Three-Dimensional Vp and Vs Models of Continental China From Joint Inversion of Body Wave, Surface Wave, and Gravity Data

Abstract:

<span"><span">To improve our understanding of the complex geological structure of continental China we need accurate depictions of the 3D structure of the crust and lithospheric mantle. Taking advantage of the increasingly dense seismograph coverage in continental China, several Vp and Vs models at various scales and resolutions have been obtained over the past decades. Tomographic models based either on body wave travel times or surface waves differ, however, in important aspects, especially for the structure beneath the Tibetan Plateau. <span">Internally consistent Vp and Vs models are needed to resolve these differences.

<span">Body wave travel time tomography and surface wave tomography each have strengths and weaknesses. Travel time tomography can yield higher resolution in regions of dense path coverage, and it generally has excellent lateral resolution beneath regions of high seismic activity or dense station distribution. In many other regions, however, the shallow subsurface cannot be resolved adequately by direct P or S travel times. In contrast, surface wave data (from earthquakes or ambient noise) generally yields better radial resolution and has better potential for resolving shallow mantle structure beneath regions that are aseismic or which are void of seismograph stations. Gravity measurements can provide constraints on spatial variations in (mass) density, but like other potential field methods interpretation of gravity anomalies is plagued by substantial ambiguity. Indeed, weak and broad structures in the shallow subsurface can produce the same gravity signal (at the surface) as a small, strong density anomaly at a larger depth.

<span">

<span">To benefit from the complementary sampling of the different data, we have developed a joint inversion scheme that uses body wave travel times, surface wave dispersion, and satellite gravity data to invert for spatial variations in Vp, Vs, and mass density (with the seismic and gravity data linked through an empirical relationship between wavespeed and density). We have tested our joint inversion at various scales, and we will present internally consistent Vp and Vs models and their implications for our understanding of the complex structure of continental China and the Tibetan Plateau.