Application of wavefield compressive sensing in surface wave tomography. Issue 3 (15th March 2018)

• Record Type: Journal Article
• Authors: Zhan, Zhongwen; Li, Qingyang; Huang, Jianping
• Is Part Of: Geophysical journal international. Volume 213:Issue 3(2018:Jun.)
• Page Start: 1731
• Page End: 1743
• Publication Date: 2018-03-15
• Subjects: Computational seismology -- Crustal imaging -- Seismic tomography -- Surface waves and free oscillations -- Wave propagation; Geophysics -- Periodicals; 550
• DOI: 10.1093/gji/ggy082
• Languages: English
• ISSNs: 0956-540X

Abstract

SUMMARY: Dense arrays allow sampling of seismic wavefield without significant aliasing, and surface wave tomography has benefitted from exploiting wavefield coherence among neighbouring stations. However, explicit or implicit assumptions about wavefield, irregular station spacing and noise still limit the applicability and resolution of current surface wave methods. Here, we propose to apply the theory of compressive sensing (CS) to seek a sparse representation of the surface wavefield using a plane-wave basis. Then we reconstruct the continuous surface wavefield on a dense regular grid before applying any tomographic methods. Synthetic tests demonstrate that wavefield CS improves robustness and resolution of Helmholtz tomography and wavefield gradiometry, especially when traditional approaches have difficulties due to sub-Nyquist sampling or complexities in wavefield.