ASSESSMENT OF AZIMUTHAL VARIATIONS IN DEPTH MIGRATED MARINE WIDE-AZIMUTH DATA: WORKFLOW DERIVATION AND EXAMPLES FROM THE EAST BREAKS REGION OF THE GULF OF MEXICO

Current seismic acquisition geometries are designed to record a broader range of azimuths. While illumination of subsalt structures is a major factor when planning these surveys, azimuthal variations due to heterogeneity and anisotropy can also significantly impact the imaging of these targets if they are not well understood. This study presents a practical workflow for detecting and assessing the significance of azimuthal variations on wide-azimuth data, in an effort to start to better utilize the azimuthal aspect of this type of survey. The assessment itself was performed in depth migrated images of wide-azimuth data from the East Breaks region of the Gulf of Mexico, which were decomposed into azimuth-sectored angle-domain gathers post-imaging. This data domain provided great flexibility to the analysis of azimuthal moveout anomalies, as different subsets of common angle volumes were evaluated depending on their level of contributions to the imaging of a given geologic target. As the derived workflow was tested in different scenarios, it became clear that analyzing a variety of attributes, volumetrically or at specific interpreted horizons, delivered qualitative results that confirmed the presence of azimuthal variations and how irregular or chaotic these were, depending on the area of interest. Throughout this study, it was quite important to develop quality control plots that aid the validation of the propose techniques. Particularly significant were the cross-correlation azimuthal residual maps and the attributes used to rule out whether an azimuthal anomaly was being induced by the migration models, since improving the fitting of the data by using a higher order of symmetry system was outside of the scope for this study. It was also concluded that while the resolution of the dataset used in this study was typical for depth migration images in the region, it was not ideal in terms of frequency and azimuthal sampling. However, this analysis did provide multiple insights into the intricacies of assessing azimuthal variations in image space.