Investigating Water Movement through Fractured Subsurface Systems: A Mesoscale Approach

Movement of fluids and contaminants through fractured subsurface systems has been of interest for many years. However, accurate modeling of the phenomena has been problematic. To provide data for the testing of predictive models and to gain experience in conducting large experiments, we designed and conducted experiments in a 2-m high by 2-m wide by 3-m long `mesoscale' cell, investigating water movement through fractured subsurface strata. In the experiments, a simulated fracture layer was packed between two coarse sand layers. Stainless steel tubes inserted through a clay matrix represented the fracture layer, which was 30-cm thick. The fracture pattern consisted of a correlated random distribution of 635 tubes with diameters ranging from 1-17 mm. Approximately, 15 metric tons of porous media were packed into the mesoscale cell. On the surface of the upper sand layer, water was applied via a 10-cm by 10-cm infiltration gallery, and it moved downward towards the fracture layer. A network of 86 probes was placed in the sand layers to measure the water pressure, which was used to monitor water movement. The experiments were focused on measuring the time-dependent spatial water arrival below the simulated fracture layer. A slight increase in the water pressure (1 cm water pressure head) was assumed to be an indicator for water arrival at the probe locations. At a few locations, probes were placed directly above and below certain fractures to characterize water movement through those fractures. An objective of the work is to use the experimental data to test modeling approaches for predicting fluid movement through fractured media. We are particularly interested in whether an effective porous medium concept can be used to describe hydraulic properties of fractured rock and to predict the general pattern of water movement through fractured rock. Results from the water infiltration experiments will be presented, including water pressure variations directly above and below fractures. At some locations, a capillary barrier appears to be operational between the upper sand layer and the fractures. At other locations, the capillary barrier effect does not appear to affect water movement into the fractures. Our study illustrates the value of mesoscale experiments for studying flow and transport processes in complex heterogeneous subsurface systems.