2014 Aurecon recruitment banner


Biological filtration for sustainable treatment of groundwater

Preekstoel Water Treatment Works

Aurecon Associate for Water Engineering, Geoff du Toit, discusses using biological filtration for the sustainable treatment of groundwater that is rich in soluble iron and manganese in the Overstrand Municipality, South Africa.

Iron and manganese is removed via biological oxidation and filtration in lieu of conventional chemical precipitation.

The biological oxidation process has been demonstrated internationally, but this is the first application of its kind in South Africa. The raw water iron and manganese concentrations allowed for in the design were 2mgFe/L and 0.5mgMn/L. These were significantly lower than the actual values, which could only be established once the boreholes had been equipped at part of the project and run at design steady state for a considerable time. These actual values ranged up to 30 mgFe/L and 5 mgMn/L; significantly higher than any of the international case studies that have been reported.

As this process is less well known in water treatment, and only applied in a few locations around the world, what was the risk of testing the Biological Filtration process in South Africa and not working?

The introduction of new technologies will always be accompanied by risk. Our role as specialist consultant and advisor to our client is to help them mitigate and manage that risk, in essence that is what engineers do. We started with a desktop study of the process and reviewed all the literature we could find, and in particular any case studies which could assist us. At the same time, a comprehensive water testing and characterisation process was undertaken to better understand the water we would be treating. This work was done in collaboration with a team of geohydrologists, whose study of the various aquifers provided critical information that overlapped with our scope. We then followed this with a field trip to visit sites where the technology is being used successfully and to meet the operational staff who are responsible for these plants. This was done to inform us of design parameters as well as operational challenges and constraints. Having acquired this information we commissioned a pilot plant of the process on the actual borehole water in order to confirm there were no, as yet, undetermined factors which would inhibit the process. As a contingency we had allowed in our design of the biofiltration plant the ability to easily retrofit chemical oxidation should the biological process not produce the required quality. Thus a systematic and comprehensive system of risk mitigation measures were put in place, at relatively little additional cost to the client. This provided the client peace of mind and the reassurance that we had undertaken due diligence to limit his exposure to risk. Often this upfront work is omitted by clients in order to save money, only to pay heavily later as risks materialise which should have been identified and mitigated at the outset.

Why was a pilot biological iron and manganese removal plant placed at another Water Treatment Plant (WTW), away from the wells that needed to be treated?

At the time, there was no convenient site to locate the pilot plant at Preekstoel WTW, as the boreholes feeding the works needed to stay in operation and the water was chemically dosed at a booster pumpstation some distance from the WTW. As a result, it would have been difficult to build and operate the pilot plant remote from the treatment works. The Kleinmond WTW draws water from a sampling borehole which is drilled into the same greater aquifer as the boreholes that would feed the Preekstoel WTW and allowed for an ideal test site, both from a security and topography perspective. The geohydrologists had advised that the water characteristics would be similar between the two sites, and any trace metals (which may cause problems) would be common. In hindsight, now that we are much more familiar with the waters, we should have reconsidered making a plan to locate the pilot plant at the Preekstoel WTW.

What results did the client and Aurecon observe at the pilot plant to commission a full-scale plant at the appropriate Water Treatment Plant?

The pilot plant results were not excellent, as we had trouble controlling the pH going into the manganese biofilter. However, we were getting almost complete iron removal which gave us confidence that the process was working, and if iron removal worked, surely with more accurate control we would be able to get the manganese removal to work, too. These results were discussed with the client, and a joint decision was taken to proceed at full-scale, armed with the information that we had gathered.

After how long did the full-scale plant perform, and did it exceed your expectations? Were there any surprises?

The commissioning process took much longer than expected, as our commissioning engineer insisted on cleaning the filters daily - which, in hindsight, inhibited the colonisation of the media by the bacteria. Once the colonisation started, the process performed reasonably well; but when spikes in the iron or manganese concentrations were experienced, we had break-through of iron and manganese (bearing in mind that the concentrations we were receiving of >30mgFe/L and >5mgMn/L are significantly higher than we had expected or have been proven elsewhere). However, over time the biology acclimatised and was able to remove greater quantities of iron and manganese until break-through no longer occurred. It took approximately three months to get complete iron and manganese removal of the lower concentration boreholes. During the commissioning period provision was made to chemically oxidise the water and retain it in the system, but operational challenges prevented this solution being successfully implemented; and the operators decided to waste the water, instead. The partially treated water was discharged to the sewer and to the wastewater treatment works.

Do you think this process could become more popular and applied to more locations where there are similar challenges?

The process is already popular in some parts of the world where it is appropriate. It has the advantages of being relatively robust, having little maintenance requirements and being cost-effective. The process is inhibited by certain water characteristics like the presence of Zn or Cu, any heavy metal, or elevated concentrations of H2S. It is, thus, not appropriate under all circumstances but should certainly be considered as an option where groundwater with elevated concentrations of iron and manganese are encountered.

To top