Abstract

Simultaneous source full wavefield inversion (SSFWI) was proposed by Krebs et al. (2009) a decade ago, which could provide orders of magnitude speed-up for full wavefield inversion (FWI) by computing gradient with one encoded shot gather instead of thousands of individual shot gathers. Since then, there have been numerous efforts to apply SSFWI to synthetic and field datasets. However, successful applications with substantial speed-ups are limited to the synthetic data because of the assumption of a fixed spread acquisition geometry for SSFWI. Among various field acquisition options, ocean bottom node (OBN) survey is closest to the fixed spread assumption. In practice, a typical OBN survey in field rolls the OBN spread due to the limited number of available nodes and for the operational efficiency, which impose additional challenges to apply SSFWI directly. In this paper, we will explain how to create fixed-spread OBN patches from a rolling OBN acquisition in the Gulf of Mexico for SSFWI. Then, we will demonstrate that SSFWI with multiple fixed-spread patches can provide a velocity model, which is almost equivalent with the velocity model by sequential source FWI (SeqFWI), with 10-20 times speed-up comparing with SeqFWI. This SSFWI workflow with OBN data could efficiently be applied to the compute-intensive FWI like broadband FWI and elastic FWI.

Full Text
Published version (Free)

Talk to us

Join us for a 30 min session where you can share your feedback and ask us any queries you have

Schedule a call