It’s based on several studies…and indeed I think if you looked at the studies you see that there is a pretty big range but they all have a best estimate of less than 0.12% of GDP growth per year.
Most interesting, I’d never considered that quirk.
Consider my ignorance fought … much appreciated.
w.
intention: Thanks for your thoughtful reply (and glad I could clear up that significant figures issue).
Well, there are lots of assumptions in this argument that I am not too confident about…e.g., over what sort of range the dependence of radiative forcing on concentration is approximately logarithmic, and over what sort of range the climate sensitivity is relatively constant. In particular, we know that for an earth in the absence of greenhouse gases, the climate sensitivity is a lot less. E.g., assuming a constant albedo and in the absence of greenhouse gases, the temperature of the earth would be 255 K with the sun at its current luminosity but only drop about 22 K to 233 K for the sun at 70% of it current luminosity. This is because of the dependence of radiative energy on the 4th power of the temperature.
I thought that there was in fact evidence that CO2 levels were generally quite a bit higher in the past and are relatively low, by historical standards, over the last several million years than they generally were before that.
That Wikipedia site explains why methane levels might have been higher a few billion years ago before there was significant oxygen in the atmosphere. And, they propose a theoretical explanation for a negative feedback whereby colder temperatures would tend to eventually lead to a rise in CO2 levels.
This is only if you need to keep things exactly in balance. But, in fact what we know is that there has been considerable variation in the earth’s climate, just that it has remained in some not-that-broad (but not that narrow either) range. And, if there is a negative feedback effect, it will tend to keep things in some sort of rough balance…or at least within certain ranges.
This part confuses me. When there is a very speculative negative feedback proposed on short time scales involving clouds, you embrace it even though the evidence from the geologically recent past, which is the best evidence available, seems to be against such strongly stabilizing effects However, when the talk turns to negative feedbacks that might have operated over longer timescales to keep the temperature from varying too far afield, you are skeptical and seem to believe that such a feedback requires some sort of miraculous balancing rather than the same sort of principles that apply to any negative feedback.
The problem that I have with your solution to the faint-sun paradox is it seems to be rejecting the current theories that best fit the available evidence that we have in favor of a new hypothesis that explains just this one piece of evidence. It seems to me that you still have to show that it can explain all of the other paleoclimate evidence…including the more recent past, for which we have much better and accurately time-resolved experimental evidence. What we know from the more recent past is that the climate has varied quite considerably in response to what appear to be pretty small forcings, and that it has been both considerably colder and considerably warmer than it is today. (There is even some evidence for a “snowball earth” and a “hothouse earth” in the more distance past.) So, it does not seem like we are particularly close to a point in either direction where a strong negative feedback might kick in (and, at any rate, the negative feedbacks that seem likely to apply would tend to operate on much longer timescales than the timescale over which we are causing the current perturbation in forcing).
jshore, a very interesting post. A quick question, more to come. You say:
I thought the luminosity was a measure of the radiative energy … not so?
w.
Depends what you mean by “based on.” Looks to me as though the 0.12 figure comes from a number that comes from ONE scenario in ONE study. Looks to me as thought there was essentially no real meta-analysis. Nor was their any uncertainty analsysis as far as I can tell.
Oh really?
Would you like to guess where the following quote comes from?
(emphasis added)
Yeah…I meant to be using them basically synonomously. What I am saying is that in the straight graybody calculation (so, in the absence of any feedbacks…or really any greenhouse gases at all), the fact that the radiative energy goes as the 4th power of the temperature means that a 30% decrease in radiative energy would correspond to about an 8.5% decrease in the absolute temperature [0.085 being is 1-(0.7)^(1/4)].
By the way, here is a paper showing what they claim to be the first direct evidence supporting the hypothesis that atmospheric CO2 levels were much higher during the early Earth’s history.
There also seems to still be some debate in the astrophysics community regarding whether the sun really was that much fainter back then.
So, in summary, while this “faint young sun paradox” is interesting, the observational and theoretical constraints do not seem strong enough to elevate solution of it above understanding the much more recent paleoclimate events that we have much better data for.
jshore, you raise some interesting points. Let me see if I can elucidate them.
On a moment to moment basis, the tropical cloud albedo regulates the energy entering the peta-watt scale heat engine we call the climate. Suppose something changed with the cloud cover, so that instead of that albedo being about 30%, it changed to 33%, a mere 10% change. What would happen?
According to orthodox theory, it would heat the planet by around 10 degrees C (eighteen degrees F) or so. And if it changed to 27%, it would cool the earth by 10°C or so.
In the last half billion years, there’s no record of an excursion like that, 10° C. A freaking great asteroid hit the planet, for goodness sake, wiped out whole lineages of species, and the planet just went back to the same temperature.
So clearly, some combination of physical processes is keeping the cloud albedo in the tropics in balance with the temperature. As I said, I see it on a daily basis. Rising humidity makes clouds and thunderstorms that cool the surface down. If that cooling system failed for a couple weeks, the town I live in (on a Pacific island at 9° S Latitude) would be unbearably hot. Hasn’t happened in recorded history.
The question isn’t whether that effect exists. The question is how strong that effect is, how tightly it constrains the range of possible temperatures.
Whether or not this explains the “Faint Early Sun” paradox is not very important to me. I’m more interested in how it works on a daily/monthly/annual/decadal/century basis.
It is noteworthy, however, that Ou’s results (in the equilibrium study I cited above) showed that with a 50% change in solar constant from todays values, the surface temperature only varied by 10°C … this does not show, or prove, or establish, that this is the solution to the paradox. It does, however, establish it as a contender.
w.
Since you apparently don’t want to guess I’ll tell you. As far as I can tell, it comes from the paper that was the basis for the 0.12 figure you have been bandying about.
Turning back to the IPCC Report, it’s clear that the IPCC never actually says something like the following:
But that’s what’s implied and that’s what you took away from it, no?
BTW that was addressed to jshore
Without the context from which that quote appears, it seems like you have just engaged in a meaningless quote-mining exercise to me. (And, we saw how you took that “case-by-case basis” quote out of context previously.)
intention: Thanks for your response. I understand your desire to use paleoclimate evidence to try to estimate the climate sensitivity, but why is it that the scientists who actually do this come to the conclusion that “the climate system is very sensitive to small perturbations and that the climate sensitivity may be even higher than suggested by models” while you, in what is frankly a very hand-waving way, come to a different conclusion?
So, are you saying that the current climate models estimate the temperature on your Pacific island to be considerably hotter than it actually is?
The more interesting question is whether Ou’s results can predict the ice age - interglacial cycles and the other recent climate changes that we have the best data for.
Lol. You gotta be kidding me.
Anyway, here’s the full section:
(my bolding)
Nonsense.
First, please dial back on the accusations of hand-waving. I have provided a citation for my claim that the Earth has an equilibrium temperature. I have provided a citation for the idea that the earth is running at a condition of maximum power production/dissipation. I have also discussed the mechanism in detail. Your insults do not strengthen your case.
In addition, your pretended shock that other scientists might come to different conclusions does not become you. The article you cited is not peer-reviewed, it is a “Perspective”. Science Magazine, where it was published, says Perspectives "are meant to express a personal viewpoint ". Why haven’t you accused the author of “hand-waving”? In any case, from the personal viewpoint you cited:
This is indeed a “cautionary lesson” … but about the climate models, not the climate sensitivity. This shows that current models don’t understand the basic fundamentals of the climate, since they cannot explain the Eocene without including … hey, guess what, cloud feedback. Which is what I have been talking about, the fact that the models don’t properly represent the cloud feedback. However, his claim that including the proper cloud feedback in the models will increase estimates of future climate change is un-cited, un-referenced, and without any basis in fact. Which is OK, after all, this is his personal viewpoint … but I hope you don’t think it is more than that, I hope you don’t mistake his viewpoint for established science.
Also, his estimates of past sensitivity to small perturbations is based on these very same climate models that can’t successfully model the Eocene … which doesn’t exactly inspire confidence.
Do “the current climate models estimate the temperature on your Pacific island to be considerably hotter than it actually is?” Haven’t you been following the discussion? The problem is not that the models don’t get the rough numbers right, they’re tuned to do just that, so of course they do. How many times do we have to go over that? THE MODELS ARE TUNED TO THE PRESENT CLIMATE, SO THEY FACT THEY GET IT RIGHT PROVES NOTHING.
The problem is that they don’t get the feedbacks right, which has much larger implications. You seem to think that the models calculate the feedbacks … not so. The feedbacks in the model are set by the modelers based on their particular beliefs. They are inputs to the models, not outputs.
That is indeed an interesting question … however, as I’ve said before, there are many interesting and unanswered questions in the study of climate, it is a newcomer to the sciences. To date, for example, I have not found a good explanation for the ~100,000 year cycles of the glaciations. Might be one, but I haven’t seen it. It is a puzzle because this frequency is not among the Milankovic cycles. Nor is there a theory about why the previous interglacial was warmer than the current one, nor why it was shorter. Heck, we don’t have an explanation as to why we’re not in an ice age right now …
w.
intention, thanks for your response. Your citations just present a very speculative hypothesis about the climate system. (One does not even seem to address the issue of how the speculative hypothesis would affect climate sensitivity, at least from what I could see on skimming it, while the other does propose in general terms that it would tend to lead to low sensitivities.) Neither address the point that I was saying you are “hand-waving” which is the claim that a study of paleoclimate data favors the idea of a low climate sensitivity (other than Ou’s vague speculations, similar to yours, along the lines of the faint sun paradox).
Well, it is not clear to me exactly what the review process is for Perspectives but you have quoted what they say about them rather selectively. Here is the full description from which you presumably took your quote:
Because they have discussed specific paleoclimate events and what the study and modeling of them suggests in terms of the climate sensitivity.
However, the point is that such feedbacks are needed not to explain why the climate was so stable but why it was so unstable. And, I don’t know why you feel that his claim is unreferenced since the previous sentence discussing the special feedbacks needed provides you with 2 references.
The point is not that the climate models are perfect…but rather to the extent that they imperfectly explain the past, the problem seems to be that the underestimate the sensitivity of the climate to perturbations.
Look, the point is that you brought up a situation that was meant to highlight something that you apparently felt was missing from the current understanding of the climate incorporated into the climate models. Now, you are admitting that in fact the current climate models get the right answer but still trying to imply that they do so only because they have been tuned to agree. However, besides the fact that your description of tuning…and the degree to which the model parameters can be tuned to fit the whole wealth of available climate data…is at odds with almost the entire climate science community, this still begs the question: Clearly, the models have processes in them that keep your Pacific island from unrealistically roasting. What exactly do you think are unphysical about these processes as currently modeled?
They set certain parametrizations that control the feedbacks but do not set the feedbacks themselves. And, this process is severely constrained by the physics of the processes themselves, as well as the huge wealth of climatic data which has orders of magnitude more degrees of freedom than the number of degrees of freedom that the climate modelers have through changing the parameters.
By the way, for what it’s worth, here was a recent paper that proposed a biological cloud feedback that they argue can explain the mystery of the extensive warmth of some of the past climates. One distressing thing, if they are correct (and clearly, their work is still pretty speculative), is that they estimate that a 4X increase in CO2 above pre-industrial levels was enough to cause dramatic declines in biological productivity.
Actually, a section that explains it a bit better is here (from the report available here):
So, in other words, the modelers were free to make plausible assumptions about what the baseline…or reference…scenario of how greenhouse gases would grow in the absence of new mitigation policies. And, then from this they also produced four different stabilization scenarios. The IPCC then apparently looked at the results, chose which of the stabilizations scenarios corresponded to their various categories of stabilization scenario (i.e., what the stabilization level for greenhouse gases was) and noted whether the baseline assumed was a higher or lower one and reported the results. I’m not sure exactly what your point is.
This, by the way, was a study conducted by the U.S. Climate Change Science Program, created by President George W. Bush.
The IPCC presented these non-predictions as predictions and completely snowed you, notwithstanding your physics degree. That’s the point. To make things worse, you would rather join in the deception of yourself rather than just admit you’ve been had.
The IPCC has always made the distinction between “predictions” and “scenarios” or “projections” given the complications of predicting future societal behavior, which is based on choices that we as societies make. And, they were clear in their report which results were based on higher baselines and which were based on lower baselines.
They can distinguish all they like, but it doesn’t change the fact that they suggested X while actually stating Y. Nor does it change the fact that you (with your physics degree) were taken in by their dishonesty. You referred to the non-prediction as a “best estimate.” When in truth and in fact, it was just a (worst case?) “scenario” based on some assumptions that were “plausible” (and others that were clearly not).