I don't know any quite as powerful as Google Earth.

Seeing beautiful patterns from above...

Dropping down into street view...

And seeing the planet in ways I would never get to see in person.

So when I came across this post on Reddit, I was fascinated.

It described “undocumented markings” in Algeria, in the middle of the Sahara

near a location called “Tebalbalet tomb.”

Visible on Google Earth.

There were 22 of them, each with 12 “surrounding things”, 42 meters in diameter

420 meters apart, at longitude 4'20 East.

It almost sounded like a joke.

But then I copied the coordinates and I looked.

There they were: identical circles in an almost perfect line.

160 kilometers from any signs of life in the world's largest desert...

in the middle of the biggest country in Africa.

This is a story about the limits of what you can find out on the internet.

About all the different ways of looking at the same thing.

And about going all the way there.

Over the course of the last 20 weeks, we filmed every step of the process

as we tried to figure out one thing...

What could these circles be?

So this whole story starts back in September 2021

when I first saw the Reddit post.

I wanted to figure out what these “markings” were

and make a video out of the entire reporting process.

No matter how long it took.

Because the answer had to be out there.

And, step one, I knew I was going to have to send some emails.

For weeks, I reached out to everyone I could think of:

Algerian experts, officials, tour groups...

even the closest hotel, in a city called Aïn Salah.

I read up on the town the circles were located closest to: Foggaret Ezzaouia.

I asked the commenters on the Reddit post...

and we even tracked down a Twitter account we thought was the same Will K

who posted this question to several subreddits before deleting his Reddit account.

I tried English and French...

organizations, academics, locals...

And then...

I waited.

But there was one easy thing to clear up first.

Were these circles real?

Or were they just some kind of satellite imaging glitch?

So I asked a teammate who works with maps a lot:

Sam, he produces our series Atlas.

And he pointed me to the company that takes a lot of the satellite pictures

for Google Earth: Maxar Technologies.

I feel very confident that those are indeed on the ground

because we see them in multiple images over multiple years.

So, I know it wasn't an artifact of the processing that Google might have done with our imagery.

And then a colleague of mine who has spent a decent amount of time studying this area

said, “You know, this is a very rich area for oil and gas.”

“This looks very similar to what we see when they're doing oil exploration.”

Oil radically changed the course of Algeria's history.

"Oil from the wastelands of the desert..."

"And it's believed that the Sahara is immensely rich in it."

When oil and gas were discovered there in 1956, companies flocked to the region

against the backdrop of a brutal decolonization war with France.

Today, Algeria is one of the world's top exporters of natural gas.

What Steve is talking about here is seismic surveys

where geophysicists analyze the Earth's surface by sending shock waves into the ground.

Depending on how those seismic waves bounce back

researchers can tell what resources can be extracted from underground.

Steve thought that, maybe, seismic pulses from a specialized vehicle

could produce something like this.

So, we had a hypothesis.

But I wanted a second opinion.

So I asked Bob Hardage at the University of Texas

one of the world's leading experts on seismic imaging.

He responded by email:

”I can assure you with 100-percent confidence that

the features in this imagery are not seismic arrays

used in oil and gas exploration.”

First, the shapes themselves weren't right.

“...there will be hundreds of thousands of receivers positioned

as either a single straight line

or as hundreds of parallel straight lines.”

I looked up pictures from NASA of seismic surveying

and you can see what he means.

Second: the fact that we could even see them meant they probably weren't a seismic survey.

“... the objective is to leave the landscape like you found it."

"If a seismic crew created something like these features

a return visit would be made to restore the landscape.”

“I have no idea what the circles in the satellite image are."

"Whatever they are, the people who created them

wanted those features to be permanent.”

“Closeout: I don't think we need to chat.”

Thanks Bob.

So I kept Googling.

I found geotagged pictures from the nearest municipality, Foggaret Ezzaouia

on a site called mapio.net.

These old stone wells sorta looked like they could be arranged in a circle.

But reverse image searches were a dead end.

I didn't know what to do next.

So we looped in Vox video's senior researcher, Melissa, to help me out.

So, I was trying to find what this thing was.

I don't know if you remember from his original post

he calls it the Tebalbalet tomb.

Do you remember that?

So I found this article.

This is from like 1985 — I mean, not 1985: 1885.

The “Well of Tebalbalet” is at the latitude 27°20 and longitude 4°38.

And that's approximately where what we're looking at is.

And it says there are two circular tumuli.

I had to Google that, I don't know that word.

-Tumuli.

What's a tumulus.

Tumulus.

It's an ancient burial mound.

Which seems... that sounds about right.

“... encompassed by two concentric mounds in the form of rings, all of great regularity."

"The two rings are respectively 30 and 21 meters in diameter, from crest to crest.”

So a document from 1885 said that, around this same area, there were

1) a bunch of wells, and

2) tombs with “rings of great regularity.”

Now, the sketches weren't an exact match.

But they got us thinking: what if these things were actually really old?

So I sent the pictures to a Tunisian archaeologist who had done research in this area.

We spoke in French because of decades of French occupation in the 19th and 20th centuries

French is still used in many contexts in Tunisia and Algeria.

And she had a new clue.

[in French] These monuments, they are without a doubt

[in French] because I know Aïn Salah very well...

[in French] These monuments are related to...

[in French] Water.

[in French] It's a desert environment, it's the Sahara.

[in French] It is practically the hottest place in the Maghreb.

[in French] It's an area which is very well known

for the difficulties of this heat there, and for the water harvest.

[in French] So the people, they dig.

[in French] It has a name: the Foggaras.

Foggara.

It's the North African name for a 2,500-year-old style of irrigation system

that goes by many names, but is often called a qanat.

Builders dig a well at an elevated point on a slope

deep enough to tap into groundwater.

They then dig parallel shafts at regular intervals.

These provide air flow for diggers as they create an underground channel

all the way back to the main well.

With a slope of 1 or 2 degrees, the channel carries water long distances

powered by gravity alone.

In a part of the world with barely any rain and no running rivers

this technology can provide water for crops, livestock, and people

year round...

making human-made oases possible.

[in French] It's curious, eh?

This was the most promising lead yet.

It explained the desert location, the circular shape, the regularity, and spacing.

Even the closest municipality's name, Foggaret Ezzaouia, is named after foggaras.

And those mapio pictures of wells started to make sense.

But I wanted to run it by more people who had studied qanats.

Qanats are actually more than just water infrastructures.

I think they are the very raison d'etre:

the basis of habitation in such harsh climates.

They start from outside of the city, but then they usually end up

into the city or into agricultural lands.

But when it came to our circles...

I have no take on it, honestly.

I'm looking at it now.

Right.

Okay, that's interesting.

There's something like 20 of them in a row.

Yeah. So that's definitely a foggara.

So at the end of that, there should be a town.

There should be an oasis or something.

But if there isn't, that means that probably the water in the qanat

or foggara has dried up since a long time.

You should talk to Dale Lightfoot.

He is the American geographer

who knows everything about qanats.

These are what we're looking at.

I couldn't even say with confidence whether these are related to water collection.

But I can tell you they're definitely not qanats.

We also found these pictures.

Do you think these could be what the circles are?

What you're showing me pictures of here looks a lot like animal-drawn wells.

I've seen these in a lot of places.

To me, this is not the same thing.

I think you're back to square one.

Back to square one, indeed.

Don't rule out space aliens.

I've heard they do crazy things, too.

So they might be wells, but probably not a qanat.

And maybe not even related to water at all.

Could we at least rule that out?

That's when Melissa found a database of oases in the Sahara.

With lists of the people who help manage their water supply.

Like Mohammed Brik, a farmer in Laghouat, Algeria.

I don't think it was done to fetch water.

Because the point of going out to look for water

is to meet the needs of the population

and agriculture.

If there's nothing for 160 kilometers

then that's not a valid hypothesis.

Right. Because there is no village, no...

There's no village.

There's no garden.

There's no oasis.

There's nothing planted.

There's no population.

We were three months in and it seemed like our most promising hypothesis yet

was probably out.

Then I got an email.

Back in early October, Steve Wood promised to send me

high-res images from Maxar's archive.

Finally, we had them.

It was the clearest look we'd had yet.

And Steve believed it showed a new detail: tire tracks.

If that was right, it would mean someone had been there within the last century.

I kept asking people.

Historians...

Algerian officials...

Archaeologists...

And nearby residents...

But after a while, I felt stuck.

Like we had exhausted what we could find out on the internet.

And there was nowhere else to go from here...

except to the circles themselves.

The longer this project went on, the more I realized that we had a choice to make.

We could keep interviewing more and more people, get more and more theories

and ultimately have no way to back them up.

Or... we could figure out a way to get someone there...

Try to film it...

and then, maybe, we could know for sure.

So I asked my teammate Christina — who works with journalists all over the world

if she knew anyone in Algeria.

And that led us to Samir Abchiche, a video journalist in Algiers.

I'm about to be a dad.

So no more adventures for me after this.

We hired Samir to be our on-the-ground journalist...

to use his expertise in the area to help us solve this mystery.

The next part took months.

We knew this wasn't going to be an ordinary video shoot.

We were asking him and his team to travel incredibly far

to go do something potentially dangerous.