Stereoradargrammetric generation of digital surface models with RADARSAT-2 in Northern Canada

Remote Predictive Mapping can help identifying potential sites for mineral exploration. Elevation information is necessary for this purpose, but current freely available elevation data north of the 60^th parallel needs improvement. This paper presents a method for creating stereoradargrammetric digital surface models (DSM) using Radarsat-2 (RS-2) images. The study site is a 23 000 km² area located in the Northwest Territories, Canada. Sub-DSM were generated from all configurations using 136 wide-ultrafine RS-2 images. Full DSM were then assembled following different criteria based on the configurations and accuracy. Assembling was key in improving the accuracy, with improvements ranging from 30% to 55%, with respect to the sub-DSM. Using opposite-side view configurations resulted in full coverage of the study site, while using only 70% of the images originally ordered. The method presented in this study can be considered an economical way to improve currently available data north of the 60^th parallel.