Global Warming, page-242

The ice core data is still high resolution data, as the gas is trapped sequentially down the hole with time, with upper records being annual and becoming coarser with time.The ice core data gives us by far the best high resolution data over the last several hundred thousand years.

The gas-age/ice-age differential is what you are talking about, where the air is not trapped until the ice is compressed to a certain degree. The pore spaces are thought to be 'connected via diffusion' to the atmosphere until encapsulated (you would think that has to be a bit rubbery) . Different areas with different snow accumulation rates will have different gas-age/ice-age variations and different closeoff depths. They determine these chronological variations by correlating the cores with other ice cores, using oxygen isotopes, methane, nitrogen, palaeo-temperature records, volcanic events, changes in orbital behaviour recorded in the ice...

For cores in areas with higher snow accumulation the gas-age/ice-age differential is lower, in areas with low accumulation, such as East Antarctica, the gas-age/ice-age is greater and uncertainties are generally greater. The Vostock core, from East Antarctica, which is commonly used to highlight these issues has a gas-age/ice-age variation of several thousand years and an uncertainty of around 1000 years and more for some parts of the core. The Greenland cores have a much lower gas-age/ice-age.

http://lgge.osug.fr/IMG/fparrenin/articles/bender-JGR2006.pdf

Anyway, this is really superfluous to the Eemian issue of cause of sea level rise and associated CO2 levels.