I’ve been getting a lot of requests to cover this story, probably 20 or so now with wonderings about “why haven’t you covered this yet?”
How quickly you all forget. WUWT was the very first to cover this story back on November 10th, 2009.
Everybody else in the media today is playing catch-up. So if you’d like to read the original press release and participate in the already ripe comments left then, see this WUWT story:
Scott (18:44:20) :
Keep in mind A has actually been changing and is not a constant. If we suppose it has been growing linearly, and is only reaching 3% now, the actually effect would be a good deal less, possibly 1.5% depending on the time constant (based on integrating a triangle instead of a rectangle). My 3% is really just a conservative bound.
You can assume 4% if you like – I think the IPCC fraction is estimated to be between about 2.4% to 5.4%. If the integrated effect is 1/2 this, we might reasonably expect that we should not get more than 5.4%/2 = 2.7% max anthropogenic induced CO2 rise.
Glad you, at least, see what I have been talking about. If you follow my modeling up at Bart (14:51:01) , you will see my attempts to modify the model in such a way that the anthropogenic attribution would make sense. But, since I have to come up with roughly an order of magnitude amplification, it doesn’t generally lead to results which appear plausible. As I say there, the model
Cdot = (Co – C)/tau + (1+Ko)*adot
where tau is the time constant, Co is the equilibrium CO2, adot is the rate of anthropogenic contribution, and Ko is a stimulated emissions gain factor, is very general, and depends only on the dynamics being “smooth” to the extent that they may be linearized about Co.
philincalifornia says:
Come on, folks. This isn’t that hard to understand if you just get rid of your mental block on it. You have a system where the equilibration time is rapid between the atmosphere, biosphere, soils, and mixed layer and much slower with the deep ocean. So, to a very good approximation, you can solve the equations for the equilibration of a slug of CO2 added to the atmosphere + biosphere + soils + mixed layer system and find how much of the CO2 is partitioned between these different systems and then consider the slower process of the exchange of this slug of CO2 with the deep oceans. (For the partitioning between ocean and atmosphere, these are the basic chemical kinetics equations. For the biosphere and soils, it is less rigorous to determine the partitioning…and the best estimates are probably empirical ones.)
If you don’t like doing it this way, you could presumably numerically solve the full equations for the whole system … but, you would get essentially the same result from looking at it the way that I described and also get considerably more insight as to what is going on.
Scott (18:44:20)
Oh, and as far as the current science, the best I have found is explained here. As best I can tell, there has been a lot of hand waving, some questionable extrapolations of lab results, and they have effectively come up with models which decouple the dynamics so that natural CO2 and anthropogenic CO2 are treated differently with different time constants. Perhaps you can make more sense of it, since you appear disposed toward chemical analysis, which I haven’t studied in about 25 years.
This does not sit well with me, because I see no reason for the sinks to discriminate between natural and anthropogenic CO2. I also suspect a misapprehension on the part of the modelers that the feedback dynamics of the natural flows can be shut down, and they will magically stay in equilibrium. In the real world, and in just about every context you can imagine, lack of resistance is always met with aggression and, like in a fixed point iteration, an autoregressive system always either ends up at a local equilibrium at which forces cancel, or it shoots off to the boundary.
Bart (17:24:12) :
Scott (17:03:52) :
You are partly correct, but you have not followed through on the calculation.
In the steady state, without anthropogenic forcing, C = E/s.
Your “A” is a rate of anthropogenic input. It is less than or equal to perhaps 3% of E within some interval of time. We can thus say (wish I could use less than/greater than signs – does anyone know how to add these without messing up html tags?)
Cdot is less than or equal to 1.03*E – sC
Except you forgot that it’s compound interest!
So if t is time in years then it’s (1.03)^t*E-sC
“…they have effectively come up with models which decouple the dynamics so that natural CO2 and anthropogenic CO2 are treated differently…”
One more thing… A smooth set of differential equations, which became more sensitive to adot as C increases, would have positive feedback, and would have become unstable in a runaway greenhouse long ago even without the anthropogenic input. So, all in all, I really think this notion that natural and anthropogenic CO2 can be treated differently is not plausible, whatever hand-waving exercises they might have come up with to convince themselves of it.
It goes without saying that such principles could never have been tested on a planet-wide scale. I think it is just a kluge which they never really thought through in its implications because they saw rising CO2, and they wanted to finger a particular culprit.
Phil. (19:51:51) :
No, Phil, that’s not it. I hope you are just trying to be funny, because it’s not even close.
Bart:
It is just stunning to me to be accused of being a primitive by people who might as well be sporting chicken bones in their noses and wearing grass skirts. Simply amazing…
I’ve experienced this as well in other forums where AGW is discussed. It’s really disconcerting that there are people out there who are convinced they have “the truth” and nothing you can say will convince them otherwise, even though their lack of modern outlook is clear, to me at least. What adds insult to injury is they make sly comments about how ignorant I am. It used to drive me up the wall.
It reminds me of a discussion I saw a year ago about the same kind of ignorance (an ignorance of thought), though about a different subject. Someone chimed in that there are people in parts of Africa who will give you a look like you’re crazy if you say that there are no such things as evil spirits that cause disease. They KNOW it’s true, like the Sun rises in the East and sets in the West. If all else fails, all you can do is accept them as they are, leave them in their ignorance, and trust that the others reading the discussion can see their contribution for what it is.
Bart (20:08:52) :
Phil. (19:51:51) :
No, Phil, that’s not it. I hope you are just trying to be funny, because it’s not even close.
Nether is your analysis, you’ve failed to account for the fact that the sources are growing faster than the sinks.
Mark Miller (20:10:49)
🙂
Phil, I’m not going to bother with you. You are dead wrong, so much so that there is no point in further discussion.
Phil. (20:22:52) :
Sorry Phil, I don’t think that’s it. Look at some of the equations shown. In the simplest version (for instance, the one I described), manmade emissions are assumed at CURRENT levels, which are far higher than the start of the “industrial period”. Thus, they grew in a step change, instantly fast.
Now, Bart did try to incorporate what would amount to delays/sluggishness in the sink response, but I don’t think that’s too necessary to prove the point. Consider one of my earlier posts on this thread but in a different line of discussion–Scott (08:46:03) :. Here I’d calculated a rate of 3.1 ppm/year increase at 2008 emissions (note the referenced post fixed an earlier calculation I gave where I had a unit conversion in error). Thus, for the rise from the supposed pre-industrial 280 ppm to the current 388 ppm, emissions at the 2008 rate would require 35 years with NO absorption/sinks of anthropogenic CO2. Consider the following graph:
http://en.wikipedia.org/wiki/File:Global_Carbon_Emission_by_Type_to_Y2004.png
If we consider a “block” of 8 gt/yr going back 35 years, how would it’s integration compare to the entire industrial period? I figure that it would be pretty similar (that’s eyeballing it, your evaluation may differ). Thus, for the entire 108 ppm change to be due to man, the equivalent of all of manmade CO2 from the mid-1800s would have to remain in the atmosphere, which is a bit ridiculous. Now, you can nitpick some of the numbers if you want. However, I think we could rerun them for an older date, say 1960, and then I imagine it would be apparent that the increase in CO2/yr was higher than all of manmade emissions. Maybe I’ll do this and see what the results are for sure, but my significant other is ready to call it a night…
Please let me know if/what I calculated something wrong.
-Scott
Phil. (20:22:52) :
Bart (20:08:52) :
Phil. (19:51:51) :
No, Phil, that’s not it. I hope you are just trying to be funny, because it’s not even close.
Nether is your analysis, you’ve failed to account for the fact that the sources are growing faster than the sinks.
——————-
Say what ?? We’re on a thread discussing Knorr et al. which is about the fact that even if the sources are growing faster than the sinks, so what ?? The previously existing (massive) sinks are not saturated, and the growth in CO2 levels (over ~ 12 – 20 ppm) is from somewhere else.
Live with it.
…. and while you’re living with it, calculate how long the next doubling is going to take, based on this data.
As if it’s even relevant anyway.
Scott (20:52:52) :
Phil. (20:22:52) :
“you’ve failed to account for the fact that the sources are growing faster than the sinks.”
Sorry Phil, I don’t think that’s it. Look at some of the equations shown. In the simplest version (for instance, the one I described), manmade emissions are assumed at CURRENT levels, which are far higher than the start of the “industrial period”. Thus, they grew in a step change, instantly fast.
No, I suggest you read Knorr, in particular Fig 1!
philincalifornia (20:53:36) :
Say what ?? We’re on a thread discussing Knorr et al. which is about the fact that even if the sources are growing faster than the sinks, so what ??
I think that Knorr will be shocked to learn that this is is your interpretation of his paper!
The previously existing (massive) sinks are not saturated, and the growth in CO2 levels (over ~ 12 – 20 ppm) is from somewhere else.
“Somewhere else”, what are you babbling about!
Ferdinand Engelbeen (11:29:41) :
anna v (10:18:42) : said:
Look at these plots for the range of CO2 values from AIRS
http://photojournal.jpl.nasa.gov/jpegMod/PIA11186_modest.jpg
Also the Japanese data
http://www.jaxa.jp/press/2009/05/20090528_ibuki_e.html#at1
Makes me suspicious that the top plot has been homogenized . Maybe the Japanese are slow in treating the data trying to homogenize them too. Climategate should make us very suspicious of the whole climate industry”
Fderdinand said:
Anna V and others, please!
Have a look at the time span: 1 (one) month for AIRS, 1 (one) week for the Japanese satellite. And even then the whole “non-homogenization” is less than 5% of the value.
“Well mixed” doesn’t mean that everywhere at any time of the year exactly the same level of CO2 is measured. It means that the amounts level out in a not too long time span, all over the world. Like a year for all measurements in one hemisphere and a few years for the whole earth… But if you have huge (seasonal) exchanges at ground level and continuous additions in one hemisphere, there will never be exact the same level at all places.
Your last paragraph applies to temperature too, exactly. Nevertheless we measure temperatures and do not call them well mixed. The percentage of temperature change in degrees kelvin 0ver the year , .2 Cper year, is much less than the CO2 differences and we, 0.01% for temperature per year, and we do not call temperature well mixed. We plot it in lumpy world plots that show anomalies. Lets see a map of CO2 anomalies, instead of values. If values of the temperatures were plotted the earth would be one color.
Ferdinand Engelbeen (11:29:41) :
anna v (10:18:42) : said:
Look at these plots for the range of CO2 values from AIRS
http://photojournal.jpl.nasa.gov/jpegMod/PIA11186_modest.jpg
Also the Japanese data
http://www.jaxa.jp/press/2009/05/20090528_ibuki_e.html#at1
Makes me suspicious that the top plot has been homogenized . Maybe the Japanese are slow in treating the data trying to homogenize them too. Climategate should make us very suspicious of the whole climate industry”
Fderdinand said:
Anna V and others, please!
Have a look at the time span: 1 (one) month for AIRS, 1 (one) week for the Japanese satellite. And even then the whole “non-homogenization” is less than 5% of the value.
“Well mixed” doesn’t mean that everywhere at any time of the year exactly the same level of CO2 is measured. It means that the amounts level out in a not too long time span, all over the world. Like a year for all measurements in one hemisphere and a few years for the whole earth… But if you have huge (seasonal) exchanges at ground level and continuous additions in one hemisphere, there will never be exact the same level at all places.
Your last paragraph applies to temperature too, exactly. Nevertheless we measure temperatures and do not call them well mixed. The percentage of temperature change in degrees kelvin 0ver the year , .2 Cper year, is much less than the CO2 differences , 0.01% for temperature per year, and we do not call temperature well mixed. We plot it in lumpy world plots that show anomalies. Lets see a map of CO2 anomalies, instead of values. If values of the temperatures were plotted the earth would be one color.
Phil. (21:54:48) :
So you think that 280 ppm was some magic number that was in a sensitive equilibrium for millennia, only to be perturbed by a gradual increase of CO2 emissions from x to 1.03x starting in 1850 ?? And that the anthropogenic emissions became additive because the sinks were somehow saturated for no apparent reason ??
You still believe in Mann’s hockey stick ??
philincalifornia (22:31:26) :
Phil. (21:54:48) :
So you think that 280 ppm was some magic number that was in a sensitive equilibrium for millennia, only to be perturbed by a gradual increase of CO2 emissions from x to 1.03x starting in 1850 ?? And that the anthropogenic emissions became additive because the sinks were somehow saturated for no apparent reason ??
You haven’t read the paper have you? The paper clearly points out that emissions rose from 2billion tons/year in 1850 to 35billion tons/year now, i.e. 17.5x, compound interest!
Phil. (21:54:48) :
Phil, you’re just not even wrong. Not only do I wrestle with the question of sensitivity extensively in this thread, but the Knorr report makes the matter moot: there is no indication of the sinks being overwhelmed or even diminishing appreciably in their power. You do not understand the article’s implications because it is clearly outside or your area of expertise, whatever that may be. Your post at 19:51:51 proves that to anyone who has a modicum of familiarity with systems theory. I see no hope of making you understand it, and no advantage in discussing it further with you.
Joel Shore (18:11:54) :
“Before you can apply systems theory, you have to understand the system.”
You have it precisely bass-ackwards. In order to understand the system, you have to understand the underlying theory of systems and how they evolve in the real world.
Scott: on this:
Bart (20:06:28) :
” A smooth set of differential equations, which became more sensitive to adot as C increases, would have positive feedback, and would have become unstable in a runaway greenhouse long ago even without the anthropogenic input.”
That is not completely true. Some models would. But, upon reflection, I find you can determine a nonlinear, negative feedback which will be stable and have increasing sensitivity to adot by taking the negative gradient of a particular class of potential functions, e.g.,
V = (1/(2*tau))*(C-Co)^2/sqrt(1+eps*(C-Co)^2)
As eps approaches zero, this produces the same feedback term as I gave earlier (Co-C)/tau. But while, with eps greater than zero, this will increase the effective time constant, leading to slower dissipation of the anthropogenic component, what it will also do is reduce the input from the natural sources as C deviates greatly from Co, and there is little reason to expect that.
But, this is something like what the system would have to be in order for the orthodoxy to be reasonable. It goes without saying that the nonlinearity would have to be very significant to get a factor of 10 increase in the sensitivity to adot. The Knorr report pretty much rules this out in our current operating regime.
anna v (22:27:10) :
Have a look at the combined plot of yearly averages from different CO2 stations from near the North Pole to the South Pole and compare that to the yearly averages of temperatures from the same places and the same time span…
Here is the one for CO2:
http://www.ferdinand-engelbeen.be/klimaat/klim_img/co2_trends.jpg
Or a yearly average difference of maximum 1% of the full range
Barrow for the same time span is average at -11 C, Hawai (sea level) at +24 C (didn’t find MLO station temperature data in a fast search) and the South Pole at -50 C. Or a near continuous yearly average difference of over 25% of the full range of absolute temperature.
The difference in temperature between different places on earth may remain more or less constant, but I don’t think that you can say that temperature levels out all over the globe in a few years time…
Richard Sharpe (17:43:38) :
I tend to have great regard for what you write because it seems so well thought out, however, I have a question.
We are told that the solubility of CO2 in water decreases with temperature.
How have you eliminated the possibility that the heat content of the oceans has been increasing recently, and therefore driving more CO2 out of solution?
A small decrease in average cloud cover (perhaps because of fewer CRs getting through, or less DMSO being released by plankton–of course all such mechanisms need to be explained) should allow more energy to get to the oceans.
Indeed the upper part of the oceans can hold less CO2 when heating up. That is partly compensated by biolife (as well in the oceans as on land), which sequester more CO2 with increased temperatures.
The historical ratio between CO2 and temperature was about 8 ppmv/K over glacials and interglacials. That includes (deep) ocean flows changes, land use changes (ice caps growth and retreat, forest growth up to the Arctic seas and retreat), with a lag of about 600 years. On shorter term (MWP-LIA) we see a similar ratio in high resolution ice cores. And nowadays we see a temperature dependent variability (Pinatubo, 1998 El Niño) in growth rate of about 4 ppmv/K around the trend.
Thus worst case is that the temperature increase of about 1 K since the LIA has changed the equilibrium setpoint with about 8 ppmv. But as we have over 100 ppmv increase by now, the ocean temperature is only a small part of the increase. And the d13C level makes it impossible that the oceans delivered the bulk of the increase.
One more thing on this: Bart (01:29:50)
The hypothesis of a nonlinear, sensitivity amplifying, feedback such as in that post in no way invalidates the simple model equation from before
Cdot = (Co-C)/tau + (1+Ko)*adot
It merely says that the linearization which led to this equation might not have a wide region of accuracy, i.e., the system would have a high degree of nonlinearity, and you might need to re-linearize with a new value of tau as C evolved outside of this region. You could integrate the equation then in a piecewise continuous fashion. Or, as an approximation (possibly more precise actually, depending on the actual complete system description), you could just substitute in the gradient feedback function given with the additive (1+Ko)*adot input.
To all:
My developments here are still evolving. But, what I have has led me to conclude that the hypothesis of anthropogenic attribution for the observed rise in atmospheric CO2 in the last 50 years requires some very unusual dynamics which do not appear to be supported by the data, in particular this study from Knorr.
These things can be stated for certain:
1) the equation
Cdot = (Co-C)/tau + (1+Ko)*adot
represents the linearized form of the system no matter its details, with only the assumption of smooth differential equations describing it, and the approximation of a linear operator K acting on adot as being adequately described by its dc gain Ko – this is very likely a reasonable assumption if adot is a very low bandwidth signal, as it appears to be. The term Ko*adot represents stimulated emissions.
As a linearization, it is accurate only in some local region about the set point Co and, if the system is highly nonlinear, it must be relinearized about a new set point when the system evolves beyond these boundaries.
2) If Ko is “small”, then the effect of anthropogenic forcing on atmospheric concentration is small, and we can state with high confidence that the effect of a 3% ratio between anthropogenic and natural forcing is effectively 3% on atmospheric concentration, the initial linearization is likely adequate, and that is just about all there is.
3) without strong nonlinearity in the system, Ko must be very large, roughly a factor of 10, to explain the observed rise. That would mean that for every unit of anthropogenic CO2, there are 10 more of natural CO2 stimulated to be released.
4) with strong nonlinearity, and a significant value of Ko to engage it, we could be in a milieu in which frequent re-linearization would be required, and we might be able to fulfill the orthodoxy in which the anthropogenic input is enough to account for the observed rise. However, it does not appear likely to me that Ko could be very large and, furthermore, the Knorr paper argues against such exotic behavior.
That is my case against the hypothesis attributing the rise to anthropogenic sources. I do not expect it to persuade those who cannot or will not understand it. But, I expect we will know within a few years who is right.
This is just one of the legs upon which CAGW depends. The other critical leg is the existence of positive feedback for CO2 temperature forcing which others are focusing on at this time. It is very unseasonably cold outside where I am tonight, as it is in many places across the northern hemisphere. Thanks to all who interacted with me in this little excursion. I am done.
Ferdinand Engelbeen (01:31:32) :
In answer to Anna V’s excellent! question:
“The difference in temperature between different places on earth may remain more or less constant, but I don’t think that you can say that temperature levels out all over the globe in a few years time…”
CO2 doesn’t level out either–it’s always lumpy and I suspect there is more that NASA hasn’t said yet about the results but we’ll see. Besides, I hope and assume the models do NOT level out temperature even though we have these ‘global surface temperature’ series that everyone talks about, no matter how majorly lumpy the individual temp readings are. I would hope they treat CO2 the same way but I assume they do NOT. In their grids I doubt they are calculating a higher CO2 level in the grid that includes Pittsburgh than the one that includes Aukland but as I understand it they do calculate the temps.
————
Ferdinand Engelbeen (01:59:23) :
“The historical ratio between CO2 and temperature was about 8 ppmv/K over glacials and interglacials.”
Please clarify something else for me. The ‘8ppmv/K’ refers to air temperature not SST, correct? or? Because as I understand it the ocean temperature did not change 10K between the glaciation and the interstitial. The question was about a rise in SST and the effect that would have on CO2, and I don’t think that was specifically answered. As temperature rose, more forests would appear which would tend to reduce, not add, CO2. So it’s not clear to me where the CO2 coming out of deep glaciations actually came from.