A study of upper mantle anisotropy from long-period surface wave data

The purpose of this research is to further our knowledge of mantle anisotropy by using long-period surface wave data. We first perform forward-modeling experiments to investigate the propagation behavior of seismic waves in an earth with an anisotropic upper mantle. Synthetic seismograms are constructed with the modal summation technique for a set of simplified earth models in order to understand better the observable aspects of anisotropy in the Earth. Our calculations suggest that upper-mantle anisotropy of a few percent can generate significant waveform anomalies in long-period seismic records of 0-20 mHz (T {dollar}>{dollar} 50s). Those anomalies are termed 'quasi-Love' and 'quasi-Rayleigh' waves and are caused by spheroidal-toroidal free oscillation (Rayleigh-Love surface wave) coupling. Azimuthal anisotropy generates these waveform anomalies more readily than do lateral variations in P and S wave velocities. Significant Quasi-Love waves are difficult to produce with levels of mantle heterogeneity consistent with global tomographic models. In addition, many modeling experiments are performed to demonstrate how spheroidal-toroidal coupling is sensitive to other complex earth structures, such as sharp lateral variations in isotropic S and P velocities, dipping slabs and earth topography. Calculations suggest that surface wave anomalies associated with mixed-type coupling in seismic records at a single station, like S-wave splitting, can be used to diagnose the presence of anisotropic structure in the upper mantle. Quasi-Love and quasi-Rayleigh waves are particularly sensitive to lateral variations in azimuthal anisotropy. The details of anomalous waveforms recorded at a seismic station are determined by a horizontal integral of anisotropic properties along the wave path, which complements the vertical integral offered by observations of shear-wave splitting. Therefore, quasi-Love wave observations are especially useful for studying mantle anisotropy in oceanic regions, where few seismic stations (and therefore shear-wave splitting analyses) are available.; Based on the theoretical results, a large number of digital 3-component seismic records from global seismic networks was analyzed to determine what constraints long-period seismic data can place on upper mantle anisotropy. The preliminary results are very encouraging. Significant waveform anomalies predicted by our theoretical calculations are observed in many seismic data and argue strongly for an anisotropic upper mantle. We have investigated long-period ({dollar}f<15{dollar} mHz) surface waves that have propagated across the Pacific Ocean region. Our observations of quasi-Love waves indicate the existence of strong anisotropic gradients in the western Pacific Ocean, in particular, near Hawaii, and in front of the Tonga-Kermadec, Kurile and Marianas-Izu-Bonin subduction zones. A lack of quasi-Love generation beneath the Phillipine Sea suggests weak azimuthal anisotropy in that region. Seismic data recorded within the Tibetan Plateau show anomalous surface wave propagation, indicating a strong horizontal anisotropic gradient beneath the central Plateau. In the southwest Pacific Ocean, the long-period quasi-Love waveforms recorded at station SNZO can be fit well with a simplified anisotropic earth model with a 90{dollar}spcirc{dollar} rotation in the orientation of the fast P velocity axis near the plate boundary. The laterally-varying anisotropy is attributed to either flow variations in the asthenosphere or variations in fossil spreading direction. (Abstract shortened by UMI.)...