Research On The Impact Of PBL Parameterizations To Simulation Of The Heavy Rainfall

The mechanism of heavy rainfall evolution is a critical problem for meteorologists. One of the most significant problems is the vertical transfer of vapor, momentum and heat in the planetary boundary layer (PBL). Thus, the importance should be attached in the PBL parameterization when the numerical weather prediction model is applied to simulate the process of heavy rainfall.In this study, four PBL parameterizations named YSU, MYJ, ACM2 and MRF in WRF are applied to simulate two cases in May and July 2013 respectively in the Yangtze River. Additionally, one case without PBL parameterization (NOPBL) is simulated as the reference. Through comparison with the observation data, the simulation capabilities of the five different kinds of experimental design were evaluated. Their simulated precipitation area, total rainfall amount and basic meteorological elements were analyzed. The results show that largest rainfall simulation bias happens in the experiments without planetary boundary layer scheme, while experiments with asymmetric convective version 2 scheme and medium-range forecast scheme produce better rainfall simulation. Especially, experiment with Mellor-Yamada-Janjic scheme yields best simulation results for small and large rainfall, while when coupled with Yonsei University scheme, the model gives highest threat score (TS) for simulation of rainfall with various intensities.Last but not least, the effects of land surface models on vertical velocity, divergence, divergence of water flux,θse field are studied. Two different LSMs named NOAH and SLAB coupled with MYJ, which has the best performance, are compared. The results showed that the configuration of MYJ PBL scheme with NOAH LSM scheme is more applicable in 721-Beijing rainstorm weather.