Theory in ecology and biogeography generally assumes that the demographic performance of species declines from the centers to the edges of species’ geographic ranges, ultimately setting their range limits. Recent meta-analysis of field studies, however, reveal this pattern to be surprisingly uncommon in plants. These results cast doubt on the relevance of important theoretical frameworks as well as major ongoing efforts to forecast climate-change-driven shifts in species distributions by up-scaling plot-level measurements of population dynamics. The reasons for this disparity between theory and empirical patterns are unclear. To address this issue, we combined measurements of individual-level demography from replicated common-garden experiments with large-scale surveys of metapopulation dynamics across the elevation ranges of two monkeyflower species in Mt. Rainier National Park (Washington, USA). We use these data to test how direct climate effects on individual performance combine with indirect effects of climate on population persistence to influence the range limits of these two taxa.
We found that climate and disturbance interact to shape the elevation range limits of our two focal species in ways that cause occupancy to decline towards range limits (as predicted from theory), while patch-level demographic performance does not. We also show that estimates of climate’s influence on individual-level demography or metapopulation dynamics in isolation are insufficient to explain the observed patterns of abundance and occupancy. Combining information from these two approaches using a hierarchical Bayesian model, however, produces equilibrium patterns of abundance and occupancy that closely match observations, as well as plausible scenarios of range change given different magnitudes of change in climate and disturbance frequency. Overall, our results suggest that taking demographic measurements at spatial and temporal scales that are too small to capture infrequent, extreme events that kill large numbers of individuals can lead to biased inference about the role of climate in shaping species ranges as well as the often-observed scenario where apparent demographic performance does not decline towards range limits. Measuring these processes at the appropriate scale for long-lived organisms like trees might require observations at scales larger than is feasible using traditional field-based approaches. Luckily, recent advances in remote-sensing technology can help overcome these obstacles.