Is the Rise in CO2 Really Not from Fossil Fuels? A Critique of the Quirk Paper

[Skyhunter]

http://icecap.us/images/uploads/TomQuirkSourcesandSinksofCO2_FINAL.pdf" is a paper published in Energy and Environment that claims the rise in CO2 is not from the combustion of fossil fuels.

I am in the process of reading it now and thought I would share it here.

The first thing that jumps right out at me is that they claim on the one hand that all the CO2 emitted in the NH is absorbed before it can travel to the SH. As evidence in supprt of this assertion they use the time lag for C¹⁴ dispersal from nuclear weapons testing.

If the CO2 is absorbed, how can it be dispersed? This seems a contradiction to me.

 

[Xnn]

The provided link is to Iceages, a denialist website.
Was this paper published in a peer reviewed science journal?
It impresses me as a violation of forum rules.

Anyhow, just glanced thru the paper. Believe it is fairly well established that ENSO events have an impact on atmospheric CO2 concentrations. However, claims that the rise in atmospheric CO2 concentrations are not from combustion of fossil fuels is simply preposterous.

Nuclear weapons testing were essentially point sources. Fossil fuel combustion is widely dispersed.

Don’t really wish to waste time reading thru junk science to figure out all of their errors and deceptions.

 

[Mike Davis]

Interesting paper from Energy and Enviorment which I believe is a peer reviewed journal. This is what the author says are the results of the study:
"The results suggest that El Nino and the Southern Oscillation events produce major changes in the carbon isotope ratio in the atmosphere. This does not favour the continuous increase of CO2 from the use of fossil fuels as the source of isotope ratio changes. The constancy of seasonal variations in CO2 and the lack of time delays between the hemispheres suggest that fossil fuel derived CO2 is almost totally absorbed locally in the year it is emitted. This implies that natural variability of the climate is the prime cause of increasing CO2, not the emissions of CO2 from the use of fossil fuels."

 

[sylas]

Checking my own work for errors... I'll be back.

Cheers -- Sylas

PS. I'm an idiot. My first post was invalid; and I'll look more carefully.

 

[sylas]

After such a mess on the first try (now deleted) I felt obliged to give a more careful response.

http://icecap.us/images/uploads/TomQuirkSourcesandSinksofCO2_FINAL.pdf" is a paper published in Energy and Environment that claims the rise in CO2 is not from the combustion of fossil fuels.

I am in the process of reading it now and thought I would share it here.

The first thing that jumps right out at me is that they claim on the one hand that all the CO2 emitted in the NH is absorbed before it can travel to the SH. As evidence in supprt of this assertion they use the time lag for C¹⁴ dispersal from nuclear weapons testing.

If the CO2 is absorbed, how can it be dispersed? This seems a contradiction to me.

I'm not quite sure what you mean. There's a carbon cycle, with carbon passing between oceans and vegetation and soil and atmosphere all the time. It's complex, and the magnitude of the fluxes are large. As carbon disperses through this cycle, it is being transferred between the different "reservoirs" all the time. Any transfer is a case where carbon is being absorbed into one reservoir and emitted from another.

Human emissions count mainly as a flux from geological reservoirs into atmospheric reservoirs. This leads to continuously increasing levels throughout the whole cycle as emissions start to move through the system; because the natural long term flux back into geological reserves is much smaller than the anthropogenic emissions flux in reverse.

The total additional carbon being removed from geological reserves is ending up all through the carbon cycle, with a net flux at present out of geological reserves and into other more dynamic reserves. For every 10 kilograms of carbon removed from geological reserves, we end up with about 5 more in the atmosphere, 2 or 3 in the ocean, and the rest in other reservoirs.

In my first comment on this paper, I made a dumb mistake (mixed up molecular weights in a conversion); and attributed the error to the author; and then compounded the error by posting too quickly. This is that post, restored and fixed up.

Figure 1A starts out with a comparison of emissions and rates of growth of atmospheric CO2. The data for Mauna Loa is available here.

To put this diagram in context; the fluxes shown indicate carbon being placed into the atmosphere at a rate rising from 3Gt/year up to 8Gt/year at present. At the same time, the natural fluxes in the carbon cycle transfer over 200 Gt/year into and out of the atmosphere. The difference is that the natural cycles are cyclic, whereas the transfer from geological reserves is cumulative. The consequence of the enormous natural fluxes is that what we add to the atmosphere gets flushed through the rest of the system. The total carbon content in the atmosphere is about 750 Gtonnes.

Comparing the natural flux to the total reservoir size, we can do a back of the envelope guess, that 25% of atmospheric carbon is recycled and replaced in a given year. At the same time, we are adding something like 1%, per year of what's there at present, based on emissions. Based on the fact that the rate of cumulative increase is about 0.5% per year, there's 0.5% per year being flushed out. It's a reasonable first order approximation to think that if we stopped emissions tomorrow, flushing rate would remain about the same, which would imply CO2 levels should start to drop at about the same level that they are presenting rising if emissions stopped dead. This rate of fall would drop off, at a rate that depends on the effective capacity of other reservoirs.

This paper depends heavily on doing subtractions. To get the annual rates of increase, they subtract absolute measures. To attribute fluxes on one thing or another, they subtract the rates. Subtractions of this form have large uncertainties. You can get useful information from trends, but at a particular year, you are likely to be overwhelmed by local effects. For example; to get the rate of increase at Mauna Loa, you just subtract reads from one year to the next. The resulting series will show a definite trend that the rate of increase is accelerating (that's a difference of differences, or second derivative) but when you look at the series over a particular year or two, you are almost certainly just looking at the local variations around Hawaii.

Here's another example. Look at figure 3 in the paper. This shows quite a stready drop in the C-13 proportion of atmospheric carbon. Conventionally, this is considered strong evidence that there's an increasing amount of carbon from geological reserves: a signature of the flux from fossil fuel use. The author seeks to discredit this notion; but how? Figure 4 shows a difference calculation. Basically, instead of plotting δ13 directly, you get the average value around a point, and plot the difference with that average. Basically, Quirk is plotting at each point the difference between that and a smoothed line through the original data. Look at the vertical scale. Figure 3 covers a depth of 1.2 from top to bottom. Figure 4 covers a span of 0.35. There may be a signal there for some effects; but this process is certainly amplifying any noise!

Figure 8 goes on to compare two sets of differences; which indicates that when the δ13 fraction goes about it's smoothed out line, the net growth rate tends to go below the smoothed line.

Here's what I think. There's a lot of natural variation going on, and if you go out of your way to remove the long term trends – which means removing the accumulation of geological CO2 into the atmosphere (!) then you see mainly the variations in natural cycles. By and large, there's a flow from the Northern Hemisphere to the South. No surface there… most of the additional flux into the atmosphere is in the North. Compare figures 9 and 10. Te first thing you note is that there's a lot more variation in the north. That would follow if carbon was not as well mixed, so there was more of a difference between the atmosphere and other reserves. The other thing that's apparent is that the North has more negative δ13 values. In the South, the values wobble around -7.4 to -7.6 at the start, and end up wobbling around -8.0. But in the North. The variations are larger, but the start is around -7.4 to -7.8, and the end is around -7.95 to -8.2. Smoothed out, there's clearly a lower δ13 values in the North – the signature of geological reserves.

So it sure as heck looks like geological carbon showing up in the North especially, and then ending up South as well by which time it is mixed up better as well.

Quirk takes differences, which looses all that information entirely because the main trends is gone – that's what differences do. And by figure 8, he's showing that an increase in the rate at which CO2 accumulates seems to align pretty well with increases in the rate at which δ13 is dropping. But isn't this still just an obvious consequence of mixing? No matter what natural cycles drive the small variations, when they help mix up the reservoirs, they'll be contributing to the mixing in of δ13 depleted geological carbon into the southern reservoirs!

Conclusion. It's hard to follow this paper, and frankly I think it is because Quirk is working so hard at avoiding the clear implication of all the main trends in the carbon cycle. The data he is using shows plainly a growth of δ13 deleted carbon it the atmosphere, correlated with growth in carbon in total. It's showing up most of all in the North, where it is preferentially in the atmosphere by contrast with other reserves, and it is mixing into the South as well; plus it is better mixed down here.

All of this supports the blistering obvious. CO2 levels are increasing mainly by virtue of a large flux of carbon into the atmosphere from geological reserves, driven by the use of fossil fuels. It's not a value judgment or a political ploy. It's just a bleeding obvious aspect of the world we live in.

In his conclusions, Quick makes three claims, which seem to be either non-sequiturs or nor valid implications of his data.

1. First claim is about peaks in δ13 correlating with peaks in CO2 concentration. Actually, that's change in the RATES at which δ13 drops with the rates at which CO2 concentration is increasing, because of the nature of his use of differences. This is a different thing. The data, as far as I can see, follows from the obvious fact that δ13 deleted carbon is showing up mainly in the north hemisphere and being mixed into the carbon cycle from there.
2. Second claim is that yearly increases in CO2 concentrations have been two orders of magnitude greater than change to seasonal variation, and that this indicates CO2 is totally absorbed the year it is emitted. I've looked over this a couple of times, and his point still escapes me. There are large cyclic seasonal variations in CO2 concentration, mainly as carbon moves in and out of other reservoirs, mainly vegetation. The Mauno Loa data shows this variation as 5 to 6 ppm peak to peak. Quirk appears to have taken a difference again, and is looking at whether the magnitude of swings is altering. His conclusion: it isn't. (And I'd add, of course not. What were you expecting?) But at the same time, total CO2 levels are increasing. (Again, of course.) Now Quirk seems to think that a take up of CO2 into other reservoirs indicates an increase in the capacity of other reservoirs, and in the seasonal variation. (Paragraph spanning pages 10-11). That looks like nonsense. It's not the capacity of other reservoirs which in increases, but their actual carrying load. There's no reason at all to predict increased seasonal variation from this. It's just weird.
3. Finally, he speaks of a lack of time delay in his third point. Here, I think he is not looking at a net flow at all, but merely changes in mixing rates.

All told: I'm unimpressed. It will play well to the usual suspects; but it seems to be an extended exercise in denying the obvious, carefully getting rid of trends before doing any analysis, and then missing the simplest explanations for what is seen anyway.

This is too long again, but I don't think the paper is worth the time it would take to clean up these comments into something more brief.

Sigh – Sylas. "Je n'ai fait celle-ci plus longue que parce que je n'ai pas eu le loisir de la faire plus courte." (Pascal, 1656)

 

[Monique]

We don't discuss papers in detail that are not published in peer-reviewed journals.