Chlorobornanes in Sediments and Fish 30 Years after Toxaphene

Oct 1, 1995 - Neil L. Rose, Sean Backus, Heidi Karlsson, and Derek C. G. Muir ... David B. Donald, Gary A. Stern, Derek C. G. Muir, Brian R. Fowler, B...
0 downloads 0 Views 1MB Size
Environ. Sci. Techno/. 1995, 29, 2490-2495

and Fish 30 Years after BRENDA M. MISKIMMIN,+ DEREK C. G. MUIR,* DAVID W. SCHINDLER," GARY A . S T E R N , * A N D NORBERT P. GRIFTt Department of Biological Sciences, University of Alberta, Edmonton, Alberta, Canada T6G 2E9,and Department of Fisheries and Oceans, Freshwater Institute, Central and Arctic Region, 501 University Crescent, Winnipeg, Manitoba, Canada R3T 2N6

Chlorobornanes (CHBs) were isolated from recent and older sediments as well as from rainbow trout muscle in two prairie lakes that were treated with toxaphene as a pesticide in the early 1960s. Cored sediments representing the years of treatment (1961/ 1962) contained high concentrations of hepta-, octa-, and nonachlorobornanes (CCHBs, 1602 and 500 ng g-l) with a similar gas chromatographic peak pattern to technical toxaphene. In contrast, negative ion mass spectrometry confirmed that the near-surface sediments contained hexa- and heptachlorobornanes at levels not normally observed in environmental samples. This suggests that some of the original toxaphene mixture had been dechlorinated and was redistributed by porewater diffusion and/or sediment focusing. Analyses of fish revealed a toxaphene "fingerprint" similar t o that of the dechlorinated material in sediments, suggesting that the chemical was both water soluble and bioavailable. Significant residual CHBs in surface sediments and in fish indicate that other toxaphene-treated lakes should be reexamined.

Introduction Toxaphene is an organochlorine insecticide (also called camphechlor) that was once widely used as a pesticide and briefly used as a fish toxicant during the late 1950s and 1960s in many North American lakes. It is composed of hepta- to decachlorocamphene derivatives, mostly chlorinated bornanes (CHBs). More than 200 peaks can be resolved by high-resolution gas chromatography (1, 2). While others have studied the residues of toxaphene in disposal sites or croplands where it was primarily used, relatively little is known about the long-term fate of * Address correspondence to this author; telephone: (403) 4921291; fax: (403) 492-9234. +

University of Alberta. Freshwater Institute.

2490 ENVIRONMENTAL SCIENCE &TECHNOLOGY / VOL. 29, NO. 10, 1995

toxaphene in aquatic environments where it was directly applied. For example,toxaphene is known to leach through sandy soils (31,topsoil (41, and anaerobic saline marsh soils and plants (3,but similar data on the distribution and movement in lake sediments are lacking. Other aspects such as short-term degradation (6, 7), the role of lake sediments in the complexation of toxaphene (€9, and the fate of various toxaphene fractions in model freshwater ecosystems (9) have been examined. Reductive dechlorination or dehydrochlorination are considered the main pathways resulting in losses of the higher chlorinated forms of toxaphene (10) however, complete degradation of the compound does not usually result. The more water-solublecomponents (earliereluting) are apparently formed relatively quickly after introduction of the parent compound to sediments (a few days, 7; 3 weeks, 11; several months, 12). No previous studies have addressed the long-term (decades)distributionin sediments and congener-specificcharacteristics of residual toxaphene from lakes that were directly treated with the chemical. Recent studies using high-resolution gas chromatography and electron capture negative ion mass spectrometry (GC-ECNIMS)have demonstrated the persistence of some CHB congeners in biota (13,141 and in air samples (15).In both biotic and abiotic samples, the pattern of CHB peaks is dominated by a limited number of hepta-, octa-, and nonachlorobornanes, indicating that the majority of the components in the original toxaphene mixture have been degraded (16). To our knowledge, specific CHB peak patterns in sediments have not previously been examined in detail by GC-ECNIMS. Because of the demonstrated persistence of these and other congeners of toxaphene as well as reports of elevated toxaphene in regions remote from direct application ( I 7191,we felt that lakes that were previously treated with the compound should be re-evaluated. The objectives of this study were to examine the distribution of toxaphene in profundal sediment cores from two treated and one untreated lake and to compare chromatographic patterns between recent and older sediments and fish from the two treated lakes. The known additions of toxaphene during the early 1960s permitted the evaluation of persistent organochlorine compounds as sedimentary date markers (critical to paleoecologists) and the assessment of current analyticalmethodology for CHBs in environmentalsamples.

Materials and Methods Sediment Collection and Preparation. Sediment cores were collected by using a freeze-corer (20,21) at the deepest site of three Albertalakes: two were treatedwith toxaphene in 1961/1962,and one was untreated. This coring process has the advantages of collecting undisturbed sediments (evidenced by preserved laminations) and freezing the sediments immediately upon collection to prevent chemical losses. Peanut Lake (54'01' N, 114'21' W mesotrophic) was treated with 7.5pg L-l toxaphene in September 1961, and Chatwin Lake (54'15' N, 110'51'W eutrophic) was treated at 18.4pg L-l in October 1962. In Peanut Lake, an 11.5-cm unlaminated interval (referred to as a "slump") occurred

0013-936x/95/0929-2490$09.00/0

D 1995 American Chemical Society

between 9.5 was 21 cm, resulting in the deep location in sedimentsofthe toxaphenemaximum. Theuntreatedlake, Peanut Lake untreated, is a distinct basin of Peanut Lake that was separated by a berm at the time of treatment to prevent native fish from reinvading the treated basin in years of high water. The upper 5 cm of sediments from the untreated basin were analyzed for toxaphene,representing our control. Other characteristics of Peanut and Chatwin Lakes are given in Miskimmin and Schindler (22). Frozen cores, about 30 cm in length, were sliced in lengthwisequarterswithabandsawandtheneitherfreezedried completely before sectioning or sectioned with the saw prior to freeze-drying. Sections usually represented 1or 2-cm slices in sediments, except surface sediments that represented 3-5 cm to provide the sample size necessary for toxaphene analysis. Nine core samples from Chatwin Lake and 11from Peanut Lake provided a toxaphene profile for each of the two treated lakes. Typically, 10 g of freezedried sediments is extracted for analysis, but the sediments from lakes directly treated with toxaphene contained enough toxaphene that we measured the compound in samples as small as 1.5 g dry wt. Sediments from both lakes were dated using 210Pb analysis (23) and varve counts (varves are annual sets of laminae) in Chatwin Lake only. Maximum toxaphene concentrations further confirmed the approximate depth representing the year of treatment. RahbowTmut Samples. Stocked rainbow trout (-2 yr old1 were collected from the two treated lakes (Peanut and Chatwin) using 3-in. gill nets. Fish were kept in frozen storage until subsampling and analyses. Thirty to forty grams of skinned dorsal muscle from three fish from each lake was removed for toxaphene analyses. Percent of dorsal musclethatwascomprised oflipidwascalculatedfollowing lipid extraction (explained below). Toxaphene (CHB) Fxtraction and GC Analyses. The procedures for extraction of CHBs were described by Muir et al. (17, 24). Briefly, fish tissue was homogenized by grinding samples with dry ice, and lipid was removed by automated gel permeation chromatography. Freeze-dried sediments were Soxhlet-extracted with dichloromethane (DCM)f o r l 6 h preparedfishtissueswereSoxhlet-extracted with 1:l hexane/DCM. Internal standards of aldrin and octachloronaphthalene(OCN)wereadded at the extraction step. Sulfur was removedusingactivated Cufilings. CHBs in sediment and fish extracts were separated from PCBs by chromatography on Florisil (17). Extractswereanalyzed bycapillarygaschromatography with ekCtrOn-CapNre detection (GC-ECD), as described by Muiret al. (17,2¶. Sampleswereinjected(splitlessmode) on a 60m x 0.25 mm i.d. DB-5 column (film thickness 0.25 pm) with an initial temperature of 100"C then programmed at 15 "C min-l to 150 and 3 "C min-l to 265 "C. Carrier gas was Hz(about 1 mL min-l) and makeup gas was N2 (40 mL min-ll. UndertheseconditionsOCN hadaretention time ofapproximately47.7min. Toxaphenewas quanaedusing a procedure similar to that described by Ribick et al. (25). Twentypeaksintheanalyticalstandard (toxaphenemixcure obtained fromthe U.S. EPA, Cincinnatti OH) were selected based on their prominence in the toxaphene standard, and a single response factor (SRF) was determined based on peak area. Toxaphene concentration was calculated by multiplyingtheareaofpeakswith the same retention times in the sample chromatogram by the SRF. Toxaphene

i

Peanut L

199oc 0-3cm

Chatwin L

1981 1978 1973

1968

1967

1963

1962

20-22

10-11

11.5

1962

125\

I

Treatment

1959

Treatment

I i

1955

::::/:: 1944 *27

O

% Qoo

600

ng g

O

VQJ

%

7%

'%0

dry weight

FIGURE 1. Toxaphene concentration profiles by depth and age in slices offreemcored sediments in Peanut Lake and Chatwin Lake. Alberta. Note that toxaphene was undetectable (