Initialization of Tropical Cyclone Structure for Operational Application

Abstract

Listen

Translate article

Abstract : The long-term goal of this project is to improve the prediction of tropical cyclone (TC) genesis, structure, and intensity changes through improved representation of 3-dimensional TC structure in operational weather forecast models. The accurate prediction of TC genesis, structure, and intensity changes is critical to Navy missions and civilian activities in coastal areas. Significant gains have been made in TC track prediction over the past decades. However, genesis and intensity forecasts have shown very little progress during the same time period. A primary contributing factor to our lack of skill in predicting TC genesis and intensity is an inadequate representation of initial axisymmetric and asymmetric TC dynamic and thermodynamic structures. TC initialization is a universal problem in nearly all current operational weather forecast systems, as indicated in a recent HFIP workshop. By conducting both idealized Observation System Simulation Experiments (OSSE) and real-case TC forecast experiments, we intend to tackle the weaknesses in current operational weather forecast models. We propose a combined TC dynamic initialization-3DVar data assimilation approach. The key component of this new scheme is a TC dynamic initialization (TCDI) package. The TCDI package includes three steps. The first step is the decomposition of the TC vortex from its environmental field. The second step is a dynamic initialization with the full nonlinear dynamics and physics, which forces the first guess field toward the observed central minimum pressure, given realistic asymmetric heating profiles. The third step is to add the newly generated TC field into the environmental field and use it in the subsequent 3DVar assimilation. After the TCDI is completed, the 3Dvar assimilation follows. A preliminary OSSE study was conducted and the results show that this new approach greatly improves both TC symmetric and asymmetric dynamic and thermodynamic patterns.