We present 3D outcrop-constrained synthetic seismic models to investigate the chronostratigraphic significance of seismic reflections of a highly-progradational mixed carbonate-siliciclastic shelf margin. Ground-based LIDAR (Light-Detection and Ranging) was paired with field mapping to construct a 3D stratigraphic framework with a known chronostratigraphic relationship. Geostatistical approaches (Truncated Gaussian Simulation with Trend and stochastic object modeling) were then applied to distribute facies in 3D, while honoring outcrop mapped surfaces, measured section control and the depositional model. During impedance modeling, we modeled both a large-scale impedance trend, which was implicitly incorporated in facies definition, and the intra-facies impedance variations. Next, we simulated seismic models using the exploding-reflector algorithm to mimic wave propagation in 3D and then migrated to the depth domain for a comparison against the geologic model. We defined chronostratigraphic correlation error (CCE) as a means to evaluate the chronostratigraphic significance of a seismic event. The CCE concept was applied to the interpreted maximum flooding surface. Results show frequency decomposition at 60Hz before event-based picking would be helpful for seismic chronostratigraphic interpretation of analogous subsurface data. In addition, by co-rendering the CCE map with seismic, facies, and impedance cross-sections, we identified time-transgression-prone regions, and their respective geologic and petrophysical causes.
Presentation Date: Wednesday, October 19, 2016
Start Time: 1:30:00 PM
Presentation Type: ORAL