Biological and Environmental Research - Earth and Environmental System Sciences
Earth and Environmental System Modeling

Modeling Extreme Precipitation over East China with a Global Variable-Resolution Modeling Framework (MPAS)

Wednesday, December 13, 2017 - 08:00
Add to Calendar

Modeling extreme precipitation requires high-resolution scales. Traditional regional downscaling modeling framework has some issues such as ill-posed boundary conditions, mismatches between the driving global and regional dynamics and physics, and the lack of regional feedback to global scales. The non-hydrostatic Model for Prediction Across Scales (MPAS), a global variable-resolution modeling framework, offers an opportunity to obtain regional features at high-resolution scales using regional mesh refinement without boundary limiting. In this study, the MPAS model is first time applied with the refined meshes over East China at various high-resolutions (16 km and 4 km) to simulate an extreme precipitation event during 26-27 June 2012. The simulations are evaluated with the ground observations from the Chinese Meteorological Administration (CMA) network and the reanalysis data. Sensitivity experiments with different physics and forecast lead time are conducted to understand the uncertainties in simulating spatial and temporal variation of precipitation. The variable-resolution simulations are also compared with the traditional global uniform-resolution simulations at a relatively low scale (~ 30 km) and a relatively high scale (~ 16 km). The analysis shows that the variable-resolution simulation can capture the high-scale feature of precipitation over East China as the uniform-resolution simulation at a relatively high scale. It also indicates that high-resolution significantly improves the capability of simulating extreme precipitation. The MPAS simulations are also compared with the traditional limited-area simulations at similar scales using the Weather Research and Forecasting Model (WRF). The difference between the simulations using these two different modeling framework is also discussed.

Link for More Information: