Evaluation of the WRF model’s performance at gray-zone resolution in simulating climate over the Tibetan Plateau

The Tibetan Plateau (TP), also known as the “Water Tower of Asia”, profoundly impacts regional and global climates. Existing climate models exhibit substantial biases over the area, primarily due to low spatial resolution, deficient driving data, and inadequate model domains. Leveraging meteorological station data, the CN05.1 gridded meteorological dataset, and various reanalysis datasets, we comprehensively evaluate the performance of the Weather Research and Forecasting model at gray-zone resolution (9 km) (hereafter WRF9km) in simulating TP air temperature and precipitation during 1980—2019, and identify bias causes. WRF9km effectively captures the spatial patterns of observed air temperature, although it exhibits a cold bias that can mainly be explained by overestimated surface albedo (accounting for 64%), along with underestimated downward radiation and ground heat fluxes. WRF9km also simulates observed spatial precipitation patterns well (seasonal correlations > 0.5, significant at the 95% level); however, its precipitation biases exhibit pronounced spatial heterogeneity, with overestimation along the slope regions of the southern and eastern TP and underestimation over the interior TP, particularly the western TP, relative to CN05.1. These biases primarily arise from inadequate characterization of wind-field dynamics and moisture transport within the model framework. Meanwhile, the WRF9km effectively captures annual precipitation variation with minor deviations, although it does not fully reproduce the temporal trends in precipitation over the TP. Overall, compared to the driving ERA5 data, WRF9km yields only marginal improvement in mitigating the cold bias but substantially reduces the regional mean precipitation wet bias by 79%, particularly over the southern and eastern slopes. This evaluation provides critical insights to advance dynamic downscaling studies in complex terrain, highlighting the need for enhanced surface albedo parameterizations and improved quality of input driving data.