Sensitivity analysis of streamflow parameters with SWAT calibrated by NCEP CFSR and future runoff assessment with developed Monte Carlo model

Abstract

The present study analyzed sensitivity of flow parameters using SWAT and effect of climate change on surface water resources, considering uncertainty with Monte Carlo. For this purpose, output of nine-model related to fifth-climate change-report during baseline period 1971–2000 was weighted. Using Monte Carlo, 100 samples of probabilistic distribution of basin temperature and rainfall were generated. LARS-WG model was used for downscaling, then temperature and precipitation were calculated under RCP2.6 and RCP8.5 for future periods 2040–2069 and 2070–2099. SWAT was calibrated using observational-data and the National Centers for Environmental Prediction (NCEP) Climate Forecast System Reanalysis (CFSR) (NCEP CFSR) global climate datasets and sensitivity of parameters affecting flow was analyzed. Results showed that observed-data had better performance than NCEP CFSR. Finally, future runoff was calculated under RCP2.6 and RCP8.5 for 2040–2069 and 2070–2099. Results showed that average annual runoff decreased by 84, 80, 82 and 80%, for 2040–2069 (RCP2.6 and RCP8.5) and for 2070–2099 (RCP2.6 and RCP8.5) relative to baseline, respectively.