Daniel sent us this one — he's asking about Pickaxe Mountain, that deeply buried nuclear facility in Iran that keeps popping up in intelligence briefings but somehow stays maddeningly opaque. He wants to know what it actually is, how deep it goes, how fortified it really is, and crucially, how much of Iran's highly enriched uranium stockpile is believed to be stashed there versus at Isfahan and other declared sites. It's a question about a mountain that might be hiding a bomb program, and we still don't know how far down it goes.
That's the thing that gets me every time I read about this site. The most fortified nuclear facility on Earth isn't in North Korea, it isn't in some Russian closed city — it's carved into a limestone mountain about thirty kilometers from Natanz, and we genuinely don't know its full depth. The IAEA quarterly report is due next week, and there are expectations it'll reveal new enrichment levels there, following the February discovery of undeclared centrifuge cascades.
The timing's not great.
The timing's never great with this facility. That's sort of its whole deal.
Before we get into the depth and the fortifications, let's step back and define what we're actually talking about. Because Pickaxe Mountain isn't the official name, obviously.
The official designation is the Shahid Alimohammadi site, and it's carved into the Kuh-e-Dasht mountain range near Natanz. But everyone in the intelligence community calls it Pickaxe Mountain — partly because of the tunneling, partly because Iran's nuclear program has a habit of naming things in ways that practically invite nicknames. The construction began in twenty eighteen, which is when satellite imagery first picked up extensive tunneling activity. Iran didn't publicly acknowledge the site until twenty twenty, and even then, they were remarkably vague about its purpose.
We're talking about something fundamentally different from the above-ground Natanz facility.
Completely different animal. Natanz is a declared enrichment site with IAEA cameras, regular inspections, the whole transparency apparatus — such as it is. Pickaxe Mountain was built to be the opposite of that. It's a purpose-built breakout facility, designed to do enrichment work that you don't want anyone watching. By twenty twenty-two, IAEA inspectors were formally denied access to parts of the site. That's when the real questions started piling up.
The IAEA knows it's there, Iran acknowledges it exists, but inspectors can't actually verify what's happening inside.
That's the shell game in a nutshell. And this is where it helps to think about what a nuclear inspection actually looks like on the ground. When IAEA inspectors show up at a declared facility like Natanz, they have a baseline inventory. They know how many centrifuges are supposed to be there, how much uranium hexafluoride gas has been fed into the cascades, what the expected output should be. They can walk through the centrifuge hall, check the seals on the equipment, verify the surveillance footage, and match the numbers. It's accounting, essentially — nuclear forensics meets audit culture. But at Pickaxe Mountain, they don't have the baseline. They don't know how many centrifuges are spinning, they don't know the feed rates, and they can't verify the seals because they can't access the chambers where the seals would be. It's less like an audit and more like standing outside a warehouse and guessing what's inside based on the heat coming off the roof.
The depth question is where it gets alarming. The confirmed depth, based on satellite radar interferometry — that's InSAR, which measures ground deformation during excavation — puts the main chambers at about eighty to a hundred meters below the surface. That's the confirmed range from the twenty twenty-three to twenty twenty-four construction phase.
Confirmed meaning we can see evidence of it from space.
But here's where it gets murky. The Institute for Science and International Security put out a report in May of last year suggesting there are deeper chambers, possibly at a hundred fifty to two hundred meters. That's based on spoil volume analysis — basically, you look at how much rock and dirt was removed during excavation, and if the volume of spoil doesn't match the confirmed chamber sizes, you've got unaccounted-for space somewhere.
Which is a very polite way of saying Iran dug more than they're telling anyone about.
The geology matters here. The Kuh-e-Dasht range is primarily sedimentary rock — limestone and dolomite. That's actually ideal for tunnel boring machines. It's soft enough to excavate efficiently but stable enough to support large chambers without excessive reinforcement. The TBMs themselves were imported through a network of front companies, which the UN Panel of Experts documented in a twenty twenty-one report. These aren't pickaxes and shovels — we're talking about industrial-scale boring machines that can chew through a mountain at surprising speed.
They used front companies to smuggle in tunnel boring machines, spent years excavating a mountain, and then told the IAEA they couldn't come in and look around. That's the kind of thing that makes inspectors develop a twitch.
Let me push on the geology point for a second. You said limestone and dolomite are ideal for tunnel boring, but doesn't that also make the site more vulnerable to certain kinds of attack? If the rock is softer, doesn't that cut both ways?
It does, and that's an important nuance. Limestone has a uniaxial compressive strength that's significantly lower than granite or basalt — we're talking maybe fifty to a hundred megapascals versus two hundred plus for hard igneous rock. So yes, a bunker buster would penetrate deeper into limestone than it would into, say, the granite under North Korea's Yongbyon facility. But you're still fighting depth. Even if you get fifty percent better penetration in limestone compared to reinforced concrete, you're looking at maybe ninety meters of penetration against a target that's potentially at two hundred meters. The math still doesn't close. The geology helps the attacker, but not enough to overcome the sheer vertical distance.
It's a marginal improvement for the offense, not a game-changer.
And the Iranians clearly understood this trade-off when they selected the site. They chose geology that made construction faster and cheaper, even knowing it offered slightly less protection per meter than harder rock, because they planned to compensate with depth. It's a volume play — make the facility so deep that the geology becomes almost irrelevant to the penetration calculation.
The fortifications are designed to make any military response extraordinarily difficult. The chambers have reinforced concrete linings two to three meters thick. The blast doors are rated for over fifty psi of overpressure — that's the kind of shockwave you'd get from a direct hit by a bunker buster. There are multiple ninety-degree turns in the access tunnels, specifically designed to attenuate shockwaves. If you detonate a bomb at the entrance, the blast wave hits a corner and loses most of its energy before reaching the sensitive equipment.
Like a muffler on a car, but for nuclear explosions.
That's not a bad way to think about it. And then there's the air defense. The site is protected by the Fifth Air Defense Base of the Islamic Revolutionary Guard Corps. They've got Bavar-373 systems, which is Iran's domestically produced long-range air defense, plus S-three hundred PMU-two systems they got from Russia. You'd have to fight your way through multiple layers of air defense just to get a bomb on target, and even then, the bomb probably can't reach deep enough.
Let's talk about that bomb problem, because this is where the math gets uncomfortable.
The US Massive Ordnance Penetrator, the GBU-fifty seven, can penetrate about sixty meters of reinforced concrete. That's the best bunker buster in the American arsenal. If the confirmed chambers are at eighty to a hundred meters, and the suspected deeper chambers are at a hundred fifty to two hundred meters, the MOP can't touch them. It's like trying to kill a bug that's burrowed into a tree trunk by swatting the bark.
Sixty meters is the maximum under ideal conditions — straight vertical drop, optimal angle, no deflection from the geology.
Limestone and dolomite aren't reinforced concrete, so you'd get somewhat better penetration, but you're still not reaching the deeper chambers. The Israelis ran a simulation exercise in twenty twenty-four called Mountain Breaker, testing a five-thousand-pound bunker buster against similar geology. The results were classified, but Janes Defence Weekly got a leak suggesting limited effectiveness. The implication was clear: current conventional munitions aren't sufficient.
This is where the operational planning gets nightmarish. Even if you could penetrate to the required depth — say you develop some new warhead that can reach two hundred meters — you still have to hit the right chamber. The facility isn't a single large cavity; it's a network of interconnected halls spread across the mountain. Satellite imagery can show you the entrance points and the ventilation shafts, but it can't tell you exactly where the deepest chambers are located in three-dimensional space. You'd need extraordinarily detailed geological intelligence, probably from human sources on the ground, to map the internal layout with enough precision to target specific chambers. And that kind of intelligence is the hardest to get.
You're bombing blind, essentially.
You're bombing with partial information, which in military terms is almost worse. You might collapse an access tunnel or destroy a ventilation shaft, which degrades operations, but you're unlikely to destroy the centrifuges themselves unless you get extraordinarily lucky. And luck is not a planning factor.
Which brings us to the centrifuge question. What's actually inside this mountain that makes it worth all this trouble?
The site is designed to host over a thousand IR-six centrifuges. For context, the IR-six produces roughly six times the output of the IR-one, which was Iran's original centrifuge design based on Pakistani technology. So a thousand IR-sixes is equivalent to about six thousand IR-ones. The IAEA's March report estimated that Pickaxe Mountain could reach sixty percent enrichment within two to three weeks of a breakout decision. From sixty percent to weapons-grade at ninety percent is a very short step.
Two to three weeks. That's the breakout timeline now.
Under the twenty fifteen JCPOA, the breakout time was estimated at twelve months. That's how dramatically the calculus has shifted. Pickaxe Mountain is the single biggest reason for that compression. It's a facility designed specifically to sprint to weapons-grade material while being essentially immune to conventional military strikes.
This is different from Fordow, which was the previous poster child for deeply buried enrichment.
Fordow is the obvious comparison, and it's instructive. The Fordow Fuel Enrichment Plant was built under a mountain near Qom, about ninety meters down. It was exposed in two thousand nine, and that exposure was a major intelligence coup. But Pickaxe Mountain is deeper, more heavily fortified, and crucially, it was designed from the ground up as a breakout facility. Fordow was a conversion of an existing military site. Pickaxe Mountain was purpose-built with modular cascades and what appears to be an underground rail system for moving uranium hexafluoride between chambers.
The rail system is a detail worth pausing on. They built a subway for uranium.
The February IAEA report noted unexplained gaps in surveillance footage from the centrifuge hall. Material accounting at the site has been described as challenging at best, impossible at worst. If you can move UF-six between chambers on an internal rail system that isn't monitored by IAEA cameras, you can shuffle material around in ways that make verification a nightmare. You enrich in one chamber, move the product to another, claim it's still feed stock — the inspectors can't track it.
The hiding techniques have evolved well past just putting a door on a cave.
The modular cascade design is key here. These centrifuge cascades can be disassembled and moved within hours. If inspectors are coming, you break down the sensitive equipment, move it to an undeclared chamber, and by the time anyone's inside, there's nothing to see. The IAEA has been complaining about this for years, but the reality is that the inspection regime was designed for fixed facilities, not modular, mobile enrichment capability.
Let's move to the stockpile question, because this is what the prompt is really driving at. How much highly enriched uranium is actually at Pickaxe Mountain versus Isfahan and other sites?
The numbers are estimates, but they're informed estimates based on satellite thermal signatures, known centrifuge counts, and production timelines. As of May this year, Iran's total enriched uranium stockpile is estimated at about sixty-five hundred kilograms, according to IAEA quarterly data. The breakdown is roughly forty-two hundred kilograms at the Isfahan Uranium Conversion Facility, about fifteen hundred kilograms at the Natanz Fuel Enrichment Plant, around eight hundred kilograms at Fordow, and an estimated thousand to fifteen hundred kilograms at Pickaxe Mountain.
Pickaxe Mountain has somewhere between fifteen and twenty-three percent of the total.
The composition matters more than the total. Highly enriched uranium — that's above twenty percent — is where the weapons potential lives. The total HEU stockpile is estimated at about two hundred fifty kilograms, and roughly a hundred kilograms of that is believed to be at Pickaxe Mountain. So about forty percent of the weapons-usable material is in the facility that we can't inspect and can't bomb.
That's the number that keeps intelligence analysts up at night.
There's also the shadow stockpile theory, which gained traction last year. The idea is that Iran has an additional fifty to a hundred kilograms of HEU already at Pickaxe Mountain, hidden from inspectors via undeclared cascades in the deeper chambers. If that theory is correct, Iran could have enough weapons-grade material for multiple devices, sitting in a mountain that conventional weapons can't reach, with no IAEA verification possible.
The shadow stockpile theory — is this based on anything concrete, or is it more of an inference from the spoil volume analysis?
It's an inference, but it's an inference that multiple intelligence agencies have made independently. The logic goes like this: the confirmed centrifuge cascades at Pickaxe Mountain have been operating for years, and the declared product doesn't match the expected output based on their capacity. Either those centrifuges are running well below capacity, which makes no economic sense, or there's undeclared product being diverted somewhere. The deeper chambers that the spoil volume analysis suggests would be the obvious place to hide that material.
Either Iran built a massively expensive enrichment facility and is using it inefficiently, or there's a hidden stockpile. Occam's razor suggests the latter.
There's precedent for this. In twenty twenty-three, the Tehran Times — which is close to the regime but not an official mouthpiece — leaked what appeared to be a Revolutionary Guard document outlining a thirty-day breakout plan specifically using Pickaxe Mountain's cascades. The document described a scenario where Iran could reach ninety percent enrichment within a month of a breakout decision, using cascades that were not declared to the IAEA. The regime denied the document's authenticity, but the level of technical detail was hard to dismiss.
A thirty-day plan is terrifyingly specific.
It aligns with the technical assessments. If you've got a thousand IR-sixes, and some of them are already running at sixty percent, the final sprint to ninety percent is measured in weeks, not months. That's the strategic reality that Pickaxe Mountain creates.
How does Isfahan compare in all of this? Because Isfahan has the largest share of the stockpile, but it's a declared site with IAEA cameras.
Isfahan is the Uranium Conversion Facility — it's where they turn yellowcake into uranium hexafluoride gas, which is the feedstock for centrifuges. It's also where they do some enrichment work. But Isfahan is above ground, it has IAEA cameras, and it's vulnerable to airstrikes. In a conflict scenario, Isfahan would be hit in the first wave. Pickaxe Mountain is the insurance policy. If Isfahan and Natanz are destroyed, the breakout capability survives underground.
The stockpile distribution reflects a deliberate hedging strategy. Keep most of the material at declared sites where it's visible and verifiable, but keep the breakout capability — the centrifuges and the weapons-usable material — in the mountain where nobody can touch it.
That's exactly what the hedging strategy looks like. And it's been remarkably effective. The international community is stuck between acknowledging that Iran is building a breakout capability and being unable to do much about it. Military strikes can't reach the deeper chambers. Diplomacy has been dead since the JCPOA collapsed. Sanctions haven't stopped the tunneling. Cyberattacks — and there have been several, including the Stuxnet precedent — can disrupt operations temporarily, but they can't destroy centrifuges that are buried under a hundred meters of rock.
Let's talk about what could actually work against this facility, because the misconception I keep seeing is that Pickaxe Mountain is invulnerable. It's not.
It's not invulnerable, it's just very, very hard to neutralize. You can bomb the entrances. If you collapse the access tunnels, you trap everything inside, and the centrifuges stop spinning because there's no power, no ventilation, no way to get personnel in or out. The problem is that entombing isn't the same as destroying. Iran could eventually dig new access tunnels. It would take years, but the material would still be there.
You're buying time, not solving the problem.
You're buying time, and that's not nothing, but it's not a permanent solution. You could also target the ventilation shafts. Underground facilities need massive air handling systems — centrifuge halls generate heat, and without ventilation, the temperature rises to the point where equipment fails. If you destroy the ventilation infrastructure, you force a shutdown. You can hit the power lines, the diesel generators, the cooling systems. None of these are as satisfying as blowing up the centrifuges themselves, but they can degrade the facility's operational capability.
There's the cyber dimension.
The cyber dimension is significant. Stuxnet showed that you can destroy centrifuges by manipulating their control systems — making them spin too fast, or oscillating the rotation speed until the rotors shatter. The challenge at Pickaxe Mountain is that the facility is likely air-gapped, meaning its control systems aren't connected to external networks. That makes cyberattacks much harder. Not impossible, but you'd need human intelligence on the inside to deliver the malware.
Which brings us to the sabotage campaigns. There's been a pattern over the last few years.
Israel has been running a sustained sabotage campaign against Iran's nuclear program for over a decade. The twenty twenty Natanz explosion that destroyed a centrifuge assembly hall. The twenty twenty-one power outage at Natanz. The assassination of Mohsen Fakhrizadeh, the program's top scientist. The poisoning of IRGC officers involved in the nuclear program. But Pickaxe Mountain represents a different challenge. You can't sabotage what you can't access. The facility's physical security is designed to make infiltration extraordinarily difficult.
We're back to the fundamental problem: we know it's there, we know roughly what's inside, and we can't do much about it.
That's the paradigm shift that Pickaxe Mountain represents. For decades, nuclear nonproliferation relied on two pillars: diplomatic verification through the IAEA, and the credible threat of military force. Pickaxe Mountain breaks both pillars. The IAEA can't verify what it can't see, and military force can't reach what's buried deep enough. This is the blueprint for twenty-first-century proliferation.
Let's get into one more misconception. There's a tendency to assume all of Iran's HEU is at Pickaxe Mountain because it's the scariest facility. But the numbers don't support that.
Only about forty percent of the HEU is believed to be there. The rest is at Isfahan and Fordow. But the distribution matters less than the capability. Pickaxe Mountain has the centrifuges and the infrastructure to take low-enriched uranium all the way to weapons-grade in a matter of weeks. Even if most of the HEU is elsewhere, the mountain holds the breakout key. That's what makes it strategically decisive.
Fordow, which used to be the big concern, is now basically the decoy.
Fordow is still significant — it's got about eight hundred kilograms of enriched material — but it's a known quantity. It was exposed, it's monitored, and its depth, while substantial, isn't beyond the reach of improved bunker busters. Pickaxe Mountain is the next generation. Deeper, more heavily fortified, modular, rail-connected, designed from the start to defeat both inspectors and bunker busters.
The North Korea comparison is instructive here.
North Korea's Yongbyon facility used similar tunneling techniques to hide plutonium production in the two thousands. The North Koreans learned early that underground facilities are the best way to defeat satellite surveillance and IAEA inspections. Iran has clearly studied the North Korean playbook and improved on it. The modular cascade design, the rail systems, the depth — these are evolutions of techniques that Pyongyang pioneered.
Now Iran is exporting that knowledge.
That's the diffusion problem. The technology and techniques used at Pickaxe Mountain — deep tunneling with TBMs, modular centrifuge cascades, underground rail systems for material movement — these are transferable. Any state with the resources to dig deep enough and the willingness to defy the IAEA can replicate this model. Pickaxe Mountain isn't just an Iranian problem; it's a template.
What does this mean for the future of nonproliferation? Let's break down the key takeaways.
The first and most important takeaway is the breakout timeline. Under the JCPOA, Iran was twelve months away from a nuclear weapon. Today, with Pickaxe Mountain operational, that timeline has collapsed to two to three weeks. That's not a gradual erosion of the agreement — that's a complete transformation of the strategic landscape.
Two to three weeks from a decision to a device.
That timeline assumes the international community detects the breakout immediately. If Iran uses the undeclared cascades in the deeper chambers, they might get a head start before anyone notices. The shadow stockpile theory suggests they might already have enough HEU for a device, which would reduce the breakout time to zero — they'd just need to assemble the weapon.
Which is a very different kind of problem.
The second takeaway is about deterrence. The US and Israel are now investing in technologies specifically designed to counter deep underground facilities. Directed energy weapons that can disable ventilation systems from standoff ranges. Seismic monitoring networks that can detect the characteristic vibrations of centrifuge cascades through solid rock. These are technologies that didn't exist at scale five years ago, and they're being developed specifically because Pickaxe Mountain has rendered traditional deterrence tools obsolete.
Seismic monitoring is particularly interesting. Centrifuges have a specific acoustic signature.
A cascade of IR-sixes spinning at supersonic speeds produces vibrations that propagate through rock. With sensitive enough seismometers and good geological modeling, you can potentially detect enrichment activity, estimate the number of operating centrifuges, and even track changes in operation over time. The US Geological Survey and the Comprehensive Nuclear-Test-Ban Treaty Organization have been deploying enhanced seismic arrays in the region. It's not as good as having cameras inside, but it's better than being completely blind.
The cat-and-mouse game continues. They dig deeper, we deploy better sensors. They build modular cascades, we develop seismic fingerprinting.
The third takeaway is about what listeners can actually do to stay informed. The IAEA quarterly reports are public documents — the next one is due June fifteenth. They're dense but readable, and they contain the official data on enrichment levels, stockpile sizes, and inspection status. The Center for Strategic and International Studies, CSIS, regularly publishes satellite imagery analysis tracking construction changes at Iranian nuclear sites. Groups like the Institute for Science and International Security put out technical assessments that are more detailed than what you'll get from news coverage.
Following the primary sources cuts through a lot of the noise.
The headlines will tell you Iran is enriching uranium. The IAEA reports will tell you exactly how much, at what level, and at which facilities. That's the difference between being alarmed and being informed.
That leaves us with one final, unsettling question. If Pickaxe Mountain is truly impenetrable to current bunker busters, what options remain?
That's the question that keeps coming up in policy circles, and there's no good answer. One option is diplomatic re-engagement — a new agreement that brings Pickaxe Mountain under some form of verification. But the trust deficit is enormous, and Iran has shown no willingness to open the facility to inspectors. Another option is acceptance — acknowledging that Iran has achieved nuclear latency and focusing on deterrence rather than prevention. That's a bitter pill for the US and Israel, but it may be where we're heading.
Or there's the third option, which nobody wants to say out loud.
The third option is military action that goes beyond conventional bunker busters. That's a threshold nobody is eager to cross. And even then, you'd need to know exactly where the deepest chambers are, which we don't. You can't target what you can't locate.
We're left with a mountain full of centrifuges, a stockpile we can't verify, and a breakout timeline measured in weeks. That's the strategic reality of Pickaxe Mountain.
It's a reality that's going to shape nonproliferation policy for the next decade, minimum. The technology that made this facility possible — the deep tunneling, the modular enrichment, the rail-based material handling — that's not going away. Other states are watching. Pickaxe Mountain is a proof of concept.
A proof of concept buried under a hundred meters of limestone. That's the world we're living in.
Now: Hilbert's daily fun fact.
Hilbert: In the nineteen eighties, during the coronation of a traditional ruler in Mali's Dogon region, the new chief was required to sit on a sacred stone while exactly thirty-three elders each placed a single grain of millet on his head, one at a time, to symbolize the weight of ancestral judgment.
...thirty-three grains of millet.
That's a remarkably specific ritual.
If you want to dig deeper into Iran's nuclear infrastructure, check out our episodes on the Fordow facility and the history of the JCPOA. Links are in the show notes. This has been My Weird Prompts. I'm Herman Poppleberry.
I'm Corn. We'll be back next week.
