More ammo for the climate sceptics

The hottest year on record? We don't know

Scientists have to be precise. It is their downfall when they try to communicate. The media don't want precision, 'if's and 'but's and error bars. They want black and white 'facts'. It's not a trivial problem. If, as a scientist, you over-simplify then you are in danger of getting a twisted message across (and losing the respect of your peers). If, on the other hand, you apply normal scientific caution people switch off and your political opponents rip you to pieces.

This is a problem that climate scientists face all the time. Every time they revise something or hedge their statements with probabilities, or admit there are competing theories, those with a vested interest in playing down climate change wade in and give it to them with all guns blazing. And it wouldn't surprise me if this happens again with the recent announcement that they've changed their mind about what was the warmest year on record. 'If they can't even decide this,' the professional sceptics will crow, 'how can they possibly say what the climate will be like in 50 years time?'

The trouble is, deciding on the warmest year is not a trivial task. It is all very well to ask how the average temperature on the Earth is varying – but how do you find out the average temperature of such a huge body, with such varied weather at any one time? It isn’t actually possible to calculate a meaningful average for the whole world. Apart from anything else, there isn’t a good enough spread of weather stations evenly across the Earth’s surface to achieve this.

Instead, what they do is make use of 'temperature anomalies.' These compare the average temperature for the required year against long term averages using the same weather stations. That way you get a like-for-like comparison and can understand the way temperature is changing without knowing the 'real' average temperature across the world.

The trouble with this approach, producing those disputes over what is the hottest year since records began, is that the chosen year will vary depending on the spread of years you use for your long term average. Hence the fact we used to think the hottest year on record was 1998, but now it is 2010. So when the anti-climate change brigade leap on this, bear in mind it isn't a mistake, it's merely refining the data. It doesn't actually matter how you cut it - the different averaging processes all say that the decade up to 2010 was the hottest since records began. But it's fairly easy to fiddle around with the specific hottest year.

Not error, just the true scientific process. As I've said many times before, science isn't about 'true and unchanging facts' - dogma should be for religion, not science - it is about our best understanding given the data we have at the moment and will always be provisional and open to change in the future.