On the relationships between bioassays and dynamics in chemically stressed, aquatic population models

Abstract

One purpose c fthis article isto synthesize some recentresults on the dynamics of mathematical models of chemically stressed aquatic populations and communities; in particular, we (1) illustrate some of the difficulties that might arise from extrapolation of bioassay results to dynamic, chemically stressed population and community models; and (2) indicate different ways in which chemicals can affect the dynamics of population models. Bioassays, an important component of ecological impact and risk assessment, can be. misleading if extrapolated to settings beyond experimental boundaries. Extrapolation of bioassays to the populations and community levels can not be direct because derived information is usually specific for a subset of individuals and obtained under experimental constraints on time and parameters. We present examples derived from a mathematical setting where consequences of bioassays, even when employed as the fundamental determinant of stress in the systeni, have no predictable relationship to the ultimate effect of the chemical on the system. The first illustration, at the population level, demonstrates that sublethal effects of a lipophilic chemical with a reversible mode of action on individuals attained at concentrations well below the LC50, indeed even below the EC50 for growth, can drive the population to extinction so that the chemically stressed population is much more severely damaged than predicted by hioassays. The second illustration at the community level indicates that results of bioassays can also indicate outcomes that (ire worse than actually occurs for the community. Finally, we compare the outcome of a spectral analysis oftime series of a sequence of chemically stressed populations, demonstrating that complex effects of lipophilic chemicals on population dynamics are not readily identifiahle from spectral signatures.