Numerical Simulation and Forecasting Techniques for Tropical Cyclones in the South China Sea

Abstract

This chapter introduces a numerical simulation and forecasting technique for tropical cyclones in the South China Sea, and describes a technique to initialize model simulation of typhoons that improves typhoon formation and forecasting in the simulation. It includes three-dimensional variational data assimilation and the curvature-modification linear balance equation. In order to determine the quality of the initial field in numerical models, a relocation method for the initial vortex is adopted as to decrease the error that is caused by bias in a tropical cyclone initial location in tropical cyclone models, and a bogus data assimilation (BDA) method is designed to assimilate typhoon bogus tangent wind into the initial wind field, and to update the typhoon’s vortex structure. Doppler radar data, satellite infrared brightness temperature data, tropical rainfall measuring mission (TRMM) retrieved rainfall rate data, and other satellite observations are integrated into the background field by means of the GRAPES-3D-VAR (Global/regional assimilation and prediction system three-dimensional variational assimilation) system, so that the initial structure of the typhoon is closer to reality. It is shown through statistical analysis that the Tropical and Marine Meteorology (Guangzhou) operational forecasting system possesses effective forecasting skill for tropical cyclones.