Towards a more robust joint migration inversion

Research output: Chapter in Book/Conference proceedings/Edited volumeConference contributionScientificpeer-review

Abstract

Conventional full waveform inversion has proven ineffective in recovering background velocity models for the targets out of the reach of refracted and diving waves. Therefore, different reflection waveform inversion (RWI) tools have been presented so far, one of which is joint migration inversion (JMI). JMI, unlike most similar tools, is parameterized by two classes of parameters, reflectivity and velocity, and equipped with decoupled imaging and tomographic sensitivity kernels. The fundamental characteristic shared by JMI or any RWI technique is the arrival-time consistency between the anchor parts of the modeled and observed data. This characteristic helps and keeps any reflection waveform tomography tool robust against cycle-skipping in near-offset data. In this paper, through examining the tomographic gradient of JMI, we show that conventional JMI suffers from cycle skipping in near-offset traces, which degenerates the fidelity of the tomographic gradient of JMI. Next, we demonstrate how the degeneracy can be lifted from two different perspectives, either by muting the misguiding tomographic wavepaths or migration isochrones built by near-offset residuals. Coupled with this, we show how excluding the cycle-skipped far-offset data significantly improves the validity of the tomographic gradient of JMI.

Original languageEnglish
Title of host publication83rd EAGE Conference and Exhibition 2022
PublisherEAGE
Pages1423-1427
ISBN (Electronic)978-171385931-4
DOIs
Publication statusPublished - 2022
Event83rd EAGE Conference and Exhibition 2022 - Madrid, Virtual, Spain
Duration: 6 Jun 20229 Jun 2022

Publication series

Name83rd EAGE Conference and Exhibition 2022
Volume2

Conference

Conference83rd EAGE Conference and Exhibition 2022
Country/TerritorySpain
CityMadrid, Virtual
Period6/06/229/06/22

Fingerprint

Dive into the research topics of 'Towards a more robust joint migration inversion'. Together they form a unique fingerprint.

Cite this