Original Research Article Using an ensemble modelling approach to predict the potential distribution of Himalayan gray goral ( Naemorhedus goral bedfordi ) in Pakistan

摘要

Global warming has negative impacts on the distribution of large ungulates, particularly for species occupying narrow distributional ranges. Knowledge of how climate change will affect future distributions is imperative for designing effective conservation action plans for at risk species such as the Himalayan gray goral (HGG), a cliff-dwelling mountainous goat. We sought to evaluate the potential distribution of Himalayan gray goral (HGG) under future climate change scenarios using ensemble modeling approaches. HGG data were obtained from previous published surveys, publications, and occurrence records ranging from 1985 to 2018. we also conducted survey in 2017–2018 using double observer method based on capture mark recapture. (Suryawanshi et?al., 2012; Tumursukh et?al., 2015). Later on we double check the record and remove double observation. After quality control screening, 139 records remained for analysis. Resulting species distribution models (SDMs) results showed sufficient internal evaluation metrics, with all TSS values being??0.7. The random forest (RF) modelling technique had on average the lowest true skill statistics (TSS) value, However the multivariate adaptive regression splines (MARS) modelling technique had the highest. The ensemble modelling internal evaluation metrics indicated adequate results with values ranging from 0.827 to 0.843. Annual mean temperature (Bio1) and annual precipitation (Bio12) were found to be the most important climatic variables impacting the potential distribution of HGG. HGG habitat determined to be suitable in both current and future climate scenarios decreased in all Representative Concentration Pathways (RCPs) scenarios with the exception of RCPs 2.6. Suitable habitat in both current and future climate scenarios remained consistent in the time periods of 2050 and 2070 under RCP4.5 while fluctuating in 2030, 2050 and 2070 under RCP 2.6. However, the suitable habitat under current and future scenarios declined in 2030 under RCPs 4.5 and in 2030, 2050, and 2070 in scenarios RCPs 8.5. Currently suitable HGG habitat was located in an area where the species is known to be locally extinct. Further work is necessary to determine the key drivers of local extinction events in an effort to mitigate population crashes. Our work will assist in formulating conservation actions for the HGG in the context of climate change, and provide a platform for continued monitoring efforts of the species.