SHEAR-WAVE SPLITTING IN REFRACTED WAVES RETURNED FROM THE UPPER-MANTLE TRANSITION ZONE BENEATH NORTHERN AUSTRALIA

Citation
C. Tong et al., SHEAR-WAVE SPLITTING IN REFRACTED WAVES RETURNED FROM THE UPPER-MANTLE TRANSITION ZONE BENEATH NORTHERN AUSTRALIA, J GEO R-SOL, 99(B8), 1994, pp. 15783-15797
Citations number
37
Categorie Soggetti
Geosciences, Interdisciplinary
Journal title
JOURNAL OF GEOPHYSICAL RESEARCH-SOLID EARTH
ISSN journal
21699313 → ACNP
Volume
99
Issue
B8
Year of publication
1994
Pages
15783 - 15797
Database
ISI
SICI code
2169-9313(1994)99:B8<15783:SSIRWR>2.0.ZU;2-2
Abstract
The broadband recording site at Warramunga (WRA) in the Northern Terri tory of Australia provides good coverage of seismic wave propagation t hrough the upper mantle for sources in the earthquake belt through Ind onesia and New Guinea. S waves recorded on the radial (SV) and tangent ial (SH) components are of comparable quality because the hard-rock re cording site minimizes the influence of coupling to P on the radial co mponent. Refracted S waves from the uppermantle transition zone show a clear advance of SH wave arrivals compared with SV. Eleven polarizati on analyses of waves returned from the transition zone yield an averag e time shift of 2.3 s with the fast direction scattered about the tran sverse direction. Nine polarization measurements of waves returned fro m the top of the lower mantle yield an average time shift of 1.7 s, ag ain with the fast direction near the transverse. No appreciable time d ifferences are observed between the radial and transverse polarization s for paths refracted within the lithospheric lid. Because the observa tions of shear wave splitting in waves passing through the low-velocit y zone, the transition zone, and the top of the lower mantle are not c oherent in their absolute polarization, the cause cannot lie in azimut hal anisotropy at shallow depths under the WRA station. The most plaus ible explanation is transverse isotropy in shear within the low-veloci ty zone under the unusually thick mantle ''lid'' under Australia. A po ssible contribution may come from anisotropy in beta-olivine at the to p of the upper mantle transition zone. Transverse isotropy in the 200- km-thick layer below the lithosphere down to the transition zone with a 1% faster shear wave speeds for horizontal polarization compared wit h vertical polarization will explain the splitting data. For this asth enospheric region the level of anisotropy is quite reasonable and the polarization is consistent with lateral flow. The geometry of the avai lable paths for waves propagating within the mantle lid is not suffici ent to place constraints on the anisotropic properties of this heterog eneous and low-loss region.