Tracing Subsurface Stormflow: Combining HYDRUS Modelling and ERT Profiles to Explore Runoff, Storage and Percolation under Intense Rainfall

In controlled rainfall experiments conducted across four catchments in Germany and Austria, rainfall simulations were conducted on 200m² large plots and 50m² small plots, all designed to detect subsurface stormflow (SSF).  At the larger plots, SSF was captured using a trench located below the irrigated area, as described in the study of Thoenes et al. (2025, hoc loco). The present study focuses on the 50m² plots, which were irrigated with an intensity of 100 mm/h for one hour. Surface runoff was collected at the downslope edge and measured in terms of both time and quantity.

Soil moisture changes were monitored using two methods: electrical resistivity tomography (ERT) along three cross-profiles, two of which intersected the rainfall area, while one was located beneath the surface runoff collection boundary. Measurements were conducted at 15-minute intervals pre-, during, and post-experiment to ensure continuous monitoring. Additionally, time-domain reflectometry (TDR) probes were installed up to a depth of 60 cm at the centre of the two ERT profiles within the rainfall area. Soil samples were collected after the experiment and analysed for physical properties, including texture, bulk density, organic content, and pF curves.

The aim of the study is to assess the potential available for deep percolation and potential SSF during intensive rainfall by employing a flexible arrangement across different hillslopes.

Using two different soil hydraulic models (single porosity model van Genuchten–Mualem and dual porosity/dual permeability model by Durner, dual van Genuchten–Mualem), the laboratory results were prepared for modelling in HYDRUS-1D. The rainfall experiments were simulated using the soil moisture data. Further calibration was performed using the measured surface runoff by adjusting the saturated hydraulic conductivity accordingly. The calibrated model allowed for a water balance calculation of the applied rainfall, partitioning it into surface runoff, soil storage, and the fraction available for deep percolation and potential SSF.

In the next step, the HYDRUS-1D simulation results were compared with the values from the ERT profiles close to the TDR measurements. Initial results confirm the findings of Pyschik et al., indicating that 80–95% of the applied rainfall is stored in the soil. The extent to which the combination of hydrological modelling and ERT profiles allows conclusions to be drawn regarding lateral water movement and SSF will subsequently be examined using HYDRUS-2D simulations in a longitudinal section.