Some problems in global tomography: modeling the core-mantle boundary and statistical analysis of travel-time data

[摘要] Core-mantle boundary (CMB) structure is mapped with travel-time data in the International Seismological Centre (ISC) catalog. Different subsets of ISC data yield inconsistent results for CMB topography. Correlations of anomalies in well covered areas indicate a common source of signal above the CMB. Allowing for heterogeneity above the CMB yields models of 4 km peak-to-peak CMB topography and 2.5% velocity variations in the D" layer. The models contain a large degree-two, zonal component. This agrees with the results of a stacking procedure employed in search of an axisymmetric pattern in the ISC data. It is concluded that the axisymmetric pattern in antipodal PKIKP data must at least in part be explained by CMB structure. A method is developed to extract estimates of random data-errors and statistical measures of the earth's structure from global, travel-time data. Application of this method to teleseismic P-wave data yields an estimate of the signal-to-noise ratio in excess of unity. Heterogeneity is found to be strongly concentrated near the earth's surface. It extends over a wide range of scales in the upper mantle, but is restricted to relatively large scales in the lower mantle. The root-mean-square level of heterogeneity is found to be of the order of 0.1% in the lower mantle, but several times higher at its top and bottom. Synthetic (2D) tests of lower-mantle and CMB travel-time inversions indicate only partial success for the lower mantle and a 5 to 10 km uncertainty in CMB mapping. This is consistent with the inconsistencies in the ISC data, when mapped onto the CMB, and suggests that the above results pertaining to CMB structure may be obscure.

[发布机构] University:California Institute of Technology;Department:Geological and Planetary Sciences

[关键词] Geology