Earth Sciences Division (ESD) Department of Energy (DOE) Lawrence Berkeley National Laboratory (LBNL)

The UFD Campaign:  UFD Natural Systems Research

Fracture network modeling

Fractures are typically not active flow paths in argillaceous rock as demonstrated by their natural capacity to act as flow barriers in hydrogeology, petroleum reservoirs, and more recently for geologic carbon sequestration. However, fractures are generated in the zone around a tunnel in an argillaceous rock known as the excavation disturbed zone (EDZ). Fracturing in the EDZ occurs as a result of changes in mechanical stress caused by the presence of the excavation openings and/or changes induced by operational factors. The presence of a fracture network could lead to an increase in hydraulic conductivity and preferential flow along repository excavations from waste emplacement tunnels, through access tunnels, to the accessible environment, degrading the favorable low-permeability characteristic of the rock for limiting flow and transport. Observations in European underground research laboratories have shown that EDZ fracturing occurs locally around excavations in various argillaceous rock environments. These observations include both soft-clay formations (Boom Clay at Mol, Belgium) and more indurated clay formations (Callovo-Oxfordian at Bure, France and Opalinus Clay at Mont Terri, Switzerland). Although the EDZ represents a region of enhanced permeability, this enhancement tends to decrease over time because of clay swelling and clay creep behavior, especially in cases where steel support and concrete lining provide a back pressure on the rock. These processes are in turn affected by water saturation, geochemical composition, and thermal conditions. Therefore, the evaluation of such a fracture network needs to address the changes in fracture hydrogeologic properties to coupled processes.

The initial work on this topic is focused on the development of a suitable discrete fracture network model. This approach is being taken because of the expected low degree of fracture network connectivity in such a network, particularly as the network changes over time, such that continuum approximations of fracture network behavior become less accurate. Methods to generate a 3-dimensional numerical grid for modeling flow and transport processes, including discrete fractures and intervening rock matrix, are being developed for the finite-volume numerical modeling scheme used in TOUGH2. The grid must allow for general fracture shapes and orientations and fracture intersections while also representing the rock matrix and fracture-matrix interactions. An approach using generalized Voronoi tessellation that partitions the domain into convex polyhedral is being investigated.

 

Voronoi grid for two fractures imbedded in rock matrix

Voronoi grid for two fractures imbedded in rock matrix

 

For more information, please contact:

Name Title ESD Department Phone Fax Email
Jens Birkholzer NE/NW Program Head Hydrogeology Department 510-486-7134 510-486-5686 JTBirkholzer@lbl.gov
Jim Houseworth Staff Scientist Hydrogeology Department 510-486-6459 510-486-5686 JEHouseworth@lbl.gov