Ice-stream Model Research Articles

Highly variable friction and slip observed at Antarctic ice stream bed

The slip of glaciers over the underlying bed is the dominant mechanism governing the migration of ice from land into the oceans, with accelerating slip contributing to sea-level rise. Yet glacier slip remains poorly understood, and observational constraints are sparse. Here we use passive seismic observations to measure both frictional shear stress and slip at the bed of the Rutford Ice Stream in Antarctica using 100,000 repetitive stick-slip icequakes. We find that basal shear stresses and slip rates vary from 104 to 107 Pa and 0.2 to 1.5 m per day, respectively. Friction and slip vary temporally over the order of hours, and spatially over 10s of metres, due to corresponding variations in effective normal stress and ice–bed interface material. Our findings suggest that the bed is substantially more complex than currently assumed in ice stream models and that basal effective normal stresses may be significantly higher than previously thought. Our observations can provide constraints on the basal boundary conditions for ice-dynamics models. This is critical for constraining the primary contribution of ice mass loss in Antarctica and hence for reducing uncertainty in sea-level rise projections.

Massive Groundwater Systems Lie Beneath Antarctic Ice

Scientists are updating ice stream models to understand the ways in which deep groundwater systems impact ice flow.

A permeameter for temperate ice: first results on permeability sensitivity to grain size

AbstractResults of ice-stream models that treat temperate ice deformation as a two-phase flow are sensitive to the ice permeability. We have constructed and begun using a custom, falling-head permeameter for measuring the permeability of temperate, polycrystalline ice. Chilled water is passed through an ice disk that is kept at the pressure-melting temperature while the rate of head decrease indicates the permeability. Fluorescein dye in the water allows water-vein geometry to be studied using fluorescence microscopy. Water flow over durations of seconds to hours is Darcian, and for grain diameter d increasing from 1.7 to 8.9 mm, average permeability decreases from 2 × 10−12 to 4 × 10−15 m2. In tests with dye on fine (d = 2 mm) and coarse (d = 7 mm) ice, average area-weighted vein radii are nearly equal, 41 and 34 μm, respectively. These average radii, if included in a theory slightly modified from Nye and Frank (1973), yield permeability values within a factor of 2.0 of best-fit values based on regression of the data. Permeability values depend on d−3.4, rather than d−2 as predicted by models if vein radii are considered independent of d. In future experiments, the dependence of permeability on liquid water content will be measured.

Synchronization of Heinrich and Dansgaard‐Oeschger Events Through Ice‐Ocean Interactions

AbstractThe cause of Heinrich events and their relationship with Dansgaard‐Oeschger (DO) events are not fully understood. Previous modeling studies have argued that Heinrich events result from either internal oscillations generated within ice sheets or ocean warming occurring during DO events. In this study, we present a coupled model of ice stream and ocean dynamics to evaluate the behavior of the coupled system with few degrees of freedom and minimal parameterizations. Both components of the model may oscillate independently, with stagnant versus active phases for the ice stream model and strong versus weak Atlantic Meridional Overturning Circulation (AMOC) phases for the ocean model. The ice sheet and ocean interact through submarine melt at the ice stream grounding line and freshwater flux into the ocean from ice sheet discharge. We show that these two oscillators have a strong tendency to synchronize, even when their interaction is weak, due to the amplification of small perturbations typical in nonlinear oscillators. In synchronized regimes with ocean‐induced melt at the ice stream grounding line, Heinrich events always follow DO events by a constant time lag. We also introduce noise into the ocean system and find that ice‐ocean interactions not only maintain a narrow distribution of timing between Heinrich and DO events, but also regulate DO event periodicity against noise in the climate system. This synchronization persists across a broad range of parameters, indicating that it is a robust explanation for Heinrich events and their timing despite the significant uncertainty associated with past ice sheet conditions.

Icepack: a new glacier flow modeling package in Python, version 1.0

Abstract. We introduce a new software package called “icepack” for modeling the flow of glaciers and ice sheets. The icepack package is built on the finite element modeling library Firedrake, which uses the Unified Form Language (UFL), a domain-specific language embedded into Python for describing weak forms of partial differential equations. The diagnostic models in icepack are formulated through action principles that are specified in UFL. The components of each action functional can be substituted for different forms of the user's choosing, which makes it easy to experiment with the model physics. The action functional itself can be used to define a solver convergence criterion that is independent of the mesh and requires little tuning on the part of the user. The icepack package includes the 2D shallow ice and shallow stream models. We have also defined a 3D hybrid model based on spectral semi-discretization of the Blatter–Pattyn equations. Finally, icepack includes a Gauss–Newton solver for inverse problems that runs substantially faster than the Broyden–Fletcher–Goldfarb–Shanno (BFGS) method often used in the glaciological literature. The overall design philosophy of icepack is to be as usable as possible for a wide a swath of the glaciological community, including both experts and novices in computational science.

Leveraging the Rapid Retreat of the Amundsen Gulf Ice Stream 13,000 Years Ago to Reveal Insight Into North American Deglaciation

AbstractWe examine the influence of different deglacial histories of the North American ice saddle, which connected the Cordilleran ice sheet and western Laurentide ice sheet, on the stability of the fast‐flowing Amundsen Gulf Ice Stream, located in the northwestern Laurentide ice sheet. We use a simplified marine‐terminating ice stream model to simulate grounding line retreat and compare our predictions to geologic evidence for the rapid collapse of the Amundsen Gulf Ice Stream at 13 ka. We show that observations of ice stream retreat can be used to distinguish between past ice unloading histories, and that the rapid retreat of the Amundsen Gulf Ice Stream near 13 ka is most dynamically consistent with substantial deglaciation of the North American ice saddle from 13 to 11.5 ka, during the Younger Dryas cooling episode. These results suggest that short‐lived changes in ice stream behavior may be used to reveal larger‐scale and longer‐duration ice sheet dynamics.

Ice-stream dynamics of the SW Barents Sea revealed by high-resolution 3D seismic imaging of glacial deposits in the Hoop area

Recent developments in seismic acquisition systems and seismic data visualization have contributed to improve the imaging of seismic geomorphologies over a broad range of topics. This study focuses on a new high-resolution P-Cable 3D seismic cube located in the Hoop area in the SW Barents Sea. The scientific motivations of the study are (1) to document glacial landforms on a meter-scale, (2) to study the link of these landforms with the subsurface, and (3) to understand ice-stream dynamics by detailed interpretation of main horizons, including the seabed, intra-glacial horizons, and Upper Regional Unconformity (URU). The horizons are assessed using the concept of seismic geomorphology, and compared with interpretations derived from conventional 3D seismic and state-of-the-art multibeam echosounder technologies to discuss the benefits and limitations of the different geophysical technologies. A well-defined intra-glacial seismic reflection is imaged in the eastern part of the P-Cable cube. This reflection has an overall NE-trending ridge-shaped morphology, which is up to 50 m high and 5–10 km wide, and interpreted as a shear margin moraine. The top of the moraine displays well-defined iceberg ploughmarks and clear slope failure events along its eastern flank. The seabed geomorphology, which has been directly shaped by glacial processes related to multiple phases of ice streaming, appears to be influenced by the buried shear margin moraine. Mega-scale glacial lineations (MSGLs) indicative of fast-flowing ice streams characterize the seabed west of the moraine. Infilled MSGLs and different sets of iceberg ploughmarks with large berms dominate the seabed geomorphology above the moraine. The seabed east of the moraine, characterized by a smooth morphology and few pockmarks, is inferred to reflect more slowly-moving ice. Not resolved by other geophysical techniques, seismic attribute maps of P-Cable 3D allowed the identification of smaller ploughmarks and corrugated ridges related to paleo tides. Eroded by the larger iceberg ploughmarks, these small ploughmarks are suggested to be linked to an earlier rapid disintegration of the Barents Sea Ice Stream. The new high-resolution 3D seismic data have led to the development of a revised ice-stream model of the Barents Sea, linking the glacial deposits to different ice stream events.

Rapid grounding line migration induced by internal ice stream variability

AbstractObservations indicate that the grounding line position of West Antarctica is sensitive to both forced and unforced ice stream variabilities. This study endeavors to characterize and understand unforced ice stream variability and associated grounding line migration. We use a flowline ice stream model with an undrained plastic bed, lateral shear stresses, and a stretched grid refined in the grounding zone. This model exhibits parameter space structure and hysteresis behavior similar to simpler ice stream models. Low prescribed temperature at the ice surface or weak geothermal heating produces thermal oscillations between active and stagnant phases. As in previous spatially resolved ice flow models, thermal activation propagates as an “activation wave.” This model's fine resolution of the grounding zone allows for accurate simulations of transient, unforced grounding line migration. Upstream of the grounding zone, horizontal grid spacing of 1 km is required to accurately resolve activation waves. Activation waves induce the grounding line to migrate over 100 km at a rate that can exceed 1 km/yr. This is followed during the active phase by retreat, which then continues for the duration of the stagnant phase. Grounding line retreat is the result of a negative mass balance near the grounding line but is not necessarily associated with negative mass balance for the entire ice stream in our simulations of internal variability. The novel approach and experiments described in this study show that there can be large excursions in grounding line position in the absence of either external forcing or retrograde slopes.

Unstable ice stream in Greenland during the Younger Dryas cold event

Past, present, and future ice sheet stability is closely linked to the dynamic behavior of major draining ice streams and surrounding ice shelves. While short observational records document the recent variability and acceleration of ice streams, the long-term dynamics of ice streams remain poorly documented. Here, we date the Pjetursson’s Moraine on Disko Island, Greenland, to 12.2 ± 0.6 ka and demonstrate that the Jakobshavn Isbrae (JI) ice stream collapsed during the middle of the Younger Dryas (YD) cold interval. We suggest that this collapse was due to the incursion of warm subsurface water under the ice shelf fronting the JI ice stream, as well as increased surface-air temperature and sea-surface temperature seasonality starting at the beginning of the YD cold interval. The triggered acceleration of the land-based JI and the delivery of icebergs into Disko Bugt potentially contributed to Heinrich Event 0 at the end of the YD. The collapse of the JI ice stream 12.2 ± 0.6 ka ago demonstrates that calving marine-based ice margins can respond rapidly to environmental changes. It provides a new benchmark for marine-terminating ice stream models.

Decadal‐scale sensitivity of Northeast Greenland ice flow to errors in surface mass balance using ISSM

The behavior of the Greenland Ice Sheet, which is considered a major contributor to sea level changes, is best understood on century and longer time scales. However, on decadal time scales, its response is less predictable due to the difficulty of modeling surface climate, as well as incomplete understanding of the dynamic processes responsible for ice flow. Therefore, it is imperative to understand how modeling advancements, such as increased spatial resolution or more comprehensive ice flow equations, might improve projections of ice sheet response to climatic trends. Here we examine how a finely resolved climate forcing influences a high‐resolution ice stream model that considers longitudinal stresses. We simulate ice flow using a two‐dimensional Shelfy‐Stream Approximation implemented within the Ice Sheet System Model (ISSM) and use uncertainty quantification tools embedded within the model to calculate the sensitivity of ice flow within the Northeast Greenland Ice Stream to errors in surface mass balance (SMB) forcing. Our results suggest that the model tends to smooth ice velocities even when forced with extreme errors in SMB. Indeed, errors propagate linearly through the model, resulting in discharge uncertainty of 16% or 1.9 Gt/yr. We find that mass flux is most sensitive to local errors but is also affected by errors hundreds of kilometers away; thus, an accurate SMB map of the entire basin is critical for realistic simulation. Furthermore, sensitivity analyses indicate that SMB forcing needs to be provided at a resolution of at least 40 km.

Diurnal and semidiurnal tide‐induced lateral movement of Ronne Ice Shelf, Antarctica

Recent GPS observations from a spatially extensive network across Ronne Ice Shelf show significant daily ice flow variations. At all sites, the almost‐synchronous horizontal displacements occur at diurnal and semidiurnal tidal periods. During spring tides, displacements, velocities and strains near the ice front have superimposed oscillations that are ±300% of their mean values and occur over a six‐hour period, resulting in regular ice shelf flow reversals. Close to ice stream grounding lines, however, the horizontal diurnal and semidiurnal signals decay and almost vanish. From our analysis, we conclude that ice shelves respond primarily elastically to tidal tilting, thus accounting for the observed diurnal and semidiurnal flow variations, and their amplification toward the ice shelf front. Our findings suggest that detailed modeling of these data could provide improved ice shelf and ice stream models for correctly simulating ice shelf flow and predicting future ice sheet evolution.

Basal conditions for Pine Island and Thwaites Glaciers, West Antarctica, determined using satellite and airborne data

AbstractWe use models constrained by remotely sensed data from Pine Island and Thwaites Glaciers, West Antarctica, to infer basal properties that are difficult to observe directly. The results indicate strong basal melting in areas upstream of the grounding lines of both glaciers, where the ice flow is fast and the basal shear stress is large. Farther inland, we find that both glaciers have ‘mixed’ bed conditions, with extensive areas of both bedrock and weak till. In particular, there are weak areas along much of Pine Island Glacier’s main trunk that could prove unstable if it retreats past the band of strong bed just above its current grounding line. In agreement with earlier studies, our forward ice-stream model shows a strong sensitivity to small perturbations in the grounding line position. These results also reveal a large sensitivity to the assumed bed (sliding or deforming) model, with non-linear sliding laws producing substantially greater dynamic response than earlier simulations that assume a linear-viscous till rheology. Finally, comparison indicates that our results using a plastic bed are compatible with the limited observational constraints and theoretical work that suggests an upper bound exists on maximum basal shear stress.

Reply to discussion by P.J. Barnett on "Generation of heavy-mineral glacial indicator dispersal trains from a diabase sill, Nipigon Region, Northwestern Ontario"

Larson and Mooers 1753 Barnett (2006) poses a number of questions regarding some of the field observations and assumptions we used in our interpretation of a heavy-mineral dispersal train derived from an isolated eastern outlier of diabase of the Nipigon sills in northwestern Ontario (Larson and Mooers 2005b). He also brings attention to some details of the regional glacial geologic setting we did not address in our analysis and interpretation of the dispersal train, and highlights new field data (Barnett and Dyer 2005) relevant to interpretation of past glacial processes acting in the broader Nipigon region. We thank Barnett for his comments, and feel they serve in part to highlight the variety of problems regarding glacial erosion and entrainment and transport to which we believe our conceptual model of glacial indicator dispersal (Larson and Mooers 2004) is able to contribute. Barnett (2006) brings attention to the fact that tills exposed at surface in the Beardmore–Geraldton region contain a wide range (< 1 up to 23%) in carbonate content (Thorleifson and Kristjansson 1993). High carbonate contents in these tills are a clear signal of long-distance transport of carbonate-bearing debris from the Hudson Bay Lowlands. The occurrence of high-carbonate tills in patches of thick, streamlined drift has been interpreted by Hicock (1988) and Hicock and others (1989) to mean that all surface tills in this region were deposited at the base of a fastflowing ice stream. However, the ice stream model does not explain the widely disparate values in till carbonate occurring over distances of only a few kilometres; we believe a different interpretation explains the same data (Larson and Mooers 2005a). Tills sampled in our limited study area contained virtually none of the Paleozoic carbonate and Proterozoic greywacke clasts characteristic of high-carbonate tills found in the region. We believe this observation is inconsistent with an origin of the till-forming material by longdistance transport of debris from the Hudson Bay Lowlands by an ice stream, and rather that it reflects local erosion and entrainment and short-distance transport. However, the origin of the tills we sampled in our study area is not relevant to the interpretations we presented. Determining whether the till in our study area was generated by erosion and entrainment into the englacial basal debris layer of the ice sheet, by erosion and entrainment in a subglacial deforming debris layer, or by another process was beyond the scope of our study. We deliberately made no explicit assertion regarding the processes by which diabase was eroded and entrained, and transported. We did, however, attempt to document the rates at which the erosion and entrainment process worked. The only explicit assumptions we incorporated into our interpretations were of the debris mass per unit bed area (m) and of the velocity at which the basal debris layer was transported (u). These values were used to estimate the erosivity (E) recorded in the till sheet, and indirectly the absolute erosion and entrainment rate (e) of diabase into the basal debris layer. Examination of the equations presented in Larson and Mooers (2004), and the modified equations presented in Larson and Mooers (2005b), shows that increasing the estimated m results in increased estimates of the values of E and e. Increasing u likewise results in increased estimates of e. However, estimates of the erosion length scale (λ) and the relative rates of erosion of the diabase and greenstone are unaffected by varying these quantities. We feel that the m and u estimates that we incorporated into our interpretation of the heavy-mineral dispersal data represent reasonable values for an ice sheet, and that the resulting e estimates are likewise reasonable. It is important to note that the absolute values of erosion and entrainment we have estimated for diabase and greenstone lithologies lie within the range documented for modern glacial systems (cf. Hallet et al. 1996). Barnett (2006) questions our conclusion that the diabase sill in our study area was significantly more resistant to glacial erosion and entrainment than the Archean greenstones into which it was intruded. Notwithstanding the higher susceptibility of gabbro–diabase to physical and chemical

Fast flow of the Lake Michigan Lobe: evidence from sediment-landform assemblages in southwestern Michigan, USA

The dynamics of the late Wisconsin Lake Michigan Lobe of the Laurentide Ice Sheet, as interpreted by sediment–landform assemblages along its southeastern margin, are consistent with fast flow driven by high subglacial pore pressures. A major advance of the Lake Michigan Lobe terminated at the Kalamazoo Moraine, where proglacial glaciotectonic activity, marginal stagnation, and release of subglacial meltwater marked the extent of the advance. A large portion of the Valparaiso “Moraine”, west of the Kalamazoo Moraine, is actually a drumlinized, till-capped upland plain rather than an ice–marginal landform assemblage. Lacustrine sediment, underlying the surficial diamicton in the Valparaiso upland, is glaciotectonically deformed to a depth of 10 m. The deformed interval consists of folded sandy and silty lacustrine sediment, which is in places sharply truncated by the overlying diamicton. Decoupling of the glacier from its bed is suggested by a discontinous thin bed of sand at the base of the diamicton, and by the sharp truncation of deformed lacustrine sediment. The association of drumlins, proglacial and subglacial deformation, marginal stagnation, and major ice-marginal outwash-fan deposition are indicative of fast flow, perhaps in a surge. These characteristics are consistent with a terrestrial ice-stream model for the Lake Michigan Lobe.

Signature of palaeo‐ice‐stream stagnation: till consolidation induced by basal freeze‐on

A combination of glaciological theory and geological observations was used to investigate the possibility of till consolidation being driven by basal freeze‐on beneath a stagnating, mid‐latitude palaeo‐ice stream. We focused on the case of the Baltic Ice Stream that advanced into Denmark at c. 15ka BP and which left behind a characteristic till sequence consisting of a strong and well‐consolidated till crust underlain by weak and poorly consolidated till. Our hypothesis is that basal freezing caused the markedly higher consolidation of the uppermost till layer. The freezing may have either triggered or simply just accompanied ice‐stream stoppage. To test the feasibility of this hypothesis, we have developed a numerical model that couples ice‐stream dynamics to time‐dependent changes of till properties. Despite relatively mild palaeo‐climatic conditions in this area during Late Pleistocene deglaciation (˜0°C), the ice‐stream model is capable of producing basal freezing when the effect of horizontal advection of cold ice is included. Our simulations of till response to basal freezing are based on thermodynamic concepts adapted from permafrost studies. Dewatering of till by basal freeze‐on may lead to overconsolidation (OCR&gt;10). Based on the history of effective pressure changes in the till, we can predict postglacial till strength profiles using the SHANSEP method. In a series of numerical experiments we have examined the response of till strength to basal freeze‐on induced beneath a decaying ice sheet. We have come reasonably close to reproducing shear strength profiles for till deposited by the Baltic Ice Stream. These observations are most consistent with palaeo‐ice‐stream stagnation triggered by basal freezing and followed by abrupt retreat (&lt;100 years) due to high surface ablation rates (&gt;10 ma‐1).

Basal mechanics of Ice Stream B, west Antarctica: 2. Undrained plastic bed model

Based on the results of our studies of the physical conditions beneath Ice Stream B, we formulate a new analytical ice stream model, the undrained plastic bed model (henceforth the UPB model). Mathematically, the UPB model is represented by a non‐linear system of four coupled equations which express the relationships among ice sliding velocity, till strength, water storage in till, and basal melt rate. We examine this system of equations for conditions of ice stream stability over short timescales that permit holding ice stream geometry constant (less than hundreds of years). Temporal variability is introduced into the UPB model only by the direct dependence of till void ratio changes (ė = ∂e/∂t) on the basal melting rate mr. Since till strength τb{e} and ice stream velocity Ub{τb} change as long as till void ratio varies, the first condition for ice stream stability is that of constant till water storage ė = 0. The second condition for ice stream stability arises from the feedback between ice stream velocity, till strength, and the basal melting rate which depends on shear heating mr{ Ub τb}. This is the “weak till” condition which requires that in a steady state till strength is a small fraction of the gravitational driving stress τb &lt; (n + 1)−1 τd. The salient feature of the UPB model is its ability to produce two thermo mechanically controlled equilibrium states, one with a strong bed and slow ice velocities (“ice sheet” mode) and one with a weak bed and fast ice velocities (“ice‐stream” mode). This bimodality of basal conditions is consistent with the available observations of subglacial conditions beneath slow and fast moving ice in West Antarctica. Basal conditions that do not correspond to these two steady states may occur transiently during switches between the two stable modes. The UPB model demonstrates that ice streams may be prone to thermally triggered instabilities, during which small perturbations in the basal thermal energy balance grow, leading to generation or elimination of the basal conditions which cause ice streaming.

Intercomparison of subglacial sediment-deformation models: application to the Late Weichselian western Barents margin

AbstractNumerical experiments, where a simple ice-sheet model was coupled with sediment-deformation models, were performed to investigate the transport of glacigenic material to the western Barents Shelf during the Late Weichselian. The ice-sheet model, and its environmental inputs, has been matched previously with a series of geological datasets relating to the maximum extent of the ice sheet (Howell and others, http://www.ggg.qub.ac.uk [rp05/1999]). Additional geological data on the volumes of sediment delivered to the Bear Island fan (Barents continental margin) are available for comparison. The experiments indicate the sensitivity of sediment transport and deposition to variations in (a) the ice-stream model and (b) a variety of model parameters. Two ice-stream models were used: (1) a height-above-buoyancy model, in which basal velocity is controlled by basal driving stress and a buoyancy-induced reduction in the normal load beneath a marine-based ice sheet; and (2) a modified version of the method presented by Alley (1990) in which basal velocity is related to pore-water pressure, sediment thickness, and driving basal stress. The results of the two different models were then compared. An extensive set of sensitivity tests was carried out to determine sediment-transport response to changes in the model’s parameters. Results indicate that, using physically realistic parameters for deforming subglacial sediment, both models reproduce the volume of Late Weichselian sediment measured on the Bear Island fan. Results from both models are sensitive to (1) cohesion of the sediment and (2) the thickness of deforming sediment beneath the ice sheet. The two models exhibited different degrees of sensitivity to the sediment parameters, with the height-above-buoyancy model proving to be less sensitive to variations in the thickness of the deforming sediment layer than the model proposed by Alley (1990). The differences between the two models examined here highlight the need for a comprehensive comparison of all the methodologies for calculating basal-ice motion currently in use.

The marginal shear stress of Ice Stream B, West Antarctica

AbstractTo ascertain whether the velocity of Ice Stream B, West Antarctica, may be controlled by the stresses in its marginal shear zones (the “Snake” and the “Dragon”), we undertook a determination of the marginal shear stress in the Dragon near Camp Up B by using ice itself as a stress meter. The observed marginal shear strain rate of 0.14 a−1is used to calculate the marginal shear stress from the flow law of ice determined by creep tests on ice cores from a depth of 300 m in the Dragon, obtained by using a hot-water ice-coring drill. The test-specimen orientation relative to the stress axes in the tests is chosen on the basis ofc-axis fabrics so that the test applies horizontal shear across vertical planes parallel to the margin. The resulting marginal shear stress is (2.2 ± 0.3) × 105Pa. This implies that 63–100% of the ice stream’s support against gravitational loading comes from the margins and only 37–0% from the base, so that the margins play an important role in controlling the ice-stream motion. The marginal shear-stress value is twice that given by the ice-stream model of Echelmeyer and others (1994) and the corresponding strain-rate enhancement factors differ greatly (E≈ 1–2 vs 10–12.5). This large discrepancy could be explained by recrystallization of the ice during or shortly after coring. Estimates of the expected recrystallization time-scale bracket the ∼1 h time-scale of coring and leave the likelihood of recrystallization uncertain. However, the observed two-maximum fabric type is not what is expected for annealing recrystallization from the sharp single-maximum fabric that would be expected in situ at the high shear strains involved (γ ∼ 20). Experimental data from Wilson (1982) suggest that, if the core did recrystallize, the prior fabric was a two-maximum fabric not substantially different from the observed one, which implies that the measured flow law and derived marginal shear stress are applicable to the in situ situation. These issues need to be resolved by further work to obtain a more definitive observational assessment of the marginal shear stress.

The marginal shear stress of Ice Stream B, West Antarctica

AbstractTo ascertain whether the velocity of Ice Stream B, West Antarctica, may be controlled by the stresses in its marginal shear zones (the “Snake” and the “Dragon”), we undertook a determination of the marginal shear stress in the Dragon near Camp Up B by using ice itself as a stress meter. The observed marginal shear strain rate of 0.14 a−1is used to calculate the marginal shear stress from the flow law of ice determined by creep tests on ice cores from a depth of 300 m in the Dragon, obtained by using a hot-water ice-coring drill. The test-specimen orientation relative to the stress axes in the tests is chosen on the basis ofc-axis fabrics so that the test applies horizontal shear across vertical planes parallel to the margin. The resulting marginal shear stress is (2.2 ± 0.3) × 105Pa. This implies that 63–100% of the ice stream’s support against gravitational loading comes from the margins and only 37–0% from the base, so that the margins play an important role in controlling the ice-stream motion. The marginal shear-stress value is twice that given by the ice-stream model of Echelmeyer and others (1994) and the corresponding strain-rate enhancement factors differ greatly (E≈ 1–2 vs 10–12.5). This large discrepancy could be explained by recrystallization of the ice during or shortly after coring. Estimates of the expected recrystallization time-scale bracket the ∼1 h time-scale of coring and leave the likelihood of recrystallization uncertain. However, the observed two-maximum fabric type is not what is expected for annealing recrystallization from the sharp single-maximum fabric that would be expected in situ at the high shear strains involved (γ ∼ 20). Experimental data from Wilson (1982) suggest that, if the core did recrystallize, the prior fabric was a two-maximum fabric not substantially different from the observed one, which implies that the measured flow law and derived marginal shear stress are applicable to the in situ situation. These issues need to be resolved by further work to obtain a more definitive observational assessment of the marginal shear stress.

A numerical model of interactions between a marine ice sheet and the solid earth: Application to a West Antarctic ice stream

A time‐dependent numerical model has been constructed that simulates retreat of a West Antarctic ice stream from the edge of the continental shelf during the Holocene period of rising sea level. This paper describes a method for computing the deformation of the solid earth caused by changes in ice and water loading during retreat of the ice stream. The relative sea level changes caused by earth deformation are incorporated as a feedback mechanism in the ice stream model. Elastic and viscous uplift of the earth, caused by thinning of the ice stream and its catchment area, is found to delay retreat of the grounding line relative to computed retreat when the ice stream was assumed to be resting on a rigid earth. Computed retreat of the grounding line began very slowly, at 15,000 B.P., because of rising eustatic sea level. Retreat accelerated after about 30 km, at 13,000 B.P., because of increasing water depth and a seabed sloping down toward the ice sheet interior. By 8000 B.P. the feedback effects of earth deformation caused retreat of the grounding line to be delayed by about 1000 years, relative to computed retreat on a rigid earth. The elastic and viscous earth response exerted a moderating influence on the computed retreat rate because uplift at the grounding line partially counteracted the effect of increasing water depth. Computed retreat of the grounding line slowed, and gradually stopped, near the present position of the ice stream grounding line. Uplift of the seafloor caused the total retreat distance to be reduced by 80 km relative to retreat computed on a rigid earth. A readvance of 20 km of the grounding line occurred between 3000 B.P. and the present because continuing viscous uplift caused sea depth to decrease. Resistance from the Ross Ice Shelf was found to be of primary importance in bringing retreat of the grounding line to a stop in the Ross Embayment, as suggested by earlier investigators. Within the context of a given ice shelf retreat history the feedback effects of earth deformation caused a reduction of the grounding‐line retreat rate, a reduction of the total computed retreat distance, and a readvance of the grounding line after eustatic sea level stopped rising.