A SIMULATION-MODEL OF HOW LIFE-HISTORY STRATEGIES MEDIATE POLLUTION EFFECTS ON FISH POPULATIONS

Pollution effects on fish populations were estimated with a simulation model, using Leslie matrices. Results from changing only first-year s urvival rate (S(o)) have already been published (Schaaf et al. 1987). This paper explores the effects of perturbing both S(o) and the adult survival rate (S(i)) for 12 spatial-temporal stocks. Most stocks exami ned are more sensitive to permanent change in S(1) than to changes in S(o). The relative importance of these two rates in determining the po pulation growth rate (lambda) depends upon the age distribution of the expected lifetime egg production of age i females (V(i)). In turn, th e vector V(i), as measured by its mean and standard deviation, is show n to vary among geographic or temporal stocks of a single species. Hen ce, we quantify the impact on population size of destroying a fixed pe rcentage of habitat, relative to where and when it occurs (i.e., relat ive impact on S(i) and S(o)). For example, destroying 1% of the Atlant ic menhaden habitat would reduce lambda by 0.8% and the population by 8.0% in 10 yr, if the impact affected only adults (e.g., offshore in w inter). If the 1% habitat destruction all occurred in the estuaries, a ffecting juveniles as well, lambda would be reduced by almost 5%, and in 10 yr drive the population down to 58% of its former equilibrium. W e show that knowledge of the mean and variance of the age distribution of V(i) permits prediction of relative sensitivity among species to p ollution. Within species, this knowledge of V(i) permits comparison of the effects of impacting different life stages, and at different time s and places.