1. Introduction

Climate change is one of the most significant environmental challenges of the 21st century, driven by increased greenhouse gas emissions. It has a widespread impact on ecosystems, agriculture, and water resources. Global warming is expected to pose significant challenges in the future. Global Climate Models (GCMs) are essential tools for simulating and assessing the effects of climate change and analyzing various atmospheric, oceanic, and land systems.

GCMs simulate the global climate response to greenhouse gas concentrations and are widely used in climate studies. Owing to structural differences and varying initial conditions, these models produce different results, even under the same emission scenarios. Therefore, evaluating and refining their output is crucial for regional studies. The CMIP6 models demonstrated improved accuracy in simulating the daily minimum and maximum temperature parameters.

According to IPCC reports, the global temperature is projected to increase by 2.1 °C to 5.3°C under intermediate scenarios and 3.3 °C to 7.5°C under pessimistic scenarios. Climate change has intensified extreme events, such as heatwaves and meteorological droughts, causing severe environmental and societal impacts. These changes affect water resources and agricultural production by increasing evaporation and transpiration, reducing soil moisture, and increasing water demand. Agriculture is highly dependent on climate conditions, and shifts in climate patterns can significantly reduce crop yields and overall productivity. Ultimately, climate change is a serious threat with major implications for natural ecosystems and agriculture. Agrometeorological phenomena, which directly influence crop production and yield, have become increasingly critical, necessitating special attention to climate and agricultural planning.

The highland provinces of the central Zagros region, including Kurdistan, Hamedan, Lorestan, Kermanshah, and Ilam, cover the study area. Based on the studies conducted, the study provinces, especially Lorestan, Hamedan, and Kurdistan, are considered to be the main areas for walnut cultivation.

2. Methodology

The Central Zagros highlands, including Kurdistan, Hamedan, Lorestan, Kermanshah, and Ilam, form the study area. Among them, Lorestan, Hamedan, and Kurdistan are the key walnut-producing regions. The area has an elevation range from 28 to 4,049 m. To assess climate change impacts on temperatures, observational data, baseline climate model outputs, and future projections were analyzed. Observational data included daily minimum and maximum temperatures, while CMIP models provided historical climate data (1985-2014) for evaluation. Future temperature projections (2021-2040, 2041-2060) were derived from CMIP models using SSP2-4.5 (moderate development) and SSP5-8.5 (high fossil fuel use) scenarios, with data sourced from ESGF.

3. Results and Discussion

In this study, 15 models from CMIP6 were selected based on their resolution, availability of required data, and climate scenarios, and their performance in simulating minimum and maximum temperatures during the baseline period (1985-2014) was evaluated. The validation results using the RMSE index showed that the minimum temperature ranged between 3.51 and 8.25, while the maximum temperature ranged between 3.25 and 12.20. The correlation coefficient (CC) for both the minimum and maximum temperatures varied between 0.97 and 1. The ACCESS-CM2 and MIROC6 models had the highest and lowest accuracy for minimum temperature, respectively, while MPI-ESM1-2-HR and MIROC6 showed the same accuracy for maximum temperature. Based on entropy calculations, the NMBD index for the minimum temperature and the NRMSE index for the maximum temperature were identified as the most suitable criteria for selecting the best model. Ultimately, ACCESS-CM2, EC-Earth3, GISS-E2-2-G, FGOALS-g3, and MRI-ESM2-0 were chosen as the best models for minimum temperature, whereas NorESM2, MPI-ESM1-2-HR, MPI-ESM1-2-LR, INM-CM5-0, and INM-CM4-8 were selected for maximum temperature.

After integrating the selected models, future changes in minimum and maximum temperatures were projected for two future periods (2021-2040 and 2041-2060) under the SSP2-4.5 and SSP5-8.5 scenarios. The results indicated that in the near future (2021-2040), the minimum temperature is expected to increase by 1 to 2°C, while the maximum temperature will rise by 0.8 to 1.1°C. In the mid-future period (2041-2060), the minimum temperature is projected to increase by 2.1 °C to 3.3°C and the maximum temperature by 1.8 °C to 2.1°C. The highest increase in minimum temperature was observed in Saqez, whereas the highest increase in maximum temperature was recorded in Qorveh and Saqez. Spatial analysis revealed that northern regions and higher elevations experienced the most significant temperature changes, whereas southern areas experienced smaller increases. These changes could have important implications for agriculture, plant phenology, and climate-related hazards.

4. Conclusion

This study assessed 15 CMIP6 models and used an entropy-based multi-model ensemble approach to analyze temperature changes at meteorological stations across western Iran for two future periods (2021-2040 and 2041-2060) under the SSP2-4.5 and SSP5-8.5 scenarios. The results showed that the ensemble method provided higher accuracy than the individual models. The minimum temperature is projected to increase by 16.49% in the near future and 29.7% in the distant future, equivalent to 1.4°C and 2.5°C above the observational period, respectively. The highest temperature increase was observed in Saqez, whereas the lowest was observed in Khorramabad. High-altitude and northern regions showed greater sensitivity to temperature changes than low-altitude and warmer areas. These findings align with those of previous studies and can aid in managing temperature-related risks, phenology, and adaptation to climate change.