MODELING THE SPATIAL AND TEMPORAL TEMPERATURE VARIATION ON GROUND SURFACE USING REMOTE SENSING DATA

Abstract

Remote Sensing observations could aid in estimating the land surface temperature (LST) by providing spatially representative data for surface characteristics and fluxes. Operational technique to derive temporal variation of LST using a modified Priestly-Taylor equation, which relies primarily on remotely sensed inputs, is described. A method allows extrapolating one time of day LST estimates continuously to values during the day. Meteorological data and NOAA-14 AVHRR observations were used to modeling LST maps over the Natori river basin in 1992. Model-computed LST values were in reasonable agreement with satellite-retrieved surface temperature map, where Channel 4 of the NOAA AVHRR images was available twice a day.