Data Integration for Fracture Model Characterization in a Middle East Carbonate Reservoir

Abstract

Abstract The work is about the challenge of modelling a complex carbonate reservoir, where the fractures network represents large part of the flow capacity. The case study is an offshore field in Ras Al Khaimah (RAK) Emirate developed by primary depletion. The objective is to integrate the fracture system identified by the 3D seismic attribute with the available well data, quite old and with only an image log, to obtain a good match of the production history. A regional assessment was performed to reconstruct the tectonic evolution of the study area and to identify the main structural domains. A final Discrete Fracture Network Model was obtained considering seismic and sub-seismic faults, through a proprietary workflow. The resulting fracture network, representing the larger scale fracture set, is not enough connected. An additional set of genetically similar fractures, at lower scale, was stochastically added to increase the intra field connectivity. Sensitivities were performed testing different ranges of the input parameters such as intensity and geometry, fitting the fracture power-law distribution. Multiple scenarios were verified within history match. The automatic lineaments features extracted were geometrical validated through a proprietary tool that allows to increase the signal/noise ratio and check the internal consistency of the lineament dataset. A calibration process, aimed to find the best parameters set-up, was performed, changing the input constraints matching the DFN with the few recorded wells mud losses. The results of the computation were the filtered set of consistent lineaments and a DFN composed by rectangular shape planes linked to them. An additional set of fractures, below the seismic resolution, was stochastically added to increase the system connectivity. Three different DFNs, with increasing number of discrete fractures, were obtained varying the fracture intensity driver. The fractures were then petrophysical characterized getting a Dual Porosity Dual Permeability model. A strong relationship between fracture intensity and well performance was observed, therefore the fracture petrophysical properties were straightforwardly tuned according to the available well test data. The DFN was able to capture the reservoir behavior in terms of gas rate, pressure and water break trough. Without the use of any numerical transmissibility barrier, the DFN allowed to reproduce the different depletion in the various areas of the reservoir. The approach proposed in this paper describes the workflow used in the company to estimate fractures contribution in a very complex environment integrating all the available data. Main challenges were represented by the description of the fractures and their permeability calibration. Simulation results confirmed the quality of the fractures characterization and observed flow capacity of the system. The adopted workflow can be helpful in similar fields with paucity of data, but long production life.