ACS PRF | ACS
All e-Annual Reports
44334-AC8
Modeling Natural Fracture Networks in the Context of Flow Simulations: Teapot Dome, Wyoming
Objective: The main objective of this research effort is to acquire an understanding of the role fracture and fault networks play in controlling oil production from the Tensleep Formation,Teapot Dome, Wyoming. The research relies on data contributed to the project by the Rocky Mountain Oil Testing Center. The center provided FMI logs, an integrated GeoGraphix data base and 3D Seismic from the field. Fracture orientation, intensity, and aperture measurements obtained from the FMI logs are used to simulate layered and composite 2D fracture networks. Analysis of post stack 3D seismic data from the producing area of the field is incorporated into fracture network characterization and, eventually, into flow simulations to gain insights into oil and water production history from the field. The initial efforts concentrated on evaluating the characteristics of the subsurface fracture networks? Fracture Systems: Schlumberger FMI log interpretations provide fracture strike and dip, bedding strike and dip, along with fracture classifications of open, healed, lithology bound or induced. Analysis of FMI logs indicates that open fracture systems in the Tensleep consist of a hinge-oblique set (N71W), a hinge-parallel (N18W) and northeast hinge-oblique (N32E) set with relative frequency of occurrence is 4.3:2.2:1, respectively. Although sampling is limited in the near-vertical boreholes it seems unlikely that the distribution of fracture orientations observed at depth could arise through random sampling of fracture sets with relative frequency similar to those observed at the surface by Cooper (2000). Cooper noted that the hinge-parallel and hinge perpendicular sets observed at the surface contain subsets that form acute angles with the vertical. If such were the case in the subsurface there would be increased probability of penetrating the hinge-parallel and hinge-perpendicular sets. The opposite is observed. Flow Simulations: The nature of fracture intensity, connectivity and directional variability in these attributes at various reservoir levels will have significant influence on production and on the strategic placement of wells during future in-field development and carbon sequestration initiatives. Through flow simulation we will attempt to define and clarify potential issues associated with these efforts. Our current plans are to use Schlumberger's Petrel software to develop fracture models for use in Eclipse flow simulations. Our initial efforts also evaluated the potential uses of the NETL codes FRACGEN and NFFLOW. Schwartz (2006) undertook numerous fracture simulations using FRACGEN which allows interactive adjustment of several parameters including clustering, aperture, length and spacing distributions. FRACGEN fracture models can be integrated directly into NFFLOW; however, the NFFLOW code currently does not incorporate a water drive which is an important reservoir parameter. Schlumberger's Eclipse simulator offers multiphase simulation suitable for modeling reservoir recovery using injected CO2. The Petrel-Eclipse interface also allows inclusion of discrete fracture systems into the flow simulation. Initial efforts with Petrel resulted in development of a multilayer model consisting of the Tensleep A and B sandstones and the intervening Dolomite B. Reservoir simulation in Petrel has a front end that is compatible with both FrontSim and Eclipse. FrontSim can now run dual porosity models and may serve as a starting point for flow simulations. A field scale fracture simulation (Figure 1) was developed from FMI log and 3D seismic interpretations. Conclusions: Fracture systems observed in the subsurface within the Tensleep differ considerably from those observed at the surface. Fracture sets within the Tensleep reservoir are dominated by NW hinge-oblique and NW hinge-parallel sets. Some open fractures are observe din the overlying Opeche Shale suggesting that seal integrity could be an issue in a future EOR or CO2 storage effort. Open fractures in the underlying Amsden Formation contain hinge-oblique and hinge-parallel sets similar to those in the Tensleep and likely facilitate ground water recharge. Several fracture models have been developed for implementation into flow simulations. The Year 2 effort will concentrate on undertaking and refining flow simulations for history matching to production from the Tensleep. Three abstracts were submitted for presentation this year: two for the AAPG annual meeting and one to the SPWLA. Bryan Schwartz completed his thesis titled “Fracture Pattern Characterization of the Tensleep Formation, Teapot Dome, Wyoming.” Acknowledgements: Appreciation is extended to several individuals working with the Rocky Mountain Oil Testing Center and National Energy Technology Center for their help with the project. Mark McKoy and Duane Smith were very helpful in getting this project started. Dr. Alan Brown (Schlumberger) provided Petrel software for use in the project and also provided our student, Valerie Smith, with a summer internship with Schlumberger where she was able to devote most of her time to learning Petrel and integrating log data and 3D seismic analysis from Teapot Dome. Early in the grant period, Brian Schwartz completed a thesis focusing on the analysis of subsurface fracture systems at Teapot Dome. Figure 1: Example fracture realization within the Tensleep. Depth converted model generated using Schlumberger's Petrel software (Smith, written communication, 2007).
