How do scientists determine how much CO2 was in the atmosphere thousands of years ago, and to what level of certainty are they able to perform these calculations?
In: Earth Science
It depends on the age of the time frame under discussion. Ice cores preserve bubbles of air that existed at the time of capture, so are good witnesses of air composition, but ice has a limited age (rare to get ice more than a couple hundred thousand years old although there are a few locations where older ice has been recovered). Precision is relatively high as far as such things go. The measurements can be pretty precise, but is the air bubble actually the same now as it was at the time the bubble formed? uncertainty enters from that.
Source material itself (the air) is not homogeneous so you need a good number of analyses to define a statistical population or you cannot really say anything about how well a single measurement is a representation of the real world (works for all analyses not just bubbles). The more data you have, the more confidence you can have.
Other methods vary from stable isotopes (carbon, oxygen, hydrogen, sulfur isotopic variations all provide clues about the nature of ocean-atmosphere chemistry), global carbonate sediment production, weathering rates of rock (CO2 is a weak acid and main source of rock weathering), and lots of other chemical or physical methods of questionable certainty but still decently useful for more ballpark understanding. Obviously, when you get into larger scale processes covering wider time frames, precision is lost. I even knew one line of thought that relied on examination of paleosoils going back well into the Proterozoic. How much should we accept the conclusions? Hard to say. Generally speaking, the changes with/over time are clear and repeatable, but how they translate into actual numbers is a harder thing to declare. High CO2 periods over low CO2 periods can be identified. Is a “high” period at 1000 ppm, 2000 ppm? or what? That is part of the interpretation and it is open to discussion, if one is honest. Trends are fairly easy to identify, but putting hard numbers to the different points in trends is a lot more challenging.
The BIG HEADS like Robert Berner and Dick Holland (both long departed from this world) were the ones who got into this sort of thing in a big way back when I was a student in the 80s. Don’t keep up so don’t know who is doing all the hot work now. Mandatory reading back then though.
Latest Answers