Hillslope-storage and rainfall-amount thresholds as controls of preferential stormflow

- Hillslope response to rainfall events was studied using 2D numerical experiments.
- Relationships between rainfall, stormflow, and leakage to bedrock were analyzed.
- Rainfall needed to initiate stormflow appeared as dynamic hillslope property.
- Hysteretic behavior of hillslope storage-discharge relationship obtained.
- Near-saturated conditions were sufficient to initiate significant stormflow.

SummaryShallow saturated subsurface flow is a dominant runoff mechanism on hillslopes of headwater catchments under humid temperate climate. Its timing and magnitude is significantly affected by the presence of preferential pathways. Reliable prediction of runoff from hillslope soils under such conditions remains a challenge.In this study, a quantitative relationship between rainfall, stormflow, and leakage to bedrock for hillslopes, where lateral preferential runoff represents a dominant part of the overall response, was sought. Combined effects of temporal rainfall distribution and initial hillslope saturation (antecedent moisture conditions) on stormflow, leakage to bedrock, and overall water balance were evaluated by conducting simulations with synthetic rainfall episodes.A two-dimensional dual-continuum model was used to analyze hydrological processes at an experimental hillslope site located in a small forested headwater catchment.Long-term seasonal simulations with natural rainfall indicated that leakage to bedrock occurred mostly as saturated flow during major runoff events. The amount of rainfall needed to initiate stormflow appeared as a dynamic hillslope property, depending on temporal rainfall distribution, initial hillslope storage, and the spatial distribution of soil water within the hillslope. No single valued rainfall threshold responsible for triggering stormflow was found. Rainfall-stormflow as well as rainfall-leakage relationships were found highly nonlinear for low initial hillslope saturations. Temporal rainfall distribution affected the amount of rainfall necessary to initiate stormflow more than it did the amounts of stormflow or leakage to bedrock. In spite of a simple hillslope geometry with constant slope and parallel soil-atmosphere and soil-bedrock interfaces considered in the analysis, the applied model predicted a hysteretic behavior of storage-discharge relationship.The results showed a mutual interplay of components of hillslope water balance exposing a nonlinear character of the hillslope response. The study provided a quantitatively coherent insight in the hydraulic functioning of hillslopes where preferential flow constitutes a dominant part of stormflow.

139