Where exactly in Africa our species evolved is one of the hottest points of contention in human evolution.
Now a new study suggests Homo sapiens emerged in a single region in the south of the continent around 300,000 years ago.
To reach their verdict, a team of researchers from South Korea and Europe simulated the Earth's climate history back 2 million years, and paired it with ancient human bone and tool discoveries.
The study was published in Nature today — but not everyone agrees with its conclusions.
Study lead author Axel Timmermann, from the Institute for Basic Science Center for Climate Physics at Pusan University, said it's the first study to link archaeological evidence with long-range climate modelling to piece together a story of early human movement and evolution.
Ian Moffat, a geoarchaeologist from Flinders University who was not involved in the work, said the study took a broad look at the "engine room" of early human evolution: Africa, Europe and parts of Asia.
"These are the places that have historically defined the narrative of early human evolution.
"It's good that questions [about human evolution] are being asked using these rigorous approaches to the climate record."
Ancestral digs
The human family tree isn't a neat and tidy series of branches. It's a tangle of forks, offshoots and dead ends, with some limbs rejoining and intermingling multiple times over the millennia as different species interbred and migrated.
What compelled our human ancestors to wander into new territory was probably climate change, which could have made a home region less hospitable, or neighbouring areas easier to explore.
That said, nailing down what the climate was like at a site that was occupied by early humans is a tricky task.
Researchers can, for instance, date sediments and analyse bits and bobs like pollen lodged in the soil. Knowing what plants grew gives researchers an idea of the climate of the time.
But the qualities that make archaeological sites like caves so good at accumulating fossilised bones and tools — they're sheltered, dry, and can remain relatively undisturbed by the outside environment — means they don't accumulate much in the way of climate evidence, Dr Moffat said.
So instead of relying on sediments from archaeological digs to provide climate information, the new study used a computer model — the same type of model used in Intergovernmental Panel for Climate Change reports, Professor Timmermann said.
"It's been used for future climate change projections as well as, now, in our case, 2 million years into the past."
Celestial climate clockwork
The climate cycles modelled by the Professor Timmerman and his team are different to climate change as we know it today, which is driven by an influx of carbon emissions into the atmosphere.
Changes in climate that encouraged our ancient ancestors to move were driven by celestial objects.
Such astronomical climate cycles are called Milankovitch cycles, named for the Serbian polymath credited with first pondering their existence in the 1920s.
There are a few types of Milankovitch cycles, including one that's dictated by Earth's distance from the Sun.
As our planet orbits the Sun, it doesn't travel in a perfect circle. The gravitational heft of Jupiter and Saturn periodically stretches our trajectory into a vague oval shape.
This means Earth is further from the Sun some of the time, and receives less warmth.
Another Milankovitch cycle arises because the Moon and Sun's gravitational tug makes the planet bulge slightly around the equator.
Then as Earth spins on its tilted axis, its equatorial bloat induces a "wobble" that affects whether the southern or northern hemisphere cops more radiation from the Sun.
Changes in the Earth's tilt, too, affect the climate, but to a lesser extent, Professor Timmermann said.
Milankovitch cycles warm and cool the planet on timescales from tens of thousands to hundreds of thousands of years (but cannot explain Earth's current warming).
And when the cycles align so the Northern Hemisphere gets less solar radiation in summer, it can kick off an ice age
One species into two
Professor Timmermann and his team simulated Earth's climate as it wobbled its way around the Sun covering most of a geological epoch called the Pleistocene.
At the beginning of the Pleistocene around 2 million years ago, the climate entered a cool cycle. It's around this time the first waves of early human migration out of Africa began.
One of the first species to likely make those initial forays was an archaic relative of modern humans known as Homo erectus.
The researchers found their climate predictions matched pretty well with physical evidence, such as ice drilled from deep in polar ice sheets, which act as proxies for past climate.
From their modelling, the team produced maps showing parts of the world that were suitable for early human habitation based on factors like rainfall and food availability.
Then, on these maps, the researchers overlaid more than 3,200 archaeological discoveries from six species of human, including our species, Neanderthals, and H. heidelbergensis, which is thought to have appeared in Africa around 800,000 years ago.
The maps allowed the researchers to see the different kinds of environments each human species preferred to dwell in.
They found, for instance, that archaic humans preferred to hang out in regions with a pretty stable climate, while humans that emerged later, such as H. heidelbergensis, were wanderers who could settle in harsher, drier conditions.
The maps also showed where where one form of early human disappeared while a second appeared, pointing to one species evolving into another.
They saw this happen to H. heidelbergensis in two places at two times. By around half a million years ago, H. heidelbergensis had spread from southern Africa throughout Europe, and maybe into Asia too.
Around 400,000 years ago, European H. heidelbergensis dwindled and Neanderthals emerged.
Then 300,000 years ago, southern African H. heidelbergensis started to be replaced by H. sapiens.
So what caused this speciation that gave rise to us?
Around the time H. sapiens appeared, Professor Timmermann said Africa was getting very dry.
"There was a major drop in habitat suitability ... and it was actually quite dramatic to the habitats in southern Africa."
It may have been that as local H. heidelbergensis populations shrank, so too did their gene pool.
Such "genetic bottlenecks" could have then led to the evolution of our species in southern Africa, Professor Timmermann said.
A single point of origin?
This single southern African origin for H. sapiens evolution is a claim that probably won't sit well with everyone in the human evolution domain.
A competing hypothesis suggests our species didn't evolve from one area, but instead arose from multiple "subdivided" populations across the African continent that sporadically bred with each other.
Michael Petraglia, head of Griffith University's Australian Research Centre for Human Evolution, is a proponent of a more "multi-regional" model.
The new study's climate models showed suitable human habitats in eastern as well as southern Africa, but there was only enough archaeological and fossil evidence to support the origin of our species in the southern region.
"And here's the fundamental problem," Professor Petraglia said.
"They're doing this based on current fossils and archaeological data. But places like Africa are hugely understudied."
In other words, evidence for our species evolving in other parts of Africa could be there. It's just not yet been dug up.
Such ancient fossil and tool scarcity is a "perpetual problem" in archaeology, Dr Moffat said.
"There's always the question of how much of what we found reflects the original distribution of the species, and how much of it is an accident of geography and geology."
How much can tools really tell us without fossils?
Another potential spanner in the study's works, Professor Petraglia wrote in an accompanying News and Views article in the journal Nature, is inferring the presence of particular human species at different sites based on tools alone.
Fossilised bones are incredibly rare, so most physical evidence of ancient humans comes from artefacts like tools they left behind.
But it can be incredibly hard, sometimes impossible, to designate a species to a tool in the absence of fossils, Professor Petraglia said.
"And in most of these 3,200 sites [in the new study], they're ... mostly archaeological sites with only stone tools, but [the researchers] are willing to categorise them as belonging to a species.
"As an example, even species that are absent in their model like Denisovans could have been using some of the toolkits that they say were being used by Homo heidelbergensis."
Finding more fossils and better categorisation of tools will bring the intricate story of human evolution into crisper focus.
New techniques that let researchers piece together genetic material extracted from fossils or even sediments will likely shake up archaeology too, Dr Moffat said.
"As ancient DNA techniques come online more, and hopefully we can push back to this era, then I think we'll see a whole other understanding of human evolution and the family tree emerge."
