Document Type : Original Article

Authors

• Negin Nabatghods ¹
• Parisa-Sadat Ashofteh ²

¹ M.Sc. Student, Department of Civil Engineering, University of Qom, Qom, Iran

² Associate Professor, Department of Civil Engineering, University of Qom, Qom, Iran

Abstract

This study examines the performance of 19 climate models from the Sixth Assessment Report (CMIP6) in simulating temperature and precipitation variables in the Qaranqoo watershed over the period 1971–2000. The primary objective is to identify the most accurate model for predicting climate variables in this region. To evaluate model performance, statistical metrics including Nash-Sutcliffe Efficiency (NSE), Kling-Gupta Efficiency (KGE), Root Mean Square Error (RMSE), Normalized Root Mean Square Error (NRMSE), Mean Absolute Error (MAE), Percent Bias (PBIAS), and correlation coefficient (r) are utilized. Results indicate that the models perform significantly better in simulating temperature compared to precipitation. The MPI-ESM1-2-LR and CMCC-ESM2 models were identified as the best models for temperature and precipitation simulation, respectively. Furthermore, sensitivity analysis and weighting based on the Analytical Hierarchy Process (AHP) revealed that the INM-CM4-8 and NORESM2-MM models hold the highest weight for precipitation simulation, while the NESM3 and MPI-ESM1-2-LR models hold the highest weight for temperature simulation. Sensitivity analysis also confirmed the stability of the results under various conditions. The findings of this study can support water planning and natural resource management in the region. Ultimately, the research highlights that using more accurate models can contribute to reducing uncertainty in climate-related decision-making.

Keywords

• CMIP6 Climate Models
• Temperature and Precipitation Simulation
• Climate Change
• Multi-Criteria Decision-Making (AHP)
• Performance Evaluation Metrics

Main Subjects

• Climate change