Tuesday
Jul032012
by 
Bishop Hill
Bishop Hill (Lack of) warming since 1995
Jul 3, 2012 Climate: Statistics Climate: Surface At the end of their chat about global warming the other day, Andrew Neil asked James Delingpole and Andrew Pendleton of FOE to send him their understandings of global temperature changes since the end of the last century. The responses have now been posted at Neil's blog.
I'm not sure they leave us any the wiser. The problem (well, one of the problems) with discussing statistical significance is that you have to have a model for what the climate does normally. Since nobody knows what model should be used, it's not clear to me that we can say anything very much about the significance of the changes in temperature in the last decades or the last hundred years.
Reader Comments (48)
Did someone say something?
I noted on the original video that JD mentioned the "tiny" warming "if any" since the first of the Jones' utterances. I think that statistically significant in layman's terms is correctly described as tiny. However, I can see how they like to get the word 'significant' in there.
Basically, bu**er-all has happened in the last 15 years - and they know it.
Cheered me up no end - best laugh of the day since Mrs Cobb announced the score at Wimbledon as bugger all.
One approach is to build lots of model approximations of the climate system and do *lots* of runs with and without CO2 forcing. The cumulative result is completely unambiguous: observed C20th warming cannot be explained using only natural forcings. This is old news.
The most likely reason for the flattening of global average temperature over the last decade is a combination of dominant La Nina conditions, aerosols and the low sunspot count during SC24 with ongoing heat transport into the deep ocean.
BBD, with all respect:
"Cannot be explained" shoud be either "cannot be explained by me," or "hasn't been explained yet."
Delingpole might have his work cut out with this lot:
http://tinyurl.com/d82m9fp
(For those who don't like clicking on links, it's a collection of vicars in the North East grinning as they proudly show off their ecclesiastical stoles embroidered with wind turbine symbols).
Just when you thought you'd seen it all.
Hilarious.
BBD, that's not "the most likely reason", it's four reasons. Until someone can take all "reasons" and make accurate predictions, the whole field will remain at the level of conjecture.
"There isn't now a global temp dataset in the World which shows no warming since 1998." au contraire!
And, as can nicely be seen, it's important to use care when picking them cherries :-)
Now the climate scientist I know would happily acknowledge that temps 'could' drop for a number years (ie a bit of naturally variability) and I could happily agree with them that this would neither prove or disprove anything...
But where would that leave the likes of FOE, Bob Ward, etc..
will they just say for each subsequent year. Its the 15th warmest, followed by the its the 17th warmest, 19th etc, etc there is no trend ;-) !
My theory is:
In several places I have looked at (UK, West Coast of North America) the occasional large enragtive anomalies stopped occurring.
Now they are occurring again.
http://sunshinehours.wordpress.com/2012/06/28/uk-5-year-averages-plotted-using-all-monthly-anomalies/
http://sunshinehours.wordpress.com/2012/06/24/usanoaa-5-year-averages-plotted-using-all-montly-anomalies/
If the warming had been unambigous, we would not be having discussions about statistical significance.
The trend in temperatures has been ambigous. That much we can say...everything else is a waste of time.
The fact that adding or deleting a few stations here and there changes a slightly cooling world into a slightly warming world is all the proof that anyone should need that nothing much has happened to the global mean over the last ten years or so.
The scientists just don't know what's caused the hiatus, or if indeed it is a hiatus. Just like they don't know where the missing heat is ( apparently hiding in deep ocean holes, presumably to evade detection. The lack of warming can also be caused by aerosols put into the atmosphere by the Chinese burning coal. (?) Either way the lack of knowledge of the earth's climate among climate scientists is staggering. We have increased the CO2 in the atmosphere by 30ppm over the last 15 years and it has had no/barkey discernible effect on temperature, and nobody knows why? $100bn well spent, at least the denizens of exotic tropical paradises think so as our friends in the environmental movement swan around the word with carbon footprints as big as Coco the Clown.
Zed and responses removed.
In any case...anybody that checks out which particular year "comes on top" with differences in the order of 1/10 or even 1/100 of a degree, is obviously unaware of the meaning of measurement in the real world.
Yes, exactly. In statistical analyses, an inference is not drawn directly from data. Rather, a statistical model is fit to the data, and inferences are drawn from the model. We sometimes see statements such as “the data are significantly increasing”, but this is loose phrasing. Strictly, data cannot be significantly increasing, only the trend in a statistical model can be.
A statistical model should be plausible on both statistical and scientific grounds. If statistical and scientific grounds for a model are not given in an analysis and are not clear from the context, then inferences drawn from the model are unfounded.
Finding grounds for a model of global temperatures has yet to be done (the only person who has even made an creditable attempt is Demetris Koutsoyiannis). That makes it impossible to evaluate significance. In other words, it makes it impossible to know if the changes in temperature can be reasonably ascribed to random variation. Both James Delingpole and Andrew Pendleton are making a basic error.
One thing puzzles me, if all the temperature records for the entire globe (and I don't understand how you can have a single figure for the whole planet) show that warming has continued, albeit at a slow rate, why do temperature records for specific places, such as those Lucy Skywalker has gathered together from a wide range of places in the Arctic, show a different picture?
Circling the Arctic
http://www.greenworldtrust.org.uk/Science/Scientific/Arctic.htm
This is not a rhetorical question. Orthodox climate scientists are presumably aware of temperature records like those in the link above and therefore they presumably have an answer to the question about how those records differ from the supposedly global trend. What is their answer?
If we can send heat into the deep ocean today, then just maybe, the heat that was sent in sometime earlier is what popped out in the latter half of the 20th century. IOW, we didn't heat it, it heated us.
Climate science is worse than biology in making up post-hoc explanations.
Case closed
http://thegwpf.org/the-observatory/5360-no-global-warming-for-15-years.html
Somewhat OT. Mr Tobis on Planet 3.0, has a video of PBS interviewing Mr Trenberth about the heat in the USA. No mention of jet stream, but linkings to Russian and Australian heatwaves of a couple of years ago, and saying that the heat records are being broken at a 10:1 rate currently. Obviously in the US it is climate change. In the UK, the distinct lack of heat, and the consistent rain is no doubt only weather.
How about - We are in the middle of a natural climate cycle that is warming from LIA to MWP temperatures, where:
LIA = bloody cold, and
MWP = loverly and warm
Meanwhile, Moonbat admits he's wrong about peak oil.
http://www.guardian.co.uk/commentisfree/2012/jul/02/peak-oil-we-we-wrong
I love the potentous way these envirobuffoons announce that, now that even they can see the bleedin' obvious, it is OK to draw conclusions from it.
It's a bit like that pillock - was it Lovelock? - who recently admitted he'd been wrong about nuclear power for 40 years, in tones that implied he expected a round of reverent applause for his belated "insight".
The best comment under the moonbat article was this one:
If you were wrong about peak oil, then you might be wrong about what happens if we use the oil...
Exactly!
407 recommends so far....
This puzzles me also. I no longer trust any of the global surface temperature datasets, and now always have a look at individual stations (with long term records) to see if they correlate with the allegedly 'unprecedented' warming of the late 20th. They usually don't, or if they do show any recent warming trend it is usually no different in magnitude to what was recorded in the early 20th Century, prior to the 1950s cooling. I also tend to think that the global surface datasets have been skewed by the high number of land stations around the North Atlantic, [which evidently has a direct influence over temperatures in the Arctic Ocean, hence all the bollocks about disappearing sea-ice and polar bears in need of suncream and liferafts]. People have short memories but thanks to researchers like Tony Brown there is good evidence that the Arctic has periodic 'mild' spells like those noted by explorers in the 1870s and 1920-30s.
http://bobtisdale.files.wordpress.com/2011/12/figure-101.png
http://noconsensus.wordpress.com/2009/06/16/historic-variation-in-arctic-ice-tony-b/
I looked at the first few responses to Neil's blog. As you might expect from a BBC blog, the true believers were out in force, outraged at any questioning of the true faith.
Thanks Barbara for the comment about the North of England Vicars with their wind turbine emblems. As you say, hilarious. It made me laugh out loud.
BBD and responses removed
BBD
One approach is to build lots of model approximations of the climate system and do *lots* of runs with and without CO2 forcing. The cumulative result is completely unambiguous: observed C20th warming cannot be explained using only natural forcings.
===========================================================================
Herein lies the problem. You start with a hypothesis, build a model based on this hypothesis, run the model, find that the output is consistent with the hypothesis, then claim this proves the hypothesis is correct. Classic circular reasoning !
BBD's argument is only valid if the models are an accurate representation of the climate system. But we know that the models are not correct and the reasons have been extensively discussed in other threads earlier this year.
Your Grace, where have you removed BBD and Zed to? While they may be dissentient to BH, and their thread jacking possibly annoying, is there not a risk that without such human sandpaper, BH becomes just such an echo chamber as RC, SKS, et al? I appreciate that they must incur a huge amount of work for you. I would, however, be interested to hear opinion on the issue.
[BH adds: I appreciate the point about echo chambers. I have been very patient with both BBD and Zed, but we have reached the point where they cannot make any comment without starting a verbal punch up. This is because they have trolled so much in the past. I let BBD comment again in the last few days as an experiment, but his past bad behaviour still seems to prevent meaningful interactions. I don't have time to police this kind of thing. Other commenters who disagree manage to do so without provoking such a reaction. I can live without these two.]
"...run the model, find that the output is consistent with the hypothesis..."
David, may I suggest a change to: ",,,,run the model, "tune" the model until the output is consistent with the hypothesis..." ?
As you say, circular reasoning - and entirely useless.
Did anyone else notice that the FoE graph conveniently ends with the 2010 uptick, claimed by some to exceed the 1998 El Nino record? Anything to spin the facts, as usual.
Dellers should have sent this link.
http://clivebest.com/blog/?p=3843#more-3843
Clive Best has reworked (as of now) that IPCC graph which showed ever shorter periods with ever steeper trends thereby "proving" accelerating warming.
I thought at the time that that cherry pick would turn into a decomposing mess if they left it lying around.
3 July: Guardian: Leo Hickman: Is it now possible to blame extreme weather on global warming?Wildfires, heatwaves and storms witnessed in the US are ‘what global warming looks like’, say climate scientists
VIDEO: ‘The odds are changing’: Kevin Trenberth, a climate scientist at the US National Center for Atmospheric Research, discusses the relationship between weather extremes and global warming on PBS Newshour.
Hickman: I put this question to a number of climate scientists…
Kerry Emanuel, professor of atmospheric science at the Massachusetts Institute of Technology…
Dr Peter Stott, head of climate monitoring and attribution, at the Met Office Hadley Centre…
Professor Michael Mann, director of the Earth System Science Center at the Penn State Department of Meteorology…
Dr Clare Goodess, senior researcher at the University of East Anglia’s Climatic Research Unit…
Dr Doug Smith, who leads decadal climate prediction research and development at the Met Office Hadley Centre…
Michael Oppenheimer, professor of geosciences and international affairs at Princeton University’s Woodrow Wilson School and Department of Geosciences…
Harold Brooks, head of the mesoscale applications group at Noaa’s National Severe Storms Laboratory…
Michael F. Wehner, staff scientist at the Lawrence Berkeley National Laboratory… http://www.guardian.co.uk/environment/blog/2012/jul/03/weather-extreme-blame-global-warming?newsfeed=true
no prizes for guessing what the abovementioned “CAGW scientists” had to say.
back in the real world:
3 July: UK Daily Mail: Graham Smith: Washout summer could lead to rickets epidemic in children not exposed to regular sunlight needed to produce vitamin D
Dr Nicola Balch, an associate specialist in child health at the British Medical Association: ‘People need just 20 to 30 minutes of sun three or four times a week to ensure they get enough vitamin D, but obviously with our weather it can be impossible to get this.’…
The miserable weather has sparked calls from doctors for vitamin D to be added to foods and supplements rolled out nationally…
http://www.dailymail.co.uk/health/article-2168264/Washout-summer-lead-rickets-epidemic-children-exposed-regular-sunlight-needed-produce-vitamin-D.html?ito=feeds-newsxml
--------------------------------------------------------------------------------
It always intrigues me that warming enthusiasts insist on showing temperature trends from 1880 when nearly all of that temperature record has nothing to do with AGW theory.
Since 1980 and after is the time period when CO2 levels supposedly become significant contributors to warming, the focus should be on that period to the present and no other temperature records:
http://www.woodfortrees.org/plot/hadcrut3vgl/last:360/plot/uah/last:360/plot/rss/last:360/plot/gistemp/last:360
Which at the current rate, assuming warming continues, we might see a 1C or so rise by 2100. (= net benefit.)
From the Ecclesiastical Uncle, an old retired bureaucrat in a field only remotely related to climate with minimal qualifications and only half a mind.
Maybe not relevant, but a read through the posts on this thread provoked a thought and an associated question in this old bureaucrat’s remaining cranium:
Thought: The present state of climate science (not the whole green super-faith) seems closely parallel to ancient religion. God occupies the more or less ubiquitous gaps in knowledge and is potentially dangerous.
Question for some historian out there: Did the number of monks, priests and the like vary across the ages and have their numbers from time to time ever been correlated with prevailing social conditions and wars, beliefs, shifts of knowledge etc?
There now seems to be an irrationally large number of climate scientists being funded to pursue their hopeless quest by societies that are losing economic dominance and have to manage gathering impoverishment.
Uncle ... I think that your are on to something here !
One thing that puzzles me is that the IPCC group claim that the temperature went up 0.45℃ between 1975 and 1995. This was made up as an average of the rise in the Northern Hemisphere of 0.65℃, and the rise in the Southern Hemisphere of 0.2℃.
Given the CO2 levels are nearly identical, does this reflect the backwardness of all those countries down south who didn't adopt a carbon tax? Will the new carbon tax in Australia lead to rapid (catch-up) warming?
It can't represent an in-balance in the sources of hot air emitted - we have in Australia quite a number of warmists with chances for a gold medal, should the Olympics include events for fastest or biggest disaster prediction.
This phrase "what global warming looks like" has been the green propaganda "meme du jour" over the few days.
It was headlined yesterday by reliable eco-doomsters Hickman and Borenstein in the UK & US - and echoed around the world by all the other green media groupies
Mysteriously, the expression was simultaneously attributed to Jonathan Overpeck, Michael Oppenheimer and various other scientactivists.
It's quite illuminating to track the manufacture and distribution of this little gem of catastrophist sloganising.
Anthony Watts actually predicted its appearance before it even happened:-
http://wattsupwiththat.com/2012/06/27/more-heated-media-prepping-tomorrow/
.....and, sure enough, it turned up the next day:-
http://wattsupwiththat.com/2012/06/28/what-global-warming-really-looks-like-michael-oppenheimer-fail/
http://www.guardian.co.uk/environment/2012/jun/29/us-wildfires-global-warming-scientists
...... and by yesterday it was around the alarmist ec(h)o -chamber:-
http://wattsupwiththat.com/2012/07/03/the-kevin-trenberth-seth-borenstein-aided-fact-free-folly-on-the-usa-heat-wave/
As well as being an excellent example of how "climate science" is really just the world's biggest activist PR operation - the phrase itself is quite carefully constructed.
"This is what global warming looks like" - can be cheerfully applied to any weather phenomenon by "serious" scientists who wouldn't dare risk making a direct attribution.
If later challenged or disproved - it can equally cheerfully disclaimed "we never said it was global warming - only that it looked like it".
Google "what global warming looks like" to see how effective this kind of lightweight brainwashing is in the meeja jungle.
(It wasn't even original anyway - they dusted off and relaunched a phrase that's been kicking around for a couple of years).
At the risk of being Godwinised - Josef Goebbels would have been proud.
BBD: explain the cooling of the N. Atlantic from AGW: http://bobtisdale.files.wordpress.com/2012/01/figure-102.png
Tisdale has looked at SSTs and they too are cooling in the N Pacific and N. Atlantic: http://bobtisdale.wordpress.com/2012/07/02/the-curious-northern-hemisphere-sea-surface-temperature-anomaly-patterns/
The real answer is probably an increase in cloud albedo.....
spartacusisfree: after-the-fact explanations are logically always-true therefore inconsequential. Leave them to Nostradamus followers.
Shub (3 Jul, 5:16PM) writes: "If we can send heat into the deep ocean today, then just maybe, the heat that was sent in sometime earlier is what popped out in the latter half of the 20th century."
Yep, I've done some work on this subject and have evidence that the Atlantic swallows heat for 99 years and the Pacific for 152. If this idea has legs, it's solar variations in Edwardian and Victorian times which govern today's sea surface temperatures and hence atmospheric. http://endisnighnot.blogspot.com/2012/03/lets-get-sorted.html
Brent, thanks for your link, that's a fascinating conjecture. I can't begin to assess it but I hope that you and others will follow up on it. I have "suspected" from a layman's perspective that there may be much that is not well understood about various natural cycles and variations, much within and above planet earth that is not yet adequately measured or accounted for, despite the (hubristic?) confidence of many.
Brent - yes, I am also impressed by your conjecture. Given that there is as much heat stored in the first 6 feet of the oceans as the planet's entire atmosphere, it is preposterous to think that the relatively slowly circulating and up to 5 mile deep oceans won't have a significant thermal lag. And your 99 years for the Atlantic and 152 for the Pacific do give a very good match. I suspect this is just another piece of the climate system jigsaw the so called climate scientists like Trenberth pretend they know all about.
I think Douglas Keenan is exactly right and I'm sure is stating the view of the statistical world. Why haven't statisticians taken much interest in the AGW debate? Perhaps it's because there is nothing to contribute. It can't be posed as a problem for statistical analysis. Climate scientists live in a parallel universe of statistical methodology, currently shared with the Public Health industry. Pointless talk of significance, and graphs generated by model simulations won't ever bring the AGW debate to a conclusion. The only hope is to gain a better understanding of the physical processes behind the changing climate.
re: the graph of Global Surface Temperatures
As an ignorant layman one thing that always bothers me, even if the GST is roughly accurate for what it represents (only a small proportion of the earth's surface in a very small time-slice), is that from around 1905 to 1942 (give or take a couple of years) the record shows a "natural" warming of around 0.7C (if I'm correct in thinking that there could be no major AGW from CO2 emissions yet??).
So if we see "nature" producing a 0.7C increase in 4 decades or less, and also some very significant increase in the century before 1880 (judging from other factors), how can anyone really be confident in the doom and gloom assessments of CAGW from the last half of the 20th century? Some contribution of CO2 driven changes, sure, probably, maybe half or more as widely asserted, I can't know. But OBVIOUSLY there are also profound natural variation, cycles etc. which do not seem to be well understood.
I am a timid "lukewarmist": I don't have any problem thinking that the "greenhouse" effect exists to some lab-measurable extent or that CO2 changes might matter. As a citizen, human, and taxpayer I have to focus mainly upon what the policy implications are supposed to be. The multi-trillion dollar questions are really all about the claimed forcings and feedbacks beyond a 1.5C (or whatever) increase from 2xCO2. I simply don't see how people can look at the "natural" records and think that the last 60 years shows something extreme and dangerous for civilization. (I also have strong doubts that we really can do much now except adapt in the future, but that's another debate)
I'm well aware that this is no scientific argument and maybe it's all well addressed somewhere, but it seems to point to a gap between arguing there has been "some" AGW and claiming all recent warming is unprecedented, catastrophic, and hurtling us along upon an accelerating train off a cliff.
That is why I emphasize the "C" in "catastrophic" AGW, because I'm not myself in any serious doubt about some AGW.... I find the blog debates about alleged problems in the physics etc. interesting but I cannot judge them. I have no reason to doubt the high competence, dedication, and sincerity of scientists like Richard Betts and Tamsin Edwards. I simply don't see that anyone is making the convincing "policy" oriented case that goes from
(1) "AGW"
to
(2) "CAGW"
to
(a much further reach)
(3) "adaptation is not enough, we all must take drastic, expensive, and almost certainly futile actions NOW" etc.
Even if the Climate Computer Models could predict the climate accurately (they can not) they would still not be Proof. Scientific Proof needs Empirical Evidence.
Given that Earths surface temperature varies between +40 degrees C and -40 degrees C 24/7...365 days of the year depending on your location (think Pole or Equator) Anybody claiming to give you an accurate mean surface temperature is giving you a pig in a poke.
Pendleton: "First: it shows a long term warming trend - and quite a dramatic one - beginning in the early 20th century and, evidently, still underway."
Isn't that one of the key points? He thinks the "warming trend" which is continuing now began in the early 20th century. The problem is, CO2 levels in the early 20th century were only slightly higher than they had been for thousands of years before that point.
So if you accept the timeframe he uses for the "warming trend", it is very probably caused by something other than CO2.
Doug Keenan,
You wrote:
" Finding grounds for a model of global temperatures has yet to be done (the only person who has even made an creditable attempt is Demetris Koutsoyiannis). That makes it impossible to evaluate significance. In other words, it makes it impossible to know if the changes in temperature can be reasonably ascribed to random variation. Both James Delingpole and Andrew Pendleton are making a basic error."
I agree with you.
If you didn't see them when first posted, these two articles were my (quite serious) attempt to frame a null with Popperian falsifiability. I failed, and continue to fail, but you will find the articles interesting, I think.
http://rankexploits.com/musings/2011/noisy-blue-ocean-blue-suede-shoes-and-agw-attribution/
http://rankexploits.com/musings/2012/more-blue-suede-shoes-and-the-definitive-source-of-the-unit-root/
It's interesting the way the word 'significant' is used in different ways. When evaluating data, the term significant refers to the fact that a change in value is above the noise, it can be regarded as a real signal, even if that is still a small one (i.e. just detectable). This is the sense that Phil Jones referred to a temperature trend being 'significant' in his famous interview. To the general public however, significant means important, or large. Not the same thing at all, but the use of the word in press releases seems to me to be aimed at maximising the confusion between the technical meaning and the common meaning to put across a quite misleading message.
Andrew Neil (don'cha just love him) is now presiding over a debate about how many angels can dance on the head of a pin. The important thing is what's NOT there in the records. There is nothing, absolutely zilch, to support any theory that we have experienced or are likely to experience, calamatous temperature rises as described and claimed by Gore et al. Its over.