Impact of assimilation of conventional and satellite meteorological observations on the numerical simulation of a Bay of Bengal Tropical Cyclone of November 2008 near Tamilnadu using WRF model

Abstract

The objective of this study is to examine the impact of assimilation of conventional and satellite data on the prediction of a severe cyclonic storm that formed in the Bay of Bengal during November 2008 with the four-dimensional data assimilation (FDDA) technique. The Weather Research and Forecasting (WRF ARW) model was used to study the structure, evolution, and intensification of the storm. Five sets of numerical simulations were performed using the WRF. In the first one, called Control run, the National Centers for Environmental Prediction (NCEP) Final Analysis (FNL) was used for the initial and boundary conditions. In the remaining experiments available observations were used to obtain an improved analysis and FDDA grid nudging was performed for a pre-forecast period of 24 h. The second simulation (FDDAALL) was performed with all the data of the Quick Scatterometer (QSCAT), Special Sensor Microwave Imager (SSM/I) winds, conventional surface, and upper air meteorological observations. QSCAT wind alone was used in the third simulation (FDDAQSCAT), the SSM/I wind alone in the fourth (FDDASSMI) and the conventional observations alone in the fifth (FDDAAWS). The FDDAALL with assimilation of all observations, produced sea level pressure pattern closely agreeing with the analysis. Examination of various parameters indicated that the Control run over predicted the intensity of the storm with large error in its track and landfall position. The assimilation experiment with QSCAT winds performed marginally better than the one with SSM/I winds due to better representation of surface wind vectors. The FDDAALL outperformed all the simulations for the intensity, movement, and rainfall associated with the storm. Results suggest that the combination of land-based surface, upper air observations along with satellite winds for assimilation produced better prediction than the assimilation with individual data sets.