Mapping Glacier Hydrology in 3D: Novel GPR Acquisition and Processing Techniques

Ground-penetrating radar (GPR) has become a well-established tool in the field of glaciology thanks to its capacity for high-resolution imaging and the excellent propagation characteristics of radar waves in snow and ice. In this context, 3D surveying and processing techniques hold significant promise for examining the internal structure and dynamics of glaciers, yet 3D studies are rarely done due to time and cost constraints. In particular, the field of glacier hydrology could immensely benefit from the acquisition and dedicated processing of high-density 3D GPR data sets, as observations of hydrological conditions inside the glacier and at its base are of critical importance for model calibration and validation. In this contribution, we attempt to exploit the full potential of high-resolution 3D GPR data to study glacier hydrology. A novel drone-based GPR acquisition system enables us to collect high-density 3D data with unprecedented spatial coverage. Our corresponding processing scheme considers two complementary components: the prominent reflected arrivals, and the faint (often neglected) diffracted wavefield. Reflection amplitudes at the ice-bedrock interface are used to delineate subglacial channels, whereas diffraction imaging methods borrowed from exploration seismology facilitate the localization of englacial conduits. We present results from two case studies in the Swiss Alps: the Haut Glacier dArolla and the Glacier dOtemma. Our workflow provides complementary maps of the subglacial drainage system and of well-developed englacial channels. For the Glacier dOtemma, we combine these results with supplementary methods (photogrammetry, dye tracing, time lapse cameras, steam drilling, and hydrological modeling) to obtain a more comprehensive characterization of the drainage system.