Resolution-oriented weighted stacking based on global optimization algorithm

Stacking is one crucial seismic data processing technique that gives a composite record made by combining traces from different shot records. The quality of stacking dramatical affects the performance of many seismic data processing tasks. The conventional equal-weight stacking method is the average of all traces in the pre-stack CMP gather, improving the signal-to-noise ratio (SNR) but reducing resolution. Most weighted stacking algorithms aim to enhance image quality by the increased SNR; however, these algorithms do not consider the resolution. Therefore, we proposed a weighted stacking algorithm with resolution enhancement, which is regarded as having maximum bandwidth and dominant frequency. Based on the genetic algorithm (GA), we calculated the stacking weights in common midpoint (CMP), or common-reflection-point (CRP) gathers. Then, we presented a weighted stacking approach to obtain the resolution-enhancement stacked data. The proposed method can obtain the resolution-enhancement stacked data by the single-trace spectrum constraint without wavelet estimation. Applications to synthetic and field seismic datasets demonstrate that compared with the traditional stacking method, the proposed method can improve the stacking resolution better, which is beneficial for subsequent interpretation.

---

Opracowanie rekordu ze środków MNiSW, umowa nr SONP/SP/546092/2022 w ramach programu "Społeczna odpowiedzialność nauki" - moduł: Popularyzacja nauki i promocja sportu (2024).