Greater Accuracy is Urgently Needed for Climate Modelling and Forecasting

Climate scientists are collaborating with experts in economic theory to improve their forecasting models and assess more accurately the impact of rising atmospheric carbon dioxide levels. Although there is broad consensus that there will be a significant rise in average global temperature, there is great uncertainty over the extent of the change, and the implications for different regions.

Greater accuracy is urgently needed to provide a sound basis for major policy decisions and to ensure that politicians and the public remain convinced that significant changes in consumption patterns and energy production are essential to stave off serious consequences in the coming decades and centuries.

The climate modelling community has become increasingly aware that some of the statistical tools that could improve their modelling of climate change may already have been developed for econometric problems, which have some of the same features. The European Science Foundation (ESF) brought these two communities together for the first time in a recent workshop, sowing the seeds for future collaboration.

"We achieved our goal of bringing together people from two very distant but equally valuable fields," said the workshop's co-convenor Peter Thejll. "It was designed as a one-way session whereby econometricians were supposed to convey knowledge of econometric methods to the climate researchers."

This has already proved highly valuable because economic and climate models require similar kinds of statistical analysis, both for example involving serial correlation where the aim is to predict the future value of a variable based upon a starting value at an earlier point in time. In economics such a variable might be the price of a commodity, while in climatology it might be temperature or atmospheric pressure. In both cases the variables change randomly during successive time intervals subject to varying constraints within a closely defined zone, and therefore can be analysed using similar "random walk" techniques.

"To solve important climate problems related to climate change and change attribution with statistics, these methods have to be used and understood by climate researchers," said Thejll. "We brought together people who understand these problems and had a great, and informative, time."

Thejll is confident the new found cooperation with the econometric community will improve climate modelling and forecasting, but first there is a need to digest some of the new tools and ideas. The aim is to introduce greater statistical sophistication into climate analysis, partly by understanding better the correlation between different aspects of change, for example how one region impacts another. "We first need to see the spread of econometric methods so that we no longer read climate research papers that ignore important statistical problems," said Thejll wryly.

This will lead to an important first step towards better climate change predictions models - understanding the limitations of existing models. "One improvement that can follow from the use of econometric methods in climate research is a better understanding of the level of ignorance we have," said Thejll.

One problem at present is that uncertainties are commonly underestimated, and this makes it very difficult to predict with much confidence even the broad climatic consequences or rising atmospheric carbon dioxide levels. But Thejll hopes and expects that by incorporating the key tools of econometric modelling, climate prediction will become much more accurate and valuable.

The ESF workshop, Econometric Time Series Analysis applied to climate research, was held in Frascati, Italy, in September 2007.

Tell Us What You Think

Do you have a review, update or anything you would like to add to this news story?

Leave your feedback
Your comment type
Submit

While we only use edited and approved content for Azthena answers, it may on occasions provide incorrect responses. Please confirm any data provided with the related suppliers or authors. We do not provide medical advice, if you search for medical information you must always consult a medical professional before acting on any information provided.

Your questions, but not your email details will be shared with OpenAI and retained for 30 days in accordance with their privacy principles.

Please do not ask questions that use sensitive or confidential information.

Read the full Terms & Conditions.