Passive seismic data Marchenko imaging: A framework for simultaneous coherent noise attenuation

View at publisher Summarise with VIRAT
Authors: Junming Zhang, Jinlong Yang, Zhongsheng Wang, Guangli Zhang
Cited by: 0
References: 8

Summary

No summary yet. VIRAT can read this paper and write a plain-language summary of its aim, methods, key findings and limitations.

Summarise with VIRAT Free to try with your Science Catalogs account.

Abstract

Marchenko imaging can obtain high-resolution subsurface imaging result at any locations through surface-recorded seismic data. In addition, it overcomes a critical limitation that conventional methods mainly rooted in the Born approximation—an assumption that seismic data only contain primary reflections. These capabilities position the Marchenko imaging as a potential solution to key challenges in passive seismic exploration where primary reflections are implicitly developed (often obscured by noise). However, the effectiveness of the Marchenko imaging is constrained by some fundamental assumptions, making it dependent on the quality of preprocessing. This issue is particularly pronounced in passive seismic exploration. The core step of Marchenko imaging involves using the propagation relationships between seismic records to obtain accurate Green’s functions. This process imposes stringent requirements on the purity of input data. However, the extensive noise present in passive seismic data, especially non-physical noise (introduced by seismic interferometry) and surface-related multiples, causing the inversion process to diverge and introducing artifacts into the results. To meet the wavefield compatibility criteria required by Marchenko method, we propose a simultaneous coherent noise attenuation framework that integrates a weak reflection signal extraction operator with an improved multiple suppression method. The operator, based on a modified collaborative filtering algorithm, leverages the distinct similarity characteristics between valid signals and coherent noise across different gather domains. By combining this operator with closed-loop surface-related multiple elimination (CLSRME), the proposed method achieves synchronous suppression of non-physical noise and free-surface multiples. Synthetic data test demonstrate that the proposed framework outperforms traditional approaches in both noise attenuation and signal preservation.

Fields of Study & Subjects

Seismic Imaging and Inversion Techniques Seismic Waves and Analysis Image and Signal Denoising Methods

Publication details

Journal / source
Second International Meeting for Applied Geoscience & Energy
Volume
1
Pages
2408–2412
Publication date
10 February 2026
Language
EN