of Carbofuran in raw
water was 0.050 |ig/L. In finished water, the maximum concentration of Carbofuran was 0.030
First Six-Year Review ofNPDWRs
Similar types of Stage 1 contaminant occurrence analyses were conducted for the first
Six-Year Review. The first Six- Year Review assessed regulated contaminant occurrence data
collected from 1993 to 1997. The first Six-Year Review data set consisted of data from 16 states
that were assembled into a "national cross-section" that was qualitatively indicative (though not
statistically representative) of national occurrence. For the first Six-Year Review (USEPA,
2003b), 9 systems (approximately 0.06% of about 14,000 systems), reported at least one
detection of carbofuran. No systems reported a detection greater than the MCL.
3-13
-------
EPA-OGWDW Contaminant Occurrence Support Document for 2" 6-Year Review Category 1 Contaminants October 2009
3.6 Summary of Data
TRI data indicate that the total off-site and on-site releases of carbofuran for the period
1998-2006 ranged from about 500 to about 2,900 pounds, with the greatest release occurring in
1998. Air emissions constituted nearly all of the releases; surface water and off-site discharges
were minimal and underground injection and releases to land were non-existent.
Ambient water data from the NAWQA Pesticide National Synthesis Project show that
most carbofuran detections occurred at agricultural sites. It was detected in 10.37% of 2,003
agricultural stream samples taken at 83 sites. It was detected in 1.71% of samples taken from
1,404 ground water agricultural wells. Additional ambient water data from the NAWQA program
indicate that carbofuran was detected in 3.8% of 30,530 ground and surface water samples and at
4.4% of 7,142 sites. The median concentration, based on detections from all sites, was 0.035
The Stage 1 analysis of occurrence in drinking water indicates that carbofuran was
detected in 0. 145% of 3 1,097 surface water and ground water systems. The median
concentration, based on detections from all systems, was 0.54 |ig/L. No systems reported
detections greater than the MCL.
In the Stage 2 analysis, no systems had an estimated mean greater than 40 |ig/L (the MCL
concentration).
Recent Agency actions to revoke all carbofuran tolerances are expected to reduce
exposure to carbofuran and its metabolite (3-hydroxycarbofuran) in food products and in water.
3.7 References
Blomquist, J.D., J.M. Denis, J.L. Cowles, J.A. Hetrick, R.D. Jones, andN.B. Birchfield. 2001.
Pesticides in Selected Water-Supply Reservoirs and Finished Drinking Water, 1999-2000:
Summary of Results from a Pilot Monitoring Program. U.S. Geological Survey Water-Resources
Investigations Report 01-456, 65 p. Available on the Internet at:
http://md.water.usgs. gov/nawqa/OFR_0 l-456.pdf.
Extension Toxicology Network (EXTOXNET). 1996. EXTOXNET Pesticide Information
Profiles-Carbofuran. Last modified June 1996. Available on the Internet at:
http://extoxnet.orst.edu/pips/carbofur.htm. Accessed July 3, 2008.
Gilliom, R.J., J.E. Barbash, C.G. Crawford, P.A. Hamilton, J.D. Martin, N. Nakagaki, L.H.
Nowell, J.C. Scott, P.E. Stackelberg, G.P. Thelin, and D.M. Wolock. 2007. The Quality of Our
Nation's Waters - Pesticides in the Nation's Streams and Ground Water, 1992-2001. Appendix
7. Statistical summaries of water-quality data. U.S. Geological Survey Circular 1291, 172 p.
Available on the Internet at: http://water.usgs.gov/nawqa/pnsp/pubs/circl291/appendix7/.
3-14
-------
EPA-OGWDW Contaminant Occurrence Support Document for 2" 6-Year Review Category 1 Contaminants October 2009
Thelin, G.P. and L.P. Gianessi. 2000. Method for Estimating Pesticide Use for County Areas of
the Conterminous United States. U.S. Geological Survey Open-File Report 00-250, 62 p.
Available on the Internet at: http://ca.water.usgs.gov/pnsp/rep/ofr00250/ofr00250.pdf.
United States Department of Agriculture (USDA). 2007. Pesticide Data Program, Annual
Summary Reports available for downloading, view ing and printing. Available on the Internet at:
http://www.ams.usda.gov/science/pdp/Download.htm.
United States Environmental Protection Agency (USEPA). 1991. National Primary Drinking
Water Regulations—Synthetic Organic Chemicals and Inorganic Chemicals; Monitoring for
Unregulated Contaminants; National Primary Drinking Water Regulations Implementation;
National Secondary Drinking Water Regulations; Final Rule Federal Register. Vol. 56, No. 30. p.
3526, January 30, 1991.
USEPA. 2003a. Non-confidential Inventory Update Reporting (IUR) Production Volume
Information - Carbofuran. Last modified September 10, 2007. Available on the Internet at:
http://www.epa.gov/oppt/iur/tools/data/2002-vol.htm. [Search for carbofuran.] Accessed
February 21, 2008.
USEPA. 2003b. Occurrence Estimation Methodology and Occurrence Findings for Six-Year
Review of National Primary Drinking Water Regulations. Office of Water. EPA-815-R-03-006.
June 2003.
USEPA. 2006. Interim Reregistration Eligibility Decision: Carbofuran. EPA-738-R-06-031.
August 2006. Available on the Internet at:
http://www.epa.gov/pesticides/reregistration/REDs/carbofuran_ired.pdf.
USEPA. 2008. TRI Explorer: Trends - Carbofuran. Released February 21, 2008. Available on
the Internet at: http://www.epa.gov/triexplorer/trends.htm. [Search for carbofuran.] Accessed
February 29, 2008.
USEPA. 2009a. The Analysis of Regulated Contaminant Occurrence Data from Public Water
Systems in Support of the Second Six-Year Review of National Primary Drinking Water
Regulations. Office of Water. EPA-815-B-09-006. October 2009.
USEPA. 2009b. Carbofuran; Final Tolerance Revocations; Final Rule. Federal Register. Vol. 74,
No. 93. p. 23046-23095, May 15, 2009.
United States Geological Survey (USGS). 2007. 2002 Pesticide Use Maps. Available on the
Internet at: http://water.usgs.gov/nawqa/pnsp/usage/maps/compound_listing.php?year=02 .
Accessed December 19, 2007.
3-15
-------
EPA-OGWDW Contaminant Occurrence Support Document for 2" 6-Year Review Category 1 Contaminants October 2009
4 Carbon Tetrachloride
This chapter on carbon tetrachloride is part of a report that is organized so that each
chapter presents information on one specific contaminant. Chapter 1 of this report, Introduction,
presents background information on all sources of data used as well as the analytical approach
used to estimate contaminant occurrence in drinking water. This contaminant chapter includes
background information (such as regulatory history, general chemical information, and
environmental fate and behavior), use and environmental release information, and carbon
tetrachloride occurrence estimates in ambient water and drinking water. All drinking water
occurrence estimates were based on data from the National Compliance Monitoring Information
Collection Request (ICR) Dataset for the Second Six-Year Review (the "Six Year Review-ICR
Dataset"). For detailed information regarding this dataset, data management, data quality
assessments, and the analytical approach used to generate all contaminant occurrence estimates
presented in this chapter, please refer to USEPA (2009).
4.1 Background
The United States Environmental Protection Agency (EPA) published the current
maximum contaminant level goal (MCLG) for carbon tetrachloride on July 8, 1987 (52 FR
25690 (USEPA, 1987)). The National Primary Drinking Water Regulations (NPDWR)
established a MCLG of zero based on a cancer classification of B2, probable human carcinogen.
The NPDWR also established a maximum contaminant level (MCL) of 5 |ig/L based on
analytical feasibility.
According to the Agency for Toxic Substances and Disease Registry (ATSDR, 2005),
carbon tetrachloride (CCU) is a clear liquid with a sweet odor. It has low flammability, but
volatilizes quite easily, typically escaping to the environment in gaseous form. However, the
environmental effects of this chemical prompted sharp declines in use since the 1970s. Carbon
tetrachloride in the atmosphere has been linked to the thinning of the ozone layer, a component
of the stratosphere responsible for shielding humans from cancer-causing ultraviolet solar
radiation (ATSDR, 2005).
Carbon tetrachloride does not occur naturally. It is expected that discharges to surface
water during production and use, and leaching into ground water from wastes deposited in
landfills are primary causes of carbon tetrachloride contamination of drinking water. Generally, it
makes its way to the air because it tends to evaporate from surface waters and soils and is not
chemically inclined to adhere to soil particles. In the air, it may take several years to break down
into other chemicals. If it becomes trapped in ground water, however, its breakdown time
accelerates to months (ATSDR, 2005).
A low background concentration of 0.1 ppb of carbon tetrachloride exists in the air
around the world, though levels in cities frequently rise to between 0.2 and 0.6 ppb. Its past use
has also resulted in low background levels in the soil and water around the world. Human
exposure to increased levels of carbon tetrachloride is typical only at specific industrial locations
where carbon tetrachloride is still used or near chemical waste sites where emissions into air,
4-1
-------
EPA-OGWDW Contaminant Occurrence Support Document for 2" 6-Year Review Category 1 Contaminants October 2009
water, or soil are not properly controlled. Carbon tetrachloride has been found in water or soil at
about 26% of the waste sites investigated under Superfund, at concentrations ranging from less
than 50 to over 1,000 ppb (ATSDR, 2005).
4.2 Use, Production, and Releases
Carbon tetrachloride has been produced in large quantities to make refrigeration fluid and
propellants for aerosol cans. It has also been used as a cleaning fluid (as a degreasing agent and
as a spot remover for clothing, furniture, and carpets). Other applications include use in fire
extinguishers and as a fumigant to kill insects in grain. Most of these uses were discontinued in
the mid-1960s. Use as a pesticide was discontinued in 1986 (ATSDR, 2005).
The Toxic Substances Control Act (TSCA) Inventory contains confidential data reported
by industry as a partial update of the Chemical Substance Inventory Update System. The data are
reported under the Inventory Update Reporting regulation. Manufacturers and importers are
required to report company information and chemical information, including production volumes
for chemicals they manufactured or imported in excess of 10,000 pounds in the immediately
preceding fiscal year.
The TSCA Chemical Substance Inventory does not provide actual production volumes;
instead, it indicates at which of five production volume ranges a chemical was manufactured or
imported. The TSCA Inventory is updated every four years. Exhibit 4-1 presents the production
of carbon tetrachloride in the United States from 1986 to 2002. Production decreased between
1990 and 1994 and has remained within the same production range since 1994.
Exhibit 4-1: Production of Carbon tetrachloride in the United States
Inventory Update Year
1986 Range
1990 Range
1994 Range
1998 Range
2002 Range
Production Volume
> 500 million - 1 billion pounds
> 500 million - 1 billion pounds
> 100 million - 500 million pounds
> 100 million - 500 million pounds
> 100 million - 500 million pounds
Source: USEPA, 2003a
Carbon tetrachloride is listed as a Toxics Release Inventory (TRI) chemical. Exhibit 4-2a
lists the environmental releases of carbon tetrachloride from 1998 to 2006. (Exhibit 4-2b
provides a graphical presentation of the environmental releases.) Air emissions constitute the
vast majority of the on-site releases and, except for a peak in 2002, remained relatively constant
over the years. Surface water discharges constituted less than 1% of total releases. Underground
4-2
-------
EPA-OGWDW Contaminant Occurrence Support Document for 2" 6-Year Review Category 1 Contaminants October 2009
injection increased until 2002 and then decreased. Off-site releases were small and generally
decreased over the period of record. On-site surface water discharges, which were highest in
1998, dropped in subsequent years. Off-site releases decreased gradually over the period of
record. These TRI data were reported from 31 states, Puerto Rico, and the Virgin Islands.
Thirteen of the 31 states reported every year.
Exhibit 4-2a: Table of Toxics Release Inventory (TRI) Data for Carbon
Tetrachloride
Year
1998
1999
2000
2001
2002
2003
2004
2005
2006
On-Site Releases (pounds)
Total Air
Emissions
275,342
232,988
281,033
288,214
444,436
228,158
208,080
246,416
218,445
Surface Water
Discharges
2,836
85
184
85
320
308
678
385
543
Underground
Injection
23,168
27,548
62,203
113,881
172,247
96,698
102,061
17,266
23,363
Releases to
Land
1,679
938
835
5,929
47
193
25
12,527
21
Off -site
Releases
(pounds)
20,251
16,566
2,546
3,557
8,493
2,067
8,906
1,647
1,487
Total On- &
Off-site
Releases
(pounds)
323,276
278,125
346,801
411,666
625,544
327,424
319,749
278,241
243,859
Source: USEPA, 2008
Exhibit 4-2b: Graph of Toxics Release Inventory (TRI) Data for Carbon
Tetrachloride
Environmental Releases of Carbon Tetrachloride
>icn nnn
Pounds Released
Ann nnn
ocn nnn
onn nnn
ocn nnn
200,000 -
150,000 -
-inn nnn
en nnn
n
• n n
111 II
n
n. INI II
_.
•
n
- ^r
TIC
1998 1999 2000 2001 2002 2003
Year
n [1 n
n N
n
II ~ llr^ IL
2004 2005 2006
•Air Emissions DSurface Water Discharges D Underground Injection DReleases to Land • Off-site Releases
Source: USEPA, 2008
4-3
-------
EPA-OGWDW Contaminant Occurrence Support Document for 2" 6-Year Review Category 1 Contaminants October 2009
4.3 Occurrence in Ambient Water
Lakes, rivers, and aquifers are the sources of most drinking water. The United States
Geological Survey (USGS) National Water Quality Assessment (NAWQA) program is a
national-scale assessment of the occurrence of contaminants in ambient surface and ground
water. Contaminant occurrence in ambient water provides information on the potential for
contaminants to adversely affect drinking water supplies. Data on the occurrence of carbon
tetrachloride in ambient water are available from the NAWQA program.
4.3.1 EPA Summary Analysis ofNA WQA Data
EPA, with the cooperation of the USGS, performed a summary analysis of all Cycle 1
(1992-2001) water monitoring data from all study units for carbon tetrachloride. Detection
frequencies were computed as the percentage of samples and sites with detections (i.e., with at
least one result equal to or greater than the reporting limit). Sample detections can be biased by
frequent sampling in areas with high (or low) occurrence. Calculating the percentage of sites with
detections can reduce this bias.
The results of the EPA analysis are presented in Exhibit 4-3. Overall, carbon tetrachloride
was detected in 1.5% of samples and at 1.2% of sites. Carbon tetrachloride was detected more
frequently in surface water than in ground water. The median concentration, based on detections
from all sites, was 0.045 |ig/L. The 99th percentile concentration, based on detections from all
sites, was 2.38 |ig/L.
Exhibit 4-3: EPA Summary Analysis of Carbon Tetrachloride Data from NAWQA
Study Units, 1992-2001
ground
water
surface
water
all
sites
Detection Frequency
(detections are results > RL1)
Number
of
Samples
4,658
1,413
6,071
% Samples
with
Detections
1.3%
2.0%
1.5%
Number
of Sites
4,159
190
4,349
% Sites
with
Detections
1.0%
4.7%
1.2%
Concentration Values
(of detections, in ug/L)
Minimum
0.01
0.016
0.01
Median
0.0535
0.03395
0.04525
95tn
Percen-
tile
1.1
0.3747
0.9
99tn
Percen-
tile
2.377
0.4023
2.377
Maxi-
mum
2.377
0.4023
2.377
1. RLs (Reporting Limits) for carbon tetrachloride varied, but did not exceed 0.016 pg/L
4.4 Occurrence in Drinking Water
Carbon tetrachloride is regulated as a volatile organic compound (VOC) in drinking
water. All non-purchased community water systems (CWSs) and non-transient non-community
4-4
-------
EPA-OGWDW Contaminant Occurrence Support Document for 2" 6-Year Review Category 1 Contaminants October 2009
water systems (NTNCWSs) are required to sample for VOCs. The maximum waiver period for
VOCs is six years for ground water systems and three years for surface water systems.
All CWSs and NTNCWSs must collect four consecutive quarterly samples during the
initial three-year compliance period.1 If all four samples are non-detections, then the system may
reduce to annual sampling. After three annual samples without a detection, a system may be
granted a waiver. If a compound is detected (but at less than the MCL), the system must take one
sample per quarter until results are below the MCL (minimum of two quarterly samples for
ground water systems and four quarterly samples for surface water systems). If all quarterly
samples are below the MCL, the system may return to annual sampling. If a compound is
detected at a level greater than the MCL, the system (whether ground water or surface water)
must take four consecutive quarterly samples until all are below the MCL. If all quarterly
samples are below the MCL, the system may return to annual sampling.
The analysis of carbon tetrachloride occurrence presented in the following section is
based on state compliance monitoring data from the Six-Year Review-ICR Dataset. This is the
largest and most comprehensive compliance monitoring data set compiled by EPA's Drinking
Water program to date.
All of the Six-Year Review-ICR states provided occurrence data for carbon tetrachloride.
These data consist of 385,864 analytical results from 50,446 public water systems (PWSs) during
the period from 1998 to 2005. The number of sample results and systems vary by state, although
the state data sets have been reviewed and checked to ensure adequacy of coverage and
completeness. The national modal minimum reporting level (MRL) for carbon tetrachloride in
the dataset is 0.5 |ig/L.
4.4.1 Stage 1A nalysis of Occurrence Data
Stage 1 analyses have been performed at the sample, sample point, and system levels.
Note that some systems have multiple sample points, such as source water sample points or entry
points to the distribution system. Exhibit 4-4 presents the number of samples, systems, and
sample points in the data set, as well as the population-served by the systems and sample points,
and the percentage of detections and the percentage of the population-served by the systems and
sample points with detections. Detections in systems provide a fundamental indication of
national occurrence, but are not representative of MCL violations. For carbon tetrachloride,
1.289% of 50,446 systems reported detections. Exhibit 4-5 lists the minimum, median, 90th
percentile, and maximum carbon tetrachloride concentrations based on detections from all
systems in the Six-Year Review-ICR Dataset. The median concentration of carbon tetrachloride,
based on detections from all systems, was 1.1 |ig/L.
1 All new systems or systems using a new water source that began operation after January 22, 2004 must demonstrate
compliance with the MCL within a period of time specified by the State. The system must also comply with the
initial sampling frequencies specified by the State to ensure that a system can demonstrate compliance with the MCL.
4-5
-------
EPA-OGWDW Contaminant Occurrence Support Document for 2" 6-Year Review Category 1 Contaminants October 2009
Exhibit 4-4: Carbon Tetrachloride Occurrence Data from the Six-Year Review-ICR
Dataset
Source Water Type
Number of
Samples
Number of
Systems
Number of
Sample Points
Pop. Served
by Systems
Pop. Served
by Sample
Points
Total Number
Ground Water
295,112
46,190
80,583
86,397,574
86,397,574
Surface Water
90,752
4,256
10,885
140,537,213
140,537,213
Combined Ground &
Surface Water
385,864
50,446
91,468
226,934,787
226,934,787
Percent with Detections
Ground Water
0.815%
0.935%
0.635%
3.419%
1 .243%
Surface Water
2.534%
5.122%
2.839%
13.773%
5.827%
Combined Ground &
Surface Water
1 .220%
1 .289%
0.898%
9.831%
4.082%
Exhibit 4-5: Carbon Tetrachloride Summary Statistics
Source Water Type
Ground Water
Detection Value (ug/L)
Minimum
0.03
Median
1.24
90tn
Percentile
3.5
Maximum
300
Surface Water
0.003
1
5.04
27
Combined Ground &
Surface Water
0.003
1.1
3.9
300
The following maps illustrate the geographic distribution of carbon tetrachloride
occurrence in drinking water. Exhibit 4-6 shows the states with systems with at least one
detection and Exhibit 4-7 shows the states with systems with at least one detection greater than
the MCL concentration. Note that five states did not submit data for use in the Six-Year Review
(Kansas2, Louisiana, Mississippi, Pennsylvania, and Washington).
The distribution of systems with detections of carbon tetrachloride is geographically
dispersed. Detection rates were generally low, although 43 states reported carbon tetrachloride
2 Kansas submitted data after the data management and analyses had been conducted; data from Kansas are not
included in the analyses.
4-6
-------
EPA-OGWDW Contaminant Occurrence Support Document for 2" 6-Year Review Category 1 Contaminants October 2009
detections. Only 4 of the 43 states reported detections in more than 5% of their systems. Twenty-
five states reported detections greater than the MCL concentration of 5 |ig/L, although only two
states reported more than 0.5% of systems with detections greater than 5 |ig/L.
Exhibit 4-6: Percentage of Systems with at Least One Carbon Tetrachloride
Detection, by State
/////. States Not Submitting Any Data
States with No Detections,
6 Leu than 1% of Systems detecting
Between 1% and 5% of Systems Detecting
More than 5% of Syftemg Delecting
4-7
-------
EPA-OGWDW Contaminant Occurrence Support Document for 2" 6-Year Review Category 1 Contaminants October 2009
Exhibit 4-7: Percentage of Systems with at Least One Carbon Tetrachloride
Detection Greater than the MCL Concentration (> 5 ug/L), by State
\
: States Not Submitting Any Data
States with No Detection* > MCL
Less than 0,1% of Systems with Detection* > MCL
M Between 0.1% and 0.5% of Systems with Detections > MCL
^B More Uian 0.5% of Systems with Detections > MCL
Exhibit 4-8 presents the Stage I analysis of carbon tetrachloride occurrence in drinking
water from systems and the population-served by those systems in the Six-Year Review-ICR
Dataset relative to several thresholds. The thresholds are: 0.5 |ig/L, 1 |ig/L, 2.5 |ig/L, and 5 |ig/L.
A total of 106 (0.210% of) systems, serving about 5.5 million persons, reported at least one
detection greater than or equal to 2.5 |ig/L; and 57 (0.113% of) systems, serving nearly 5 million
persons, reported at least one detection exceeding 5 |ig/L. Detections greater than the thresholds
were more common in ground water systems than surface water systems. However, because
surface water systems tend to be larger, a greater population was served by surface water systems
reporting at least one detection greater than the thresholds.
Exhibit 4-8: Carbon Tetrachloride Stage 1 Analysis - Systems and Population
with at Least One Threshold Exceedance
Source
Water Type
Ground
Water
Threshold
> 5 |jg/L
>2.5|jg/L
> 1 |jg/L
> 0.5 |jg/L
Number of
Systems
Exceeding
Threshold
49
81
179
312
Percent of
Systems
Exceeding
Threshold
0.106%
0.175%
0.388%
0.675%
Total Population-
Served by
Systems
Exceeding
Threshold
146,334
313,180
1,010,901
1,813,677
Percent of
Population-
Served by
Systems
Exceeding
Threshold
0.169%
0.362%
1.170%
2.099%
4-8
-------
EPA-OGWDW Contaminant Occurrence Support Document for 2" 6-Year Review Category 1 Contaminants October 2009
Source
Water Type
Surface
Water
Threshold
> 5 |jg/L
>2.5|jg/L
> 1 H9/L
> 0.5 |jg/L
Number of
Systems
Exceeding
Threshold
8
25
82
147
Percent of
Systems
Exceeding
Threshold
0.188%
0.587%
1.927%
3.454%
Total Population-
Served by
Systems
Exceeding
Threshold
4,821,230
5,174,513
7,071,752
12,609,728
Percent of
Population-
Served by
Systems
Exceeding
Threshold
3.431%
3.682%
5.032%
8.973%
Combined
Ground &
Surface
Water
> 5 |jg/L
>2.5|jg/L
> 1 H9/L
> 0.5 |jg/L
57
106
261
459
0.113%
0.210%
0.517%
0.910%
4,967,564
5,487,693
8,082,653
14,423,405
2.189%
2.418%
3.562%
6.356%
The occurrence estimates presented above (in Exhibit 4-8) are based on the conservative
assumption that if a carbon tetrachloride detection is found in a single entry point/sample point in
a system, then the entire population-served by the system is assumed to potentially be exposed to
carbon tetrachloride (even though other entry points/sampling points at the system, and the
population-served by those entry/sample points, may exhibit no occurrence of carbon
tetrachloride). The sample point analysis is a less conservative estimate of the population-served
by a system with a detection of carbon tetrachloride because it estimates the potentially exposed
population-served based only on the proportion of the population-served by entry points/sample
points with detections of carbon tetrachloride (rather than the entire population-served by the
system).
Exhibit 4-9 presents the Stage 1 analysis of carbon tetrachloride occurrence in drinking
water from sample points and population-served by those sample points in the Six-Year Review-
ICRDataset. A total of 135 (0.148% of) sample points, serving an estimated 578,000 persons,
reported at least one detection greater than or equal to 2.5 |ig/L; and 67 (0.073% of) sample
points, serving an estimated 160,000 persons, reported at least one detection exceeding 5 |ig/L.
The percentages of sample points exceeding 5 |ig/L were small for both ground water and surface
sample points, but a larger population was served by surface water sample points with threshold
exceedances.
4-9
-------
EPA-OGWDW Contaminant Occurrence Support Document for 2" 6-Year Review Category 1 Contaminants October 2009
Exhibit 4-9: Carbon Tetrachloride Stage 1 Analysis - Sample Points and
Population with at Least One Threshold Exceedance
Source
Water Type
Ground
Water
Threshold
> 5 pg/L
>2.5|jg/L
> 1 |jg/L
> 0.5 |jg/L
Number of
Sample Points
Exceeding
Threshold
52
90
202
375
Percent of
Sample Points
Exceeding
Threshold
0.065%
0.112%
0.251%
0.465%
Total Population-
Served by Sample
Points Exceeding
Threshold
29,169
76,401
266,717
552,059
Percent of
Population-
Served by
Sample Points
Exceeding
Threshold
0.034%
0.088%
0.309%
0.639%
Surface
Water
> 5 pg/L
>2.5|jg/L
> 1 |jg/L
> 0.5 |jg/L
15
45
135
221
0.138%
0.413%
1 .240%
2.030%
130,663
501,204
1,599,997
4,333,382
0.093%
0.357%
1.138%
3.083%
Combined
Ground &
Surface
Water
> 5 pg/L
>2.5|jg/L
> 1 |jg/L
> 0.5 |jg/L
67
135
337
596
0.073%
0.148%
0.368%
0.652%
159,832
577,605
1,866,714
4,885,441
0.070%
0.255%
0.823%
2.153%
4.4.2 Stage 2 A nalysis of Occurrence Data
For the Stage 2 analyses, three different substitution values -zero, one-half the MRL
value, and the full MRL value- were used to replace each non-detection record. Three arithmetic
mean carbon tetrachloride concentrations were calculated at each system using the zero, one-half
MRL and full MRL substitution values. These mean calculations were done for all systems with
data in the Six-Year Review-ICR dataset. Then the percentages of all systems with a mean
concentration greater than each threshold were calculated.
Exhibit 4-10 presents the Stage 2 analysis of estimated system mean concentrations for
carbon tetrachloride occurrence in drinking water relative to several thresholds. The thresholds
are: 0.5 |ig/L, 1 |ig/L, 2.5 |ig/L, and 5 |ig/L. Exhibit 4-11 presents similar information based on
population-served by the systems. Based on the Stage 2 analyses, no more than 13 ground water
systems (approximately 0.026% of all systems) had an estimated system mean greater than or
equal to 2.5 |ig/L. These 13 systems serve 5,808 persons. Five systems (approximately 0.010% of
all systems), serving 1,746 persons, had an estimated system mean greater than the MCL
concentration. No system mean exceedances of the MCL concentration occurred in surface water
systems.
4-10
-------
EPA-OGWDW Contaminant Occurrence Support Document for 2" 6-Year Review Category 1 Contaminants October 2009
Exhibit 4-10: Carbon Tetrachloride Stage 2 Analysis - Systems with a System
Mean Threshold Exceedance
Source
Water Type
Ground
Water
Threshold
> 5 |jg/L
>2.5|jg/L
> 1 |jg/L
> 0.5 |jg/L
Number of Systems with Mean
Exceeding Threshold1
Full MRL
5
13
54
293
1/2 MRL
5
12
46
101
Zero
5
12
36
76
Percent of Systems with Mean
Exceeding Threshold1
Full MRL
0.011%
0.028%
0.117%
0.634%
1/2 MRL
0.011%
0.026%
0.100%
0.219%
Zero
0.011%
0.026%
0.078%
0.165%
Surface
Water
> 5 |jg/L
>2.5|jg/L
> 1 |jg/L
> 0.5 |jg/L
0
0
5
133
0
0
4
17
0
0
4
8
0.000%
0.000%
0.117%
3.125%
0.000%
0.000%
0.094%
0.399%
0.000%
0.000%
0.094%
0.188%
Combined
Ground &
Surface
Water
> 5 pg/L
>2.5|jg/L
> 1 |jg/L
> 0.5 |jg/L
5
13
59
426
5
12
50
118
5
12
40
84
0.010%
0.026%
0.117%
0.844%
0.010%
0.024%
0.099%
0.234%
0.010%
0.024%
0.079%
0.167%
1 The results shown in this table are based on system arithmetic means that were calculated using all sample
detection records and all non-detection records, with the non-detection records replaced (substituted) with
concentration values equal to the full MRL, 14 MRL, and zero.
Exhibit 4-11: Carbon Tetrachloride Stage 2 Analysis - Population-Served by
Systems with a System Mean Threshold Exceedance
Source
Water Type
Ground
Water
Threshold
> 5 |jg/L
>2.5|jg/L
> 1 |jg/L
> 0.5 |jg/L
Population-Served by Systems with
Mean Exceeding Threshold1
Full MRL
1,746
5,808
60,156
1,638,978
1/2 MRL
1,746
5,508
58,138
159,018
Zero
1,746
5,508
35,883
131,766
Percent of Population-Served by
Systems with Mean Exceeding
Threshold1
Full MRL
0.002%
0.007%
0.070%
1.897%
1/2 MRL
0.002%
0.006%
0.067%
0.184%
Zero
0.002%
0.006%
0.042%
0.153%
Surface
Water
> 5 |jg/L
>2.5|jg/L
> 1 |jg/L
> 0.5 |jg/L
0
0
204,708
9,450,131
0
0
153,958
590,725
0
0
153,958
236,026
0.000%
0.000%
0.146%
6.724%
0.000%
0.000%
0.110%
0.420%
0.000%
0.000%
0.110%
0.168%
4-11
-------
EPA-OGWDW Contaminant Occurrence Support Document for 2" 6-Year Review Category 1 Contaminants October 2009
Source
Water
Type
Combined
Ground &
Surface
Water
Threshold
> 5 |jg/L
>2.5|jg/L
> 1 H9/L
> 0.5 |jg/L
Population-Served by Systems with
Mean Exceeding Threshold1
Full MRL
1,746
5,808
264,864
11,089,109
1/2 MRL
1,746
5,508
212,096
749,743
Zero
1,746
5,508
189,841
367,792
Percent of Population-Served by
Systems with Mean Exceeding
Threshold1
Full MRL
0.001%
0.003%
0.117%
4.886%
1/2 MRL
0.001%
0.002%
0.093%
0.330%
Zero
0.001%
0.002%
0.084%
0.162%
1. The results shown in this table are based on system arithmetic means that were calculated using all sample
detection records and all non-detection records, with the non-detection records replaced (substituted) with
concentration values equal to the full MRL, 14 MRL, and zero.
4.5 Additional Occurrence Data
Volatile Organic Compounds in the Nation's Ground Water and Drinking Water Supply Wells
The USGS conducted a national assessment of 55 VOCs, including carbon tetrachloride,
in samples collected from drinking water supply wells, specifically domestic and public wells
(Zogorski et a/., 2006). This occurrence assessment was based on analyses of samples collected
at the well head, and before any treatment or blending, from about 2,400 domestic wells and
about 1,100 public wells.
In domestic wells, carbon tetrachloride had a detection frequency of 0.21% at an
assessment level of 0.2 |ig/L (based on 2,400 samples). The median concentration of detections
was 0.043 |ig/L. Three detections were within one order of magnitude of the MCL concentration
(of 5 |ig/L). No detections were greater than the MCL concentration.
In public wells, carbon tetrachloride had a detection frequency of 0.73% at an assessment
level of 0.2 jig/L (based on 1,096 samples). The median concentration of detections was 0.38
|ig/L. Five detections were within one order of magnitude of the MCL concentration. No
detections were greater than the MCL concentration.
First Six-Year Review ofNPDWRs
Similar types of Stage 1 contaminant occurrence analyses were conducted for the first
Six-Year Review. The first Six-Year Review assessed regulated contaminant occurrence data
collected from 1993 to 1997. The first Six-Year Review data set consisted of data from 16 states
that were assembled into a "national cross-section" that was qualitatively indicative (though not
statistically representative) of national occurrence. For the first Six-Year Review (USEPA,
2003b), 458 systems (approximately 1.99% of about 23,000 systems), reported at least one
detection of carbon tetrachloride. Forty-six systems (approximately 0.2% of all systems),
reported at least one detection greater than the MCL concentration.
4.6 Summary of Data
TRI data indicate that total off-site and on-site releases of carbon tetrachloride for the
period 1998-2006 ranged from about 244,000 pounds to about 626,000 pounds, with a peak in
4-12
-------
EPA-OGWDW Contaminant Occurrence Support Document for 2" 6-Year Review Category 1 Contaminants October 2009
2002. Most of the releases were in the form of air emissions. Underground injection was the
second most important source of releases. Surface water releases, releases to land, and off-site
releases were relatively minor.
Ambient water data from the NAWQA program indicate that carbon tetrachloride was
detected in 1.5% of 6,071 ground and surface water samples and at 1.2% of 4,349 sites. The
median concentration, based on detections from all sites, was 0.045 |ig/L.
The Stage 1 analysis of occurrence in drinking water indicates that carbon tetrachloride
was detected in 1.289% of 50,446 surface water and ground water systems. The median
concentration, based on detections from all systems, was 1.1 |ig/L. In total, 0.113% of systems
(serving 2.189% of the population) reported at least one detection greater than the MCL
concentration of 5 |ig/L. Detections were more common in ground water systems than in surface
water systems. However, because surface water systems tend to be larger, a greater population
was served by surface water systems reporting at least one detection greater than the thresholds.
The pattern was similar for sample points.
In the Stage 2 analysis, 0.010% of systems (serving 0.001% of the total population) had
an estimated system mean concentration of carbon tetrachloride greater than the MCL
concentration. No surface water systems had mean values exceeding the MCL concentration.
4.7 References
Agency for Toxic Substances and Disease Registry (ATSDR). 2005. Toxicological Profile for
Carbon Tetrachloride. August 2005. Available on the Internet at:
http ://www. atsdr. cdc. gov/toxprofiles/tp3 O.html.
United States Environmental Protection Agency (USEPA). 1987. National Primary Drinking
Water Regulations—Synthetic Organic Chemicals; Monitoring for Unregulated Contaminants;
Final Rule. Federal Register. Vol. 52, No. 130. p. 25690, July 8, 1987.
USEPA. 2003a. Non-confidential Inventory Update Reporting (IUR) Production Volume
Information - Carbon tetrachloride. Last modified September 10, 2007. Available on the Internet
at: http://www.epa.gov/oppt/iur/tools/data/2002-vol.htm. [Search for carbon tetrachloride.]
Accessed February 21, 2008.
USEPA. 2003b. Occurrence Estimation Methodology and Occurrence Findings for Six-Year
Review of National Primary Drinking Water Regulations. Office of Water. EPA-815-R-03-006.
June 2003.
USEPA. 2008. TRI Explorer: Trends - Carbon Tetrachloride. Released February 21, 2008.
Available on the Internet at: http://www.epa.gov/triexplorer/trends.htm. [Search for carbon
tetrachloride.] Accessed February 29, 2008.
USEPA. 2009. The Analysis of Regulated Contaminant Occurrence Data from Public Water
Systems in Support of the Second Six-Year Review of National Primary Drinking Water
Regulations. Office of Water. EPA-815-B-09-006. October 2009.
4-13
-------
EPA-OGWDW Contaminant Occurrence Support Document for 2" 6-Year Review Category 1 Contaminants October 2009
Zogorski, J.S., J.M. Carter, T. Ivahnenko, W.W. Lapham, MJ. Moran, B.L. Rowe, P.J.
Squillace, and P.L. Toccalino. 2006. Volatile organic compounds in the nation's ground water
and drinking water supply wells. U.S. Geological Survey Circular 1292, 101 p. Available on the
Internet at: http://pubs.er.usgs.gov/pubs/cir/cirl292.
4-14
-------
EPA-OGWDW Contaminant Occurrence Support Document for 2" 6-Year Review Category 1 Contaminants October 2009
5 Chlordane
This chapter on chlordane is part of a report that is organized so that each chapter
presents information on one specific contaminant. Chapter 1 of this report, Introduction, presents
background information on all sources of data used as well as the analytical approach used to
estimate contaminant occurrence in drinking water. This contaminant chapter includes
background information (such as regulatory history, general chemical information, and
environmental fate and behavior), use and environmental release information, and chlordane
occurrence estimates in ambient water and drinking water. All drinking water occurrence
estimates were based on data from the National Compliance Monitoring Information Collection
Request (ICR) Dataset for the Second Six-Year Review (the "Six Year Review-ICR Dataset").
For detailed information regarding this dataset, data management, data quality assessments, and
the analytical approach used to generate all contaminant occurrence estimates presented in this
chapter, please refer to USEPA (2009).
5.1 Background
The United States Environmental Protection Agency (EPA) published the current
National Primary Drinking Water Regulations (NPDWR) for chlordane on January 30, 1991 (56
FR 3526 (USEPA, 1991)). The NPDWR established a maximum contaminant level goal
(MCLG) of zero based on a cancer classification of B2, probable human carcinogen. The
NPDWR also established a maximum contaminant level (MCL) of 2 |ig/L based on analytical
feasibility.
According to the Agency for Toxic Substances and Disease Registry (ATSDR, 1994),
chlordane (CioHeClg) is not a single chemical but a mixture of many related chemicals used as a
pesticide in the United States between 1948 and 1988. Some of the major components are trans-
chlordane, cis-chlordane, beta-chlordene, heptachlor, and trans-nonachlor. The mixture forms a
thick liquid, which may range in color from clear to amber, and may vary in smell between
odorless and irritatingly acrid. Chlordane is produced by chlorinating cyclopentadiene to form
hexachlorocyclopentadiene and condensing the latter with cyclopentadiene to form chlordene.
The chlordene is further chlorinated at high temperature and pressure to chlordane. Because of
wide-ranging health and environmental concerns, chlordane's use was restricted to termite
control in the 1980s and has since been banned entirely. Nonetheless, production for export is
still permitted (ATSDR, 1994).
When chlordane is applied to surface soils, about half of it evaporates within two to three
days. However, once the soil is tilled, it may remain underground for over 20 years, especially if
the soil is heavy, clayey, or largely organic. Because it attaches strongly to particles in the upper
layers of soil, chlordane is not likely to infiltrate into ground water. It is not known to what extent
chlordane breaks down in soil or water. In the air, it breaks down by reacting with light and other
chemicals, but it is long-lived enough that it may redeposit on land or water far away from its
source. Chlordane is remarkably persistent in the environment, and even though it has long been
banned in the United States, it was still detectable in the fat offish, birds, mammals, and almost
all humans as of the 1990s (ATSDR, 1994).
5-1
-------
EPA-OGWDW Contaminant Occurrence Support Document for 2" 6-Year Review Category 1 Contaminants October 2009
Since use of chlordane has been banned, those at the highest risk for high exposure are
individuals living in houses that were treated for termites using chlordane. Over 50 million
persons in the United States have lived in such homes, where maximum concentrations have
been measured as high as 0.06 mg/m3. Chlordane also remains in the food supply because much
of the farmland was treated with chlordane in the 1960s and 1970s and the chemical persists in
the soil. A survey conducted by the Food and Drug Administration (FDA) determined the daily
intake of chlordane from food to be between 0.0005 and 0.0015 ng/kg for teenagers and adults
(ATSDR, 1994).
5.2 Use, Production, and Releases
The commercial use of chlordane in the United States was banned April 14, 1988;
however it is still manufactured for export. Chlordane and a chlordane-heptachlor combination
were used in Canada and the United States as insecticides on various types of agricultural crops
and vegetation, in the home, on food and nonfood crops, and on lawns and gardens (ASTDR,
1994). In the mid 1970's 35% of chlordane was used by pest control operators, mostly on
termites; 28% was used on agricultural crops, including corn and citrus; 30% was used for home
lawn and garden use; and 7% was applied on turf and ornamentals (ASTDR, 1994). The only
commercial use of chlordane products still permitted is for fire ant control in power transformers.
The Chemical Substance Inventory Update System contains confidential data reported by
industry as a partial update of the Toxic Substances Control Act (TSCA) Inventory. The TSCA
Chemical Substance Inventory does not report production data for chlordane.
Chlordane is listed as a Toxics Release Inventory (TRI) chemical. Exhibit 5-la lists the
environmental releases of chlordane from 1998 to 2006. (Exhibit 5-lb provides a graphical
presentation of the environmental releases.) The majority of chlordane releases were releases to
land, with generally minimal releases to air and surface water. There has been no underground
injection except for in 1998, when large amounts of chlordane were both injected underground
and released to land. Off-site releases have also been generally only a few percent of the total.
These TRI data for chlordane were reported from 20 states, with only 4 of the 20 states
(Arkansas, New Jersey, Ohio, and Texas) reporting every year.
Exhibit 5-1 a: Table of Toxics Release Inventory (TRI) Data for Chlordane
Year
1998
1999
2000
2001
2002
2003
On-Site Releases (pounds)
Total Air
Emissions
45
8
14
15
501
12
Surface Water
Discharges
1
1
0
80
147
107
Underground
Injection
20,106
0
0
0
0
0
Releases
to Land
25,548
0
8,948
3,648
1,659
3,483
Off-Site
Releases
(pounds)
22
40
47
336
35
67
Total On- &
Off-site
Releases
(pounds)
45,721
49
9,008
4,079
2,343
3,669
5-2
-------
EPA-OGWDW Contaminant Occurrence Support Document for 2" 6-Year Review Category 1 Contaminants October 2009
Year
2004
2005
2006
On-Site Releases (pounds)
Total Air
Emissions
16
16
7
Surface Water
Discharges
35
1
0
Underground
Injection
0
0
0
Releases
to Land
3,928
4,355
6,321
Off-Site
Releases
(pounds)
22
811
164
Total On- &
Off-site
Releases
(pounds)
4,001
5,184
6,493
Source: USEPA, 2008
1. There is a hyphen in this cell because the facility that reported releases left that particular cell blank in its data
submission (a zero in a cell denotes either that the facility reported "0" or "NA" in its data submission).
Exhibit 5-1 b: Graph of Toxics Release Inventory (TRI) Data for Chlordane
Environmental Releases of Chlordane
30,000
25,000
o 20,000
(0
03
I 15,000
•o 10,000
Ł 5,000
n
n__ n n n n. IL
1998
1999
2000
2001
2002
Year
2003
2004
2005
2006
lAir Emissions DSurface Water Discharges D Underground Injection DReleases to Land BOff-site Releases
Source: USEPA, 2008
5.3 Occurrence in Ambient Water
No National Water Quality Assessment (NAWQA) Program data are available from the
United States Geological Survey (USGS) on the occurrence of chlordane in ambient waters from
the Pesticide National Synthesis Project (Gilliom et a/., 2007) or the EPA Summary Analysis of
Cycle 1 data.
5.4 Occurrence in Drinking Water
Chlordane is regulated as a synthetic organic chemical (SOC) in drinking water. All non-
purchased community water systems (CWSs) and non-transient non-community water systems
(NTNCWSs) are required to sample for SOCs. The maximum waiver period for SOCs is three
years, but this waiver can be renewed indefinitely. (Systems may receive a waiver at any time
5-3
-------
EPA-OGWDW Contaminant Occurrence Support Document for 2" 6-Year Review Category 1 Contaminants October 2009
without having collected an SOC sample. With a waiver, systems are not required to collect an
SOC sample.)
All CWSs and NTNCWSs without an SOC waiver must collect four consecutive
quarterly samples during the initial three-year compliance period.1 If all four samples are non-
detections, then a system serving less than 3,300 persons may reduce to one sample during each
consecutive compliance period; a system serving more than 3,300 persons may reduce to two
quarterly samples within a 12-month period during a 3-year compliance period. If a chemical is
detected (but is less than the MCL), the system must take one sample per quarter until results are
below the MCL (minimum of two quarterly samples for ground water systems and four quarterly
samples for surface water systems). If all quarterly samples are below the MCL, the system may
return to annual sampling. If a chemical is detected at a level greater than the MCL, the system
(whether ground water or surface water) must take quarterly samples until four consecutive
quarters are below the MCL. If all quarterly samples are below the MCL, the system may return
to annual sampling.
The analysis of chlordane occurrence presented in the following section is based on state
compliance monitoring data from the Six-Year Review-ICR Dataset. This is the largest and most
comprehensive compliance monitoring data set compiled by EPA's Drinking Water program to
date.
Forty-four of the Six-Year Review-ICR states provided occurrence data for chlordane.
(There were no chlordane data from Oklahoma.) These data consist of 128,453 analytical results
from 31,841 public water systems (PWSs) during the period from 1998 to 2005. The number of
sample results and systems vary by state, although the state data sets have been reviewed and
checked to ensure adequacy of coverage and completeness. The national modal minimum
reporting level (MRL) for chlordane in the dataset is 0.2 |ig/L.
5.4.1 Stage 1A nalysis of Occurrence Data
Stage 1 analyses have been performed at the sample, sample point, and system levels.
Note that some systems have multiple sample points, such as source water sample points or entry
points to the distribution system. Exhibit 5-2 presents the number of samples, systems, and
sample points in the data set, as well as the population-served by the systems and sample points,
and the percentage of detections and the percentage of the population-served by the systems and
sample points with detections. Detections in systems provide a fundamental indication of
national occurrence, but are not representative of MCL violations. For chlordane, 0.210% of
31,841 systems reported detections. Exhibit 5-3 lists the minimum, median, 90th percentile, and
maximum chlordane concentrations based on detections from all systems in the Six-Year
Review-ICR Dataset. The median concentration of chlordane, based on detections from all
systems, was 0.38 |ig/L.
1 All new systems or systems using a new water source that began operation after January 22, 2004 must demonstrate
compliance with the MCL within a period of time specified by the State. The system must also comply with the
initial sampling frequencies specified by the State to ensure that a system can demonstrate compliance with the MCL.
5-4
-------
EPA-OGWDW Contaminant Occurrence Support Document for 2" 6-Year Review Category 1 Contaminants October 2009
Exhibit 5-2: Chlordane Occurrence Data from the Six-Year Review-ICR Dataset
Source Water Type
Number of
Samples
Number of
Systems
Number of
Sample Points
Pop. Served
by Systems
Pop. Served
by Sample
Points
Total Number
Ground Water
101,389
28,808
45,986
66,936,780
66,936,780
Surface Water
27,064
3,033
6,686
115,100,195
115,100,195
Combined Ground &
Surface Water
128,453
31,841
52,672
182,036,975
182,036,975
Percent with Detections
Ground Water
0.152%
0.205%
0.139%
4.124%
3.984%
Surface Water
0.059%
0.264%
0.194%
0.571%
0.374%
Combined Ground &
Surface Water
0.132%
0.210%
0.146%
1.878%
1.701%
Exhibit 5-3: Chlordane Summary Statistics
Source Water Type
Ground Water
Detection Value (ug/L)
Minimum
0.005
Median
0.4
90tn
Percentile
1.6
Maximum
5.4
Surface Water
0.02
0.1
0.5
0.9
Combined Ground &
Surface Water
0.005
0.38
1.59
5.4
The following maps illustrate the geographic distribution of chlordane occurrences in
drinking water. Exhibit 5-4 shows the states with systems with at least one detection and Exhibit
5-5 shows the states with systems with at least one detection greater than the MCL concentration.
Note that five states did not submit data for use in the Six-Year Review (Kansas2, Louisiana,
Mississippi, Pennsylvania, and Washington). Oklahoma did submit Six-Year data for most
contaminants; however, there is a statewide waiver for chlordane in Oklahoma so no chlordane
data were available from that state.
2 Kansas submitted data after the data management and analyses had been conducted; data from Kansas are not
included in the analyses.
5-5
-------
EPA-OGWDW Contaminant Occurrence Support Document for 2" 6-Year Review Category 1 Contaminants October 2009
The distribution of systems with detections of chlordane is geographically dispersed.
Detection rates were low overall, with a few states on the eastern seaboard and in the southeast
having somewhat higher detection rates (0.5% to 1% of their systems). Two states had detections
greater than the MCL concentration of 2 |ig/L.
Exhibit 5-4: Percentage of Systems with at Least One Chlordane Detection, by
State
\
States with Waivers or Not Submitting
States with No Detections
Loss than 0.5% of Sysloms detecting
| Between 0.5% and 1% of Systems Detecting
j More than 1% of Systems Detecting
5-6
-------
EPA-OGWDW Contaminant Occurrence Support Document for 2" 6-Year Review Category 1 Contaminants October 2009
Exhibit 5-5: Percentage of Systems with at Least One Chlordane Detection
Greater than the MCL Concentration (> 2 ug/L), by State
States with Waivers or Not Submitting
States with No Detection* > MCL
Less than 0,1% of Systems with Detection* > MCL
M Between 0.1% and 0.5% of Systems with Detections > MCL
^B More Uian 0.5% of Systems with Detections > MCL
Exhibit 5-6 presents the Stage I analysis of chlordane occurrences in drinking water from
systems and the population-served by those systems in the Six-Year Review-ICR Dataset relative
to two thresholds: 2 |ig/L and 1 |ig/L. Twelve (0.038% of) systems, serving about 1,600 persons,
reported at least one detection greater than or equal to 1 u.g/L; four systems (0.013% of all
systems), serving 625 persons, reported at least one detection greater than 2 |ig/L. All detections
of chlordane greater than or equal to 1 |ig/L and greater than 2 |ig/L were in ground water
systems.
Exhibit 5-6: Chlordane Stage 1 Analysis - Systems and Population with at Least
One Threshold Exceedance
Source
Water Type
Ground
Water
Threshold
> 2 ug/L
> 1 ug/L
N umber of
Systems
Exceeding
Threshold
4
12
Percent of
Systems
Exceeding
Threshold
0.014%
0.042%
Total Population -
Served by
Systems
Exceeding
Threshold
625
1,645
Percent of
Population-Served
by Systems
Exceeding
Threshold
0.001%
0.002%
Surface
Water
> 2 ug/L
* 1 ug/L
0
0
0.000%
0.000%
0
0
0.000%
0.000%
5-7
-------
EPA-OGWDW Contaminant Occurrence Support Document for 2" 6-Year Review Category 1 Contaminants October 2009
Source
Water Type
Combined
Ground &
Surface
Water
Threshold
> 2 ug/L
> 1 ug/L
N umber of
Systems
Exceeding
Threshold
4
12
Percent of
Systems
Exceeding
Threshold
0.013%
0.038%
Total Population -
Served by
Systems
Exceeding
Threshold
625
1,645
Percent of
Population-Served
by Systems
Exceeding
Threshold
0.0003%
0.001%
The occurrence estimates presented above (in Exhibit 5-6) are based on the conservative
assumption that if a chlordane detection is found in a single entry point/sample point in a system,
then the entire population-served by the system is assumed to potentially be exposed to chlordane
(even though other entry points/sampling points at the system, and the population-served by
those entry/sample points, may exhibit no occurrence of chlordane). The sample point analysis is
a less conservative estimate of the population-served by a system with a detection of chlordane
because it estimates the potentially exposed population-served based only on the proportion of
the population-served by entry points/sample points with detections of chlordane (rather than the
entire population-served by the system).
Exhibit 5-7 presents the Stage 1 analysis of chlordane occurrences in drinking water from
sample points and population-served by those sample points in the Six-Year Review-ICR
Dataset. Sixteen (0.030% of) sample points, serving an estimated 1,053 persons, had at least one
result greater than or equal to 1 u.g/L; and six (0.011% of) sample points, serving an estimated
433 persons, had at least one result exceeding 2 u.g/L. Similar to the system-level occurrence
analysis, all detections of chlordane greater than or equal to 1 u.g/L and greater than 2 u.g/L were
in sample points from ground water systems.
Exhibit 5-7: Chlordane Stage 1 Analysis - Sample Points and Population with at
Least One Threshold Exceedance
Source
Water Type
Ground
Water
Threshold
> 2 ug/L
> 1 ug/L
Number of
Sample Points
Exceeding
Threshold
6
16
Percent of
Sample Points
Exceeding
Threshold
0.013%
0.035%
Total Population-
Served by Sample
Points Exceeding
Threshold
433
1,053
Percent of
Population-
Served by
Sample Points
Exceeding
Threshold
0.001%
0.002%
Surface
Water
> 2 ug/L
> 1 ug/L
0
0
0.000%
0.000%
0
0
0.000%
0.000%
5-8
-------
EPA-OGWDW Contaminant Occurrence Support Document for 2" 6-Year Review Category 1 Contaminants October 2009
Source
Water Type
Combined
Ground &
Surface
Water
Threshold
> 2 |jg/L
> 1 |jg/L
Number of
Sample Points
Exceeding
Threshold
6
16
Percent of
Sample Points
Exceeding
Threshold
0.011%
0.030%
Total Population-
Served by Sample
Points Exceeding
Threshold
433
1,053
Percent of
Population-
Served by
Sample Points
Exceeding
Threshold
0.0002%
0.001%
5.4.2 Stage 2 A nalysis of Occurrence Data
For the Stage 2 analyses, three different substitution values -zero, one-half the MRL
value, and the full MRL value- were used to replace each non-detection record. Three arithmetic
mean chlordane concentrations were calculated at each system using the zero, one-half MRL and
full MRL substitution values. These mean calculations were done for all systems with data in the
Six-Year Review-ICR dataset. Then the percentages of all systems with a mean concentration
greater than each threshold were calculated.
Exhibit 5-8 presents the Stage 2 analysis of estimated system mean concentrations for
chlordane in drinking water relative to two thresholds: 2 |ig/L and 1 |ig/L. Exhibit 5-9 presents
similar information based on population-served by the systems. Based on the Stage 2 analyses,
no more than two ground water systems (approximately 0.006% of all systems) had estimated
system means greater than or equal to 1 |ig/L. These two systems serve about 116 persons. One
ground water system (0.003%), serving about 80 persons, had estimated system means greater
than 2 |ig/L. No surface water systems had estimated means greater than either threshold.
Exhibit 5-8: Chlordane Stage 2 Analysis - Systems with a System Mean Threshold
Exceedance
Source
Water Type
Ground
Water
Threshold
> 2 |jg/L
> 1 |jg/L
Number of Systems with Mean
Exceeding Threshold1
Full MRL
1
2
1/2 MRL
1
2
Zero
1
1
Percent of Systems with Mean
Exceeding Threshold1
Full MRL
0.003%
0.007%
1/2 MRL
0.003%
0.007%
Zero
0.003%
0.003%
Surface
Water
> 2 |jg/L
> 1 H9/L
0
0
0
0
0
0
0.000%
0.000%
0.000%
0.000%
0.000%
0.000%
5-9
-------
EPA-OGWDW Contaminant Occurrence Support Document for 2" 6-Year Review Category 1 Contaminants October 2009
Source
Water Type
Combined
Ground &
Surface
Water
Threshold
> 2 |jg/L
> 1 |jg/L
Number of Systems with Mean
Exceeding Threshold1
Full MRL
1
2
1/2 MRL
1
2
Zero
1
1
Percent of Systems with Mean
Exceeding Threshold1
Full MRL
0.003%
0.006%
1/2 MRL
0.003%
0.006%
Zero
0.003%
0.003%
1. The results shown in this table are based on system arithmetic means that were calculated using all sample
detection records and all non-detection records, with the non-detection records replaced (substituted) with
concentration values equal to the full MRL, 1A MRL, and zero.
Exhibit 5-9: Chlordane Stage 2 Analysis - Population-Served by Systems with a
System Mean Threshold Exceedance
Source
Water Type
Ground
Water
Threshold
> 2 |jg/L
> 1 H9/L
Population -Served by Systems
with Mean Exceeding Threshold1
Full MRL
80
116
1/2 MRL
80
116
Zero
80
80
Percent of Population -Served by
Systems with Mean Exceeding
Threshold1
Full MRL
0.0001%
0.0002%
1/2 MRL
0.0001%
0.0002%
Zero
0.0001%
0.0001%
Surface
Water
> 2 |jg/L
> 1 H9/L
0
0
0
0
0
0
0.000%
0.000%
0.000%
0.000%
0.000%
0.000%
Combined
Ground &
Surface
Water
> 2 |jg/L
> 1 pg/L
80
116
80
116
80
80
0.00004%
0.0001%
0.00004%
0.0001%
0.00004%
0.00004%
1. The results shown in this table are based on system arithmetic means that were calculated using all sample
detection records and all non-detection records, with the non-detection records replaced (substituted) with
concentration values equal to the full MRL, 14 MRL, and zero.
5.5 Additional Occurrence Data
First Six-Year Review ofNPDWRs
Similar types of Stage 1 contaminant occurrence analyses were conducted for the first
Six-Year Review. The first Six-Year Review assessed regulated contaminant occurrence data
collected from 1993 to 1997. The first Six-Year Review data set consisted of data from 16 states
that were assembled into a "national cross-section" that was qualitatively indicative (though not
statistically representative) of national occurrence. For the first Six-Year Review (USEPA,
2003), 157 systems (approximately 1.19% of about 13,000 systems), reported at least one
detection of chlordane. One system (approximately 0.01% of all systems), reported at least one
detection greater than the MCL concentration.
5-10
-------
EPA-OGWDW Contaminant Occurrence Support Document for 2" 6-Year Review Category 1 Contaminants October 2009
5.6 Summary of Data
TRI data indicate that the total off-site and on-site releases of chlordane for the period
1998-2006 ranged from about 50 pounds to about 46,000 pounds; most years had total releases
between approximately 2,000 and 9,000 pounds. Releases to land constituted the majority of
releases. Surface water discharges and air emissions were much less than releases to land, and
underground injection was non-existent except for 1998.
No data are available from the NAWQA program on chlordane in ambient waters.
The Stage 1 analysis of occurrences in drinking water indicates that chlordane was
detected in 0.210% of 31,841 surface water and ground water systems. The median
concentration, based on detections from all systems, was 0.38 |ig/L. In total, 0.013% of systems
(serving 0.0003% of the total population) reported at least one detection greater than the MCL
concentration of 2 |ig/L. No surface water systems reported detections greater than the MCL
concentration. A similar pattern was seen for ground water and surface water sample points.
In the Stage 2 analysis, 0.003% of all water systems (serving 0.00004% of the total
population) had an estimated system mean concentration of chlordane greater than the MCL
concentration. There were no system mean threshold exceedances for surface water. Only one
ground water system (serving 80 persons) had a mean exceeding the MCL concentration.
5.7 References
Agency for Toxic Substances and Disease Registry (ATSDR). 1994. Toxicological Profile for
Chlordane. May 1994. Available on the Internet at:
http://www.atsdr.cdc.gov/toxprofiles/tp31 .html.
Gilliom, R.J., I.E. Barbash, C.G. Crawford, P.A. Hamilton, J.D. Martin, N. Nakagaki, L.H.
Nowell, J.C. Scott, P.E. Stackelberg, G.P. Thelin, and D.M. Wolock. 2007. The Quality of Our
Nation's Waters - Pesticides in the Nation's Streams and Ground Water, 1992-2001. Appendix
7. Statistical summaries of water-quality data. U.S. Geological Survey Circular 1291, 172 p.
Available on the Internet at: http://water.usgs.gov/nawqa/pnsp/pubs/circ 129l/appendix7/.
United States Environmental Protection Agency (USEPA). 1991. National Primary Drinking
Water Regulations—Synthetic Organic Chemicals and Inorganic Chemicals; Monitoring for
Unregulated Contaminants; National Primary Drinking Water Regulations Implementation;
National Secondary Drinking Water Regulations; Final Rule Federal Register. Vol. 56, No. 30. p.
3526, January 30, 1991.
USEPA. 2003. Occurrence Estimation Methodology and Occurrence Findings for Six-Year
Review of National Primary Drinking Water Regulations. Office of Water. EPA-815-R-03-006.
June 2003.
5-11
-------
EPA-OGWDW Contaminant Occurrence Support Document for 2" 6-Year Review Category 1 Contaminants October 2009
USEPA. 2008. TRI Explorer: Trends - Chlordane. Released February 21, 2008b. Available on
the Internet at: http://www.epa.gov/triexplorer/trends.htm. [Search for chlordane.] Accessed
February 29, 2008.
USEPA. 2009. The Analysis of Regulated Contaminant Occurrence Data from Public Water
Systems in Support of the Second Six-Year Review of National Primary Drinking Water
Regulations. Office of Water. EPA-815-B-09-006. October 2009.
5-12
-------
EPA-OGWDW Contaminant Occurrence Support Document for 2" 6-Year Review Category 1 Contaminants October 2009
6 l,2-Dibromo-3-chloropropane (DBCP)
This chapter on l,2-dibromo-3-chloropropane (DBCP) is part of a report that is organized
so that each chapter presents information on one specific contaminant. Chapter 1 of this report,
Introduction, presents background information on all sources of data used as well as the
analytical approach used to estimate contaminant occurrence in drinking water. This contaminant
chapter includes background information (such as regulatory history, general chemical
information, and environmental fate and behavior), use and environmental release information,
and DBCP occurrence estimates in ambient water and drinking water. All drinking water
occurrence estimates were based on data from the National Compliance Monitoring Information
Collection Request (ICR) Dataset for the Second Six-Year Review (the "Six Year Review-ICR
Dataset"). For detailed information regarding this dataset, data management, data quality
assessments, and the analytical approach used to generate all contaminant occurrence estimates
presented in this chapter, please refer to USEPA (2009).
6.1 Background
The United States Environmental Protection Agency (EPA) published the current
National Primary Drinking Water Regulations (NPDWR) for DBCP on January 30, 1991 (56 FR
3526 (USEPA, 1991)). The NPDWR established a maximum contaminant level goal (MCLG) of
zero based on a cancer classification of B2, probable human carcinogen. The NPDWR also
established a maximum contaminant level (MCL) of 0.2 |ig/L based on analytical feasibility.
According to the Agency for Toxic Substances and Disease Registry (ATSDR, 1992),
DBCP (C3H5Br2Cl) is a colorless liquid with a sharp smell. It does not evaporate quickly and is
only weakly soluble in water.
DBCP is not usually found in the environment. It was once produced in large amounts as
a pesticide, but this use has been discontinued in the United States since the 1980s. Some
industries still use DBCP to make fire-resistant substances. Because production is very limited, it
is difficult to determine exactly how much is being used currently. DBCP is sometimes found in
the soil and ground water near cropland where it has been used. It may also enter the
environment during production or use in industry and research (ATSDR, 1992).
When DBCP is deposited on soil or released to surface water, much of it evaporates to
the air, where it breaks down over the course of months. DBCP applied to soil may also penetrate
downward into deeper soils and ground water. In such cases it may take a number of years for the
chemical to break down. Based on the limited usage in the past 5 to 10 years, it is expected that
levels where the chemical has not been used or discarded are either low or nonexistent (ATSDR,
1992).
6.2 Use, Production, and Releases
Until 1977, DBCP was used extensively as a soil fumigant and nematocide on over 40
different crops in efforts to protect field crops, vegetables, fruits and nuts, nursery and
6-1
-------
EPA-OGWDW Contaminant Occurrence Support Document for 2" 6-Year Review Category 1 Contaminants October 2009
greenhouse crops, and turf from pests. From 1977 to 1979, EPA suspended registration for all
DBCP-containing products except for use on pineapples in Hawaii. In 1985, EPA canceled all
registrations for DBCP, including use on pineapples (ATSDR, 1992).
The Chemical Substance Inventory Update System contains confidential data reported by
industry as a partial update of the Toxic Substances Control Act (TSCA) Inventory. The TSCA
Chemical Substance Inventory does not report production data for DBCP.
DBCP is listed as a Toxics Release Inventory (TRI) chemical. Exhibit 6-la lists the
environmental releases of DBCP from 1998 to 2006. (Exhibit 6-lb provides a graphical
presentation of the environmental releases.) Releases were negligible until 2001. Almost all
DBCP releases were off-site. In isolated instances, negligible quantities were released to the air.
No DBCP was reported as being injected underground, or as being released to the land or surface
water. Arkansas and Ohio are the only two states that reported any releases of DBCP for the
period of record. (Ohio reported releases for all years. Arkansas reported releases for all years
except 1998.)
Exhibit 6-1 a: Table of Toxics Release Inventory (TRI) Data for DBCP
Year
19981
1999
2000
2001
2002
2003
2004
2005
2006
On-Site Releases (pounds)
Total Air
Emissions
0
6
6
1
0
0
2
0
Surface Water
Discharges
0
0
0
0
0
0
0
0
Underground
Injection
0
0
0
0
0
0
0
0
Releases
to Land
0
0
0
0
0
0
0
0
Off-Site
Releases
(pounds)
6
10
505
255
250
255
255
255
Total On- &
Off -site
Releases
(pounds)
0
6
16
511
256
250
255
257
255
Source: USEPA, 2008
1. There are decimal points in this row because the facility certified that its total annual reportable amount is less
than 500 pounds, and does not manufacture, process, or otherwise use more than 1 million pounds.
6-2
-------
EPA-OGWDW Contaminant Occurrence Support Document for 2" 6-Year Review Category 1 Contaminants October 2009
Exhibit 6-1 b: Graph of Toxics Release Inventory (TRI) Data for DBCP
Environmental Releases of DBCP
•n
(0
03
-------
EPA-OGWDW Contaminant Occurrence Support Document for 2" 6-Year Review Category 1 Contaminants October 2009
Exhibit 6-2: EPA Summary Analysis of 1,2-Dibromo-3-chloropropane (DBCP) Data
from NAWQA Study Units, 1992-2001
ground
water
surface
water
all
sites
Detection Frequency
(detections are results > RL1)
Number
of
Samples
4,920
1,393
6,313
% Samples
with
Detections
2.2%
0.2%
1.7%
Number
of Sites
4,148
185
4,333
% Sites
with
Detections
1.4%
1 .6%
1.4%
Concentration Values
(of detections, in ug/L)
Minimum
0.0252
0.09
0.0252
Median
1.2885
0.1
1.2
95th
Percen-
tile
3.2
0.11
3.2
99th
Percen-
tile
4
0.11
4
Maximum
6.4
0.11
6.4
1. RLs (Reporting Limits) for DBCP varied, but did not exceed 0.03 ug/L.
6.4 Occurrence in Drinking Water
DBCP is regulated as a synthetic organic chemical (SOC) in drinking water. All non-
purchased community water systems (CWSs) and non-transient non-community water systems
(NTNCWSs) are required to sample for SOCs. The maximum waiver period for SOCs is three
years, but this waiver can be renewed indefinitely. (Systems may receive a waiver at any time
without having collected an SOC sample. With a waiver, systems are not required to collect an
SOC sample.) Statewide waivers are often granted for DBCP.
All CWSs and NTNCWSs without an SOC waiver must collect four consecutive
quarterly samples during the initial three-year compliance period.1 If all four samples are non-
detections, then a system serving less than 3,300 persons may reduce to one sample during each
consecutive compliance period; a system serving more than 3,300 persons may reduce to two
quarterly samples within a 12-month period during a 3-year compliance period. If a chemical is
detected (but is less than the MCL), the system must take one sample per quarter until results are
below the MCL (minimum of two quarterly samples for ground water systems and four quarterly
samples for surface water systems). If all quarterly samples are below the MCL, the system may
return to annual sampling. If a chemical is detected at a level greater than the MCL, the system
(whether ground water or surface water) must take quarterly samples until four consecutive
quarters are below the MCL. If all quarterly samples are below the MCL, the system may return
to annual sampling.
The analysis of DBCP occurrence presented in the following section is based on state
compliance monitoring data from the Six-Year Review-ICR Dataset. This is the largest and most
comprehensive compliance monitoring data set compiled by EPA's Drinking Water program to
date.
All new systems or systems using a new water source that began operation after January 22, 2004 must demonstrate
compliance with the MCL within a period of time specified by the State. The system must also comply with the
initial sampling frequencies specified by the State to ensure that a system can demonstrate compliance with the MCL.
6-4
-------
EPA-OGWDW Contaminant Occurrence Support Document for 2" 6-Year Review Category 1 Contaminants October 2009
Forty-four of the Six-Year Review-ICR states provided occurrence data for DBCP.
(Oklahoma provided no DBCP data.) These data consist of 231,169 analytical results from
37,618 public water systems (PWSs) during the period from 1998 to 2005. The number of
sample results and systems vary by state, although the state data sets have been reviewed and
checked to ensure adequacy of coverage and completeness. The national modal minimum
reporting level (MRL) for DBCP in the dataset is 0.02 |ig/L.
6.4.1 Stage 1A nalysis of Occurrence Data
Stage 1 analyses have been performed at the sample, sample point, and system levels.
Note that some systems have multiple sample points, such as source water sample points or entry
points to the distribution system. Exhibit 6-3 presents the number of samples, systems, and
sample points in the data set, as well as the population-served by the systems and sample points,
and the percentage of detections and the percentage of the population-served by the systems and
sample points with detections. Detections in systems provide a fundamental indication of
national occurrence, but are not representative of MCL violations. For DBCP, 1.029% of 37,618
systems reported detections. Exhibit 6-4 lists the minimum, median, 90th percentile, and
maximum DBCP concentrations based on detections from all systems in the Six-Year Review-
ICR Dataset. The median concentration of DBCP, based on detections from all systems, was 0.3
Exhibit 6-3: DBCP Occurrence Data from the Six-Year Review-ICR Dataset
Source Water Type
Number of
Samples
Number of
Systems
Number of
Sample Points
Pop. Served
by Systems
Pop. Served
by Sample
Points
Total Number
Ground Water
178,958
34,602
59,021
73,552,992
73,552,992
Surface Water
52,211
3,016
7,762
120,196,347
120,196,347
Combined Ground &
Surface Water
231,169
37,618
66,783
193,749,339
193,749,339
Percent with Detections
Ground Water
4.675%
0.922%
1.166%
9.820%
6.454%
Surface Water
21.806%
2.255%
5.037%
8.929%
4.555%
Combined Ground &
Surface Water
8.544%
1.029%
1.616%
9.268%
5.276%
6-5
-------
EPA-OGWDW Contaminant Occurrence Support Document for 2" 6-Year Review Category 1 Contaminants October 2009
Exhibit 6-4: DBCP Summary Statistics
Source Water Type
Ground Water
Detection Value (ug/L)
Minimum
0.02
Median
0.3
90tn
Percentile
2
Maximum
70
Surface Water
0.009
0.29
1.1
230
Combined Ground &
Surface Water
0.009
0.3
1.5
230
The following maps illustrate the geographic distribution of DBCP occurrence in drinking
water. Exhibit 6-5 shows the states with systems with at least one detection and Exhibit 6-6
shows the states with systems with at least one detection greater than the MCL concentration.
Note that five states did not submit data for use in the Six-Year Review (Kansas2, Louisiana,
Mississippi, Pennsylvania, and Washington). Oklahoma did submit Six-Year data for most
contaminants; however, there is a statewide waiver for DBCP in Oklahoma so no DBCP data
were available from that state.
The distribution of systems with detections of DBCP is geographically dispersed.
Detection rates were generally low, with only three states having greater than 5% of their systems
detecting DBCP. Fourteen states reported at least one detection greater than the MCL
concentration of 0.2 |ig/L, although only 9 of these states had more than 0.5% of systems with at
least one detection greater than the MCL concentration.
2 Kansas submitted data after the data management and analyses had been conducted; data from Kansas are not
included in the analyses.
6-6
-------
EPA-OGWDW Contaminant Occurrence Support Document for 2" 6-Year Review Category 1 Contaminants October 2009
Exhibit 6-5: Percentage of Systems with at Least One DBCP Detection, by State
States with Waivers or Not Submitting
States with No Detections
Less than 1% of Systems Detecting
Between 1% and 5% of Systems Detecting
More than 5% of Systems Detecting
Exhibit 6-6: Percentage of Systems with at Least One DBCP Detection Greater
than the MCL Concentration (> 0.2 vglL), by State
State* with Waivers or Not Submitting
States with No Detections > MCL
BLe*S than 0.1% Of Systems with Detections > MCL
Between 0.1 % and 0.5% of Systems with Detections > MCL
g More tfian 0.5% of Systems with Detections > MCL
Exhibit 6-7 presents the Stage 1 analysis of DBCP occurrence in drinking water from
systems and the population-served by those systems in the Six-Year Review-ICR Dataset relative
to two thresholds: 0.2 |ig/L and 0.1 |ig/L. A total of 201 (0.534% of) systems, serving about 7
5-7
-------
EPA-OGWDW Contaminant Occurrence Support Document for 2" 6-Year Review Category 1 Contaminants October 2009
million persons, reported at least one detection greater than or equal to 0.1 |ig/L; and 146
(0.388% of) systems, serving 3 million persons, reported at least one detection greater than 0.2
|ig/L. Detections greater than the thresholds occurred in more ground water than surface water
systems. However, because surface water systems tend to be larger, more people were served by
surface water systems reporting at least one detection at greater than threshold values.
Exhibit 6-7: DBCP Stage 1 Analysis - Systems and Population with at Least One
Threshold Exceedance
Source
Water Type
Ground
Water
Threshold
> 0.2 |jg/L
>0.1 |jg/L
Number of
Systems
Exceeding
Threshold
120
168
Percent of
Systems
Exceeding
Threshold
0.347%
0.486%
Total Population-
Served by
Systems
Exceeding
Threshold
1,032,709
1,370,317
Percent of
Population-
Served by
Systems
Exceeding
Threshold
1 .404%
1.863%
Surface
Water
> 0.2 |jg/L
>0.1 |jg/L
26
33
0.862%
1.094%
2,118,782
5,730,210
1.763%
4.767%
Combined
Ground &
Surface
Water
> 0.2 |jg/L
>0.1 |jg/L
146
201
0.388%
0.534%
3,151,491
7,100,527
1.627%
3.665%
The occurrence estimates presented above (in Exhibit 6-7) are based on the conservative
assumption that if a DBCP detection is found in a single entry point/sample point in a system,
then the entire population-served by the system is assumed to potentially be exposed to DBCP
(even though other entry points/sampling points at the system, and the population-served by
those entry/sample points, may exhibit no occurrence of DBCP). The sample point analysis is a
less conservative estimate of the population-served by a system with a detection of DBCP
because it estimates the potentially exposed population-served based only on the proportion of
the population-served by entry points/sample points with detections of DBCP (rather than the
entire population-served by the system).
Exhibit 6-8 presents the Stage 1 analysis of DBCP occurrence in drinking water from
sample points and population-served by those sample points in the Six-Year Review-ICR
Dataset. A total of 540 (0.809% of) sample points, serving an estimated 3.8 million persons,
reported at least one detection greater than or equal to 0.1 |ig/L; and 346 (0.518% of) sample
points, serving an estimated 818,000 persons, reported at least one detection exceeding 0.2 |ig/L.
-------
EPA-OGWDW Contaminant Occurrence Support Document for 2" 6-Year Review Category 1 Contaminants October 2009
Exhibit 6-8: DBCP Stage 1 Analysis - Sample Points and Population with at Least
One Threshold Exceedance
Source
Water Type
Ground
Water
Threshold
> 0.2 |jg/L
>0.1 |jg/L
Number of
Sample Points
Exceeding
Threshold
199
318
Percent of
Sample Points
Exceeding
Threshold
0.337%
0.539%
Total Population-
Served by Sample
Points Exceeding
Threshold
217,321
394,293
Percent of
Population-
Served by
Sample Points
Exceeding
Threshold
0.295%
0.536%
Surface
Water
> 0.2 |jg/L
>0.1 |jg/L
147
222
1.894%
2.860%
600,798
3,365,848
0.500%
2.800%
Combined
Ground &
Surface
Water
> 0.2 |jg/L
>0.1 |jg/L
346
540
0.518%
0.809%
818,119
3,760,140
0.422%
1.941%
6.4.2 Stage 2 A nalysis of Occurrence Data
For the Stage 2 analyses, three different substitution values -zero, one-half the MRL
value, and the full MRL value- were used to replace each non-detection record. Three arithmetic
mean DBCP concentrations were calculated at each system using the zero, one-half MRL and full
MRL substitution values. These mean calculations were done for all systems with data in the Six-
Year Review-ICR dataset. Then the percentages of all systems with a mean concentration greater
than the each threshold were calculated.
Exhibit 6-9 presents the Stage 2 analysis of estimated system mean concentrations for
DBCP occurrence in drinking water relative to two thresholds: 0.2 |ig/L and 0.1 |ig/L. Exhibit 6-
10 presents similar information based on population-served by the systems. Based on the Stage 2
analyses, no more than 97 ground water and surface water systems (approximately 0.258%) had
an estimated system mean greater than or equal to 0.1 |ig/L. These 97 systems serve about 1.4
million persons. A total of 42 systems (approximately 0.112% of all systems), serving 25,000
persons, had an estimated system mean greater than 0.2 |ig/L. Similar to the Stage 1 analysis,
more ground water than surface water systems had exceedances of both thresholds. However,
nearly the same number of people was served by ground water and surface water systems with
mean concentrations greater than 0.2 |ig/L.
6-9
-------
EPA-OGWDW Contaminant Occurrence Support Document for 2" 6-Year Review Category 1 Contaminants October 2009
Exhibit 6-9: DBCP Stage 2 Analysis - Systems with a System Mean Threshold
Exceedance
Source
Water Type
Ground
Water
Threshold
> 0.2 |jg/L
>0.1 |jg/L
Number of Systems with Mean
Exceeding Threshold1
Full MRL
38
85
1/2 MRL
38
82
Zero
38
81
Percent of Systems with Mean
Exceeding Threshold1
Full MRL
0.110%
0.246%
1/2 MRL
0.110%
0.237%
Zero
0.110%
0.234%
Surface
Water
> 0.2 |jg/L
>0.1 |jg/L
4
12
4
11
4
11
0.133%
0.398%
0.133%
0.365%
0.133%
0.365%
Combined
Ground &
Surface
Water
> 0.2 |jg/L
>0.1 |jg/L
42
97
42
93
42
92
0.112%
0.258%
0.112%
0.247%
0.112%
0.245%
1 The results shown in this table are based on system arithmetic means that were calculated using all sample
detection records and all non-detection records, with the non-detection records replaced (substituted) with
concentration values equal to the full MRL, 14 MRL, and zero.
Exhibit 6-10: DBCP Stage 2 Analysis - Population-Served by Systems with a
System Mean Threshold Exceedance
Source
Water Type
Ground
Water
Threshold
> 0.2 |jg/L
>0.1 |jg/L
Population -Served by Systems
with Mean Exceeding Threshold1
Full MRL
15,947
391,923
1/2 MRL
15,947
385,285
Zero
15,947
195,179
Percent of Population -Served by
Systems with Mean Exceeding
Threshold1
Full MRL
0.022%
0.533%
1/2 MRL
0.022%
0.524%
Zero
0.022%
0.265%
Surface
Water
> 0.2 |jg/L
>0.1 |jg/L
9,246
990,387
9,246
986,013
9,246
986,013
0.008%
0.824%
0.008%
0.820%
0.008%
0.820%
Combined
Ground &
Surface
Water
> 0.2 |jg/L
>0.1 |jg/L
25,193
1,382,310
25,193
1,371,298
25,193
1,181,192
0.013%
0.713%
0.013%
0.708%
0.013%
0.610%
1. The results shown in this table are based on system arithmetic means that were calculated using all sample
detection records and all non-detection records, with the non-detection records replaced (substituted) with
concentration values equal to the full MRL, 14 MRL, and zero.
6-10
-------
EPA-OGWDW Contaminant Occurrence Support Document for 2" 6-Year Review Category 1 Contaminants October 2009
6.5 Additional Occurrence Data
First Six-Year Review ofNPDWRs
Similar types of Stage 1 contaminant occurrence analyses were conducted for the first
Six-Year Review. The first Six-Year Review assessed regulated contaminant occurrence data
collected from 1993 to 1997. The first Six-Year Review data set consisted of data from 16 states
that were assembled into a "national cross-section" that was qualitatively indicative (though not
statistically representative) of national occurrence. For the first Six-Year Review (USEPA,
2003), 226 systems (approximately 1.61% of about 14,000 systems), reported at least one
detection of DBCP. One hundred and twenty-eight systems (approximately 0.91% of all
systems), reported at least one detection greater than the MCL concentration.
6.6 Summary of Data
TRI data indicate that total off-site and on-site release of DBCP for the period 1998-2006
were minimal and ranged from 0 to about 500 pounds. There were no releases to surface water,
to land, or through underground injection. Air emissions of DBCP were extremely low. DBCP
releases were almost exclusively off-site.
Ambient water data from the NAWQA program indicate that DBCP was detected in 1.7%
of 6,313 ground and surface water samples and at 1.4% of 4,333 sites. The median concentration,
based on detections from all sites, was 1.2 |ig/L.
The Stage 1 analysis of occurrence in drinking water indicates that DBCP was detected in
1.029% of 37,618 surface and ground water systems. The median concentration, based on
detections from all systems, was 0.09 |ig/L. In total, 0.388% of systems (serving 1.627% of the
total population) reported at least one detection greater than the MCL concentration of 0.2 |ig/L.
Detections are more common in ground water systems than surface water systems. However,
because surface water systems tend to be larger than ground water systems, a greater population
was served by surface water systems with at least one detection greater than the MCL
concentration. A similar pattern was seen for sample points.
In the Stage 2 analysis, 0.112% of all systems (serving 0.013% of the population) had an
estimated system mean concentration of DBCP greater than the MCL concentration. More
ground water systems had mean concentrations exceeding the MCL concentration, but the
population-served by ground water systems and surface water systems were similar.
6.7 References
Agency for Toxic Substances and Disease Registry (ATSDR). 1992. Toxicological Profile for
1,2-Dibromo-3-chloropropane. September 1992. Available on the Internet at:
http://www.atsdr.cdc.gov/toxprofiles/tp36.html.
6-11
-------
EPA-OGWDW Contaminant Occurrence Support Document for 2" 6-Year Review Category 1 Contaminants October 2009
United States Environmental Protection Agency (USEPA). 1991. National Primary Drinking
Water Regulations—Synthetic Organic Chemicals and Inorganic Chemicals; Monitoring for
Unregulated Contaminants; National Primary Drinking Water Regulations Implementation;
National Secondary Drinking Water Regulations; Final Rule Federal Register. Vol. 56, No. 30. p.
3526, January 30, 1991.
USEPA. 2003. Occurrence Estimation Methodology and Occurrence Findings for Six-Year
Review of National Primary Drinking Water Regulations. Office of Water. EPA-815-R-03-006.
June 2003.
USEPA. 2008. TRI Explorer: Trends - l,2-Dibromo-3-chloropropane. Released February 21,
2008b. Available on the Internet at: http://www.epa.gov/triexplorer/trends.htm. [Search for 1,2-
dibromo-3-chloropropane.] Accessed February 29, 2008.
USEPA. 2009. The Analysis of Regulated Contaminant Occurrence Data from Public Water
Systems in Support of the Second Six-Year Review of National Primary Drinking Water
Regulations. Office of Water. EPA-815-B-09-006. October 2009.
6-12
-------
EPA-OGWDW Contaminant Occurrence Support Document for 2" 6-Year Review Category 1 Contaminants October 2009
7 1,2-Dichloroethane
This chapter on 1,2-dichloroethane is part of a report that is organized so that each
chapter presents information on one specific contaminant. Chapter 1 of this report, Introduction,
presents background information on all sources of data used as well as the analytical approach
used to estimate contaminant occurrence in drinking water. This contaminant chapter includes
background information (such as regulatory history, general chemical information, and
environmental fate and behavior), use and environmental release information, and 1,2-
dichloroethane occurrence estimates in ambient water and drinking water. All drinking water
occurrence estimates were based on data from the National Compliance Monitoring Information
Collection Request (ICR) Dataset for the Second Six-Year Review (the "Six Year Review-ICR
Dataset"). For detailed information regarding this dataset, data management, data quality
assessments, and the analytical approach used to generate all contaminant occurrence estimates
presented in this chapter, please refer to USEPA (2009).
7.1 Background
The United States Environmental Protection Agency (EPA) published the current
National Primary Drinking Water Regulations (NPDWR) for 1,2-dichloroethane on July 8, 1987
(52 FR 25690 (USEPA, 1987)). The NPDWR established a maximum contaminant level goal
(MCLG) of zero based on a cancer classification of B2, probable human carcinogen. The
NPDWR also established a maximum contaminant level (MCL) of 5 jig/L based on analytical
feasibility.
According to the Agency for Toxic Substances and Disease Registry (ATSDR, 2001),
1,2-dichloroethane (CiELiCb) is a clear, manufactured liquid that evaporates quickly at room
temperature. It has a pleasant smell and sweet taste and burns with a smoky flame. Other names
for 1,2-dichloroethane are ethylene dichloride, dichloroethane, and EDC (ATSDR, 2001).
There are no known natural sources of 1,2-dichloroethane. The major routes of entry of
1,2-dichloroethane to the environment are consequences of industrial activity. Emissions to the
atmosphere comprise the largest fraction of all releases of 1,2-dichloroethane to the environment.
Once in the atmosphere, it can last more than five months before being broken down by sunlight.
During this time, it can travel large distances and be washed out by precipitation far from its
point of origin. It is believed that discharges to surface water and leaching from solid waste are
the primary causes of 1,2-dichloroethane contamination in drinking water. It is not known how
long 1,2-dichloroethane lasts in surface water; however, it has been observed to last longer in
lakes than in rivers. Once underground, it can travel long distances and contaminate wells
(ATSDR, 2001).
Human exposure to 1,2-dichloroethane typically occurs because of improper disposal, but
1,2-dichloroethane also has been detected in the air in urban and industrial areas at levels
between 0.003 and 1.5 ppb. There is also a small chance that humans might be exposed to 1,2-
dichloroethane through contact with old chemical products made with it, such as leaded gasoline,
cleaning agents, pesticides, and household adhesives (ATSDR, 2001).
7-1
-------
EPA-OGWDW Contaminant Occurrence Support Document for 2" 6-Year Review Category 1 Contaminants October 2009
7.2 Use, Production, and Releases
Although large amounts of 1,2-dichloroethane are produced today, most of it is used as a
chemical intermediate for vinyl chloride. This accounts for 98% of its usage. 1,2-Dichloroethane
is also used in smaller amounts to synthesize other organic compounds such as vinylidene
chloride, 1,1,1-trichloroethane, trichloroethylene, tetrachloroethylene, aziridines, ethylene
diamines, and various chlorinated solvents. In the past it was added to leaded gasoline to remove
the lead (ATSDR, 2001).
Some uses of 1,2-dichloroethane have been discontinued, especially in many consumer
products. It is no longer used in varnish and finish removers, soaps and scouring compounds,
metal degreasers, ore flotation, paints, coatings, or adhesives. Its use in organic synthesis for
extraction and cleaning purposes has also been discontinued. It was also formerly used as a grain,
household, and soil fumigant (ATSDR, 2001).
Every year 17.5 million tons of 1,2-dichloroethane are produced in the United States,
Western Europe, and Japan (Field and Sierra-Alvarez, 2004). The Toxic Substances Control Act
(TSCA) Inventory contains confidential data reported by industry as a partial update of the
Chemical Substance Inventory Update System. The data are reported under the Inventory Update
Reporting regulation. Manufacturers and importers are required to report company information
and chemical information, including production volumes for chemicals they manufactured or
imported in excess of 10,000 pounds in the immediately preceding fiscal year.
The TSCA Chemical Substance Inventory does not provide actual production volumes;
instead, it indicates at which of five production volume ranges a chemical was manufactured or
imported. The TSCA Inventory is updated every four years. Exhibit 7-1 presents the production
of 1,2-dichloroethane in the United States from 1986 to 2002. Note that the production of 1,2-
dichloroethane in the United States has remained in excess of 1 billion pounds since 1986.
Exhibit 7-1: Production of 1,2-Dichloroethane in the United States
Inventory Update Year
1986 Range
1990 Range
1994 Range
1998 Range
2002 Range
Production Volume
> 1 billion pounds
> 1 billion pounds
> 1 billion pounds
> 1 billion pounds
> 1 billion pounds
Source: USEPA, 2003a
7-2
-------
EPA-OGWDW Contaminant Occurrence Support Document for 2" 6-Year Review Category 1 Contaminants October 2009
1,2-Dichloroethane is listed as a Toxics Release Inventory (TRI) chemical. Exhibit 7-2a
lists the environmental releases of 1,2-dichloroethane from 1998 to 2006. (Exhibit 7-2b provides
a graphical presentation of the environmental releases.) In most years, on-site releases were
greater than off-site releases, except in 1999. Air emissions and underground injection accounted
for most on-site releases. Surface water discharges and releases to land were always well under
1% of total releases. The TRI data for 1,2-dichloroethane were reported from 30 states, plus
Puerto Rico, with 18 states and Puerto Rico reporting releases in all 9 years.
Exhibit 7-2a: Table of Toxics Release Inventory (TRI) Data for 1,2-Dichloroethane
Year
1998
1999
2000
2001
2002
2003
2004
2005
2006
On-Site Releases (pounds)
Total Air
Emissions
710,621
548,798
525,366
451,555
453,569
367,312
441,171
500,487
448,864
Surface Water
Discharges
2,608
904
839
2,227
4,797
864
1,377
2,785
1,923
Underground
Injection
58,994
65,465
171,423
13,463
237,560
235,948
298,546
220,963
26,134
Releases to
Land
886
2,983
2,271
141
574
148
69
68
38
Off-site
Releases
(pounds)
185,688
683,304
444,832
23,386
30,538
253,746
252,616
254,288
121,616
Total On- &
Off -site
Releases
(pounds)
958,797
1,301,454
1,144,731
490,772
727,038
858,018
993,780
978,591
598,576
Source: USEPA, 2008
7-3
-------
EPA-OGWDW Contaminant Occurrence Support Document for 2" 6-Year Review Category 1 Contaminants October 2009
Exhibit 7-2b: Graph of Toxics Release Inventory (TRI) Data for 1,2-Dichloroethane
Pounds Released
Environmental Releases of
onn nnn _,
vnn nnn -I
finn nnn
AC\C\ nnn
onn nnn
onn nnn
-I nn nnn
n
1
,2-Dichloroethane
1 n
i—
n
II n 1
| m
n
• • •
n
•— i n
p.
n
-
_
i-i
1 998 1 999 2000 2001 2002 2003
Year
•Air Emissions DSurface Water Discharges D Underground
i-i
2004
njection D Releases to Lane
i-i
2005
•Off-site
2006
Releases
Source: USEPA, 2008
7.3 Occurrence in Ambient Water
Lakes, rivers, and aquifers are the sources of most drinking water. The United States
Geological Survey (USGS) National Water Quality Assessment (NAWQA) program is a
national-scale assessment of the occurrence of contaminants in ambient surface and ground
water. Contaminant occurrence in ambient water provides information on the potential for
contaminants to adversely affect drinking water supplies. Data on the occurrence of 1,2-
dichloroethane in ambient water are available from the NAWQA program.
7.3.1 EPA Summary Analysis ofNA WQA Data
EPA, with the cooperation of the USGS, performed a summary analysis of all Cycle 1
(1992-2001) water monitoring data from all study units for 1,2-dichloroethane. Detection
frequencies were computed as the percentage of samples and sites with detections (i.e., with at
least one result equal to or greater than the reporting limit). Sample detections can be biased by
frequent sampling in areas with high (or low) occurrence. Calculating the percentage of sites with
detections can reduce this bias.
The results of the EPA analysis are presented in Exhibit 7-3. Overall, 1,2-dichloroethane
was detected in 0.3% of samples and at 0.5% of sites. 1,2-Dichloroethane was detected more
frequently in ground water than in surface water. The median concentration, based on detections
from all sites, was 0.377 |ig/L. The 99th percentile concentration, based on detections from all
sites, was 33.4 |ig/L.
7-4
-------
EPA-OGWDW Contaminant Occurrence Support Document for 2" 6-Year Review Category 1 Contaminants October 2009
Exhibit 7-3: EPA Summary Analysis of 1,2-Dichloroethane Data from NAWQA
Study Units, 1992-2001
ground
water
surface
water
all
sites
Detection Frequency
(detections are results > RL1)
Number
of
Samples
4,622
1,402
6,024
% Samples
with
Detections
0.4%
0.2%
0.3%
Number
of Sites
4,137
182
4,319
% Sites
with
Detections
0.4%
1 .6%
0.5%
Concentration Values
(of detections, in ug/L)
Minimum
0.035
0.0291
0.0291
Median
0.3987
0.0512
0.3765
95th
Percen-
tile
33.4
0.247
3
99th
Percen-
tile
33.4
0.247
33.4
Maximum
33.4
0.247
33.4
1. RLs (Reporting Limits) for 1,2-dichloroethane varied, but did not exceed 0.035 pg/L
7.4 Occurrence in Drinking Water
1,2-dichloroethane is regulated as a volatile organic compound (VOC) in drinking water.
All non-purchased community water systems (CWSs) and non-transient non-community water
systems (NTNCWSs) are required to sample for VOCs. The maximum waiver period for VOCs
is six years for ground water systems and three years for surface water systems.
All CWSs and NTNCWSs must collect four consecutive quarterly samples during the
initial three-year compliance period.1 If all four samples are non-detections, then the system may
reduce to annual sampling. After three annual samples without a detection, a system may be
granted a waiver. If a compound is detected (but at less than the MCL), the system must take one
sample per quarter until results are below the MCL (minimum of two quarterly samples for
ground water systems and four quarterly samples for surface water systems). If all quarterly
samples are below the MCL, the system may return to annual sampling. If a compound is
detected at a level greater than the MCL, the system (whether ground water or surface water)
must take four consecutive quarterly samples until all are below the MCL. If all quarterly
samples are below the MCL, the system may return to annual sampling.
The analysis of 1,2-dichloroethane occurrence presented in the following section is based
on state compliance monitoring data from the Six-Year Review-ICR Dataset. This is the largest
and most comprehensive compliance monitoring data set compiled by EPA's Drinking Water
program to date.
All of the Six-Year Review-ICR states provided occurrence data for 1,2-dichloroethane.
These data consist of 375,636 analytical results from 50,442 public water systems (PWSs) during
the period from 1998 to 2005. The number of sample results and systems vary by state, although
the state data sets have been reviewed and checked to ensure adequacy of coverage and
All new systems or systems using a new water source that began operation after January 22, 2004 must demonstrate
compliance with the MCL within a period of time specified by the State. The system must also comply with the
initial sampling frequencies specified by the State to ensure that a system can demonstrate compliance with the MCL.
7-5
-------
EPA-OGWDW Contaminant Occurrence Support Document for 2" 6-Year Review Category 1 Contaminants October 2009
completeness. The national modal minimum reporting level (MRL) for 1,2-dichloroethane in the
datasetis 0.5 |ig/L.
7.4.1 Stage 1A nalysis of Occurrence Data
Stage 1 analyses have been performed at the sample, sample point, and system levels. Note that
some systems have multiple sample points, such as source water sample points or entry points to
the distribution system. Exhibit 7-4 presents the number of samples, systems, and sample points
in the data set, as well as the population-served by the systems and sample points, and the
percentage of detections and the percentage of the population-served by the systems and sample
points with detections. Detections in systems provide a fundamental indication of national
occurrence, but are not representative of MCL violations. For 1,2-dichloroethane, 0.745% of
50,442 systems reported detections. Exhibit 7-5 lists the minimum, median, 90th percentile, and
maximum 1,2-dichloroethane concentrations based on detections from all systems in the Six-
Year Review-ICR Dataset. The median concentration of 1,2-dichloroethane, based on detections
from all systems, was 0.9 |ig/L.
Exhibit 7-4: 1,2-Dichloroethane Occurrence Data from the Six-Year Review-ICR
Dataset
Source Water Type
Number of
Samples
Number of
Systems
Number of
Sample Points
Pop. Served
by Systems
Pop. Served
by Sample
Points
Total Number
Ground Water
290,308
46,185
80,541
86,392,793
86,392,793
Surface Water
85,328
4,257
10,862
140,541,198
140,541,198
Combined Ground &
Surface Water
375,636
50,442
91,403
226,933,991
226,933,991
Percent with Detections
Ground Water
0.526%
0.686%
0.559%
3.361%
1.377%
Surface Water
0.661%
1.386%
1 .234%
7.909%
2.867%
Combined Ground &
Surface Water
0.557%
0.745%
0.639%
6.178%
2.300%
Exhibit 7-5: 1,2-Dichloroethane Summary Statistics
Source Water Type
Ground Water
Detection Value (ug/L)
Minimum
0.01
Median
1
90tn
Percentile
3.6
Maximum
135.8
7-6
-------
EPA-OGWDW Contaminant Occurrence Support Document for 2" 6-Year Review Category 1 Contaminants October 2009
Source Water Type
Detection Value (ug/L)
Minimum
Median
90tn
Percent! le
Maximum
Surface Water
0.005
0.745
2
47
Combined Ground &
Surface Water
0.005
0.9
3.1
135.8
The following maps illustrate the geographic distribution of 1,2-dichloroethane
occurrence in drinking water. Exhibit 7-6 shows the states with systems with at least one
detection and Exhibit 7-7 shows the states with systems with at least one detection greater than
the MCL concentration. Note that five states did not submit data for use in the Six-Year Review
(Kansas2, Louisiana, Mississippi, Pennsylvania, and Washington).
The distribution of systems with detections of 1,2-dichloroethane is geographically
dispersed. Detection rates were generally low; only seven states reported detections in more than
2% of their systems. Twenty-seven states reported detections in less than 1% of their systems.
Eighteen states reported detections greater than the MCL concentration of 5 |ig/L, although only
4 of those states reported detections greater than 5 jig/L in more than 0.2% of their systems.
Exhibit 7-6: Percentage of Systems with at Least One 1,2-Dichloroethane
Detection, by State
Estates Not Submitting Any Data
States with No Detections
Less than 1% of Systems Detecting
Between 1% and 2% of Systems Detecting
^B More than 2% of Systems Detecting
Kansas submitted data after the data management and analyses had been conducted; data from Kansas are not
included in the analyses.
7-7
-------
EPA-OGWDW Contaminant Occurrence Support Document for 2" 6-Year Review Category 1 Contaminants October 2009
Exhibit 7-7: Percentage of Systems with at Least One 1,2-Dichloroethane
Detection Greater than the MCL Concentration (> 5 ug/L), by State
'////, States Not Submitting Any Data
States witn No Detections > MCL
Less than 0.1% of Systems with Detections > MCL
H Between 0.1% and 0.2% of Systems with Detections > MCL
^H More than 0.2% of Systems with Detections > MCL
Exhibit 7-8 presents the Stage 1 analysis of 1,2-dichloroethane occurrence in drinking
water from systems and the population-served by those systems in the Six-Year Review-ICR
Dataset relative to several thresholds. The thresholds are: 5 |ig/L, 2.5 |ig/L, 1 |ig/L, and 0.5 |ig/L.
A total of 66 systems (0.131% of all systems), serving about 4.8 million persons, reported at
least one detection greater than or equal to 2.5 ng/L; and 36 systems (0.071% of all systems),
serving nearly 438,000 persons, reported at least one detection greater than 5 |ig/L. Detections
greater than the thresholds were more common in more ground water than surface water systems.
However, with the exception of the 5 jig/L threshold, the population-served by the surface water
systems reporting at least one detection greater than each threshold was greater the population-
served by ground water systems. This is likely due to the fact that surface water systems tend to
be larger.
Exhibit 7-8: 1,2-Dichloroethane Stage 1 Analysis - Systems and Population with
at Least One Threshold Exceedance
Source
Water Type
Ground
Water
Threshold
> 5 pg/L
>2.5|jg/L
> 1 |jg/L
> 0.5 |jg/L
Number of
Systems
Exceeding
Threshold
33
58
139
214
Percent of
Systems
Exceeding
Threshold
0.071%
0.126%
0.301%
0.463%
Total Population-
Served by Systems
Exceeding
Threshold
247,240
337,994
858,498
1,399,602
Percent of
Population-
Served by
Systems
Exceeding
Threshold
0.286%
0.391%
0.994%
1.620%
-------
EPA-OGWDW Contaminant Occurrence Support Document for 2" 6-Year Review Category 1 Contaminants October 2009
Source
Water Type
Threshold
Number of
Systems
Exceeding
Threshold
Percent of
Systems
Exceeding
Threshold
Total Population-
Served by Systems
Exceeding
Threshold
Percent of
Population-
Served by
Systems
Exceeding
Threshold
Surface
Water
> 5 pg/L
>2.5|jg/L
> 1 |jg/L
> 0.5 |jg/L
3
8
25
41
0.070%
0.188%
0.587%
0.963%
190,972
4,412,609
5,428,439
7,551,832
0.136%
3.140%
3.863%
5.373%
Combined
Ground &
Surface
Water
> 5 pg/L
>2.5|jg/L
> 1 |jg/L
> 0.5 |jg/L
36
66
164
255
0.071%
0.131%
0.325%
0.506%
438,212
4,750,603
6,286,937
8,951,434
0.193%
2.093%
2.770%
3.945%
The occurrence estimates presented above (in Exhibit 7-8) are based on the conservative
assumption that if a 1,2-dichloroethane detection is found in a single entry point/sample point in
a system, then the entire population-served by the system is assumed to potentially be exposed to
1,2-dichloroethane (even though other entry points/sampling points at the system, and the
population-served by those entry/sample points, may exhibit no occurrence of 1,2-
dichloroethane). The sample point analysis is a less conservative estimate of the population-
served by a system with a detection of 1,2-dichloroethane because it estimates the potentially
exposed population-served based only on the proportion of the population-served by entry
points/sample points with detections of 1,2-dichloroethane (rather than the entire population-
served by the system).
Exhibit 7-9 presents the Stage 1 analysis of 1,2-dichloroethane occurrence in drinking
water from sample points and population-served by those sample points in the Six-Year Review-
ICRDataset. A total of 82 (0.090% of) sample points, serving an estimated 189,000 persons,
reported at least one detection exceeding 2.5 |ig/L; and 48 (0.053% of) sample points, serving an
estimated 73,000 persons, reported at least one detection exceeding 5 |ig/L. Similar to the
system-level occurrence analysis, 1,2-dichloroethane was detected in more ground water than
surface water sample points at levels greater than 5 |ig/L.
7-9
-------
EPA-OGWDW Contaminant Occurrence Support Document for 2" 6-Year Review Category 1 Contaminants October 2009
Exhibit 7-9: 1,2-Dichloroethane Stage 1 Analysis - Sample Points and Population
with at Least One Threshold Exceedance
Source
Water Type
Ground
Water
Threshold
> 5 |jg/L
>2.5|jg/L
> 1 H9/L
> 0.5 |jg/L
Number of
Sample Points
Exceeding
Threshold
45
72
186
316
Percent of
Sample Points
Exceeding
Threshold
0.056%
0.089%
0.231%
0.392%
Total Population-
Served by Sample
Points Exceeding
Threshold
40,325
55,370
196,208
346,010
Percent of
Population-
Served by
Sample Points
Exceeding
Threshold
0.047%
0.064%
0.227%
0.401%
Surface
Water
> 5 |jg/L
>2.5|jg/L
> 1 H9/L
> 0.5 |jg/L
3
10
45
111
0.028%
0.092%
0.414%
1.022%
32,358
133,818
682,144
2,596,932
0.023%
0.095%
0.485%
1.848%
Combined
Ground &
Surface
Water
> 5 |jg/L
>2.5|jg/L
> 1 Mg/L
> 0.5 |jg/L
48
82
231
427
0.053%
0.090%
0.253%
0.467%
72,683
189,188
878,353
2,942,942
0.032%
0.083%
0.387%
1.297%
7.4.2 Stage 2 Analysis of Occurrence Data
For the Stage 2 analyses, three different substitution values -zero, one-half the MRL
value, and the full MRL value- were used to replace each non-detection record. Three arithmetic
mean 1,2-dichloroethane concentrations were calculated at each system using the zero, one-half
MRL and full MRL substitution values. These mean calculations were done for all systems with
data in the Six-Year Review-ICR dataset. Then the percentages of all systems with a mean
concentration greater than each threshold were calculated.
Exhibit 7-10 presents the Stage 2 analysis of estimated system mean concentrations for
1,2-dichloroethane occurrence in drinking water relative to several thresholds. The thresholds
are: 5 |ig/L, 2.5 |ig/L, 1 |ig/L, and 0.5 |ig/L. Exhibit 7-11 presents similar information based on
population-served by the systems. Based on the Stage 2 analyses, no more than 9 ground water
and surface water systems (approximately 0.018% of all systems) had an estimated system mean
greater than or equal to 2.5 |ig/L. These 9 systems serve 873 persons. Three systems (0.006% of
all systems), serving 151 persons, had an estimated system mean greater than 5 |ig/L. No surface
water systems had estimated means greater than 5 |ig/L.
7-10
-------
EPA-OGWDW Contaminant Occurrence Support Document for 2" 6-Year Review Category 1 Contaminants October 2009
Exhibit 7-10: 1,2-Dichloroethane Stage 2 Analysis - Systems with a System Mean
Threshold Exceedance
Source
Water
Type
Ground
Water
Threshold
> 5 pg/L
>2.5|jg/L
> 1 |jg/L
> 0.5 |jg/L
Number of Systems with Mean
Exceeding Threshold1
Full MRL
3
9
43
190
1/2 MRL
3
9
35
77
Zero
3
9
29
60
Percent of Systems with Mean
Exceeding Threshold1
Full MRL
0.006%
0.019%
0.093%
0.411%
1/2 MRL
0.006%
0.019%
0.076%
0.167%
Zero
0.006%
0.019%
0.063%
0.130%
Surface
Water
> 5 pg/L
>2.5|jg/L
> 1 |jg/L
> 0.5 |jg/L
0
0
3
38
0
0
2
5
0
0
1
3
0.000%
0.000%
0.070%
0.893%
0.000%
0.000%
0.047%
0.117%
0.000%
0.000%
0.023%
0.070%
Combined
Ground &
Surface
Water
> 5 |jg/L
>2.5|jg/L
> 1 |jg/L
> 0.5 |jg/L
3
9
46
228
3
9
37
82
3
8
30
63
0.006%
0.018%
0.091%
0.452%
0.006%
0.018%
0.073%
0.163%
0.006%
0.016%
0.059%
0.125%
1 The results shown in this table are based on system arithmetic means that were calculated using all sample
detection records and all non-detection records, with the non-detection records replaced (substituted) with
concentration values equal to the full MRL, 14 MRL, and zero.
Exhibit 7-11: 1,2-Dichloroethane Stage 2 Analysis - Population-Served by
Systems with a System Mean Threshold Exceedance
Source
Water
Type
Ground
Water
Threshold
> 5 pg/L
>2.5|jg/L
> 1 |jg/L
> 0.5 |jg/L
Population-Served by Systems
with Mean Exceeding Threshold1
Full MRL
151
873
99,274
963,134
1/2 MRL
151
873
90,942
135,637
Zero
151
833
82,396
119,232
Percent of Population-Served by
Systems with Mean Exceeding
Threshold1
Full MRL
0.0002%
0.001%
0.115%
1.115%
1/2 MRL
0.0002%
0.001%
0.105%
0.157%
Zero
0.0002%
0.001%
0.095%
0.138%
Surface
Water
> 5 pg/L
>2.5|jg/L
> 1 |jg/L
> 0.5 |jg/L
0
0
90,390
7,359,474
0
0
54,260
141,310
0
0
4,750
90,390
0.000%
0.000%
0.064%
5.237%
0.000%
0.000%
0.039%
0.101%
0.000%
0.000%
0.003%
0.064%
7-11
-------
EPA-OGWDW Contaminant Occurrence Support Document for 2" 6-Year Review Category 1 Contaminants October 2009
Source
Water Type
Combined
Ground &
Surface
Water
Threshold
> 5 |jg/L
>2.5|jg/L
> 1 H9/L
> 0.5 |jg/L
Population -Served by Systems
with Mean Exceeding Threshold1
Full MRL
151
873
189,664
8,322,608
1/2 MRL
151
873
145,202
276,947
Zero
151
833
87,146
209,622
Percent of Population-Served by
Systems with Mean Exceeding
Threshold1
Full MRL
0.0001%
0.0004%
0.084%
3.667%
1/2 MRL
0.0001%
0.0004%
0.064%
0.122%
Zero
0.0001%
0.0004%
0.038%
0.092%
1 The results shown in this table are based on system arithmetic means that were calculated using all sample
detection records and all non-detection records, with the non-detection records replaced (substituted) with
concentration values equal to the full MRL, 1A MRL, and zero.
7.5 Additional Occurrence Data
Volatile Organic Compounds in the Nation's Ground Water and Drinking water Supply Wells
The USGS conducted a national assessment of 55 VOCs, including 1,2-dichloroethane, in
samples collected from drinking water supply wells, specifically domestic and public wells
(Zogorski et a/., 2006). This occurrence assessment was based on analyses of samples collected
at the well head, and before any treatment or blending, from about 2,400 domestic wells and
about 1,100 public wells.
In domestic wells, 1,2-dichloroethane had a detection frequency of 0.21% at an
assessment level of 0.2 jig/L (based on 2,383 samples). The median concentration of detections
was 1.3 |ig/L. Three detections were within one order of magnitude of the MCL concentration (5
|ig/L). No detections were greater than the MCL concentration.
In public wells, 1,2-dichloroethane had a detection frequency of 0.56% at an assessment
level of 0.2 |ig/L (based on 1,073 samples). The median concentration of detections was 0.39
|ig/L. Two detections were within one order of magnitude of the MCL concentration. No
detections were greater than the MCL concentration.
First Six-Year Review ofNPDWRs
Similar types of Stage 1 contaminant occurrence analyses were conducted for the first
Six-Year Review. The first Six-Year Review assessed regulated contaminant occurrence data
collected from 1993 to 1997. The first Six-Year Review data set consisted of data from 16 states
that were assembled into a "national cross-section" that was qualitatively indicative (though not
statistically representative) of national occurrence. For the first Six-Year Review (USEPA,
2003b), 302 systems (approximately 1.31% of about 23,000 systems), reported at least one
detection of 1,2-dichloroethane. Thirty systems (approximately 0.13% of all systems), reported at
least one detection greater than the MCL concentration.
7-12
-------
EPA-OGWDW Contaminant Occurrence Support Document for 2" 6-Year Review Category 1 Contaminants October 2009
7.6 Summary of Data
TRI data indicate that total off-site and on-site releases of 1,2-dichloroethane for the
period 1998-2006 ranged from approximately 491,000 pounds to about 1.3 million pounds. Off-
site releases and air emissions were the largest contributors. Discharges to surface water and
releases to land were the smallest fractions.
Ambient water data from the NAWQA program indicate that 1,2-dichloroethane was
detected in 0.3% of 6,024 ground and surface water samples and at 0.5% of 4,319 sites. The
median concentration, based on detections from all sites, was 0.38 |ig/L.
The Stage 1 analysis of occurrence in drinking water indicates that 1,2-dichloroethane
was detected in 0.745% of 50,442 surface water and ground water systems. The median
concentration, based on detections from all systems, was 0.9 |ig/L. In total, 0.071% of systems
(serving 0.193% of the population) reported at least one detection greater than the MCL
concentration of 5 |ig/L.
In the Stage 2 analysis, 0.006% of ground water systems (serving 0.0001% of the
population) had an estimated system mean concentration of 1,2-dichloroethane greater than the
MCL concentration. No surface water systems had mean concentrations exceeding the MCL
concentration.
7.7 References
Agency for Toxic Substances and Disease Registry (ATSDR). 2001. Toxicological Profile for
1,2-Dichloroethane. September 2001. Available on the Internet at:
http ://www. atsdr.cdc.gov/toxprofiles/tp3 8 .html.
Field, J.A. and R. Sierra-Alvarez. 2004. Biodegradability of chlorinated solvents and related
chlorinated aliphatic compounds. Reviews in Environmental Science and Biotechnology. 3:185-
254.
United States Environmental Protection Agency (USEPA). 1987. National Primary Drinking
Water Regulations—Synthetic Organic Chemicals; Monitoring for Unregulated Contaminants;
Final Rule. Federal Register. Vol. 52, No. 130. p. 25690, July 8, 1987.
USEPA. 2003a. Non-confidential Inventory Update Reporting (IUR) Production Volume
Information - 1,2-Dichloroethane. Last modified September 10, 2007. Available on the Internet
at: http://www.epa.gov/oppt/iur/tools/data/2002-vol.htm. [Search for 1,2-dichloroethane.]
Accessed February 21, 2008.
USEPA. 2003b. Occurrence Estimation Methodology and Occurrence Findings for Six-Year
Review of National Primary Drinking Water Regulations. Office of Water. EPA-815-R-03-006.
June 2003.
7-13
-------
EPA-OGWDW Contaminant Occurrence Support Document for 2" 6-Year Review Category 1 Contaminants October 2009
USEPA. 2008. TRI Explorer: Trends - 1,2-Dichloroethane. Released February 21, 2008.
Available on the Internet at: http://www.epa.gov/triexplorer/trends.htm. [Search for 1,2-
dichloroethane.] Accessed February 29, 2008.
USEPA. 2009. The Analysis of Regulated Contaminant Occurrence Data from Public Water
Systems in Support of the Second Six-Year Review of National Primary Drinking Water
Regulations. Office of Water. EPA-815-B-09-006. October 2009.
Zogorski, J.S., J.M. Carter, T. Ivahnenko, W.W. Lapham, MJ. Moran, B.L. Rowe, PJ. Squillace,
and P.L. Toccalino. 2006. Volatile organic compounds in the nation's ground water and drinking
water supply wells. U.S. Geological Survey Circular 1292, 101 p. Available on the Internet at:
http://pubs.er.usgs.gov/pubs/cir/cirl292.
7-14
-------
EPA-OGWDW Contaminant Occurrence Support Document for 2" 6-Year Review Category 1 Contaminants October 2009
8 Dichloromethane (Methylene Chloride)
This chapter on dichloromethane (methylene chloride) is part of a report that is organized
so that each chapter presents information on one specific contaminant. Chapter 1 of this report,
Introduction, presents background information on all sources of data used as well as the
analytical approach used to estimate contaminant occurrence in drinking water. This contaminant
chapter includes background information (such as regulatory history, general chemical
information, and environmental fate and behavior), use and environmental release information,
and dichloromethane occurrence estimates in ambient water and drinking water. All drinking
water occurrence estimates were based on data from the National Compliance Monitoring
Information Collection Request (ICR) Dataset for the Second Six-Year Review (the "Six Year
Review-ICR Dataset"). For detailed information regarding this dataset, data management, data
quality assessments, and the analytical approach used to generate all contaminant occurrence
estimates presented in this chapter, please refer to USEPA (2009).
8.1 Background
The United States Environmental Protection Agency (EPA) published the National
Primary Drinking Water Regulations (NPDWR) for dichloromethane on July 17, 1992 (57 FR
3 1776 (USEPA, 1992)). The NPDWR established a maximum contaminant level goal (MCLG)
of zero based on a cancer classification of B2, probable human carcinogen. The NPDWR also
established a maximum contaminant level (MCL) of 5 |ig/L based on analytical feasibility.
According to the Agency for Toxic Substances and Disease Registry (ATSDR, 2000),
dichloromethane is a colorless liquid with a mild, sweet odor. It does not occur naturally in the
environment.
Dichloromethane (CH^Cb) is typically released into the environment in air as it is highly
volatile. It is also found in minimal amounts in water and soil. The major releases of
dichloromethane come from its use as an industrial solvent and paint stripper. Releases also
come from certain aerosol and pesticide products, its use in manufacturing photographic film, as
well as some spray paints, automotive cleaners, and other household products. In air,
dichloromethane reacts readily with other chemicals in the presence of light. In water, it is broken
down by bacterial action or through reactions with other chemicals (ATSDR, 2000).
The general population may be exposed to dichloromethane through contaminated air or
exposure to consumer products containing dichloromethane. On an average, adults in the urban
United States take in an estimated 30 to 309 jig of dichloromethane daily from outdoor air.
However, the highest and most common form of exposure is in workplaces where the chemical is
used in manufacturing. It can be particularly damaging if workplaces are not adequately
ventilated. Although not as common, exposure may also occur through contaminated water or
food (ATSDR, 2000).
8-1
-------
EPA-OGWDW Contaminant Occurrence Support Document for 2" 6-Year Review Category 1 Contaminants October 2009
8.2 Use, Production, and Releases
Dichloromethane is most often used as a paint remover. It is also used as a solvent and
cleaning agent in a variety of industries, and approved as a fumigant for strawberries and grains,
and to extract substances from foodstuffs. It may also occur in aerosols, pesticide products, spray
paints, automotive cleaners, and other household products. It is also used in the manufacture of
photographic film. Dichloromethane is produced from methane gas or wood alcohol. Most
dichloromethane releases to the environment are from its use as an end product by various
industries, and from the use of aerosol products and paint removers in the home (ATSDR, 2000).
The Toxic Substances Control Act (TSCA) Inventory contains confidential data reported
by industry as a partial update of the Chemical Substance Inventory Update System. The data are
reported under the Inventory Update Reporting regulation. Manufacturers and importers are
required to report company information and chemical information, including production volumes
for chemicals they manufactured or imported in excess of 10,000 pounds in the immediately
preceding fiscal year.
The TSCA Chemical Substance Inventory does not provide actual production volumes;
instead, it indicates at which of five production volume ranges a chemical was manufactured or
imported. The TSCA Inventory is updated every four years. Exhibit 8-1 presents the production
of dichloromethane in the United States from 1986 to 2002. Note that the production of
dichloromethane in the United States decreased between 1990 and 1994 and has remained within
the same range since then.
Exhibit 8-1: Production of Dichloromethane in the United States
Inventory Update Year
1986 Range
1990 Range
1994 Range
1998 Range
2002 Range
Production Volume
> 500 million - 1 billion pounds
> 500 million - 1 billion pounds
> 100 million - 500 million pounds
> 100 million - 500 million pounds
> 100 million - 500 million pounds
Source: USEPA, 2003a
Dichloromethane is listed as a Toxics Release Inventory (TRI) chemical. Exhibit 8-2a
lists the environmental releases of dichloromethane from 1998 to 2006. (Exhibit 8-2b provides a
graphical presentation of the environmental releases.) Air emissions generally comprised more
than 90% of all releases for the period of record, and they decreased substantially overtime.
Underground injection and off-site releases were the next largest categories, contributing a few
percent each to the total. Surface water discharges were the smallest portion, and they decreased
-------
EPA-OGWDW Contaminant Occurrence Support Document for 2" 6-Year Review Category 1 Contaminants October 2009
from their highest value in 1998 over the period of record. These TRI data for dichloromethane
were reported from 43 states plus Puerto Rico. Forty states and Puerto Rico reported
dichloromethane releases every year between 1998 and 2006.
Exhibit 8-2a: Table of Toxics Release Inventory (TRI) Data for Dichloromethane
Year
1998
1999
2000
2001
2002
2003
2004
2005
2006
On-Site Releases (pounds)
Total Air
Emissions
41,025,447
35,761,445
31,045,357
21,651,731
11,192,849
8,567,874
7,345,231
6,113,157
5,781,544
Surface Water
Discharges
15,754
12,072
10,292
4,668
4,149
4,610
6,774
5,703
3,896
Underground
Injection
490,664
166,859
199,809
218,992
354,396
284,248
274,967
195,903
214,187
Releases to
Land
183,719
61,574
838,111
63,523
4,423
3,642
3,382
100,717
102,681
Off-site
Releases
(pounds)
389,826
576,516
367,190
376,324
190,825
94,115
231,712
281,171
699,803
Total On- &
Off -site
Releases
(pounds)
42,105,410
36,578,466
32,460,759
22,315,238
11,746,642
8,954,489
7,862,066
6,696,653
6,802,111
Source: USEPA, 2008
Exhibit 8-2b: Graph of Toxics Release Inventory (TRI) Data for Dichloromethane
Pounds Released
_i._i.|\j|\JCOCO-t>.-t>.
Environmental Releases of Dichloromethane
5 nnn nnn
Onnn nnn
5 nnn nnn
n nnn nnn
c nnn nnn
Onnn nnn
c nnn nnn
Onnn nnn
c nnn nnn
n
0 1 n n n
II II II n n _
1998 1999 2000 2001 2002 2003 2004 2005 2006
Year
• Air Emissions DSurface Water Discharges n Underground Injection n Releases to Land BOff-site Releases
Source: USEPA, 2008
-------
EPA-OGWDW Contaminant Occurrence Support Document for 2" 6-Year Review Category 1 Contaminants October 2009
8.3 Occurrence in Ambient Water
Lakes, rivers, and aquifers are the sources of most drinking water. The United States
Geological Survey (USGS) National Water Quality Assessment (NAWQA) program is a
national-scale assessment of the occurrence of contaminants in ambient surface and ground
water. Contaminant occurrence in ambient water provides information on the potential for
contaminants to adversely affect drinking water supplies. Data on the occurrence of
dichloromethane in ambient water are available from the NAWQA program.
8.3.1 EPA Summary Analysis ofNA WQA Data
EPA, with the cooperation of the USGS, performed a summary analysis of all Cycle 1
(1992-2001) water monitoring data from all study units for dichloromethane. Detection
frequencies were computed as the percentage of samples and sites with detections (i.e., with at
least one result equal to or greater than the reporting limit). Sample detections can be biased by
frequent sampling in areas with high (or low) occurrence. Calculating the percentage of sites with
detections can reduce this bias.
The results of the EPA analysis are presented in Exhibit 8-3. Overall, dichloromethane
was detected in 5.6% of samples and at 4.4% of sites. Dichloromethane was detected more
frequently in surface water than in ground water. The median concentration, based on detections
from all sites, was 0.037 |ig/L. The 99th percentile concentration, based on detections from all
sites, was 4.09 |ig/L.
Exhibit 8-3: EPA Summary Analysis of Dichloromethane Data from NAWQA Study
Units, 1992-2001
ground
water
surface
water
all
sites
Detection Frequency
(detections are results > RL1)
Number
of
Samples
4,653
1,414
6,067
% Samples
with
Detections
2.9%
14.6%
5.6%
Number
of Sites
4,154
191
4,345
% Sites
with
Detections
3.2%
31.9%
4.4%
Concentration Values
(of detections, in ug/L)
Minimum
0.008
0.0055
0.0055
Median
0.05
0.03455
0.037
95tn
Percen-
tile
1.6
1
1.2
99tn
Percen-
tile
21.6
1.55
4.09
Maxi-
mum
25.8
34
34
1. RLs (Reporting Limits) for dichloromethane varied, but did not exceed 0.008 ug/L.
8.4 Occurrence in Drinking Water
Dichloromethane is regulated as a volatile organic compound (VOC) in drinking water.
All non-purchased community water systems (CWSs) and non-transient non-community water
systems (NTNCWSs) are required to sample for VOCs. The maximum waiver period for VOCs
is six years for ground water systems and three years for surface water systems.
8-4
-------
EPA-OGWDW Contaminant Occurrence Support Document for 2" 6-Year Review Category 1 Contaminants October 2009
All CWSs and NTNCWSs must collect four consecutive quarterly samples during the
initial three-year compliance period.1 If all four samples are non-detections, then the system may
reduce to annual sampling. After three annual samples without a detection, a system may be
granted a waiver. If a compound is detected (but at less than the MCL), the system must take one
sample per quarter until results are below the MCL (minimum of two quarterly samples for
ground water systems and four quarterly samples for surface water systems). If all quarterly
samples are below the MCL, the system may return to annual sampling. If a compound is
detected at a level greater than the MCL, the system (whether ground water or surface water)
must take four consecutive quarterly samples until all are below the MCL. If all quarterly
samples are below the MCL, the system may return to annual sampling.
The analysis of dichloromethane occurrence presented in the following section is based
on state compliance monitoring data from the Six-Year Review-ICR Dataset. This is the largest
and most comprehensive compliance monitoring data set compiled by EPA's Drinking Water
program to date.
All of the Six-Year Review-ICR states provided occurrence data for dichloromethane.
These data consist of 372,009 analytical results from 50,169 public water systems (PWSs) during
the period from 1998 to 2005. The number of sample results and systems vary by state, although
the state data sets have been reviewed and checked to ensure adequacy of coverage and
completeness. The national modal minimum reporting level (MRL) for dichloromethane in the
datasetis 0.5 |ig/L.
8.4.1 Stage 1A nalysis of Occurrence Data
Stage 1 analyses have been performed at the sample, sample point, and system levels.
Note that some systems have multiple sample points, such as source water sample points or entry
points to the distribution system. Exhibit 8-4 presents the number of samples, systems, and
sample points in the data set, as well as the population-served by the systems and sample points,
and the percentage of detections and the percentage of the population-served by the systems and
sample points with detections. Detections in systems provide a fundamental indication of
national occurrence, but are not representative of MCL violations. For dichloromethane, 3.903%
of 50,169 systems reported detections. Exhibit 8-5 lists the minimum, median, 90th percentile,
and maximum dichloromethane concentrations based on detections from all systems in the Six-
Year Review-ICR Dataset. The median concentration of dichloromethane, based on detections
from all systems, was 0.9 |ig/L.
1 All new systems or systems using a new water source that began operation after January 22, 2004 must demonstrate
compliance with the MCL within a period of time specified by the State. The system must also comply with the
initial sampling frequencies specified by the State to ensure that a system can demonstrate compliance with the MCL.
8-5
-------
EPA-OGWDW Contaminant Occurrence Support Document for 2" 6-Year Review Category 1 Contaminants October 2009
Exhibit 8-4: Dichloromethane Occurrence Data from the Six-Year Review-ICR
Dataset
Source Water Type
Number of
Samples
Number of
Systems
Number of
Sample Points
Pop. Served
by Systems
Pop. Served
by Sample
Points
Total Number
Ground Water
287,552
45,918
80,241
86,307,048
86,307,048
Surface Water
84,457
4,251
10,869
140,537,012
140,537,012
Combined Ground &
Surface Water
372,009
50,169
91,110
226,844,060
226,844,060
Percent with Detections
Ground Water
0.874%
3.513%
2.357%
10.521%
3.902%
Surface Water
0.909%
8.116%
4.481%
20.240%
5.707%
Combined Ground &
Surface Water
0.882%
3.903%
2.610%
16.542%
5.020%
Exhibit 8-5: Dichloromethane Summary Statistics
Source Water Type
Ground Water
Detection Value (ug/L)
Minimum
0.0041
Median
0.9
90tn
Percentile
3.62
Maximum
315.9
Surface Water
0.005
1
3.2
31
Combined Ground &
Surface Water
0.0041
0.9
3.4
315.9
The following maps illustrate the geographic distribution of dichloromethane occurrence
in drinking water. Exhibit 8-6 shows the states with systems with at least one detection and
Exhibit 8-7 shows the states with systems with at least one detection greater than the MCL
concentration. Note that five states did not submit data for use in the Six-Year Review (Kansas2,
Louisiana, Mississippi, Pennsylvania, and Washington).
The distribution of systems with detections of dichloromethane is geographically
dispersed. Detection rates were generally high; all states that submitted data except for Hawaii
2 Kansas submitted data after the data management and analyses had been conducted; data for Kansas are not
included in the analyses.
8-6
-------
EPA-OGWDW Contaminant Occurrence Support Document for 2" 6-Year Review Category 1 Contaminants October 2009
reported detections. Only five states reported detections in more than 10% of their systems. Eight
states reported detections in between 5 and 10% of their systems. A total of 37 states reported
detections greater than the MCL concentration of 5 |ig/L. Only 4 of these 37 states reported that
more than 1% of their systems reported detections greater than 5 |ig/L.
Exhibit 8-6: Percentage of Systems with at Least One Dichloromethane Detection,
by State
States Not Submitting Any Data
States with No Detections
Loss than 5% of System* Detecting
| Between 5% and 10% of Systems Detecting
j More than 10% of Systems Detecting
3-7
-------
EPA-OGWDW Contaminant Occurrence Support Document for 2" 6-Year Review Category 1 Contaminants October 2009
Exhibit 8-7: Percentage of Systems with at Least One Dichloromethane Detection
Greater than the MCL Concentration (> 5 ug/L), by State
^% Stales Not Submitting Any Data
States with No Detections > MCL
• Less than 0.5% of Systems with Detections > MCL
Between 0.5V, and 1% of Systems with Detection* > MCL
| Mara than 1% of Systems with Detections > MCL
Exhibit 8-7 presents the Stage 1 analysis of dichloromethane occurrence in drinking water
from systems and the population-served by those systems in the Six-Year Review-ICR Dataset
relative to several thresholds. The thresholds are: 5 |ig/L, 2.5 |ig/L, 1 |ig/L, and 0.5 |ig/L. A total
of 398 systems (0.793% of all systems), serving about 8.1 million persons, reported at least one
detection greater than or equal to 2.5 |ig/L; and 147 systems (0.293% of all systems), serving 2.9
million persons, reported at least one detection exceeding 5 |ig/L. Detections greater than the
thresholds were more common in more ground water than surface water systems. However,
because surface water systems tend to be larger, a greater population was served by the systems
reporting at least one detection greater than the thresholds.
Exhibit 8-7: Dichloromethane Stage 1 Analysis - Systems and Population with at
Least One Threshold Exceedance
Source
Water Type
Ground
Water
Threshold
> 5 |jg/L.
>2.5|jg/L
> 1 H9/L
> 0.5 |jg/L.
Number of
Systems
Exceeding
Threshold
125
319
783
1,319
Percent of
Systems
Exceeding
Threshold
0.272%
0.695%
1.705%
2.873%
Total Population-
Served by
Systems
Exceeding
Threshold
426,870
1,996,422
4,262,402
6,991,178
Percent of
Population-Served
by Systems
Exceeding
Threshold
0.495%
2.313%
4.939%
8.100%
8-8
-------
EPA-OGWDW Contaminant Occurrence Support Document for 2" 6-Year Review Category 1 Contaminants October 2009
Source
Water Type
Surface
Water
Threshold
> 5 |jg/L
>2.5|jg/L
> 1 |jg/L
> 0.5 |jg/L
Number of
Systems
Exceeding
Threshold
22
79
193
313
Percent of
Systems
Exceeding
Threshold
0.518%
1.858%
4.540%
7.363%
Total Population-
Served by
Systems
Exceeding
Threshold
2,516,579
6,112,617
11,712,322
25,538,096
Percent of
Population-Served
by Systems
Exceeding
Threshold
1.791%
4.349%
8.334%
18.172%
Combined
Ground &
Surface
Water
> 5 |jg/L
>2.5|jg/L
> 1 |jg/L
> 0.5 |jg/L
147
398
976
1,632
0.293%
0.793%
1.945%
3.253%
2,943,449
8,109,039
15,974,724
32,529,274
1 .298%
3.575%
7.042%
14.340%
The occurrence estimates presented above (in Exhibit 8-7) are based on the conservative
assumption that if a dichloromethane detection is found in a single entry point/sample point in a
system, then the entire population-served by the system is assumed to potentially be exposed to
dichloromethane (even though other entry points/sampling points at the system, and the
population-served by those entry/sample points, may exhibit no occurrence of dichloromethane).
The sample point analysis is a less conservative estimate of the population-served by a system
with a detection of dichloromethane because it estimates the potentially exposed population-
served based only on the proportion of the population-served by entry points/sample points with
detections of dichloromethane (rather than the entire population-served by the system).
Exhibit 8-8 presents the Stage 1 analysis of dichloromethane occurrence in drinking water
from sample points and population-served by those sample points in the Six-Year Review-ICR
Dataset. A total of 448 (0.492% of) sample points, serving an estimated 3 million persons,
reported at least one detection greater than or equal to 2.5 |ig/L; and 164 (0.180% of) sample
points, serving an estimated 1.2 million persons, reported at least one detection exceeding 5
|ig/L. Similar to the system-level occurrence analysis, more ground water sample points had
threshold exceedances. However, because surface water systems tend to be larger, a greater
population was served by sample points with threshold exceedances.
-------
EPA-OGWDW Contaminant Occurrence Support Document for 2" 6-Year Review Category 1 Contaminants October 2009
Exhibit 8-8: Dichloromethane Stage 1 Analysis - Sample Points and Population
with at Least One Threshold Exceedance
Source
Water Type
Ground
Water
Threshold
> 5 |jg/L
>2.5pg/L
> 1 pg/L
> 0.5 |jg/L
Number of
Sample Points
Exceeding
Threshold
137
351
894
1,535
Percent of
Sample
Points
Exceeding
Threshold
0.171%
0.437%
1.114%
1.913%
Total Population -
Served by Sample
Points Exceeding
Threshold
164,062
1,167,914
2,018,857
2,815,719
Percent of
Population-Served
by Sample Points
Exceeding
Threshold
0.190%
1.353%
2.339%
3.262%
Surface
Water
> 5 |jg/L
>2.5pg/L
> 1 pg/L
> 0.5 |jg/L
27
97
267
440
0.248%
0.892%
2.457%
4.048%
1,081,905
1,875,293
4,355,464
7,099,967
0.770%
1.334%
3.099%
5.052%
Combined
Ground &
Surface
Water
> 5 |jg/L
>2.5pg/L
> 1 pg/L
> 0.5 pg/L
164
448
1,161
1,975
0.180%
0.492%
1.274%
2.168%
1,245,967
3,043,207
6,374,321
9,915,686
0.549%
1.342%
2.810%
4.371%
8.4.2 Stage 2 A nalysis of Occurrence Data
For the Stage 2 analyses, three different substitution values -zero, one-half the MRL
value, and the full MRL value- were used to replace each non-detection record. Three arithmetic
mean dichloromethane concentrations were calculated at each system using the zero, one-half
MRL and full MRL substitution values. These mean calculations were done for all systems with
data in the Six-Year Review-ICR dataset. Then the percentages of all systems with a mean
concentration greater than each threshold were calculated.
Exhibit 8-9 presents the Stage 2 analysis of estimated system mean concentrations for
dichloromethane occurrence in drinking water relative to several thresholds. The thresholds are:
5 |ig/L, 2.5 |ig/L, 1 |ig/L, and 0.5 |ig/L. Exhibit 8-10 presents similar information based on
population-served by the systems. Based on the Stage 2 analyses, no more than 53 ground water
and surface water systems (approximately 0.106% of all systems) had an estimated system mean
greater than or equal to 2.5 |ig/L. These 53 systems serve 43,953 persons. No more than 17
(approximately 0.034% of all systems) systems serving 12,019 persons had an estimated system
mean greater than 5 |ig/L. The percentages of systems with MCL concentration exceedances
were similarly low for both surface water and ground water systems, but a larger number of
ground water systems had an estimated system mean greater than the MCL concentration of 5
8-10
-------
EPA-OGWDW Contaminant Occurrence Support Document for 2" 6-Year Review Category 1 Contaminants October 2009
Exhibit 8-9: Dichloromethane Stage 2 Analysis - Systems with a System Mean
Threshold Exceedance
Source
Water Type
Ground
Water
Threshold
> 5 |jg/L
>2.5|jg/L
> 1 |jg/L
> 0.5 |jg/L
Number of Systems with Mean
Exceeding Threshold1
Full MRL
16
48
240
1,251
1/2 MRL
15
46
181
494
Zero
12
42
151
329
Percent of Systems with Mean
Exceeding Threshold1
Full MRL
0.035%
0.105%
0.523%
2.724%
1/2 MRL
0.033%
0.100%
0.394%
1.076%
Zero
0.026%
0.091%
0.329%
0.716%
Surface
Water
> 5 |jg/L
>2.5|jg/L
> 1 |jg/L
> 0.5 |jg/L
1
5
36
306
1
5
27
85
1
4
18
54
0.024%
0.118%
0.847%
7.198%
0.024%
0.118%
0.635%
2.000%
0.024%
0.094%
0.423%
1.270%
Combined
Ground &
Surface
Water
> 5 |jg/L
>2.5|jg/L
> 1 |jg/L
> 0.5 |jg/L
17
53
276
1,557
16
51
208
579
13
46
169
383
0.034%
0.106%
0.550%
3.104%
0.032%
0.102%
0.415%
1.154%
0.026%
0.092%
0.337%
0.763%
1 The results shown in this table are based on system arithmetic means that were calculated using all sample
detection records and all non-detection records, with the non-detection records replaced (substituted) with
concentration values equal to the full MRL, 14 MRL, and zero.
Exhibit 8-10: Dichloromethane Stage 2 Analysis - Population-Served by Systems
with a System Mean Threshold Exceedance
Source
Water Type
Ground
Water
Threshold
> 5 pg/L
>2.5|jg/L
> 1 |jg/L
> 0.5 |jg/L
Population-Served by Systems with
Mean Exceeding Threshold1
Full MRL
10,219
36,518
1,273,387
6,645,442
1/2 MRL
10,119
33,009
1,152,919
1,910,810
Zero
9,148
31,607
791,105
1,385,316
Percent of Population-Served by
Systems with Mean Exceeding
Threshold1
Full MRL
0.012%
0.042%
1 .475%
7.700%
1/2 MRL
0.012%
0.038%
1.336%
2.214%
Zero
0.011%
0.037%
0.917%
1.605%
Surface
Water
> 5 pg/L
>2.5|jg/L
> 1 |jg/L
> 0.5 |jg/L
1,800
7,435
243,772
25,125,784
1,800
7,435
232,890
1,586,371
1,800
7,350
154,950
458,789
0.001%
0.005%
0.173%
17.878%
0.001%
0.005%
0.166%
1.129%
0.001%
0.005%
0.110%
0.326%
8-11
-------
EPA-OGWDW Contaminant Occurrence Support Document for 2" 6-Year Review Category 1 Contaminants October 2009
Source
Water Type
Combined
Ground &
Surface
Water
Threshold
> 5 |jg/L
>2.5|jg/L
> 1 |jg/L
> 0.5 |jg/L
Population-Served by Systems with
Mean Exceeding Threshold1
Full MRL
12,019
43,953
1,517,159
31,771,226
1/2 MRL
11,919
40,444
1,385,809
3,497,181
Zero
10,948
38,957
946,055
1,844,105
Percent of Population-Served by
Systems with Mean Exceeding
Threshold1
Full MRL
0.005%
0.019%
0.669%
14.006%
1/2 MRL
0.005%
0.018%
0.611%
1.542%
Zero
0.005%
0.017%
0.417%
0.813%
1 The results shown in this table are based on system arithmetic means that were calculated using all sample
detection records and all non-detection records, with the non-detection records replaced (substituted) with
concentration values equal to the full MRL, 1A MRL, and zero.
8.5 Additional Occurrence Data
Volatile Organic Compounds in the Nation's Ground Water and Drinking Water Supply Wells
The USGS conducted a national assessment of 55 VOCs, including dichloromethane
(referred to as methylene chloride in the study) in samples collected from drinking water supply
wells, specifically domestic and public wells (Zogorski et a/., 2006). This occurrence assessment
was based on analyses of samples collected at the well head, and before any treatment or
blending, from about 2,400 domestic wells and about 1,100 public wells.
In domestic wells, dichloromethane had a detection frequency of 0.67% at an assessment
level of 0.2 jig/L (based on 2,398 samples). The median concentration of detections was 0.029
|ig/L. Nine detections were within one order of magnitude of the MCL concentration (5 |ig/L).
No detections were greater than the MCL concentration.
In public wells, dichloromethane had a detection frequency of 0.46% at an assessment
level of 0.2 |ig/L (based on 1,094 samples). The median concentration of detections was 1.6
|ig/L. Two detections were within one order of magnitude of the MCL concentration and two
detections were greater than the MCL concentration.
First Six-Year Review ofNPDWRs
Similar types of Stage 1 contaminant occurrence analyses were conducted for the first
Six-Year Review. The first Six-Year Review assessed regulated contaminant occurrence data
collected from 1993 to 1997. The first Six-Year Review data set consisted of data from 16 states
that were assembled into a "national cross-section" that was qualitatively indicative (though not
statistically representative) of national occurrence. For the first Six-Year Review (USEPA,
2003b), 1,850 systems (approximately 8.59% of about 22,000 systems), reported at least one
detection of dichloromethane. One hundred and forty-four systems (approximately 0.67% of all
systems), reported at least one detection greater than the MCL concentration.
-------
EPA-OGWDW Contaminant Occurrence Support Document for 2" 6-Year Review Category 1 Contaminants October 2009
8.6 Summary of Data
TRI data indicate that total off-site and on-site releases of dichloromethane ranged from
about 6.7 million pounds to about 42 million pounds, decreasing steadily with time. Air
emissions generally constituted over 90% of the total. Surface water discharges were the smallest
fraction.
Ambient water data from the NAWQA program indicate that dichloromethane was
detected in 5.6% of 6,067 ground and surface water samples and at 4.4% of 4,345 sites. The
median concentration, based on detections from all sites, was 0.037 |ig/L.
The Stage 1 analysis of occurrence in drinking water indicates that dichloromethane was
detected in 3.903% of 50,169 surface water and ground water systems. The median
concentration, based on detections from all systems, was 0.9 |ig/L. In total, 0.293% of systems
(serving 1.298% of the population) reported at least one detection greater than the MCL
concentration of 5 |ig/L. Detections greater than the MCL concentration were more common in
ground water systems than surface water systems. However, because surface water systems tend
to be larger, a greater population was served by surface water systems with at least one detection
greater than the MCL concentration. A similar pattern was seen for sample points.
In the Stage 2 analysis, 0.034% of surface water and ground water systems (serving
0.005% of the population) had an estimated system mean concentration greater than the MCL
concentration. System mean exceedances of all thresholds were greater for ground water systems,
and a larger population was served by ground water systems with means exceeding the
thresholds.
8.7 References
Agency for Toxic Substances and Disease Registry (ATSDR). 2000. Toxicological Profile for
Methylene Chloride. September 2005. Available on the Internet at:
http://www.atsdr.cdc.gov/toxprofiles/tpl4.html.
United States Environmental Protection Agency (USEPA). 1992. Drinking Water; National
Primary Drinking Water Regulations—Synthetic Organic Chemicals and Inorganic Chemicals;
National Primary Drinking Water Regulations Implementation; Final Rule. Federal Register.
Vol. 57, No. 138. p. 31776, July 17, 1992.
USEPA. 2003a. Non-confidential Inventory Update Reporting (IUR) Production Volume
Information - Dichloromethane. Last modified September 10, 2007. Available on the Internet at:
http://www.epa.gov/oppt/iur/tools/data/2002-vol.htm. [Search for dichloromethane.] Accessed
February 21, 2008.
USEPA. 2003b. Occurrence Estimation Methodology and Occurrence Findings for Six-Year
Review of National Primary Drinking Water Regulations. Office of Water. EPA-815-R-03-006.
June 2003.
-------
EPA-OGWDW Contaminant Occurrence Support Document for 2" 6-Year Review Category 1 Contaminants October 2009
USEPA. 2008. TRI Explorer: Trends - Dichloromethane. Released February 21, 2008. Available
on the Internet at: http://www.epa.gov/triexplorer/trends.htm. [Search for dichloromethane.]
Accessed February 29, 2008.
USEPA. 2009. The Analysis of Regulated Contaminant Occurrence Data from Public Water
Systems in Support of the Second Six-Year Review of National Primary Drinking Water
Regulations. Office of Water. EPA-815-B-09-006. October 2009.
Zogorski, J.S., J.M. Carter, T. Ivahnenko, W.W. Lapham, MJ. Moran, B.L. Rowe, PJ. Squillace,
and P.L. Toccalino. 2006. Volatile organic compounds in the nation's ground water and drinking
water supply wells. U.S. Geological Survey Circular 1292, 101 p. Available on the Internet at:
http://pubs.er.usgs.gov/pubs/cir/cirl292.
8-14
-------
EPA-OGWDW Contaminant Occurrence Support Document for 2" 6-Year Review Category 1 Contaminants October 2009
9 2,4-Dichlorophenoxyacetic Acid (2,4-D)
This chapter on 2,4-Dichlorophenoxyacetic acid (2,4-D) is part of a report that is
organized so that each chapter presents information on one specific contaminant. Chapter 1 of
this report, Introduction, presents background information on all sources of data used as well as
the analytical approach used to estimate contaminant occurrence in drinking water. This
contaminant chapter includes background information (such as regulatory history, general
chemical information, and environmental fate and behavior), use and environmental release
information, and 2,4-D occurrence estimates in ambient water and drinking water. All drinking
water occurrence estimates were based on data from the National Compliance Monitoring
Information Collection Request (ICR) Dataset for the Second Six-Year Review (the "Six Year
Review-ICR Dataset"). For detailed information regarding this dataset, data management, data
quality assessments, and the analytical approach used to generate all contaminant occurrence
estimates presented in this chapter, please refer to USEPA (2009).
9.1 Background
The United States Environmental Protection Agency (EPA) published the National
Primary Drinking Water Regulations (NPDWR) for 2,4-D on January 30, 1991 (56 FR 3526
(USEPA, 1991)). The NPDWR established a maximum contaminant level goal (MCLG) and a
maximum contaminant level (MCL) of 70 |ig/L. EPA developed the MCLG based on a reference
dose (RfD) of 10 jig/kg-day (0.01 mg/kg-day) and a cancer classification of D, not classifiable as
to human carcinogenicity. Note that in 2005, EPA updated its health effects assessment of 2,4-D
(USEPA, 2005). The assessment reduced the RfD from 10 |ig/kg-day (0.01 mg/kg-day) to 5
|ig/kg-day (0.005 mg/kg-day).
According to the Extension Toxicology Network (EXTOXNET, 1996), 2,4-D
(CgHeCbOs) is a white powder used as a systemic herbicide. While there are many forms or
derivatives of 2,4-D including esters, amines, and salts, toxicological and ecological information
is generally given for the acid form. It may be formulated as an emulsion, in aqueous solution
(salts), and as a dry compound (EXTOXNET, 1996).
2,4-D is degraded by microbes in soil and water. The half-life in soil is less than seven
days. In water, breakdown rates increase with increased nutrients, sediment load, and dissolved
organic carbon. Under oxygenated conditions the half-life is one week to several weeks. Despite
its short half-life in soil and in aquatic environments, the compound has been detected in
groundwater supplies in at least five states. Very low concentrations have also been detected in
surface waters throughout the United States (EXTOXNET, 1996).
9.2 Use, Production, and Releases
2,4-D has been used as a systemic herbicide to control many types of broadleaf weeds. It
has been used in cultivated agriculture, in pasture and rangeland applications, forest
management, home gardens, and to control aquatic vegetation (EXTOXNET, 1996).
9-1
-------
EPA-OGWDW Contaminant Occurrence Support Document for 2" 6-Year Review Category 1 Contaminants October 2009
Exhibit 9-1, compiled by the United States Geological Survey (USGS, 2007), shows the
geographic distribution of estimated average annual 2,4-D use in the United States from 1999
through 2004. A breakdown of use by crop is also included. The map was created by the USGS
using state-level data sets on pesticide use rates from 1999-2004 compiled by the CropLife
Foundation, at the Crop Protection Research Institute, combined with county-level data on
harvested crop acreage obtained from the 2002 Census of Agriculture. Due to the nature of the
data sources, non-agricultural uses are not reflected here, and variations in use at the county-level
are also not well represented (Thelin and Gianessi, 2000). The USGS (2007) estimates that
nearly 40 million pounds of 2,4-D active ingredient were used on average each year between
1999 and 2004. The greatest use of 2,4-D is in the middle of the country, from North Dakota to
Texas.
Exhibit 9-1: Estimated Annual Agricultural Use for 2,4-D (c. 2002)
2,4-D - herbicide
2002 e-ii/r iritd emal agrtutu'»l use
Average annual use af
aclTse ngredenl
(pounSt i»r squa'9 mil* oi aiyictihral
landkioouily}
—i no estimated lisa
I 0.031 lo 1.084
G 1.086 b 3.241
J 3.24S to 7.029
~3 7.027 to 13.156
• >= 13.169
Crap*
Tdtai
pwrntft app*rt
Z1S17BBO
otiwr hay
Gugztane
b*fer*rg
Bofghuir
1 sii-Uii-
142B739
S54CB8
S257U
5S.H
ia.es
fJDS
sja
.;.-..'
a.TB
3.U
1.41
IJ9
Source: L/SGS, 2007
The Toxic Substances Control Act (TSCA) Inventory contains confidential data reported
by industry as a partial update of the Chemical Substance Inventory Update System. The data are
reported under the Inventory Update Reporting regulation. Manufacturers and importers are
required to report company information and chemical information, including production volumes
for chemicals they manufactured or imported in excess of 10,000 pounds in the immediately
preceding fiscal year.
The TSCA Chemical Substance Inventory does not provide actual production volumes;
instead, it indicates at which of five production volume ranges a chemical was manufactured or
imported. The TSCA Inventory is updated every four years. Exhibit 9-2 presents the production
9-2
-------
EPA-OGWDW Contaminant Occurrence Support Document for 2" 6-Year Review Category 1 Contaminants October 2009
of 2,4-D in the United States from 1986 to 2002. Note that, although the production of 2,4-D in
the United States increased since 1986, production remained within the same range since 1990.
Exhibit 9-2: Production of 2,4-D in the United States
Inventory Update Year
1986 Range
1990 Range
1994 Range
1998 Range
2002 Range
Production Volume
> 10 million - 50 million
> 50 million - 100 million
> 50 million - 100 million
> 50 million - 100 million
> 50 million - 100 million
Source: USEPA, 2003a
2,4-D is listed as a Toxics Release Inventory (TRI) chemical. Exhibit 9-3a lists the
environmental releases of 2,4-D from 1998 to 2006. (Exhibit 9-3b provides a graphical
presentation of the environmental releases.) Underground injection was the most significant
mode of release of 2,4-D in 1998, 1999, and 2002, and was much less significant in the other
years. Surface water discharges remained steady and small relative to other modes of release. Air
releases remained relatively steady, and were generally greater than underground injection and
surface water discharges. Releases to land increased dramatically after 2002. The TRI data for
2,4-D were reported from 21 states, with 13 states reporting (Alabama, Arkansas, California,
Iowa, Illinois, Kansas, Michigan, Missouri, Montana, Nebraska, Ohio, Pennsylvania, and Texas)
in all 9 years.
Exhibit 9-3a: Table of Toxics Release Inventory (TRI) Data for 2,4-D
Year
1998
1999
2000
2001
2002
2003
2004
2005
2006
On-Site Releases (pounds)
Total Air
Emissions
3,738
7,913
6,816
2,247
3,706
3,765
1,884
3,622
3,607
Surface Water
Discharges
88
74
256
13
13
10
165
10
125
Underground
Injection
31,209
22,290
250
250
14,250
250
250
250
5
Releases
to Land
1,798
1,798
1,831
1,509
2,100
250,458
187,431
263,396
154,491
Off-Site
Releases
(pounds)
5,112
12,994
23,964
9,597
6,851
12,452
4,897
1,879
3,032
Total On- &
Off-site
Releases
(pounds)
41,945
45,069
33,117
13,616
26,920
266,935
194,627
269,156
161,260
Source: USEPA, 2008
9-3
-------
EPA-OGWDW Contaminant Occurrence Support Document for 2" 6-Year Review Category 1 Contaminants October 2009
Exhibit 9-3b: Graph of Toxics Release Inventory (TRI) Data for 2,4-D
Pounds Released
Environmental Releases of 2,4-D
onn nnn
ocn nnn
onn nnn
•icn nnn
-inn nnn
en nnn
. .
ll_ r, FLrn „ _TI rn n_
~
1998 1999 2000 2001 2002 2003 2004 2005 2006
Year
• Air Emissions DSurface Water Discharges D Underground Injection D Releases to Land BOff-site Releases
Source: USEPA, 2008
9.3 Occurrence in Ambient Water
Lakes, rivers, and aquifers are the sources of most drinking water. The USGS National
Water Quality Assessment (NAWQA) program is a national-scale assessment of the occurrence
of contaminants in ambient surface and ground water. Contaminant occurrence in ambient water
provides information on the potential for contaminants to adversely affect drinking water
supplies. Data on the occurrence of 2,4-D in ambient water are available from the NAWQA
program for both the Pesticide National Synthesis Project and a summary analysis of all Cycle 1
water monitoring data.
9.3.1 NA WQA Pesticide National Synthesis Project
The NAWQA Pesticide National Synthesis Project is a national-scale assessment of the
occurrence and behavior of pesticides in streams and ground water of the United States and the
potential for pesticides to adversely affect drinking water supplies or aquatic ecosystems. Under
the National Synthesis Project, the USGS monitored 2,4-D between 1992 and 2001 in streams
and wells across the country (Gilliom etal., 2007). The maximum long-term method detection
limit for 2,4-D was 0.080 |ig/L.
In NAWQA stream samples (Exhibit 9-4), 2,4-D was found at frequencies ranging from
0.38% of samples in undeveloped areas to 6.38% of samples in mixed land use settings, 13.08%
of samples in agricultural settings, and 13.10% of samples in urban areas. The 95th percentile
concentration was less than the method detection limit in undeveloped areas, 0.080 |ig/L in
9-4
-------
EPA-OGWDW Contaminant Occurrence Support Document for 2" 6-Year Review Category 1 Contaminants October 2009
mixed land use settings, 0.270 |ig/L in agricultural settings, and 0.450 |ig/L in urban land use
settings. The highest concentration, 15.0 |ig/L, was found at an agricultural site.
Exhibit 9-4: USGS National Synthesis Summary of NAWQA Monitoring of 2,4-D in
Streams, 1992-2001
Land Use
Type
Agricultural
Mixed
Undeveloped
Urban
No. of Samples
(No. of Sites)
1,465(62)
801 (36)
101 (5)
523(19)
Detection
Frequency
13.08%
6.38%
0.38%
13.10%
50tn Percent! le
(Median)
Concentration
(in ug/L)
ND1
ND1
ND1
ND1
95th Percentile
Concentration
(in ug/L)
0.270
0.080
ND1
0.450
Maximum
Concentration
(in ug/L)
15.000
1.400
0.060
5.530
Source: Gilliom et a/., 2007
1. ND = not detected (Concentration is less than the maximum long term method detection limit and is expected to
be less than any higher percentile concentration shown in the table)
In ground water NAWQA samples (Exhibit 9-5), 2,4-D was found at frequencies ranging
from 0% of samples in undeveloped areas to 0.38% in mixed land use settings, 0.66% of samples
in agricultural settings, and 0.75% of samples in urban areas. The 95th percentile concentrations
were less than the method detection limit in all settings. The highest concentration, 14.8 |ig/L,
was detected at an urban site.
Exhibit 9-5: USGS National Synthesis Summary of NAWQA Monitoring of 2,4-D in
Ground Water, 1992-2001
Land Use Type
Agricultural
Mixed
Undeveloped
Urban
Number
of Wells
1,208
1,568
22
666
Detection
Frequency
0.66%
0.38%
0%
0.75%
50tn Percentile
(Median)
Concentration
(in ug/L)
ND1
ND1
ND1
ND1
95th Percentile
Concentration
(in ug/L)
ND1
ND1
ND1
ND1
Maximum
Concentration
(in ug/L)
4.54
0.15
ND1
14.8
Source: Gilliom et a/., 2007
1. ND = not detected (Concentration is less than the maximum long term method detection limit and is expected to
be less than any higher percentile concentration shown in the table)
9.3.2 EPA Summary Analysis ofNA WQA Data
Whereas the NAWQA program often uses the most representative data for a site to
calculate summary statistics, EPA, with the cooperation of USGS, performed a summary analysis
of all Cycle 1 (1992-2001) water monitoring data from all study units for 2,4-D. Detection
frequencies were computed as the percentage of samples and sites with detections (i.e., with at
9-5
-------
EPA-OGWDW Contaminant Occurrence Support Document for 2" 6-Year Review Category 1 Contaminants October 2009
least one result equal to or greater than the reporting limit). Sample detections can be biased by
frequent sampling in areas with high (or low) occurrence. Calculating the percentage of sites with
detections can reduce this bias.
The results of the EPA analysis are presented in Exhibit 9-6. Overall, 2,4-D was detected
in 8.1% of samples and at 5.1% of sites. 2,4-D was detected more frequently in surface water
than in ground water. The median concentration, based on detections from all sites, was 0.2 |ig/L.
The 99th percentile concentration, based on detections from all sites, was 5.83 |ig/L.
Exhibit 9-6: EPA Summary Analysis of 2,4-D Data from NAWQA Study Units, 1992-
2001
ground
water
surface
water
all
sites
Detection Frequency
(detections are results > RL1)
Number
of
Samples
4,018
5,535
9,553
% Samples
with
Detections
0.5%
13.6%
8.1%
Number
of Sites
3,645
894
4,539
% Sites
with
Detections
0.5%
23.8%
5.1%
Concentration Values
(of detections, in ug/L)
Minimum
0.01
0.007
0.007
Median
0.11
0.2
0.2
95th
Percen-
tile
14.8
1.76
1.8
99th
Percen-
tile
14.8
5.53
5.83
Maximum
14.8
74.5718
74.5718
1. RLs (Reporting Limits) for 2,4-D varied, but did not exceed 0.01 pg/L Note that because this EPA analysis
involves more data points than the USGS analyses presented above, a direct comparison is not possible.
9.4 Occurrence in Drinking Water
2,4-D is regulated as a synthetic organic chemical (SOC) in drinking water. All non-
purchased community water systems (CWSs) and non-transient non-community water systems
(NTNCWSs) are required to sample for SOCs. The maximum waiver period for SOCs is three
years, but this waiver can be renewed indefinitely. (Systems may receive a waiver at any time
without having collected an SOC sample. With a waiver, systems are not required to collect an
SOC sample.)
All CWSs and NTNCWSs without an SOC waiver must collect four consecutive
quarterly samples during the initial three-year compliance period.1 If all four samples are non-
detections, then a system serving less than 3,300 persons may reduce to one sample during each
consecutive compliance period; a system serving more than 3,300 persons may reduce to two
quarterly samples within a 12-month period during a 3-year compliance period. If a chemical is
detected (but is less than the MCL), the system must take one sample per quarter until results are
below the MCL (minimum of two quarterly samples for ground water systems and four quarterly
samples for surface water systems). If all quarterly samples are below the MCL, the system may
1 All new systems or systems using a new water source that began operation after January 22, 2004 must demonstrate
compliance with the MCL within a period of time specified by the State. The system must also comply with the
initial sampling frequencies specified by the State to ensure that a system can demonstrate compliance with the MCL.
9-6
-------
EPA-OGWDW Contaminant Occurrence Support Document for 2" 6-Year Review Category 1 Contaminants October 2009
return to annual sampling. If a chemical is detected at a level greater than the MCL, the system
(whether ground water or surface water) must take quarterly samples until four consecutive
quarters are below the MCL. If all quarterly samples are below the MCL, the system may return
to annual sampling.
The analysis of 2,4-D occurrence presented in the following section is based on state
compliance monitoring data from the Six-Year Review-ICR Dataset. This is the largest and most
comprehensive compliance monitoring data set compiled by EPA's Drinking Water program to
date.
Forty-three of the Six-Year Review-ICR states provided occurrence data for 2,4-D.
(There were no 2,4-D data from Hawaii or Oklahoma.) These data consist of 124,285 analytical
results from 33,187 public water systems (PWSs) during the period from 1998 to 2005. The
number of sample results and systems vary by state, although the state data sets have been
reviewed and checked to ensure adequacy of coverage and completeness. The national modal
minimum reporting level (MRL) for 2,4-D in the dataset is 0.1 |ig/L.
9.4.1 Stage 1A nalysis of Occurrence Data
Stage 1 analyses have been performed at the sample, sample point, and system levels.
Note that some systems have multiple sample points, such as source water sample points or entry
points to the distribution system. Exhibit 9-7 presents the number of samples, systems, and
sample points in the data set, as well as the population-served by the systems and sample points,
and the percentage of detections and the percentage of the population-served by the systems and
sample points with detections. Detections in systems provide a fundamental indication of
national occurrence, but are not representative of MCL violations. For 2,4-D, 0.901% of 33,187
systems reported detections. Exhibit 9-8 lists the minimum, median, 90th percentile, and
maximum 2,4-D concentrations based on detections from all systems in the Six-Year Review-
ICR Dataset. The median concentration of 2,4-D, based on detections from all systems, was 0.3
Exhibit 9-7: 2,4-D Occurrence Data from the Six-Year Review-ICR Dataset
Source Water Type
Number of
Samples
Number of
Systems
Number of
Sample Points
Pop. Served
by Systems
Pop. Served
by Sample
Points
Total Number
Ground Water
101,329
30,172
48,417
67,200,878
67,200,878
Surface Water
22,956
3,015
6,980
120,250,305
120,250,305
Combined Ground &
Surface Water
124,285
33,187
55,397
187,451,183
187,451,183
Percent with Detections
Ground Water
0.283%
0.669%
0.489%
5.317%
4.452%
9-7
-------
EPA-OGWDW Contaminant Occurrence Support Document for 2" 6-Year Review Category 1 Contaminants October 2009
Source Water Type
Number of
Samples
Number of
Systems
Number of
Sample Points
Pop. Served
by Systems
Pop. Served
by Sample
Points
Surface Water
0.640%
3.217%
1 .605%
10.613%
6.221%
Combined Ground &
Surface Water
0.349%
0.901%
0.630%
8.714%
5.587%
Exhibit 9-8: 2,4-D Summary Statistics
Source Water Type
Ground Water
Detection Value (ug/L)
Minimum
0.02
Median
0.3
90tn
Percentile
2
Maximum
70
Surface Water
0.009
0.29
1.1
230
Combined Ground &
Surface Water
0.009
0.3
1.5
230
The following maps illustrate the geographic distribution of 2,4-D occurrence in drinking
water. Exhibit 9-9 shows the states with systems with at least one detection and Exhibit 9-10
shows the states with systems with at least one detection greater than the MCL concentration.
Note that five states did not submit data for use in the Six-Year Review (Kansas2, Louisiana,
Mississippi, Pennsylvania, and Washington). Hawaii and Oklahoma did submit Six-Year data for
most contaminants but neither state submitted data for 2,4-D.
The distribution of systems with detections of 2,4-D is geographically dispersed.
Detection rates were generally low, with only two states having greater than 5% of their systems
detecting 2,4-D. Only one state reported detections greater than the MCL concentration of 70
2 Kansas submitted data after the data management and analyses had been conducted; data from Kansas are not
included in the analyses.
-------
EPA-OGWDW Contaminant Occurrence Support Document for 2" 6-Year Review Category 1 Contaminants October 2009
Exhibit 9-9: Percentage of Systems with at Least One 2,4-D Detection, by State
States with Waivers or Not Submitting
States with No Detections
Less than 1*/, of Systems Detecting
| Between 1% and 5% of Systems Detecting
j More than 5% of Systems Detecting
Exhibit 9-10: Percentage of Systems with at Least One 2,4-D Detection Greater
than the MCL Concentration (> 70 M9/L), by State
tffif Sttte* with Waivers or Hot Submitting
States with No Detections T MCL
I Less than 0.1 % of Systems with Defections > MCL
| Between 0.1% and 0.5% of Systems with Detections > MCL
j More tfian O.S% of Systems with Detections > MCL
Exhibit 9-11 presents the Stage 1 analysis of 2,4-D occurrence in drinking water from
systems and the population-served by those systems in the Six-Year Review-ICR Dataset relative
-------
EPA-OGWDW Contaminant Occurrence Support Document for 2" 6-Year Review Category 1 Contaminants October 2009
to several thresholds. The thresholds are: 70 |ig/L, 40 |ig/L, and 35 |ig/L. Three (0.009% of)
systems, serving about 16,000 persons, reported at least one detection greater than or equal to 35
|ig/L; and one (0.003% of) systems, serving about 14,000 persons, reported at least one detection
greater than 70 |ig/L. The numbers of surface water and ground water systems reporting at least
one detection greater than the thresholds were similarly low. Only one surface water system,
serving 13,933 persons, reported at least one detection greater than any of the thresholds. This
system serves a larger population than the ground water systems.
Exhibit 9-11: 2,4-D Stage 1 Analysis - Systems and Population with at Least One
Threshold Exceedance
Source
Water
Type
Ground
Water
Threshold
> 70 |jg/L
> 40 |jg/L
> 35 |jg/L
Number of
Systems
Exceeding
Threshold
0
1
2
Percent of
Systems
Exceeding
Threshold
0.000%
0.003%
0.007%
Total Population-
Served by
Systems
Exceeding
Threshold
0
1,810
1,845
Percent of
Population -Served
by Systems
Exceeding
Threshold
0.000%
0.003%
0.003%
Surface
Water
> 70 |jg/L
> 40 |jg/L
> 35 |jg/L
1
1
1
0.033%
0.033%
0.033%
13,933
13,933
13,933
0.012%
0.012%
0.012%
Combined
Ground &
Surface
Water
> 70 |jg/L
> 40 |jg/L
> 35 |jg/L
1
2
3
0.003%
0.006%
0.009%
13,933
15,743
15,778
0.007%
0.008%
0.008%
The occurrence estimates presented above (in Exhibit 9-11) are based on the conservative
assumption that if a 2,4-D detection is found in a single entry point/sample point in a system,
then the entire population-served by the system is assumed to potentially be exposed to 2,4-D
(even though other entry points/sampling points at the system, and the population-served by
those entry/sample points, may exhibit no occurrence of 2,4-D). The sample point analysis is a
less conservative estimate of the population-served by a system with a detection of 2,4-D because
it estimates the potentially exposed population-served based only on the proportion of the
population-served by entry points/sample points with detections of 2,4-D (rather than the entire
population-served by the system).
Exhibit 9-12 presents the Stage 1 analysis of 2,4-D occurrence in drinking water from
sample points and population-served by those sample points in the Six-Year Review-ICR
Dataset. Three (0.005% of) sample points, serving 15,778 persons, reported at least one detection
9-10
-------
EPA-OGWDW Contaminant Occurrence Support Document for 2" 6-Year Review Category 1 Contaminants October 2009
exceeding 35 |ig/L; and one (0.005% of) sample points, serving 13,933 persons, reported at least
one detection exceeding 70 |ig/L.
Exhibit 9-12: 2,4-D Stage 1 Analysis - Sample Points and Population with at Least
One Threshold Exceedance
Source
Water
Type
Ground
Water
Threshold
> 70 |jg/L
> 40 |jg/L
> 35 |jg/L
Number of
Sample Points
Exceeding
Threshold
0
1
2
Percent of
Sample Points
Exceeding
Threshold
0.000%
0.002%
0.004%
Total Population-
Served by Sample
Points Exceeding
Threshold
0
1,810
1,845
Percent of
Population-
Served by
Sample Points
Exceeding
Threshold
0.000%
0.003%
0.003%
Surface
Water
> 70 |jg/L
> 40 |jg/L
> 35 |jg/L
1
1
1
0.014%
0.014%
0.014%
13,933
13,933
13,933
0.012%
0.012%
0.012%
Combined
Ground &
Surface
Water
> 70 |jg/L
> 40 |jg/L
> 35 |jg/L
1
2
3
0.002%
0.004%
0.005%
13,933
15,743
15,778
0.007%
0.008%
0.008%
9.4.2 Stage 2 A nalysis of Occurrence Data
For the Stage 2 analyses, three different substitution values -zero, one-half the MRL
value, and the full MRL value- were used to replace each non-detection record. Three arithmetic
mean 2,4-D concentrations were calculated at each system using the zero, one-half MRL and full
MRL substitution values. These mean calculations were done for all systems with data in the Six-
Year Review-ICR dataset. Then the percentages of all systems with a mean concentration greater
than the each threshold were calculated.
Exhibit 9-13 presents the Stage 2 analysis of estimated system mean concentrations for
2,4-D occurrence in drinking water relative to several thresholds. The thresholds are: 70 |ig/L, 40
|ig/L, and 35 |ig/L. Exhibit 9-14 presents similar information based on population-served by the
systems. Based on the Stage 2 analyses, no systems had an estimated system mean greater than or
equal to any of the thresholds.
9-11
-------
EPA-OGWDW Contaminant Occurrence Support Document for 2" 6-Year Review Category 1 Contaminants October 2009
Exhibit 9-13: 2,4-D Stage 2 Analysis - Systems with a System Mean Threshold
Exceedance
Source
Water Type
Ground
Water
Threshold
> 70 |jg/L
> 40 |jg/L
> 35 |jg/L
Number of Systems with Mean
Exceeding Threshold1
Full MRL
0
0
0
1/2 MRL
0
0
0
Zero
0
0
0
Percent of Systems with Mean
Exceeding Threshold1
Full MRL
0.000%
0.000%
0.000%
1/2 MRL
0.000%
0.000%
0.000%
Zero
0.000%
0.000%
0.000%
Surface
Water
> 70 |jg/L
> 40 |jg/L
> 35 |jg/L
0
0
0
0
0
0
0
0
0
0.000%
0.000%
0.000%
0.000%
0.000%
0.000%
0.000%
0.000%
0.000%
Combined
Ground &
Surface
Water
> 70 |jg/L
> 40 |jg/L
> 35 |jg/L
0
0
0
0
0
0
0
0
0
0.000%
0.000%
0.000%
0.000%
0.000%
0.000%
0.000%
0.000%
0.000%
1 The results shown in this table are based on system arithmetic means that were calculated using all sample
detection records and all non-detection records, with the non-detection records replaced (substituted) with
concentration values equal to the full MRL, 14 MRL, and zero.
Exhibit 9-14: 2,4-D Stage 2 Analysis - Population-Served by Systems with a
System Mean Threshold Exceedance
Source
Water Type
Ground
Water
Threshold
> 70 |jg/L
> 40 |jg/L
> 35 |jg/L
Population-Served by Systems with
Mean Exceeding Threshold1
Full MRL
0
0
0
1/2 MRL
0
0
0
Zero
0
0
0
Percent of Population-Served by
Systems with Mean Exceeding
Threshold1
Full MRL
0.000%
0.000%
0.000%
1/2 MRL
0.000%
0.000%
0.000%
Zero
0.000%
0.000%
0.000%
Surface
Water
> 70 |jg/L
> 40 |jg/L
> 35 |jg/L
0
0
0
0
0
0
0
0
0
0.000%
0.000%
0.000%
0.000%
0.000%
0.000%
0.000%
0.000%
0.000%
9-12
-------
EPA-OGWDW Contaminant Occurrence Support Document for 2" 6-Year Review Category 1 Contaminants October 2009
Source
Water Type
Combined
Ground &
Surface
Water
Threshold
> 70 |jg/L
> 40 |jg/L
> 35 |jg/L
Population-Served by Systems with
Mean Exceeding Threshold1
Full MRL
0
0
0
1/2 MRL
0
0
0
Zero
0
0
0
Percent of Population-Served by
Systems with Mean Exceeding
Threshold1
Full MRL
0.000%
0.000%
0.000%
1/2 MRL
0.000%
0.000%
0.000%
Zero
0.000%
0.000%
0.000%
1 The results shown in this table are based on system arithmetic means that were calculated using all sample
detection records and all non-detection records, with the non-detection records replaced (substituted) with
concentration values equal to the full MRL, 1A MRL, and zero.
9.5 Additional Occurrence Data
United States Department of Agriculture (USDA) Pesticide Data Program
2,4-D was included in the 2004 annual summary of the USDA Pesticide Data Program
(PDF; USDA, 2006). In finished drinking water, 2,4-D was detected in 134 (35.4 percent) of 379
samples; concentrations ranged from 0.006 |ig/L to 0.401 |ig/L. In untreated drinking water, 2,4-
D was detected in 126 (33.2 percent) of 380 samples; concentrations ranged from 0.006 |ig/L to
0.493 |ig/L.
USGS 1999 Pilot Monitoring Program
2,4-D was an analyte in the 1999 Pilot Monitoring Program (Blomquist et al., 2001). 2,4-
D was detected in approximately 57 percent of the 312 raw water samples and almost 44 percent
of the 225 finished water samples. In raw water, the 95th percentile value was 0.158 |ig/L and the
maximum concentration was 0.414 |ig/L. In finished water, the 95th percentile value was 0.132
|ig/L and the maximum concentration of 2,4-D was 0.634 |ig/L.
First Six-Year Review ofNPDWRs
Similar types of Stage 1 contaminant occurrence analyses were conducted for the first
Six-Year Review. The first Six-Year Review assessed regulated contaminant occurrence data
collected from 1993 to 1997. The first Six-Year Review data set consisted of data from 16 states
that were assembled into a "national cross-section" that was qualitatively indicative (though not
statistically representative) of national occurrence. For the first Six-Year Review (USEPA,
2003b), 19 systems (approximately 0.12% of about 16,000 systems), reported at least one
detection of 2,4-D. Three systems (approximately 0.02% of all systems), reported at least one
detection greater than the MCL concentration.
9.6 Summary of Data
TRI data indicate that total off-site and on-site releases of 2,4-D for the period 1998-2006
ranged from about 14,000 pounds to about 269,000 pounds, with a substantial increase from
9-13
-------
EPA-OGWDW Contaminant Occurrence Support Document for 2" 6-Year Review Category 1 Contaminants October 2009
2003 to 2006. The increase was due to a steep rise in releases to land. Underground injection
constituted most on- and off-site releases between 1998 and 1999. Off-site releases were greater
than on-site releases from 2000 to 2002.
Ambient water data from the NAWQA Pesticide National Synthesis Project show that
most 2,4-D detections occurred at agricultural and urban sites. It was detected in 13.08% of 1,465
agricultural stream samples taken at 62 sites. It was detected in 13.10% of 523 urban stream
samples taken at 19 sites. 2,4-D was detected in 0.66% of samples taken from 1,208 ground
water agricultural wells and in 0.75% of samples taken from 666 urban wells. Additional ambient
water data from the NAWQA program indicate that 2,4-D was detected in 8.1% of 9,553 ground
and surface water samples and at 5.1% of 4,539 sites. The median concentration, based on
detections from all sites, was 0.2 |ig/L.
The Stage 1 analysis of occurrence in drinking water indicates that 2,4-D was detected in
0.901% of 33,187 surface water and ground water systems. The median concentration, based on
detections from all systems, was 0.3 |ig/L. Only one surface water system (serving almost 14,000
persons) reported at least one detection greater than the MCL concentration. No ground water
systems reported detections greater than the MCL concentration.
In the Stage 2 analysis, no systems had an estimated mean concentration exceeding the
MCL concentration.
9.7 References
Blomquist, J.D., J.M. Denis, J.L. Cowles, J.A. Hetrick, R.D. Jones, andN.B. Birchfield. 2001.
Pesticides in Selected Water-Supply Reservoirs and Finished Drinking Water, 1999-2000:
Summary of Results from a Pilot Monitoring Program. U.S. Geological Survey Water-Resources
Investigations Report 01-456, 65 p. Available on the Internet at:
http://md.water.usgs.gov/nawqa/OFR 01-456.pdf.
Extension Toxicology Network (EXTOXNET). 1996. EXTOXNET Pesticide Information
Profiles-2,4-D. Last modified June 1996. Available on the Internet at:
http://extoxnet.orst.edu/pips/24-D.htm. Accessed February 12, 2008.
Gilliom, R.J., J.E. Barbash, C.G. Crawford, P.A. Hamilton, J.D. Martin, N. Nakagaki, L.H.
Nowell, J.C. Scott, P.E. Stackelberg, G.P. Thelin, and D.M. Wolock. 2007. The Quality of Our
Nation's Waters - Pesticides in the Nation's Streams and Ground Water, 1992-2001. Appendix
7. Statistical summaries of water-quality data. U.S. Geological Survey Circular 1291, 172 p.
Available on the Internet at: http://water.usgs.gov/nawqa/pnsp/pubs/circ 129l/appendix7/.
Thelin, G.P. and L.P. Gianessi. 2000. Method for Estimating Pesticide Use for County Areas of
the Conterminous United States. U.S. Geological Survey Open-File Report 00-250, 62 p.
Available on the Internet at: http://ca.water.usgs.gov/pnsp/rep/ofr00250/ofr00250.pdf.
9-14
-------
EPA-OGWDW Contaminant Occurrence Support Document for 2" 6-Year Review Category 1 Contaminants October 2009
United States Department of Agriculture (USDA). 2006. Pesticide Data Program, Annual
Summary, Calendar Year 2004. Available on the Internet at:
http://www.ams.usda.gov/science/pdp/Summary2004.pdf.
United States Environmental Protection Agency (USEPA). 1991. National Primary Drinking
Water Regulations—Synthetic Organic Chemicals and Inorganic Chemicals; Monitoring for
Unregulated Contaminants; National Primary Drinking Water Regulations Implementation;
National Secondary Drinking Water Regulations; Final Rule Federal Register. Vol. 56, No. 30. p.
3526, January 30, 1991.
USEPA. 2003a. Non-confidential Inventory Update Reporting (IUR) Production Volume
Information - 2,4-D. Last modified September 10, 2007. Available on the Internet at:
http://www.epa.gov/oppt/iur/tools/data/2002-vol.htm. [Search for 2,4-D.] Accessed February 21,
2008.
USEPA. 2003b. Occurrence Estimation Methodology and Occurrence Findings for Six-Year
Review of National Primary Drinking Water Regulations. Office of Water. EPA-815-R-03-006.
June 2003.
USEPA. 2005. Reregistration Eligibility Decision (RED) - 2,4-D. June 2005. EPA-738-R-05-
002. Available on the Internet at: http://www.epa.gov/oppsrrdl/REDs/24d_red.pdf.
USEPA. 2008. TRI Explorer: Trends - 2,4-D. Released February 21, 2008b. Available on the
Internet at: http://www.epa.gov/triexplorer/trends.htm. [Search for 2,4-D.] Accessed February 29,
2008.
USEPA. 2009. The Analysis of Regulated Contaminant Occurrence Data from Public Water
Systems in Support of the Second Six-Year Review of National Primary Drinking Water
Regulations. Office of Water. EPA-815-B-09-006. October 2009.
United States Geological Survey (USGS). 2007. 2002 Pesticide Use Maps. Available on the
Internet at: http://water.usgs.gov/nawqa/pnsp/usage/maps/compound_listing.php?year=02.
Accessed December 19, 2007.
9-15
-------
EPA-OGWDW Contaminant Occurrence Support Document for 2" 6-Year Review Category 1 Contaminants October 2009
10 1,2-Dichloropropane
This chapter on 1,2-dichloropropane is part of a report that is organized so that each
chapter presents information on one specific contaminant. Chapter 1 of this report, Introduction,
presents background information on all sources of data used as well as the analytical approach
used to estimate contaminant occurrence in drinking water. This contaminant chapter includes
background information (such as regulatory history, general chemical information, and
environmental fate and behavior), use and environmental release information, and 1,2-
dichloropropane occurrence estimates in ambient water and drinking water. All drinking water
occurrence estimates were based on data from the National Compliance Monitoring Information
Collection Request (ICR) Dataset for the Second Six-Year Review (the "Six Year Review-ICR
Dataset"). For detailed information regarding this dataset, data management, data quality
assessments, and the analytical approach used to generate all contaminant occurrence estimates
presented in this chapter, please refer to USEPA (2009).
10.1 Background
The United States Environmental Protection Agency (EPA) published the current
National Primary Drinking Water Regulations (NPDWR) for 1,2-dichloropropane on January 30,
1991 (56 FR 3526 (USEPA, 1991)). The NPDWR established a maximum contaminant level
goal (MCLG) of zero based on a cancer classification of B2, probable human carcinogen. The
NPDWR also established a maximum contaminant level (MCL) of 5 |ig/L based on analytical
feasibility.
According to the Agency for Toxic Substances and Disease Registry (ATSDR, 1989),
1,2-dichloropropane (also known as 1,2-DCP) is a colorless organic liquid with a chloroform-like
odor. It is moderately soluble in water and evaporates readily at room temperature (ATSDR,
1989).
1,2-Dichloropropane (CsHeCb) is a man-made chemical whose presence in the
environment results from anthropogenic activity. The major releases of 1,2-dichloropropane are
to the atmosphere and to soil. Most of the 1,2-dichloropropane released into the environment
ends up in the air or ground water. The half-life of 1,2-dichloropropane in air is not known
exactly, but it is longer than 23 days, which means that 1,2-dichloropropane can spread to areas
far from where it is released. Nonetheless, air levels of 1,2-dichloropropane are usually quite low
- in urban areas of the United States, the average amount in air is about 22 parts per trillion (ppt).
In ground water, the half-life of 1,2-dichloropropane is estimated to be between six months and
two years. Private wells in farming areas where 1,2-dichloropropane was once used as a soil
fumigant have the greatest risk for contamination (ATSDR, 1989).
10.2 Use, Production, and Releases
1,2-Dichloropropane use today is limited primarily to research and industry as a chemical
intermediate to make perchloroethylene and several other related chlorinated chemicals. Before
10-1
-------
EPA-OGWDW Contaminant Occurrence Support Document for 2" 6-Year Review Category 1 Contaminants October 2009
the early 1980s, 1,2-dichloropropane was used in farming as a soil fumigant and was found in
some paint strippers, varnishes, and furniture finish removers (ATSDR, 1989).
The Toxic Substances Control Act (TSCA) Inventory contains confidential data reported
by industry as a partial update of the Chemical Substance Inventory Update System. The data are
reported under the Inventory Update Reporting regulation. Manufacturers and importers are
required to report company information and chemical information, including production volumes
for chemicals they manufactured or imported in excess of 10,000 pounds in the immediately
preceding fiscal year.
The TSCA Chemical Substance Inventory does not provide actual production volumes;
instead, it indicates at which of five production volume ranges a chemical was manufactured or
imported. The TSCA Inventory is updated every four years. Exhibit 10-1 presents the production
of 1,2-dichloropropane in the United States from 1986 to 2002. Note that the production of 1,2-
dichloropropane in the United States has been in the same range for much of the reporting period.
Exhibit 10-1: Production of 1,2-Dichloropropane in the United States
Inventory Update Year
1986 Range
1990 Range
1994 Range
1998 Range
2002 Range
Production Volume
> 100 million - 500 million pounds
> 50 million - 100 million pounds
No Reports
> 100 million - 500 million pounds
> 100 million - 500 million pounds
Source: USEPA, 2003a
1,2-Dichloropropane is listed as a Toxics Release Inventory (TRI) chemical. Exhibit 10-
2a lists the environmental releases of 1, 2-dichloropropane from 1998 to 2006. (Exhibit 10-2b
provides a graphical presentation of the environmental releases.) On-site releases were greater
than off-site releases. Air emissions generally constituted greater than 95% of total releases and
declined considerably during the period of record. Surface water discharges, releases to land, and
off-site releases fluctuated during the period of record but were minor. There were virtually no
underground injection releases. The TRI data for 1,2-dichloropropane were reported from 13
states, with six states (Louisiana, New Jersey, New York, Ohio, Texas, and Virginia) reporting in
all 9 years.
10-2
-------
EPA-OGWDW Contaminant Occurrence Support Document for 2" 6-Year Review Category 1 Contaminants October 2009
Exhibit 10-2a: Table of Toxics Release Inventory (TRI) Data for 1,2-
Dichloropropane
Year
1998
1999
2000
2001
2002
2003
2004
2005
2006
On-Site Releases (pounds)
Total Air
Emissions
298,158
249,678
264,196
161,265
144,703
146,803
157,499
126,643
93,224
Surface Water
Discharges
1,122
9,243
436
463
1,229
560
2,921
5,585
4,768
Underground
Injection
0
0
5
0
0
0
0
0
0
Releases to
Land
32
30
382
6,608
2,950
2,515
138
98
8
Off -site
Releases
(pounds)
267
6,869
16
6,662
283
3,331
309
259
255
Total On- &
Off-site
Releases
(pounds)
299,579
265,820
265,035
174,998
149,165
153,209
160,867
132,585
98,255
Source: USEPA, 2008
Exhibit 10-2b: Graph of Toxics Release Inventory (TRI) Data for 1,2-
Dichloropropane
Environmental Releases of 1,2-Dichloropropane
350,000
1!
(0
&
v>
•a
D
o
a.
1998
1999
2000 2001
2002
Year
2003
2004 2005
2006
lAir Emissions DSurface Water Discharges D Underground Injection DReleases to Land BOff-site Releases
Source: USEPA, 2008
10-3
-------
EPA-OGWDW Contaminant Occurrence Support Document for 2" 6-Year Review Category 1 Contaminants October 2009
10.3 Occurrence in Ambient Water
Lakes, rivers, and aquifers are the sources of most drinking water. The United States
Geological Survey (USGS) National Water Quality Assessment (NAWQA) program is a
national-scale assessment of the occurrence of contaminants in ambient surface and ground
water. Contaminant occurrence in ambient water provides information on the potential for
contaminants to adversely affect drinking water supplies. Data on the occurrence of 1,2-
dichloropropane in ambient water are available from the NAWQA program.
10.3.1 EPA Summary Analysis of NAWQA Data
EPA, with the cooperation of the USGS, performed a summary analysis of all Cycle 1
(1992-2001) water monitoring data from all study units for 1,2-dichloropropane. Detection
frequencies were computed as the percentage of samples and sites with detections (i.e., with at
least one result equal to or greater than the reporting limit). Sample detections can be biased by
frequent sampling in areas with high (or low) occurrence. Calculating the percentage of sites with
detections can reduce this bias.
The results of the EPA analysis are presented in Exhibit 10-3. Overall, 1,2-
dichloropropane was detected in 1.7% of samples and at 1.8% of sites. It was detected more
frequently in ground water than in surface water. The median concentration, based on detections
from all sites, was 0.129 |ig/L. The 99th percentile concentration, based on detections from all
sites, was 11.4 |ig/L.
Exhibit 10-3: EPA Summary Analysis of 1,2-Dichloropropane Data from NAWQA
Study Units, 1992-2001
ground
water
surface
water
all
sites
Detection Frequency
(detections are results > RL1)
Number
of
Samples
4,658
1,413
6,071
% Samples
with
Detections
1.9%
0.8%
1.7%
Number
of Sites
4,159
190
4,349
% Sites
with
Detections
1.8%
1 .6%
1.8%
Concentration Values
(of detections, in ug/L)
Minimum
0.002
0.01
0.002
Median
0.1649
0.02
0.1288
95th
Percen-
tile
6.6
0.091
1.6
99th
Percen-
tile
19.4
0.091
11.4
Maximum
19.4
0.091
19.4
1. RLs (Reporting Limits) for 1,2-dichloropropane varied, but did not exceed 0.01 pg/L
10.4 Occurrence in Drinking Water
1,2-Dichloropropane is regulated as a volatile organic compound (VOC) in drinking
water. All non-purchased community water systems (CWSs) and non-transient non-community
water systems (NTNCWSs) are required to sample for VOCs. The maximum waiver period for
VOCs is six years for ground water systems and three years for surface water systems.
10-4
-------
EPA-OGWDW Contaminant Occurrence Support Document for 2" 6-Year Review Category 1 Contaminants October 2009
All CWSs and NTNCWSs must collect four consecutive quarterly samples during the
initial three-year compliance period.1 If all four samples are non-detections, then the system may
reduce to annual sampling. After three annual samples without a detection, a system may be
granted a waiver. If a compound is detected (but at less than the MCL), the system must take one
sample per quarter until results are below the MCL (minimum of two quarterly samples for
ground water systems and four quarterly samples for surface water systems). If all quarterly
samples are below the MCL, the system may return to annual sampling. If a compound is
detected at a level greater than the MCL, the system (whether ground water or surface water)
must take four consecutive quarterly samples until all are below the MCL. If all quarterly
samples are below the MCL, the system may return to annual sampling.
The analysis of 1,2-dichloropropane occurrence presented in the following section is
based on state compliance monitoring data from the Six-Year Review-ICR Dataset. This is the
largest and most comprehensive compliance monitoring data set compiled by EPA's Drinking
Water program to date.
All of the Six-Year Review-ICR states provided occurrence data for 1,2-dichloropropane.
These data consist of 374,409 analytical results from 50,437 public water systems (PWSs) during
the period from 1998 to 2005. The number of sample results and systems vary by state, although
the state data sets have been reviewed and checked to ensure adequacy of coverage and
completeness. The national modal minimum reporting level (MRL) for 1,2-dichloropropane in
the dataset is 0.5 |ig/L.
10.4.1 Stage 1 Analysis of Occurrence Data
Stage 1 analyses have been performed at the sample, sample point, and system levels. Note that
some systems have multiple sample points, such as source water sample points or entry points to
the distribution system. Exhibit 10-4 presents the number of samples, systems, and sample points
in the data set, as well as the population-served by the systems and sample points, and the
percentage of detections and the percentage of the population-served by the systems and sample
points with detections. Detections in systems provide a fundamental indication of national
occurrence, but are not representative of MCL violations. For 1,2-dichloropropane, 0.480% of
50,437 systems reported detections. Exhibit 10-5 lists the minimum, median, 90th percentile, and
maximum 1,2-dichloropropane concentrations based on detections from all systems in the Six-
Year Review-ICR Dataset. The median concentration of 1,2-dichloropropane, based on
detections from all systems, was 1 |ig/L.
1 All new systems or systems using a new water source that began operation after January 22, 2004 must demonstrate
compliance with the MCL within a period of time specified by the State. The system must also comply with the
initial sampling frequencies specified by the State to ensure that a system can demonstrate compliance with the MCL.
10-5
-------
EPA-OGWDW Contaminant Occurrence Support Document for 2" 6-Year Review Category 1 Contaminants October 2009
Exhibit 10-4: 1,2-Dichloropropane Occurrence Data from the Six-Year Review-ICR
Dataset
Source Water Type
Number of
Samples
Number of
Systems
Number of
Sample Points
Pop. Served
by Systems
Pop. Served
by Sample
Points
Total Number
Ground Water
289,938
46,180
80,535
86,395,709
86,395,709
Surface Water
84,471
4,257
10,856
140,516,425
140,516,425
Combined Ground &
Surface Water
374,409
50,437
91,391
226,912,134
226,912,134
Percent with Detections
Ground Water
0.231%
0.468%
0.308%
1 .644%
0.765%
Surface Water
0.300%
0.611%
0.405%
6.229%
0.704%
Combined Ground &
Surface Water
0.246%
0.480%
0.320%
4.483%
0.727%
Exhibit 10-5: 1,2-Dichloropropane Summary Statistics
Source Water Type
Ground Water
Detection Value (ug/L)
Minimum
0.01
Median
0.9
90tn
Percentile
3.1
Maximum
11.3
Surface Water
0.005
1.1
2.5
26.2
Combined Ground &
Surface Water
0.005
1
2.9
26.2
The following maps illustrate the geographic distribution of 1,2-dichloropropane
occurrence in drinking water. Exhibit 10-6 shows the states with systems with at least one
detection and Exhibit 10-7 shows the states with systems with at least one detection greater than
the MCL concentration. Note that five states did not submit data for use in the Six-Year Review
(Kansas2, Louisiana, Mississippi, Pennsylvania, and Washington).
The distribution of systems with detections of 1,2-dichloropropane is geographically
dispersed. Detection rates were generally low; although 37 states reported detections, only six
2 Kansas submitted data after the data management and analyses had been conducted; data from Kansas are not
included in the analyses.
10-6
-------
EPA-OGWDW Contaminant Occurrence Support Document for 2" 6-Year Review Category 1 Contaminants October 2009
states reported detections in more than 1% of their systems. Only eight states reported detections
greater than the MCL concentration of 5 |ig/L.
Exhibit 10-6: Percentage of Systems with at Least One 1,2-Dichloropropane
Detection, By State
States Not Submitting Any Data
States with No Detections
Less than 0.5% of Systems Detecting
| Between 0.5% and 1% of Systems Detecting
| More than 1% of Systems Detecting
10-7
-------
EPA-OGWDW Contaminant Occurrence Support Document for 2" 6-Year Review Category 1 Contaminants October 2009
Exhibit 10-7: Percentage of Systems with at Least One 1,2-Dichloropropane
Detection Greater than the MCL Concentration (> 5 pg/L), by State
States Mot Submitting Any Data
States with No Detections > MCL
Less than D.I % of Systems with Detections > MCL
Between 0.1 % and 0.5% of Systems with Detections > MCL
More than 0.5% of Systems with Detections > MCL
Exhibit 10-8 presents the Stage 1 analysis of 1,2-dichloropropane occurrence in drinking
water from systems and the population-served by those systems in the Six-Year Review-ICR
Dataset relative to several thresholds. The thresholds are: 5 tig/L, 2.5 tig/L, 1 tig/L, and 0.5 tig/L.
Thirty-nine (0.077% of) systems, serving about 1.1 million persons, reported at least one
detection greater than or equal to 2.5 tig/L; and 10 (0.020% of) systems, serving about 352,000
persons, reported at least one detection greater than 5 tig/L. Detections greater than the thresholds
were more common in ground water systems than surface water systems. However, because
surface water systems tend to be larger, a greater population was served by surface water systems
reporting detections greater than the thresholds.
10-8
-------
EPA-OGWDW Contaminant Occurrence Support Document for 2" 6-Year Review Category 1 Contaminants October 2009
Exhibit 10-8: 1,2-Dichloropropane Stage 1 Analysis - Systems and Population
with at Least One Threshold Exceedance
Source
Water Type
Ground
Water
Threshold
> 5 |jg/L
>2.5pg/L
> 1 pg/L
> 0.5 |jg/L
Number of
Systems
Exceeding
Threshold
8
33
84
150
Percent of
Systems
Exceeding
Threshold
0.017%
0.071%
0.182%
0.325%
Total Population-
Served by
Systems
Exceeding
Threshold
16,691
43,469
464,180
666,040
Percent of
Population -Served
by Systems
Exceeding
Threshold
0.019%
0.050%
0.537%
0.771%
Surface
Water
> 5 |jg/L
>2.5pg/L
> 1 pg/L
> 0.5 |jg/L
2
6
14
15
0.047%
0.141%
0.329%
0.352%
334,996
1,094,057
7,662,775
8,053,925
0.238%
0.779%
5.453%
5.732%
Combined
Ground &
Surface
Water
> 5 |jg/L
>2.5pg/L
> 1 pg/L
> 0.5 pg/L
10
39
98
165
0.020%
0.077%
0.194%
0.327%
351,687
1,137,526
8,126,955
8,719,965
0.155%
0.501%
3.582%
3.843%
The occurrence estimates presented above (in Exhibit 10-8) are based on the conservative
assumption that if a 1,2-dichloropropane detection is found in a single entry point/sample point
in a system, then the entire population-served by the system is assumed to potentially be exposed
to 1,2-dichloropropane (even though other entry points/sampling points at the system, and the
population-served by those entry/sample points, may exhibit no occurrence of 1,2-
dichloropropane). The sample point analysis is a less conservative estimate of the population-
served by a system with a detection of 1,2-dichloropropane because it estimates the potentially
exposed population-served based only on the proportion of the population-served by entry
points/sample points with detections of 1,2-dichloropropane (rather than the entire population-
served by the system).
Exhibit 10-9 presents the Stage 1 analysis of 1,2-dichloropropane occurrence in drinking
water from sample points and population-served by those sample points in the Six-Year Review-
ICR Dataset. Forty-two (0.046% of) sample points, serving an estimated 228,000 persons,
reported at least one detection greater than or equal to 2.5 |ig/L; and 13 (0.014% of) sample
points, serving an estimated 182,000 persons, reported at least one detection exceeding 5 |ig/L.
Similar to the system-level occurrence analysis, more ground water sample points had threshold
exceedances. However, because surface water systems tend to be larger, a greater population was
served by sample points with threshold exceedances.
10-9
-------
EPA-OGWDW Contaminant Occurrence Support Document for 2" 6-Year Review Category 1 Contaminants October 2009
Exhibit 10-9: 1,2-Dichloropropane Stage 1 Analysis - Sample Points and
Population with at Least One Threshold Exceedance
Source
Water Type
Ground
Water
Threshold
> 5 pg/L
>2.5|jg/L
> 1 |jg/L
> 0.5 |jg/L
Number of
Sample Points
Exceeding
Threshold
10
35
99
169
Percent of
Sample Points
Exceeding
Threshold
0.012%
0.043%
0.123%
0.210%
Total Population-
Served by Sample
Points Exceeding
Threshold
3,884
24,739
95,285
151,688
Percent of
Population-
Served by
Sample Points
Exceeding
Threshold
0.004%
0.029%
0.110%
0.176%
Surface
Water
> 5 pg/L
>2.5|jg/L
> 1 |jg/L
> 0.5 |jg/L
3
7
21
29
0.028%
0.064%
0.193%
0.267%
178,346
203,597
523,178
641,646
0.127%
0.145%
0.372%
0.457%
Combined
Ground &
Surface
Water
> 5 pg/L
>2.5|jg/L
> 1 |jg/L
> 0.5 |jg/L
13
42
120
198
0.014%
0.046%
0.131%
0.217%
182,230
228,336
618,463
793,334
0.080%
0.101%
0.273%
0.350%
10.4.2 Stage 2 Analysis of Occurrence Data
For the Stage 2 analyses, three different substitution values -zero, one-half the MRL
value, and the full MRL value- were used to replace each non-detection record. Three arithmetic
mean 1,2-dichloropropane concentrations were calculated at each system using the zero, one-half
MRL and full MRL substitution values. These mean calculations were done for all systems with
data in the Six-Year Review-ICR dataset. Then the percentages of all systems with a mean
concentration greater than each threshold were calculated.
Exhibit 10-10 presents the Stage 2 analysis of estimated system mean concentrations for
1,2-dichloropropane occurrence in drinking water relative to several thresholds. The thresholds
are: 5 |ig/L, 2.5 |ig/L, 1 |ig/L, and 0.5 |ig/L. Exhibit 10-11 presents similar information based on
population-served by the systems. Based on the Stage 2 analyses, two ground water systems
(approximately 0.004% of all systems) had an estimated system mean greater than or equal to 2.5
|ig/L. No surface water system had an estimated system mean greater than or equal to 2.5 |ig/L.
These two systems serve 120 persons. No systems had an estimated system mean greater than 5
10-10
-------
EPA-OGWDW Contaminant Occurrence Support Document for 2" 6-Year Review Category 1 Contaminants October 2009
Exhibit 10-10: 1,2-Dichloropropane Stage 2 Analysis - Systems with a System
Mean Threshold Exceedance
Source
Water Type
Ground
Water
Threshold
> 5 |jg/L
>2.5|jg/L
> 1 |jg/L
> 0.5 |jg/L
Number of Systems with Mean
Exceeding Threshold1
Full MRL
0
2
26
138
1/2 MRL
0
2
23
59
Zero
0
2
20
46
Percent of Systems with Mean
Exceeding Threshold1
Full MRL
0.000%
0.004%
0.056%
0.299%
1/2 MRL
0.000%
0.004%
0.050%
0.128%
Zero
0.000%
0.004%
0.043%
0.100%
Surface
Water
> 5 |jg/L
>2.5|jg/L
> 1 |jg/L
> 0.5 |jg/L
0
0
1
14
0
0
1
2
0
0
1
1
0.000%
0.000%
0.023%
0.329%
0.000%
0.000%
0.023%
0.047%
0.000%
0.000%
0.023%
0.023%
Combined
Ground &
Surface
Water
> 5 pg/L
>2.5|jg/L
> 1 |jg/L
> 0.5 |jg/L
0
2
27
152
0
2
24
61
0
2
21
47
0.000%
0.004%
0.054%
0.301%
0.000%
0.004%
0.048%
0.121%
0.000%
0.004%
0.042%
0.093%
1 The results shown in this table are based on system arithmetic means that were calculated using all sample
detection records and all non-detection records, with the non-detection records replaced (substituted) with
concentration values equal to the full MRL, 14 MRL, and zero.
Exhibit 10-11: 1,2-Dichloropropane Stage 2 Analysis - Population-Served by
Systems with a System Mean Threshold Exceedance
Source
Water Type
Ground
Water
Threshold
> 5 pg/L
>2.5|jg/L
> 1 H9/L
> 0.5 |jg/L
Population -Served by Systems
with Mean Exceeding Threshold1
Full MRL
0
120
20,445
647,056
1/2 MRL
0
120
20,057
86,764
Zero
0
120
18,656
30,337
Percent of Population-Served by
Systems with Mean Exceeding
Threshold1
Full MRL
0.000%
0.0001%
0.024%
0.749%
1/2 MRL
0.000%
0.0001%
0.023%
0.100%
Zero
0.000%
0.0001%
0.022%
0.035%
Surface
Water
> 5 pg/L
>2.5|jg/L
> 1 H9/L
> 0.5 |jg/L
0
0
265,696
7,377,671
0
0
265,696
407,696
0
0
265,696
265,696
0.000%
0.000%
0.189%
5.250%
0.000%
0.000%
0.189%
0.290%
0.000%
0.000%
0.189%
0.189%
10-11
-------
EPA-OGWDW Contaminant Occurrence Support Document for 2" 6-Year Review Category 1 Contaminants October 2009
Source
Water Type
Combined
Ground &
Surface
Water
Threshold
> 5 |jg/L
>2.5|jg/L
> 1 H9/L
> 0.5 |jg/L
Population -Served by Systems
with Mean Exceeding Threshold1
Full MRL
0
120
286,141
8,024,727
1/2 MRL
0
120
285,753
494,460
Zero
0
120
284,352
296,033
Percent of Population-Served by
Systems with Mean Exceeding
Threshold1
Full MRL
0.000%
0.0001%
0.126%
3.536%
1/2 MRL
0.000%
0.0001%
0.126%
0.218%
Zero
0.000%
0.0001%
0.125%
0.130%
1 The results shown in this table are based on system arithmetic means that were calculated using all sample
detection records and all non-detection records, with the non-detection records replaced (substituted) with
concentration values equal to the full MRL, 1A MRL, and zero.
10.5 Additional Occurrence Data
Volatile Organic Compounds in the Nation's Ground Water and Drinking Water Supply Wells
The USGS conducted a national assessment of 55 VOCs, including 1,2-dichloropropane,
in samples collected from drinking water supply wells, specifically domestic and public wells
(Zogorski et a/., 2006). This occurrence assessment was based on analyses of samples collected
at the well head, and before any treatment or blending, from about 2,400 domestic wells and
about 1,100 public wells.
In domestic wells, 1,2-dichloropropane had a detection frequency of 0.58% at an
assessment level of 0.2 jig/L (based on 2,400 samples). The median concentration of detections
was 0.30 |ig/L. Six detections were within one order of magnitude of the MCL concentration (5
|ig/L) and three detections were greater than the MCL concentration.
In public wells, 1,2-dichloropropane had a detection frequency of 0.74% at an assessment
level of 0.2 |ig/L (based on 1,078 samples). The median concentration of detections was 0.17
|ig/L. Two detections were within one order of magnitude of the MCL concentration. No
detections were greater than the MCL concentration.
First Six-Year Review ofNPDWRs
Similar types of Stage 1 contaminant occurrence analyses were conducted for the first
Six-Year Review. The first Six-Year Review assessed regulated contaminant occurrence data
collected from 1993 to 1997. The first Six-Year Review data set consisted of data from 16 states
that were assembled into a "national cross-section" that was qualitatively indicative (though not
statistically representative) of national occurrence. For the first Six-Year Review (USEPA,
2003b), 148 systems (approximately 0.67% of about 22,000 systems), reported at least one
detection of 1,2-dichloropropane. Fifteen systems (approximately 0.07% of all systems), reported
at least one detection greater than the MCL concentration.
10-12
-------
EPA-OGWDW Contaminant Occurrence Support Document for 2" 6-Year Review Category 1 Contaminants October 2009
10.6 Summary of Data
TRI data indicate that total off-site and on-site releases of 1,2-dichloropropane for the
period 1998-2006 ranged from approximately 98,000 pounds to about 300,000 pounds, with a
steady decrease over time. Releases were almost entirely emissions to air. Surface water
discharges, releases to land, and off-site releases were minor. Releases via underground injection
were virtually non-existent.
Ambient water data from the NAWQA program indicate that 1,2-dichloropropane was
detected in 1.7% of 6,071 ground and surface water samples and at 1.8% of 4,349 sites. The
median concentration, based on detections from all sites, was 0.13 |ig/L.
The Stage 1 analysis of occurrence in drinking water indicates that 1,2-dichloropropane
was detected in 0.480% of 50,437 surface water and ground water systems. The median
concentration, based on detections from all systems, was 1 |ig/L. Approximately 0.020% of all
systems (serving 0.155% of the population) reported at least one detection greater than the MCL
concentration of 5 |ig/L. The number of systems reporting at least one detection greater than the
MCL concentration was greater for ground water systems. However, because surface water
systems tend to be larger than ground water systems, a greater population was served by surface
water systems reporting at least one detection greater than the MCL concentration. A similar
pattern was seen for sample points.
In the Stage 2 analysis, 0.004% of ground water systems (serving 0.0001% of the
population) had an estimated system mean concentration of 1,2-dichloropropane greater than 2.5
|ig/L. No surface or ground water system had a mean concentration exceeding the MCL
concentration.
10.7 References
Agency for Toxic Substances and Disease Registry (ATSDR). 1989. Toxicological Profile for
1,2-Dichloropropane. December 1989. Available on the Internet at:
http://www.atsdr.cdc.gov/toxprofiles/tpl34.html.
United States Environmental Protection Agency (USEPA). 1991. National Primary Drinking
Water Regulations—Synthetic Organic Chemicals and Inorganic Chemicals; Monitoring for
Unregulated Contaminants; National Primary Drinking Water Regulations Implementation;
National Secondary Drinking Water Regulations; Final Rule Federal Register. Vol. 56, No. 30. p.
3526, January 30, 1991.
USEPA. 2003a. Non-confidential Inventory Update Reporting (IUR) Production Volume
Information - 1,2-Dichloropropane. Last modified September 10, 2007. Available on the Internet
at: http://www.epa.gov/oppt/iur/tools/data/2002-vol.htm. [Search for 1,2-dichloropropane.]
Accessed February 21, 2008.
10-13
-------
EPA-OGWDW Contaminant Occurrence Support Document for 2" 6-Year Review Category 1 Contaminants October 2009
USEPA. 2003b. Occurrence Estimation Methodology and Occurrence Findings for Six-Year
Review of National Primary Drinking Water Regulations. Office of Water. EPA-815-R-03-006.
June 2003.
USEPA. 2008. TRI Explorer: Trends - 1,2-Dichloropropane. Released February 21, 2008.
Available on the Internet at: http://www.epa.gov/triexplorer/trends.htm. [Search for 1,2-
dichloropropane.] Accessed February 29, 2008.
USEPA. 2009. The Analysis of Regulated Contaminant Occurrence Data from Public Water
Systems in Support of the Second Six-Year Review of National Primary Drinking Water
Regulations. Office of Water. EPA-815-B-09-006. October 2009.
Zogorski, J.S., J.M. Carter, T. Ivahnenko, W.W. Lapham, MJ. Moran, B.L. Rowe, PJ. Squillace,
and P.L. Toccalino. 2006. Volatile organic compounds in the nation's ground water and drinking
water supply wells. U.S. Geological Survey Circular 1292, 101 p. Available on the Internet at:
http://pubs.er.usgs.gov/pubs/cir/cirl292.
10-14
-------
EPA-OGWDW Contaminant Occurrence Support Document for 2" 6-Year Review Category 1 Contaminants October 2009
11 Endothall
This chapter on endothall is part of a report that is organized so that each chapter presents
information on one specific contaminant. Chapter 1 of this report, Introduction, presents
background information on all sources of data used as well as the analytical approach used to
estimate contaminant occurrence in drinking water. This contaminant chapter includes
background information (such as regulatory history, general chemical information, and
environmental fate and behavior), use and environmental release information, and endothall
occurrence estimates in ambient water and drinking water. All drinking water occurrence
estimates were based on data from the National Compliance Monitoring Information Collection
Request (ICR) Dataset for the Second Six-Year Review (the "Six Year Review-ICR Dataset").
For detailed information regarding this dataset, data management, data quality assessments, and
the analytical approach used to generate all contaminant occurrence estimates presented in this
chapter, please refer to USEPA (2009).
11.1 Background
The United States Environmental Protection Agency (EPA) published the current
National Primary Drinking Water Regulations (NPDWR) for endothall on July 17, 1992 (57 FR
3 1776 (USEPA, 1992)). The NPDWR established a maximum contaminant level goal (MCLG)
and a maximum contaminant level (MCL) of 100 |ig/L. EPA developed the MCLG based on a
reference dose (RfD) of 20 jig/kg-day (0.02 mg/kg-day) and a cancer classification of D, not
classifiable as to human carcinogenicity.
According to USEPA (2005), endothall (CgHioOs) is an organic solid consisting of white,
odorless crystals. It is applied as either a dipotassium salt or an N, N-dimethylalkylamine salt; the
latter form accounts for the most use. Endothall acid is not directly applied, but rather is formed
as a breakdown product resulting from application of the salt forms (USEPA, 2005).
Endothall is quickly broken down by microbes once deposited on soil or water. In
aqueous environments it is expected to persist for less than ten days. In terrestrial environments,
endothall degrades with a half-life of approximately ten days. Endothall does not bioaccumulate
(USEPA, 2005).
11.2 Use, Production, and Releases
According to USEPA (2005), endothall is most often used as an aquatic herbicide to
control submerged vegetation and algae in lakes, ponds, and irrigation canals. It is also used
terrestrially as an herbicide, where it is applied to cotton, potatoes, hops, clover, and alfalfa
(USEPA, 2005).
Exhibit 11-1, compiled by the United States Geological Survey (USGS, 2007), shows the
geographic distribution of estimated average annual endothall use in the United States from 1999
through 2004. A breakdown of use by crop is also included. The map was created by the USGS
using state-level data sets on pesticide use rates from 1999-2004 compiled by the CropLife
ll-l
-------
EPA-OGWDW Contaminant Occurrence Support Document for 2" 6-Year Review Category 1 Contaminants October 2009
Foundation, at the Crop Protection Research Institute, combined with county-level data on
harvested crop acreage obtained from the 2002 Census of Agriculture. Due to the nature of the
data sources, non-agricultural uses are not reflected here, and variations in use at the county-level
are also not well represented (Thelin and Gianessi, 2000). The USGS (2007) estimates that about
37,000 pounds of endothall active ingredient were used on average each year between 1999 and
2004. The greatest use of endothall is in the Midwest, the Pacific Northwest, and California.
Exhibit 11-1: Estimated Annual Agricultural Use for Endothall (c. 2002)
ENDOTHALL - herbicide
2002 estimated annual agricultural use
Average annual use of
active ingredient
(pounds per square mils of agricultural
land in county)
EH no estimated jse
D 0.001 to 0.009
• 0.01 to 0.035
D 0.036 to 0.083
D 0.084 to 0.307
• >= 0.308
Crops
cotton
potatoes
hops
Total
Pounds Applied
20037
14966
1730
Percent
National Use
64.66
40.74
4.71
Source: USGS, 2007
The Chemical Substance Inventory Update System contains confidential data reported by
industry as a partial update of the Toxic Substances Control Act (TSCA) Inventory. The TSCA
Chemical Substance Inventory does not report production data for endothall.
Endothall is not listed as a Toxics Release Inventory (TRI) chemical; thus no TRI release
records are available. Therefore, the use of endothall (described in the previous section) may
provide the primary indication of where releases are most likely to occur.
11.3 Occurrence in Ambient Water
No National Water Quality Assessment (NAWQA) Program data are available from the
United States Geological Survey (USGS) on the occurrence of endothall in ambient waters from
11-2
-------
EPA-OGWDW Contaminant Occurrence Support Document for 2" 6-Year Review Category 1 Contaminants October 2009
the Pesticide National Synthesis Project (Gilliom et a/., 2007) or the EPA Summary Analysis of
Cycle 1 data.
11.4 Occurrence in Drinking Water
Endothall is regulated as a synthetic organic chemical (SOC) in drinking water. All non-
purchased community water systems (CWSs) and non-transient non-community water systems
(NTNCWSs) are required to sample for SOCs. The maximum waiver period for SOCs is three
years, but this waiver can be renewed indefinitely. (Systems may receive a waiver at any time
without having collected an SOC sample. With a waiver, systems are not required to collect an
SOC sample.) Statewide waivers are often granted for endothall.
All CWSs and NTNCWSs without an SOC waiver must collect four consecutive
quarterly samples during the initial three-year compliance period.1 If all four samples are non-
detections, then a system serving less than 3,300 persons may reduce to one sample during each
consecutive compliance period; a system serving more than 3,300 persons may reduce to two
quarterly samples within a 12-month period during a 3-year compliance period. If a chemical is
detected (but is less than the MCL), the system must take one sample per quarter until results are
below the MCL (minimum of two quarterly samples for ground water systems and four quarterly
samples for surface water systems). If all quarterly samples are below the MCL, the system may
return to annual sampling. If a chemical is detected at a level greater than the MCL, the system
(whether ground water or surface water) must take quarterly samples until four consecutive
quarters are below the MCL. If all quarterly samples are below the MCL, the system may return
to annual sampling.
The analysis of endothall occurrence presented in the following section is based on state
compliance monitoring data from the Six-Year Review-ICR Dataset. This is the largest and most
comprehensive compliance monitoring data set compiled by EPA's Drinking Water program to
date.
Thirty-eight of the Six-Year Review-ICR states provided occurrence data for endothall.
(There were no endothall data from Hawaii, Minnesota, Missouri, North Carolina, New
Hampshire, Oklahoma, and Texas.) These data consist of 49,820 analytical results from 14,156
public water systems (PWSs) during the period from 1998 to 2005. The number of sample results
and systems vary by state, although the state data sets have been reviewed and checked to ensure
adequacy of coverage and completeness. The national modal minimum reporting level (MRL) for
endothall in the dataset is 9 |ig/L.
11.4.1 Stage 1 Analysis of Occurrence Data
Stage 1 analyses have been performed at the sample, sample point, and system levels.
Note that some systems have multiple sample points, such as source water sample points or entry
points to the distribution system. Exhibit 11-2 presents the number of samples, systems, and
1 All new systems or systems using a new water source that began operation after January 22, 2004 must demonstrate
compliance with the MCL within a period of time specified by the State. The system must also comply with the
initial sampling frequencies specified by the State to ensure that a system can demonstrate compliance with the MCL.
11-3
-------
EPA-OGWDW Contaminant Occurrence Support Document for 2" 6-Year Review Category 1 Contaminants October 2009
sample points in the data set, as well as the population-served by the systems and sample points,
and the percentage of detections and the percentage of the population-served by the systems and
sample points with detections. Detections in systems provide a fundamental indication of
national occurrence, but are not representative of MCL violations. For endothall, 0.233% of
14,156 systems reported detections. Exhibit 11-3 lists the minimum, median, 90th percentile, and
maximum endothall concentrations based on detections from all systems in the Six-Year Review-
ICR Dataset. The median concentration of endothall, based on detections from all systems, was
9.8
Exhibit 11-2: Endothall Occurrence Data from the Six-Year Review-ICR Dataset
Source Water Type
Number of
Samples
Number of
Systems
Number of
Sample Points
Pop. Served
by Systems
Pop. Served
by Sample
Points
Total Number
Ground Water
38,414
12,632
20,422
44,124,258
44,124,258
Surface Water
11,406
1,524
3,593
74,412,543
74,412,543
Combined Ground &
Surface Water
49,820
14,156
24,015
118,536,801
118,536,801
Percent with Detections
Ground Water
0.081%
0.158%
0.118%
7.222%
7.130%
Surface Water
0.149%
0.853%
0.417%
3.934%
1.885%
Combined Ground &
Surface Water
0.096%
0.233%
0.162%
5.158%
3.837%
Exhibit 11-3: Endothall Summary Statistics
Source Water Type
Ground Water
Detection Value (ug/L)
Minimum
1.5
Median
11.5
90tn
Percentile
13.6
Maximum
450
Surface Water
0.1
9
171.2
548.1
Combined Ground &
Surface Water
0.1
9.8
55
548.1
11-4
-------
EPA-OGWDW Contaminant Occurrence Support Document for 2" 6-Year Review Category 1 Contaminants October 2009
The following maps illustrate the geographic distribution of endothall occurrence in
drinking water. Exhibit 11-4 shows the states with systems with at least one detection and
Exhibit 11-5 shows the states with systems with at least one detection greater than the MCL
concentration. Note that five states did not submit data for use in the Six-Year Review (Kansas2,
Louisiana, Mississippi, Pennsylvania, and Washington). Hawaii, Minnesota, Missouri, North
Carolina, New Hampshire, Oklahoma, and Texas did submit Six-Year data for most
contaminants; however, these states did not submit endothall data.
The distribution of systems with detections of endothall is geographically dispersed.
Detection rates were generally low; only three states reported detections in more than 1% of their
systems. Only one state reported detections greater than the MCL concentration of 100 |ig/L.
Exhibit 11-4: Percentage of Systems with at Least One Endothall Detection,
by State
tffft States with Waivers or Not Submitting
States with No Detections
Less than 1% of Systems Detecting
^B Between 1% and 5% of Systems Detecting
|^| More than 5% of Systems Detecting
2 Kansas submitted data after the data management and analyses had been conducted; data from Kansas are not
included in the analyses.
11-5
-------
EPA-OGWDW Contaminant Occurrence Support Document for 2" 6-Year Review Category 1 Contaminants October 2009
Exhibit 11-5: Percentage of Systems with at Least One Endothall Detection
Greater than the MCL Concentration (> 100 ug/L), by State
Slates with Waivers or Not Submitting
States with No Detections > MCL
Less than 0.1% of Systems with Detections > MCL
Between 0.1% and 0.5% of Systems with Detections > MCL
More than 0.5% of Systems with Detections > MCL
Exhibit 11-6 presents the Stage 1 analysis of endothall occurrence in drinking water from
systems and the population-served by those systems in the Six-Year Review-ICR Dataset relative
to 100 |ig/L and 50 |ig/L. Four systems (0.028% of all systems), serving about 87,000 persons,
had at least one detection greater than or equal to 50 |ig/L; and three systems (0.021% of all
systems), serving 75,000 persons, had at least one detection greater than 100 |ig/L. Only one
ground water system, which serves 20,000 persons, reported at least one detection of endothall
greater than or equal to 50 |ig/L or greater than 100 |ig/L. Two surface water systems (serving
about 55,000 persons) reported at least one detection of endothall greater than 100 |ig/L.
Exhibit 11-6: Endothall Stage 1 Analysis - Systems and Population with at Least
One Threshold Exceedance
Source
Water
Type
Ground
Water
Threshold
> 100|jg/L
> 50 |jg/L
Number of
Systems
Exceeding
Threshold
1
1
Percent of
Systems
Exceeding
Threshold
0.008%
0.008%
Total Population-
Served by Systems
Exceeding
Threshold
20,000
20,000
Percent of
Population -Served
by Systems
Exceeding
Threshold
0.045%
0.045%
Surface
Water
> 100 |jg/L
> 50 |jg/L
2
3
0.131%
0.197%
54,657
66,657
0.073%
0.090%
11-6
-------
EPA-OGWDW Contaminant Occurrence Support Document for 2" 6-Year Review Category 1 Contaminants October 2009
Source
Water
Type
Combined
Ground &
Surface
Water
Threshold
>100|jg/L
> 50 |jg/L
Number of
Systems
Exceeding
Threshold
3
4
Percent of
Systems
Exceeding
Threshold
0.021%
0.028%
Total Population-
Served by Systems
Exceeding
Threshold
74,657
86,657
Percent of
Population -Served
by Systems
Exceeding
Threshold
0.063%
0.073%
The occurrence estimates presented above (in Exhibit 11-6) are based on the conservative
assumption that if an endothall detection is found in a single entry point/sample point in a
system, then the entire population-served by the system is assumed to potentially be exposed to
endothall (even though other entry points/sampling points at the system, and the population-
served by those entry/sample points, may exhibit no occurrence of endothall). The sample point
analysis is a less conservative estimate of the population-served by a system with a detection of
endothall because it estimates the potentially exposed population-served based only on the
proportion of the population-served by entry points/sample points with detections of endothall
(rather than the entire population-served by the system).
Exhibit 11-7 presents the Stage 1 analysis of endothall occurrence in drinking water from
sample points and population-served by those sample points in the Six-Year Review-ICR
Dataset. Four (0.017% of) sample points (serving approximately 29,000 persons) had at least one
detection greater than or equal to 50 |ig/L; and three (0.012% of) sample points (serving
approximately 28,000 persons) had at least one detection exceeding 100 |ig/L.
Exhibit 11-7: Endothall Stage 1 Analysis - Sample Points and Population with at
Least One Threshold Exceedance
Source
Water
Type
Ground
Water
Threshold
> 100 |jg/L
> 50 |jg/L
Number of
Sample
Points
Exceeding
Threshold
1
1
Percent of
Sample
Points
Exceeding
Threshold
0.005%
0.005%
Total Population-
Served by Sample
Points Exceeding
Threshold
20,000
20,000
Percent of
Population -Served
by Sample Points
Exceeding
Threshold
0.045%
0.045%
Surface
Water
>100|jg/L
> 50 |jg/L
2
3
0.056%
0.083%
7,977
8,549
0.011%
0.011%
Combined
Ground &
Surface
Water
>100|jg/L
> 50 |jg/L
3
4
0.012%
0.017%
27,977
28,549
0.024%
0.024%
11-7
-------
EPA-OGWDW Contaminant Occurrence Support Document for 2" 6-Year Review Category 1 Contaminants October 2009
11.4.2 Stage 2 Analysis of Occurrence Data
For the Stage 2 analyses, three different substitution values -zero, one-half the MRL
value, and the full MRL value- were used to replace each non-detection record. Three arithmetic
mean endothall concentrations were calculated at each system using the zero, one-half MRL and
full MRL substitution values. These mean calculations were done for all systems with data in the
Six-Year Review-ICR dataset. Then the percentages of all systems with a mean concentration
greater than each threshold were calculated.
Exhibit 11-8 presents the Stage 2 analysis of estimated system mean concentrations for
endothall occurrence in drinking water relative to 100 |ig/L and 50 |ig/L. Exhibit 11-9 presents
similar information based on population-served by the systems. Based on the Stage 2 analyses,
none of the systems had an estimated mean greater than 100 |ig/L. None of the ground water
systems had a mean exceeding 50 |ig/L. Among the surface water systems, one system had a
mean equal to or greater than 50 |ig/L. That system serves about 10,000 persons.
Exhibit 11-8: Endothall Stage 2 Analysis - Systems with a System Mean
Threshold Exceedance
Source Water
Type
Ground Water
Threshold
> 100|jg/L
> 50 |jg/L
Number of Systems with Mean
Exceeding Threshold1
Full MRL
0
0
1/2 MRL
0
0
Zero
0
0
Percent of Systems with Mean
Exceeding Threshold1
Full MRL
0.000%
0.000%
1/2 MRL
0.000%
0.000%
Zero
0.000%
0.000%
Surface Water
> 100|jg/L
> 50 |jg/L
0
1
0
1
0
1
0.000%
0.066%
0.000%
0.066%
0.000%
0.066%
Combined
Ground &
Surface Water
> 100|jg/L
> 50 |jg/L
0
1
0
1
0
1
0.000%
0.007%
0.000%
0.007%
0.000%
0.007%
1. The results shown in this table are based on system arithmetic means that were calculated using all sample
detection records and all non-detection records, with the non-detection records replaced (substituted) with
concentration values equal to the full MRL, 1A MRL, and zero.
11-8
-------
EPA-OGWDW Contaminant Occurrence Support Document for 2" 6-Year Review Category 1 Contaminants October 2009
Exhibit 11-9: Endothall Stage 2 Analysis - Population-Served by Systems with a
System Mean Threshold Exceedance
Source Water
Type
Ground Water
Threshold
> 100|jg/L
> 50 |jg/L
Population-Served by Systems
with Mean Exceeding
Threshold1
Full MRL
0
0
1/2 MRL
0
0
Zero
0
0
Percent of Population-Served by
Systems with Mean Exceeding
Threshold1
Full MRL
0.000%
0.000%
1/2 MRL
0.000%
0.000%
Zero
0.000%
0.000%
Surface Water
> 100|jg/L
> 50 |jg/L
0
10,000
0
10,000
0
10,000
0.000%
0.013%
0.000%
0.013%
0.000%
0.013%
Combined
Ground &
Surface Water
> 100|jg/L
> 50 |jg/L
0
10,000
0
10,000
0
10,000
0.000%
0.008%
0.000%
0.008%
0.000%
0.008%
1 The results shown in this table are based on system arithmetic means that were calculated using all sample
detection records and all non-detection records, with the non-detection records replaced (substituted) with
concentration values equal to the full MRL, 1A MRL, and zero.
11.5 Additional Occurrence Data
First Six-Year Review ofNPDWRs
Similar types of Stage 1 contaminant occurrence analyses were conducted for the first
Six-Year Review. The first Six-Year Review assessed regulated contaminant occurrence data
collected from 1993 to 1997. The first Six-Year Review data set consisted of data from 16 states
that were assembled into a "national cross-section" that was qualitatively indicative (though not
statistically representative) of national occurrence. For the first Six-Year Review (USEPA,
2003), 13 systems (approximately 0.15% of about 9,000 systems), reported at least one detection
of endothall. Three systems (approximately 0.03% of all systems), reported at least one detection
greater than the MCL concentration.
11.6 Summary of Data
Endothall is not listed as a TRI chemical; no release records are available. There are also
no ambient water data available from the NAWQA program for endothall.
The Stage 1 analysis of occurrence in drinking water indicates that endothall was detected
in 0.233% of 14,156 surface water and ground water systems. The median concentration, based
on detections from all systems, was 9.8 |ig/L. In total, 0.021% of systems (serving 0.063% of the
population) reported at least one detection greater than the MCL concentration of 100 |ig/L. Only
one ground water system, serving 20,000 persons, reported at least one detection greater than the
100 |ig/L or 50 |ig/L. Two surface water systems, serving about 55,000 persons, reported at least
11-9
-------
EPA-OGWDW Contaminant Occurrence Support Document for 2" 6-Year Review Category 1 Contaminants October 2009
one detection greater than 100 |ig/L. Three surface water systems, serving about 67,000 persons,
reported at least one detection greater than 50 |ig/L.
In the Stage 2 analysis, no ground water systems had a system mean concentration of
endothall exceeding either 100 |ig/L or 50 |ig/L. Surface water system means did not exceed 100
Hg/L. However, one surface water system (serving 10,000 persons) had a mean concentration
greater than 50 |ig/L.
11.7 References
Gilliom, R.J., I.E. Barbash, C.G. Crawford, P.A. Hamilton, J.D. Martin, N. Nakagaki, L.H.
Nowell, J.C. Scott, P.E. Stackelberg, G.P. Thelin, and D.M. Wolock. 2007. The Quality of Our
Nation's Waters - Pesticides in the Nation's Streams and Ground Water, 1992-2001. Appendix
7. Statistical summaries of water-quality data. U.S. Geological Survey Circular 1291, 172 p.
Available on the Internet at: http://water.usgs.gov/nawqa/pnsp/pubs/circ 129l/appendix7/.
Thelin, G.P. and L.P. Gianessi. 2000. Method for Estimating Pesticide Use for County Areas of
the Conterminous United States. U.S. Geological Survey Open-File Report 00-250, 62 p.
Available on the Internet at: http://ca.water.usgs.gov/pnsp/rep/ofr00250/ofr00250.pdf.
United States Environmental Protection Agency (USEPA). 1992. Drinking Water; National
Primary Drinking Water Regulations—Synthetic Organic Chemicals and Inorganic Chemicals;
National Primary Drinking Water Regulations Implementation; Final Rule. Federal Register.
Vol. 57, No. 138. p. 31776, July 17, 1992.
USEPA. 2003. Occurrence Estimation Methodology and Occurrence Findings for Six-Year
Review of National Primary Drinking Water Regulations. Office of Water. EPA-815-R-03-006.
June 2003.
USEPA. 2005. Reregistration Eligibility Decision (RED)-Endothall. Washington, DC: Office of
Prevention, Pesticides, and Toxic Substances. EPA-738-R-05-008. 210 p. Available on the
Internet at: http://www.epa.gov/oppsrrdl/REDs/endothall red.pdf
USEPA. 2009. The Analysis of Regulated Contaminant Occurrence Data from Public Water
Systems in Support of the Second Six-Year Review of National Primary Drinking Water
Regulations. Office of Water. EPA-815-B-09-006. October 2009.
United States Geological Survey (USGS). 2007. 2002 Pesticide Use Maps. Available on the
Internet at: http://water.usgs.gov/nawqa/pnsp/usage/maps/compound listing.php?year=02.
Accessed December 19, 2007.
11-10
-------
EPA-OGWDW Contaminant Occurrence Support Document for 2" 6-Year Review Category 1 Contaminants October 2009
12 Heptachlor
This chapter on heptachlor is part of a report that is organized so that each chapter
presents information on one specific contaminant. Chapter 1 of this report, Introduction, presents
background information on all sources of data used as well as the analytical approach used to
estimate contaminant occurrence in drinking water. This contaminant chapter includes
background information (such as regulatory history, general chemical information, and
environmental fate and behavior), use and environmental release information, and heptachlor
occurrence estimates in ambient water and drinking water. All drinking water occurrence
estimates were based on data from the National Compliance Monitoring Information Collection
Request (ICR) Dataset for the Second Six-Year Review (the "Six Year Review-ICR Dataset").
For detailed information regarding this dataset, data management, data quality assessments, and
the analytical approach used to generate all contaminant occurrence estimates presented in this
chapter, please refer to USEPA (2009).
12.1 Background
The United States Environmental Protection Agency (EPA) published the current
National Primary Drinking Water Regulations (NPDWR) for heptachlor on January 30, 1991 (56
FR 3526 (USEPA, 1991)). The NPDWR established a maximum contaminant level goal
(MCLG) of zero based on a cancer classification of B2, probable human carcinogen. The
NPDWR also established a maximum contaminant level (MCL) of 0.4 |ig/L based on analytical
feasibility.
According to the Agency for Toxic Substances and Disease Registry (ATSDR, 2007),
heptachlor (CioHsCly) is a white to tan powder that smells somewhat like camphor. It is not very
soluble in water and does not explode easily. In the past it was both a breakdown product and a
component of the pesticide chlordane (approximately 10% by weight). Since 1988, heptachlor
has been banned from all uses except fire ant control. The principal breakdown product of
heptachlor is heptachlor epoxide (ATSDR, 2007).
From 1953 to 1974, heptachlor entered the soil and surface water when farmers used it to
kill insects in seed grains and on crops. It also entered the air and soil due to use by homeowners
and professional exterminators for termite eradication. Heptachlor can enter the air, soil, ground
water, and surface water from leaks at hazardous waste sites or landfills (ATSDR, 2007).
Heptachlor adheres to soil very strongly and evaporates slowly into the air. When
heptachlor is released to the air, it can travel very long distances and can be deposited elsewhere.
In soils and water, heptachlor tends to break down into heptachlor epoxide, which, in addition to
being more toxic, can persist in the environment for many years. Both heptachlor and heptachlor
epoxide reside in fatty tissues and can build up in fish, cattle, and humans (ATSDR, 2007).
Exposure to heptachlor and heptachlor epoxide most commonly occurs through ingestion
of contaminated foods, which might include fish, shellfish, dairy products, meat, and poultry.
Contaminated fish and shellfish have been found to contain 2 to 750 ppb of heptachlor. Exposure
12-1
-------
EPA-OGWDW Contaminant Occurrence Support Document for 2" 6-Year Review Category 1 Contaminants October 2009
may also occur through water, air, and soil. In particular, those living in homes treated for
termites using heptachlor and those living near contaminated soils may inhale heptachlor and
heptachlor epoxide (ATSDR, 2007).
12.2 Use, Production, and Releases
Until 1988, heptachlor was used extensively for killing insects in homes, buildings, and
on food crops. Currently it can only be used for fire ant control in underground power
transformers (ATSDR, 2007).
The Chemical Substance Inventory Update System contains confidential data reported by
industry as a partial update of the Toxic Substances Control Act (TSCA) Inventory. The TSCA
Chemical Substance Inventory does not report production data for heptachlor.
Heptachlor is listed as a Toxics Release Inventory (TRI) chemical. Exhibit 12-la lists the
environmental releases of heptachlor from 1998 to 2006. (Exhibit 12-lb provides a graphical
presentation of the environmental releases.) On-site releases of heptachlor exceeded off-site
releases in every year except in 1999. Releases to land constituted most of the emissions for the
period of record (generally in excess of 95%) except in 1998, when air emissions peaked.
Relative to releases to land, releases to air, via underground injection, and by discharge to surface
water contributed negligible amounts. Off-site releases remained low throughout the period of
record. The TRI data for heptachlor were reported from 15 states, with 4 states (Arkansas, New
Jersey, Ohio, and Texas) reporting releases in all 9 years.
Exhibit 12-1 a: Table of Toxics Release Inventory (TRI) Data for Heptachlor
Year
1998
1999
2000
2001
2002
2003
2004
2005
2006
On-Site Releases (pounds)
Total Air
Emissions
137
5
7
6
33
8
2
5
5
Surface Water
Discharges
0
1
0
0
30
25
9
2
0
Underground
Injection
5
0
0
0
0
0
0
0
0
Releases to
Land
0
0
2,373
272
378
1,254
3,138
3,608
1,207
Off-site
Releases
(pounds)
12
14
10
28
2
2
1
3
1
Total On- &
Off-site
Releases
(pounds)
154
20
2,390
306
442
1,289
3,150
3,619
1,212
Source: USEPA, 2008
12-2
-------
EPA-OGWDW Contaminant Occurrence Support Document for 2" 6-Year Review Category 1 Contaminants October 2009
Exhibit 12-1b: Graph of Toxics Release Inventory (TRI) Data for Heptachlor
Environmental Releases of Heptachlor
A nnn
Pounds Released
•3 ^nn
•3 nnn
o ^nn
o nnn
1 ^nn
1 nnn
^nn
. .
•
~
n_ Fl
l-l
1998 1999 2000 2001 2002 2003
Year
i-i
I I I
2004 2005 2006
• Air Emissions DSurface Water Discharges D Underground Injection DReleases to Land BOff-site Releases
Source: USEPA, 2008
12.3 Occurrence in Ambient Water
No National Water Quality Assessment (NAWQA) Program data are available from the
United States Geological Survey (USGS) on the occurrence of heptachlor in ambient waters from
the Pesticide National Synthesis Project (Gilliom et a/., 2007) or the EPA Summary Analysis of
Cycle 1 data.
12.4 Occurrence in Drinking Water
Heptachlor is regulated as a synthetic organic chemical (SOC) in drinking water. All non-
purchased community water systems (CWSs) and non-transient non-community water systems
(NTNCWSs) are required to sample for SOCs. The maximum waiver period for SOCs is three
years, but this waiver can be renewed indefinitely. (Systems may receive a waiver at any time
without having collected an SOC sample. With a waiver, systems are not required to collect an
SOC sample.)
All CWSs and NTNCWSs without an SOC waiver must collect four consecutive
quarterly samples during the initial three-year compliance period.1 If all four samples are non-
detections, then a system serving less than 3,300 persons may reduce to one sample during each
consecutive compliance period; a system serving more than 3,300 persons may reduce to two
quarterly samples within a 12-month period during a 3-year compliance period. If a chemical is
detected (but is less than the MCL), the system must take one sample per quarter until results are
1 All new systems or systems using a new water source that began operation after January 22, 2004 must demonstrate
compliance with the MCL within a period of time specified by the State. The system must also comply with the
initial sampling frequencies specified by the State to ensure that a system can demonstrate compliance with the MCL.
12-3
-------
EPA-OGWDW Contaminant Occurrence Support Document for 2" 6-Year Review Category 1 Contaminants October 2009
below the MCL (minimum of two quarterly samples for ground water systems and four quarterly
samples for surface water systems). If all quarterly samples are below the MCL, the system may
return to annual sampling. If a chemical is detected at a level greater than the MCL, the system
(whether ground water or surface water) must take quarterly samples until four consecutive
quarters are below the MCL. If all quarterly samples are below the MCL, the system may return
to annual sampling.
The analysis of heptachlor occurrence presented in the following section is based on state
compliance monitoring data from the Six-Year Review-ICR Dataset. This is the largest and most
comprehensive compliance monitoring data set compiled by EPA's Drinking Water program to
date.
Forty-four of the Six-Year Review-ICR states provided occurrence data for heptachlor.
(There were no heptachlor data from Oklahoma.) These data consist of 133,726 analytical results
from 33,020 public water systems (PWSs) during the period from 1998 to 2005. The number of
sample results and systems vary by state, although the state data sets have been reviewed and
checked to ensure adequacy of coverage and completeness. The national modal minimum
reporting level (MRL) for heptachlor in the dataset is 0.04 |ig/L.
12.4.1 Stage 1 Analysis of Occurrence Data
Stage 1 analyses have been performed at the sample, sample point, and system levels.
Note that some systems have multiple sample points, such as source water sample points or entry
points to the distribution system. Exhibit 12-2 presents the number of samples, systems, and
sample points in the data set, as well as the population-served by the systems and sample points,
and the percentage of detections and the percentage of the population-served by the systems and
sample points with detections. Detections in systems provide a fundamental indication of
national occurrence, but are not representative of MCL violations. For heptachlor, 0.796% of
33,020 systems reported detections. Exhibit 12-3 lists the minimum, median, 90th percentile, and
maximum heptachlor concentrations based on detections from all systems in the Six-Year
Review-ICR Dataset. The median concentration of heptachlor, based on detections from all
systems, was 0.1 |ig/L.
Exhibit 12-2: Heptachlor Occurrence Data from the Six-Year Review-ICR Dataset
Source Water Type
Number of
Samples
Number of
Systems
Number of
Sample Points
Pop. Served
by Systems
Pop. Served
by Sample
Points
Total Number
Ground Water
106,106
29,965
48,681
67,922,151
67,922,151
Surface Water
27,620
3,055
6,733
116,521,633
116,521,633
Combined Ground &
Surface Water
133,726
33,020
55,414
184,443,784
184,443,784
12-4
-------
EPA-OGWDW Contaminant Occurrence Support Document for 2" 6-Year Review Category 1 Contaminants October 2009
Source Water Type
Number of
Samples
Number of
Systems
Number of
Sample Points
Pop. Served
by Systems
Pop. Served
by Sample
Points
Percent with Detections
Ground Water
0.491%
0.761%
0.768%
4.819%
4.179%
Surface Water
0.275%
1.146%
0.713%
4.528%
3.838%
Combined Ground &
Surface Water
0.446%
0.796%
0.762%
4.635%
3.964%
Exhibit 12-3: Heptachlor Summary Statistics
Source Water Type
Ground Water
Detection Value (ug/L)
Minimum
0.001
Median
0.1
90tn
Percentile
0.1
Maximum
0.5
Surface Water
0.001
0.1
0.12
0.5
Combined Ground &
Surface Water
0.001
0.1
0.1
0.5
The following maps illustrate the geographic distribution of heptachlor occurrence in
drinking water. Exhibit 12-4 shows the states with systems with at least one detection and
Exhibit 12-5 shows the states with systems with at least one detection greater than the MCL
concentration. Note that five states did not submit data for use in the Six-Year Review (Kansas2,
Louisiana, Mississippi, Pennsylvania, and Washington). Oklahoma did submit Six-Year data for
most contaminants; however, there is a statewide waiver for heptachlor in Oklahoma, so no
heptachlor data were available from that state.
The distribution of systems with detections of heptachlor is geographically dispersed.
Detection rates were generally low, with only three states detecting heptachlor in more than 1%
of their systems. Only one state reported detections of heptachlor greater than the MCL
concentration of 0.4 |ig/L.
2 Kansas submitted data after the data management and analyses had been conducted; data from Kansas are not
included in the analyses.
12-5
-------
EPA-OGWDW Contaminant Occurrence Support Document for 2" 6-Year Review Category 1 Contaminants October 2009
Exhibit 12-4: Percentage of Systems with at Least One Heptachlor Detection, by
State
\
•y/y/. States with Waivers or Not Submitting
States with No Detections
Less than 0.5% of Systems Detecting
BBj Between 0.5% and 1 % of Systems Detecting
BH More than 1% of Systems Detecting
Exhibit 12-5: Percentage of Systems with at Least One Heptachlor Detection
Greater than the MCL Concentration (> 0.4 \igiL), by State
States with Waivers or Not SubmRUng
States with Ho Detections > MCL
Less than 0.5% of Systems with Detections > MCL
•• Between 0.5% and 1% of Systems with Detections > MCL
JB More ttian 1% of Systems with Detections > MCL
Exhibit 12-6 presents the Stage 1 analysis of heptachlor occurrence in drinking water
from systems and the population-served by those systems in the Six-Year Review-ICR Dataset
12-6
-------
EPA-OGWDW Contaminant Occurrence Support Document for 2" 6-Year Review Category 1 Contaminants October 2009
relative to several thresholds. The thresholds are: 0.4 |ig/L, 0.2 |ig/L, and 0.1 |ig/L. Seven
(0.021% of) systems, serving about 268,000 persons, reported at least one detection greater than
or equal to 0.2 |ig/L; and 2 (0.006% of) systems, serving almost 4,700 persons, reported at least
one detection greater than 0.4 |ig/L. The numbers of surface and ground water systems reporting
at least one detection greater than 0.4 |ig/L were similarly low; however, because surface water
systems tend to be larger, more people were served by surface water systems than ground water
systems reporting at least one detection greater than 0.4 |ig/L.
Exhibit 12-6: Heptachlor Stage 1 Analysis - Systems and Population with at Least
One Threshold Exceedance
Source
Water Type
Ground
Water
Threshold
> 0.4 |jg/L
> 0.2 |jg/L
> 1 H9/L
Number of
Systems
Exceeding
Threshold
1
3
207
Percent of
Systems
Exceeding
Threshold
0.003%
0.010%
0.691%
Total Population-
Served by
Systems
Exceeding
Threshold
325
5,615
470,220
Percent of
Population-Served
by Systems
Exceeding
Threshold
0.000%
0.008%
0.692%
Surface
Water
> 0.4 |jg/L
>
> 1
1
4
27
0.033%
0.131%
0.884%
4,374
262,286
3,435,360
0.004%
0.225%
2.948%
Combined
Ground &
Surface
Water
> 0.4 |jg/L
>
> 1
2
7
234
0.006%
0.021%
0.709%
4,699
267,901
3,905,580
0.003%
0.145%
2.117%
The occurrence estimates presented above (in Exhibit 12-6) are based on the conservative
assumption that if a heptachlor detection is found in a single entry point/sample point in a
system, then the entire population-served by the system is assumed to potentially be exposed to
heptachlor (even though other entry points/sampling points at the system, and the population-
served by those entry/sample points, may exhibit no occurrence of heptachlor). The sample point
analysis is a less conservative estimate of the population-served by a system with a detection of
heptachlor because it estimates the potentially exposed population-served based only on the
proportion of the population-served by entry points/sample points with detections of heptachlor
(rather than the entire population-served by the system).
Exhibit 12-7 presents the Stage 1 analysis of heptachlor occurrence in drinking water
from sample points and population-served by those sample points in the Six-Year Review-ICR
Dataset. Nine (0.016% of) sample points, serving an estimated 261,000 persons, reported at least
12-7
-------
EPA-OGWDW Contaminant Occurrence Support Document for 2" 6-Year Review Category 1 Contaminants October 2009
one detection greater than or equal to 0.2 |ig/L and two (0.004% of) sample points, serving an
estimated 2,500 persons, reported at least one detection exceeding 0.4 |ig/L.
Exhibit 12-7: Heptachlor Stage 1 Analysis - Sample Points and Population with at
Least One Threshold Exceedance
Source
Water Type
Ground
Water
Threshold
> 0.4 |jg/L
>
> 1
Number of
Sample Points
Exceeding
Threshold
1
3
350
Percent of
Sample Points
Exceeding
Threshold
0.002%
0.006%
0.719%
Total Population-
Served by Sample
Points Exceeding
Threshold
325
1,186
206,377
Percent of
Population-
Served by
Sample Points
Exceeding
Threshold
0.000%
0.002%
0.304%
Surface
Water
> 0.4 |jg/L
>
> 1
1
6
36
0.015%
0.089%
0.535%
2,187
260,099
1,789,231
0.002%
0.223%
1.536%
Combined
Ground &
Surface
Water
> 0.4 |jg/L
>
> 1
2
9
386
0.004%
0.016%
0.697%
2,512
261,285
1,995,607
0.001%
0.142%
1.082%
12.4.2 Stage 2 Analysis of Occurrence Data
For the Stage 2 analyses, three different substitution values -zero, one-half the MRL
value, and the full MRL value- were used to replace each non-detection record. Three arithmetic
mean heptachlor concentrations were calculated at each system using the zero, one-half MRL and
full MRL substitution values. These mean calculations were done for all systems with data in the
Six-Year Review-ICR dataset. Then the percentages of all systems with a mean concentration
greater than each threshold were calculated.
Exhibit 12-8 presents the Stage 2 analysis of estimated system mean concentrations for
heptachlor in drinking water relative to several thresholds. The thresholds are: 0.4 |ig/L, 0.2
|ig/L, and 0.1 |ig/L. Exhibit 12-9 presents similar information based on population-served by the
systems. Based on the Stage 2 analyses, one ground water system (approximately 0.003% of all
systems) had an estimated system mean greater than or equal to 0.2 |ig/L and 0.4 |ig/L. This
system serves 325 persons.
12-8
-------
EPA-OGWDW Contaminant Occurrence Support Document for 2" 6-Year Review Category 1 Contaminants October 2009
Exhibit 12-8: Heptachlor Stage 2 Analysis - Systems with a System Mean
Threshold Exceedance
Source
Water Type
Ground
Water
Threshold
> 0.4 ug/L
>
> 1
Number of Systems with Mean
Exceeding Threshold1
Full MRL
1
1
41
1/2 MRL
1
1
41
Zero
1
1
41
Percent of Systems with Mean
Exceeding Threshold1
Full MRL
0.003%
0.003%
0.137%
1/2 MRL
0.003%
0.003%
0.137%
Zero
0.003%
0.003%
0.137%
Surface
Water
> 0.4 ug/L
>
> 1
0
0
1
0
0
1
0
0
1
0.000%
0.000%
0.033%
0.000%
0.000%
0.033%
0.000%
0.000%
0.033%
Combined
Ground &
Surface
Water
> 0.4 ug/L
>
> 1
1
1
42
1
1
42
1
1
42
0.003%
0.003%
0.127%
0.003%
0.003%
0.127%
0.003%
0.003%
0.127%
1. The results shown in this table are based on system arithmetic means that were calculated using all sample
detection records and all non-detection records, with the non-detection records replaced (substituted) with
concentration values equal to the full MRL, 14 MRL, and zero.
Exhibit 12-9: Heptachlor Stage 2 Analysis - Population-Served by Systems with a
System Mean Threshold Exceedance
Source
Water Type
Ground
Water
Threshold
> 0.4 ug/L
>
> 1
Population-Served by Systems
with Mean Exceeding Threshold1
Full MRL
325
325
23,112
1/2 MRL
325
325
23,112
Zero
325
325
23,112
Percent of Population -Served by
Systems with Mean Exceeding
Threshold1
Full MRL
0.0005%
0.0005%
0.034%
1/2 MRL
0.0005%
0.0005%
0.034%
Zero
0.0005%
0.0005%
0.034%
Surface
Water
> 0.4 ug/L
>
> 1
0
0
8,380
0
0
8,380
0
0
8,380
0.000%
0.000%
0.007%
0.000%
0.000%
0.007%
0.000%
0.000%
0.007%
Combined
Ground &
Surface
Water
> 0.4 ug/L
> 0.2 ug/L
> 1
325
325
31,492
325
325
31,492
325
325
31,492
0.0002%
0.0002%
0.017%
0.0002%
0.0002%
0.017%
0.0002%
0.0002%
0.017%
1. The results shown in this table are based on system arithmetic means that were calculated using all sample
detection records and all non-detection records, with the non-detection records replaced (substituted) with
concentration values equal to the full MRL, 1A MRL, and zero.
12-9
-------
EPA-OGWDW Contaminant Occurrence Support Document for 2" 6-Year Review Category 1 Contaminants October 2009
12.5 Additional Occurrence Data
First Six-Year Review ofNPDWRs
Similar types of Stage 1 contaminant occurrence analyses were conducted for the first
Six-Year Review. The first Six-Year Review assessed regulated contaminant occurrence data
collected from 1993 to 1997. The first Six-Year Review data set consisted of data from 16 states
that were assembled into a "national cross-section" that was qualitatively indicative (though not
statistically representative) of national occurrence. For the first Six-Year Review (USEPA,
2003), 12 systems (approximately 0.08% of about 14,000 systems), reported at least one
detection of heptachlor. One system (approximately 0.01% of all systems), reported at least one
detection greater than the MCL concentration.
12.6 Summary of Data
TRI data indicate that total off-site and on-site releases of heptachlor for the period 1998-
2006 ranged from 20 pounds to about 3,600 pounds. Releases to land constituted the
overwhelming majority of material released, except during 1998 and 1999. Amounts released
off-site were minor compared to releases to land.
There are no ambient water data available from the NAWQA program for heptachlor.
The Stage 1 analysis of occurrence in drinking water indicates that heptachlor was
detected in 0.796% of 33,020 surface and ground water systems. The median concentration,
based on detections from all systems, was 0.1 |ig/L. In total, 0.006% of systems (serving
approximately 0.003% of total the population) reported at least one detection greater than the
MCL concentration of 0.4 |ig/L. Similar numbers of ground water and surface water systems
reported at least one detection greater than the MCL concentration. However, because surface
water systems tend to be larger, a greater population was served by surface water systems
reporting at least one detection greater than the MCL concentration. A similar pattern was seen
for ground water and surface water sample points.
In the Stage 2 analysis, one ground water system (approximately 0.003% of all systems,
serving 0.0002% of the population) had an estimated system mean concentration of heptachlor
greater than the MCL concentration.
12.7 References
Agency for Toxic Substances and Disease Registry (ATSDR). 2007. Toxicological Profile for
Heptachlor and Heptachlor Epoxide. August 2007. Available on the Internet at:
http://www.atsdr.cdc.gov/toxprofiles/tpl2.html.
12-10
-------
EPA-OGWDW Contaminant Occurrence Support Document for 2" 6-Year Review Category 1 Contaminants October 2009
Gilliom, R.J., I.E. Barbash, C.G. Crawford, P.A. Hamilton, J.D. Martin, N. Nakagaki, L.H.
Nowell, J.C. Scott, P.E. Stackelberg, G.P. Thelin, and D.M. Wolock. 2007. The Quality of Our
Nation's Waters - Pesticides in the Nation's Streams and Ground Water, 1992-2001. Appendix
7. Statistical summaries of water-quality data. U.S. Geological Survey Circular 1291, 172 p.
Available on the Internet at: http://water.usgs.gov/nawqa/pnsp/pubs/circ 129l/appendix7/.
United States Environmental Protection Agency (USEPA). 1991. National Primary Drinking
Water Regulations—Synthetic Organic Chemicals and Inorganic Chemicals; Monitoring for
Unregulated Contaminants; National Primary Drinking Water Regulations Implementation;
National Secondary Drinking Water Regulations; Final Rule Federal Register. Vol. 56, No. 30. p.
3526, January 30, 1991.
USEPA. 2003. Occurrence Estimation Methodology and Occurrence Findings for Six-Year
Review of National Primary Drinking Water Regulations. Office of Water. EPA-815-R-03-006.
June 2003.
USEPA. 2008. TRI Explorer: Trends - Heptachlor. Released February 21, 2008. Available on the
Internet at: http://www.epa.gov/triexplorer/trends.htm. [Search for heptachlor.] Accessed
February 29, 2008.
USEPA. 2009. The Analysis of Regulated Contaminant Occurrence Data from Public Water
Systems in Support of the Second Six-Year Review of National Primary Drinking Water
Regulations. Office of Water. EPA-815-B-09-006. October 2009.
12-11
-------
EPA-OGWDW Contaminant Occurrence Support Document for 2" 6-Year Review Category 1 Contaminants October 2009
13 Heptachlor Epoxide
This chapter on heptachlor epoxide is part of a report that is organized so that each
chapter presents information on one specific contaminant. Chapter 1 of this report, Introduction,
presents background information on all sources of data used as well as the analytical approach
used to estimate contaminant occurrence in drinking water. This contaminant chapter includes
background information (such as regulatory history, general chemical information, and
environmental fate and behavior), use and environmental release information, and heptachlor
epoxide occurrence estimates in ambient water and drinking water. All drinking water occurrence
estimates were based on data from the National Compliance Monitoring Information Collection
Request (ICR) Dataset for the Second Six-Year Review (the "Six Year Review-ICR Dataset").
For detailed information regarding this dataset, data management, data quality assessments, and
the analytical approach used to generate all contaminant occurrence estimates presented in this
chapter, please refer to USEPA (2009).
13.1 Background
On January 30, 1991, the United States Environmental Protection Agency (EPA)
published the current National Primary Drinking Water Regulations (NPDWR) for heptachlor
epoxide, a product of heptachlor degradation (56 FR 3526 (USEPA, 1991)). The NPDWR
established a maximum contaminant level goal (MCLG) of zero based on a cancer classification
of B2, probable human carcinogen. The NPDWR also established a maximum contaminant level
(MCL) of 0.2 |ig/L based on analytical feasibility.
According to the Agency for Toxic Substances and Disease Registry (ATSDR, 2007),
heptachlor epoxide (CioHsClyO) is a white powder and the principal breakdown product of the
pesticide heptachlor, which is itself a breakdown product and a component of the pesticide
chlordane. Within hours after application, bacteria and other organisms degrade about 20% of the
applied heptachlor into heptachlor epoxide (ATSDR, 2007).
From 1953 to 1974, heptachlor epoxide entered the soil and surface water when farmers
used heptachlor to kill insects in seed grains and on crops. It also entered the air and soil when
homeowners and professional insect exterminators used heptachlor to kill termites. Heptachlor
epoxide can enter the air, soil, ground water, and surface water from leaks at hazardous waste
sites or landfills. It is more toxic than heptachlor and can persist in the environment for many
years. When heptachlor epoxide is released to the air, it can travel very long distances and can be
deposited elsewhere. Both heptachlor and heptachlor epoxide accumulate in fish, cattle, and
humans and can reside in fatty tissues for years (ATSDR, 2007).
Exposure to heptachlor and heptachlor epoxide most commonly occurs through ingestion
of contaminated foods, which might include fish, shellfish, dairy products, meat, and poultry.
Contaminated fish and shellfish have been found to contain 0.1 to 480 ppb of heptachlor epoxide.
Heptachlor epoxide has been found in human milk samples at levels ranging from 0.13 to 128
ppb. Exposure may also occur through water, air, and soil. In particular, those living in homes
13-1
-------
EPA-OGWDW Contaminant Occurrence Support Document for 2" 6-Year Review Category 1 Contaminants October 2009
treated for termites using heptachlor and those living near contaminated soils may inhale
heptachlor and heptachlor epoxide (ATSDR, 2007).
13.2 Use, Production, and Releases
Until 1988, heptachlor was used extensively for killing insects in homes, buildings, and
on food crops. Currently it can only be used for fire ant control in underground power
transformers (ATSDR, 2007). Heptachlor epoxide is not produced commercially, but rather is
formed by the chemical and biological transformation of heptachlor in the environment.
The Chemical Substance Inventory Update System contains confidential data reported by
industry as a partial update of the Toxic Substances Control Act (TSCA) Inventory. The TSCA
Chemical Substance Inventory does not report production data for heptachlor epoxide.
Heptachlor epoxide is not listed as a Toxics Release Inventory (TRI) chemical; thus no
TRI release records are available. Therefore, the use of heptachlor epoxide (described in the
previous section) may provide the primary indication of where releases are most likely to occur.
13.3 Occurrence in Ambient Water
No National Water Quality Assessment (NAWQA) Program data are available from the
United States Geological Survey (USGS) on the occurrence of heptachlor epoxide in ambient
waters from the Pesticide National Synthesis Project (Gilliom et a/., 2007) or the EPA Summary
Analysis of Cycle 1 data.
13.4 Occurrence in Drinking Water
Heptachlor epoxide is regulated as a synthetic organic chemical (SOC) in drinking water.
All non-purchased community water systems (CWSs) and non-transient non-community water
systems (NTNCWSs) are required to sample for SOCs. The maximum waiver period for SOCs is
three years, but this waiver can be renewed indefinitely. (Systems may receive a waiver at any
time without having collected an SOC sample. With a waiver, systems are not required to collect
an SOC sample.)
All CWSs and NTNCWSs without an SOC waiver must collect four consecutive
quarterly samples during the initial three-year compliance period.1 If all four samples are non-
detections, then a system serving less than 3,300 persons may reduce to one sample during each
consecutive compliance period; a system serving more than 3,300 persons may reduce to two
quarterly samples within a 12-month period during a 3-year compliance period. If a chemical is
detected (but is less than the MCL), the system must take one sample per quarter until results are
1 All new systems or systems using a new water source that began operation after January 22, 2004 must demonstrate
compliance with the MCL within a period of time specified by the State. The system must also comply with the
initial sampling frequencies specified by the State to ensure that a system can demonstrate compliance with the MCL.
13-2
-------
EPA-OGWDW Contaminant Occurrence Support Document for 2" 6-Year Review Category 1 Contaminants October 2009
below the MCL (minimum of two quarterly samples for ground water systems and four quarterly
samples for surface water systems). If all quarterly samples are below the MCL, the system may
return to annual sampling. If a chemical is detected at a level greater than the MCL, the system
(whether ground water or surface water) must take quarterly samples until four consecutive
quarters are below the MCL. If all quarterly samples are below the MCL, the system may return
to annual sampling.
The analysis of heptachlor epoxide occurrence presented in the following section is based
on state compliance monitoring data from the Six-Year Review-ICR Dataset. This is the largest
and most comprehensive compliance monitoring data set compiled by EPA's Drinking Water
program to date.
Forty-four of the Six-Year Review-ICR states provided occurrence data for heptachlor
epoxide. (There were no heptachlor epoxide data from Oklahoma.) These data consist of 134,708
analytical results from 33,015 public water systems (PWSs) during the period from 1998 to 2005.
The number of sample results and systems vary by state, although the state data sets have been
reviewed and checked to ensure adequacy of coverage and completeness. The national modal
minimum reporting level (MRL) for heptachlor epoxide in the dataset is 0.02 |ig/L.
13.4.1 Stage 1 Analysis of Occurrence Data
Stage 1 analyses have been performed at the sample, sample point, and system levels.
Note that some systems have multiple sample points, such as source water sample points or entry
points to the distribution system. Exhibit 13-1 presents the number of samples, systems, and
sample points in the data set, as well as the population-served by the systems and sample points,
and the percentage of detections and the percentage of the population-served by the systems and
sample points with detections. Detections in systems provide a fundamental indication of
national occurrence, but are not representative of MCL violations. For heptachlor epoxide,
0.221% of 33,015 systems reported detections. Exhibit 13-2 lists the minimum, median, 90th
percentile, and maximum heptachlor epoxide concentrations based on detections from all
systems in the Six-Year Review-ICR Dataset. The median concentration of heptachlor epoxide,
based on detections from all systems, was 0.03 |ig/L.
Exhibit 13-1: Heptachlor Epoxide Occurrence Data from the Six-Year Review-ICR
Dataset
Source Water Type
Number of
Samples
Number of
Systems
Number of
Sample Points
Pop. Served
by Systems
Pop. Served
by Sample
Points
Total Number
Ground Water
106,551
29,961
48,713
67,983,066
67,983,066
Surface Water
28,157
3,054
6,810
116,494,709
116,494,709
13-3
-------
EPA-OGWDW Contaminant Occurrence Support Document for 2" 6-Year Review Category 1 Contaminants October 2009
Source Water Type
Combined Ground &
Surface Water
Number of
Samples
134,708
Number of
Systems
33,015
Number of
Sample Points
55,523
Pop. Served
by Systems
184,477,775
Pop. Served
by Sample
Points
184,477,775
Percent with Detections
Ground Water
0.148%
0.194%
0.146%
4.608%
4.395%
Surface Water
0.071%
0.491%
0.279%
0.656%
0.503%
Combined Ground &
Surface Water
0.132%
0.221%
0.162%
2.112%
1.937%
Exhibit 13-2: Heptachlor Epoxide Summary Statistics
Source Water Type
Ground Water
Detection Value (ug/L)
Minimum
0.003
Median
0.03
90tn
Percentile
0.09
Maximum
2
Surface Water
0.001
0.03
0.2
0.5
Combined Ground &
Surface Water
0.001
0.03
0.1
2
The following maps illustrate the geographic distribution of heptachlor epoxide
occurrence in drinking water. Exhibit 13-3 shows the states with systems with at least one
detection and Exhibit 13-4 shows the states with systems with at least one detection greater than
the MCL concentration. Note that five states did not submit data for use in the Six-Year Review
(Kansas2, Louisiana, Mississippi, Pennsylvania, and Washington). Oklahoma did submit Six-
Year data for most contaminants; however, there is a statewide waiver for heptachlor epoxide in
Oklahoma so no data were available from that state.
The distribution of systems with detections of heptachlor epoxide is geographically
dispersed. Detection rates were generally low, with only four states having greater than 1% of
their systems detecting heptachlor epoxide. Only three states reported detections greater than the
MCL concentration of 0.2 |ig/L.
Kansas submitted data after the data management and analyses had been conducted; data from Kansas are not
included in the analyses.
13-4
-------
EPA-OGWDW Contaminant Occurrence Support Document for 2" 6-Year Review Category 1 Contaminants October 2009
Exhibit 13-3: Percentage of Systems with at Least One Heptachlor Epoxide
Detection, by State
States with Waivers or Not Submitting
States with No Detections
Less than 0.5% of Systems Detecting
| Between 0.5% and 1% of Systems Detecting
j More than 1% of Systems Detecting
Exhibit 13-4: Percentage of Systems with at Least One Heptachlor Epoxide
Detection Greater than the MCL Concentration (> 0.2 pg/L), by State
St*t« with Wslvert or Not Submitting
States with No Detections > MCL
] Less than 0 1% of System* with Detections > MCL
| Between 0.1% and 0.5% of Systems with Detections > MCL
j Hone *ian 0.5% of Systems wfth Detections > MCL
13-5
-------
EPA-OGWDW Contaminant Occurrence Support Document for 2" 6-Year Review Category 1 Contaminants October 2009
Exhibit 13-5 presents the Stage 1 analysis of heptachlor epoxide occurrence in drinking
water from systems and the population-served by those systems in the Six-Year Review-ICR
Dataset relative to two thresholds: 0.2 |ig/L and 0.1 |ig/L. Fifteen (0.045% of) systems, serving
more than 462,000 persons, reported at least one detection greater than or equal to 0.1 |ig/L; and
six (0.018% of) systems, serving more than 19,000 persons, reported at least one detection
greater than 0.2 |ig/L. More people were served by surface water systems with at least one
detection of heptachlor epoxide greater than or equal to 0.1 |ig/L, but more people were served
by ground water systems with at least one detection greater than 0.2 |ig/L.
Exhibit 13-5: Heptachlor Epoxide Stage 1 Analysis - Systems and Population with
at Least One Threshold Exceedance
Source
Water Type
Ground
Water
Threshold
> 0.2 |jg/L
> 1 |jg/L
Number of
Systems
Exceeding
Threshold
5
11
Percent of
Systems
Exceeding
Threshold
0.017%
0.037%
Total Population -
Served by
Systems
Exceeding
Threshold
14,973
22,458
Percent of
Population-Served
by Systems
Exceeding
Threshold
0.022%
0.033%
Surface
Water
> 0.2 |jg/L
> 1
1
4
0.033%
0.131%
4,374
439,554
0.004%
0.377%
Combined
Ground &
Surface
Water
> 0.2 |jg/L
> 1
6
15
0.018%
0.045%
19,347
462,012
0.010%
0.250%
The occurrence estimates presented above (in Exhibit 13-5) are based on the conservative
assumption that if a heptachlor epoxide detection is found in a single entry point/sample point in
a system, then the entire population-served by the system is assumed to potentially be exposed to
heptachlor epoxide (even though other entry points/sampling points at the system, and the
population-served by those entry/sample points, may exhibit no occurrence of heptachlor
epoxide). The sample point analysis is a less conservative estimate of the population-served by a
system with a detection of heptachlor epoxide because it estimates the potentially exposed
population-served based only on the proportion of the population-served by entry points/sample
points with detections of heptachlor epoxide (rather than the entire population-served by the
system).
Exhibit 13-6 presents the Stage 1 analysis of heptachlor epoxide occurrence in drinking
water from sample points and population-served by those sample points in the Six-Year Review-
ICR Dataset. Sixteen (0.029% of) sample points, serving an estimated 357,000 persons, reported
at least one detection greater than or equal to 0.1 |ig/L; and 6 (0.011% of) sample points, serving
an estimated 4,900 persons, reported at least one detection exceeding 0.2 |ig/L.
13-6
-------
EPA-OGWDW Contaminant Occurrence Support Document for 2" 6-Year Review Category 1 Contaminants October 2009
Exhibit 13-6: Heptachlor Epoxide Stage 1 Analysis - Sample Points and
Population with at Least One Threshold Exceedance
Source
Water Type
Ground
Water
Threshold
> 0.2 |jg/L
> 1
Number of
Sample Points
Exceeding
Threshold
5
11
Percent of
Sample Points
Exceeding
Threshold
0.010%
0.023%
Total Population-
Served by Sample
Points Exceeding
Threshold
2,703
4,062
Percent of
Population-
Served by
Sample Points
Exceeding
Threshold
0.004%
0.006%
Surface
Water
> 0.2 |jg/L
> 1 |jg/L
1
5
0.015%
0.073%
2,187
352,546
0.002%
0.303%
Combined
Ground &
Surface
Water
> 0.2 |jg/L
> 1
6
16
0.011%
0.029%
4,890
356,608
0.003%
0.193%
13.4.2 Stage 2 Analysis of Occurrence Data
For the Stage 2 analyses, three different substitution values -zero, one-half the MRL
value, and the full MRL value- were used to replace each non-detection record. Three arithmetic
mean heptachlor epoxide concentrations were calculated at each system using the zero, one-half
MRL and full MRL substitution values. These mean calculations were done for all systems with
data in the Six-Year Review-ICR dataset. Then the percentages of all systems with a mean
concentration greater than each threshold were calculated.
Exhibit 13-7 presents the Stage 2 analysis of estimated system mean concentrations for
heptachlor epoxide occurrence in drinking water relative to two thresholds: 0.2 |ig/L and 0.1
|ig/L. Exhibit 13-8 presents similar information based on population-served by the systems.
Based on the Stage 2 analyses, three ground water systems (approximately 0.009% of all
systems) had an estimated system mean greater than or equal to 0.1 |ig/L. These three systems
serve about 14,419 persons. One ground water system (0.003% of all systems), serving 325
persons, had an estimated system mean greater than 0.2 |ig/L. No surface water systems had
means exceeding the thresholds.
13-7
-------
EPA-OGWDW Contaminant Occurrence Support Document for 2" 6-Year Review Category 1 Contaminants October 2009
Exhibit 13-7: Heptachlor Epoxide Stage 2 Analysis - Systems with a System Mean
Threshold Exceedance
Source
Water Type
Ground
Water
Threshold
> 0.2 |jg/L
> 1
Number of Systems with Mean
Exceeding Threshold1
Full MRL
1
3
1/2 MRL
1
3
Zero
1
3
Percent of Systems with Mean
Exceeding Threshold1
Full MRL
0.003%
0.010%
1/2 MRL
0.003%
0.010%
Zero
0.003%
0.010%
Surface
Water
> 0.2 |jg/L
> 1
0
0
0
0
0
0
0.000%
0.000%
0.000%
0.000%
0.000%
0.000%
Combined
Ground &
Surface
Water
> 0.2 |jg/L
> 1
1
3
1
3
1
3
0.003%
0.009%
0.003%
0.009%
0.003%
0.009%
1 The results shown in this table are based on system arithmetic means that were calculated using all sample
detection records and all non-detection records, with the non-detection records replaced (substituted) with
concentration values equal to the full MRL, 14 MRL, and zero.
Exhibit 13-8: Heptachlor Epoxide Stage 2 Analysis - Population-Served by
Systems with a System Mean Threshold Exceedance
Source
Water Type
Ground
Water
Threshold
> 0.2 |jg/L
> 1
Population -Served by Systems
with Mean Exceeding Threshold1
Full MRL
325
14,419
1/2 MRL
325
14,419
Zero
325
14,419
Percent of Population -Served by
Systems with Mean Exceeding
Threshold1
Full MRL
0.0005%
0.021%
1/2 MRL
0.0005%
0.021%
Zero
0.0005%
0.021%
Surface
Water
> 0.2 |jg/L
> 1
0
0
0
0
0
0
0.000%
0.000%
0.000%
0.000%
0.000%
0.000%
Combined
Ground &
Surface
Water
> 0.2 |jg/L
> 1
325
14,419
325
14,419
325
14,419
0.0002%
0.008%
0.0002%
0.008%
0.0002%
0.008%
1. The results shown in this table are based on system arithmetic means that were calculated using all sample
detection records and all non-detection records, with the non-detection records replaced (substituted) with
concentration values equal to the full MRL, 1A MRL, and zero.
13-8
-------
EPA-OGWDW Contaminant Occurrence Support Document for 2" 6-Year Review Category 1 Contaminants October 2009
13.5 Additional Occurrence Data
First Six-Year Review ofNPDWRs
Similar types of Stage 1 contaminant occurrence analyses were conducted for the first
Six-Year Review. The first Six-Year Review assessed regulated contaminant occurrence data
collected from 1993 to 1997. The first Six-Year Review data set consisted of data from 16 states
that were assembled into a "national cross-section" that was qualitatively indicative (though not
statistically representative) of national occurrence. For the first Six-Year Review (USEPA,
2003), 13 systems (approximately 0.09% of about 14,000 systems), reported at least one
detection of heptachlor epoxide. Four systems (approximately 0.03% of all systems), reported at
least one detection greater than the MCL concentration.
13.6 Summary of Data
Heptachlor epoxide is not listed as a TRI chemical; no release records are available. No
data are available from the NAWQA program for heptachlor epoxide.
The Stage 1 analysis of occurrence in drinking water indicates that heptachlor epoxide
was detected in 0.221% of 33,015 surface water and ground water systems. The median
concentration, based on detections from all systems, was 0.03 |ig/L. In total, 0.018% of systems
(serving 0.010% of the population) reported at least one detection greater than the MCL
concentration of 0.2 |ig/L. A larger population was served by ground water systems than surface
water systems reporting at least one detection greater than the MCL concentration.
In the Stage 2 analysis, one ground water system (0.003% of systems, serving 0.0002% of
the total population) had an estimated system mean concentration of heptachlor epoxide greater
than the MCL concentration. No surface water systems had mean concentrations exceeding the
MCL concentration.
13.7 References
Agency for Toxic Substances and Disease Registry (ATSDR). 2007. Toxicological Profile for
Heptachlor and Heptachlor Epoxide. August 2007. Available on the Internet at:
http://www.atsdr.cdc.gov/toxprofiles/tpl2.html.
Gilliom, R.J., I.E. Barbash, C.G. Crawford, P.A. Hamilton, J.D. Martin, N. Nakagaki, L.H.
Nowell, J.C. Scott, P.E. Stackelberg, G.P. Thelin, and D.M. Wolock. 2007. The Quality of Our
Nation's Waters - Pesticides in the Nation's Streams and Ground Water, 1992-2001. Appendix
7. Statistical summaries of water-quality data. U.S. Geological Survey Circular 1291, 172 p.
Available on the Internet at: http://water.usgs.gov/nawqa/pnsp/pubs/circl291/appendix7/.
13-9
-------
EPA-OGWDW Contaminant Occurrence Support Document for 2" 6-Year Review Category 1 Contaminants October 2009
United States Environmental Protection Agency (USEPA). 1991. National Primary Drinking
Water Regulations—Synthetic Organic Chemicals and Inorganic Chemicals; Monitoring for
Unregulated Contaminants; National Primary Drinking Water Regulations Implementation;
National Secondary Drinking Water Regulations; Final Rule Federal Register. Vol. 56, No. 30. p.
3526, January 30, 1991.
USEPA. 2003. Occurrence Estimation Methodology and Occurrence Findings for Six-Year
Review of National Primary Drinking Water Regulations. Office of Water. EPA-815-R-03-006.
June 2003.
USEPA. 2009. The Analysis of Regulated Contaminant Occurrence Data from Public Water
Systems in Support of the Second Six-Year Review of National Primary Drinking Water
Regulations. Office of Water. EPA-815-B-09-006. October 2009.
13-10
-------
EPA-OGWDW Contaminant Occurrence Support Document for 2" 6-Year Review Category 1 Contaminants October 2009
14 Hexachlorobenzene
This chapter on hexachlorobenzene is part of a report that is organized so that each
chapter presents information on one specific contaminant. Chapter 1 of this report, Introduction,
presents background information on all sources of data used as well as the analytical approach
used to estimate contaminant occurrence in drinking water. This contaminant chapter includes
background information (such as regulatory history, general chemical information, and
environmental fate and behavior), use and environmental release information, and
hexachlorobenzene occurrence estimates in ambient water and drinking water. All drinking water
occurrence estimates were based on data from the National Compliance Monitoring Information
Collection Request (ICR) Dataset for the Second Six-Year Review (the "Six Year Review-ICR
Dataset"). For detailed information regarding this dataset, data management, data quality
assessments, and the analytical approach used to generate all contaminant occurrence estimates
presented in this chapter, please refer to USEPA (2009).
14.1 Background
The United States Environmental Protection Agency (EPA) published the current
National Primary Drinking Water Regulations (NPDWR) for hexachlorobenzene on July 17,
1992 (57 FR 31776 (USEPA, 1992)). The NPDWR established a maximum contaminant level
goal (MCLG) of zero based on a cancer classification of B2, probable human carcinogen. The
NPDWR also established a maximum contaminant level (MCL) of 1 |ig/L based on analytical
feasibility.
According to the Agency for Toxic Substances and Disease Registry (ATSDR, 2002),
hexachlorobenzene (CeCle) is a white crystalline solid that does not occur naturally in the
environment. It is formed as a by-product during the manufacturing of chemical solvents, other
chlorine-containing compounds, and pesticides. Small amounts of hexachlorobenzene can also be
produced as a by-product during combustion processes, such as waste incineration, and during
certain wood preservation processes (ATSDR, 2002).
Hexachlorobenzene breaks down very slowly in the environment. Its half-life in soil is
three to six years. In surface water its half-life is 2.7 to 5.7 years, and in ground water it is 5.3 to
11.4 years. It tends to bind to sediments at the bottom of lakes because it is relatively insoluble in
water. Hexachlorobenzene tends not to evaporate, but once in the air, it may be carried over large
distances (ATSDR, 2002).
Exposure to hexachlorobenzene can occur near industrial sites where it is produced as an
unintentional by-product or as a minor part of another chemical product. Levels in urban air are
typically very low (about 0.03 ppb). Exposure by inhalation is estimated to be 0.01 |ig/kg/year.
Contact with contaminated soils and the ingestion of contaminated foods are other modes of
exposure. Hexachlorobenzene's poor solubility in water helps mitigate its occurrence in drinking
water (ATSDR, 2002).
14-1
-------
EPA-OGWDW Contaminant Occurrence Support Document for 2" 6-Year Review Category 1 Contaminants October 2009
14.2 Use, Production, and Releases
Although it is not used commercially in the United States today, hexachlorobenzene was
widely used as a pesticide until 1965. It was also used to make fireworks, ammunition, and
synthetic rubber (ATSDR, 2002).
The Toxic Substances Control Act (TSCA) Inventory contains confidential data reported
by industry as a partial update of the Chemical Substance Inventory Update System. The data are
reported under the Inventory Update Reporting regulation. Manufacturers and importers are
required to report company information and chemical information, including production volumes
for chemicals they manufactured or imported in excess of 10,000 pounds in the immediately
preceding fiscal year.
The TSCA Chemical Substance Inventory does not provide actual production volumes;
instead, it indicates at which of five production volume ranges a chemical was manufactured or
imported. The TSCA Inventory is updated every four years. Exhibit 14-1 presents the production
of hexachlorobenzene in the United States from 1986 to 2002. Note that the production of
hexachlorobenzene in the United States was not documented in the Chemical Update System
until 1998 and its production volume has remained within the same range since then.
Exhibit 14-1: Production of Hexachlorobenzene in the United States
Inventory Update Year
1986 Range
1990 Range
1994 Range
1998 Range
2002 Range
Production Volume
No Reports
No Reports
No Reports
10,000-500,000 pounds
10,000-500,000 pounds
Source: USEPA, 2003a
Hexachlorobenzene is listed as a Toxics Release Inventory (TRI) chemical. Exhibit 14-2a
lists the environmental releases of hexachlorobenzene from 1998 to 2006. (Exhibit 14-2b
provides a graphical presentation of the environmental releases.) Most releases were to land or
off-site. Air emissions were 5% or less of the total; surface water discharges and releases via
underground injection were even smaller. The TRI data for hexachlorobenzene were reported
from 37 states, with 7 states (California, Kentucky, Louisiana, New Jersey, Ohio, Tennessee, and
Texas) reporting releases in all 9 years.
14-2
-------
EPA-OGWDW Contaminant Occurrence Support Document for 2" 6-Year Review Category 1 Contaminants October 2009
Exhibit 14-2a: Table of Toxics Release Inventory (TRI) Data for
Hexachlorobenzene
Year
1998
1999
2000
2001
2002
2003
2004
2005
2006
On-Site Releases (pounds)
Total Air
Emissions
386
571
1,994
1,199
1,138
1,197
1,169
842
412
Surface Water
Discharges
4
8
331
322
397
120
133
69
71
Underground
Injection
0
0
48
22
12
7
19
429
2
Releases to
Land
96
13,023
22,700
23,525
11,725
15,214
28,733
34,366
27,377
Off -site
Releases
(pounds)
13,328
13,559
14,253
18,073
8,999
18,795
18,570
27,307
6,497
Total On- &
Off-site
Releases
(pounds)
13,814
27,161
39,327
43,141
22,270
35,334
48,624
63,013
34,359
Source: USEPA, 2008
Exhibit 14-2b: Graph of Toxics Release Inventory (TRI) Data for
Hexachlorobenzene
Pounds Released
Environmental Releases of Hexachlorobenzene
oc nnn -,
on nnn
oc nnn
9n nnn -
-i c nnn
•in nnn
^ nnn -
Im
ri
„ t
-
•
^
~
:
\
1998 1999 2000 2001 2002 2003 2004 2005 2006
Year
•Air Emissions D Surface Water Discharges DUnderground Injection DReleases to Land BOff-site Releases
Source: USEPA, 2008
14.3 Occurrence in Ambient Water
No National Water Quality Assessment (NAWQA) Program data are available from the
United States Geological Survey (USGS) on the occurrence of hexachlorobenzene in ambient
14-3
-------
EPA-OGWDW Contaminant Occurrence Support Document for 2" 6-Year Review Category 1 Contaminants October 2009
waters from the Pesticide National Synthesis Project (Gilliom et a/., 2007) or the EPA Summary
Analysis of Cycle 1 data.
14.4 Occurrence in Drinking Water
Hexachlorobenzene is regulated as a synthetic organic chemical (SOC) in drinking water.
All non-purchased community water systems (CWSs) and non-transient non-community water
systems (NTNCWSs) are required to sample for SOCs. The maximum waiver period for SOCs is
three years, but this waiver can be renewed indefinitely. (Systems may receive a waiver at any
time without having collected an SOC sample. With a waiver, systems are not required to collect
an SOC sample.)
All CWSs and NTNCWSs without an SOC waiver must collect four consecutive
quarterly samples during the initial three-year compliance period. * If all four samples are non-
detections, then a system serving less than 3,300 persons may reduce to one sample during each
consecutive compliance period; a system serving more than 3,300 persons may reduce to two
quarterly samples within a 12-month period during a 3-year compliance period. If a chemical is
detected (but is less than the MCL), the system must take one sample per quarter until results are
below the MCL (minimum of two quarterly samples for ground water systems and four quarterly
samples for surface water systems). If all quarterly samples are below the MCL, the system may
return to annual sampling. If a chemical is detected at a level greater than the MCL, the system
(whether ground water or surface water) must take quarterly samples until four consecutive
quarters are below the MCL. If all quarterly samples are below the MCL, the system may return
to annual sampling.
The analysis of hexachlorobenzene occurrence presented in the following section is
based on state compliance monitoring data from the Six-Year Review-ICR Dataset. This is the
largest and most comprehensive compliance monitoring data set compiled by EPA's Drinking
Water program to date.
Forty-four of the Six-Year Review-ICR states provided occurrence data for
hexachlorobenzene. (There were no hexachlorobenzene data from Oklahoma.) These data
consist of 135,512 analytical results from 32,826 public water systems (PWSs) during the period
from 1998 to 2005. The number of sample results and systems vary by state, although the state
data sets have been reviewed and checked to ensure adequacy of coverage and completeness.
The national modal minimum reporting level (MRL) for hexachlorobenzene in the dataset is 0.1
1 All new systems or systems using a new water source that began operation after January 22, 2004 must demonstrate
compliance with the MCL within a period of time specified by the State. The system must also comply with the
initial sampling frequencies specified by the State to ensure that a system can demonstrate compliance with the
MCL.
14-4
-------
EPA-OGWDW Contaminant Occurrence Support Document for 2" 6-Year Review Category 1 Contaminants October 2009
14.4.1 Stage 1 Analysis of Occurrence Data
Stage 1 analyses have been performed at the sample, sample point, and system levels.
Note that some systems have multiple sample points, such as source water sample points or entry
points to the distribution system. Exhibit 14-3 presents the number of samples, systems, and
sample points in the data set, as well as the population-served by the systems and sample points,
and the percentage of detections and the percentage of the population-served by the systems and
sample points with detections. Detections in systems provide a fundamental indication of
national occurrence, but are not representative of MCL violations. For hexachlorobenzene,
0.341% of 32,826 systems reported detections. Exhibit 14-4 lists the minimum, median, 90th
percentile, and maximum hexachlorobenzene concentrations based on detections from all
systems in the Six-Year Review-ICR Dataset. The median concentration of hexachlorobenzene,
based on detections from all systems, was 0.1 |ig/L.
Exhibit 14-3: Hexachlorobenzene Occurrence Data from the Six-Year Review-ICR
Dataset
Source Water Type
Number of
Samples
Number of
Systems
Number of
Sample Points
Pop. Served
by Systems
Pop. Served
by Sample
Points
Total Number
Ground Water
106,879
29,793
48,753
67,957,193
67,957,193
Surface Water
28,633
3,033
6,791
116,167,630
116,167,630
Combined Ground &
Surface Water
135,512
32,826
55,544
184,124,823
184,124,823
Percent with Detections
Ground Water
0.067%
0.205%
0.135%
4.099%
3.946%
Surface Water
0.234%
1 .682%
0.825%
0.988%
0.878%
Combined Ground &
Surface Water
0.103%
0.341%
0.220%
2.136%
2.011%
14-5
-------
EPA-OGWDW Contaminant Occurrence Support Document for 2" 6-Year Review Category 1 Contaminants October 2009
Exhibit 14-4: Hexachlorobenzene Summary Statistics
Source Water Type
Ground Water
Detection Value (ug/L)
Minimum
0.002
Median
0.045
90tn
Percentile
0.5
Maximum
10
Surface Water
0.01
0.5
0.5
2
Combined Ground &
Surface Water
0.002
0.1
0.5
10
The following maps illustrate the geographic distribution of hexachlorobenzene
occurrence in drinking water. Exhibit 14-5 shows the states with systems with at least one
detection and Exhibit 14-6 shows the states with systems with at least one detection greater than
the MCL concentration. Note that five states did not submit data for use in the Six-Year Review
(Kansas2, Louisiana, Mississippi, Pennsylvania, and Washington). Oklahoma did submit Six-
Year data for most contaminants. There is, however, a statewide waiver for hexachlorobenzene
in Oklahoma so no hexachlorobenzene data were available from that state.
The distribution of systems with detections of hexachlorobenzene is geographically
dispersed. Detection rates were generally low, with only six states reporting more than 0.5% of
their systems with hexachlorobenzene detections. Only four states reported detections greater
than the MCL concentration of 1 |ig/L.
Kansas submitted data after the data management and analyses had been conducted; data from Kansas are not
included in the analyses.
14-6
-------
EPA-OGWDW Contaminant Occurrence Support Document for 2" 6-Year Review Category 1 Contaminants October 2009
Exhibit 14-5: Percentage of Systems with at Least One Hexachlorobenzene
Detection, by State
State* with Waivers or Not Submitting
States with No Detections
Less than 0.1% of Systems Detecting
^BJ Between 0.1 % and 0.5% of Syste ms Detecting
^B More than 0-5% of Systems Detecting
Exhibit 14-6: Percentage of Systems with at Least One Hexachlorobenzene
Detection Greater than the MCL Concentration (> 1 pg/L), by State
St*t« with Wslvert or Not Submitting
States wl MCL
| Between 0.1% and O.S% of Systems with Detections > MCL
j Hflne dran 0.5% of Systems with Detection 5 * MCL
14-7
-------
EPA-OGWDW Contaminant Occurrence Support Document for 2" 6-Year Review Category 1 Contaminants October 2009
Exhibit 14-7 presents the Stage 1 analysis of hexachlorobenzene occurrence in drinking
water from systems and the population-served by those systems in the Six-Year Review-ICR
Dataset relative to several thresholds. The thresholds are: 1 |ig/L, 0.5 |ig/L, and 0.1 |ig/L. Forty-
six systems (0.140% of all systems), serving about 617,000 persons, reported at least one
detection greater than or equal to 0.5 |ig/L; and four systems (0.012% of all systems), serving
about 5,000 persons, reported at least one detection greater than the MCL concentration of 1
|ig/L. Hexachlorobenzene occurred in more surface water systems than ground water systems at
levels greater than or equal to 0.5 jig/L and 0.1 |ig/L. Very few systems of either type reported at
least one detection greater than 1 |ig/L.
Exhibit 14-7: Hexachlorobenzene Stage 1 Analysis - Systems and Population with
at Least One Threshold Exceedance
Source
Water Type
Ground
Water
Threshold
> 1 pg/L
> 0.5 |jg/L
>0.1 |jg/L
Number of
Systems
Exceeding
Threshold
3
12
18
Percent of
Systems
Exceeding
Threshold
0.010%
0.040%
0.060%
Total Population -
Served by
Systems
Exceeding
Threshold
2,273
18,893
23,963
Percent of
Population -Served
by Systems
Exceeding
Threshold
0.003%
0.028%
0.035%
Surface
Water
> 1 pg/L
>
>0.1 |jg/L
1
34
39
0.033%
1.121%
1 .286%
2,671
597,830
912,379
0.002%
0.515%
0.785%
Combined
Ground &
Surface
Water
> 1 pg/L
>
>0.1 pg/L
4
46
57
0.012%
0.140%
0.174%
4,944
616,723
936,342
0.003%
0.335%
0.509%
The occurrence estimates presented above (in Exhibit 14-7) are based on the conservative
assumption that if a hexachlorobenzene detection is found in a single entry point/sample point in
a system, then the entire population-served by the system is assumed to potentially be exposed to
hexachlorobenzene (even though other entry points/sampling points at the system, and the
population-served by those entry/sample points, may exhibit no occurrence of
hexachlorobenzene). The sample point analysis is a less conservative estimate of the population-
served by a system with a detection of hexachlorobenzene because it estimates the potentially
exposed population-served based only on the proportion of the population-served by entry
points/sample points with detections of hexachlorobenzene (rather than the entire population-
served by the system).
Exhibit 14-8 presents the Stage 1 analysis of hexachlorobenzene occurrence in drinking
water from sample points and population-served by those sample points in the Six-Year Review-
ICR Dataset. Forty-six (0.083% of) sample points, serving an estimated 605,000 persons,
14-8
-------
EPA-OGWDW Contaminant Occurrence Support Document for 2" 6-Year Review Category 1 Contaminants October 2009
reported at least one detection greater than or equal to 0.5 |ig/L; and four (0.007% of) sample
points, serving an estimated 4,900 persons, reported at least one detection exceeding 1 |ig/L.
Similar to the system-level occurrence analysis, very few sample points reported detections at
levels greater than 1 |ig/L.
Exhibit 14-8: Hexachlorobenzene Stage 1 Analysis - Sample Points and
Population with at Least One Threshold Exceedance
Source
Water Type
Ground
Water
Threshold
> 1 pg/L
>
>0.1 |jg/L
Number of
Sample Points
Exceeding
Threshold
3
12
18
Percent of
Sample Points
Exceeding
Threshold
0.006%
0.025%
0.037%
Total Population-
Served by Sample
Points Exceeding
Threshold
2,273
15,610
20,250
Percent of
Population-
Served by
Sample Points
Exceeding
Threshold
0.003%
0.023%
0.030%
Surface
Water
> 1 pg/L
> 0.5 |jg/L
>0.1 |jg/L
1
34
41
0.015%
0.501%
0.604%
2,671
589,039
858,215
0.002%
0.507%
0.739%
Combined
Ground &
Surface
Water
> 1 pg/L
>
>0.1 pg/L
4
46
59
0.007%
0.083%
0.106%
4,944
604,649
878,465
0.003%
0.328%
0.477%
14.4.2 Stage 2 Analysis of Occurrence Data
For the Stage 2 analyses, three different substitution values -zero, one-half the MRL
value, and the full MRL value- were used to replace each non-detection record. Three arithmetic
mean hexachlorobenzene concentrations were calculated at each system using the zero, one-half
MRL and full MRL substitution values. These mean calculations were done for all systems with
data in the Six-Year Review-ICR dataset. Then the percentages of all systems with a mean
concentration greater than each threshold were calculated.
Exhibit 14-9 presents the Stage 2 analysis of estimated system mean concentrations for
hexachlorobenzene occurrence in drinking water relative to several thresholds. The thresholds
are: 1 |ig/L, 0.5 |ig/L, and 0.1 |ig/L. Exhibit 14-10 presents similar information based on
population-served by the systems. Based on the Stage 2 analyses, four systems (approximately
0.012% of all systems) had an estimated system mean greater than or equal to 0.5 |ig/L.
Together, these four systems serve about 4,944 persons. Three systems (approximately 0.009%
of all systems), serving 2,273 persons, had an estimated system mean greater than the MCL
concentration of 1 |ig/L.
14-9
-------
EPA-OGWDW Contaminant Occurrence Support Document for 2" 6-Year Review Category 1 Contaminants October 2009
Exhibit 14-9: Hexachlorobenzene Stage 2 Analysis - Systems with a System Mean
Threshold Exceedance
Source
Water Type
Ground
Water
Threshold
> 1 pg/L
>
>0.1 |jg/L
Number of Systems with Mean
Exceeding Threshold1
Full MRL
3
3
20
1/2 MRL
3
3
8
Zero
3
3
8
Percent of Systems with Mean
Exceeding Threshold1
Full MRL
0.010%
0.010%
0.067%
1/2 MRL
0.010%
0.010%
0.027%
Zero
0.010%
0.010%
0.027%
Surface
Water
> 1 pg/L
>
>0.1 |jg/L
0
1
39
0
1
8
0
1
1
0.000%
0.033%
1 .286%
0.000%
0.033%
0.264%
0.000%
0.033%
0.033%
Combined
Ground &
Surface
Water
> 1 pg/L
>
>0.1 pg/L
3
4
59
3
4
16
3
4
9
0.009%
0.012%
0.180%
0.009%
0.012%
0.049%
0.009%
0.012%
0.027%
1. The results shown in this table are based on system arithmetic means that were calculated using all sample
detection records and all non-detection records, with the non-detection records replaced (substituted) with
concentration values equal to the full MRL, 1A MRL, and zero.
Exhibit 14-10: Hexachlorobenzene Stage 2 Analysis - Population-Served by
Systems with a System Mean Threshold Exceedance
Source
Water Type
Ground
Water
Threshold
> 1 pg/L
>
>0.1 pg/L
Population-Served by Systems
with Mean Exceeding Threshold1
Full MRL
2,273
2,273
639,313
1/2 MRL
2,273
2,273
6,741
Zero
2,273
2,273
6,741
Percent of Population-Served by
Systems with Mean Exceeding
Threshold1
Full MRL
0.003%
0.003%
0.941%
1/2 MRL
0.003%
0.003%
0.010%
Zero
0.003%
0.003%
0.010%
Surface
Water
> 1 pg/L
>
>0.1 pg/L
0
2,671
2,485,099
0
2,671
87,448
0
2,671
2,671
0.000%
0.002%
2.139%
0.000%
0.002%
0.075%
0.000%
0.002%
0.002%
14-10
-------
EPA-OGWDW Contaminant Occurrence Support Document for 2" 6-Year Review Category 1 Contaminants October 2009
Source
Water Type
Threshold
Population-Served by Systems
with Mean Exceeding Threshold1
Full MRL
1/2 MRL
Zero
Percent of Population-Served by
Systems with Mean Exceeding
Threshold1
Full MRL
1/2 MRL
Zero
Combined
Ground &
Surface
Water
> 1 pg/L
>
>0.1 |jg/L
2,273
4,944
3,124,412
2,273
4,944
94,189
2,273
4,944
9,412
0.001%
0.003%
1.697%
0.001%
0.003%
0.051%
0.001%
0.003%
0.005%
1. The results shown in this table are based on system arithmetic means that were calculated using all sample
detection records and all non-detection records, with the non-detection records replaced (substituted) with
concentration values equal to the full MRL, 14 MRL, and zero.
14.5 Additional Occurrence Data
First Six-Year Review ofNPDWRs
Similar types of Stage 1 contaminant occurrence analyses were conducted for the first
Six-Year Review. The first Six-Year Review assessed regulated contaminant occurrence data
collected from 1993 to 1997. The first Six-Year Review data set consisted of data from 16 states
that were assembled into a "national cross-section" that was qualitatively indicative (though not
statistically representative) of national occurrence. For the first Six-Year Review (USEPA,
2003b), 13 systems (approximately 0.09% of about 14,000 systems), reported at least one
detection of hexachlorobenzene. One system (approximately 0.01% of all systems), reported at
least one detection greater than the MCL concentration.
14.6 Summary of Data
TRI data indicate that total off-site and on-site releases of hexachlorobenzene for the
period 1998-2006 ranged from approximately 14,000 pounds to approximately 63,000 pounds.
Releases to land and off-site releases constituted the majority of total on- and off-site releases.
Air emissions and underground injection were less than 5% of the total releases.
Ambient water quality data from the NAWQA program were not available for
hexachlorobenzene.
The Stage 1 analysis of occurrence in drinking water indicates that hexachlorobenzene
was detected in 0.341% of 32,826 surface and ground water systems. The median concentration,
based on detections from all systems, was 0.1 |ig/L. In total, 0.012% of systems (serving 0.003%
of the population) reported at least one detection greater than the MCL concentration of 1 |ig/L.
In the Stage 2 analysis, approximately 0.009% of systems (serving 0.001% of the
population) had an estimated system mean concentration of hexachlorobenzene greater than the
MCL concentration. No surface water systems had mean concentrations exceeding the MCL
14-11
-------
EPA-OGWDW Contaminant Occurrence Support Document for 2" 6-Year Review Category 1 Contaminants October 2009
concentration. Only 3 ground water systems, serving 2,273 persons, had mean concentrations
exceeding the MCL concentration.
14.7 References
Agency for Toxic Substances and Disease Registry (ATSDR). 2002. Toxicological Profile for
Hexachlorobenzene. September 2002. Available on the Internet at:
http://www.atsdr.cdc.gov/toxprofiles/tp90.html.
Gilliom, R.J., I.E. Barbash, C.G. Crawford, P.A. Hamilton, J.D. Martin, N. Nakagaki, L.H.
Nowell, J.C. Scott, P.E. Stackelberg, G.P. Thelin, and D.M. Wolock. 2007. The Quality of Our
Nation's Waters - Pesticides in the Nation's Streams and Ground Water, 1992-2001. Appendix
7. Statistical summaries of water-quality data. U.S. Geological Survey Circular 1291, 172 p.
Available on the Internet at: http://water.usgs.gov/nawqa/pnsp/pubs/circ 129l/appendix7/.
United States Environmental Protection Agency (USEPA). 1992. Drinking Water; National
Primary Drinking Water Regulations—Synthetic Organic Chemicals and Inorganic Chemicals;
National Primary Drinking Water Regulations Implementation; Final Rule. Federal Register.
Vol. 57, No. 138. p. 31776, July 17, 1992.
USEPA. 2003a. Non-confidential Inventory Update Reporting (IUR) Production Volume
Information - Hexachlorobenzene. Last modified September 10, 2007. Available on the Internet
at: http://www.epa.gov/oppt/iur/tools/data/2002-vol.htm. [Search for hexachlorobenzene.]
Accessed February 21, 2008.
USEPA. 2003b. Occurrence Estimation Methodology and Occurrence Findings for Six-Year
Review of National Primary Drinking Water Regulations. Office of Water. EPA-815-R-03-006.
June 2003.
USEPA. 2008. TRI Explorer: Trends - Hexachlorobenzene. Released February 21, 2008.
Available on the Internet at: http://www.epa.gov/triexplorer/trends.htm. [Search for
hexachlorobenzene.] Accessed February 29, 2008.
USEPA. 2009. The Analysis of Regulated Contaminant Occurrence Data from Public Water
Systems in Support of the Second Six-Year Review of National Primary Drinking Water
Regulations. Office of Water. EPA-815-B-09-006. October 2009.
14-12
-------
EPA-OGWDW Contaminant Occurrence Support Document for 2" 6-Year Review Category 1 Contaminants October 2009
15 Hexachlorocyclopentadiene
This chapter on hexachlorocyclopentadiene is part of a report that is organized so that
each chapter presents information on one specific contaminant. Chapter 1 of this report,
Introduction, presents background information on all sources of data used as well as the
analytical approach used to estimate contaminant occurrence in drinking water. This contaminant
chapter includes background information (such as regulatory history, general chemical
information, and environmental fate and behavior), use and environmental release information,
and hexachlorocyclopentadiene occurrence estimates in ambient water and drinking water. All
drinking water occurrence estimates were based on data from the National Compliance
Monitoring Information Collection Request (ICR) Dataset for the Second Six-Year Review (the
"Six Year Review-ICR Dataset"). For detailed information regarding this dataset, data
management, data quality assessments, and the analytical approach used to generate all
contaminant occurrence estimates presented in this chapter, please refer to USEPA (2009).
15.1 Background
The United States Environmental Protection Agency (EPA) published the current
National Primary Drinking Water Regulations (NPDWR) for hexachlorocyclopentadiene on July
17, 1992 (57 FR 31776 (USEPA, 1992)). The NPDWR established a maximum contaminant
level goal (MCLG) and a maximum contaminant level (MCL) of 50 |ig/L. EPA based the MCLG
on a reference dose (RfD) of 7 jig/kg-day (0.007 mg/kg-day) and a cancer classification of D, not
classifiable as to human carcinogenicity.
According to the Agency for Toxic Substances and Disease Registry (ATSDR, 1999a),
hexachlorocyclopentadiene (CsCle^s a manufactured chemical that does not occur naturally. It is
a light, lemon-yellow liquid with a sharp musty odor that easily evaporates into the air. The vapor
looks like a blue haze. Hexachlorocyclopentadiene is also known as HCCPD (ATSDR, 1999a).
Hexachlorocyclopentadiene is released as vapor during manufacture and use.
Hexachlorocyclopentadiene degrades quickly in both soil and in water. It doesn't dissolve readily
in water and will evaporate from the water surface. Hexachlorocyclopentadiene that gets into soil
binds to decaying plant and animal matter and can move through sandy soil to reach ground
water. The half life in soil ranges from one to two weeks, while in water about half of the
hexachlorocyclopentadiene is broken into other chemicals by light in as little as four minutes
(ASTDR, 1999a).
The general population's exposure to hexachlorocyclopentadiene is thought to be
insignificant. Exposure occurs primarily in occupational settings. Individuals in the general
population who live in the vicinity of industrial facilities or hazardous waste sites where
contamination has been detected may be exposed to potentially higher levels of
hexachlorocyclopentadiene. Young children may be exposed if they play in or around
contaminated soil or sediment in the vicinity of production facilities or hazardous waste sites.
(ATSDR, 1999b).
15-1
-------
EPA-OGWDW Contaminant Occurrence Support Document for 2" 6-Year Review Category 1 Contaminants October 2009
15.2 Use, Production, and Releases
Hexachlorocyclopentadiene is used in the manufacture of certain pesticides, including
heptachlor, chlordane, aldrin, dieldrin, and endrin (USEPA, 2000). It is also used to make flame
retardants, resins that do not burn, shock-proof plastics, esters, ketones, fluorocarbons, and dyes.
Most of the hexachlorocyclopentadiene in the environment results from releases during its
production and disposal (ATSDR, 1999a).
The Toxic Substances Control Act (TSCA) Inventory contains confidential data reported
by industry as a partial update of the Chemical Substance Inventory Update System. The data are
reported under the Inventory Update Reporting regulation. Manufacturers and importers are
required to report company information and chemical information, including production volumes
for chemicals they manufactured or imported in excess of 10,000 pounds in the immediately
preceding fiscal year.
The TSCA Chemical Substance Inventory does not provide actual production volumes;
instead, it indicates at which of five production-volume ranges a chemical was manufactured or
imported. The TSCA Inventory is updated every four years. Exhibit 15-1 presents the production
of hexachlorocyclopentadiene in the United States from 1986 to 2002. Note that the production
of hexachlorocyclopentadiene in the United States has consistently been between 10 million and
50 million pounds since 1986.
Exhibit 15-1: Production of Hexachlorocyclopentadiene in the United States
Inventory Update Year
1986 Range
1990 Range
1994 Range
1998 Range
2002 Range
Production Volume
> 10 million - 50 million pounds
> 10 million - 50 million pounds
> 10 million - 50 million pounds
> 10 million - 50 million pounds
> 10 million - 50 million pounds
Source: USEPA, 2003a
Hexachlorocyclopentadiene is listed as a Toxics Release Inventory (TRI) chemical.
Exhibit 15-2a lists the environmental releases of hexachlorocyclopentadiene from 1998 to 2006.
(Exhibit 15-2b provides a graphical presentation of the environmental releases.) Most releases
were to air or off-site. Underground injection and releases to land were only present in 1998.
Surface water discharges were negligible for all years. The TRI data for
hexachlorocyclopentadiene were reported from 11 states, with 3 states (New York, Ohio, and
Tennessee) reporting releases in all 9 years.
15-2
-------
EPA-OGWDW Contaminant Occurrence Support Document for 2" 6-Year Review Category 1 Contaminants October 2009
Exhibit 15-2a: Table of Toxics Release Inventory (TRI) Data for
Hexachlorocyclopentadiene
Year
1998
1999
2000
2001
2002
2003
2004
2005
2006
On-Site Releases (pounds)
Total Air
Emissions
5,801
1,114
1,310
1,229
1,208
1,133
1,265
1,000
786
Surface Water
Discharges
0
2
0
0
5
61
54
0
0
Underground
Injection
250
0
0
0
0
0
0
0
0
Releases to
Land
5,520
0
0
0
0
0
0
0
0
Off-site
Releases
(pounds)
1,067
1,196
2,186
849
560
464
433
430
503
Total On- &
Off -site
Releases
(pounds)
12,638
2,312
3,496
2,078
1,773
1,659
1,752
1,430
1,289
Source: USEPA, 2008
Exhibit 15-2b: Graph of Toxics Release Inventory (TRI) Data for
Hexachlorocyclopentadiene
Environmental Releases of Hexachlorocyclopentadiene
5nnn
•D
*> A nnn
ro
•D
c
3 o nnn
Q.
•
-
• • n n n n n n
n n .I .1 .1 .1 •
1998 1999 2000 2001 2002 2003 2004 2005 2006
Year
• Air Emissions D Surface Water Discharges D Underground Injection D Releases to Land • Off-Site Releases
Source: USEPA, 2008
15.3 Occurrence in Ambient Water
No National Water Quality Assessment (NAWQA) Program data are available from the
United States Geological Survey (USGS) on the occurrence of hexachlorocyclopentadiene in
ambient waters from the Pesticide National Synthesis Project (Gilliom et a/., 2007) or the EPA
Summary Analysis of Cycle 1 data.
15-3
-------
EPA-OGWDW Contaminant Occurrence Support Document for 2" 6-Year Review Category 1 Contaminants October 2009
15.4 Occurrence in Drinking Water
Hexachlorocyclopentadiene is regulated as a synthetic organic chemical (SOC) in
drinking water. All non-purchased community water systems (CWSs) and non-transient non-
community water systems (NTNCWSs) are required to sample for SOCs. The maximum waiver
period for SOCs is three years, but this waiver can be renewed indefinitely. (Systems may receive
a waiver at any time without having collected an SOC sample. With a waiver, systems are not
required to collect an SOC sample.)
All CWSs and NTNCWSs without an SOC waiver must collect four consecutive
quarterly samples during the initial three-year compliance period.1 If all four samples are non-
detections, then a system serving less than 3,300 persons may reduce to one sample during each
consecutive compliance period; a system serving more than 3,300 persons may reduce to two
quarterly samples within a 12-month period during a 3-year compliance period. If a chemical is
detected (but is less than the MCL), the system must take one sample per quarter until results are
below the MCL (minimum of two quarterly samples for ground water systems and four quarterly
samples for surface water systems). If all quarterly samples are below the MCL, the system may
return to annual sampling. If a chemical is detected at a level greater than the MCL, the system
(whether ground water or surface water) must take quarterly samples until four consecutive
quarters are below the MCL. If all quarterly samples are below the MCL, the system may return
to annual sampling.
The analysis of hexachlorocyclopentadiene occurrence presented in the following section
is based on state compliance monitoring data from the Six-Year Review-ICR Dataset. This is the
largest and most comprehensive compliance monitoring data set compiled by EPA's Drinking
Water program to date.
Forty-four of the Six-Year Review-ICR states provided occurrence data for
hexachlorocyclopentadiene. (There were no hexachlorocyclopentadiene data from Oklahoma.)
These data consist of 135,290 analytical results from 32,801 public water systems (PWSs) during
the period from 1998 to 2005. The number of sample results and systems vary by state, although
the state data sets have been reviewed and checked to ensure adequacy of coverage and
completeness. The national modal minimum reporting level (MRL) for
hexachlorocyclopentadiene in the dataset is 0.1 |ig/L.
15.4.1 Stage 1 Analysis of Occurrence Data
Stage 1 analyses have been performed at the sample, sample point, and system levels.
Note that some systems have multiple sample points, such as source water sample points or entry
points to the distribution system. Exhibit 15-3 presents the number of samples, systems, and
sample points in the data set, as well as the population-served by the systems and sample points,
and the percentage of detections and the percentage of the population-served by the systems and
1 All new systems or systems using a new water source that began operation after January 22, 2004 must demonstrate
compliance with the MCL within a period of time specified by the State. The system must also comply with the
initial sampling frequencies specified by the State to ensure that a system can demonstrate compliance with the MCL.
15-4
-------
EPA-OGWDW Contaminant Occurrence Support Document for 2" 6-Year Review Category 1 Contaminants October 2009
sample points with detections. Detections in systems provide a fundamental indication of
national occurrence, but are not representative of MCL violations. For
hexachlorocyclopentadiene, 0.686% of 32,801 systems reported detections. Exhibit 15-4 lists the
minimum, median, 90th percentile, and maximum hexachlorocyclopentadiene concentrations
based on detections from all systems in the Six-Year Review-ICR Dataset. The median
concentration of hexachlorocyclopentadiene, based on detections from all systems, was 0.15
Exhibit 15-3: Hexachlorocyclopentadiene Occurrence Data from the Six-Year
Review-ICR Dataset
Source Water Type
Number of
Samples
Number of
Systems
Number of
Sample Points
Pop. Served
by Systems
Pop. Served
by Sample
Points
Total Number
Ground Water
106,621
29,767
48,670
67,981,852
67,981,852
Surface Water
28,669
3,034
6,771
116,756,578
116,756,578
Combined Ground &
Surface Water
135,290
32,801
55,441
184,738,430
184,738,430
Percent with Detections
Ground Water
0.098%
0.262%
0.173%
5.641%
4.329%
Surface Water
0.949%
4.845%
2.452%
9.882%
6.297%
Combined Ground &
Surface Water
0.278%
0.686%
0.451%
8.321%
5.573%
Exhibit 15-4: Hexachlorocyclopentadiene Summary Statistics
Source Water Type
Ground Water
Detection Value (ug/L)
Minimum
0.005
Median
0.275
90tn
Percentile
10
Maximum
15
Surface Water
0.001
0.13
10
10
Combined Ground &
Surface Water
0.001
0.15
10
15
15-5
-------
EPA-OGWDW Contaminant Occurrence Support Document for 2" 6-Year Review Category 1 Contaminants October 2009
The following maps illustrate the geographic distribution of hexachlorocyclopentadiene
occurrence in drinking water. Exhibit 15-5 shows the states with systems with at least one
detection and Exhibit 15-6 shows the states with systems with at least one detection greater than
the MCL concentration. Note that five states did not submit data for use in the Six-Year Review
(Kansas2, Louisiana, Mississippi, Pennsylvania, and Washington). Oklahoma submitted Six-Year
data for most contaminants; however, there is a statewide waiver for hexachlorocyclopentadiene
in Oklahoma so no hexachlorocyclopentadiene data were available from that state.
The distribution of systems with detections of hexachlorocyclopentadiene is
geographically dispersed and detection rates were generally low. Ten states reported detections in
more than 1% of their systems. In three states, between 0.5% and 1% of systems reported
detections. Fourteen states reported no detections. No states reported detections greater than the
MCL concentration of 50 |ig/L.
Exhibit 15-5: Percentage of Systems with at Least One
Hexachlorocyclopentadiene Detection, by State
A States with Waivers or Not Submitting
States with No Detections
Less than 0.5% of Systems Detecting
Between 0.5% and 1 % of Systams Detecting
More than 1% of Systems Detecting
Exhibit 15-7 presents the Stage 1 analysis of hexachlorocyclopentadiene occurrence in
drinking water from systems and the population-served by those systems in the Six-Year Review-
ICR Dataset relative to several thresholds. The thresholds are: 50 |ig/L, 40 |ig/L, and 25 |ig/L. No
systems reported detections greater than any of the thresholds.
2 Kansas submitted data after the data management and analyses had been conducted; data for Kansas are not
included in the analyses.
15-6
-------
EPA-OGWDW Contaminant Occurrence Support Document for 2" 6-Year Review Category 1 Contaminants October 2009
Exhibit 15-7: Hexachlorocyclopentadiene Stage 1 Analysis - Systems and
Population with at Least One Threshold Exceedance
Source Water
Type
Ground Water
Threshold
> 50 |jg/L
> 40 |jg/L
> 25 |jg/L
Number of
Systems
Exceeding
Threshold
0
0
0
Percent of
Systems
Exceeding
Threshold
0.000%
0.000%
0.000%
Total Population-
Served by
Systems
Exceeding
Threshold
0
0
0
Percent of
Population -Served
by Systems
Exceeding
Threshold
0.000%
0.000%
0.000%
Surface Water
> 50 |jg/L
>
>
0
0
0
0.000%
0.000%
0.000%
0
0
0
0.000%
0.000%
0.000%
Combined
Ground &
Surface Water
> 50 |jg/L
>
>
0
0
0
0.000%
0.000%
0.000%
0
0
0
0.000%
0.000%
0.000%
The occurrence estimates presented above (in Exhibit 15-7) are based on the conservative
assumption that if a hexachlorocyclopentadiene detection is found in a single entry point/sample
point in a system, then the entire population-served by the system is assumed to potentially be
exposed to hexachlorocyclopentadiene (even though other entry points/sampling points at the
system, and the population-served by those entry/sample points, may exhibit no occurrence of
hexachlorocyclopentadiene). The sample point analysis is a less conservative estimate of the
population-served by a system with a detection of hexachlorocyclopentadiene because it
estimates the potentially exposed population-served based only on the proportion of the
population-served by entry points/sample points with detections of hexachlorocyclopentadiene
(rather than the entire population-served by the system).
Exhibit 15-8 presents the Stage 1 analysis of hexachlorocyclopentadiene occurrence in
drinking water from sample points and population-served by those sample points in the Six-Year
Review-ICR Dataset. No sample points reported detections greater than any of the thresholds.
Exhibit 15-8: Hexachlorocyclopentadiene Stage 1 Analysis - Sample Points and
Population with at Least One Threshold Exceedance
Source Water
Type
Ground Water
Threshold
> 50 |jg/L
>
Number of
Sample
Points
Exceeding
Threshold
0
0
Percent of
Sample
Points
Exceeding
Threshold
0.000%
0.000%
Total Population-
Served by Sample
Points Exceeding
Threshold
0
0
Population-
Served by
Sample Points
Exceeding
Threshold
0.000%
0.000%
15-7
-------
EPA-OGWDW Contaminant Occurrence Support Document for 2" 6-Year Review Category 1 Contaminants October 2009
Source Water
Type
Threshold
>
Number of
Sample
Points
Exceeding
Threshold
0
Percent of
Sample
Points
Exceeding
Threshold
0.000%
Total Population-
Served by Sample
Points Exceeding
Threshold
0
Percent of
Population-
Served by
Sample Points
Exceeding
Threshold
0.000%
Surface Water
> 50 |jg/L
>
>
0
0
0
0.000%
0.000%
0.000%
0
0
0
0.000%
0.000%
0.000%
Combined
Ground &
Surface Water
> 50 |jg/L
>
>
0
0
0
0.000%
0.000%
0.000%
0
0
0
0.000%
0.000%
0.000%
15.4.2 Stage 2 Analysis of Occurrence Data
For the Stage 2 analyses, three different substitution values -zero, one-half the MRL
value, and the full MRL value- were used to replace each non-detection record. Three arithmetic
mean hexachlorocyclopentadiene concentrations were calculated at each system using the zero,
one-half MRL and full MRL substitution values. These mean calculations were done for all
systems with data in the Six-Year Review-ICR dataset. Then the percentages of all systems with
a mean concentration greater than each threshold were calculated.
Exhibit 15-9 presents the Stage 2 analysis of estimated system mean concentrations for
hexachlorocyclopentadiene in drinking water relative to several thresholds. The thresholds are:
50 |ig/L, 40 |ig/L, and 25 |ig/L. Exhibit 15-10 presents similar information based on population-
served by the systems. Based on the Stage 2 analyses, no systems had an estimated system mean
greater than or equal to any of the thresholds.
Exhibit 15-9: Hexachlorocyclopentadiene Stage 2 Analysis - Systems with a
System Mean Threshold Exceedance
Source
Water Type
Ground
Water
Threshold
> 50 |jg/L
>
>
Number of Systems with Mean
Exceeding Threshold
Full MRL
0
0
0
1/2 MRL
0
0
0
Zero
0
0
0
Percent of Systems with Mean
Exceeding Threshold
Full MRL
0.000%
0.000%
0.000%
1/2 MRL
0.000%
0.000%
0.000%
Zero
0.000%
0.000%
0.000%
Surface
Water
> 50 |jg/L
>
>
0
0
0
0
0
0
0
0
0
0.000%
0.000%
0.000%
0.000%
0.000%
0.000%
0.000%
0.000%
0.000%
15-8
-------
EPA-OGWDW Contaminant Occurrence Support Document for 2" 6-Year Review Category 1 Contaminants October 2009
Source
Water Type
Combined
Ground &
Surface
Water
Threshold
> 50 |jg/L
>
>
Number of Systems with Mean
Exceeding Threshold
Full MRL
0
0
0
1/2 MRL
0
0
0
Zero
0
0
0
Percent of Systems with Mean
Exceeding Threshold
Full MRL
0.000%
0.000%
0.000%
1/2 MRL
0.000%
0.000%
0.000%
Zero
0.000%
0.000%
0.000%
1. The results shown in this table are based on system arithmetic means that were calculated using all sample
detection records and all non-detection records, with the non-detection records replaced (substituted) with
concentration values equal to the full MRL, 14 MRL, and zero.
Exhibit 15-10: Hexachlorocyclopentadiene Stage 2 Analysis - Population-Served
by Systems with a System Mean Threshold Exceedance
Source
Water Type
Ground
Water
Threshold
> 50 |jg/L
>
>
Population-Served by Systems
with Mean Exceeding Threshold
Full MRL
0
0
0
1/2 MRL
0
0
0
Zero
0
0
0
Percent of Population -Served by
Systems with Mean Exceeding
Threshold
Full MRL
0.000%
0.000%
0.000%
1/2 MRL
0.000%
0.000%
0.000%
Zero
0.000%
0.000%
0.000%
Surface
Water
> 50 |jg/L
>
>
0
0
0
0
0
0
0
0
0
0.000%
0.000%
0.000%
0.000%
0.000%
0.000%
0.000%
0.000%
0.000%
Combined
Ground &
Surface
Water
> 50 |jg/L
>
>
0
0
0
0
0
0
0
0
0
0.000%
0.000%
0.000%
0.000%
0.000%
0.000%
0.000%
0.000%
0.000%
1. The results shown in this table are based on system arithmetic means that were calculated using all sample
detection records and all non-detection records, with the non-detection records replaced (substituted) with
concentration values equal to the full MRL, 14 MRL, and zero.
15.5 Additional Occurrence Data
First Six-Year Review ofNPDWRs
Similar types of Stage 1 contaminant occurrence analyses were conducted for the first
Six-Year Review. The first Six-Year Review assessed regulated contaminant occurrence data
collected from 1993 to 1997. The first Six-Year Review data set consisted of data from 16 states
that were assembled into a "national cross-section" that was qualitatively indicative (though not
statistically representative) of national occurrence. For the first Six-Year Review (USEPA,
2003b), 124 systems (approximately 0.89% of about 14,000 systems), reported at least one
detection of hexachlorocyclopentadiene. No systems reported detections greater than the MCL
concentration.
15-9
-------
EPA-OGWDW Contaminant Occurrence Support Document for 2" 6-Year Review Category 1 Contaminants October 2009
15.6 Summary of Data
Hexachlorocyclopentadiene is listed as a Toxics Release Inventory (TRI) chemical. TRI
data indicate that total off-site and on-site releases of hexachlorocyclopentadiene for the period
1998-2006 ranged from about 1,300 pounds to approximately 13,000 pounds, with a decrease
over time. In most years, on-site releases were greater than off-site releases.
Ambient water quality data from the NAWQA program were not available for
hexachl orocy cl opentadi ene.
The Stage 1 analysis of occurrence in drinking water indicates that
hexachlorocyclopentadiene was detected in 0.686% of 32,801 surface and ground water systems.
The median concentration, based on detections from all sites, was 0.15 |ig/L. No systems
reported detections greater than the MCL concentration of 50 |ig/L.
In the Stage 2 analysis, no systems had an estimated system mean greater than the MCL
concentration.
15.7 References
Agency for Toxic Substances and Disease Registry (ATSDR). 1999a. ToxFAQsfor
Hexachlorocyclopentadiene (HCCPD). June 1999. Available on the Internet at:
http://www.atsdr.cdc.gov/tfactsll2.html.
ATSDR. 1999b. Toxicological Profile for Hexachlorocyclopentadiene (HCCPD). July 1999.
Available on the Internet at: http://www.atsdr.cdc.gov/toxprofiles/tpll2.html.
Gilliom, R.J., J.E. Barbash, C.G. Crawford, P.A. Hamilton, J.D. Martin, N. Nakagaki, L.H.
Nowell, J.C. Scott, P.E. Stackelberg, G.P. Thelin, and D.M. Wolock. 2007. The Quality of Our
Nation's Waters - Pesticides in the Nation's Streams and Ground Water, 1992-2001. Appendix
7. Statistical summaries of water-quality data. U.S. Geological Survey Circular 1291, 172 p.
Available on the Internet at: http://water.usgs.gov/nawqa/pnsp/pubs/circ 129l/appendix7/.
United States Environmental Protection Agency (USEPA). 1992. Drinking Water; National
Primary Drinking Water Regulations—Synthetic Organic Chemicals and Inorganic Chemicals;
National Primary Drinking Water Regulations Implementation; Final Rule. Federal Register.
Vol. 57, No. 138. p. 31776, July 17, 1992.
USEPA. 2000. Hexachlorocyclopentadiene. 77-47-4. Technology Transfer Network Air Toxics
Web Site. Last modified November 6, 2007. Available on the Internet at:
http://www.epa.gov/ttn/atw/hlthef/hexa-die.html. Accessed March 7, 2008.
USEPA. 2003a. Non-confidential Inventory Update Reporting (IUR) Production Volume
Information -Hexachlorocyclopentadiene. Last modified September 10, 2007. Available on the
Internet at: http://www.epa.gov/oppt/iur/tools/data/2002-vol.htm. [Search for
hexachlorocyclopentadiene.] Accessed July 8, 2008.
15-10
-------
EPA-OGWDW Contaminant Occurrence Support Document for 2" 6-Year Review Category 1 Contaminants October 2009
USEPA. 2003b. Occurrence Estimation Methodology and Occurrence Findings for Six-Year
Review of National Primary Drinking Water Regulations. Office of Water. EPA-815-R-03-006.
June 2003.
USEPA. 2008. TRI Explorer: Trends - Hexachlorocyclopentadiene. Released February 21, 2008.
Available on the Internet at: http://www.epa.gov/triexplorer/trends.htm. [Search for
hexachlorocyclopentadiene.] Accessed June 16, 2008.
USEPA. 2009. The Analysis of Regulated Contaminant Occurrence Data from Public Water
Systems in Support of the Second Six-Year Review of National Primary Drinking Water
Regulations. Office of Water. EPA-815-B-09-006. October 2009.
15-11
-------
EPA-OGWDW Contaminant Occurrence Support Document for 2" 6-Year Review Category 1 Contaminants October 2009
16 Oxamyl (Vydate)
This chapter on oxamyl (vydate) is part of a report that is organized so that each chapter
presents information on one specific contaminant. Chapter 1 of this report, Introduction, presents
background information on all sources of data used as well as the analytical approach used to
estimate contaminant occurrence in drinking water. This contaminant chapter includes
background information (such as regulatory history, general chemical information, and
environmental fate and behavior), use and environmental release information, and oxamyl
occurrence estimates in ambient water and drinking water. All drinking water occurrence
estimates were based on data from the National Compliance Monitoring Information Collection
Request (ICR) Dataset for the Second Six-Year Review (the "Six Year Review-ICR Dataset").
For detailed information regarding this dataset, data management, data quality assessments, and
the analytical approach used to generate all contaminant occurrence estimates presented in this
chapter, please refer to USEPA (2009).
16.1 Background
The United States Environmental Protection Agency (EPA) published the current
National Primary Drinking Water Regulations (NPDWR) for oxamyl on July 17, 1992 (57 FR
31776 (USEPA, 1992)). The NPDWR established a maximum contaminant level goal (MCLG)
and a maximum contaminant level (MCL) of 200 |ig/L. EPA based the MCLG on a reference
dose (RfD) of 25 jig/kg-day (0.025 mg/kg-day) and a cancer classification of E, evidence of non-
carcinogenicity for humans.
According to the Extension Toxicology Network (EXTOXNET, 1996), oxamyl (also
known as Vydate) is a colorless or white crystalline solid with a garlic-like odor. It is a restricted
use pesticide that is available in both liquid and granular form; however, the granular form is
banned in the United States (EXTOXNET, 1996).
Oxamyl (CyH^NsOsS) is released directly to the environment through its use as an
insecticide and during its manufacture, handling, and storage. Oxamyl is rapidly broken down in
water, with a half-life of one to two days. Degradation is also rapid in soils by aerobic and
anaerobic bacteria. As a result, it is more likely to be found in groundwater than in surface water.
Oxamyl is excreted rapidly and does not accumulate in mammalian tissue (EXTOXNET, 1996).
16.2 Use, Production, and Releases
Oxamyl is widely used for control of insects, mites and nematodes on field crops, fruits
and ornamentals. There are no residential uses of oxamyl (USEPA, 2000).
Exhibit 16-1, compiled by the United States Geological Survey (USGS, 2007), shows the
geographic distribution of estimated average annual oxamyl use in the United States from 1999
through 2004. A breakdown of use by crop is also included. The map was created by the USGS
using state-level data sets on pesticide use rates from 1999-2004 compiled by the CropLife
Foundation, at the Crop Protection Research Institute, combined with county-level data on
16-1
-------
EPA-OGWDW Contaminant Occurrence Support Document for 2" 6-Year Review Category 1 Contaminants October 2009
harvested crop acreage obtained from the 2002 Census of Agriculture. Due to the nature of the
data sources, non-agricultural uses are not reflected here, and variations in use at the county-level
are also not well represented (Thelin and Gianessi, 2000). The USGS (2007) estimates that more
than 700,000 pounds of oxamyl active ingredient were used on average each year between 1999
and 2004. The greatest use of oxamyl is in the Pacific Northwest, California, and the South.
Exhibit 16-1: Estimated Annual Agricultural Use for Oxamyl (c. 2002)
OXAMYL - Insecticide
2002 esti "latsd annual a^kaJtunil use
Average annual usewl
actwe *igred ait
(pounds par square nite of ogrteuRuroJ
bnQ h courty)
H no estimated use
J OJ301 O 0.003
H 0.004 ti 0.016
H ox>i7 to o.oee
LJ 0,069 to 0 645
• >= 0.648
Oro«»
cotton
swdriot*
ri.llV. :il
drycnions
Umetoei
cftwFut
ealwy
*Pffc*
WOtf.
bolpoopore
•nut
sounds awlW
350964
W»11
•1S651
3B/W3
19751
1WS
13137
9029
em
6140
rWCCrt
latonal J5e
49.23
zeisi
MB
s.«
S.77
1JD
IJt
1^7
ua
0.72
Source: USGS, 2007
The Chemical Substance Inventory Update System contains confidential data reported by
industry as a partial update of the Toxic Substances Control Act (TSCA) Inventory. The TSCA
Chemical Substance Inventory does not report production data for oxamyl.
Oxamyl is not listed as a Toxics Release Inventory (TRI) chemical; thus no TRI release
records are available. Therefore, information on the use of oxamyl (described in the previous
section) may provide the primary indication of where releases are most likely to occur.
16.3 Occurrence in Ambient Water
Lakes, rivers, and aquifers are the sources of most drinking water. The USGS National
Water Quality Assessment (NAWQA) program is a national-scale assessment of the occurrence
of contaminants in ambient surface and ground water. Contaminant occurrence in ambient water
provides information on the potential for contaminants to adversely affect drinking water
16-2
-------
EPA-OGWDW Contaminant Occurrence Support Document for 2" 6-Year Review Category 1 Contaminants October 2009
supplies. Data on the occurrence of oxamyl in ambient water are available from the NAWQA
program for both the Pesticide National Synthesis Project and a summary analysis of all Cycle 1
water monitoring data.
16.3.1 NA WQA Pesticide National Synthesis Project
The NAWQA Pesticide National Synthesis Project is a national-scale assessment of the
occurrence and behavior of pesticides in streams and ground water of the United States and the
potential for pesticides to adversely affect drinking water supplies or aquatic ecosystems. Under
the National Synthesis Project, the USGS monitored oxamyl between 1992 and 2001 in streams
and wells across the country (Gilliom etal., 2007). The maximum long-term method detection
limit for oxamyl was 0.080 |ig/L.
In NAWQA stream samples (Exhibit 16-2), oxamyl was only detected at agricultural sites
(at a frequency of 0.57%). The highest concentration at an agricultural site was 0.160 |ig/L.
Exhibit 16-2: USGS National Synthesis Summary of NAWQA Monitoring of Oxamyl
in Streams, 1992-2001
Land Use
Type
Agricultural
Mixed
Undeveloped
Urban
No. of Samples
(No. of Sites)
1,469(62)
800 (36)
100(5)
524(19)
Detection
Frequency
0.57%
0.00%
0.00%
0.00%
50tn Percent! le
(Median)
Concentration
(in ug/L)
ND1
ND1
ND1
ND1
95th Percentile
Concentration
(in ug/L)
ND1
ND1
ND1
ND1
Maximum
Concentration
(in ug/L)
0.160
ND1
ND1
ND1
Source: Gilliom et a/., 2007
1. ND = not detected (Concentration is less than the maximum long term method detection limit and is expected to
be less than any higher percentile concentration shown in the table)
In ground water NAWQA samples (Exhibit 16-3), oxamyl was found at frequencies
ranging from 0% of samples in undeveloped areas to 0.06% in mixed land use settings, 0.15% in
urban areas, and 0.85% in agricultural settings. The 95th percentile concentrations were less than
the method detection limit in all settings. The highest concentration, 2.06 |ig/L, was found at an
agricultural site.
Exhibit 16-3: USGS National Synthesis Summary of NAWQA Monitoring of Oxamyl
in Ground Water, 1992-2001
Land Use Type
Agricultural
Mixed
Number
of Wells
1,174
1,542
Detection
Frequency
0.85%
0.06%
50tn Percentile
(Median)
Concentration
(in ug/L)
ND1
ND1
95th Percentile
Concentration
(in ug/L)
ND1
ND1
Maximum
Concentration
(in ug/L)
2.06
0.03
16-3
-------
EPA-OGWDW Contaminant Occurrence Support Document for 2" 6-Year Review Category 1 Contaminants October 2009
Land Use Type
Undeveloped
Urban
Number
of Wells
23
653
Detection
Frequency
0%
0.15%
50tn Percentile
(Median)
Concentration
(in ug/L)
ND1
ND1
95th Percentile
Concentration
(in ug/L)
ND1
ND1
Maximum
Concentration
(in ug/L)
ND1
0.26
Source: Gilliom et a/., 2007
1. ND = not detected (Concentration is less than the maximum long term method detection limit and is expected to
be less than any higher percentile concentration shown in the table)
16.3.2 EPA Summary Analysis ofNAWQA Data
EPA, with the cooperation of the USGS, performed a summary analysis of all Cycle 1
(1992-2001) water monitoring data from all study units for oxamyl. Detection frequencies were
computed as the percentage of samples and sites with detections (i.e., with at least one result
equal to or greater than the reporting limit). Sample detections can be biased by frequent
sampling in areas with high (or low) occurrence. Calculating the percentage of sites with
detections can reduce this bias.
The results of the EPA analysis are presented in Exhibit 16-4. Overall, oxamyl was
detected in 0.3% of samples and at 0.3% of sites. Oxamyl was detected at more ground water
sites than surface water sites. The median concentration, based on detections from all sites, was
0.155 |ig/L. The 99th percentile concentration, based on detections from all sites, was 2.06 |ig/L.
Exhibit 16-4: EPA Summary Analysis of Oxamyl Data from NAWQA Study Units,
1992-2001
ground
water
surface
water
all
sites
Detection Frequency
(detections are results > RL1)
Number
of
Samples
3,985
5,522
9,507
% Samples
with
Detections
0.3%
0.2%
0.3%
Number
of Sites
3,609
894
4,503
% Sites
with
Detections
0.3%
0.2%
0.3%
Concentration Values
(of detections, in ug/L)
Minimum
0.02
0.01
0.01
Median
1.245
0.11
0.155
95th
Percen-
tile
2.06
0.16
1.55
99th
Percen-
tile
2.06
0.16
2.06
Maximum
2.06
0.16
2.06
1. RLs (Reporting Limits) for oxamyl varied, but did not exceed 0.012 pg/L Note that because this EPA analysis
involves more data points than the USGS analyses presented above, a direct comparison is not possible.
16.4 Occurrence in Drinking Water
Oxamyl is regulated as a synthetic organic chemical (SOC) in drinking water. All non-
purchased community water systems (CWSs) and non-transient non-community water systems
(NTNCWSs) are required to sample for SOCs. The maximum waiver period for SOCs is three
years, but this waiver can be renewed indefinitely. (Systems may receive a waiver at any time
16-4
-------
EPA-OGWDW Contaminant Occurrence Support Document for 2" 6-Year Review Category 1 Contaminants October 2009
without having collected an SOC sample. With a waiver, systems are not required to collect an
SOC sample.)
All CWSs and NTNCWSs without an SOC waiver must collect four consecutive
quarterly samples during the initial three-year compliance period.1 If all four samples are non-
detections, then a system serving less than 3,300 persons may reduce to one sample during each
consecutive compliance period; a system serving more than 3,300 persons may reduce to two
quarterly samples within a 12-month period during a 3-year compliance period. If a chemical is
detected (but is less than the MCL), the system must take one sample per quarter until results are
below the MCL (minimum of two quarterly samples for ground water systems and four quarterly
samples for surface water systems). If all quarterly samples are below the MCL, the system may
return to annual sampling. If a chemical is detected at a level greater than the MCL, the system
(whether ground water or surface water) must take quarterly samples until four consecutive
quarters are below the MCL. If all quarterly samples are below the MCL, the system may return
to annual sampling.
The analysis of oxamyl occurrence presented in the following section is based on state
compliance monitoring data from the Six-Year Review-ICR Dataset. This is the largest and most
comprehensive compliance monitoring data set compiled by EPA's Drinking Water program to
date.
All of the Six-Year Review-ICR states provided occurrence data for oxamyl. These data
consist of 113,380 analytical results from 30,876 public water systems (PWSs) during the period
from 1998 to 2005. The number of sample results and systems vary by state, although the state
data sets have been reviewed and checked to ensure adequacy of coverage and completeness. The
national modal minimum reporting level (MRL) for oxamyl in the dataset is 2 |ig/L.
16.4.1 Stage 1 Analysis of Occurrence Data
Stage 1 analyses have been performed at the sample, sample point, and system levels. Note that
some systems have multiple sample points, such as source water sample points or entry points to
the distribution system. Exhibit 16-5 presents the number of samples, systems, and sample points
in the data set, as well as the population-served by the systems and sample points, and the
percentage of detections and the percentage of the population-served by the systems and sample
points with detections. Detections in systems provide a fundamental indication of national
occurrence, but are not representative of MCL violations. For oxamyl, 0.230% of 30,876 systems
reported detections. Exhibit 16-6 lists the minimum, median, 90th percentile, and maximum
oxamyl concentrations based on detections from all systems in the Six-Year Review-ICR
Dataset. The median concentration of oxamyl, based on detections from all systems, was 1 ng/L.
1 All new systems or systems using a new water source that began operation after January 22, 2004 must demonstrate
compliance with the MCL within a period of time specified by the State. The system must also comply with the
initial sampling frequencies specified by the State to ensure that a system can demonstrate compliance with the MCL.
16-5
-------
EPA-OGWDW Contaminant Occurrence Support Document for 2" 6-Year Review Category 1 Contaminants October 2009
Exhibit 16-5: Oxamyl Occurrence Data from the Six-Year Review-ICR Dataset
Source Water Type
Number of
Samples
Number of
Systems
Number of
Sample Points
Pop. Served
by Systems
Pop. Served
by Sample
Points
Total Number
Ground Water
93,766
28,100
45,236
61,701,854
61,701,854
Surface Water
19,614
2,776
5,989
105,676,577
105,676,577
Combined Ground &
Surface Water
113,380
30,876
51,225
167,378,431
167,378,431
Percent with Detections
Ground Water
0.075%
0.203%
0.137%
6.233%
5.100%
Surface Water
0.092%
0.504%
0.267%
2.303%
1.114%
Combined Ground &
Surface Water
0.078%
0.230%
0.152%
3.752%
2.583%
Exhibit 16-6: Oxamyl Summary Statistics
Source Water Type
Ground Water
Detection Value (ug/L)
Minimum
0.18
Median
1
90tn
Percentile
5
Maximum
5.01
Surface Water
0.5
2
40
531.1
Combined Ground &
Surface Water
0.18
1
5
531.1
The following maps illustrate the geographic distribution of oxamyl occurrences in
drinking water. Exhibit 16-7 shows the states with systems with at least one detection and
Exhibit 16-8 shows the states with systems with at least one detection greater than the MCL
concentration. Note that five states did not submit data for use in the Six-Year Review (Kansas2,
Louisiana, Mississippi, Pennsylvania, and Washington).
The distribution of systems with detections of oxamyl is geographically dispersed.
Detection rates were generally low, with only 3 states reporting detections in more than 2% of
2 Kansas submitted data after the data management and analyses had been conducted; data from Kansas are not
included in the analyses.
16-6
-------
EPA-OGWDW Contaminant Occurrence Support Document for 2" 6-Year Review Category 1 Contaminants October 2009
their systems. Only one state reported detections greater than the MCL concentration of 200
Exhibit 16-7: Percentage of Systems with at Least One Oxamyl Detection, by
State
States Not Submitting Any Data
States with No Detections
Loss than 1% of Systems Detecting
| Between 1% and 2% of Systems Detecting
I More than 2% of Systems Detecting
Exhibit 16-8: Percentage of Systems with at Least One Oxamyl Detection Greater
than the MCL Concentration (> 200 \\glL), by State
States Not Submitting Any Data
States with No Detection* > MCL
Less than D.I % Of Systems with Detections > MCL
| Between 0.1% and 0.2% of Systems with Detections > MCL
I More liian 0.2% of Systems with Detection s > MCL
16-7
-------
EPA-OGWDW Contaminant Occurrence Support Document for 2" 6-Year Review Category 1 Contaminants October 2009
Exhibit 16-9 presents the Stage 1 analysis of oxamyl occurrences in drinking water from
systems and the population-served by those systems in the Six-Year Review-ICR Dataset relative
to several thresholds. The thresholds are: 200 |ig/L, 100 |ig/L, and 2 |ig/L. One surface water
system (0.003% of all systems), serving 200 persons, reported at least one detection greater than
or equal to 100 |ig/L and greater than the MCL concentration of 200 |ig/L. No ground water
systems reported detections greater than 100 |ig/L or 200 |ig/L.
Exhibit 16-9: Oxamyl Stage 1 Analysis - Systems and Population with at Least
One Threshold Exceedance
Source
Water
Type
Ground
Water
Threshold
> 200 |jg/L
>1
> 2 Mg/L
Number of
Systems
Exceeding
Threshold
0
0
14
Percent of
Systems
Exceeding
Threshold
0.000%
0.000%
0.050%
Total Population-
Served by
Systems
Exceeding
Threshold
0
0
38,142
Percent of
Population -Served
by Systems
Exceeding
Threshold
0.000%
0.000%
0.062%
Surface
Water
> 200 |jg/L
>1
> 2 |jg/L
1
1
4
0.036%
0.036%
0.144%
200
200
258,348
0.0002%
0.0002%
0.244%
Combined
Ground &
Surface
Water
> 200 |jg/L
>1
> 2 |jg/L
1
1
18
0.003%
0.003%
0.058%
200
200
296,490
0.0001%
0.0001%
0.177%
The occurrence estimates presented above (in Exhibit 16-9) are based on the conservative
assumption that if an oxamyl detection is found in a single entry point/sample point in a system,
then the entire population-served by the system is assumed to potentially be exposed to oxamyl
(even though other entry points/sampling points at the system, and the population-served by
those entry/sample points, may exhibit no occurrence of oxamyl). The sample point analysis is a
less conservative estimate of the population-served by a system with a detection of oxamyl
because it estimates the potentially exposed population-served based only on the proportion of
the population-served by entry points/sample points with detections of oxamyl (rather than the
entire population-served by the system).
Exhibit 16-10 presents the Stage 1 analysis of oxamyl occurrences in drinking water from
sample points and population-served by those sample points in the Six-Year Review-ICR
Dataset. One (0.002%) sample point, serving an estimated 29 persons, had at least one detection
greater than or equal to 100 |ig/L and greater than the MCL concentration of 200 |ig/L.
16-8
-------
EPA-OGWDW Contaminant Occurrence Support Document for 2" 6-Year Review Category 1 Contaminants October 2009
Exhibit 16-10: Oxamyl Stage 1 Analysis - Sample Points and Population with at
Least One Threshold Exceedance
Source
Water
Type
Ground
Water
Threshold
> 200 |jg/L
>1
> 2 |jg/L
Number of
Sample Points
Exceeding
Threshold
0
0
16
Percent of
Sample Points
Exceeding
Threshold
0.000%
0.000%
0.035%
Total Population -
Served by Sample
Points Exceeding
Threshold
0
0
35,868
Percent of
Population-
Served by
Sample Points
Exceeding
Threshold
0.000%
0.000%
0.058%
Surface
Water
> 200 |jg/L
>1
> 2 |jg/L
1
1
5
0.017%
0.017%
0.083%
29
29
173,356
0.00003%
0.00003%
0.164%
Combined
Ground &
Surface
Water
> 200 |jg/L
> 1 |jg/L
> 2 |jg/L
1
1
21
0.002%
0.002%
0.041%
29
29
209,223
0.00002%
0.00002%
0.125%
16.4.2 Stage 2 Analysis of Occurrence Data
For the Stage 2 analyses, three different substitution values -zero, one-half the MRL
value, and the full MRL value- were used to replace each non-detection record. Three arithmetic
mean oxamyl concentrations were calculated at each system using the zero, one-half MRL and
full MRL substitution values. These mean calculations were done for all systems with data in the
Six-Year Review-ICR dataset. Then the percentages of all systems with a mean concentration
greater than each threshold were calculated.
Exhibit 16-11 presents the Stage 2 analysis of estimated system mean concentrations for
oxamyl occurrences in drinking water relative to several thresholds. The thresholds are: 200
|ig/L, 100 |ig/L, and 2 |ig/L. Exhibit 16-12 presents similar information based on population-
served by the systems. Based on the Stage 2 analyses, no systems had an estimated system mean
greater than or equal to 100 |ig/L and greater than the MCL concentration of 200 |ig/L.
Exhibit 16-11: Oxamyl Stage 2 Analysis - Systems with a System Mean Threshold
Exceedance
Source
Water Type
Ground
Water
Threshold
> 200 |jg/L
>1
> 2 |jg/L
Number of Systems with Mean
Exceeding Threshold1
Full MRL
0
0
14
1/2 MRL
0
0
11
Zero
0
0
6
Percent of Systems with Mean
Exceeding Threshold1
Full MRL
0.000%
0.000%
0.050%
1/2 MRL
0.000%
0.000%
0.039%
Zero
0.000%
0.000%
0.021%
16-9
-------
EPA-OGWDW Contaminant Occurrence Support Document for 2" 6-Year Review Category 1 Contaminants October 2009
Source
Water Type
Surface
Water
Threshold
> 200 |jg/L
>1
> 2 Mg/L
Number of Systems with Mean
Exceeding Threshold1
Full MRL
0
0
3
1/2 MRL
0
0
2
Zero
0
0
2
Percent of Systems with Mean
Exceeding Threshold1
Full MRL
0.000%
0.000%
0.108%
1/2 MRL
0.000%
0.000%
0.072%
Zero
0.000%
0.000%
0.072%
Combined
Ground &
Surface
Water
> 200 |jg/L
>1
> 2 |jg/L
0
0
17
0
0
13
0
0
8
0.000%
0.000%
0.055%
0.000%
0.000%
0.042%
0.000%
0.000%
0.026%
1. The results shown in this table are based on system arithmetic means that were calculated using all sample
detection records and all non-detection records, with the non-detection records replaced (substituted) with
concentration values equal to the full MRL, 1A MRL, and zero.
Exhibit 16-12: Oxamyl Stage 2 Analysis - Population-Served by Systems with a
System Mean Threshold Exceedance
Source
Water Type
Ground
Water
Threshold
> 200 |jg/L
> 1 |jg/L
> 2 |jg/L
Population-Served by Systems with
Mean Exceeding Threshold1
Full MRL
0
0
38,142
1/2 MRL
0
0
37,979
Zero
0
0
3,669
Percent of Population -Served by
Systems with Mean Exceeding
Threshold1
Full MRL
0.000%
0.000%
0.062%
1/2 MRL
0.000%
0.000%
0.062%
Zero
0.000%
0.000%
0.006%
Surface
Water
> 200 |jg/L
>1
> 2 |jg/L
0
0
257,848
0
0
3,386
0
0
3,386
0.000%
0.000%
0.244%
0.000%
0.000%
0.003%
0.000%
0.000%
0.003%
Combined
Ground &
Surface
Water
> 200 |jg/L
>1
> 2 |jg/L
0
0
295,990
0
0
41,365
0
0
7,055
0.000%
0.000%
0.177%
0.000%
0.000%
0.025%
0.000%
0.000%
0.004%
1. The results shown in this table are based on system arithmetic means that were calculated using all sample
detection records and all non-detection records, with the non-detection records replaced (substituted) with
concentration values equal to the full MRL, 14 MRL, and zero.
16.5 Additional Occurrence Data
USGS 1999 Pilot Monitoring Program
Oxamyl was an analyte in the 1999 Pilot Monitoring Program (Blomquist et a/., 2001);
however, it was not detected in any raw or finished water samples.
16-10
-------
EPA-OGWDW Contaminant Occurrence Support Document for 2" 6-Year Review Category 1 Contaminants October 2009
First Six-Year Review ofNPDWRs
Similar types of Stage 1 contaminant occurrence analyses were conducted for the first
Six-Year Review. The first Six-Year Review assessed regulated contaminant occurrence data
collected from 1993 to 1997. The first Six-Year Review data set consisted of data from 16 states
that were assembled into a "national cross-section" that was qualitatively indicative (though not
statistically representative) of national occurrence. For the first Six-Year Review (USEPA,
2003), 10 systems (approximately 0.08% of about 13,000 systems), reported at least one
detection of oxamyl. No systems reported a detection greater than the MCL concentration.
16.6 Summary of Data
Oxamyl is not listed as a TRI chemical; no release records are available.
Ambient water data from the NAWQA Pesticide National Synthesis Project show that
most oxamyl detections occurred at agricultural sites. It was detected in 0.57% of 1,469
agricultural stream samples taken at 62 sites and was detected in 0.85% of samples taken from
1,174 ground water agricultural wells. Additional ambient water data from the NAWQA program
indicate that oxamyl was detected in 0.3% of 9,507 ground and surface water samples and at
0.3% of 4,503 sites. The median concentration, based on detections from all sites, was 0.16 |ig/L.
The Stage 1 analysis of occurrences in drinking water indicates that oxamyl was detected
in 0.230% of 30,876 surface water and ground water systems. The median concentration, based
on detections from all systems, was 1 ng/L. One surface water system (0.003% of all systems),
serving 200 persons, reported at least one detection greater than the MCL concentration of 200
Hg/L. No ground water systems reported detections greater than the MCL concentration.
In the Stage 2 analysis, no systems had an estimated system mean greater than or equal to
100 |ig/L and greater than the MCL concentration of 200 |ig/L.
16.7 References
Blomquist, J.D., J.M. Denis, J.L. Cowles, J.A. Hetrick, R.D. Jones, andN.B. Birchfield. 2001.
Pesticides in Selected Water-Supply Reservoirs and Finished Drinking Water, 1999-2000:
Summary of Results from a Pilot Monitoring Program. U.S. Geological Survey Water-Resources
Investigations Report 01-456, 65 p. Available on the Internet at:
http://md.water.usgs. gov/nawqa/OFR_0 l-456.pdf.
Extension Toxicology Network (EXTOXNET). 1996. EXTOXNET Pesticide Information
Profiles-Oxamyl. Last modified June 1996. Available on the Internet at:
http://extoxnet.orst.edu/pips/oxamyl.htm. Accessed February 18, 2008.
16-11
-------
EPA-OGWDW Contaminant Occurrence Support Document for 2" 6-Year Review Category 1 Contaminants October 2009
Gilliom, R.J., I.E. Barbash, C.G. Crawford, P.A. Hamilton, J.D. Martin, N. Nakagaki, L.H.
Nowell, J.C. Scott, P.E. Stackelberg, G.P. Thelin, and D.M. Wolock. 2007. The Quality of Our
Nation's Waters - Pesticides in the Nation's Streams and Ground Water, 1992-2001. Appendix
7. Statistical summaries of water-quality data. U.S. Geological Survey Circular 1291, 172 p.
Available on the Internet at: http://water.usgs.gov/nawqa/pnsp/pubs/circ 129l/appendix7/.
Thelin, G.P. and L.P. Gianessi. 2000. Method for Estimating Pesticide Use for County Areas of
the Conterminous United States. U.S. Geological Survey Open-File Report 00-250, 62 p.
Available on the Internet at: http://ca.water.usgs.gov/pnsp/rep/ofr00250/ofr00250.pdf.
United States Environmental Protection Agency (USEPA). 1992. Drinking Water; National
Primary Drinking Water Regulations—Synthetic Organic Chemicals and Inorganic Chemicals;
National Primary Drinking Water Regulations Implementation; Final Rule. Federal Register.
Vol. 57, No. 138. p. 31776, July 17, 1992.
USEPA. 2000.1.R.E.D Factsheet: OxamylFacts. EPA-738-F-00-013. November 2000.
Available on the Internet at:
http://www.epa.gov/pesticides/reregistration/REDs/factsheets/0253iredfact.pdf
USEPA. 2003. Occurrence Estimation Methodology and Occurrence Findings for Six-Year
Review of National Primary Drinking Water Regulations. Office of Water. EPA-815-R-03-006.
June 2003.
USEPA. 2009. The Analysis of Regulated Contaminant Occurrence Data from Public Water
Systems in Support of the Second Six-Year Review of National Primary Drinking Water
Regulations. Office of Water. EPA-815-B-09-006. October 2009.
United States Geological Survey (USGS). 2007. 2002 Pesticide Use Maps. Available on the
Internet at: http://water.usgs.gov/nawqa/pnsp/usage/maps/compound_listing.php?year=02.
Accessed December 5, 2007.
16-12
-------
EPA-OGWDW Contaminant Occurrence Support Document for 2" 6-Year Review Category 1 Contaminants October 2009
17 Tetrachloroethylene (PCE)
This chapter on tetrachloroethylene (PCE) is part of a report that is organized so that each
chapter presents information on one specific contaminant. Chapter 1 of this report, Introduction,
presents background information on all sources of data used as well as the analytical approach
used to estimate contaminant occurrence in drinking water. This contaminant chapter includes
background information (such as regulatory history, general chemical information, and
environmental fate and behavior), use and environmental release information, and
tetrachloroethylene occurrence estimates in ambient water and drinking water. All drinking water
occurrence estimates were based on data from the National Compliance Monitoring Information
Collection Request (ICR) Dataset for the Second Six-Year Review (the "Six Year Review-ICR
Dataset"). For detailed information regarding this dataset, data management, data quality
assessments, and the analytical approach used to generate all contaminant occurrence estimates
presented in this chapter, please refer to USEPA (2009).
17.1 Background
The United States Environmental Protection Agency (EPA) published the current
National Primary Drinking Water Regulations (NPDWR) for tetrachloroethylene on January 30,
1991 (56 FR 3526 (USEPA, 1991)). The NPDWR established a maximum contaminant level
goal (MCLG) of zero based on a cancer classification of B2, probable human carcinogen. The
NPDWR also established a maximum contaminant level (MCL) of 5 |ig/L based on analytical
feasibility.
According to the Agency for Toxic Substances and Disease Registry (ATSDR, 1997),
tetrachloroethylene ^CU) is a manufactured organic compound that has a sharp, sweet odor. It
is a nonflammable liquid at room temperature and evaporates easily into the air. Other names for
tetrachloroethylene include perchloroethylene, PCE, and tetrachloroethene (ATSDR, 1997).
Tetrachloroethylene typically enters the environment by evaporating into the air during
use. It can also leach into water supplies and soil during disposal of sewage sludge and factory
waste and when leaking from underground storage tanks. Tetrachloroethylene can remain in the
air for several months before it is broken down into other chemicals or is brought back down to
the soil and water by rain. Much of the tetrachloroethylene that gets into water and soil will
evaporate. However, because tetrachloroethylene can travel through soils quite easily, it can
contaminate underground drinking water supplies. If it infiltrates ground water, it may stay there
for many months without breaking down. Under certain conditions, ambient bacteria break down
tetrachloroethylene, but, some of the chemicals formed may also be harmful. Tetrachloroethylene
also may adhere to the soil under some conditions and remain there (ATSDR, 1997).
Tetrachloroethylene levels in air are generally higher in cities or industrial areas of higher
use compared to more rural or remote areas. It can be smelled at levels of 1,000 ppb in air, but
background levels in air are usually less than 1 ppb. The air close to dry cleaning shops and
chemical waste sites typically has levels higher than background levels but less than 1,000 ppb.
17-1
-------
EPA-OGWDW Contaminant Occurrence Support Document for 2" 6-Year Review Category 1 Contaminants October 2009
Both surface water and ground water may contain tetrachloroethylene, typically at levels less than
1 ppb (ATSDR, 1997).
17.2 Use, Production, and Releases
According to ATSDR (1997), tetrachloroethylene was first commercially produced in the
United States in 1925. Its greatest uses are as a solvent in the textile and dry-cleaning industries.
It is a frequently used chlorinated hydrocarbon solvent and plays a pivotal role as a chemical
intermediate in the production of chlorofluorocarbons. Tetrachloroethylene is also used as a
vapor degreasing agent in metal-cleaning operations (ATSDR, 1997).
The Toxic Substances Control Act (TSCA) Inventory contains confidential data reported
by industry as a partial update of the Chemical Substance Inventory Update System. The data are
reported under the Inventory Update Reporting regulation. Manufacturers and importers are
required to report company information and chemical information, including production volumes
for chemicals they manufactured or imported in excess of 10,000 pounds in the immediately
preceding fiscal year.
The TSCA Chemical Substance Inventory does not provide actual production volumes;
instead, it indicates at which of five production volume ranges a chemical was manufactured or
imported. The TSCA Inventory is updated every four years. Exhibit 17-1 presents the production
of tetrachloroethylene in the United States from 1986 to 2002. The production of
tetrachloroethylene increased after 1994 and has remained in the same range since then.
Exhibit 17-1: Production of Tetrachloroethylene in the United States
Inventory Update Year
1986 Range
1990 Range
1994 Range
1998 Range
2002 Range
Production Volume
> 500 million - 1 billion pounds
> 500 million - 1 billion pounds
> 100 million - 500 million pounds
> 100 million - 500 million pounds
> 100 million - 500 million pounds
Source: USEPA, 2003a
Tetrachloroethylene is listed as a Toxics Release Inventory (TRI) chemical. Exhibit 17-2a
lists the environmental releases of tetrachloroethylene between 1998 and 2006. (Exhibit 17-2b
provides a graphical presentation of the environmental releases.) Air emissions comprised the
majority of on- and off-site releases. They decreased steadily over the period of record, as did
total releases. Underground injection, releases to land, and off-site releases varied, reaching
peaks in 2002 and 2003. The TRI data for tetrachloroethylene were reported from 46 states, plus
17-2
-------
EPA-OGWDW Contaminant Occurrence Support Document for T 6-Year Review Category 1 Contaminants October 2009
Puerto Rico and the Virgin Islands, with 41 states and the Virgin Islands reporting releases in all
9 years. No tetrachloroethylene releases were reported from Alaska, Maine, South Dakota, or
Vermont.
Exhibit 17-2a: Table of Toxics Release Inventory (TRI) Data
Tetrachloroethylene
for
Year
1998
1999
2000
2001
2002
2003
2004
2005
2006
On-Site Releases (pounds)
Total Air
Emissions
5,693,129
4,044,192
3,548,773
3,003,539
2,375,578
2,053,684
2,116,601
1,963,224
1,816,903
Surface Water
Discharges
1,741
1,794
1,159
556
812
1,064
519
673
663
Underground
Injection
6,560
9,185
60,391
111,606
149,077
87,075
136,528
69,701
9,565
Releases to
Land
18,986
104,885
13,262
23,082
144,522
116,064
78,470
115,763
149,569
Off -site
Releases
(pounds)
172,389
217,741
50,061
58,129
363,765
992,243
140,650
95,467
184,590
Total On- &
Off-site
Releases
(pounds)
5,892,805
4,377,797
3,673,646
3,196,912
3,033,754
3,250,131
2,472,768
2,244,828
2,161,291
Source: USEPA, 2008
Exhibit 17-2b: Graph of Toxics Release Inventory (TRI) Data for
Tetrachloroethylene
Pounds Released
Environmental Releases of Tetrachloroethylene
3nnn nnn -,
^ nnn nnn
i nnn nnn
3 nnn nnn
~> nnn nnn
1 nnn nnn
II n
In n n
_J _n
1998 1999 2000 2001 2002 2003 2004 2005 2006
Year
• Air Emissions DSurface Water Discharges D Underground Injection DReleases to Land • Off-site Releases
Source: USEPA, 2008
17-3
-------
EPA-OGWDW Contaminant Occurrence Support Document for 2" 6-Year Review Category 1 Contaminants October 2009
17.3 Occurrence in Ambient Water
Lakes, rivers, and aquifers are the sources of most drinking water. The United States
Geological Survey (USGS) National Water Quality Assessment (NAWQA) program is a
national-scale assessment of the occurrence of contaminants in ambient surface and ground
water. Contaminant occurrence in ambient water provides information on the potential for
contaminants to adversely affect drinking water supplies. Data on the occurrence of
tetrachloroethylene in ambient water are available from the NAWQA program.
17.3.1 EPA Summary Analysis ofNA WQA Data
EPA, with the cooperation of the USGS, performed a summary analysis of all Cycle 1
(1992-2001) water monitoring data from all study units for tetrachloroethylene. Detection
frequencies were computed as the percentage of samples and sites with detections (i.e., with at
least one result equal to or greater than the reporting limit). Sample detections can be biased by
frequent sampling in areas with high (or low) occurrence. Calculating the percentage of sites with
detections can reduce this bias.
The results of the EPA analysis are presented in Exhibit 17-3. Overall,
tetrachloroethylene was detected in 19.6% of samples and at 13.2% of sites. Tetrachloroethylene
was detected more frequently in surface water than in ground water. The median concentration,
based on detections from all sites, was 0.037 |ig/L. The 99th percentile concentration, based on
detections from all sites, was 29 |ig/L.
Exhibit 17-3 EPA Summary Analysis of Tetrachloroethylene Data from NAWQA
Study Units, 1992-2001
ground
water
surface
water
all
sites
Detection Frequency
(detections are results > RL1)
Number
of
Samples
4,578
1,413
5,991
% Samples
with
Detections
12.4%
42.9%
19.6%
Number
of Sites
4,105
190
4,295
% Sites
with
Detections
11.7%
44.7%
13.2%
Concentration Values
(of detections, in ug/L)
Minimum
0.002
0.001
0.001
Median
0.05
0.0339
0.037
95th
Percen-
tile
6.5
0.4119
2.14
99th
Percen-
tile
130
2.3
29
Maximum
6,000
7.2
6,000
1. RLs (Reporting Limits) for tetrachloroethylene varied, but did not exceed 0.002 ug/L.
17.4 Occurrence in Drinking Water
Tetrachloroethylene is regulated as a volatile organic compound (VOC) in drinking water.
All non-purchased community water systems (CWSs) and non-transient non-community water
17-4
-------
EPA-OGWDW Contaminant Occurrence Support Document for 2" 6-Year Review Category 1 Contaminants October 2009
systems (NTNCWSs) are required to sample for VOCs. The maximum waiver period for VOCs
is six years for ground water systems and three years for surface water systems.
All CWSs and NTNCWSs must collect four consecutive quarterly samples during the
initial three-year compliance period.1 If all four samples are non-detections, then the system may
reduce to annual sampling. After three annual samples without a detection, a system may be
granted a waiver. If a compound is detected (but at less than the MCL), the system must take one
sample per quarter until results are below the MCL (minimum of two quarterly samples for
ground water systems and four quarterly samples for surface water systems). If all quarterly
samples are below the MCL, the system may return to annual sampling. If a compound is
detected at a level greater than the MCL, the system (whether ground water or surface water)
must take four consecutive quarterly samples until all are below the MCL. If all quarterly
samples are below the MCL, the system may return to annual sampling.
The analysis of tetrachloroethylene occurrence presented in the following section is based
on state compliance monitoring data from the Six-Year Review-ICR Dataset. This is the largest
and most comprehensive compliance monitoring data set compiled by EPA's Drinking Water
program to date.
All of the Six-Year Review-ICR states provided occurrence data for tetrachloroethylene.
These data consist of 410,042 analytical results from 50,436 public water systems (PWSs) during
the period from 1998 to 2005. The number of sample results and systems vary by state, although
the state data sets have been reviewed and checked to ensure adequacy of coverage and
completeness. The national modal minimum reporting level (MRL) for tetrachloroethylene in the
dataset is 0.5 |ig/L.
17.4.1 Stage 1 Analysis of Occurrence Data
Stage 1 analyses have been performed at the sample, sample point, and system levels.
Note that some systems have multiple sample points, such as source water sample points or entry
points to the distribution system. Exhibit 17-4 presents the number of samples, systems, and
sample points in the data set, as well as the population-served by the systems and sample points,
and the percentage of detections and the percentage of the population-served by the systems and
sample points with detections. Detections in systems provide a fundamental indication of
national occurrence, but are not representative of MCL violations. For tetrachloroethylene,
2.502% of 50,436 systems reported detections. Exhibit 17-5 lists the minimum, median, 90th
percentile, and maximum tetrachloroethylene concentrations based on detections from all
systems in the Six-Year Review-ICR Dataset. The median concentration of tetrachloroethylene,
based on detections from all systems, was 1.6 |ig/L.
1 All new systems or systems using a new water source that began operation after January 22, 2004 must demonstrate
compliance with the MCL within a period of time specified by the State. The system must also comply with the
initial sampling frequencies specified by the State to ensure that a system can demonstrate compliance with the MCL.
17-5
-------
EPA-OGWDW Contaminant Occurrence Support Document for 2" 6-Year Review Category 1 Contaminants October 2009
Exhibit 17-4: Tetrachloroethylene Occurrence Data from the Six-Year Review-ICR
Dataset
Source Water Type
Number of
Samples
Number of
Systems
Number of
Sample Points
Pop. Served
by Systems
Pop. Served
by Sample
Points
Total Number
Ground Water
309,579
46,180
81,142
86,396,173
86,396,173
Surface Water
100,463
4,256
11,030
140,612,382
140,612,382
Combined Ground &
Surface Water
410,042
50,436
92,172
227,008,555
227,008,555
Percent with Detections
Ground Water
6.961%
2.362%
2.780%
15.923%
4.798%
Surface Water
13.200%
4.018%
6.192%
12.940%
4.299%
Combined Ground &
Surface Water
8.490%
2.502%
3.189%
14.076%
4.489%
Exhibit 17-5: Tetrachloroethylene Summary Statistics
Source Water Type
Ground Water
Detection Value (ug/L)
Minimum
0.01
Median
1.6
90tn
Percentile
6.5
Maximum
177
Surface Water
0.001
1.6
14
88
Combined Ground &
Surface Water
0.001
1.6
8.5
177
The following maps illustrate the geographic distribution of tetrachloroethylene
occurrence in drinking water. Exhibit 17-6 shows the states with systems with at least one
detection and Exhibit 17-7 shows the states with systems with at least one detection greater than
the MCL concentration. Note that five states did not submit data for use in the Six-Year Review
(Kansas2, Louisiana, Mississippi, Pennsylvania, and Washington).
The distribution of systems with detections of tetrachloroethylene is geographically
dispersed. Detection rates were high; systems in all states except Hawaii reported detections of
2 Kansas submitted data after the data management and analyses had been conducted; data from Kansas are not
included in the analyses.
17-6
-------
EPA-OGWDW Contaminant Occurrence Support Document for 2" 6-Year Review Category 1 Contaminants October 2009
tetrachloroethylene, although only 5 states reported detections in more 5% of their systems.
Thirty-six states reported at least one system that detected tetrachloroethylene greater than the
MCL concentration of 5 |ig/L.
Exhibit 17-6: Percentage of Systems with at Least One Tetrachloroethylene
Detection, by State
/////t States Not Submitting Any Data
States with No Detections,
6 Leu than 1% of Systems detecting
Between 1% and 5% of Systems Detecting
More than 5% of Syftemg Delecting
Exhibit 17-7: Percentage of Systems with at Least One Tetrachloroethylene
Detection Greater than the MCL Concentration (> 5 [igIL), by State
States Not Submitting Any Data
States with No Detection* > MCL
Less than D.I % Of Systems with Detections > MCL
| Between 0.1% and 0.5% of Systems with Detections > MCL
I More Uian 0.5% of Systems with Detection? > MCL
17-7
-------
EPA-OGWDW Contaminant Occurrence Support Document for 2" 6-Year Review Category 1 Contaminants October 2009
Exhibit 17-8 presents the Stage 1 analysis of tetrachloroethylene occurrence in drinking
water from systems and the population-served by those systems in the Six-Year Review-ICR
Dataset relative to several thresholds. The thresholds are: 5 |ig/L, 2.5 |ig/L, 1 |ig/L, and 0.5 |ig/L.
A total of 404 systems (0.801% of all systems), serving more than 17.5 million persons, reported
at least one detection greater than or equal to 2.5 |ig/L; and 241 systems (0.478% of all systems),
serving about 14.6 million persons, reported at least one detection greater than the MCL
concentration of 5 |ig/L. More ground water systems than surface water systems reported at least
one detection greater than thresholds. However, because surface water systems tend to be larger,
a greater population was served by the surface water systems reporting at least one detection
greater than thresholds.
Exhibit 17-8: Tetrachloroethylene Stage 1 Analysis - Systems and Population
with at Least One Threshold Exceedance
Source
Water Type
Ground
Water
Threshold
> 5 |jg/L
>2.5pg/L
> 1 pg/L
> 0.5 |jg/L
Number of
Systems
Exceeding
Threshold
185
331
615
915
Percent of
Systems
Exceeding
Threshold
0.401%
0.717%
1.332%
1.981%
Total Population-
Served by
Systems
Exceeding
Threshold
3,823,729
6,157,866
10,479,911
12,472,505
Percent of
Population -Served
by Systems
Exceeding
Threshold
4.426%
7.127%
12.130%
14.436%
Surface
Water
> 5 |jg/L
>2.5pg/L
> 1 pg/L
> 0.5 |jg/L
56
73
118
148
1.316%
1.715%
2.773%
3.477%
10,814,217
11,524,206
15,318,905
17,083,237
7.691%
8.196%
10.894%
12.149%
Combined
Ground &
Surface
Water
> 5 |jg/L
>2.5pg/L
> 1 pg/L
> 0.5 pg/L
241
404
733
1,063
0.478%
0.801%
1.453%
2.108%
14,637,946
17,682,072
25,798,816
29,555,742
6.448%
7.789%
11.365%
13.020%
The occurrence estimates presented above (in Exhibit 17-8) are based on the conservative
assumption that if a tetrachloroethylene detection is found in a single entry point/sample point in
a system, then the entire population-served by the system is assumed to potentially be exposed to
tetrachloroethylene (even though other entry points/sampling points at the system, and the
population-served by those entry/sample points, may exhibit no occurrence of
tetrachloroethylene). The sample point analysis is a less conservative estimate of the population-
served by a system with a detection of tetrachloroethylene because it estimates the potentially
exposed population-served based only on the proportion of the population-served by entry
points/sample points with detections of tetrachloroethylene (rather than the entire population-
served by the system).
17-8
-------
EPA-OGWDW Contaminant Occurrence Support Document for 2" 6-Year Review Category 1 Contaminants October 2009
Exhibit 17-9 presents the Stage 1 analysis of tetrachloroethylene occurrence in drinking
water from sample points and population-served by those sample points in the Six-Year Review-
ICR Dataset. A total of 921 (0.999% of) sample points, serving an estimated 4.2 million persons,
reported at least one detection greater than or equal to 2.5 |ig/L; and 481 (0.522% of) sample
points, serving an estimated 2.8 million persons, reported at least one detection exceeding 5
|ig/L. Similar to the system-level occurrence analysis, more ground water sample points had
threshold exceedances. However, because surface water systems tend to be larger, a greater
population was served by sample points with threshold exceedances.
Exhibit 17-9: Tetrachloroethylene Stage 1 Analysis - Sample Points and
Population with at Least One Threshold Exceedance
Source
Water Type
Ground
Water
Threshold
> 5 |jg/L
>2.5pg/L
> 1 pg/L
> 0.5 |jg/L
Number of
Sample Points
Exceeding
Threshold
324
649
1,360
1,973
Percent of
Sample Points
Exceeding
Threshold
0.399%
0.800%
1 .676%
2.432%
Total Population-
Served by Sample
Points Exceeding
Threshold
806,193
1,347,724
2,621,338
3,679,955
Percent of
Population-
Served by
Sample Points
Exceeding
Threshold
0.933%
1.560%
3.034%
4.259%
Surface
Water
> 5 |jg/L
>2.5pg/L
> 1 pg/L
> 0.5 |jg/L
157
272
479
640
1 .423%
2.466%
4.343%
5.802%
1,944,714
2,823,497
4,243,036
5,346,124
1.383%
2.008%
3.018%
3.802%
Combined
Ground &
Surface
Water
> 5 |jg/L
>2.5pg/L
> 1 pg/L
> 0.5 pg/L
481
921
1,839
2,613
0.522%
0.999%
1.995%
2.835%
2,750,906
4,171,221
6,864,374
9,026,080
1.212%
1.837%
3.024%
3.976%
17.4.2 Stage 2 Analysis of Occurrence Data
For the Stage 2 analyses, three different substitution values -zero, one-half the MRL
value, and the full MRL value- were used to replace each non-detection record. Three arithmetic
mean tetrachloroethylene concentrations were calculated at each system using the zero, one-half
MRL and full MRL substitution values. These mean calculations were done for all systems with
data in the Six-Year Review-ICR dataset. Then the percentages of all systems with a mean
concentration greater than each threshold were calculated.
Exhibit 17-10 presents the Stage 2 analysis of estimated system mean concentrations for
tetrachloroethylene occurrence in drinking water relative to several thresholds. The thresholds
are: 5 |ig/L, 2.5 |ig/L, 1 |ig/L, and 0.5 |ig/L. Exhibit 17-11 presents similar information based on
17-9
-------
EPA-OGWDW Contaminant Occurrence Support Document for 2" 6-Year Review Category 1 Contaminants October 2009
population-served by the systems. Based on the Stage 2 analyses, no more than 75 ground water
and surface water systems (approximately 0.149% of all systems) had an estimated system mean
greater than or equal to 2.5 |ig/L. These 75 systems serve about 1.7 million persons. No more
than 25 systems (0.050% of all systems), serving about 1.1 million persons, had an estimated
system mean greater than the MCL concentration of 5 |ig/L.
Exhibit 17-10: Tetrachloroethylene Stage 2 Analysis - Systems with a System
Mean Threshold Exceedance
Source
Water Type
Ground
Water
Threshold
> 5 pg/L
> 2.5 pg/L
> 1 pg/L
> 0.5 |jg/L
Number of Systems with Mean
Exceeding Threshold1
Full MRL
20
61
244
860
1/2 MRL
20
58
214
446
Zero
20
55
190
356
Percent of Systems with Mean
Exceeding Threshold1
Full MRL
0.043%
0.132%
0.528%
1.862%
1/2 MRL
0.043%
0.126%
0.463%
0.966%
Zero
0.043%
0.119%
0.411%
0.771%
Surface
Water
> 5 pg/L
> 2.5 pg/L
> 1 pg/L
> 0.5 |jg/L
5
14
42
145
3
13
37
73
3
13
30
56
0.117%
0.329%
0.987%
3.407%
0.070%
0.305%
0.869%
1.715%
0.070%
0.305%
0.705%
1.316%
Combined
Ground &
Surface
Water
> 5 pg/L
> 2.5 pg/L
> 1 pg/L
> 0.5 pg/L
25
75
286
1,005
23
71
251
519
23
68
220
412
0.050%
0.149%
0.567%
1.993%
0.046%
0.141%
0.498%
1.029%
0.046%
0.135%
0.436%
0.817%
1. The results shown in this table are based on system arithmetic means that were calculated using all sample
detection records and all non-detection records, with the non-detection records replaced (substituted) with
concentration values equal to the full MRL, 1A MRL, and zero.
Exhibit 17-11: Tetrachloroethylene Stage 2 Analysis - Population-Served by
Systems with a System Mean Threshold Exceedance
Source
Water Type
Ground
Water
Threshold
> 5 pg/L
> 2.5 pg/L
> 1 pg/L
> 0.5 pg/L
Population-Served by Systems with
Mean Exceeding Threshold1
Full MRL
81,488
461,738
2,498,326
11,851,080
1/2 MRL
81,488
455,716
2,122,956
4,827,257
Zero
81,488
411,244
1,933,039
3,399,163
Percent of Population-Served by
Systems with Mean Exceeding
Threshold1
Full MRL
0.094%
0.534%
2.892%
13.717%
1/2 MRL
0.094%
0.527%
2.457%
5.587%
Zero
0.094%
0.476%
2.237%
3.934%
17-10
-------
EPA-OGWDW Contaminant Occurrence Support Document for 2" 6-Year Review Category 1 Contaminants October 2009
Source
Water Type
Surface
Water
Threshold
> 5 |jg/L
>2.5|jg/L
> 1 H9/L
> 0.5 |jg/L
Population-Served by Systems with
Mean Exceeding Threshold1
Full MRL
992,445
1,244,255
8,207,545
16,303,826
1/2 MRL
546,718
1,236,118
8,053,939
9,744,394
Zero
546,718
1,236,118
7,691,532
9,009,278
Percent of Population-Served by
Systems with Mean Exceeding
Threshold1
Full MRL
0.706%
0.885%
5.837%
11.595%
1/2 MRL
0.389%
0.879%
5.728%
6.930%
Zero
0.389%
0.879%
5.470%
6.407%
Combined
Ground &
Surface
Water
> 5 |jg/L
>2.5|jg/L
> 1 H9/L
> 0.5 |jg/L
1,073,933
1,705,993
10,705,871
28,154,906
628,206
1,691,834
10,176,895
14,571,651
628,206
1,647,362
9,624,571
12,408,441
0.473%
0.752%
4.716%
12.403%
0.277%
0.745%
4.483%
6.419%
0.277%
0.726%
4.240%
5.466%
1 The results shown in this table are based on system arithmetic means that were calculated using all sample
detection records and all non-detection records, with the non-detection records replaced (substituted) with
concentration values equal to the full MRL, 1A MRL, and zero.
17.5 Additional Occurrence Data
Volatile Organic Compounds in the Nation's Ground Water and Drinking Water Supply Wells
The USGS conducted a national assessment of 55 VOCs, including tetrachloroethylene
(referred to as perchloroethene in the study), in samples collected from drinking water supply
wells, specifically domestic and public wells (Zogorski et a/., 2006). This occurrence assessment
was based on analyses of samples collected at the well head, and before any treatment or
blending, from about 2,400 domestic wells and about 1,100 public wells.
In domestic wells, tetrachloroethylene had a detection frequency of 2.0% at an assessment
level of 0.2 jig/L (based on 2,371 samples). The median concentration of detections was 0.058
|ig/L. Seventeen detections were within one order of magnitude of the MCL concentration (5
|ig/L) and five detections were greater than the MCL concentration.
In public wells, tetrachloroethylene had a detection frequency of 5.3% at an assessment
level of 0.2 |ig/L (based on 1,093 samples). The median concentration of detections was 0.20
|ig/L. Twenty-six detections were within one order of magnitude of the MCL concentration and
nine detections were greater than the MCL concentration.
Occurrence of Water-Quality Constituents in Domestic Wells across the United States
The USGS, in collaboration with the Centers for Disease Control and Prevention (CDC),
released an online report on the occurrence of 11 priority water-quality constituents of possible
health concern in domestic wells located in 16 states across the United States (Bartholomay et
a/., 2007). Two water-quality data sets were used to define the water quality of domestic-water
supplies: (1) data from the NAWQA Program, and (2) USGS state data. The common assessment
level used to quantify the detection frequency of tetrachloroethylene was 0.2 |ig/L.
17-11
-------
EPA-OGWDW Contaminant Occurrence Support Document for 2" 6-Year Review Category 1 Contaminants October 2009
A total of 1,039 wells were sampled for tetrachloroethylene (referred to as
perchloroethene (PCE) in the study). The detection frequency for tetrachloroethylene was 1.3%.
Less than 1% of samples had concentrations of tetrachloroethylene less than but within one order
of magnitude of the MCL concentration (5 |ig/L); 0.1% of samples had concentrations greater
than the MCL concentration.
First Six-Year Review ofNPDWRs
Similar types of Stage 1 contaminant occurrence analyses were conducted for the first
Six-Year Review. The first Six-Year Review assessed regulated contaminant occurrence data
collected from 1993 to 1997. The first Six-Year Review data set consisted of data from 16 states
that were assembled into a "national cross-section" that was qualitatively indicative (though not
statistically representative) of national occurrence. For the first Six-Year Review (USEPA,
2003b), 751 systems (approximately 3.36% of about 22,000 systems), reported at least one
detection of tetrachloroethylene. One hundred and seventy-four systems (approximately 0.78% of
all systems), reported at least one detection greater than the MCL concentration.
17.6 Summary of Data
TRI data indicate that the total off-site and on-site releases of tetrachloroethylene for the
period 1998-2006 ranged from nearly 2.2 million pounds to about 5.9 million pounds. The
majority of releases (63-97%) were air emissions. Surface water discharges were the smallest
fraction. Air emissions and total releases decreased during the period of record.
Ambient water data from the NAWQA program indicate that tetrachloroethylene was
detected in 19.6% of 5,991 ground and surface water samples and at 13.2% of 4,295 sites. The
median concentration, based on detections from all sites, was 0.037 |ig/L.
The Stage 1 analysis of occurrence in drinking water indicates that tetrachloroethylene
was detected in 2.502% of 50,436 surface water and ground water systems. The median
concentration, based on detections from all systems, was 1.6 |ig/L. In total, 0.478% of systems
(serving 6.448% of the population) reported at least one detection greater than the MCL
concentration of 5 |ig/L. Detections greater than the MCL concentration were more common in
more ground water systems than surface water systems. However, because surface water systems
tend to be larger, more people were served by surface water systems reporting at least one
detection greater than the MCL concentration. A similar pattern was seen for sample points.
In the Stage 2 analysis, 0.050% of all systems (serving 0.473 of the population) had an
estimated system mean of tetrachloroethylene greater than the MCL concentration. The number
of systems with threshold exceedances was greater for ground water systems, but larger
population were served by surface water systems with mean exceedances.
17-12
-------
EPA-OGWDW Contaminant Occurrence Support Document for 2" 6-Year Review Category 1 Contaminants October 2009
17.7 References
Agency for Toxic Substances and Disease Registry (ATSDR). 1997. Toxicological Profile for
Tetrachloroethylene. September 1997. Available on the Internet at:
http ://www. atsdr.cdc. gov/toxprofiles/tp 18 .html.
Bartholomay, R.C., J.M. Carter, S.L. Qi, PJ. Squillace, and G.L. Rowe. 2007. Summary of
selected U.S. Geological Survey data on domestic well water quality for the Centers for Disease
Control's National Environmental Public Health Tracking Program: U.S. Geological Survey
Scientific Investigations Report 2007-5213, 57 p. Available on the Internet at:
http://pubs.usgs.gov/sir/2007/5213/.
United States Environmental Protection Agency (USEPA). 1991. National Primary Drinking
Water Regulations—Synthetic Organic Chemicals and Inorganic Chemicals; Monitoring for
Unregulated Contaminants; National Primary Drinking Water Regulations Implementation;
National Secondary Drinking Water Regulations; Final Rule Federal Register. Vol. 56, No. 30. p.
3526, January 30, 1991.
USEPA. 2003a. Non-confidential Inventory Update Reporting (IUR) Production Volume
Information - Tetrachloroethylene. Last modified September 10, 2007. Available on the Internet
at: http://www.epa.gov/oppt/iur/tools/data/2002-vol.htm. [Search for tetrachloroethylene.]
Accessed February 21, 2008.
USEPA. 2003b. Occurrence Estimation Methodology and Occurrence Findings for Six-Year
Review of National Primary Drinking Water Regulations. Office of Water. EPA-815-R-03-006.
June 2003.
USEPA. 2008. TRI Explorer: Trends - Tetrachloroethylene. Released February 21, 2008.
Available on the Internet at: http://www.epa.gov/triexplorer/trends.htm. [Search for
tetrachloroethylene.] Accessed February 29, 2008.
USEPA. 2009. The Analysis of Regulated Contaminant Occurrence Data from Public Water
Systems in Support of the Second Six-Year Review of National Primary Drinking Water
Regulations. Office of Water. EPA-815-B-09-006. October 2009.
Zogorski, J.S., J.M. Carter, T. Ivahnenko, W.W. Lapham, M.J. Moran, B.L. Rowe, P.J. Squillace,
and P.L. Toccalino. 2006. Volatile organic compounds in the nation's ground water and drinking
water supply wells. U.S. Geological Survey Circular 1292, 101 p. Available on the Internet at:
http://pubs.er.usgs.gov/pubs/cir/cirl292.
17-13
-------
EPA-OGWDW Contaminant Occurrence Support Document for 2" 6-Year Review Category 1 Contaminants October 2009
18 Toluene
This chapter on toluene is part of a report that is organized so that each chapter presents
information on one specific contaminant. Chapter 1 of this report, Introduction, presents
background information on all sources of data used as well as the analytical approach used to
estimate contaminant occurrence in drinking water. This contaminant chapter includes
background information (such as regulatory history, general chemical information, and
environmental fate and behavior), use and environmental release information, and toluene
occurrence estimates in ambient water and drinking water. All drinking water occurrence
estimates were based on data from the National Compliance Monitoring Information Collection
Request (ICR) Dataset for the Second Six-Year Review (the "Six Year Review-ICR Dataset").
For detailed information regarding this dataset, data management, data quality assessments, and
the analytical approach used to generate all contaminant occurrence estimates presented in this
chapter, please refer to USEPA (2009).
18.1 Background
The United States Environmental Protection Agency (EPA) published the current
National Primary Drinking Water Regulations (NPDWR) for toluene on January 30, 1991 (56 FR
3526 (USEPA, 1991)). The NPDWR established a maximum contaminant level goal (MCLG)
and a maximum contaminant level (MCL) of 1,000 |ig/L. The Agency based the MCLG on a
reference dose (RfD) of 200 jig/kg-day (0.2 mg/kg-day) and a cancer classification of D, not
classifiable as to human carcinogenicity.
According to the Agency for Toxic Substances and Disease Registry (ATSDR, 2000),
toluene (CeHsCHa) is a clear, colorless liquid with a distinctive smell. Toluene occurs naturally
in crude oil and in the tolu tree. It is also produced in the process of making gasoline and other
fuels from crude oil, in making coke from coal, and as a by-product in the manufacture of styrene
(ATSDR, 2000).
Toluene evaporates very easily and enters the air when products that contain it are used. It
enters surface water and ground water from spills of solvents and petroleum products as well as
from leaking underground storage tanks at gasoline stations. Because toluene is frequently used
to dissolve other substances, it often accompanies other chemicals to waste sites, where it can
contaminate surrounding soil and water. Once in the environment, toluene tends to break down
rapidly, especially in the presence of microorganisms. It does not tend to accumulate in animals
or plants because it is metabolized (ATSDR, 2000).
Persons who smoke or who work with gasoline, kerosene, heating oil, paints, and
lacquers are at the greatest risk of toluene exposure. Common household products, such as nail
polish, cosmetics, rubber cement, stain removers, fabric dyes, inks, and adhesives, also provide
exposure to toluene. Waste sites are a risk factor as well. EPA reported in 1991 that toluene was
found in well water, surface water, or soil at 63% of the hazardous waste sites surveyed. Toluene
concentrations, inside and outside, are usually less than 1 ppm when not close to industry. Non-
smokers who do not work with toluene-containing products are probably only exposed to about
18-1
-------
EPA-OGWDW Contaminant Occurrence Support Document for 2" 6-Year Review Category 1 Contaminants October 2009
300 jig of toluene a day. Smoking a pack of cigarettes a day can add another 1,000 jig to daily
exposure (ATSDR, 2000).
18.2 Use, Production, and Releases
Toluene is a good solvent and is often added to gasoline along with benzene and xylene
to improve octane ratings (ATSDR, 2000). In addition, it is used in the manufacture of paints,
paint thinners, fingernail polish, lacquers, adhesives, and rubber, as well as in some printing and
leather tanning processes.
The Toxic Substances Control Act (TSCA) Inventory contains confidential data reported
by industry as a partial update of the Chemical Substance Inventory Update System. The data are
reported under the Inventory Update Reporting regulation. Manufacturers and importers are
required to report company information and chemical information, including production volumes
for chemicals they manufactured or imported in excess of 10,000 pounds in the immediately
preceding fiscal year.
The TSCA Chemical Substance Inventory does not provide actual production volumes;
instead, it indicates at which of five production volume ranges a chemical was manufactured or
imported. The TSCA Inventory is updated every four years. Exhibit 18-1 presents the production
of toluene in the United States from 1986 to 2002. Note that the production of toluene in the
United States exceeded 1 billion pounds since 1986.
Exhibit 18-1: Production of Toluene in the United States
Inventory Update Year
1986 Range
1990 Range
1994 Range
1998 Range
2002 Range
Production Volume
> 1 billion pounds
> 1 billion pounds
> 1 billion pounds
> 1 billion pounds
> 1 billion pounds
Source: USEPA, 2003a
Toluene is listed as a Toxics Release Inventory (TRI) chemical. Exhibit 18-2a lists the
environmental releases of toluene from 1998 to 2006. (Exhibit 18-2b provides a graphical
presentation of the environmental releases.) Air emissions constituted more than 90% of total
releases and declined steadily over the period of record. Off-site releases were the next largest
fraction and generally comprised 1 to 3% of the total. Surface water discharges, underground
injection, and releases to land contributed variable but small fractions. The TRI data for toluene
were reported from all 50 states, plus American Samoa, Guam, the Northern Mariana Islands,
18-2
-------
EPA-OGWDW Contaminant Occurrence Support Document for 2" 6-Year Review Category 1 Contaminants October 2009
Puerto Rico and the Virgin Islands. All states and territories except for American Samoa reported
toluene releases in all nine years.
Exhibit 18-2a: Table of Toxics Release Inventory (TRI) Data for Toluene
Year
1998
1999
2000
2001
2002
2003
2004
2005
2006
On-Site Releases (pounds)
Total Air
Emissions
99,562,995
91,706,170
82,067,078
71,820,887
64,034,648
55,454,167
53,079,139
52,948,091
45,002,472
Surface Water
Discharges
43,783
44,609
40,494
75,415
28,798
28,578
266,254
30,222
31,971
Underground
Injection
724,075
728,385
513,810
265,020
1,281,792
732,951
915,605
1,389,513
468,857
Releases to
Land
94,628
310,517
148,061
127,657
87,574
938,753
1,112,241
1,232,729
2,072,125
Off -site
Releases
(pounds)
1,797,274
2,295,834
1,615,834
1,891,934
1,129,282
1,067,495
1,389,014
1,660,673
1,312,985
Total On- &
Off-site
Releases
(pounds)
102,222,755
95,085,515
84,385,276
74,180,913
66,562,093
58,221,944
56,762,253
57,261,229
48,888,410
Source: USEPA, 2008
Exhibit 18-2b: Graph of Toxics Release Inventory (TRI) Data for Toluene
Pounds Released
Environmental Releases of Toluene
ion nnn nnn -,
-inn nnn nnn
on nnn nnn
fin nnn nnn
An nnn nnn
on nnn nnn
^^^-^-^^
^HH^t
TnHHHn
1998 1999 2000 2001 2002 2003 2004 2005 2006
Year
•Air Emissions DSurface Water Discharges D Underground Injection DReleases to Land BOff-site Releases
Source: USEPA, 2008
18-3
-------
EPA-OGWDW Contaminant Occurrence Support Document for 2" 6-Year Review Category 1 Contaminants October 2009
18.3 Occurrence in Ambient Water
Lakes, rivers, and aquifers are the sources of most drinking water. The United States
Geological Survey (USGS) National Water Quality Assessment (NAWQA) program is a
national-scale assessment of the occurrence of contaminants in ambient surface and ground
water. Contaminant occurrence in ambient water provides information on the potential for
contaminants to adversely affect drinking water supplies. Data on the occurrence of toluene in
ambient water are available from the NAWQA program.
18.3.1 EPA Summary Analysis of NAWQA Data
EPA, with the cooperation of the USGS, performed a summary analysis of all Cycle 1
(1992-2001) water monitoring data from all study units for toluene. Detection frequencies were
computed as the percentage of samples and sites with detections (i.e., with at least one result
equal to or greater than the reporting limit). Sample detections can be biased by frequent
sampling in areas with high (or low) occurrence. Calculating the percentage of sites with
detections can reduce this bias.
The results of the EPA analysis are presented in Exhibit 18-3. Overall, toluene was
detected in 26.3% of samples and at 15.9% of sites. Toluene was detected more frequently in
surface water than in ground water. The median concentration, based on detections from all sites,
was 0.052 |ig/L. The 99th percentile concentration, based on detections from all sites, was 3.48
Exhibit 18-3: EPA Summary Analysis of Toluene Data from NAWQA Study Units,
1992-2001
ground
water
surface
water
all
sites
Detection Frequency
(detections are results > RL1)
Number
of
Samples
4,545
1,394
5,939
% Samples
with
Detections
13.1%
69.4%
26.3%
Number
of Sites
4,061
182
4,243
% Sites
with
Detections
13.9%
60.4%
15.9%
Concentration Values
(of detections, in ug/L)
Minimum
0.003
0.006
0.003
Median
0.0356
0.06
0.0515
95th
Percen-
tile
0.8845
0.42
0.6
99th
Percen-
tile
12
1.289
3.483
Maximum
1300
11
1300
1. RLs (Reporting Limits) for toluene varied, but did not exceed 0.006 pg/L
18.4 Occurrence in Drinking Water
Toluene is regulated as a volatile organic compound (VOC) in drinking water. All non-
purchased community water systems (CWSs) and non-transient non-community water systems
(NTNCWSs) are required to sample for VOCs. The maximum waiver period for VOCs is six
years for ground water systems and three years for surface water systems.
18-4
-------
EPA-OGWDW Contaminant Occurrence Support Document for 2" 6-Year Review Category 1 Contaminants October 2009
All CWSs and NTNCWSs must collect four consecutive quarterly samples during the
initial three-year compliance period.1 If all four samples are non-detections, then the system may
reduce to annual sampling. After three annual samples without a detection, a system may be
granted a waiver. If a compound is detected (but at less than the MCL), the system must take one
sample per quarter until results are below the MCL (minimum of two quarterly samples for
ground water systems and four quarterly samples for surface water systems). If all quarterly
samples are below the MCL, the system may return to annual sampling. If a compound is
detected at a level greater than the MCL, the system (whether ground water or surface water)
must take four consecutive quarterly samples until all are below the MCL. If all quarterly
samples are below the MCL, the system may return to annual sampling.
The analysis of toluene occurrence presented in the following section is based on state
compliance monitoring data from the Six- Year Review-ICR Dataset. This is the largest and most
comprehensive compliance monitoring data set compiled by EPA's Drinking Water program to
date.
All of the Six-Year Review-ICR states provided occurrence data for toluene. These data
consist of 374,250 analytical results from about 50,451 public water systems (PWSs) during the
period from 1998 to 2005. The number of sample results and systems vary by state, although the
state data sets have been reviewed and checked to ensure adequacy of coverage and
completeness. The national modal minimum reporting level (MRL) for toluene in the dataset is
0.5
18.4.1 Stage 1 Analysis of Occurrence Data
Stage 1 analyses have been performed at the sample, sample point, and system levels.
Note that some systems have multiple sample points, such as source water sample points or entry
points to the distribution system. Exhibit 18-4 presents the number of samples, systems, and
sample points in the data set, as well as the population-served by the systems and sample points,
and the percentage of detections and the percentage of the population-served by the systems and
sample points with detections. Detections in systems provide a fundamental indication of
national occurrence, but are not representative of MCL violations. For toluene, 5.764% of 50,45 1
systems reported detections. Exhibit 18-5 lists the minimum, median, 90th percentile, and
maximum toluene concentrations based on detections from all systems in the Six-Year Review-
ICR Dataset. The median concentration of toluene, based on detections from all systems, was
0.94
1 All new systems or systems using a new water source that began operation after January 22, 2004 must demonstrate
compliance with the MCL within a period of time specified by the State. The system must also comply with the
initial sampling frequencies specified by the State to ensure that a system can demonstrate compliance with the MCL.
18-5
-------
EPA-OGWDW Contaminant Occurrence Support Document for 2" 6-Year Review Category 1 Contaminants October 2009
Exhibit 18-4: Toluene Occurrence Data from the Six-Year Review-ICR Dataset
Source Water Type
Number of
Samples
Number of
Systems
Number of
Sample Points
Pop. Served
by Systems
Pop. Served
by Sample
Points
Total Number
Ground Water
289,635
46,193
80,577
86,403,546
86,403,546
Surface Water
84,615
4,258
10,870
140,551,755
140,551,755
Combined Ground &
Surface Water
374,250
50,451
91,447
226,955,301
226,955,301
Percent with Detections
Ground Water
1.381%
5.535%
3.561%
9.361%
2.912%
Surface Water
0.779%
8.243%
3.947%
15.584%
4.391%
Combined Ground &
Surface Water
1 .245%
5.764%
3.606%
13.215%
3.828%
Exhibit 18-5: Toluene Summary Statistics
Source Water Type
Ground Water
Detection Value (ug/L)
Minimum
0.0077
Median
1
90tn
Percentile
6
Maximum
5,100
Surface Water
0.002
0.7
2.9
740
Combined Ground &
Surface Water
0.002
0.94
5.4
5,100
The following maps illustrate the geographic distribution of toluene occurrence in
drinking water. Exhibit 18-6 shows the states with systems with at least one detection and
Exhibit 18-7 shows the states with systems with at least one detection greater than the MCL
concentration. Note that five states did not submit data for use in the Six-Year Review (Kansas2,
Louisiana, Mississippi, Pennsylvania, and Washington).
The distribution of systems with detections of toluene is geographically dispersed.
Detection rates were high; systems in all states except Hawaii reported detections of toluene,
although only 8 states reported detections in more 10% of their systems. Although toluene
2 Kansas submitted data after the data management and analyses had been conducted; data from Kansas are not
included in the analyses.
18-6
-------
EPA-OGWDW Contaminant Occurrence Support Document for 2" 6-Year Review Category 1 Contaminants October 2009
detection rates were high, only Michigan reported detections greater than the MCL concentration
of 1,000 iig/L.
Exhibit 18-6: Percentage of Systems with at Least One Toluene Detection, by
State
States Not Submitting Any Data
States with No Detections
Loss than 5% of Systems Detecting
| Between 5% and 10% of Systems Detecting
| More than 10% of Systems Detecting
Exhibit 18-7: Percentage of Systems with at Least One Toluene Detection Greater
than the MCL Concentration (> 1,000 ug/L), by State
States Not Submitting Any Data
States with No Detections > MCL
Less than 0.5% erf Systems with Detections > MCL
I Between 0,5% and 1% of Systems wim Detections > MCL
More inan 1% qf Systems with Detections > MCL
18-7
-------
EPA-OGWDW Contaminant Occurrence Support Document for 2" 6-Year Review Category 1 Contaminants October 2009
Exhibit 18-8 presents the Stage 1 analysis of toluene occurrence in drinking water from
systems and the population-served by those systems in the Six-Year Review-ICR Dataset relative
to several thresholds. The thresholds are: 1,000 |ig/L, 600 |ig/L, and 500 |ig/L. Eighteen systems
(0.036% of all systems), serving about 22,000 persons, reported at least one detection greater
than or equal to 500 |ig/L; and one system, serving 500 persons, reported at least one detection
greater than the MCL concentration of 1,000 |ig/L. Toluene was detected in more ground water
systems than surface water systems at levels greater than all thresholds. No surface water systems
detected toluene at a level greater than 1,000 |ig/L.
Exhibit 18-8: Toluene Stage 1 Analysis - Systems and Population with at Least
One Threshold Exceedance
Source
Water Type
Ground
Water
Threshold
> 1,000|jg/L
>
> Mg/L
Number of
Systems
Exceeding
Threshold
1
14
15
Percent of
Systems
Exceeding
Threshold
0.002%
0.030%
0.032%
Total Population -
Served by
Systems
Exceeding
Threshold
500
15,689
15,757
Percent of
Population -Served
by Systems
Exceeding
Threshold
0.001%
0.018%
0.018%
Surface
Water
> 1,000|jg/L
>
>
0
1
3
0.000%
0.023%
0.070%
0
1,612
6,212
0.000%
0.001%
0.004%
Combined
Ground &
Surface
Water
>1,000|jg/L
>
>
1
15
18
0.002%
0.030%
0.036%
500
17,301
21,969
0.000%
0.008%
0.010%
The occurrence estimates presented above (in Exhibit 18-8) are based on the conservative
assumption that if a toluene detection is found in a single entry point/sample point in a system,
then the entire population-served by the system is assumed to potentially be exposed to toluene
(even though other entry points/sampling points at the system, and the population-served by
those entry/sample points, may exhibit no occurrence of toluene). The sample point analysis is a
less conservative estimate of the population-served by a system with a detection of toluene
because it estimates the potentially exposed population-served based only on the proportion of
the population-served by entry points/sample points with detections of toluene (rather than the
entire population-served by the system).
18-8
-------
EPA-OGWDW Contaminant Occurrence Support Document for 2" 6-Year Review Category 1 Contaminants October 2009
Exhibit 18-9 presents the Stage 1 analysis of toluene occurrence in drinking water from
sample points and population-served by those sample points in the Six-Year Review-ICR
Dataset. Eighteen (0.020% of) sample points, serving an estimated 11,500 persons, had at least
one detection greater than or equal to 500 |ig/L; and one sample point, serving an estimated 500
persons, had at least one detection exceeding 1,000 |ig/L. Similar to the system-level occurrence
analysis, toluene was detected in more ground water than surface water sample points at all
thresholds.
Exhibit 18-9: Toluene Stage 1 Analysis - Sample Points and Population with at
Least One Threshold Exceedance
Source
Water Type
Ground
Water
Threshold
> 1,000|jg/L
>
>
Number of
Sample Points
Exceeding
Threshold
1
14
15
Percent of
Sample
Points
Exceeding
Threshold
0.001%
0.017%
0.019%
Total Population-
Served by
Sample Points
Exceeding
Threshold
500
6,243
6,311
Percent of
Population -Served
by Sample Points
Exceeding
Threshold
0.001%
0.007%
0.007%
Surface
Water
> 1,000|jg/L
>
>
0
1
3
0.000%
0.009%
0.028%
0
1,612
5,212
0.000%
0.001%
0.004%
Combined
Ground &
Surface
Water
> 1,000|jg/L
>
>
1
15
18
0.001%
0.016%
0.020%
500
7,855
11,523
0.0002%
0.003%
0.005%
18.4.2 Stage 2 Analysis of Occurrence Data
For the Stage 2 analyses, three different substitution values -zero, one-half the MRL
value, and the full MRL value- were used to replace each non-detection record. Three arithmetic
mean toluene concentrations were calculated at each system using the zero, one-half MRL and
full MRL substitution values. These mean calculations were done for all systems with data in the
Six-Year Review-ICR dataset. Then the percentages of all systems with a mean concentration
greater than each threshold were calculated.
Exhibit 18-10 presents the Stage 2 analysis of estimated system mean concentrations for
toluene occurrence in drinking water relative to several thresholds. The thresholds are: 1,000
|ig/L, 600 |ig/L, and 500 |ig/L. Exhibit 18-11 presents similar information based on population-
served by the systems. Based on the three substitution methods, two ground water systems
18-9
-------
EPA-OGWDW Contaminant Occurrence Support Document for 2" 6-Year Review Category 1 Contaminants October 2009
(approximately 0.004% of all systems) had estimated system means greater than or equal to 500
Hg/L. These 2 systems serve 840 persons. One ground water system (0.002% of all systems),
serving 500 persons, had an estimated system mean greater than the MCL concentration of 1,000
Hg/L. None of the surface water systems had a mean greater than or equal to any of the
thresholds.
Exhibit 18-10: Toluene Stage 2 Analysis - Systems with a System Mean Threshold
Exceedance
Source
Water Type
Ground
Water
Threshold
> 1,000|jg/L
>
>
Number of Systems with Mean
Exceeding Threshold1
Full MRL
1
2
2
1/2 MRL
1
2
2
Zero
1
2
2
Percent of Systems with Mean
Exceeding Threshold1
Full MRL
0.002%
0.004%
0.004%
1/2 MRL
0.002%
0.004%
0.004%
Zero
0.002%
0.004%
0.004%
Surface
Water
> 1,000|jg/L
>
>
0
0
0
0
0
0
0
0
0
0.000%
0.000%
0.000%
0.000%
0.000%
0.000%
0.000%
0.000%
0.000%
Combined
Ground &
Surface
Water
> 1,000|jg/L
>
>
1
2
2
1
2
2
1
2
2
0.002%
0.004%
0.004%
0.002%
0.004%
0.004%
0.002%
0.004%
0.004%
1. The results shown in this table are based on system arithmetic means that were calculated using all sample
detection records and all non-detection records, with the non-detection records replaced (substituted) with
concentration values equal to the full MRL, 1A MRL, and zero.
Exhibit 18-11: Toluene Stage 2 Analysis - Population-Served by Systems with a
System Mean Threshold Exceedance
Source
Water Type
Ground
Water
Threshold
> 1,000|jg/L
>
>
Population-Served by Systems with
Mean Exceeding Threshold1
Full MRL
500
840
840
1/2 MRL
500
840
840
Zero
500
840
840
Percent of Population-Served by
Systems with Mean Exceeding
Threshold1
Full MRL
0.001%
0.001%
0.001%
1/2 MRL
0.001%
0.001%
0.001%
Zero
0.001%
0.001%
0.001%
Surface
Water
> 1,000|jg/L
>
>
0
0
0
0
0
0
0
0
0
0.000%
0.000%
0.000%
0.000%
0.000%
0.000%
0.000%
0.000%
0.000%
18-10
-------
EPA-OGWDW Contaminant Occurrence Support Document for 2" 6-Year Review Category 1 Contaminants October 2009
Source
Water Type
Combined
Ground &
Surface
Water
Threshold
> 1,000|jg/L
>
>
Population-Served by Systems with
Mean Exceeding Threshold1
Full MRL
500
840
840
1/2 MRL
500
840
840
Zero
500
840
840
Percent of Population-Served by
Systems with Mean Exceeding
Threshold1
Full MRL
0.0002%
0.0004%
0.0004%
1/2 MRL
0.0002%
0.0004%
0.0004%
Zero
0.0002%
0.0004%
0.0004%
1. The results shown in this table are based on system arithmetic means that were calculated using all sample
detection records and all non-detection records, with the non-detection records replaced (substituted) with
concentration values equal to the full MRL, 1A MRL, and zero.
18.5 Additional Occurrence Data
First Six-Year Review ofNPDWRs
Similar types of Stage 1 contaminant occurrence analyses were conducted for the first
Six-Year Review. The first Six-Year Review assessed regulated contaminant occurrence data
collected from 1993 to 1997. The first Six-Year Review data set consisted of data from 16 states
that were assembled into a "national cross-section" that was qualitatively indicative (though not
statistically representative) of national occurrence. For the first Six-Year Review (USEPA,
2003b), 1,133 systems (approximately 4.73% of about 24,000 systems), reported at least one
detection of toluene. No systems reported a detection greater than the MCL concentration.
18.6 Summary of Data
TRI data indicate that total off-site and on-site releases of toluene for the period 1998-
2006 ranged from approximately 49 million to 102 million pounds, with an overall decrease with
time. Air emissions constituted almost all of the volume released. Off-site releases and the other
on-site releases comprised only a few percent of the total amounts released.
Ambient water data from the NAWQA program indicate that toluene was detected in
26.3% of 5,939 ground and surface water samples and at 15.9% of 4,243 sites. The median
concentration, based on detections from all sites, was 0.05 |ig/L.
The Stage 1 analysis of occurrence in drinking water indicates that toluene was detected
in 5.764% of 50,451 surface water and ground water systems. The median concentration, based
on detections from all systems, was 0.94 |ig/L. Only one ground water system (serving 500
persons) reported at least one detection of toluene greater than the MCL concentration. That
system represents 0.002% of all systems and 0.0002% of the population. None of the surface
water systems reported any detections greater than the MCL concentration. The same pattern was
seen for ground water and surface water sample points.
In the Stage 2 analysis, one ground water system (serving 500 persons) had a system
mean concentration of toluene greater than the MCL concentration. No surface water systems
had a system mean exceeding the MCL concentration.
18-11
-------
EPA-OGWDW Contaminant Occurrence Support Document for 2" 6-Year Review Category 1 Contaminants October 2009
18.7 References
Agency for Toxic Substances and Disease Registry (ATSDR). 2000. Toxicological Profile for
Toluene. September 2000. Available on the Internet at:
http://www.atsdr.cdc.gov/toxprofiles/tp56.html.
United States Environmental Protection Agency (USEPA). 1991. National Primary Drinking
Water Regulations—Synthetic Organic Chemicals and Inorganic Chemicals; Monitoring for
Unregulated Contaminants; National Primary Drinking Water Regulations Implementation;
National Secondary Drinking Water Regulations; Final Rule Federal Register. Vol. 56, No. 30. p.
3526, January 30, 1991.
USEPA. 2003a. Non-confidential Inventory Update Reporting (IUR) Production Volume
Information - Toluene. Last modified September 10, 2007. Available on the Internet at:
http://www.epa.gov/oppt/iur/tools/data/2002-vol.htm. [Search for toluene.] Accessed February
21,2008.
USEPA. 2003b. Occurrence Estimation Methodology and Occurrence Findings for Six-Year
Review of National Primary Drinking Water Regulations. Office of Water. EPA-815-R-03-006.
June 2003.
USEPA. 2008. TRI Explorer: Trends - Toluene. Released February 21, 2008. Available on the
Internet at: http://www.epa.gov/triexplorer/trends.htm. [Search for toluene.] Accessed February
29, 2008.
USEPA. 2009. The Analysis of Regulated Contaminant Occurrence Data from Public Water
Systems in Support of the Second Six-Year Review of National Primary Drinking Water
Regulations. Office of Water. EPA-815-B-09-006. October 2009.
18-12
-------
EPA-OGWDW Contaminant Occurrence Support Document for 2" 6-Year Review Category 1 Contaminants October 2009
19 Toxaphene
This chapter on toxaphene is part of a report that is organized so that each chapter
presents information on one specific contaminant. Chapter 1 of this report, Introduction, presents
background information on all sources of data used as well as the analytical approach used to
estimate contaminant occurrence in drinking water. This contaminant chapter includes
background information (such as regulatory history, general chemical information, and
environmental fate and behavior), use and environmental release information, and toxaphene
occurrence estimates in ambient water and drinking water. All drinking water occurrence
estimates were based on data from the National Compliance Monitoring Information Collection
Request (ICR) Dataset for the Second Six-Year Review (the "Six Year Review-ICR Dataset").
For detailed information regarding this dataset, data management, data quality assessments, and
the analytical approach used to generate all contaminant occurrence estimates presented in this
chapter, please refer to USEPA (2009).
19.1 Background
The United States Environmental Protection Agency (EPA) published the current
National Primary Drinking Water Regulations (NPDWR) for toxaphene on January 30, 1991 (56
FR 3526 (USEPA, 1991)). The NPDWR established a maximum contaminant level goal
(MCLG) of zero based on a cancer classification of B2, probable human carcinogen. The
NPDWR also established a maximum contaminant level (MCL) of 3 |ig/L based on analytical
feasibility.
According to the Agency for Toxic Substances and Disease Registry (ATSDR, 1996),
toxaphene (CioHioClg) is a manufactured insecticide containing over 670 chemicals. In its
original form, toxaphene is a yellow to amber waxy solid that smells like turpentine, but it may
also exist as a gas. It evaporates when in solid form or when mixed with liquids, but it does not
burn (ATSDR, 1996).
Toxaphene enters the environment after it is applied to a crop or poured into a lake (to
kill unwanted fish). It does not dissolve well in water, so it is more likely to be found in air, soil,
or the sediment at the bottom of lakes and streams. Toxaphene breaks down very slowly in the
environment, so it still represents a health risk even though it has not been widely used for over
20 years. Levels may be high in some predatory fish and mammals because toxaphene
accumulates in fatty tissues (ATSDR, 1996).
Because the use of toxaphene is now banned in the United States, individuals can
probably only be exposed to it in areas where it is concentrated, such as waste sites. Wildlife in
such areas may be dangerous to eat. Toddlers, who are likely to put things in their mouths, are at
risk of eating contaminated soil. Though one can be exposed to toxaphene by inhaling
contaminated air, concentrations are likely too low to cause health effects (ATSDR, 1996).
19-1
-------
EPA-OGWDW Contaminant Occurrence Support Document for 2" 6-Year Review Category 1 Contaminants October 2009
19.2 Use, Production, and Releases
Toxaphene was one of the most heavily used insecticides in the United States until 1982.
It was used primarily in the southern United States to control insect pests on cotton and other
crops. Toxaphene was also used to control insect pests on livestock and to kill unwanted fish in
lakes (ATSDR, 1996).
The Chemical Substance Inventory Update System contains confidential data reported by
industry as a partial update of the Toxic Substances Control Act (TSCA) Inventory. The TSCA
Chemical Substance Inventory does not report production data for toxaphene.
Toxaphene is listed as a Toxics Release Inventory (TRI) chemical. Exhibit 19-la lists the
environmental releases of toxaphene from 1998 to 2006. (Exhibit 19-lb provides a graphical
presentation of the environmental releases.) On-site releases of toxaphene consistently exceeded
off-site releases. Releases to land constituted most of the on-site releases, except in 1999 when
no toxaphene was released to land. Toxaphene was only injected underground in 2004 at
negligible levels. Surface was discharges were also negligible, except for 2003. Off-site releases
remained at low levels during the period of record, except for an anomalously high value in 2001.
The TRI data for toxaphene were reported from 19 states, with Arkansas, Ohio and Texas
reporting releases in all nine years.
Exhibit 19-1 a: Table of Toxics Release Inventory (TRI) Data for Toxaphene
Year
1998
1999
2000
2001
2002
2003
2004
2005
2006
On-Site Releases (pounds)
Total Air
Emissions
13
16
21
43
38
7
7
4
9
Surface Water
Discharges
0
1
2
6
24
547
1
1
1
Underground
Injection
0
0
0
0
0
0
10
0
0
Releases to
Land
25,476
0
5,928
3,074
5,666
1,802
6,699
5,881
4,584
Off-site
Releases
(pounds)
113
14
55
1,090
18
19
70
22
23
Total On- &
Off-site
Releases
(pounds)
25,602
31
6,006
4,213
5,747
2,374
6,787
5,908
4,616
Source: USEPA, 2008
19-2
-------
EPA-OGWDW Contaminant Occurrence Support Document for 2" 6-Year Review Category 1 Contaminants October 2009
Exhibit 19-1b: Graph of Toxics Release Inventory (TRI) Data for Toxaphene
Pounds Released
Environmental Releases of Toxaphene
•an nnn -,
oc; nnn
on nnn -
-ic nnn
-in nnn -
^ nnn
n -
Dn F
nU [] _n [
1998 1999 2000 2001 2002 2003 2004
Year
__
2005 2006
•Air Emissions DSurface Water Discharges D Underground Injection D Releases to Land • Off-site Releases
Source: USEPA, 2008
19.3 Occurrence in Ambient Water
No National Water Quality Assessment (NAWQA) Program data are available from the
United States Geological Survey (USGS) on the occurrence of toxaphene in ambient waters from
the Pesticide National Synthesis Project (Gilliom et a/., 2007) or the EPA Summary Analysis of
Cycle 1 data.
19.4 Occurrence in Drinking Water
Toxaphene is regulated as a synthetic organic chemical (SOC) in drinking water. All non-
purchased community water systems (CWSs) and non-transient non-community water systems
(NTNCWSs) are required to sample for SOCs. The maximum waiver period for SOCs is three
years, but this waiver can be renewed indefinitely. (Systems may receive a waiver at any time
without having collected an SOC sample. With a waiver, systems are not required to collect an
SOC sample.)
All CWSs and NTNCWSs without an SOC waiver must collect four consecutive
quarterly samples during the initial three-year compliance period.1 If all four samples are non-
detections, then a system serving less than 3,300 persons may reduce to one sample during each
1 All new systems or systems using a new water source that began operation after January 22, 2004 must demonstrate
compliance with the MCL within a period of time specified by the State. The system must also comply with the
initial sampling frequencies specified by the State to ensure that a system can demonstrate compliance with the MCL.
19-3
-------
EPA-OGWDW Contaminant Occurrence Support Document for 2" 6-Year Review Category 1 Contaminants October 2009
consecutive compliance period; a system serving more than 3,300 persons may reduce to two
quarterly samples within a 12-month period during a 3-year compliance period. If a chemical is
detected (but is less than the MCL), the system must take one sample per quarter until results are
below the MCL (minimum of two quarterly samples for ground water systems and four quarterly
samples for surface water systems). If all quarterly samples are below the MCL, the system may
return to annual sampling. If a chemical is detected at a level greater than the MCL, the system
(whether ground water or surface water) must take quarterly samples until four consecutive
quarters are below the MCL. If all quarterly samples are below the MCL, the system may return
to annual sampling.
The analysis of toxaphene occurrence presented in the following section is based on state
compliance monitoring data from the Six-Year Review-ICR Dataset. This is the largest and most
comprehensive compliance monitoring data set compiled by EPA's Drinking Water program to
date.
Forty-four of the Six-Year Review-ICR states provided occurrence data for toxaphene.
(There were no toxaphene data from Oklahoma.) These data consist of 115,403 analytical results
from 30,387 public water systems (PWSs) during the period from 1998 to 2005. The number of
sample results and systems vary by state, although the state data sets have been reviewed and
checked to ensure adequacy of coverage and completeness. The national modal minimum
reporting level (MRL) for toxaphene in the dataset is 1 |ig/L.
19.4.1 Stage 1 Analysis of Occurrence Data
Stage 1 analyses have been performed at the sample, sample point, and system levels.
Note that some systems have multiple sample points, such as source water sample points or entry
points to the distribution system. Exhibit 19-2 presents the number of samples, systems, and
sample points in the data set, as well as the population-served by the systems and sample points,
and the percentage of detections and the percentage of the population-served by the systems and
sample points with detections. Detections in systems provide a fundamental indication of
national occurrence, but are not representative of MCL violations. For toxaphene, 0.132% of
30,387 systems reported detections. Exhibit 19-3 lists the minimum, median, 90th percentile, and
maximum toxaphene concentrations based on detections from all systems in the Six-Year
Review-ICR Dataset. The median concentration of toxaphene, based on detections from all
systems, was 1 |ig/L.
Exhibit 19-2: Toxaphene Occurrence Data from the Six-Year Review-ICR Dataset
Source Water Type
Number of
Samples
Number of
Systems
Number of
Sample Points
Pop. Served
by Systems
Pop. Served
by Sample
Points
Total Number
Ground Water
94,872
27,718
44,769
60,886,049
60,886,049
19-4
-------
EPA-OGWDW Contaminant Occurrence Support Document for 2" 6-Year Review Category 1 Contaminants October 2009
Source Water Type
Surface Water
Number of
Samples
20,531
Number of
Systems
2,669
Number of
Sample Points
5,945
Pop. Served
by Systems
99,126,085
Pop. Served
by Sample
Points
99,126,085
Combined Ground &
Surface Water
115,403
30,387
50,714
160,012,134
160,012,134
Percent with Detections
Ground Water
0.062%
0.094%
0.058%
4.344%
4.328%
Surface Water
0.107%
0.525%
0.353%
0.901%
0.669%
Combined Ground &
Surface Water
0.070%
0.132%
0.093%
2.211%
2.061%
Exhibit 19-3: Toxaphene Summary Statistics
Source Water Type
Ground Water
Detection Value (ug/L)
Minimum
0.04
Median
1.5
90tn
Percentile
5
Maximum
6.8
Surface Water
0.07
1
2.5
2.5
Combined Ground &
Surface Water
0.04
1
4.9
6.8
The following maps illustrate the geographic distribution of toxaphene occurrence in
drinking water. Exhibit 19-4 shows the states with systems with at least one detection and
Exhibit 19-5 shows the states with systems with at least one detection greater than the MCL
concentration. Note that five states did not submit data for use in the Six-Year Review (Kansas2,
Louisiana, Mississippi, Pennsylvania, and Washington). Oklahoma did submit Six-Year data for
most contaminants; however, there is a statewide waiver for toxaphene in Oklahoma so no
toxaphene data were available from that state.
The distribution of systems with detections of toxaphene is geographically dispersed.
Detection rates were generally low; only three states reported detections in more than 1% of their
Kansas submitted data after the data management and analyses had been conducted; data from Kansas are not
included in the analyses.
19-5
-------
EPA-OGWDW Contaminant Occurrence Support Document for 2" 6-Year Review Category 1 Contaminants October 2009
systems. Eight states reported detections in less than 1% of their systems. Only two states
reported detections greater than the MCL concentration of 3 |ig/L.
Exhibit 19-4: Percentage of Systems with at Least One Toxaphene Detection,
by State
I *
States with Waivers or Not Submitting
States with No Detections
Less than 1% of Systems Detecting
Between 1% and 1% of Systems Detecting
More than 2% of Systems Detecting
19-6
-------
EPA-OGWDW Contaminant Occurrence Support Document for 2" 6-Year Review Category 1 Contaminants October 2009
Exhibit 19-5: Percentage of Systems with at Least One Toxaphene Detection
Greater than the MCL Concentration (> 3 ug/L), by State
States with Waivers or Not Submitting
States with No Detections > MCL
Less than 0.1% of Systems with Detections > MCL
| Between 0.1 °/» and 1% of Systems with Detections > MCL
| More than 1% of Systems with Detections > MCL
Exhibit 19-6 presents the Stage 1 analysis of toxaphene occurrence in drinking water
from systems and the population-served by those systems in the Six-Year Review-ICR Dataset
relative to several thresholds. The thresholds are: 3 |ig/L, 1.5 |ig/L, and 1 ng/L. Nine systems
(0.030% of all systems), serving approximately 285,000 persons, reported at least one detection
greater than or equal to 1.5 ng/L; and five systems (0.016% of all systems), serving about 27,000
persons, reported at least one detection greater than the MCL concentration of 3 |ig/L. All of the
systems reporting at least on detection greater than 3 jig/L were ground water systems.
Exhibit 19-6: Toxaphene Stage 1 Analysis - Systems and Population with at Least
One Threshold Exceedance
Source
Water Type
Ground
Water
Threshold
> 3 ug/L
>1.5|jg/L
> 1 H9/L
N umber of
Systems
Exceeding
Threshold
5
7
7
Percent of
Systems
Exceeding
Threshold
0.018%
0.025%
0.025%
Total Population -
Served by
Systems
Exceeding
Threshold
27,063
27,713
27,713
Percent of
Population -Served
by Systems
Exceeding
Threshold
0.044%
0.046%
0.046%
19-7
-------
EPA-OGWDW Contaminant Occurrence Support Document for 2" 6-Year Review Category 1 Contaminants October 2009
Source
Water Type
Surface
Water
Threshold
> 3 ug/L
>1
> 1 ug/L
Number of
Systems
Exceeding
Threshold
0
2
2
Percent of
Systems
Exceeding
Threshold
0.000%
0.075%
0.075%
Total Population -
Served by
Systems
Exceeding
Threshold
0
257,737
257,737
Percent of
Population -Served
by Systems
Exceeding
Threshold
0.000%
0.260%
0.260%
Combined
Ground &
Surface
Water
> 3 ug/L
:> 1.5 ug/L
> 1 ug/L
5
9
9
0.016%
0.030%
0.030%
27,063
285,450
285,450
0.017%
0.178%
0.178%
The occurrence estimates presented above (in Exhibit 19-6) are based on the conservative
assumption that if a toxaphene detection is found in a single entry point/sample point in a system,
then the entire population-served by the system is assumed to potentially be exposed to
toxaphene (even though other entry points/sampling points at the system, and the population-
served by those entry/sample points, may exhibit no occurrence of toxaphene). The sample point
analysis is a less conservative estimate of the population-served by a system with a detection of
toxaphene because it estimates the potentially exposed population-served based only on the
proportion of the population-served by entry points/sample points with detections of toxaphene
(rather than the entire population-served by the system).
Exhibit 19-7 presents the Stage 1 analysis of toxaphene occurrence in drinking water
from sample points and population-served by those sample points in the Six-Year Review-ICR
Dataset. Eleven (0.022% of) sample points, serving an estimated 283,000 persons, reported at
least one detection greater than or equal to 1.5 u.g/L; and 5 (0.010% of) sample points, serving an
estimated 24,000 persons, reported at least one detection greater than 3 u.g/L.
Exhibit 19-7: Toxaphene Stage 1 Analysis - Sample Points and Population with at
Least One Threshold Exceedance
Source
Water Type
Ground
Water
Threshold
> 3 ug/L
>1
> 1 ug/L
N umber of
Sample Points
Exceeding
Threshold
5
7
7
Percent of
Sample Points
Exceeding
Threshold
0.011%
0.016%
0.016%
Total Population-
Served by Sample
Points Exceeding
Threshold
24,198
24,848
24,848
Percent of
Population-
Served by
Sample Points
Exceeding
Threshold
0.040%
0.041%
0.041%
19-8
-------
EPA-OGWDW Contaminant Occurrence Support Document for 2" 6-Year Review Category 1 Contaminants October 2009
Source
Water Type
Threshold
N umber of
Sample Points
Exceeding
Threshold
Percent of
Sample Points
Exceeding
Threshold
Total Population-
Served by Sample
Points Exceeding
Threshold
Percent of
Population-
Served by
Sample Points
Exceeding
Threshold
Surface
Water
> 3 ug/L
>1
> 1 pg/L
0
4
4
0.000%
0.067%
0.067%
0
257,737
257,737
0.000%
0.260%
0.260%
Combined
Ground &
Surface
Water
> 3 ug/L
>1
> 1 pg/L
5
11
11
0.010%
0.022%
0.022%
24,198
282,585
282,585
0.015%
0.177%
0.177%
19.4.2 Stage 2 Analysis of Occurrence Data
For the Stage 2 analyses, three different substitution values -zero, one-half the MRL
value, and the full MRL value- were used to replace each non-detection record. Three arithmetic
mean toxaphene concentrations were calculated at each system using the zero, one-half MRL and
full MRL substitution values. These mean calculations were done for all systems with data in the
Six-Year Review-ICR dataset. Then the percentages of all systems with a mean concentration
greater than each threshold were calculated.
Exhibit 19-8 presents the Stage 2 analysis of estimated system mean concentrations for
toxaphene occurrence in drinking water relative to several thresholds. The thresholds are: 3 |ig/L,
1.5 |ig/L, and 1 |ig/L. Exhibit 19-9 presents similar information based on population-served by
the systems. Based on the Stage 2 analyses, five ground water systems (approximately 0.016% of
all systems) had an estimated system mean greater than or equal to 1.5 |ig/L. These five systems
serve 23,429 persons. No more than 4 systems (0.013% of all systems), serving 22,989 persons,
had an estimated system mean greater than the MCL concentration of 3 |ig/L.
Exhibit 19-8: Toxaphene Stage 2 Analysis - Systems with a System Mean
Threshold Exceedance
Source
Water Type
Ground
Threshold
> 3 |jg/L
Number of Systems with Mean
Exceeding Threshold1
Full MRL
4
1/2 MRL
3
Zero
3
Percent of Systems with Mean
Exceeding Threshold1
Full MRL
0.014%
1/2 MRL
0.011%
Zero
0.011%
19-9
-------
EPA-OGWDW Contaminant Occurrence Support Document for 2" 6-Year Review Category 1 Contaminants October 2009
>1
> 1 pg/L
5
7
5
5
5
5
0.018%
0.025%
0.018%
0.018%
0.018%
0.018%
Surface
Water
> 3 |jg/L
>1
> 1 pg/L
0
0
2
0
0
0
0
0
0
0.000%
0.000%
0.075%
0.000%
0.000%
0.000%
0.000%
0.000%
0.000%
Combined
Ground &
Surface
Water
> 3 |jg/L
>1
> 1 pg/L
4
5
9
3
5
5
3
5
5
0.013%
0.016%
0.030%
0.010%
0.016%
0.016%
0.010%
0.016%
0.016%
1 The results shown in this table are based on system arithmetic means that were calculated using all sample
detection records and all non-detection records, with the non-detection records replaced (substituted) with
concentration values equal to the full MRL, % MRL, and zero.
Exhibit 19-9: Toxaphene Stage 2 Analysis - Population-Served by Systems with a
System Mean Threshold Exceedance
Source
Water Type
Ground
Water
Threshold
> 3 |jg/L
>1
> 1 pg/L
Population-Served by Systems
with Mean Exceeding Threshold1
Full MRL
22,989
23,429
27,713
1/2 MRL
22,691
23,429
23,429
Zero
22,691
23,429
23,429
Percent of Population-Served by
Systems with Mean Exceeding
Threshold1
Full MRL
0.038%
0.038%
0.046%
1/2 MRL
0.037%
0.038%
0.038%
Zero
0.037%
0.038%
0.038%
Surface
Water
> 3 |jg/L
>1
> 1 pg/L
0
0
257,737
0
0
0
0
0
0
0.000%
0.000%
0.260%
0.000%
0.000%
0.000%
0.000%
0.000%
0.000%
Combined
Ground &
Surface
Water
> 3 pg/L
>1
> 1 pg/L
22,989
23,429
285,450
22,691
23,429
23,429
22,691
23,429
23,429
0.014%
0.015%
0.178%
0.014%
0.015%
0.015%
0.014%
0.015%
0.015%
1. The results shown in this table are based on system arithmetic means that were calculated using all sample
detection records and all non-detection records, with the non-detection records replaced (substituted) with
concentration values equal to the full MRL, 14 MRL, and zero.
19-10
-------
EPA-OGWDW Contaminant Occurrence Support Document for 2" 6-Year Review Category 1 Contaminants October 2009
19.5 Additional Occurrence Data
First Six-Year Review ofNPDWRs
Similar types of Stage 1 contaminant occurrence analyses were conducted for the first
Six-Year Review. The first Six-Year Review assessed regulated contaminant occurrence data
collected from 1993 to 1997. The first Six-Year Review data set consisted of data from 16 states
that were assembled into a "national cross-section" that was qualitatively indicative (though not
statistically representative) of national occurrence. For the first Six-Year Review (USEPA,
2003), 11 systems (approximately 0.08% of about 14,000 systems), reported at least one
detection of toxaphene. One system (approximately 0.01% of all systems), reported at least one
detection greater than the MCL concentration.
19.6 Summary of Data
TRI data indicate that total off-site and on-site releases of toxaphene for the period 1998-
2006 were typically on the order of several thousand pounds. Releases to land constituted almost
all releases, except in 1999, when no such releases were reported.
No data are available from the NAWQA program for toxaphene in ambient waters.
The Stage 1 analysis of occurrence in drinking water indicates that toxaphene was
detected in 0.132% of 30,387 surface water and ground water systems. The median
concentration, based on detections from all systems, was 1 |ig/L. In total, 0.016% of systems
(serving about 27,000 persons) reported at least one detection greater than the MCL
concentration of 3 |ig/L. All of the systems reporting at least one detection greater than the MCL
concentration were ground water systems. A small number of surface water systems reported at
least one detection greater than the other thresholds (1 |ig/L and 1.5 |ig/L).
In the Stage 2 analysis, approximately 0.013% of all systems (serving 0.014% of the
population) had an estimated system mean concentration of toxaphene greater than the MCL
concentration.
19.7 References
Agency for Toxic Substances and Disease Registry (ATSDR). 1996. Toxicological Profile for
Toxaphene. August 1996. Available on the Internet at:
http://www.atsdr.cdc.gov/toxprofiles/tp94.html.
Gilliom, R.J., I.E. Barbash, C.G. Crawford, P.A. Hamilton, J.D. Martin, N. Nakagaki, L.H.
Nowell, J.C. Scott, P.E. Stackelberg, G.P. Thelin, and D.M. Wolock. 2007. The Quality of Our
Nation's Waters - Pesticides in the Nation's Streams and Ground Water, 1992-2001. Appendix
7. Statistical summaries of water-quality data. U.S. Geological Survey Circular 1291, 172 p.
Available on the Internet at: http://water.usgs.gov/nawqa/pnsp/pubs/circ 129l/appendix7/.
19-11
-------
EPA-OGWDW Contaminant Occurrence Support Document for 2" 6-Year Review Category 1 Contaminants October 2009
United State Environmental Protection Agency (USEPA). 1991. National Primary Drinking
Water Regulations—Synthetic Organic Chemicals and Inorganic Chemicals; Monitoring for
Unregulated Contaminants; National Primary Drinking Water Regulations Implementation;
National Secondary Drinking Water Regulations; Final Rule Federal Register. Vol. 56, No. 30. p.
3526, January 30, 1991.
USEPA. 2003. Occurrence Estimation Methodology and Occurrence Findings for Six-Year
Review of National Primary Drinking Water Regulations. Office of Water. EPA-815-R-03-006.
June 2003.
USEPA. 2008. TRI Explorer: Trends - Toxaphene. Released February 21, 2008. Available on
the Internet at: http://www.epa.gov/triexplorer/trends.htm. [Search for toxaphene.] Accessed
February 29, 2008.
USEPA. 2009. The Analysis of Regulated Contaminant Occurrence Data from Public Water
Systems in Support of the Second Six-Year Review of National Primary Drinking Water
Regulations. Office of Water. EPA-815-B-09-006. October 2009.
19-12
-------
EPA-OGWDW Contaminant Occurrence Support Document for 2" 6-Year Review Category 1 Contaminants October 2009
20 l,l?2-Trichloroethane
This chapter on 1,1,2-trichloroethane is part of a report that is organized so that each
chapter presents information on one specific contaminant. Chapter 1 of this report, Introduction,
presents background information on all sources of data used as well as the analytical approach
used to estimate contaminant occurrence in drinking water. This contaminant chapter includes
background information (such as regulatory history, general chemical information, and
environmental fate and behavior), use and environmental release information, and 1,1,2-
trichloroethane occurrence estimates in ambient water and drinking water. All drinking water
occurrence estimates were based on data from the National Compliance Monitoring Information
Collection Request (ICR) Dataset for the Second Six-Year Review (the "Six Year Review-ICR
Dataset"). For detailed information regarding this dataset, data management, data quality
assessments, and the analytical approach used to generate all contaminant occurrence estimates
presented in this chapter, please refer to USEPA (2009).
20.1 Background
The United States Environmental Protection Agency (EPA) published the current
National Primary Drinking Water Regulations (NPDWR) for 1,1,2-trichloroethane on July 17,
1992 (57 FR 31776 (USEPA, 1992)). The NPDWR established a maximum contaminant level
goal (MCLG) of 3 jig/L based on a reference dose (RfD) of 4 jig/kg-day (0.004 mg/kg-day) and a
cancer classification of C, possible human carcinogen. The NPDWR also established a maximum
contaminant level (MCL) of 5 jig/L based on analytical feasibility.
According to the Agency for Toxic Substances and Disease Registry (ATSDR, 1989),
1,1,2-trichloroethane (C^HsCls) is a colorless, sweet-smelling liquid. It does not burn easily and
boils at a higher temperature than water. 1,1,2-Trichloroethane is also known as ethane
trichloride, p-trichloroethane, vinyl trichloride, and 1,2,2-trichloroethane (ATSDR, 1989).
It is difficult to determine precisely how 1,1,2-trichloroethane enters the environment;
however, surveys of industrial wastewater show 1,1,2-trichloroethane in discharges from the
timber products industry, plastics and synthetics industries, and laundries. Studies of 1,1,2-
trichloroethane below the soil surface and in ground water indicate that it may take months to
years to break down. Low levels of 1,1,2-trichloroethane may be found in outdoor air, most likely
related to its use as a solvent. Limited data show that 1,1,2-trichloroethane is present in a quarter
to a half of city air samples. Where 1,1,2-trichloroethane is found, the samples usually contain 10
to 50 parts per trillion (ppt). In the air, 1,1,2-trichloroethane has a half-life of 49 days, providing
enough time for it to spread far and re-deposit before breaking down. It may also be formed in
landfills when 1,1,2,2-tetrachloroethane degrades (ATSDR, 1989).
20.2 Use, Production, and Releases
The principal use of 1,1,2-trichloroethane is as a chemical intermediate in the production
of 1,1-dichloroethylene. Industries that discharge 1,1,2-trichloroethane include the timber
20-1
-------
EPA-OGWDW Contaminant Occurrence Support Document for 2" 6-Year Review Category 1 Contaminants October 2009
products industry, plastics and synthetics industries, and laundries. (ATSDR, 1989). 1,1,2-
Trichloroethane also has limited use as a solvent for fats, oils, waxes, and resins. Information
indicates that 1,1,2-trichloroethane has been sold for use in consumer products, but it is not
known for which products, nor how extensive the use (ATSDR, 1989).
The Toxic Substances Control Act (TSCA) Inventory contains confidential data reported
by industry as a partial update of the Chemical Substance Inventory Update System. The data are
reported under the Inventory Update Reporting regulation. Manufacturers and importers are
required to report company information and chemical information, including production volumes
for chemicals they manufactured or imported in excess of 10,000 pounds in the immediately
preceding fiscal year.
The TSCA Chemical Substance Inventory does not provide actual production volumes;
instead, it indicates at which of five production volume ranges a chemical was manufactured or
imported. The TSCA Inventory is updated every four years. Exhibit 20-1 presents the production
of 1,1,2-trichloroethane in the United States from 1986 to 2002. Note that except for a decrease
for the 1994 data, the production of 1,1,2-trichloroethane in the United States has remained
within the range of 100 million to 500 million since 1986.
Exhibit 20-1: Production of 1,1,2-Trichloroethane in the United States
Inventory Update Year
1986
1990
1994
1998
2002
Range
Range
Range
Range
Range
Production
> 100
> 100
> 10
> 100
> 100
million -
million -
million -
million -
million -
Volume
500 million
500 million
50 million
500 million
500 million
Source: USEPA, 2003a
1,1,2-Trichloroethane is listed as a Toxics Release Inventory (TRI) chemical. Exhibit 20-
2a lists the environmental releases of 1,1,2-trichloroethane between 1998 and 2006. (Exhibit 20-
2b provides a graphical presentation of the environmental releases.) Total releases were highest
in 1998 and declined steadily with time. Air emissions generally constituted over 90% of the
total material released. Land releases were negligible except for sharp increases in 1999, 2000,
and 2001. Surface water discharges and off-site releases contributed only a few percent of the
total, and underground injection was essentially nonexistent. The TRI data for 1,1,2-
trichloroethane were reported from 20 states, with 6 states (Arkansas, Iowa, Kentucky, Louisiana,
Ohio, and Texas) reporting releases in all 9 years.
20-2
-------
EPA-OGWDW Contaminant Occurrence Support Document for 2" 6-Year Review Category 1 Contaminants October 2009
Exhibit 20-2a: Table of Toxics Release Inventory (TRI) Data for 1,1,2-
Trichloroethane
Year
1998
1999
2000
2001
2002
2003
2004
2005
2006
On-Site Releases (pounds)
Total Air
Emissions
280,213
199,121
91,189
84,383
39,046
28,101
20,492
16,268
24,989
Surface Water
Discharges
790
926
572
15
1,256
84
412
239
92
Underground
Injection
5
0
5
0
0
0
0
0
0
Releases to
Land
1
13,788
11,813
554
16
19
0
15
0
Off -site
Releases
(pounds)
2,326
484
7,967
1,274
570
2,423
796
1,062
433
Total On- &
Off-site
Releases
(pounds)
283,335
214,319
111,546
86,226
40,888
30,627
21,701
17,584
25,514
Source: USEPA, 2008
Exhibit 20-2b: Graph of Toxics Release Inventory (TRI) Data for 1,1,2-
Trichloroethane
Pounds Released
Environmental Releases of 1,1,2-Trichloroethane
onn nnn
Ten nnn
9nn nnn
-------
EPA-OGWDW Contaminant Occurrence Support Document for 2" 6-Year Review Category 1 Contaminants October 2009
20.3 Occurrence in Ambient Water
Lakes, rivers, and aquifers are the sources of most drinking water. The United States
Geological Survey (USGS) National Water Quality Assessment (NAWQA) program is a
national-scale assessment of the occurrence of contaminants in ambient surface and ground
water. Contaminant occurrence in ambient water provides information on the potential for
contaminants to adversely affect drinking water supplies. Data on the occurrence of 1,1,2-
trichloroethane in ambient water are available from the NAWQA program.
20.3.1 EPA Summary Analysis of NAWQA Data
EPA, with the cooperation of the USGS, performed a summary analysis of all Cycle 1
(1992-2001) water monitoring data from all study units for 1,1,2-trichloroethane. Detection
frequencies were computed as the percentage of samples and sites with detections (i.e., with at
least one result equal to or greater than the reporting limit). Sample detections can be biased by
frequent sampling in areas with high (or low) occurrence. Calculating the percentage of sites with
detections can reduce this bias.
The results of the EPA analysis are presented in Exhibit 20-3. Overall, 1,1,2-
trichloroethane was detected in ground water at 0.1% of samples and at 0.1% of sites. 1,1,2-
Trichloroethane was never detected in surface water. The median concentration, based on
detections from ground water sites, was 0.0399 |ig/L. The 99th percentile concentration, based on
detections from ground water sites, was 0.0785 |ig/L.
Exhibit 20-3: EPA Summary Analysis of 1,1,2-Trichloroethane Data from NAWQA
Study Units, 1992-2001
ground
water
surface
water
all
sites
Detection Frequency
(detections are results > RL1)
Number
of
Samples
4,544
1,408
5,952
% Samples
with
Detections
0.1%
0.0%
0.1%
Number
of Sites
4,127
190
4,317
% Sites
with
Detections
0.1%
0.0%
0.1%
Concentration Values
(of detections, in ug/L)
Minimum
0.028
0.028
Median
0.0399
0.0399
95th
Percen-
tile
0.0785
0.0785
99th
Percen-
tile
0.0785
0.0785
Maximum
0.0785
0.0785
1. RLs (Reporting Limits) for 1,1,2-trichloroethane varied, but did not exceed 0.06 ug/L.
20.4 Occurrence in Drinking Water
1,1,2-Trichloroethane is regulated as a volatile organic compound (VOC) in drinking
water. All non-purchased community water systems (CWSs) and non-transient non-community
20-4
-------
EPA-OGWDW Contaminant Occurrence Support Document for 2" 6-Year Review Category 1 Contaminants October 2009
water systems (NTNCWSs) are required to sample for VOCs. The maximum waiver period for
VOCs is six years for ground water systems and three years for surface water systems.
All CWSs and NTNCWSs must collect four consecutive quarterly samples during the
initial three-year compliance period.l If all four samples are non-detections, then the system may
reduce to annual sampling. After three annual samples without a detection, a system may be
granted a waiver. If a compound is detected (but at less than the MCL), the system must take one
sample per quarter until results are below the MCL (minimum of two quarterly samples for
ground water systems and four quarterly samples for surface water systems). If all quarterly
samples are below the MCL, the system may return to annual sampling. If a compound is
detected at a level greater than the MCL, the system (whether ground water or surface water)
must take four consecutive quarterly samples until all are below the MCL. If all quarterly
samples are below the MCL, the system may return to annual sampling.
The analysis of 1,1,2-trichloroethane occurrence presented in the following section is
based on state compliance monitoring data from the Six-Year Review-ICR Dataset. This is the
largest and most comprehensive compliance monitoring data set compiled by EPA's Drinking
Water program to date.
All of the Six-Year Review-ICR states provided occurrence data for 1,1,2-
trichloroethane. These data consist of 372,386 analytical results from 50,195 public water
systems (PWSs) during the period from 1998 to 2005. The numbers of sample results and
systems vary by state, although the state data sets have been reviewed and checked to ensure
adequacy of coverage and completeness. The national modal minimum reporting level (MRL) for
1,1,2-trichloroethane in the dataset is 0.5 |ig/L.
20.4.1 Stage 1 Analysis of Occurrence Data
Stage 1 analyses have been performed at the sample, sample point, and system levels.
Note that some systems have multiple sample points, such as source water sample points or entry
points to the distribution system. Exhibit 20-4 presents the number of samples, systems, and
sample points in the data set, as well as the population-served by the systems and sample points,
and the percentage of detections and the percentage of the population-served by the systems and
sample points with detections. Detections in systems provide a fundamental indication of
national occurrence, but are not representative of MCL violations. For 1,1,2-trichloroethane,
0.179% of 50,195 systems reported detections. Exhibit 20-5 lists the minimum, median, 90th
percentile, and maximum 1,1,2-trichloroethane concentrations based on detections from all
systems in the Six-Year Review-ICR Dataset. The median concentration of 1,1,2-trichloroethane,
based on detections from all systems, was 0.6 |ig/L.
1 All new systems or systems using a new water source that began operation after January 22, 2004 must demonstrate
compliance with the MCL within a period of time specified by the State. The system must also comply with the
initial sampling frequencies specified by the State to ensure that a system can demonstrate compliance with the MCL.
20-5
-------
EPA-OGWDW Contaminant Occurrence Support Document for 2" 6-Year Review Category 1 Contaminants October 2009
Exhibit 20-4: 1,1,2-Trichloroethane Occurrence Data from the Six-Year Review-
ICR Dataset
Source Water Type
Number of
Samples
Number of
Systems
Number of
Sample Points
Pop. Served
by Systems
Pop. Served
by Sample
Points
Total Number
Ground Water
288,218
45,942
80,274
86,314,890
86,314,890
Surface Water
84,168
4,253
10,845
140,536,579
140,536,579
Combined Ground &
Surface Water
372,386
50,195
91,119
226,851,469
226,851,469
Percent with Detections
Ground Water
0.033%
0.128%
0.077%
1.016%
0.579%
Surface Water
0.169%
0.729%
0.424%
0.842%
0.268%
Combined Ground &
Surface Water
0.064%
0.179%
0.119%
0.908%
0.386%
Exhibit 20-5: 1,1,2-Trichloroethane Summary Statistics
Source Water Type
Ground Water
Detection Value (ug/L)
Minimum
0.01
Median
0.54
90tn
Percentile
2.3
Maximum
36
Surface Water
0.04
0.7
2.4
40
Combined Ground &
Surface Water
0.01
0.6
2.4
40
The following maps illustrate the geographic distribution of 1,1,2-trichloroethane
occurrence in drinking water. Exhibit 20-6 shows the states with systems with at least one
detection and Exhibit 20-7 shows the states with systems with at least one detection greater than
the MCL concentration for 1,1,2-trichloroethane. Note that five states did not submit data for use
in the Six-Year Review (Kansas2, Louisiana, Mississippi, Pennsylvania, and Washington).
Kansas submitted data after the data management and analyses had been conducted; data from Kansas are not
included in the analyses.
20-6
-------
EPA-OGWDW Contaminant Occurrence Support Document for 2" 6-Year Review Category 1 Contaminants October 2009
The distribution of systems with detections of 1,1,2-trichloroethane is geographically
dispersed. Detection rates were generally low; only two states reported detections in more 1% of
their systems. Only five states reported detections greater than the MCL concentration of 5 |ig/L.
Exhibit 20-6: Percentage of Systems with at Least One 1,1,2-Trichloroethane
Detection, by State
States Not Submitting Any Data
States with No Detections
Less than 0.5% of Sysloms Detecting
| Between 0.5% and 1 % of Systems Detecting
| More than 1% of Systems Detecting
20-7
-------
EPA-OGWDW Contaminant Occurrence Support Document for 2" 6-Year Review Category 1 Contaminants October 2009
Exhibit 20-7: Percentage of Systems with at Least One 1,1,2-Trichloroethane
Detection Greater than the MCL Concentration (> 5 \iglL}, by State
States Mot Submitting Any D*ta
States with No Detection* > MCL
Less than 01% of Systems with detections > MCL
| Between 0.1% and 0.5% of Systems with Detections > MCL
j Mote dian 0.5% of Systems with Defections > MCL
Exhibit 20-8 presents the Stage 1 analysis of 1,1,2-trichloroethane occurrence in drinking
water from systems and the population-served by those systems in the Six-Year Review-ICR
Dataset relative to several thresholds. The thresholds are: 5 |ig/L, 3 |ig/L, and 2.5 |ig/L. Thirteen
systems (0.026% of all systems), serving about 353,000 persons, reported at least one detection
greater than or equal to 2.5 |ig/L; and seven systems (0.014% of all systems), serving about
126,000 persons, reported at least one detection greater than the MCL concentration of 5 |ig/L.
Similar numbers of surface water and ground water systems reported at least one detection
greater than the thresholds. However, because surface water systems tend to be larger, a greater
population was served by surface water systems reporting at least one detection greater than the
thresholds.
20-8
-------
EPA-OGWDW Contaminant Occurrence Support Document for 2" 6-Year Review Category 1 Contaminants October 2009
Exhibit 20-8: 1,1,2-Trichloroethane Stage 1 Analysis - Systems and Population
with at Least One Threshold Exceedance
Source
Water
Type
Ground
Water
Threshold
> 5 |jg/L
>
>
Number of
Systems
Exceeding
Threshold
5
7
8
Percent of
Systems
Exceeding
Threshold
0.011%
0.015%
0.017%
Total Population-
Served by
Systems
Exceeding
Threshold
20,910
31,838
31,882
Percent of
Population -Served
by Systems
Exceeding
Threshold
0.024%
0.037%
0.037%
Surface
Water
> 5 |jg/L
>
>
2
5
5
0.047%
0.118%
0.118%
105,441
320,808
320,808
0.075%
0.228%
0.228%
Combined
Ground &
Surface
Water
> 5 |jg/L
>
>
7
12
13
0.014%
0.024%
0.026%
126,351
352,646
352,690
0.056%
0.155%
0.155%
The occurrence estimates presented above (in Exhibit 20-8) are based on the conservative
assumption that if a 1,1,2-trichloroethane detection is found in a single entry point/sample point
in a system, then the entire population-served by the system is assumed to potentially be exposed
to 1,1,2-trichloroethane (even though other entry points/sampling points at the system, and the
population-served by those entry/sample points, may exhibit no occurrence of 1,1,2-
trichloroethane). The sample point analysis is a less conservative estimate of the population-
served by a system with a detection of 1,1,2-trichloroethane because it estimates the potentially
exposed population-served based only on the proportion of the population-served by entry
points/sample points with detections of 1,1,2-trichloroethane (rather than the entire population-
served by the system).
Exhibit 20-9 presents the Stage 1 analysis of 1,1,2-trichloroethane occurrence in drinking
water from sample points and population-served by those sample points in the Six-Year Review-
ICR Dataset. Nineteen (0.021% of) sample points, serving an estimated 32,000 persons, reported
at least one detection greater than or equal to 2.5 |ig/L; and 11 (0.012% of) sample points,
serving an estimated 15,000 persons, reported at least one detection exceeding 5 |ig/L. As with
the system-level analysis, a larger population was served by surface water systems with
exceedances.
20-9
-------
EPA-OGWDW Contaminant Occurrence Support Document for 2" 6-Year Review Category 1 Contaminants October 2009
Exhibit 20-9: 1,1,2-Trichloroethane Stage 1 Analysis - Sample Points and
Population with at Least One Threshold Exceedance
Source
Water
Type
Ground
Water
Threshold
> 5 |jg/L
>
>
Number of
Sample Points
Exceeding
Threshold
5
7
8
Percent of
Sample Points
Exceeding
Threshold
0.006%
0.009%
0.010%
Total Population-
Served by Sample
Points Exceeding
Threshold
5,774
8,273
8,295
Percent of
Population-
Served by
Sample Points
Exceeding
Threshold
0.007%
0.010%
0.010%
Surface
Water
> 5 pg/L
> 3 |jg/L
>
6
9
11
0.055%
0.083%
0.101%
9,630
19,462
23,292
0.007%
0.014%
0.017%
Combined
Ground &
Surface
Water
> 5 pg/L
>
>
11
16
19
0.012%
0.018%
0.021%
15,404
27,735
31,587
0.007%
0.012%
0.014%
20.4.2 Stage 2 Analysis of Occurrence Data
For the Stage 2 analyses, three different substitution values -zero, one-half the MRL
value, and the full MRL value- were used to replace each non-detection record. Three arithmetic
mean 1,1,2-trichloroethane concentrations were calculated at each system using the zero, one-
half MRL and full MRL substitution values. These mean calculations were done for all systems
with data in the Six-Year Review-ICR dataset. Then the percentages of all systems with a mean
concentration greater than each threshold were calculated.
Exhibit 20-10 presents the Stage 2 analysis of estimated system mean concentrations for
1,1,2-trichloroethane occurrence in drinking water relative to several thresholds. The thresholds
are: 5 |ig/L, 3 |ig/L, and 2.5 |ig/L. Exhibit 20-11 presents similar information based on
population-served by the systems. Based on the Stage 2 analyses, one ground water system
(approximately 0.002% of all systems) had an estimated system mean greater than or equal to 2.5
|ig/L. This system serves about 700 persons. None of the ground water systems had an estimated
system mean greater than the MCL concentration of 5 |ig/L. None of the surface water systems
had a mean greater than or equal to any of the thresholds.
20-10
-------
EPA-OGWDW Contaminant Occurrence Support Document for 2" 6-Year Review Category 1 Contaminants October 2009
Exhibit 20-10: 1,1,2-Trichloroethane Stage 2 Analysis - Systems with a System
Mean Threshold Exceedance
Source Water
Type
Ground
Water
Threshold
> 5 pg/L
>
>
Number of Systems with Mean
Exceeding Threshold1
Full MRL
0
1
1
1/2 MRL
0
1
1
Zero
0
1
1
Percent of Systems with Mean
Exceeding Threshold1
Full MRL
0.000%
0.002%
0.002%
1/2 MRL
0.000%
0.002%
0.002%
Zero
0.000%
0.002%
0.002%
Surface
Water
> 5 pg/L
>
>
0
0
0
0
0
0
0
0
0
0.000%
0.000%
0.000%
0.000%
0.000%
0.000%
0.000%
0.000%
0.000%
Combined
Ground &
Surface
Water
> 5 pg/L
>
>
0
1
1
0
1
1
0
1
1
0.000%
0.002%
0.002%
0.000%
0.002%
0.002%
0.000%
0.002%
0.002%
1. The results shown in this table are based on system arithmetic means that were calculated using all sample
detection records and all non-detection records, with the non-detection records replaced (substituted) with
concentration values equal to the full MRL, 14 MRL, and zero.
Exhibit 20-11: 1,1,2-Trichloroethane Stage 2 Analysis - Population-Served by
Systems with a System Mean Threshold Exceedance
Source Water
Type
Ground
Water
Threshold
> 5 |jg/L
> 3 ug/L
> H9/L
Population-Served by Systems
with Mean Exceeding
Threshold1
Full MRL
0
700
700
1/2 MRL
0
700
700
Zero
0
700
700
Percent of Population-Served by
Systems with Mean Exceeding
Threshold1
Full MRL
0.000%
0.001%
0.001%
1/2 MRL
0.000%
0.001%
0.001%
Zero
0.000%
0.001%
0.001%
Surface
Water
> 5 |jg/L
>
>
0
0
0
0
0
0
0
0
0
0.000%
0.000%
0.000%
0.000%
0.000%
0.000%
0.000%
0.000%
0.000%
Combined
Ground &
Surface
Water
> 5 |jg/L
>
>
0
700
700
0
700
700
0
700
700
0.000%
0.0003%
0.0003%
0.000%
0.0003%
0.0003%
0.000%
0.0003%
0.0003%
1. The results shown in this table are based on system arithmetic means that were calculated using all sample
detection records and all non-detection records, with the non-detection records replaced (substituted) with
concentration values equal to the full MRL, 1A MRL, and zero.
20-11
-------
EPA-OGWDW Contaminant Occurrence Support Document for 2" 6-Year Review Category 1 Contaminants October 2009
20.5 Additional Occurrence Data
Volatile Organic Compounds in the Nation's Ground Water and Drinking Water Supply Wells
The USGS conducted a national assessment of 55 VOCs, including 1,1,2-trichloroethane,
in samples collected from drinking water supply wells, specifically domestic and public wells
(Zogorski et a/., 2006). This occurrence assessment was based on analyses of samples collected
at the well head, and before any treatment or blending, from about 2,400 domestic wells and
about 1,100 public wells.
In domestic wells, 1,1,2-trichloroethane was not detected at an assessment level of 0.2
Hg/L (based on 2,156 samples). The median concentration of detections was 0.028 |ig/L. No
detections were greater than or within one order of magnitude of the MCL concentration (5
In public wells, 1,1,2-trichloroethane was not detected at an assessment level of 0.2 jig/L
(based on 1,078 samples). The median concentration of detections was 0.018 |ig/L. No detections
were greater than, or within one order of magnitude of, the MCL concentration.
First Six-Year Review ofNPDWRs
Similar types of Stage 1 contaminant occurrence analyses were conducted for the first
Six-Year Review. The first Six- Year Review assessed regulated contaminant occurrence data
collected from 1993 to 1997. The first Six-Year Review data set consisted of data from 16 states
that were assembled into a "national cross-section" that was qualitatively indicative (though not
statistically representative) of national occurrence. For the first Six-Year Review (USEPA,
2003b), 138 systems (approximately 0.62% of about 22,000 systems), reported at least one
detection of 1,1,2-trichloroethane. Nine systems (approximately 0.04% of all systems), reported
at least one detection greater than the MCL concentration.
20.6 Summary of Data
TRI data indicate that the total off-site and on-site releases of 1,1,2-trichloroethane ranged
from about 17,600 pounds to approximately 283,000 pounds. Nearly all of the material released
was in the form of air emissions. Surface water and off-site discharges contributed a few percent
to the total. Releases to land and underground injections were minor.
Ambient water data from the NAWQA program indicate that 1,1,2-trichloroethane was
detected in 0. 1% of 5,952 ground and surface water samples and at 0. 1% of 4,3 17 sites. The
median concentration, based on detections from all sites, was 0.04 |ig/L.
The Stage 1 analysis of occurrence in drinking water indicates that 1,1,2-trichloroethane
was detected in 0.179% of 50,195 surface water and ground water systems. The median
concentration, based on detections from all systems, was 0.6 |ig/L. In total, 0.014% of systems
(serving 0.056% of the population) reported at least one detection greater than the MCL
20-12
-------
EPA-OGWDW Contaminant Occurrence Support Document for 2" 6-Year Review Category 1 Contaminants October 2009
concentration of 5 |ig/L. 1,1,2-Trichloroethane was detected in similar numbers of ground water
and surface water systems. However, because surface water systems are generally larger, a
greater population was served by surface water systems at least one detection greater than the
MCL concentration.
In the Stage 2 analysis, no systems had an estimated mean concentration of 1,1,2-
trichloroethane greater than MCL concentration.
20.7 References
Agency for Toxic Substances and Disease Registry (ATSDR). 1989. Toxicological Profile for
1,1,2-Trichloroethane. December 1989. Available on the Internet at:
http ://www. atsdr.cdc. gov/toxprofiles/tp 148 .html.
United States Environmental Protection Agency (USEPA). 1992. Drinking Water; National
Primary Drinking Water Regulations—Synthetic Organic Chemicals and Inorganic Chemicals;
National Primary Drinking Water Regulations Implementation; Final Rule. Federal Register.
Vol. 57, No. 138. p. 31776, July 17, 1992.
USEPA. 2003a. Non-confidential Inventory Update Reporting (IUR) Production Volume
Information - 1,1,2-Trichloroethane. Last modified September 10, 2007. Available on the
Internet at: http://www.epa.gov/oppt/iur/tools/data/2002-vol.htm. [Search for 1,1,2-
trichloroethane.] Accessed February 21, 2008.
USEPA. 2003b. Occurrence Estimation Methodology and Occurrence Findings for Six-Year
Review of National Primary Drinking Water Regulations. Office of Water. EPA-815-R-03-006.
June 2003.
USEPA. 2008. TRI Explorer: Trends - 1,1,2-Trichloroethane. Released February 21, 2008.
Available on the Internet at: http://www.epa.gov/triexplorer/trends.htm. [Search for 1,1,2-
trichloroethane.] Accessed February 29, 2008.
USEPA. 2009. The Analysis of Regulated Contaminant Occurrence Data from Public Water
Systems in Support of the Second Six-Year Review of National Primary Drinking Water
Regulations. Office of Water. EPA-815-B-09-006. October 2009.
Zogorski, J.S., J.M. Carter, T. Ivahnenko, W.W. Lapham, M.J. Moran, B.L. Rowe, P.J. Squillace,
and P.L. Toccalino. 2006. Volatile organic compounds in the nation's ground water and drinking
water supply wells. U.S. Geological Survey Circular 1292, 101 p. Available on the Internet at:
http://pubs.er.usgs.gov/pubs/cir/cirl292.
20-13
-------
EPA-OGWDW Contaminant Occurrence Support Document for 2" 6-Year Review Category 1 Contaminants October 2009
21 Trichloroethylene (TCE)
This chapter on trichloroethylene (TCE) is part of a report that is organized so that each
chapter presents information on one specific contaminant. Chapter 1 of this report, Introduction,
presents background information on all sources of data used as well as the analytical approach
used to estimate contaminant occurrence in drinking water. This contaminant chapter includes
background information (such as regulatory history, general chemical information, and
environmental fate and behavior), use and environmental release information, and
trichloroethylene occurrence estimates in ambient water and drinking water. All drinking water
occurrence estimates were based on data from the National Compliance Monitoring Information
Collection Request (ICR) Dataset for the Second Six-Year Review (the "Six Year Review-ICR
Dataset"). For detailed information regarding this dataset, data management, data quality
assessments, and the analytical approach used to generate all contaminant occurrence estimates
presented in this chapter, please refer to USEPA (2009).
21.1 Background
The United States Environmental Protection Agency (EPA) published the current
National Primary Drinking Water Regulations (NPDWR) for trichloroethylene on July 8, 1987
(52 FR 25690 (USEPA, 1987)). The NPDWR established a maximum contaminant level goal
(MCLG) of zero based on a cancer classification of B2, probable human carcinogen. The
NPDWR also established a maximum contaminant level (MCL) of 5 |ig/L based on analytical
feasibility.
According to the Agency for Toxic Substances and Disease Registry (ATSDR, 1997),
trichloroethylene ^HCb) is a nonflammable, colorless liquid at room temperature that has a
somewhat sweet odor, and a sweet, burning taste. Other names for trichloroethylene include TCE
and trichloroethene (ATSDR, 1997).
By far, the largest source of trichloroethylene in the environment is evaporation from
factories that use it to remove grease from metals. In the air, trichloroethylene has a half-life of
approximately one week, breaking down into a lung irritant called phosgene. Chemical waste
sites may leak trichloroethylene to surface water, which slows break down to days or weeks. In
ground water and soils, degradation is much slower than that (ATSDR, 1997).
Exposure to trichloroethylene may occur through contact with household products such as
typewriter correction fluids, paints and paint removers, glues, spot removers, rug cleaning fluids,
and metal cleaners. Factory workers, those living near factories, and those living near hazardous
waste sites may also be exposed. Trichloroethylene is found in the air, but typically at
concentrations much lower than 1 ppm. Trichloroethylene in foods is believed to come from
contamination of the water used in food processing or from food processing equipment cleaned
with trichloroethylene (ATSDR, 1997).
21-1
-------
EPA-OGWDW Contaminant Occurrence Support Document for 2" 6-Year Review Category 1 Contaminants October 2009
21.2 Use, Production, and Releases
According to ATSDR (1997), the primary use of trichloroethylene in the United States is
as a vapor degreasing agent for fabricated metal parts. It is an excellent extraction solvent for
greases, oils, fats, waxes, and tars and is used by the textile processing industry to scour cotton,
wool, and other fabrics. As a general solvent, or as a component of solvent blends,
trichloroethylene is used with adhesives, lubricants, paints, varnishes, paint strippers, pesticides,
and cold metal cleaners. Before 1977, it was used as a surgical disinfectant, as a pet food
additive, and as an agent to decaffeinate coffee; however, these uses were banned by the Food
and Drug Administration (ATSDR, 1997).
The Toxic Substances Control Act (TSCA) Inventory contains confidential data reported
by industry as a partial update of the Chemical Substance Inventory Update System. The data are
reported under the Inventory Update Reporting regulation. Manufacturers and importers are
required to report company information and chemical information, including production volumes
for chemicals they manufactured or imported in excess of 10,000 pounds in the immediately
preceding fiscal year.
The TSCA Chemical Substance Inventory does not provide actual production volumes;
instead, it indicates at which of five production volume ranges a chemical was manufactured or
imported. The TSCA Inventory is updated every four years. Exhibit 21-1 presents the production
of trichloroethylene in the United States from 1986 to 2002. Aside from a decrease in the 1994
data, the production of trichloroethylene in the United States has remained within the range of
100 million to 500 million since 1986.
Exhibit 21-1: Production of Trichloroethylene in the United States
Inventory Update Year
1986 Range
1990 Range
1994 Range
1998 Range
2002 Range
Production
> 100 million -
> 100 million -
> 50 million -
> 100 million -
> 100 million -
Volume
500 million
500 million
100 million
500 million
500 million
Source: USEPA, 2003a
Trichloroethylene is listed as a Toxics Release Inventory (TRI) chemical. Exhibit 21-2a
lists the environmental releases of trichloroethylene between 1998 and 2006. (Exhibit 21-2b
provides a graphical presentation of the environmental releases.) Trichloroethylene releases
declined steadily during the period of review. Air emissions constituted more than 95% of all
releases. Off-site releases were generally the second largest fraction and surface water discharges
21-2
-------
EPA-OGWDW Contaminant Occurrence Support Document for 2" 6-Year Review Category 1 Contaminants October 2009
generally contributed the smallest amounts. The TRI data for trichloroethylene were reported
from 45 states, with 40 states reporting releases in all 9 years. No releases were reported from
Alaska, Hawaii, Idaho, North Dakota or West Virginia.
Exhibit 21-2a: Table of Toxics Release Inventory (TRI) Data for Trichloroethylene
Year
1998
1999
2000
2001
2002
2003
2004
2005
2006
On-Site Releases (pounds)
Total Air
Emissions
13,274,099
10,645,692
9,861,009
8,421,733
8,197,393
7,084,772
6,181,340
5,535,452
5,612,675
Surface Water
Discharges
882
1,034
593
406
579
595
216
533
482
Underground
Injection
593
0
47,877
98,220
140,190
90,971
123,637
86,817
0
Releases to
Land
800
148,867
9,607
12,609
230
150,642
2
4,711
77,339
Off-site
Releases
(pounds)
136,766
192,385
171,952
133,531
139,398
66,894
71,780
60,074
294,071
Total On- &
Off -site
Releases
(pounds)
13,413,140
10,987,978
10,091,038
8,666,499
8,477,790
7,393,873
6,376,975
5,687,587
5,984,567
Source: USEPA, 2008
Exhibit 21-2b: Graph of Toxics Release Inventory (TRI) Data for Trichloroethylene
Pounds Released
Environmental Releases of Trichloroethylene
-------
EPA-OGWDW Contaminant Occurrence Support Document for 2" 6-Year Review Category 1 Contaminants October 2009
21.3 Occurrence in Ambient Water
Lakes, rivers, and aquifers are the sources of most drinking water. The United States
Geological Survey (USGS) National Water Quality Assessment (NAWQA) program is a
national-scale assessment of the occurrence of contaminants in ambient surface and ground
water. Contaminant occurrence in ambient water provides information on the potential for
contaminants to adversely affect drinking water supplies. Data on the occurrence of
trichloroethylene in ambient water are available from the NAWQA program.
21.3.1 EPA Summary Analysis of NAWQA Data
EPA, with the cooperation of the USGS, performed a summary analysis of all Cycle 1
(1992-2001) water monitoring data from all study units for trichloroethylene. Detection
frequencies were computed as the percentage of samples and sites with detections (i.e., with at
least one result equal to or greater than the reporting limit). Sample detections can be biased by
frequent sampling in areas with high (or low) occurrence. Calculating the percentage of sites with
detections can reduce this bias.
The results of the EPA analysis are presented in Exhibit 21-3. Overall, trichloroethylene
was detected in 12.8% of samples and at 6.2% of sites. Trichloroethylene was detected more
frequently in surface water than in ground water. The median concentration, based on detections
from all sites, was 0.058 |ig/L. The 99th percentile concentration, based on detections from all
sites, was 17.33 |ig/L.
Exhibit 21-3: EPA Summary Analysis of Trichloroethylene Data from NAWQA
Study Units, 1992-2001
ground
water
surface
water
all
sites
Detection Frequency
(detections are results > RL1)
Number
of
Samples
4,655
1,413
6,068
% Samples
with
Detections
5.4%
37.2%
12.8%
Number
of Sites
4,159
190
4,349
% Sites
with
Detections
4.6%
41.6%
6.2%
Concentration Values
(of detections, in ug/L)
Minimum
0.002
0.003
0.002
Median
0.177
0.0409
0.058
95th
Percen-
tile
7.9
1.1
2.2
99th
Percen-
tile
103.5
2.2
17.33
Maximum
230
17.33
230
1. RLs (Reporting Limits) for trichloroethylene varied, but did not exceed 0.003 ug/L.
21.4 Occurrence in Drinking Water
Trichloroethylene is regulated as a volatile organic compound (VOC) in drinking water.
All non-purchased community water systems (CWSs) and non-transient non-community water
systems (NTNCWSs) are required to sample for VOCs. The maximum waiver period for VOCs
is six years for ground water systems and three years for surface water systems.
21-4
-------
EPA-OGWDW Contaminant Occurrence Support Document for 2" 6-Year Review Category 1 Contaminants October 2009
All CWSs and NTNCWSs must collect four consecutive quarterly samples during the
initial three-year compliance period.l If all four samples are non-detections, then the system may
reduce to annual sampling. After three annual samples without a detection, a system may be
granted a waiver. If a compound is detected (but at less than the MCL), the system must take one
sample per quarter until results are below the MCL (minimum of two quarterly samples for
ground water systems and four quarterly samples for surface water systems). If all quarterly
samples are below the MCL, the system may return to annual sampling. If a compound is
detected at a level greater than the MCL, the system (whether ground water or surface water)
must take four consecutive quarterly samples until all are below the MCL. If all quarterly
samples are below the MCL, the system may return to annual sampling.
The analysis of trichloroethylene occurrence presented in the following section is based
on state compliance monitoring data from the Six-Year Review-ICR Dataset. This is the largest
and most comprehensive compliance monitoring data set compiled by EPA's Drinking Water
program to date.
All of the Six-Year Review-ICR states provided occurrence data for trichloroethylene.
These data consist of 403,609 analytical results from 50,432 public water systems (PWSs) during
the period from 1998 to 2005. The number of sample results and systems vary by state, although
the state data sets have been reviewed and checked to ensure adequacy of coverage and
completeness. The national modal minimum reporting level (MRL) for trichloroethylene in the
datasetis 0.5 |ig/L.
21.4.1 Stage 1 Analysis of Occurrence Data
Stage 1 analyses have been performed at the sample, sample point, and system levels.
Note that some systems have multiple sample points, such as source water sample points or entry
points to the distribution system. Exhibit 21-4 presents the number of samples, systems, and
sample points in the data set, as well as the population-served by the systems and sample points,
and the percentage of detections and the percentage of the population-served by the systems and
sample points with detections. Detections in systems provide a fundamental indication of
national occurrence, but are not representative of MCL violations. For trichloroethylene, 2.009%
of 50,432 systems reported detections. Exhibit 21-5 lists the minimum, median, 90th percentile,
and maximum trichloroethylene concentrations based on detections from all systems in the Six-
Year Review-ICR Dataset. The median concentration of trichloroethylene, based on detections
from all systems, was 1.5 |ig/L.
1 All new systems or systems using a new water source that began operation after January 22, 2004 must demonstrate
compliance with the MCL within a period of time specified by the State. The system must also comply with the
initial sampling frequencies specified by the State to ensure that a system can demonstrate compliance with the MCL.
21-5
-------
EPA-OGWDW Contaminant Occurrence Support Document for 2" 6-Year Review Category 1 Contaminants October 2009
Exhibit 21-4: Trichloroethylene Occurrence Data from the Six-Year Review-ICR
Dataset
Source Water Type
Number of
Samples
Number of
Systems
Number of
Sample Points
Pop. Served
by Systems
Pop. Served
by Sample
Points
Total Number
Ground Water
304,776
46,177
80,631
86,393,209
86,393,209
Surface Water
98,833
4,255
10,904
140,514,477
140,514,477
Combined Ground &
Surface Water
403,609
50,432
91,535
226,907,686
226,907,686
Percent with Detections
Ground Water
4.855%
1.867%
1.969%
12.214%
4.333%
Surface Water
14.934%
3.549%
5.805%
13.512%
4.748%
Combined Ground &
Surface Water
7.323%
2.009%
2.426%
13.018%
4.590%
Exhibit 21-5: Trichloroethylene Summary Statistics
Source Water Type
Ground Water
Detection Value (ug/L)
Minimum
0.0018
Median
1.3
90tn
Percentile
7.5
Maximum
159
Surface Water
0.001
1.8
14.2
50
Combined Ground &
Surface Water
0.001
1.5
10.5
159
The following maps illustrate the geographic distribution of trichloroethylene occurrence
in drinking water. Exhibit 21-6 shows the states with systems with at least one detection and
Exhibit 21-7 shows the states with systems with at least one detection greater than the MCL
concentration. Note that five states did not submit data for use in the Six-Year Review (Kansas2,
Louisiana, Mississippi, Pennsylvania, and Washington).
The distribution of systems with detections of trichloroethylene is geographically
dispersed. Detection rates were high; systems in all states except Hawaii reported detections of
2 Kansas submitted data after the data management and analyses had been conducted; data from Kansas are not
included in the analyses.
21-6
-------
EPA-OGWDW Contaminant Occurrence Support Document for 2" 6-Year Review Category 1 Contaminants October 2009
trichloroethylene, although only 6 states reported detections in more 5% of their systems. Thirty-
two states reported at least one detection greater than the MCL concentration of 5 |ig/L, although
only 8 of these states had more than 0.5% of systems with at least one detection greater than 5
Exhibit 21-6: Percentage of Systems with at Least One Trichloroethylene
Detection, by State
=
States Not Submitting Any Data
States with No Detections
L*s* than 1% of Systems Detecting
Between 1% and 5% of Systems Detecting
More than 5% of Systems Detecting
21-7
-------
EPA-OGWDW Contaminant Occurrence Support Document for 2" 6-Year Review Category 1 Contaminants October 2009
Exhibit 21-7: Percentage of Systems with at Least One Trichloroethylene
Detection Greater than the MCL Concentration (> 5 pg/L), by State
\
: States Not Submitting Any Data
States with No Detection* > MCL
Less than 0,1% of Systems with Detection* > MCL
M Between 0.1% and 0.5% of Systems with Detections > MCL
^B More Uian 0.5% of Systems with Detections > MCL
Exhibit 21-8 presents the Stage 1 analysis of trichloroethylene occurrence in drinking
water from systems and the population-served by those systems in the Six-Year Review-ICR
Dataset relative to several thresholds. The thresholds are: 5 |ig/L, 2.5 |ig/L, 1 |ig/L, and 0.5 |ig/L.
A total of 338 systems (0.670% of all systems), serving almost 16 million persons, reported at
least one detection greater than or equal to 2.5 |ig/L; and 195 systems (0.387% of all systems),
serving about 12 million persons, reported at least one detection greater than 5 |ig/L.
Trichloroethylene was detected in more ground water systems than surface water systems at
levels greater than all thresholds. However, because surface water systems tend to be larger, a
greater population was served by surface water systems reporting at least one detection greater
than all thresholds.
21-8
-------
EPA-OGWDW Contaminant Occurrence Support Document for 2" 6-Year Review Category 1 Contaminants October 2009
Exhibit 21-8: Trichloroethylene Stage 1 Analysis - Systems and Population with
at Least One Threshold Exceedance
Source
Water Type
Ground
Water
Threshold
> 5 |jg/L
>2.5|jg/L
> 1 H9/L
> 0.5 |jg/L
Number of
Systems
Exceeding
Threshold
157
283
490
682
Percent of
Systems
Exceeding
Threshold
0.340%
0.613%
1.061%
1.477%
Total Population-
Served by Systems
Exceeding
Threshold
2,241,272
4,817,431
7,008,143
8,714,465
Percent of
Population-
Served by
Systems
Exceeding
Threshold
2.594%
5.576%
8.112%
10.087%
Surface
Water
> 5 pg/L
>2.5|jg/L
> 1 H9/L
> 0.5 |jg/L
38
55
93
122
0.893%
1.293%
2.186%
2.867%
9,703,812
11,131,487
16,021,388
17,384,229
6.906%
7.922%
1 1 .402%
12.372%
Combined
Ground &
Surface
Water
> 5 pg/L
>2.5|jg/L
> 1 |jg/L
> 0.5 |jg/L
195
338
583
804
0.387%
0.670%
1.156%
1.594%
11,945,084
15,948,918
23,029,531
26,098,694
5.264%
7.029%
10.149%
11.502%
The occurrence estimates presented above (in Exhibit 21-8) are based on the conservative
assumption that if a trichloroethylene detection is found in a single entry point/sample point in a
system, then the entire population-served by the system is assumed to potentially be exposed to
trichloroethylene (even though other entry points/sampling points at the system, and the
population-served by those entry/sample points, may exhibit no occurrence of trichloroethylene).
The sample point analysis is a less conservative estimate of the population-served by a system
with a detection of trichloroethylene because it estimates the potentially exposed population-
served based only on the proportion of the population-served by entry points/sample points with
detections of trichloroethylene (rather than the entire population-served by the system).
Exhibit 21-9 presents the Stage 1 analysis of trichloroethylene occurrence in drinking
water from sample points and population-served by those sample points in the Six-Year Review-
ICR Dataset. A total of 769 (0.840% of) sample points, serving an estimated 4.5 million persons,
reported at least one detection greater than or equal to 2.5 |ig/L; and 433 (0.473% of) sample
points, serving an estimated 3.3 million persons, reported at least one detection exceeding 5
|ig/L. Similar to the system-level occurrence analysis, thresholds were exceeded in more ground
21-9
-------
EPA-OGWDW Contaminant Occurrence Support Document for 2" 6-Year Review Category 1 Contaminants October 2009
water than surface water sample points. However, because surface water systems tend to be
larger, a greater population was served by sample points with threshold exceedances.
Exhibit 21-9: Trichloroethylene Stage 1 Analysis - Sample Points and Population
with at Least One Threshold Exceedance
Source
Water Type
Ground
Water
Threshold
> 5 |jg/L
>2.5|jg/L
> 1 H9/L
> 0.5 |jg/L
Number of
Sample Points
Exceeding
Threshold
245
478
910
1,301
Percent of
Sample Points
Exceeding
Threshold
0.304%
0.593%
1.129%
1.614%
Total Population-
Served by Sample
Points Exceeding
Threshold
436,582
855,067
1,987,709
2,800,442
Percent of
Population-
Served by
Sample Points
Exceeding
Threshold
0.505%
0.990%
2.301%
3.242%
Surface
Water
> 5 |jg/L
>2.5|jg/L
> 1 |jg/L
> 0.5 |jg/L
188
291
459
582
1.724%
2.669%
4.209%
5.337%
2,858,740
3,692,244
5,214,082
6,059,333
2.034%
2.628%
3.711%
4.312%
Combined
Ground &
Surface
Water
> 5 pg/L
>2.5|jg/L
> 1 |jg/L
> 0.5 |jg/L
433
769
1,369
1,883
0.473%
0.840%
1 .496%
2.057%
3,295,321
4,547,311
7,201,791
8,859,775
1.452%
2.004%
3.174%
3.905%
21.4.2 Stage 2 Analysis of Occurrence Data
For the Stage 2 analyses, three different substitution values -zero, one-half the MRL
value, and the full MRL value- were used to replace each non-detection record. Three arithmetic
mean trichloroethylene concentrations were calculated at each system using the zero, one-half
MRL and full MRL substitution values. These mean calculations were done for all systems with
data in the Six-Year Review-ICR dataset. Then the percentages of all systems with a mean
concentration greater than each threshold were calculated.
Exhibit 21-10 presents the Stage 2 analysis of estimated system mean concentrations for
trichloroethylene occurrence in drinking water relative to several thresholds. The thresholds are:
5 |ig/L, 2.5 |ig/L, 1 |ig/L, and 0.5 |ig/L. Exhibit 21-11 presents similar information based on
population-served by the systems. Based on the Stage 2 analyses, no more than 70 ground water
21-10
-------
EPA-OGWDW Contaminant Occurrence Support Document for 2" 6-Year Review Category 1 Contaminants October 2009
and surface water systems (approximately 0.139% of all systems) had an estimated system mean
greater than or equal to 2.5 |ig/L. These 70 systems serve almost 4.8 million persons. Twenty-
five systems (0.050% of all systems), serving about 410,000 persons, had an estimated system
mean greater than 5 |ig/L. More ground water systems than surface water systems had an
estimated system mean greater than the MCL concentration, although a larger population was
served by surface water systems with means exceeding the MCL concentration.
Exhibit 21-10: Trichloroethylene Stage 2 Analysis - Systems with a System Mean
Threshold Exceedance
Source
Water Type
Ground
Water
Threshold
> 5 |jg/L
>2.5|jg/L
> 1 |jg/L
> 0.5 |jg/L
Number of Systems with Mean
Exceeding Threshold1
Full MRL
18
60
209
624
1/2 MRL
18
58
181
334
Zero
18
54
160
267
Percent of Systems with Mean
Exceeding Threshold1
Full MRL
0.039%
0.130%
0.453%
1.351%
1/2 MRL
0.039%
0.126%
0.392%
0.723%
Zero
0.039%
0.117%
0.346%
0.578%
Surface
Water
> 5 |jg/L
>2.5|jg/L
> 1 |jg/L
> 0.5 |jg/L
7
10
31
115
7
10
27
54
7
10
22
43
0.165%
0.235%
0.729%
2.703%
0.165%
0.235%
0.635%
1.269%
0.165%
0.235%
0.517%
1.011%
Combined
Ground &
Surface
Water
> 5 |jg/L
>2.5|jg/L
> 1 |jg/L
> 0.5 |jg/L
25
70
240
739
25
68
208
388
25
64
182
310
0.050%
0.139%
0.476%
1 .465%
0.050%
0.135%
0.412%
0.769%
0.050%
0.127%
0.361%
0.615%
1. The results shown in this table are based on system arithmetic means that were calculated using all sample
detection records and all non-detection records, with the non-detection records replaced (substituted) with
concentration values equal to the full MRL, 14 MRL, and zero.
Exhibit 21-11: Trichloroethylene Stage 2 Analysis - Population-Served by
Systems with a System Mean Threshold Exceedance
Source
Water Type
Ground
Water
Threshold
> 5 pg/L
>2.5|jg/L
> 1 H9/L
> 0.5 |jg/L
Population -Served by Systems with
Mean Exceeding Threshold1
Full MRL
50,410
307,930
1,911,987
8,000,226
1/2 MRL
50,410
234,229
1,686,897
2,929,896
Zero
50,410
140,973
1,347,211
2,411,718
Percent of Population-Served by
Systems with Mean Exceeding
Threshold1
Full MRL
0.058%
0.356%
2.213%
9.260%
1/2 MRL
0.058%
0.271%
1.953%
3.391%
Zero
0.058%
0.163%
1.559%
2.792%
21-11
-------
EPA-OGWDW Contaminant Occurrence Support Document for 2" 6-Year Review Category 1 Contaminants October 2009
Source
Water Type
Surface
Water
Threshold
> 5 |jg/L
>2.5|jg/L
> 1 H9/L
> 0.5 |jg/L
Population -Served by Systems with
Mean Exceeding Threshold1
Full MRL
359,713
4,456,573
8,458,791
16,251,310
1/2 MRL
359,713
4,456,573
6,594,945
9,936,227
Zero
359,713
4,456,573
6,051,956
9,472,157
Percent of Population-Served by
Systems with Mean Exceeding
Threshold1
Full MRL
0.256%
3.172%
6.020%
11.566%
1/2 MRL
0.256%
3.172%
4.693%
7.071%
Zero
0.256%
3.172%
4.307%
6.741%
Combined
Ground &
Surface
Water
> 5 |jg/L
>2.5|jg/L
> 1 H9/L
> 0.5 |jg/L
410,123
4,764,503
10,370,778
24,251,536
410,123
4,690,802
8,281,842
12,866,123
410,123
4,597,546
7,399,167
11,883,875
0.181%
2.100%
4.570%
10.688%
0.181%
2.067%
3.650%
5.670%
0.181%
2.026%
3.261%
5.237%
1 The results shown in this table are based on system arithmetic means that were calculated using all sample
detection records and all non-detection records, with the non-detection records replaced (substituted) with
concentration values equal to the full MRL, 1A MRL, and zero.
21.5 Additional Occurrence Data
Volatile Organic Compounds in the Nation's Ground Water and Drinking Water Supply Wells
The USGS conducted a national assessment of 55 VOCs, including trichloroethylene, in
samples collected from drinking water supply wells, specifically domestic and public wells
(Zogorski et a/., 2006). This occurrence assessment was based on analyses of samples collected
at the well head, and before any treatment or blending, from about 2,400 domestic wells and
about 1,100 public wells.
In domestic wells, trichloroethylene had a detection frequency of 0.92% at an assessment
level of 0.2 jig/L (based on 2,400 samples). The median concentration of detections was 0.14
|ig/L. Nine detections were within one order of magnitude of the MCL concentration (5 |ig/L)
and six detections were greater than the MCL concentration.
In public wells, trichloroethylene had a detection frequency of 4.3% at an assessment
level of 0.2 |ig/L (based on 1,093 samples). The median concentration of detections was 0.52
|ig/L. Twenty-seven detections were within one order of magnitude of the MCL concentration
and nine detections were greater than the MCL concentration.
Occurrence of Water-Quality Constituents in Domestic Wells across the United States
The USGS, in collaboration with the Centers for Disease Control and Prevention (CDC),
released an online report on the occurrence of 11 priority water-quality constituents of possible
health concern in domestic wells located in 16 states across the United States (Bartholomay et
a/., 2007). Two water-quality data sets were used to define the water quality of domestic-water
supplies: (1) data from the NAWQA Program, and (2) USGS state data. The common assessment
level used to quantify the detection frequency of trichloroethylene was 0.2 |ig/L.
21-12
-------
EPA-OGWDW Contaminant Occurrence Support Document for 2" 6-Year Review Category 1 Contaminants October 2009
A total of 1,041 wells were sampled for trichloroethylene (referred to as trichloroethene
(TCE) in the study). The detection frequency for trichloroethylene was 0.8%. Less than 0.5% of
samples had concentrations of trichloroethylene less than, but within one order of magnitude of,
the MCL concentration (5 |ig/L); 0.1% of samples had concentrations greater than the MCL
concentration.
First Six-Year Review ofNPDWRs
Similar types of Stage 1 contaminant occurrence analyses were conducted for the first
Six-Year Review. The first Six-Year Review assessed regulated contaminant occurrence data
collected from 1993 to 1997. The first Six-Year Review data set consisted of data from 16 states
that were assembled into a "national cross-section" that was qualitatively indicative (though not
statistically representative) of national occurrence. For the first Six-Year Review (USEPA,
2003b), 602 systems (approximately 2.61% of about 23,000 systems), reported at least one
detection of trichloroethylene. One hundred and fifty systems (approximately 0.65% of all
systems), reported at least one detection greater than the MCL concentration.
21.6 Summary of Data
TRI data indicate that the total off-site and on-site releases of trichloroethylene for the
period 1998-2006 ranged from approximately 5.7 million pounds to about 13.4 million pounds.
The maximum releases of trichloroethylene were in 1998, and amounts declined steadily with
time. Air emissions constituted over 95% of the total releases.
Ambient water data from the NAWQA program indicate that trichloroethylene was
detected in 12.8% of 6,068 ground and surface water samples and at 6.2% of 4,349 sites. The
median concentration, based on detections from all sites, was 0.06 |ig/L.
The Stage 1 analysis of occurrence in drinking water indicates that trichloroethylene was
detected in 2.009% of 50,432 surface water and ground water systems. The median
concentration, based on detections from all systems, was 1.5 |ig/L. In total, 0.387% of systems
(serving 5.264% of the population) reported at least one detection greater than the MCL
concentration of 5 |ig/L. Detections greater than the MCL concentration were more common in
ground water than surface water systems. However, because surface water systems tend to be
larger, a greater population was served by surface water systems reporting at least one detection
greater than the MCL concentration. A similar pattern was seen for sample points.
In the Stage 2 analysis, 0.050% of all systems (serving 0.181% of the population) had an
estimated system mean concentration of trichloroethylene greater than the MCL concentration.
The number of systems with threshold exceedances was greater for ground water systems, but
larger population were served by surface water systems with mean exceedances.
21-13
-------
EPA-OGWDW Contaminant Occurrence Support Document for 2" 6-Year Review Category 1 Contaminants October 2009
21.7 References
Agency for Toxic Substances and Disease Registry (ATSDR). 1997. Toxicological Profile for
Trichloroethylene. September 1997. Available on the Internet at:
http://www.atsdr.cdc.gov/toxprofiles/tpl9.html.
Bartholomay, R.C., J.M. Carter, S.L. Qi, PJ. Squillace, and G.L. Rowe. 2007. Summary of
selected U.S. Geological Survey data on domestic well water quality for the Centers for Disease
Control's National Environmental Public Health Tracking Program: U.S. Geological Survey
Scientific Investigations Report 2007-5213, 57 p. Available on the Internet at:
http://pubs.usgs.gov/sir/2007/5213/.
United States Environmental Protection Agency (USEPA). 1987. National Primary Drinking
Water Regulations—Synthetic Organic Chemicals; Monitoring for Unregulated Contaminants;
Final Rule. Federal Register. Vol. 52, No. 130. p. 25690, July 8, 1987.
USEPA. 2003a. Non-confidential Inventory Update Reporting (IUR) Production Volume
Information - Trichloroethylene. Last modified September 10, 2007. Available on the Internet at:
http://www.epa.gov/oppt/iur/tools/data/2002-vol.htm. [Search for trichloroethylene.] Accessed
February 21, 2008.
USEPA. 2003b. Occurrence Estimation Methodology and Occurrence Findings for Six-Year
Review of National Primary Drinking Water Regulations. Office of Water. EPA-815-R-03-006.
June 2003.
USEPA. 2008. TRI Explorer: Trends - Trichloroethylene. Released February 21, 2008. Available
on the Internet at: http://www.epa.gov/triexplorer/trends.htm. [Search for trichloroethylene.]
Accessed February 29, 2008.
USEPA. 2009. The Analysis of Regulated Contaminant Occurrence Data from Public Water
Systems in Support of the Second Six-Year Review of National Primary Drinking Water
Regulations. Office of Water. EPA-815-B-09-006. October 2009.
Zogorski, J.S., J.M. Carter, T. Ivahnenko, W.W. Lapham, M.J. Moran, B.L. Rowe, P.J. Squillace,
and P.L. Toccalino. 2006. Volatile organic compounds in the nation's ground water and drinking
water supply wells. U.S. Geological Survey Circular 1292, 101 p. Available on the Internet at:
http://pubs.er.usgs.gov/pubs/cir/cirl292.
21-14
-------
EPA-OGWDW Contaminant Occurrence Support Document for 2" 6-Year Review Category 1 Contaminants October 2009
22 Vinyl Chloride
This chapter on vinyl chloride is part of a report that is organized so that each chapter
presents information on one specific contaminant. Chapter 1 of this report, Introduction, presents
background information on all sources of data used as well as the analytical approach used to
estimate contaminant occurrence in drinking water. This contaminant chapter includes
background information (such as regulatory history, general chemical information, and
environmental fate and behavior), use and environmental release information, and vinyl chloride
occurrence estimates in ambient water and drinking water. All drinking water occurrence
estimates were based on data from the National Compliance Monitoring Information Collection
Request (ICR) Dataset for the Second Six-Year Review (the "Six Year Review-ICR Dataset").
For detailed information regarding this dataset, data management, data quality assessments, and
the analytical approach used to generate all contaminant occurrence estimates presented in this
chapter, please refer to USEPA (2009).
22.1 Background
The United States Environmental Protection Agency (EPA) published the current
National Primary Drinking Water Regulations (NPDWR) for vinyl chloride on July 8, 1987 (52
FR 25690 (USEPA, 1987)). The NPDWR established a maximum contaminant level goal
(MCLG) of zero based on a cancer classification of A, known human carcinogen. The NPDWR
also established a maximum contaminant level (MCL) of 2 |ig/L based on analytical feasibility.
According to the Agency for Toxic Substances and Disease Registry (ATSDR, 2006),
vinyl chloride (C2H3C1) is a colorless, flammable gas. It has a mild, sweet odor, and can be tasted
in water at concentrations greater than 3.4 parts per million. Other names for vinyl chloride
include chloroethene, chloroethylene, ethylene monochloride, and monochloroethylene (ATSDR,
2006).
Most vinyl chloride enters the environment from manufacturing plants that release it into
the air and water according to regulatory limits. Once in the air, vinyl chloride breaks down into
simple carbon compounds within a few days. Vinyl chloride evaporates easily, so contaminated
surface soils rapidly release it to the air. The presence of vinyl chloride in ground water is
frequently the result of subsurface reactions among other chemicals. Vinyl chloride is not thought
to accumulate in plant and animal tissues (ATSDR, 2006).
Inhalation of vinyl chloride is typical only in the vicinity of manufacturing plants, waste
sites, and landfills. Tobacco smoke also contains vinyl chloride, but only at trace levels. Surveys
of the presence of vinyl chloride in ground water have found it in less than 1% of wells. No
current data are available on whether vinyl chloride might leach from polyvinyl chloride (PVC)
pipes (ATSDR, 2006).
22-1
-------
EPA-OGWDW Contaminant Occurrence Support Document for 2" 6-Year Review Category 1 Contaminants October 2009
22.2 Use, Production, and Releases
According to ATSDR (2006), vinyl chloride once enjoyed a wide variety of uses.
However, since the 1970s it has been used primarily for the production of PVC, a common
plastic product. Vinyl chloride is also produced in the natural environment when other
manufactured chemicals are degraded by microorganisms (ATSDR, 2006).
The Toxic Substances Control Act (TSCA) Inventory contains confidential data reported
by industry as a partial update of the Chemical Substance Inventory Update System. The data are
reported under the Inventory Update Reporting regulation. Manufacturers and importers are
required to report company information and chemical information, including production volumes
for chemicals they manufactured or imported in excess of 10,000 pounds in the immediately
preceding fiscal year.
The TSCA Chemical Substance Inventory does not provide actual production volumes;
instead, it indicates at which of five production volume ranges a chemical was manufactured or
imported. The TSCA Inventory is updated every four years. Exhibit 22-1 presents the production
of vinyl chloride in the United States from 1986 to 2002. Note that the production of vinyl
chloride in the United States has exceeded 1 billion pounds since 1986.
Exhibit 22-1: Production of Vinyl Chloride in the United States
Inventory Update Year
1986 Range
1990 Range
1994 Range
1998 Range
2002 Range
Production Volume
> 1 billion pounds
> 1 billion pounds
> 1 billion pounds
> 1 billion pounds
> 1 billion pounds
Source: USEPA, 2003a
Vinyl chloride is listed as a Toxics Release Inventory (TRI) chemical. Exhibit 22-2a lists
the environmental releases of vinyl chloride between 1998 and 2006. (Exhibit 22-2b provides a
graphical presentation of the environmental releases.) Total releases declined steadily during the
period of review. Most releases were due to air emissions (ranging from approximately 82% to
98%). Off-site releases and underground injection provided most of the remainder of releases;
both were highly variable. Surface water discharges and releases to land comprised less than 1%
of the total. The TRI data for vinyl chloride were reported from 26 states, with 10 states
(Delaware, Illinois, Kentucky, Louisiana, Michigan, Mississippi, Missouri, New Jersey, Ohio and
Texas) reporting releases in all 9 years.
22-2
-------
EPA-OGWDW Contaminant Occurrence Support Document for 2" 6-Year Review Category 1 Contaminants October 2009
Exhibit 22-2a: Table of Toxics Release Inventory (TRI) Data for Vinyl Chloride
Year
1998
1999
2000
2001
2002
2003
2004
2005
2006
On-Site Releases (pounds)
Total Air
Emissions
885,387
855,691
773,826
734,259
670,992
669,404
653,837
566,475
559,292
Surface Water
Discharges
78
103
177
102
543
389
228
156
111
Underground
Injection
154
405
43,650
96,042
139,462
80,154
84,114
0
0
Releases to
Land
0
1
0
1
0
10
0
0
2,400
Off-site
Releases
(pounds)
69,214
14,015
623
1,002
349
25,777
27,918
28,018
23,587
Total On-&
Off -site
Releases
(pounds)
954,833
870,215
818,276
831,406
811,346
775,734
766,097
594,648
585,390
Source: USEPA, 2008
Exhibit 22-2b: Graph of Toxics Release Inventory (TRI) Data for Vinyl Chloride
Pounds Released
Environmental Releases of Vinyl Chloride
1 nnn nnn
onn nnn
snn nnn
vnn nnn
finn nnn
cnn nnn
/inn nnn
onn nnn
onn nnn
-i nn nnn
n
•
n
1 1 •
II 1 n
11 n II II 11
J-, LL
i in im.im-ii .
^
1998 1999 2000 2001 2002 2003 2004 2005 2006
Year
• Air Emissions DSurface Water Discharges D Underground Injection DReleases to Land • Off-site Releases
Source: USEPA, 2008
22.3 Occurrence in Ambient Water
Lakes, rivers, and aquifers are the sources of most drinking water. The United States
Geological Survey (USGS) National Water Quality Assessment (NAWQA) program is a
national-scale assessment of the occurrence of contaminants in ambient surface and ground
22-3
-------
EPA-OGWDW Contaminant Occurrence Support Document for 2" 6-Year Review Category 1 Contaminants October 2009
water. Contaminant occurrence in ambient water provides information on the potential for
contaminants to adversely affect drinking water supplies. Data on the occurrence of vinyl
chloride in ambient water are available from the NAWQA program.
22.3.1 EPA Summary Analysis of NAWQA Data
EPA, with the cooperation of the USGS, performed a summary analysis of all Cycle 1
(1992-2001) water monitoring data from all study units for vinyl chloride. Detection frequencies
were computed as the percentage of samples and sites with detections (i.e., with at least one
result equal to or greater than the reporting limit). Sample detections can be biased by frequent
sampling in areas with high (or low) occurrence. Calculating the percentage of sites with
detections can reduce this bias.
The results of the EPA analysis are presented in Exhibit 22-3. Overall, vinyl chloride was
detected in 1.1% of samples and at 0.8% of sites. Vinyl chloride was detected more frequently in
surface water than in ground water. The median concentration, based on detections from all sites,
was 0.2 ng/L. The 99th percentile concentration, based on detections from all sites, was 34 |ig/L.
Exhibit 22-3: EPA Summary Analysis of Vinyl Chloride Data from NAWQA Study
Units, 1992-2001
ground
water
surface
water
all
sites
Detection Frequency
(detections are results > RL1)
Number
of
Samples
4,659
1,414
6,073
% Samples
with
Detections
0.8%
1.8%
1.1%
Number
of Sites
4,159
191
4,350
% Sites
with
Detections
0.6%
4.7%
0.8%
Concentration Values
(of detections, in ug/L)
Minimum
0.01
0.01
0.01
Median
0.45
0.043
0.2
95th
Percen-
tile
29
1.6
8.13
99th
Percen-
tile
32
34
34
Maximum
32
34
34
1. RLs (Reporting Limits) for vinyl chloride varied, but did not exceed 0.01 ug/L.
22.4 Occurrence in Drinking Water
Vinyl chloride is regulated as a volatile organic compound (VOC) in drinking water. All
non-purchased community water systems (CWSs) and non-transient non-community water
systems (NTNCWSs) are required to sample for VOCs. The maximum waiver period for VOCs
is six years for ground water systems and three years for surface water systems.
All CWSs and NTNCWSs must collect four consecutive quarterly samples during the
initial three-year compliance period. * If all four samples are non-detections, then the system may
1 All new systems or systems using a new water source that began operation after January 22, 2004 must demonstrate
compliance with the MCL within a period of time specified by the State. The system must also comply with the
initial sampling frequencies specified by the State to ensure that a system can demonstrate compliance with the MCL.
22-4
-------
EPA-OGWDW Contaminant Occurrence Support Document for 2" 6-Year Review Category 1 Contaminants October 2009
reduce to annual sampling. After three annual samples without a detection, a system may be
granted a waiver. If a compound is detected (but at less than the MCL), the system must take one
sample per quarter until results are below the MCL (minimum of two quarterly samples for
ground water systems and four quarterly samples for surface water systems). If all quarterly
samples are below the MCL, the system may return to annual sampling. If a compound is
detected at a level greater than the MCL, the system (whether ground water or surface water)
must take four consecutive quarterly samples until all are below the MCL. If all quarterly
samples are below the MCL, the system may return to annual sampling.
The analysis of vinyl chloride occurrence presented in the following section is based on
state compliance monitoring data from the Six-Year Review-ICR Dataset. This is the largest and
most comprehensive compliance monitoring data set compiled by EPA's Drinking Water
program to date.
All of the Six-Year Review-ICR states provided occurrence data for vinyl chloride. These
data consist of 373,161 analytical results from 50,411 public water systems (PWSs) during the
period from 1998 to 2005. The number of sample results and systems vary by state, although the
state data sets have been reviewed and checked to ensure adequacy of coverage and
completeness. The national modal minimum reporting level (MRL) for vinyl chloride in the
datasetis 0.5 |ig/L.
22.4.1 Stage 1 Analysis of Occurrence Data
Stage 1 analyses have been performed at the sample, sample point, and system levels.
Note that some systems have multiple sample points, such as source water sample points or entry
points to the distribution system. Exhibit 22-4 presents the number of samples, systems, and
sample points in the data set, as well as the population-served by the systems and sample points,
and the percentage of detections and the percentage of the population-served by the systems and
sample points with detections. Detections in systems provide a fundamental indication of
national occurrence, but are not representative of MCL violations. For vinyl chloride, 0.411% of
50,411 systems reported detections. Exhibit 22-5 lists the minimum, median, 90th percentile, and
maximum vinyl chloride concentrations based on detections from all systems in the Six-Year
Review-ICR Dataset. The median concentration of vinyl chloride, based on detections from all
systems, was 0.89 |ig/L.
Exhibit 22-4: Vinyl Chloride Occurrence Data from the Six-Year Review-ICR
Dataset
Source Water Type
Number of
Samples
Number of
Systems
Number of
Sample Points
Pop. Served
by Systems
Pop. Served
by Sample
Points
Total Number
Ground Water
288,671
46,158
80,487
86,057,502
86,057,502
Surface Water
84,490
4,253
10,861
140,406,921
140,406,921
22-5
-------
EPA-OGWDW Contaminant Occurrence Support Document for 2" 6-Year Review Category 1 Contaminants October 2009
Source Water Type
Number of
Samples
Number of
Systems
Number of
Sample Points
Pop. Served
by Systems
Pop. Served
by Sample
Points
Combined Ground &
Surface Water
373,161
50,411
91,348
226,464,423
226,464,423
Percent with Detections
Ground Water
0.137%
0.349%
0.245%
1.388%
0.783%
Surface Water
0.183%
1.082%
0.552%
1 .663%
0.669%
Combined Ground &
Surface Water
0.147%
0.411%
0.281%
1.558%
0.712%
Exhibit 22-5: Vinyl Chloride Summary Statistics
Source Water Type
Ground Water
Detection Value (ug/L)
Minimum
0.01
Median
0.8
90tn
Percentile
4.6
Maximum
46.5
Surface Water
0.3
1
2.48
37.7
Combined Ground &
Surface Water
0.01
0.89
3.545
46.5
The following maps illustrate the geographic distribution of vinyl chloride occurrence in
drinking water. Exhibit 22-6 shows the states with systems with at least one detection and
Exhibit 22-7 shows the states with systems with at least one detection greater than the MCL
concentration. Note that five states did not submit data for use in the Six-Year Review (Kansas2,
Louisiana, Mississippi, Pennsylvania, and Washington).
The distribution of systems with detections of vinyl chloride is geographically dispersed.
Detection rates were generally low; although 36 states reported detections of vinyl chloride, only
4 states reported detections in more than 1% of their systems. Twenty-two states reported
detections greater than the MCL concentration of 2 |ig/L, but none of these states reported more
than 0.5% of systems with detections greater than 2 |ig/L.
2 Kansas submitted data after the data management and analyses had been conducted; data from Kansas are not
included in the analyses.
22-6
-------
EPA-OGWDW Contaminant Occurrence Support Document for 2" 6-Year Review Category 1 Contaminants October 2009
Exhibit 22-6: Percentage of Systems with at Least One Vinyl Chloride Detection,
by State
: , , : States Nol Submitting Any DM*
Slate) wMh No Detections
L«i than or Equal to 0.5% of Syitwns Detecting
Mor« tlwi 1 % of $yrt*n» Detecting
Exhibit 22-7: Percentage of Systems with at Least One Vinyl Chloride Detection
Greater than the MCL Concentration (> 2 pg/L), by State
St*tet Not SubmlBlng Any Drta
States with No Detections > MCL
Lees than 0.1% of Systems with Defections > MCL
Between 0,1% and 0.5% of Systems with Detections > MCL
More Ulan 0.5% of Systems with Defection s > MCL
Exhibit 22-8 presents the Stage 1 analysis of vinyl chloride occurrence in drinking water
from systems and the population-served by those systems in the Six-Year Review-ICR Dataset
22-7
-------
EPA-OGWDW Contaminant Occurrence Support Document for 2" 6-Year Review Category 1 Contaminants October 2009
relative to several thresholds. The thresholds are: 2 |ig/L, 1 |ig/L, and 0.5 |ig/L. A total of 84
systems (0.167% of all systems), serving about 896,000 persons, reported at least one detection
greater than or equal to 1 |ig/L; and 41 systems (0.081% of all systems), serving approximately
482,000 persons, reported at least one detection greater than the MCL concentration of 2 u.g/L.
Exhibit 22-8: Vinyl Chloride Stage 1 Analysis - Systems and Population with at
Least One Threshold Exceedance
Source
Water Type
Ground
Water
Threshold
> 2 ug/L
>1
> 0.5 |jg/L
Number of
Systems
Exceeding
Threshold
31
63
113
Percent of
Systems
Exceeding
Threshold
0.067%
0.136%
0.245%
Total Population-
Served by Systems
Exceeding
Threshold
334,543
427,229
825,343
Percent of
Population-
Served by
Systems
Exceeding
Threshold
0.389%
0.496%
0.959%
Surface
Water
> 2 ug/L
>1
> 0.5 ug/L
10
21
33
0.235%
0.494%
0.776%
147,536
468,912
1,621,211
0.105%
0.334%
1.155%
Combined
Ground &
Surface
Water
> 2 ug/L
>1
> 0.5 ug/L
41
84
146
0.081%
0.167%
0.290%
482,079
896,141
2,446,554
0.213%
0.396%
1.080%
The occurrence estimates presented above (in Exhibit 22-8) are based on the conservative
assumption that if a vinyl chloride detection is found in a single entry point/sample point in a
system, then the entire population-served by the system is assumed to potentially be exposed to
vinyl chloride (even though other entry points/sampling points at the system, and the population-
served by those entry/sample points, may exhibit no occurrence of vinyl chloride). The sample
point analysis is a less conservative estimate of the population-served by a system with a
detection of vinyl chloride because it estimates the potentially exposed population-served based
only on the proportion of the population-served by entry points/sample points with detections of
vinyl chloride (rather than the entire population-served by the system).
Exhibit 22-9 presents the Stage 1 analysis of vinyl chloride occurrence in drinking water
from sample points and population-served by those sample points in the Six-Year Review-ICR
Dataset. A total of 93 (0.102% of) sample points, serving an estimated 432,000 persons, reported
at least one detection greater than or equal to 1 |ig/L; and 44 (0.048% of) sample points, serving
an estimated 281,000 persons, reported at least one detection exceeding the MCL concentration
of 2 |ig/L.
22-8
-------
EPA-OGWDW Contaminant Occurrence Support Document for 2" 6-Year Review Category 1 Contaminants October 2009
Exhibit 22-9: Vinyl Chloride Stage 1 Analysis - Sample Points and Population with
at Least One Threshold Exceedance
Source
Water Type
Ground
Water
Threshold
> 2 |jg/L
>1
> 0.5 |jg/L
Number of
Sample Points
Exceeding
Threshold
33
70
137
Percent of
Sample Points
Exceeding
Threshold
0.041%
0.087%
0.170%
Total Population-
Served by Sample
Points Exceeding
Threshold
220,102
280,092
416,459
Percent of
Population-
Served by
Sample Points
Exceeding
Threshold
0.256%
0.325%
0.484%
Surface
Water
> 2 |jg/L
>1
> 0.5 |jg/L
11
23
38
0.101%
0.212%
0.350%
60,566
151,515
757,509
0.043%
0.108%
0.540%
Combined
Ground &
Surface
Water
> 2 |jg/L
>1
> 0.5 |jg/L
44
93
175
0.048%
0.102%
0.192%
280,669
431,607
1,173,969
0.124%
0.191%
0.518%
22.4.2 Stage 2 Analysis of Occurrence Data
For the Stage 2 analyses, three different substitution values -zero, one-half the MRL
value, and the full MRL value- were used to replace each non-detection record. Three arithmetic
mean vinyl chloride concentrations were calculated at each system using the zero, one-half MRL
and full MRL substitution values. These mean calculations were done for all systems with data in
the Six-Year Review-ICR dataset. Then the percentages of all systems with a mean concentration
greater than each threshold were calculated.
Exhibit 22-10 presents the Stage 2 analysis of estimated system mean concentrations for
vinyl chloride occurrence in drinking water relative to several thresholds. The thresholds are: 2
|ig/L, 1 |ig/L, and 0.5 |ig/L. Exhibit 22-11 presents similar information based on population-
served by the systems. Based on the Stage 2 analyses, no more than 21 ground water and surface
water systems (approximately 0.042% of all systems) had an estimated system mean greater than
or equal to 1 |ig/L. These 21 systems serve 56,161 persons. Eleven systems (0.022% of all
systems), serving 13,843 persons, had an estimated system mean greater than the MCL
concentration of 2 |ig/L.
22-9
-------
EPA-OGWDW Contaminant Occurrence Support Document for 2" 6-Year Review Category 1 Contaminants October 2009
Exhibit 22-10: Vinyl Chloride Stage 2 Analysis - Systems with a System Mean
Threshold Exceedance
Source
Water Type
Ground
Water
Threshold
> 2 |jg/L
>1
> 0.5 |jg/L
Number of Systems with Mean
Exceeding Threshold1
Full MRL
10
20
104
1/2 MRL
9
17
41
Zero
7
14
28
Percent of Systems with Mean
Exceeding Threshold1
Full MRL
0.022%
0.043%
0.225%
1/2 MRL
0.019%
0.037%
0.089%
Zero
0.015%
0.030%
0.061%
Surface
Water
> 2 |jg/L
>1
> 0.5 |jg/L
1
1
32
1
1
8
1
1
4
0.024%
0.024%
0.752%
0.024%
0.024%
0.188%
0.024%
0.024%
0.094%
Combined
Ground &
Surface
Water
> 2 |jg/L
>1
> 0.5 |jg/L
11
21
136
10
18
49
8
15
32
0.022%
0.042%
0.270%
0.020%
0.036%
0.097%
0.016%
0.030%
0.063%
1 The results shown in this table are based on system arithmetic means that were calculated using all sample
detection records and all non-detection records, with the non-detection records replaced (substituted) with
concentration values equal to the full MRL, 14 MRL, and zero.
Exhibit 22-11: Vinyl Chloride Stage 2 Analysis - Population-Served by Systems
with a System Mean Threshold Exceedance
Source
Water Type
Ground
Water
Threshold
> 2 |jg/L
> 1 H9/L
> 0.5 |jg/L
Population-Served by Systems
with Mean Exceeding Threshold1
Full MRL
12,478
54,796
712,162
1/2 MRL
10,793
21,736
338,964
Zero
4,688
16,526
71,559
Percent of Population -Served by
Systems with Mean Exceeding
Threshold1
Full MRL
0.014%
0.064%
0.828%
1/2 MRL
0.013%
0.025%
0.394%
Zero
0.005%
0.019%
0.083%
Surface
Water
> 2 |jg/L
>1
> 0.5 |jg/L
1,365
1,365
1,363,671
1,365
1,365
426,745
1,365
1,365
411,827
0.001%
0.001%
0.971%
0.001%
0.001%
0.304%
0.001%
0.001%
0.293%
Combined
Ground &
Surface
Water
> 2 |jg/L
>1
> 0.5 |jg/L
13,843
56,161
2,075,833
12,158
23,101
765,709
6,053
17,891
483,386
0.006%
0.025%
0.917%
0.005%
0.010%
0.338%
0.003%
0.008%
0.213%
1. The results shown in this table are based on system arithmetic means that were calculated using all sample
detection records and all non-detection records, with the non-detection records replaced (substituted) with
concentration values equal to the full MRL, 1A MRL, and zero.
22-10
-------
EPA-OGWDW Contaminant Occurrence Support Document for 2" 6-Year Review Category 1 Contaminants October 2009
22.5 Additional Occurrence Data
Volatile Organic Compounds in the Nation's Ground Water and Drinking Water Supply Wells
The USGS conducted a national assessment of 55 VOCs, including vinyl chloride, in
samples collected from drinking water supply wells, specifically domestic and public wells
(Zogorski et a/., 2006). This occurrence assessment was based on analyses of samples collected
at the well head, and before any treatment or blending, from about 2,400 domestic wells and
about 1,100 public wells.
In domestic wells, vinyl chloride had a detection frequency of 0.083% at an assessment
level of 0.2 jig/L (based on 2,401 samples). The median concentration of detections was 0.74
|ig/L. Two detections were within one order of magnitude of the MCL concentration (2 |ig/L).
No detections were greater than the MCL concentration.
In public wells, vinyl chloride had a detection frequency of 0.18% at an assessment level
of 0.2 |ig/L (based on 1,096 samples). The median concentration of detections was 3.2 |ig/L. No
detections were within one order of magnitude of the MCL concentration; however, two
detections were greater than the MCL concentration.
First Six-Year Review ofNPDWRs
Similar types of Stage 1 contaminant occurrence analyses were conducted for the first
Six-Year Review. The first Six-Year Review assessed regulated contaminant occurrence data
collected from 1993 to 1997. The first Six-Year Review data set consisted of data from 16 states
that were assembled into a "national cross-section" that was qualitatively indicative (though not
statistically representative) of national occurrence. For the first Six-Year Review (USEPA,
2003b), 149 systems (approximately 0.61% of about 25,000 systems), reported at least one
detection of vinyl chloride. Twenty-seven systems (approximately 0.11% of all systems),
reported at least one detection greater than the MCL concentration.
22.6 Summary of Data
TRI data indicate that the total off-site and on-site releases of vinyl chloride for the period
1998 to 2006 ranged from about 585,000 pounds to about 955,000 pounds, declining steadily
with time. Air emissions constituted nearly all of the releases; off-site releases and underground
injection contributed a few percent to the total. Releases to land and surface water discharges
were a minimal fraction of the total releases.
Ambient water data from the NAWQA program indicate that vinyl chloride was detected
in 1.1% of 6,073 ground and surface water samples and at 0.8% of 4,350 sites. The median
concentration, based on detections from all sites, was 0.2 |ig/L.
The Stage 1 analysis of occurrence in drinking water indicates that vinyl chloride was
detected in 0.411% of 50,411 surface water and ground water systems. The median
concentration, based on detections from all systems, was 0.89 |ig/L. In total, 0.081% of systems
22-11
-------
EPA-OGWDW Contaminant Occurrence Support Document for 2" 6-Year Review Category 1 Contaminants October 2009
(serving 0.213% of the population) reported at least one detection greater than the MCL
concentration of 2 |ig/L.
In the Stage 2 analysis, 0.022% of all systems (serving 0.006% of the population) had an
estimated system mean concentration of vinyl chloride greater than the MCL concentration.
These included 10 ground water systems and one surface water system, serving about 12,478
persons and 1,365 persons, respectively.
22.7 References
Agency for Toxic Substances and Disease Registry (ATSDR). 2006. Toxicological Profile for
Vinyl Chloride. July 2006. Available on the Internet at:
http://www.atsdr.cdc.gov/toxprofiles/tp20.html.
United States Environmental Protection Agency (USEPA). 1987. National Primary Drinking
Water Regulations—Synthetic Organic Chemicals; Monitoring for Unregulated Contaminants;
Final Rule. Federal Register. Vol. 52, No. 130. p. 25690, July 8, 1987.
USEPA. 2003a. Non-confidential Inventory Update Reporting (IUR) Production Volume
Information - Vinyl Chloride. Last modified September 10, 2007. Available on the Internet at:
http://www.epa.gov/oppt/iur/tools/data/2002-vol.htm. [Search for vinyl chloride.] Accessed
February 21, 2008.
USEPA. 2003b. Occurrence Estimation Methodology and Occurrence Findings for Six-Year
Review of National Primary Drinking Water Regulations. Office of Water. EPA-815-R-03-006.
June 2003.
USEPA. 2008. TRI Explorer: Trends - Vinyl Chloride. Released February 21, 2008. Available
on the Internet at: http://www.epa.gov/triexplorer/trends.htm. [Search for vinyl chloride.]
Accessed February 29, 2008.
USEPA. 2009. The Analysis of Regulated Contaminant Occurrence Data from Public Water
Systems in Support of the Second Six-Year Review of National Primary Drinking Water
Regulations. Office of Water. EPA-815-B-09-006. October 2009.
Zogorski, J.S., J.M. Carter, T. Ivahnenko, W.W. Lapham, M.J. Moran, B.L. Rowe, P.J. Squillace,
and P.L. Toccalino. 2006. Volatile organic compounds in the nation's ground water and drinking
water supply wells. U.S. Geological Survey Circular 1292, 101 p. Available on the Internet at:
http://pubs.er.usgs.gov/pubs/cir/cirl292.
22-12
-------
EPA-OGWDW Contaminant Occurrence Support Document for 2" 6-Year Review Category 1 Contaminants October 2009
23 Xylenes (Total)
This chapter on xylenes (total) is part of a report that is organized so that each chapter
presents information on one specific contaminant. Chapter 1 of this report, Introduction, presents
background information on all sources of data used as well as the analytical approach used to
estimate contaminant occurrence in drinking water. This contaminant chapter includes
background information (such as regulatory history, general chemical information, and
environmental fate and behavior), use and environmental release information, and total xylenes
occurrence estimates in ambient water and drinking water. All drinking water occurrence
estimates were based on data from the National Compliance Monitoring Information Collection
Request (ICR) Dataset for the Second Six-Year Review (the "Six Year Review-ICR Dataset").
For detailed information regarding this dataset, data management, data quality assessments, and
the analytical approach used to generate all contaminant occurrence estimates presented in this
chapter, please refer to USEPA (2009).
23.1 Background
The United States Environmental Protection Agency (EPA) published the current
National Primary Drinking Water Regulations (NPDWR) for total xylenes on January 30, 1991
(56 FR 3526 (USEPA, 1991)). The NPDWR established a maximum contaminant level goal
(MCLG) and a maximum contaminant level (MCL) of 10,000 |ig/L. The Agency based the
MCLG on a reference dose (RfD) of 2,000 jig/kg-day (2 mg/kg-day) and a cancer classification
of D, not classifiable as to human carcinogenicity.
According to the Agency for Toxic Substances and Disease Registry (ATSDR, 2007),
xylene is a colorless, flammable liquid with a sweet odor. There are three isomers of xylene:
meta-xylene, ortho-xylene, and para-xylene (m-, o-, and p-xylene, respectively). The term "total
xylenes" refers to all three isomers of xylene. Commercial or mixed xylene typically contains
about 40-65% m-xylene and up to 20% each of o-xylene, p-xylene, and ethylbenzene (ATSDR,
2007). Xylene is a synthetic chemical primarily produced from petroleum. Xylene also occurs
naturally in petroleum and coal tar and, to a small extent, is formed during forest fires (ATSDR,
2007).
Xylene can enter the environment when it is manufactured, packaged, shipped, or used.
Because xylene evaporates easily, most xylene that gets into soil and water will enter the air,
where it is broken down by sunlight within a couple of days. Xylene below the soil surface may
travel down through the soil and enter ground water, where it may remain for several months
before being broken down by microorganisms (ATSDR, 2007).
Humans may come into contact with xylene when using a variety of consumer products,
including gasoline, paint, varnish, shellac, rust preventatives, and cigarette smoke. Workers who
routinely come into contact with xylene-containing solvents are at the highest risk. Xylene has
also been detected in many types of food at levels ranging from 1 to 100 parts per billion (ppb).
Typical levels of xylene measured in outdoor air in the United States range from 1 to 30 ppb
(ATSDR, 2007).
23-1
-------
EPA-OGWDW Contaminant Occurrence Support Document for 2" 6-Year Review Category 1 Contaminants October 2009
23.2 Use, Production, and Releases
Xylene is primarily used as a solvent in the printing, rubber, and leather industries. Along
with other solvents, xylene is widely used as a cleaning agent and paint thinner. It is also found in
small amounts in airplane fuel and gasoline (ATSDR, 2007).
The Toxic Substances Control Act (TSCA) Inventory contains confidential data reported
by industry as a partial update of the Chemical Substance Inventory Update System. The data are
reported under the Inventory Update Reporting regulation. Manufacturers and importers are
required to report company information and chemical information, including production volumes
for chemicals they manufactured or imported in excess of 10,000 pounds in the immediately
preceding fiscal year.
The TSCA Chemical Substance Inventory does not provide actual production volumes;
instead, it indicates at which of five production volume ranges a chemical was manufactured or
imported. The TSCA Inventory is updated every four years. Exhibit 23-1 presents the production
of total xylenes in the United States from 1986 to 2002. Note that the production of total xylenes
in the United States has remained greater than 1 billion pounds since 1986.
Exhibit 23-1: Production of Xylenes (Total) in the United States
Inventory Update Year
1986 Range
1990 Range
1994 Range
1998 Range
2002 Range
Production Volume
> 1 billion pounds
> 1 billion pounds
> 1 billion pounds
> 1 billion pounds
> 1 billion pounds
Source: USEPA, 2003a
Xylenes (mixed isomers) is listed as a Toxics Release Inventory (TRI) chemical. Exhibit
23-2a lists the environmental releases of xylenes from 1998 to 2006. (Exhibit 23-2b provides a
graphical presentation of the environmental releases.) Total releases declined steadily during the
period of record. Air emissions comprised more than 90% of the total. Off-site releases were
generally the second most important component. Land releases and underground injection
abruptly increased in 2002 and 2003. Surface water discharges and underground injection were
generally less than 1% of the total. The TRI data for xylenes (mixed isomers) were reported from
all 50 states, plus American Samoa, Guam, the Northern Mariana Islands, Puerto Rico, and the
Virgin Islands. All states/entities except for American Samoa reported xylene (mixed isomers)
releases in all nine years.
23-2
-------
EPA-OGWDW Contaminant Occurrence Support Document for 2" 6-Year Review Category 1 Contaminants October 2009
Exhibit 23-2a: Table of Toxics Release Inventory (TRI) Data for Xylene
(Mixed Isomers)
Year
1998
1999
2000
2001
2002
2003
2004
2005
2006
On-Site Releases (pounds)
Total Air
Emissions
70,752,076
68,221,056
58,994,999
47,569,339
42,797,703
37,826,840
35,408,702
31,527,622
28,806,761
Surface Water
Discharges
57,504
30,098
82,264
21,732
42,840
27,383
28,551
20,812
14,075
Underground
Injection
124,285
87,440
151,148
80,941
1,274,919
850,266
991,687
943,046
866,934
Releases to
Land
110,051
252,417
142,558
177,289
220,451
1,149,247
1,034,448
1,404,603
1,322,519
Off-site
Releases
(pounds)
1,510,412
2,072,728
1,338,504
1,137,209
690,481
1,112,708
1,186,781
1,036,594
1,000,118
Total On- &
Off -site
Releases
(pounds)
72,554,328
70,663,739
60,709,473
48,986,510
45,026,395
40,966,443
38,650,169
34,932,677
32,010,408
Source: USEPA, 2008
Exhibit 23-2b: Graph of Toxics Release Inventory (TRI) Data for Xylene
(Mixed Isomers)
Environmental Releases of Xylene (Mixed Isomers)
1!
&
u>
•a
o
a.
80,000,000
70,000,000
60,000,000 -
50,000,000 -
40,000,000 -
30,000,000 -
20,000,000 -
10,000,000 -
0
1998
1999
2000
2001
2002
Year
2003
2004
2005
2006
lAir Emissions DSurface Water Discharges nUnderground Injection n Releases to Land • Off-site Releases
Source: USEPA, 2008
23.3 Occurrence in Ambient Water
Lakes, rivers, and aquifers are the sources of most drinking water. The United States
Geological Survey (USGS) National Water Quality Assessment (NAWQA) program is a
23-3
-------
EPA-OGWDW Contaminant Occurrence Support Document for 2" 6-Year Review Category 1 Contaminants October 2009
national-scale assessment of the occurrence of contaminants in ambient surface and ground
water. Contaminant occurrence in ambient water provides information on the potential for
contaminants to adversely affect drinking water supplies. Data on the occurrence of total xylenes
in ambient water are available from the NAWQA program.
23.3.1 EPA Summary Analysis of NAWQA Data
EPA, with the cooperation of the USGS, performed a summary analysis of all Cycle 1
(1992-2001) water monitoring data from all study units for total xylenes. Detection frequencies
were computed as the percentage of samples and sites with detections (i.e., with at least one
result equal to or greater than the reporting limit). Sample detections can be biased by frequent
sampling in areas with high (or low) occurrence. Calculating the percentage of sites with
detections can reduce this bias.
The results of the EPA analysis are presented in Exhibit 23-3. Overall, xylenes (total)
were detected in 1.4% of samples and at 1.4% of sites. Xylenes were detected more frequently in
surface water than in ground water. The median concentration, based on detections from all sites,
was 0.7 |ig/L. The 99th percentile concentration, based on detections from all sites, was 300 |ig/L.
Exhibit 23-3: EPA Summary Analysis of Xylenes (Total) Data from NAWQA Study
Units, 1992-2001
ground
water
surface
water
all
sites
Detection Frequency
(detections are results > RL1)
Number
of
Samples
1,840
36
1,876
% Samples
with
Detections
1.3%
11.1%
1.4%
Number
of Sites
1,731
19
1,750
% Sites
with
Detections
1.2%
15.8%
1.4%
Concentration Values
(of detections, in ug/L)
Minimum
0.2
0.2
0.2
Median
0.8
0.25
0.7
95th
Percen-
tile
42
5.2
42
99th
Percen-
tile
300
5.2
300
Maximum
300
5.2
300
1. RLs (Reporting Limits) for xylenes (total) varied, but did not exceed 0.2pg/L
23.4 Occurrence in Drinking Water
Xylenes are regulated as volatile organic compounds (VOCs) in drinking water. All non-
purchased community water systems (CWSs) and non-transient non-community water systems
(NTNCWSs) are required to sample for VOCs. The maximum waiver period for VOCs is six
years for ground water systems and three years for surface water systems.
23-4
-------
EPA-OGWDW Contaminant Occurrence Support Document for 2" 6-Year Review Category 1 Contaminants October 2009
All CWSs and NTNCWSs must collect four consecutive quarterly samples during the
initial three-year compliance period.1 If all four samples are non-detections, then the system may
reduce to annual sampling. After three annual samples without a detection, a system may be
granted a waiver. If a compound is detected (but at less than the MCL), the system must take one
sample per quarter until results are below the MCL (minimum of two quarterly samples for
ground water systems and four quarterly samples for surface water systems). If all quarterly
samples are below the MCL, the system may return to annual sampling. If a compound is
detected at a level greater than the MCL, the system (whether ground water or surface water)
must take four consecutive quarterly samples until all are below the MCL. If all quarterly
samples are below the MCL, the system may return to annual sampling.
The analysis of xylenes occurrence presented in the following section is based on state
compliance monitoring data from the Six-Year Review-ICR Dataset. This is the largest and most
comprehensive compliance monitoring data set compiled by EPA to date.
Forty-four of the Six-Year Review-ICR states provided occurrence data for xylenes.
(There were no data for xylenes from Hawaii.) These data consist of 363,466 analytical results
from 47,698 public water systems (PWSs) during the period from 1998 to 2005. The number of
sample results and systems vary by state, although the state data sets have been reviewed and
checked to ensure adequacy of coverage and completeness. The national modal minimum
reporting level (MRL) for xylenes in the dataset is 0.5 |ig/L.
23.4.1 Stage 1 Analysis of Occurrence Data
Stage 1 analyses have been performed at the sample, sample point, and system levels.
Note that some systems have multiple sample points, such as source water sample points or entry
points to the distribution system. Exhibit 23-4 presents the number of samples, systems, and
sample points in the data set, as well as the population-served by the systems and sample points,
and the percentage of detections and the percentage of the population-served by the systems and
sample points with detections. Detections in systems provide a fundamental indication of
national occurrence, but are not representative of MCL violations. For xylenes, 7.592% of 47,698
systems reported detections. Exhibit 23-5 lists the minimum, median, 90th percentile, and
maximum xylenes concentrations based on detections from all systems in the Six-Year Review-
ICR Dataset. The median concentration of xylenes, based on detections from all systems, was 1.1
1 All new systems or systems using a new water source that began operation after January 22, 2004 must demonstrate
compliance with the MCL within a period of time specified by the State. The system must also comply with the
initial sampling frequencies specified by the State to ensure that a system can demonstrate compliance with the MCL.
23-5
-------
EPA-OGWDW Contaminant Occurrence Support Document for 2" 6-Year Review Category 1 Contaminants October 2009
Exhibit 23-4: Xylenes (Total) Occurrence Data from the Six-Year Review-ICR
Dataset
Source Water Type
Number of
Samples
Number of
Systems
Number of
Sample Points
Pop. Served
by Systems
Pop. Served
by Sample
Points
Total Number
Ground Water
281,787
43,726
76,361
83,748,512
83,748,512
Surface Water
81,679
3,972
10,162
134,323,023
134,323,023
Combined Ground &
Surface Water
363,466
47,698
86,523
218,071,535
218,071,535
Percent with Detections
Ground Water
2.811%
7.275%
5.336%
13.222%
4.896%
Surface Water
1 .654%
11.078%
5.757%
20.134%
5.806%
Combined Ground &
Surface Water
2.551%
7.592%
5.386%
17.480%
5.457%
Exhibit 23-5: Xylenes (Total) Summary Statistics
Source Water Type
Ground Water
Detection Value (ug/L)
Minimum
0.01
Median
1.1
90tn
Percentile
8.1
Maximum
4,000
Surface Water
0.002
1
8.3
1,860
Combined Ground &
Surface Water
0.002
1.1
8.1
4,000
Exhibit 23-6 illustrates the geographic distribution of xylenes occurrence in drinking
water, identifying the states with systems having at least one detection. Note that five states did
not submit data for use in the Six-Year Review (Kansas2, Louisiana, Mississippi, Pennsylvania,
and Washington). Hawaii did submit Six-Year data for most contaminants, but did not submit
any data for xylenes.
The distribution of systems with detections of xylenes is geographically dispersed.
Detection rates were generally high; all states reported detections in at least one of their systems.
2 Kansas submitted data after the data management and analyses had been conducted; data from Kansas are not
included in the analyses.
23-6
-------
EPA-OGWDW Contaminant Occurrence Support Document for 2" 6-Year Review Category 1 Contaminants October 2009
Five states reported detections in more than 20% of their systems. No states reported detections
greater than the MCL concentration of 10,000 |ig/L.
Exhibit 23-6: Percentage of Systems with at Least One Xylenes Detection,
by State
States with Waivers or Not Submitting
States with No Detection*
Less than 10% of Systems Detecting
| Between 10% and 20% of Systems Detecting
| More than 20% of Systems Detecting
Exhibit 23-7 presents the Stage 1 analysis of xylenes occurrence in drinking water from
systems and the population-served by those systems in the Six-Year Review-ICR Dataset relative
to several thresholds. The thresholds are: 10,000 |ig/L, 5,000 |ig/L, and 1,000 |ig/L. Xylenes
occurred in more ground water systems than surface water systems, which reported at least one
detection greater than 1,000 |ig/L. No ground or surface water systems reported detections of
xylenes greater than or equal to 5,000 |ig/L or 10,000 |ig/L.
Exhibit 23-7: Xylenes (Total) Stage 1 Analysis - Systems and Population with at
Least One Threshold Exceedance
Source
Water
Type
Ground
Water
Threshold
> 1 0,000 |jg/L
> 5,000 |jg/L
> 1 |jg/L
Number of
Systems
Exceeding
Threshold
0
0
9
Percent of
Systems
Exceeding
Threshold
0.000%
0.000%
0.021%
Total Population-
Served by
Systems
Exceeding
Threshold
0
0
2,344
Percent of
Population -Served
by Systems
Exceeding
Threshold
0.000%
0.000%
0.003%
23-7
-------
EPA-OGWDW Contaminant Occurrence Support Document for 2" 6-Year Review Category 1 Contaminants October 2009
Source
Water
Type
Surface
Water
Threshold
> 1 0,000 |jg/L
>
>1
Number of
Systems
Exceeding
Threshold
0
0
1
Percent of
Systems
Exceeding
Threshold
0.000%
0.000%
0.025%
Total Population-
Served by
Systems
Exceeding
Threshold
0
0
1,785
Percent of
Population -Served
by Systems
Exceeding
Threshold
0.000%
0.000%
0.001%
Combined
Ground &
Surface
Water
> 1 0,000 |jg/L
>
>1
0
0
10
0.000%
0.000%
0.021%
0
0
4,129
0.000%
0.000%
0.002%
The occurrence estimates presented above (in Exhibit 23-7) are based on the conservative
assumption that if a detection of xylenes is found in a single entry point/sample point in a system,
then the entire population-served by the system is assumed to potentially be exposed to xylenes
(even though other entry points/sampling points at the system, and the population-served by
those entry/sample points, may exhibit no occurrence of xylenes). The sample point analysis is a
less conservative estimate of the population-served by a system with a detection of xylenes
because it estimates the potentially exposed population-served based only on the proportion of
the population-served by entry points/sample points with detections of xylenes (rather than the
entire population-served by the system).
Exhibit 23-8 presents the Stage 1 analysis of xylenes occurrence in drinking water from
sample points and population-served by those sample points in the Six-Year Review-ICR
Dataset. Similar to the system-level occurrence analysis, xylenes were detected in more ground
water than surface water sample points at levels greater than 1,000 |ig/L. Xylenes were never
detected at levels greater than or equal to 5,000 jig/L or 10,000 |ig/L.
Exhibit 23-8: Xylenes (Total) Stage 1 Analysis - Sample Points and Population
with at Least One Threshold Exceedance
Source
Water
Type
Ground
Water
Threshold
> 1 0,000 |jg/L
>
>1
Number of
Sample Points
Exceeding
Threshold
0
0
9
Percent of
Sample Points
Exceeding
Threshold
0.000%
0.000%
0.012%
Total Population-
Served by Sample
Points Exceeding
Threshold
0
0
1,525
Percent of
Population-
Served by
Sample Points
Exceeding
Threshold
0.000%
0.000%
0.002%
23-8
-------
EPA-OGWDW Contaminant Occurrence Support Document for 2" 6-Year Review Category 1 Contaminants October 2009
Source
Water
Type
Surface
Water
Threshold
>1 0,000 |jg/L
>
>1
Number of
Sample Points
Exceeding
Threshold
0
0
1
Percent of
Sample Points
Exceeding
Threshold
0.000%
0.000%
0.010%
Total Population-
Served by Sample
Points Exceeding
Threshold
0
0
1,785
Percent of
Population-
Served by
Sample Points
Exceeding
Threshold
0.000%
0.000%
0.001%
Combined
Ground &
Surface
Water
>1 0,000 |jg/L
>
>1
0
0
10
0.000%
0.000%
0.012%
0
0
3,310
0.000%
0.000%
0.002%
23.4.2 Stage 2 Analysis of Occurrence Data
For the Stage 2 analyses, three different substitution values -zero, one-half the MRL
value, and the full MRL value- were used to replace each non-detection record. Three arithmetic
mean xylene concentrations were calculated at each system using the zero, one-half MRL and
full MRL substitution values. These mean calculations were done for all systems with data in the
Six-Year Review-ICR dataset. Then the percentages of all systems with a mean concentration
greater than each threshold were calculated.
Exhibit 23-9 presents the Stage 2 analysis of estimated system mean concentrations for
xylenes occurrence in drinking water relative to several thresholds. The thresholds are: 10,000
|ig/L, 5,000 |ig/L, and 1,000 |ig/L. Exhibit 23-10 presents similar information based on
population-served by the systems. Based on the Stage 2 analyses, no ground water or surface
water systems had estimated system means greater than or equal to any of the thresholds.
Exhibit 23-9: Xylenes (Total) Stage 2 Analysis - Systems with a System Mean
Threshold Exceedance
Source Water
Type
Ground
Water
Threshold
> 10,000 |jg/L
>
>1
Number of Systems with Mean
Exceeding Threshold1
Full MRL
0
0
0
1/2 MRL
0
0
0
Zero
0
0
0
Percent of Systems with Mean
Exceeding Threshold1
Full MRL
0.000%
0.000%
0.000%
1/2 MRL
0.000%
0.000%
0.000%
Zero
0.000%
0.000%
0.000%
Surface
Water
> 10,000 |jg/L
>
>1
0
0
0
0
0
0
0
0
0
0.000%
0.000%
0.000%
0.000%
0.000%
0.000%
0.000%
0.000%
0.000%
23-9
-------
EPA-OGWDW Contaminant Occurrence Support Document for 2" 6-Year Review Category 1 Contaminants October 2009
Source Water
Type
Combined
Ground &
Surface
Water
Threshold
> 1 0,000 |jg/L
> H9/L
>1
Number of Systems with Mean
Exceeding Threshold1
Full MRL
0
0
0
1/2 MRL
0
0
0
Zero
0
0
0
Percent of Systems with Mean
Exceeding Threshold1
Full MRL
0.000%
0.000%
0.000%
1/2 MRL
0.000%
0.000%
0.000%
Zero
0.000%
0.000%
0.000%
1. The results shown in this table are based on system arithmetic means that were calculated using all sample
detection records and all non-detection records, with the non-detection records replaced (substituted) with
concentration values equal to the full MRL, 1A MRL, and zero.
Exhibit 23-10: Xylenes (Total) Stage 2 Analysis - Population-Served by Systems
with a System Mean Threshold Exceedance
Source Water
Type
Ground
Water
Threshold
> 1 0,000 |jg/L
>
>1
Population-Served by Systems with
Mean Exceeding Threshold1
Full MRL
0
0
0
1/2 MRL
0
0
0
Zero
0
0
0
Percent of Population-Served
by Systems with Mean
Exceeding Threshold1
Full MRL
0.000%
0.000%
0.000%
1/2 MRL
0.000%
0.000%
0.000%
Zero
0.000%
0.000%
0.000%
Surface
Water
> 1 0,000 |jg/L
>
>1
0
0
0
0
0
0
0
0
0
0.000%
0.000%
0.000%
0.000%
0.000%
0.000%
0.000%
0.000%
0.000%
Combined
Ground &
Surface
Water
> 1 0,000 |jg/L
>
>1
0
0
0
0
0
0
0
0
0
0.000%
0.000%
0.000%
0.000%
0.000%
0.000%
0.000%
0.000%
0.000%
1. The results shown in this table are based on system arithmetic means that were calculated using all sample
detection records and all non-detection records, with the non-detection records replaced (substituted) with
concentration values equal to the full MRL, 14 MRL, and zero.
23.5 Additional Occurrence Data
Volatile Organic Compounds in the Nation's Ground Water and Drinking Water Supply Wells
The USGS conducted a national assessment of 55 VOCs, including o-, m- and/>-xylenes,
in samples collected from drinking water supply wells, specifically domestic and public wells
(Zogorski et a/., 2006). This occurrence assessment was based on analyses of samples collected
at the well head, and before any treatment or blending, from about 2,400 domestic wells and
about 1,100 public wells.
In domestic wells, total xylenes had a detection frequency of 0.21% at an assessment
level of 0.2 jig/L (based on 2,388 samples). The median concentration of detections was 0.020
|ig/L. No detections were greater than or within one order of magnitude of the MCL
concentration (10,000 |ig/L).
23-10
-------
EPA-OGWDW Contaminant Occurrence Support Document for 2" 6-Year Review Category 1 Contaminants October 2009
In public wells, total xylenes had a detection frequency of 0.56% at an assessment level of
0.2 |ig/L (based on 1,069 samples). The median concentration of detections was 0.44 |ig/L. No
detections were greater than or within one order of magnitude of the MCL concentration.
First Six-Year Review ofNPDWRs
Similar types of Stage 1 contaminant occurrence analyses were conducted for the first
Six-Year Review. The first Six-Year Review assessed regulated contaminant occurrence data
collected from 1993 to 1997. The first Six-Year Review data set consisted of data from 16 states
that were assembled into a "national cross-section" that was qualitatively indicative (though not
statistically representative) of national occurrence. For the first Six-Year Review (USEPA,
2003b), 919 systems (approximately 4.16% of about 22,000 systems), reported at least one
detection of xylenes. No systems reported a detection greater than the MCL concentration.
23.6 Summary of Data
TRI data indicate that the total off-site and on-site releases of xylenes for the period 1998-
2006 ranged from about 32 million pounds to about 72.5 million pounds. Releases were greatest
in 1998 and decreased steadily with time. Air emissions comprised more than 90% of the total
releases. Off-site releases and releases to land were the second and third most important
contributors. Surface water discharges and underground injection were a very small fraction of
the total releases.
Ambient water data from the NAWQA program indicate that xylenes were detected in
1.4% of 1,876 ground and surface water samples and at 1.4% of 1,750 sites. The median
concentration, based on detections from all sites, was 0.7 |ig/L.
The Stage 1 analysis of occurrence in drinking water indicate that xylenes were detected
in 7.592% of 47,698 surface water and ground water systems. The median concentration, based
on detections from all systems, was 1.1 |ig/L. No systems reported detections greater than 5,000
|ig/L or the MCL concentration of 10,000 |ig/L. Nine ground water systems and one surface
water system reported at least one detection greater than 1,000 |ig/L. The pattern was the same
for sample points.
In the Stage 2 analysis, no systems had an estimated system mean concentration of
xylenes greater than any of the thresholds.
23.7 References
Agency for Toxic Substances and Disease Registry (ATSDR). 2007. Toxicological Profile for
Xylene. August 2007. Available on the Internet at:
http ://www. atsdr.cdc.gov/toxprofiles/tp71 .html.
23-11
-------
EPA-OGWDW Contaminant Occurrence Support Document for 2" 6-Year Review Category 1 Contaminants October 2009
United States Environmental Protection Agency (USEPA). 1991. National Primary Drinking
Water Regulations—Synthetic Organic Chemicals and Inorganic Chemicals; Monitoring for
Unregulated Contaminants; National Primary Drinking Water Regulations Implementation;
National Secondary Drinking Water Regulations; Final Rule Federal Register. Vol. 56, No. 30. p.
3526, January 30, 1991.
USEPA. 2003a. Non-confidential Inventory Update Reporting (IUR) Production Volume
Information - Xylenes. Last modified September 10, 2007. Available on the Internet at:
http://www.epa.gov/oppt/iur/tools/data/2002-vol.htm. [Search for xylenes.] Accessed February
21,2008.
USEPA. 2003b. Occurrence Estimation Methodology and Occurrence Findings for Six-Year
Review of National Primary Drinking Water Regulations. Office of Water. EPA-815-R-03-006.
June 2003.
USEPA. 2008. TRI Explorer: Trends - Xylenes. Released February 21, 2008. Available on the
Internet at: http://www.epa.gov/triexplorer/trends.htm. [Search for xylenes (mixed isomers).]
Accessed February 29, 2008.
USEPA. 2009. The Analysis of Regulated Contaminant Occurrence Data from Public Water
Systems in Support of the Second Six-Year Review of National Primary Drinking Water
Regulations. Office of Water. EPA-815-B-09-006. October 2009.
Zogorski, J.S., J.M. Carter, T. Ivahnenko, W.W. Lapham, M.J. Moran, B.L. Rowe, P.J. Squillace,
and P.L. Toccalino. 2006. Volatile organic compounds in the nation's ground water and drinking
water supply wells. U.S. Geological Survey Circular 1292, 101 p. Available on the Internet at:
http://pubs.er.usgs.gov/pubs/cir/cirl292.
23-12
------- |