Traveltimes are conventionally evaluated by solving the zero-order approximation of the Wentzel, Kramers and Brillouin (WKB) expansion of the wave equation. This high frequency approximation is good enough for most imaging applications and provides us with a traveltime equation called the eikonal equation. The eikonal equation is a non-linear partial differential equation which can be solved by any of the familiar numerical methods. Among the most popular of these methods is the method of characteristics which yields the ray tracing equations and the finite difference approaches. In the first part of the Master Thesis, we use the ray tracing method to solve the eikonal equation to get P-waves traveltimes for orthorhombic models with arbitrary orientation of symmetry planes. We start with a ray tracing procedure specified in curvilinear coordinate system valid for anisotropy of arbitrary symmetry. The coordinate system is constructed so that the coordinate lines are perpendicular to the symmetry planes of an orthorohombic medium. Advantages of this approach are the conservation of orthorhombic symmetry throughout the model and reduction of the number of parameters specifying the model. We combine this procedure with first-order ray tracing and dynamic ray tracing equations for P waves propagating in smooth, inhomogeneous, weakly anisotropic media. The first-order ray tracing and dynamic ray tracing equations are derived from the exact ones by replacing the exact P-wave eigenvalue of the Christoffel matrix by its first-order approximation. In the second part of the Master Thesis, we compute traveltimes using the fast marching method and we develop an approach to estimate the anisotropy parameters. The idea is to relate them analytically to traveltimes which is challenging in inhomogeneous media. Using perturbation theory, we develop traveltime approximations for transversely isotropic media with horizontal symmetry axis (HTI) as explicit functions of the anellipticity parameter and the symmetry axis azimuth in inhomogeneous background media. Specifically, our expansion assumes an inhomogeneous elliptically anisotropic background medium, which may be obtained from well information and stacking velocity analysis in HTI media. This formulation has advantages on two fronts: on one hand, it alleviates the computational complexity associated with solving the HTI eikonal equation, and on the other hand, it provides a mechanism to scan for the best fitting parameters without the need for repetitive modeling of traveltimes, because the traveltime coefficients of the expansion are independent of the perturbed parameters.

This study is concerned with the computation of seismic first-arrival traveltimes in anisotropic media using finite difference eikonal methods. For this purpose, different numerical schemes that directly solve the eikonal equation are implemented and assessed numerically. Subsequently, they are used for pre-stack depth migration on synthetic and field data. The thesis starts with a detailed examination of different finite difference methods that have gained popularity in scientific literature for computing seismic traveltimes in isotropic media. The most appropriate for an extension towards anisotropic media are found to be the so-called Fast Marching/Sweeping methods. Both schemes rely on different iteration strategies, but incorporate the same upwind finite difference Godunov schemes that are implemented up to the second order. As a result, the derived methods exhibit high numerical accuracy and perform robustly even in highly contrasted velocity models. Subsequently, the methods are adapted for transversely isotropic media with vertical (VTI) and tilted (TTI) symmetry axes, respectively. Therefore, two different formulations for approximating the anisotropic phase velocities are tested, which are the weakly-anisotropic and the pseudo-acoustic approximation. As expected, the pseudo-acoustic formulation shows superior accuracy especially for strongly anisotropic media. Moreover, it turns out that the tested eikonal schemes are generally more accurate than anisotropic ray tracing approaches, since they do not require an approximation of the group velocity. Numerical experiments are carried out on homogeneous models with varying strengths of anisotropy and the industrial BP 2007 benchmark model. They show that the computed eikonal traveltimes are in good agreement with independent results from finite difference modelling of the isotropic and anisotropic elastic wave equations, and traveltimes estimated by ray-based wavefront construction, respectively. The computational performance of the TI eikonal schemes is largely increased compared to their original isotropic implementations, which is due to the algebraic complexity of the anisotropic phase velocity formulations. At this point, the Fast Marching Method is found to be more efficient on models containing up to 50 million grid points. For larger models, the anisotropic Fast Sweeping implementation gradually becomes advantageous. Here, both techniques perform independently well of the structural complexity of the underlying velocity model. The final step of this thesis is the application of the developed eikonal schemes in pre-stack depth migration. A synthetic experiment over a VTI/TTI layer-cake model demonstrates that the traveltime computation leads to accurate imaging results including a tilted, strongly anisotropic shale layer. The experiment shows further that the estimation of anisotropic velocity models solely from surface reflection data is highly ambiguous. In a second example, the eikonal solvers are applied for depth imaging of two-dimensional field data that were acquired for geothermal exploration in southern Tuscany, Italy. The developed methods also produce clear imaging results in this setting, which illustrates their general applicability for pre-stack depth imaging, particularly in challenging environments.

This study is concerned with the computation of seismic first-arrival traveltimes in anisotropic media using finite difference eikonal methods. For this purpose, different numerical schemes that directly solve the eikonal equation are implemented and assessed numerically. Subsequently, they are used for pre-stack depth migration on synthetic and field data. The thesis starts with a detailed examination of different finite difference methods that have gained popularity in scientific literature for computing seismic traveltimes in isotropic media. The most appropriate for an extension towards anisotropic media are found to be the so-called Fast Marching/Sweeping methods. Both schemes rely on different iteration strategies, but incorporate the same upwind finite difference Godunov schemes that are implemented up to the second order. As a result, the derived methods exhibit high numerical accuracy and perform robustly even in highly contrasted velocity models. Subsequently, the methods are adapted for transversely isotropic media with vertical (VTI) and tilted (TTI) symmetry axes, respectively. Therefore, two different formulations for approximating the anisotropic phase velocities are tested, which are the weakly-anisotropic and the pseudo-acoustic approximation. As expected, the pseudo-acoustic formulation shows superior accuracy especially for strongly anisotropic media. Moreover, it turns out that the tested eikonal schemes are generally more accurate than anisotropic ray tracing approaches, since they do not require an approximation of the group velocity. Numerical experiments are carried out on homogeneous models with varying strengths of anisotropy and the industrial BP 2007 benchmark model. They show that the computed eikonal traveltimes are in good agreement with independent results from finite difference modelling of the isotropic and anisotropic elastic wave equations, and traveltimes estimated by ray-based wavefront construction, respectively. The computational performance of the TI eikonal schemes is largely increased compared to their original isotropic implementations, which is due to the algebraic complexity of the anisotropic phase velocity formulations. At this point, the Fast Marching Method is found to be more efficient on models containing up to 50 million grid points. For larger models, the anisotropic Fast Sweeping implementation gradually becomes advantageous. Here, both techniques perform independently well of the structural complexity of the underlying velocity model. The final step of this thesis is the application of the developed eikonal schemes in pre-stack depth migration. A synthetic experiment over a VTI/TTI layer-cake model demonstrates that the traveltime computation leads to accurate imaging results including a tilted, strongly anisotropic shale layer. The experiment shows further that the estimation of anisotropic velocity models solely from surface reflection data is highly ambiguous. In a second example, the eikonal solvers are applied for depth imaging of two-dimensional field data that were acquired for geothermal exploration in southern Tuscany, Italy. The developed methods also produce clear imaging results in this setting, which illustrates their general applicability for pre-stack depth imaging, particularly in challenging environments.

Les temps de trajet, amplitudes et angles à la source des ondes sismiques sont utilisés dans de nombreuses applications telles que la migration, la tomographie, l'estimation de la sensibilité de détection et la localisation des microséismes. Dans le contexte de la microsismicité, il est nécessaire de calculer en quasi temps réel ces attributs avec précision. Nous avons développé ici un ensemble d'algorithmes rapides et précis en 3D pour des modèles à fort contraste de vitesse.Nous présentons une nouvelle méthode pour calculer les temps de trajet, les amplitudes et les angles à la source des ondes correspondant aux premières arrivées. Plus précisément, nous résolvons l'équation Eikonal, l'équation de transport et l'équation des angles en nous basant sur une approche par différences finies pour des modèles de vitesse en 3D. Nous proposons une nouvelle méthode hybride qui bénéficie des avantages respectifs de plusieurs approches existantes de résolution de l'équation Eikonal. En particulier, les approches classiques proposent généralement de résoudre directement les équations et font l'approximation localement d'une onde plane. Cette approximation n'est pas bien adaptée au voisinage de la source car la courbure du front d'onde est importante. Des erreurs de temps de trajet sont alors générées près de la position de la source, puis propagées à travers tout le modèle de vitesse. Ceci empêche de calculer correctement les amplitudes et les angles à la source puisqu'ils reposent sur les gradients des temps. Nous surmontons cette difficulté en introduisant les opérateurs sphériques ; plus précisément nous reformulons les temps de trajet, amplitudes et angles à la source par la méthode des perturbations.Nous validons nos nouvelles méthodes pour différents modèles à fort contraste de vitesse en 2D et 3D et montrons notre contribution par rapport aux approches existantes. Nos résultats sont similaires à ceux calculés en utilisant la modélisation de la forme d'onde totale alors qu'ils sont bien moins coûteux en temps de calcul. Ces résultats ouvrent donc de nouvelles perspectives pour de nombreuses applications telles que la migration, l'estimation de la sensibilité de détection et l'inversion des mécanismes au foyer. / Traveltimes, amplitudes and take-off angles of seismic waves are used in many applications such as migration, tomography, detection sensitivity estimation and microseism location. In the microseismicty context it is necessary to compute in near real time accurately these attributes. Here we developed a set of fast and accurate algorithms in 3D for highly contrasted velocity models.We present a new accurate method for computing first arrival traveltimes, amplitudes and take-off angles; more precisely we solve the Eikonal, transport and take-off angle equations based on a finite difference approach for 3D velocity models. We propose a new hybrid method that benefits from the advantages of several existing Eikonal solvers. Common approaches that solve directly these equations assume that we are locally propagating a plane wave. This approximation is not well adapted in the neighborhood of the source since the wavefront curvature is important. Travel times errors are generated near the source position and then propagated through the whole velocity model. This prevents from properly calculating the amplitudes and the take-off angles since they rely on the travel time gradients that are not accurate. We overcome this difficulty by introducing spherical operators. Indeed we reformulate the traveltimes, amplitudes and take-off angles with the perturbation method.We validate our new methods on various highly contrasted velocity models in 2D and 3D and show our contribution compared to other existing approaches. Our results are similar to those computed using full waveform modeling while they are obtained in a much shorter CPU time. These results open thus new perspectives for several applications such as migration, detection sensitivity estimation and focal mechanism inversion.