Automated Mapping of Ice-wedge Polygon Troughs in the Continuous Permafrost Zone using Commercial Satellite Imagery

Abstract:

Very high spatial resolution (VHSR) commercial satellite imagery enables transformational venues to observe, map, and document the micro-topographic transitions occurring in polygonal tundra at multiple spatial and temporal frequencies. Production of imagery-enabled Arctic-science ready products at regional scales is yet new and actively evolving. The slow uptake of VHSR imagery in arctic science products is largely due to prevailing knowledge gaps in sophisticated image analysis methodology that can effectively manage the inherent scene complexities. The central objective of this exploratory study is to develop an object-based image analysis workflow to automatically extract ice-wedge polygon troughs from VHSR commercial satellite imagery. We employed a systematic experiment to understand the degree of interoperability of knowledge-based workflows across tundra cover units focusing on the same semantic class. In our multi-scale trough modelling workflow, we coupled mathematical morphological filtering with segmentation process to enhance the quality of image object candidates and classification accuracies. Results suggest that the trough modelling workflow exhibit substantial interoperability across the terrain while producing promising classification accuracies. This reflects the potential generalizability of the mapping workflow over a larger domain.