Using advanced machine-learning and climate models, researchers have shown that the ancestors of crops like wheat, barley, and rye probably were much less widespread in the Middle East 12,000 years ago than previously believed. This challenges traditional assumptions about the geography of early plant domestication and agriculture.
In a new study in the journal Open Quaternary, researchers from the University of Copenhagen and the University of the Basque Country have reconstructed the likely ancient geographic ranges of 65 wild plant species closely associated with early farming in West Asia. These include the wild ancestors of wheat, barley, rye, lentils, and other crops that sparked the agricultural revolution more than 10,000 years ago.
"The first farming societies were established in the Middle East about 12,000 years ago. We know this from the artifacts, seeds, and animal bones that archaeologists have recovered from excavations. But we know little about the natural background vegetation in these areas, which means that we also don't know exactly where the Neolithic peoples found the plants that they eventually domesticated," says archaeologist and lead author Joe Roe from the University of Copenhagen. He adds:
"Based on our new data, it looks like the ancestors of some of the plants most important to modern agriculture - wheat, rye and barley, etc. - did not grow where we expected and also that they were much less widespread than we thought."
Joe Roe and his co-author, archaeobotanist Amaia Arranz-Otaegui, were surprised to find that many early crop ancestors appear to have been concentrated in the Mediterranean coast of the Levant, suggesting this area acted as "refugium" during the rather extreme climate of the late Ice Age.
"This suggests that many wild crops were well-adapted to quite cold and dry conditions and did not necessarily expand with the arrival of the warmer and wetter climate in which the first farming communities established themselves," says Amaia Arranz-Otaegui.
Together, these findings provide the clearest picture yet of where the world's earliest agricultural plants once grew and the kind of landscapes ancient communities lived in when they transitioned from foraging to farming.
A methodological breakthrough
The study also marks an important step forward for how researchers model past ecosystems. By combining large, open datasets on where specific plant species grow today with advanced computer simulations of past global climate, they were able to create detailed maps showing where ancient plants were likely to have grown.
"Essentially, we used the same climate simulations that IPPC uses to predict our future climate, just turned backwards, and combined them with a machine learning model of what kinds of environment these plants are adapted to," says Amaia Arranz-Otaegui.
According to the researchers, this modelling approach represents a new line of evidence for understanding the ecological context of early agriculture. Because it does not rely on archaeological preservation, which can be distorted by burial, human activity, and recovery biases, it offers an independent and complementary picture of ancient plant environments.
"This gives us a whole new window onto the ecological backdrop of the world's first farmers," the two authors conclude.
The article Biogeography of Crop Progenitors and Wild Plant Resources in the Terminal Pleistocene and Early Holocene of West Asia, 14.7-8.3 ka has been published in the open access journal Open Quaternary.
The article is an outcome of the ERC funded research project PalaeOrigins - Tracing the Epipalaeolithic origins of plant management in southwest Asia conducted at the University of the Basque Country, in collaboration with researchers from University of Copenhagen.
