Where Grey Matter meets Dark Matter
Episode 21 - 9 December 2009
Gigolos get lonely too...
How old is stuff? Well you could check the 'Use by' date and add a few weeks or maybe a year on top of that. But what about if stuff is really old, before we had laws requiring the labeling of said stuff in such a fashion? And what about if you want to know other things, like what happened back in 'the day'? Well then, there's a few ways.
Counting tree rings (a process called 'dendochronology') is one of the dating methods that pretty much everybody learns about as a kid. A tree puts a on a new ring each year, so to figure out how old your favourite tree is (the tree that was your only friend in your darkest hours), simply saw it off near the base and then count the number of rings. Of course, you can then warm yourself by the glowing remains of your erstwhile BFFT.
Source: California Academy of Sciences
Ice cores are a 'cool' method. It amounts to drilling down into the ice and pulling out a cylinder of ice. If you know roughly the rate that new ice is deposited on the surface of a glacier or ice sheet, then you can drill down a certain depth (which corresponds to a certain time in the past), and examine the ice to see what sort of filth was in the atmosphere and find out about the climate at the time.
Fossils are another important layer in the strata of dating methods. Say you know how old a fossil is. If you find that fossil in some rock, then that rock has to be the same age (with an error bar). Conversely, if you know how old a layer of rock is, and you find a wicked new fossil in it, then it must be about the same age as the rock (since creatures don't tend to find themselves teleported into the middle of solid rock formations - unless Scotty got drunk again).
Where did we dig up this old fossil?
But even if you don't know anything about a particular fossil or the rock it's in, then you can still compare it to another unknown rock-fossil combination. If you know something about evolution, and the two fossils are closely related, then by knowing which is the more recent species, you can say which rock is older. And, of course, vice versa, if you know that one rock was deposited later than the other, then it's fossil must be younger than the other one.
But if you want to get down and really funky then radiometric dating is for you. This is where you use the known decay rates of nuclei (the middle bits of atoms) to tell you how long a rock has been sitting there. When the rock is liquid, it freely exchanges material with its nearby environment, keeping the ratio of different nuclei in equilibirum. But, when it cools past its 'closure temperature' the atoms are locked in, and new atoms can't get in. For that point on, you can use the different decay rates of different nuclei to tell how old it is. For instance, if you know there should be 1% of Unobtanium in this rock, but you only found 0.5%, then you know half of it decayed away. If you know how long it takes to decay on average, you can figure out how long it was since the rock was at its closure temperature. This method only dates back to the time that the rock was last heated. It might have sat there for a billion years, but if it was heated above its closure temperature only 10 years ago (by a volcano, say) then dating the rock will only tell you it's 10 years old. If the cosmetic industry knew of this, they'd take over the world.
One classic type of radiometric dating is carbon dating. When plants are alive, they take carbon dioxide from the air, and use that carbon to build themselves. This means that ratio of different carbon ISOTOPES (still carbon but with a different number of neutrons) will be the same as in the air. However, when the plant dies, it stops taking carbon from the air, which is exactly like the 'closure temperature' of a rock. From that point on, using the different decay rates of carbon isotopes, you can figure out when the thing died.
This is why the jobs of cartoonist and scientist are not the same.
The most important thing about all dating methods though, is that you need some sort of calibration in order to get going. In each of the cases above, we say things like 'if you know such-and-such, then this-that-and-the-other'. Even in the simple case of tree rings, you need to know that the tree makes one ring per year. If the tree only made one ring every three years, then you would need to multiply the number of rings by three to get the age. More worrying is the thought that the tree might make a random number of rings per year, in which case, counting the rings gives no information about its age at all.
Every method has its own calibration, and setting it up is a delicate task. It relies on finding that the weight of evidence from several dating methods all points to the same age. Now this is where a lot of people (especially Divine Creationists) run into trouble - they see these calibrations and they see a whole interlocking web of faulty mechanisms, each one lying to support the lies of the others. However, this misses some key points.
i) If the problem is that the data is 'noisy' (meaning that the dating methods are just plain randomly unreliable) then it is unlikely in the extreme that they would largely agree with each other - that in itself is something that would need to be carefully explained.
ii) If they aren't random, and all the ages agree, yet the age is still wrong, then that is also something that needs detailed explanation.
For either case i) or ii), there are HUGE implications for every other area of science, since the sort of explanations you would need would overturn all established physics and chemistry. And so far they're working pretty well. So let's not be too hasty.
Chris mentioned a dating method used to figure the age of songs released by the band Faith No More. Here is the table used:
- Sweet free Lecture Presentations: Eurispet-Williams (2008), and Geoscience Australia: Williams (2004).
- Dickin, A. P. (1995), Radiogenic isotope geology, Cambridge University Press.
- Radiometric Dating in a nutshell: http://www.encyclopedia.com/doc/1O112-isotopicdating.html.
The end of the world is nigh! Of course. When isn't it? The Mayans, smugly secure in their now extinct civilisation, predicted the end of ours. Those Mayan competistors thought they were so good.
All the hubbub about the latest descruction blockbuster 2012 is from the Mayan calendar, which apparently runs out of days in December 2012. Holy turn-the-page, Batman, we're doomed! Ridiculous? Of course not.
Here's a good description of what it's all about.
As far as the film goes: the acting was wooden, the storyline was weak, the science was unrealistic and the effects were over-the-top. It was alright.
But maybe this is what the Mayans were really worried about:
- Brian Dunning's (Skeptoid) take on it: http://skeptoid.com/episodes/4093.
- 220 end-of-the-world predictions: http://www.bible.ca/pre-date-setters.htm.