A simple model for estimating the daily upward longwave surface radiation flux from NOAA-AVHRR data

A simple physically-based method for estimating the daily surface outgoing longwave radiative flux during cloudless days is presented. The proposed algorithm is well-suited for remote sensing purposes, in the framework of surface net radiation and evapotranspiration studies: it only requires the NOAA-AVHRR derived surface temperature acquired in the early afternoon, the minimum air temperature easily obtainable from the meteorological network, and the day-length computed from the site coordinates and the date. The algorithm features one single empirical coefficient, adjusted statistically on a large data set generated by a coupled heat and mass transfer model between the soil and the atmosphere. The method is then validated against ten different experimental data sets, covering a wide range of surface, climate and weather conditions. The validation is performed at the daily and five-day scales and is shown to lead to a good precision. The sources of errors are then discussed and a map of the surface outgoing longwave radiation flux established from a NOAA-AVHRR image is finally presented.