Peter C. Leary

Source parameters of small earthquakes recorded at 2.5 km depth, Cajon Pass, southern California: Implications for earthquake scaling

A 2.5 km deep triaxial seismometer at Cajon Pass in southern California has recorded several hundred earthquakes <ML4.0 occurring within the San Andreas fault system. At 2.5 km seismic background noise is below amplifier sensitivity and the 2–250 Hz spectral range of recorded seismic motion is wider and higher than that of most natural event catalogs. Compared with downhole recorded motion, seismic amplitudes at the surface are amplified below 10 Hz and severely attenuated above 30 Hz. We estimate that QS is at least 1000 for wave motion at 2.5 km and below and QP is over 2000. The range of source dimensions in the downhole recorded catalog is ∼10 m to ∼70 m (ML∼−2.0, Mo∼108 Nm to ML∼2.7, Mo∼1013 Nm). The plot of log(source-radius) vs log(moment) has a straight line trend compatible with earthquake scaling at constant stress drop; inferred stress drops are scattered between 1 and 500 bars. There is no evidence in the catalog for the proposed minimum source dimension at ∼100 m. When the Cajon Pass borehole catalog, containing some of the smallest recorded natural earthquakes, is combined with 800 larger events from previous studies, the moment-radius trend suggests that natural earthquakes are self-similar over a magnitude range M∼−2 to ∼8. We suggest that inferences of minimum source dimension are more likely due to bias in bandlimited individual catalogs than to properties of the seismic crust.

Seismic trapped modes in the Oroville and San Andreas Fault zones

Three-component borehole seismic profiling of the recently active Oroville, California, normal fault and microearthquake event recording with a near-fault three-component borehole seismometer on the San Andreas fault at Parkfield, California, have shown numerous instances of pronounced dispersive wave trains following the shear wave arrivals. These wave trains are interpreted as fault zone-trapped seismic modes. Parkfield earthquakes exciting trapped modes have been located as deep as 10 kilometers, as shallow as 4 kilometers, and extend 12 kilometers along the fault on either side of the recording station. Selected Oroville and Parkfield wave forms are modeled as the fundamental and first higher trapped SH modes of a narrow low-velocity layer at the fault. Modeling results suggest that the Oroville fault zone is 18 meters wide at depth and has a shear wave velocity of 1 kilometer per second, whereas at Parkfield, the fault gouge is 100 to 150 meters wide and has a shear wave velocity of 1.1 to 1.8 kilometers per second. These low-velocity layers are probably the rupture planes on which earthquakes occur.

Joint inversion of fault zone head waves and direct P arrivals for crustal structure near major faults

A newly recognized class of seismic phases, namely fault zone head waves arising from refraction at a transfault velocity contrast, provides additional constraints to direct P waves in inversions of near-fault arrival time data for earthquake location and velocity structure. Incorporation of fault zone head waves in the inversion process increases the available data and broadens the spatial coverage of ray paths between sources and receivers. Equally important, the explicit recognition of fault zone head waves eliminates a source of error resulting from misidentifying head wave first arrivals as direct P waves. A simple algorithm for the joint inversion of fault zone head waves and direct P phases is illustrated with observed and synthetic data. Application to a small data set recorded by the borehole seismic network at Parkfield, California, provides separate seismic velocity depth profiles for the crustal blocks on the two sides of the San Andreas fault. For depths less than 3 km the obtained velocity contrast is 10–20%. For greater depths it decreases to 3–7%. Numerical tests with synthetic data show that for sources with known locations and for well located local earthquakes (randomized mislocation errors of up to 600 m) proper use of fault zone head waves can significantly improve the accuracy of the fitted velocity structure. The results of this work motivate the inclusion of fault zone head waves in state-of-the-art joint structure-hypocenter inversions using high quality near-fault data.

Frequency dependent crustal scattering and absorption at 5–160 Hz from coda decay observed at 2.5 km Depth

A triaxial 10 Hz seismometer at 2.5 km depth in the Cajon Pass borehole near the San Andreas fault in southern California records shear-wave coda motion from small local events for over 20 seconds duration. The passband of recorded seismic motion is 5 Hz to 200 Hz. To measure the rate of coda energy decay as a function of frequency, we filter the vector velocity seismograms of seven events into five octave-wide frequency bands (mean frequencies ≈ 7, 14, 28, 56 and 112 Hz) and square the filtered seismograms. The observed energy decay in each passband is well approximated by first and second order scattering plus intrinsic attenuation as formulated by Zeng at al. (JGR 1991). The fits determine two energy decay parameters expressed as inverse lengths, sscat for scattering and sintr for absorption. Because the source-receiver distance is less than the thickness of the upper crust and the receiver is at depth, the direct body wave is uncomplicated by refracted energy and/or surface waves and allows accurate recording of coda energy relative to source pulse energy. The coda/source energy ratio directly defines the scattering attenuation parameter sscat and voids the need for multiple offset observations. The observed frequency behavior of scattering is sscat≈f1.3; the intrinsic attenuation parameter has frequency dependence sintr≈f0.43. Normalized to energy lost per unit cycle (quality factor Q), intrinsic attenuation decreases with frequency, Qintr≈f0.57, while scattering attenuation increases with frequency, Qscat≈1/f0.3 with Qscat > 10Qintr at all observed frequencies.

Seismic reflection constraints on the structure of the crust beneath the San Bernardino Mountains, Transverse Ranges, southern California

A 30 km-long N-S seismic reflection line was shot by California Consortium for Crust Studies (CALCRUST) across the southern Mojave Desert and onto the northern flank of the San Bernardino Mountains in southern California. On the northern end of the seismic section, the reflectivity increases markedly in the midcrust at a depth corresponding to a two-way travel time of 4 to 5 s (12–15 km), suggesting a transition between nonreflecting brittle upper crust and reflecting ductile lower crust. The high reflectivity disappears at about 8 s (24 km) and may be correlated with a change in seismic velocity in the lower crust from 6.3 km/s to 6.8 km/s. A band of reflectivity between 9.5 and 10 s (27–30 km) is believed to represent the Moho. The midcrustal relectivity transition and Moho both deflect downward toward the San Bernardino Mountains uplift over the entire length of the profile. The deflection of the midcrustal transition (12°) appears greater than that of the Moho (6°), resulting in a thinning of the lower crust to the south beneath the uplift. In addition, the midcrustal transition coincides with the base of the seismogenic zone (brittle-ductile transition?) which is also dipping southward beneath the San Bernardino Mountains, while the Moho deflection is consistent with elastic flexure resulting from edge loading by the San Bernardino Mountains which have been thrust over the Mojave block. It is suggested that the thinning of the lower crust beneath the San Bernardino Mountains is a result of north directed ductile flow in response to loading by the over thickened upper crust. Since a portion of the load is transmitted through the lower crust to the Moho, the time constant for flow equilibrium must be of the order of or greater than that for the time of uplift (≥2 m.y.).

Fractures and Physical Heterogeneity in Crustal Rock

Despite decades of effort to effectively model fracture distributions in crustal rock, in situ flow and failure structures continue to defeat model predictions. The absence of spatially accurate predictive models seriously degrades hydrocarbon recovery (Oil and Gas Industry Task Force, 2001; Francis and Pennington, 2001), protection and remediation of groundwater aquifers (National Research Council, 1990; Bear et al,1993), and risk assessment of hazardous waste repositories (Haszeldine and Smythe, 1996). Geotechnical surveys of mines (Inazaki et al, 1999; Kneib et al, 2000) and civil-hazard programs for major crustal earthquakes (Geller, 1997) likewise continue to face large uncertainties and costly risks from the potential for rock failure along unknown fracture systems.

Fractal fracture scattering origin of S‐wave coda: Spectral evidence from recordings at 2.5 km

Local earthquake seismograms recorded at a depth of 2.5 Km in the Cajon Pass borehole near the San Andreas fault, southern California, yield body-wave and coda-wave amplitude spectra at frequencies between 10 and 200 Hz without interference from either near-surface attenuation or surface waves. The coda-wave spectra resemble the shear-wave source spectra except that above the corner frequencies fo ≈ 20–30 Hz coda spectra decay by power-law exponent n ≈ −2.3±0.1 while the source shear-wave spectra decay by cubic power-law (mean power-law exponent n ≈ −3.1±0.1). Assuming a cubic source power-law spectral decay, the high frequency power-law enrichment of coda amplitudes relative to source amplitudes implies a power-law distribution of scatterers that increases with frequency as ≈ f0.7±0.1. The distribution of acoustic reflectivity deduced from the Cajon Pass well log has a power-law density ≈ ν0.6 at the relevant spatial frequencies ν. The agreement between the temporal and spatial frequency power-law exponents may be explained by first order scattering in fractal fracture-heterogeneous material.

A Pilot Vertical Seismic Profiling Experiment in the Cajon Pass Deep Scientific Drillhole

The Cajon Pass Deep Scientific Drillhole (see Henyey et al., this volume) is expected to provide one of the first opportunities for a comprehensive study of the crystalline crust using the vertical seismic profiling (VSP) technique. The hole is being drilled to a depth of 5 km through a diverse suite of basement rocks adjacent to the San Andreas fault in southern California. Major low-angle structures of regional importance related to the fault and nearby Transverse Ranges uplift (Fig. 1) are inferred to exist in the Cajon Pass area (e.g. Weldon 1986). Numerous fracture zones (faults?) and petrologic discontinuities have already been intersected in the first 2 km of hole.

Fluid Flow and Heat Transport Computation for Power-Law Scaling Poroperm Media

In applying Darcy’s law to fluid flow in geologic formations, it is generally assumed that flow variations average to an effectively constant formation flow property. This assumption is, however, fundamentally inaccurate for the ambient crust. Well-log, well-core, and well-flow empirics show that crustal flow spatial variations are systematically correlated from mm to km. Translating crustal flow spatial correlation empirics into numerical form for fluid flow/transport simulation requires computations to be performed on a single global mesh that supports long-range spatial correlation flow structures. Global meshes populated by spatially correlated stochastic poroperm distributions can be processed by 3D finite-element solvers. We model wellbore-logged Dm-scale temperature data due to heat advective flow into a well transecting small faults in a Hm-scale sandstone volume. Wellbore-centric thermal transport is described by Peclet number ≡ ( = wellbore radius, = fluid velocity at , = mean crustal porosity, and = rock-water thermal diffusivity). The modelling schema is (i) 3D global mesh for spatially correlated stochastic poropermeability; (ii) ambient percolation flow calibrated by well-core porosity-controlled permeability; (iii) advection via fault-like structures calibrated by well-log neutron porosity; (iv) flow ~ 0.5 in ambient crust and ~ 5 for fault-borne advection.

Developing stress-monitoring sites

Theory, laboratory and field evidence reported elsewhere demonstrate that shear-wave splitting monitors the low-level pre-fracturing deformation of the crustal rock which is driven by the response of fluids in cracked rock.