Characterization of fractures in an arkosic sandstone from the western damage zone of the San Andreas Fault (SAF) at San Andreas Fault Observatory at Depth (SAFOD) was used to better understand the origin of damage and to determine the scale dependence of fracture fabric and fracture density. Samples for this study were acquired from core taken at approximately 2.6 km depth during Phase 1 drilling at SAFOD. Petrographic sections of samples were studied using an optical petrographic microscope equipped with a universal stage and digital imaging system, and a scanning electron microscope with cathodoluminescence (SEM-CL) imaging capability. Use of combined optical imaging and SEM-CL imaging was found to more successfully acquire true fracture density at the grain scale. Linear fracture density and fracture orientation were determined for transgranular fractures at the whole thin section scale, and intragranular fractures at the grain scale. The microscopic scale measurements were compared to measurements of mesoscopic scale fractures in the same core, as well as to published data from an ancient, exhumed trace of the SAF in southern California. Fracturing in the damage zone of the SAF fault follows simple scaling laws from the grain scale to the km scale. Fracture density distributions in the core from SAFOD are similar to distributions in damaged arkosic sandstone of the SAF along other traces. Transgranular fractures, which are dominantly shear fractures, indicate preferred orientation approximately parallel to the dominant sets of the mesoscale faults. Although additional work is necessary to confirm general applicability, the results of this work demonstrate that fracture density and orientation distribution over a broad range of scales can be determined from measurements at the mesoscopic scale using empirical scaling relations.

Recovered core samples from the San Andreas Fault Observatory at Depth (SAFOD), located near Parkfield, CA, offer a unique opportunity to study the products of faulting and to learn about the mechanisms of slip at 3 km depth. Casing deformation reflects active creep along two strands of the San Andreas Fault (SAF) at SAFOD. The two fault strands are referred to as the Southwest Deforming Zone (SDZ) at 3194 m measured depth (MD) and the Central Deforming Zone (CDZ) at 3301 m MD. The SDZ and CDZ contain remarkably similar gouge layers, both of which consist of a clay-bearing, ultrafine grain matrix containing survivor clasts of sandstone and serpentinite. The two gouges have sharp boundary contacts with the adjacent rocks. We have used X-ray Computed Tomography (XCT) imaging, at two different sampling resolutions, to investigate the mesoscale and microscale structure of the fault zone, specifically to characterize the shape, preferred orientation, and size distribution of the survivor clasts. Using various image processing techniques, survivor clast shape and size are characterized in 3D by best-fit ellipsoids. Renderings of survivor clasts illustrate that survivor clasts have fine tips reminiscent of sigma type tails of porphyroclasts observed in myolonites. The resolution of the XCT imaging permits characterization of survivor clasts with equivalent spherical diameters greater than 0.63 mm. The survivor clast population in both the SDZ and CDZ gouge layers have similar particle size distributions (PSD) which fit a power law with a slope of approximately -3; aspect ratio (major to minor axis ratios) distributions also are similar throughout ranging between 1.5 and 4, with the majority occurring between 2-2.5. The volume- and shape- distributions vary little with position across the gouge zones. A strong shape preferred orientation (SPO) exists in both creeping zones. In both the SDZ and CDZ the minor axes form a SPO approximately normal to the plane of the San Andreas Fault (SAF), and the major axes define a lineation in the plane of the SAF. The observation that the size-, shape- and orientation-distributions of mesoscale, matrix-supported clasts are similar in the SDZ and CDZ gouge layers, and vary little with position in each gouge layer, is consistent with the hypothesis that aseismic creep in the SDZ and CDZ is achieved by distributed, shearing. The consistency between the SPO and simple-shear, strike-slip kinematics, and the marked difference of PSD, fabric, cohesion and clast lithology of the gouge with that of the adjacent rock, is consistent with the hypothesis that the vast majority of the shear displacement on the SAF at SAFOD is accommodated within the gouge layers and the gouge displays a mature, nearly steady-state structure.

Les processus physiques et chimiques activés pendant le cycle sismique déterminent l'évolution des propriétés mécaniques des failles, à court terme (pendant un séisme) comme à long terme (la récupération des propretés élastiques des roches de faille après un seisme). L'étude des roches de faille naturelles est un moyen pour identifier les processus actives pendant les diverses phases des cycle séismique. En cette thèse, échantillons prévenants de deux failles séismiques sont étudiés: la Faille de San Andreas (California, USA), une faille séismique active, et la faille de Gole Larghe (Alpes Méridionales, Italie), une faille séismique exhumée. La Faille de San Andreas a été forée jusqu'à 2.7km de profondeur. Les échantillons montrent une superposition de: pression-dissolution - hydrofracturation - pression dissolution. La succession des évents est compatible avec la formation de sacs de fluides dans zones de basse perméabilité dans la faille, ou la pression de fluides augmente à cause de le progressif compactage de le gouge de faille, jusqu'à la nucléation de une rupture. La faille de Gole Larghe est une faille exhumée, qui a préservé des pseudotachylytes (roches fondues par le chaleur de friction pendant une frottement séismique) formées à 9 - 11 km de profondeur il y a 30 millions d'années. Deux argumentes sont traités: (i) l'évolution des microstructures des cataclasites associées à les pseudotachylytes, pour identifier les processus qui peuvent porter à la formation de instabilités frictionnelles pendant les premières phases de croissance de une faille. (ii) L'origine des fluides en failles séismiques et pendant la fusion pour friction. La formation de un système de failles à cataclasites permit la percolation de un fluide aqueux de profondeur. La composition isotopique des pseudotachylytes (calculé sans la component de hydratation) est proche à celle des pseudotachylytes reproduites en expériences du laboratoire (sans fluides). La principale source de fluides pendant la fusion pour friction est donc la déshydratation des minéraux hydraté des roches autour de la faille. / The time recurrence of earthquakes is the result of the feedback between the tectonic loading and the evolution of fault strength during the seicmic cycle. This thesis aims to identify the chemical and physical processes in fault rocks from the modern seismogenic San Andreas Fault (California, USA) and the ancient seismogenic Gole Larghe Fault (Southern Alps, Italy). The San Andreas Fault was drilled to 2.7 km depth, and samples were extracted from the depth of nucleation of repeating microearthquakes. A cyclic recurrence of pressure-solution creep – hydrofracture - pressure solution creep supports the idea that isolated compartments of high fluid pressure might cause the nucleation of small to moderate size earthquakes, associated with the dominant creeping activity in this fault segment. The Gole Larghe Fault Zone was active 30 Ma ago at 9 – 11 km depth. The occurrence of pseudotachylytes witnesses its seismic behavior. Two topics were investigated: (i) The fabric evolution of cataclastic rocks with increasing deformation, to identify the processes potentially leading to the onset of unstable slip at the early stages of fault growth. (ii) The origin of fluids involved in seismic faulting and frictional melting. The formation of a cataclastic fault network allows the ingression of external hydrous fluids, probably of deep origin. The similar isotopic composition of natural pseudotachylytes and pseudotachylytes produced in dry conditions suggests that the fluid source is the dehydration of OH-bearing minerals in the wall rocks induced by coseismic frictional heating.

The quest to answer fundamental questions and solve complex problems is a principal tenet of Earth science. The pursuit of scientific knowledge has generated profuse research, resulting in a plethora of information-rich resources. This phenomenon offers great potential for scientific discovery. However, a deficiency in information connectivity and processing standards has become evident. This deficiency has resulted in a demand for tools to facilitate and process this upsurge in information. This ontology project is an answer to the demand for information processing tools. The primary purpose of this domain-specific ontology and knowledge base is to organize, connect, and correlate research data related to brittle deformation microstructures. This semantically enabled ontology may be queried to return not only asserted information, but inferred knowledge that may not be evident. In addition, its standardized development in OWL-DL (Web Ontology Language-Description Logic) allows the potential for sharing and reuse among other geologic science communities.

The aim of the San Andreas Fault Observatory at Depth (SAFOD) project, a component of the NSF Earthscope Initiative, is to directly observe active fault processes at seismogenic depths through the drilling of a 3 km deep (true vertical depth) inclined borehole across San Andreas fault. Preliminary subsurface models based on surface mapping and geophysical data predicted different lithologies than were actually encountered. At 1920 meters measured depth (mmd), a sequence of well-indurated, interbedded arkosic conglomerates, sandstones, and siltstones was encountered. We present a detailed lithologic and structural characterization as a step toward understanding the complex geologic history of this fault-bounded block of arkosic sedimentary rocks. We divide the arkosic section into three lithologic units with different compositional, structural, and sedimentary features: the upper arkose, 1920-2530 mmd, the clay-rich zone, 2530-2680 Illtlld, and the lower arkose, 2680-3150 mmd. We interpret the section to have been deposited in a Salinian transtensional basin, in either a subaqueous or subaerial fan setting. We suggest four different possibly equivalent sedimentary units to the SAFOD arkoses, the locations of which are dependent on how the San Andreas fault system has evolved over time in the vicinity of the SAFOD site. Detailed analysis of three subsidiary faults encountered in the arkosic section at 1920 mmd, 2530 mmd, and 3060 mmd, shows that subsurface faults have similar microstructures and composition as exhumed faults at the surface, with less evidence of alteration from extensive fluid flow.

Spot core from the San Andreas Fault Observatory at Depth (SAFOD) borehole provides the opportunity to characterize and quantify damage and mineral alteration of siliciclastics within an active, large-displacement plate-boundary fault zone. Deformed arkosic, coarse-grained, pebbly sandstone, and fine-grained sandstone and siltstone retrieved from 2.55 km depth represent the western damaged zone of the San Andreas Fault, approximately 130 m west of the Southwest Deforming Zone (SDZ). The sandstone is cut by numerous subsidiary faults that display extensive evidence of repeating episodes of compaction, shear, dilation, and cementation. The subsidiary faults are grouped into three size classes: 1) small faults, 1 to 2 mm thick, that record an early stage of fault development, 2) intermediate-size faults, 2 to 3 mm thick, that show cataclastic grain size reduction and flow, extensive cementation, and alteration of host particles, and 3) large subsidiary faults that have cemented cataclastic zones up to 10 mm thick. The cataclasites contain fractured host-rock particles of quartz, oligoclase, and orthoclase, in addition to albite and laumontite produced by syn-deformation alteration reactions. Five structural units are distinguished in the subsidiary fault zones: fractured sandstones, brecciated sandstones, microbreccias, microbreccias within distinct shear zones, and principal slip surfaces. We have quantified the particle size distributions and the particle shape of the host rock mineral phases and the volume fraction of the alteration products for these representative structural units. Shape characteristics vary as a function of shear strain and grain size, with smooth, more circular particles evolving as a result of increasing shear strain. Overall, the particle sizes are consistent with a power law distribution over the particle size range investigated (0.3 µm < d < 400 µm). The exponent (fractal dimension, D) is found to increase with shear strain and volume fraction of laumontite. This overall increase in D and evolution of shape with increasing shear strain reflects a general transition from constrained comminution, active at low shear strains to abrasion processes that dominate at high shear strains.

Les processus physiques et chimiques activés pendant le cycle sismique déterminent l'évolution des propriétés mécaniques des failles, à court terme (pendant un séisme) comme à long terme (la récupération des propretés élastiques des roches de faille après un seisme). L'étude des roches de faille naturelles est un moyen pour identifier les processus actives pendant les diverses phases des cycle séismique. En cette thèse, échantillons prévenants de deux failles séismiques sont étudiés: la Faille de San Andreas (California, USA), une faille séismique active, et la faille de Gole Larghe (Alpes Méridionales, Italie), une faille séismique exhumée. La Faille de San Andreas a été forée jusqu'à 2.7km de profondeur. Les échantillons montrent une superposition de: pression-dissolution - hydrofracturation - pression dissolution. La succession des évents est compatible avec la formation de sacs de fluides dans zones de basse perméabilité dans la faille, ou la pression de fluides augmente à cause de le progressif compactage de le gouge de faille, jusqu'à la nucléation de une rupture. La faille de Gole Larghe est une faille exhumée, qui a préservé des pseudotachylytes (roches fondues par le chaleur de friction pendant une frottement séismique) formées à 9 - 11 km de profondeur il y a 30 millions d'années. Deux argumentes sont traités: (i) l'évolution des microstructures des cataclasites associées à les pseudotachylytes, pour identifier les processus qui peuvent porter à la formation de instabilités frictionnelles pendant les premières phases de croissance de une faille. (ii) L'origine des fluides en failles séismiques et pendant la fusion pour friction. La formation de un système de failles à cataclasites permit la percolation de un fluide aqueux de profondeur. La composition isotopique des pseudotachylytes (calculé sans la component de hydratation) est proche à celle des pseudotachylytes reproduites en expériences du laboratoire (sans fluides). La principale source de fluides pendant la fusion pour friction est donc la déshydratation des minéraux hydraté des roches autour de la faille.