Brandwacht River: Stabilization of severe river bank erosion

Site description and problems

The Brandwacht River flows southwards from the Outeniqua mountains, and discharges into the sea a few kilometres from Hartenbos, near George in the Eden District of the Western Cape. There are many small farmers along the river producing a variety of crops. As can be seen on the historical Google Earth images, a significant change in alien vegetation growing in the river and along it’s banks took place after the 2006 flood. The Google Earth images show clearly a marked increase in the presence of black wattle (Acacia mearnsii) in the period 2005 to 2011 (there are however many other species of alien vegetation such as the Spanish Reed (Arundo donax ) and others ). The 2011 Google Earth image also shows the onset of river bank erosion at sites opposite where the black wattle is growing (this erosion occurred most likely mostly during the 2008 flood events). The 2013 Google Earth image (post the 2013 flood) shows that the bank erosion has become much more severe and that in places the right bank has moved completely to where the left bank was 10 years previously. Not only was agricultural soil being lost on the bank of the river, but large quantities of sediment were washing downstream as a result of the impact of the alien vegetation.

After inspections in the field, as well as that of historical satellite imagery, it was clear that once the river was disturbed by bulldozing and the flourishing of invasive alien vegetation (probably starting in the late 20th century), the natural stability of the river became ever more degraded as one progressed down the river. A 3-kilometre-long stretch of river was selected as the target area because that was where the instability was seen to be developing – in other words there were zones of vigorous alien vegetation development as well as zones where serious soil loss and mobilization of sediment load were taking place. With the funds available it was not possible to address the river further down, but it was believed that by reducing the sediment flow down the river from the target area, the whole riparian community downstream would benefit from the work.

Objectives for undertaking rehabilitation

The primary objective of the work was to prevent the abnormal movement of sediment down the river (in other words sediment that was mobilized by the ingress of invasive alien vegetation), and so protect the agricultural production potential of land downstream of the site. There are a host of secondary benefits of the work including: –

  • The protection of agricultural production potential on the site itself (by preventing the washing away of the river banks).
  • The provision of a work opportunity for unemployed locals during the project, and for the maintenance of the river after the project.
  • The protection of the natural environment by the removal of alien vegetation and the partial restoration of the natural flora at the worksite.
  • The protection of the local road and bridge infrastructure, which impacts on the local agricultural economy.

Intervention that was undertaken

 A detailed topographical survey of the 3-kilometre-long stretch of river was carried out, and a hydrological study of the catchment was done. This information was used to determine flood levels, to design a rehabilitated river channel (along the route that existed about 10 years previously), and to design groyne structures to support the rehabilitated channel. The width and shape of the rehabilitated channel was designed such flow velocities during floods were mild enough to not promote erosion, especially once the indigenous wetland vegetation had been re-established.

Groynes were chosen as river training structures (as opposed to riprap or other longitudinal protection), partly because it is a lot more economical, but also because the use of groynes promotes a wider and slower flowing watercourse during floods, and the spaces between the groynes are used for trapping sediment and the establishment of indigenous wetland vegetation. The height of the groyne structures above the river bed were kept low so that large sections of the groynes would be overtopped during moderate floods and double the river’s flow width.

Rock filled gabions were selected as the construction material for two reasons. The first being that the material is flexible enough to accommodate foundation movements should that occur during an extreme flood. The second being that it is a material that is easy to work with, and this facilitates the prescription that the contractor’s labour force must be comprised of at least 70% local persons.

The planning of the work was completed in 2016. Construction began in March 2017 and was completed in August 2018. The total project cost was in the order of R24 m. A total of 27 groyne structures were constructed and for that just over 9 000 m3 of rock was packed into gabion baskets and mattresses.

Specific design and construction issues

The design technique for groynes developed in the Western Cape has been used far more often for boulder bed rivers, and not so much for sand-bed rivers such as the Brandwacht River. It would be expected that the sediment of a sand-bed river would be mobilized to a greater depth during a flood than would be a cobble-bed river. To make sure that the tips of the groynes at Brandwacht will stable during floods, it was therefore essential to keep the foundations at least one meter below the lowest part of the pools in the surveyed pool-and-riffle sequence. This caused problems during construction because the anchor block under the mattress around the foundation of the groyne is at least 2 meters below the river bed.

The contractor found that conventional techniques of pumping water out of the excavation so that the anchor block could be installed just did not work as the inflow of sub-soil water was just too strong. It was then decided to employ a curtain of well-points around the excavation and a very large pump, which dropped the water table enough to enable the construction to go ahead as planned.

Outcomes of the rehabilitation action

Post the completion of construction, several small floods have been experienced, but nothing significant to test the project yet. The indigenous wetland vegetation has (re-established and naturally appearing) is establishing very well. The banks of the river are now clearly stable, and it is expected that the project will achieve its goals with ease.



Survey and design: Western Cape Department of Agriculture

Funding: National Department of Agriculture – Disaster Relief

Main contractor: B-Waitabasa


Project management: CASIDRA