What’s Wrong with the World

I recommend spending some extended time here (the blog www.wattsupwiththat.com) for perspective.

Michael Flynn has a nice takedown on the relation of solar activity and global temperature. As well as looking at a broader slice of time than merely the last geological "shake" since 1880.

http://m-francis.livejournal.com/105847.html

8. We need to get rid of the fraudulent grant-seekers in Climatology in order to have any real science done in the field.

http://bishophill.squarespace.com/

You might check his book out for a start.

Incidentally, it's increasingly clear state sponsored science isn't a good idea. Questionable and highly unverifiable science is happening in many fields- nutrition, psychology, sociology, etc...
The list is long and the research tends to justify the existence of a huge government that constantly interferes.

By contrast, something like the discovery of spontaneous order working with traffic,

http://marginalrevolution.com/marginalrevolution/2010/08/spontaneous-order-on-the-road.html

actually works in the real world (Climatology is plagued with computer simulations that we are somehow supposed to take seriously despite the number of variables involved) and tends to suggest less government interference gives us better outcomes.

What seems fairly certain is that "climate science" is in fact not science at all and cannot claim the authority our society gives to scientific conclusions. Mencius Moldbug has some very useful things to say about AGW:

That is: is the official truth of AGW, which claims the high credibility produced by Popperian falsifiability in a functioning system of critical feedback, in fact justified in claiming this credibility?

The answer is easy: no.

To understand the impact of increased CO2, we need to know the climate sensitivity. Q: how can scientists, at least Popperian scientists, evaluate the climate sensitivity? A: they can't. There is no falsifiable procedure which can estimate climate sensitivity.

To estimate climate sensitivity, all you need is an accurate model of Earth's atmosphere. Likewise, to get to Alpha Centauri, all you have to do is jump very high. The difference between the computing power we have, and the computing power we would need in order to accurately model Earth's atmosphere, is comparable to the difference between my vertical leap and the distance to Alpha Centauri. For all practical purposes, climate modeling is the equivalent of earthquake prediction: an unsolvable problem.

If you want to see this argument laid out in detail, read Pat Frank's article in Skeptic. To my mind, all this detail about error bars simply obfuscates the fact of an unsolvable problem. The GCMs that purport to simulate climate are interesting experiments, and it's not unimpressive that they can be made to produce results that look at least reasonable. But they model the atmosphere with grid cells 100 miles on a side, and attempt to use this to predict the state of the atmosphere - a chaotic system - for the next century. This does not pass the laugh test.

There is simply no scientific way to verify or falsify the accuracy of any such piece of software. It is not practical to perturb Earth's climate, perturb your model's climate, and test that they both respond in the same way. And there is no other way to test a model. In the end, all you have is a curve that records past temperature, and a piece of software that generates future temperature. Perhaps if we could watch the predicted and actual curves match up for a century or so, we could generate something like statistical significance. But we can't. And hindcasting - fitting the models to data from the past - overfits, and is completely worthless.

http://unqualified-reservations.blogspot.com/2009_01_01_archive.html

It's real? Good, only 7 more degrees (C) and we'll be where it's optimal. I'm sick of living in the lingering end of this ice age.

I'm glad to see more conservatives starting to take the possibility of global warming seriously, but it looks like you are just going to get knee-jerk reactions against the idea because "conservatism" these days means little more than defending the interests of major corporations.

Also, I find it interesting that someone would suggest that we cannot trust government-backed research, but we can trust research done by liberals who are ideologically committed to the idea that the "free market" will solve all of our problems. I'm not saying that we shouldn't take liberal arguments seriously, but if you are going to reject government-sponsored research because it might be biased, I don't see why you shouldn't do the same for the neoliberals.

Warren: Been listening to Watts for years. Are you local? Thanks for the link.

Alan: I'll check it out, thanks.

August: Please explain why profit-motivated science, or agenda-driven non-profit science, should be inherently more objective than state-sponsored science.

Andrew E: Extrapolations are always dangerous and too frequently misapplied. Nevertheless, historical data correlations either mean something or they don't. It seems to me that Moldbug lacks curiosity: "we can't have absolute certainty, we can't have perfect climate models, we can't run falsifiable experiments, etc., therefore we should forget about it". But there are other ways to determine probable outcomes, and it seems to me that that's all we're looking for here: probabilities.

Jeff,

...therefore we should forget about it

No, I think Moldbug's main point is not that we should forget about it but that we can't call climatology science.

Have I missed anything?

Yes:

GW is real and it does matter. Rising average temperatures will be necessary to produce more food globally, especially as carbon inputs to agriculture become scarce.

Not saying I believe either one, but neither is implausible.

We should be reducing our dependency on fossil fuels anyway.

#1: This is not a self-evident truth.

#2: If the anthropogenic claim is pseudo-science, this statement is frustrating in this context. Compare: If someone came up with a kooky theory that fornication was causing plants to die in impoverished countries, and if that theory were roundly refuted empirically, there would be something rather odd, to put it mildly, about moral conservatives saying, "Oh, well, people should not be fornicating anyway." Welll, yes, but why bring that up just _here_, when you are discussing the theory that said activity has a particular causal effect? Please remember that plenty of liberals have used this type of statement to justify _not worrying_ about justifiable scientific doubts concerning the anthropogenic claim. _That_ approach amounts to something uncomfortably close to a justification for fraud: "We want to induce people to do, or not do, X anyway, so let's tell them something that we have plausible reason to believe is empirically false to get them to go along with our other ideas about their actions."

James: I think you're right on all counts.

Andrew E: That's fine with me. I'm happy just calling it climatology. Even so, climatology depends upon science and draws upon science, so there's no escaping the association.

Nevertheless, historical data correlations either mean something or they don't.

Like the Medieval warm period? Hmmm? The one that had to be "hidden" to make the hockey stick graph?

It also seems indisputable that...rising CO2 levels correlate starkly with rising global temperatures.

That's _indisputable_? I'm trying to figure out how you can possibly think this, Jeff. You _must_ know that it is by no means indisputed, that it's disputed directly by the people that the "warmists" decry as "denialists."

rising CO2 levels correlate starkly with rising global temperatures

I think this is relatively uncontroversial. The trouble is in working out the direction of causation. A warmer earth produces more carbon based lifeforms (cf. Cambrian Explosion), thereby unlocking carbon.

That's right, Steve. I've seen "denialists" dispute causation, and even promote reverse causation, but I've never seen anyone deny the correlation. That doesn't mean they're not out there. Lydia, if you have some contrary data set in mind, please share it. I'm expecting to learn something in this thread.

"We should be reducing our dependency on fossil fuels anyway."

#1: This is not a self-evident truth.

It's a slam-dunk as far as I can tell. Why would you dispute this?

GW is real and it does matter. Rising average temperatures will be necessary to produce more food globally, especially as carbon inputs to agriculture become scarce.

Steve, I like it, and readily admit its plausibility. In this view GW is a benevolent act of Providence. On the other hand, we could be getting what we deserve instead ...

The trouble is in working out the direction of causation.

Well, that's the trick isn't it. I suppose this could be achieved either deductively or inductively. A deductive methodology requires a working model of the earth's atmosphere. An inductive methodology would require some fundamental axioms, known a priori, regarding the dynamics of the earth's climate much like proper economic theory is built up using praxeology.

Jeff, I'm thinking there specifically about the MWP.

Why would I dispute it about fossil fuels? Well, because for one thing, I'm also not convinced that "peak oil is real." I'm not convinced that we know how much oil there is in the ground, and we're not by any means doing all that we can even to make use of the oil resources we have within our own sovereign control. (Thanks, environmentalists.) The use of fossil fuels is just another technology, with tradeoffs, benefits, pros and cons, like anything else. It's not uniquely bad or uniquely something we need to be getting away from. It may be that we'll find good energy options that actually work other than that and that work better and are economically viable, and if so, fine. But to make it a priority as an end in itself will almost certainly result in the artificial pushing of technologies not ready for prime-time or harmful in other respects--possibly even to the environment, for that matter. Frankly, if I really thought that reducing our dependency on fossil fuels were an A-level priority, I would push for much more use of nuclear power. One of our biggest problems here is that the same people telling us that it's self-evident that we need to reduce our dependency on fossil fuels are utterly closed to that option. In fact, when one finds that again and again the options that would allow first-world humans to keep their quality of life are the very ones that we "must" not use, one begins to think this isn't a coincidence--that reducing first-world quality of life _is_ the goal. I'm not saying for you, Jeff, but I mean plausibly for the people you are reading and being influenced by in these environmental areas.

Do I have to drag out Feynman's Cargo Cult Science speech?

That is: is the official truth of AGW, which claims the high credibility produced by Popperian falsifiability in a functioning system of critical feedback, in fact justified in claiming this credibility?

Popperian falsifiability as a sine qua non for science has been known to be false for thirty years. In fact, there are some undecidable propositions (read: unfalsifiable), such as the Continuum Hypothesis in set theory and the Halting Problem in computer science that still, despite this, allowed many significant breakthroughs to be made.

In fact, if we could control the weather, then the falsifiability of AGW would be irrelevant.

To estimate climate sensitivity, all you need is an accurate model of Earth's atmosphere. Not necessarily. To estimate the sensitivity of water to heating, all you need is a pan, a thermometer, and a watch. You don't need a detailed knowledge of every molecule of water in the pan.

But they model the atmosphere with grid cells 100 miles on a side, and attempt to use this to predict the state of the atmosphere - a chaotic system - for the next century. This does not pass the laugh test.

There are so many things wrong with this statement it is embarrassing. To begin with, what does the size of the spacial grid have to do with the time-dependent behavior of a computer model? Grid size has to do with resolution - spacial resolution, not temporal resolution. He has screwed up the space and time domains.

Secondly, one needs to use both grid and step size to estimate a time-dependent system and the size of the grid is directly related to the uncertainty in the fineness of the solution, but chaotic systems are not what this man thinks they are. In fact, there is no computer program, anywhere, that will every be able to model a chaotic system, because it takes an infinite precision to do that and all finite systems have finite precision. One can, however, find a strange attractor and in fact, sometimes find an exact solution to the equations of a chaotic system and from these, one can make bounded estimations.

In fact, there is a million dollar Clay Prize waiting for someone who can find a closed-form solution to the Navier-Stokes equations, which is the governing equation for all of fluid dynamics. We can solve, exactly, certain limiting cases, but not the general case. If we could, then we would have the necessary equations to say something definite about the earth's weather systems, even though, paradoxically, we still would not be able to predict the weather. Chaotic systems have exactly definable dynamics, but unpredictable time series, because the solution space is larger than the function space, and so the solution folds back in on itself with non-integer, fractal dimensions.

Nevertheless, one can place bounds on chaotic systems. For instance, I can say with 100% certainty that the temperature of the water vapor on the planet, without external heating from extraordinary circumstances (such as a pressure wave induced by the collision of rogue asteroid hitting the planet), can never get above 110 degrees, C. Likewise, the oceans will never get colder than -10 degrees, C (the higher and lower temperatures are because of contamination of the water and the effects of pressure). All of these things allow for bounds on certain questions.

Oh, and it has been known for at least ten years that certain chaotic systems are controllable.

People should really learn something about chaos theory before they start making unfounded commentary. I spent several years in graduate school studying non-linear dynamical systems theory and chaos because chaos is seem in the neural activity in the brain in such things as epilepsy.

There is simply no scientific way to verify or falsify the accuracy of any such piece of software.

The man does not know what he is taking about. To begin with there are whole graduate courses on error estimation in computer programs (including those that deal with chaos). The software can be shown to have whatever accuracy it has. What he probably means is that without knowing the real values of the input data, there is no way to know whether or not the equations correspond to reality, but that is not a function of the software. One can, in some cases, use multivariable Bayesian analysis to refine the predictions based on improvements in data as it becomes available.

It is not practical to perturb Earth's climate, perturb your model's climate, and test that they both respond in the same way.

He has not proven that one has to perturb the ENTIRE system to show that the model works. If that were the case, to test Einstein's General Theory of Relativity, one would have to perturb the entire universe each time one ran a test. It all depends on things like the locality and topology of the theory. One could test some computer programs that model the earth's atmosphere in a freezer. It all depends on what one is looking for. His statement is too broad brush.

And hindcasting - fitting the models to data from the past - overfits, and is completely worthless. Not necessarily. One can do this in equilibrium systems all of the time. The question is how far the earth's atmosphere is from equilibrium. Since it is chaotic, it is non-equilibrium, but there are close and far from equilibrium systems. I don't know the measure on the earth's atmosphere and I suspect the author doesn't, either.

From August's comments: something like the discovery of spontaneous order working with traffic,

I'll bet you didn't know that that can be modeled using something similar to fluid flow, so the two problems are not unrelated.

All I want to say is that the general public and politicians can gripe and moan and make histrionic statements on both sides of the fence and I hold them accountable for the fog of discussion (I'm looking at you, Al Gore). Scientists, if left alone, will, eventually ask the right questions. The problem is that they are seldom left alone. The lure of money or fame tempts many good scientists, but then this marches, lock-step with the greed and pride of politicians. Jefferson wanted a high wall of separation between Church and State, but today we need a high wall of separation between Science and the State, but come to think of it, since both theology and climatology are sciences, with precision being highly important, while in politics, the fuzzier the better, perhaps that wall should have been recognized to have been really the same wall, all along. Both theology and science have always been the worse when they have had to work with temporal powers.

Sorry, for going on so long. I have about five hours more of things I have prudentially decided not to say. I just wish we could have more clarity between humans. It would certainly lead to a clearer understanding of the weather.

The Chicken

Welll, yes, but why bring that up just _here_, when you are discussing the theory that said activity has a particular causal effect?

Just a reminder: Fornication is a sin; reducing fossil fuel consumption is not.

In this case doing what is certainly right, for one reason, has the added benefit of doing what is possibly right (and certainly not wrong) for another.

Oh, before anyone asks: I have no idea about AGW. I say we find a nice little exoplanet and run an experiment. Who's with me :)

The Chicken

Ok, so basically The Chicken says building a useful model of the earth's climate sensitivity might be possible, but not necessarily so. Interesting.

I'm afraid I can't help thinking when seeing this post of Mark Steyn's recent comments about Mitt Romney, who has embraced AGW "two years after the peer-review hit the fan in East Anglia, Copenhagen, and at the IPCC."

http://www.nationalreview.com/corner/268929/global-warming-mitt-mark-steyn

Jeff, I think you misunderstood my analogy concerning fornication. I was saying that even if _not_ reducing fossil fuel use were a sin (as fornication is), it would be worrisome to see people bring that up in the context of a discussion of whether that particular sin has a particular empirical consequence. If the claim that it has a particular empirical consequence has been strongly empirically refuted, it's troublesome to hear people bring up the wrongness while discussing the empirical question. I wouldn't do that if someone had a kooky theory about fornication causing some dire consequence to crops. I wouldn't say, "Oh, well, we should do that anyway, and it might be having this effect on crops, so we might be doing something good for the world, too." Because if I did it might sound like I was open (as I'm afraid some liberals are in the current case) to pushing something scientifically dubious for reasons of stopping the sin. Therefore, I counsel _not_ saying things in the context of discussing the AGW theory like, "Oh, well, we should do that anyway."

FWIW (I haven't checked out the details), there's this statement:

according to The New York Times of November 7, 2006, two ice ages having apparently occurred in the face of carbon levels in the atmosphere 16 times greater than that of today, millions of years before mankind’s appearance on earth

http://cambioclimaticoglobal.com/climate-change-news/do-you-want-to-hear-the-other-side-of-the-climate-change-debate

I don't know what the independent evidence (as opposed to conjecture based on the assumption of correlation) is concerning CO2 levels during the Medieval Warm Period, but whatever they were, they weren't caused by the kinds of human activities we are now being told we must stop or reduce.

Not much time at the moment, Lydia, but a couple of thoughts:

Why would anyone think that atmospheric conditions on an uninhabited and uninhabitable earth millions of years ago is relevant to the question? I mean, apart from proving that high C02 levels make the planet unfit for human habitation?

Also, the extrapolations required to make this kind of statement - that two ice ages occurred millions of years ago with atmospheric C02 levels 16 times greater than today's - are so shamelessly wild and riddled with untenable assumptions that one hardly knows how to respond without ridicule. I'm sorry, but this kind of objection goes absolutely nowhere.

so shamelessly wild and riddled with untenable assumptions that one hardly knows how to respond without ridicule.

My reaction precisely to many of the things said by warmists, Jeff. I certainly hope you're not under the impression that their conclusions are straightforward, simple, and objective and not involving questionable assumptions and extrapolations. For example, a link I was just looking at (but haven't been able to re-find) claimed to be able to tell us the carbon dioxide levels for many hundred thousand years past--and of course, this was supposed to tell us how dangerous things are now, because they are now higher than ever before. Then there are the pieces of counterevidence to various concrete predictions that have actually been made, such as about glaciers and such.

Ironically, when you put the claim about CO2 levels over hundreds of thousands of years together with a study linked (honestly enough) by this warmist:

http://www.nicholas.duke.edu/thegreengrok/hockeystick-revisited

you find a counterexample to the correlation claim. The Oppo study he mentions would seem to indicate that the MWP was similar in temperature to current temperatures. But in that case, we have higher CO2 levels now than ever before without having higher temperatures than ever before.

Now, I think myself that this is _all_ highly conjectural and probably based on untenable or at a minimum independently poorly supported assumptions. We're making a whole lot of guesses, and what astonishes me is to see you, especially in the aftermath of the absolute mess of Climategate, seeming to take the warmists for the objective and careful scientists!

This is especially amazing as I know that you are, of all things, a YEC, and yet you do not even seem to realize the extent to which the kinds of strong extrapolations to which you object in the creation debate are being carried out by warmists themselves.

Jeff - are you aware of any warmist who admits what it would *really* take to reduce global CO2 admissions sufficiently to make any great long-run difference, given his own claims about the current situation and its causes?

As opposed to the usual suspects who advocate lots more power for the state plus lots more funding for "scientists" coupled with utterly trivial adjustments a la Kyoto?

I'm not.

Yeah, the kinds of CO2 reductions we would need, I hear, are analogous to those of the US budget shortfalls. Anyone who isn't talking about realistic reductions (based on their own models) probably aren't actually concerned with climate change/national debt.

I've got the perfect solution for global warming: everyone should hold their breath, then we should, immediately, blow off the atmosphere with a series of precisely-timed nukes,mynah very one should exhale. Problem solved :)

The Chocken

We can't even trust the measurements offered, and there are systematic efforts by folks to hide their data, rather than share it for peer review.
If the folks who support it won't treat it like science, why on EARTH should anyone else?

Chicken-
The setup of "pan, thermometer, watch, water" is an accurate model of "heating water." The model MUST be accurate to within a set amount-- for example, the pan of water is going to be inaccurate by the tolerance of the thermometer, the tolerance of the watch, and the interference offered by the metal of the pan.
The model's accuracy can be tested by seeing if observed effects can be used to make accurate projections-- perhaps with a different pan, different amounts of water, other known variables, then it can be used as a "known value."
Current computer climate models aren't able to "predict" things that previously happened when we input old measurements, so they're known to be inaccurate. (when you have to tune them during the running to pass, it doesn't count)
Calibration, I know-- I am a trained and experienced calibration technician. Fiddling with variables during a test invalidates the text. Coyote's "plug variables" would be like me putting my thumb on the scale to make sure it read the right weight for all the tests.
(I get rather emotional about this stuff, because one of the first things I did on ship was clean out the gear of the Marines we lost when a helo crashed-- someone had miscalibrated the equipment used to calibrate their altimeter, which didn't usually matter...if they hadn't been in the middle of a sand storm. Accuracy MATTERS, burn it, especially with something used for making big decisions, be it policy or "how high are we off the ground.")

Jeff, state sponsored research tends to be inherently biased in a particular way. Other types of research may be biased, but the biases would tend to cancel each other out. In other words, state sponsored stuff sends a strong signal to find state sponsored solutions, but without the state we'd get a variety of signals. Repeatability becomes more important again- we wouldn't have CRU people destroying data, for instance, or using code that would create a hockey stick if you ran the phone book through it. These are games played to appeal to powerful politicians. Without state interference, wealthy individuals might get taken in by charlatans, but the damage would be limited to the individual's fortune and not the entire world population- again, because without political power scientists have to rely on repeatability of results and persuasion rather than state sanctioned forms of coercion.

Just think of the field of economics. Governments hire economists; economists ignore basic economic reality and provide various fairytale reasons as to why whatever the politicians want to do will work.

Chicken, I did know spontaneous order in traffic could be modeled. One of the reasons I can't stand climate models is that I've read about Lorenz. He discovered how such tiny changes in one variable can make a massive difference in the models; and then we are talking about how many variables? Not to mention variables we don't know about? I wish I could find that paper on the DNA modeling that has been going on- apparently, instead of doing direct testing, someone came up with a simulation and lots of researchers used the simulation because it is cheaper. Then someone checked the simulation against real world testing- the results seem to suggest a lot of revision will be coming in DNA science. I believe the situation to range from problematic to dire in many fields of research.

To go back to the beginning,

Global Warming is Real...That much seems indisputable.

Ahh. Ahem. Well, no, I don't think so. It is, in fact, disputed by people who don't have a personal iron in the fire.

Let's take the simple data shown in the table. Oh, excuse me, the ADJUSTED data that sits in the table. Adjusted? Yes, of course. It is very likely adjusted for urbanization creep. 150 years ago, at least 70% of the recording stations (ones in use both then and now) were in far less urbanized locales than those very stations are now. It is well known that urban areas are warmer than the surrounding countryside, because of all the sources of man-made heat - i.e. heat that is not part of the weather. What is not known is by how much this urbanization affects the temps recorded. As a result, we have all these recording places that show higher values for temps now than they did 150 years ago - due to being next to an apartment building in the middle of a large suburb instead of being 10 miles outside of town, for example.

For some long-range reports and graphs, scientists take the recorded values for temps for these now-urban locations and adjust them. It is a matter of guestimation as to the "right" amount of adjustment: there is no way to be sure you have an adequate (and not excessive) adjustment formula, and there is no way of being certain of how much variation (for different kinds of cities) in the adjustment formula we need. For example, to avoid circularity with respect to the global warming problem, you cannot say "these city temps are 5 degrees higher now than they were in 1880, so the right adjustment amount is 5 degrees", since you don't know whether global warming changed the base (weather) values separately from the man-made heat. If you compare the local records to out-lying but similarly situated other recording stations, you either have (a) other stations close enough to the one in doubt that they are also affected by urbanization creep, or (b) other stations so far away that you cannot be confident that they adequately reflect a valid base-line for the weather that is going on at the local station you know you need to adjust - they are too far away to be subject to the exact same weather, and therefore they may be unable to guide you to the correct adjustment amount. (After all, part of the reason for using many weather recording stations is that there are many different weathers in many places, not all the same and not all reflecting overall climate in the same manner.)

Like it or not, we simply cannot say with very high confidence that we actually know what the temps are doing in detail, compared to 1880, in a way that excludes directly man-made recording variation.

To take another measurement / interpretation problem: the temp listed in the table is "average global temperature". This is a constructed composite that may not really "mean" anything that is in the least bit important in understanding weather, climate, or anything else. For example, let's suppose (I am just hypothesizing one possible way of constructing the average listed, I don't know what method was used) that the average uses daily averages from recording stations: today's high of 90 and today's low of 68 get averaged out to a daily value of 79. You can get 365 days of daily values to get an annual value for this station. Average all the stations recording, and that's your global average for the year.

But wait a minute. If 90% of the recording stations are in America, and 5% in Europe, and 5% in the rest of the world, then what the "global" average really shows is an American average. If 99% of the stations are over land and only 1% are over water, then it shows a "land average" not a global average. Given that more than 70% of the surface is water, that's not a very good stand-in for the global average.

Wait another minute. If the high today is 90 and the low is 68, but the day has 6 hours between 88 and 90, that's a very different weather showing than a day with the same high and low but only 2 hours above 86. The mid-point between the high and the low simply may not be sufficient data to say what the weather is actually doing in a meaningful way. The total amount of time above X or below Y may be far more meaningful data than the mid-point between the 2 extremes for the day. Is it known if daily temperatures keep closely to a curve than on a daily basis balances the amount of day above the mid-point and the amount below the mid-point?

Let's take one last item: actually measuring the temp: different ways of measuring are used, and they are all subject to certain amounts of discrepancy / error. A mercury thermometer will show errors in different ways than an alcohol one, versus a metal one, or a digital recorder, and so on, so a time-related change in equipment could result in a time-related discrepancy in values. The calibration theory and process was a little gamey in 1880 compared to today. A given recording station may have variation that has a bias: a recorder in the Yukon may accurately record his readings 340 days per year, and simply guess on the worst 25 days because it is "too damn cold to go out there and get the stinking temperature." Just a few years ago, there was a national study that determined that the so-called normal "body-temperature" value of 98.6 F had been mis-stated for decades: it was re-stated downwards by more than a tenth of a degree, I can't recall the exact amount.

I am afraid that without a LOT more attention to detail, a graph that purports to show the change in global temperatures since 1880 is not terribly confidence-building. It may be a beginning of an argument that GW is real, but it cannot carry the argument alone.

Thank you, Tony. Just one set of several sets of types of problems with the whole argument on that side.

August,

If you want to read the original paper by Lorenz that started the chaos craze, it can be found, here. He used a truncated model of convection that just happens to show chaos. I assume you've played with the Lorenz Attractor. There are numerous on-line sites. I do not know if anyone has studied the solution if the equations are not truncated at three equation. No one has really proven that chaos exists in weather systems, only in approximations to weather systems (if I'm wrong, someone tell me, as I can't keep up with everything and I might have overlooked research that suggests otherwise).

I don't doubt that there are poor models that make poor predictions, but we use quantum mechanical calculations all of the time in chemistry and physics and trust them. If a model is wrong, it's wrong. The DNA modeling being wrong seems a bit strange because we know how to do that and it is not a chaotic system, but it has been known to happen. This is not a flaw in the science, but the application. It is difficult to model complex systems by computer. It is more of an art than a science and we don't always know who the good artists are. One wrong model in one field does not guarantee wrong models in others. For instance, fluid flow modeling for automobiles is quite refined. Unfortunately, some fluid systems, especially those with certain ranges of Reynolds numbers, can be turbulent and this is a chaotic process.

I just wonder what right I have to hold an opinion on global warming without reviewing the massive amounts of data and learning meteorology.

The Chicken

Like it or not, we simply cannot say with very high confidence that we actually know what the temps are doing in detail, compared to 1880, in a way that excludes directly man-made recording variation.

Yes, we can, by infrared satellite measurements. I do agree, however, with the massive problems of terrestrially-based measurements.

The Chicken

Some Climategate humor, on the hockey stick graph:

http://www.mandm.org.nz/2009/11/climate-change-statistics-humour.html

Yes, we can, by infrared satellite measurements.

I don't understand. We can't possibly have those for 1880, or for a long time thereafter.

Jeff C.,

There are two main sets of arguments related to AGW:

1) we can argue about the science, which I don't think is as settled as the mainstream media or scientific community would have us all believe (this is another good website for skeptics: http://www.climate-skeptic.com/);

2) or you can accept the scientific consensus and argue, like Jim Manzi, that the economics still suggest we should basically do nothing more than study the problem:

http://theamericanscene.com/category/Climate-Change/

I urge you to read some of the longer Manzi articles, because he is scathingly good on the economics. For example, just like Lydia, I would question your assumption that "[w]e should be reducing our dependency on fossil fuels anyway." Says who and why? There are economic trade-offs to not using fossil fuels which you ignore and which I would want us to consider before suggesting such a radical idea.

Yes, Foxfier, I agree that calibration is important and the measurement systems for recording temperature on the ground are pretty poor. Feymann made a related point in his Cargo Cult Science speech.

As for the satellite data, of course we don't have data from the 1880s. I was just pointing out that we don't have to rely of terrestrial data.

As for alternative fuel, fusion is the only way to go.

Personally, I don't really like fossil fuels and I think we have become fat and lazy because of them. I wish people lived close enough to work/play to bike there. We would have healthier people and who has ever heard of "bike rage?". I realize that we need energy, but without fossil fuels and nuclear fuels, there is scarce left other than beasts of burden or solar. I think we could build massive microphone transducers to convert all of the noise coming from politicians into electricity. Why, that, alone, would probably satisfy out needs :)

The Chicken

Jeff, you are wise to be concerned. I assume you want to leave your land to your kids and all the maps I've seen show the Central Valley getting hammered (not so much for the North Coast temperature wise and if you are at a decent elevation).

All this is probably academic anyway. Given the current and foreseeable political and economic realities, this cake is already baked and, for out grandchildren at least, most of the then living will curse us and envy the dead..

Why get off oil as much as possible? A meditation on the externalities - trillions on three recent wars and maintaining a huge standing military alone - should be enough.

Re: peak oil, Sullivan has an interesting collection of links which interested folks should follow.

http://andrewsullivan.thedailybeast.com/2011/06/running-out-of-oil.html

"I believe there are lots of folks on both sides of this issue whose confidence is, shall we say, unjustified by their actual knowledge. What I don't understand is why conservatives don't at least take the issue a little more seriously."

An excellent observation and question. More likely than not, the reality is that, as with, and way more so than with things economic. everyone reading this thread lacks the chops to deal with the data. Hence the references to Moldbug and arguments that depend on off-the-wall analogies (apologies to any of you who have a PhD in climatology AND who have done your work in this area).

The rest of us take our cues from our political and cultural orientations. The problem for liberals and conservatives alike is that it seems likely that those orientations have less to do with our analytical abilities than with our wiring.

http://neuropolitics.org/default.asp

The challenge for thoughtful conservatives is that reality currently (N.B. I said currently) has a leftward tilt. Recall that cap and trade (like mandates) was a conservative idea (liberals said OK, lets try it) from the days of acid rain to the 2008 presidential election, today opposition is a litmus test. The trend line shows conservatism to be hopeless on this issue anytime soon (and the window is fast closing).

On a personal level, and like others, you need to decide on your priorities.

We would have healthier people and who has ever heard of "bike rage?"

*raises her hand* You haven't had the "pleasure" of being run down by self righteous morons that think traffic laws are for other people?
A blanker on a bike is the same as a blanker in a vehicle, just with a bit less mass and a bit more self-entitlement. I sincerely doubt cities were a lot friendly when people were being run down by horses instead of cars.

Satellite temperature measurement doesn't address the issue of man-made temperature variations like the urban heat island effect, although it would help avoid the "temperature station by an air conditioner exhaust" type problems.

One wrong model in one field does not guarantee wrong models in others.

Being made by scientists is what guarantees wrong models in others-- science is all about failing over and over until you get it right, no? If you had enough data to make correct models all the time right off the bat, the need for them would be far lower.

We're not talking about one wrong model in one field-- we're talking about all the climate models thus far designed being unable to demonstrate unadjusted accuracy when tested against reality. The predictions don't pan out, the data gets hidden or flat-out destroyed, and they have to actually remove KNOWN facts to make their theories hold up. (a search for IPCC "Little ice age" "Medieval warm period" 1990 or similar should give a wide range of results; I'm trying to keep this out of the spam file)

everyone reading this thread lacks the chops to deal with the data.

I'm pretty sure I have the chops to deal with the data. Many scientists do. What we don't have is the time or the interest. There is a certain equality in scientific training, so that a good computer modeler in, say, quantum physics or chemistry, could probably be a good modeler in a related field. Just how different is climatology than, say, chemical hydrology or even musical acoustics? Musical acoustics uses very similar equations to model air flow. Stirred tank reactors in chemistry use similar flow equations to those in meteorology. It is physical chemistry, not meteorology, that explains why clouds form. Water vapor is a chemical phenomenon and thermodynamics of heat transfer is a problem in engineering and physics. Climatology is an interdisciplinary field and certainly experts in the related fields can both understand the research and make contributions. Want information about ice core data? Ask a paleontologist. Want to know about CO2 scavenging in the ocean? Ask a chemist.

I said I have no opinion on GW because I haven't done the research, not that I can't. There are only so many things a person can do in life and while I have a great interest in the weather, I do not have an interest in getting mixed up in politics and it is almost impossible not to get mixed up in politics when doing research in GW (thanks, Al Gore, for not keeping you mouth shut and letting scientist do their work without the interference of politics).

My main concerns in debates like these is to make sure that popular science doesn't get substituted for real science.

The Chicken

Briefly, Foxfier, your last two comments have both been held up. I've been on hand to release them quickly ... no guarantees next time! Carry on ...

Being made by scientists is what guarantees wrong models in others-- science is all about failing over and over until you get it right, no? If you had enough data to make correct models all the time right off the bat, the need for them would be far lower.

I understand your frustration. Some science has been oversold, but with only a few scrapes of data I could calculate the cooking time of a 28 lbs. turkey and be pretty close. It all depends on the model system. Non-linear systems are, generally, harder to model because of the sensitivity to initial conditions. The ordinary partial differential equations for heat flow are linear and the turkey is a relatively simple composite system.

I do theory, mainly, but I have spent my time doing extensive computer modeling of the mechanics of biological systems. I once reconstructed the biomechanics of a million year old fossil fern for a paleobotanist colleague of mine when we were in graduate school. How do I know the model was basically correct, even though no on has ever seen this fern - because the mechanical theory behind the modeling works just as well for plants as for buildings. Stress and strain are the same concepts today as a million years ago. Yes, I had to make some estimates, but plants are constrained by biology both as to their water content and their moduli of elasticity, so I was guaranteed to be pretty close.

Theory and applied people think differently. Sometimes you need the one when the other is impractical.

I know that the computer models in climatology are suspect. The earth is a very complex system. Of course, if you think climatology is hard, you should try doing neurodynamical modeling. We can't even do invasive measurements. If we had similar data to climatologists, I tell you, we be making much faster progress. The brain is orders of magnitude harder to model than the earth and the equations are just as nonlinear. It's amazing, but we do get thing right. We have pretty good model of how they eye works, but not the nose. Of course, computer scientists don't get the glory or blame when things go wrong in the brain - usually it's the neurosurgeons.

The Chicken

Some day, I'm going to just start typing and see how hilariously the auto spell distorts my writing. I used to be such a good speller. My name is Masked Chicken and I am a spellcheckaholic.

The Chicken

Global Warming is Real. That much seems indisputable. It also seems indisputable that atmospheric carbon dioxide levels have been steadily rising, and that rising CO2 levels correlate starkly with rising global temperatures.

Actually, the only thing I would consider indisputable here is that the graphs you've posted show sudden and drastic increases in warming and CO2. After Climategate, and all we've learned about data being persistently hidden, destroyed, corrupted, and manipulated, I would not bet a red cent that those graphs represent actual reality.

Jeff - yeah! They got out very quickly, I was just trying to make it so they didn't get snagged in the first place! Oh well.

TMC-
I think you misunderstand me; I'm not frustrated about any part of the model thing except for people putting their faith in models that we know are bad.
That we can't trust the data we're using to build the models only makes it worse. (IIRC, Russia's data was recently revealed to have been largely made up out of whole cloth. It's like trying to model economics and believing China's growth reports!)

"I said I have no opinion on GW because I haven't done the research, not that I can't."

Which was my point.

If things turn out badly, Al Gore will be compared to MG Billy Mitchell.

If things turn out badly, Al Gore will be compared to MG Billy Mitchell.

If things turn out badly in the climate sense, the likelihood that it will turn out badly under the specific mechanism that Algore promoted is quite modest. If it comes through any of several other modes of climate adjustment, Algore will be like the Jewish German 'prophet' in 1927 warning "the end is near" because he noted that the Jews were buying BMWs instead of Mercedes, and this portended grave results for German Jews.

@ al: "Given the current and foreseeable political and economic realities, this cake is already baked..."

For once, we agree. Even those who are loudest in proclaiming their faith in AGW totally fail to suggest any solutions that are even remotely within the ballpark of political plausibility + potential usefulness.

Something really dramatic will have to happen before that equation changes.

"...for ou[r] grandchildren at least, most of the[m] then living will curse us and envy the dead..."

I think that's fairly likely. But not because of global warming.

The Chicken said:

"Popperian falsifiability as a sine qua non for science has been known to be false for thirty years. In fact, there are some undecidable propositions (read: unfalsifiable), such as the Continuum Hypothesis in set theory and the Halting Problem in computer science that still, despite this, allowed many significant breakthroughs to be made."

This comment is a bit off-topic but here goes anyway. First, unfalsifiable statements are of the domain of the empirical sciences while undecidable statements are in the domain of mathematics. The examples given are from mathematics, so why does that show that Popperian falsifiability as a sine qua non is false? For the empirical sciences that is, which is the context in which the criterion makes sense.

And when speaking of undecidable statements one must be more careful. First, the halting problem is not undecidable in the same sense that the Continuum Hypothesis (CH for short) is. Turing's result can even be seen as a positive one -- although admittedly, I am being a little perverse here. Second, there is no absolute sense of the term "undecidable" as it relates to CH; the qualifier is always relative to some formal theory. We know that the Continuum Hypothesis is independent from the standard ZFC set theory: Goedel showed that ZFC cannot disprove CH, then Cohen showed that ZFC cannot prove CH. But if we consider the formal theory ZFC + CH, then this theory is consistent (assuming ZFC is) and trivially proves CH. Of course, this does not exactly count as a proof of CH so let me add a few more, slightly less trivial remarks. There has been intense work to try and unlock additional set-theoretic axioms that could settle the truth-value of CH. Probably the most common view among the specialists is that CH is false -- Goedel (a platonist) and Cohen (a formalist) were of this opinion. It has long been known that the constructibility axiom proves CH true, but my impression is that most set-theorists view the axiom as false, in part because it negates even relatively weak large cardinal axioms; at any rate it is definitely not an intuitive axiom and you will not see it used in standard mathematical practice. More recently, both Freiling and Woodin have given complex arguments (that I cannot even begin to understand) showing CH is false but the jury is still out.

When considering AGW:

1. Check the newspapers of the early 1880's. Seriously, even Hell's Bible(aka New York Times) articles are of great interest. It seems to have been unseasonably cold. Worst blizzards in living history, coldest ever, etc. And we start our data selection when?

2. The Dust Bowl. Again, check the ... wait it's all been digitized. I was about to say microfiche. Another sign I'm getting old. If it is warm now because of evil, icky, corporate carbon, what explains the early '30's? Hmm?

3. Medieval Warm Period. It required special cherry picking of bristlecone pines (which are useless as proxy data according to dendrologists) for Michael Mann (inventor of "The Hockeystick") to obfuscate.

4. Roman Warm Period. Even warmer than the Medieval Warm Period. Less CO2 to blame as well.

5. Minoan Warm Period. Even warmer still than the Roman Warm Period. Even less Anthropogenic CO2 to blame.

6. Yamal. This alone is a textbook example of a dendrology/climatology study gone wrong for the sake of whoring to AGW confirmation bias, er government grants. Please notice that Yamal is heavily relied upon in nearly all AGW studies.

7. Climategate. Read it. All of it. Get back to me afterwards.

As an addendum:

Being that I have the blessed luck of living somewhere in the same climate zone as Cocytus, especially these last two years, I am particularly enraged by the priggish nature of Greeniacs and their Socialist kin. I wish it was warming. Really warming. I wish, I beg, I pray that the global average was 5 degrees warmer.

It is only in the pursuit of gub'ment grants do these slimes of the earth, filthy hippies that they are, manage to "prove" AGW, and then also somehow "prove" that it would be a bad thing.

It is cold that kills. It is cold that destroys. It is cold that prevents life. It is the proximity to the freezing point that causes severe weather. Cold is the enemy. Oh, and cold in the near future given the recurrence of a massive solar minimum followed by a tepid peak.

This comment is a bit off-topic but here goes anyway. First, unfalsifiable statements are of the domain of the empirical sciences while undecidable statements are in the domain of mathematics. The examples given are from mathematics, so why does that show that Popperian falsifiability as a sine qua non is false? For the empirical sciences that is, which is the context in which the criterion makes sense.

The two are separate but relatable topics. My point was to make a comparison.

The insufficiency of the falsifiability criteria was argued in a way not related to uncertainty under the Quine-Duhem Thesis that theories are so interconnected that a falsification does not necessarily invalidate the current theory, but perhaps one of the sub-theories, so it is impossible to derive a clear answer from the experiment as to what exactly is false.

Back to undecidability: if something is undecidable, it is unfalsifiable, necessarily, although the reverse is not always true, since string theory is decidable, but not falsifiable, at least in practice. I did not mean to imply that the CH is really useful in laboratory science, just that any science that might depend on it (I have no idea what that might be) will find situations where it could turn out to give both positive and negative answers to experiments if CH is undecidable. Nevertheless, undecidability has not stopped math from progressing (that the Halting problem is undecidable has not stopped programmers from writing code) and neither has the inability to create a rigorously falsifiable test in every case stopped the march of science. That was my point. Too much weight is given to falsifiability. It is a useful heuristic, but not an absolute criteria.

Falsifiability is not a property of empirical science, only. Whenever one finds a counterexample in math, one has falsified the proof, so to speak. A criteria for falsifiability for any math proof is simply to find a counterexample. Math and science are so interrelated these days, with things like applied math, that an undecidable condition in the math will lead to a lack of falsifiability in the science. A classic example is the Selberg Trace Formula which is used to model the correlation between energy levels in quantum mechanics and classical periodic orbits in quantum chaos. The zeros of the Riemann Zeta Function map the zeros of the eigenfunctions in quantum chaos. IF Riemann's Hypothesis, that the real parts of the zeros of the Zeta Function are on the 1/2 line turns out to be undecidable, that might mean that there are some problems in quantum chaos that are undecidable and, hence, unfalsifiable. If one cannot tell that the formula one is using to make a measurement is true or consistent, then no matter how many false readings one gets based on that formula, it is not enough to discount the phenomenon.

Sorry...back to global warming. G Rodrigues and I can take our discussion out in the hallway.

The Chicken

Has anybody commented on the fact that if the government really believed in global warming that they would be paying the relatively modest amount necessary to install accurate temperature sensors in every town or even at regular intervals on the map couple to a wi-fi transmitter so that real-time information could be collected? That they are not tells me, unequivocally, they they really aren't serious about proving or disproving the hypothesis. A million data points per second could pretty easily give a great deal of information.

The Chicken

I think I'm going to bow out of the discussion. Ignorance and pride (my own) are like alcohol and barbiturates - a deadly combination.

The Chicken

Something I've stumbled onto as a result of this conversation ... Dr. Richard Muller of the Berkeley Earth Surface Temperature group, a climate-change skeptic and hockey-stick whistleblower - and a supporter of the work of fellow skeptic Anthony Watts - has launched a new project with the intent of settling the controversy. He presented some preliminary findings to congress earlier this year. You can read the report here:

http://berkeleyearth.org/Resources/Muller_Testimony_31_March_2011

To summarize: he addresses Tony's concerns about poor temperature station quality and, while admitting that station quality was poor for early data sets, concludes that the global warming estimates using these stations "were not exaggerated". He also confirms his belief that a percentage of global warming is human caused. Apart from bloggers and talk radio personalities, even the "denialists" admit the phenomenon is real. So I think it's fair to start the conversation on that premise.

Here is a quotation from a paper in which Muller criticizes the hockey stick (and good for him for that):

My own reading of the literature and study of paleoclimate suggests strongly that carbon dioxide from burning of fossil fuels will prove to be the greatest pollutant of human history. It is likely to have severe and detrimental effects on global climate.

If you think Muller is a "denialist," that pretty much shows that he isn't. He's not just talking about "a percentage" (which could be negligible). He's a firm believer in AGW. You can say that he's probably right or whatever, but you can hardly say that "denialists" agree that human burning of fossil fuels "is likely to have severe and detrimental effects on global climate." If they did, they wouldn't be "denialists."

"...for ou[r] grandchildren at least, most of the[m] then living will curse us and envy the dead..."

I think that's fairly likely. But not because of global warming.

If nihilism is inherited you might be onto something here.

Dr. Richard Muller wasn't a whistle-blower, he simply admitted that McIntyre was right in his evidence that the hockey stick was a fraud-- his whole shtick is "those on either side of me are anti-scientific." I wouldn't call him a supporter of Watt's work, even though he requested some to help him make the new BEST network. (He's also not a skeptic at all-- the BEST program was started with the basic assumption that "global warming is responsible for warming." http://wattsupwiththat.com/2011/02/11/new-independent-surface-temperature-record-in-the-works/ )

Incidentally, his willingness to give out preliminary results before they published so much as a single peer-reviewed paper, such a short time after the program got started, didn't win him any points in the science-not-a-circus category. Somebody that STARTS with the assumption that AGW is real, but is willing to take a swing at both sides, isn't a "climate-change skeptic."

Incidentally, can we stop with the "denialist" and "denier" stuff? It's basic bloody scientific skepticism, not the holocaust-evoking labels. I'm skeptical about yeti, Big Foot and the Mongolian Death Worm, but I don't get called a "Cryptid Denier" as a result.

Sorry, Foxfier, I'm probably responsible for using the term "denialist" first in this thread, and as a full-fledged skeptic, I used it tongue-in-cheek. I totally agree that it's been set up with offensive purposes and with the connection to Holocaust denial in mind. I was acquiescing in it much as I might in the same spirit call myself an "Islamophobe" without thereby agreeing that my opinions about Islam reflect an irrational fear. I suppose that was why I've used "warmist" in this thread as the parallel term--to put "ist" on the end of both, which I thought might seem equitable in an amusing way.

Just so tired of letting the other side choose the names and thus frame the debate.

I think that's fairly likely. But not because of global warming.

Yep. But my grandchildren will know that I opposed what was done to them.

Maybe this was mentioned above somewhere, but to be honest I haven't had time to read the 64 comments in entirity.

I read several years ago that while GW was real, the greatest (and may be only) contribution to GW was primarily in urban areas. The move from an agricultural society to an industrial society is a major factor.

And this move is accelerated by consumerism. I know the solution, but don't know how to get there.

"GISS is located at Columbia University in New York City. The institute is a laboratory in the Earth Sciences Division of NASA's Goddard Space Flight Center and a unit of the Columbia University Earth Institute."

Columbia? that settles it for me, they have to be right.

We should be reducing our dependency on fossil fuels anyway.

Direct hit! Regardless of where one comes down on the issue, conservation and diversification of energy sources is, at a minimum, prudent. And heading toward mandatory, I would argue.

It will take a while to achieve this, though. Moreover, it will require everybody's shibboleths to be examined, and in many cases, discarded.

Moreover, it will require everybody's shibboleths to be examined, and in many cases, discarded.

That should be intended as a joke, but I'm afraid it's not. The self-referential problem here is sort of glaring: What about reexamining and possibly discarding the shibboleth that we should be reducing our dependency on fossil fuels?

Fine--you scored gotcha points. But not much more.

Given what we know about American oil reserves and the development of such sources as oil sands/shale, what's the argument for the status quo? I'm not particularly fond of helping to endow another madrassa every time I gas up. But, as they say, mileage varies.

http://www.theregister.co.uk/2011/06/14/ice_age/

Seemed appropriate.

Well, Dale, if you really want to know what real energy independence means, I suggest you look into applications of depolymerization of anything from sewage to trash to coal as starters, then move onto thorium based nuclear power.

American oil reserves are also much larger than the environmentalist lobby (using intentionally outdated and biased data) would have you believe. By all means, let's put the screws to OPEC, but let's do it with energy that works, as opposed to worthless windmills and carbon-indulgence trading.

As a side not, I really wish global warming WAS for real. Sadly, it is not and Wisconsin remains in the same climate zone as Cocytus.

And as I mentioned above, it's not as though we're exactly _rushing_ to make the fullest use of all our domestic oil sources. Far from it. I'd rather fund Sarah Palin's home state than a Madrassa any day. :-)

The greater problem is trying to make new energy sources enforced by the UN. And that looks like where we're heading to.
It's all well that developed countries should change their habits if they really think they are melting the world.
The real problem arises for the African or Asian city that is still thinking how it will get some eletric light for the first time. They know they have some coal, and luckily some petroleum, but he asks the American embassador for some help with setting up a power plant and discovers everybody think he should use solar panels and wind turbines that are as costly as he is poor.