To investigate the effect of fracture surface roughness on solute transport in a rough fracture and the effect on the hydrodynamic dispersion coefficient D_L and transport velocity V_t, roughness and hydraulic aperture of various fractures under constant head were calculated using fractal theory. The study discussed the influence of roughness on solute transport in rough fractures, as well as the seepage flow and solute velocity with different fractal dimensions. Breakthrough curves were generated by integrating the Navier-Stokes equation and the advection-diffusion equation (ADE). The breakthrough curves from direct simulation were coupled using the classical ADE approach. Subsequently, D_L and V_t were determined. The study explored the control exerted by D_L and V_t on solutes and their correlation with roughness. A prediction model based on fractal theory for solute transport in rough fractures was established and compared with direct simulation and geometric feature-based models. Results indicate that fractal dimension D and amplitude parameter K_v effectively characterize surface roughness, with a strong correlation between D_L, V_t, and fractal parameter DK_v. The established model demonstrates high applicability and accuracy in simulating solute transport in rough fractures.

During the operational phases of the upper reservoir in a pumped storage power station, the water level, leakage area, and hydraulic gradient of the upper reservoir alter dynamically due to the cyclic pumping and draining activities. The rising groundwater level during storage introduces distinct leakage conditions within the reservoir basin, characterized by unsaturated, partially saturated, and saturated states. Consequently, reservoir basin leakage exhibits variability across these states. To address this issue, this study formulated rational assumptions corresponding to the three leakage states in a reservoir basin and derived analytical expressions for seepage calculation based on Darcy's law and the principles governing groundwater flow refraction. A case study was conducted to investigate the relationship between various factors and leakage. The results showed that leakage primarily depended on the permeability of the impermeable layer in the reservoir basin. The upper reservoir leakage was estimated, and the calculated leakage generally agreed with the measurements, offering insights into the leakage mechanism of the Liyang pumped storage power station. In addition, the reasons for disparities between measured and calculated leakage were analyzed, and the reliability of the developed method was validated. The findings of this study provide a foundation for the seepage control design of upstream reservoirs in similar projects.