Splitting Delay Research Articles

Seismic anisotropy under central Alaska from SKS splitting observations

Seismic anisotropy under central Alaska is studied using shear wave splitting observations of SKS waves recorded on the Broadband Experiment Across the Alaska Range (BEAAR), 1999–2001. Splitting results can be divided into two distinct regions separated by the 70 km contour of the subducting Pacific plate. Waves that travel through the thicker mantle wedge show fast directions that are parallel to the strike of the slab. These slab‐parallel directions appear to indicate along‐strike flow in the mantle wedge, and splitting delay times increase with path length in the mantle wedge, suggesting anisotropy of 7.9% ± 0.9%. The region of along‐strike flow corresponds to high seismic attenuation and hence high temperatures. Along‐strike flow here may be driven by secular shallowing of the slab driven by subduction of buoyant Yakutat terrane crust or by toroidal flow around the east end of the Aleutian slab. Waves traveling southeast of the 70 km contour sample the Pacific plate and the nose of the mantle wedge; they show fast directions that parallel the direction of plate motion. These fast directions are most likely due to flow under the subducting Pacific plate and/or anisotropy within the subducting Pacific lithosphere. The high splitting delay times (0.8–1.7 s) associated with these convergence‐parallel directions cannot be produced in the mantle wedge, which is 10–30 km thick here. Thus, anisotropy shows a sharp 90° change in fabric associated with the onset of high‐temperature wedge flow.

Upper mantle anisotropy in SE and Central Brazil from SKS splitting: Evidence of asthenospheric flow around a cratonic keel

We present results of upper mantle anisotropy derived from measurements of core refracted shear wave splitting (mainly SKS phases) recorded at 48 stations covering the major tectonic provinces in Central and SE Brazil, such as the Tocantins Province (Paraguay–Araguaia and Brasília belts between the Amazon and São Francisco cratons), the Paraná intracratonic basin, the southern part of the São Francisco craton, and the Mantiqueira province (with the coastal Ribeira belt). Although the fast polarization directions vary across the region, consistent orientations are observed over hundreds of kilometers. The fast polarization directions tend to be close to the absolute plate motion given by the hot-spot reference model HS3-NUVEL1A. However, correlations with geological structures are also observed in the southern Brasilia belt and in the Ribeira belt, respectively located SW and S of the São Francisco craton. On the other hand, in the northern Tocantins province, the fast shear-wave direction (∼ N60°S) is oblique to the SW–NE trend of the geological units and faults, and no anisotropy contribution from lithospheric sources can be clearly identified. Overall, the fast polarization directions show a fan-shaped pattern strongly suggesting asthenosperic flow around a thick and stiff keel in the southern part of the São Francisco craton, consistent with the high-velocity anomaly revealed by recent surface-wave tomography. The observed NW–SE directions in the southern part of the Brasília belt may also be interpreted as resulting from asthenospheric flow channeled between the São Francisco craton and a cratonic block beneath the Paraná basin. The largest splitting delays observed in the southern Brasília belt and in the Ribeira belt (up to 2.4 s) suggest contributions from both lithospheric and asthenospheric sources in those two areas. Our preferred model for the anisotropy causing the observed pattern of SKS splitting consists of flow-induced deformation in the asthenosphere caused by the absolute plate motion (roughly towards W or WSW) with flow channeled around a thick cratonic keel. In the southern Brasília and in the Ribeira fold belts, the large delays and the parallelism with geological structures indicate additional anisotropy contribution from lithospheric sources.

Depleted swell root beneath the Cape Verde Islands

The hotspot swell—an area of uplifted bathymetry or topography surrounding regional volcanism—is a defining hotspot characteristic, yet its origin is poorly understood. To test current ideas about swell formation, we studied the crust and shallow mantle structure of the Cape Verdes in a passive seismic experiment. The Cape Verde Islands are ∼450 km west of Senegal in the tropical Atlantic Ocean and are on the southwest flank of the Cape Verde Rise, the largest bathymetric anomaly in the oceans, rising ∼2 km above the surrounding seafloor ([Crough, 1982][1]). The archipelago occupies a unique position, an approximately stationary one in the hotspot frame of reference ([Gripp and Gordon, 2002][2]) and therefore with respect to the melting source believed to have produced it. Here we present an analysis of compressional to shear (P to S) converted seismic phases, recorded on a temporary network of seismograph stations on the Cape Verde Islands, that indicate a crust thickened to 22 km is underlain by a high-velocity, low-density layer, which overlies a zone of low shear-wave velocity starting at ∼80 km depth. We also measured shear-wave splitting delay times for teleseismic SKS phases, which are ∼0.81 s, compatible with an origin in this same layer. We interpret these observations as effects of hotspot melting, which produces a thickened crust and a depleted swell root that buoys the ocean floor and spreads laterally as it grows over time. [1]: #ref-5 [2]: #ref-11

Variations in shear‐wave splitting in young Pacific seafloor

Analysis of SKS phases recorded by ocean‐bottom seismometers in the GLIMPSE Experiment has expanded the region of in situ measurements of shear‐wave splitting in young seafloor near the southern East Pacific Rise. Average splitting times at individual GLIMPSE sites range from 1.1 to 2.2 s with fast direction azimuths trending approximately in the direction of absolute plate motion and fossil spreading for the Pacific plate. The GLIMPSE study area at 12 to 14°S is adjacent to the MELT Experiment. The new observations demonstrate that splitting delays are not uniform on the Pacific plate and that the contrast between the Nazca and Pacific plates is not a fundamental consequence of the asymmetric plate motion. The variations in degree of splitting could be caused by small‐scale convection, variations in water content, or rate of deformation in the asthenospheric return flow to the East Pacific Rise.

Detailed mapping of seismic anisotropy with local shear waves in southeastern Kamchatka

SUMMARY The Kamchatka Peninsula lies over a vigorous subduction zone where Pacific and North American plates converge at a rate of almost 80 mm yr −1 . Earthquakes associated with the subduction process provide an excellent source of seismic data for the study of anisotropic properties of the upper mantle and crust overlying the downgoing lithospheric slab. We collected a large set of shear waves from events within the slab recorded by a variety of seismic stations in Kamchatka. Data from permanent and temporary networks cover the entire ∼700 km length of the subduction zone, with 50‐200 km spacing between observing sites, resulting in an unprecedented coverage of the supraslab mantle wedge. We estimated shear wave splitting in selected S waves using two techniques and applied quality tests to ensure measurement stability. Fast directions vary from station to station, and they can vary with backazimuth at individual stations and with direction of propagation for individual sources. In over 350 measurements we recovered meaningful splitting delays, up to 1 s, with most delays in the 0.2‐0.6 s range. Additionally, in nearly 200 measurements splitting could not be resolved, yielding ‘NULL’ observations. Anisotropic properties of the Kamchatka supraslab mantle wedge vary greatly along the volcanic arc and forearc of the subduction zone. Observed anisotropic indicators in the arc and forearc correlate spatially with some tectonic features (e.g. volcanoes). Inland of the volcanic arc most splitting values indicate trench-parallel fast polarization. We do not observe depth dependence in local S-wave splitting delays, consistent with a shallow coherence of anisotropic texture. In the vicinity of Petropavlovsk-Kamchatsky observed anisotropic indicators are coherently trench-normal, and thus consistent with 2-D corner flow. However, splitting above the fragmented slab edge near the Klyuchevskoy volcanic centre is variable and trench-oblique. Birefringence between Petropavlovsk and Klyuchevskoy is weak. Overall, our observations are incompatible with a regional slab-driven corner flow regime.

Relationship between crustal finite strain and seismic anisotropy in the mantle, Pacific-Australia plate boundary zone, South Island, New Zealand

SUMMARY The relationship between crust and mantle deformation in plate boundary zones is an outstanding problem in geodynamics. New Zealand provides a rare opportunity to examine the way strike-slip faults relate to deep-seated zones of lower crustal and mantle flow. A conspicuous bend deflects elongated terranes such as the Dun Mountain ophiolite through >70 ◦ in the continental crust, which we interpret as being the result of distributed dextral shear between the Pacific and Australian plates in the Cenozoic. We utilized variations in the strike of two different geological markers (ophiolite terrane and fold belts) towards the Alpine fault to calculate finite strains in three crustal domains. The deflection is best matched by a transpressional, rather than a simple shear deformation. These transpressional models predict maximum horizontal finite strain azimuths (±10 ◦ ) that trend anticlockwise ∼30 ◦ to ∼10 ◦ from the Alpine fault. These azimuths match published fast polarization azimuths of SKS and local ( 2.1, and strains (es) > 1.9 without being reoriented towards the shear plane by recrystallization. In agreement with our modelling of mantle flow, the large SKS shear wave splitting delays (>2 s) on the South Island suggest that the direction of maximum finite stretch in the mantle is more likely horizontal than vertical. This inference is consistent with the 3-D strain calculated from deflected markers in the crust.

5-ps, 10-GHz pulse generation from an all-optical semiconductor switch embedded in a ring cavity

We observed spontaneous generation of nearly transform-limited pulses (5 ps, 10 GHz, 1558 nm) from a semiconductor ring oscillator. This oscillator is completely different from conventional mode-locked lasers in that the pulse width is passively determined by a split delay time inside its ring cavity. This is because the ring cavity includes an all-optical semiconductor switch whose window width is determined by the split delay time.

Shear-wave velocity structure beneath North Island, New Zealand, from Rayleigh-wave interstation phase velocities

Fundamental-mode Rayleigh-wave dispersion velocities have been determined for interstation paths along the strike of the Hikurangi Margin in eastern North Island, New Zealand, for the period range 17 to 73 s. We invert the observed surface-wave dispersion velocities to show S-wave velocities of 4.2–4.4 km s−1 (Vp∼ 7.5–7.8 km s−1) within a subducting crust fixed at 12 km thickness, overlying an upper-mantle lid with S-wave velocities between 4.7 and 4.9 km s−1 (Vp∼ 8.3–8.7 km s−1), some 40 to 20 km, thick respectively. The inherent non-uniqueness in the inversion controls the surface-wave resolution at these depths to a minimum layer thickness and thickness increment of about 20 km. Earlier body-wave studies divide this upper-mantle lid into a 10 km thick layer of normal mantle velocities (Vp∼ 8.2 km s−1) overlying a very high-velocity layer (Vp∼8.9 km s−1). The implication from this surface-wave study is that this latter layer cannot be much thicker than about 10 km. The lithospheric thickness of the subducting Pacific plate is of the order of 30 to 50 km, which is at the lower end of the thicknesses found elsewhere for >100 Ma oceanic lithosphere. The depth to the S-wave upper-mantle low-velocity zone of approximately 60 to 80 km correlates with the maximum depth of seismicity along the belt. The thickness and S-wave velocity of the low-velocity zone is controlled by the inversion to be 60–80 km thick with a velocity of 4.6± 0.1 km s−1 (Vp∼ 8.2± 0.15 km s−1). Anisotropy of oriented olivine has recently been used to explain the presence of high velocities in the subducted slab's uppermost mantle in eastern North Island. However, large SKS splitting delays require the anisotropy to be more extensive in depth. We have carried out inversion of Rayleigh-wave interstation phase-velocity dispersion data assuming an anisotropic mantle with pseudo-hexagonal symmetry. Results indicate that mantle velocities derived from the isotropic inversion are overestimates of the S-wave velocity in the vertical plane of the anisotropic structure, perpendicular to the symmetry axis, by less than 3 per cent; this is within the error bounds of the isotropic inversion results. The LVZ is less well pronounced in anisotropic inversions.

Variation of Shear-wave Residuals and Splitting Parameters from Array Observations in Southern Tibet

A tight array of seismographs spanning a 500 km traverse of southern Tibet resolved anisotropy from SKS with a spatial variation of its direction and an increase northward of the splitting delay, as well as of its first arrival residual. Both waves split by velocity anisotropy are slow relative to P and their waveform analysis may be interpreted to suggest attenuation anisotropy. The array here provides examples of residuals and splitting of other S waves which do not tightly conform to the anisotropy assumed in the simplest model of olivine of transverse isotropy with horizontal symmetry axis. S waves are also split, with parameters which vary along the array, and hence are relevant to near-receiver structure like those of SKS. Their splitting delay, for non-vertical incidence and polarization, appears larger than that for SKS. Residuals of S-wave first arrivals and splitting delays increase less northwards for S than SKS. Anisotropy with a slow vertical axis may account for these observations. Its origin may be related to horizontal shear or flow in low-velocity layers.

High-resolution Y-axis readout for delay-line microchannel anodes

We have devised a method to obtain a high-resolution Y-axis event position determination from microchannel plate delay-line detectors. The method is based on the double-delay-line wedge–wedge charge partition principle of Lampton et al. [M. Lampton, O. H. W. Siegmund, and R. Raffanti, IEEE Trans. Nucl. Sci. 37, 1548 (1990)], where the X axis is read out by a pair of side-by-side delay lines and the Y-axis coordinate is determined by comparing the charges on the two parallel delay lines. However, our new method abandons the common field return path of the two delay lines, and splits them into two adjacent but independent transformer coupled timing circuits having no common ground connection except via the charge-measurement amplifiers. Thus we eliminate the large electrostatic capacitance of the delay lines from the input of the charge-measuring system. Due to the very low output return capacitance of the split delay lines, the Y-axis noise performance is improved enormously. Over its limited working field, the spatial resolution in the Y coordinate can easily equal the resolution provided in the X coordinate by the delay line. Applications to extreme and far ultraviolet photon-counting spectroscopy are envisioned.