3.9 Global CO₂ levels and Antarctic temperatures

Figure 18 shows that over the last nine glacial cycles, the global CO₂ and Antarctic temperature appear to be positively and very closely correlated, showing the same patterns of change.

Maximise

Figure 18 Past Antarctic temperature changes (top) and global atmospheric CO₂ concentrations (bottom) going back through nine ice ages, derived from the EPICA ice core (Luthi et al., 2008)

Show description|Hide description

Two graphs of data from the EPICA ice core. The top panel is identical to the figure described in Figure 13, that is the temperature change inferred from the ice core over the last 800 000 years BP. The bottom panel shows how the atmospheric carbon dioxide trapped in the ice core varies over the same time period as the top panel. The two plots appear correlated, with warm temperature changes occurring at the same time as high atmospheric CO₂ concentrations of typically around 280 ppm, and the cold periods occurring at the same time as low atmospheric CO₂ concentrations of typically about 180 ppm. In a similar way to that described for Figure 13, the amplitude of the variation in atmospheric CO₂ concentrations reduces in periods earlier than 450 000 years BP. The warm periods then have a maximum atmospheric CO₂ concentration of typically 20 ppm less at around 260 ppm whilst the minimum is similar at about 180 ppm.

Figure 18 Past Antarctic temperature changes (top) and global atmospheric CO₂ concentrations (bottom) going back through nine ice ages, ...

You now have data that you could possibly use to predict what might happen as a result of the increasing CO₂ concentration that Keeling detected. You could theoretically plot a graph of temperature against CO₂ concentration to highlight the correlation and, from this, read off the temperature for any given CO₂ concentration. Unfortunately, there is a problem with this. The current atmospheric CO₂ concentration is higher than at any time in the previous 800 000 years, so even if you had a graph of the mathematical relationship between temperature and CO₂ concentration from the earlier data, it would not include the current (and much less, any possible future increased) CO₂ concentration. You would have to extrapolate (that is, extend) the graph beyond the available set of values, and you do not know enough from this data alone to be sure that the relationship will hold outside these limits. This means that it is difficult to use information from these earlier periods to predict what may happen in the future.

It is possible to be fairly sure that Milankovitch cycles amplified by greenhouse gases are at least partly responsible for the coming and going of ice ages; it is the best theory and one to which almost all climate scientists subscribe. But as you have seen, it is not a complete explanation, and some of the earlier cycles do not conform to this theory. To make useful predictions for the near future, and hence to suggest actions to protect the environment, you need to look for some more detailed information and more accurate scientific models.