For the past three years, the water in South Lake, North Lake and False Bay has been coming back under 6 parts per thousand. Seawater is 35. Below 3, according to the two scientists who have been watching those readings longest, the marine fish in Lake St Lucia start dying in numbers nobody will be able to count properly.
That is the whole story in one measurement, and it is worth sitting with for a second, because almost everything else about this argument is contested and that figure is not. Lake St Lucia, on the KwaZulu-Natal coast of South Africa, is the largest estuarine lake system in Africa. It covers roughly 35,000 hectares and is on average less than a metre deep. It accounts for about half of all the estuarine surface area in South Africa, which means it is the nursery for a very large share of the fish that eventually turn up on hooks and in nets along a long stretch of that coastline. It sits inside the iSimangaliso Wetland Park, inscribed as a World Heritage Site in 1999, and the local name translates roughly as place of wonder.
Professor Alan Whitfield has been studying it, and estuaries like it, since 1978. That was the year he joined what was then the JLB Smith Institute of Ichthyology, which became the South African Institute for Aquatic Biodiversity. He retired from it in 2018 and holds the title of emeritus chief scientist there, along with an honorary associate professorship at Rhodes University. Four degrees, all in zoology or ichthyology, the last of them a doctorate of science awarded in 2011. If you wanted to design a person to tell you what has happened to South African estuaries over the last half century, you would end up with something close to him.
Speaking at a seminar in Makhanda during Heritage Month, Whitfield said the lake is about to undergo what he calls an ecosystem flip. Within the next decade, on his account, the estuary becomes a freshwater wetland, the connection to the sea effectively ends, and the marine fish trapped on the wrong side of it die. He put the number at thousands and thousands, probably hundreds of thousands, and the hedge in that phrasing is his own, not something added afterwards. What replaces the system, he said, is a much simpler place with far less variety in it, and in his opinion not a place that warrants World Heritage status at all.
The iSimangaliso Wetland Park Authority, which is responsible for the site, has said in public that it does not share the view that the St Lucia estuary is approaching an imminent ecological tipping point. That position was given to a South African newspaper by a spokesperson who was not named. When a second newsroom approached the authority months later, it did not respond before publication. So one of the two parties in this disagreement is on the record through an unnamed official and has since stopped answering questions, which is worth saying plainly because it shapes what a reader can and cannot judge here.
It would be easy to write that as a story about a brave scientist and an institution with something to hide. It is not that, and the reason it is not is more interesting than the simple version.
Start with how the lake got here, because the mechanism is the argument. The uMfolozi is the biggest single source of water into the system. In 1952 its mouth was diverted away from the lake, for two reasons that were both perfectly rational at the time. The river carries an enormous sediment load out of a catchment that has been getting steadily more degraded, and farmers had been planting sugar cane on the floodplain since around 1911 and did not want to be flooded when the river mouth closed. So the river was sent elsewhere, dredgers were brought in through the 1950s and 1960s, hard structures were built at the estuary mouth, and for sixty years the mouth was opened artificially whenever it closed.
Then two things happened. In 1984 Cyclone Domoina destroyed the structures at the mouth and wrecked a canal that had been dug in the late 1970s to feed clean fresh water into the lake, and the canal was never repaired. And in the early 2000s, deprived of its largest source of fresh water, the lake came close to drying out completely for the first time anyone could remember. Parts of it dried out again in 2016. Dry means hypersaline first, and hypersaline kills things too.
So in July 2012 the uMfolozi was reconnected to the estuary, with funding routed through the World Bank’s Global Environment Facility. It looked like a correction of a seventy year old mistake. Whitfield has said that at the time he and his colleagues thought they were finally on the right track.
What came down the reconnected river was sediment. The mouth then stayed closed almost continuously from the beginning of 2015 to the end of 2020, and during those six years silt and clay piled up in the estuary and in the long channel called the Narrows that links lake to sea. Reeds moved out from the banks into the narrowing channel. The abandoned link canal from the 1970s, which was supposed to be decommissioned, got broken into by uMfolozi floods and now delivers sediment straight into the estuary near a spot called Honeymoon Bend, where the deposits and the reeds together have built a constriction that stops tidal water pushing upstream. Further in, vegetation has choked the Potter’s Channel and Brodie’s Shallows area so thoroughly that lake water can only seep through it rather than flow.
That is the physical reason the salinity readings are what they are. Salt water cannot get in. And one of the clearest signs is something you can see rather than measure: freshwater reeds along the shoreline are thriving, and those reeds cannot survive water saltier than about 12 parts per thousand.
The fish have already started voting. Adult flathead mullet used to make an annual spawning run out of False Bay and North Lake, through South Lake and down the Narrows, in such numbers that crocodiles and flocks of pelicans gathered to meet them. That run appears to have been lost in recent decades. Nobody staged it. It simply stopped happening.
Whitfield and Dr Ricky Taylor, writing together, and with Professor Janine Adams and Caroline Fox in a paper in the African Journal of Aquatic Science, have set out three things they want done, in order. Block the two points where the uMfolozi has broken into the old link canal, and then dredge out the sediment that has built up at Honeymoon Bend. Recreate Potter’s Channel so lake water can actually move. And breach the combined mouth artificially if it stays closed for more than three years. They also want a proper hydro and ecological monitoring programme, which they say has been largely absent for the past decade, and that complaint is not a small one, because it means both sides of this argument are working with less data than they should have.
Here is where the simple version falls apart. Whitfield’s written case and his spoken one are not the same case. In the estuaries feature he and Taylor published with SAIAB earlier this year, they describe St Lucia as close to an ecological tipping point and then say, in as many words, that when it will occur they do not know, because it depends on how connected the channels are, on episodic river floods, on sea level rise, and on how fast sediment accumulates. The decade figure is not in the paper. It is what he said at a seminar. That matters for how a reader should hold it: the mechanism is measured, the direction is measured, and the timing is a projection by one scientist with a long record rather than a consensus finding.
The two sides are also not using the word tipping point to mean the same thing. Whitfield and Taylor use it geologically. St Lucia, they argue, has always been travelling from deep estuary towards coastal floodplain, freshwater pans and swamp, and the only question is how fast, with human decisions having compressed thousands of years of change into decades. On that reading the lake becomes an ephemeral shallow freshwater system that dries out entirely in long dry spells, there is no more seawater exchange, sediment keeps building, and the change is irreversible. The authority’s denial is about imminence. It says it does not accept that the estuary is approaching an imminent tipping point, and it says decisions about a World Heritage Site require long term monitoring, specialist investigations, environmental authorisations, independent expert review and observation of the ecosystem as it currently is. Those are two different propositions. Somebody can be right about a trajectory and wrong about a date, and an institution can be right to want authorisations before it sends in machinery and still be presiding over a decline.
And there is a third voice that makes the custodian versus scientist frame collapse altogether. Professor Anthony Forbes, a marine and estuarine researcher, has called the new proposals fiddling with nature. His argument is not that the lake is fine. It is that the sixty year artificial breaching policy that ran from 1952 to 2012 is precisely what left the lake with no fresh water and nearly dried it out in the early 2000s, and that the people now proposing new engineering are proposing a version of the thing that caused the last disaster. When the authority assisted a breach of the mouth early this decade after river flooding, a group of scientists, some of whom had worked at St Lucia since the 1970s, said publicly that it was done in apparent disregard of the authority’s own management strategy, which since 2012 had been to stop artificial breaching and let the estuary re establish its own dynamics. Forbes was among the names attached to that criticism.
So read those two episodes together. The authority was attacked by scientists for breaching the mouth. It is now being attacked by scientists for not being willing to intervene. Whitfield’s own recommendation includes artificial breaching under defined conditions. These are not two camps. This is a genuinely hard problem on which qualified people have been disagreeing for half a century, and the authority sits in the middle of it holding the permits.
The government has had a look too. A panel of experts was convened to examine the scientific basis for breaching the St Lucia mouth, and its report was released by the national environment minister a few years ago. It did not side with the people who called the breach a scandal. It found ecological reasons that supported opening the mouth, including how long it had been shut off from the sea, the sediment that had accumulated, invasive species, loss of biodiversity and a weakened nursery function. It recorded that after the opening the nursery function was restored, marine fish came back in numbers that mattered to tourism and recreational fishing, large animals returned to the mouth area, and the lake shifted from fresh towards brackish. It also recorded what did not happen: no change in sediment removal, no improvement with the reeds or the invasive species, no help for the mangroves, and no relief for the farmers upstream. Its recommendations were cautious. Maintenance breaching could continue only in clearly defined exceptional circumstances, there should be an intensive monitoring plan, there should be vegetation and sediment clearing in specific channels, and technical documents should be translated so that the people who live there can read them.
That report is the strongest thing in the authority’s hand, and it is also, read closely, an argument for something very like what Whitfield is asking for. Clear the channels. Monitor properly. Breach when the circumstances demand it. The disagreement that remains is about how close the deadline is and therefore how fast anyone has to move, and on that the authority has the slower clock.
One thing that should be said flatly, because the language around this story invites error: iSimangaliso has not been placed on the List of World Heritage in Danger. There is no UNESCO decision doing that. The site has state of conservation reports on file, and its boundaries were substantially modified last year when it was joined with Maputo National Park across the border, but in danger is a formal designation and it has not been applied here. What exists is scientists warning that it should be in danger, which is a different sentence.
Everything else that makes this place worth arguing about is still there, for now. More than 500 bird species. Nile crocodiles. The hippos, which are the reason a game reserve was declared here in 1896, when this was one of the last major refuges for hippo herds in the country. Pelicans, cormorants, terns, herons, gulls. A tour boat that takes visitors up the Narrows to look for them. Skiboats launching near the mouth. And out at sea to the south, on the Thukela Banks, a prawn fishery that was once very active and depends on an estuary nursery that is slowly closing.
Whitfield is not asking for the lake to be made permanent. He is not even arguing that the flip can be prevented forever, because his own written position is that the trajectory is geological and the destination is a swamp. What he is arguing is that people pushed the system down that road faster than it was going, and that people can slow it down again, and that the window for doing so is narrower than the people responsible for the place believe.
The last time anyone opened that mouth deliberately, they did it with backhoes, and they were criticised for it by scientists who had been working there since before the men on the machines were born. The time before that, the policy was to keep it open, and that nearly dried the lake out. Nobody in this story has a clean record, and nobody has an obviously right answer.
What there is, is a mouth that has spent most of the last ten years shut, a channel full of reeds, a canal nobody closed, three years of readings under 6 parts per thousand, and a man who started measuring this water in 1978 saying he thinks it has about ten years left as the thing he has spent his life describing. He could be wrong about the ten years. He has been wrong before and said so, out loud, about 2012. The fish on the freshwater side of that sandbar do not get to wait and find out who wins the argument.