Seismic full waveform inversion (waveform tomography) is a method to reconstruct the underground velocity field in high resolution using seismic data. The method was first introduced during the 1980’s and became computationally feasible during the late 1990’s when the method was implemented in the frequency domain. This work presents three case studies and one synthetic benchmark of full waveform inversion applications. Two of the case studies are focused on time-lapse cross-well and 2D reflection seismic data sets acquired at the Ketzin CO2 geological storage site. These studies are parts of the CO2SINK and CO2MAN projects. The results show that waveform tomography is more effective than traveltime tomography for the CO2 injection monitoring at the Ketzin site for the cross-well geometry. For the surface data sets we find it is difficult to recover the true value of the velocity anomaly due to the injection using the waveform inversion method, but it is possible to qualitatively locate the distribution of the injected CO2. The results agree well with expectations based upon conventional 2D CDP processing methods and more extensive 3D CDP processing methods in the area. A further investigation was done to study the feasibility and efficiency of seismic full waveform inversion for time-lapse monitoring of onshore CO2 geological storage sites using a reflection seismic geometry with synthetic data sets. The results show that waveform inversion may be a good complement to standard CDP processing when monitoring CO2 injection. The choice of method and strategy for waveform inversion is quite dependent on the goals of the time-lapse monitoring of the CO2 injection. The last case study is an application of the full waveform inversion method to two crooked profiles at the Forsmark site in eastern central Sweden. The main goal of this study was to help determine if the observed reflections are mainly due to fluid filled fracture zones or mafic sills. One main difficulty here is that the profiles have a crooked line geometry which corresponds to 3D seismic geometry, but a 2D based inversion method is being used. This is partly handled by a 3D to 2D coordinate projection method from traveltime inversion. The results show that these reflections are primarily due to zones of lower velocity, consistent with them being generated at water filled fracture zones.

L'inversion de forme d'onde complète (FWI) est un processus non-linéaire et mal posé d’ajustement de données, dans notre cas, issues d’acquisitions simiques. Cette technique cherche à reconstruire, à partir d’un modèle initial obtenu à faible nombre d’onde (faible résolution), des paramètres constitutifs contrôlant la propagation des ondes à grands nombres d’ondes (forte résolution). Durant ce processus itératif, certains artéfacts peuvent altérer la qualité du modèle reconstruit. Afin de diminuer ces artéfacts et d’assurer une reconstruction des paramètres qui soit cohérente d’un point de vue géologique, différentes techniques de pré-conditionnement ou de régularisation peuvent être proposées.Cette thèse se focalise sur le potentiel de nouveaux filtres multi-dimensionnels construits dans l’espace des nombres d’ondes et orientés suivant les structures géologiques. Une stratégie de pré-conditionnement a été mise au point à l’aide de ces filtres et a été appliquée avec succès à la problématique FWI. La formulation analytique 1D de l’opérateur inverse de covariance laplacienne (Tarantola, 2005) constitue la base de la formulation d’opérateurs de dimension supérieure qui sont validés ici en les comparants avec l’opérateur analytique de covariance laplacienne 1D. Nous avons utilisé cette fonction analytique inverse 1D comme la base de filtrage de dimension supérieure, via l’addition de multiples fonctions inverses orientées orthogonalement. Ces fonctions laplaciennes inverses additionnelles (AIL) sont obtenues pour des configurations 2D et 3D après discrétisation par des techniques de différences finies. Nous montrons que l’on peut calculer un filtre en nombre d’onde de manière rapide et robuste en résolvant le système linéaire associé à ces opérateurs inverses. Lorsque des pentes sont inclues à l’étape de discrétisation par différences finies, il est alors possible d’utiliser ces opérateurs comme des filtres en nombre d’ondes orientés vers les structures géologiques, ceci avec une grande efficacité.Ce filtre (AIL) montre des propriétés rapides de convergence et des performances indépendantes du vecteur à filtrer. Nous montrons notamment comment ce filtre peut être utilisé comme un opérateur utile pour le gradient associé à la FWI. Le pré-conditionnement du gradient peut atténuer les effets du problème mal-posé qui vont s’étendre dans l’espace des modèles. Deux exemples synthétiques (Valhall et Marmousi) calculés dans l’espace des fréquences sont proposés dans cette thèse. Le pré-conditionnement AIL s’avère efficace pour atténuer d’une part la signature mal-posée provenant de la présence de bruit ambient dans les données observées et d’autre part d’artéfacts liés aux effets de repliement spatial liés aux conditions d’imagerie par FWI. La possibilité d’inclure des pentes permet de filtrer de manière préférentielle en considérant des pendages géologiques. Cette stratégie de filtrage permet l’atténuation d’artéfacts, tout en préservant le contenu en nombre d’ondes de la stratigraphie orthogonale au pendage.Un cas réel d’inversion 2D FWI est finalement abordé permettant tout d’abord d’illustrer la sensibilité des résultats d’inversion au modèle initial. Celui-ci est d’importance majeure, particulièrement dans les régions profondes dépassant la pénétration maximale des ondes transmises. L’application de la technique FWI à cette acquisition sismique a permis d’améliorer de manière significative la cohérence sur une image migrée par renversement du temps (RTM). Nous montrons également que le pré-conditionneur AIL permet une décroissance significative du nombre de tirs requis à modéliser dans la boucle d’inversion, sans pour autant dégrader le contenu en nombre d’onde des structures géologiques principales dans les résultats finaux obtenus après inversion. / Full waveform inversion (FWI) is a non-linear, ill-posed, local data fitting technique. FWI looks to moves from an initial, low-wavenumber representation of the earth parameters to a broadband representation. During this iterative process a number of undesirable artifacts can map into our model parameter reconstruction. To mitigate these artifacts and to ensure a geologically consistent model parameter reconstruction, various preconditioning and/or regularization strategies have been proposed.This thesis details the construction of new, efficient, multi-dimensional, structurally-orientated wavenumber filters. A preconditioning strategy has been devised using these filters that we have successfully applied to FWI. The 1D analytical inverse Laplacian covariance operator (Tarantola, 2005) forms the basis of higher dimensional operators and is initially validated by comparing to the 1D analytical Laplacian covariance operator. We use this analytical 1D inverse function as the basis for higher dimensional filtering via the addition of multiple, orthogonally orientated inverse functions. These additive inverse laplacian functions (AIL) are shown in 2D and 3D configurations and are discretized using finite-difference techniques. We show that one can calculate, a rapid and robust wavenumber filter, by solving the linear system associated with these inverse operators. When dip is included at the finite difference discretization stage, it is possible to use these operators as highly efficient, structurally orientated wavenumber filters.The AIL filter is shown to be rapid to converge and its performance is independent of the vector to be filtered. We show, that the filter can be a useful preconditioning operator for the FWI gradient. Preconditioning the gradient can mitigate against ill-posed effects mapping into the model-space. Two synthetic (Valhall and Marmousi) frequency domain FWI example are shown in this thesis. The AIL preconditioner has success at mitigating the ill-posed imprint coming from ambient noise in the observed data and also artifacts from spatial aliasing effects in the FWI imaging condition. The ability to include dip, allows one to preferentially filter along geological dip. This filtering strategy allows the mitigation of artifacts, while simultaneously preserving the stratigraphic based wavenumber content that is orthogonal to dip.A 2D, real data FWI case-study is also shown and we highlight the sensitivity of the inversion result to the initial model. The initial model is of key importance, this especially true in the areas deeper than the maximum penetration of transmitted waves. The application of FWI on this line is able to significantly improve gather alignment on a RTM, migrated image. We also see that the AIL preconditioner can allows us to significantly decrease the number of shot records we are required to model in our inversion workflow without degrading the key geological wavenumber content in the final inversion result.

Title: The source process of Greek earthquakes Author: Dana K ížová Department: Department of Geophysics Supervisor of the doctoral thesis: Prof. RNDr. Ji í Zahradník DrSc., Department of Geophysics Abstract: Investigations of moment tensor (MT) and its uncertainty are topical. This thesis is focused on isotropic component of three shallow earthquakes: Event A in Cretan Sea (Mw 5.3) and two events near Santorini island, B (Mw 4.9) and C (Mw 4.7). MT is inverted from full waveforms in an assumed 1D velocity model. The inverse problem is non-linear in centroid depth and time, and linear in six MT parameters, one is the MT-trace. Uncertainty of isotropic component is studied by a new approach (K ížová et al., 2013). The trace is systematically varied, and remaining parameters are optimized. The method reveals tradeoffs between the isotropic component, depth, time, and focal mechanism. From two existing velocity models, we prefer the one with lower condition number, in which a (positive) isotropic component is indicated for event B. To rapidly assess a likely existence of isotropic component, an empirical method is proposed (K ížová et al., 2016). It is based on comparison between depth- dependences of waveform correlation in full and deviatoric modes. Based on extensive synthetic tests, the method confirms a...

The ever-expanding need for energy, including those related to fossil fuels, is behind the drive to explore more complicated regions, such as salt and subsalt provinces. This exploration quest relies heavily on recorded surface seismic data to provide precise and detailed subsurface properties. However, conventional seismic processing algorithms including the state-of-the-art full-waveform inversion (FWI) fail to recover those features in many areas of salt provinces. Even the industrial solution with substantial involvement of manual human-interpretation has faced challenges in many regions. In this thesis, I attempt to replace those manual, and somewhat erroneous, steps to the velocity building in salt provinces with a mathematically robust algorithm under the FWI machinery. I, specifically, regularize FWI by penalizing the velocity drops with depth with a new more flexible function. Although promising, FWI is computationally very expensive, especially for large 3D seismic data. It updates an initial guess of the model iteratively using the gradient of the misfit function, which requires lengthy computations and large memory space/disc storage. Based on the adjoint state method, gradient computation usually requires us to store the source wavefield, or include an additional extrapolation step to propagate the source wavefield from its temporary storage at the boundary. To mitigate this computational overburden, I propose an amplitude excitation gradient calculation based on representing the source wavefield history by a single, specifically the most energetic arrival. In this thesis, I also propose a novel-multiscale scheme based on ux-corrected transport (FCT) to reduce artifacts in the gradient direction due to the noise present in seismic data. FCT comprises of two finite difference schemes: a transport and a diffusion to compute the flux at a grid point. I observe a couple of benefits in FCT-based FWI. First, it yields a smooth gradient at the earlier iterations of FWI by promoting the lower frequency content of the seismic data. Second, it is easily compatible with the existing FWI code, and with any objective function. The multiscale strategy starts with a large smoothing parameter and relaxes it progressively to zero to achieve the final inverted model from traditional FWI.

Dans cette thèse, un algorithme d'inversion de formes d'ondes (FWI) est développé pour l'imagerie 3D des paramètres élastiques de la lithosphère à partir des enregistrements télésismiques dans le but d'accroître la résolution des images lithosphériques. La modélisation sismique est effectuée par un méthode hybride d'injection de champ d'ondes. Une première modélisation est effectuée dans une Terre globale avec le logiciel AxiSEM pour déterminer les champs d’ondes aux bords de la cible lithosphérique. Ces solutions sont ensuite propagées dans cette cible par une méthode aux éléments finis spectraux. Le problème inverse est résolu avec un algorithme d’optimisation locale de type quasi-Newton (l-BFGS). La sensibilité de la méthode à la configuration expérimentale (paramétrisation du milieu, modèle initial, géométrie et échantillonnage du dispositif de capteurs) est tout d’abord analysée avec un modèle synthétique réaliste des Alpes Occidentales. L’algorithme est finalement appliqué à neuf événements de la campagne CIFALPS dans les Alpes occidentales jusqu’à une fréquence de 0.2Hz. Les modèles de vitesses P et S et de densité révèlent les grandes structures lithosphériques de la chaîne alpine, en particulier le corps d’Ivrée et la géométrie des Moho européen et adriatique. Plus profondément, deux anomalies de vitesses lentes sont imagées dans le manteau et sont interprétées comme la signature d’une remontée asthénosphérique et la localisation du détachement du panneau plongeant européen. Ces résultats corroborent l’hypothèse d’une subduction continentale de la croûte européenne et d’une éventuelle déchirure du panneau plongeant européen lors de la phase de collision. / In this thesis, a full-waveform inversion (FWI) algorithm is developed with the aim to image the elastic properties (Vp, Vs and density) of 3D lithospheric models from teleseismic recordings with a spatial resolution of the order of the wavelength. Seismic modeling is performed with a wavefield injection hybrid approach. A first simulation is performed in a global radially symmetric Earth with the AxiSEM code to compute the wavefields on the borders of the lithospheric target. Then, these wavefields are propagated in the target with the spectral finite-element method. After linearization, the inverse problem is solved with a quasi-Newton (1-BFGS) optimization algorithm. The sensitivity of the teleseismic FWI to the experimental setup (subsurface parameterization, initial model, sampling and geometry of the station layout) is first assessed with a realistic synthetic model of the Western Alps. The method is finally applied to nine events of the CIFALPS experiment carried out in the Western Alps, up to a frequency of 0.2Hz. Reliable models of P and S wave speeds and density reveal with an unprecedented resolution the crustal and lithospheric structures of the Alpine Belt, in particular the geometry of the Ivrea body, and the European and Adriatic Mohos. Deeper, two slow velocity anomalies beneath the Western Alps are imaged in the mantle. The first, to the west of the chain, is interpreted as the signature of an asthenospheric upwelling, the second near the location of the Ivrea body indicates the European slab break-off. The study supports the hypothesis of the European continental crust subduction and confirms the possible tearing of the European slab.

Maintenance of aging buried infrastructure and reinforced concrete are critical issues in the United States. Inexpensive non-destructive techniques for mapping and imaging infrastructure and defects are an integral component of maintenance. Ground penetrating radar (GPR) is a widely-used non-destructive tool for locating buried infrastructure and for imaging rebar and other features of interest to civil engineers. Conventional acquisition and interpretation of GPR profiles is based on the arrival times of strong reflected/diffracted returns, and qualitative interpretation of return amplitudes. Features are thereby generally well located, but their material properties are only qualitatively assessed. For example, in the typical imaging of buried pipes, the average radar wave velocity through the overlying soil is estimated, but the properties of the pipe itself are not quantitatively resolved. For pipes on the order of the radar wavelength (<5-35 cm), pipe dimensions and infilling material remain ambiguous. Full waveform inversion (FWI) methods exploit the entire radar return rather than the time and peak amplitude. FWI can generate better quantitative estimates of subsurface properties. In recent decades FWI methods, developed for seismic oil exploration, have been adapted and advanced for GPR with encouraging results. To date, however, FWI methods for GPR data have not been specifically tuned and applied on surface collected common offset GPR data, which are the most common type of GPR data for engineering applications. I present an effective FWI method specifically tailored for common-offset GPR data. This method is composed of three main components, the forward modeling, wavelet estimation and inversion tools. For the forward modeling and iterative data inversion I use two open-source software packages, gprMax and PEST. The source wavelet, which is the most challenging component that guarantees the success of the method, is estimated with a novel Sparse Blind Deconvolution (SBD) algorithm that I have developed. The present dissertation indicates that with FWI, GPR can yield better quantitative estimates, for example, of both the diameters of small pipes and rebar and their electromagnetic properties (permittivity, conductivity). Also better estimates of electrical properties of the surrounding media (i.e. soil or concrete) are achieved with FWI.

京都大学 / 0048 / 新制・課程博士 / 博士(工学) / 甲第20061号 / 工博第4249号 / 新制||工||1658(附属図書館) / 京都大学大学院工学研究科社会基盤工学専攻 / (主査)教授 三ｹ田 均, 教授 小池 克明, 教授 木村 亮 高梨 将 / 学位規則第4条第1項該当 / Doctor of Philosophy (Engineering) / Kyoto University / DFAM

The main objective of this thesis is to develop novel misfit functions for full-waveform inversion such that (a) the estimation of the long-wavelength model will less likely stagnate in spurious local minima and (b) the inversion is immune to wavelet inaccuracy. First, I investigate the pros and cons of misfit functions based on optimal transport theory to indicate the traveltime discrepancy for seismic data. Even though the mathematically well-defined optimal transport theory is robust to highlight the traveltime difference between two probability distributions, it becomes restricted as applied to seismic data mainly because the seismic data are not probability distribution functions. We then develop a misfit function combining the local cross-correlation and dynamic time warping. This combination enables the proposed misfit automatically identify arrivals associated with a phase shift. Numerical and field data examples demonstrate its robustness for early arrivals and limitations for later arrivals.%, which means that a proper pre-processing step is still required. Next, we introduce differentiable dynamic time warping distance as the misfit function highlighting the traveltime discrepancy without non-trivial human intervention. Compared to the conventional warping distance, the differentiable version retains the property of representing the traveltime difference; moreover, it can eliminate abrupt changes in the adjoint source, which helps full-waveform inversion converge to geologically relevant estimates. Finally, we develop a misfit function entitled the deconvolutional double-difference measurement. The new misfit measures the first difference by deconvolution rather than cross-correlation. We also present the derivation of the adjoint source with the new misfit function. Numerical examples and mathematical proof demonstrate that this modification makes full-waveform inversion with the deconvolutional double-difference measurement immune to wavelet inaccuracy.

Full Waveform Inversion (FWI) is a widely used optimization technique for subsurface imaging where the goal is to estimate the seismic wave velocity beneath the Earth's surface from the observed seismic data at the surface. The problem is primarily governed by the wave equation, which is a non-linear second-order partial differential equation. A number of approaches have been developed for FWI including physics-based iterative numerical solvers as well as data-driven machine learning (ML) methods. Existing numerical solutions to FWI suffer from three major challenges: (1) sensitivity to initial velocity guess (2) non-convex loss landscape, and (3) sensitivity to noise. Additionally, they suffer from high computational cost, making them infeasible to apply in complex real-world applications. Existing ML solutions for FWI only solve for the inverse and are prone to yield non-unique solutions. In this work, we propose to solve both forward and inverse problems jointly to alleviate the issue of non-unique solutions for an inverse problem. We study the FWI problem from a new perspective and propose a novel approach based on Invertible Neural Networks. This type of neural network is designed to learn bijective mappings between the input and target distributions and hence they present a potential solution to solve forward and inverse problems jointly. In this thesis, we developed a data-driven framework that can be used to learn forward and inverse mappings between any arbitrary input and output space. Our model, Invertible X-net, can be used to solve FWI to obtain high-quality velocity images and also predict the seismic waveforms data. We compare our model with the existing baseline mod- els and show that our model outperforms them in velocity reconstruction on the OpenFWI dataset. Additionally, we also compare the predicted waveforms with a baseline and ground truth and show that our model is capable of predicting highly accurate seismic waveforms simultaneously. / Master of Science / Recent advancements in deep learning have led to the development of sophisticated methods that can be used to solve scientific problems in many disciplines including medical imaging, geophysics, and signal processing. For example, in geophysics, we study the internal structure of the Earth from indirect physical measurements. Often, these kind of problems are challenging due to existence of non-unique and unstable solutions. In this thesis, we look at one such problem called Full Waveform Inversion which aims to estimate velocity of mechanical wave inside the Earth from wave amplitude observations on the surface. For this problem, we explore a special class of neural networks that allows to uniquely map the input and output space and thus alleviate the non-uniqueness and instability in performing Full Waveform Inversion for seismic imaging.

Wissenschaftliche Untersuchungen von terrestrischen und aquatischen Ökosystemen erfordern präzise Informationen über die dreidimensionale Struktur des ökologischen Systems. Full-waveform Flugzeuglaserscannerdaten eignen sich hervorragend zur Charakterisierung von Ökosystemen und bilden eine ideale Basis für die vollständige volumetrische Repräsentation der Vegetations- und Gewässerstruktur in einem Voxelraum. Die Voxelattribute werden dabei aus der digitalisierten Wellenform abgeleitet. Jeder emittierte Laserpuls wird von Dämpfungseffekten beeinflusst, die durch Teilreflexionen auf seinem Weg durch die unterschiedlichen Vegetations- oder Wasserschichten entstehen. Dadurch ist die Struktur im unteren Bereich der empfangenen Rohsignale unterrepräsentiert. Die im Rahmen dieser Arbeit entwickelten innovativen Methoden zur Analyse von full-waveform Daten ermöglichen die Generierung einer radiometrisch korrigierten Voxelraumrepräsentation. Voraussetzung dafür ist die numerisch stabile Rekonstruktion des effektiven differentiellen Rückstreuquerschnitts mit geeigneten Entfaltungs- und Regularisierungsverfahren. Das Kernstück der Analyse bildet die Beschreibung der Signaldämpfung mit Hilfe geeigneter Modelle. Auf Grundlage dieser Modelle wurden neuartige Korrekturverfahren zur Kompensation der Signaldämpfung erarbeitet, wobei der Korrekturterm direkt aus dem differentiellen Rückstreuquerschnitt abgeleitet wird. Die Grundidee der entwickelten Methode ist das schrittweise Anheben der Signalintensität in Abhängigkeit von der individuellen Historie jedes Laserpulses. Die Resultate der vorliegenden Arbeit tragen dazu bei, die in full-waveform Daten enthaltenen Informationen über die Vegetations- und Gewässerstruktur zugänglich zu machen. Weiterhin zeigen die hier präsentierten Ergebnisse, dass die Limitierungen bestehender Auswertemethoden, welche weitgehend auf die Extraktion diskreter Maxima und die Erzeugung volumetrischer Repräsentationen aus diskreten 3D Punktwolken beschränkt sind, überwunden werden können. / The scientific investigation of terrestrial and aquatic ecosystems requires precise information on the three-dimensional structure of the ecologic system. Full-waveform airborne laser scanner data are an ideal basis for the complete volumetric representation of vegetation and water structure in a voxel space. Due to attenuation effects, caused by partial reflections during the laser pulse propagation through the vegetation or water column, each individual laser pulse echo is significantly modified. As a result, the structure in the lower parts of the vegetation or water column is underrepresented in the digitized waveform. Within this research, novel and innovative methods were developed, which enable the generation of a radiometrically correct voxel space representation. Therefore, a numerically stable reconstruction of the effective differential backscattering cross section utilizing appropriate deconvolution and regularization techniques is required. The essential element of the analysis is the description of the signal attenuation using applicable mathematical models. For this purpose, novel correction methods compensating the signal attenuation based on these models were developed. The correction term is directly derived from the differential backscatter cross section. The basic idea is a gradually increase of the signal amplitudes depending on the individual history of each laser pulse. The results gained in this work contribute to an improved access to the information on vegetation and water structure, contained in full-waveform laser scanner data. Furthermore, it is possible to overcome limitations of existing approaches, which are mainly based on the extraction of discrete maxima.