Bishop Hill Climate cuttings 44
Dec 5, 2010 With quite a lot of climate and weather stories around, here is another roundup for you.
David Rose in the Mail on Sunday sticks the boot into the Met Office, noting the failure of reality to keep up with their incessant predictions of warming.
Strangely enough, comedian David Mitchell - a man who would normally expected to be "right-on" on these issues - is also letting off a few pot-shots at Britain's weather forecasters, wondering if they would "get it right more often if it stuck to the facts rather than suppositions."
No surprises from Christopher Booker, who reckons it's time for global warming enthusiasts to call it a day.
Leo Hickman looks at a new enviro-TV show for children on the BBC. Apparently Santa's runway is melting and children are to be encouraged to get involved by making online energy pledges. Amusingly, Hickman says that the show is sure "to raise hackles in certain predictable corners". Whoever can he mean?
Judith Curry is also looking at the subject of eco-indoctrination, discussing Michael Olson's post on An Inconvenient Truth, which tears into AGW proponents for not viewing the movie in a critical fashion - not in terms of its factual content - Olson is a green - but in terms of the way it dismissed sceptics:
And then there was Climategate. Literally overnight the, “there is no debate,” voice vanished. The science and environmental communities finally learned there is a debate — not through effective leadership and communication, but by having their noses shoved in it.
Pierre Gosselin was in Berlin for a conference of German sceptics, and hoping to get a cut of the oil money. He was disappointed, but enjoyed the conference.
Reader Comments (44)
Just read the MoS article. Should reach a wide audience and is timed well with the recent Met Office campaign pitched against the back-drop of snow. If temperatures really have been flat for 15 years then surely the uncertainties must be building?
How solid is the data for a flat temperature for 10-15 years?
Pierre Gosselin 's 'conference recap' is a highly recommended read.
James Delingpoles latest... anymore tips, and I'm going to have to charge royalties on Telegraph webtraffic.
http://blogs.telegraph.co.uk/news/jamesdelingpole/100066749/campaign-against-climate-change-a-christmas-appeal/
David Mitchell's Grauniad piece is quite amusing but is careful not to go too far off message.
No recognition of the fact that the agenda-driven MET Office "scientists" use similar equipment and techniques to predict weather conditions in a month's time to the ones used to "project" what the climate will be in 40 or 90 years time.
Was amused to see that in "Peep Show" on C4 David Mitchell's character uses the issue of renewables to start an argument between his desired girlfriend and her current boyfriend with the comment that "someone is making money out of climate change". A joke maybe but perhaps a small sign that a little AGW blasphemy is creeping into the mainstream.
The Now Show Radio 4
Occasionally funny but mainly standard BBC paid left wing comics. My heart sank when I heard the announcer tell us that the well known Climate Scientist, Marcus Brigstock was going to explain how the freezing conditions were caused by global warming. I couldn't face anymore.
See article in today's ST about bigger and better wind turbines. When will the message get through?
Regards
Paul
Barely two months after the inauguration ceremony for Germany’s first pilot offshore wind farm, “Alpha Ventus” in the North Sea, all six of the newly installed wind turbines were completely idle, due to gearbox damage. Two turbines must be replaced entirely; the other four repaired.
http://www.masterresource.org/2010/12/german-offshore-wind-problems/
Any trawler captains out there what to diversify their income by setting up a 'Green Flag' for the North sea?
I would presume the work is dangerous but the income stream would be very generous and extremely regular!
Bish tells me off for the language I use when I consider the likes of Marcus Brigstock. Please don't mention him again or my potty mouth will reveal all.........
@ Lord Beaverbrook
Those trawler skippers could also get into the business of shortening the windmill blades on an annual basis to allow rotation to continue as sea levels rise :-)
Anybody know which Mann paper the MoS is referring to here:
Earlier this year, a paper by Michael Mann - for years a leading light in the IPCC, and the author of the infamous 'hockey stick graph' showing flat temperatures for 2,000 years until the recent dizzying increase - made an extraordinary admission: that, as his critics had always claimed, there had indeed been a ' medieval warm period' around 1000 AD, when the world may well have been hotter than it is now.
For the Mann paper, see here
http://thegwpf.org/the-observatory/163-mann-discovers-medieval-warm-period.html
A bit Off Topic, I know, but I wanted to share my joy at this discovery.
Watching a video of Professor Kevin Anderson's lecture at Leeds Uni, in the subsequent Q&A session a member of the audience suggested reforestation of the Sahara desert. (This is at 36:20 of the 69-minute video at http://www.youtube.com/watch?v=ycD38HRc1gY.) The gentleman explains that it will become rainforest. Nobody laughs. This is not a comedy sketch. I quote: "Get rid of the Sahara and make a great big rainforest there." The Prof wasn't keen on this, concerned about the timber from such projects rotting and causing methane. But if the timber could be incorporated into buildings it would be a better carbon sink. Hmmm.. is he advocating that we should live in log cabins?
The Good Professor explained in his talk that if global temperatures rise by 4C it will be the end of civilization, and that one of the few solutions would be a "planned economic retraction" (at 31:40; he said, "I'm not gonna use the word recession - too negative") on the scale of the collapse of the Soviet Union, but worldwide.
This seems to confirm that an antidevelopment agenda underlies the global warming movement.
@Paul
I also had the misfortune to listen to the Marcus Brigstock rant. It makes your toes curl.
The BBC is turning into a political party. No its worse than a political party - as least the pols have to stand for election and at some stage have to face the real world.
ssat;
It's already built into the design, there's a hand crank that has to be turned twice every Sunday morning to increase the height of the windmill in mm by the current level of CO2 in the atmosphere.
DECC have applied for a new ministerial position to allocate funds to train passing mariners. They will become the Minister for Master Mariner hand cranking and will be expected to provide literature to enable the hand cranking to operate efficiently. I beleive Miss Pope from the MET has offered to assist in a purely informal arrangement.
There is a delightful letter in the 28th July edition of Professional Engineering, to whit "After several weeks of investigation into the optimum number of blades on wind turbines, the obvious conclusion is zero. After witnessing large numbers of these edifices stationary for weeks on end, there can be no other conclusion."
The author then goes on to write that the only maintenance they would require in such condition, would be the occasional lick of paint and for them to be used to fly flags.
See Wind Power: Change of use.
http://5at.co.uk/uploads/Articles%20and%20papers/PE%20Letter%20July%2028.pdf
@ Perry: "See Wind Power: Change of use."
A better idea may be as pier supports;
http://www.piers.org.uk/
And as they are connected to the grid, there will be plenty of power for the illuminations and coincidently, many more venues for comedians such as Marcus Bridgstocke to perform at.
:-)
@ Argusfreak asks:
Depends on who you ask.
Roy Spencer, of UAH, who oversees one of the two interpretations of the satellite atmospheric temperature record says the last ten years is a noisy data set without any discernible trend.
See latest update here:
http://www.drroyspencer.com/wp-content/uploads/UAH_LT_1979_thru_Nov_10.gif
UEA's Phil Jones gave an interesting interview to the BBC in which he said:
See the rest here:
http://news.bbc.co.uk/1/hi/8511670.stm
Others will berate you for your stupidity and tell you that you are misrepresenting the data and should be ashamed etc etc.
@ Argusfreak
I’ve plotted HADCRUT3 global average temperature 2000 – 2010 using the data tools at WoodForTrees.org.
There is a tiny but certainly not statistically significant trend of ~0.02C/ decade.
The trend steepens if you go back to 1995 as Prof. Jones did because of the step change in global average temperatures following the 1998 El Nino. Nevertheless, he is clear that it is - just - statistically insignificant.
The link ran out of the frame so I have had to split it. If you cut and paste it back together and if it still works, you can see this clearly:
http://woodfortrees.org/plot/hadcrut3vgl/
from:2000/to:2010/plot/hadcrut3vgl/from:2000/to:2010/trend/plot/none
No change in annual average global temperature.
Well summed up here
http://thegwpf.org/the-observatory/1626-nothing-wrong-with-our-graph.html
Dec 5, 2010 at 2:39 PM | @ BBD
To calculate a trend you should start a a peak (1998) and end at a peak (2010).
If you start and end at random, you would get a trend from a sine-curve.
Alexej
http://woodfortrees.org/plot/hadcrut3vgl/from:1998/to:2010/plot/hadcrut3vgl/from:1998/to:2010/trend
Sorry. Link overshoot...
http://woodfortrees.org/plot/hadcrut3vgl/
from:1998/to:2010/plot/hadcrut3vgl/from:1998/to:2010/trend
And the trend is - about zero. Which may well have been your point ;-)
Evelyn Waugh on the present madness:
"Despite their promises at the last Election, the politicians had not yet changed the climate. The State Meteorological Institute had so far produced only an unseasonable fall of snow.... The weather varied from day to day and from county to county as it had done of old, most anomalously."
From "Love Among the Ruins", a short story published 56 years ago.
BBD, people get upset when I use 1998 as a start point, so now I go for the last decade:-
http://woodfortrees.org/plot/hadcrut3vgl/from:2001/to:2011/plot/hadcrut3vgl/from:2001/to:2011/trend
BBD - just checked the woodfortrees link you kindly provided and I notice the chart used doesn't have upper and lower control lines as commonly usd in Statistical Process Control Charts - forgive my ignorance but is there a reason for that? I am asking because I am more used to putting time series data into SPC charts (Shewhart charts or control charts) which show whether the "process" is in control and help identify normal and special cause variation (which I'm sure you know!). I have just started to take an interest in the climate change discussions and having looked at a number of time series charts in various links have noticed they tend to be just that rather than Shewhart control charts - I am curious as to why control charts are not used rather than time series charts given they were designed to present data in a way that takes cognisance of variation and help identify what is "normal" and what is Special Cause Variation?
Serge
I’m not sure. While the Shewhart-Deming control chart is a tool for process evaluation there would seem to be a case for upper and lower control limits (eg 3 standard error) for highlighting special cause variation in climate data.
As you say, the best you are likely to find is a mean and bars for 1SD. Possibly the control chart approach is regarded as inappropriately deterministic for climate data?
Can anyone help answer Serge’s question?
WFT provides the basic tools and very useful it is too, but as far as I can see it will only generate a mean. The good news is that it acts as a data portal (like the even more wonderful Climate Explorer maintained by the KNMI here: http://climexp.knmi.nl/start.cgi?someone@somewhere).
So you can import the raw data into Excel (or whatever you are using for data visualisation) and add the UCL/LCL yourself.
Green Sand
Why do they get upset with 1998 as a start point?
As Alexej Buergin reminds above:
Obviously both sides of the debate have made dodgy decisions with selective start and end points for calculating trends in a single time series.
Ole Humlum has some perceptive things to say about this in his critique of the way this was done in AR4 (http://www.climate4you.com/)
His site doesn't let you link directly to the relevant section, but its easy to find.
Look at the blue index bar on the left of the screen. Click the final entry 'Climate reflections', then click '20080911: Is the global temperature increase 1981-2005 unique compared to the general temperature rise since the end of the Little Ice Age ?'
If we make every effort to be fair, then perhaps the best approach is to use the WoodForTrees averaged temperature index (WTI) and go from 1998 to 2011. This is influenced by GISTEMP and UAH/RSS so the decadal trend is higher, about 0.08C/decade, but still not significant.
http://woodfortrees.org/
plot/wti/from:1998/to:2011/plot/wti/from:1998/to:2011/trend
Thanks BBD - I have Minitab and have pasted the raw data into it. It produced a mean of 0.4120, an Upper control Limit of 0.5799 and a Lower Control Limit of 0.2442. What that would mean is that any variation with a value between the UCL and LCL is "normal" (or predictable) in this run of data. There are 16 points where the data indicates Special Cause Variation outside of the UCL or LCL. The moving range (the variation between consecutive points) has an UCL of 0.2062 and a LCL of 0 with a mean of 0.0631. Again this indicates that any variation between consecutive data points within these control limits is "normal". There were 6 occasions when the variation was "Special Cause" and went beyond the UCL.
These were the only SPC tests I applied to the data. As with all data it's interpreting what it means of course!
Serge
Thank you very much for sharing this. Were you using HADCRUT3 global average for 1998 - 2011? And did the excursions cluster in a suggestive fashion or were they apparently random?
BBD - no problem at all, it only takes seconds in minitab to produce SPC charts. I used the data your link provided then the llink from that page to the raw data. that ends at 2009.92. The following is at the end of the data:#File: hadcrut3vgl.txt
#
#Time series (hadcrut3) from 1850 to 2010.83
#Selected data from 1998
#Selected data up to 2010
#Least squares trend line; slope = 0.000296986 per year
1998 0.410272
2010 0.413836
#Data ends
#Number of samples: 2
#Mean: 0.412054
Is that what you mean?
BBD: "The link ran out of the frame so I have had to split it. If you cut and paste it back together and if it still works, you can see this clearly"
BBD: "Sorry. Link overshoot"
You (and others on this thread) appear to have adopted a self-denying ordinance in respect of standard HTML hyperlink syntax. But SquareSpace makes it available -- see the prompt near the end of the "Post a New Comment" form.
BBD
"If we make every effort to be fair...."
Agreed, also thanks to you and Serge for your comments, they are helpful.
BBD additional - from my reckoning the variation between the two stated averages in 1998 and 2010 is 0.003564, which is well within the UCL of the moving range and therefore could be considered "normal" (and cerainly would not be considered statistically significant). It goes without saying that the one constant about the climate is that it has significant variation. In my book when looking at data from a system or process with variation in (and I don;t know one that doesn't) then it is misleading to make comparisons with another period in time (ie comparing 2010 with 1998) because we can't say which, if either was "normal", which, if either was nearer the UCL and which if either was nearer the LCL, and hence the usefulness of SPC. Likewise, using an average is unhelpful for the same reason.
When I look at the minitab chart I can see 3 distinct phases - phase 1 1998 - 2000.92, phase 2 2001 - 2007.67 and phase 3 2007.75 to the end of this data. Phase 1 shows an obvious fall from the high of 1998, phase 2 is stable and phase 3 shows a drop then a rise back to normal parameters. It would be useful to know if these phases correspond with anything climate related, but to make things more complicated any such correlation might have been years before it manifested in these changes. That any help?
Jane Coles
Thanks for the nudge. I knew I was being an idiot over this.
Serge
I'm off to bed so more tomorrow. But, phase 1 1998 - 2000.92 that you identify is the very large El Nino event of 1998. So when you ask: 'It would be useful to know if these phases correspond with anything climate related' then the answer is yes.
If you have time, google 'El Nino' and look at the effects it and its counterpart, La Nina, have on global temperatures.
Sorry to butt in, but:
Why are you all doing statistics and charts over such a short time frame?
I thought most people on both sides of the arguement agree that this is only weather, not climate.
There is a major 60 plus year beat or cycle in temperature.
You need data for some multiples of that to determine where you are in the cycle,
And wether there is any secular trend beyong these 60 odd year fluctiations.
Oh! I nearly forgot -
You must be able to accurately remove UHI from the data before you begin.
Sorry once more for my poor presentation.
My careless typing tends to mask my inability to spell.
I do always proof read my contributions, which does remove about 90% of my errlrd or errors.
AusieDan asks:
If you look back to the original question posed by Argusfreak at comment #1 in this thread, you will find this:
Happy now?
Serge
Sorry for bailing out last night but I have a really horrible cold and had to call it a day.
Anyway, you were asking if and how changes in average global temperature over short intervals relate to climate.
Easily the strongest variable signal in the global average temperature series is the Pacific El Nino/La Nina oscillation (formally El Nino Southern Oscillation or ENSO).
See here
And here
Billy Kessler’s NOAA ENSO FAQ (second link) can be particularly helpful in sorting out specifics.
El Nino causes average lower tropospheric temperatures to rise; La Nina cools them.
To see this in action, go to WoodForTrees and plot the UAH satellite tropospheric temperature from 1995 – 2011.
Eg here
Noting the peaks and troughs in the curve, you see the plot begin with a warming peak caused by the 1994/95 El Nino, then the onset of cooling from the 1995/96 La Nina.
Then comes the towering spike of the ‘super’ El Nino event of 1997/98, followed by three years of cool La Nina conditions 1998/99, 1999/00, 2000/01.
A fairly weak EN in 2002/03 elevates temperatures again. Two more EN of similar scale follow at 2004/05 and 2006/07 interspersed with cooling episodes that were not officially classified as full-blown La Nina.
2007/08 saw the onset of a moderately strong cooling LN and the record ends with the strong warming from the 2009/10 EN. Global average temperatures are now falling and a strong LN is underway.
Sorry this is a little long-winded (and probably not of interest to everyone here), but I hope you can now ‘read’ the temperature curve and understand the pre-eminence of ENSO in the integration of weather events into climate.
Apologies to AusieDan.
Was and is crass.
Sorry.
BBD - no problem, hope the cold is on the mend :) Thanks for the info, very helpful. Any more data you want an SPC of give me a shout.
Serge
Very happy to have been able to help.
You might well find a long lurk in the comments at The Blackboard a good way of getting up to speed on some of the technical aspects of climate - especially discussions of temperature (http://rankexploits.com/musings/).
For scientific and political background framing an important part of the climate 'debate' I cannot recommend our host's book too highly (The Hockey Stick Illusion: see right bar, above).
Have fun...
Dominic
My eyes are getting bad. I misread David Mitchell's quote in the opening essay as "get it right more often if it stuck to the facts rather than suppositories."