Climate modelling research requires rebalancing if it is to move quickly to providing better information for tomorrow's decision makers, say researchers from LSE's Centre for the Analysis of Time Series (CATS).
A paper is published today, Monday 13 August, by Professor David Stainforth of Exeter University, a visiting research fellow at the LSE's CATS, Professor Lenny Smith, director of CATS, CATS researcher Edward Tredger and Myles Allen of Oxford University. It is one of a number published in the current issue of the journal The Philosophical Transactions of the Royal Society on climate change.
Their paper discusses the sources of uncertainty in the interpretation of climate model simulations as projections of the future. Assessing these uncertainties is fundamental to extracting information which is of value in industrial and governmental decision making. Over-interpretation of today's climate models risks undermining the value of climate science just as it is beginning to provide decision relevant information.
The authors argue for the advantages of closing the loop between climate science and societal decision makers, and making a clearer distinction between the output of model experiments designed for improving the model and those of immediate relevance for decision making.
'A frank and open discussion of just how stable we think today's model projection are would be of great value to decision makers,' said Professor Leonard Smith. 'There is a lack of clarity regarding the space and time scales on which users should pay attention to model forecasts. The models are improving fast: we can expect regional forecasts for 2050 to change a lot in the next ten years. But how do we maintain appropriate confidence in the models as these expected improvements come online?'
'The answer is more comprehensive assessments of uncertainty if we are to provide better information for today's policy makers,' said Professor David Stainforth. 'Such assessments would help steer the development of climate models and focus observational campaigns. Together this would improve our ability to inform decision makers in the future.'
For the full article, see http://www.lse.ac.uk/collections/cats/papersPDFs/75_Stainforth
See also Climateprediction.net.
Summaries of this paper, and others, appear in the current edition of the journal, Philosophical Transactions of the Royal Society, to be published collectively online. To obtain copies of the papers please contact Laura Dibb on 020 7451 2250 or Clare Kingston on 020 7451 2508 or alternatively email firstname.lastname@example.org
The Centre for the Analysis of Time Series (CATS) was established in 2000 and is based within the Department of Statistics at LSE. The School has a long and distinguished history in time series analysis. See http://www.lse.ac.uk/collections/cats/
For more on uncertainty modelling and climate change, see LSE Magazine, winter 2006, http://www2.lse.ac.uk/LSEMagazine/previousIssues.aspx
Climate change - in June Sir Nicholas Stern, author of the Stern Review on the Economics of Climate Change, rejoined LSE as director of LSE's Asia Research Centre and head of its new India Observatory, see
Nicholas Stern returns to the London School of Economics
Dr Simon Dietz, of the School's Geography and Environment Department, also contributed to the Stern Review.
Professor Leonard Smith is also participating in a British Antarctic Survey event about complexity in Cambridge from 13 to 17 August 2007.
Too late to escape climate disaster? (18 Aug)
Some climate tipping points may already have been passed, and others may be closer than we thought, it emerged this week. At the Cambridge meeting Lenny Smith, a statistician at the London School of Economics, warned about the 'naive realism' of current climate modelling. 'Our models are being overinterpreted and misinterpreted,' he said. 'They are getting better; I don't want to trash them per se. But as we change our predictions, how do we maintain the credibility of the science?
Earth Log - Complex lesson (17 Aug)
The main topic at a conference in Cambridge on Monday on 'complexity' in climate science was why nature in general, and climate in particular, is too complex to capture in even the biggest statistical models. Lenny Smith of the London School of Economics was the pithiest. Resorting to a political lexicon, he said the models were 'not fit for purpose' because the real world contained 'too many unknown unknowns'.
Gambling on tomorrow (16 Aug)
Modelling the Earth's climate mathematically is hard already. Now a new difficulty is emerging
13 August 2007