Examining the implications of spatial variability of saturated soil hydraulic conductivity on direct surface runoff hydrographs

Abstract

Saturated soil hydraulic conductivity (Ksat) and soil moisture content (Θ) are key soil hydraulic properties for the understanding of hydrological processes at the watershed scale. Previous approaches in the use of Ksat to describe hydrological processes have mainly applied standardized or derived Ksat values, although the high spatial variability of this soil parameter has been frequently reported. Hence, this study aims to assess the influence of Ksat and Θ spatial variabilities on direct surface runoff (DSR) hydrographs resulting from isolated rainfall events using the Limburg Soil Erosion Model (LISEM) in conjunction with measured Ksat data. Data sets were obtained from a grid of 179 points established in the Ellert creek watershed (ECW), with a drainage area of 0.7 km2. The hydrological monitoring in the ECW consists of an automatic rain gauge and a water level gauging station, which provides rainfall-runoff data every 5 min. Ordinary Kriging was used to map Ksat in the area. Five 5-minute interval rainfall-runoff events were selected from 2017 to 2019 for the hydrological modeling in LISEM, whose water level did not go beyond the reliable applicability of the stage-discharge rating curve generated for the ECW’s outlet. The rainfall depth of these events varied from 24.0 to 66.4 mm, and peak discharge varied from 0.49 to 4.92 m3/s. The effects of Ksat spatial variability and its uncertainties on DSR hydrographs were assessed using the Sequential Gaussian Simulation. As the depth and average intensity of rainfall decreased, there was a greater propagation of the uncertainties of Ksat to peak discharges. The simulated DSR hydrographs and their peak discharges were adequately represented by LISEM. However, LISEM was unable to simulate the hydrological response of the watershed to a two peaks event separated by lower rainfall intensity.