An assessment of the critical source areas and transport pathways of diffuse pollution in the Umngeni Catchment, South Africa.

Abstract

The difficulty in locating and managing diffuse pollution sources and their transport pathways is one of the reasons for the continued degradation of surface water in South Africa. Dealing with this problem is complex, as the sources and transport pathways of the pollutants are often not known because of the diffuse nature of the pollution. This study demonstrates the constraints of conventional diffuse pollution assessment approaches in identifying the Critical Source Areas (CSAs) and transport pathways of diffuse pollution, as applied in the uMngeni Catchment, South Africa. The use of various risk-based modelling approaches are reviewed for identifying the risk of diffuse pollution generation and transportation across a catchment landscape. The Sensitive Catchment Integrated Modelling and Analysis Platform (SCIMAP) Model is a risk-based tool that was developed to give a spatial representation of diffuse pollution sources. In this study, the SCIMAP Model was applied to identify and prioritise the protection and control of nutrient CSAs and transport pathways within the uMngeni Catchment. The results of the study were displayed in a catchment scale web map. The hydrological connectivity risk in the catchment was higher in the high-lying western areas and lower in the middle-eastern areas. The upper and middle parts of the catchment that are dominated by commercial agriculture and built-up urban areas were identified as the most impactful CSAs for intervention. The results are immediately applicable to water managers in the catchment and are strongly linked to the investment efforts in ecological infrastructure. A walkover survey revealed that the SCIMAP Model was able to direct the CSA investigations to the nutrient sources at four out of five locations. The survey also revealed that the accuracy of the modelled transport pathways increased with an increase in the elevation difference. The sensitivity of the SCIMAP Model to input land cover weightings was assessed, using an objective function. A high sensitivity of the modelled high-risk areas was observed on the intermediate diffuse pollution risk map, and a slight sensitivity of the modelled high-risk areas on the final diffuse pollution risk map, when the input landcover weightings were increased and decreased by 5%, 10% and 15%. This implies that caution should be practised in the formulation of the input land cover weightings, as they are a potential source of error in the model outputs. It is concluded that SCIMAP is a valuable tool for identifying the CSAs and transport pathways of diffuse pollution in a catchment. The results of the model can better inform the management of diffuse pollution and guide investments in the protection of the ecological infrastructure in the uMngeni Catchment. However, the establishment of input land cover weightings is very important and should receive priority in similar studies in the future.