Managing nutrient flows into the Zandvlei Estuary, Cape Town using Sustainable Drainage Systems (SuDS)

Master Thesis


Permanent link to this Item
Journal Title
Link to Journal
Journal ISSN
Volume Title
The Zandvlei estuary is the only functional estuary on the False Bay coastline which forms part of the southeast boundary of the City of Cape Town (CCT) on the southwest coast of South Africa. It provides an important habitat for fledgling fish species in the area and is a popular recreational site for birdwatchers, sailors and conservation-enthusiasts. The estuary has experienced water quality problems for many years. It drains three upstream sub-catchments which are largely comprised of conventional drainage systems which carry stormwater pollutants into the watercourses that then drain into the estuary. Of the many pollutants that flow into the estuary, elevated nutrient levels (specifically phosphorus and nitrogen) are a major concern because they can lead to eutrophication and excessive plant growth. In recent years the estuary has experienced algal blooms resulting in adverse effects on local plants, and animal and human life. Sustainable Drainage Systems (SuDS) are holistic drainage practices that address, amongst other things, water quality issues like those experienced by the Zandvlei estuary by treating the stormwater. The aim of this study was to identify areas of raised nutrient concentrations and model various SuDS measures to improve the water quality of runoff. Surface water sampling in the catchment and a study of historical water quality data was undertaken to identify areas with higher nutrient concentrations. Phosphorus concentrations, in the form of SRP, was found to be above the eutrophic threshold in most samples, while nitrogen concentrations, in the form of TIN, were found to mostly be beneath the eutrophic threshold. The locations of higher nutrient concentrations coincided with suspected sources of raised nutrients based on a desktop study of the catchment that identified agricultural areas, golf courses that use treated effluent for irrigation, and commercial/industrial areas. Water level sensors were developed and placed in each sub-catchment to compare nutrient concentrations during storms to stream flow during storms. A hydrological and hydraulic model was built representing two of the river systems feeding into the Zandvlei estuary – the Keysers and Westlake River sub-catchments. The PCSWMM software package developed by CHI, which uses the SWMM engine developed by USEPA, was employed. Data regarding historical rainfall, land use, stream flow, surface cover, soil types, geology and historical temperatures were used to construct the model. The model was calibrated against historical stream flow data obtained from CCT. Various SuDS measures were modelled and grouped into three scenarios: Scenario 1 modelled local controls targeting the suspected source areas, Scenario 2 involved modelling large-scale regional controls, and Scenario 3 was a combination of the first two. The local controls modelled were swales and bioretention areas. The regional controls modelled were wetlands, primarily because the study area already contains wetland areas that are currently disconnected from the river systems. Wetlands were modelled with upper and lower limit treatment equations. The simulated wetlands were able to reduce the SRP and TIN concentrations to below the eutrophic threshold (below 0.25 mg/ℓ and 2.5 mg/ℓ respectively) for three years out of the 8-year simulation period. The various scenarios achieved pollutant reductions ranging from approximately 25-80%. The CCT requires a TP reduction of 45% and a TSS reduction 80%. For TP, this was achieved throughout the simulation period when wetlands are functioning efficiently (i.e. the upper treatment limit); while for TSS, this requirement was only achieved in some years when the wetlands are functioning efficiently. Scenario 1 showed the least pollutant removal overall, while Scenario 2 and 3 showed similar pollutant removal and flow reduction, suggesting that the effect of the proposed local controls is insignificant compared to that of the large-scale regional controls.