Reducing the loss of life from flooding in the uMngeni river basin

It’s mid to late summer in South Africa, the time when we might expect floods.

A few years ago, a friend was recounting memories from his childhood in Greyling Street, Pietermaritzburg. His earliest memory was of being woken by his father one night in 1943 and, when he placed his feet on the floor, he encountered water. He and the rest of the family moved rapidly onto the roof of their home as the Dorpspruit came down in flood. It took a little while for the implications of what he was saying to sink in. If Greyling Street was flooded then the entire showgrounds area, the railway shunting yard, several electrical substations and various road systems would also have been under water. Chaos would have reigned.

Fast forward to the 1987 floods, and there is a mark high on the wall of the local scuba diving club which indicates the level to which the water rose. Kershaw Park and surrounds were entirely under water. Fast forward again to Christmas Day 1995; I recall in graphic detail the carnage left by the flooding of the Slangspruit in Edendale – 160 people died that day.

Occasional flooding, with its accompanying death and destruction, is a very real issue in Pietermaritzburg and it is not surprising – we are in a river basin. Are we prepared for these floods and are we able to deal with them when they occur?

Climate change models indicate that KwaZulu-Natal is getting hotter and wetter and the risk of extreme rainfall events is increasing. Therefore, we are likely to see bigger and more intense floods more often.

Also, our catchment areas are no longer pristine grasslands and indigenous forests. They comprise lots of hard surfaces resulting in more rapid run-off of rainfall. This further increases the intensity of flooding. In addition, our rivers are crossed by numerous bridges. These slow the flow rate but cause back-flooding. You won’t be swept away but you will still drown.

Following the Christmas Day flood we worked closely with the municipality. Our specific focus was to reduce the loss of life associated with flooding – ensuring people were out of harm’s way. Our main message was simple – stop people building their homes on floodplains. That is easier said than done. Floodplains are not settled precisely because they are subject to flooding but are particularly enticing to poor informal settlers with nowhere else to go.

I did a quick Google Earth tour of Edendale and was encouraged to see that, since 1995, very little settlement in low-lying areas has taken place. I was particularly heartened to see that nobody had resettled in the area where most of the deaths occurred during the Christmas Day flood.

But then I moved downstream and encountered, on the Dorpspruit, the Jika Joe informal settlement and its associated temporary accommodation. If, or rather when, we have a repeat of the Dorpspruit flood, Jika Joe will be submerged. People will die and many will catch disease as the floodwaters will be laced with raw sewage. It is imperative that these people, the most vulnerable and marginalized of Pietermaritzburg’s community, be moved out of harm’s way.

I understand that a flood early warning system is being developed by the district municipality and Umgeni Water. That is encouraging but will not entirely solve the problem. The time between a storm in the Winterskloof valley and the floodwaters reaching town is so short as to make a warning system very difficult to implement.

Duncan Hay is associate research fellow at UKZN and the executive director of the Institute of Natural Resources. This blogpost is an edited version of an article that was first published in The Witness on 11 January 2018.