This paper describes a workflow that fully utilizes the pre-stack and post-stack seismic attributes to derive reliable geologic and fracture models that are validated by multiple blind wells and reservoir simulation. The first step in the workflow is to run post-stack seismic processes, which includes post-stack inversion and spectral imaging. The second step consists of applying a pre-stack seismic process called elastic inversion which will lead to various key seismic properties that provide good discrimination between reservoir facies. The third step consists of using the various post-stack and pre-stack seismic cubes to derive 3D geologic and fracture models. The fourth step is to use the derived models in a reservoir simulator to verify the validity of the models. This workflow was applied to a complex fractured carbonate field in offshore Tunisia which produces oil from the El Gueria formation. A large number of post-stack and pre-stack seismic attributes were generated in time and then depth converted within a 3D geocellular grid. These seismic attributes were used as input in REFRACT TM , Prism Seismic fracture modeling software, to create geologic and fracture models. The resulting porosity and permeability models were put into Eclipse TM reservoir simulator software. Individual well performances were matched at eighty percent of the wells, confirming the reliability and accuracy of the derived geologic and fracture models and the usefulness of the workflow.

Use of sector models with fine grids that preserve the boundary conditions of the full field model has been of particular benefit to studying well coning behaviour for the different well geometries while allowing detail studies of the physics of flow and to optimize production rate by different well designs. The objective of this project was to carry out simulation studies to investigate the pattern of gas coning and water encroachment for a bilateral well with the primary aim of producing oil from a reservoir overlain by a large gas cap in Field A. High precision local refinement studies in the simulation model were undertaken to help place the wells and optimize the completion design at the same time capturing the global field behaviour. This methodology was also used to properly simulate multilateral wells containing inflow control devices, allowing for pressure losses along the wellbore to be equalized and to minimize gas and water coning. Prior to undertaking the simulation studies, several sensitivities were carried out to determine how other parameters such as boundary conditions and grid refinements could affect the output of near well bore models. By taking advantage of the time savings resulting from the generation of reduced fine grid models, several simulations were run to investigate the impact of different well configurations and operations due for instance to close/opening of valves or laterals. The simulation studies resulted in the determination of the pattern of gas coning, water encroachment, optimum vertical placement of the oil lateral and the orientation of the gas lateral as they affect total recovery. The use of Inflow Control Device (ICD) was determined to be of benefit especially in controlling water and gas influx while providing a uniform production profile along the wellbore that delay gas and water coning and this is being incorporated now in the plan of development. Introduction These simulation studies are part of an ongoing reservoir development project for a gas/oil field with an oil rim about 40m thick and large gas cap. The objective of this study was to carry out several sensitivity studies so as to optimize the production of oil using multilateral well and smart completions, while accounting for the uncertainties in developing the model. The main uncertainties that have been identified while developing the static model include determination of the exact fluid contacts GOC and GWC, depth conversion methods, porosity and permeability distribution, which are directly related to facies distribution. Uncertainties in the dynamic model include permeability distribution, relative permeability data, impact of fractures, direction and density, aquifer size and connectivity, and the transmissibility in the z-direction. The main focus of this project was to try to model capture the effects of some of these uncertainties and how they impact the production of oil from the well. The multilateral wells are bilateral with the top lateral drilled through the gas cap with the primary aim of producing gas which will serve as the means for meeting the gas requirements for a natural gas lift system and gas producers when oil production finishes. The well completion is such as to make the gas available at the main bore and at the appropriate depth to lighten the liquid column with the adequate adjustment of the inlet control valves if water breakthrough fractures.

Ashtart, a large fracture enhanced carbonate oil field offshore Tunisia, required a new reservoir model with high prediction confidence to optimize tail end production and to evaluate its remaining upside potential requiring drilling or tertiary recovery methods. Due to the complexity of the field, a synergistic approach was adopted by teaming up the different disciplines and partners.This approach entailed the creation of a fine scaled structural and petrophysical 3D matrix geomodel. The model was subsequently scaled up and supplemented with observed dynamically acting features such as fault or fracture flow paths and flow barriers. Modeled matrix permeabilities were systematically adjusted to match established well productivities, without arbitrarily changing flow conditions in the vicinity of wells.Crucial to this process were not only thorough studies of fracture distributions and attributes, but also the integration of all observed fluid dynamic aspects. Pressure transient analyses proved to be a valuable tool, leading to consistent effective permeabilities and the identification of previously unknown lateral and vertical boundaries. Superior history matches following iterative numerical modeling demonstrate the success of this single permeability modeling approach with improved prediction capabilities.

No abstract

Ashtart, a large fracture enhanced carbonate oil field offshore Tunisia, required a new reservoir model with high prediction confidence to optimize tail end production and to evaluate its remaining upside potential requiring drilling or tertiary recovery methods. Due to the complexity of the field, a synergistic approach was adopted by teaming up the different disciplines and partners.This approach entailed the creation of a fine scaled structural and petrophysical 3D matrix geomodel. The model was subsequently scaled up and supplemented with observed dynamically acting features such as fault or fracture flow paths and flow barriers. Modeled matrix permeabilities were systematically adjusted to match established well productivities, without arbitrarily changing flow conditions in the vicinity of wells.Crucial to this process were not only thorough studies of fracture distributions and attributes, but also the integration of all observed fluid dynamic aspects. Pressure transient analyses proved to be a valuable tool, leading to consistent effective permeabilities and the identification of previously unknown lateral and vertical boundaries. Superior history matches following iterative numerical modeling demonstrate the success of this single permeability modeling approach with improved prediction capabilities.