Genetically Effective Population Size of Large Mammals: An Assessment of Estimators

Abstract: We calculated genetically effective population sizes (N e ) for simulated grizzly bear (Ursus arctos) populations by tracing the loss of heterozygosity through time and compared them with estimates of N, produced by applying published formulas to demographic results from the simulation. Estimates of N e using different formulas on identical data varied wadely. Equations published by Hill (1972), and modifications of those used by Ryman et al. (1981) and Reed et al. (1986), provided the most accurate estimates Minor population fluctuations had little effect on N e but variation in lifetime reproductive success among mules (V km ) greatly reduced N, from its expectation under random mating success All methods to estimate N e for populations with complex demographics require extensive data, but estimates for V km in polygynous species are especially difficult to obtain. We suggest that simulation modeling may provide alternative methods to estimate V km and N e