Per-decade probabilities are easy to under-read, so do the compounding slowly. Jehn cites estimates for a repeat of the 1859 Carrington storm at roughly 1% to 12% per decade [7]. Treating decades as independent draws, the century figure is 1 - (1 - p)^10: about 9.6% at the low end, about 72% at the high [13]. Set that beside the 1-in-6 estimate for a magnitude-7 eruption and the chance of at least one of the two arriving this century runs from about 25% to about 77% [14]. The width of that range is the honest part of the answer. Independence across decades is my assumption rather than a measurement, so treat the top of the band as soft.
What the arithmetic does not give you is severity mapping. A 1-in-6 eruption is not a 1-in-6 chance of losing 90% of global calories; that figure belongs specifically to the most extreme nuclear-war scenario researchers have modelled [8]. The events share a mechanism rather than a magnitude: particles lofted into the upper atmosphere, then cooler and drier growing seasons, with losses that can run as far as total harvest failure [9]. A long blackout reaches the same destination by a different route, taking out fertilizer, pesticides and fuel where the growing is densest [10].
The omission looks less like ignorance than like scope discipline. A register is a planning instrument, and a scenario with no available response tends not to survive drafting. The UK document holds its reasonable worst case at over a thousand deaths and tens of billions of pounds [5], which is a magnitude a national resilience system can actually be sized against. Jehn concedes that the UK and Switzerland are the strong performers here and still fall short [12], which is the uncomfortable part of the claim rather than a throwaway.
A small-scale rehearsal is already running. A significant share of the world's traded fertilizer moves through the Strait of Hormuz, and the UN Food and Agriculture Organization has warned that the resulting shortfall will cut yields into 2027 [11]. That is the same transmission path a sunlight loss would use, at a magnitude the system can absorb. It confirms the ripple is real; it says nothing about the tail.
Where the piece thins out is price. It asks for funded pilots of foods that can be produced at scale without much sunlight or without fertilizer, alongside emergency trade rules agreed before they are needed [3], and it neither costs the pilots nor names the candidate foods [15]. Two of the three steps consume analyst hours and diplomatic capital. The third needs a budget line, and nothing here says how large. This is one researcher's argument in an opinion slot [2], and its central criticism is checkable against a published document, which is more than most risk commentary offers.
