Seismic chronostratigraphy for reservoir characterization : modeling and applications

Date

2017-05

Authors

He, Yawen

Journal Title

Journal ISSN

Volume Title

Publisher

Abstract

The assumption of the chronostratigraphic significance of seismic reflections serves as a fundamental premise in interpreting stratigraphy from seismic images. This hypothesis proposed in 1977 was initially applied to delineate depositional sequences as the basic interpretive unit, and then to reconstruct Wheeler Diagram and regional sea level curves. After a further comparison against with global eustatic events, these regional curves can further facilitate predicting the age, distribution, and facies of depositional sequence before drilling in a seismic-covered area during petroleum exploration. With a boom in reservoir-level seismic applications, for obtaining significant high frequency sequence (HFS) surfaces as the bounding surfaces in static reservoir model construction, this fundamental assumption was inevitably extended to characterize HFS and even high-frequency cycles (HFC) during seismic reservoir characterization.

For an ultimate improvement in constructing reservoir-bounding surfaces, the author targeted at evaluating the validity of this fundamental assumption as applied to high-order seismic stratigraphy. The author conducted the entire project via the forward seismic modeling upon geologic models with known chronostratigraphic relationship. Besides, these input models carefully honor the reservoir geology for meaningful discussions on (1) shallow marine siliciclastic reservoirs in Starfak Field, GoM, (2) shallow-water mixed carbonate/clastic Upper San Andres-Grayburg reservoirs in Permian Basin, and (3) shallow-water carbonate Abo shelf margin-Clear Fork platform in Permian Basin.

This study has achieved three-fold contributions. On the aspect of realistic geocellular, property and seismic modeling at the reservoir scale, the author integrated high-resolution sequence stratigraphic framework, published 3D depositional model, intra-facies heterogeneity in 3D modeling to selectively apply advanced geostatistical methods to model hierarchical heterogeneity. Subsequently, the author proposed an evaluation scheme with a defined parameter ('time-correlation error/TCE') to assess HFS-scale reservoir-bounding surfaces. These assessments revealed an interactive influence from (1) stratal geometry, (2) lateral lithofacies variation, (3) lithofacies-sonic velocity relationship in pure- versus mixed-lithology successions, (4) intra-facies heterogeneity, and (5) seismic frequency. Finally, based on these forward modeling results, the author proposed a decision tree to determine valid interpretation strategy in seismic chronostratigraphic correlation in scenarios with geoscientists’ expert knowledge and recommended an attribute-driven volumetric picking scheme to improve published algorithms for scenarios without prior knowledge.

Description

LCSH Subject Headings

Citation