From John Mitchell, Chief Scientist, Met Office and Robert Watson, Chief Scientific Adviser, Defra
Your article, “We need better forecasts – and fast” (3 May, p 8), is written solely from a weather forecaster’s perspective rather than from the perspective of a climate-change researcher. This is unhelpful at best. It is not true to say that climate models “cannot reproduce El Niños in the Pacific Ocean… nor the Atlantic storm tracks and blocking high pressure zones that determine whether western Europe is wet or dry”.
Most models, including those from the Met Office Hadley Centre, UK, do simulate these features. Climate models will never be able to reproduce features on the same scale as weather forecasting models, but they can still produce helpful results, using multiple model techniques to quantify the uncertainties. The International Panel on Climate Change’s (IPCC) Working Group 1 report assessed regional (but not local) climate change. It gives a clear indication of the level of uncertainty in current climate models.
The article suggests that the IPCC’s long review process means that its assessments are not up to date. However, the IPCC report had a deadline for journal acceptance of July 2006 – and made use of papers with formal publication dates through to early 2007 – not 2005 as stated. The process ensures all the scientific evidence, especially new research, is properly assessed before it is used to inform policy.
Finally, improvements in prediction require an increase in computing power as you state, but this has to be accompanied by a greatly improved understanding of cloud behaviour and the terrestrial biosphere. Until we have that improved understanding, any increase in computing power is best distributed between a number of centres. This maintains a diversity of approach, ensuring we can quantify the uncertainty and risk within our climate predictions.
Advertisement
From Bob Muirhead
You report James Hansen as saying that the Antarctic ice cap only appeared when atmospheric carbon dioxide levels fell to about 425 parts per million, 50 million years ago. He goes on to say that CO2 levels may now rise back to 425 ppm within two decades.
He then suggests that this will represent a tipping point that might trigger rapid loss of the Antarctic ice cap. However, it is unlikely that the state of the ice sheets at 425 ppm of atmospheric CO2 is independent of whether we approach that level from above – as was the case 50 million years ago – or from below, as now.
We are dealing with very complicated non-linear systems with many variables that are also changing in non-linear ways.
Port Melbourne, Victoria, Australia
Exeter, UK , London, UK
