Return to Website

Number Watch Web Forum

This forum is about wrong numbers in science, politics and the media. It respects good science and good English.

Number Watch Web Forum
Start a New Topic 
View Entire Thread
Re: Update on CACC sceptic alerts

...Best Wishes To John for 2011.

....Thanks for this post,David. Most interesting.
.Keep 'em coming!

I watched the BBC's Horizon~"What is One Degree?"~
An hour long programme which would have been much improved if it had been condensed to 15 mins at most.
Naturally Global Warming/Climate Change had to be brought into it.
At 49.00 mins, I feel that Prof. David Spiegelhalter is pulling some graphical sleigh of hand but I cannot put my finger on it? Any ideas?

Re: Update on CACC sceptic alerts

In reply to Edward, I did see the programme. The Spiegelhalter bit was, if I remember correctly, where he was arguing that you could tell that the mean temperature was going up by noting if the number of very hot days being experienced was going up, and gave an analogy that average height must be increasing if very tall people seemed to be more common. That sounds reasonably OK, the 'model' of weather that most of us probably have in our heads is that temperature, rainfall, etc are random about some mean and that the mean could be systematically shifting. An objection to this might be raised by somebody like Piers Corbyn who claims that extreme weather is caused by solar-lunar effects on the jetstream, and if he is right then that would mean that weather is less random than we tend to think it is.

My overall impression of the documentary was that the probable main purpose of it was as a trial run for using Ben Miller as a science presenter. There seems to be a pattern in recent years where the BBC is trying out science presenters with physics degrees who are connected with the entertainment industry, like Brian Cox (who used to be in a 1990s pop group), Dara O'Briain (Irish comedian who co-presented 'Stargazing Live' a few weeks ago) and Ben Miller (of the Armstrong & Miller sketch comedy show). The reason given for making the documentary (explained at the start) looked a bit bogus to me. Miller supposedly had a conversation at a dinner party with some woman on global warming, and he then realised that he didn't understand how temperature was measured very well. However the content of the first half of the documentary in particular was at about 'O' level physics standard, and it seems unlikely that someone as well qualified as Miller would have forgotten this kind of material, like temperature being a measure of how fast atoms and molecules are bouncing around.

One misleading thing I noticed in the documentary was that anybody watching it might think weather station temperatures are being recorded to 1/50 deg C accuracy, which is the calibration accuracy quoted by Miller for the thermometer. I think the official figure for the accuracy of weather station temperatures is ±0.2 deg C, and that's the standard deviation, the range would be something like ±0.5 deg C. The accuracy range would be expected to be even higher than that for measuring sea surface temperature using relatively crude methods like the 'bucket method' and the 'engine intake method'.

Re: Update on CACC sceptic alerts

Anthony Watts and his crowd over at have been doing some work on temperature measurement at USHCN and the results are pretty horrifying. From memory, about 40% of stations are no better than (bugger, where's the plus-or-minus key on a Wintel keyboard?) +/- 5 deg. C.
One of several other problems with temperature measurement is the recent reduction in the number of stations used to compile the average temperature (if that has any sensible meaning). It is probably valid to adjust temperatures by the adiabatic lapse rate for changes in altitude, but to use urbanised temperature stations when we don't know the magnitude of the urban heat island effect (which probably varies wildly with many local factors anyway) is just asking for problems.
My preferred method of seeing if there has been any general warming is simply to look at long records from those few well-sited and maintained purely rural stations, as Michael Crichton did in "State of Fear". Those won't show you the overall temperature of the planet, but who cares? They do show the trend over time at one spot and if the trends are generally in agreement that is a good indication of the way the world is going.