EMPIRICAL MODELING OF PHOTOSYNTHETICALLY ACTIVE RADIATION (PAR) USING WIND SPEED AND AEROSOL DATA IN MAKURDI, NIGERIA

Abstract

This study introduces a novel data-driven approach to PAR estimation by integrating wind speed and aerosol dynamics, addressing gaps in tropical Savannah regions in Nigeria. Recognizing the critical role of PAR in agricultural productivity and ecological dynamics, the research investigates the reliability of using wind speed and aerosol concentrations including PM_2.5, black carbon and dust as predictive variables for PAR estimation. Eight empirical models were developed using statistical modeling and validated with statistical metrics such as Root Mean Square Error (RMSE), Mean Absolute Error (MAE), Mean Absolute Percentage Error (MAPE) and Pearson’s correlation coefficient, R. Results obtained show that Model 4 emerged as the most accurate and reliable, exhibiting the lowest error values and strongest correlation with observed data. In contrast, some models with high theoretical correlation, such as Model 5, performed poorly in prediction accuracy, underscoring the complexity of atmospheric interactions affecting PAR. The findings highlight the significance of model validation and selection tailored to local climatic conditions. This research contributes to the advancement of cost-effective PAR estimation methods and provides valuable insights for agricultural planning, environmental monitoring, and climate-smart strategies in sub-Saharan Africa.