Since failures in climate science claims are on the rise, can we start naming climate prediction failures after scientists and activists? I can think of a few: The Hansen Hiatus, for example.
Climate campaigners seem to think they have a winner with this takedown of elected officials who reject global warming science, in which fake news reports talk of the turmoil and tragedy created by Hurricane Marco Rubio, Hurricane James Inhofe, Hurricane John Boehner and more.
The trouble is, the science on a connection between hurricanes and global warming is going in the opposite direction, if the near-final draft of next month’s climate science assessment from the Intergovernmental Panel on Climate Change is any indication.
Andrew Revkin at NYT’s DOT Earth
=============================================================
Twitter / RyanMaue: El Reno tornado reclassified …
El Reno tornado reclassified from EF-5 to EF-3 highlighting some debate at the time about using radar wind data http://www.your4state.com/story/national-weather-service-re-classifies-el-reno-ok-tornado-to-ef-3/d/story/9i6cjSZVfUWnisUjHByPSg …
See also this WUWT story bringing the “widest tornado” claim into question.
==============================================================
What? Polar bears are out on that disappearing sea ice in late summer?
To put all this into perspective, note that research in this region between 2000 and 2005 determined that, on average, only 3.7% of all Southern Beaufort polar bears spent time on land between mid-September and the end of October (Schliebe et al. 2008). As the estimated total population at that time was 1,526 bears (and still is), it means that on average, only about 56 bears spent time on land each summer in the early 2000s.
See Susan Crockford’s blog: Ten out of ten polar bears being tracked this summer in the Beaufort Sea are on the ice
===============================================================
What? Algae don’t have enough atmospheric CO2 so they have to make their own?
A paper published today in Nature finds that marine algae, which evolved and thrived with atmospheric CO2 levels 15 times higher than the present, required a novel adaptation to adjust to the relatively low CO2 levels during the Cenozoic era, when CO2 levels were still more than twice current levels. According to the paper, this novel adaptation was to manufacture their own CO2 at the reaction site for photosynthesis, required due to a paucity of CO2 in the atmosphere. Algae evolved more than 500 million years ago, when CO2 levels were ~15-17 times higher than the present; current CO2 levels are near the lowest levels of the past 500 million years.
Algae evolved more than 500 million years ago, when CO2 levels were ~15-17 times higher than the present.
http://hockeyschtick.blogspot.com/2013/08/new-paper-finds-algae-have-to.html
==============================================================
Rowers give up. Another “save the planet from global warming” Arctic trek gets a reality check from Nature:
Severe weather conditions hindered our early progress and now ice chokes the passage ahead.
Our ice router Victor has been very clear in what lies ahead. He writes, “Just to give you the danger of ice situation at the eastern Arctic, Eef Willems of “Tooluka” (NED) pulled out of the game and returning to Greenland. At many Eastern places of NWP locals have not seen this type ice conditions. Residents of Resolute say 20 years have not seen anything like. Its, ice, ice and more ice. Larsen, Peel, Bellot, Regent and Barrow Strait are all choked. That is the only route to East. Already West Lancaster received -2C temperature expecting -7C on Tuesday with the snow.”
Richard Weber, my teammate to the South Pole in 2009 and without doubt the most accomplished polar skier alive today, is owner and operator of Arctic Watch on Cunningham Inlet at the northern end of Somerset Island. Arctic Watch faces out onto our proposed eastern route. Richard dropped me a note the other day advising: “This has been the coldest season with the most ice since we started Arctic Watch in 2000. Almost no whales. The NWPassage is still blocked with ice. Some of the bays still have not melted!”
…we’d require at least another 50-60 days to make it to Pond Inlet. Throw in the issues of less light, colder temperatures, harsher fall storms and lots of ice blocking the route and our decision is easy.
…Our message remains unaffected though, bringing awareness to the pressing issues of climate change in the arctic.
Despite that admission of failure due to so much ice, readers are hard pressed to find many pictures of it in their photo stream: http://www.flickr.com/photos/95019072@N08/
I suppose showing pictures of ice is counter-productive to their mission, since like many fools before them, they expected the Arctic to be mostly ice free due to global warming. – Anthony
=============================================================
Uh, oh
New paper finds cloud assumptions in climate models could be incorrect by factor of 2
More problems for climate models: A new paper published in the Journal of Geophysical Research-Atmospheres finds that models must take into account not only the presence or absence of clouds but also how clouds are stacked vertically. The authors find that changes in vertical stacking of clouds can change radiative forcing assumptions by a factor of two [100%]. However, state of the art climate models do not take vertical stacking into consideration, and most global datasets of cloudiness also do not contain this information. “Clouds, which can absorb or reflect incoming radiation and affect the amount of radiation escaping from Earth’s atmosphere, remain the greatest source of uncertainty in global climate modeling,” and according to this paper, that uncertainty has just doubled from what was previously thought.
See: THE HOCKEY SCHTICK
Why the forthcoming UN IPCC Report is already toast
The IPCC is set to release its latest Assessment Report 5 [AR5] in about 1 month, yet the report will be dead on arrival and hopelessly out-of-date in light of recent inconvenient peer-reviewed papers published after the cut-off date for inclusion, as well as papers published before the cut-off date which the UN continues to ignore. Since almost the entire report hinges on the output of climate models, and those models have recently been falsified at a confidence level of >98% over the past 15 years, and falsified at a confidence level of 90% over the past 20 years, the entire report and its Summary for Policymakers are already invalidated even before publication.
http://hockeyschtick.blogspot.com/2013/08/why-forthcoming-un-ipcc-report-is.html
rgb, I just wish you would stop shilly-shallying and tell us what you really think 🙂
Seriously, a masterly demolition with lots of educational highlights for non-scientists like me. I suspect that Oldberg’s feint is the first of many to try to explain why the CAGW models have been so spectacularly wrong. I have saved your posts in case somebody tries to run this line again elsewhere.
Thanks.
This is because he makes irrational assertions based on
(a) his own definition of terms
and
(b) refusal to provide explanation of the terms he uses.
He often claims he can “expand” or “explain” what he says but – when pressed – he never does.
Time honored ways of trying to win an argument, at least in your own mind, of course. I have nothing against the idea of disambiguating terms as this can indeed be a major problem in any philosophical or logical debate, and it is closely tied to e.g. shifting sands fallacies and so on that often underlie it (inherent in the refusal to provide explanation and example of the new “definitions” inserted). Hence my request for a concrete example, ideally in context, for what he has in mind. They’d better not be GCMs, of course, because those are painfully obviously ordinary predictive models and a perfect match for NON-ambiguous definition 1 from the wordnet list of definitions of models. They embody many hypotheses (so that the hypothesis itself is quite complicated) all of them ultimately based on observation and data because all of science including the laws of physics are based on observation and data in a consistent Bayesian network or empirically supported reasoning.
GCMs in particular are further fueled by the use of empirically set, tunable parameters and are more or less without exception individually validated on a training set of climate data used to tune these parameters. When released into the wild, they promptly have diverged from the future, and were never ever able to hindcast the past. Their use has openly been to predict the dire climate future, and has invariably been accompanied by spurious, erroneous, and deliberately misleading claims of the “probability” of that future in quite quantitative terms, made in precisely the context where they would have the maximum political impact and be safely shielded from any sort of scientific or statistical criticism.
If there is something else that he wants to consider a “modele”, all he has to do is — without obscenity, ad hominem, or shrillness — present one or more clear examples and indicate how they cannot fit into the list of accepted definitions of the term model and how they are relevant to the discussion of the general circulation models used as the fundamental statistical predictive elements of the politicized side of climate science. Or, perhaps, his point is that they should not be so used because one cannot make a specific Bayesian argument with full knowledge of their priors, but that is a common feature of nearly all predictive models and doesn’t stop them from being models built on top of a collective hypothesis, it just means that when they fail a hypothesis test it is reasonable to reject that hypothesis and try try again. More generally, one should weaken the conclusion of a complex argument to the extent that the prior probabilities are unknown or poorly known, which is why multivariate predictive modeling is a goddamn hard game to play — it doesn’t take a lot to weaken a complex argument to damn near a coin flip, and it means that a wise person will want a lot of weak verification before he trusts his life or fortune to its predictions. Taleb’s The Black Swan presents a lovely argument for why in his polemic against the abuse of the normal, against ceteris paribus, against egregious assertion of Bayesian priors that are more properly unknown and sometimes unknowable.
rgb
rgbatduke:
Thank you for taking the time to respond.
Like many words in the English vernacular, “model” is polysemic. That it possesses more than one meaning causes trouble of a logical kind when: a) this word is used in making an argument and b) the word changes meaning in the midst of this argument. That it changes meaning makes of this argument an “equivocation.” Upon superficial examination, an equivocation looks like a syllogism. However, while the conclusion of a syllogism is true, the conclusion of an equivocation is false or unproved. That it is not true is, however, concealed by the appearance that the argument is of the form of a syllogism.
The following example of an equivocation may be instructive:
Major premise: A plane is a carpenter’s tool.
Minor premise: A Boeing 737 is a plane.
Conclusion: A Boeing 737 is a carpenter’s tool.
The false conclusion that a Boeing 737 is a carpenter’s tool is produced when the polysemic term “plane” changes meaning in the midst of the argument.
By rule of logic, a proper conclusion cannot be drawn from an equivocation. To draw an IMPROPER conclusion from an equivocation is the “equivocation fallacy.”
An equivocation fallacy cannot exist absent polysemic terms in the associated argument. Hence, the prospect for reaching a valid conclusion from an argument is enhanced when the language of this argument is disambiguated such that all terms are monosemic. While disambiguation is wholly beneficial to the end of reaching valid conclusions from global warming arguments, there are people who resist disambiguation of the terms of global warming arguments. Among these people have been yourself and richardscourtney. I’d like to change your minds.
Logic is the field that contains what is known about the rules by which correct inferences may be distinguished from incorrect ones. That this is so lends logical significance to the distinction between a model that makes a predictive inference and a model that does not make one. A model of the former type is susceptible to being validated (in the disambiguated sense of this word), makes predictions (in the disambiguated sense of this word) and conveys information to a policy maker about the outcomes from his or her policy decisions. A model of the latter type is insusceptible to being validated, makes no predictions and conveys no information to a policy maker. A model of the former type is suitable for use in making policy decisions on greenhouse gas emissions while a model of the latter type is unsuitable for this use. The climate models of AR4 are entirely of the latter type.
Global warming research has produced no climate models that are suitable for policy making and thus has failed in its intended purpose. Nonetheless, policy makers have used the climate models that HAVE been produced by this research in making policy. That this is so has been obscured through the mechanism of drawing improper conclusions from equivocations. In the article at http://wmbriggs.com/blog/?p=7923 I catch the climatologist Gavin Schmidt in the act of doing so.
These fallacies are based, in part, upon the polysemic term “model.” They are additionally based upon the polysemic terms science, scientist, scientific, predict/project, prediction/projection, validate/evaluate and validation/evaluation ( ibid ). The latter four terms are word pairs wherein the words in each word-pair have differing meanings but the two words are used as synonyms in making the associated argument.
Terry Oldberg:
Your long-winded and mostly meaningless but entirely pointless post at September 3, 2013 at 8:46 am includes an unjustified assertion; viz.
the climate models cited in the AR4 are of an undefined type which does not make predictions.
Clearly, they do make predictions. And, clearly, the AR4 states those predictions.
Unfortunately, you failed to mention why you think the models of the AR4 do not make predictions.
Perhaps you would be willing to correct that oversight so rgb has something in your post which can be addressed?
Richard
richardscourtney:
Like many of your arguments the one of Sept. 3 at 9:12 am is an example of an equivocation; in this case, the polysemic term that changes meaning is “predict.” If your aim were to ensure that people were not mislead by your arguments you could accomplish this aim through replacement of polysemic with monosemic terms. This, however, you steadfastly refuse to do. That you refuse makes it unprofitable to attempt logical discourse with you. You consistently thwart logical discourse through applications of the equivocation fallacy.
Terry Oldberg:
Your post at September 4, 2013 at 9:10 am purports to be an answer to my post at September 3, 2013 at 9:12 am.
My post asked you a question which you seem to have missed. Just to be clear, it was this
Please be so kind as to answer the question.
Insults to me are not an answer to the question.
Saying I use the word “predict” is not an answer to the question.
Saying why you think the models of the AR4 do not make predictions would be an answer to the question.
Richard
rgbatduke:
I am trapped in Alice’s rabbit hole again.
I would be grateful for a rescue attempt.
My pea shooter cannot get me out but one of your artillery barrages may blast the hole apart.
Richard
The only reply I have time to make is that the metaphor (rather than reply) given about concerning disambiguation of the word “plane”, in context, would have us reasoning about the kind of clothes GCMs wear as they parade down runways.
In other words, only a complete idiot could look at the content, context, and purpose of general circulation computational predictive models and then confuse them with any meaning of the term “model” but meaning 1 in my list above, which does not under any circumstances require disambiguation either logically or semantically. There is a name for this particular logical fallacy — could it be “straw man”? Inventing an issue out of whole cloth, writing entire articles on the subject intended to provide apologia to something that isn’t a model/modele issue but is just poorly done science, improperly used by scientists to influence political issues under the guise of settled science when in fact it not only isn’t settled, it fails basic hypothesis tests applied to the GCMs one at a time as hypothesis based predictive models.
There is no need to kick the GCMs by calling them modeles (a term that no one will understand because you just made it up, Terry) — their failure as mere models is quite straightforward and unambiguous. And no, I’m not calling you an idiot — I’m saying that both your example and mine illustrate that there isn’t the slightest shred of contextual ambiguity here. If you want to work on disambiguating various terms used in religious discussion, such as whether or not atheism is a theistic belief, or the term Universe (meaning everything that has objectively real existence) does or does not (by definition) include God and hence logically precludes any possibility of the Universe being created by a dualistic God from a manifestly contradictory complement to the set of everything that actually exists, I’m all over that — endless, harmless fun in an issue that has no real empirical basis and hence is fundamentally undecidable at least with the data in hand. Trying to play this game in predictive modeling is simply unnecessary, confusing, and a bit silly.
Beyond that, Richard really doesn’t need my help, but we’d all appreciate you making your point. Are general circulation models models in the commonly accepted sense of the term in this context or not? If not, be well-prepared to defend your assertion otherwise, on the basis not of your made-up term “modele” which we have not yet agreed is logically or semantically necessary. If it is not necessary, adding it is unacceptable and confusing and violates several precepts of efficient argumentation. Note that there isn’t any doubt what the creators of those models think that they are, what AR1-4 assert that they are, what AR5 is ready to assert what they are, what the scientists who wrote the summaries for policy makers claim that they are, and what they are used for, without exception, worldwide.
They are failed predictive models, but I would argue that there is no doubt that they are predictive models in design, intent, purpose, application, and all subsequent use. Where is the ambiguity?
rgb
rgbatduke:
re your post at September 4, 2013 at 1:52 pm.
Thankyou that is precisely the kind of barrage I requested to blast the ‘rabbit hole’ apart. And it included a broadside that has sunk the good ship Oldberg beneath the waves.
Thankyou.
Richard
rgbatduke:
A phrase that you use in your latest post, “predictive model,” is notable for employing two polysemic terms. They are “predictive” and “model.” If you wished, you could assist the cause of rationality in debates over global warming by switching to the use of monosemic terms. For you to serve this cause, it is not necessary for you to use the specific words (including modele) that I use in my papers. You could, for example, use made up words. The essential aspect of a disambiguation is not the words that are chosen but rather that the chosen words are monosemic in their description of the methodology of the research. I’m flabbergasted at your resistance to joining me in this endeavor. Does this mean that you prefer irrationality?
A phrase that you use in your latest post, “predictive model,” is notable for employing two polysemic terms. They are “predictive” and “model.” If you wished, you could assist the cause of rationality in debates over global warming by switching to the use of monosemic terms. For you to serve this cause, it is not necessary for you to use the specific words (including modele) that I use in my papers. You could, for example, use made up words. The essential aspect of a disambiguation is not the words that are chosen but rather that the chosen words are monosemic in their description of the methodology of the research. I’m flabbergasted at your resistance to joining me in this endeavor. Does this mean that you prefer irrationality?
Hey dude, some of us have heard of these really nifty semantic devices called “adjectives”. We learned of them when we needed to disambiguate things like red sweaters from green sweaters where we didn’t want to invent terms like “sweaters” for red sweaters only and “swatters” for green sweaters only and “swutters” for purple sweaters only. It allowed us to make an entirely desirable compression of language wherein class modifiers and context are used to provide fine tuning — when necessary — in human reason and communication. After all, there are times that Mom told us “be sure to wear a sweater” because — and pay attention, because this is verbal reasoning — it was cold outside. She didn’t much care if it was red or green or purple. Well, truth be told she always hated the purple sweater but the point is that the color of the sweater was irrelevant to the purpose of her communication, and if we disambiguated the general class of sweaters with distinct words, we would actually hinder the use of logic and reason when it applies appropriately to the whole class. We’d also fill several more OED-sized dictionaries, make it far more difficult to learn language at all, utterly destroy poetry and puns and the richness of human communication that is possible in part because one word often stands for several things, allowing terms to carry denotation and well as connotation.
None of which matters in this context! There isn’t a single person properly involved in the climate debate who does not know what a predictive model is, and that GCMs are predictive models. They are physics-based mathematical computational models used to make predictions. About the future. All of that is what everybody in any scientific or mathematical or statistical or economic field would properly recognize from the term “predictive model”. There is no need whatsoever for disambiguation, and if there were, it would surely be better accomplished through the use of adjectives that make the descriptors sufficiently precise to handle arbitrary degrees of subclassification while still allowing discussions, logic, reason, communication at the class level.
Dude, you might learn about tree structures and perhaps a bit about semantics before you go trumpeting your knowledge of “logic” and seeking to use it to add completely unnecessary words to the English language. Yes, there are discussions in science and philosophy that happen because people use the same term in different ways and talk to cross purposes as a result. There are more efficient (and, might I say less arrogant) ways to resolve the issue than inventing words and browbeating people who are perfectly happy with the humble adjectival modifier with them.
It is said that Eskimos have a zillion words for snow, and back when I studied formal logic and the philosophy of language and read Malinoski, Dorothy Lee, Sapir, Whorf, and all of those guys I actually wrote a paper that covered stuff like the fact that Trobriand islanders have 30-odd words for “yam”, because their language evolved to call every stage of the yam plant and tuber’s development by a separate word — noun rich, adjective poor. Yams were (obviously) really important in their culture, and equally obviously their language was incredibly poor. There’s a reason English has become the de facto global language instead of a rationally designed language like Esperanto. You might think on that and try to puzzle out why.
Don’t make me disambiguate “people who are addicted to pointlessly screwing around with language” by inventing a word like “epistemoholic” to apply to you.
Terms like that are catchy. They could go viral. Unlike “modele”, not before the heat death of this Universe.
rgb
rgbatduke:
In your latest post, you set up a strawman and knock it down. I have nowhere stated that adjectives cannot be used in disamguating the terms of an argument. The fact is that you steadfastly refuse to use adjectives or any other means of disambiguation in making arguments about global warming. A consequence is for you to repeatedly draw conclusions from equivocations in violation of a logical principle.
In your latest post, you set up a strawman and knock it down. I have nowhere stated that adjectives cannot be used in disamguating the terms of an argument. The fact is that you steadfastly refuse to use adjectives or any other means of disambiguation in making arguments about global warming. A consequence is for you to repeatedly draw conclusions from equivocations in violation of a logical principle.
No, specifically, you stated and objection to using “polysemic terms”, with the clear implication that in the specific case of “predictive modeling” there was something that needed to be disambiguated. Your entire article — which I would count as your statements even if they occurred elsewhere — proposes the term “modele” to disambiguate things that can be and easily are handled with adjectives, and that nobody seriously misunderstands anyway. The invention of new words to (arguably pointlessly) replace terms that are perfectly clear in context and indeed are self-defining — a GCM cannot possibly be mistaken because it is precisely defined by its code and its actual application in human affairs — suggests an unhealthy degree of obsession with nuances of language that are not, in the end fruitful of any greater understanding or insight.
I am very interested, of course, in hearing precise instances of where I have “refused to use adjectives or other means of disambiguation in making arguments about global warming”. Oh, wait, you said (quite unambiguously) “steadfastly”, so I guess I never do anything else.
Clearly I need to go back to my dictionary, as I’ve apparently been nouning too many adjectives and didn’t even realize it, in all my discussions of global warming.
Oh, wait. Warming is a noun. Global must be a disambiguating adjective! Sometimes I even through in the term “anthropogenic”! Or both “catastrophic” and “anthropogenic” at the same time (how daring of me).
So pull the other one.
rgb
Terry Oldenberg, what puffery. You mistake such “speak” for elegant argument. Divining the splitting of the width of a gnat’s a** hair is ill-elegant logic and invites derision, and at best, certainly not 4 marks.