Neanderthals may have crossed the Arabian Desert — scientists have found circumstantial evidence

However, this is still only circumstantial evidence. No Neanderthal bones or DNA have been found in Arabia, so scientists cannot claim that it was the Neanderthals who made these tools. Moreover, the discovery does not prove that Neanderthals literally traversed the entire Arabian Peninsula — it points to the possible penetration of this human group deep into what is now the Nefud Desert.

The trace was found near an ancient lake

The Jebel Katefeh-1 site, or JKF-1, is situated near Jubba in the north of modern-day Saudi Arabia. Today, it is one of the arid regions of the Nefud Desert, but tens of thousands of years ago, the climate here changed repeatedly.

Archaeologists have long known that there was an ancient lake near Jebel Katefeh. Hundreds of Middle Palaeolithic stone artefacts had previously been found at the site, but the overly broad range of dates meant it was not possible to link them with certainty to a specific ancient human population.

The new study has refined the age of the archaeological layers using optically stimulated luminescence — a method that determines when the mineral grains in the sediment were last exposed to light.

Two of the key dates were 59.2 ± 5.1 thousand years and 60.1 ± 3.8 thousand years. Taking the entire stratigraphy into account, the authors date the site’s main period of occupation to approximately 55 thousand years ago.

The tools were compared with finds from Africa and the Levant

As no human bones were found at the site, the researchers attempted to identify the possible makers based on stone-working techniques.

The analysis focused on artefacts created using the Levallois technique. In this technique, the ancient craftsman would first prepare the stone core in a specific way, so that a flake of the desired shape could then be obtained with a single or a few strikes.

It is important to note that both Neanderthals and Homo sapiens used the Levallois technique. Therefore, the mere discovery of such a tool says nothing about the biological affiliation of its maker.

The authors went a step further and compiled a comparative set of 704 intact Levallois blanks from sites in North-East Africa, Arabia and the Levant, dating from approximately 100 to 50 thousand years ago. They then carried out a statistical comparison of their shapes and investigated which factors best explained the differences: site age, environmental conditions, raw materials, collection size, site type or the history of human populations.

In comparative collections where the presumed makers of the tools were already known from the archaeological context, the model distinguished between tool assemblages associated with Neanderthals and Homo sapiens with over 93 per cent accuracy. It was then applied to finds from Jebel Katefe-1.

The result was quite striking: the median probability of Neanderthal attribution for the JKF-1 tools was 78 per cent.

The model attributed only four of the flakes examined to Homo sapiens with greater probability. A further approximately 24 per cent of the artefacts were assigned a probability of association with Neanderthals exceeding 95 per cent.

According to the authors, the technological differences between these tools and earlier assemblages from Arabia are best explained by a change in, or the presence of, another human population.

And Neanderthals emerge as the most likely candidates for this period.

It is easy to misinterpret this figure

78 per cent does not represent the probability that Neanderthals were actually present in Saudi Arabia. It is a statistical classification of the shape of stone flakes within a specific model and the selected comparative dataset.

The authors themselves draw attention to a significant limitation: their comparative database lacked open-air Neanderthal sites, whereas JKF-1 is precisely such an open-air site. This may influence the shape of the stone artefacts found and complicates interpretation.

The researchers attempted to statistically separate the influence of site type from presumed population affiliation and found an independent effect for both factors. Nevertheless, in the absence of human remains, the conclusion remains probabilistic.

Therefore, the correct formulation is that the tools show a strong technological similarity to those associated with Neanderthals, rather than ‘scientists have proven the presence of Neanderthals’.

Nefud looked very different back then compared to today

There is another important caveat. It would be wrong to imagine Neanderthals making their way through a modern, scorching, waterless desert.

The artefacts date from a period of climatic transition, when, following a drier phase, conditions became wetter and the ancient lake filled up again. Such a climate could have created temporary green corridors with water, vegetation and animals within what is now a desert.

The site was located right next to such a palaeo-lake.

Therefore, the possible presence of Neanderthals in the Nefud primarily demonstrates their ability to utilise more diverse landscapes than the cold regions of Europe traditionally associated with them. However, this is not yet proof of life in an extremely arid, modern-type desert.

Neanderthals may have ventured further south

Neanderthals are known from confirmed fossil finds in the Levant and the Zagros Mountains. During the period to which JKF-1 is now attributed, Neanderthal populations were present at several sites to the north of Arabia.

If these new stone tools were indeed made by them, their known geographical range would have to be extended considerably further south.

This also undermines the old stereotype of Neanderthals as people adapted almost exclusively to the cold European climate. Their remains are already known far beyond the borders of Western Europe, and this new study suggests an even broader environmental range.

That said, the authors do not claim that this particular group lived permanently in Arabia. Wet periods may have opened up temporary routes along which people and animals ventured deep into the peninsula, and with the return of drought, such corridors would have disappeared again.

The discovery has an even more intriguing implication.

The Arabian Peninsula was situated between Africa and Eurasia and repeatedly served as a route for the dispersal of ancient humans. Archaeological evidence shows that Homo sapiens had entered Arabia well before the period under consideration. For example, a human phalanx from Al-Wusta dates back to approximately 85,000–90,000 years ago.

If Neanderthals did indeed reach northern Arabia around 55,000 years ago, the peninsula may have been part of a much wider area through which several human populations moved at different times.

The authors even suggest that contact and interbreeding between Neanderthals and Homo sapiens may have occurred across a more extensive region than is commonly assumed.

However, the new study has found no evidence of such an encounter. At JKF-1, there are no remains of individuals from both groups, no ancient DNA or other direct evidence of contact. This is a possible consequence of the new geographical picture, rather than a proven event.

The key evidence is still missing

The most reliable way to settle the matter would be to find human remains or ancient DNA directly within the archaeological layers of Arabia.

But this is particularly difficult. The hot conditions of the low latitudes are not conducive to the long-term preservation of DNA, and fossilised human remains on the Arabian Peninsula are extremely rare.

Researchers are therefore attempting to extract demographic information from stone tools – one of the most durable types of archaeological evidence.

New research shows that statistical analysis of their shape can reveal patterns that were previously assessed mainly by visual inspection. However, the most cautious conclusion so far is that, around 55,000 years ago, there was a stone tool industry in northern Arabia which, based on a whole range of characteristics, is significantly closer to known Neanderthal assemblages than to comparable collections of Homo sapiens.

It is not yet known for certain who exactly held these stones in their hands.

Source

Study:“Neanderthals in the desert? Population history best explains Middle Palaeolithic technological heterogeneity in Arabia”.

Authors: James Blinkhorn, Huw S. Groucutt, W. Christopher Carleton, Eric Andrieux, Simon J. Armitage, Ceri Shipton, Paul S. Breeze, Nick Drake, Abdulaziz Al-Omari, Faisal Al-Jibreen, Abdullah M. Alsharekh, Eleanor M. L. Scerri, Simon Blockley, Michael D. Petraglia.

Journal: *Quaternary Science Reviews*, vol. 390, article 110196. The paper was accepted on 28 July and published online on 1 August 2026.