C31B-0313 0800h
Sub-glacier ocean properties and mass balance estimates of Petermann Gletscher's floating tongue in Northwestern Greenland
Petermann Gletscher is the largest and most influential outlet glacier in central northern Greenland. Located at 81 N, 60 W, it drains an area of 71,580 km2, with a discharge of 12 cubic km of ice per year into the Arctic Ocean. We finished a third field season in spring 2004 collecting in situ data on local climate, ice velocity, ice thickness profiles and bottom melt rates of the floating ice tongue. In addition, water properties (salinity and temperature profiles) in large, channel-like bottom cavities beneath the floating ice tongue were measured. The melt rates in these "channels" are in excess of 10 m/y and probably responsible for most of the mass loss of the Petermann Gletscher. The ocean measurements will be discussed in comparison with other ocean-profile soundings in the region. The bottom topography of the floating ice tongue has been mapped for some regions using surface-based ground penetrating radar at 25 MHz frequency and NASA aircraft radar profiles. A new map showing these under-ice features will be presented. GPS tidal motion has been measured over one lunar cycle at the flex zone and on the free floating ice tongue. These results will be compared to historic measurements made at the beginning of last century. A "worm-like" sheer feature of 80 m in height and several km in length has been studied using differential GPS readings. The mean velocity of the floating tongue ice is 1.08 km/y in that region, whereas the ice along the margin has a 30%-reduced flow speed, resulting in this strange looking sheer feature. Finally, the mass balance of the floating ice tongue will be discussed based on in situ measurements, aircraft profiles, satellite data, and model approximations.
C31B-0314 0800h
West Antarctic Balance Fluxes: Impact of Smoothing, Algorithm and Topography.
Grid-based calculations of balance flux and velocity have been widely used to understand the large-scale dynamics of ice masses and as indicators of their state of balance. This research investigates a number of issues relating to their calculation for the West Antarctic Ice Sheet (see below for further details): 1) different topography smoothing techniques; 2) different grid based flow-apportioning algorithms; 3) the source of the flow direction, whether from smoothed topography, or smoothed gravitational driving stress; 4) different flux routing techniques and 5) the impact of different topographic datasets. The different algorithms described below lead to significant differences in both ice stream margins and values of fluxes within them. This encourages caution in the use of grid-based balance flux/velocity distributions and values, especially when considering the state of balance of individual ice streams. 1) Most previous calculations have used the same numerical scheme (Budd and Warner, 1996) applied to a smoothed topography in order to incorporate the longitudinal stresses that smooth ice flow. There are two options to consider when smoothing the topography, the size of the averaging filter and the shape of the averaging function. However, this is not a physically-based approach to incorporating smoothed ice flow and also introduces significant flow artefacts when using a variable weighting function. 2) Different algorithms to apportion flow are investigated; using 4 or 8 neighbours, and apportioning flow to all down-slope cells or only 2 (based on derived flow direction). 3) A theoretically more acceptable approach of incorporating smoothed ice flow is to use the smoothed gravitational driving stress in x and y components to derive a flow direction. The flux can then be apportioned using the flow direction approach used above. 4) The original scheme (Budd and Warner, 1996) uses an elevation sort technique to calculate the balance flux contribution from all cells to each individual cell. However, elevation sort is only successful when ice cannot flow uphill. Other possible techniques include using a recursive call for each neighbour or using a sparse matrix solution. 5) Two digital elevation models are used as input data, which have significant differences in coastal and mountainous areas and therefore lead to different calculations. Of particular interest is the difference in the Rutford Ice Stream/Carlson Inlet and Kamb Ice Stream (Ice Stream C) fluxes.
C31B-0315 0800h
Iceberg surveillance: preliminary analysis of iceberg drift data
Automatic weather stations equipped with GPS (AWS/GPS) have recently been deployed on giant tabular icebergs in the Ross Sea. The aim of the project is to use the {\em{in situ}} data to formulate an accurate dynamical iceberg drift model for the Southern Ocean. A prerequisite for future dynamical drift models is a systematic evaluation of the data's integrity and its relationship to existing data sources. To this end a statistical analysis of the last 2 years is performed on the meteorological data from iceberg AWS/GPS stations and surrounding Antarctic stations. The results of this analysis are presented as a guide for future drift models.
C31B-0316 0800h
Hydroacoustic Observations of Antarctic-Derived Ice-Generated Tremor in the Indian Ocean
Hydroacoustic data from two sets of moored hydrophones within the Indian Ocean Basin exhibit two distinct tremor signals that can be associated with drifting icebergs and glacial features along the Wilkesland coast of eastern Antarctica. The first type of signal is a variable harmonic tremor (VHT) with spectral peaks (4-100 Hz) that fluctuate through time and can persist from several minutes to over an hour. The second signal is a cusped pulsed tremor (CPT) of regularly spaced packets of curved harmonic spectral bands (4-40 Hz) with the duration of a single pulse ranging from 25-60 seconds, the interval between pulses ranging from 30-90 seconds and a series of pulses lasting anywhere from 10 minutes to over an hour. Sound-channel based observations indicate fundamental frequencies for both signals are typically within the 4-10 Hz range, but can reach as high as 30 Hz in some instances. Received levels at two hydrophone stations ($34\deg$ S $114\deg$ E and $7\deg$ S $72\deg$ E) have maximum values of 142 and 133 dB re $1\mu$Pa (peak-to-peak), respectively, with propagation as a surface-reflected phase at lower latitudes shifting to a sound channel phase after passing through the convergence zone. While VHT signal source locations appear to include most of the Wilkesland coast ($85\deg$ E - $149\deg$ E longitude), we have been able to successfully correlate a subset of the loudest VHT signals, for which the signal azimuth could be constrained at two hydrophone stations, with the westerly-migrating iceberg B-15D from early 2002 through 2003. Most of the VHT signals, however, cannot be correlated with large icebergs, suggesting that these signals also may be sourced from smaller unnamed icebergs and coastal outlet glaciers. The CPT signal appears to source from five specific locations along the Antarctic coast where satellite imagery show active ice movement from ice streams and outlet glaciers. Because of the harmonic nature of both signals, resonance initiated by excitation of an ice mass or fluid-filled cavity within the ice could be a viable source mechanism for the tremor signals.
C31B-0317 0800h
RADARSAT Antarctic Mapping System 2 - System Overview
The Modified Antarctic Mapping Mission (MAMM), the second mission of the RADARSAT 1 Antarctic Mapping Project, occurred during the fall of 2000, after the first Antarctic Mapping Mission was successfully completed in 1997 and yielded the first high-resolution radar mosaic of Antarctica. The two goals of the MAMM were to produce high-resolution image mosaics of Antarctica north of 80$^{o}$S latitude and to measure the surface velocity field over coherent and/or trackable areas of the ice sheet north of 80$^{o}$S latitude for ice dynamics studies. The Radarsat Antarctic Mapping System 2 (RAMS2) is a custom-designed processing system to convert the MAMM data acquired during the fall of 2000 into map-quality image mosaics, coherence mosaics and velocity fields. The system starts with the processed single look complex (SLC) data sets and has two processing phases: image mosaic phase and interferometric processing phase. Image mosaic phase includes data ingest, image mosaic and coherence mosaic. Interferometric processing phase includes frame-based phase 1 processing, data take-based phase 2 processing and block-based phase 3 velocity mosaic. The frame-based interferometric processing includes image registration and interferogram generation for each framed SLC data pair. The phase 2 interferometric processing mosaics all the framed interferograms of each data take into one big mosaiced interferogram and do interferogram filtering, phase unwrapping and velocity field estimation on the data take bases. The phase 3 velocity mosaic produces block-based velocity maps by mosaicing dozens or hundreds of mosaiced interferograms. In this paper we present the system overview of both the image mosaic and interferometric processing phases, describe algorithms and innovations involved in the interferometric processing phase and the products RAMS2 provides for the science community. Some examples of coherence mosaics and velocity mosaics are also presented.
http://www.vexcel.com
C31B-0318 0800h
Building an Integrated View of Antarctica Using Products from the Radarsat Antarctic Mapping Project
From the IGY through the early 1970's, scientific knowledge about Antarctica was gleaned primarily from observations made at points along traverse routes or along sparsely spaced aircraft flight lines. Analysis of these observations provided tantalizing clues about the nature and behavior of the ice sheet and the bedrock beneath. But the data themselves could not provide a continental scale assessment of how different glaciological and geophysical regimes interacted. The situation began to change rapidly in the mid-1970's with the launch of several active and passive sensors on satellites positioned in polar orbit. Still, none of the instruments alone or together could provide important large-scale information on several key geophysical variables. For that reason, NASA and the Canadian Space Agency began planning in the early 1980's for an imaging campaign using the Synthetic Aperture Radar to be carried by RADARSAT 1. The science driving the campaign was to obtain a benchmark for detecting changes in the continent by comparison with earlier and subsequent data. The resulting RAMP Antarctic image mosaic helped reach that goal and has been used to map the coastline of the continent in great detail, to study and contrast glaciological regimes about the continent, and to examine glaciological process such as the evolution of ice shelves by observing ice shelves at various stages of development about the continent. But a single instrument or data set is rarely ever able to answer complex scientific questions. For that reason it becomes useful and very interesting to use the RAMP mosaic as a basis for integrating other continental scale observations into a common framework referencable to features observable on the surface. In this paper, we present an ensemble of several continental data sets including velocities so far computed from the RAMP data set. We compare the RAMP image mosaic with: surface topography patterns using the RAMP Digital Elevation Model; BEDMAP compiled basal topography; a RAMP prepared model of surface balance velocities; and interferometric surface velocities derived from the RAMP data set. We find good correlations between the basal topography and backscatter strength in the vicinity of the Belgica Highlands. The result is exciting because it suggests that inferences about basal topography and properties can be made using the image mosaic in regions where basal topography data are sparse or completely absent. We also find that there is good correlation between image-mosaic-inferred glaciological structures and the surface balance velocity field. For example, the modeled velocities patterns capture the extensive network of ice streams draining into the Filchner Ice Shelf. These ice streams are revealed in the RAMP mosaic by the intense crevassing that occurs along their margins and which appears bright in radar imagery. Comparisons between the balance velocity model and measured RAMP velocities also reveal where the ice flow is more complex than predicted. West of the Amery ice shelf, measured velocities reveal the presence of numerous small ice streams or outlet glaciers that snake through the region but which are absent in the velocity model.
C31B-0319 0800h
A New MODIS-based Mosaic of Antarctica: MOA
A digital image map of the Antarctic continent assembled from ~100 MODIS (Moderate Resolution Imaging Spectroradiometer)band 1 and band 2 orbit swaths is presented. The map is comprised of both Terra and Aqua MODIS images acquired during the 2003-2004 austral summer. Multiple images of all areas are combined in an image-enhancement scheme to provide surface resolutions between 125 and 250 meters. Images were destriped, georeferenced, and resampled using the MS2GT software available at NSIDC. Acquisition times were limited to 0500 - 1300 UT continent-wide to provide a more uniform illumination direction of ice surface morphology across image seams, while at the same time clearly image linear snow features of every orientation. Regions of cloud cover and intense forward scattering from snow were removed. Both channels are compiled and an approximate grain size algorithm is applied in a separate data layer. The mosaic is rendered in a 125-meter grid identical to the RAMP (Radarsat Antarctic Mapping Project) Antarctic Mapping Mission - 1 (AMM-1) 125-meter mosaic. The image map reveals a host of detailed features on the ice sheet, ice shelves and fast ice of the continent, and provides an accurate rendition of mountain and nunatak features. In particular, snow dune features, ice flowlines and streaklines, crevasse regions, and ice shelf rifts are imaged clearly. Mean grain size information was also extracted from an normalized difference of the band 1 and band 2 images, highlighting blue ice areas. The MOA image map is complimentary to the radar-image-based RAMP mosaic, highlighting true surface morphology without a subsurface volume backscattering component. This results in a better discrimination between accumulation- and crevasse-related subsurface changes and surface features.
C31B-0320 0800h
Superimposed ice accumulation zone as mapped by GPR and satellite SAR
Ground-Penetrating Radar (GPR) profiles have been recorded annually along the centerline of Kongsvegen, Svalbard since 1999, and a dense grid was obtained in the upper ablation and firn areas in Spring 2003. The GPR operates at 500 MHz, and grid profiles reach depths equivalent to about 24 and 19 m in firn and ice respectively. GPR profiles reveal distinct layering in the firn area, starting upglacier from around the mean equilibrium line altitude, and increasing in thickness upglacier. These easily detected layers correspond to previous years' summer surfaces, and can be correlated in the firn area with the net balance measured at stakes as part of the mass balance monitoring program. Below the firn-ice transition (FIT) the layers are due to superimposed ice formation. Here the layers are more irregular, are characterized by weaker back-scatter, and appear as prograding deposits, the latter consistent with the mechanism of superimposed ice formation in which snowpack water flowing downglacier is frozen onto the glacier in thin sheets. The FIT as imaged by GPR coincides in location with a transition from medium to high back-scatter observed in satellite SAR imagery. We compare the medium backscatter zone in the SAR imagery to a thickness map of the superimposed ice layers obtained from the GPR grid, and to ice cores. From the area extent derived by SAR imagery and the depth distribution imaged by GPR, and a DEM of the glacier surface, it seems that the accumulation of superimposed ice is governed strongly by the surface topography. Furthermore, time-series of SAR images and GPR centreline profiles show that the location of the FIT is retreating, in response to a record number of years with negative mass balance.
C31B-0321 0800h
Hubbard Glacier and the Potential Damming of Russell Fiord from Analysis of Small-Aircraft Laser Altimeter Data and NASA SRTM-USGS-Canada DEMs
Small-aircraft Global Positioning System located laser altimeter profile data acquired in Aug. 2003 with a NASA - German Aerospace Center Shuttle Radar Topography Mission (SRTM) X-Band interferometic synthetic radar digital elevation model (DEM) acquired in Feb. 2000 are combined with U.S. Geological Survey and Natural Resource Canada photogrammetric DEMs whose data derive from 1961 and 1976 to estimate elevation changes on Hubbard Glacier and the noted tributary glaciers Valerie and Cathedral. Systematic errors in the earlier DEMs are estimated using fixed terrain. Winter snow depth from Sept. 1999 to 22 Feb. 2000 is estimated and used to adjustment the SRTM DEM elevations. The SRTM DEM elevations are adjusted to a mean sea level reference frame using GEOID99-Alaska (National Geodetic Survey geoid model). Observed surface elevation changes are spatially non-uniform. Valerie and Cathedral show markedly different water equivalent thickening and thinning characteristics apparently due to glacier dynamics. Preliminary results indicate the main flow band of Hubbard shows a draw-down in the upper ablation area of about 1.7 $\pm$ 1 m/yr, whereas in the lower accumulation area there is a build-up of about 2.0 $\pm$ 1 m/yr in the 1999 to 2003 time period. The middle accumulation area shows a net thickening of about 1.0 $\pm$ 1 m/yr, in the 1976 to 2003 time period. The upper accumulation area shows small net thinning of about 0.1 $\pm$ 1 m/yr in the same time period. Along the main flow band of Hubbard Glacier a mean thickening of about 0.8 $\pm$ 1 m/yr in the 1961/76 to 2003 time period is observed. These changes have occurred in spite of the brief June 2002 damming of Russell Fiord.
C31B-0322 0800h
A Time Domain Reflectometry System for Characterizing the Dielectric Constant Versus Unfrozen Water Content of Glacial Ice
Thinning of the world's ice masses has contributed significantly to sea level rise over the last century and is likely to have a much more significant impact on sea level in the future. Hence, a key challenge is to predict the manner in which ice masses will respond to climate change, using dynamic models. These models require detailed knowledge of glacier mechanical properties, which depend on the proportion and distribution of unfrozen water at ice grain boundaries. Water content and distribution is also the strongest influence on the velocity of electromagnetic (radar) waves in ice, which makes radar a very useful tool in modern glaciology. In order to interpret radar data to give ice properties it is necessary to characterize the relationship between the unfrozen water content within ice, and its dielectric constant. This can be done using laboratory measurements of glacial ice-core dielectric constant coupled with thin section microscopic analysis of ice cores to measure porewater contents. Design and testing of the equipment for characterization of ice core dielectric constant using Time Domain Reflectometry (TDR) is reported here. TDR is a technique developed for characterization of the dielectric constant, and hence the water content, of soils. Conventionally, the TDR waveguides are embedded in the medium under investigation. However, it is difficult to insert TDR electrodes into solids such as rock and ice. To circumvent this difficulty, `press on' TDR waveguides have been developed for use in rock; here we present a modified version for use with ice cores. The problems associated with TDR measurements on glacial ice compared with soils are i) that the unfrozen water content is relatively low, leading to low dielectric constant and short two-way travel times, and ii) that some ice facies have relatively large crystal sizes compared with the sensed volume of conventional probe designs. Several press-on probe designs have been developed to counter these problems and aspects of their performance are reported.
C31B-0323 0800h
Borehole Optical Stratigraphy and Neutron Scattering Density Measurements at Summit, Greenland
We have made side-by-side measurements in several boreholes at Summit, Greenland, using Borehole Optical Stratigraphy (BOS) and Neutron-Scattering density logging techniques. The BOS logs show a high degree of correlation at shallow depths with Neutron Scattering logs taken in the same borehole. This supports the hypothesis that BOS is detecting changes in both density and grain size, as well as possibly other parameters. The correlation between returned brightness and density decreases with depth and finally becomes negative, and we investigate the possible reasons for this inversion, which may be related to changes in densification regime from grain-boundary sliding to pressure sintering.
C31B-0324 0800h
Local climate on and around a glacier - a case study of Storglaci\"{a}ren
It is sometimes necessary to transform the climate data from a station to another station on a glacier. However, it is generally not so easy to do so since a glacier has its own specific microclimate. At Storglaci\"{a}ren in the summer 2003, air temperature and wind speed were measured at two weather stations set up near the center of the glacier and at the ridge of the bordering valley wall 300 m above the glacier surface. Additional continuous measurements are made at a weather station at Tarfala Research Station, which is located 1 km down glacier (1135 m a.s.l.). The result show a slight temperature difference between ridge and glacier stations because of the cooling effect by the glacier. Thus, temperature lapse rate is different. Wind speed on the ridge was higher than the other two in most cases, and the difference was largest during periods of high wind speed. The correlation between wind speed at the ridge and the other sites is weak.
C31B-0325 0800h
A Multifaceted View of a Glacier Speed-Up Event
We present a suite of observations that elucidate the subglacial and englacial conditions leading to an eight fold speed-up of a valley glacier. Data were collected on the 7.5 km long temperate Bench Glacier, Chugach Mountains, Alaska. We documented the speed-up event, which lasted for approximately one week of June, with data that include: 1) subglacial water pressure measured in a grid of 16 closely spaced boreholes to the bed; 2) water pressure measured in an additional 31 boreholes at 14 locations spanning the length of the glacier; 3) discharge of the terminus outlet stream; 4) GPS measurements of surface velocity and elevation at 4 locations spaced along the glacier; 5) optical surveys of the velocity and uplift of a network of 24 stakes; 5) vertical strain along two boreholes; 6) borehole slug and pump tests; and, 7) direct observation of englacial and subglacial water flow via borehole video camera. The speed-up event consisted of a two stage velocity increase to a maximum level eight times the background deformational velocity. The speed-up was associated a slight drop in subglacial water pressure (not a pressure rise). Numerous anomalous events occurred in conjunction with the accelerated sliding, for example: pressure variations that were unusually uniform over space, surface uplift due to bed separation, and an episode of massive up-welling and refreezing of englacial waters. Our interpretation of the speed-up cycle is that it began with an initial enhanced sliding event that propagated up-glacier as an insatiability within a thin but widespread layer of subglacial water. After a small displacement, stability was regained and sliding temporarily terminated. This initial event, however, initiated connections between the bed and a large volume of englacial water contained within void spaces, which caused widespread basal flooding, bed separation, and resumed sliding. Sliding accelerated until well developed drainage pathways were established, leading to draining of the bed and a flood at the terminus.
http://research.gg.uwyo.edu/joelh/benchglacier
C31B-0326 0800h
Time/Space Migration of Diurnal Cycles Related to Hydrology and Ice Dynamics: Bench Glacier, Alaska
We present a time/space analysis of the diurnal fluctuations in basal water pressure, surface velocity, stream discharge, and melt water input along the length of Bench Glacier, Alaska. Basal water pressure was measured in 43 different boreholes and surface velocities were recorded at four locations using differential GPS. The melt water input was computed using a surface energy balance model calibrated to surface ablation measurements. The diurnal fluctuations in basal water pressure were often large (80% of overburden pressure). The surface velocity of Bench Glacier throughout the summer commonly varied from 0.01 to 0.05m/day on a diurnal cycle. Preliminary analysis suggests that basal water pressure and surface velocity, however, did not demonstrate a simple correlation. The time of peak basal water pressure did not occur synchronously at all boreholes, and tended to be slightly later at boreholes higher on the glacier. The time of peak basal water pressure also migrated steadily throughout the first half of the melt season from near mid-night (0:00-2:00) to late afternoon (16:00-18:00), and remained relatively constant for the remainder of the melt season. This suggests that the hydrology system developed during the first half of the melt season to accommodate the melt water input more efficiently.
C31B-0327 0800h
Clast Plowing as a Potential Flow Instability for Soft-bedded Glaciers
Fast flow of most glaciers is controlled by conditions at the ice-bed interface. When glaciers move over deformable till, clasts at the ice-bed contact provide the roughness that resists glacier motion. These clasts, however, plow through the bed surface when basal water pressure is high. Factors that control plowing resistance, including plowing speed, are therefore important to study. A ring-shear device was used to study the resistive forces on plowing clasts as a function of plowing speed. Hemispheres of different sizes were dragged through water-saturated till at speeds ranging from 15 to 380 m/year. Resistive forces on the hemispheres and pore-water pressure in adjacent till were measured. Tests were conducted with two tills that differed in hydraulic diffusivity by two orders of magnitude. Results indicated that shear resistance decreased dramatically with increased plowing speed. At high plowing speeds, excess pore-water pressures were generated in front of the hemispheres because the rate of till compaction exceeded the rate at which pore pressure could diffuse. The excess pore-water pressures weakened the till in front of plowing clasts and decreased resistive forces. Using measured pore pressures, resistive forces on a plowing hemisphere were estimated successfully with a geotechnical theory of cone penetration. These laboratory results indicate how plowing speed controls resistive forces on plowing clasts. Increases in glacier sliding speed will reduce plowing resistance promoting even faster glacier flow: a potential flow instability for glaciers on soft beds that has not been noted previously.
C31B-0328 0800h
An investigation of subglacial processes with a wireless multi-sensor probe from Briksdalsbreen, Norway
This paper presents the results from eight subglacial wireless probes inserted into the till and ice beneath Briksdalsbreen during summer 2004. This is a new technique in glaciology, to understand glacier processes, and the relationship between glacier dynamics and climate change. These probes were installed in the ice where it was approximately 70m deep and they record temperature, pressure, stress, outside resistivity and tilt. The sensors are held within a small probe which mimics a natural clast. They send their data back by radio communications to a Base Station on the surface of the glacier, which is then sent to the Internet as part of an Environmental Sensor Network. Data was also collected from the surface of the glacier (weather, ice movement, images). Once the probes were installed they began to transmit their data immediately. This is combined with other data (GPR, video camera images and sediment sampling) to investigate the processes occurring beneath a temperate glacier resting on a thin deforming bed.
C31B-0329 0800h
Laboratory Study of Frictional and Rheological Properties of Till From Matanuska Glacier, Alaska
Recent studies show that small-amplitude stress changes can trigger ice sheet motion and subglacial seismicity. Deformation in the subglacial region plays a key role in determining slip behavior, including creep, transient slip, and stick-slip (seismic) motion. However, progress in understanding these phenomena is limited by uncertainty in the rheology and frictional properties of glacial till. We report on detailed laboratory experiments to measure the frictional constitutive properties of till from the Matanuska Glacier. Experiments were conducted in a servo-controlled, double direct shear apparatus with dried till samples sheared at displacement rates of 1 to 300 $\mu$m/s and normal stresses of 0.5 to 5.0 MPa. Till was sheared in a three-block arrangement in which two layers are sandwiched between a central forcing block and two stationary blocks. The nominal frictional contact area is constant during shear. We studied the effect of layer thickness, shear-zone surface roughness, sliding history, normal stress, and slip velocity. In one set of experiments, the frictional contact area was 100 cm$^{2}$ and the forcing blocks were grooved perpendicular to the shear direction with a depth of 0.8 mm and wavelength of 1 mm. A second set of forcing blocks had contact dimensions of 15 cm x 15 cm, and these surfaces had two wavelengths of roughness: grooves of wavelength 1 mm and depth 0.7 mm were superimposed on grooves of wavelength of 10 mm and depth 0.7 mm. All experiments were done at room temperature and humidity. The till had grains ranging from 6.3 mm to finer than .063 mm with a mean of 2.67 mm. Sample thicknesses were 0.7 and 1 cm for the smaller set of blocks and 3 and 6 cm for the larger blocks. In all cases the grain size is less that 1/10 the layer thickness to minimize boundary effects. In the thicker layers, we investigated the role of shear localization by intentionally constraining the shear zone width. Preliminary results at normal stress of 0.5 MPa show stable sliding and give residual values of sliding friction of 0.65 at 1 $\mu$m/s increasing linearly with the log of velocity to a maximum of 0.77 at 300 $\mu$m/s. This trend indicates velocity-strengthening behavior, in which friction increases with sliding velocity. Increasing normal stresses to 1.0 and 5.0 MPa yields values for sliding friction of 0.61 to 0.64 and 0.56 to 0.60, respectively, showing that the degree of velocity-strengthening varies with normal stress. Coulomb-Mohr failure envelopes give an internal friction angle of 31.5 degrees and cohesion of 4.5 kPa. Our results indicate velocity-strengthening and thus imply that shear of the Matanuska glacial till will be dominantly stable under the conditions studied.
C31B-0330 0800h
Subglacial Depositional Processes in the Port Askaig Formation (Neoproterozoic) of Ireland
The Port Askaig Formation was deposited during the Vendian glaciation (c. 650 Ma) and is a range of tillites that outcrop discontinuously from Banffshire (Scotland) to Connemara (Ireland). Sedimentary structures commonly observed include dropstones and sediment drapes, interpreted as deposition from a floating glacial ice shelf in a shallow marginal sea. Other structures, such as intersecting clastic dikes, have been interpreted as evidence for subaerial exposure of the tillite surface. Exposures of the Port Askaig Formation were examined at its Irish type area at Kiltyfanned Lough, County Donegal. Here, homogeneous sandy beds with internal planar bedding structures are separated by laminated fine sand beds which have erosional upper surfaces. The laminated beds are clast-free and individual laminae are laterally continuous and undisturbed. Larger clasts lie bed-parallel and are draped by overlying beds. Occasionally drapes are asymmetric with a thickened sediment prow, suggestive of flow direction. The clastic dikes are polygonal in plan view, may be isolated or interconnected, and are often arranged in parallel sheets which pinch out laterally. Internally, the clastic dikes are infilled with coarse sand to gravel. Infills are often aligned parallel to dike margins. The presence of draped and deformed sediments suggest a subglacial environment with free water availability. The flat-lying morphology of clasts also favours a subglacial rather than a full marine environment. The morphology and disposition of clastic dikes is interpreted as due to subglacial hydrofracturing of a till sheet and upward passage of sediment-charged water through the fracture zone, which is known from late Pleistocene and Precambrian tillites elsewhere. Variations in water availability can be reconciled by a sub-ice shelf depositional model with spatial and temporal changes in tidally-induced ice-bed coupling.
C31B-0331 0800h
Extreme Weichselian Output Rates of Subglacial Till at the Front of the Norwegian Channel Ice Stream
High-resolution seismic and core data have been used to study the North Sea Fan, located on the continental slope at the outlet of the Norwegian Channel, northern North Sea. The North Sea Fan is dominated by glacigenic debris flows sourced from subglacial till brought to the shelf break by the Norwegian Channel Ice Stream, which drained a major part of the southwestern Fennoscandian Ice Sheet. Thickness maps of glacigenic debris flows and evidence from well dated cores show that during the last glacial cycle close to 6000 km3 of glacigenic debris flows were deposited. Evidence from cores along the upper slope and in the Norwegian Channel suggests that the ice stream was not active before ~28 ka or after 15 14C ka BP. This, along with other data constraining the late Weichselian glaciation history of Southern Scandinavia suggests that the Norwegian Channel Ice Stream was in total active between c. 9 and 5 k.y. from ~28 ka to 15 14C ka BP. This activity period implies output rates between 6500 and 11600 m3 per metre of the ice stream front each year. These output rates are higher than what has been calculated from presently active ice streams in Antarctica. Implications for the thickness of the deformable till layer beneath the ice stream, sediment budget and ice stream stability will be discussed.
C31B-0332 0800h
Geomorphic Evidence for Compound Moraines, Ice Streaming, and Surging of the Laurentide Ice Sheet in Northern Wisconsin
Reconstructions of late-Wisconsin glacial events in the Lake Superior region have typically assumed broadly-synchronous lobes of the Laurentide Ice Sheet (LIS). End moraines are often thought to have formed by a single cycle of ice advance and retreat under steady-state flow conditions. However, our geologic mapping in northern Wisconsin during about the last decade shows a much more complex history for southern lobes of the LIS in this region (namely the Superior Lobe, Chippewa Lobe, Wisconsin Valley Lobe, and Langlade Lobe). Landform and stratigraphic evidence shows that terminal moraines produced by some of these lobes are composite features resulting from two or more ice-flow events. Associated with these events were large changes in ice-flow direction (sixty degrees or more) and the development of landforms that indicate the former presence of ice-margin offsets and intralobe shear margins, similar to those described on several modern surging glaciers and ice streams. More specifically, although these LIS lobes were broadly synchronous when advancing southward out of the Lake Superior basin, cross-cutting relationships of end moraines show interlobe variability in the timing of ice advances. Especially in the case of the Chippewa and Wisconsin Valley Lobes, ice-surface-slope and ice-flow indicators (i.e. drumlins) show large changes in ice-flow direction (approximately ninety degrees) during at least two distinct flow events of each lobe. Broad, hummocky, compound, terminal moraines indicate two or more flow events of each lobe into their marginal zones. The terminal zones of all the lobes must have carried a substantial volume of englacial and supraglacial sediment to result in such broad, hummocky moraines (which include many ice-walled-lake plains). In summary, the landforms and landscapes produced by southern lobes of the Laurentide Ice Sheet in northern Wisconsin are broadly consistent with criteria recently proposed for recognizing paleo-ice streams and are also consistent with features observed on modern surging glaciers. Additionally, reconstruction of paleo-ice-flow directions shows asynchronous advances of the LIS lobes in northern Wisconsin along with complex intralobe flow. Such ice-sheet behavior was likely in direct response to changes in the Superior Lobe in the Lake Superior basin; however, multiple factors still need to be evaluated.
C31B-0333 0800h
Longitudinal Deviatoric Stresses in Streaming Flow: results from a simple model
Longitudinal deviatoric stresses are assuming a new importance in the modeling of ice-stream flow and ice-sheet stability. Recent ice-shelf changes have been followed by inland-ice changes in several places, and strong tidal effects are observed propagating up ice streams. New model results and data show the importance of transmission of longitudinal deviatoric stresses across regions of enhanced basal lubrication and feedbacks on the maintenance of that lubrication. Here, we report additional model results on longitudinal-stress effects in ice-stream flow, showing further the importance of these phenomena. We suggest that longitudinal deviatoric stresses should not be neglected when assessing future sea-level rise under warming climate scenarios.
C31B-0334 0800h
Forecast models of Arctic coastal changes during the next 100 years
Predicted global climate warming will significantly effect Arctic coastal dynamics due to the reduction of the sea-ice cover, the amplification of wave processes and the degradation of coastal permafrost. This poster presents scenarios of Arctic coastal changes based on forecast models for the next 100 years. The results indicate that the most essential changes will take place within permafrost environments. Starting in the Norwegian Sea and moving eastward, the results are summarized in the following. In the Norwegian and Barents Seas coastal changes will be minimal. In the Pechora Sea, in the Varanday area, the total sediment flux to the sea from coastal erosion will increase up to 5 million tons per year. In the Kara Sea a longer ice-free period will promote the amplification of wave processes. Thus, within the northern part of Yamal Peninsula the estimated shoreline retreat rates will reach 2.2 meter per year during the first half of the next century and 3.4 m per year during the second half. At the end of the century the average shoreline retreat rates will be 2.5-2.6 times higher than today. In the Laptev Sea the total activity of wave processes will increase by a factor of 1.5-2.0. By the end of the 21st century the estimated rates of thermoabrasive processes for example at Muostakh Island and Shirokostan Peninsula will be 1.25-1.4 times higher than today and average coastal retreat rates will reach 6.4 m per year. Shirokostan's shoreline will retreat 550-600 m and the associated sediment flux will increase up to 18-21 million tons per year, which is comparable to the modern fluvial sediment discharge. In the East Siberian Sea the destruction of the long-term sea-ice cover in the offshore area (100-400 km) could be activated. Thermoabrasive processes are likely to increase by a factor of 4. Using mathematical modeling the increase of thermoabrasion at the Crestovskiy Cape can be predicted as 2.6-4.3 m per year during the first part of the century and 3.4-5.2 during the second part, which is equivalent to a total increase by a factor of 1.4-1.5. In the Chukchi Sea any catastrophic destruction of accumulative bars is not expected but the average wave height in the near shore zone will increase from 1 m to ca. 3 m, corresponding with a twice longer wave period. As a result, the relief of onshore bars will significantly change and new coastal accumulative banks could be formed. This study has been performed under the framework of the Arctic Coastal Dynamics (ACD) project and was supported by INTAS (grant no. 2001-2332).