Scale effect in macroscopic permeability of jointed rock mass using a coupled stress-damage-flow method

The scale effect holds true for both the strength and permeability of a rock mass, especially for a jointed rock mass. Thus, the size of a rock mass structure must be considered during design, and the rock mass permeability obtained in laboratory must be modified against the scale effect before being applied in engineering practice. Based on the progressive failure process analysis and the fluid–solid coupled theory, this paper reproduces variations of macroscopic permeability of jointed rock masses with various sizes using numerical methods. The relationship between the observation scale and the seepage directionality or randomicity is presented, and the sensitivity of the scale effects of rock mass strength and permeability is compared. The results from this study reveal that for jointed rock masses with randomly distributed orthogonal joints, as the observation scale increases, the macroscopic permeability increases following a negative exponential equation. The variation of permeability along different directions gradually attenuates, and the rock mass permeability tends to be isotropic at the macroscopic level. When the observation scale approaches the characteristic size, the macroscopic permeability becomes stable and isotropic. Moreover, it is found that the rock mass strength is more sensitive to the scale effect than the permeability, and the characteristic size corresponding to strength is smaller than that for permeability. Furthermore, the rock mass permeability suddenly increases as the rock mass fails under compression, which is caused by coalescence of the seepage channels. In this case, the scale effect becomes less important. Thus, during the design of rock structures, it is necessary to comprehensively consider the scale effects of both rock mass strength and permeability, and the parameters shall be calibrated according to the laboratory test results.