How do we know what the average temperature on Earth was thousands of years ago?
In: Earth Science
Sadly, neither prehistoric humans nor animals even further back were kind enough to leave either written records or stopped examples of thermometers.
However, scientists can make good guesses about past global temperatures in a few ways, mostly by measuring things that correlate with global temperature.
For example, glaciers and ice sheets can trap small pockets of air in them, and form in layers that can be counted or otherwise fairly reliably dated. These air pockets can be removed and analyzed, for example showing the carbon dioxide levels in the atmosphere at that time (carbon dioxide levels are more-or-less the same globally).
The composition of the snow/ice itself actually varies based on global temperature! Oxygen comes in 2 major isotopes, O-16 and O-18. What an “isotope” is is beyond the ELI5 of this explanation, but the upshot is that more O-18 in the ice means it was colder, in a very reliable manner.
Many things like trees or corals grow more when it’s warm than when it’s cold, so you can take core samples and examine the width of the bands to check the local temperature at a certain time. While this seems like it could only go back for the life of a single tree, scientists can actually match up patterns of bands between trees (finding bands at the end of one tree’s life that match the beginning of another’s) and make a chain of trees going back a long time.
The sediment on the bottom of lakes also forms reliable discrete layers, and the composition varies with the composition of the water and atmosphere at the time. The chemical composition of preserved mollusk shells also reflect the temperature at the time of their creation, and if they can be dated, that provides more info.
There’s some looser ones, too, like “XYZ plant was found here, and we know its seeds only sprout above 23 degrees, so it must have been at least that hot here”, but those are less precise, because things like local sunlight or wind can affect them.
I’m sure that there are other “climate proxies” that I’m missing, but this should give a good feel for it.
Essentially, by studying a ton of sources – tree rings, ice cores, rock structure, sediments, fossils etc. and comparing them to what we know about growth in different temperatures.
By plotting all these variables on a graph we can get a pretty good idea, albeit they need to be adjusted for other factors such as the atmosphere at the time.
E.g. if it is colder you would expect more rapid ice growth. Slower tree growth. Tropical animals not found in temperate waters.
It may sound inexact but most of the time we are only narrowing temperatures to a century or so. It isn’t possible to predict individual years, but to see long term trends. This is not least because dating techniques become increasingly less specific the further you go in the past.
Latest Answers