Le climat continental et froid de la Béringie lors de la glaciation du Wisconsinien a conduit à la formation d’une forme relique de pergélisol syngénétique nommé yedoma. Ces dépôts ont permis la préservation d’indicateurs environnementaux très diversifiés qui peuvent être employés pour reconstituer la dynamique climatique et écologique de la Béringie avant le dernier maximum glaciaire. À ce jour, peu d’études ont été réalisées au nord de la chaîne de montagnes Brooks (Alaska) et l’hétérogénéité écologique régionale de la Béringie Est lors de la glaciation du Wisonsinien reste mal définie. Ce mémoire porte sur une reconstitution paléoenvironnementale de plus de 39 ka du nord de l’Alaska réalisée à partir de sédiments provenant du Yedoma de la rivière Itkillik. Les objectifs sont (1) de reconstituer l’histoire de la végétation avec l’analyse pollinique; (2) de reconstituer les températures de juillet, le contraste de température saisonnier et l’ensoleillement de juillet avec la technique des analogues modernes et (3) de mettre les données biogéochimiques et glaciologiques du site en lien avec le climat reconstitué. L’étude montre que vers 35 ka BP (Interstade du Wisconsinien Moyen), des conditions climatiques semblables à l’actuel ont favorisé l’accumulation de tourbe riche en carbone organique. À partir de 29,7 ka BP, les températures de juillet reconstituées diminuent, alors que la continentalité du climat semble augmenter. Le contenu en glace des sédiments est plus alors plus faible et la pluie pollinique devient dominée par Poaceae, Artemisia et autres herbacés non graminoïdes. Ces indicateurs suggèrent des conditions environnementales plus xériques qu’aujourd’hui. Les anomalies isotopiques de 18O, 2H et l’excès de deutérium confirment un épisode d’avancée glaciaire (Wisconsinien Tardif). Après 17,9 ka BP (Tardiglaciaire), les températures de juillet et le contraste saisonnier augmentent. Les valeurs de contenu en carbone organique des sédiments sont plus élevées et la plus grande disponibilité en eau favorise l’établissement d’un couvert herbacé moderne dominé par les Cyperaceae. / The cold-arid climate associated with the Wisconsinan glaciation in Beringia has led to the formation of a relict form of syngenetic permafrost, termed yedoma. These deposits contain various environmental proxies that can be used to reconstruct the climatic and ecological dynamics across Beringia prior to the Last Glacial Maximum (LGM). To date, only a few studies have attempted to reconstruct LGM climate north of the Brooks Range and the regional ecological heterogeneity of eastern Beringia is still poorly understood. The present thesis focuses on paleoenvironmental reconstructions of northern Alaska spanning about 39 ka, based on sediments from the Itkillik river Yedoma. The objectives are (1) to reconstruct the regional vegetation history from pollen analysis; (2) to reconstruct the July temperatures, seasonal temperature contrast and July sunshine based on the modern analogue technique applied to pollen and (3) to link the biogeochemical and glaciological records to the reconstructed climate. The study shows that around 35 ka BP (Middle Wisconsinan), climate conditions were similar than modern and favored the accumulation of peat and organic carbon. From 29.7 ka BP, July temperature decreased as continentality increased. Ice content was low and the vegetation was dominated by Poaceae, Artemisia and other non-graminoid indicators of xeric environmental conditions. Isotopic anomalies of 18O, 2H and deuterium excess indicate a glacial advance (Late Wisconsinan). Improving climate and ecological conditions is recorded after 17.9 ka BP (Late Glacial). Overall, the results are more similar to reconstructions of other sites located in northern and interior Alaska than those from interior Yukon or western Beringia.

Le réchauffement climatique affecte fortement les régions nordiques du Canada où le dégel du pergélisol discontinu à sa limite sud est accompagné du mouvement de la limite des arbres vers le nord en zone de pergélisol continu. Ces altérations faites aux paysages de la Taïga des Plaines sont le point de départ de plusieurs rétroactions puisque les changements apportés aux caractéristiques de la surface (au niveau de l’albédo, l’humidité du sol et la rugosité de la surface) vont à leur tour entraîner des modifications biophysiques et éventuellement influencer l’augmentation ou la diminution subséquente des températures et de l’humidité de l’air. Seulement, il y a un nombre important de facteurs d’influence qu’il est difficile de projeter toutes les boucles rétroactives qui surviendront avec les présents changements climatiques en régions nordiques. Dans le but de caractériser les échanges d’eau et d’énergie entre la surface et l’atmosphère de trois sites des Territoires du Nord-Ouest subissant les conséquences de l’augmentation des températures de l’air, la méthode micro-météorologique de covariance des turbulences fut utilisée en 2013 aux sites de Scotty Creek (forêt boréale et tourbière nordique en zone de pergélisol sporadique-discontinu), de Havikpak Creek (forêt boréale nordique en zone de pergélisol continu) et de Trail Valley Creek (toundra arctique en zone de pergélisol continu). En identifiant les procédés biotiques et abiotiques (ex. intensité lumineuse, disponibilité en eau, etc.) d’évapotranspiration aux trois sites, les contrôles par l’eau et l’énergie furent caractérisés et permirent ainsi de projeter une augmentation de la limitation en eau, mais surtout en énergie du site de Trail Valley Creek. La répartition de l’énergie projetée est semblable à celle de Havikpak Creek, avec une augmentation de la proportion du flux de chaleur sensible au détriment de celui latent suite aux modifications des caractéristiques de la surface (albédo, rugosité et humidité du sol). L’augmentation relative du flux d’énergie sensible laisse présager une boucle rétroactive positive de l’augmentation des températures de l’air à ce site. Ensuite, en comparant des données modelées de la hauteur de la couche limite planétaire et des données provenant de profils atmosphériques d’Environnement Canada entre les trois sites, les changements de hauteur de cette couche atmosphérique furent aussi projetés. Trail Valley Creek pourrait connaître une hausse de la hauteur de sa couche limite planétaire avec le temps alors que Scotty Creek connaîtrait une diminution de celle-ci. Ces changements au niveau des couches atmosphériques liés à la répartition des flux d’énergie dans les écosystèmes se répercuteraient alors sur le climat régional de façon difficile à déterminer pour l’instant. Les changements apportés désignent une boucle rétroactive positive des températures de l’air à Trail Valley Creek et l’inverse à Scotty Creek. Les deux axes d’analyse arrivent donc aux mêmes conclusions et soulignent aussi l’importance de l’influence mutuelle entre le climat et les caractéristiques spécifiques des écosystèmes à la surface. / Along the southern margin of permafrost, the boreal forest is underlain by ice-rich and relatively warm permafrost which is converted into permafrost-free peatlands and lake ecosystems due to warmer temperatures and increased thaw rates. At the same time, in the continuous permafrost zone the tree-line of the boreal forest is advancing northward into what is currently Arctic tundra. Both land cover changes in the Taiga Plains ecozone are affecting the magnitude of complex feedback loops, including regional biophysical feedbacks through altered net water vapor and heat exchanges caused by changes in land surface albedo, hydrology and surface roughness. Changes affecting the ecosystems are numerous and it is currently hard to estimate the direction (positive or negative) and magnitude of the resulting biophysical feedbacks. To improve our understanding of implications arising from land cover changes, the energy and water exchanges between surface and atmosphere at three sites in the Northwest Territories, Canada are characterized: Scotty Creek (boreal forest-peatland landscape with sporadic permafrost), Havikpak Creek (boreal forest with continuous permafrost) and Trail Valley Creek (tundra with continuous permafrost). The results of this study are based on measurements of water vapor and heat fluxes obtained with the eddy covariance technique, in addition to supporting ancillary measurements (e.g., net radiation, ground heat flux). For the growing season of 2013, biotic and abiotic controls (ex. light intensity, water availability, etc.) of evapotranspiration at the three sites were identified and analyzed leading to a projected increase in water and energy limitation for Trail Valley Creek. This limitation can be explained by increased energy repartition to sensible heat than to latent heat, following alterations of the land surface as the treeline moves towards the arctic tundra landscape. The relative increase in the sensible heat flux is an indication for an amplified positive feedback of rising air temperature. A comparison of modeled planetary boundary layer heights with Environment Canada atmospheric profiles for the sites leads to the same projection of a positive air temperature feedback. As the treeline moves north, at Trail Valley Creek, an increase of its planetary boundary layer is expected and the opposite phenomenon is expected at Scotty Creek. Albedo, hydrology and surface roughness will be modified, affecting energy partitioning and atmospheric layers which in turn will influence climate. The two methods have led to the same conclusion and highlight the importance of mutual influence between climate and land surface characteristics.

Les tourbières ont contribué à refroidir le climat terrestre pendant l’Holocène en accumulant un réservoir de carbone important. Dans la forêt boréale canadienne, les sols gelés en permanence (pergélisols) sont répandus et ceux-ci sont principalement localisés dans les tourbières où ils forment des plateaux surélevés. Le dégel du pergélisol, causé entre autres par le réchauffement atmosphérique ou d’autres perturbations, provoque l’effondrement des plateaux et la saturation en eau du sol ce qui modifie entre autres le couvert végétal et le cycle du carbone. Les modélisations suggèrent que les latitudes nordiques seront les plus affectées par le réchauffement climatique alors qu’on y observe déjà un recul du couvert du pergélisol. Il est primordial de comprendre comment le dégel du pergélisol affecte la fonction de puits de carbone des tourbières puisque des rétroactions sur le climat sont possibles si une grande quantité de gaz à effet de serre est émise ou séquestrée. J’utilise une chronoséquence représentant le temps depuis le dégel d’un plateau de pergélisol des Territoires du Nord-Ouest pour comprendre les facteurs influençant l’aggradation et la dégradation du pergélisol dans les tourbières et évaluer l’effet du dégel sur l’accumulation de carbone et la préservation du carbone déjà accumulé. Les taux d’accumulation de carbone associés à la présence de pergélisol dans le passé et au présent sont lents, et la tourbe est moins décomposée dans les secteurs ayant été affectés plus longtemps par le pergélisol. En somme, le pergélisol réduit l’accumulation de carbone en surface mais permet une meilleure préservation du carbone déjà accumulé. / Peatlands have contributed to cool the Earth's climate during the Holocene by accumulating a large carbon pool. In the Canadian boreal forest, perennially frozen soils (permafrost soils) are abundant and they are located mainly in peatlands where they form elevated plateaus. Thawing permafrost caused by atmospheric warming or other disturbances lead to the collapse of plateaus and soil saturation, impacting vegetation cover and carbon cycling. Models suggest that northern latitudes will be the most severely affected by global warming as we are already observing a decline in permafrost cover. It is important to understand how permafrost thaw affects the peatland carbon sink function as feedbacks on the climate are possible if a large amount of greenhouse gas is emitted or sequestered. I use a chronosequence representing the time since permafrost in a Northwest Territories peatland to understand the factors influencing aggradation and degradation of permafrost in peatlands and to evaluate the effect of thawing on the carbon accumulation and preservation. The carbon accumulation rates associated with the presence of permafrost in the past and present are slow, and the peat is less decomposed in areas that have been affected by permafrost longer. In sum, permafrost reduces surface carbon accumulation but allows for better preservation of the carbon already accumulated.

Ice and frost wedges are a geomorphological phenomenon which is directly related to periglacial environment and permafrost (Murton, 2007). The presence of permafrost in the territory of the Czech Republic in the Pleistocene period is directly proven by polygonal nets of ice and frost wedge pseudomorphs, which are clearly visible in some remote sensing images. Among others, they can also be used as indicators of paleoenvironmental conditions for the period in which their recent forms originated and developed and for the period of their secondary infilling (Sekyra, 1958). The present thesis focuses on the spatial distribution and morphology of polygonal nets of ice and frost wedge pseudomorphs which were created in the territory of the Czech Republic at the end of Pleistocene and the beginning of Holocene. The analysis of the spatial distribution of polygonal nets was carried out with freely available remote sensing images provided by the GoogleEarth Pro application (Google Inc., 2011). The number of locations with a potential presence of pseudomorphs was 629. Out of these, 49 were subjected to a morphometric analysis of polygonal nets and their corresponding landscape. Statistical data analysis showed that the described polygonal nets of ice and frost wedge pseudomorphs in the territory of the Czech Republic...

About 24 % of the land surface in the northern hemisphere are underlayed by permafrost in various states. Permafrost aggradation occurs under special environmental conditions with overall low annual precipitation rates and very low mean annual temperatures. Because the general permafrost occurrence is mainly driven by large-scale climatic conditions, the distribution of permafrost deposits can be considered as an important climate indicator. The region with the most extensive continuous permafrost is Siberia. In northeast Siberia, the ice- and organic-rich permafrost deposits of the Ice Complex are widely distributed. These deposits consist mostly of silty to fine-grained sandy sediments that were accumulated during the Late Pleistocene in an extensive plain on the then subaerial Laptev Sea shelf. One important precondition for the Ice Complex sedimentation was, that the Laptev Sea shelf was not glaciated during the Late Pleistocene, resulting in a mostly continuous accumulation of permafrost sediments for at least this period. This shelf landscape became inundated and eroded in large parts by the Holocene marine transgression after the Last Glacial Maximum. Remnants of this landscape are preserved only in the present day coastal areas.<br><br> Because the Ice Complex deposits contain a wide variety of palaeo-environmental proxies, it is an excellent palaeo-climate archive for the Late Quaternary in the region. Furthermore, the ice-rich Ice Complex deposits are sensible to climatic change, i.e. climate warming. Because of the large-scale climatic changes at the transition from the Pleistocene to the Holocene, the Ice Complex was subject to extensive thermokarst processes since the Early Holocene.<br><br> Permafrost deposits are not only an environmental indicator, but also an important climate factor. Tundra wetlands, which have developed in environments with aggrading permafrost, are considered a net sink for carbon, as organic matter is stored in peat or is syn-sedimentary frozen with permafrost aggradation. Contrary, the Holocene thermokarst development resulted in permafrost degradation and thus the release of formerly stored organic carbon. Modern tundra wetlands are also considered an important source for the climate-driving gas methane, originating mainly from microbial activity in the seasonal active layer.<br><br> Most scenarios for future global climate development predict a strong warming trend especially in the Arctic. Consequently, for the understanding of how permafrost deposits will react and contribute to such scenarios, it is necessary to investigate and evaluate ice-rich permafrost deposits like the widespread Ice Complex as climate indicator and climate factor during the Late Quaternary. Such investigations are a pre-condition for the precise modelling of future developments in permafrost distribution and the influence of permafrost degradation on global climate.<br><br> The focus of this work, which was conducted within the frame of the multi-disciplinary joint German-Russian research projects "Laptev Sea 2000" (1998-2002) and "Dynamics of Permafrost" (2003-2005), was twofold. First, the possibilities of using remote sensing and terrain modelling techniques for the observation of periglacial landscapes in Northeast Siberia in their present state was evaluated and applied to key sites in the Laptev Sea coastal lowlands. The key sites were situated in the eastern Laptev Sea (Bykovsky Peninsula and Khorogor Valley) and the western Laptev Sea (Cape Mamontovy Klyk region). For this task, techniques using CORONA satellite imagery, Landsat-7 satellite imagery, and digital elevation models were developed for the mapping of periglacial structures, which are especially indicative of permafrost degradation. The major goals were to quantify the extent of permafrost degradation structures and their distribution in the investigated key areas, and to establish techniques, which can be used also for the investigation of other regions with thermokarst occurrence. Geographical information systems were employed for the mapping, the spatial analysis, and the enhancement of classification results by rule-based stratification. The results from the key sites show, that thermokarst, and related processes and structures, completely re-shaped the former accumulation plain to a strongly degraded landscape, which is characterised by extensive deep depressions and erosional remnants of the Late Pleistocene surface. As a results of this rapid process, which in large parts happened within a short period during the Early Holocene, the hydrological and sedimentological regime was completely changed on a large scale. These events resulted also in a release of large amounts of organic carbon. Thermokarst is now the major component in the modern periglacial landscapes in terms of spatial extent, but also in its influence on hydrology, sedimentation and the development of vegetation assemblages. Second, the possibilities of using remote sensing and terrain modelling as a supplementary tool for palaeo-environmental reconstructions in the investigated regions were explored. For this task additionally a comprehensive cryolithological field database was developed for the Bykovsky Peninsula and the Khorogor Valley, which contains previously published data from boreholes, outcrops sections, subsurface samples, and subsurface samples, as well as additional own field data. The period covered by this database is mainly the Late Pleistocene and the Holocene, but also the basal deposits of the sedimentary sequence, interpreted as Pliocene to Early Pleistocene, are contained. Remote sensing was applied for the observation of periglacial strucures, which then were successfully related to distinct landscape development stages or time intervals in the investigation area. Terrain modelling was used for providing a general context of the landscape development. Finally, a scheme was developed describing mainly the Late Quaternary landscape evolution in this area. A major finding was the possibility of connecting periglacial surface structures to distinct landscape development stages, and thus use them as additional palaeo-environmental indicator together with other proxies for area-related palaeo-environmental reconstructions. In the landscape evolution scheme, i.e. of the genesis of the Late Pleistocene Ice Complex and the Holocene thermokarst development, some new aspects are presented in terms of sediment source and general sedimentation conditions. This findings apply also for other sites in the Laptev Sea region. / Die vorliegende Arbeit wurde im Rahmen der multidisziplinären Deutsch-Russischen Verbundprojekte "Laptev See 2000" (1998-2002) und "Dynamik des Permafrost" (2003-2005) erstellt.<br> Etwa 24 % der Landoberfläche der Erde sind von Permafrost unterlagert. Die ausgedehntesten Permafrostgebiete befinden sich heute in Sibirien. In Nordostsibirien, das während der letzten Eiszeit nicht von Inlandeismassen bedeckt bedeckt war, lagerten sich während dieser Zeit mächtige eisreiche Permafrostsedimente ab. Die durch den nacheiszeitlichen Meeresspiegelanstieg um ca. 120 Meter nur noch in den heutigen Küstengebieten erhaltenen Ablagerungen sind zum Teil hervorragende Paläoklimaarchive, die verschiedenste fossile organische Überreste der Eiszeitlichen Fauna und Flora konserviert haben. Aber auch die Sedimente und das enthalten Grundeis enthalten Klimainformationen z.B. die aus Mineralogie, Ablagerungsmilieu oder geochemischer und isotopenchemischer Zusammensetzung gewonnen werden können.<br><br> Der hohe Eisgehalt in den Sedimenten führte mit Beginn der holozänen Warmzeit zur Bildung von Thermokarst und Thermo-Erosion, d.h. zu starken Zersetzungserscheinungen durch Auftauen und Erosion. Thermokarst beschreibt das Schmelzen des Grundeises und die gleichzeitig stattfindende tiefe Absenkung der betroffenen Landoberfläche. Thermokarst geht mit der Bildung von Thermokarstseen einher, deren Wasserkörper ein zusätzlicher Wärmespeicher ist und das Auftauen des darunter liegenden Permafrost verstärken kann. In Sibirien, aber auch anderen Regionen der Arktis, sind weite Gebiete von Thermokarst betroffen. Der Einfluss dieser klimabedingten großräumigen Landschaftsveränderungen in Permafrostgebieten auf den lokalen, regionalen und auch globalen Stoff- und Energiehaushalt ist bisher nur wenig untersucht. Die vorliegende Arbeit beschäftigt sich mit der Charakterisierung und Evolution von periglazialen Landschaften im nordsibirischen Laptevsee-Gebiet, die seit dem Beginn des Holozän von solchen klimatisch bedingten Veränderungen betroffen sind, und liefert damit ein Puzzleteil zum einen für die Rekonstruktion der Landschaft und Landschaftsentwicklung als auch Vorraussetzungen für das Verständnis der großräumig wirkenden geologischen und geomorphologischen Veränderungsprozesse. Die generellen Schwerpunkte, für die die vorliegende Arbeit Informationen liefert, sind die Charakterisierung von periglazialen Relief- und Oberflächentypen und die Bestimmung ihrer räumlichen Verbreitung, die Identifizierung und Quantifizierung einzelner geologischer und geomorphologischer Prozesse in diesen Landschaften, und die Rekonstruktion der Entwicklung periglazialer Landschaften im Spätquartär für Schlüsselgebiete im Küstengebiet der nordsibirischen Laptevsee.<br><br> Um diese generellen Schwerpunkte zu erreichen, werden verschiedene Einzelziele in der Arbeit verfolgt:<br><br> Die Entwicklung and Anwendung von Satellitenfernerkundungstechniken zur Analyse periglazialer Landschaften in Nordsibirien. Dazu werden hochauflösende Corona-Satellitendaten und multispektrale Landsat-7 Satellitendaten verwendet.<br> Die Untersuchung von Satellitenbildern, mit dem Schwerpunkt auf Oberflächen, die von der Zersetzung des eisreichen Permafrosts betroffen sind<br> Die Entwicklung von hochauflösenden digitalen Geländemodellen für die geomorphologische Analyse in zwei Schlüsselgebieten<br> Die räumliche Untersuchung der gewonnenen Daten mit Hilfe von geographischen Informationssystemen, mit einem Schwerpunkt auf Form, Verteilung und Außmaß von holozänem Thermokarst<br> Das Sammeln und Auswerten von Felddaten, mit Schwerpunkt auf Oberflächeneigenschaften periglazialer Landschaften und der Zusammensetzung der Permafrostablagerungen<br> Die Anwendung der gewonnenen Daten zur Unterstützung, Verbesserung und Ausweitung der lokal gewonnenen Felddaten und Paläoumweltrekonstruktionen, sowie die datengestützte Entwicklung von Vorstellungen zur Landschaftsgenese<br><br> Weite, Permafrost-dominierte Küstentiefländer der heutigen Laptevsee in Nordost-Sibirien sind durch die spätpleistozänen Ablagerungen des Eiskomplex aufgebaut. Diese zumeist schluffig bis mittelsandigen Ablagerungen sind durch einen sehr großen Eisgehalt in Form von verteiltem Grundeis und großer syngenetischer Eiskeile, sowie einem relativ hohen Anteil an organischen Resten gekennzeichnet. Mit Beginn der holozänen Klimaerwärmung kam es zur weitläufigen Bildung von Thermokarst.

Numerous glacial advances over the past 2 million years have covered the entire Canadian and Fennoscandian Shield outcrop. During glacial advance and retreat, permafrost is expected to form in front of the glacier. The question of how permafrost and freezing impact the formation and evolution of brines in natural systems may be vital to understanding the chemistry of groundwater in crystalline rocks. Investigations of groundwater conditions beneath thick permafrost can provide valuable information that can be applied to assessing safety of deep, underground nuclear waste repositories and understanding analogues to potential life-bearing zones on Mars. However, very little scientific investigation of cryogenic processes and hydrogeology deep within crystalline systems has been published. The purpose of this research is to evaluate the impacts of thick permafrost (>300m) formation on groundwater chemical and flow system evolution in the crystalline rock environment over geologic timescales. A field investigation was conducted at the Lupin Mine in Nunavut, Canada, to characterize the physical and hydrogeochemical conditions within and beneath a thick permafrost layer. Taliks, or unfrozen channels within the permafrost, are found beneath large lakes in the field area, and provide potential hydraulic connections through the permafrost. Rock matrix waters are dilute and do not appear to affect groundwater salinity. Permafrost waters are Na-Cl and Na-Cl-SO4 type, and have been contaminated with chloride and nitrate by mining activities. Sulfide oxidation in the permafrost may be naturally occurring or is enhanced by mining activities. Basal permafrost waters (550 to 570 mbgs) are variably affected by mining. The less contaminated basal waters have medium sulfate concentrations and are Ca-Na dominated. This is similar to deeper, uncontaminated subpermafrost waters, which are Ca-Na-Cl or Na-Ca-Cl type with a wide range of salinities (2.6 to 40 g•L-1). The lower salinity subpermafrost waters are attributed to dissociation of methane hydrate and drawdown of dilute talik waters by the hydraulic gradient created by mine dewatering. This investigation was unable to determine the influence of talik waters to the subpermafrost zone in undisturbed conditions. Pressures are also highly variable, and do not correlate with salinity. Fracture infillings are scarce and calcite δ18O and δ13C values have a large range. Microthermometry indicates a large range in salinities and homogenization temperatures as well, indicative of a boiling system. In situ freezing of fluids and methane hydrate formation may have concentrated the remaining fluids. Field activities at the Lupin mine also provided an opportunity to study the nature of gases within crystalline rocks in a permafrost environment. Gases were generally methane-dominated (64 to 87), with methane δ13C and δ2H values varying between -56 and -42‰ VPDB and -349 to -181 ‰ VSMOW, respectively. The gases sampled within the Lupin mine have unique ranges of chemical and isotopic compositions compared with other Canadian and Fennoscandian Shield gases. The gases may be of thermogenic origin, mixed with some bacteriogenic gas. The generally low δ2H-CH4 ratios are somewhat problematic to this interpretation, but the geologic history of the site, a metaturbidite sequence, supports a thermogenic gas origin. The presence of gas hydrate in the rock surrounding Lupin was inferred, based on temperature measurements and hydrostatic pressures. Evidence also suggests fractures near the mine have been depressurized, likely due to mine de-watering, resulting in dissipation of methane hydrate near the mine. Modeling results indicate methane hydrates were stable throughout the Quaternary glacial-interglacial cycles, potentially limiting subglacial recharge. The effects of deep permafrost formation and dissipation during the Pleistocene glacial/interglacial cycle to deep groundwaters in the Canadian Shield were also investigated by compiling data from thirty-nine sites at twenty-four locations across the Canadian Shield. Impacts due to glacial meltwater recharge and surficial cryogenic concentration of fluids, which had been previously considered by others, and in situ freeze-out effects due to ice and/or methane hydrate formation were considered. At some Canadian Shield sites, there are indications that fresh, brackish, and saline groundwaters have been affected by one of these processes, but the data were not sufficient to differentiate between mixed, intruded glacial meltwaters, or residual waters resulting from either permafrost or methane hydrate formation. Physical and geochemical data do not support the cryogenic formation of Canadian Shield brines from seawater in glacial marginal troughs. The origin and evolution of Canadian and Fennoscandian Shield brines was explored with a survey of chlorine and bromine stable isotope ratios. The δ37Cl and δ81Br isotopic ratios varied between -0.78 ‰ and 1.52 ‰ (SMOC) and 0.01 ‰ and 1.52 ‰ (SMOB), respectively. Variability of chlorine and bromine isotope ratios decreases with increasing depth. Fennoscandian Shield groundwaters tend to be more enriched than Canadian Shield groundwaters for both 37Cl and 81Br. Other sources and processes which may affect δ37Cl and δ81Br composition are also explored. Primary processes such as magmatic and/or hydrothermal activity are thought to be responsible for the isotopic composition of the most concentrated fluids at each site. Positive correlations between δ81Br, and δ37Cl with δ2H-CH4 and δ13C-CH4 were noted. At this time the cause of the relationship is unclear, and may be a result of changing redox, pH, temperature, and/or pressure conditions during hydrothermal, metamorphic, or volcanogenic processes. The data suggest solute sources and fluid evolution at individual sites would be better constrained utilizing a multi-tracer investigation of δ37Cl, δ81Br, and 87Sr/86Sr ratios comparing fluids, rocks, and fracture filling minerals (including fluid inclusions).

Numerous glacial advances over the past 2 million years have covered the entire Canadian and Fennoscandian Shield outcrop. During glacial advance and retreat, permafrost is expected to form in front of the glacier. The question of how permafrost and freezing impact the formation and evolution of brines in natural systems may be vital to understanding the chemistry of groundwater in crystalline rocks. Investigations of groundwater conditions beneath thick permafrost can provide valuable information that can be applied to assessing safety of deep, underground nuclear waste repositories and understanding analogues to potential life-bearing zones on Mars. However, very little scientific investigation of cryogenic processes and hydrogeology deep within crystalline systems has been published. The purpose of this research is to evaluate the impacts of thick permafrost (>300m) formation on groundwater chemical and flow system evolution in the crystalline rock environment over geologic timescales. A field investigation was conducted at the Lupin Mine in Nunavut, Canada, to characterize the physical and hydrogeochemical conditions within and beneath a thick permafrost layer. Taliks, or unfrozen channels within the permafrost, are found beneath large lakes in the field area, and provide potential hydraulic connections through the permafrost. Rock matrix waters are dilute and do not appear to affect groundwater salinity. Permafrost waters are Na-Cl and Na-Cl-SO4 type, and have been contaminated with chloride and nitrate by mining activities. Sulfide oxidation in the permafrost may be naturally occurring or is enhanced by mining activities. Basal permafrost waters (550 to 570 mbgs) are variably affected by mining. The less contaminated basal waters have medium sulfate concentrations and are Ca-Na dominated. This is similar to deeper, uncontaminated subpermafrost waters, which are Ca-Na-Cl or Na-Ca-Cl type with a wide range of salinities (2.6 to 40 g•L-1). The lower salinity subpermafrost waters are attributed to dissociation of methane hydrate and drawdown of dilute talik waters by the hydraulic gradient created by mine dewatering. This investigation was unable to determine the influence of talik waters to the subpermafrost zone in undisturbed conditions. Pressures are also highly variable, and do not correlate with salinity. Fracture infillings are scarce and calcite δ18O and δ13C values have a large range. Microthermometry indicates a large range in salinities and homogenization temperatures as well, indicative of a boiling system. In situ freezing of fluids and methane hydrate formation may have concentrated the remaining fluids. Field activities at the Lupin mine also provided an opportunity to study the nature of gases within crystalline rocks in a permafrost environment. Gases were generally methane-dominated (64 to 87), with methane δ13C and δ2H values varying between -56 and -42‰ VPDB and -349 to -181 ‰ VSMOW, respectively. The gases sampled within the Lupin mine have unique ranges of chemical and isotopic compositions compared with other Canadian and Fennoscandian Shield gases. The gases may be of thermogenic origin, mixed with some bacteriogenic gas. The generally low δ2H-CH4 ratios are somewhat problematic to this interpretation, but the geologic history of the site, a metaturbidite sequence, supports a thermogenic gas origin. The presence of gas hydrate in the rock surrounding Lupin was inferred, based on temperature measurements and hydrostatic pressures. Evidence also suggests fractures near the mine have been depressurized, likely due to mine de-watering, resulting in dissipation of methane hydrate near the mine. Modeling results indicate methane hydrates were stable throughout the Quaternary glacial-interglacial cycles, potentially limiting subglacial recharge. The effects of deep permafrost formation and dissipation during the Pleistocene glacial/interglacial cycle to deep groundwaters in the Canadian Shield were also investigated by compiling data from thirty-nine sites at twenty-four locations across the Canadian Shield. Impacts due to glacial meltwater recharge and surficial cryogenic concentration of fluids, which had been previously considered by others, and in situ freeze-out effects due to ice and/or methane hydrate formation were considered. At some Canadian Shield sites, there are indications that fresh, brackish, and saline groundwaters have been affected by one of these processes, but the data were not sufficient to differentiate between mixed, intruded glacial meltwaters, or residual waters resulting from either permafrost or methane hydrate formation. Physical and geochemical data do not support the cryogenic formation of Canadian Shield brines from seawater in glacial marginal troughs. The origin and evolution of Canadian and Fennoscandian Shield brines was explored with a survey of chlorine and bromine stable isotope ratios. The δ37Cl and δ81Br isotopic ratios varied between -0.78 ‰ and 1.52 ‰ (SMOC) and 0.01 ‰ and 1.52 ‰ (SMOB), respectively. Variability of chlorine and bromine isotope ratios decreases with increasing depth. Fennoscandian Shield groundwaters tend to be more enriched than Canadian Shield groundwaters for both 37Cl and 81Br. Other sources and processes which may affect δ37Cl and δ81Br composition are also explored. Primary processes such as magmatic and/or hydrothermal activity are thought to be responsible for the isotopic composition of the most concentrated fluids at each site. Positive correlations between δ81Br, and δ37Cl with δ2H-CH4 and δ13C-CH4 were noted. At this time the cause of the relationship is unclear, and may be a result of changing redox, pH, temperature, and/or pressure conditions during hydrothermal, metamorphic, or volcanogenic processes. The data suggest solute sources and fluid evolution at individual sites would be better constrained utilizing a multi-tracer investigation of δ37Cl, δ81Br, and 87Sr/86Sr ratios comparing fluids, rocks, and fracture filling minerals (including fluid inclusions).

L'importance climatique des hautes latitudes est exacerbée par le contexte actuel du réchauffement climatique, de part de leur forte sensibilité à ces changements et en raison des rétroactions globales majeures qu'elles sont susceptibles d'engendrer. La modélisation offre un moyen d'estimer ces impacts dans les temps passés, présents et futurs, tout en quantifiant les incertitudes procédant des imperfections de notre connaissance de ces environnements et de leur représentation. L'amélioration et l'évaluation de la représentation des hautes latitudes dans les modèles de climat globaux répondent donc à de forts enjeux scientifiques et sociétaux : c'est dans ce cadre précis que s'inscrit mon travail de thèse. Le gel des sols est une spécificité majeure des régions circum-arctiques, porteuse d'implications climatiques aux plans thermiques, hydrologiques et biogéochimiques. Une paramétrisation des impacts hydrologiques du gel des sols a été introduite dans le schéma hydrologique multi-couches du modèle de surfaces continentales ORCHIDEE : ses effets sur le contenu en eau des sols et le régime hydrologique des principaux bassins de l'Arctique ont été évalués par comparaison à des données de terrain, révélant la plus-value d'une telle représentation mais aussi les lacunes résiduelles de la modélisation, qui touchent à l'absence de représentation des réservoirs temporaires d'eau de surface et, sans doute, d'une paramétrisation sous-maille du gel des sols. Parallèlement, une représentation des effets thermiques du gel des sols développée pour un modèle antérieur à ORCHIDEE a été révisée et évaluée à différentes échelles spatiales par comparaison à des données observationnelles : si la représentation de l'énergie de chaleur latente augmente la température des sols soumis au gel saisonnier, un biais froid subsiste dans la modélisation, imputable à une représentation imparfaite de la neige. Une étude de sensibilité conduite sur cette variable en confirme les implications thermiques mais aussi biogéochimiques à l'échelle des régions circum-arctiques, sous-tendues par les importantes quantités de matière organique que ces régions renferment. Alors que les caractéristiques de la neige sont souvent représentées comme spatialement uniformes dans les modèles de climat globaux, la simple prise en compte du caractère particulièrement isolant de la neige de taïga engendre des changements importants dans le cycle du carbone aux hautes latitudes, et souligne les incertitudes entachant notre représentation actuelle de ces écosystèmes. Les propriétés thermiques de la neige n'en sont pas l'unique vecteur, mais une évaluation détaillée de notre modélisation sur un site de permafrost arctique (station de Bayelva, Svalbard) désigne la neige comme une source majeure des incertitudes associées à notre modélisation des hautes latitudes, au travers de représentations inadaptées de son albédo, sa rugosité de surface, son contenu variable en eau liquide pouvant accommoder de l'eau de pluie. En termes hydrologiques, l'absence de représentation spécifique des zones de montagne, des caractéristiques hydrauliques des sols à granulométrie grossière du Haut-Arctique, et des nombreuses étendues d'eau libre des régions circum-arctiques, limite notre capacité à représenter raisonnablement des principales caractéristiques de l'hydrologie de surface de ces régions. Le diagnostique de ces limites définit autant de potentiels d'amélioration de la modélisation des hautes latitudes, sources possibles de développements futurs. / Focus has recently increased on high-latitude climatic processes as awareness rose about the extreme sensitivity of the Arctic to climate change and its potential for major positive climate feedbacks. Modelling offers a powerful tool to assess the climatic impact of changes in the northern high-latitude regions, as well as to quantify the range of uncertainty stemming from the limits of our knowledge and representation of these environments. My PhD project, dedicated to the improvement of a land-surface model for high-latitude regions and the evaluation of its performances, tackles therefore an issue of concern both for science and society. Soil freezing is a major physical process of boreal regions, with climatic implications. Here, a parameterization of the hydrological effects of soil freezing is developed within the multi-layer hydrological scheme of the land-surface model ORCHIDEE, and its performance is evaluated against observations at different scales, including remotely-sensed data. Taking the hydrological impact of soil freezing into account improves our representation of soil moisture and river discharges over the pan-Arctic land-surface area. However, residual inaccuracies suggest that potential for improvement lies in the representation of temporary surface water reservoirs like floodplains, surface ponding, and, possibly, the introduction of a subgrid variability in soil freezing. Hydrological modelling at high latitudes would also benefit from a specific treatment of mountainous areas and a revision of soil textural input parameters to account for abundant coarse-grained soils in the High-Arctic. Concomitantly, the thermal parameterization of soil freezing in ORCHIDEE is revised and evaluated against field data: latent heat effects yield a reduction but no suppression of a model cold bias in winter soil temperatures, part of which is imputed to the coarse representation of snow in the model. A sensitivity study performed on the insulative properties of taiga vs. tundra snow over the pan-Arctic terrestrial domain confirms the thermal implications of snow and outlines its consequences for carbon cycling at high-latitudes, calling for an appropriate representation of snow-vegetation interactions. Snow is furthermore implicated in identified flaws of the modelled surface energy balance, the components of which are precisely compared with a one-year high quality dataset collected at an Arctic permafrost site in Svalbard. Inaccuracies are diagnosed to stem from the representation of albedo, surface roughness and liquid water percolation and phase change within the snowpack. These diverse

L’étude des instabilités gravitaires profondes de versant (nommées ici RVR : Ruptures de Versant Rocheux) s’effectue généralement sous forme d’un suivi instrumenté à l’échelle d’un site jugé dangereux, parfois à l’échelle d’une vallée ou d’un massif. Plus rares sont les études qui apprécient la diversité, la taille et la distribution spatiale des RVR à l’échelle d’une chaîne de montagne. C’est ce que propose cette thèse pour les Alpes occidentales. Il s’agit tout d’abord d’un inventaire construit de manière systématique par imagerie satellite à l’aide d’un outil en accès libre : Google Earth Pro™, et d’une série de méthodes de détection visuelles assorties de vérifications sur le terrain. Une typologie qui s’appuie sur des classifications existantes, mais qui s’adapte au cortège de RVR observés dans l’aire d’étude, a permis de retenir cinq grandes catégories de RVR : les EAR (Eboulements et avalanches rocheuses), GR (Glissements rocheux), GC (Glissements-coulées), DGCVR (Déformations gravitaires profondes de versants rocheux) et DDV (Déformation de versant). Nous élaborons sur cette base une étiologie des RVR en fonction de grands facteurs préparatoires réputés mettre en mouvement les masses rocheuses : lithologie, structure géologique (contacts anormaux), sismicité, pente topographique, relief local, intensité du paléoenglacement würmien, précipitations actuelles, dégradation du pergélisol. Sur un inventaire exhaustif de 1400 RVR, les résultats montrent que la susceptibilité lithologique est le premier facteur qui conditionne l’occurrence des RVR, mais qu’il se cumule avec l’amplitude du relief exacerbée par le paléoenglacement quaternaire. Ce dernier fournit le potentiel gravitationnel localement nécessaire à la mise en mouvement des masses rocheuses. Les autres facteurs examinés présentent des degrés d’importance moindres à l’échelle régionale, avec toutefois des exceptions intéressantes à l’échelle locale et pour des catégories de RVR particulières. Ainsi, on peut noter des RVR en lien avec certaines failles et fronts de chevauchements, ainsi qu’avec la dégradation du pergélisol — mais uniquement dans le cas des éboulements. Parmi l’ensemble des facteurs, le pouvoir explicatif des totaux de précipitations demeure le plus faible. Dans une optique de valorisation scientifique du catalogue des RVR inventoriés, nous proposons des perspectives de mise en valeur géo-patrimoniale de certaines RVR sur la base de leurs caractéristiques morphologiques, ou du risque que certaines masses rocheuses font peser sur les enjeux économiques et humains des populations. Nous présentons ainsi une galerie de RVR remarquables, retenues pour leur caractère singulier, ou dangereux, ou éducatif, ou emblématique à divers titres. / The study of rockslope failure (RSF) is usually focused on the instrumental monitoring of hazardous sites, sometimes extended to a population of RSF in a valley or massif. Few studies survey and analyse RSF at the much broader scale of a mountain range. Here we produce a systematic inventory of RSF in the Western Alps based on satellite imagery provided by the open-access platform Google Earth Pro™, and using a series of ground-truth-tested visual detection methods. Based on a categorisation inspired by existing classifications but adapted to the range of RSF observed in the study area, five main RSF types were identified: rockfalls and rock avalanches, rockslides, earthflows, deep-seated gravitational slope deformations (DSGSD), and slope deformations. We analyse the spatial incidence of those five categories in relation to a range of likely cumulative causes. The analysis covers lithology and rock fabric, geological structure (faults, thrust fronts), seismicity, slope angle, local relief, the intensity of Würmian glaciation, modern rainfall patterns, and permafrost degradation. Results from a total population of 1400 RSF occurrences show that RSF incidence and mode are overwhelmingly susceptible to rock type, but that local relief enhanced by past glaciation generates the gravitational potential needed to move the rock masses. Other conditional factors receive lower rankings at the regional scale, but stronger connections appear in local settings. At places, RSF size or density are seen to correlate with faults, thrust fronts, and with permafrost degradation (restricted, however, to the rockfall category). Among all the likely causes of RSF, rainfall totals represent the weakest link. Among the 1400 sites we focus on a subset of flagship RSF occurrences that we consider relevant to either geoheritage or land-use planning concerns. The criteria were selected on the basis of morphological characteristics (uniqueness, educational and scientific value) or from the perspective of the hazards that some of the displaced rock masses may present to human life and infrastructure.

Pendant les transitions glaciaire-interglaciaires,on observe une augmentation en partie abrupte de près de 100 ppm du CO2atmosphérique, indiquant une redistribution majeure entre les réservoirs de carbone des continents, de l'océan et de l'atmosphère.Expliquer les flux de carbone associés à ces transitions est un défi scientifique, qui nécessite une meilleure compréhension du stock de carbone ‘initial’ dans la biosphère terrestre au cours de la période glaciaire. L’objectif de cette thèse est d’améliorer la compréhension du fonctionnement des écosystèmes terrestres et des stocks de carbone au cours du dernier maximum glaciaire (LGM, il y a environ21.000 ans), à travers plusieurs nouveaux développements dans le modèle global de végétation ORCHIDEE-MICT, pour améliorer la représentation de la dynamique de la végétation, la dynamique du carbone dans le sol du pergélisol et les interactions entre les grands herbivores et la végétation dans le modèle de la surface terrestre.Pour la première partie, la représentation de la dynamique de la végétation dans ORCHIDEEMICT pour les régions des moyennes et hautes latitudes, a été calibrée et évaluée avec un ensemble de données spatiales de classes de végétation, production primaire brute, et de biomasse forestière pour la période actuelle.Des améliorations sont obtenues avec la nouvelle version du modèle dans la distribution des groupes fonctionnels de végétation. Ce modèle a ensuite été appliqué pour simuler la distribution de la végétation au cours de laLGM, montrant un accord général avec les reconstructions ponctuelles basées sur des données de pollen et de macro-fossiles de plantes.Une partie du pergélisol (sols gelés en permanence) contient des sédiments épais,riches en glace et en matières organiques appelés Yedoma, qui contiennent de grandes quantités de carbone organique, et sont des reliques des stocks de carbone du Pléistocène.Ces sédiments ont été accumulés sous des climats glaciaires. Afin de simuler l'accumulation du carbone dans les dépôts de Yedoma, j’ai proposé une nouvelle paramétrisation de la sédimentation verticale dans le module de carbone dans le sol de ORCHIDEE-MICT. L'inclusion de ce processus a permis de reproduire la distribution verticale de carbone observée sur des sites de Yedoma. Une première estimation du stock de carbone dans le pergélisol au cours du LGM est obtenue, de l’ordre de ~ 1550 PgC, dont 390 ~446 PgC sous forme de Yedoma encore intacts aujourd’hui (1,3 millions de km2).Potentiellement, une plus grande surface de Yedoma pourrait être présente pendant leLGM, qui a disparue lors de la déglaciation.Pour la troisième partie, à la lumière des impacts écologiques des grands animaux, et le rôle potentiel des méga-herbivores comme une force qui a maintenu les écosystèmes steppiques pendant les périodes glaciaires, j'ai incorporé un modèle de d’herbivores dans ORCHIDEE-MICT, basé sur des équations physiologiques pour l'apport énergétique et les dépenses, le taux de natalité, et le taux de mortalité pour les grands herbivores sauvages.Le modèle a montré des résultats raisonnables de biomasse des grands herbivores en comparaison avec des observations disponibles aujourd’hui sur des réserves naturelles. Nous avons simulé un biome de prairies très étendu pendant le LGM avec une densité importante de grands herbivores. Les effets des grands herbivores sur la végétation et le cycle du carbone du LGM ont été discutés, y compris la réduction de la couverture forestière, et la plus grande productivité des prairies.Enfin, j’ai réalisé une estimation préliminaire du stock total de carbone dans le permafrost pendant le LGM, après avoir tenu compte des effets des grands herbivores et en faisant une extrapolation de l'étendue spatiale des sédiments de type Yedoma basée sur des analogues climatiques et topographiques qui sont similaires à la région de Yedoma actuelle. / During the repeated glacialinterglacialtransitions, there has been aconsistent and partly abrupt increase of nearly100 ppm in atmospheric CO2, indicating majorredistributions among the carbon reservoirs ofland, ocean and atmosphere. A comprehensiveexplanation of the carbon fluxes associatedwith the transitions is still missing, requiring abetter understanding of the potential carbonstock in terrestrial biosphere during the glacialperiod. In this thesis, I aimed to improve theunderstanding of terrestrial carbon stocks andcarbon cycle during the Last Glacial Maximum(LGM, about 21,000 years ago), through aseries of model developments to improve therepresentation of vegetation dynamics,permafrost soil carbon dynamics, andinteractions between large herbivores andvegetation in the ORCHIDEE-MICT landsurface model.For the first part, I improved theparameterization of vegetation dynamics inORCHIDEE-MICT for the northern mid- tohigh-latitude regions, which was evaluatedagainst present-day observation-based datasetsof land cover, gross primary production, andforest biomass. Significant improvements wereshown for the new model version in thedistribution of plant functional types (PFTs),including a more realistic simulation of thenorthern tree limit and of the distribution ofevergreen and deciduous conifers in the borealzone. The revised model was then applied tosimulate vegetation distribution during theLGM, showing a general agreement with thepoint-scale reconstructions based on pollen andplant macrofossil data.Among permafrost (perennially frozen) soils,the thick, ice-rich and organic-rich siltysediments called yedoma deposits hold largequantities of organic carbon, which areremnants of late-Pleistocene carbonaccumulated under glacial climates. In order tosimulate the buildup of the thick frozen carbonin yedoma deposits, I implemented asedimentation parameterization in the soilcarbon module of ORCHIDEE-MICT. Theinclusion of sedimentation allowed the modelto reproduce the vertical distribution of carbonobserved at the yedoma sites, leading toseveral-fold increase in total carbon. Simulatedpermafrost soil carbon stock during the LGMwas ~1550 PgC, among which 390~446 PgCwithin today’s known yedoma region (1.3million km2). This result was still anunderestimation since the potentially largerarea of yedoma during the LGM than todaywas not yet taken into account.For the third part, in light of the growingevidence on the ecological impacts of largeanimals, and the potential role of megaherbivoresas a driving force that maintainedthe steppe ecosystems during the glacialperiods, I incorporated a dynamic grazingmodel in ORCHIDEE-MICT, based onphysiological equations for energy intake andexpenditure, reproduction rate, and mortalityrate for wild large grazers. The model showedreasonable results of today’s grazer biomasscompared to empirical data in protected areas,and was able to produce an extensive biomewith a dominant vegetation of grass and asubstantial distribution of large grazers duringthe LGM. The effects of large grazers onvegetation and carbon cycle were discussed,including reducing tree cover, enhancinggrassland productivity, and increasing theturnover rate of vegetation living biomass.Lastly, I presented a preliminary estimation ofpotential LGM permafrost carbon stock, afteraccounting for the effects of large grazers, aswell as extrapolations for the spatial extent ofyedoma-like thick sediments based on climaticand topographic features that are similar to theknown yedoma region. Since these results werederived under LGM climate and constantsedimentation rate, a more realistic simulationwould need to consider transient climate duringthe last glacial period and sedimentation ratevariations in the next step.