Scientists question computerised climate-change models[fr] 

Recognising that powerful computer-based simulations are a key element in predicting climate change, a new Institute of Physics (IOP ) report, published on 26 September 2007, shows that leading climate-physicists' views on the reliability of these models differ. The IOP is also urging world leaders "to remain alert to the latest scientific thought on climate change". 

Entitled Climate change prediction: a robust or flawed process? , the report is based on a seminar held in June 2007 that debated the link between global warming and human activity. The aim was to discuss whether climate-change prediction models are sufficiently robust to influence government policies or whether such models are flawed and the planet is experiencing natural changes that humans are unable to influence.

Massachusetts Institute of Technology (MIT ) Professor Richard S Lindzen believes that the "current climate models exaggerate the impact of CO2 on temperature because of a poor understanding and representation of the feedback effects due to clouds and water vapour".

However, Professor Alan J Thorpe from the Natural Environment Research Council (NERC ) said that "there is no reason a priori that model uncertainties should cause systematic overprediction of global average warming". He supports the findings of the recent Intergovernmental Panel on Climate Change's (IPCC) report and concludes that "there is a risk of dangerous climate change resulting from the emission of greenhouse gases". 

However, Professor Thorpe recognises that there is a need for further research on basic climate processes, the biosphere and the interactions between vegetation, marine life and the climate system.

According to the Institute of Physics, the computer models are built around a long-accumulated understanding of the underlying physical processes, but "there are still many aspects that are less well understood".

"Even though there is evidence around us of climate change, and we need to continue to take action now - greater computing power and further debate is needed in order to make the modelling processes more robust to remove current uncertainties," said Tajinder Panesor from the IOP. 

Later this year in December, world leaders will gather in Bali to try to hammer out a successor to the Kyoto Protocol.