全球变化知识资源中心

Aim We use current models of species distribution to predict the future habitat suitability for an endangered, keystone rodent (Dipodomys ingens, giant kangaroo rat). We incorporate the possibility of local adaptation and novel competitive interactions to improve the predictive accuracy of our Maxent models. Location San Joaquin Valley, California, USA. Methods We created Maxent models of two isolated populations of D. ingens. Using local surveys and state-wide data, we also modelled California ground squirrels (Otospermophilus beecheyi), a potential novel competitor. Models included landscape variables (slope and soil composition) and climate variables (temperature, precipitation and climatic water deficit). We used Warren's I to evaluate niche overlap between species as well as between isolated populations of the same species. We then projected each set of models into the future to evaluate D. ingens response to climate change. Results Niche overlap between the D. ingens populations was moderate (I = 0.44), suggesting that they already experience different climatic regimes and providing support for population-level modelling. Projecting individual populations separately into the future, under a high emission climate change scenario (CCSM4, rcp8.5), resulted in less predicted range contraction than modelling the species as a whole. However, forecasted distributions showed areas of increasing niche overlap between the Panoche population of D. ingens and O. beecheyi (I = 0.63 to 0.74), indicating competition could be a novel range limit. Main conclusions Modelling the distributions of isolated populations separately provides a more accurate estimation of the species' contemporary niche, capturing not just overlap between the populations, but individually suitable areas as well. Projecting the individual contemporary and historical models into the future allows us to estimate future habitat suitability. Predicting the distribution of a novel competitor characterizes areas of potential competition. Subsequently, management can target suitable areas for preservation and avoid areas of potential novel competition.