Global circulation models predict not only future changes in mean temperature and precipitation, but also increased climate variability. Increased environmental variability will likely increase the variability of survival and recruitment (i.e., vital rates), often resulting in decreased population growth. Life history theory and stochastic demography can aid in forecasting the ecological impacts of climate change on natural populations. As a first step towards understanding how plant populations will respond to projected climate change, we used a long-term dataset collected in mixed-grass prairie to construct stochastic matrix models for ten forbs and conducted elasticity analyses to examine the potential effects of changes in climate means (E^sμ) versus changes in climate variability (E^sσ) on the long-term stochastic population growth rate, λ_s. We also conducted a Life Table Response Experiment to examine how survival and recruitment contributed to variation in past population growth. Our objectives were to (1) determine the relative contribution of survival and recruitment to λ_s, (2) determine the impact of variability in vital rates to λ_s, and (3) to synthesize the results across species using life history traits.

Results/Conclusions

Elasticity and contribution values were higher for recruitment than for survival in all species. Nevertheless, elasticity and contribution values for recruitment decreased with generation time, indicating recruitment was relatively more important for population growth in shorter-lived species. Both E^sμ and | E^sσ | for recruitment decreased with generation time. Total elasticity to variance (T^σ) increased with increasing mean and increasing variability in recruitment across species. Our results indicate that longer-lived plants should be less responsive to changes in climate mean and variability. In addition, plants with a high recruitment rate and/or high variability in recruitment will be most affected by changes in climate variability.