Page 666 - The Toxicology of Fishes

646 The Toxicology of Fishes

by fish and other organisms by diverse routes such as respiration, food ingestion, and dermal absorption.
It is important to determine and quantify the dominant routes of uptake that control the dose received
by the fish, which in turn ultimately controls the effect of the toxicant. For most organic toxicants, the
chemical is absorbed from food or respired water into the bloodstream and becomes distributed among
organs and tissues, especially those of high lipid content. Metals, on the other hand, may specifically
partition to respiring surfaces, resulting in interference with critical transport processes.
When a toxicant is suspected of having adverse effects on a fish population, a standard response is
to sample and analyze the water or sediment and infer the likelihood of effects by comparing measured
concentrations with those known to cause adverse effects in laboratory exposures or bioassays. Unfor-
tunately, such analyses are often fraught with difficulties. First, a satisfactory analytical method for
measuring the concentration may not be available; for example, only recently have methods become
available to analyze the alkyl phenolic substances described by Tyler in Chapter 25. Appreciable spatial
and temporal variability in concentrations may occur due to the nature of the environment. Also, such
analyses may not adequately discriminate between a truly dissolved organic substance and that which
is associated with dissolved matter such as fulvic acids. These two different forms found in the envi-
ronment are likely to differ in bioavailability. Finally, chemical analyses of these type can be prohibitively
expensive for the amount of useful information gained.
When the presence of a toxicant is established and approximate levels are known, it may be useful
to supplement measurements by mass balance models, which provide a basis for calculating the con-
centrations, partitioning, and process rates in the aquatic system. These deductions should be validated
by comparing predictions with actual measurements, but in some cases the only method of estimating
concentrations is by calculation. A major benefit of compiling a comprehensive, validated mass balance
model is that it can provide a complete picture of the fate of the toxicant, including estimates of certain
quantities such as evaporation rates that cannot be measured directly. Such a model allows predictions
to be made of the effect of reducing loadings (namely, by how much) and how quickly concentrations
will fall. The model may help to explain seasonal variations of concentration, bioavailability, and
exposure. It is becoming increasingly evident that environmental management or regulation to restore
viable fish populations with minimal contamination is greatly assisted by having available a full,
quantitative description of the sources and fate of the toxicant and, accordingly, the actual exposure
experienced by aquatic biota.
A large number of models is available, many of which can be downloaded from the Internet and are
supported by agencies such as the U.S. Environmental Protection Agency (EPA). It is impossible to
describe these models in detail, so our focus here is to present and illustrate the basic principles that
apply in a simple single-box model and to outline briefly how this approach can be extended to treat
more complex systems. We emphasize that models are necessarily simplifications of a complex reality;
thus, they are no substitutes for direct measurement. The equations inherent in the model represent the
modeler’s best attempts to express the phenomena quantitatively, but these expressions may be incomplete
or flawed. All models rely on estimates of input parameters such as temperature, flow rates, rate constants
for degrading reactions, and partition coefficients expressing equilibrium concentration ratios between
phases. All are subject to variation, error, and uncertainty. The result is that the model output is rarely
more accurate than a factor of 2 or 3. The modeler has an obligation to convey this uncertainty to the
user to avoid over-interpretation of the results or over-reliance on the findings. Models are entirely
complementary to monitoring efforts, not a substitute for them. It is, however, very satisfying when
agreement between the two occurs, as this suggests that the system is both reliably monitored and
quantitatively understood.

A Historical Note

It is appropriate to provide a brief historical perspective on mass balance modeling of aquatic systems
and to acknowledge the pioneering efforts of early workers in this field. These models date back to the
1920s, when it was recognized that the discharge of degradable organic wastes into rivers was depleting
dissolved oxygen levels and causing fish mortality. The studies by Streeter and Phelps (1925) were the