A series of analogue flume experiments are used to investigate initiation, flow and deposition of static piles of polymict materials, the sorting during transport, and the three dimensional geometry of the resulting deposits. Sequential charges are used to investigate the effects and extent of reworking. The particle heterogeneity is designed to simulate typical PDC make-up, with analogues for juvenile pumice and lithic clasts, as well as the fine-grained pumiceous material which makes up the bulk of the flow. Analogue flume experiments are used to investigate the generation of complex facies variations typical of pyroclastic density current (PDC) deposits. Polymict charges are developed to behave as analogues for the particle size and density contrasts present in PDC (i.e. lithic and juvenile pumice clasts), and investigate the effect of granular sorting during flow on the geometry of deposit architectures. Multiple charges are used to simulate pulses or sequences of separate PDC in order to assess the extent and effects of reworking. 3D visualisation of the resulting deposits reveals stratigraphies analogous to those seen in PDC, including pumice ‘rafting' or over-passing and inverse grading of pumice, and normal grading of lithics by simple gravitational granular sorting. Reworking between differentially-coloured layers makes several complex shear-derived Kelvin-Helmholtz instability features apparent, from fully developed rotational eddies, to less developed recumbent flame structures. The implications for the formation of these in PDC are assessed, including the potential influences on temperature proxy data, radiogenic dating by included phenocrysts (40Ar/39Ar) or charcoals (14C), calculation of eruptive volumes, sedimentation rates and flow velocity.

Cette thèse présente les résultats et conclusions issues d’une série de modèles analogiques de systèmes de compressif à différentes échelles : Les expériences réalisées à l’échelle crustale montrent que la symétrie de structures compressives, de type plis et chevauchements avec 3 niveaux de décollement, est fortement dépendante de la vitesse de sédimentation. Les résultats ont été appliqués au Subandin Bolivien. Les expériences réalisées à l’échelle lithosphérique simulent la subduction et l’extension arrière-arc dans un système subissant une compression parallèle à la marge continent-océan (COB). Ces modèles démontrent que la différence de densité entre les plaques continentales et océaniques est le paramètre clé pour expliquer l'extension arrière-arc: plus petite est la différence de densité, plus faible est l'extension produite. Les résultats ont été appliqués al ‘Anatolie. Enfin, ces modèles ont été utilisés pour tester la reproductibilité et les limites de la modélisation analogique. / This thesis presents the results and conclusions from a series of analogue modelling of deformation in compressive and subduction zones (crustal scale and lithospheric scale) : The experiments carried out at the crustal scale show that the symmetry of compressive structures, folds and trust belts with 3 levels of décollement is strongly dependent on the rate of sedimentation. The results were applied to the Subandin Bolivian. The experiments carried out at the lithospheric scale simulate subduction and back-arc extension in a system under compression parallel to the continent - ocean margin (COB). These models show that the density ratio between the continental and oceanic plates is the key factor to explain the back-arc extension: as smaller the difference in density is, less extension occurred. The results were applied to Anatolia. Finally, these models were used to test the reproducibility and limits for analog modeling.

Les chaînes plissées d'avant-pays sont des objets géologiques qui se forment dans un contexte compressif et représentent la partie externe des orogènes.Elles sont composées de nombreuses structures plissées associées à des chevauchements généralement enracinés au sein d'un niveau de décollement peu profond situé dans la partie basale de la couverture sédimentaire. Ces objets géologiques ont été beaucoup étudiés au XXème siècle.Ils ont été modélisés cinématiquement, analogiquement et mécaniquement.S’il est indispensable de tenir compte de l’aspect mécanique en géologie structurale, les modèles mécaniques restent cependant trop peu utilisés par le géologue structuraliste.L'objectif de cette thèse est de montrer comment apporter des contraintes mécaniques à l'étude des structures géologiques, grâce à une théorie mécanique facilement utilisable en géologie structurale.Cette théorie, le calcul à la rupture (Limit Analysis), représente en effet un bon intermédiaire entre les modèles cinématiques et les modèles mécaniques en éléments finis, très complets mais relativement complexes d'utilisation.Nous étudions ici des exemples appartenant à la chaîne plissée d’avant-pays du Jura et utilisons les logiciels Optum G2 et SLAMTec.Nous procédons alors selon deux approches. La première approche, présentée en première partie de ce manuscrit consiste à étudier la déformation passée.Nous nous focalisons sur la région de Saint-Ursanne dans le Nord-Est du Jura, en Suisse et réalisons dans un premier temps une coupe géologique d'une structure que nous étendons ensuite en 3D via une série de coupes balayant l'ensemble de la zone d'étude.Pour réaliser ces coupes nous utilisons à la fois des principes d’équilibrage cinématique, des modélisations mécaniques par calcul à la rupture et des modélisations analogiques en boîte à sable.Nous montrons ainsi l’importance de l’héritage tectonique sur la mise en place des structures de cette région et modélisons mécaniquement l’influence d’un décalage du niveau de décollement par une faille normale héritée, lors d’un épisode compressif ultérieur.Nous montrons alors que ce décalage représente un générateur de failles ainsi qu’un point d’accroche qui ralenti la propagation du front de déformation, contrôlant par ailleurs la direction de la structure qui se développera.Nous montrons également que la topographie joue un rôle majeur dans la mise en place des structures géologiques.Dans la deuxième partie du manuscrit nous abordons une seconde approche qui consiste à étudier la déformation actuelle.Nous changeons cette fois-ci d’échelle pour nous intéresser à la tectonique actuelle de l’ensemble du Jura.Nos modélisations prédisent que cette chaîne plissée, formée par une tectonique de couverture, est aujourd’hui partiellement affectée par une tectonique de socle.Alors que la partie ouest du Jura serait toujours affectée par une tectonique de couverture, le socle devrait aujourd’hui être impliqué dans la déformation au niveau de la partie nord-est.Nous montrons cependant qu’une tectonique de socle n’empêche pas l’activation simultanée des niveaux de décollements situés dans la couverture (les évaporites triasiques, dans le cas du Jura).L’intérêt pratique de ces méthodes est illustré par des questions industrielles : le dernier chapitre présente une étude des coupes géologiques d’un terrain d’étude de la Nagra, situé dans l’est du Jura.Nous regardons la réponse mécanique actuelle en compression de l’interprétation structurale proposée.Les outils numériques utilisés dans l’ensemble de ce travail ont, par ailleurs, fait l’objet d’un tutoriel réalisé dans le cadre d’une mission au sein de l’entreprise Total. / Fold-and-thrust-belts form in a compressive geological context and represent the external part of orogens.They are composed of numerous folds and thrusts generally rooted in a shallow décollement level located in the basal part of the sedimentary cover.These belts have been studied extensively during the XX century. They have been modeled kinematically, analogically and mechanically.The mechanical aspect of deformation must be taken into account in structural geology, but yet, mechanical models remain underused by the structural geologist.The objective of this thesis is to show how to bring mechanical constraints to the study of geological structures with a mechanical theory easily usable in structural geology, the Limit Analysis.This theory represents a good compromise between the kinematic models and the finite element mechanical models, very complete but relatively complicated to use.We study here examples from the Jura fold-and-thrust belt and use the softwares Optum G2 and SLAMTec.We proceed with two different approaches.The first one is presented in the first part of this manuscript and consists in studying the past deformation.We focus on Saint-Ursanne area, in the North-East Switzerland Jura.We construct first a 2D geological cross section of the Mont Terri structure and then we extend this work in 3D by a series of cross sections that scan the whole Saint Ursanne area.These cross-sections are drawn using kinematical rules, mechanical modeling with Limit Analysis and analog sandbox modeling.We show the importance of the tectonic inheritance on the onset of structures in the studied region and we mechanically model the influence of a décollement offset due to an inherited normal fault during a subsequent compressive episode.We show that this offset represents a “fault generator” and a catching point that slows the propagation of the deformation front and furthermore controls the orientation of the structure that develops.We also show that the topography plays a major role in the establishment of geological structures.In the second part of the manuscript we follow a second approach that consists in studying the current deformation.We change of scale and study the current tectonics of the entire Jura.Our models predict that this belt, formed in thin-skin tectonics, is currently partially affected by thick-skin tectonics.The western Jura would be still affected by a thin-skin tectonics, but the basement should be involved in the deformation in the north-east Jura.However, we show that a thick-skin tectonics does not preclude the simultaneous activation of the shallower décollements contained by the sedimentary cover (Triassic evaporites, in the Jura).Practical interest of these methods is illustrated by industrial questions: the last chapter presents a study of geological cross-sections from a studied area of Nagra in the eastern Jura.We look at the current mechanical answer under compression of the proposed structural interpretations.The numerical tools used in this work have been the subject of a tutorial made during a mission at Total company.

Analogue modelling of strike-slip faulting provides insight into the development and behaviour of surface ruptures with accumulated slip, with relevance for understanding how information recorded in paleoseismic trenches relates to the earthquake behaviour of active faults. Patterns of surface deformation were investigated in analogue experiments using cohesive and non-cohesive granular materials above planar strike-slip basement faults. Surface deformation during the experiments was monitored by 3D PIV (particle image velocimetry) and 2D time lapse photography. Analysis focused on fault zone morphology and development, as well as the relationship of the models to surface deformation observed at the Greendale Fault that resulted from the 2010 Darfield earthquake. Complex rupture patterns with similar characteristics to the Greendale Fault (e.g. en echelon fractures, Riedel shears, pop-up structures, etc.) can be generated by a simple fault plane of uniform dip, slip, and frictional properties. The specific structures and the style of their development are determined by the properties of the overburden and the nature of the material surface. The width of the zone of distributed deformation correlates closely with sediment thickness, while the width of discrete fracturing is controlled by the material properties as well as the thickness of the overburden. The overall deformation zone width increases with the growth of initial, oblique fractures and subsequently narrows with time as strain localizes onto discrete fractures parallel to the underlying basement fault. Mapping the evolution of fracture patterns with progressive strain reveals that Riedel shears, striking at 90-120° (underlying fault strike = 90°) are more frequently reactivated during multiple earthquake cycles, and are thus most likely to provide reliable paleoseismic records. This will help identify suitable locations for paleoseismic trenches and interpret trench records on the Greendale Fault and other active, strike-slip faults in analogous geologic settings. These results also highlight the tendency of trenching studies on faults of this type to underestimate the number and displacement of previous ruptures, which potentially leads to an underestimate of the magnitude potential and recurrence interval of paleoearthquakes.

Superimposition of contractional structures is widely observed in different scales in the world. Superimposed structures form due to different processes: change in strain accommodation from one type of structure to another during a single progressive shortening; successive coaxial shortening phases separated by an unconformity; superimposition of different non-coaxial shortening phases. Using results of a series of systematic analogue models and detailed field structural mapping, this thesis focuses on the geometry and kinematics of such superimposed structures that are formed by these three processes. During a single progressive folding, thrusts develop within a fold to accommodate stain variations in different regime of the fold. Limited displacement along these thrusts does not significantly modify the geometry of the fold. However, during multiple shortening phases (coaxial or non-coaxial), early formed structures are modified by the later phase ones. The later thrusts can cut and displace the pre-existing structures. The early folds are tightened or interfered by the later folding phase. Pre-existing thrusts may be reactivated either in dip direction and/or along strike during the later shortening. The pre-existing structures in turn influence development of the later structures, which results in change in structure spacing. An angular unconformity between two shortening phases clearly truncates the early phase structures and separates structures of different levels. Unlike in the post-erosional layers, in the layers below the unconformity, complicated superimposed structures are visible. This thesis shows that geometry and sequence of structures formed during one progressive shortening or multiple shortening phases strongly depend on the mode of the superimposition (coaxial, orthogonal or oblique) and the orientation of pre-existing structures.

The Grängesberg Mining District (GMD) is located in the western part of the Bergslagen province in south central Sweden, and is the only known apatite-iron oxide mineralization south of Norrbotten. The Grängesberg deposit is the largest of the mineralizations along GMD and consists of a line of steeply to moderately dipping 20-100 m wide lenses extending to a depth of 1.7 km. It is hosted by felsic volcanic rocks that are phyllosilicate altered in the vicinity of the ore. Field relationships indicate that a competent granitoid was structurally emplaced on top of the ore during D2, and that the less competent phyllosilicate-rich host rocks accommodated large parts of the strain. These D2 structures described from the area around the ore bodies in Grängesberg include stretching lineation at the tapering edges of the lenses, asymmetric folds with opposite vergence along strike, sheath folds and possibly fold interference patterns. These structures are proposed to have formed in response to the competence contrast and the reverse movements on a large scale. The hypothesis is that the competence contrast between the competent ore and granitoid and the less competent host rocks could explain the formation and location of the key D2-structures in Grängesberg. To test the hypothesis, four analogue tectonic models were run in the centrifuge at the Hans Ramberg Tectonic Laboratory (HRTL), Uppsala University. The competent bodies were represented by an acrylic glass wedge with two lenses corresponding to the ore lenses, with an inclination of either 60° or 45° mimicking the approximate dip of the ore. Plasticine was used as the less competent host rocks. On the top surface, circles and squares were printed and used as strain markers. The models were run at c. 300-400 G until penetrative deformation and shortening ranging from 32.5 to 39.2% was reached. During centrifuging, the wedge indented the plasticine, resembling the possible deformation during thrusting of the deep granitoid. The models were then cut to reveal the structures formed, and one model was digitalized using the software Move. After shortening, the strain markers had deformed to show a change in direction of strain around the lenses. Vertical sections perpendicular to the shortening direction showed that stretching lineation developed between the lenses. Horizontal sections revealed asymmetric non-cylindrical folds with opposite vergence along strike and fold interference patterns. These results show that the observed field relationships were reproducible with the model setup, and that the competence contrast between the ore bodies, the granitoid and the host rocks could be the controlling factor for localization of shear and sites of stretching in the area.

Applying the anisotropy of magnetic susceptibility to analogue models provides detailed insights into the strain distribution and quantification of deformation within contractional tectonic settings like fold-and-thrust belts (FTBs). Shortening in FTBs is accommodated by layer-parallel shortening, folding, and thrusting. The models in this research reflect the different deformation processes and the resulting magnetic fabric can be attributed to thrusting, folding and layer-parallel shortening. Thrusting develops a magnetic foliation parallel to the thrust surface, whereas folding and penetrative strain develop a magnetic lineation perpendicular to the shorting direction but parallel to the bedding. These fabric types can be observed in the first model of this study, which simulated a FTB shortened above two adjacent décollements with different frictional properties. The different friction coefficients along the décollements have not only an effect on the geometric and kinematic evolution of a FTB, but also on the strain distribution and magnitude of strain within the belt.  The second series of models performed in this study show the development of a thrust imbricate and the strain distribution across a single imbricate in more detail. Three models, with similar setup but different magnitudes of bulk shortening, show strain gradients by gradual changes in principal axes orientations and decrease in degree of anisotropy with decreasing distance to thrusts and kinkzones. These models show that at the beginning of shortening, strain is accommodated mainly by penetrative strain. With further shortening, formation of thrusts and kinkzones overprint the magnetic fabric locally and the degree of anisotropy is decreasing within the deformation zones. At thrusts, an overprint of the magnetic fabric prior deformation towards a magnetic foliation parallel to the thrust surfaces can be observed. A rather complex interplay between thrusting and folding can be analysed in the kinkzones. In general, this thesis outlines the characteristics of magnetic fabric observed in FTBs, relates different types of magnetic fabric to different processes of deformation and provides insights into the strain distribution of FTBs.

Non renseigné / Most mega-earthquakes (i.e. earthquakes with Mw ≥ 8.5) occur along subduction mega-thrusts, the interfaces between the subducting - and the overriding plates in convergent margins. These events may have catastrophic impact on human societies due to their destructive potential. For this reason being able to predict the timing and size of these earthquakes became one goal of the international scientific community. The subduction seismic cycle is influenced by many different parameters. The interplay between these parameters governing the frequency and size of megathrust earthquakes still remains unclear, mainly due to the short (i.e. limited to the last century) seismic record.The seismogenic part of the subduction thrust fault spans between depths of 11±4 and ± 51 km (Heuret et al. 2011). In this zone a combination of temperature, pressure and rocks characteristics creates conditions favourable for seismic behaviour. Whether a specific area in the subduction thrust fault has the ability to trigger mega-earthquakes can be expressed using the degree of seismic coupling, i.e. the amount of slip that occurs with respect to the total amount of plate convergence (e.g. Scholz 1998; Scholz & Campos 2012). When a fault is fully coupled, all of the fault slip occurs during earthquakes instead of also during aseismic behaviour (e.g. slow slip events). The internal structure of the interplate fault zone mainly determines whether an area within a subduction zone behaves seismic or aseismic (Wang & Bilek 2011). This is influenced by the topography of the plate interface (e.g. subducting seamounts; Wang & Bilek 2014), but also subducted sediments and fluids in the subduction channel may play an important role.The main goal of this project is to understand which parameters affect the behaviour of mega-earthquake ruptures. This will be done by comparing natural data (e.g. seafloor roughness, sediment thickness and fluid content in the subduction channel) to rupture characteristics of major recent earthquakes. With this analysis also more knowledge can be gained on the triggering of slow earthquakes instead of mega-earthquakes. These are slow slip events with lower frequencies and longer durations than ‘regular’ earthquakes (Saffer & Wallace 2015).The database of natural data, implemented by the long-term scientific joint venture between the Univ. Montpellier and the LET (Roma Tre) will be used for the analysis. Ongoing work is done on determining a method for estimating the seafloor roughness, i.e. the distribution of high, low and smooth areas (by Michel Peyret in collaboration with Serge Lallemand, Univ. Montpellier). Also data is available on the trench sediment thickness around the world (Heuret et al. 2011). In the frame of this project, information on the roughness of the seafloor will be added to the database. In addition the rupture characteristics of major recent earthquakes will be collected. By performing a multiparametric statistical analysis of the database, a conceptual model will be realized, exploring the possible link between all the different parameters. The aim is to validate this model in the lab using scaled 3D analogue models. This will be done both at the LET and at Univ. Montpellier by using a broad range of geometries and contact materials with different rheologies (e.g. gelatin, foam rubber and a new analogue material; Caniven et al. 2015; Corbi et al. 2013). This jointed experimental approach with both the Univ. Montpellier and the LET involved creates a rich environment where differences and similarities of the two different approaches can be used to validate the results.

This research project on volcanic debris avalanches aims to provide a better understanding of the influence of the volume, fall height and topography on the deposit location and morphology. This will enable improvements in delineation of the areas at risk from volcanic debris avalanches, and improvements in management of a disaster should it occur. Undertaken to fulfil the requirements for a double degree (Geological Engineering and MSc in Hazard and Disaster Management) this work is the result of a collaboration between Polytechnic Institute LaSalle-Beauvais in France and the University of Canterbury in New Zealand. Following a brief introduction to the topic, statistical analyses of volcanic debris avalanche deposits are undertaken. Multiple variables analyses (Principal Components Analyses and Regressions) were carried out using a database of 298 volcanic debris avalanches derived from modification of Dufresne’s recent database. It was found that the volume has the main influence on the deposits rather than the fall height; the latter seems to have greater effect on avalanches of small volume. The topography into which the deposit is emplaced mainly determines its geometrical characteristics. These statistical results were compared with the results of laboratory-scale analogue modelling. A model similar to that used by Shea in 2005 provided data indicating similar trends of the influence of volume, fall height and topography on mass movement deposits at all scales. The final aspect of this project was a numerical simulation of a large debris avalanche from the north flank of the Taranaki volcano in the direction of the city of New Plymouth. The numerical code VolcFlow developed by Kelfoun in 2005 was used, after being tested against the laboratory experiments to verity its accuracy. The simulations showed that the Pouaki range protects the city of New Plymouth form major impacts from Taranaki collapses, but also indicated some potential problems with the hazard zoning and evacuation zones presently in place.

This Thesis reports on the integration of different modelling techniques to construct a unified and better constrained conceptual model of structural evolution of the Pico del Águila anticline (External Sierras, Southern Pyrenees, Spain). The structure is a well-known example of detachment fold, which exhibits a N-S structural trend, parallel to the direction of tectonic transport in the Southern Pyrenees. Based on field observations of an unevenly distributed Triassic décollement, analogue modelling show how to generate orogen-perpendicular structures which may result in transverse anticlines. Numerical models investigate the effect of a complex mechanical stratigraphy, characterized by an interlayering of competent and incompetent layers, plus syn-kinematic sedimentation in the fold growth. Based on field data and seismic interpretations, a 3D reconstruction and sequential geomechanical restoration of the Pico del Águila anticline reports the coexistence of multiple folding mechanisms occurring simultaneously in different units and structural domains of the fold, leading to a complex strain pattern that can not be assessed by simplistic kinematic 2D approaches. By integrating the presented models with the previous data in the region, we discuss about the benefits and drawbacks of each modelling technique and present an integrated model of structural evolution for the Pico del Águila anticline. This brings a better comprehension of the structure as well as the processes that drove the evolution of the N-S detachment anticlines in the External Sierras of the Southern Pyrenees. / L’anticlinal del Pico del Águila és un conegut exemple de plec de desenganxament amb sedimentació marina a fluvio-deltaica associada, amb una tendència estructural N-S, paral•lela a la direcció de transport tectònic dels Pirineus Meridionals. Basat en observacions de camp que indiquen una distribució heterogènia del nivell de desenganxament Triàsic, els models analògics mostren el procés de generació d’estructures perpendiculars al sistema orogènic que poden donar lloc als anticlinals N-S descrits a les Sierras Exteriores Aragonesas. Els models numèrics investiguen l’efecte d’una estratigrafia mecànica complexa, caracteritzada per una intercalació d’unitats competents i incomptetents (amb marcades diferències en el grau de competència, per tant) i de la sedimentació sin-cinemàtica en el creixement i evolució de l’anticlinal. Basat en dades de camp i interpretacions de perfils sísmics s’ha portat a terme una reconstrucció i restitució geomecànica tridimensional de l’anticlinal del Pico del Águila. D’aquestes se’n deriva la coexistència de de múltiples mecanismes de plegament actuant simultàniament en diferents unitats i dominis estructurals, la qual implica al seu torn un patró i distribució de la deformació que no poden ser avaluats mitjançant aproximacions o tècniques de modelització cinemàtiques o/i bidimensionals. S’integren també els resultats obtinguts a partir de les esmentades tècniques de modelització amb les dades i coneixements previs de la regió, es discuteixen els beneficis, desavantatges i limitacions de cadascuna d’aquestes tècniques de modelització, i es presenta un model integrat d’evolució estructural del Pico del Águila. Aquest anàlisi crític dels resultats i aquest esforç d’integració porten sense dubte cap a una millor comprensió de l’estructura i dels processos que menaren l’evolució dels plecs de desenganxament N-S de les Sierras Exteriores Aragonesas dels Pirineus Meridionals.