Abstract
In recent decades, severe and long duration drought events have serious impacts on northern China. Consequently, this study is motivated by investigating the projection and uncertainties of drought change over northern China based on 20 CMIP5 multi‐model ensemble (AMME) and best multi‐model ensemble (BMME) under middle and high emission scenarios. BMME selected in term of the spatial patterns and interannual variability of aridity index (AI) against the observations shows better performance on the simulation of dryland especially for hyper‐arid areas in Northwest China (NWC) and sub‐humid regions in North China (NC). In the middle of 21st century, the projections from two model ensembles illustrate the aggravation of drought in loess plateau and west of Xinjiang but reduction of acridity over NWC. Northeast China (NEC) becomes more arid under RCP8.5 that is sensitive with the high emission. The uncertainties increase between two model ensembles at the end of 21st century when the opposite trend can be found over NC under RCP4.5 and NEC under RCP8.5, respectively. The increasing rate of precipitation and PET is the key factor for AI change. The larger contribution of precipitation than PET results in the humid tendency over NWC. Moreover, the equivalent contributions of precipitation and PET induce the increased uncertainties in AI change over NC and NEC. Similarly, the temporal evaluation of AI indicates decreasing trend of drought over NWC but the large uncertainties can be found after middle of this century over NEC and in the early of the century over NC.
Funding Information
  • National Natural Science Foundation of China (41905085, 41530530, 41875168)

This publication has 51 references indexed in Scilit: