DEVELOPMENT OF SURFACE REFLECTANCE PRODUCTS FROM RESOURCESAT-2 SENSOR DATA AND VALIDATION USING GROUND AND LANSDAT- 8 OLI SURFACE REFLECTANCES

摘要

Surface Reflectance (SR) products are fundamental inputs for any remote sensing based application. We have implemented an atmospheric correction procedure based on Second Simulation of Satellite Signal in the Solar Spectrum (6S) radiative transfer code for the retrieval of SR products from Resourcesat-2 (RS-2) Advanced Wide Field Sensor (AWiFS) and Linear Imaging Self Scanner (LISS-3) sensor data. The correction is based on lambertian assumption and corrects for aerosol and molecular absorption and scattering effects. The atmospheric characterization parameters are retrieved from MODIS and INSAT climatology products. Coefficients that are required for atmospheric correction are delivered by the 6S code using the above inputs, which are then used to correct RS-2 imagery on a per pixel basis. The surface reflectance product thus generated was evaluated by cross comparison with contemporaneous United States Geological Survey (USGS) provisional Landsat-8 (L8) Operational Land Imager (OLI) SR product and also using ground measured spectral reflectances. Temporal validation over a wide range of target reflectances showed a good correlation between SR retrieved from Landsat-8 and the RS-2 SR product in every band, with R2 higher than 0.98. Retrieved SR values were also found to be consistent with ground based spectroradiometer measurements. The fact that atmospheric correction do matter is illustrated using NDVI(Normalized Difference Vegetation Index) as an example.