Predicting the potential geographic distribution of Camellia sinensis var. shan under multiple climate change scenarios in Van Chan District Vietnam

The study focuses on the potential geographic distribution of a valuable green tea plant for sustainable agriculture development in the mountainous northwestern region of Vietnam, called Camellia sinensis var. shan (Css). It also discusses how the spatial simulation of species distribution through environmental variables and Maximum Entropy (MaxEnt) model at a local scale. In the context of climate change, its potential distribution was modeled under current and future scenarios of 04 representative concentration pathways (RCP 2.6, RCP 4.5, RCP 6.0, and RCP 8.5) for 2050. From the field survey, 70 sample points were used for the procedure, including 75% and 25% of total for training and validation, respectively. Understanding the demand of downscaled CMIP5 bioclimatic data to higher resolution, the prediction-mapping results of Css meet the accuracy requirement of Area Under Curve, which ranges from 0.9016 +/- 0.0147. In which, the mean diurnal range (Bio 2) was contributed significantly (85.5%) to prediction results. The highest increase of Css was observed from prediction map, which has included 24.87% for RCP 2.6 (at high suitable degree), 2.11% for RCP 4.5 (at medium suitable degree), 1.85% (at very less suitable degree) and more than 3.19 times (at very high suitable degree) for RCP 6.0, and 32.03% for RCP 8.5 (at less suitable degree). Prediction results in 2050 revealed that the slight gain at high and very high suitable degree couldn't compensate for the "less suitable" expansion of Css in north-western Vietnam. This research could provide useful information about potential distribution assessment under multiple climate change scenarios for local development strategies as well as conservation management.