Rayleigh wave phase velocities, small‐scale convection, and azimuthal anisotropy beneath southern California

Abstract

[1] We use Rayleigh waves to invert for shear velocities in the upper mantle beneath southern California. A one-dimensional shear velocity model reveals a pronounced lowvelocity zone (LVZ) from 90 to 210 km. The pattern of velocity anomalies indicates that there is active small-scale convection in the asthenosphere and that the dominant form of convection is three-dimensional (3-D) lithospheric drips and asthenospheric upwellings, rather than 2-D sheets or slabs. Several of the features that we observe have been previously detected by body wave tomography: these anomalies have been interpreted as delaminated lithosphere and consequent upwelling of the asthenosphere beneath the eastern edge of the southern Sierra Nevada and Walker Lane region; sinking lithosphere beneath the southern Central Valley; upwelling beneath the Salton Trough; and downwelling beneath the Transverse Ranges. Our new observations provide better constraints on the lateral and vertical extent of these anomalies. In addition, we detect two previously undetected features: a high-velocity anomaly beneath the northern Peninsular Range and a low-velocity anomaly beneath the northeastern Mojave block. We also estimate the azimuthal anisotropy from Rayleigh wave data. The strength is $1.7% at periods shorter than 100 s and decreases to below 1% at longer periods. The fast direction is nearly E-W. The anisotropic layer is more than 300 km thick. The E-W fast directions in the lithosphere and sublithosphere mantle may be caused by distinct deformation mechanisms: pure shear in the lithosphere due to N-S tectonic shortening and simple shear in sublithosphere mantle due to mantle flow.

A New Perspective on the Geometry of the San Andreas Fault in Southern California and Its Relationship to Lithospheric Structure

“…C1a). This postulated orientation of olivine does not, however, explain the 1%-2% anisotropy modeled from Rayleigh-wave tomography (Yang and Forsyth, 2006). Significantly, however, V P =V S for this orientation of olivine would be relatively high (∼2:0; Fig.…”

“…This effect is not well resolved spatially, but could originate chiefly in the Transverse Ranges, which are central to the Pn study area of Hearn (1984). If the fast direction of mantle olivine grains (a-axes) is oriented downward in steeply dipping shear zones in the mantle beneath the Transverse Ranges due to downwelling PAC, then: (1) the HVB would be apparent both from steeply inclined P and S imaging rays and from Rayleigh waves (Sv) (see images of Humphreys and Clayton, 1990;Yang and Forsyth, 2006;and Schmandt and Humphreys, 2010) ; (2) Pn anisotropy would be developed as observed (Hearn, 1984); and (3) SKS splitting would not change from outside to inside the footprint of the HVB, as is observed (e.g., Polet and Kanamori, 2002), because both vibration directions for steeply emergent S waves would have the same speed in the model shown (Fig. C1a).…”

“…11). A shear-wave HVB image has also been produced from Rayleigh-wave tomography (Yang and Forsyth, 2006), although the model indicates a depth extent for the HVB of only 150 km. Schmandt and Humphreys (2010) have imaged this body with both teleseismic P and S waves using approximate 3D sensitivity kernels in three frequency bands.…”

Anisotropy from SKS splitting across the Pacific-North America plate boundary offshore southern California

“…The magnitude of peak-to-peak anisotropy is 1.7 per cent at periods below 40 s, decreasing to 1 per cent for longer periods, predicting a shear wave splitting measurement of only 0.25-0.4 s in the Anisotropy offshore southern California 245 lithosphere (assuming a 90 km lithospheric depth). This implies that the remaining 1.1 s observed in splitting measurements occurs at sublithospheric depths as deep as ∼300 km (Yang & Forsyth 2006). In the oceanic mantle, we can expect lithospheric anisotropy to be stronger (up to ∼3.5 per cent peak-to-peak) than in continental lithosphere (Weeraratne et al 2007).…”

“…Surface wave azimuthal anisotropy averaged over the area covered by the permanent Southern California Seismic Network varies uniformly in this unusually complex region, with a uniform EW fast direction for periods between 20-40 s that changes slightly to NW-SE at longer periods up to 110 s (Yang & Forsyth 2006). The magnitude of peak-to-peak anisotropy is 1.7 per cent at periods below 40 s, decreasing to 1 per cent for longer periods, predicting a shear wave splitting measurement of only 0.25-0.4 s in the Anisotropy offshore southern California 245 lithosphere (assuming a 90 km lithospheric depth).…”