Pollution

Point sources of pollution in an area relate to land uses like cemeteries, landfill/waste disposal sites and a more recently installed power station which runs on diesel fuel. Non-point sources include informal settlements which are un-serviced, small scale farming practices and industrial activities. Emerging contaminants only came to light through a special study in Atlantis, indicating that trace organic compounds and pharmaceuticals are present within the system.

Water Quality Improvement and its Monitoring

Water quality is measured intensively, both in the aquifer and in the urban water management system. The aquifer water quality is assessed by sampling boreholes within the developed wellfields as well as monitoring points spread throughout the Atlantis water management area on a five-week cycle. For the urban water management system, critical monitoring points (control points) were selected that take into account the various steps within the MAR system at which significant changes could occur (Tredoux et al., 2009a). These critical control points are explained in Section 3.5. The full extent of water quality monitoring of the area is shown in Figure 10.

Electrical conductivity in the Witzand wellfield ranges between 40 and 120 mS/m. The northern part contains more calcium bicarbonate (CaHCO₃) water while the southern parts contain more sodium chloride (NaCl) rich water. Water abstracted from this wellfield often has to undergo softening to remove temporary hardness. Several of the boreholes in the southern most parts of the wellfield have been decommissioned to avoid abstraction of water with high salinity and sulfate content (Bugan et al., 2016).

Electrical conductivity within the Silwerstroom wellfield for the last few decades ranges between 40 and 180 mS/m with an average of about 80 mS/m. This wellfield has lower natural calcium (Ca) levels compared to the Witzand wellfield. Only electrical conductivity and iron (which can occur at levels of 0.1 to 0.3 mg/L) fall outside of the drinking water standards in South Africa (SANS, 2011; DWAF, 1998), but these concentrations are removed in the final softening treatment.

Figure 10 – Monitoring and production borehole locations throughout the Atlantis system (Bugan et al., 2016).

The monitoring system has shown that potassium (K) and nitrate (NO₃^–), which are frequently used as tracers of wastewater, are elevated in the wastewater. However, all parameter concentrations are reduced considerably as water moves through the water management system. This demonstrates the efficiency of soil as a natural filter during infiltration of water.

Of particular importance for human health is the impact of the treatment system on bacteria. Stormwater tends to have higher microbial counts than the treated wastewater effluent. Bacterial counts at various points in the system illustrate the importance of the subsurface passage as a safety barrier in the system (Figure 11). Tredoux and others (2009a) report that not only do indicator organisms like E. coli decrease substantially as water moves through the system, but pathogens, including viruses, follow a similar pattern of logarithmic reduction, providing the necessary safety margins for the recycling system.

Figure 11 – Bacterial removal in the recycling system of Atlantis (Tredoux et al., 2009a).

Emerging contaminants are not routinely monitored, and occur at very low concentrations. An order of magnitude reduction in concentrations of measured parameters was observed as they move from surface to groundwater to abstracted water to the distribution point. Although some degree of degradation or adsorption on the geological material in the subsurface is expected, it would seem that most of the reduction could be ascribed to dilution. The results demonstrate subsurface passage of water as a successful method to reduce the health risk for potable reuse (Tredoux et al., 2009a).

Since the 1990s, borehole clogging has been a recurring problem in the Atlantis wellfields, which resulted in reduced borehole yields. The occurrence of borehole clogging may be attributed to (Bugan et.al., 2016):

• the presence of naturally available iron and manganese in the aquifer (Smith, 2006);
• the occurrence of natural soil bacteria which accumulate iron and manganese due to fluctuating groundwater levels and dissolved oxygen concentrations;
• borehole construction; and,
• the pumping schedule (e.g., over pumping during drought conditions).

In-situ iron removal to prevent iron clogging in production boreholes was recently investigated in the Atlantis Aquifer (Robey and Tredoux, 2013). Ozone was generated on-site and injected back into the aquifer by means of well points. Results of the experiment showed that iron and manganese concentrations in the dissolved phase were reduced considerably. A full-scale implementation of this experiment is currently underway in the area. Successful implementation of this method could significantly reduce the cost of water treatment and production well maintenance costs.