Submarine melting is one of the major mechanisms of ice loss from marine-terminating glaciers and ice shelves, but its contribution is yet to be fully understood. Here, we demonstrate the feasibility of monitoring melting using passive underwater acoustics, by sensing the loud crackling sound produced during melting due to the release of pressurised ice-trapped bubbles. We profile the acoustic field in glacial bays in Svalbard using a hydrophone array, and show that the sound level in the bay contains clues on the melt activity. The sound level's interpretation is hindered by its spatial variability, which we suppress using a model of melt-induced acoustic activity. Thereby, we show that the sound generated at the glacier terminus is correlated with the ablation rate at the calving glacier front and the water temperature, and thus linked to the melt rate. This marks a step forward in using passive acoustics to monitor submarine melt, paving the way for an autonomous, long-term, large-scale monitoring tool providing data that can inform assessments and simulations of ice sheet loss and sea level rise.

The term “water pocket” describes invisible en– and subglacial water reservoirs that can cause sudden glacial outburst floods. However, there is currently no consensus on its definition and the formation and rupture mechanisms of water pockets remain poorly understood. This study aims to understand the mechanisms behind water pocket outburst floods (WPOFs) from alpine glaciers by analyzing their spatial and temporal distribution, pre-event meteorological conditions, and the glacio-geomorphic features of the glaciers from which the floods originate. To this end, we updated an inventory of known WPOFs in the Swiss Alps to 91 events from 37 individual glaciers. Most WPOFs occurred between June and September. Meteorological data indicate anomalously high temperatures during the days preceding most events and heavy precipitation on 25% of days for which WPOFs occur, indicating that water pockets typically rupture during periods of high water input. We propose four mechanisms of water pocket formation: temporary subglacial channel blockage (which is the mechanism suggested most often for our inventory), hydraulic barriers, water-filled crevasses, and accumulation of liquid water behind barriers of cold ice (thermal barriers). Overall, our analysis highlights the challenge of understanding WPOFs due to the sub-surface nature of water pockets, emphasizing the need for field-based research to improve their detection.