Michael E. Wysession

Shear wave splitting, continental keels, and patterns of mantle flow

In this study we investigated the origin of seismic anisotropy in the mantle beneath North America. In particular, we evaluated whether shear wave splitting patterns in eastern North America are better explained by anisotropy caused by lithospheric deformation, anisotropy due to mantle flow beneath the lithosphere, or a combination of both. We examined new measurements of shear wave splitting from the Missouri to Massachusetts broadband seismometer array (MOMA), the North American Mantle Anisotropy and Discontinuity experiment (NOMAD), as well as splitting parameters from several previous studies. We developed a simple finite difference model that approximates mantle flow around a complex, three-dimensional continental lithospheric keel. To evaluate potential anisotropy from mantle flow beneath the lithosphere in eastern North America, we compared shear wave splitting observations to predicted splitting parameters calculated using this mantle flow model. Our results indicate that a significant portion of observed shear wave splitting in eastern North America can be explained by mantle flow around the continental keel. However, shear wave splitting patterns in a few regions of eastern North America indicate that a component of lithospheric anisotropy must exist, particularly in regions containing the largest keel thicknesses. For eastern North America, as well as for splitting observations in Australia, Europe, and South America, we favor a model in which anisotropy is controlled by a combination of both lithospheric deformation and subcontinental mantle flow.

The structure of the core-mantle boundary from diffracted waves

Diffracted P and S waves (Pd, Sd) traveling around the core-mantle boundary (CMB) of the Earth give us information about the velocity structure and therefore the thermochemistry of D″, the base of the Earths mantle. By examining Pd and Sdarrivals we determined the apparent ray parameter for different regions at the base of the mantle. By comparing the data slownesses to those found from reflectivity synthetic seismograms we were able to quantify D″ average velocities. Using these averaged velocities with a thermochemical modeling of lower mantle minerals using a Birch-Mumaghan equation of state, we have been able to make chemical and physical inferences as to the causes of lateral variations at the CMB. Examinations found significant lateral heterogeneity at the base of the mantle, amounting to ≈ 4% for both P and S velocities. These velocities did not always vary in parallel, and the Poisson ratio varied regionally by almost 6%. The most unusual region of the CMB found was under Indonesia, where velocities 3% slower than the preliminary reference Earth models were found adjacent to a region of faster than average velocities. These regions currently correspond to areas of core up welling and down welling (respectively) found by Voorhies (1986), which if mostly held in place by core-mantle coupling might cause a flux of heat and iron into the mantle, making the anomaly both thermally and chemically derived. At the CMB under the northern Pacific rim the fastest shear velocities were found, but the same region yielded slower than average P velocities. While the presence of fast shear velocities here would support the idea that we are seeing the cold dregs of mantle convection, perhaps continuing down from the North Pacific subduction zones, the presence of slow P velocities suggests additional complications. Our thermochemical modeling suggests that the D″ Poisson ratio is very sensitive to variations in the silicate/oxide ratio and that a decrease in the amount of perovskite relative to magnesiowustite may play an important role in this region.

Mantle discontinuities and temperature under the North American continental keel

A ubiquitous feature of upper-mantle seismic velocity models has been the presence of high-velocity ‘keels’ beneath stable continental interiors. Uncertainty remains, however, regarding the maximum depth to which continental keels extend, the degree to which they have cooled the mantle that surrounds them and their role in mantle flow. Here we investigate thermal anomalies across the eastern margin of the North American continental keel by imaging the seismic discontinuities at depths of 410 and 660 km with compressional-to-shear converted waves recorded by a 1,500-km-long seismometer deployment in the eastern United States. The thickness of the transition zone (the region nominally between depths of 410 and 660 km) and the depth to the ‘410-km’ discontinuity indicate that cold keel material and sub-keel downwellings must be largely confined to the upper mantle and may impinge on the transition zone only in localized regions and with thermal anomalies of less than ∼150 K. A 20-km depression of the ‘660-km’ discontinuity to the south of the westernmost stations coincides with a region of fast velocity in the deep transition zone and may be associated with the remnants of the subducted Farallon plate,,.

Intraplate Seismicity of the Pacific Basin, 1913–1988

We establish here a comprehensive database of intraplate seismicity in the Pacific Basin. Relocation and analysis of 894 earthquakes yield 403 reliable intraplate earthquakes during 1913–1988. These numbers do not include earthquake swarms, which account for another 838 events. Most of the remainder (304 events) are actually plate boundary earthquakes that have been erroneously located in intraplate regions. A significant number occur in recent years when location capabilities should have guarded against this situation. Relocations involve a careful linear inversion ofP andS arrivals, accompanied by a Monte Carlo statistical analysis. We have also attentively removed the high number of clerical errors and nuclear tests that exist in epicenter bulletins.A geographical examination of the relocated epicenters reveals several striking features. There are three NW-SE lineaments north of the Fiji Plateau and in Micronesia; diffuse seismicity and incompatible focal mechanisms argue against the southernmost, discussed byOkalet al. (1986) andKroenke andWalker (1986), as the simple relocation of the Solomon trench to the North. Besides another striking lineament, along the 130°W meridian, there is also a strong correlation between seismicity and bathymetry in certain parts of the Basin. In the Eastcentral Pacific and Nazca plates there are many epicenters on fracture zones and fossil spreading ridges, and hot spot traces like the Louisville, Nazca and Cocos Ridges also display seismicity.

Lowermost mantle anisotropy beneath the Pacific: Imaging the source of the Hawaiian plume

We utilized recordings of seismic shear phases provided by several North American broadband seismometer arrays to provide unique constraints on shear wave anisotropy beneath the northern and central Pacific Ocean. Using a new analysis method that reduces measurement errors and enables the analysis of a larger number of available waveforms, we examined relative travel times of teleseismic S and Sdiff that sample a large area of lowermost mantle structure. The results of this study provide evidence for small-scale lateral and depth variations in shear wave anisotropy for a broad region of the lowermost mantle beneath the Pacific Ocean. In particular, we image a localized zone of anomalously strong anisotropy whose strength increases toward the top of DQ beneath Hawaii. Our results, combined with a previous study of VP/VSH ratios, indicate that ancient subducted slab material may be responsible for observations beneath the northern Pacific, while lenses or layers of core^mantle boundary reaction products or partial melt, oriented by horizontal inflow of mantle material to the Hawaiian plume source, can explain observations beneath the central Pacific. fl 2001 Elsevier Science B.V. All rights reserved.

Mapping the lowermost mantle using core-reflected shear waves

A map of laterally varying D″ velocities is obtained for the region from 50°S to 50°N in latitude and 70°E to 190°E in longitude. Velocities are found using an analysis of the differential travel time residuals from 481 ScS-S and 266 sScS-sS phase pairs. The long-period data are taken from the Global Digital Seismograph Network digital waveform catalog for the time period of January 1980 to March 1987. Each differential travel time is found by a cross correlation of the S phase ground displacement, corrected to simulate differential attenuation, with all following phases. Travel times are corrected for ellipticity and mantle heterogeneity outside of their D″ paths, and the remaining residuals are interpreted as the result of D″ heterogeneity. Ray-tracing tests are made to check the validity of converting travel time residuals into velocity path anomalies.The resulting map reveals significant long-wavelength D″ structure including a 3% low-velocity region beneath northeastern Indonesia, surrounded by three identified high-velocity zones beneath northwestern Pacifica (+4%), Southeast Asia (+3%), and Australia (+3–5%). This structure is of continent/ocean spatial scales and is most likely created by dynamic processes dominant in the lower mantle. The low-velocity region may have both chemical and thermal origins and is very possibly the site of an incipient lower mantle plume where mature D″ rock which has been heated by the core has become gravitationally unstable and begun to rise. A chemical component possibly exists as a chemical boundary layer is dragged laterally toward the plume site, much the way continents are dragged toward subduction zones. The high-velocity zones possibly result from the downward convection of cold lower mantle plumes, which pond at the core-mantle boundary. These seismic anomalies may also contain a chemical signature from faster iron-poor materials brought down through the lower mantle or the additional presence of SiO2 stishovite, perhaps in its higher-pressure polymorph.

Seismic Evidence for Subduction-Transported Water in the Lower Mantle

We use seismic attenuation tomography to identify a region at the top of the lower mantle that displays very high attenuation consistent with an elevated water content. Tomography inversions with >80,000 differential travel-time and attenuation measurements yield 3D whole-mantle models of shear velocity (V S ) and shear quality factor (Q μ ). The global attenuation pattern is dominated by the location of subducting lithosphere. The lowest Q μ anomaly in the whole mantle is observed at the top of the lower mantle (660-1400 km depth) beneath eastern Asia. The anomaly occupies a large region overlying the high-Q μ sheet-like features interpreted as subducted oceanic lithosphere. Seismic velocities decrease only slightly in this region, suggesting that water content best explains the anomaly. The subducting of Pacific oceanic lithosphere beneath eastern Asia likely remains cold enough to transport stable dense hydrous mineral phase D well into the lower mantle. We propose that the eventual decomposition of phase D due to increased temperature or pressure within the lower mantle floods the mantle with water, yielding a large low-Q μ anomaly.

Using MOMA Broadband Array ScS‐S data to image smaller‐scale structures at the base of the mantle

ScS-S residuals obtained at stations of the Missouri-to-Massachusetts (MOMA) temporary broadband seismic array are used to delineate variations in seismic velocity structure above the core-mantle boundary (CMB) at scales smaller than observable with tomographic models. South American earthquakes recorded at MOMA reveal a slow-velocity anomaly that is at least as small as the limit of the resolution of ScS waves, about 300 km across. This is modeled as being within a region of fast velocities in whole-mantle models. The slow ScS-S residuals correlate well with a peak in ScS/S relative amplitudes. The small region of slow shear velocity at the CMB could be a pocket of lower mantle rock trapped beneath the descending Farallon slab, or evidence of chemical boundary layer variations.

Intraplate earthquakes in the southwest Pacific Ocean Basin and the seismotectonics of the southern Tasman Sea

An examination of 311 intraplate earthquakes in the Australian plate portion of the Pacific Ocean basin reported from 1918 to 1990 reveals that only 113 events are reliably intraplate, with most of the rest relocating to active trenches and transforms. The non-random distribution of the reliably intraplate events gives insight into the tectonic stresses present. The central Tasman Sea is mostly aseismic except for a swarm of activity at the predicted site of the Tasmantid hot spot. To the north, the broad regions of the Coral Sea, South Fiji Basin and Lord Howe Rise show very little intraplate seismicity, yet the narrow Norfolk Ridge and Three Kings Rise, caught between the double convergence of the New Hebrides and Tonga subduction zones, support many more earthquakes. High levels of intraplate seismicity in the southern Tasman Sea adjacent to the Macquarie Ridge Complex (MRC) indicate that this region may be undergoing internal deformation due to the unusual nature of the Australia-Pacific plate boundary. Additional support exists in the form of intraplate focal mechanisms similar to those at the plate boundary and a set of parallel gravity rolls which are observed in recent geoid maps. Some aftershocks of the Mw = 8.2 Macquarie Ridge earthquake of 1989 occurred in a fracture zone west of the Macquarie Ridge Complex [Das, 1992], but we have found several earthquakes from as early as 1924 which relocate to this feature, suggesting that its reactivation may be more significant than previously thought. This reactivation of a fossil fracture zone may be the result of the increasing amount of oblique convergence between the Australia and Pacific plates at the Macquarie Ridge Complex, formerly a spreading center, and the stresses associated with subducting recently formed Australian ocean crust beneath the older Pacific plate.

Investigating the base of the mantle using differential travel times

Abstract Several techniques using differential seismic travel times to map lateral structure in the lowermost mantle are discussed. Results are shown for recent studies involving the established techniques of core-reflected phases (ScS-S and PcP-P) and diffracted phase profiles (Sdiff), and new techniques involving the differential times of both core-transmitted and core-diffracted phases (PKP-Pdiff and Sdiff-SKS-SKKS) are described. The recent databases of digital seismograms have allowed for a study of D″ velocities in the Eastern Hemisphere using ScS-S and sScS-sS differential times from the many Western Pacific earthquakes. The result is an image at a resolution of a few hundred kilometers of a slow velocity anomaly of 2500 km width beneath Micronesia (−2% relative to the Preliminary Reference Earth Model (PREM)) that is surrounded on three sides by fast D″ rock that is 3% faster than PREM. A study using the differential arrivals of core-diffracted S waves (Sdiff) from digital records is providing information about long-wavelength variations in D″ shear velocities, though the rigorous earthquake-station geometry requirements limit the study to particular regions of the globe. Another study is using over 40 000 PcP-P differential travel times as reported to the International Seismological Centre to map global P velocities at the base of the mantle, and it shows that global coverage of the core-mantle boundary (CMB) is very poor. Though there are some regions (Northern Asia, Northern Pacific, Central America) with enough data sampling to allow a quantification of average D″ P velocities (with a total robust range of 4% lateral variation), they cover only a small portion of the total CMB. As a means of increasing our understanding of the long-wavelength variations of seismic velocities, a description is given of two techniques that will take advantage of totally different sets of earthquake-station geometries from the core-reflected phase studies. In the distance range from 120° to beyond 165° the differential times of PKP and Pdiff can be used to map long-wavelength D″ P velocities. These two phases are very different in shape and frequency content, so the differential times are found by a waveform cross correlation with reflectivity synthetic counterparts. In the distance range from 105° to beyond 135° the differential times of Sdiff, SKS, and SKKS can be examined simultaneously to test models of velocity structure above and below the CMB, also through comparison with synthetic counterparts.