Advanced Conceptual Models for Unsaturated and Two-Phase Flow in Fractured Rock

Abstract

Research Objective: The Department of Energy Environmental Management Program is faced with two major issues involving two-phase flow in fractured rock; specifically, transport of dissolved contaminants in the Vadose Zone, and the fate of Dense Nonaqueous Phase Liquids (DNAPLs) below the water table. Conceptual models currently used to address these problems do not correctly include the influence of the fractures, thus leading to erroneous predictions. Recent work has shown that it is crucial to understand the topology, or 'structure' of the fluid phases (air/water or water/DNAPL) within the subsurface. It has also been shown that even under steady boundary conditions, the influence of fractures can lead to complex and dynamic phase structure that controls system behavior, with or without the presence of a porous rock matrix. Complicated phase structures within the fracture network can facilitate rapid transport, and lead to a sparsely populated and widespread distribution of concentrated contaminants; these qualities are highly difficult to describe with current conceptual models. The focus of our work is to improve predictive modeling through the development of advanced conceptual models for two-phase flow in fractured rock.Research Approach: Preliminary experiments have shown that behavior at fracture intersections is key, thus we are employing systematic experimentation to identify and classify behavior at intersections. Understanding gained at the scale of individual fracture intersections will be used to augment a Modified Invasion Percolation Model (MIP) that has shown significant promise in predicting flow through fracture networks. Development of the MIP model will be iterative, in that we will use numerical simulations to design critical physical experiments at the network scale, which will challenge the model. The augmented, and fully tested MIP model will be exercised on realistic fracture networks for the purpose of understanding large-scale development of phase structure. Results of the large-scale network simulations will be abstracted to the data sets from the INEEL Vadose Zone Research Park so that critical features may be included in conceptual models used by the Environmental Management Science Program and the Idaho Closure Project.