Lev Vinnik

Global‐scale analysis of the mantle Pds phases

A global-scale analysis of travel times of the mantle Pds phases (converted from P to S from the discontinuity at a depth d beneath the receiver) is presented. Most of the data are related to P′410′s and P′660′s, recorded by the broadband stations of Geoscope, Incorporated Research Institutions for Seismology (IRIS) and other networks. In order to provide a good global-scale coverage, published measurements at 37 stations are complemented by our measurements at 45 stations. Lateral resolution of our data is many times higher than that of SS and its precursors, previously used for similar purposes, but we have fewer data in the oceans. Lateral variations of travel times of the major Pds phases are strongly correlated with the lateral P and S velocity variations in the uppermost mantle, as revealed by tomographic studies. The scaling coefficient between the teleseismic S and P residuals, as found from the variations of travel times of the major Pds phases, is around 3.7 for North America and around 3.3 for the rest of the world. Almost all the differential travel times t(P′660′s) – t(P′410′s) are in the range of ±1 s, relative to the average value of 24.0 s. If the lateral variations of the velocities within the mantle transition zone (MTZ) can be neglected, the corresponding variations of thickness of the MTZ are in the range of ±10 km, relative to the standard depth of 250 km, and do not show significant correlation with either velocities in the upper mantle above the MTZ or with the variations of thickness of the MTZ, as found in some recent studies of precursors to SS. There is a weak but marginally significant positive correlation between the variations of thickness of the MTZ inferred from our data and the S velocity variations in the MTZ, as given by Su et al. [1994]. This correlation suggests that both variables contain temperature-dependent components. Lateral variability of the MTZ is also manifested by extreme weakness of P′410′s at several locations. This may indicate either anomalous topography on 410 km discontinuity or strong broadening of the otherwise sharp discontinuity. At some stations there are arrivals that can be interpreted as P′520′s. This interpretation implies that the strengh and/or sharpness of 520 km discontinuity, as well as its depth, are laterally variable in a broad range.

Receiver functions for the Tien Shan Analog Broadband Network: Contrasts in the evolution of structures across the Talasso-Fergana Fault

We determined the structure of the crust beneath the central and western Tien Shan by analyzing broadband (1–20 s) analog seismograms of converted P-SV phases generated by earthquakes at teleseismic distances and recorded by 14 seismograph stations. The one-dimensional structures that best explain the waveforms reveal pronounced differences in crustal velocities east and west of the Talasso-Fergana fault. Specifically, the transition zone between the crust and the mantle east of the fault is about 2 times broader than that west of the fault. Also, velocities at depths between 10 and 35 km are about 10% lower in the east, although the depth range of these lower velocities is not well resolved. The Talasso-Fergana fault is an important boundary for other observables; in addition to the structural discontinuities observed at the surface, the areas east of the Talasso-Fergana fault are associated with abnormally low mantle velocities, outcrops of basalt, low Q, and short-wavelength variations in anisotropy. Integrating these observations, we interpret the broad mantle gradient as being due to vertical intrusions of mantle material into the lower crust. Likewise, the low velocities in the midcrust could be thermally induced or due to the introduction of magmatically derived fluids. The gross contrast in structure east and west of the Talasso-Fergana fault could reflect a contrast in the dynamics of mountain building in these two regions. We postulate that while crustal shortening is the dominant mechanism controlling topography west of the Talasso-Fergana fault, vertical uplift caused by a mass deficiency in the upper mantle may contribute significantly to the generation and maintenance of the high elevations east of the fault.

Seismic anisotropy in the D″ layer

We present observations of diffracted SV (SVd) for a path between the Fiji-Tonga islands and the eastern coast of North America at distances greater than 110°. Observed features of S diffracted suggest that coupling between SVd and SHd can be ruled out as a first order effect for this path. Arrivals of SHd are late relative to IASP91 travel-times by about 10 s, and those of SVd are late relative to SHd by 3 s, for most records. The slope of the log(SVd/SHd) spectral ratio is around 3Hz -1 in the range 0.06-0.15 Hz. A transversely isotropic low-velocity layer in the lowermost mantle with a thickness of 200-300 km may account for most of the observed properties of SVd.

Anisotropy in the center of the inner core

We have assembled a collection of PKP data from broadband records of the Geoscope network. This collection is unique because, for the first time, it includes polar paths at epicentral distances between 172° and 177°, for which PKPDF samples the central part of the inner core. After Hilbert transforming the DF branch, the waveforms of PKPAB and PKPDF usually become very similar, and we measure differential travel times with an accuracy of a fraction of a second. The differential (AB-DF) times for equatorial paths are close to those predicted by PREM, whereas for the polar paths, they are larger by 3 to 6 sec. Absolute DF times confirm that the effect is primarily in the inner core. These observations are compatible with a model of cylindrical anisotropy in the inner core with the axis of symmetry aligned with the Earths spin axis and an amplitude of 3.5%. They require that the anisotropy extend to the central part of the inner core, confirming extrapolations made by Creager (1992) and ruling out models where anisotropy is confined to the outer 300 km of the inner core (Tromp, 1993).

Anisotropic structures at the base of the Earth's mantle

The D″ shell at the base of the Earths mantle is thought to be a thermal and compositional boundary layer where vigorous dynamical processes are taking place. An important property of D″ is its seismic anisotropy, expressed as different velocities for horizontally and vertically polarized shear waves that have been diffracted or reflected at the core–mantle boundary,. The nature of this anisotropy has been the subject of debate. Here we present an analysis of various seismic phases, generated in the Kermadec–Fiji–Tonga zone and recorded at stations in North America, which reveal a region at the base of the mantle beneath the southwest Pacific Ocean where horizontally propagating vertically polarized waves are slower (by at least 10 per cent) than horizontally polarized waves. This observed anisotropy is an order of magnitude larger than that previously thought to exist in the lower mantle, and corresponds to lateral variations in horizontally polarized shear-wave velocity which are also of about 10 per cent. We speculate that this anisotropy may be the result of the mixing and shearing of strongly heterogeneous material in the boundary layer.

Sharpness of the mantle discontinuities

Seismic estimates of sharpness of the mantle discontinuities are important for constraining models of composition and temperature in the deep Earth but these data are difficult to obtain. We explore a possibility to determine sharpness of the major mantle discontinuities (those at depths around 400 and 650 km) from the broad-band records of phases converted from P to S underneath the receiver. Our estimates are obtained from a comparative analysis of waveforms of the converted phases and those of the P waves in the teleseismic records of BRV (Kazakhstan), GRF (Germany), NRE0 (Norway) and YKW (Canada). The data indicate that the shape of the pulse converted at the 400-km discontinuity is close to that of the P wave whereas the pulse of the 650-km conversion is substantially broader. We infer from this comparison that the 400-km discontinuity is sharp (width of the transition zone is less than 5–7 km). The 650-km transition is modelled by a linear gradient zone 20–30 km thick.

Shear wave splitting in the mantle Ps phases

Distribution with depth of mantle anisotropy is uncertain. To better constrain this distribution, we analyse records of the mantle Ps phases at the GRF array in Germany. Two components of these phases (SV and T) are detected by stacking teleseismic records of events well distributed in azimuth and distance. The detected T component signals are generated not so much by splitting of SV as by conversion from P to SH at the mantle discontinuities. To reconcile these observations with previously known observations of splitting in SKS, at least three anisotropic layers are required. The data favour a model with NNE oriented fast velocity at the top of the mantle. This is consistent with the results of other seismic studies. Beneath this layer the fast direction of anisotropy changes by around 90°. Our analysis suggests as well that a layer at the base of the mantle transition zone (MTZ) is anisotropic. This anisotropy might arise in the convective boundary layer between the upper and the lower mantle.

Evidence for a stagnant plume in the transition zone

The structure of the mantle transition zone (MTZ) beneath the Pacific ocean is investigated by using Pds waves, converted from P to S from seismic discontinuities at depths d in the receiver region. At stations within the South Pacific superswell, the pulse of P660s is anomalously broad. To explain this effect, we assume that the 660 km phase boundary may present a barrier for a thermochemical plume ascending from the lower mantle. Then the low-velocity material of the plume accumulates just below this boundary, and forms a large-scale lens, which broadens the pulse of P660s. The phase boundary is uplifted by heating, but the related travel time anomaly is cancelled by the increased effective width of the transition.

Shear wave splitting in the records of the German regional seismic network

Estimates of the parameters of shear-wave splitting in the records of SKS and SKKS of the new German Regional Seismograph Network (GRSN) for one-layer model show large azimuthal variations at some stations of the network. It is found that the variations are compatible with the presence of two anisotropic mantle layers in the region of the South German Triangle (SGT); the fast direction in the upper layer is between N-S and NE-SW, close to the previously reported estimates of the fast direction in the sub crustal lithosphere of the SGT. The fast direction in the lower, presumably asthenospheric layer is close to E-W. The estimates of the effective fast direction for one-layer model at the stations of the GRSN confirm the previously noted tendency of the fast direction of mantle anisotropy in Central Europe to change from 50°–70° in the western part of the region to 100°–120° in the eastern part. A correlation of this change with a bending of the present-day direction of maximum horizontal stress in the crust suggests that anisotropy in the asthenosphere of Central Europe can be related to the Alpine orogeny.

DEEP SEISMIC STRUCTURE OF THE KAAPVAAL CRATON

The Kaapvaal craton is remarkable for the intensive Jurassic-Cretaceous kimberlite magmatism. There are indications that the hot-spot heating of the upper mantle in southern Africa continued in the Cenozoic. However, surface wave data, in spite of their sensitivity to temperature, failed to provide any evidence of thermal agitation at depths less than 350 km beneath the Kaapvaal craton (Bloch et al., 1969). We investigated the S velocity structure in the depth range of the mantle transition zone, with the aid of teleseismic P-to-S converted phases. Our analysis suggests that a layer of anomalously low S velocity is present in the 370–470 km depth interval beneath the Kaapvaal craton. The anomalous layer is most likely of thermal origin; its upper boundary may separate the deep mantle root of the craton from the underlying mantle. The anomaly could arise beneath the craton already in the Mesozoic, and melting within this layer could contribute to the kimberlite magmatism.