Performance evaluation of PBL and cumulus parameterization schemes of WRF ARW model in simulating severe thunderstorm events over Gadanki MST radar facility — Case study

Abstract

In the present study, an attempt has been made to simulate three severe thunderstorm events that occurred over Gadanki (13.5° N, 79.2° E) region of the Mesosphere–Stratosphere–Troposphere (MST) Radar facility using Weather Research Forecasting (WRF ARW version 3.2) model. We examined the performance of five planetary boundary layer (PBL) parameterization schemes namely, the Yonsei University (YSU), Mellor–Yamada–Janjic (MYJ), Mellor–Yamada Nakanishi and Niino Level 2.5 PBL (MYNN2), and Medium-Range Forecast (MRF) and Asymmetric Convective Model version 2 (ACM2) and three cumulus parameterization schemes Kain–Fritisch (KF), Betts–Miller–Janjic (BMJ) and Grell–Devenyi ensemble scheme (GD) in simulating boundary layer parameters, thermodynamic structure and vertical velocity profiles on the days of the thunderstorm events. Triple nested domain having the inner-most domain of 3 km grid resolution over the study area is considered. The model simulated parameters are validated with the available in situ meteorological observations obtained from micro-meteorological tower, radiosonde, MST radar wind profiler and observed rainfall along with the surface fluxes at Gadanki. After validating the model simulations with the available PBL observations and the statistical assessment reveal that the MYJ scheme could be able to capture the characteristic variations of surface meteorological variables such as air temperature, relative humidity, wind component, vertical profiles of wind, relative humidity and equivalent potential temperature and surface layer fluxes during the study period. Cores of strong convective updrafts with a time lag and lead of one and half hour are better represented by the model with MYJ scheme with GD as seen in the vertical velocity profiles obtained from MST radar observations. The present study advocates that the MYJ–GD combination is suitable for the simulation of thunderstorm events over the study region.