Re: How do we know with precision, how old a rock is?

Alex:

There are many different ways to learn the age of a rock, and each method 
has its own degree of accuracy and precision. All absolute rock ages depend 
ultimately upon radiometric age dates because this is the only method that 
yields ages in years. 

Radiometric age dating depends on the fact that radioactive isotopes decay 
to other elements at a constant rate. In other words, for each atom of a 
radioactive isotope, there is a fixed probability that it will decay during 
a given length of time. If you consider a group of such atoms, such as the 
billions of them present in a small rock sample, we know the amount of time 
it takes half of the radioactive atoms to decay. This is the same amount of 
time it takes half of the remaining half (1/4 of the original set) to 
decay, and so on. So, the amount of remaining radioactive isotope can be 
compared to the amounts of its decay products, to find the amount of time 
it has been sitting in the rock, and isolated from sources of new 
radioactive atoms.

Other methods yield relative ages. 

However, there are a variety of radioactive materials. For instance, 
Carbon-14 decays rapidly, and is useful to date objects thousands to a few 
tens of thousands of years old. Uranium 235 decays much more slowly to 
lead, and this decay series is useful for dating rocks a few million to 
many millions of years old. Other radioactive isotopes have their own 
useful age ranges. So, the basic idea is to find a rock that contains a 
certain radioactive isotope and its decay products. This rock might be the 
very rock you're interested in, or it might be another rock whose age can 
be related to that of the rock of interest. This is where the "many 
different ways" mentioned in the first sentence come in. 

For instance. Suppose you are interested in the age of a sandstone unit. On 
top of the sandstone unit is a bentonite, a very special clay layer that 
used to be volcanic ash. Volcanic eruptions last for days to months, and 
cover large areas with thin layers. These ashes contain radioactive 
isotopes and can be dated. If the bentonite is dated, the sandstone beneath 
it is probably only slightly older. If there appears to be no break in 
sedimentation (indicating missing time) between the sandstone and the 
bentonite, then for practical purposes they are the same age. 

Here's another example. The sandstone may have been deposited in the sea, 
and contain fossils. One of these fossils could be a particular species of 
ammonite, known from previous work elsewhere (including multiple 
radiometric age dates) to have lived only during a period of 800,000 years. 
This is a very short time, geologically, and the sandstone has been dated 
both accurately and precisely.

If there is no bentonite, and no age-diagnostic fossils, then we can't be 
as precise. However, we can still estimate the age. If you go to younger or 
older rock units, you will find layers that can be dated one way or 
another. Then, you know the sandstone is younger than one age, and older 
than another. You can estimate the amount of time represented by the 
various layers and surfaces between the two dated layers, and be a little 
more precise about the age of the sandstone. However, if the two dated 
layers are much different in age from one another, then the age estimate 
for the sandstone can't be extremely precise. 

I hope this has helped. 

David C. Kopaska-Merkel
Geological Survey of Alabama
PO Box O
Tuscaloosa AL 35486
USA
(205) 349-2852
FAX (205) 349-2861
Email: davidkm@ogb.gsa.tuscaloosa.al.us