Impact of Cumulus Convective and Plantery Boundery Layer Parameterization Schemes on The Simulation Of Tropical Cyclone Using Weather Research And Forcasting (WRF) Model

Abstract

The convection and planetary boundary layer (PBL) processes play significant role in the genesis and intensification of tropical cyclones (TCs). Three convective and two PBL parameterization schemes have been incorporated in the present study for numerical weather prediction models. A systematic inter comparison of performance of parameterization schemes is essential to select a model for Bangladesh and its surroundings. In this context, a comprehensive sensitivity analysis on physical parameterization schemes of Weather Research and Forecasting (WRF) model have been carried out for the prediction of track and intensity of three the tropical cyclones: TC Nargis (26 April- 02 May, 2008), TC Thane (25-3 1 December, 2011) and Mahasen (10-16 May, 2013). For this purpose, making six possible combinations of physical parameterization schemes of two PBL Schemes (Yonsei University Scheme and Mellor Yamada Jainjic Scheme) and three convection schemes (Kain-Fritch, Betts-Miller, and Grell- Devenyi) of WRF-ARW model are employed to obtain the best combination for the prediction of TCs over Bay of Bengal. In this regard, at first it is considered TC Nargis as the case- 1 (26 April- 02 May, 2008) and the WRF model is run using two types of Planetary Boundary Layer (PBL) YSU & MYJ respectively along with three Cumulus Parameterization (CP) KF, BM & GD schemes respectively and six independent run is made. The model domain is consisting of single nested with 9 km horizontal resolution with 28 vertical sigma levels. It is repeated the procedure for two other cases TC Thane (25-3 1 December, 2011) and TC Mahasen (10 – 16 May, 2013). The simulated track and intensity (pressure & wind) are compared with the IMD observed data and select the best combination of the parameterization schemes. The tracks are also compared with those provided by the operational centers like NCEP and IMD. It is found that the combination of YSU PBL scheme with KF convection scheme (YKF) provides a better prediction of intensity, track, and rainfall consistently. The average Root Mean Square Error (RMSE) of intensity in terms SLP is 0.782243 hPa for all TCs, for 10-m wind is 2.2, 2.02, and 3.79 ms-1 for Nargis, Thane and Mahasen respectively and mean track errors are 33.44 km for Mahasen, 21.16 km for Thane and 55.26 km for Nargis respectively. The errors are found to be least in all aspects with YKF combination. Finally, the pressure, wind (vector and scalar horizontal wind, radial, tangential and vertical wind), vorticity, temperature anomaly, water vapor mixing ratio, rainfall and track have been studied to understand the cyclonic structure and dynamics.