Detection of a shallow-buried rock obstruction using 2D full waveform inversion. Issue 143 (April 2021)

• Record Type: Journal Article
• Authors: Liu, Yaohui; Zhang, Yunhuo; Nilot, Enhedelihai; Yannick, C.H. Ng; Yang, Ping; Ku, Taeseo
• Is Part Of: Soil dynamics and earthquake engineering. Issue 143(2021)
• Publication Date: 2021-04
• Subjects: FWI -- Rayleigh wave -- Cross-correlation -- MASW -- Geotechnical site investigation; Soil dynamics -- Periodicals; Earthquake engineering -- Periodicals; Sols -- Dynamique -- Périodiques; Génie parasismique -- Périodiques; 624.176205
• DOI: 10.1016/j.soildyn.2021.106644
• Languages: English
• ISSNs: 0267-7261

Abstract

Abstract: Shallow-buried rock mass/obstructions in a reclaimed sand site have seriously delayed a construction project in eastern Singapore, and seismic surveys were conducted to estimate the location and range of the rock. Active-source and passive-source seismic data were recorded using an array with 24 geophones at 1 m intervals. A 2D full waveform inversion (FWI) approach, multichannel analysis of surface waves (MASW), and cross-correlation seismic interferometry were used to process the collected seismic data. Results show that MASW and cross-correlation provide useful information, but limited to 1D S-wave velocity profiles and the upper surface of shallow buried rock. Using the full information of seismic waveform, the FWI approach can reflect even the range of rock. Sequential inversion with a low-pass frequency filter at incremental corner frequencies can increase the precision of the FWI approach. With higher precision, the FWI of surface waves is considered to be an attractive and effective method to detect underground obstruction. Highlights: FWI is used to process seismic data for underground obstruction survey. An accurate 2D S-wave velocity structure can be acquired through FWI. MASW can only reflect the upper surface of granite. Accuracy of crosscorrelation appears lower than MASW and FWI at shallow depth. Sequential application of low pass frequency filters can improve accuracy of FWI.