at higher concentrations in sewage sludges than aldrin and endrin,. In order to investigate the temporal variation of these compounds in sewage sludge, three ...
OCCURRENCE ORGANOCHLORINE
AND DISTRIBUTION COMPOUNDS
OF P E R S I S T E N T
IN U.K. S E W A G E S L U D G E S
A, E. M c I N T Y R E and J. N. L E S T E R
Public Health Engineering Laboratory, Civil Engineering Department, Imperial College, London SW7 2BU, U.K.
(Received November 8, 1983; revised February 21, 1984) Abstract. A total of 444 sewage sludges, sampled from sewage treatment works in the U.K., were analyzed for polychlorinated biphenyls (PCB) and the organochiorine insecticides v-Hexachlorocyclohexane (7-HCH), aldrin, dieldrin and endrin. PCB, 7-HCH and dieldrin were found to occur more frequently and at higher concentrations in sewage sludges than aldrin and endrin, In order to investigate the temporal variation of these compounds in sewage sludge, three sewage treatment works were sampled at monthly intervals over an 18 mo period. PCB, 7-HCH and dieldrin were again the most frequently detected compounds. Concentrations of aldrin and endrin were only sporadically detected. The results of a serial correlation analysis of these data indicated that some short term correlation of PCB and dieldrin concentrations above and below the mean value was evident. The data acquired in the course of this survey have been compared with the results of similar work conducted previously in the U.K. and other countries. The possible implications of the presence of persistent organochlorine compounds in sewage sludges which are disposed to the environment are discussed.
I. Introduction
Concern over the public health implications of organic micropollutants in the hydrological cycle has prompted the detailed study of particular classes of compounds in recent years. The sources, occurrence, behaviour and removal of such contaminants in the water cycle have been reviewed recently (Bedding et al., 1982; Bedding et al., 1983a), as have possible health effects and legislation (Bedding et al., 1983b) and analytical methodology (McIntyre and Lester, 1983). Polychlorinated biphenyls (PCB) and organochlorine insecticides (OCL) constitute two classes of compounds which are of prime concern, due to their extreme persistence in the environment and potential for biomagnification (Hutzinger et al., 1974). Although the production and usage of PCB and certain OCL have been severely limited in some developed countries, residues of the more persistent compounds remain in the environment as a legacy of their past use. Of the large number of insecticidal compounds produced, v-HCH aldrin, dieldrin, endrin and components of the DDT group have been those most frequently detected in environmental samples (Babish et al., 1981) although residues of Dichorodiphenyltrichloroethane (DDT) and its analogues have declined in recent years as a result of legislative proscription of their use (United States Environmental Protection Agency, 1971). The occurrence of PCB in waste waters may arise from a number of sources, both industrial and domestic (Lester, 1983), including aqueous effluents from transformer and capacitor manufacturing (United States Environmental Protection Agency, 1976a), investment casting plants (Stratton and Sosebee, 1976) and waste paper recycling Wasp, Air, and Soil Pollution 23 (1984) 397-415. © 1984 by D. Reidel Publishing Company.
0049-6979/84/0234-0397502.85.
398
A . E . MCINTYRE AND J. N. LESTER
operations (United States Environmental Protection Agency, 1976b). In addition, relatively low concentrations of PCB have also been detected in domestic waste waters in the U.S.A. (Schmidt et al., 1976), Canada (Shannon etal., 1976; Lawrence and Tosine, 1977) and the U.K. (McIntyre et al., 1981a). Owing to the hydrophobic nature of PCB and OCL, their residues in waste waters will be associated almost exclusively with solid material and will therefore be removed to a significant degree during primary and secondary sedimentation processes in sewage treatment, concentrating in sewage sludges (Garcia-Gutierrez et al., 1982; Lester, 1982). In the U.K., approximately 1.24 x 106 tonne of dry sewage solids are disposed per annum; of this quantity, approximately 74 ~o is disposed to land, 23 ~o is dumped at sea and the remaining 3 ~ is incinerated (Government of Great Britain, 1978). Of the proportion disposed to land, 56~/o is utilized in agriculture and therefore a total of approximately 41.1 ~o (some 0.51 x 106 tonne) is disposed to farmland and the oceans. The use of sewage sludge on agricultural land is considered to be beneficial in terms of the return of organic matter and inorganic nutrients to the soil (Government of Great Britain, 1970, 1977). Dumping of sewage sludge at sea is also an important disposal option in economic terms where it exists as the only viable alternative for large sewage treatment works situated near the coast, when the costs of transporting large volumes of liquid sludge by road are considered (Government of Great Britain, 1978). It is apparent that an important route for contamination of food chains is disposal of sludge to pasture land (Lindsay, 1982) and subsequent uptake by grazing animals (Healy, 1968). In a limited number of cases, this has resulted in elevated concentrations in cow's milk (Bergh and Peoples, 1977; De Haan, 1977; Lindsay, 1979), physiological damage in sheep which were fed cabbage grown on PCB-contaminated sludge/soil mixtures (Haschek et al., 1979) and accumulation of organochlorine residues in beef cattle (Fries, 1977). Although organochlorine residues may be taken up by crop plants from soil (Lichtenstein, 1959; Nash, 1968; Lawrence and Tosine, 1977), the greater proportion will be tightly bound to humic materials (Beall and Nash, 1971; Jacobs et al., 1976) and will therefore be largely unavailable. Sea disposal of contaminated sludge has been shown to result in elevated concentrations of PCB in sediments and benthic fauna (Halcrow et al., 1974) and in certain fish (Waddington et aI., 1973). If marine disposal of sludge is carried out in areas where dispersion and dilution are rapid, few or no significant deleterious effects may be detected. To date, the only legislation designed to control the disposal of sludges contaminated with organic micropollutants has been promulgated in the U.S.A. (United States Environmental Protection Agency, 1979), where a limit of 10 mg kg-~ for PCB in sewage sludge applied to arable land used for growing food-chain crops has been established. In the U.K., no legislation or guidelines currently exist, although a proposal for a Council Directive on the use of sewage sludge in agriculture has recently been promulgated within the European Economic Community (Commission of the European Communities, 1982).
ORGANOCHLORINES IN SEWAGE SLUDGE
399
With the continuation of present practices, monitoring of organic contaminant concentrations in sewage sludges may be desirable both to provide information on the dissemination of micropollutants in the environment and to achieve a degree of protection against the accumulation of such residues to potentially toxic levels in soils, marine sediments and fauna, with further potential for biomagnification.
2. Materials and Methods
2.1. SAMPLING The sampling programme was designed to obtain approximately 450 grab samples of sewage sludge in the 'as disposed' condition from U.K. treatment plants, reflecting the size distribution of plants within individual water authority areas, the geographical area of water authorities and the range of normal sludge treatment procedures currently employed in the U.K. Hence, the samples obtained varied from raw primary and humus sludges collected from small treatment plants in rural areas serving < 1000 population, to chemically-conditioned dewatered and anaerobically digested sludges from treatment plants located in industrialized urban areas serving > 1 x 10 6 population. In addition to the main survey, samples of sewage sludge were collected on a monthly basis from three sewage treatment plants over an 18 mo period, in order to investigate the variation with time in the concentrations of persistent organochlorine compounds. Sewage works A incorporated primary sedimentation with activated sludge secondary biological treatment and anaerobic digestion of the mixed primary/surplus activated sludge. Total population served is approximately 2.4 x 10 6 with an average daily flow of 1137 ML and an industrial effluent flow contribution of 12.5 70. Sewage works B consisted of primary sedimentation and percolating filter secondary treatment with chemical conditioning and filter-press dewatering of the mixed primary/humus sludge. Total population served is 13 000, with an average daily flow of 2.73 ML and an industrial effluent flow contribution of 13~o. Sewage works C incorporated primary sedimentation with activated sludge secondary treatment and anerobic digestion of the mixed primary/surplus activated sludge. Total population served is 49 000 with an average daily flow of 34 ML and an industrial effluent flow contribution of 18 ~o. 2.2. ANALYSIS OF SEWAGE SLUDGES Concentrations of PCB, ?-H.CH, aldrin, dieldrin and endrin in sewage sludge samples were determined using previously developed methods (McIntyre et al., 1980; GarciaGutierrez et al., 1982). The analytical procedure involved homogenisation of the sample, extraction of an aliquot with hexane, clean-up of the extract by alumina/silver nitrate and silica gel column chromatography and analysis by packed column electron-capture gas chromatography. Where samples were received in dried cake form, these were prepared for analysis by the method of McIntyre el aL (1981b). The detection limit of
400
A.E. MCINTYRE AND J. N. LESTER
the method is 0.01 lag g - 1 for each component with typical relative standard deviations of 2.5 to 6.2~o. 2.3. STATISTICAL TREATMENT OF RESULTS Mean, median and percentile concentrations were calculated for the five compounds, together with standard deviations. Log-normal distribution plots were also constructed. In order to extract further information from the accumulated data, the results for each determinand were categorised according to sludge type and mode of treatment (if any) and analysis of variance was conducted between the categories. Correlation analyses were also carried out to detect any relationship between determinand concentrations and population size served or trade effluent received by the sewage treatment works sampled. Results from the analysis of sludges from three sewage treatment works sampled over an 18 mo period were subjected to a serial correlation analysis, as described by Chatfield (1975).
3. Results 3.1. DISTRIBUTION
OF PCB AND O C L CONCENTRATIONS IN UK SEWAGE SLUDGES
A statistical summary of the results from the analysis of sewage sludge samples from 444 U K treatment plants is included in Table I. TABLE I Statistical summary of concentrations of persistent organochlorine compounds in U.K. sewage sludges Compound
PCB ?-HCH Aldrin Dieldrin Endrin
Percentiles, gg g - 1 Mean conc. ggg-i
Standard deviation ggg-1
5
50 median
95
Range gg g 1
0.34 0.41 0.03 0.50 0.11
1.23 3.58 0.03 3.07 0.16
0.03 0.01 0.01 0.02 0.01
0.14 0.09 0.02 0.13 0.06
0.95 0.67 0.09 0.97 0.30
0.01-21.5 0.01-70.00 0.01-00.21 0.01-52.94 0.01-00.71
PCB concentrations exhibited a mean value of 0.34 gg g - 1 a standard deviation of 1.23 gg g - 1 and a 50th percentile value (median) of 0.14 gg g - 1. The data were not normally distributed, as confirmed by both the mean : median ratio (2.4) and by the observation that the majority of the distribution did not lie between + 3 on of the mean value. 7-HCH concentrations varied between 0.01 and 70.0 gg g - 1 with a mean value of 0.41 gg g - 1 , a standard deviation (on) of 3.58 gg g - 1 and a 50th percentile value (median) of 0.09 lag g - 1. The mean : median ratio of 4.6 indicates that the data were
ORGANOCHLORINES IN SEWAGE SLUDGE
401
abnormally distributed. Aldrin concentrations in the 444 sludge samples were generally lower than the other compounds, with a mean of 0.03 lag g - ~, a standard deviation of 0.03 lag g - 1 and a 50th percentile value (median) of 0.01 lag g - 1. The limited range in the concentrations measured (0.01 to 0.21 lag g - ~) and the relatively large number of non-detectable values precluded a comprehensive appreciation of the statistical distribution of the results. Concentrations of dieldrin varied between 0.01 and 52.94 lag g - 1 with a mean value of 0.50 lagg-1, a standard deviation of 3.07 lagg-1 and a 50th percentile (median) of 0.13 lag g - l . The m e a n : m e d i a n ratio (3.8) indicated that the data did not approximate to a normal distribution. Endrin concentrations measured in the sludge samples lay in the more limited range 0.01 to 1.71 lag g - 1, with a mean value of0.11 lag g - ~, a standard deviation of 0.16 lag g - 1 and a 50th percentile concentration (median) of 0.06 lag g - 1. The data are not normally distributed, as is confirmed by the mean : median ratio of 1.8. In an attempt to define more closely the form of the results, data from analyses for the five determinands were plotted on lognormal probability graphs which are illustrated in Figures 1 to 5. It is evident that the data approximated tolerably well to straight lines on these plots and it therefore appeared that the results of this survey were best described by the log-normal distribution form. 3.2. A N A L Y S I S OF VARIANCE BETWEEN SEWAGE S L U D G E TYPES
In order to evaluate any possible differences in the concentrations of persistent organochlorine substances between the various types of sludges sampled (whether primary, mixed primary/secondary biological sludges or digested sludges), the data acquired from analysis of the 444 sludge samples were divided on the basis of the three aforementioned sludge types and analysis of variance was conducted for each determinand. No significant differences were observed between the means, at the 95~o significance level, of any of the five determinands for the three sludge types, with the exception of dieldrin concentrations in digested sludges. This result was found subsequently to be caused by three particularly elevated concentrations recorded for digested sludges obtained from treatment works known to receive trade effluents containing dieldrin. Since the difference between the means could not therefore be attributed to sludge treatment factors, the test was repeated with the exclusion from the calculations of the three elevated values and a negative result was obtained. 3.3.
C O R R E L A T I O N ANALYSIS OF
PCB
AND O C L
CONCENTRATIONS IN U.K.
SEWAGE SLUDGES
Simple linear correlation analyses between determinand concentration, relative proportions of trade effluent received and the population served by the respective 444 sewage treatment works sampled were carried out for each determinand. Little or no correlation was observed between any of the compounds considered and either of the two sewage parameters; typical examples of correlations obtained for PCB, 7-HCH, aldrin, dieldrin and endrin with trade effluent received and population served are included in Table II. The correlation coefficients (r) vary between - 0.021 and + 0. I80,
402
A. E. M C I N T Y R E A N D J. N. L E S T E R
10"0
PC,~, ugg-1
1.0
•
II
•'I
•
; t i 0 It 0.1
f-
i~ lw A
v
A W
0.0'~
20
50
00
98
99
percentiles Fig. 1. Log-normal probability plot of PCB concentrations in U.K. sewage sludges.
ORGANOCHLORINES
~y/
IN
SEWAGE
403
SLUDGE
10-0
--HCH ug g-1
I~ 1.0
It •
A v
0110 0.1
I~ A v
A v
A w
,~
0-0'
A v
20
50
90
95
98
percent i les Fig. 2. Log-normal probability plot of ~-HCH concentrations in U.K. sewage sludges.
404
A, E. MCINTYRE AND J. N. LESTER
10"0
aldrin ug g-1
1.0
¢* 1) 0-1
A W
A~F~
V ~
~A~,
~qW
..~.-.; F ~
0-01
•
== ~ ' ~ J ' "
20
• IF1
50
90
percentiles
Fig. 3. Log-normal probability plot of aldrin concentrations in U.K. sewage sludges.
ORGAN•CHLORINES
1N S E W A G E
SLUDGE
40'5
10"0 dieldrin ugg-1
1.0
)
O
~D • •
~
•
• 0-1
A~ w
~
~ ~
0"01
20
~
50
90
98
percentiles
99
Fig. 4. Log-normal probability plot of dieldrin concentrations in U.K. sewage sludges.
406
,~. E. M C I N T Y R E
A N D J. N, L E S T E R
10"0
endrin ug g-l,
1.0
~D •
~D
el'
A W
0.1
A v
A w
....
Adlt, w'qv
~,A , , , ~
0"01
L
20
50
90
95
98
99
percent i les Fig. 5. Log-normal probability plot of endrin concentrations in U.K. sewage sludges.
ORGANOCHLORINES IN SEWAGE SLUDGE
407
TABLE II Correlation analysis of determinands with industrial effluent received and population served Compound
Correlation coefficients (r)
PCB 7-HCH Aldrin Dieldrin Endrin
Industrial effluent
Population
+ +
- 0.013 - 0.018 + 0.002 - 0.017 + 0.176
0.010 0.013 0.021 0.029 0.180
indicating that little or no correlation existed with, perhaps, the exception of endrin, where positive correlations of + 0.180 and + 0.176 were obtained for both industrial effluent received and population served. The reasons for these correlations are unclear at present.
1-o0"750,50'25-
ugg-1
~
HCH
0.20-150,10,05"
0,75-
0.5-
0,25-
0 a
m
j
j
a 198t
s
o
n
I
J
f
m
a
m
j
j
a
1983.
Fig. 6. Monthly variation in concentrations of PCB, y-HCH and dieldrin in sludges sampled from sewage works A.
408
A . E . MCINTYRE AND J. N. LESTER
3.3.1. Variation of Persistent Organochlorine Concentrations in Sludges from Three
Sewage Treatment Works Concentrations of PCB, 7-HCH, aldrin, dieldrin and endrin measured in 18 discrete monthly samples from each of these sewage treatments works are summarised in Figures 6, 7, and 8. PCB concentrations in works A sludges varied between 0.11 ~g g - ~ and 0.93 pg g - 1 over the sampling period with a mean concentration of 0.26 ~tg g - 1. Concentrations of 7-HCH varied between 0.01 and 0.21 ~tg g - 1 over the sampling period with a mean value of 0.10 pg g - ~ . Aldrin was only detected in 6 samples at concentrations of 0.01 to 0.04 ~tg g - 1. Dieldrin concentrations varied between 0.21 and 0.91 ~tgg-1 with a mean of 0.60 ~tg g - 1 . Endrin was only detected in one sample (October 1981) at a concentration of 0.02 ~tg g - 1. Concentrations of PCB in the sludges sampled from sewage treatment works B varied between 0.03 and 0.45 pg g - i with a mean value of 0.12 pg g - 1 and were non-detectable on five occasions, v - H C H concentrations lay in the range 0.02 to 0.61 ~tg g - 1 with a mean of 0.20 ~tg g - 1. Aldrin was only detected on five occasions, at concentrations of 0.01 to 0.02 ~g g - 1. Dieldrin concentrations varied between 0.01 and 0.31 pg g - 1 with a mean value of 0.08 ~tg g - 1 over the sampling period. Endrin was not detected on seven occasions and concentrations varied between 0.01 and 0.19 ~tg g - 1, with a mean value of 0.09 ~tg g - 1, over the remaining 11 samples.
0S .N PCB 0.~5-
0.25~0-3-
/
~
.
._
j
f
.~--~---4~.
~
DIELDRIN
~.1-
m
i
J
a
s
o
n
]
198t ~i~. 7. ~o~tbl~ ~afi~fio~ i~ ~ o ~ e ~ t r ~ t i o ~ of ~ B , ~ - ~ wo~k~ B.
m
a
m
j
j
a
1982 ~ d di~]dd~ i~ sl~d~es samp]~d from s~wa~e
ORGANOCHLORINES INSEWAGESLUDGE
409
In sludges from sewage treatment works C, PCB concentrations varied between 0.04 and 0.51 ~tg g - 1with a mean value of 0.22 lag g - 1 and one non-detectable result in May 1982. 7-HCH concentrations were non-detectable on two occasions and, in the remaining 16 samples, varied between 0.01 and 0.23 ~tg g - 1 with a mean concentration of 0.06 lag g 1. Aldrin concentrations were non-detectable on 13 occasions and varied between 0.01 and 0.04 lag g-1 in the remaining 5 samples. Concentrations of dieldrin lay in the range 0.02 to 0.30 p,g g - 1 with a mean of 0.14 ~tg g - ~ and one non-detectable result. Endrin was only detected on seven occasions and concentrations varied between 0.01 and 1.17 lag g - 1 with a mean value of 0.25 pg g - 1. Further examination of the results reveals possible trends in the data for some determinands and over the eighteen month sampling period. In works A sewage sludge samples distinct peaks in concentration are evident for ?-HCH over the sampling period, around August to November 1981 and over May to June 1982. A broadly similar trend is also apparent for dieldrin, where peaks in concentration occur around September to November 1981 and March
0'5-0,4_~
~
~
PCB
0.2-
0.t-
~
ugg-1
"~/-HCH
o.1-
_.J o.3-
DIELDRIN
0,2-
04-
a
m
j 1981
j
a
s
o
n
I
J
f
m
a
m
l
J
a
1982
Fig.8. Monthlyvariationin concentrationsofPCB,~-HCHand dieldrinin sludgessampledfromsewage worksC.
410
A. E. MCINTYRE AND J. N. LESTER
1982. No apparent trend exists for concentrations of PCB. In works B sewage sludges, 7-HCH and dieldrin concentrations again exhibit similar trends, with elevated concentrations in the period July to December 1981. PCB concentrations displayed a small increase over this period and a well defined peak over the period May to September 1982, when no similar increases were observed for 7-HCH and dieldrin. 7-HCH concentrations in the sewage sludges sampled from works C exhibited two distinct peaks over June to October 1981 and April to July 1982. A minor peak in dieldrin concentrations was also observed over June to September 1981. Elevated PCB concentrations also occurred in May 1981 and October 1981, during the same overall period of elevated 7-HCH and dieldrin concentrations. In order to elucidate any underlying trends in the monthly data from analysis of sewage sludges taken from these three locations, a serial correlation (autocorrelation) analysis was conducted, according to the method described by Chatfield (1975). It is possible, from the shape and form of the correlograms, to assess the nature of the variations inherent in the raw data. However, unless these correlograms conform closely to a small number of pre-established forms, interpretation of the underlying trends in the time series data is difficult. For the samples obtained from the three sewage treatment works over the 18 mo period, correlograms were derived only for PCB, 7-HCH and dieldrin, since low and non-detectable concentrations of aldrin and endrin were observed frequently. PCB concentrations in sewage sludges from works A exhibited a random time series distribution, with autocorrelation coefficients (rk). A similar pattern was also obtained for 7-HCH concentrations, where little or no correlation between successive values at different time lags was evident. Dieldrin concentrations displayed a somewhat different correlogram, the form of which suggested a degree of short-term correlation, i.e. one for which an observation above the mean tends to be followed by one or more further observations above the mean, and similarly for observations below the mean. This may also be termed as an autoregressive time series. For the works B samples, both the PCB and dieldrin concentrations exhibit a degree of short term correlation, as did those of 7-HCH, although to a lesser degree. Concentrations of PCB, 7-HCH and dieldrin in the samples from works C approximated closely to a random series, with the autocorrelation coefficients of all these determinands at different time lags tending to zero. Although some general indication were present as to the trends in the time series data for the sewage sludge samples from the three works, insufficient raw data points were available to achieve a comprehensive analysis. Since only 5 time lags may be properly extracted from the available data, the power of the statistical analysis was substantially reduced. 4. Discussion Polychlorinated biphenyls and the organochlorine insecticides 7-HCH and dieldrin have been found to be common micro-contaminants of U.K. sewage sludge. Aldrin and endrin were observed to occur less frequently in sewage sludges and at concentrations
ORGANOCHLORINES IN SEWAGE S L U D G E
411
close to the detection limit of the analytical method, which may be explained by the limited use of these two compounds in the U.K. In the case of aldrin, the reported epoxidation of this compound to dieldrin in aquatic media may also explain the low concentrations observed (Moriarty, 1975). Although information pertaining to concentrations of persistent organochlorine compounds in sewage sludges is limited both in terms of the number of surveys conducted and numbers of samples analyzed, it is of value here to compare results of the present survey with those of previous exercises carried out in the U.K. and other countries. Table III includes a summary of relevant data arising from previous work. Mean concentrations of PCB measured during this survey are broadly comparable with earlier results obtained in the U.K. in 1979 (McIntyre and Lester, 1982) and 1972-1974 (Water Research Centre, 1973; Water Research Centre, 1974). 7-HCH and dieldrin concentrations recorded in the 1979 U.K. work were approximately one-half those detected in this survey, although this may be accounted for by the relatively small number of samples analyzed. The results of an analytical survey conducted during 1978 in the Netherlands (Fieggen, 1981) exhibited lower mean concentrations of both 7-HCH and dieldrin, but higher average concentrations Of PCB. A survey in Sweden of 146 sewage sludges in 1971 (De Haan, 1976) reported lower concentrations of 7-HCH and significantly higher mean concentrations of dieldrin and PCB. Concentrations of PCB and OCL reported from minor surveys in Canada (Shannon etal., 1976), Japan (Kowase etal., 1977) and the U.S.A. (Furr etal., 1976; Babish etal., 1981) varied considerably in the range 0.02 to 16.04 ~tg g- 1 for individual compounds and therefore results from all these surveys appear to be broadly comparable. The results of the 18 mo survey of sewage sludges from three selected sewage treatment works have established typical variations in the concentrations of PCB and OCL with time. The serial correlation analyses of these data have been largely inconclusive owing to the relatively small number of samples taken. There are, however, some indications that short-term correlations existed for certain compounds, particularly for PCB and dieldrin in the works B samples. In these cases, the occurrence of consecutive values above or below the mean value demonstrated that this was not due to random variation, but to some definite event, perhaps associated with a discharge of residues from industrial usage to the sewers. The statistical distribution of PCB and OCL has been found to approximate to the log-normal distribution form principally because of the relatively large number of low concentrations observed, which were also influenced by the necessary imposition of an analytical detection limit, and the small number of high concentrations encountered. No major differences in concentrations were evident between the various geographical regions in the U.K., with the exception of rural areas, where consistently low concentrations of organochlorine compounds were observed. On a few occasions, particularly high concentrations of PCB and OCL were observed, notably those of dieldrin in several samples and sporadic occurrences of relatively high concentrations of PCB and 7-HCH. In a number of cases, the occurrence of these elevated concentrations may be related to known industrial uses of certain compounds (such as dieldrin usage for moth-proofing
412
A. E. MCINTYRE AND J. N. LESTER
~ ~ ~
~~ - ~~
~
~
~
~
~
~~
~
~
~
~
