African rainforests vanished for 600 years, then bounced back—why?
Three thousand years ago, dense old-growth rainforests covered most of central Africa. But around 2,600 years ago, an event that ecologists call the Late Holocene Rainforest Crisis occurred, and the forests suddenly gave way to savannas dotted with islands of trees. Six hundred years later, the forests grew back almost as swiftly as they had vanished.
But for the last 20 years, paleoecologists have debated what caused the Rainforest Crisis. Most thought that the region’s climate changed, bringing either less annual rain or a longer dry season with a short but intense monsoon. That climate shift, many paleoecologists argue, devastated the rainforests of central Africa but created perfect conditions for savannas. But a new study proposes that humans may actually have been the culprits.
A question of timing
Around the time of the Rainforest Crisis, farmers from northern Africa started migrating southward, bringing with them an advanced culture of pearl-millet cultivation, ironworking, and palm-oil harvesting, all of which take a toll on the landscape. Those northern farmers spoke Bantu languages, which are still spoken by about 300 million people in Africa today.
The currently accepted version among paleoecologists is that warming sea-surface temperatures in the Gulf of Guinea caused a shift in the region’s monsoon cycle, leading to a longer, drier dry season, so people migrated south to farm millet on the open grasslands. But University of Potsdam paleoecologist Yannick Garcin and his colleagues, who just published a new study on the Rainforest Crisis, argue that it happened the other way around: people moved south into the rainforest and cleared land to plant millet. And when their population crashed 600 years later, the rainforest rebounded.
Most of the debate hinges on whether those millet farmers moved south during the Rainforest Crisis or before and on whether there’s evidence of climate shift at the same time. And among paleoecologists and archaeologists, the debate is heating up. If you’re not a paleoecologist, the whole debate may seem a little esoteric, but it’s a great example of the scientific process at work, with both sides presenting their own sets of evidence and debating which is most accurate and whose interpretation best fits the data.
Paleoarchaeologist Jean Maley and his colleagues published a paper in October 2017 in which they argued for the climate hypothesis, citing sediment layers found in lakes in Ghana, Gabon, and the Congo. These show evidence of increased erosion around 2,650 years ago—presumably thanks to more intense monsoon rains. That lines up well with the other evidence for the climate shift.
And an earlier study sampled pollen from the sediments at the bottom of Lake Victoria, which showed that the water level 2,200 years ago was much lower than it is today and that savanna had taken over land formerly shaded by rainforest canopies.
On the other hand, Garcin and his colleagues recently took a sediment core from Lake Barombi in Cameroon. The 12-meter-long cylinder of mud held 10,500 years’ worth of accumulated sediment layers, which contained microscopic bits of material called plant waxes. Plants secrete waxy mixtures of lipids to protect their outer cells, and these waxes can last for thousands of years in soil. Conveniently for paleoecologists, they record what ratio of hydrogen isotopes the plant got from its water, as well as how the plant handled carbon.
Woody plants like trees and shrubs obtain carbon differently from grasses, so their plant waxes end up with different ratios of the stable isotope carbon-13. Prior to 2,600 years ago, the plant waxes in Lake Barombi seemed to be from mostly trees and shrubs, exactly what you’d expect in a thriving rainforest. But within a century, carbon-13 ratios in the plant waxes washing into Lake Barombi started looking much more like grassland than forest, which matched the pollen data from other studies. After about 600 years, though, the carbon-14 signature of forests replaced the grasslands.
Crisis or cause?
Those findings more or less confirmed what the pollen studies had to say about the timing of the Rainforest Crisis, but it didn’t say anything about what caused the event. But the ratio of hydrogen isotopes in plant waxes can reveal information about climate because, on a scale of decades, those ratios generally line up with average annual rainfall. In the Lake Barombi sediment core, that evidence pointed to a long, gradual drying tend from 7,000 to 2,000 years ago, but there didn’t seem to be any sudden climate shift near the beginning of the Rainforest Crisis. In fact, according to the Lake Barombi data, the area was actually wetter during the Rainforest Crisis than it is now, and today it’s still mostly covered with rainforest.
Tiny shells preserved in sediment cores from the Gulf of Guinea didn’t contain any evidence of a change in sea surface temperature, according to Garcin and his colleagues.
But when they examined a database of 460 archaeological sites from around the region, they found that very few sites had been dated to earlier than 4000 years ago. Human activity seems to have started picking up in the region around then, and it really exploded around 2,600 years ago. Garcin and his colleagues say that’s evidence of a major population increase right around the beginning of the Rainforest Crisis.
They claim this study is clear evidence that people, not climate, caused the Late Holocene Rainforest Crisis. But not everyone is convinced. Maley told Ars Technica that the most important argument against the blame-the-humans hypothesis and in favor of a shift in climate is the sheer geographic scale of the Rainforest Crisis—it happened at nearly the same time from the Equator to the southern Sahara. Humans, on the other hand, didn’t migrate south into central Africa’s rainforests with same kind of synchronization.
In other words, the debate over what happened to central Africa’s Late Holocene rainforests is far from settled.
Science in progress
Paleoecologists, archaeologists, and even linguists are still weighing in with new lines of evidence, and it’s likely to be some time before they reach a consensus. Part of the uncertainty is due to the limited resolution of dating methods, whether it’s radiocarbon dating or counting layers in a sediment core.
For instance, some of the evidence for the climate hypothesis comes from pollen and diatoms from the bottom of nearby Lake Ossa. But Garcin and his colleagues claim that, due to aged carbon in the sediment itself, there’s about a 400-year uncertainty in dating those layers. And in Garcin’s data, the Rainforest Crisis appears to have started at Lake Barombi about 200 years before Bantu agricultural sites started turning up in the region.
“We answer that we are at the limit of the precision of the dating method used to resolve such a small lag,” said Garcin, “and since the Lake Barombi is at the northern edge of Central Africa, it may have witnessed first those human impacts compared to the rest of the whole region.”
One thing scientists on all sides agree on is that it’s important to understand the cause of past events like this, because reconstructing past climate events can help predict how humans, climate, and vulnerable ecosystems might interact in the future. And in some ways, the Rainforest Crisis is an encouraging story, because it means the rainforest can bounce back from deforestation.
“Rainforest ecosystems are very sensitive to disturbances but also resilient,” said Garcin.