The only thing we can be sure of about the future is that it will be absolutely fantastic. So if what I say now seems to you to be very reasonable, then I have failed completely. Only if what I tell you appears absolutely unbelievable, have we any chance of visualizing the future as it really will happen.

- Arthur C. Clarke (h/t Brin)

Wednesday, September 15, 2010

Duck Curry

Judith Curry from the Currysphere:
"evidence for a hypothesis is represented as green, evidence against is represented as red, and the white area reflecting uncommitted belief that can be associated with uncertainty in evidence or unknowns....

As a litmus test question, I prefer the following:

Will the climate of the 21st century will [sic] be dominated by anthropogenic warming (green) or natural variability (solar, volcanoes, natural internal oscillations)?

which is the question with the greatest policy relevance, IMO. My scores on this one are

  • green 25%
  • white 50%,
  • red 25%.
Hmmm. Never mind the tiny sensitivity of climate implied. Just consider the clarity of thought involved here.
Do you think this creature is a duck?

Duck: 25%
Not duck: 25%
Unsure: 50%

I am 50% uncertain whether it is a duck, and I am equally 25% certain that it is a duck and that it isn't a duck.
Well, that clears it up, thanks.

21 comments:

Steve Scolnik said...

Just the assessment you might expect from someone with policy assessments for sale to the widest possible market. Follow the money:
Climate Forecast Applications Network

rab said...

Ah. Much clearer now. Thanks, Steve.

bluegrue said...

But, but, but, you are misunderestimating the subtlety of that flaginism. She's saying that anthropognenic warming will account for 25% to 75% of the climate in the 21st century and natural variabilty will take of the rest. Or something like that.

It still leaves the question, what the poor Italians have done to Curry to deserve this kind of treatment.

Michael Tobis said...

No, BG, I wish she had said that, but she said that she is 25% sure that anthropogenic forcing will be more important and 25% sure that natural variability will be more important and 50% unsure.

If she had said what you say she said, it would have the advantage over what she actually said that it would be almost surely wrong, and one could argue with it. As it stands I fear that it is in the "not even wrong" category.

Michael Tobis said...

That Curry has a forecasting business with Webster bothers me not in the least. I don't understand what is supposed to be wrong with that.

It does put her in the old school climatology camp with Gray and Knappenberg and that bunch, which may explain a lot.

I may need to explain this culture gap some day.

Steve Bloom said...

Well, CFAN would give Judy a reason to want to increase her public profile, and to the extent that the reactionary GTech milieu has any relationship to their client list it helps our understanding of what she's about. It's interesting to see that Peter isn't touching any of this, at least not publically.

Michael, do you really mean to say that in and of itself having a forecasting business puts them in the same camp as the parties you mention? Peter is pointedly not in that camp.

Michael Tobis said...

Not necessarily in the ideological camp.

But the traditional techniques of long term forecasting are neither traditionally meteorological nor traditionally climatological (both being more rigorous).

They are heuristic and statistical, having something of the flavor of the charts and graphs day traders mess around with. More old school, in the tradition of Lamb and Bryson. Bryson was a delight and I expect Lamb was too, but they were no more mathematical than a smart tenth grader.

What's more, if climate change really takes off, the heuristics will break. So they have something at stake in not believing in severe climate change.

Do you know Webster?

Arthur said...

I don't understand what the flag is supposed to mean either. It's introduced this way, via the AR4 canonical statement:

" * Most of the observed increase in global average temperatures since the mid-20th century is very likely due to the observed increase in anthropogenic greenhouse gas concentrations.

This statement is often used as a litmus test for belief regarding global warming, i.e. you believe this statement (consensus) or you don’t (skeptic). Very likely denotes a probability of anthropogenic influence between 90 and 99% (lets pick 95%) and I interpret most to mean between 51 and 90% (lets pick 70%), with the remainder (30%) associated with natural variability. Hence, the Italian flag analysis could represent this in the following way:

* 5% assigned to uncommitted belief (white),
* 67% assigned to anthropogenic forcing (green),
* 28% assigned to natural variability (red).
"

Is the 67% supposed to come from 70% multiplied by 95%, and the 28% from 30% times 95%?

But the 5% uncertainty vs the other two are quantified on completely different dimensions, I don't see how you can justify conflating them like that!

For example, what does the 28% actually mean? Suppose 30% of the temperature trend can be ascribed to non-human causes, with 95% confidence. How does that become 28% (or 28.5%) of something else? I don't get it. Anybody else?

bluegrue said...

I've just suggested the creation of an analogon to the geek code. I hope I havn't created a monster.
http://judithcurry.com/2010/09/15/doubt/#comment-836

I think it will be interesting to see some of the codes, if the idea gets picked up.

Pangolin said...

Pardon me. Am I supposed to understand that the numbers given in the OP apply to the mindset of one person?

Or is this some sort of magical-thinking, risk-assessment model that has numbers plugged into it to justify a classic WAG.

Help the layman out here.

William T said...

"Will the climate of the 21st century will [sic] be dominated by anthropogenic warming (green) or natural variability (solar, volcanoes, natural internal oscillations)?"

Well, presumably climate fluctuations over 2-5 (or 10?) years will be dominated by the occurrence or not of those natural events. The trend over 100 years will be almost certainly be dominated by whatever CO2e forcing humans impose. So what does "dominated" mean in this question?

Pangolin said...

"So what does "dominated" mean in this question?"-William T.

That brings up another question. Obviously the climate of Pakistan was "dominated" on a numerical basis by natural variations. However that one month of extreme weather that appears to have been the result of climate forcing pretty much destroyed the last two decades of infrastructure improvements in Pakistan.

It's not the "dominant" pattern that we should worry about. It's the possibility of increases in outliers. How do we forecast/account for those in our calculations?

David B. Benson said...

Pressed duck.

Squeeze the fat out first before making

duck curry.

(Or was this comment supposed to go on a cooking blog? Isn't this one? I"m 50% unsure.)

Daniel said...

Thought for sure someone would've posted this one already: Dr. Boli and the Duck.

Situation rectified.

The Yooper

Steve Bloom said...

Michael, I'm somewhat familar with his work, and have paid close attention to the comments he's made at RC over the years, but otherwise I don't know him. I think you're correct about the generally statistical approach, but note that he has a number of papers featuring modeling work (albeit all with co-authors such that it's hard to know who did what). Re the heuristics, in sharp contrast to Judy he seems to have an expectation that they will change.

Also, I notice that he currently heads the AGU atmo sciences section, which I assume must mean he has fairly wide respect among more than just scientists in the tradition you describe.

I'll probably continue thinking positive thoughts about him so long as he continues to offer no public support for Judy's excesses.

Aaron said...

The heuristics are broken! What snowfall do you plan for in Washington, D.C.? What flood crest do you plan for on the Mississippi? (The 5 highest in history have all been within the last decade!)

People just have not admitted that as the climate has warmed, the weather has changed. As the climate keeps warming, the weather will keep changing.

OK, it is a noisy signal. Noisy signals are why we invented statistics. Time to get out the full statistics tool box, and use the right tools for the job.

rustneversleeps said...

I see the "mt's current thinking on ice and SLR" thread is closed for comments. So, I'll just use this unrelated thread to bookmark a low-keyed little 1/2 hour video gem that may not get much notice: Dick Peltier, U of Toronto. Ice Sheets - The Polar Canaries. Aug. 14, 2010.

This was part of 3-day weekend symposium held recently at UofT. Other presenters included Andrew Weaver, Bill McKibben, Jim Prall and many others.

Hank Roberts said...

http://people-press.org/report/?pageid=1549
"A large majority of the public (76%) and nearly all scientists (97%) say that it is appropriate for scientists to become actively involved in political debates on controversial issues...."

hat tip to:
http://bigthink.com/ideas/24012
(cough*Nisbet and Pielke ...)

Hank Roberts said...

> http://www.cfanclimate.com/
amazingly heavy on the Curry

Lou Grinzo said...

Wait a minute -- CFAN was founded in 2006, according to their site, and people are just now discovering Curry's connection to it?

I hate to say it, but I have the strong suspicion that everyone on the reality-enhanced side of this discussion (call it the Truth Tribe, if you must) has been played like a fiddle as part of nothing more or less than a marketing push.

Michael Tobis said...

What's the problem with CFAN? It just looks like a bucket for small consulting gigs.