Search

Nitrate in Antarctic snow has seasonal cycles in its nitrogen and oxygen isotopic ratios that reflect its sources and atmospheric formation processes, and as a result, nitrate archived in Antarctic ice should have great potential to record atmospheric chemistry changes over thousands of years. However, sunlight that strikes the snow surface results in photolytic nitrate loss and isotopic fractionation that can completely obscure the nitrate’s original isotopic values. To gain insight into how photolysis overwrites the seasonal atmospheric cycles, we collected 244 snow samples along a 850 km transect of East Antarctica during the 2013–2014 CHICTABA traverse. The CHICTABA route’s limited elevation change, consistent distance between the coast and the high interior plateau, and intermediate accumulation rates offered a gentle environmental gradient ideal for studying the competing pre- and post-depositional influences on archived nitrate isotopes. We find that nitrate isotopes in snow along the transect are indeed notably modified by photolysis after deposition, and drier sites have more intense photolytic impacts. Still, an imprint of the original seasonal cycles of atmospheric nitrate isotopes is still present in the top 1–2 m of the snowpack and likely preserved through archiving in glacial ice at these sites. Despite this preservation, reconstructing past atmospheric values from archived nitrate along CHICTABA and in similar transitional regions remains a difficult challenge without having an independent proxy for photolytic loss to correct for post-depositional isotopic changes. Nevertheless, nitrate isotopes should function as a proxy for snow accumulation rate in such regions if multiple years of deposition are aggregated to remove the seasonal cycles, and this application can prove highly valuable in its own right.

The temperature of the earth is one of the most important climate parameters. Proxy records of past climate changes, in particular temperature, are a fundamental tool for exploring internal climate processes and natural climate forcings. Despite the excellent information provided by ice core records in Antarctica, the temperature variability of the past 2000 years is difficult to evaluate from the low accumulation sites in the Antarctic continent interior. Here we present the results from the Aurora Basin North (ABN) ice core (71° S, 111° E, 2690 m a.s.l.) in the lower part of the East Antarctic plateau where accumulation is substantially higher than other ice core drilling sites on the plateau, and provide unprecedented insight in East Antarctic past temperature variability. We reconstructed the temperature of the last 2000 years using two independent methods: the widely used water stable isotopes (δ18O), and by inverse modelling of borehole temperature and past temperature gradients estimated from the inert gas stable isotopes (δ40Ar and δ15N). This second reconstruction is based on three independent measurement types: borehole temperature, firn thickness, and firn temperature gradient. The δ18O temperature reconstruction supports stable temperature conditions within 1 °C over the past 2000 years, in agreement with other ice core δ18O records in the region. However, the gas and borehole temperature reconstruction suggest that surface conditions 2 °C cooler than average prevailed in the 1000–1400 CE period, and support a 20th century warming of 1 °C. These changes are remarkably consistent with reconstructed Southern Annular Mode (SAM) variability, as it shows colder temperatures during the positive phase of the SAM in the beginning of the last millennium, with rapidly increasing temperature as the SAM changes to the negative phase. The transition to a negative SAM phase after 1400 CE is however not accompanied by a warming in West Antarctica, which suggests an influence of Pacific South American modes, inducing a cooling in West Antarctica while ABN is warming after this time. A precipitation hiatus during cold periods could explain why water isotope temperature reconstruction underestimates the temperature changes. Both reconstructions arguably record climate in their own way, with a focus on atmospheric and hydrologic cycles for water isotopes, as opposed to surface temperature for gases isotopes and borehole. This study demonstrates the importance of using a variety of sources for comprehensive paleoclimate reconstructions.

In coastal Antarctica, outlet glaciers exhibit complex dynamics materialized by intense internal deformation, enhanced basal sliding and strong thermo-mechanical interactions with the ocean. Here we aim to use seismic observations to unravel these various processes and their link with glacier and ocean dynamics. As part of the SEIS-ADELICE project (2020-2024) supported by the French Polar Institute IPEV, in January 2022 we deployed four permanent and six temporary (1 month long) broadband seismic stations on and around the Astrolabe Glacier (Terre Adélie, East Antarctica), as well as four ocean-bottom seismometers at sea near the terminus of the floating tongue. In January 2023 we will be supplementing this setup by a temporary network of 50 seismic nodes above the grounding line of the glacier. Preliminary detection and classification of seismic events reveals a wide variety of cryo-seismic signals. The most pervasive events correspond to icequakes, are located close to the surface, and exhibit clear tidal modulation. We interpret these events as being generated by the brittle fracturing of ice associated with crevasse opening. We also observe numerous short and similar repetitive events of much lower amplitude that are located at few restricted locations near the ice-bedrock interface. These events are likely produced by basal stick-slip over punctual bedrock asperities. Finally, we observe glacial tremors which could result from hydraulic sources at the ice-bedrock interface, although further analysis is required to confirm this hypothesis. This preliminary work provides useful grounds for deeper analysis to be done in the future on source characteristics and their more quantitative links with glacier dynamics.

Better understanding the behaviour of tidewater outlet glaciers fringing marine ice sheets is of paramount importance to simulate Antarctica‘s future response to global warming. Addressing the processes underlying these glaciers dynamics (ice motion, crack propagation, basal melting, sea ice interaction, calving events, etc) is a mean of constraining their ice discharge to the sea and hence the ice sheet global mass balance. We here focus on the Astrolable glacier located in Terre Adélie (140°E, 67°S) near the Dumont d'Urville French research station. In January 2019, a large crack of around 3km length was observed in the western shore of the glacier potentially leading to a calving of ca. 28 km2.The fissure has progressively grown until November 2021 when an iceberg of 20km2 was eventually released. The location of the glacier outlet at the proximity of the Dumont D’Urville French research station is an asset to collect in-situ observations such as GNSS surveys and seismic monitoring. Satellite optical imagery also provides numerous acquisitions from the early nineties till the end of 2021 thanks to the Landsat and Sentinel-2 missions. We used two monitoring techniques: optical remote sensing and seismology to analyze changes in the activity of the glacier outlet. We computed the displacement of the ice surface with MPIC-OPT-ICE service available on the ESA Geohazards Exploitation Platform (GEP) and derived the velocity and strain rates from the archive of multispectral Sentinel-2 imagery from 2017 to the end of 2021. The images of the Landsat mission are used to map the limit of the ice front in order to retrieve the calving cycle of the Astrolabe. We observe that the ice front had significantly advanced toward the sea (4 km) since September 2016 and such an extension has not been observed in the previous years (since 2006) despite minor calving episodes. The joint analysis of the seismological data and the velocity and strain maps are discussed with the recent evolution of the glacier outlet. The strain maps show complex patterns of extension and compression areas. The number of calving events detected in the seismological dataset significantly increased during 2016-2021 in comparison with the period 2012-2016. Since the beginning of 2021, both datasets show an acceleration. The number of calving events increased exponentially from June 2021 until the rupture in November 2021 and the velocity of the ice surface accelerated from 1 m.day-1 to 4 m.day-1 in the part of the glacier that detached afterward. This calving event is the first one of this magnitude ever documented over the Astrolabe glacier.

The response of very-high-elevation glaciated areas on Mont Blanc to climate change has been analysed using observations and numerical modelling over the last 2 decades. Unlike the changes at low elevations, we observe very low glacier thickness changes, of about −2.6 m on average since 1993. The slight changes in horizontal ice flow velocities and submergence velocities suggest a decrease of about 10 % in ice flux and surface mass balance. This is due to less snow accumulation and is consistent with the precipitation decrease observed in meteorological data. Conversely, measurements performed in deep boreholes since 1994 reveal strong changes in englacial temperature reaching a 1.5 ∘C increase at a depth of 50 m. We conclude that at such very high elevations, current changes in climate do not lead to visible changes in glacier thickness but cause invisible changes within the glacier in terms of englacial temperatures. Our analysis from numerical modelling shows that glacier near-surface temperature warming is enhanced by increasing melt frequency at high elevations although the impact on surface mass balance is low. This results in a non-linear response of englacial temperature to currently rising air temperatures. In addition, borehole temperature inversion including a new dataset confirms previous findings of similar air temperature changes at high and low elevations in the Alps.

The Sentinel Application Platform (SNAP) architecture facilitates Earth Observation data processing (http://step.esa.int/main/toolboxes/snap/). In this work we present results from a new Snow Processor for SNAP. We also describe physical principles behind the developed snow property retrieval technique based on the analysis of Ocean and Land Colour Instrument (OLCI) onboard Sentinel-3A/B measurements over clean and polluted snow fields. Using OLCI spectral reflectance measurements in the range 400-1020nm, we derive important snow properties such as spectral and broadband albedo, snow specific surface area, snow extent and grain size on the spatial grid of 300m. The algorithm also incorporates cloud screening and atmospheric correction procedures over snow surfaces. We present validation results using ground measurements from Antarctica, the Greenland ice sheet and the French Alps. We find the spectral albedo retrieved with accuracy of better than 3% on average, making our retrievals sufficient for a variety of applications. Broadband albedo is retrieved with the average accuracy of about 5% over snow. Therefore, the uncertainties of satellite retrievals are close to experimental errors of ground measurements. The retrieved surface grain size shows good agreement with ground observations. Snow specific surface area observations are also consistent with our OLCI retrievals. We present snow albedo and grain size mapping over the inland ice sheet of Greenland for areas including dry snow, melted/melting snow and impurity rich bare ice. The algorithm can be applied to OLCI Sentinel-3 measurements providing an opportunity for creation of long – term snow property records essential for climate monitoring and data assimilation studies - especially in the Arctic region, where we face rapid environmental changes including reduction of snow/ice extent and, therefore, planetary albedo.

The response of very high elevation glaciated areas on Mont Blanc to climate change has been analyzed using observations and numerical modeling. Unlike the changes at low elevations, we observe very low glacier thickness changes, of about −2.6 m on the average since 1993. The slight changes in horizontal ice flow velocities and submergence velocities suggest a decrease of about 10 % in ice flux and surface mass balance. This is due to snow accumulation changes and is consistent with the precipitation decrease observed in meteorological data. Conversely, measurements performed in deep boreholes since 1994 reveal strong changes in englacial temperature reaching 1.5 °C at a depth of 50 m. We conclude that at such very high elevations, current changes in climate do not lead to visible changes in glacier thickness but cause invisible changes within the glacier in terms of englacial temperatures. Our analysis from numerical modeling shows that glacier near-surface temperature warming is enhanced by increasing melt-frequency at high elevations although the impact on surface mass balance is low. This results in a non-linear response of englacial temperature to currently rising air temperatures. In addition, borehole temperature inversion including a new dataset confirms previous findings of similar air temperature changes at high and low elevations in the Alps.

Results from ground-penetrating radar (GPR) measurements and shallow ice cores carried out during a scientific traverse between Dome Concordia (DC) and Vostok stations are presented in order to infer both spatial and temporal characteristics of snow accumulation over the East Antarctic Plateau. Spatially continuous accumulation rates along the traverse are computed from the identification of three equally spaced radar reflections spanning about the last 600 years. Accurate dating of these internal reflection horizons (IRHs) is obtained from a depth–age relationship derived from volcanic horizons and bomb testing fallouts on a DC ice core and shows a very good consistency when tested against extra ice cores drilled along the radar profile. Accumulation rates are then inferred by accounting for density profiles down to each IRH. For the latter purpose, a careful error analysis showed that using a single and more accurate density profile along a DC core provided more reliable results than trying to include the potential spatial variability in density from extra (but less accurate) ice cores distributed along the profile. The most striking feature is an accumulation pattern that remains constant through time with persistent gradients such as a marked decrease from 26 mm w.e. yr−1 at DC to 20 mm w.e. yr−1 at the south-west end of the profile over the last 234 years on average (with a similar decrease from 25 to 19 mm w.e. yr−1 over the last 592 years). As for the time dependency, despite an overall consistency with similar measurements carried out along the main East Antarctic divides, interpreting possible trends remains difficult. Indeed, error bars in our measurements are still too large to unambiguously infer an apparent time increase in accumulation rate. For the proposed absolute values, maximum margins of error are in the range 4 mm w.e. yr−1 (last 234 years) to 2 mm w.e. yr−1 (last 592 years), a decrease with depth mainly resulting from the time-averaging when computing accumulation rates.

International audience; The isotopic compositions of oxygen and hydrogen in ice cores are invaluable tools for the reconstruction of past climate variations. Used alone, they give insights into the variations of the local temperature, whereas taken together they can provide information on the climatic conditions at the point of origin of the moisture. However, recent analyses of snow from shallow pits indicate that the climatic signal can become erased in very low accumulation regions, due to local processes of snow reworking. The signal-to-noise ratio decreases and the climatic signal can then only be retrieved using stacks of several snow pits. Obviously, the signal is not completely lost at this stage, otherwise it would be impossible to extract valuable climate information from ice cores as has been done, for instance, for the last glaciation. To better understand how the climatic signal is passed from the precipitation to the snow, we present here results from varied snow samples from East Antarctica. First, we look at the relationship between isotopes and temperature from a geographical point of view, using results from three traverses across Antarctica, to see how the relationship is built up through the distillation process. We also take advantage of these measures to see how second-order parameters (d-excess and 17 O-excess) are related to δ 18 O and how they are controlled. d-excess increases in the interior of the continent (i.e., when δ 18 O decreases), due to the distillation process, whereas 17 O-excess decreases in remote areas, due to kinetic fractionation at low temperature. In both cases, these changes are associated with the loss of original information regarding the source. Then, we look at the same relationships in precipitation samples collected over 1 year at Dome C and Vos-tok, as well as in surface snow at Dome C. We note that the slope of the δ 18 O vs. temperature (T) relationship decreases in these samples compared to those from the traverses, and thus caution is advocated when using spatial slopes for past Published by Copernicus Publications on behalf of the European Geosciences Union. 838 A. Touzeau et al.: Acquisition of isotopic composition for surface snow in East Antarctica climate reconstruction. The second-order parameters behave in the same way in the precipitation as in the surface snow from traverses, indicating that similar processes are active and that their interpretation in terms of source climatic parameters is strongly complicated by local temperature effects in East Antarctica. Finally we check if the same relationships between δ 18 O and second-order parameters are also found in the snow from four snow pits. While the d-excess remains opposed to δ 18 O in most snow pits, the 17 O-excess is no longer positively correlated to δ 18 O and even shows anti-correlation to δ 18 O at Vostok. This may be due to a strato-spheric influence at this site and/or to post-deposition processes .

Using the finite-element code Elmer, we show that the full Stokes modeling of the ice-sheet/ice-shelf transition we propose can give consistent predictions of grounding-line migration. Like other marine ice-sheet models our approach is highly sensitive to the chosen mesh resolution. However, with a grid size down to

SPOT and Pleiades images acquired since 2002 are used to describe velocity and elevation changes on Astrolabe Glacier, East Antarctica. Multi-temporal pairs of images are used to generate velocity fields by automatically tracking surfaces features. Stereo-pairs are used to create DEMs. Using three DEMs (2003, 2007 and 2013), we describe a surprising surface elevation increase since 2002, that reached a mean rate of 1.8 m/yr between 2003 and 2007. Conversely, the velocity fields did not reveal any major change in velocity so the origin of the strong increase in elevation remains non elucidated

Several different approaches of various complexities have been used in glacier and ice sheet modelling studies. Amongst them, owing to its simplicity, the Shallow Ice Approximation appears to be the most widely adopted method. This approach, essentially used for ice sheets, owes its success to the shallow aspect of the modelled ice mass embodied in an aspect ratio ζ. When considering smaller ice bodies like alpine-type glaciers, the question arises as to whether the SIA is still valid, given that the method is all the more accurate as ζ is small. In order to test the domain of applicability of the method, results of a SIA finite difference model are compared to those of a finite element model in which the flow equations are fully considered. From a set of two-dimensional flow tests, it is shown that the accuracy of the method is much more deteriorated with increasing bedrock slopes than it is with increasing accumulation rates, even if higher accumulations lead to thicker glaciers with a larger ζ. This leads to the conclusion that when slopes become pronounced, it is a bedrock-related aspect ratio that becomes of relevance such that the bedrock slope should be the most important parameter to consider for assessing the validity of the SIA Method. A 3-dimensional simulation shows that longitudinal shear stresses explain a large part of the misfit between SIA and full-Stokes approaches. To cite this article: E. Le Meur et al., C. R. Physique 5 (2004).

The acoustic impedance of the subglacial material beneath 7.2 km profiles on four ice streams in Antarctica has been measured using a seismic technique. The ice streams span a wide range of dynamic conditions with flow rates of 35–464 m a–1. The acoustic impedance indicates that poorly lithified or dilated sedimentary material is ubiquitous beneath these ice streams. Meanacoustic impedance across each profile correlates well with basal shear stress and the slipperiness of the bed, indicating that acoustic impedance is a good diagnostic not only for the porosity of the subglacial material, but also for its dynamic state (deforming or non-deforming). Beneath two of the ice streams, lodged (non-deforming) and dilated (deforming) sediment coexist but their distribution is not obviously controlled by basal topography or ice thickness. Their distribution may be controlled by complex material properties or the deformation history. Beneath Rutford Ice Stream, lodged and dilated sediment coexist and are distributed in broad bands several kilometres wide, whileon Talutis Inlet there is considerable variability over much shorter distances; this may reflect differences in the mechanism of drainage beneath the ice streams. The material beneath the slow-moving Carlson Inlet is probably lodged but unlithified sediment; this is consistent with the hypothesis that Carlson Inlet was once a fast-flowing ice stream but is now in a stagnant phase, which could possibly be revivedby raised basal water content. The entire bed beneath fast-flowing Evans Ice Stream is dilated sediment.

Satellite altimetric time series allow high-precision monitoring of ice-sheet mass balance. Understanding elevation changes in these regions is important because outlet glaciers along ice-sheet margins are critical in controlling flow of inland ice. Here we discuss a new airborne altimetry dataset collected as part of the ICECAP (International Collaborative Exploration of the Cryosphere by Airborne Profiling) project over East Antarctica. Using the ALAMO (Airborne Laser Altimeter with Mapping Optics) system of a scanning photon-counting lidar combined with a laser altimeter, we extend the 2003–09 surface elevation record of NASA’s ICESat satellite, by determining cross-track slope and thus independently correcting for ICESat’s cross-track pointing errors. In areas of high slope, cross-track errors result in measured elevation change that combines surface slope and the actual Δz/Δt signal. Slope corrections are particularly important in coastal ice streams, which often exhibit both rapidly changing elevations and high surface slopes. As a test case (assuming that surface slopes do not change significantly) we observe a lack of ice dynamic change at Cook Ice Shelf, while significant thinning occurred at Totten and Denman Glaciers during 2003–09.

The bedrock isostatic response exerts a strong control on ice-sheet dynamics and is therefore always taken into account in ice-sheet models. This paper reviews the various methods normally used in the ice-sheet modelling community to deal with the bedrock response and compares these with a more sophisticated full-Earth model. Each of these bedrock treatments, five in total, is coupled with a three-dimensional thermomechanical ice-sheet model under the same forcing conditions to simulate the Antarctic ice sheet during the last glacial cycle. The outputs of the simulations are compared on the basis of the time-dependent behaviour for the total ice volume and the mean bedrock elevation during the cycle and of the present rate of uplift over Antarctica. This comparison confirms the necessity of accounting for the elastic bending of the lithosphere in order to yield realistic bedrock patterns. It furthermore demonstrates the deficiencies inherent to the diffusion equation in modelling the complex deformation within the mantle. Nevertheless, when characteristic parameters are varied within their range of uncertainty, differences within one single method are often of the same order as those between the various methods. This overview finally attempts to point out the main advantages and drawbacks of each of these methods and to determine which one is most appropriate depending on the specific modelling requirements.

[1] Temperatures were measured in two deep boreholes drilled at the same location in the ice at Col du Dome (4250 m) in 1994 and 2005, providing clear evidence of atmospheric warming. The 1994 temperature profile was already far from steady state conditions. Results from a heat transfer model reveal that the englacial temperature increase cannot be explained solely by atmospheric temperature rise. The latent heat produced by the refreezing of surface meltwater below the surface also contributes to the englacial temperature increase. Although surface melting is normally very low at this altitude, this contribution became significant after 1980 for temperatures at the top of the borehole. Simulations for different climatic scenarios show that glaciated areas located between 3500 and 4250 m could become temperate in the future. This warming could have a major impact on the stability of hanging glaciers frozen to their beds if the melting point is reached.

International audience; This paper reports on glacier variations in two mountainous regions of the world, the Alps and the tropical Andes. Available records of snout position and glacier mass balance are compared and interpreted on a climatological basis. In both regions, there is a long-term decreasing trend over the 20th century. The yield of this trend is different from one glacier to the other, depending on geographic and geometric characteristics. Analysing the surface energy balance, net all wave radiation is the main energy flux at the glacier surface. The turbulent fluxes represent an important term with strong positive sensible heat flux in the Alps and strong negative latent heat flux (sublimation) in the Andes. Tropical glaciers are sensitive to inter-annual variations in solid precipitation that affects the albedo, whereas Alpine glaciers are strongly influenced by air temperature changes in the Alps.

International audience; The causes and timing of the Little Ice Age (fifteenth to nineteenth centuries) are still unclear (Crowley, 2000; Bond et al., 2001; Shindell et al., 2001). During the last part of this event (1760–1830), the advance of glaciers in the Alps conflicts with the summer temperature signal (Intergovernmental Panel on Climate Change, 2001). This paper attempts to solve this paradox. From glacier fluctuations and monthly temperature data, we show that mean winter precipitation was higher by at least 25% during this final phase compared to the twentieth century average and that glacier recession after 1830 clearly resulted from a winter precipitation decrease and not a temperature increase. Conversely, since the beginning of the twentieth century, glacier changes have been driven mainly by temperature change.

A two-dimensional ice-flow model based on the shallow-ice approximation (SIA) is used to investigate the dynamics of Glacier de Saint-Sorlin, France. This glacier is well suited for this kind of study. First, the particular geometry of the glacier itself as well as that of the bedrock surface allows for correct applicability of the SIA (zeroth-order equations), provided that thickness changes and termini positions rather than short-scale dynamics are considered. Secondly, the wealth of available data for the glacier including mass-balance series and records of glacier changes provides a reliable forcing and a powerful constraining set for the model. Steady-state simulations show realistic results and the capabilities of the model in reproducing the glacier extent at the beginning of the 20th century. An extensive parameter study of ice rheology and sliding intensity is also carried out and the results are checked against the thickness changes as well as the glacier termini positions since 1905. It is possible to find a parameter combination that best matches these two types of data with an ice-flow rate factor of 2 × 10−24 Pa−3 s−1 and a Weertman-type sliding factor of 5 × 10−14 m8 N−3 a−1 which both appear to be in agreement with similar inferences from recent modelling attempts.

This paper deals with present-day gravity changes in response to the evolving Greenland ice sheet. We present a detailed computation from a 3-D thermomechanical ice sheet model that is interactively coupled with a self-gravitating spherical viscoelastic bedrock model. The coupled model is run over the last two glacial cycles to yield the loading evolution over time. Based on both the ice sheet's long-term history and its modern evolution averaged over the last 200 years, results are presented of the absolute gravity trend that would arise from a ground survey and of the corresponding geoid rate of change a satellite would see from space. The main results yield ground absolute gravity trends of the order of ±1 µgal yr−1 over the ice-free areas and total geoid changes in the range between −0.1 and +0.3 mm yr−1. These estimates could help to design future measurement campaigns by revealing areas of strong signal and/or specific patterns, although there are uncertainties associated with the parameters adopted for the Earth's rheology and aspects of the ice sheet model. Given the instrumental accuracy of a particular surveying method, these theoretical trends could also be useful to assess the required duration of a measurement campaign. According to our results, the present-day gravitational signal is dominated by the response to past loading changes rather than current mass changes of the Greenland ice sheet. We finally discuss the potential of inferring the present-day evolution of the Greenland ice sheet from the geoid rate of change measured by the future geodetic GRACE mission. We find that despite the anticipated high-quality data from satellites, such a method is compromised by the uncertainties in the earth model, the dominance of isostatic recovery on the current bedrock signal, and other inaccuracies inherent to the method itself.

The paper presents a discussion of evolution patterns of present-day changes of ice thickness, surface elevation, and bedrock elevation over the Greenlandand Antarctic continents. These patterns were obtained from calculations with dynamic 3-D thermomechanic ice sheet models which were coupled to aself-gravitating spherical visco-elastic Earth model. The experiments were initialized with simulations over the last two glacial cycles and subsequentlyanalyzed over the last 200 years to obtain the present evolution. The calculations brought to light that the Antarctic ice sheet is still adjusting to the lastglacial-interglacial transition yielding a decreasing ice volume and a rising bedrock elevation of the order of several cm per year. The Greenland ice sheet, onthe other hand, was found to be close to a stationary state with a mean thickness change of only a few mm per year. However, the calculations revealed largespatial differences. Patterns over Greenland are characterized by a small thickening over the ice-sheet interior and a general thinning of the ablation areatogether with a concomitant concentric pattern of rising bedrock elevations around the Greenland margin and a small sinking below central Greenland. InAntarctica, almost all of the changes are concentrated in the West Antarctic ice sheet, which is still retreating at both the Weddell and Ross Sea margins. Overmost of both ice sheets, the surface elevation trend is dominated by ice thickness changes rather than by bedrock elevation changes.

Two main contributions dominate the mass balance of Antarctica: surface mass balance, represented by all processes of gain and loss of mass acting at the upper surface (precipitations, melting, sublimation, wind transport...) and dynamical transport from grounded ice to the sea. This transfer takes place through outlet glaciers and represents 90% of the total loss of the whole ice sheet volume. These coastal systems act as regulators for the evolution of the ice sheet mass balance and for its contribution to sea level rise. Recently, observations emphasize a relevant decrease of mass balance in certain zones of Antarctica such as the West Coast, and an associated contribution to sea level rise from the ice sheet which increases significantly with respect to thermal expansion. Studying dynamics of outlet glaciers by modelisation thus participates at a better understanding of involved processes and enables to predict their response to any external sollicitations and to assess their potential impact on sea level budget. This work aims at providing with new elements for constraining these ice flow models for Antarctica. It is composed of two main parts. The first one concerns the implementation of physical processes into numerical models, in particular represented by grounding line migration, delimitating the grounded part from the floating one. It is based on 2 dimensionnal synthetic cases. The diversity of flow line ice sheet models is emphasized, with the associated differences and similarities. Most of these models lie on two strong assumptions, e.g. steadyness and dominance of basal sliding on ice flow, which are not always fulfilled. Moreover, the intercomparison work reveals discrepancies between models in terms of sea level contribution suggesting particular caution to be taken when considering corresponding results. Efforts have now to converge towards the validity of assumptions and on methods for implementing grounding line migration. The second part consists in applying the so-called 'full-Stokes' Elmer/Ice model to the 3D real case of the Astrolabe Glacier situated in Adélie Land in the east part of Antarctica. This application allows us to conduct sensitivity tests of the model to input incertainties such as the ones of bedrock description. This sensitivity appears to be significant, recommending a good knowledge of bedrock elevations and appropriate methods for its determination on the mesh nodes associated to the model., Le bilan de masse de l'Antarctique dépend de deux contributions principales: le bilan de masse de surface constitué par l'ensemble des processus de perte et de gain de masse agissant en surface (précipitations, fonte, sublimation, tranport par le vent...) et le transfert dynamique de glace de la calotte vers la mer. Ce transfert a lieu au niveau des glaciers émissaires côtiers et représente 90% de la perte du volume total de glace de la calotte. Ces systèmes côtiers constituent ainsi les principaux régulateurs de l'évolution du bilan de masse de la calotte et de la contribution de cette dernière à l'élévation du niveau des mers. Les observations récentes révèlent une diminution accrue du bilan de masse en certaines zones de la calotte polaire telle que la partie Ouest et une contribution à l'élévation du niveau des mers de la part de la calotte qui augmente progressivement par rapport à l'expansion thermique. L'étude de la dynamique des glaciers émissaires par modélisation permet ainsi de mieux comprendre leur fonctionnement, de prédire leur réponse à une quelconque sollicitation, et d'évaluer l'impact potentiel sur l'élévation du niveau des mers. Ce travail de thèse vise à apporter de nouveaux éléments pour mieux contraindre ces modèles d'écoulement de la calotte antarctique. Il se scinde en deux axes principaux. La première partie s'intéresse à l'implémentation des processus physiques dans le modèle numérique, en particulier représenté par la migration de la ligne d'échouage qui sépare la partie posée de la partie flottante. Elle se base sur des cas synthétiques 2D. Nous mettons en évidence dans cette partie la diversité des modèles d'écoulement de calotte impliqués et les différences et similitudes associées. En particulier, la majorité de ces modèles repose sur deux hypothèses fortes que sont la stationnarité et la prédominance du glissement basal sur l'écoulement dont le bien-fondé est remis en question. L'étude d'intercomparaison révèle certaines divergences entre les modèles en terme de contribution à l'élévation du niveau des mers, les résultats nécessitent donc d'être pris avec précaution et les efforts doivent se concentrer sur la validité des hypothèses et l'implémentation du processus de migration de la ligne d'échouage. La deuxième partie consiste à appliquer le modèle Elmer/ice, qualifié de 'full-stokes', au cas réel 3D du Glacier de l'Astrolabe situé en Terre Adélie en Antarctique de l'Est.Cette application nous a permis de tester la sensibilité du modèle full-Stokes aux incertitudes sur les données d'entrées telles que la description du socle rocheux. Cette sensibilité s'avère significative recommandant une bonne connaissance de l'élévation du socle sous jacent à la glace et des méthodes appropriées pour sa détermination sur les noeuds du maillage associé au modèle.

Modeling glacier dynamics needs proper knowledge of a number of processes which are responsible for the displacement observed at the surface of glaciers. Some of these mechanisms are well known and yet implemented into ice-flow models. On the other hand, processes that work at the base of the glaciers are much less controlled. Thus, even if glacier sliding has been observed since the early twentieth century its accurate modeling is still a current issue. In order to determine glacier sliding, friction laws that are currently used in ice flow models are only depending upon the basal shear stress. This simple relationship needs a precise fitting of the parameters which vary both in time and space so as to yield surface velocities compatible with data. Field observations also show that subglacial water pressure plays a crucial role in glacier dynamics. Furthermore, water pressure is closely related to the volume of water present at the bed of the glacier and, therefore to the production of water. The objective of this thesis is to develop a subglacial hydrological model which enables the computation of water pressure at the base of glaciers and to couple it to an ice flow model through a friction law. We choose to implement an equivalent porous media which, according to the choose parameters, features both efficient and inefficient components of the system. The sensitivity experiments show that the proposed method can reproduce the characteristics of a subglacial drainage system. Finally, the robustness of the model arose from its ability to qualitatively reproduce an extreme glaciological phenomenon under the form of a jökulhlaup., La modélisation de la dynamique glaciaire passe par la compréhension et la reproduction des processus physiques responsables des déplacements observés à la surface des glaciers. Certains de ces processus, et en particuliers ceux qui oeuvrent à la base des glaciers, sont moins bien maîtrisés. Ainsi, même si le glissement à la base des glaciers à été observé dès le début du XXe siècle sa modélisation reste un problème actuel. La majorité des modèles de dynamiques glaciaires utilisent des lois de frottement uniquement basées sur la contrainte basale tangentielle pour déterminer les vitesses de glissement. Il est alors nécessaire de faire varier en temps et en espace le paramètre de la loi de frottement pour obtenir un champ de vitesse comparable aux données mesurées. Par ailleurs, de nombreuses études ont montré que la pression et donc le volume d'eau à la base des glaciers jouait un rôle important sur la vitesse de glissement des glaciers. L'objectif de cette thèse est de mettre en place un modèle capable de calculer la pression d'eau à la base des glaciers et de le coupler à un modèle d'écoulement glaciaire par l'intermédiaire d'une loi de frottement. On utilise pour cela une approche utilisant des milieux poreux analogues représentant les deux composantes (inefficace et efficace) du système de drainage. Les expériences de sensibilité présentées montrent que cette méthode permet de reproduire les spécificités d'un système de drainage sous-glaciaire. Enfin, la reproduction qualitative d'un phénomène glaciologique extrême de jökulhlaup (vidange de lac sous-glaciaire) a permis de vérifier la robustesse du modèle.

In this work, various aspects of glacier flow modelling are addressed.The model developed by Le Meur and Vincent (2003) is used at different occasions. It is upgraded by switching to a semi-implicit scheme, mass conservation is improved and other shortcomings are resolved (ice-thickness that may become negative and an unrealistic ice settlement above the bergschrund).A comparison between two different approaches for mass balance fields is performed on the St. Sorlin glacier (France) : mass balance from measurements and mass balance from a model. The future behaviour of the glacier under a climatic scenario is also predicted.An intercomparison is done with the goal to determine the type of model that is the most appropriate for a given type of glacier geometry.The models used are : the SIA model from Le Meur and Vincent (2003), the SIA model and the higher-order model from Pattyn (2003) and a Full Stokes model (Elmer). Different synthetic geometries are used as well as a real case. The synthetic tests show on the one hand the limits of the applicability of the SIA. On the other hand a rough comparison of CPU times shows the gain in CPU time. Conversely, the increase in CPU time turns out to be reduced when switching from a higher-order model to a Full Stokes model.The simulations on the St. Sorlin glacier give an insight into the validity of the SIA on this glacier. Even if the large-scale evolution is correctly reproduced, neither the velocity field nor some small structures in the surface geometry can be properly reproduced. Simulations are compared to observations for snout position and surface velocities. A last chapter deals with the Cotopaxi glacier (Andes)., Des aspects variés de la modélisation de l'écoulement glaciaire sont abordés. Le modèle de Le Meur et Vincent (2003) est utilisé à plusieurs reprises. Plusieurs améliorations sont effectuées (conservation de la masse, traitement des épaisseurs négatives, traitement de la glace située en amont de la rimaye).Sur le glacier de St. Sorlin (France) une comparaison entre le champ de bilan issu des mesures et celui issu d'un modèle est effectuée. Le comportement du glacier sous un scénario climatique futur est prédit.Une intercomparaison entre différents types de modèles est effectuée avec pour objectif la détermination du type de modèle le plus approprié en fonction du type de glacier. Les modèles testés sont le modèle SIA de Le Meur et Vincent (2003), le modèle SIA et le modèle d'ordre supérieur de Pattyn (2003) et un modèle Full Stokes (Elmer). Des géométries synthétiques sont utilisées ainsi qu'un cas réel. Les tests synthétiques montrent les limites de l'applicabilité de la SIA. Par contre, une comparaison rapide montre le gain considérable en temps CPU. D'un autre côté, l'augmentation du coût en terme de temps CPU ne s'avère pas très importante lors du passage d'un modèle d'ordre supérieur à un modèle Full Stokes.Les simulations effectuées sur le glacier de St. Sorlin donnent un aperçu des limites de la validité de la SIA sur ce glacier. Même si elle reproduit globalement l'évolution observée du glacier, elle ne reproduit pas correctement le champ de vitesse ni certaines structures de la géométrie.Les simulations sont comparées avec les observations pour la position du front et les vitesses de surface. Un dernier chapitre est consacré au glacier Cotopaxi (Andes).