Temperature lags CO2 in ice cores

The notion that air T lags 800 years behind CO2 increases (from glacial ice) is prominent among ‘CO2 does not increase T’ memes. Y’all must have heard of this. Previously, I have posted here newer analyses suggesting that the lag is shorter or possibly even nothing:

Centennial-scale changes in the global carbon cycle during the last deglaciation. Nature 514, 7524 (2014)

Global warming preceded by increasing carbon dioxide concentrations during the last deglaciation. Nature 484, 49–54 (2012)

I had earlier missed reading GregP507’s reference to this in our ‘Global Air Temps’ thread:

Rise in temperatures and CO2 follow each other closely in climate change – University of Copenhagen

Which actually leads to:

Tightened constraints on the time-lag between Antarctic temperature and CO2 during the last deglaciation. Climates of the Past, 8, 1213-1221 (2012)

And for missing that, I must apologize. As Gandalf the wizard said, “Help often comes unlooked for”.

The ‘air T lags behind CO2’ meme may be losing its legs. Science does progress. Please, somebody, inform the affinity websites.

 

Way back in the days where ice cores were disputed as appropriate CO2 chronosequences , the argument went like this. First the snow falls, then new overlaying snow squishes it into 'firn', and finally the squish into ice is completed. The firn stage could be a short time or long. Trapped CO2 can move around (including vertically which is the time axis) in firn. Later (lower) the gas bubbles get locked in place.

The temperature signal is 18O isotopes in the frozen water, and it lacks this vertical mobility.

This was an early disputandum about ice-core CO2 chronologies. It does correctly lead to fuzziness in the CO2/time relationship. As far as I can tell, everyone knows that now except folks still clinging to 'air T lags CO2'. It is an interesting thing to have forgotten, given that it was a skeptics' meme at the outset.

I presume that the ice-core people are improving their thin-slicing technology. With that done, see which cores show the sharpest transitions in any stored gas (could be CH4 for example instead of CO2). Sharp transitions suggest less vertical migrations in the firn stage. This seems like the Nature 2014 study.

A different approach is to use sediment cores instead of ice. Then there is no firn, this was done in the Nature 2012 study.

Anyway this aspect of science appears to be moving forward and narrowing uncertainties. Just like it's supposed to do. I can help people try to keep up, but cannot guarantee success.

My impression is that ice sheets with the thinnest firns (least time spent in transition) are those with the most rapid new snowfall. Better to say most rapid during the times slices one is most interested in. It seems up to the core studiers to identify those, slice them thinly and do good analytical chemistry on the micro scale. Let us all hope therefore that they are not part of some global conspiracy. That they do not drink too much coffee.

Steady hands, lads, and follow the science wherever it leads.

 

The volcanic record is sulfuric acid, dissolved in water.It can also 'spread out' in firn. But as we know from science, liquids diffuse more slowly than gases. So it does not spread out as much as CO2.

Bob's quote refers to borehole temperatures. Can be made in ice, soil or rock. Thermal conductivity is quite low in such solids, so the T signal doesn't spread out much here either. If it did, boreholes would be much less useful. Sometimes they are not very useful, as when liquid water is flowing through your solid. Liquid water is really good at transferring heat.

Also you use a real fancy thermometer like a PTD, with resolution about 0.001 degree. Way better than an ARGO thermometer