Predominant Runoff Components During Heavy Rainfall Events on Cultivated Catchment

Abstract:

The fact that flash floods initiated in arable catchments are often accompanied by massive sediment and nutrient loads often leads to the assumption that surface runoff is the principle pathway by which runoff reaches watercourses. But the hydrology of cultivated catchments has its specific features due to the temporary variable topsoil properties and a sharp divide between topsoil and compacted subsoil. Under various conditions the prevailing runoff mechanisms may vary from surface runoff to subsurface runoff or deep percolation. On the basis of an evaluation of several rainfall-runoff events in a representative agricultural catchment (Nucice, Czech Republic), we show that runoff from cultivated land may be generated in a way similar to that seen on forested slopes, where shallow subsurface runoff is the predominant pathway. To identify the predominant runoff pathway, we employed a combination of turbidity measurements and stream discharge data. Although we observed temporal variability of topsoil properties attributable to seasonal weather changes and agricultural activities, e.g. bulk density and porosity, runoff generation was mainly driven by precipitation characteristics and the initial catchment saturation. The concept of the runoff formation was also observed during plot scale experiments with rainfall simulator. Various initial soil moisture conditions, and vegetation stages delimited the simulations. Variable proportions of both monitored runoff components were observed in relation to rainfall intensity and duration, ranging from zero surface runoff to a distinct dominance of surface runoff. Even with the highest tested precipitation intensities, surface runoff always formed due to saturation excess of the topsoil, irrespective of the topsoil properties and crops. The experiments were numerically modelled and analysed to understand the effect of temporal variability in the macropores and intra-aggregate voids ratio within the topsoil. We used a combination of physically based macroscopic models S1D and HYPO. In the S1D the dual permeability approach with two coupled Richards equations is used, the simultaneously operating HYPO code is based on a diffusion wave (Boussinesq eq.). The research was performed within the framework of project No. 13-20388P.