Regularization with differential evolution-based common-offset common-reflection surface: A case study for field onshore seismic data

We introduce a novel approach for seismic pre-stack data regularization using the common-offset common-reflection-surface method with a global attribute search strategy, employing a bio-inspired differential evolution optimization algorithm. We compare the global common-offset common-reflection-surface approach with the conventional sequential attribute search, focusing on their performance in pre-stack data regularization. Tests on synthetic and onshore (field) seismic datasets demonstrate that the global search approach significantly improves performance, enhancing signal-to-noise ratio and coherently filling missing traces. We show that the global common-offset common-reflection-surface method effectively addresses spatial irregularities, reconstructing reflections without artefacts, filling data gaps and highlighting geological details even in complex areas. In contrast, the sequential common-offset common-reflection-surface method, while capable of reconstructing missing traces, shows lower interpolation quality and fails to adequately highlight complex geological features such as high-dip reflections.