San Andreas Fault Observatory At Depth Site Research Articles

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.

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.

A microstructural study of fault rocks from the SAFOD: Implications for the deformation mechanisms and strength of the creeping segment of the San Andreas Fault

The San Andreas Fault zone in central California accommodates tectonic strain by stable slip and microseismic activity. We study microstructural controls of strength and deformation in the fault using core samples provided by the San Andreas Fault Observatory at Depth (SAFOD) including gouge corresponding to presently active shearing intervals in the main borehole. The methods of study include high-resolution optical and electron microscopy, X-ray fluorescence mapping, X-ray powder diffraction, energy dispersive X-ray spectroscopy, white light interferometry, and image processing.The fault zone at the SAFOD site consists of a strongly deformed and foliated core zone that includes 2–3 m thick active shear zones, surrounded by less deformed rocks. Results suggest deformation and foliation of the core zone outside the active shear zones by alternating cataclasis and pressure solution mechanisms. The active shear zones, considered zones of large-scale shear localization, appear to be associated with an abundance of weak phases including smectite clays, serpentinite alteration products, and amorphous material. We suggest that deformation along the active shear zones is by a granular-type flow mechanism that involves frictional sliding of microlithons along phyllosilicate-rich Riedel shear surfaces as well as stress-driven diffusive mass transfer. The microstructural data may be interpreted to suggest that deformation in the active shear zones is strongly displacement-weakening. The fault creeps because the velocity strengthening weak gouge in the active shear zones is being sheared without strong restrengthening mechanisms such as cementation or fracture sealing. Possible mechanisms for the observed microseismicity in the creeping segment of the SAF include local high fluid pressure build-ups, hard asperity development by fracture-and-seal cycles, and stress build-up due to slip zone undulations.

Low-temperature deformation in calcite veins of SAFOD core samples (San Andreas Fault) — Microstructural analysis and implications for fault rheology

The microstructures of four core samples from the San Andreas Fault Observatory at Depth (SAFOD) were investigated with optical and transmission electron microscopy. These samples, consisting of sandstone, siltstone, and fault gouge from phase III of the drilling campaign (3141–3307 m MD), show a complex composition of quartz, feldspar, clays, and amorphous material. Microstructures indicate intense shearing and dissolution–precipitation as main deformation processes. The samples also contain abundant veins filled with calcite. Within the inspected veins the calcite grains exhibit different degrees of deformation with evidence for twinning and crystal plasticity. Dislocation densities (ranging from ≈ 3 · 10 12 m − 2 to ≈ 3 · 10 13 m − 2 ) and twin line densities (≈ 22 mm − 1 –165 mm − 1 ) are used as paleo-piezometers. The corresponding estimates of differential stresses vary between 33 and 132 MPa, deduced from dislocation density and 92–251 MPa obtained from twin density, possibly reflecting chronologically different maximum stress states and/or grain scale stress perturbations. Mean values of stress estimates are 68 ± 46 MPa and 168 ± 60 MPa, respectively, where estimates from dislocation density may represent a lower bound and those from twin density an upper bound. The stress estimates are also compatible with residual lattice strains determined with microfocus Laue diffraction yielding equivalent stresses of 50–300 MPa in twinned calcite. The lower stress bound agrees with stress estimates from borehole breakout measurements performed in the pilot hole. From these data and assuming hydrostatic pore pressure and a low intermediate principal stress close to the overburden stress, frictional sliding of the San Andreas Fault at the SAFOD site is constrained to friction coefficients between 0.24 and 0.31. These low friction values may be related to the presence of clays, talc, and amorphous phases found in the fault cores and support the hypothesis of a weak San Andreas Fault.

Low strength of deep San Andreas fault gouge from SAFOD core

The San Andreas fault accommodates 28-34 mm yr(-1) of right lateral motion of the Pacific crustal plate northwestward past the North American plate. In California, the fault is composed of two distinct locked segments that have produced great earthquakes in historical times, separated by a 150-km-long creeping zone. The San Andreas Fault Observatory at Depth (SAFOD) is a scientific borehole located northwest of Parkfield, California, near the southern end of the creeping zone. Core was recovered from across the actively deforming San Andreas fault at a vertical depth of 2.7 km (ref. 1). Here we report laboratory strength measurements of these fault core materials at in situ conditions, demonstrating that at this locality and this depth the San Andreas fault is profoundly weak (coefficient of friction, 0.15) owing to the presence of the smectite clay mineral saponite, which is one of the weakest phyllosilicates known. This Mg-rich clay is the low-temperature product of metasomatic reactions between the quartzofeldspathic wall rocks and serpentinite blocks in the fault. These findings provide strong evidence that deformation of the mechanically unusual creeping portions of the San Andreas fault system is controlled by the presence of weak minerals rather than by high fluid pressure or other proposed mechanisms. The combination of these measurements of fault core strength with borehole observations yields a self-consistent picture of the stress state of the San Andreas fault at the SAFOD site, in which the fault is intrinsically weak in an otherwise strong crust.

Evidence of transient increases of fluid pressure in SAFOD phase III cores

[1] The San Andreas Fault Observatory at Depth (SAFOD) in Parkfield, central California, has been drilled through a fault segment that is actively deforming through creep and microearthquakes. Creeping is accommodated in two fault strands, the Southwest and Central Deforming Zones, embedded within a damaged zone of deformed shale and siltstone. During drilling, no pressurized fluids have been encountered, even though the fault zone acts as a permeability barrier to fluid circulation between the North American and Pacific plates. Microstructural analysis of sheared shales associated with calcite and anhydrite-bearing veins found in SAFOD cores collected at 1.5m from the Southwest Deforming Zone, suggests that transient increases of pore fluid pressure have occurred during the fault activity, causing mode I fracturing of the rocks. Such build-ups in fluid pressure may be related to permeability reduction during fault creep and pressure-solution processes, resulting in localized failure of small fault zone patches and providing a potential mechanism for the initiation of some of the microearthquakes registered in the SAFOD site.

Three‐dimensional passive seismic waveform imaging around the SAFOD site, California, using the generalized Radon transform

We apply a three‐dimensional (3D) generalized Radon transform (GRT) to scattered P‐waves from 575 local earthquakes recorded at 68 temporary network stations for passive‐source imaging of (near‐vertical) structures close to the San Andreas Fault Observatory at Depth (SAFOD) site. The GRT image profiles through or close by the SAFOD site reveal near‐vertical reflectors close to the fault zone as well as in the granite to the southwest and the Franciscan mélange to the northeast of the main fault. Although slightly lower in resolution, these structures are generally similar to features in 2D images produced with steep‐dip prestack seismic migration of data from active source seismic reflection and refraction surveys. Our GRT images, however, also reveal several vertical reflectors to the northeast of the SAF that do not appear in the migration images but which are consistent with local geology. These results suggest that in a seismically active area, inverse scattering of earthquake data (for instance with the GRT) can be a viable and, in 3D, economic alternative to an active source survey.

Three-Dimensional Seismic Attenuation Structure around the SAFOD Site, Parkfield, California

Abstract We present models of the three-dimensional (3D) seismic attenuation structure, both Q p and Q s , for a 16 km 2 area centered on the San Andreas Fault Observatory at Depth (SAFOD). The P - and S -wave t * -values used in the inversion were determined from local earthquake data recorded by seismic network and portable array stations within the Parkfield region by inverting arrival spectra for source parameters, t * , and site response. Two techniques for determining the site response, the joint and alternating methods, were compared and it was found that the alternating method significantly underestimated site response variations. The t * -values were inverted to obtain 3D frequency-independent Q p and Q s models using 3D V p and V s models and associated event locations. A shallow low- Q area ( Q p and Q s about 50–75) on the southwest edge of both models is attributed to the low-velocity Cenozoic sedimentary rocks that overlie the Salinian basement rock. A high- Q feature ( Q p and Q s about 250 to 300) abuts this area and is interpreted as the Salinian basement. Adjacent to the San Andreas fault (SAF) trace, on its southwest side, there is a low- Q feature ( Q p and Q s about 50–80) attributed to a wedge of sedimentary rocks; uniformly low Q p - and Q s -values suggest that the wedge is fluid rich. A low- Q basin feature ( Q p and Q s about 50–75) on the northeast side of the SAF is interpreted as a fluid rich zone. Beneath this area there is a high- Q feature ( Q p and Q s about 220–300), which may be caused by crack closure due to increased pressure with depth in the rocks of the Franciscan formation. Given these high Q -values, it seems unlikely that this area acts as a fluid pathway for fluids entering the fault zone from the east into the seismogenic zone of the SAF.

Refining the image of the San Andreas Fault near Parkfield, California using a finite difference travel time computation technique

The Parkfield Area Seismic Observatory (PASO) was a dense, telemetered seismic array that operated for nearly 2 years in a 15 km aperture centered on the San Andreas Fault Observatory at Depth (SAFOD) drill site. The main objective of this deployment was to refine the locations of earthquakes that will serve as potential targets for SAFOD drilling and in the process develop a high (for passive seismological techniques) resolution image of the fault zone structure. A challenging aspect of the analysis of this data set was the known existence of large (20–25%) contrasts in seismic wavespeed across the San Andreas Fault. The resultant distortion of raypaths could challenge the applicability of approximate ray tracing techniques. In order to test the sensitivity of our hypocenter locations and tomographic image to the particular ray tracing and inversion technique employed, we compare an initial determination of locations and structure developed using a coarse grid and an approximate ray tracer [Thurber, C., Roecker, S., Roberts, K., Gold, M., Powell, M.L. , and Rittger, K., 2003. Earthquake locations and three-dimensional fault zone structure along the creeping section of the San Andreas fault near Parkfield, CA: Preparing for SAFOD, Geophys. Res. Lett., 30 3, 10.1029/2002GL016004.] with one derived from a relatively fine grid and an application of a finite difference algorithm [Hole, J.A., and Zelt, B.C., 1995. 3-D finite-difference reflection traveltimes, Geophys. J. Int., 121, 2, 427–434.]. In both cases, we inverted arrival-time data from about 686 local earthquakes and 23 shots simultaneously for earthquake locations and three-dimensional Vp and Vp/Vs structure. Included are data from an active source seismic experiment around the SAFOD site as well as from a vertical array of geophones installed in the 2-km-deep SAFOD pilot hole, drilled in summer 2002. Our results show that the main features of the original analysis are robust: hypocenters are located beneath the trace of the fault in the vicinity of the drill site and the positions of major contrasts in wavespeed are largely the same. At the same time, we determine that shear wave speeds in the upper 2 km of the fault zone are significantly lower than previously estimated, and our estimate of the depth of the main part of the seismogenic zone decreases in places by ∼ 100 m. Tests using “virtual earthquakes” (borehole receiver gathers of picks for surface shots) indicate that our event locations near the borehole currently are accurate to about a few tens of meters horizontally and vertically.

Structure of the San Andreas Fault at SAFOD from P‐wave tomography and fault‐guided wave mapping

Fault‐guided waves reveal a low‐velocity fault segment a few hundred meters southwest of the main strand of the San Andreas Fault (SAF) system. In 2004, the San Andreas Fault Observatory at Depth (SAFOD) Main Hole was drilled 2.5 km underground and 0.7 km west of the SAF surface trace. A 3‐component, 4.5‐Hz seismograph was installed near the bottom of this hole. This instrument recorded fault zone guided (Fg) waves originating from earthquakes along the main SAF ∼2 km north and 3 km south of the SAFOD site. This ∼5 km length corresponds to a distinctive low‐velocity structure imaged in 2003 using microearthquakes recorded on the Pilot Hole array. Because this structure transmits Fg‐waves from the main fault, it is probably connected to the main SAF and is most likely a major, unmapped fault.

The SAFOD Pilot Hole seismic array: Wave propagation effects as a function of sensor depth and source location

In July 2002 we installed a vertical array of seismometers in the San Andreas Fault Observatory at Depth (SAFOD) Pilot Hole (PH). The bottom of this 32 level, 1240 m long array of 3‐ components is located at a depth of ∼2100 m below ground. Surface‐explosion and microearthquake seismograms recorded by the array give valuable insights into the structure of the SAFOD site. The ratios of P‐ and S‐wave velocities (Vp/Vs) along the array suggest the presence of two faults intersecting the PH. The Vp/Vs ratios also depend on source location, with high values to the NW, and lower ones to the SE, correlating with high and low creep rates along the SAF, respectively. Since higher ratios can be produced by increasing fluid saturation, we suggest that this effect might account for both our observations and their correlation with the creep distribution.

Crustal structure across the San Andreas Fault at the SAFOD site from potential field and geologic studies

We present newly compiled magnetic, gravity, and geologic datasets from the Parkfield region around the San Andreas Fault Observatory at Depth (SAFOD) pilot hole in order to help define the structure and geophysical setting of the San Andreas Fault (SAF). A 2‐D cross section of the SAF zone at SAFOD, based on new, tightly spaced magnetic and gravity observations and surface geology, shows that as drilling proceeds NE toward the SAF, it is likely that at least 2 fault bounded magnetic slivers, possibly consisting of magnetic granitic rock, serpentinite, or unusually magnetic sandstone, will be encountered. The upper 2 km of the model is constrained by an order of magnitude increase in magnetic susceptibility at 1400 m depth observed in pilot hole measurements. NE of the SAF, a flat lying, tabular body of serpentinite at 2 km depth separates two masses of Franciscan rock and truncates against the SAF.

High-Resolution Seismic Velocities and Shallow Structure of the San Andreas Fault Zone at Middle Mountain, Parkfield, California

A 5-km-long, high-resolution seismic imaging survey across the San Andreas fault (SAF) zone and the proposed San Andreas Fault Observatory at Depth (SAFOD) drill site near Parkfield, California, shows that velocities vary both laterally and vertically. Velocities range from 4.0 km/sec) probably correspond to granitic rock of the Salinian block, which is exposed a few kilometers southwest of the SAF. The depth to the top of probable granitic rock varies laterally along the seismic profile but is about 600 m below the surface at the proposed SAFOD site. We observe a prominent, lateral low-velocity zone (LVZ) beneath and southwest of the surface trace of the SAF. The LVZ is about 1.5 km wide at 300-m depth but tapers to about 600 m wide at 750-m depth. At the maximum depth of the velocity model (750 m), the LVZ is centered approximately 400 m southwest of the surface trace of the SAF. Similar velocities and velocity gradients are observed at comparable depths on both sides of the LVZ, suggesting that the LVZ is anomalous relative to rocks on either side of it. Velocities within the LVZ are lower than those of San Andreas fault gouge, and the LVZ is also anomalous with respect to gravity, magnetic, and resistivity measurements. Because of its proximity to the surface trace of the SAF, it is tempting to suggest that the LVZ represents a zone of fractured crystalline rocks at depth. However, the LVZ instead probably represents a tectonic sliver of sedimentary rock that now rests adjacent to or encompasses the SAF. Such a sliver of sedimentary rock implies fault strands on both sides and possibly within the sliver, suggesting a zone of fault strands at least 1.5 km wide at a depth of 300 m, tapering to about 600 m wide at 750-m depth. Fluids within the sedimentary sliver are probably responsible for observed low-resistivity values.