Improving the Spatiotemporal Resolution of Land Surface Temperature Data Using Disaggregation and Fusion Techniques: A Comparison

Land Surface Temperature (LST) and its diurnal variation are important parameters for several applications. Thermal sensors in polar orbiting and geostationary orbiting satellites can provide LST data at high spatial and temporal resolutions respectively. This study aims to generate high spatiotemporal LST by combining the coarse resolution geostationary satellite data (INSAT 3D) with the Moderate Resolution Imaging Spectroradiometer (MODIS) LST product using spatial disaggregation (DisTrad model) and spatiotemporal fusion (STITFM model) techniques. In addition, the ability of these two methods to properly represent the diurnal temperature cycle (DTC) is also examined. It was found that the spatial disaggregation method provided relatively better results than spatiotemporal fusion technique in improving the spatiotemporal resolution of LST.