Geology of the National Parks

Textbook 10.2: The Grand Canyon


More on the Age of the Earth—Uniformitarianism and The Grand Canyon

The Grand Canyon from above
The Grand Canyon. This is not the deepest or the steepest canyon on Earth, but it is huge and spectacular.
Credit: R.B. Alley
Map of the U.S. with Grand Canyon National Park highlighted in north-central Arizona.
Grand Canyon National Park Location
Credit: R.B. Alley

The Grand Canyon is a mile-deep, 18-mile wide, 277-mile long (1.6 km x 29 km x 446 km) gash in the Earth. The colorful spires, the rocky cliffs, the hidden pocket canyons, the pristine springs laying down lovely deposits, the roaring thunderstorms and arching rainbows are to many the quintessence of the U.S. West. The Grand Canyon is neither the deepest nor the steepest canyon of the planet, but the Grand Canyon indeed is grand, and defines “canyon” for many people.

When the author, his cousin Chuck, and his sister, Sharon, were hiking the Bright Angel Trail from the North Rim into the canyon, a snake crossed the trail and slithered into some dry grass just at the trail edge. Chuck and I, in the lead, could see quite clearly that this snake ended in a “harmless” tail. Sharon, just behind, was not aware of the snake until it stuck its head out and rattled the grass just at her feet. Deciding that discretion was the better part of valor, and that if it rattles like a rattler it might actually be one, she made one mighty leap backward, landing in a cloud of dust on a switchback below.

Sharon almost certainly was not concerned with the rocks about her at that instant, but she had leaped backward through history. And what a history it is.

The Longest Story

At the bottom, the inner canyon is cut through the Precambrian Vishnu and Brahma Schists. The older Vishnu has the appearance and chemical composition of metamorphosed sediments. The lava flows of the Brahma preserve the pillow structure of submarine eruptions, but the interbedded volcanic airfall material shows that at times the region was exposed as dry land. The total thickness of three miles of lava flows and interbedded layers, now standing almost on end although they initially were deposited almost horizontally, speaks of an important, long-lasting interval of deposition.

These oldest lava flows and sediments of the Grand Canyon have been "cooked," and are now of metamorphic types that form only in the hearts of mountain ranges at very high pressures and temperatures. During and after the metamorphism, melted rock (magma) squirted into cracks in these rocks, and then froze to form the pretty pink Zoroaster Granite. Yet this whole package of rocks was then brought back to the surface as the rocks of the mountains above them were eroded, with the erosion producing a very smooth, nearly horizontal plain on top of them, and weathering/soil formation causing changes that extend deep into these rocks.

The sea next advanced across this plain, first picking up and carrying and rolling pieces of the rocks and soils on the erosion surface to form a conglomerate, then giving way to sandstones, shales, and limestones that piled up to a thickness of two miles or so. (Such a great thickness does not mean that the sea was two miles deep; rather, in this case, the water stayed relatively shallow, but the warping of the crust by the drifting plates and other processes caused the sea floor to sink as the muds and other deposits piled up.) These rocks include mud cracks, ripple marks, casts of salt crystals that formed when the sea water evaporated in nearshore environments, and stromatolites, which are algal-mat deposits in which the algae trap mud, grow up through it, and trap more mud. All of these are similar to modern features, and indicate gradual accumulation (a layer, then drying for mud cracks, then more mud, then ripples from water flow, then drying for salt casts, and on and on and on).

Death-Valley-type pull-apart faulting then dropped and rotated these layers, so that they now slant (see the figure below). Long-term weathering and erosion then occurred, leading to a plain broken by a few higher points where especially resistant rocks did not erode as rapidly. Again, deep weathering speaks of long exposure. In some places, the sediments were entirely removed down to the metamorphic rocks beneath, but the sediments are preserved where they were dropped by faulting.

The sea returned, again reworking materials on the erosion surface to make a basal conglomerate, followed by beach sandstone, then offshore shale, and limestone from farther offshore. As the sea deepened and the beach moved towards the land, shale was deposited on sandstone and then limestone on shale. These three layers form the Tonto Plateau that is so evident on the south side of the canyon. The rocks of the Tonto Plateau include fossils of marine animals such as trilobites, and even trilobite tracks. Again, all evidence is of deposition by processes just like those operating today, over long periods of time. A layer with a trilobite track must have been exposed long enough for a trilobite to crawl across it. The thousands and thousands of different layers in the rocks, with ripples and tracks and fossils, indicate long times.

Time then passed of which we have no record in the Grand Canyon, except that stream channels were carved on top of the limestone, indicating that the region was raised out of the sea and erosion was occurring. Fossils from two of the periods of the Paleozoic are missing, indicating that much time passed. When deposition resumed, the first rocks put down were limestones in the stream valleys, but another time gap sits on top of those in-the-channel rocks. The limestones in the channels include coral and shellfish (brachiopod) fossils, and plates from armored fish.

Diagram showing the rock units exposed in the walls of the Grand Canyon, and the erosional surfaces (called unconformities) that separate some of the rock units.
Diagram showing the rock units exposed in the walls of the Grand Canyon, and the erosional surfaces (called unconformities) that separate some of the rock units.
Credit: R.B. Alley

The marine Redwall Limestone was deposited next, so-named because it makes a red wall. The limestone is gray, with the red (rust and clay) dripping down from red rocks above. The Redwall Limestone contains fossils of corals, sea lilies (crinoids) and shellfish (brachiopods), but with notable differences from the fossils of those general types found in limestones below, and both sets of fossils differ from those in limestones above. The Redwall Limestone contains caves and sinkholes, which in turn contain sediments associated with the rocks above. Caves generally form on land or possibly very close to land under shallow water, not beneath the open ocean, so the rocks were lifted near or above sea level and eroded after Redwall deposition.

Then, the sea flooded in, at least in the region that would become the western part of the Canyon, and deposited the Surprise Canyon limestone in erosional stream channels in the top of the Redwall. These rocks were not even described until the 1980s, and are only reachable by helicopter or arduous climbing. These rocks are not indicated in the diagram, above, which is what you would see on the Bright Angel Trail in the central Grand Canyon. Erosion cut the top of the Surprise Canyon before deposition of more layers.

Next are sandstones, siltstones and shales with plant fossils, footprints, etc., at various levels through the rocks, indicating deposition on land in floodplain conditions. Insect fossils appear on the upward trip through the rocks, and then great dunes with sand-blasted, wind-frosted grains and occasional lizard footprints. You might imagine the sand dunes of the Sahara spreading across the flood plain of the Nile for these rocks. Marine conditions then returned, providing mostly limestones with sponge fossils and shark’s teeth as well as corals, crinoids and brachiopods, finally reaching the top of the canyon.

If you're on the North Rim of the Canyon, gaze farther north. The rocks you're on slant downward to the north, and you are looking at rows of cliffs with younger rocks, up through the cliffs of Zion from the age of the dinosaurs, up through the lakes of Bryce, up and up and up until finally you reach the trees and Native American sites older than the historical chronologies of Archbishop Ussher.

(By the way, if you’re interested in the carving of the Grand Canyon, have a look in the Unit 10 Enrichment.)

How Old?

A pile of rocks like those in the Grand Canyon does not reveal its age easily. But we have evidence of seas, mountain building, mountain erosion, more seas, more distortion, more erosion, and more, and more, and more. The rocks involved are old friends—similar things are forming today. Using the principle of uniformitarianism—the present is the key to the past—we can make some estimate as to how long events take. The schists at the bottom were buried miles deep in mountain ranges and later brought to the surface by erosion, and fast erosion rates require a million years to strip off a mile, for example.

The geologists of the 1700s, working primarily in Europe, pieced together stories such as this. They tried to estimate the times involved. One difficulty was that they could not tell how much time was in the erosional time gaps, or unconformities—was erosion fast, or slow? And they could not really unravel all of the stories in the oldest rocks because metamorphism had erased some of the stories.

These early geologists eventually estimated that the rocks told of events that required AT LEAST tens of millions of years to hundreds of millions of years. Just depositing the sedimentary rocks would take about that long, with much more time represented by the unconformities and the oldest really-messed-up rocks. This is deep time—the Earth is not the historical thousands of years, or even the ice-layer/tree-ring tens of thousands of years. History was written and trees grew on the relics of vastly greater histories. Looking into that history is one of the great joys of geology, but it brings us to the edge of a mental cliff from which some people do not wish to look. In the next section, we will see just how high that cliff really is.

Radioactive Clocks

The techniques of layer counting and uniformitarianism are useful in dating, but the real workhorse these days is radiometric or radioactive dating. The Earth is host to a wide range of naturally occurring radioactive elements. An atom of a radioactive element eventually will spontaneously change to some other type of atom, by emitting radioactive energy.

Radioactive decay occurs in various ways. The easiest to understand is when a nucleus splits into two parts, kicking out a part of itself. Remember that heat causes molecules in water to bounce around and occasionally evaporate; atoms or molecules in rocks are also bouncing around, but are so tightly bound that very, very few break free at the Earth’s surface. In somewhat the same, way, the protons and neutrons in the nucleus of an atom are always wiggling and bouncing around; most nuclei are so tightly bound that this wiggling doesn’t change anything, but some types of nuclei are weakly enough bound that occasionally some protons and neutrons “evaporate.” We call those types of atoms that “evaporate” radioactive, and those that do not stable. (A real nuclear physicist would probably yell at us because we oversimplified a bit too much, but this should do for introductory geology.)

Commonly, a nucleus that “evaporates” emits a group of two protons and two neutrons, which is the nucleus of a helium atom and also is called an alpha particle, for historical reasons. Emitting an alpha particle changes the remainder of the nucleus to the type of atom that is two protons lighter. Other types of radioactive changes also occur, including splitting of a nucleus into nearly equal-sized chunks, change of a neutron to a proton plus an electron that is emitted, or capture of an electron by a proton to change into a neutron. All of these change the type of atom from one element to another. All are explainable by well-known physical principles, and all are as natural and regular as the downward fall of your pencil if you drop it off your desk.

The behavior of any one atom is not predictable, but in large groups the average behavior is easily predictable. The basic rule is that, if you watch for some specified time such as one hour, the more radioactive atoms you start with, the more radioactive atoms you will see change. If you start your stopwatch when you have some number of a given type of radioactive atom, and stop the watch when half have changed, you will have estimated the half-life of the radioactive type. Each radioactive isotope has a distinctive half-life, which can be measured in the laboratory. (Note that you do not need to wait for an entire half-life to measure it; the Enrichment section shows mathematically that you need to wait only long enough for enough atoms to change to be measured accurately—the rule that more change in some time if you start with more is all you need.)

Suppose you start with 2000 atoms of the parent type. These decay into offspring (most textbooks refer to these offspring as daughters). After one half-life, 1000 parent atoms remain and 1000 offspring have been produced. After another half-life, half of those 1000 parent atoms have changed to offspring, leaving 500 parents and giving 1000+500=1500 offspring. After a third half-life, half of the remaining parents have changed, so that now only 250 parents remain and 1500+250=1750 offspring have been produced. During the fourth half-life, half of the remaining parents decay, leaving only 125 parents and giving 1750+125=1875 offspring.

Now, we really need to deal with large numbers, so add ten zeros to the end of each of the numbers in the previous paragraph. Such numbers of radioactive atoms are common in even relatively small samples of rock; the total number of atoms in a fist-sized chunk of rock is about 1 followed by 24 zeros.

As noted, there are many different parent types with different half-lives. Some half-lives are very short—seconds or less. Others are very long—billions of years or more. Some of the radioactive parents are left over from the explosions of stars that produced the stuff of which the Earth is made. Other radioactive parents are created by cosmic rays that strike atoms on Earth. Some radioactive decays produce offspring that are themselves radioactive parents for a further generation, and several such decays may be required to produce a stable offspring. And radioactive decays may damage neighboring atoms, producing radioactive types.

Telling Time

Consider the example of potassium-40 and argon-40. Argon-40 has 18 protons and 22 neutrons in its nucleus, for a total of 40 particles. Potassium-40 has 19 protons and 21 neutrons, also totaling 40. Potassium-40 is a parent with a half-life of 1.3 billion years. Potassium is abundant on Earth, and occurs in many common minerals, and some of the potassium is the radioactive parent potassium-40. The offspring, argon-40, is a gas. If lava flows out on the surface of the Earth, the argon escapes. Thus, a lava flow will start with some parent potassium-40 but no offspring argon-40. As time passes, the potassium-40 breaks down to argon-40, which builds up in the rock. If today the rock has as many potassium-40 as argon-40 atoms, then one half-life has passed and the rock is 1.3 billion years old. Whatever the ratio is, the math is not that difficult and gives the age.

It is possible, of course, for argon-40 to leak out of the mineral. If it does, we will think that the rock is younger than it really is. But if leakage is occurring from a mineral grain, then the outside of the grain will contain less argon-40 than the inside does, and this can be measured, revealing the problem. A mineral grain that grew in slowly cooling melted rock far down in the Earth and that then was erupted may have begun trapping argon-40 before the eruption occurred, in which case the age obtained will be the time when the grain started growing rather than the time when the eruption occurred. Scientists do not blindly apply dating techniques; they think about what is being measured, and apply a little common sense.

But isn’t it possible that the radioactive “clocks” ran at a different rate in the past? Some of the most free-thinking physicists have suggested slight changes in physical “constants” over time; couldn’t that affect the clocks? A critical difficulty with this seemingly “easy” idea is that the forces controlling the stability of atomic nuclei (hence the rate of radioactive decay) are the forces involved in all sorts of other processes including energy generation in the sun and other stars, so if you tweak the “clock” very much, you have turned off or blown up the sun, and we know that has not happened.

There is an easier argument against changing decay rates. The techniques of radiometric dating have been tested against layer-counting ages and historically documented ages of events, and found to agree beautifully. Radiometric dating techniques also have been tested against uniformitarian results, and found to be fully consistent. The techniques have been tested against each other—one sample can be dated using several different parent-offspring pairs—and good agreement is found. (Short-lived parent-offspring pairs are used to date young things—in old rocks, all of the parents are gone—and long-half-life pairs are used to date old things—in young things, the number of offspring is too small to allow accurate measurement—but enough different types exist that intercomparisons are still possible.)

You won’t have to look very far on the web to find sites—usually attached to fundamentalist Christian ideas—complaining about errors in radiometric dating. (And Dr. Alley was once shown a published tract pointing out how stupid Dr. Alley himself must be to think that he could count more annual layers in an ice core than the total age of the Earth!) Some of the objections to radiometric dating are fairly silly, and even some of the young-Earth sites have put up notes asking followers to avoid using certain common arguments against scientists because those arguments are just wrong. The 5000-year-old living clam falls in this category. (See the Enrichment if you want to get the low-down on “old” living clams.) The bottom line is that radiometric dating is useful, practical, successful, matches written records as far back as they go, matches other indications beyond that, and reveals a deep and fascinating history. Radiometric dating is not perfect, it does include errors, practitioners have to know what they’re doing and think about it, but it works.

The oldest rocks found on Earth are about 4 billion years old, and some of those contain mineral grains recycled from even older rocks. The active Earth has almost certainly erased the record of its very earliest rocks. Meteorites probably formed from the solar nebula at about the same time as the Earth did, and since have fallen on the Earth. The oldest meteorites are about 4.6 billion years old, and that is our best estimate for the age of the Earth. Careful analyses of the changing lead isotopic ratios over time (from decay of uranium) also yield that number for the Earth. And 4.6 billion years is, indeed, is deep time.