On the prediction of tropical cyclones over the indian region using a synthetic vortex scheme in a mesoscale model

摘要

The tropical cyclones form over the oceanic regions where conventional meteorological observations are not available. This contributes to a poor initial analysis of the cyclonic vortex and hence inadequate forecast. One way of overcoming the above problem is to modify the initial analysis by replacing the weak and ill-defined vortex in the initial analysis with a synthetic vortex having the correct size and intensity at the correct location. In this study we are investigating the effect of inclusion of a synthetic vortex based on Rankine as well as on Holland wind profiles, using NCAR-AFWA bogussing scheme for the prediction of four tropical cyclones, which formed over the Bay of Bengal during November 2002 and 2005, December 2005 and over the Arabian Sea during May 2004, using the MM5 model. Two numerical experiments are designed in this study for each of the above four cyclones. In the first experiment the model is integrated with a synthetic vortex based on Rankine wind profile while in the second experiment we utilize the Holland wind profile. For the November 2002 cyclone, in both the experiments the model is integrated from 10 November 2002 18 UTC to 12 November, 2002 12 UTC with the synthetic vortex inserted at the initial time. The results of the study for the November 2002 cyclone show that the model simulation with the Holland vortex has produced a stronger cyclone in terms of minimum sea-level pressure and maximum wind speed. Also, the results for the November 2002 cyclone with the Holland vortex showed a better longitudinal height section of the horizontal wind speed across the center of the cyclone. The track error of the cyclone for the November 2002 cyclone is less in the model simulation with the Holland vortex at the initial time and at 24 hours of forecast. The results for the November 2002 cyclone with the Rankine vortex showed greater vertical wind speed as compared to the Holland vortex. However, for the November 2002 cyclone there were no significant differences in the spatial distribution of precipitation for both the experiments. In order to provide an adequate number of case studies for a good statistical sample, the present study is extended for three additional cyclones over the Indian region. All four cyclones studied here show that the Holland vortex has produced a stronger cyclone in terms of the minimum sea-level pressure and maximum wind speed. The Holland vortex showed a better vertical structure of wind speed in the longitudinal height section at 24 hours of forecast for the November 2005 cyclone while the structure was better for the Rankine vortex for the remaining two cyclones. There were no significant differences in the spatial distribution of precipitation for the two experiments corresponding to all four cyclones. Some statistical results pertaining to all four cyclones are provided such as the average track error as well as the average difference between the observed and the model minimum sea-level pressure and the maximum wind speed. The statistical results corresponding to the average of all the four cyclones are at only a slight variance with the results corresponding to the November 2002 cyclone.