Cryospheric surface feature classification is one of the widely used applications in the field of polar remote sensing. Precise surface feature maps derived from remotely sensed imageries are the major requirement for many geoscientific applications in polar regions. The present study explores the capabilities of C-band dual polarimetric (HH & HV) SAR imagery from Indian Radar Imaging Satellite (RISAT-1) for land cryospheric surface feature mapping. The study areas selected for the present task were Larsemann Hills and Schirmacher Oasis, East Antarctica. RISAT-1 Fine Resolution STRIPMAP (FRS-1) mode data with 3-m spatial resolution was used in the present research attempt. In order to provide additional context to the amount of information in dual polarized RISAT-1 SAR data, a band HH+HV was introduced to make use of the original two polarizations. In addition to the data calibration, transformed divergence (TD) procedure was performed for class separability analysis to evaluate the quality of the statistics before image classification. For most of the class pairs the TD values were comparable, which indicated that the classes have good separability. Fuzzy and Artificial Neural Network classifiers were implemented and accuracy was checked. Nonparametric classifier Support Vector Machine (SVM) was also used to classify RISAT-1 data with an optimized polarization combination into three land-cover classes consisting of sea ice/snow/ice, rocks/landmass, and lakes/waterbodies. This study demonstrates that C-band FRS1 image mode data from the RISAT-1 mission can be exploited to identify, map and monitor land cover features in the polar regions, even during dark winter period. For better landcover classification and analysis, hybrid polarimetric data (cFRS-1 mode) from RISAT-1, which incorporates phase information, unlike the dual-pol linear (HH, HV) can be used for obtaining better polarization signatures.
Present study employs reef-up approach to map coral reef zones along the Sentinel Island of Andaman using high spectral resolution offered by hyper spectral imagery by Hyperion mission of NASA. This data consisting of 242 spectral bands, provide a unique ability to identify Coral substrate based on their spectral properties. We applied atmospheric correction with the help of Fast Line-of-sight Atmospheric Analysis of Hypercubes (FLAASH) module of ENVI software. This atmospherically corrected was used to extract Coral Reef Zones (CRZ) based on specific threshold limits after subtracting data of 782.95nm band from 579.45nm band of Hyperion imagery. Both of these bands were chosen due to their property of exhibiting maximum spectral contrast that determines threshold limits to distinguish a coral area from its non-coral counterpart. These CRZs were compared with the coral reef zones base map developed using LISS-III data by INCOIS, Hyderabad and SAC, Ahmadabad under CZS project. We observed that extracted CRZ area was 85.25 m<sup>2 </sup> and 110.1 m<sup>2</sup> using LISS-III and Hyperion Data respectively. Despite the overestimation of CRZ by Hyperion data as compared to LISS-III, the spatial distribution of CRZ showed reasonable similarity in both.