--> Three Rock-Typing Methods and Implementation as Part of the Reservoir Characterization and Uncertainty Assessment: An Example from the Arab Formation (Upper Jurassic), Onshore Field United Arab Emirates

AAPG Middle East Region Geoscience Technology Workshop:
3rd Edition Carbonate Reservoirs of the Middle East

Datapages, Inc.Print this page

Three Rock-Typing Methods and Implementation as Part of the Reservoir Characterization and Uncertainty Assessment: An Example from the Arab Formation (Upper Jurassic), Onshore Field United Arab Emirates

Abstract

The paper explores some rock-typing approaches to characterize the reservoir quality in the Arab formation in onshore field of the UAE. The analysis aims to capture the heterogeneity of the reservoir, lateral continuity and link to the sedimentary and diagenetic settings. The data base used were the core analysis (RCA and SCAL), slab and thin section description and well logs. That information was assembled/integrated employing different Rock-typing approaches defined. The main focus was in the upper section of the Arab Formation dominated by dolomitic limestone intercalated with anhydrites. In most of the cases, the precursor rock fabric was preserved or at least interpreted from the thin sections. However, the diagenesis was important enough to have a strong overprint on the rock-quality of the reservoir and needed to be considered. Being the data concentrated in the crest of the structure, the challenge was to link the rock-typing to pre-conditioned sedimentary setting that once defined, it is expected to be control the 3D distribution of the rock-types in the reservoir model. The approaches have in common two main stages, the 1D modeling (at well level) and the 3D extrapolation. It has been considered that in the 1D modeling, rock-type definition, goes in to three layers of analysis: cored, uncored wells and an integration layer. Basically, the 1D models (rock-typing approaches) were calibrated with the core data to be able to calculate the rock-types in the uncored wells. Three main approaches were used: (1) Lucia’s (1995-2007), (2) PC-Types: FZI iso lines/classes or GHE (Cortez and Corbett, 2005) combined with MICP data families which it is called in this paper PC-Types, and (3) Lithotypes, based on the lithological description which represents a more genetic approach. Lucia’s method explores the textural aspects of the rock and aims to translate it into a RFN class that links Poro/Perm transforms and SW estimations to the texture of the rock. The PC-Types on the other hand, based its rock-type classes according to the families of SW-height curves and pore throat distributions. If they are transform into J-functions, a derivation of PC-Types is then linked to porosity/permeability relationships (GHE-classes), in which case a correspondence analysis is performed between the PC-Types and the GHE classes. Finally, the Lithotypes explore the lithology classes identified in the core description, partitioned in different categories. They were extrapolated to the uncored wells using different multivariable techniques (e. i. NN and Cluster Analysis). Each Lithotype has a corresponding poro/perm model and SW estimation functions calibrated with MICP data. The final resulting rock-type models will use the poro/perm relationships and SW-H functions defined in 1D modeling stage. They represent scenarios that are carry on in the 3D modeling and uncertainty analysis. The link between the rock-types and conceptual sedimentary model will allow a more realistic extrapolation of the rock-types beyond well control leading to more consistent 3D rock-type models and as per as consequence a more robust 3D property models linked to Static Rock Types (SRT) were classified through distinctive sets of geologic and petrophysical groups. This classification resulted in four SRTs. SRT 1 exhibiting enhanced reservoir properties product of early diagenesis, SRT2 is dominated by neutral diagenetic processes that preserve reservoir properties, SRT3 and SRT4 are both associated to late diagenetic property reducing processes that distort arrangement of minerals and pore structure. The major achievement of this rock-typing approach resumed in the integration of the Geology and Petrophysics. This integration enable finding significant evidence to understand reservoir properties at depositional stage, properties alteration product of diagenetic processes and reservoir dynamic behavior links to a geologic concept. This rock-typing approach changes the traditional practice, formerly used to model this particular reservoir, which was limited only to the classification of petrophysical patterns, instead, this approach allows associating a particular petrophysical pattern to a singular geologic facies, feature or event. Ultimately, via the integration of dynamic and static data, reservoir models become more predictive. Similarly, the basis of the rock-typing approach presented herein brings together a solid static understanding in order to delineate the origin of particular reservoir dynamic behaviors. This fit-for- purpose approach built from the premise of integration provides a complete basis for reservoir simulation, management, and forecasting, and at the same time contributes reducing reservoir uncertainties by means of enhancing heterogeneities predictability, dynamic flow understanding, which all combined yields organically into optimized field development strategies.