If Temperature Leads CO₂, Can CO₂ Affect Temperature?

(c) 2009 by Barton Paul Levenson

A common argument of global warming deniers goes like this:

Analysis of Greenland, Siberian and Antarctic ice cores shows that, although temperature and carbon dioxide correlate over long periods of time, temperature appears to lead CO₂ by an average of 800 years. Cause and effect can't reach back in time, so clearly temperature is the cause and CO₂ the effect, not vice versa. Therefore, we don't have to worry about carbon dioxide causing global warming.

Sounds plausible, doesn't it? And the temperature lead time in the ice cores is definitely there. Have the global warming deniers finally found a clinching argument?

Well, no.

Let's examine what causes ice ages. Climatologists agree that these are initiated by the so-called "Milankovic Cycles." These are periodic variations in the Earth's orbit and axial tilt that affect the distribution of sunlight over the Earth's surface:

• The Earth's orbital eccentricity, or variation from a circle, varies between 0.005 and 0.058 (the present value is about 0.017). Cycles involving eccentricity are 95,000; 136,000; and 413,000 years long, and combine to create an overall cycle about 100,000 years old.
• The tilt of the Earth's axis varies between about 22.1° and 24.5° (the present value is about 23.5°) with a mean period of about 41,000 years.
• The direction of the Earth's rotational pole precesses (wobbles) with a period of about 25,800 years.

These cycles alter the distribution of sunlight over the Earth's surface, regularly increasing and decreasing the relative amount that falls on the planet's ice-covered poles. Serbian astronomer Milutin Milankovic proposed in 1930 that these variations might account for the Quaternary ice ages that have governed Earth's climate for about the last 3,000,000 years. By 1970 or so his explanation was generally accepted.

But detailed calculation shows that these variations are not enough to account for the swings in temperature experienced in the ice ages, another variable measured from ice core data. The swing in mean global annual surface temperature (M-GAST) across the ice ages is about 5-6 K. Milankovic cycles only account for about 3 K.

Something amplifies the temperature swing. The most likely candidate is carbon dioxide, which is a greenhouse gas. More CO₂ in the air raises the surface temperature, all else being equal. For an explanation as to why, try here.

Let's examine what happens in a natural deglaciation. Milankovic cycles start to warm the poles, and the Earth's temperature slowly rises. Carbon dioxide exists in large amounts in solution in the world's oceans. But the warmer the water gets, the less CO₂ it can hold. CO₂ bubbles out of the oceans, and it in turn raises the temperature more. The temperature does not run away because the feedback between warming and CO₂ release is a converging series, not a diverging one.

In short, it's a feedback. Temperature raises CO₂, which raises temperature. Carbon dioxide is both a cause and an effect.

But is this always what happens when the Earth warms? Does it account for the present global warming? No and no.

There have been other episodes in Earth's geological history where carbon dioxide increased first, e.g. during episodes of excessive volcanism, and temperature followed. One example is the Paleocene-Eocene Thermal Maximum (PETM), which occurred about 55 million years ago and caused a mass extinction.

And another example is the present global warming. The new carbon dioxide in the air is NOT coming out of the oceans. We know this for three reasons:

1. The radioisotope signature of the new CO₂ shows that it is coming primarily from burning fossil fuels. Cosmic ray bombardment continuously turns some of the nitrogen into the air from nitrogen-14 to radioactive carbon-14, to the point where about 1 in 1 trillion carbon atoms in the Earth system is ¹⁴C as opposed to ¹²C. Ocean CO2 has a normal complement of ¹⁴C. But fossil fuel carbon does not. ¹⁴C has a half-life of about 5,730 years. Fossil fuels are 100-300 million years old; their carbon-14 has long ago decayed away to nitrogen-14. There is also a clue from the ratio of ¹³C to ¹²C. Plants preferentially take in ¹²C in photosynthesis, and fossil fuels came primarily from dead plants. The new CO₂ in the air is ¹³C deficient. Ocean carbon is not.
2. We're burning an immense amount of fossil fuels. All the carbon dioxide so generated has to go somewhere, since in non-nuclear reactions matter can neither be created nor destroyed. We know how much is going into sinks--about half. The rest stays in the atmosphere.
3. From gas exchange measurements, the ocean is presently a net SINK for carbon dioxide, not a net SOURCE. The world's oceans give off about 90 gigatons of carbon a year, but take in 92.

There is also a clue from the famous 800-year time lag. If this were a natural deglaciation, we would expect to see a lag centuries long between the temperature increase and the carbon dioxide increase. Instead, the two are highly correlated (r = 0.87) in the same year:

So, in short, the argument "temperature leads CO₂, so it's the cause and CO₂ is the effect" fails on several grounds. In brief:

• The relationship is a feedback. Temperature affects CO₂, but CO₂ also affects temperature, so the denialist argument is like saying "We know chickens come from eggs, so eggs can't come from chickens."
• Temperature doesn't always lead CO₂. Sometimes (as in the PETM), CO₂ leads temperature.
• Present CO₂ is coming from burning fossil fuels, not from the oceans as in a natural deglaciation.

And that's that. Carbon dioxide is the main cause of the present global warming, and it's a serious problem.