ESTIMATION OF ALL-NET RADIATION OVER AKURE, SOUTH WESTERN NIGERIA

Abstract

Earth's net radiation, sometimes called net flux, is the balance between incoming and outgoing solar radiation energy from the atmosphere. This work is focused on the use of tropospheric variables to estimate net radiation and investigate its temporal (diurnal) and daily variation. In this study, a five-year data of tropospheric variables (solar radiation, air temperature, surface temperature, maximum temperature, minimum temperature and dew point temperature) which spans January 2007 to December 2011 were obtained from the on-going measurements of some weather variables by the Communication and Atmospheric Physics Research Group of the Department of Physics, FUTA. The site of the measurement is the old site of the Nigerian Television Authority, NTA Akure at Iju (7.15oN, 5.12oE), in Akure North Local Government area of Ondo State. Results obtained shows that the net radiation exhibits a radiant energy surplus during daytime and a deficit at nighttime. The daily mean maximum of the hourly net radiation flux is about 550 Wm-2 for the dry season the value occurring around 12:00 hrs and 380 Wm-2 for the wet season the value occurring around 12:30 hrs. A time lag between net radiation and temperature was also observed. The net radiative flux reaches its maximum at 12:00 hrs while the temperature reading attained its peak value around 15:00 hrs. The daily averages of net radiation was also found to have a double peak with values of about 160 Wm-2day-1 around February and November and a minimum of about 50 Wm-2day-1, while the daily average temperature has only one peak value occurring in February (the hottest month of the year). Daily net radiation values were analysed from 10 models that differ in structure and complexity. Nine models were compared with the parameterized relationship of Jegede et al., (2011) which was taken as the standard model, and most of the models were found to underestimate the Jegede’s model, because they were location dependent. The regression equations could have performed well, if locally calibrated constants were available as they are simple and do not require too much input and computational intensity.