Environ. Sci. Technol. 2001, 35, 1312-1319
An Historical Record of Toxaphene and Its Congeners in a Remote Lake in Western Europe NEIL L. ROSE* Environmental Change Research Centre, University College London, 26 Bedford Way, London WC1H 0AP, U.K. SEAN BACKUS, HEIDI KARLSSON, AND DEREK C. G. MUIR National Water Research Institute, Environment Canada, Burlington, Ontario L7R 4A6, Canada
Historical trends in the concentrations of toxaphene, its component homologue groups, and individual chlorobornane congeners were determined in a dated sediment core from a mountain lake, Lochnagar, in Scotland, U.K., representing the first such profiles outside of North America. The profile of total toxaphene showed a bimodal distribution with maxima in the mid-1970s and early 1990s unlike the unimodal PCB profile (maximum 1973) in the same core. The earlier toxaphene peak shows agreement with the U.S. source curve and therefore may correspond to modeled global patterns while the later peak may be due to long-range transport from eastern and southern Europe or from still lower latitudes. Sediment toxaphene concentrations (14 ng/g dry wt (dw) at surface; 40 ng/g dw at maximum) and accumulation rates (surface 0.42 ng cm-2 yr-1; maximum 1.6 ng cm-2 yr-1) were considerably higher than levels in untreated sites in the Great Lakes region and the Canadian Arctic and are equivalent to those reported for the Great Lakes themselves where there have been additional riverine inputs. Two toxaphene congeners, B6923 and B7-1001, accounted for most of the hexa- and heptachlorobornanes, respectively. B6-923 and B7-1001 abundance ratios (relative to the sum of 12 chlorobornane congeners) had doubling times of 17 ( 3 and 12 ( 3 yr, respectively. Given the remoteness of the site from areas of toxaphene production and usage, the high sediment levels raise concerns over toxaphene levels in areas of Europe closer to sources, especially where sites are fished for human consumption. Further European data are needed for comparison.
Introduction Toxaphene is a complex mixture of polychlorinated bornanes and camphenes (1) and was widely used as an insecticide following the ban on DDT in the 1970s (2, 3). It was particularly used in the cotton-growing industry (1, 2) but was also used as a piscicide to remove undesirable fish, where it was found to cause damage to nontarget organisms (1, 4, 5). It is considered carcinogenic (2), mutagenic (6), and has been found to affect the central nervous system, immune system, * Corresponding author telephone: +44 20 7679 5543; fax: +44 20 7679 7565; e-mail:
[email protected]. 1312
9
ENVIRONMENTAL SCIENCE & TECHNOLOGY / VOL. 35, NO. 7, 2001
and a range of internal organs (3). It has therefore been placed on the United Nations Economic Commission for Europe (UNECE) list of banned persistent organic compounds and the forthcoming United Nations Environment Program (UNEP) global ban list (7). Increasing attention has been focused on toxaphene following concerns about the relationship between human health and consumption of fish (1). In lakes that were directly treated with toxaphene as part of fish eradication programs, it was found that restocking could not be successfully undertaken for years after treatment (4) and that dechlorinated components remained bioavailable 30 yr later (8). Largest usage was in the United States and the former Soviet Union while in Europe use has been limited (9, 10). However, the volatile, persistent, and lipophilic nature of toxaphene has resulted in global atmospheric distribution, and traces have been found in areas remote from any treated sites. Indeed, Kidd et al. (11) reported that levels in an Arctic lake, resulting entirely from atmospheric deposition, have biomagnified to an extent that the fish are considered hazardous to human health. Other studies have confirmed this potential for long-range transboundary movement, for example, to the High Arctic (12) and Lake Baikal in eastern Siberia (13), resulting in the contamination of both lake waters and biota (14, 15). However, although some contemporary levels of toxaphene in biota [e.g., marine fish (1, 16, 17) and mammals (1); riverine birds (18); and freshwater fish (16, 19)] have been recorded in Europe, there are few studies on lake ecosystems and none on long-term trends despite the use of toxaphene in southern and eastern regions. The extent and duration of contamination at untreated freshwater sites in Europe is therefore unknown, as is the extent to which toxaphene may have bioaccumulated to harmful levels in freshwater fish stocks. This paper uses the lake sediments of a U.K. mountain lake, remote from areas of toxaphene production and use, as a natural archive. We present the first historical record of toxaphene outside North America as well as the first longterm record of individual chlorobornane (CHB) congeners in sediment contaminated solely by atmospheric deposition. Because historical deposition of polychlorinated biphenyls (PCBs) is, by comparison with toxaphene, relatively welldocumented in the U.K. and western Europe, we also include the profile for PCBs in the same core.
Site and Methodologies Site. Remote mountain lakes are excellent indicators of pollutant deposition and its effects as they are not influenced by direct forms of disturbance such as land-use change and wastewater input. Furthermore, geological and climatic factors combined with shallow soils and high relief results in their being more sensitive and vulnerable to such pollutants. The sediment records of mountain lakes are known to provide reliable archives of atmospherically deposited pollutants (20) while the ability to produce sediment core chronologies using radionuclides (e.g., 210Pb, 137Cs, 241Am) allows the date of depositional events and the rate of any recorded change to be calculated. Lochnagar (56°57.48′ N, 3°13.08′ W) (Figure 1) is a remote corrie lake in the Grampian Mountains of Scotland lying in the center of a granitic massif at an altitude of 785 m. It is part of the U.K.’s Acid Waters Monitoring and Environmental Change Networks and the UNECE’s International Cooperative Programme on Assessment and Monitoring of Acidification of Rivers and Lakes and hence is one of the most intensely studied and monitored freshwater bodies in 10.1021/es0015895 CCC: $20.00
2001 American Chemical Society Published on Web 02/27/2001
FIGURE 1. Map of the North Atlantic with (inset) the location of the study site Lochnagar in the United Kingdom. the U.K. Toxaphene has been neither used nor produced in the U.K., and while some production was undertaken in France and Germany (1955-1990, main production in the early 1980s) (2), much of this was exported to eastern Europe and Cuba (10). Therefore, Lochnagar and its sediment record are ideally placed to monitor the long-range transport and historical deposition of this pollutant to an area of Europe remote from any direct treatment. Sediment Coring and Dating. A sediment core was taken from the deep area of the lake (23 m) using a modified Glew gravity corer (21) from a securely anchored inflatable boat. The core was extruded vertically in the field in 0.5-cm slices directly into sterile Whirlpak bags that were immediately sealed and stored cold (4 °C) prior to all analyses. Homogeneous portions of 18 samples from the core were treated using a variation on the Eakins and Morrison (22) polonium distillation procedure. Further details on the 210Pb analysis are given in Turner (23). The sedimentation rate of the core varied between 0.08 and 0.17 cm yr-1, while the accuracy of the dating ranged from ( 1 yr in the surface slices to ( 5 yr in slices corresponding to the late 19th century. Organochlorine Analysis. Sediment extraction to determine toxaphene, PCBs, and other organochlorines pesticides was performed as described by Muir et al. (24, 25) with minor modifications. Wet sediments were centrifuged to remove excess water and then combined with anhydrous Na2SO4 to yield a dry powder. Samples were extracted on an accelerated solvent extractor (Dionex Instruments, Oakville, ON) using dichloromethane (DCM). The conditions involved two separate 10-min extractions at elevated temperature (100 °C) and pressure (2000 psi). PCB 30 and octachloronaphthalene (OCN) were added to all sediments prior to extraction and used as internal standards. Combined DCM extracts were mixed with mercury to remove sulfur. Extracts were then transferred into hexane and were chromatographed on an activated silica column. PCBs were eluted with hexane, and most CHBs (except B8-1413 which eluted with hexane) were eluted along with HCH, chlordane-related compounds, p,p′DDD, and p,p′-DDT with (1:1) hexane:DCM. Extracts were
then carefully reduced in volume and taken up in isooctane for gas chromatographic analysis. Gas chromatography using electron capture negative ion mass spectrometry (GC-ECNI/MS) analysis of sediment core extracts was carried out using a Hewlett-Packard (Avondale PA) mass selective detector (HP5973 MSD) in negative ion mode. Ion source, quadruple and transfer liner were held at 150, 100, and 250 °C, respectively. Methane of ultrahigh purity was used as an ionization gas at a pressure of 1.9 × 10-4 Torr in the ion source. Separation was performed with a HP5 column (30 m × 0.25 mm i.d.; film thickness, 0.25 mm) using a temperature program optimized earlier by Glassmeyer et al. (26) and splitless injection. The following ions, corresponding to fragment ions (M - Cl)-, (M - Cl + 2)-, and (M - Cl + 4)- for the hexa- (m/z 307, 309, and 311), hepta (m/z 341, 343, and 345), octa- (m/z 375, 377, and 379) and nonaCHBs (m/z 411 and m/z 413) were monitored. [13C-γ]chlordane was used as an internal performance standard. Quantitative analysis of total toxaphene and homologues was carried out using an external standard of “Hercules” technical toxaphene (Ultra Scientific). In addition, 20 individual hexa- to nonachlorobornane congeners were determined using standards from Dr. Ehrenstorfer (Augsburg Germany) or Promochem (Wesel, Germany) of which 12 were detected in one or more slices (B6-923, B7-1001, P25, B81413, P31, B8-789, B8-531, B8-1414/1945, B8-809, B8-2229, B9-1679, and B9-715). For additional information on nomenclature, see ref 1. In brief, the letter “B” refers to chlorobornane and the number refers to Cl substitution. The 4-digit number codes for chlorine positioning are according to the method of Andrews and Vetter (27). Some congeners continue to be referred to by their “P” or Parlar numbers (28) because structures are unknown. PCB congeners (104 peaks including coeluting congeners) were quantified by highresolution GC with 63Ni electron capture detection (24, 25) using a Hewlett-Packard 6890 GC equipped with a 30 m × 0.25 mm (0.25 mm film thickness) DB-5 column (J&W Scientific) with H2 carrier gas and electronic pressure control. VOL. 35, NO. 7, 2001 / ENVIRONMENTAL SCIENCE & TECHNOLOGY
9
1313
FIGURE 2. (A) Total toxaphene chronological profiles from the dated Lochnagar sediment core. (B) Toxaphene production in East Germany (1955-1990) (10) and in the United States (interpolated from ref 30). (C) Total PCBs in the same core are shown for comparison. (D) The U.K. PCB production taken from ref 38. Quality assurance included the analysis of laboratory blanks consisting of all reagents and a certified reference sediment (EC5, National Laboratory for Environmental Testing, Burlington, ON) every 6 samples. In addition, sediment core slices dated to pre-1900 were analyzed as a further check on background levels. All silica gel chromatography was conducted in a clean room equipped with carbon and HEPA-filtered air. Method detection limits (MDLs) for individual chlorobornane congeners were approximately 0.01 ng/g (dry wt; dw) and 0.1 ng/g dw for total toxaphene and homologue groups. MDLs for PCB congeners ranged from 0.02 to 0.2 ng/g dw with the median of 0.04 ng/g dw.
Results and Discussion The sediment core showed a uniform profile of excess 210Pb profile with cumulative dry weight (23). The sedimentation rate, calculated using the Constant Rate of Supply model (29), was estimated to be 0.012 ( 0.002 g cm-2 yr-1. The historical concentrations of total toxaphene and the hexa- to nonachloro-homologue groups are shown in Figures 2 and 3, respectively. Total concentrations are primarily driven by the hepta group with lesser inputs from the octahomologues. The other groups make only minor contributions. Prior to 1960, total concentrations were low but uniform at all analyzed sediment levels. A dramatic increase in concentration occurred in the 1960s, stabilizing for a period between the mid-1960s and the mid-1980s before reaching a peak in 1990 and then declining to the sediment surface. Rice and Evans (30) reported production data for toxaphene (Figure 2), and Eisenreich and Rapaport (31) produced a source curve for toxaphene in their study on U.S. and Canadian peat cores. The East German toxaphene production data reported by Heinisch et al. (10) can also be used as an indicator of European sources although, as noted previously, much of the production was exported. To our knowledge, 1314
9
ENVIRONMENTAL SCIENCE & TECHNOLOGY / VOL. 35, NO. 7, 2001
there are no toxaphene sediment profiles outside of North America against which to compare our data. As the United States was the major user of toxaphene, it might be assumed that profiles and source curves from North America represent global patterns. Models using U.S. data calculate that peak atmospheric input for toxaphene occurred in 1975 (31, 32), and temperate climatic zone lake sediment concentrations should maximize around 1978 (32). This is in reasonable agreement with the dated sediment records of untreated lakes in the United States and Canada (33, 34) as well as at sites in the Great Lakes at which nonatmospheric inputs form a major component (35). At Lochnagar, there appears to be a bimodal distribution with peaks in both the 1970s and 1990. The earlier peak shows agreement with the U.S. source curve and hence may correspond to the modeled global input patterns, whereas the latter may reflect the status of toxaphene usage in Europe. East German production peaked in 1982 at 2630 t and declined to 434 t in 1990 (10). As late as 1992, toxaphene was still neither banned nor restricted in Spain and Hungary and only restricted in Poland and some other countries of eastern Europe (9). These are areas from which it is known that atmospheric pollutants can impact on this region of the U.K. (35, 36). Additional sources may be from lower latitudes, as it is reported that toxaphene continues to be used in some African countries (1). The sediment record of total PCBs is shown in Figure 2 for the same Lochnagar sediment core. PCB concentrations begin to increase in levels dated to the mid-1930s and undergo a large increase during the 1960s. This increase coincides very well with the known production of PCBs in the U.K. (37). The trend of PCBs with age of the core slices also agrees well with a recent study of PCB deposition to Esthwaite Water, a semi-rural lake in northwest England about 295 km directly south of Lochnagar (38). Unlike the toxaphene profile, PCBs show a more uniform distribution with depth between 1960
FIGURE 3. Toxaphene homologue profiles in the dated Lochnagar sediment core and proportions of the homologues in recent and older slices. The proportions of the homologues in technical toxaphene (Hercules) standard (34) are presented for comparison.
TABLE 1. Reported Lake Sediment Toxaphene Data for Comparison with Lochnagara surface surface toxaphene b concn max concn accum rate (ng/g dw) (ng/g dw) (ng cm-2 yr-1)
site
region
core date
Lochnagar L382 L375 Far Lake Hawk Lake Amituk Lake Hazen Lake Kusawa Lake Siskiwit Outer Island Laberge 2 Laberge 3 Fox Lake Lake Michigan
Scotland, U.K. northwest Ontario northwest Ontario Canadian Arctic Canadian Arctic Canadian Arctic Canadian Arctic Yukon Great Lakes Canadian Arctic Yukon Yukon Yukon Great Lakes
1997 1988 1989 1988 1988 1989 1990 1992 1991 1991 1992 1992 1993 1991-1992
14.0 2.6 5.3 7.4 17 2.3 0.9 0.61 8.9 4.0 0.28 0.07 1.3 15-45
40.1 18 5.8 10 19 2.3 0.9 0.64 18 4.9 0.53 0.56 2.0 20-48
0.42 0.039 0.07 0.07 0.08 0.0095 0.012 0.003 0.25 0.065 0.023 0.005 0.016 0.24-1.01
Lake Superior
Great Lakes
1991-1994 2.8-15
2.8-29
0.097-0.14
Hanson Lake Peanut Lake Chatwin Lake
Yukon, Canada Alberta, Canada Alberta, Canada
1994 1985 1990
8000 500 1602
3.4 2.3 1.7
600 112 53
max toxaphene accum rate (ng cm-2 yr-1); date 1.6; 1990 0.19; 1980 0.08; 1984 0.09; 1970 0.10; 1961 0.0095; 1989 0.012; 1990 0.018; 1979 0.51; 1971 0.072; 1972 0.043; 1975 0.041; 1971 0.024; 1981 0.32-1.07; 1969-1989 0.097-0.25; 1974-1990 24 10 51
inventoryc toxaphene (ng cm-2) source
ref
18.4 a a a a a a 0.6 11.7 2.8 1.6 1.5 0.8 7.5-43
d d d d d d d d d d e e e e
this study 24 24 24 24 24 24 11, 49 34 34 11, 49 11, 49 11, 49 34
3.1-7.9
e
34
a a a
f f f
41,42 8 8
a Data not reported. b All toxaphene results by GC-electron capture negative ion mass spectrometry except for Canadian Arctic and Northwest Ontario lakes (L382, L375, Hawk, Far, Amituk, and Hazen), which were by GC-electron capture detection, a less specific technique. c Inventories reported as calculated by respective authors. For this study: inventory ) total mass/cm2 of toxaphene in post-1900 slices. d Atmospheric only. e Atmospheric + some riverine. f Treated.
and the mid-1990s. Following a maximum in the slice with median age of 1973 ( 2 yr, PCB levels decline with a 50% disappearance time of 20 ( 7 yr between 1973 and 1993. The PCB profiles in dated sediment cores collected in remote lakes in northern Finland (68-69° N) also show post-1950 deposition of PCBs with maxima in the 1970s (39). A peat core from an ombrotropic bog in northwest England also showed maximum PCBs in slices dated to the late 1960s and an approximately 50% decline in concentrations between
1970 and 1990 (40). The historical record of PCBs in Lochnagar is thus in good agreement with other sediment cores from the U.K. and northern Europe, with a peat core and with production data. It is quite likely therefore that Lochnagar is representative of the historical regional deposition of other persistent organochlorines with similar semivolatile properties, such as toxaphene. Surface sediment concentrations of toxaphene were 14 ng/g dw in the Lochnagar sediment core. This is at the upper VOL. 35, NO. 7, 2001 / ENVIRONMENTAL SCIENCE & TECHNOLOGY
9
1315
FIGURE 4. Historical concentration profiles (ng/g dry wt) of major individual congeners in the dated Lochnagar sediment core. B6-923 (Hx-Sed) and B7-1001 (Hp-Sed) are dechlorination products of other higher chlorinated congeners. Octachlorobornanes B8-789 and B8-2229 and nonachlorobornanes B9-1679 and B9-715 were the most prominent individual toxaphene congeners.
TABLE 2. Change in Abundance of Major Individual CHB Congeners Relative to the Sum of 12 Major Chlorobornanesa along with Changes in Ratios of o,p′- and p,p′-DDD to Total DDTb in Lochnagar Sediments congenerc
time period (yr)
trend description
abundance ratio doubling time or half-life ( SEd
R2
probabilitye
B6-923 B7-1001 B8-1413 (P26) B8-789 (P38) B8-1414/1945 (P40/41) B8-806/809 (P42) B8-2229 (P44) B9-1679 (P50) B9-715 (P58) o,p-DDD5 p,p-DDD5
0-49 0-35 0-35 0-20 0-39 0-39 0-44 0-35 0-35 3-35 3-35
increase increase decline no significant trend increase increase no significant trend weak decline weak increase increase increase
17 ( 3 12 ( 1 -16 ( 5 -14 ( 10 38 ( 9 39 ( 13 62 ( 40 -23 ( 11 29 ( 14 41 ( 5 25 ( 4
0.707 0.925 0.541 0.284 0.648 0.481 0.183 0.324 0.335 0.885 0.860
