Impact of data assimilation on a calibrated WRF model for the prediction of tropical cyclones over the Bay of Bengal

The main objective of the present study is to examine the impact of three-dimensional variational data assimilation utilizing the multivariate background error covariance (BEC) estimates, in combination with the model calibration, for the simulations of seven tropical cyclones over the Bay of Bengal region. The study indicates that the utilization of multivariate BEC in assimilation influences the model forecasts in terms of wind speed at 10 m height, precipitation, cyclone tracks and cyclone intensity. The assimilation experiments conducted with a previously calibrated model combined with the control variable option 6 (cv6) of BEC have reduced the overall ischolar_main mean square error (RMSE) of 10 m wind speed by 17.02%, precipitation by 11.14%, cyclone track by 41.93% and the intensity by 25.5% when compared to the default model simulations without assimilation. The best experimental set-up is then used for the operational forecast of a recent cyclone Gulab. The results show an RMSE reduction of 18.61% in the cyclone track and 28.99% in intensity forecasts. These results also confirm that the utilization of cv6 BEC in the assimilation of conventional and radiance observations on a calibrated model improves the forecast of tropical cyclones over the Bay of Bengal region.