Evaluation CMIP5 Models In Order to Simulate Rainfall by using a Combination of Precipitation data Network Aphrodit and Satellite Precipitation Persiann-cdr In Khuzestan Province

One of the most important Limitation General Circulation Models , Large scale are being simulation of climatic variables. So should With Various method are downscaled, The ability to have identified a study area. Choose a suitable GCM model for the study area Very important role In the simulation  parameter (precipitation) is intended for future. In this research of CMIP5 Models Contains BCC-CSM1.1.M , MPI-ESM-MR and MPI-ESM-LR was used In order to evaluate three models in the baseline period And determine the best model In order to fit into the study area .The first the Historical and Observation data are divided in to two Periodes: the first half (1983-1992) the period baseline and the second half(1994-2003) the period Validation. Index verification Mean absolute error and MAE Skill score Calculate between models and observation data in the evaluation period before and after Downscaling seasonal in Software Matlab For all Pixels Khuzestan Province. Results show that Befor applying the change factor method BCC-CSM1.1.M Model Error was Less than the other two models , and The other two were similar in terms of error rate. After applying change factor, BCC-CSM1.1.M Model the improvement was driven self and MPI-ESM-MR and MPI-ESM-LR models were ranked as second and third suitable models. So most trusted after the downscaling change factor method for the future under the scenario RCP4.5 for Khuzestan Province was assigned to BCC-CSM1.1.M model.