Hourly Air Temperature Modeling Based on Atmospheric Pressure, Global Solar Radiation and Relative Humidity Data

This paper is focusing on hourly air temperature estimation model (MAT) using available meteorological measured data located in Laghouat (Algeria). The hourly air temperature defined by the present model can be calculated at any time of the night or the day period based on atmospheric pressure, global solar radiation and relative humidity data. This work was compared with three published models from the literature as: Wave, Idliman and Double cosine. Fifteen months of hourly atmospheric pressure, global solar radiation, relative humidity and air temperature data collected during the period (January 2015 to March 2016) were used to test the accuracy of the various models studied. The analysis of the days selected randomly showed that the MAT model gave substantially good fit to the observed data. The RMSE of the MAT model is less than 0.5 oC during all the period of study than the other models studied ranged in the interval (2 oC, 4 oC). The estimated results are compared to the measured ones by using statistical parameters tests such as the mean bias error (MBE), the mean percentage error (MPE), the mean absolute error (MAE), the root mean square error (RMSE) and the coefficient of determination (R2).