Stress-dependent permeability of fractured rock masses: a numerical study

Abstract

We investigate the stress-dependent permeability issue in fractured rock masses considering the effects of nonlinear normal deformation and shear dilation of fractures using a two-dimensional distinct element method program, UDEC, based on a realistic discrete fracture network realization.A series of "numerical" experiments were conducted to calculate changes in the permeability of simulated fractured rock masses under various loading conditions. Numerical experiments were conducted in two ways: (1) increasing the overall stresses with a fixed ratio of horizontal to vertical stresses components; and (2) increasing the differential stresses (i.e., the difference between the horizontal and vertical stresses) while keeping the magnitude of vertical stress constant.These numerical experiments show that the permeability of fractured rocks decreases with increased stress magnitudes when the stress ratio is not large enough to cause shear dilation of fractures, whereas permeability increases with increased stress when the stress ratio is large enough.Permeability changes at low stress levels are more sensitive than at high stress levels due to the nonlinear fracture normal stress-displacement relation. Significant stress-induced channeling is observed as the shear dilation causes the concentration of fluid flow along connected shear fractures.Anisotropy of permeability emerges with the increase of differential stresses, and this anisotropy can become more prominent with the influence of shear dilation and localized flow paths. A set of empirical equations in closed-form, accounting for both normal closure and shear dilation of the fractures, is proposed to model the stress-dependent permeability. These equations prove to be in good agreement with the results obtained from our numerical experiments.2

Polyaxial stress-dependent permeability of a three-dimensional fractured rock layer

“…The simulation results show consistency with previous field observations: only a small portion of fractures are highly conductive (Tsang and Neretnieks 1998;Follin et al 2014); critically stressed faults tend to have much higher hydraulic conductivity (Barton et al 1995;Zoback 2007). The observation that the shear dilation of preexisting fractures has dominant effects on flow localisation based on our simulations is also consistent with the results of previous numerical studies (Min et al 2004;Baghbanan and Jing 2008;Zhao et al 2011). Compared to the results based on an idealised 3D persistent fracture network (Lei et al 2015a), the permeability of this fractured layer is less sensitive to stress changes, because the matrix blocks of this limestone layer are partially bounded by impersistent fractures and tend to be more difficult to rotate under more restrictive interlocking between blocks.…”

Monte Carlo simulation of radionuclide migration in fractured rock for the performance assessment of radioactive waste repositories

“…The same case study is then extended to consider that the aperture of the fractures may stochastically vary in time due to the mechanical stresses induced by earthquakes hitting the migration domain [27]. The apertures of fractures may change due to normal stress-induced closures or openings and to shear stress-induced dilations; thus, the permeability of fracture rock masses is stress-dependent [24] [26]. This indirect hydro-mechanical coupling is particularly important in fractured rock masses, since stress-induced changes in permeability can be large (several orders of magnitude) and irreversible under, in general, perturbations resulting from various natural and human activities [24].…”

“…The apertures of fractures may change due to normal stress-induced closures or openings and to shear stress-induced dilations; thus, the permeability of fracture rock masses is stress-dependent [24] [26]. This indirect hydro-mechanical coupling is particularly important in fractured rock masses, since stress-induced changes in permeability can be large (several orders of magnitude) and irreversible under, in general, perturbations resulting from various natural and human activities [24]. For simplicity, but with no loss of generality, we now assume that i) the times at which the earthquakes occurr is distributed according to an exponential distribution with parameter [43], which is the average return rate of magnitude 4/5 earthquakes of the site proposed in [35] and ii) each earthquake occurrence induces a fracture permeability change which, on average, conservatively increases the pore velocity by a 10% factor, starting from the initial value 0 , f v .…”

“…In particular, in a realistic setting, the geometric and hydrologic properties of the transport media may be subject to changes which can be both continuous and discrete in time. For example, such modifications may include the closure of fractures in salt deposits due to the plasticity of the medium [19] [20], the decrease in fracture aperture due to slow deposition [21], seasonal water content changes in seasons due to rainfalls [22], thermo-hydraulic-mechanical-chemical (THMC) couplings which can affect the permeability of the media and the average hydraulic gradient [23] [24][25] [26] or the abrupt changes in the fracture aperture due to shock events (e.g. earthquakes, volcanic eruptions, etc.)…”

Naturally fractured hydrocarbon reservoir simulation by elastic fracture modeling

“…Taron et al (2014) tested fracture dilation in a geothermal system. However, Min et al (2004) and Farahmand et al (2015) derived relationships between aperture dilation and applied stress. Min et al (2004) studied completely all cases and we used the results that they derived in their complete study.…”