I hear a lot of interesting things when I play the “fly-on-the-wall scientist.” Most statements uttered casually between friends can be tested scientifically; at the very least, research has already been done and one only needs to dig a little to find out whether the statement is true. There are many things in life that can be demonstrated true and false. Grab a bowl; let’s pop a serving of buttery science!
Carbon dioxide (CO2) levels lag behind changes in global temperature. Therefore, CO2 is not a greenhouse gas and/or other causes are at work in global climate change and therefore the current changes are not human-induced (“anthropogenic”).
After concluding a review and Q&A session for my introductory physics course, I had to remain late at SMU until Jodi returned from a formal dinner event. I was killing time (trying not to think about the norovirus that was playing havoc with my intestines that same day) when a conversation between two colleagues at the other end of the room caught my attention. One of my colleagues was arguing, apparently as he usually does, that anthropogenic climate change has little or no scientific basis. To support his claim, he repeated a statement I have heard before from anthropogenic climate-change deniers (ACCDs): CO2 levels are a lagging, not a leading indicator (that is, the rise in CO2 levels TRAILS the change in temperature), and therefore CO2 is either not a greenhouse gas or global climate change is induced by factors other than CO2, so humans cannot be responsible.
I decided that this was a real opportunity to me to investigate the background of this statement. So here we go!
A look at the chart
Here is a great example of a plot that gets referred to by many ACCDs (reproduced from ).
Above, we see a lot of data being shown at once. Concentrate on a, b, and c. Curves a and b are the levels of CO2 and CH4 (methane), both greenhouse gases (that is, these are gases that can trap heat). We see the levels rising and falling over long periods of time, but if we compare to curve b, the isotopic atmospheric temperature, we see that changes in temperature for warming lead the changes in CO2 and CH4. For a slightly marked up version of this plot which helps see where warming trends fall and how insolation changes match to those, see the appendix below.
Why can warming ever lead greenhouse gases?
Before we proceed, it’s definitely worth understanding how the researchers came to these results. They obtained and studied ice cores from Antarctica. Ice traps bubbles of atmospheric gas, effectively encasing a snap-shot of atmospheric conditions over vast periods of time. Studying the trapped gases taught the researchers about atmospheric CO2 and CH4 levels. How do they determine temperature changes? They compare the level of Deuterium present at each level of the ice core. Heavy water isotopes are always present in water; however, water vapor will tend to contain lower levels of heavy isotopes and high levels of light ones; precipitation will contain more of the heavy isotopes and lower levels of the light ones. Changes in deuterium levels therefore are a proxy for temperature, and changes relative to modern temperatures (and the Deuterium levels at present) can be used to chart temperature back in time.
Finally, what is “insolation” shown in curve e in the above plot? According to Wikipedia :
Insolation is a measure of solar radiation energy received on a given surface area in a given time. It is commonly expressed as average irradiance in watts per square meter (W/m2) or kilowatt-hours per square meter per day (kW·h/(m2·day)) (or hours/day).
As the Earth’s relationship to the sun changes due to the eccentricity of the orbit, the tilt of the rotation axis, and the precession of the earth’s orbit, so does the amount of radiation received per unit area. There is a fairly clear understanding of how this change influences Earth’s climate in regular cycles, called “Milankovitch cycles.”  The above graph shows that insolation, specifically increases in Joules striking the Earth, can initiate warming trends.
Increases in solar radiation can initiate warming trends by causing ice-age glaciers to recede. Melting/receding glaciers reduce Earth albedo – the property of reflectivity that, rather than trapping solar radiation simply reflects it back into space. Reduced albedo means more heating. As the heating starts to increase, CO2 and CH4 are released from storage on or in the earth (in sea water, for instance, where under cooler temperatures the gases remain trapped in the water). We see then how CO2 and CH4 levels can begin to rise after warming has begun. This then creates a feedback loop; higher levels of heat-trapping gases trap more heat, and heating begins to accelerate, causing more glacier to melt. This defines the sharp warming we see in the graph above. Subsequent re-cooling of the earth can take tens of thousands of years, after only a couple of thousand years of this warming. Then the whole thing repeats again. This, according to Antarctic ice samples, has been repeating every 100,000 years or so for the last 400,000 years.
It is important to note that the lead that temperature has on CO2 levels, historically, is a small one. Current estimates range between 200-1000 years of leading. However, the warming trend lasts thousands of years – so it does not end when CO2 and CH4 enter the atmosphere in increased amounts, it continues strongly thereafter.
What’s happening now?
So we see how insolation (the change in earth’s reception of energy from the sun) can lead to warming (decreasing glaciation), which then releases greenhouse gases after the warming is initiated and begins a few thousand year runaway period of warming.
Is that what is happening in the current period? Are ACCDs right?
The answer, quite simply, is no. Here is why.
First, look at the above graph. Our current era is at the left of the graph. We see that solar insolation is at a minimum right now – that is, we are in a period where energy from the sun is at a low-point. Previously, warming trends were always initiated on the increasing side of insolation, not on the down-side or at a minimum. So already we see a distinction between the last 400,000 years and the current warming trend.
Second, atmospheric CO2 levels have not lagged the current warming – they have led it. This is very clear in all of the data that has been analyzed looking at CO2 levels and global temperature over the past 200 years, since anthropogenic expulsion of CO2 into the atmosphere began to ramp during the American and European industrial revolutions. The story is told in the CO2 data shown in the insets of the figure below.
The above figure is taken from Ref. . We see that rapid increases in atmospheric CO2 levels (the top plot) began in the 1800s and have continued unabated since. The real up-tick in levels began in 1900, when the slope of the change increased. The next increase in slope was then in about 1955-1960. Let’s then compare that to when temperature changes began over the same period. The figure below, also from , shows these relative to the average temperature between 1850-2006:
Whether we look at global average temperature or just the average in the Northern or Southern Hemispheres, we see the same trend. Temperature seemed to be going in regular cycles around a flat average from the start of the graph (1850) until about 1920. Then in 1920, the average around which regular cycles occurred obtained a new slope – no longer zero (flat), but positive (increasing). From about 1920 to 1980, a period of 60 years, the global average temperature increased by 0.2 degrees Celcius. From 1980-2006, a period of 26 years, global average temperature increased by an additional 0.2 degrees Celcius. The time to double the temperature change cut in half between 1920-1980 and 1980-2006. That’s stunning.
But CO2 levels began increasing long before these temperature changes took off. So in the current era of warming, CO2 is driving the change and not the other way around. CO2 now leads temperature.
Is that CO2 really from human activity?
A final statement that ACCDs might make about all of this is the following: so what . . . there’s no proof that this CO2 comes from human activity, so maybe this is just a natural cycle where CO2 happens to lead temperature for a change.
Actually, there is a clear physics answer to this question: isotope ratios. I have commented on this before  based on studies like those in reference . CO2 that remains in the atmosphere for a very long time, or which is present in biological material on the surface of the earth, is constantly exposed to cosmic ray radiation. This radiation can generate the isotope of stable carbon-12, unstable carbon-14, in a ratio that is actually easily measurable to high precision. In other words, CO2 that derives from close to the surface of the earth has a larger C-14 content ratio than carbon that is more shielded from cosmic ray radiation.
Carbon that has been underground for a long time, in the form of fossil fuels, is just such a kind of carbon that will have a lower C-14 level. By studying the level of C-14 in the air trapped in tree rings and other sources, we can watch to see if over time the atmosphere loses C-14 as more C-12-rich fossil-fuel carbon is pumped into the air. Sure enough, this is exactly what is seen. C-14 is a fingerprint, and we observe that since the 1800s the level of C-14 has decreased dramatically in the atmosphere. This tells us that less radiogenic carbon is entering the atmosphere – just the kind of carbon obtained from deep sequestered fossil fuel. The only species burning lots of deep-sequestered carbon is humans, and thus the modern CO2 levels are anthropogenic in origin.
So, does CO2 lag temperature? Yes, it can, and historically it has. The ice core record is clear on that. However, in the modern warming cycle CO2 and other gases have led the warming trend. In addition, that CO2 is rich in C-12 but depleted in C-14 because it comes from reserves deep in the earth. Anthropogenic warming is the status quo, and acting otherwise is wishing your doctor hadn’t diagnosed you with cancer and shopping around for a doctor that will say otherwise.
4/17/2011: Based on a reader comment, I produced an annotated version of the Vostok Ice Core data plot. The blue lines help you to see how significant warming trends match to locations on the insolation curve. I made the plot “blind”: I zoomed way in on the temperature curve, marked the start of significant warming trends with blue lines, and when I drew all lines I zoomed out and extended the lines to the axes. That way I wouldn’t only choose warming trends that definitely correlated with high radiation from the sun. The blue lines show that this appears to be the case, but any bias I possess should not be a factor in that relationship.
 J. R. Petit, J. Jouzel, D. Raynaud, N. I. Barkov, J.-M. Barnola, I. Basile, M. Bender, J. Chappellaz, M. Davis, G. Delaygue, M. Delmotte, V. M. Kotlyakov, M. Legrand, V, Y Lipenkov, C. Lorius, L. Pepin, C. Ritz, E, Saltzman, and M. Stievenard, “Climate and history of the past 420,000 years from the Vostok ice core, Antarctica,” Nature 399 (1999) 429-436
 “Contribution of Working Group I to the Fourth Assessment Report of the Intergovernmental Panel on Climate Change, 2007,” Solomon, S., D. Qin, M. Manning, Z. Chen, M. Marquis, K.B. Averyt, M. Tignor and H.L. Miller (eds.) http://www.ipcc.ch/publications_and_data/ar4/wg1/en/contents.html
 “Seen and Unseen” http://steve.cooleysekula.net/blog/2010/05/25/seen-and-unseen/
 http://adsabs.harvard.edu/abs/1981E&PSL..53..349S [“Atmospheric C changes resulting from fossil fuel CO2 release and cosmic ray flux variability.” Stuiver, M. and Quay, P. D. Earth and Planetary Science Letters, vol. 53, no. 3, May 1981, p. 349-362. ]