How are scientists able to tell how old a rock is?



How are scientists able to tell how old a rock is?

In: Chemistry

You have probably heard of radiocarbon dating which is used to date organic elements based on how much carbon-14 have decayed. This is not usable on rocks because it does not contain carbon. However there are a number of similar radioisotope dating techniques for other elements besides carbon-14. For example potasium-argon dating. This works on any mineral that contains potasium. Some of this potasium is the isotope potasium-40 which with a half life of over a billion year turns into argon-40. You would not expect there to be any argon in molten or dissolved rock because argon is a gas that will just bubble up to the surface and escape into the atmpsohere. So all of the argon-40 must come from potasium-40. By measuring the ratio between these two isotopes you know how much have decayed and therefore how long it have taken. There are plenty of these types of dating techniques. Which ones work depend on the age and composition of the rock. You would usually try a few different techniques and then compare these results to each other for an accurate result.

There are already several very fine answers to this question so I want to address some issues that non-scientists sometimes have trouble with, The first is the idea of accuracy vs precision. Accuracy is the closeness to the correct answer, precision is a measure of the smallest unit that can be determined. It is thus possible to have a very precise answer that is absurdly inaccurate. Getting an accurate answer requires one set of conditions, getting a precise answer requires a different set. Both sets must be met to have a good answer.

Another issue that comes up is the age of the rock vs the age of its constituent grains or minerals. In the case of volcanics getting an age date on the time of crystallization of some of the minerals is not the same as getting the time of eruption. And in sedimentary rocks there is the time of deposition of the grains, the time of solidification of the cementing minerals, the possibility that some of the cementing minerals may be at least partially dissolved and then new ones precipitated in the voids. All three will give different ‘ages’ for the rock.

A third issue that affects sedimentary rocks is that of erosion, transportation, and redeposition. This one can be particularly tricky if one or a very few fossils are used to give a date. In a particular instance I encounterred a few years ago the law of superposition indicated one age but the included fossils indicated a much older (like about 100 my older) date. We resolved the issue by comparing the fossils in questionable bed to those in other beds and noting that the fossils in one were heavily eroded as would be expected in the stream deposited sands and pebbles of the younger formation while they were vey shap and well defined in the older formation from which they had been eroded and subsequently transported.

Geochronology is a wondrous thing but it is neither the worthless joke some maintain nor the infallible ‘truth’ some non-geologists think it is when they read a tag on a specimen in a museum.

Radiometric dating is one of the best methods. Rocks can incorporate certain elements into their crystal structures when they form while rejecting others. For example Zircon can incorporate uranium atoms but strongly rejects lead atoms when it forms.

However once the rock has formed its crystal structure has effectively set, uranium is radioactive and so over time it will decay down to lead (with a half life of hundreds of millions to billions of years). Any lead in zircon must have been formed from uranium which has decayed because it couldn’t have been there when the rock formed. This means that by measuring the amount of lead compared to the amount of uranium you can work out how long the rock has been solid for, i.e its age.

If it’s a male rock you can just straight up ask it. If it’s a female rock and especially if it looks more on the older side then it’s considered kind of a social faux pas.

Well you have the absolute dating of a rock with the radiometric method described by the others here. But what us also important is relative dating. If you dig into the ground you get layers with specific characteristics. How deeper you go how older the (sedimentary) rock grnerally. You can try to describe this sequence of layers and correlate it with the sequence of layers in the region with similar conditions. And you don’t have to dig hundreds of meters deep either, generally these layers are not flat and level like they used to be but they are deformed and not level with the ground. Because most peaks from geological history rock from different geological periods will be found on the surfafe at different locations, snd so you can kinda try to puzzle together this sequence of characteristic layers. Although you have to be careful with this as well. This layer profile cannot be extended too far because conditions naturally are different in different regions now, also in the geological past. The further two locations are apart the harder it gets. Also depositions can also vary gradually in space and not in time, eg. when a river deposits a sandy layer in the nearby sea and clay somewhat further for a long period of time while the coast moves, then these sand and clay will not correspobd with the same period.
Also nice is to use fossils to correlate layers, because species appear, die out, become less and more common over time.