3D-var assimilation of GTS observation with the gravity wave drag scheme improves summer high resolution climate simulation over the Tibetan Plateau

The Tibetan Plateau (TP) is one of the most complicated orographic regions worldwide, and due to the lack of quantitative observations, the different simulation biases are still existing via various climate models over the TP. In this study, a one-summer-month 6-km dynamical downscaling simulation is conducted to evaluate the improvement of the new gravity wave drag (GWD) scheme and cycled assimilation of observations from the Global Telecommunications System (GTS) by the three-dimensional variational data assimilation (3D-Var) method. The updated GWD scheme provides better simulation results for surface and vertical winds, temperature, and humidity. 3D-Var cycled assimilation of GTS observations with the GWD scheme further improves the wind forecasting at the upper atmosphere levels but enhances temperature cold bias over the TP, and the latter may be partly related to less surface sensible and latent heat flux after assimilation. Notably, it obviously performs better at the spatial distribution and temporal variation of daily precipitation, thus effectively reduces the precipitation wet bias, especially for the eastern TP. This benefits from the more accurate simulation of different precipitation categories by data assimilation, especially for the light rain (1–10 mm/day). The mechanism of less precipitation wet bias is that assimilation of GTS observation results in weaker monsoon flow water vapor transport from low-latitude oceans, weaker Tibetan High with more accurate circulation fields and less upward vertical velocity. This research may provide guidance for establishing a downscaling dataset of higher spatiotemporal resolution for the TP.