The occurrence of surface melt in Antarctica has hitherto been associated with the austral summer season, when the dominant source of melt energy is provided by solar radiation. We use in situ and satellite observations from a previously unsurveyed region to show that events of intense surface melt on Larsen C Ice Shelf occur frequently throughout the dark Antarctic winter, with peak intensities sometimes exceeding summertime values. A regional atmospheric model confirms that in the absence of solar radiation, these multiday melt events are driven by outbreaks of warm and dry föhn wind descending down the leeside of the Antarctic Peninsula mountain range, resulting in downward turbulent fluxes of sensible heat that drive sustained surface melt fluxes in excess of 200 W/m². From 2015 to 2017 (including the extreme melt winter of 2016), ∼23% of the annual melt flux was produced in winter, and spaceborne observations of surface melt since 2000 show that wintertime melt is widespread in some years. Winter melt heats the firn layer to the melting point up to a depth of ∼3 m, thereby facilitating the formation of impenetrable ice layers and retarding or reversing autumn and winter cooling of the firn. While the absence of a trend in winter melt is consistent with insignificant changes in the observed Southern Hemisphere atmospheric circulation during winter, we anticipate an increase in winter melt as a response to increasing greenhouse gas concentration.

Ice shelves play a key role in the dynamics of marine ice sheets by buttressing grounded ice and limiting rates of ice flux to the oceans. In response to recent climatic and oceanic change, ice shelves fringing the West Antarctic Ice Sheet (WAIS) have begun to fragment and retreat, with major implications for ice-sheet stability. Here, we focus on the Thwaites Eastern Ice Shelf (TEIS), the remaining pinned floating extension of Thwaites Glacier. We show that TEIS has undergone a process of fragmentation in the last 5 years, including brittle failure along a major shear zone, formation of tensile cracks on the main body of the shelf, and a release of tabular bergs on both the eastern and western flanks. Simulations with the Helsinki Discrete Element Model (HiDEM) show that this pattern of failure is associated with high backstress from a submarine pinning point at the distal edge of the shelf. We show that a significant zone of shear, upstream of the main pinning point, developed in response to the rapid acceleration of the shelf between 2002 and 2006, seeding damage on the shelf. Subsequently, basal melting and positive feedback between damage and strain rates weakened TEIS, allowing damage to accumulate. Thus, although backstress on TEIS has likely diminished over time as the pinning point shrunk, accumulation of damage has ensured that the ice in the shear zone remained the weakest link in the system. Experiments with the BISICLES ice-sheet model indicate that additional damage to or unpinning of TEIS is unlikely to trigger significantly increased ice loss from WAIS, but the calving response to the loss of TEIS remains highly uncertain. It is widely recognised that ice-shelf fragmentation and collapse can be triggered by hydrofracturing and/or unpinning from ice-shelf margins or grounding points. Our results indicate a third mechanism, backstress triggered failure, that can occur if and when an ice shelf is no longer able to withstand stress imposed by pinning points. In most circumstances, pinning points are essential for ice-shelf stability, but as ice shelves thin and weaken, the concentration of backstress in damaged ice upstream of a pinning point may provide the seeds of their demise.

Accelerating ice loss from Thwaites Glacier is contributing approximately 5% of global sea‐level rise, and could add tens of centimeters to sea level over the coming centuries. We use an ocean model to calculate sub‐ice melting for a succession of Digital Elevation Models of the main trunk of Thwaites Glacier from 2011 to 2022. The ice evolution during this period induces a strong geometrical feedback onto melting. Ice thinning and retreat provides a larger melting area, thicker and better‐connected sub‐ice water column, and steeper ice base. This leads to stronger sub‐ice ocean currents, increasing melting by over 30% without any change in forcing from wider ocean conditions. This geometrical feedback over just 12 years is comparable to melting changes arising from plausible century‐scale changes in ocean conditions and subglacial meltwater inflow. These findings imply that ocean‐driven ice loss from Thwaites Glacier may only be weakly influenced by anthropogenic emissions mitigation.

Marine-terminating outlet glacier terminus traces, mapped from satellite and aerial imagery, have been used extensively in understanding how outlet glaciers adjust to climate change variability over a range of timescales. Numerous studies have digitized termini manually, but this process is labor intensive, and no consistent approach exists. A lack of coordination leads to duplication of efforts, particularly for Greenland, which is a major scientific research focus. At the same time, machine learning techniques are rapidly making progress in their ability to automate accurate extraction of glacier termini, with promising developments across a number of optical and synthetic aperture radar (SAR) satellite sensors. These techniques rely on high-quality, manually digitized terminus traces to be used as training data for robust automatic traces. Here we present a database of manually digitized terminus traces for machine learning and scientific applications. These data have been collected, cleaned, assigned with appropriate metadata including image scenes, and compiled so they can be easily accessed by scientists. The TermPicks data set includes 39 060 individual terminus traces for 278 glaciers with a mean of 136 ± 190 and median of 93 of traces per glacier. Across all glaciers, 32 567 dates have been digitized, of which 4467 have traces from more than one author, and there is a duplication rate of 17 %. We find a median error of ∼ 100 m among manually traced termini. Most traces are obtained after 1999, when Landsat 7 was launched. We also provide an overview of an updated version of the Google Earth Engine Digitization Tool (GEEDiT), which has been developed specifically for future manual picking of the Greenland Ice Sheet.

By the end of 2018 Kangerlussuaq Glacier in southeast Greenland had retreated further inland than at any time in the past 80 years and its terminus was approaching a region of retrograde bed slope from where further rapid retreat would have been inevitable. Here we show that the retreat occurred because the glacier failed to advance during the winters of 2016/17 and 2017/18 owing to a weakened proglacial mélange. This mixture of sea ice and icebergs is normally rigid enough to inhibit calving in winter, but for 2 consecutive years it repeatedly collapsed, allowing Kangerlussuaq Glacier to continue to calve all year round. The mélange break-ups followed the establishment of anomalously warm surface water on the continental shelf during 2016, which likely penetrated the fjord. As calving continued uninterrupted from summer 2016 to the end of 2018 the glacier accelerated by 35 % and thinned by 35 m. These observations demonstrate the importance of near-surface ocean temperatures in tidewater glacier stability and show that it is not only deep-ocean warming that can lead to glacier retreat. During winter 2019 a persistent mélange reformed and the glacier readvanced by 3.5 km.