Demographic and environmental stochasticity

Abstract

This chapter defines and formulates demographic and environmental stochasticity, and illustrates statistical methods for estimating them from field data. Population fluctuations in most species are produced by demographic and environmental stochasticity, rather than by internally driven cycles or chaos. Demographic stochasticity results from chance independent events of individual mortality and reproduction, causing random fluctuations in population growth rate, primarily in small populations. Environmental stochasticity results from temporal fluctuations in mortality and reproductive rates of all individuals in a population in the same or similar fashion, causing population growth rate to fluctuate randomly in populations of all sizes. In populations much larger than the ratio of the demographic variance to the environmental variance, demographic stochasticity can be neglected. The demographic variance can be estimated from data on individual survival and reproduction, and using this, the environmental variance can then be estimated from population time series. Under density-independent growth in a random environment, the eventual rate of increase or decrease of population size is given by the long-run growth rate, the mean rate of increase of log population size, which is reduced by stochasticity. Density-dependent population growth in a stochastic environment produces temporal autocorrelation in population size, even in the absence of temporal autocorrelation in the environment. Time series for terrestrial populations of birds and mammals show little evidence of temporal environmental autocorrelation.