FIVE CM SOIL TEMPERATURE ESTIMATION USING LANDSAT IMAGES AND WEATHER SYNOPTIC STATION DATA

The thermal behavior of the soil (near surface) as a function of depth and time is difficult to estimate due to the involvement of many parameters such as short term weather variations, seasonal variations, soil moisture content, soil thermal conductivity etc. Space-based remote sensing offers great potential in contributing large-scale monitoring of the Earth s surface temperature. If properly utilized, satellite systems can offer the spatial, temporal, and spectral resolution necessary for consistent and continuous coverage of the whole Earth environment and its surrounding atmospheric parameters including temperature. Land surface temperature may be derived by several remote sensing methods, including infrared and microwave techniques. However, satellite sees only the very surface while in this work we intend to retrieve the temperature at 5cm depth. The method involves soil temperature at 5cm depth measured at the synoptic weather station, soil moisture at 5cm depth measured at the synoptic weather station and the distance between this station and the pixel. The RMSE of the estimated soil temperature at 5cm depth and those measure at synoptic weather station was found to be 1.1 Celsius with a correlation coefficient (R) of 0.99. it is believed that this temperature can be efficiently used in surface energy balance calculation.