A surface-wave study of the Pacific Ocean basin

Abstract

SUMMARY Group velocity dispersion data from Rayleigh waves is used to map the crustal and upper-mantle structure beneath the Pacific Ocean basin. Data from the Global Digital Seismic Network (GDSN) provides dispersion information of a broad period range (15–300 s) necessary to achieve adequate resolution down to the upper mantle. No a priori assumption on regionalization is made; a stochastic grid-inversion method is used instead that allows the extraction of pure-path dispersion data from mixed-path measurements. The pure-path grid size is 10°× 10°, and the entire Pacific Ocean basin is divided into more than 100 grids. The large number of wave paths used permits sufficient resolution in the process of pure-path decomposition. A second generalized inversion is applied to the pure-path dispersion data for each individual grid element, resulting in shear velocity as a function of depth. The combined output of the velocity-depth functions of all grid elements gives a three dimensional crustal and upper-mantle structure of the Pacific Ocean basin. This enables cross-sections and depth slicings of the region to be made down to a depth of 300 km. Results show a detailed picture on the lateral heterogeneity of the crustal and upper mantle. The average thickness of the oceanic crust is less than 10 km. A few regions in the ocean basin are identified as having a continent-like crust. A pronounced low-velocity zone with shear velocity below 4.3 km s-1 exists nearly everywhere in the depth range from 70 to 200 km. The depth and thickness of this low-velocity layer vary over the Pacific Ocean basin, but are in general conformance with the known surface tectonic picture. The lateral heterogeneity and the regional distribution of velocity variations provide good baseline data for quantitative discussions of an upper mantle flow pattern and the associated dynamic process.