Tj. Iannuzzi et al., DISTRIBUTIONS OF KEY EXPOSURE FACTORS CONTROLLING THE UPTAKE OF XENOBIOTIC CHEMICALS IN AN ESTUARINE FOOD-WEB, Environmental toxicology and chemistry, 15(11), 1996, pp. 1979-1992
A critical evaluation of literature on the behavior physiology, and ec
ology of common estuarine or organisms was conducted in an attempt to
develop probabilistic distributions for those variables that influence
the uptake of xenobiotic chemicals from sediments, water, and food so
urces. The ranges, central tendencies, and distributions of several ke
y parameter values were identified for dominant organisms from various
trophic levels, including the polychaete Nereis virens, mummichog (Fu
ndulus heteroclitus), blue crab (Callinectes sapidus), and striped bas
s (Morone saxatilis). The exposure factors of interest included ingest
ion rate for various food sources, growth rate, respiration rate, excr
etion rare, body weight, wet/dry weight ratio, lipid content, chemical
assimilation efficiency, and food assimilation efficiency. These expo
sure factors are critical to the execution of mechanistic food web mod
els, which, when properly calibrated, can be used to estimate tissue c
oncentrations of nonionic chemicals in aquatic organisms based on know
ledge of the bioenergetics and feeding interactions within a food web
and the sediment and water concentrations of chemicals. In this articl
e we describe the use of distributions for various exposure factors in
the context of a mechanistic bioaccumulation model that is amenable t
o probabilistic analyses for multiple organisms within a food web. A c
ase study is provided which compares the estimated versus measured con
centrations of five polychlorinated biphenyl (PCB) congeners in a repr
esentative food web from the tidal portion of the Passaic River, New J
ersey, USA. The results suggest that the model is accurate within an o
rder of magnitude or less in estimating the bioaccumulation of PCBs in
this food web without calibration. The results of a model sensitivity
analysis suggest that the input parameters which most influence the o
utput of the model are both chemical and organism specific.