Prestack seismic data enhancement with partial common-reflection-surface (CRS) stack

Abstract

We developed a new partial common-reflection-surface (CRS) stacking method to enhance the quality of sparse low-fold seismic data. For this purpose, we use kinematic wavefield attributes computed during the automatic CRS stack. We apply a multiparameter CRS traveltime formula to compute partial stacked CRS supergathers. Our algorithm allows us to generate NMO-uncorrected gathers without the application of inverse NMO/DMO. Gathers obtained by this approach are regularized and have better signal-to-noise ratio compared with original common-midpoint gathers. Instead of the original data, these improved prestack data can be used in many conventional processing steps, e.g., velocity analysis or prestack depth migration, providing enhanced images and better quality control. We verified the method on 2D synthetic data and applied it to low-fold land data from northern Germany. The synthetic examples show the robustness of the partial CRS stack in the presence of noise. Sparse land data became regularized, and the signal-to-noise ratio of the seismograms increased as a result of the partial CRS stack. Prestack depth migration of the generated partially stacked CRS supergathers produced significantly improved common-image gathers as well as depth-migrated sections.

Enhancement of prestack diffraction data and attributes using a traveltime decomposition approach

“…The dataset contains a total of 401 CMPs and both the lateral extension of the seismic line and the maximum offset are 5 km. Figure 5 compares various finite-offset semblance sections ranging from 500 m to 4000 m offset provided by FO prediction and the zero-offset-based partial CRS method (Baykulov and Gajewski, 2009). The results clearly reveal that the zero-offset operators used by partial CRS (Fig.…”

“…As in the previous examples, we applied partial CRS (Baykulov and Gajewski, 2009) and the new diffraction traveltime decomposition approach (FO prediction) to the data. The previous zero-offset CRS processing was carried out using a global optimization scheme for the attribute search (Walda and Gajewski, 2015) which accounted for a maximum of nine conflicting events at each sample.…”

“…Based on the zero-offset CRS stack, Baykulov and Gajewski (2009) recently introduced the so-called partial CRS stack. In this method, finite-offset traces are simulated by application of zero-offset operators for local finite-offset stacks, which corresponds to the extrapolation of traveltimes from zero-offset.…”

“…Based on the redundancy of zero-offset and finite-offset information for diffractions, this new approach allows the direct prediction of finite-offset diffraction attributes from zero-offset CRS processing results (Schwarz et al, 2014a;Bauer, 2014). In this fashion, the full prestack data volume is accurately characterized without carrying out computationally expensive generic finite-offset CRS processing (Zhang et al, 2001) or partial CRS stacking (Baykulov and Gajewski, 2009), where finite-offset traveltimes are extrapolated from zero-offset results.…”

Passive seismic source localization via common-reflection-surface attributes

“…Geod., 60 (2016) neighboring midpoints are jointly described by a traveltime surface (Jäger et al, 2001). While a decrease in the noise level is a virtue on its own, the estimated CRS operators can also be used for efficient prestack data enhancement, like trace interpolation or regularization (Baykulov and Gajewski, 2009). In addition, the CRS traveltime moveout expression is parameterized in terms of kinematic attributes of fictitious one-way propagation experiments (Hubral, 1983), which can be utilized for various applications, ranging from multiple suppression (Dümmong and Gajewski, 2008), diffraction separation (Dell and Gajewski, 2011), geometrical spreading compensation (Hubral, 1983), Fresnel zone estimation for migration to advanced depth velocity model building via NIP tomography (Duveneck, 2004).…”