Fractal Roughness-Controlled Solute Transport and Apparent Local Retention in Sheared Rock Fractures Under Different Hydraulic Conditions

Abstract

Solute transport in rough fractures is governed by surface geometry, shear-induced aperture redistribution, and hydraulic conditions. This study presents a comparative three-dimensional numerical investigation of conservative solute transport in three individual self-affine fracture realizations with JRC values of 9.43, 13.18, and 17.37, five prescribed shear displacements (Us = 1–5 mm), and seven Peclet numbers (Pe = 0.1–1500), yielding 105 finite-element cases. For the selected realizations, increasing roughness and shear displacement reorganized the aperture field, enhanced streamline tortuosity, and increased flow heterogeneity, particularly in the medium- and high-roughness geometries. Under the investigated low-Reynolds-number conditions, the pressure-gradient-flow-rate relationships remained approximately linear. Mechanical aperture, hydraulic aperture, and intrinsic permeability evolved asynchronously: hydraulic transport capacity remained nearly stable in the low-roughness realization but decreased progressively in the medium- and high-roughness realizations. Pe primarily controlled the breakthrough regime. Diffusion dominated at Pe = 0.1, mixed advection-diffusion behavior occurred at Pe = 1–10, and advection dominated at Pe ≥ 100. Increasing shear displacement mainly affected the early and late portions of the breakthrough curves, whereas the central transition changed comparatively little. At high roughness and high Pe, concentration isosurfaces showed stronger preferential transport and delayed migration in low-velocity or weakly connected regions. The resulting apparent local retention denotes transient storage and delayed exchange within the fracture void space. No independent experimental validation, multilevel mesh-convergence test, stochastic ensemble analysis, systematic parameter-sensitivity analysis, or breakthrough-curve-derived residence-time, dispersivity, tailing, or transport-aperture metrics were performed. The findings should therefore be interpreted as comparative trends for the three investigated rigid-wall, isolated-fracture, steady-flow, and conservative-transport models, rather than as statistically representative JRC-class relationships or field-scale predictions.