Alan V. Morgan
One of the "advantages" of living to be a more "senior" person is the hindsight gained in looking back over your lifetime. In the half-century (plus a little) that I have been interested in geology, the geological time scale has been refined time and time again. For example, when I started university, back in 1961, the base of the Cambrian period was 600 million years, then placed at 530 million (1982), back again to 590 million (also in 1982), then to 570 (1990) and now, perhaps finally, placed at 542 ± 0.2 million years (Ma). Remember that the Cambrian and younger rock sequences only represent a small part of the geological time scale that has been established over 200 years of geologic observations. But perhaps I am getting a little beyond myself.
Certainly at the age of about 10 or 11 I was looking at rock sequences in my hometown of Barry in South Wales. I would like to share a few of these observations since they throw some light on the perplexing problem of sorting out the age of the world. I was really fortunate in living in a part of Britain where the geological structures were only moderately affected by earth movements in the past and so it was very easy to observe the juxtaposition of beds. This brings me to the first aspect of dating termed:
Relative dating
Figure 1: Horizontal basal Jurrasic strata, Old Harbour, Barry, South Wales.
Figure 1 shows a rock sequence exposed in the sea cliffs about one kilometre from my childhood home. The beds here are horizontal and they display three of the fundamental laws that cover stratigraphy. The Principle of Superposition first expounded (or at least documented) by Nicolaus Steno over 300 years ago states that the oldest (first deposited) bed is always at the bottom of a given sequence and the youngest (last deposited) bed is always at the top. (The inherent assumption is that the sequence has not been totally inverted). The Principle of Horizontality is another assumption that the sediments that form the bed were laid down horizontally. In a sequence that is as well defined and undisturbed it is easy to see the fundamentals of a third principle. This is the Principle of Lateral Continuity; where one can follow a bed for hundreds of metres (or even kilometres) until the bed finally either grades into another lithological unit, thins out and vanishes, or is cut off by some other factor. It might drop below beach level, be cut by topsoil at the ground surface or terminate in a rock break such as a fault or a sharp contact with another rock type.
Figure 2: Angular unconformity at Friar's Point, Barry Island, South Wales.
Figure 2 illustrates another outcrop on a nearby headland. Again the geology is relatively simple but it shows a fourth important principle. This is the Principle of Cross-cutting Relationships. The Principle of Cross-cutting Relationships states that in a given rock sequence if "something" (and the something might be an igneous intrusion, a mineral vein, a fault or younger strata) cuts through a sequence of beds then the "something" must be younger than the beds through which it cuts. In Figure 2 it is fairly easy to see that the lower beds (originally deposited horizontally) have been subsequently tilted and then cut by a second (horizontal) sequence. Even as a younger person it was not difficult to assume that the tilted sequence was older than the horizontal strata above the dipping beds. More importantly this illustrates the concept of relative ages. Let's put this in order of events, so we start at the bottom (the oldest part).
|
Stage 8: Formation of topsoil and growth of grass. |
By working your way from one local outcrop to the next "up" or "down" through each sequence, it is possible to find additional regionally younger and older units and to determine how beds might grade laterally from one lithology into another. This provides a "relative" way of assessing the oldest and youngest beds in the area being studied. Perhaps one of the most famous examples in the world is "Hutton's unconformity" at Siccar Point in Berwickshire, Scotland (Figure 3). It was at this locality that James Hutton recognized the true immensity of geological time and helped the natural historians of the day to break away from the religious dogma that supposedly (according to the Irish Bishop Ussher) dictated that the world was created in 4004 BC. If you are dealing with metamorphic or igneous terranes that have no fossils then you can probably at least determine which are the oldest and youngest rocks and the age progression of each by using cross-cutting relationships. An example is illustrated in Figure 4.
Figure 3 (ABOVE): "Hutton's Unconformity" in Scotland.
In sedimentary units your observations will not only include the lithology of the beds that you are mapping (which allows a determination of the type of environment that existed when the sediments were deposited), but also the fossil content of the beds concerned. Fossils may be blatantly obvious, or they may be totally absent. Most sedimentary rocks do contain macroscopic or microscopic fossils and a careful determination of the families, genera and species might also allow the "age" of the beds to be determined. Certain fossil groups are "generalists" existing for huge blocks of geologic time with little or no apparent change. The small intertidal brachiopod Lingula, looks the same today is it did in rocks as old as 500 million years. On the other hand, more rapidly-evolving organisms such as certain graptolites or ammonites, changed morphologically over "short" periods of geologic time (perhaps 500,000 years). We use the fossilised remains of such rapidly-evolving animals or plants as zonal fossils.
Figure 4 (BELOW): Igneous rock near Kingston, ON. The light angular unconformity (grey overlying beds) contains clasts from the lower sequence, and must be younger.
Zonal fossils have certain requirements. In a rather simplistic way, these are: rapid evolution (they exist for short time frames); easily recognised (usually a function of hard parts that are easily preserved); wide geographic distribution (found in many different parts of the world), and presence in rocks of different lithologies (many depositional environments). Many excellent zonal fossils were "drifters" in marine sequences of the past. Because zonal fossils existed for short widows of geologic time one can use their presence to correlate one outcrop with another. This is true in regional contexts but might also be true for world-wide correlation. Lingula, for example, would make a most unsuitable zonal fossil because finding specimens in a rock sequence would not provide much information about age (only that the rock sequence falls somewhere in the last 542 million years of geologic time)! On the other hand finding Psiloceras planorbis (an ammonite) would be extremely useful over much of western Europe since it only exists in the lowest zone of the Jurassic (199.6 +/- 0.6 Ma). Putting a geological age such as this on a fossil takes us out of the realm of relative dating and into:
Absolute dating
"Absolute dating" implies "absolute dates" and this is not quite correct. To a geologist it illustrates that a certain radiometric decay series has been used (sometimes more than one) to determine the rock age. It does not imply precise calendar years, but it does provide a framework which, with limitations, provides an approximation of the ages of rocks. For example the age determinations on the junction between Cretaceous and Paleogene (Tertiary) rocks are 65.5 +/- 0.3 Ma. Multiple determinations using different dating techniques on this boundary from many different areas of our planet average at this age. Rocks older than this (Cretaceous) contain dinosaur remains. Rocks younger than this (Paleogene in age) do not contain dinosaur remains, unless odd bones have been weathered out and are incorporated in younger sediments. If you wander the badlands of Alberta you will find dinosaur bones (sometimes lots of them; - see What On Earth, Fall issue 2003), but this is not an indication that dinosaurs are alive today. Dinosaur bones can therefore be incorporated into sediments that are being laid down today in the Red Deer River valley, but to a future (insectoid?) geologist it should not be an indication that dinosaurs were alive in the Quaternary Period (the time frame that we live in today). On the other hand, bones of humans cannot be found in sediments of Cretaceous age because humans simply were not alive on Earth at that time.
Table 1: Main radiometric dating techniques
Elements and isotopes involved |
Half-life |
Age range provided> |
Dates obtained from |
| Uranium - Lead Series | |||
| 238U - 206Pb | 4.6 Billion years | Age of Earth | Zircon |
| 235U - 207Pb | 713 Million years | Age of Earth | Zircon |
| Potassium - Argon | |||
| 40K - 40Ar | 1.3 Billion years | Age of Earth | Micas (volcanic rocks) |
| Radiocarbon dating | |||
| C14 - N14 | 5,730 years | -150 to ~50,000 | Organic materials |
Painstaking analyses of different rock sequences all over the world by generations of geologists have provided observations that allow us to state, with a fair degree of confidence, what types of fossils should occur when and where. Of course Mother Nature throws us an occasional odd curve ball, such as the discovery of living coelacanths in the Indian Ocean, when the last fossil specimens found in rock sequences date back to about 70 Ma!
I have implied above that we use different types of dating methods to provide ages on rock sequences. A discussion of each of them is beyond the scope of this article, but I would like to mention the main techniques and the ages most suited to that determination. These are summarised in Table 1.
A detailed discussion of each of these methods is inappropriate. Suffice to say that the parent product will decay to a daughter product (often through a whole series of subsidiary decay series). The measurement of the remaining "parent" isotope to the abundance of "daughter" isotope provides a measurable difference that allows a calculation of the "age" of the rock under examination. If you are really interested in following this topic I refer you to the webpage of the International Stratigraphic Commission.
I have provided a simplified example using radiocarbon-14 as an example. This is also known as Carbon-14 or radiocarbon dating. Radiocarbon dating is extremely useful to Quaternary geologists looking at the younger events of the last ice age and to archeologists, studying the past 40,000 to 50,000 years since practically all organic materials - shells, bones, teeth, skin and hair, peat, wood, (and structures made from wood), or charcoal - can be dated. There are some problems with the technique. Human use of fossil fuels (coal and oil particularly) has added vast quantities of "infinitely old" carbon to our atmosphere over the past 200 years. Furthermore, above-ground testing of large nuclear bombs, especially in the 1960's, added radioactive contaminants that complicate analyses of natural decay rates. For these reasons dates prior to about 1800 are considered reliable and dates following 1800 are not usually attempted. At the more distant end of the methodology dates beyond about 50,000 are also considered unreliable.
How does this technique work? Cosmic rays from our Sun, enter the upper atmosphere forcing the conversion of Nitrogen14 to Carbon14. Most Carbon14 is oxidized to form Carbon Dioxide (CO2). The CO2 is then absorbed by plants, which are then eaten by other organisms. However, some of the original Carbon14 is not altered and the ratio of original Carbon14 (the unstable form) to CO2 (the stable form) is about 1 atom to one trillion atoms. When the plant or animal dies it stops taking in CO2 and the decay of Carbon14 begins. Since the dead plant or animal cannot take in any more CO2 or Carbon14 the "radiometric clock" within the organism begins to count down. We know the precise decay rate of Carbon14 since it has been measured against annual rings taken from long-living organisms, sections of giant Sequoia trees, for example that are >2,000 years old or Bristlecone Pine (>4,000 years). The Carbon14 in the sample decays to Nitrogen14 in a constant manner, with 50% changing (or "one-half life" changing) every 5,730 years. Put in another way, 5,730 years after the organism dies only 50% of the Carbon14 contained within the organism remains. After another 5,730 years (or at 11,460 years after the organism died) only 25% remains. Unfortunately by 7 half-lives (age 40,110 years) less than 1% of daughter isotope is left. This explains why dates greater than 50,000 years before present are considered unreliable, since the amount of measurable activity is exceedingly small. Incidentally, "before present", is by convention before 1950. Conversion scales for "carbon-14 years" to real "calendar years" are also available in the very youngest part of time.
Radiometric dating, therefore, depends on when the "parent" material was formed and the rate of decay of that parent material to the daughter product. In the case of igneous rocks it depends on the time of crystallization of the parent mineral in the rock that is being dated. Let's take the case of a granite, an igneous intrusion that is, in reality, two billion years old. The atomic clock starts counting down when the magma cools. Unfortunately the rock sequence is involved in a later reheating (perhaps another mountain-building episode) at one billion years. The atomic clock within the granite is "reset" by this later reheating and partial melting episode, and acquires the "new" later age. Geologists must be aware that they are dating this later episode of mountain building and not the original age of the granite. Careful mapping and dating elsewhere in the region can usually resolve such conflicts.
When the granite disintegrates a geologist might date grains of feldspars (using Potassium40 dating) that are trapped in sedimentary rocks. The sedimentary rocks are, perhaps, 600 million years old. However, the geologist would have to be careful to realize that they might be dating grains that come from the original granite that cooled 2 billion years ago, or from the "reset" feldspar ages that were derived from the altered granites one billion years ago. Neither of these ages would reflect the "true" age of the sedimentary sequence that is 600 million years old.
The net result is that with reliable dating methods that have been developed largely over the past three or four decades, coupled with meticulous geological mapping over the past 200 years, the geoscience community has been able to unlock the history of our world. We know how old it is, when life forms began to inhabit the Earth; when major catastrophes and calamities have overwhelmed our planet; when new species have appeared or evolved, in many cases, provide the rates of evolution. Finally we have been able to say when certain mineral deposits (including all the energy minerals, so necessary for our survival) were created and also how our world has physically evolved. Much remains for future generations of geologists! The current geological timescale is featured in the centrefold of this issue.
Some comments on the Timescale and Teminologies use
The timescale illustrated in the centrefold is the latest version of many. This is the currently accepted scale with the latest refinements using absolute time. The meeting of the 32nd International Congress, to be held in August of 2004 in Florence, Italy, will ratify a number of the latest amendments.
Please note that I have not used the correct colours for the various divisions of geological time (there are indeed designated colours) and I have introduced a few of my own views. For example, in the left side of the diagram in the Proterozoic I have kept the "Aphebian" "Hadrinian" and "Helikian", rather than just absorbing the newer terms (Siderian to Ediacaran); although the latter are included. The Edicaran (latest period of the latest Precambrian) is now known to have many fossils in various parts of the world. My suspicion is that sometime in the future this might be absorbed into the Phanerozoic Eon, although I am unaware of any suggestions for this to take place. At the very bottom of the timescale I have included the oldest radiometrically dated rocks (extra-terrestrial and on Earth). Note that some terrestrial minerals (zircons) are dated to between 4.3 and 4.4 billion years.
In the Paleozoic (Carboniferous Period) I have indicated the current time break between the "lower" (Mississippian) and "upper" (Pennsylvanian) subdivisions used extensively in North America.
In the Cenozoic, one of the most important changes will be to delete the long-familiar "Tertiary Period" and replace it with the Paleogene and the Neogene, divided at 23.03 million years. There is also a proposal to delete the Quaternary and replace it with the "Pleistogene". I have not included this in the chart.