Effect of stage-specific vital rates on population growth rates and effective population sizes in an endangered iteroparous plant.

Effective population size (N(e)) determines the strength of genetic drift and can influence the level of genetic diversity a population can maintain. Assessing how changes in demographic rates associated with environmental variables and management actions affect N(e) thus can be crucial to the conservation of endangered species. Calculation of N(e) through demographic models makes it possible to use elasticity analyses to study this issue. The elasticity of N(e) to a given vital rate is the proportional change in N(e) associated with a proportional increase in that vital rate. In addition, demographic models can be used to study N(e) and population growth rate (λ) simultaneously. Simultaneous examination is important because some vital rates differ diametrically in their associations with λ and N(e). For example, in some cases increasing these vital rates increases λ and decreases N(e). We used elasticity analysis to study the effect of stage-specific survival and flowering rates on N(e), annual effective population size (N(a)), and λ in seven populations of the endangered plant Austrian dragonhead (Dracocephalum austriacum). In populations with λ ≥ 1, the elasticities of N(e) and N(a) were similar to those of λ. Survival rates of adults were associated with greater elasticities than survival rates of juveniles, flowering rates, or fecundity. In populations with λ < 1, N(e) and N(a) exhibited greater elasticities to juvenile than to adult vital rates. These patterns are similar to those observed in other species with similar life histories. We did not observe contrasting effects of any vital rate on λ and N(e); thus, management actions that increase the λ of populations of Austrian dragonhead will not increase genetic drift. Our results show that elasticity analyses of N(e) and N(a) can complement elasticity analysis of λ. Moreover, such analyses do not require more data than standard matrix models of population dynamics.