Variation among Individuals and Reduced Demographic Stochasticity.

Abstract

Population viability analysis ( PVA) is a technique that employs stochastic demographic models to predict extinction risk. All else being equal, higher variance in a demographic rate leads to a greater extinction risk. Demographic stochasticity represents variance due to differences among individuals. Current implementations of PVAs, however, assume that the expected fates of all individuals are identical. For example, demographic stochasticity in survival is modeled as a random draw from a binomial distribution. We developed a simple conceptual model showing that if there is variation among individuals in expected survival, then existing PVA models overestimate the variance due to demographic stochasticity in survival. This is a consequence of Jensen's inequality and the fact that the binomial demographic variance is a concave function of mean survival. The effect of variation among individuals on demographic stochasticity in fecundity depends on the mean-variance relationship for individual reproductive success, which is not presently known. If fecundity patterns mirror those of survival, then variation among individuals will reduce the extinction risk of small populations.