Automated delineation and morphometry of unclassified subglacial bedforms

Abstract

We designed an automated tool to delineate and analyse the shape of subglacial bedforms using a recently defined land surface parameter, the volumetric obscurance. The tool is based on the assumption that the diversity of subglacial bedform shapes reflects a continuum; therefore, unlike traditional methods, no pre- or post-mapping classification of bedforms is performed. It uses digital elevation models and optical satellite images to generate regional morphological maps (bedform outlines and crestlines) and regional morphometric maps (spatialized statistical analysis of bedform morphometrics). We tested the tool on the ArcticDEM, over a portion of the former Laurentide Ice Sheet bed that displays a wide diversity of bedform shapes (Keewatin Ice Dome, northern Canada). The produced morphological maps are consistent, with a correspondence of approximately 75% on individual bedform outlines, with two reference maps digitized manually by two different glacial geomorphologists. Despite the 25% difference between individual bedform outlines generated automatically and manually, the derived morphometric maps are similar. They can be interpreted in the context of subglacial deformation and hydrology, providing a new potential for palaeoglaciological reconstructions at the ice-sheet scale. The tool was developed in Python and is freely accessible.