|dc.description.abstract||Hydraulic aperture of joints is an important parameter for estimating the fluid flow capability of rock mass. However, the measurements of hydraulic apertures are not common and uncertainty encountered accordingly in the groundwater flow modelling for safety assessment of the radioactive waste final disposal. A synthetic hydraulic aperture dataset was used to evaluate the statistical characteristics of hydraulic apertures via different analysis methods. First of all, reasonable statistical distributions of different parameters of joints (joint roughness coefficient JRC, joint compressive strength JCS, and joint orientation) are assumed according to the data collected from specific sites where the sandstone and argillite outcropped. Secondly, the joints with different orientation, JRC and JCS are randomly generated on the basis of different lithology (sandstone and argillite) at different depth. Thirdly, the Barton-Bandis model was used to calculate the mechanical apertures of each joint under assigned in-situ stresses. Finally, synthetic hydraulic aperture dataset can be obtained according to the mechanical aperture and empirical function described the relation between hydraulic and mechanical apertures. In addition to the synthetic case, we collected a real case where the hydraulic apertures at different depth are available.
According to the testing results of synthetic case and real case, the probability density distribution of the hydraulic aperture conforms to the lognormal distribution. In engineering practice, a normal distribution assumption of hydraulic apertures could be problematic. The standard deviations of the hydraulic apertures are largest when the influence of depth, lithology, and joint characteristics (joint orientation, JRC and JCS) is neglected. When the hydraulic apertures of different lithology were separately analyzed, the uncertainty dropped. If the decreasing trend of hydraulic apertures with depth was considered, the uncertainty dropped further. When the multivariate regression analysis model considering the depth and joint characteristics was used to analysis the synthetic hydraulic apertures of sandstone and argillite separately, the standard deviation is the lowest among others. According to the cubic law, the standard deviation of the joint transmissivity can be reduced to 0.56% of the one where neglecting all of the influential factors when making statistical analysis of the synthetic hydraulic fracture. The analysis result of real case also shows a significant reduced uncertainty when the decreasing trend of hydraulic apertures with increasing depth was considered.||en_US|