Comparative evaluation of different solar radiation models with Angstrom-Prescott model for Hazaribagh, Jharkhand

Abstract

This study evaluates the performance of six solar radiation models for Hazaribagh in Jharkhand by comparing their estimates with those derived from the Angstrom-Prescott (A-P) model, which served as the benchmark reference. The results revealed significant variability in model performance on both a monthly and seasonal basis. The Togrul-Onat and Ertekin-Xaldiz models tended to overestimate solar radiation, particularly during the summer months, while underestimating it in the remaining months. In contrast, the Ogelman model consistently underestimated solar radiation throughout the entire year. The Almorox-Hontoria model showed only minor overestimations in certain months, while the Chen model primarily overestimated during the spring and early summer. On a monthly scale, all selected models showed a positive correlation with the standard Angstrom-Prescott (A-P) model, with R² values ranging from 0.52 to 0.99. Notably, the Almorox-Hontoria model exhibited the strongest positive correlation (R² = 0.993) with the A-P model, identifying it as the most reliable for estimating solar radiation. On a seasonal scale, the models generally performed well, with R² values ranging from 0.85 to 0.99. However, the Togrul-Onat and Ertekin-Xaldiz models exhibited weaker correlations with the A-P model, particularly during the Zaid season, indicating their limitations in accurately estimating average daily solar radiation during this period. These results highlight the necessity of careful model selection and calibration to account for seasonal variability. Overall, the Almorox-Hontoria model demonstrated the highest accuracy and consistency across both monthly and seasonal scales, emphasizing the importance of adjusting models to specific temporal and geographic conditions.

Keywords

• Model Performance
• Solar Radiation
• Seasonal Variation
• Regression analysis
• Angstrom-Prescott