Rainfall forecasting using variational assimilation of radar data in numerical cloud models

In this paper, a variational data assimilation procedure for initialization of a cloud model using radar reflectivity and radial velocity observations and its impact on short term rainfall forecasting are investigated in a si mulation framework. The procedure is based on the formulation of an objecti ve function, which consists of a linear combination of the squared differen ces between model forecasts and actual observations. The model's initializa tion requires the gradient-based minimization of the objective function. Th ree sources of errors in assimilation and forecasting are considered. These include observational errors, mathematically ill-posed structure (the obse rvations are incomplete and do not uniquely determine the cloud model's ini tial conditions) and optimization related issues (the observations are comp lete, but the optimization procedure involved in the initialization fails t o find the best solution). It is found that the reflectivity observation er rors have significant impact on forecasts and that the mathematical structu re and optimization caused errors are related. Conclusions and recommendati ons for future work are formulated. (C) 2000 Elsevier Science Ltd. All righ ts reserved.