C42A-01 INVITED
Basal Conditions for Pine Island and Thwaites Glaciers Determined using Satellite and Airborne Data
We used models constrained by remotely sensed data for Pine Island and Thwaites glaciers to infer basal properties, which are difficult to observe directly. The results indicate strong basal melting in areas above the grounding lines of both glaciers where the speeds are fast and the basal shear stresses are high. Farther inland, both glaciers have mixed bed conditions with extensive apparent areas of both crystalline bed rock and weak till. In particular, there are weak areas along much of its main trunk that could prove unstable if Pine Island Glacier retreats past the band of strong bed just above its current grounding line. In agreement with earlier studies, our model shows a strong sensitivity to small perturbations in the grounding line position. These results also reveal a strong sensitivity to the assumed sliding model, with non-linear sliding laws producing substantially greater dynamic response than earlier simulations that assume a linear-viscous till. Finally, our results using a plastic-bed are at least compatible with the limited observational constraints, which is consistent with recent theoretical work that shows basal shear stress cannot grow without bound.
C42A-02
Persistent drainage from a subglacial lake reducing glacier sliding
We present velocity observations of a glacier outlet in Vatnajökull, Iceland, deduced from interferometric SAR (InSAR) data obtained during the ERS1/2 tandem mission in 1995-2000. Up to 70% decrease in glacier motion was observed subsequent to a large jökulhlaup from the subglacial lake Grímsvötn in November 1996 and it had not reached its former flow rate at the end of our study period. The jökulhlaup damaged the lake ice-dam causing persistent drainage from Grímsvötn. InSAR observations of water accumulation within the lake suggest that a leakage of >3 m3 s-1 prevailed throughout our study period. Our interpretation of the observed reduction in glacier motion is that the water drained underneath the whole length of the glacier outlet through tunnels at low water pressure. Further, the tunnel flow dried up a linked cavity system at higher water pressure, underneath the upper and centre part of the glacier, which prior to the jökulhlaup sustained basal sliding.
C42A-03
Subglacial Meltwater Drainage at Paakitsoq, West Greenland: Insights From a Distributed, Physically Based Numerical Model
Recent results indicate that surface melting influences the dynamics of the Greenland Ice Sheet margin through meltwater input to a subglacial drainage system, but the hydrological characteristics of this drainage system and the degree to which variations in subglacial water pressure enhance or impede ice flow remain uncertain. Investigating the hydrology of this relatively inaccessible subglacial system requires a numerical modeling approach in which spatial and temporal variations in subglacial water pressures are calculated in response to the main controlling variables (subglacial drainage system structure and morphology and surface water inputs). We present the preliminary findings of such a study for the Paakitsoq region of W. Greenland, north of Jakobshavn Isbrae. Recent airborne radar data are used to construct surface and bed DEMs of the region from which patterns of subglacial hydraulic potential are derived. These are used to define the subglacial drainage system structure (the location, alignment and interconnection of major drainage pathways). Water flow along these pathways is modeled using a component of the United States Environmental Protection Agency's Storm Water Management Model (SWMM) modified to allow for enlargement and closure of ice walled channels (cf. Arnold et al., Hydrol. Processes, 12, 1998). We assess the model's ability to deal with two types of input: rapid lake drainage events, and diurnally varying melt inputs calculated from a degree-day model. We perform sensitivity tests to determine the effects of model parameters on modeled channel cross-sectional area, water pressure and subglacial flow. Finally, we simulate drainage beneath the ice sheet for a summer melt season and compare the results with measured proglacial stream discharges. Preliminary results suggest that channelized flow is only possible close to the ice sheet margin where ice is thin and water inputs are large. Distributed drainage is predicted beneath thicker inland regions.
C42A-04 INVITED
Understanding changes in the subglacial environment through modeling of internal layers
Observations from ice-penetrating radar data indicate that internal layers in ice sheets (isochrones) generally conform to bed topography in slow-moving regions but can be deformed due to changes (both spatial and temporal) in boundary conditions. The shapes of internal layers provide information about which boundary conditions dominate; changes at the surface (accumulation and/or strain rates) have a greater impact on near-surface layers while changes at the bed (basal melt or sliding rates) have a greater impact on deeper internal layers. We use a simple kinematic model to understand observations of deformed, deep internal layers in Greenland and Antarctica to help interpret processes that affect ice dynamics in each location. In Greenland, radar data reveal several englacial drainage features, likely moulins, which may connect to the subglacial system. Strong evidence of a supra to subglacial connection is found where these moulin-features are coincident with strongly downwarped, truncated internal layers as model results suggest that localized subglacial melting generates downwarped layers. In Antarctica, we also observe strongly downwarped internal layers, but here they are not coincident with englacial drainage features. Instead, downwarped layers are coincident with multiple basal crevasses indicating floatation of this region in the past. That is, while layer downwarping we have observed in Antarctica also appears to be in response to subglacial melting, the mechanism for subglacial melt is likely very different to our observations in Greenland. We find that model results can provide important insight into past and present boundary conditions but our simple model is limited in that it cannot determine the mechanism for melt nor detail the timing of past changes. As a result, models of internal layers should be used in conjunction with other observations that can provide this information.
C42A-05
Surface Energy Balance Model of the Helheim Glacier, Southeast Greenland
We present a distributed surface energy balance model of Helheim Glacier based on data collected by Automatic Weather Stations (AWSs) in the summer of 2008. Measurements were made at two sites in the ablation zone, at different elevations, overlapping in operation period. Using a digital elevation model we quantify the melt over the entire glacier taking topographic shading and reflection effects into account. A spatial model of the melt rates can be used to search for a correlation between these and occurrences of rapid glacier movement as a test of the hypothesis of melt water-lubricated enhanced flow. This correlation can also provide a basis for estimating the time-lag between melt water generation at the surface and its arrival at the bed through moulins, channels and the like. In addition to quantifying direct mass-loss processes, the model output has multiple uses in future efforts, e.g. crevasse-spreading models or salinity studies of the fjord.
C42A-06
Numerical model studies of melting and freezing in lakes, canyons and crevasses on the surface of an ice sheet.
Recent observations of the Greenland ice sheet suggest that meltwater movement both on the surface and within englacial conduits can have a significant effect on the ice sheet's flow and possibly its mass balance. Motivated by these observations, we investigate, using numerical models, the thermal and geometrical evolution of several "canonical" types of meltwater features typically observed on the Greenland ice sheet. These features include multi-year meltwater lakes that are subject to seasonal environmental and solar forcing, surface meltwater channels of various scales, and standing meltwater in surface cracks and nascent moulins. The model physics includes heat transfer (including radiative and convective effects) within the combined ice and water domains, and features prediction of evolving ice/water phase boundaries that can either change the geometry of the ice/water domains or even completely eliminate the water phase (e.g., in the case of a failed moulin). Initial numerical experiments examine the evolution of water-filled cracks that are either located subaerially, immediately at the surface of the ice sheet, or at the bottom of a larger standing meltwater feature, such as a meltpond. These initial experiments shed light on the question of whether large standing meltwater features are necessary in the initial formation of moulins.
C42A-07
Ice Stream Shear Margin Basal Melting, West Antarctica
Basal water lubricates and enables the anomalous flow feature of ice streams in the West Antarctic Ice Sheet. As surface melt is insufficient to supply the base with the volume of water known to be there, basal melting must be the source of this water. How basal melt patterns vary spatial can be an insight into the dynamics of ice streams, which remain incompletely described by glaciological theory. Through a heuristic model extended from the work of Whillans and Van der Veen (2001) and Van der Veen et al. (2007) a spatial pattern of basal melt for the Whillans Ice Stream with high melt rates (20 to 50 mm/a) located under or directly adjacent to shear margins emerged. This pattern offers hypotheses for the onset of streaming flow, shear crevasse development and observed morphological changes of a slowing and widening ice stream. The limitations and the uncertainties of this model make the determination of exact basal melt rates difficult, but the patterns of melt rate distribution are robust. This allows for a perspective to better understand current dynamics and how basal melt may play a role in the ice stream's future development.
C42A-08 INVITED
Melt-flow Acceleration Effects on the Greenland Ice Sheet
The first observations of accelerated ice-sheet flow at the equilibrium line in West-central Greenland during summer melt periods of 1996 to 1999 indicated that surface melt-water rapidly propagated to the base and enhanced the basal sliding (Zwally et al., 2002). Additional research has confirmed that melt-water from lake drainage rapidly propagates to the base causing ice acceleration (Das et al., 2008) and that the melt- acceleration effect is widespread in ice terminates on land and the flow in outlet glaciers (Joughin et al., 2008; van de Wal et al., 2008). However, the climatic significance of the effect has been questioned ( van de Wal et al., 2008; Price et al., 2008), as well as whether accelerations at the equilibrium line are caused by local meltwater injection or by longitudinal coupling of downstream melt-acceleration (Price et al., 2008). Additional velocity measurements since 1999 show further characteristics of the melt-acceleration in the ice flowline though Swiss Camp, which terminates on land, and in a nearby flowline that terminates in an outlet glacier. Accelerations as large as three times the average winter velocity are observed during stronger melt events. At downstream locations, accelerations begin earlier in the melt season, but accelerations at multiple sites along a flow line later in the season occur simultaneously. At the equilibrium line, a short period of surface uplift of about 50 cm occurs when the flow abruptly changes from acceleration to deceleration, apparently caused by ice compression during the transition. At downstream locations, the surface rises at the beginning of the melt season and drops at the end of seasons suggesting an uplift forced by sub-ice water and sediment. The net additional displacement of several meters during the summer is about 3 to 5 % of the total annual displacement, and is increasing as summer temperatures and surface melting increases. Equivalence of the net additional displacement at upstream and downstream sites indicates no net longitudinal ice strain after the acceleration-deceleration periods. Approximate equivalence of the ratio of peak summer velocities to average winter velocities along the flowline indicate that local melt-acceleration is occurring at and above the equilibrium as well as from longitudinal coupling of downstream effects.