The Dotson Ice Shelf has resisted acceleration and ice-front retreat despite high basal-melt rates and rapid disaggregation of the neighboring Crosson Ice Shelf. Because of this lack of acceleration, previous studies have assumed that Dotson is stable. Here we show clear evidence of Dotson's destabilization as it decelerates, contrary to the common assumption that ice-flow deceleration is synonymous with stability. Ungrounding of a series of pinning points initiated acceleration in the Upper Dotson in the early 2000s, which subsequently slowed ice flow in the Lower Dotson. Discharge from the tributary Kohler Glacier into Crosson increased, but non-proportionally. Using ICESat and ICESat-2 altimetry data we show that ungrounding of the remaining pinning points is linked to a tripling in basal melt rates between 2006-2016 and 2016-2020. Basal melt rates on Crosson doubled over the same period. The higher basal melt at Lower Dotson is consistent with the cyclonic ocean circulation in the Dotson cavity, which tends to lift isopycnals and allow warmer deep water to interact with the ice. Given current surface-lowering rates, we estimate that several remaining pinning points in the Upper Dotson will unground within one to three decades. The grounding line of Kohler Glacier will retreat past a bathymetric saddle by the late 2030s and merge into the Smith West Glacier catchment, raising concern that reconfiguration of regional ice-flow dynamics and new pathways for the intrusion of warm modified Circumpolar Deep Water could further accelerate grounding-line retreat in the Dotson-Crosson Ice Shelf System.

Outlet glaciers account for almost half of the Greenland Ice Sheet’s mass loss since 1990. Warming subsurface Atlantic Water (AW) has been implicated in much of that loss, particularly along Greenland’s southeastern coast. However, oceanographic observations are sparse prior to the last decade, making it difficult to diagnose changes in AW properties reaching the glaciers. Here, we investigate the use of sea surface temperatures (SST) to quantify ocean temperature variability on the continental shelf near Sermilik Fjord and Helheim Glacier. We find that after removing the short-term, atmospheric-driven variability in non-winter months, regional SSTs provide a reliable upper ocean temperature record. In the trough region near Sermilik Fjord, the adjusted SSTs correlate well with moored ocean measurements of the water entering the fjord at depth and driving glacier melting. Using this relationship, we reconstruct the AW variability on the shelf dating back to 2000, eight years before the first mooring deployments. Seasonally, AW reaches close to the fjord’s mouth in fall and winter and further offshore in spring. Interannually, the AW temperatures in the trough do not always track properties in the source waters of the Irminger Current. Instead, the properties of the waters found at the fjord mouth depend on both variations in the source AW and, also, in the Polar Water that flows into the region from the Arctic Ocean. Satellite-derived SSTs, although dependent on local oceanography, have the potential to improve understanding around previously unanswered glacier-ocean questions in areas surrounding Greenland and Antarctica.

Enhanced transport of warm subsurface Atlantic Waters (AW) into Greenland fjords has driven glacier mass loss, but the mechanisms transporting AW to the fjords remain poorly characterized. Here, we identify a wind-driver for AW inflow toward Sermilik Fjord abutting Helheim Glacier, one of Greenland’s largest glaciers. Often associated with the passing of cyclones and subsequent sea surface lowering, a weakening or reversal of northeasterly alongshore winds stimulates coastal ocean upwelling that, through interactions with Sermilik’s bathymetric trough on the continental shelf, leads to enhanced AW upwelling and inflow along the trough. These intrusions produce ocean warming at deep moorings near Sermilik Fjord mouth (0.31±0.18°C) and within the fjord (250m: 0.30±0.19°C; 350m: 0.17±0.09°C) that is not diminished by subsequent coastal downwelling. Similar wind-driven processes at other bathymetric trough regions around Greenland may play a substantial role in ocean heat transport towards much of the Greenland Ice Sheet.