Simulating Preferential Flow by A Multi-dimensional Process-based HYDRUS Model

Preferential flow (PF) processes are controlled by subsurface structures with a hierarchical organization across scales, but there is a lack of multiscale model validation using field data. In this study, a comprehensive dataset collected in the forested Shale Hills catchment was used to test and validate PF simulations with the 2-dimensional HYDRUS-2D model at the hillslope scale. The simulations were also compared with the 1-dimensional results at the pedon scale (HYDRUS-1D) and 3-dimensional results at the catchment scale (HYDRUS-3D). There was a good agreement between the 1D simulations and soil moisture measurements, which were mainly affected by the vertical change in porosity/permeability with depth and precipitation characteristics. However, short-term fluctuations due to PF were poorly captured. Notably, 2D and 3D simulations, accounting for PF controlled by slope position and shallow fractured bedrock, provided better results than the 1D simulations. The dual-porosity or anisotropic model provided more accurate soil moisture predictions than the single-porosity or isotropic model due to the more realistic representation of local soil and fractured shale. Consequently, our study shows the importance of multi-dimensional model approaches and the need to adequately represent the bedrocks' soil structure and fractured nature for the PF simulation. The multi-dimensional modeling approaches can represent PF pathways to the first-order stream and shows the benefits of the 3D simulation with detailed information to identify the dominant hydrological process.