Ice Flow Research Articles

Terminus thinning drives recent acceleration of a Greenlandic lake-terminating outlet glacier

Abstract Ice-contact proglacial lakes affect ice dynamics and the transition of glacier termini from land- to lake-terminating has been shown to cause ice flow acceleration. In recent decades, the number and size of Greenlandic ice-marginal lakes has increased, highlighting the need to further understand these lake-terminating ice-margins as their influence on ice sheet mass balance increases. Here, time series of satellite-derived observations of ice velocity, surface elevation, and terminus position were generated at a lake-terminating outlet glacier, Isortuarsuup Sermia, and the nearby land-terminating Kangaasarsuup Sermia in south-west Greenland. At Isortuarsuup Sermia, annual surface velocity at the terminus increased by a factor of 2.5 to 214 ± 4 m yr−1 (2013–2021), with the magnitude of this acceleration declining with distance up-glacier. Meanwhile, near-terminus surface elevation changed at a rate of −2.3 ± 1.1 m yr−1 (2012–2021). Conversely, velocity change at Kangaasarsuup Sermia was minimal, while surface elevation change was approximately half at comparable elevations (−1.2 ± 0.3 m yr−1). We attribute these dynamic differences to thinning at Isortuarsuup Sermia and subsequent retreat from a stabilising sublacustrine moraine, and emphasise the potential of proglacial lakes to enhance future rates of mass loss from the Greenland Ice Sheet.

Ross Ice Shelf Displacement and Elastic Plate Waves Induced by Whillans Ice Stream Slip Events

AbstractIce shelves are assumed to flow steadily from their grounding lines to the ice front. We report the detection of ice‐propagating extensional Lamb (plate) waves accompanied by pulses of permanent ice shelf displacement observed by co‐located Global Navigation Satellite System receivers and seismographs on the Ross Ice Shelf. The extensional waves and associated ice shelf displacement are produced by tidally triggered basal slip events of the Whillans Ice Stream, which flows into the ice shelf. The propagation velocity of 2,800 m/s is intermediate between shear and compressional ice velocities, with velocity and particle motions consistent with predictions for extensional Lamb waves. During the passage of the Lamb waves the entire ice shelf is displaced about 60 mm with a velocity more than an order of magnitude above its long‐term flow rate. Observed displacements indicate a peak dynamic strain of 10−7, comparable to that of earthquake surface waves that trigger ice quakes.

Applications of ArcticDEM for measuring volcanic dynamics, landslides, retrogressive thaw slumps, snowdrifts, and vegetation heights

Topographical changes are of fundamental interest to a wide range of Arctic science disciplines faced with the need to anticipate, monitor, and respond to the effects of climate change, including geohazard management, glaciology, hydrology, permafrost, and ecology. This study demonstrates several geomorphological, cryospheric, and biophysical applications of ArcticDEM – a large collection of publicly available, time-dependent digital elevation models (DEMs) of the Arctic. Our study illustrates ArcticDEM's applicability across different disciplines and five orders of magnitude of elevation derivatives, including measuring volcanic lava flows, ice cauldrons, post-failure landslides, retrogressive thaw slumps, snowdrifts, and tundra vegetation heights. We quantified surface elevation changes in different geological settings and conditions using the time series of ArcticDEM. Following the 2014–2015 Bárðarbunga eruption in Iceland, ArcticDEM analysis mapped the lava flow field, and revealed the post-eruptive ice flows and ice cauldron dynamics. The total dense-rock equivalent (DRE) volume of lava flows is estimated to be (1431 ± 2) million m3. Then, we present the aftermath of a landslide in Kinnikinnick, Alaska, yielding a total landslide volume of (400 ± 8) × 103 m3 and a total area of 0.025 km2. ArcticDEM is further proven useful for studying retrogressive thaw slumps (RTS). The ArcticDEM-mapped RTS profile is validated by ICESat-2 and drone photogrammetry resulting in a standard deviation of 0.5 m. Volume estimates for lake-side and hillslope RTSs range between 40,000 ± 9000 m3 and 1,160,000 ± 85,000 m3, highlighting applicability across a range of RTS magnitudes. A case study for mapping tundra snow demonstrates ArcticDEM's potential for identifying high-accumulation, late-lying snow areas. The approach proves effective in quantifying relative snow accumulation rather than absolute values (standard deviation of 0.25 m, bias of −0.41 m, and a correlation coefficient of 0.69 with snow depth estimated by unmanned aerial systems photogrammetry). Furthermore, ArcticDEM data show its feasibility for estimating tundra vegetation heights with a standard deviation of 0.3 m (no bias) and a correlation up to 0.8 compared to the light detection and ranging (LiDAR). The demonstrated capabilities of ArcticDEM will pave the way for the broad and pan-Arctic use of this new data source for many disciplines, especially when combined with other imagery products. The wide range of signals embedded in ArcticDEM underscores the potential challenges in deciphering signals in regions affected by various geological processes and environmental influences.

Ice volume and thickness of all Scandinavian glaciers and ice caps

Abstract We present a new map of bed topography and ice thickness together with a corresponding ice volume estimate representative of the years ~2010 for all Scandinavian ice caps and glaciers. Starting from surface observations, we invert for ice thickness by iteratively running an innovative ice dynamics model on a distributed grid and updating bed topography until modelled and observed glacier dynamics as represented by their rate of surface elevation change (dh/dt) fields align. The ice flow model used is the instructed glacier model (Jouvet and Cordonnier, 2023, Journal of Glaciology 1–15), a generic physics-informed deep-learning emulator that models higher-order ice flow with high-computational efficiency. We calibrate the modelled thicknesses against >11 000 ice thickness observations, resulting in a final ice volume estimate of 302.7 km3 for Norway, 18.4 km3 for Sweden and 321.1 km3 for the whole of Scandinavia with an error estimate of ~ $\pm 11\%$ . The validation statistics computed indicate good agreement between modelled and observed thicknesses (RMSE = 55 m, Pearson's r = 0.87, bias = 0.8 m), outperforming all other ice thickness maps available for the region. The modelled bed shapes thus provide unprecedented detail in the subglacial topography, especially for ice caps where we produce the first maps that show ice-dynamically realistic flow features.

Evolution of sub-ice-shelf channels reveals changes in ocean-driven melt in West Antarctica

Abstract Basal channels, which are troughs carved into the undersides of ice shelves by buoyant plumes of water, are modulators of ice-shelf basal melt and structural stability. In this study, we track the evolution of 12 large basal channels beneath ice shelves of the Amundsen and Bellingshausen seas region in West Antarctica using the Landsat record since its start in the 1970s through 2020. We observe examples of channel growth, interactions with ice-shelf features, and systematic changes in sinuosity that give insight into the life cycles of basal channels. We use the last two decades of the record, combined with contemporary ice-flow velocity datasets, to separate channel-path evolution into components related to advection by ice flow and those controlled by other forcings, such as ocean melt or surface accumulation. Our results show that ice-flow-independent lateral channel migration is overwhelmingly to the left when viewed down-flow, suggesting that it is dominated by Coriolis-influenced ocean melt. By applying a model of channel-path evolution dominantly controlled by ice flow and ocean melt, we show that the majority of channels surveyed exhibit non-steady behavior that serves as a novel proxy for increased ocean forcing in West Antarctica starting at least in the early 1970s.

Reducing uncertainty in source area exploration of mineralized glacial erratics using terrestrial cosmogenic radionuclide dating

Mineral exploration often relies on sedimentary indicators as first signs of a potential nearby mineralization in bedrock. The transportation of sediment by glacial and fluvial processes introduces uncertainty into the tracing of potential source areas of the sediment. This is particularly challenging in areas where glacial erosion has been weak, resulting in reworking of sediments and multiple directions of transportation. In this study we explore the use of Terrestrial Cosmogenic Nuclides (TCN) alongside conventional mineral exploration methods to better link indicators with their potential source areas.The study focuses on a mineral exploration project site in northern Finland in a region with stratigraphic evidence of multiple glaciations. The project targets the source of discovered Au mineralized erratic boulders that have been deposited atop the most recent till unit. The superposition suggests transportation in the latest glacial event. However, the 10Be (35.9 ± 1.3 and 30.3 ± 1.1 ka) and 26Al TCN ages analyzed from the mineralized erratics precede the latest glacial event, suggesting the possibility of multiple stages of transportation. The local bedrock TCN inventories (48.9 ± 1.9 and 85.3 ± 2.8 ka) are well preserved, suggestive of weak glacial erosion and therefore short recent transportation distances for the mineralized erratics. By combining the TCN interpretation and historical ice flow directions derived from the till stratigraphy, we suggest a nearby source locating NNW of the dated erratics.We find that TCN dating can limit some of the transportation related uncertainties in glaciated terrain. Through the analysis of TCN inventories from bedrock and boulders it is possible to characterize glacial erosion and boulder transportation, and to identify repeated exposure events, i.e., the possibility of multi-staged transportation. The method benefits from combined use of stratigraphic investigations that can identify both the local transportation directions and local glacial coverage history. TCN sampling has a minimal environmental impact, can be used in remote areas and can provide information about the transport history already in the early stages of exploration. Although the analysis is time consuming, the survey is light to conduct and informative even with a small number of samples.

Impact of boundary conditions on the modeled thermal regime of the Antarctic ice sheet

Abstract. A realistic initialization of ice flow models is critical for predicting future changes in ice sheet mass balance and their associated contribution to sea level rise. The initial thermal state of an ice sheet is particularly important, as it controls ice viscosity and basal conditions, thereby influencing the overall ice velocity. Englacial and subglacial conditions, however, remain poorly understood due to insufficient direct measurements, which complicate the initialization and validation of thermal models. Here, we investigate the impact of using different geothermal heat flux (GHF) datasets and vertical velocity profiles on the thermal state of the Antarctic ice sheet and compare our modeled temperatures to in situ measurements from 15 boreholes. We find that the temperature profile is more sensitive to vertical velocity than to GHF. The basal temperature of grounded ice and the amount of basal melting are influenced by both selection of GHF and vertical velocity. More importantly, we find that the standard approach, which consists of combining basal sliding speed and incompressibility to derive vertical velocities, provides reasonably good results in fast-flow regions (ice velocity >50 m yr−1) but performs poorly in slow-flow regions (ice velocity <50 m yr−1). Furthermore, the modeled temperature profiles in ice streams, where bed geometry is generated using a mass conservation approach, show better agreement with observed borehole temperatures compared to kriging-based bed geometry.

The staggered retreat of grounded ice in the Ross Sea, Antarctica, since the Last Glacial Maximum (LGM)

Abstract. The retreat of the West Antarctic Ice Sheet (WAIS) in the Ross Sea after the Last Glacial Maximum (LGM) was more significant than for any other Antarctic sector. Here we combined the available chronology of retreat with new mapping of seismically resolvable grounding zone wedges (GZWs). Mapping GZWs is important because they record the locations of former stillstands in the extent of grounded ice for individual ice streams during the overall retreat. Our analysis shows that the longest stillstands occurred early in the deglacial period and had millennial durations. Stillstands ended abruptly with retreat distances measured in the tens to hundreds of kilometers creating deep embayments in the extent of grounded ice across the Ross Sea. The location of embayments shifted through time. The available chronological data show that cessation of WAIS and East Antarctic Ice Sheet (EAIS) stillstands was highly asynchronous across at least 5000 years. There was a general shift to shorter stillstands throughout the deglacial period. The asynchronous collapse of individual catchments during the deglacial period suggests that the Ross Sea sector would have contributed to multiple episodes of relatively small-amplitude sea-level rise as the WAIS and EAIS retreated from the region. The high sinuosity of the modern grounding zone in the Ross Sea suggests that this style of retreat persists.

60-YEAR REMOTE SENSING OBSERVATIONS REVEALED CROSS-CYCLE EVOLUTION OF THE BRUNT ICE SHELF IN THE CONTEXT OF GLOBAL WARMING

Abstract. The Antarctic Ice Sheet (AIS) holds the potential to raise global sea levels by a staggering 57 m. Ice shelves, extensions of these ice sheets into the ocean, play a crucial role in stabilizing the upstream ice sheet. The Brunt Ice Shelf (BIS), located in the Southern Weddell Sea region and proximal to the continental shelf break, stands as a vital indicator for the region's response to global climate warming. Hence, this study systematically employs a range of remote sensing and field data, including remote sensing data from 1963 to 2023 (ARGON images, Landsat1-9 images, Sentinel-1 images, ICESat-2 altimetry data, etc.) and in-situ ice flow velocity data since 1956, to conduct a comprehensive analysis of the BIS's calving cycle and evolution over nearly six decades. The results reveal a significant exacerbation of both dynamical and structural instabilities in the BIS compared to the calving cycle in the 1970s. Combining analyses of air temperature from meteorological station, water temperature from ocean models and in-situ measurements such as Conductivity-Temperature-Depth profiles, the study suggests that the increased instability of the BIS is likely linked to the intrusion of warmer currents over the continental shelf. This research provides valuable insights into the evolving behavior of the BIS in response to changing climate conditions, underscoring the critical role of remote sensing data in advancing our understanding of these processes on a global scale.

The effect of landfast sea ice buttressing on ice dynamic speedup in the Larsen B embayment, Antarctica

Abstract. We observe the evacuation of 11-year-old landfast sea ice in the Larsen B embayment on the East Antarctic Peninsula in January 2022, which was in part triggered by warm atmospheric conditions and strong offshore winds. This evacuation of sea ice was closely followed by major changes in the calving behaviour and dynamics of a subset of the ocean-terminating glaciers in the region. We show using satellite measurements that, following a decade of gradual slow-down, Hektoria, Green, and Crane glaciers sped up by approximately 20 %–50 % between February and the end of 2022, each increasing in speed by more than 100 m a−1. Circumstantially, this is attributable to their transition into tidewater glaciers following the loss of their ice shelves after the landfast sea ice evacuation. However, a question remains as to whether the landfast sea ice could have influenced the dynamics of these glaciers, or the stability of their ice shelves, through a buttressing effect akin to that of confined ice shelves on grounded ice streams. We show, with a series of diagnostic modelling experiments, that direct landfast sea ice buttressing had a negligible impact on the dynamics of the grounded ice streams. Furthermore, we suggest that the loss of landfast sea ice buttressing could have impacted the dynamics of the rheologically weak ice shelves, in turn diminishing their stability over time; however, the accompanying shifts in the distributions of resistive stress within the ice shelves would have been minor. This indicates that this loss of buttressing by landfast sea ice is likely to have been a secondary process in the ice shelf disaggregation compared to, for example, increased ocean swell or the drivers of the initial landfast sea ice disintegration.

Subsurface warming in the Antarctica’s Weddell Sea can be avoided by reaching the 2∘C warming target

Recently, seasonal pulses of modified Warm Deep Water have been observed near the Filchner Ice Shelf front in the Weddell Sea, Antarctica. Here, we investigate the temperature evolution of subsurface waters in the Filchner Trough under four future scenarios of carbon dioxide emissions using the climate model AWI-CM. Our model simulates these warm intrusions, suggests more frequent pulses in a warmer climate, and supports the potential for a regime shift from cold to warm Filchner Trough in two high-emission scenarios. The regime shift is governed in particular by decreasing local sea ice formation and a shoaling thermocline. Cavity circulation is not critical in triggering the change. Consequences would include increased ice shelf basal melting, reduced buttressing of fast-flowing ice streams, loss of grounded ice and an acceleration of global sea level rise. According to our simulations, the regime shift can be avoided and the Filchner Trough warming can be restricted to 0.5 ∘C by reaching the 2 ∘C climate goal.

Geomorphological analysis of grounding-zone wedges reveals glacial retreat processes on the Labrador Shelf

Newly acquired high-resolution geophysical data from four glacial cross-shelf troughs of the Labrador Shelf, namely the Okak, Hopedale, Makkovik and Cartwright troughs, are presented in this study. These cross-shelf troughs were repeatedly excavated by the former Laurentide Ice Sheet (LIS) during past glaciations. We map and describe three previously unknown grounding-zone wedges (GZWs) from an area where only sparse information about the retreat of the LIS is available. Based on our geomorphological analysis, we derive the dynamics and the direction of the ice retreat. For the Okak, Makkovik, and Cartwright troughs, the newly identified GZWs indicate fast-flowing ice streams with a high sediment load. We propose that a stable ice grounding-zone paralleled the present-day coastline in the Okak and Makkovik troughs, while in the Cartwright Trough, an ice-margin retreat towards the outlet of Lake Melville is more likely. We date available sediment samples of the deposits that onlap one of the GZWs to 14.5 ka cal BP and can show that the discovered GZWs must be older. This new information updates our understanding of the dynamics of the LIS, particularly at its eastern boundary, the Labrador Shelf and Sea, which is important to better reconstruct the effects of the LIS dynamics on global ocean currents and climate.

Link between glacial striation morphology and induced drag

Abrasion acts to smooth glacial terrains and leaves behind linear scratch-like features (striations) on bedrock landscapes. Striations are often used as measures of glacier flow directions, but their morphology can also provide information about the subglacial stress conditions that produced the features. While striations are often abundant in the field, the processes that create them can be opaque and hard to examine in situ because they occur under thick layers of flowing ice. To alleviate that difficulty and provide information for interpretation of the populations of striations that are observed in the field, we conducted a set of laboratory experiments in which a ring of temperate debris-laden ice was slid atop a planar marble bed under various contact force conditions that led to the creation of hundreds of striations. During the experiment, numerous glaciological properties were continuously measured, including the resistive drag. Following the completion of the experiments, the marble beds were extracted, and the striations were measured for length and categorized by morphological type, and a subset was measured using a high-resolution white-light profilometer. These experiments showed that, similar to field observations, type 2 striations were initially the most abundant; however, we found that type 3 striations became the most abundant at large displacements. We found good correlation between the abundance of striations as a function of displacement and measured drag as a function of displacement. When taken together, these results suggest that, in natural settings, ice flow around small roughness elements in glacier beds can “reset” the basal debris field, causing striations to become more abundant in their wake. As roughness is linked to quarrying, abrasion rates may increase in areas of increased quarrying.

Laurentide Ice Sheet configuration in southern Ontario, Canada during the last glaciation (MIS 4 to 2) from stratigraphic drilling and LIDAR-based surficial mapping

Regional subsurface mapping of glacial depositional systems preserved in buried bedrock paleovalleys, and quantitative analysis of new LiDAR imagery of surface glacial landforms using machine learning techniques, when combined, are powerful tools for assessing the dynamics of the Laurentide Ice Sheet (LIS) during the last (Wisconsinan) glaciation in southern Ontario. While age dating of deposits preserved below Last Glacial Maximum tills (LGM: marine isotope stage (MIS) 2 < c.24 000 years B.P. (ybp)) is still sparse, newly available sedimentological data derived by cored drilling, combined with legacy outcrop data, identify thick (100 m+) successions of glaciolacustrine sediments and a lack of till(s), indicating that the ice sheet margin did not extend beyond the Niagara Escarpment at the western end of Lake Ontario, during the earliest phases of the glaciation (MIS 4) or the ensuing mid-Wisconsinan (MIS 3). Ice was able to extend into New York State blocking the Rome outlet to the Hudson Valley ponding deep proglacial lakes in the glacio-isostatically depressed Huron–Ontario–Erie basins recorded by thick glaciolacustrine sediments in paleovalleys. These were cannibalized by an expanding Late Wisconsinan ice sheet after ∼24 000 ybp recorded by extensive till sheets resting on a marked erosional unconformity, with drumlinized surfaces. Analysis and visualization of LiDAR data identifies discrete statistically validated flow sets of highly elongated streamlined bedforms (mega-scale glacial lineations (MSGLs)). These provide key evidence of a major reorganization of the ice sheet margin during deglaciation into lobate paleo ice streams shortly after 17 400 ybp. MSGLs are cut across earlier LGM drumlinized tills creating widespread “palimpsest” surfaces. At least two principal phases of fast ice flow can be identified, marked by large fluxes of sediment and the rapid building of large gravel and sand-dominated moraine complexes within interlobate depocentres, the largest glacial landforms in southern Ontario. Analysis of LiDAR data further reveals the common presence of DeGeer moraines where ice margins retreated in water, and iceberg scours. Future work using LiDAR mapping has the objective of fully documenting the number, extent, and timing of ice streams to enhance glaciological modelling when the ice sheet rapidly lost mass.

Shear margins in upper half of Northeast Greenland Ice Stream were established two millennia ago

Only a few localised ice streams drain most of the ice from the Greenland Ice Sheet. Thus, understanding ice stream behaviour and its temporal variability is crucially important to predict future sea-level change. The interior trunk of the 700 km-long North-East Greenland Ice Stream (NEGIS) is remarkable due to the lack of any clear bedrock channel to explain its presence. Here, we present a 3-dimensional analysis of the folding and advection of its stratigraphic horizons, which shows that the localised flow and shear margins in the upper NEGIS were fully developed only ca 2000 years ago. Our results contradict the assumption that the ice stream has been stable throughout the Holocene in its current form and show that upper NEGIS-type development of ice streaming, with distinct shear margins and no bed topography relationship, can be established on time scales of hundreds of years, which is a major challenge for realistic mass-balance and sea-level rise projections.

Abrupt Holocene ice loss due to thinning and ungrounding in the Weddell Sea Embayment

The extent of grounded ice and buttressing by the Ronne Ice Shelf, which provides resistance to the outflow of ice streams, moderate West Antarctic Ice Sheet stability. During the Last Glacial Maximum, the ice sheet advanced and was grounded near the Weddell Sea continental shelf break. The timing of subsequent ice sheet retreat and the relative roles of ice shelf buttressing and grounding line changes remain unresolved. Here we use an ice core record from grounded ice at Skytrain Ice Rise to constrain the timing and speed of early Holocene ice sheet retreat. Measured δ18O and total air content suggest that the surface elevation of Skytrain Ice Rise decreased by about 450 m between 8.2 and 8.0 kyr before 1950 ce (±0.13 kyr). We attribute this elevation change to dynamic thinning due to flow changes induced by the ungrounding of ice in the area. Ice core sodium concentrations suggest that the ice front of this ungrounded ice shelf then retreated about 270 km (±30 km) from 7.7 to 7.3 kyr before 1950 ce. These centennial-scale changes demonstrate how quickly ice mass can be lost from the West Antarctic Ice Sheet due to changes in grounded ice without extensive ice shelf calving. Our findings both support and temporally constrain ice sheet models that exhibit rapid ice loss in the Weddell Sea sector in the early Holocene.

Palaeoglacier and palaeoclimate reconstruction since the Neoglacial of the central Gangdise Mountains, southern Tibetan Plateau

Palaeoglacier modelling is a crucial approach for reconstructing palaeoclimate. Although the timing of glaciations during the Little Ice Age (LIA) and Neoglacial in the central Gangdise Mountains has been partially determined, there are still undated fresh-looking moraines. This study utilized coupled mass balance and ice flow models to simulate the palaeoclimate during the five LIA glacial events (LIA-1 to LIA-5) and two Neoglacial glacial events (Neo-1 and Neo-2). The results revealed that the glacial volumes in the Gangdise Mountains during the LIA and Neoglacial periods ranged from ∼1.10 to 1.57 km3. Comparing the reconstructed climate with other proxies, it was found that Neo-2 and Neo-1 exhibited temperatures 0.54 to 0.82 °C and 0.7 to 1.0 °C lower than the present, respectively, with precipitation levels at 110–140% of the present. The temperatures during the LIA periods were 0.36 to 1 °C lower than the present, with precipitation levels at 80–120% of the present. The interannual variability of climate at decadal or centennial scales may have played a significant role in the formation of the LIA moraines.

Validating ensemble historical simulations of Upernavik Isstrøm (1985–2019) using observations of surface velocity and elevation

Abstract The future of tidewater glaciers in response to climate warming is one of the largest sources of uncertainty in the contribution of the Greenland ice sheet to global sea-level rise. In this study, we investigate the ability of an ice-sheet model to reproduce the past evolution of the velocity and surface elevation of a tidewater glacier, Upernavik Isstrøm, by prescribing front positions. To achieve this, we run two ensembles of simulations with a Weertman and a regularised-Coulomb friction law. We show that the ice-flow model has to include a reduction in friction in the first 15 km upstream of the ice front in fast-flowing regions to capture the trends observed during the 1985–2019 period. Without this process, the ensemble model overestimates the ice flow before the retreat of the front in 2005 and does not fully reproduce its acceleration during the retreat. This results in an overestimation of the total mass loss between 1985 and 2019 of 50% (300 vs 200 Gt). Using a variance-based sensitivity analysis, we show that uncertainties in the friction law and the ice-flow law have a greater impact on the model results than surface mass balance and initial surface elevation.

Sub-regional variability in the influence of ice-contact lakes on Himalayan glaciers

Abstract Ice-contact lakes modify glacier geometry and dynamics by shifting the majority of mass loss from the ice surface to the terminus. Lake-terminating glaciers are known to experience greater thinning rates and higher velocities than land-terminating glaciers, but the controls on variability in surface elevation change and ice flow between lake-terminating glaciers in different regions remain poorly explored. We combined existing datasets of glacier velocity, surface elevation change and glacial lake area to characterise the evolution of 352 lake-terminating and land-terminating glaciers within three Himalayan sub-regions between 2000 and 2019. These analyses show that the influence of ice-contact lakes propagates up-glacier across only the lowermost 30% of the hypsometric distribution, even where lakes are well established. We find that ice-contact lakes only affect glacier behaviour when the lakes reach an advanced evolutionary stage; most clearly manifested in the Eastern Himalaya by statistically robust differences in glacier-wide surface elevation change between lake-terminating (–0.68 ± 0.05 m a–1) and land-terminating (–0.54 ± 0.04 m a–1) glaciers. These differences are driven by the presence of a greater number of well-developed ice-contact lakes in the Eastern Himalaya compared to in the Western and Central Himalaya, resulting from greater mass loss rates to date.

Graphics-processing-unit-accelerated ice flow solver for unstructured meshes using the Shallow-Shelf Approximation (FastIceFlo v1.0.1)

Abstract. Ice-sheet flow models capable of accurately projecting their future mass balance constitute tools to improve flood risk assessment and assist sea-level rise mitigation associated with enhanced ice discharge. Some processes that need to be captured, such as grounding-line migration, require high spatial resolution (under the kilometer scale). Conventional ice flow models mainly execute on central processing units (CPUs), which feature limited parallel processing capabilities and peak memory bandwidth. This may hinder model scalability and result in long run times, requiring significant computational resources. As an alternative, graphics processing units (GPUs) are ideally suited for high spatial resolution, as the calculations can be performed concurrently by thousands of threads, processing most of the computational domain simultaneously. In this study, we combine a GPU-based approach with the pseudo-transient (PT) method, an accelerated iterative and matrix-free solution strategy, and investigate its performance for finite elements and unstructured meshes with application to two-dimensional (2-D) models of real glaciers at a regional scale. For both the Jakobshavn and Pine Island glacier models, the number of nonlinear PT iterations required to converge a given number of vertices (N) scales in the order of 𝒪(N1.2) or better. We further compare the performance of the PT CUDA C implementation with a standard finite-element CPU-based implementation using the price-to-performance metric. The price of a single Tesla V100 GPU is 1.5 times that of two Intel Xeon Gold 6140 CPUs. We expect a minimum speedup of at least 1.5 times to justify the Tesla V100 GPU price to performance. Our developments result in a GPU-based implementation that achieves this goal with a speedup beyond 1.5 times. This study represents a first step toward leveraging GPU processing power, enabling more accurate polar ice discharge predictions. The insights gained will benefit efforts to diminish spatial resolution constraints at higher computing performance. The higher computing performance will allow for ensembles of ice-sheet flow simulations to be run at the continental scale and higher resolution, a previously challenging task. The advances will further enable the quantification of model sensitivity to changes in upcoming climate forcings. These findings will significantly benefit process-oriented sea-level-projection studies over the coming decades.