Geological Dating

An animal doesn't simply drop dead on the ground, lie around for a lot of years and fossilize. The flesh gets eaten; the bones are crushed, weathered and pulverized. The unfortunate creatures have to be buried in some way, and either quickly or in an environment deprived of oxygen. Chemical and physical processes may then set about gradually replacing the remains with minerals. Only very rarely are any organic traces preserved. And also gradually, the burying medium, be it mud, sand, lime or volcanic ash, will be compressed under the weight of the accumulating material above, until it's formed into a rock strata.

Previously, a different sediment might have been deposited at the same place; a marine sediment, for example. Maybe a further marine bed will form above.

Now we have three distinct levels, with the remains of our Deadosaurus sandwiched in the middle. We have already got some useable information. The top bed must have been deposited most recently and the lower strata can only be the earliest. Even if later folding has tilted the rocks, or even tipped them upsidedown in extreme cases, there are virtually always 'this-way-up' indicators, eg. ripple marks, which show the original sequence. In general, remains found in any level will correspond with the age of the sedimentation, except in rare cases of secondary deposit. This doesn't tell us much perhaps, but it's a start. Fossils in the top bed are the newest, whilst those in the deepest are the oldest. This is the foundation of relative dating.

Because many people find the subject interesting, and because the information obtained can be very useful for prospecting purposes and construction projects, an immense amount of time and effort has gone into the study of stratification over the past two centuries. Dedicated amateur fossilers have built up highly detailed geological summaries of many localities. Engineering projects, (eg, road and railway construction), and industrial activities, (eg, mining and quarrying), have allowed broader details to be examined. Geologists and mapmakers have augmented this wealth of material with field studies of their own. One way or another, many thousands of people have contributed to the study of stratification, and continue to do so. This has resulted in a remarkably detailed picture of some past events.

There is an order evident in the rock record, which cannot be reasonably denied. The earliest deposits contain no fossils of multi-celled organisms. The remains of vaguely woodlouse-like sea creatures, the trilobites, are found in layers which are older, and never younger, than those containing mammals. No mammoth bones have ever been deposited in or beneath beds containing dinosaurs. The various strata were laid down in sequence, and each contains the remains of animals and plants which died during the time concerned.

Ammonites were common marine creatures of the Mesozoic, though they're extinct now. Their shells have frequently been preserved as fossils. There is a great variety of different kinds, many of which are relatively easy to identify, given the necessary knowledge, experience and reference books. You get small lumpy-bumpy ones, large whirly-twirly ones and so on. Often, a particular type will only be found in a certain sediment bed. It doesn't occur either above or below. This suggests it only lived in the area for a relatively short time, geologically speaking.

Sometimes, the same ammonite genus might be identified in association with similar fauna in another narrow bed, hundreds, or thousands of miles away. It may even turn up at numerous sites. This gives us something really useful; a clearly identifiable animal, which lived over a wide area, but only for a relatively short timespan. This is a first-class Leitfossil. Many such ammonites are known. Leitfossils, (from the German for guide-fossils), are an important means for tying sites chronologically together. Even if a strata contains no Leitfossils itself, it might still be roughly dated with reference to any found above or below. It's such clues which show that the Portland Limestone in Dorset (England), the Solnhofen Plattenkalk in Bavaria (Germany), and the Morrison Formation in Colorado and Utah (USA) are all about the same age; Tithonian - uppermost Jurassic.

What's also of interest about Leitfossils, (which are often called trace fossils), is that this form of dating isn't actually dependent upon any one particular organism. The same ammonite crops up within a broadly similar community. In practice, it's the whole assemblage which provides the decisive evidence, rather than simply the presence of one particular fossil. Furthermore, it's often possible to use entirely different kinds of remains to verify the results. It's not uncommon to read of the age of the same location being tested and resolved by independent studies of insect, crayfish, crab and various other critters.

Stratification is the basis of knowing that dinosaurs seem to have evolved in the Middle Triassic, Diplodocus lived in the Upper Jurassic and Triceratops during the Upper Cretaceous. What it doesn't do directly is to provide any numbers. That's where the death certificates come in.

I've recently read several articles which inform me that carbon dating is not a reliable method of determining the age of dinosaur bones. I fully concur with this view. Nevertheless, to illustrate the principles involved in radiometric dating, and because carbon dating is a reasonably well known term in the wider world, let's briefly look at how and if and why it works.

Carbon dating is used to assess the approximate age of organic material. Carbon-14 is a radioactive isotope of carbon, which is produced in the upper atmosphere by the reaction of nitrogen, under bombardment from cosmic rays. Because it's in the air, we and all other terrestrial organisms absorb small amounts of it. When an animal dies, it stops eating and breathing. This absorption ceases.

Radioactive material is, by definition, unstable. Over time it decays down to daughter products. In the case of C-14 about 50% of the original material decays within 5,715 years. This is termed the half-life. If you know how much C-14 was present when the animal died and measure how much there is now, you can theoretically deduce how long ago the funeral took place. This applies to all terrestrial organic remains, not just animals.

The rate of C-14 in the atmosphere is not fixed, but nor does it appear to be prone to wild fluctuations. Unless there's fuller information available, (eg. C-14 levels as recorded by tree rings, ice layers in Greenland or annual lakebed sediments), the present atmospheric level is used as the starting point for calculations. This means the technique is less than 100% accurate, but not generally much less.

If the half-life was variable, that would be a problem. Extensive and continuing testing strongly evidences, however, that it isn't. Several atomic half-lives have been observed to vary fractionally, but only under most peculiar circumstances involving extreme pressures and speeds, in line with Einstein's theory of relativity, apparently. (I could supply more details, though I don't understand them.) None of these circumstances are applicable to any dating methods. "Radioactive atoms used for dating have been subjected to heat, cold, pressure, vaccum, acceleration, and strong chemical reactions without any measurable change. " (1)

Our best information is, based on nearly a century of observation, radioactive half-lives are astonishingly regular. That's why atomic clocks work so well.

I mention C-14 in particular because these essays I've been reading seem obsessed with it. They overlook a rather significant point. Dinosaur remains are rarely organic in any degree. Pick up a convenient Mesozoic fossil and you will notice it's composed of stone. That's why it's heavy. That carbon dating is not appropriate should be obvious, with a little thought. It makes me wonder why some people, Dr Kent Honvid (2) for example, seek to make such an issue of it. Stone doesn't eat or breathe. Therefore, it absorbs no C-14 from the atmosphere. That alone makes the technique unusable for dinosaurs, regardless of time-scale considerations. Appropriate dating methods work in accordance to much the same principles, however, but are based upon the original chemical composition of the subject material.

There are over fourty methods in use for dating rock samples. Each involves a radioactive isotope, (the parent), and a product, (the daughter). Each calculated half-life is subject to rigorous scrutiny and the uncertainty factor is at most 5% (rhenium), and generally less than 2%. Here are some naturally occuring examples: (1)

Parent	Product	Half-Life in Years
Samarium-147	Neodymium-143	106 billion
Rubidium-87	Strontium-87	48,8 billion
Rhenium-187	Osmium-187	42 billion
Lutetium-176	Hafnium-176	38 billion
Thorium-232	Lead-208	14 billion
Uranium-238	Lead-206	4,5 billion
Potassium-40	Argon-40	1,26 billion
Uranium-235	Lead-207	0,7 billion
----------	----------	----------
Beryllium-10	Boron-10	1,52 million
Chlorine-36	Argon-36	300,000
Carbon-14	Nitrogen-14	5,715

"Notice one other important detail about radioactive isotopes. Most of the naturally occuring radioactive isotopes mentioned above have very long half-lives, on the order of billions of years. The only ones with shorter half-lives are those which have a source constantly replenishing them, such as carbon-14, beryllium-10 and chlorine-36 produced by cosmic rays. We can make hundreds of other radioactive isotopes with half-lives shorter than a billion years, but they do not occur naturally on earth.

Occassionally there is evidence that these isotopes existed at some point in the past, but have since decayed completely away." (1)

Not all rock is suitable for each, or indeed for any of the techniques available. Uranium- Lead demands the presence of uranium, otherwise it's clearly useless. The exact methods employed have to match the qualities of the particular sample in question.

The best candidates for radiometric dating are igneous, (volcanic), rocks. The lava cools, and the atoms are constrained within what is virtually a closed system. Any resultant daughter atoms are trapped and ready to be counted. However, there might already have been some daughter atoms present in the source material. Some volcanic lava, in Hawaii for instance, can be very inconvenient. It's low in potassium whilst being rich in argon, because its source material happened to be argon rich. For the results to be meaningful, you have to be able to determine how much daughter product was already present, before crystalization. If this can't be done, then that particular rock can't be sensibly dated. Usually, a comparison of the parent-daughter ratios in different minerals will provide the answer. Sometimes, more sophisticated methods are called for, involving other daughter products or isotopes and perhaps a handy nuclear reactor. For a far fuller account, see the paper by Dr Roger C Wiens (1).

Strictly speaking, radioactive decay isn't actually used to date Mesozoic fossils. It's primarily employed to date igneous rock, where possible. This would appear to pose a bit of a problem. Fossils are typically found in sedimentary rocks, which are generally not suitable for radiometric dating. Igneous rocks, which can be dated, are not likely to contain any fossils. As a consequence, fossils, or indeed fossil bearing strata, cannot usually be directly so assessed. The solution lies in the application of stratification, whereby igneous materials serve as chronological bookmarks.

Earlier, we left our Deadosaurus fossilizing in a layer, sandwiched between two marine sediments. We know it's something like middling old, in terms of the local geology, whilst recognizing that the three levels were probably deposited, and subsequently eroded, at different rates. Given no further clues, something like middling old will have to suffice.

Underneath the earliest marine bed there happens to be some igneous rock, which dates to 200 million years. Our dinosaur is clearly not as old as that. There's no igneous rock above our upper marine bed. However, at a location 300 miles away, there's another deposit with the corresponding Leitfossils present. Overlying volcanic material there yields a date of 180 Ma. Applying the information which is locally available, the information from the corresponding site and a bit of common sense mathematics, we can deduce our Deadosaurus died something like middling between those two dates; approximately 190 Ma.

It's sometimes said that rocks are dated according to the fossils contained within them. This is true, as far as relative dating and Leitfossils are concerned. That technique has just been employed in the hypothetical example above. It doesn't apply on igneous strata, however, when radiometric dating methods are used. There are no fossils to compare.

Nor should unexpected results be casually disregarded. They should be examined and explained. In practice, contamination of the sample might be the cause. Sometimes, however, a surprising result is correct, or at least indicative of something mineralogically interesting.

The dates which really matter are those which are testable and verifiable, using a variety fo radiometric methods. The Hell Creek formation in North America, where there's also considerable evidence of past volcanic activity, provides an interesting example (3). Numerous tests by various teams employing different techniques on a number of occassions at a variety of sites have produced broadly similar results, +/- a couple of hundred thousand years. The border between the Cretaceous and the Tertiary strata, which is clearly identifiable at Hell Creek, is shown as being 65 million years old. The border between the last two epochs os the Upper Cretaceous, the Campanian and the Maastrichtian, is about 72 years old, +/- a bit. There must be a reasonable explanation for this consistency and there is. The results are correct.

If fossils are found above the K-T border, they're younger than 65 million years. If they're discovered just below it, they're around 65 Ma. If they're found a bit below the Maastrichtian, then they're something like 72 million years old or so. Nobody is claiming any greater degree of precision, as far as I'm aware.

Having established a couple of trustworthy dates, we can now employ stratification again. Other sites which are Maastrichtian should also be round about 65 - 72 miilion years old. It's quite conceivable there may actually be some regional variation in all this. Maybe European ammonites survived some thousands of years longer than Australian ammonites, or vice versa, or not. I've no idea. But a possible few thousand years are of no significance, because the dates are only meant as approximations anyway. No method is known which is more precise. However, given that there is overwhelming evidence of global extinctions at, or near, the K-T border, it does seem reasonable to assume much the same date for every K-T site. And this border is especially well-defined on account of the ammonites and their subsequent absence from the fossil record, but also because of the numerous examples of an iridium abnormality recorded around the world, etc.

Most dates for the Mesozoic aren't perhaps as well established as the K-T border, because they haven't been as extensively examined. Nevertheless, they're also far from arbitrary. They are based upon testable data. All are constantly subject to revision and, if the evidence demands it, adjustments to the theoretical geological ages occur. No major alterations have been made for many years, however, because no major discrepencies have been found. Should any arise..., now that would be interesting.

(2) Doesn't carbon dating or Potassium Argon dating prove the Earth is millions of years old? by Dr Kent Hovind

How to date dinosaurs