Predicting the spatiotemporal characteristics of flash droughts with downscaled CMIP5 models in the Jinghe River basin of China.

Climate warming greatly affects the frequency and intensity of flash droughts, which can cause huge damage to agriculture. It is important to understand the changing rules of future flash droughts and take precautionary measures in advance. Thus, we focused on the flash drought characteristic of the Jinghe River basin using variable infiltration capacity (VIC) model and four-model ensemble in the two representative concentration pathway scenarios. Four-model ensemble mean can well capture hydrological changes in the reference period. The heat wave flash drought (HWFD) and the precipitation deficit flash drought (PDFD) mainly occur in the northern during reference period. The HWFD and PDFD have shown a linear growth trend in the future and both shown higher growth rates in the RCP8.5 scenario. The frequency of occurrence (FOC) increments of flash droughts were relatively high in the southern Jinghe River basin. And the HWFD and the PDFD mainly occurred in May-September. Further results indicate that the contribution of the maximum temperature to HWFD was the biggest (greater than 0.7), followed by evapotranspiration (ET) and soil moisture (SM). The contribution of maximum temperature to PDFD was the biggest (greater than 0.5), followed by precipitation and ET. Global warming in the twenty-first century is likely to lead to intensification of flash droughts. Therefore, measures and suggestions were proposed to effectively respond to flash droughts in our study.