Assessment and Modeling of High Permeability Areas in Carbonate Reservoirs

Abstract

Abstract In carbonate reservoirs, heterogeneity is usually driven both by the depositional and the diagenetic patterns. When original reservoir properties have been strongly modified by dissolution, the assessment and the modeling of karstic high permeability areas constitute key issues during reservoir appraisal. In this paper, a very innovative methodology is proposed for the evaluation and modeling of reservoirs in which dissolution occurred at an early stage, and was controlled by sequence stratigraphy. Using stochastic models of depositional facies built using Neptune (a modeling methodology/tool that uses cubes of facies proportions derived from paleobathymetry), the probability for facies to be affected by early karstification is evaluated. This can be possible thanks to the 3D grid of paleobathymetry previously computed with Neptune. Then, the distinction is made between the regions potentially exposed to dissolution, and the regions with no chance to be exposed. The last stage consists in simulating petrophysics according to the depositional facies and its diagenetic behavior. Such a methodology enables to take into account the 4D aspect of the reservoir construction. The prediction and the management of the areas of emersion are possible, and contribute to obtain significant advances in the realistic aspect of the 3D distribution of reservoir properties in stochastic models. Moreover, consistency between sedimentology, diagenesis, and petrophysics is input at a very high level in the model. This makes possible to preserve the key heterogeneity in reservoir models for fluids flow simulation. Application to an oil field in Middle East illustrates the methodology. Introduction Building 3D reservoir models with respect to geology constitutes a recurrent challenge in Oil &Gas Industry. Stochastic simulation techniques are usually required to face this problem, through the simulation of depositional facies prior to the simulation of petrophysics. However, when residual reservoir properties result from complex multi-stage geological processes, the realistic extrapolation of scarse well data becomes an unreachable quest. Carbonate reservoirs generally deal with such a description, as far as diagenesis significantly overprints depositional features. The way to render very realistic images of reservoir heterogeneity is to model successively the geological phenomena that drive 3D distribution of petrophysics. For depositional heterogeneity, a new methodology has been developed, that enables to compute cubes of facies proportions consistently with sequence stratigraphy, the depositional model, and the experimental data. This can be possible, thanks to a prior modeling of accommodation potential (increment of available space for sedimentation), and of paleobathymetry of sedimentation. Therefore, Neptune methodology provides models of depositional facies constructed on real stratigraphic grids, which reproduce in the whole field the progradational or transgressive features recorded at wells. However, in carbonates, there is no bi-univoque relationship between the depositional facies and the rock-type. This is caused by the effects of diagenesis that strongly modify the initial reservoir properties. Consequently, a valid approach for carbonate reservoir modeling requires the ability of modeling diagenetic effects. This can be possible when the diagnostic concerning the origin and mechanism of diagenesis has been made by sedimentologists, and when guidelines do exist for extrapolating diagenetic effects. These conditions are respected when early dissolution occurs during the construction of the reservoir : the 3D distribution of epikarstic zones is then driven by sequence stratigraphy. A new methodology is needed to address the modeling of those reservoirs in which karstic features constitute the key heterogeneity that drives fluid flows. This paper describes this methodology, derived from Neptune, for modeling high-perm layers in such reservoirs, which are frequent in Peritethyan carbonates.