Have you considered adding a filtered event listener for atomic landmine destruction, to create chain reactions?

I have to look into real-world physics first, as nuclear fission is real-world science.
I have some physics background, but not a nuclear scientist, so I could be wrong in the analysis.

Both gun-type (e.g., Little Boy) and implosion-type (e.g., Fat Man) are finely-designed bombs. They are triggered by some circuitry that ignite all explosives together to create compression such that fission is initiated.
Some external explosions are more likely to create some rupture and scatter the well-placed components, thus disabling the bomb.
Triggering by external neutron sources (like from another atomic bomb explosion) can at most cause a minor part of fissile materials to go fission, thus incomplete detonation. The yield is going to be much lower than a properly triggered atomic bomb. So the scenario of various atomic bombs to get chain triggered is quite unlikely in my educated guess.
They have to be placed very closely to each other, if such effect has a chance to happen at all. Even if it happens, the implicated "slave" bombs are going to have incomplete detonation, their yield is going to be much lower than their designed yield.

The atomic landmine is using very similar trigger mechanism as atomic bomb, as the above still applies.
So I do not think it is a plausible, in terms of science.

Ah, perhaps I shouldn't have used what sounded like a formalism in chain reacting with the title. You're right about the mechanics being questionably physical, unless the detonator is itself highly unstable, and the detonation goes through the normal channels, set off by the blast.

Mainly I was focused on the gameplay aspect. Vanilla reactors detonate if destroyed at maximum temperature, despite the ridiculousness of that. Having landmines which are supposed to detonate in that manner but do not chain react in the same way is inconsistent with the base game, but entirely your call.

Yes, reactor meltdown only means rupture of core, breach of containment structure and escape of radioactive materials into the environment. It is NOT equivalent to explosions made by atomic bombs. In Chernobyl incident, the staff casualties are mainly from the ionizing radiation. ONLY two died from the explosion itself. The explosion blast is really nothing compared to an atomic bomb. (More like the scale of a conventional explosion.)

Thus, I have no interest to further expand such an unrealistic (or wrong) mechanism.
If I were to modify reactor meltdown and add atomic landmine chain explosion myself, I would only attach the vanilla explosion of the vanilla explosive rocket. But I guess this is not what you want.

Indeed not. In fact, if anything I figured it would be more realistic for atomic landmines to have the full blast, compared to the base game's use of it. Not truly because of a chain reaction, but as a consequence of the trigger mechanism being set off (as if by proximity detection false positive, or some other form of instability causing an explosion as designed in function, if not in cause).

Anti-personnel landmines are triggered on pressure. They are designed to trigger on human stepping on them, not small animals like a dog, mouse, etc.
Anti-tank landmines are usually also triggered on pressure (and magnetic). They are adjusted to trigger on heavier vehicles only (as they are much more expensive than anti-personnel landmines). So an infantry with heavy equipment will not set them off.
And no army are so dumb to design the landmines to trigger on blasts so easily. Otherwise mine sweeping will be a very easy job. Just firing a explosive shell into the minesweeping area to clear all landmines? Such things never happen.

I don't know any design of real-world atomic/nuclear landmines. (Their design and deployment would be classified information, I am nowehre to get them.) But in common sense, I doubt if army will allow them to be so sensitive that they will be triggered by blasts. Some skirmish fights firing some explosive shells in the area will then set off all atomic landmines? No, I don't think any army will be so dumb to deploy such things on the battlefield. It does not worth it! Such sensitive landmines will set off by one or two of those flying shells, well before the target enemy tank formations even reach the area.

Guessing from my military knowledge. The primary use of atomic landmines are used to stop enemy tank formations, which are the most valuable targets they can get on land. So the trigger settings are going to be like anti-tank landmines, only triggered by pressure (and magnetic) of tanks, and won't be set off by just blasts.

Conclusion: By design, anti-personnel landmines are already the most sensitive in triggering. But even for that, blasts are not effective to set them off. I doubt atomic landmines would set off by blasts at all.

PS: Maybe they can be finely controlled by remote triggering too, but then they are no longer landmines. (They would then be called remote bombs or charges, which are out of the scope of this mod.)