Evaluation of Medium-Term Forecast of TIGGE Numerical Weather Prediction Models in Karun Basin

The flood has become a serious issue Over the past decades, due to urban development and climate change and therefore, international interest in flood forecast has increased subsequently. On the other hand, due to the fact that the main cause of the flood is precipitation, its precise prediction in hydrological applications is important. This study evaluates the predicted rainfall data of the global numerical models of the TIGGE database and their improvement using bias correction in the Karun Basin. Meteorological assessments were carried out in point and regional scale and the estimation of displacement error, volume and spatial pattern of precipitation of the models were performed using the CRA Object Oriented Method. At the next step, the predicted precipitation was improved using quantile mapping method. In assessing the initial data of the TIGGE database of global models, the ECMWF numerical model has dominated in the point and region evaluation with respect to the indices compared to other models, and the CMC model has a poor performance compared to other models. The evaluation of biased correction data by quantile mapping method also indicates an improvement in evaluation of the indices compared to the results before applying this method. Furthermore, in estimating the displacement error, the volume and spatial pattern of the global numerical models, the ECMWF, NCEP, and CMC models have shown to perform better than the UK model, respectively