Elastodynamic Marchenko Green's function retrieval from two-sided reflection and transmission data

J.R. van der Neut, J. Brackenhoff, Giovanni Angelo Meles, E.C. Slob, C.P.A. Wapenaar

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

4 Downloads (Pure)

Abstract

Green’s functions in an unknown elastic layered medium can be retrieved from single-sided reflection data by solving a Marchenko equation. This methodology requires a priori knowledge of all forward-scattered (non-converted and converted) waveforms. Moreover, the medium should satisfy stringent monotonicity conditions, which are often not met in realistic scenarios. In this contribution, we show that the situation is significantly less cumbersome if two-sided reflection and transmission data are recorded (for instance in laboratory settings). A novel methodology is presented to retrieve elastodynamic Green’s functions from such data. Apart from the two-sided reflection and transmission responses, our methodology requires knowledge of the direct non-converted PP- and SS-transmissions (a priori knowledge of forward-scattered converted waveforms is not needed). We demonstrate the success of our methodology by conducting a numerical experiment in an elastic layered medium that violates the monotonicity conditions of the Marchenko equation for single-sided reflection data. The limitations of the methodology and the sensitivity to errors in our initial estimates require further investigation.
Original languageEnglish
Title of host publication83rd annual EAGE meeting
Number of pages5
DOIs
Publication statusPublished - 2022
Event83rd annual EAGE meeting - online meeting, madrid, Spain
Duration: 6 Jun 20219 Jun 2021
Conference number: 83rd

Conference

Conference83rd annual EAGE meeting
Country/TerritorySpain
Citymadrid
Period6/06/219/06/21

Fingerprint

Dive into the research topics of 'Elastodynamic Marchenko Green's function retrieval from two-sided reflection and transmission data'. Together they form a unique fingerprint.

Cite this