Effects of a Nuclear Detonation in Low Earth Orbit

104 points by PaulHoule 23 hours ago on hackernews | 87 comments

knotruby671 | 21 hours ago

had to model EMP coupling for a sat com board once, the margins you need for hardening are brutal even for small yields

[OP] PaulHoule | 21 hours ago

What I understand is that EMP hardening is a bear. Like if there is the slightest crack or imperfection it will get through. Also the effects are said to be a bit unpredictable. Poorly protected hardware might survive if you get lucky, well protected hardware might fail if you get unlucky.

Of course that's an entirely different phenomenon: X-Rays from an exoatmospheric explosion are absorbed in the atmosphere and cause a plasma explosion with widespread ionization and charge moving around... with a huge RF electromagnetic effect.

The main effect that this paper is talking about is beta particles from decaying fission products get trapped in the magnetosphere and damage electronics for a sustained time.

consumer451 | 21 hours ago

This is even worse than I had imagined. I always thought that "the good news" is that no country with on-orbit assets would do this, as it would take out their satellites as well.

Aside from the last thing you do before deciding on a nuclear assault on land targets, that's a fair assumption, isn't it?

lumost | 21 hours ago

It depends a bit on whether you can hurt the opponent more than yourself in the strike. e.g. If you have a small satellite network or can survive without it, then why not pull the trigger? There are few methods of using Nuclear weapons which only target viable military/hardware targets.

OgsyedIE | 21 hours ago

Nuclear exchanges are, in theory, the last resort used when the low probability of the state partially surviving in bunkers is less small than the low probability of winning whatever current big-picture conflict is considered existential through conventional military means. It is not practical for states to differentiate each kind of exchange as not "all-or-nothing", they can only safely assume that they are all "all-or-nothing" and plan within that framework.

This is the kind of topic that necessitates a visualization of what the stakes are. If you haven't seen Threads and have two hours free, go watch it.

If you haven't got two hours but at least have half an hour, some Qwen agents comparing many YouTube transcripts to the film script have suggested to me that the summary-type video essay about Threads by the YouTube creator Pugsr is the summary-type video essay that most preserves the thematic content and chronological ordering of the events depicted in Threads, which I canaried by skimming the transcript now and endorse.

The nuclear planners in states and relevant executive actors (at least in the majority of cases) grapple with sober acknowledgement of the kind of downsides depicted in Threads and which, if any, non-exchange outcomes are even worse off for themselves. There aren't many, but most people would falsely assume there aren't any at all.

fhdkweig | 21 hours ago

phainopepla2 | 20 hours ago

> If you haven't got two hours but at least have half an hour, some Qwen agents comparing many YouTube transcripts to the film script have suggested to me that the summary-type video essay about Threads by the YouTube creator Pugsr is the summary-type video essay that most preserves the thematic content and chronological ordering of the events depicted in Threads, which I canaried by skimming the transcript now and endorse.

This sentence is giving me a headache.

order-matters | 20 hours ago

I, on the other hand, greatly appreciate the transparency given here. I think its a bit unnecessary if it was going to lead to a personal endorsement anyway though

DeluluDon | 18 hours ago

Reading comprehension gives you headaches?

OgsyedIE | 15 hours ago

Sorry, 'now and' at the end should read 'and now'.

PowerElectronix | 20 hours ago

As a fun fact, France has a first nuclear strike doctrine, this is, if you invade France, the first thing France will do is to nuke a small village in your country, as a warning to stop.
Will it do that over someone invading French Guiana?

devmor | 18 hours ago

That seems like a bad idea given how many countries have a “if someone launches a nuke, obliterate everything that’s looked at us funny” doctrine.

__MatrixMan__ | 16 hours ago

The French do seem to be a bit more likely than others to call a powerful man's bluff.

AnthonyMouse | 13 hours ago

> The French do seem to be a bit more likely than others to call a powerful man's bluff.

That's assuming that a) they're actually bluffing and b) activating a bunch of never before used execution paths in a complicated, sprawling bureaucratic system whose mandate is to immediately do the thing you're betting it won't is guaranteed to have an outcome you can reliably predict.

PowerElectronix | 10 hours ago

The doctrine is made thinking in Germany and the soviet union. Germany had no nukes so they would stop any aggression on the spot and the soviet union wouldn't risk retaliating as the US armament was ready to erase them from history on a minute notice.

France could never match the destructive power of the soviet union arsenal in a nuclear exchange, but they only needed the soviet union to not want to risk dragging the US into one, the only way for that is a first strike doctrine with vague criteria.

ahartmetz | 9 hours ago

The French doctrine works even without the US. Russia's population is fairly concentrated, especially its economic and cultural elites. France could kill more than half of it. Terrible deal for Russia to risk that in order to conquer France.

ShadowOfThePit | 4 hours ago

As if Russia couldn't wipe out a third of France's economy by nuking Paris.

ahartmetz | 3 hours ago

It sure can, but the point is that the retaliation makes it not worth it.

jjk166 | 15 hours ago

That's not France's doctrine. France's doctrine is if a conflict escalates substantially they will fire a nuclear warning shot in an attempt to regain deterrence. They purposefully leave ambiguous what circumstances would warrant such a warning shot.

https://www.elysee.fr/emmanuel-macron/2026/03/02/deplacement...

PowerElectronix | 10 hours ago

Yes, that is more precise and maybe a land invasion doesn't trigger it. I find the doctrine both absolutely insane and the most effective at deterring aggression.

rcxdude | 9 hours ago

Yeah, the idea is to show that they're serious enough to break the nuclear taboo while still leaving some room to climb down the escalation ladder instead of things escalating to a full nuclear exchange.

PowerElectronix | 20 hours ago

A nuke in space would most likely be viewed more as a conventional attack than a nuclear strike. You are just taking stuff off space, not causing a any direct deaths and not irradiating soil on any country.

I would be surprised if a space nuclear strike isn't carried in the very first stages of a US-China armed conflict. There's just too much to be gained and very little to lose (nobody will retaliate with a nuke over a population to that)

pixl97 | 18 hours ago

Um... depends where the nuke goes off at and how high. In LEO over the US would put us in the dark age. So yea, it's a great way to get flash fried.

hammock | 21 hours ago

This is based on declassified info, of which the last real-world data we got is over 60 years old (Starfish Prime).

What can be done today is certainly very different.

lukan | 21 hours ago

This is partly based on declassified info and laws of physics did not change so much.

hammock | 21 hours ago

No so much for the scenario studied. More so that it’s probable (and undisclosed) that there now exist nuclear weapons 30x smaller than known, with more controllable effects (EMP, neutron, blast radius, fusion ratio, etc)

rbanffy | 20 hours ago

I don't think this makes the result any better. While the destruction can be a little more tailored to a specific target, it still results in a massive infrastructure collapse.

ASalazarMX | 19 hours ago

I thought the holy grail of nuclear weapons during the Cold War was the neutron bomb, because they thought it would erase life while preserving most of the infrastructure. I don’t know if such weapon was achieved, or even if it’s possible, but declassified tech is usually decades behind.

taneq | 17 hours ago

I think it’s an interesting assumption that ‘infrastructure’ is going to be a fungible asset that an aggressor could productively take over. Even in the 70s the actual operation of a power plant or manufacturing facility would have depended on huge amounts of institutional knowledge. These days, think about the effort that goes into onboarding a single employee and the time taken for them to become truly a net positive. Now consider replacing the entire staff with zero handover, I’d expect the effort required to be on the same order as just building from scratch.

lukan | 12 hours ago

"I’d expect the effort required to be on the same order as just building from scratch"

Not really. If you already have machines and manuals, it just takes some smart people with dedication to figure out most of the rest.

Harder then being introduced, but way way cheaper and easy then building new. Most production is not classified experimental research, but operating standard equipment.

Not really. There aren't so many industrial equipment manufacturers. You just need people that worked with the same equipment elsewhere.

I work in a medical lab. We have the exact same analyzers as a lot of other labs over the world: Sysmex XN for hematology, Cobas for biochemistry, Siemens Microscan or Vitek for microbiology, etc. Even using a completely new one is not an issue. They all work the same. Passwords could be a problem, but 99% of the time they're either not used or written on a sticker on the monitor.

rbanffy | 7 hours ago

> 99% of the time they're either not used or written on a sticker on the monitor.

At least disaster recovery is easy.

LorenPechtel | 18 hours ago

Probably not.

The problem is there is a minimum amount of material to make a bomb. To some extent you can substitute higher pressure for less material but mostly you make a very small bomb by a lower multiplication factor causing less of the material to fission. But that is inherently a random thing, the closer you get to not working the bigger the effect of a very minor difference early in the detonation. Thus, while you can make a very small bomb you can't predict it's yield.

ranger_danger | 17 hours ago

Dial-a-Yield has existed since the 50s.

https://en.wikipedia.org/wiki/Variable_yield

beloch | 21 hours ago

This is one reason why weaponizing space is a bad idea. Even hardened orbital weapon platforms will be easy to destroy with imprecise detonations, as shown here, so space-based weapon capabilities will be short-lived in any meaningful conflict. However, their mere presence in orbit makes it far more likely that all other satellites will be destroyed, possibly even in minor conflicts. A strike against offensive military satellites (with all civilian satellites in the area as collateral damage) would likely be viewed as less of a taboo than a strike against cities.

This is a bad path to go down.

PowerElectronix | 20 hours ago

I guess the next best place for a weapons system in space is orbiting the moon. You can detect threats from earth with enough time to intercept and your presence there is don't threaten (indirectly, that is) the earth orbiying satellites.

zzril | 20 hours ago

Makes me think of Stanislaw Lem, Peace on Earth.

nancyminusone | 19 hours ago

Intercept what now? Being at the moon will take 10s of hours to get to earth.

PowerElectronix | 10 hours ago

I mean attacks against the weapons system, but yeah, interception of attacks by the weapons system would be just as easy. Maybe the best position is close to earth but far enough from the van Allen belts to be unaffected by the lingering effects of a nuke.

rbanffy | 20 hours ago

> This is one reason why weaponizing space is a bad idea.

It's a bad idea for anyone who depends on orbital infrastructure. For anyone who doesn't, has no orbital assets of relevance, has a small orbital launcher, and a compact nuke, it levels the playfield by ruining the party for everyone else.

[OP] PaulHoule | 20 hours ago

You don't even need the orbital launcher, you just need a suborbital launcher which can loft the bomb high enough without attaining orbital insertion speed. For instance, this anti-satellite weapon

https://en.wikipedia.org/wiki/ASM-135_ASAT

wasn't bigger than other missiles you could fire from an F-15 but could take down a satellite.

echoangle | 18 hours ago

> wasn't bigger than other missiles you could fire from an F-15

Technically not quite true for the version of the F-15 that was used.

The ASM-135 was 1180 kg.

The heaviest missile the base eagle is carrying is an AIM-120 with 160 kg or so.

The strike eagle (which was developed from the eagle but wasn’t introduced until 1988, the same year the ASM-135 project ended) can carry heavier stuff like the AGM-130 with around 1300 kg though.

dopa42365 | 18 hours ago

Anything that can carry a 1180kg bomb load (not that impressive, any fighter from the last 50+ years can do much more than that) can also carry a 1180kg bomb with engine at one end (missile).

As it turns out, advanced missiles that require the most smartest secretive military grade integration in only the bestest finest aircraft can be controlled just fine from a tablet in a garbage ancient analog MiG-29 if shit hits the fan.

dylan604 | 14 hours ago

Is that really true? Just because a plane can handle a total weight load does not mean it can attach all of that weight to a single mounting spot.

bell-cot | 5 hours ago

Technically - you are correct.

Practically - "able to drop 2000 lbs. bombs" has been a baseline expectation for warplanes for a long time. Wikipedia notes that the F-4E Phantom (a common 1960's interceptor and fighter/bomber) was rated for "Up to 18,650 lb (8,480 kg) of weapons on nine external hardpoints, including general-purpose bombs, cluster bombs, TV- and laser-guided bombs, rocket pods, air-to-ground missiles, anti-ship missiles, gun pods, and nuclear weapons." A bit of division - even if all 9 hardpoints are the same, that's >2,000 lbs. per hardpoint.

rbanffy | 3 hours ago

> Technically - you are correct.

The best type of correct.

Sorry. I’ll show myself out.

asdff | 16 hours ago

Maybe you wouldn't even need that. Find some random satellite with some security vulnerability you could exploit, and maneuver it somewhere to start tipping over all the dominoes.

__MatrixMan__ | 16 hours ago

Another route would be to have some friends in the right places so you can have a contractor deliver the payload to orbit for you.

manarth | 13 hours ago

  > "For anyone who doesn't [depend on orbital infrastructure]"
Which limits it to possibly some isolated tribal communities with little to no contact with globalised society.

Much of globalised society is dependent on orbital infrastructure – even countries which don't have satellites deployed will have long-tail impacts on the loss of orbital communication and sensing.

- Weather forecasts

- Crop planning

- Disaster response

- Mapping and location services

- Communication

- Aviation

- Science observations (sea-water temperatures; currents; AMOC)

- Entertainment broadcasting

- Mind control (ok, I might be making that one up)

rbanffy | 7 hours ago

Using something for convenience doesn't mean depending on it. Most of human recent history happened before satellites.

And we are thinking of state agents here. If a terrorist group gets a rocket and a nuke, they can vaporize a city or vaporize all satellites.

rexskimmer | 19 hours ago

Although LEO would be a widespread scorched-Earth style attack, a capable military would probably consider more strategic attacks targeting GPS satellites in MEO and communications satellites in GEO.

taurath | 18 hours ago

> This is a bad path to go down.

We're there - the people our system has put in charge are the people who care the least for humanity.

asdff | 17 hours ago

Interesting how communications platforms are now pretty much straight dual use with military. Are you going to knock out starlink, used to guide drones, but also deliver needed internet connectivity to the most remote parts of the globe?

The idea that an orbital conflict would last probably hours if that makes the recent interest in the Chinese and Russian lunar base efforts all the more interesting. What is the moon if not a pre made orbital platform that is far more impervious to bombardment? It might not even be such a complicated construction project, hiding some missile chuckers of some design in some lava tubes of sufficient geometry perhaps. I'm not sure if it is too far away to be impractical, however.

King-Aaron | 15 hours ago

I would wager that any conflict that starts knocking out satellites would quickly follow with several megatons falling on population centres. By the time that happens all bets are off and remote area connectivity is probably low on the list of priorities.

defrost | 15 hours ago

Falling _on_ .. or falling _towards_ ?

  In total, 1,344 Starlink satellites have re-entered the atmosphere, according to astronomer Jonathan McDowell. The de-orbiting has raised concerns that vaporized satellites could release harmful ozone-depleting gases, prompting researchers to urge further study of the issue
~ https://au.pcmag.com/networking/118589/260-starlink-satellit...

King-Aaron | 15 hours ago

I mean any shooting war that sees satellites being shot down would also involve nuclear attacks

defrost | 15 hours ago

Would it though?

Taking satellites down is about removing over watch and reducing communications ability .. that doesn't have to lead to to a nuclear exchange.

eg: Russia-Ukraine war has seen satellite capabilities reduced (although not by frag or EMP bursts) w/out escalation to nuclear exchange.

Further, Pakistan in 1998 demo'd air to target nuclear warheads not in the megatonne range.

King-Aaron | 15 hours ago

I would absolutely say so, if we're talking about detonating nuclear weapons in space as per this discussion.

mr_toad | 8 hours ago

> harmful ozone-depleting gases,

Such as?

defrost | 8 hours ago

bawolff | 15 hours ago

Isn't that already true? Spy satellites are already critical in conflict and would probably be destroyed quickly in any peer conflict of major powers.

jdprgm | 13 hours ago

Thinking about this wouldn't a better approach than blowing them up just be a small rocket goes up and just pushes a satellite back towards earth a bit where it burns up on re-entry?

bell-cot | 4 hours ago

Orbital rendezvous, docking, and de-orbiting is vastly more difficult than destruction with a kinetic weapon.

So if your goal is knocking the satellite out of action, you do the latter. Vs. if your objective is demonstrating both your advanced orbital capabilities and your willingness to violate norms - okay, then the rendezvous thing might make sense.

bell-cot | 4 hours ago

> ... space-based weapon capabilities will be short-lived in any meaningful conflict.

That's less an obstacle than we might wish it to be. Especially since the major players will still have both surface-to-orbit weapons, and replacement space-based weapons which they can quickly launch.

> This is a bad path to go down.

True. Unfortunately, history shows that humans all-too-often chose to go down bad paths.

stymaar | 21 hours ago

Here's North Korea own Hormuz Straight.
other than the super-powers, North Korea could/would do it

it's not just what's in LEO that would be taken out

everything within line-of-sight like continental USA would have an EMP that would fry everything

Starfish Prime took out Hawaii streetlights and that's before today's extremely dense chips

Probably would take out a bunch of neocloud datacenters, all GPUs fried

Just like the clean skies after 9/11 grounding, planet would be peaceful for a little while until wars break out

thrance | 19 hours ago

Why would North Korea do it? That would bring much undesired attention to their little insular kingdom. Their strategy has been to talk big while doing nothing for a while now, and it's worked pretty well in leaving them alone.
Two days in a row with nuclear weapons posts! I don't remember the last time this happened...

If you are interested in nuclear weapons like me, you may enjoy the free nuclear war simulator I released a couple of weeks ago:

https://galantrixgames.itch.io/thousand-suns

Chamix | 19 hours ago

How does it compare to nuclear war simulator (https://store.steampowered.com/app/1603940)? I felt it was decent enough, but it got a little lazy in actually modeling what an actual realpolitik etc modern countervalue/force nuclear exchange would look like - it at least included https://github.com/davidteter/OPEN-RISOP as a loadable scenario but didn't actually model individual launch points (among other fidelity omissions) or attempt to model any NATO/US response.

I get it is hard to augment (let alone recreate) the work of an actual strategic target planner, but I feel with LLM-schlepping the RISOP database could be somewhat reasonably updated/inferred for modern targets and then, more ambitiously, ported to China->US, and perhaps ultimately, full NATO v BRICS.

NWS has a more accurate fallout model, but it's not real-time like Thousand Suns. I can't speak of their general damage model because I couldn't find many details on that. Its scenario editor appears to be very flexible but, to be honest, I don't understand that product very well because of the UI.

I made Thousand Suns for basically one thing: simulate the impact of an all out war with a civilian focus. I modeled many aspects of this problem in much detail (the page has more info, as well as the scenario creation guide). Optimal trajectories, MIRVs, interceptions, radar modeling... It's all there.

This all out scenario isn't super plausible but I was interested in showing how destructive even a single submarine could be, and how alternative strategies make some difference but not that much!

I made the configuration file LLM friendly so you could get external sources and create different scenarios. The data model has allowance for power plants, ports, etc but they are not exposed yet. That would make the RISOP more interesting for inclusion but right now the counterforce is focused on nuclear threats and countervalue solely on civilian deaths (using the RISOP would allow more economic destruction with fewer deaths in theory).

PBS Space Time has an awesome explainer on the actual physics of EMP

* https://www.youtube.com/watch?v=ZMjo_UY8vEM

Luke2642 | 20 hours ago

This is actually kinda comforting. If we do need to go back to a simpler time for some technological apocalypse reason, it won't actually be too hard to organise. An ICBMs apogee is easily high enough, even 4500km is possible.

LorenPechtel | 18 hours ago

That would probably be an extinction event.

Alien1Being | 15 hours ago

For human civilsation perhaps.

The animals, birds, fish, insects, slime molds, plants and bacteria would be fine after a few generations to recover from the climate changes.

Schiendelman | 14 hours ago

They wouldn't.

The problem with humans is that we can eat almost everything. And when our agriculture, which is very good at high density food production, fails, we will go back to foraging.

It would be worse than what we do now, to an extent that would be difficult to imagine. 1 billion starving people will eat everything.

0xWTF | 19 hours ago

So, I'm back to going full off grid, well water and HAM radio? Any pointers to good venison recipes?

defrost | 15 hours ago

Back to?

Can't say we ever really left being self sufficient; that said, I can only offer pretty decent treatments for kangaroo tail, legs, etc.

orbital-decay | 19 hours ago

Actual detonation of a 300kt device at 290km, just to give some intuition of what it would cause on the ground. Disabling LEO satellites seems benign compared to that:

https://en.wikipedia.org/wiki/Soviet_Project_K_nuclear_tests

>The electromagnetic pulse (EMP) fused all of the 570-kilometer monitored overhead telephone line with measured currents of 1500 to 3400 amperes during the 22 October 1962 test. [...] The EMP from the same test caused the destruction of the Karaganda power plant, and shut down 1,000 km (620 mi) of shallow-buried power cables between Astana (then called Aqmola) and Almaty (then called Alma-Ata).[3]

fifici375wu | 19 hours ago

Which of the technology leaders of today would you believe would follow through on their scorched earth ultimatums and how many of them have bunkers

wolvoleo | 19 hours ago

Um yeah but this will affect everyone's sats equally so any party that does this will hurt themselves too.

Edit: Except maybe North Korea which doesn't have any sats worth mentioning. But they exist at the convenience of China and Russia and that would become a lot less convenient in that case.

hinkley | 17 hours ago

How much atmosphere do you strip off of the planet with each explosion I wonder.

Clearly they've thought about perturbing the top of the atmosphere enough to increase deorbiting of LEO satellites, but how much gets high enough to be shaved by the solar wind?

Alien1Being | 15 hours ago

Too late now.

All the big players ( US, Russia, China, India) are thought / strongly rumoured to have nuclear weapons in orbit. ( There has been no official confirmation of this)

Realistically, if you are a country worried about being invaded by the USA, it would be foolish to NOT have space nuclear weapons.

Of course, this is contrary to the Outer Space Treaty, signed in 1967 by the United States, the Soviet Union (now Russia), the United Kingdom, and over 115 other nations, explicitly bans putting nuclear warheads or WMDs.

chmod775 | 13 hours ago

Rumor source: some scare-TV pseudo-documentary, whose viewing figures are mainly driven by their famous accuracy on military capabilities*.

*Thanks to their underpaid writer's god-given insights. A few heretics claim the channel's statisticians merely discovered a correlation between viewership and hard alcohol in the staff cafeteria.

nine_k | 15 hours ago

The actual problematic part is not even nuke-specific: not the EMP, not the gamma / x-ray blast. It's electrons trapped in the magnetic field and creating a low-lying and intense additional "radiation belt". It quickly renders satellites inoperable.

A nuclear explosion is one way to produce it, but I can imagine different, slower and likely more precise ways, maybe e.g. by ionizing some innocuous gas released by a spacecraft with a laser / maser from Earth's surface, where the requisite gigawatts are available much more cheaply and discreetly.

incognition | 12 hours ago

There was this concept in science fiction, I think it was from Neal Stephenson's Seveneves, that when a moon shatters, there's enough orbital debris to form a debris field around the Earth and no satellites can get through. The Kessler effect

frumiousirc | 6 hours ago

> Seveneves

That's also where I first learned of it (great book!). The idea was first raised in 1978.

https://en.wikipedia.org/wiki/Kessler_syndrome

Reminding me of this idea got me curious where we are currently given Kessler's model.

The model defines a critical density of satellites. Less dense and debris is pulled to earth and burns up faster than the debris can produce more debris by taking out more satellites. Above that density, a chain reaction of debris breaking up more satellites producing more debris, etc occurs. This critical density depends on altitude, naturally. Lower orbits can drain away debris more quickly than higher orbits. In 2025, Kessler and Lewis found essentially all altitudes populated by satellites are above the critical density for runaway destruction. This was driven by an increase in satellite density, with Starlink called out as the major contributor.

Seems pretty dire to me. However, the model does not predict over what time scale a runaway would progress and does not include satellites that can maneuver to avoid collision.