What happens if a nuclear missile is air-destroyed before detonation?

It turns into a modest dirty bomb of weapons grade plutonium. This isn't actually as nasty as it sounds--in the 80s Iraq and Saddam Hussein apparently did a lot of research and testing of actual dirty bombs and found they really were not effective at all (they weren't even found to be likely to kill anyone, it would just make a small area slightly hot radioactively but was easily cleaned up like any other small nuclear fuel mishap/incident). Initiating the full nuclear chain reaction takes a very specific and controlled process which would not occur without the bomb intending it to happen.

In the 80s a Titan 2 ICBM in the south actually exploded in its silo from a fuel mishap. The missile had a full hydrogen bomb on board and it was blown out of the silo and into a nearby field, still intact and unexploded. So in practice we've seen there's low risk for an unintended chain reaction. This is a great documentary on the incident: https://www.pbs.org/wgbh/americanexperience/films/command-an...

Keep in mind that in order for a nuclear explosion to occur, a very precisely timed series of explosions must occur within the device to trigger fission (and then potentially fusion, depending on the type of device). If the inbound missile is kinetically damaged or destroyed before it gets the chance to trigger, then the conventional explosives will likely detonate within the missile in a way that does not reach criticality, resulting in a conventional explosion with a small amount of fallout from the nuclear core that wasn't triggered being dispersed.

Speaking from the experience of being under regular attacks from Russian cruise and ballistic rockets for the last year, I can tell you, that with enough missiles launched, you’ll have all the possible results:

  - some will explode in the mid air with full force
  - some will explode in the mid air with reduced force
  - some will disassemble on the intercept and won’t explode at all
  - some will fall to the ground and explode with variety level of force
  - some will fall and won’t explode at all
  - some will fall to the seas and rivers and may explode much later and make all other kinds of problems

Successfully intercepted ballistic missiles on the final phase will arrive very close to the target, so chances for it to cause damage are much higher.

Are we talking about boost phase, midcourse or terminal phase defense?

In all of these cases, a successful intercept will happen tens if not thousands of kilometers from the target.

A boost phase interception would happen above the launch area. A midcourse intercept would essentially happen in space and a terminal intercept would happen at the edge of the atmosphere.

I'd recommend reading about Aegis, THAAD and GMD. Also see https://en.wikipedia.org/wiki/Ballistic_missile_flight_phase...

Of course there are also stealth bombers and cruise missiles. In the case of stealth bombers, you either shoot them before the bombs are dropped or you don't (Barring some extremely conveniently located point defenses, like C-RAM)

Cruise missiles? Shot down by conventional air defenses hundreds of kilometers from the target if terrain allows.

Hypersonic cruise missiles? Maybe shot down by conventional air defenses.

As an aside, if you're interested in these sort of issues read Command and Control [1]. Far too many accidents have happened and we've been lucky that there have been no accidental detonations. We've leaned heavily into safety systems that were hard to test and often poorly built. Plane crashes, fires, lost inventory, it's all in this book. An entertaining read.

1 - https://www.penguinrandomhouse.com/books/303337/command-and-...

Scenario I: The too-badly-damaged-to-properly-detonate nuclear warhead is now a piece (or maybe lots of pieces) of dangerous radioactive junk. Bad. But many orders of magnitude less bad than a successful nuclear detonation would have been.

Scenario II: Your defenses knocked out the warhead's delivery system, but did not damage the actual (small, tough) nuclear warhead or its control systems badly enough to disable those. My bet is that the standard design spec. is "when fully armed and near-ish to the target(s), if a critical failure occurs in the delivery system, then detonate immediately". Hopefully your missile defenses intercepted it far, far away from anything that you care much about.

If the fissile core is anything like the original Trinity test one, it relies on a solid ball of plutonium metal being compacted to reach criticality (modern ones may be hollow). This means the explosion wavefront needs to arrive simultaneously all around the ball. This is hard! A ton of the computational effort legendarily described by Feynman with his IBM tabulator/multiple coloured cards story was, I believe, just to make this work. Link to the relevant part (the whole lecture is a hoot).

https://youtu.be/uY-u1qyRM5w?t=2800

Anyway the original bomb had a lot of detonators all around that needed to be fired right. Modern ones have only two, from the little I've seen in published imagery.

https://en.wikipedia.org/wiki/Pit_%28nuclear_weapon%29#Safet...

One hopes this is arranged so simultaneous firing of both detonators results in a fizzle, and a precise stagger is needed. Either way though, an external explosive force will just disassemble the thing, not cause it to go off.

The essence of MAD is that a strike is never "the" nuclear missile. It will consist of many missiles, more warheads, and yet more decoys.

https://youtu.be/2yfXgu37iyI?t=136

I guess you would get a small cloud of radioactive dust without fission yield.

Generally you need to create sufficient neutron flow of neutrons of very specific energy to start fission chain reaction. This happens by bringing more of the fissile material together - more neutrons cause more fissile reactions, leading to more neutrons-boom. An explosive fission will push matter away. If you create an explosion that parts the fissile material before most of the nuclei have split, it’s a dud. To get a really high yield you have to shape the geometry of the fissile material very precisely at a very specific moment so that most of the nuclei get flow of neutrons at exactly correct energy. It does not suffice to bring heat to nuclear material, or just lump it randomly together, you need to be very precise so that there is enough of the fissile at the right place at the right time.

The non-nuclear detonators explosives in nukes serve two purposes: push the fissile material together fast enough, and shape the fissile material just-so due to the pressure effect. It’s really, really inmprobable to trigger the nuke by a random explosion other than the trigger one.

Well, after resolving some obvious issues (like “the warhead should execute its firing sequence precisely or nothing will happen”), we enter the realm of highly classified information, and can only speculate how the modern designs would behave (modern nuclear weapons are the product of decades of adversarial evolution, with ABMs, counter-ABM protection, and counter-counter-ABM measures being the focus of research efforts).

Like, some designs were “one point unsafe”, but could be this exploited deliberately as counter-ABM protection? Or: we know that there were many efforts to allow nuclear warheads to detonate simultaneously and close to each other (so neutron flux from earlier detonations would not disrupt other devices’ firing sequences), but how exactly it relates to ABM protection? And so on and so on.

Exact solutions are classified, and what we can learn from unclassified sources is mostly a speculation.

A cannonball of Uranium is ejected at high speed from the asymmetric detonation of its warhead, possibly slightly dented but otherwise intact and subcritical. If it lands in anything soft (a field, sand, etc), it ends up beneath the small crater, if anything like a building it'll have the same effect as said cannonball.

Most interceptor systems are designed to hit missiles in the middle of their flight plan. For the US, that means over the Atlantic or Pacific ocean. So the warhead(s) likely break up and their radioactive materials are dumped into those oceans and the atmosphere above them, either in mostly intact chunks of debris, or in the form of dust when they burn up. If they did detonate (unlikely but not impossible) it would be well away from populated areas...

Bare in mind that no one has actually managed to intercept

Asking because of the renewed interest in a global missile defense system by the US? https://wikipedia.org/wiki/Space_Development_Agency

Lots of references in there, some seem to answer your question.