Abstract. A procedure involving simultaneous experimental and numerical research is described for the purposes of understanding the mechanisms involved when extreme precipitation is transformed to flood stream discharge. It is shown that experiments and model applications by themselves are not sufficient for process identification, but that their combined application provides considerable insight into the subsurface flow processes. The proposed approach is unconventional in that a numerical model, based on stringent continuity and momentum criteria, is used as a tool for process identification only. Unlike other studies, it is not intended to demonstrate the applicability of the utilised model for general hydrological applications, or to provide evidence of the suitability of particular model simplifications. Rather, different and sometimes conflicting model realisations are used to examine the plausibility of flow processes which may occur on natural hill slopes. Hereby, small scale effects such as those relating to the mechanisms of water entry into the macropores, and the movement of water to the surrounding matrix are identified from the results of well instrumented field experiments.