Somehow a paper arguing that the increase in atmospheric CO2 is mostly natural has managed to pass peer-review. Gavin Schmidt’s already covered it in a Realcimate post. Gavin Cawley’s paper is, in a sense, a pre-emptive response to this new paper. I’ll make a few comments similar to what Gavin has already said in the Realclimate post and then make a somewhat broader point.
The summary of the paper says
- The average residence time of CO2 in the atmosphere is found to be 4 years.
The confusion here is between residence time and adjustment timescale. Given the various fluxes of CO2, an individual CO2 molecule will only stay in the atmosphere for a few years before being taken up by one of the natural sinks. However, this doesn’t mean that an enhancement of atmospheric CO2 will decay in only a few years, because there is both a flux of CO2 out of the atmosphere and into the atmosphere – the molecule leaving the atmosphere is replaced. The residence time might only be a few years, but the adjustment timescale of ~100 years, or longer. I discuss this in more detail in these two posts.
The paper then says two related things:
- The anthropogenic fraction of CO2 in the atmosphere is only 4.3%.
- Human emissions only contribute 15% to the CO2 increase over the Industrial Era.
The key point is that atmospheric CO2 has increased by about 40% since pre-industrial times and it is all anthropogenic. The short residence time of an atmospheric CO2 molecule, however, means that not all of the enhancement will be made of up of molecules that had an anthropogenic origin. This, however, does not mean that the enhancement is somehow not anthropogenic; without our emission there would be no enhancement in the first place.
I thought, however, that I would comment on something that appears to be often misunderstood. The paper says it is estimated that the removal of the additional emissions from the atmosphere will take a few hundred thousand years and implies that this is wrong (through determining a very short residence time). I discuss some of this in these two posts, but I’ll elaborate a bit more here.
There are quite a large number of timescales associated with drawing down atmospheric CO2, but – in a simple sense – when we emit CO2 into the atmosphere, it mixes between the various reservoirs (atmosphere, ocean, biosphere) until – on a timescale of centuries – it reaches a new equilibrium, which is then drawn down over a timescale of thousands of years via weathering (ultimately taking more than 100 thousand years to fully recover). Some seem to think that it should settle back to the intial concentration, but it can’t because we’ve essentially added new CO2 to the system. Eli has a nice animation in this post.
That it will take more than one hundred thousands years for atmospheric CO2 to return to pre-industrial values is partly based on past changes (such as the Paleocene-Eocene Thermal Maximum – PETM) and partly on the carbonate chemistry of seawater. If you work through the carbonate chemistry calculation, you can show that there is something that is now called the Revelle factor (which I discuss here). This is the ratio of the fractional change in atmospheric CO2 to the fractional change in dissolved inorganic carbon in the ocean, and it is about 10.
This tells us that if we add CO2 to the system, once it’s distributed through the ocean/atmosphere system, the fractional change in atmopheric CO2 will be 10 times greater than the fractional change in dissolved inorganic carbon in the oceans. I discuss some of this in this post. Also, if you consider the amount of carbon in the ocean and atmosphere, you can show that between 15% and 30% of our emissions will remain in the atmosphere once ocean invasion is complete. At this stage, atmospheric CO2 is further drawn down through weathering, which is very slow and, hence, it will take more than 100 thousand years to ultimately return to pre-industrial levels.
The point I’m getting at is that the long timescale over which atmospheric CO2 will slowly return to pre-industrial levels is a consequence of the carbonate chemistry of seawater and weathering; you can’t assess this by simply considering the short-timescales fluxes into, and out of, the various reservoirs. So, not only does this new paper confuse residence time and adjustment timescale (amongst various other confusions) it also infers things about the long timescale over which atmospheric CO2 will recover using an analysis that is completely inappropriate. If you want to read a paper this does this analysis properly, you should read the atmospheric lifetime of fossil fuel carbon dioxide, by Archer et al. (2009).
Of course, some might argue that this post wasn’t really necessary, as any paper suggesting that the rise in atmospheric CO2 is not anthropogenic is obviously nonsense, but sometimes it’s worth delving into this in more detail, although maybe this is more for my own benefit than for the benefit of others. It is my blog, though 🙂