Flooding of the Somerset West Hospital not at all surprising

18 November 2013

Somerset West’s Vergelegen Hospital took hydrotherapy to a new high this week when the complex flooded and resulted in it having to be closed.  Given that the hospital was built in the flow path of a former arm of the Lourens River – and the warnings  that predated the location of the hospital in this spot – this was not an unexpected event.  It just needed a big enough flood event – which Nature provided on Friday 15 November 2013 – when six inches of rain fell within 8 hours.

Historical map of Somerset West showing the river splitting into two arms just upstream of the town as it was then.

Historical map of Somerset West showing the river splitting into two arms just upstream of the town as it was then.

The Lourens River used to have two arms, the river having historically divided into an an inverted ‘Y’ just upstream of where the hospital is located.  The split can be seen in the historical map above.  When the hospital foundations were being excavated it was very clear that it was in an old riverbed (the eastern arm), as was also shown to be the case during excavations for the retirement village across the road just to the southeast.  More to the point, the evidence showed that a substantial river had come this way in the past, one that would not be able to fit into the western arm alone.  In fact, the historic bridge can only allow one-third to one-fifth of a major summer storm discharge to pass through – guess where the rest goes…

Herewith an extract from Rivers and Wetlands of Cape Town (Brown and Magoba, eds., pg 289), which contains a reprint of a picture I took during the conversion of the Morkel farm de Bos into a business park and retirement village (Heritage Park):

“The second flood channel had not gone unnoticed by the engineers, and the 1995 team led by Professor Rooseboom had earmarked a route for a flood alleviation canal for the Lourens River, which followed the course of the original [second] channel for some of its way.  Unfortunately the land was never reserved [surprise surprise] and by 2000, a Medi-Clinic and an old-age home occupied the space where the Lourens River flood channel originally ran – hence the scheme could not be implemented”.

Somerset West is now reaping the rewards of bad planning decisions.  Rivers have a habit of trying to get back to where they want to flow, no matter how much engineering is placed in their path!  The following graph from one of our weather stations showed how the rain started suddenly and with intent at 16:30 on Friday afternoon:

Rainfall rate, Somerset West, Friday November 15th, 2013

Rainfall rate, Somerset West, Friday November 15th, 2013

Over the past 29 years, November 2013 (154 mm so far and more due tomorrow) has now outclassed the previous 2009 total of 146 mm (difference being that in 2009 the rain did not all come in one short burst). Over the 29 year period, the November average is 38 mm.  2013 (1091 mm so far) is also not yet the wettest year in 29, this accolade still held (but not for very much longer) by 2007 (1100 mm) [Rainfall data are for the Helderberg College area, measured using an automatic tipping bucket rainguage].

Comparison of Somerset West monthly average rainfall totals vs 2013 values (Source: DHEC).

Comparison of Somerset West monthly average rainfall totals vs 2013 values (Source: DHEC).

Somerset West Annual Rainfall Totals (1984-2013) Source: DHEC

Somerset West Annual Rainfall Totals (1984-2013) Source: DHEC


Stabilization structures (gabions) collapsing in the Lourens River following the mid-2013 high flow event (Photo: Bill Harding)

Stabilization structures (gabions) collapsing in the Lourens River following the mid-2013 high flow event (Photo: Bill Harding)

10 Responses to Flooding of the Somerset West Hospital not at all surprising

  1. Christa says:

    Wow, absolute ongelooflike feite! Mens kan nie glo dat ingeligte mense so ‘n dom besluit geneem het wat vandag verreikende gevolge het nie.

    • Bill Harding says:

      This type of problem is more prevalent than many may perhaps realise. In this particular case, as I understand the situation, the local authority was not happy with the findings of the first hydraulic study (which indicated the need for a flood relief channel to mimic the role of the original eastern arm of the river), so they ignored it and commissioned a second. As it confirmed the findings of the first, they had a third go and yet again the same conclusion was reached! As a result Somerset West has its own “waterfront” at times like this.

      Currently the hospital parking is a mass of soggy carpets and a carport area full of hospital beds! I expect that the renovation and carpet providers are going to have a good Xmas as a result of this 😉

  2. Arthur Chapman says:

    Yes, building in flood plains is surprisingly common and the responsible people exhibit a strange disregard for the inevitable. It also appears that the return periods for extreme rainfall is shortening, for most of the Western Cape at least. We (de Waal, Chapman, Kemp) have a paper in review which we think shows this for the Western Cape. We argue against stationarity in the rainfall records and suggest that flood line evaluations may be out of date.

  3. Hans Nieuwmeyer says:

    Comment from Mr Hans Nieuwmeyer, former Chief Engineer at AECI (Paardevlei):

    In the 1980’s I was the Chief Engineer in AECI and we, [together] with other owners of river bank properties became alarmed about the planning being done and the loss of control of the owners of the river banks. We formed the “Lourens River Conservation Society (the LRCS).

    We as owners first based our arguments on the High Court ruling dated in the 1930’s but found that the municipalities (Somerset West and Strand) were not interested.

    The Society then contracted, at the member’s expense, a [recognized] river expert to evaluate the river planning and to provide advice for future action to avoid disasters. The expert provided us with a detailed report and by his calculations concluded that severe floods would occur about 3 times a century. (The AECI factory was flooded in the 1920’s).

    The expert in the report stated that the planned Hospital and Old Age home are in the wrong place and will be flooded.

    A copy of the report was sent to the two municipalities mentioned above but they took no action. There should be copies of the report in the municipal archives. The copy of the report and the minutes of the meetings, held by AECI, have been sent to the AECI museum in Modderfontein.

    • Bill Harding says:

      The work commissioned by the LRCS came to the same conclusion as that made by three other experts. In all, the City of Cape Town commissioned three studies to assess the Lourens River flooding risk. After the first by Albert Rooseboom, they contracted Drs Guy Pegram and David Pitman to provide their own assessments, both of which concurred with Prof Rooseboom’s findings – probably to the chagrin of the City of Cape Town. All the while the costs of the obvious flood alleviation works were becoming increasingly costly. The City of Cape Town cannot now claim to not have known – they had the findings of three of the best hydraulic engineers in South Africa – but they didn’t like what the reports said. They also ignored the findings of an expert contracted by the LRCS. Allowing hospitals and retirement villages to be built in the floodplain AND channel of a major river – which has had half of its capacity removed – is probably not the brightest decision to make.

      But, even two weeks after the event, the City of Cape Town still casts red herrings by admitting that the hospital has been built in a floodplain but creates an excuse: [back then] “we thought about stormwater management differently”. What absolute rubbish. Building in a floodplain has always been a stupid decision, in this case the City had the best professional advice available, yet ignored it.

      Looking at the debris lines on the Hathersage Farm upstream of Bizweni, it is clear where the river decided to go overland, pretty much along the alignment of its now missing left arm.

  4. gerry brand says:

    This revelation will no doubt leave residents with null and void cover in terms with their insurers. Would the individuals and organisations responsible for this cover up be ultimately liable for future loss of life and property ?

  5. G.P.Robertson says:

    I would like to know if anyone has a copy of a report on the Lourens River written by Dr Dave Beaumont – who used to work for [the firm] Hill Kaplan Scott [now GIBB Consulting Engineers). At the time he was very concerned with the dams (24) that are on the estate above the river.

    Can anyone tell me if the [November 2013] flood was a 10,20,50 or 100 year flood?

  6. Buks Kotze says:

    I live in Somerset West, in the area named Kalamunda (Geering Street). We have been flooded 3 times, (actually 4 times if counting the river bursting its banks on 14/15 Nov 2003) since Oct 2012. Extensive damage have been sustained in Protea, Meadow Lane and Geering Streets.

    With the last flood, if the wall between the open erf and Graceland Place did not collapse, acting as sluice, even more damage would have been done.

    Myself and other residents feel that we should stand together in protecting our families and homes. Also since it seems as if the most of Heathersage Farm is going to be developed into a school (1,800 students) and 3 residential complexes – the study is in the process of being conducted.

    • Bill Harding says:

      I have had several comments from property owners who have now discovered that they live below the 1:50 year flood line – some substantially below it. Some of these owners foresee a loss in the value of their property given the not insubstantial flooding they recently experienced and may well experience again. The November 2013 event was a 1:50 year magnitude, imagine what a bigger storm event would achieve?

      The area you mention lies downstream of Hathersage Farm, on which property it is likely that the Lourens River may have divided into two channels. Post-flood evidence suggests that the river went overland along the presumed southeast alignment, straight towards Meadow Lane and the Bizweni area.

      I would assume that the onus on the City of Cape Town to ensure the protection of life and property, vis a vis the completion of the Lourens River Flood Management Scheme, remains as strong now as it did back in the 90s when the scheme was planned.

Leave a Reply

Your email address will not be published. Required fields are marked *