Long Lead Rainfall Prediction Using Statistical Downscaling and Artificial Neural Network Modeling

Long lead rainfall prediction is important in the management and operation of water resources and many models have been developed for this purpose. Each of the developed models has its special strengths and weaknesses that must be considered in real time applications. In this paper, field and General Circulation Models (GCM) data are used with the Statistical Downscaling Model (SDSM) and the Artificial Neural Network (ANN) model for long lead rainfall prediction. These models have been used for the prediction of rainfall for 5 months (from December to April) in a study area in the south eastern part of Iran. The SDSM model considers climate change scenarios using the selected climate parameters in rainfall prediction, but the ANN models are driven by observed data and do not consider physical relations between variables. The results show that SDSM outperforms the ANN model.