Thread by @sovietologist

Russia's reverses in Ukraine and Putin's nuclear saber-rattling have stoked anxieties that Russia might use its non-strategic nuclear weapons. An under-appreciated consideration is that we can't predict the physical consequences of such an attack confidently. 1/25

What we think of as "nuclear weapons effects" such as blast and fallout are incredibly complicated physical phenomena that result from the interaction of the radiation and materials emitted by the detonating weapon with the matter in the surrounding environment. 2/25

Modeling nuclear weapons effects from first principles is extremely difficult even with modern supercomputers. 3/25

But if that's the case, how did analysts during the Cold War develop the nuclear weapons models we use today, such as those in Glasstone and Dolan's The Effects of Nuclear Weapons? 4/25 atomicarchive.com/resources/docu…

Part of the answer, of course, is data from atmospheric nuclear weapons tests. But only a few nuclear weapons effects models are derived directly from atmospheric testing data 5/25

Most nuclear tests during the atmospheric testing era were intended for weapons development, *not* for studying nuclear weapons effects. And the tests conducted did not represent many likely military use cases for reasons of convenience or safety 6/25

So the models found in Glasstone or in old military manuals were mostly derived from simplified physical/computational models and validated where possible from the available test data, *not* derived from first principles 7/25

As a consequence, these models could potentially be extremely misleading and fail to predict catastrophic collateral effects 8/25

One such effect is rainout/washout from a nuclear explosion, which could result from the use of low-yield nuclear weapons such as those possessed by Russia 9/25

Glasstone suggests that there is an easily-predicted "fallout-free burst height" above which a nuclear explosion will not produce significant radiological hazards on the surface 10/25

This is based upon the premise that if the fireball does not intersect with the ground and entrain soil to form micrometer-sized fallout particles, those particles that do form will be too small to undergo gravitational settling 11/25 atomicarchive.com/resources/docu…

But gravitational settling (fallout) isn't the only way that radiological contamination from a nuclear explosion can be deposited on the ground: rain or other precipitation can also do so, and much faster if it does 12/25

This matters because of the rapid decay of the radioactivity from a nuclear explosion. 13/25

If that radioactivity is deposited on the surface minutes after a nuclear explosion by precipitation rather than after hours by fallout, a low-yield burst can result in doses to people that would usually be associated with high-yield bursts 14/25

Weapons effects experts at the national labs have been warning for decades that rainout/washout could cause enormous hazards in the aftermath of tactical nuclear explosions 15/25 inis.iaea.org/collection/NCL…

For example, a 1974 study from Lawrence Livermore found that hazards 10km downwind from a 1-kiloton burst could be on the order of 25,000 R/hr. That's ~50x a lethal radiation dose! 16/25

Admittedly that's very much a worst-case analysis, but more realistic modeling also suggests that low-yield nuclear bursts at "fallout-free" heights could result in disastrous radiological contamination on the ground 17/25 osti.gov/biblio/6163450

A 1979 LLNL study modeled a 1-kiloton free airburst over Germany and predicted dose rates on the surface of 600+ rads/hr kilometers downwind. Still potentially very lethal. 18/25

The possibility of rainout/washout unnerving because weather isn't entirely predictable, but it gets worse--it might occur even when it isn't already raining 19/25

Back in the 80s a couple of researchers at Argonne made an argument that a nuclear detonation in sufficiently humid conditions would produce its own pyrocumulus and might rain out its own fission products! 20/25 citeseerx.ist.psu.edu/viewdoc/downlo…

If this analysis is correct, low-yield bursts in relatively humid conditions (including those occurring in Europe much of the time) may be likely to produce these rainout effects even if they are conducted at heights where the old models say they won't produce fallout 21/25

Unnervingly, state-of-the-art modeling has upheld the self-rainout hypothesis. Last year LLNL researchers published a study adapting the Weather Research and Forecasting (WRF) model to nuclear explosions 22/ sciencedirect.com/science/articl…

They use this model to simulate self-induced rainout at Nagasaki, which was a burst *well* above the so-called "fallout-free height" 23/25 ars.els-cdn.com/content/image/…

Rainout is just one of the "known unknowns" that might turn out to be nasty surprises in the event of battlefield nuclear use! Our understanding of nuclear weapons effects just isn't very complete 24/25

As this 2015 Johns Hopkins APL study concluded, "Absent the actual use of nuclear weapons, tremendous uncertainties will inevitably remain in our understanding of the consequences of nuclear weapons use." 25/25 jhuapl.edu/Content/docume…