For years, my less-green friends have pointed to the long ice cores in Antarctica, where annual striations go back hundreds of thousands of years. The temperature can be inferred from the relative abundance of common oxygen (16O) and its rarer isotope with two additional neutrons, 18O. In most cases, temperature changes before carbon dioxide does, seemingly inverting the normal cause and effect for the current climate rage.
Not so, say Jeremy Shakun, from Harvard, and his eight colleagues from around the world. They gathered up 80 so-called “proxy” indicators of local temperature (including the Antarctic ice cores), and compared them to the amount of carbon dioxide trapped in the ice. It appears that—after a little non-CO2 temperature rise of speculative origin—carbon dioxide increases indeed preceded the massive deglaciation that began about 20,000 years ago.
Shakun et al. argue that the Antarctic ice cores are a special case, where through a very complicated (and hard to verify) chain of causation, temperature will in fact increase before carbon dioxide.
Enter the internet, one of the greatest things that has ever happened to science. Because of requirement for increased data transparency (pushed, in no small part, by the Climategate gang’s reluctance to release theirs), the numbers behind important papers are now available online.
Willis Eschenbach, over at www.wattsupwiththat.com (a blog anyone interested in the climate issue should bookmark) got into the raw data—before it was sliced and diced into the very compelling graphics by Shakun et al. Warning: the following image is scientifically graphic and you might want to keep your greener friends away from it.
Each black dot is a temperature data point from the 80 proxies. The data are “standardized”—a common technique, to render their metrics commensurate. The average of each proxy is zero and the scatter about that mean is scaled to a value of one. The red dots are the carbon dioxide concentration in bubbles trapped in Greenland’s ice, about which there is little serious dispute.
It’s fair to say that the central core of the proxies is pretty concurrent with the carbon dioxide increase. It’s also fair to say that there is an awful lot of scatter in the proxy data. Finally, it’s fair to ask what all of this means for the future.
The third author on this paper, Andreas Schmittner, published something related two months ago, where, using a somewhat more global sample of proxies, he concluded that the likelihood of a very large warming this century from carbon dioxide is exceedingly small.
Willis Eschenbach’s plot of all the proxy data in Shakun et al., with ice-core carbon dioxide superimposed on a similar scale.
Schmittner’s study bolstered the policy argument that technological changes that will occur in this century should be sufficient to prevent a disastrous warming without drastic and expensive policy diktats. (One example is the unforeseen discovery of centuries-worth of accessible natural gas, which produces power with substantially lower emissions of carbon dioxide than coal.)
I have cautioned my friends for years that, whether or not temperature changes predated carbon dioxide changes in Antarctica, there’s no serious dispute that it is a greenhouse gas that will cause some surface warming. Consequently, while it is a really diligent piece of work, I don’t see the earth-shaking implications of Shakun’s new paper. And I am even more skeptical than I was about the usefulness of proxy climate indicators.
With regard to carbon dioxide, it’s not the heat, it’s the sensitivity (of temperature), where multiple lines of evidence are now converging on a low value.
