United States
Environmental Protection
Agency
Office of Water
Office of Pesticides and
Toxic Substances
Fall 1990
National Pesticide Survey
Project Summary
,.... National
Survey of its
Kind
Summary of
Phase INPS
Results
What Was
EPA Looking
For?
The U.S. Environmental Protection Agency (EPA) has completed its five-year
National Survey of Pesticides in Drinking Water Wells (NPS). This fact sheet provides
an overview of the Survey and a summary of Survey findings.
A joint project of EPA's Office of Drinking Water (ODW) and Office of Pesticide
Programs (OPP), the Survey is the first national study of pesticides, pesticide
degradates, and nitrate in drinking water wells. The Survey has two principal
objectives: (1) to determine the frequency and concentration of the presence of
pesticides and nitrate in drinking water wells nationally; and (2) to improve EPA's
understanding of how the presence of pesticides and nitrate in drinking water wells is
associated with patterns of pesticide use and the vulnerability of ground water to
contamination. The findings will help EPA set priorities, prepare guidance, and
develop and implement regulatory programs.
Development of the Survey's statistical design started in 1984. In 1987, EPA
conducted a pilot study in California, Mississippi, and Minnesota to test Survey
implementation and analytical procedures. Sampling for the full Survey began in April
1988 and the final sample was collected in February 1990. More than 1300 wells were
sampled, some in every State. The total cost of the Survey was approximately $12
million.
EPA released a Phase I Report in Fall 1990. This report covers Survey design,
implementation, analytical chemistry, quality assurance, and summary results. In
Spring 1991, EPA will release a Phase II Report containing results and findings from
analyses of the relationships among chemical detections, well construction, ground-
water vulnerability, pesticide use, and other factors.
The NPS provides EPA's first national estimates of the frequencies and
concentrations of pesticides and nitrate in community water system (CWS) wells and
rural domestic drinking water wells.
These Survey results indicate that the proportion of wells nationwide found to
"contain any particular pesticide or pesticide degradate is low. Considering only the
proportion of wells containing pesticides over the EPA levels of health concern, Survey
results do not demonstrate any immediate widespread health problem. Survey resuns
also show, however, that substantial numbers of wells, particularly rural domestic
wells, could be affected by the presence of one or more pesticides. In addition.
substantial numbers of wells are affected both by the presence of nitrate and by
nitrate over EPA levels of health concern. EPA believes that these results indicate trvar
there is need for continued attention and additional analysis of the issue.
The Survey analyzed a statistically representative sample of wells to provk* a
national assessment of the presence of pesticides and nitrate in drinking water w«t».
The Survey does not assess the presence of pesticides or nitrate at the local, county
or State level, nor does it assess the presence of pesticides or nitrate in ground or
surface water generally. The Survey focused on the quality of water in drinking w««r
NPS Project Summary
-------
wells before treatment rather than the quality of drinking water at the tap. If a water
treatment system was connected to the well, EPA sampled prior to the treatment
system.
Water samples collected by EPA were analyzed for 101 pesticides, 25 pesticide
degradates, and nitrate (for a total of 127 analytes). Because the NPS was designed
to assess the presence of a wide range of pesticides in drinking water wells, multi-
residue methods of chemical analysis were used to provide a broad-range scanning of
chemicals in collected samples. The analytical method selected by EPA for nitrate
analyzed the combined presence of nitrate and nitrite in collected samples measured
as nitrogen (N), which is reported as a single concentration of nitrate. Exhibit 1 lists
the NPS analytes and analytic methods.
What Did EPA EPA estimates that there are approximately 38,300 community water systems
with operating wells nationally, which together comprise approximately 94,600 CWS
wells, and that there are about 10,500,000 rural domestic wells in the United States.
Initial Survey results include several findings. EPA estimates that:
About 9,850 (10.4%) of the nation's CWS wells and about 446,000
(4.2%) of the nation's rural domestic wells contain at least one
pesticide above Survey minimum reporting limits. The most
commonly found chemicals, other than nitrate, are the acid
metabolites of the pesticide OCPA, and atrazine.
• The levels of pesticides and pesticide degradates found in wells
were usually lower than levels of drinking water health concern.
Over half of all CWS wells and rural domestic wells nationally
contain nitrate above the NPS minimum reporting limit (0.15 mg/L).
About 1,130 (1.2%) CWS wells and 254,000 (2.4%) rural domestic
wells contain concentrations above EPA's Maximum Contaminant
Level (MCL) for nitrate of 10 mg/L, the maximum permissible level
of a contaminant in water that is delivered to any user of a public
water system.
• The maximum concentrations of nitrate detected were
approximately 13 mg/L for CWS wells and 120 mg/L for rural
domestic wells. (There may be other wells not sampled by the
Survey that contain nitrate at concentrations higher than the levels
detected in the sampled wells.)
• The median concentrations of nitrate detected, above the Survey's
minimum reporting limit, were approximately 1.6 mg/L for both CWS
wells and rural domestic wells.
Selected Survey findings for wells with at least one pesticide detected are
summarized in Exhibit 2. Pesticide concentrations are reported in units of micrograms
per liter (pg/L), the equivalent of parts per billion (ppb). Exhibit 2 also presents
selected Survey results for wells containing nitrate above the minimum reporting limit.
Nitrate concentrations are reported in units of milligrams of nitrogen per liter (mg/L),
which is equivalent to parts per million (ppm).
The degree of precision of the Survey is reflected in the confidence intervals
presented in Exhibits 2 and 3. The confidence intervals represent ranges, specified by
an upper and lower bound, that EPA is reasonably confident contain the national
estimates. For example, EPA estimates that about 9,850 (10.4%) CWS wells
NPS Pro/«ct Summary
-------
Exhibit 1
Analytes Detectable By Method
NPS METHOD 1: Gas Chromatography with a Nitrogen-Phosphorous Detector
Alachlor Diphenamid Methyl paraoxon
Ameiryn Oisulfoton* .. Metolachlor
Atraton Disulfoton sulfbne* Metribuzin
Atrazine Oisulfoton sutfoxide* Mevmphos
Bromacil EPIC Molinate
Butachlor Ethoprop Napropamide
Butylate Fenamiphos Norflurazon
Carboxin Fenarimol Pebulate
Chlorpropham Fluridone Prometon
Cycloate Hexazinone Prometryn
Diazinon" MGK 264 Pronamide*
Dichlorvos Merphos* Propazine
NPS METHOD 2: Gas Chromatography with an Electron Capture Detector
4,4-OOD Oieldrin Heptachlor epoxide
4,4-ODE Endosulfan I Hexachlorobenzene
4,4-ODT Endosulfan II . Methoxychlor
Aldrin Endosulfan sulfate Propachlor
Chlorobenzilate* Endrin Trifluralin
Chloroneb Endrin Aldehyde alpha - HCH
Chlorothalonil Etridiazole beta • HCH
DCPA Heptachlor delta - HCH*
NPS METHOD 3: Gas Chromatography with an Electron Capture Detector
2,4-0 4-Nitrophenol* . Oalapon*
2,4-DB Acifluorfen* Dicamba
2,4,5-TP Bentazon Dicamba, 5-hydroxy-
2,4,5-T Chloramben* Dichlorprop
3,5-Dichlorobenzoic acid DCPA acid metabolites Dinoseb
NPS METHOD 4: High Performance Liquid Chromatography with an Ultraviolet Detector
Atrazine, deethylated Diuron Metribuzin OA
Barban Fenamiphos sulfone Metribuzin DADK*
Carbofuran, phenol-3-keto- Fenamiphos sulfoxide Metribuzin OK*
Carbofuran, phenol Fluometuron Neburon
Cyanazine LJnuron Pronamide metabolite
NPS METHOD 5: Direct Aqueous Injection HPLC with Post-Column Derivatizatlon
Aldicarb Baygon Carbofuran, 3-hydroxy-
Aldicarb sulfone Carbaryl Methiocarb
Aldicarb sulfoxide Carbofuran Methomyl
NPS METHOD 6: Qw Chromatography with a Nitrogen-Phosphorous Detector
Ethylene thiourea (ETU)
NPS METHOD 7: Microextraction and Gas Chromatography
Ethylene dibromide (EDB) 1 ,2 - dichloropropane** trans -1,3-
Dibromochloropropane (DBCP) cis - 1,3 - dichloropropene** dichloropropene**
NPS METHOD 9: Automated Cadmium Reduction and Colorimetrlc Detection
Nitrate and nitrite measured as nitrogen (N)
* Qualitative only.
** Method 8 dropped. Analytes previously included in Method 8 also detectable by Method 7.
(46 Analytes)
Simazine
Simeiryn
Stirofos
Tebuthiuron
Terbacil
Terbufos*
Terbutryn
Triademefon
Tricyclazole
Vernolate
(29 Analytes)
gamma - HCH
alpha-Chlordane
gamma-Chlordane
cis - Permethrin
trans - Permethrin
(17 Analytes)
Pentachlorophenol (PCP)
Picloram
(18 Analytes)
Propanil
Propham
Swep
(10 Analytes)
Oxamyl
(1 Analyte)
(5 Analytes)
(1 Analyte)
NPS Project Summary
-------
Exhibit 2
National Estimates for Number and Percent of Wells
For Pesticides and Nitrate
PESTICIDES
CWS walls nationally with at least
one pesticide
CWS wells above HAL**
CWS wells above MCL***
Rural domestic wells nationally with
at least one pesticide
Rural domestic wells above HAL**
Rural domestic wells above
MCL***
NITRATE
CWS wells nationally
CWS wells above MCL***
Rural domestic wells nationally
Rural domestic wells above
MCL***
Estimated
Number
9,850
0
0
446,000
19,400
60.900
49,300
1.130
5,990.000
254,000
95% Confidence
Interval
(Lower • Upper)
(6,330-13,400)
(0 - 750)
(0 • 750)
(246,000 - 647.000)
(170- 131,000)
'• (9,430 - 199,000)
(45,300 - 53,300)
(370 • 2.600)
(5.280,000 - 6,700,000)
(122.000 -464,000)
Estimated
Percent
10.4
0
0
4.2
0.2
0.6
52.1
1.2
57.0
2.4
95% Confidence
Interval
(Lower - Upper)*
(6.8-14.1)
(0 - 0.8)
(0 - 0.8)
(2.3 - 6.2)
(<0.1 • 1.2)
(0.1 - 1.9)
(48.0 - 56.3)
(0.4 - 2.7)
(50.3 - 63.8)
(1.2-2.4)
Numbers between zero and 0.05 are reported as less than 0.1 (<0.1).
Health Advisory Level (HAL) is the concentration of a contaminant in water that may be consumed over a person's
lifetime without harmful effects. HALs are non-enforceable health-based guideline* that consider only non-cancer toxic
effects. Only pesticides with HALs were included in estimating the number of wells containing pesticides above the
HALs.
Maximum Contaminant Level (MCL) is the maximum permissible level of a contaminant in water that is delivered to any
user of a public water system. MCLs are enforceable standards. Only pesticides with MCLs were included in
estimating the number of wells containing pesticides above the MCLs. Although the MCL is not legally applicable to
rural domestic wells, it was used as a standard of quality for drinking water.
NPS Pro/«ct Summary
-------
Exhibit 3
National Estimates for Number and Percent of Wells Containing
Detectable Amounts of Pesticides and Pesticide Degradates Analyzed
by the National Pesticide Survey*
Community Water System
Well*
OCPA acid metabolite*
Atrazine
Simazine
Prometon
Hexachlorobenzene""
Dibromochloropropene
(DBCP)""
Oinoseb""
Rural Domestic W«U«
OCPA acid metabolites
Atrazine
Dibromochloropropane
(DBCP)""
Prometon
Simazine
Ethylene dibromide (ED8)""
Gamma-HCH (Undane)
Ethylene thiourea (ETU)
Bentazon
Alachlor
99% Confidence
Estimated Interval
Number (Lower • Upper)
6,010 (3.170 • 8,840)
1,570 (420-2,710)
1,080 (350- £540)
520 (78-1,710)
470 (61 - 1 .830)
370 (33-1.480)
25 (1 • 870)
264,000 (129,000-477,000)
70,800 (13,300-214,000)
38,400 (2.740 -164,000)
25,600 (640-142,000)
25,100 (590-141,000)
19,200 (160-131,000)
13.100 (14 - 120,000)
8.470 (1-111.000)
7.160 (1 • 109,000)
3.140 (1 - 101,000)
95% Confidence
Estimated Interval
Percent (Lower - Upper)**
6.4 (3.4 - 9.3)
1.7 (0.5-2.9)
1.1 (0.4-2.7}
0.5 (0.1-1.8)
0.5 (0.1-1.7)
0.4 (<0.1 - 1 .6)
<0.1 (<0.1 • 0.9)
2.5 (1.2-4.5)
0.7 (0.1 - 2.0)
0.4 (<0.1-1.6)
0.2 -. (<0.1-1.4)
0.2 (<0.1-1.3)
0.2 (<0.1-1.2)
0.1 (<0.1-1.1)
0.1 (<0.1-1.1)
0.1 (<0.1 - 1.0)
<0.1 (<0.1 - 1.0)
NPS
RcpOftinQ
Umlt***
0.10M9A
0.12 jig/L
0.38 jig/L
0.15
-------
nationally contain detectable levels of at least one pesticide. Considering the
precision of the Survey, this number could be as low as 6,330 (6.8%), or as high as
13,400 (14.1%) as indicated by the 95% confidence intervals.
Exhibit 3 provides national estimates for the number of CWS and rural domestic
wells containing individual detected pesticides and pesticide degradates, along with
the associated 95% confidence intervals. Of all the pesticides and pesticide
degradates detected in the Survey, OCPA acid metabolites and atrazine were the most
commonly found. For OCPA acid metabolites, which are degradates of OCPA, the
maximum concentrations- detected were approximately 7.2 /jg/L for CWS wells and 2.4
pg/L for rural domestic wells. The median concentrations for DCPA acid metabolites
from the wells sampled in the Survey were about 0.34 /jg/L for CWS wells and 0.38
A/g/L for rural domestic wells. All OCPA acid metabolite detections were at
concentration levels of 8 pgJL or less, a small fraction (0.2%) of the HAL limit of 4,000
jjgJL The maximum concentrations of atrazine in Survey drinking water samples were
about 0.92 A/g/L for CWS wells and 7.0 ng/L for rural domestic wells. The median
concentrations of detected atrazine, from the wells sampled in the Survey were about
0.26 pg/L for CWS wells and 0.29 ^g/L for rural domestic wells. The proposed MCL
for atrazine is 3 pg/L A total of five pesticides, alachlor, atrazine,
dibromochloropropane (08CP), ethylene dibromide (EDB), and gamma-HCH (lindane)
were detected in rural domestic wells at levels above their respective maximum
contaminant levels (MCLs). MCLs are enforceable standards established by EPA for
public water systems to protect human health. Although the MCL is not legally
applicable to rural domestic wells, it was used as a standard of quality for drinking
water. EPA notified well owners within 24 hours when detections were above health-
based guidelines or standards. None of the detections of pesticides or pesticide
degradates for CWS wells were above the MCL or HAL The NFS Survey Analytes
Fact Sheet contains a list of all Survey analytes with their Minimum Reporting Limits,
MCLs, and HALs.
How Did EPA The Survey was designed to yield results that are statistically representative of
Select Wells? a" cws wells and rural domestic wells in the United States. EPA used statistical
survey methods to select a representative group of CWS wells and rural domestic
wells for sampling.
First, EPA characterized all counties in the U.S. according to pesticide use and
relative ground-water vulnerability - two critical factors affecting the presence of
pesticides in drinking water wells. EPA concentrated on agricultural pesticide use,
specified as high, moderate, low, or uncommon pesticide use. EPA characterized
ground-water vulnerability by using a numerical classification system called DRASTIC,
which considers depth of water, recharge, aquifer media, soil media, topography,
impact of vadose zone, and hydraulic conductivity of the aquifer.
To identify CWS wells, EPA randomly selected 7,083 community water systems
from a list containing information on all public water supply systems. For the CWS
well survey, EPA was interested in obtaining very accurate estimates of pesticide
occurrence in counties that are more vulnerable to ground-water contamination. This
was achieved by slightly over-representing wells from these counties in the set of
selected CWS wells. EPA conducted telephone interviews with representatives of the
7,083 selected systems to determine their operating status, confirm the number of
wells, and obtain cooperation for sampling. Based on the results of the screening
process, EPA collected water samples from 566 wells, 540 of which were used in data
analysis based on well samples that passed quality assurance requirements.
When selecting rural domestic wells, EPA randomly chose 90 counties as areas
for sampling to represent the nation's wide range of agricultural pesticide use and
ground-water vulnerability as measured by DRASTIC. EPA used the DRASTIC
6 NPS Project Summary
-------
New Testing
Methods
Developed
Sampling
Across the
Country
In the Lab
classification system to score ground-water vulnerability within counties and subcounty
areas. EPA collected information from county agricultural extension agents on
cropping intensity to further subdivide the counties into areas that are more vulnerable
or less vulnerable to the presence of pesticides in ground water. For the rural
domestic well survey, EPA was interested in obtaining very accurate estimates of
pesticide occurence in locations with high pesticide use and high vulnerability to
ground-water contamination. This was achieved by slightly over-representing wells
from these locations in the set of selected rural domestic wells. A total of 864 eligible
wells were selected for sampling. Of these selected wells, 783 wells were sampled
based on homeowner participation; 752 of which were used in data analysts based on
well samples that passed quality assurance requirements.
Once the wells were selected, EPA developed a sampling schedule to visit each
well once. EPA scheduled sample collection so that well visits were spread out across
the 22 month sampling period. This schedule provided well water samples during all
seasons and pesticide application cycles. This approach was used to minimize the
effect of seasonal variability. Eligible CWS wells were defined as wells in systems of
piped drinking water that either have at least 15 connections or serve 25 permanent
residents. To be eligible, a system must have had at least one operating well (at the
time of sampling) that was being used to supply drinking water.
Each sample taken in the National Pesticide Survey was tested for 127 analytes
(see Exhibit 1). Through extensive literature searches, consultation with scientific
experts, and assessments of methods in the laboratory, EPA evaluated existing
laboratory methods for testing the analytes. Because of the wide variety of
procedures used in these methods and the large number of analytes to be included in
the Survey, EPA needed methods .that could efficiently test for several analytes.
Ultimately, EPA used two existing analytical methods and developed six new methods
specifically for the Survey. Five of these six are multi-residue methods, each capable
of detecting ten or more pesticides.
From April 1988 to February 1990, EPA collected water samples and well
information from over 1300 community water system wells and rural domestic wells.
EPA sampled each well once, collecting a minimum of 17 bottles of well water. State
agencies across the country provided the sampling teams to collect samples from
community water system wells. EPA's contractors conducted the domestic well
sampling. CWS wells were sampled in every State and domestic wells were sampled
in 38 states.
At each well sampled, questionnaires were used to collect data necessary for
the interpretation of NPS results. Data included:
• observations about the well sampled and the surrounding area;
• information from the owner/operator about well construction and
agricultural and non-agricultural pesticide use on the property
where the well was located; and
• information from local area experts (such as a county agricultural
extension agent) about crops, pesticide use, and land use within
one-half mile of the well.
The chemical analyses of NPS water samples were performed at five contract
laboratories and three EPA laboratories. The contract laboratories were responsible
for chemical analyses of water samples collected in the field by one or more of the
established methods. Two EPA laboratories were responsible for managing contract
laboratories, confirming detections of pesticides in samples, and ensuring that quality
NPS Project Summery
-------
Putting
Quality First
Communica-
tions Network
Summary of
Planned
Phase II
Analyses
control standards were maintained; the third EPA laboratory performed chemical
analyses.
EPA made an extraordinary commitment to quality early in the Survey. A quality
assurance (QA) program covering every major component of the Survey ensured that
the Survey produced high quality, statistically valid data useful to both scientists and
policy makers. EPA developed QA Project Plans for well selection, data collection and
analysis, well sampling, interviewing, and laboratory analysis. During the Survey, EPA
conducted numerous audits of field and laboratory activities to ensure that procedures
approved as part of the QA program were followed. EPA conducted performance
evaluation studies of the analytical laboratories to monitor laboratory capability.
Communication between EPA, Survey participants, and interested parties was
key to the successful completion of the Survey.
In EPA's ten Regions, an NPS contact was identified to enlist the cooperation of
State water supply and pesticide agencies and to answer questions from the media
elected officials, and organizations interested in the Survey. At the local level, county
agricultural extension agents and health officials contributed to Survey planning and
implementation and provided a two-way flow of information between EPA and
participating local communities.
EPA maintained communications with interested parties through a variety of
outreach techniques, including briefings for the media, Congress, and governors'
representatives, presentations at national and regional association meetings, articles in
technical and scientific journals, and a continuing dialogue with key representatives in
the agricultural, environmental, and industrial communities. Periodic 'Project Updates'
were distributed to over 3,000 interested individuals and organizations.
To assist in notifying Survey participants of sampling results, EPA prepared one-
page Hearth Advisory Summaries to explain the potential health effects of exposure to
pesticides in non-technical terms. These Summaries are based on longer, more
technical scientific documents called Health Advisories. EPA sent appropriate Hearth
Advisory Summaries to well owners and operators along with Survey sampling results.
EPA plans a number of statistical analyses using the data compiled from NPS
questionnaires and other sources. These analyses will study the association of
pesticides and nitrate in drinking water wells with such factors as fertilizer and
pesticide use on the property where the well is located, use within one-half mile of tne
well, and use within the county. Studies will also address the association of pesticides
and nitrate in wells with ground-water vulnerability characteristics including depth to
groundwater, recharge, aquifer media, soil media, topography, impact of unsaturaieo
zone, and hydraulic conductivity. The Phase II Report will analyze the questionnaire
databases, the first-stage and second-stage DRASTIC stratification scoring results
nitrogen fertilizer and pesticide sales, and the Survey analytical results, to investigate
the potential causes and consequences of pesticide residues in drinking water wens
These analyses will include a study of whether there is an association among
questionnaire responses. The hypotheses that might be tested include the following
Are there more analyte detections of chemicals associated with unconfined aquifers
than with confined aquifers? Are nitrate detections associated with septic units on :**
property? Are there more analyte detections of chemicals associated with areas
where irrigation is used? EPA is planning scientific investigations to:
NPS Profit
-------
Determine trie association between detections and weU characteristics;
Where to Go
for More
Information
• Correlate DRASTIC subscores by individual factors (e.g., depth to ground
water) with rural domestic well detections;
• Determine associations between pesticide use and pesticide detections in
water samples;
• Prepare tables showing pesticide detections by county pesticide use
estimates for specific analytes that were detected in the Survey;
• Prepare summary statistics such as frequencies of analytes detected in
areas where crops with known associated pesticide use are grown;
• Conduct regression modeling for analytes with sufficient detections.
These analyses will seek to identify significant relationships of variables
that are not readily discernible from the analyses of well characteristics
presented in the Phase I Report;
• Evaluate analytes with few detections by non-statistical review of all
available data such as marginal comments on questionnaires; and
• Compare NPS findings with other studies.
This fact sheet is part of a series of NPS outreach materials, fact sheets and
reports. The following additional NPS fact sheets are available through EPA's Public
Information Center (401 M Street SW, Washington DC 20460, 202-382-2080):
Survey Design
Survey Anafytea
Fact Sheet for each
detected anatyte
Analytical Method*
How EPA Will Use
The NPS ftesufts
Summary Results
Gfos*ary
Quality Assurance/
Quality Control
Additional information on the Survey and on pesticides in general can be
obtained from the following sources:
U.S. EPA Safe Drinking Water Hotline
1-800-426-4791 (In Washington, DC - 382-5533)
Monday-Friday, 8:30 am to 4:30 pm Eastern Time
National Pesticide Telecommunications Network
1-800-858-7378
24 hours a day
U.S. EPA Office of Pesticide Programs (OPP) Docket
401 M Street, SW Room NEG004
Washington, DC 20460
(202) 382-3587
Information on regulation of
pesticides in drinking
water
Information on health
effects and safe
handling of pesticides
Background documents
for Survey (available
for review)
NPS Project Summary
-------
National Technical Information Service (NT1S) Copies of the
5285 Port Royal Road NFS Phase I Report
Springfield, VA 22161 (available 1991)
(703) 487-4650 and
NPS Phase II Report
(when available)
If you are concerned about the presence of pesticides and nitrate in your private water
well, contact your local ornate hearth department. Other experts in your State
environmental agency or agriculture and health department may also be helpful to
you If you receive your drinking water from a community water system and have
questions about your water quality, contact your local community water system
owner/operator or the State water supply agency.
NPS Pro/«ct Summary.
to
------- |