This is a blog about paleontology (the study of the history of life on Earth through the fossil record) with an emphasis on vertebrate paleontology, the study of extinct vertebrates (animals with backbones). The methodology and findings of paleontology will be discussed, as well as related issues such as evolutionary theory. The blogger is a vertebrate paleontologist specializing in the Triassic Period, the Beginning of the Age of Dinosaurs.
My posts are presented as opinion and commentary and do not represent the views of LabSpaces Productions, LLC, my employer, or my educational institution.
In the last two blogs (click here for Part I and Part II), we've been talking about ways of determining the relative ages of sedimentary rocks and fossils (e.g. species A lived some time before species B), without determining their exact numeric ages (in thousands, millions, or billions of years). This is referred to as "relative age dating." If we start applying numeric ages, we are talking about "absolute age dating." The main method by which this is done is called radioisotopic dating. Explaining how radioisotopic dating works is going to require that I hop around a bit between subjects, but it will all come together in the end.
Just to give a little basic physics recap, an atom is made up of three types of particles: protons, neutrons, and electrons. The number of protons determines the type of atom (and element); for example, an atom of potassium (K) always has 19 protons, which is its "atomic number". The number of neutrons only changes the mass of the atom; atoms with the same number of protons but different numbers of neutrons are called "isotopes" of an element. The number of protons and neutrons together is the isotope's "mass number", and is written next to the symbol for the element, usually as a superscript. For example, K39 has 19 protons and 20 neutrons (19 + 20 = 39), while K40 has 19 protons and 21 neutrons (19 + 21 = 40).
Some isotopes are unhappy with the exact numbers of protons and neutrons that they have, and change them; these are called "radioactive isotopes" or "radioisotopes". Some change a proton into a neutron, some change a neutron into a proton, and others kick off two of each. All of these changes alter the number of protons (and also involve the release of charged particles, which is what makes them radioactive), so all also change the type of atom (and element). The original radioisotope is called the "parent", while the isotope it converts into is called the "daughter". For example, parent K40 breaks down into daughter Ar40 (argon has 18 protons, which tells you that Ar40 has 22 neutrons (18 + 22 = 40); this also tells you that when K40 decays, it does so by converting a proton into a neutron).
The important thing for our purposes is that all radioisotopes decay at a constant rate (which was determined by measuring the rate in a lab), and that the rate is different for different radioisotopes. This rate is usually expressed as "half life", which is the amount of time it takes for 50% of any amount of parent to convert into daughter. For K40, the half-life is about 1.25 billion years. SO, if you had 10 moles of K40, exactly 5 moles would convert to Ar40 after 1.25 billion years, and 7.5 moles would have converted to Ar40 (leaving 2.5 moles of K40) after 2.5 billion years. The decay takes place along a smooth exponential curve, like this:
SO, if we have a certain amount of K40, and a certain amount of Ar40, and we know that all the Ar40 started off as K40, then we can calculate how long it has been since all of it was K40. It's kind of like those math questions like "If a train leaves the station at 12 AM and is travelling at 50 mph, how long will it take to travel 150 miles?" OK? Moving on...
Let's talk about igneous rocks. Unlike sedimentary rocks, igneous rocks are formed from the cooling of molten rock (magma). Volcanic rocks are therefore igneous rocks. When magma erupts from a volcano (at which point it is called lava), and cools, it forms volcanic rocks like basalt and andesite. Some volcanoes (largely because of the chemistry of the magma, and the amount of water it contains) produce explosive eruptions which emit massive quantities of volcanic "ash". Volcanic ash is composed of tiny fragments of glass and rock, which formed from the rapid cooling of the magma and were then blasted into pieces by the explosive force of the eruption.
Here is a short film of the 1980 eruption of Mt. Saint Helens, which was originally a series of still photographs run through a morphing program. The dark clouds erupting toward the end are volcanic ash. Below is a picture taken from the other side of the mountain after the initial blast, showing volcanic ash continuing to be emitted by the volcano.
Millions of tons of volcanic ash can be produced by a single eruption and travel for thousands of miles before settling down over subsequent days and weeks. This will be important later.
Let's take a closer look at what happens when igneous rocks form. Igneous rocks are composed of various types of minerals which form as the magma cools. These minerals are composed of atoms arranged in a repeating geometric structure, and different minerals have different chemical compositions (i.e., they are composed of different types of atoms). Initially, the magma, being a fluid, is composed of different types of atoms moving around independently.
As the magma cools (which occurs rapidly during an eruption, once the magma is exposed to air and water), atoms begin to form bonds, resulting in solid minerals, each with their own chemical compositions. The formula for the mineral biotite is K(Mg,Fe)3(AlSi3O10)(F,OH)2, so when a crystal of biotite forms, it absorbs a lot of potassium (including the radioisotope K40) into its crystal structure...but little or no argon.
What this means is that any Ar40 you find in a biotite crystal formed through the decay of K40. So, if you can measure the amount of K40 remaining, and the amount of Ar40 present, you can calculate how long it has been since the crystal formed (i.e. how long since all of the Ar40 was also K40, i.e. about how long it has been since the eruption). The same thing is true of other minerals and radioisotopes. For example, U235 decays (through a long series of intermediates) into Pb206. The mineral zircon has a formula of ZrSiO4, but also incorporates uranium and thorium into its crystal structure (it substitutes for Zr)...but little or no lead. Therefore, you can use the relative amounts of U235 and Pb206 to determine how long it has been since the zircon crystal formed, just as with the biotite.
Better yet, tiny minerals, including biotites and zircons, can get transported in volcanic ash, and settle down thousands of miles from the volcano on top of sedimentary rock layers which formed before the eruption (obviously, or they wouldn't be there for the ash to settle on). The ash layer can then be buried by layers of sediments after the eruption. This means that you can use the Law of Superposition to establish the relative age of an ash layer compared to the strata around it...and if you can date minerals within the ash layer and get a numeric date, strata below the ash layer will be older than that date, and strata above it will be younger than the date. Ha Ha!
Remember the exercise in biostratigraphic correlation I used in the last post? Let’s introduce an additional complication. In three of our measured sections, we have volcanic ash layers which have been dated radioisotopically to 100 Ma (this abbreviation stands for "millions of years ago"), 70 Ma, and 50 Ma.
Now, let's correlate them biostratigraphically like we did before:
...and suddenly, our time scale is numerically calibrated. The purple starfish lived before 100 Ma, the blue oyster lived somewhat before and after 100 Ma and coexisted with the green belemnite between 70 Ma and 100 Ma (perhaps about 85 Ma?), and the red snail appeared between 50 Ma and 70 Ma (perhaps at about 65 Ma?)...and so on.
A couple more points about radioisotopic dating. Observe that when I put the three ash beds that I dated in chronological order within the composite time scale, I did it using the Law of Superposition and biostratigraphic correlation...not the dates that I calculated from the ash layers. If radioisotopic dating didn't work, there is no reason to expect the dates calculated to be in the right order (the top layer could have been 100 Ma, and the lowest 50 Ma)...but they do. The real radioisotopic dates we derive from the rock record also occur in the chronological order one would expect from their superpositional relationships. Moreover, dates calculated using different minerals and radioisotopes corroborate each other. Contrary to what creationists allege, there is no compelling evidence that radioisotopic dating doesn't work, and plenty of evidence that it does.
Another point is that the method discussed here is NOT radiocarbon dating. Radiocarbon dating is a very peculiar type of radioisotopic dating. Instead of dating minerals in igneous rocks, it dates organic material. Radioactive C14 is continuously absorbed by living things throughout their lives. Once organisms die, the C14 is no longer replenished, and breaks down into its daughter product N14. Radiocarbon dating therefore gives you an estimate for the amount of time which has elapsed since the organism died. It is used by archeologists to date materials like wood and animal hide, but is only good for materials about 60,000 years old or less. The reason for this is that C14 has such a short half life (about 5,700 years) that it breaks down far too quickly to be good for dating anything older than 60,000 years old; there is so little parent C14 left in older materials that it can't be measured accurately (the inverse is the reason why we can't use U235 of K40 to date things which are thousands or tens of thousands of years old; the half life is so long that too little daughter has been produced to measure accurately). Radiocarbon dating is therefore useless to most geologists and paleontologists (except for those studying relatively recent rocks and fossils), and anyone who tells you that radiocarbon dating is used to date things millions of years old (like dinosaurs) doesn't know what they are talking about.
What I've outlined over the last three posts is how the geologic time scale was constructed and calibrated, and how we have reconstructed the history of life on Earth. When paleontologists say anything about the history of life (e.g. dinosaurs went extinct 65 million years ago, long before the first modern humans, who only appeared about 250,000 years ago), they rely primarily on the three techniques I've outlined in the last three posts.
Now that I have given an overview of how the history of life was reconstructed, let’s take a brief walk through that history. Or we might talk about something else first, I dunno. I'll get back to you.
This post has been viewed: 5541 time(s)
Great post! Thanks Jeff!
How precise is the measurement of the decay of the isotopes? How is that done? For example, do you scrape a piece of the layer you want to test and measure the radioactivity with an instrument?
Are the isotopes in sedimentary rocks also? How do you do radioisotopic dating if you don't have igneous rock layers (no volcano in the vicinity)?
Radioactivity is sometimes measured directly in a couple ways, using either a giger counter, or by examining the amount of damage which has been caused to the crystal by radioactivity, to get an estimate for how much of the parent material has decayed, and how much remains. However, it is more common to measure the amounts of parent and daughter directly using mass spectometry, in which the crystal is partially vaporized and the atoms released in the process are identified by the spectometer.
I asked Randall Irmis at the Utah Natural History Museum, who has more experience with radioisotopic dating, your question about precision. Here is his respone:
"Precision for those methods is a very complex question, and it totally depends on the type of analytical method used to make the measurements. The simple answer is that the most precise methods for U/Pb and Ar/Ar reach per mil precision (i.e., 1 parts per million in measuring the isotope ratios). This translates to +/- .2 Ma for Triassic samples, but obviously a higher uncertainty further back in time, and lower uncertainty in younger rocks."
FYI: Ar/Ar dating is a more complicated but allegedly accurate variation of K-Ar dating, which measures the ratios of different isotopes of Argon (those produced by decay, as opposed to stable isotopes). Although I am not terribly familiar with the detailes of the method, it involves irradiating the sample first to converts other isotopes of potassium into isotopes of argon, and the ratio of different argon isotopes is used to estimate the original amount of K40.
I'd love to use some of these images in my Oceanography course, we're looking at Relative and Absolute dating. I usually don't go into much detail but these posts inspired me to cover it more. Can I use some of your images in a powerpoint and maybe a handout? I'll cite with the blog's general address and then your name, city, state, country, as in your profile.
I don't have a problem with anyone using any of the figures I prepare for the blog, which are most of the drawings. Other images (including photos) mostly come off of Wikipedia commons.