Seven Years Ago, An IPCC Lead Author Exposed Critical Weaknesses of the IPCC Foretelling Tools

In the interests of “piling-on” let me add another example of Trenbreth’s refreshing candor:
Trenbreth, too, has criticized the bureaucratic nature of the IPCC process, suggesting that it may have outgrown its usefulness. “I do think we should have declared success and moved on, after the last report,” he said. “There are too many scientists involved who are not leading researchers–owing to the demands for new people, and geographical, national, and gender equity.” (Google: “huffington post tom zeller 12/14/2012 leaked ipcc report excites skeptics”).
Uh-oh! So did Kevin say what I thought he said? Hotwhopper’s gonna have fun with this one, I’m thinkin’.

Most (all?) of the models are variants on one, different parameter settings by grad students. I hear there is silence about the 3% that come within shouting distance of observations because their ECS settings are so low.

“The burden of proof that a model result is not valid should be on the critic, not the modeller……”
Pseudoscience.

Correct me if I’m wrong, but isn’t this the same “burden of proof” rule that applies to crystal ball readings, tea leaf readings, entrail readings, and other “analog” methods of predicting the future?
How could someone even write that? It’s like saying the burden of proof that you murdered someone should be the responsibility of the murder victim.

The burden of proof that a model result is not valid should be on the critic, not the modeller……

Say what?
a) Models are uncertain.
b) No, that’s not strong enough. Models are wrong, but some can be useful.
c) But it isn’t the responsibility of the modeller to validate their own model. That job is left up to the “critic”. Otherwise, we should simply uncritically accept model results as being wrong but useful.
The problem with this is simple. What do you do if there are no critics? How do you maintain the honesty required for science to work at all? To quote Feynman:

… But there is one feature I notice that is generally missing in cargo cult science. That is the idea that we all hope you have learned in studying science in school–we never say explicitly what this is, but just hope that you catch on by all the examples of scientific investigation. It is interesting, therefore, to bring it out now and speak of it explicitly. It’s a kind of scientific integrity, a principle of scientific thought that corresponds to a kind of utter honesty–a kind of leaning over backwards. For example, if you’re doing an experiment, you should report everything that you think might make it invalid–not only what you think is right about it: other causes that could possibly explain your results; and things you thought of that you’ve eliminated by some other experiment, and how they worked–to make sure the other fellow can tell they have been eliminated.
Details that could throw doubt on your interpretation must be given, if you know them. You must do the best you can–if you know anything at all wrong, or possibly wrong–to explain it. If you make a theory, for example, and advertise it, or put it out, then you must also put down all the facts that disagree with it, as well as those that agree with it. There is also a more subtle problem. When you have put a lot of ideas together to make an elaborate theory, you want to make sure, when explaining what it fits, that those things it fits are not just the things that gave you the idea for the theory; but that the finished theory makes something else come out right, in addition.
In summary, the idea is to give all of the information to help others to judge the value of your contribution; not just the information that leads to judgement in one particular direction or another.

It seems that Feynman would disagree with Trenberth that the onus of proof is on the “critics” to disprove a model, or to be even less obtrusive and merely request that the model be validated in some way against reality before using it to dispose of a hundred billion dollars or so a year for the indefinite future at the expense of fixing all of the other problems in the world that could be fixed for 8 billion dollars a month). Y’know, like curing malaria. Vaccinating all the children in the world against common childhood diseases. Building supplies of clean, safe, and plentiful water to prevent the transmission of water-born fecal-oral illnesses. Installing sewage treatment systems to prevent the contamination of water supplies in the first place. Building schools worldwide. Bringing about world peace by creating world prosperity.
Back in my day, people hated the Viet Nam war because:
a) We were supporting an incredibly corrupt government that basically perpetuated the system of de facto slavery and economic suppression practiced by the colonial French. From an ethical point of view, we were fighting for the wrong side.
b) Old guys voted for the war. Young guys had to go fight it, and (for most of the duration of the war) couldn’t even vote.
c) The war was never declared, and indeed was fought with rules that prohibited us from actually invading North Viet Nam, systematically eliminating its military assets, and winning. We literally could not win as long as North Viet Nam refused to give up and could not be invaded, while they could invade South Viet Nam at will.
d) The war was ultimately supported by a mix of the large multinational corporations and organized crime! It (like the Gulf wars) was a money launderer’s dream. It is estimated that $0.50 of every $1.00 spent on Viet Nam by congress was skimmed or stolen. The “military-industrial complex” that Eisenhower warned us about sucked that teat dry. (And wept when it finally went away, to be sustained only by the cold war for sixteen increasingly desolate years, when even that went away.)
e) Was justified by an esoteric “domino theory” that said that if we didn’t oppose the spread of communism with military force, one nation after another would become communist. Viet Nam, of course, was direct evidence that this theory was simply not correct. Not to mention the complete collapse of militarized global communism 16-18 years later.
Notice the similarities. An esoteric theory. An infinite money sink endorsed by congress to support a “cause”. A nation’s worth of productive power perfectly happy to line up with satchels open for each to receive its helping of manna from heaven to prevent the imagined horrendous consequences of failing to act on the basis of the theory. Money laundering, support of the military-industrial complex, graft to be skimmed, pork to be barrelled, and all supported by a bit of wink wink nod nod, you know what I mean you know what I mean when even the proponents of the theory refuse to look at it too hard lest the entire thing dry up and blow away when the political climate changes.
At least we aren’t sending our children to be killed over it.
No, instead we are simply ensuring that somebody else’s children will, as a direct consequence of the gravy train so carefully erected to divert tax money and disposable income into a boondoggle, die of preventable illnesses, while their parents and surviving siblings will spend yet another generation living in 17th century pre-industrial poverty waiting for the cheap energy that might make them prosperous and free.
Honesty really is the best policy. There is (in my opinion) absolutely no doubt that there is some anthropogenic contribution to the generally observed global warming. Indeed, I can and have personally fit carbon dioxide concentration to HadCRUT4 data and determined a truly excellent fit across all 164 years with the single formula:

The residual standard error of the resulting fit (built using R) is 0.121 on 163 degrees of freedom with a two parameter, purely physics-based fit, and one of those two parameters is merely matching the mutual scales, since an anomaly has no absolute origin. To the extent that the carbon dioxide increase can be attributed to human activity (which may be arguable but almost certainly isn’t zero) there is literally zero statistical doubt in the reality of anthropogenic global warming. One is fitting a formula derived from basic physics to observational data and obtaining agreement with parameters that have a p-value of zero (less than ) in a two parameter fit, one of which is mere alignment of the scales and could be eliminated by a suitable transformation of the data (e.g. by taking a derivative and fitting that instead).
The only problem is that this fit predicts 2.3 K of additional warming by 2100 if the worst case scenario of increasing carbon dioxide is realized, one that carries us to around 900 ppm by 2100 (call it RCP9.5 from AR5). That scenario requires CO_2 to increase like the exponential of the date displacement cubed for the next 85 years. It is difficult to imagine any future growth scenario that leaves this rate (which fits Mauna Loa data adequately) intact, if indeed it is the correct but non-unique description of ML data. If one fits the Mauna Loa data optimally with an extrapolable exponential of a power form, one obtains the exponential of roughly the date displacement to the 1.3 power, which leads to between 600 and 700 ppm CO_2 in 2100. That results in only 1.4 K of additional warming by 2100.
Even this scenario (roughly their RCP6.5) is probably unduly pessimistic. Solar power unsubsidized is coming into its own and will almost certainly continue being implemented both at the private and the public utility level simply because there is acceptable ROI in many parts of the world and fuel is expensive and likely to become more so over time. Lockheed-Martin is publicly claiming that they have a working fusion reactor design and are starting to build a working prototype that they expect to demonstrate in five years. A number of other groups are similarly claiming (perhaps with less credibility but that doesn’t make their claims untrue) that they either have a solution or are within an ace of one, functional prototypes if not patents pending. There are literally mountains of Thorium in the world (including a whole range of them in western North Carolina) and LFTR is another potential game-changing technology. Batteries continue to improve. There isn’t one single reason to believe that we will still be building coal burning power plants in fifty years even if we do nothing to artificially increase the cost of carbon based energy now, even if we do nothing by way of artificially subsidizing renewable resources with a poor ROI and major unsolved not-ready-for-prime-time technological problems.
I really should do one final carbon scenario — the logistic scenario where one fits Mauna Loa to a symmetric logistic function that has its inflection point around 2050. Carbon would then peak at around 500 ppm. That would produce 2.62*log(1.2) = 0.47 K of additional warming. It assumes that we’ve already built around 1/3 of the carbon-based generation facilities that will ever be built, and that by 2050 (after two full more decades of frenetic building by the Chinese and India and African countries) one or more of the technologies that would make burning carbon for energy utterly obsolete will mature and have such an enormous comparative ROI that coal burning plants are allowed to age out (if stored solar ultimately is cost-effective) or are converted if e.g. fusion really is made feasible anytime in the next thirty years! as it will instantly make coal not only obsolete, but so obsolete that the last coal burning plant in the world will be decommissioned within a decade or at most two of commercially productive fusion.
Honesty! We are betting the certain and immediate demise of millions of people a year that we could save with one tenth of what we are pissing away panicking over carbon dioxide against warming that a sane, one parameter fit that works nearly perfectly historically extrapolates to be anywhere from 0.5 to 2.5 degrees kelvin, with the latter a very unlikely worst case scenario and the former the most plausible one!
It is difficult even assess the expectation cost of the current policy fully amortized to 2100 against the worst case scenario given the enormous and certain cost steadily accumulating today.
If you gave the proponents of severe measures to stop an imagined “catastrophic” anthropogenic global warming a knife and a child in the third world, and told them that they could prevent the possible catastrophe with complete certainty if only they take the knife and cut out the heart of the child, I suspect that you would have very few takers. Even fewer if you were told that you would have to repeat this sacrifice every ten seconds and rip the still beating heart out of some 3 million children a year, every year, for the indefinite future, adding another quarter of a billion children to the pile of 20 or 30 million we’ve stacked up already by the year 2100. Yet they do not hesitate to just as certainly condemn those children to death by simultaneously making energy more expensive worldwide and by diverting most of the elective wealth in the world away from the simple measures that would prevent those utterly preventable deaths.
Perhaps the voters of the world would appreciate an opportunity to honestly appraise the honest self-assessments of the model builders, the ones where they bend over backwards to point out the flaws in their own models and the reasons they should not be trusted as a basis for public policy, instead of deliberately concealing those flaws in the misapplication of statistics and deceptive language that currently forms the basis of the Summary for Policy Makers in the various AR’s of the IPCC.
In the meantime, the (frankly) unexpected success I’ve had in fitting a pure null-hypothesis carbon-dioxide only model to past data demonstrates that a simple two parameter physics-based model blows the pants off of the collective MME mean of all of the models in CMIP5. That doesn’t make it predictive — for one thing, no model simple or otherwise can easily fit the Holocene, or even the last 2000 years, in no small part because we have only a terribly inaccurate idea derived from multiple, inconsistent proxies of what global temperatures and climate were like outside of the modern historical instrumental era. For another, the model ignores all sorts of local phenomena that affect short term climate trends (e.g. big volcanoes), all sorts of nonlinear feedback that the system could at some point express even though for 164 years there is no statistical reason to reject the carbon-dioxide only null hypothesis, all of the longer time scale climate dynamics, and hence could work well but have lousy predictive skill.
Even so, it could hardly do worse in the future than the CMIP5 MME mean, which is already showing no predictive skill and is lousy at hindcasting HadCRUT4, as well, let alone any of the component models that contribute to it.
rgb

Actual after he made this claim some one went through the IPCC report and pointed out they did indeed make predictions quite a few time , Trenberth like the rest of the Team mistakes the massive size of ego for proof of his claims.

One of the chief problems with alarmist predictions is the bad statistics.

One last time, Well Said, Sir!
Indeed, it should be graven with a needle into the corners of the eye as a warner to all who would be warned. One day climate science will be held up to world-spanning mockery for its utter disregard of the laws of statistics and its common rules of best practice.
Oh, wait. It already is, at least by such luminaries as William Briggs.
I’m trying to do my bit as well.
rgb