Iterative passive-source location estimation and velocity inversion using geometric-mean reverse-time migration and full-waveform inversion. Issue 3 (9th September 2020)

• Record Type: Journal Article
• Authors: Lyu, Bin; Nakata, Nori
• Is Part Of: Geophysical journal international. Volume 223:Issue 3(2020)
• Page Start: 1935
• Page End: 1947
• Publication Date: 2020-09-09
• Subjects: Numerical modelling -- Numerical solutions -- Waveform inversion -- Wave propagation; Geophysics -- Periodicals; 550
• DOI: 10.1093/gji/ggaa428
• Languages: English
• ISSNs: 0956-540X

Abstract

SUMMARY: Passive-seismic provides useful information for reservoir monitoring and structural imaging; for example, the locations of microseismic events are helpful to understand the extension of the hydraulic fracturing. However, passive-seismic imaging still faces some challenges. First, it is not easy to know where the passive-seismic events happened, which is known as passive-source locating. Additionally, the accuracy of the subsurface velocity model will influence the accuracy of the estimated passive-source locations and the quality of the structural imaging obtained from the passive-seismic data. Therefore the velocity inversion using the passive-seismic data is required to provide the velocity with higher accuracy. Focusing on these challenges, we develop an iterative passive-source location estimation and velocity inversion method using geometric-mean reverse-time migration (GmRTM) and full-waveform inversion (FWI). In each iteration, the source location is estimated using a high-resolution GmRTM method, which provides a better focusing of passive-source imaging compared to conventional wavefield scanning method. The passive-source FWI is then followed to optimize the velocity model using the estimated source location provided by GmRTM. The source location estimation and velocity inversion are implemented sequentially. We evaluate this iterative method using the Marmousi model data set. The experiment result and sensitivity analysis indicate that the proposed method is effective to locate the sources and optimize velocity model in the areas with complicated subsurface structures and noisy recordings.