Comparing the variable flow velocity advection-dispersion model with the continuous time random walk model: Insights from a soil column salt transport experiment

Abstract

The transport of salt in low-permeability media can significantly alter permeability. To investigate the impact of permeability changes on solute transport simulations, this study conducted a salt transport experiment using a soil column with an upper sand layer, a middle clayey silt layer, and a lower sand layer. The experimental results showed that, under a constant hydraulic head difference, the flow rate decreased linearly, with a reduction exceeding 35 %. The solute penetration time increased significantly along the flow path. Using the velocity decay data, an advection-dispersion model that accounts for variations in flow velocity was developed. Additionally, an advection-dispersion model with constant flow velocity and a continuous time random walk (CTRW) model were also developed, both of which ignored velocity changes. Simulation results indicated that the variable flow velocity advection-dispersion model yielded the best results, followed by the CTRW model, with the constant flow velocity advection-dispersion model performing least effectively. When measured data on flow velocity changes are available, the variable flow velocity advection-dispersion model is recommended. If measured velocity change data are unavailable but the range of CTRW model parameters can be ascertained, the CTRW model is advisable. In the absence of ascertainable CTRW model parameter ranges and when the constant flow velocity advection-dispersion model is necessary, the advection-dispersion model with a constant maximum flow velocity should be employed.