Regional climate model intercomparison over the Tibetan Plateau in the GEWEX/LS4P Phase I
Jianping Tang,Yongkang Xue,Mengyuan Long,Mengnan Ma,Xin-Zhong Liang,Shiori Sugimoto,Kun Yang,Zhenming Ji,Jinkyu Hong,Jeongwon Kim,Haoran Xu,Xu Zhou,Tomonori Sato,Hiroshi G. Takahashi,Shuyu Wang,Guiling Wang,Sin Chan Chou,Weidong Guo,Miao Yu,Xiaoduo Pan
DOI: https://doi.org/10.1007/s00382-023-06992-4
IF: 4.901
2023-11-05
Climate Dynamics
Abstract:Results from eight regional climate models (RCMs) participating in the Impact of Initialized Land Temperature and Snowpack on Sub-seasonal to Seasonal Prediction (LS4P) initiative of the Global Energy and Water Exchanges (GEWEX) are examined and compared with observations over the Tibetan Plateau (TP). The RCM common domain covers most areas of East Asia with a horizontal resolution of 20–30 km. The model simulation covers a period from April to September in each year between 1991 and 2015. This study explores the RCMs' ability for seasonal climate simulation over the TP, focusing on the summer monsoon climate as part of the LS4P initiative. An intercomparison is made among eight RCMs for precipitation, surface air temperature, mid-troposphere atmospheric circulation, moisture conditions, and surface energy fluxes. It shows that the downscaling characteristics differ significantly between two major RCM types. The RegCM4 models show positive precipitation biases over the entire TP, especially over the south and southeast TP, while the WRF models mostly show both positive and negative precipitation biases over the TP with relatively high spatial correlation between simulated and observed precipitation. The multi-model ensemble mean produces overall smaller precipitation biases than most individual RCMs, with the largest biases over the southeastern TP, and smaller surface air temperature biases over most areas of the TP, especially over the central and southwestern TP. Moreover, the ensemble mean can better reproduce the inter-annual variation of precipitation and surface air temperature than most RCMs with proper magnitude. Sensitivity analyses using RegCM4 with different physics parameterizations show that varying land and cumulus schemes may induce large precipitation differences over the TP by affecting moisture and atmospheric circulation conditions in the lower and upper troposphere, respectively. Moreover, turbulent heat and radiation fluxes differences are associated with the temperature differences between different RegCM4 models.
meteorology & atmospheric sciences