San Andreas Fault Observatory At Depth Research Articles

A study of fluid overpressure microstructures from the creeping segment of the San Andreas fault

Evidence of episodic fluid overpressure events noted in samples from the San Andreas Fault Observatory at Depth (SAFOD) have remained largely uncorrelated in terms of their collective significance for seismic history of the fault zone. The compositional and microstructural correlations sought in this study could shed light on questions about potential for major seismic events in the creeping segment of the SAF in central California. We used quantitative energy dispersive spectroscopy (EDS), Cathodoluminescence (CL) and Scanning Electron Microscope (SEM) imaging, and electron backscatter diffraction (EBSD) analysis to acquire geochemical and microstructural data from a suite of twenty SAFOD core samples including the damage zone and the active core of the fault. The results indicate intermittent coseismic fluid overpressure events that overprint the background aseismic creep across the fault. Analysis of trace elements and deformation in the coseismic calcite vein generations and their associated hydrothermal mineral phases indicate progressive uplift and exhumation followed by an asymmetric incursion of meteoric water into the damage zone. The same analysis suggests that the actively creeping intervals act as permeability barriers. Our results are in overall agreement with recent studies of the SAF in central California that indicate large seismic events have occurred intermittent with aseismic creep in recent geological time or suggest future potential for such events.

Spatial-temporal characterization of the San Andreas Fault by fault-zone trapped waves at seismic experiment site, Parkfield, California

In this article, we review our previous research for spatial and temporal characterizations of the San Andreas Fault (SAF) at Parkfield, using the fault-zone trapped wave (FZTW) since the middle 1980s. Parkfield, California has been taken as a scientific seismic experimental site in the USA since the 1970s, and the SAF is the target fault to investigate earthquake physics and forecasting. More than ten types of field experiments (including seismic, geophysical, geochemical, geodetic and so on) have been carried out at this experimental site since then. In the fall of 2003, a pair of scientific wells were drilled at the San Andreas Fault Observatory at Depth (SAFOD) site; the main-hole (MH) passed a ~200-m-wide low-velocity zone (LVZ) with highly fractured rocks of the SAF at a depth of ~3.2 km below the wellhead on the ground level (Hickman et al., 2005; Zoback, 2007; Lockner et al., 2011). Borehole seismographs were installed in the SAFOD MH in 2004, which were located within the LVZ of the fault at ~3-km depth to probe the internal structure and physical properties of the SAF. On September 28 2004, a M6 earthquake occurred ~15 km southeast of the town of Parkfield. The data recorded in the field experiments before and after the 2004 M6 earthquake provided a unique opportunity to monitor the co-mainshock damage and post-seismic heal of the SAF associated with this strong earthquake. This retrospective review of the results from a sequence of our previous experiments at the Parkfield SAF, California, will be valuable for other researchers who are carrying out seismic experiments at the active faults to develop the community seismic wave velocity models, the fault models and the earthquake forecasting models in global seismogenic regions.

Geospatial Management and Analysis of Microstructural Data from San Andreas Fault Observatory at Depth (SAFOD) Core Samples

Core samples obtained from scientific drilling could provide large volumes of direct microstructural and compositional data, but generating results via the traditional treatment of such data is often time-consuming and inefficient. Unifying microstructural data within a spatially referenced Geographic Information System (GIS) environment provides an opportunity to readily locate, visualize, correlate, and apply remote sensing techniques to the data. Using 26 core billet samples from the San Andreas Fault Observatory at Depth (SAFOD), this study developed GIS-based procedures for: 1. Spatially referenced visualization and storage of various microstructural data from core billets; 2. 3D modeling of billets and thin section positions within each billet, which serve as a digital record after irreversible fragmentation of the physical billets; and 3. Vector feature creation and unsupervised classification of a multi-generation calcite vein network from cathodluminescence (CL) imagery. Building on existing work which is predominantly limited to the 2D space of single thin sections, our results indicate that a GIS can facilitate spatial treatment of data even at centimeter to nanometer scales, but also revealed challenges involving intensive 3D representations and complex matrix transformations required to create geographically translated forms of the within-billet coordinate systems, which are suggested for consideration in future studies.

Passive Seismic Imaging of Near Vertical Structures Around the SAFOD Site, California, Jointly Using Scattered P and SH Waves

AbstractTo better illuminate structural discontinuities around the San Andreas Fault Observatory at Depth (SAFOD) site, following the previous study of Zhang et al. (2009), we have extended the Generalized Radon Transform (GRT) method to jointly use scattered P and SH waves from abundant local earthquakes recorded by a local dense seismic network. A hybrid imaging condition is applied to extract consistent structures from scattered P and SH images. In this way, more robust structure information can be determined from separate images with potential artifacts. For the transverse component waveforms, coherent S‐S scattered waves after the direct SH waves can be clearly identified, and they are less interfered with by other scattered and converted waves compared to scattered P‐P waves on the vertical component. Similar to Zhang, Wang, et al. (2009), near‐vertical reflectors are imaged on both sides of the San Andreas Fault (SAF) with scattered SH waves, which is generally consistent with the imaging results using scattered P‐P waves from local earthquakes and a wide‐angle active seismic reflection profile. Compared to the P‐P scattering imaging result, the imaging result using scattered SH waves has higher resolution due to shorter S wavelength and cleaner S‐S scattered waveforms, and the SAF, as well as other reflectors around it, is better resolved. Although the resolution of the joint image obtained by combining separate imaging results from different scattered waves may be degraded, it is able to more robustly characterize strong structure discontinuities using passive seismic sources.

Core surprise: What's inside a plate boundary?

Despite the fact that 90% of global seismicity occurs at plate boundary faults, our understanding of their internal structure is lacking. It’s not easy to see inside a plate boundary fault – typically composed of a high-strain fault core surrounded by a fractured damage zone – and when we can, it often requires expensive drilling projects that yield limited information on the internal structure of the whole fault. Understanding the internal structure of large faults is crucial, because their chemical and mechanical properties control how and where earthquakes rupture, nucleate and propagate. This in turn limits the size of the earthquake or the amount of radiated seismic energy, and consequently the severity of surface damage. The 1999 magnitude 7.7 earthquake along the Chelungpu plate boundary fault, for example – the second deadliest earthquake in Taiwan’s recorded history – saw significant variations in slip and ground motion at different locations along the fault which resulted in large local variations in casualties and damage. Subsequent field investigations related these variations to changes in the fault’s structure (i.e., clay width, geometry), which in turn controlled how the fault moved.

Serpentinite‐Rich Gouge in a Creeping Segment of the Bartlett Springs Fault, Northern California: Comparison With SAFOD and Implications for Seismic Hazard

AbstractAn exposure of a creeping segment of the Bartlett Springs Fault (BSF), part of the San Andreas Fault system in northern California, is a ~1.5‐m‐wide zone of serpentinite‐bearing fault gouge cutting through Late Pleistocene fluvial deposits. The fault gouge consists of porphyroclasts of antigorite serpentinite, talc, chlorite, and tremolite‐actinolite, along with some Franciscan metamorphic rocks, in a matrix of the same materials. The Mg‐mineral assemblage is stable at temperatures above 250–300 °C. The BSF gouge is interpreted to have been tectonically incorporated into the fault from depths near the base of the seismogenic zone and to have risen buoyantly to the surface where it is now undergoing right‐lateral displacement. The ultramafic‐rich composition, frictional properties, and inferred mode of emplacement of the BSF serpentinitic gouge correspond to those of the creeping traces of the San Andreas Fault identified in the SAFOD (San Andreas Fault Observatory at Depth) drill hole. This suggests a common origin for creep at both locations. A tectonic model for the source of the ultramafic‐rich materials in the BSF is proposed that potentially could explain the distribution of creep throughout the northernmost San Andreas Fault system.

Continuous Measurement of Stress‐Induced Travel‐Time Variations at SAFOD

Author(s): Yang, C; Niu, F; Daley, TM; Taira, T | Abstract: In situ stress measurement at seismogenic depth is critically important for deciphering fault zone processes. In this study, we conducted a second active-source crosswell field experiment at the Parkfield San Andreas Fault Observatory at Depth (SAFOD) drill site to investigate the detectability of stress-induced seismic velocity changes at the top part of the seismogenic zone. We employed the same configuration of our previous experiments, which deployed a piezoelectric source and a three-component (3C) accelerometer at 1 km deep inside the pilot and main holes, respectively. We also added a hydrophone, which is attached to the source, to monitor the repeatability of the source waveforms. Over a 40-day recording period, we confirmed an ∼0:04% travel-time variation in S wave and coda that roughly follows the fluctuation of barometric pressure. We attributed this correlation to stress sensitivity of seismic velocity and the stress sensitivity is estimated to be 2:0 × 10 −7 Pa −1 , which is approximately two orders of magnitude higher than those measured in laboratory with dry rock samples, but is consistent with our previous results. Our results confirm the hypothesis that substantial cracks and/or pore spaces exist at seismogenic depths and thus may be used to monitor the subsurface stress field with active-source crosswell seismic.

A study of secondary pyrite deformation and calcite veins in SAFOD damage zone with implications for aseismic creep deformation mechanism at depths >3 km

Previous studies of the San Andreas Fault damage-zone samples from the San Andreas Fault Observatory at Depth (SAFOD) have identified a variety of tectonic microstructures including pressure solution cleavage, calcite-sealed fractures vein fabric, and pyrite and anhydrite hydrothermal fracture sealing. Understanding the deformation provenance of the damage zone rocks and operative deformation mechanism(s) based on preserved microstructures provide insight into overall deformation behavior of the entire seismogenic zone in the creeping section of this transform fault. We analysed the deformation of hydrothermal secondary pyrite in connection with network of calcite veins in a sample of foliated ultracataclasites bordering the actively creeping Southwestern Deforming Zone (SDZ), using SEM, EBSD and CL microscopy. The results show that calcite veins associated with the pressure solution cleavage are crosscut by the secondary pyrite deformed under a range of P-T conditions. Relatively undeformed secondary pyrite is found sealing implosion microbreccia. Our review of previously available data indicates that the damage zone rocks may represent a collage of structural and compositional domains from both locked and creeping sections of the SAF. This interpretation together with results of this study suggest that weak-clay frictional deformation mechanism(s) is likely to be the predominant aseismic creep mechanism at depths below the SAFOD.

Grain size‐dependent strength of phyllosilicate‐rich gouges in the shallow crust: Insights from the SAFOD site

AbstractThe San Andreas Fault Observatory at Depth (SAFOD) drilling project directly sampled a transitional (between creeping and locked) segment of the San Andreas Fault at 2.7 km depth. At the site, changes in strain rate occur between periods of coseismic slip (>10−7 s−1) and interseismic creep (10−10 s−1) over decadal scales (~30 years). Microstructural observations of core retrieved from the SAFOD site show throughgoing fractures and gouge‐rich cores within the fractures, evidence of predominantly brittle deformation mechanisms. Within the gouge‐rich cores, strong phases show evidence of deformation by pressure solution once the grain size is reduced to a critical effective grain size. Models of pressure solution‐accommodated creep for quartz‐phyllosilicate mixtures indicate that viscous weakening of quartz occurs during the interseismic period once a critical effective grain size of 1 μm is achieved, consistent with microstructural observations. This causes pronounced weakening, as the strength of the mixture is then controlled by the frictional properties of the phyllosilicate phases. These results have pronounced implications for the internal deformation of fault zones in the shallow crust, where at low strain rates, deformation is accommodated by both viscous and brittle deformation mechanisms. As strain rates increase, the critical effective grain size for weakening decreases, localizing deformation into the finest‐grained gouges until deformation can no longer be accommodated by viscous processes and purely brittle failure occurs.

Nucleation process of magnitude 2 repeating earthquakes on the San Andreas Fault predicted by rate‐and‐state fault models with SAFOD drill core data

AbstractRecent laboratory shear‐slip experiments conducted on a nominally flat frictional interface reported the intriguing details of a two‐phase nucleation of stick‐slip motion that precedes the dynamic rupture propagation. This behavior was subsequently reproduced by a physics‐based model incorporating laboratory‐derived rate‐and‐state friction laws. However, applying the laboratory and theoretical results to the nucleation of crustal earthquakes remains challenging due to poorly constrained physical and friction properties of fault zone rocks at seismogenic depths. Here we apply the same physics‐based model to simulate the nucleation process of crustal earthquakes using unique data acquired during the San Andreas Fault Observatory at Depth (SAFOD) experiment and new and existing measurements of friction properties of SAFOD drill core samples. Using this well‐constrained model, we predict what the nucleation phase will look like for magnitude ∼2 repeating earthquakes on segments of the San Andreas Fault at a 2.8 km depth. We find that despite up to 3 orders of magnitude difference in the physical and friction parameters and stress conditions, the behavior of the modeled nucleation is qualitatively similar to that of laboratory earthquakes, with the nucleation consisting of two distinct phases. Our results further suggest that precursory slow slip associated with the earthquake nucleation phase may be observable in the hours before the occurrence of the magnitude ∼2 earthquakes by strain measurements close (a few hundred meters) to the hypocenter, in a position reached by the existing borehole.

Experimental constraints on the relationship between clay abundance, clay fabric, and frictional behavior for the Central Deforming Zone of the San Andreas Fault

AbstractThe presence of smectite (saponite) in fault gouge from the Central Deforming Zone of the San Andreas Fault at Parkfield, CA has been linked to low mechanical strength and aseismic slip. However, the precise relationship between clay mineral structure, fabric development, fault strength, and the stability of frictional sliding is not well understood. We address these questions through the integration of laboratory friction tests and FIB‐SEM analysis of fault rock recovered from the San Andreas Fault Observatory at Depth (SAFOD) borehole. Intact fault rock was compared with experimentally sheared fault gouge and different proportions of either quartz clasts or SAFOD clasts extracted from the sample. Nano‐textural measurements show the development of localized clay particle alignment along shear folia developed within synthetic gouges; such slip planes have multiples of random distribution (MRD) values of 3.0–4.9. The MRD values measured are higher than previous estimates (MRD 1.5) that show lower degrees of shear localization and clay alignment averaged over larger volumes. The intact fault rock exhibits less well‐developed nano‐clay fabrics than the experimentally sheared materials, and MRD values decrease with smectite content. We show that the abundance, strength, and shape of clasts all influence fabric evolution via strain localization: quartz clasts yield more strongly developed clay fabrics than serpentine‐dominated SAFOD clasts. Our results suggest that (1) both clay abundance and the development of nano‐scale fabrics play a role in fault zone weakening and (2) aseismic creep is promoted by slip along clay shears with >20 wt % smectite content and MRD values ≥2.7.

Elevated time-dependent strengthening rates observed in San Andreas Fault drilling samples

The central San Andreas Fault in California is known as a creeping fault, however recent studies have shown that it may be accumulating a slip deficit and thus its seismogenic potential should be seriously considered. We conducted laboratory friction experiments measuring time-dependent frictional strengthening (healing) on fault zone and wall rock samples recovered during drilling at the San Andreas Fault Observatory at Depth (SAFOD), located near the southern edge of the creeping section and in the direct vicinity of three repeating microearthquake clusters. We find that for hold times of up to 3000 s, frictional healing follows a log-linear dependence on hold time and that the healing rate is very low for a sample of the actively shearing fault core, consistent with previous results. However, considering longer hold times up to ∼350,000 s, the healing rate accelerates such that the data for all samples are better described by a power law relation. In general, samples having a higher content of phyllosilicate minerals exhibit low log-linear healing rates, and the notably clay-rich fault zone sample also exhibits strong power-law healing when longer hold times are included. Our data suggest that weak faults, such as the creeping section of the San Andreas Fault, can accumulate interseismic shear stress more rapidly than expected from previous friction data. Using the power-law dependence of frictional healing on hold time, calculations of recurrence interval and stress drop based on our data accurately match observations of discrete creep events and repeating Mw=2 earthquakes on the San Andreas Fault.

Constraints on paleofluid sources using the clumped-isotope thermometry of carbonate veins from the SAFOD (San Andreas Fault Observatory at Depth) borehole

The San Andreas Fault Observatory at Depth (SAFOD), near Parkfield, California, is a borehole drilled through two active deforming zones of the San Andreas fault, the Southwest Deforming Zone (SDZ) and the Central Deforming Zone (CDZ). These zones accommodate displacement by seismic slip and aseismic creep. Elevated fluid pressures and fluid–rock interactions have been proposed to explain the low apparent strength and aseismic creep observed, but the origin of the fluids and existence of high fluid pressures remains uncertain. We use clumped-isotope thermometry and δ18O–δ13C compositions of calcite in veins to constrain the origin of paleofluids and compare these results to the isotopic composition of modern-day pore fluids from the SAFOD borehole and nearby areas. We observe that: (1) calcite vein temperatures vary from 81 to 134°C, which overlaps the current ambient borehole temperatures of 110–115 °C at sampled depths; (2) vein calcite is not in carbon isotope equilibrium with modern-day pore fluids; (3) the δ18O values of paleofluids close to the SDZ and CDZ, calculated from vein δ18O and temperature data, are not in equilibrium with local modern-day pore waters but approach equilibrium with modern pore waters far from these zones; and (4) syntectonic vein calcite is only in C- and O-isotopic equilibrium with their host rocks within the SDZ and CDZ. Spatial patterns of δ18O and δ13C show little evidence for across-fault fluid-flow. Clumped isotope temperatures are consistent with locally-derived fluid sources, but not with continuous or episodic replenishment of fluids from shallow sedimentary brines or deep fluid sources. Our findings are compatible with flow of meteoric fluids from the southwestern damage zone into the SDZ and CDZ, which would have favored the formation of weak phyllosilicates and contributed to the present day weakness of the two actively deforming zones.

Low resistivity and permeability in actively deforming shear zones on the San Andreas Fault at SAFOD

AbstractThe San Andreas Fault Observatory at Depth (SAFOD) scientific drill hole near Parkfield, California, crosses the San Andreas Fault at a depth of 2.7 km. Downhole measurements and analysis of core retrieved from Phase 3 drilling reveal two narrow, actively deforming zones of smectite‐clay gouge within a roughly 200 m wide fault damage zone of sandstones, siltstones, and mudstones. Here we report electrical resistivity and permeability measurements on core samples from all of these structural units at effective confining pressures up to 120 MPa. Electrical resistivity (~10 Ω‐m) and permeability (10−21 to 10−22 m2) in the actively deforming zones were 1 to 2 orders of magnitude lower than the surrounding damage zone material, consistent with broader‐scale observations from the downhole resistivity and seismic velocity logs. The higher porosity of the clay gouge, 2 to 8 times greater than that in the damage zone rocks, along with surface conduction were the principal factors contributing to the observed low resistivities. The high percentage of fine‐grained clay in the deforming zones also greatly reduced permeability to values low enough to create a barrier to fluid flow across the fault. Together, resistivity and permeability data can be used to assess the hydrogeologic characteristics of the fault, key to understanding fault structure and strength. The low resistivities and strength measurements of the SAFOD core are consistent with observations of low resistivity clays that are often found in the principal slip zones of other active faults making resistivity logs a valuable tool for identifying these zones.

Interpretation ofSWaves Generated by Near-Surface Chemical Explosions at SAFOD

A series of near-surface chemical explosions conducted at the San Andreas Fault Observatory at Depth (SAFOD) were recorded by high-frequency downhole receiver arrays in separate experiments in November 2003 and May 2005. The 2003 experiment involved ∼100 kg shots detonated along a 46-km-long line (Hole–Ryberg line) centered on SAFOD and recorded by 32 three-component geophones in the pilot hole between 0.8 and 2.0 km depth. The 2005 experiment involved ∼36 kg shots detonated at Parkfield Area Seismic Observatory (PASO) stations (at ∼1–8 km offset) recorded by 80 three-component geophones in the main hole between the surface and 2.4 km depth. These data sample the downgoing seismic wavefield and constrain the shallow velocity and attenuation structure, as well as the first-order characteristics of the source. Using forward modeling on a velocity structure designed for the near field, both observed P - and S -wave energy for the PASO shots are identified with the travel times expected for direct and/or reflected phases. Larger-offset recordings from shots along the Hole–Ryberg line reveal substantial SV and SH energy, especially southwest of SAFOD from the source as indicated by P -to- S amplitude ratios. The generated SV energy is interpreted to arise chiefly from P -to- S conversions at subhorizontal discontinuities. This provides a simple mechanism for often-observed low P -to- S amplitude ratios from nuclear explosions in the far field, as originating from strong near-field wave conversions. Online Material: Figures of observed and synthetic seismogram record sections, snapshots of synthetic wavefields, and models of crustal structure.

San Andreas fault zone velocity structure at SAFOD at core, log, and seismic scales

AbstractThe San Andreas Fault (SAF), like other mature brittle faults, exhibits a zone of low seismic velocity hypothesized to result from fluid pressure effects and/or development of a damage zone. To address the relative contributions of these mechanisms in developing low‐velocity zones, we measured P and S wave velocities ultrasonically at elevated confining and pore pressures on core samples from the San Andreas Fault Observatory at Depth (SAFOD). We compared those data to wireline and seismic‐scale velocities to examine the scale dependence of acoustic properties of the fault core and damage zone. Average laboratory P and S wave velocities of the fault gouge at estimated in situ conditions are 3.1 and 1.5 km/s, respectively, consistent with the sonic log from the same intervals. These data show that fault core has intrinsically low velocity, even if no anomalous pore pressure is assumed, due to alteration and mechanical damage. In contrast, laboratory average P and S wave velocities for the damage zone are 4.7 and 2.5 km/s, up to 41% greater than the sonic log in the damage zone. This scale dependence indicates that stress conditions or macroscale features dominate the damage zone's acoustic properties, although velocity dispersion could play a role. Because no pressure anomaly was detected while drilling the SAFOD borehole, we infer that damage at a scale larger than core samples controls the elastic properties of the broader damage zone. This result bolsters other independent lines of evidence that the SAF does not contain major pore fluid overpressure at SAFOD.

Properties and Processes of Crustal Fault Zones: Volume II

Active faults and earthquake rupture zones are largely buried under miles of rocks. Due to the effects of attenuation, typical seismological and other geophysical data lack detailed information on faulting processes. For these reasons as well as the inherent complexity of faulting, many fundamental questions on the structure and processes of fault zones in relation to earthquake properties and generated ground motion remain unanswered. To discuss the state of the art in the field, the 40th Workshop of the International School of Geophysics, titled ‘‘Properties and Processes of Crustal Fault Zones’’, was held May 18‐24, 2013, in the Ettore Majorana Foundation and Centre for Scientific Culture of Erice, Sicily. The papers in a previous volume I and the present volume II discuss observational and theoretical results based on presentations at the workshop and additional contributions that advance the understanding of earthquakes and faults. Topics covered in this volume include imaging of fault zones and the crust with seismic and other signals, microstructural analyses of fault zone rocks, long paleoseismic records of large earthquakes, inferences on stress, stress drops and fault geometries from seismological and geological data, properties of dynamic ruptures, generation and healing of rock damage, temporal changes of seismic properties, postseismic deformation, and scaling of earthquake rupture areas with moment. In the first paper of the volume, Zigone et al. use Rayleigh and Love waves constructed from the ambient seismic noise to obtain a 3D model of S wave velocities in the southern California plate-boundary region. The results show clear damage zones in the top few kilometres around the San Andreas, San Jacinto, and Elsinore faults, and velocity contrasts across various fault sections, which augment earthquake tomography images at greater depth. Additional findings include 2-h azimuthal anisotropy with fast directions parallel to geometrically simple fault sections and mixed directions in complex areas. Bennington et al. develop a method for joint inversion of P wave velocity and electrical resistivity using a normalized cross-gradient constraint, and apply it to obtain a structural profile across the San Andreas Fault Observatory at Depth (SAFOD). The results provide strong constraints on several key geological units and are used to infer possible fluid-rich regions in the fault zone structure. Bradbury et al. describe the microstructure and chemical composition of faultrelated rocks from the SAFOD borehole and surface outcrops that may be useful analogs for San Andreas Fault rocks. The observed rock textures are consistent with largely aseismic deformation punctuated by seismic slip. The mineralogy and whole-rock geochemistry indicate that the fault zone experienced transient fluid‐rock interactions. Ferrarini et al. constrain the fault segmentation pattern and regional stress tensor in the intra-Apennine area of central Italy using geological and seismological fault-slip data associated with the Colfiorito 1997 Mw 6.0 and L’Aquila 2009 Mw 6.1 seismic sequences. The results indicate a prevailing tensional regime involving a subhorizontal NE‐SW minimum stress axis and only minor strike-slip regimes. The findings are at odds with results of other studies indicating wide areas characterized by strike

Joint Inversion of Seismic and Magnetotelluric Data in the Parkfield Region of California Using the Normalized Cross-Gradient Constraint

We present jointly inverted models of P-wave velocity (Vp) and electrical resistivity for a two-dimensional profile centered on the San Andreas Fault Observatory at Depth (SAFOD). Significant structural similarity between main features of the separately inverted Vp and resistivity models is exploited by carrying out a joint inversion of the two datasets using the normalized cross-gradient constraint. This constraint favors structurally similar Vp and resistivity images that adequately fit the seismic and magnetotelluric (MT) datasets. The new inversion code, tomoDDMT, merges the seismic inversion code tomoDD and the forward modeling and sensitivity kernel subroutines of the MT inversion code OCCAM2DMT. TomoDDMT is tested on a synthetic dataset and demonstrates the code’s ability to more accurately resolve features of the input synthetic structure relative to the separately inverted resistivity and velocity models. Using tomoDDMT, we are able to resolve a number of key issues raised during drilling at SAFOD. We are able to infer the distribution of several geologic units including the Salinian granitoids, the Great Valley sequence, and the Franciscan Formation. The distribution and transport of fluids at both shallow and great depths is also examined. Low values of velocity/resistivity attributed to a feature known as the Eastern Conductor (EC) can be explained in two ways: the EC is a brine-filled, high porosity region, or this region is composed largely of clay-rich shales of the Franciscan. The Eastern Wall, which lies immediately adjacent to the EC, is unlikely to be a fluid pathway into the San Andreas Fault’s seismogenic zone due to its observed higher resistivity and velocity values.

A scaling law to characterize fault-damage zones at reservoir depths

We analyze fracture-density variations in subsurface fault-damage zones in two distinct geologic environments, adjacent to faults in the granitic SSC reservoir and adjacent to faults in arkosic sandstones near the San Andreas fault in central California. These damage zones are similar in terms of width, peak fracture or fault (FF) density, and the rate of FF density decay with distance from the main fault. Seismic images from the SSC reservoir exhibit a large basement master fault associated with 27 seismically resolvable second-order faults. A maximum of 5 to 6 FF/m (1.5 to 1.8 FF/ft) are observed in the 50 to 80 m (164 to 262 ft) wide damage zones associated with second-order faults that are identified in image logs from four wells. Damage zones associated with second-order faults immediately southwest of the San Andreas Fault are also interpreted using image logs from the San Andreas Fault Observatory at Depth (SAFOD) borehole. These damage zones are also 50–80 m wide (164 to 262 ft) with peak FF density of 2.5 to 6 FF/m (0.8 to 1.8 FF/ft). The FF density in damage zones observed in both the study areas is found to decay with distance according to a power law . The fault constant is the FF density at unit distance from the fault, which is about 10–30 FF/m (3.1–9.1 FF/ft) in the SSC reservoir and 6–17 FF/m (1.8–5.2 FF/ft) in the arkose. The decay rate ranges from 0.68 to 1.06 in the SSC reservoir, and from 0.4 to 0.75 in the arkosic section. This quantification of damage-zone attributes can facilitate the incorporation of the geometry and properties of damage zones in reservoir flow simulation models.

Comparative mineral chemistry and textures of SAFOD fault gouge and damage-zone rocks

Creep in the San Andreas Fault Observatory at Depth (SAFOD) drillhole is localized to two foliated gouges, the central deforming zone (CDZ) and southwest deforming zone (SDZ). The gouges consist of porphyroclasts of serpentinite and sedimentary rock dispersed in a foliated matrix of Mg-smectite clays that formed as a result of shearing-enhanced reactions between the serpentinite and quartzofeldspathic rocks. The CDZ takes up most of the creep and exhibits differences in mineralogy and texture from the SDZ that are attributable to its higher shearing rate. In addition, a ∼0.2-m-wide sector of the CDZ at its northeastern margin (NE-CDZ) is identical to the SDZ and may represent a gradient in creep rate across the CDZ. The SDZ and NE-CDZ have lower clay contents and larger porphyroclasts than most of the CDZ, and they contain veinlets and strain fringes of calcite in the gouge matrix not seen elsewhere in the CDZ. Matrix clays in the SDZ and NE-CDZ are saponite and corrensite, whereas the rest of the CDZ lacks corrensite. Saponite is younger than corrensite, reflecting clay crystallization under declining temperatures, and clays in the more actively deforming portions of the CDZ have better equilibrated to the lower-temperature conditions.