We recently ran two articles that were quite critical of aspects of the Guardian’s coverage of the stolen emails. This is a response from Dr. James Randerson, the editor of the Guardian’s environmental website.
I edit the Guardian’s environment website and was part of the editorial team that produced the 12-part investigation by veteran science journalist Fred Pearce into the hacked East Anglia climate emails. I’m very grateful to RealClimate for giving us the opportunity to respond to the recent posts on the investigation: “The Guardian Disappoints” and “Close Encounters of the Absurd Kind”.
I should say first that we hold RealClimate in very high regard. The site is part of the Guardian Environment Network, a collection of more than 20 hand-picked websites including Grist and Nature’s Climate Feedback blog with whom we have a mutual content sharing agreement. Under the arrangement, the Guardian website republishes RealClimate blogs regularly. We take seriously your criticisms and are considering them carefully. The Guardian has a commitment to accuracy and correcting factual errors.
Such is the public interest in this story that ever since the emails were released in November, there has been a strong demand for an in-depth journalistic account of what they tell us about how climate scientists operate. As RealClimate rightly pointed out, the response from much of the media has been lazy to the point of “pathology”.
No other media organisation has come close to producing such a comprehensive and carefully researched attempt to get to the bottom of the emails affair. The investigation tries to reflect the complexity and historical context of the story, and runs to some 28,000 words – of which around half appeared in the printed newspaper.
Dr. Schmidt did not mince his words though when he said that Fred’s investigation falls, “well below the normal Guardian standards of reporting”, while Dr Ben Santer wrote, “I am taking this opportunity to correct Mr. Pearce’s omissions, to reply to the key allegations, and to supply links to more detailed responses.” Both have also criticised our experimental online exercise to harness the expertise of people with a special knowledge of the emails in order to create a “peer reviewed” account of what they tell us.
More on that later, but it is wrong to suggest that this is a lazy substitute for traditional journalistic standards and that key protagonists were not invited to respond prior to publication. On the contrary, the investigation was subject to rigorous editorial checking and Fred contacted numerous individuals in the course of his research. Many (particularly those at UEA) declined to comment.
The other side of the story
The RealClimate commentary reads like a distorted fairground mirror of the Guardian investigation – one that highlights the uncomfortable bits but blurs the rest. The posts did point out that “Some of the other pieces in this series are fine” but do not reflect the large amount of analysis in the investigation of the way the emails have been misused by those with a political agenda and the extensive context we included to indicate the pressure scientists writing those emails were under from time-consuming requests for data.
In part 2 (How the ‘climategate’ scandal is bogus and based on climate sceptics’ lies), for example, we detail how the “hide the decline” email has been misused by Sarah Palin, Senator James Inhofe and others to create, apparently deliberately, the impression that climate scientists had fiddled the figures.
Almost all the media and political discussion about the hacked climate emails has been based on soundbites publicised by professional sceptics and their blogs. In many cases, these have been taken out of context and twisted to mean something they were never intended to.
In part 1 (Battle over climate data turned into war between scientists and sceptics) and in a separate piece that appeared in the newspaper (Climate scientists have long been targets for sceptics) Fred outlines the tactics and motivations of some on the “sceptic” side of the debate.
All this happened against the backdrop of a long-term assault by politically motivated, and commercially funded, climate-change deniers against the activities of many of the key scientists featuring in the emails.
Similarly in Part 7 (Victory for openness as IPCC climate scientist opens up lab doors) Fred explains how the emails give a special insight into what being on the end of that assault was like.
In the leaked emails, [Ben Santer] is seen sharing those experiences with other victims of hectoring and abuse by the more rabid climate sceptics. Others had their own horror stories, including Mike Mann over his hockey stick graph, Kevin Trenberth over his analysis of hurricanes and warming in the aftermath of Katrina, and later Jones over his escalating data wars. In each case, they argue, legitimate debates about scientific analysis and access to researchers’ data have been turned into vindictive character assassination.
And in the concluding part of the investigation (Part 12: Climate science emails cannot destroy argument that world is warming, and humans are responsible), Fred lays out unequivocally that nothing in the emails casts doubt on the case for climate change being attributable to human actions.
Is the science of climate change fatally flawed by the climategate revelations? Absolutely not. Nothing uncovered in the emails destroys the argument that humans are warming the planet. None of the 1,073 emails plus 3,587 files containing documents, raw data and computer code upsets the 200-year-old science behind the “greenhouse effect” of gases such as carbon dioxide, which traps solar heat and warms the atmosphere. Nothing changes the fact that carbon dioxide is accumulating in the atmosphere thanks to human emissions from burning carbon-based fuels such as coal and oil. Nor the calculations by physicists that for every square metre of the Earth’s surface, 1.6 watts more energy enters the atmosphere than leaves it.
And we know the world is warming as a result. Thousands of thermometers in areas remote from any conceivable local urban influences tell us that. The oceans are warming too. The great majority of the world’s glaciers are retreating, Arctic sea ice is disappearing, sea levels are rising ever faster, trees are climbing up hillsides and permafrost is melting.
These are not statistical artefacts or the result of scientists cherry-picking data.
Looking under every rock
There are few, if any newspapers in the world with a stronger commitment to action on climate change than the Guardian and our sister paper the Observer. We have a team of 6 full-time environment correspondents as well as three editors and a collection of bloggers and columnists.
It was the Guardian that orchestrated a global editorial carried by 56 newspapers in 45 countries on 7th December 2009 to call for action from world leaders at Copenhagen. [RC: Also at RealClimate]
And we have been instrumental in supporting the 10:10 climate change campaign which aims to inspire individuals, organisations and businesses to cut their carbon emissions by 10% in 2010. The UK branch of 10:10 has signed up nearly 60,000 people and over 4000 businesses and organisations.
But only by looking thoroughly under every rock can those of us pressing for action on climate change maintain with confidence that the scientific case remains sound. Fred’s investigation shows that confidence is indeed well placed, but to claim that the emails do not throw up some troubling issues looks like the inward-looking mentality that is sometimes (perhaps understandably) expressed in the emails themselves.
The two posts published so far on RealClimate come to over 8500 words and it has been suggested that a line by line response to each of the points made would not be productive. I say again that we are totally unembarrassed about correcting genuine errors, but many of the points raised at RealClimate are differences of interpretation. There were implications that the investigation omitted some key information which in fact appeared in Fred’s pieces – for example that the data on Chinese weather station locations from the Phil Jones et al 1990 Nature data were eventually released publicly and that the two studies Jones had threatened to keep out of the IPCC AR4 report were in fact cited there.
However, I would like to make four points:
- Dr Phil Jones, the head of the Climatic Research Unit at UEA has said in an interview with Nature that the handling of the records of the Chinese weather station data from his 1990 Nature paper (which Fred wrote about in part 5 of the investigation) was “not acceptable… [it’s] not best practice,” and he acknowledged that that stations “probably did move”. He added that he was considering a correction to Nature. To our knowledge, no other media organisation or blogger had used the emails to shed light on the controversy over the 1990 paper so a correction would not be on the table without the Pearce investigation.
- Dr. Schmidt states that we imply Dr Tom Wigley supported allegations of “fabrication” from climate sceptic Douglas Keenan. We do not make that assertion in the piece. Also, Dr Schmidt does not reproduce the most eye-catching quotes from a May 2009 email from Wigley to Jones in which he raises serious doubts about the quality of Jones’s scientific team and his handling of the Chinese weather station data.The hacked emails do not include a response from Jones if there was one.
- As Dr. Schmidt pointed out, we have made three small corrections to the piece “Controversy behind climate science’s ‘hockey stick’ graph” at the request of Dr Michael Mann, but none changed the main point the article was making, which was that in 1999, Mann’s hockey-stick reconstruction was the subject of intense academic debate amongst climate scientists.
- Neither of the RealClimate blogs dealt with Fred’s piece on FOI requests, but a statement from the UK’s deputy information commissioner Graham Smith has made clear that he believes that FOI legislation was not followed correctly. He wrote, “The emails which are now public reveal that [climate sceptic David] Holland’s requests under the Freedom of Information Act were not dealt with as they should have been under the legislation. Section 77 of the Freedom of Information Act makes it an offence for public authorities to act so as to prevent intentionally the disclosure of requested information.” This is a serious issue worthy of discussion and debate.
Peer-reviewed journalism
I mentioned above our attempt to create a definitive account of the emails by leveraging the expertise of people involved or with a special knowledge of the messages and the issues they discuss. This account will eventually be expanded into a book. In practice, this means us adding annotations from people to the online versions of the articles so that readers can watch a form of living peer-review in progress. Click on the yellow highlights in the pieces themselves to read the annotations.
This represents an extraordinary commitment to transparency that we believe is unique in journalism. What other news organisation would open itself to direct criticism in this way including, for example, annotations that read “this is absolutely false” and “this is really bad”? The respected Columbia Journalism Review has praised the approach. “Regardless of whether you agree with Pearce or Schmidt, the Guardian’s approach appropriately acknowledges that evidence leaves room for some degree of interpretation. It is this kind of detailed, intellectually honest (even technologically innovative) reporting that news outlets like The New York Times should be striving for,” it wrote.
In the same spirit we have showcased diverse critical opinions on the issues and our own coverage of them, including from Dr Myles Allen, Dr Vicky Pope, Dr Mike Hulme and the Guardian’s environment correspondent Dr David Adam. Again few newspapers would have reflected such diverse viewpoints.
The reaction from some to our online annotation exercise has been hostile though. On our letters pages Dr Myles Allen and Dr Ben Santer wrote last week:
Claiming to produce “the definitive” analysis now is a brazen attempt to pre-empt the inquiries’ conclusions…What is wrong with the old-fashioned approach of checking facts before publication? When the final version is published, you will no doubt make much of the fact that “everyone had a chance to comment”, implying that any statement that was not challenged must therefore be true.
Our intention is not to undermine or pre-empt the ongoing inquiries into the CRU emails. Each of those has a very specific remit and none is attempting to produce a detailed account that uses the emails to shed light on recent climate controversies. Nor is this an exercise in blackmailing scientists into fact-checking on the cheap – if it were then it would be a monumental false-economy.
In truth, this is a serious-minded attempt to make sense of a large volume of new information about a complex and highly charged issue. No other newspaper has ever offered its journalism up for very public and exacting scrutiny in this way. We sincerely invite those involved who know the issues most thoroughly to contribute.
Jack says in #394:
Correlation to 0.1 accuracy between thermometers and proxy seems rather meaningless when neither measurement is reliable even to +/- 1.0 accuracy.
You’re missing the difference between precision and accuracy. If you want to know the exact temperature you need accuracy. If you want to know changes in temperature you need precision. Accuracy is how close the measurement is to the actual thing being measured. Precision just means how many decimal places one can consistently measure.
For example, my bathroom scale may weigh to the ounce and still be off 10 pounds. That scale would be quite precise but not accurate at all. As long as I use the same scale I can tell very well whether I’m gaining or losing weight – just not how much I actually weigh.
That’s what the scientists have done with the temperature record. Say they get the diary of someone who lived in the 1800’s who recorded the temperature every day. The thermometer has long since been lost, and you don’t really know how it was sited. But if you have 100 of them scattered about and you average them appropriately you can determine that some years were hotter or colder than others, and by about how much.
Re #391:
If CO2 emissions stopped tomorrow, certainly the CO2 would remain abouve pre-industrial levels for a long time (forever in theory, ignoring all other factors), but the Earth would not get hotter, as the current CO2 forcing would be continually reduced. This would more than compensate for any increased saturation in the ocean’s ability to absorb heat; that is to say, the theoretical ‘backlog’ of warming due to the ocean not having reached thermal equilibrium would not matter. That is the basic logic of the situation as I understand it. The backlog only becomes the deciding factor in a scenario where atmospheric CO2 concentration remains constant.
Sea level would indeed continue to rise as more heat went into the oceans. But then they’ve been rising at a fairly constant rate for most of a century, and until recently it bothered nobody, even in Tuvalo.
Your link relating to the ice is too pessimistic, it relates to higher emission scenarios. Doubtless melting at current rates would continue (though depending on ice dynamics a new steady state could emerge) but that does not seem very serious; it is only a fraction of current sea level rise.
Re #397:
Of course you are correct about the SO2, and I guess a sudden stop in CO2 emissions without a concomitant drop in aerosols is even more unlikely than a sudden stop in CO2 emissions. Still, the question here is the thermal inertia related to CO2, and it seems correct to me to isolate such presumed inertia factors to determine the real meaning of each. Anyway this was all pretty much thrashed out in the related thread.
# Re 398
It should have been from “the first link in #309.”
Re: Jim’s comments on #384:
I won’t comment on the first two comments because I think I cannot put my points more clearly and people can choose to agree either with me or with Jim.
“WRT to divergence’s possible influence, do a sensitivity analysis incorporating reasonable estimates of pre-historical divergence if you’re concerned about it. This will require very careful attention to methods of course” – Jim
Cheap and cheerful method: instrumental records available for 140 years or so. Divergence has been observed for 20 years in the case of two series and 40 in the case of the third. Furthermore the proxies are to a large extent calibrated to the rest of the instrumental data. But even if we ignore this, the divergence covers almost one fifth of the time when instrumental data was available, and the cause is unknown. If I sample 6 swans and one is black, I can’t state with 95% confidence that an unknown population of swans is white!
[Response: Cheerful is good. Not even remotely close to rigorous enough unfortunately. See here for some ideas and references, esp Cook et al., 2004.–Jim]
“If you were really interested in what the science has to say about past millenial temps instead of criticizing Phil Jones” – Jim
Can’t I do both?
[Response: As long as it’s legitimate and unbiased.–Jim]
@RayLadbury (#397)
>
> There was an interesting discussion
> about the recent paper showing little
> warming if CO2 emissions stopped.
>
I assume you are making reference to this :-
https://www.realclimate.org/index.php/archives/2010/03/climate-change-commitments/
“CO2 concentrations would start to fall immediately since the ocean and terrestrial biosphere would continue to absorb more carbon than they release as long as the CO2 level in the atmosphere is higher than pre-industrial levels (approximately). And subsequent temperatures (depending slightly on the model you are using) would either be flat or slightly decreasing. With this definition then, there is no climate change commitment because of climate inertia.”
This does contrast with this view :-
http://www.ipcc.ch/publications_and_data/ar4/wg1/en/ch10s10-7-2.html
“Using a similar approach, Friedlingstein and Solomon (2005) show that even if emissions were immediately cut to zero, the system would continue to warm for several more decades before starting to cool. It is important also to note that ocean heat content and changes in the cryosphere evolve on time scales extending over centuries.”
However, the key point is that it’s not going to happen, is it ? There is actually no way that emissions can be cut to zero overnight. Well, there is, but it wouldn’t be pretty.
The best we can hope for, probably, is roughly stabilisation of the concentration of Greenhouse Gases in the Atmosphere, hopefully under 450ppmv, although it might have to be under 400ppmv to stop Greenland’s terminaly melt; although if we can find a non-poisonous Geoengineering technique (thinks…blue champagne ?) then we might be able to go below to 350ppmv which some seriously think the only safe future.
This is where the “CC” constant concentrations scenario comes in. There would definitely be future warming under CC, so we had better stabilise sooner rather than later. Here is an “idealised” later :-
http://www.ipcc.ch/publications_and_data/ar4/wg1/en/figure-10-34.html
Re #388:
I have no vendetta against Phil Jones. In fact I have some sympathy for him. Nor am I trying to elevate the relatively small issue of the WMO cover graph into a great conspiracy.
However I do believe that the truth of the issue should be clarified. There are many here who wish to insist that there was nothing incorrect about Jones’s procedures, because it aids their own vendettas against those who have criticised him and others.
[Response: If your point was that prominent graphics should be clearly labeled as to what they depict, we can definitely agree to that. The problem is people ascribe nefarious intent to these kinds of things and that’s where the real trouble comes in–Jim]
Comment by Bob — 24 March 2010 @ 7:35 PM
“I agree with #5 It is a shame you have to waste your time and TAXPAYER’S MONEY with this sort of stuff rather then focusing on your real TAXPAYER’S work !!”
—-
What???!!! Who funds the Real Climate website? Is it taxpayers? Get a grip!
I am pleased to see how upset many of you AGW believers have been by the Guardian articles. Maybe now you will have a better sense of just how frustrated many sceptical, legitimate climate scientists have felt for the last couple of decades.
P.S.
The Gulf Stream is doing fine and Arctic sea ice extent is doing very nicely thank you!!
Jimbo, it’s the end of March.
Is the Arctic sea ice extent near maximum?
It’s barely getting almost to average right now, at its peak.
Remember how low it went last winter?
Remember what they said 51 weeks ago?
For your review, this year is like last year, when they wrote:
“April 6, 2009
Arctic sea ice younger, thinner as melt season begins.
Arctic sea ice extent has begun its seasonal decline towards the September minimum. Ice extent through the winter was similar to that of recent years, but lower than the 1979 to 2000 average. More importantly, the melt season has begun with a substantial amount of thin first-year ice, which is vulnerable to summer melt.”
http://nsidc.org/arcticseaicenews/2009/040609.html
I don’t know if you’re the same “Jimbo” who’s posted the same mistake before, but in case you are looking for educational material, consider:
“…. sea ice remains much thinner than in the past, and so is more vulnerable to further decline. The data suggest that the ice reached a record low volume in 2008, and has thinned even more in 2009. Sea ice extent normally varies from year to year, much like the weather changes from day to day. But just as one warm day in October does not negate a cooling trend toward winter, a slight annual gain in sea ice extent over a record low does not negate the long-term decline.
In addition, ice extent is only one measure of sea ice. Satellite measurements from NASA show that in 2008, Arctic sea ice was thinner than 2007, and likely reached a record low volume. So, what would scientists call a recovery in sea ice? … In a recovery, scientists would also expect to see a return to an Arctic sea ice cover dominated by thicker, multiyear ice.”
http://nsidc.org/arcticseaicenews/faq.html
You can copypaste wrong information more easily than you can look up good information. Make the extra effort.
#40 Re: the P.S.
It seems that you now expect all the papers will soon be giving “fair treatment for all trends”. Place a mirror on the horizontal axis and show every trend together with its reflection. That will show the climatologists Whats What and be really fair.
http://tamino.wordpress.com/2008/08/13/sea-ice-hyperbole/
As for the Gulf Stream , the implications of that scare was misunderstood by the Guardian, (outside its main reports) but not by Realclimate.
Point taken, Gerry. I need to go back and re-read that thread on climate change commitments.
Jimbo says: 30 March 2010 at 5:26 PM
“…Arctic sea ice extent is doing very nicely thank you!!”
Well, extent once again is not looking so great compared to the historical record. Those encouraged by the last few weeks’ upswing would do well to ask how thick is all that suddenly apparent ice? By extension of what Hank points out, it’s better to reserve any judgment until September or October; the significance of extent is diminishing as volume increasingly becomes the issue.
Extent versus volume is not complicated to understand. Try driving on a frozen lake too early or too late and you’ll quickly realize with a cold sinking feeling how deceiving first appearances can be.
RE: Comment #401 Comment by David Miller
“That’s what the scientists have done with the temperature record. Say they get the diary of someone who lived in the 1800’s who recorded the temperature every day. The thermometer has long since been lost, and you don’t really know how it was sited. But if you have 100 of them scattered about and you average them appropriately you can determine that some years were hotter or colder than others, and by about how much.”
To 0.1 accuracy? That’s a very precise “about how much”! By “averaging” a wide variety of non-standard and often uncalibrated instruments, in a wide range of uncontrolled and sometimes changing sites, monitored by persons of widely varying training? “Averaging” inaccurate data does not necessarily make it more accurate! Even today’s USHCN sites, with calibrated instruments and controlled siting, can have errors far greater than 0.1.
It is the blithe certitude with which claims like this are made that allows skeptics to cast a shadow of doubt on the AGW hypothesis and the state of today’s climate science.
[Response: Your fundamental points/understanding are wrong. Please, read up on this topic, dropping the fixation with 0.1 as a magic number.–Jim]
Jack Maloney,
What happens to the error on the mean as your number of measurements, n, increases? It decreases as 1/sqrt(n), right? For this not to happen, you would have to have systematic errors that biased ALL of the measurements.
Jack, #412
You’re still missing the point. Averaging a bunch of inaccurate thermometers doesn’t make the result accurate.
Averaging the *changes* to each individual reading tells you what the average change is, and that’s what we’re all trying to measure.
It’s like my scale being off 10 pounds, and yours being off 5 pounds, and someone elses being right on the money. Averaging our weights on these scales doesn’t tell us how much we really weigh – individually or collectively. But if all three of us weigh ourselves every day and note the results it’s easy to determine whether we, as a group, are gaining weight or not.
[Response: Well said. This is the difference between precision and accuracy. If you are measuring change, you don’t need accuracy at all, you just need precision.–eric]
I downloaded the seasonal data from here, and the daily data for 2009 through the 30th of March2010 from here, and created seasonal averages to fill in 2009 and “winter” 2010(one day short). The plot from Appleworks is here, and doesn’t show what I would call a nice recovery. The top plot is the seasonal averages through 2009 plus the yearly average, and the bottom plot is average winter sea ice cover through 2010.
To-day’s report by James Randerson on outcome of Parliamentary inquiry into UEA.
http://www.guardian.co.uk/environment/2010/mar/31/climate-mails-inquiry-jones-cleared
This refers to the width of tree rings which involve a contribution to the carbon sinks.
I notice that the anti-AGW crowd don’t ask questions about any implications this might have for the future climate, especially if the divergence were to increase drastically and spread. By the way I am no expert in such matters.
It is obvious that the hockey stick skeptics would like everyone to speculate about the handle rather than looking at the blade.
Hunt Janin (366),
http://www.ncdc.noaa.gov/oa/climate/globalwarming.html
http://tamino.wordpress.com
http://BartonPaulLevenson.com/Climatology.html
With regard to the saintly James Randerson and his sanctimonious point:
“There are few, if any newspapers in the world with a stronger commitment to action on climate change than the Guardian and our sister paper the Observer. We have a team of 6 full-time environment correspondents as well as three editors and a collection of bloggers and columnists.”
It is worth pointing out why the Guardian is able to afford to employ so many full-time hacks to cover the environment (and even then pay for a freelance, Fred Pearce, to write up stuff its own journos don’t have the time to cover). The Guardian group sold off a 50% stake in a second-hand car magazine called Auto Trader, which does nothing more than publish ads for selling/buying automobiles. The sale raised a neat £200 million which Alan Rusbridger,the Guardian’s Editor, is now happily burning through. So, hardly an environment-friendly income stream, and one that The Guardian and its hacks rarely, if ever, talk about.
The newspaper is written by, and read by hippocrites — I should know because I live in Islington and I read it!
What have you done to the cat? It looks half dead!
–Mrs. Schrodinger
Jimbo: I am pleased to see how upset many of you AGW believers have been by the Guardian articles. Maybe now you will have a better sense of just how frustrated many sceptical, legitimate climate scientists have felt for the last couple of decades.
BPL: Yeah, the same way I’m sympathetic to “physicists” who want to disprove relativity or quantum mechanics.
Read my lips: AGW Denial is on a scientific par with believing aliens built the pyramids. Got it?
To crunch this thread back on topic the Guardian has today published a rather nasty piece from Pearce on the Parliamentary Report .There really seems to be some unspoken animus around.
I wonder if there may be tension around the British FOI act and this story is being used as proxy .
Comment by Hank Roberts – 30 March 2010 @ 6:11 PM
“Jimbo, it’s the end of March.
Is the Arctic sea ice extent near maximum?
It’s barely getting almost to average right now, at its peak.
Remember how low it went last winter?
Remember what they said 51 weeks ago?”
—–
RESPONSE:
Use your own lyin eyes Hank!
http://www.ijis.iarc.uaf.edu/en/home/seaice_extent.htm
http://nsidc.org/data/seaice_index/images/daily_images/N_stddev_timeseries.png
http://arctic-roos.org/observations/satellite-data/sea-ice/ice-area-and-extent-in-arctic
http://arctic.atmos.uiuc.edu/cryosphere/IMAGES/seaice.recent.arctic.png
http://ocean.dmi.dk/arctic/icecover.uk.php
Jack Maloney:
You have strayed from merely having a different interpretation to making errors of basic fact. Please look at the actual precision used by, for example, the HadCET record (one of the few temperature records sparse enough and old enough for this to be a real issue). You will see early values are to 1 degree or 0.5 degree precision, not 0.1.
By fixating on a minor detail, you have glossed right over all the real difficulties and challenges in working out the probable error range for early temperature records, which include uncertainty about the properties of the thermometers used and gaps in the records.
As a result, if you look at the error bars for HadCET (or any of the global products such as GISS) you will see bigger error bars earlier on. Exactly as we expect.
If you want more detail, you are going to have to start reading the actual papers.
Jimbo, your eyes are misleading you.
Use more of your brain than the pattern recognition part.
Our brains are pre-loaded to detect patterns even when they don’t exist.
Do you believe in evolution? Consider where you came from and why. Pattern detection kept our ancestors alive by detecting threats like leopards in the dappled shade. Extremely hypersensitive pattern detection was vital, because there was little cost for frequent false alarms but a great risk of becoming a snack by missing detecting a real leopard.
Our brains find patterns but can also think through what we’re seeing. You can do this.
NSIDC — the gray band marks 2 standard deviations. Know what that means?
Think of anything that matters to you; if the measure is off by even one standard dev. do you notice? Do you know _how_far_off_ 2 st. dev. is???
Take Statistics 101, and your view of the world changes. Seriously, man.
ijis — most recent eight years.
arctic-roos — that’s the most recent four years
cryosphere — that’s the most recent two years
ocean.dmi.dk — that’s the most recent five years
If you’re only comfortable looking at pictures, look at the trend pictures.
You know how to find them. Read the links and quotes I already gave you.
32 years: http://arctic.atmos.uiuc.edu/cryosphere/IMAGES/current.anom.jpg
Trend: http://nsidc.org/images/arcticseaicenews/20100303_Figure3.png
Think: http://moregrumbinescience.blogspot.com/2010/03/wuwt-trumpets-result-supporting-climate.html
Here is one method I used to see how much data one can wring out of long term results, where the data accuracy is not that great. In this case I started with the longest record (Central England 1659-2008), found the anomaly from the 1961-1990 period,adjusted the raw data and compared it to the Hadcet data from 1850-2008. I then filtered both anomalies with a Fourier Convolution filters, with a cut-off of 0.025 cycles/year or 40 year period.
I chose this method in that it includes the end points, better then the MOV and recursive filters. Also it reflects many of the periodic or almost periodic physical processes. This resulted in the following graph:
http://www.imagenerd.com/uploads/lt-temp-1650-2008-1-Rxrdy.gif
Here one sees the interesting periodic motion in the 50 year period, in the Central English data, and somewhat similar motions in the Hadcet data. With the Hadcet data, the additional of stations along the way from different parts of the world could modify the periods. Also thre is a apparent peaking of both sets of data about and after 2000.
The next step was to include the four longest running data sets before 1750 (England, DeBilt (1706-2008), Uppsala (1722-2008), Berlin (1701-2008). The latter two from Rimfrost –
http://www.rimfrost.no/
Each data set was anomalized to it’s 1961-1990 average . This anomalized group was then averaged on a yearly basis, ignoring missing data. This resulted in the following graph:
http://www.imagenerd.com/uploads/lt-temp-1750-2008-4-EyvXd.gif
Here we see again the ~50 year cycle, and a increase in the most recent 50 years from the single Central England data set. Urban effects of Holland, Uppsala, Berlin (?).
The last group used temperatures whose records started before 1800. These included the previous plus Geneve (1753-2008), Praha (1775-2005),Stockholm (1755-2005), Budapest (1780-2008), up to Warsaw (1792-2008). At total of 14 in all, and all from weatern or central Europe.
http://www.imagenerd.com/uploads/lt-temp-1800-2008-14-9ZSv8.gif
Again the ~50 year cycle keeps showing up, and this this case the last 25 years is down from the previous group, closer to the Hadcet. So while the temperature series are from different location, and unspecified accuracy, some of the patters seem to hold pretty constant.
JBob,
I see oscillation, but no evidence of periodicity. Have you run tests for statistical significance?
Re #404:
But note that 2004 > 1999. Jones didn’t know much apart from Briffa’s report that before the divergence started, agreement was good back to 1860, i.e. 140 years. That puts the issue much closer to my swan analogy. I do not accept that a 95% confidence interval of 0.1 degC could have been justified in the circumstances.
Even now, given the nature of proxies – for which the word ‘independence’ must often seem like the concept of a crazed idealist – I doubt that it is possible to say with anything approaching certainty that the current ‘divergence’ phenomenon has never happened before during the last millenium.
Re #405:
The “system” presumably includes the oceans, which will still be absorbing heat under all scenarios. I don’t think that contradicts the notion that surface temperatures would be expected to fall immediately.
True enough, these are all idealisations – but they can help make the physics clear.
“I doubt that it is possible to say with anything approaching certainty that the current ‘divergence’ phenomenon has never happened before during the last millenium.”
Got any reason for that doubt?
After all, why would some tree ring proxies be giving the wrong temperature since 1960 and be divergent from other proxies now, whereas in this hypothetical past incident where the tree ring proxies were divergent, every other proxy was also divergent IN THE SAME DIRECTION and AT THE SAME TIME.
Or just hoping, were you?
“JBob, I see oscillation, but no evidence of periodicity. Have you run tests for statistical significance?”
Comment by Ray Ladbury — 31 March 2010 @ 10:20 AM
Already done, and in the peer reviewed literature.
“Peaks at 23.5-yr,36-yr, and 65-yr dominate only in one or the other of the shorter intervals, thus showing the features are not periodic.”
“Figure 2 shows the temporal variation of the two dominant spectral components, 23.5 yr …and 102 yr … These components are not strictly periodic; they are strongly amplitude and phase modulated.” [maths that I can’t retype omitted – BD]
Time scales and trends in the Central England Temperature data (1659-1990): A Wavelet analysis, Baliunas, Frick, Sokoloff, and Soon, Geophys. Res. Lett., 24, 1351–1354
http://cfa-www.harvard.edu/~wsoon/myownPapers-d/BFSS97GRLwaveletonCET.pdf
#427 Ray, periodicity would depend on the tolerances one puts on it. If you want to get fussy about it, sunrise and sunset may or may not be periodic, depending on the tolerance, or accuracy one wants to use. Sunspot activity is considered periodic or almost periodic, same as a normal heart beat. One does not need a statistical analysis to recognize that. A simple ruler will do.
JBob,
If you don’t know a driver for your periodic phenomenon, you’d better get fussy, or you’ll delude yourself.
Re #429:
Perhaps you should take a look at the proxy data. You seem to have a somewhat naive idea of how consistent the different proxies actually are!
Look for example at Figure 3 of Mann 2008. The different proxies wibble and wobble considerably, often agreeing with each other, but often disagreeing, too.
So my reason for the doubt is that it would take some pretty heavyweight argument and statistics to get anywhere near certainty on this issue. There is certainly some evidence that this particular form of divergence, at least, is novel – but does it reach near-certainty? Can you tell me?
#431 Brian
Interesting ref.. I haven’t used the wavelet that much to comment one way or the other. However in this Fourier case, frequencies above 0.025 cycles/year ( 40 year periods) were nulled out. So how could components with periods below 40 years contribute to the filtered result? In this case, I’m looking for longer period patterns to correlate between the three different time series.
Here is a comparison of the different filter types, using the raw Ave14 data. Three filter methods are used, 40 yr MOV, 40 yr Fourier, and a two direction 2 pole Chebushev with 0.025 cycle cutoff. The Chebushev was run forward and then backward so as to compensate for the phase delay. It, doesn’t work great around the end points, like the MOV, but works well in the center. This is known as the “filtfilt” in the MATLAB signal conditioning Toolbox.
http://www.imagenerd.com/uploads/ave14-smoothed-rev_cheb-j0m9Y.gif
Note that in the center the “filtfilt” is very close to the Fourier, and is smoother then the MOV.
Again, the presence of those longer period oscillations could indicate that the method of merging these long term European temperature stations may not be far off. Even though we don’t know how accurate the stations were, they do seem to follow long term trends. Which was my basic point. These old stations may not be the best, but it’s all we have been delt.
#433 Ray, could be AMO effects or even the PDO. Remember we are dealing with a very non-linear system here. Non-linear systems have a way of changing response frequencies from the forcing frequencies.
“James Lovelock: Humans are too stupid to prevent climate change
In his first in-depth interview since the theft of UEA emails, the scientist blames inertia and democracy for lack of action” article continues at:
http://www.guardian.co.uk/science/2010/mar/29/james-lovelock-climate-change
The Guardian gets an A for the day for publishing an interview of James Lovelock.
Since I find it difficult to disagree with James Lovelock, . . .
Almost all humans are too stupid to prevent climate change. The remainder are scientists. Only Scientists should move to Mars, just in case.
I just started reading “Denialism” by Michael Specter. The bad news is that Michael Specter is NOT a psychologist or psychiatrist. So far, there is one thing worth mentioning: Most people confuse what corporations do with what scientists and engineers do. It is very important that we point out that it is the CEO, the managers and the stockholders, NOT THE SCIENTISTS AND ENGINEERS, who make those fatal [for you] decisions. Cases: Vioxx, Pinto gas tank, Corvair suspension system, Thalidomide, space shuttle disaster, etc. etc. etc. We had better say this loudly and often. In order for things to go better for the people in general, the scientists and engineers have to be in charge. Scientists and engineers in charge would be upside down from the present situation.
Scientists and engineers ARE EMPLOYEES. Managers ARE DECISION MAKERS. Managers over-rule the better decisions made by the scientists and engineers. That is where disasters come from. Pinto gas tanks and Corvair suspension systems were designed with the extra parts. Management removed those parts.
Engineers who design cars and other things to last too long are known as “hard core unemployable,” or government employees in my case.
Fred Pearce was let loose again. He can’t help himself.
Has he got a personal grudge against Prof Jones or is he just trying to save his upcoming book? This is one very nasty article. I won’t be reading anything he writes again. He’s clearly not quite rational on this topic.
http://www.guardian.co.uk/environment/cif-green/2010/mar/31/hacked-climate-email-inquiry-phil-jones
Jimbo wrote: “Maybe now you will have a better sense of just how frustrated many sceptical, legitimate climate scientists have felt for the last couple of decades.”
Yes, it’s true — skeptical, legitimate climate scientists like the ones who run this site have been very frustrated by the deliberately deceitful pseudoscience, outright lies — and most recently vicious personal attacks against them — that have been cranked out for the last couple of decades by fossil fuel industry-funded frauds and cranks and given unwarranted legitimacy by the mass media, and regurgitated ad nauseum on blogs everywhere by Ditto-Heads who unquestioningly believe whatever drivel is spoon-fed to them by the phony “conservative” media, and call themselves “skeptics” for doing so.
“Perhaps you should take a look at the proxy data. You seem to have a somewhat naive idea of how consistent the different proxies actually are!”
Perhaps you didn’t check out the work done to correlate the different proxies and how a SUBSET of tree ring proxies IN LIMITED AREAS deviate from ALL OTHER PROXIES AFTER 1960-ish.
Nothing there to say that the deviations happened before.
You made that claim, prove it.
“Remember we are dealing with a very non-linear system here. Non-linear systems have a way of changing response frequencies from the forcing frequencies.” J. Bob — 31 March 2010 @ 10:28 PM
I agree, but I would go even further. The non-linearities are emergent properties of the complex interconnectivity of the many parts of the earth climate system – The presence and amplitude of El Nino changes the rainfall in the southeastern US, which affects evapotranspiration and cloudiness over the warm parts of the Atlantic, which affects the amount of sensible and latent heat which goes into the atmosphere, which the prevailing winds carry to Great Britain, and so on through a chain of events which eventually influence the barometric pressure differences between the Eastern and Western Pacific which drive the ENSO. The strength, duration, and timing with respect to seasonal shifts in the ITCZ affect the rate and magnitude of propagation of the events, and the feedbacks at various scales. How many Atlantic hurricanes, what their tracks, intensities, and timing seasonally and with respect to the phase of el Nino eventually effect the rate and magnitude of the wobble to the next La Nina and following El Nino. Because the system has large (and non-linear) dissipation, we don’t see 40 degree ENSO SST swings, or 300 meter tall wind driven waves, or 500 km/hr sustained hurricane winds. Even though the strongest determinants of things like the ENSO periods filter the responses to a range of frequencies, the variable sum of the other influences means that the duration of an El Nino event can be from about 6 to about 18 months. We learn about period, frequency, time constants, phase, amplitude, from simple systems – pendulum; a mass, a spring, and a dashpot; an inductor, a capacitor, and a resistor – and we want to use those words and the tools that go along with them (Fourier transforms, wavelet analysis) even though we know they are inadequate, limited, and inexact when used to describe climate events.
Even deceptively simple systems can have devilishly complex behavior – here’s a simple demonstration that anyone can set up in their living room:
Take a roll of electrical tape, stick the free end to a doorway or light fixture, an unroll about a meter – viola, a pendulum with simple harmonic motion.
Take a spool of thread, unroll about 3/4 meter, refix it so it wont unroll further, and tie the free end to the spool of tape; two simple pendulums, coupled, periods not harmonically related unless you are very lucky. Give the tape roll a poke to set the system in motion, and watch the path of the thread spool – sudden complexity, especially if you put the tape spool in an elliptical instead of straight path.
Stick a light pie pan to the tape just above the spool, an position a fan so that it’s blowing on the pan and the tape. Now the motion is chaotic, but bounded. It’s not fractal, because the filtering effect of the masses limits the high frequency motions. even though the tape flutters at high frequencies because its aerodynamic properties (flat, thin, flexible). It doesn’t transfer much energy to the masses, because it’s stretchy; and flat, thin and stretchy* aren’t variables in the equations of harmonic motion. As the pie plate twists in the breeze, it captures or dissipates variable amounts of energy, and changes the orientation and dynamic tension on the tape, which changes the amplitude and frequency of the tape flutter, and causes the thread spool to swing in different directions and trajectories. In the setup I have in my living room, the path of the thread spool covers a range of about 1/4 meter, plus it bounces up and down as the tape flutters, about 1 cm, about 1 Hz. *[ “stretchy” does enter into the mass+spring bouncy harmonic motion, but electrical tape stretchiness is a very nonlinear and dissipative spring – I did warn you about deceptively simple yet devilishly complex &;>)]. I can post a picture if anybody’s that interested.
If we spend too much time trying to find explanations for what might be 4 or 5 cycles of what might be a periodic climate forcing in the time over which we have data (CET), which we know is a measure of one state variable (temperature) at one location (Central England) of a complicated non linear climate system, we risk spending our time chasing hares hidden in the weeds instead of shooting the geese flying overhead.
Re my #353.
When I tried to do a brief but fair summary of Fred P’s previous errors,that had come my way, I only mentioned his earliest reports of the hockey stick controversy. But I completely forgot the Latif affair:
http://thingsbreak.wordpress.com/2009/09/11/of-moles-and-whacking-mojib-latif-predicted-two-decades-of-cooling/
which I think was I think a rather serious lapse. This did not involve the technicalities of the hockey stick. Since Fred P’s method depends so much on the reporting of conversations, and since Latif was perfectly willing to communicate why did Fred P fail to check ? e.g by email or telephone? Andrew Gilligan was sacked by the BBC (perhaps unfairly) because of that kind of charge. It was the about the WMD issue.
#440 Brian, remember, the basic point was old temperature records, lack of accuracy data, and how it might be possible to wring some more information out of that data. The method I used was a way to compare long term data sets of different record lengths, and how adding stations distorts the average. However with filtering, some periodic or almost periodic oscillations still seem to show up, that might indicate there is a reasonable amount of correlation with the stations of central and western Europe, 200 to 350 years ago. That is, expand the individual station data from that era to maybe get a broader picture of what was going on in that era.
Yes I have seen some very unexpected non-linear responses. It is exceptionally difficult working through the basic partial differential equations, to the practical application of how many of these systems are linked together. On the compound pendulum, you can make it much more interesting if you replace the thread with “soft” springs.
A good systems person lives in two worlds, the micro/macro and the overall view. Not easy always, but interesting.
Happy Easter.
Re ‘Completely Fed up’ #439:
So “LIMITED” that they skewed the entire data series? It would seem that these “limited areas” comprised most of the trees under consideration. But yes, they found that trees in some regions tracked instrumentally measured temperatures. And they even found that the two sets of tree proxies matched each other “reasonably” well further back than 1860.
Some evidence, then, that this may have been a one-off recent event.
I didn’t make that claim. What I said was that it would be difficult to prove that such a divergence did not happen before. Can you prove that?
And this thread’s Paul R. Ehrlich Apocalyptic Award goes to:
“YOU are going to die a horrible death SOON unless we stop burning coal NOW.”
Sou says: 1 April 2010 at 12:24 PM
Fred Pearce was let loose again. He can’t help himself.
Stick your neck out too far and it tends to get stuck. Climbing down is hard to do, virtually impossible for most of us.
George Monbiot put in an appearance at Climate Progress just a couple of days ago, playing a mournful violin sonata about how Phil Jones had disrespected the work of selfless journalists by losing his patience with toddlers bearing reams of FOI requests. Monbiot’s got his neck stuck on “Out”, too, and can’t pull it back.
378 Ike Solem: “Blanket promotions for nuclear ignore the issues – fuel supply, waste disposal, construction costs, and need for large amounts of cooling water.”
We have 5000 years worth of fuel if we recycle and breed fuel.
There is no such thing as nuclear waste. It is perfectly good fuel that is being wasted for political reasons.
Construction costs are almost total cost. Averaged over a 40 year life, you get a 30% savings in price compared to coal.
Cooling water is NOT needed. Air cooling is a perfectly reasonable substitute. Water cooling is used when it is easy.
Quit shilling for the coal industry. Did you realize that you were repeating coal industry propaganda? Did you know that coal fired power plants put 100 to 400 TIMES as much radiation into the environment as nuclear power plants are allowed to? Coal contains URANIUM, ARSENIC, THORIUM, LEAD, Antimony, Cobalt, Nickel, Copper, Selenium, Barium, Fluorine, Silver, Beryllium, Iron, Sulfur, Boron, Titanium, Cadmium, Magnesium, Thorium, Calcium, Manganese, Vanadium, Chlorine, Aluminum, Chromium, Molybdenum and Zinc. There is so much of these elements in coal that cinders and coal smoke are actually valuable ores. We should be able to get all the uranium and thorium we need to fuel nuclear power plants for centuries by using cinders and smoke as ore.
444 J: And how are YOU going to get their attention?
379 Geoff Wexler: “Lovelock’s earlier wild alarmism”:
Lovelock’s ideas may be unpopular, but he is too smart to dismiss. If Lovelock’s intuition says there is a non-linearity, go look for it, just to be safe.
“What I said was that it would be difficult to prove that such a divergence did not happen before. Can you prove that?”
Yes. Go to the literature that explains the divergence problem.
Bingo.
Now prove that your statement about how it’s not possible to proclaim accuracy from proxy records.
“So “LIMITED” that they skewed the entire data series?”
They didn’t.
They left the data out.
This does not proclaim that the data would be skewed by their inclusion any more than the surfacestations assertion that bad stations were the cause of the warming trend in the US temperature record.
When those “bad” stations were taken out, the dataset was not skewed there either.