资源环境科技发展态势分析平台

Forecasting the impacts of climate change on species persistence in diverse natural communities requires a way to account for indirect effects mediated through species interactions. In particular, we expect species to experience major changes in competition as they track favorable climates. Here, we combine experimental data with a recently developed theoretical framework based on coexistence theory to measure the impact of climate‐driven range shifts on alpine plant persistence under climate change. We transplanted three co‐dominant alpine perennial species to five elevations, creating a maximum of 5°C increase in average growing‐season temperature. We statistically modeled species' demographic rates in response to the environment and interpolated species' intrinsic ranges—the environmental mapping of reproduction in the absence of competition. We used low‐density population growth rates—species' initial rate of invasion into an established community—as a metric of persistence. Further analysis of low‐density growth rates (LGRs) allowed us to parse the direct impacts of climate change from indirect impacts mediated by shifting competition. Our models predict qualitatively different range shifts for each species based on the climate conditions under which growth rates are maximized and where they are zero. Overall, climate change is predicted to increase the intrinsic (competition free) growth rates of all species, as warmer and wetter conditions increase the favorability of alpine habitat. However, these benefits are entirely negated by increased competition arising from greater overlap between competitors in their intrinsic ranges. Species were highly dispersal limited, which can prevent species from tracking shifting intrinsic ranges by reducing population spread rates. Yet dispersal limitation also promoted species' persistence because it promotes persistence mechanisms. Our study demonstrates the complex pathways by which climate change impacts species' persistence by altering their competitive environment, and highlights how a persistence framework based on LGRs can help disentangle impacts.