assessment process because bioaccumulation information is a key element in
assessing risk. Therefore, this category receives a moderate ranking in Table 1.
Sediment bioaccumulation tests, can, however, be used to directly and easily
measure the potential for bioaccumulation of these compounds from sediment.
Bioavailability of contaminants from water for ecological receptors.
Adsorption of hydrophobic COCs onto DOC in surface waters reduces
bioavailability (Landrum et al. 1987; McCarthy and Jimenez 1985). Kinetic
limitation to accumulation of hydrophobic COCs into phytoplankton may reduce
bioavailability of COCs from water (Swackhamer and Skoglund 1993).
Partitioning of hydrophobic COCs onto DOC can be easily estimated, if
measurements of the concentration of DOC are available. However, kinetic
limitations to uptake in phytoplankton are more difficult to predict. Because of
this uncertainty in the uptake kinetics, this category is given a ranking of
moderate uncertainty that is moderately difficult to quantify.
Exposure models for aquatic organisms: model uncertainty. Several
models are available for predicting concentrations of contaminants in tissues of
exposed organisms from concentrations in the environment. Two models that are
often used to predict tissue concentrations from concentrations in sediment are
the models of Gobas (1993) and Thomann (1989). These models estimate steady-
state concentrations of hydrophobic organic substances in various aquatic
organisms. In these models, fish absorb chemicals directly from water via the
gills and through the consumption of food via the gastrointestinal tract.
Chemicals are lost via the gills to the surrounding water, via egestion of fecal
matter, and by metabolism. Growth of the fish also results in dilution of tissue
concentrations and is modeled as a loss process.
Presently, why the ability to metabolize organic xenobiotics varies among
taxonomic groups is not understood (Nebert, Nelson, and Feyereisen 1989), and
few models can estimate the rate of metabolism and elimination of these
contaminants. For some persistent organochlorine chemicals, the rate of
metabolism is small compared to the rate of uptake or elimination via other
routes, and metabolism can be ignored. For other chemicals, such as PAHs,
which are readily transformed by fish, use of the Gobas (1993) and Thomann
(1989) models without consideration of metabolism is not recommended. This is
a large source of uncertainty in the use of these models in risk assessment. Risk
to a receptor could be overestimated if actual body burdens of PAHs, for
example, were lower than predicted due to metabolism. Alternatively, risk could
be underestimated when metabolites accumulate in tissue and exert a greater
toxic effect than the unmetabolized parent compound.
a. Equilibrium models and time-varying conditions. Equilibrium conditions
are often used to predict bioaccumulation of contaminants from sediment
and water into aquatic organisms, such as in the Gobas (1993) and
Thomann (1989) models. These steady-state models provide an
analytical solution to time-varying models under the assumption that
contaminant concentrations are constant (dC/dT = 0). These models may
be appropriate for estimating long-term expected concentrations of
contaminants, but will not be suitable for episodic events.
Chapter 5 Uncertainty in Tier IV Risk Assessments