4.3 Brittle Deformation History and Current Fracture Network Flow Properties

The architecture of the connected network and its flow properties (e.g., anisotropy, transmissivity) depend on the geometric characteristics of the fractures, namely orientation, aperture, size (or persistence), spacing, and fracture interactions. These, in turn, are determined by the characteristics of the affected rock and by its brittle deformation history, i.e., the number of tectonic events as well as their duration, stress magnitude and orientation.

The influence of fracture persistence on variation of hydraulic conductivity (K) of a fractured aquifer over physical dimensions ranging from meters to kilometers was studied by Shapiro and others (2007). Single-hole hydraulic tests conducted at a scale of a few meters showed a range of K from 10−10 to 10−4 m/s. Cross-borehole hydraulic tests at distances ranging from 10 to 100 m showed the presence of highly transmissive fractures with a K of approximately 10−4 m/s. However, for distances over 100 m, the bulk K was approximately 10−7 m/s because it was controlled by the less conductive fractures. The same values were found up to distances of kilometers. The authors attributed this behavior to the poor connectivity of the fracture network which was imparted by the short trace lengths of the fractures that rarely exceeded 10 m, as was observed on the surveyed road cuts.

The current stress field may control the final stage of the brittle deformation history and can influence the aperture and transmissivity of fractures (Banks et al., 1994; Barton et al., 1995; Ferril et al., 1999; Morin & Savage, 2003; Morin et al., 2006; DesRoches et al., 2014). Barton and others (1995) demonstrated that there is a relationship between in-situ stress and fluid flow. The analysis of data obtained from three boreholes indicated that, in the highly fractured crystalline rocks they studied, the critically stressed faults appear to be the most important hydraulic conduits. These faults are either parallel to or at an acute angle (up to 30° to 40°) with the orientation of SHmax at a specific site. This means that they are optimally oriented for reactivation by opening or by shear in the current stress field. Similarly, Ferril and others (1999) proposed that “faults with favorable orientations for slip or dilation are potential fluid flow pathways” and demonstrated that it is consistent with “anisotropic transmissivity controlled by faults and fractures active in the present-day in-situ stress field.” The transmissivity values were obtained from data derived from a long-term aquifer pumping test.

In terms of the effect of topography on the in-situ stress, Morin and others (2006) demonstrated that it can locally influence the transmissivity of fractures and concluded that high dip fractures, parallel and close to a steep topographic slope, were more transmissive due to the influence of the laterally free slope face on the in-situ stress. This topography effect is quite local, being present within only 100 m of the slope face. Morin & Savage (2003) considered the effects of regional and local stresses to better understand the hydrologic system of a fractured-rock aquifer. Gravity and tectonic stresses vary with depth and with the specific location of a site. For example, along a slope fracture connectivity systematically increases with depth, whereas it increases only moderately with depth below a valley floor.

Flow evidence along fractures observed in outcrops and wells shows that the orientation of more transmissive fractures can vary substantially over tens of kilometers (e.g., Fernandes et al., 2016b), or abruptly at the same location (Talbot & Sirrat, 2001). Thus, the simple extrapolation of results from one region to another is not recommended; on the contrary, local data should always be collected. For example, Fernandes & Rudolph (2001) show that wells on NW lineaments are more productive in certain structural domains, whereas wells in NNE lineaments are more productive in other domains. The authors concluded that such domains are related to two Quaternary tectonic events having different stress orientations. Lineament interpretation has been used in hydrogeological studies as an auxiliary method for structural characterization (e.g., Mabee et al., 1994; Gleeson & Novakowski, 2009). Considerations regarding the limitations, reach, and scope of lineament studies are provided in Box 2.