Numerical Modeling of Strong Heterogeneity in the Deep Mantle

University Of California-Berkeley, Berkeley CA

Investigators

Abstract

Abstract for proposal EAR0106000 (PH #38x) Title: Numerical Modeling of Strong Heterogeneity in the Deep Mantle PI's: B. Romanowicz, University of California, Berkeley This project addresses the issue of how to model teleseismic S waveforms to accurately account for strong 3D heterogeneity, such as recently found in D", that cannot be handled using standard approaches, in view of recent developments in numerical modeling, and with the goal of improving our understanding of deep mantle dynamics. In the last few years, much progress has been made in the development of numerical methods adapted to spherical geometry and able to compute waves emanating from a realistic seismic source, reaching, within reasonable computational time, periods of interest for teleseismic studies, making no assumptions on the strength of velocity contrasts, and able to handle interface waves and interface topography. The most promising new method is the spectral element method. It combines the advantages of the flexibility of a finite element method with the precision of a pseudo-spectral method. Nevertheless this method has an excessive numerical cost for realistic frequencies (i.e. at least 15-25 sec). Further progress has been made recently by Capdeville (2000), who developed a hybrid method that couples spectral element computations with a normal mode solution, so that the spectral elements are used only in the target strongly heterogeneous regions. This allows one to focus on the target region and extend computations to much higher frequencies than if spectral elements are used for the whole earth. In collaboration with Dr. Capdeville, we will apply this method to waveform modeling of S-type waves (S, Sdiff, SKS, ScS) interacting with the base of the mantle and the core-mantle boundary. We will particularly be interested in modeling waveform distortion due to heterogeneity, such as that observed in SPdKS, and SVdiff waveforms. In the process, we will also address issues of crust and uppermost mantle effects on teleseismic body and mantle waves. This work will involve adapting Capdeville's hybrid method to the specific case of D", collecting an appropriate global dataset of S waveforms and forward modeling D" structure to match these waveforms.