C23A-0975 1340h
Velocity field of mountain glaciers measured by correlating SPOT5 images
A complete and detailed map of the velocity field on mountain glaciers is derived by cross-correlating SPOT5 optical images. This approach offers an alternative to SAR interferometry and speckle tracking, because no present or planned RADAR mission provides data with a temporal separation short enough to measure the displacements of mountain glaciers. A DEM and an accurate orbital model permit to project one of the SPOT5 images into the geometry of the other one before cross-correlation. The strong ablation occurring during summer time on glaciers requires a correction to obtain unbiased estimates of displacement. We assess the accuracy of our measurement by comparison with nearly simultaneous differential GPS surveys. If the images have similar incidence angles and correlate well, the accuracy is on the order of 0.5 m, or one fifth of the pixel size. Similar results are also obtained without ground control points and with the low resolution GTOPO30 DEM. Monitoring of remote glaciers is thus possible. On the studied glaciers (French Alps and Iceland), velocities are ranging from a few meters per year to over 500 m/year. The high accuracy and the large spatial coverage of our measurements allow monitoring the gglacial acceleration due to summertime surface melting. Our methodology, applied to SPOT5 images, can also be used to measure the deformation of the Earth's surface caused by landslides, earthquakes and volcanos.
C23A-0976 1340h
Rapid thinning of Mont Blanc glaciers (Alps) derived from satellite images
Mountain glaciers are good indicators of climate change: their mass balance is directly connected to the interannual variability of climatic parameters such as temperature and precipitations. Mass balance is also the key parameter to estimate the contribution of glaciers to sea level rise. A first step toward the measurement of glacier mass balance from space is an accurate measurement of glacier elevation changes. In this study, we measured from satellite images the rapid thinning of glaciers in the French Alps. From pairs of satellite images, we compute Digital Elevation Models (DEMs) for the Mont Blanc area in 1979, 1994, 2000, and 2003. To register the DEMs to the 2003 reference DEM, we adjust their longitude, latitude and altitude over motionless areas. Substracting the DEMs yields the thickness change of glaciers. Inconsistent values are removed assuming a log-normal distribution of our measurements. The uncertainty of the thickness change measurement is greatly reduced by averaging over areas covering altitude intervals of 50 m. Comparisons with topographic profiles and a differential DEM from aerial photographs obtained on the Mer de Glace indicate an overall accuracy of 1m. Below 2100 m, satellite DEMs show an evolution of the thinning rate from 1±0.4 m/a (years 1979-1994) to 4.1±1.7 m/a (2000-2003). Our methodology provides a complete view of elevation changes of glaciers during the last 25 years. We discuss how the slope and orientation of glaciers control their shrinkage. We also present a first attempt to derive the cumulated mass balance for the Mer de Glace and Argentiere glaciers over the 1979-2003 period. In the near future, applying our method to satellite optical images with meter or sub-meter resolution could monitor the thickness change of remote glaciers with a precision of the order of 10 cm.
C23A-0977 1340h
Modelling the Re-freezing of Surface Meltwater and the Formation of Superimposed Ice on an Arctic Glacier
A detailed surface energy-mass balance model including an explicit calculation of meltwater refreezing and superimposed ice formation is run with meteorological input from the past 30 years at Midre Lovenbreen - a high Arctic valley glacier, Spitsbergen, Svalbard. The model is then validated through comparison of the results with field measurements and observations. Further comparisons are then drawn between the results of this physically based model and those of simpler, more easily applicable models of refreezing both new and taken from the literature. Model results indicate that superimposed ice alone accounts for 33 % (on average) of the total net accumulation at Midre Lovenbreen under present conditions. Correspondingly the removal of all forms of meltwater refreezing from the mass balance calculation is shown to result in a 68 % increase in the rate of modelled mass loss from this glacier. Net mass balance output from the coupled model is found to be highly sensitive to changes in mean annual air temperature but much less sensitive to changes in annual precipitation. The detailed model presented here closely reproduces the field measurements of superimposed ice. The simpler models, however, vary in their agreement to the field data, with highly schematic approaches found to transfer poorly from the conditions under which they were developed. An additional scheme is therefore presented here based on a simplified thermodynamic approach to calculating the mass of meltwater refreezing at any site. This requires only annual climatic input data, yet is transferable to a range of conditions due to the physical basis of the calculation.
C23A-0978 1340h
Calving of the Brunt Ice Shelf, the Risk of Losing Halley
The British Antarctic Survey base station, Halley is situated on the Brunt Ice Shelf on the eastern edge of the Weddell Sea in West Antarctica. The risk of losing Halley to calving stems mainly from crack extension on the ice shelf or iceberg collision from an advancing ice stream further north, called Stancomb-Wills. The risk of calving will significantly increase as Halley approaches and passes the location of the most recent calving front. Halley is currently only 8km from this calving front and at its present velocity (550m a-1) will reach it in about 14 years. Furthermore, there has been a dramatic slowing down of the ice shelf, from about 750 m a-1 in 1999 to 550m a-1 in 2004, as the ice front wraps itself around a grounded area called McDonald Ice Rumples. The ice shelf is currently slowing at a rate of approximately 40m a-1, the effect of this on the stability of the shelf is as yet unknown. The monitoring programme established through the Lifetime of Halley project provides a means of examining and monitoring the ice shelf movement and fluctuations over both short and long time scales, thus providing a crucial early warning system should a calving event occur. This programme, now in its third year, complements observations already established and improves the overall understanding of the Brunt Ice Shelf in terms of its dynamics, stability and fracture mechanics. The following provides information regarding the behavioural history of the Brunt Ice Shelf, the calving problem and the Lifetime of Halley monitoring programme. Some future work will also be outlined with the ultimate aim of providing an improved understanding of crack propagation.
C23A-0979 1340h
Tracking the Duration and Magnitude of Snow Albedo and Temperature coupling during the Ablation Season
Remote sensing research has in the last two decades led to significant progress in monitoring and measuring certain snow hydrologic processes in particular, estimating the timing of snowmelt run-off can be improved by monitoring when the snowpack is ripe for melt. Diagnostic information about the evolving spring time snow pack can be determined using snow surface optical and thermal properties. Optical and thermal characteristics can only be useful if changes in these parameters are indicative of the overall state of the snowpack. This analysis is interested in investigating duration and magnitude of snow surface albedo and average snow pack temperature through the melt season. Automated snow surface albedo and average snowpack temperature were acquired from the Mammoth Mountain Energy Balance Monitoring Site in the California Sierra Nevada Mountain Range for the melt season of 2002. Incoming and outgoing shortwave radiation measured from 0.285-2.8 m, and average snowpack temperature data collected at this site were downloaded from the Mammoth Mountain Ski Area (MMSA) database website. Surface albedo was calculated using daily incoming and outgoing direct shortwave radiation. Maximum snow accumulation occurred at this site on approximately March 24, 2002. Therefore, analysis of snow surface albedo and average pack temperature coupling were examined from peak accumulation through June 30, 200 when the pack was fully depleted. The difference between daily albedo and average temperature were computed for each day. The first derivative of albedo and temperature difference was computed for the time series and used as an indicator of temporal coupling. The first derivative anomaly time series exhibited strong small perturbations (strong coupling) through the course of the ablation season with significant variation occurring on June 3, 2002 when the snowpack had depleted to approximately 80 cm in depth. After this date, significant decoupling between surface albedo and snowpack temperature occurs.
C23A-0980 1340h
Rheology of the Ronne Ice Shelf, Antarctica, inferred from satellite radar interferometry data using an inverse control method.
Ice shelves floating around the Antarctic Ice Sheet spread under their own weight into the ocean. The ice flow is controlled by the rigidity of ice, and a delicate interaction with bottom and surface accumulation. Rigidity ( or flow law parameter $B$ ) depends mainly on temperature [{\em Paterson, 1994}], and fabrics. This study presents an inverse control method developped to infer $B$ on ice shelves. The method is based on finding the best fit with observations of ice velocity from satellite radar interferometry. The model was tested on the Ronne Ice Shelf, and the results show flow law parameter $B$ varying between 300 kPa a$^{1/3}$ and 900 kPa a$^{1/3}$. Minimums appear along the ice margins which could be due to ice softening (viscous heating). High values are found in the wake of large glaciers which advect large quantities of cold ice. Some areas near the grounding lines experience basal melting, which increases rigidity. Melting near the icefront corresponds to areas of decreased rigidity. This method allows the modeller to account for variations in the distribution of $B$ in ice flow models. We thank the California Institude of Technology for making this study possible.
C23A-0981 1340h
East Asian Monsoon variations directly recorded in the Greenlandic North-GRIP ice core
Ice cores from central Greenland are the temporally best resolved records of northern hemisphere climate of the last glacial period. Here we show that during the last glacial period (MIS 4 to MIS 2) East Asian Monsoon variations are directly recorded in the North-GRIP ice core (north-central Greenland). The East Asian Winter Monsoon (EAWM) strength as inferred from Chinese loess deposits correlates well with the concentration and size distribution of insoluble mineral particles in the North-GRIP core, and we infer that the ice core microparticle record can be taken as a proxy for EAWM. Further, variations of the East Asian Summer Monsoon (EASM) strength are coincident with changes of the Ca2+/particle ratio at North-GRIP. In light of these findings and taking d18O as a proxy for North Atlantic temperature we investigate the rapid climatic transitions during the last glacial period (Dansgaard-Oeschger events). We discuss differences between the changes of EAWM, EASM, and North Atlantic temperature regarding their timing at these rapid transitions; this can be done with unforeseen accuracy because all proxies are from the same archive and do not underlie relative dating uncertainties. Furthermore we find that before 75ka BP (i.e. before the MIS 4/5-transition) atmospheric circulation patterns in the northern hemisphere as well as teleconnections between North Atlantic circulation and the East Asian Monsoon system must have been substantially different. They underwent considerable reorganisation, which may be linked to the build up of large ice sheets in the northern hemisphere during this time.
C23A-0982 1340h
Continuous Measurements of Air Content: First Results from the EDML Ice Core, Antarctica
The air content in cold glacier ice depends on the pore space volume and on local air pressure and temperature at the time of pore close-off. A new method to measure the air content of glacier ice has been developed and will be presented. The measurement has been integrated into the Continuous Flow Analysis set-up which is used for chemical profiling of the EPICA ice cores. It allows the continuous determination of the air content at high depth resolution (<1cm). The air content has been measured continuously in the deep ice core from Dronning Maud Land, Antarctica from 700 m to 2450 m depth, which corresponds to approx. 10 ka to 150 ka before present. First results indicate that the smoothed profile is not dominated by the 100 ka cycle but instead by the 40 ka insolation cycle. The high resolution record exhibits considerable variability of the air content even on the 1 cm scale. These variations may partly be seasonal, and we will discuss the possible importance of different causalities that may influence the air content of glacier ice. The small scale variations of air content may suggest heterogeneous pore close-off and eventually may aid its reconstruction.
C23A-0983 1340h
Analysis of near-zero temperature anomalies in high latitude regions
Daily temperatures follow quasi-sinusoidal time-series over annual periods. It is expected that the daily temperature frequencies form bimodal distributions corresponding to the winter and summer temperature modes. Unlike the expectation, analysis of high latitude temperature datasets revealed tri-modal temperature frequencies in many areas. The third mode falls into the near-zero temperature range and it is most likely a buffer resulting from the latent heat of phase transition of water during snow melt. There is no statistically considerable change in the slope on the daily time-series curves around near-zero temperatures. We have analyzed daily climatology modes for individual days during the melt and freeze periods and found bimodal distributions for those days which are unique relative to all other days during the year. In these cases, one of the modes is linked to the zero temperature mark. A statistical algorithm was developed that allows for the identification and quantification of the influence of water phase transition on the daily temperature frequencies that cause these anomalies. Mapping of the sub-zero temperature anomalies over the Pan-Arctic domain showed that they tend to be in non-mountainous regions under continental climate conditions. Milder near-ocean climates and mountainous regions do not show these near-zero anomalies. Southern parts of the pan-Arctic are also affected by the near-zero anomalies in such a way that the warmer winter mode is eroded and skewed toward the anomaly.
C23A-0984 1340h
Analysis of multi scale radiometric data acquired during the NASA Cold Land Process Experiment
One of the main goals of the NASA Cold Land Processes Experiment is to explore scaling issues associated with microwave remote sensing of snow-packs. During the experiment, microwave radiometer observations were made at scales from the plot scale up to satellite pixels (25x25 Km2). In this paper, brightness temperatures acquired by the University of Tokyo's Ground Based Microwave Radiometer-7 (GBMR-7), the NOAA Polarimetric Scanning Radiometer (PSR/A), the SSM/I and AMSR-E satellite radiometers during the third Intensive Observation Period (IOP3, dry snow, February 19 - 25, 2003) have been analyzed and compared. Differences among observations are discussed. The effects of topography, as well as different antenna footprint positions, observation times and calibrations are considered and used to understand the differences observed at the different scales. In addition, inter-satellite differences were observed.
C23A-0985 1340h
Automated Laser-Light Scattering measurements of Impurities, Bubbles, and Imperfections in Ice Cores
Laser- light scattering (LLS) on polar ice, or on polar ice meltwater, is an accepted method for measuring the concentration of water insoluble aerosol deposits (dust) in the ice. LLS on polar ice can also be used to measure water soluble aerosols, as well as imperfections (air bubbles and cavities) in the ice. LLS was originally proposed by Hammer (1977a, b) as a method for measuring the dust concentration in polar ice meltwater. Ram et al. (1995) later advanced the method and applied it to solid ice, measuring the dust concentration profile along the deep, bubble-free sections of the Greenland Ice Sheet Projetct 2 (GISP2) ice core (Ram et al., 1995, 2000) from central Greenland. In this paper, we will put previous empirical findings (Ram et al., 1995, 2000) on a theoretical footing, and extend the usability of LLS on ice into the realm of the non-transparent, bubbly polar ice. For LLS on clear, bubble-free polar ice, we studied numerically the scattering of light by soluble and insoluble (dust) aerosol particles embedded in the ice to complement previous experimental studies (Ram et al., 2000). For air bubbles in polar ice, we calculated the effects of multiple light scattering using Mie theory and Monte Carlo simulations, and found a method for determining the bubble number size and concentration using LLS on bubbly ice. We also demonstrated that LLS can be used on bubbly ice to measure annual layers rapidly in an objective manner. Hammer, C. U. (1977a), Dating of Greenland ice cores by microparticle concentration analyses., in International Symposium on Isotopes and Impurities in Snow and Ice, pp. 297-301, IAHS publ. no. 118. Hammer, C. U. (1977b), Dust studies on Greenland ice cores, in International Symposium on Isotopes and Impurities in Snow and Ice, pp. 365-370, IAHS publ. no. 118. Ram, M., M. Illing, P. Weber, G. Koenig, and M. Kaplan (1995), Polar ice stratigraphy from laser-light scattering: Scattering from ice, Geophys. Res. Lett., 22(24), 3525-3527. Ram, M., J. Donarummo, M. R. Stolz, and G. Koenig (2000), Calibration of laser-light scattering measurements of dust concentration for Wisconsinan GISP2 ice using instrumental neutron activation analysis of aluminum: Results and discussion, J. Geophys. Res., 105(D20), 24,731--24,738.
C23A-0986 1340h
Subsurface ice properties, genesis, and preservation in the Dry Valleys, Antarctica
The presence of subsurface ice is an outstanding problem in the Antarctic Dry Valleys. Current models indicate that ice in soils should sublimate to several meters below the surface within a few thousand years. However, ice is commonly found within the top decimeters in soils that are up to millions of years old. Subsurface ice is important because it is a fundamental component of diverse landscape processes including patterned ground formation (surface renewal, micro-relief generation), downslope soil motion, and pedogenesis. Ice-cemented soils are subject to contraction-expansion as they cool and warm and, as a result they tend to form distinctive polygonal patterns delineated by troughs over thermal contraction cracks. Furthermore, subsurface ice may provide an archive of paleoenvironmental climate, and it can be an important source of water for biological, as well as geomorphic processes. We examine the various forms, characteristics, and mechanisms of the formation of subsurface ice in the Dry Valleys. Data on stable isotope, particulate debris, and ions in the various forms of ice are shown for the various types of ground ice, which include interstitial ice in soils, buried glacial ice, and according recent findings, segregation ice at the lower elevations. Based on vertical profiles of salts and of stable isotope in ice, we have developed a model that explains formation and preservation of ground ice for long periods of time. This is of particular interest in light of previous sublimation models that indicate that the ground ice is not stable, and should disappear relatively quickly due to sublimation. Lessons learned from the Dry Valleys have direct implication for studies of ground ice on Mars.
C23A-0987 1340h
Distributed Energy-Balance Models at the Surface of the Taylor Glacier, Antarctica
As part of the effort to understand the climate history of Antarctica and its possible responses to anthropogenic climate change, our research group has undertaken a measurement and modeling effort on the Taylor Glacier, Antarctica. The Taylor Glacier has a unique potential for informing the way we think about how Antarctic glaciers have responded to past climate changes for the following reasons: an ice core was drilled at the source of the glacier, at Taylor Dome, in the mid 1990's; the glacier terminates on land so we have information about past glacier terminus positions from a series of moraines; and biology in the Taylor Valley depends on melt water and nutrients from the glacier. Among other field activities, we set up 6 permanent weather stations on the Taylor Glacier in the 2003-4 field season and made ablation measurements at about 250 stakes. We present results from the first year's weather station data. The stations are transmitting the data back to us via the ARGOS satellite system. The weather data has been incorporated into a few previously-published energy-balance models (e.g. the AMUNDSEN model and the Excel-based model by Brock and Arnold) that have been modified to account for the polar nature of the Taylor Glacier. We will present results from these models along with a new model that uses wind speed data at 2 levels and temperature and relative humidity at 1 level to predict ablation. Maps of predicted ablation are compared with measurements from ablation stakes. Animations of the weather station data and relevant AVHRR satellite imagery are presented and are also available on the web at the URL below.
http://geography.berkeley.edu/~abliss/
C23A-0988 1340h
An Integrated Geoscience Assessment of the Influence of Geologic Processes and Climate on the Permafrost Environment of the Mackenzie Delta, N.W.T.
The Mackenzie River delta of Canada's Northwest Territories is characterized by some of the most complex permafrost conditions in the Arctic. Within a 400 square kilometre area of the outer delta, permafrost varies from less than 80m to over 600m in thickness. This variability in both permafrost thickness and ground temperature conditions is dictated by significant variations in the history of surface temperatures caused by the interplay between climatic effects and variations in large-scale geologic processes. During the past 100,000 years the geothermal regime of this area has been influenced by glaciation, long periods of sub-aerial exposure, periods of marine transgression/regression, fluvial deposition, coastal erosion, and lake development and drainage. This paper will introduce an integrated geoscience approach to the assessment of permafrost and related geotechnical considerations in the Mackenzie Delta. A key objective of the research was establish constraints to the late Quaternary history of the area by utilizing shallow core samples collected from lacustrine, deltaic and shallow marine environments. The current variability in annual temperature regime of aquatic environments was assessed through the deployment of small ground temperature loggers in lakes, ponds and river channels. Shallow ground temperatures were also available from a variety of terrestrial observation sites. An assessment of the paleo-temperature regime was undertaken through terrain mapping and coring of contemporary lake basins to constrain the history of lake formation and drainage. These data served as a basis for geothermal modeling to determine the theoretical distribution of permafrost in this area.
C23A-0989 1340h
Sedimentology and Permafrost Characteristics of Pingo-Like Features (PLFs) from the Beaufort Sea shelf, NWT, Canada
Pingo-like features (PLFs) are rounded positive relief features commonly found on Beaufort Sea shelf, NWT. PLFs occur in water depths from 20 to 200m, are typically a few hundred meters in diameter and rise 10 to 35m above the seafloor. In the fall of 2003, an MBARI-USGS-GSC-DFO coring and geophysical study was undertaken of a number of PLFs. The crests, flanks and moats of 8 PLFs, as well as background shelf sites, were vibra-cored. Upon recovery, core temperatures of moat sediments ranged from 2.0 to -0.5 deg C and no ice bonding was observed. Sediments consisted of dark-olive-grey to black muds with shells. Sedimentary structures were rare with some finely laminated to finely-color-banded beds. Intense bioturbation, in situ marine shells and a lack of terriginous macrofossils suggest moat sediments were deposited in a shallow coastal environment. In some instances, a down core grain size coarsening was observed with higher organic content suggesting a gradational environment towards more lagoonal conditions. Core temperatures from the 8 PLFs were 0 to -1.7 deg C, significantly colder than the moat sediments. Ice-bonded permafrost was encountered within 1m of the seabed with visible ice content up to 40% by volume. Several ice-bonded intervals were preserved frozen for detailed investigation in the lab. The observed ground ice in the cores was quite unique when compared with visible ice forms commonly seen in regional terrestrial sections. The ice gave the core a vuggy texture with individual ice-filled vugs 10 to 200 mm3. Vugs were typically flattened to ovoid. When thawed, the ice produced excess water resulting in a very soft texture. In many cases the vuggy texture was maintained with sediment voids forming where the ice was. PLF crest sediments were massive silty clays with clayey silts and muddy fine sand interbeds. They generally lack sedimentary structures, although this may have been due to sediment structure loss upon thawing. The background seafloor sediments consisted of unfrozen, massive silty sands and sandy silts and were distinct from the crest and moat sediments. In several cores, a sharp transition was noted to well-sorted sands. This lower unit may represent a transgressed terrestrial sequence. Research continues to determine the origin of the PLFs and quantify the role of permafrost and ice formation.