Tala Maria Aabø
Øster Voldgade 10, 1350 København K, Område III, Building: 03-1-355
PhD project; Fractures and Their Influence on Fluid Flow in Chalk
Objective: The Rørdal quarry (Aalborg, Denmark) offers a unique opportunity for studying an onshore natural fracture network in chalk. In this project a digital 3D dataset of the quarry site is generated through a holistic integrated workflow, enabling an extensive digital mapping of structural lineaments from cm- to m scale. The resultant data is used for increasing our understanding of local stress regime, fracture orientation and fracture connectivity of the system. Moreover, results will provide generic knowledge of fracture interaction and fluid flow in chalk systems, which is a crucial part of e.g. modelling of geothermal-, hydrocarbon- and groundwater reservoirs.
Approach: the ~1km N-S oriented Rørdal quarry is currently site to cement production. As excavation progresses, new vertical sections or time-slices of the chalk wall are exposed. Multiple generations of such time-slices are reproduced as virtual outcrops, which are primarily based on georeferenced imagery collected by drone and processed in Agisoft Metashape. The virtual outcrops are combined in VR Geological Studio, which allows for tracing of digital fracture-, fault- and bedding interpretations from one model generation to the next, thus creating a 2.5D coverage of the Rørdal quarry. A 3D structural model (static) is constructed by integrating the interpretations from VR Geological Studio with GPR data in Petrel. This framework in turn serves as input to a geo-mechanically based discrete fracture network model (DFN), which is currently under development at DTU (Technical University of Denmark). The output of the DFN provides a full 3D coverage of fractures in Rørdal, which is founded upon geological principles for fracture propagation and termination within a layer-bound principle. This data is in turn subjected to fluid flow modelling and compared to real fluid flow observations from GPR in a collaboration project. One of the virtual outcrops have been mapped in detail (covering 1000s of fractures) for analysis in FracPaQ (Matlab). The large number of mapped fractures in the FracPaQ plugin has potential for big data analyses.