The seismic structure of the Tonga-Hawaii corridor has been investigated by combining two data sets: Revenaugh and Jordan's reflectivity profile from ScS reverberations, which provides travel times to and impedance contrasts across the major mantle discontinuities, and 1500 new observations of frequency-dependent phase delays for the three-component S, SS, and SSS body waves and the R1 and G1 surface waves, which constrain the velocity structure within this layered framework. The shear waves turning in the upper mantle showed significant splitting of the SH and SV components, indicative of shallow polarization anisotropy. The data set was inverted in conjunction with attenuation and mineralogical constraints to obtain a complete spherically symmetric, radially anisotropic structure. The final model, PA5, is characterized by a highvelocity, anisotropic lid, bounded at 68 km depth by a large (negative) G discontinuity; a lowvelocity, anisotropic layer below G, extending to a small L discontinuity at 166 km; an isotropic, steep-gradient region between 166 km and 415 km; and transition-zone discontinuities at 415, 507, and 651 km. The depth of the radial anisotropy in PA5 is shallower than in most previous studies based on surface waves and higher modes. The average value of radial shear anisotropy in the lid, +3.7%, is consistent with the magnitude expected from the spreading-controlled models of olivine orientation, while anisotropy in the low-velocity zone, which is required by our data set, could be induced either by paleostrains that took place near the ridge crest or by shearing in the asthenosphere as a result of present-day plate motions. On the basis of recent work by Hirth and Kohlstedt, we suggest that the G discontinuity is caused by a rapid increase in the water content of mantle minerals with depth, marking the fossilized lower boundary of the melt separation zone active during crust formation. The high-gradient zone between 200 and 400 km is a characteristic feature of convecting oceanic upper mantle and is probably controlled by a steady decrease in the homologous temperature over this depth interval. The average shear-velocity gradient in the transition zone is lower than in most previous seismic models, in better agreement with the predictions for a pyrolitic composition. [e.g., Forsyth, 1975a; Regan and Anderson, 1984; Farra and Vinnik, 1994], and discontinuity structure [e.g., Revenaugh and Jordan, 1987, 1989; Vidale and Benz, 1992] of the upper mantle. Most recently, global and regional 3-D tomography has begun to provide more detailed maps of the lateral heterogeneity in the Pacific mantle [e.g., Woodhouse and Dziewonski, 1984; Montagner and Tanimoto, 1991; Zhang and Tanimoto, 1993; Su et al., 1994]. While such tomographic studies have yielded useful images of the major geographic features, their data sets and model parameterizations have thus far suffered from inadequate Copyright 1996 by the American Geophysical Union. P•nor number O6IRO 1 gg9. 0148-0227/96/96JB-01882509.00 resolution of verti...

We have formulated a new waveform-analysis procedure to recover phase and amplitude information frorn individual seismograms that makes use of the ability to compute complete seismograms from realistic earth models. The basic tool is the isolation filter, a composite waveform constructed to select data from a desirable portion of the seismogram. When the cross-correlation between this synthetic waveform and an observed seismogram is localized in the time domain by windowing and in the frequency domain by narrow-band filtering, the resulting cross-correlagram can be approximated by a five-parameter Gaussian wavelet. One of these five parameters is the bandwidth of the correlagram, specified by the narrow-band filter; the other four define a set of time-like, frequency-dependent quantities { br, : x = q, p, a, g}, which are functionals of earth structure. bt, is the differential phase delay and at, is the differential group delay of the observed waveform relative to the synthetic, and bt, and 6t, are the corresponding frequency-dependent amplitude parameters. We have developed a procedure for measuring the four generalized seismological data functionals by fitting a Gaussian wavelet to the windowed, filtered cross-correlagram. To relate the GSDFs to earth structure, we apply corrections to the differential times for the effects of windowing and filtering. Solving a linear system of four equations in four unknowns yields a set of differential dispersion parameters { bz, : x = q, p, a, g}. Formulae expressing the perturbations of the GSDFs in terms of the perturbations to the dispersion parameters for the individual component waveforms, including all interference effects, have been derived. Under a set of approximations valid for a large class of isolation filters, these can be simplified to yield easily computed expressions for the FrCchet kernels of the 6~~' s .The calculation of these FrCchet kernels requires no high-frequency approximations, and it can be extended to the investigation three-dimensional earth structure.

The 25 April 1992 magnitude 7.1 Cape Mendocino thrust earthquake demonstrated that the North America-Gorda plate boundary is seismogenic and illustrated hazards that could result from much larger earthquakes forecast for the Cascadia region. The shock occurred just north of the Mendocino Triple Junction and caused strong ground motion and moderate damage in the immediate area. Rupture initiated onshore at a depth of 10.5 kilometers and propagated up-dip and seaward. Slip on steep faults in the Gorda plate generated two magnitude 6.6 aftershocks on 26 April. The main shock did not produce surface rupture on land but caused coastal uplift and a tsunami. The emerging picture of seismicity and faulting at the triple junction suggests that the region is likely to continue experiencing significant seismicity.

No abstract

Cover images: Photographs showing potential beneficiaries of ShakeAlert, the earthquake early warning system for the West Coast of the United States. Clockwise from the upper left: ShakeAlert may provide enough warning (top left) to slow or stop taxiing planes (photograph from Ersin Ergin/ Shutterstock.com), (top middle) for surgeons or other medical professionals to stop delicate procedures (photograph from Andrei_R./Shutterstock. com), and (top right) for water companies to close valves and preserve precious water reserves (photograph from cpaulfell/Shutterstock.com). In addition, ShakeAlert may provide enough warning for (bottom left) people to drop, cover, and hold on (photograph from ChameleonsEye/Shutterstock. com); (bottom middle) first responders to open bay doors in advance of the shaking (photograph from Seattle Fire Department); and (bottom right) train operators to slow or stop trains (photograph from James Kirkikus/Shutterstock.com).