Stormwater treatment during infiltration and its effect on shallow urban groundwater quality in the Cape Flats Aquifer: field study and soil-column experiments

Schneuwly, Rachelle

Stormwater treatment during infiltration and its effect on shallow urban groundwater quality in the Cape Flats Aquifer: field study and soil-column experiments

Thesis / Dissertation

2025

Publisher

University of Cape Town

Department

Department of Civil Engineering

Faculty

Faculty of Engineering and the Built Environment

Abstract

In 2019 the City of Cape Town published a Water Strategy following a 3-year drought during which time the city nearly ran out of potable water. The Strategy includes among its five commitments a commitment to diversify water supply sources and a commitment to become a water sensitive city. Stormwater harvesting and recharge of the Cape Flats Aquifer are steps toward both commitments. This study formed part of a larger transdisciplinary research project where a stormwater detention pond in Mitchell's Plain, Cape Town was retrofitted with an infiltration swale and used as a study site. The aims of the study were to evaluate the potential for stormwater treatment during infiltration through the soil, and to investigate whether stormwater infiltration could result in deterioration of groundwater quality in the underlying Cape Flats Aquifer. Research was conducted in the field by monitoring stormwater and groundwater quality at the study site, and in the laboratory using large-scale soil column experiments. During the field study stormwater contaminants were detected at concentrations typical of an urban residential catchment with mean values of dissolved ammonium (0.35 mg/L), dissolved nitrate (0.7 mg/L) dissolved phosphorous (0.16 mg/L), dissolved aluminium (36 μg/L), dissolved zinc (81 μg/L), and total organic carbon (25 mg/L). Groundwater at the study site was shallow ranging from 2.4 to 3.7 m below ground level of the catchment and 0.1 to 1.4 m below the infiltration swale, although it remained below 0.5 m below the swale for most of the study period. Background groundwater at the study site was high in nitrate (mean 15.2 mg/L), low in total organic carbon (mean 2.7 mg/L) and low in dissolved oxygen (mean 1.4 mg/L) while the stormwater had a mean dissolved oxygen concentration of 7.4 mg/L. The field study showed that the infiltration processes resulted in an overall mean reduction in the concentrations of dissolved contaminants ammonium, phosphorus, aluminium, zinc, and total organic carbon, by 92%, 92%, 83%, 91%, and 84% respectively after accounting for dilution effects. The infiltration process also resulted in consumption of 54% of the stormwater dissolved oxygen and 37% of the groundwater nitrate after accounting for dilution effects. Furthermore, the mean groundwater nitrate concentration in the background groundwater was significantly higher (p<0.001) than nitrate concentrations in all other wells. This suggested that the organic carbon input from the stormwater and soil profile stimulated microbial activity which led to the consumption of the available oxygen and groundwater denitrification. Some localised and time restrained mobilisation of iron and manganese was found, however significant increase from background contaminant concentrations was found in groundwater at the study site. In addition, stormwater outflow from the study site had lower concentrations of inorganic nutrients (N and P), organic carbon and most metals compared to the incoming stormwater. Column experiments were conducted using synthetic stormwater with concentrations of dissolved contaminants typical of a residential area, and five times higher concentrations to represent poorer stormwater quality which may be encountered in other parts of the Cape Flats. Results showed that 0.5 m of soil was able to effectively retain P, Zn, Pb, Ni, Cr and Cu during infiltration of synthetic stormwater. Removal of total nitrogen was driven by denitrification and improved in the saturated zone. Denitrification during infiltration was improved by the addition of organic carbon to the synthetic stormwater, however this also resulted in slightly higher concentrations of geogenic contaminants arsenic (13.6 vs 6.9 μg/L), iron (843 vs 657 μg/L), and manganese (21.4 vs 18.5 μg/L) in the column effluent. The concentrations used did not significantly influence the effluent concentrations of soil adsorbable contaminants (P, Zn, Pb, Ni, Cr and Cu) after infiltration through the soil or the concentration to total nitrogen in the saturated zone. Based on this research enhanced infiltration of stormwater from relatively small (neighbourhood scale) formal residential catchments such as the one in this study is unlikely to pose a significant threat to groundwater quality, therefore this should be encouraged. Stormwater with higher contaminant concentrations may also be suitable for infiltration, however further research is required to determine the concentration limits and determining factors for suitability of a particular site.

Keywords

Stormwater

Cape Flats

Reference:

Collections

Masters

Full item page