Home – Home – Jornals – Contemporary Problems of Ecology 2020 number 5

Accurate predictions of species’ range shifts in response to ongoing climate change is, in our opinion, a key starting point in an assessment of a species vulnerability to climate change. In this study, we used Maxent 3.4.1 to construct a Daurian pika climatic niche distribution model using 273 species presence points throughout its whole distribution range, together with 19 bioclimatic layers, and examined the pika vulnerability to the effects of global climate change under two emission scenarios RCP 2.6 and RCP 8.5. It was found that the models chosen using statistical selection criteria (AUC and AIC) displayed reduced transferability to new territories, which can substitute for time in predicting climate-change effects on biodiversity. To obtain more transferable model we used cross-validation in which data were assigned non-randomly to 3 groups that were spatially distinct. The constructed model was validated using two completely independent field-collected datasets. (1) Of the three unexplored previously areas with optimal climatic conditions, only one was inhabited by the pika. Other areas were revealed to have edaphic factors unsuitable for pika living. (2) Comparison of range projections for time periods of 140-120 Ka, 21 Ka and 6 Ka with pika fossil locations in the corresponding periods showed good fitting. The climatic niche range changes in the year 2070 do not look critical for Ochotona dauurica due to the significant overlap between current and predicted ranges. However, the internal structure of species’ range is predicted to change significantly, especially pronounced in the RCP 8.5 scenario, when the largest deviation from the current distribution is expected along with an increase in fragmentation within the predicted range