Print this pageImproving Subsalt Imaging Using 3D Basin Model Derived Velocities
Stephan PETMECKY1, Martin L. ALBERTIN2, and Nick BURKE2
1BP Exploration, Sunbury on Thames, Surrey, UK
2BP Exploration, Houston, Texas
The quality of depth imaging is directly related to the accuracy of the underlying velocity model. In most subsalt settings, lack of angular illumination severely degrades the resolution and accuracy of velocity information derived from the seismic data itself. A standard approach for building a starting velocity model in this difficult subsalt setting uses more reliable velocity information outboard of salt which is subsequently extrapolated to populate the subsalt regions. The shortcoming of this method lies in the assumption that the effective stress observed outboard of salt can be extrapolated beneath salt solely as a function of depth below mudline.
A more suitable approach for building the initial subsalt velocity model uses constrained estimates of effective stress to empirically derive subsalt velocities. Several techniques can be used, ranging in complexity from perturbed models based on the extrapolated starting velocity model to full 3D basin simulations (Albertin et.al., 2006). The applicability of each technology depends on how well the subsalt geology is known as well as on time constraints related to each project.
Albertin et al., 2002, already showed that velocities derived from basin model effective stress data show excellent large-scale agreement to seismic velocities. 3D basin modeling represents the most rigorous approach for determining subsalt effective stresses because it accounts for depositional history, sand distribution and connectivity, variable shale properties, timing of salt emplacement, etc. - all of which influence the magnitude and distribution of abnormal pore pressure. In the present case, numerical 3D basin modeling was used to predict pressures for an exploration well in the Central Gulf of Mexico (Mississippi Canyon protraction area). A calibrated present day effective stress cube from a basin model was used to calculate subsalt velocities. The comparison between basin modeling derived velocities and the original velocity field showed remarkable differences, particularly in that the former indicated a far more complex subsalt velocity distribution than the latter. Next, basin model sediment velocities were utilized to re-migrate the seismic data set, resulting in a more detailed image as well as changed source rock geometries. Where the original image showed a deep defocus at the source rock level, the new geometry would focus hydrocarbons into the identified trap. Based on the results of the subsequently drilled exploration well, it has to be assumed that the re-migrated data, aided by basin modeling derived velocities, provide a more accurate image of the subsalt geology.
Depending on the quality of geologic information available, basin models can be calibrated to produce good velocity matches outboard of salt. Therefore, it is reasonable to expect that basin models can provide useful stress estimates for determining velocity models for subsalt imaging, as demonstrated here. However, this only holds true in the absence of major unresolved geologic features which could significantly alter the expected overpressure distribution beneath salt.
References
Albertin, M.L., Burke, N., Petmecky, S., Etgen, J., Billette, F., 2006, Effective Stress Modeling Techniques for Estimating Subsalt Velocities: 2006 SEG Summer Research Workshop on Geopressure, Abstracts.
AAPG Search and Discover Article #90066©2007 AAPG Hedberg Conference, The Hague, The Netherlands