Online Journal for E&P Geoscientists

Sirat, Manhal ^*1; Walia, Samir ²; Al Daghar, Khadija A.¹; Al Mazrooei, Suhaila H.¹; Ramadan, Yasin ¹; Ali, Abduladim Z.¹; Morad, Sadoon ³
(1) EUFR, Abu Dhabi Company for Onshore Operations, Abu Dhabi, United Arab Emirates. (2) Geosciences, Emerson/ Roxar, Abu Dhabi, United Arab Emirates. (3) Geosciences, Petroleum Institute, Abu Dhabi, United Arab Emirates.

Characterization of fractured carbonate reservoirs aims to provide realistic static model, which can be used for simulation purposes in order to history match the production and to predict reservoir behavior in the near- and far future. This characterization is commonly achieved by integrating both extensive direct and indirect data sets, including seismic, well cores and logs, petrophysical and geomechanical data, drilling and production data and data taken from analog outcrops and regional maps. Most of these geological data are descriptive and static, and can therefore only be integrated into dynamic model by quantification approach. Hence, our ultimate objective of quantifying fracture data is to provide fracture porosity and permeability of various reservoirs within a giant hydrocarbon field in Abu Dhabi.

We used in this paper specific quantitative fracture attributes taken from the above tools and from analog outcrops such as fractures aperture, length and width, density and fracture-surface morphology, production and pressure data. Furthermore, we used basic assumptions of Geomechanics, paleostress and in-situ stresses to control the openness and closeness potential of each individual fracture set. Finally, we build a structural conceptual model for fracture network size and permeability that honors these observations and assumptions.

Based on stereographic contouring methods of poles to fracture planes, five vertical to sub-vertical fracture sets were identified and represent 90% of a population of ca 4000 fracture measurements. Three dominant sets striking N45W, N75W and N-S respectively represent 77% of the vertical to sub-vertical population compared to less dominant sets striking N30E and N60E, which together represent 23% of the vertical fracture sets. Fractures length, width, and density are controlled by bed thickness and distance to seismic faults, whereas aperture is controlled by mineralization and reactivation mode throughout geological time. Fracture sets have undergone several stages of reactivations, where three sets are currently prone to be opened whereas the other two sets are prone to be closed by the ambient stress. These finding have been incorporated into the dynamic model, which resulted in a considerable improvement of the history matching results.