Exploring englacial hydrology with surface nuclear magnetic resonance

The amount and distribution of liquid water inside a glacier are relevant for its dynamics, related natural hazards or for sediment transport. Experimentally investigating the glacier's hydrology is challenging because of restricted accessibility, investigation depth, material properties, and environmental factors. In addition, the subglacial drainage network is highly dynamic and undergoes diurnal and seasonal changes.

This contribution investigates the application of surface nuclear magnetic resonance (SNMR) to characterize the liquid water distribution within Swiss Alpine glaciers. Analogous to magnetic resonance imaging (MRI) in medicine, SNMR utilizes an oscillating magnetic field to excite nuclear spins of hydrogen atoms within water molecules. The subsequent spin relaxation is then analyzed, providing insights into the probed material. In simpler terms, this process allows us to directly detect liquid water in ice and gain information on its spatial distribution.

We conducted a first SNMR field survey on Rhonegletscher in the summer of 2023. During this survey, we tested various measurement configurations, including separate-loop measurements and the application of noise-compensation loops. The latter proved crucial for subsequent data processing. After carefully optimizing the processing scheme, we extracted SNMR signals in several recordings despite the poor signal-to-noise ratio. The results were compared to 1D forward-modelled data, suggesting that the average water content in the survey area lay between 0.7 and 1.2 %. In addition, we could show that a homogenous water distribution over the entire ice column cannot explain the observed data and that we need to consider more complex subsurface models including at least one additional water layer. Specifically, our ongoing research aims to identify which configurations of the subglacial water distribution (e.g., homogenous water distribution vs layered water-ice structure resulting from an englacial water channel) are distinguishable experimentally. Moreover, the study seeks to optimize measurement design and data processing methodologies to acquire information more efficiently, and effectively handle the expected low signal-to-noise ratios.

In future field campaigns, we intend to deploy SNMR for selected glaciological case studies within the Swiss Alps. A primary focus will be on efficiently detecting water pockets that may pose a potential risk of downstream flooding upon rupture. Similarly, we want to investigate the extent to which we can distinguish cold from temperate ice.