Regional climate dynamical downscaling over the Tibetan Plateau—From quarter-degree to kilometer-scale.

The Tibetan Plateau (TP) possesses the largest cryosphere in the world outside of the Arctic and Antarctic, and is the source of nine major rivers in Asia. The surface environment of the TP has undergone significant changes against the background of global warming. It is projected that the continuation of climate change in the future will result in most of the glaciers and frozen soil disappearing by the end of this century, and freshwater resources will be greatly reduced, on which 22% of the world's population depends. These environmental changes are of great concern to global society given the influences of the TP on the climate at the global scale. However, great uncertainties exist in global climate simulations over the TP, which affects our ability to properly understand the associated water security crisis. Based on atmospheric dynamics and physical processes, dynamical downscaling can characterize surface conditions more accurately than global simulations, and better simulate and predict regional or local weather and climate situations. With advances in supercomputing, the grid spacing of dynamical downscaling simulations has been continuously increasing, marching the technique into the kilometer-scale era. In this paper, the origin and development of dynamical downscaling in the TP region from the quarter-degree to kilometer scale is firstly introduced, including an assessment of the advantages and disadvantages of dynamical downscaling at the kilometer scale over the TP. Then, the main land surface factors affecting the performance of dynamical downscaling over the TP are described, as well as a brief introduction to a land surface model with specific plateau characteristics. Specifically, it has emerged that perfecting the land surface model and improving the performance of land-atmosphere interaction are the most effective ways to advance the performance of dynamic downscaling in this region. Finally, the challenges and some recommended future research directions are discussed and proposed. [ABSTRACT FROM AUTHOR]