Thursday, May 3, 2007

Reply to James Annan

James Annan posted this. Here is my reply:

C02 has almost no effect? Implied belief that C02 has a negligible effect? Hardly. Straw man, James.

My guest post on Backseat Driving explains why I bet:
  • The evidence in favor of carbon emissions as the main cause of global warming has weakened substantially, especially now that higher-resolution ice core data shows that past temperatures rises occurred before atmospheric carbon increases. This weakening is not widely recognized.
  • The political climate is running strongly in favor of carbon emissions as the main cause.
So now is a good time to bet against carbon emissions, at the best odds available.

Here is part of what I said: "I think that it is possible that carbon emissions are the dominant cause of global warming, but in light of the weakening evidence I judge that probability to be about 20% rather than almost 90% as estimated by the IPCC."

From which, given that global temperatures are currently increasing at about 0.2C/decade, one would presumably conclude that I believe that atmospheric carbon is responsible for at least a 20% * 0.2C/decade = 0.04C/decade temperature rise, at least in a probabilistic sense. Hardly negligible. In any case, what I believe (this year!) is: atmospheric carbon has a non-negligible effect, but is probably not the main cause of global warming.

James if I read your post correctly, we would probably agree that:
  • There is a warming effect due to the extra carbon we humans have put into the atmosphere.
  • There are other causes for the global temperature to change, including human aerosols and maybe cosmic rays.
  • The current warming, about 0.2C/decade, is due to the extra carbon and the other effects.
  • If warming due to extra carbon exceeds 0.15C/decade, carbon is the main cause of global warming and I will lose my bets.
  • If warming due to extra carbon is less than 0.05/C decade, carbon is not the main cause of global warming and I will win my bets (provided the other causes reverse themselves in the next decade).

And where we disagree is in what the contribution of the extra carbon is:
  • I think it likely that the current global warming is mainly due some other cause, which may subside in time, but that the warming due to the extra carbon will continue (since we keep increasing the extra carbon). The increase in extra carbon would put an upward bias on the natural temperature variability, an upward-sloping channel on a graph of global temperature. We might want to curb carbon emissions some day.
  • You presumably think it likely that current global warming is mainly due to the increasing extra carbon, and that global warming will therefore continue at much the same rate, with some relatively minor fluctuations due to natural variability, until we curb carbon emissions substantially.

So let's focus on quantifying the warming effect of the extra carbon. I agree with you, this is crucial.

You quote a ballpark warming figure due to a doubling of atmospheric carbon, ignoring feedback and clouds but including extra water vapor, of 2C. Can you explain this figure, at a level of physics in say Resnick and Halliday or Feynman's Lectures, so that it is convincing to an interested person with a modest scientific background? Can you then add in all the complications of a real atmosphere and feedbacks? What are the lowest and highest values of warming due to extra carbon that come out of that process? I'd be delighted if someone did that. If the minimum possible warming was not small, the assumptions were watertight, and the data inputs all incontrovertible, then you would persuade a lot of skeptics like me over to your side.

Alas, the only figures offered to me are from opaque models. I find that unsatisfactory of course, because I do not know the assumptions built into those models, the input data, or how the calculations were made. How confident can we be that nothing was omitted (after all, we only discovered global dimming a few years ago)? And the intrinsic difficulty of convincing me by calculation and model is massive because they are so complex -- I haven't got the time to look over every detail and check every assumption.

The warming effect of the extra carbon, to date, is presented as an argument by authority. So naturally we wonder about the political situation of those authorities. All a bit unsatisfactory and unconvincing.

What we need is to measure the warming due to extra carbon, to tie it in with some observations. How are we going? Presumably the parts of the atmosphere with the extra carbon should be showing a warming trend. There has been a big effort trying to measure these changes, with balloons since 1958 and with satellites since 1979.

As we all know, science works by falsifying hypotheses by experiment or observation. (Warning: The following point is relevant for the process of science, as I will note below. It is not offered as an argument against carbon-causes-warming.) As pointed out in the first substantial point in The Great Global Warming Swindle, the observations collected on atmospheric warming contradicted the greenhouse hypothesis. As noted here however, due to better data and corrections to old data, by 2006 those observations no longer rule out the possibility of greenhouse warming in the atmosphere (except that in the tropics the observed results still contradict current climate models in some important respects). That is, until 2006 the results of direct measurements on warming in the atmosphere were falsifying the hypothesis that the extra carbon was causing warming! And now the data does not contradict the hypothesis. But (to get back to the main point), the data doesn't really allow us to directly measure the amount of warming due to the extra carbon.

So, in the absence of good observational data, we simply don't know how much the extra carbon is warming the world.

Now let's turn to cosmic rays. Svensmark's experiments in 2006 put the cosmic ray theory on roughly the same basis as the atmospheric carbon theory: we know they work in a laboratory, but the jump to quantifying that in our atmosphere is a bit hard. (Well ok, we will know a lot more about how cosmic rays perform in test situations after the CERN experiments, scheduled for 2010.)

Research into the cosmic-ray possibility should be given a fair go. Will the proponents of carbon as the main cause resist looking at alternative causes? Will they give other possibilities a thorough examination? Bear in mind that until last year the atmospheric observations noted above appeared to falsify atmospheric carbon as the cause, yet we persisted with blaming global warming on carbon emissions. So even if we find some data that falsifies cosmic rays, we should continue researching them in a well-funded and enthusiastic fashion for a good few years -- if only to be fair.

Well James, in the absence of good observational data about how much warming is caused by the extra carbon, we can each demand that the burden of proof is on the other -- either because of the precautionary principle, or because of the competing principle of not meddling unless you're sure of the cause.

But, having seen the way we got here and seeing the political climate, I bet against warming intensifying. And, in the groundbreaking first major bet on global warming, you bet the other way. Bring on more evidence!

(Shorter url for this post: http://tinyurl.com/24zt52)

6 comments:

C W Magee said...

If you want to make models less opaquem you can read the paper for the GISS 2 here:
http://pubs.giss.nasa.gov/abstracts/1983/Hansen_etal.html

And download the PC version to play with yourself here:
http://edgcm.columbia.edu/

My opionion of cosmic ray climatology, as a mineralogist who sometimes helps set up paleoclimate analyses but doesn't really study climate, is here:
http://lablemminglounge.blogspot.com/2006/11/jan-veizens-cosmic-ray-climatology.html

Heiko said...

I can see why you'd think that the IPCC argue from authority, well, they are in practise, because I can't follow the calculation for forcing, and it seems neither can you or some 99.99% of the world's population.

However, your argument that the IPCC must be wrong with their climate sensitivity estimate of 1.5C to 4.5C seems pretty flimsy to me, I mean you haven't presented any alternative calculation of how much warming a doubling of CO2 should cause.

I don't think it's enough to say "I don't understand the models, so the IPCC argue from authority", it's not enough when you want to prove that the IPCC estimate is wrong.

I certainly wouldn't bet money against the IPCC, well actually I don't like betting, so I wouldn't even want to take the other side of the bet ;-)

Heiko said...

And while I trust the IPCC, and think a bet against you has got pretty good odds on that basis, it's still quite possible that you win in spite of climate sensitivity being 3C, say because of aerosol caused cooling.

guthrie said...

I have no idea how you can say:

"So, in the absence of good observational data, we simply don't know how much the extra carbon is warming the world."

by dismissing the observational data that has been presented, error checked, and all the rest of it. Thus, I think your being a bit silly.

David Evans said...

Heiko - The essence of the point I was clumsily making is that science, historically, has rarely progressed by calculations and models. Rather it has been by repeatable experiments and by observation.

You know the usual cautionary litany of theories held by authorities that turned out to be spectacularly wrong: heavier-than-air flight not possible, sun orbits the earth,...

In this context, I am not saying the IPCC carbon sensitivities are wrong necessarily, I'm saying that I can't see or follow their calculations -- and anyway I wouldn't trust them to have not left out something significant. But if you had clear observational data, I would of course by convinced in an instant.

David Evans said...

Heiko - Yes, as you observe, the terms of the bet are imperfect. Either of us win despite being wrong. But we deliberately set the critical level at 0.15C/decade to minimize that possibility. I figured temperature was about the only unambiguous measure we could bet on.