We investigate upper crustal anisotropy of the Endeavour Segment of the Juan de Fuca Ridge using shear wave splitting measurements of ~3000 earthquakes recorded during three years using the Keck seafloor seismic network. We apply a new cluster analysis of shear-wave splitting measurements to our database. The methodology reduces the use of subjective criteria and improves the accuracy of measurements in the presence of noisy data. Fast polarization directions at a given seismic station are constant and stable during the deployment; however, fast-polarization directions between stations vary significantly. We presume that the lack of consistency of shear wave splitting among seismic stations reflects the spatial distribution of anisotropy in the vicinity of the ridge axis. We infer that the variation of fast polarization directions and delay times is caused by spatial variations in shallow hydrogeological structures and the stress field. Local faults and fissures are unlikely to be the primary cause of this anisotropy since most of the fast polarization directions are not consistent with the ridge parallel trend of faults. Stress perturbations induced by magmatic injection into the axial magma chamber or spatial variation in the rates of a hydrothermal heat transfer may contribute to the observed heterogeneity in seismic anisotropy.

The influence of elastic anisotropy on seismic wave propagation is often neglected for the sake of simplicity. However, ignoring anisotropy may lead to significant errors in the processing of seismic data and ultimately in a poor image of the subsurface. This is especially true in wide-aperture Vertical Seismic Profiles where waves travel both vertically and horizontally. Anisotropy has been neglected in wavefront construction methods of seismic ray-tracing until Gibson (2000), who showed they are powerful tools to simulate seismic wave propagation in three-dimensional anisotropic subsurface models. The code is currently under development using a C++ object oriented programming approach because it provides high flexibility in the design of new components and facilitates debugging and maintenance of a complex algorithm. So far, the code was used to simulate propagation in homogeneous or simple heterogeneous anisotropic velocity models mainly designed for testing purposes. In particular, it has never been applied to simulate a field dataset. We propose here an analytical method involving little algebra and that allows the design of realistic heterogeneous anisotropic models using the C++ object oriented programming approach. The new model class can model smooth multi-layered subsurface with gradients or models with many dip variations. It has been used to model first arrival times of a wide-aperture VSP dataset from the Gulf of Mexico to estimate the amount of anisotropy. The proposed velocity model is transversely isotropic. The anisotropy is constant throughout the model and is defined via Thomsen's parameters. Values in the final model are epsilon = 0.055 and delta = -0.115. The model is compatible with the a priori knowledge of the local geology and reduces the RMS average time difference between measured and computed travel times by 51% in comparison to the initial isotropic model. These values are realistic and are similar to other measurements of anisotropy in the Gulf of Mexico.

The influence of elastic anisotropy on seismic wave propagation is often neglected for the sake of simplicity. However, ignoring anisotropy may lead to significant errors in the processing of seismic data and ultimately in a poor image of the subsurface. This is especially true in wide-aperture Vertical Seismic Profiles where waves travel both vertically and horizontally. Anisotropy has been neglected in wavefront construction methods of seismic ray-tracing until Gibson (2000), who showed they are powerful tools to simulate seismic wave propagation in three-dimensional anisotropic subsurface models. The code is currently under development using a C++ object oriented programming approach because it provides high flexibility in the design of new components and facilitates debugging and maintenance of a complex algorithm. So far, the code was used to simulate propagation in homogeneous or simple heterogeneous anisotropic velocity models mainly designed for testing purposes. In particular, it has never been applied to simulate a field dataset. We propose here an analytical method involving little algebra and that allows the design of realistic heterogeneous anisotropic models using the C++ object oriented programming approach. The new model class can model smooth multi-layered subsurface with gradients or models with many dip variations. It has been used to model first arrival times of a wide-aperture VSP dataset from the Gulf of Mexico to estimate the amount of anisotropy. The proposed velocity model is transversely isotropic. The anisotropy is constant throughout the model and is defined via Thomsen's parameters. Values in the final model are epsilon = 0.055 and delta = -0.115. The model is compatible with the a priori knowledge of the local geology and reduces the RMS average time difference between measured and computed travel times by 51% in comparison to the initial isotropic model. These values are realistic and are similar to other measurements of anisotropy in the Gulf of Mexico.

This dissertation presents the development of a method for quantitative integration of seismic (elastic) anisotropy attributes with reservoir performance data as an aid in characterization of systems of natural fractures in hydrocarbon reservoirs. This new method incorporates stochastic Discrete Feature Network (DFN) fracture modeling techniques, DFN model based fracture system hydraulic property and elastic anisotropy modeling, and non-linear inversion techniques, to achieve numerical integration of production data and seismic attributes for iterative refinement of initial trend and fracture intensity estimates. Although DFN modeling, flow simulation, and elastic anisotropy modeling are in themselves not new technologies, this dissertation represents the first known attempt to integrate advanced models for production performance and elastic anisotropy in fractured reservoirs using a rigorous mathematical inversion. The following new developments are presented: . • Forward modeling and sensitivity analysis of the upscaled hydraulic properties of realistic DFN fracture models through use of effective permeability modeling techniques. . • Forward modeling and sensitivity analysis of azimuthally variant seismic attributes based on the same DFN models. . • Development of a combined production and seismic data objective function and computation of sensitivity coefficients. . • Iterative model-based non-linear inversion of DFN fracture model trend and intensity through minimization of the combined objective function. This new technique is demonstrated on synthetic models with single and multiple fracture sets as well as differing background (host) reservoir hydraulic and elastic properties. Results on these synthetic control models show that, given a well conditioned initial DFN model and good quality field production and seismic observations, the integration procedure results in convergence of both fracture trend and intensity in models with both single and multiple fracture sets. Tests show that for a single fracture set convergence is accelerated when the combined objective function is used as compared to a similar technique using only production data in the objective function. Tests performed on multiple fracture sets show that, without the addition of seismic anisotropy, the model fails to converge. These tests validate the importance of the new process for use in more realistic reservoir models.

This dissertation explores subduction zone-related deformation both on short time scales in the form of subduction zone earthquakes and over larger time and geographical scales in the form of subduction rollback or detachment of the subducting lithosphere. The study presented here is composed of two parts. First, we analyzed the source-rupture processes of the April 1, 2007 Solomon Islands Earthquake (Mw=8.1) using a body-wave inversion technique. Our analysis indicated that the earthquake ruptured approximately 240 km of the southeast Pacific subduction zone in two sub-events.In the second part of this study, we used shear-wave splitting analysis to investigate the effects of the subducting African lithosphere on the upper-mantle flow field beneath the Anatolian Plate in the Eastern Mediterranean region. Our shear-wave splitting results are consistent with relatively uniform southwest-directed flow towards the actively southwestward-retreating Aegean slab. Based on spatial variations in observed delay times we identified varying flow speeds beneath Anatolia and we attribute this variation to the differential retreat rates of the Aegean and the Cyprean trenches.Finally, we used teleseismic P-wave travel-time tomography to image the geometry of the subducting African lithosphere beneath the Anatolia region. Our tomograms show that the subducting African lithosphere is partitioned into at least two segments along the Cyprean and the Aegean trenches. We observed a gap between the two segments through which hot asthenosphere ascends beneath the volcanic fields of western Anatolia. Our results show that the Cyprean slab is steeper than the Aegean slab. We inferred that this steep geometry, in part, controls the flow regime of asthenosphere beneath Anatolia causing variations in flow speeds inferred from shear-wave splitting analysis.

The lithosphere of northwestern Canada recorded more than 2.5 Gy of complex tectonic evolution, from the formation of the ancient cores of the continental lithosphere such as the Slave craton to the Phanerozoic Cordilleran orogeny with substantial variations in crust and upper mantle structures that led to the concentration of natural resources (i.e., diamonds in cratons). Present-day northwestern Canada juxtaposes a thin and hot Cordilleran lithosphere to the thick and cold cratonic lithosphere, which has important implications for regional geodynamics. Recently, seismic station coverage has drastically increased across northwestern Canada, allowing the development of seismic tomography models and other passive-source seismic methods at high resolution in order to investigate the tectonic evolution and dynamics of the lithosphere in this region. The P- and S-wave upper mantle structures of northwestern Canada reveal that the distribution of kimberlite fields in the Slave craton correlates with the margin of fast and slow seismic mantle anomalies, which could delineate weak zones in the lithosphere. Based on our tomographic models we identify two high-velocity seismic anomalies straddling the arcuate Cordillera Deformation Front that have controlled its regional deformation, including a newly identified Mackenzie craton characterized by high seismic velocities extending from the lower crust to the upper mantle to the north of the Mackenzie Mountains. Furthermore, our P-wave tomography model shows sharp velocity contrasts beneath the surface trace of the Tintina Fault. Estimates of seismic anisotropy show a progressive rotation of fast-axis directions when approaching the fault zone. Together, they provide seismic evidence for the trans-lithospheric nature of the Tintina Fault. We further propose that the Tintina Fault has chiseled off small pieces of the Laurentian craton between the Late Cretaceous and the Eocene, which would imply that large lithospheric-scale shear zones are able to cut through small pieces of refractory cratonic mantle and transport them over several hundred kilometers.

Les progrès de l'imagerie sismique ces trente dernières années ont permis de révéler la structure complexe de la graine : une anisotropie cylindrique de quelques pourcents dont la structure fine présente des variations radiales et latérales. Ce travail de thèse s’est concentré sur les différentes dynamiques susceptibles de provoquer une telle structure. [Il y a autre chose dans cette structure]Nous avons revisité la dynamique induite par la force de Lorentz, discutant les conditions aux limites, la croissance de la graine et la stratification en densité. La déformation n’est raisonnablement suffisante que pour des viscosités inférieures à 1012 Pa.s, dans la fourchette basse des estimations. Les modèles d'écoulements globaux dans la graine peuvent se classer en deux grandes catégories. Pour un profil de densité stable, seul un forçage extérieur, tel que la force de Lorentz, peut induire un écoulement. Dans le cas instable, la dynamique est contrôlée principalement par des instabilités de convection. Les nouvelles estimations de la diffusion thermique limitent les instabilités thermiques à des âges de graine de l'ordre de la centaine de millions d'années. En se focalisant sur deux paramètres, la viscosité et l'âge de la graine, nous avons construit un diagramme de régime qui compare quantitativement les différents modèles proposés dans la littérature. En croisant amplitude du taux de déformation et géométrie attendues, on peut raisonnablement restreindre les domaines qui pourraient engendrer la structure observée. Pourtant, aucun modèle n'est pour l'instant capable d'expliquer à la fois l'amplitude et la géométrie de l'anisotropie sismique et encore moins la dichotomie Est-Ouest. La couche F est une anomalie dans le noyau externe : d’une épaisseur de 200 km environ, à la base du noyau externe, elle présente des vitesses d’ondes sismiques plus faibles que celles prédites pour un noyau liquide parfaitement mélangé. Elle est interprétée comme une zone chimiquement appauvrie, en contradiction avec la cristallisation de la graine qui libère des éléments légers à la surface même de la graine. Nous étudions la possibilité d’une cristallisation en volume dans cette couche. Les particules de fer solides sédimentent en croissant dans un liquide de plus en plus appauvri en éléments légers. Cette neige de fer est stable sous certaines conditions, étudiées ici. / In the past thirty years, our understanding of the inner core structure has increased with the number of seismic studies. Observations reveal a global anisotropy with a cylindrical symmetry and radial and lateral heterogeneities. In this work, I have studied different hypothesis on the dynamics of the inner core to explain these observations. Revisiting the dynamics induced by the Lorentz force, we studied the effect of new boundary conditions, the effect of stratification and growth rate. However, the obtained flow is not strong enough to deform the media for viscosities larger than 1012 Pa.s, in the lower bound of the published inner core estimates. Deformation mechanisms can be subdivided between natural convection (arising from unstable thermal or compositional gradients) and externally forced flows, like the one induced by the Lorentz force. Recent estimates of the thermal diffusivity of iron at high pressure limit the possibility of thermal convection to an age of the inner core lower than a hundred of millions years. Two key parameters emerge for the inner core dynamics: the sign and strength of the density stratification and the viscosity of the inner core. We construct a regime diagram for the Earth’s inner core dynamics that compares the different published models in term of maximum instantaneous deformation rate. This diagram allows us to compare both expected strain rate and deformation geometry with the seismic observations. However, we find that no published model can explain all the seismic observations. The inner core anisotropy and the hemispherical dichotomy are especially difficult to reconcile with these models. The F-Layer is a 200km anomalous layer at the bottom of the outer that presents low P-Wave velocities compared to well-Mixed model. It has been interpreted as a layer depleted in light elements, whereas we usually consider that light elements are expelled at the surface of the inner core by freezing of the outer core alloy. We study the hypothesis of freezing in the bulk of the layer, with iron particles growing and settling in an increasingly depleted liquid.

The upper mantle structure of Swedish part of Baltic Shield with its isotropic and anisotropic seismic velocity characteristics is investigated using telesesismic body waves (i.e. P waves and shear waves) recorded by the Swedish National Seismological Network (SNSN). Nonlinear high-resolution P and SV and SH wave isotropic tomographic inversions reveal velocity perturbations of ± 3 % down to at least 470 km below the network. Separate SV and SV models indicate several consistent major features, many of which are also consistent with P-wave results. A direct cell by cell comparison of SH and SV models reveals velocity differences of up to 4%. Numerical tests show that differences in the two S-wave models can only be partially caused by noise and limited resolution, and some features are attributed to the effect of large scale anisotropy. Shear-wave splitting and P-travel time residual analyses also detect anisotropic mantle structure. Distinct back-azimuth dependence of SKS splitting excludes single-layer anisotropy models with horizontal symmetry axes for the whole region. Joint inversion using both the P and S data reveals 3D self-consistent anisotropic models with well-defined mantle lithospheric domains. These domains of differently oriented anisotropy most probably retain fossil fabric since the domains' origin, supporting the idea of the existence of an early form of plate tectonics during formation of continental cratons already in the Archean. The possible disturbing effects of anisotropy on seismic tomography studies are investigated, and found to be potentially significant. P-wave arrival times were adjusted based on the estimates of mantle anisotropy, and re-inverted. The general pattern of the velocity-perturbation images was similar but changed significantly in some places, including the disappearance of a slab-like structure identified in the inversion with the original data. Thus the analysis demonstrates that anisotropy of quite plausible magnitude can have a significant effect on the tomographic images, and should not be ignored. If, as we believe, our estimates of anisotropy are reasonably correct, then the model based on the adjusted data should give a more robust and correct image of the mantle structure.

Estimation of elastic anisotropy, which is usually caused by rock fabrics and mineral orientation, has an important role in exploration seismology and better understanding of crustal seismic reflections. If not properly taken care of during processing steps, it may lead to wrong interpretation or distorted seismic image. In this thesis, a state-of-the-art under the development Laser Doppler Interferometer (LDI) device is used to measure phase velocities on the surface of rock samples from a major deformation zone (Österbybruk Deformation Zone) in the Bergslagen region of eastern Sweden. Then, a general inversion code is deployed to invert measured phase velocities to obtain full elastic stiffness tensors of two samples from the major deformation zone in the study area. At the end, results are used to correct for the anisotropy effects using three dimensionless Tsvankin's parameters and a non-hyperbolic moveout equation. The resulting stacked section shows partial reflection improvement of the deformation zone compared with the isotropic processing section. This suggests that rock anisotropy may also contribute to the generation of reflections from the deformation zones in the study area but requires further investigations.

In the first article, the seismic properties for a suite of rocks along the West Cycladic Detachment System (Greece) are calculated, using Electron backscatter diffraction (EBSD) measurements and the minerals’ elastic stiffness tensors. Muscovite and glaucophane well defined crystallographic preferred orientation increases the seismic anisotropy. Maximum Pwave velocities have the same orientation as the Miocene extension and maximum S-wave anisotropy is subhorizontal, parallel with mineral alignment, suggesting strong radial anisotropy with a slow subvertical axis of symmetry. In the second article, teleseismic receiver functions are calculated for an array of stations in the Cyclades and decomposed into back-azimuth harmonics to visualise the variations in structure and anisotropy across the array. Synthetic receiver functions are modeled using the first order structural observations of seismic discontinuities and EBSD data. They indicate 5% of anisotropy with slow symmetry axis in the upper crust, and demonstrate the importance of rock textural constraints in seismic velocity profile interpretation.