A Comparative Study for the Estimation of Downward Longwave Radiation under Clear Sky Condition across the Midland Climatic Zone of Nigeria

Abstract

Downward Longwave Radiation (DLR) is a crucial component of the Earth's surface radiation budget, important for both local weather forecasting and global climate modeling. This study investigates the clear-sky DLR in Minna and Zaria, with the aim at comparing and estimating the DLR under a cloudless sky for the region. The specific objectives include examining the influence of DLR on relative humidity and air temperature, evaluating the performance of eleven widely used existing empirical models, developing five new locally calibrated regression models, and ranking all sixteen models to ascertain their reliability and applicability in the study area. Monthly average daily meteorological data of relative humidity, air temperature, and DLR, spanning a forty-year period (1984–2023) were utilized for the study. Model performance was assessed using five statistical indicators: Mean Bias Error (MBE), Root Mean Square Error (RMSE), Mean Percentage Error (MPE), t-test, and Index of Agreement (IA), supplemented by correlation coefficient (R) for the proposed modified models. For Minna, maximum DLR of 424.7500 Wm^-2 was recorded in April, while the minimum of 360.6771 Wm^-2 occurred in December. At Zaria, the peak DLR of 412.0104 Wm^-2 was observed in May, with the lowest value of 333.6875 Wm^-2 recorded in December. Among the eleven existing models evaluated, the Idso model ranked best for both Minna and Zaria with overall ranking scores of 12 and 5 respectively, while the Dilley and O'Brien model performed least. Among the five newly developed modified regression models, the modified Idso model proved most accurate for Minna, and the modified Guest model was found most suitable for Zaria. Generally, the developed modified DLR models outperformed the existing models, which can be recommended for the estimation of DLR for midland climatic zone in the absence of the measured DLR.