After decades of research on stochastic reservoir characterization, mapping of average properties and representation of heterogeneities are in frequent use in the petroleum companies. Methodologies for data integration and uncertainty ar e also available. Focus of the current project is on identification of high-contrast spatial features in the reservoir formation. Each feature unit will typically occupy a small volume of the formation under study, although the number of features may be l arge. The features of interest have extreme properties and spatial continuity which may entail strong impact on hydrocarbon depletion.
The research will focus on:
- Identification of thin, laterally extensive beds of shale or calsite, or of
high-perm eability sediments, which have high impact on fluid flow.
- Identification of pockets of high-residual hydrocarbon left after depletion, which
increase the recovery potential.
The methodology will also be relevant for other challenges related to hi gh-contrast spatial features in reservoir evaluation:
- Identification of hydrocarbon accumulation in geological formations during
- Identification of fault zones in reservoir characterization.
- Identification of high-pressure pockets and extreme-compact beds during well
All these geoscience features have several characteristics in common:
- relatively small volume of each feature unit
- spatial features being thin with relatively large areal extent
- high-contrast p roperties of features relative to background
- hard to identify from one data type alone
- not target of traditional characterization which focus on average properties
The solutions will be sought in a Bayesian inversion framework.