Multichannel analysis of passive surface waves based on crosscorrelations

Abstract

Passive seismic methods in highly populated urban areas have gained much attention from geophysics and civil engineering communities because traditional seismic surveys, especially in complex urbanized environments, might be improperly applied. In passive seismic methods, directional noise sources will inevitably bring azimuthal effects and spatial aliasing to dispersion measurements due to the fact that true randomness of ambient noise cannot be achieved in reality. To solve these problems, multichannel analysis of passive surface (MAPS) waves based on long noise sequence crosscorrelations is proposed. We have introduced a hybrid method of seismic interferometry and the roadside passive multichannel analysis of surface waves (MASW) using crosscorrelation to produce common virtual source gathers from 1 h multichannel noise records. Common virtual source gathers are then used to do dispersion analysis with an active scheme based on phase-shift measurement. Synthetic tests demonstrated the advantages of this method with azimuthal adjustment and dispersion imaging for directional noise source distribution. Two field applications were conducted, and results from the roadside passive MASW, MAPS, and spatial autocorrelation method were compared. Our study indicated the superiority of MAPS over the roadside passive MASW on the validity of azimuth detection, feasibility of combining the active MASW and MAPS, and accuracy in determining dispersion energy trends, especially at a relative low-frequency range ([Formula: see text]) in urban areas.