United States
Environmental Protection
Agency
Office of Water
4606
EPA816-R-99-006
November 1999
A Review of Contaminant
Occurrence in Public
Water Systems
\
Printed on Recycled Papei
-------
-------
CONTENTS
TABLES AND FIGURES
ACRONYMS ix
EXECUTIVE SUMMARY xi
ACKNOWLEDGMENTS xiii
DISCLAIMER xv
I. INTRODUCTION 1
LA. Background 1
n. DATA USED AND METHODS OF ANALYSIS 5
It. A. Data Management 8
n.B. Data Elements 9
H.B.1. Analytical Results Data Elements 10
H.B.2. System and Sample Elements 11
H.C. Data Quality and Consistency Issues 12
H.D. URCIS Data Quality 13
n.E. Basic Analysis and Review 15
DDL TOWARD A REPRESENTATIVE SAMPLE 15
in.A. Representativeness 16
ffl.A.1. Manufacturing Indicators 17
ffl.A.2. Agricultural Indicators 18
ffl.B. Representative Cross-Section of States 19
IV. THE SDWA OCCURRENCE DATA 20
FV.A. Contaminant Groups 21
FV.A.1, SOC Groups 21
IV.A.2. VOC Groups 22
FV.B. Overview of Results 23
IV.C. Contaminant Occurrence in Surface Water and Ground Water Systems 25
IV.C.l. Ground Water Vulnerability 30
IV.D. National Cross-Section: Perspectives and Comparison with Other Data 34
IV.D.I. URCIS Data 34
IV.D.2. Novartis Atrazine and Simazine Data 34
FV.D.3. USGS Data 36
-------
A Review of Contaminant Occurrence in Public Water Systems
V. CONTAMINANT OCCURRENCE: SYSTEM SIZE AND OTHER VULNERABILITY
FACTORS 39
V.A. System Size and Contaminant Occurrence 39
V.A.I. Systems By The Numbers; Other Perspectives On Occurrence 44
V.A.l.a. System Numbers Perspective 44
V.A.l.b. Population Perspective 45
V.B. Contaminant Sources and Occurrence 48
V.C. Temporal Variability and Vulnerability 51
V.C.I. SOCs 52
V.C.2. VOCs 57
V.C.3. Implications For Monitoring 62
VI. CO-OCCURRENCE 65
VI.A. Correlation Methods 66
VLB. Correlation Results 66
VD. SUMMARY AND CONCLUSIONS 68
REFERENCES CITED 75
APPENDIX A.
APPENDIX B.
APPENDIX C.
APPENDIX D.
Summary Tables of Contaminant Occurrence Data
USGS Studies Reviewed for Contaminant Occurrence Information
General Data Quality Issues
Summary Tables V.A.1 through V.A.8
IV
-------
TABLES AND FIGURES
Table LA. La. Inorganic chemicals (13) regulated in PWSs under SDWA Phase rules 3
Table LA. 1 .b. Synthetic organic chemicals (30) regulated in PWSs under SDWA
Phase rules 4
Table I.A.I.e. Volatile organic chemicals (21) regulated in PWSs under SDWA
Phase rules 5
Table D. Principal State and supplemental databases used for analysis in this report 7
Table Ht. States with water quality data included in the drinking water occurrence
assessment 20
Table IV.B. 1. Five Phase rule contaminants which occur most frequently at concentrations
greater than their MCL in either surface water or ground water systems 25
Table IV.C. 1. Summary of occurrence of selected Phase D/V contaminants in water
systems using surface water, from national cross-section States; ranges from all
States studied 27
Table IV.C.2. Summary of occurrence of selected Phase WV contaminants in water
systems using ground water, from national cross-section States; ranges from all
States studied 28
Table TV.C.3. Summary comparison of occurrence of selected Phase n/V contaminants in
water systems using surface water vs. ground water, from national cross-section
States 29
Table IV.C.4. Summary of national MCL violations for surface water and ground water
systems 30
Table IV.C.5. Percentage of Iowa municipal water-supply wells with detections of
various contaminants: for wells of various depths and for wells finished in
different aquifers 31
Table IV.C.6. Percentage of Iowa public water systems with detections of various
contaminants or contaminant groups, for systems using different source water,
aquifers, and wells of different depths 32
-------
A Review of Contaminant Occurrence in Public Water Systems
Figure IV.C.l. Nitrate-N concentrations versus well depth, from water-quality analyses of
private wells in north-central (NC) and northeastern (NE) Iowa, illustrating the
general inverse relationship between well depth (or ground-water depth) and
contaminant occurrence 33
Table IV.D.l. Comparison of national occurrence estimates from URCIS and the national
cross-section of State SDWA data 35
Table IV.D.2. Comparison of national estimates of occurrence (percentage of water systems
with detections) of atrazine and simazine from the Novartis study of 21 high-use
States and from the national cross-section (8 States) compiled in this study 36
Table FV.D.3. Percentage detections of atrazine in different occurrence studies and the
minimum reporting levels used 38
Table IV.D.4. Pesticide detections from public water system wells at different minimum
reporting levels 38
Figure V.A.I. Summary of the percentage of systems with detections (>MRL) and
exceedances (>MCL) of selected VOCs, comparing ground-water and surface-water
supplied systems, by size of system (population served categories), for A) a high-
occurrence state, and B) a low-occurrence state 40
Figure V. A.2. Summary of the percentage of systems with detections (>MRL) and
exceedances (>MCL) of selected SOCs, comparing ground-water and surface-water
supplied systems, by size of system (population served categories), for A) a high-
occurrence state, and B) a low-occurrence state 41
Figure V.A.3. Summary of the percentage of systems with detections (>MRL) and
exceedances (>MCL) of selected A) VOCs, and B) SOCs, comparing ground-water
and surface-water supplied systems, by size of system (population served categories),
from the national URCIS database 42
Table V. A.9. Summary of national MCL violations by system size, for surface water and
ground water systems. Percent MCL violations derived from SDWIS information
for 1/1/93-3/31/1998 43
Table V.A.10. Total number of nonpurchased, community and non-transient non-community
water systems (CWSs and NTNCWSs), by source water and system size 45
Table V.A.I 1. Community and non-transient non-community water systems (CWSs and
NTNCWSs) with detections of PCE >MRL and >MCL from URCIS, by system size .. 46
Table V.A.12. Summary comparison of occurrence of selected Phase n/V contaminants in
water systems using surface water vs. ground water, from national cross-section
States 47
VI
-------
A Review of Contaminant Occurrence in Public Water Systems
Table V.B. 1. Summary of the percentage of systems with detections (>MRL) of selected
VOC contaminants, comparing surface-water and ground-water supplied systems,
by the State ranking for total TRI release (in pounds per square mile) 49
Table V.B.2. Summary of the percentage of systems with detections (>MRL) of selected
SOC contaminants, comparing surface-water and ground-water supplied systems,
by the State ranking for Agricultural Chemical Expenditures 50
Figure V.B.I. Percentage of CWS in a State with detections (>MRL) of atrazine related to
the State's national rank (l=highest rank, i.e., greatest amount applied) for pounds
of atrazine applied, for surface-water and ground-water supplied systems (data from
Novartis study, Clarkson et al., 1997) 51
Figure V.C. 1. Summary of monthly average total pesticide concentrations in streams sampled
in the USGS NAWQA Program, for streams affected by runoff from agricultural and
urban lands. (After Larson, Gilliom, and Capel, 1999) 53
Figure V.C.2. Number of community water systems with monthly mean atrazine
concentrations above 3.0 fj.g/L (in raw water) 55
Figure V.C.3. Percentage of systems with detections and maximum concentration detected,
by month, for various herbicides, for surface water systems in Ohio 56
Figure V.C.4. Percentage of systems with detections and maximum concentration detected,
by month, for various herbicides, for surface water systems in Ohio 57
Figure V.C.5. Percentage of systems with detections of xylene, by month, for three States ... 59
Figure V.C.6. Percentage of systems with detections of xylene, toluene, and benzene, by
month, for Alabama 60
Figure V.C.7. Percentage of systems with detections of tetrachloroethylene and
trichloroethylene, by month, for Illinois 61
Figure V.C.8. Percentage of systems with detections (>MRL, >0.5MCL) of any of the 21
regulated VOCs, by month, for Iowa 62
Figure V.C.9. Schematic annual contaminant concentration profile, derived from actual data
from a Midwestern stream, with three sampling scenarios (A, B, and C) noted (with
four sampling times for each) 64
Table V.C.I. Percentage of Monte Carlo sampling simulations that are within, over, or
under the tolerance of the time-weighted annual mean atrazine concentration
calculated from detailed field sampling 65
Table VI.B.1. Occurrence and co-occurrence of SOCs and VOCs in two States 67
Vll
-------
I
-------
ACRONYMS
Advanced Notice of Proposed Rule Making (ANPRM)
Alternative Monitoring Guidelines (AMG)
American Water Works Service Company (AWWSC)
Chemical Monitoring Reform (CMR)
Comma Separated Values (CSV)
Community Water System (CWS)
Dense Non-aqueous Phase Liquids (DNAPLS)
Dibromochloropropane (DBCP)
Environmental Protection Agency (EPA)
Ethylene Dibromide (EDB)
High-occurrence Synthetic Organic Chemical (HiSOC)
Inorganic Chemical (IOC)
Light Non-Aqueous Phase Liquids (LNAPLs)
Low-occurrence Synthetic Organic Chemical (LoSOC)
Maximum Contaminant Level (MCL)
Method Detection Limit (MDL)
Minimum Reporting Level (or Limit, MRL)
National Primary Drinking Water Regulations (NPDWRs)
National Water Quality Assessment Program (NAWQA)
National Alachlor Well Water Survey (NAWWS)
National Pesticide Survey (NPS)
IX
-------
A Review- of Contaminant Occurrence in Public Water Systems
Non-Transient Non-Community Water System (NTNCWS)
North-central (NC)
Northeastern (NE)
Office of Ground Water and Drinking Water (OGWDW)
Percentage of Systems with Exceedances (>MCL)
Percentage of Systems with Detections (>MRL)
Permanent Monitoring Relief (PMR)
Public Water System (PWS)
Public Water System Identifier (PWSID)
Regulatory Implementation Branch (RIB)
Safe Drinking Water Act (SDWA)
Safe Drinking Water Information System (SDWIS)
Source Water Assessment Program (SWAP)
Synthetic Organic Chemical (SOC)
Tetrachloroethylene (PCE)
Toxic Release Inventory (TRI)
Transient Non-Community Water System (TNCWS)
Tricloroethylene (TCE)
United States Geological Survey (USGS)
Unregulated Contaminant Monitoring Information System (URCIS)
Volatile Organic Chemical (VOC)
I
,
-------
USEPA, Office of Water Report: EPA 816-R-99-006, November 1999
A REVIEW OF CONTAMINANT OCCURRENCE IN
PUBLIC DRINKING WATER SYSTEMS:
EXECUTIVE SUMMARY
This study represents the Environmental Protection Agency's (EPA's) most extensive analysis to
date of the occurrence of contaminants in Public Water Systems (PWSs) regulated under the Safe
Drinking Water Act (SDWA). EPA's Office of Ground Water and Drinking Water conducted
this study to provide a better scientific basis for considering changes to the chemical monitoring
requirements in the drinking water program.
State databases, comprised of SDWA compliance-monitoring data from PWSs, were the primary
data sources for this analysis. Data from 12 states were analyzed in detail. Data from 8 States
were used to develop a national cross-section of contaminant occurrence. The States were
selected to represent the national range of hydrologic regimes and pollution potential. The data
used represent more than 10.7 million analytical results from nearly 26,000 PWSs. More than
70% of the data are from 1993 or later; reflecting the beginning of the Phase n/V requirements.
While the national cross-section cannot be stated to be "statistically representative," the results
are clearly indicative of the national values for occurrence. The eight cross-section States
represent over 25% of the population using public water supplies, as well. Comparisons with
other limited national data suggest that the cross-section occurrence values are slightly high yet
very comparable to other national averages.
Additional sources of national data were also used, such as: EPA's Unregulated Contaminant
Monitoring Information System (URCIS), which includes data from 40 States and territories
from the first round of unregulated contaminant monitoring; and U.S. Geological Survey studies
of ambient monitoring, including results from the National Water Quality Assessment Program,
covering parts of 42 States.
Data management and handling were an important component of this effort because each State
database required unique editing and formatting to facilitate consistent analysis and valid
comparisons among States. Variables were cross-checked for consistency and unique issues
were resolved in consultation with the States. Whenever errors or ambiguities could not be
resolved, data were eliminated to avoid aberrant results. This analysis summarizes contaminant
occurrence by systems, to avoid the skew that is inherent in summaries by sample. Only standard
SDWA compliance samples were used.
Contaminant occurrence in drinking water systems is highly variable. All 64 Phase n/V
contaminants have been detected in drinking water systems, however, the frequency of
occurrence in water systems ranges widely. Fifty-nine of the 64 contaminants have a reported
detection at greater than half their Maximum Contaminant Level (MCL); but MCL exceedances
are not common. The five that do not have detections >0.5 MCL are all Synthetic Organic
Chemicals (SOCs). Fifty-five of the contaminants occur in more than 1% of PWSs. The nine
that occur in less than 1% of PWSs are also all SOCs: alachlor, carbofuran, chlordane,
glyphosate, hexachlorobenzene, hexachloropentadiene, oxamyl, toxaphene, and PCBs (though
PCBs have not been monitored as intensively as other SOCs).
-------
A Review of Contaminant Occurrence in Public Water Systems
Twenty-nine of the 30 regulated SOCs have not been detected at all in some States (only
phthalate occurs in every State, but its occurrence relates, in part, to sample contamination). The
greatest range for any of the organic chemicals is for atrazine, which ranges from no systems
with detections in some States to 97% of surface water systems showing detections in
Midwestern States.
Seven of the 21 Volatile Organic Chemicals (VOCs) occur in every State studied: ethylbenzene,
cis-1,2- dichloroethane, tetrachloroethylene (PCE), trichloroethylene (TCE), vinyl chloride,
1,1,1- trichlorethane, and xylenes. It is common for many VOCs to occur in about 30 percent of
surface or ground water systems.
Surface water systems tend to be more vulnerable than ground water systems for most
contaminants. SOCs are far more common in surface water systems than ground water systems
and surface water systems have more MCL exceedances as well. VOCs are more common than
most SOCs in both surface and ground water systems. VOCs show greater general occurrence in
surface water but surface water systems and ground water systems are nearly equal for the
occurrence of exceedances. Ground water systems show slightly more MCL violations for
VOCs. Inorganic Chemicals (lOCs) are about equally common in surface water and ground
water systems, but ground water systems generally have higher average concentrations. lOCs
have relatively high occurrence rates because they occur naturally.
For most VOCs and many SOCs, larger systems show proportionately greater contaminant
occurrence than smaller systems, and a tendency, at least for VOCs, to have proportionately more
systems with exceedances. This trend is most consistent for ground water systems. To get the
greater ground water yields needed, larger systems typically must use unconfined ground water
supplies, which are typically shallow and more vulnerable to contaminant occurrence. While
large systems have proportionately greater occurrence the number of small systems is much
greater. Hence, there is a large number of small systems affected by contaminants.
Ground water systems are complex and it is difficult to make generalizations about vulnerability.
Information on aquifer characteristics, well depth, well casing and construction details, wellhead
protection information, as well as system and well maintenance history are some of the factors
that might be considered to establish specific monitoring approaches for a ground water system.
Many SOCs (pesticides, in particular) exhibit strong seasonal patterns because their application
or discharge into the environment is concentrated seasonally. In contrast, VOCs do not typically
show such seasonality. Targeting monitoring to such vulnerable periods can improve the
effectiveness of compliance monitoring and the accuracy of exposure estimates.
Statistical reviews suggest that sampling strategies can be designed to account for seasonal peak
occurrence. While vulnerable periods may vary from region to region, the data suggest that only
one vulnerable period would likely need to be targeted in a given region. Vulnerable periods are
apparent for surface waters and shallow, vulnerable ground water systems. For deeper ground
water systems few generalizations can be made, except to note that as well depth increases the
degree of temporal variability usually decreases.
Xll
-------
ACKNOWLEDGMENTS
The compilation and analysis of data presented in this report was undertaken by EPA's Office of
Ground Water and Drinking Water (OGWDW) to enhance the scientific understanding of the
occurrence of chemical contaminants in public drinking water systems and to refine the basis for
the monitoring of these contaminants. This effort was directed by Mr. Edwin Thomas of
OGWDW. It began under the management of Mr. Michael Muse.
We would like to thank the many States, as well as the American Water Works Service Company
that contributed data sets and valuable advice. Thanks also to the many public water systems that
conducted the monitoring. EPA's Office of Pesticide Programs and the U.S. Geological Survey
contributed valuable supplemental studies and data. Mr. Lewis Summers and Mr. Charles Job of
OGWDW managed the completion of EPA's URCIS data base. The report received extensive
review from EPA and State personnel, as well as external stakeholders and expert peer reviewers.
In particular, thanks are due Robert Libra (Geological Survey Bureau, Iowa Department of
Natural Resources), Robert Gilliom and John Zogorski (Water Resources Division, National
Water Quality Assessment, US Geological Survey), and Kenneth Cantor (Occupational
Epidemiology Branch, National Cancer Institute) for their constructive reviews and suggestions.
The Cadmus Group, Inc. served as the prime contractor for this project, supporting the data
analysis and report development. Dr. George Hallberg served as Cadmus' Project Manager. For
further information, contact Ed Thomas, USEPA, OGWDW, 202-260-0910 (fax, 202-401-2345)
or e-mail, thomas.edwin@epa.gov.
X11J
-------
-------
DISCLAIMER
This report does not constitute U.S. Environmental Protection Agency Policy. Mention of trade
names or commercial products does not constitute endorsement or recommendation for use.
This document is designed to provide technical background for the Office of Ground Water and
Drinking Water's program. The document does not, however, substitute for the Safe Drinking
Water Act or EPA's regulations nor is this document a regulation itself. Thus, it cannot impose
legally-binding requirements on EPA, States, or the regulated community, and may not apply to a
particular situation based on the circumstances.
XV
-------
-------
A REVIEW OF CONTAMINANT OCCURRENCE
IN PUBLIC DRINKING WATER SYSTEMS
I. INTRODUCTION
The purpose of this report is to review the analysis of data on the occurrence of chemical
contaminants in public drinking-water systems regulated under the Safe Drinking Water Act
(SDWA). This is the most extensive analysis EPA has conducted to date for these contaminants
and it provides an overview of several occurrence issues that can contribute improved scientific
understanding for management of SDWA programs.
LA. Background
EPA's Office of Ground Water and Drinking Water (OGWDW) is required by the Safe Drinking
Water Act (SDWA) to protect public health and safety by setting standards for the quality of
water provided by public water systems. To meet the requirements of the SDWA, EPA has
published the National Primary Drinking Water Regulations (NPDWRs) to establish Federal
standards for drinking water contaminants and rules for monitoring those contaminants. EPA
promulgated the first regulations, for nine inorganic chemicals, six organic chemicals, total
coliform bacteria, and turbidity, in 1975. Subsequently, EPA has published numerous
amendments and standards for additional contaminants.
The Phase rules regulate 64 chemical contaminants (Table I. A.I) that are of primary concern
because of their potential to cause chronic health effects and their known occurrence in the
source waters of public water systems. The NPDWRs for Phase I were promulgated in July
1987. The Phase n regulations, which modified the requirements for Phase I contaminants, were
promulgated in 1991. Standards for additional contaminants were established with the Phase DOB
regulation of 1991 and the Phase V regulation of July 1992. The Phase rules have been
challenged by some as overly prescriptive, complicated, and costly. Critics have contended that
nationally uniform rules often result in requirements for monitoring contaminants that do not
occur in many different regions of the country. However, there has been limited data or analysis
of data to support these criticism or to inform deliberations for reform.
The Regulatory Implementation Branch (RIB) (formerly the Drinking Water Branch) of the
OGWDW is responsible for implementing these regulations and recommending revisions to
improve the efficiency of implementation and reduce the regulatory burden. In 1995, the RIB
began a review to assess ways to simplify requirements and facilitate implementation of its many
programs under SDWA. Simultaneously, OGWDW was responding to a Presidential directive to
improve all Agency programs and reduce regulatory burden. Streamlining the monitoring
requirements of the Phase n, IEB, and V rules, described as Chemical Monitoring Reform
(CMR), offers an opportunity to respond to both initiatives.
RIB established the CMR State Workgroup to develop a framework for proposed CMR changes
to the Phase rules. During the deliberations to develop CMR, Congress passed the 1996
Amendments to SDWA. In these Amendments, Congress authorized States that have received
-------
A Review of Contaminant Occurrence in Public Water Systems
EPA approval of their Source Water Assessment Programs (SWAP) to offer Permanent
Monitoring Relief (PMR) to public water systems. The provisions of the PMR have been
considered under the Alternative Monitoring Guidelines (AMG), and most components have
been incorporated into the CMR options that EPA is considering. The 1996 Amendments also
directed EPA to review the monitoring requirements for not fewer than 12 contaminants
identified by the Administrator and to promulgate any necessary modifications by August 6,
1998. These actions altered the time line for development and proposal of the CMR.
In June 1997, an Advanced Notice of Proposed Rule Making (ANPRM) was issued for CMR and
AMG. The ANPRM described some options under consideration and asked for additional public
and stakeholder input. EPA's limited analysis of occurrence data, developed for the ANPRM, as
well as the comments and response to the ANPRM and the CMR State Workgroup, all indicated
the need for a more comprehensive review of contaminant occurrence and assessment of the
implications of these data for designing chemical monitoring reform options. The CMR
revisions were delayed until occurrence data could be further evaluated. On August 5,1997, the
Agency announced the final guidelines for AMG.
A preliminary review of occurrence data was presented at a CMR Stakeholders' Meeting in
Spring 1998. The Stakeholders also concurred that further action on CMR should be postponed
until the data could be analyzed more fully. In July 1998, a Federal Register Notice announced
that no further action was warranted or would be taken on CMR until the occurrence data had
been analyzed and the results evaluated.
This report summarizes the results of extensive additional analysis of contaminant occurrence
data and reviews these findings to inform continuing deliberations for possible changes in
chemical monitoring strategies.
-------
A Review of Contaminant Occurrence in Public Water Systems
Table I.A.l.a. Inorganic chemicals (13) regulated in PWSs under SOW A Phase rules.
Contaminant
MCL
mg/L
Common Sources of Contaminant in
Drinking Water
Potential Health
Effects
Inorganic Chemicals
Antimony (total)
Arsenic
Asbestos
Barium
Beryllium (tola!)
Cadmium
Chromium
Cyanide
Fluoride
Mercury
Nickel
Selenium
Thallium (total)
0.0060
0.050
7mf/L
2.0
0.0040
0.0050
0.10
0.20
4.0
0.0020
0.10
0.050
0.0020
Fire retardants, ceramics, electronics,
fireworks, solder
Natural deposits, smelters, glass, electronic
wastes, old orchards
Natural deposits; asbestos cement in water
systems
Natural deposits; pigments, epoxy sealants,
spent coal
Electrical, aerospace, defense industries
Galvanized pipe corrosion; natural deposits;
batteries, paints
Natural deposits; mining, electroplating,
pigments
Electroplating, steel, plastics, mining,
fertilizer
Natural deposits; fertilizer, aluminum
industry; H20 additive
Crop runoff; natural deposits; batteries,
electrical switches
Metal alloys, electroplating, batteries,
chemical production
Natural deposits; mining, smelting, coal/oil
combustion
Electronics, drugs, alloys, glass
Cancer
Skin, nervous system
toxicity
Cancer
Circulatory system
effects
Bone, lung damage
Kidney effects
Liver, kidney, circulatory
disorders
Thyroid, nervous system
damage
Skeletal and dental
fluorosis
Kidney, nervous system
disorders
Heart, liver damage
Liver damage
Kidney, liver, brain,
intestinal disorders
-------
A Review of Contaminant Occurrence in Public Water Systems
Table I.A.l.b. Synthetic organic chemicals (30) regulated in PWSs under SDWA Phase rules.
Contaminant
MCL
mg/L
Common Sources of Contaminant in
Drinking Water
Potential Health
Effects
Synthetic Organic Chemicals
Alachlor (Lasso)
Atrazine
Benzo[a]pyrene
bis(2-ethylhexyl) Adipate
bis(2-ethylhexyl) Phthalale
Carbofuran (Furadan)
Chlordane
2,4-D
Daiapon
Dibromochloropropane
(1 ,2-Dibromo-3-chloropropane,
orDBCP)
Dinoseb
Diquat
Endothall
Endrin
Ethylene dibromide
(1,2-Dibromoethane; orEDB)
Glyphosate (Roundup)
fieptachlor
Heptachlor Epoxide
fiexachlorobenzene
riexachlorocyclopeniadiene
Jndane
Methoxychlor
Oxamyl (Vydate)
*CBs as Decachlorobiphenyl;
PCB total as DCBP
'entachlorophenol
Picloram (Tordon)
Simazine
2,3,7,8-TCDD(Dioxin)
2.4,5-TP (Silvex)
Toxaphene
0.0020
0.0030
0.0002
0.40
0.0060
0.040
0.0020
0.070
0.20
0.0002
0.0070
0.020
0.10
0.0020
0.00005
0.70
0.00040
0.00020
0.00 JO
0.050
0.00020
0.040
0.20
0.00050
0.00100
0.50
0.0040
3x10-8
0.050
0.0030
Herbicide on corn, soybeans, other crops
Herbicide on corn and non-cropland
Coal tar coating; burning organic matter;
volcanoes, fossil fuels
Synthetic robber, food packaging, cosmetics
PVC and other plastics
Soil fumigant on com and cotton; restricted in
some areas
Leaching from soil treatment for termites
Herbicide on wheat, corn, rangelands. lawns
Herbicide on orchards, beans, coffee, lawns,
road/railways
Soil fumigant on soybeans, cotton, pineapple,
orchards
Herbicide on crop and non-crop applications
Herbicide on land & aquatic weeds
Herbicide on crops, land/aquatic weeds; rapidly
degraded
Pesticide on insects, rodents, birds; restricted
since 1980
Leaded gas additives; leaching of soil fumigant
Herbicide on grasses, weeds, brush
Leaching of insecticide for termites, very few
crops
Biodegradaiion of heptachlor
Pesticide production waste by-product
Pesticide production intermediate
Insecticide on cattle, lumber, gardens; restricted
1983
Insecticide for fruits, vegetables, alfalfa,
livestock, pets
Insecticide on apples, potatoes, tomatoes
Coolant oils from electrical transformers;
plasticizers
Wood preservatives, herbicide, cooling tower
wastes
Herbicide on broadleaf and woody plants
Herbicide on grass sod, some crops, aquatic algae
Chemical production by-product; impurity in
lerbicides
Herbicide on crops, right-of-way, golf courses;
canceled 1983
Insecticide on cattle, cotton, soybeans; canceled
1982
Cancer
Mammary gland tumors
Cancer
Decreased body weight
Cancer
Nervous, reproductive
system disorders
Cancer
Liver and kidney damage
Liver, kidney disorders
Cancer
Thyroid, reproductive organ
damage
Li ver, kidney, eye effects
Liver, kidney,
gastrointestinal disorders
Liver, kidney, heart damage
Cancer
Liver, kidney damage
Cancer
Cancer
Cancer
Kidney, stomach damage
Liver, kidney, nerve,
immune, circulatory system
Growth, liver, kidney, nerve
disorders
Kidney damage
Cancer
Cancer, liver and kidney
effects
Kidney, liver damage
Cancer
Cancer
Liver and kidney damage
Cancer
-------
A Review of Contaminant Occurrence in Public Water Systems
Table LA.l.c. Volatile organic chemicals (21) regulated in PWSs under SDWA Phase rules.
Contaminant
MCL
mg/L
Common Sources of Contaminant in
Drinking Water
Potential Health
Effects
Volatile Organic Chemicals
Benzene
Carbon tetrachloride
Chlorobenzene
o-Dichlorobenzene
p-Dichlorobenzene
1 .2-Dichloroethane
cis- 1 ,2-DichloroethyIene
trans- 1 ,2-Dichloroethylene
U-Dichloroethene
(1,1 -Dichloroethylene)
1 ,2-Dichloropropane
Ethyl benzene
Methylene chloride
(Dichloromethane)
Styrene
Tetrachloroethylene
Toluene
1 ,2,4-TrichIorobenzene
1 , 1 ,1 -Trichloroethane
1,1,2-TrichIoroethane
Trichloroethene
(Trichloroethylene)
Vinyl chloride
Xylenes (TotaJ)
0.0050
0.0050
0.10
0.60
0.0750
0.0050
0.070
0.10
0.0070
0.0050
0.70
0.0050
0.10
0.0050
1.0
0.070
0.2000
0.0050
0.0050
0.0020
10.0
Some foods; gas, drugs, pesticide, paint, plastic
industries
Solvents and their degradation products
Waste solvent from metal degreasing processes
Paints, engine cleaning compounds, dyes,
chemical wastes
Room and water deodorants, and "mothballs"
Leaded gas. fumigants. paints
Waste industrial extraction solvents
Waste industrial extraction solvents
Plastics, dyes, perfumes, paints
Soil fumigant; waste industrial solvents
Gasoline; insecticides; chemical manufacturing
wastes
Paint stripper, metal degreaser, propellant,
extraction
Plastics, rubber, resin, drug industries; leachate
from city landfills
Improper disposal of dry cleaning and other
solvents
Gasoline additive; manufacturing and solvent
operations
Herbicide production; dye carrier
Adhesives, aerosols, textiles, paints, inks, metal
degreasers
Solvent in rubber, other organic products;
chemical production wastes
Textiles, adhesives and metal degreasers
May leach from PVC pipe; formed by solvent
breakdown
By-product of gasoline refining; paints, inks,
detergents
Cancer
Cancer
Nervous system and liver
Liver, kidney, blood cell
damage
Cancer
Cancer
Liver, kidney, nervous,
circulatory disorders
Liver, kidney, nervous,
circulatory disorders
Cancer, liver & kidney
effects
Liver, kidney effects; cancer
Liver, kidney, nervous
system disorders
Cancer
Liver, nervous system
damage
Cancer
Liver, kidney, nervous,
circulatory disorders
Liver, kidney damage
Liver, nervous system
effects
Kidney, liver, nervous
system
Cancer
Cancer
Liver, kidney; nervous
system disorders
II. DATA USED AND METHODS OF ANALYSIS
The data included in the CMR contaminant occurrence analysis came from various sources
including State occurrence databases, EPA databases, private water system databases (e.g.,
-------
A Review of Contaminant Occurrence in Public Water Systems
American Water Works Service Company), special studies conducted by States or industry, and
published data and reports of the U.S. Geological Survey (USGS). These major data sources,
which were analyzed in detail, are summarized in Table H Additional data were analyzed and
other findings were reviewed from various States, industry and published research. These
additional sources are not included in Table D but will be introduced and referenced where
appropriate in this report. These additional databases were not completely analyzed for several
reasons. In some cases necessary data elements were incomplete, formatting problems impeded
analysis, or their data was of limited applicability. In other cases they were redundant with more
complete databases.
The State databases, providing SDWA compliance monitoring data from public drinking water
systems, were the most important sources for this analysis. These data form the core of the
results presented in this report. As summarized in Table It, data from 12 States were analyzed in
detail, providing the summary information described in this report. These data represent more
than 10.7 million analytical results from nearly 26,000 public water systems. In most cases, the
State databases contained additional data that were not included either because they were not
appropriate for this analysis or because they posed various data quality problems (as discussed in
later sections). (As noted on Table IH, data from 14 States were reviewed.)
As shown in Table JO, various States may include data from different time periods. However, the
majority of data are from 1993 and later (especially the data used in this summary analysis),
which marks the beginning of Phase H/V monitoring. Initial screening of the data showed that
most data quality problems were in pre-1993 data. Therefore, in many of the data sets, results
gathered before 1993, or 1990 in particular, were eliminated from these analyses. More than
70% of all data utilized in this "report are from 1993 or later; this proportion is even greater for
most synthetic organic chemicals (SOCs).
These databases represent State primacy agency SDWA monitoring results as provided by the
States. The data represent the analytical results for Community Water Systems (CWSs) and
Non-Transient Non-Community Water Systems (NTNCWSs) that are required to monitor for the
Phase n/V chemicals. Some States included data from Transient Non-Community Water
Systems (TNCWSs), but these systems are not required to monitor for the Phase n/V
contaminants. The TNCWSs' data are almost exclusively for nitrate, which is not included in
these discussions. Some State data sets are not complete for all systems. Massachusetts had only
VOC data available from approximately one-third of their CWSs. The Ohio database only
includes results from a special study of selected herbicides (SOCs) in surface water systems
(Ohio makes these data available on an internet site, http://www.epa.state.oh.us/
ddagw/pestspst.html). Michigan provided an excellent database for SOCs and VOCs (volatile
organic chemicals), but it does not include IOC data (inorganic chemicals).
States provided these data sets voluntarily; some were available publicly; and a few were
requested to broaden the spatial coverage of the analysis. For example, Iowa and California have
published reports of their analysis of SDWA data (Hallberg et al., 1996; California DHS, 1997).
California also makes their data publicly available on CD ROM. But for purposes of this
analysis (so data could be analyzed in a consistent manner among states for compatibility) it was
necessary to get a copy of the data in a different format.
-------
A Review of Contaminant Occurrence in Public Water Systems
Table II. Principal State and supplemental databases used for analysis in this report.
State
Alabama
California
Illinois
Indiana
Iowa
Massachusetts
Michigan (CWS)
Michigan
(NTNC)
Montana
Hew Jersey
New Mexico
Ohio
Oregon
STATE
TOTALS
URCIS (40 States
& Territories)
Novartis (21
States)
AWWSC(19
States)
USGS (42 States)
TOTALS
Contaminant Groups
Represented (lOCs,
SOCs, VOCs,
O=other
contaminants)1
IOC, SOC, VOC, O
IOC, SOC, VOC, O
IOC, SOC, VOC, 0
IOC, SOC, VOC, O
IOC, SOC, VOC, O
(published report)
VOCs only
SOC, VOC, O
SOC, VOC, O
IOC, SOC, VOC, 0
IOC, SOC, VOC, 0
IOC, SOC, VOC, O
SOC
(only selected herbicides
in SW systems)
IOC, SOC, VOC, 0
some SOC, VOC, O
SOC (atrazine and
simazine)
IOC, SOC, VOC
some SOC, VOC, O
Number of
Analytical
Results
included in
this Analysis
708,569
3,897,362
2,967,946
257,428
458,320
76,737
346,181
339,540
276,675
980,915
266,262
6,646
169,521
10,752,102
3,492,480
128,603
55,526
14,428,711
Number of
PWSs with
Results for this
Analysis
731
6,414
1,392
1,488
2,090
322
1,146
2,106
1,786
4,503
1,299
144
2,345
25,766
24,357
15,766
137
Time Period
1985-1998
1984-1998
1987-1997
1982-1997
1988-1995
1993-1997
1993-1997
1993-1997
1993-1998
1993-1998
1992-1996
1996-1997
1990-1998
1983-1992
1993-1996
1995-1996
1992-1997
'Includes data for: lOCs = the 13 regulated inorganic chemicals in Table I.A.I; SOCs = the 30 regulated
synthetic organic chemicals in Table I.A.1; VOCs = the 21 regulated volatile organic chemicals in Table I.A.1; Other
= other regulated or unregulated chemicals not listed in Table I.A.I.
-------
A Review of Contaminant Occurrence in Public Water Systems
Table n also summarizes four other major sources of data. Coupled with the State information,
the data reviewed for this report include more than 14 million analytical results. (The number of
systems from the States and these supplemental sources is not totaled because there is overlap
among these special databases and some States, and hence many systems would be double-
counted.) The supplemental databases include:
• EPA's Unregulated Contaminant Monitoring Information System (URCIS). Includes
data from about 40 States and Territories from the first round of unregulated contaminant
monitoring (1989-1993). These data also include samples taken prior to 1989 which were
grandfathered into the database. (The values shown here for URCIS may differ from
other EPA reviews because of the screening conducted for this study to ensure
consistency in analysis.)
• A special regional analysis of atrazine and simazine occurrence in CWSs, conducted by
Novartis Crop Protection, Inc., the pesticides' registrant. This study was conducted as
part of a special study under direction from EPA's Office of Pesticide Programs. While
limited to two contaminants, this study is important because it compiled data from CWS
monitoring conducted between 1993 and 1996, from 21 States across the country, for one
of the most widely occurring contaminants. It represents one of the most complete
assessments of drinking water occurrence data of its kind (Clarkson et al., 1997).
• Drinking water data from systems operated by the American Water Works Service
Company. AWWSC operates systems in many states, and while it keeps relatively few
records (compared to State occurrence databases), its multi-state nature helped to provide
insights early in this analysis, and to serve as a comparison to the representative cross-
section results. (AWWSC also collects various supplemental data and maintains its own
data system, which provided useful insight on various data quality issues.)
• Studies of the U.S. Geological Survey (USGS), particularly results from the ongoing
National Water Quality Assessment Program (NAWQA) and other special contaminant
studies (e.g., the Mid-Continent Herbicide Study, the Toxics Program). The USGS
conducts extensive monitoring activities under nationally consistent protocols. This
makes the data very valuable for reviewing national patterns of occurrence that can
provide perspectives on drinking water problems. As further discussed in this report, the
majority of these data aire from ambient water sources and not drinking water. In all,
more than 75 USGS reports on surface and ground water quality in parts of 42 States
were reviewed. (A listing of USGS reports reviewed is given in Appendix B.)
II.A. Data Management
There are numerous data handling and management issues, as well as data quality issues, that had
to be addressed to enable the analysis presented in this report. The primary objective was to
develop a consistent and repeatable approach that would allow valid comparisons between and
among the various data sets, allowing the data to be jointly evaluated to provide an overview of
occurrence patterns at the national level. Some of these issues are reviewed below as a preface to
understanding the results.
-------
A Review of Contaminant Occurrence in Public Water Systems
In general, States and other sources did not reorganize or reformat data, but simply transmitted
the data in whatever manner was easiest. For example, while the Phase WV compliance data
from 1993-1995 were of greatest interest, in many cases it was easier for the State simply to
transmit their entire data set, which generally contained information on all chemical
contaminants, over a greater span of years (as noted Table n). In addition to the regulated
contaminants, every database included data on additional contaminants which varied among the
databases.
Data were transferred using three main media: FTP, e-mail, and diskettes (including zip-disks or
CD-ROM). Evident from the number of analytical results presented on Table n, these are very
large databases, often several megabytes in size, and transmission was often complicated. Many
of the data sets were received "as is" and had not been formatted by the State in any way. The
data were received in a number of file types including spreadsheet files, DBF files, and THM
files. Each database was unique in format, layout, custom codes, and data element usage.
After receipt, an initial review of the information in each database was performed. In most cases,
the data was not accompanied by a protocol outlining each variable. In many cases, the variable
headings could be determined by examination. In every instance, follow-up with the State/data
source was necessary to decipher variable headings or contaminant codes. When all variables
were understood, a formatting plan for the data was developed. Nearly all of the data sets
required some type of formatting to facilitate analysis. Data formatting problems varied from
one dataset to another. (Some of the most common data problems are outlined in Appendix C.)
All statistical analyses were conducted in SAS® statistical software. Data formatting problems
were corrected in Microsoft® Excel with the aid of specialized programs written in Visual Basic®
or were corrected directly in SAS before the analysis began.2 Data formatting was the most time
consuming and labor intensive part of the data analysis. Each data set presented unique
challenges. While analysis of the data was consistent from one data set to another, each data set
required some unique editing and filtering because of differences among basic data elements (as
described below).
II.B. Data Elements
Each drinking water contaminant database was reviewed to ensure it contained the basic data
elements (data fields) necessary to conduct a consistent analysis for this study. These elements
were reviewed with State/data-source contacts both before and after data were received to ensure
consistent and appropriate interpretations. While the presence of such elements enables the
various data sets to be analyzed in a similar manner, each also may be used in unique ways by the
individual States/sources. A brief review of these elements, examples of some of the common
problems encountered, and some of the data editing that was required, are also presented.
2 SAS is a registered trademark of the SAS Institute, Inc. Excel and Visual Basic are trademarks of the
Microsoft Corporation.
-------
A Review of Contaminant Occurrence in Public Water Systems
HE. 1. Analytical Results Data Elements
Contaminant Identifier (ID). States use different identifiers including EPA OGWDW
Contaminant IDs, STORET contaminant ID numbers, and unique State contaminant identifiers,
and/or some combination of the above. Some include numeric identification codes as well as the
chemical or trade name of the contaminant. In some cases, multiple IDs referred to the same
contaminant. In most cases it was necessary to confirm the "translation" of IDs to particular
contaminants. CAS numbers were added to ensure accurate identification and processing.
Crosswalk tables had to be constructed to ensure that all entries could be converted to uniform
IDs.
Analytical Method. In general, this was entered as the EPA method number used for the analysis,
but this also varies among States, with some using their own unique method IDs. In several
cases there was no record for this element. At the State level, the assumption was that the State
certified labs use only approved methods, so this field was not necessary for every sample.
Detection or Reporting Limit. Knowing the Minimum Reporting Level (or Limit, MRL) for a
contaminant-method combination can be very helpful for understanding some aspects of the
results. Very few databases contained a field for this element. A few provided either a
crosswalk table or a general paper copy. (The MRL can sometimes be surmised from the data, if
less-than fields and data values are recorded. Also, the MRL is commonly referred to as the
Method Detection Limit (MDL) but this is not strictly correct. The MRL is the minimum
concentration that can be reported with confidence and is based on the MDL. The MDL is a
function of the method and equipment used to determine the presence of an analyte in relation to
background noise. The MRL is often 5 to 10 times greater, or a statistical multiplier greater, than
the MDL.) Although most databases did not provide the MRLs, some basic information is
available for compliance analyses. Within for the Drinking Water program EPA specifies MDLs
(MRLs) for all VOCs; for SOCs trigger levels are specified that are essentially multi-method
practical quantitation limits that serve as MRLs for most States. lOCs are the most variable
because of the diversity of methods in use.
Analytical Result's Elements. The actual analytical results generally comprise multiple fields:
sometimes the sign of the results; sometimes the unit of measure; and the actual numerical value
of the analysis. One critical component of the results sign in such databases is how "
-------
A Review of Contaminant Occurrence in Public Water Systems
The other fields were carefully reviewed to ensure that all results were entered in (or were
converted to) the same units of measure. For this analysis, all results were processed and are
reported in mg/L (parts per million). State databases handled this in different ways, also. Some
recorded a results field using both mg/L and ug/L, or other units as appropriate; some States
recorded a uniform string of numerals with a code for where the decimal place should be located;
while some entered all data in the same units and had no record; and other States assumed that
certain contaminants were always entered in the same units (e.g., lOCs in mg/L, SOCs in ug/L).
Again, careful review was required to ensure correct units were used. In nearly all databases,
some small number of values appeared that were significant outliers, suggesting that a unit
conversion or original reporting problem was likely. In some cases, these could be confirmed
with the State and then eliminated. These data were rare and do not affect any general statistics
unless the maximum concentration value was of interest. In one case, for a few analytes, a State
knew aberrant data existed (because of a reporting error), and results greater than the 99th
percentile concentration were deleted.
H.B.2. System and Sample Elements
Public Water System Identifier (PWSID). Most States did use the SDWIS PWSID but some used
unique State identifiers, while others used both. In some cases, crosswalk tables had to be
constructed from other State databases to ensure consistency. Many databases also included the
system name, which was helpful for cross-checking problems with system IDs (e.g., when an ED
was entered incompletely or in error). In a few databases, elements were included identifying a
particular facility or treatment plant belonging to the PWS.
Source Water Type. The source of the raw water was identified in all datasets. The identification
was typically by a code for surface water, ground water, both, purchased, or other (and a myriad
of minor codes, unique to different States). For this analysis a primary objective was to
subdivide and summarize data by systems using surface water and ground water. Systems using
both, and systems using ground water under the influence of surface water, were included with
surface water data. Systems classified as purchased or consecutive, and those with no source
identified, or "other" sources were not included in the analysis. This was a small component of
the data.
Other System Features. A few States include other information in their databases that were used
in the analysis (or could be useful in future analyses). These included data on the population
served by the system, system type (CWS, NTNCWS), or other locational data such as latitude
and longitude.
Sample Identifiers. Sample IDs and other sample elements were also handled in a variety of
ways by the States. Many desired elements were included in some databases, including: a unique
sample ID (number); sampling point identifier (name or number for a specific site where a
sample was collected); sample location type (e.g., entry point to distribution system; distribution
line; well head); sample type (e.g., compliance, confirmation, duplicate, special, finished/treated,
raw, and many others); sample collection date; and sometimes date of analysis. Sample location
type and sample type were often a single field, with limited information. On the minimal side,
sometimes only the sample type and date of sampling were included. Date had to be combined
II
-------
A Review of Contaminant Occurrence in Public Water Systems
with PWSID to form a discrete sample ID to facilitate these analyses. The dates were also used
in conducting various temporal analyses and in correlating samples to assess co-occurrence of
contaminants.
ILC. Data Quality and Consistency Issues
There are a myriad of data quality issues involved in a project such as this. Many have been
noted in the preceding discussions. A few key issues are reviewed here in further detail.
As noted, every State database contained unique data elements or unique treatment of common
elements. Even after initial screening and conversion, unique factors were always uncovered
during data analysis. These were resolved in consultation with the State/source. As a general
rule, when errors or ambiguities in various data elements could not be resolved, the result was
eliminated from the analysis to avoid aberrant results. This eliminated very few observations
(compared to the thousands of analytical results included in the databases).
There are always concerns with the quality of the analytical data entered in such databases. This
study only used data from State primacy agencies -- official data from the regulated drinking
water program. Laboratories are certified for drinking water programs, which assumes the use of
various quality-control and quality-assurance procedures. Certainly problems exist, but the base
assumption is that the analytical data are of good quality.
When computing basic occurrence statistics, such as the number or percent of samples or systems
with detections of a given contaminant, the value of the minimum reporting level (concentration)
can have important consequences. The lower the reporting limit, the greater the number of
detections. Within reason, MRLs can even vary from lab to lab using the same method, or can
vary with sample batch, etc. There can be more dramatic variation if different methods are used
for the same contaminant. Within the drinking water program, methods have become well
standardized so this was not a significant issue for this study, particularly for the SOCs and
VOCs. For example, the VOCs EPA requires an MRL (MDL) of 0.5 vg/L. For SOCs, trigger
levels are specified that functionally have ben used as multi-method MRLs by many States. The
compounds most affected are the lOCs, where many more methods are available, with a much
greater variance in MRLs. (This will be discussed further in the review of results.) MRLs are
not specified in many State databases, as discussed. Hence it was not possible to "standardize"
the detections to a fixed MRL.
Another issue that can affect a large-scale summary of results, is the different sampling schedules
that may be used by different PWSs. A PWS with a known contaminant problem usually has to
sample more frequently than a PWS that has never detected the contaminant. Obviously, a
simple computation of the percentage of samples with detections (or other statistics) can be
skewed by the more frequent sampling at the contaminated site. This analysis is focused on
occurrence at the system level, which avoids the skewness inherent in the sample data,
particularly over the multi-year period covered in the summary statistics. Also, as noted above,
the data used in this analysis were standardized to use only data with confirmed water source and
sampling type information. Only standard SDWA compliance samples were used; "special"
samples, or "investigation" samples (investigating a contaminant problem that would bias
12
-------
A Review of Contaminant Occurrence in Public Waier Systems
results), or samples of unknown type were not used in the analysis. Various quality control and
review checks were made of the results, including review by the States providing the data. Many
of the most problematic data quality problems encountered occurred with older data (especially,
pre-1990-1993). In some cases, as noted, these were simply eliminated from the analysis.
Another data management issue involves short-hand codes used for the most fundamental data
element: the analytical results. Many States have special codes they enter for multiple
contaminants that are covered under the same analytical method. One of the most common
approaches summarizes the results of a single method with an 'ND' or zero (for all contaminants
not detected) and individual observations only for those contaminants with a positive result. This
system is best illustrated using the 21 VOCs covered under method 502.2 as an example. If none
of the 21 VOCs were detected, a State might enter '21 VOCs1 in the contaminant ID column and
'0' in the results column, as if it were a single observation. If one or more of the 21 VOCs were
detected, these individual contaminants would be entered in the contaminant ID column and the
concentration detected would be entered in the analytical results field as individual observations.
It then is assumed, in this case, that the other contaminants were tested for but not detected.
While this system works well for data storage, it does not work well for data analysis. To
statistically analyze the occurrence of contaminants a result is needed for every observation
(sample). Hence, special processing was required for every data set to unravel these codes.
When "21 VOCs" were 0, the contaminant ID and the 0 actually had to be entered in the dataset
for all 21 of the contaminants, to fill out the record. Likewise, when two of the contaminants
from the VOCs were entered, the processing had to recognize which two, and then fill out the
dataset for the additional 19. Although the example above may seem straightforward, the reality
of implementation was often more complicated. It was not unusual to find caveats in the
database which needed to be further defined by the State before analysis could begin (e.g.,
sometimes the rules for processing such entries changed over time). This is done differently
among States, and it is typically is not entirely consistent within a State dataset.
All of these controls and data editing were performed to make the results as reliable and
consistent as possible and to ensure that they were unambiguously from finished drinking water,
representing the product delivered to the public. Also, elimination of data with inconsistent
elements helps to ensure that the analysis is relatively repeatable, for future consideration.
II.D. URCIS Data Quality
As noted, for this study we also reviewed some results from the URCIS database. There are data
quality issues unique to URCIS. These are reviewed to put URCIS results, and their limitations,
in perspective. EPA's Unregulated Contaminant Monitoring Information System (URCIS) is a
compilation of public water system monitoring results for contaminants without NPDWRs (i.e.,
unregulated contaminants) that are also collected under the authority of SDWA (some of the
contaminants included in early rounds of this monitoring later became regulated contaminants
with NPDWRs). EPA collected these data from States and has been working on the data clean-
up and analysis since 1992. The version of the database used here was a current edition of data
(from 1997) selected for other on-going analyses. The values shown here for URCIS may be
somewhat different than for other EPA reviews because of the screening conducted for this study
13
-------
A Review of Contaminant Occurrence in Public Water Systems
to ensure consistency in analysis. The data includes VOC and some SOC results from 39 States
and the Virgin Islands.
Data from URCIS used in this analysis range in date from 1983 to 1992. The majority of the data
are from the first round of required, unregulated contaminant monitoring that began in 1987.
Because of the age of these data (in relationship to the rapid improvements that have been made
in data processing systems) the quality of data received by EPA for URCIS was highly variable.
The data were supplied by States on a variety of media, ranging from photocopies of hand-
written files to electronic files on magnetic tape or diskette, and in many different formats and
software configurations. Some data were electronically transferrable, other data had to be
manually entered or re-entered; legibility was occasionally a problem in entering the data,
introducing the possibility of transcription errors.
The sampling point designation in the original data is often unclear or omitted and many samples
are not identified as coming from the source water, an entry point, or elsewhere in the
distribution system. It is not always clear whether a sample was of raw or finished water. The
method of analyzing samples was not always reported or was reported incompletely or in a code
other than EPA method numbers. Not all samples had certification number for laboratories
conducting the analyses. In some cases, the system water source (surface or ground water) was
not included, but most of these elements have been completed by matching URCIS records with
SDWIS.
All of the original data entered into the URCIS database had the concentration units converted to
^g/L; some data sets, however, included data with mixed units (fj,g/L and mg/L) and the
conversion resulted in abnormally high concentrations for some observations. While the
observations with the highest concentrations were excluded from the analysis as suspect, there
may be observations remaining in the database with values up to three orders of magnitude
greater than the actual concentration. While these observations affect a review of the maximum
concentration values of a contaminant, and other parametric calculations (e.g., means) they will
have limited impact on the important occurrence statistics reviewed in this report. As with the
State SDWA data, entries without source water codes or other key elements, or those having
undecipherable codes, were eliminated from the analysis. There are other features of the data
that also must be understood.
Results from seven States in URCIS include only chemical analyses with positive detections.
Results from six other States are all between 88% and 99% detections; results for these States are
probably all from multi-analyte samples with a positive detection for at least one analyte. As a
result, detection rates in URCIS are biased toward a greater percentage of contaminant detections
than is actually the case. In addition, the full range of contaminants tested for is not discernable
for those States which reported mostly detections instead of all results. However, for States that
submitted complete data, the URCIS results are very consistent with the results from the
complete State SDWA databases collected for this study (for the analytes they have in common).
In spite of these shortcomings, URCIS is such a large database, with data from so many States,
that it has considerable value. The sheer quantity of data in URCIS somewhat compensates for
some of the quality problems, and the spectrum of coverage (40 States and Territories) warrant
review. Further, most of the data quality problems bias the data in the same, conservative
14
-------
A Review of Contaminant Occurrence in Public Water Systems
direction: the occurrence of contaminants is likely over-stated in both frequency and maximum
concentrations. Thus, aggregate values should not underestimate occurrence.
II.E. Basic Analysis and Review
As noted, the basic data manipulations and statistical analyses were all performed using SAS
programs. The basic data summaries output for each of the datasets included the following
statistics for each analyte (and some groups of analytes) by ground water and surface water:
• Number of samples
• Number and percentage of samples with reported values greater than the MRL
• Number and percentage of samples with a reported value greater than half the MCL
• Number and percentage of samples with a reported value greater than the MCL
• Number of systems (sampling for a given analyte)
• Number and percentage of systems with reported values greater than the MRL
• Number and percentage of systems with a reported value greater than half the MCL
• Number and percentage of systems with a reported value greater than the MCL
• The concentration of the median, 95th, and 99th percentile (in mg/L)
Note that a reported anayltical value greater than the MCL does not constitute an MCL violation.
An MCL violation occurs when the MCL is exceeded by the average of four quarterly samples or
confirmation samples as required by the primacy State. Occurrence values were calculated based
on the number of samples, primarily for evaluating the data. The results described in this report
are those calculated on a system basis (i.e. the percentage of systems with detections) to avoid the
bias and skew inherent in the sample data.
The results were reviewed and cross-checked with some duplicate programming. Results were
compared with published or unpublished analyses from the states, when possible. Summary
reports and data tables were returned to the States/sources for their review and comment.
ID. TOWARD A REPRESENTATIVE SAMPLE
As discussed, this evaluation is one of the first and most extensive analyses EPA has conducted
on the occurrence of chemical contaminants regulated under the Safe Drinking Water Act. One
primary purpose is to begin to provide better data to address issues related to occurrence and
monitoring of public drinking-water systems. Another objective, even at this initial stage of
analysis, is to provide an overview of occurrence patterns for the 64 regulated chemicals at the
national level. Currently, there is no complete record of analytical results collected under SDWA
that can be processed for a comprehensive national overview. EPA's Safe Drinking Water
Information System (SDWIS) maintains a variety of water system inventory and operation
information as well as compliance program information. For most contaminants, the only
analytical results filed in SDWIS are those related to violations of an MCL. The analytical
results from monitoring of the Phase rule chemicals, and most other contaminants, are stored in
the individual State databases. There has been no feasible way to access these data to construct a
15
-------
A Review of Contaminant Occurrence in Public Water Systems
nationally representative sample except, as undertaken here, through analyzing the individual
State datasets.
As noted, most of the data for this review were provided voluntarily by States. Obviously,
constructing a "representative" view from such data can be problematic, so additional data were
requested from a few States to broaden the spatial coverage of the analysis. The 14 States whose
data were reviewed are shown on Table HI (the 12 States whose data were analyzed in detail are
shown on Table II). While 14 of 50 States is a substantial sample, it is not necessarily
representative. The data from ail these States could simply be aggregated to compute a
composite occurrence value for a contaminant. However, the result would significantly over-
represent Midwestern "Combelt" States (i.e., the sample could contain Kansas, Missouri, Iowa,
Illinois, Indiana, and Ohio, over 40% of the States represented). Hence, various means were
evaluated to enable the construction of a cross-section from the available State databases that
would provide a reasonable first view of national occurrence.
There are many sophisticated statistical methods that can be applied to analyze limited (and
biased) data. However, at this early stage of evaluating SDWA occurrence data this did not seem
warranted. For initial analysis, we attempted to develop a simple approach, that would be clear
and repeatable, resulting in aggregate numbers that could be easily understood. When data from
more States can be compiled, more involved methods may be warranted, especially after some
initial determination of the variance for the different contaminants.
IILA. Representativeness
Two broad factors were considered in the assessment: geographic or spatial diversity — the wide
range of climatic and hydrologic conditions across the United States must be represented to some
degree; and pollution potential — the States should represent the range of the likelihood of
contaminant occurrence. Many past EPA studies have shown that some simple measures, such
as population (or population density) are valid indicators of pollution, because it is human
activity and land use — be it manufacturing or agriculture — that is the source of most
pollutants, particularly the organic chemicals. Various demographic or other factors were
evaluated as independent measures or indicators of pollution potential. In general,
manufacturing/industrial activity, as well as population density, are considered the major sources
of many VOCs (degreasers, solvents, petroleum compounds). Most SOCs are pesticides, and
agriculture is the largest user of these compounds. While lOCs have various uses in
manufacturing, they also occur naturally and ambient concentrations can be enhanced by mining
or other diffuse activities. Natural geologic sources of lOCs were not directly considered in the
assessment for representativeness, in part because whole States needed to be evaluated and such
sources are often localized. However, by including geographic or spatial coverage across the
Untied States as a factor (e.g., from Illinois to Montana) a range of geologic conditions are
inherently included (as well as the hydrogeologic and climatic variability).
16
-------
A Review of Contaminant Occurrence in Public Water Systems
More than thirty-five different factors that might be useful indicators of pollution in each State3
were considered in this study. These ranged from Census data on manufacturing, agriculture, and
population density, to indices such as EPA's Section 106 allocation factors or the 1991-1992
Green Index: A State by State Guide to the Nation's Environmental Health (prepared by the
Institute for Southern Studies). Two methods were considered for evaluating the States'
comparable pollution potential. The first was development of a singular numerical index,
incorporating factors such as manufacturing in the State, total pounds of chemicals released, and
pesticides used, into a comprehensive ranking for each State. However, such a ranking for all
sources requires various factors to be weighted, and the meaning of the resultant number can be
difficult to understand, as well as argumentative.
For this initial analysis, a second, simpler method was used to evaluate the pollution potential
and the representativeness of the States. One primary factor was chosen to indicate the potential
pollution from manufacturing and one factor for agriculture in each State. States were then
ranked from 1 -50 for each factor and divided into quartiles based on the ranking. The rankings
were reviewed to assess if States could be selected in approximate balance from each quartile. In
addition, a number of secondary factors were also considered to further insure that the data were
representative. This method does not, of course, avoid all of the problems discussed above, but it
does provide a simple look at the conditions represented by the States. The resultant rankings are
summarized on Table ffl; the factors considered are discussed briefly, below.
HI.A. 1. Manufacturing Indicators
Numerous factors were considered as potential indicators of manufacturing-related pollution,
including EPA's Toxic Release Inventory (TRI) (including total releases, releases per square
mile, and releases excluding air releases), the number of manufacturing establishments, the
number of manufacturing establishments per square mile, the number of manufacturing
employees, the value added by manufacturers, and the value added per capita (Annual Survey of
Manufacturers, 1995; Census of Manufacturers, 1992; and Toxic Release Inventory, 1995).
These factors were each considered in terms of their inherent value as pollution potential
indicators, their range and variance (in providing a relative ranking of the States), and their inter-
relationships.
The total TRI releases per square mile, number of manufacturing establishments per square mile,
and value added per capita were considered the three most useful indicators. The TRI was
considered useful because it is a measure of how many pounds of toxic chemicals are released
within the State. While there are problems with the TRI (e.g., some inconsistent release
estimation techniques; omission of many small establishments, or those with releases below
specified thresholds), it can validly be used as a direct indicator of potential pollutants released.
The number of manufacturing establishments per square mile takes into account how many
factories are actually engaged in manufacturing and thus how many establishments potentially
3The data are analyzed on a statewide basis, thus any determination of representativeness must be based on
whether the States for which information is available are representative of the nation as a whole. There are, of course,
problems with using States to determine representativeness since States are large, diverse entities; however it is not
practical to break the data down any further.
17
-------
A Review of Contaminant Occurrence in Public Water Systems
contribute to pollution. By breaking down the number of manufacturing establishments per
square mile, the size of the State is also taken into account. The final factor that was considered
to be viable was the value added by manufacturers per capita. This is an attractive measure
because the value added should correlate well with how much is actually produced within the
State. The problem with the measure, and this is also a problem with the number of
manufacturing establishments per square mile, is that it does not take into account the variation
in pollution released by different industries. For example, an industry that adds a lot of value to a
product may cause little pollution while another industry that does not add much value may
contribute more pollution.
The data clearly show a close correlation between the number of manufacturing establishments
per square mile and the population density in each State, as well as a clear linear association with
the total TRI pounds released/square mile, number of manufacturing employees, and total value
added. Hence, the number of manufacturing establishments per square mile was used as the
primary indicator. The other key reason for choosing this factor was that it is a simple measure
of how many establishments are actually engaged in manufacturing and thus are potentially
polluting sources of drinking water. The TRI total pounds released per square mile was used as a
secondary factor in determining representativeness. Squill ace and others (1999), from the USGS,
found a significant correlation between VOC occurrence in ambient ground water and population
density, in a national NAWQA study. As noted, population density and manufacturing density
are highly correlated. Manufacturing density and TRI data were used in this ranking because
they were considered more direct measures of pollution potential for this study.
ffi.A.2. Agricultural Indicators
There is no complete measure of pesticide usage by States that is readily available. So, a variety
of factors were considered to assess potential organic chemical pollution from agriculture in each
State. These included the percent of the State's population that is classified as rural, the percent
of land in the State that is cropland, the percent of land that is grassland pasture and rangeland (a
possible inverse indicator), and total farm agricultural chemical expenses. Like the
manufacturing factors, these agricultural variables were considered in terms of their value in
indicating potential sources of pollution and were plotted against one another to determine how
closely they are correlated.
Of these factors, total farm agricultural chemical expenses was considered to be the best indicator
of potential pollution. The percent of the State's population that lives in rural areas does not
necessarily relate to agricultural chemical use or cropland. There is, of course, a correlation
between cropland and agricultural chemical use, but there are notable exceptions such as Florida
and California which use a large amount of agricultural chemicals despite having more limited
cropland area. While there are some incomplete surveys of pesticide use, the Census of
Agriculture (1992) measure of dollars spent on agricultural chemicals, is a more consistent and
complete measure.
is
-------
A Review of Contaminant Occurrence in Public Water Systems
III.B. Representative Cross-Section of States
Table ffl summarizes the quartile rankings of the States selected to approximate the national
cross-section used in this summary report. Eight States were selected, from the 14 whose data
were evaluated, as providing the best balanced cross-section, based on geographic coverage,
relative rankings for pollution potential (i.e., potential for contaminant occurrence), and data
quality and completeness. These States provide representation from across the U.S., from
Alabama to Oregon to New Jersey. They provide some representation from all quartiles of
pollution potential indicators, and, hopefully, a balance in potential occurrence. The data from
these eight States were used to compute aggregate occurrence values (i.e., the percentage of
water systems that had a detection of contaminant X) as an approximation of a national cross-
section. For balance, all the SDWA data (see Table H) were used to show the range of
occurrence in the results summaries. However, the ranges for only a few of the 64 contaminants
were extended by the additional data. While the data from these Cross-Section States cannot be
stated to be "statistically representative," their distribution should provide a clear indication of
central tendency. The results will be compared with other data for perspective and as a
comparative check. It is also worthy of note that the 8 Cross-Section States represent over 25%
of the U.S. population using PWSs, and over 20% of the PWSs, a substantial sample.
(Considering the other States evaluated and used in the range estimates, the data include over
40% of the population and nearly one-third of PWSs.) Again, if additional State databases can
be added to this analysis, other statistical techniques can be utilized to develop national estimates
of occurrence.
19
-------
A Review of Contaminant Occurrence in Public Water Systems
Table III. States with water quality data included in the drinking water occurrence assessment.
States are listed by their national rank order (l=highest, 50=Iowest) for the number of
manufacturing establishments per square mile. The rank for TRI releases in pounds per square
mile is in parentheses. Also listed is the State's rank for total expenditures for agricultural
chemicals. States on the left are those selected for the national cross-section compilation; States
on the right are others whose data were evaluated and are included in the data ranges
summarized.
Quartiies
for rank-
order of
all States
1
2
3
4
National
Cross-
Section
States
New Jersey
Illinois
California
Michigan
Alabama
Oregon
New
Mexico
Montana
National Rank:
Number of
Manufacturing
Facilities per
square mile
2(8)
10(11)
11(38)
13(16)
25(7)
34 (39)
44(40)
48 (34)
National Rank:
Total
Expenditures,
Agricultural
Chemicals
37
2
1
18
26
22
40
34
Other States;
Data Used for
Ranges and
Check on
Cross-Section
Massachusetts
Ohio
Indiana
Missouri
Iowa
Kansas
National Rank:
Number of
Manufacturing
Facilities per
square mile
4(22)
6(2)
15(6)
26 (26)
36 (28)
39 (37)
National
Rank: Total
Expenditures,
Agricultural
Chemicals
43
11
7
12
3
16
IV. THE SDWA OCCURRENCE DATA
In the following sections of this report we summarize the results of the analysis of the drinking
water contaminant data. The summary discussions will focus on basic occurrence patterns, and
discuss the percentage of water supply systems in the national cross-section (or individual states,
to illustrate particular points) that had detections of a given contaminant greater than the MRL
(>MRL; simply put, systems with detections), systems with detections greater than 50% of the
MCL (>0.5MCL), and systems with detections greater than the value of the MCL (>MCL).
In this study, data were analyzed for approximately 280 different contaminants. This includes
the 64 contaminants regulated under the Phase rules (Table LA. 1), six other regulated
contaminants (e.g., nitrate, lead and copper), 34 unregulated contaminants that were formally
required for monitoring under past regulations, 14 contaminants that were discretionary for
monitoring under these regulations, and approximately 160 other unregulated contaminants. Of
these other unregulated contaminants, 22 are on EPA's Contaminant Candidate List, including 16
that appear on the proposed new unregulated contaminant monitoring list. The remaining 140
are other unregulated contaminants that were contained in the State databases. Beyond the
20
-------
A Review of Contaminant Occurrence in Public Water Systems
contaminants that were required for monitoring under SOW A, the data are very incomplete,
sometimes consisting of only a single analysis. Of the 10.7 million analytical records processed,
about 73% were for required monitoring.
As noted earlier, data were processed from approximately 26,000 water systems. This does not
mean that there are data from 26,000 systems for each contaminant. As discussed, not every
State provided data for all systems or all contaminants. Even within a State, some systems may
have waivers for particular contaminants or groups of contaminants. The maximum number of
systems for a particular contaminant is approximately 16,000 for many VOCs and SOCs. For the
regulated contaminants the number of analyses varies from about 50,000 to 150,000 for each
contaminant.
The narrative summary will not attempt to discuss results for all 280 contaminants or even the 64
Phase rule chemicals. The summary will focus on illustrating key issues and answering key
questions. Most of the summary tables will focus on a few representative IOC, SOC, and VOC
contaminants. (More complete summary tables are presented in Appendix A.) In addition to
data for individual contaminants, summary data will be presented by contaminant groups.
IV.A. Contaminant Groups
The group classifications are partly related to the nature and source of the contaminants and
partly to their chemical properties and general method of analysis. The aggregate groups provide
additional perspective because many of the compounds are inter-related by source or fate and
transport characteristics. For this analysis, the occurrence of any contaminant in the group is
counted system by system. A system may detect one or all contaminants in the group, but is only
counted once. The analysis by contaminant groups provides an indication of the aggregate
occurrence and some insight into co-occurrence. The contaminants are grouped by their standard
categories, lOCs, SOCs, and VOCs, and also into smaller categories.
FV.A.1. SOC Groups
The SOCs are divided into high-occurrence SOCs (HiSOCs) and low-occurrence SOCs
(LoSOCs). The HiSOCs and LoSOCs Groups were used for the impact analysis estimates of the
draft CMR. SOCs Groups 1,2, and 3 are further subdivisions that will be used for the current
analysis. SOCs Group 1 aggregates the data for the commonly occurring herbicides that often
dominate the occurrence of HiSOCs; these are alachlor, atrazine, and simazine. These three
contaminants are often analyzed from the same sample (using a multi-analyte method). SOCs
Group 2 includes adipate, phthalate, and benzo(a)pyrene, which vary in occurrence. In some
studies, phthalate and adipate may have high rates of occurrence, but this has, in part, been
attributed to sample contamination resulting from the use of plastic piping or even plastic
sampling and lab equipment. SOCs Group 3 includes the remainder of the SOCs (with the
exception of the PCBs and Dioxin). These tend to be either low or localized in occurrence,
related to the contaminant's chemical properties and regional use patterns.
21
-------
A Review of Contaminant Occurrence in Public Water Systems
IV.A.2. VOC Groups
While all of the 21 currently regulated volatile organic compounds (VOCs) are similar (i.e. they
evaporate readily), it can be useful to differentiate between two smaller groups of VOCs: VOCs
Group 1 and Group 2. While not all of the 21 regulated compounds are included in these
subclassifications, distinguishing between these two groups is very helpful, as many of the
compounds within each group have similar environmental sources and fates.
The Group 1 VOCs include the lighter, non-halogenated compounds: benzene, toluene,
ethylbenzene, and the xylenes. Aside from having many structural similarities (all contain a
benzene ring substructure), all of these compounds are either natural components or derivatives
of crude oil (Howard, 1997). Thus, major environmental sources of these compounds are
gasoline and other petroleum fuels. Other potential sources for Group 1 VOCs are industrial
solvents, carriers or solvents in pesticides and paints (Howard, 1997). Because of their low water
solubilities (ranging from 0.16 to 1.78 g/L at 25 °C; Bloemen and Bums, 1993) and low densities
(ranging from 0.86 to 0.88 g/cm3 at 20°C; Lide, 1996), these compounds often form light non-
aqueous phase liquids (LNAPLs) which tend to "float" on the water surface.
As gasoline and other petroleum fuels are common sources of Group 1 VOCs related to fuel
spills and surface run-off, these compounds are more likely to be found in surface waters than in
ground waters. Because Group 1 VOCs share a potential source, the likelihood of Group 1
compounds co-occurring with other Group 1 compounds is greater than with other regulated
VOCs, as none of the other 17 regulated VOCs are commonly found in gasoline (Howard, 1997).
In surface waters, there is often a pattern of frequent Group 1 VOC occurrence, but at low
concentrations. Also, as "floaters", when they occur in ground water they tend to be found more
often in shallow ground water than in deeper ground water (Hallberg et al., 1996). It is possible
that this pattern is at least partially attributable to the ubiquity of their environmental sources
coupled with the fact that they form LNAPLS, which facilitates their volatilization, and leads to
lower concentrations in surface waters. Also, these lighter compounds are more subject to
biological degradation and there is greater residence time in the soil-ground-water system
allowing for biological processing.
The chemicals which compromise the Group 2 VOCs are the heavier, halogenated organic
compounds. The six compounds included in this group are tetrachloroethylene (PCE),
trichloroethylene (TCE), cis and trans 1,2,-dichloroethylene, 1,1,-dichloroethylene, and vinyl
chloride. As with Group 1 compounds, all of these compounds are structurally similar: they all
have a double-bonded, two-carbon-chain backbone, and all contain at least one chlorine atom.
None of the Group 2 VOCs are known to occur naturally. While many of these compounds are
used as metal degreasers and as industrial solvents, some may also be used for dry cleaning
(PCE: Howard, 1997; TCE: Parsons, et al., 1984), in paint and ink formation (TCE; Howard,
1997), and in synthetic polymers such as PVC (vinyl chloride; Howard, 1997) and plastic wrap
(1,1-dichloroethylene; Howard, 1997). As with Group 1, Group 2 VOCs tend to form non-
aqueous phase liquids because of their low water solubilities. However, with the exception of
vinyl chloride, Group 2 compounds tend to "sink" in the water column because of their higher
densities (ranging from 1.21 to 1.62 g/cm3 at 20°C; Lide, 1996), and thus are classified as dense
non-aqueous phase liquids (DNAPLs). When present in ground water, this behavior provides
22
-------
A Review of Contaminant Occurrence in Public Water Systems
limited access to the air/water interface, and thus these compounds may tend to volatilize less
than Group 1 VOCs.
Although each Group 2 compound may have a discrete environmental source, many of the
compounds are components of the PCE degradation series. Upon its release into the
environment, PCE may undergo microbial degradation, losing chlorine atoms one by one and
forming first trichloroethylene (Parsons et al., 1984), then the dichloroethylenes (cis- and trans-
1,2-dichloroethylene: Parsons et al., 1984; 1,1-dichloroethylene: Howard, 1997), and eventually
vinyl chloride (Howard, 1997). Additional degradation paths are possible, some eventually
leading to carbon tetrachloride and dichloromethane, two other regulated VOCs.
Because of their potential sources in the environment, and because they tend to "sink," Group 2
VOCs are more likely to be found in ground waters than in surface waters. Like Group 1 VOCs,
these compounds may be more likely to co-occur with each other than with other regulated
VOCs, not because of their common potential source, but because they are part of the same
degradation series. Co-occurrence in this case may be the occurrence of two or more distinct
compounds in the same location but at different times (i.e., spatial but not temporal co-
occurrence). For example, PCE might be found in ground water taken from a well subject to
local PCE contamination. Over time, the level of PCE might diminish or disappear altogether,
while the concentrations of TCE, the dichloroethylenes, and vinyl chloride, which might have
been previously undetected, could rise due to microbial degradation.
There are other VOCs which share potential environmental sources, and thus might also co-
occur. For example, 1,2-dichloroethane is a common additive to leaded gas, and so it might co-
occur with the Group 1 VOCs if the major contaminant source contained leaded gasoline.
Additionally, 1,1,1-trichloroethane may be transformed, via hydrolysis or microbial degradation,
to 1,1,-dichloroethylene (Howard, 1997), which is part of the PCE degradation series. However,
these associations are not as strong as the associations between the constituents of Group 1 and
Group 2.
IV.B. Overview of Results
If there is a generalization that can be made about contaminant occurrence in drinking water
systems, it is that occurrence is highly variable. It is important to note that all 64 contaminants
regulated under the Phase rules (Table LA. 1) have been detected in drinking water systems. Only
five of the 64 contaminants do not have a reported detection at >0.5 their MCL. They are all
SOCs: glyphosate, hexachlorobenzene, methoxychlor, oxamyl, and 2,4,5-TP. All other lOCs,
SOCs, and VOCs have detections >0.5 MCL.
The frequency of occurrence in water systems ranges widely. All lOCs, except asbestos, have
been detected in 90 -100% of systems in some States. Only three lOCs (antimony, beryllium,
and cyanide) have not been detected in at least one system in at least one State (in either surface
or ground water). The high occurrence rates are not surprising for lOCs because they occur
naturally. Occurrence and concentration of lOCs can vary greatly related to natural geologic
sources (sometimes enhanced by mineral extraction activities). Often such sources, and effects,
are localized and are beyond the scope of this analysis.
23
-------
A Review of Contaminant Occurrence in Public Water Systems
Twenty-nine of the 30 regulated SOCs have not been detected at all in some States; only
phthaJate occurs in every State, but its high occurrence relates, in part, to sample contamination
from plastics used in sampling or lab work (or system plumbing). Because of this problem,
phthalate is not included among some of the other summary values presented. (Dioxin is not
included among these summary values either, because dioxin only rarely has been included in
monitoring.) The greatest range for any of the organic chemicals is for atrazine, which ranges
from zero systems in some States to >97% occurrence in surface water systems in Midwestern
States (up to 9% in ground water systems).
Fourteen of the 21 VOCs have not been detected in some of the participating States. The seven
that occur in every State studied, in either surface or ground water systems, are: ethylbenzene,
cis-l,2-dichloroethane, tetrachloroethyJene (PCE), trichloroethylene (TCE), vinyl chloride, 1,1,1-
trichlorethane, and xylenes. Many VOCs occur in up to, or more than, 30% of surface or ground
water systems in various States.
Only nine of the 64 contaminants occur in fewer than 1% of either surface water or ground water
systems in all States. These are all SOCs: alachlor, carbofuran, chlordane, glyphosate,
hexachlorobenzene, hexachloropentadiene, oxamyl, toxaphene, and PCBs (though PCBs have
not been monitored as intensively as the other SOCs). In contrast, the five compounds which
occur most frequently at concentrations greater than the value of their MCL (which does not in
itself constitute an MCL violation) are shown in Table IV.B.l. (As noted, phthalates and dioxin
are excluded.)
Several observations are warranted. First, MCL exceedances are not common, even for this top-
five list. The compounds with the most common exceedances are a mix of VOCs and SOCs.
Fluoride makes the ground water list, but fluoride detections tend to be a result of natural
occurrence and treatment additions. Three of the compounds occur both on the surface water and
ground water list, suggesting their widespread and common impact. The data in Table IV.B. 1
illustrate a general trend: exceedances for SOCs tend to be more common for surface water
systems; for VOCs, exceedances tend to be more common for ground water systems (though the
differences from surface water are not great; see Table IV.C.1-3, also). (It should be noted,
however, that the detection of methylene chloride and EDB can be confounded with other
organic compounds, and it is suspected that the occurrence values reported may be too great.)
24
-------
A Review of Contaminant Occurrence in Public Water Systems
Table IV.B.1. Five Phase rule contaminants which occur most frequently at concentrations
greater than their MCL in either surface water or ground water systems. The five contaminants
are those with the greatest percentage of systems showing exceedances in the national cross-
section.
Surface- Water Supplied
Systems
Contaminant
atrazine
EDB
methylene chloride
PCE*
TCE*
% Systems
with MCL**
exceedances
10.7%
3.7%
4.7%
1.7%
1.2%
Ground-Water Supplied
Systems
Contaminant
DBCP
methylene chloride
PCE*
TCE*
1,1,1 -trichloro- ethane; and fluoride
% Systems
with MCL**
exceedances
2.0%
2.3%
1.8%
1.5%
1.3%
*PCE- tetrachloroethylene; TCE- trichloroethylene
** % Systems with MCL exceedances indicates the proportion of systems with any analytical results exceeding the
concentration value of the MCL; it does not necessarily indicate an MCL violation. An MCL violation occurs when
the MCL is exceeded by the average results from four quarterly samples or confirmation samples as required by the
primacy State.
IV.C. Contaminant Occurrence in Surface Water and Ground Water Systems
Developing a further understanding of the differences in occurrence patterns, and the
susceptibility or vulnerability of a water supply to contamination, is a primary objective of this
work. Understanding the vulnerability of water supplies and water systems—the presence or
probability of contaminant occurrence—is a key to developing cost-effective monitoring
programs. Recognition of the differences between surface water and ground water systems has
always been a basic element in developing monitoring strategies and protecting public health.
Surface water systems are, in essence, vulnerable to any contaminant with a source in the
upstream watershed, but they have the potential factor of dilution working for them to dampen
impacts. Ground water sources are typically vulnerable to a smaller number of sources (those on
flow paths leading to the supply wells) but have less capacity for dilution. As noted above, there
are basic differences in occurrence patterns between these water sources, but these differences
vary by contaminant. For example, 14% of the regulated SOCs are detected in more than 5% of
surface water systems in the national cross-section. In contrast, no SOCs occur in more than 5%
of ground water systems. Nearly two-thirds of SOCs are detected in >1% of surface water
systems, but over three-fourths of these SOCs are detected in only a very few (<1%) ground
water systems. For VOCs, the occurrence patterns reverse to some degree, but VOCs are more
common than most SOCs in both surface and ground water systems. VOCs are detected in
surface water systems more frequently than is generally recognized.
25
-------
A Review of Contaminant Occurrence in Public Water Systems
Tables IV.C. 1 through IV.C.3 summarize occurrence patterns for select Phase H/V contaminants
and the contaminant groups. The individual contaminants represent a range of occurrence
patterns within each contaminant group, e.g., SOCs ranging from atrazine to glyphosate. Tables
JTV.C. 1 and 2 summarize the national cross-section occurrence values, with ranges shown from
all State data, for surface and ground water. Table FV.C.3 shows a comparison of national cross-
section values for surface and ground water. lOCs are about equally common (detected >MRL)
in surface water systems and ground water systems, but ground water systems have higher
concentrations, i.e., a greater proportion of systems exceeding 0.5 MCL and the MCL. SOCs are
far more common in surface water systems than ground water systems and surface water systems
have more exceedances as well. (The Group 2 SOCs, which have more equal occurrence in
surface water systems and ground water systems, include the phthalates and their attendant
problems.) VOCs show much greater general occurrence in surface water systems but surface
water systems and ground water systems are nearly equal for the occurrence of exceedances.
Whether a greater proportion of surface water systems or ground water systems show detections
varies by individual VOC or group. In total, ground water systems show slightly more MCL
violations for VOCs (Table FV.C.4).
26
-------
A Review of Contaminant Occurrence in Public Water Systems
Table IV.C.l. Summary of occurrence of selected Phase n/V contaminants in water systems
using surface water, from national cross-section States; ranges from all States studied. Percent
MCL violations derived from SDWIS information for 1/1/93-3/31/1998.
Contaminant
Percent > MRL
Cross-
Section
Range
(All States)
Percent > l/z MCL
Cross-
Section
Range
(All States)
Percent > MCL*
Cross-
Section
Range
(All States)
National
Percent MCL
Violations
Individual Contaminants (in surface water systems)
lOCs
Barium
Cadmium
Mercury
SOCs
2.4-D
Atrazine
Benzo[a]pyrene
Carbofuran
(Furadan)
Ethylene Dibromide
(EDB)
Glyphosate
(Roundup)
Simazine
VOCs
Benzene
Styrene
Tetrachloroethylene
Vinyl Chloride
Xylenes (Total)
49.1%
5.1%
9.0%
11.2%
21.1%
0.5%
0.8%
4.2%
0.0%
15.9%
3.9%
4.1%
7.1%
3.1%
12.3%
22.1%- 100.0%
0.0%- 100.0%
0.0%- 100.0%
0.0% - 50.5%
0.0%- 97.1%
0.0% - 3.7%
0.0% - 1.8%
0.0% - 9.4%
0.0% - 0.9%
0.0% - 67.0%
0.0% - 17.1%
0.0% - 14.5%
0.0% - 16.2%
0.0% - 13.7%
1.9%- 29.1%
Group Summaries
lOCs
lOCs-Regulated
SOCs
SOCs-Group 1
SOCs-Group 2
VOCs
VOCs-Regulated
VOCs-Group 1
VOCs-Group 2
83.7%
21.9%
20.4%
41.1%
19.5%
11.1%
69.7%- 100.0%
0.0%- 97.1%
0.0% - 88.9%
8.3% - 89.7%
2.1%- 42.9%
0.0% - 30.3%
0.6%
1.3%
1.3%
0.2%
13.2%
0.0%
0.0%
3.8%
0.0%
2.5%
0.5%
0.0%
2.5%
0.3%
0.2%
0.0%- 6.1%
0.0%- 9.1%
0.0%- 6.1%
0.0%- 0.6%
0.0% - 62.4%
0.0%- 1.9%
0.0% - 0.0%
0.0%- 8.5%
0.0%- 0.0%
0.0%- 11.9%
0.0% - 3.0%
0.0%- 0.0%
0.0%- 15.2%
0.0%- 3.0%
0.0%- 1.9*
0.5%
0.2%
0.5%
0.0%
10.7%
0.0%
0.0%
3.7%
0.0%
1.0%
0.3%
0.0%
1.7%
0.3%
0.0%
0.0% - 3.0%
0.0%- 1.9%
0.0%- 1.9%
0.0%- 0.0%
0.0%- 51.4%
0.0% - 0.0%
0.0% - 0.0%
0.0% - 8.5%
0.0%- 0.0%
0.0% - 4.6%
0.0%- 3.0%
0.0%- 0.0%
0.0%- 15.2%
0.0% - 3.0%
0.0%- 0.0%
6.8%
13.2%
1.2%
15.4%
1.7%
4.5%
0.0%- 92.6%
0.0%- 62.4%
0.0%- 5.6%
0.0%- 60.6%
0.0%- 12.1%
0.0%- 30.3%
2.5%
10.7%
0.9%
8.2%
0.9%
2.9%
0.0%- 6.1%
0.0%- 51.4%
0.0%- 2.8%
0.0%- 57.6%
0.0%- 6.1%
0.0% - 30.3%
0.0%
<0.1%
<0.1%
0.0%
0.8%
0.0%
0.0%
0.1%
0.0%
0.0%
<0.1%
0.0%
0.1%
0.0%
0.0%
0.3%
1.0%
0.2%
* % > MCL indicates Che proportion of systems with any analytical results exceeding the concentration value of the MCL; it does not necessarily
indicate an MCL violation. An MCL viotation occurs when the MCL is exceeded by the average results from four quarterly samples or
confirmation samples as required by the primacy State.
IOC-Regulated: includes all the regulated lOCs.
SOCs-Group 1: includes alachlor. atrazine, and simazine.
SOCs-Group 2: includes bis(2-ethylhexyl)phthalate, bis(2-ethylhexyl)adipate. and benzo{a)pyrene.
VOCs-Regulated: includes all the regulated VOCs
VOCs-Group I: includes benzene, ethyl benzene, toluene, and total xylenes (LNAPLs).
VOCs-Group 2: includes cis-1,2-dichtoroethylene. trans-1.2-dichloroethylene. 1,1-dichloroethene, tetrachloroethylene, trichloroethene, and vinyl
chloride (DNAPLs).
27
-------
A Review of Contaminant Occurrence in Public Water Systems
Table IV.C.2. Summary of occurrence of selected Phase n/V contaminants in water systems
using ground water, from national cross-section States; ranges from all States studied. Percent
MCL violations derived from SDWIS information for 1/1/93-3/31/1998.
Contaminant
Percent > MRL
Cross-
Section
Range
(AU States)
Percent >»/2 MCL
Cross-
Section
Range
(All States)
Percent > MCL*
Cross-
Section
Range
(AH States)
National
Percent MCL
Violations
Individual Contaminants (in ground water systems)
lOCs
Barium
Cadmium
Mercury
SOCs
2,4-D
Atrazine
Benzo[a]pyrene
Carbofuran
(Furadan)
Ethylene Dibromide
(EDB)
Glyphosate
(Roundup)
Simazine
VOCs
Benzene
Styrene
Tetrachloroethylene
Vinyl Chloride
Xylenes (Total)
47.3%
4.9%
4.5%
1.2%
2.0%
0.5%
0.1%
1.0%
0.1%
1.4%
1.2%
2.1%
4.3%
0.5%
3.9%
31.5%- 98.6%
0.4%- 93.9%
0.4%- 93.3%
0.0% - 4.7%
0.0% - 9.2%
0.0% - 2.9%
0.0%- 0.1%
0.0% - 2.6%
0.0% - 0.5%
0.0% - 3.2%
0.0% - 4.2%
0.0%- 8.1%
0.6%- 14.2%
0.0% - 3.0%
0.4%- 15.3%
0.8%
1.2%
0.7%
0.0%
0.3%
0.1%
0.0%
1.0%
0.0%
0.0%
0.5%
0.2%
2.3%
0.2%
0.2%
0.0%- 6.9%
0.0%- 21.7%
0.0%- 3.2%
0.0%- 0.1%
0.0% - 2.0%
0.0%- 0.3%
0.0%- 0.1%
0.0%- 4.2%
0.0%- 0.0%
0.0%- 0.1%
0.0%- 1.8%
0.0%- 0.9%
0.1%- 7.6%
0.0%- 0.6%
0.0%- 1.0%
0.2%
0.6%
0.4%
0.0%
0.1%
0.1%
0.0%
0.7%
0.0%
0.0%
0.4%
0.2%
1.8%
0.2%
0.1%
0.0%- 3.1%
0.0%- i.5%
0.0%- 2.0%
0.0%- 0.1%
0.0%- 0.5%
0.0%- 0.3%
0.0%- 0.0%
0.0%- 1.9%
0.0%- 0.0%
0.0%- 0.0%
0.0%- 1.8%
0.0%- 0.9%
0.0%- 5.7%
0.0% - 0.6%
0.0%- 0.8%
0.1%
0.1%
<0.1%
<0.1%
<0.1%
0.0%
0.0%
<0.1%
0.0%
0.0%
<0.1%
0.0%
0.1%
<0.1%
0.0%
Group Summaries
lOCs
lOCs-Regulated
SOCs
SOCs-Group 1
SOCs-Group 2
VOCs
VOCs-Regulated
VOCs-Group 1
VOCs-Group 2
83.5%
2.4%
L 13.4%
19.9%
6.6%
6.4%
73.5%- 100.0%
0.0% - 9.2%
0.0%- 54.6%
6.2%- 65.5%
2.0%- 28.6%
2.1%- 21.1%
9.2%
0.3%
1.9%
7.9%
1.3%
3.9%
0.0% - 58.9%
0.0%- 2.0%
0.0%- 9.1%
1.3%- 28.7%
0.0%- 6.2%
0.9%- 12.9%
4.2%
0.1%
1.0%
6.1%
0.9%
3.2%
0.0%- 6.2%
0.0%- 0.6%
0.0%- 9.1%
0.7% - 27.9%
0.0%- 4.4%
0.0%- 12.9%
0.6%
0.1%
0.4%
*%> MCL indicates the proportion of systems with any analytical results exceeding the concentration value of the MCL; it does not necessarily
indicate an MCL violation. An MCL violation occurs when the MCL is exceeded by the average results from four quarterly samples or
confirmation samples as required by the primacy State.
IOC-Regulated: includes all the regulated lOCs
SOCs-Group 1: includes alachlor. atrazine, and simazine.
SOCs-Group 2: includes bis(2-ethy!hexyl)phthalate, bis(2-ethylhexyl)adipate, and benzo(a)pyrene.
VOCs-Regulated: includes all the regulated VOCs
VOCs-Group 1: includes benzene, ethyl benzene, toluene, and total xylenes (LNAPLsj.
VOCs-Group 2: includes cis-l,2-dichloroethylene, trans-t,2-dkhloroethylene, 1,1-dichloroethene, tetrachloroethylene, trichloroetbene, and vinyl
chloride (DNAPLs).
28
-------
A Review of Contaminant Occurrence in Public Water Systems
Table IV.C.3. Summary comparison of occurrence of selected Phase n/V contaminants in water
systems using surface water vs. ground water, from national cross-section States. Percent MCL
violations derived from SDWIS information for 1/1/93-3/31/1998.
Contaminant
Percent > MRL
Surface
Water
Ground
Water
Percent > Vz MCL
Surface
Water
Ground
Water
Percent > MCL*
Surface
Water
Ground
Water
National %
MCL Viol.-
SW
National %
MCL Viol.-
GW
Individual Contaminants
lOCs
Barium
Cadmium
Mercury
SOCs
2,4-D
Atrazine
Benzo[a]pyrene
Carbofuran
(Furadan)
Ethylene Dibromjde
(EDB)
Glyphosate
(Roundup)
Simazine
VOCs
Benzene
Styrene
Tetrachloroethylene
Vinyl Chloride
Xylenes (Total)
49.1%
5.1%
9.0%
11.2%
21.1%
0.5%
0.8%
4.2%
0.0%
15.9%
3.9%
4.1%
7.1%
3.1%
12.3%
47.3%
4.9%
4.5%
1.2%
2.0%
0.5%
0.1%
1.0%
0.1%
1.4%
1.2%
2.1%
4.3%
0.5%
3.9%
0.6%
1.3%
1.3%
0.2%
13.2%
0.0%
0.0%
3.8%
0.0%
2.5%
0.5%
0.0%
2.5%
0.3%
0.2%
0.8%
1.2%
0.7%
0.0%
0.3%
0.1%
0.0%
1.0%
0.0%
0.0%
0.5%
0.2%
2.3%
0.2%
0.2%
0.5%
0.2%
0.5%
0.0%
10.7%
0.0%
0.0%
3.7%
0.0%
1.0%
0.3%
0.0%
1.7%
0.3%
0.0%
0.2%
0.6%
0.4%
0.0%
0.1%
0.1%
0.0%
0.7%
0.0%
0.0%
0.4%
0.2%
1.8%
0.2%
0.1%
0.0%
<0.5%
<0.1%
0.0%
0.8%
0.0%
0.0%
0.1%
0.0%
0.0%
<0.1%
0.0%
0.1%
0.0%
0.0%
0.1%
0.1%
<0.1%
<0.1%
<0.1%
0.0%
0.0%
<0.1%
0.0%
0.0%
<0.1%
0.0%
0.1%
<0.1%
0.0%
Group Summaries
IOCS
lOCs-Regulated
SOCs
SOCs-Group 1
SOCs-Group 2
VOCs
VOCs-Regulated
VOCs-Group 1
VOCs-Group 2
83.7%
21.9%
20.4%
41.1%
19.5%
11.1%
83.5%
2.4%
13.4%
19.9%
6.6%
6.4%
6.8%
13.2%
1.2%
15.4%
1.7%
4.5%
9.2%
0.3%
1.9%
7.9%
1.3%
3.9%
2.5%
10.7%
0.9%
8.2%
0.9%
2.9%
4.2%
0.1%
1.0%
6.1%
0.9%
3.2%
0.3%
1.0%
0.2%
0.6%
0.1%
0.4%
* % > MCL indicates the proportion of systems with any analytical results exceeding the concentration value of the MCL; it does not necessarily
indicate an MCL violation. An MCL violation occurs when (he MCL is exceeded by the average results from four quarterly samples or
confirmation samples as required by the primacy State.
IOC-Regulated: includes all the regulated IOCS.
SOCs-Group 1: includes alachlor. atrazine, and simazine.
SOCs-Group 2: includes bis(2-ethylbexyl)phthalate, bis(2-ethylhexyl)adipate, and benzo
-------
A Review of Contaminant Occurrence in Public Water Systems
Table IV.C.4. Summary of national MCL violations for surface water and ground water
systems. Percent MCL violations derived from SDWIS information for 1/1/93-3/31/98.
Contaminant Group
IOCS
SOCs
VOCs
Percentage of systems with MCL violations
Surface Water Systems
0.3 %
1.0%
0.2%
Ground Water Systems
0.6%
0.1%
0.4%
In short, surface water systems tend to be more vulnerable than ground water systems for many
contaminants, but they are not necessarily more vulnerable to occurrences exceeding an MCL. A
greater percentage of surface water systems show detections of SOCs than ground water systems,
in particular, but surface water systems show greater occurrence of VOCs as well. Surface water
systems even show a larger proportion of systems with exceedances of VOC-MCL
concentrations, though ground water systems show a slightly larger proportion of systems with
MCL violations. Partly, this may be related to the greater temporal variability in contaminant
occurrence and transport in surface waters. Surface water is vulnerable to many problems related
to land surface discharges of contaminants through runoff, spills, and even shallow ground water
recharge. Yet even in the areas most vulnerable to SOC contamination, variability is still a key
and confounding factor. In Midwest States, where 90-97% of surface water systems show
occurrence of pesticides, such as atrazine, there is still the small percentage of systems that have
no detections. These are often small watersheds (reservoirs) that are protected from contaminant
sources.
IV.C. 1. Ground Water Vulnerability
Ground water systems are much more variable in nature than surface water systems because
ground water adds other dimensions to the vulnerability equation. Aquifer type and well depth
are key hydrogeologic factors affecting ground water vulnerability. Confined aquifers have a
mantle of material that inhibits the recharge of water with modern contaminants into an aquifer.
Depth of a well and the resultant depth of the ground water influences vulnerability. Depth, in a
general sense, can be related to the time it takes for recharge water to reach a well intake in an
aquifer. The water tapped by a deep well may simply be too old to contain any modern, synthetic
organic contaminants. Deeper wells are prone to occurrence of natural contaminants, such as
some ICCs or radiochemicals, but are less susceptible to occurrence of anthropogenic
contaminants such as pesticides or many VOCs.
Aggregating all ground water systems in a vulnerability assessment does not provide adequate
information. Most public water system databases contain little information that can be used to
further unravel ground water system vulnerability. Further complicating such studies, is that
many public water systems operate multiple wells; some of which may be shallow and vulnerable
to contaminants, others may be deep and protected. When analyzing finished water data it is
30
-------
A Review of Contaminant Occurrence in Public Water Systems
often not possible to discern the ultimate source of a sample. Vulnerability is best defined using
water quality data from a particular well, where aquifer, depth, and other factors can be evaluated
independently. To illustrate, data from Iowa will be used where investigators have matched basic
aquifer and well properties to SDWA compliance monitoring data (Hallberg et al., 1996).
In Iowa, the Iowa Department of Natural Resources and the U.S. Geological Survey have
sampled municipal water-supply wells for common contaminants (Kolpin et al., 1997). These
data show that wells less than 100 feet deep have greater detections of nitrate and some common
herbicides, for example (Table IV.C.5). These studies also show that there are differences in
occurrence rates among different aquifer types; alluvial aquifers having more common detections
than bedrock aquifers, with glacial-drift aquifers also exhibiting lower vulnerability. This trend
in aquifer type reflects two inter-related variables: depth and confinement. Alluvial aquifers are
very shallow, by definition, and typically have a thin confining cover, if any. The glacial-drift
aquifers have a variable thickness of glacial deposits that are confining beds, and tend to be
relatively deeply buried. Bedrock aquifers are quite variable in setting and depth. The data are
for the raw water drawn from the wells. Although most of these ground water systems would not
have treated water in a way that would affect the occurrence of these contaminants, they may
have blended water that would lower the concentration and/or the occurrence.
Table FV.C.5. Percentage of Iowa municipal water-supply wells with detections of various
contaminants: for wells of various depths and for wells finished in different aquifers. (After
Kolpin, Sneck-Fahrer, Hallberg, and Libra, 1997.)
Systems using:
Wells <50 feet deep
Wells 51-100 feet deep
Wells > 100 feet deep
Percent
nitrate
76%
76%
28%
PWS wells
alachlor
4%
1%
2%
with detections
of:
atrazine cyanazine
16%
22%
8%
4%
9%
0%
metolachlor
7%
29%
3%
Systems using wells finished in:
Alluvial Aquifers
Bedrock Aquifers
Glacial-Drift Aquifers
74%
46%
35%
3%
2%
0%
21%
15%
0%
7%
0%
0%
17%
5%
0%
Iowa studies have also linked SDWA compliance monitoring data from finished drinking water
with various well and aquifer characteristics. These water-quality data come from standard
drinking water samples (entry points to distribution systems). Similar occurrence patterns can be
seen in these data (Table IV.C.6). First, a greater occurrence of contaminants in surface water
systems than ground water systems is pronounced, except for the Group 2 VOCs, (the dense
VOCs: PCE, etc.), which are more prevalent in ground water systems. Within the systems using
ground water, those using alluvial aquifers generally show the highest occurrence of SOCs and
VOCs, followed closely by the vulnerable karst aquifers. The exception, VOCs-Group 1, the
"light" petroleum-related compounds, are essentially equal in the various aquifers.
31
-------
A Review of Contaminant Occurrence in Public Water Systems
These data illustrate the problems with analyzing ground water as a single entity. For example,
overall, 9% of ground water systems have detections of atrazine, but 30% of systems using
alluvial aquifer wells less than 50 feet deep show detections. Within each aquifer type, there is a
general decrease in contaminant occurrence as well depth increases. Ground water systems using
alluvial aquifers show the most pronounced change with well-depth.
In the bedrock aquifers, contaminants are evident even in systems with wells over 250 feet deep.
There are two factors that contribute to this. First, many of these systems use multiple wells
from different sources. Many have a deeper well (>250 feet) and a shallower well (<150 feet).
The finished water quality data from the composite system confounds this analysis because the
contaminant occurrence cannot be uniquely related to wells of a given depth. Some of these
systems do show contaminant occurrence at greater depth, particularly in the karst aquifer group.
The hydrogeologic conditions where these aquifers are prevalent promote more rapid ground
water circulation, to greater depth, than in most areas. Hence, younger water, with SOCs, VOCs,
or other human-induced contamination, extends to greater depth than in many areas. The
combination of aquifer characteristics, hydrogeologic setting, and well depth must all be factored
to understand some of these patterns.
Table FV.C.6. Percentage of Iowa public water systems with detections of various contaminants
or contaminant groups, for systems using different source water, aquifers, and wells of different
depths. (Hallberg, Riley, Kantamneni, Weyer, and Kelley, 1996.)
Systems using:
All Surface Water
All Ground Water
Alluvial Aquifer (all)
With Wells < 50 ft deep
51-100 feet
101-150 feet
Fractured-Karst Bedrock Aquifer (a) (all)
With Wells < 150 ft deep
15 1-250 feet
>250 feet
Sandstone Bedrock Aquifer (b) (all)
With Wells < 150 ft deep
15 1-250 feet
>250 feet
Percent systems with detections of:
Any Any VOCs-1
atrazine SOCs VOCs ("light") (
97%
9%
23%
30%
10%
0%
14%
22%
24%
7%
. 2%
2%
11%
1%
97%
10%
29%
33%
18%
10%
14%
22%
24%
7%
3%
2%
11%
2%
37%
16%
23%
26%
18%
7%
21%
87%
22%
15%
20%
50%
33%
18%
23%
9%
13%
14%
13%
0%
14%
16%
15%
13%
14%
50%
22%
13%
VOCs-2
"heavy")
3%
4%
9%
9%
8%
7%
6%
9%
10%
3%
4%
40%
0%
3%
(a) Silurian-Devonian Aquifer; (b) Cambrian-Ordovician Aquifer
32
-------
A Review of Contaminant Occurrence in Public Water Systems
Figure IV.C. 1 provides a graphic illustration, showing the occurrence of nitrate in wells of
various depths in two regions of Iowa. These data are from private and public wells, but help to
illustrate these points. In the north-central region (NC), there is a pronounced relationship
between nitrate concentration and depth; nitrate levels decline sharply in wells >100 feet deep.
The karst aquifer hydrogeologic settings, described above, are from the northeastern region (NE).
Here the hydrogeologic setting promotes deeper circulation of ground water, and the nitrate-
depth relationship, while still apparent, is not as sharply defined. Nitrate occurs at high
concentrations to greater depth in this environment. The graphs also illustrate the variability in
ground water systems; while high concentrations of contaminants may be more prevalent at
shallower depths, there is still a mode of low occurrence, because of the range and variability of
hydrogeologic controls.
Given the variability in ground water settings, and system level variations, developing improved
monitoring requires refined information at the sub-state and system levels. Information on
aquifer characteristics, well depth, well casing and construction details, wellhead protection
information, as well as system and well maintenance history are some of the factors that should
be considered. Several States have utilized such information to improve the cost-effectiveness of
their monitoring program. The NJ Department of Environmental Protection, after conducting a
study of drinking water vulnerability with the USGS, estimated that waiving monitoring for wells
and intakes that are not vulnerable to pesticide contamination will save an estimated $5.1 million
dollars per year. The Washington State Department of Health, which also conducted a study
with the USGS, estimated that pesticide monitoring waivers for public supply wells would save
an estimated $6.0 million dollars per year (Vowinkel et al., 1996; Ryker and Williamson, 1996).
Name Concent radon as ND}-N (tng/L)
3 10 20 30 40 50 60
0
Nitrate Concentration as ND,-N (mg/L)
0 10 20 30 44 50 80
0
Figure FV.C.1. Nitrate-N concentrations versus well depth, from water-quality analyses of
private wells in north-central (NC) and northeastern (NE) Iowa, illustrating the general inverse
relationship between well depth (or ground-water depth) and contaminant occurrence. (Kross et
al., 1990, 1993; Hallberg et al., 1992).
33
-------
A Review of Contaminant Occurrence in Public Water Systems
IV.D. National Cross-Section: Perspectives and Comparison with Other Data
Before proceeding with further analysis of the State SDWA data, some perspective on the value
of the national cross-section can be offered. As noted, the occurrence estimates developed from
the cross-section cannot be considered a truly "representative" sample of nation-wide occurrence
of a given contaminant. If the States included do represent an adequate cross-section, these
values should be indicative of the average values. With the relatively low occurrence rates and
variability apparent in the data, the cross-section provides what is most likely a high estimate of
contaminant occurrence. There are few other sources of data available to use as a check on this
sample. Three primary sources of data will be used to provide perspective: (1) the URCIS
database; (2) the atrazine and simazine studies by Novartis; and (3) data from the USGS.
IV.D.1. URCIS Data
The URCIS database, as described above, is one of the few national compilations of drinking
water contaminant data. While URCIS has data quality limitations, it also has greater State
representation (some data from 39 States and 1 territory) than this study. As noted, for States
that had relatively complete data in both URCIS and this study, data comparisons were made and
the differences were found to be minor. (The URCIS database is a subset of the data used for
this study.) Some of the contaminants that were monitored under the original unregulated
contaminant monitoring regime (i.e., the URCIS data) have since become regulated
contaminants; hence, there is more data in the current State databases than in URCIS.
Comparing national occurrence values computed from URCIS with the national cross-section
shows that, in the majority of cases (about 67%), the cross-section has a slightly higher
proportion of systems with detections of a contaminant with overlapping data (Table IV.D.1),
and for about one-third of contaminants, URCIS shows an equal or greater percentage. In all
cases, the values are comparable; no values for systems with detections stand out as markedly or
unexpectedly different. In each case, contaminants that stand out as exceptionally high or low,
are proportionally similar in the other data set (e.g., methylene chloride, xylenes, or p-
chlorotoluene). This suggests that if additional State data were added, in a representative
fashion, that many of the national cross-section values would decline slightly. An EPA survey
for VOCs conducted in the 1980s showed similar results also, in terms of the proportions of
systems with VOCs and the general order of occurrence (Westrick et al., 1984).
IV.D.2. Novartis Atrazine and Simazine Data
Novartis Crop Protection, Inc. conducted a special regional analysis of atrazine and simazine
occurrence in CWSs, as part of a special study under direction from EPA's Office of Pesticide
Programs. While limited to two contaminants, this study is important because it compiled data
from CWS monitoring, conducted from 1993 through 1996, from 21 States across the country,
for one of the most widely occurring compounds (Clarkson et al., 1997). It represents one of the
most complete assessments of drinking water occurrence data of its kind. The 21 States selected
were the States with the greatest use for these pesticides, so occurrence estimates from this study
should be high (relative to occurrence for all 50 States, for example). The Novartis study
includes data from several States in the cross-section (California, Illinois and Michigan) and
from other State data reviewed and used in the ranges presented in this study (Indiana, Iowa,
34
-------
A Review of Contaminant Occurrence in Public Water Systems
Ohio). The individual State estimates matched quite closely. A comparison of the national
estimates (Table IV.D.2.) shows that the cross-section provides a lower estimate for atrazine and
a very comparable estimate for simazine. The ranges are nearly identical, which suggests that
data reviewed for this study capture a similar range of conditions as the Novartis sample of 21
States.
Table IV.D.1. Comparison of national occurrence estimates from URCIS and the national cross-
section of State SDWA data.
SURFACE WATER
SYSTEMS
URCIS
%>MRL
National Cross-Section
%>MRL
RANGE
GROUND WATER
SYSTEMS
URCIS
%>MRL
National Cross-Section
%>MRL
SOCs - Regulated
Ethylene Dibromide (EDB)
Dibromoch loropropane
(DBCP)
2.5%
2.6%
4.1%
5.3%
0%-9.4%
0%-13.2%
1.1%
2.5%
1.2%
3.0%
RANGE
0%-4.0
0%-9.3%
VOCs - Regulated
Benzene
Methylene chloride
Styrene
Tetrachloroeihylene
Toluene
Trichloroethylene
Vinyl chloride
Xylenes (Total)
2.5%
11.9%
1.4%
3.1%
8.2%
6.7%
1.2%
11.9%
4.2%
24.3%
4.5%
7.6%
10.8%
5.4%
3.4%
13.3%
0%-17.l%
0%-73.5%
0%-14.5%
0%-16.2%
0%-30.0%
0%-21.2%
0%-13.7%
0%-29.1%
1.9%
3.8%
0.6%
4.2%
3.8%
4.2%
0.5%
3.1%
1.3%
9.6%
2.3%
4.4%
4.0%
2.9%
0.5%
4.2%
0.5%-4.2%
I.8%-57.8%
0%-8.1%
0%-14.2%
1.6%- 14.4%
0.6%- 12.8%
0%-3.0%
0.9%- 15.3%
VOCs - Unregulated
(m-) 1 ,3-Dichlorobenzene
i . 1 , 1 ,2-Tetrachloroethane
Chloromethane
1 ,2,3-Trichloropropane
Bromomethane
o-Chlorotoluene
p-Chlorotoluene
0.8%
0.6%
2.4%
0.3%
1.3%
0.6%
0.3%
0.8%
1.0%
6.5%
0.8%
1.0%
0.3%
0.3%
0%-2.9%
0%-6.1%
0%-28.6%
0%-9.1%
0%-3.0%
0%-1.9%
0%-1.9%
0.2%
O.i%
1.5%
0.3%
0.7%
0.2%
0.2%
0.2%
0.2%
1.1%
0.3%
0.3%
0.4%
0.3%
0%-I.3%
0%-0.7%
0%-9.7%
0%-0.9%
0%-2.3%
0%-.9%
0%-0.4%
35
-------
A Review of Contaminant Occurrence in Public Water Systems
Table IV.D.2. Comparison of national estimates of occurrence (percentage of water systems
with detections) of atrazine and simazine from the Novartis study of 21 high-use States and from
the national cross-section (8 States) compiled in this study.
National cross-section surface
water systems
Novartis surface water systems
National cross-section ground
water systems
Novartis ground water systems
Percentage of public water systems with:
Atrazine
% >MRL
21.1%
51.7%
2.0%
3.0%
Range
0-97.1%
0-98.4%
0-9.2%
0-15.0%
Simazine
%>MRL
15.9%
15.1%
1.4%
0.6%
Range
0-67%
0-69%
0-3.2%
0-3.8%
IV.D.3. USGSData
A third source of national data are various studies by the USGS, particularly their National Water
Quality Assessment Program (NAWQA). Results from USGS studies have been brought up
repeatedly in CMR deliberations, noting that USGS estimates of pesticide ocurrences in surface
and ground water are greater than EPA estimates. The USGS conducts various monitoring
activities under nationally consistent protocols. This makes the data very valuable for reviewing
national patterns of occurrence that can provide perspectives on drinking water problems. One
important difference, however, is that the majority of the USGS data are from ambient water (i.e.,
raw water) and not drinking water. EPA estimates, such as in this study, are for drinking water
(i.e., finished water, after treatment, that may have been blended from multiple raw water
sources). While this precludes direct comparisons, the USGS work still provides some pertinent
perspectives. The review of the USGS findings provide a comparitive national perspective and
also illustrate why conclusions from the USGS ambient data are inherently different than
drinking water analyzed by EPA.
In their review of the NAWQA data collected to date, Kolpin et al. (1998) note that detections of
pesticide compounds are widespread in the shallow ground water sampled throughout the U.S.,
with one or more pesticide compounds detected in over 54% of the wells and springs sampled.
Thirty-nine different pesticide compounds were detected; 26 at concentrations above 0.01 ug/L.
This is a much higher frequency of occurrence than might be expected in the SDWA data. But
further scrutiny is required to put these values in perspective. The USGS is analyzing for a broad
spectrum of compounds, and the 54% value is for their entire contaminant group. Only five of
the compounds are regulated under SDWA Phase n/V rules. Five more are on the current
unregulated contaminant list that is required for monitoring by PWSs; twelve more are on the list
for monitoring under the proposed new Unregulated Contaminant Monitoring Regulation.
36
-------
A Review of Contaminant Occurrence in Public Water Systems
Consistent with findings from previous large scale studies, more than 95% of the pesticide
detections were at concentrations less than 1.0 ng/L. This concentration is below most of the
MCLs for the regulated compounds; hence there are few exceedances. In contrast, the overall
frequency of pesticide detection in the NAWQA studies is considerably higher than those
reported by two previous large-scale, multi-state studies of pesticide occurrence in ground water
across the United States: EPA's National Pesticide Survey (NPS) and National Alachlor Well
Water Survey (NAWWS).
The USGS authors note that the contrasts in pesticide detection frequencies between the
NAWQA data and the NPS and NAWWS investigations are attributable primarily to: "(a) low
detection threshold characteristics of the analytical method used for NAWQA and (b) the
relatively young age of ground water sampled in the NAWQA land-use studies." These are
important factors for understanding the USGS information and putting it into perspective for
drinking water programs: (1) USGS uses much lower minimum reporting levels than normal
SDWA labs; (2) USGS studies are targeted to relatively vulnerable conditions to establish the
link between ground water quality and land use, not drinking water and land use; and (3) the
sampling is targeted at shallow wells, monitoring wells, and is not typically from deeper drinking
water well (and thus the water sampled is relatively young).
The effects of targeting shallow, vulnerable ground water in drinking water quality studies were
illustrated in the previous section. As the Iowa data show, 9% of all systems using ground water
had atrazine detections, but 30% of systems using alluvial aquifer wells <50 feet deep had
detections (Table IV.C.7). Even these drinking water supply wells would be somewhat deeper
than many research monitoring wells.
The effect of the lower MRLs used in the USGS studies is to increase the number (and
percentage) of detections. The USGS MRL is about 10-times lower than the MRL used in the
EPA-NPS study (and most SDWA analyses), and the percentage of atrazine detections was
nearly 40 times greater (Table IV.D.3). In a more direct comparison from a drinking water study,
the State of Washington and the USGS cooperated on an evaluation of the vulnerability of PWS
ground water systems to improve Washington's waiver system for required monitoring
(Washington Div. of Drinking Water, 1995; Ryker and Williamson, 1996). Public water system
wells were sampled and analyzed for various pesticides (and other SOCs). The special waiver
study used lower MRLs than normally required for standard EPA or SDWA analyses in order to
provide added quality assurance to their evaluation. This resulted in a substantial increase in the
number of detections (Table IV.D.4). For example, of the 51 detections of atrazine with the
special, lower MRL, only 17 would be detections using the normal SDWA MRLs.
37
-------
A Review of Contaminant Occurrence in Public Water Systems
Table IV.D.3. Percentage detections of atrazine in different occurrence studies and the
minimum reporting levels used. (In part, from Kolpin et al., 1998.)
Study
USGS NAWQA
National Alachlor Well Water
USEPA National Pesticide Survey
Typical SDWA Analysis
MRL-ug/L
0.01
0.03
0.12
0.1
% of Wells with
Detections
28%
12%
0.7%
Table IV.D.4. Pesticide detections from public water system wells at different minimum
reporting levels. From a special vulnerability and waiver study conducted by the State of
Washington and the USGS (Washington Div. of Drinking Water, 1995; Ryker and Williamson,
1996).
Pesticide
2,4-D
2,4,5-TP
Atrazine
Lindane
Picloram
Simazine
Number of detections
with lower MRLs from
special waiver study
6
10
51
1
7
17
Number of detections
above standard
EPA/SDWA MRLs
6
1
17
0
2
2
While the USGS studies have important implications for many aspects of the drinking water
program, the findings on occurrence in raw ambient water can not be directly related to drinking
water.
In summary, the aggregated national cross-section of contaminant occurrence data compiled in
this report appears to provide a conservative, but reasonable approximation of national
occurrence values. The comparisons with other data suggest mat the cross-section summary
occurrence values are likely slightly high. Yet the data are representative enough that some basic
principles can be derived to guide further data analysis and collection efforts.
38
-------
A Review of Contaminant Occurrence in Public Water Systems
V. CONTAMINANT OCCURRENCE: SYSTEM SIZE AND OTHER VULNERABILITY
FACTORS
In addition to evaluating the fundamental differences and vulnerability of surface water and
ground water supplied systems, other factors were evaluated to assess occurrence patterns and
vulnerability. This section discusses system size, contaminant source issues, and temporal
(seasonal) variability issues.
V.A. System Size and Contaminant Occurrence
Concerns are often raised about the greater problems of contaminant occurrence in small
systems. In part, this is related to the more limited resources and options that small systems have
when dealing with contaminant problems. Some questions raised during CMR deliberations
indicate a perception that small systems in general may have greater contaminant occurrence.
Some of the State data and all of the URCIS data were analyzed in relation to system size (by
standard categories of the population served) and subdivided by water source (surface or ground)
to assess whether any consistant trends are apparent in the data.
Surprisingly consistent patterns were revealed in all the data sets analyzed. Results for the
selected Phase n/V contaminants (introduced in Tables IV.C.2-4) and various URCIS data are
summarized in Tables V.A.1 through V.A.8 and in Figures V.A.I through V.A.3 (Tables V.A.1
through V.A.8 are located in Appendix D). The Figures provide a summary of the trends in
occurrence patterns. In general, the same consistent patterns emerged from high contaminant
occurrence States and low contaminant occurrence States, particularly for the organic chemicals
— VOCs and SOCs. Few consistent trends were evident for the lOCs.
The proportion of systems with detections of VOCs increases with system size, from small to
large systems, particularly for ground water systems. The same trend is evident for surface water
systems, but it is not as consistent for all contaminants. In a survey conducted in the 1980s, EPA
also found that large ground water systems had significantly greater frequency of VOC
occurrence than small systems (Westrick et al., 1984).
The trend is more evident for percent detections (>MRL) than for the percentage of systems with
MCL exceedances (>MCL). This can be explained, in part, because the number of systems with
exceedances becomes so low that they may only occur in one size class in a State. The URCIS
data, with its greater national representation, suggest that for ground water systems the
proportion of systems with VOCs >MCL also increases with system size (Fig. V.A.3; Tables
V.A.6-8 — see Appendix D).
The SDWIS MCL violation data shown in Table V.A.9 illustrate this point. Again, there is a
consistent increase in the percentage of systems with VOC violations from small to large ground
water systems. While the patterns are not as consistent for other contaminant groups, there are
no cases where the smallest systems (<500 served) have the greatest percentage of violations.
The one exception in all State data reviewed, was when a system serving <500 persons was the
only size category (or at most, one other category) that detected the contaminant.
39
-------
A Review of Contaminant Occurrence in Public Water Systems
A) High-occurrence State
> MBL-Benz»ne
<500 SOI -3.300 3.30T - 10.00-- > 50.000
10.000 50.000
Population Sarvad
DSuitace Walar
• Ground Witir ,
> MHL-Totrachloroethylone
50.000
10,000 50.000
Population Sarvod
> IIRL-Xytanej
< SOO SOt -3.300 3,301 -
10.000
10.001 •
50.000
» 50,000
Population Sorvad
B) Low-occurrence States
> MRL-Boniai»
< SOO 501 -3.3OO 3.301 •
10,000
10.001 - > 50.000
50.000
Population Sarvad
> MRL-Totnchtoroatnytana
< 500 SOt -3.3OO 3.301- 10.OO1 - > 50.000
10,000 SO.OOO
Population Sarvad
> MCL-B«nzene
< 500 501 -3.300 3.301 - 10.001 - > SO.OOO
10.0DO 50.000
Population Sorvod
> MCL-Tatrachloroethjftane
<500 501-3.300 3.301- 10.001- > $0,000
10,000 so.ooo
Population Sorvod
No systems with xylene concentrations above !h« MCL
> UCL-Bonzana
%o(Syil«ni
5
i
•"••
< SOO S01 -3.300 3.301 • 10.001 • > 50.000
10,000 50,000
Population Sorvod
> MCL-TvtracMormthyton*
500 501 -3.300 3,301 -
10.000
10.DOI - > 50.OOO
so.ooo
Population Sarvod
Figure VA.1. Summary of the percentage of systems with detections (>MRL) and exceedances
(>MCL) of selected VOCs, comparing ground-water and surface-water supplied systems, by size
of system (population served categories), for A) a high-occurrence state, and B) a low-occurrence
state.
40
-------
A Review of Contaminant Occurrence in Public Water Systems
A) High-occurrence State
> URL-EOB
> MHL-Cjrbolur»n
< 500 501 -3.300 3.301 - 10,001 •
10.000 50.000
Population Sorvod
, 50.000
B) Low-occurrence States
> MRL-EDB
Population Servcrf
> MCL-Atrazine
100%
I 80%
I S0%
2 40%
J 20%
TT
H
: 500 501 -3.300 3.301 - 10.001 - > SO 000
10.000 50.000
Population !
> UCL-EDB
Populntion $*iv*d
No systems with carbofuran concentrations above the MCL
> MCL-Atnzine
SOI -3.30O 3,301- 10.001-
10.000 so.ooo
Population Sfrrmd
> MCL-EDB
00 3,301 -
10,000
Populttton S*rv«d
10.001 -
$0,000
> 50.000
! V.A.2. Summary of the percentage of systems with detections (>MRL) and exceedances
(>MCL) of selected SOCs, comparing ground-water and surface-water supplied systems, by size
of system (population served categories), for A) a high-occurrence state, and B) a low-occurrence
state.
41
-------
A Review of Contaminant Occurrence in Public Water Systems
A) VOCs in URCIS
> MRL-T«lr»chloroelhylene
< 500 501 -3,300 3,301 - 10.001
10,000 50.000
Population Served
|DSuf!««« WlWr
IBJGcoundW.i.r
> MRL-TrichloroethytMie
50%
S01 -3.300 3.301 - 10.001 •
10.000 50.000
Population Svrvad
> 50.000
. MRL-Xyl«n*s
< 500 SOI -3.300 3.301 • 10.001 • » 50.000
10,000 50.000
Population Sarvad
B) SOCs in URCIS
> MRL-EDB
< 50O 501 -3.3OO 3.301 - 10.001 -
10.000 50.000
Population Sarvad
3. MRL-DBCP
< 500 SOI -3,300 3.301 - 10.001 -
10.000 50.000
Population Samd
> 50.000
< 500 501 -3.300 3.301 - 10,001 -
tO.OOO 50.000
Population Sarvad
> MCL-TrieMoroethytoim
« 500 S01 -3.300
3.301 - 10.001 -
10.000 50.000
Population Sanaa1
No systems with xylene concentrations above (he MCL
> MCL-EDB
' 500 SOI -3.300 3.301 - 10.001 - > 50.000
10.000 $O.OOO
Population Sarvad
> MCL-DBCP
SOI -3.3OO
3.301 • 10.001 -
10.000 50.000
Population Semd
> 50.000
Figure V.A3. Summary of the percentage of systems with detections (>MRL) and exceedances
(>MCL) of selected A) VOCs, and B) SOCs, comparing ground-water and surface-water
supplied systems, by size of system (population served categories), from the national URCIS
database.
42
-------
A Review of Contaminant Occurrence in Public Water Systems
Table V.A.9. Summary of national MCL violations by system size, for surface water and ground
water systems. Percent MCL violations derived from SDWIS information for 1/1/93-3/31/1998.
System Size
(Population served)
<500
501-3,300
3,301-10,000
10,000-50,000
>50,000
All
Percentage of water systems nationally, with MCL violations
summarized for contaminant groups, by source water and
system size (population served)
lOCs
SW
0.2%
0.4%
0.3%
0.5%
0.0%
0.3%
GW
0.5%
0.5%
0.9%
1.7%
0.0%
0.6%
SOCs
SW
0.5%
1.7%
1.2%
0.5%
0.3%
1.0%
GW
0.1%
0.1%
0.1%
0.1%
0.0%
0.1%
VOCs
SW
0.2%
0.2%
0.0%
0.4%
0.3%
0.2%
GW
0.4%
0.3%
0.7%
1.8%
1.1%
0.4%
This same trend in ground water is apparent for many SOC contaminants, but not all (Fig. V.A.2
and V.A.3). For high-occurrence States and URCIS — both of which have more data —
increasing occurrence from small to large systems is clearly illustrated. However, there are often
too few systems with SOC detections for a pattern to emerge. For example, in Figure V.A.2A,
carbofuran only occurs in one system size category. For a low-occurrence State (Fig. V.A.2B),
EDB only occurs in two size classes of ground water systems and is predominant in systems
serving >10,000 persons, but in URCIS (Fig. VA.3) the trend from small to large is clearly
defined, both for occurrence (>MRL) and exceedances (>MCL).
Patterns for surface water systems are variable, sometimes increasing, sometimes decreasing with
system size, again, in part because there are relatively few systems with exceedances. The
pattern for systems with exceedances (>MCL) generally follow those for detections. For
atrazine, on Figure V.A.2, the largest ground water systems have the greatest proportion of
systems >MCL, but the trend is the opposite for surface water systems.
The data clearly indicate that, on a proportional basis, small systems do not show a greater
incidence of contaminant occurrence. For many contaminants, the larger systems show greater
occurrence and a tendency, at least for VOCs, to have proportionately more systems with
exceedances. This trend is most consistent for ground water systems and is a logical pattern in
many respects. First, systems using ground water to supply larger populations must produce
greater volumes of water. To pump large, consistent volumes of ground water most systems use
relatively shallow, often unconfined, aquifers that can produce the water needed. Many larger
systems use alluvial aquifer well fields. As already described, shallow, unconfined aquifers, such
as alluvial aquifers, are also the ground water settings most vulnerable to contamination. The
greater pumping rates may further contribute to greater contributing area to these wells. Many
times, larger systems serve and are located around major population centers, which typically have
a greater number and density of contaminant sources, particularly for VOCs. Many monitoring
43
-------
A Review of Contaminant Occurrence in Public Water Systems
requirements arc more stringent for systems serving larger populations, in large part because they
do serve the largest proportion of the population. The data support that general approach.
V.A.I. Systems By The Numbers; Other Perspectives On Occurrence
The percentage of systems with occurrence by size or source water categories provides insight to
the relative, inherent vulnerability to contamination. There are other important perspectives in
these numbers that also should be kept in mind, such as the number of systems involved or the
population affected. These are briefly discussed below.
V.A.1.a. System Numbers Perspective. Table V.A. 10 summarizes the number of systems within
each system-size/source-water category (for CWSs and NTNCWSs). In the U.S. there are many
more small systems than large systems. For example, there are approximately 46,000 ground
water systems serving <500 people and about 200 serving >50,000 people. One percent of
ground water systems serving <500 people represents a larger number of systems (-460) than the
total number of ground water systems serving >50,000 people. Thus, lower proportional
occurrence in small systems may still translate into a greater number of systems with problems,
directly impacting the resources needed for program oversight for States. Tables V.A.11 and
V.A. 12 provide perspective on the number of systems with contaminant detections in relation to
the percentages of systems.
Table V.A.I 1 illustrates the perspective by CWS and NTNCWS system size, by surface water
and ground water for the occurrence of PCE (tetrachlorethylene). For this example the number
of systems with detections is calculated by multiplying the percentage of systems with detections
from the URCIS database times the national inventory numbers in Table V.A.10. As described
previously, on a proportional basis, there is a substantial and consistent trend with occurrence
increasing from small to large ground water systems: 43% of the largest ground water systems
(those serving more than 50,000 people) have detections of PCE while only 2% of the smallest
ground water systems (those serving less than 500 people) have detections. This indicates that
87 of the largest ground water systems detect PCE while approximately 1,000 of the smallest
ground water systems detect PCE, even though there are proportionately 40% fewer systems with
detections in this size class. Hence, while the proportion of systems with detections steadily
increases with increasing size, the number of systems with detections declines.
The comparative perspective of percentage of systems and number of systems also is pertinent to
the comparison of source-water categories. Table V.A.12, shows the summary comparison of
surface water and ground water systems from the national cross-section (after Table IV.C.3),
including both percentages and number of systems nationally. The number of systems is
calculated by multiplying the percent occurrence from the national cross-section times the
national inventory (Table V.A.10). For atrazine there is a significant difference in vulnerability:
21% of surface water systems have detections, compared to 2% of ground water systems. This
translates to nearly identical number of systems—1,275 for surface water and 1,247 for ground
water—because of the 10-fold difference in system-source water numbers. For xylenes, the
percentages still indicate greater vulnerability for surface water systems: over 12% detecting
xylenes, compared to 4% for ground water. From a program implementation standpoint, this
equals about 740 surface water systems vs. 2,400 ground water systems, however.
44
-------
A Review of Contaminant Occurrence in Public Water Systems
Both the proportional occurrence and the actual number of systems with problems are important
perspectives to understanding the implications of these data. The percentage of systems with
occurrence of a contaminant by system size or by source water illustrates the relative
vulnerability of systems. The actual number of systems affected directly relates to the level of
effort required for program implementation, the necessary oversight, and the total costs involved.
VA.l.b. Population Perspective. Exposure assessment is not the focus of this report, however,
as noted in previous sections, one of the most important perspectives on occurrence and system
size is related to population exposure. A relatively small portion of the approximately 68,000
CWSs and NTNCWSs serve the majority of the national population. As shown on Table
V.A.10,96% of the total CWSs and NTNCWSs are small systems, each serving less than 10,000
people, while 4% are large systems, each serving greater than 10,000 people. However, the 4%
comprised of large systems serve nearly 80% of the population, while the 96% comprised of
small systems, serve about 20% of the population. Hence, the increased proportion of large
systems with occurrence of contaminants also translates into a disproportionately greater portion
of the population being exposed. All of these perspectives on occurrence have bearing on
considerations for design of monitoring schemes and other facets of drinking water programs.
Table V.A.10. Total number of nonpurchased, community and non-transient non-community
water systems (CWSs and NTNCWSs), by source water and system size. From January 1997
SDWIS database.
Size Category
500 and under
501 to 3,300
3,301 to 10,000
Subtotal * 10,000
10,001-50,000
over 50,001
Subtotal > 10,000
TOTAL
# Systems
Surface Water
1,900
1,820
1,006
4,726
927
389
1,316
6,042
# Systems
Ground Water
46,200
12,306
2,404
60.910
1,254
204
1,458
62,368
Total f Systems
TOTAL
48,100
14,126
3,410
65,636
2,181
593
2,774
68,410
45
U.S. EPA Headquarters Library
Mail code 3201
1200 Pennsylvania Avenue NW
Washington DC 20460
-------
A Review of Contaminant Occurrence in Public Water Systems
Tabie V.A.11. Community and non-transient non-community water systems (CWSs and
NTNCWSs) with detections of PCE >MRL and >MCL from URCIS, by system size. Number of
systems affected nationally extrapolated from SDWIS inventory in Table V.A.10.
Percentage and numbers of water systems nationally, with detections >MRL and detections
>MCL* for PCE, by source water and system size {population served)
System size
(Population served)
<500
500-3,300
3,301-10,000
10,001-50,000
>50,000
SW > MRL
1.1%
2.0%
1.3%
5.5%
11.8%
22
36
13
51
46
GW > MRL
2.2%
3.8%
9.6%
23.8%
42.8%
1,001
470
230
298
87
SW > MCL*
0.0%
0.4%
0.8%
1.2%
3.7%
0
8
8
11
14
GW > MCL*
0.4%
1.0%
2.9%
9.2%
23.3%
198
126
70
115
48
*%> MCL indicates the proportion of systems with any analytical results exceeding the concentration value of the MCL; it does not
necessarily indicate an MCL violation. An MCL violation occurs when the MCL is exceeded by the average results from four
quarterly samples or confirmation samples as required by the primacy State
46
-------
A Review of Contaminant Occurrence in Public Water Systems
Table V.A.12. Summary comparison of occurrence of selected Phase H/V contaminants in water
systems using surface water vs. ground water, from national cross-section States. Shows
percentage of systems and number of systems (after Table IV.C.3).
Contaminant
Percent > MRL
Surface
Water
Ground
Water
#>MRL
Surface
Water
Ground
Water
Percent > MCL*
Surface
Water
Ground
Water
#>MCL
Surface
Water
Ground
Water
Individual Contaminants
IOCS
Barium
Cadmium
Mercury
SOCs
2,4-D
Atrazine
Benzo[a]pyrene
Carbofuran
(Furadan)
Ethylene Dibromide
(EDB)
Glyphosate
(Roundup)
Simazine
VOCs
Benzene
Styrene
Tetrachloroethyiene
Vinyl Chloride
Xylenes (Total)
Group Summaries
IOCS
lOCs-ReguIated
SOCs
SOCs-Group 1
SOCs-Group 2
VOCs
VOCs-Regulated
VOCs-Group 1
VOCs-Group 2
49.1%
5.1%
9.0%
11.2%
21.1%
0.5%
0.8%
4.2%
0.0%
15.9%
3.9%
4.1%
7.1%
3.1%
12.3%
47.3%
4.9%
4.5%
1.2%
2.0%
0.5%
0.1%
1.0%
0.1%
1.4%
1.2%
2.1%
4.3%
0.5%
3.9%
83.7%
21.9%
20.4%
41.1%
19.5%
11.1%
83.5%
2.4%
13.4%
19.9%
6.6%
6.4%
2,967
308
544
677
1,275
30
48
254
0
961
236
248
429
187
743
29,500
3,056
3,368
748
1,247
312
62
624
62
873
748
1,310
2,682
312
2,432
0.5%
0.2%
0.5%
0.0%
10.7%
0.0%
0.0%
3.7%
0.0%
1.0%
0.3%
0.0%
1.7%
0.3%
0.0%
5,057
1,323
1,233
2,483
1,178
671
52,077
1,497
8,357
12,411
4,116
3,992
2.5%
10.7%
0.9%
8.2%
0.9%
2.9%
0.2%
0.6%
0.4%
0.0%
0.1%
0.1%
0.0%
0.7%
0.0%
0.0%
0.4%
0.2%
1.8%
0.2%
0.1%
4.2%
0.1%
1.0%
6.1%
0.9%
3.2%
30
12
30
0
646
0
0
224
0
60
18
0
103
18
0
125
374
249
0
62
62
0
437
0
0
249
125
1,123
125
62
151
646
54
495
54
175
2,619
62
624
3,804
561
1,996
* % > MCL indicates the proportion of systems with any analytical results exceeding the concentration value of the MCL; it does not necessarily
indicate an MCL violation. An MCL violation occurs when the MCL is exceeded by the average results from four quarterly samples or
confirmation samples as required by the primacy State.
IOC-Regulated: includes all the regulated lOCs.
SOCs-Group 1: includes alachlor, atrazme, and simazine.
SOCs-Group 2: includes bis(2-ethylhexyl)phthalate, bis(2-ethylhexyl)adipate. and bcnzo(a)pyrene
VOCs-Regulaied: includes all the regulated VOCs
VOCs-Group I: includes benzene, ethyl benzene, toluene, and total xylenes (LNAPLs).
VOCs-Group 2: includes cis-1,2-dichloroethyiene, trans-l.2-dichloroethylene. 1,1-dichloroethene. tetrachloroethylene. trichloroethene, and vinyl
chloride (DNAPLs).
47
-------
A Review of Contaminant Occurrence in Public Water Systems
V.B. Contaminant Sources and Occurrence
The prevalence of contaminants in drinking water can often be related to the use of chemical
contaminants or the natural occurrence of minerals and metals within the recharge zone or
watershed of a public water system. The relative contaminant source contribution should be
considered when States revise their monitoring programs. Little information was collected for
this study to specifically address this issue. Some observations can be summarized to illustrate
the effects of contaminant sources on the occurrence of contaminants in drinking water. As
noted, the occurrence of lOCs can vary greatly related to natural geologic sources and inter-
related mining activities. Often such sources are localized (sub-state) and such analysis is
beyond the scope of this study. At the State level patterns are not readily evident. Hence, the
analysis focuses on the organic chemicals.
Table V.B.I summarizes the average drinking water occurrence rates of selected VOCs for 8
different States, with their corresponding national ranking for total TRI releases (in pounds) per
square mile. There is a consistent increase in the percentage of water systems with VOCs >MRL
from low TRI States to high TRI States. The trend is best expressed by the VOC Groups 1 and 2
summaries. For individual compounds it is less consistent, but still evident. Again, the patterns
are most consistent for ground water systems.
Table V.B.2 summarizes the average drinking water occurrence rates of selected SOC's for 8
different States with their corresponding ranking for total agricultural chemical expenditures.
Here there is no evident consistency. While the very highest occurrence values (e.g., SOCs
Group-1, or 2,4-D) are in the number two ranked State (i.e., high chemical expenditures, high
pesticide use State), the highest ranked State has generally low values for occurrence. The
pattern for occurrence in relation to agrichemical expenditures (and total pesticide use) differs for
different compounds, largely because individual pesticide use varies greatly depending on crop
type. For example, the compounds used in Illinois for corn and soybean production are not the
same as those used for specialty crops and fruits in California. Illinois shows 96% of surface
water systems with detections of atrazine, but California only shows 1 %. Illinois ranks first
nationally in atrazine use, California is not even among the top 20 States for atrazine use. In
addition, many widely used pesticides are not included in SDWA monitoring requirements. The
pesticide compounds in use are much more variable than the general nature of VOCs related to
urban and industrial contaminant sources. Thus, it is difficult to establish clear associations
between SDWA contaminant occurrence patterns and aggregate pesticide use.
Where more details are known for a particular pesticide, more useful relationships may be
apparent. Figure V.B.1 shows the percentage of CWSs in a State with detections of atrazine
plotted against the ranks of States by amount of atrazine applied to all crops for the twenty
highest atrazine-use States (data from the Novartis study; Clarkson et al., 1997). For surface
water there is a strong linear relationship. Atrazine is land applied, over large areas, which is
why such a relationship is apparent. The graph also illustrates the much lower occurrence in, and
lower vulnerability of, ground water systems. With the lesser occurrence values, in relation to
the complexity of ground water settings, there is no significant trend for ground water systems.
48
-------
A Review of Contaminant Occurrence in Public Water Systems
The USGS NAQWA studies also have summarized some aspects of contaminant source factors
related to occurrence (URL for the NAWQA National Synthesis: http://wwwrvares.er.usgs.gov/
nawqa/natsyn.html; see Appendix B). Many of the compounds included in their studies are not
regulated for drinking water, but summary observations are illustrative of considerations for
evaluating risks to drinking water systems. In their national synthesis to date, they have found
that herbicides are more common in water sources in agricultural areas and insecticides are more
common in urban watersheds. Two times as many insecticides are detected in surface water as
ground water. (Insecticides tend to be less mobile and less soluble than many herbicides.)
Atrazine was the most commonly detected pesticide in both surface and ground water, which
coincides with the national cross-section estimates, as well. Also illustrated is that various
pesticide compounds are more prevalent in shallow ground water, in different agricultural
regions, related to crop type and the types of pesticides used (Kolpin et ah, 1998). Petroleum-
related VOCs occurred much more frequently in shallow wells in urban areas than agricultural
areas (Squillace et ah, 1998).
While not surprising, for the first time substantial information is becoming available to support
and refine such observations. These can be useful factors to guide monitoring reform
considerations.
Table V.B.I. Summary of the percentage of systems with detections (>MRL) of selected VOC
contaminants, comparing surface-water and ground-water supplied systems, by the State ranking
for total TRI release (in pounds per square mile); States range from low (fewest pounds released)
to high (most pounds released) rank, from left to right.
Contaminant
State Rank-Toxic Release Inventory
40
39
38 1 34
16 i 11
8
7
Surface Water (% > MRL)
VOCs-Groups
Group 1
Group 2
VOCs-Individual
Benzene
Tetrachloroethylene
5.9%
0.0%
0.0%
0.0%
3.8%
5.1%
0.6%
2.8%
8.0%
6.4%
0.0%
5.0%
7.6%
1.9%
0.0%
1.9%
30.5%
0.0%
1.7%
0.0%
39.3%
22.2%
17.1%
16.2%
30.3%
30.3%
3.0%
15.2%
42.9%
25.7%
2.9%
12.9%
Ground Water (% > MRL)
VOCs-Groups
Group 1
Group 2
VOCs-Individual
Benzene
Tetrachloroethylene
6.3%
2.4%
0.9%
2.0%
2.2%
3.6%
0.5%
6.9%
2.5%
5.3%
0.5%
4.0%
2.0%
2.1%
0.3%
1.8%
5.6%
3.4%
1.0%
1.8%
19.5%
8.8%
4.2%
5.5%
9.9%
13.0%
1.8%
5.7%
28.6%
15.4%
2.5%
9.7%
Group 1 VOCs-Benzene, Ethyl benzene, Toluene, Xylenes (total)
Group 2 VOCs-cis-l,2-DichlorethyIene, trans-1,2-Dichlorethylene, 1,1-Dichloroethene,
Tetrachloroethylene, Trichloroethene (Trichloroethylene), Vinyl chloride
49
-------
A Review of Contaminant Occurrence in Public Water Systems
Table V.B.2. Summary of the percentage of systems with detections (>MRL) of selected SOC
contaminants, comparing surface-water and ground-water supplied systems, by the State ranking
for Agricultural Chemical Expenditures; States range from low (least expenditures) to high (most
expenditures) rank, from left to right.
Contaminant
State Rank-Agricultural Chemical Expenditures
40
37 | 34 | 26
22
18
2
1
Surface Water (% > MRL)
SOCs-Groups
Group 1
SOCs-lndividual
2,4-D
Carbofuran (Furadan)
Ethylene Dibromide
(EDB)
0.0%
0.0%
0.0%
6.7%
0.0%
0.0%
0.0%
9.4%
0.0%
0.0%
0.0%
2.0%
13.2%
11.8%
0.0%
4.4%
0.0%
1.3%
1.4%
0.7%
7.0%
0.0%
0.0%
0.0%
96.3%
50.5%
1.8%
9.4%
1.3%
1.9%
0.0%
5.1%
Ground Water (% > MRL)
SOCs-Groups
Group 1
SOCs-lndividual
2,4-D
Carbofuran (Furadan)
Ethylene Dibromide
(EDB)
0.3%
0.0%
0.0%
2.5%
0.0%
0.0%
0.0%
0.5%
0.1%
0.1%
0.0%
0.0%
3.4%
4.7%
0.0%
2.6%
0.0%
0.2%
0.0%
0.9%
0.5%
0.0%
0.0%
0.0%
8.1%
3.5%
0.1%
1.3%
2.6%
1.0%
0.2%
1.3%
Group 1 SOCs-Alachlor (Lasso), Atrazine, and Simazine
50
-------
A Review of Contaminant Occurrence in Public Water Systems
» GW Systems
B SW Systems
Linear Fit (SW
Systems)
- - Linear Fit (GW
Systems)
State Rank by Pounds of Atrazine Applied to All Crops
Figure VJJ.1. Percentage of CWS in a State with detections (>MRL) of atrazine related to the
State's national rank (l=highest rank, i.e., greatest amount applied) for pounds of atrazine
applied, for surface-water and ground-water supplied systems (data from Novartis 1997).
V.C. Temporal Variability and Vulnerability
Water quality studies and monitoring throughout the United States have clearly shown that
occurrence and/or concentrations for some contaminants may vary over time, both seasonally as
well as from year to year. The seasonally of contaminant occurrence, or period of peak
concentration, commonly varies with seasonal changes in the hydrologic cycle in relation to the
source of contaminants and their fate and transport characteristics. Particularly for land-applied
or land-disposed contaminants, the seasonal increase in the flux of water (e.g., spring rains) can
mobilize contaminants and move them into surface or ground water flow systems. For the most
vulnerable of water systems, such as surface waters, unconfined shallow ground water, and karst
flow systems, for example, contaminant occurrence or peak concentrations typically occur during
annual runoff and recharge periods. Targeting monitoring to these vulnerable time periods can
improve the effectiveness of compliance monitoring and accuracy of exposure estimates.
However, there are concerns about the cost effectiveness of seasonal targeting approaches. If, for
example, many of the 64 Phase IFV contaminants exhibit different seasonal patterns, trying to
51
-------
A Review of Contaminant Occurrence in Public Water Systems
seasonally target all the different contaminants could lead to a very complex and ultimately more
costly monitoring regimen.
For much of the United States east of the Rocky Mountains many studies have shown the season
of greatest vulnerability for contaminant occurrence is the late-spring, early-summer runoff-
recharge period. This has been well established from detailed source water monitoring data,
particularly for contaminants such as pesticides and nitrate (e.g., Larson et al., 1997; Barbash and
Resek, 1996; Hallberg, 1989a,b). For example, Figure V.C.I summarizes pesticide
concentrations in streams from the USGS NAWQA studies. This national summary shows the
concentration of pesticides in agricultural areas peaking from May through July. For streams
draining urban areas the concentrations are lower and they do not show such pronounced
seasonally, though May through July would still include most of the peak period.
For deeper, more confined ground water systems, defining vulnerable periods is much more
difficult. The exact flow path is more complex, and the time of travel much greater, and these
are dependent upon many factors unique to a particular well and aquifer setting (e.g., Hallberg
and Keeney, 1993). However, as depth of ground water increases (and vulnerability decreases),
seasonal variability typically decreases (e.g., Barbash and Resek, 1996). There is no seasonal
generality that can be applied to these deeper ground water settings.
V.C.I. SOCs
The State SDWA occurrence data were analyzed for seasonal patterns that might provide insight
into drinking water monitoring schedules. Unraveling such patterns from data aggregated from
many different water sources and systems is difficult, at best, as discussed below. The clearest
examples are for high-occurrence pesticides. Figure V.C.2 illustrates the typical seasonal pattern
for atrazine occurrence which peaks in May-July, but the number of CWSs with high monthly
means decreases slowly through the fall and winter. This is one way to look at occurrence
patterns.
52
-------
A Review of Contaminant Occurrence in Public Water Systems
_J
s"5
<0
O>
§3
o
0>
!5 2
^
£1
"o rt
_L-,_JL I
L—JiZZli-1--155-
jan mar
— !
— -n
r— '
— I
—\
r*
may
f«b apr
1
L__ — i : j •~" | i | i i j— ..—
j"l
jun
au£
sep nov
oct dec
IMAX
75%
I*1CLH«IM
T25%
WIN
Agricultural streams
i5
CO
£ 4
ci
§3
o
S 2
o
"5?
en
33 o
L,J
u
I . i=i— j I? -~. L__j_J
jan mar
1 .
'n
r~
may
feb apr
Urban streams
jul
jun
aug
j JL
| j , j | 1 ^_
sep nov
oct dec
i MAX
§
-^ 75%
MEDIAN
T25%
M1N
Figure V.C.I. Summary of monthly average total pesticide concentrations in streams sampled in
the USGS NAWQA Program, for streams affected by runoff from agricultural and urban lands.
(After Larson, Gilliom, and Capel, 1999).
Figures V.C.3 and V.C.4 also illustrate seasonal patterns for pesticides, as well as the problems
that can be encountered in using drinking water data to conduct such analyses. These data are
from the special Ohio studies of occurrence in surface water systems (Ohio EPA, 1998).
Particularly for pesticides that occur on a more intermittent basis, May-July peaks in the
percentage of systems with detections are evident. For atrazine, however, the greatest percentage
53
-------
A Review of Contaminant Occurrence in Public Water Systems
of systems with detections appears to be September and December. This is largely an artifact of
the sampling regimen, and of the systems that were required to sample. Not all systems sampled
in the fall and winter; only those systems that were known, or suspected of having year-round
occurrence had to sample. Hence, in September 100% of systems sampling had detections. The
seasonal occurrence pattern is more clearly defined looking at the maximum concentrations
detected by month, where May-July clearly stand out.
As illustrated by this example, analyzing the State PWS data is complicated, because many
sources of variation may be aggregated — many different systems, with different source
characteristics, different sampling schedules, over various years, all in relation to various
contaminant source characteristics. These can result in "smoothing" out the seasonal variation
(e.g., percentage of systems with atrazine, Fig. V.C.3), especially for persistent contaminants that
may be present all year. The aggregation of systems and source characteristics particularly
confound analysis of ground water systems, but also affects analysis of surface water systems.
For example, detailed studies by the USGS and others have shown that the seasonal response in
reservoirs may be very different than in streams — and these are both typically identified simply
as surface water sources in the PWS databases.
Small streams are more immediately affected by runoff events, and contaminant concentrations
are generally greater than in large streams (which integrate a greater area). While this changes
the details of temporal patterns at the daily-weekly level, the seasonal patterns are similar.
Reservoirs, however, store these runoff-related events, and contaminant variations appear to be
dampened. The high concentrations that enter reservoirs during runoff events may be stored for
some time (e.g., months), and year-to-year variation may be more important than seasonal
variations in reservoirs and lakes, depending upon reservoir size, land use in the watershed, and
the turnover rate (Battaglin and Goolsby, 1998; Larson et al., 1997; Scribner et al., 1996).
Some studies have also shown secondary peak concentrations of some pesticides in fall and
winter months with discharge from urban areas, but these are of much lesser magnitude than the
spring period occurrence peaks (Coupe et al., 1995). Also, seasonal patterns are different in the
Pacific west, for example, where fall and winter are important rainfall and recharge periods.
Patterns also can be complicated by irrigation schedules or releases from irrigation storage
reservoirs in the arid west (e.g., Larson et al, 1997; Kuivila and Foe, 1995).
54
-------
A Review of Contaminant Occurrence in Public Water Systems
Month
Figure V.C.2. Number of community water systems with monthly mean atrazine concentrations
above 3.0 jUg/L (in raw water), from special sampling study in Iowa, by Novartis Crop
Production (Novartis, 1997; Clarkson et ah, 1997).
55
-------
A Review of Contaminant Occurrence in Public Water Systems
*
| to*.
t m
i
»*
a*
X
fc
p
1
MS
M
ft
«ns
newil al Syit»mi With D*l»cllani of Alacftloi, t y M onlh
I_
I
1
1
•
•
• -
NS MC N$ M» JW JJ W S
-------
A Review of Contaminant Occurrence in Public Water Systems
'. I
i«
$
s M
i 1.
: ?
i E
i
ercent ot syftiemi w»h Detection* et MetolocMot. ly Month
i •
III •
III
II 1 1
II 1 1
NS NS MO NS Mo/ JUi AJ NS Sw NS MS Dec
'00*
j 80%
j5
20%
CK
Percent ogeet SytFemi With Detection* of M*tilbuz)n otd
Cyonatlrt*. ey Month
-™" ~ " ' " " '
•
ill
111 1 I
III 1 1
NS NS Mo NS Mot Jui JU NS S«J NS NS Dec
|
j
MoadmumConeentFOllonf or Metotoehlor Detected, By Month t
s!
» »
1 20
NS ME Ma NS Mot JU
i
| _
i JU NS Sao NS NS Da=
X
U
l'°
0
MarimumCoficwtlrotloni ot Mcttlbiuln and C«aimli»
DottelM. ty Month
I.
:
J B
NS NS Ma NS M» Jin JJ NS Sap NS NS OK
j
' B iQ%
. * ^^
| W
0*-
Poreonleotot Systwni WHh DM«ctloni ot AectochJor, >y
Month
~<
|
1
AC* • MS MS Mo MS M
-------
A Review of Contaminant Occurrence in Public Water Systems
Figures V.C.5 through V.C.8 summarize various occurrence data by month. The VOC data, like
the SOC data were analyzed several ways, including monthly number or percent of samples and
systems with detections (>MRL, >0.5MCL, >MCL), the percentage of detections per month as a
function of all detections or the portion of systems sampling per month; and monthly
concentrations (median, 95th percentile, maximum). Even individual systems with common
occurrence were isolated to assess possible temporal trends. No systematic trends were apparent.
All the results look similar to the examples in Figures V.C.5 through V.C.8. There are no
consistent seasonal patterns that emerge for VOCs. Figure V.C.5, shows monthly charts for
xylene for several States. From the ground water systems from Dlinois and the surface water
systems from Michigan a 'bell-shaped' occurrence pattern, peaking in mid-summer, might be
surmised. However, the Illinois surface water systems alternate peaks and declining values.
Oregon shows a peak in December, although this could be a function of Oregon's different
climatic regimen. Other analysis in these States suggest that the patterns are more related to what
systems are sampling, rather than a seasonal pattern, especially when groups of related
contaminants are viewed. For example, Figure V.C.6, shows several VOC Group-1
contaminants for one State. Hints of seasonal patterns for one contaminant are out of phase with
others. For ground water systems, for xylene, an overall "bell-shaped" pattern occurs, except that
the lowest month is in the middle of the bell. Figure V.C.7 and 8, show a similar lack of pattern
for the heavier VOCs (PCE, and TCE), and all VOCs aggregated. The USGS has completed an
in-depth review of daily VOC data from a ten-year period from water utilities along the Ohio
River (Lundgren and Lopes, in press) and also have found no significant seasonal patterns for
VOC occurrence. While some data suggest that light VOCs might be more prevalent in cooler
months in surface waters, the patterns are not broadly evident.
While there are undoubtedly individual water systems or watersheds where seasonal patterns
could be productively targeted, this would need to be developed by individual States and systems
from detailed, local information. (There are likely seasonal patterns for light petroleum fuel
related compounds in reservoirs or water bodies with heavy boating use, for example.) Unlike
many SOCs, no general patterns for VOCs are evident on a regional, let alone a national basis.
58
-------
A Review of Contaminant Occurrence in Public Water Systems
Ground Water Systems in Mmois-Xylen* Detections
£ "T-
3
V .
n n L = J 1 * _R_n
** 1 1 n j~~ 1 1 K ;- iri i
j Jan Feb Mar Apr May Jun Jul Aug Sep Oct Nov Dec
20%
16%
16%
| 14%
1 «*
M 10%
! 8%
c
fe 6%
u oy*
2% -
0% i
Surface Water Systems in H!ino>*-Xyt*n« Detections
r-i
n n
-
-
- n
'• -4T
r
ns
n :1 I-M
Jan Feb Mr Apr kby Jun Jul Aug Sep Oct Nov Dec
Ground Water Systems in Mfchigan-Xytone Detections
35% . ,., - ,
30%
£ 10%
Jan Feb Mar Apr May Jun Ju) Aug Sep Oct Nov Dec
Ground Water Systems in Or»gon«Xyton>o Detections
tyy
»CT
U I I nil I
Jan feb Mar Apr May Jun Jul Aug Sep Oct Nov Dae
i
j
i
Surface Water Systems in Mfehigeft-Xyfen* Ptteetkms ;
I
1 20%
• 15%
• 10%
0.
5%
0%
n Fl
H II )
_ r
i
!
-"
I
- - ~
"
_n
f II 1
Jan Feb Mar Apr May Jun Jul Aug Sep Oct Nov Dec
12%
10%
1 **
* 6%
2
Surface Water Syltemi in Oagon-Xylen* Datections
n
j
•*
Jan Feb P*r Apr May Jun ,
1
1
: ; ''•> \
''. j ". i
M Aug Sep Oei No* Dec
Figure V.C.5. Percentage of systems with detections of xylene, by month, for three States.
59
-------
A Review of Contaminant Occurrence in Public Water Systems
Ground water Syitams in Atabanw-XyJemi Detections
9%
8%
7%
6%
° 5%
0%
;?
;
t;
C
n
s
r
-
4tn^
_HJ j
J»n Fed Mar Apr May Jun Jut Aug Sep Oct Nov Dec
Ground Water Sy«Wms in Aiabama-Totiiene Detections
12%
10%
8%
f-H
!~
1
Jan Fee Uv Apr May Jun Jul Aug Sep Oct Nov Dec
Ground Water Syitemiln *labani»a*»Bim Cwecoon*
7%
6%
3%
Bi
H
Jan Feb Mar Apr May Jun Jul Aug Sep Oct Nov Dec
9% 1
Surface Waters
y«
—
terns m Mabame-XyHme Detections
i-.
--;' :-
54%
3%
0%
ff
Jan Feb Mar Apr May Jun Jul Aug S«p Oet NOT Dec
I 8%-
1
n
-
"
n Fi
J
i-
t:
:'
?
dtkei
n
i
!
j
i
^
^ * l^
Surface W«Ur Symmt in Mnnimi n«nnm DmrtloiM
7%
«%•-
Jan Fed Mar Apr May Jun Jul Aug Sep Oct Nov Dec
Figure V.C.6. Percentage of systems with detections of xylene, toluene, and benzene, by month,
for Alabama.
60
-------
A Review of Contaminant Occurrence in Public Water Systems
16% •
14% •
12%
>.
o
= 6% •
0
Ground Water Systems in Illinois
Detections
-TBlractiiorotthyte ne
-TUn
D n i_n3
n n
; Jan Feb Mr Apr Stay Jun J
14% •
Q> 8% •
0
£ 4A
„
11 n H n n
LJI Aug Sep Ocl Nov Dec
Ground Water Systems In IIHnois-Trlcrttoroethyren*
Detections
ftrrfrti
irii
.
TTrflTTn1
Jan Feb Mar Apr May Jun Jul Aug Sep On Nov Dec
Gr
14%
12%
I 10%
I-
| 6%
§ 4%
£
2%
0%
ound Water Systems in HKnois-Vinyl Chloride Detections
nmnPlra Flra ,-, :
JBn Fob Mar Apr
htoy Jun ,
U Aug Sep Oct NOV Dec
16%-
14% -
o
Surface Water Systems in Nttnais-Tetrschloroethylene
Detections
on
II
;
n JTI
' P 1 i
Jan Feb Mar Apr May Jun Jul Aug Sep Ocl Nov Dec
14%
*>
(A <"•
B e%
g
i
Surface Water Systems in Mtools-TrlcMorotUiyKne
Detections
',
•
n n
n
1
: 'i
' n '
n n I.I n i
Jan Feb Mar Apr May Jun Jul Aug Sep Oct Nov Dec
Percent o» Systems
3S*33SSS
Surface WW»r SysUmt in Wmote-Vlnyl ChlorMe
Dettctions
-
^
K
i B
': 1
i n ;
Jan Feb Ktar Apr Ntey Jun Jut Aug Sap Oct Nov Dec
Figure V.C.7. Percentage of systems with detections of tetrachloroethylene and
trichloroethylene, by month, for Illinois.
61
-------
A Review of Contaminant Occurrence in Public Water Systems
Surface Water Systems In low a-Detections for 21
VOCs
35%
30% - -
25% ••
c 20% ••
S. 15%
10%
5% -t-
0%
HTTP
Ho Surface Water Systems in Iowa with Detections > 1/2 MCL (or
21 VOCs
Jan Feb Mar Apr May Jun Jul Aug Sep Ocl Nov Dec i
I
Ground Water Systems in Iowa-Detections for 21
VOCs
30%
25%
20%
•£
g 15%
*,0%
0%
1-
»
S,
J,
z
5
1
, «,
?'
™
A
*i
1
?
s
| ""
••c
«*
k,
»
§
-
:
•i
li
•r
j?
?
1
•
i
!
|
i
Ground Water Systems in km a-Detections > 1/2
MCL for 21 VOCs
Jan Feb Mar Apr May Jun Jul Aug Sep Oct Nov Dec
6%
5%
4%
3%
2%
1%
0%
1.
Jan Fat) Mar Apr May Jun Jul Aug Sep Ocl Nov Dec
Figure V.C.8. Percentage of systems with detections (>MRL, >0.5MCL) of any of the 21
regulated VOCs, by month, for Iowa.
V.C.3. Implications For Monitoring
The impetus to target monitoring to vulnerable time periods has been the recognition that simple
quarterly monitoring often does not adequately account for seasonal peak concentrations for
compliance analysis or exposure estimates. Figure V.C.9 shows a schematic annual
concentration record (generated from herbicide concentration data from a Midwestern river with
high-ground water baseflow). This configuration is typical of the strong seasonal occurrence
patterns that have been studied. Without special water treatment by the PWS, these
concentration trends would also be apparent in finished drinking water that uses this source
(Hallberg, 1989a; Hallberg et al., 1996). Quarterly water samples could be collected at times
labeled A during the year. This sampling regimen would, by chance or by choice, significantly
underestimate the annual average concentration. Even scenario B, which collects one quarterly
sample during the May-July peak period, would underestimate occurrence.
Statistical studies of sampling strategies in surface water (e.g., Battaglin and Hay, 1996) show
that incorporating sampling during spring and early summer runoff periods provides a more
accurate representation of annual occurrence than random quarterly sampling (that can avoid
these months) — as in scenario A on Fig. V.C.9. In these studies, the USGS evaluated the
62
-------
A Review of Contaminant Occurrence in Public Water Systems
effects of 10 sampling strategies on the estimates of annual mean concentration of herbicides,
such as atrazine. The accuracy of the strategy was computed by comparing time-weighted annual
mean concentrations calculated from detailed water sampling at 17 locations with the annual
mean estimated by each sampling strategy, using 1,000 Monte Carlo simulations for each
strategy. (In other words, each sampling strategy was simulated using 1,000 different
combinations of sampling times throughout the year.) The results were compared to a tolerance
value around the actual mean from the detailed water-quality data. Pertinent results are
summarized in Table V.C.I. A value of ± 0.75/^g/L around the actual mean (which is 25% of the
MCL of 3^ig/L for atrazine) is used for the tolerance. The table summarizes the percentage of
sampling simulations within the tolerance, or that are over (overestimating) or under
(underestimating) the actual mean plus or minus the tolerance value. Quarterly sampling
underestimated the mean in 20% of the random simulations; it was within the tolerance 63% of
the time — assuming a random distribution. (The quarterly results appear much more accurate
than scenario A would imply because the random simulation results in at least one-third of the
simulations collecting samples during peak months.) Monthly sampling was the most accurate,
but this would increase sampling requirements three-fold from the current 4 samples/year
(quarterly). However, three scenarios are nearly as accurate as monthly, and would not require
additional sampling. Strategies sampling once each in May and June, and considering the other
10 months as zeros, or once in April, May, and June with 9 zeroes, or once each in April, May,
June, and July with 8 zeros, range from 81 % to 84% within the tolerance of the actual annual
mean. A sampling scenario such as C, on Figure V.C.9, could provide a much more accurate
view of drinking water quality, while still only requiring four samples per year. This scenario
targets 3 samples in the typically vulnerable months, and collects a fourth sample during the off-
season to provide a more complete record. With the type of concentration range often seen in
strongly seasonal contaminants, the fall-winter background sample would have a similar
numerical effect as the "zero" assumption in the simulations, but would provide a more
continuous record.
Ground water studies (e.g., Pinsky et al, 1997) suggest that the more vulnerable ground water
settings also show peaks during these periods, and such targeting could also be appropriate in
these settings. From the data and literature reviewed, such a targeting strategy for SOCs would
be adequate to improve the effectiveness of compliance monitoring and accuracy of exposure
estimates. Most of the data suggest that most organic contaminants will vary in the same cycle
— or, as with many VOCs, will show little systematic or seasonal variability. Hence, VOCs
could be sampled on a similar schedule with SOCs and not lose resolution. However, this
approach will always be most effective if States and systems use their local knowledge to define
seasonal vulnerability patterns, and adjust such schedules. For example, in the Pacific west,
some pesticides show peak concentrations in fall-winter, because of the use of pesticides on
orchards during their dormant season, this is the rain/runoff season in this climate, and, in some
cases, related to reservoir release schedules (e.g., Kuivila and Foe, 1995; Larson et al., 1997).
There is no simple, single guideline that fits all situations that will improve the accuracy of
monitoring while balancing it against the burden of monitoring. All the factors discussed in this
report should be considered to devise strategies to refine monitoring to target high risk situations
and reduce monitoring burden on those systems with lower levels of risk.
63
-------
A Review of Contaminant Occurrence in Public Water Systems
10-
= 1.0-
0.1-
1st Quarter
Figure V.C.9. Schematic annual contaminant concentration profile, derived from actual data
from a Midwestern stream, with three sampling scenarios (A, B, and C) noted (with four
sampling times for each).
64
-------
A Review of Contaminant Occurrence in Public Water Systems
Table V.C.I. Percentage of Monte Carlo sampling simulations that are within, over, or under
the tolerance of the time-weighted annual mean atrazine concentration calculated from detailed
field sampling. Battaglin and Hay, 1996.
Percentage of Monte Carlo simulations within, over, or under ± 0.75 ^g/L of the time-weighted
annual mean atrazine concentration; 1,000 simulations per sampling strategy, for all sites.
Sampling Strategy
1 each, April, May, June, July
Quarterly
1 in June w/ 1 1 zeros
1 each in May, June, w/ 10 zeros
1 each in April, May, June w/ 9 zeros
1 each in April, May, June, July w/ 8 zeros
Monthly
Percentage of si
M&
within
39%
63%
58%
81%
82%
84%
85%
mutations within to]
^L(±25%oftheM<
over
53%
16%
1%
5%
5%
6%
7%
eranceof ±0.75
:D
under
7%
20%
40%
14%
13%
9%
8%
After Battaglin, W.A. and L.E. Hay. "Effects of Sampling Strategies on Estimates of Annual Mean Herbicide
Concentrations in Midwestern Rivers." Environmental Science & Technology, v. 30, pp. 889-896. 1996.
VI. CO-OCCURRENCE
An analysis of the co-occurrence of contaminants, i.e., the correlation of occurrence of pairs of
contaminants, was also conducted. This was of interest to assess if various contaminants are
associated with one another for consideration of monitoring scenarios, or prediction of exposure,
and/or to assess if some contaminants co-occur commonly enough that one might proxy for
another as a trigger for more complete monitoring. Some States, for example, have considered
using the detection of certain key organic chemicals as indicators to require more complete
monitoring for VOCs and SOCs.
There are two basic approaches that were considered for this assessment. The first approach
would consider if two contaminants were or were not detected in the same sample (i.e., in
samples collected from the same location on the same date). This is the most restrictive test that
could be applied. The second approach would consider if two contaminants were or were not
detected in the same system, over some period of time. This more liberal approach was taken as
the first step, because occurrence of contaminants is generally quite low and co-occurrence will
be difficult to evaluate and demonstrate. The less restrictive system-based analysis should yield
higher statistical correlations than a sample-based analysis. This could then be used as an
indicator of associations that might warrant further investigation. As discussed below, there were
few significant findings from this analysis.
65
-------
A Review of Contaminant Occurrence in Public Water Systems
VIA. Correlation Methods
An analysis of the correlation of contaminant occurrence by system was conducted for the Phase
n/V contaminants and required unregulated contaminants from data from some National Cross-
Section States and on the URCIS database. Conducting the analysis by system, rather than
sample, removed the restriction of time from the study. Thus, an occurrence of two contaminants
in the same system two years apart would produce a correlation even if the events are unrelated.
However, this method will also capture associations between contaminants where one compound
might be a degradation by-product of another and simply occurs in the same system at a later
date.
The data were organized in such a way that if a compound was detected in a system at least once
that compound was given a score of 'one'. A contaminant which appears multiple times in a
system is still counted only once. If a contaminant did not appear in a water system over the
course of a sampling period that compound variable was given a score of 'zero'. In other words,
concentrations were not used or compared, just simple presence-absence. Correlation matrices
(cross tabulations) were developed with all the contaminants compared to each other in paired
fashion. (This produces a 64-by-64 matrix of results for Phase n/V contaminants for each State,
for example.) Non-parametric correlation tests were computed (considering the simple 1-0
approach noted) using the SAS statistical program.
Spearman correlation coefficients (rho, or rs) were computed, the values of which vary between 1
(complete concordance) and -1 (complete discordance). The correlation coefficient is an
indication of how frequently two compounds occur in the same system as well as how often
neither occurs in a system. When a pair of compounds occur together and don't occur together
the correlation coefficient will be high. When two compounds frequently occur (and not occur)
independently of one another the coefficient will be low. If each compound only appears in a
system where the other is absent, the coefficient may be negative.
VLB. Correlation Results
Very few strong correlations were found, even though the probability values are highly
significant for many associations. Every correlation coefficient greater than 0.3 was statistically
significant, at p = 0.001. The magnitude of the p values is, in part, related to the large number of
data. In essence, while there were many significant associations among contaminants the
associations were not strong predictors. A correlation coefficient of less than 0.8 is not very
useful as a predictor of co-occurrence. The vast majority of r values were less than 0.5 and this is
analogous to an r2 value that only explains approximately 25% of the variance.
While a high correlation coefficient is a strong indication of frequent co-occurrence within a
group of systems, a low coefficient does not necessarily mean that occurrence of one compound
is not related to the other. For instance, in a given State, 100% of systems which have detections
of ethyl benzene also have detections of total xylenes but only 33% of systems which detected
total xylenes detected ethyl benzene. The correlation coefficient in this case is only 0.57; the
statistic doesn't give any indication that, in this State, the presence of one contaminant only takes
66
-------
A Review of Contaminant Occurrence in Public Water Systems
place in the presence of the other. It is also true that contaminants which are commonly found in
a State's water system may be highly correlated even if occurrence is mutually unrelated.
The few stronger correlations that appeared, that were repeated in more than one State, were not
particularly enlightening. There were some associations among lOCs. The lOCs occur naturally,
and are more ubiquitous than most organic chemicals, so this was not surprising. The others
showed associations among: SOCs-Group 1 contaminants, particularly atrazine and simazine;
VOCs-Group-1, the LNAPLs or "floaters", related to gasoline compounds. Others had obvious
explanations, such as associations among o-xylene and p-xylene, or o-xylene and total xylenes.
A few larger scale examples, at the contaminant group level, may be illustrative. These examples
also suggest the other inferences that may be drawn from the group data. The national cross-
section shows that 41 % of all surface water systems have detections of VOCs (any VOC). Of
these systems, 20% have detections of VOCs-Group 1, and 11% have detections for VOCs-
Group 2. Group 1 and 2 total 31 %. The two groups comprise the majority of the VOCs. If the
total percentage of systems with detections of VOCs were 20% (i.e., less than the total of the two
groups) it would suggest substantial co-occurrence between Group 1 and 2 compounds.
However, the total is 41%, considerably more than the total of the two Groups, suggesting little
co-occurrence.
The other relationship of interest raised by workgroup and stakeholders was the relationship of
SOCs and VOCs. Table VI.B.1. summarizes some results and illustrates the variability among
States, as well as the effect of low occurrence rates on assessment of co-occurrence. In the high-
occurrence State, 92% of surface water systems have detections of some SOC. Hence, there has
to be some substantive overlap with systems that also have detections of VOCs — particularly
when 90% of systems had detections of some VOC.
Table VI.B.1. Occurrence and co-occurrence of SOCs and VOCs in two States.
% Systems with detections of:
Any SOC
Any VOC
Any SOC AND VOC
Only SOCs
Only VOCs
AnvSOC/lM)/O/?VOC
High-Occurrence State
SW systems
92%
90%
82%
9%
7%
99%
GW systems
16%
66%
12%
5%
53%
69%
Low-Occurrence State
SW systems
4%
15%
0%
4%
15%
19%
GW systems
3%
13%
1%
2%
12%
15%
Indeed, 82% of surface water systems showed co-occurrence, i.e., the systems had detections of
both an SOC and a VOC. This left 9% of systems with detections of just SOCs, and 7% of
systems with only VOCs. These values are additive, and in sum, 99% of surface water systems
67
-------
A Review of Contaminant Occurrence in Public Water Systems
had detections of either an SOC and/or a VOC. For ground water systems there is not nearly
such a strong association. Only 12% of ground water systems show co-occurrence (any SOC and
VOC), while 58% had only an SOC or only a VOC.
Similarly, the low-occurrence State, shows the difference of occurrence patterns. For surface
water systems, there is no co-occurrence at all between SOCs and VOCs, and only 1% co-
occurrence for ground water systems. This further illustrates the generally different source
relationships between SOCs and VOCs, as discussed elsewhere in this report. The analysis of
co-occurrence does not provide any added insight for development of monitoring strategies.
Even at the group level, the analysis indicates that SOCs and VOCs require separate
consideration.
VII. SUMMARY AND CONCLUSIONS
This study provides the most extensive analysis EPA has conducted to date for the occurrence of
contaminants regulated under the National Primary Drinking Water Regulations, implemented by
EPA's Office of Ground Water and Drinking Water under the Safe Drinking Water Act (SOWA).
The primary impetus for the study has been the interest in Chemical Monitoring Reform (CMR)
in response to concerns about the efficacy of monitoring, the associated regulatory burden, and
the need to consider Alternative Monitoring Guidelines ("Permanent Monitoring Relief) for
PWSs under the 1996 Amendments to SDWA. A more comprehensive review of occurrence
data was needed to inform the deliberation of chemical monitoring strategy options. It is hoped
that the findings and results of this study will be helpful in other assessments to make continuous
improvements in drinking water programs.
Data Sources
The most important sources for this analysis were the State databases that provided SDWA
compliance-monitoring data from public drinking water systems. The data includes the
analytical results from Community Water Systems (CWSs) and Non-Transient Non-Community
Water Systems (NTNCWSs) required to monitor for Phase WV chemicals. Data from 14 states
were reviewed. Data from 12 states were analyzed in detail. The data used in this study
represent more than 10.7 million analytical results from nearly 26,000 public water systems.
More than 70 percent of the data utilized date from 1993 or later; this proportion is even greater
for most synthetic organic compounds (SOCs).
Four additional sources of national data were included:
• Drinking water data from systems operated by the American Water Works Service
Company, which provided a multi-state perspective as a comparison to the representative
cross-section results, and provided useful insight on various data quality issues.
• Regional analysis of atrazine and simazine occurrence in CWSs conducted by Novartis
Crop Protection under a special study for EPA's Office of Pesticide programs, which
compiled CWS monitoring data between 1993 and 1996.
68
-------
A Review of Contaminant Occurrence in Public Water Systems
* EPA's Unregulated Contaminant Monitoring Information System (URCIS) which
includes data from about 40 States and territories from the first round of unregulated
contaminant monitoring (1989-1993).
* USGS Studies, particularly results from National Water Quality Assessment Program.
While these data were useful for reviewing national occurrence patterns, the majority of
these data represent ambient water and not drinking water. These data were extracted
from 75 reports and covered parts of 42 States.
Data Management and Handling
The large, unique databases used presented many challenges that could affect the resultant
analysis. Hence, some data management issues warrant review. In developing an approach to
data management and handling, the primary objective was to develop a consistent and repeatable
analytical method that would allow valid comparisons between and among the various data sets,
thereby enabling an overview of occurrence patterns at the national level. AH data sets required
some formatting and editing to facilitate analysis. Variables were cross-checked for consistency
and to assure that each was represented or converted to represent the same units. While analysis
of the data was consistent from one data set to another, each data set required some unique
editing and filtering because of differences among basic data elements. These were corrected
prior to analysis. All statistical analyses were conducted in SAS statistical software, with data
formatting problems corrected in Microsoft Excel supplemented by specialized programs written
in Visual Basic prior to analysis.
Basic data elements were reviewed to ensure consistent and comparable analyses, including:
• Contaminant Identifiers: In general, EPA method numbers were used, adjusted for
unique method ID numbers used in some States.
• Detection or Reporting Limits: Minimum Reporting Levels (MRLs) for contaminant
methods were also cross-checked.
* Analytical Results Elements: Special care was taken to assure that results were reported
or translated to identical units.
• System and Sample Elements: These included source water type (surface water, ground
water, both, purchased or other); other system features included by individual States (such
as population served, system type or other locational data); and a variety of sample
identifiers.
• Other Codes: These were carefully reviewed to determine their applicability for inclusion
and to ensure that they were entered or converted to the same units.
In addition to the data quality and consistency issues addressed above, a number of States
included unique data elements or provided a unique treatment of common elements. These were
resolved in consultation with the States. Whenever errors or ambiguities could not be resolved,
69
-------
A Review of Contaminant Occurrence in Public Water Systems
data were eliminated from the analysis to avoid aberrant results. However, it should be noted
that very few data were excluded for these reasons, when compared to the thousands of analytical
results included in these data bases.
Differences in sampling frequencies were also identified as a potential analysis problem.
Systems with contaminant problems sample more frequently than those with no problems. This
analysis summarizes contaminant occurrence by systems, to avoid the skew that is inherent in
summaries by sample. For this analysis, only standard SDWA compliance samples were used;
"special" samples, or "investigation" samples that would bias results or other samples of
unknown type were not used.
Finally, there were a number of unique data quality concerns with the URCIS data. Most of the
data quality problems identified would bias the data in the same, conservative direction: the
occurrence of contaminants is likely overstated in both frequency and maximum concentrations.
Thus, aggregate values should not underestimate occurrence. Because URCIS is such a large
database, it has considerable value. The sheer quantity of data in URCIS overwhelms many
minor quality problems, and the spectrum of coverage (39 States) warrants review.
Sample Representativeness
While 14 of 50 States are a substantial sample, the States with available data do not constitute a
representative sample of the U.S. Therefore, various means were evaluated to construct a cross-
section from the available State databases that would provide a reasonable first view of national
occurrence. For this initial analysis, a simple approach was adopted that would be clear and
replicable, resulting in aggregate numbers that could be easily understood. Two broad factors
were considered: geographic or spatial diversity and pollution potential.
To attempt a representative view from the available State data the geographic diversity of
climatic and hydrologic conditions across the United States needed to be represented. In
addition, population and land use or activity factors were considered as indicators of pollution
potential. Measures of manufacturing, industrial, and agricultural (pesticide use) activity were
used as major indicators. From these considerations, data from 8 of the States were used to
construct the national cross-section. While the national cross-section cannot be stated to be
"statistically representative," the results clearly should be indicative of the central tendency of
national values for occurrence. Further, the data from the 8 Cross-Section States represent over
25% of the U.S. population using PWSs and over 20%of the PWSs sampling.
The national cross-section results were compared to the other limited national data, the URCIS
results, the Novartis study, and the USGS ambient water results. In short, the aggregated State
data, forming the national cross-section of contaminant occurrence compiled for this report,
appears to provide a conservative but reasonable approximation of national occurrence values.
The comparisons with other data suggest that the cross-section summary occurrence values are
likely slightly high. Yet the data are representative enough that the summary data are valuable
and some basic principles can be derived to guide further analysis and collection efforts.
70
-------
A Review of Contaminant Occurrence in Public Water Systems
Analysis Results
Data were analyzed for approximately 280 contaminants, including the 64 contaminants
regulated under the Phase rules, six other regulated contaminants (e.g., nitrate, lead and copper);
34 unregulated contaminants that were required to be monitored; 14 unregulated contaminants
that were considered discretionary for monitoring; and approximately 160 other unregulated
contaminants. Of the 160 others, 22 are on EPA's Contaminant Candidate List, including 16 that
appear on the proposed new unregulated contaminant monitoring list. The remaining 140 are
other unregulated contaminants reported in the State databases. Of the 10.7 million analytical
records, about 73 percent were for required monitoring.
If there is a generalization that can be made about contaminant occurrence in drinking water
systems, it is that occurrence is highly variable. All 64 contaminants currently regulated under
the Phase rules have been detected in drinking water systems, however, the frequency of
occurrence in water systems ranges widely. Only five of the 64 contaminants, all SOCs, do not
have a reported detection at greater than half their MCL. All lOCs, except asbestos, have been
detected in 90-100% of systems in some States. Only three lOCs (antimony, beryllium, and
cyanide) have not been detected in at least one system in at least one State (in either surface or
ground water). The high occurrence rates are not surprising for lOCs because they occur
naturally.
Twenty-nine of the 30 regulated SOCs have not been detected at all in some States; only
phthalate occurs in every State, but its occurrence relates, in part, to sample contamination from
plastics used in sampling, lab work, or system plumbing. (Because of various problems such as
this phthalate, adipate, and dioxin are not included in the summary discussions.) The greatest
range for any of the organic chemicals is for atrazine, which ranges from zero systems in some
States to 97 percent occurrence in surface water systems in Midwestern States (up to 9 percent in
ground water systems).
Fourteen of the 21 VOCs have not been detected in some of the participating States. The seven
that occur in every State studied, in either surface or ground water systems, are ethylbenzene,
cis-l,2-dichloroethane, tetrachloroethylene (PCE), tricloroethylene (TCE), vinyl chloride,
1,1,1-trichlorethane, and xylenes. Many VOCs occur in up to, or more than 30 percent of surface
or ground water systems in various States.
Only nine of the 64 contaminants occur in less than 1 percent of either surface water or ground
water systems. These are all SOCs: alachlor, carbofuran, chlordane, glyphosate,
hexachlorobenzene, hexachloropentadiene, oxamyl, toxaphene, and PCBs (though PCBs have
not been monitored as intensively as the other SOCs).
The analysis showed that MCL exceedances are not common. The compounds with the most
common exceedances are a mix of VOCs and SOCs. Three compounds occur both on the
surface water and ground water list, suggesting their widespread and common impact. The
general trend that emerges is that exceedances for SOCs tend to be more common for surface
water systems; for VOCs, exceedances tend to be more common for ground water systems
71
-------
A Review of Contaminant Occurrence in Public Water Systems
(though the differences from surface water are not great). Fluoride makes the ground water list,
but fluoride detections tend to be a result of natural occurrence and treatment additions.
Comparison of Contaminant Occurrence in Surface Water and Ground Water Systems
Fourteen percent of the regulated SOCs are detected in more than 5 percent of surface water
systems in the national cross-section. In contrast, no SOCs occur in more than 5 percent of
ground water systems. Nearly two-thirds of the 30 SOCs are detected in more than 1 percent of
surface water systems, but over three-fourths of these SOCs are detected in only a very few (less
than 1 percent) ground water systems. SOCs are far more common in surface water systems than
ground water systems and surface water systems have more exceedances as well.
For VOCs, the occurrence patterns reverse to some degree, but VOCs are more common than
most SOCs in both surface and ground water systems. VOCs are detected in surface water
systems more frequently than is generally recognized. In addition, VOCs show much greater
general occurrence in surface water systems but surface water systems and ground water systems
are nearly equal for the occurrence of exceedances. Ground water systems show slightly more
MCL violations for VOCs.
lOCs are about equally common (detected at concentrations greater than the MRL) in surface
water systems and ground water systems, but ground water systems generally have higher
average concentrations.
In short, surface water systems tend to be more vulnerable than, or equally vulnerable to ground
water systems for many contaminants, but they do not necessarily have more occurrences
exceeding an MCL. In particular, a greater percentage of surface water systems show detections
of SOCs and surface water systems show greater occurrence of VOCs, as well.
Ground Water Vulnerability
If variability is a confounding factor for surface water systems, it is the rule for ground water
systems. Ground water systems are highly variable in nature because ground water adds other
dimensions to the vulnerability equation. Aquifer type and well depth are key hydrogeologic
factors to ground water vulnerability.
The relative vulnerability of public water systems using ground water can be evaluated to better
target and customize monitoring requirements. Because of the variability in ground water
settings and system level variations, developing improved monitoring requires refined
information at the sub-State and system levels. Information on aquifer characteristics, well
depth, well casing and construction details, wellhead protection information, as well as system
and well maintenance history are some of the factors that should be considered.
72
-------
A Review of Contaminant Occurrence in Public Water Systems
System Size and Other Vulnerability Factors
In general, the same consistent patterns emerge from States with high contaminant occurrence
and from States with low contaminant occurrence. Consistent patterns were not observed for
lOCs, but were for organic chemicals - VOCs and SOCs.
The proportion of systems with detections of VOCs increases with system size from small to
large systems, particularly for ground water systems. While the same trend is evident for surface
water systems, it is not consistent for all contaminants. The trend is more evident for percent
detections (MRL) than for the percentage of systems with MCL exceedances (MCL). This can
be expected, in part, because the number of systems with exceedances becomes so low that they
may only occur in one size-class in a State.
This same trend in ground water is apparent for many, but not all SOC contaminants. For
high-occurrence States or URCIS, both of which have more data, increasing occurrence from
small to large systems is clearly illustrated. Patterns for surface water systems are variable -
sometimes increasing, sometimes decreasing with system size, again, in part because there are
relatively few systems with exceedances.
The data clearly indicate that, on a proportional basis, small systems do not show a greater
incidence of contaminant occurrence. For many contaminants, the larger systems show greater
occurrence and a tendency, at least for VOCs, to have proportionately more systems with
exceedances. This trend is most consistent for ground water systems and is a logical pattern in
many respects. Large PWSs need larger volumes of water. To get the larger ground water yields
needed, larger systems typically use unconfmed ground water supplies, that are typically shallow
and more vulnerable to contaminant occurrence.
The prevalence of contaminants in drinking water can often be related to the use of chemical
contaminants within the recharge zone or watershed of a public water system. The relative
contaminant source contribution should be considered when States revise their monitoring
programs.
Temporal Variability and Vulnerability
Water quality studies and monitoring throughout the US have clearly shown that occurrence
and/or concentration for some contaminants may vary over time, both seasonally and from year
to year. Targeting monitoring to vulnerable times can improve the effectiveness of compliance
monitoring and the accuracy of exposure estimates. However, there are concerns associated with
the cost-effectiveness of seasonal targeting patterns.
Many SOCs (the pesticide compounds, in particular) exhibit strong seasonal patterns because
their application or discharge into the environment is concentrated seasonally. In contrast, VOCs
do not typically show such seasonally in either source or discharge into the environment.
Studies of individual water systems or hydrologic settings sometimes show VOC patterns that
parallel seasonal hydrologic patterns; however, on a regional scale, no clear, general patterns
emerge. While there are undoubtedly individual water systems or watersheds where seasonal
73
-------
A Review of Contaminant Occurrence in Public Water Systems
patterns of VOCs could be productively targeted, this would need to be developed by individual
States and systems from detailed, local information.
For SOCs, statistical and modeling data suggest that sampling strategies could be designed to
better target and account for vulnerable, seasonal peak occurrence, without necessarily increasing
sampling frequency and burden. Few data suggest that multiple vulnerable periods would need
to be targeted in the same region. Vulnerable periods are apparent for surface waters and
shallow, vulnerable ground water systems. For deeper ground water systems few generalizations
can be made at the scale of this study, except to note that as well depth increases the degree of
temporal variability usually decreases.
Conclusions
The data compiled and analyzed for this study comprise the largest analysis of contaminant
occurrence data from Public Water Systems that EPA has conducted to date. The national cross-
section developed from State drinking water databases provide a reasonable overview of the
occurrence of the Phase H/V contaminants. These summary data provide an improved scientific
basis for evaluating monitoring and sampling programs.
74
-------
REFERENCES CITED
Barbash, I.E., and E.A. Resek. Pesticides in Ground Water, volume two of the series Pesticides
in the Hydrologic System. Ann Arbor Press, Inc., Chelsea, Michigan. 1996.
Battaglin, W., and L. Hay. "Effects of Sampling Strategies on Estimates of Annual Mean
Herbicide Concentrations in Midwestern Rivers." Environmental Science & Technology.
30:889-896. 1996.
Battaglin, W., and D. Goolsby. "Regression Models of Herbicide Concentrations in Outflow
from Reserviors in the Midwestern USA, 1992-1993." Journal of the American Water
Works Association. 1369:34-6. 1998.
Bloemen, HJ.Th., and J. Burn, eds. Chemistry and .Analy_sis_of^Volatile Organic Compounds in
the Environment. London: Chapman and Hall. 1993.
California Department of Health Services (DHS). Drinking Water Quality Monitoring Data
1984-1996; Annual Status Report. Monitoring and Evaluation Unit, California
Department of Health Services. November 1997.
Clarkson, J.R., N.A. Hines, D.P. Tierney, and B.R. Christensen. Human Exposure toAtrazine
and Simazine via Ground and Surface Drinking Water. Novartis Crop Protection, Inc.
Novartis Study No. 696-95. EPA MRID No. 44152122. 1997.
Coupe, R.H., D.A. Goolsby, J.L. Iverson, D.J. Markovchik, and S.D. Zaugg. U.S. Department of
the Interior. Pesticide, Nutrient, Water-Discharge, and Physical-Property Data for the
Mississippi River and Some of Its Tributaries, April 1991-September 1992. U.S.
Geological Survey Open File Report 93-657. 1995.
Hallberg, G.R., and D. Keeney. "Nitrate." In W.A. Alley, Regional Ground-Water Quality:
Chapter 2, p. 297-322. Van Nostrand Reinhold, New York, NY. 1993.
Hallberg, G.R., D.G. Riley, J.R. Kantamneni, P.J. Weyer, and R.D. Kelley. Assessment of Iowa
Safe Drinking Water Act Monitoring Data: 1987-1995. University Hygienic Laboratory
Research Report 97-1, p. 132. University of Iowa, Iowa City, IA. 1996.
Hallberg, G.R. 1989a. "Pesticide Pollution of Groundwater in the Humid United States." In H.
Bouwer and R.S. Bowman, eds., Effect of Agriculture on Groundwater; Agriculture,
Ecosystems, and Environment. Special edition, v. 26, p. 299-367. 1989.
x— ./
Hallberg, G.R. 1989b. "Nitrate in Groundwater in the United States." In R.F. Follett, ed.,
Nitrogen^Management and Groundwater Protection: Chapter 3, p. 35-74. Elsevier
Scientific Publications, Amsterdam, The Netherlands. 1989.
75
-------
A Review of Contaminant Occurrence in Public Water Systems
Hallberg, G.R., K. Woida, R.D. Libra, K.D. Rex, K.D. Sesker, B.C. Kross, L.S. Seigley, B.K.
Nations, D.J. Quade, D.R. Bruner, H.F. Nicholson, J.K. Johnson, and K.L. Cherryholmes.
The Iowa State-Wide Rural Well Water Survey: Site and Well Characteristics and Water
Quality. Iowa Department of Natural Resources, Geologic Survey Bureau, Technical
Information Series 23. 43p. 1992.
Howard, P.H., ed. Handbook of Environmental Fate and Exposure Data For Organic
Chemicals. Vol. I-V. Boca Raton: CRC Press. 1989-97.
Kalkhoff, S.J., D.W. Kolpin, E.M. Thurman, I. Ferrer, and D. Barcelo. "Degradation of
Chloroacetanilide Herbicides: The Prevalence of Sulfonic and Oxanilic Acid Metabolites
in Iowa Groundwaters and Surface Waters." Environmental Science & Technology.
32(11): 1738-1740. 1998.
Kolpin, D.W., D. Sneck-Fahrer, G.R. Hallberg, and R.D. Libra. "Temporal Trends of Selected
Agricultural Chemicals in Iowa's Groundwater, 1982-95: Are things getting better?"
Journal of Environmental Quality. 26:1007-1017. 1997.
Kolpin, D.W., I.E. Barbash, and RJ. Gilliom. "Occurrence of Pesticides in Shallow
Groundwater of the United States: Initial Results from the National Water-Quality
Assessment Program." Environmental Science & Technology. 32(5): 558-566. 1998.
Kolpin, D.W., SJ. Kalkhoff, D.A. Goolsby, D.A. Sneck-Fahrer, and E.M. Thurman.
"Occurrence of Selected Herbicides and Herbicide Degradation Products in Iowa's
GroundWater, 1995." Ground Water. 35-4, p. 679-687. 1997.
Kolpin, D.W., E.M. Thurman, and D.A. Goolsby. "Occurrence of Selected Pesticides and Their
Metabolites in Near-Surface Aquifers of the Midwestern United States." Environmental
Science & Technology. 30(1): 335-340. 1996.
Kross, B.C., G.R. Hallberg, D.R. Bruner, R.D. Libra, K.D, Rex, L.M.B. Weih, M.E. Vermace,
L.F. Burmeister, N.H. Hall, K.L. Cherryholmes, J.K. Johnson, M.I. Selim, B.K. Nations,
L.S. Seigley, DJ. Quade, A.G. Dudler, K.D. Sesker, M.A. Gulp, C.F. Lynch, H.F.
Nicholson, and J.P. Hughes. The Iowa State-Wide Rural Well-Water Survey, Water-
Quality Data: Initial Analysis. Iowa Department of Natural Resources, Geologic Survey
Bureau, Technical Information Series 19. 142 p. 1990.
Kross, B.C., G.R. Hallberg, D.R. Bruner, K. Cherryholmes, and J.K. Johnson. "The nitrate
contamination of private well water in Iowa." American Journal of Public Health. 83, p.
270-272. 1993.
Kuivila, K.M., and C.G. Foe. "Concentrations, Transport, and Biological Effects of Dormant
Spray Pesticides in the San Francisco Estuary California." Environmental Toxicology
and Chemistry. 14 (7), p. 1141-1150. 1995
76
-------
A Review of Contaminant Occurrence in Public Water Systems
Larson, S.J., RJ. Gilliom, and P.O. Capel. U.S. Department of the Interior. Pesticides in
Streams of the United States — Initial Results from the National Water Quality
Assessment Program. U.S. Geological Survey. Water-Resources Investigations Report
98-4222,92 p. 1999.
Larson, S.J., P.O. Capel, and M.S. Majewski. Pesticides in Surface Waters, volume three of the
series Pesticides in the Hvdrologic System. Ann Arbor Press, Inc., Chelsea, Michigan.
1997.
Lide, D.R, ed. CRC Handbook of Chemistry and Physics. 77th Edition. Boca Raton: CRC Press.
1996.
Lundgren, R.F., and T.J. Lopes. Occurrence, Distribution, and Trends of Volatile Organic
Compounds (VOCs) in the Ohio River and Its Major Tributaries, 1987-96. U.S.
Geological Survey Water Resources Investigations Report 99-XXX. 1999, in press.
Majewski, M.S., and P.D. Capel. Pesticides in the Atmosphere, volume one of the series
Pesticides in the Hvdrologic System. Ann Arbor Press, Inc., Chelsea, Michigan. 1995.
Novartis Crop Protection, Inc. Voluntary Atrazine Monitoring Program at Selected Community
Water Systems: Iowa 1996. Technical Report 6-97, Environmental and Public Affairs
Department. 1997.
Ohio Environmental Protection Agency. Pesticide Special Study. Division of Drinking and
Ground Waters. Columbus, 1998. Available on internet at
http://www.epa.state.oh.us/ddagw/pestspst.html. 1998.
Parsons, F., P.R. Wood, and J. DeMarco. "Transformations of Tetrachloroethene and
Trichloroethene in Microcosms and Groundwater." Journal of the American Water
Works Association. 76:56-9. 1984.
Pinsky, P., M. Lorber, K. Johnson, B. Kross, L. Burmeister, A. Wilkins, and G.R. Hallberg. "A
Study of the Temporal Variability of Atrazine in Private Well Water." Environmental
Monitoring and Assessment. 47, p. 197-221. 1997.
Ryker, S.J. and A.K. Williamson. Pesticides in Public Supply Wells of Washington State. U.S.
Geological Survey Fact Sheet 122-96. May 1996.
Schaap, B.D., and S.M. Linhart. Quality of Ground Water Used for Selected Municipal Water
Supplies in Iowa, 1982-96 Water Years. U.S. Geological Survey Open-File Report 98-3.
May 1998.
Scribner, E.A., D.A. Goolsby, E.M. Thurman, M.T. Meyer, and W.A. Battaglin. Concentrations
of Selected Herbicides, Herbicide Metabolites, and Nutrients in Outflow from Selected
Midwestern Reservoirs, April 1992 Through September 1993. U.S. Geological Survey
Open File Report 96-363. 1996.
77
-------
A Review of Contaminant Occurrence in Public Water Systems
Squillance, P.J., D.A. Pope, and C.V. Price. Occurrence of the Gasoline Additive MTBE in
Shallow Ground Water in Urban and Agricultural Areas. U.S. Geological Survey Fact
Sheet 114-95. 1996.
Squillance, P.J., M.J. Moran, W.W. Lapham, C.V. Price, R.M. Clawges, and J.S. Zogorski.
"Volatile Organic Compounds in Untreated Groundwater of the United States, 1985-
1995." Environmental Science & Technology. 33(23): 4176-4187. 1999.
Vowinkel, E.F., R.M. Clawges, D.E. Buxton, D.A. Stedfast, and J.B. Louis. Vulnerability of
Public Drinking Water Supplies in New Jersey to Pesticides. U.S. Geological Survey
Fact Sheet 165-96. 1996.
Westrick, J.J., J.W. Mello, and R.F. Thomas. "The Groundwater Supply Survey." Jour.
American WaterworksAssoc. 76,52-59, 1984.
78
J
-------
APPENDIX A
Summary Tables of Contaminant Occurrence Data
Table A-l: Summary comparison of occurrence of Phase n/V contaminants in water systems
using surface water vs. ground water, from a national cross-section of States. Percent
MCL violations derived from SDWIS information for 1/1/93-3/31/1998.
Table A-2: Summary of occurrence of Phase n/V contaminants from a national cross-section of
States. Range of values from all States.
(A-2a - Surface Water Systems
A-2b - Ground Water Systems)
Table A-3: Summary of occurrence of other regulated contaminants from a national cross-section
of States. Range of values from all States.
(A-3a - Surface Water Systems
A-3b - Ground Water Systems)
Table A-4: Summary of occurrence of required unregulated contaminants [1986-1999] from a
national cross-section of States. Range of values from all States.
(A-4a - Surface Water Systems
A-4b - Ground Water Systems)
Table A-5: Summary of occurrence of contaminants on Contaminant Candidate List and NEW
Proposed UCMR from a national cross-section of States. Range of values from all States.
(A-5a - Surface Water Systems
A-5b - Ground Water Systems)
Table A-6: Summary of occurrence of other not-regulated contaminants from a national cross-
section of States. Range of values from all States.
(A-6a - Surface Water Systems
A-6b - Ground Water Systems)
Table A-7: Summary of contaminant occurrence from the Unregulated Contaminant Monitoring
Information System (URCIS).
(A-7a - Surface Water Systems
A-7b - Ground Water Systems)
-------
-------
A Review of Contaminant Occurrence in Public Water Systems
Table A-l. Summary comparison of occurrence of Phase n/V contaminants in water systems
using surface water vs. ground water, from national cross-section States. Percent MCL
violations derived from SDWIS information for 1/1/93-3/31/1998.
Contaminant
Percent > MRL
Surface
Water
Ground
Water
Percent > Vt MCL
Surface
Water
Ground
Water
Percent > MCL*
Surface
Water
Ground
Water
National
%MCL
VioI.-SW
National
%MCL
VioL-GW
Individual Contaminants
lOCs
Antimony (total)
Arsenic
Asbestos (MCL in fibers per
liter)
Barium
Beryllium (total)
Cadmium
Chromium
Cyanide
Fluoride
Mercury
Nickel
Selenium
Thallium (total)
SOCs
2,3,7,8-TCDD (Dioxin)
2,4,5-TP (Silvex)
2,4-D
Alachlor (Lasso)
Atrazine
Benzo(a]pyrene
bis(2-ethylhexyl) adipate
bis(2-ethylhexyl) phthalate
Carbofuran (Furadan)
Chlordane
DaJapon
Dibromochloropropane (DBCP)
Dinoseb
Diquat
Endothall
Endrin
Ethylene Dibromide (EDB)
Glyphosate (Roundup)
Heptachlor
Heptachior Epoxide
Hexachlorobenzene
Hexachlorocyclopentadiene
Lindane
Methoxychlor
Oxamyl (Vydate)
Pentachlorophenol
Ptcloram (Tordon)
Simazine
Total PCBs
Aroclor 10] 6
Aroclor 1221
4.2%
13.0%
8.9%
49.1%
2.5%
5.1%
1 0.5%
5.1%
77.8%
9.0%
11.8%
11.2%
2.5%
0.0%
2.0%
11.2%
7.3%
21.1%
0.5%
4.9%
28.9%
0.8%
0.0%
9.4%
4.8%
2.5%
3.5%
0.3%
2.1%
4.2%
0.0%
0.4%
0.4%
0.4%
9.6%
1.2%
1.0%
0.0%
3.1%
3.7%
15.9%
0.2%
0.0%
0.0%
3.2%
19.3%
8.5%
47.3%
2.1%
4.9%
13.2%
2.0%
72.5%
4.5%
10.6%
8.6%
3.6%
1.3%
0.7%
1.2%
0.3%
2.0%
0.5%
6.8%
14.9%
0.1%
0.1%
0.8%
2.6%
0.4%
0.8%
0.2%
0.2%
1.0%
0.1%
0.2%
0.2%
0.0%
0.1%
0.3%
0.2%
0.1%
0.7%
0.5%
1.4%
0.2%
0.1%
0.1%
0.8%
0.6%
0.7%
0.6%
0.4%
1.3%
0.3%
0.4%
0.8%
1.3%
1.2%
0.2%
0.8%
0.0%
0.0%
0.2%
1.5%
13.2%
0.0%
0.5%
3.2%
0.0%
0.0%
0.2%
1.1%
0.0%
0.0%
0.0%
0.2%
3.8%
0.0%
0.0%
0.0%
0.0%
1.0%
0.3%
0.0%
0.0%
0.4%
0.0%
2.5%
0.2%
0.0%
0.0%
1.2%
1.6%
0.9%
0.8%
0.5%
1.2%
0.5%
0.5%
3.4%
0.7%
0.9%
0.3%
1.3%
1.3%
0.0%
0.0%
0.1%
0.3%
0.1%
0.4%
2.7%
0.0%
0.0%
0.0%
2.3%
0.0%
0.0%
0.0%
0.1%
1.0%
0.0%
0.1%
0.0%
0.0%
0.0%
0.1%
0.0%
0.0%
0.1%
0.0%
0.0%
0.1%
0.0%
0.0%
0.2%
0.5%
0.7%
0.5%
0.0%
0.2%
0.2%
0.0%
0.5%
0.5%
0.4%
0.0%
0.0%
0.0%
0.0%
0.0%
0.2%
10.7%
0.0%
0.5%
2.8%
0.0%
0.0%
0.2%
1.1%
0.0%
0.0%
0.0%
0.2%
3.7%
0.0%
0.0%
0.0%
0.0%
0.6%
0.3%
0.0%
0.0%
0.2%
0.0%
1.0%
0.2%
0.0%
0.0%
0.4%
0.9%
0.4%
0.2%
0.2%
0.6%
0.2%
0.2%
1.3%
0.4%
0.4%
0.2%
0.4%
1.3%
0.0%
0.0%
0.0%
0.1%
0.1%
0.3%
1.7%
0.0%
0.0%
0.0%
2.0%
0.0%
0.0%
0.0%
0.0%
0.7%
0.0%
0.0%
0.0%
0.0%
0.0%
0.1%
0.0%
0.0%
0.0%
0.0%
0.0%
0.1%
0.0%
0.0%
0.08%
0.00%
0.00%
0.00%
0.02%
0.05%
0.00%
0.02%
0.02%
0.03%
0.02%
0.03%
0.12%
0.00%
0.00%
0.00%
0.00%
0.83%
0.00%
0.02%
0.03%
0.00%
0.00%
0.00%
0.00%
0.00%
0.00%
0.00%
0.02%
0.07%
0.00%
0.00%
0.00%
0.00%
0.00%
0.00%
0.00%
0.00%
0.00%
0.00%
0.00%
0.00%
0.00%
0.00%
0.06%
0.06%
0.01%
0.06%
0.02%
0.06%
0.01%
0.01%
0.18%
0.04%
0.01%
0.07%
0.06%
0.00%
0.00%
<0.01%
0.00%
0.01%
0.00%
0.00%
0.01%
0.00%
0.00%
0.00%
<0.01%
0.00%
0.00%
0.00%
0.00%
0.04%
0.00%
0.00%
0.00%
0.00%
0.00%
0.00%
0.00%
0.00%
<0.01%
<0.01%
0.00%
<0.01%
0.00%
0.00%
A-l
-------
A Review of Contaminant Occurrence in Public Water Systems
Contaminant
Arodor 1232
Aroclor 1242
Arodor 1248
Arodor 1254
Aroclor 1260
Toxaphene
VOCs
l,],l-Trichloroethane
1 , 1 ,2-Trichloroethane
1,1-Dichloroethene
1 ,2,4-Trichlorobenzene
1,2-Dichloroethane
1 ,2-Dichloropropane
Benzene
Carbon tetrachloride
Chlorobenzene
cis- 1 ,2-Dichloroethylene
Ethyl benzene
Methylene chloride
(Dichloromethane)
o-Dichlorobenzene
p-Dichlorobenzene
Styrene
Tetrachloroethylene (PCE)
Toluene
trans- 1 ,2-Dichloroethylene
Trichloroethene
(Trichloroethylene, TCE)
Vinyl chloride
Xylenes (Total)
Percent > MRL
Surface
Water
0.0%
0.0%
0.0%
0.0%
0.0%
0.3%
7.3%
5.7%
2.9%
L_ 3.1%
3.1%
3.2%
3.9%
9.0%
8.1%
3.3%
7.3%
25.6%
3.7%
6.2%
4.1%
7.1%
11.9%
2.7%
5.6%
3.1%
12.3%
Ground
Water
0.1%
0.1%
0.1%
0.2%
0.0%
0.1%
3.3%
0.7%
1.5%
1.0%
1.4%
1.0%
1.2%
1.7%
1.0%
1.9%
2.2%
11.1%
1.2%
2.0%
2.1%
4.3%
3.8%
0.7%
3.1%
0.5%
3.9%
Percent >VzMCL
Surface
Water
0.0%
0.0%
0.0%
0.0%
0.0%
0.0%
0.9%
0.6%
0.3%
0.0%
0.3%
0.6%
0.5%
1.6%
0.2%
0.3%
0.3%
10.4%
0.2%
0.0%
0.0%
2.5%
1.0%
0.0%
1.9%
0.3%
0.2%
Ground
Water
0.0%
0.0%
0.0%
0.0%
0.0%
0.0%
1.3%
0.4%
1.0%
0.4%
0.6%
0.4%
0.5%
0-6%
0.3%
0.7%
0.3%
3.3%
0.6%
0.9%
0.2%
2.3%
0.7%
0.2%
1.8%
0.2%
0.2%
Percent > MCL*
Surface
Water
0.0%
0.0%
0.0%
0.0%
0.0%
0.0%
0.9%
0.3%
0.3%
0.0%
0.3%
0.5%
0.3%
1.1%
0.2%
0.3%
0.3%
4.7%
0.0%
0.0%
0.0%
1.7%
0.5%
0.0%
1.2%
0.3%
0.0%
Ground
Water
0.0%
0.0%
0.0%
0.0%
0.0%
0.0%
1.3%
0.3%
0.9%
0.4%
0.4%
0.3%
0.4%
0.4%
0.3%
0.6%
0.1%
2.3%
0.5%
0.8%
0.2%
1.8%
0.4%
0.2%
1.5%
0.2%
0.1%
National
%MCL
VioL-SW
0.00%
0.00%
0.00%
0.00%
0.00%
0.00%
0.00%
0.00%
0.02%
0.00%
0.00%
0.00%
0.02%
0.00%
0.00%
0.00%
0.00%
0.03%
0.00%
0.00%
0.00%
0.13%
0.00%
0.00%
0.05%
0.00%
0.00%
National
%MCL
VioL-GW^
0.00%P
0.00%
0.00%
0.00%
0.00%
<0.01%
0.01%
0.01%
0.04%
0.00%
0.02%
0.01%
0.04%
0.02%
<0.01%
0.01%
<0.01%
0.05%
0.00%
0.00%
0.00%
0.14%
0.00%
<0.01%
0.12%
0.02%
0.00%
Group Summaries
lOCs
HOCs-AlI Regulated
SOCs
SOCs-Group 1
SOCs-Group 2
VOCs
VOCs-All Regulated
VOCs-Group 1
VOCs-Group 2
83.7%
21.9%
20.4%
41.1%
19.5%
11.1%
83.5%
2.4%
13.4%
19.9%
6.6%
6.4%
6.8%
13.2%
1.2%
15.4%
1.7%
4.5%
9.2%
0-3%
1.9%
7.9%
1.3%
3.9%
2.5%
10.7%
0.9%
8.2%
0.9%
2.9%
4.2%
0.1%
1.0%
6.1%
0.9%
3.2%
0.3%
1.0%
0.2%
0.6%
0.1%
0.4%
•
* % > MCL indicates the proportion of systems with any analytical results exceeding the concentration value of the MCL; it does not necessarily
indicate an MCL violation. An MCL violation occurs when (be MCL is exceeded by the average results from four quarterly samples or
confirmation samples as required by the primacy State.
IOC-Regulated: includes all the regulated lOCs.
SOCs-Gronp I: includes alachlor. atrazine, and simazine.
SOCs-Group 2: includes bis(2-ethylhexyl)phthalate, bis(2-ethyibexy|)»dipate, and benzo(a)pyrene.
VOCs-Regulated: includes all die regulated VOCs
VOCs-Group 1: includes benzene, ethyl benzene, toluene, and total xylenes (LNAPLEs)
VOCs-Group 2: includes cis-l,2-dichloroethylene, trans-1,2-dichloroethylene. l.l-dichloroethene, tetrachloroethylene, trichloroethene, and vinyl
chloride (DNAPLEs).
A-2
-------
.4 /?evi^w of Contaminant Occurrence in Public Water Systems
u
CX5
u
a
^
8 8
CO SS
i2 o
g^=
11
U o
•a "
O 00
s §
OS
I
o
v;
C3
O.
"S
i
u
ca
E
a
tfl
cs
I
i
i
s
u
U.
s« 3
(Ni O
ci! o
g
gg
g
g
g
S S
g
g
g
f\ S*
»|o>
gig
o o
8* S*
•nt in
01 o
a« 3?
n
g
a
1
gg
a? s«
gg
s
I
aS s!
° i
S
a! a!
gg
$ Si
f^i p'l (
o; o| 01 o
g.g.|
C4\ Of O
s*i;
g.gji
£ S
m CM
g
f f
Si
fig
gg
g
aS a« a«
S 9
gg
s?1 as
0*. U>
ol o>
#1 a?
o\ p
rjjTpf _j
o^l 3*-f 3s
CM| ix -
o! rt O
g
g
S
a? a*
1:
§
«
s*
— e
g
s.
I
g
a?
oil —
g
S
s
s
O; Oj
0(01
i|p
S
<= P
|Synt
lj
Oibr
be
A-3
-------
A Review of Contaminant Occurrence in Public Water Systems
ra
55
"o
c
Cfl
o
U
"5
o
'is
«
.£
- ^-v
«> T3
« "
«
il
al
o
ffft
u
u
3
on
4)
I
u
S
2 2 3
g
g
g
ggg
sie
g
8? S«i
CM n
I
g
ig
g
o«
O
g
g
si
ri
S§
8 :
Ot I
P3 I
S ;
o
>
1
*
i
U
S
ij
!
j
A-4
-------
A Review of Contaminant Occurrence in Public Water Systems
* g
o =
(4 IX
g
s?
# #1 a«
P-. p* <£
«-; c4 o
3NS5a^S*2?3?t^:^3^i5^o~«^5-
iw(M^rcv»-CMCV'pOS|cOC15«-;O
^bb^bbbbbbbbb
u
en
U
13
_o
CS
a
u
P
§Z
c «
'^ S
C3
C
2 C3
BO
3,0s
57
O.
UM
O
o
3
U
O
S* S!
i oF o
g
61
s?
o CD
3« #
r
»i
gli
g.9
g.g.
8*
3Si i
:gg
g
3? 3«
N O
g
g
9 T
~ICM;
o!
sSi a!
oi o
g
bi b
gg
M
o
it
3*i
§-s
g
g
g
g
«#
38 s!
g
4*
I
O
g
g
gg
3? 3!
g
£8
£ sli
§|S
S
i
1
ii
£
8
b
S
i
ss
1!
i
i§
i
S
S
s
s s
SIS
"I
«
ii
a
s
s
g s
S3S
rop
A-5
-------
A Review of Contaminant Occurrence in Public Water Systems
CO
(**
o
JO
s
§
"3
CO
5/3
« 9"
=
« CO
ll
•O
o
8
£
en
1
m
o o
g
gg
ri ri
I
1
gg
8
oi V
IS
g
i
"
a? s?
« -
S
I
§
o
1
I
s-
o
A-6
-------
A Review of Contaminant Occurrence in Public Water Systems
S
tww
o
oo
W3
o
u
cs
z
cs
C
00
B
S3
c E
2 >
c <~
o o
I
I
u
u
o
"o
^
CO
*
f^
<
^
A
£
£
o
u
I I
8
i 09
I I oo
I
a
o
U
»
c
a
O
ItJ
01 ID
O O
*\ 8s
O p
O O
rite
1
I
S
s?
8
6i '
I o ^
Ss
\ -§5
-%
Ibv
A-7
-------
A Review of Contaminant Occurrence in Public Water Systems
s
55
<*-
o
c
"I
CO
2
U
eo
I
e E
« £
e ""
te
|£
J
o
8
SI?
wa
*?
Ji
to .
k i
1
•D
I-'
o
I
I
M
1
U
!i
-------
u
S
A Review of Contaminant Occurrence in Public Water Systems
E
u
S
1>
u
, w
OT!
£
§
I °
If
S
S
oo
•o
Si
0£
w^
O
u
o
tl
"!;A
i
pi
X
s^
to
:|
i§
S? i?
jg
I d
1
S
o
1
s s
i« S»
o o
gg
E
g
SI
§
i
g
2 £
#1*
pi r-
T
gg
8
l§
g
g.;g.
o' o o
gg
Iff
"I I
gig
i
*! => =>i
•^ ol o
gg
g
S
§
A-9
-------
A Review of Contaminant Occurrence in Public Water Systems
2
O
es
o
"i
ra
.2
o>
MG
s|
c .E
00 VI
<3\ —
^ ~9
£ E
I!
C U
2-1
r^ **
o>
o
8
o
>,
3
a
I
s
o
§
i
z; 2
1 tra
If
8
11
fl
I!
f 5
?i
** •
A-10
-------
A Review of Contaminant Occurrence in Public Water Systems
c
U
CS
nj
c
o
£
i/>
1
1
o .
<= 8
5:1
d 5
0,0
6? !S
£i
S* 88
S Z
gglg
s* s«
z i
& ll
ia ~W
gi ««
2|g
|2
|g;
I o! 6
2 85
II
il
sS a«
gg
S! #
gg
* *
8
gg
8 81
p! .
o o
gg
gg
e ^
O
1
I
O
I
gj5
o
Si
n $ n
R S
«0
S
S SS S
-J.s
ss s
s s
s s
s s
s
-i-'
s s s
8»
ss
ss
glgls
-f
V
N
i_f
^-ki 5
A-I2
-------
A Review of Contaminant Occurrence in Public Water Systems
o
1
S
u
o
•a .
§=0
i %
O.5
e c
— o
5TU
vo 2
QC &T3
OS _
s
•o
-h
I
OS
0
O
I
qs
g
s
I
II
IS
!!
?i
B
!«•
A-13
-------
A Review of Contaminant Occurrence in Public Water Systems
«
CJ
<3
Cfl
C
u
§3
"O W3
s
Cvl U
T3 aa
1 s
go:
u .
1
c i—
ca o
1 1
§ 2
o --
U J
U r-
u E
C O
m\ .
I
VI
<
:rf i
ii ;
s?
u
1 I
i ">!
s«
gi =
88
w
u
l»
ds
O) CO
OJ
«=
col
S?
ql q
d d
&
zi
I
s
Si S
1 § «' :
r *i § ~
: m
3*
£
m
lbenz
s?
CM
d
*S" ~5
3*' b*
o —
£
3?
g
S
s
i
i
-S
P
violati
I
I
value of
f
S
-J
i
§
s
I
&£
f i
ii
A-14
-------
A Review of Contaminant Occurrence in Public Water Systems
CT)
IM
4J
"a
1
o
Oi
U
"2 gj
-------
A Review of Contaminant Occurrence in Public Water Systems
5
55
.2
1
co
6
II
«5 C
£ o
5 «
- b
Is2
^ -o
u H
w .
= fi
IA «
1 -
-
w
J
g£g|
88
5
§5
S S
S 3
# jt
*!*g
gg
8
g
S
8
§
88
g?
$3
ggf
8
ff|ssc
<*
gig
-1°
s si
si
g
s
g
ggg
gg
gg
?is
g
ggg
g
g
gggggg
6 o
ggggg
gg
S SSSS
s
o o
gg
gg
o
g
oj ei
g
S £33
3
s
gl
gg:
gg
odd
S£3
s:
o
u
O
GO
s
!!
s
Q ul
i
A-16
-------
A Review of Contaminant Occurrence in Public Water Systems
CO
Cfc«
o
>
u
or)
U
II
13 .£
Z c
« O
C ^
i ^
M
§
n
.= I
UQ ft>
C Q,
1 S
cd *s
fc E
•£ 2
>
"3
|
to
ve
gjggg
b b
§i
iii
g
g
I;
S .
u dol
i
g
g
g
#! 3?
O1 O
Oi O
g
I
g.ig.
O O
§;g
S 8
9
g
g.g.
g
S
g
gg.
g
g
o
g
o
g
O O
g.g.
g]
C3 (
11
§i
f
I
I
•s
I
J
i
j
A-17
-------
A Review of Contaminant Occurrence in Public Water Systems
o
e
u
es
.0
E .
E §
5 "
^ -O
•a 2
B C
§ 8
E «
O S.
.
(8 «8
CJ —
J* J
E>
O "tr
r
•o
ii
g
Is
SI S
al a s si si K
222
S 9
§
8
g
g
ss
§
8
ri ri
ii
6 o
g
o o e> 6
o 6
gg
g§gg
iSS PS
g
3
CC
I
i
Q Ul
tvi
A-18
U.S. EPA Headquarters Library
Mail code 3201
1200 Pennsylvania Avenue NW
Washington DC 20460
-------
A Review of Contaminant Occurrence in Public Water Systems
o
u
CO
en
co
e
u
1.5
1
-------
A Review of Contaminant Occurrence in Public Water Systems
Table A-7.a. Summary of Contaminant Occurrence in Surface Water Systems from the
Unregulated Contaminant Monitoring Information System (URCIS).
Contaminant
Phase II/V Regulated Contaminants
Synthetic Organic Compounds
Dibromochloropropane (DBCP)
Ethyjene Dibromide (EDB)
Volatile Organic Compounds
1,1,1 -Thchloroethane
1 ,1 ,2-Trichloroethane
1 ,1 -Dichloroethene
1 ,2,4-Trichlorobenzene
1 ,2-Dichloroethane
1 ,2-Dichloropropane
Benzene
Carbon tetrachloride
Chlorobenzene
cis-1 ,2-Dichloroethylene
Ethyl benzene
Methylene chloride (Dichloromethane)
o-Dichlorobenzene
p-Dichlorobenzene
Styrene
Tetrachtoroethylene
Toluene
trans- 1 ,2-Dichloroethylene
Trichloroethene (Trichloroethylene)
Vinyl chloride
Xylenes (Total)
o-Xylene
M-Xylene
P-Xylene
Required Unregulated Contaminants
(nv) 1,3-Dichlorobenzene
1,1,1 ,2-Tetrachloroethane
1 ,2,3-Trichloropropane
1 .3-DichloroDrooane
Chloroethane
Chloromethane
Dibromo methane
o-Chlorotoluene
p-Chlorotoluene
Surface Water Systems
No. of
Analyses
6,211
5,517
7,603
8,892
7,743
6.837
7,541
9.373
7.101
7,534
9,291
8,852
9.168
9,034
9.053
7.429
8.928
9,500
9,165
9,3431
7,851
6,438
1,906
6,993
6,578
5,958
9,153
8.669
8.698
8.709
9.160
9,322
8,635
8,273
7,973
No. of
Systems
1,606
1,395
1,393
2,007
1.539
1,691
1,372
1.972
1,288
1,385
2.017
1.989
2.056
2.008
1.948
1.498
1,985
1,816
2,065
1,974
1,394
1,363
687
1.609
1.404
1,357
1,971
1,905
1,910
1.899
2.035
2,051
1,872
1,854
1,841
>MRL
N
41
35
78
36
36
12
25
15
32
70
68
36
88
238
19
46
28i
56
169
23
93
17
82
47
31
33
15
12
5
2
26
49
34
11
6
%
2.55%
2.51%
5.60%
1 .79%
2.34%
0.71%
1.82%
0.76%
2.48%
5.05%
3.37%
1.81%
4.28%
11.85%
0.98%
3.20%
1.41%
3.08%
8.18%
1.17%
6.67%
1.25%
11.94%
2.92%
2.21%
2.43%
0.76%
0.63%
0.26%
0.11%
1.28%
2.39%
1.82%
0.59%
0.33%
>0.5 MCL
N
18
22
1
6
10
0
7
0
%
>l
N
1.12%| 15
1.58% 19
0.07%
0.30%
0.65%
0.00%
0.51%
0.00%
11 i 0.85%
11
0
2
0
119
0
0
0
22
0
0
30
11
0
0.79%
0.00%
0.10%
0.00%
5.93%
0.00%
0.00%
, 0.00%
1.21%
0.00%
0.00%
2.15%
0.81%
0.00%
0
4
4
0
7
0
4
5
0
0
0
67
0
0
0
16
0
0
18
8
0
MCL"
%
0-93%
1.36%
0.00%
0.20%
0.26%
0.00%
0.51%
0.00%
0.31%
0.36%
0.00%
0.00%
0.00%
3.34%
0.00%
0.00%
0.00%
0.88%
0.00%
0.00%
1.29%
0.59%
0.00%
N/A
N/A
N/A
0
0
0
0.00%
0.00%
0.00%
0
0
0
0.00%
0.00%
0.00%
N/A
N/A
32
1.56%
20 1 0.98%
N/A
0
0
0.00%
0.00%
0
0
0.00%
0.00%
A-20
-------
A Review of Contaminant Occurrence in Public Water Systems
Table A-7.a. Summary of Contaminant Occurrence in Surface Water Systems from the
Unregulated Contaminant Monitoring Information System (URCIS) (Continued).
Contaminant
Discretionary Compounds
1 ,2.3-Trichlorobenzene
Surface Water Systems
No. of
Analyses
No. of
Systems
6,437 1,542
1,3,5-TrimethyJbenzene 6,154
Bromochlorornethane €.637
Diehlorodifluoromethane 6,933
isopropvlbenzene
n-Butvl benzene
n-Propylbenzene
p-lsopropyltoluene
sec-Butvlbenzene
6,314
6.483
6,467
6,060
6.216
tert-Butvlbenzene | 6,214
TricWorofluoromethane 7,353
cis-1 .3-DichloroDropene
Total 1.2-Dichloroethene
trans-1 .3-Dichioronropene
Other Reautated Contaminants
irornodichlorornethane
Brornoform
Chlorodibromometriane
Chloroform
Total THMs
Contaminant Candidate List
1 ,1.2,2-Tetrachloroethane
1,1-Oichloroethane
1 , 1 -Dichtorooropene
3,038
3.900
11.419
10,179
10,639
11,987
8,988
9.471
8,607
1.2.4-Trimetriylbenzene 6,276
2.2-DiohloroDroDane
Bromobenzene
Bromomethane
Hexachtorobutadiene
Naphthalene
Total 1.3-Dichlorooropene
8,802
6,928
9.298
5,958
6.466
3.771
1,524
1.538
1,744
1,524
1.523
>MRL
N
%
10: 0.65%
14! 0.92%
18
29
6
13
1.516! 5
1.435 3
1,489 3
1,489
1,792
953
1.073
2.611
2,146
2.378
2.736
2,020
2,040
1,899
1,519
1,905
1.878
2.039
1.538
1.620
947
1.17%
1.66%
0.39%
0.85%
0.33%
0.21%
0.20%
3! 0.20%
46 2.57%
14
14
2,113
474
1,535
2,341
21
39
8
21
5
11
27
10
, 33
r e
1.47%
1.30%
80.93%
22.09%
64.55%
85.56%
1.04%
1.91%
0.42%
1.38%
0.26%
0.59%
1.32%
0.65%
2.04%
0.63%
>0.5 MCL
N
%
>MCL*
N
%
N/A
N/A
2 1 0.13%
0
0.00%
1
0
0.07%
0.00%
N/A
N/A
N/A
N/A
N/A
N/A
N/A
N/A
N/A
N/A
N/A
63
2.65%
7[ 0.29%
N/A
N/A
N/A
N/A
N/A
N/A
N/A
5
7
2
0.25%
0.46%
0.12%
K
4
6
0
0.20%
0.39%
0.00%
I/A
N/A = There is no MCUHAL for this contaminant.
* % > MCL indicates the proportion erf systems with any
an MCL violation. An MCL violation occurs when the Ml
rewired bv the Drimacv State
3L is exceeded by the average
concentration value of tfte MCL; it does not necessarily indicate
results from lour quarterly samples or confirmation samples as
A-21
-------
A Review of Contaminant Occurrence in Public Water Systems
Table A-7.b. Summary of Contaminant Occurrence in Ground Water Systems from the
Unregulated Contaminant Monitoring Information System (URCIS).
Contaminant
Phase li/V Requlated Contaminants
Synthetic Organic Compounds
Dibromochloropropane (DBCP)
Ethylene Dibromide (EDB)
Volatile Organic Compounds
1,1,1-Trichloroethane
1,1.2-Trichloroethane
1,1-Dichloroethene
1 ,2,4-Trichlorobenzene
1 ,2-Dichloroethane
1 ,2-Dichloropropane
Benzene
Carbon tetrachloride
Chlorobenzene
cis-1 ,2-Dichloroethylene
Ethyl benzene
Methylene chloride (Dichloromethane)
o-Dichlorobenzene
p-Dichloro benzene
Styrene
Tetrachloroethylene
Toluene
trans- 1 ,2-Dichloroethylene
Trichloroethene (Trichtoroethylene)
Vinyl chloride
Xylenes (Total)
o-Xylene
M-Xylene
P-Xylene
Reauired Unreaulated Contaminants
(m-) 1 ,3-Dichlorobenzene
1 .1 ,1 ,2-Tetrachloroethane
1 ,2.3-TrichloroprDDane
1 ,3-Dichloropropane
Chloroethane
Chloromethane
Dibromomethane
o-Chlorotoluene
p-Chlorotoiuene
Ground Water Systems
No. of
Analyses
44,492
40,568
40.293
55,913
39,985
39,915
38,602
56,122
38,718
39,453
57,052
49.313
56,726
55,033
56.725
39,647
47,684
63,710
56.879
56,299
45,407
37.337
26,985
38.692
[ 33.628
27.364
58,780
49,387
49,810
49,269
57,469
58,214
48,437
44,052
42.912
No. of
Systems
11,730
10,579
14,697
18,625
14,592
12,349
14,640
18,286
14,403
14.622
18.714
15,434
18,825
18.053
18,673
14,668
15,327
18,808
18,870
18.653
14.695
14.479
8,913
13,076
10.484
9.219
19,127
15,702
16.139
15.702
18,892
18,964
15.323
14,516
14,416
>MRL
N
299
114
662
59
204
57
236
125
267
235
56
276
365
689
42
171
97
782
725
130
621
73
279
301
247
155
39
22
47
21
70
285
40
27
26
%
2.55%
1 .08%
>0.5 MCL
N
191
94
4.50% 13
0.32%! 12
1.40%! 66
0.46%
1.61%
0.68%
1 .85%
1.61%
0.30%
1 .79%
1.94%
3.82%
0.22%
1.17%
0.63%
4.16%
3.84%
0.70%
4.23%
0.50%
3.13%
2.30%
2.36%
1.68%
0.20%
0.14%
0.29%
0.13%
0.37%
1.50%
0.26%
0.19%
0.18%
0
65
34
99
71
1
17
2
220
0
2
1
353
5
3
270
38
0
%
1.63%
0.89%
0.09%
0.06%
0.45%
0.00%
0.44%
0.19%
0.69%
0.49%
0.01%
0.11%
0.01%
1.22%
0.00%
0.01%
0.01%
1.88%
0.03%
0.02%
1.84%
0.26%
0.00%
>MCL*
N
161J
73
6
5
37
0
37
16
63
40
1
8
0
125
0
1
0
229^
1
1
190
28
0
%
1.37%
0.69%
0.04%
0.03%
0.25%
0.00%
0.25%
0.09%
0.44%
0.27%
0.01%
0.05%
0.00%
0.69%
0.00%
0.01%
0.00%
1.22%
0.01%
0.01%
1.29%
0.19%
0.00%
N/A
N/A
N/A
0
0
2
0.00%
0.00%
0.01%
0
0
1
0.00%
0.00%
0.01%
N/A
N/A
144
0.76%
93
0.49%
N/A
0
0
0.00%
0.00%
0
0
0.00%
0.00%
A-22
-------
A Review of Coniaminam Occurrence in Public Water Systems
Table A-7.b. Summary of Contaminant Occurrence in Ground Water Systems from the
Unregulated Contaminant Monitoring Information System (URCIS) (Continued).
Contaminant
Discretionary Compounds
1 ,2,3-Trichlorobenzene
1 .3.5-Trimethvlbenzene
Bromochloromethane
Dichlorodifluoromethane
Ground Water Systems
No. of
Analyses
36,383
36,324
38,250
45,692
Isoproovlbenzene 37,526
n-Butvlbenzene 37,413
n-Propylbenzene 37,412
p-lsopropyttoluene
sec-Butylbenzene
tert-Butvlbenzene
Trichlorofluorometfiane
Other Not-Reaulated Contaminants
cis-1 ,3-Dichloropropene
Total 1,2-Dtchloroethene
trans- 1 ,3-Dichloroprooene
Other Regulated Contaminants
Bromodichloromethane
Bromoform
Chlorodibromometriane
Chloroform
Total THMs
. . ,
*ontaminant candidate List
1.1 ,2,2-Tetrachloroethane
1.1-DJchloroetriane
1 ,1 -Dichloropropene
1 ,2,4-Trimethvlbenzene
2,2-Dichloropropane
Bromobenzene
Bromomethane
Hexachtorobutadiene
Naphthalene
Total 1,3-DtetiloroproDene
36,196
36,436
36,442
47,691
22,845
24,227
49.848
49,719
49,776
52,083
58,339
59,044
37.206
48,534
48,930
58.536
37,286
38,837
26,235
No. of
Systems
11,920
11.736
1 1 ,925
14.994
11.840
11,827
11.791
11.308
11.428
11.435
15,725
8.707
9.280
18.443
18,262
18,337
18,771
19.070
19.202
11.820
15,497
15.225
18.838
11.334
12.336
8.404
>MRL
N
54
86
55
223
41
37
43
32
29
24
243
45
15
2,931
1.484
2.431
4.483
80
278
110
25
29
140
36
146
14
%
0.45%
0.73%
0.46%
1.49%
>0.,
N
>MCL
%
>MCL*
N
%
N/A
N/A
6
0
0.05%
0.00%
3| 0.03%
oi 0.00%
0.35% N/A
0.31%
0.36%
0.28%
0.25%
0.21%
1.55%
0.52%
0.16%
15.89%
8.13%
13.26%
23.88%
N/A
N/A
N/A
N/A
N/A
N/A
N/A
t
N/A
N/A
N/A
26
0.14%
4
0.02%
N/A
I
0.42%
1.45%
0.93%
0.16%
0.19%
0.74%1
0.32%
1.18%
0.17%J
N/A
N/A
N/A
N/A
N/A
N/A
26
13
12
0.14%
0.11%
0.10%
15
6
4
0.08%
0.05%
0.03%
N/A
N/A = There is no MCUHAL for this contaminant.
• % > MCL indicates the proportion of systems with any analytical results exceeding the concentration value of the MCL; it does not necessarily
indicate an MCL violation. An MCL violation occurs when the MCL is exceeded by the average results from four quarterly samples or confirmation
saffiotes a&jeouired bv the orimacv State.
A-23
-------
APPENDIX B
USGS Studies Reviewed for Contaminant Occurrence Information.
-------
-------
USGS STUDIES REVIEWED FOR CONTAMINANT OCCURRENCE INFORMATION
Adamski, J. U.S. Department of the Interior. Nitrate and Pesticides in Ground Water of the
Ozark Plateaus in Arkansas, Kansas, Missouri, and Oklahoma. Fact Sheet # 182-96,
U.S. Geological Survey. 1997.
Adamski, J. U.S. Department of the Interior. Nutrients and Pesticides in Ground Water of the
Ozark Plateaus in Arkansas, Kansas, Missouri, and Oklahoma. Water Resources
Investigations Report # 96-4313, U.S. Geological Survey. 1997.
Anderolm, S.K., MJ. Radell, and S.F. Richey. U.S. Department of the Interior. Water Quality
Assessment of the Rio Grande Valley Study Unit, Colorado, New Mexico and Texas--
Analysis of Selected Nutrient, Suspended-Sediment, and Pesticide Data. Water
Resources Investigations Report # 94-4061, U.S. Geological Survey. 1995.
Ator, S.W., J.D. Blomquist, J.W. Brakebill, J.M. Denis, MJ. Ferrari, C.V. Miller, and H. Zappia.
U.S. Department of the Interior. Water Quality in the Potomac River Basin; Maryland,
Pennsylvania, Virginia, West Virginia, and the District of Columbia. Circular 1166, US
Geological Survey. 1998.
Ator, S.W. and MJ. Ferrari. U.S. Department of the Interior. Nitrate and Selected Pesticides in
Ground Water of the Mid-Atlantic Region. Water Resources Investigations Report 97-
4139, United States Geological Survey. 1997.
Barbash, I.E. U.S. Department of the Interior. Pesticides in Ground Water, Current
Understanding of Distribution and Major Influences. Fact Sheet 244-95, US Geological
Survey. 1996.
Barbash, J.E., and E.A. Resek. Pesticides in Ground Water, volume two of the series Pesticides
in the Hvdrologic System. Ann Arbor Press, Inc., Chelsea, Michigan. 1996.
Battaglin, W., and L. Hay. "Effects of Sampling Strategies on Estimates of Annual Mean
Herbicide Concentrations in Midwestern Rivers." Environmental Science & Technology.
30:889-896. 1996.
Battaglin, W., and D. Goolsby. "Regression Models of Herbicide Concentrations in Outflow
from Reserviors in the Midwestern USA, 1992-1993." Journal of the American Water
Works Association. 1369:34-6. 1998.
Berndt, M.P., H.H. Hatzell, C.A. Crandall, M. Turtora, J.R. Pittman, and E.T. Oaksford. U.S.
Department of the Interior. Water Quality in the Georgia-Florida Coastal Plain, Georgia
and Florida, 1992-1996. Circular 1151, U.S. Geological Survey. 1998.
B-l
-------
A Review of Contaminant Occurrence in Public Water Systems
Bevans, H.E., M.S. Lico, and S.J. Lawrence. U.S. Department of the Interior. Water Quality in
the Las Vegas Valley Area and the Carson and Truckee River Basins; Nevada and
California. Circular 1170, US Geological Survey. 1998.
Brown, M.F. U.S. Department of the Interior. Water-Quality Assessment of the Trinity River
Basin, Texas • Pesticides in a Coastal Prairie Agricultural Area. Open File Report # 96-
124, U.S. Geological Survey. February 1996.
Brown, M.F. U.S. Department of the Interior. Water-Quality Assessment of the Trinity River
Basin, Texas - Pesticides in a Suburban Watershed, Arlington, 1993-94. Fact Sheet #
159-95, U.S. Geological Survey. June 1995.
Burkart, M.R. and D.W. Kolpin. "Hydrologic and Land-Use Factors Associated with Herbicides
and Nitrate in Near-Surf ace Aquifers." Journal of Environmental Quality. 22(4): 646-
656. 1993.
Clark, G.M. U.S. Department of the Interior. Assessment of Nutrients, Suspended Sediment, and
Pesticides in Surface Water of the Upper Snake River Basin, Idaho and Western
Wyoming, Water Years 1991-95. Water Resources Investigations Report # 97-4020, U.S.
Geological Survey. 1997.
Clark, G.M. U.S. Department of the Interior. Occurrence and Flux of Selected Pesticides in
Surface Water of the Upper Snake River Basin, Idaho and Western Wyoming. Water
Resources Investigations Report # 97-4020, U.S. Geological Survey. 1997.
Clark, G.M., T.R. Maret, M.G. Rupert, M.A. Maupin, W.H. Low, and D.S. Ott. U.S.
Department of the Interior. Water Quality in the Upper Snake River Basin, Idaho and
Wyoming, 1992-95. Circular # 1160, U.S. Geological Survey. 1998.
Coupe, R.H., D.A. Goolsby, J.L. Iverson, D.J. Markovchik, and S.D. Zaugg. U.S. Department of
the Interior. Pesticide, Nutrient, Water-Discharge, and Physical-Property Data for the
Mississippi River and Some of Its Tributaries, April 1991-September 1992. Open File
Report 93-657, U.S. Geological Survey. 1995.
Crepeau, K.L., K.M. Kuivila, and J.L. Domagalski. U.S. Department of the Interior.
"Concentrations of Dissolved Rice Pesticides in the Colusa Basin Drain and Sacramento
River, CA, 1990-1992." In USGS Toxic Substances Hydrology Program-Proceedings of
the Technical Meeting, Colorado Springs, CO, September 20-24, 1993, edited by D.W.
Morganwalp and D.A. Aronson, p. 711-718. Water Resources Investigation Report 94-
4015, U.S. Geological Survey. 1996.
Dubrovsky, N. M., C.R. Kratzer, L.R. Brown, J.M. Gronberg, and K. R. Burow. U.S.
Department of the Interior. Water Quality in the San Joaquin-Tulare basins; California,
Circular 1159, US Geological Survey. 1998.
Fenelon, J.M. U.S. Department of the Interior. Water Quality in the White River Basin: Indiana.
Circular 1150, US Geological Survey. 1998.
B-2
-------
A Review of Contaminant Occurrence in Public Water Systems
Fisher, G.T. U.S. Department of the Interior. Selected Herbicides in Major Streams in the
Potomac River Basin Upstream from Washington. NAWQA Fact Sheet 107-95, U.S.
Geological Survey. 1995.
Frick, E.A., D.J. Hippe, G.R. Buell, C.A. Couch, E.H. Hopkins, DJ. Wangsness, and J.W.
Garrett. U.S. Department of the Interior. Water Quality in the Apalachicola-
Chattahoochee-Flint River Basin. Circular 1164, US Geological Survey. 1998.
Gilliom, R.J., W.M. Alley, and M.E. Gurtz. U.S. Department of the Interior. Design of the
National Water-Quality Assessment Program: Occurrence and Distribution of Water-
Quality Conditions. Circular 1112, US Geological Survey. 1995.
Goolsby, D.A. and W.A. Battaglin. "Occurrence and Distribution of Pesticides in Rivers of the
Midwestern United States." Chapter 16 in M.L. Leng, E.M.K. Leovey, and P.L. Zubkoff,
eds., Agrochemical Environmental Fate; State of the An. American Chemical Society.
Washington, DC. April, 1995.
Goolsby, D.A., R.C. Coupe, and DJ. Markovchick. U.S. Department of the Interior.
Distribution of Selected Herbicides and Nitrate in the Mississippi River and its Major
Tributaries, April through June 1991. Water Investigations Report 91-4163, U.S.
Geological Survey. 1991.
Goolsby, D.A. and W.E. Pereira. U.S. Department of the Interior. "Pesticides in the Mississippi
River." In R.H. Meade, ed., Contaminants in the Mississippi River. Circular 1133, U.S.
Geological Survey. 1995.
Graffy, E.A., D.R. Helsel, and D.K. Mueller. U.S. Department of the Interior. Nutrients in the
Nations Waters: Identifying Problems and Progress. Fact Sheet 218-96, US Geological
Survey. 1996.
Hippe, DJ. U.S. Department of the Interior. "Pesticide Occurrence in the Upper Floridan
Aquifer in the Dougherty Plain and Marianna Lowlands Districts, Southwestern Georgia
and Adjacent Areas of Alabama and Florida." In Proceedings of the 1997 Georgia Water
Resources Conference, Athens, GA, Institute of Ecology, The University of Georgia,
March 20-22,1997, edited by KJ. Hatcher, p. 49-50. U.S. Geological Survey. 1997.
Hippe, DJ. and J.W. Garrett. U.S. Department of the Interior. "The Spatial Distribution of
Dissolved Pesticides in Surface Water of the Apalachicola-Chattahoochee-Flint River
Basin in Relation to Land Use and Pesticide Runoff-Potential Ratings, May 1994." In
Proceedings of the 1997 Georgia Water Resources Conference, Athens, GA, Institute of
Ecology, The University of Georgia, March 20-22, 1997, edited by KJ. Hatcher, p. 4-13.
U.S. Geological Survey. 1997.
Kilroy, K.C. and S.A. Watkins. U.S. Department of the Interior. Pesticides in Surface Water,
Bottom Sediment, Crayfish, and Shallow Ground Water in Las Vegas Valley Area,
Carson River Basin, and Truckee River Basin, Nevada and California, 1992-1995. Fact
Sheet 75-97, U.S. Geological Survey. June 1997.
B-3
-------
A Review of Contaminant Occurrence in Public Water Systems
Kalkhoff, S.J., D.W. Kolpin, E.M. Thurman, I. Ferrer, and D. Barcelo. "Degradation of
Chloroacetanilide Herbicides: The Prevalence of Sulfonic and Oxanilic Acid Metabolites
in Iowa Groundwaters and Surface Waters." Environmental Science & Technology.
32(11): 1738-1740. 1998.
Kolpin, D.W., D. Sneck-Fahrer, G.R. Hallberg, and R.D. Libra. "Temporal Trends of Selected
Agricultural Chemicals in Iowa's Groundwater, 1982-95: Are things getting better?"
Journal of Environmental Quality. 26: 1007-1017. 1997.
Kolpin, D.W., S. J. Kalkhoff, D.A. Goolsby, D.A. Sneck-Fahrer, and E.M. Thurman.
"Occurrence of Selected Herbicides and Herbicide Degradation Products in Iowa's
Ground Water, 1995." Ground Water. 35-4: 679-687. 1997.
Kolpin, D.W., E.M. Thurman, and D.A. Goolsby. "Occurrence of Selected Pesticides and Their
Metabolites in Near-Surface Aquifers of the Midwestern United States." Environmental
Science & Technology. 30(1): 335-340. 1996.
Kolpin, D.W. "Agricultural Chemicals in Ground Water of the Midwestern U.S.: Relations to
Land Use." Journal of Environmental Quality. 26(4). 1997.
Kolpin, D.W., J.E. Barbash, and R. J. Gilliom. "Occurrence of Pesticides in Shallow
Groundwater of the United States: Initial Results from the National Water-Quality
Assessment Program." Environmental Science & Technology. 32(5): 558-566. 1998.
Kolpin, D.W., M.R. Burkart, and E.M. Thurman. U.S. Department of the Interior. Herbicides
and Nitrate in Near-Surface Aquifers in the Midcontinental United States 1991. Water
Supply Paper 2413, US Geological Survey. 1994.
Kolpin, D.W., G.R. Hallberg, D.A. Sneck-Fahrer, and R.D. Libra. U.S. Department of the
Interior. Agricultural Chemicals in Iowa's Ground Water, 1982-95—What are the
Trends? Fact Sheet 116-97, US Geological Survey. 1997.
Kolpin, D.W., PJ. Squillance, J.S. Zogorski, and I.E. Barbash. "Pesticides and Volatile Organic
Compounds in Shallow Urban Groundwater of the United States." In Groundwater in the
Urban Environment: Problems. Processes, and Management. 469-474. Rotterdam: A.A.
Balkema Publishers. 1997.
Kolpin, D.W., E.M. Thurman, and D.A. Goolsby. "Occurrence of Selected Pesticides and Their
Metabolites in Near-Surface Aquifers of the Midwestern United States." Environmental
Science & Technology. 30(1): 335-340. 1996.
Koterba, M.T., W.S.L. Banks, and R.J. Shedlock. "Pesticides in Shallow Ground Water in the
Delmarva Peninsula." Journal of Environmental Quality. 22: 500-518. 1993
Land, L.F. U.S. Department of the Interior. Water-Quality Assessment of the Trinity River
Basin, Texas - Nutrients and Pesticides in the Watersheds ofRichland and Chambers
-------
A Review of Contaminant Occurrence in Public Water Systems
Creeks, 1993-95. Water Resources Investigations Report # 97-4132, U.S. Geological
Survey. 1997.
Land, L.F. U.S. Department of the Interior. Water-Quality Assessment of the Trinity River
Basin, Texas - Pesticides in Urban and Agricultural Streams, 1993-95. Fact Sheet # 178-
96, U.S. Geological Survey. July 1996.
Land, L.F. and M.F. Brown. U.S. Department of the Interior. Water-Quality Assessment of the
Trinity River Basin, Texas - Pesticides in Streams Draining an Urban and an
Agricultural Area, 1993-95. Water Resources Investigations Report # 96-4114, U.S.
Geological Survey. 1995.
Land, L.F. U.S. Department of the Interior. Water-Quality in the Trinity River Basin, Texas -
1992-95. Circular #1171, U.S. Geological Survey. 1998.
Larson, S.J., R.J. Gilliom, and P.D. Capel. U.S. Department of the Interior. Pesticides in
Streams of the United States — Initial Results from the National Water Quality
Assessment Program. U.S. Geological Survey. Water-Resources Investigations Report
98-4222, 92 p. 1999.
Larson, S.J., P.D. Capel, and M.S. Majewski. Pesticides in Surface Waters, volume three of the
series Pesticides in the Hvdrologic System. Ann Arbor Press, Inc., Chelsea, Michigan.
1997.
Larson, S.J., P.D. Capel, and M.S. Majewski. "Analysis of Key Topics — Sources, Behavior, and
Transport." Chap. 5 in Pesticides in Surface Waters, volume three of the series Pesticides
in the Hvdrologic System. Ann Arbor Press, Inc., Chelsea, Michigan. 1997.
Leahy, P.P. and T.H. Thompson. U.S. Department of the Interior. The National Water-Quality
Assessment Program. Open-File Report 94-70, US Geological Survey. 1994.
Levings, G.W., D.F. Healey, S.F. Richey, and L.F. Carter. U.S. Department of the Interior.
Water Quality in the Rio Grande Valley, Colorado, New Mexico, and Texas, 1992-95.
Circular # 1162, U.S. Geological Survey. 1998.
Lindsey, B.D., K.J. Breen, M.D. Bilger, and R.A. Brightbill. U.S. Department of the Interior.
Water Quality in the Lower Susquehanna River Basin; Pennsylvania and Maryland.
Circular 1168, US Geological Survey. 1998.
Lopes, T.J. and D.A. Bender. U.S; Department of the Interior. "Nonpoint Sources of Volatile
Organic Compounds in Urban Areas - Relative Importance of Land Surfaces and Air."
Environmental Pollution. 1998. In Press.
Lopes, T.J., E.T. Furlong, and J.W. Pritt. "Occurrence and Distribution of Semivolatile Organic
Compounds in Stream Bed Sediments, United States, 1992-95." Environmental
Toxicology and Risk Assessment. Volume 7. American Society for Testing and
Materials. 1998.
B-5
-------
A Review of Contaminant Occurrence in Public Water Systems
Majewski, M.S., and P.D. Capel. Pesticides in the Atmosphere, volume one of the series
Pesticides in the Hvdrologic System. Ann Arbor Press, Inc., Chelsea, Michigan. 1995.
Nolan, B.T. and B.C. Ruddy. U.S. Department of the Interior. Nitrate in Ground Waters of the
United States-Assessing Risk. Fact Sheet FS-092-96, US Geological Survey. 1996.
Oaksford, E.T. U.S. Department of the Interior. National Water-Quality Assessment Program-
Preliminary Results: Agricultural Chemicals in the Suwannee River Basin, Georgia-
Florida Coastal Plain Study Unit. Open-File Report 94-103, U.S. Geological Survey.
1994.
Ohio Environmental Protection Agency. Pesticide Special Study. Division of Drinking and
Ground Waters. Columbus, 1998. Available on internet at
http://www.epa.state.oh.us/ddagw/pestspst.html.
Petersen, J.C., J.C. Adamski, J.V. Davis, S.R. Femmer, D.A. Freiwald, and R.L. Joseph. U.S.
Department of the Interior. Water Quality of the Ozark Plateaus; Arkansas, Kansas,
Missouri, and Oklahoma. Circular 1158, US Geological Survey. 1998.
Phillips, P.J., and D.W. Hanchar. U.S. Department of the Interior. Water-Quality Assessment of
the Hudson River Basin in New York and Adjacent States: Analysis of Available Nutrient,
Pesticide, Volatile Organic Compound, and Suspended Sediment Data, 1970-1990.
Water-Resources Investigations Report 96-4065, U.S. Geological Survey. 1996.
Puckett, L. J. U.S. Department of the Interior. Nonpoint and Point Sources of Nitrogen in Major
Watersheds of the United States. Water Resources Investigation Report 94-4001 and Fact
Sheet, US Geological Survey. 1996.
Reutter, D.C. U.S. Department of the Interior. Water-Quality Assessment of the Trinity River
Basin, Texas - Well and Water-Quality Data From the Outcrop of the Woodbine Aquifer
in Urban Tarrant County, 1993. Open File Report # 96-413, U.S. Geological Survey.
1996.
Rupert, M.G. U.S. Department of the Interior. Analysis of Data on Nutrients and Organic
Compounds in Groundwater in the Upper Snake River Basin, Idaho and Western
Wyoming, 1980-91. Water Resources Investigations Report # 94^135, U.S. Geological
Survey. 1994.
Rupert, M.G. U.S. Department of the Interior. Major Sources of Nitrogen Input and Loss in the
Upper Snake River Basin, Idaho and Western Wyoming, 1990. Water Resources
Investigations Report # 96-4008, U.S. Geological Survey. 1996.
Rupert, M.G. U.S. Department of the Interior. Nitrate in Ground Water of the Upper Snake
River Basin, Idaho and Western Wyoming, 1991-95. Water Resources Investigations
Report # 97^174, U.S. Geological Survey. 1997.
B-6
-------
A Review of Contaminant Occurrence in Public Water Systems
Ryker, S.J. and A.K. Williamson. U.S. Department of the Interior. Pesticides in Public Supply
Wells of the Central Columbia Plateau. Fact Sheet 205-96, U.S. Geological Survey.
October 1996.
Ryker, S.J. and A.K. Williamson. U.S. Department of the Interior. Pesticides in Public Supply
Wells of Washington State. Fact Sheet 122-96, U.S. Geological Survey. May 1996.
Schaap, B.D., and S.M. Linhart. U.S. Department of the Interior. Quality of Ground Water Used
for Selected Municipal Water Supplies in Iowa, 1982-96 Water Years. Open-File Report
98-3, U.S. Geological Survey. May 1998.
Scribner, E.A., D.A. Goolsby, E.M. Thurman, M.T. Meyer, and W.A. Battaglin. U.S.
Department of the Interior. Concentrations of Selected Herbicides, Herbicide
Metabolites, and Nutrients in Outflow from Selected Midwestern Reservoirs, April 1992
Through September 1993. Open File Report 96-363, U.S. Geological Survey. 1996.
Scribner, E.A., D.A. Goolsby, E. M. Thurman, M.T. Meyer, and M.L. Pomes. U.S. Department
of the Interior. Concentrations of Selected Herbicides, Two Triazine Metabolites, and
Nutrients in Storm Runoff from Nine Stream Basins in the Midwestern United States,
1990-1992. Open File Report 94-396, U.S. Geological Survey. 1994.
Shipp, A. U.S. Department of the Interior. Water-Quality Assessment of the Trinity River Basin,
Texas - Pesticide Occurrence in Streams, Winter and Spring -1994. Fact Sheet #160-95,
U.S. Geological Survey. June 1995.
Spruil, T.B., D.A. Harned, P.M. Ruhl, J.L. Eimers, G. McMahon, K.E. Smith, D.R. Galeone, and
M.D. Woodside. U.S. Department of the Interior. Water Quality in the Albemarle-
Pamlico Drainage Basin; North Carolina and Virginia. Circular 1157, US Geological
Survey. 1998.
Squillance, P.J., D.A. Pope, and C.V. Price. U.S. Department of the Interior. Occurrence of the
Gasoline Additive MTBE in Shallow Ground Water in Urban and Agricultural Areas.
Fact Sheet 114-95, U.S. Geological Survey. 1996.
Thurman, E.M., D.A. Goolsby, M.T. Meyer, and D.W. Kolpin. "Herbicides in Surface Waters of
the Midwestern United States: The Effect of Spring Flush." Environmental Science &
Technology. 25: 1794-1796. 1991.
U.S. Department of the Interior. National Synthesis: Pesticides in Ground Water. Fact Sheet FS
244-95, US Geological Survey. 1995.
U.S. Department of the Interior. Nutrients in the Nation's Waters—Too Much of a Good Thing?
Circular 1136, U.S. Geological Survey. 1996.
U.S. Department of the Interior. Pesticides in Surface and Ground Water of the United States:
Preliminary Results of the National Water Quality Assessment Program (NAWQA).
Pesticide National Synthesis Project, U.S. Geological Survey. 1998.
B-7
-------
A Review of Contaminant Occurrence in Public Water Systems
U.S. Department of the Interior. Pesticides in Surface Waters, Current Understanding of
Distribution and Major Influences. Fact Sheet 039-97, US Geological Survey. 1997.
U.S. Department of the Interior. VOCs in Ground Water of the United States: Preliminary
Results of the National Water Quality Assessment Program (NAWQA). VOC National
Synthesis Project, U.S. Geological Survey. 1998.
Vowinkel, E.F., R.M. Clawges, D.E. Buxton, D.A. Stedfast, and J.B. Louis. U.S. Department of
the Interior. Vulnerability of Public Drinking Water Supplies in New Jersey to Pesticides.
Fact Sheet 165-96, U.S. Geological Survey. 1996.
Wagner, R.J. J.C. Ebbert, and L.M. Roberts. U.S. Department of the Interior. Are Agricultural
Pesticides in Surface Waters of the Central Columbia Plateau? Fact Sheet # 241-95,
U.S. Geological Survey. 1995.
Wagner, R.J., J.C. Ebbert, L.M. Roberts, and S.J. Ryker. U.S. Department of the Interior.
Agricultural Pesticide Applications and Observed Concentrations in Surface Waters from
Four Drainage Basins in the Central Columbia Plateau, Washington, and Idaho. Water-
Resources Investigations Report 95-4285, U.S. Geological Survey. 1996.
Wall, G.R. and P.J. Phillips. U.S. Department of the Interior. Pesticides in Surface Waters of
the Hudson River Basin, New York and Adjacent States. Fact Sheet 238-96, US
Geological Survey. April 1997.
Wall, G.R., K. Riva-Murray, and P.J. Phillips. U.S. Department of the Interior. Water Quality in
the Hudson River Basin; New York and Adjacent States. Circular 1165, US Geological
Survey. 1998.
Washington Department of Health. Results of the Areawide Groundwater Monitoring Project.
Division of Drinking Water. Spring 1995.
Williamson, A.K., M.D. Munn, S.J. Ryker, R.J. Wagner, J.C. Ebbert, and A.M. Vanderpool.
U.S. Department of the Interior. Water Quality of the Columbia River Plateau,
Washington and Idaho. Circular 1144, US Geological Survey. 1998.
Zogorski, J. U.S. Department of the Interior. VOCs in Ground Water of the U. S.: Preliminary
Results of the NAWQA Program. Fact Sheet, US Geological Survey. 1998.
B-8
-------
APPENDIX C
General Data Quality Issues
-------
-------
GENERAL DATA QUALITY ISSUES
To add perspective to the types of data quality issues encountered while working with the
occurrence data, some of the most common concerns are outlined below. While this is far from a
complete list of all the quality issues encountered, it should provide some idea of the various
complexities and complications involved in preparing the data.
Every State database contained unique data elements or a unique treatment of common elements.
Even after initial screening and conversion, factors were always uncovered during data analysis.
These were resolved in consultation with the State or data source. As a general rale, when errors
or ambiguities in various data elements could not be resolved, the observation was eliminated
from the analysis to avoid aberrant results. These observations made up only a very small
portion of the large number of records included in most data sets.
Structure of the Data
Most of the data sets are "vertically" designed, meaning there is one column for the system
identification number, one column for the date sampled, one column for the name of the
contaminant analyzed, one column for the results, etc. Other data sets are "horizontally"
designed, meaning there is one column for the system identification number and multiple
columns for the contaminants analyzed (each contaminant has a separate column). The results
for each contaminant are thus displayed along a single data row under the appropriate
contaminant heading. A horizontal data structure is far more difficult to analyze in the manner
necessary for this study and requires extensive processing to transform into a more suitable
vertical structure.
File Format
Files which were submitted in CSV (comma separated values) format were problematic if proper
care was not taken to exclude or modify data fields (usually text fields) which might contain
commas within them (for example, a field holding the address of a water system might contain
one or more commas). This field would often separate into two or more columns as the
computer read the file. Unless every observation contained the same number of text commas
within each field, the following data fields would no longer line up by column between
observations. Making the necessary adjustments to allow the computer to properly recognize and
process the correct data fields is a tedious and time intensive task, even with the use of
specialized programs.
Multiple Data Sets
The size and number of data sets required to make up a complete occurrence database for a State
can vary considerably. While some States maintain one database for all water system monitoring
results, it is not unusual for a State to keep a number of databases for various data subsets (e.g., a
separate database for each of the contaminant groups - lOCs, SOCs, and VOCs). Other States
have very elaborate subsets consisting of multiple (10 or more) separate data sets for all of the
compliance data for the State. Sometimes each subset will have a unique format and structure
and require significant formatting before it is compatible with the rest of the State data. Many
C-l
-------
A Review of Contaminant Occurrence in Public Water Systems
States have an individual data manager for each of the contaminant groups and coordinating data
transfer requires communication with more than one contact person.
Contaminant Codes
Analytes were identified by a variety of codes including EPA codes, CAS numbers, STORET
codes (which might have multiple codes for a given contaminant), State-specific or laboratory-
specific codes, or by chemical name. In every case where a contaminant was identified by any
system other than EPA codes, the proper EPA contaminant designation was found and added to
the record.
Some States have special coding systems for contaminants that are covered under the same
analytical method. One of the most common systems summarizes the results of a single method
with an 'ND' or zero for all contaminants not detected and individual observations only for those
contaminants with a positive result. This system is best illustrated using the 21 VOCs covered
under method 502.2 as an example. If none of the 21 VOCs were detected, a State might enter
'21 VOCs' in the contaminant column and '0' in the results column of a single observation. If
one or more of the 21 VOCs were detected, these individual contaminants would be entered in
the contaminant column and the value they were detected at would be entered in the results
column of individual observations. It is assumed that every contaminant not recorded
individually was tested for and not detected. Although the example above may seem
straightforward, the rules can become complicated. It is not unusual to find caveats in the
database which need to be further defined by the State before analysis can begin. This is done
differently between States, and it is often not entirely consistent within a State data set.
Reporting non-Detections
There are numerous ways that States report non-detections in their data systems. Some States
report a value of zero when a contaminant is not detected, others have a column in which they
enter a less-than sign and then enter the method detection limit or reporting level in the results
column. A third common method for reporting no detects is the inclusion of a separate text
variable that will read *ND'. A result value of ND would be converted to zero for the purposes
of this study, while less-than values were kept as they are to provide a sense of the various
MRL/MDLs.
Units
States differ in the units they use to enter contaminant concentration results and, at times, the
reporting units change within a State. The most common unit for reporting results is in
milligrams per liter, but at times only the SOCs are reported in milligrams per liter and the rest of
the database will be in micrograms per liter. In some cases, units are not identified in a data set.
Other Data
Most of the databases coming from the States are not designed for data analysis and are generally
"electronic filing cabinets." The databases contain special sampling data, raw water data,
compliance data, and other sampling data. To analyze only the compliance data, it is necessary
to understand the coding system the State uses to distinguish the compliance sampling from the
other types of sampling. Most coding systems are unique to each State.
C-2
-------
APPENDIX D
Summary Tables V.A.I through V.A.8.
See text for details.
-------
-------
A Review of Contaminant Occurrence in Public Water Systems
Table V.A.1. Summary of the percentage of systems with detections (>MRL) and exceedances
(>MCL*) of selected contaminants, comparing ground-water supplied, community-water systems
and non-transient, non-cornmunity water systems in Michigan.
Contaminant
SOCs
2,4-D
Atrazine
Benzo[a]pyrene
Carbofuran (Furadan)
Ethylene Dibromide (EDB)
Glyphosate (Roundup)
Simazine
VOCs
Benzene
Styrene
Tetrachloroethylene
Vinyl Chloride
XyJenes (Total)
CWSs
Percent > MRL
0.0%
0.3%
0.0%
0.0%
0.0%
0.0%
0.3%
0.9%
8.1%
1.8%
0.2%
1.8%
Percent > MCL*
0.0%
0.1%
0.0%
0.0%
0.0%
0.0%
0.0%
0.2%
0.0%
0.2%
0.0%
0.1%
NTNCWSs
Percent > MRL
0.0%
0.4%
0.0%
0.0%
0.0%
0.1%
0.9%
1.5%
0.6%
0.0%
1.6%
Percent > MCL*
0.0%
0.0%
0.0%
0.0%
0.0%
0.0%
0.3%
0.0%
0.1%
0.0%
0.0%
* % > MCL indicates the proportion of systems with any analytical results exceeding the concentration value of the MCL: it does not
necessarily indicate an MCL violation. An MCL violation occurs when the MCL is exceeded by the average results from four quarterly samples
or confirmation samples as required by the primacy State.
D-l
-------
A Review of Contaminant Occurrence in Public Water Systems
Table V.A.2. Summary of the percentage of systems with detections (>MRL) of selected
contaminants, comparing ground-water and surface-water supplied systems, by size of system
(population served categories), for a high-occurrence state.
Contaminant
Population Served
<500
500-3^00
3,301-10,000
10,001-50,000
> 50,000
Surface Water (% > MRL)
lOCs
Barium
Cadmium
Mercury
SOCs
2,4-D
Atrazine
Benzo[a]pyrene
Carbofuran (Furadan)
Ethylene Dibromide (EDB)
Glyphosate (Roundup)
VOCs
Benzene
Styrene
Tetrachloroethylene
Vinyl Chloride
Xytenes (Total)
100.0%
20.0%
0.0%
44.4%
100.0%
0.0%
0.0%
0.0%
0.0%
0.0%
0.0%
0.0%
0.0%
22.2%
100.0%
0.0%
3.2%
65.8%
100.0%
0.0%
5.3%
2.6%
0.0%
5.1%
2.6%
2.6%
2.6%
17.9%
100.0%
0.0%
0.0%
53.8%
96.2%
0.0%
0.0%
20.0%
0.0%
33.3% j
26.7%
30.0%
23.3%
43.3%
100.0%
0.0%
0.0%
29.6%
96.3%
0.0%
0.0%
10.7%
0.0%
17.9%
17.9%
25.0%
17.9%
32.1%
100.0%
0.0%
0.0%
44.4%
55.6%
11.1%
0.0%
9.1%
0.0%
27.3%
27.3%
18.2%
27.3%
27.3%
Ground Water (% > MRL)
IOCS
Barium
Cadmium
Mercury
SOCs
2,4-D
Atrazine
Benzo[a]pyrene
Carbofuran (Furadan)
Ethylene Dibromide (EDB)
Glyphosate (Roundup)
VOCs
Benzene
Styrene
Tetrachloroethylene
Vinyl Chloride
Xylenes (Total)
98.2%
0.0%
0.0%
1.3%
2.5%
0.6%
0.2%
0.6%
0.0%
1.6%
3.3%
2.2%
1.6%
11.1%
98.6%
4.2%
1.9%
5.3%
11.3%
0.0%
0.0%
0-9%
0.0%
4.1%
4.4%
5.1%
2.1%
18.4%
95.2%
0.0%
0.0%
6.0%
12.0%
0.0%
0.0%
3.8%
0.0%
13.3%
10.5%
16.2%
11.4%
19.4%
90.9%
0.0%
0.0%
6.4%
21.3%
0.0%
0.0%
5.8%
0.0%
15.4%
3.8%
25.0%
7.7%
32.7%
100.0%
0.0%
0.0%
33.3%
33.3%
33.3%
0.0%
33.3%
0.0%
66.7%
66.7%
100.0%
66.7%
66.7%
D-2
-------
A Review of Contaminant Occurrence in Public Water Systems
Table V.A.3. Summary of the percentage of systems with exceedances (>MCL*) of selected
contaminants, comparing ground-water and surface-water supplied systems, by size of system
(population served categories), for a high-occurrence state.
Contaminant
Population Served
<500
500-3,300
3,301-10,000
10,001-50,000
> 50,000
Surface Water (% > MCL*)
IOCS
Barium
Cadmium
Mercury
SOCs
2,4-D
Atrazine
Benzo[a]pyrene
Carbofuran (Furadan)
Ethylene Dibromide (EDB)
Glyphosate (Roundup)
VOCs
Benzene
Styrene
Tetrachloroethylene
Vinyl Chloride
Xyienes (Total)
0.0%
0.0%
0.0%
0.0%
77.8%
0.0%
0.0%
0.0%
0.0%
0.0%
0.0%
0.0%
0.0%
0.0%
Ground Water (% > MCL*)
lOCs
Barium
Cadmium
Mercury
SOCs
2,4-D
Atrazine
Benzo(a]pyrene
Carbofuran (Furadan)
Ethylene Dibromide (EDB)
Glyphosate (Roundup)
VOCs
Benzene
Styrene
Tetrachloroethylene
Vinyl Chloride
Xyienes (Total)
0.0%
0.0%
0.0%
0.0%
0.0%
0.0%
0.0%
0.0%
0.0%
0.1%
0.0%
0.4%
0.0%
0.0%
0.0%
0.0%
0.0%
0.0%
71.1%
0.0%
0.0%
0.0%
0.0%
2.6%
0.0%
0.0%
0.0%
0.0%
0.0%
0.0%
0.0%
0.0%
57.7%
0.0%
0.0%
20.0%
0.0%
0.0%
0.0%
0.0%
0.0%
0.0%
0.0%
0.0%
0.0%
0.0%
18.5%
0.0%
0.0%
10.7%
0.0%
0.0%
0.0%
0.0%
0.0%
0.0%
2.8%
1.4%
1.9%
0.0%
1.1%
0.0%
0.0% ]
0.9%
0.0%
0.5%
0.0%
0.7%
0.2%
0.0%
4.8%
0.0%
0.0%
0.0%
0.0%
0.0%
0.0%
2.9%
0.0%
1.0%
0.0%
0.0%
1.9%
0.0%
18.2%
0.0%
0.0%
0.0%
0.0%
0.0%
0.0%
1.9%
0.0%
0.0%
0.0%
3.8%
0.0%
0.0%
25.0%
0.0%
0.0%
0.0%
22.2%
0.0%
0.0%
9.1%
0.0%
0.0%
0.0%
0.0%
0.0%
0.0%
0.0%
0.0%
0.0%
0.0%
33.3%
33.3%
0.0%
33.3%
0.0%
0.0%
0.0%
33.3%
0.0%
0.0%
* %> MCL indicates the proportion of systems with any analytical results exceeding the concentration value of the MCL; it does not necessarily
indicate an MCL violation. An MCL violation occurs when the MCL is exceeded by the average results from four quarterly samples or
confirmation samples as required by the primacy Slate.
D-3
-------
A Review of Contaminant Occurrence in Public Water Systems
Table V.A.4. Summary of the percentage of systems with detections (>MRL) of selected
contaminants, comparing ground-water and surface-water supplied systems, by size of system
(population served categories), for a low-occurrence state.
Contaminant
Population Served
<500
500-3,300
3,301-10,000
10,001-50,000
> 50,000
Surface Water (% > MRL)
lOCs
Barium
Cadmium
Mercury
SOCs
2,4-D
Atrazine
Benzo[a]pyrene
Carbofuran (Furadan)
Ethylene Dibromide (EDB)
Glyphosate (Roundup)
VOCs
Benzene
Styrene
Tetrachloroethylene
Vinyl Chloride
Xylenes (Total)
29.6%
5.6%
11.3%
0.0%
0.0%
0.0%
0.0%
0.0%
0.0%
0.0%
0.0%
4.5%
0.0%
2.3%
Ground Water (% > MRL)
IOCS
Barium
Cadmium
Mercury
SOCs
2,4-D
Atrazine
Benzo[a]pyrene
Carbofuran (Furadan)
Ethylene Dibromide (EDB)
Glyphosate (Roundup)
VOCs
Benzene
Styrene
Tetrachloroethylene
Vinyl Chloride
Xyienes (Total)
29.3%
5.4%
6.7%
0.3%
0.0%
0.0%
0.0%
0.9%
0.0%
0.2%
0.0%
6.5%
0.0%
1.5%
33.3%
1.6%
17.5%
3.3%
0.0%
0.0%
0.0%
0.0%
0.0%
1.6%
0.0%
3.5%
1.6%
3.5%
45.8%
6.7%
9.2%
0.0%
0.0%
0.0%
0.0%
0.0%
0.0%
0.8%
0.0%
10.3%
0.0%
3.9%
66.7%
16.7%
16.7%
0.0%
0.0%
0.0%
0.0%
0.0%
0.0%
0.0%
5.6%
0.0%
5.6%
0.0%
36.4%
4.5%
9.1%
0.0%
0.0%
0.0%
0.0%
0.0%
0.0%
4.5%
0.0%
0.0%
0.0%
0.0%
36.4%
4.5%
9.1%
0.0%
0.0%
0.0%
4.8%
4.8%
0.0%
0.0%
0.0%
0.0%
0.0%
0.0%
64.7%
29.4%
23.5%
0.0%
0.0%
0.0%
0.0%
6.3%
0-0%
5.6%
0.0%
0.0%
0.0%
0.0%
50.0%
0.0%
25.0%
0.0%
0.0%
0.0%
25.0%
0.0%
0.0%
0.0%
0.0%
0.0%
0.0%
0.0%
33.3%
0.0%
33.3%
0.0%
0.0%
0.0%
0.0%
0.0%
0.0%
0.0%
0.0%
33.3%
0.0%
0.0%
D-4
-------
A Review of Contaminant Occurrence in Public Water Systems
Table V.A.5. Summary of the percentage of systems with exceedances (>MCL*) of selected
contaminants, comparing ground-water and surface-water supplied systems, by size of system
(population served categories), for a low-occurrence state.
Contaminant
Population Served
50,000
Surface Water (% > MCL*)
lOCs
Barium
Cadmium
Mercury
SOCs
2,4-D
Atrazine
Benzo[a]pyrene
Carbofuran (Furadan)
Ethylene Dibromide (EDB)
Glyphosate (Roundup)
VOCs
Benzene
Styrene
Tetrachloroethylene
Vinyl Chloride
Xylenes (Total)
Ground Water (% > MCL*)
lOCs
Barium
Cadmium
Mercury
SOCs
2,4-D
Atrazine
Benzo[a]pyrene
Carbofuran (Furadan)
Ethylene Dibromide (EDB)
Giyphosate (Roundup)
VOCs
Benzene
Styrene
Tetrachloroethylene
Vinyl Chloride
Xylenes (Total)
0.0%
0.0%
0.0%
0.0%
0.0%
0.0%
0.0%
0.0%
0.0%
0.0%
0.0%
2.3%
0.0%
0.0%
0.1%
0.3%
0.1%
0.0%
0.0%
0.0%
0.0%
0.3%
0.0%
0.1%
0.0%
1.8%
0.0%
0.0%
1.6%
0.0%
3.2%
0.0%
0.0%
0.0%
0.0%
0.0%
0.0%
0.0%
0.0%
0.0%
0.0%
0.0%
0.0%
1.7%
0.0%
0.0%
0.0%
0.0%
0.0%
0.0%
0.0%
0.0%
0.0%
0.0%
0.0%
0.0%
0.0%
0.0%
0.0%
0.0%
0.0%
0.0%
0.0%
0.0%
0.0%
0.0%
0.0%
0.0%
0.0%
0.0%
0.0%
0.0%
0.0%
0.0%
0.0%
0.0%
0.0%
0.0%
0.0%
0.0%
0.0%
0.0%
0.0%
0.0%
0.0%
0.0%
0.0%
0.0%
0.0%
0.0%
0.0%
0.0%
0.0%
0.0%
0.0%
0.0%
0.0%
0.0%
0.0%
0.0%
0.0%
0.0%
0.0%
0.0%
0.0%
0.0%
0.0%
0.0%
0.0%
0.0%
0.0%
0.0%
0.0%
0.0%
0.0%
0.0%
0.0%
0.0%
0.0%
0.0%
0.0%
0.0%
0.0%
0.0%
0.0%
0.0%
0.0%
0.0%
0.0%
0.0%
0.0%
0.0%
0.0%
0.0%
0.0%
0.0%
0.0%
0.0%
0.0%
0.0%
* % > MCL indicates the proportion of systems with any analytical results exceeding the concentration value of the MCL; it does not necessarily
indicate an MCL violation. An MCL violation occurs when the MCL is exceeded by the average results from four quarterly samples or
confirmation samples as required by the primacy State.
D-5
-------
A Review of Contaminant Occurrence in Public Water Systems
Table V.A.6. Summary of the percentage of systems with detections (>MRL) of selected
contaminants, comparing ground-water and surface-water supplied systems, by size of system
(population served categories), from the national URCIS database.
Contaminant
Population Served
<500
500-3,300
3301-10,000
10,001-50,000
> 50,000
Surface Water (% > MRL)
SOCs
Dibromochloropropane
(DBCP)
Ethylene Dibromide (EDB)
VOCs
Methylene Chloride
Tetrachloroethylene
Toluene
Trichloroethylene
Vinyl Chloride
Xylenes (Total)
1.5%
3.9%
11.1%
1.1%
7.7%
0.5%
1.4%
15.8%
Ground Water (% > MRL)
SOCs
Dibromochloropropane
(DBCP)
Ethylene Dibromide (EDB)
VOCs
Methylene Chloride
Tetrachloroethylene
Toluene
Trichloroethylene
Vinyl Chloride
Xylenes (Total)
1.5%
0.3%
2.7%
2.2%
2.9%
2.1%
0.2%
1.8%
1.7%
0.3%
10.3%
2.0%
6-2%
4.1%
0.6%
7.0%
2.5%
1.1%
3.4%
3.8%
3.6%
4.0%
0.5%
5.0%
1.1%
1.3%
14.5%
1.3%
7.3%
7.2%
1.2%
9.8%
4.8%
3.4%
8.0%
9.6%
8.2%
10.4%
0.8%
7.9%
3.3%
3.4%
14.6%
5.5%
8.2%
12.6%
2.1%
11.8%
12.2%
6.8%
14.0%
23.8%
10.3%
23.3%
2.9%
6.9%
10.7%
10.1%
17.1%
13.8%
13.1%
22.1%
3.1%
17.2%
21.9%
13.3%
24.5%
42.8%
17.3%
39.4%
5.5%
10.2%
D-6
-------
A Review of Contaminant Occurrence in Public Waier Systems
Table V.A.7. Summary of the percentage of systems with exceedances (>MCL*) of selected
contaminants, comparing ground-water and surface-water supplied systems, by size of system
(population served categories), from the national URCIS database.
Contaminant
Population Served
<500
500-3,300
3301-10,000
10,001-50,000
> 50,000
Surface Water (% > MCL*)
SOCs
Dibromochloropropane
(DBCP)
Ethylene Dibromide (EDB)
VOCs
Methylene Chloride
Tetrachloroethylene
Toluene
Trichloroethylene
Vinyl Chloride
Xylenes (Total)
2.5%
0.5%
4.3%
0.0%
0.0%
0.0%
0.0%
0.0%
0.0%
0.9%
2.9%
0.4%
0.0%
0.3%
0.3%
0.0%
0.7%
0.7%
4.1%
0.8%
0.0%
0.4%
0.8%
0.0%
0.3%
1.6%
3.9%
1.2%
0.0%
2.0%
1.3%
0.0%
3.4%
5.8%
4.8%
3.7%
0.0%
8.7%
2.1%
0.0%
Ground Water (% > MCL*)
SOCs
Dibromochloropropane
(DBCP)
Ethylene Dibromide (EDB)
VOCs
Methylene Chloride
Tetrachloroethylene
Toluene
Trichloroethylene
Vinyl Chloride
Xylenes (Total)
0.9%
0.2%
0.4%
0.4%
0.0%
0.4%
0.1%
0.0%
1.2%
0.5%
0.7%
1.0%
0.0%
1.0%
0.1%
0.0%
2.6%
2.1%
1.7%
2.9%
0.0%
3.5%
0.1%
0.0%
6.2%
5.6%
2.8%
9.2%
0.0%
8.4%
1.4%
0.0%
8.8%
6.3%
5.4%
23.3%
0.0%
22.4%
3.7%
0.0%
* % > MCL indicates the proportion of systems with any analytical results exceeding the concentration value of the MCL; it does not necessarily
indicate an MCL violation. An MCL violation occurs when the MCL is exceeded by the average results from four quarterly samples or
confirmation samples as required by the primacy State.
D-7
-------
A Review of Contaminant Occurrence in Public Water Systems
Table V.A.8. Summary of the percentage of systems with detections (>MRL) of selected
unregulated contaminants, comparing ground-water and surface-water supplied systems, by size
of system (population served categories), from the national URCIS database.
Contaminant
Population Served
<500
500-3300
3301-10,000
10,001-50,000
> 50,000
Surface Water (% > MRL)
VOCs
(m-) 1,3-Dichlorobenzene
1,1,1 ,2-Tetrachloroethane
1 ,2,3-Trichloropropane
Bromomelhane
Chloromelhane
o-Chlorotoluene
p-ChlorotoIuene
Ground Water (% > MRL)
VOCs
(m-> 1,3-Dichlorobenzene
1,1,1 ,2-Tetrachloroethane
1 ,2,3-Trichloropropane
Bromomethane
Chloromethane
o-Chlorotoluene
p-Chlorotoluene
1.7%
0.4%
0.4%
2.0%
1.7%
0.7%
0.8%
0.2%
0.2%
0.2%
0.2%
2.6%
0.4%
0.0%
0.3%
1.3%
0.3%
0.5%
2.5%
0.5%
0.3%
0-8%
0.8%
0.3%
2.1%
2-4%
1.1%
0.8%
0.2%
0.0%
0.2%
0.7%
1.3%
0.1%
0.1%
0.1%
0.1%
0.1%
0.6%
1.6%
0.1%
0.1%
0.3%
0.2%
0.7%
0.7%
3.2%
0.5%
0.5%
0.7%
1.5%
1.9%
1.8%
3.3%
0.4%
0.9%
2.1%
1.3%
0.7%
3.7%
5.0%
0.0%
0.0%
2.2%
4.0%
4.4%
3.3%
4.9%
4.9%
3.0%
D-8
------- |