No description available.

L'objectif de cette thèse est de comprendre la formation, la structuration et les processus de réactivation d'une chaîne de montagne intracontinentale atypique : les Longmen Shan, situés dans la province du Sichuan, en Chine. Localisés à la limite entre le craton du Yangtze où s'est déposé le bassin du Sichuan (au Sud-Est) et le bloc du Songpan Garze appartenant au plateau tibétain (au Nord-Ouest), les Longmen Shan se sont majoritairement structurés lors de l'orogénèse indosinienne, à la fin du Trias et ont ensuite subit plusieurs réactivations.Cette chaîne de montagne est donc un endroit privilégié pour étudier la réactivation et l'héritage structural et thermique d'une structure intracontinentale, en relation étroite avec la formation et l'évolution du plateau tibétain. Tout d'abord, pour contraindre les paramètres crustaux, une imagerie crustale détaillée le long d'une coupe à travers cette chaîne est proposée. Une réseau sismologique de 35 stations a été déployé pendant plus de 2 ans le long d'un profil dense. La technique des fonctions récepteurs a été appliquée et les données gravimétriques ont été utilisées conjointement pour renforcer l'imagerie obtenue. Un saut de Moho abrupt de 20km a été imagé, entre une croûte tibétaine épaisse d'environ 63km et la croûte du craton du Yangtze , épaisse de 45km. Ce résultat traduit la confrontation de deux lithosphères d'épaisseurs et de propriétés physiques contrastées. Les rapports Vp/Vs ainsi que les mesures d'anisotropie crustale et mantellique ont montré l'absence d'une zone à faible vitesse ou d'une zone de fluage important au sein de la croûte, ce qui réfute les modèles de déformation de la croûte tibétaine impliquant un chenal de déformation au sein de la croûte tibétaine. L'imagerie crustale a donc mis en évidence un important contraste à l'échelle lithosphérique. Le second axe de ce travail a consisté à étudier cette région à plus long terme en menant une étude stratigraphique, tectonique et métamorphique afin de déduire l'importance de l'héritage géologique dans sa structuration actuelle. Dès le début du Paléozoïque, la marge passive qui sera ensuite inversée présentait déja probablement une variation abrupte de l'épaisseur crustale. Un premier contraste d'épaisseur lithosphérique au niveau des Longmen Shan se situaient donc à la limite entre deux domaines paléogéographiques différents. A la fin du Trias, lors de la fermeture de la Paléotéthys, l'épais prisme sédimentaire du Songpan Garze a débordé sur la marge passive de la bordure Ouest du craton du Yangtze, dans la région des Longmen Shan. Cependant, il n'y a aucune évidence de subduction dans cette chaîne et le métamorphisme associé à cette phase de déformation correspond à des moyennes températures (jusqu'à plus de 600°C) pour des pressions relativement modestes. Les données métamorphiques ont montré un pic de pression (relativement faible, inférieur à 8kbar) suivi d'un pic de température pouvant conduire à une migmatisation associée à une exhumation variable en fonction de la localisation au sein de la chaîne. Les variations latérales de l'exhumation sont interprétées comme directement associées à la dynamique de la mise en place de la nappe du Songpan Garze sur la marge Ouest du craton du Yangtze. L'apex des Longmen Shan correspond donc au front de la nappe du Songpan Garze et délimite deux domaines structuraux et métamorphiques contrastés.Cette étude met en évidence une phase de réactivation à la fin du Crétacé de la chaîne, probablement associée à la rotation dans le sens horaire du craton du Yangtze. Enfin, la dernière phase de déformation affectant les Longmen Shan est une répercussion de la collision entre l'Inde et l'Eurasie qui fini de structurer cette chaîne.Nous avons donc montré qu'une limite paléogéographique majeure, héritée d'une structure en marge passive transtensive peut subir le débordement de nappes sédimentaires provenant d'un prisme distant de très grande taille. Ce débordement a provoqué une inversion de relief et un surépaississement crustal en conséquence de la superposition de ces épaisses nappes. Une fois cette limite tectonique formée, la région va subir plusieurs réactivations liées à des déformations annexes comme l'orogénèse Yanshanienne ou la collision entre l'Inde et l'Eurasie. Cette chaîne est encore active aujourd'hui comme l'a démontré le séisme du Sichuan du 12 Mai 2008 qui a eu lieu dans les Longmen Shan avec des caractéristiques atypiques. / The aim of this study is to understand formation, evolution and reactivation processes of an atypic intracontinental mountain range : the Longmen Shan (Sichuan, China). The Longmen Shan are located at the boundary between the Yangtze craton and the tibetan crust and were mostly formed during the indosinian orogeny, at the end of the Triassic. After this orogeny, the Longmen Shan were reactivated by several deformation phases. This mountain range is a key area to study reactivation processes and structural and thermal inheritance. Furthermore, the Longmen Shan tectonic history is linked with the formation and the evolution of the Tibetan Plateau.At first, a detailed seismological imagery was performed along a cross-section through the Longmen Shan, passing by the epicenter of the Sichuan earthquake. 35 seismological stations were deployed during more than 2 years with a small interstation distance. Using the receiver function method and gravimetric data, a steep Moho step of about 20km between the 63km-thick Tibetan crust and the 45km-thick Yangtze craton was imaged. This geometry is the result of the confrontation between the thick and soft tibetan lithosphere abutting the resistant Yangtze lithosphere. Vp/Vs ratio and crustal and mantellic anisotropic measurements indicate that there is no extensive zone of partial melting inside the tibetan crust, which is in disagreement with models that proposed the occurence of a channel flow inside the crust. The second part of this word was focused on the long-term study of the deformation in the Longmen Shan using stratigraphic, tectonic and metamorphic data. This part highlights the importance of the geological inheritance in the present-day crustal geometry of the mountain range. Since the beginning of the Paleozoïc, the Western passive margin of the Yangtze crust was probably already associated with a crustal thickness step, as a consequence of the transtensive openning context. The Longmen Shan were located at a paleogeographic boundary. At the end of the Triassic, the closure of the Paleotethys is responsible of the formation of the thick Songpan Garze accretionary wedge which overflowed on the Western part of the Yangtze craton margin, in the Longmen Shan area. There is no evidence of subduction in this belt and the associated metamorphism consists of middle temperatures and relatively low pressures. Metamorphic data indicates that a pressure maximum (lower than 8kbar) was followed by a temperature maximum which could be associated with partial melting, as observed in Danba metamorphic complexe. Lateral variations of the recorded exhumation are interpreted as a consequence of the dynamics of the setting up of the Songpan Garze nappe on the Yangtze craton margin. The Longmen Shan are located at the front of the Songpan Garze nappe and marks a transition zone between two contrasted tectonic and metamorphic provinces.This study highlights a cretaceous reactivation phase probably associated with the clockwise rotation of the Yangtze craton. At the end, the last deformation phase is a consequence of the India/Eurasia collision.The Longmen Shan were a major paleogeographic boundary at the end of the Paleozoïc which were subject to the overflow of the thick Songpan Garze accretionary wedge. This overflow is responsible of a relief inversion and a crustal thickenning as a consequence of superposition of the sedimentary pile. After this episode, the region will be reactivated by the distant Yanshanian and the Himalayan orogenies.

Orogenic belts are fundamental features of plate tectonics. The crustal structure of orogens around the world shows a wide range of deformation styles from narrow, asymmetric wedges like the Pyrenees to wide, plateau-like orogens such as the Zagros or the Himalaya. The primary controlling factor on the size and structure of an orogen is the amount of convergence between the colliding plates. However, there are important additional factors providing major controls on the structural development of orogens. Among the potential parameters that can affect the style of deformation are the crustal strength, inherited weaknesses, and the surface processes. These parameters have been studied extensively in the past but their relative importance remains unclear. The aim of this thesis is to improve our understanding of: (1) How surface processes affect mountain building, with a special focus on the relationship between thin-skinned foreland and thick-skinned internal deformation of orogens. (2) How inherited extensional structures affect mountain building. The study was carried out using the Pyrenees as a special reference case. To answer our research questions we have used a wide range of state-of-the-art numerical modelling tools. In paper 1 we present a new method where we couple a structural-kinematic model and a thermo-kinematic model to evaluate the consistency of existing balanced section reconstructions with independent thermochronology data. In papers 2 and 3 we use 2D lithospheric scale thermo-mechanical models with surface process algorithms. Using the above toolset, we demonstrate that syntectonic sedimentation results in longer basement thrust sheets as well as longer thin-skinned thrust sheets and a generally wider orogen. Conversely erosion tends to narrow the wedge and reduce the orogenic loading of the colliding plates, limiting the space available for deposition in the flexural foreland deeps. We also demonstrate that inherited extensional structures play a crucial role in mountain building as they facilitate the migration of deformation into the undeformed basement of the overriding plate. Moreover, a significant amount of lower-crustal/mantle-lithospheric material is preserved at shallow depths only in the presence of extensional inheritance, but significant erosion is needed in order to bring this material to the surface. Our models also show that thin-skinned thrust sheets are generally rooted in the footwall of basement thrusts as they form outward-propagating sequences. As soon as a new basement thrust forms, the thin-skinned sequence situated on top of the new basement thrust-sheet is abandoned in favour of starting a new sequence in the footwall of the new thrust. Regarding our case study, it was possible to reproduce the section restoration using a structural-kinematic model with high accuracy up to the 36-Ma time slice and with limited accuracy up to the 50-Ma time slice. The thermochronometric ages predicted by the thermo-kinematic modelling are generally in good agreement with both the high- and low-temperature thermochronology data available in the Central Pyrenees; hence we conclude that the restoration is to a first order consistent with these datasets. The predicted thermochronological ages approximate the available low-temperature thermochronology data better by taking into account the late-stage burial and re-excavation scenario affecting the southern flank of the Pyrenean wedge presented by Coney et al. (1996), and quantified by Fillon and van der Beek (2012). In conclusion, our model experiments suggest, that extensional inheritance played a prime role in the structural evolution of the Pyrenees, with the major characteristics of the North Pyrenean Unit, including the presence of steep, inverted normal faults, the relative tectonic quiescence of the area after the early inversion and the presence of a lower-crustal body at shallow depth below the unit, best recaptured by our accordion models. / Orogenic belts are fundamental features of plate tectonics. The crustal structure of orogens around the world shows a wide range of deformation styles from narrow, asymmetric wedges like the Pyrenees to wide, plateau-like orogens such as the Zagros or the Himalaya. The primary controlling factor on the size and structure of an orogen is the amount of convergence between the colliding plates. However, there are important additional factors providing major controls on the structural development of orogens. Among the potential parameters that can affect the style of deformation are the crustal strength, inherited weaknesses, and the surface processes. These parameters have been studied extensively in the past but their relative importance remains unclear. The aim of this thesis is to improve our understanding of: (1) How surface processes affect mountain building, with a special focus on the relationship between thin-skinned foreland and thick-skinned internal deformation of orogens. (2) How inherited extensional structures affect mountain building. The study was carried out using the Pyrenees as a special reference case. To answer our research questions we have used a wide range of state-of-the-art numerical modelling tools. In paper 1 we present a new method where we couple a structural-kinematic model and a thermo-kinematic model to evaluate the consistency of existing balanced section reconstructions with independent thermochronology data. In papers 2 and 3 we use 2D lithospheric scale thermo-mechanical models with surface process algorithms. Using the above toolset, we demonstrate that syntectonic sedimentation results in longer basement thrust sheets as well as longer thin-skinned thrust sheets and a generally wider orogen. Conversely erosion tends to narrow the wedge and reduce the orogenic loading of the colliding plates, limiting the space available for deposition in the flexural foreland deeps. We also demonstrate that inherited extensional structures play a crucial role in mountain building as they facilitate the migration of deformation into the undeformed basement of the overriding plate. Moreover, a significant amount of lower-crustal/mantle-lithospheric material is preserved at shallow depths only in the presence of extensional inheritance, but significant erosion is needed in order to bring this material to the surface. Our models also show that thin-skinned thrust sheets are generally rooted in the footwall of basement thrusts as they form outward-propagating sequences. As soon as a new basement thrust forms, the thin-skinned sequence situated on top of the new basement thrust-sheet is abandoned in favour of starting a new sequence in the footwall of the new thrust. Regarding our case study, it was possible to reproduce the section restoration using a structural-kinematic model with high accuracy up to the 36-Ma time slice and with limited accuracy up to the 50-Ma time slice. The thermochronometric ages predicted by the thermo-kinematic modelling are generally in good agreement with both the high- and low-temperature thermochronology data available in the Central Pyrenees; hence we conclude that the restoration is to a first order consistent with these datasets. The predicted thermochronological ages approximate the available low-temperature thermochronology data better by taking into account the late-stage burial and re-excavation scenario affecting the southern flank of the Pyrenean wedge presented by Coney et al. (1996), and quantified by Fillon and van der Beek (2012). In conclusion, our model experiments suggest, that extensional inheritance played a prime role in the structural evolution of the Pyrenees, with the major characteristics of the North Pyrenean Unit, including the presence of steep, inverted normal faults, the relative tectonic quiescence of the area after the early inversion and the presence of a lower-crustal body at shallow depth below the unit, best recaptured by our accordion models.