Résumé
Structural discontinuities in low-permeability media significantly influence flow and transport processes as well as a large range of coupled processes. Characterization of fracture zones is crucial for understanding numerous systems of environmental, geological and industrial importance. However, the accessibility of fractured media is generally restricted to boreholes and tunnels. This implies that fracture properties, such as their location and orientation relative to the borehole, and rock properties can be well resolved only at certain discrete point locations within the borehole. We show here how space-time distributed temperature measurements with Fiber-Optic Distributed Temperature Sensing (FO-DTS) during thermal field tests can be used to characterize fracture connectivity, to infer the profile of rock thermal properties along the borehole and to determine geometrical properties of individual hydraulically-active fractures. We performed a series of thermal experiments in a well-characterized, moderately fractured crystalline rock at the Grimsel Test Site (GTS) in Switzerland, where complementary datasets including core samples, borehole geophysics, tracer and hydraulic tests exist for comparison and validation. The first type of thermal tests are thermal dilution experiments, where the borehole is flushed with water of contrast temperature and then the temperature recovery is monitored for several hours. Thermal dilution tests were carried out under both natural and forced (i.e. cross-hole flow) conditions and are shown to enable detection of cross-flowing fractures. Moreover, we derive analytical expressions for estimating formation thermal conductivity. The second type of thermal tracer tests are cross-borehole heat tracer tests. Based on analytical method, we show how discrete borehole thermal anomalies can constrain the orientation of sparse, permeable fractures embedded in intact rock mass. The derived parameters (i.e. the number of hydraulically important fractures, their positioning, connectivity and mean orientation of discrete transport pathways) provide information on the spatial organization of discrete flow paths, which may be valuable to calibrate flow and transport simulations based on discrete fracture networks.