Modelling of Fractured Granitic Geothermal Reservoirs: Use of Deterministic and stochastic methods in Discrete Fracture Networks and a Coupled Processes Modeling Framework
Authors/Creators
- 1. B2R, Institut Polytechnique UniLaSalle Beauvais, Geosciences Department, 19 Rue Pierre Waguet, F-60026 Beauvais, France
- 2. Division of Hydrologic Sciences, Desert Research Institute, 2215 Raggio Parkway, Reno, NV 89512, United States
- 3. Technische Universität Darmstadt, Institute of Applied Geosciences, Schnittspahnstraße 9, D-64287 Darmstadt, Germany
Description
The MEET project (Multidisciplinary and multi-context demonstration of EGS exploration and Exploitation Techniques and potentials) is an interdisciplinary study aiming at the improvement of Enhanced Geothermal System (EGS) technology across Europe.
The present work contributes to the MEET project in the form of a 3D structural model and Discrete Fracture Network (DFN) analysis, coupled with multi-physics models along key transects, of an analog for the Soultz-sous-Forêts reservoir (Upper Rhine Graben,
France). The Noble Hills (Death Valley CA, USA) are chosen as analog due to the common transtensional tectonic setting (rifting zone), an arid region of near total outcrop exposure and making it ideal for Structure from photogrammetry images. In comparison
to the Soultz-sous-Forêts reservoir, the Noble Hills has a similar granitic lithology, which has been fractured and hydrothermally altered. The front of the range is bounded by the Southern Death Valley Fault Zone (SDVFZ), a regional dextral fault, which places Tertiary sediments against granite. The photogrammetric models were produced from a combination of ground-based, unmanned aerial vehicle (UAV), and aerial photography. This multi-scale approach of image collection provided resolutions ranging from cm to dm scale. Planar feature extraction and detection was conducted using a combination of multi-point analysis, automatic facet
detection, and manual polyline extraction. This planar data was extrapolated to construct a geometric model of the Noble Hills fracture network and ultimately create a representative 3D DFN model. Given the range of fracture size and relative complexity of fracture
network, we implemented multiple workflows to address different aspects of the methodology. The DFNs were generated in 3D using a deterministic approach and the fracture planes were further projected onto orthogonal 2D slices for computing flow through
interconnected network of fractures.