my skepticism of climate change has to do with its underlying cause, not its existence.

NASA has reliable data (not commonly not reported in mainstream media) that all of the planets in our solar system are heating up at the same rate. makes you wonder why?

perhaps you might shed some light on this phenomenon. naturally, i have my thoughts.

Anyway, not much to disagree with because what you say is nearly a tautology: changing the material properties changes the response. Since what that response actually is remains obscure it can be stated even more simply: changing the equation changes the equation.

It’s funny, I am at the same time both more optimistic of our ability (continually improving) to model turbulence, and less convinced of the usefulness of the spherical cow climate model than you.

The primary difference between me and you is in our (un)willingness to leap from physics to moral imperative. Any shared understanding of climate processes we may have will never change that.

It would help if you would at least acknowledge the point — the radiative balance model is sufficient to at least show that there’s an effect, and suspect that it could be quite significant. The details of what changes occur, at high resolution, and an accurate predsixtion of magnitude, is the domain of future work, and remains to be discovered. But the threat of a large magnitude effect is established.

I put forward the basic radiative balance model as a starting point, and show that with the addition of greenhouse gases there’s no way around making the world either hotter or grayer. It’s a pretty simple argument —

You may be interested in this. Pressing a simple model of a complex system too hard is a good way to convince yourself of nonsense. Spherical cows are pretty useful rhetorical tools though ; – )

Tom said:

…let’s get back to the issue: you do need an “accurate” model to arrive at an accurate prediction. Which models do you assess as being accurate?

Part of my lit review has been focused on climate model verification and validation. I tend to focus on the verification part in online discussions because that’s just math, and it’s less likely to devolve into heated polemics. What I found was pretty surprising to me: none of the climate models are yet capable of demonstrating grid converged solutions for even some globally averaged functionals. Initial work on doing that sort of demonstration by increasing resolution was “discouraging.” I think such a demonstration is just a matter of time and improved technique though.

While we don’t have a complete model of the person, we can predict that consciousness will fade — perhaps fatally. The prediction can be accurate, despite a completely lacking or incorrect model of the system. We might model a conscious being as the work of homunculi that strike in intemperate weather.

–

As such, I put forward the basic radiative balance model as a starting point, and show that with the addition of greenhouse gases there’s no way around making the world either hotter or grayer. It’s a pretty simple argument — and that’s the point — I think people are getting lost in the details. The first point to get across that we’re changing the planet — it is not infinite. The second is that we seem to be changing it dramatically.

I feel like any such discussion must address the fact that, globally, phytoplankton has declined by 40% in the past century. That’s 40% of the base of the food pyramid for all ocean life. That’s 20% of the plant life in the world, and accounts for 20% of the oxygen we breathe. Whatever is causing it, something significant is happening to the ecosystems of earth.

http://wormlab.biology.dal.ca/ramweb/papers-total/Boyce_etal_2010.pdf

He also talks about barriers to communication between different specialists. I think this email fiasco, if nothing else, highlights the importance of knocking down those barriers.

BTW, the links about the ‘truth’ of the emails are based on RealClimate.org’s ‘analysis’, and those guys don’t exactly come across as bastions of academic integrity in the leaked emails either. The responses by Zorita and Curry are much more what I would expect from people interested in getting back to the science.

The whole thing reeks. I am getting worried that things will degenerate into the same kind of partisan, or worse, tribal conflicts that make up domestic politics and international conflicts. The Palestine/Israel conflict comes to mind, and understanding is only worsening as attention is directed towards slights, attacks, and atrocities of the past.

It will take a lot of pioneering work to recover a reasonable scientific effort and debate.

I downloaded the email zip file a while ago; there are lots of very unprofessional and unethical things these guys were colluding on. I don’t even care about the ‘hide the decline’ email; the ones about obfuscating data and methods, manipulating editorial personnel on journals, making sure that the ‘right’ folks are reviewing the in-crowd’s papers, and redefining the definition of what counts as peer review. Those are the things that bother me (aside from all of the assorted buffoonery those guys displayed that was unrelated to technical competence and professional integrity).

I’d encourage you to read them yourself. If you have a technical / research background at all you will be appalled.

For your information, the “chicanery” and allegations about deception are overblown. Of all the e-mails taken from the climate research center that got hacked, two sentences were cherry picked and interpreted out of context to show some sort of conspiracy. No amount of research conspiracy can bring back the glaciers that have melted, ice shelves have broken off, nor change the fact that CO2 is a greenhouse gas, nor change the fact that its concentration and the concentration of other potent GHGs are increasing in our atmosphere due to human activity. All the counter-examples cited of temperature falling only shows that the effect on global temperatures is not a monotonic function, not that the long term trend since industrialization hasn’t been an increase in temperature.

Here is one explanation given as a counterpoint to the mis-interpretation of those e-mails from someone familiar with the material.
_________________________________
The email “trick of adding in the real temps to each series … to hide the decline” has nothing to do with “hiding evidence of global cooling.” The email is referring to the tree ring divergence problem, a well-known and actively studied phenomenon where dendrochronological proxy reconstructions diverge from instrumental data starting at around 1950. In other words, studying the relative width of tree rings from very old trees can give reconstructions of the climate during those years, except this stops working for trees in Northern-hemisphere forests after 1950. There are many hypotheses as to why this occurred, but there is no definite answer yet. A good overview of the problem can be found here:
http://www.ldeo.columbia.edu/%7Eliepert/pdf/DArrigo_etal.pdf

Direct instrumental data clearly shows a warming trend since 1950, as well as other proxy reconstructions, however because of the divergence problem, tree ring proxies from the Northern hemisphere do not correlate with any of these. Therefore, when creating the chart of proxy compilations, tree ring data was only included up until 1960. The chart is here:
http://commons.wikimedia.org/wiki/File:1000_Year_Temperature_Comparison.png

The lightest-blue line is the tree-ring proxy. Notice that it correlates quite well with all the other proxies, although it stops at 1960. If the omitted data was included, it would have shown a clearly juxtaposed decline. This is what “hide the decline” referred to.

If you refer to the above pdf, on page 3 they have a chart that shows the full proxy, where the decline can be seen. It also shows how the proxy diverges with instrumental temperature records. No data is being hidden from the public, it is readily available (it took me all of 2 minutes to find it).

Lastly, the “trick” from the email was referring to plotting instrumental data along with the proxy data. This is a very common practice to show that instrumental data correlates with proxy data. In the chart above, instrumental data is the solid black line, which as can be seen, follows the exact same pattern as the most recent proxy data.
_______________________________________________
See this as well:
http://www.treehugger.com/files/2009/11/truth-hacked-climate-email-controversy.php

Your right about skepticism revealing a specific state of knowledge, but not about the coherence of that state. I tend to take a more Bayesian view of the reasons for disagreement, rather than attributing motivations to folks I disagree with, like “abdication of human responsibility”.

With the recently revealed chicanery by some prominent alarmists, are prior probabilities for deception all that unreasonable?

The bottom line is that the way we keep from fooling ourselves (and others) is by being as open and honest about the raw data, and the manipulations we perform on the data as we can. Being less than transparent only tends to enforce the prior probability for deception.

One of the biggest skeptics with a scientific background is Fred Singer. His skepticism took this progression:
* In the 90′s, he claimed that the globe was not warming at all.
* More recently, he claimed that it is warming, but it is not our fault and has nothing to do with CO2.
* Now, he has co-authored and published a book claiming that the world is experiencing “unstoppable global warming” due to a natural cycle that happens every few thousand years and that there is nothing we can do about it.
* His co-author is now touring about arguing that we are experiencing global cooling.

What this tolerance for gross inconsistency (not warming/warming but not our fault/ global cooling) suggests is that there is an abdication of human responsibility for climate change; the skeptics have embraced a number of contradictory stances towards climate change with no apparent concern for the contradictions as long as their stance does not admit the one thing unacceptable to them–holding humans responsible. Someone with this attitude cannot be convinced by reason and evidence because they’ve concluded (at an emotional level) that they will not be held responsible.

What I’m talking about here is dishonest skepticism. Honest skepticism is entitled to debate over reason and evidence, but dishonest skepticism can only be contained; once someone embraces it, nothing changes their mind.

Thanks! It’s been an enjoyable conversation for me as well. I do want to write a more detailed follow up to your further points and questions — it will have to wait until I’m past some deadlines. But thank you very much for your links :-)

Cheers,

Danielle

First, thanks for the provocative post; it has really gotten me much more into this interesting literature. The interface between computational physics, decision theory and public policy is an exciting one.

Second, Yale 360 has an interesting article on the economics of climate change policy you might be interested in. I wrote up a little response as well.

This points to the larger issue of ‘decisions under uncertainty’, The MIT group seems to be the one out front on applying those tools to this field. Good food for thought for anyone making decisions based on physical predictions (isn’t that most of us?).

Cheers.

I think if this were the case you’d see a lot of the rancour disappear, people aren’t reacting to climate change science because they hate science, they react to unvalidated models being used to fear monger just a bit and as a justification for massive expansion of government regulation/taxation.

Your bombs and jet engines examples are near and dear to my heart, and illustrate what I said earlier. We don’t (or at least it’s not best practice to) bend metal on those types of hardware development programs based on model predictions that haven’t been calibrated to the reality of interest (blades in cascades, components in rigs, engines in static test stands, flight hardware). That validation process doesn’t occur by observing engines in the wild, we design experiments so that we protect ourselves from being fooled by random/uncontrolled variation and correlation of the inputs (this is difficult to do with the earth). You won’t see Pratt & Whitney, GE or Rolls Royce making large capital expenditures based on colorful fluid dynamics unconstrained by experimental benchmarking (and I say this as an unabashed CFD nerd).

Having said that, I agree that the “we can’t predict every turbulent eddy, so models are useless” argument is silliness. We can get good wags on the time averages without getting every little eddy right (but this requires turbulence model calibration/validation).

It seems (at least to me) like the folks who are closest to the actual measurements of the climate (satellites, radiosondes, etc) tend to also be the ones who are quite sceptical of the tales of impending doom from the modellers. Am I mistaken?

Danielle also said: “… you don’t even need a general appraisal of a situation to ascertain risk worthy of action.”

I disagree, situational awareness is pretty key when you are in a “shoot, move or communicate” sort of predicament. Running open loop is generally not a recipe for success in a risky, complex environment. YMMV.

What’s wrong with a little skepticism? There is a whole lot of bs passed off as science regarding climate change. In particular the models which claim to predict the temperature centuries form now by modeling a chaotic system with boxes 100mi in width using parameters with +-10% error _each year_ is ridiculous.

Plus what do you want us to do? Spend trillions we don’t have to try and reduce our oil consumption by a fraction? Other countries will just burn anything we don’t.

—

To which I reply

—

Not if alternatives achieve unsubsidized competitiveness, which is precisely what some are trying to achieve.

The internet is a chaotic system if you choose packets as your unit of granularity, and yet this message still came over the wire intact. The wiring inside a bomb impacted by an outside force is a chaotic system with plenty of uncertainty as to what’s inside, but you would still be unwise to kick it. Gas flow in jet engines is a highly turbulent and chaotic system with measured temperature variation from the best models exceeding in many cases 20%, and yet jets will still fly.

You don’t need a perfect model to arrive at an accurate prediction, you don’t need a terribly accurate prediction in order to arrive at a general appraisal of a situation, and you don’t even need a general appraisal of a situation to ascertain risk worthy of action.

Frankly, it should be probably be enough to say “we know that most of the energy the Earth receives is from the sun, and we know that most of the energy is dissipates is through infrared. We also know that the Earth is an extremely complex, strongly coupled, and chaotic system, in which the details of every component can never be known, and being able to predict with a high level of confidence what the effects of any major change would be is unlikely. Maybe we should probably stop fucking with the Earth’s primary means of cooling?”

There is nothing wrong with skepticism, per se. Skepticism is a positive, and it is a very worthy thing to root out the bs from the science. But it must be weighed against two things: the need to come to some decisions in a limited timeframe, and the fact that people often managed to justify to themselves, using any means at their disposal, their continued action or inaction. Too often ‘climate change skeptics’ claim that due to unacceptable uncertainty, they’re going to continue doing what they’re doing — or — that it isn’t happening at all — or — that it’s just part of the natural variation, when simple physics tells us that if you significantly modify a primary means of forcing a system then at least something about it must change.

Your ‘special circumstances’:

1. … cyclic climate change irrelevant to long term …, Agreed, that’s just talking heads muddying the water.

2. … earth becoming more reflective …, true, but I think some of the other feedback mechanisms are more interesting and haven’t been modeled very accurately yet, for instance Dr Spencer’s work on precipitation systems:
http://www.nypost.com/p/news/opinion/opedcolumnists/not_that_simple_gqs0XKDYmQhnILuB6aM9hK
(disclaimer: I used to be Roy’s neighbour)

3. … rational to disagree about the extent of the problem …, Agreed, I guess you could categorize me as a “problem” skeptic rather than an “effect” skeptic. Thanks, that’s a useful distinction.

4. Since you mentioned Dyson earlier, his review of a couple books is interesting: http://www.nybooks.com/articles/21494

You said, “*Any* physical prediction, no matter how careful the work, has, necessarily, some uncertainty. This case isn’t just a matter of academic curiosity – we cannot just hold our breath and see what happens.” I agree with the first half of your statement. Given the uncertain predictions of the future, any policy response ought to balance the expected risks/rewards. You may disagree with the economic analysis that Nordhaus has done (see Dyson’s book review), but it is telling that the policy suggestions to combat global warming that get a lot of press (Stern’s and Gore’s suggestions) are an *order of magnitude worse*, economically speaking, than doing nothing, and doing nothing is only slightly worse than the optimal strategy of a gradually increasing carbon tax. So, I think the second half of your statement doesn’t follow, doing nothing might be close to the right choice. Nordhaus could certainly be wrong, but at least it’s an honest effort to make reasoned policy choices that take into account the cost of our actions and our uncertainty about the future.

I think the ‘special circumstance’ that you left out, that is most fundamentally important, is that we can’t validate climate models to the same level as other models. We can’t perform controlled experiments on the reality of interest. This means that, unlike most of our efforts in applied physics which are quite successful, we won’t have *quantitative* estimates of the uncertainty of our predictions. We won’t know which modeling choices were well founded and which were just a mistake of history (biased by the available observations, a sort of selection bias).

By the way, I see that you do take validation seriously, your post Keeping Prediction Honest is a good one too.

Thank you for your comments. I do mention, explicitly, that the black box might turn more reflective: to my mind this is the most plausible objection to the climate models, since those feedback effects are, as you point out, nonlinear and difficult to predict. But most skeptics don’t mention this, and point out climate’s cyclical nature, instead. Furthermore, an increased albedo means a dimmer world, and probably with more clouds. It is reasonable to disagree that this is necessarily a bad thing, but not to pretend that there are no consequences.

It is in fact true that most of the light of the sun is turned to infrared — the sun spectrum albedo of the earth is, on average, about 0.3. For short periods the albedo can vary considerably between about 0.05 and 0.6, but it does stay most of the time below 0.5. [1][2]

The other issue is that the overall, full spectrum albedo of Earth have been measured, by the satellite CERES, as declining, recently, though they suggest it is due to variation in aerosols, which are more sensitive to market forces, technology, and weather than most other pieces.[2][3]

I believe strongly in skepticism. I believe in the validation of models, piecemeal and together. And I believe that one should bend over backwards to not fool oneself. I also believe that this is, earnestly, carefully, extensively, being done. And in this case, there are special circumstances.

First, the cyclical climate change that is so often bandied about by ‘skeptics’ is quite irrelevant to the overall, long term problem.

Second, that the other holes in the reasoning I gave above — primarily the Earth becoming more reflective are already being questioned and checked and measured by many projects and scientists.

Third, saying that the Earth will react to the changes by becoming more reflective has substantial implications for life on Earth. Saying that radiation will increase due to an increase in the four power average temperature implies increased temperatures *or* increased extreme weather events. When people say that the Earth will adjust itself to the greenhouse gases by a variety of natural phenomena, they are not climate change skeptics, they are climate *problem* skeptics. It’s dishonest to claim otherwise. It’s a rational conversation to disagree about the extent of that as a problem, but it’s not honest to claim that nothing’s happening and nothing will happen.

Fourth, we only have one Earth. *Any* physical prediction, no matter how careful the work, has, necessarily, some uncertainty. This case isn’t just a matter of academic curiosity – we cannot just hold our breath and see what happens. If a credible, careful, extensive effort to produce a range of predictions shows that some significant chance of a catastrophic event (be it Greenland melting, the permanent loss of coral reefs or rainforests, or climate change causing rapid evolution and spread of diseases and vectors,) we had better come up with some reasoning for why the chance isn’t significant or why the problems aren’t catastrophic. Or we should act.

[1] – http://www.spenvis.oma.be/spenvis/ecss/ecss06/ecss06.html#_TOC641
[2] – http://earthobservatory.nasa.gov/IOTD/view.php?id=5484
[3] – http://www.sciencemag.org/cgi/content/abstract/308/5723/825

I think many technically minded folks in engineering and hard science are used to validating the predictions of their models with designed experiments. Going from observational studies and un-validated model predictions (we can’t do designed experiments on the earth) to hysterics about eliminating hydrocarbon based energy sources to “save life as we know it” seems very unscientific to them.

I’m sure you’d expect the engineers of your start-up to validate their scooter-performance-models by testing components and prototypes before using them to justify spending bunches of money on materials, tooling and production facilities. At least the same level of healthy “skepticism” ought to be applied to climate change predictions. Feynman’s first principle comes to mind: “The first principle is that you must not fool yourself – and you are the easiest person to fool.” His take on scientists taking extra care to not mislead the lay public with hyperbole is also germane

I like Dyson’s view of the situation, and his carbon eating tree idea is interesting. Solutions like that are far more inspiring than what normally gets bandied about (taxing carbon, how boring!). Maybe you could even make a profit from solar-powered carbon sequestering trees, that would actually be sustainable, and wouldn’t require world government or heavy taxation of productive economic activity (or any pseudo-religious hysterics about saving the planet; I know, I know, how boring!).

Nice essay; good luck with your start-up!