A Greenland-to-Europe heat link that the models leave out, and nobody has priced

A planner can do nothing with an underestimate whose size is unstated. The assertion here is that the ensembles behind European heat projections omit a process, not that they run cold by two degrees or by fifteen years ^[4]. The practical effect is to make a symmetric uncertainty band one-sided: the plausible error runs hot, and the paper offers no figure for how far. A design value pinned to the median of a model ensemble still sits at that median. What is now in question is whether the distribution it came from has a right tail.

The chain of events carries a lag, and the lag is the part worth watching. Meltwater leaves the ice sheet in summer, cools the surface through autumn and winter as its freshness blocks mixing with warmer deep water ^[6], sharpens the front against warmer water to the south ^[7], feeds cyclones that make the jet wavier ^[8], and shows up as ridging over Europe the following summer ^[5]. That is roughly a year between an observable quantity and its claimed consequence ^[15]. Greenland mass loss is measured continuously, so if the link holds, a large melt season is a statement about the next European summer rather than a retrospective one.

That is also where the disagreement bites. John Methven of the University of Reading, an author of a July study that found the same southward Atlantic jet shift but attributed it to natural fluctuation due to reverse ^[10], says the connection between the cold blob and Greenland melt is not established and calls it the weak link, pointing to AMOC weakening or atmospheric circulation change as alternatives ^[11]. The two accounts differ in shape as much as in cause. A weakening overturning circulation drifts; meltwater arrives in discrete summers. The preprint's year-matching statistic, an average 81 billion tonnes of extra melt in the summer preceding heavy ridging years ^[2], is the thing that either survives peer review as a pulse signal or does not. If it does not, there is no annual precursor to watch, and the underestimate argument loses its clock.

The stakes are set by what ridging years have already done. The Europe-wide event of 2003 killed 70,000 people ^[9]. Marilena Oltmanns of the National Oceanography Centre, one of the preprint's authors, says the mechanism played a major role in this year's heatwaves ^[3] and expects more drought in some places and flooding in others, with rising variability on top ^[13]. The same North Atlantic cooling is linked in the research to the 2022 Mediterranean derecho, with winds above 60 metres per second, about 216 kilometres per hour, and hail that killed 12 people ^[12][14]. Heat planning is the narrow reading of this; the wider one is that the hazard set attached to a wavier jet is not only hot.