[Kay Ludlow]

I wonder how closely they can narrow down a time period when they are looking back across that many years. Really, the C02 levels that they talking about today as so high represent a few years - 10, 20 or so. Averaged across the last century they likely are still really normal. Can they find a section that correlates to a particular century? I wouldn't think so - I think they'd be lucky to be able to identify particular 1000 year time spans. That means that statistically it's meaningless to compare this years C02 to anything they find in the cores.

That would be like taking the last few letters of a book and comparing them to the entire book - "THE END has a far greater representation the the capital E than the rest of the book, demonstrating a trend toward increasing capitalized vowels". Comparing this year to nearly a million years worth of ice core is just as meaningless.

[blam]

Frequent identifiable volcano ash layers in the ice core help some with the dating.

[Doctor Stochastic]

Ice cores are layered. You can look it up. There are various methods. Besides counting visually (and comparing to annulary tree growth rings), there is a difference in pH between summer and winter depositions so that electrical conductivity can be used. This is good to about 1 or 2 mm. Also one can analyze the H vs D and O16 vs O18 ratios to get temperatures. Various metal residues (Be for example) can be used to date certain parts against known dates. Cores can also be checked with coral cross-sections as well as cores from other parts of the world (Greenland, Siberia, Antarctica.)

There are articles in the usual journals.