Regional-scale differential time tomography methods : development and application to the siberia data set
Haijiang Zhang,K. Mackey,K. Fujita,C. Thurber,L. Steck,C. Rowe,S. Roecker,M. Toksoz
2009-01-01
Abstract:This is the final year of a collaborative project by the University of Wisconsin-Madison (UW-Madison), Massachusetts Institute of Technology (MIT), Michigan State University (MSU), and Los Alamos National Laboratory (LANL). There are four main tasks in this project: (1) an extension of our development of double-difference (DD) seismic tomography to the use of station-pair residual differences, including incorporation of a new method for resolution matrix calculation; (2) testing, refinement, and adaptation of a method for spherical-earth finite-difference (SEFD) travel time calculations for use in DD tomography; (3) an extension of our Cartesian adaptive-grid DD tomography algorithm to spherical coordinates; and (4) collaborative work among the UW-Madison, MIT, MSU, and LANL groups to apply these analysis tools to the Siberia data set. As reported in previous Monitoring Research Review proceedings, we have successfully incorporated station-pair differential times into the extended DD tomography code. A new resolution matrix calculation method based on the PROPACK package is also incorporated and tested, which is able to efficiently and accurately estimate singular values and vectors for large matrices based on the Lanczos bidiagonalization with partial reorthogonalization. In our tomography algorithm, we use a spherical-Earth finite-difference (SEFD) travel time method to calculate travel times and trace rays. The basic concept is the extension of a standard Cartesian FD travel time algorithm to the spherical case through development of a mesh in radius, co-latitude, and longitude, expression of the finite difference (FD) derivatives in a form appropriate to the spherical mesh, and the construction of “stencils” to calculate extrapolated travel times. We benchmarked the SEFD method against the “sphere-in-a-box” Cartesian FD travel time algorithm (Flanagan et al., 2007). We have applied the extended DD tomography algorithm separately to the southern (Baikal and Amur regions) and northern (Magadan and eastern Yakutsk regions) parts of the eastern Siberia. The velocity models in both parts show strong heterogeneities at shallow depths, as expected for the variable and complicated nature of the crust. For this last year, we focus our effort on extending the spherical regular-grid DD tomography code to the adaptive version, in which the inversion grid nodes are adapted according to the data distribution. In addition to the data we already collected for the Baikal, Amur, Magadan and Yakutsk regions, we also collect the data for the Kamchatka region. We first apply the spherical regular-grid DD tomography method to the whole data set to obtain a seamless velocity model for eastern Siberia. The adaptive-grid DD tomography method will also be applied to the whole data set. The model will be tested using the peaceful nuclear explosions (PNEs) that occurred and were recorded in the region. 2009 Monitoring Research Review: Ground-Based Nuclear Explosion Monitoring Technologies