The Blackboard

“Hone your communications skills; we all need to communicate more effectively.”

One way to achieve effective communication is to communicate honestly.

Andrew

Well, that is another move in the right direction. However, as long as Judith keeps using the term “skeptic” to refer to all those that disagree with what she obviously still believes is a “consensus,” then she is STILL effectively circling the wagons.

Also, there is this statement:

“There is no question that there is a political noise machine in existence that feeds on research and statements from climate change skeptics.”

That, again, reveals a bias and a “holier than thou attitude” that gets in the way of real communication. That is an insinuation of corruption and politics only on the “skeptic” side. Drop the damn “big oil crap! There are politics on both sides!!!!

Please fix links. e.g.
Currie’s presentation on the integrity of climate research can be found here
http://www.pacinst.org/topics/....._Curry.pdf

When I hear they label me a Skeptic I say thank you for including me in such august company and Copernicus, Galileo and Einstein. All of them were Skeptics of long held Truths. Copernicus is one of the best examples of Skepticism. The Earth Centric solar sytem Theory even had a “working” model that could produce relatively accurate predictions but Copernicus overturned that becasue he was a skeptic. Aristotle had heavier objects falling faster then lighter ones, but Galileo was skeptical of that. Einstein took on what was probably the most sancrosanct scientific truth in the Laws of Gravity, but showed that even that is not immune to improvement.

So thank you Dr. Curry I’ll take that compliment and maybe someday you will join us Copernicus, Galileo, Einstein and I.

The only real response to skeptics is to be honest in all aspects of the work, to have work that is reproducible, and to let the facts speak for themselves. AGW promoters have done none of this.

Lucia,
Reading the report on hurricanes shows that while Dr. Curry is demanding honesty, her work as represented in her hurricane study is, to say the least, fallacious and misleading.

But I am surprised that Romm has let her publish at his site. I am disappointed that she would want to.
I wonder if Dr. Curry would support a demand for all the organs of the AGW community to be forced to be open?

I think one thing that could be addressed is how come that there is alienation, not just of those on the margins, but those just outside the keep.

It seems you are either on the bus or off the bus.

Speaking personally, one thing that I find disturbing is the lack of highlighting of buried bodies by people who are not skeptical. If those that are convinced of the strength of the science took firm and public ownership of things that do not fit the models pre-emptively and not just when they have to, I for one would not be so skeptical.

Do it post fact and it looks like an admission, like the fact that the models can not predict temperature trends except for periods longer than a decde or more.

The artic sea ice collapse was nearly a good moment, as it was unpredicted, and that was admitted, so if extents increase then only the people that jumped on the tipping-point angle will look stupid.

This might be a good moment to clear the decks of a lot of dead bodies, like perhaps that the post 1950s warming in Europe, seems greater than that predicted by the models, which could mean that the models both fail to predict accurately and underestimate a regional warming pattern, and that cuts both ways. If it was spun for all it is worth as an underestimation (things are worse than we thought) that would remove wiggle room, if temperatures started to fall in Europe. If it was just presented as something that cannot be adequately explained or modelled that would leave the presentation of the science less likley to be a hostage to fortune.

Europe is, on the whole, very receptive to the concept of AGW, simply because much of the region has become perceptibly hotter in the last 50 years. You do not need a thermometer to judge this. It might be a smart move to make people aware that some of that might be non AGW in origin, just in case we are in for decdes of temperature stasis or decline.

Alex

Her error is palpable: she describes the two groups as climate scientists and sceptics. She never acknowledges that many sceptics are indeed climate scientists.

I was surprised Romm let her post there, but I think Judy was wise to do so. Some of these things need to appear on the activists blogs.

I would like to congratulate Dr. Curry for her small baby step in starting on the long journey toward open-minded science and an honest understanding of the abusive atmosphere to which she has been a party. We can only hope that the scales will eventually fall from her eyes and allow her to see the political landscape as it really is. Slandering her opponents while calling for honesty in her field only qualifies as a step forward when the field is one as diseased as climate science.

Well 2 cheers for Dr Curry. Excellent sentiments, and no reason to doubt their sincerity.

I note that when challenged at CA over whether she had published all her data, metadata and code, she said that she had not, but would seek to do so fully in future. She explained that her website had not been updated with the requisite archive because of budget cuts, or lack of resources, or whatever.

I find this inconsistent with her call to arms. However, I am sure she will be most generously judged on her future performance in this respect.

That’s ok as far as it goes, but she’s still refusing to acknowledge that there’s a “noise machine” on the alarmist side that’s at least as noisy as the “noise machine” on the skeptical side, and I maintain that what got them into trouble in the first place was listening to the alarmist polemical “noise machine” rather than taking a disinterested scientific approach. The “circling the wagons” explanation only makes sense if you can explain why they were so married to a certain set of conclusions in the first place, that they saw criticism as an attack on their sense of identity.

I’m not sure exactly what’s the cause of this, but someone on another blog mentioned the Kübler-Ross model. I get a strong sense that she and many others are stuck in stage 3 (bargaining), and doesn’t want to got to stage 5 (acceptance) without getting some concession out of the deal.

FWIW, I’m also seeing evidence, primarily in the non-scientific world, of many people stuck in stages 1 and 2 (denial and anger, respectively). And of course, the ones in stage 4 (depression) aren’t talking.

Unfortunately this letter doesn’t deserve applause. The carbon cultists have been trying for the last two years to improve their propaganda techniques. Several editorials in New Scientist have pushed for this, sounding much like Curry. There is no scientific humility at all, no indication that she is dealing with theories that must withstand falsification. Only a fear that the public is beginning to doubt the Holy Axioms of Received Truth.

Unfortunately this letter doesn’t deserve applause. The carbon cultists have been trying for the last two years to improve their propaganda techniques. Several editorials in New Scientist have pushed for this, sounding much like Curry. There is no scientific humility at all, no indication that she is dealing with theories that must withstand falsification. Only a fear that the public is beginning to doubt the Holy Axioms of Received Truth.

So I wonder if the good Dr. still believes this testimony she made in 2006 to the US Congress.

“In addition to my own personal research experiences in the Arctic, a series of national and international assessments undertaken by the Intergovernmental Panel on Climate Change (IPCC),the U.S. National Academies, and the U.S. Climate Change Science have made it very difficult to maintain a credible position of scientific skepticism regarding the influence of humans on global warming. The past year has seen striking resolutions to two controversies involving the data record of climate change that support anthropogenic greenhouse warming: the synthesis report on the surface temperature reconstructions over the past two millennia the (NAS, 2006) and the synthesis and assessment report on temperature trends in the lower atmosphere (CCSP, 2006).
Further, the draft IPCC 4th Assessment Report presents climate model simulations that are far more sophisticated and accurate than were available in prior assessments, substantially increasing the credibility of such simulations and the associated projections. The cautious conclusions of the large body of scientists contributing to these assessment reports by evaluating a large body of published research are extremely important in providing a balanced overview of the state of
knowledge in the scientific research community. Based upon these assessments, our understanding of how the climate system works, while incomplete, is more than sufficiently robust to afford a basis for rational action.”

That is a lot of faith based on IPCC reports we now know are based on corrupted science and models that cook the data to produce desired results.

Perhaps she should send a new letter to Congress.

One of the issues that bugs me is this constant claim that anyone who disagrees with ANYTHING about the religion is a shill of “big oil” or some other industrial group. This comes from the unbelievably naieve attitude on the part of the Left that industry is “evil” and government is “good.” I think some of these dreamers are now waking up to the fact that their investments and RETIREMENT funding (yes, school teachers, yours too) is backed by–ohmygod–industry—and OHMYGOD, even “big oil.” Everyone, including the liberal environmentalists had better PRAY that industry (which is the ONLY source of wealth) becomes healthy again.

To support (one should be accurate here!) –

“In terms of my own publications, i am including as supplementary material any new data set that we create that is used in our analyses. My web site is badly out of date (victim of budget cuts), but our most recent paper by Belanger et al. on hurricane induced tornadoes does include the data as supplemental material.”

Along with others above, I am less than impressed. Her focus is still on “dealing with the skeptics” rather than simply telling the truth. You don’t have to strategize on how to counter misinformation if you have the truth on your side. When scientists go beyond telling the truth, no ends-justify-the-means argument will hold. What’s needed is not to start being nice. What’s needed is independent auditors from outside climate science to start going over the papers of the Team one by one, line by line, in public. That’s what the team really fears – to loose their gatekeeper status and to be judged by independent outsiders.

It is in the New York Times with an introduction by Andy Revkin and a reply by Mike Hulme.

I would applaud Ms. Curry not for the letter which is reasonable enough but still not there, but for her comment under her post on Romm’s blog:

http://climateprogress.org/200.....ent-213630

”
Ian, during autumn 2005 i and my colleagues did suffer slanderous and libelous comments about our work on the hurricane and global warming topic. So I don’t think naive is an appropriate characterization. I reserve the word “deniers” for people that are explicitly associated with advocacy groups that are politicizing this issue (CEI comes immediately to mind). Skeptics that are doing analysis and publishing their research (in journals or the blogosphere) deserve to be called skeptics, even if their analyses and research provides fuel for the deniers. It is the failure of the many in the climate community to draw this distinction between deniers and skeptics that has resulted in this problem. In my classification system, Steve McIntyre is a bona fide skeptic, not a denier. Researchers who explicitly affiliate themselves with advocacy groups can be classified as deniers, but if they put forward a serious argument, we need to look at it.
“

Personally I’d be ashamed of having sided with people who behaved as they did – revsing, barricading, stifling of communication, bullying, you name it.
After reading Willis Eschenbach’s FOI Request at CA, I could only respond as follows (as posted at CA):
“If there was ever a case of violating the FOIA, this is it.
I’m particularly concerned by the overall psychology that permeated throughout the scientists surrounding and including Mann and Jones. I see a lot of delusions and paranoia.
This is not uncommon for people who reach dictatorial stages. History has always shown this – I’ll spare the obvious examples. Clearly we see a science here that indeed became extremely insecure, and later defensive, secretive and dictatorial.”

Indeed you can’t call that science anymore, can you.
Sorry, but the behaviour at CRU is very troubling and is a threat to the field of ciimate science.
Sceptics are the best scientists – as mentioned by one reader here.

I made the mistake of investing some time in writing a comment in reply to Judith Curry’s gest blog. I attempted to lay out a reasoned skeptical position (a ‘lukewarmer position’ if you will). It was snipped completely.

Don’t waste your time trying to post with Joe, he is never going to allow even the most reasonable of comments.

Has anybody noticed google has removed any search suggestions relating to the AGW scandal? I actually noticed this today because I now have to type in my entire search string. I found someone who actually posted on this.

http://www.smalldeadanimals.co.....12721.html

This has me very worried given googles massive control over information flow.

I guess I’m not sure what the point of all of this is.

If you look at the first few pages of the Copenhagen Treaty, there is a paragraph which makes the signer agree to all of the conclusions of the UN IPCC. The conclusion is that humans are responsible for changing the climate of course. Once a country has agreed to this paragraph, there will be no need for further research into climate change. In other words, it is a paragraph to stop the discussion, to stop the research and debate, it’s a shut-up paragraph. So the funding for climate science will be cut off, there will be no need for it. Besides if the funding continues, climate science might inadvertantly discover the truth. So anyway once the funding is gone, climate science will return to the obscurity from whence it came. Copenahgen is the end of climate science.

Fred from Canuckistan:

The cautious conclusions of the large body of scientists contributing to these assessment reports by evaluating a large body of published research are extremely important in providing a balanced overview of the state of
knowledge in the scientific research community. Based upon these assessments, our understanding of how the climate system works, while incomplete, is more than sufficiently robust to afford a basis for rational action.

And with that, she stepped over the line. That’s the problem a lot of people hare having with this. Once you start advocating policy, you’ve forfeited you objectivity.

Scientists and policy makers both can stand to take a page from the lawyers. In court, you find facts, you render a verdict, and only when all of that is done, do you deal with remedies. It was set up that way for a reason. What I see happening here is what happens when you don’t have that firewall between facts and action – the perceived need for action starts to drive the finding of fact.

This is exactly what started this train wreck (with the active support of political activist organizations, btw), And this needs to be totally off limits.

This is bigger than climatology, btw. When the history books are written, this event will go down as a turning point in scientifically advised policy in other areas as well (for example bioethics). This is THE huge unanswered question of the current epoch, and it desperately needs a good answer, NOW. It’s interesting to read in the bioethics literature how they are struggling with similar issues; i.e. is it possible to be objective if you’re in the employ of a pharmaceutical company. Oddly, they too seem to not see a similar conflict of interest when academic researchers are beholden to government grants.

Judy, Thank you for your posting. As a retired MS-ChE and scientist with over 40 yrs of model design experience my ultimate aim was to make sure my models, projections and finding were sufficiently robust to withstand scrutiny from all quarters, which I knew would surely follow. As you might guess, I was very careful to make sure that I repeatedly checked and rechecked my data, procedures and calculations. If climate science (and science in general) is not open and transparent such that others can challenge it, the entire field will lose credibility, as is currently happening. As such, the rest of the scientific field, of which I am still an active member, wish to have all pertinent data, assumptions and basic calculation procedures/methodology posted & available such that I can verify the veracity of calculations and results.

Lucia. Please block off the student’s one paragraph letter to distinguish it from the rest of Curry’s opinion/blog, as at Romm’s blog.

Despite “Climategate” showing 10,700,000 results, there’s no quicksearch for that either. Type in the whole word and all you get is “Climate Guatemala”! Very odd indeed!

Calvin Ball (Comment#25060)

And with that, she stepped over the line. That’s the problem a lot of people are having with this. Once you start advocating policy, you’ve forfeited you objectivity.

The problem here is reading correctly. That para does not advocate policy. It says the science is good enough to be a basis for forming policy.

Nick Stokes,

“the science is good enough to be a basis for forming policy”

This is clearly an advocation for policy. It has the words “for forming policy” in it. That’s what the science is good enough for. If it wasn’t an advocation for something, it would just say “the science is good.”

Andrew

Andrew_KY (Comment#25067)
What policy does that para advocate?

Nick Stokes,

It doesn’t specify any details of any particular policy. But it does specify policy. You know, it leaves room for the whole “we have to do something” stupidity policy.

Andrew

Nick Stokes:

“For forming policy” implies that she’s already concluded that some form of policy is needed. I concede that this is ambiguous, because if a scientist is asked the narrow question regarding what is the level of confidence of a conclusion, that’s right on the cusp of advocating policy, but perhaps not. This raises another question though, which is why was she so certain, in light of recent events? Was she deceived, or what was the alternative?

As I mentioned later, this is a messy and emerging question that’s going to have to be answered in areas other than climate. And where exactly the proper lines are, is going to have to be debated. I would argue that it would be one thing for a scientist to advise that more research is needed in certain specified areas, and another thing entirely to advocate direct action. I think when a proper policy is finally formulated, it probably should make these kinds of distinctions; it is the scientists who know where the weaknesses in the knowledge are, and where the most effective allocations of research funds are going to be.

Jumping directly to action policy proposals leaves out a step, as well as it leaves out certain expertise (for example economics and engineering) that should be brought into play when policy is finally discussed.

In fact, aside from the general dishonesty and sense of betrayal from the public, the real wrong turn that Jones and Mann all the others made in this episode is skipping a step. Instead of insisting on the resources needed to answer the questions so that what Dr. Curry said about the certainty would have been true, they simply fudged.

A really interesting question would be why? What pressures were brought to bear that made them decide to fudge rather than recommend more research?

I have always respected Dr. Curry’s professional integrity. It has been apparent since her last foray into the hurricane debate a few years back that she has tried hard to maintain a very center of the road approach in the reporting of her research. I must agree however, that her continued “us vs the skeptics” mentality is curious, and needs some clarification. What is a skeptic? If by skeptic she means a flat-earth believing ignorant political hack, who is not at all serious about truly understanding the science, but rather engages in cheap political propaganda on the right, then I find myself agreeing more with Gavin Schmidt. Research scientists should not be forced to waste their time responding to countless “gotcha attempts”, so that some political operative might score quick media points. However, I fear that the above definition of skeptic is not at all what she has in mind. It is more likely that what she refers to in derogatory terms as a “skeptic”, may in fact be any informed earth scientist or engineer who sees strong a priori and empirical reason to doubt the robustness of the hypothesis that holds that the human addition of CO2 is the dominant driver of climate on all scales, global to regional.

These other types of “skeptics” of whom I fear Dr. Curry may refer, are often trained scientists or engineers who fully understand that CO2 is a first-order climate forcing which warms the planet (all else remaining equal), but they also realize that the full body of scientific research literature contains countless high-quality peer-reviewed papers which have demonstrated conclusively that all things do not remain equal. Humans perturb the climate system in many other important ways besides adding CO2. Additionally, this neglected body of research literature suggests strongly that the climate system is a highly non-linear multi-component system, with certain parts such as the ocean having long memory times, leading to important natural variability. These complex interactions mean that very non-intuitive feedbacks between its various components are commonly “normal” behavior. Consequently, this second group of “skeptics” often point out that skillful prediction of how the climate system will evolve over the coming decades to centuries is an enormous technical challenge, and this daunting challenge may even prove to lie beyond theoretical limits of skillful prediction. Some in this group of skeptics has pointed out that claims of skillful predictability “oversell” what the climate science community is actually capable of delivering to the public, and that this overselling might eventually backfire on the community when the public realizes that they have been sold a bill of goods.

My reluctant conclusion from this entire Climategate scandal is that the recent hacked e-mails show a definite tendency by a few of the most well-known and politically connected within the climate science community to willingly look for ways to quash the reporting in peer-reviewed journals of the broader scientific research conclusions reached by this second group of respectable scientists and engineers. This recent window into the internal workings of a few powerful people within climate science world has been an incredibly troubling revelation. Even though Dr. Curry’s calls for more transparency seem at first glance to be a logical and even-handed response, I sense from her letter that she remains woefully naive about what is really at stake here, and the full extent that she herself has been duped.

Calvin Ball: “What pressures were brought to bear that made them decide to fudge rather than recommend more research?”
.
I think “Harry” from the CRU answers this question pretty well:
.
“We’re looking at an *unprecedented* acceleration in temperature, and it’s not due to a sudden lack of volvanic eruptions. Even if it turns out to be naturally-occurring, who’s willing to take that chance? We should be trying to wean ourselves off of unsustainable energy generation and use anyway.”
.
The ends justify the means.

“as long as Judith keeps using the term ’skeptic’ to refer to all those that disagree with what she obviously still believes is a ‘consensus’”

I agree with that. It appears that they are the “skeptics”, not the other way around. They seem to refuse to believe actual observations that do not fit their models and seem “spring loaded” to “adjust” the data to fit the model rather than adjust the model to fit the data.

What bothers me most is what I see as a lack of willingness to accept that there are natural cyclical activities that influence climate and that what has been seen recently tends to support that notion. These cycles seem to operate on longer time scales than is current practice for the determination of meteorological “normals”.

For example, I don’t believe CRUT needs to get rid of current baseline but they should extend it another 30 years back in time. It is becoming apparent that there might be a natural cycle that operates at about a 60 year period that influences temperatures over a lot of the area covered by surface measurement. It seems to me that we should be using a 60 year baseline rather than a 30 year. So if “normal” is the average of 1931 to 1990, things look a lot different than from 1961 to 1990.

These natural cycles were not well known when 30 years was selected for determining what “average” is. They are now known and so it should be the most responsible thing to take them into account when calculating what the “normal” temperature is and to gauge whether climate is warming or cooling outside the range of those normal cycles.

I believe the one single step of going to a 60 year baseline period changes everything but I suppose what I am saying might be heard as “if things were different, then they wouldn’t be the same”. Sometimes one must recognize that they have learned something, make a change and then go forward. Climate science seems to want to do this by “changing” the observations to fit the hypothesis and that is intellectually dishonest.

Could it be that some people see a lot of top jobs becoming available? Those that remain untainted will be polishing up their resume soon… I suspect. I have heard the Brits are looking for a “neutral, respected by both camps” climate scientist to chair an investigation into the CRU affair. Nominations anyone??

Checked out the Climate Progress site for the first time. All I can tell are first look is that most visitors to this site know how to quote the Green Peace Exon blog. If this is her first choice to post, forget it…..

Judith Curry if you could get off your advocacy high horse for a moment, you might want to encourage future climate scientists, who might be temporarily disillusioned with the peripheral email issue, with some advice about sticking to the science – you know the part they must love and want to excel in. Tell them to leave the advocacy to others and that that advocacy might eventually interfere with their scientific judgments. And above all tell those who might have chosen science to further some political cause that they might want to reconsider entering a field that could more directly impact that cause. You might want to tell them that we already have more than enough scientists/advocates.

G. Karst

Uhh……Lucia?

DaveJR (Comment#25074)
Calvin Ball: “What pressures were brought to bear that made them decide to fudge rather than recommend more research?”
I think “Harry” from the CRU answers this question pretty well:

This is ridiculous. Do Harry’s other comments sound as if they were made because of management pressure? This guy is your hero, and he’s saying what he believes.

Nick:

“This is ridiculous. Do Harry’s other comments sound as if they were made because of management pressure? This guy is your hero, and he’s saying what he believes.”

Gawd, what planet are you from? This is “unspinnable!”

“There is no question that there is a political noise machine in existence that feeds on research and statements from climate change skeptics.”

This is a depressingly naive statement – global warming is a minor hypothesis that has been hijacked by left wing activists, politicians, and journalists to promote their view of society – it is a left wing invention. Then the vested interests have boosted it along – publishers wanting doom and gloom stories for their newspapers, carbon traders in banks, alternative energy businesses etc etc. If it wasn’t for the left wing “political noise machine”, it would have been identified as as a weak theory, full of uncertainties and assumptions.
As time goes on, particular implications such as those on infectious disease (1) and hurricanes (2) have been completely refuted. This means the economic justification for action is profoundly weak, and Lomborg has shown that other issues deserve a higher priority.

Hopefully Dr Curry will listen as she proceeds to “3. Take the “high ground:” engage the skeptics”. I can assure her that her side is strong on assertions but the skeptic scientists are very strong on objective analysis.

1. http://www.co2science.org/articles/V12/N4/C2.php
2. http://www.publications.parlia.....12we21.htm

Ms. Curry made a number of reasonable points but I remain a bit skeptical. An idealized version of climate science in which the practitioners generate numbers and conclusions in a detached fashion and then seem mildly surprised that these results have enormous political implications is not credible.

AGW is not a single legislative issue like gun control or federal health care. This is about proposals to effect the greatest reorganization of political, technological and economic power in human history. Given a choice to be (a) at the core of the ’science’ that is the basis for this historic global transformation rather than (b) a physics geek who studies rather elusive weather patterns, I gotta believe (a) has a distinct if not irresistible appeal not to mention the likelihood of vastly greater funding.

I believe Ms. Curry is also part of the “more hurricanes” faction in climate science which is about the weakest claim made on behalf of CAGW and thus not a good credential for arguing on behalf of scientific before policy considerations.

Good on her for addressing the issues is a reasonable way. However, I think she needs to look a little deeper at a larger root problem in her field.

Vincent Grey: “There was proof of fraud all along.”

http://pajamasmedia.com/blog/v.....exclusive/

My goodness. Sounds like it’s time to start to work on redefining what the mainstream media is:

http://online.wsj.com/article/.....ns_opinion

OA

Yep, the mainstream media really needs some redefinition:

http://www.ft.com/cms/s/0/48e1.....abdc0.html

Looks like being certain of your cause is hard.

OA

Nick: “Do Harry’s other comments sound as if they were made because of management pressure?”
.
Calvin asks about “pressure”, you assume “management pressure”.
.
Nick: “he’s saying what he believes.”
.
Spot on! The pressure is “belief” that your cause is right and just. That even if you’re wrong you’re still doing the right thing. Or, in summary, that “the ends justify the means”.
.
The email, incidentally, was to a mailing list @norwichgreenparty.org titled: “Re: [ngp-list] Press Release ‘Global Warning’ talk establishes West Norfolk Green Party”.

And not so mainstream, but informative none the less:

http://blogs.telegraph.co.uk/n.....sh-begins/

OA

Why doesn’t she just read Feynman’s Cargo Cult speech? It avoids all the tribal language, makes the same point far more clearly (and is way more entertaining), and wasn’t written by a member of a disgraced cabal?

For Nick and Dr. Curry,

another gem from the CRU E-Mails courtesy of Steve Milloy at Junkscience:

“On Oct 14, 2009, at 10:17 AM, Kevin Trenberth wrote:

Hi Tom
How come you do not agree with a statement that says we are no where close to knowing where energy is going or whether clouds are changing to make the planet brighter. We are not close to balancing the energy budget. The fact that we can not account for what is happening in the climate system makes any consideration of geoengineering quite hopeless as we will never be able to tell if it is successful or not! It is a travesty!
Kevin”

Especially note the date of Oct 14, 2009.

Dr. Curry needs to take note of the continuing lack of supportive FACTS on her side and adjust her obvious attitude toward sceptics and us deniers.

Her letter reeks of a person being forced by circumstance to engage those they think lower than themselves.

That mail has been repeatedly reproduced before now, kunkhat. I’m not sure what point you think you’re making.

Trenberth is correct, IMV – it is indeed a travesty that as 2009 closes we do not have the necessary observations to determine what is happening with the required degree of accuracy. The Glory mission will be the first satellite to provide such accuracy for tropospheric aerosols, but won’t launch before Oct 2010 -

http://glory.gsfc.nasa.gov/

The DSCOVR (Triana) mission, which would have provided the same for stratospheric aerosol observation, was mothballed by the Bush administration years ago. An FOI request aimed at discovering the grounds for that was rejected by the Bush White House (ironic, huh?) -

http://en.wikipedia.org/wiki/DSCOVR

Observations of the oceans, especially at depth, are not up to the standard needed for accurate analysis. And so on.

I trust that you, like Trenberth, would like to see our research knowledge advanced further. Whatever the message of such observations may be, we certainly need them.

Now we need a skeptic to write a similar open letter for fellow skeptics. McIntyre? Mosh? Bender?

Boris (Comment#25154)
November 28th, 2009 at 11:18 am

Now we need a skeptic to write a similar open letter for fellow skeptics. McIntyre? Mosh? Bender?

Anyone receive any e-mails from a graduate student complaining that skeptics are driving them out of science?

Boris–
Who should it address? What do you propose it says?

Judy wrote in response to letters she is getting from graduate students asking her to address a specific question. What’s the analog for any ’skeptic’?

Judith says ” A better understanding of the enormous policy implications of our field should imbue in all of us a greater responsibility for upholding the highest standards of research ethics.”

This lofty prose from a department head sounds great, but it is not credible. Without enforcement, its nothing more than empty words read from a teleprompter. I see NONE of the usual mechanisms to enforce this either in place or even being proposed. There are no independent assessments; no double-blind tests; no legal ramifications tied to whether future climate predictions are accurate; no licensing process as found in other professions such as medicine, engineering, architecture, law; AFAIK no requirement for ethics education in undergraduate or graduate programs in climate science.

It is plainly obvious to anyone who reads the peer-reviewed literature that catastrophist climate scientists can publish whatever crap will justify the next grant proposal, with no negative repercussions from the climate community.

Until this is cleaned up, climate science deserves no place at the policy table. It has striven hard to earn the well-deserved reputation of carrying out policy-based evidence-making*. Otherwise known as junk science.

*see any of the major EPA regulatory decisions for egregious examples of this process.

chris y (Comment#25163)

It is plainly obvious to anyone who reads the peer-reviewed literature that catastrophist climate scientists can publish whatever crap will justify the next grant proposal…

Oh really? Give me an example, please, from, let’s say, the study of ice cap mass loss for starters (we can work our way through each field thereafter). It shouldn’t be hard for you to demonstrate what is “plainly obvious….crap”. Looking forward to your revelations (either of what I’ve missed as “obvious…crap” or else of the fact that you are talking, er, something similar).

Dr. Curry, A good first step then, would be for climate science to take a big step backward, and acknowledge that the world does not have a reliable GMST record prior to the satellite era. IMHO, this is the single biggest issue facing all, warmer, luke warmer (me) and skeptic. CRU simply has no credibility any longer and E. M. Smith has ably demonstrated GISS credibility +/- 1C and no better, and quite likely worse. I would be all in favour of an international project to determine a consistent method of reporting data, and a transparent multidisciplinary process to determine a method of synthesizing a product the world can rely upon. Such method might well be able to rescue past data and give us something to work with.

Dr.Curry is writing from the perspective of a school counselor; if or when all this blows up, where will these students go after college?

” E. M. Smith has ably demonstrated GISS credibility +/- 1C ”

Lol! You take this to be a fact universally acknowledged, beyond reasonable doubt and thus justifying such an unsceptical assertion? Yo – let’s get rid of all this stuff lacking ‘credibility’ and replace it with unqualified certainties culled from a blog like that! (Who needs credibility anyway when you’re attacking GISS, HadCRUT or whatever? It’s open season for just declaring whatever doesn’t suit you to be suspect!).

Simon, False precision. That alone chases GISS to the ones place. They use instuments which measure to the ones, and then report to hundredths? High School students know better. Don’t need to get into the systematic bias demonstrated by the migration and dropout of thermometers, although unlike large positive water vapor feedbacks, these are documented.

Simon Evans – Can you help me understand the Briffa MXD reconstructions? I think I know how the basic mechanism is supposed to work but I just can´t figure out how to interpret the results. Presumably it is possible to look at how closely the observed temperatures match with the tree ring properties during the calibration period. The correlation between them would be a basis for calculating error bars going into the reconstruction.

In this particular case there seems to be a difficulty because the complete instrumental period shows that the assumption of a linear relationship cannot be sustained. However Briffa has decided to remove the later part of the instrumental period where the parabolic relationship emerges. This means that the reconstruction as performed very much hinges on the assumption that the later part of the instrumental period represents some form of aberration that was not in play during either the calibration or reconstruction periods.

Surely the uncertainties about the validity of this assumption should show up somehow in the error bars of the plotted reconstruction. Perhaps that would be mixing frequentist and Bayesian statistics in an inappropriate manner but I do not see how I can give any credence at all to the reconstruction if the validity of the assumption cannot be quantified.

It seems to be a proposition of the form: If A is true then B follows. B might be very interesting, but if the truth of A is not addressed by anything other than a subjective expert opinion I am not sure that B should be accepted as more than a thought-provoking conjecture.

Perhaps you can help me cut through my confusion.

Judy wrote in response to letters she is getting from graduate students asking her to address a specific question. What’s the analog for any ’skeptic’?

You are absolutely right, Lucia. There are apparently no skeptics who have problems with Pat Michaels erasing graphs or his embrace of The Khilyuk and Chilingar paper. There are no skeptics who agitate for the UAH code. There are no skeptics.

Is Mike Hulme the whistleblower?

http://dotearth.blogs.nytimes......e-11377”

Just wondering.

Dr. Curry does make a pretty good start, but she clearly regards her efforts as missionary work. That is not a good attitude for a scientist. As it is, most of Joe Romm’s commenters seem to have missed her point in any case. They seem to see her as legitimizing skeptics.

Robert, it seems many warmers have morphed into believers, and as such AGW for them is a faith based proposition rather than a science question. Dr. Curry is communicating with heretics and blasphemors, fundamentalist churches don’t tolerate such behaviour. I suspect this will end badly for the faithful, as this time the inquisition will be on the side of science.

Sorry to be a bit off topic but I think this interview with Mike Hulme should be very interesting to the regulars at this blog: http://www.theregister.co.uk/2.....interview/

Robert–
Judy acts as an advocate at times and admits it. In this she is more honest than others.

She believes she keeps her scientific judgment and advocacy separate, and probably succeed about as well as anyone. Let’s face it, all scientists have political opinions. Some are fortunate to work in areas where those opinions can’t possibly affect their interpretation of any results. But in some fields that’s difficult. Judy has admitted to some advocacy and I think honestly seeks for transparency.

“At the heart of this issue is how climate researchers deal with skeptics. I have served my time in the “trenches of the climate war” in the context of the debate on hurricanes and global warming.” (Quote Dr.Curry)

Sorry, as long as Dr.Curry utters phrases like “deal with the sceptics” and ““We won the war — the Intergovernmental Panel on Climate Change was awarded a Nobel Peace Prize, and climate and energy legislation is near the top of the U.S. agenda,why keep fighting all these silly battles and putting ourselves in this position?” then I till don’t really believe she is honestly prepared to talk with people like McSteve on an even level.

The AGW cabal has become a liability for saving the planet and Dr.Curry and increasingly more climate scientists (like Dr.Hulme, Dr.Von Storch, Dr.Zorita) and long time supporters like George Monbiot are now separate themselves from the Mann/Jones clique because of this liability. Clearly the Mann/Jones cabal have lost their credibility and their papers/theories/models will be quickly dismissed if it is not thoroughly peer reviewed by experts of all kinds and free access to all codes and mega data.

History is rewritten on climate science and hopefully the early reactions of Dr.Curry and George Monbiot are an example for other scientists to reach out for better science and less politics. I must say that I do have respect for both for their actions knowing what flak they’ll receive from the AGW zealots. Too many have digged their heels way to far in the sand already to change their mind without a massive loss of face.

Boris (Comment#25154) November 28th, 2009 at 11:18 am
Now we need a skeptic to write a similar open letter for fellow skeptics. McIntyre? Mosh? Bender?

Well, I have no problem writing an open letter to skeptics, but I’m not a skeptic. I don’t know why this is so hard for people to understand. I start like so. Based on my work experience and understanding I believe in radiative physics. Adding GHGs to the atmosphere will warm it. If people want to drag me into debates about that it’s about as interesting as debates about the reality of my hand. Fun, but I don’t have the time. In my mind the next logical thing to look at is the temperature record. And so I try to do so in a methodical fashion. Let’s see the data and the code.
Let’s look at siting, instruments and all those pesky details.

I try not to be diverted by people who are ignorant about the reality of radiative physics. I also don’t read speculative junk about sunspots or cosmic rays or whatever. I do waste sometime with talk about climate models. Why this approach? Well, in the end whatever you say about sunspots and the climate is going to come back to a discussion of the temperature record. Same for cosmic rays and same for the climate models. Boris, you also think that you have some point about UAH. How can I put this to you so that you will understand. In due course. Of course I will push for UAH code. In the case of CRU code I actually will be able to understand the code. I do have some sympathy with researchers who worried about releasing code to people who had no knowledge of code or what it meant. I think that Jeff Id or somebody like him might have a better shot at understanding UAH code than I do. In any case what am I going to do with UAH code? Well, the first thing I will do is claim that UAH has priority over RSS on the basis of transparency.
Second, the next thing I would do is compare the UAH record with the land record and sea record. Dang, that means I have to have the code and data for the that. You see how everything comes back to the thing I clamor for? And you wonder why I focus like a hedgehog on that? It’s brutally simple. Whatever line of evidence you take to support global warming I am going to show how that line of evidence “depends” on the temperature record.
Glacier retreat? Hmm why are they retreating? its getting warmer! really? how much warmer? I dunno, lets see the code and data. The science of global warming makes sense because of coherence between the various strands of evidence. The core strand of evidence is the temperature record. All other claims either rely on it or cohere with it.

My open letter to skeptics. You effin idiots. For the past two years I have told you that you should focus all your efforts on getting the code and data for the global temp so that you can actually have a real debate. Listen up.

Best Irony: if jones et al go down in flames it won’t be because they perpetrated a hoax. It will be because they tried to thwart the FOI process. Even better, when that data and code does get released, the result of an audit will be inconsequential. So, they will go down because they refused to release data and code that would have exonerated them. They were slaves to PR principles that destroyed them while the scientific principles they ignored would have saved them.

steven mosher (Comment#25191)
November 28th, 2009 at 4:10 pm

Even better, when that data and code does get released, the result of an audit will be inconsequential.

Looks like it’s too late for the data.

http://www.timesonline.co.uk/t.....936289.ece

This weekend it emerged that the unit has thrown away much of the data. Tucked away on its website is this statement: “Data storage availability in the 1980s meant that we were not able to keep the multiple sources for some sites … We, therefore, do not hold the original raw data but only the value-added (ie, quality controlled and homogenised) data.”

Old news for the CA crowd.

Simon Evans,

” unqualified certainties culled from a blog like that! ”

Your religion is showing again. When have you or anyone you trust even attempted to audit the work at Chiefio?? Please give us a reference.

Ad Homs are poor form at best.

Simon Evans,

I will keep repeating that even Climate Science insiders do not KNOW enough about the earth to push any policy other than learning more.

How can you determine what to DO, when you don’t know how the thing you are trying to adjust WORKS?!?!?!?!?!?!

Would you like me to try ADJUSTING your brain?? I didn’t think so. In fact, I doubt you would be overjoyed about having any of our best surgical brain specialists fiddle with your brain unless it is one of the PROVEN procedures and it is absolutely necessary. Proven with much empirical experimentation before the first try on a human by the way!! That is the same way I feel about people, with or without religious convictions, messing with our society and environment!!!

Dr. Curry’s attitude is that of religious belief. Even though those in her community WHO SHOULD KNOW DON’T, she proceeds as if SHE DOES KNOW!!The other option is just as bad. She is proceeding because the policy is her real goal, not the science that alledgedly supports the policy.

You doubt me??

” For the past 3 years, I have been trying to figure out how to engage skeptics effectively in the context of #3, which I think is a method that can be effective in countering the arguments of skeptics, while at the same time being consistent with our core research values.”

What is there to figure out?? Free the code. Free the data. Open the discussion. What is there in her core research values that would slow this reasoning?? Why would this not fit into her core research values??

Notice, she puts this in the perspective of EFFECTIVE IN COUNTERING THE ARGUMENTS OF SKEPTICS!!!! There is no doubt in this mind that she and associates are right and everyone else is wrong. She is a religious believer. One can not laud the token bows to Science from someone like this.

Nowhere in Dr. Curry’s statements are there any hint that there is something to be gained from a discussion with the sceptics. Only better ways to counter them.

DeNilistism #3 –

“chicken bones, tea leaves and tree rings can all be read. To find their books, look in the fiction section”

Steve Mosher, would you agree that since the data for CRUTem is unavailable, and presumably destroyed, would you agree that it should be stricken from the record, and never referred to again?

Peter (Comment#25215) November 28th, 2009 at 8:46 pm

My recommendation made here some while ago while we were debating the FOI was as follows: use GISStemp. At least with GISSTEMP we have the code for calculating the anomaly. Further GISStemp have clear documentation on which sites they use and their data sources. However, many many researchers publish their findings using CRU as a source. If we throw out CRU as an index then all those studies go in the trash bin as well. So be it. If people published their code ,”recompiling” their science with GISSTEMP would be relatively easy.

To be clear I already reject CRU since they are not transparent. In my world no paper would reference CRU. Those that have or those that do are similarly rejected. confidential data; IP protected code. Don’t use it.

WRT GISSTEMP the process of validating that code is presently underway by EM Smith. When he finishes then we have a whole LOWER LEVEL of data and code to request.. data and code from NOAA. But until that happens if somebody must use data I’d suggest that they use GISSTEMP. I have no opinion on UAH or RSS. If it’s impossible due to IP restrictions for RSS to release their code, then I’d suggest using UAH.

For me the guiding principle is transparency. For example, some people slag on GISSTEMP because it shows a higher warming trend. I shrug. GISSTEMP is more open and so going forward I think it best that scientists use the data and methods that are most transparent, that is most open to improvement.

Steven Mosher, the intellectual consistency and transparency that you demonstrate and espouse would never fly in climate “science”. Where is the shock and awe?

Boris,

“There are no skeptics who agitate for the UAH code. There are no skeptics.”

And how does this question show, or make, the ground systems record accurate enough for climate science??

All it does is suggest the possibility that the Satellite systems may not be accurate enough either. I can accept that. Can you??

But then, we really need to be measuring the energy in a parcel of air, not just the temperature!!!

Steven Mosher: “But until that happens if somebody must use data I’d suggest that they use GISSTEMP”. As you say, that’s not a very attractive option either. As I understand it, the correction for UHI outside the US is either flawed or absent and it’s probably going to be hard to fix that since most stations in ROW are situated in towns or airports?
And even if we get the code straightened out, we still need to know the influence of microsite biases (surfacestations.org). No?

steven mosher (Comment#25220) November 28th, 2009 at 9:12 pm
Niels A Nielsen (Comment#25239) November 29th, 2009 at 4:22 am

Agree with both of you. From limited glimpses at the CRU code available now, GIStemp seems preferable for so many reasons. It is ‘bottom up’ where at least they start with identifiable data. I reckon the most of the ‘warming’ is in the homogenisation steps, both at individual station and gridbox level. I think the negative UHI corrections (see here for examples) may have more effect than the normal (positive) UHI corrections, certainly outside the US. Why else would they be tolerated? The CRU code on the other hand seems to be model first, add real data afterwards to improve the model, at least that is my take on it after the Harry Read Me file. The stations used are in the documents file FOIA\documents\cru-code\linux\obs\_ref for comparison with GHCN data.

For a lighter take on what E.M.Smith is doing see: GIStemp reloaded (I apologise in advance!)

Peter (Comment#25176) November 28th, 2009 at 1:50 pm

Simon, False precision. That alone chases GISS to the ones place. They use instuments which measure to the ones, and then report to hundredths? High School students know better

Peter, High School students would need to grasp the distinction between the accuracy of area-averaged temperature anomalies and the accuracy of area-averaged absolute temperature. I don’t know how you (or your source) derive “GISS credibility as +/- 1C and no better, and quite likely worse” – perhaps you’d like to give your evidence. GISS does not present an estimate of absolute global mean temperature as being accurate to
0.1°C. I trust we are discussing the same issue.

Jorge (Comment#25177) November 28th, 2009 at 2:11 pm

Simon Evans – Can you help me understand the Briffa MXD reconstructions?…..
……Perhaps you can help me cut through my confusion.

I doubt it, Jorge, since I agree that there remains significant uncertainty as to the cause(s) of divergence. Briffa’s exclusion of recent divergence from calibration limits verification opportunity. I think it’s a problematic issue, which the dendros are continuing to work on, and personally I wouldn’t put much weight on any direct temperature comparison of late 20th century warming with, say, the MWP, when it’s based on divergence-impacted tree-ring data. Just my view.

kuhnkat (Comment#25194)

I criticised the certainty of an assertion, that is not ad hom.

kuhnkat (Comment#25197)

Would you like me to try ADJUSTING your brain?? I didn’t think so. In fact, I doubt you would be overjoyed about having any of our best surgical brain specialists fiddle with your brain unless it is one of the PROVEN procedures and it is absolutely necessary. Proven with much empirical experimentation before the first try on a human by the way!! That is the same way I feel about people, with or without religious convictions, messing with our society and environment!!!

Then we agree – we should not be “messing with our…environment” by pumping further GHGs into the atmosphere. I look forward to you proof that such ‘messing’ will be inconsequential. The point is that your position is not one of taking a ‘no change until proven’ default, it is one of preferring continuing change as a default, even though you assert that you don’t know how the system works and thus can’t calculate the effect of continuing to change the conditions.

steven mosher,
Since GISS in many ways agrees with what CRU claims their temp record shows, I think the real lesson is that we need to find out what mischief Hansen’s team has been doing to the data.
We already know Hansen is far past being an objective scientist at all. He is selling policy ideas based on his conviction taht we are nearing an apocalypse.
We already know that Hansen made crazy predication about Manhattan and sea level rises that were to have taken place by now.
Since his basic work product agrees in large part with what we now know is dubious at best, we are reasonable to question anything that look similar to it.

Peter (Comment#25176) November 28th, 2009 at 1:50 pm

Simon, False precision. That alone chases GISS to the ones place. They use instuments which measure to the ones, and then report to hundredths? High School students know better

Peter, High School students would need to grasp the distinction between the accuracy of area-averaged temperature anomalies and the accuracy of area-averaged absolute temperature. I don’t know how you (or your source) derive “GISS credibility as +/- 1C and no better, and quite likely worse” – perhaps you’d like to give your evidence. GISS does not present an estimate of absolute global mean temperature as being accurate to
0.1°C. I trust we are discussing the same issue.

Jorge (Comment#25177) November 28th, 2009 at 2:11 pm

Simon Evans – Can you help me understand the Briffa MXD reconstructions?…..
……Perhaps you can help me cut through my confusion.

I doubt it, Jorge, since I agree that there remains significant uncertainty as to the cause(s) of divergence. Briffa’s exclusion of recent divergence from calibration limits verification opportunity. I think it’s a problematic issue, which the dendros are continuing to work on, and personally I wouldn’t put much weight on any direct temperature comparison of late 20th century warming with, say, the MWP, when it’s based on divergence-impacted tree-ring data. Just my view.

hunter (Comment#25254)

Since his basic work product agrees in large part with what we now know is dubious at best, we are reasonable to question anything that look similar to it.

Who’s “we”? Would that be “you” and other people like you?

Anyway, that’s great for you, hunter – you’ve now decided that HadCRUT is dodgy so anything that correlates with it must be dodgy. So RSS must be dodgy. UAH must be dodgy too! Shucks, Spencer and Christy are in on the global warming conspiracy! Who’d have thought?

Simon – Thanks for the reply. At least, that is one of the “multiple lines of evidence” that I can strike out for now.

Simon,

I am only discussing anomalies, the absolute temperature is irrelevant. The issue is not accuracy it is precision. E.M. Smith explained it to a doubter thusly, which completely concurs with my high school chemistry, in university they didn’t explain it, they simply chopped your work to bits if you didn’t get it: “Please take at look at the issue of False Precision. The original data are recorded in full degree F, there can be no meaning in any number more precise than that. There is no ‘made better’ or ‘made worse’ from any changes of any sort to the right of the decimal point; there is only a different error inside the error bands of the calculations.” That is not complicated. These are floating point numbers. You may do do any mathematical operation that I am aware of and recover precision which did not exist in the underlying data.

Peter,

i am not particular familiar with E.M. Smith’s take on this, so please excuse my ignorance. Of course, I understand the argument about false precision. However, if you were to imagine ten observations rounding to the nearest degree F (which is not the case in any case), nine of which round to 2 and one of which rounds to 1, then it is not ‘false precision ‘ to take an average of those, it is simply averaging. The average may be expressed to as many places as you like – this is not an expression of absolute precision but simply an expression of the average. The precision is (unarguably) a matter of the averaging rather than a matter of the absolute.

The notion that you’ve proposed, that GISS is only good for +/-!C is, frankly, bonkers. How do you account for its correlation with the satellite record if you think that?

Simon, I cannot stop you from making a fool of yourself, but I would ask you to. You cannot gain precision from averaging. GISS raw dat is reported in whole degrees. Please read about significant digits befor replying.

Seth Pinto (Comment#25058) November 27th, 2009 at 1:47 pm

Has anybody noticed google has removed any search suggestions relating to the AGW scandal? I actually noticed this today because I now have to type in my entire search string. I found someone who actually posted on this.

http://www.smalldeadanimals.co…..12721.html

This has me very worried given googles massive control over information flow.

Go to Bing, plenty of hits there, this is what happens when you get one dominant search engine, the only reason google became No1 was it`s friendly colourful look.

hunter (Comment#25254) November 29th, 2009 at 10:33 am

“Since GISS in many ways agrees with what CRU claims their temp record shows, I think the real lesson is that we need to find out what mischief Hansen’s team has been doing to the data.”

Personally I don’t think there is any mischief at play in GISS. By mischief I mean deliberate and conscious decisions to warm
the record. There are errors in his code. There are analysis decisions made that need to be examined, but I would not
call this mischief. That is motive laden.

“We already know Hansen is far past being an objective scientist at all. He is selling policy ideas based on his conviction taht we are nearing an apocalypse.
We already know that Hansen made crazy predication about Manhattan and sea level rises that were to have taken place by now.”

None of this matters to an analysis of the code and what the code does. bad people can write good code. When hansen turned over his code HE REMOVED his personality, interests, past record, etc from the equation. We now have the code. the person who wrote it doesnt matter. what matters now is this: WHAT does the code do. That is the benefit of turning code over. It now doesnt matter who wrote it. What matters is what does the code do. Is it defensible? are there errors? do the errors matter? how do we improve it.

“Since his basic work product agrees in large part with what we now know is dubious at best, we are reasonable to question anything that look similar to it.”

I don’t agree with your approach to analysis here. I don’t know that GISSTEMP is dubious or that CRUTemp is dubious. I’ve said before that I think that there is a strong possiblity that UHI infects the record. There is only one way I know to the bottom of this problem. A full audit of the data and code: from GISSTEMP then on to GHCN itself.

Hint: it doesnt stop with GISSTEMP code and data.

GISS raw dat is reported in whole degrees.

That is complete, absolute and total bullshit. Don’t talk to me about “making a fool of yourself”! Randomly, see here, for example (Ostrov Vize):

http://data.giss.nasa.gov/work.....tation.txt

How can you make such horribly wrong statements?

Simon, You seem willing to persist, so be it. The data you have presented are monthly averages. The raw data are twice daily readings, T-min, and T-max. The fact that GISS adds the first digit of false precision at this level does not make it OK. The fact that having done so, they wait til the data gets smeared around the globe, and then use this infilled, adjusted, averaged soup to add a furhter digit of false precision in the hundredths place says so much.

Peter,

You have an issue with averaging. That’s nice for you.

How about answering my question? If you think that GISS is only good for +/- 1C (so that is really crap, yes?) how do you account for its correlation with the satellite record? Looking forward to your answer.

Simon, You are confusing two issues. I was speaking to the issue of false precision. Do you get that? I have no issue with averaging, I am stating the scientific fact that you may not start with numbers with precision in the ones place, and do anything to them that creates meaningful information in the tenths place, let alone the hundredths place. Period. Do you get that? The fact that Gistemp correlates to the satellite record (so you say) could not be more irrelevant to this discussion.

Peter,

You said that “E. M. Smith has ably demonstrated GISS credibility +/- 1C and no better”. I am not especially interested in your school-level nattering about false precision. The fact of correlation is entirely relevant to your claim. Your assertion is ludicrous.

Simon, having demonstrated your utter lack of knowledge regarding this matter, your lack of interest is hardly surprising. As tedious as this may be I will try again. A temperature is measured, lets say 20 degrees. Do you understand that what that means is that the temperature is between 19.5 and 20.4 degrees and nothing more, i.e. 20 +/- 0.5? That is the limit of the precision of the instrument. If you cannot understand this simple point, we cannot go on.

Peter,

You are demonstrating your refusal to answer a question put to you before. Don’t put to me silly questions as a means of disguising your refusal. Answer the question put to you first, that’s the honest way.

Simon, what can I say? If you cannot demonstrate that you understand what I have told you, the rest of the discussion is moot. Precison is part of the larger answer. Why go to the trouble to explain complicated concepts, if you have demonstrated that you cannot grasp simple ones?

Peter,

you are embarrassing yourself by not answering the pertinent question, which undermines your ludicrous assertion that GISS is good only for +/- 1c.

Don’t be so entirely ridiculous – of course I understand your boringly obvious statements about precision. What insight do you wish to offer us about the usefulness of those trivial observations? Apparently you won’t even answer a straightforward question.

Simon,

I have to side with Peter. From reading your posts, you can be described as a committed AGW advocate, and we all have experienced the mental gymnastics required of such in this venue and others. I would not accuse others of embarassing themselves, if I was such an obvious political advocate myself.

Andrew

Peter,

the simple matter is that, regardless of your feeble attempts at ad homs, your case is false. If the temperature in grid square x is 10 degrees and the temperature in grid square y is 9 degrees then the average temperature over those two grid squares is 9.5 degrees. That is not ‘false precision’, it is simple averaging, regardless. Please do explain to us what problem you have with this notion.

Oh, and justs to help you out a bit here, Peter, since we’re doing school-level thinking -

Yup, the real temperature in box x could have been 9.6 degrees….

…and the real temperature in box y could have been 9.4 degrees…

and, ho hum, what’s the average? Golly gosh, seems to be the same!

Simon, I have given you every opportunity to stop displaying your ignorance, you have declined. So be it. Grid square x temp is 10, measured with thermometer(s) with whole degree precision so the actual value is 9.5-10.4 Similarly grid square y temp is 9 which means 8.5-9.4 So the average is actually a range with 9.5 +8.5 = 18/2 or 9 on the low end and 10.4 + 9.4 + 19.8/2 or 9.9 on the high end. Thus we can say that the average is between 9 and 10 and nothing more. Actually you can say anything you want, but serious people will not pay attention.

But Peter, you have not established that temperatures are measured with thermometers which only have whole degree precision (that is, there is no physical justification for rounding up or rounding down). Are you serious? Please give your references for this interesting claim.

Regardless, a set of whole degree precise only measurements over an area will give an average which is closer to the actual average

Woops, that was an incomplete post. I’ll tell you this, Peter – when people resort to ad homs, as you have consistently done, then I think little of them.

You’re somebody who won’t answer a straight question, it seems.

So, I am bored with your assertions that those who generate GISS are much stupider than you are.

Simon is quite right. The point is that you know the average with higher precision than you know the individual readings.

An analogy is polling voters. A survey reports that 53% will vote for A, 47% for B. Now any individual voter votes 100% for A or B. And the basic polling measurement consists of asking just one voter. Only two possible results, 100% or 0% for A. That’s low precision. But aggregate 1000 measurements, and you can say A has 53%, with an error of about 3%.

That’s meaningful for an election outcome. And the average temperature, to higher precision, is meaningful as a measure of total heat.

Nick, Do you want to go there? Are you seriously going to try to tell me that sampling a large enough series of discrete events with a yes or no answer to obtain a certain statistical confidence has any bearing on increasing precision by averaging a large number of floating point measurements which initially lacked that level of precision? Are you????

Nick,
Peter is correct you are confusing discrete versus continuous measurement please check a a first year physics (physics 101) or engineering or stats text for a full discussion of the difference of precision

Peter,

At first glance, I might be inclined to agree with Nick. Is it a very foolish position in your opinion?

Oliver

Peter (Comment#25322)
Nick, Do you want to go there?
Absolutely. Yes/no – in quantitative terms it’s 100.0% or 0.0%.

Peter, you should learn a little humility here. There are a lot of people who know a lot about data precision and averaging. GST averages have been around for a long time, and all those people can see it. There is virtually no scientific controversy about it. If you want to make the case you’re making, you’ll have to assemble a much more informed argument.

Andrew Kennett (Comment#25323)
Andrew, I have over thirty years experience as a research scientist in our division of Mathematics and Statistics. And I do remember those books. Peter’s point is that temp measurements are in fact discrete – effectively, integer values (in degrees) are returned.

Nick my point was the opposite, that temperature readings are not integers but measured ranges. Counting voters yield integers. You can average integers and retain precision to infinity (but you know that) you may not do so with temperature readings. I don’t need to make the case, since they are clearly measured numbers with indisputable ranges. You wish to treat them as integers……..you have the case to make, sir.

Peter, the case is elementary. Yes, a thermometer could return a fairly continuous range of results. But the result of the actual reading process is a number from a discrete set of values. That’s what finite precision means. This is familiar in the computer representation of real numbers by finite precision discrete approximations.

Those discrete values form the basic data set that is used for averaging.

Nick, my request is simple. You are claiming that information can be created which did not originally exist, i.e. temperature data in the tenth’s and hundredth’s place, out of temperature readings in the ones place? Please refer to a peer reviewed mathematical paper which makes this case.

Once you start advocating policy, you’ve forfeited you objectivity.

Does that mean that if you advocate against a particular policy or the need for policy in general, you lose objectivity?

I can do better. Here are the NIST guidelines. Check 2.2. The error of combined results is to be expressed as effectively the RMS value. That becomes a reduced value when you form an average. It’s the same as you do with polling error. For the average of N values, it goes down as sqrt(N).

It’s more complicated with GST, because the errors aren’t the same, and the sites are spread out. But the basic argument for reducing precision applies.

I think that the issue of rounding and significant figures is not remotely significant at this point.

I think we need to worry more about a (deliberate?) failure to correct for UHI, historical data revisionism, increased dependence on suspect correction algorithms as the number of actual measurement sites declined all present a more significant set of issues with respect to the reliability of our surface temp records. Whether they were properly averaging bogus data is not terribly important.

Until HadCrut and GISS are under the control and direction of professional scientists not given to political activism and data manipulation we won’t know how significant or reliable the reported results are.

Nick, that is a poor first pass. If, big if, this related to thousands of measurements of the same thing i.e temperature at station “x” at the same time, you would be correct. Since this is not, but rather one measurement at a time, in unique locations this is irrelevant. But then you know this. Next?

Peter,
The discretization errors at stations Xi and Xj maybe be taken to be uncorrelated (unless you think that one station is more likely to be close to an integer temperature than another). The average you compute from N such stations is the sum of the real values plus the sum of the errors which are uniformly distributed on (-0.5,0.5), all divided by N.

So why do the individual readings need to be from the same station for such an “average” to be useful (in the sense of reducing the random noise)?

Just a bit of background on the temperature thing.

In the US historically temperature is recorded once a day. At the time of observation the observer records two numbers: maximum temp during the last 24 hours and minimum temp during the last 24 hours. The observer rounds to the nearest degree F.

Then these two numbers are added and divided by 2.
That result is also rounded to the nearest F.

The daily numbers are then summed and that result is divided by the number of days in the month.

Brohan & Jones et al (2005) argue that the measurement error should be calculated as .2C / Sqrt(60) as they argue that there are 60 measurements made each month with a thermometer that has a .2C accuracy. But are there 60 measurements?
or are there just 30? dunno I’ve had a couple of tenured stats professors say there are only 30 measurements. I’ve had some say that it makes no sense since there is no such thing as the “monthly” temp that is being measured “60″ or “30″ times.

To continue:

Every day the thermometer records two values:

Tmax ( which is good to about .2C 1 sd)
Tmin ( which is good to about .2C 1 sd)

Tmax is then rounded and tmin is rounded. Then they are added.
then they are divided by 2. then that result is rounded.

calculate the measurement error in round( round(Tmax))+ round(Tmin))/2

Then calculate the error when you sum these for a month.

Do you all see something odd here?

Suddenly I know the monthly temperature to a fairly high accuracy? Hmm. Why not calculate the measurement error on a yearly basis. According to Brhan and jones logic the measurement error on the yearly basis would be .2C/sqrt(730) Does that seem funny to you? Why not calculate the average on a DECADE basis.
There are 3650 days in a decade. hence there are 7300 observations in a decade, therefore the measurement error on the decadel figure according to brohan jones logic = .2C/sqrt(7300). Seem odd.

BTW I hate this precision accuracy debate. makes my head hurt.

Nick Stokes:

It’s more complicated with GST, because the errors aren’t the same, and the sites are spread out. But the basic argument for reducing precision applies.

…And because the simple formula applies to normally-distributed, uncorrelated errors, not likely to be the case here.

The other thing to keep in mind here, though, is that we are computing temperature anomalies, rather than absolute temperature. Subtracting the mean temperature removes a lot of sins when it comes to measurement errors.

steven mosher (Comment#25336)
You divide by 60. The answer is in that NIST doc that I cited above. For any of these averages, you’re adding some number of readings (N say) and dividing by N. When you add independent random variables, the variance (square of std error) of the total is the sum of the component variances. So the variance of N values is N times that for one (if they are all the same). To get back to std error, you take the sqrt, and then divide by N (averaging). The nett result is .2/sqrt(N).

So going from month to year, etc, the std error diminishes more, reflecting the fact that you have more observations, even though the temperatures themselves vary a lot.

If they are not independent, the error of the average diminishes less than if they are.

Simon Evans (Comment#25315) November 29th, 2009 at 6:36 pm
“But Peter, you have not established that temperatures are measured with thermometers which only have whole degree precision (that is, there is no physical justification for rounding up or rounding down). Are you serious? Please give your references for this interesting claim.
Regardless, a set of whole degree precise only measurements over an area will give an average which is closer to the actual average”

Lemme see:

http://www.nws.noaa.gov/om/coo.....-notes.htm

You will see the instructions for a B91 given here. basically this is the form that observers fill out when recording observations.

Carrick (Comment#25338)
Normality is not assumed. Correlation has an effect, but it’s reading error that we’re talking about, so they are likely to be largely uncorrelated across sites.

Subtracting a mean doesn’t reduce the error – in fact it increases it, by an amount that may be small.

Nick Stokes (Comment#25339) November 29th, 2009 at 11:14 pm

Sorry Nick the stats profs I talked to said that you don’t divide by 60 you divide by 30.

You have 30 measurements of the MAX
you have 30 measurements of the MIN

You have 30 measurements of (tmax+tmin)/2 or Tave.

See?

steven,
When you expand your expression for Tave, what you have is 60 measurements of some kind of t (max or min) + 60 error terms which (unless there is something funny going on) don’t particularly care whether they were maxes or mins. Thus the discretization error for Tave should look like something gotten for 60 readings. Is that a fair way to look at it?

steven mosher (Comment#25343) November 29th, 2009 at 11:21 pm

You have 30 measurements of (tmax+tmin)/2 or Tave.

Yes, but is the sd of Tave for the month the same, greater than or less than the sd’s of Tmax and Tmin? Off the top of my head, I would guess that Tave has a lower sd by a factor of sqrt(2), not coincidentally exactly equivalent to dividing the variance of Tmax or Tmin by sqrt(60).

steven mosher (Comment#25343)

Sorry Nick the stats profs I talked to said that you don’t divide by 60 you divide by 30.
You have 30 measurements of the MAX
you have 30 measurements of the MIN
You have 30 measurements of (tmax+tmin)/2 or Tave.

Yes, but the error of the mean (tmax+tmin)/2 is .2/sqrt(2). That’s what you have to divide by sqrt(30). Nett result – divide by sqrt(60).

That’s the great thing about sum squares. Stick to it and it all works itself out.

Nick Stokes:

Normality is not assumed. Correlation has an effect, but it’s reading error that we’re talking about, so they are likely to be largely uncorrelated across sites.

Actually, normality is assumed in the interpretation of “the error of the mean”, computed as sigma/Sqrt(N), in terms of its usual confidence interval (68.3%).

This confidence interval is given by sigma/Sqrt(N) for normal, uncorrelated noise. For other distributions, the factor is different than 1/Sqrt(N) (though for large N, it approaches 1/Sqrt(N) of course, by the central limit theorem).

Uniformly distributed random errors is a good place to start in showing a counter-demonstration to your claim that it doesn’t depend on the distribution being normal.

Correlation has an effect, but it’s reading error that we’re talking about, so they are likely to be largely uncorrelated across

Um… no.

One of the largest correlated error is systematic bias in the instrumentation. If you are using the same equipment across a large number of sites (or even have similar station location systematic errors), then the correlation associated with this error will approach 100%.

Likewise if you have a series of poorly-sited urban stations that you average over, the result will have the same systematic offset (as individual stations) from the “true climate temperature”.

Subtracting a mean doesn’t reduce the error – in fact it increases it, by an amount that may be small.

If you have large systematic uncertainties associated with site selection, then subtracting the mean removes the scatter associated with that site selection. For large N (e.g. the problem at hand here), the growth in the variance from subtracting the mean is completely negligible of course.

So if you look at the scatter associated with temperature residuals, this is generally much smaller, than if you plot average temperatures for each site. Take home message is your job is much easy in climatology than in meteorology, because you aren’t trying to measure temperature gradients between sites, but rather variation in temperature residuals over time.

Yes, but is the sd of Tave for the month the same, greater than or less than the sd’s of Tmax and Tmin? Off the top of my head, I would guess that Tave has a lower sd by a factor of sqrt(2), not coincidentally exactly equivalent to dividing the variance of Tmax or Tmin by sqrt(60).

Tavg has the same expectation value as (Tmax+Tmin)/2 if the mode equals the mean (this is almost certainly not true of temperature data).

For other distributions, (Tmax+Tmin)/2 is generally a better (in the sense of unbiased) estimator of central tendency than Tavg.

Carrick (Comment#25347)
You only need a distribution to get the tail probabilities. The arithmetic of variances and standard errors is distribution-free.

I’ve talked of reading errors. I didn’t include systematic biases of instruments etc, since they just shift the mean and don’t affect the variance (unless they change). But yes, subtracting the mean, as with anomalies, does help to remove their effect.

In my last comment I hadn’t yet read DeWitt’s remark, which was spot on. But he’s just defining Tave as (Tmax+Tmin)/2.

Nick Stokes (Comment#25346) November 30th, 2009 at 12:19 am

Dang it. I’m gunna have to go back to 2007 comments that RomanM made over at CA. Maybe I misrembered how he explained it to me. wouldnt be the first time.

The search at CA sucks and it was 2 years ago so until I find the comment I’ll accept your explanation. Still doesn’t sit well.

Nick:

You only need a distribution to get the tail probabilities. The arithmetic of variances and standard errors is distribution-free.

Not really… but yeah, you could define formulas to calculate “variance” and “error of the mean” in terms of formulas that do not depend on the distribution (though that isn’t standard in statistics, actually).

But so what? The only reason to compute the standard error of the mean is in terms of understanding the associated confidence interval.

Most of the time when we quote uncertainties of a derived quantity experimentally (like a mean), we are referring to the 68.3% CL. In that sense, error of the mean is related to the square-root of the variance by a 1/sqrt(N) only if N is very large or if the distribution is nearly normal.

For myself, I generally quote at the 95% CL (you can’t take the 68.3% CL errors of mean and multiply them by two in general… the tails are where the distributions usually look the least Gaussian).

I’ve talked of reading errors

And until you announced that in response to my comment, how would anybody know this?

What you are saying holds for that of course, but in any case, one needs to consider the realistic error budget here, not just one element of it that is very small compared to other error sources.

Arrg Nick. It may have been UC who made the comment. Can’t find it.

Slight follow on:

The definition of variance (for a continuous random variable X) is:

Var(X) = E[(X-µ)^2] = int_a^b (x-mu)^2 P_X(x) dx

where P_X(x) is defined on the interval (a,b).

So, yes, in terms of statistical definitions, Var(X) most certainly depends on the distribution being considered.

Also,

µN = (X1 + X2 + … + XN)/N

can be treated as another random variable, whose distribution is given by another distribution P_µN(x), and it’s variance (the square of the error of the mean) is calculated in exactly the same manner as above. And in general N Var(µN) != Var(X).

So at least in standard jargon of statistics, this isn’t a true statement: “The arithmetic of variances and standard errors is distribution-free.”

Typo: I mean to say “And in general N Var(µN) != Var(X)”. I corrected it after submitting, but the correction doesn’t seem to be taking hold.

Carrick (Comment#25352)
And until you announced that in response to my comment, how would anybody know this?
Well, you did. You quoted it
“but it’s reading error that we’re talking about, ”
#25354
If you want to express the variance sum relation for independent X1, X2: var(X1+X2)=var(X1)+var(X2) (which is really all I’m using) in terms of the second moment integrals you can, and it’s still true regardless of P_X1 and P_X2. The reason is that the pdf of X1+X2 is the convolution of P_X1 and P_X2, and if you work out the second moment for that, it gives the above relation (the moment generating function helps).

Comment #25342 in response to me was the first time you made a distinction, which to me seemed artificial and overly constraining. Anyway…this is getting argumentative, let’s move on, sorry.

The reason is that the pdf of X1+X2 is the convolution of P_X1 and P_X2, and if you work out the second moment for that, it gives the above relation (the moment generating function helps).

Ah, that’s a good point. Learn something new every day!

My comments about the important of CLs and the interpretation of the square root of variance still holds of course.

Again a follow-up – here’s the Wiki version. This sum property is the basis of ANOVA.