ABSTRACT

Sequence stratigraphy in the seismic interpretation workflow helps in the understanding of the basin analysis and the spatial distribution of reservoirs, seal and source rocks. Classical methods consist in identifying seismic discontinuities corresponding to the reflection terminations (downlap, toplap, onlap, and truncation) to define stratigraphic sequences and their system tracts. Such task is a labour intensive process mainly based on a limited number of auto-tracked horizons. Recently new approaches have been proposed to optimize this workflow. In this paper, we propose to analyze the thickness variations of a geological model, computed with a global approach based on a minimization process between the seismic relationships (Pauget et al 2009). Given the fact, the geological model is continous; the variation of the thickness can be computed for any seismic voxel. The thinning zones of the geological model enhance stratigraphic discontinuities and provides to the interpreter a high level of precision in the identification of the sequences. We have applied this method on the block F3, located in the Dutch sector of the North Sea, presenting relevant large-scale sigmoidal bedding. The analysis of the thickness attribute enhanced zones of convergence of the seismic reflections packages corresponding to the observed downlaps and toplaps. A sub division into stratigraphic sequences could be achieved by mapping and thresholding the thickness values. Convergence zones of the different reflection packages were modeled in three dimensions for a better understanding of the spatial depositional process. This case study has shown the rapidity, robustness and the accuracy of the geo-model approach in the analysis of the stratigraphic sequences. These results suggest that the method could be used to optimize the level of detection of subtle traps, seals and reservoirs, at an early stage in the interpretation process.

This content is only available via PDF.
You can access this article if you purchase or spend a download.