Document 6wa0YjVDwGzQmwK6wpKwnpXmo

CHLORINATED HYDROCARBONS IN THE LAKE ONTARIO ECOSYSTEM By Clarence L. Haile Water Chemistry Program University of Wisconsin Madison, .Wisconsin 53706 Principal Investigators Gilman D. Veith National Water Quality Laboratory Duluth, Minnesota 55804 April 1. 1972 to August 10, 1972 . ' G. Fred Lee Institute for Environmental Sciences University of Texas-Dallas Richardson, Texas 75080 August 10, 1972 to August 10, 1973 William C. Boyle Department of Civil ar.d Environmental Engineering University of Wisconsin Madison, Wisconsin 53706 August 10, 1973 to June 30, 1974 Project Officer Michael 0. Mull in U.S. Environmental Protection Agency Grosse He Laboratory Grosse He, Michigan 48138 Prepared for Office of Research and Monitoring U.S. Environmental Protection Agency : Washington, D.C. 20460 HONS 067595 ABSTRACT Lake Ontario fish, water, sediment, net plankton, Cladophora, and benthos were examined for DDT group pesticides, dieldrin, and PCBs. Endrin, BliC group pesticides, and heptachlor were also identified in some fish samples. Average concentrations ranged from 23 ng/1 (t-DDT), 4.8 ng/1 (dieldrin), and 55 ng/1 (PC8s as Aroclor 1254 equivalent) for water to 1.40 ug/g (t-DDT), 0.07 ug/g (dieldrin), and 5.15 ug/g (PC3s) for whole fish. DDE levels were generally similar to t-DDT levels, except for sediments where DDD and DDT contributed significantly to t-DDT values. PCB/t-DDT ratios averaged 2.6 for all samples except for sediment (7.0) and benthos (5.3). This report was submitted in fulfillment of Project Number R-800CC8, by the University of Wisconsin, under the (partial) sponsorship of the Environmental Protection Agency. Work was completed as of August, 1974. 1i HONS 067596 Abstract List of Tables Acknowledgements CONTENTS . Sections ' I Conclusions . II Recommendations III Introduction IV Methods V Results and Discussion VI References Page .H . . iv v 1 -3 , '4 ' t 5 13 . 28 tti NONS 067597 TABLES .. Gas Chromatographic Columns and Conditions. Chlorinated Hydrocarbons and Fat in Lake Ontario Fish DDT and PCBs In Lake Ontario Fish Fat. Chlorinated Hydrocarbons inLake Ontario Hater. Chlorinated Hydrocarbons inLake Ontario Sediment. Chlorinated Hydrocarbons inLake Ontario Net Plankton. Chlorinated Hydrocarbons in Lake Ontario Cladophora. Chlorinated Hydrocarbons inLake Ontario Benthic Fauna. Page 10 14 16 19 21 23 2526 1v HONS 0675S8 ACKNOWLEDGEMENTS Ke wish to thank John Carr, Nelson Thoir.as, Sara Mozley, Don McNaught, and the officers and crews of the Researcher. Advance II. and Kaho for their assistance in sample collection. Technical assistance was provided by Richard Pyter, Walter Gunthier, Katherine Benkert, Judy Capelli and Paul Choitz. We are especially grateful to David Armstrong for his review and comments on the manuscript. v MCNS 06759$ SECTION I CONCLUSIONS Lake Ontario fish, water, sediment, r.et plankton, Cladophora. and benthic fauna contained significant concentrations of DOT group pesticides, dieldrin, and PCBs. Endrin, heptachlor, and BHC group pesticides (especially lindane, y BHC) were also identified in sore fish. <( ' Lake-wide concentrations of t-DDT, dieldrin, and PCBs(Aroclor 1254 equivalent) for fish (alev/Ives, smelt, and slimy sculpin) ranged from 0.35 to 1.40 ug/g, 0.C4 to 0.07 ug/g, and 2.35 to 5.13 ug/g, respectively. Lake Ontario water was found to contain "total" concentrations (dissolved + particulate) of 28 r.g/1, 4.8 ng/1, and 55 ng/1 for t-007, dieldrin, and PC3s. Water collected off Oswego contained comparatively high levels of DDT group pesticides, dieldrin, and PCBs, while waters off Hamilton contained higher t-DDT levels, and waters off the mouth of the Niagara River showed higher PCB concentrations. Average sediment t-DDT, dieldrin, and PCB concentrations were 22, 1.2, ana 120 ng/g, respectively. Sediment off the mouth of the Welland Canal showed higher levels of all three contaminants while sediments off the mouth of the Niagara River contained higher levels of PCBs and dieldrin. Sediments off Oswego and at an eastern mid-lake site shewed higher levels of PCBs and dieldrin, respectively. Average concentrations in net plankton v/ere 3.5 ug/g (t-DDT), 0.12 ug/g (dieldrin), and 7.2 ug/g (PCBs). Corresponding concentrations in Ck~P"h3ra were 229, 13,and 515 ng/g for t-DDT, dieldrin, and PCBs, respectively. Lake Ontario benthic fauna v/ere found to contain 99. L.S, and 471 ng/g t-DDT, dieldrin, and PCBs, respectively, with tenth:s taken off Hamilton exhibiting levels approximately four times benthos off Rochester and Oswego. .' High concentrations of PCBs in waters and sediments off the mouth of the Niagara River and Oswego indicate the importance of the Niagara and Oswego Rivers as Inputs of PCBs associated with settlable particulates. i HOWS 067600 In most cases, t-DDT concentrations were similar to concentrations of the ODT metabolite, DOE, except in sediments where DDT and ODD contributed much larger fractions. FCfi/t-DDT ratios for all samples fell in the range of 1.9 to 3.1 except for sediment (7.0) and benthos (5.3). MONS 067601 2 '' SECTION II . RECOWiENCATIONS '' ' . Although Lake Ontario fish contained accumulated chlorinated hydrocarbons at levels several orders of magnitude higher than amounts tr. tr.eir food organisms and the lake water, processes of uptake and elimination of these contaminants by fish are uncertain. The mechanisms controlling these processes should be determined and their relative importance evaluated to aid in understanding the relationships between contaminant concentrations in water and in the associated biota. The elucidation of these relationships will allow an assessment of the probable impact on the aquatic ecosystem when contaminant inputs to the lake are altered. .. ' '. . Since chlorinated hydrocarbons associated with the dissolved and particulate (organic and inorganic) fractions of lake waters may interact differently within the lake ecosystem, analytical methods should be developed to separate and quantitate the contaminant concentrations in these fractions. .. , Lake Ontario waters and sediments near the mouths of the Niagara and Oswego Rivers were found to contain significantly higher concentrations of ?CBs than the other waters and sediments sampled. This indicates .that the levels and forms (dissolved or associated with particulates) of PCSs in these rivers should be determined to allow an assessment of their importance as sources of PCBs to the Lake Ontario ecosystem. Since fish are capable of accumulating chlorinated hydrocarbon 'concentrations several orders of magnitude higher than the surrounding water, Lake Ontario fish should be extensively examined for halogerated '.vdrocybcns not previously Identified or confirmed in fish. Some compounds, although present ir. lake waters at undetected levels, may bo of'considerable importance duo to their toxicities even at very low levels. An investigation of the possible presence of previously unidentified halogenated hydrocarbons is currently In progress as an extension of the research presented in this report. ' '' HONS 067602 3 . ' SECTION III ' . INTRODUCTION' Measurements of pesticide residues in the Great Lakes have shown excessive levels of several chlorinated hydrocarbons. Polychlorinated biphenyls (PCDs) and DDT group pesticides (i.e., DDT, DDE, DDD, and - rcli.ted isomers) have been confirmed in Lake Michigan fish at levels exceeding U.S. Food and Drug Administration action limits, and dicldrin has beer, found at levels approaching the action limit (Veith, 197G; F.oir.ert, 1570). The pesticide contamination problem in Lake Ontario his received less attention than in Lake Michigan where residue concentrations in several segments of the ecosystem may be higher (Rcinert, 1970; Veith, 1973). Although studies of chlorinated hydrocarbon contamination in Lake Ontario fish have been conducted (Reinert, 1970; Kaiser,1974), information concerning other segments of the ecosystem .is -incomplete. . .- ..... . . This study was conducted as a part of the International Field Year for the Great Lakes to provide baseline information on the levels of CDT group pesticides, dieldrin, and PCBs in Lake Ontario fish, water, sediment, net plankton, Cladcphora, and benthic fauna. This information 'will allow a more complete assessment of the chlorinated hydrocarbon problem in Lake Ontario and contribute to an understanding of chlorinated hydrocarbon transport in aquatic ecosystems. , MCNS 067603 4 . .sa::pli.\s SECTION IV METHODS . '. . -:. ;v . . . . : ; . ' ' . Ives (Alosa pr.eudoharenous). smelt (Osnerus mordax), slimy sculpin (Cc-kus r.: m tus), water, sediment, net plankton, CUdophora. and benches samples were obtained from several near-shore and mid-lake sites on Lake Ontario during the summer of 1972. Sampling was more intensive near Rochester, Oswego, and Hamilton. Samples were also collected off Cobourg and Olcott, at the eastern end of the lake, and at four mid-lake sites. Figure 1 shows the locations of the sampling stations. Kater samples were taken (Van Dorn type sampler) just below the surface, .at 10 m- below the surface, and at 10 m above .no sediment at each station. Fish were trawl-netted at 10 to 73 ra end mixed plankton netted (64 um mesh opening) at 5 to 10 m. Clarionhora was gathered at 1 to 2 m depths. Sediments were sampled with a Ponar grab, ar.d benthos were captured using a opibcnthic sled. All samples wore transported and stored frozen or near 4C in glass or metal containers to minimize chemical changes and contamination. . EXTRACTION AND CLEANUP ' ' ;\V. Fish f - - ' Extraction and cleanup of whole fish were conducted according to the procedures described by Veith (1970). Frozen fish (a combined scrnple of at least eight individuals for each species captured at each site) were ground twice to homogenize the flesh before weighing out six 10 g sub-scir.ples. Each sub-sample was blended with 70 g anhydrous Na^SO^ and extracted for at least 4 hr with 170 ml of 1:1 ethyl ether-hexane (v/v) In an all-glass Soxhlet extractor. The extracts were concentrated to .20 ml In an air stream and 2 ml aliquots were removed for analysis of non-volitlle fats and oils (residue after . i-vcporct'lon at 15CC for 20 min). The remaining extracts were subjected to liquid chromatographic cleanup and fractionation. The extracts were placed on 20 g columns of florisil (Fisher F-100, 60-100 mesh washed : . . . > . ' ' HONS 067604 HONS 0 6 7 6 0 5 f-Fish, w-Water, s-Sedlment, p-Net Plankton, C-CIadophora, b-Benthos. Station numbers are IFYGL Station Identifiers, wit."*, hexane and activated by heating to 650C for 2 hr) and topped with anhydrous. NajjSO,, to prevent deactivation of tha florisil by water. The columns were sequentially eluted with 200 ml each of 6, 12, and 502 ether in hexar.e (v/v). Since preliminary gas chromatographic analysis o' the 12 ar.d 5G2 eluates failed to provide identification of compounds of interest, notably dicldrin and endrin, which v/ere identified in the i.2 eludes, only t.i)c 62 eluates were considered for subsequent examination. :he presence of dicldrin and endrig in the less polar fractions likely resulted from some deactivation of the florisil by water vapor prior to use. The 62 eluates were concentrated (air stream) to 50 ml before removing 5 ml aliquots for preliminary gas chromatographic determination if ODE {generally the largest peak on the chromatogram). The remaining . portions were concentrated to less than 10 ml before placing on 20 g col u.v.ns of silicic acid (washed with hexane, dried at 130C overnight, ana partially deactivated with 2.IS water). Elution with 250 ml hexar.e produced the PCD fractions. Further elution with 200 ml 3:1 dichloro- mothsne-hexane (v/v) allowed elution of the chlorinated pesticide fractions. These fractions were evaporated in an air stream, hexane ' was. added, and the fractions were re-evaporated. This process was repeated several times to insure complete removal of the dichloromethar.e. 'deter ' . ... _ ; v - \ . _ ' Each water sample (10 1) v/as extracted at the collection site by passara through a column of six polyurethane foam plugs at a flow rate of 250 ml/rdr, usir.g the procedure described by lithe et al_. (1972). Preparation of the plugs involved Soxhlet extractions (at least 4 hr) with 1:1 ethyl ethor- hexene (v/v) to remove contaminants followed by coating the plugs with a 12 solution of DC-200 silicone oil in hexane and air-drying. Following extraction of water samples, the plugs were removed from the column and again extracted (4 hr) with 1:1 ether-hexane (v/v). The columns were rinsed several tiroes with acetone. The combined extracts from the plugs ar.d column rinsings were reduced to about 5 ml in an air stream before cleanup by liquid chromatographic procedures similar to those for cleanup of fish extracts, . HONS 067606 7. Simples ivc re ill owed to air dry at room temperature before weighing out six 25 g sub-samples for'analysis. The sub-samples were thoroughly ground by mortar and pestle, mixed with anhydrous ,'ia,,S04, and extracted .:itr. 17C ml 1:1 ether-hexane (v/v) In all-glass Soxhlet extractor: for t hr. The extracts were concentrated to about 10 ml before liquid chromatographic cleanup in a manner similar to that used for fish extracts. .' lint ?'. tnhton nncl Cladcphora plankton and Clarionhora samples were transferred to tared centrifuge tubus and sub-divided into sub-samples of about 1 g where appropriate. Aftor centrifuging at 2000 rpm for 25 mir., the supernatant was decanted r.piuly into separatory funnels, 2 ml acetone was added to each tube, and the samples were allowed to air-cry. The tubes were weighed ar.d sample dry weights wore determined by difference. Water decanted from each tube was extracted twice with 25 ml of hexane to recover materials released from the cells. A 35 ml portion of the extract was added to the corresponding sample tube and the tube was shaken periodically over & 30 hr period. The extract was decanted and the residue extracted a second time with a 35 ml portion of fresh hexane. The combined extracts were concentrated to 5 ml for cleanup by procedures similar tc those for fish extracts. . . ; Enrthas .' ' ' ' " Hued benthic fauna (largely Pontinoreia affinis) ware dried at roam temperature and separated into at least three 10 g aliquots before extraction and cleanup by procedures identical to those used for sediccn samples. _. >. . ?"St:c:oe oetekkikation f'.ialil.-uiv.j Determination ' '. besticide fractions of all fish and several water, sediment, plankton, Cla.'rrV.rn.and benthos extracts from silicic acid cleanup were MCNS 067607 c.r'C.i.atogrsphcd (Verier Aerograph 1 SCO or 170C) on four gas . chromatograph!;: (EC) color,ins eluting into electron capture detectors or i) to allow multiple column peak-matching identification of fr.aior. components. The GC columns and column conditions utilises arc ,:t:c in Table 1. To facilitate peak matching, peak retention times converted to relative retention times (relative to the retention si,:: of p.p'-COE) for each set of column and conditions. A file of ral.tive retention times of many chlorinated pesticides and metabolites os.-silod from chromatograms of'single- and multi-component pesticide siar-.i solutions and from retention time data reported by Thompson jj_. (1359). Positions of significant peaks on chromatograms cf cl to fractions on the four columns ware converted to relative retention cnnas for comparison with the relative retention times of sUnde.re: in t^c file. Peak identity assignments were made based on four matched relative retention times for peaks of similar height. Several pesticide ar.d PC3 fractions from silicic acid liquid " chrosatograp.no cleanup of fish extracts were examined by gas chromatograpny/ mass spectromstric (GC/KS) limited mass range scan techniques for COE, diclx'riri, and DOT. The fractions were chromatographed (Varian Aerograph KCD) cn a 1.3 m x 4 mm ID glass column of 32 DC-2C0 on GO-GO mesh Oas-Chrom Q, with column and injector temperatures of 175C ar.d 2C0C, respectively, eluting (by a helium flow of 10 ml/min) directly into the icr. source of a quacrupole mass spectrometer (Finnigan 1015C) focused on narrow m/e ranges characteristic of the degradation patterns cf DOE, dleldrin, or DOT. DOE ana COT wore monitored by focusing on m/e ranges cf D-if.-250 and 235-229, respectively, adapting procedures described by .one'ili (1072). Oieldrir. was detected by monitoring the m/e range of 2S1--5?. Identities ware based on the response of this selective catcctb:* consistent with appropriate, retention times. fr-',fi-,>tir TeUrnir.aticn .' for determination of DOE in fish, the 5 ml aliquots taken from extract;: sfac:* fltrisil cleanup were diluted to 10 ml and chromatographed on a l.i n x. 2 sea'ID glass column of 32 0C-200 on 80-100 mesh Chromasorb K, and MOWS 067608 9 Table 1. GAS CHROMATOGRAPHIC COLUMNS AND CONDITIONS. Stationary Phase* Column Length (m)** Column Temperature (C) Detector Temperature (C) Injector Temperature (C) Carrier Gas (Ng) Plow (ml/mln) . 1.5* OV-17/1.95* QF-1 2.1 185 210 . 210 . 15 . aSolld support was 100-120 mesh Gas-Chront Q. ^All columns are 2 urn ID colled glass tubes. 2* 0V-101/3* QF-1 1* QF-1 1.5 150 190 210 . 12 3 185 210 210 ' 36 . 31 0V-17 2.1 185 210 210 ' 24 HONS 0 6 7 6 0 9 elated with Ng carrier flow of 40 ml/min into an electron capture detector (3il). The column, detector, and Injector temperatures were 200, 210,and 225C, respectively. The relatively large p,p'-DDE peak of the chromatograms allowed its determination through peak height conparison with standards. Pesticide fractions from silicic acid cleanup of fish, water, sediment, net plankton, Cladophora.and benthos extracts were chromatographed on a 3 m x 2 nm ID glass column of 1% QF-1 on 100-120 mesh Gas-Chrom Q to determine the DDT group pesticides and dieldrin. Column, detector, and injector temperatures were 180, 200,and 215C, respectively, and the flow was 30 ml/min. An electron capture detector (3H or ^N1) was used. All fractions were diluted to 2 ml (with hexane) before determination except for fish extracts which were diluted to 25 ml. Areas of peaks identified as DDT group pesticides and dieldrin were neasured by a disc Integrator and compared with standards. Since DDE Is not fractionated cleanly into the pesticide fraction during silicic acid chromatography of extractsof water .net plankton, Cladophora. and benthos, the PCB fractions were also chromatographed as above to determine their DDE content. Sediment PCB fractions could not be simflarily examined because of severe interferences. The DDE content of several fish PCB fractions from silicic acid cleanup was determined by GC/MS limited mass range scan techniques. Chromatographic conditions were similar to those described for the qualitative determination of DDE. Peak heights were compared with those of standards. .. PCB DETERMINATION Perch!orination . PCB fractions from silicic acid cleanup of fish, water, sediment, net plankton, benthos, and Cladophora extracts were perchlorinated by procedures described by Veith (1973). Fractions were evaporated to dryness in glass vials, SbCl5 (0.2 ml) was added to each vial, and the vials were sealed with teflon-lined screw caps before heating to 180C .' . ;: . 11 MOWS 067610 for 6 hr. After cooling to near 0C'in an ice bath, 1 ml of 61[ HC1 was added to each vial to destroy the residual SbCl5. The reaction solutions were extracted with five 1 ml portions of hexane. The hexane extracts were combined and passed through a disposable pipet containing anhydrous Na2S04 to remove traces of water. Some highly colored extracts from sediment PCB fractions were washed with 1 ml of 102 KOH in anhydrous methanol to remove interfering inorganics. All ' perchlorinated extracts were reduced to 2 ml for GC determination of decachlorobiphenyl (DCB) except fish extracts which were analyzed at 25 ml. ' Determination of Decachlorobiphenyl - Perchlorinated PCB fractions were chromatographed on a 1.5 m x 2 mm ID glass column of 1.52 OV-17/1.95X QF-T on 100-120 mesh Gas-Chrom Q with N? flow rate of 50 ml/min into an electron capture detector ( H or Ni). Column, detector, and injector temperatures were 200, 210,and 230C, respectively. Peak heights of DCB in the extracts were compared with those of standards and the DCB content was converted numerically to equivalent concentrations of Aroclor 1254. The.results of these analytical methods are reported without correction with respect to recovery during sample extraction and extract cleanup. . HONS 067611 12 SECTION V RESULTS AMD DISCUSSION FISH All pesticide fractions from silicic acid cleanup of fish extracts contatned DDT group pesticides and dieldrin, based on identification by multiple column GC and peak-matching techniques. In addition, several other common chlorinated hydrocarbon pesticides were Identified in the extracts of fish from some Sampling sites using this technique. Notably, endrin was Identified in all fish taken off Prince Edward Point and from Mexico Bay, but was apparently absent in fish from a transect between Galloo Island and Stoney Island, the other eastern lake sampling site. Endrin was not identified in fish collected from western lake sites off Hamilton, Olcott, and Rochester. The BHC family pesticides were identified in fish taken off Olcott, Rochester, Prince Edward Point, and in Mexico Bay, with lindane (y BHC) generally the major constituent. Heptachlor was Identified in slimy sculpin taken off Rochester. Although identity assignments from a limited file of compounds were based on peak-matching which is subject to some of the ambiquity inherent in complex chromatograms, the use of a four-.column system tended to decrease the incidence of these ambiguities. The GC/MS limited mass range identity assignments of extract components provided positive identification of chlorinated hydrocarbons in fish. Using this technique, all fish pesticide fractions examined were shown to contain DDE, ODD, DDT, and dieldrin. In addition, the PCB fractions of extracts from most fish contained significant levels of DDE, while ODD, DDT, and dieldrin were undetectable (less than 2 ng/g on a whole fish basis). . Concentrations of t-DDT (sum of DDT, DDE, and DDD), dieldrin, and PCBs (expressed as Aroclor 1254 equivalent) found In whole fish (i.e. wet weight basis) and the extractable fat contents of the fish are shown in Table 2. The DDE values shown resulted from preliminary DDE HONS 067612 13 Table 2 CHLORINATED HYDROCARBONS AND FAT IN LAKE ONTARIO FISH* Species Location AT ewi fe A1 ewife . Alewife Alewife Alewife . Alewife ' Smelt Smelt Smelt ' Smelt . Smelt Slimy Sculpin Slimy Sculpin Slimy Sculpin Sliqy Sculpin Slimy Sculpin Slimy Sculpin Hami1 ton . Olcott Rochester ' Mexico Bay Prince Edward Pt. Galloo-Stoney. Hamilton Olcott Rochester Prince Edward Pt. Galloo-Stoney Hamilton Olcott Rochester Mexico 3ay . Prince Edward Pt. Galloo-Stoney Fat DDE ODD DDT Total DDT Dieldrin PC8 % ug/g whole fish 3.6 0.46 5.2 0.77 3.4 0.71 3.1 0.79 1.2 0.81 2.4 0.96 . 4.9 . 1.36 3.0 0.85 4.1 1.37 6.7 0.86 6.0 0.91 9.8 0.94 5.1 1.10 4.3 1.11 5.7 . 1.28 7.6 0.83 8.6 0.60 0.07 0.14 0.07 0.16 0.10 0.18 0.07 ' 0.13 N.D. N.D. 0.08 0.18 0.06 0.23 0.05 0.20 0.13 0.29 0.10 0.23 0.10 0.24 N.D. N.D. 0.15 0.29 0.10 0.26 N.D. 0.26 0.15 0.25 0.12 0.17 0.67 1.00 0.99 0.99 0.81 .1.22 1.65 1.10 1.79 1.19 1.25 0.94 1.54 1.41 1.54 1.23 0.89 ' 0.04 3.12 0.03 1.73 0.04 4.36 0.03 0.94 0.03 0.14 0.04 3.81 0.04 2.47 0.02 2.62 0.03 3.25 0.06 3.49 0.07 1.40 N.D. 2.89 0.06 9.17. 0.05 4.32 o.io - 6.49 0.11 ' 1.58 0.04 3.33 aN.D. Indicates that no determination was made. HONS 0 6 7 6 1 3 determinations on the extracts following florisil liquid chromatographic cleanup, while the DDD and DOT levels were obtained from determinations on pesticide fractions following silicic acid cleanup. DDE was also determined on the silicic acid pesticide fractions by electron capture GC and on the silicic acid PCB fractions by limited mass range GC/MS techniques. The sum of DDE in the silicic acid pesticide and PCB fractions was considerably less than the amount in the preceding florisil eluate, in some cases as much as 50% less. However since DOE was the largest peak for chromatograms of extracts after florisil cleanup, PCB contribution to the DDE peak was probably small. In most cases, the major contribution to t-DDT values was from DDE. Quantitatively, DDT and DDD were minor constituents, making up less than 26% and 14% of the total, respectively. Although dieldrin concentrations shown (Table 2) are of interest, the low levels observed do not allow evaluation of possible station-to-station or species-to-specles relationships. The higher levels of t-DDT observed provide a better basis for comparisons. Individual variation in the chlorinated hydrocarbon contents of the fish were decreased by sampling an aggregate of ground whole fish of several age-weight classes for each species collected at each site. Because chlorinated hydrocarbon levels have been related to the fat content of the fish (Veith, 1973), these variations likely are further decreased by examining chlorinated hydrocarbon concentrations in relation to the extractable fats contents of the whole fish. Table 3 shows t-DDT and PCB levels in fish based on fat content. This data indicates that the more migratory alewives and smelt accumulate higher t-DDT levels on a fat basis (averages of 36.2 ug/g and 30.5 ug/g) than the less migratory slimy sculpin (16.9 ug/g) (Scott and Crossman, 1973). Relative standard deviations for t-DDT levels (fat basis) for the species are 53, 36, and 66% for alewives, smelt, and slimy sculpin, respectively. Comparison of these variations with the average relative deviation for analytical replicates for extracts of an agggregated sample (14%) suggests that these variations are partly related to differences in chlorinated hydrocarbon levels among the sampling sites. - 15 HONS 06761-* ' Table 3 DOT AND PCBs IN LAKE ONTARIO FISH FAT (ug/g) Species Alewlfe Alewlfe Alewlfe Alewl fe Alewlfe Alewlfe Smelt Smelt Smelt Smelt Smelt Slimy Sculpln Slimy Sculpln Slimy Sculpln Slimy Sculpln Slimy Sculpln slimy Sculpln Location Total DDT PCB PCB/Total DDT Hamilton Olcott Rochester Mexico Bay 1 Prince Edward Pt. Galloo-Stoney Hamilton Olcott Rochester Prince Edward Pt. Galloo-Stoney - Hamilton Olcott Rochester Mexico Bay Prince Edward Pt. Galloo-Stoney 18.6 19.2 29.1 31.9 67.5 SO. 8 33.7 36.7 43.7 17.8 20.8 9.6 30.2 32.8 27.0 16.2 10.4 86.7 33.3 128.2 30.3 11.7 1S8.8 . 50.4. 87.3 79.3 52.1 23.3 29.5 179.8 100.5 113.9 60.3 38.7 4.7 1.7 4.4 0.9 0.2 3.1 1.5 2.4 1.8 2.9 1.1 3.1 6.0 3.1 4.2 . 3.7 3.7 i 5- 16 HONS 06761s This Is supported by the highest relative standard deviation for the slimy sculpin. Considering the slimy sculpin data shown in Table 3, the waters off Olcott, Rochester, and of Mexico Bay may contribute to greater accumulation of DOT group pesticides by slimy sculpin than water off Hamilton, Prince Edward Point.and between Galloo and Stoney Islands. The lake bottom characteristics of these areas may also be Important factors in the chlorinated hydrocarbon accumulation in the bottom-feeding sculpin. The lake-wide average for t-DDT (whole fish) In smelt (1.40 ug/g) for 1972 compares favorably with that reported by Reinert (1970) for Lake Ontario fish captured from 1965 to 1968 (1.58 ug/g) although the alewlfe value (0.95 ug/g) is less than half that reported in 1970 (1.99 ug/g). Reinert (1970) did not report on Lake Ontario slimy sculpin. Dieldrin levels reported by Reinert (1970) for alewives and smelt captured from 1967 to 1968 (0.11 ug/g and 0.06 ug/g, respectively) are comparable to levels in.fish for 1972 (0.04 ug/g for alewlfe and smelt). PC8 concentrations {2.65 ug/g) found in smelt (Table 2) were similar to those reported by Veith (1973) for Lake Michigan smelt (2.7 ug/g). However, concentrations found in alewives (2.35 ug/g) were considerably lower than values reported for Lake Michigan alewives (4.6 ug/g). For both lakes, PCB levels were expressed as Arochlor 1254 equivalent. Slimy sculpin were not included In the Lake Michigan report. . Lake Ontario slimy sculpin exhibited highly variable PCB and t-DDT concentrations on a fat basis. The relative standard deviation for PCB levels in slimy sculpin (fat basis) was 65% about a mean of 97.1 ug/g.' Corresponding relative standard deviation values for alewives and smelt were about 80S (mean * 74.8 ug/g) and 44X (mean = 58.5 ug/g), respectively, This indicates greater statlon-to-station variation for PCB accumulation in slimy sculpin and alewives than In smelt. Furthermore, the waters near Olcott, Rochester and in Mexico Bay contributed to greater accumulation of PCB concentrations by sculpin than waters off Hamilton, 17 (SONS O*7616 Prince Edward Point, and between Galloo and Stoney Islands, which is in agreement with the trend for DOT accumulation. . A large variation was observed in PCB/t-DDT ratios for fish (Table 3). Average values, however, (2.5 for alewives, 1.9 for smelt, and 4.0 for slimy sculpin), were comparable to those reported by Veith (1973) (1.4 for alewives and 2.6 for smelt) for Lake Michigan fish. PCB/t-DDT ratios are often cited due to chemical similarities between PCBs and persistent DDT metabolites. Although their importance has not been well established, these ratios may be indicative of the relative accumulation and/or removal rates of PCBs and DDT group pesticides and may contribute to an assessment of the relative importance of Industrial and rural contaminant inputs to specific lake regions. HATER - The t-DDT, dieldrin, and PCB concentrations for water are shown in Table 4. Since there v/ere no apparent relationships between concentrations and depth, with the one exception as discussed below, determinations from different depths v/ere treated as replicates for each site and ' averaged. Lake-wide averages for t-DDT and dieldrin were 23 ng/1 and 4.8 ng/1, respectively. Relative standard deviations for t-DDT and dieldrin were 542 and 852, respectively, indicating considerable site-to-site variation. Waters off Hamilton, Cobourg, and Oswego showed high levels of t-DDT. Dieldrin levels were highest off Cobourg and Oswego. The anomalously high dieldrin level for waters of Oswego resulted from a very high surface water value (34.9 ng/1) which was averaged with the much lower levels found in samples of deeper v/aters (1.7 ng/1 and 1.1 ng/1). In all v/aters, except off Oswego, DDE v/as the major component of the DOT group pesticides. DDD contributed 0 to 192, while DDT represented 6 to 132 in most cases. The t-DDT in v/aters off Oswego contained higher proportions of DDD and DDT, 282 and 262, respectively. Waters from the Deep Hole area of the Rochester Basin showed a t-DDT level of 16 ng/1 with DDT contributing over 402. V. . MONS 067617 18 Table 4. CHLORINATED HYDROCARBONS IN LAKE ONTARIO WATER (ng/1) Location Hamilton Toronto Niagara River Olcott Cobourg Rochester Deep Hole Oswego IFYGL Station Identifier Station Depth ID 1 8 13 30 36 60 75 . . 90 ` 33 . 76 13 . 24 24 . 25 229 .21 ODE 37.4 20.5 13.9 26.6 45.2 29.9 9.4 22.4 ODD 2.5 1.6 0.9 7.1 4.5 < 0.5 < 0.5 13.8 DDT Total DOT . . Dleldrln . PCB . 4.5 1.4 2.4 4.6 7.2 ' 2.3 6.5 12.8 44 3.1 49 24 3.5 35 . 17 2.1 97 38 3.9 44 '57 9.9 45 32 2.2 ' 40 16 1.3 56 .49 ... ....12.6 ........ ...77 HONS 06 PCB concentrations for Lake Ontario waters were between 35 and 56 ng/1, except for v/aters off Oswego (77 ng/1) and the mouth of the Niagara River (97 ng/1). The lake-wide average concentration was 55 ng/1. Keglccting the two highest concentrations, the lake-wide average becomes 45 ng/1 with a relative standard deviation of 162. Other than these south-shore areas, the PC8 content of Lake Ontario waters appears relatively uniform. The PCB/t-DDT ratios for Lake Ontario waters averaged 2.1, a value comparable to that for Lake Ontario fish. Since the water sampling and extraction procedures did not discriminate between dissolved and particulate fractions, the t-DDT, dieldrin, and PCB concentrations shown in Table 4 represent "total" concentrations. The contribution of contaminants associated with net plankton (i.e., particles larger than 64 um) to the "total" levels found in Lake Ontario waters is estimated to be less than 12. This estimate was calculated from the contaminant concentrations found in net plankton (Table 6) from three areas of Lake Ontario, phytoplankton cell counts determined for these areas during the month these areas were sampled for chlorinated hydrocarbons (Stoermer, 1973), and dry weight per cell values reported for a laboratory algal culture (Lee et a1_., 1971). Thus DOT group pesticides, dieldrin, and PCBs in the Lake Ontario waters sampled are likely dissolved or associated with particles that will pass through a 64 um net (e.g., nanoplankton and small inorganic particles). . '. SEDIMENT - Sediment concentrations of t-DDT, dieldrin, and PCBs are shown in Table 5. Lake-wide averages for sediment t-DDT and dieldrin are 22 ng/g end 1.2 ng/g, respectively, with relative standard deviations of 722 and 612, respectively. These variations are indicative of significant slte-to-site differences. Sediments taken off the mouth of the Welland Canal and at the eastern mid-lake site showed higher levels of dieldrin and t-DDT. Sediment from three of the eight sites showed DDE contributions -to t-DDT levels of less than 50%. ODD was the major contributor to t-DDT *CNS 067619 20 Table S. CHLORINATED HYDROCARBONS IN LAKE ONTARIO SEDIMENT (ng/g dry sediment)1 Location Welland Canal Niagara River Olcott Cobourg Hid-lake Rochester Oswego Mid-lake East IFYGL Station Identifier. 12 13 30 36 46 60 91 93 DOE 12 . 11 4.8 . 8.0 - 11 . 8.0 ' . 9.0 : 16 D00 15 3.5' 5.6 0.9 5.4 1.5 5.1 31 aN.D. Indicates no determination was made. DDT 12 . 0.7 1.2 0.9 2.8 0.2 3.8 7.4 Total DDT 39 15 12 10 19 10 . 18 54 01 el drIn 2.6 1.4 . 0.9 0.6 0.5 0.9 0.8 ' 2.1 PCB 245 155 80 43 79 . 84 158 N.D. SHOW In sediments from the eastern mid-lake site and off the Welland Canal. The DOE levels shown in Table 5 must be considered underestimates, however, since ft was not possible to measure the DDE contents of sediment PCB fractions due to interferences. Lake-wide t-DDT and dleldrin averages were comparable generally to those found by Leland et al_. (1973) for southern Lake Michigan top interval sediments (18.5 ng/g and 2.0 ng/g, respectively). However, the Lake Michigan sediments showed major DDT contributions (ca. 50X) to the t-DDT, while Lake Ontario data showed DDT contributions no greater than 312. Sediment PCB concentrations averaged 120 ng/g. PC8 concentrations in sediments off the mouths of the Welland Canal and Niagara River and off Oswego, (mean * 184 ng/g) averaged more than twice the concentrations found at the four other sites (mean = 72 ng/g). Considering the high PCB contents found for lake water off the mouth of the Niagara River and off Oswego, the Niagara and Oswego Rivers may be Important sources to Lake Ontario of PCBs associated with settlable particulates. The PCB/t-DDT ratios exhibited a low relative standard deviation of 33X about a mean of 7.0. This mean may be too high due to underestimation of DDE levels. It Is unlikely, however, that the DDE content of the sediment extracts could be great enough to decrease PCB/t-DDT ratios to levels similar to those for the fish and water. NET PLANKTON ' DDT group pesticides, dleldrin, and PCB levels (dry weight basis) found In mixed net plankton are shown in Table 4. Although the samples were predominantly viable phytoplankton, the sampling method also collects smaller zooplankton, detritus, and suspended Inorganic matter. Contributions of these minor components could not be assessed. Lake-wide averages for t-DDT and dleldrin were 3.5 ug/g and 0.12 ug/g on a dry . weight basis, respectively, with corresponding relative standard deviations of 422! and 91%. In nearly every case, DDE comprised over 75X of the t-DDT concentrations. . ' 22 ONS 676Zl Table 6. CHLORINATED HYDROCARBONS IN LAKE ONTARIO NET PLANKTON (ng/g dry weight)* Location Hamilton Mid-lake West Cobourg Mid-lake Rochester Deep Hole Mid-lake East IFYGL Station Identifier 1 10 36 45 60 75 96 . DDE 4.00 3.52 3.26 1.49 1.19 5.89 2.45 DOD 0.09 0.37 <0.05 0.07 <0.05 . 0.04 0.09 aN.D. Indicates that no determination was made. DDT 0.040.12 < 0.05 0.7B . < 0.05 < 0.05 0.86 Total DDT 4.1 4.0 3.3 2.3. 1.2 5.9 3.4 Dleldrln 0.24 0.25 < 0.05 0.16 0.02 0.02 0.18 PCB 3.4 10.6 7.6 3.6 N.D. 11.8 6.0 HONS 0 6 7 6 2 2 Net plankton PCB levels averaged 7.2 ug/g. Plankton from the Deep Hole of the Rochester Basin and the mid-lake west site, show high PCB levels contributing to a relative standard deviation of 49" for the lake-wide average PCB concentration. The average PCB/t-DOT ratio (3.1) was sooewhat higher than for water or fish. Chlorinated hydrocarbon levels for plankton appeared high compared to concentrations in fish. However,, plankton concentrations were expressed on a dry weight basis while fish concentrations were on a whole fish (wet weight) basis. Assuming whole1 fish are about 20% dry weight (F.D.A., 1969), typical fish t-DDT, dieldrin, and PCB concentrations were 7.0 ug/g, 0.3 ug/g, and 13.2 ug/g dry weight (smelt), respectively. On this basis, net plankton concentrations of these chlorinated hydrocarbons were about 50% less than concentrations in fish. CLAOOPHORA ' The results of determinations of DDT group pesticides, dieldrin, and PCB levels in extracts of Cladophora are shown in Table 7. Host samples, except those off Toronto and in the Black River Bay, were from the south shore of Lake Ontario. The exact sampling site locations of four of the samples is uncertain. The average t-DDT, dieldrin, and PCB levels (dry weight basis) were 229, 13,and 515 ng/g, respectively,, with relative standard deviations of 40, 64.and 39% respectively. These concentrations are 10 to 100 times less than for net plankton. The average PCB/t-DDT ratio of 2.3 was near the corresponding averages for fish and water. .. . , BENTHOS ' DDT group pesticide, dieldrin, and PCB levels determined for extracts Of benthic fauna, largely Pontiporeia, are shown in Table 8. Of the three samples taken, one off Hamilton showed t-DDT, dieldrin, and PCB levels approximately four times those determined for benthos off Oswego and Rochester. Also, 000 made a significantly larger contribution to the t-DOT in benthos off Rochester than other samples. Average t-DDT and PCB levels, 99 ng/g and 471 ng/g on a dry weight basis, respectively, '. 24 HONS 067623 Table 7, CHLORINATED HYDROCARBONS IN LAKE ONTARIO CLAOOPHOKA (ng/g dry weight) Location DOE 000 DDT Black River Bay Black River Bay South Shore* South-Shore South Shore South Shore Rochester Toronto 344 4.8 7.4 129 0.45 1.9 97 19 2.5 194 26 4.7 347 17 1.1 192 30 16 165 5.2 6.0 196 13 8.0 Specific sampling site unknown. ' Total DOT 357 131 119 225 365 238 176 217 Dieldrin 1.9 6.5 14 14 4.0 25 16 21 PCB 860 333 232 607 436 576 411 666 C J. HONS 067624 25 Table 8. CHLORINATED HYDROCARBONS IN LAKE ONTARIO BENTHIC FAUNA (ng/g dry weight) Location Haoii 1 ton Rochester Oswego DOE DDD 124. 34 26 26 1.8 3.4 DDT 59 6.1 2.4 Total DDT 209 42 32 Dieldrln 14.8 2.9 3.0 PCB 976 341 97 were quite comparable to levels reported by Flotard (1974) for Pontiporeia of the western shore of Lake Michigan, 130 ng/g and 310 ng/g, respectively. The PCB/t-ODT ratios averaged 5.3. This value is somewhat higher than that calculated from the concentrations reported by Flotard (1974) or ratios calculated for all other samples reported except sediment. Although some sediment contamination may be indicated, the benthos PCB/t-OOT ratio is comparable to that for slimy sculpin (4.0), whose diet is primarily benthic fauna (Scott and Crossman, 1973). HONS 067626 27 SECTION VI REFERENCES Bonelll, E.J. 1972. Gas Chromatograph/Hass Spectrometer Techniques for Determination of Interferences In Pesticide Analysis. Anal. Chem. 44:603-606. Flotard, R.D. 1974. PCBs In Lake Sediments. Ph.D. Thesis (Water Chemistry), University of Wisconsin, Madison (in progress). Kaiser, K.L.E. 1974. Mirex: an Unrecognized Contaminant of Fishes from Lake Ontario. Science 185:523-525. Lee, C.C., R.F. Harris, J.K. Syers, and D.E. Armstrong. 1971. Adenosine Triphosphate Content of Selenastrum capricornutum. Appl. Microbiol. 21:957-958. Lei and, H.V., W.N. Bruce, and N.F. Shimp. 1973. Chlorinated Hydrocarbon Insecticides in Sediments of Southern Lake Michigan. Environ. Sci. Technol. 7:833-838. ' Reinert, R.E. 1970. Pesticide Concentrations in Great Lakes Fish. Pestic. Konit. J. 3:233-240. Scott, W.8. and E.J. Crossman. 1973. Freshwater Fishes of Canada. Fish. Res. Bd. Can. Bulletin 184, Ottawa, 966 p. Stoermer, E.F. 1973.' Analysis of Phytoplankton Composition and Abundance During IFYGL. In: First Annual Reports of the EPA, IFYGL, Environmetnal Protection Agency Report 660/3-73-021, p. 90-109. Thompson, J.F., A.C. Walker, and R.F. Moseman. 1969. Evaluation of Eight Gas Chromatographic Columns for Chlorinated Pesticides. J.A.O.A.C. 52:1263-1277. U.S. Food and Drug Administration. 1969. Pesticide Analytical Manual. U.S. Department of Health, Education and Welfare. Uthe, J.F., J. Reinke, and II. Gesser. 1972. Extraction of Organochlorine Pesticides from Water by Porous Polyurethane Coated with Selective . Absorbent. Environ. Letters. 3:117-135. ' HONS 007627 28 Veith, G.D. 1970. Environmental Chemistry of PCBs in the Milwaukee River. Ph.D. Thesis (Water Chemistry), University of Wisconsin, Madison, 180p. Veith, G.D. 1973. Chlorinated Hydrocarbons in Fish from Lake Michigan. U.S. Environmental Protection Agency Project 16020P8E, 129p. HONS 067628 29