A Fracture Never Comes Alone: Associations of Fractures and Stylolites in Analogue Outcrops Improve Borehole Image Interpretations of Fractured Carbonate Geothermal Reservoirs

Abstract. Natural discontinuity networks control convective fluid flow in carbonate geothermal reservoirs with low matrix porosity and permeability. The network can be separated into discontinuities that formed due to local drivers (e.g. faults/folds) and the background network formed by far-field stresses, each with different scaling behaviour. Borehole data are the only source to sample the subsurface network, as the majority of the discontinuities are of sub-seismic scale. Borehole images are the most cost-effective way of sampling the network, but the limited sample area and image resolution hamper the identification of the background network in this dataset. Analogue outcrops may complement the borehole data, but only after the analogy between outcrop and subsurface reservoir is established. In this study, we present a method that uses associations of discontinuity sets to establish a robust link between the outcrop and the subsurface. A discontinuity association comprises up to 4 discontinuity sets that can form coeval in a single stress field, a well-known concept that is rarely applied for subsurface characterization of discontinuities. We use the orientations and type of discontinuity associations as paleostress indicators in order to map out principal stress trajectories of regional discontinuity-forming events that created the background discontinuity network. We demonstrate this methodology in the Geneva Basin, Switzerland, where the naturally fractured Lower Cretaceous pre-foredeep carbonates are targeted for geothermal exploitation. Outcrops in the mountain ranges that surround the basin, consistently reveal two multiscale discontinuity-forming events that formed prior to Alpine fold-and-thrusting and thus constitute the regional scale background network. Therefore, based on the analogy principle, we predict that the target reservoir is also affected by these events. We use this prediction to isolate background-related discontinuities on image logs from two borehole that penetrate the target reservoir in the Geneva Basin. This analysis reveals that ∼45 % of the observed discontinuities can be understood in the framework of the regional-scale background. In this way, we demonstrate that DAs in outcrops are a powerful tool to predict the geometry of natural discontinuity networks in the subsurface and subsequently can be used to develop geothermal exploitation strategies in naturally fractured reservoirs.