Pages

Thursday 27 September 2012

Met Office coin-tossers

This is the Met Office effort at predicting global temperatures taken from their own website. I don’t propose to go into it in any great detail, but I think it’s worth remembering how much it costs us to create graphs like this.


 Figure 1: Observed (black, from Hadley Centre, GISS and NCDC) and predicted global average annual surface temperature difference relative to 1971-2000. Previous predictions starting from June 1985, 1995 and 2005 are shown as white curves, with red shading representing their probable range, such that the observations are expected to lie within the shading 90% of the time. The most recent forecast (thick blue curve with thin blue curves showing range) starts from September 2011. All data are rolling annual mean values. The gap between the black and blue curves arises because the last observed value represents the period November 2010 to October 2011 whereas the first forecast period is September 2011 to August 2012.

The poor performance is obscured a little by making the graph much wider than it need be. If we magnify the right hand section we get this.



As anyone may see simply by looking at the actual temperature graph (black lines), there has been no significant trend up or down since 1998 – 14 years and counting. The Met Office model kiddies (white line) missed this hiatus, so the next prediction from 2011 (thick blue line) is for a very sharp rise indeed. To catch up with the earlier 2005 prediction I suppose. Forget the evidence, just keep the narrative going – it’s the bureaucratic way.

As we all know, the Met Office thinks it should have a more powerful computer to generate these fantasies.

Why? Won't a new pencil do just as well?

Their best effort was the 1995 prediction. The 2005 effort was considerably worse because of the continued temperature hiatus which they clearly didn’t expect, can’t explain and in any event disproves their CO2 theory.

How the 2011 prediction turns out remains to be seen, but the Met Office coin-tossers must be keeping their fingers crossed. Actually it is their only option – bet on a temperature rise. Remember that those making this decision will be senior staff and ten years closer to retirement when their latest prediction is compared to climate reality.

They may be lucky, they may be unlucky, only the climate knows and the climate isn’t giving anything away - not even to the mighty Met Office.

7 comments:

Mark Wadsworth said...

Nice one.

Anonymous said...

Now looking at this I thought 'naughty AK - playing graph-games - at his age too'. Then I ran my straight-edge over the 1950 to 2000 numbers and thought - well there seems a bit of a rise. Then I thought well it was flattish from 1950 - 1980 then went up a bit then went flattish from 2000 - 2010 so it might just as easily go back down again - or not.

I must say that in the absence of other data the blue line seems a bit er speculative. But 1950 - 1980 is 30 years and 2000 - 2010 a
mere 10 years. The 1980 - 2000 rise may or may not mean anything. So to be honest I reckon to wait until about 2040 before drawing any conclusion.

BTW I found an old alcohol thermometer a while back that didn't work and the liquid had gone pale brown. So I clipped off the tip and ran out the liquid - it smelled like vinegar. Anyone know if this is common with this type of themometer?

Anonymous said...

Perhaps we should demand the return of the coins they toss as the first step in austerity.

Demetrius said...

Remember 1258!

James Higham said...

This will be linked to in tomorrow's piece about the AGW arbitrary "data".

Unknown said...

O/T ...but sad ( if true). I found the 'old fool' blog when reading your blog and was concerned when he said he wasn't feeling well. The latest entry on his blog is from a family member reporting that he had sadly passed away. Devastating if true as I loved reading his stories. Did you hear anything more ?

A K Haart said...

Mark - thanks.

Roger - I agree - too little good data to make predictions. I'd chuck the ground-based data and stick with satellite. 2040 sounds about right, but more data may still be required.

Maybe the glass in your thermometer was thin enough to be slightly permeable to oxygen? If so, job done.

David - they would have to be sovs!

Demetrius - yes, that's a genuine cause for alarm. No models needed.

James - good - let's keep at it however tedious is sometimes seems.

Lord Monty - I know nothing apart from the comments. I hope it isn't true because I loved reading his stories too. I think I'll keep checking his blog anyway.