An important new paper published today in Global and Planetary Change finds that changes in CO2 follow rather than lead global air surface temperature and that “CO2 released from use of fossil fuels have little influence on the observed changes in the amount of atmospheric CO2” The paper finds the “overall global temperature change sequence of events appears to be from 1) the ocean surface to 2) the land surface to 3) the lower troposphere,” in other words, the opposite of claims by global warming alarmists that CO2 in the atmosphere drives land and ocean temperatures. Instead, just as in the ice cores, CO2 levels are found to be a lagging effect ocean warming, not significantly related to man-made emissions, and not the driver of warming. Prior research has shown infrared radiation from greenhouse gases is incapable of warming the oceans, only shortwave radiation from the Sun is capable of penetrating and heating the oceans and thereby driving global surface temperatures.
The highlights of the paper are:
► The overall global temperature change sequence of events appears to be from 1) the ocean surface to 2) the land surface to 3) the lower troposphere.
► Changes in global atmospheric CO2 are lagging about 11–12 months behind changes in global sea surface temperature.
► Changes in global atmospheric CO2 are lagging 9.5-10 months behind changes in global air surface temperature.
► Changes in global atmospheric CO2 are lagging about 9 months behind changes in global lower troposphere temperature.
► Changes in ocean temperatures appear to explain a substantial part of the observed changes in atmospheric CO2 since January 1980.
► CO2 released from use of fossil fuels have little influence on the observed changes in the amount of atmospheric CO2, and changes in atmospheric CO2 are not tracking changes in human emissions.
The paper:
The phase relation between atmospheric carbon dioxide and global temperature
- a Department of Geosciences, University of Oslo, P.O. Box 1047 Blindern, N-0316 Oslo, Norway
- b Department of Geology, University Centre in Svalbard (UNIS), P.O. Box 156, N-9171 Longyearbyen, Svalbard, Norway
- c Telenor Norway, Finance, N-1331 Fornebu, Norway
- d Department of Physics and Technology, University of Tromsø, N-9037 Tromsø, Norway
Abstract
Using data series on atmospheric carbon dioxide and global temperatures we investigate the phase relation (leads/lags) between these for the period January 1980 to December 2011. Ice cores show atmospheric CO2 variations to lag behind atmospheric temperature changes on a century to millennium scale, but modern temperature is expected to lag changes in atmospheric CO2, as the atmospheric temperature increase since about 1975 generally is assumed to be caused by the modern increase in CO2. In our analysis we use eight well-known datasets; 1) globally averaged well-mixed marine boundary layer CO2 data, 2) HadCRUT3 surface air temperature data, 3) GISS surface air temperature data, 4) NCDC surface air temperature data, 5) HadSST2 sea surface data, 6) UAH lower troposphere temperature data series, 7) CDIAC data on release of anthropogene CO2, and 8) GWP data on volcanic eruptions. Annual cycles are present in all datasets except 7) and 8), and to remove the influence of these we analyze 12-month averaged data. We find a high degree of co-variation between all data series except 7) and 8), but with changes in CO2 always lagging changes in temperature. The maximum positive correlation between CO2 and temperature is found for CO2 lagging 11–12 months in relation to global sea surface temperature, 9.5-10 months to global surface air temperature, and about 9 months to global lower troposphere temperature. The correlation between changes in ocean temperatures and atmospheric CO2 is high, but do not explain all observed changes.
Stephen Wilde says:
September 11, 2012 at 7:59 am
As per your earlier business analogy. More turnover (emissions) less more expenses (local absorption) comes to zero and the natural underlying balance remains unaffected.
More turnover = more balanced inputs and outputs, but the net balance is more own money than is seen as gain at the end of the year. No matter how fast your own money goes into the drain, your bussiness makes a loss…
Smokey says:
September 11, 2012 at 10:24 am
John Finn,
The trend in CO2 follows the trend in temperature. Look at this graph, and you will see that is a fact. That cause and effect relationship occurs on all time scales, out to at least 400,000 years.
Smokey
I realise you struggle with this stuff but data prior to the last 150 years is not terribly relevant to what’s happening at the moment. Basically we (humans) are providing an additional source of atmospheric CO2 which is totally independent of temperature. The other natural exchanges are still taking place – as we can see from the small annual fluctuations (i.e the CO2 rates of change) but the the long term trend of ~2ppm per year is due to the burning of fossil fuels.
John Finn says:
September 11, 2012 at 11:10 am
“Basically we (humans) are providing an additional source of atmospheric CO2 which is totally independent of temperature.”
Attribution on that basis employs the logical fallacy of post hoc, ergo propter hoc.
However, the fact that it is independent of temperature is what allows us to discount its effect on overall atmospheric levels, because that level is temperature dependent.
John Finn says:
September 11, 2012 at 10:09 am
Bart is 100% convinced that some unknown natural source (maybe the deep oceans) is responsible for a continuous increasing trend of now 100+ ppmv in the atmosphere (70+ since 1960). By using a completely arbitrary offset, the same factor gives both the short term variability and the trend. He doesn’t accept in any way that the trend may be completely spurious, despite that this violates about all known observations. Thus in his opinion, these observations must be wrong.
– It violates the mass balance.
– It violates the 13C/12C balance.
– It violates the measured uptake by the oceans.
– It violates the calculated uptake by the biosphere (oxygen balance).
– It should include processes which are quite modest on rapid changes (4-5 ppmv/°C), give huge responses (over 100 ppmv/°C) for medium speed changes and back to modest (including eating away the medium fast responses) for medium to very long time changes (8 ppmv/°C). Quite remarkable for natural processes.
Where he goes wrong is that he thinks that one process is responsible for both the fast response and the medium response (in that case, the same variable may be responsible for variability and trend). But the fast response is from the ocean surface layer (and fast changes in vegetation), which only can absorb 10% of the change in the atmosphere, 90% is going into much slower processes: deep oceans, more permanent sequestering in land carbon,…
And last but not least, there are two variables at work: temperature and human emissions. The latter alone is already average twice the observed increase… That leaves no room for fantasies of a continuous increasing inflow of CO2 from an unknown proces from a sustained temperature difference of a few tenths of a °C…
Ferdinand Engelbeen says:
September 11, 2012 at 11:37 am
“By using a completely arbitrary offset, the same factor gives both the short term variability and the trend.”
Your way doesn’t match in phase.
“He doesn’t accept in any way that the trend may be completely spurious…”
You have no evidence that it is, and it is quite impossible when the correlation is seamless across all frequency components.
“It violates the mass balance.”
As we have demonstrated ad nauseum in this thread, this argument is completely fallacious.
“It violates the 13C/12C balance.”
There are many possible explanations for the balance, as others have brilliantly commented on this thread.
“It violates the measured uptake by the oceans.”
Measurements which are low accuracy, were never intended for the purpose, and which do not measure the pertinent information.
“It violates the calculated uptake by the biosphere (oxygen balance).”
Of which we do not have a full accounting.
“It should include processes which are quite modest on rapid changes…”
Your own strawman.
“Where he goes wrong is that he thinks that one process is responsible for both the fast response and the medium response…”
It is not what I think. That is what the temperature-CO2 rate of change data indicate – the response is seamless across frequency.
“The latter alone is already average twice the observed increase…i>
Again with the mass balance fallacy. Your edifice is built on a foundation of sand.
Bart says:
September 11, 2012 at 11:34 am
However, the fact that it is independent of temperature is what allows us to discount its effect on overall atmospheric levels, because that level is temperature dependent.
In your fantasy, based on a completely arbitrary offset. The correlation with the human emissions is a near fit, while temperature is far more variable:
http://www.ferdinand-engelbeen.be/klimaat/klim_img/temp_co2_1900_2004.jpg
and
http://www.ferdinand-engelbeen.be/klimaat/klim_img/acc_co2_1900_2004.jpg
Maybe just coincidence, but I don’t know of any natural process that can mimic the human emissions both in ratio of quantity and d13C level in such a linear way…
Bart says:
September 11, 2012 at 10:56 am
Sure it does.
No it doesn’t. The clue should be in the rate of temperature change which is positive some times and negative others, i.e.
Temp goes UP -> CO2 goes UP
Temp goes DOWN -> CO2 goes UP
There is a disconnect between Temperature and CO2. This becomes obvious when we look at the data over several years or decades, e.g. between 1982 and 2102 (i.e. 30 years)
Temp change = ~0.5 deg
CO2 change = 50 ppm (341ppm -> 391ppm)
Perhaps you think the 0.5 deg temp rise has produced a CO2 rise of 50ppm – giving a relationship of 100 ppm per deg C ?? Ok what about the 1998-2012 period
Temp change = ~0.1 deg
CO2 change = 25ppm (366ppm->391ppm)
Now we‘ve got a relationship of 250ppm per deg C.
You’ve got a blind spot with this stuff. The comments in the rest of your post suggests you think that if temperatures dropped by say half a degree (roughly to1970s levels) then CO2 would continue to rise indefinitely, i.e. they would continue to rise above the current 391ppm level and that this would have nothing to do with fossil fuel burning. This implies you don’t acknowledge that some equilibrium level would eventually be reached – despite what all the evidence tells us. Sheeesh!!
Ferdinand Engelbeen says:
September 11, 2012 at 11:49 am
“The correlation with the human emissions is a near fit, while temperature is far more variable”
Those are not “near fits”. They are coarse measures of integrated variables. Integration is a low pass filtering operation, which severely attenuates the fine detail, and you are left with easy to fit low order polynomials. It’s a coin toss at that point – emissions are going up. If CO2 happens also to be going up, you will get a similar result via scaling.
You cannot say anything about the goodness of the fit unless you can match other detail which is far less likely to match unless there is a genuine relationship. And, that is what the temperature-CO2 rate of change relationship does.
John Finn says:
September 11, 2012 at 12:05 pm
“The clue should be in the rate of temperature change which is positive some times and negative others, i.e.
Temp goes UP -> CO2 goes UP
Temp goes DOWN -> CO2 goes UP “
Does anyone here understand calculus?
The relationship is
dCO2/dt = k*(T – T0)
The integral of the scaled and offset temperature is, to a very high degree of accuracy, an almost perfect replication of the CO2 record.
“This implies you don’t acknowledge that some equilibrium level would eventually be reached – despite what all the evidence tells us.”
Read what I wrote again, Sparky.
John Finn says:
September 11, 2012 at 12:05 pm
Temp change = ~0.1 deg
CO2 change = 25ppm (366ppm->391ppm)
Now we‘ve got a relationship of 250ppm per deg C.
What Bart says is indeed what you calculated:
dCO2/dt = k*(T – T0)
No matter how long the period: the rate of change of CO2 is positive as long there is an offset with an arbitrary baseline temperature level. No response from increased levels in the atmosphere. No response from increasing sinks. The sky is the limit…
Think of the 5000 years of increase in temperature between a glacial and an interglacial period or the 10,000 years above the base temperature during an interglacial or the 100,000 years of a glacial below the baseline…
But of course the ice core data are not accepted by Bart…
As I have shown before, there is an excellent fit between the differential equation temperature model and CO2. This plot was using GISTEMP. As others have noted (I think it was Tallbloke), the agreement in the derivative is even better using the satellite measurements. And, since there is a 1-1 mapping between the derivative and the integral of the derivative, the agreement in the integrated value would likewise be better.
But, I’m not going to do that exercise for you (meaning Ferdinand and John Finn). It is not my job to spoon feed you to make up for the deficiencies in your educations. I have shown you that Ferdinand’s arguments are shaky at best, and totally wrong at worst (e.g., the silly “mass balance” argument). I have shown you that the correlation with temperature explains every significant quality about the CO2 rise. If you choose to ignore what I have patiently (and, I am getting very impatient now) explained, then just go ahead and do it. It’s not my problem, it is yours.
Ferdinand Engelbeen says:
September 11, 2012 at 12:36 pm
Now, you are being completely disingenuous. You know quite well that the full model is
dCO2/dt = -CO2/tau + k*(T-To)
where “tau” is a long time constant. You know I have explained at length about the non-stationarity of the model, and the potential for regime changes. If you have sunk so low in desperation that you are willfully misrepresenting my position to make headway with random observers, then we really have reached the end of the road.
Bart says:
September 11, 2012 at 12:51 pm
Now, you are being completely disingenuous. You know quite well that the full model is
dCO2/dt = -CO2/tau + k*(T-To)
It doesn’t matter that you include the -CO2/tau with a long time constant, the effect is the same: a continuous increase of CO2 as long as there is a temperature difference with the baseline. While the human emissions should disappear rapidely.
But that includes that the extra natural inputs are extremely high, as good as the extra natural outputs. The net effect is a lot of extra throughput, but still no substantial contribution of the natural cycle. That is what the mass balance shows. Even if you ridicule that, it simply proves that your theory is bogus and that the trend is not caused by temperature, but by a different process.
But that all indeed is repeatedly said to no avail…
John Finn says:
September 11, 2012 at 12:05 pm
For an alternative view of the increase, based on observations, see the estimates of the contribution of temperature and precipitation in the rate of change of the CO2 increase by Pieter Tans, at his speech during the festivities of 50 year Mauna Loa CO2 measurements, page 11 and following:
http://esrl.noaa.gov/gmd/co2conference/pdfs/tans.pdf
Friends:
I write to provide a reality check.
The data does not enable a definitive conclusion as to the cause of the recent rise in atmospheric CO2 concentration because it allows several understandings of the cause.
Therefore,
(a) it is very reasonable for anybody to say what he thinks is the cause and why he thinks that,
but
(b) it is unreasonable to say others must be wrong about the cause because they weight the same data differently.
Richard
“(a) it is very reasonable for anybody to say what he thinks is the cause and why he thinks that,
but
(b) it is unreasonable to say others must be wrong about the cause because they weight the same data differently.”
Agreed.
The position should become clear in due course when we have had time to observed the response of both ocean heat content trends and atmospheric CO2 trends during the later stages of solar cycle 24.
I don’t think we will need to wait as long as 15 years. Probably less than 5 years since there seem to be changes in the rates of acceleration already.
Ferdinand Engelbeen says:
September 11, 2012 at 1:27 pm
“It doesn’t matter that you include the -CO2/tau with a long time constant, the effect is the same: a continuous increase of CO2 as long as there is a temperature difference with the baseline. “
Wrong!
Wrong, wrong, wrong, wrong, wrong!
You do not know what you are talking about. You are just not qualified, and further discussion is futile.
richardscourtney says:
September 11, 2012 at 2:49 pm
“The data does not enable a definitive conclusion as to the cause of the recent rise in atmospheric CO2 concentration because it allows several understandings of the cause.”
I’m sorry, Richard, but you are wrong. The data are very clear. But, in the land of the blind, the one-eyed man is a raving lunatic spouting nonsense that he can “see”, whatever that means.
Bart says:
September 11, 2012 at 12:51 pm
Ferdinand Engelbeen says:
September 11, 2012 at 12:36 pm
Now, you are being completely disingenuous. You know quite well that the full model is
dCO2/dt = -CO2/tau + k*(T-To)
where “tau” is a long time constant. You know I have explained at length about the non-stationarity of the model, and the potential for regime changes.
Ahhh I see – I’m guessing the “potential for regime changes” is the catch all for periods when the model doesn’t work (which are probably quite plentiful). Brilliant! – derive a simple first order differential equation which fits the available data for a given period then when the model is shown to be complete cobblers for other periods in earth’s history it can simply be put down to “regime changes”.
Tell me, Bart, has anyone actually been convinced by your ‘model’?
John Finn says:
September 11, 2012 at 4:08 pm
What a stupid comment. This is a nonlinear, non-autonomous system. Of course there are different operating conditions, which lead to different linearized models. We don’t have enough information to say how it behaves in the large. But, we do have enough to know how it behaves right here and now, or at least since 1958. And, that behavior rules out human attribution.
I’m “late to the party” but let me see if I can shed a little light by going back to the actual CO2 derivative. The data shows three clear drivers of CO2 — plants, people, and climate.
1) Here is the actual derivative of CO2 vs time (along with scaled temperature anomalies).
http://www.woodfortrees.org/plot/esrl-co2/derivative/from:1979/plot/uah/scale:0.25
There are HUGE swings every year in the CO2 derivative. This, of course, is the annual uptake/release of CO2 by plants (one of the things Al Gore got right in his movie). So it you want to know the slope at any given time, the best single predictor is the time of year.
So what BESIDES the annual growth/death of plants drives CO2?
2) Here is the smoothed derivative of CO2 vs time.
http://www.woodfortrees.org/plot/esrl-co2/derivative/mean:12/from:1979/plot/uah/scale:0.25/plot/none
By averaging over 12 month, the intra-annual changes disappear, leaving only the interannual changes.
Perhaps the single most important feature is that the smoothed derivative is ALWAYS POSITIVE. That means that year-in and year-out, the CO2 is sloping up. There has never (in the data available) been a 12 month period where the CO2 has dropped. Furthermore, the smoothed slope is slowly increasing over all. What would cause this trend — where CO2 is increasing at an accelerating rate? Burning of fossil fuels comes to mind. Perhaps some other natural process ALSO happens to be putting out more and more CO2 over the last 50 years. But Occam’s Razor suggests that fossil fuels are a more logical answer (and I am pretty sure isotope analysis also supports this, but I don’t have enough expertise off-hand to be more sure).
So if you want the best estimate of slope for CO2 BESIDES plants, the next best predictor is a steady, slowing increasing function of time (matching the increasing use of fossil fuels).
So what BESIDES plants and the burning of fossil fuels drives CO2?
3) Here is the smoothed and offset derivative of CO2 vs time.
http://www.woodfortrees.org/plot/esrl-co2/derivative/mean:12/from:1979/offset:-.14/plot/uah/scale:0.25
Smoothing removes the effect of plants. The offset removes the long-term effect of fossil fuels. What is left is the third (and smallest) effect — that response of nature to temperature. To be sure, the effect shows a strong correlation and is almost certainly real. Yes, nature DOES respond to temperature increases by pumping up CO2. There is indeed a feedback where temperature drives CO2.
That does not in anyway negate the possibility that CO2 could drive temperatures as well. By fiddling with averaging and slopes and offsets, it would be simple to smooth the upward (but jagged) trend in temperature to be and upward but smooth trend that matches CO2 quite well, too.
I am rather sure that BOTH (CO2 drives temperature) and (temperature drives CO2) are real and both are important, since both are pretty clear in the data.
Tim Folkerts says:
September 11, 2012 at 7:17 pm
“What would cause this trend — where CO2 is increasing at an accelerating rate? Burning of fossil fuels comes to mind. Perhaps some other natural process ALSO happens to be putting out more and more CO2 over the last 50 years. But Occam’s Razor suggests that fossil fuels are a more logical answer …”
Not possible. Emissions over time are not constant. They would add additional curvature. There is no room for that when you have removed the part due to temperature.
The temperature you are plotting already has an arbitrary baseline subtracted out from it. The odds that it is the appropriate baseline for determining the effect of temperature are vanishingly small. Hence, you have another variable to match. When you do, the temperature relationship is all you need to accurately reconstruct CO2.
“I am rather sure that …CO2 drives temperature…”
Where is this evident? Nowhere in these plots. The rate of change of CO2 precisely matches the scaled and re-baselined temperature anomaly. Thus, the integrated CO2 level always lags the temperature. The at-least-dominant arrow of causality points necessarily in the direction of: temperature to CO2.
Put another way, it would be absurd to claim that CO2 rate of change drives temperature. Again, the at-least-dominant arrow of causality points necessarily in the direction of: temperature to CO2.
How can this be, when we know, all things being equal, that an increase in CO2 must cause a temperature rise? Because all things are not equal. This is a complex system, with many active feedback paths, which is necessarily stable with dominant negative feedback. Such systems generally resist attempts to move them from where they want to be.
It is also worth noting that, if it were true that higher temperatures drive CO2 higher (which we can see to be true by inspection of the plots) and that added CO2 raises temperature, then this would set up a classic self-reinforcing cycle of positive feedback, which would have seared the surface of the Earth ages ago if the dynamic were not being severely attenuated in some fashion.
Tim Folkerts says:
September 11, 2012 at 7:17 pm
Tim, the main discussion between Bart and me is that Bart thinks that near all medium-term extra CO2 comes from an offset between the observed increase in temperature over time and an arbitrary offset with a baseline temperature.
It is easy to fit both curves, inclusing their curvatory, simply because temperature over the period 1960-2012 in general is upgoing. That doesn’t leave much room for any human influence.
But the trend part is even better reconstructed by a linear relationship of the increase in the atmosphere above equilibrium (where the equilibrium is what the ice cores show as CO2 levels for the past temperatures). That leaves not much room for the temperature influence.
These diametrically opposite positions are now going back and forth for a few years.
The position of Bart violates about all known observations, but he rejects them all, including ice core measurements, the isotope ratio’s, etc.
If you look at the mass balance, if Bart is right, then near all human CO2 is fastly removed, but that implies that near all of the natural input is removed too (the total mass removed is the natural input + halve the human input), thus at the end, there is a lot of throughput, but still no net input from the natural carbon cycle. See here and here
The main point is if the fast response to fast temperature changes and the response to longer term variations have the same cause. Nature has a lot of CO2 sources and sinks, some fast (ocean surface, fast vegetation responses, 1-3 years half life time), others are slower (deep ocean exchanges, more permanent carbon storage via the biosphere, ~40 years half life), and others are very slow (rock weathering, sedimentation).
The interesting point is that the fast processes are limited in capacity: a CO2 change of 100% in the atmosphere gives only a change of 10% in the oceans surface layer (and plant alveoles water?), that is the Revelle factor or buffer factor. Thus only 10% of any change in the atmosphere is captured by the oceans surface. The rest is removed by slower processes.
In my informed opinion, the fast processes are responsible for the fast response of CO2 rate of change to temperature changes. The slow processes, like deep ocean exchanges, are far less influenced by temperature changes and mainly respond to the overall change in CO2 levels, compared to the equilibrium level.
Bart’s one-factor-fits-all implies that the same fast processes are responsible for both the variability of the rate of change and the height and trend of the rate of change.
But such fast processes don’t exist…
Bart says:
September 11, 2012 at 11:25 pm
It is also worth noting that, if it were true that higher temperatures drive CO2 higher (which we can see to be true by inspection of the plots) and that added CO2 raises temperature, then this would set up a classic self-reinforcing cycle of positive feedback, which would have seared the surface of the Earth ages ago if the dynamic were not being severely attenuated in some fashion.
Depends of the factors involved: if the response of CO2 levels to temperature is modest and the response of temperature to CO2 levels is modest, the mutual increase is modest too. As long as the overall gain is less than 2, there is no runaway. Here a plot of the difference between no feedback of temperature on CO2 levels and with a small feedback, both with CO2 lagging temperature changes:
http://www.ferdinand-engelbeen.be/klimaat/klim_img/feedback.jpg
Bart says:
September 11, 2012 at 6:26 pm
John Finn says:
September 11, 2012 at 4:08 pm
What a stupid comment. This is a nonlinear, non-autonomous system. Of course there are different operating conditions, which lead to different linearized models. We don’t have enough information to say how it behaves in the large. But, we do have enough to know how it behaves right here and now, or at least since 1958. And, that behavior rules out human attribution.
No it doesn’t, Bart. We know that CO2 levels have remained in a narrow range during the 800 or so years before ~1900 despite temperature fluctuations over that time.
But let me guess …… you don’t believe ice core data. How convenient!