Let's start with the age of the ice/bubbles/CO2. If you cut a tree down, it is very straightforward that the number of tree rings is directly related to the age of the tree. 35 rings means the tree is 35 years old. Trees are fairly simple as they grow during the spring/summer and go dormant during the winter, which results in contrasting layers and easy to see annual rings. Yeah, some trees don't make rings very well and trees typically don't live very long.

If you take a core into an ice cap (Greenland, Antartica), you also find "rings" in the form of layers, but these layers represent 1 year of snowfall. Fresh snow starts of soft, fluffy, and mostly air. As that layer of snow gets buried by more seasons worth of snow, pressure from the snow above compacts the snow (what was once maybe a couple feet of snow is now maybe 6 inches of snow). It's more of a slushie texture. More snow builds up above, more pressure, more compaction, and you end up with a layer of ice maybe a half inch thick or thereabouts. But it's not just pure ice, air bubbles are throughout the ice and preserve a sample of the atmosphere when that layer of snow was first deposited.



Take a core in 2019 for example and start counting layers down and you get the age of each layer, similar to counting the rings in a tree. Valid questions would be: How do you know each layer is one year? What if there are a bunch of layers missing? Confidence is added when you find anomalies in the ice core, such as a thin layer of volcanic ash. If you're in Greenland, then that ash layer might be due to a volcanic eruption in nearby Iceland. Maybe the eruption was in 1950 (just making up a random date), and the ash layer was 69 layers/years down. Maybe you find another ash layer that's a couple hundred layers down that's really close to a historical eruption, but they are off by 3-4 years. Do you throw out the whole idea because somewhere along the line you counted what should have been two separate layers as just 1? The age errors on these ice core records are usually accurate to about 1 or 2%. Ash layer shown below.



How do we know that something funky didn't happen on a particular year? Valid question. Scientists don't put all their eggs in one basket. We've collected many many ice cores from Greenland and Antartica and get similar records of CO2 and other gasses trapped in the ice. Are there subtle differences? Sure. Does not being an exact perfect match negate their records? No. Also, none of these analyzes are of a single air bubble.

For what it is worth, this is research that has been going on for decades, not some emerging science that we're still trying to figure out. If you want the research papers I can get them to you somehow. Below are a couple citations. For testing the reliability of the measurements, one would simply need to measure the gas trapped in a layer of snow from a particular year, let's say 1970, and then compare it to the CO2 concentration measured directly from the atmosphere from the same year. Then do the same for other years. The atmosphere is mixed well enough that the CO2 concentration in Alabama is about the same as it is in Hawaii, and Greenland, and Antartica, etc. for any given time.

U. Siegenthaler & H. Oeschger (1987) Biospheric CO2 emissions during the past 200 years reconstructed by deconvolution of ice core data, Tellus B: Chemical and Physical Meteorology, 39:1-2, 140-154, DOI: 10.3402/tellusb.v39i1-2.15331

1983 "Comparison of CO2 measurements by two laboratories on air from bubbles in polar ice" Nature

Here is a short video on the topic. Do we honestly think that people would go to some of the most miserable places on earth to collect these samples just to make up false or meaningless data?