United States
Environmental Protection
Agency
Office of Water (WH-550)
Office of Pesticides and
Toxic Substances (H-7501C)
EPA 570/9-91 -021
Winter 1992
SEPA National Pesticide Survey
Update and Summary of Phase II
Results
First National
Survey of its
Kind
Summary of
Phase II
Analyses
No Single
Factor
Predominates
Everyone
Must Help
Address The
Problem
The U.S. Environmental Protection Agency (EPA) has completed the Phase II
Report of Its National Survey of Pesticides in Drinking Water Wells (NPS). This fact
sheet provides an update on Survey activities and a summary of the Phase II results.
The Survey is a joint project of EPA's Office of Ground Water and Drinking
Water (OGWDW) and Office of Pesticide Programs (OPP). It is the first national study
of pesticides, pesticide degradates, and nitrate in drinking water wells. EPA designed
the Survey with two principal objectives. In Phase I, EPA developed national estimates
of the frequency and concentration of the presence of pesticides and nitrate in
drinking water wells. In Phase II, EPA carried out statistical analyses of the NPS data
and pertinent data from non-NPS databases 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 sensitivity of ground water to contamination. After
peer review of the report by the Scientific Advisory Panel established under the
Federal Insecticide, Fungicide, and Rodenticide Act (FIFRA), EPA will use the findings
to help implement pesticide registration and water protection programs.
In Phase II, EPA studied how detections and concentrations of pesticides and
nitrate in drinking water wells are affected by the sensitivity of ground water to
contamination, use of fertilizers and pesticides, precipitation, irrigation, the chemical
characteristics of pesticides, and the age, depth, construction, and location of drinking
water wells.
The Phase II analyses found associations that support several of the commonly
accepted theories concerning the presence of chemicals in water from drinking water
wells. In addition, some expected relationships were not found. The Survey identified
statistical associations between the presence of pesticides and nitrate in drinking
water wells and agricultural activity, the use of fertilizers, and livestock operations. The
amount of a pesticide applied for non-agricultural purposes was linked to the chance
of a nearby contaminated well. A number of factors affecting transport of chemicals to
ground water, including precipitation, the presence of surface water close to the
sampled well, and other wells operating near the sampled well, were found to be
related to the presence of pesticides and nitrate in well water. Older wells and
shallower wells were also found to be more likely to contain detectable amounts of
pesticides and nitrate. The probability of detecting pesticides or nitrate was found to
be greater in wells with low water temperature or pH. Pesticides with longer half-lives
were more likely to be detected.
Because of the many factors that can influence contamination of drinking water
wells and ground water and because ground-water contamination is irreversible, an
approach that concentrates on pollution prevention is the most effective means of
protecting drinking water wells from contamination. Using the Phase II statistical
analyses, EPA concluded that a variety of environmental conditions and human
activities combine to affect the occurrence of pesticides ana nitrate in drinking water
NPS Project Summary
-------
How EPA
Conducted
The Analysis
Major
Findings
wells and that no single factor alone can explain the presence of pesticides or nitrates.
Among the steps that need to be considered are appropriate reductions in the use of
pesticides and fertilizers, site-specific assessments to accurately target vulnerable
ground water, identification and protection of ground-water recharge areas and
protection of wellhead areas; more careful use of flood irrigation; and continued efforts
to identify problem pesticides and other materials and to establish more protective use
requirements for them.
EPA used standard statistical procedures to evaluate the data gathered by the
Survey. The data came from a broad range of sources. They included the records of
detections of pesticides or nitrate in wells sampled by the Survey, and information
from questionnaires such as pesticide and fertilizer use near wells; crops grown near
wells; the age, depth, construction, and location of wells; and information about the
location of lakes and rivers near the wells. EPA assessed ground-water pollution
potential and obtained data from sources other than the Survey questionnaires to use
in the analysis. For example, The National Oceanic and Atmospheric Administration
(NOAA) provided information on rainfall from weather stations in the counties where
wells were located that enabled EPA to develop data on precipitation for five years
prior to the Survey. NOAA also provided an index of drought conditions that enabled
EPA to investigate whether precipitation and drought affected the Survey results. EPA
used data from the Department of Agriculture's Census of Agriculture, from Resources
for the Future, and from the National Fertilizer and Environmental Research Center to
investigate how agricultural activities, livestock production, pesticide sales, and
fertilizer sales are related to detections of pesticides or nitrate in drinking water wells.
EPA used this broad range of information to determine whether individual
factors were associated with pesticide and nitrate occurrence in drinking water wells.
EPA used the results of these initial tests to investigate how the combined factors
influence the detections and concentrations of pesticides and nitrate.
EPA tested first for associations between detections and individual factors
considered one at a time. Significant results were frequently from community water
system (CWS) wells with nitrate detections. The relatively large number of
associations involving nitrate detections is related to a significantly larger number of
nitrate detections than pesticide detections were obtained. Because most of the
detections were for nitrate in CWS wells, most of the statistically significant
associations apply to that group. The Survey analysis concluded that the complex
phenomena and interactions examined in the NPS are not easily described by simple,
aggregated measures.
Site-specific assessment needed for pollution potential of well water
EPA used the Agricultural DRASTIC system to evaluate the pollution potential of
ground water in different counties and sub-county areas. DRASTIC scores are based
on seven factors: depth to ground water, recharge, aquifer media, soil media,
topography, impact of the unsaturated zone, and hydraulic conductivity. In the Phase
II analysis, EPA investigated whether overall DRASTIC scores or any of the seven
subscores, either for counties or for sub-county areas, were associated with detections
ol pesticides or nitrate in drinking water wells. EPA concluded that DRASTIC, as it
was used by the Survey, generally had not identified drinking water wells with a
greater likelihood of detections. Localized or site-specific assessments appear to be
necessary to obtain adequate evaluations of the sensitivity of drinking water wells to
contamination.
NPS Project Summary
-------
Certain measures of agricultural activity, pesticide use, and fertilizer sales shown to be
associated with detections
The Survey used several different sources of data on pesticide and nitrate use
near sampled wells. EPA collected information on agricultural crops in counties
across the country and pesticide sales to identify locations where pesticide use was
likely. Survey personnel also collected information on crops grown and pesticides and
fertilizer used near the wells from county agricultural extension agents, well owners,
and community water system operators. EPA also obtained data from other sources
on pesticide sales, nitrogen fertilizer sales, and agricultural activities to provide indirect
measures of pesticide and nitrate use.
The indirect measures of pesticide and nitrate use showed strong associations
between use and detections. The market vaJue of crops was highly related to
pesticide detections in rural domestic wells. Pesticide detections were less likely for
areas with fertilized pasture and rangeland and less likely to be associated with
numbers of beef cattle. Data on use of pesticides by urban applicators and golf
courses showed a relation between detections of DCPA acid metabolites and the rate
of DCPA application. DCPA acid metabolites were the most frequently detected
pesticide analytes and result from the degradation of the pesticide DCPA. The market
value of livestock was related to nitrate detections. The amount of fertilizer sold in a
county was found to be associated with concentrations of nitrate in wells in that
county.
Pesticide use data from Survey did not show strong associations with detections
The Survey's questionnaire data provided direct measures of agricultural
pesticide use near the sampled wells. The pesticide use data from the questionnaires
did not show the same strong associations with detections. Pesticides were detected
where they were not reported as used, and reported as used where they were not
detected. The data collected about each of the sampled wells did not show that
pesticide detections could be linked to misuse of the pesticide (accidental spills or
leaks, or mixing or disposal in an improper location). In considering how detections
might occur without reports of use of that pesticide, EPA concluded that the time
period (three or five years) about which it asked for use information might not have
been long enough, that the geographic area about which it asked for information
might not have included the area of recharge for the ground water that was the source
for the well, or that respondents' information might sometimes be mistaken or
inaccurate
Moist conditions associated with fewer detections and lower concentrations of nitrates
The Phase II analysis obtained information from the National Oceanic and
Atmospheric Administration (NOAA) about precipitation in the counties where wells
had been sampled. NOAA also provided an index of mcist ard drought conditions
(the Palmer Draught Index) for those counties. EPA concluded that detections were
less likely in areas with increased precipitation. The analysis of the Palmer Drought
Index data showed ihat fewer nitrate detections and lower concentrations of nitrate
occurred in moist areas, at least in CWS wells.
Shallower wells and older wells associated with detections
Many details of well construction and condition were obtained from the Survey's
questionnaires and analyzed in Phase II. Shallower wells were shown to be
associated with more frequent detections. The analysis showed that wells located
near surface bodies of water such as lakes and rivers were less likely to have
NPS Project Summary
-------
pesticides or nitrate detected in them. Weaker results also indicated that older wells
were associated with more frequent detections.
Results for In addition to analyzing single factors, EPA also studied whether combinations
Combinations of variables are good predictors of the occurrence of pesticides and nitrate. The
of Factors following table shows that several factors, when analyzed together, are statistically
related.
TABLE 1: LEADING FACTORS ASSOCIATED WITH CONTAMINATION
OF DRINKING WATER WELLS*
CWS Wells
Pesticide
Detections
• Fertilized
Pasture and
Rangeland
• Well Depth
• Other operating
wells
Nitrate Detections
• Fertilized Pasture
and Rangeland
• Monthly
Precipitation
• Well Water pH
• Property Farmed
Nitrate
Concentrations
• Monthly
Precipitation
• Well Water
Conductivity
• Total Nitrogen
Fertilizer Sales
• Well Depth
• Palmer Drought
Index Score
• Market Value of
Crops
Rural Domestic Wells
Pesticide
Detections
• Market Value of
Crops
• Number of Beef
Cows
Nitrate Detections
• Well Age
• Monthly
Precipitation
• Well Water pH
• Drainage Ditch
Within 1/2 Mile
• Fertilized
Pasture and
Rangeland
Nitrate
Concentrations
• Well Depth
• Market Value of
Crops
• Surface Water
Within 1/2 Mile
• DRASTIC
Topography
Score
• Total Nitrogen
Fertilizer Sales
The NFS Phase II Report gives a complete description of the relative importance and direction
of associations.
Exposure
Assessment
Estimates of potential health ejects remain low
EPA used the data generated by the Survey to estimate national concentration
distributions for the two chemicals detected most frequently: nitrate and DCPA acid
metabolites. Such estimates also had been prepared in Phase I using only
NPS Project Summary
-------
Recommen-
dations for
Design of
Future
Studies
concentrations that exceeded minimum reporting levels. Due to possible occurrence
below the minimum reporting limits, the Phase II estimates indicate that the frequency
of occurrence of these chemicals is somewhat greater than indicated by the estimates
reported in Phase I. Approximately 10.4 million people in the United States are served
by CWS wells or rural domestic wells^that contain DCPA ,. "id metabolites, but none
are expected to be exposed above the Lifetime Health Advisory Level (HAL) of 4,000
parts per billion (ppb). About 4.5 million people in the U.S., including 66,000 infants
under one year of age, are served by CWS wells or rural domestic wells that exceed
the maximum contaminant level (MCL) for nitrate of to parts per million (ppm).
Persons with high levels of nitrate in their private wells should consult their
pediatricians and may wish to obtain water from alternate sources that have less than
10 mg/L of nitrate to help protect infants from the risk of methemoglobinemia (blue-
baby syndrome). Physicians are usually well informed about the risks to infants of
high levels of nitrate in drinking water and are able to provide medical treatment.
Public water supplies that violate the maximum contaminant level of 10 mg/L for nitrate
are required to notify their customers about the violation, and the adverse health
effects caused by nitrate. Systems that apply for variances or exemptions while in
violation of the standard may be required by the state to provide bottled water or
point-of-use or point-of-entry devices to avoid unreasonable risks to health. Local and
state health authorities are the best source of information concerning alternate sources
of drinking water for infants.
EPA estimates that about 10.4 percent of CWS wells and 4.2 percent of rural
domestic wells contain detectable levels of one or more pesticides. The Phase II
study estimated the chance that a well that contains one or more pesticides also
exceeds the MCL or HAL for those chemicals for which an MCL or HAL has been
established. Although the Survey did not identify any CWS wells exceeding health
based limits for pesticides. EPA estimates that no more than 7.3 percent of the 10.4
percent of CWS wells that contain one or more pesticides could exceed an MCL or
health advisory. Similarly, no more than 28.3 percent of the 4.2 percent of rural
domestic wells that contain detectable levels of one or more pesticides are also
expected to exceed a health based limit. In summary, about 1 percent of all drinking
water wells in the U.S. are estimated to exceed a health based limit. EPA concluded
that the overall chance of a given well exceeding a level of concern for a pesticide is
low. If a well contains a detectable amount of one or more pesticides, it has a slightly
higher risk of also exceeding a health based limit. EPA recommends that well owners
that know or suspect that their well is affected by pesticides have the water tested to
ensure that any pesticides are present at levels below the MCLs or health advisories.
The NPS was the first survey of the presence of pesticides and nitrate in public
and private drinking water wells throughout the United States. Its results provide
several useful lessons for the design of future studies. The Survey design functioned
effectively to produce the data upon which the Phase I national population estimates
were based. To provide the best data for complex statistical analysis of survey results,
EPA recommends that future studies carefully consider the following points.
• Survey designs should only call for sampling a higher proportion of wells in
areas that meet certain previously specified criteria (such as areas of high
pesticide use or high ground-water sensitivity) in cases when it is known that
the selected criteria have a proven and measurable influence and the
characteristics of those areas can be defined and measured with accuracy.
Pilot studies should be conducted that test statistical analytic approaches as
well as planned laboratory and survey data collection procedures.
• Specifications for survey size and precision and the limits established in
chemical analytic procedures for reporting detections (such as minimum
reporting limits) should be chosen to ensure that a sufficient number of
NPS Project Summary
-------
Further
Studies
Needed
Access to
Databases
Where to Go
for More
Information
detections are likely to be obtained to support planned statistical procedures.
All laboratory results should be reported and all detections should be confirmed
by the appropriate chemical analytic methods.
Site-specific data on pesticide use and ground-water sensitivity should be
obtained. In addition to determining the sensitivity of ground water, data on the
recharge patterns for particular wells is also desirable. Detailed, publicly
accessible, data on actual pesticide use should include non-farm as well as
farm pesticides, and data should also be gathered on both farm and non-farm
uses of fertilizers.
Pesticide metabolites should be included in sample analysis.
The Survey analysis identified a number of additional topics for future study.
They include studies of seasonal and temporal effects on contamination, analysis of
links between surface and ground-water contamination, and collection and evaluation
of site-specific data on soil characteristics and recharge and their association with
contamination patterns in wells.
The data collected by the Survey, documented computer files, and records of
the analyses carried out to prepare the Phase I and Phase II results will be available.
Contact the EPA Office of Pesticide Programs Docket for access to the data,
documentation, and records.
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 Safe
Drinking Welter Hotline, 1-800-426-4791):
Survey Design
Survey Analytes
Fact Sheet for each
detected anatyte
Analytical Methods
How EPA Will Use
The NPS Results
Summary Results
Glossary
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
Public Information Branch (H-7506 C)
401 M Street, SW Room NEG004
Washington, DC 20460
(703) 557-2805
National Technical Information Service (NTIS)
5285 Port Ftoyal Road
Springfield, VA 22161
(703) 487-4-650 or
1 -800-336-4700
Information on regulation of
pesticides in drinking
water
Information on health
effects and safe
handling of pesticides
Background documents
for Survey and documented
databases (available for
review)
Copies of the
NPS Phase I Report
(NTIS #PB91-125765)
and
NPS Phase II Report
(NTIS #»<
(available 1992)
WPS Project Summary
------- |