Application of hydropedological insights in hydrological modelling of the Stevenson-Hamilton Research Supersite, Kruger National Park, South Africa

Abstract

Soil information is increasingly sought after for hydrological modelling, as the importance of soil in the hydrological cycle is understood better. In this paper the output of a digital soil mapping exercise was used as the soil input into a distributed hydrological model (ACRU) for a test site within the Stevenson-Hamilton Research Supersite, Kruger National Park (South Africa). The aim was to determine the effect of parameterising a hydrological model with increased levels of soil information, at different scales. To accommodate this aim, ACRU was run in 3 different modes, each with increasing levels of input, on 3 catchments, including a 1st, 2nd and 3rd order catchment. The outputs evaluated included both streamflow and soil water content at selected soil profiles. Simulation accuracy increased with higher levels of soil input, as well as with increasing catchment size. The improved accuracy with increased soil input underscores the value of detailed soil information in modelling, while the improved results with increased catchment size show that the optimal scale for including soil information has not yet been reached.