Guest Post by Willis Eschenbach
There seem to be a host of people out there who want to discuss whether humanoids are responsible for the post ~1850 rise in the amount of CO2. People seem madly passionate about this question. So I figure I’ll deal with it by employing the method I used in the 1960s to fire off dynamite shots when I was in the road-building game … light the fuse, and run like hell …
First, the data, as far as it is known. What we have to play with are several lines of evidence, some of which are solid, and some not so solid. These break into three groups: data about the atmospheric levels, data about the emissions, and data about the isotopes.
The most solid of the atmospheric data, as we have been discussing, is the Mauna Loa CO2 data. This in turn is well supported by the ice core data. Here’s what they look like for the last thousand years:
Figure 1. Mauna Loa CO2 data (orange circles), and CO2 data from 8 separate ice cores. Fuji ice core data is analyzed by two methods (wet and dry). Siple ice core data is analyzed by two different groups (Friedli et al., and Neftel et al.). You can see why Michael Mann is madly desirous of establishing the temperature hockeystick … otherwise, he has to explain the Medieval Warm Period without recourse to CO2. Photo shows the outside of the WAIS ice core drilling shed.
So here’s the battle plan:
I’m going to lay out and discuss the data and the major issues as I understand them, and tell you what I think. Then y’all can pick it all apart. Let me preface this by saying that I do think that the recent increase in CO2 levels is due to human activities.
Issue 1. The shape of the historical record.
I will start with Figure 1. As you can see, there is excellent agreement between the eight different ice cores, including the different methods and different analysts for two of the cores. There is also excellent agreement between the ice cores and the Mauna Loa data. Perhaps the agreement is coincidence. Perhaps it is conspiracy. Perhaps it is simple error. Me, I think it represents a good estimate of the historical background CO2 record.
So if you are going to believe that this is not a result of human activities, it would help to answer the question of what else might have that effect. It is not necessary to provide an alternative hypothesis if you disbelieve that humans are the cause … but it would help your case. Me, I can’t think of any obvious other explanation for that precipitous recent rise.
Issue 2. Emissions versus Atmospheric Levels and Sequestration
There are a couple of datasets that give us amounts of CO2 emissions from human activities. The first is the CDIAC emissions dataset. This gives the annual emissions (as tonnes of carbon, not CO2) separately for fossil fuel gas, liquids, and solids. It also gives the amounts for cement production and gas flaring.
The second dataset is much less accurate. It is an estimate of the emissions from changes in land use and land cover, or “LU/LC” as it is known … what is a science if it doesn’t have acronyms? The most comprehensive dataset I’ve found for this is the Houghton dataset. Here are the emissions as shown by those two datasets:
Figure 2. Anthropogenic (human-caused) emissions from fossil fuel burning and cement manufacture (blue line), land use/land cover (LU/LC) changes (white line), and the total of the two (red line).
While this is informative, and looks somewhat like the change in atmospheric CO2, we need something to compare the two directly. The magic number to do this is the number of gigatonnes (billions of tonnes, 1 * 10^9) of carbon that it takes to change the atmospheric CO2 concentration by 1 ppmv. This turns out to be 2.13 gigatonnes of carbon (C) per 1 ppmv.
Using that relationship, we can compare emissions and atmospheric CO2 directly. Figure 3 looks at the cumulative emissions since 1850, along with the atmospheric changes (converted from ppmv to gigatonnes C). When we do so, we see an interesting relationship. Not all of the emitted CO2 ends up in the atmosphere. Some is sequestered (absorbed) by the natural systems of the earth.
Figure 3. Total emissions (fossil, cement, & LU/LC), amount remaining in the atmosphere, and amount sequestered.
Here we see that not all of the carbon that is emitted (in the form of CO2) remains in the atmosphere. Some is absorbed by some combination of the ocean, the biosphere, and the land. How are we to understand this?
To do so, we need to consider a couple of often conflated measurements. One is the residence time of CO2. This is the amount of time that the average CO2 molecule stays in the atmosphere. It can be calculated in a couple of ways, and is likely about 6–8 years.
The other measure, often confused with the first, is the half-life, or alternately the e-folding time of CO2. Suppose we put a pulse of CO2 into an atmospheric system which is at some kind of equilibrium. The pulse will slowly decay, and after a certain time, the system will return to equilibrium. This is called “exponential decay”, since a certain percentage of the excess is removed each year. The strength of the exponential decay is usually measured as the amount of time it takes for the pulse to decay to half its original value (half-life) or to 1/e (0.37) of its original value (e-folding time). The length of this decay (half-life or e-folding time) is much more difficult to calculate than the residence time. The IPCC says it is somewhere between 90 and 200 years. I say it is much less, as does Jacobson.
Now, how can we determine if it is actually the case that we are looking at exponential decay of the added CO2? One way is to compare it to what a calculated exponential decay would look like. Here’s the result, using an e-folding time of 31 years:
Figure 4. Total cumulative emissions (fossil, cement, & LU/LC), cumulative amount remaining in the atmosphere, and cumulative amount sequestered. Calculated sequestered amount (yellow line) and calculated airborne amount (black) are shown as well.
As you can see, the assumption of exponential decay fits the observed data quite well, supporting the idea that the excess atmospheric carbon is indeed from human activities.
Issue 3. 12C and 13C carbon isotopes
Carbon has a couple of natural isotopes, 12C and 13C. 12C is lighter than 13C. Plants preferentially use the lighter isotope (12C). As a result, plant derived materials (including fossil fuels) have a lower amount of 13C with respect to 12C (a lower 13C/12C ratio).
It is claimed (I have not looked very deeply into this) that since about 1850 the amount of 12C in the atmosphere has been increasing. There are several lines of evidence for this: 13C/12C ratios in tree rings, 13C/12C ratios in the ocean, and 13C/12C ratios in sponges. Together, they suggest that the cause of the post 1850 CO2 rise is fossil fuel burning.
However, there are problems with this. For example, here is a Nature article called “Problems in interpreting tree-ring δ 13C records”. The abstract says (emphasis mine):
THE stable carbon isotopic (13C/12C) record of twentieth-century tree rings has been examined1-3 for evidence of the effects of the input of isotopically lighter fossil fuel CO2 (δ 13C~-25‰ relative to the primary PDB standard4), since the onset of major fossil fuel combustion during the mid-nineteenth century, on the 13C/12C ratio of atmospheric CO2(δ 13C~-7‰), which is assimilated by trees by photosynthesis. The decline in δ13C up to 1930 observed in several series of tree-ring measurements has exceeded that anticipated from the input of fossil fuel CO2 to the atmosphere, leading to suggestions of an additional input ‰) during the late nineteenth/early twentieth century. Stuiver has suggested that a lowering of atmospheric δ 13C of 0.7‰, from 1860 to 1930 over and above that due to fossil fuel CO2 can be attributed to a net biospheric CO2 (δ 13C~-25‰) release comparable, in fact, to the total fossil fuel CO2 flux from 1850 to 1970. If information about the role of the biosphere as a source of or a sink for CO2 in the recent past can be derived from tree-ring 13C/12C data it could prove useful in evaluating the response of the whole dynamic carbon cycle to increasing input of fossil fuel CO2 and thus in predicting potential climatic change through the greenhouse effect of resultant atmospheric CO2 concentrations. I report here the trend (Fig. 1a) in whole wood δ 13C from 1883 to 1968 for tree rings of an American elm, grown in a non-forest environment at sea level in Falmouth, Cape Cod, Massachusetts (41°34’N, 70°38’W) on the northeastern coast of the US. Examination of the δ 13C trends in the light of various potential influences demonstrates the difficulty of attributing fluctuations in 13C/12C ratios to a unique cause and suggests that comparison of pre-1850 ratios with temperature records could aid resolution of perturbatory parameters in the twentieth century.
This isotopic line of argument seems like the weakest one to me. The total flux of carbon through the atmosphere is about 211 gigtonnes plus the human contribution. This means that the human contribution to the atmospheric flux ranged from ~2.7% in 1978 to 4% in 2008. During that time, the average of the 11 NOAA measuring stations value for the 13C/12C ratio decreased by -0.7 per mil.
Now, the atmosphere has ~ -7 per mil 13C/12C. Given that, for the amount of CO2 added to the atmosphere to cause a 0.7 mil drop, the added CO2 would need to have had a 13C/12C of around -60 per mil.
But fossil fuels in the current mix have a 13C/12C ration of ~ -28 per mil, only about half of that requried to make such a change. So it is clear that the fossil fuel burning is not the sole cause of the change in the atmospheric 13C/12C ratio. Note that this is the same finding as in the Nature article.
In addition, from an examination of the year-by-year changes it is obvious that there are other large scale effects on the global 13C/12C ratio. From 1984 to 1986, it increased by 0.03 per mil. From ’86 to ’89, it decreased by -0.2. And from ’89 to ’92, it didn’t change at all. Why?
However, at least the sign of the change in atmospheric 13C/12C ratio (decreasing) is in agreement with with theory that at least part of it is from anthropogenic CO2 production from fossil fuel burning.
CONCLUSION
As I said, I think that the preponderance of evidence shows that humans are the main cause of the increase in atmospheric CO2. It is unlikely that the change in CO2 is from the overall temperature increase. During the ice age to interglacial transitions, on average a change of 7°C led to a doubling of CO2. We have seen about a tenth of that change (0.7°C) since 1850, so we’d expect a CO2 change from temperature alone of only about 20 ppmv.
Given all of the issues discussed above, I say humans are responsible for the change in atmospheric CO2 … but obviously, for lots of people, YMMV. Also, please be aware that I don’t think that the change in CO2 will make any meaningful difference to the temperature, for reasons that I explain here.
So having taken a look at the data, we have finally arrived at …
RULES FOR THE DISCUSSION OF ATTRIBUTION OF THE CO2 RISE
1. Numbers trump assertions. If you don’t provide numbers, you won’t get much traction.
2. Ad hominems are meaningless. Saying that some scientist is funded by big oil, or is a member of Greenpeace, or is a geologist rather than an atmospheric physicist, is meaningless. What is important is whether what they say is true or not. Focus on the claims and their veracity, not on the sources of the claims. Sources mean nothing.
3. Appeals to authority are equally meaningless. Who cares what the 12-member Board of the National Academy of Sciences says? Science isn’t run by a vote … thank goodness.
4. Make your cites specific. “The IPCC says …” is useless. “Chapter 7 of the IPCC AR4 says …” is useless. Cite us chapter and verse, specify page and paragraph. I don’t want to have to dig through an entire paper or an IPCC chapter to guess at which one line you are talking about.
5. QUOTE WHAT YOU DISAGREE WITH!!! I can’t stress this enough. Far too often, people attack something that another person hasn’t said. Quote their words, the exact words you think are mistaken, so we can all see if you have understood what they are saying.
6. NO PERSONAL ATTACKS!!! Repeat after me. No personal attacks. No “only a fool would believe …”. No “Are you crazy?”. No speculation about a person’s motives. No “deniers”, no “warmists”, no “econazis”, none of the above. Play nice.
OK, countdown to mayhem in 3, 2, 1 … I’m outta here.
Here is what AIRS itself is saying:
Significant Findings from AIRS Data
1. ‘Carbon dioxide is not homogeneous in the mid-troposphere; previously it was thought to be well-mixed
2. ‘The distribution of carbon dioxide in the mid-troposphere is strongly influenced by large-scale circulations such as the mid-latitude jet streams and by synoptic weather systems, most notably in the summer hemisphere
3. ‘There are significant differences between simulated and observed CO2 abundance outside of the tropics, raising questions about the transport pathways between the lower and upper troposphere in current models
4. ‘Zonal transport in the southern hemisphere shows the complexity of its carbon cycle and needs further study
http://airs.jpl.nasa.gov/AIRS_CO2_Data/About_AIRS_CO2_Data/
They talk of middle troposphere for their results. Not column averaged as the Japanese data I linked to. Even at this level AIRS does not see well mixing.
Not well mixed means one needs 3d plots.
Did you watch the experiment with pouring CO2?
John Finn says:
June 7, 2010 at 3:40 pm
Willis Eschenbach says
June 7, 1:20 pm
You are again conflating residence time (6-8 years or so) with the half life (much longer)
Not so – residence time is half life / ln 2 (which is 0.693). Thus t1/2 is a bit shorter than residence time (tau).
There might be some confusion over the definition of ‘residence time’ here. I think Willis is referring to the average lifetime (residence time) of a CO2 molecule in the atmosphere.
Thats what I’m talking about too – the residence time tau. If we choose to define CO2 clearence from the atmosphere as exponential (rather than immortalise CO2 in the atmosphere as a CAGW deity) then we have to play by the rules of exponential decay. Again, residence time tau = halflife / 0.693.
This is only a few years. However, this is not the same thing as the time taken for a pulse of CO2 to be removed from the atmosphere. In this case, both the half-life and e-folding time are much longer.
How long does a pulse of CO2 stay in the atmosphere? Forever, until judgement day, or until CO2 starts causing global warming, however you choose to define an infinite time. Exponential decay never reaches zero.
This discussion on exponential CO2 removal is very important, the CAGW establishment try to wriggle out of discussion of CO2 removal from the atmosphere and thus tacitly deify CO2 as an immortal angry sky god. This discussion will make them uncomfortably so is to be applauded – indeed it will seem blasphemous to them to think that CO2 can leave the atmosphere.
But why introduce the esoteric term “e folding time” when all that is needed is the exponential decay term or the half life?
In fact CO2 removal from the atmosphere would seem likely to be multi-term or multi-compartmental with several exponential terms (although I looked in Google scholar but could not find any reference). There may be short, medium and long term compartments. This could explain the anomalously smooth curve of CO2 in air with time. However I share the suspicion of some posters about the credibility of such a smooth CO2 curve. If human CO2 emission is indeed now a dominant signal, then changes in CO2 output caused for instance by an economic recession, should leave a visible signal in the CO2 curve. But they dont.
Willis Eschenbach says:
June 8, 2010 at 12:52 am
… the main issue is not whether humans are raising the CO2. It is whether that CO2 rise will affect the temperature. I see no evidence that it will, and lots of evidence that it won’t. So I don’t want to dispute what I see as an inconsequential issue for which, as I said, I think the preponderance of evidence is large in favor of the idea that humans caused the CO2 rise.
Fair enough, and I agree with you that it’s a side issue. Still one that tax raisers will use to beat us with though. If the human part of the atmosperic co2 flux increased 1.3% over the modern warming period, we are only talking about 0.00049% of the atmosphere. I won’t be losing any sleep.
Cheers
Keith Minto, sorry I posted an inaccessible link, might have left you wondering what I was talking about, here’s the correct one, and correctly labeled NASA’s, not IPCC’s, they just approved it! 🙂 (it’s a spoof!)
http://www.vermonttiger.com/content/2008/07/nasa-free-energ.html
I always read about the 800-year timelag of a CO2 increase following a rise in temperatures, but I hardly ever find anybody arguing that the present level of CO2 was caused 800 years ago. Strange!
Willis:
At (June 7, 2010 at 1:13 pm) you write:
“My impression (haven’t read too much on the stomatal measurements) is that there are some problems with the stomatal data. One that I haven’t seen discussed much is this: …”
Previously, at (June 7, 2010 at 2:43 am) I wrote:
“As an aside, I address your point concerning the ice-core data because I think it is a distraction. …
[snip]
There is – at very least – adequate reason to assess the recent changes in atmospheric CO2 concentration as indicated at Mauna Loa, Barrow, etc. on the basis of the behaviour of the carbon cycle since 1958 (when measurements began at Mauna Loa).
Comparison of the recent rise in atmospheric CO2 concentration with paleo data merely provides a debate as to
(a) the validity of the ice-core data (which provides the ‘hockey stick’ graph you reproduce above)
and
(b) the validity of the stomata data that shows the recent rise in atmospheric CO2 concentration is similar to rises that have repeatedly happened previously.”
Quod erat demonstrandum.
Richard
Z says June 7, 2010 at 5:56 pm:
Yes, this is a very cogent point. The lack of pressure is found to be a big problem when oceanographers/biologists try to study deep sea creatures when the creatures are exposed to our 1 bar atmosphere.
As soon as the core is exposed – unless it is but into a pressure chamber and kept there until and including the sampling tests are undertaken – the “steady state” is erased and all bets are off as to what was really there when it was at depth. No matter HOW careful they are with procedures, the baseline shifted in unknown ways, throwing serious doubt on any data.
If they then USE that data, biff! GIGO.
John Finn:
At (June 7, 2010 at 3:28 pm ) pertaining to Beck’s data, you ask:
“So a drop of ~80 ppm in a few years around 1940 is believable is it?”
I answer:
Yes, it is.
Please see my above comment at (June 7, 2010 at 3:23 am). I wrote there:
“Your point about “uptake of CO2” by the oceans cuts both ways. The great bulk of carbon flowing around the carbon cycle is in the oceans. An equilibrium state exists between the atmospheric CO2 concentration and the carbon concentration in the ocean surface layer. So, all other things being equal, if the atmospheric CO2 concentration increases then – as you say – the ocean surface layer will dissolve additional CO2 and alkalinity of the layer will reduce. However, the opposite is also true.
If the alkilinity of the ocean surface layer reduces then the equilibrium state will alter to increase the atmospheric CO2 concentration and to reduce the carbon in the ocean surface layer. The pH change required to achieve all of the recent rise in atmospheric CO2 concentration (i.e. since 1958 when measurements began at Mauna Loa) is less than 0.1 which is much, much too small for it to be detectable. And changes of this magnitude can be expected to occur.
Surface waters sink to ocean bottom, travel around the globe for ~800 years then return to ocean surface. They can be expected to dissolve S and Cl from exposure to undersea volcanism during their travels. So, the return to the surface of these waters will convey the S and Cl ions to the surface layer centuries after their exposure to the volcanism, and this could easily reduce the surface layer pH by more than 0.1. Hence, variations in undersea volcanism centuries ago could be completely responsible for the recent rise in atmospheric CO2 concentration.
Please note that the fact that these volcanic variations could be responsible for the recent rise does not mean they are responsible (which is the same logic as the fact that the anthropogenic emissions could be responsible does not mean that they are).
However, Tom Quirk observes that the geographical distribution of atmospheric carbon isotopes provides a better fit to the undersea volcanism hypothesis than to the anthropogenic hypothesis as a cause of the rise: see
http://climaterealists.com/attachments/database/A%20Clean%20Demonstration%20of%20Carbon%2013%20Isotope%20Depletion.pdf
There are many possible causes of the recent rise in atmospheric CO2 concentration. They each warrant investigation, and there is not sufficient evidence to champion any one of them.”
I hope you will take note of my final paragraph in my above quotation: I am not championing the ‘volcanic variation hypothesis’ but am using it here to demonstrate that “a drop of ~80 ppm in a few years around 1940” is physically possible.
Seasonal variation differs greatly from place to place. It is minimal at Mauna Loa (which is why Keeling chose to monitor atmospheric CO2 concentration there) where it is typically more than 8 ppmv each year. And, at that location, this variation is attributed to oceanic emission and sequestration.
So, a change to the pH of the ocean surface layer that deverely inhibited oceanic emission (but not sequestration) could easily induce “a drop of ~80 ppm” in less than a decade.
This is yet another example of how closed minds inhibit investigation of the true behaviours of the carbon cycle by asserting that known possibilities are not “believable”.
Richard
Gail Combs says June 7, 2010 at 6:46 pm:
Wow. I have always furrowed my brow at why they would put a CO2 station on top of a CO2 spewing volcano and try to convince us all that the data is pristine. Out in the middle of the ocean makes a lot of sense, but couldn’t they find an island WITHOUT a simmering volcano?
THANK YOU for this bit of information. I don’t trust their data at all anymore. Even the CO2 data coming from Mauna Loa has to be “adjusted” – and guess who gets to do the adjusting?
Unless I’ve missed it, I notice no-one has mentioned the AIRS data from satellite observations. These are the oservations of CO2 in the mid-troposphere. From what I recall they seem to provide strong independent support for that the Mauna Loa readings are accuarate and are representaive of the “well-mixed” atmosphere.
This link
http://airs.jpl.nasa.gov/AIRS_CO2_Data/
shows the CO2 map for July 2009 which shows the average concentration at roughly 387 ppm. The July 2009 reading at ML was 387.74 ppm.
If we can establish that ML and the ice core data are reliable, then it will become a relatively straightforward task to show that human emissions are mainly responsible for the rise.
That low blue point (360ppm) on the coast of western Australia contrasts with a solitary ‘spot’ in the desert of central Australia and one near Brisbane at 385ppm. 385ppm is also recorded by Cape Grim on the west coast of Tasmania in 2009.
This is more than “a percent or two” of variation across a continent and worthy of more attention, perhaps in a new thread.
I could be wrong but I think that anna v is, like me is interested in a vertical profile of CO2 mixing,or, more likely, layering.
Wayne, I will digest that Nicol article a little more and not comment on it now so as not to go OT, thanks anyway for the link.
“Willis
“I’ve never been able to find a physical explanation of how this works in the real world.”
The reason is that exponential decay just isn’t the right model for diffusion of CO2 into the sea. If you want to think of it as a set of layers of equal thickness and diffusivity respobding to a rise of CO2 in the air, then the top layer absorbs CO2 fairly quickly, since there’s not much resistance. When it approaches capacity, the next layer comes into play. But to get there, CO2 has to pass through the top layer, so it fills more slowly, with a longer time constant. And so on down.
This is solving the diffusion equation, and instead of convolving with a decaying exponential, which as Bart says is what you are doing, you should be convolving with a function of form t^(-3/2). This decays much more slowly than an exponential, but you could consider it approximately made up as a sum of exponentials of varying time constant. That’s where the Bern model comes from.
anna v says:
June 7, 2010 at 11:47 pm
We had discussed it here at the time it appeared, http://wattsupwiththat.com/2009/09/13/some-results-from-gosat-co2-hot-spots-in-interesting-places/
Thanks for that reference…. which includes some interesting comments regarding the validity of the MLO data (besides their data processing procedures)…
OsandZsChemist says:
September 14, 2009 at 5:01 am
It is interesting to correlate this with historical pollution tracking and jetstream patterns http://www.cgd.ucar.edu/cms/pjr/pubs/2000JD900842.pdf which imply a dumpout between Hong Kong (lat. 22N) and Hawaii (lat. 21N). Although an early reach, it is possible to speculate that the Mauna Loa CO2 measurements are biased by the huge CO2 emissions from China and the Middle East. Anyone tried a correlation between ramp up of fossil fuel use in China and the Mauna Loa http://www.esrl.noaa.gov/gmd/ccgg/trends/co2_data_mlo.html measurements?
John says:
September 14, 2009 at 11:58 am
Concerning Mauna Loa, Jeffrey A. Glassman (http://www.rocketscientistsjournal.com/) has for years insisted that Mauna Loa was a very poor choice for monitoring CO2. The island lies in the middle of a discharge plume as cool North Pacific water warming on its trip southward swings westward. The warming results in a steady elevation in atmospheric CO2. This map seems to confirm that Mauna Loa overestimates global atmospheric CO2 levels.
=====================================================
So it looks like the MLO and Ice Core data have all sorts of associated issues….
So this CO2 analysis seems to have very questionable foundations… so say the least.
John Finn:
At (June 8, 2010 at 2:56 am) you say:
“If we can establish that ML and the ice core data are reliable, then it will become a relatively straightforward task to show that human emissions are mainly responsible for the rise.”
I agree it is a pity that we cannot determine the reliability of either of them.
However, in the light of his comment at (June 7, 2010 at 7:09 pm) , dr.bill would probably be interested to know that in 1859 Michael Faraday was the discoverer of the major problem with ice cores as reliable stores of atmospheric CO2 concentrations: i.e. all ice surfaces (including surfaces of ice crystals) are coated in a liquid water phase at all temperatures down to about -40 deg.C.
(ref. Rosenburg R, ‘Why is ice slippery’, Physics Today, Dc.2005).
Liquid water dissolves CO2 and ionic diffusion occurs through liquid water.
Richard
Re: Gail Combs on June 7, 2010 at 5:15 pm
Gail, I tossed that out there for general consideration without speculating myself, but there was something about the ice core low CO2 levels that was tickling my brain. The plant data show much lower levels, and yes I agree that could be a local source/sink issue but there it is. Well, for a long time now I’ve read speculation here about CO2 diffusing out of high-concentration air bubbles in the ice cores, yielding too-low readings. Where does it go? If there was an “averaging” occurring in the ice in all those layers over all that time, wouldn’t the low-concentration bubbles gain some CO2? And there it is, the plant data showing much lower levels than what the ice cores do.
Of course in some ways it sounded silly, as with that graph (link to US-B&W version) not getting smoother further back in time, and sudden swings from low to high being preserved. So it doesn’t look like such “averaging” is a long-term effect due to the lack of long-term smoothing, and swing preservation says it’s not short term acting on “young” ice. However, I can’t rule out the high and low points of the swings weren’t clipped, especially with the lower plant numbers, thus can’t rule out a short-term issue, when the young ice wasn’t “set” and under so much pressure… So I posted it without speculation to see if someone else might get that same tickle.
Looking at that graph again though, I’ve noticed a major irritation. The CO2, temperature, and CH4 lines match up well, and this graph has too coarse a time scale to really see any CO2/temperature lag. But what is noticeable is that insolation line. Going by the peaks, first temperature and the other two peak, then insolation peaks. It’s not a perfect match, and they don’t always line up that way, especially on minor peaks where other factors might be in play, but that certain pattern is far more likely to be seen than otherwise. Further, quite often insolation will be rising, then the others peak, then insolation peaks. Around 160,000 BP there’s an oddball, temp and CO2 have a minor peak while insolation is going down, but methane is also going down. Nearly always the pattern is there, insolation rises, the three peak, then insolation peaks.
Which brings about the following question: Is there some negative feedback mechanism keeping insolation from raising the other readings past a certain point? Looks to me like an albedo change is indicated, namely increased cloud cover. Past a certain temperature, increased cloud cover due to increased evaporation acts to dampen any further increases in temperature. (One may choose to argue it is CO2 and/or methane, both GHG’s, that are somehow triggering some sort of temperature-limiting mechanism. Which would be interesting.)
…
Ah crud, and here I was, I really was, trying to keep near the “attribution” topic while simply mentioning something about the ice cores I had found that makes me question them somewhat, and noting the huge pre-SUV rises. Oh well. Around here there tends to be a love/hate relationship to ice cores, we like the CO2/temp lag and the higher temps, don’t like the presumably-low CO2 measurements. Guess we’ll have to treat them like we do the historical temperature records, we have reasons to (strongly) suspect they could be a mess but they are what we have to work with, so let’s see what real info we can tease out of them. And look there, possible evidence of a global temperature-limiting mechanism!
#
#
Michael Larkin says:
June 7, 2010 at 6:17 pm
This is coming from left field, but if no one has mentioned it, I wondered if it might have relevance. If not all “fossil fuel” is in fact biogenic, then wouldn’t that affect C12/C13 ratios?
I have no opinion either way, incidentally, about abiogenic oil or anthopogenic CO2 increase. The more I read pro- and contra, the more I come to realise that agnosticism is the only sane position for me.
________________________________________________________________________
I looked into that a while ago and you run into a confounding factor – microbes living in the coal
Carbon-14 in Coal Deposits
“…it would also seem to represent a problem for the established geologic timescale, as conventional thought holds that coal deposits were largely if not entirely formed during the Carboniferous period approximately 300 million years ago. Since the halflife of carbon-14 is 5,730 years, any that was present in the coal at the time of formation should have long since decayed to stable daughter products. The presence of 14C in coal therefore is an anomaly that requires explanation….”
bacteria/fungi hypothesis: Lowe then makes a reasonable case for fungi and bacteria – there are fungi that can degrade lignite (Polyporus versicolor and Poria montiola), as well as autotrophic “thiobacillus-like” bacteria that oxidize pyrites in coal, and he points out that bacteria have been found 3km underground apparently living on granite. Lowe states that fungal and bacterial activity is particularly likely in warm, damp coal exposed to air, and he points out that microbial action only has to result in the deposition of ~0.1% by weight of modern carbon in the coal to produce an apparent age of 45,000 years for the specimen.
Since Lowe’s paper, there have been many more reports of deep subterranean bacteria, which apparently form a heretofore unrecognized ecosystem deep below the earth in rocks and in oils (abstracts below). Presumably most of these bacteria never interact with the “modern” 14C of the atmosphere. But some deep bacterial activity apparently can result in increased concentrations of 13C….”
On the next Ice Age: Everyone seems to forget the default setting for the Earth is a deep freeze.
About the onset of an ice age, this comment from E.M.Smith:
From “Ice Age” by John and Mary Gribbin (a wonderful read, gives the richness of the characters in the discovery of the ice ages, the history of the process, and a gentle introduction to some of the science involved.):
Pg.53: […]the single most important thing to emerge from these discussions was Koppen’s realization of the key season in the Ice Age saga. Adhemar and Croll had thought that the decisive factor in encouraging Ice to spread across the Northern Hemisphere must be the occurrence of extremely cold winters, resulting in increased snowfall. At first, Milankovitch had shared this view. But it was Koppen who pointed out that it is always cold enough for snow to fall in the Arctic winter, even today, and that the reason that the Northern Hemisphere is not in the grip of a full Ice Age is because the ‘extra’ snow melts away again in the summer.
[EMS: Note that the Southern Hemisphere is similarly irrelevant to the ice age cycle since it is always cold enough for snow to stay frozen. It just doesn’t change enough to matter.]
Pg 54: He reasoned that the way to encourage the ice to spread would be to have a reduction in summer warmth, because then less of the winter snowfall would melt. If less snow melted in summer than fell in winter, the ice sheets would grow – and once they had started to grow, the feedback effect of the way the ice and snow reflect away incoming solar energy would enhance the process.
Pg 57: It isn’t so much that Ice Ages occur when the astronomical influences conspire to produce particularly cool summers, rather what matters is that Interglacials only occur when the astronomical influences conspire to produce unusually warm summers, encouraging the ice to retreat. Without all three of the astronomical rhythms working in step this way, the Earth stays in a deep freeze.
End Quote.
So, to summarize:
1) The south pole doesn’t matter to the process, it’s always frozen.
2) We are normally in a long ice age and only pop out for short intervals when conditions are just right.
3) The ‘just right’ is Northern Hemisphere summers warm enough to melt the snow and ice.
4) Warm enough is when the N. hemisphere: must be pointed at the sun in summer when: at close approach to the sun with the right elliptical shape, with the pole tilted over far enough, with… or we freeze.
I would add a note that I think it is particularly illuminating that we are near the end of an Interglacial (next stop is an ice age), the only thing that keeps it away is the summer Arctic ice and snow melt. So what is the AGW crowd in histrionics about? That the Arctic ice and snow are not sticking through the summer… Think about it…
Right now Earth is nearest the sun during northern winters, farthest from the sun during northern summers.
E.M.Smith sums up my concerns very nicely. Thank you Mr. Smith and also Willis for giving us a chance to thrash out the pros and cons of the increase in CO2 is caused by man hypothesis.
Richard Courtney June 8, 2010 at 2:50 am made the point about the possibility of CO2 concentration dropping substantially and by implication of course the other way round.
This was discussed at great length on my thread on this subject over at Air Vent when Becks figures around 1940 were dissected.
The Oceans are a giant source or sink dependent on temperatures. I believe Richard himself gave a very good indicator of the Billions of Tonnes that could theoretically cycle between ocean and atmosphere on another thread a few months ago.
So a considerable and rapid change in atmospheric concentration is theoretically possible IF the right conditions prevail. It does happen sometimes that the Arctic and Antarctic can be warming or cooling in unison.
http://i43.tinypic.com/a4wiu8.png
If this should occur at the same time there and in most other parts of the world the absorption or outgasing can be very considerable. Of course the time scale of cause and effect needs to be taken into account-1 year? 1000 years?
I have asked before if anyone knew of the amount of outgassing or absorption of CO2 per one tenth of a degree of oceanic temperature change, in order to try to determine if the temperature changes we witness in the graph can possibly have the effect of significantly affecting the atmospheric concentration. However to date no one has come up with an appropriate formula.
Tonyb
Willis Eschenbach says:
June 8, 2010 at 12:52 am
tallbloke says:
June 7, 2010 at 11:19 pm
1) Why no account taken of the work done by Georg Ernst Beck in gathering empirical records which show that co2 levels were much more upsy downsy in the C19th and early C20th than the warmists would like to have us know?
I discussed Beck’s work (and Beck himself commented) on my previous thread.
I think your reasons for rejecting his work don’t hold up in the light of his latest paper:
http://www.biokurs.de/treibhaus/CO2_versus_windspeed-review-1-FM.pdf
It shows the levels of this “well mixed” gas were at around 332 back in the 1800’s and as high as today in 1940. I trust direct air measuements more than ice core proxies from the south pole where co2 levels are generally lower anyway.
Richard S Courtney says:
June 8, 2010 at 4:24 am
I agree it is a pity that we cannot determine the reliability of either of them.
1. Do you think that the agreement between ML, Barrow, South Pole and AIRS data is a coincidence or doy you believe they are all wrong for the same reason?
2. Do you think the same argument applies to all 8 ice core datasets?
I have learned quite a lot from the many commenters on this thread, most notably (but not exclusively) Richard S Courtney and tonyb. I also share the concerns of anna v regarding both energy and the mixing issue, and am not nearly as confident about the procedures at Mauna Loa (and elsewhere) as Willis seems to be, but I am grateful for the clear way in which he has outlined exactly what they do with the initial measurements.
In tonyb’s first post yesterday, he included a link to a graph that I think sums up a lot regarding the effects of CO2 on ‘global warming’.
I have long had a poster-sized version of the equivalent of this graph hung up on my office wall, except that the curves have been labelled f(t) and g(t), and no reference is made to climate issues. As people come and go, I ask them what they think of the connection between the two functions. So far, the most common responses have been:
– No connection.
– Very little connection.
– Is this a trick question?
– You’re joking, right?
/dr.bill
Fred H. Haynie:
At June 7, 2010 at 6:20 pm you say:
“A more informative exercise is to use the Scripps seasonally adjusted monthly CO2 averages, convert to global gigitons/year (annual difference or accumulation rate) and compare the cyclic behavior to the relatively straight line for anthropogenic emissions. They average about the same but the natural cycles vary by orders of magnitude. This does not play well for cause and effect.”
With respect it is not true that “This does not play well for cause and effect”, but it does deny the CO2 ‘budget’ analyses used by e.g. the IPCC to determine “accumulation” of anthropogenic CO2.
Your comment pertains to the fact that the annual pulse of anthropogenic CO2 into the atmosphere should relate to the annual increase of CO2 in the atmosphere if one is directly causal of the other, but their variations greatly differ from year to year. This necessary relationship is because the direct causation would require that the carbon cycle were so near to saturation that the system could not sequester all the anthropogenic addition.
However, the rates of the seasonal variations to atmospheric CO2 concentration demonstrate that the system is not near such saturation.
(ref. Rorsch A, Courtney RS & Thoenes D, ‘The Interaction of Climate Change and the Carbon Dioxide Cycle’ E&E v16no2 (2005) ).
A caveat is that the use of annual data for anthropogenic CO2 may be an error. Some data on e.g. fuel consumption may not be collated in time so may be misallocated to an adjacent year, so 2-year smoothing of the data is justifiable. And some countries may use different 12-month periods for their accounting years which – together with the reason for 2-year smoothing – provides justification for 3-year smoothing. But smoothing of the data over 4 or more years is not justifiable.
The IPCC uses 5-year smoothing to get the data to ‘fit’ its model of ‘accumulation’ of anthropogenic emissions to ‘fit’ the observed rise in atmospheric CO2 concentration as determined at Mauna Loa.
However, in our paper that I cite here and outline at June 7, 2010 at 2:43 am above, we provided six models that each match the annual data for the anthropogenic emission to the observed rise in atmospheric CO2 concentration as determined at Mauna Loa. None of these models used any smoothing or other adjustment to any of the data. As I explained above (at June 7, 2010 at 2:43 am):
“Our paper then used attribution studies to model the system response. Those attribution studies used three different basic models to emulate the causes of the rise of CO2 concentration in the atmosphere in the twentieth century. They each assumed
(a) a significant effect of the anthropogenic emission
and
(b) no discernible effect of the anthropogenic emission.
Thus we assessed six models.
These numerical exercises are a caution to estimates of future changes to the atmospheric CO2 concentration. The three basic models used in these exercises each emulate different physical processes and each agrees with the observed recent rise of atmospheric CO2 concentration. They each demonstrate that the observed recent rise of atmospheric CO2 concentration may be solely a consequence of the anthropogenic emission or may be solely a result of, for example, desorption from the oceans induced by the temperature rise that preceded it. Furthermore, extrapolation using these models gives very different predictions of future atmospheric CO2 concentration whatever the cause of the recent rise in atmospheric CO2 concentration.”
This provides an apparent paradox. The annual anthropogenic emission of CO2 should relate to the annual increase of CO2 in the atmosphere if one is directly causal of the other but these two parameters do not correlate. But – using each of our different models – we were able to model the increase of CO2 in the atmosphere as being a function solely of the annual anthropogenic emission of CO2. And we did not use any ‘fiddle factors’ such as the 5-year-averageing used by the IPCC to get a ’fit’. (Adoption of that smoothing really is a disgrace. There can be no justification for it because there is no known physical mechanism that would have that effect).
The apparent paradox is resolved by considering the calculated equilibrium CO2 concentration values. These show an important difference between the three models. They diverge.
But each model indicates that, for each year, the calculated CO2 concentration for the equilibrium state is considerably above the value of the observed CO2 concentration in the air. This demonstrates that each model indicates there is a considerable time lag required to reach the equilibrium state when there is no accumulation of CO2 in the atmosphere.
The short term sequestration processes can easily adapt to sequester the anthropogenic emission in that year. But, according to our models, the total emission of any year affects the equilibrium state of the entire system. Some processes of the system are very slow with rate constants of years and decades. Hence, the system takes decades to fully adjust to the new equilibrium. And the models predict the atmospheric CO2 concentration slowly rising in response to the changing equilibrium condition.
Simply, we demonstrated that it is possible that the total natural flux of CO2 from the Earth to the air may increase over time as a response to increasing anthropogenic emission. And this provides an explanation of why the apparent accumulation of CO2 in the atmosphere continued when in two subsequent years the anthropogenic flux into the atmosphere decreased (this happened, for example, in the years 1973-1974, 1987-1988, and 1998-1999).
So, in summation, your observation does disprove the IPCC model of “accumulating” anthropogenic CO2 in the air, but it does not negate the possibility that the anthropogenic emission is responsible for the recent (i.e. since 1958) rise in atmospheric CO2 concentration. And our models demonstrate that the cause of the recent rise may be entirely natural, or entirely anthropogenic, or some combination of anthropogenic and natural causes.
So, a question:
Is the cause of the rise in atmospheric CO2 concentration.natural or anthropogenic in part or in whole?
Answer:
God alone knows.
Richard
Malaga View says:
June 8, 2010 at 3:44 am
……
of a discharge plume as cool North Pacific water warming on its trip southward swings westward. The warming results in a steady elevation in atmospheric CO2. This map seems to confirm that Mauna Loa overestimates global atmospheric CO2 levels.
Then so does the AIRS satellite data from the mid-troposphere – and ‘coincidentally’ – by almost exactly the same amount.
=====================================================
So it looks like the MLO and Ice Core data have all sorts of associated issues….
So this CO2 analysis seems to have very questionable foundations… so say the least.
ML data is fine. It is validated by a host other data. Ice Core data is consistent across all datasets. Ice age concentrations show there is a lot of variability – just not so much in the inter-glacial periods, i.e. +/-20ppm.
Malaga View says:
June 8, 2010 at 12:45 am
Have you actually looked at the scale on your linked maps. We do know about the seasonal cycle, by the way.
tonyb says:
June 8, 2010 at 5:04 am
I have asked before if anyone knew of the amount of outgassing or absorption of CO2 per one tenth of a degree of oceanic temperature change, in order to try to determine if the temperature changes we witness in the graph can possibly have the effect of significantly affecting the atmospheric concentration. However to date no one has come up with an appropriate formula.
The most recent evaluation of this in a Nature paper came out with a median value of ~8 ppm CO2/ºC (upper limit ~20).
David C. Frank, Jan Esper, Christoph C. Raible, Ulf Büntgen, Valerie Trouet, Benjamin Stocker, & Fortunat Joos. Ensemble reconstruction constraints on the global carbon cycle sensitivity to climate. Nature, 2010; 463 (7280): 527 DOI: 10.1038/nature08769
The term well mixed in some cases has been misused. As concerning the Scripps data, they tend to define it as when there is enough turbulance in the surface boundary layer that concentrations are relatively uniform and are representive of a background concentration. That layer is roughly around a kilometer between the surface and the base of clouds. They only include measurements in their monthly averages that meet these criteria. “Well mixed” and “E-fold” should not be used to try to explain the global uniformity of these measurements. There are other physical processes controlling the atmospheric concentration of CO2. For example, how long does it take for CO2 “out-gassing” from the tropical ocean to be absorbed into clouds and then be returned to the ocean as rain? What fraction is returned to the ocean and what fraction is transported to the top of the atmosphere in towering clouds? How do wind patterns carry CO2 from it’s tropical source to it’s Artic sink? How and why do both source and sink change with time? Until these processes are better understood, it is not much better than speculation to attribute some fraction of the rise in background CO2 levels to anthropogenic emissions.