Wide-azimuth OBC Seismic Data Optimization for AVO and AVAz Analysis, Offshore Abu Dhabi, UAE

Abstract

The lack of high-resolution subsurface images from poor seismic imaging quality leads to inaccurate AVO/AVAz analysis and fault interpretation, which are critical for reserves estimation and de-risking any imminent drilling decisions. In a developing filed, acquiring and processing a new seismic data often falls outside the time frame of ongoing field development, as it will require great efforts in overcoming logistics challenges along with additional costs. In this case study, in offshore Abu Dhabi, revisiting the vintage data with careful and detailed reprocessing whilst utilizing the latest technologies has proven to be an effective, practical and cost-efficient method in improving the fault resolution and reservoir characterization. In this case study, it is observed that the deeper events in the vintage data were masked by the strong surface wave energy. The irregular acquisition geometry of the seismic data caused the aliasing of the surface wave. The application of harsh de-noising techniques in the vintage processing further deteriorated the fault definition. Thus, to tackle the aliasing problem, 5D trace densification and regularization was applied to increase the input data and create a de-aliased surface wave model. This allowed for subsequent subtraction of the strong surface wave, without damaging the body wave. Further cascaded surface wave attenuation algorithm improved the image quality. Modern fault preserving residual noise attenuation was applied along with 5D Fourier reconstruction mitigated the residual noise content. It has been proven in the case study that multi-dimensional data densification and 5D reconstruction of the signal enhanced the fault delineation. By leveraging the modern signal-processing innovations, the final results produced a better overall reflection image focused on the angle/azimuth stacks suitable for fault interpretation, AVO and AVAz analysis. In conclusion, poorly vintage seismic data has been shown to possess a high value despite its irregular geometry and low resolution.