Improved understanding and modelling of hydrological responses in formal and informal urban areas: case study in Tshwane, South Africa.

Abstract

It is widely accepted that urban development results in larger flood peak discharges with faster catchment response times, larger total runoff volumes and lower base flow volumes. However, these effects have not previously been studied in the South African context, where typical development characteristics in both formal and informal settlements differ from those reported in previous studies. In order to configure accurate hydrological models for South African conditions, it is important to understand the impacts of typical South African urban development types on runoff. This study aims to gain new understanding of hydrological processes in the diverse range of South African urban environments and to use this knowledge to develop an improved approach for hydrological modelling in South African urban areas. This study investigates the effects of urban development on runoff, including: (a) total runoff volumes, (b) stormflow volumes, (c) base flow volumes, (d) flood peaks and (e) catchment response time from South African urban catchments, using case studies from Gauteng Province. It is shown that total runoff, stormflow volumes and base flow all increase in the study catchments regardless of whether the upstream development is formal or informal. However, contrary to current scientific consensus, changes in flood peaks and catchment response times with development are statistically insignificant in most catchments with formal urban development in this study. The seeming paradox of increased flood volumes without corresponding increase in flood peaks and decreased catchment response time is attributed to the temporary storage and subsequent attenuation of peaks in these catchments. The catchment with the largest proportion of informal development shows increasing trends in both volume and flood peaks, although catchment response times do not show significant change. It is postulated that the complexity of increased imperviousness without increased connectivity of impervious areas to drainage systems, as typically found in informal settlements, could contribute to this result. In order to assess the impact of the improved understanding and quantification of urban responses on modelling, the Storm Water Management Model (SWMM) was configured for two catchments, first using conventional parameter sets, then incorporating measured and derived imperviousness and connectivity data from this study. The simulation results show that the parameter values proposed in this study improve the results in both catchments. Considerable additional improvement is achieved through the incorporation of pseudo storages to simulate unexpected attenuation in the catchments. Based on the results it is recommended that attenuation be incorporated in hydrological models of urban areas in ungauged catchments with similar characteristics. Recommendations are made for further investigation into reasons for the findings, as well as further studies in other South African urban areas, when additional data is available.