Les glaciers rocheux sont l'une des expressions visibles du pergélisol de montagne et sont à ce titre l'objet d'études nombreuses et variées depuis plus de 20 ans. Deux principaux thèmes sont actuellement étudiés par la communauté scientifique : i) mieux appréhender les mouvements qui affectent les glaciers rocheux ainsi que les mécanismes qui les régissent et évaluer l'impact des changements atmosphériques globaux sur ces formes périglaciaires. Dans ces travaux de thèse nous proposons de contribuer au premier de ces deux thèmes de recherche en utilisant un des outils offerts par la télédétection, déjà utilisé dans d'autres pays pour l'étude des glaciers rocheux : l'interférométrie radar différentielle (D-InSAR). L'interférométrie radar différentielle (D-InSAR) est une méthode basée sur la mesure de la différence de phase entre deux images radar qui couvrent la même zone à différents intervalles de temps et depuis des orbites quasi-similaires. Déjà été utilisée dans d'autres domaines, notamment en glaciologie, volcanologie et sismologie, cette technique crée des interférogrammes, cartes de la déformation du sol en deux dimensions dans la ligne de visée du satellite, qui permettent de détecter et de quantifier un déplacement de quelques centimètres survenu au sol entre deux acquisitions radar. Des recherches en Suisse ont également montré que cette technique pouvait être utilisée pour semi-quantifier les mouvements de glaciers rocheux sur de vastes secteurs et parfois les quantifier (sous réserve de respecter certains critères lors de la création et l'analyse des interférogrammes). Dans ce travail de thèse, nous avons acquis des images radar provenant des satellites ERS (datant de 1991 à 1995) et TerraSAR-X (datant de l'été 2012) grâce à l'ESA (agence spatiale européenne) et au DLR (Deutschen Zentrums für Luft und Raumfahrt, agence spatiale Allemande) dans le but de créer et d'analyser des interférogrammes dans le secteur des Alpes françaises. Nous nous intéressons principalement à la détection des glaciers rocheux. Les archives des satellites ERS-1 et 2 nous ont permis de travailler à l'échelle de l'ensemble des Alpes françaises avec une résolution moyenne (25m en géométrie sol). Les données provenant de TerraSAR-X (TSX) ont permis des analyses plus fines grâce à une meilleure résolution en géométrie sol (10m). Cependant le secteur d'études s'est restreint à la Haute Maurienne/Haute Tarentaise à cause du coût des données et du temps nécessaire à la création et à l'analyse des interférogrammes. Pour les données ERS, nous avons pris en compte l'ensemble des images disponibles au-dessus des Alpes françaises et choisi celles adaptées pour notre étude. Finalement 9 interférogrammes ont été créés. Pour analyser ces données deux méthodes ont été employées : i) une analyse visuelle par des géomorphologues dans un SIG (aidée par des ortho-images et des données topographiques) ; ii) une évaluation de la pertinence des résultats par l'intermédiaire d'une comparaison entre les résultats D-InSAR et des inventaires de glaciers rocheux existants. Finalement une carte de l'ensemble des glaciers rocheux détectés dans les Alpes françaises a été produite. Les données TSX ont été analysées sensiblement de la même manière. Enfin, une troisième étude s'est concentrée sur le cas peu commun du glacier rocheux « déstabilisé » de Pierre Brune (mouvements > 2m/an). Une reconstitution historique des déplacements a été réalisée à partir d'images optiques d'archive et des données GPS ont été acquises durant l'été 2012 et comparées aux données obtenues par D-InSAR. En nous basant sur ces études aux résolutions d'image et aux échelles spatiales variées, nous proposons une discussion sur la pertinence de l'utilisation de la méthode D-InSAR pour du « monitoring » à moyen et long termes des glaciers rocheux mais également sur les atouts et inconvénients de la méthode. / Rockglaciers are one the visual expressions of mountain permafrost and have been the focus of numerous and various studies in the last two decades. Two main topics are studied by the scientific community: i) better understanding the movements of active rockglaciers and the phenomena that generate those movements; ii) assessing impact of global atmospheric change on these periglacial shapes. Here we propose to contribute to the first topic by using remote sensing method of displacement measurements already used in other countries for rockglaciers studies: Differential Interferometric Synthetic Aperture Radar (D-InSAR). D-InSAR is a method of measurement based on the phase difference between two radar images, which represent the same area but at different time intervals. The technique generates interferograms, maps of surface deformation in two-dimensions allowing for the detection and quantification (in centimeters) of variations in distance between the target and the radar between two different data acquisitions. Recent research has shown that the InSAR technique can be used to semi-quantify rockglacier deformation (under the assumption that certain conditions are respected with regard to generating and interpreting the interferograms). In the present thesis, ERS radar images (dating from 1991 to 1995) and TerraSAR-X data (dating from summer 2012) were obtained in courtesy of ESA (European Space Agency) and DLR (Deutschen Zentrums für Luft- und Raumfahrt, German Space Agency) with the aim of generating interferograms. We are interested by the detection of rockglacier movements. The ERS archives allowed us to work at French Alps scale with moderate resolution (25m in ground geometry) whereas TerraSAR-X data have better ground resolution (10m) but our analysis are more local (Haute Maurienne/Haute Tarentaise) due to the cost of the data and the time-consuming nature of the analyzes. With ERS, we selected all archives data and chose the more relevant of them. Finally 9 interferograms were generated. To analyse this amount of data two methods were employed: i) a GIS analysis of interferograms by geomorphologists (helped by ortho-images and topographic data), ii) a comparison between interferograms and existing rockglaciers shape inventory to evaluate the quality of the radar detections. At the end of the analysis a map of the French Alps with all detected rockglacier movements was produced. With TerraSAR-X data, the method of analysis was almost the same. Analyzes focused on the Haute Maurienne/Haute Tarentaise massif. A third scale of analysis is focus on Pierre Brune rockglacier which has been detected on ERS interferograms as destabilized rockglaciers (movement > 2 m/year). Further investigations have been carried out on this site (historical movements reconstitution and GPS acquisitions). Based on these studies with different scales and resolutions, we proposed a discussion about suitability of D-InSAR measurements method for long term rockglaciers monitoring as well as drawbacks and benefits of the method.

L’objectif de ce mémoire est d’acquérir une connaissance détaillée sur l’évolution spatiale de la température de surface du sol (GST) au mont Jacques-Cartier et sur la réponse thermique de son îlot de pergélisol alpin aux changements climatiques passés et futurs. L’étude est basée sur un ensemble de mesures de température (GST, sous-sol) et de neige, ainsi que des modèles spatiaux de distribution potentielle de la GST et des simulations numériques du régime thermique du sol. Les résultats montrent que la distribution de la GST sur le plateau est principalement corrélée avec la répartition du couvert nival. Au-dessus de la limite de la végétation, le plateau est caractérisé par un couvert de neige peu épais et discontinu en hiver en raison de la topographie du site et l’action des forts vents. La GST est alors couplée avec les températures de l’air amenant des conditions froides en surface. Dans les îlots de krummholz et les dépressions topographiques sur les versants SE sous le vent, la neige soufflée du plateau s’accumule en un couvert très épais induisant des conditions de surface beaucoup plus chaude que sur le plateau dû à l’effet isolant de la neige. En raison de la quasi-absence de neige en hiver et de la nature du substrat, la réponse du pergélisol du sommet du mont Jacques-Cartier au signal climatique est très rapide. De 1978 à 2014, la température du sol a augmenté à toutes les profondeurs au niveau du forage suivant la même tendance que les températures de l’air. Si la tendance au réchauffement se poursuit telle que prévue par les simulations climatiques produites par le consortium Ouranos, le pergélisol pourrait disparaître d’ici à 2040-2050. / The objective of the study was to acquire detailed knowledge of the spatial evolution of the ground surface temperature (GST) on Mont Jacques-Cartier and the thermal response of its marginal permafrost body to the past and future climate changes. The study is based on temperature (GST, underground) and snow measurements, and spatial modeling of the potential GST distribution and numerical modeling of the ground thermal regime. The result showed that the spatio-temporal variability of the GST over the summit is mainly correlated with the snowpack distribution. On the wind-exposed plateau, the snowpack is thin and discontinuous. The GST is thus closely connected to the very cold air temperature in winter. In the krummholz patches and in the topographic depression on the leeward slope, drifted snow accumulation is significant leading to surface condition warmer than over the wind-exposed plateau. Because of the near snow-free condition of the plateau summit and the highly conductive nature of its bedrock, the response of the permafrost to the climate signal is rapid. From 1978 to 2014, the ground warmed at all depths. If the recent trend continues as predicted by the climate simulations provided by the Ouranos consortium, the complete disappearance of the permafrost body at Mont Jacques-Cartier could occur around 2040-2050.

Basal Temperature of Snow (BTS) measurements were used as the primary inputs to a high resolution (30 x 30 m grid cells) empirical-statistical regional permafrost probability model for the southern and central Yukon, and northernmost British Columbia (59° - 65°N). Data from seven individual study areas distributed across the region were combined using a blended distance decay technique, with an eighth area used for validation. The model predictions are reasonably consistent with previous permafrost maps for the area with some notable differences and a much higher level of detail. The modelling gives an overall permafrost probability of 52%. North of 62°N, permafrost becomes more extensive in the lowland areas whereas farther south permafrost is typically common only above treeline. Significant differences exist between the mountain environments of the Yukon and the Swiss Alps where the BTS method originated and as a result different modelling approaches had to be developed. This work therefore: (1) develops additional explanatory variables for permafrost probability modelling, the most notable of which is equivalent elevation, (2) confirms the use of ground truthing as a requirement for empirical-statistical modelling in the Yukon and (3) uses a combination of models for the region in order to spatially predict between study areas. The results of this thesis will be of use to linear infrastructure route-planning, geohazard assessment and climate change adaptation strategies. Future work employing the model will allow the effects of scenario-based climate warming to be examined.

Basal Temperature of Snow (BTS) measurements were used as the primary inputs to a high resolution (30 x 30 m grid cells) empirical-statistical regional permafrost probability model for the southern and central Yukon, and northernmost British Columbia (59° - 65°N). Data from seven individual study areas distributed across the region were combined using a blended distance decay technique, with an eighth area used for validation. The model predictions are reasonably consistent with previous permafrost maps for the area with some notable differences and a much higher level of detail. The modelling gives an overall permafrost probability of 52%. North of 62°N, permafrost becomes more extensive in the lowland areas whereas farther south permafrost is typically common only above treeline. Significant differences exist between the mountain environments of the Yukon and the Swiss Alps where the BTS method originated and as a result different modelling approaches had to be developed. This work therefore: (1) develops additional explanatory variables for permafrost probability modelling, the most notable of which is equivalent elevation, (2) confirms the use of ground truthing as a requirement for empirical-statistical modelling in the Yukon and (3) uses a combination of models for the region in order to spatially predict between study areas. The results of this thesis will be of use to linear infrastructure route-planning, geohazard assessment and climate change adaptation strategies. Future work employing the model will allow the effects of scenario-based climate warming to be examined.

Basal Temperature of Snow (BTS) measurements were used as the primary inputs to a high resolution (30 x 30 m grid cells) empirical-statistical regional permafrost probability model for the southern and central Yukon, and northernmost British Columbia (59° - 65°N). Data from seven individual study areas distributed across the region were combined using a blended distance decay technique, with an eighth area used for validation. The model predictions are reasonably consistent with previous permafrost maps for the area with some notable differences and a much higher level of detail. The modelling gives an overall permafrost probability of 52%. North of 62°N, permafrost becomes more extensive in the lowland areas whereas farther south permafrost is typically common only above treeline. Significant differences exist between the mountain environments of the Yukon and the Swiss Alps where the BTS method originated and as a result different modelling approaches had to be developed. This work therefore: (1) develops additional explanatory variables for permafrost probability modelling, the most notable of which is equivalent elevation, (2) confirms the use of ground truthing as a requirement for empirical-statistical modelling in the Yukon and (3) uses a combination of models for the region in order to spatially predict between study areas. The results of this thesis will be of use to linear infrastructure route-planning, geohazard assessment and climate change adaptation strategies. Future work employing the model will allow the effects of scenario-based climate warming to be examined.

Basal Temperature of Snow (BTS) measurements were used as the primary inputs to a high resolution (30 x 30 m grid cells) empirical-statistical regional permafrost probability model for the southern and central Yukon, and northernmost British Columbia (59° - 65°N). Data from seven individual study areas distributed across the region were combined using a blended distance decay technique, with an eighth area used for validation. The model predictions are reasonably consistent with previous permafrost maps for the area with some notable differences and a much higher level of detail. The modelling gives an overall permafrost probability of 52%. North of 62°N, permafrost becomes more extensive in the lowland areas whereas farther south permafrost is typically common only above treeline. Significant differences exist between the mountain environments of the Yukon and the Swiss Alps where the BTS method originated and as a result different modelling approaches had to be developed. This work therefore: (1) develops additional explanatory variables for permafrost probability modelling, the most notable of which is equivalent elevation, (2) confirms the use of ground truthing as a requirement for empirical-statistical modelling in the Yukon and (3) uses a combination of models for the region in order to spatially predict between study areas. The results of this thesis will be of use to linear infrastructure route-planning, geohazard assessment and climate change adaptation strategies. Future work employing the model will allow the effects of scenario-based climate warming to be examined.

<p>Examinator Professor Peter Jansson, Institutionen för naturgeografi</p> / Research school for teachers on climate and environment grant 2017-06037

This thesis analyses ice-rich mountain permafrost with regard to its genesis, distribution, deformation and interaction with other environmental factors. The processes influencing ground ice formation in ice-rich and ice-poor mountain permafrost are highlighted. Factors influencing the presence of ice-rich permafrost are identified and their individual or combined effect on frozen ground is determined. Based on these findings, a new permafrost distribution map of Switzerland was created, which specifies permafrost temperature and ice contents and considers rock glacier creep paths. The deformation of rock glaciers is investigated with newly developed monitoring systems and concepts. This enables a better understanding of the processes leading to rock glacier acceleration at different time scales.

This thesis analyses ice-rich mountain permafrost with regard to its genesis, distribution, deformation and interaction with other environmental factors. The processes influencing ground ice formation in ice-rich and ice-poor mountain permafrost are highlighted. Factors influencing the presence of ice-rich permafrost are identified and their individual or combined effect on frozen ground is determined. Based on these findings, a new permafrost distribution map of Switzerland was created, which specifies permafrost temperature and ice contents and considers rock glacier creep paths. The deformation of rock glaciers is investigated with newly developed monitoring systems and concepts. This enables a better understanding of the processes leading to rock glacier acceleration at different time scales.