They found evidence of 37 cases of regional abrupt changes in the ocean, sea ice, snow cover, permafrost and terrestrial biosphere. Many of these events occur for global warming levels of less than two degrees, a threshold sometimes presented as a safe limit. Although most models turn out to predict one or more abrupt regional shifts, any specific shift typically appears in only a few models.
"This illustrates the high uncertainty in predicting tipping points" says lead author Sybren Drijfhout. "More precisely, our results show that the different state-of-the-art models agree that abrupt changes are likely, but that predicting when and where they will occur remains very difficult". "Also, our results show that no safe limit exists and that many abrupt shifts already occur for global warming levels much lower than two degrees."
Examples of detected climate tipping include abrupt shifts in sea ice and ocean circulation patterns as well as abrupt shifts in vegetation and marine productivity. Sea-ice abrupt changes are particularly common in the climate simulations. However, various models also predict abrupt changes in Earth system elements such as the Amazon forest, tundra permafrost and snow on the Tibetan plateau.
"Interestingly, abrupt events could come out as a cascade of different phenomena" adds Victor Brovkin, CliSAP researcher and co-author from Max Planck Institute for Meteorology. "For example, a collapse of permafrost in Arctic is followed by a rapid increase in forest area there. This kind of domino effect should have implications not only for natural system, but also for society".
"The majority of the detected abrupt shifts are distant from the major population centers of the planet, but their occurrence could have implications over large distances" says Martin Claussen, also CliSAP, director at the MPI-M and one of the co-authors. "Our work is only a starting point. Now we need to look deeper into mechanisms of tipping points and design an approach to diagnose them during the next round of climate model simulations for IPCC".
Chris Huntingford of the UK Centre for Ecology and Hydrology adds: "We appeal to those designing the CMIP6 experiment format to consider a requirement for larger ensembles of simulations. This will enable better understanding as to how robust our specific timings are for each identified system tipping point".
Drijfhout, S., Bathiany, S., Beaulieu, C., Brovkin, V., Claussen, M., Huntingford, C., Scheffer, M., Sgubin, G., and Swingedouw, D. (2015): A catalogue of abrupt shifts in IPCC climate models. PNAS early online publication,