Page 667 - The Toxicology of Fishes

Exposure Assessment and Modeling in the Aquatic Environment 647

first to characterize the process quantitatively in the form of the oxygen sag equation, which is essentially
a riverine oxygen mass balance. Later, these principles were applied to lakes and estuaries and to other
substances, notably hydrophobic organics, pesticides, metals, radionuclides, and nutrients. Much of the
pioneering work in North America was done at Manhattan College in New York by O’Connor and his
colleagues, notably Mueller, Thomann, Di Toro, and Connolly. Thomann (1998) reviewed the evolution
of these models as they have grown in scope, complexity, and utility. The models now routinely treat
water, sediment, the atmosphere, and aquatic food webs by segmenting the system into numerous
connected boxes and allowing for time-dependent changes in hydrodynamics and temperature. Episodic
events such as spills can be simulated. Thomann (1998) suggested that we are now entering a “golden
age” of aquatic modeling with the potential to contribute to the rational and effective management of
aquatic and marine systems. Currently available models are designed to treat organic substances, metals,
nutrients, and radionuclides. They vary greatly in complexity. They can be applied to lakes, rivers,
wetlands, estuaries, groundwater, and oceanic systems. Some include estimates of bioaccumulation and
toxicity. They all have the common objective of seeking to describe or simulate the fate of the chemical
in an aquatic system.

Fundamental Concepts

The simplest aquatic system unit is a small lake that consists of a well-mixed body of water, underlain
by a layer of well-mixed sediment with an overlying atmosphere. Use of the concept of well-mixed
compartments enables the amount of toxicant to be characterized by a single concentration rather than
a distribution of values. This is obviously a simplification, but it greatly facilitates mathematical simu-
lation. The water column is usually treated as consisting of water, suspended solids, and, in some cases,
dissolved organic matter. Discrimination between dissolved and suspended is operationally defined by
the pore size of a filter (usually a fraction of a micrometer); thus, smaller particles such as fulvic and
humic acid agglomerations may be treated as being dissolved when in reality they are separate phases.
The bottom sediment is usually treated as a mixture of water, mineral solids, and organic matter.
Vertebrate and invertebrate biota may be present in both water and sediment. Finally, the atmosphere
above the lake is treated as consisting of air and aerosol particles with chemical transport facilitated by
rainfall.
Toxicants can move through this system via two general transport processes. Any toxicant associated
with particles in air, water, or sediment will move with these particles as they deposit, resuspend, or
flow with the water. Toxicant will also flow with the water in dissolved form. These are essentially
advection processes. In addition, the toxicant is in a continuous state of diffusion through the system,
with a tendency to approach thermodynamic equilibrium (i.e., equal chemical potential or fugacity).
Concentrations in water and air, for example, will tend to approach equilibrium as dictated by Henry’s
law. At all times, the substance is subject to transformation or degradation processes such as microbial
conversion, photolysis, hydrolysis, or oxidation in all media.
In most, but not all, cases the mass balance model can ignore biota (except for biodegradation) because
the mass of substance associated with biota is usually insignificant. Exceptions would include the
transport of chemicals associated with migrating salmon or bird guano. The concentration in biota can
be large multiples of that in water, but because the volume of biota is relatively small (e.g., 1 to 10 parts
in 1 million) the mass in biota is also small. This simplifies the model by allowing the dominant abiotic
fate processes to be deduced. Concentrations in biotic phases such as fish can be estimated later in a
separate calculation.
Figure 14.1 illustrates the processes to which a substance may be subjected in a simple unit lake.
Table 14.1 lists these processes and suggests the structure of the equations used to describe the equilibria
or kinetics. The modeler’s objective is to write equations describing all these processes in terms of the
local concentration of the substance, combine them, and then solve the overall mass balance. This can
be viewed as an accounting problem in which the aim is to balance the numerous inputs and outputs.
Different modelers write the various expressions describing the rates of these processes in different
ways; however, they share the common goal of quantifying all process rates accurately. These expressions