Finally, a study that shows unreliable models are the root of overwrought 'extreme weather' events

Dr. Judith Curry tips me to this new study in GRL:

Unreliable climate simulations overestimate attributable risk of extreme weather and climate events

Omar Bellprat, Francisco Doblas-Reyes


Event attribution aims to estimate the role of an external driver after the occurrence of an extreme weather and climate event by comparing the probability that the event occurs in two counterfactual worlds. These probabilities are typically computed using ensembles of climate simulations whose simulated probabilities are known to be imperfect. The implications of using imperfect models in this context are largely unknown, limited by the number of observed extreme events in the past to conduct a robust evaluation. Using an idealized framework, this model limitation is studied by generating large number of simulations with variable reliability in simulated probability. The framework illustrates that unreliable climate simulations are prone to overestimate the attributable risk to climate change. Climate model ensembles tend to be overconfident in their representation of the climate variability which leads to systematic increase in the attributable risk to an extreme event. Our results suggest that event attribution approaches comprising of a single climate model would benefit from ensemble calibration in order to account for model inadequacies similarly as operational forecasting systems.


0 0 votes
Article Rating
Newest Most Voted
Inline Feedbacks
View all comments
February 22, 2016 3:27 pm

Bingo. Now, there’s a game design/development approach.

Reply to  Evan Jones
February 22, 2016 3:36 pm

Which is why gaming and virtual reality has such a dominant presence in what is being pushed in K-12 education all over the world. The media is perfectly aware that education is pushing the visual over symbol systems like math and genuine print fluency. The Next Generation Science Standards in the US fully intend to make students believe that weather, Climate, and social systems like cities and economies function just like a circulatory or other physical system.
With students exposed to graphics repeatedly that model just that, they will perceive that the world works in ways it does not. It’s not what is true that drives us, but what we believe to be true. Every social scientist and educator interested in social transformation knows that.
It’s such a part of the global reconstruction plans that UNESCO actually combines the two useful areas of communication for transformation as ‘media education.’

Reply to  Robin
February 22, 2016 5:06 pm

Well said Robin. Belief produces votes. Votes control elections. Election control Government policies.
Most people “get” that bad models will yield bad conclusions but few people understand “models” well enough to judge them.
Nearly the same subject: Repeat a lie (or just B. S.) enough times and it becomes a “canon” of a society. The unwillingness of the knowledgeable to confront the left about the Global Warming Hoax allows the watermelons to destroy free enterprise and representative government.

Reply to  Robin
February 22, 2016 7:07 pm

Today an article on…And they are not stopping there!

Reply to  Robin
February 23, 2016 9:29 am

The Next Generation Science Standards in the US fully intend to make students believe that weather, Climate, and social systems like cities and economies function just like a circulatory or other physical system.
You neglected to mention my personal favorite among systems that can be modeled just like any physical system … The Stock Market :^)

Reply to  Robin
February 27, 2016 6:07 pm

Don’t underestimate game theory. Every iota of work I have done on the surface stations paper is grounded in game theory. And I have to say, this one’s the best game I’ve ever played. The most fun, too.

george e. smith
Reply to  Evan Jones
February 23, 2016 12:07 pm

“””””….. The implications of using imperfect models in this context are largely unknown, limited by the number of observed extreme events in the past to conduct a robust evaluation. …..”””””
Anybody got an English language translation of this riddle ??

Reply to  george e. smith
February 23, 2016 1:16 pm

I think it means
“Yeah, we’re trying to reverse sides, and keep funding, so we have enough unknowable unknowns to prove we need more money. Much more money!”
I may be wrong but – follow the funding [or cherchez la femme]!

Reply to  george e. smith
February 27, 2016 6:11 pm

Well, as Boss in The Great McGinty said when asked what to do about the guy who voted 37 times (@ a buck a shot), “Pay him!”

February 22, 2016 3:44 pm

This lays out UNESCO’s position and how it ties to Climate hype, education, dirigisme, and the important role of multimedia.
People do not appreciate that what are called ‘tests’ will now actually be what are called ‘formative assessments’ monitoring what students do in open-ended problem solving in virtual reality.

Reply to  Robin
February 22, 2016 4:35 pm

The trick is to get the analysis right. Including quantification of unknowns.

February 22, 2016 3:58 pm

Reblogged this on Climatism and commented:
And yet still, draconian climate policy and literally trillions of dollars worth of ‘green’ schemes and scams are dictated and directed around overheated IPCC and CSIRO climate models that *do not* observe reality and are simply broken.

February 22, 2016 4:32 pm

…this model limitation is studied by generating large number of simulations with variable reliability in simulated probability.

Simulated climate with variable reliability and simulated probabilities.
We’re not in Kansas anymore, Toto.

Reply to  H.R.
February 22, 2016 4:41 pm

Adjustment to raw data is necessary. Getting those right is an exercise in game theory. NOAA got it wrong, wrong, wrong. But that does not not mean that competent folks cannot get it (roughly) right.
Game theory makes it obvious what homogenization (without regard to microsite) is doing and why, and it does so even without detailed knowledge of what is going on inside the obtuse mess of the Black Box.
Game theory makes it easy to make corrections that are open, clear, understandable — and replicable by anyone.

Reply to  H.R.
February 22, 2016 4:42 pm

Aha! That first one jumped out at me so hard that I rushed by the very next sentence.

The framework illustrates that unreliable climate simulations are prone to overestimate the attributable risk to climate change.

If they are really sure about that then the models can be completely relied upon… to overestimate.

February 22, 2016 4:51 pm

To paraphrase. Implications of imperfect climate models unknown. So we investigated, and it is worse than we thought…weather extremes are not so extreme.
Well, given that climate models do no reproduce regional climates (they do not downscale), nor reproduce global average features (temps, temp anomalies [the pause]), rainfall, clouds…this result is not surprising, except to warmunists.

February 22, 2016 5:30 pm

Do they think this is a mistake?

Michael Jankowski
February 22, 2016 5:37 pm

It’s only a matter of time before the inability to accurately model extreme weather events will be blamed on the unpredictability of extreme weather events caused by global warming/climate change.

Reply to  Michael Jankowski
February 22, 2016 5:49 pm

Good point but maybe should have ended with “caused by CO2”.

Reply to  eyesonu
February 22, 2016 10:41 pm

I know this a bit of topic but this just showed up regarding the CO2 story, perhaps Dr Tisdale could check it out.They are trying their best, just the latest:

Eugene WR Gallun
Reply to  Michael Jankowski
February 22, 2016 7:08 pm

Michael Jankowski
It is only a matter of time before the inability to accurately model extreme weather events will be said to have occurred due to weather become totally chaotic — due to CO2 — and therefore IT IS WORSE THAN WE THOUGHT! Weather is now so chaotic it can no longer be modeled!!!!
Eugene WR Gallun

February 22, 2016 5:53 pm

Never made much sense.
Extreme weather is a function of intense gradients, not higher temperatures.
But attribution can be plausible and that’s enough for unscrupulous fear mongers.

February 22, 2016 6:19 pm

This site has an amazing investigative article which exposes the mainstream media’s completely unsubstantiated claim that Cyclone Winston was the ‘strongest ever’ in the southern hemisphere:
It seems that Dr Jeff Masters has been at it again.

Reply to  MP
February 22, 2016 7:17 pm

The media reports on this storm were terrible, there were some that actually mentioned “possible” gusts of up to 360 kms/ hr and so on . Look it was a bad storm but the media aren’t doing themselves any favors in this case. I hope your report get to them at some point.

Reply to  tobias smit
February 22, 2016 9:05 pm

Thanks Tobias. Everyone – please share this story with whoever you can. The media has so far been given a complete free pass on this. We should not be made to swallow this.

Reply to  MP
February 23, 2016 2:08 am

NBC News meteorologist Bill Karins warned that Patricia would be “the most devastating storm to ever hit Mexico” with “catastrophic damage” likely between the posh resort of Puerto Vallarta and the bustling port city of Manzanillo. This was the biggest flop ever and all media should be embarrassed.

Reply to  MP
February 23, 2016 6:15 am

Ok. Last nail in the coffin of my fandom of Jeff masters. Disappointing because experience shows he does such good work on tropical prediction but misreprresenting data is the last straw. Funny that he seems to understand the vagaries of the hurricane models. But his website does get hard to read with the global warming drumbeat in th background. Sigh

February 22, 2016 6:24 pm

I followed the abstract well until reaching the last sentence where I crashed and burned. The last sentence of the abstract is, “Our results suggest that event attribution approaches comprising of a single climate model would benefit from ensemble calibration in order to account for model inadequacies similarly as operational forecasting systems.”

Reply to  John Whitman
February 22, 2016 6:49 pm

Well if you got that far, could you explain an earlier sentence to me?
Using an idealized framework, this model limitation is studied by generating large number of simulations with variable reliability in simulated probability.
Anyone? Rud?
This paper is well over my head, but I cannot shake the feeling that they’ve taken models that they know are wrong (but don’t know why or by how much) and simulated a whole bunch of data using a statistical model that they cannot validate…. So a model of a model of models from which they somehow draw a conclusion that the models are wrong about this one thing.
Well we already knew that! ACE and other measures of extreme weather have been in decline for decades and we know that from actual data! So what does this paper actually tell us that is new?

Reply to  davidmhoffer
February 22, 2016 6:54 pm

So what does this paper actually tell us that is new? And is their methodology sound enough to draw a valid conclusion?

4 eyes
Reply to  davidmhoffer
February 22, 2016 10:23 pm

FWIW, I reacted to that same sentence the same as Mr Hoffer. I am shaking my head trying to work out what is being said in the abstract

Bill Parsons
Reply to  John Whitman
February 22, 2016 10:05 pm

Translation: Our apocalypse is The Official Apocalypse – Get with the program! Meanwhile…
Ah-one, ah-two, ah-one, two, three: There’s no business like show business, there’s no business I know-ow-ow…

george e. smith
Reply to  Bill Parsons
February 23, 2016 12:13 pm

Well you can tack it up on a wall, and throw darts at it, or better yet and more utile, place it in the bottom of the parrot’s cage, and let the bird drop rear end darts at it.

Greg Cavanagh
February 22, 2016 6:42 pm

“The implications of using imperfect models in this context are largely unknown…”
The money quote. They don’t know what the consequences are when you use wrong models to model stuff (especially when those models are believed).
The rest of us would say you get a wrong answer, but hey, they’re trying.

george e. smith
Reply to  Greg Cavanagh
February 23, 2016 12:17 pm

Well my response to the authors would be: “Unfortunately we cannot determine how much we should charge you out of your grant fund, for writing this Bulwer-Lytton prize winning crappola.”

February 22, 2016 7:10 pm
The Washington Postage Stamp newspaper has a HUGE front page article about our global warming warrior, Mann, howling about how we are all going to drown and the hockey stick is now the oceans, not the temperatures!
Quote: The new research also forecasts that no matter how much carbon dioxide we emit, 21st-century sea level rise will still greatly outstrip what was seen in the 1900s. Nonetheless, choices made today could have a big impact. For a low emissions scenario, it finds that seas might only rise between 24 and 61 centimeters. In contrast, for a high emissions scenario — one that the recent Paris climate accord pledged the world to avert — they could rise as much as 52 to 131 centimeters, or, at the very high end, 4.29 feet.
However, Kopp notes that the methods used to project these totals may not fully capture what happens over the course of this century. “We have a model that’s calibrated against a period when a certain set of processes, largely thermal expansion and glaciers, were dominant,” he says, “and we’re looking forward to a period when other factors will be dominant.”…The new study follows in the footsteps of a 2011 study that looked at the ocean and climate records contained in salt marshes in North Carolina to infer the history of sea level rise over the past 2,100 years — research that had many of the same authors. That study, too, found that the recent sea level rise is unprecedented over that time period.
Mann of Penn State, one originator of the “hockey stick” reference and a co-author on the 2011 study (but not the current one), said by email that he thinks the current work is an “incremental advance” on that prior study, albeit one he agreed with in broad outline…end quote.
They are still blasting away with the mainstream media howling along with them.

Reply to  emsnews
February 22, 2016 7:14 pm

Notice that this latest news hysteria is all about models and the cute crew doing this are vague about their timeline and sort of talk about the last ‘thousand years’ but then jump to saying it is warmer than the Roman Warm Period right now! Crafty guys, they know perfectly well it is no where near as warm as back then. These guys never give an inch, they have this hockey stick and are high sticking us and no one is benching them.

Reply to  emsnews
February 23, 2016 4:28 am

They refuse to read the vaulable science because they are not peer-reviewed. In a devastating critique of peer-review, he said it had become “a way to maintain the status quo”. Noakes presented significant scientific evidence for issues central to this case, that aim straight at the heart of conventional medical and dietary “wisdom” and the status quo, and threaten careers, reputations, livelihoods and funding. (It is, after all, “difficult to get a man to understand something, when his salary depends on his not understanding it”, as US author Uptom Sinclair once said

February 22, 2016 7:11 pm

Well, now that Australia noticed the science is settled…

February 22, 2016 7:39 pm

‘The Distribution of Climate Change Public Opinion in Canada’, Feb.2016
Expected to be published after peer review. Abstract and paper at:
Poll shows high level of support for climate change and cap-and-trade in Canada.
Partly funded by:
Skoll Global Threats Fund, U.S.
Energy Foundation, U.S.
Grantham Foundation, U.S.
Authors are from, U.Montreal, Yale U., U.Calif.SB, U.Utah.
Jeffrey Skoll, OC, films include ‘An Inconvenient Truth’ with Al Gore
Foundation money coming from the U.S. to Canada to support the Climate Change agenda in Canada.

Reply to  Barbara
February 23, 2016 4:49 am

Barbara The U of M was unrestrained, extravagant, on all radio stations in Montreal yesterday on this study. You must buy in too hell with the truth.

Reply to  Russell
February 23, 2016 6:20 pm

Support for the above poll study:
Trottier Energy Institute
Sustainable Prosperity Canada
Canada 2020
Public Policy Forum
Bet who supported this poll paper weren’t mentioned?
Maybe this paper should be looked at? But Canadians have been fed so much climate scare this might be true.

Reply to  Russell
February 24, 2016 9:22 am

Bet who supported this poll paper weren’t mentioned?

Hi Barbara,
Yeah, the financiers are mentioned at the end of the first page of the paper.

Thanks to Christopher Borick, Barry Rabe, and Anthony Leiserowitz for their support of this project, and the collection of its underlying data. Thanks to the Social Sciences and Humanities Research Council of Canada, the Fonds de Recherche du Québec – Société et Culture, the Skoll Global Threats Fund, the Energy Foundation, and the Grantham Foundation for the Protection of the Environment for financial support. Funding for individual survey waves was provided by the Ministère des Relations internationales et de la Francophonie, the Public Policy Forum, Sustainable Prosperity, Canada 2020, l’Institut de l’énergie Trottier and la Chaire d’études politiques et économiques américaines.

It’s purely a propaganda puff piece, of course.
Perhaps it’s leveraged off this April 2013 Skoll Discussion Piece at Yale?
Because . . advocacy.

Reply to  Russell
February 24, 2016 6:52 pm

Maybe from Canada 2020 too?
The “bet” is about what the general public knows about the parties that supported this poll study.

Reply to  Russell
February 24, 2016 7:19 pm

Yale Program, Climate Change Communication
People & Partners
A team of psychologists, geographers, political scientists, statisticians ,pollsters, and etc who investigate why U.S. citizens and those around the world are not responding to climate change.
Archived so wait for page to come up.

Reply to  Russell
February 24, 2016 8:10 pm

Canada 2020
‘2013 Canada-US Comparative Climate Opinion Survey’, March 3, 2014
Erick Lachaplle
Christopher Borick
Barry G. Rabe

February 22, 2016 8:04 pm

At some point, that time long ago for many including myself, the debate over why mainstream climate science is left-wing political drivel was concluded. See the Dr. Ball’s outline just down thread for a good summary of motives.
I’m somewhat amused the link to Dr. Curry, she only became politically self aware of the climate agenda political motives, according to her story, the past decade. Dismally late with little contrition, still not ready to close the casino because she found gambling going on. Pretty representative of mainstream skeptics which is the other half of 50 plus years of Greenshirt expansion.
The discussion should be mostly about “why” climate science is so knowingly corrupt rather then “how” if you’re hoping to actually see a reversal. This is where a large section of skeptics fall away and how the monster grows without actual science or reason behind it.

February 22, 2016 8:23 pm
Three time series of average summer daily maximum temperature (TMax JJA) are developed for three interior regions of Alabama (AL) from stations with varying periods-of-record and unknown inhomogeneities. The time frame is 1883-2014. … Varying the parameters of the construction methodology creates 333 time series with a central trend-value based on the largest group of stations of -0.07 °C decade-1 with a best-guess estimate of measurement uncertainty being -0.12 to -0.02 °C decade-1. This best-guess result is insignificantly different (0.01 C decade-1) from a similar regional calculation using NOAA nClimDiv data beginning in 1895. … Finally, 77 CMIP-5 climate model runs are examined for Alabama and indicate no skill at replicating long-term temperature and precipitation changes since 1895.

Reply to  kennethrichards
February 22, 2016 8:26 pm

Alabama cooling at a rate of -0.07 °C decade since the 1890s. 77 climate models “indicate “no skill” at replicating this long-term cooling.

February 22, 2016 9:19 pm

Meanwhile the southern hemisphere just had its most powerful cyclone (Winston) to make landfall in recorded history and the second most powerful on the planet after Haiyan. In fact 6 of the top 11 have happened this century.

Reply to  Simon
February 22, 2016 10:59 pm

That one has been completely and utterly debunked:

Reply to  Simon
February 23, 2016 2:28 am
Bryan A
Reply to  Simon
February 23, 2016 12:34 pm

Amazing how each of the bottom six have exactly the same sustained wind speeds

Reply to  Bryan A
February 23, 2016 5:55 pm

It’s really disappointing that engineered tables like this get traction with the mainstream media – to the point where we can’t have a proper discussion of ideas of how we can better treat the planet. Instead, we get to talk about the carbon boogie man and line the UN’s pockets with billions of dollars:

February 22, 2016 9:35 pm

Isn’t the news here the tacit admission that “the models are known to be imperfect”? That’s something the IPCC would not want to admit. They are not talking about “imperfect” in the sense of zero error range, but rather imperfect in the sense of being unreliable to use for realistic analysis and leading to exaggerated claims, as they seem to demonstrate. That’s what the deniers have been claiming all along, right?
If I was a newspaper editor again, my headline here would be:
Climate Models Unreliable, Lead to Exaggerated Claims, Scientists Say

February 23, 2016 12:43 am

Look it is like when you multiply negative numbers you get a positive.
With climate models you just multiply all the wrong models against each other and you get the right answer.
Ask any climate scientist, It is just math :-!

February 23, 2016 2:24 am

Alarmists frequently claim we have loaded the dice in favour of extreme climate events. Turns out that the climate models are loaded dice. Nobody will be shocked by this news.

Bryan A
Reply to  Charlie
February 23, 2016 12:38 pm

I am Shocked that Climate Model Dice are Loaded

February 23, 2016 5:24 am

100.000% of CAGW modelS have been WRONG by many standard deviations for many decades for all severe weather events: hurricanes, typhoons, cyclones, droughts, floods, thunderstorms, tropical storms, sub-tropical storms, tornadoes, precipitation, etc..
And yet Leftist CAGW politicians and grant grubbers insist severe weather frequency and intensity is getting worse and worse, despite the empirical evidence showing servers weather trends are flat or falling….
Why is this Leftist scam still a thing???
Eventually, scientists outside of climatology will have to blow the whistle on this CAGW scam.

Smart Rock
February 23, 2016 9:47 am

Based on reading the abstract (why bother reading the full paper, my mind is made up anyway), these guys seem to be saying that “event attribution” based on a single model is inappropriate.
Who needs a model to attribute an event?
If “event attribution” means saying in any kind of public forum that “Storm XYZ was the worst in the last 217 years and is clear proof that climate change is real and blah blah blah….” – that doesn’t need a model at all. It just requires bold assertion in the absence of any kind of fact-checking or even a knowledge of weather history.

Gunga Din
February 23, 2016 2:12 pm

Groundhog Day was February 2nd. I read somewhere that Punxsutawney Phil’s rather vague predictions of “6 more weeks of winter” is accurate about 40% of the time.
He has a better track record than the Climate Models.

February 23, 2016 7:21 pm

The climate models are proven to be grossly wrong, over estimating, so simply change the temperature data . TaDa problem solved , even NOAA could figure that one out . You just have to wonder who were the visionaries that place weather stations on asphalt beside air conditioning rejected heat ? What gene pool would think yeah that’s the place ? Probably the same crowd who burned people on the logic that they might be witches . You never know let’s be safe .

February 27, 2016 9:42 am

The same crew which have a habit of misplacing surface temperature stations only near or in excessive hear sources and never in a cooler, stream or freezer, are now revealed displaying similar tricks biasing their computer programs with the same result. How surprising.

%d bloggers like this: