Surficial materials mapping in Nunavut, Canada with multibeam RADARSAT-2 dual-polarization C-HH and C-HV, LANDSAT-7 ETM+, and DEM data

Our study assesses the use of multibeam RADARSAT-2 dual-polarization C-HH and C-HV, LANDSAT-7 ETM+, and Digital Elevation Model (DEM) data for mapping surficial materials in Nunavut, Canada. The RADARSAT-2 images were acquired using standard beam 1 and 7 modes in both ascending and descending orbits. RADARSAT-2 images were filtered to reduce speckle and then orthorectified. Representative training areas of distinct surficial deposits (bedrock, boulders, organic deposits, sand and gravel, thick till with dense vegetation, thick till with sparse vegetation, and thin till) were identified from field information and by interpreting panchromatic aerial photographs and LANDSAT-7 ETM+ images. Maximum likelihood supervised classifications were conducted on the RADARSAT-2 C-HH and C-HV images, individually and combined with LANDSAT-7 ETM+ and (or) DEM data. The best overall classification accuracy was obtained by combining RADARSAT-2 C-HH and C-HV images with LANDSAT-7 ETM+ and DEM data. Confusion between several surface materials was reduced, but confusion between "bedrock" and "boulders" and between "sand and gravel" and "organic deposits" or "thin till" class still remains. Another limitation of the study includes the lack of a field survey to validate training areas for three of the four detailed analysis areas. Nonetheless, this study produced surficial materials maps covering eight 1:250 000 scale National Topographic System of Canada sheets that provide important geological information about this remote area.