EPA/TSD/OW-83
The Ground Water Supply Survey
Summary of Volatile Organic Contaminant Occurrence Data
by
James J. Westrick
J. Wayne Mello
Robert F. Thomas
Technical Support Division
Office of Drinking Water
Office of Water
Environmental Protection Agency
Cincinnati, Ohio
January 1983
-------
Executive Summary
In response to the increasing nationwide concern over contamination
of ground water supplies by volatile organic chemicals (VOCs), the
Office of Drinking Water (ODW) of the Environmental Protection Agency is
considering possible regulatory options for limiting VOCs in drinking
water. In order to strengthen the VOC occurrence data base and to
encourage state involvement in the VOC problem, ODW conducted a sampling
and analysis program on finished water from 945 water supplies which use
underground sources. The sampling program was conducted from December 1980
to December 1981. One hundred eighty-six (186) supplies from a random
list of systems serving more than 10,000 persons and 280 supplies
selected at random from systems serving less than 10,000 persons were
sampled to provide estimates of national occurrence. An additional 479
supplies were selected by state agencies for sampling and analysis.
This group of supplies was designated the nonrandom sample. The states
were encouraged to choose supplies for the nonrandom sample for which no
prior VOC data were available and which the state agencies believed had
a higher than normal probability of contamination by VOCs, based upon
their knowledge of local conditions, i.e., proximity to landfills,
industrial activity, etc.
An extensive quality assurance protocol was followed which provided
careful monitoring, control, and documentation of the quality of the
analytical data and resulted in a high degree of confidence in the
identification and quantisation of VOCs.
-------
-------
The percentages of supplies containing various levels of VOCs are
shown below.
toimber of X of Supplies with Summed Concentrations
Supplies of VOCs Greater than the Value Shown
Random Sample
< 10,000
> 10,000
Nonrandom Sample
< 10,000
> 10,000
280
186
321
158
NOTE: Concentrations are
OL - Lower
> OL >1_.0 > 5.0 >_10 >_50 > 100
16.8 7.1 2.9 1.8 0.4 0
28.0 14.0 6.5 3.4 0.5 0
22.4 12.8 4.7 3.1 0 0
37.3 27.2 17.7 13.9 3.8 1.9
in ug/1
quantitation limit
Of the small systems in the random sample, 16.8% contained at least
one VOC above its quantitation limit, 2.9% had levels totaling more than
5.0 ug/1, and 0.4% had summed VOC concentrations exceeding 50 ug/1.
Twenty-eight percent (28%) of the randomly selected systems serving more
than 10,000 persons contained measurable VOCs, while 6.5% had summed VOC
concentrations above 5.0 ug/1 and 0.5% were contaminated with VOCs
totaling more than 50 ug/1.
The nonrandom sample showed higher frequencies of occurrence for
both large and small systems at all levels. The percentages of large
and small systems contaminated at any level in the nonrandom sample were
both only 33% higher than those in the random sample. However, in the
ii
-------
-------
higher levels of contamination, the percentages of "hits" were two to
four times greater in the state-selected sample than in the random
sample.
The five compounds that occurred most frequently in the samples
analyzed during the survey are shown below, along with their frequencies
of occurrence in the subsets of the random and nonrandom samples.
Frequency of
Random
trichloroethylene
1,1,1-trichloroethane
tetrachloroethylene
ds/trans-l,2-dichloroethylene
1,1-dichloroethane
< in, ooo
3.2
4.3
4.6
1.1
3.6
> 10,000
11.3
8.1
11.3
7.0
4.3
Occurrence,
%
Nonrandom
< 10,000
7.2
8.1
8.4
3.4
1.9
> 10,000
24.1
15.8
11.4
17.1
10.8
In the random sample, trichloroethylene occurred at levels greater
than 5 ug/1 in 2 of 280 small systems and in 3.2% of the large systems
(6 of 186). Trichloroethylene exceeded 50 ug/1 in one of the randomly
selected systems.serving more than 10,000 persons. Eight other compounds
exceeded 5 ug/1 at least once in the random sample.
In the state selected sample there were 52 occurrences of single
compounds above 5 ug/1, with trichloroethylene, cis/trans-l,2-dichloroethylene,
and tetrachloroethylene accounting for 18, 11, and 10 of these 52 occurrences,
respectively. Trichloroethylene was found 3 times at concentrations
iii
-------
-------
exceeding 50 ug/1. Tetrachloroethylene and cis/trans-l,2-dichloroethylene
were each found once at levels above 50 ug/1 in the nonrandom supplies.
Thirty-six of the 52 occurrences over 5 ug/1 were in systems serving
more than 10,000 persons.
iv
-------
-------
The Ground Water Supply Survey
Summary of Volatile Organic Contaminant Occurrence Data
Background
Volatile organic contaminants (VOCs) are a general category of
synthetic organic chemicals which include low molecular weight, volatile
aliphatic and aromatic hydrocarbons, many of which are halogenated. The
presence of VOCs in ground water supplies has been reported with increasing
frequency as the use of the sophisticated analytical techniques required
for their detection grows, along with public concern over the potential
for contamination of ground water by chemical spills, improper chemical
waste disposal practices and other of man's activities.
In order to supplement existing data on the occurrence of VOCs in
drinking water, for purposes of development of regulatory alternatives,1
the EPA, Office of Drinking Water (ODW), Technical Support Division
(TSD), Cincinnati, Ohio, has conducted an extensive sampling and analysis
program to examine the occurrence of VOCs in drinking water from ground
water sources. The sampling program was initiated in December 1980 and
completed in December 1981. The major objectives of the program were
twofold: 1) provide additional data for the estimation of the nationwide
occurrence of VOCs in drinking water supplied from underground sources;
and 2) to collect information and the physical characteristics of the
well field and the surrounding area in an effort to develop a predictive
capability for locating contaminated ground water. Only the occurrence
data is discussed in this report.
-------
-------
The survey was divided into two parts. One half the sampling and
analytical resources were dedicated to the development of data from a
random sample of ground water supplies drawn from the inventory of
public water systems maintained in the Federal Reporting Data System
(FRDS). A list of 500 supplies was selected at random from the national
inventory. The list was developed from two subsets. Three hundred
supplies were drawn from systems serving fewer than 10,000 persons,
while the remaining 200 supplies served more than 10,000 persons. A
second randomly generated two-part list of supplies provided replacement
sampling sites in cases where the supplies on the primary list of 500
were inappropriate or non-existent. This could occur if, for example, a
utility had recently begun to purchase its water from another utility
which used a surface water source.
VOC occurrence data gathered by TSD from analysis of samples col-
lected from ground water systems during the Community Water Supply
o
Survey (CWSS) of 1978 were used to estimate the necessary sample sizes.
The frequency of occurrence of any of the 10 VOC parameters for which
analyses were conducted in samples from 301 ground water systems in the
CWSS serving less than 10,000 persons was 0.12 (12% of the systems).
Forty-five percent (0.45) of the 29 large ground water systems studied
contained at least one VOC. Using these occurrence frequencies as
estimates of what might be found in this survey, it was determined that
sample sizes of 300 small systems and 200 large systems would allow at
least 0.95 confidence that the size of the error of estimating occurrence
frequencies would be no more than 0.036 (30% of the CWSS point estimate)
for small systems and 0.069 (15% of the CWSS point estimate) for large
3
systems.
-------
-------
The second part of the survey was utilized to encourage the state
agencies to attempt to identify problem supplies. This would not only
expand state involvement in the VOC problem, but also provide the Office
of Drinking Water with some information on the frequency and extent of
serious problems that could be found based on the state agency's knowl-
edge of local conditions. Each state was assigned a number of sampling
sites (supplies) roughly proportional to its fraction of the total
number of ground water systems nationwide; this was designated the
nonrandom sample. The target number of nonrandom sites was also 500.
The state agencies were encouraged to select supplies that they sus-
pected might be contaminated by VOCs because of the proximity of the
well field to industries, landfills, or other potential sources of
contamination. They were also encouraged to choose supplies for which
no VOC data were available in an effort to maximize the discovery of
previously unknown contamination.
Approach
To obtain information from a maximum number of supplies within the
available resources, it was decided to collect one sample of finished
water from each utility at a point near the entrance to the distribution
system. The VOC concentrations in water supplied from a single well
that is not pumped continuously can vary depending on pumping rate and
schedule and the hydrodynamics of the plume of contamination. If mul-
tiple wells supply a system at a single entry point and some wells are
contaminated while others are not, the VOC concentration in the sample
-------
-------
at the entry point could vary greatly, depending on which wells were in
operation at the time of sampling. In systems with more than one entry
point, a single sample would obviously represent only those wells con-
tributing to that entry point. With these limitations in mind, the
sample of finished water taken at or near a point of entry provides a
reasonable compromise between the information obtained from a single
sample from a single well and that from multiple samples taken throughout
the system.
State drinking water agencies played a major role in the planning
and execution of this survey. Each state with primary Public Water
System enforcement responsibility (primacy) was contacted through the
regional EPA drinking water offices. Most of the states indicated a
willingness to assist in the project. State involvement in planning
consisted of reviewing the primary random list for errors or inappropriate/
non-existent systems, filling in missing information on the random
supplies, selecting systems for inclusion in the nonrandom portion of
the survey, and providing scheduling information to the TSD project
engineer. In most cases, state personnel traveled to the sampling
sites, collected the samples and site information and shipped the samples
in ice to the TSD laboratory in Cincinnati, Ohio, using sampling supplies
(bottles, insulated shipping boxes, etc.) provided by TSD. In non-
primacy states and states that were unable to assist in the planning or
the sampling because of budgetary or other constraints, regional EPA
personnel provided the necessary assistance. TSD personnel collected
some samples that were obtainable within reasonable driving distance
-------
-------
from Cincinnati or when it was not possible for either state or regional
EPA personnel to travel to the sampling locations.
After the first round of sampling and analysis was completed, the
states were offered the opportunity to resample finished water and water
from various wells in systems found, in the initial round of sampling,
to be contaminated. This would not only be of assistance to the states
1n their investigation of the extent of contamination, but would also
provide additional data to EPA. An examination of the resample data
could provide insight as to the variability of contaminant concentration
with time and also lend confidence in the results obtained from the
initial round of sampling and analysis.
Logist4cs
A sampling kit was prepared at TSD for each sampling location.
Amber bottles of 60 ml and 250 ml capacity were dosed with a preserva-
tive (mercuric chloride at 10 mg/1), capped with teflon septa and screw
caps, labelled with preprinted labels which had been stamped with the
sample identification numbers, and secured in "styrofoam" boxes. A
shipping blank (250 ml bottle containing organic-free water and preserva-
tive) was also included with the sampling kit. The shipping blanks were
to remain with the sampling kit through all stages of transportation and
storage. Any possibilities of contamination from the surroundings could
be investigated by analysis of the shipping blank. The styrofoam boxes
had been custom molded to hold the proper number of bottles. The bottles,
along with a plastic bag and tie, a sampling site data sheet, sampling
and shipping instructions, and shipping labels and forms were shipped to
-------
-------
the sample collectors on a schedule which had been prearranged with the
states. The sample collectors took the samples, filled in the labels
and site data sheets, iced and secured the boxes, and delivered them to
an overnight freight delivery service. All shipping costs were paid by
EPA.
Samples were received at TSD on the day after they were collected.
They were unpacked, logged in, and any unusual circumstances were noted.
The sample bottles were then placed in storage in a cold room free of
organic vapor contamination until they were repacked for shipment to the
chemical analysis contract laboratory, SRI International, Inc.,
Menlo Park, California. Replicate samples were collected at each site
so half the bottles were shipped to the contract laboratory and half
were held in cold storage at TSD. This was necessary for occasional
analysis of sample duplicates by TSP chemists or for quick-response, in-
house verification of contract laboratory results.
When the samples were shipped to the contract laboratory, the
pertinent sample information was entered into the TSD data system for
tracking purposes. Primary analysis of the samples was completed within
30 days of collection. The contract laboratory was provided access to
the data system, so upon completion of the analyses for a sample the
data were entered at the contractor's terminal and retrieved by the TSD
project engineer in Cincinnati. The results for each sample were examined
by the TSD contract project officer and verified by agreement between
the project officer and the SRI project leader after consideration of
quality assurance information. The verified data were then entered into
a confirmed data file for data analysis and reporting.
-------
-------
Analytical Considerations
A total of thirty-four parameters were selected for analysis by
purge and trap gas chromatographic methods. The parameters are listed
in Table 1. Any discussion of VOCs in this report includes only the 29
non-trihalomethane compounds listed. Five trihalomethanes (THMs),
chloroform, bromodichloromethane, dibromochloromethane, dichloroiodomethane,
and bromoform were included in the analysis, even though their presence
nearly always results from the reaction of chlorine and precursor substances
upon disinfection. The samples had not been dosed with a reducing
agent, so the THM formation reaction proceeded until the time of analysis
or until the depletion of either chlorine or precursor material.
The two isomers of 1,2-dichloroethylene could not be separately
determined by the analysis and thus are considered in this report to be
one parameter. The same is true for ortho- and para-xylenes. Methylene
chloride originally was to be determined, but this compound is a very
widely used laboratory solvent and appears frequently as a laboratory
contaminant. Because it was found in all the analyzed shipping blanks,
it was virtually impossible to ascertain whether the methylene chloride
detected in a sample was originally present or if the sample had become
contaminated from the surrounding atmosphere. Therefore, results for
methylene chloride could not be validated and none are included in this
report.
The purge and trap GC analyses were conducted according to EPA
4 5
methods 50,2.1 and 503.1 with a significant modification. The non-
destructive photoionization detector for analysis of aromatics and the
Coulson electrolytic conductivity detector for the analysis of halocarbons
-------
-------
were directly coupled in series, allowing analysis for the complete list
of thirty-four compounds with one sample purge. A comparability study
conducted prior to the survey showed that the serial analysis gave
results equivalent to separate analyses for the two types of compounds.
This technique proved to be very beneficial in terms of the time for and
cost of analysis. An additional benefit resulted from the acquisition
of extra information by using the two detectors in series. The photoionization
detector can assist in identifying and quantifying compounds that co-
elute from the primary GC column or that have poor responses with the
Coulson detector.
Residual chlorine was measured by a colorimetric kit in samples
from supplies that practice chlorination. This measurement was done to
provide information supplemental to the THM data. As a general indicator
of non-specific organic content, total organic carbon (TOC) was measured
on all samples by means of a low level carbon analyzer. Because TOC and
residual chlorine measurements are of secondary interest, those data are
not presented here but will be included in a later report.
Duality Assurance
When the contract for analytical services was written for the Ground
Water Supply Survey, a detailed quality assurance protocol was included
to monitor and maintain the quality of data generated in the analyses.
This protocol was followed throughout the survey and the validating data
were continuously scrutinized by the Project Officer. Table 2 is a
summarized list of the quality assurance employed.
-------
-------
EPA Reference Samples
The precision and accuracy for analysis of the EPA halocarbon and
aromatic reference samples, which were analyzed weekly, easily met the QA
specifications. This was true for both the primary and confirmatory
analytical schemes. The EPA reference samples contained known concentrations
of compounds Including the four common trihalomethanes and nine
frequently found VOCs. The precision measure for the analysis of reference
samples used herein is coefficient of variation, the relative standard
deviation. This is the standard deviation of approximately 50 analyses
divided by the mean of those values. The precision of the primary analysis
of reference samples at levels below 5 ug/1 averaged +13% with a range of
+8% for tetrachloroethylene to +22% for 1,1,1-trichloroethane. For reference
samples containing levels greater than 5 ug/1, the precision ranged from
+6% for trichloroethylene to +20% for 1,1,1-trichloroethane with an
average precision of +11%. Accuracy is indicated by the percent error,
that is the difference between the mean of the measured values and the
expected (true) value divided by the expected value. This parameter
ranged from 0% for tetrachloroethylene at 5.9 ug/1 to -19% for
dibromochloromethane at 2.1 ug/1 with averages of -9% below 5 ug/1
and -4% above 5 ug/1. Negative error indicates that the mean of the
measured values was less than the expected value.
Duplicate Analyses
As another gauge of precision, the contract called for duplicate
analyses to be performed on"a minimum of 10% of the samples. The duplicates
-------
-------
were to agree within 40% for compounds present below 5 ug/1 and within
20% above 5 ug/1. The precision measure used here is the percent
difference, 100 times the absolute difference in the duplicate values
divided by the mean of the t>*o values. Data were gathered on 16 individual
compounds present collectively in the duplicate analyses. A total of
84 quantifiable results less than 5 ug/1 were duplicated with all but five
meeting the precision criterion. The average percent difference for the
quantifiable low level duplicate results was 17%. Eighteen quantifiable
pairs of duplicate results greater than 5 ug/1 were reported, with four of
the eighteen falling outside the precision limits. The average precision
of the eighteen pairs of higher lever determinations was 13%. Although
these figures cannot be compared directly to the precision obtained
with the EPA reference samples, they are of a similar magnitude and reflect
the consistency of replicate analyses done in the contractor's laboratory.
Confirmatory Analyses
All samples found or suspected to contain purgeable aromatic and halo-
carbon compounds other than the THMs were reanalyzed using different chroma-
tographic columns that elute the compounds in different orders. In addition,
samples containing chloroform at concentrations greater than 40 ug/1 were
reanalyzed using the confirmatory column because chloroform at this concen-
tration level could mask small quantities of 1,2-dichloroethane. Approximately
33% of all samples were reanalyzed by second column chromatography for
halocarbons and 6% for aromatics. Although it was not required by the
10
-------
-------
contract, the contract laboratory reported not only confirmatory identi-
fication but also concentration of compounds run by second column chroma-
tography. Precision and accuracy for the analyses of 19 EPA reference
samples for halocarbons and 11 EPA reference samples for aromatics were
documented. All accuracy values were within the contract limits for the
primary analyses and the precision values for all but two of the compounds
fell within the primary analysis error limits. In addition, approximately
5% of all the samples were reanalyzed by gas chromatography/mass
spectrometry for additional confirmation and tentative identification
of unknown peaks.
Blind Samples
Five blind samples were prepared by TSD in the initial phase of
the survey to ascertain the contractor's ability to qualitatively
identify particular compounds and quantitatively measure them. The
blinds consisted of five different mixtures of compounds, spiked into
organic-free distilled water. These were periodically sent to the
contractor early in the survey period disguised as survey samples.
The mixtures were designed to pose selected anomalies in the
analytical system, such as interferences or compounds with similar gas
chromatographic retention times. Prior to shipment, the blinds were
analyzed by TSD and these results were compared to those subsequently
reported by the contractor. In every case, the contractor correctly
identified the spiked compounds. Although there were no quantitative
criteria established for the blind samples, the percent differences
11
-------
-------
between SRI's results and TSD-determined concentrations were within
the error limits for duplicates for 27 of 32 pairs of values.
TSD Analysis of Duplicate Samples
Replicate samples were collected in separate bottles and stored
at TSD so they could be analyzed as an additional check on the contract
laboratory results. The TSD-analyzed check samples were chosen from
those that the contractor had reported to contain one or more of the
purgeable organics. Although there were no quantitative criteria
established for inter-laboratory analysis of duplicate samples, the
percent differences between SRI's results and those of TSD were within
the error limits established for duplicate analyses for 48 of 64 pairs
of values above the lower quantitation limits. The error limits used
for the comparison here are those established for a single laboratory
conducting duplicate analyses of the same sample. Larger percent
differences were expected for this comparison since the analyses were
done on duplicate samples and analyzed by totally independent systems.
It is important to note that these duplicate samples often contained
several compounds at widely varying concentrations, from <1 ug/1 to
over 100 ug/1.
The quality assurance program was a major, critical part of the
analytical activity. It consumed a significant fraction of the analytical
resources expended by the contract laboratory and required considerable
time and effort by TSD personnel. Careful attention to the monitoring,
control, and documentation of the data quality resulted in a high
12
-------
-------
degree of confidence that the identification and quantisation of compounds
were accurate.
The data supporting the preceding discussion and an in-depth
description of the analytical quality assurance program for this survey
can be found in the contractor's final report and in a recent paper
by Kingsley.
Results
The distribution of both random and nonrandom samples is shown in
Table 3 and Figures 1 and 2. The final numbers of random systems sampled
were 280 serving fewer than 10,000 persons and 186 serving more than
10,000, 34 systems less than the target of 500 random systems. The
final tally for state-selected sites was 479. Because of the random
selection process, the number of random sample sites located in each
state should be roughly proportional to the number of ground water
systems in that state. The number of nonrandom samples allocated to a
state was also based approximately on its number of ground water systems.
Figures 1 and 2 are pictorial representations of all sampling locations.
The open circles represent sampling sites where no VOC contamination was
detected and the closed circles represent locations of samples that
contained at least one VOC above the quantitation limit. The quantisation
limits are not the same for all compounds as can be seen from subsequent
tables. In most cases, the quantitation limit is either 0.2 ug/1 or
0.5 ug/1. This difference in lower quantitation limits can confuse the
interpretation of data somewhat, and the occurrence data presented
herein should be viewed with the different quantitation limits in mind.
13
-------
-------
For purposes of this report, unless otherwise stated, an occurrence is
defined as any specific organic parameter which is found at or above its
lower quantitation limit.
Tables 4 and 5 provide summary occurrence data from the random
sample for each of the 34 parameters. Table 4 contains data for all
specific parameters from the random sample of systems serving fewer than
10,000 persons and Table 5 contains the results for systems serving over
10,000 persons. The quantitation limit for each parameter, the fre-
quency of occurrence (number and percent of samples in which each con-
taminant was found), the median concentration of the positive values of
each compound and the highest concentration found are presented.
Because the two parts of the random sample were selected independ-
ently and because a much higher percentage of large systems than small
systems were included (15% of roughly 1200 systems serving greater than
10,000 vs. 0.6% of nearly 48,000 systems serving fewer than 10,000
persons), the data from the large systems and the small systems are not
combined for analysis. The normal curve approximation to the binomial
distribution for large samples was used to conduct tests of significance
of the difference in frequency of occurrence of compounds. The large
system frequency of occurrence was significantly greater than the small
system frequency for trichloroethylene, 1,2-dichloroethylene, and
tetrachloroethylene at the 0.01 significance level, and for 1,2-dichloro-
propane, carbon tetrachloride, and 1,1,1-trichloroethane at the 0.05
significance level. No other significant differences in the occurrence
of specific parameters could be discerned between the large system and
small system samples.
14
-------
-------
While the data clearly indicate that the frequency of occurrence of
several of the compounds is higher among the larger communities sampled,
a similar inference cannot be drawn regarding the severity of contami-
nation from a casual observation of the data in Tables 3 and 4. The
highest concentrations of 1,2-dichloropropane, trichloroethylene, and
benzene were found in samples from the larger communities, while small
system samples contained the highest levels of 1,1,1-trichloroethane,
carbon tetrachloride, and tetrachloroethylene. Because of the possibly
large variability in concentration at each sample point anything more
than a cursory examination of the concentration distribution is beyond
the scope of the available data.
Trihalomethane data are also shown in the tables. Trihalomethanes
occurred more frequently in the larger system samples, but this could
result from the higher percentage of large systems that chlorinate their
water supplies (85% of the large systems sampled vs. 56% of the smaller
systems sampled). The THM concentrations were generally low, as demon-
strated by the median values, while it is also evident that some ground
waters can produce very high THM values. Again, the variable lower
quantitation limits must be recognized in considering both the frequency
of occurrence and the median of the positive values for the THMs. Since
the samples were normally analyzed after about 1-4 weeks of low tem-
perature storage, the THM concentrations reported are undoubtedly higher
than they would have been had the THM formation reaction been stopped by
a reducing agent at the time of sampling and may or may not be repre-
sentative of concentrations in the distribution systems. However, the
15
-------
-------
data do provide a good indication of the tendency for THM formation in
ground water supplies.
Examination of the individual supply data and site information
indicated that there may have been a few THM occurrences which were not
the result of chlorination of the water supply (probably no more than
six, all with very low concentrations). Because the focus of this
survey was on VOCs other than THMs, the THM data will not be addressed
further 1n this report.
There is no evidence in the literature that chlorination of drinking
water causes the formation of any of the non-THM VOCs. There have been
reports that commercial chlorine can contain traces of carbon tetrachloride
which can contaminate chlorinated drinking water. Therefore, the carbon
tetrachloride occurrence data presented herein should be qualified by
the possibility that the cause of some carbon tetrachloride occurrences
may be contaminated chlorine.
The data from the random sample of systems serving less than 10,000
persons were examined for any other possible effects of chlorination.
There was no significant difference in occurrence frequencies for any of
the VOCs between small systems that chlorinate and those that do not.
The larger systems which do not chlorinate are too few in number to
provide a valid comparison with larger systems which do chlorinate.
The number and percentage of contaminated supplies in each part of
the random sample are listed by the number of contaminants in Table 6.
Of 280 small systems, 47 contained one or more of the 29 VOCs included
in the analysis. Of those 47 supplies, 19 supplies had multiple contaminants
16
-------
-------
above the quanfitation limit. Of the 186 larger systems, 52 contained
at least one contaminant. Twenty-five of those 52 supplies contained
more than one VOC.
Water samples from 16.8% of the systems serving less than 10,000
persons and 28.0% of the larger systems contained at least one VOC.
Therefore, based on data from this survey alone, the point estimates of
the probability that systems serving under and over 10,000 persons are
contaminated with at least one VOC above its quantisation limit are
16.8% and 28*0% respectively. A computer program was written to calcu-
late the confidence limits of the estimates based on the binomial distri-
bution. The confidence interval is a function simply of the observed
frequency and the sample size and does not account for any uncertainty
due to analytical variability or variation in water quality. The frequency
of occurrence in all systems serving less than 10,000 persons can be
estimated with 95% confidence to lie within the range 12.9% - 21.7%.
Likewise, the large system frequency of occurrence can be estimated to
lie within the 95% confidence interval 22.1% - 35.0%. The frequency of
occurrence for the large systems was greater than that for the small
systems at the 0.01 significance level.
Information on the concentration distributions of the VOCs that
occurred in the random sample is further broken down by smaller population
categories in Table 7. There were 16 occurrences greater than 5.0 ug/1.
Trichloroethylene was found at > 5.0 ug/1 in eight supplies, once at
> 50 ug/1.
17
-------
-------
Table 8 is a summary of the data by the same population categories
and concentration cells used In Table 7, but showing the number of
supplies with a summed concentration falling within various concentra-
tion ranges. For example, there were 88 supplies in the population
category 101-500. Of those 88 supplies, 77 contained no VOC above its
quantitation limit, 9 contained one or more contaminants with the sums
of the concentrations less than 5.0 ug/1, and two supplies had a summed
VOC concentration 1n the range 11 ug/1 - 50 ug/1. The point estimate of
the probability of a system that serves more than 10,000 people con-
taining a summed VOC concentration of more than 5.0 ug/1 was 6.5% (12 of
186) with a 95% confidence interval of 3.8% - 11%. Eight of 280 small
systems (2.9%) contained a VOC total of > 5.0 ug/1, resulting in a 95%
confidence Interval for the estimate of 1.5% - 5.6%. The frequency of
occurrence of summed VOCs > 5.0 ug/1 was greater in large systems than
in small systems at the 0.05 significance level.
The Nonrandom Sample
Because of the nature of the nonrandom sample, where the states
selected sites that they suspected might be contaminated, no statistical
interpretation of the data will be attempted here. All the nonrandom
sample data are presented in Tables 9 through 14 in the same format as
the random sample data. Obviously higher frequencies and concentrations
were found in this sample set than in the random sample. Nearly one-
fourth of the large systems and 7% of the small systems selected by the state
agencies were contaminated with trichloroethylene. Other compounds
appearing frequently included cis/trans-l,2-dichloroethylene, 1,1,1-
trichloroethane, tetrachloroethylene, and 1,1-dichloroethane. Of the
18
-------
-------
131 systems found to be contaminated with VOCs, over half showed the
presence of more than one contaminant; the water from one smaller com-
munity contained eight VOCs. Trichloroethylene, tetrachloroethylene,
and cis/trans-l,2-dichloroethylene were found at levels greater than
5.0 ug/1 18, 11, and 10 times, respectively. Trichloroethylene occurred
three times at greater than 50 ug/1, tetrachloroethylene and cis/trans-
1,2-dichloroethylene once each. All xylene occurrences were in supplies
serving less than 10,000 persons; in fact, only seven occurrences of
aromatic compounds were found in the larger supplies. Although 1,1,1-
trichloroethane was the second most frequently found compound, it was
found only four times over 5.0 ug/1. Thirty-seven percent of the larger
supplies selected by the States did indeed have at least one measurable
VOC and 18% had a summed VOC concentration of > 5.0 ug/1. Of the
smaller systems selected, 22% showed some contamination with the summed
VOC concentration exceeding 5.0 ug/1 in 5% of the samples.
Resampling of Contaminated Supplies
Approximately 100 contaminated supplies were resampled. The states
were asked to resample the finished water and were also given the oppor-
tunity to collect several raw water samples of their choosing. In many
cases, the original sample point was not resampled or the sampling
points were not described well enough to enable comparison of the
original sample and the resample.
An example of data from a supply which did resample finished water
serves to point out some interesting aspects of ground water VOC data.
The concentrations of VOCs found in samples of finished water from a
single well owned by a small town (City A) collected nine months apart
are shown in Table 15.
19
-------
-------
These two samples show very much the same pattern of contamination
with a possible slight decrease in concentration. Trichloroethylene,
which was counted as an occurrence in the original sample, was not found
above the quantification limit in the resample. Table 15 illustrates
several commonly occurring circumstances. When the original sample and
the resample represented water from a single well, both usually con-
tained nearly identical patterns of contamination. This increases con-
fidence that the original results were accurate. It also reinforces the
belief, widely held, that ground water contamination levels usually
change very slowly. It was not uncommon that the occurrence of a
compound at or near the quantitation limit was not repeated in the
resample. For example in 37 supplies for which the original sample
point was resampled, there were 25 occurrences which did not recur in
the resample (many of the supplies resampled had multiple occurrences).
However, there were also 16 instances when a compound that was not found
in the original sample was quantified in the resampled finished water.
This non-repeatability occasionally occurred in well defined, single well
samples such as those shown in Table 15 for very low level contaminants.
This situation could result from either the normal analytical variability
or from actual changes in concentration at the well. It was more common
in larger systems where the finished water was a blend of water from
multiple wells with varied levels of contamination. In these cases,
changes in the contaminant concentrations could result from changes in
the relative contributions of the various wells as determined by their
pumping rates. Temporal concentration changes could also result from
20
-------
-------
relatively rapid movement of the plume of contamination, which could
occur under certain conditions of recharge and withdrawal in a highly
permeable aquifer.
The data from resampling finished water from individual wells
reinforce confidence that the identifications and quantisation of compounds
in the samples were accurate. The data from larger, multiple well
systems indicate that concentrations of compounds in a finished water
can vary considerably over a period of time. Sampling a large number of
supplies, as was done in this survey, is necessary to provide an accurate
representation of the percentage of systems with water containing VOCs
and a good indication of the magnitude of the concentration levels.
Finished water quality variability in ground water supplies is
virtually site specific and not amenable to definition by a national
survey such as this. It was recognized in the beginning that, in the
face of uncertainty, the sampling approach taken would be a compromise
between broad national coverage and a high degree of representativeness.
Subjective evaluation of the resample data has substantiated that approach.
Conclusion
The Ground Water Supply Survey was undertaken principally to
strengthen the body of data on the occurrence of VOCs in ground water
supplies. Careful attention was paid to all aspects of quality assur-
ance in order to provide a reliable representation of VOC occurrence in
the Nation's ground water supplies. The frequencies of occurrence of 29
volatile compounds in samples collected from 466 randomly selected
communities and 479 communities selected by the State agencies were
determined. The three most frequently detected compounds were trichloro-
ethylene, tetrachloroethylene, and 1,1,1-trichloroethane. The percent-
ages of supplies containing at least one VOC above its quantisation
21
-------
-------
limit 1n the various subsets of the survey were: random < 10,000 persons
- 16.8%, random > 10,000 persons - 28.0%, nonrandom < 10,000 persons - 22.4%,
and nonrandom > 10,000 persons - 37.3%. The percentages of supplies
whose finished water contain summed VOC concentrations greater than 5
ug/1 were: random < 10,000 persons 2.9%, random > 10,000 persons - 6.5%,
nonrandom < 10,000 persons - 4.7%, and nonrandom > 10,000 persons - 17.7%.
Simple statistical tests, based on the random parts of the survey
revealed significant differences in the frequencies of occurrence in the
larger and smaller community subsets of the random sample. The results
of the random sample were also used to construct confidence limits
around estimates of probabilities of occurrence.
The nonrandom portion of the sample provides additional data on the
"high side" of the occurrence curve, since the sites were selected in
hopes of finding a greater frequency of higher level contamination. For
example, six of the eight supplies with summed VOC concentrations > 50
ug/1 were from the nonrandom sample, including all three of the supplies
with summed VOC concentrations > 100 ug/1.
Resampling of contaminated supplies tended to strengthen confidence
in the quality of the analytical data. It also showed that finished
water quality, with respect to VOCs, can vary with time especially in
larger systems with multiple wells.
This report is offered as a summary discussion of the Ground Water
Supply Survey and its results. Additional analysis of the data generated
by the survey will appear in the occurrence documents prepared by the
Agency in support of VOC regulatory activity.
22
-------
-------
Acknowledgements
The authors gratefully acknowledge the contributions of the many
others who assisted in carrying out this sampling and analysis program:
the SRI International analysis team headed by Dr. Barbara Kingsley;
members of the Office of Drinking Water in Washington and Cincinnati
who participated in the planning and execution of the project and in
preparation and review of this document, including Victor J. Kimm,
Lowell A. Van Den Berg, Dr. Joseph A. Cotruvo, Dr. Arnold M. Kuzmack,
Dr. Herbert 0. Brass, Craig Vogt, Dr. David Schnare, Eric Bissonette,
Richard Johnston, Waymon Wallace, Jane Gruber, Audrey Kroner, Dale Ruhter,
and William Coniglio; the EPA regional water supply personnel who coordi-
nated and monitored and, in some cases, carried out the activities in
their regions; and especially the personnel of the state drinking water
agencies, without whose excellent cooperation the project could not have
been accomplished.
23
-------
-------
References
1. U.S. Environmental Protection Agency, "National Revised Primary
Drinking Water Regulations, Volatile Synthetic Organic Chemicals in
Drinking Water; Advanced Notice of Proposed Rulemaking." Federal
Register 47: 9350, March 4, 1982.
2. Brass, H.J., Weisner, M.J. and Kingsley, B.A., "Community Water
Supply Survey: Sampling and Analysis for Purgeable Organics and
Total Organic Carbon." Presented at the American Water Works Association
Annual Conference, St. Louis, MO, June 1981.
3. Miller, I. and Freund, J.E., Probability and Statistics for
Engineers. Prentice-Hall, Inc., Englewood Cliffs, NJ, 1965.
4. U.S. Environmental Protection Agency, "The Determination of Halogenated
Chemicals in Water by the Purge and Trap Method." Method 502.1,
EPA #600/4-81-059, April 1981.
5. U.S. Environmental Protection Agency, "The Analysis of Aromatic
Chemicals in Water by the Purge and Trap Method." Method 503.1,
EPA #600/4-81-057, May 1980.
6. SRI International, Final Report, EPA Contract No. 68-03-3031,
"Determination of the Ouality of Ground Water Supplies."
December 1982.
7. Kingsley, B.A., "Quality Assurance in a Contract Laboratory."
Presented at the American Water Works Association, Water Ouality
Technology Conference, Nashville, TN, December 1982.
-------
-------
Table 1
Specific Organic Parameters
Ground Water Supply Survey
(December 1980 - December 1981)
vinyl chloride
1,1-di chloroethylene
1,1-dichloroethane
cis and/or trans-l,2-dichloroethylene
1,2-dichloroethane
1,1,1-t ri chloroethane
carbon tetrachloride
1,2-dichloropropane
trichloroethylene
tetrachloroethylene
benzene
toluene
ethyl benzene
bromobenzene
m-xylene
o+p-xylene
p-dichlorobenzene
1,1,2-trichloroethane
1,1,1,2-tetrachloroethane
1,1,2,2-tetrachloroethane
chlorobenzene
1,2-di bromo-3-chloropropane
n-propylbenzene
o-chlorotoluene
p-chlorotoluene
m-dichlorobenzene
o-di chlorobenzene
styrene
isopropylbenzene
chloroform
bromodi chloromethane
di bromochloromethane
dichloroiodomethane
bromoform
25
-------
-------
Table 2
Quality Assurance Protocol for the Analysis of VOCs
ro
QA Analysis
EPA Reference Samples
Duplicate Analyses
Confirmatory Analyses
Blind Samples
TSD Analysis of
Duplicate Samples
Frequency or Amount
Specified Contract Limits
<5 uq/1 >5 ug/1
10% of samples
100% of positives
variable
10% of positives
Agreement: 40%
Qualitative Agreement
none specified
none specified
Source of Sample
I/week for each
Instrument
Precision:
Accuracy:
+40%
+40%
+20%
^20%
EMSL-C1n
20% Survey samples
Survey samples
TSD generated
Survey samples
-------
-------
Table 3
Number of Supplies Sampled
By State
Ground Water Supply Survey
(December 1980 - December 1981)
State
Al abama
Alaska
An zona
Arkansas
California
Colorado
Connecticut
Delaware
Florida
Georgl a
Hawaii
Idaho
Illinois
Indiana
Iowa
Kansas
Kentucky
Louisiana
Maine
Maryland
Massachusetts
Michigan
Minnesota
Mississippi
Missouri
Montana
Random
7
4
8
3
34
2
8
1
44
14
6
15
8
12
11
4
14
v
4
11
8
10
14
6
4
Non-random
5
4
9
4
30
6
7
1
31
13
2
8
12
8
13
6
3
10
2
6
4
12
9
14
10
5
State
Nebraska
Nevada
New Hampshire
New Jersey
New Mexico
New York
North Carolina
North Dakota
Ohio
Oklahoma
Oregon
Pennsylvania
Rhode Island
South Carolina
South Dakota
Tennessee
Texas
Utah
Vermont
Virginia
Washington
West Virginia
Wisconsin
Wyoming
Puerto Rico
Random
8
2
2
17
1
22
13
_
14
4
7
16
1
5
4
6
41
8
2
9
19
4
7
-
2
Non-random
6
3
4
5
6
25
31
3
15
5
7
26
2
11
4
4
33
2
3
18
10
5
13
2
2
27
-------
-------
Summary of Occurrence Data9
Random Sample
Supplies Serving Less Than 10,000 Persons'3
Ground Water Supply Survey
December 1980 - December 1981
Ouantltatlon Occurrences
Parameter Limit No. X
vinyl chloride
1 ,l-d1chloroethy1ene
1 ,1-dlchloroethane
c1s and/or trans-
1 ,2-41 chl oroethyl ene
l,2-d1chloroethane
1.1,1-trlchloroethane
carbon tetrachloHde
1 ,2-d1ch1oropropane
tr1 chl oroethyl ene
tet rachl oroethyl ene
benzene
toluene
ethyl benzene
bromobenzene
m-xylene
o+p-xylene
p-d 1 ch 1 orobenzene
1 ,1 ,2-tr1 chloroethane
1,1,1 ,2-tetrachloroethane
1,1,2,2-tetrachloroethane
chl orobenzene
l,2-d1bromo-3-chloropropane .
n-propyl benzene
o-chlorotoluene
p-chlorotoluene
m-dl chl orobenzene
o-d1 chl orobenzene
styrene
Isopropyl benzene
chloroform
bromod 1 chl oromethane
fl 1 bromoch 1 oromet hane
d 1 chl orol odomethane
bromoform
a All concentrations are In ug/1
b 280 sample sites
1.0
0.2
0.2
0.2
0.5
0.2
0.2
0.2
0.2
0.2
0.5
0.5
0.5
0.5
0.2
0.2
0.5
0.5
0.2
0.5
0.5
5.0
0.5
0.5
0.5
0.5
0.5
0.5
0.5
0.2
0.2
0.5
1.0
1.0
0
4
. 10
3
0
12
5
1
9
13
1
4
2
3
6
6
2
0
0
0
0
1
0
0
0
0
0
0
0
104
100
87
2
44
0
1.4
3.6
1.1
0
4.3
1.8
0.4
3.2
4.6
0.4
1.4
0.7
1.1
2.1
2.1
0.7
0
0
0
0
0.4
0
0
0
0
0
0
0
37.1
35.7
31.1
0.7
15.7
Median of
Positives
1.2
0.51
0.23
-
0.32
0.37
0.75
0.88
0.35
0.61
0.62
0.94
1.9
0.32
0.34
0.60
-
-
-
-
5.5
-
-
-
-
-
-
-
1.4
1.4
2.1
2.8
2.4
Maximum
Value
.
6.3
3.2
1.7
-
18
16
0.75
40
23
0.61
0.85
1.1
5.8
1.5
0.59
0.68
-
-
-
-
5.5
-
-
-
-
-
-
-
140
60
52
4.1
54
-------
-------
Table 5
Summary of Occurrence Data3
Random Sample
Supplies Serving More Than 10,000 Persons
Ground Water Supply Survey
December 1980 - December 1981
Quantltatlon Occurrences
Parameter Limit No. %
vinyl chloride
1.1-dlchloroethylene
I,l-d1ch1oroethane
els and/or trans-
1 ,2-d1chloroethylene
1 ,2-d1chloroethane
1.1,1-trlchloroethane
carbon tetrachlorlde
l,2-d1chloropropane
trlchloroethylene
tet rachl oroethyl ene
benzene
toluene
ethyl benzene
bromobenzene
m-xylene
o+p-xylene
p-dlchlorobenzene
1 ,1 ,2-trfchloroethane
1 ,1 ,1 ,2-tetrachloroethane
1,1,2,2-tetrachloroethane
chlorobenzene
1 ,2-d1bromo-3-chloropropane
n-propyl benzene
o-chlorotoluene
p-chlorotoluene
m-dlchlorobenzene
o-dl chlorob«nzene
styrene
Isopropyl benzene
chloroform
bromodlchloromethane
dlbromochloromethane
d1 chlorol odomethane
bromoform
* All concentrations are In uq/1
b 186 sample sites
1.0
0.2
0.2
0.2
0.5
0.2
0.2
0.2
0.2
0.2
0.5
0.5
0.5
0.5
0.2
0.2
0.5
0.5
0.2
" 0.5
0.5
5.0
0.5
0.5
0.5
0.5
0.5
0.5
0.5
0.2
0.2
0.5
1.0
1.0
1
5
8
13
3
15
10
5
21
21
2
2
1
1
2
2
3
0
0
0
0
0
0
0
0
0
0
0
0
106
101
96
3
57
0.5
2.7
4.3
7.0
1.6
8.1
5.4
2.7
11.3
11.3
1.1
1.1
0.5
0.5
1.1
1.1
1.6
0
0
0
0
0
0
0
0
0
0
0
0
57.0
54.3
51.6
1.6
30.6
Median of
Positives
1.1
0.28
0.54
1.1
0.57
1.0
0.32
0.96
1.0
0.52
9.0
2.6
0.74
1.7
0.46
0.59
0.66
-
-
-
-
-
-
-
-
-
-
-
-
1.6
1.6
2.9
1.8
3.8
Maximum
Value
1.1
2.2
1.2
2.0
0.95
3.1
2.8
21
78
5.9
15
2.9
0.74
1.7
0.61
0.91
1.3
-
-
-
-
-
-
-
-
-
-
-
-
300
71
59
4.1
50
29
-------
-------
Table 6
Summary of Multiple Occurrences
Random Sample
Ground Water Supply Survey
December 1980 - December 1981
Supplies Containing Listed Number
Number of Contaminants
0
1
2
3
4
5
6
7
<
No.
233
28
10
6
1
1
0
1
280
Population
10,000
%
83.2
10.0
3.5
2.1
0.4
0.4
0
0.4
100%
Category
>
No.
134
27
8
6
5
3
2
1
^^^M
186
of Contaminants
10,000
%
72.1
14.5
4.3
3.2
2.7
1.6
1.1
0.5
100%
30
-------
-------
Table 7
Occurrence of Purgeable Synthetic Organic Contaminants9
by Population Category and Concentration
Random Sample
Ground Water Supply Survey
December 1980 - December 1981
Number of Supplies With Listed Concentration0
Parameter
tMchloroethylene (0.2)
tetrachloroethylene (0.2)
1,1,1-tMchloroethane (0.2)
c1s and/or trans-
1,2-dlchloroethylene (0.2)
Population
< 100
101 - 500
501 - 1,000
1.001 - 2,500
2,501 - 5,000
5,001 - 10,000
10,001 - 100,000
> 100,000
Total
< 100
101 - 500
501 - 1,000
1,001 - 2,500
2,501 - 5,000
5,001 - 10,000
10,001 - 100,000
> 100,000
Total
< 100
101 - 500
501 - 1.000
1,001 - 2,500
2,501 - 5,000
5,001 - 10,000
10,001 - 100,000
> 100,000
Total
< 100
101 . 500
501 - 1,000
1,001 - 2,500
2,501 - 5,000
5.001 - 10,000
10,001 100,000
> 100,000
Total
Below QL
78
87
28
30
33
15
154
11
436
76
85
28
30
33
15
153
12
432
76
88
27
29
33
15
158
13
439
79
88
28
31
35
16
162
11
450
QL - 5.0
1
1
-
1
2
2
13
2
22
4
2
-
1
3
2
19
1
32
4
-
1
1
2
2
15
-
25
1
-
-
-
1
1
11
2
16
5.1 - 10
.
.
.
.
-
3
-
3
_
-
-
-
-
1
-
1
-
1
-
.
-
-
1
_
-
-
-
-
-
-
-
-
11 - 50
1
_
.
.
1
.
2
-
4
1
-
-
-
-
-
-
1
_
-
-
-
1
-
-
-
1
.
-
-
-
-
-
-
-
-
51 - 100
_
_
_
-
1
-
1
_
.
-
.
-
-
-
-
-
.
-
.
.
-
-
.
_
.
-
-
.
-
-
.
> 100
^
_
.
.
-
-
^
.
-
.
-
-
-
-
-
_
-
-
-
-
-
-
-
-
_
.
-
-
-
-
-
-
-
31
-------
-------
Table 7 (Cont'd.)
Occurrence of Purgeable Synthetic Organic Contaminants8
by Population Category and Concentration
Random Sample
Ground Water Supply Survey
December 1980 - December 1981
Parameter
I,l-d1chloroethane (0.2)
carbon tetrachlorfde (0.2)
I,l-d1chloroethy1ene (0.2)
o+p-xylene (0..2)
Population
< 100
101 - 500
501 - 1,000
1,001 - 2,500
2,501 - 5,000
5,001 - 10,000
10,001 - 100,000
> 100,000
Total
< 100
101 - 500
501 - 1,000
1,001 - 2,500
2,501 - 5,000
5,001 - 10,000
10,001 - 100,000
> 100,000
Total
< 100
101 - 500
501 - 1,000
1,001 - 2,500
2,501 - 5,000
5,001 - 10,000
10.001 - 100,000
> 100.000
Total
< 100
101 - 500
501 - 1,000
1,001 - 2,500
2,501 - 5,000
5,001 - 10,000
10,001 - 100,000
> 100,000
Total
Number
Below Ql QL
78
84
28
30
35
15
166
12
448
80
87
27
29
35
17
163
13
451
79
87
27
31
35
17
169
12
457
78
87
28
30
34
17
171
13
458
of Supplies With Listed Concentration6
- 5.0 5.1 - 10 11 - 50 51 - 100
2
4
...
1 ...
1 ...
2 ...
7
1 ...
18
. ...
1
1
2 ...
1 ...
...
10
.
14 - 1 -
1 ...
1 ...
1
.
1 ...
...
4
1 ...
81-.
2 ...
1 -
...
1 ...
2 ...
-
2
...
8 ...
> 100
_
_
-
-
.
_
-
-
.
.
.
-
.
-
-
-
.
_
.
-
-
-
-
-
-
.
_
.
.
.
-
-
-
.
32
-------
-------
Table 7 (Cont'd.)
Occurrence of Purgeable Synthetic Organic Contaminants9
by Population Category and Concentration
Random Sample
Ground Water Supply Survey
December 1980 - December 1981
Parameter
1,2-dlchloroethane (0.5)
m-xylene (0.2)
1,2-dlchloropropane (0.2)
benzene (0.5)
Population
< 100
101 - 500
501 - 1,000
1,001 - 2,500
2,501 - 5,000
5,001 - 10,000
10.001 - 100,000
> 100.000
Total
< 100
101 - 500
501 - 1,000
1,001 - 2,500
2,501 - 5,000
5,001 . 10,000
10,001 - 100,000
> 100,000
Total
< 100
101 - 500
501 - 1,000
1,001 - 2,500
2,501 - 5,000
5.001 - 10.000
10,001 . 100.000
> 100,000
Total
< 100
101 - 500
501 - 1.000
1,001 - 2,500
2,501 . 5,000
5,001 . 10,000
10,001 . 100,000
> 100,000
Total
Number
Below QL QL
80
88
28
31
36
17
170
13
463
78
87
28
30
34
17
171
13
458
80
87
28
31
36
17
168
13
460
80
88
28
31
36
16
171
13
463
of Supplies With Listed Concentration0
- 5.0 5.1 - 10 11 - 50 51 . 100
- ...
...
...
...
...
...
3 ...
...
3 ...
2 ...
1 ...
...
1
2
...
2 ...
...
8 ...
_ _
1 ...
...
...
...
...
4 - 1
_ - -
5 - 1
. . . -
...
...
...
_ - -
1 ...
1 - 1 -
. - -
2 - 1
> 100
m
_
.
_
_
_
-
-
_
.
.
_
_
-
-
_
.
.
-
.
.
-
-
.
-
.
.
.
-
-
_
33
-------
-------
Table 7 (Cont'd.)
Occurrence of Purgeable Synthetic Organic Contaminants8
by Population Category and Concentration
Random Sample
Ground Water Supply Survey
December 1980 - December 1981
Parameter
toluene (0.5)
p-d1ch1orobenzene (0.5)
vinyl chloride (1.0)
ethyl benzene (0.5)
Population
< 100
101 - 500
501 - 1,000
1,001 - 2,500
2,501 - 5,000
5,001 - 10,000
10,001 - 100,000
> 100,000
Total
< 100
101 - 500
501 - 1,000
1,001 - 2,500 '
2,501 - 5,000
5,001 - 10,000
10,001 - 100,000
> 100,000
Total
< 100
101 - 500
501 - 1,000
1,001 - 2,500
2,501 - 5,000
5,001 - 10.000
10,001 - 100,000
> 100,000
Total
< 100
101 - 500
501 - 1,000
1,001 - 2.500
2,501 - 5,000-
5.001 - 10,000
10,001 - 100,000
> 100,000
Total
Number
Below QL QL
79
87
28
30
36
16
171
13
460
80
87
28
31
36
16
170
13
461
80
88
28
31
36
17
172
13
465
78
88
28
31
36
17
172
13
463
of Supplies With Listed Concentration0
- 5.0 5.1 - 10 11 - 50 51 - 100
1 ...
1
...
1 ...
...
1
2
...
6 ...
- ...
1
...
. ...
1 - - -
3 ...
...
5 - -
- ...
...
...
_
...
.
1 ...
...
1 ...
2
...
...
...
...
.
1
...
3 ...
> 100
.
.
.
_
.
.
-
-
.
_
-
.
_
-
-
_
-
_
.
-
-
_
-
-
.
_
.
.
.
.
-
_
34
-------
-------
Table 7 (Cont'd.)
Occurrence of Purgeable Synthetic Organic Contaminants8
by Population Category and Concentration
Random Sample
Ground Water Supply Survey
December 1980 - December 1981
Parameter Population
bromobenzene (0.5)
< 100
101 - 500
501 - 1,000
1,001 - 2,500
2,501 - 5,000
5,001 - 10,000
10,001 - 100,000
> 100,000
Total
l,2-d1bromo-3-chloropropane (5.0)
< 100
101 - 500
501 - 1,000
1,001 - 2,500
2,501 - 5,000
5,001 - 10,000
10,001 - 100,000
> 100.000
Total
Number
Below QL QL
80
87
26
31
36
17
172
13
462
80
88
28
31
35
17
173
13
465
of Supplies
- 5.0 5.1
_
1
1
-
-
-
1
-
3
_
-
-
-
-
-
-
-
With Listed Concentration0
-10 11 - 50 51 - 100
- -
- _
1
...
-
-
-
...
1
...
...
...
...
1
-
...
...
1
> 100
.
_
-
.
.
-
-
-
_
.
.
.
.
.
-
-
* Does not Include THMs
b 466 supplies
c All concentrations are In ug/1
Number In parenthesis Is the lower quantItatlon limit, QL,
1n ug/1
35
-------
-------
Table 8
Occurrence Data for the Summation of
Purgeable Synthetic Organic Contaminants
by Population Category and Summed Concentration
Random Sample
Ground Water Supply Survey
December 1980 December 1981
Number of Supplies
Population
< 100
101 - 500
501 - 1,000
1,001 - 2,500
2,501 - 5,000
5,001 - 10,000
10,001 - 100,000
> 100,000
Below QLb
70
77
24
26
26
10
123
11
QL - 5.0
9
9
2
4
8
7
38
2
with
5.1 -
0
0
2
0
1
0
5
0
Listed Summed
10 11 - 50
1
2
0
1
0
0
6
0
Concentration
51 - 100
0
0
0
0
1
0
1
0
> 100
0
0
0
0
0
0
0
0
a Summed concentration is summation of all purgeable
synthetic organic contaminants, exclusive of THMs,
in ug/1.
No contaminants found above the
quantitation
limit.
36
-------
-------
Table 9
Summary of Occurrence Data*
Nonrandom Sample
Ground Water Supply Survey
December 1980 - December 1981
Quantltatlon Occurrences
Parameter Limit No. X
vinyl chloride
I,l-d1 chl oroethyl ene
I,l-d1 chl oroethane
c1s and/or trans-
1,2-dlchloroethylene
1 , 2-dl chl oroethane
1,1,1-trlchloroethane
carbon tetrachlorlde
1,2-dlchloropropane
trlchl oroethyl ene
tetrachl oroethyl ene
benzene
toluene
ethyl benzene
bromobenzene
m-xyl ene
o+p-xylene
p-dl chl orobenzene
1,1,2-trl chl oroethane
1,1,1,2-tetrachl oroethane
1 , 1 , 2, 2-tet rachl oroethane
chl orobenzene
1 , 2-d1 bromo-3-chl oropropane
n-nropyl benzene
o-chlorotoluene
p-chlorotoluene
m-d1chl orobenzene
o-d1chlorobenzene
styrene
Isopropyl benzene
chloroform
bromodlchl oromethane
dl bromochl oromethane
d1 chl orol odomethane
bromoform
* All concentrations are In ug/1
b 479 sample sites
1.0
0.2
0.2
0.2
0.5
0.2
0.2
0.2
0.2
0.2
0.5
0.5
0.5
0.5
0.2
0.2
0.5
0.5
0.2
0.5
0.5
5.0
0.5
0.5
0.5
0.5
0.5
0.5
0.5
0.2
0.2
0.5
1.0
1.0
6
15
23
38
7
51
15
7
61
45
8
5
3
2
8
10
4
0
0
0
1
0
1
1
0
0
2
0
0
255
244
222
13
148
1.3
3.1
4.8
7.9
1.5
10.6
3.1
1.5
12.7
9.4
1.7
1.0
0.6
0.4
1.7
2.1
0.8
0
0
0
0.2
0
0.2
0.2
0
0
0.4
0
0
53.2
50.9
46.3
2.7
30.9
Median of
Positives
2.7
0.35
0.63
1.7
2.5
1.0
0.45
1.2
1.4
0.73
1.6
0.73
0.87
0.97
0.38
0.44
0.73
-
-
-
2.7
-
0.98
2.4
-
-
2.4
-
-
1.9
2.1
3.9
1.2
4.2
Maximum
Value
8.4
3.0
4.2
120
9.8
21
15
18
130
69
12
1.5
0.95
1.2
0.83
2.5
0.90
-
-
-
2.7
-
0.98
2.4
-
-
2.7
-
-
430
110
63
4.2
110
37
-------
-------
Table in
Summary of Occurrence Data8
Nonrandom Sample
Supplies Serving Less Than 10,000 Persons
Ground Water Supply Survey
December 1980 - December 1981
Quantltatlon Occurrences
Parameter Limit No. 1
vinyl chloride
1 ,1-dl chl oroethyl ene
1,1-dlchloroethane
els and/or trans-
1,2-dlchloroethylene
l,2-d1chloroethane
1,1,1-tMchloroethane
carbon tetrachlorfde
1 ,2-d1 chl oropropane
trl chl oroethyl ene
tetrachl oroethyl ene
benzene
toluene
ethyl benzene
bromobenzene
m-xylene
o+p-xylene
p-dlchlorobenzene
1,1,2-trlchloroethane
1 ,1 ,1 ,2-tetrachloroethane
1,1,2 ,2-tetrachloroethane
chlorohenzene
1 ,2-d1bromo-3-chl oropropane
n-p ropy 1 benzene
o-chlorotoluene
p-chlorotoluene
m-dlchlorobenzene
o-dl chl orobenzene
styrene
1 sop ropy! benzene
chloroform
bromodl chloromethane
d1 bromochl oromethane
dlchlorolodomethane
bromoform
' All concentrations are In ug/1
b 32! sample sites
1.0
0.2
0.2
0.2
0.5
0.2
0.2
0.2
0.2
0.2
0.5
0.5
0.5
0.5
0.2
0.2
0.5
0.5
0.2
0.5
0.5
5.0
0.5
0.5
0.5
0.5
0.5
0.5
0.5
0.2
0.2
0.5
1.0
1.0
0
5
6
11
3
25
9
3
23
27
5
4
3
2
8
10
4
0
0
0
1
0
1
0
0
0
1
0
0
155
144
135
5
88
0
1.6
1.9
3.4
0.9
7.8
2.8
0.9
7.2
8.4
1.6
1.2
0.9
0.6
2.5
3.1
1.2
0
0
0
0.3
0
0.3
0
0
0
0.3
0
0
48.3
44.9
42.1
1.6
27.4
Median of
Positives
0.35
0.62
1.3
2.9
1.2
0.44
1.2
1.2
0.79
1.6
0.67
0.87
0.97
0.38
0.44
0.74
-
-.
-
2.7
-
0.98
-
-
-
2.2
-
-
1.6
2.0
3.5
1.4
3.7
Maximum
Value
.
3.0
1.2
17
3.4
8.2
15
1.4
29
21
12
0.79
0.95
1.2
0.83
2.5
0.90
-
-
-
2.7
-
0.98
-
-
-
2.2
-
-
100
49
63
4.2
110
38
-------
-------
Table 11
Summary of Occurrence Data'
Nonrandom Sample
Supplies Serving More Than 10,000 Persons'1
Ground Water Supply Survey
December 1980 - December 1981
Parameter
vinyl chloride
1,1-dlchloroethylene
1 , 1-dl chl oroethane
els and/or trans-
1,2-dlchloroethylene
1 ,2-d1 chl oroethane
1,1. 1-trlchl oroethane
carbon tetrachlorlde
l,2-d1chloropropane
trlchloroethylene
tetrachl oroethy 1 ene
benzene
toluene
ethyl benzene
bromobenzene
m-xylene
o*p-xylene
p-dlchlorobenzene
1,1,2-trlchloroethane
1,1,1,2-tetrachloroethane
1,1,2, 2-tetrachl oroethane
chlorobenzene
1 , 2-dl bromo-3-chl oropropane
n-propyl benzene
o-chlorotoluene
p-chlorotoluene
n-dlchl orobenzene
o-dl chl orobenzene
styrene
Isopropyl benzene
chloroform
bronodl chl oromethane
dlbromochl oromthane
dl chl orol odomethane
bromofonn
* All concentrations are In ug/1
b 158 sample sites
Quantltatlon Occurrences
Limit No. X
1.0
0.2
0.2
0.2
0.5
0.2
0.2
0.2
0.2
0.2
0.5
0.5
0.5
0.5
0.2
0.2
- 0.5
0.5
0.2
0.5
0.5
5.0
0.5
0.5
0.5
0.5
0.5
0.5
0.5
0.2
0.2
0.5
1.0
1.0
6
10
17
27
4
26
6
4
38
18
3
1
0
0
0
0
0
0
0
0
0
0
0
1
0
0
1
0
0
100
100
87
8
60
3.8
6.3
10.8
17.1
2.5
16.5
3.8
2.5
24.1
11.4
1.9
0.6
0
0
0
0
0
0
0
0
0
0
0
0.6
0
0
0.6
0
0
63.3
63.3
55.1
5.1
38.0
Median of
Positives
2.7
0.34
0.87
2.7
1.8
0.93
0.70
0.70
1.5
0.66
2.7
1.5
-
-
-
-
-
-
-
-
-
-
-
2.4
-
-
2.7
-
-
2.1
2.2
4.6
1.2
5.1
Maximum
Value
8.4
0.64
4.2
120
9.8
21
9.4
18
130
69
12
1.5
-
-
-
-
-
-
-
-
-
-
-
2.4
-
-
2.7
-
-
430
110
51
4.1
68
39
-------
-------
Table 12
Summary of Multiple Occurrences
Nonrandom Sample
Ground Water Supply Survey
December 1980 - December 1981
Supplies Containing Listed Number
Number of Contaminants
0
1
2
3
4
5
6
7
8
<
No.
249
35
15
11
7
2
0
1
1
321
Population
10,000
%
77.6
10.9
4.7
3.4
2.2
0.6
0
0.3
0.3
100%
Category
>
No.
99
19
14
7
7
7
4
1
0
158
of Contaminants
10,000
%
62.7
12.0
8.9
4.4
4.4
4.4
2.6
0.6
0
100%
40
-------
-------
Table 13
Occurrence of Purgeable Synthetic Organic Contaminants8
by Population Category and Concentration
Nonrandom Sample
Ground Water Supply Survey
December 1980 - December 1981
Parameter
Population
Number of Supplies With Listed Concentration0
Below QL QL - 5.0 5.1 - 10 11 - 50 51 - 100 > 100
trlchloroethylene (0.2)
tetrachloroethylene (0.2)
1,1,1-trlchloroethane (0.2)
< 100
101 - 500
501 - 1,000
1,001 - 2,500
2,501 - 5,000
5.001 - 10,000
10,001 - 100,000
> 100,000
Total
< 100
101 - 500
501 - 1,000
1,001 - 2,500
2,501
5,001
5,000
10,000
10,001 . 100,000
> 100,000
Total
< 100
101 - 500
501 - 1,000
1,001 - 2,500
2,501 . 5,000
5,001 - 10,000
10.001 - 100,000
> 100,000
Total
35
41
33
69
55
65
104
16
418
29
41
31
70
56
67
125
15
434
32
41
32
68
53
70
117
15
428
1
7
8
23
2
43
5
1
3
1
5
5
11
3
34
3
2
2
3
9
5
18
5
47
2
2
1
1
1
2
6
1
11
2
2
els and/or trans-
l,2-d1chloroethylene (0.2)
< 100 35
101 - 500 43
501 - 1,000 33
1,001 - 2,500 70
2.501 - 5,000 60
5,001 - 10,000 69
10,001 - 100,000 114
> 100.000 17
Total 441
1
1
3
5
16
2
28
41
-------
-------
Table 13 (Cont'd.)
Occurrence of Purgeable Synthetic Organic Contaminants9
by Population Category and Concentration
Nonrandom Sample
Ground Water Supply Survey
December 1980 - December 1981
Parameter
I,l-d1ch1oroethane (0.2)
carbon tetrachlorlde (0.2)
I,l-d1chloroethylene (0.2)
o-»-p-xylene (0.2)
Population
< 100
101 - 500
501 - 1.000
1,001 . 2,500
2,501 - 5,000
5,001 - 10,000
10,001 - 100,000
> 100.000
Total
< 100
101 - 500
501 - 1,000
1,001 - 2,500
2,501 - 5,000
5,001 - 10,000 '
10,001 - 100,000
> 100,000
Total
< 100
101 - 500
501 - 1,000
1,001 - 2,500
2,501 - 5,000
5,001 - 10,000
10,001 - 100.000
> 100,000
Total
< 100
101 - 500
501 - 1,000
1,001 - 2,500
2,501 - 5,000
5.001 - 10,000
10,001 - 100,000
> 100,000
Total
Number
Below QL QL
35
43
32
71
60
74
123
18
456
35
43
30
70
61
73
133
19
464
35
43
32
71
61
74
128
20
464
34
43
33
70
59
72
138
20
469
of Supplies With Listed Concentration0
- 5.0 5.1 - 10 11 - 50 51 - 100
- . _ .
-
2
-
3 ...
1
15
2 ...
23 ...
»
* . - -
3 - 1
1 ...
2 ...
2 - -
41--
1 ...
13 1 1
. . . .
...
2
...
2 ...
1 ...
10 ...
- - -
15 ...
1 ...
...
1 ...
1 ...
4 ...
3 ...
_
...
10
> 100
.
_
_
_
.
_
.
.
.
_
.
-
.
-
.
.
_
.
_
.
.
.
.
-
.
_
_
.
_
-
.
-
42
-------
-------
Table 13 (Cont'd.)
Occurrence of Purgeable Synthetic Organic Contaminants9
by Population Category and Concentration
Nonrandom Sample
Ground Water Supply Survey
December 1980 - December 1981
Parameter
bromobenzene (0.5)
chlorobenzene (0.5)
o-d1chlorobenzene (0.5)
n-propyl benzene- (0.5)
Population
< 100
101 - 500
501 - 1,000
1,001 - 2,500
2,501 - 5.000
5.001 - 10,000
10,001 - 100,000
> 100.000
Total
< 100
101 - 500
501 - 1,000
1.001 - 2,500
2,501 - 5,000
5.001 - 10.000
10.001 - 100.000
> 100.000
Total
< 100
101 - 500
501 - 1,000
1.001 - 2,500
2,501 - 5,000
5,001 - 10,000
10,001 - 100.000
> 100,000
Total
< 100
101 - 500
501 - 1.000
1,001 - 2.500
2,501 - 5,000
5,001 - 10,000
10,001 - 100,000
> 100,000
Total
Number
Below QL QL
35
43
33
71
62
75
138
20
477
35
43
34
70
63
75
138
20
478
35
43
34
70
63
75
137
20
477
34
43
34
71
63
75
138
20
478
of Supplies With Listed Concentration0
- 5.0 5.1-10 11 - 50 51 - 100
. . . .
_
1 ...
...
1 ...
...
...
...
2 ...
. ...
...
...
1 ...
...
...
...
...
1 ...
. ...
.
_ ...
1 ...
_
...
1 - . - -
...
2 ...
1 ...
...
...
...
...
-
...
...
1 ...
> 100
^
.
-
-
_
-
.
-
-
.
.
.
-
-
-
-
-
.
.
.
.
-
-
. .
-
-.
.
-
-
.
-
-
.
-
-
43
-------
-------
Table 13 (Cont'd.)
Occurrence of Purgeable Synthetic Organic Contaminants8
by Population Category and Concentration
Nonrandom Sample
Ground Water Supply Survey
December 1980 - December 1981
Parameter
l,2-d1chloroethane (0.5)
m-xylene (0.2)
1,2-dlchloropropane (0.2)
benzene (0.5)
Population
< 100
101 . 500
501 - 1,000
1,001 - 2,500
2,501 - 5.000
5.001 - 10,000
10,001 - 100,000
> 100,000
Total
< 100
101 - 500
501 - 1,000
1,001 - 2,500
2,501 - 5,000
5,001 - 10,000
10,001 - 100,000
> 100,000
Total
< 100
101 - 500
501 . 1,000
1,001 - 2,500
2,501 . 5,000
5,001 - 10,000
10,001 - 100,000
> 100,000
Total
< 100
101 500
501 - 1,000
1,001 - 2,500
2,501 - 5,000
5,001 - 10,000
10,001 - 100,000
> 100,000
Total
Number
Below QL QL
35
43
34
71
61
74
134
20
472
34
42
33
70
61
73
138
20
471
34
43
34
69
63
75
134
20
472
35
43
34
70
59
75
135
20
471
of Supplies With Listed Concentration0
- 5.0 5.1 - 10 11 - 50 51 - 100
-
. - - _
...
...
2
1
3 1 - -
.
6 1
1 ...
1 ...
1 ...
1 ...
2 ...
2 ...
...
...
8 ...
1 ...
.
.
2 -
_ .
.
3 1 -
...
6 - 1
. ...
...
...
1 ...
3 . 1 .
_ ...
2 - 1
...
6 - 2
> 100
.
.
.
-
-
_
.
-
-
.
.
_
.
-
.
.
-
-
.
.
-
.
-
-
.
-
-
.
-
.
.
.
.
-
-
.
44
-------
-------
Table 13 (Cont'd.)
Occurrence of Purgeable Synthetic Organic Contaminants9
by Population Category and Concentration
Nonrandom Sample
Ground Water Supply Survey
December 1980 - December 1981
Parameter
toluene (0.5)
p-d1chlorobenzene (0.5)
vinyl chloride (1.0)
ethyl benzene {0.5)
Population
< 100
101 - 500
501 - 1,000
1,001 - 2,500
2,501 - 5.000
5,001 - 10,000
10,001 - 100,000
> 100,000
Total
< 100
101 - 500
501 - 1,000
1,001 - 2,500
2,501 - 5,000
5,001 - 10,000
10,001 - 100,000
> 100,000
Total
< 100
101 - 500
501 - 1,000
1,001 - 2.500
2,501 - 5,000
5,001 - 10,000
10,001 - 100,000
> 100,000
Total
< 100
101 - 500
501 - 1.000
1,001 - 2.500
2.501 - 5,000
5,001 - 10,000
10,001 - 100,000
> 100,000
Total
Number
Below QL QL
35
43
34
70
62
73
137
20
474
35
43
34
70
61
74
138
20
475
35
43
34
71
63
75
132
20
473
35
42
34
70
63
74
138
20
476
of Supplies With Listed Concentration0
- 5.0 5.1 - 10 11 - 50 51 - 100
. . _ _
- -
-
1
1
2
1 ...
...
5 ...
- ...
. ...
...
1 - - -
2 ---
1 ...
...
...
4 ...
...
...
...
.
...
...
3 3
-
33--
. ...
1 ...
...
1 ...
- -
1 ...
...
...
3 ...
> 100
.
_
.
.
.
-
-
-
_
_
.
-
_
.
-
-
-
.
_
-
-
.
.
-
-
-
.
.
-
-
.
-
-
-
.
45
-------
-------
Table 13 (Cont'd.)
Occurrence of Purgeable Synthetic Organic Contaminants9
by Population Category and Concentration
Nonrandom Sample
Ground Water Supply Survey
December 1980 - December 1981
Parameter Population
o-chlorotoluene (0.5)
< 100
101 . 500
501 - 1,000
1,001 - 2,500
2,501 - 5,000
5,001 - 10,000
10,001 - 100,000
> 100,000
Total
Number
Below QL QL
35
43
34
71
63
75
137
20
478
of Supplies With Listed Concentration0
- 5.0 5.1 - 10 11 - 50 51 - 100
- - _
-
- - _ .
...
...
...
1 ...
...
1 ...
> 100
_
_
_
.
.
-
-
a Does not Include THMs
b 479 supplies
c All concentrations are In ug/1
d Number 1n parenthesis 1s the lower quant 1tat1 on Hm1t, QL,
1n ug/1
-------
-------
Table 14
Occurrence Data for the Summation of
Purgeable Synthetic Organic Contaminants
by Population Category and Summed Concentration
Nonrandom Sample
Ground Water Supply Survey
December 1980 - December 1981
Number of Supplies with Listed Summed Concentration
Population
< 100
101 - 500
501 - 1,000
1,001 - 2,500
2,501 - 5,000
5,001 - 10,000
10,001 - 100,000
> 100,000
Below QLb
24
38
27
62
43
55
85
14
QL - 5.0 5.1
10
4
5
8
18
12
29
2
a Summed concentration is summation of all purgeabl
synthetic organic contaminants, exclusive of THMs
in ug/1.
No contaminants found above the
- 10
1
0
0
0
0
4
6
0
e
»
11 - 50
0
1
2
1
2
4
13
3
51 - 100
0
0
0
0
0
0
3
0
> 100
0
0
0
0
0
0
2
1
quantitation limit.
47
-------
-------
Table 15
Volatile Organic Contaminants
Found in Original and Resample
City A
Parameter June 1981 March 1982
1,1-dichloroethane 0.62 0.51
1,1,1-trichloroethane 1.9 1.4
trichloroethylene 0.21 < 0.2
tetrachloroethylene 1.3 0.94
All concentrations ug/1
48
-------
-------
8
o
o o
KIT:
MM «M no VOC
Ml* «Nh M toMt «w
e
e o
e o
e
, 0"
LOCATION OF
RANDOM SAMPLING SITES
GROUND WATER SUPPLY SURVEY
I 1M1)
Fiqure 1
-------
-------
in
o
oo o
Oo
r
KIT:
t
M* with M voc
8
o
o .
LOCATION OF
NONRANOOM SAMPLING SITES
GROUND WATER SUPPLY SURVEY
1MO - DCCIMMH 1H1I
0
c
Figure 2
------- |