Atlit-Yam — A Neolithic Village Beneath the Mediterranean
Atlit-Yam is a submerged Neolithic village off the Carmel coast near Atlit, south of Haifa, Israel. It lies 8–12 m deep and dates to about 6900–6300 BC, making it one of the oldest and best-preserved submerged settlements in the world. It contains rectangular stone houses, water wells, a semicircle of seven megaliths, human burials and evidence of farming and fishing. It was discovered in 1984 by marine archaeologist Ehud Galili.
How old is Atlit-Yam?
Radiocarbon dates place it at roughly 6900–6300 BC, about 9,000 years ago, in the Pre-Pottery Neolithic C period.
At that time sea level in the eastern Mediterranean was around 10–12 m lower than today, and the village stood on dry coastal land a few hundred metres from the shore. Rising seas after the last Ice Age eventually drowned it, along with a string of other prehistoric sites along the Carmel coast.
| Date | Event |
|---|---|
| ~6900 BC | Village established on the coastal plain |
| ~6900–6300 BC | Farming, herding and fishing community |
| ~6300 BC | Abandonment |
| After 6000 BC | Sea level rises and the site is submerged |
| 1984 | Ehud Galili discovers the site after storms strip sand from the seabed |
| 2008 | Study reports earliest human tuberculosis in a mother and child burial |

What has been found at Atlit-Yam?
Archaeologists found rectangular stone houses, stone-paved areas, a well about 5.5 m deep (one of the oldest known), hearths, tools of flint, bone and wood, and dozens of human burials (published counts range from about 65 to more than 90). A semicircle of seven stone megaliths, some over 1 m tall with cup-marks, surrounds a freshwater spring and may have had a ritual purpose. Remains of wheat, barley, lentils, domesticated animals and thousands of fish bones show a mixed economy of farming, herding and fishing.
Why is Atlit-Yam important for medical history?
In 2008, Israel Hershkovitz and colleagues reported DNA and bone evidence of Mycobacterium tuberculosis in the skeletons of a woman and child buried together at Atlit-Yam, the earliest confirmed human cases of tuberculosis. The find supports the idea that the disease spread with the crowding and domestic animals of early farming villages.
Was Atlit-Yam destroyed by a tsunami?
A 2006 study by Maria Teresa Pareschi suggested that a tsunami from a collapse of Mount Etna’s flank around 6000 BC could have struck the eastern Mediterranean and caused Atlit-Yam’s sudden abandonment, citing piles of fish found in the site. Many archaeologists favour gradual sea-level rise and salinisation of wells as the main reason. The question remains open.
Can you visit Atlit-Yam?
The site is protected by the Israel Antiquities Authority and is not a public dive site; its remains are usually covered by sand and exposed only after storms. Finds are studied and curated by the University of Haifa and the Israel Antiquities Authority.
For the oldest submerged town with a street plan, see Pavlopetri; for a legend, Atlantis.
Where exactly is Atlit-Yam, and what did the landscape look like?
Atlit-Yam lies roughly 300 to 400 metres off the modern beach in the Bay of Atlit, on the northern Carmel coast about 15 km south of Haifa. Today the ruins sit 8 to 12 metres down on a sandy, rocky seabed. Nine thousand years ago the scene was completely different. Sea level was lower, the shoreline lay further west, and the village stood on a low coastal plain close to the mouth of the Oren stream, with freshwater springs and patches of marsh nearby. Pollen and the remains of marsh plants recovered from the site confirm that wetlands fringed the settlement.
That setting explains much of the village’s success. The Carmel ridge, only a few kilometres inland, supplied flint, limestone and timber. The marshes and springs supplied fresh water. The sea supplied fish. A community here could combine early farming, herding, hunting and fishing within walking distance of home, which is exactly the mixed economy the excavated remains reveal.
How was Atlit-Yam discovered?
The Carmel coast has a long history of chance discoveries. From the 1960s onward, large-scale quarrying of beach sand stripped away the protective blanket that normally covers the shallow seabed, and storms then exposed shipwrecks and the remains of drowned settlements. In 1984 the marine archaeologist Ehud Galili, surveying for shipwrecks, noticed prehistoric structures in the Bay of Atlit. Over the following decades he led a long series of underwater surveys and excavations on behalf of the Israel Antiquities Authority, working with the University of Haifa’s Recanati Institute for Maritime Studies.
The first major report, by Galili and colleagues in the Journal of Field Archaeology in 1993, described the site as a prehistoric settlement on the sea floor. A 2011 underwater field school run under the European SPLASHCOS programme (Submerged Prehistoric Archaeology and Landscapes of the Continental Shelf) returned to the site to train a new generation of underwater archaeologists, and in 2025 a dedicated multi-chapter volume on Atlit-Yam, covering its architecture, wells, stone tools, fishing and research history, appeared in Springer’s “When the Land Meets the Sea” series.
How do archaeologists excavate a village under 10 metres of water?
Excavating Atlit-Yam is slow and weather-dependent work. The remains are usually buried under sand and become visible only after winter storms scour the seabed, so the team has to move quickly when conditions allow. Divers remove sediment with water dredges and airlifts, record every feature with measured drawings and photographs, and lift finds in bags and boxes for conservation ashore. Visibility in the shallow, surf-stirred water is often poor, and diving time at 10 metres is limited by air supply and cold.
The sea also protects. Once the village was drowned and sealed under sand, it was shielded from ploughing, later building and looting, which on land often destroy Neolithic sites. Waterlogged conditions preserved organic remains such as wood, seeds and plant fibres that rarely survive in dry soil. That combination of hard work and exceptional preservation is the reason drowned sites like this are so valuable to prehistorians.
What was the Pre-Pottery Neolithic C?
Archaeologists divide the early Neolithic of the Levant into phases. In the Pre-Pottery Neolithic B (PPNB), roughly 8800 to 7000 BC, people in the region were building large farming villages, such as ‘Ain Ghazal in Jordan, and domesticating wheat, barley, goats and sheep, all without pottery. Around 7000 BC many of the biggest inland villages shrank or were abandoned. The phase that followed, before pottery spread, is called the Pre-Pottery Neolithic C (PPNC).
Atlit-Yam is one of the best-preserved PPNC sites anywhere. It shows a community that was small by PPNB standards but well organised, living in rectangular houses, digging wells, storing grain and burying its dead under or beside its homes. For researchers it is a rare snapshot of how people adapted during a period of social change, and of how quickly farming spread down to the coast and was combined with marine resources.
What did the people of Atlit-Yam eat?
The plant and animal remains point to a broad, flexible diet. Excavators found a pit holding thousands of charred grains of emmer and naked wheat near one of the structures, along with granary weevils, a sure sign of stored grain. Waterlogged seeds of fig, grape, lentil and almond were recovered, and charcoal shows that olive, carob, oak and date wood were burned on the hearths.
The animal bones are dominated by goat, followed by cattle and pig. Some specialists argue that at least part of the cattle and goat population was not yet fully domesticated, which would make Atlit-Yam a window on incipient domestication rather than mature herding. Fish were clearly important too. More than 6,000 fish remains were recovered, and the great majority of the identifiable bones belong to the grey triggerfish, alongside groupers and drums. Bone fishing hooks and a drilled bone needle add to the picture of a village that knew the sea well.
Was Atlit-Yam a fishing village?
Galili and his colleagues describe Atlit-Yam as one of the earliest known fishing villages on the Mediterranean coast. The triggerfish dominate so strongly that they suggest a targeted fishery, perhaps using nets or traps set close to shore, rather than casual line fishing. In one part of the site a concentration of fish bones and scales in pristine condition was found. It has been interpreted as a pile of fish prepared for storage or exchange that was left behind when the inhabitants departed, although some researchers consider that natural accumulation cannot be ruled out.
An intriguing find from the well was a seed of Styrax officinalis, a shrub whose crushed fruits were used for centuries in the eastern Mediterranean as a fish poison to stun fish in pools and streams. A single seed proves nothing on its own, but it hints that fishing methods at Atlit-Yam may have been more varied than hooks and nets.
What do the skeletons reveal about life in the village?
Atlit-Yam is unusual among Mediterranean underwater sites because it contains human burials still in their original positions. Most bodies were laid in a flexed position, a typical Levantine Neolithic custom, and many were placed close to or inside the houses. The skeletons have been studied for evidence of health, work and disease.
Several skeletons show auditory exostoses, small bony growths in the ear canal that develop in people who repeatedly immerse their heads in cold water. The condition is common today among surfers and divers and is often called “surfer’s ear”. Its presence suggests that some villagers regularly dived or worked in the sea. Heavy wear on the teeth may reflect using the mouth as a third hand when making cord or nets, and wear patterns at the elbow have been linked to repetitive tasks such as paddling. Together they sketch a population whose daily work was physically demanding and closely tied to the water.
Why does the tuberculosis discovery matter so much?
The 2008 tuberculosis study led by Israel Hershkovitz of Tel Aviv University and Helen Donoghue of University College London examined the remains of a woman and an infant buried together and thought to be mother and child. Both showed bone lesions characteristic of tuberculosis. The team recovered ancient DNA of Mycobacterium tuberculosis and also detected the bacterium’s distinctive cell-wall lipids, two independent lines of evidence that pushed the confirmed history of human TB back by about 3,000 years.
The finding also helped settle an evolutionary question. An older theory held that human tuberculosis evolved from the bovine form after cattle were domesticated. The Atlit-Yam DNA belonged to the human-type lineage and already carried a deletion found in a common modern family of strains. In the researchers’ words, the infecting organism was definitely the human strain. That fits genetic studies suggesting the bovine form branched off later than the human one, and it shows that TB was already established in a community that kept animals but had not yet developed dairying.
How old are the wells, and why did they fail?
Atlit-Yam’s dry-stone well is about 1.5 metres across and about 5.5 metres deep, and it is among the oldest known water wells in the world. Excavators identified several phases of use. The upper fill of the well contained food waste, small stones and shells rather than clean silt, which suggests that the shaft was turned into a rubbish pit once its water was no longer drinkable.
Galili’s explanation is that the rising sea pushed salt water into the coastal aquifer. As the shoreline crept closer, the wells grew brackish and the village’s freshwater supply failed. Without fresh water a coastal settlement cannot survive, however rich the fishing, so salinisation offers a straightforward reason for abandonment that does not require a catastrophe.
How fast did the sea rise around Atlit-Yam?
At the height of the last Ice Age, around 20,000 years ago, global sea level stood roughly 120 metres below today’s level. As the ice sheets melted the sea rose rapidly, by tens of metres over a few thousand years in the early Holocene, before slowing down after about 7,000 years ago. Along the Carmel coast that rise first drowned the land beyond the shore and then the coastal plain itself.
Atlit-Yam was probably submerged by about 7,000 years ago. Later, during the Pottery Neolithic, villages of the Wadi Rabah culture were built closer to the new shoreline at sites such as Neve-Yam, Kfar Samir, Kfar Galim and Tel Hreiz. Those settlements in turn were drowned as the sea continued to rise, which is why the Carmel coast has one of the densest clusters of submerged prehistoric sites in the world.
What happened at the neighbouring drowned villages?
The younger sites add context to Atlit-Yam’s story. At Kfar Samir, south of Haifa, divers found pits full of crushed olive stones and pulp that Galili and Sharvit interpreted as evidence of olive-oil production around 7,000 years ago, among the earliest known in the world. At Tel Hreiz, a team led by Galili reported in 2019 a line of boulders about 100 metres long that they interpreted as a sea wall, possibly the earliest known coastal defence built against sea-level rise. If that interpretation is right, the people of the Carmel coast were already trying to hold back the sea a few centuries after Atlit-Yam had been lost to it.
What was the stone semicircle for?
The seven megaliths at Atlit-Yam are limestone blocks, each estimated at around 600 kg, set in a semicircle around a freshwater spring. Some carry cup-marks, small hollows pecked into the surface. Because the arrangement surrounds a source of water, Galili proposed that it had a ritual role connected with water, perhaps a shrine or meeting place. Nothing about the stones proves a specific belief, but the effort required to move and set them shows that the community could organise collective work.
It is natural to compare the semicircle with Stonehenge, but Atlit-Yam is roughly 4,000 years older than the earliest phase of that monument. It is younger, however, than the carved pillars of Göbekli Tepe in Turkey, which date to around 9500 BC. Atlit-Yam belongs to a long Near Eastern tradition of monumental stone settings that began well before farming was fully established.
How does Atlit-Yam compare with other drowned prehistoric sites?
Submerged prehistoric sites are now known from many continental shelves. In the North Sea, the drowned plain called Doggerland once linked Britain to mainland Europe. At Bouldnor Cliff, off the Isle of Wight, divers found an 8,000-year-old Mesolithic site where sedimentary DNA studies in 2015 reported wheat DNA, a claim that remains debated. In Greece, the town of Pavlopetri preserves Bronze Age streets and buildings in shallow water.
Atlit-Yam stands out among these for its combination of age, architecture, wells, stone monuments and in-situ human burials. Few underwater sites allow researchers to study how people lived, worked, fell ill and were buried in the same place, which is why the village keeps appearing in textbooks of prehistory and paleopathology.
How reliable is the Etna tsunami idea?
Maria Teresa Pareschi and colleagues ran computer simulations in 2006 showing that a giant debris avalanche from Mount Etna’s eastern flank, entering the Ionian Sea, could have sent a tsunami across the eastern Mediterranean in the early Holocene. They suggested that Atlit-Yam might have been abandoned, about 8,000 years ago, because of such a wave. The proposal attracted wide media attention, partly because of the fish pile that seemed to imply a sudden departure.
The idea met immediate scientific criticism. A formal comment published in Geophysical Research Letters in 2008 questioned the dating and the link to the marine deposits used as evidence. More fundamentally, no tsunami deposit has been identified at Atlit-Yam itself, and the dates of the village’s abandonment and of the Etna collapse are not precise enough to show they coincided. Most archaeologists working on the site therefore regard gradual sea-level rise and failing wells as the better-supported explanation, while keeping the tsunami as an untested possibility.
Why does Atlit-Yam matter today?
Atlit-Yam is a reminder that sea-level rise is not new to human societies. People on the Carmel coast watched their wells turn brackish, moved inland, and built again, and their abandoned homes now provide some of the best evidence for early farming, fishing and disease in the Mediterranean. For anyone interested in the deep history behind modern coastal change, it is one of the most important underwater archaeological sites in the world.
How old is Atlit-Yam?
About 9,000 years old, dated to roughly 6900–6300 BC.
Where is Atlit-Yam?
Off the coast of Atlit, near Haifa, Israel, 8–12 m underwater.
Who discovered Atlit-Yam?
Marine archaeologist Ehud Galili in 1984.
What is the stone circle at Atlit-Yam?
A semicircle of seven megaliths around a freshwater spring, possibly ritual.
Why did Atlit-Yam sink?
Mainly post-Ice Age sea-level rise; a tsunami hypothesis has been proposed but is debated.
Can you dive Atlit-Yam?
No, it is a protected archaeological site and usually buried by sand.
- Galili, E. et al. (1993). Atlit-Yam: a prehistoric site on the sea floor off the Israeli coast. Journal of Field Archaeology 20: 133–157.
- Hershkovitz, I. et al. (2008). Detection and molecular characterization of 9,000-year-old Mycobacterium tuberculosis from a Neolithic settlement. PLoS ONE 3(10): e3426.
- Pareschi, M.T. et al. (2006). Lost tsunami. Geophysical Research Letters 33.
- Galili, E., Benjamin, J., Hershkovitz, I. et al. (2017). Atlit-Yam: a unique 9000 year old prehistoric village submerged off the Carmel coast, Israel – the SPLASHCOS field school (2011). In: Under the Sea: Archaeology and Palaeolandscapes of the Continental Shelf. Springer.
- Yerkes, R.W., Galili, E. & Barkai, R. (2014). Activities at final Pre-Pottery Neolithic (PPNC) fishing village revealed through microwear analysis of bifacial flint tools from the submerged Atlit-Yam site, Israel. Journal of Archaeological Science.
- Vigliotti, L. (2008). Comment on “Lost tsunami” by Maria Teresa Pareschi et al. Geophysical Research Letters.
- Galili, E. & Sharvit, J. (1994). Evidence of olive oil production from the submerged site at Kfar Samir, Israel. Journal of the Israel Prehistoric Society.
- Galili, E. & Horwitz, L.K. (eds) (2025). Atlit-Yam, a Submerged Pre-Pottery Neolithic C Site off the Carmel Coast, Israel. When the Land Meets the Sea series, Springer.
Blane Perun has been exploring Earth’s oceans and marine life for over 25 years. Founder of TheSea.Org in 1999, underwater photographer, coral aquaculture pioneer, and explorer of 80+ countries.
About Blane Perun →More on Sunken Cities →