Evaluation of the Ability of CMIP6 Global Climate Models to Simulate Precipitation in the Yellow River Basin, China

Choosing an appropriate global climate model (GCM) is of great significance for the simulation of the hydrological cycle over a basin under the future climate scenarios. In this study, the rank score method (RS) with eight indicators was applied to comprehensively evaluate the ability of 19 GCMs in the Sixth Global Atmosphere and Coupled Model Intercomparison Project (CMIP6) to simulate the properties of precipitation during 1961-2014 over the Yellow River Basin. The results indicated that: (1) The GCMs in the CMIP6 differed greatly in their ability to simulate precipitation in the basin, with the six higher-scoring GCMs as follows: MRI-ESM2-0, ACCESS-CM2, CNRM-CM6-1, CNRM-ESM2-1, FGOALS-f3-L, MPI-ESM1-2-HR. (2) Most GCMs overestimated the precipitation in the basin, and their ability to simulate the phase distribution of monthly extreme precipitation was poor. Although most GCMs in the CMIP6 could simulate the variation of annual precipitation in the basin, the simulated wet season was too long and the precipitation during the period was too high. (3) Most GCMs could simulate the variation of precipitation in summer and winter, but they underestimated the precipitation of summer and overestimated that of spring in the basin. (4) The GCMs in the CMIP6 well simulated the spatial distribution of the average annual precipitation in the basin, but there was an obvious overestimation in the source area of the Yellow River, and an underestimation in the northern part of the middle reaches of the basin. (5) Though the GCMs performed well to simulate the spatial characteristics of modal on the average annual precipitation, most GCMs overestimated the variability of the first mode in the northern part and the second mode in the source area of the basin.