Eirik Myrvoll-Nilsen
← All publications

Extrapolation from historical data cannot reliably predict the time of a potential AMOC collapse

Morr, Andreas and Ben-Yami, Maya and Groenke, Brian and Schötz, Christof and Cotronei, Alessandro and Myrvoll-Nilsen, Eirik and Bathiany, Sebastian and Rypdal, Martin and Boers, Niklas

arXiv , 2026

DOI PDF

Part of my work on Early warning signals .

Abstract

Ditlevsen and Ditlevsen (Nature Communications, 2023) propose a statistical framework to estimate the timing of a potential collapse of the Atlantic Meridional Overturning Circulation (AMOC) based on extrapolating information from observed sea-surface temperature (SST) variability. Here we examine the sensitivity of their results and argue that four types of uncertainty are insufficiently explored: structural uncertainty associated with the assumed low-order bifurcation model, statistical uncertainty in the model fit, uncertainty in the representativeness of SST-based fingerprints as proxies for the high-dimensional AMOC dynamics, and uncertainty in the underlying data arising from non-stationary observational coverage and dataset preprocessing. Using synthetic experiments and a systematic analysis of alternative fingerprints and observational products, we show that the estimated tipping times are highly sensitive to these uncertainties, and extend several millennia into the future when they are thoroughly propagated.

BibTeX

@article{morr2026extrapolation,
  title = {Extrapolation from historical data cannot reliably predict the time of a potential {AMOC} collapse},
  author = {Morr, Andreas and Ben-Yami, Maya and Groenke, Brian and Sch{\"o}tz, Christof and Cotronei, Alessandro and Myrvoll-Nilsen, Eirik and Bathiany, Sebastian and Rypdal, Martin and Boers, Niklas},
  journal = {arXiv},
  year = {2026},
  type = {preprint},
  url = {https://arxiv.org/abs/2604.20341},
  doi = {https://doi.org/10.48550/arXiv.2604.20341}
}