Capturing response differences of species distribution to climate and human pressures by incorporating local adaptation: Implications for the conservation of a critically endangered species

Considering local adaptation has been increasingly involved in forecasting species distributions under climate change and the management of species conservation. Herein, we take the critically endangered Chinese giant salamander (Andrias davidianus) that has both a low dispersal ability and distinct population divergence in different regions as an example. Basin-scale models that represent different populations in the Huanghe River Basin (HRB), the Yangtze River Basin (YRB), and the Pearl River Basin (PRB) were established using ensemble species distribution models. The species ranges under the future human population density (HPD) and climate change were predicted, and the range loss was evaluated for local basins in 2050 and 2070. Our results showed that the predominant factors affecting species distributions differed among basins, and the responses of the species occurrence to HPD and climate factors were distinctly different from northern to southern basins. Future HPD changes would be the most influential factor that engenders negative impacts on the species distribution in all three basins, especially in the HRB. Climate change will likely be less prominent in decreasing the species range, excluding in the YRB and PRB under the highest-emissions scenario in 2050. Overall, the high-emissions scenario would more significantly aggravate the negative impacts produced by HPD change in both 2050 and 2070, with maximum losses of species ranges in the HRB, YRB, and PRB of 83.4%, 60.0%, and 53.5%, respectively, under the scenarios of the combined impacts of HPD and climate changes. We proposed adapted conservation policies to effectively protect the habitat of this critically endangered animal in different basins based on the outcomes. Our research addresses the importance of incorporating local adaptation into species distribution modeling to inform conservation and management decisions.