Discrete Fracture Network Modeling of a Giant Middle-East Fractured Carbonate Reservoir, the Bibi Hakimeh Field, SW Iran

Ole Petter Wennberg¹*, Mohammad Keramati², Mehran Azizzadeh², Frank Mengel¹, Kellfrid Lyslo¹, Daniel Berge Sollien¹, Claus von Winterfeld¹**
¹ Statoil ASA, 4035 Stavanger, Norway.
² RIPI, Tehran, Iran.
* [email protected]
** Present address: Petroleum Development Oman LLC, Muscat, Oman

The aim of this study was to establish an understanding of the distribution of open fractures in The Bibi Hakimeh Field, a giant Middle East fractured carbonate reservoir, and their influence on fluid flow behaviour. Further, the aim was to apply this understanding in realistic discrete fracture network (DFN) models, and to generate full field fracture parameters for the dual porosity fluid flow simulation model. The investigated field is located in the outer part of the Zagros Mountain Chain in the Dezful Embayment area in SW Iran. The field represents a classical asymmetric NW-SE trending Zagros anticline with a steep SW limb and a shallowly dipping NE limb. Main data sources in this study have been: core, bore-hole image, well productivity index (PI), PLT-logs, mudloss information, structural model and outcrop analogue data. 3D-seismic data have not been available. The high degree of flow heterogeneity in this field is shown by 3 orders of magnitude variation in the initial PI values. For modelling purposes fractures have been divided into fracture swarms and diffuse fracturing. Fracture swarms represent short intervals in a well, which are responsible for a major part of its hydrocarbon production. These features may represent clustered fractures, faults or single fractures with extreme apertures, and cut through large parts of the reservoir unit. Diffuse fracturing are the distributed minor fractures within the reservoir, and occurs in general as strata-bound fractures with a high angle to the bedding. 2D lateral variation in fracture density has been modeled by use of fracture density maps based on the structural model (dip and curvature) and on PI from wells. In the different full field DFN-model realisations, the fracture density maps have been combined with a 3D fracture facies model derived from well PLT-logs and mudloss information. Finally, the DFN-models were up-scaled to fracture parameters to be used in the dual porosity dynamic simulation. The effective fracture conductivities were defined and optimised in an iterative process with the reservoir simulation activity.

Figure 1