Measuring the strength of demographic stochasticity

According to a well-established paradigm in conservation biology, demographic stochasticity as such is an improbable cause of extinction in populations larger than 100 individuals. However, the strength of demographic stochasticity depends not only on population size, but on the life history characteristics of the population as well. Consequently, neglecting demographic stochasticity is not appropriate without specific justification. This paper introduces two new measures, demographic effective population size N-D and strength of demographic stochasticity D. Thus, the rule of 100 individuals being relatively safe from extinction caused by demographic stochasticity is reformulated by considering the effective population size instead of simply counting the number of individuals. Also, the measures render it possible to compare vulnerabilities to extinction caused by different types of life histories. Five examples, taking advantage of the theory of branching processes and Markov chains, illustrate the use of the measures. (C) 1996 Academic Press Limited