On September 24, 1996, President Clinton signed the Comprehensive Test Ban Treaty (CTBT), which bans the testing of all nuclear weapons thereby limiting their future development. Seismology is the primary tool used for the detection and identification of underground explosions and thus, will play a key role in monitoring a CTBT. The detection and identification of low yield explosions requires seismic stations at regional distances (1500 km). However, because the regional wavefield propagates within the extremely heterogeneous crustal waveguide, the seismic waveforms are also very complicated. Therefore, it is necessary to have a solid understanding of how the phases used in regional discriminants develop within different tectonic regimes. Thus, the development of the seismic phases Pn and Lg, which compose the seismic discriminant Pn/Lg, within the western U.S. from the Non-Proliferation Experiment are evaluated. The most fundamental discriminant is event location as 90% of all seismic sources occur too deep within the earth to be unnatural. France resumed its nuclear testing program after a four year moratorium and conducted six tests during a five month period starting in September of 1995. Using teleseismic data, a joint hypocenter determination algorithm was used to determine the hypocenters of these six explosions. One of the most important problems in monitoring a CTBT is the detection and location of small seismic events. Although seismic arrays have become the central tool for event detection, in the context of a global monitoring treaty, there will be some dependence on sparse regional networks of three-component broadband seismic stations to detect low yield explosions. However, the full power of the data has not been utilized, namely using phases other than P and S. Therefore, the information in the surface wavetrain is used to improve the locations of small seismic events recorded on a sparse network in Bolivia. Finally, as a discrimination example in a complex region, P to S ratios are used to determine source parameters of the Mw 8.3 deep Bolivia earthquake.

Modern broadband seismic instrumentation has provided enormous opportunities to retrieve the information in almost any frequency band of seismic interest. In this thesis, we have investigated the long period responses of the broadband seismometers and the problem of recovering actual groundmotion. For the first time, we recovered the static offset for an earthquake from dynamic seismograms. The very long period waves of near- and intermediate-field term from 1994 large Bolivian deep earthquake (depth = 630km, MW and 1997 large Argentina deep earthquake (depth = 285km, MW are successfully recovered from the portable broadband recordings by BANJO and APVC networks. These waves provide another dynamic window into the seismic source process and may provide unique information to help constrain the source dynamics of deep earthquakes in the future. We have developed a new method to locate global explosion events based on broadband waveform stacking and simulated annealing. This method utilizes the information provided by the full broadband waveforms. Instead of "picking times", the character of the wavelet is used for locating events. The application of this methodology to a Lop Nor nuclear explosion is very successful, and suggests a procedure for automatic monitoring. We have discussed the problem of deep earthquakes from the viewpoint of rock mechanics and seismology. The rupture propagation of deep earthquakes requires a slip-weakening process unlike that for shallow events. However, this process is not necessarily the same as the process which triggers the rupture. Partial melting due to stress release is developed to account for the slip-weakening process in the deep earthquake rupture. The energy required for partial melting in this model is on the same order of the maximum energy required for the slip-weakening process in the shallow earthquake rupture. However, the verification of this model requires experimental work on the thermodynamic properties of rocks under non-hydrostatic stress. The solution of the deep earthquake problem will require an interdisciplinary study of seismology, high pressure rock mechanics, and mineralogy.

This dissertation uses a finite element technique to explore the role of viscoelastic behavior in a wide range of plate tectonic processes. We consider problems associated with spreading centers, earthquake triggering, and subduction zone dynamics. We simulated the evolution of a slow-spreading center upon cessation of active spreading in order to predict long-term changes in the axial valley morphology. Results suggest that the axial valley created at a slow-spreading center persists because the crust is too strong to deform ductily and because no effective mechanism exists to reverse the topography created by rift-bounding normal faults. These results suggest that the persistence of axial valleys at extinct spreading centers is consistent with a lithospheric stretching model based on dynamic forces for active slow-spreading ridges. In our study of earthquake triggering, results suggest that if a ductile lower crust or upper mantle flows viscously following a thrust event, relaxation may cause a transfer of stress to the upper crust. Under certain conditions this may lead to further increases and a lateral expansion of high Coulomb stresses along the base of the upper crust. Analysis of experimentally determined non-Newtonian flow laws suggests that wet granitic, quartz, and feldspar aggregates may yield a viscosity on the order of 1019 Pa-s. The calculated rate of stress transfer from a viscous lower crust or upper mantle to the upper crust becomes faster with increasing values of the power law exponent and the presence of a regional compressive strain rate. In our study of subduction zone dynamics, we model the density and strength structures that drive the Nazca and South American plates. Results suggest that chemical buoyancy and phase changes associated with a cool subducting slab strongly influence the magnitude of driving forces, and the downgoing slab behaves weaker than the strength that would be expected based solely on temperature. Additionally, results suggest that large stresses are produced on the western margin of South American due to forces associated with asthenospheric cornerflow. These forces may be responsible for the high topography of the South American Cordilleran.

I have developed a three-dimensional, lithospheric-scale model of the Bolivian Andes at ∼20°S. The model is based on tomographic images of velocity and attenuation for both P- and S-waves. Observations of travel-time and attenuation for this study are from regional, mantle earthquakes in the subducted Nazca plate recorded on a portable, broadband seismic array in Bolivia. The shallow mantle under the Altiplano from ∼18°S to ∼20°S is high velocity, but seismic Q is relatively low (Vp∼8.3, Vs∼4.9, Qp∼150, Qs∼100). These seismic properties suggest lithospheric mantle, approaching solidus conditions. High velocity material in the Altiplano extends to a depth of ∼150 km at 18.5°S, shallowing to ∼100 km at 20.5°S. Anomalously low velocity and Q anomalies are imaged in the mantle beneath the Eastern (fold and thrust) and Western (volcanic arc) Cordilleras of Bolivia. In the Western Cordillera, velocity and attenuation anomalies are locally strong (Vp∼7.8, Vs∼4.25, Qp∼80, Qs∼20), consistent with partial melt conditions. However, there is a segment of higher velocity and Q between 19°S and 20°S, that is correlated with reduced Quaternary arc volcanism. In the Eastern Cordillera, shallow mantle velocity and Q generally decrease from Altiplano values, but there is a localized low velocity and Q anomaly (Vp∼7.8, Vs∼4.1, Qp∼50, Qs∼10) underneath the Los Frailes volcanic center. The strong velocity and attenuation anomalies and the spatial correlation with the volcanic complex favor an interpretation of partial melt. From the subduction trench into parts of the Western Cordillera, processes associated with Nazca Plate subduction dominate shallow mantle structure. Structural constrains from the tomographic results and the geologic history of the Bolivian Andes favor a model of lithospheric shortening for the development of shallow-mantle structure in the Eastern Cordillera and Altiplano. A delamination/partial-continental-subduction process is favored for the production of both shallow mantle structure and volcanism in the Eastern Cordillera. This process may remove mafic components in thickened lower crust, refining the crust towards a more felsic composition.

<p>The convergent motion of the Pacific and North American Plates in Alaska has produced geologic features associated with subduction zones and has transported displaced terranes along the Queen Charlotte-Fairweather fault system that forms the northeastern boundary of the Pacific Plate. These subduction features stop abruptly at the edge of the Yakutat Block displaced terrane, approximately 300 km from the Queen Charlotte-Fairweather fault. The purpose of this study was to determine the type and geometry of the boundary between the Yakutat Block and North American as well as the cause of the offset volcanic arc and missing Wadati-Benioff zone. </p><p> I calculated P and S-wave receiver functions for 57 broadband seismic stations located in southeast Alaska. S-wave data was migrated using a Common Conversion Point procedure. P-wave data was imaged via a three-dimensional, pre-stack migration using plane-wave decomposition weighted by an inverse generalized Radon transform to calculate the scattering potential for each event. I also calculated the temperatures at the top of the Yakutat slab and mantle wedge using three different analytical thermal models. </p><p> The 3-D images and animations I produced show that the Yakutat Block is being subducted, continuous with the Pacific slab, and extends to the Queen Charlotte-Fairweather strike-slip fault systems. The subducted slab extends north to the Wrangell Volcanic Field with a dip than increases gradually from 10&deg; in the west to 15&deg; in the east, stripping approximately 15 km of overlying sediments. The location of the Wrangell Volcanic Field and lack of Wadati-Benioff zone are consistent with the temperatures I calculated for the top of slab and mantle wedge after stripping 15 km of sediment. </p>

While plate tectonic theory has been developed considerably in the last four decades, the of initiation of subduction remains an enigma. In this thesis, I investigate the possible initiation of subduction of North Atlantic oceanic lithosphere at the Scotian basin off east coast of Canada. / On the basis of the relationships between topography, gravity and magnetic anomalies, in both space and Fourier domains, a crustal model is proposed to explain the long- and intermediate-wavelength components of the gravity anomalies of eastern Canada which takes into account the topography, density perturbations in the crust and upper mantle and Moho undulation with certain degree of compensation. No pronounced correlation is observed between the reduced-to-pole magnetic anomalies and topography, as well as between the magnetic anomalies and the vertical gradient of gravity anomalies. The low magnetic susceptibility beneath suture zones indicates thermal demagnetization of the high-magnetic crustal root. / The above topography and gravity analysis method is used to determine the compensation state of the Scotian basin. A 3-D density model that incorporates seawater, sediments, oceanic and continental crust and the lithosphere down to 35 km depth is established, which together with density perturbations inside the crust and mantle of the model, produce the observed gravity anomalies. Regional compensation is required to explain the observations. A 2-D thin elastic plate model is adopted to investigate the flexural response of the lithosphere subjected to the loads due to the sediments and excess density perturbations. The rigidity of the lithosphere beneath the basin is estimated to be about one to two orders of magnitude smaller than elsewhere. The weakening is due to thermal blanketing effects of the thick sediments and the temperature-dependent rheology of lithosphere. The weak lithosphere would have a controlling effect on reactivation of normal faults at the ocean-continent boundary. The possible re-orientation of the spreading ridge of North Atlantic ocean may produce tensional stresses that would decouple the oceanic lithosphere from the continental one at the weak zone, allowing the lithosphere beneath the basin to bend more. Subduction initiation would be facilitated when taking into account further weakening of the lithosphere by extra sediments and subsequent compression. (Abstract shortened by UMI.)

<p> Seismic attenuation is sensitive to the physical properties of the subsurface, which makes attenuation analysis a useful tool for reservoir characterization. In this thesis, I present algorithms for estimating directionally dependent attenuation coefficients and perform asymptotic and numerical analysis of wave propagation in attenuative anisotropic media. </p><p> First, I introduce a methodology to estimate the S-wave interval attenuation coefficient by extending the layer-stripping method of Behura and Tsvankin (2009) to mode-converted (PS) waves. Kinematic reconstruction of pure shear (SS) events in the target layer and the overburden is performed by combining velocity-independent layer stripping with the PP+PS=SS method. Then, application of the spectral-ratio method and the dynamic version of velocity-independent layer stripping to the constructed SS reflections yields the S-wave interval attenuation coefficient in the target layer. The attenuation coefficient estimated for a range of source-receiver offsets can be inverted for the interval attenuation-anisotropy parameters. The method is tested on synthetic data generated with the anisotropic reflectivity method for layered VTI (transversely isotropic with a vertical symmetry axis) media and vertical symmetry planes of orthorhombic media. </p><p> Then, I analyze a cross-hole data set generated by perforation shots set off in a horizontal borehole to induce hydraulic fracturing in a tight gas reservoir. The spectral-ratio method is applied to pairs of traces to set up a system of equations for directionally-dependent effective attenuation. Although the inversion provides clear evidence of attenuation anisotropy, the narrow range of propagation directions impairs the accuracy of anisotropy analysis. The observed variations of the attenuation coefficient between different perforation stages appear to be related to changes in the medium due to hydraulic fracturing and stimulation. </p><p> Important insights into point-source radiation in attenuative anisotropic media can be gained by applying asymptotic methods. I derive the asymptotic Green's function in homogeneous, attenuative, arbitrarily anisotropic media using the steepest-descent method. The saddle-point condition helps describe the behavior of the far field slowness and group-velocity vectors and evaluate the inhomogeneity angle (the angle between the real and imaginary parts of the slowness vector). The results from the asymptotic analysis are compared with those from the ray-perturbation method for P-waves in TI media. </p><p> Finally, I address the problem of efficient viscoelastic modeling in heterogeneous anisotropic media. The Kirchhoff scattering integral is employed to generate reflected P-waves, with the required Green's functions computed by summation of Gaussian beams. The influence of attenuation on the Gaussian beams is incorporated using ray-perturbation theory. The method is applied to generate synthetic data from a highly attenuative VTI medium above a horizontal reflector and a structurally complex acoustic model with a salt body.</p>

Among the most significant features of satellite magnetic anomaly maps is the association of negative anomalies with deep sedimentary basins, as observed in the Nova Scotia basin. A possible explanation is partial demagnetization of the oceanic lithosphere by thermal blanketing of sediments and subsidence of the lower crust and uppermost mantle. I test this hypothesis by computing magnetization of oceanic lithosphere beneath the Nova Scotia basin based on a detailed thermal evolution model which includes rifting, subsidence and sea-floor spreading. The results confirm the possibility of demagnetization, however it is not substantial to explain the observed negative magnetic anomaly. As an alternative explanation, I propose a model in which the oceanic crust and uppermost mantle have age-varying magnetization and the continent has an average total magnetization contrast of 20000 A greater than the oceanic. This model produces a magnetic anomaly map having first order features which are in good agreement with satellite magnetic anomaly maps.

During the 1940's, the probability of an earthquake occurring with seismic moment, $M sp prime$, was found to follow: Pr($M sp prime > M) sim M sp{-q sb{D}}$ where $q sb{D}$ is the Gutenberg-Richter exponent. Such distributions are connected with spatial scaling; they are now known to be generic features of multifractals, specifically connected with first order multifractal phase transitions. This behaviour has also been taken as the basic feature of self-organized criticality. / Seismographs measure seismic wave magnitudes, seismic moments are then estimated by semi-empirical power law relations. We first show that the horizontal structure of earthquake seismic moment is in accord with our model of it as a space-time multifractal. Evidence is also presented that suggests it is a universal multifractal. An exponent $ eta$ which varies the relation between magnitude and moment is introduced and $q sb{D, eta}$, the Gutenberg-Richter exponent for $ eta$, is calculated. Related to $q sb{D, eta}$, is the effective dimension, $D sb{dressing}$ = 1.3 of the spatial averaging which was determined, along with $q sb{D, eta}$, using earthquake data from the South-Western United States.

<p> Estimating an accurate velocity model is crucial for seismic imaging to obtain a good understanding of the subsurface structure. The objective of this thesis is to investigate methods of velocity analysis by optimizing seismic images. </p><p> A conventional seismic image is obtained by zero-lag crosscorrelation of wavefields extrapolated from a source wavelet and recorded data on the surface using a velocity model. The velocity model provides the kinematic information needed by the imaging algorithm to position the reflectors at correct locations and to focus the image. In complex geology, wave-equation migration is a powerful tool for accurately imaging the earth's interior; the quality of the output image, however, depends on the accuracy of the velocity model. Given such a dependency between the image and model, analyzing the velocity information from the image is still not intuitive and often ambiguous. If the nonzero space- and time-lags information are preserved in the crosscorrelation, the output are image hypercube defined as extended images. Compared to the conventional image, the extended images provide a straightforward way to analyze the image quality and to characterize the velocity model accuracy. </p><p> Understanding the reflection moveout is the key to developing velocity model building methods using extended images. In the extended image space, reflections form coherent objects which depend on space (lags) and time (lags). These objects resemble cones which ideally have their apex at zero space and time lags. The symmetry axis of the cone lies along the time-lag axis. The apex of the cone is located at zero lags only if the velocity model is accurate. This corresponds to the situation when reflection energy focuses at origin in both the space- and time-lag common-image gathers (the slices at zero time and space lags, respectively). When the velocity model is inaccurate, the cone shifts along the time-lag axis. This results in residual moveout in space-lag gathers (zero time-lag slice) and defocusing in time-lag gathers (zero space-lag slice). These phenomena are correlated, and they are a rich source of information for velocity model updates. </p><p> The extended image distortions caused by velocity model errors can be used to design velocity model building algorithms. When the extended image cones shift, the distance and direction of their apex away from zero time lag constrain model errors. This information can be used to construct an image perturbation, from which a slowness perturbation is inverted under the framework of linearized wave-equation migration velocity analysis. Alternatively, one can formulate a non-linear optimization problem to reconstruct the model by minimizing this image error. This approach requires the adjoint-state method to compute the gradient of the objective function, and iteratively update the model in the steepest-descent direction. </p><p> The space-lag subset of extended images has been used to reconstruct the velocity model by differential semblance optimization for a decade. The basis of the method is to penalize the defocusing in the gathers and to focus the reflection energy at zero lags by optimizing the model. The assumption that defocusing is caused by velocity model error is violated where the subsurface illumination is uneven. To improve the robustness and accuracy of the technique, the illumination compensation must be incorporated into the model building. The illumination compensation effectively isolates the defocusing due to uneven illumination or missing data. The key is to construct an illumination-based penalty operator by illumination analysis. Such a penalty automatically downweights the defocusing from illumination effects and allows the inversion to suffer less from the effects of uneven illumination and to take into account only the image error due to inaccurate velocity models. </p><p> One major issue for differential semblance optimization with space-lag gathers is the cost of computing and storing the gathers. To address the problem, extended space- and time-lag point gathers can be used as an alternative to the costlier common-image gathers. The point gathers are subsets of extended images constructed sparsely in subsurface on reflectors. The point gathers share similar reflection moveout characteristics with space-lag gathers, and thus differential semblance optimization can be implemented with such gathers. The point gathers reduce the computational and storage cost required by space-lag gathers especially in 3-D applications. Furthermore, the point gathers avoid the dip limitation in space-lag gathers and more accurately characterize the velocity information for steep reflections.</p>