Hunt and Bischoff provided the first data demonstrating progressive bioaccumulation and increases in concentrations of the chlorinated insecticide DDD through an aquatic food web. DDD was applied to Clear Lake, California during three administration events in 1949, 1951, and 1957. Administrations were designed to achieve a nominal concentration of DDD in water of 50mgl~ , although reportedly water residues never achieved such levels. Mortalities of fish-eating birds were observed within months after the second and third applications, with the population of western grebes decreasing from 1000 pairs prior to DDD administration to less than 30 pairs in 1960. Food web sampling and residue analysis indicated phytoplankton achieved concentrations of approximately 5mgg~\ pelagic fish contained between 50 and 300 mgg-1 and a brown bullhead contained 2500 mgg-1 of DDD. DDD concentrations in western grebes and California gulls were reported at more than 2000 mgg~ . Soon after, other studies began documenting DDT bioaccumulation in different food webs. Woodwell et al. determined DDT concentrations in water, soil, plankton, invertebrates, mussels, fish, and fish-eating birds in a salt marsh south of Long Island, New York. DDT increased from 0.04mgg_1 in plankton to 75mgg_1 in ring-billed gulls. Plankton concentrations were 800-fold higher than residues measured in water. Invertebrates and fish exhibited intermediate concentrations of DDT compared to plankton and birds, consistent with their trophic status. Unfortunately, the above studies did not determine lipid concentrations of samples submitted for insecticide residues. As such, these data could not be used to test the thermodynamic criteria associated with equilibrium partitioning theory and biomagnification.
Advancements in analytical technology in the 1980s, particularly with the development of capillary gas chro-matography columns, greatly increased the ability of environmental scientists to examine individual chemical concentrations in more complex field matrices. This led to a plethora of food web data sets documenting biomagnification of other organic contaminants including polychlorinated biphenyls (PCBs). Two major studies documenting food web biomagnification of individual PCB congeners were published in 1988. Oliver and Niimi measured individual PCB concentrations in water, sediments, amphipods, slimy sculpin, alewife, and lake trout from Lake Ontario. The authors also measured lipid contents in the biological samples allowing them to directly test predictions of the equilibrium partitioning theory. Their data demonstrated increases in lipid-nor-malized PCB concentrations with increasing trophic status (see Figure 1). Salmonids were also shown to have fivefold higher lipid-normalized concentrations than predicted from equilibrium partitioning theory based on residues in water. Herring gulls from the same lake collected 1 year later in another study demonstrated lipid-normalized PCB concentrations that were tenfold higher than measured for salmonids by Oliver and Niimi. Conolly and Pederson also demonstrated that the fugacity ratio of rainbow trout/water exceeded a value of 1 for PCBs having a log KqW value of 4 and greater in Lake Ontario. The authors demonstrated that the trout/water fugacity ratio for PCBs increased with increasing chemical K0W up to values from 10 to 100 for PCBs having log K0W values of 6 and higher. The same authors also demonstrated progressive increases in the animal/water fugacity ratio for PCBs with animal trophic status in the Lake Michigan food web. PCB animal/water fugacity ratios ranged from 3 to 5 for white fish and chub occupying a trophic level of 2 and up to a value of 14 for fish occupying a trophic level of 4. Similar case studies of food web biomagnification using lipid-normalized data sets have subsequently been demonstrated in several other aquatic systems including all five Great Lakes, Lake Baikal, and in agricultural and arctic terrestrial ecosystems.
Other data sets have shown the relationship between hydrophobic organic contaminant bioaccumulation and food chain length or number of trophic steps within the system. Using data generated from the Ontario sport fish contaminant surveillance program, Rasmussen et al. examined PCB bioaccumulation in lake trout from a large number of lakes in Ontario, Canada. The authors demonstrated that planktivorous lake trout from lakes lacking suitable forage fish exhibited lower PCB bioaccumulation compared to piscivorous lake trout from lakes containing forage fish. Finally, lake trout from lakes containing both forage fish and mysids achieved the highest contaminant residues. Lakes were also categorized and analyzed by location to remove confounding factors associated with different loading sources to individual systems. The authors attributed the lake to lake differences in PCB bioaccumulation by lake trout to reflect differences in food chain length. A more recent study documented enriched toxaphene bioaccumulation in fish from Lake
Labarge, Yukon Territory, Canada as contrasted with other subarctic lakes which showed much lower bioaccumulation of the same contaminants in fish. Lake Labarge was isolated from known pollutant sources and thought to receive most of its inputs via atmospheric deposition. Water concentrations of toxaphene were also found to be similar in Lake Labarge relative to the other lakes which showed lower toxaphene bioaccumulation in fish. The major difference noted for Lake Laberge lake trout, burbot, and lake whitefish was that fish from this lake were feeding at higher trophic levels as revealed both by diet analysis and trophic enrichment of stable nitrogen isotopes. This study, similar to that of Rasmussen's work provided the empirical linkage between ecosystem structure, number of trophic links, and magnitude of biomagnification realized in top predator fish.
Was this article helpful?