Performance Assessment of 4D-VAR Microphysics Schemes in Simulating the Track and Intensity of Super Cyclonic Storm “Amphan”

Abstract

The Four-Dimensional Variational (4DVar) data assimilation system of the Advanced Research Weather Research and Forecasting (WRF) model, developed by the international community dedicated to data assimilation research and operations, is customised to simulate the super cyclonic storm "Amphan" formed over the Bay of Bengal during May 16, 2020, to May 21, 2020. Five simulations are conducted using five different microphysics schemes namely, Kessler, Lin et al., WRF Single Moment 3-class (WSM3), WSM5, and WSM6 at a horizontal resolution of 18 km, keeping the Kain–Fritsch cumulus and the Yonsei University planetary boundary-layer scheme fixed. The model simulated features of "Amphan" are compared with observational data from the India Meteorological Department (IMD), the Global Precipitation Measurement mission (GPM), and the 5th generation European Centre for Medium-Range Weather Forecasts (ECMWF) Atmospheric Reanalysis (ERA-5) over the specified region. Among all the schemes, Lin et al. scheme shows track remarkably close to the observed track. Lin et al. (WSM5) scheme shows least along track (AT) error of 7.47 km at 24-h forecast length. Lin et al. shows least AT error of 5.8 km (28.12 km) for 48-h (72-h) forecast length. All schemes except Kessler and WSM3 show the spatial distribution of maximum sustained wind (MSW) surrounding the eye of the cyclone which is similar with ERA5 data. All the schemes underestimate the 10m-MSW during the entire life of the storm. However, the Kessler scheme simulates higher 10m-MSW during 00 UTC 18 May to 12 UTC 19 May in comparison to other schemes and further the simulated MSW matches with IMD observation up to 06 UTC 20 May. The Kessler scheme overestimates the MSLP for the intensity level ESCS-VSCS-SCS-CS valid 09 UTC on 19 May to 00 UTC on 21 May and other schemes underestimate during this period. The analysis carried out with the Method for Object-Based Diagnostic Evaluation tool reveals that the Lin et al. (WSM6) scheme indicates enhanced forecast proficiency for accumulation valid 00 UTC 20 May (21 May) 2020. The analysis of vertically integrated moisture transport (VIMT) and vertically integrated moisture divergence (VIMD) suggests that the greater moisture transport is quite evident in Lin et al. scheme during the SuCS intensity level. Kessler scheme is efficient in simulating warm-rain process and high intensity storm.