More of the past needed for accurate climate predictions
Press release issued: 1 March 2008
In research published today in Earth and Planetary Science Letters, The Open University Professor of Earth Sciences Bob Spicer casts doubt on our ability to accurately predict future climate changes because we simply don’t know enough about the past. The paper was co-authored by Paul Valdes, Professor of Physical Geography at the University of Bristol and colleagues in Leeds, Sweden, Russia and Austria.
Professor Spicer said: “Our understanding of climate change is anchored to the climate we have today. By studying extreme greenhouse climates of the past (for example, when the dinosaurs lived and there were virtually no polar ice caps) we have found that there is a huge mismatch between geological data and predictions from state-of-the-art climate models, similar to those used by the Intergovernmental Panel on Climate Change (IPCC) to make future predictions.”
Professor Paul Valdes, from Bristol’s School of Geographical Sciences continued: “The cause of the mismatch could be due to weaknesses of the climate models or because we do not understand how plants respond to climate change and higher carbon dioxide. In either case, it suggests that we are probably underestimating the degree of change in the future. The biggest errors are in key areas such as the poles and continental interiors (where we grow our food for example)."
Professor Spicer added: "Despite over 30 years of climate model development there has always existed a mismatch between model results and geological data. It was thought that this was due to model simplicity and that as the models became more comprehensive these differences would diminish or disappear altogether. Unfortunately this has not been the case and suggests that Intergovernmental Panel on Climate Change (IPCC) predictions may well represent "best case" views - the future change is likely to be far greater than the IPCC suggests“.
What Professor Spicer and his colleagues report in the paper quantifies the model data mismatch for a continental interior (Siberia) and shows that the difference is far greater than conceived uncertainties in both the methods used to determine ancient climate and the models. These huge differences exist despite using a wide ensemble of ancient geographies, atmospheric compositions and even energy fluxes from the sun, suggesting that the cause is a fundamental weakness in our understanding of warm environments.
The Late Cretaceous continental interior of Siberia: a challenge for climate models by Robert A. Spicer, Anders Ahlberg, Alexei B. Herman, Christa-Charlotte Hofmann, Michail Raikevich, Paul J. Valdes, Paul J. Markwick, Earth and Planetary Science Letters