Impact of Boundary Layer Physics on Tropical Cyclone Simulations in the Bay of Bengal Using the WRF Model

Abstract

In this work, the sensitivity of tropical cyclone (TC) simulations over the Bay of Bengal to planetary boundary layer (PBL) physics in the WRF model is investigated. Numerical simulations are performed with WRF-ARW model using NCEP GFS data for five very severe cyclonic storms (Vardha, Hudhud, Phailin, Lehar and Thane). Five conceptually different PBL schemes (YSU, MYJ, QNSE, MYNN and BouLac) are evaluated. Results of 25 sensitivity experiments showed that PBL physics mainly affected the intensity while producing small variations in track prediction. The QNSE, followed by MYJ and BouLac, produced highly intensified storms, and MYNN produced weakly intensified storms. The YSU scheme showed better comparisons with IMD best track estimates. From the analysis of five cyclones, it is found that the YSU produced minimum errors for central pressure (−5.4, −0.8, −2.6, −5.25 hPa), maximum wind (19, 7.6, −0.96, −0.77 m/s) and track (66, 146, 182, 217 km) at 24-, 48-, 72- and 96-h forecast intervals. Analysis of various thermodynamical and dynamical parameters clearly showed that the PBL physics impacts the predictions by variation of (1) surface energy fluxes, (2) convergence, (3) inflow/outflow, (4) tangential winds, (5) vertical motion and (6) strength of the warm core and associated storm structure. A detailed analysis conducted in the case of Hudhud indicated that the PBL schemes influenced the intensity predictions through a WISHE type of feedback by the variation of convergence, radial inflow, vertical motion, and surface fluxes. While the YSU and MYNN schemes produced moderate values of radial inflow, the QNSE, MYJ and BouLac schemes produced stronger inflow. The stronger inflow, spin-up and stronger wind-induced transport of energy fluxes in the QNSE, MYJ and BouLac schemes led to a stronger convection and a higher intensification of TCs in these simulations.