We analyze recordings of seismic events induced by underground mining operations at the Moonee Colliery, located in Australia, and at the Springfield Pike Limestone Quarry, located in the United States. The data were recorded underground in the vicinity of active mining operations and were initially used by the monitoring systems at the two mines to evaluate the potential for roof failure based on the temporal and spatial distribution of the seismic activity. In an earlier study, the roof failures at the two underground mine locations were found to be the result of two distinct failure processes, both of which were correlated with escalated seismic activity before the roof collapsed. In this study, we reexamine the recordings of these seismic emissions for a further assessment of the state of instability in the roof. We estimate the static seismic moment and radiated seismic energy for each recorded seismic event induced by mining operations at the two underground mine locations. These seismic source parameters are estimated from source spectra have been corrected for the instrument response, propagation effects and bandwidth limitations. The apparent stress, which provides an estimate of the stress drop (or stress release) associated with a seismic event, is then determined from the product between the modulus of rigidity and the ratio between the radiated seismic energy and static seismic moment. The validity of constant stress drop scaling for the seismic events at the two underground mine locations is tested. Estimation of the seismic source parameters indicate that the stress drop of the mining induced events increases over three orders of magnitude of increasing seismic moment (106 N·m ≤ M0 ≤ 109 N·m) and indicate a divergence from constant stress drop scaling. When these results are compiled with the results from seven other independent studies, which analyzed the seismicity associated with a variety of seismogenic environments, this trend is found to span over ten orders of magnitude of seismic moment (106 N·m ≤ M0 ≤ 1016 N·m). The observation that the mining induced events do not conform to constant stress drop scaling may assist in gaining a better understanding of the evolution of the roof failure process. We have found that the stress drop at one of the studied mines appears to increase through time prior to a roof collapse. More data are necessary to test this hypothesis. If this hypothesis is validated, it would have important implications for monitoring roof stability. Incorporation of near-real-time estimates of the stress drop into the existing seismic monitoring protocol may provide improved warning of imminent roof collapse hazards. / Master of Science

To better understand the ground motion and associated seismic hazard of earthquakes in the central and eastern United States (CEUS), this dissertation focuses on the source parameters and wave propagation characteristics of both tectonic earthquakes and induced earthquakes in the CEUS. The infrequent occurrence of significant earthquakes in the CEUS limits the necessary observations needed to understand earthquake processes and to reduce uncertainty in seismic-hazard maps. The well-recored aftershock sequence of the 2011 Mineral, Virginia, earthquake offers a rare opportunity to improve our understanding of earthquake processes and earthquake hazard in this populous region of the United States. Moreover, the rapid increase of seismicity in the CEUS since 2009 that has been linked to wastewater injection has raised concern regarding the potential hazard. In this dissertation, I first present a detailed study of the aftershock sequence of the 2011 Mw 5.7 Mineral, Virginia earthquake. It involves the hypocenter locations of ~3000 earthquakes, ~400 focal mechanism solutions, statistics of the aftershock sequence, and the Coulomb stress modeling that explains the triggering mechnanism of those aftershocks. Second, I examine the S-wave attenuation at critical short hypocentral distances (< 60 km) using the aftershock data. The observed S-wave amplitudes decay as a function of hypocenter distance R according to R^-1.3 - R^-1.5, which is substantially steeper than R^-1 for a homogeneous whole space. Finally, I propose and apply a stable multi-window coda spectral ratio method to estimate corner frequencies and Brune-type stress drops for the 2011 Mineral, Virginia mainshock and aftershocks, as well as induced earthquakes in Oklahoma. The goal of this comparative study is to find out whether or not there are systematical differences in source parameters between tectonic earthquakes and induced earthquakes in the CEUS. I found generally much higher stress drops for the Mineral, Virginia sequence. However, the stress drops for those induced earthquakes in Oklahoma exhibit large varation among individual earthquake sequences, with the large mainshocks having high stress drops (20-30 MPa, Brune-type) except for the 2011 Mw 5.6 Prague, Oklahoma earthquake. And spatially varying stress drops indicates strong fault heterogeneity, which in the case of induced earthquakes may be influenced by the injection of fluids into the subsurface. / Ph. D.

The state of stress in forearc regions depends on the balance of two competing factors: the plate coupling force that generates margin-normal compression, and the gravitational force, that generates margin-normal tension. Widespread reversal of the focal mechanisms of small earthquakes after the 2011 Tohoku-oki earthquake indicate a reversal in the dominant state of stress of the forearc, from compressive before the earthquake to tensional afterwards. This implies that the plate coupling force dominated before the earthquake, and that the coseismic weakening of the fault lowered the amount of stress exerted on the forearc, such that the gravitational force became dominant in the post-seismic period. This change requires that the average stress drop along the fault represents a significant portion of the fault strength. Two cases are possible: (1) The fault was strong and the stress drop was large or nearly-complete (e.g. from 50 MPa to 10 MPa), or (2) that the fault was weak and the stress drop was small (e.g. from 15 MPa to 10 MPa). The first option appears to be consistent with the dramatic weakening associated with high-rate rock friction experiments, while the second option is consistent with seismological observations that large earthquakes are characterized by low average stress drops. In this work, we demonstrate that the second option is correct. A very weak fault, represented by an apparent coefficient of friction of 0.032, is sufficient to put the Japan Trench forearc into margin-normal compression. Lowering this value by ~0.01 causes the reversal of the state of stress as observed after the earthquake. A slightly stronger fault, with a strength of 0.045, does not agree well with the observed spatial extent of normal faulting for the same coseismic reduction in strength. We also calculate distributions of stress change on the fault and average stress drop values for the Tohoku-oki earthquake, as predicted from 20 published rupture models which were constrained by seismic, tsunami, and geodetic data. Our results reconcile seismic observations that average stress drops for large megathrust events are low with laboratory work on high-rate weakening that predicts very high or complete stress drop. We find that, in all rupture models, regions of high stress drop (20 – 55 MPa) are probably indicative of dynamic weakening during seismic slip, but that the heterogeneous nature of fault slip does not allow these regions to become widespread. Also, coseismic stress increase on the fault occurs in many parts of the fault, including parts of the area that experienced high slip (> 30 m). These two factors ensure that the average stress drop remains low (< 5 MPa). The low average stress drop during the Tohoku earthquake, consistent with values reported for other large earthquakes, makes it unambiguous that the Japan Trench megathrust is very weak. / Graduate

Cette thèse traite de l'étude de la profondeur des séismes modérés (4 < M < 5.5), à travers l'utilisation de données telesismiques, c'est à dire à partir d'enregistrements de sismographes situes a des distances de 3000 jusqu'à 9000 km de la source. L'estimation de la profondeur d'un séisme s'obtient en comparant les temps d'arrivée de l'onde directe (P) générée par ce séisme, avec ceux des ondes réfléchies sur la surface de la Terre, au-dessus du foyer du séisme (les phases de profondeur pP, sP). Nous utilisons le réseau du CTBTO (Comprehensive Test-Ban-Treaty-Organization, composé de mini-réseaux (10-20 sismographes), nous permettant de développer deux nouvelles méthodes d'estimation de la profondeur. La première est une inversion complète du mécanisme au foyer, simultanément a la profondeur. La seconde est une méthode d'analyse spectrale : le cepstre. A l'aide d'exemples de séismes en zone intracontinentale, nous montrons que ces deux méthodes sont complémentaires et qu'elles apportent une information nouvelle sur l'estimation de la profondeur pour les séismes de régions peu instrumentées. Dans un second temps, une optimisation de la méthode cepstrale a permis d'étendre l'estimation de la profondeur a l'utilisation de stations isolées provenant du réseau mondial IRIS. Cette nouvelle méthode, complètement automatique, a permis de localiser en profondeur les séismes de magnitude supérieure à 4.5 pour la zone de subduction de Guerrero (Mexique). Nous avons ainsi mis en évidence une répartition homogène des profondeurs des séismes dans la zone du gap de Guerrero. En combinant ces estimations de la profondeur de l'interface avec celles obtenues a l'aide d'une relocalisation de l'ensemble de la sismicité (par la méthode de l'ISC-Locator), nous proposons une imagerie de la géométrie de la subduction. Nous avons ensuite étudié l'influence de la profondeur sur les propriétés sismiques des séismes. En particulier, nous avons évalué la relation entre la profondeur et la chute de contrainte pour les séismes récents et modérés de la plaine du Pô, en Italie. Pour cela, nous nous sommes appuyés sur le réseau accéléromètrique régional de l'INGV Milan pour estimer les spectres sources de ces séismes, puis en déduire les fréquences coins et les chutes de contraintes associées. Ces spectres sources s'obtiennent à l'aide d'une méthode d'inversion qui sépare simultanément l'effet de l'atténuation, les effets de sites et l'effet de la source sur les spectres des ondes S, générées par les séismes et enregistrées en surface par les accéléromètres. Après inversion, nous trouvons une faible augmentation de la chute de contrainte avec la profondeur. / This thesis deals with depth estimations of moderate earthquakes (4 < M < 5.5), observed through the use of teleseismic data. At teleseismic distances (from 3000 to 9000 km), estimations of earthquake depths come from the estimation of the delays between the arrival time of the direct P-wave and the arrival times of the waves which have been reflected on the surface above the source (known as : depth phases pP and sP). The CTBTO (Comprehensive Test-Ban- Treaty-Organization) monitoring system allows the detection of these teleseismic phases for weak events, with magnitudes below 5, since this global network is composed by arrays (10-20 close single stations, in the same location). Using this network, two methods were developed for teleseismic depth estimation. First, a depth-phase recognition method is applied, based on a new improved cepstral analysis. In addition, we have developed a focal mechanism genetic algorithm inversion. We have applied these two methods for earthquakes occurring in intracontinental areas and we have proved that these new depth determinations provide new and complementary information about the source for barely instrumented areas. We have also developed another method, completely blind and automatic, which consists in an improvement of the cepstral analysis. The reliability of this method to improve depth estimation has been proved by relocating the recent moderate seismicity of the Guerrero subduction area (Mexico). In the Guerrero area, this cepstral analysis efficiently clusters event locations. We then use teleseismic waves reflected on the subduction interface to evaluate the depth of the interface above the focal hypocenter, which allows to specify the subduction interface geometry. We have finally conducted an exhaustive analysis of the depth distributions, based on a relocalization of the ISC catalogue using the new ISC-Locator algorithm and we provide an improved image of the subduction. Finally, we have conducted a study of the relationship between earthquake depths and seismic properties. In particular, we have focused on the burning issue about the links between stress drops and focal depths. We have found an interesting case study with the Po Plain recent seismicity (Northern Italy). Brune's stress drops and magnitudes have been estimated from a generalized parametric inversion using the regional accelerometric network from INGV Milano. We have shown that there is only a slight dependency between stress drops and earthquake depths in the Po Plain.