Coupling of Stress-Strain Behavior and Seepage Flow - Rock Mechanics Based on an Anisotropic Jointed Rock Model - page 172

The change of permeability of discontinuities subjected to a normal compressive stress
and a shear stress can be measured in so-called “coupled shear-fl ow tests”. Such tests
were carried out inter alia by Makurat (1985), Dillo (1991), Gentier et al. (1997), Gen-
tier et al. (1998), Olsson (1998), Olsson (1999), Esaki et al. (1999) and Li et al. (2008).
For example, Esaki et al. (1999) conducted coupled shear-fl ow tests under a constant
hydraulic gradient and different constant normal stress levels on granite specimens with
artifi cially created rough joints. In these tests at fi rst a shear displacement δ s,D up to

20 mm was accomplished and then a shear displacement was conducted in the reverse
direction. Permeability was determined at each step of shearing by keeping the shear
displacement constant until steady-state fl ow conditions were achieved. Shear stress τ res

and normal displacement of the specimen δ n were recorded continuously. Normal dis-

placement of the discontinuity δ n,D , and thus mean aperture , was determined by

subtracting the compression of the intact rock from the total normal displacement of
the specimen. Hydraulic aperture (2a i ) h was back-calculated from fl ow rate measure-
ments using the cubic law for specifi c fl ow Q s according to Equation (6.16).
In Fig 7.6 (left) mean and hydraulic apertures versus δ s,D are represented for a test with

an applied normal stress of σ n = 1 MPa. Accordingly, the mean aperture increased with

increasing shear displacement. After a shear displacement of 20 mm a dilatant normal
displacement of some 3 mm occurred. In contrast, the hydraulic aperture approached a
constant value with increasing shear displacement.
The permeability of the discontinuity k d calculated according to Equation (6.19) in
tests with stress levels of σ n = 20 MPa increased with increasing shear

displacement by one to two orders of magnitude for the fi rst 5 mm and then gradually
approached to a constant value. During reverse shearing the permeability took on sim-
ilar values to those for forward shearing (Fig. 7.6, right).

σ n = 1 MPa and

Figure 7.6 Results of coupled shear-fl ow tests on a rough artifi cial joint in granite (Esaki et al. 1999)

Similar results were obtained in coupled shear-fl ow tests carried out on natural granite
joints by Olsson (1998 and 1999).
The constant hydraulic aperture and permeability after reaching a particular shear
displacement was explained by roughness degradation and reduction of joint vol-