Imaging Attenuation From Array Analysis of Surface Waves

Abstract

Anelastic attenuation provides key insight in our understanding of thermal and rheological structures and the associated deformation and dynamic mechanisms of the Earth's deep interior. Unfortunately, attenuation tomography is advanced far behind wave-speed tomography due to the challenge in properly excluding the complex effects of elastic heterogeneities on seismic wave amplitude. By taking advantage of phase tracking in seismic array analysis, here we derive a new theory of Helmholtz tomography that well accounts for attenuation, source radiation, and scattering, etc., and present a technique called Helmholtz Multi-Event Tomography (HelMET) to retrieve the attenuation properly. The effectiveness of this method is then validated by synthetic inversions. Our synthetic seismograms are calculated using a newly developed three-dimensional finite-difference algorithm that accounts for physical dispersion and dissipation in anelastic media and remains accurate and stable even if strong attenuation exists. Compared to the traditional method poorly performed in the synthetic inversion, the HelMET well recovers the input attenuation anomalies, suggesting that this method can be used to successfully isolate attenuation from the complicated effects of elastic heterogeneities. Our results underline the implication of the new theory and method in accurately imaging high-resolution attenuation structures and unambiguously interpreting the anelastic heterogeneities of the Earth by array-based earthquake and ambient noise data with inexpensive computation.