Super-resolution Snow-cover Mapping of Ku-Band based ISRO’s SCATSAT-1 Data using Spectral Mixture Analysis

The Scatterometer Satellite (SCATSAT-1), launched by ISRO has been utilized in many remote sensing applications to deliver near-real-time monitoring services. To represent the SCATSAT-1 applicability, the major scientific domains include hydrological studies, cryospheric applications, understanding the oceanic dynamics and agriculture applications. SCATSAT-1 offers various levels of data products with the best spatial resolution of ~2 km (Level-4), enhanced via the Scatterometer image reconstruction (SIR) technique. For the same reason, it may restrict the applicability of SCATSAT-1 in extracting different land-cover types. In this work, super-resolution mapping (SRM) has been utilized to estimate and improve the spatial distribution of land cover classes at the sub-pixel level. To achieve the SRM, the linear spectral mixing (LSM) model has been utilized in snow-cover mapping using SCATSAT-1 and MOD02 (MODIS) based normalized difference snow index (NDSI) dataset. This study has been conducted over a part of the rugged terrain Himalayan region (Himachal Pradesh State, India). Results show that satisfactory thematic maps have been generated via the SRM approach. The use of SRM is, therefore, a promising method to improve the spatial representation of the SCATSAT-1 dataset which may enhance the scope of ISRO’s SCATSAT-1 not only in the cryosphere but also in other scientific domains.