People send me stuff. In this case I have received an embargoed paper and press release from Nature from another member of the news media who wanted me to look at it.
The new paper is scheduled to be published in Nature and is embargoed until 10AM PDT Sunday morning, July 20th. That said, Bob Tisdale and I have been examining the paper, which oddly includes co-authors Dr. Stephan Lewandowsky and Dr. Naomi Oreskes and is on the topic of ENSO and “the pause” in global warming. I say oddly because neither Lewandowsky or Oreskes concentrates on physical science, but direct their work towards psychology and science history respectively.
Tisdale found a potentially fatal glaring oversight, which I verified, and as a professional courtesy I have notified two people who are listed as authors on the paper. It has been 24 hours, and I have no response from either. Since it is possible that they have not received these emails, I thought it would be useful to post my emails to them here.
It is also possible they are simply ignoring the email. I just don’t know. As we’ve seen previously in attempts at communication with Dr. Lewandowsky, he often turns valid criticisms into puzzles and taunts, so anything could be happening behind the scenes here if they have read my email. It would seem to me that they’d be monitoring their emails ahead of publication to field questions from the many journalists who have been given this press release, so I find it puzzling there has been no response.
Note: for those that would criticize my action as “breaking the embargo” I have not even named the paper title, its DOI, or used any language from the paper itself. If I were an author, and somebody spotted what could be a fatal blunder that made it past peer review, I’d certainly want to know about it before the paper press release occurs. It is about 24 hours to publication, so they still have time to respond, and hopefully this message on WUWT will make it to them.
Here is what I sent (email addresses have been link disabled to prevent them from being spambot harvested):
===============================================================
From: Anthony
Sent: Friday, July 18, 2014 9:01 AM
To: james.risbey at csiro.au
Subject: Fw: Questions on Risbey et al. (2014)
Hello Dr. Risbey,
At first I had trouble finding your email, which is why I sent it to Ms.Oreskes first. I dare not send it to professor Lewandowsky, since as we have seen by example, all he does is taunt people who have legitimate questions.
Can you answer the question below?
Thank you for your consideration.
Anthony Watts
—–Original Message—–
From: Anthony
Sent: Friday, July 18, 2014 8:48 AM
To: oreskes at fas.harvard.edu
Subject: Questions on Risbey et al. (2014)
Dear Dr. Oreskes,
As a climate journalist running the most viewed blog on climate, I have been graciously provided an advance copy of the press release and paper Risbey et al. (2014) that is being held under embargo until Sunday, July 20th. I am in the process of helping to co-author a rebuttal to Risbey et al. (2014) I think we’ve spotted a major blunder, but I want to check with a team member first.
One of the key points of Risbey et al. is the claim that the selected 4 “best” climate models could simulate the spatial patterns of the warming and cooling trends in sea surface temperatures during the hiatus period.
But reading and re-reading the paper we cannot determine where it actually identifies the models selected as the “best” 4 and “worst” 4 climate models.
Risbey et al. identifies the 18 originals, but not the other 8 that are “best” or “worst”.
Risbey et al. presented histograms of the modeled and observed trends for the 15-year warming period (1984-1998) before the 15-year hiatus period in cell b of their Figure 1. So, obviously, that period was important. Yet Risbey et al. did not present how well or poorly the 4 “best” models simulated the spatial trends in sea surface temperatures for the important period of 1984-1998.
Is there some identification of the “best” and “worst” referenced in the paper that we have overlooked, or is there a reason for this oversight?
Thank you for your consideration.
Anthony Watts
WUWT
============================================================
UPDATE: as of 10:15AM PDT July 20th, the paper has been published online here:
http://www.nature.com/nclimate/journal/vaop/ncurrent/full/nclimate2310.html
Well-estimated global surface warming in climate projections selected for ENSO phase
Abstract
The question of how climate model projections have tracked the actual evolution of global mean surface air temperature is important in establishing the credibility of their projections. Some studies and the IPCC Fifth Assessment Report suggest that the recent 15-year period (1998–2012) provides evidence that models are overestimating current temperature evolution. Such comparisons are not evidence against model trends because they represent only one realization where the decadal natural variability component of the model climate is generally not in phase with observations. We present a more appropriate test of models where only those models with natural variability (represented by El Niño/Southern Oscillation) largely in phase with observations are selected from multi-model ensembles for comparison with observations. These tests show that climate models have provided good estimates of 15-year trends, including for recent periods and for Pacific spatial trend patterns.
of interest is this:
UPDATE2: rebuttal has been posted
Discover more from Watts Up With That?
Subscribe to get the latest posts sent to your email.
“Not really. in fact they are biased and weirdly averaging them gives you the best answer. just fact.”
It’s a profound assertion, but it runs into a logical contradiction.
If it is true, we’d never throw out old model predictions (regardless of bias and other issues). Any new model results would be inferior, and we could only use the by adding them into the superior “grand ensemble average”. And each addition of inferior results would improve the “grand ensemble average”.
This would need to be confirmed by a robust validation methodology (there is no escaping this requirement).
But if the methodology confirms that one set of results is inferior by blending this into a superior set of results (both cases using the same methodology and tests to determine which is inferior and superior).
In other words, the best thing to do with inferior model results is to throw them away if we have superior results to hand.
It appears that the missing ingredient is the rigorous validation of the models. Until we have this, assertions that the average of model results is better than individual results is not supportable.
It leaves the same questions hanging over the above paper: why does adding three “also-rans” improve their analysis compared to just using the “winner”?
If you average models and there is one halfway right model in there it will track better than any ensemble of anonymous incorrect models is my reading of above comments. Still not a very good model but?
…and you find these in Mosher??
poptech,
You assert it is not possible to look bad while stating facts, yet you manage to do just that.
And going to your website is to tour an example that supports Willis’ argument against anonymity.
You shred someone while hiding behind your anonymity. You make those of us who support anonymity on the internet look bad. You actually posted Steve’s picture along with a questionable interpretation of his CV. But why stop there? His home address is a “fact”. His car is a “fact”. His kid’s names and pictures are “facts”. Why don’t you do like the climate thugs here in Houston and put on a mask and go stand in front of his house and tell him how bad he is?
You are demonstrating that the climate obsessed true believers are not the only ones who can do boorish low class extremist behavior.
If they don’t name the 4 best then it becomes harder to check their work.
kcrucible commented on A courtesy note ahead of publication for Risbey et al. 2014.
>>If they don’t name the 4 best then it becomes harder to check their work<<
_________________________________
It is a matter of belief and climate religion, hence withstanding all checks and logic thinking. And the alarwarmists don't want to be checked and their formidable prejudices destroyed by hard facts: 17 years + 10months…
“It leaves the same questions hanging over the above paper: why does adding three “also-rans” improve their analysis compared to just using the “winner”?”
The obvious answer is that even “the winner” has problems, which are obscured by the outputs of the others.
However, given that they’re not naming the 4 best that they’re citing (how could that possibly have gotten through peer review?? That’s not even science, just a writing obviousness.), it could be that increasing the number of elements increases the validation-complexity… much like password length increases the degree of brute-force hacking time required.
I await with bated breath to see how they further convulse and tie themselves in knots trying to explain away the halt in global warming. I expect we’ll see more of an emphasis on the phrase “climate change” and “unusual weather”, as if the CO2 has somehow, (by magic one can only presume) morphed into those other, undefinable qualities.
Even the “best” climate models have a fundamental, fatal flaw; they simply assume that CO2 is a major driver of climate. They can tweak and fiddle with the knobs until kingdom come, and they will still be totally wrong.
Steven Mosher says:
July 19, 2014 at 3:12 pm
Do you mean that as a general observation, or is the scope of that remark confined to the 18 climate models in question here?
1. general observation about all the models
By “better tool” do you mean more consistent with observations? How do you judge performance? Do you account for differences in inflection points in your measurement?
1. pick your skill metric.. but more consistent yes.
Is not an average of a bunch of models simply another model?
1. A+ answer
Does that imply that some kind of averaging process internal to a model makes it a better model?
1. no
How so? Is it always the case that increasing the number of models in the “average” increases the accuracy? Is it a linear improvement or something else?
1. Not always the case. I never looked at the improvement stats
To make it a “better tool”, do you have to apply weights (non-unit)? How are these weights derived? What kind of average is it? Arithmetic? Geometric? Harmonic?
1. weights are a big debate. currently no weights
I’d be interested to know on what theory you base your assertion, because, for the life of me, I can’t see it.
1. No theory. pure fact. If you take the mean of the models you get a better fit. why? dunno.
just a fact.
=================================
You never looked at any data on how much “better” the average is than the individual model prediction, but somehow you just know the average is “better”?
Well I admit I don’t know very much, but this sounds a little sub-scientific to me. I can somewhat understand someone how has a solid theory of operation being overconfident to the point where they don’t feel they need to look at the data, but you’re telling me you have no theory as to why it works, and you haven’t looked at the data to see if, in fact, it does work. Yet you confidently make the assertion that the average being “better” than a single model is fact. You’re having me on, right?
“I don’t know why that beetle in the matchbox wiggles when it’s about to rain, but it’s a fact…”
Mosher writes “The issue is the four worst on this test will be the best on Some other test”
The issue is that none of the models do well on all of the tests and therefore cant be modelling the way the climate changes.
Defences like “based on physics” are hilarious when they’re all based on physics but all get very different results. Getting a fit will work with any series of inputs. Just because a few of them test well means nothing when others dont and cant to get the model’s optimum results.
Maybe what they meant was they did a BEST splice of output of four different climate models and found corrlelation with something or another.
Mosher also writes a bit later “Not really. in fact they are biased and weirdly averaging them gives you the best answer. just fact.”
Rubbish. Taking the “best” 4 and averaging them apparently gives a better result. That’s another fact from this paper. In fact if you have a bunch of random results and take the “best” of them you’ll always get a better result. And if you average the lot then that average will be better than half of them.
Climate models must deny century scale chaos or they have no predictive ability.
Yet climate is long term chaotic from first principles of massive century scale ocean fluid dynamics.
Climate models have few real data inputs, merely solar output that is too steady to matter, the greenhouse effect that is useless since equivalent warming occurred in the beginning of the global average plot, and pollution which can’t explain mid-century cooling since now we have another multidecade pause after pollution cleared up.
The simplest act of real scientists would be to use the *measured* climate sensitivity to now recalibrate their positive feedbacks into more neutral ones. Then nearly all the models would show the pause as just another bit of noise in a much less warm future.
But where did they get their climate sensitivity via positive water vapor feedback in the first place? They made it up! It’s a constant added to their software.
Just plug in Richard Linden’s updated feedback estimate of nearly no positive feedback and you are done doing proper science. Alarm is then called off. Another recent paper estimates feedback as near null as well:
http://www.worldscientific.com/doi/abs/10.1142/S0217979214500957
The alarmists keep lying about how dangerous future warming is locked in due to the physics of the standard greenhouse effect but it’s really their amplification of it instead that adds degrees to it and that amplification is now two decades falsified. They use willful slander to label all skeptics greenhouse effect denying Sky Dragons, and if they are that desperately dishonest, that is quite telling.
What does Mosher’s splitting of hairs here accomplish in the face of that? It distracts from news of the basic falsification of high climate sensitivity. It distracts from the lie of how the mellow and thus beneficial raw greenhouse effect has been turned into Godzilla by a single line in a computer program. It distracts from laypersons finding out that the government and its scientific enablers have become Enron. Don’t let these guys distract you from loudly exposing their refusal to simply empirically downgrade their climate sensitivity now that it is the only rational and moral thing to do.
-=NikFromNYC=-, Ph.D. in carbon chemistry (Columbia/Harvard)
Bob Tisdale says:
I’m looking forward to hearing your remarks. I hope the discussion on that thread doesn’t get hijacked by a discussion of how Steven Mosher dresses. :/
If I recall correctly &/or as I understand it, “their” fundamental premise behind AGW/C^3 (anthrocentric global warming/cataclysmic climate change) is that prior to industrialized man (i.e. coal fired power plants) the atmospheric CO2 concentration was in natural balance, sources and sinks in perfect harmony, at 268.36490 ppm by molecular volume in molar carbon equivalents.
The rapid increase in atmospheric CO2 concentrations as measured by the Keeling curve at Mauna Loa (data which must be “adjusted” to account for nearby volcanic outgassing) could only be due to mankind’s industrial activity (CFPPs). The Keeling curve and the global temperature hockey stick were then combined into sufficient coincidence to equal cause for concern.
Now “they” are offering an explanation for the 17 year hiatus in global warming while atmospheric CO2 concentrations continue to climb, zipping past 350 ppm and past 400 ppm several years ago at NOAA’s inland tall towers. (never hear about them) The ocean, “they” now admit, is more of a CO2/temperature sink than “they” previously understood. Well that pretty much trashes “their” fundamental premise. If “they” don’t really understand the sinks it stands to reason “they” also don’t understand the sources. IPCC AR5 pretty much admits the same in TS.6 Key Uncertainties.
The Keeling curve atmospheric CO2 concentrations and industrialized mankind’s contributions (CFPPS) when considered on a geologic time scale (at least 10,000 years) are completely lost in the data cloud of natural variations.
No melting ice caps, no rising sea levels, no extreme weather, no rising temperature. “They” were, are, and continue to be wrong. Get over it, the sooner the better.
Mosher writes “The issue is the four worst on this test will be the best on Some other test”
Here we have 18 climate models.
These 4 here reproduce the pause, but show accelerated CAGW in the future.
What criteria can we come up with, post hoc, to justify calling these 4 models the “best”?
OK, write it up.
Steve Mosher, so far your name appears 77 times on this thread, and looking through the comments, it doesn’t appear that many persons agree with you, rightly or wrongly.
Note to the others: If I may suggest, please drop the ad homs with respect to Steve. You’re not adding anything relevant to the discussion.
The reason the average of many runs of a single model is better is that the individual model runs are all over the place and so the odd excursions cancel out. A possible reason averaging multiple runs from multiple models MAY give you better answers for some questions is that since the models all have some differences (else they would not be different models), some may capture some effects while others capture different effects. For many projections, the averaged models do not give very good results (IMO). There has been some discussion of “throwing out” the worst performing, most highly warming models, but this has not occurred yet. What “democracy of the models” means IMO is that no one wants to put themselves on the record criticizing anyone else’s model. Eventually, people may realize their model is too far off and begin to change it and of course will get more publications from doing so. In many fields other than climate, scientists would be more critical and more open about which models performed poorly. If it turns out to be true that this paper does not “name names”, then this would be a sad statement about the state of climate science. Reminds me of the Harry Potter novels and “He Who Must Not Be Named” and with the same implications. People are scared of offending the powerful and connected. But rather than fearing the “Avada Kedavra” curse, they fear losing grant funding and the scientific ostracization and harassment so recent experienced by Dr. Lennart Bengtsson.
Kate Forney, sounds like you’re new here. Welcome. With respect to model outputs, you wrote, “You never looked at any data on how much “better” the average is than the individual model prediction…”
Not to be nitpicky, but the outputs of climate models are not data. Full definition (1) from Merriam- Websters:
“factual information (as measurements or statistics) used as a basis for reasoning, discussion, or calculation <the data is plentiful and easily available"
http://www.merriam-webster.com/dictionary/data
Climate model outputs are definitely not "factual information."
More generally, it's best not to use the term data when talking about climate model outputs so that readers can differentiate between observations (data) and model outputs (computer-aided conjecture).
Dear Watts et al:
Your criticisism appears to be beg the question because you assume, as they assert, that the models are meaningful. Suppose they tell you, with supporting evidence, which models are best/worst, would this improve the paper?
The right answer is No, because the models are constructed to hindcast and the data used to calibrate them is highly suspect. Look inside one of these things and what you find is some 1960s fortran and a great many (thousands in the one I took apart) of encrusted adjustments designed to add or modify the model’s behavior – and all of it parametrized to fit some data set.
Unfortunately the data is suspect – I am now quite sure that there may be a decline, but there is no pause: early data has been adjusted downward, later data upward – and that limits the predictive power of these models to coincidental concordance arising from the narrowness of the predictive band.
Paul Murphy says: “Your criticisism appears to be beg the question because you assume, as they assert, that the models are meaningful. Suppose they tell you, with supporting evidence, which models are best/worst, would this improve the paper?”
The findings of their paper can not be reproduced unless the models they selected are known.
Why is Mosher given free reign to troll in the comments? Because that’s all he ever contributes here.
[Because he contributes and doesn’t contravene the site rules.. . mod]
Bob Tisdale says:
July 20, 2014 at 7:38 am
========================
Thank you Bob. I’ll bear that in mind.
How better might I have phrased the question, the point of which was to interrogate Mr. Mosher regarding how he could know an “average of the models” was more informative than any single model?
He admits he hasn’t looked at any performance measures, nor does he have any plausible theory with respect to how his assertion could be, so I can’t comprehend the basis for his confidence.
Thank you for the pleasure of your replies.
Now, let me reverse your “averages are more accurate” summary – though I know the total answer is more than just that.
We have “one list of data” – that of temperatures recorded to various degrees of accuracy at very specific over the past years, and a much longer set of proxy temperatures of varying degrees of accuracy (inaccurate temperatures, and inaccurate dates of each inaccurate temperature) over a much longer period of time.
Now, has ANY single run of ANY model at ANY time reproduced today’s actual record of temperatures over the past 150 years of measured temperature data across the continental US?
The past 100 years across India?
The past 250 years of measured temperature data across the northeast US and Canada?
The past 350 years of measured data across central England?
That is, has any climate model at any time actual reproduced any temperature record at specific regions over a long period of time?
Supposedly, a “climate model” duplicates the earth’s “average” climate by numerically breaaking up the earth’s into zones for boundary-value “exchanges” of that box with other boxes above, below, right-left-north-south of each box. The results then are grouped togther to define that date-time-group’s “average” total earth anomaly, then everything is reset, and everything is run again.
So => ALL “boxes” are known, therefore, you can get a list of temperatures for any length of time for any region on earth. Each computer run is a unique calculation, so you can’t pretend that the results of the tens of thousands of model runs on each of the 18 or 21 or 23 climate models is “not available”.
Has any model actually worked over any lengthy period of time – outside of the “forced” programming times of varying input forcings (deliberately modifying cloud, solar, particles, etc) designed to yield results that mimic the temperature record?
Now, separately, Paul Murphy very correctly adds a critique similar to mine:
July 20, 2014 at 7:51 am
Dear Watts et al:
That is, if the model is “calibrated” by artificially changing past forcings so past calculated temperatures are “correct” and “do” match the temperature record,
… (2) is the temperature record they are trying to match actually corrected, or actually corrupted, by your fellow bureaucrats’ constant work as they change the past recorded temperatures?
… (3) Do the model runs (even with artificially padded and subtracted forcings) duplicate the past temperature records over long period of time? Or are they really nothing more than “if this year is 1915, then the average global temperature = 24.5 degrees after the model run”
2. After a 15 year run, what is the actual result of a single model run?
Show us the winds, temperatures, humidities, aerosols, the box-by-box sizes and shapes, ice coverage, cloud coverage, and the hourly pressures and temperatures after “32 years of model run 07-16-2014” … All that is ever reported is a final temperature difference at a mythical date in a mythical future free of future changes except CO2 levels.
From Bob Tisdale on July 20, 2014 at 7:23 am:
FWIW, I’ve been defending the person against libel, not agreeing with what he said which I was only peripherally aware of from other comments.
How much of a internet arrogant bully and elitist snob must one be to call Mosher computer illiterate? That’s like saying someone who regularly converses and corresponds in English is illiterate because they lack an English degree. It should be pretty clear that having said degree ain’t no guarantee you can always speak English good.
I could not agree more with the above post/comment by Bob Tisdale. Well put.
However, the “data sets” put out by the government agencies are now so “adjusted” by incompetence, bias, half-assed computer algorithm, “in-filling”, zombie stations, and so on that I don’t think the word “data” fits there either.
For just one example, this very morning I read: “TOBS Update: Something Seriously Wrong At USHCN” http://stevengoddard.wordpress.com/2014/07/20/something-seriously-wrong-at-ushcn/
We need a good word for that stuff that should be data but is not data.