Search

Large-ensemble perturbed-parameter forward ice-flow modeling can provide useful insights to uncertainties in inversions for basal drag or other ice-flow parameters. Inversion and data assimilation provide estimates of poorly known parameters that are essential for accurate prognostic modeling. Because ice flow depends on many such parameters with their associated uncertainties, which may interact in nonlinear ways, full uncertainty assessment for parameter estimates is challenging. With rising computational power, it is increasingly practicable to explore co-dependencies and sensitivities. Here, we use a well-characterized higher-order flowline model configured for a well-lubricated ("shelfy") ice stream to run large ensembles, perturbing the magnitude and spatial pattern of basal drag, basal topography, and input flux. We find for steady state that ice-stream velocity and thickness along the entire domain are especially correlated to drag at the downstream end, but with greater local correlation during transients. The modeled effects of basal topographic perturbations on velocity and ice thickness are primarily local. Perturbations of input ice fluxes interact with the others in interesting ways. These insights point to the value of inversions informed by concentrated observations during forced transients such as lake-drainage events, accumulation-rate fluctuations or ice-shelf losses, and to the care needed when interpreting local results of some inversions for basal-drag parameters. • Perturbed-parameter ensembles reveal ice-flow-model uncertainties and interactions. • Basal-friction perturbations at the downstream end especially affect ice-stream flow. • During transients, basal-friction perturbations influence local ice-stream flow more. • Basal-topography perturbations mainly affect ice-stream thickness locally. • Large uncertainties in input flux can obscure other ice-flow controls. [ABSTRACT FROM AUTHOR]

Abstract: New model results show that a glacier advance caused by deposition of a rock avalanche on the ice will be followed by stagnation of the advanced ice lobe, producing distributed, hummocky deposits quite different from the single moraine ridges typically dated in paleoclimatic reconstructions. We obtain these results using a numerical flow-line glacier model with superimposed rock debris. In particular, our results imply that the single moraine ridge known as the Waiho Loop of the Franz Josef Glacier, New Zealand, is not the result of an advance triggered by a rock avalanche, else there would be a widespread, hummocky, glacial stagnation deposit associated with it. [Copyright &y& Elsevier]

Abstract: The great ice sheets covering Antarctica and Greenland were, traditionally, believed to take thousands of years to respond to external forcing. Recent observations suggest, however, that major changes in the dynamics of parts of the ice sheets are taking place over timescales of years. These changes were not predicted by numerical models, and the underlying cause(s) remains uncertain. It has been suggested that regional oceanic and/or atmospheric warming are responsible but separating the influence and importance of these two forcings has not been possible. In most cases, the role of atmospheric versus oceanic control remains uncertain. Here, we review the observations of rapid change and discuss the possible mechanisms, in the light of advances in numerical modelling and our understanding of the processes that may be responsible. [Copyright &y& Elsevier]

• Timing and rate of glacier retreat is dictated by basal rheology and topography. • Retreat rate across mixed-rheology beds depends on spacing between basal bumps. • Ice-ocean interactions around isolated topographic highs impact stability. The spatially variable basal flow law suggested by geophysical data from Thwaites Glacier, West Antarctica produces modeled ice-flow response to warming that differs notably from the response of commonly assumed spatially uniform flow laws, with implications for future sea-level rise. Unlike many ice-sheet outlets, Thwaites flows across rather than along prominent topography, with hard stoss faces where form drag is understood to give a low-stress-exponent basal flow law and soft lee-side tills likely to give nearly-plastic behavior. Applying the PSU 2-D higher-order flowline model to a range of idealized Thwaites-like topographies, we test the impact of nearly plastic, nearly viscous, and spatially variable basal rheology on forced grounding-line evolution. In agreement with previous findings, for spatially uniform basal rheology, the timing and rate of grounding-line retreat depend strongly on the bed exponent, with plastic rheology promoting longer stability at the current grounding line but faster retreat once destabilized. Additionally, we find that retreat over mixed-rheology beds is sensitive to the wavelength of the basal topography. For outlet-glacier flow across long-wavelength bumps (≳10 km), behavior falls between that for uniform viscous and plastic beds, yet closer to plastic. However, over shorter-wavelength topography, at times, grounding-line retreat is slower than for either uniform-rheology end-member. Accurately accounting for variable basal conditions in ice-sheet models thus is critical for improving projections of both the timing and magnitude of retreat. [ABSTRACT FROM AUTHOR]

Abstract Seismic measurements on Thwaites Glacier show a spatially variable bed character, with implications for ice-sheet stability. The West Antarctic Ice Sheet is losing mass rapidly through outlet glaciers and ice streams in the Amundsen Sea Embayment, including Thwaites Glacier, where limited observations and modeling suggest that ice-flow rates depend on bed properties. Here we characterize bed properties of Thwaites Glacier based on amplitude analysis of reflection-seismic data from a ∼40-km-long profile collected in the approximate flow direction and two ∼10-km-long profiles transverse to flow. The upstream portion of the seismic profile reveals a ∼12-km long sedimentary basin with ice-flow-aligned bedforms capped by a continuous till layer that is likely soft and deforming (porosity ∼0.4–0.45), with several locations where water has pooled at the bed. Downstream of the sedimentary basin, the bed rises by ∼400 m over ∼25 km into subglacial highlands. Our seismic survey of these subglacial highlands reveals strong spatial variations in bed character across rugged topography (∼200 m amplitude at ∼2- to 5-km wavelength) resembling crag-and-tails. Till on the stoss sides (facing upglacier) of topographic highs is more consolidated (porosity ∼0.3–0.35 or lower), whereas the lee sides (facing downglacier) and flat regions exhibit porosity similar to the till of the upstream sedimentary basin. Modeling studies could use the observed correlation between bed character and bed aspect and slope to extend our observations to other parts of Thwaites Glacier, resulting in more-realistic models of future grounding-line retreat. Our findings highlight the need for more geophysical constraints on bed properties for important outlets in Antarctica and Greenland. Highlights • P -wave reflectivities show bed character varies with subglacial morphology. • Relatively flat, ∼12-km-long basin is "soft" bedded, capped by dilatant till. • "Hard" beds on stoss, "soft" beds on lee sides of topography in subglacial highlands. • Localized water layers are scattered beneath upper Thwaites Glacier. [ABSTRACT FROM AUTHOR]

Rapid retreat of Antarctic ice shelves in a warmer climate remains challenging to predict, contributing to uncertainties in projections of sea level rise. The collapse of the Larsen B Ice Shelf on the Antarctic Peninsula in 2002 was perhaps the most dramatic example of ice shelf retreat on record. Despite a general consensus that regional climate warming is associated with ice shelf collapse, knowledge of the precise details leading to and following collapse is limited. Here we examine a comprehensive set of satellite observations, modeling experiments, and climate reanalysis data to elucidate calving behavior, flow dynamics, and mechanical conditions of the northern Larsen Ice Shelf since the mid-1960s. Our analysis reveals that the Larsen B collapse in March 2002 was the last phase of a calving sequence that began in 1998, and was controlled by both atmospheric and oceanic anomalies that weakened the ice shelf structure over time. The collapse was preceded by a transition from infrequent calving of large tabular-icebergs and localized flow acceleration to frequent calving of small-icebergs and ice-shelf-wide acceleration. The intermittent pulses of the calving sequence were triggered by failure of the northern shear margin, most likely due to ocean-driven weakening by incision of sub-ice-shelf channels, and were further controlled by the location of effective buttressing sources and preexisting crevasses. The limited observational data indicate a similar process for the Larsen A ice shelf. These calving events corresponded with warm anomalies associated with a La Niña / positive Southern Annular Mode teleconnection pattern. The results suggest that warm climate anomalies control the occurrence of calving, while the extent and speed of calving are governed by ice shelf geometry and mechanical conditions, in particular, the sturdiness of the "weakest" shear margin. Sudden widespread flow acceleration and frequent small-iceberg calving may serve as quantifiable precursors for ice shelf destabilization. • A transition in calving and acceleration style preceded collapse of Larsen Ice shelf. • Destabilization of the ice shelves was controlled by the weakest shear margin(s). • A La Niña / positive Southern Annular Mode teleconnection pattern promotes calving. • Sturdiness of shear margins is critical in ice shelf vulnerability to climate change. [ABSTRACT FROM AUTHOR]

The largest accelerations of glaciers and ice sheets are caused by changes in basal slip. Here we examine glacier speed and rain-induced accelerations using a near-continuous 26-month-long GNSS time series from a large maritime glacier (Tasman Glacier, New Zealand). During periods of high rain-rate we observe short-term increases in 24-hour speeds to up to 15-times background speed. Speeds calculated over 3-hour intervals increase to up to 36-times background speed. Acceleration events correspond with times when bed separation also increases rapidly indicating that the acceleration is associated with the growth of water-filled cavities at the bed. Glacier speeds then decrease prior to the reduction in bed separation, indicating cavity growth, not cavity extent, controls the acceleration. The short-term accelerations are superimposed on longer-term periods of enhanced velocity that persist for days to weeks and decay at similar rates to bed separation estimates and proglacial lake levels. A power-law relationship between observed rain-rate and speed exists at the glacier front and exhibits no apparent upper bound. Overall, we estimate that rain-induced accelerations account for 11–14% of Tasman Glacier's displacement during the observation period. The rain-rate–velocity relationship and the rainfall record since 2001 indicate that rain-induced speed-up events result in interannual variability in glacier displacement of 2%–13%. [ABSTRACT FROM AUTHOR]

Geological data indicate that global mean sea level has fluctuated on 10 3 to 10 6 yr time scales during the last ∼25 million years, at times reaching 20 m or more above modern. If correct, this implies substantial variations in the size of the East Antarctic Ice Sheet (EAIS). However, most climate and ice sheet models have not been able to simulate significant EAIS retreat from continental size, given that atmospheric CO 2 levels were relatively low throughout this period. Here, we use a continental ice sheet model to show that mechanisms based on recent observations and analysis have the potential to resolve this model–data conflict. In response to atmospheric and ocean temperatures typical of past warm periods, floating ice shelves may be drastically reduced or removed completely by increased oceanic melting, and by hydrofracturing due to surface melt draining into crevasses. Ice at deep grounding lines may be weakened by hydrofracturing and reduced buttressing, and may fail structurally if stresses exceed the ice yield strength, producing rapid retreat. Incorporating these mechanisms in our ice-sheet model accelerates the expected collapse of the West Antarctic Ice Sheet to decadal time scales, and also causes retreat into major East Antarctic subglacial basins, producing ∼17 m global sea-level rise within a few thousand years. The mechanisms are highly parameterized and should be tested by further process studies. But if accurate, they offer one explanation for past sea-level high stands, and suggest that Antarctica may be more vulnerable to warm climates than in most previous studies. [ABSTRACT FROM AUTHOR]

We present radio-echo sounding (RES), global positioning system (GPS), and active-source seismic data across the central portion of the Northeast Greenland Ice Stream (NEGIS). NEGIS widens downglacier from a small region of high geothermal flux near the ice divide. Our data reveal high-porosity (40+%) water-saturated till lubricating the ice stream. Ice accelerates and thins as it flows into NEGIS, producing marginal troughs in surface topography. These troughs create steep gradients in the subglacial hydropotential that generate parallel "sticky" and "slippery" bands beneath the shear margins. The low-porosity "sticky" sediment bands limit ice entrainment across the margins and thus restrict further widening, producing the long, narrow, and relatively stable ice stream. However, the observed relations among surface elevation, basal water routing, broad sedimentary drape, and till dilatancy suggest that rapid shifts in ice dynamics are possible, including rapid transmission of ocean forcing inland. The source and routing of the subglacial till are unclear, but our data help constrain hypotheses. [ABSTRACT FROM AUTHOR]

Abstract: Viscoelastic models of ice-shelf flexure and ice-stream velocity perturbations are combined into a single efficient flowline model to study tidal forcing of grounded ice. The magnitude and timing of ice-stream response to tidally driven changes in hydrostatic pressure and/or basal drag are found to depend significantly on bed rheology, with only a perfectly plastic bed allowing instantaneous velocity response at the grounding line. The model can reasonably reproduce GPS observations near the grounding zone of Bindschadler Ice Stream (formerly Ice Stream D) on semidiurnal time scales; however, other forcings such as tidally driven ice-shelf slope transverse to the flowline and flexurally driven till deformation must also be considered if diurnal motion is to be matched. [Copyright &y& Elsevier]

Abstract: We develop a model of an ice shelf-ice stream system as a viscoelastic beam partially supported by an elastic foundation. When bedrock near the grounding line acts as a fulcrum, leverage from the ice shelf dropping at low tide can cause significant uplift in the first few kilometers of grounded ice. This uplift and the corresponding depression at high tide lead to basal pressure variations of sufficient magnitude to influence subglacial hydrology. Tidal flexure may thus affect basal lubrication, sediment flow, and till strength, all of which are significant factors in ice-stream dynamics and grounding-line stability. Under certain circumstances, our results suggest the possibility of seawater being drawn into the subglacial water system. The presence of seawater beneath grounded ice would significantly change the radar reflectivity of the grounding zone and complicate the interpretation of grounded versus floating ice based on ice-penetrating radar observations. [Copyright &y& Elsevier]

Abstract: We describe efforts to model the Holocene extent of the Rhone Glacier, Switzerland, using four paleoclimate records as templates for paleo-equilibrium line altitude to identify candidate driving mechanisms of glaciers in the Alps. We evaluate the success of each paleoclimate template by comparing cosmogenic 10Be and 14C concentrations in pro-glacial bedrock derived from modeled glacier configurations to measured values. An adequate fit can be obtained using mean summer insolation for 46.5°N. However, use of the Dongee Cave, China, speleothem record yields the best fit by accounting for both sub-millennial (e.g. Little Ice Age and Medieval Warm Period) and multi-millennial climate variations (summer insolation). Our result indicates that glaciers in the Alps primarily responded to changes in insolation during the Holocene were smaller than today during the early Holocene when insolation was relatively high, and became larger during the mid to late Holocene. Superimposed on the first-order insolation response were shorter, sometimes large amplitude, length changes in response to short-lived climate events such as the Medieval Warm Period and the LIA. [Copyright &y& Elsevier]

Abstract: We present data on the triple Ar isotope composition in trapped gas from clean, stratigraphically disturbed ice between 2800 and 3040m depth in the GISP2 ice core, and from basal dirty ice from 3040 to 3053m depth. We also present data for the abundance and isotopic composition of O2 and N2, and abundance of Ar, in the basal dirty ice. The Ar/N2 ratio of dirty basal ice, the heavy isotope enrichment (reflecting gravitational fractionation), and the total gas content all indicate that the gases in basal dirty ice originate from the assimilation of clean ice of the overlying glacier, which comprises most of the ice in the dirty bottom layer. O2 is partly to completely depleted in basal ice, reflecting active metabolism. The gravitationally corrected ratio of 40Ar/38Ar, which decreases with age in the global atmosphere, is compatible with an age of 100–250ka for clean disturbed ice. In basal ice, 40Ar is present in excess due to injection of radiogenic 40Ar produced in the underlying continental crust. The weak depth gradient of 40Ar in the dirty basal ice, and the distribution of dirt, indicate mixing within the basal ice, while various published lines of evidence indicate mixing within the overlying clean, disturbed ice. Excess CH4, which reaches thousands of ppm in basal dirty ice at GRIP, is virtually absent in overlying clean disturbed ice, demonstrating that mixing of dirty basal ice into the overlying clean ice, if it occurs at all, is very slow. Order-of-magnitude estimates indicate that the mixing rate of clean ice into dirty ice is sufficient to maintain a steady thickness of dirty ice against thinning from the mean ice flow. The dirty ice appears to consist of two or more basal components in addition to clean glacial ice. A small amount of soil or permafrost, plus preglacial snow, lake or ground ice could explain the observations. [ABSTRACT FROM AUTHOR]

• 21,915 glacier-bed icequakes reveal dynamics of transient glacier slip processes. • Water pressure transient intensity directly influences (a)seismicity of glacier slip. • Glacier bed "sticky-spots" may represent the strongest—but not only—asperities. • Debris effects on drag and seismicity explain slip observation and theory misfits. Glacier slip modulates sea-level rise. Most slip models assume quasi-equilibrium process dynamics and debris-free basal ice, but observations regularly show debris-rich basal ice sliding under transient external forcings, placing theory and observation in tension. Seismicity generated by slip (icequake activity) provides a window into subglacial dynamics, but basal icequakes are often too sparse to resolve diagnostic trends at scales relevant to reveal slip-facilitating processes. We present a rich icequake catalog from an alpine glacier with sub-diurnal and decametric resolution that overcomes these limitations. We find characteristic trends between subglacial hydrology, slip, and seismicity that corroborate emerging models of transient slip over hard beds and highlight the importance of basal debris in slip dynamics. Specifically, icequakes are more abundant when subglacial cavities are likely large—a configuration that favors reduced drag in debris-free models—but this occurs as slip decelerates. Drag between entrained debris and the bed is enhanced under the same conditions and favors seismogenesis, which reconciles this discrepancy and may indicate slip-stabilizing feedback not considered in most models of glacier motion. Furthermore, we characterize spatial, temporal, and magnitude scaling patterns in this seismic catalog consistent with trends observed in previous glacier seismicity studies, suggesting that the slip-facilitating process dynamics characterized at Saskatchewan Glacier are relevant in other parts of the cryosphere. [ABSTRACT FROM AUTHOR]

We predict the perturbation to the Earth's length-of-day (LOD) over the Common Era using a recently derived estimate of global sea-level change for this time period. We use this estimate to derive a time series of “clock error”, defined as the difference in timing of two clocks, one based on a theoretically invariant time scale (terrestrial time) and one fixed to Earth rotation (universal time), and compare this time series to millennial scale variability in clock error inferred from ancient eclipse records. Under the assumption that global sea-level change over the Common Era is driven by ice mass flux alone, we find that this flux can reconcile a significant fraction of the discrepancies between clock error computed assuming constant slowing of Earth's rotation and that inferred from eclipse records since 700 CE. In contrast, ice mass flux cannot reconcile the temporal variability prior to 700 CE. [ABSTRACT FROM AUTHOR]

Understanding the processes that affect streaming ice flow and the mass balance of glaciers and ice sheets requires sound knowledge of their subglacial environments. Previous studies have shown that an extensive deformable subglacial sediment layer favors fast ice-stream flow. However, areas of high basal drag, termed sticky spots, are of particular interest because they inhibit the fast flow of the overriding ice. The stick-slip behavior of Whillans Ice Stream (WIS) is perhaps the most conspicuous manifestation of a subglacial sticky spot. We present new ice-thickness and seismic-reflection measurements collected over the main sticky spot in the ice plain of WIS, allowing us to elucidate its role in the behavior of the ice stream. Ice-thickness and surface-elevation data show that the sticky spot occupies a subglacial topographic high. Water flow in response to the hydrological potential gradient will be routed around the sticky spot if effective pressures are similar on the sticky spot and elsewhere. The seismic experiment imaged a laterally continuous basal layer approximately 6 m thick, having compressional wave velocities of greater than 1800 m s −1 and density greater than 1800 kg m −3 , indicative of a till layer that is stiffer than corresponding till beneath well-lubricated parts of the ice stream. This layer likely continues to deform under the higher shear stress of the sticky spot, and some water may be pumped up onto the sticky spot during motion events. [ABSTRACT FROM AUTHOR]

Pine Island Glacier (PIG) in the Amundsen Sea sector of the West Antarctic Ice Sheet (WAIS) is losing mass and contributing to global sea-level rise at an accelerating rate. Although recent observations and modeling have identified the incursion of relatively warm Circumpolar Deep Water (CDW) beneath the PIG ice shelf (PIGIS) as the main driver of this ice-mass loss, the lack of precise bathymetry limits furthering our understanding of the ice–ocean interactions and improving the accuracy of modeling. Here we present updated bathymetry and sediment distribution beneath the PIGIS, modeled by the inversion of aerogravity data with constraints from active-source seismic data, observations from an autonomous underwater vehicle, and the regional gravity-anomaly field derived from satellite gravity observations. Modeled bathymetry shows a submarine ridge beneath the middle of PIGIS that rises ∼350 to 400 m above the surrounding sea floor, with a minimum water-column thickness of ∼200 m above it. This submarine ridge continues across the whole width of the 45-km wide ice shelf, with no deep troughs crossing it, confirming the general features of the previously predicted sub-ice-shelf ocean circulation. However, the relatively low resolution of the aerogravity data and limitations in our inversion method leave a possibility that there is an undetected, few-kilometers-wide or narrower trough that may alter the predicted sub-ice-shelf ocean circulation. Modeled sediment distribution indicates a sedimentary basin of up to ∼800 m thick near the current grounding zone of the main PIG trunk and extending farther inland, and a region seaward of the submarine ridge where sediments are thin or absent with exposed crystalline basement that extends seaward into Pine Island Bay. Therefore, the submarine ridge marks the transition from a thick sedimentary basin providing a smooth interface over which ice could flow easily by sliding or sediment deformation, to a region with no to little sediments and instead a rough interface over which ice flows mainly by deformation. We hypothesize that the post-Last Glacial Maximum retreat of PIG stabilized at this location because of the spatial transition in basal conditions. This in turn supports the hypothesis that the recent retreat of PIG was strongly forced, probably by changes in ocean circulation, rather than occurring because of ongoing response to the end of the ice age or other changes inland of or beneath PIG. [ABSTRACT FROM AUTHOR]

Abstract: Ice-penetrating radar and kinematic GPS observations from Subglacial Lake Whillans (SLW), West Antarctica, reveal a shallow lake that is confined by steep basal topographic features. Radar imaging of SLW, although indicating wet basal conditions, is consistent with a water column depth of only ~6m or less during the near low-stand state at the time of the survey. Kinematic GPS profiles reveal that SLW is generally defined by a ~15m surface depression centered at S 84.237° W 153.614°. This point coincides with the area of lowest hydropotential in the lake basin and also the largest surface elevation range in ICESat data. Therefore this location appears to be an opportune site for subglacial access drilling of this active subglacial lake. A distinct basal topographic ridge on the grid south side of the basin is coincident with a strong contrast in relative basal reflectivity (~6dB), which we interpret as the lake boundary. Mapped hydropotential (calculated assuming hydrostatic equilibrium) shows that water enters the lake from the upstream direction and drains downstream. We hypothesize that a lake-level rise of ~5m plus flexural effects is sufficient to overtop a drainage divide. Thus SLW acts as a temporary storage basin for water beneath Whillans Ice Stream. [Copyright &y& Elsevier]

Abstract: Active subglacial lakes concentrate the distribution of water beneath ice sheets in both space and time. Seismic and surface observations from Subglacial Lake Whillans (SLW), West Antarctica, reveal that this active lake forms a persistent, albeit fluctuating, reservoir beneath Whillans Ice Stream. Imaging and phase observations using active-source seismic data show that SLW is a perpetually shallow feature. When surveyed near its low-stand, a water column was resolvable by seismic techniques along only 5km of the 45km profiled, with a maximum depth of less than 8m. Satellite altimetry shows that the high-stand adds no more than 3–4m to this. This water column presents a suitable drill site at S 84.240° W 153.694°. Elsewhere, the majority of the bed appears wet with soft sediment or water thicknesses of less than the imaging resolution of our data of approximately 2m. The surface expression of the active lake, previously revealed by ICESat elevation data and image differencing, generally corresponds to the seismic estimate of soft sediment or water, with notable exceptions occurring at the upstream and downstream ends of the lake. These exceptions indicate that SLW''s water column is very shallow or absent in places at low-stands, or has disconnected or transiently active and inactive portions. [Copyright &y& Elsevier]

Abstract: The motion of Bindschadler Ice Stream, West Antarctica, is dominated by sliding over a nearly plastic bed, according to analysis of kinematic GPS data using a new viscoelastic flowline model. Inversions of time-averaged velocity data with viscous ice-flow models can be consistent with multiple sliding laws, but propagation of velocity perturbations in a viscoelastic model can distinguish between sliding laws with different exponents. We develop such a model and apply it to a time series of velocity for the tidally modulated flow of Bindschadler Ice Stream (formerly Ice Stream D). Observed velocity perturbations are found to be consistent with a flow-law exponent m ≥8, which indicates basal motion with a relatively weak till bed; lower exponents consistent with motion dominated by deformation within the ice over a hard or frozen bed are found to be unlikely. This result suggests that Bindschadler Ice Stream would respond rapidly and significantly to any future loss of buttressing from the Ross Ice Shelf. [Copyright &y& Elsevier]

Abstract: The stick–slip motion and associated seismic emissions of Whillans Ice Stream (WIS), West Antarctica are two of the many recent observations of unexpected ice sheet behavior that are challenging traditional models of rapid glacier motion. Here we find that the WIS slip events repeatedly nucleate from a sticky-spot located in the middle of the ice stream, acting similar to an asperity in traditional models of earthquake physics. This region shows less motion than surrounding areas during the inter-slip periods, thus, concentrating stress and producing a pulse of seismic energy at the onset of slip. The propagating rupture breaks through an additional asperity in the northern part of the ice stream, producing another pulse of seismic energy 6–12min after initiation. Both asperities are regions of higher hydraulic potential than surrounding regions, suggesting they may have greater bed friction due to reduced water lubrication. Tidal pacing of the stress accumulation combined with fault healing controls the applied stress at failure, with higher stress giving faster propagation of the rupture front and higher slip velocities; these differences are reflected in the timing of the teleseismic arrivals. Our results highlight both the great sensitivity of large ice streams to small changes in external forcing and the importance of limited regions of the subglacial bed in controlling their motion, as well as providing insights to the mechanics of repeating earthquakes. [Copyright &y& Elsevier]

Abstract: The initial retreat of ice shelf grounding lines stabilized on seaward-sloping beds is influenced by the rheology of these beds, according to new model results. We apply a fully-coupled process model to investigate how the response of an ice stream to increased ocean temperature beneath its ice shelf depends on the assumed form of its basal flow law. For the same applied oceanic warming, the increase in grounding-line flux can be twice as great for an effectively-plastic bed as for a linear-viscous bed, suggesting that improved knowledge of the basal flow law of ice streams is necessary for predicting ice-sheet response to climatic forcing. [Copyright &y& Elsevier]

Thwaites Glacier (TG), West Antarctica, is losing mass in response to oceanic forcing. Future evolution could lead to deglaciation of the marine basins of the West Antarctic Ice Sheet, depending on ongoing and future climate forcings, but also on basal topography/bathymetry, basal properties, and physical processes operating within the grounding zone. Hence, it is important to know the distribution of bed types of TG's interior and grounding zone, and to incorporate them accurately in models in order to improve estimates of retreat rates and stability. Here we estimate properties of the bed by determining its acoustic impedance from amplitude analysis of reflection seismic data. We report on the results from two lines – a longitudinal (L-Line) and a transverse (N-Line) – on a central flowline of TG ∼100 km inland from the grounding zone. The data show considerable spatial variability in bed forms and properties, similar to results from a comparable survey farther inland. Notably, we find the same pattern here of hard (presumed bedrock) material on the stoss side of bumps and soft (presumed till) on the lee side. Physical understanding indicates the basal flow law describing motion over different regions of TG's bed likely varies from nearly-viscous over the bedrock regions to nearly-plastic over till regions, providing guidance for modeling. • The "hard" stoss and "soft" lee bed pattern persists downstream of prior observations. • Localized regions of water occur in the lees of bumps and the downstream end of lower Thwaites Glacier. • "Stoss-side-moats" in regions with high topographic relief are "hard" and may be locations of increased scour. [ABSTRACT FROM AUTHOR]