Global climate change is projected to continue and result in prolonged and more intense droughts, which can
increase soil water repellency (SWR). To be able to estimate the consequences of SWR on vadose zone hydrology, it is
important to determine soil hydraulic properties (SHP). Sequential modeling using HYDRUS (2D/3D) was performed on
an experimental field site with artificially imposed drought scenarios (moderately M and severely S stressed) and a control
plot. First, inverse modeling was performed for SHP estimation based on water and ethanol infiltration experimental data, followed by model validation on one selected irrigation event. Finally, hillslope modeling was performed to assess water balance for 2014. Results suggest that prolonged dry periods can increase soil water repellency. Inverse modeling was successfully performed for infiltrating liquids, water and ethanol, with R2 and model efficiency (E) values both > 0.9. SHP derived from the ethanol measurements showed large differences in van Genuchten-Mualem (VGM) parameters for the M and S plots compared to water infiltration experiments. SWR resulted in large saturated hydraulic conductivity (Ks) decrease on the M and S scenarios. After validation of SHP on water content measurements during a selected irrigation event, one year simulations (2014) showed that water repellency increases surface runoff in non-structured soils at hillslopes.
Capillary barriers are an interesting alternative component for cover systems of landfills and contaminated sites. Provided they are sufficiently validated, soil hydrological models could be fast and powerful tools for the dimensioning of capillary barriers. Outflow rates measured in a 10 m long tipping trough for one material combination and two slopes from stationary periods were compared to simulation results of HYDRUS (2D/3D), Version 2.05. The measured outflow rates show a typical pattern with slope-dependent threshold values indicating the efficiency of the capillary barrier. This flow pattern could not be reproduced with HYDRUS (2D/3D) that for different input setups produced smooth patterns without thresholds. The input setup was varied for different soil hydraulic models (van Genuchten-Mualem vs. Brooks-Corey), homogeneous and heterogeneous transport domains (no scaling vs. stochastically distributed scaling factors considering the Miller-Miller similitude), different HYDRUS versions (standard vs. alternative; i.e., with material properties assigned either to finite element nodes or finite elements, respectively), and different lower boundary conditions (seepage face vs. free drainage). Differences between measured and simulated outflow patterns could be caused by the measurements, the application of the model, or by the model itself. The van Genuchten-Mualem model may not be suitable to describe the soil hydrological relationships of these particular materials. The reason for the mismatch, however, could not be identified yet.