Simulation study of the effects of model uncertainty in variational assimilation of radar data on rainfall forecasting

Recent developments in data assimilation techniques make use of cloud model s initialized with radar data an attractive alternative for real-time quant itative precipitation forecasting (QPF). Before such approaches are used op erationally. there are a number of aspects that need to be addressed to und erstand better the benefits and drawbacks of the practical applications of cloud models in QPF. One aspect is the effect of various sources of uncerta inty on the forecasting performance. Data assimilation formulations based o n variational techniques allow accounting for uncertainty in observations b ut have no efficient mechanism of accounting for uncertainty in the model o n which they are based. To investigate the issue. a simulation-based Monte Carlo methodology, suitable for the analysis of complex nonlinear models, i s used. A one-dimensional stochastic-dynamic cloud model, derived by consid ering stochastic terms in a physically based cloud model is used to simulat e rainfall and radar reflectivity data. A deterministic version of the mode l is then initialized by a variational assimilation technique and used for forecasting. The differences between the forecasts and actual realizations of the stochastic cloud model are statistically analyzed to assess the effe ct of cloud model uncertainty on forecasting. In this paper this methodolog y is also used to study additional effects of other types of uncertainty, s uch as those in radar observations and in the description of rain drop size distribution, for a more complete understanding of the impact of uncertain ties on rainfall forecasting. Based on the scenarios investigated in the pa per, conclusions and recommendations concerning the use of complex cloud mo dels in real-world applications are made. (C) 2000 Elsevier Science B.V. Al l rights reserved.