Investigating Contaminated Sites on Fractured Rock Using the DFN Approach

This presentation provides an overview of a major field focused program of studies aimed at improved investigation methods and understanding organic contaminant source zones and plumes in fractured porous sedimentary rock. This research began in 1997, when intensive field studies were initiated at a TCE contaminated site on steeply dipping and faulted sandstone in California. Now, with collaborations involving several disciplines (analytical chemistry, mathematical modelling, geophysics, microbiology), the program includes three other sites contaminated with chlorinated solvents in addition to the California site: a Wisconsin site on flat-lying sandstone and two sites in Ontario on flat-lying dolostone. These four sites have important differences so that they are broadly representative of sedimentary rock but they have several aspects in common, including: much site data from earlier conventional investigations, contamination initially caused decades ago by DNAPL flow into the rock, sufficient matrix porosity (2-20 %) to allow diffusion-driven chemical mass transfer between fractures and the rock matrix causing strong influence on contaminant behaviour, deep contaminant occurrence (greater than 350 m below ground surface at one site), and each site receives much regulatory attention. Also, the plume fronts advance at rates much slower than the average linear groundwater velocity in the fracture networks. Based on the field results obtained to date, a general conceptual model for the formation and long-term evolution of source zones and plumes in fractured porous sedimentary rock is proposed. This conceptual framework is being tested and the various processes quantified through field investigations using a suite of high-resolution techniques that I refer to as the discrete-fracture network (DFN) investigation approach. This then allows application of DFN numerical models, such as FRACTRAN and HydroGeoSphere, to simulate flow and contaminant transport at these sites. Conventional field methods used in fractured rock studies are poorly suited for plume delineation or characterization and therefore, new methods are being developed and used at all of the sites. In the DFN approach, emphasis for data acquisition is on data specific to individual fractures and the fracture network as well as the rock matrix blocks between fractures so that the characteristics and interactions between these two domains can be discerned. Hence, the spatial scale of measurements on continuous rock core and also in the core holes must be exceptionally detailed. Rock core contaminant analyses at each site confirm that nearly all of the contaminant mass now resides in the lowpermeability rock matrix although the down-gradient transport occurs in numerous, wellconnected fractures. Therefore, quantifying the interactions between these domains is