T51D-2907
Multiscale Finite-frequency Seismic Imaging of the Southern Alaska Subduction Zone

Friday, 18 December 2015
Poster Hall (Moscone South)
Xin Song, University of Toronto, Toronto, ON, Canada, Shu-Huei Hung, NTU National Taiwan University, Taipei, Taiwan, Ping Tong, Stanford University, Stanford, CA, United States and Qinya Liu, University of Toronto, Physics, Toronto, ON, Canada
Abstract:
Southern Alaska is one of the most seismically active regions in north America as the Pacific plate subducts northward beneath North America plate along the Aleutian trench. In this study, we determine 3-D variations of P- and S-wave speed and Possion's ratio (Vp/Vs) perturbations of the southern Alaska subduction zone based on broadband tele-seismic data recorded by 198 seismic stations for about 2000 events with magnitudes greater than 5.5 during the period from June 2000 to December 2014. Relative arrival times of P and S phases bwtween stations are accurately measured by adapting the efficient multi-channel cross-correlation (MCCC) technique. The obtained arrival-time data are then used to tomographically image the Vp and Vs structures beneath the stations based on 3-D finite-frequency sensitivity kernels and a wavelet-based multi-scale model parameterization. Our results show strong positive velocity anomalies in the crust and upper mantle starting at a depth of about 50km and extending to northwestward down to a depth of 200 km and covering about 350 km in horizontal distance. The high velocity feature interpreted as a cold slab has a thickness of about 50km and a subducting angle of about 45o, consistent with some previous studies of southern Alaska. We also plan to further obtain high-resolution seismic imaging of southern Alaska subduction zone by utilizing the converted and coda waves of tele-seismic main phases (e.g., P and S) based on a hybrid tomographic technique combining spectral-element method (SEM) and frequency-wavenumber (FK) method. The 3D Vp and Vs models obtained from finite-frequency traveltime tomography thus can serve as a proper starting velocity model for the hybrid SEM-FK imaging to further reveal high-resolution details of the subduction zone.