Control options for river water quality improvement: a case study of TDS and inorganic nitrogen in the Crocodile River (South Africa)

Using a simple conceptual dynamic river water quality model, the effects of different basin-wide water quality management options on downstream water quality improvements in a semi-arid river, the Crocodile River (South Africa) were investigated. When a river is impacted by high rates of freshwater withdrawal (in its upstream reaches), and receives polluted side-stream inflows and wastewater effluent discharges (in the middle reaches), river water quality can deteriorate seriously over time. This study focused on two water quality problems: Progressive increases in the concentrations of total dissolved solids (TDS) as a measure of salinity, and the concentrations of nitrate-plus-nitrite and ammonia (as inorganic nitrogen) as a measure of eutrophication. Based on a lowflow analysis for the period prior to construction of the Kwena Dam (1960 to 1979), the 7d low flows that could be expected to occur every 10 years (7Q10) are generally very low (< 0.5 m3·s-1), both in the upstream (Montrose Weir) and the downstream (Kruger National Park) sections of the Crocodile River. During such critical periods of low river flow, very low effluent standard limits would be required to prevent adverse river water quality. However, these options are not economically feasible. Furthermore, inflows from the highly polluted tributary stream, the Kaap River, which drains an area where considerable gold mining takes place, govern water quality in the Crocodile River downstream of the Crocodile-Kaap confluence. Subsequently, two additional water quality control options (setting limits for maximum water withdrawal and low-flow augmentation) were analysed. The results show that a decrease in maximum water withdrawal could reduce the TDS concentration. Furthermore, controlling water release patterns from a dam at the Montrose Weir can have a remarkably positive effect on the downstream river water quality. On the basis of the 1989/90 monitoring data, a minimum flow of 5 m3·s-1 at the Montrose Weir can reduce concentrations of TDS and ammonia nitrogen by about 20% and 60%, respectively, in the Kruger National Park (at the downstream point of the considered river). However, this management option does not reduce nitrate nitrogen concentrations. The proposed model used in this study is relatively simple and can be used as a tool for the evaluation of short-term (monthly) basin-wide water quality management options. Keywords: Dynamic model, Flow regulation, Water quality management, Tank in series model (WaterSA: 2003 29(2): 209-218)