Simulation of severe land-falling Bay of Bengal Cyclones during 1995–1999 using Mesoscale model MM5

Abstract

In this study, the nonhydrostatic version of Pennsylvania State University (PSU)/National Center for Atmospheric Research (NCAR) mesoscale model MM5 is used to simulate the severe land-falling Bay of Bengal cyclones. The cyclonic storms associated with maximum sustained wind of 48 knots or more are considered severe cyclones. The main objective of the study is to improve model initialization and evaluate the model performance towards prediction of intensity, track, and landfall of these storms. The model configuration used in the present study is primarily based on sensitivity studies conducted earlier by the authors. The vortex specification in the large scale global analysis is rectified using synthetic data in preparation of high resolution reanalysis. The vortex initialization in the model is done through 12 hours nudging to the prepared high-resolution reanalysis. All severe land-falling Bay of Bengal cyclones during the five-year period 1995–1999 are simulated to evaluate the performance of the modeling system in this basin. The storms are simulated at least up to their landfall. The model simulated mean sea level pressure, horizontal wind, and rainfall are compared with observations/best-fit estimation. The simulated tracks of the storms are compared with the best-fit tracks. The average errors in track forecast in the present study are compared with the average error in representing the tracks in the National Center for Environmental Prediction (NCEP) reanalysis, errors in similar other predictions, operational prediction, and forecast difficulty level (FDL) in North Indian Basin. It indicates that the tracks of these storms are relatively better simulated in the present study with the errors less than the FDL in the North Indian Basin and present operational track forecast errors in this basin. Most importantly, the landfall points of these storms are well simulated by the model though the time of landfall delays from actual landfall time as reported by India Meteorology Department (IMD) at an average by five hours. The intensity of these storms is also simulated reasonably well by the model though the sharp deepening of some explosively deepening storms is not well captured.