SEISMIC ANISOTROPY OF OCEANIC UPPER-MANTLE - SHEAR-WAVE SPLITTING METHODOLOGIES AND OBSERVATIONS

Citation
Cj. Wolfe et Pg. Silver, SEISMIC ANISOTROPY OF OCEANIC UPPER-MANTLE - SHEAR-WAVE SPLITTING METHODOLOGIES AND OBSERVATIONS, J GEO R-SOL, 103(B1), 1998, pp. 749-771
Citations number
58
Categorie Soggetti
Geochemitry & Geophysics","Geosciences, Interdisciplinary","Astronomy & Astrophysics",Oceanografhy,"Metereology & Atmospheric Sciences
Journal title
JOURNAL OF GEOPHYSICAL RESEARCH-SOLID EARTH
ISSN journal
21699313 → ACNP
Volume
103
Issue
B1
Year of publication
1998
Pages
749 - 771
Database
ISI
SICI code
2169-9313(1998)103:B1<749:SAOOU->2.0.ZU;2-Y
Abstract
We develop methodologies to obtain accurate measurements of shear wave splitting and apply these techniques to examine the pattern of oceani c upper mantle anisotropy. To obtain high-quality estimates of receive r splitting at island stations, we devise a stacking method that finds the optimum splitting parameters and 95% error region for teleseismic shear-wave phases from a suite of earthquake events. To obtain additi onal measurements in oceanic regions, we develop techniques to measure splitting parameters and errors for SS phases that sample upper mantl e anisotropy at their bounce points. However, we find the data are oft en of low resolution, and anomalous characteristics are sometimes foun d that make splitting difficult to interpret. Ten second splitting of SS is observed across the BANJO seismic array, but we cannot unambiguo usly attribute this signal to mantle anisotropy at the bounce. The rec eiver splitting methods are used to assess the adequacy of a two aniso tropic layer model for the Pacific region, with fast polarization azim uth (phi) in the lithosphere oriented in the fossil spreading directio n and phi in the asthenosphere oriented in the absolute plate motion d irection. This model has been proposed to explain surface wave data in the Pacific Ocean, but our splitting results demonstrate that oceanic anisotropy patterns are more heterogeneous than would be predicted. W hile the island splitting measurements could reflect the influence of individual hotspot upwellings, hotspot effects do not appear to be uni versally dominant. We propose that splitting observations alternativel y indicate broad-scale differences in the underlying character of ocea nic upper mantle anisotropy, associated with coherant patters of litho spheric structure and asthenospheric flow. In particular, splitting, s urface wave models, and regional studies all support a model where lit hospheric anisotropy throughout the South Pacific has been erased or r eoriented toward the absolute plate motion direction, whereas more lim ited observations in the North Pacific indicate that the fossil lithos pheric signature appears to be preserved.