Guest Post by Willis Eschenbach
Under the radar, and un-noticed by many climate scientists, there was a recent study by the National Academy of Sciences (NAS), commissioned by the US Government, regarding climate change. Here is the remit under which they were supposed to operate:
Specifically, our charge was
1. To identify the principal premises on which our current understanding of the question [of the climate effects of CO2] is based,
2. To assess quantitatively the adequacy and uncertainty of our knowledge of these factors and processes, and
3. To summarize in concise and objective terms our best present understanding of the carbon dioxide/climate issue for the benefit of policymakers.
Now, that all sounds quite reasonable. In fact, if we knew the answers to those questions, we’d be a long ways ahead of where we are now.
But as it turned out, being AGW supporting climate scientists, the NAS study group decided that they knew better. They decided that to answer the actual question they had been asked would be too difficult, that it would take too long.
Now that’s OK. Sometimes scientists are asked for stuff that might take a decade to figure out. And that’s just what they should have told their political masters, can’t do it, takes too long. But noooo … they knew better, so they decided that instead, they should answer a different question entirely. After listing the reasons that it was too hard to answer the questions they were actually asked, they say (emphasis mine):
A complete assessment of all the issues will be a long and difficult task.
It seemed feasible, however, to start with a single basic question: If we were indeed certain that atmospheric carbon dioxide would increase on a known schedule, how well could we project the climatic consequences?
Oooookaaaay … I guess that’s now the modern post-normal science method. First, you assume that there will be “climatic consequences” from increasing CO2. Then you see if you can “project the consequences”.
They are right that it is easier to do that than to actually establish IF there will be climatic consequences. It makes it so much simpler if you just assume that CO2 drives the climate. Once you have the answer, the questions get much easier …
However, they did at least try to answer their own question. And what are their findings? Well, they started out with this:
We estimate the most probable global warming for a doubling of CO2 to be near 3′C with a probable error of ± 1.5°C.
No surprise there. They point out that this estimate, of course, comes from climate models. Surprisingly, however, they have no question and are in no mystery about whether climate models are tuned or not. They say (emphasis mine):
Since individual clouds are below the grid scale of the general circulation models, ways must be found to relate the total cloud amount in a grid box to the grid-point variables. Existing parameterizations of cloud amounts in general circulation models are physically very crude. When empirical adjustments of parameters are made to achieve verisimilitude, the model may appear to be validated against the present climate. But such tuning by itself does not guarantee that the response of clouds to a change in the CO2 concentration is also tuned. It must thus be emphasized that the modeling of clouds is one of the weakest links in the general circulation modeling efforts.
Modeling of clouds is one of the weakest links … can’t disagree with that.
So what is the current state of play regarding the climate feedback? The authors say that the positive water vapor feedback overrules any possible negative feedbacks:
We have examined with care ail known negative feedback mechanisms, such as increases in low or middle cloud amount, and have concluded that the oversimplifications and inaccuracies in the models are not likely to have vitiated the principal conclusion that there will be appreciable warming. The known negative feedback mechanisms can reduce the warming, but they do not appear to be so strong as the positive moisture feedback.
However, as has been the case for years, when you get to the actual section of the report where they discuss the clouds (the main negative feedback), the report merely reiterates that the clouds are poorly understood and poorly represented … how does that work, that they are sure the net feedback is positive, but they don’t understand and can only poorly represent the negative feedbacks? They say, for example:
How important the overall cloud effects are is, however, an extremely difficult question to answer. The cloud distribution is a product of the entire climate system, in which many other feedbacks are involved. Trustworthy answers can be obtained only through comprehensive numerical modeling of the general circulations of the atmosphere and oceans together with validation by comparison of the observed with the model-produced cloud types and amounts.
In other words, they don’t know but they’re sure the net is positive.
Regarding whether the models are able to accurately replicate regional climates, the report says:
At present, we cannot simulate accurately the details of regional climate and thus cannot predict the locations and intensities of regional climate changes with confidence. This situation may be expected to improve gradually as greater scientific understanding is acquired and faster computers are built.
So there you have it, folks. The climate sensitivity is 3°C per doubling of CO2, with an error of about ± 1.5°C. Net feedback is positive, although we don’t understand the clouds. The models are not yet able to simulate regional climates. No surprises in any of that. It’s just what you’d expect a NAS panel to say.
Now, before going forwards, since the NAS report is based on computer models, let me take a slight diversion to list a few facts about computers, which are a long-time fascination of mine. As long as I can remember, I wanted a computer of my own. When I was a little kid I dreamed about having one. I speak a half dozen computer languages reasonably well, and there are more that I’ve forgotten. I wrote my first computer program in 1963.
Watching the changes in computer power has been astounding. In 1979, the fastest computer in the world was the Cray-1 supercomputer. In 1979, a Cray-1 supercomputer, a machine far beyond anything that most scientists might have dreamed of having, had 8 Mb of memory, 10 Gb of hard disk space, and ran at 100 MFLOPS (million floating point operations per second). The computer I’m writing this on has a thousand times the memory, fifty times the disk space, and two hundred times the speed of the Cray-1.
And that’s just my desktop computer. The new NASA climate supercomputer “Gaea” shown in Figure 1 runs two and a half million times as fast as a Cray-1. This means that a one-day run on “Gaea” would take a Cray-1 about seven thousand years to complete …
Now, why is the speed of a Cray-1 computer relevant to the NAS report I quoted from above?
It is relevant because as some of you may have realized, the NAS report I quoted from above is called the “Charney Report“. As far as I know, it was the first official National Academy of Science statement on the CO2 question. And when I said it was a “recent report”, I was thinking about it in historical terms. It was published in 1979.
Here’s the bizarre part, the elephant in the climate science room. The Charney Report could have been written yesterday. AGW supporters are still making exactly the same claims, as if no time had passed at all. For example, AGW supporters are still saying the same thing about the clouds now as they were back in 1979—they admit they don’t understand them, that it’s the biggest problem in the models, but all the same but they’re sure the net feedback is positive. I’m not sure clear that works, but it’s been that way since 1979.
That’s the oddity to me—when you read the Charney Report, it is obvious that almost nothing of significance has changed in the field since 1979. There have been no scientific breakthroughs, no new deep understandings. People are still making the same claims about climate sensitivity, with almost no change in the huge error limits. The range still varies by a factor of three, from about 1.5 to about 4.5°C per doubling of CO2.
Meanwhile, the computer horsepower has increased beyond anyone’s wildest expectations. The size of the climate models has done the same. The climate models of 1979 were thousands of lines of code. The modern models are more like millions of lines of code. Back then it was atmosphere only models with a few layers and large gridcells. Now we have fully coupled ocean-atmosphere-cryosphere-biosphere-lithosphere models, with much smaller gridcells and dozens of both oceanic and atmospheric layers.
And since 1979, an entire climate industry has grown up that has spent millions of human-hours applying that constantly increasing computer horsepower to studying the climate.
And after the millions of hours of human effort, after the millions and millions of dollars gone into research, after all of those million-fold increases in computer speed and size, and after the phenomenal increase in model sophistication and detail … the guesstimated range of climate sensitivity hasn’t narrowed in any significant fashion. It’s still right around 3 ± 1.5°C per double of CO2, just like it was in 1979.
And the same thing is true on most fronts in climate science. We still don’t understand the things that were mysteries a third of a century ago. After all of the gigantic advances in model speed, size, and detail, we still can say nothing definitive about the clouds. We still don’t have a handle on the net feedback. It’s like the whole realm of climate science got stuck in a 1979 time warp, and has basically gone nowhere since then. The models are thousands of times bigger, and thousands of times faster, and thousands of times more complex, but they are still useless for regional predictions.
How can we understand this stupendous lack of progress, a third of a century of intensive work with very little to show for it?
For me, there is only one answer. The lack of progress means that there is some fundamental misunderstanding at the very base of the modern climate edifice. It means that the underlying paradigm that the whole field is built on must contain some basic and far-reaching theoretical error.
Now we can debate what that fundamental misunderstanding might be.
But I see no other explanation that makes sense. Every other field of science has seen huge advances since 1979. New fields have opened up, old fields have moved ahead. Genomics and nanotechnology and proteomics and optics and carbon chemistry and all the rest, everyone has ridden the computer revolution to heights undreamed of … except climate science.
That’s the elephant in the room—the incredible lack of progress in the field despite a third of a century of intense study.
Now me, I think the fundamental misunderstanding is the idea that the surface air temperature is a linear function of forcing. That’s why it was lethal for the Charney folks to answer the wrong question. They started with the assumption that a change in forcing would change the temperature, and wondered “how well could we project the climatic consequences?”
Once you’ve done that, once you’ve assumed that CO2 is the culprit, you’ve ruled out the understanding of the climate as a heat engine.
Once you’ve done that, you’ve ruled out the idea that like all flow systems, the climate has preferential states, and that it evolves to maximize entropy.
Once you’ve done that, you’ve ruled out all of the various thermostatic and homeostatic climate mechanisms that are operating at a host of spatial and temporal scales.
And as it turns out, once you’ve done that, once you make the assumption that surface temperature is a linear function of forcing, you’ve ruled out any progress in the field until that error is rectified.
But that’s just me. You may have some other explanation for the almost total lack of progress in climate science in the last third of a century, and if so, all cordial comments gladly accepted. Allow me to recommend that your comments be brief, clear and interesting.
PS—Please do not compare this to the lack of progress in something like achieving nuclear fusion. Unlike climate science, that is a practical problem, and a devilishly complex one. The challenge there is to build something never seen in nature—a bottle that can contain the sun here on earth.
Climate, on the other hand, is a theoretical question, not a building challenge.
PPS—Please don’t come in and start off with version number 45,122,164 of the “Willis, you’re an ignorant jerk” meme. I know that. I was born yesterday, and my background music is Tom o’Bedlam’s song:
By a host of furious fancies Whereof I am commander With a sword of fire, and a steed of air Through the universe I wander. By a ghost of rags and shadows I summoned am to tourney Ten leagues beyond the wild world's end Methinks it is no journey.
So let’s just take my ignorance and my non compos mentation and my general jerkitude as established facts, consider them read into the record, and stick to the science, OK?