Recursive Interferometric Surface-wave Suppression For Improved Reflection Imaging

Faezeh Shirmohammadi, Deyan Draganov, Ranajit Ghose, Eric Verschuur, Jan Thorbecke, Kees Wapenaar · 2024

Read Recursive Interferometric Surface-wave Suppression For Improved Reflection Imaging by Faezeh Shirmohammadi, Deyan Draganov, Ranajit Ghose, Eric Verschuur, Jan Thorbecke, Kees Wapenaar (2024) in the Tower of Babel, a free online library of public-domain works preserved in clean, searchable text. No sign-up. The full text is served as a plain-text file at a predictable URL, where text is present: https://antarctic-labs.com/tower-of-babel/arxiv-2411-02620.txt.

Description

Abstract High-resolution seismic reflections are essential for imaging and monitoring applications. In seismic land surveys using sources and receivers at the surface, surface waves often dominate, masking the reflections. In this study, we demonstrate the efficacy of a two-step procedure to suppress surface waves in an active-source reflection seismic dataset.

Catalog record

CREATORFaezeh Shirmohammadi, Deyan Draganov, Ranajit Ghose, Eric Verschuur, Jan Thorbecke, Kees Wapenaar
YEAR2024
COLLECTIONPAPERS
FORMATTXT
FILE SIZE39 KB
VERSION1.0
LICENSEhttp://creativecommons.org/licenses/by/4.0/
RIGHTSCC BY
SOURCEarXiv

Source: https://arxiv.org/abs/2411-02620

Browse the full library · About the Tower of Babel

Full text

The opening of the work, from the library's hosted text:

Recursive Interferometric Surface-wave Suppression For Improved Reflection Imaging Faezeh Shirmohammadi, Deyan Draganov, Ranajit Ghose, Eric Verschuur, Jan Thorbecke, Kees Wapenaar Abstract High-resolution seismic reflections are essential for imaging and monitoring applications. In seismic land surveys using sources and receivers at the surface, surface waves often dominate, masking the reflections. In this study, we demonstrate the efficacy of a two-step procedure to suppress surface waves in an active-source reflection seismic dataset. First, we apply seismic interferometry (SI) by cross-correlation, turning receivers into virtual sources to estimate the dominant surface waves. Then, we perform adaptive subtraction to minimise the difference between the surface waves in the original data and the result of SI. We propose a new approach where the initial suppression results are used for further iterations, followed by adaptive subtraction. This technique aims to enhance the efficacy of data-driven surface-wave suppression through an iterative process. We use a 2D seismic reflection dataset from Scheemda, situated in the Groningen province of the Netherlands, to illustrate the technique’s efficiency. A comparison between the data after recursive interferometric surface-wave suppression and the original data across time and frequency-wavenumber domains shows significant suppression of the surface waves, enhancing visualization of the reflections for following subsurface imaging and monitoring studies. Keywords : Data processing, Seismic Interferometry, Seismics, Surface waves, Groningen, Reflections. 1. Introduction Seismic data acquired on land is often contaminated by surface waves, which are a significant noise source for reflection seismic studies. The surface waves often have a velocity and frequency content similar to those of the investigated reflections, which makes it challenging to suppress them from the dataset. Conventionally, surface waves are suppressed during data processing using methods such as frequency-offset (f-x) ( Yilmaz, 2001a ), frequency-wavenumber (f-k), or bandpass filtering. However, these approaches can prove ineffective when surface waves are scattered and overlap with the frequency and moveout of the reflected body waves that we intend to preserve. The f-k filter may result in artefacts due to signal distortion and spatial correlation of the background noise because the surface-wave energy may be distributed over a significant range of the f-k spectrum ( Konstantaki et al., 2015a ), thus further lowering the quality of the reflections. Recently, the prediction of surface waves with seismic interferometry (SI) and their subsequent adaptive subtraction from seismic reflection data has emerged as a technique for suppressing surface waves ( Dong et al., 2006a ; Halliday et al., 2010a ). In SI, seismic observations from various receiver locations are, for example, cross-correlated to retrieve new seismic responses from virtual sources positioned at the receivers’ locations ( Lobkis & Weaver, 2001a ; Campillo & Paul, 2003a ; Wapenaar & Fokkema, 2006a ; Manen et al., 2006a ; Curtis et al., 2006a ). This process enables the retrieval of seismic responses between pairs of receivers. For suppression, the retrieved responses are then subtracted from the original field recordings using least-squares matching, resulting in data with suppressed surface waves. This suppression technique is usually called interferometric surface-wave suppression. In previous studies, the interferometric surface-wave suppression was applied in a non-recursive way on the data. Halliday et al. (2010a) demonstrated its effectiveness in the context of hydrocarbon exploration, while Konstantaki et al. (2015a) and Liu et al. (2018a) showcased its utility for near-surface applications. Moreover, Balestrini et al. (2020a) demonstrated its application for deep mineral explorations. Here, we propose a new approach using the first output of the interferometric surface-wave suppression for more iterations. We term this technique "Recursive Interferometric Surface-wave Suppression" (RISS). This technique aims to enhance the efficacy of the data-driven surface-wave suppression through an iterative procedure. In this study, we demonstrate RISS on a 2D reflection dataset acquired in Scheemda, Groningen province, the Netherlands. By using RISS, we aim to enhance the visualisation of reflections, which can provide clearer images of the subsurface structures and enhance the overall interpretation of the seismic data. Such advancements are particularly critical for Groningen, where gas production has resulted in induced seismicity since 1963 ( Muntendam-Bos et al., 2022a ). We evaluate the RISS results in comparison with those from other techniques such as time muting and f-k filtering. Below, we first present in Section 2 the methodology of RISS. This will be followed by a description of the seismic data acquisition in Section 3, the results in Section 4, and then a discussion and conclusions. 2. Methodology In our proposed approach, SI is employed first to retrieve the dominant surface waves. The retrieved surface-wave energy is subsequently adaptively subtracted from the dataset. Following this, the obtained data is utilised to iterate through these two steps, contributing to the improvement of the reflection resolution. This section outlines the implementation of RISS. 2.1 Surface-wave retrieval by seismic interferometry SI refers to the method of retrieving new seismic responses, for example between two receivers, using most commonly cross-correlation, and the result creates a virtual source at one of the receiver locations ( Wapenaar & Fokkema, 2006a ; Larose et al., 2006a ; Schuster & Zhou, 2006a ). In an active-source survey, this process is usually achieved by cross-correlating the recordings at two receivers and then stacking the individual virtual-source traces over all available active sources ( Halliday & Curtis, 2008a ). So, the retrieved virtual-source response between two receivers at positions \mathbf{x}_{A} and \mathbf{x}_{B} can be expressed in the time domain in its simplest form as: G(\mathbf{x}_{B},\mathbf{x}_{A},t)+G(\mathbf{x}_{B},\mathbf{x}_{A},-t)=\sum_{n=1}^{N}G(\mathbf{x}_{B},\mathbf{x}_{n},t)*G(\mathbf{x}_{A},\mathbf{x}_{n},-t), (1) where G(\mathbf{x}_{B},\mathbf{x}_{n},t) is the response of a recording at receiver \mathbf{x}_{B} and G(\mathbf{x}_{A},\mathbf{x}_{n},-t) is the time-reversed response of a recording at receiver \mathbf{x}_{A} , both from a source at \mathbf{x}_{n} at the Earth’s surface. The left-hand side of the equation represents the response and its time-reversal between the two receivers at \mathbf{x}_{A} and \mathbf{x}_{B} at the surface, implying that the receiver at \mathbf{x}_{A} has been turned into a virtual source. N represents the total number of active sources at the surface and * denotes convolution. In a laterally homogeneous 2D medium, sources at points in line with the receivers contribute to the retrieval of direct surface-wave arrivals since they are all in the stationary-phase region. So, the results retrieved by SI will be dominated by surface waves ( Balestrini et al., 2020a ). Figure 1 shows a schematic representation of SI for retrieving direct arrivals, including surface waves. By correlating the recording at \mathbf{x}_{B} from the active source at \mathbf{x} in Figure 1 a with a recording at \mathbf{x}_{A} in Figure 1 c, the virtual response between \mathbf{x}_{B} and \mathbf{x}_{A} is retrieved, as illustrated by the purple arrow in Figure 1 d, at causal times (the causal part refers to times later than the zero time). Similarly, the virtual response between another receiver at \mathbf{x}_{B^{\prime}} and a receiver at \mathbf{x}_{A} is retrieved by correlating the response at \mathbf{x}_{B^{\prime}} in Figure 1 b with that at \mathbf{x}_{A} in Figure 1 c, as depicted by the orange arrow in Figure 1 d, at acausal times (the acausal part refers to times earlier than the zero time). In both cases, the receiver at \mathbf{x}_{A} acts as a virtual source as shown by the blue explosion in Figure 1 d. We repeat this procedure for all active sources, e.g., as shown in Figure 1 e for another active source at

Read the complete text