Investigation of land surface temperature algorithms using NOAA AVHRR images Original Research Article

Several split-window equations have been developed to derive the land surface temperature (LST) using NOAA/AVHRR images. In this work, we present the results of a comparative test of LST algorithms. The calculation of land surface temperature requires the knowledge of the land surface emissivity. First, we estimated the land surface emissivity using satellite information by two different methods: using the temperature independent thermal infrared spectral indices (TISI) and the correlation between the emissivity and normalized difference vegetation index (NDVI). The comparison of these two methods to calculate the emissivity was performed for the area of Carpathian Basin. We investigated the sensitivity of the different split-window equations to the emissivity values. The temperatures derived by the different split-window equations were compared to ground truth measurements.