Search

We investigate the physical basis of the crevasse-depth (CD) calving law by analysing relationships between glaciological stresses and calving behaviour at Sermeq Kujalleq (Store Glacier), Greenland. Our observations and model simulations show that the glacier has a stable position defined by a compressive arch between lateral pinning points. Ice advance beyond the arch results in calving back to the stable position; conversely, if melt-undercutting forces the ice front behind the stable position, it readvances because ice velocities exceed subaqueous melt rates. This behaviour is typical of self-organising criticality, in which the stable ice-front position acts as an attractor between unstable super-critical and sub-critical regimes. This perspective provides strong support for a 'position-law' approach to modelling calving at Sermeq Kujalleq, because any calving 'rate' is simply a by-product of how quickly ice is delivered to the critical point. The CD calving law predicts ice-front position from the penetration of surface and basal crevasse fields, and accurately simulates super-critical calving back to the compressive arch and melt-driven calving into the sub-critical zone. The CD calving law reflects the glaciological controls on calving at Sermeq Kujalleq and exhibits considerable skill in simulating its mean position and seasonal fluctuations. [ABSTRACT FROM AUTHOR]

Mass loss due to dynamic changes in ice sheets is a significant contributor to sea level rise, and this contribution is expected to increase in the future. Numerical codes simulating the evolution of ice sheets can potentially quantify this future contribution. However, the uncertainty inherent in these models propagates into projections of sea level rise is and hence crucial to understand. Key variables of ice sheet models, such as basal drag or ice stiffness, are typically initialized using inversion methodologies to ensure that models match present observations. Such inversions often involve tens or hundreds of thousands of parameters, with unknown uncertainties and dependencies. The computationally intensive nature of inversions along with their high number of parameters mean traditional methods such as Monte Carlo are expensive for uncertainty quantification. Here we develop a framework to estimate the posterior uncertainty of inversions and project them onto sea level change projections over the decadal timescale. The framework treats parametric uncertainty as multivariate Gaussian and exploits the equivalence between the Hessian of the model and the inverse covariance of the parameter set. The former is computed efficiently via algorithmic differentiation, and the posterior covariance is propagated in time using a time-dependent model adjoint to produce projection error bars. This work represents an important step in quantifying the internal uncertainty of projections of ice sheet models. [ABSTRACT FROM AUTHOR]

Mass loss due to dynamic changes in ice sheets is a significant contributor to sea level rise, and this contribution is expected to increase in the future. Numerical codes simulating the evolution of ice sheets can potentially quantify this future contribution. However, the uncertainty inherent in these models propagates into projections of sea level rise, and hence is crucial to understand. Key variables of ice sheet models, such as basal drag or ice stiffness, are typically initialized using inversion methodologies to ensure that models match present observations. Such inversions often involve tens or hundreds of thousands of parameters, with unknown uncertainties and dependencies. The computationally intensive nature of inversions along with their high number of parameters mean traditional methods such as Monte Carlo are expensive for uncertainty quantification. Here we develop a framework to estimate the posterior uncertainty of inversions, and project them onto sea level change projections over the decadal timescale. The framework treats parametric uncertainty as multivariate Gaussian, and exploits the equivalence between the Hessian of the model and the inverse covariance of the parameter set. The former is computed efficiently via algorithmic differentiation, and the posterior covariance is propagated in time using a time-dependent model adjoint to produce projection error bars. This work represents an important step in quantifying the internal uncertainty of projections of ice-sheet models. [ABSTRACT FROM AUTHOR]

The response of tidewater glaciers to ocean warming remains a key uncertainty in sea level rise predictions. Here we use a 3‐D numerical model to examine the response of an idealized tidewater glacier to spatial variations in submarine melt rate. While melting toward the center of the terminus causes only a localized increase in mass loss, melting near the lateral margins triggers increased calving across the width of the glacier, causing the terminus to retreat at several times the width‐averaged melt rate. This occurs because melting near the margins has a greater disruptive impact on the compressive stress arch that transfers resistance from the side walls to the body of the glacier. We suggest that the rate of terminus advance or retreat may thus be governed by the difference between ice velocity and submarine melting in the slow‐flowing zones away from the glacier center. Plain Language Summary: The rapid retreat of tidewater glaciers (i.e., glaciers that drain directly into the ocean) has become an increasing source of concern in recent years. Increased melting of the submerged parts of these glaciers by warming ocean waters is thought to be an important driver of this retreat, but exactly how glaciers respond to this submarine melting remains unclear. Using a numerical model, we find that tidewater glaciers may be most sensitive to melting close to the valleysides, which reduces structural support for the central section of the glacier and so triggers an increase in iceberg calving. By better constraining the relationships between submarine melting, calving, and glacier retreat, our findings allow improved prediction of the ice loss expected from tidewater glaciers as the climate continues to warm. Key Points: We use a 3‐D numerical model to examine the response of an idealized tidewater glacier to spatial variations in submarine melt rateThe glacier is most sensitive to melting near the lateral margins, which triggers increased calving across the width of the terminusTerminus retreat may thus be paced by the difference between ice velocity and submarine melting across slow flowing marginal zones [ABSTRACT FROM AUTHOR]

Iceberg calving accounts for between 30 % and 60 % of net mass loss from the Greenland Ice Sheet, which has intensified and is now the single largest contributor to global sea level rise in the cryosphere. Changes to calving rates and the dynamics of calving glaciers represent a significant uncertainty in projections of future sea level rise. A growing body of observational evidence suggests that calving glaciers respond rapidly to regional environmental change, but predictive capacity is limited by the lack of suitable models capable of simulating calving mechanisms realistically. Here, we use a 3-D full-Stokes calving model to investigate the environmental sensitivity of Store Glacier, a large outlet glacier in West Greenland. We focus on two environmental processes: undercutting by submarine melting and buttressing by ice mélange, and our results indicate that Store Glacier is likely to be able to withstand moderate warming perturbations in which the former is increased by 50 % and the latter reduced by 50 %. However, severe perturbation with a doubling of submarine melt rates or a complete loss of ice mélange destabilises the calving front in our model runs. Furthermore, our analysis reveals that stress and fracture patterns at Store's terminus are complex and varied, primarily due to the influence of basal topography. Calving style and environmental sensitivity vary greatly, with propagation of surface crevasses significantly influencing iceberg production in the northern side, whereas basal crevasses dominate in the south. Any future retreat is likely to be initiated in the southern side by a combination of increased submarine melt rates in summer and reduced mélange strength in winter. The lateral variability, as well as the importance of rotational and bending forces at the terminus, underlines the importance of using the 3-D full-Stokes stress solution when modelling Greenland's calving glaciers. [ABSTRACT FROM AUTHOR]

Social connectedness among neighbours impacts health and well-being, especially during stressful life events like a pandemic. An activity such as neighbourhood walking enables urban inhabitants to engage in incidental sociability and acts of "neighbouring"--that is, authentic social interactions with neighbours--to potentially bolster the social fabric of neighbourhoods and strengthen relationships. With the potential of neighbourhood walking in mind, this article investigates how everyday encounters while engaged in routine neighbourhood walks strengthen and/or weaken social ties among neighbours. To this end, the article draws on three sources of qualitative data from neighbourhood walkers in Southwestern Ontario, Canada: (a) "walking diaries" in which participants took note of their walking routes, the people they observed on their walks, and other details of their walking experiences; (b) maps of their neighbourhoods that outlined the boundaries of their self-identified neighbourhoods, their routine walking routes, and the people they recognized during their neighbourhood walks; and (c) one-on-one interviews during which participants provided crucial context and meaning to the maps and their walking experiences. The findings provide evidence of how interactions among inhabitants, while engaged in neighbourhood walking, help generate greater social connectedness. [ABSTRACT FROM AUTHOR]

A new calving algorithm is developed in the glacier model Elmer/Ice that allows unrestricted calving and terminus advance in 3D. The algorithm uses the meshing software Mmg to implement anisotropic remeshing and allow mesh adaptation at each time step. The development of the algorithm, along with the implementation of the crevasse depth law, produces a new full-Stokes calving model capable of simulating calving and terminus advance across an array of complex geometries. Using a synthetic tidewater glacier geometry, the model is tested to highlight the numerical model parameters that can alter calving when using the crevasse depth law. For a system with no clear attractor at a pinning point, the model time step and mesh resolution are shown to alter the simulated calving. In particular, the vertical mesh resolution has a large impact, increasing calving, as the frontal bending stresses are better resolved. However, when the system has a strong attractor, provided by basal pinning points, numerical model parameters have a limited effect on the terminus evolution. Conversely, transient systems with no clear attractors are highly influenced by the choice of numerical model parameters. The new algorithm is capable of implementing unlimited terminus advance and retreat, as well as unrestricted calving geometries, applying any vertically varying melt distribution to the front for use in conjunction with any calving law or potentially advecting variables downstream. In overcoming previous technical hurdles, the algorithm opens up the opportunity to improve both our understanding of the physical processes and our ability to predict calving. [ABSTRACT FROM AUTHOR]

The ongoing retreat of many tidewater glaciers has been attributed to the increased submarine melting of calving fronts in response to warming oceans and increased meltwater runoff. The mechanisms through which submarine melting affects the stability of tidewater glaciers remain poorly understood however. In particular, the existence of a 'calving amplifier effect', in which submarine melting leads to a further increase in ice loss through calving, remains uncertain. Numerical modelling experiments have shown that this effect may occur when a glacier calving front is undercut due to vertical variation in the rate of submarine melting. There has however been relatively little investigation of the impact on calving, and thus total mass loss, of lateral variation in melting across the calving front.Here, we use a three-dimensional, full-Stokes glacier calving model implemented in Elmer/Ice to examine the response of an idealised tidewater glacier to spatially variable submarine melting. The model uses a crevasse-depth calving criterion that triggers calving when surface and basal crevasses penetrate the full thickness of the glacier. We subject the model glacier to spatially discrete areas of submarine melting representing the zones of intense localised melting that occur at tidewater glaciers where the input of subglacial runoff drives vigorous plumes adjacent to the calving front. We find that submarine melting may have both a positive and a negative impact on the calving rate, depending on a range of variables including the intensity of the melting, the extent of the calving front subject to melting, and location of these zones of intense melting along the calving front. In particular, we find that the greatest positive impact on calving occurs when submarine melting is applied close to the ice margins rather than in the centre of the glacier. This occurs because as the melting incises notches into the calving front, it disrupts the natural compressive stress-arch that provides support to the centre of the glacier, isolating it from the stabilising influence of the fjord walls and thus triggering calving until a new stable geometry is formed. It is thus possible that through this calving amplifier effect, small areas of intense plume-driven melting could trigger enhanced calving and retreat even at large and fast-flowing glaciers at which the average submarine melt rate remains below the glacier velocity. [ABSTRACT FROM AUTHOR]

The dynamic response of calving glaciers to warming climate is one of the greatest uncertainties in predictions of future sea level rise. A growing body of observational evidence suggests that the calving glaciers which drain the Greenland and Antarctic ice sheets are highly sensitive to environmental forcing, but the current generation of glacier models is unable to adequately resolve this critical interaction between ice sheets and climate. We present results from a new approach combining the continuum model Elmer/Ice and the discrete element/particle model HiDEM, applied to Store Glacier, a large calving glacier in West Greenland. We use Elmer/Ice to investigate the dynamic/stress response to both submarine melting and ice mélange buttressing of the calving front, and use HiDEM to investigate fracture propagation and calving for specific geometries. We implement a new approach to investigating ice mélange; by allowing the glacier to calve repeatedly into the fjord, we effectively 'build' a mélange from previously calved icebergs.Our results indicate that lateral support from valley walls plays a critical role in the stability of the calving front; a persistent compressive stress arch exists just behind the current terminus position, transmitting resistive stress from sidewalls to centreline. Furthermore, we find that minimal mélange buttressing applied near the lateral margins is sufficient to advance this compressive arch, allowing the terminus to advance in winter. With respect to submarine melting, we find that a modest melt undercut (60m) can promote calving at an otherwise stable terminus position. Our HiDEM simulations for Store Glacier closely match the calving style observed in satellite imagery; the ability to accurately resolve calving at the event scale is a major step forward in our understanding of calving processes. By comparing stress patterns in Elmer/Ice with fracture patterns in HiDEM, we seek to formulate improved calving laws which can be implemented in ice-sheet scale continuum models. [ABSTRACT FROM AUTHOR]

We investigate the subglacial hydrology of Store Glacier in West Greenland, using the open-source, full-Stokes model Elmer/Ice in a novel 3D application that includes a distributed water sheet, as well as discrete channelised drainage, and a 1D model to simulate submarine plumes at the calving front. At first, we produce a baseline winter scenario with no surface meltwater. We then investigate the hydrological system during summer, focussing specifically on 2012 and 2017, which provide examples of high and low surface-meltwater inputs, respectively. We show that the common assumption of zero winter freshwater flux is invalid, and we find channels over 1 m 2 in area occurring up to 5 km inland in winter. We also find that the production of water from friction and geothermal heat is sufficiently high to drive year-round plume activity, with ice-front melting averaging 0.15 m d -1. When the model is forced with seasonally averaged surface melt from summer, we show a hydrological system with significant distributed sheet activity extending 65 and 45 km inland in 2012 and 2017, respectively; while channels with a cross-sectional area higher than 1 m 2 form as far as 55 and 30 km inland. Using daily values for the surface melt as forcing, we find only a weak relationship between the input of surface meltwater and the intensity of plume melting at the calving front, whereas there is a strong correlation between surface-meltwater peaks and basal water pressures. The former shows that storage of water on multiple timescales within the subglacial drainage system plays an important role in modulating subglacial discharge. The latter shows that high melt inputs can drive high basal water pressures even when the channelised network grows larger. This has implications for the future velocity and mass loss of Store Glacier, and the consequent sea-level rise, in a warming world. [ABSTRACT FROM AUTHOR]

Abstract. A new calving algorithm was developed in the glacier model Elmer/Ice that allows unrestricted calving and terminus advance in 3D. The algorithm used the meshing software Mmg to implement anisotropic remeshing and allow mesh adaptation at each timestep. The development of the algorithm along with the implementation of the crevasse depth law produced a new full-Stokes calving model capable of simulating calving and terminus advance across an array of complex geometries. Using a synthetic tidewater glacier geometry the model was tested to highlight the non-physical parameters that can alter calving. For a system with no clear attractor, model timestep and mesh resolution are shown to alter the simulated calving. In particular vertical mesh resolution had a large impact, increasing calving, as the frontal bending stresses are better resolved. However, when the system had a strong attractor, provided by basal pinning points, non-physical parameters have a limited affect on the terminus evolution. The new algorithm is capable of implementing unlimited terminus advance and retreat as well as unrestricted calving geometries, applying any melt field to the front, use in conjunction with any calving law or potentially advecting variables downstream. [ABSTRACT FROM AUTHOR]

Ice sheet models are the main tool to generate forecasts of ice sheet mass loss, a significant contributor to sea level rise; thus, knowing the likelihood of such projections is of critical societal importance. However, to capture the complete range of possible projections of mass loss, ice sheet models need efficient methods to quantify the forecast uncertainty. Uncertainties originate from the model structure, from the climate and ocean forcing used to run the model, and from model calibration. Here we quantify the latter, applying an error propagation framework to a realistic setting in West Antarctica. As in many other ice sheet modelling studies we use a control method to calibrate grid-scale flow parameters (parameters describing the basal drag and ice stiffness) with remotely sensed observations. Yet our framework augments the control method with a Hessian-based Bayesian approach that estimates the posterior covariance of the inverted parameters. This enables us to quantify the impact of the calibration uncertainty on forecasts of sea level rise contribution or volume above flotation (VAF) due to the choice of different regularization strengths (prior strengths), sliding laws, and velocity inputs. We find that by choosing different satellite ice velocity products our model leads to different estimates of VAF after 40 years. We use this difference in model output to quantify the variance that projections of VAF are expected to have after 40 years and identify prior strengths that can reproduce that variability. We demonstrate that if we use prior strengths suggested by L -curve analysis, as is typically done in ice sheet calibration studies, our uncertainty quantification is not able to reproduce that same variability. The regularization suggested by the L curves is too strong, and thus propagating the observational error through to VAF uncertainties under this choice of prior leads to errors that are smaller than those suggested by our two-member "sample" of observed velocity fields. [ABSTRACT FROM AUTHOR]

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. [ABSTRACT FROM AUTHOR]

The structure of subglacial hydrological drainage systems beneath large Greenlandic tidewater glaciers influences ice velocity due to its effect on basal traction, and it may also influence calving by its control on melting at the ice-ocean interface. Understanding these systems is critical to being able to accurately predict the evolution of the Greenland Ice Sheet and the resulting sea-level rise, as the fifteen largest Greenland outlet glaciers are responsible for 77% of the additional mass loss from the ice sheet due to acceleration since 2000. In this study, we use numerical modelling and observations of Store Glacier in West Greenland to constrain the form of the subglacial drainage system and melt rates from the resulting plumes, both of which are poorly known.We investigate the evolution of Store Glacier's subglacial hydrology using the open-source, full-Stokes model Elmer/Ice in a novel 3D application, the GlaDS hydrological module, which includes a distributed sheet and the ability to form concentrated channels when the sheet locally reaches sufficient thickness to initiate melting in cavities. At first, we produce a baseline winter scenario in which the hydrological system contains only basally derived meltwater produced primarily from friction at the bed. We then investigate the hydrological system during summer, focussing specifically on 2012 and 2017, which provide examples of high and low inputs of surface meltwater, respectively. When the model is forced with constant average runoff from summer (Jun-Aug) 2012, outputs show a hydrological system with significant sheet activity extending 55km inland and channels with a cross-sectional area higher than 1 m2 forming up to 45 km from the margin. However, we also find the hydrological system to be active in winter, when significant sheet flux is evident up to 35 km inland, and channels form as far as 25 km inland. The discharge of meltwater into the fjord produces convective plumes throughout the year, and these drive average submarine ice-front melt rates of up to 0.15 m d-1 in summer and 0.12 m d-1 in winter, rising to 1.6 m d-1 and 1.07 m d-1, respectively, for average temporal maximum melt rates. When the model is forced with daily time-series of summer runoff, we find a higher intensity of plume melting, reaching 0.16 m d-1 on average (rising to 2.21 m d-1 for the average temporal maximum). Our study shows that plume-induced ice-front melting is substantial even in winter, and we hypothesise that the high spatial variability and intensity during summer may drive calving through plume-induced notches cutting into the ice front as well as by undercutting of the terminus as a whole. [ABSTRACT FROM AUTHOR]

Three-dimensional ice flow modelling requires a large number of computing resources and observation data, such that 2-D simulations are often preferable. However, when there is significant lateral divergence, this must be accounted for (2.5-D models), and a flow tube is considered (volume between two horizontal flowlines). In the absence of velocity observations, this flow tube can be derived assuming that the flowlines follow the steepest slope of the surface, under a few flow assumptions. This method typically consists of scanning a digital elevation model (DEM) with a moving window and computing the curvature at the centre of this window. The ability of the 2.5-D models to account properly for a 3-D state of strain and stress has not clearly been established, nor their sensitivity to the size of the scanning window and to the geometry of the ice surface, for example in the cases of sharp ridges. Here, we study the applicability of a 2.5-D ice flow model around a dome, typical of the East Antarctic plateau conditions. A twin experiment is carried out, comparing 3-D and 2.5-D computed velocities, on three dome geometries, for several scanning windows and thermal conditions. The chosen scanning window used to evaluate the ice surface curvature should be comparable to the typical radius of this curvature. For isothermal ice, the error made by the 2.5-D model is in the range 0-10% for weakly diverging flows, but is 2 or 3 times higher for highly diverging flows and could lead to a non-physical ice surface at the dome. For non-isothermal ice, assuming a linear temperature profile, the presence of a sharp ridge makes the 2.5-D velocity field unrealistic. In such cases, the basal ice is warmer and more easily laterally strained than the upper one, the walls of the flow tube are not vertical, and the assumptions of the 2.5-D model are no longer valid. [ABSTRACT FROM AUTHOR]

Three-dimensional ice flow modelling requires a lot of computing resources and observation data, such that 2D simulations are often preferable, at least when the stream lines are parallel; otherwise the lateral divergence of the flow should be accounted for (2.5D models). Assuming that the stream lines follow the steepest slope of the surface, the width variations of a flow tube are computed thanks to the surface curvature. The ability of the 2.5D models to account properly for a 3D state of strain and stress has not clearly been established, especially their sensitivity on how the ice surface curvature is determined (scanning window on a DEM), and on the geometry of the ice surface. In particular, these models might fail for divergent flows, and need to be more clearly defined. A twin experiment is here carried out, comparing 3D and 2.5D computed velocities, on three dome geometries, for several scanning windows and thermal conditions. The chosen scanning window used to evaluate the ice surface curvature should be comparable to the typical size of the measured ice relief. For isothermal ice, the error made by the 2.5D model is in the range 0-10% but for highly diverging flows the errors are 2 or 3 times higher and could lead to a non-physical reversed surface convexity at the dome. For non-isothermal ice, assuming a realistic temperature profile, the presence of a sharp ridge leads to a partly reversed velocity profile. The warmer bottom ice is more deformed than the upper ice, and this results in the non-verticality of the walls of the flow tube, violating the 2.5D assumptions. [ABSTRACT FROM AUTHOR]

Calving mechanisms are still poorly understood. The size and the frequency of calving events may vary by several orders of magnitude, making the development of an universal calving model challenging. At Bowdoin Glacier, Northwest Greenland, most of the yearly mass loss by calving is due to a few large events. Here, we analyse two major calving events in detail, which occurred nearly at the same location and followed a strikingly similar fracturing pattern. Our analysis relies on data obtained by interferometric radar and UAV photogrammetry during two summer fieldwork campaigns in July 2015 and July 2017. The crevasses likely deepened by hydro-fracturing, since a supraglacial river supplied them with water. Besides that, our high temporal and spatial resolution data reveals the influence of tides on the opening of the crack.Here, we used the ice flow model Elmer/Ice to analyse the observations further. Modeling crevasse opening required the development of a dedicated remeshing routine for the crevasse. We use Elmer/Ice as a diagnostic, inverse model to identify key drivers of the crevasse opening and to determine possible combinations of water level inside the crevasse and crevasse depth that prevailed prior to the calving event. As a result, we find that the water level acts as a first-order control on crevasse opening rates. Finally, we apply the discrete element model HiDEM to study crevasse initiation and crevasse propagation with time and we test HiDEM's capability of reproducing the calving style observed at Bowdoin Glacier. [ABSTRACT FROM AUTHOR]

Marine Ice Cliff Instability (MICI) has been proposed as a mechanism through which marine-based ice sheets could undergo rapid collapse and mass loss. MICI is based on the idea that the finite strength of ice places a limit on ice cliff height above sea level (freeboard); therefore, if buttressing ice shelves disintegrate and expose deep calving cliffs, widespread catastrophic ice-sheet disintegration could occur. Ice sheet models incorporating MICI processes predict rapid disintegration of the Amundsen Sea sector of the West Antarctic Ice Sheet under high CO2 scenarios (Pollard et al., 2015; DeConto and Pollard, 2016). These models, however, employ highly simplified parameterizations of key calving processes, and possible rates of future ice loss and sea level rise remain subject to large uncertainties. MICI invokes ice-front behaviors beyond the current observed range, complicating attempts to validate model results. DOMINOS (Disintegration of Marine Ice Sheets: Novel Optimized Simulations; part of the International Thwaites Glacier Collaboration) aims to develop detailed process-based models of MICI and its impact on the sensitivity of the West Antarctic Ice Sheet. Here we present preliminary results from the Helsinki Discrete Element Model (HiDEM), which aim to understand the mechanics of ice-cliff instability and determine the spatial and temporal scales of failure events. Collapse of deep ice cliffs (>90 m freeboard) occurs in response to temporally varying longitudinal and buoyant forces, and involves complex mixed-mode rheology including brittle failure, viscous deformation, and enhanced viscous flow along shear zones. Calved bergs form floating tongues of mélange, which may provide stabilizing backstress if jamming occurs against lateral or basal pinning points. In the absence of backstress, mechanical instability propagates upglacier, causing rapid, unstable retreat where basins deepen inland. [ABSTRACT FROM AUTHOR]