EPICA Ice Core Data and temperature/CO2 relationships

March 1, 2010 by jason

One of the central battlefields of the global warming debate today, is the history of CO2 concentrations and their relationships to temperature. In my previous article on the climate change debate I illustrated that there are strong theoretical as well as empirical (ie. historical) indications that CO2 is in fact one of the drivers of past temperature cycles. In this article I want to draw you a timeline, and show you how these relationships have acted in the past.

The European Project for Ice Coring in Antarctica (EPICA)

The European Project for Ice Coring in Antarctica (EPICA) is a multinational European project for deep ice core drilling in Antarctica, with the aim of obtaining detailed information of past climate conditions, specifically atmospheric composition and temperature. For more detailed information on the project, and how the data was obtained, refer to my article on the EPICA Project.

Why it might seem like CO2 is just a bystander

Central to the typical argument of someone who denies global warming, is usually the idea that CO2 has lagged temperature in the past. This is true to some degree, but as I explained in my global warming article, there are other factors at work, and (not so) recent science suggests that although CO2 was not the initiating force behind some of history's temperature spikes, it was one of the main reasons the spikes (called interglacials) were so long and intense. This is well established science and echoed by many respected and recent studies (Siegenthaler et. al (2005), Hönisch et. al 2009, and Mudelsee (2001) to name a few).

To deny this theory, some argue that there is no evidence that CO2 was the driver for any of the major temperature increases in the past, or at least not for any of those in the past million years. To support this argument, we often see this graph:

This is a commonly cited graph, representing a small extract of data obtained directly from the EPICA Dome C antarctic survey (source: myff.org)

The arrows indicate clearly that, based on the data shown in this graph, concentration maxima occured consistently after temperature maxima. Global warming deniers will often state that because CO2 spikes and gradient changes always followed temperature changes, it could not have been the source of feedback (ie. it could not even have helped to increase the temperature).

The truth is basically that there is no proof that CO2 has a significant impact on temperature, there is just an overwhelming and growing collection of good science that points at CO2 as being the single biggest factor affecting our climate today. Reading this article won't make you a guru on all things CO2, but it will help you to understand what really happened in the last 400,000 years (at least in terms of temperature and CO2).

The Chronology of the early EPICA CO2 history

The period of cycling glaciation which we see in the EPICA Ice Core records are the later section of what is called the Pleistocene, a period ranging from 1.8 million years ago to about 10,000BC. A very detailed study on the EPICA Ice Core records, as well as an explanation for the events that took place, is provided by Siegenthaler et al. (2005), to which I will reference to in outlining the chronology of the EPICA Ice Core records. First, let's take a look at the EPICA Ice Core record data:

Source: Siegenthaler et al. (2005)

The above graph can basically be considered a relatively accurate representation of CO2 concentrations as well as temperature levels throughout the given time period, at least for the accuracies we will be needing for this discussion. The data was obtained by simply analyzing porous ice components drilled from the EPICA site to obtain the gas components. Based on the gas components the temperature and CO2 concentrations can be identified, with the depth indicating the age of the sample. Based on data by Petit et al. (1999) and Fischer et al. (1999), the lowest values for each glaciation cycle are 182 ± 4 ppmv, and the highest values during the deglaciation periods are 296 ± 7 ppmv. This is incredibly stable, given the complexity of the earth's climatic system, and is what lead scientists to suspect that some controlling feedback mechanisms are acting on the global temperature system to regulate the peaks and troughs (see Falkowski et al. (2000)).

Temperature and CO2 proxies for the periods 650,000 years BP to 400,000 years BP. Glacial terminations are given in roman numerals (V,VI and VII), and marine isotope stages (MIS) are given in arabic numerals as reference (Source: Siegenthaler et al. (2005))

The oldest atmospheric data obtained from EPICA Ice Core Records, at 650,000 years BP, also represents the period of lowest CO2 concentration levels (182ppmv at 644,000 years BP). At MIS 630,000 years BP, CO2 concentrations are at about 190 ppmv just prior to glacial termination VII. The increase in temperature and CO2 concentrations at this glacial termination occurred very quickly (within 3000 years).

In regards to the upward trend of CO2, it can be divided into 2 main regions. The first is the rapid increase in CO2, to about 235 ppmv within a period of less than 2000 years. The second period is more prolonged, increasing in CO2 by about 20 ppmv over approximately 5000 years. This second CO2 peak is very similar to the magnitude and time-scale of the Holocene deglaciation. This is important because it shows that the holocene warming is not unprecedented or unusual, as Ruddiman (2003), among others, suggested. What is more likely is that the increases in CO2 are a response of the carbon cycle to massive changes in global biomass levels (Joos (2004)). The deglaciation maximum for this first period is reached at 620,000 years BP, where CO2 levels reach their peak at 260 ppmv.

The deglaciation is then interrupted, and sinks to almost 200 ppmv, remaining very turbulent between 610,000 and 590,000 years BP, although the conditions are near glacial. This raises the question of whether the period between 620,000 and 560,000 years BP was a single interglacial or in fact several. The increases of CO2 concentrations leading out of this possible glacial period (leading up to 580,000 years BP) take about 5000 years each. One unexpected feature of this period is the long and stable warm period between 580,000 and 560,000 years BP, strongly contrasting other records suggesting increases in global ice coverage for this same period (see Lisiecki (2005)). It should be noted that this region of stability is not limited to CO2 and temperature data, but also CH4 and aerosol levels. This period thus represents the most stable carbon cycle of our known history, extending for a total of 28,000 years. This stable warm time period is therefore the subject of much research to better understand orbital pattern influence on global climate. It cannot however, be stated as yet what the causes of this stable period were.

The decline in temperature observed at the end of this long warm period begins at about 555,000 years BP, and is interrupted by 2 short and pronounced spikes in temperature. this turbulence is followed by a deglaciation event at 510,000 years BP, which has a magnitude among the lowest of all the deglaciation periods of the past 650,000 years. As with the termination VII, termination VI can be divided into 2 regions, with a small semi-glaciation period (this time at 490,000 years BP) breaking up the deglaciation phase. The minimum CO2 concentration occurs at 481,000 years, lagging the temperature minimum by about 10,000 years.

The glaciation following this event is again broken up by 2 temperature and CO2 level peaks. These peaks are comparable both in duration and magnitude to the CO2 spikes that can be observed during the past antarctic warming events in the last glacial, indicating again that recent events were not at all out of the ordinary (see Wagner (1999)). A more detailed analysis of this data by Siegenthaler et al. (2005) further showed that the CO2 concentrations at the latest interglacial observed here (at 400,000 years BP) were very similar to those of the last holocene.

The CO2 lag and its complexities

Many people will tell you that the CO2 lag is simply a response of CO2 concentrations to temperature, but even historically it isn't that simple. Aside from the fact that CO2 causes radiative forcing and thus logically contributes at least to some global warming, we can observe changes in the response of CO2 to temperature from the EPICA Ice Core Records, indicating that the relationship is more complex than a simple response to temperature.

For example, taking the glacial terminations V, VI and VII (those used in this article), yields lags of CO2 levels against temperature of 800, 1600 and 2800 years, respectively. It is interesting to note that there appears to be a decreasing trend here (ie. the lag is getting lower). Fischer et al. (1999) concluded that the average lag of CO2 to global temperature during the past 3 glacial terminations (I, II and III) were between 200 and 1000 years. An interesting observation is also the apparent lead of CO2 over temperature between 535,000 and 548,000 years BP (by 1500 to 2500 years).

Why it's not the milankovich cycle...

It might also be relevant to include here a graph of the affect of changes in irradiation levels caused by the milankovich cycle. I described the nature of the milankovich cycle in more detail in an earlier article on why CO2 lags temperature. The milankovich cycle represents changes in radiation forcing on the earth through variations in the earth's orbital pattern. I don't want to dwell on this too long, but just to show that the milankovich cycle is unlikely to be a significant driver for the magnitudes and time periods of glacials and interglacials that we are seeing in the EPICA Ice Core records, here is a map of the milankovich cycle patterns against global temperatures:

Milankovich cycle forcing patterns against global temperature levels from the Vostock Ice Core Records (Source: Wikimedia Commons)

It is plainly evident that the Milankovich cycle alone could not be the major driver behind these sudden and exponential increases in global temperature, and the best explanation for a controlled, cycled and exponential increase in global temperature is, at present, CO2. I have already illustrated why it is not reasonable to suggest that cosmic rays are the cause for global warming, so I will not go over this again.

What does it all mean?

If you take anything from this article, let it be this: We don't know everything. Scientists are doing their best to uncover the truth, but until then all we have is information, interpolation and speculation. What we do see is an incredibly strong bond between global temperature levels and CO2 concentrations, one that appears almost inseparable. There is no known historic example of a time when CO2 was known to lead temperature. This does not mean that CO2 does not influence temperature, in fact the ranges of temperature increases that we see in the terminations are simply too rapid and powerful to be explained by anything other than a feedback loop with CO2. The fact that CO2 levels were never higher than 300 ppmv during the past 650,000 years, and are now at an alarming 380 ppmv, should be of great concern.

It seems evident that increasing CO2 concentrations will result in increasing temperatures, but to what degree, and how soon, we simply don't know.