B. L. N. Kennett

Global azimuthal seismic anisotropy and the unique plate-motion deformation of Australia

Differences in the thickness of the high-velocity lid underlying continents as imaged by seismic tomography, have fuelled a long debate on the origin of the ‘roots’ of continents. Some of these differences may be reconciled by observations of radial anisotropy between 250 and 300 km depth, with horizontally polarized shear waves travelling faster than vertically polarized ones. This azimuthally averaged anisotropy could arise from present-day deformation at the base of the plate, as has been found for shallower depths beneath ocean basins. Such deformation would also produce significant azimuthal variation, owing to the preferred alignment of highly anisotropic minerals. Here we report global observations of surface-wave azimuthal anisotropy, which indicate that only the continental portion of the Australian plate displays significant azimuthal anisotropy and strong correlation with present-day plate motion in the depth range 175–300 km. Beneath other continents, azimuthal anisotropy is only weakly correlated with plate motion and its depth location is similar to that found beneath oceans. We infer that the fast-moving Australian plate contains the only continental region with a sufficiently large deformation at its base to be transformed into azimuthal anisotropy. Simple shear leading to anisotropy with a plunging axis of symmetry may explain the smaller azimuthal anisotropy beneath other continents.

Joint seismic tomography for bulk sound and shear wave speed in the Earth's mantle

High-quality P and S travel times are now available from careful reprocessing of data reported to international agencies. A restricted data set has been extracted for which comparable ray coverage is achieved for P and S, and used for a joint inversion to produce a three-dimensional model for shear and bulk sound velocities represented in terms of 2° × 2° cells and 18 layers in depth through the mantle. About 106 times for each of P and S are combined to produce 312,549 summary rays for each wave type. Linearizing about the ak135 reference model, 583,200 coupled tomographic equations are solved using an iterative partitioned scheme. Clear high-resolution images are obtained for both bulk-sound speed and shear wavespeed. The bulk and shear moduli have differing sensitivity to temperature and mineral composition, and so the images of the two velocity distributions help to constrain the nature of the processes which produce the variations. Different heterogeneity regimes can be recognised in the upper mantle, the transition zone, most of the lower mantle, and the lowermost mantle. In the upper mantle, many features can be explained by thermal effects; but in some orogenic zones (e.g. western North America), the opposite sense of the bulk-sound and shear wave speed variation requires compositional effects or volatiles to outweigh any thermal effects. In the lower mantle, pronounced narrow structures which may represent remnant subduction are most marked in shear. The level of large-scale variations in bulk sound speed compared to shear diminishes with depth in the lower mantle reaching a minimum near 2000 km. Below this depth, the variability of both wave speeds increases. Near the core-mantle boundary the variations of the two wave speeds show little concordance, suggesting the presence of widespread chemical heterogeneity.

The Australian continental upper mantle structure and deformation inferred from surface waves

We present a new three-dimensional model for the SV wave hetero-geneities and azimuthal anisotropy in the upper mantle of the Australasian region. The model is constrained by the waveforms of 2194 Rayleigh waves seismograms with a dense ray coverage that ensure a lateral resolution of the order of few hundred of kilometers. The use of higher modes allows the resolution of the structure down to depths of at least 400 km. In the upper 200 km of the model, seismic velocities are lower on the eastern Phanerozoic margin of the continent compared to the Precambrian central and western cratons, in agreement with previous results for Australia. The boundary between Phanerozoic and Precambrian Australia is not clear, especially in the south, where a broad positive seismic anomaly underlays the Lachlan Fold Belt. The high-velocity lid beneath the continent shows significant variations in thickness. Locally, it may extend down to a depth of 300 km in the mantle, but for most of Precambrian Australia the lithospheric thickness oscillates around 200 km, while it is thinner on the eastern Phanerozoic margin. We found significant SV wave azimuthal anisotropy in the upper 250 km of the mantle, with a drastic change in the organization of anisotropy between the upper 150 km of the model and the deeper part, as revealed in a preliminary inversion. In the upper 150 km of the mantle, azimuthal anisotropy appears more likely to be related to past deformation frozen in the lithosphere, and in central Australia we found clear evidence that deformation is preserved since the Alice Springs orogeny. Below 150 km, a smoother pattern of anisotropy is observed, more likely to be related to present-day deformation due to the northward motion of the Australian plate. Our current data set allows constraint of the anisotropic directions at different depths with an unprecedented lateral resolution. The observation of significant changes of anisotropic directions with depth in the Australian continental mantle suggests that care should be taken in the interpretation of anisotropy from SKS observations.

Genetic algorithm inversion for receiver functions with application to crust and uppermost mantle structure beneath eastern Australia

Genetic algorithm (GA) inversion, a nonlin- ear global optimization technique, has been applied to determine crustal and uppermost mantle velocity struc- ture from teleseismic receiver functions. With a new vicinity of the receiver. The influence of the source can be largely eliminated by source equalization in which the radial component of motion is deconvolved with the vertical component to generate a receiver function (Langston,1979) which isolates conversions from P to

Anisotropy in the Australasian upper mantle from Love and Rayleigh waveform inversion

generated at boundaries beneath the recording site. The receiver function waveform can be inverted in the

Multi-component autoregressive techniques for the analysis of seismograms

Abstract Records of both Rayleigh and Love waves have been analyzed to determine the pattern of anisotropy in the Australasian region. The approach is based on a two-stage inversion. Starting from a smooth PREM model with transverse isotropy about a vertical symmetry axis, the first step is an inversion of the waveforms of surface waves to produce path specific one-dimensional (1-D) upper mantle models. Under the assumption that the 1-D models represent averages along the paths, the results from 1584 Love and Rayleigh wave seismograms are combined in a tomographic inversion to provide a representation of three-dimensional structure for wavespeed heterogeneities and anisotropy. Polarization anisotropy with SH faster than SV is retrieved in the upper 200–250 km of the mantle for most of Precambrian Australia. In this depth interval, significant lateral variations in the level of polarization anisotropy are present. Locally, the anisotropy can be large, reaching an extreme value of 9% that is difficult to reconcile with current mineralogical models. However, the discrepancy may be explained in part by the presence of strong lateral heterogeneities along the path, or by effects introduced by the simplifying assumption of transverse isotropy for each path. The consistency between the location of polarization and azimuthal anisotropy in depth suggests that both observations share a common origin. The observation of polarization anisotropy down to at least 200 km supports a two-layered anisotropic model as constrained by the azimuthal anisotropy of SV waves. In the upper layer, 150 km thick, anisotropy would be related to past deformation frozen in the lithosphere while in the lower layer, anisotropy would reflect present day deformation due to plate motion.

Joint bulk-sound and shear tomography for Western Pacific subduction zones

Abstract Autoregressive methods provide a very useful means of characterising a seismic record; calculating the power spectra of a seismic record and determining the onset time of different classes of arrivals. The representation of a time series with an autoregressive (AR) process of low order can be applied to both multi-component and single-component traces of broadband and short period seismograms. In three-component analysis the AR coefficients are represented as second-order tensors and include potential cross-coupling between the different components of the seismogram. Power spectrum estimation using autoregressive methods is demonstrated to be effective for both signal and noise and has the advantage over FFT methods in that it is smoother and less susceptible to statistical noise. The order of the AR process required to resolve the detail of the spectra is higher for a complex signal than for the preceding noise. This variation in the weighting of the AR coefficients provides an effective way to characterise data in a similar way to Spectragrams and Vespagrams and can be achieved with as few as five AR coefficients. For three-component analysis a display of the nine AR coefficients can be readily organised with three AR-grams for each of the original data components. The various elements of the AR tensor coefficients reflect different changes in the seismogram. The presence of secondary phases is often clearer on a cross-correlation AR-gram (NE or EN) than on the autocorrelation AR-gram (NN or NE). The variations in the weighting of the AR coefficients can be exploited in two different styles of approach to onset time estimation (phase picking). In the first method, two different AR representations are constructed for different portions of the record and the onset time is estimated from the point of transition. In the second method, a single AR representation is constructed and the onset time estimation is based on the growth of a component which is not represented by the AR process. Both methods can be applied to both single and three-component data. For large impulsive P phases, both methods picked the onset time within two samples of the manually estimated onset time. For S phases, where the energy is present on all three components, three-component AR onset time estimation is preferred to that using a single component. The approach is very robust with the three-component method picking the onset time of a very small S phase on a broad-band record to within 0.5 s of the best manual estimate.

The crustal thickness of Australia

Abstract Detailed regional body wave tomographic inversion of the Western Pacific region has been performed using P and S travel times from common sources and receivers, with a joint inversion in terms of bulk-sound and shear wave-speed variations in the mantle. This technique allows the separation of the influence of bulk and shear moduli, and hence a more direct comparison with mineral physics information. The study region is parameterized with cells of side 0.5° to 2° and 19 layers to a depth of 1500 km, while the rest of the mantle was parameterized with 5×5° cells with 16 layers between the surface and the core–mantle boundary. A simultaneous inversion is made for regional and global structures to minimize the influence of surrounding structures on the regional image. A nested iterative inversion scheme is employed with local linearization and three-dimensional ray tracing through the successive model updates. The results of the regional tomographic inversion reveal the penetration of a subducted slab below the 660 km discontinuity at the Kurile–Kamchatka trench, while flattening of slabs above this depth is observed in the Japan and Izu–Bonin subduction zones on both the bulk-sound and shear wave-speed images. The penetration of a subducted slab down to a depth of at least 1200 km is seen below the southern part of the Bonin trench, Mariana, Philippine, and Java subduction zones. Fast shear wave-speed perturbations associated with the subducted slabs, down to the 410 km transition zone, are larger than the comparable bulk-sound perturbations for all these subduction zones except the Philippines. The bulk-sound signature for the subducted slab is more pronounced than for shear in the Philippines, Talaud, New Guinea, Solomon, and Tonga subduction zones, where penetration of the slab into the middle mantle is observed. Variation in the amplitude ratio between bulk-sound and shear wave-speed anomalies correlates well with the subduction parameters of the descending slab. Slabs younger than 90 Ma at the trench show bulk-sound dominance in the upper mantle, while older slabs have a stronger shear wave-speed signature. Spreading of the fast shear wave-speed zone between 800 and 1000 km is observed in the areas of deep subducted slab penetration, but has no comparable expression in the bulk-sound images. This high-velocity feature may reflect physical or chemical disequilibria introduced to the lower mantle by subducted slabs.

Subduction zone guided waves and the heterogeneity structure of the subducted plate: Intensity anomalies in northern Japan

We investigate the crustal structure of the Australian continent using the temporary broadband stations of the Skippy and Kimba projects and permanent broadband stations. We isolate near-receiver information, in the form of crustal P-to-S conversions, using the receiver function technique. Stacked receiver functions are inverted for S velocity structure using a Genetic Algorithm approach to Receiver Function Inversion (GARFI). From the resulting velocity models we are able to determine the Moho depth and to classify the width of the crust-mantle transition for 65 broadband stations. Using these results and 51 independent estimates of crustal thickness from refraction and reflection profiles, we present a new, improved, map of Moho depth for the Australian continent. The thinnest crust (25 km) occurs in the Archean Yilgarn Craton in Western Australia; the thickest crust (61 km) occurs in Proterozoic central Australia. The average crustal thickness is 38.8 km (standard deviation 6.2 km). Interpolation error estimates are made using kriging and fall into the range 2.5–7.0 km. We find generally good agreement between the depth to the seismologically defined Moho and xenolith-derived estimates of crustal thickness beneath northeastern Australia. However, beneath the Lachlan Fold Belt the estimates are not in agreement, and it is possible that the two techniques are mapping differing parts of a broad Moho transition zone. The Archean cratons of Western Australia appear to have remained largely stable since cratonization, reflected in only slight variation of Moho depth. The largely Proterozoic center of Australia shows relatively thicker crust overall as well as major Moho offsets. We see evidence of the margin of the contact between the Precambrian craton and the Tasman Orogen, referred to as the Tasman Line.

A low seismic wavespeed anomaly beneath northwestern India: a seismic signature of the Deccan plume?

[1] The subducting Pacific plate acts an efficient waveguide for high-frequency signals and often produces anomalously large intensity on the eastern seaboard of northern Japan during deep earthquakes. The waveform records in the region of high intensity show a low-frequency (f 2 Hz) later arrivals with a long coda. This behavior is not explained by a simple subduction zone model comprising a high-velocity plate with low attenuation. From the analysis of observed broadband waveforms and numerical simulation of seismic wave propagation in the Pacific subduction zone we demonstrate that the high-frequency guided waves traveling in the subducting plate arise from the scattering of seismic waves by heterogeneity in plate structure. Our preferred model of the heterogeneity has elongated scatterers parallel to the plate margin described by a von Karmann function with a downdip correlation length of about 10 km and much shorter correlation length of about 0.5 km in thickness. The standard deviation of wave speed fluctuations from the averaged background model is about 2%. This new heterogeneous plate model generates significant scattering of seismic waves with wavelengths shorter than correlation distance in thickness, but low-frequency waves, with long wavelengths, can easy tunnel through such lamina structure. The result is frequency-selective propagation characteristics with a faster low-frequency phase followed by large and high-frequency signals with very long coda. A low-wave speed channel effect from the former oceanic crust at the top of the subducting slab is not necessary to explain the observed dispersed signals and the very long high-frequency coda. Three-dimensional simulations, using the Earth simulator supercomputer for modeling of high-frequency seismic wave propagation in the Pacific subduction zone including plate heterogeneity, clearly demonstrate the scattering waveguide effects for high-frequency seismic waves traveling in the plate. The region of large intensity for the heterogeneous model migrates away from the hypocenter into northern Japan with an elongated zone along the Pacific coast, almost comparable to the observations from deep events in the Pacific plate.