Fracture networks and fault zones play an important role for subsurface fluid flow. Estimating the impact on the permeability field of such structures is of high interest for reservoir characterization, groundwater management, and exploration targeting. Of particular interest is the anisotropic permeability resulting from fracture sets, which can be represented as a tensor. These permeabilities of small- to mesoscale fracture networks can be utilized in macroscale models of macroscale fault networks. Obtaining exact permeability values from fractured rocks through laboratory experiments is challenging and subject to large uncertainties. Numerical methods can help estimating reliable values but are computationally expensive. On the large scale it is desirable to predict the dominant pathways in fault networks as these zones can strongly affect localization of mineralizing fluids, can affect the productivity of reservoirs (hydrocarbon and geothermal), and has implications for groundwater management. Here, we present a methodology that (1) allows for estimating permeability anisotropy, and (2) predicting dominant fluid pathways from regional scale maps. Assuming that the permeability of fractured or faulted media is governed by the connectivity of the network entities, we will show how permeability anisotropy and dominant pathways can be obtained from a graph representation of 2D discontinuity networks. The graph metrics we base our analysis on are the betweenness centrality and maximum flow.
Open-Access Online Publication: March 03, 2023