There is a mound in eastern Turkey on a hillside in a remote corner of Agri province about 29 km south of the summit of Mount Ararat.
It sits at roughly 6,300 ft above sea level.
It is 515 ft long.
It has two tapered ends and a raised central ridge.
From the air, from almost any angle, it looks exactly like a boat.
Most people who hear that description assume the story ends there.
Another exotic claim filed away somewhere between ancient astronauts and Atlantis.
The scientific consensus has long said this place is just an eroded fold in the rock.
A geological curiosity, nothing more.
But in September 2026, a team of 20 researchers from five countries drilled 16 holes into that mound.
They drilled some of them to a depth of 18 m.
At one point, four to 5 meters down, a percussion drill bit struck something so hard it snapped clean.
And they lowered a small robot, a camera on wheels, built originally to crawl through wombat tunnels in Tasmania down into the darkness beneath the surface.
What it returned from down there has split credible people into two camps that cannot both be right.
The central question this video is about is not whether Noah’s ark exists.
The central question is this.
What exactly is underneath that mound?
Because both sides now agree something is there.
They just disagree completely on what it is.
To understand why this matters, you have to go back to May 1948.
A Kurdish shepherd named Rashid Sarhan was working the hillside above Dogayazit village after a series of three earthquakes had shifted the earth.
When the ground settled, something had been exposed.
A long oval form rising from the surrounding mud flow.
Sadhan noted it.
Local people passed the story around.
For 11 years, the wider world did not know it existed.
Then in October 1959, a Turkish army captain named Ilhan Durupinar was reviewing aerial photographs taken during a NATO mapping mission.
He was a cgrapher by training.
His job was to read landscapes.
And in one photograph, he spotted it.
That unmistakable outline, a shape that to anyone looking at it with fresh eyes does not seem to belong on a hillside.
Gurupinar reported it to the Turkish government immediately.
The photographs were published in Life magazine in [music] 1960.
They reached a professor of photoggramometry at Ohio State University named Arthur Brandenburgger.
He examined them and said the site deserved investigation.
A small expedition was organized.
In September 1960, a team including Brandenburgger, Durupinard himself, and members of the Archaeological Research Foundation dug into the formation for two days.
They used dynamite in places, blasting into the mound’s edge.
They found soil and rock.
They went home.
And here is where the official story, the we looked, there was nothing story, gets its first crack.
Because that 1960 expedition did not conduct systematic scientific analysis.
They dug.
They used explosives.
They did not find a wooden ship sitting intact in the ground.
And so they left.
The conclusion that the site was nothing was not based on rigorous geological testing.
It was based on 2 days of digging and the absence of obvious wreckage.
For the next several decades, that absence of obvious wreckage was treated as the end of the matter.
It wasn’t.
In 1977, an American amateur explorer named Ron Wyatt visited the site.
Wyatt was not a credentialed scientist.
He was a nurse anesthetist from Tennessee who had become obsessed with biblical archaeology.
He looked at the mound and became convinced it was Noah’s ark.
He promoted that claim loudly on television, in self-published books, through a ministry he ran.
He brought equipment, he said detected iron in a grid pattern beneath the surface.
He said he had found a ship’s ribbing.
Almost no professional geologists accepted Wyatt’s claims.
His methods were questioned.
His readings were disputed, but he kept coming back.
And in 1985, he was joined at the site by a man named David Foldold.
Fold was different from Wyatt.
He was a professional marine salvage expert, someone [music] who spent his career locating and recovering sunken vessels.
He had technical credentials.
[music] When he looked at the formation and listened to Wyatt’s interpretation, he initially agreed.
In 1988, Fel published a book called the ark of Noah, arguing at length that the Durupanar formation was the remains of the biblical vessel.
And then 8 years later, he changed his mind completely.
[music] In 1996, Fold co-authored a paper in the peer-reviewed journal of geocscience [music] education with geologist Lawrence Collins.
The title of the paper was direct bogus Noah’s Ark from Turkey exposed as a common geologic structure.
Their conclusion, after microscopic analysis of the materials Wyatt had claimed were artificial, was [music] unambiguous.
The formation was a doubly plunging sincline, a natural folding and erosion of sedimentary rock layers that happens to produce an oval, vaguely boat-shaped [music] ridge.
The supposed iron grid was weathered volcanic mineral.
There was no fossilized wood.
There were no traces of wood, reed, or elemental carbon.
It was rock.
It had always been rock.
Fold had once written a whole book arguing this was the ark.
Now he was co-writing a paper calling that conclusion bogus.
His Australian biographer June Daws later maintained that Fold privately continued to believe he had found something significant that [music] his public reversal and his private conviction never fully aligned.
Whether that is true is impossible to verify now.
Fold died in 1998 2 years after the paper was published.
Whatever he believed in his final years, he took it with him.
That is the official scientific position as of 1996.
And for most of the world, that settled it.
What most of the world does not know is that something began happening at the site in 2019 that changed [music] the terms of the debate.
Andrew Jones is an independent researcher and software developer based in the United States [music] who has been studying Durupinad for decades.
In 2019, he collaborated with a company called TOPA 3D out of Oregon to conduct a systematic ground penetrating radar survey of the entire formation.
Ground penetrating radar works by sending pulses of electromagnetic energy into the ground and recording the reflections.
Different materials reflect differently.
Voids reflect differently than solid rock.
Constructed walls reflect differently than undifferentiated sediment.
The data comes back as a cross-section of what lies beneath the surface.
What the 2019 survey showed was not what you would expect to see if the formation were a simple [music] uniform block of compressed volcanic clay and limestone.
The radar data showed three distinct layers.
It showed what appeared to be a long central corridor approximately 13 m in length running along the interior of the mound.
It showed angular structures, lines meeting at right angles at depths [music] between 8 and 20 m.
Right angles don’t occur randomly in geological formations.
[music] They require force acting from multiple directions or they require [music] construction.
Jones was careful in how he described this.
He did not say it proved Noah’s ark.
He said it was not what [music] you would expect to see if this site is just a solid block of rock or an accumulation of random debris from a mud flow.
He said it was what you would expect to see if this was a constructed object.
The geological counterargument, and it is a real one, worth taking seriously, is that fractures and joints within rock can produce linear and angular features that look constructed but aren’t.
A sincline under stress can fracture in regular patterns.
Radar reads those fractures and they can look like walls.
Lawrence Collins, who co-wrote the 1996 paper, has pointed out exactly this.
Finding straight lines and right angles underground, he has said, is not by itself evidence of a man-made structure.
Both of those things can be simultaneously [music] true.
The radar shows something.
What that something is, that is what the 2026 expedition went to find out.
We’ll come back to what the radar specifically showed the team where to drill.
But first, there is a detail about the soil at this site that almost nobody was talking about, and it turns out to be one of the more striking pieces of data in this whole story.
In 2024, the Noah’s Arc scans team working with soil scientists collected samples from 22 points across and around the formation.
They analyzed the organic content, the amount of decomposed biological material in the soil.
Inside the perimeter of the boatshaped formation, the organic matter was roughly three times higher than in soil collected just outside it.
On the same hillside, a few feet past the boundary of the shape, the numbers dropped sharply.
The potassium levels were also elevated inside.
The pH was different.
The chemistry of the soil within the outline of the formation was systematically distinct from the chemistry immediately surrounding it.
Andrew Jones described the difference as statistically staggering.
[music] He was explicit about the interpretation.
Wood decomposes as it decays over thousands of years.
The organic matter it was made of doesn’t vanish.
It migrates into the surrounding soil.
If there were once large quantities of timber buried here, you would expect to find elevated organic content in the soil above and around it.
You might push back on that, and the push back is fair.
Elevated organic matter inside a depression in a hillside has multiple possible natural explanations.
Vegetation accumulates differently in low-lying areas.
Water drains into depressions and carries organic sediment.
A natural hollow could collect biological material over centuries entirely without human intervention.
The researchers acknowledge this.
They say the soil results are an investigative lead.
They are not claiming the organic matter proves anything.
They are saying it is consistent with decayed wood and that it adds one more layer of anomaly to a site that is accumulating anomalies.
Hold that thought because in the summer of 2026, the Turkish government did something it had never done before at this site.
The previous investigations, [music] Wyatts in the 1980s, fossolds, the various surface surveys had all been conducted without formal government authorization for deep excavation.
The Turkish government had allowed surface observations.
[music] It had not allowed drilling.
In 2026, that changed.
The Ministry of Environment, Urbanization, and Climate Change issued a permit.
The project was placed under the direction of Professor Jenker Atila, an archaeologist in the department of archaeology at Civas Jumhuret University.
That matters.
This was not a group of enthusiasts with a shovel.
This was a permitted university-led excavation, the first time in the history of the site that a formal academic institution had official government authorization to go underground and [music] see what was actually there.
Professor Atila said it plainly.
This was the first time anyone with government permission had been able to go underground and see what was there.
Every previous claim about this site, every photograph, every radar image, every surface survey had been made without that access.
The 2026 expedition was the first to have it.
The team assembled 20 researchers from five countries, including Turkish archaeologists, geologists, and soil scientists, alongside the American-based Noah’s Arc Scans organization.
They brought militarygrade ground penetrating radar developed by Dr.
Kosro Baktar, a scientist whose radar technology had been used by the US Air Force, the Navy, and NORAD.
They brought LAR mapping, electrical resistivity tomography, drone imagery.
They built a full three-dimensional model of the surface.
They established GPS coordinates at hundreds of points across a grid laid over the entire formation.
And then on September 23rd, 2026, they started drilling.
The drilling plan was based on the radar data.
Bktar’s new survey done with equipment more advanced than what Topa 3D had used in 2019 had confirmed the earlier findings and added to them.
The radar showed what appeared to be voids and tunnels at depths between 3 and 13 [music] m.
It showed what the team interpreted as possible corridors and chamberlike spaces.
One possible passage, they said, extended more than [music] 250 ft through the interior of the formation.
They also described something the radar had shown them that they could not easily dismiss.
A thick subsurface layer divided into three [music] distinct sections, three horizontal levels.
The biblical description of the ark in the book of Genesis specifies that it had three [music] decks.
The team is careful to say that a three-layered subsurface feature does not prove the formation is the ark.
Such a match between data and scripture could be coincidence.
[music] A sincline erodess in layers.
Sedimentary deposits form in horizontal strata.
None of that is unusual, but the team notes it and [music] it adds to the texture of the anomalies they are trying to explain.
16 bore holes, 12 inside the boundary of the formation, four outside as controls.
Core samples extracted to depths of up to 18 m.
More than 360 soil samples which have been sent to laboratories in Ankura for analysis.
The results are expected in early 2027.
[music] And that is a piece of this story we will return to.
But before the full drilling results come back, two things happened during the fieldwork that the team [music] says they did not expect.
The first was the drill bit.
At one of the holes inside the formation, 4 to 5 m below the surface, the drilling system had already penetrated the softer overlying material.
Then it stopped.
The percussion bit had [music] struck something so hard that it broke.
Whatever was down there was harder than limestone.
Jones told reporters, “Petrified or highly mineralized wood are both a possibility we want to investigate through third party lab testing.”
[music] He was precise.
That is a working hypothesis, not a confirmed finding.
The team does not know what broke the drill.
No laboratory has yet identified the material.
But the team notes that petrified wood is typically harder than limestone.
A drill bit snapping 4 to 5 m down inside a formation that looks like a boat is an arresting data point.
The second unexpected finding was more fundamental.
In every bore hole drilled inside the perimeter of the formation, the team encountered the same result.
No bedrock, no solid limestone.
The rock behavior outside the formation boundary was different from the rock behavior inside it.
Standard bedrock appeared in the control holes outside the shape.
Inside there was something else.
Rich dark soil, layers of sediment, organic material, but no bedrock.
Jones said it directly.
No bedrock or solid limestone has so far been identified in the holes drilled inside the formation.
These current findings challenge claims that this is simply a natural shape made of limestone or bedrock.
A natural sincline is by definition rock.
The 1996 Collins and Fhold paper described it as a rock formation.
If the interior of the formation is not rock, if it is something else chemically distinct from the surrounding hillside, that is a question the 1996 study does not answer because the 1996 study did not drill.
If you have been watching until now, you deserve to know about the robot.
And to understand what the robot found, you need to understand what it is and where it came from.
If you came to this channel for stories like this one, for the real evidence, the honest disputes, [music] and the things that most documentaries either sensationalize or bury, subscribe now.
We don’t announce discoveries before they’re confirmed, and we don’t hide the counterarguments.
Dr.
Robert Ross is an associate professor of electronics engineering at LRO University in Melbourne, Australia.
In 2020 and 2021, he designed and built a small remotec controlled robot called Wombot, a machine specifically engineered to crawl through narrow, muddy Wombat burrows in Tasmania without damaging them.
Equipped with front and rear cameras, environmental sensors, and a chassis small enough to navigate tight underground spaces, the wombot was published in the journal SN Applied Sciences in 2021.
It was used to study whether wombat burrow conditions helped spread sarcoptic mange among the animals.
Ross saw a news item in 2025 from a group looking for a robot that could explore narrow underground spaces at a site in eastern Turkey.
He reached out.
He spent 9 months collaborating with the Noah’s Arc team.
He built a smaller adapted version of the wombot named Gopher after the gopher wood mentioned in the biblical description of the ark with front and rear waterproof wide-angle cameras, environmental sensors, and the ability to operate up to 25 m below the surface.
The robot left Australia in late July 2026.
It had spent its entire existence before that moment exploring the burrows of marsupials in Tasmania.
Now it was headed for a hillside above a small Turkish town toward a hole drilled into one of the most debated patches of Earth on the planet.
Ross was clear about what he hoped his technology would do.
It could be the archaeological discovery of the century, he said.
But whether the site turns out to be Noah’s Ark or something completely different, it’s exciting to see our technology helping researchers investigate and better understand what’s there.
That is the measured voice of a scientist who wants to know what is in the ground.
Not one committed to a predetermined answer.
The gopher was lowered into one of the drilled bore holes 12 m down.
The radar had mapped what appeared to be voids beneath it.
The robot’s camera was rolling.
The first attempt ran immediately into a problem.
The cavern floor below the borehole was covered in extremely loose sediment.
The robot’s wheels struggled to find traction on material that shifted under them.
The footage from this first deployment was limited by the robot’s inability to maneuver effectively in the loose material.
Think about what that detail means on its own.
Loose sediment on a cavern floor.
If this were a solid block of compressed natural rock, there would be no cavern floor to struggle on.
There would be no cavern at all.
The team’s end of season video shows the gopher being lowered into [music] the borehole, a small wheeled machine about the size of a large book descending on a cable into darkness.
The footage it captured shows the walls of the drilled shaft, then the opening into a void, then loose, dark sediment, and then something that the geological interpretation of this site needs to account for.
The robot’s cameras recorded that there was in fact an open underground space beneath the formation, not bedrock, not compressed clay, an accessible void at depth inside the perimeter of the mound.
The team reports that the organic material visible in the core samples and in the robot’s footage appears rich and dark, consistent, they say, with centuries or millennia of biological decomposition.
Engineers are now modifying the gopher’s mobility system so that it can maneuver more effectively on loose sediment.
The next field season will send it deeper into the mapped corridors and chamberlike spaces.
That footage has not yet been released.
What has been released is this.
A robot camera went underground inside the Durupanar formation and documented that there is material down there that is not what a uniform natural rock formation should contain.
Think about what that means.
Even in the most conservative possible interpretation, if this is simply an eroded geological fold, if the 1996 paper’s conclusion is the complete story, then the interior of the fold should be rock.
Compressed sedimentary rock folded by tectonic pressure, weathered by millennia of freeze thaw cycles at 6,300 ft of elevation.
You drill into it, you find rock, consistent rock.
No voids, no anomalous carbon, no organic rich dark soil, no layer hard enough to snap a drill bit at 5 m.
The team does not have an explanation for all of these things.
Neither do the geologists who say the site is natural.
Both sides are now waiting for the same thing.
The laboratory analysis of the 360 soil samples and core materials expected from Ankara University and Sivas Jumury University in early 2027.
Those results will not definitively prove or disprove that this is Noah’s ark.
Let’s be precise about what they could do.
If the carbon dating of organic material inside the formation returns [music] a date consistent with the proposed timeline of the biblical flood approximately 2,300 BC or about 4,300 years ago, that would be significant.
If the hard layer the drill broke on is identified as petrified timber by laboratory analysis, that would be significant.
If it turns out to be a naturally occurring hard mineral deposit, that would be equally significant in the other direction.
Right now, neither conclusion has been reached.
And that brings us to where the honest version of this story lives.
Not in the headlines that say 100% confirmed and not in the dismissals that say this is obviously nothing.
Andrew Jones, who founded Noah’s Arc Scans and who has been studying this site longer than almost anyone alive, told Newsweek in late September 2026 that the team had not yet identified any material [music] as ancient wood.
He said no laboratory results were in yet.
He said, and this is the voice of a careful researcher, not a promoter, that they had found a hard layer that broke a drill bit, and that petrified wood was one possibility they wanted to investigate through third-party testing.
Professor Attilla, the archaeologist from Civas Chumhuri University, directing the permitted excavation, cautioned explicitly against claims circulating on social media and in some news coverage that the research had already confirmed the presence of Noah’s Ark.
It had not.
The fieldwork had produced data that required analysis.
That analysis was underway.
He was also clear that the project’s findings would be published [music] in peer-reviewed papers and at international conferences under the title Durupun Formation 2026 research.
Whatever the lab’s return, it will go through the standard process of scientific scrutiny.
That is not nothing.
That is exactly how it should work.
On the other side, geologist Lawrence Collins, who co-wrote the 1996 paper and [music] remains the most credentialed scientific voice arguing against the ark interpretation, has noted that right angle features in underground radar data can be produced by fractures in a sincline under geological stress.
He found no fossilized wood in the 1990s [music] analysis.
He found no traces of reed or elemental carbon.
The fact that the 2026 team is drilling with better equipment does not by itself invalidate what the 1996 analysis found through microscopy.
There is a fair question to raise here.
If this is just a natural geological feature, why does it have a different chemical [music] signature than the surrounding rock?
Why does it have three times the organic content of soil collected 5 ft outside its [music] perimeter?
Why is there no bedrock inside the formation when bedrock appears immediately outside it?
Why did a drill bit break on something harder than limestone [music] 4 m down?
Collins would likely answer that those questions have natural explanations [music] that will emerge from the lab results.
He might point to the drainage patterns on the hillside, to the particular mineral composition of this volcanic region, to the way erosion creates differential chemical environments.
He has been careful, consistent, and peer-reviewed in his skepticism.
Both men could be right about some things and wrong about others.
That is how science actually works.
The narrator’s position stated plainly, “This site has not been confirmed as Noah’s ark.
The 2026 expedition has not discovered the biblical vessel.
What it has found is that the interior of this formation does not behave like the solid natural rock formation the 1996 consensus described.
That gap between the 1996 conclusion and the 2026 data is real.
It deserves investigation.
The lab results due in 2027 will not resolve the theological question of whether Noah’s ark existed.
But they may tell us for the first time whether there is something buried inside that mound that has no natural explanation.
That is a meaningful distinction and it is why the story is not over.
There is one detail about the 2026 expedition that almost no coverage of it has mentioned and it is the one that stays with you.
The pottery.
During the fieldwork, researchers found ceramic fragments in the vicinity of the Durupinar Formation.
Ancient pottery, not inside the formation itself.
Jones was careful to clarify that the fragments were found beside it, not within it, but they are there on the hillside.
Pottery that may indicate ancient human settlements in the area at an elevation and in a location that would have been remote and difficult to reach.
Who was up there?
Why?
Pottery near an ancient, striking, unmistakably boat-shaped landmark suggests that whoever lived in this region thousands of years ago knew this place, came to it, perhaps considered it significant.
That is not evidence of anything specific, but it is a human detail.
The kind that reminds you that this is not just a geological puzzle.
People have been looking at that mound for a very long time.
They may not have had ground penetrating radar, they had eyes.
And here is where the story closes its first loop and opens the next one.
In October 1959, Ilhan Durupinard looked at an aerial photograph and saw something on a hillside that he could not explain.
He was a trained cgrapher.
He knew what mountains looked like from the air.
He reported it because in his professional judgment, it was anomalous.
He was not a believer in a simple sense.
He was a man who looked at evidence and could not make it fit the expected pattern.
In September 2026, a robot built to explore wombat burrows in Tasmania went underground inside that same mound and came back with footage showing something that the expected geological explanation does not easily accommodate.
67 years separate those two moments.
In between, a marine salvage expert wrote a book arguing the site was Noah’s Ark, then co-wrote a paper saying it was bogus.
A military captain filed a report about an anomalous aerial photograph to a government that didn’t quite know what to do with it.
Researchers came and left.
Headlines appeared and faded.
The mound stayed on the hillside.
The laboratory results from 2026 are due in early 2027.
If the carbon analysis returns a date, if the hard layer is identified, if the organic matter inside the formation is confirmed as decomposed wood or confirmed as something perfectly ordinary, the scientific record will be one step clearer than it is today.
The question of what is buried in that hillside south of Mount Ararat is not answered.
The question of why the inside of that formation is chemically different from everything around it is not answered.
The footage the robot brought back from underground is not the end of this story.
It is the beginning of the part where we find out whether the anomaly has a name.
The next video covers a site two countries away where researchers in 2023 made a discovery that the Durupinar team has quietly been comparing notes with.
That one does have lab results and they are strange.
