Guest Post By Frank Lansner, civil engineer, biotechnology.
More words on the topic first presented here: http://icecap.us/images/uploads/FlaticecoreCO2.pdf
I wrote:
It appears from this graph that CO2 concentrations follows temperature with approx 6-9 months. The interesting part is off course that the CO2 trends so markedly responds to temperature changes.
To some, this is “not possible” as we normally see a very smooth rise on CO2 curves. However, the difference in CO2 rise from year to year is quite different from warm to cold years, and as shown differences are closely dependent on global temperatures. Take a closer look:
For this writing I have slightly modified the presentation of UAH data vs. Mauna Loa data:
The relatively rough relationship between CO2 growth per year and global temperatures (UAH) is:
1979: CO2 growth (ppm/year) = 3,5 * Temp.anomaly(K) + 0,7
2008: CO2 growth (ppm/year) = 3,5 * Temp.anomaly(K) + 1,2
1979-2008:
CO2 growth (ppm/year) = 3,5 * Temp.anomaly(K) + 0,95
For 2007, a UAH temperature anomaly approximately – 0,32 K should lead to CO2 rise/year = 0 , that is, CO2-stagnation.
These equations are useful for overall understanding, but so far they don’t give a fully precise and nuanced picture, of course. On the graph, I have illustrated that there is a longer trend difference between CO2 and Temperature. Thus, the “constant” of the equation should be a variable as it varies with time (1979: 0,7 2008: 1,2).
The trend difference means, that from 1979 to 2008 the CO2-rise per year compared to the global temperatures has fallen 0,5 ppm/year, or the other way around: It now takes approx. +0,15 K global temperature anomaly more to achieve the same level of CO2 rise/year as it did in 1979.
How can this be? The CO2 rise/year now takes higher temperatures to achieve?
With the human emissions rising in the time interval 1979-2008, one could imagine that it would be the other way around, that CO2 rises came with still smaller temperature rises needed. But no, its becoming “harder and harder” to make CO2 rise in the atmosphere.
So generally, the human emissions effect appears inferior to other effects in this context at least.
Which effects could hold CO2 rise/year down as we see?
The fact that we today have higher CO2 concentration in the atmosphere than in 1979 does not favour more CO2 release from the oceans. However the fact that we approx 500 million years ago had several thousand ppm CO2 in the atmosphere implies that the 385 ppm today hardly does a big difference.
My guess is, that what we see is mainly the effect of the growing biosphere.
In short: A period with higher temperatures leads to higher CO2 rises/year and thus of course after some years higher CO2 concentration in the atmosphere.
In the period of rising temperatures and CO2 concentration, the biosphere has grown extremely much.
The results of trend analyses of time series over the Sahel region of seasonally integrated NDVI using NOAA AVHRR NDVI-data from 1982 to 1999:
Source: http://www.eoearth.org/article/Greening_of_the_Sahel
Even if we put every European in “Plant a tree”-projects we could never reach a fraction of what mother nature has achieved in Sahel alone over these few years. In Addition, in these areas lots of more precipitation is occurring now. ( If we here have a “point of no return” im not sure Africans would ever want to come back to “normal”. We Europeans want so much to help Africans – but take away the CO2? What kind of help is that? )
In addition, the seas are much more crowded with life, plankton etc.
The biosphere is blooming due to CO2: http://wattsupwiththat.com/2008/06/08/surprise-earths-biosphere-is-booming-co2-the-cause/
So today we have a larger biosphere. Every single extra plant or plankton cell will demand its share of CO2. It takes more CO2 to feed a larger biosphere. More CO2 is pulled out of the atmosphere today than earlier. An enormous negative feedback on CO2 levels. Roughly: Any human CO2-influence would cause bigger biosphere that eventually omits the human CO2-influence.
A rather interesting scenario: What happens if temperatures go down below approx – 0,3 K UAH??
Well first it appears from my rough equation that CO2 levels will go down. We will have negative CO2 rise / year. But the bigger biosphere is still there (!!!) even though temperature and thus CO2 levels suddenly should drop and it will still demand its bigger share of CO2. And more, in these days of Cold PDO and especially more precipitation due to the solar condition, we might see more CO2 washed faster out of the atmosphere.
This adds up to my belief, that a cooling after a longer warming trend, mostly due to the bigger biosphere, could be accompanied by quite rapid fall in CO2 levels. Faster that temperature raise leads to CO2 rise? In short, I postulate: CO2 often falls quicker than it rises:
(I am very aware that the data Ernst-Georg Beck has gathered has had a lot of critic. I will not here be a judge, but I think its fair to show that Becks data to some degree matches my expectations, even though the level of CO2 appears high. But I am no judge of what is too high etc.)
So what to expect now? First of all, how about the present cooling??
We should be able to see the big Jan 2008 dive in global temperature in CO2? Well yes, this dive should 6-9 months appear thereafter. And if we take a look at Mauna Loa data released Aug 3, nicely in the 6-9 months time frame after Jan 2008, we saw a dive.
However, this dive was mostly removed from Mauna Loa data 4 Aug 2008, so its hard to judge anything about 2008.
Antarctic ice core data shows that in the period 1890-1940 there was a flat development approx 8 ppm from 300 ppm to 308 ppm.
We have seen first in this writing, that the CO2 is very responsive to temperature changes 1979-2008. So how come the warmer temperatures 1920-40´s has no effect at all on the extremely straight Antarctic CO2 curve?
Is there a mismatch between extremely flat Antarctic CO2 data on one side and Mauna Loa data/UAH data on the other side? If so, which data sets are correct? Mauna Loa/UAH or Antarctic ice cores?
Dear Frank,
Had some connection problems, therefore the testmessage… Now I will put the message in two parts, probably the message was too long, or too many links in it…
I think we are going to agree somewhere…
I agree that longer term processes as well as in the oceans as in the biosphere have their influence, but these too are limited: once the temperature of the full ocean (including the deep oceans) is increased by a certain temperature, the related increase of CO2 in the atmosphere will hinder a further increase of CO2 from the oceans. Thus a new equilibrium will be reached. The same for land occupation and sea algues: a doubling of CO2 will increase the amount of carbon fixed over decades, but that is less than double: even with all other necessities available in unlimited quantities (which are the limiting factors in many cases), that gives about 20-40% extra growth, not 100%. Thus even there, a new equilibrium is reached in a few decades (for extra CO2 over current land occupation) to millennia (for ice sheet retraction and plant spread).
The combined effects of oceans and vegetation are known from ice cores: dCO2/dT is about 8 ppmv/°C, pretty constant over 4 ice age – interglacial cycles in 420,000 years and surprisingly linear, despite that a number of players in this game are acting far from linear. See:
http://www.ferdinand-engelbeen.be/klimaat/klim_img/Vostok_trends.gif
No effect is seen from prolonged, sustained cold or warm periods over thousands of years, thus T has no effect, only dT has an effect. That is true for the whole pre-industrial period, including the MWP-LIA cooling, as seen in the Law Dome (and other) ice cores. See:
http://www.ferdinand-engelbeen.be/klimaat/klim_img/law_dome_1000yr.jpg
What about the current period? Based on the ice core dCO2/dT relationship, the increase in temperature since the LIA has added not more than 6 ppmv to the atmosphere to reach a new equilibrium. That is all. Humans meanwhile have added about twice as much CO2 to the atmosphere as what is measured as increase in the atmosphere. But let us see what is important: dCO2/T or dCO2/dT.
Here the graph of dCO2/T from your formula:
http://www.ferdinand-engelbeen.be/klimaat/klim_img/lans_trend.jpg
I had to adjust the “constant” somewhat, as the early decades of the Hadcrut3gl are colder, thus needed a higher constant and the last decades did give an overshoot. The start for the best fit was at about 1.2 ppmv/yr the end at 0.7 ppmv/yr linearly decreasing in between.
Part two:
Here the graph of the dCO2/dT from my formula:
http://www.ferdinand-engelbeen.be/klimaat/klim_img/egbn_trend.jpg
I have increased the “baseline” of my formula from 0.55*emissions/year to 0.57*emissions/year, as it seems that the ratio of what rests in the atmosphere is slightly increasing.
If you compare the graphs, both follow more or less the temperature variability and there is little difference in performance. But what about the “constant” compared to the “baseline” approach?
The “constant” need to decrease over time to allow the temperature trend over 45 years to follow the increase. As the “constant” is a mixture of everything except temperature, the emissions are included, but these are increasing markedly over the full period. Thus something else must pull out the extra CO2. That may be vegetation growth, but as that is measured via oxygen measurements, vegetation indeed is a sink for CO2, but by far not large enough to absorb the extra near 3 ppmv/yr which is the difference of emissions with the “constant” at the end of the period. See:
http://www.sciencemag.org/cgi/reprint/287/5462/2467.pdf and
http://www.bowdoin.edu/~mbattle/papers_posters_and_talks/BenderGBC2005.pdf
Moreover, the “constant” occupies about 2/3rd of the yearly increase of CO2 in average, thus temperature is only responsible for 1/3rd of the increase, the rest anyway is from the emissions (or one need even more sink).
Part three:
The “baseline” approach is in line with what is found in the ice cores and reflects what happens if we may assume that the CO2/temperature relationship is a dynamic equilibrium. In that case, the trend (baseline) itself is (near) fully caused by the emissions and temperature changes only influence the variability around the trend, not the trend itself.
We can use both calculated trends of dCO2/yr with the total accumulation of CO2 in the atmosphere:
http://www.ferdinand-engelbeen.be/klimaat/klim_img/lans_egbn_acc_trend.jpg
Both formulas follow the real trend pretty well, where the 57% emissions approach is a near fit, while the temperature approach is slightly slow at first and a little too fast at the end.
As you can see, the “baseline” approach fits all CO2-T relationships over all periods of the past near million years, the “constant” approach need a highly variable “constant” to match the CO2 trend in different periods and is not even applicable for pre-industrial times.
Thus it looks like that the temperature influence is limited in time and isn’t responsible for the current trend. That both T and dT give a nice correlation with dCO2/yr is not coincidence: both are highly variable and don’t show any period where T and dT are flat for more than a few years. Thus it is impossible to know which one is the driver for the variability of dCO2/yr. But from ice cores we know that it is dT.
About emissions and temperature as drivers in ice cores, here are two graphs comparing total emissions with total CO2 increase and temperature with total CO2 increase, as measured in ice cores (Law Dome and others) 1900-1959:
http://www.ferdinand-engelbeen.be/klimaat/klim_img/temp_co2_1900_1959.jpg
and
http://www.ferdinand-engelbeen.be/klimaat/klim_img/acc_co2_1900_1959.jpg
Again, the emissions trend, be it more irregular, due to the resolution and accuracy of the ice core measurements and emission inventories, is far superior over the temperature-CO2 trend.
Thus there is no reason the doubt the value of the ice core CO2 measurements and all observations support that dT is a limited driver of CO2 levels, from fast (3 ppmv/°C) on short term (months) to slow (8 ppmv/°C) over millennia. The current increase is quite certainly caused by the emissions and not caused by temperature changes, although there is a (spurious) correlation with temperature, as both temperature (irregularly) and CO2 levels (very regularly) did go up in the past 50 years.
Regards,
Ferdinand