Predicting the effects of climate change on prospective Banj oak (Quercus leucotrichophora) dispersal in Kumaun region of Uttarakhand using machine learning algorithms

All sort of vegetation is highly responsive to climatic factors and therefore distribution and redistribution of vegetation is bound to be affected by the change in the climatic conditions. The present episode of climate change is rapid in nature with it fastest temperature rise in Himalayas after poles on the earth, rendering vegetation of this region vulnerable to redistribution in space and time. Therefore, accurate modeling of the potential distribution of plants native to the Himalayan area is essential. Machine learning has improved the accuracy of species distribution models to a greater extent. The effects of climate change on the spread of Banj oak, a prominent tree species of the mid-Himalayas in Uttarakhand's Kumaun area, were simulated in this study. The generalized linear model (GLM), boosted regression tree (BRT), and maximum entropy (MaxEnt) were used to achieve this. The models' accuracies were calculated and compared. The accuracy was determined using the area under the curve (AUC) and receiver operating characteristics (ROC) curves. The MaxEnt model outperformed the rest two models and therefore it was utilized for modeling and prediction of potential distribution of Banj oak for the present and future. The results with higher accuracy (i.e., AUC > 0.95) model suggested that the areal expansion of potential distribution of Banj oak is going to crunch down by more than 1000 sq. km. as compared to today by the year of 2070, highlighting the gravity of climate change. This areal reduction of broadleaf tree is limited in the lower latitude. Higher altitudes were predicted to enjoy expansion of the aforesaid species. This study is a stand-alone contribution to the species distribution modeling of Quercus leucotrichophora in the mid-elevations of the Central Himalayas in India.