Caged Fish for Estimating Concentrations of Trace Substances in Natural Waters

Abstract

This paper attempts to establish a theoretical background for using caged fish as estimators of concentrations of biologically accumulable trace substances in natural waters. The major motive for using this monitoring technique is that, for most trace substances only a certain, and usually unknown, fraction of the total concentration can be expected to be biologically available and that, therefore, conventional analytical methods may yield values that are biologically irrelevant. It is shown that, in general, if (i) the metabolic rate of the experimental fish is constant, (ii) the turnover factor of the trace substance in the fish is numerically known and either constant or numerically small, and (iii) the efficiency of oxygen withdrawal is constant and numerically known, caged fish can be used to estimate the mass concentration (relative to oxygen) of accumulable trace substances in the ambient water. As illustration, the concentration of methyl mercury that is biologically available in a lake is estimated, yielding figures of the order of 1 ng l-−1. The precision problem is discussed with regard to optimal experimental design and selection of experimental fish.