HR: 1340h
AN: T33C-0564    [Abstracts]
TI: Sensitivity of Predicted Upper Mantle Seismic Anisotropy to LPO and Finite Strain Modelling Techniques
AU: * Nippress, S
EM: nippress@liverpool.ac.uk
AF: University of Liverpool, Department of Earth and Ocean Sciences, Jane Herdman Laboratories, Liverpool, L69 3GP United Kingdom
AU: Kusznir, N
EM: sr11@liverpool.ac.uk
AF: University of Liverpool, Department of Earth and Ocean Sciences, Jane Herdman Laboratories, Liverpool, L69 3GP United Kingdom
AU: Kendall, M
EM: gljmk@bristol.ac.uk
AF: University of Bristol, Department of Earth Sciences Wills Memorial Building Queen's Road , Bristol, BS8 1RJ United Kingdom
AB: Observations of shear wave splitting, and hence seismic anisotropy, offer insights into the nature of plate boundaries. Both finite strain and lattice preferred orientation (LPO) methods have been developed by numerous authors in order to predict observed upper mantle seismic anisotropy. The goal of this study is to compare the seismic anisotropy and hence shear-wave splitting predicted using LPO and finite strain methods for a mid-ocean ridge and subduction mantle wedge. We use finite element (FE) modelling to predict subduction zone flow and deformation arising from incompressible fluid flow driven by the body forces arising from the negative buoyancy of the subducting slab. The FE model is driven dynamically by the excess density (50kg/m³) of the subducting slab within the upper mantle, rather than using plate-like velocity boundary conditions. Ocean ridge flow is modelled using a corner flow solution. We use the resulting mantle flow and deformation fields to model seismic anisotropy and shear wave splitting in each setting using both LPO and finite strain methods. We find that seismic anisotropy and shear wave splitting predicted using LPO theory is extremely sensitive to mineral input parameters, such as grain boundary mobility, grain boundary sliding and mineral composition, however experimental constraints greatly reduce the range of predicted values. Defining mantle mineralogy is extremely important in the LPO modelling of seismic anisotropy. For example, shear wave splitting, with a 100% olivine aggregate predicts much larger splitting than that observed, while an aggregate of 70% olivine and 30% enstatite predicts splitting that compares better with observations. We test three conversion schemes (Ribe, 1992; Zhang et al., 2000; Ismail and Mainprice, 1998) from finite strain to seismic anisotropy and find that only the Ribe [1992] scheme compares well to observations. The other schemes predict magnitudes of shear wave splitting well in excess of the observations. When the mantle viscosity structure is altered the predicted seismic anisotropy near to the top of the subducting slab varies while at distance it remains the same. Sensitivity of predicted seismic anisotropy to the parameter space was also tested. We find that when appropriate mineral input parameters and mineralogy are used in the LPO modelling, and the Ribe [1992] conversion is used for the finite strain modelling the two schemes predict similar magnitudes of shear-wave splitting to each other and to observations in both mid-ocean ridge and mantle-wedge settings. Tracing seismic rays from the slab edge to the surface in our subduction model shows that the degree of splitting is sensitive to the distance from the slab and the depth of the source. For teleseismic synthetic rays, the predicted shear wave splitting is 0.5-1.0sec for our mid-ocean ridge and subduction mantle wedge models.
DE: 7208 Mantle (1212, 1213, 8124)
DE: 7240 Subduction zones (1207, 1219, 1240)
DE: 7245 Mid-ocean ridges
DE: 8170 Subduction zone processes (1031, 3060, 3613, 8413)
SC: Tectonophysics [T]
MN: Fall Meeting 2005