On the improvement of precipitation forecast skill from physical initialization

Abstract

This study explores the impact of physical initialization on the numerical weather prediction of tropical rainfall. The goal is to improve the definition of the initial state by assimilation of proposed or currently available surface and satellite-based observations during a pre-integration phase using a global spectral model. Physical initialization refers to the use of reverse algorithms consistent with the physics of the numerical model which can provide a modification of the initial state via incorporation of an analysis of tropical rainrates. This modification of the initial state variables is accomplished in an assimilation phase of the model forecast. The physical initialization process produces, in a diagnostic sense, a thermodynamic consistency between the humidity variable, the surface fluxes, rainfall distributions, diabatic heating and the clouds. A diabatic initialization is achieved by a Newtonian relaxation of the above diagnosed humidity variable where the divergent wind is permitted to evolve in response to the imposed surface fluxes and the condensation heating in a consistent manner. An important finding of this study is related to the absolute correlation of the observation only based rainfall and the model-based rainfall at the initial time and at the end of a one day forecast which are significantly improved with the use of physical initialization. The “observed” rainrates are obtained from algorithms that translate satellite-based measurements of outgoing longwave radiation and radiances for an array of microwave frequencies. In addition, the available raingauge records over the land area are incorporated to define the “observed” rainfall over the gaussian transform grid squares of a global spectral model at a high resolution (T106). Thus the rainfall measures are averages over roughly 100 × 100 km² and 7.5 min which is the time step of the spectral model. It is for this averaged representation that we are able to demonstrate a very marked improvement in nowcasting and one day forecasts of tropical rainfall. The monthly mean rainfall climatology, thus obtained, nearly replicates the rainfall analyses provided to the physical initialization.