20 million years of CO2 and ice sheet/sea level correlation

Posted on November 2, 2009


iceageWhen you look at the ice core record, there’s a significant amount of correlation between sea level rise and the amount of carbon dioxide (CO2) in the air at the time. But the ice core record goes back less than a million years. A study published a couple of weeks ago in the journal Science measured proxy data for CO2 concentration in the ocean and compared that data to other data on the stability of ice sheets. The authors discovered that there is strong correlation between the two going back at least 20 million years.

One of the challenges that the authors had was the fact that few available previous studies didn’t show correlation between the amount of CO2 in the air and the global climate prior to the start of ice core data. The authors hypothesized that this was a problem with the other datasets and developed a set of tests to check their hypothesis.

First they found two sites in the Pacific where they concluded – based on prior published studies – that the effects on marine sediments would be relatively unchanged over the last 20 million years due to specific geologic and oceanographic factors (limited upwelling, geologic stability, low biological productivity, et al). And they measured three different proxies from marine fossils that enabled them to estimate pH, sea surface temperature, and the amount of CO2 in the water.

Then they compared their results to the ice core data in order to estimate the accuracy of their measurements. What they found was that their reconstruction of the amount of CO2 in the air independently reproduced the ice core measurements to within the known error in the ice core measurements themselves. The importance of this fact was mentioned specifically in the paper:

[F]ew pCO2 proxies have replicated the ice core data of the past 0.8 Ma. (NOTE: “pCO2” is defined as the partial pressure of CO2 and is thus a measurement of the amount of CO2 in the atmosphere. “Ma” is a shorthand unit for “millions of years ago.”)

icecoresectionAs a result of this new reconstruction, the authors claim that “[r]esults for the Miocene and Late Pliocene support a close coupling between pCO2 and climate.”

In addition, the paper finds that a climatic optimum from 14-16 million years ago have the highest estimated CO2 concentrations in the paper’s data, and that during the optimum is the only period in the entire 20 million year dataset that has higher CO2 concentrations than the present.

The authors don’t claim to have answered everything, and like all good scientists, they point out that they haven’t proven causation, only shown very high correlation. Attribution studies to determine whether CO2 was a cause, an effect, or both will require more research.

Even so, the paper has a number of important conclusions. First, the data supports “the hypothesis that greenhouse gas forcing was an important modulator of climate over [the past 20 million years] via direct and indirect effects.” Second, the new reconstruction has sufficient resolution to define rough thresholds of CO2 concentration in the atmosphere for different degrees of ice sheet size and stability, and thus sea level. Specifically, the last time that there was this much CO2 in the air, there was little to no sea ice in the Arctic, Greenland had little to no ice, there was essentially no ice on West Antarctica, and even East Antarctica was mostly ice-free. And finally, the reconstruction may indicate that the global climate is highly sensitive to the amount of CO2 in the atmosphere.

At the climate optimum described in the study, “global surface temperatures were on average 3 to 6°C warmer than present.” If this study’s results are corroborated, then this paleoclimate reconstruction will be yet another study supporting the widespread understanding that climate is very sensitive to CO2 concentrations. In addition, the study will stand out as another example of “climate disruption is worse than we figured” as it points to the near complete melting of both Greenland and both sides of Antarctica. That would raise sea level by nearly 70 meters (~230 feet).

Other studies have shown that it takes hundreds to thousands of years for that much ice to melt, but if it starts this century, there may not be much humanity can do about it but move inland.

Thanks to lead author Dr. Tripati for a review copy of her paper. For the supplemental online information, click here.

Image Credits:
W Berner/University of Bern, via NewScientist.com

Posted in: Uncategorized