You are here: Home Research Projects & Joint Activities Projects Archive ASSERT

ASSERT – Assessment of Uncertainty in Climate Change Projections

Can we constrain climate sensitivity by observational data from past climate changes? To which degree is our analysis model-independent? How should Monte Carlo ensembles of climate projections be designed to be most useful for decision under uncertainty? The project derives the effects of climate model parameter uncertainties on climate projections for the example of the CLIMBER model hierarchy. Bayesian learning from present-day climatologies will reduce projection uncertainty. ASSERT will utilise a rare strength of CLIMBER2 and add glacial information into this procedure, hence will narrow down projection uncertainty further. Sampling strategies will be derived in order to upscale this information for the more comprehensive CLIMBER3 as well as to serve the needs of several Integrated Assessment projects using CLIMBER-output. Competence on the diagnosis of coexisting equilibria, bifurcations, and adequate sampling methods, given such features, will be strengthened.

Thomas Schneider von Deimling (project leader), Email to:Thomas.Schneider(at)

Eva Bauer, Andrey Ganopolski, Hermann Held (project leader until 2006), Vladimir Petoukhov, Stefan Rahmstorf

ASSERT aims to provide global warming projections that are robust under climate model uncertainty, thereby working at the interface of “climate modelling” and “Integrated Assessment of climatic change under uncertainty”. In practical terms, ASSERT constrains model parameter uncertainty by running huge ensembles of perturbed parameter physics. The focus hereby is on climate sensitivity (CS), a key model characteristic of high importance both to the modelling as well as to the Integrating Assessment (of climatic change) community. Recent studies of CS uncertainty has left the community with a very broad range for CS that would make Integrated Assessment of climatic change a formidable task: societies would have to take into account too many possibilities (degrees of global warming) to adjust to.

A key objective of ASSERT is to reduce uncertainty in CS by accounting for various climatic archives, especially paleo-data from the LGM (Last Glacial Maximum, 21 kyrs BP) and regional temperature changes of 20th Century warming.

Results gained in ASSERT will be used for project PRIMAP for a probabilistic assessment of emission paths.


  • A. Lorenz, H. Held, E. Bauer, T. Schneider von Deimling (2009): "Constraining Ocean Diffusivities from the 8.2 ka event".  Climate Dynamics (in press)
  • Andrey Ganopolski and Thomas Schneider von Deimling. Comment on „Aerosol radiative forcing and climate sensitivity deduced from the Last Glacial Maximum to Holocene transition” by Petr Chylek and Ulrike Lohmann,  Geophys. Res. Lett. 35, (2008)
  • Schneider von Deimling, T., A. Ganopolski, H. Held, S. Rahmstorf. Are paleo-proxy data helpful for constraining future climate change?, PAGES News, 16, (2008)
  • Schneider von Deimling, T., H. Held, A. Ganopolski, S. Rahmstorf. Climate sensitivity estimated from ensemble simulations of glacial climate, Climate Dynamics 27, (2006).
  • H. Held, T. Schneider von Deimling, Transformation of Possibility Functions in a Climate Model of Intermediate Complexity, Advances in Soft Computing 6, 337-345 (2006).
  • T. Schneider von Deimling, A. Ganopolski, H. Held, S. Rahmstorf, How cold was the Last Glacial Maximum?, Geophys. Res. Lett. 33, L14709, (2006).
  • H. Held and T. Kleinen, Detection of climate system bifurcations by degenerate fingerprinting, Geophys. Res. Lett. 31, L23207 (2004).

Document Actions