In its broadest sense, geology is the study of Earth — its interior and its exterior surface, the rocks and other materials that are around us, the processes that have resulted in the formation of those materials, the water that flows over the surface and lies underground, the changes that have taken place over the vastness of geological time, and the changes that we can anticipate will take place in the near future. Geology is a science, meaning that we use deductive reasoning and scientific methods (see Box 1.1) to understand geological problems. It is, arguably, the most integrated of all of the sciences because it involves the understanding and application of all of the other sciences: physics, chemistry, biology, mathematics, astronomy, and others. But unlike most of the other sciences, geology has an extra dimension, that of time — deep time — billions of years of it. Geologists study the evidence that they see around them, but in most cases, they are observing the results of processes that happened thousands, millions, and even billions of years in the past. Those were processes that took place at incredibly slow rates — millimetres per year to centimetres per year — but because of the amount of time available, they produced massive results.

Geology is displayed on a grand scale in mountainous regions, perhaps nowhere better than the Rocky Mountains in Canada (Figure 1.1). The peak on the right is Rearguard Mountain, which is a few kilometres northeast of Mount Robson, the tallest peak in the Canadian Rockies (3,954 m). The large glacier in the middle of the photo is the Robson Glacier. The river flowing from Robson Glacier drains into Berg Lake in the bottom right. There are many geological features portrayed here. The sedimentary rock that these mountains are made of formed in ocean water over 500 million years ago. A few hundred million years later, these beds were pushed east for tens to hundreds of kilometres by tectonic plate convergence and also pushed up to thousands of metres above sea level. Over the past two million years this area — like most of the rest of Canada — has been repeatedly glaciated, and the erosional effects of those glaciations are obvious. The Robson Glacier is now only a small remnant of its size during the Little Ice Age of the 15th to 18th centuries, as shown by the distinctive line on the slope on the left. Like almost all other glaciers in the world, it is now receding even more rapidly because of human-caused climate change.

Figure 1.1  Rearguard Mountain and Robson Glacier in the Rocky Mountains of British Columbia [SE]

Geology is also about understanding the evolution of life on Earth; about discovering resources such as metals and energy; about recognizing and minimizing the environmental implications of our use of those resources; and about learning how to mitigate the hazards related to earthquakes, volcanic eruptions, and slope failures. All of these aspects of geology, and many more, are covered in this textbook.

Box 1.1 What are scientific methods?

There is no single method of inquiry that is specifically the “scientific method”; furthermore, scientific inquiry is not necessarily different from serious research in other disciplines. The key feature of serious inquiry is the creation of a hypothesis (a tentative explanation) that could explain the observations that have been made, and then the formulation and testing (by experimentation) of one or more predictions that follow from that hypothesis. For example, we might observe that most of the cobbles in a stream bed are well rounded (see photo in this box), and then derive the hypothesis that the rocks become rounded during transportation along the stream bed. A prediction that follows from this hypothesis is that cobbles present in a stream will become increasingly rounded over time as they are transported downstream. An experiment to test this prediction would be to place some angular cobbles in a stream, label them so that we can be sure to find them again later, and then return at various time intervals (over a period of months or years) to carefully measure their locations and roundness. A critical feature of a good hypothesis and any resulting predictions is that they must be testable.  For example, an alternative hypothesis to the one above is that an extraterrestrial organization creates rounded cobbles and places them in streams when nobody is looking. This may indeed be the case, but there is no practical way to test this hypothesis. Most importantly, there is no way to prove that it is false, because if we aren’t able to catch the aliens at work, we still won’t know if they did it!

The simple answer to this question is that Earth is our home — our only home for the foreseeable future — and in order to ensure that it continues to be a great place to live, we need to understand how it works. Another answer is that some of us can’t help but study it because it’s fascinating. But there is more to it than that:

• We rely on Earth for valuable resources such as soil, water, metals, industrial minerals, and energy, and we need to know how to find these resources and exploit them sustainably.
• We can study rocks and the fossils they contain to understand the evolution of our environment and the life within it.
• We can learn to minimize our risks from earthquakes, volcanoes, slope failures, and damaging storms.
• We can learn how and why Earth’s climate has changed in the past, and use that knowledge to understand both natural and human-caused climate change.
• We can recognize how our activities have altered the environment in many ways and the climate in increasingly serious ways, and how to avoid more severe changes in the future.
• We can use our knowledge of Earth to understand other planets in our solar system, as well as those around distant stars.

An example of the importance of geological studies for minimizing risks to the public is illustrated in Figure 1.2. This is a slope failure that took place in January 2005 in the Riverside Drive area of North Vancouver. The steep bank beneath the house shown gave way, and a slurry of mud and sand flowed down, destroying another house below and killing one person. This event took place following a heavy rainfall, which is a common occurrence in southwestern B.C. in the winter.

Figure 1.2  The aftermath of a deadly debris flow in the Riverside Drive area of North Vancouver in January, 2005 [The Province, used with permission]

The irony of the 2005 slope failure is that the District of North Vancouver had been warned in a geological report written in 1980 that this area was prone to slope failure and that steps should be taken to minimize the risk to residents. Very little was done in the intervening 25 years, and the results were deadly.

Geologists are involved in a range of widely varying occupations with one thing in common: the privilege of studying this fascinating planet. In Canada, many geologists work in the resource industries, including mineral exploration and mining and energy exploration and extraction. Other major areas where geologists work include hazard assessment and mitigation (e.g., assessment of risks from slope failures, earthquakes, and volcanic eruptions); water supply planning, development, and management; waste management; and assessment of geological issues on construction projects such as highways, tunnels, and bridges. Most geologists are employed in the private sector, but many work for government-funded geological organizations, such as the Geological Survey of Canada or one of the provincial geological surveys. And of course, many geologists are involved in education at the secondary and the postsecondary levels.

Some people are attracted to geology because they like to be outdoors, and it is true that many geological opportunities involve fieldwork in places that are as amazing to see as they are interesting to study. But a lot of geological work is also done in offices or laboratories. Geological work tends to be varied and challenging, and for these reasons and many others, geologists are among those who are the most satisfied with their employment.

Figure 1.3 Geologists examining ash-layer deposits at Kilauea Volcano, Hawaii [SE photo]

In Canada, most working geologists are required to be registered with an association of professional geoscientists. This typically involves meeting specific postsecondary educational standards and gaining several years of relevant professional experience under the supervision of a registered geoscientist. Information about the Association of Professional Engineers and Geoscientists of British Columbia can be found at: https://www.apeg.bc.ca/Home.

The rest of this chapter is devoted to a brief overview of a few of the important aspects of physical geology, starting with minerals and rocks. This is followed by a review of Earth’s internal structure and the processes of plate tectonics, and an explanation of geological time.

Like everything else in the universe, Earth is made up of varying proportions of the 90 naturally occurring elements — hydrogen, carbon, oxygen, magnesium, silicon, iron, and so on. In most geological materials, these combine in various ways to make minerals. Minerals will be covered in some detail in Chapter 2, but here we will briefly touch on what minerals are, and how they are related to rocks.

A mineral is a naturally occurring combination of specific elements that are arranged in a particular repeating three-dimensional structure or lattice.Terms in bold are defined in the glossary at the end of the book. The mineral halite is shown as an example in Figure 1.4. In this case, atoms of sodium (Na: purple) alternate with atoms of chlorine (Cl: green) in all three dimensions, and the angles between the bonds are all 90°. Even in a tiny crystal, like the ones in your salt shaker, the lattices extend in all three directions for thousands of repetitions. Halite always has this composition and this structure.

Figure 1.4 The lattice structure and composition of the mineral halite (common table salt) [SE]

There are thousands of minerals, although only a few dozen are mentioned in this book. In nature, minerals are found in rocks, and the vast majority of rocks are composed of at least a few different minerals. A close-up view of granite, a common rock, is shown in Figure 1.5. Although a hand-sized piece of granite may have thousands of individual mineral crystals in it, there are typically only a few different minerals, as shown here.

Figure 1.5 A close-up view of the rock granite and some of the minerals that it typically contains (H = hornblende (amphibole), Q = quartz and F = feldspar). The crystals range from about 0.1 to 3 mm in diameter. Most are irregular in outline, but some are rectangular. [SE]

Rocks can form in a variety of ways. Igneous rocks form from magma (molten rock) that has either cooled slowly underground (e.g., to produce granite) or cooled quickly at the surface after a volcanic eruption (e.g., basalt). Sedimentary rocks, such as sandstone, form when the weathered products of other rocks accumulate at the surface and are then buried by other sediments. Metamorphic rocks form when either igneous or sedimentary rocks are heated and squeezed to the point where some of their minerals are unstable and new minerals form to create a different type of rock. An example is schist.

A key point to remember is the difference between a mineral and a rock. A mineral is a pure substance with a specific composition and structure, while a rock is typically a mixture of several different minerals (although a few types of rock may include only one type of mineral).  Examples of minerals are feldspar, quartz, mica, halite, calcite, and amphibole. Examples of rocks are granite, basalt, sandstone, limestone, and schist.

Plate tectonics is the model or theory that has been used for the past 60 years to understand Earth’s development and structure — more specifically the origins of continents and oceans, of folded rocks and mountain ranges, of earthquakes and volcanoes, and of continental drift. It is explained in some detail in Chapter 10, but is introduced here because it includes concepts that are important to many of the topics covered in the next few chapters.

Key to understanding plate tectonics is an understanding of Earth’s internal structure, which is illustrated in Figure 1.6. Earth’s core consists mostly of iron. The outer core is hot enough for the iron to be liquid. The inner core, although even hotter, is under so much pressure that it is solid. The mantle is made up of iron and magnesium silicate minerals. The bulk of the mantle, surrounding the outer core, is solid rock, but is plastic enough to be able to flow slowly. Surrounding that part of the mantle is a partially molten layer (the asthenosphere), and the outermost part of the mantle is rigid. The crust — composed mostly of granite on the continents and mostly of basalt beneath the oceans — is also rigid. The crust and outermost rigid mantle together make up the lithosphere. The lithosphere is divided into about 20 tectonic plates that move in different directions on Earth’s surface. (For a more accurate depiction of the components of the Earth’s interior, please see Figure 9.2.)

An important property of Earth (and other planets) is that the temperature increases with depth, from close to 0°C at the surface to about 7000°C at the centre of the core. In the crust, the rate of temperature increase is about 30°C/km. This is known as the geothermal gradient.

Figure 1.6  The structure of Earth’s interior showing the inner and outer core, the different layers of the mantle, and the crust [Wikipedia]

Heat is continuously flowing outward from Earth’s interior, and the transfer of heat from the core to the mantle causes convection in the mantle (Figure 1.7). This convection is the primary driving force for the movement of tectonic plates. At places where convection currents in the mantle are moving upward, new lithosphere forms (at ocean ridges), and the plates move apart (diverge). Where two plates are converging (and the convective flow is downward), one plate will be subducted (pushed down) into the mantle beneath the other. Many of Earth’s major earthquakes and volcanoes are associated with convergent boundaries.

Figure 1.7  A model of convection within Earth’s mantle [http://upload.wikimedia.org/wikipedia/commons/thumb/2/27/Oceanic_spreading.svg/1280px-Oceanic_spreading.svg.png]

Earth’s major tectonic plates and the directions and rates at which they are diverging at sea-floor ridges, are shown in Figure 1.8.

Exercises

Exercise 1.2 Plate Motion During Your Lifetime

Using either a map of the tectonic plates from the Internet or Figure 1.8, determine which tectonic plate you are on right now, approximately how fast it is moving, and in what direction. How far has that plate moved relative to Earth’s core since you were born?

Figure 1.8 Earth’s tectonic plates and tectonic features that have been active over the past 1 million years [http://commons.wikimedia.org/wiki/File:Plate_tectonics_map.gif]

In 1788, after many years of geological study, James Hutton, one of the great pioneers of geology, wrote the following about the age of Earth: The result, therefore, of our present enquiry is, that we find no vestige of a beginning — no prospect of an end.Hutton, J, 1788. Theory of the Earth; or an investigation of the laws observable in the composition, dissolution, and restoration of land upon the Globe. Transactions of the Royal Society of Edinburgh. Of course he wasn’t exactly correct, there was a beginning and there will be an end to Earth, but what he was trying to express is that geological time is so vast that we humans, who typically live for less than a century, have no means of appreciating how much geological time there is. Hutton didn’t even try to assign an age to Earth, but we now know that it is approximately 4,570 million years old. Using the scientific notation for geological time, that is 4,570 Ma (for mega annum or “millions of years”) or 4.57 Ga (for giga annum or billions of years). More recent dates can be expressed in ka (kilo annum); for example, the last cycle of glaciation ended at approximately 11.7 ka or 11,700 years ago. This notation will be used for geological dates throughout this book.

Unfortunately, knowing how to express geological time doesn’t really help us understand or appreciate its extent. A version of the geological time scale is included as Figure 1.9. Unlike time scales you’ll see in other places, or even later in this book, this time scale is linear throughout its length, meaning that 50 Ma during the Cenozoic is the same thickness as 50 Ma during the Hadean—in each case about the height of the “M” in Ma. The Pleistocene glacial epoch began at about 2.6 Ma, which is equivalent to half the thickness of the thin grey line at the top of the yellow bar marked “Cenozoic.” Most other time scales have earlier parts of Earth’s history compressed so that more detail can be shown for the more recent parts. That makes it difficult to appreciate the extent of geological time.

Figure 1.9 The geological time scale [SE]

To create some context, the Phanerozoic Eon (the last 542 million years) is named for the time during which visible (phaneros) life (zoi) is present in the geological record. In fact, large organisms — those that leave fossils visible to the naked eye — have existed for a little longer than that, first appearing around 600 Ma, or a span of just over 13% of geological time. Animals have been on land for 360 million years, or 8% of geological time. Mammals have dominated since the demise of the dinosaurs around 65 Ma, or 1.5% of geological time, and the genus Homo has existed since approximately 2.2 Ma, or 0.05% (1/2,000th) of geological time.

Geologists (and geology students) need to understand geological time. That doesn’t mean simply memorizing the geological time scale; instead, it means getting your mind around the concept that although most geological processes are extremely slow, very large and important things can happen if such processes continue for enough time.

For example, the Atlantic Ocean between Nova Scotia and northwestern Africa has been getting wider at a rate of about 2.5 cm per year. Imagine yourself taking a journey at that rate — it would be impossibly and ridiculously slow. And yet, since it started to form around 200 Ma (just 4% of geological time), the Atlantic Ocean has grown to a width of over 5,000 km!

A useful mechanism for understanding geological time is to scale it all down into one year. The origin of the solar system and Earth at 4.57 Ga would be represented by January 1, and the present year would be represented by the last tiny fraction of a second on New Year’s Eve. At this scale, each day of the year represents 12.5 million years; each hour represents about 500,000 years; each minute represents 8,694 years; and each second represents 145 years. Some significant events in Earth’s history, as expressed on this time scale, are summarized on Table 1.1.

Table 1.1  A summary of some important geological dates expressed as if all of geological time was condensed into one year [SE]

The topics covered in this chapter can be summarized as follows:

Introduction

Minerals are all around us: the graphite in your pencil, the salt on your table, the plaster on your walls, and the trace amounts of gold in your computer. Minerals can be found in a wide variety of consumer products including paper, medicine, processed foods, cosmetics, and many more. And of course, everything made of metal is also derived from minerals.

As defined in Chapter 1, a mineral is a naturally occurring combination of specific elements arranged in a particular repeating three-dimensional structure (Figure 2.1).

“Naturally occurring” implies that minerals are not artificially made, although many naturally occurring minerals (e.g., diamond) are also made in laboratories. That doesn’t disqualify them from being minerals.

“Specific elements” means that most minerals have a specific chemical formula or composition. The mineral pyrite, for example, is FeS₂ (two atoms of sulphur for each atom of iron), and any significant departure from that would make it a different mineral. But many minerals have variable compositions within a specific range. The mineral olivine, for example, can range all the way from Fe₂SiO₄ to Mg₂SiO₄. Intervening compositions are written as (Fe,Mg)₂SiO₄ meaning that Fe and Mg can be present in any proportion. This type of substitution is known as solid solution.

Most important of all, a mineral has a specific “repeating three-dimensional structure” or “lattice,” which is the way in which the atoms are arranged. We’ve already seen in Chapter 1 how sodium and chlorine atoms in halite alternate in a regular pattern. That happens to be about the simplest mineral lattice of all; most mineral lattices are much more complicated, as we’ll see.

Figure 2.1 Crystals of native sulphur at an outlet of volcanic gases, Kilauea volcano, Hawaii

All matter, including mineral crystals, is made up of atoms, and all atoms are made up of three main particles: protons, neutrons, and electrons. As summarized in Table 2.1, protons are positively charged, neutrons are uncharged and electrons are negatively charged. The negative charge of one electron balances the positive charge of one proton. Both protons and neutrons have a mass of 1, while electrons have almost no mass.

Table 2.1 Charges and masses of the particles within atoms

The element hydrogen has the simplest atoms, each with just one proton and one electron. The proton forms the nucleus, while the electron orbits around it. All other elements have neutrons as well as protons in their nucleus, such as helium, which is depicted in Figure 2.2. The positively charged protons tend to repel each other, and the neutrons help to hold the nucleus together. The number of protons is the atomic number, and the number of protons plus neutrons is the atomic mass. For hydrogen, the atomic number is 1 because there is one proton and no neutrons. For helium, it is 4: two protons and two neutrons.

For most of the 16 lightest elements (up to oxygen) the number of neutrons is equal to the number of protons. For most of the remaining elements, there are more neutrons than protons, because extra neutrons are needed to keep the nucleus together by overcoming the mutual repulsion of the increasing numbers of protons concentrated in a very small space. For example, silicon has 14 protons and 14 neutrons. Its atomic number is 14 and its atomic mass is 28. The most common isotope of uranium has 92 protons and 146 neutrons. Its atomic number is 92 and its atomic mass is 238 (92 + 146).

Figure 2.2 A depiction of a helium atom.

The dot in the middle is the nucleus, and the surrounding cloud represents where the two electrons might be at any time. The darker the shade, the more likely that an electron will be there. An angstrom (Å) is 10^-10m . A femtometre (fm) is 10^-15m. In other words, a helium atom’s electron cloud is about 100,000 times bigger than its nucleus.

Electrons orbiting around the nucleus of an atom are arranged in shells — also known as “energy levels.” The first shell can hold only two electrons, while the next shell holds up to eight electrons. Subsequent shells can hold more electrons, but the outermost shell of any atom holds no more than eight electrons. The electrons in the outermost shell play an important role in bonding between atoms. Elements that have a full outer shell are inert in that they do not react with other elements to form compounds. They all appear in the far-right column of the periodic table: helium, neon, argon, etc. For elements that do not have a full outer shell, the outermost electrons can interact with the outermost electrons of nearby atoms to create chemical bonds. The electron shell configurations for 29 of the first 36 elements are listed in Table 2.2.

Table 2.2 Electron shell configurations of some of the elements up to element 36. (The inert elements, with filled outer shells, are bolded.)

Attributions

Figure 2.2
Helium Atom by Yzmo is under CC-BY-SA-3.0

As we’ve just seen, an atom seeks to have a full outer shell (i.e., eight electrons for most elements, or two electrons for hydrogen and helium) to be atomically stable. This is accomplished by transferring or sharing electrons with other atoms. Elements that already have their outer orbits filled are considered to be inert; they do not readily take part in chemical reactions.

Sodium has 11 electrons: two in the first shell, eight in the second, and one in the third (Figure 2.3). Sodium readily gives up the third shell electron; when it loses this one negative charge, it becomes positively charged. By giving up its lone third shell electron, sodium ends up with a full outer second shell. Chlorine, on the other hand, has 17 electrons: two in the first shell, eight in the second, and seven in the third. Chlorine readily accepts an eighth electron to fill its third shell, and therefore becomes negatively charged because of an imbalance between the number of protons (17) and electrons (18). In changing their number of electrons, these atoms become ions — the sodium loses an electron to become a positive ion or cation, and the chlorine gains an electron to become a negative ion or anion (Figure 2.3). Since negative and positive charges attract, sodium and chlorine ions stick together, creating an ionic bond. Electrons can be thought of as being transferred from one atom to another in an ionic bond. Common table salt (NaCl) is a mineral composed of chlorine and sodium linked together by ionic bonds (Figure 1.4). The mineral name for NaCl is halite.

Figure 2.3 A very simplified electron configuration of sodium and chlorine atoms (top). Sodium gives up an electron to become a cation (bottom left) and chlorine accepts an electron to become an anion (bottom right).

An element like chlorine can also form bonds without forming ions. For example, two chlorine atoms, which each seek an eighth electron in their outer shell, can share an electron in what is known as a covalent bond, to form chlorine gas (Cl₂) (Figure 2.4). Electrons are shared in a covalent bond.

Figure 2.4 Depiction of a covalent bond between two chlorine atoms. The electrons are black, in the left atom and blue in the right atom. Two electrons are shared (one black and one blue) so that each atom “appears” to have a full outer shell. [SE]

Exercises

Exercise 2.1 Cations, Anions, and Ionic Bonding

A number of elements are listed below along with their atomic numbers. Assuming that the first electron shell can hold two electrons and subsequent electron shells can hold eight electrons, sketch in the electron configurations for these elements. Predict whether the element is likely to form a cation (+) or an anion (–), and what charge it would have (e.g., +1, +2, –1). The first one is done for you.

Fluorine (9)    anion (-1)	Lithium (3) ________
Magnesium (12) ________	Argon (18) ________
Chlorine (17) ________	Beryllium (3) ________
Oxygen (8) ________	Sodium (11) ________

An uncharged carbon atom has six protons and six electrons; two of the electrons are in the inner shell and four in the outer shell (Figure 2.5). Carbon would need to gain or lose four electrons to have a filled outer shell, and this would create too great a charge imbalance for the ion to be stable. On the other hand, carbon can share electrons to create covalent bonds. In the mineral diamond, the carbon atoms are linked together in a three-dimensional framework, where one carbon atom is bonded to four other carbon atoms and every bond is a very strong covalent bond. In the mineral graphite, the carbon atoms are linked together in sheets or layers (Figure 2.5), and each carbon atom is covalently bonded to three others. Graphite-based compounds, which are strong because of the strong intra-layer covalent bonding, are used in high-end sports equipment such as ultralight racing bicycles. Graphite itself is soft because the bonding between these layers is relatively weak, and it is used in a variety of applications, including lubricants and pencils.

Figure 2.5 The electron configuration of carbon (above) and the sharing of electrons in covalent C bonding of diamond (right). The electrons shown in blue are shared between adjacent C atoms. Although shown here in only two dimensions, diamond has a three-dimensional structure as shown on Figure 2.7.

Silicon and oxygen bond together to create a silica tetrahedron, which is a four-sided pyramid shape with O at each corner and Si in the middle (Figure 2.6). This structure is the building block of the many important silicate minerals. The bonds in a silica tetrahedron have some of the properties of covalent bonds and some of the properties of ionic bonds. As a result of the ionic character, silicon becomes a cation (with a charge of +4) and oxygen becomes an anion (with a charge of –2). The net charge of a silica tetrahedron (SiO₄) is –4. As we will see later, silica tetrahedral (plural of tetrahedron) link together in a variety of ways to form most of the common minerals of the crust.

Figure 2.6 The silica tetrahedron, the building block of all silicate minerals (Because the silicon has a charge of +4 and the four oxygens each have a charge of -2, the silica tetrahedron has a net charge of -4.)

Most minerals are characterized by ionic bonds, covalent bonds, or a combination of the two, but there are other types of bonds that are important in minerals, including metallic bonds and weaker electrostatic forces (hydrogen or Van der Waals bonds). Metallic elements have outer electrons that are relatively loosely held. (The metals are highlighted on the periodic table in Appendix 1.) When bonds between such atoms are formed, these electrons can move freely from one atom to another. A metal can thus be thought of as an array of positively charged atomic nuclei immersed in a sea of mobile electrons. This feature accounts for two very important properties of metals: their electrical conductivity and their malleability (they can be deformed and shaped).

Molecules that are bonded ionically or covalently can also have other weaker electrostatic forces holding them together. Examples of this are the force holding graphite sheets together and the attraction between water molecules.

What’s with all of these “sili” names?

The element silicon is one of the most important geological elements and is the second-most abundant element in Earth’s crust (after oxygen). Silicon bonds readily with oxygen to form a silica tetrahedron (Figure 2.6). Pure silicon crystals (created in a lab) are used to make semiconductive media in electronic devices. A silicate mineral is one in which silicon and oxygen are present as silica tetrahedra. Silica also refers to a chemical component of a rock and is expressed as % SiO₂. The mineral quartz is made up entirely of silica tetrahedra, and some forms of quartz are known as silica. Silicone is a synthetic product (e.g., silicone rubber, resin, or caulking) made from silicon-oxygen chains and various organic molecules. To help you keep the “sili” names straight, here is a summary table:

As described in Chapter 1, all minerals are characterized by a specific three-dimensional pattern known as a lattice or crystal structure. These structures range from the simple cubic pattern of halite (NaCl) (Figure 1.4), to the very complex patterns of some silicate minerals. Two minerals may have the same composition, but very different crystal structures and properties. Graphite and diamond, for example, are both composed only of carbon, but while diamond is the hardest substance known, graphite is softer than paper. Their lattice structures are compared in Figure 2.7.

Figure 2.7 A depiction of the lattices of graphite and diamond. [Long Description]

Figure 2.8 Cubic crystals (left) and right-angle cleavage planes (right) of the mineral halite. If you look closely at the cleavage fragment in the middle, you can see where it would break again (cleave) along a plane parallel to the existing surface.

Mineral lattices have important implications for mineral properties, as exemplified by the relative hardnesses of diamond and graphite. Lattices also determine the shape that mineral crystals grow in and how they break. For example, the right angles in the lattice of the mineral halite (Figure 1.4) influence both the shape of its crystals (typically cubic), and the way those crystals break (Figure 2.8).

Attributions

Figure 2.8
Image on left: Halite by Rob Lavinsky, iRocks.com is used under a CC-BY-SA-3.0

Long Descriptions

Figure 2.7 long description: Graphite is a mixture of strong covalent bonds and weak inter-layer bonds. In diamonds, all bonds are strong covalent bonds. [Return to Figure 2.7].

Most minerals are made up of a cation (a positively charged ion) or several cations and an anion (a negatively charged ion (e.g., S^2–)) or an anion complex (e.g., SO₄^2–)_. For example, in the mineral hematite (Fe₂O₃), the cation is Fe₃⁺ (iron) and the anion is O^2– (oxygen). We group minerals into classes on the basis of their predominant anion or anion group. These include oxides, sulphides, carbonates, silicates, and others. Silicates are by far the predominant group in terms of their abundance within the crust and mantle. (They will be discussed in Section 2.4). Some examples of minerals from the different mineral groups are given in Table 2.3.

Oxide minerals have oxygen (O^2–) as their anion, but they exclude those with oxygen complexes such as carbonate (CO₃^2–), sulphate (SO₄^2–), and silicate (SiO₄^4–). The most important oxides are the iron oxides hematite and magnetite (Fe₂O₃ and Fe₃O₄, respectively). Both of these are important ores of iron. Corundum (Al₂O₃) is an abrasive, but can also be a gemstone in its ruby and sapphire varieties. If the oxygen is also combined with hydrogen to form the hydroxyl anion (OH^–) the mineral is known as a hydroxide. Some important hydroxides are limonite and bauxite, which are ores of iron and aluminium respectively. Frozen water (H₂O) is a mineral (an oxide), but liquid water is not because it doesn’t have a regular lattice.

Sulphides are minerals with the S^–2 anion, and they include galena (PbS), sphalerite (ZnS), chalcopyrite (CuFeS₂), and molybdenite (MoS₂), which are the most important ores of lead, zinc, copper, and molybdenum respectively. Some other sulphide minerals are pyrite (FeS₂), bornite (Cu₅FeS₄), stibnite (Sb₂S₃), and arsenopyrite (FeAsS).

Sulphates are minerals with the SO₄^–2 anion, and these include anhydrite (CaSO₄) and its cousin gypsum (CaSO₄.2H₂O) and the sulphates of barium and strontium: barite (BaSO₄) and celestite (SrSO₄). In all of these minerals, the cation has a +2 charge, which balances the –2 charge on the sulphate ion.

The halides are so named because the anions include the halogen elements chlorine, fluorine, bromine, etc. Examples are halite (NaCl), cryolite (Na₃AlF₆), and fluorite (CaF₂).

The carbonates include minerals in which the anion is the CO₃^–2 complex. The carbonate combines with +2 cations to form minerals such as calcite (CaCO₃), magnesite (MgCO₃), dolomite ((Ca,Mg)CO₃), and siderite (FeCO₃). The copper minerals malachite and azurite are also carbonates.

In phosphate minerals, the anion is the PO₄^–3 complex. An important phosphate mineral is apatite (Ca₅(PO₄)₃(OH)), which is what your teeth are made of.

The silicate minerals include the elements silicon and oxygen in varying proportions ranging from Si : O₂ to Si : O₄. These are discussed at length in Section 2.4.

Native minerals are single-element minerals, such as gold, copper, sulphur, and graphite.

The vast majority of the minerals that make up the rocks of Earth’s crust are silicate minerals. These include minerals such as quartz, feldspar, mica, amphibole, pyroxene, olivine, and a great variety of clay minerals. The building block of all of these minerals is the silica tetrahedron, a combination of four oxygen atoms and one silicon atom. These are arranged such that planes drawn through the oxygen atoms form a tetrahedron (Figure 2.6). Since the silicon ion has a charge of +4 and each of the four oxygen ions has a charge of –2, the silica tetrahedron has a net charge of –4.

In silicate minerals, these tetrahedra are arranged and linked together in a variety of ways, from single units to complex frameworks (Figure 2.9). The simplest silicate structure, that of the mineral olivine, is composed of isolated tetrahedra bonded to iron and/or magnesium ions. In olivine, the –4 charge of each silica tetrahedron is balanced by two divalent (i.e., +2) iron or magnesium cations. Olivine can be either Mg₂SiO₄ or Fe₂SiO₄, or some combination of the two (Mg,Fe)₂SiO₄. The divalent cations of magnesium and iron are quite close in radius (0.73 versus 0.62 angstromsAn angstrom is the unit commonly used for the expression of atomic-scale dimensions. One angstrom is 10–10 m or 0.0000000001 m. The symbol for an angstrom is Å.). Because of this size similarity, and because they are both divalent cations (both have a charge of +2), iron and magnesium can readily substitute for each other in olivine and in many other minerals.

Tetrahedron Configuration		Example Minerals
	Isolated (nesosilicates)	Olivine, garnet, zircon, kyanite
	Pairs (sorosilicates)	Epidote, zoisite
	Rings (cyclosilicates)	Tourmaline
	Single chains (inosilicates)	Pyroxenes, wollastonite
	Double chains (inosilicates)	Amphiboles
	Sheets (phyllosilicates)	Micas, clay minerals, serpentine, chlorite
3-dimensional structure	Framework (tectosilicates)	Feldspars, quartz, zeolite

Figure 2.9 Silicate mineral configurations. The triangles represent silica tetrahedra.

Exercises

Exercise 2.3 Make a Tetrahedron

Cut around the outside of the shape (solid lines and dotted lines), and then fold along the solid lines to form a tetrahedron.

If you have glue or tape, secure the tabs to the tetrahedron to hold it together. If you don’t have glue or tape, make a slice along the thin grey line and insert the pointed tab into the slit.

If you are doing this in a classroom, try joining your tetrahedron with others into pairs, rings, single and double chains, sheets, and even three-dimensional frameworks.

In olivine, unlike most other silicate minerals, the silica tetrahedra are not bonded to each other. They are, however, bonded to the iron and/or magnesium as shown on Figure 2.10.

Figure 2.10 A depiction of the structure of olivine as seen from above. The formula for this particular olivine, which has three Fe ions for each Mg ion, could be written: Mg0.5Fe1.5SiO4.

As already noted, the +2 ions of iron and magnesium are similar in size (although not quite the same). This allows them to substitute for each other in some silicate minerals. In fact, the common ions in silicate minerals have a wide range of sizes, as shown in Figure 2.11. All of the ions shown are cations, except for oxygen. Note that iron can exist as both a +2 ion (if it loses two electrons during ionization) or a +3 ion (if it loses three). Fe²⁺ is known as ferrous iron. Fe³⁺ is known as ferric iron. Ionic radii are critical to the composition of silicate minerals, so we’ll be referring to this diagram again.

Figure 2.11 The ionic radii (effective sizes) in angstroms, of some of the common ions in silicate minerals

The structure of the single-chain silicate pyroxene is shown on Figures 2.12 and 2.13. In pyroxene, silica tetrahedra are linked together in a single chain, where one oxygen ion from each tetrahedron is shared with the adjacent tetrahedron, hence there are fewer oxygens in the structure. The result is that the oxygen-to-silicon ratio is lower than in olivine (3:1 instead of 4:1), and the net charge per silicon atom is less (–2 instead of –4), since fewer cations are necessary to balance that charge. Pyroxene compositions are of the type MgSiO₃, FeSiO₃, and CaSiO₃, or some combination of these. Pyroxene can also be written as (Mg,Fe,Ca)SiO₃, where the elements in the brackets can be present in any proportion. In other words, pyroxene has one cation for each silica tetrahedron (e.g., MgSiO₃) while olivine has two (e.g., Mg₂SiO₄). Because each silicon ion is +4 and each oxygen ion is –2, the three oxygens (–6) and the one silicon (+4) give a net charge of –2 for the single chain of silica tetrahedra. In pyroxene, the one divalent cation (2+) per tetrahedron balances that –2 charge. In olivine, it takes two divalent cations to balance the –4 charge of an isolated tetrahedron.

The structure of pyroxene is more “permissive” than that of olivine — meaning that cations with a wider range of ionic radii can fit into it. That’s why pyroxenes can have iron (radius 0.63 Å) or magnesium (radius 0.72 Å) or calcium (radius 1.00 Å) cations.

Figure 2.12 A depiction of the structure of pyroxene. The tetrahedral chains continue to left and right and each is interspersed with a series of divalent cations. If these are Mg ions, then the formula is MgSiO3.

Figure 2.13 A single silica tetrahedron (left) with  four oxygen ions per silicon ion (SiO4). Part of a single chain of tetrahedra (right), where the oxygen atoms at the adjoining corners are shared between two tetrahedra (arrows). For a very long chain the resulting ratio of silicon to oxygen is 1 to 3 (SiO3).

In mica structures, the silica tetrahedra are arranged in continuous sheets, where each tetrahedron shares three oxygen anions with adjacent tetrahedra. There is even more sharing of oxygens between adjacent tetrahedra and hence fewer charge-balancing cations are needed for sheet silicate minerals. Bonding between sheets is relatively weak, and this accounts for the well-developed one-directional cleavage (Figure 2.14). Biotite mica can have iron and/or magnesium in it and that makes it a ferromagnesian silicate mineral (like olivine, pyroxene, and amphibole). Chlorite is another similar mineral that commonly includes magnesium. In muscovite mica, the only cations present are aluminum and potassium; hence it is a non-ferromagnesian silicate mineral.

Figure 2.14 Biotite mica (left) and muscovite mica (right). Both are sheet silicates and split easily into thin layers along planes parallel to the sheets. Biotite is dark like the other iron- and/or magnesium-bearing silicates (e.g., olivine, pyroxene, and amphibole), while muscovite is light coloured. (Each sample is about 3 cm across.)

Apart from muscovite, biotite, and chlorite, there are many other sheet silicates (or phyllosilicates), which usually exist as clay-sized fragments (i.e., less than 0.004 mm). These include the clay minerals kaolinite, illite, and smectite, and although they are difficult to study because of their very small size, they are extremely important components of rocks and especially of soils.

All of the sheet silicate minerals also have water in their structure.

Silica tetrahedra are bonded in three-dimensional frameworks in both the feldspars and quartz. These are non-ferromagnesian minerals — they don’t contain any iron or magnesium. In addition to silica tetrahedra, feldspars include the cations aluminum, potassium, sodium, and calcium in various combinations. Quartz contains only silica tetrahedra.

The three main feldspar minerals are potassium feldspar, (a.k.a. K-feldspar or K-spar) and two types of plagioclase feldspar: albite (sodium only) and anorthite (calcium only). As is the case for iron and magnesium in olivine, there is a continuous range of compositions (solid solution series) between albite and anorthite in plagioclase. This is because the calcium and sodium ions are almost identical in size (1.00 Å versus 0.99 Å). Any intermediate compositions between CaAl₂Si₃O₈ and NaAlSi₃O₈ can exist (Figure 2.15). This is a little bit surprising because, although they are very similar in size, calcium and sodium ions don’t have the same charge (Ca²⁺ versus Na+). This problem is accounted for by corresponding substitution of Al³⁺ for Si⁴⁺. Therefore, albite is NaAlSi₃O₈ (one Al and three Si) while anorthite is CaAl₂Si₂O₈ (two Al and two Si), and plagioclase feldspars of intermediate composition have intermediate proportions of Al and Si. This is called a “coupled-substitution.”

The intermediate-composition plagioclase feldspars are oligoclase (10% to 30% Ca), andesine (30% to 50% Ca), labradorite (50% to 70% Ca), and bytownite (70% to 90% Ca). K-feldspar (KAlSi₃O₈) has a slightly different structure than that of plagioclase, owing to the larger size of the potassium ion (1.37 Å) and because of this large size, potassium and sodium do not readily substitute for each other, except at high temperatures. These high-temperature feldspars are likely to be found only in volcanic rocks because intrusive igneous rocks cool slowly enough to low temperatures for the feldspars to change into one of the lower-temperature forms.

Figure 2.15 Compositions of the feldspar minerals

In quartz (SiO₂), the silica tetrahedra are bonded in a “perfect” three-dimensional framework. Each tetrahedron is bonded to four other tetrahedra (with an oxygen shared at every corner of each tetrahedron), and as a result, the ratio of silicon to oxygen is 1:2. Since the one silicon cation has a +4 charge and the two oxygen anions each have a –2 charge, the charge is balanced. There is no need for aluminum or any of the other cations such as sodium or potassium. The hardness and lack of cleavage in quartz result from the strong covalent/ionic bonds characteristic of the silica tetrahedron.

In order for a mineral crystal to grow, the elements needed to make it must be present in the appropriate proportions, the physical and chemical conditions must be favourable, and there must be sufficient time for the atoms to become arranged.

Physical and chemical conditions include factors such as temperature, pressure, presence of water, pH, and amount of oxygen available. Time is one of the most important factors because it takes time for atoms to become ordered. If time is limited, the mineral grains will remain very small. The presence of water enhances the mobility of ions and can lead to the formation of larger crystals over shorter time periods.

Most of the minerals that make up the rocks around us formed through the cooling of molten rock, known as magma. At the high temperatures that exist deep within Earth, some geological materials are liquid. As magma rises up through the crust, either by volcanic eruption or by more gradual processes, it cools and minerals crystallize. If the cooling process is rapid (minutes, hours, days, or years), the components of the minerals will not have time to become ordered and only small crystals can form before the rock becomes solid. The resulting rock will be fine-grained (i.e., crystals less than 1 mm). If the cooling is slow (from decades to millions of years), the degree of ordering will be higher and relatively large crystals will form. In some cases, the cooling will be so fast (seconds) that the texture will be glassy, which means that no crystals at all form. Volcanic glass is not composed of minerals because the magma has cooled too rapidly for crystals to grow, although over time (millions of years) the volcanic glass may crystallize into various silicate minerals.

Minerals can also form in several other ways:

• Precipitation from aqueous solution (i.e., from hot water flowing underground, from evaporation of a lake or inland sea, or in some cases, directly from seawater)
• Precipitation from gaseous emanations (e.g., in volcanic regions as shown in Figure 2.1)
• Metamorphism — formation of new minerals directly from the elements within existing minerals under conditions of elevated temperature and pressure
• Weathering — during which minerals unstable at Earth’s surface may be altered to other minerals
• Organic formation — formation of minerals within shells (primarily calcite) and teeth and bones (primarily apatite) by organisms (these organically formed minerals are still called minerals because they can also form inorganically)

Opal is a mineraloid, because although it has all of the other properties of a mineral, it does not have a specific structure. Pearl is not a mineral because it can only be produced by organic processes.

Minerals are universal. A crystal of hematite on Mars will have the same properties as one on Earth, and the same as one on a planet orbiting another star. That’s good news for geology students who are planning interplanetary travel since we can use those properties to help us identify minerals anywhere. That doesn’t mean that it’s easy, however; identification of minerals takes a lot of practice. Some of the mineral properties that are useful for identification are as follows:

Colour

For most of us, colour is one of our key ways of identifying objects. While some minerals have particularly distinctive colours that make good diagnostic properties, many do not, and for many, colour is simply unreliable. The mineral sulphur (Figures 2.1 and 2.16) is always a distinctive and unique yellow. Hematite, on the other hand, is an example of a mineral for which colour is not diagnostic. In some forms hematite is deep dull red, but in others it is black and shiny metallic (Figure 2.16). Many other minerals can have a wide range of colours (e.g., quartz, feldspar, amphibole, fluorite, and calcite). In most cases, the variations in colours are a result of varying proportions of trace elements within the mineral. In the case of quartz, for example, yellow quartz (citrine) has trace amounts of ferric iron (Fe³⁺), rose quartz has trace amounts of manganese, purple quartz (amethyst) has trace amounts of iron, and milky quartz, which is very common, has millions of fluid inclusions (tiny cavities, each filled with water).

Figure 2.16 Examples of the colours of the minerals sulphur and hematite

Streak

In the context of minerals, “colour” is what you see when light reflects off the surface of the sample. One reason that colour can be so variable is that the type of surface is variable. If we grind a small amount of the sample to a powder we get a much better indication of its actual colour. This can easily be done by scraping a corner of the sample across a streak plate (a piece of unglazed porcelain). The result is that some of the mineral gets ground to a powder and we can get a better impression of its “true” colour (Figure 2.17).

Figure 2.17 The streak colours of earthy hematite (left) and specular hematite (right). Although the specular hematite streak looks close to black, it does have red undertones that you can see if you look closely. [SE]

Lustre

Lustre is the way light reflects off the surface of a mineral, and the degree to which it penetrates into the interior. The key distinction is between metallic and non-metallic lustres. Light does not pass through metals, and that is the main reason they look “metallic.” Even a thin sheet of metal — such as aluminum foil — will prevent light from passing through it. Many non-metallic minerals may look as if light will not pass through them, but if you take a closer look at a thin edge of the mineral you can see that it does. If a non-metallic mineral has a shiny, reflective surface, then it is called “glassy.” If it is dull and non-reflective, it is “earthy.” Other types of non-metallic lustres are “silky,” “pearly,” and “resinous.” Lustre is a good diagnostic property, since most minerals will always appear either metallic or non-metallic. There are a few exceptions to this (e.g., hematite in Figure 2.16).

Hardness

One of the most important diagnostic properties of a mineral is its hardness. In 1812 German mineralogist Friedrich Mohs came up with a list of 10 reasonably common minerals that had a wide range of hardness. These minerals are shown in Figure 2.18, with the Mohs scale of hardness along the bottom axis. In fact, while each mineral on the list is harder than the one before it, the relative measured hardnesses (vertical axis) are not linear. For example apatite is about three times harder than fluorite and diamond is three times harder than corundum. Some commonly available reference materials are also shown on this diagram, including a typical fingernail (2.5), a piece of copper wire (3.5), a knife blade or a piece of window glass (5.5), a hardened steel file (6.5), and a porcelain streak plate (7). These are tools that a geologist can use to measure the hardness of unknown minerals. For example, if you have a mineral that you can’t scratch with your fingernail, but you can scratch with a copper wire, then its hardness is between 2.5 and 3.5. And of course the minerals themselves can be used to test other minerals.

Figure 2.18 Minerals and reference materials in the Mohs scale of hardness. The “measured hardness” values are Vickers Hardness numbers.

Crystal Habit

When minerals form within rocks, there is a possibility that they will form in distinctive crystal shapes if they are not crowded out by other pre-existing minerals. Every mineral has one or more distinctive crystal habits, but it is not that common, in ordinary rocks, for the shapes to be obvious. Quartz, for example, will form six-sided prisms with pointed ends, but this typically happens only when it crystallizes from a hot water solution within a cavity in an existing rock (Figure 2.19). Pyrite can form cubic crystals (Figure 2.19), but can also form crystals with 12 faces, known as dodecahedra (“dodeca” means 12). The mineral garnet also forms dodecahedral crystals (Figure 2.19).

Figure 2.19 Hexagonal prisms of quartz (left), cubic crystals of pyrite (centre), and a dodecahedral crystal of garnet (right)

Because beautiful well-formed crystals are rare in ordinary rocks, habit isn’t as useful a diagnostic feature as one might think. However, there are several minerals for which it is important. One is garnet, which is common in some metamorphic rocks and typically displays the dodecahedral shape. Another is amphibole, which forms long thin crystals, and is common in igneous rocks like granite (Figure 1.5).

Mineral habit is often related to the regular arrangement of the molecules that make up the mineral. Some of the terms that are used to describe habit include bladed, botryoidal (grape-like), dendritic (branched), drusy (an encrustation of minerals), equant (similar in all dimensions), fibrous, platy, prismatic (long and thin), and stubby.

Cleavage and fracture

Crystal habit is a reflection of how a mineral grows, while cleavage and fracture describe how it breaks. These characteristics are the most important diagnostic features of many minerals, and often the most difficult to understand and identify. Cleavage is what we see when a mineral breaks along a specific plane or planes, while fracture is an irregular break. Some minerals tend to cleave along planes at various fixed orientations, some do not cleave at all (they only fracture). Minerals that have cleavage can also fracture along surfaces that are not parallel to their cleavage planes.

As we’ve already discussed, the way that minerals break is determined by their atomic arrangement and specifically by the orientation of weaknesses within the lattice. Graphite and the micas, for example, have cleavage planes parallel to their sheets (Figures 2.7 and 2.14), and halite has three cleavage planes parallel to the lattice directions (Figure 2.8).

Quartz has no cleavage because it has equally strong Si^–O bonds in all directions, and feldspar has two cleavages at 90° to each other (Figure 1.5).

One of the main difficulties with recognizing and describing cleavage is that it is visible only in individual crystals. Most rocks have small crystals and it’s very difficult to see the cleavage within a small crystal. Geology students have to work hard to understand and recognize cleavage, but it’s worth the effort since it is a reliable diagnostic property for most minerals.

Density

Density is a measure of the mass of a mineral per unit volume, and it is a useful diagnostic tool in some cases. Most common minerals, such as quartz, feldspar, calcite, amphibole, and mica, have what we call “average density” (2.6 to 3.0 g/cm³), and it would be difficult to tell them apart on the basis of their density. On the other hand, many of the metallic minerals, such as pyrite, hematite, and magnetite, have densities over 5 g/cm³. They can easily be distinguished from the lighter minerals on the basis of density, but not necessarily from each other. A limitation of using density as a diagnostic tool is that one cannot assess it in minerals that are a small part of a rock with other minerals in it.

Other properties

Several other properties are also useful for identification of some minerals. For example, calcite is soluble in dilute acid and will give off bubbles of carbon dioxide. Magnetite is magnetic, so will affect a magnet. A few other minerals are weakly magnetic.

Attributions

Figure 2.19
Quartz Bresil by Didier Descouens is under CC BY 3.0
Pyrite cubic crystals on marlstone by Carles Millan is under CC BY SA 3.0
Almandine garnet by Eurico Zimbres (FGEL/UERJ) and Tom Epaminondas (mineral collector) is under CC BY SA 2.0

The topics covered in this chapter can be summarized as follows.

Introduction

Figure 3.1 A fine-grained mafic dyke (dark green) intruded into a felsic dyke (pink) and into coarse diorite (grey), Quadra Island, B.C. All of these rocks are composed of more than one type of mineral. The mineral components are clearly visible in the diorite, but not in the other two rock types. [SE photo]

A rock is a consolidated mixture of the same or different minerals. By consolidated, we mean hard and strong; real rocks don’t fall apart in your hands! A mixture of minerals implies the presence of more than one mineral grain, but not necessarily more than one type of mineral (Figure 3.1). A rock can be composed of only one type of mineral (e.g., limestone is commonly made up of only calcite), but most rocks are composed of several different minerals. A rock can also include non-minerals, such as fossils or the organic matter within a coal bed or in some types of mudstone.

Rocks are grouped into three main categories based on how they form:

Igneous: formed from the cooling and crystallization of magma (molten rock)

Sedimentary: formed when weathered fragments of other rocks are buried, compressed, and cemented together, or when minerals precipitate directly from solution

Metamorphic: formed by alteration (due to heat, pressure, and/or chemical action) of a pre-existing igneous or sedimentary rock

The rock components of the crust are slowly but constantly being changed from one form to another and the processes involved are summarized in the rock cycle (Figure 3.2). The rock cycle is driven by two forces: (1) Earth’s internal heat engine, which moves material around in the core and the mantle and leads to slow but significant changes within the crust, and (2) the hydrological cycle, which is the movement of water, ice, and air at the surface, and is powered by the sun.

The rock cycle is still active on Earth because our core is hot enough to keep the mantle moving, our atmosphere is relatively thick, and we have liquid water. On some other planets or their satellites, such as the Moon, the rock cycle is virtually dead because the core is no longer hot enough to drive mantle convection and there is no atmosphere or liquid water.

Figure 3.2 A schematic view of the rock cycle. [SE]

In describing the rock cycle, we can start anywhere we like, although it’s convenient to start with magma. As we’ll see in more detail below, magma is rock that is hot to the point of being entirely molten. This happens at between about 800° and 1300°C, depending on the composition and the pressure, onto the surface and cool quickly (within seconds to years) — forming extrusive igneous rock (Figure 3.3).

Figure 3.3 Magma forming pahoehoe basalt at Kilauea Volcano, Hawaii [SE]

Magma can either cool slowly within the crust (over centuries to millions of years) — forming intrusive igneous rock, or erupt onto the surface and cool quickly (within seconds to years) — forming extrusive igneous rock. Intrusive igneous rock typically crystallizes at depths of hundreds of metres to tens of kilometres below the surface. To change its position in the rock cycle, intrusive igneous rock has to be uplifted and exposed by the erosion of the overlying rocks.

Through the various plate-tectonics-related processes of mountain building, all types of rocks are uplifted and exposed at the surface. Once exposed, they are weathered, both physically (by mechanical breaking of the rock) and chemically (by weathering of the minerals), and the weathering products — mostly small rock and mineral fragments — are eroded, transported, and then deposited as sediments. Transportation and deposition occur through the action of glaciers, streams, waves, wind, and other agents, and sediments are deposited in rivers, lakes, deserts, and the ocean.

Figure 3.4 Cretaceous-aged marine sandstone overlying mudstone, Gabriola Island, B.C. [SE]

Unless they are re-eroded and moved along, sediments will eventually be buried by more sediments. At depths of hundreds of metres or more, they become compressed and cemented into sedimentary rock. Again through various means, largely resulting from plate-tectonic forces, different kinds of rocks are either uplifted, to be re-eroded, or buried deeper within the crust where they are heated up, squeezed, and changed into metamorphic rock.

Figure 3.5 Metamorphosed and folded Triassic-aged limestone, Quadra Island, B.C. [SE]

Magmas can vary widely in composition, but in general they are made up of only eight elements; in order of importance: oxygen, silicon, aluminum, iron, calcium, sodium, magnesium, and potassium (Figure 3.6). Oxygen, the most abundant element in magma, comprises a little less than half the total, followed by silicon at just over one-quarter. The remaining elements make up the other one-quarter. Magmas derived from crustal material are dominated by oxygen, silicon, aluminum, sodium, and potassium.

The composition of magma depends on the rock it was formed from (by melting), and the conditions of that melting. Magmas derived from the mantle have higher levels of iron, magnesium, and calcium, but they are still likely to be dominated by oxygen and silicon. All magmas have varying proportions of elements such as hydrogen, carbon, and sulphur, which are converted into gases like water vapour, carbon dioxide, and hydrogen sulphide as the magma cools.

Figure 3.6 Average elemental proportions in Earth’s crust, which is close to the average composition of magmas within the crust [SE]

Virtually all of the igneous rocks that we see on Earth are derived from magmas that formed from partial melting of existing rock, either in the upper mantle or the crust. Partial melting is what happens when only some parts of a rock melt; it takes place because rocks are not pure materials. Most rocks are made up of several minerals, each of which has a different melting temperature. The wax in a candle is a pure material. If you put some wax into a warm oven (50°C will do as the melting temperature of most wax is about 40°C) and leave it there for a while, it will soon start to melt. That’s complete melting, not partial melting. If instead you took a mixture of wax, plastic, aluminum, and glass and put it into the same warm oven, the wax would soon start to melt, but the plastic, aluminum, and glass would not melt (Figure 3.7a). That’s partial melting and the result would be solid plastic, aluminum, and glass surrounded by liquid wax (Figure 3.7b). If we heat the oven up to around 120°C, the plastic would melt too and mix with the liquid wax, but the aluminum and glass would remain solid (Figure 3.7c). Again this is partial melting. If we separated the wax/plastic “magma” from the other components and let it cool, it would eventually harden. As you can see from Figure 3.7d, the liquid wax and plastic have mixed, and on cooling, have formed what looks like a single solid substance. It is most likely that this is a very fine-grained mixture of solid wax and solid plastic, but it could also be some other substance that has formed from the combination of the two.

Figure 3.7 Partial melting of “pretend rock”: (a) the original components of white candle wax, black plastic pipe, green beach glass, and aluminum wire, (b) after heating to 50˚C for 30 minutes only the wax has melted, (c) after heating to 120˚C for 60 minutes much of the plastic has melted and the two liquids have mixed, (d) the liquid has been separated from the solids and allowed to cool to make a “pretend rock” with a different overall composition. [SE]

In this example, we partially melted some pretend rock to create some pretend magma. We then separated the magma from the source and allowed it to cool to make a new pretend rock with a composition quite different from the original material (it lacks glass and aluminum).

Of course partial melting in the real world isn’t exactly the same as in our pretend-rock example. The main differences are that rocks are much more complex than the four-component system we used, and the mineral components of most rocks have more similar melting temperatures, so two or more minerals are likely to melt at the same time to varying degrees. Another important difference is that when rocks melt, the process takes thousands to millions of years, not the 90 minutes it took in the pretend-rock example.

Contrary to what one might expect, and contrary to what we did to make our pretend rock, most partial melting of real rock does not involve heating the rock up. The two main mechanisms through which rocks melt are decompression melting and flux melting. Decompression melting takes place within Earth when a body of rock is held at approximately the same temperature but the pressure is reduced. This happens because the rock is being moved toward the surface, either at a mantle plume (a.k.a., hot spot), or in the upwelling part of a mantle convection cell.Mantle plumes are described in Chapter 4 and mantle convection in Chapter 9. The mechanism of decompression melting is shown in Figure 3.8a. If a rock that is hot enough to be close to its melting point is moved toward the surface, the pressure is reduced, and the rock can pass to the liquid side of its melting curve. At this point, partial melting starts to take place. The process of flux melting is shown in Figure 3.8b. If a rock is close to its melting point and some water (a flux that promotes melting) is added to the rock, the melting temperature is reduced (solid line versus dotted line), and partial melting starts.

Figure 3.8 Mechanisms for (a) decompression melting (the rock is moved toward the surface) and (b) flux melting (water is added to the rock) and the melting curve is displaced. [SE]

The partial melting of rock happens in a wide range of situations, most of which are related to plate tectonics. The more important of these are shown in Figure 3.9. At both mantle plumes and in the upward parts of convection systems, rock is being moved toward the surface, the pressure is dropping, and at some point, the rock crosses to the liquid side of its melting curve. At subduction zones, water from the wet, subducting oceanic crust is transferred into the overlying hot mantle. This provides the flux needed to lower the melting temperature. In both of these cases, only partial melting takes place — typically only about 10% of the rock melts — and it is always the most silica-rich components of the rock that melt, creating a magma that is more silica-rich than the rock from which it is derived. (By analogy, the melt from our pretend rock is richer in wax and plastic than the “rock” from which it was derived.) The magma produced, being less dense than the surrounding rock, moves up through the mantle, and eventually into the crust.

Figure 3.9 Common sites of magma formation in the upper mantle. The black circles are regions of partial melting. The blue arrows represent water being transferred from the subducting plates into the overlying mantle. [SE, after USGS (http://pubs.usgs.gov/gip/dynamic/Vigil.html)]

As it moves toward the surface, and especially when it moves from the mantle into the lower crust, the hot magma interacts with the surrounding rock. This typically leads to partial melting of the surrounding rock because most such magmas are hotter than the melting temperature of crustal rock. (In this case, melting is caused by an increase in temperature.) Again, the more silica-rich parts of the surrounding rock are preferentially melted, and this contributes to an increase in the silica content of the magma.

At very high temperatures (over 1300°C), most magma is entirely liquid because there is too much energy for the atoms to bond together. As the temperature drops, usually because the magma is slowly moving upward, things start to change. Silicon and oxygen combine to form silica tetrahedra, and then, as cooling continues, the tetrahedra start to link together to make chains (polymerize). These silica chains have the important effect of making the magma more viscous (less runny), and as we’ll see in Chapter 4, magma viscosity has significant implications for volcanic eruptions. As the magma continues to cool, crystals start to form.

The minerals that make up igneous rocks crystallize at a range of different temperatures. This explains why a cooling magma can have some crystals within it and yet remain predominantly liquid. The sequence in which minerals crystallize from a magma is known as the Bowen reaction series (Figure 3.10 and Who was Bowen).

Of the common silicate minerals, olivine normally crystallizes first, at between 1200° and 1300°C. As the temperature drops, and assuming that some silica remains in the magma, the olivine crystals react (combine) with some of the silica in the magma (see Box 3.1) to form pyroxene. As long as there is silica remaining and the rate of cooling is slow, this process continues down the discontinuous branch: olivine to pyroxene, pyroxene to amphibole, and (under the right conditions) amphibole to biotite.

At about the point where pyroxene begins to crystallize, plagioclase feldspar also begins to crystallize. At that temperature, the plagioclase is calcium-rich (anorthite) (see Figure 2.15). As the temperature drops, and providing that there is sodium left in the magma, the plagioclase that forms is a more sodium-rich variety.

Figure 3.10 The Bowen reaction series describes the process of magma crystallization [SE]

Who was Bowen, and what’s a reaction series?

Norman Levi Bowen, born in Kingston Ontario, studied geology at Queen’s University and then at MIT in Boston. In 1912, he joined the Carnegie Institution in Washington, D.C., where he carried out groundbreaking experimental research into the processes of cooling magmas. Working mostly with basaltic magmas, he determined the order of crystallization of minerals as the temperature drops. The method, in brief, was to melt the rock to a magma in a specially made kiln, allow it to cool slowly to a specific temperature (allowing some minerals to form), and then quench it (cool it quickly) so that no new minerals form (only glass). The results were studied under the microscope and by chemical analysis. This was done over and over, each time allowing the magma to cool to a lower temperature before quenching.

The Bowen reaction series is one of the results of his work, and even a century later, it is an important basis for our understanding of igneous rocks. The word reaction is critical. In the discontinuous branch, olivine is typically the first mineral to form (at just below 1300°C). As the temperature continues to drop, olivine becomes unstable while pyroxene becomes stable. The early-forming olivine crystals react with silica in the remaining liquid magma and are converted into pyroxene, something like this:

Mg₂SiO₄ + SiO₂  ——>     2MgSiO₃

olivine                                       pyroxene

This continues down the chain, as long as there is still silica left in the liquid. [image from Wikipedia: http://en.wikipedia.org/wiki/File:NormanLBowen_1909.jpg]

In cases where cooling happens relatively quickly, individual plagioclase crystals can be zoned from calcium-rich in the centre to more sodium-rich around the outside. This occurs when calcium-rich early-forming plagioclase crystals become coated with progressively more sodium-rich plagioclase as the magma cools. Figure 3.11 shows a zoned plagioclase under a microscope.

Figure 3.11 A zoned plagioclase crystal. The central part is calcium-rich and the outside part is sodium-rich: [Sandra Johnstone, used with permission]

Finally, if the magma is quite silica-rich to begin with, there will still be some left at around 750° to 800°C, and from this last magma, potassium feldspar, quartz, and maybe muscovite mica will form.

The composition of the original magma is critical to magma crystallization because it determines how far the reaction process can continue before all of the silica is used up. The compositions of typical mafic, intermediate, and felsic magmas are shown in Figure 3.12. Note that, unlike Figure 3.6, these compositions are expressed in terms of “oxides” (e.g., Al₂O₃ rather than just Al). There are two reasons for this: one is that in the early analytical procedures, the results were always expressed that way, and the other is that all of these elements combine readily with oxygen to form oxides.

Figure 3.12 The chemical compositions of typical mafic, intermediate, and felsic magmas and the types of rocks that form from them. [SE]

Mafic magmas have 45% to 55% SiO₂, about 25% total of FeO and MgO plus CaO, and about 5% Na₂O + K₂O. Felsic magmas, on the other hand, have much more SiO₂ (65% to 75%) and Na₂O + K₂O (around 10%) and much less FeO and MgO plus CaO (about 5%).

As a mafic magma starts to cool, some of the silica combines with iron and magnesium to make olivine. As it cools further, much of the remaining silica goes into calcium-rich plagioclase, and any silica left may be used to convert some of the olivine to pyroxene. Soon after that, all of the magma is used up and no further changes takes place. The minerals present will be olivine, pyroxene, and calcium-rich plagioclase. If the magma cools slowly underground, the product will be gabbro; if it cools quickly at the surface, the product will be basalt (Figure 3.13).

Felsic magmas tend to be cooler than mafic magmas when crystallization begins (because they don’t have to be as hot to remain liquid), and so they may start out crystallizing pyroxene (not olivine) and plagioclase. As cooling continues, the various reactions on the discontinuous branch will proceed because silica is abundant, the plagioclase will become increasingly sodium-rich, and eventually potassium feldspar and quartz will form. Commonly even very felsic rocks will not have biotite or muscovite because they may not have enough aluminum or enough hydrogen to make the OH complexes that are necessary for mica minerals. Typical felsic rocks are granite and rhyolite (Figure 3.13).

The cooling behaviour of intermediate magmas lie somewhere between those of mafic and felsic magmas. Typical intermediate rocks are diorite and andesite (Figure 3.13).

Figure 3.13 Examples of the igneous rocks that form from mafic, intermediate, and felsic magmas. [SE]

A number of processes that take place within a magma chamber can affect the types of rocks produced in the end. If the magma has a low viscosity (i.e., it’s runny) — which is likely if it is mafic — the crystals that form early, such as olivine (Figure 3.14a), may slowly settle toward the bottom of the magma chamber (Figure 3.14b). The means that the overall composition of the magma near the top of the magma chamber will become more felsic, as it is losing some iron- and magnesium-rich components. This process is known as fractional crystallization. The crystals that settle might either form an olivine-rich layer near the bottom of the magma chamber, or they might remelt because the lower part is likely to be hotter than the upper part (remember, from Chapter 1, that temperatures increase steadily with depth in Earth because of the geothermal gradient). If any melting takes place, crystal settling will make the magma at the bottom of the chamber more mafic than it was to begin with (Figure 3.14c).

Figure 3.14 An example of crystal settling and the formation of a zoned magma chamber [SE]

If crystal settling does not take place, because the magma is too viscous, then the process of cooling will continue as predicted by the Bowen reaction series. In some cases, however, partially cooled but still liquid magma, with crystals in it, will either move farther up into a cooler part of the crust, or all the way to the surface during a volcanic eruption. In either of these situations, the magma that has moved toward the surface is likely to cool much faster than it did within the magma chamber, and the rest of the rock will have a finer crystalline texture. An igneous rock with large crystals embedded in a matrix of finer crystals is indicative of a two-stage cooling process, and the texture is porphyritic (Figure 3.15).

Figure 3.15 Porphyritic textures: volcanic porphyry (left – olivine crystals in Hawaiian basalt) and intrusive porphyry (right) [SE]

As has already been described, igneous rocks are classified into four categories, based on either their chemistry or their mineral composition: felsic, intermediate, mafic, and ultramafic. The diagram in Figure 3.16 can be used to help classify igneous rocks by their mineral composition. An important feature to note on this diagram is the red line separating the non-ferromagnesian silicates in the lower left (K-feldspar, quartz, and plagioclase feldspar) from the ferromagnesian silicates in the upper right (biotite, amphibole, pyroxene, and olivine). In classifying intrusive igneous rocks, the first thing to consider is the percentage of ferromagnesian silicates. That’s relatively easy in most igneous rocks because the ferromagnesian minerals are clearly darker than the others. At the same time, it’s quite difficult to estimate the proportions of minerals in a rock.

Based on the position of the red line in Figure 3.16, it is evident that felsic rocks can have about 1% to 20% ferromagnesian silicates (the red line intersects the left side of the felsic zone 1% of the distance from the top of the diagram, and it intersects the right side of the felsic zone 20% of the distance from the top). Intermediate rocks have between 20% and 50% ferromagnesian silicates, and mafic rocks have 50% to 100% ferromagnesian silicates. To be more specific, felsic rocks typically have biotite and/or amphibole; intermediate rocks have amphibole and, in some cases, pyroxene; and mafic rocks have pyroxene and, in some cases, olivine.

Figure 3.16 A simplified classification diagram for igneous rocks based on their mineral compositions [SE]

If we focus on the non-ferromagnesian silicates, it is evident that felsic rocks can have from 0% to 35% K-feldspar, from 25% to 35% quartz (the vertical thickness of the quartz field varies from 25% to 35%), and from 25% to 50% plagioclase (and that plagioclase will be sodium-rich, or albitic). Intermediate rocks can have up to 25% quartz and 50% to 75% plagioclase. Mafic rocks only have plagioclase (up to 50%), and that plagioclase will be calcium-rich, or anorthitic.

Figure 3.17 provides a diagrammatic representation of the proportions of dark minerals in light-coloured rocks. You can use that when trying to estimate the ferromagnesian mineral content of actual rocks, and you can get some practice doing that by completing Exercise 3.6.

Figure 3.17 A guide to estimating the proportions of dark minerals in light-coloured rocks

Exercises

Exercise 3.6 Proportions of Ferromagnesian Silicates

The four igneous rocks shown below have differing proportions of ferromagnesian silicates. Estimate those proportions using the diagrams in Figure 3.17, and then use Figure 3.16 to determine the likely rock name for each one.

___%	___%	___%	___%
__________	__________	__________	__________

Igneous rocks are also classified according to their textures. The textures of volcanic rocks will be discussed in Chapter 4, so here we’ll only look at the different textures of intrusive igneous rocks. Almost all intrusive igneous rocks have crystals that are large enough to see with the naked eye, and we use the term phaneritic (from the Greek word phaneros meaning visible) to describe that. Typically that means they are larger than about 0.5 mm — the thickness of a strong line made with a ballpoint pen. (If the crystals are too small to distinguish, which is typical of most volcanic rocks, we use the term aphanitic.) The intrusive rocks shown in Figure 3.13 are all phaneritic, as are those shown in Exercise 3.6. 

In general, the size of crystals is proportional to the rate of cooling. The longer it takes for a body of magma to cool, the larger the crystals will be. It is not uncommon to see an intrusive igneous rock with crystals up to a centimetre long. In some situations, especially toward the end of the cooling stage, the magma can become water rich. The presence of liquid water (still liquid at high temperatures because it is under pressure) promotes the relatively easy movement of ions, and this allows crystals to grow large, sometimes to several centimetres (Figure 3.18). As already described, if an igneous rock goes through a two-stage cooling process, its texture will be porphyritic (Figure 3.15).

Figure 3.18 A pegmatite with mica, quartz, and tourmaline (black) from the White Elephant mine, South Dakota [from http://en.wikipedia.org/wiki/Pegmatite#mediaviewer/File:We-pegmatite.jpg]

In most cases, a body of hot magma is less dense than the rock surrounding it, so it has a tendency to move very slowly up toward the surface. It does so in a few different ways, including filling and widening existing cracks, melting the surrounding rock (called country rock“Country rock” is not necessarily music to a geologist’s ears. The term refers to the original “rock of the country” or region, and hence the rock into which the magma intruded to form a pluton.), pushing the rock aside (where it is somewhat plastic), and breaking the rock. Where some of the country rock is broken off, it may fall into the magma, a process called stoping. The resulting fragments, illustrated in Figure 3.19, are known as xenoliths (Greek for “strange rocks”).

Figure 3.19 Xenoliths of mafic rock in granite, Victoria, B.C. The fragments of dark rock have been broken off and incorporated into the light-coloured granite. [SE]

Some upward-moving magma reaches the surface, resulting in volcanic eruptions, but most cools within the crust. The resulting body of rock is known as a pluton. Plutons can have various different shapes and relationships to the surrounding country rock as shown in Figure 3.20.

Figure 3.20 Depiction of some of the types of plutons. a: stocks (if they coalesce at depth then they might constitute a batholith), b: sill (a tabular body, in this case parallel to bedding), c: dyke (cross-cuts bedding), d: laccolith (a sill that has pushed up the overlying rock layers), e: pipe (a cylindrical conduit feeding a volcano). The two features labelled f could be pipes or dykes, but from this perspective it’s not possible to determine if they are cylindrical or tabular. [SE drawing]

Large irregular-shaped plutons are called either stocks or batholiths. The distinction between the two is made on the basis of the area that is exposed at the surface: if the body has an exposed surface area greater than 100 km², then it’s a batholith; smaller than 100 km² and it’s a stock. Batholiths are typically formed only when a number of stocks coalesce beneath the surface to create one large body. One of the largest batholiths in the world is the Coast Range Plutonic Complex, which extends all the way from the Vancouver region to southeastern Alaska (Figure 3.21). More accurately, it’s many batholiths.

Tabular (sheet-like) plutons are distinguished on the basis of whether or not they are concordant with (parallel to) existing layering (e.g., sedimentary bedding or metamorphic foliation) in the country rock. A sill is concordant with existing layering, and a dyke is discordant. If the country rock has no bedding or foliation, then any tabular body within it is a dyke. Note that the sill-versus-dyke designation is not determined simply by the orientation of the feature. A dyke can be horizontal and a sill can be vertical (if the bedding is vertical). A large dyke can be seen in Figure 3.21.

A laccolith is a sill-like body that has expanded upward by deforming the overlying rock.

Finally, a pipe is a cylindrical body (with a circular, ellipitical, or even irregular cross-section) that served as a conduit for the movement of magma from one location to another. Most known pipes fed volcanoes, although pipes can also connect plutons. It is also possible for a dyke to feed a volcano.

Figure 3.21 The Stawamus Chief, part of the Coast Range Plutonic Complex, near to Squamish, B.C. The cliff is about 600 m high. Most of the dark stripes are a result of algae and lichen growth where the surface is frequently wet, but there is a large (about 10 m across) vertical dyke that extends from bottom to top. [SE photo]

As discussed already, plutons can interact with the rocks into which they are intruded, sometimes leading to partial melting of the country rock or to stoping and formation of xenoliths. And, as we’ll see in Chapter 7, the heat of a body of magma can lead to metamorphism of the country rock. The country rock can also have an effect on the magma within a pluton. The most obvious such effect is the formation of a chilled margin along the edges of the pluton, where it came in contact with country rock that was significantly colder than the magma. Within the chilled margin, the magma cooled more quickly than in the centre of the dyke, so the texture is finer and the colour may be different. An example is shown in Figure 3.22.

Figure 3.22 A mafic dyke with chilled margins within basalt at Nanoose, B.C. The coin is 24 mm in diameter. The dyke is about 25 cm across and the chilled margins are 2 cm wide.

The topics covered in this chapter can be summarized as follows:

Introduction

A volcano is any location where magma comes to the surface, or has done so within the past several million years. This can include eruptions on the ocean floor (or even under the water of lake), where they are called subaqueous eruptions, or on land, where they are called subaerial eruptions. Not all volcanic eruptions produce the volcanic mountains with which we are familiar; in fact most of Earth’s volcanism takes place along the spreading ridges on the sea floor and does not produce volcanic mountains at all — not even sea-floor mountains.

Canada has a great deal of volcanic rock, but most of it is old, some of it billions of years old. Only in B.C. and the Yukon are there volcanoes that have been active within the past 2.6 Ma (Pleistocene or younger), and the vast majority of these are in B.C. We’ll look at those in some detail toward the end of this chapter, but a few of them are shown on Figures 4.1 and 4.2.

The study of volcanoes is critical to our understanding of the geological evolution of Earth, and to our understanding of significant changes in climate. But, most important of all, understanding volcanic eruptions allows us to save lives and property. Over the past few decades, volcanologists have made great strides in their ability to forecast volcanic eruptions and predict the consequences — this has already saved thousands of lives.

Figure 4.1 Mt. Garibaldi, near Squamish B.C., is one of Canada’s tallest (2,678 m) and most recently active volcanoes. It last erupted approximately 10,000 years ago. [SE photo]

Figure 4.2 Mt. Garibaldi (background left, looking from the north) with Garibaldi Lake in the foreground. The volcanic peak in the centre is Mt. Price and the dark flat–topped peak is The Table. All three of these volcanoes were active during the last glaciation. [SE photo]

The relationships between plate tectonics and volcanism are shown on Figure 4.3. As summarized in Chapter 3, magma is formed at three main plate-tectonic settings: divergent boundaries (decompression melting), convergent boundaries (flux melting), and mantle plumes (decompression melting).

Figure 4.3 The plate-tectonic settings of common types of volcanism. Composite volcanoes form at subduction zones, either on ocean-ocean convergent boundaries (left) or ocean-continent convergent boundaries (right). Both shield volcanoes and cinder cones form in areas of continental rifting. Shield volcanoes form above mantle plumes, but can also form at other tectonic settings. Sea-floor volcanism can take place at divergent boundaries, mantle plumes and ocean-ocean-convergent boundaries. [SE, after USGS (http://pubs.usgs.gov/gip/dynamic/Vigil.html)]

The mantle and crustal processes that take place in areas of volcanism are illustrated in Figure 4.4. At a spreading ridge, hot mantle rock moves slowly upward by convection (cm/year), and within about 60 km of the surface, partial melting starts because of decompression. Over the triangular area shown in Figure 4.4a, about 10% of the ultramafic mantle rock melts, producing mafic magma that moves upward toward the axis of spreading (where the two plates are moving away from each other). The magma fills vertical fractures produced by the spreading and spills out onto the sea floor to form basaltic pillows (more on that later) and lava flows. There is spreading-ridge volcanism taking place about 200 km offshore from the west coast of Vancouver Island.

Exercises

Exercise 4.1 How Thick Is the Oceanic Crust?

Figure 4.4a shows a triangular zone about 60 km thick; within this zone, approximately 10% of the mantle rock melts to form oceanic crust. Based on this information, approximately how thick do you think the resulting oceanic crust should be?

Figure 4.4 The processes that lead to volcanism in the three main volcanic settings on Earth: (a) volcanism related to plate divergence, (b) volcanism at an ocean-continent boundary*, and (c) volcanism related to a mantle plume. [SE, after USGS (http://pubs.usgs.gov/gip/dynamic/Vigil.html)]
*Similar processes take place at an ocean-ocean convergent boundary.

At an ocean-continent or ocean-oceanAt an ocean-continent convergent boundary, part of a plate that is made up of oceanic crust is subducting beneath part of another plate made up of continental crust. At an ocean-ocean convergent boundary, oceanic crust is being subducted beneath another oceanic-crust plate. convergent boundary, oceanic crust is pushed far down into the mantle (Figure 4.4b). It is heated up, and while there isn’t enough heat to melt the subducting crust, there is enough to force the water out of some of its minerals. This water rises into the overlying mantle where it contributes to flux melting of the mantle rock. The mafic magma produced rises through the mantle to the base of the crust. There it contributes to partial melting of crustal rock, and thus it assimilates much more felsic material. That magma, now intermediate in composition, continues to rise and assimilate crustal material; in the upper part of the crust, it accumulates into plutons. From time to time, the magma from the plutons rises toward surface, leading to volcanic eruptions. Mt. Garibaldi (Figures 4.1 and 4.2) is an example of subduction-related volcanism. 

A mantle plume is an ascending column of hot rock (not magma) that originates deep in the mantle, possibly just above the core-mantle boundary. Mantle plumes are thought to rise at approximately 10 times the rate of mantle convection. The ascending column may be on the order of kilometres to tens of kilometres across, but near the surface it spreads out to create a mushroom-style head that is several tens to over 100 kilometres across. Near the base of the lithosphere (the rigid part of the mantle), the mantle plume (and possibly some of the surrounding mantle material) partially melts to form mafic magma that rises to feed volcanoes. Since most mantle plumes are beneath the oceans, the early stages of volcanism typically take place on the sea floor. Over time, islands may form like those in Hawaii.

Volcanism in northwestern B.C. (Figures 4.5 and 4.6) is related to continental rifting. This area is not at a divergent or convergent boundary, and there is no evidence of an underlying mantle plume. The crust of northwestern B.C. is being stressed by the northward movement of the Pacific Plate against the North America Plate, and the resulting crustal fracturing provides a conduit for the flow of magma from the mantle. This may be an early stage of continental rifting, such as that found in eastern Africa.

Figure 4.5 Volcanoes and volcanic fields in the Northern Cordillera Volcanic Province, B.C. (base map from Wikipedia (http://commons.wikimedia.org/wiki/File:South-West_Canada.jpg). Volcanic locations from Edwards, B. & Russell, J. (2000). Distribution, nature, and origin of Neogene-Quaternary magmatism in the northern Cordilleran volcanic province, Canada. Geological Society of America Bulletin. pp. 1280-1293[SE]Cordillera Volcanic Province, B.C.

Figure 4.6 Volcanic rock at the Tseax River area, northwestern B.C. [SE]

As noted in the previous section, the types of magma produced in the various volcanic settings can differ significantly. At divergent boundaries and oceanic mantle plumes, where there is little interaction with crustal materials and magma fractionation to create felsic melts does not take place, the magma tends to be consistently mafic. At subduction zones, where the magma ascends through significant thicknesses of crust, interaction between the magma and the crustal rock — some of which is quite felsic — leads to increases in the felsic character of the magma.

As shown in Figure 4.7, several processes can make magma that is stored in a chamber within the crust more felsic, and can also contribute to development of vertical zonation from more mafic at the bottom to more felsic at the top. Partial melting of country rock and country-rock xenoliths increases the overall felsic character of the magma; first, because the country rocks tends to be more felsic than the magma, and second, because the more felsic components of the country rock melt preferentially. Settling of ferromagnesian crystals from the upper part of the magma, and possible remelting of those crystals in the lower part can both contribute to the vertical zonation from relatively mafic at the bottom to more felsic at the top.

Figure 4.7 The important processes that lead to changes in the composition of magmas stored within magma chambers within relatively felsic rocks of the crust. [SE]

From the perspective of volcanism there are some important differences between felsic and mafic magmas. First, as we’ve already discussed, felsic magmas tend to be more viscous because they have more silica, and hence more polymerization. Second, felsic magmas tend to have higher levels of volatiles; that is, components that behave as gases during volcanic eruptions. The most abundant volatile in magma is water (H₂O), followed typically by carbon dioxide (CO₂), and then by sulphur dioxide (SO₂). The general relationship between the SiO₂ content of magma and the amount of volatiles is shown in Figure 4.8. Although there are many exceptions to this trend, mafic magmas typically have 1% to 3% volatiles, intermediate magmas have 3% to 4% volatiles, and felsic magmas have 4% to 7% volatiles.

Figure 4.8 Variations in the volatile compositions of magmas as a function of silica content [SE after Schminke, 2004, (Schminke, H-U., 2004, Volcanism, Springer-Verlag, Heidelberg)]

Differences in viscosity and volatile level have significant implications for the nature of volcanic eruptions. When magma is deep beneath the surface and under high pressure from the surrounding rocks, the gases remain dissolved. As magma approaches the surface, the pressure exerted on it decreases. Gas bubbles start to form, and the more gas there is in the magma, the more bubbles form. If the gas content is low or the magma is runny enough for gases to rise up through it and escape to surface, the pressure will not become excessive. Assuming that it can break through to the surface, the magma will flow out relatively gently. An eruption that involves a steady non-violent flow of magma is called effusive.

If the magma is felsic, and therefore too viscous for gases to escape easily, or if it has a particularly high gas content, it is likely to be under high pressure. Viscous magma doesn’t flow easily, so even if there is a way for it to move out, it may not flow out. Under these circumstances pressure will continue to build as more magma moves up from beneath and gases continue to exsolve. Eventually some part of the volcano will break and then all of that pent-up pressure will lead to an explosive eruption.

Mantle plume and spreading-ridge magmas tend to be consistently mafic, so effusive eruptions are the norm. At subduction zones, the average magma composition is likely to be close to intermediate, but as we’ve seen, magma chambers can become zoned and so compositions ranging from felsic to mafic are possible. Eruption styles can be correspondingly variable.

There are numerous types of volcanoes or volcanic sources; some of the more common ones are summarized in Table 4.1.

Table 4.1 A summary of the important types of volcanism

The sizes and shapes of typical shield, composite, and cinder-cone volcanoes are compared in Figure 4.9, although, to be fair, Mauna Loa is the largest shield volcano on Earth; all others are smaller. Mauna Loa rises from the surrounding flat sea floor, and its diameter is in the order of 200 km. Its elevation is 4,169 m above sea level. Mt. St. Helens, a composite volcano, rises above the surrounding hills of the Cascade Range. Its diameter is about 6 km, and its height is 2,550 m above sea level. Cinder cones are much smaller. On this drawing, even a large cinder cone is just a dot.

Figure 4.9 Profiles of Mauna Loa shield volcano, Mt. St. Helens composite volcano, and a large cinder cone [SE]

Cinder Cones

Cinder cones, like Eve Cone in northern B.C. (Figure 4.10), are typically only a few hundred metres in diameter, and few are more than 200 m high. Most are made up of fragments of vesicular mafic rock (scoria) that were expelled as the magma boiled when it approached the surface, creating fire fountains. In many cases, these later became effusive (lava flows) when the gases were depleted. Most cinder cones are monogenetic, meaning that they formed during a single eruptive phase that might have lasted weeks or months. Because cinder cones are made up almost exclusively of loose fragments, they have very little strength. They can be easily, and relatively quickly, eroded away.

Figure 4.10 Eve Cone, situated near to Mt. Edziza in northern B.C., formed approximately 700 years ago [Wikipedia, http://en.wikipedia.org/wiki/Eve_Cone# mediaviewer/File:Symmetrical_Eve_Cone.jpg]

Composite Volcanoes

Composite volcanoes, like Mt. St. Helens in Washington State (Figure 4.11), are almost all associated with subduction at convergent plate boundaries — either ocean-continent or ocean-ocean boundaries (Figure 4.4b). They can extend up to several thousand metres from the surrounding terrain, and, with slopes ranging up to 30˚, are typically up to 10 km across.   At many such volcanoes, magma is stored in a magma chamber in the upper part of the crust. For example, at Mt. St. Helens, there is evidence of a magma chamber that is approximately 1 km wide and extends from about 6 km to 14 km below the surface (Figure 4.12). Systematic variations in the composition of volcanism over the past several thousand years at Mt. St. Helens imply that the magma chamber is zoned, from more felsic at the top to more mafic at the bottom.

Figure 4.11 The north side of Mt. St. Helens in southwestern Washington State, 2003 [SE photo]. The large 1980 eruption reduced the height of the volcano by 400 m, and a sector collapse removed a large part of the northern flank. Between 1980 and 1986 the slow eruption of more mafic and less viscous lava led to construction of a dome inside the crater.

Figure 4.12 A cross-section through the upper part of the crust at Mt. St. Helens showing the zoned magma chamber. [SE, after Pringle, 1993]

Mafic eruptions (and some intermediate eruptions), on the other hand, produce lava flows; the one shown in Figure 4.13b is thick enough (about 10 m in total) to have cooled in a columnar jointing pattern (Figure 4.14). Lava flows both flatten the profile of the volcano (because the lava typically flows farther than pyroclastic debris falls) and protect the fragmental deposits from erosion. Even so, composite volcanoes tend to erode quickly. Patrick Pringle, a volcanologist with the Washington State Department of Natural Resources, describes Mt. St. Helens as a “pile of junk.” The rock that makes up Mt. St. Helens ranges in composition from rhyolite (Figure 4.13a) to basalt (Figure 4.13b); this implies that the types of past eruptions have varied widely in character. As already noted, felsic magma doesn’t flow easily and doesn’t allow gases to escape easily. Under these circumstances, pressure builds up until a conduit opens, and then an explosive eruption results from the gas-rich upper part of the magma chamber, producing pyroclastic debris, as shown on Figure 4.13a. This type of eruption can also lead to rapid melting of ice and snow on a volcano, which typically triggers large mudflows known as lahars (Figure 4.13a). Hot, fast-moving pyroclastic flows and lahars are the two main causes of casualties in volcanic eruptions. Pyroclastic flows killed approximately 30,000 people during the 1902 eruption of Mt. Pelée on the Caribbean island of Martinique. Most were incinerated in their homes. In 1985 a massive lahar, triggered by the eruption of Nevado del Ruiz, killed 23,000 people in the Colombian town of Armero, about 50 km from the volcano.

In a geological context, composite volcanoes tend to form relatively quickly and do not last very long. Mt. St. Helens, for example, is made up of rock that is all younger than 40,000 years; most of it is younger than 3,000 years. If its volcanic activity ceases, it might erode away within a few tens of thousands of years. This is largely because of the presence of pyroclastic eruptive material, which is not strong.

Figure 4.13 Mt. St. Helens volcanic deposits: (a) lahar deposits (L) and felsic pyroclastic deposits (P) and (b) a columnar basalt lava flow. The two photos were taken at locations only about 500 m apart. [SE]

Figure 4.14 Figure 4.14 The development of columnar jointing in basalt, here seen from the top looking down. As the rock cools it shrinks, and because it is very homogenous it shrinks in a systematic way. When the rock breaks it does so with approximately 120˚ angles between the fracture planes. The resulting columns tend to be 6-sided but 5- and 7-sided columns also form. [SE]

Shield Volcanoes

Most shield volcanoes are associated with mantle plumes, although some form at divergent boundaries, either on land or on the sea floor. Because of their non-viscous mafic magma they tend to have relatively gentle slopes (2 to 10˚) and the larger ones can be over 100 km in diameter. The best-known shield volcanoes are those that make up the Hawaiian Islands, and of these, the only active ones are on the big island of Hawaii. Mauna Loa, the world’s largest volcano and the world’s largest mountain (by volume) last erupted in 1984. Kilauea, arguably the world’s most active volcano, has been erupting, virtually without interruption, since 1983. Loihi is an underwater volcano on the southeastern side of Hawaii. It is last known to have erupted in 1996, but may have erupted since then without being detected.

All of the Hawaiian volcanoes are related to the mantle plume that currently lies beneath Mauna Loa, Kilauea, and Loihi (Figure 4.15). In this area, the Pacific Plate is moving northwest at a rate of about 7 cm/year. This means that the earlier formed — and now extinct — volcanoes have now moved well away from the mantle plume. As shown on Figure 4.15, there is evidence of crustal magma chambers beneath all three active Hawaiian volcanoes. At Kilauea, the magma chamber appears to be several kilometres in diameter, and is situated between 8 km and 11 km below surface.Lin, G, Amelung, F, Lavallee, Y, and Okubo, P, 2014, Seismic evidence for a crustal magma reservoir beneath the upper east rift zone of Kilauea volcano, Hawaii. Geology. V.

Figure 4.15 Mauna Kea from near to the summit of Mauna Loa, Hawaii [http://upload.wikimedia.org/wikipedia/commons/f/f1/Hawaii_hotspot_cross-sectional_diagram.jpg]

Although it is not a prominent mountain (Figure 4.9), Kilauea volcano has a large caldera in its summit area (Figure 4.16). A caldera is a volcanic crater that is more than 2 km in diameter; this one is 4 km long and 3 km wide. It contains a smaller feature called Halema’uma’u crater, which has a total depth of over 200 m below the surrounding area. Most volcanic craters and calderas are formed above magma chambers, and the level of the crater floor is influenced by the amount of pressure exerted by the magma body. During historical times, the floors of both Kilauea caldera and Halema’uma’u crater have moved up during expansion of the magma chamber and down during deflation of the chamber.

Figure 4.16 Aerial view of the Kilauea caldera. The caldera is about 4 km across, and up to 120 m deep. It encloses a smaller and deeper crater known as Halema’uma’u. [http://upload.wikimedia.org/wikipedia/commons/b/b4/Kilauea_ali_2012_01_28.jpg]

One of the conspicuous features of Kilauea caldera is rising water vapour (the white cloud in Figure 4.16) and a strong smell of sulphur (Figure 4.17). As is typical in magmatic regions, water is the main volatile component, followed by carbon dioxide and sulphur dioxide. These, and some minor gases, originate from the magma chamber at depth and rise up through cracks in the overlying rock. This degassing of the magma is critical to the style of eruption at Kilauea, which, for most of the past 30 years, has been effusive, not explosive.

Figure 4.17 A gas-composition monitoring station (left) within the Kilauea caldera and at the edge of Halema’uma’u crater. The rising clouds are mostly composed of water vapour, but also include carbon dioxide and sulphur dioxide. Sulphur crystals (right) have formed around a gas vent in the caldera. [SE photos]

The Kilauea eruption that began in 1983 started with the formation of a cinder cone at Pu’u ’O’o, approximately 15 km east of the caldera (Figure 4.18). The magma feeding this eruption flowed along a major conduit system known as the East Rift, which extends for about 20 km from the caldera, first southeast and then east. Lava fountaining and construction of the Pu’u ’O’o cinder cone (Figure 4.19a) continued until 1986 at which time the flow became effusive. From 1986 to 2014, lava flowed from a gap in the southern flank of Pu’u ’O’o down the slope of Kilauea through a lava tube (Figure 4.19d), emerging at or near the ocean. Since June 2014, the lava has flowed northeast (see Exercise 4.4).

Figure 4.18 Satellite image of Kilauea volcano showing the East rift and Pu’u ’O’o, the site of the eruption that started in 1983. The puffy white blobs are clouds. [SE after, http://en.wikipedia.org/wiki/Hawaii_(island)#mediaviewer/File:Island_of_Hawai%27i_-_Landsat_mosaic.jpg]

The two main types of textures created during effusive subaerial eruptions are pahoehoe and aa. Pahoehoe, ropy lava that forms as non-viscous lava, flows gently, forming a skin that gels and then wrinkles because of ongoing flow of the lava below the surface (Figure 4.19b, and “lava flow video”). Aa, or blocky lava, forms when magma is forced to flow faster than it is able to (down a slope for example) (Figure 4.19c). Tephra (lava fragments) is produced during explosive eruptions, and accumulates in the vicinity of cinder cones.

Figure 4.19d is a view into an active lava tube on the southern edge of Kilauea. The red glow is from a stream of very hot lava (~1200°C) that has flowed underground for most of the 8 km from the Pu’u ’O’o vent. Lava tubes form naturally and readily on both shield and composite volcanoes because flowing mafic lava preferentially cools near its margins, forming solid lava levées that eventually close over the top of the flow. The magma within a lava tube is not exposed to the air, so it remains hot and fluid and can flow for tens of kilometres, thus contributing to the large size and low slopes of shield volcanoes. The Hawaiian volcanoes are riddled with thousands of old lava tubes, some as long as 50 km.

Figure 4.19 Images of Kilauea volcano taken in 2002 (b & c) and 2007 (a & d) [SE photos] (a) Pu’u’O’o cinder cone in the background with tephra in the foreground and aa lava in the middle, (b) Formation of pahoehoe on the southern edge of Kilauea, (c) Formation of aa on a steep slope on Kilauea, (d) Skylight in an active lava tube, Kilauea.

Kilauea is approximately 300 ka old, while neighbouring Mauna Loa is over 700 ka and Mauna Kea is over 1 Ma. If volcanism continues above the Hawaii mantle plume in the same manner that it has for the past 85 Ma, it is likely that Kilauea will continue to erupt for at least another 500,000 years. By that time, its neighbour, Loihi, will have emerged from the sea floor, and its other neighbours, Mauna Loa and Mauna Kea, will have become significantly eroded, like their cousins, the islands to the northwest (Figure 4.15).

Large Igneous Provinces

While the Hawaii mantle plume has produced a relatively low volume of magma for a very long time (~85 Ma), other mantle plumes are less consistent, and some generate massive volumes of magma over relatively short time periods. Although their origin is still controversial, it is thought that the volcanism leading to large igneous provinces (LIP) is related to very high volume but relatively short duration bursts of magma from mantle plumes. An example of an LIP is the Columbia River Basalt Group (CRGB), which extends across Washington, Oregon, and Idaho (Figure 4.20). This volcanism, which covered an area of about 160,000 km2 with basaltic rock up to several hundred metres thick, took place between 17 and 14 Ma.

Figure 4.20 A part of the Columbia River Basalt Group at Frenchman Coulee, eastern Washington. All of the flows visible here have formed large (up to two metres in diameter) columnar basalts, a result of relatively slow cooling of flows that are tens of m thick. The inset map shows the approximate extent of the 17 to 14 Ma Columbia River Basalts, with the location of the photo shown as a star. [SE – photo and drawing]

Most other LIP eruptions are much bigger. The Siberian Traps (also basalt), which erupted at the end of the Permian period at 250 Ma, are estimated to have produced approximately 40 times as much lava as the CRBG.

The mantle plume that is assumed to be responsible for the CRBG is now situated beneath the Yellowstone area, where it leads to felsic volcanism. Over the past 2 Ma three very large explosive eruptions at Yellowstone have yielded approximately 900 km³ of felsic magma, about 900 times the volume of the 1980 eruption of Mt. St. Helens, but only 5% of the volume of mafic magma in the CRBG.

Sea-Floor Volcanism

Some LIP eruptions occur on the sea floor, the largest being the one that created the Ontong Java plateau in the western Pacific Ocean at around 122 Ma. But most sea-floor volcanism originates at divergent boundaries and involves relatively low-volume eruptions. Under these conditions, hot lava that oozes out into the cold seawater quickly cools on the outside and then behaves a little like toothpaste. The resulting blobs of lava are known as pillows, and they tend to form piles around a sea-floor lava vent (Figure 4.21). In terms of area, there is very likely more pillow basalt on the sea floor than any other type of rock on Earth.

Figure 4.21 Modern and ancient sea-floor pillow basalts (left) Modern sea-floor pillows in the south Pacific [NOAA, from http://en.wikipedia.org/wiki/ Basalt#mediaviewer/File:Pillow_basalt_crop_l.jpg] (right) Eroded 40 to 50 Ma pillows on the shore of Vancouver Island, near to Sooke. The pillows are 30 to 40 cm in diameter. [SE]

Kimberlites

While all of the volcanism discussed so far is thought to originate from partial melting in the upper mantle or within the crust, there is a special class of volcanoes called kimberlites that have their origins much deeper in the mantle, at depths of 150 km to 450 km. During a kimberlite eruption, material from this depth may make its way to surface quickly (hours to days) with little interaction with the surrounding rocks. As a result, kimberlite eruptive material is representative of mantle compositions: it is ultramafic.

Kimberlite eruptions that originate at depths greater than 200 km, within areas beneath old thick crust (shields), traverse the region of stability of diamond in the mantle, and in some cases, bring diamond-bearing material to the surface. All of the diamond deposits on Earth are assumed to have formed in this way; an example is the rich Ekati Mine in the Northwest Territories (Figure 4.22).

Figure 4.22 Ekati diamond mine, Northwest Territories, part of the Lac de Gras kimberlite field [http://upload.wikimedia.org/wikipedia/commons/8/88/Ekati_mine_640px.jpg]

The kimberlites at Ekati erupted between 45 and 60 Ma. Many kimberlites are older, some much older. There have been no kimberlite eruptions in historic times. The youngest known kimberlites are in the Igwisi Hills in Tanzania and are only about 10,000 years old. The next youngest known are around 30 Ma old.

There are two classes of volcanic hazards, direct and indirect. Direct hazards are forces that directly kill or injure people, or destroy property or wildlife habitat. Indirect hazards are volcanism-induced environmental changes that lead to distress, famine, or habitat destruction. Indirect effects of volcanism have accounted for approximately 8 million deaths during historical times, while direct effects have accounted for fewer than 200,000, or 2.5% of the total. Some of the more important types of volcanic hazards are summarized in Table 4.2.

Table 4.2 A summary of the important volcanic hazards

Volcanic Gas and Tephra Emissions

Large volumes of tephra (rock fragments, mostly pumice) and gases are emitted during major plinian eruptions (large explosive eruptions with hot gas a tephra columns extending into the stratosphere) at composite volcanoes, and a large volume of gas is released during some very high-volume effusive eruptions. One of the major effects is cooling of the climate by 1° to 2°C for several months to a few years because the dust particles and tiny droplets and particles of sulphur compounds block the sun. The last significant event of this type was in 1991 and 1992 following the large eruption of Mt. Pinatubo in the Philippines. A drop of 1° to 2°C may not seem like very much, but that is the global average amount of cooling, and cooling was much more severe in some regions and at some times.

Over an eight-month period in 1783 and 1784, a massive effusive eruption took place at the Laki volcano in Iceland. Although there was relatively little volcanic ash involved, a massive amount of sulphur dioxide was released into the atmosphere, along with a significant volume of hydrofluoric acid (HF). The sulphate aerosols that formed in the atmosphere led to dramatic cooling in the northern hemisphere. There were serious crop failures in Europe and North America, and a total of 6 million people are estimated to have died from famine and respiratory complications. In Iceland, poisoning from the HF resulted in the death of 80% of sheep, 50% of cattle, and the ensuing famine, along with HF poisoning, resulted in more than 10,000 human deaths, about 25% of the population.

Volcanic ash can also have serious implications for aircraft because it can destroy jet engines. For example, over 5 million airline passengers had their travel disrupted by the 2010 Eyjafjallajökull volcanic eruption in Iceland.

Pyroclastic Density Currents

In a typical explosive eruption at a composite volcano, the tephra and gases are ejected with explosive force and are hot enough to be forced high up into the atmosphere. As the eruption proceeds, and the amount of gas in the rising magma starts to decrease, parts will become heavier than air, and they can then flow downward along the flanks of the volcano (Figure 4.23). As they descend, they cool more and flow faster, reaching speeds up to several hundred km/h. A pyroclastic density current (PDC) consists of tephra ranging in size from boulders to microscopic shards of glass (made up of the edges and junctions of the bubbles of shattered pumice), plus gases (dominated by water vapour, but also including other gases). The temperature of this material can be as high as 1000°C. Among the most famous PDCs are the one that destroyed Pompeii in the year 79 CE, killing an estimated 18,000 people, and the one that destroyed the town of St. Pierre, Martinique, in 1902, killing an estimated 30,000.

The buoyant upper parts of pyroclastic density currents can flow over water, in some cases for several kilometres. The 1902 St. Pierre PDC flowed out into the harbour and destroyed several wooden ships anchored there.

Figure 4.23 The plinian eruption of Mt. Mayon, Philippines. in 1984. Although most of the eruption column is ascending into the atmosphere, there are pyroclastic density currents flowing down the sides of the volcano in several places. [USGS photo from: http://upload.wikimedia.org/wikipedia/commons/7/73/Pyroclastic_flows_at_Mayon_Volcano.jpg]

Pyroclastic Fall

Most of the tephra from an explosive eruption ascends high into the atmosphere, and some of it is distributed around Earth by high-altitude winds. The larger components (larger than 0.1 mm) tend to fall relatively close to the volcano, and the amount produced by large eruptions can cause serious damage and casualties. The large 1991 eruption of Mt. Pinatubo in the Philippines resulted in the accumulation of tens of centimetres of ash in fields and on rooftops in the surrounding populated region. Heavy typhoon rains that hit the island at the same time added to the weight of the tephra, leading to the collapse of thousands of roofs and to at least 300 of the 700 deaths attributed to the eruption.

Lahar

A lahar is any mudflow or debris flow that is related to a volcano. Most are caused by melting snow and ice during an eruption, as was the case with the lahar that destroyed the Colombian town of Armero in 1985 (described earlier). Lahars can also happen when there is no volcanic eruption, and one of the reasons is that, as we’ve seen, composite volcanoes tend to be weak and easily eroded.

In October 1998, category 5 hurricane Mitch slammed into the coast of central America. Damage was extensive and 19,000 people died, not so much because of high winds but because of intense rainfall — some regions received almost 2 m of rain over a few days! Mudflows and debris flows occurred in many areas, especially in Honduras and Nicaragua. An example is Casita Volcano in Nicaragua, where the heavy rains weakened rock and volcanic debris on the upper slopes, resulting in a debris flow that rapidly built in volume as it raced down the steep slope, and then ripped through the towns of El Porvenir and Rolando Rodriguez killing more than 2,000 people (Figure 4.24). El Porvenir and Rolando Rodriguez were new towns that had been built without planning approval in an area that was known to be at risk of lahars.

Figure 4.24 Part of the path of the lahar from Casita Volcano, October 30, 1998. [USGS photo from: http://volcanoes.usgs.gov/hazards/lahar/casita.php]

Sector Collapse and Debris Avalanche

In the context of volcanoes, sector collapse or flank collapse is the catastrophic failure of a significant part of an existing volcano, creating a large debris avalanche. This hazard was first recognized with the failure of the north side of Mt. St. Helens immediately prior to the large eruption on May 18, 1980. In the weeks before the eruption, a large bulge had formed on the side of the volcano, the result of magma transfer from depth into a satellite magma body within the mountain itself. Early on the morning of May 18, a moderate earthquake struck nearby; this is thought to have destabilized the bulge, leading to Earth’s largest observed landslide in historical times. The failure of this part of the volcano exposed the underlying satellite magma chamber, causing it to explode sideways, which exposed the conduit leading to the magma chamber below. The resulting plinian eruption — with a 24 km high eruption column — lasted for nine hours.

In August 2010, a massive part of the flank of B.C.’s Mt. Meager gave way and about 48 million cubic metres of rock rushed down the valley, one of the largest slope failures in Canada in historical times (Figure 4.25). More than 25 slope failures have taken place at Mt. Meager in the past 8,000 years, some of them more than 10 times larger than the 2010 failure.

Figure 4.25 The August 2010 Mt. Meager rock avalanche, showing where the slide originated (arrow, 4 km upstream), its path down a steep narrow valley, and the debris field (and the stream that eventually cut through it) in the foreground.   (Mika McKinnon photo, http://www.geomika.com/blog/2011/ 01/05/the-trouble-with-landslides/ Used with permission (mika@geomika.com)

Lava Flows

As we saw in Exercise 4.4, lava flows at volcanoes like Kilauea do not advance very quickly, and in most cases, people can get out of the way. Of course, it is more difficult to move infrastructure, and so buildings and roads are typically the main casualties of lava flows.

Exercises

Exercise 4.5 Volcanic Hazards in Squamish

The town of Squamish is situated approximately 10 km from Mt. Garibaldi, as shown in the photo. In the event of a major eruption of Mt. Garibaldi, which of the following hazards has the potential to be an issue for the residents of Squamish or for those passing through on Highway 99? [SE after Google Earth]

Hazard	Yes or No, and Brief Explanation
Tephra emission
Gas emission
Pyroclastic density current
Pyroclastic fall
Lahar
Sector collapse
Lava flow

In 2005 USGS geologist Chris Newhall made a list of the six most important signs of an imminent volcanic eruption. They are as follows:

1. Gas leaks — the release of gases (mostly H₂O, CO₂, and SO₂) from the magma into the atmosphere through cracks in the overlying rock
2. Bit of a bulge — the deformation of part of the volcano, indicating that a magma chamber at depth is swelling or becoming more pressurized
3. Getting shaky — many (hundreds to thousands) of small earthquakes, indicating that magma is on the move. The quakes may be the result of the magma forcing the surrounding rocks to crack, or a harmonic vibration that is evidence of magmatic fluids moving underground.
4. Dropping fast — a sudden decrease in the rate of seismicity, which may indicate that magma has stalled, which could mean that something is about to give way
5. Big bump — a pronounced bulge on the side of the volcano (like the one at Mt. St. Helens in 1980), which may indicate that magma has moved close to surface
6. Blowing off steam — steam eruptions (a.k.a. phreatic eruptions) that happen when magma near the surface heats groundwater to the boiling point. The water eventually explodes, sending fragments of the overlying rock far into the air.

With these signs in mind, we can make a list of the equipment we should have and the actions we can take to monitor a volcano and predict when it might erupt.

Assessing seismicity: The simplest and cheapest way to monitor a volcano is with seismometers. In an area with several volcanoes that have the potential to erupt (e.g., the Squamish-Pemberton area), a few well-placed seismometers can provide us with an early warning that something is changing beneath one of the volcanoes, and that we need to take a closer look. There are currently enough seismometers in the Lower Mainland and on Vancouver Island to provide this information.See: http://www.earthquakescanada.nrcan.gc.ca/stndon/CNSN-RNSC/stnbook-cahierstn/index-eng.php?tpl_sorting=map&CHIS_SZ=west

If there is seismic evidence that a volcano is coming to life, more seismometers should be placed in locations within a few tens of kilometres of the source of the activity (Figure 4.26). This will allow geologists to determine the exact location and depth of the seismic activity so that they can see where the magma is moving.

Figure 4.26 A seismometer installed in 2007 in the vicinity of the Nazco Cone, B.C. [photo Cathie Hickson, used with permission]

Detecting gases: Water vapour quickly turns into clouds of liquid water droplets and is relatively easy to detect just by looking, but CO₂ and SO₂ are not as obvious. It’s important to be able to monitor changes in the composition of volcanic gases, and we need instruments to do that. Some can be monitored from a distance (from the ground or even from the air) using infrared devices, but to obtain more accurate data, we need to sample the air and do chemical analysis. This can be achieved with instruments placed on the ground close to the source of the gases (see Figure 4.17), or by collecting samples of the air and analyzing them in a lab.

Measuring deformation: There are two main ways to measure ground deformation at a volcano. One is known as a tiltmeter, which is a sensitive three-directional level that can sense small changes in the tilt of the ground at a specific location. Another is through the use of GPS (global positioning system) technology (Figure 4.27). GPS is more effective than a tiltmeter because it provides information on how far the ground has actually moved — east-west, north-south, and up-down.

Figure 4.27 A GPS unit installed at Hualalai volcano, Hawaii. The dish-shaped antenna on the right is the GPS receiver. The antenna on the left is for communication with a base station. [from USGS at: http://hvo.wr.usgs.gov/volcanowatch/view.php?id=173]

By combining information from these types of sources, along with careful observations made on the ground and from the air, and a thorough knowledge of how volcanoes work, geologists can get a good idea of the potential for a volcano to erupt in the near future (months to weeks, but not days). They can then make recommendations to authorities about the need for evacuations and restricting transportation corridors. Our ability to predict volcanic eruptions has increased dramatically in recent decades because of advances in our understanding of how volcanoes behave and in monitoring technology. Providing that careful work is done, there is no longer a large risk of surprise eruptions, and providing that public warnings are issued and heeded, it is less and less likely that thousands will die from sector collapse, pyroclastic flows, ash falls, or lahars. Indirect hazards are still very real, however, and we can expect the next eruption like the one at Laki in 1783 to take an even greater toll than it did then, especially since there are now roughly eight times as many people on Earth.

As shown on the Figure 4.28, three types of volcanic environments are represented in British Columbia:

• The Cascade Arc (a.k.a. the Garibaldi Volcanic Belt in Canada) is related to subduction of the Juan de Fuca Plate beneath the North America plate.
• The Anahim Volcanic Belt is assumed to be related to a mantle plume.
• The Stikine Volcanic Belt and the Wells Gray-Clearwater Volcanic Field are assumed to be related to crustal rifting.
  

  Figure 4.28 Major volcanic centres in British Columbia (base map from Wikipedia (http://commons.wikimedia.org/wiki/File:South-West_Canada.jpg). Volcanic locations from Wood, D., 1993, Waiting for another big blast – probing B.C.’s volcanoes, Canadian Geographic, based on the work of Cathie Hickson)

Subduction Volcanism

Southwestern British Columbia is at the northern end of the Juan de Fuca (Cascadia) subduction zone, and the volcanism there is related to magma generation by flux melting in the upper mantle above the subducting plate. In general, there has been a much lower rate and volume of volcanism in the B.C. part of this belt than in the U.S. part. One reason for this is that the northern part of the Juan de Fuca Plate (i.e., the Explorer Plate) is either not subducting, or is subducting at a slower rate than the rest of the plate. There are several volcanic centres in the Garibaldi Volcanic Belt: the Garibaldi centre (including Mt. Garibaldi and the Black Tusk-Mt. Price area adjacent to Garibaldi Lake (Figures 4.1 and 4.2), Mt. Cayley, and Mt. Meager (Figure 4.25). The most recent volcanic activity in this area was at Mt. Meager. Approximately 2,400 years ago, an explosive eruption of about the same magnitude as the 1980 Mt. St. Helens eruption took place at Mt. Meager. Ash spread as far east as Alberta. There was also significant eruptive activity at Mts. Price and Garibaldi approximately 12,000 and 10,000 years ago during the last glaciation; in both cases, lava and tephra built up against glacial ice in the adjacent valley (Figure 4.29). The Table in Figure 4.2 at the beginning of this chapter is a tuya, a volcano that formed beneath glacial ice and had its top eroded by the lake that formed around it in the ice.

Figure 4.29 Perspective view of the Garibaldi region (looking east) showing the outlines of two lava flows from Mt. Price. Volcanism in this area last took place when the valley in the foreground was filled with glacial ice. The cliff known as the Barrier formed when part of the Mt. Price lava flow failed after deglaciation. The steep western face of Mt. Garibaldi formed by sector collapse, also because the rocks were no longer supported by glacial ice. [SE after Google Earth]

Mantle Plume Volcanism

The chain of volcanic complexes and cones extending from Milbanke Sound to Nazko Cone is interpreted as being related to a mantle plume currently situated close to the Nazko Cone, just west of Quesnel. The North America Plate is moving in a westerly direction at about 2 cm per year with respect to this plume, and the series of now partly eroded shield volcanoes between Nazco and the coast is interpreted to have been formed by the plume as the continent moved over it.

The Rainbow Range, which formed at approximately 8 Ma, is the largest of these older volcanoes. It has a diameter of about 30 km and an elevation of 2,495 m (Figure 4.30). The name “Rainbow” refers to the bright colours displayed by some of the volcanic rocks as they weather.

Figure 4.30 Rainbow Range, Chilcotin Plateau, B.C. (http://upload.wikimedia.org/wikipedia/commons/f/fd/Rainbow_Range_Colors.jpg).

Rift-Related Volcanism

While B.C. is not about to split into pieces, two areas of volcanism are related to rifting — or at least to stretching-related fractures that might extend through the crust. These are the Wells Gray-Clearwater volcanic field southeast of Quesnel, and the Northern Cordillera Volcanic Field, which ranges across the northwestern corner of the province (as already discussed in section 4.1). This area includes Canada’s most recent volcanic eruption, a cinder cone and mafic lava flow that formed around 250 years ago at the Tseax River Cone in the Nass River area north of Terrace.  According to Nisga’a oral history, as many as 2,000 people died during that eruption, in which lava overran their village on the Nass River. Most of the deaths are attributed to asphyxiation from volcanic gases, probably carbon dioxide.

The Mount Edziza Volcanic Field near the Stikine River is a large area of lava flows, sulphurous ridges, and cinder cones. The most recent eruption in this area was about 1,000 years ago. While most of the other volcanism in the Edziza region is mafic and involves lava flows and cinder cones, Mt. Edziza itself (Figure 4.31) is a composite volcano with rock compositions ranging from rhyolite to basalt. A possible explanation for the presence of composite volcanism in an area dominated by mafic flows and cinder cones is that there is a magma chamber beneath this area, within which magma differentiation is taking place.

Figure 4.31 Mount Edziza, in the Stikine area, B.C., with Eve Cone in the foreground. (http://upload.wikimedia.org/wikipedia/commons/5/54/Mount_Edziza%2C_British_Columbia.jpg).

The topics covered in this chapter can be summarized as follows:

Introduction

Weathering is what takes place when a body of rock is exposed to the “weather” — in other words, to the forces and conditions that exist at Earth’s surface. With the exception of volcanic rocks and some sedimentary rocks, most rocks are formed at some depth within the crust. There they experience relatively constant temperature, high pressure, no contact with the atmosphere, and little or no moving water. Once a rock is exposed at the surface, which is what happens when the overlying rock is eroded away, conditions change dramatically. Temperatures vary widely, there is much less pressure, oxygen and other gases are plentiful, and in most climates, water is abundant (Figure 5.1).

Weathering includes two main processes that are quite different. One is the mechanical breakdown of rock into smaller fragments, and the other is the chemical change of the minerals within the rock to forms that are stable in the surface environment. Mechanical weathering provides fresh surfaces for attack by chemical processes, and chemical weathering weakens the rock so that it is more susceptible to mechanical weathering. Together, these processes create two very important products, one being the sedimentary clasts and ions in solution that can eventually become sedimentary rock, and the other being the soil that is necessary for our existence on Earth.

The various processes related to uplift and weathering are summarized in the rock cycle in Figure 5.2.

Figure 5.2 Weathering can take place once a rock is exposed at surface by uplift and the removal of the overlying rock. [SE]

Intrusive igneous rocks form at depths of several hundreds of metres to several tens of kilometres. Sediments are turned into sedimentary rocks only when they are buried by other sediments to depths in excess of several hundreds of metres. Most metamorphic rocks are formed at depths of kilometres to tens of kilometres. Weathering cannot even begin until these rocks are uplifted through various processes of mountain building — most of which are related to plate tectonics — and the overlying material has been eroded away and the rock is exposed as an outcrop.To a geologist, an outcrop is an exposure of bedrock, the solid rock of the crust.

The important agents of mechanical weathering are:

• The decrease in pressure that results from removal of overlying rock
• Freezing and thawing of water in cracks in the rock
• Formation of salt crystals within the rock
• Cracking from plant roots and exposure by burrowing animals

When a mass of rock is exposed by weathering and removal of the overlying rock, there is a decrease in the confining pressure on the rock, and the rock expands. This unloading promotes cracking of the rock, known as exfoliation, as shown in the granitic rock in Figure 5.3.

Figure 5.3 Exfoliation fractures in granitic rock exposed on the west side of the Coquihalla Highway north of Hope, B.C. [SE]

Figure 5.4 Exfoliation of slate at a road cut in the Columbia Mountains west of Golden, B.C. [SE photo]

Granitic rock tends to exfoliate parallel to the exposed surface because the rock is typically homogenous, and it doesn’t have predetermined planes along which it must fracture. Sedimentary and metamorphic rocks, on the other hand, tend to exfoliate along predetermined planes (Figure 5.4).

Frost wedging is the process by which water seeps into cracks in a rock, expands on freezing, and thus enlarges the cracks (Figure 5.5). The effectiveness of frost wedging is related to the frequency of freezing and thawing. Frost wedging is most effective in a climate like Canada’s. In warm areas where freezing is infrequent, in very cold areas where thawing is infrequent, or in very dry areas, where there is little water to seep into cracks, the role of frost wedging is limited.

Figure 5.5 The process of frost wedging on a steep slope. Water gets into fractures and then freezes, expanding the fracture a little. When the water thaws it seeps a little farther into the expanded crack. The process is repeated many times, and eventually a piece of rock will be wedged away. [SE]

In many parts of Canada, the transition between freezing nighttime temperatures and thawing daytime temperatures is frequent — tens to hundreds of times a year. Even in warm coastal areas of southern B.C., freezing and thawing transitions are common at higher elevations. A common feature in areas of effective frost wedging is a talus slope — a fan-shaped deposit of fragments removed by frost wedging from the steep rocky slopes above (Figure 5.6).

Figure 5.6 An area with very effective frost-wedging near Keremeos, B.C. The fragments that have been wedged away from the cliffs above have accumulated in a talus deposit at the base of the slope. The rocks in this area have quite varied colours, and those are reflected in the colours of the talus. [SE]

A related process, frost heaving, takes place within unconsolidated materials on gentle slopes. In this case, water in the soil freezes and expands, pushing the overlying material up. Frost heaving is responsible for winter damage to roads all over North America.

When salt water seeps into rocks and then evaporates on a hot sunny day, salt crystals grow within cracks and pores in the rock. The growth of these crystals exerts pressure on the rock and can push grains apart, causing the rock to weaken and break. There are many examples of this on the rocky shorelines of Vancouver Island and the Gulf Islands, where sandstone outcrops are common and salty seawater is readily available (Figure 5.7). Salt weathering can also occur away from the coast, because most environments have some salt in them.

Figure 5.7 Honeycomb weathering of sandstone on Gabriola Island, B.C. The holes are caused by crystallization of salt within rock pores, and the seemingly regular pattern is related to the original roughness of the surface. It’s a positive-feedback process because the holes collect salt water at high tide, and so the effect is accentuated around existing holes. This type of weathering is most pronounced on south-facing sunny exposures. [SE]

The effects of plants and animals are significant in mechanical weathering. Roots can force their way into even the tiniest cracks, and then they exert tremendous pressure on the rocks as they grow, widening the cracks and breaking the rock (Figure 5.8). Although animals do not normally burrow through solid rock, they can excavate and remove huge volumes of soil, and thus expose the rock to weathering by other mechanisms.

Figure 5.8 Conifers growing on granitic rocks at The Lions, near Vancouver, B.C. [SE]

Mechanical weathering is greatly facilitated by erosion, which is the removal of weathering products, allowing for the exposure of more rock for weathering. A good example of this is shown in Figure 5.6. On the steep rock faces at the top of the cliff, rock fragments have been broken off by ice wedging, and then removed by gravity. This is a form of mass wasting, which is discussed in more detail in Chapter 15. Other important agents of erosion that also have the effect of removing the products of weathering include water in streams (Chapter 13), ice in glaciers (Chapter 16), and waves on the coasts (Chapter 17).

Chemical weathering results from chemical changes to minerals that become unstable when they are exposed to surface conditions. The kinds of changes that take place are highly specific to the mineral and the environmental conditions. Some minerals, like quartz, are virtually unaffected by chemical weathering, while others, like feldspar, are easily altered. In general, the degree of chemical weathering is greatest in warm and wet climates, and least in cold and dry climates. The important characteristics of surface conditions that lead to chemical weathering are the presence of water (in the air and on the ground surface), the abundance of oxygen, and the presence of carbon dioxide, which produces weak carbonic acid when combined with water. That process, which is fundamental to most chemical weathering, can be shown as follows:

H₂O + CO₂ —->H₂CO₃    then   H₂CO₃ —-> H⁺     + HCO₃^–,

water + carbon dioxide —-> carbonic acid   then    carbonic acid  —-> hydrgen ion + carbonate ion

Here we have water (e.g., as rain) plus carbon dioxide in the atmosphere, combining to create carbonic acid. Then carbonic acid dissociates (comes apart) to form hydrogen and carbonate ions. The amount of CO₂ in the air is enough to make only very weak carbonic acid, but there is typically much more CO₂ in the soil, so water that percolates through the soil can become significantly more acidic.

There are two main types of chemical weathering. On the one hand, some minerals become altered to other minerals. For example, feldspar is altered — by hydrolysis — to clay minerals. On the other hand, some minerals dissolve completely, and their components go into solution. For example, calcite (CaCO₃) is soluble in acidic solutions.

The hydrolysis of feldspar can be written like this:

CaAl₂Si₂O₈  + H₂CO₃  + ½O₂ —-> Al₂Si₂O₅(OH)₄ +   Ca²⁺ +  CO₃^2-

plagioclase + carbonic acid —-> kaolinite + dissolved calcium + carbonate ions

This reaction shows calcium plagioclase feldspar, but similar reactions could also be written for sodium or potassium feldspars. In this case, we end up with the mineral kaolinite, along with calcium and carbonate ions in solution. Those ions can eventually combine (probably in the ocean) to form the mineral calcite. The hydrolysis of feldspar to clay is illustrated in Figure 5.9, which shows two images of the same granitic rock, a recently broken fresh surface on the left and a clay-altered weathered surface on the right. Other silicate minerals can also go through hydrolysis, although the end results will be a little different. For example, pyroxene can be converted to the clay minerals chlorite or smectite, and olivine can be converted to the clay mineral serpentine.

Figure 5.9 Unweathered (left) and weathered (right) surfaces of the same piece of granitic rock. On the unweathered surfaces the feldspars are still fresh and glassy-looking. On the weathered surface the feldspar has been altered to the chalky-looking clay mineral kaolinite. [SE]

Oxidation is another very important chemical weathering process. The oxidation of the iron in a ferromagnesian silicate starts with the dissolution of the iron. For olivine, the process looks like this, where olivine in the presence of carbonic acid is converted to dissolved iron, carbonate, and silicic acid:

Fe₂SiO₄+ 4H₂CO₃ —> 2Fe^₂+   +  4HCO₃^–       +  H₄SiO₄

olivine + (carbonic acid) —> dissolved iron + dissolved carbonate + dissolved silicic acid

In the presence of oxygen, the dissolved iron is then quickly converted to hematite:

2Fe^₂+  + 4HCO₃– + ½ O₂     +  2H₂O —->Fe₂O₃   + 4H₂CO₃

dissolved iron + bicarbonate + oxygen + water—->hematite + carbonic acid

The equation shown here is for olivine, but it could apply to almost any other ferromagnesian silicate, including pyroxene, amphibole, or biotite. Iron in the sulphide minerals (e.g., pyrite) can also be oxidized in this way. And the mineral hematite is not the only possible end result, as there is a wide range of iron oxide minerals that can form in this way. The results of this process are illustrated in Figure 5.10, which shows a granitic rock in which some of the biotite and amphibole have been altered to form the iron oxide mineral limonite.

Figure 5.10 A granitic rock containing biotite and amphibole which have been altered near to the rock’s surface to limonite, which is a mixture of iron oxide minerals. [SE]

A special type of oxidation takes place in areas where the rocks have elevated levels of sulphide minerals, especially pyrite (FeS₂). Pyrite reacts with water and oxygen to form sulphuric acid, as follows:

2FeS₂  + 7O₂ +2H₂O —–> 2Fe²⁺   H₂SO₄+ 2H⁺

pyrite + oxygen + water —–> iron ions + sulphuric acid + hydrogen ions

The runoff from areas where this process is taking place is known as acid rock drainage (ARD), and even a rock with 1% or 2% pyrite can produce significant ARD. Some of the worst examples of ARD are at metal mine sites, especially where pyrite-bearing rock and waste material have been mined from deep underground and then piled up and left exposed to water and oxygen. One example of that is the Mt. Washington Mine near Courtenay on Vancouver Island (Figure 5.11), but there are many similar sites across Canada and around the world.

Figure 5.11 Exposed oxidizing and acid generating rocks and mine waste at the abandoned Mt. Washington Mine, B.C. (left), and an example of acid drainage downstream from the mine site (right). [SE]

At many ARD sites, the pH of the runoff water is less than 4 (very acidic). Under these conditions, metals such as copper, zinc, and lead are quite soluble, which can lead to toxicity for aquatic and other organisms. For many years, the river downstream from the Mt. Washington Mine had so much dissolved copper in it that it was toxic to salmon. Remediation work has since been carried out at the mine and the situation has improved.

The hydrolysis of feldspar and other silicate minerals and the oxidation of iron in ferromagnesian silicates all serve to create rocks that are softer and weaker than they were to begin with, and thus more susceptible to mechanical weathering.

The weathering reactions that we’ve discussed so far involved the transformation of one mineral to another mineral (e.g., feldspar to clay), and the release of some ions in solution (e.g., Ca²⁺). Some weathering processes involve the complete dissolution of a mineral. Calcite, for example, will dissolve in weak acid, to produce calcium and bicarbonate ions. The equation is as follows:

CaCO₃  + H⁺   + HCO₃^–  —–>   Ca²⁺  + 2HCO₃^–

calcite + hydrogen ions + bicarbonate —–>  calcium ions + bicarbonate

Calcite is the major component of limestone (typically more than 95%), and under surface conditions, limestone will dissolve to varying degrees (depending on which minerals it contains, other than calcite), as shown in Figure 5.12. Limestone also dissolves at relatively shallow depths underground, forming limestone caves. This is discussed in more detail in Chapter 14, where we look at groundwater.

Figure 5.12 A limestone outcrop on Quadra Island, B.C. The limestone, which is primarily made up of the mineral calcite, has been dissolved to different degrees in different areas because of compositional differences. The buff-coloured bands are volcanic rock, which is not soluble. [SE]

The products of weathering and erosion are the unconsolidated materials that we find around us on slopes, beneath glaciers, in stream valleys, on beaches, and in deserts. The nature of these materials — their composition, size, degree of sorting, and degree of rounding — is determined by the type of rock that is being weathered, the nature of the weathering, the erosion and transportation processes, and the climate.

A summary of the weathering products of some of the common minerals present in rocks is provided in Table 5.1.

Table 5.1 A list of the typical weathering products of some of the minerals in common rocks [SE]

Some examples of the products of weathering are shown in Figure 5.13. They range widely in size and shape depending on the processes involved. If and when deposits like these are turned into sedimentary rocks, the textures of those rocks will vary significantly. Importantly, when we describe sedimentary rocks that formed millions of years in the past, we can use those properties to make inferences about the conditions that existed during their formation.

We’ll talk more about the nature and interpretation of sediments and sedimentary rocks in Chapter 6, but it’s worth considering here why the sandy sediments shown in Figure 5.13 are so strongly dominated by the mineral quartz, even though quartz makes up less than 20% of Earth’s crust. The explanation is that quartz is highly resistant to the types of weathering that occur at Earth’s surface. It is not affected by weak acids or the presence of oxygen. This makes it unique among the minerals that are common in igneous rocks. Quartz is also very hard, and doesn’t have cleavage, so it is resistant to mechanical erosion.

As weathering proceeds, the ferromagnesian silicates and feldspar are very likely to be broken into small pieces and converted into clay minerals and dissolved ions (e.g., Ca²⁺, Na⁺, K⁺, Fe²⁺, Mg²⁺, and H₄SiO₄). In other words, quartz, clay minerals, and dissolved ions are the most common products of weathering. Quartz and some of the clay minerals tend to form sedimentary deposits on and at the edges of continents, while the rest of the clay minerals and the dissolved ions tend to be washed out into the oceans to form sediments on the sea floor.

Figure 5.13 Products of weathering and erosion formed under different conditions. [SE]

Exercises

Exercise 5.3 Describing the Weathering Origins of Sands

In the left side of the following table, a number of different sands are illustrated and. On the right side, describe some of the important weathering processes that might have led to the development of these sands. [SE photos]

Image	Description and Location	Possible Weathering Processes
	Fragments of coral, algae, and urchin from a shallow water area (~2 m depth) near a reef in Belize. The grains are between 0.1 and 1 mm.
	Angular quartz and rock fragments from a glacial stream deposit near Osoyoos, B.C. The grains are between 0.25 and 0.5 mm across.
	Rounded grains of olivine (green) and volcanic glass (black) from a beach on the big island of Hawaii. The grains are approximately 1 mm across.

Weathering is a key part of the process of soil formation, and soil is critical to our existence on Earth. In other words, we owe our existence to weathering, and we need to take care of our soil!

Many people refer to any loose material on Earth’s surface as soil, but to geologists (and geology students) soil is the material that includes organic matter, lies within the top few tens of centimetres of the surface, and is important in sustaining plant growth.

Soil is a complex mixture of minerals (approximately 45%), organic matter (approximately 5%), and empty space (approximately 50%, filled to varying degrees with air and water). The mineral content of soils is variable, but is dominated by clay minerals and quartz, along with minor amounts of feldspar and small fragments of rock. The types of weathering that take place within a region have a major influence on soil composition and texture. For example, in a warm climate, where chemical weathering dominates, soils tend to be richer in clay. Soil scientists describe soil texture in terms of the relative proportions of sand, silt, and clay, as shown in Figure 5.14. The sand and silt components in this diagram are dominated by quartz, with lesser amounts of feldspar and rock fragments, while the clay component is dominated by the clay minerals.

Figure 5.14 The U.S. Department of Agriculture soil texture diagram. This diagram applies only to the mineral component of soils, and the names are textural descriptions, not soil classes. [http://en.wikipedia.org/wiki/ Soil#media viewer/File:SoilTexture_USDA.png]

Soil forms through accumulation and decay of organic matter and through the mechanical and chemical weathering processes described above. The factors that affect the nature of soil and the rate of its formation include climate (especially average temperature and precipitation amounts, and the consequent types of vegetation), the type of parent material, the slope of the surface, and the amount of time available.

Climate

Soils develop because of the weathering of materials on Earth’s surface, including the mechanical breakup of rocks, and the chemical weathering of minerals. Soil development is facilitated by the downward percolation of water. Soil forms most readily under temperate to tropical conditions (not cold) and where precipitation amounts are moderate (not dry, but not too wet). Chemical weathering reactions (especially the formation of clay minerals) and biochemical reactions proceed fastest under warm conditions, and plant growth is enhanced in warm climates. Too much water (e.g., in rainforests) can lead to the leaching of important chemical nutrients and hence to acidic soils. In humid and poorly drained regions, swampy conditions may prevail, producing soil that is dominated by organic matter. Too little water (e.g., in deserts and semi-deserts), results in very limited downward chemical transportation and the accumulation of salts and carbonate minerals (e.g., calcite) from upward-moving water. Soils in dry regions also suffer from a lack of organic material (Figure 5.15).

Figure 5.15 Poorly developed soil on wind-blown silt (loess) in an arid part of northeastern Washington State [SE]

Parent Material

Soil parent materials can include all different types of bedrock and any type of unconsolidated sediments, such as glacial deposits and stream deposits. Soils are described as residual soils if they develop on bedrock, and transported soils if they develop on transported material such as glacial sediments. But the term “transported soil” is misleading because it implies that the soil itself has been transported, which is not the case. When referring to such soil, it is better to be specific and say “soil developed on unconsolidated material,” because that distinguishes it from soil developed on bedrock.

Quartz-rich parent material, such as granite, sandstone, or loose sand, leads to the development of sandy soils. Quartz-poor material, such as shale or basalt, generates soils with little sand.

Parent materials provide important nutrients to residual soils. For example, a minor constituent of granitic rocks is the calcium-phosphate mineral apatite, which is a source of the important soil nutrient phosphorus. Basaltic parent material tends to generate very fertile soils because it also provides phosphorus, along with significant amounts of iron, magnesium, and calcium.

Some unconsolidated materials, such as river-flood deposits, make for especially good soils because they tend to be rich in clay minerals. Clay minerals have large surface areas with negative charges that are attractive to positively charged elements like calcium, magnesium, iron, and potassium — important nutrients for plant growth.

Slope

Soil can only develop where surface materials remain in place and are not frequently moved away by mass wasting. Soils cannot develop where the rate of soil formation is less than the rate of erosion, so steep slopes tend to have little or no soil.

Time

Even under ideal conditions, soil takes thousands of years to develop. Virtually all of southern Canada was still glaciated up until 14 ka, and most of the central and northern parts of B.C., the prairies, Ontario, and Quebec were still glaciated at 12 ka. Glaciers still dominated the central and northern parts of Canada until around 10 ka, and so, at that time, conditions were still not ideal for soil development even in the southern regions. Therefore, soils in Canada, and especially in central and northern Canada, are relatively young and not well developed.

The same applies to soils that are forming on newly created surfaces, such as recent deltas or sand bars, or in areas of mass wasting.

Soil Horizons

The process of soil formation generally involves the downward movement of clay, water, and dissolved ions, and a common result of that is the development of chemically and texturally different layers known as soil horizons. The typically developed soil horizons, as illustrated in Figure 5.16, are:

O — the layer of organic matter

A — the layer of partially decayed organic matter mixed with mineral material

E— the eluviated (leached) layer from which some of the clay and iron have been removed to create a pale layer that may be sandier than the other layers

B — the layer of accumulation of clay, iron, and other elements from the overlying soil

C — the layer of incomplete weathering

Although rare in Canada, another type of layer that develops in hot arid regions is known as caliche (pronounced ca-lee-chee). It forms from the downward (or in some cases upward) movement of calcium ions, and the precipitation of calcite within the soil. When well developed, caliche cements the surrounding material together to form a layer that has the consistency of concrete.

Figure 5.16 Soil horizons in a podsol from a site in northeastern Scotland. O: organic matter A: organic matter and mineral material E: leached layer B: accumulation of clay, iron etc. C: incomplete weathering of parent material [SE after http://commons.wikimedia.org/wiki/File:Podzol_-_geograph.org.uk_-_218892.jpg]

Like all geological materials, soil is subject to erosion, although under natural conditions on gentle slopes, the rate of soil formation either balances or exceeds the rate of erosion. Human practices related to forestry and agriculture have significantly upset this balance. 

Soils are held in place by vegetation. When vegetation is removed, either through cutting trees or routinely harvesting crops and tilling the soil, that protection is either temporarily or permanently lost. The primary agents of the erosion of unprotected soil are water and wind.

Water erosion is accentuated on sloped surfaces because fast-flowing water obviously has greater eroding power than still water (Figure 5.17). Raindrops can disaggregate exposed soil particles, putting the finer material (e.g., clays) into suspension in the water. Sheetwash, unchannelled flow across a surface carries suspended material away, and channels erode right through the soil layer, removing both fine and coarse material.

Figure 5.17 Soil erosion by rain and channelled runoff on a field in Alberta. [from Alberta Agriculture and Rural Development, http://www1.agric.gov.ab.ca/$department/deptdocs.nsf/all/agdex9313, used with permission]

Wind erosion is exacerbated by the removal of trees that act as wind breaks and by agricultural practices that leave bare soil exposed (Figure 5.18).

Tillage is also a factor in soil erosion, especially on slopes, because each time the soil is lifted by a cultivator, it is moved a few centimetres down the slope.

Figure 5.18 Soil erosion by wind in Alberta. [from Alberta Agriculture and Rural Development, http://www1.agric.gov.ab.ca/$department/deptdocs.nsf/all/agdex9313, used with permission]

Up until the 1950s, the classification of soils in Canada was based on the system used in the United States. However, it was long recognized that the U.S .system did not apply well to many parts of Canada because of climate and environmental differences. The Canadian System of Soil Classification was first outlined in 1955 and has been refined and modified numerous times since then.

There are 10 orders of soil recognized in Canada. Each one is divided into groups, and then families, and then series, but we will only look at the orders, some of which are summarized in Table 5.2. The distribution of these types of soils (and a few others) in Canada is shown in Figure 5.19.

Table 5.2 The nature, origins and distributions of the more important soil orders in Canada

There is an excellent website on Canadian soils, with videos describing the origins and characteristics of the soils, at: http://soilweb.landfood.ubc.ca/classification/.

As we’ve discussed, the processes of soil formation are dominated by the downward transportation of clays and certain elements, and the nature of those processes depends in large part on the climate. In Canada’s predominantly cool and humid climate (which applies to most places other than the far north), podsolization is the norm. This involves downward transportation of hydrogen, iron, and aluminum (and other elements) from the upper part of the soil profile, and accumulation of clay, iron, and aluminum in the B horizon. Most of the podsols, luvisols, and brunisols of Canada form through various types of podsolization.

Figure 5.19 The soil order map of Canada. [from The Department of Soil Science, University of Saskatchewan, http://www.soilsofcanada.ca/ used with permission]

In the grasslands of the dry southern parts of the prairie provinces and in some of the drier parts of southern B.C., dark brown organic-rich chernozem soils are dominant. In some parts of these areas, weak calcification takes place with leaching of calcium from the upper layers and accumulation of calcium in the B layer. Development of caliche layers is rare in Canada.

Organic soils form in areas with poor drainage (i.e., swamps) and a rich supply of organic matter. These soils have very little mineral matter.

In the permafrost regions of the north, where glacial retreat was most recent, the time available for soil formation has been short and the rate of soil formation is very slow. The soils are called cryosols (cryo means “ice cold”). Permafrost areas are also characterized by the churning of the soil by freeze-thaw processes, and as a result, development of soil horizons is very limited.

Earth has two important carbon cycles. One is the biological one, wherein living organisms — mostly plants — consume carbon dioxide from the atmosphere to make their tissues, and then, after they die, that carbon is released back into the atmosphere when they decay over a period of years or decades. A small proportion of this biological-cycle carbon becomes buried in sedimentary rocks: during the slow formation of coal, as tiny fragments and molecules in organic-rich shale, and as the shells and other parts of marine organisms in limestone. This then becomes part of the geological carbon cycle, a cycle that actually involves a majority of Earth’s carbon, but one that operates only very slowly.

The geological carbon cycle is shown diagrammatically in Figure 5.20. The various steps in the process (not necessarily in this order) are as follows:

Figure 5.20 A representation of the geological carbon cycle (a: carbon in organic matter stored in peat, coal and permafrost, b: weathering of silicate minerals converts atmospheric carbon dioxide to dissolved bicarbonate, c: dissolved carbon is converted to calcite by marine organisms, d: carbon compounds are stored in sediments, e: carbon-bearing sediments are transferred to longer-term storage in the mantle, and f: carbon dioxide is released back to atmosphere during volcanic eruptions.) [SE]

During much of Earth’s history, the geological carbon cycle has been balanced, with carbon being released by volcanism at approximately the same rate that it is stored by the other processes. Under these conditions, the climate remains relatively stable.

During some periods of Earth’s history, that balance has been upset. This can happen during prolonged periods of greater than average volcanism. One example is the eruption of the Siberian Traps at around 250 Ma, which appears to have led to strong climate warming over a few million years.

A carbon imbalance is also associated with significant mountain-building events. For example, the Himalayan Range was formed between about 40 and 10 Ma and over that time period — and still today — the rate of weathering on Earth has been enhanced because those mountains are so high and the range is so extensive. The weathering of these rocks — most importantly the hydrolysis of feldspar — has resulted in consumption of atmospheric carbon dioxide and transfer of the carbon to the oceans and to ocean-floor carbonate minerals. The steady drop in carbon dioxide levels over the past 40 million years, which led to the Pleistocene glaciations, is partly attributable to the formation of the Himalayan Range.

Another, non-geological form of carbon-cycle imbalance is happening today on a very rapid time scale. We are in the process of extracting vast volumes of fossil fuels (coal, oil, and gas) that was stored in rocks over the past several hundred million years, and converting these fuels to energy and carbon dioxide. By doing so, we are changing the climate faster than has ever happened in the past.

The topics covered in this chapter can be summarized as follows:

 Introduction

Figure 6.1 The Cretaceous Dinosaur Park Formation at Dinosaur Provincial Park, Alberta, one the world’s most important sites for dinosaur fossils. The rocks in the foreground show cross-bedding, indicative of deposition in a fluvial (river) environment

In Chapter 5, we talked about weathering and erosion, which are the first two steps in the transformation of existing rocks into sedimentary rocks. The remaining steps in the formation of sedimentary rocks are transportation, deposition, burial, and lithification (Figure 6.2). Transportation is the movement of sediments or dissolved ions from the site of erosion to a site of deposition; this can be by wind, flowing water, glacial ice, or mass movement down a slope. Deposition takes place where the conditions change enough so that sediments being transported can no longer be transported (e.g., a current slows). Burial occurs when more sediments are piled onto existing sediments, and layers formed earlier are covered and compacted. Lithification is what happens — at depths of hundreds to thousands of metres — when those compacted sediments become cemented together to form solid sedimentary rock.

Figure 6.2 The rock cycle, showing the processes related to sedimentary rocks on the right-hand side.

In this textbook, we divide sedimentary rocks into two main types: clastic and chemical. Clastic sedimentary rocks are mainly composed of material that has been transported as solid fragments (clasts). Chemical sedimentary rocks are mainly composed of material that has been transported as ions in solution. It’s important not to assume that mechanical weathering leads only to clastic sedimentary rocks, while chemical weathering leads only to chemical sedimentary rocks. In most cases, millions of years separate the weathering and depositional processes, and both types of sedimentary rocks tend to include at least some material derived from both types of weathering.

A clast is a fragment of rock or mineral, ranging in size from less than a micronA micron is a millionth of a metre. There are 1,000 microns in a millimetre. (too small to see) to as big as an apartment block. Various types of clasts are shown in Figure 5.12 and in Exercise 5.3. The smaller ones tend to be composed of a single mineral crystal, and the larger ones are typically composed of pieces of rock. As we’ve seen in Chapter 5, most sand-sized clasts are made of quartz because quartz is more resistant to weathering than any other common mineral. Most clasts that are smaller than sand size (<1/16 mm) are made of clay minerals. Most clasts larger than sand size (>2 mm) are actual fragments of rock, and commonly these might be fine-grained rock like basalt or andesite, or if they are bigger, coarse-grained rock like granite or gneiss.

Grain-Size Classification

Geologists that study sediments and sedimentary rocks use the Udden-Wentworth grain-size scale for describing the sizes of the grains in these materials (Table 6.1).

Table 6.1 The Udden-Wentworth grain-size scale for classifying sediments and the grains that make up sedimentary rocks

There are six main grain-size categories; five are broken down into subcategories, with clay being the exception. The diameter limits for each successive subcategory are twice as large as the one beneath it. In general, a boulder is bigger than a toaster and difficult to lift. There is no upper limit to the size of boulder.The largest known free-standing rock (i.e., not part of bedrock) is Giant Rock in the Mojave Desert, California. It’s about as big as an apartment building — seven storeys high! A small cobble will fit in one hand, a large one in two hands. A pebble is something that you could throw quite easily. The smaller ones — known as granules — are gravel size, but still you could throw one. But you can’t really throw a single grain of sand. Sand ranges from 2 mm down to 0.063 mm, and its key characteristic is that it feels “sandy” or gritty between your fingers — even the finest sand grains feel that way. Silt is essentially too small for individual grains to be visible, and while sand feels sandy to your fingers, silt feels smooth to your fingers but gritty in your mouth. Clay is so fine that it feels smooth even in your mouth.

If you drop a granule into a glass of water, it will sink quickly to the bottom (less than half a second). If you drop a grain of sand into the same glass, it will sink more slowly (a second or two depending on the size). A grain of silt will take several seconds to get to the bottom, and a particle of fine clay may never get there. The rate of settling is determined by the balance between gravity and friction, as shown in Figure 6.3.

Figure 6.3 The two forces operating on a grain of sand in water. Gravity is pushing it down, and the friction between the grain and the water is resisting that downward force. Large particles settle quickly because the gravitational force (which is proportional to the mass, and therefore to the volume of the particle) is much greater than the frictional force (which is proportional to the surface area of the particle). For small particles it is only slightly greater, so they settle slowly.

Transportation

One of the key principles of sedimentary geology is that the ability of a moving medium (air or water) to move sedimentary particles, and keep them moving, is dependent on the velocity of flow. The faster the medium flows, the larger the particles it can move. This is illustrated in Figure 6.4. Parts of the river are moving faster than other parts, especially where the slope is greatest and the channel is narrow. Not only does the velocity of a river change from place to place, but it changes from season to season.

During peak dischargeDischarge of a stream is the volume of flow passing a point per unit time. It’s normally measured in cubic metres per second (m3/s). at this location, the water is high enough to flow over the embankment on the right, and it flows fast enough to move the boulders that cannot be moved during low flows.

Figure 6.4 Variations in flow velocity on the Englishman River near Parksville, B.C. When the photo was taken the river was not flowing fast enough anywhere to move the boulders and cobbles visible here, but it is fast enough when the discharge is higher.

Clasts within streams are moved in several different ways, as illustrated in Figure 6.5. Large bedload clasts are pushed (by traction) or bounced along the bottom (saltation), while smaller clasts are suspended in the water and kept there by the turbulence of the flow. As the flow velocity changes, different-sized clasts may be either incorporated into the flow or deposited on the bottom. At various places along a river, there are always some clasts being deposited, some staying where they are, and some being eroded and transported. This changes over time as the discharge of the river changes in response to changing weather conditions.

Other sediment transportation media, such as waves, ocean currents, and wind, operate under similar principles, with flow velocity as the key underlying factor that controls transportation and deposition.

Figure 6.5 Transportation of sediment clasts by stream flow. The larger clasts, resting on the bottom (bedload), are moved by traction (sliding) or by saltation (bouncing). Smaller clasts are kept in suspension by turbulence in the flow. Ions (depicted as + and – in the image, but invisible in real life) are dissolved in the water.

Clastic sediments are deposited in a wide range of environments, including glaciers, slope failures, rivers — both fast and slow, lakes, deltas, and ocean environments — both shallow and deep. Depending on the grain size in particular, they may eventually form into rocks ranging from fine mudstone to coarse breccia and conglomerate.

Lithification is the term used to describe a number of different processes that take place within a deposit of sediment to turn it into solid rock. One of these processes is burial by other sediments, which leads to compaction of the material and removal of some of the intervening water and air. After this stage, the individual clasts are all touching one another. Cementation is the process of crystallization of minerals within the pores between the small clasts, and also at the points of contact between the larger clasts (sand size and larger). Depending on the pressure, temperature, and chemical conditions, these crystals might include calcite, hematite, quartz, clay minerals, or a range of other minerals.

The characteristics and distinguishing features of clastic sedimentary rocks are summarized in Table 6.2. Mudrock is composed of at least 75% silt- and clay-sized fragments. If it is dominated by clay, it is called claystone. If it shows evidence of bedding or fine laminations, it is shale; otherwise it is mudstone. Mudrocks form in very low energy environments, such as lakes, river backwaters, and the deep ocean.

Group	Examples	Characteristics
Mudrock	mudstone	>75% silt and clay, not bedded
	shale	>75% silt and clay, thinly bedded
Coal		dominated by fragments of partially decayed plant matter, often enclosed between beds of sandstone or mudrock
Sandstone	quartz sandstone	dominated by sand, >90% quartz
	arkose	dominated by sand, >10% feldspar
	lithic wacke	dominated by sand, >10% rock fragments, >15% silt and clay
Conglomerate		dominated by rounded clasts, pebble size and larger
Breccia		dominated by angular clasts, pebble size and larger

Table 6. 2 The main types of clastic sedimentary rocks and their characteristics.

Most coal forms in fluvial or delta environments where vegetation growth is vigorous and where decaying plant matter accumulates in long-lasting swamps with low oxygen levels. To avoid oxidation and breakdown, the organic matter must remain submerged for centuries or millennia, until it is covered with another layer of either muddy or sandy sediments.

It is important to note that in some textbooks coal is described as an “organic sedimentary rock.” In this book, coal is classified with the clastic rocks for two reasons: first, because it is made up of fragments of organic matter; and second, because coal seams (sedimentary layers) are almost always interbedded with layers of clastic rocks, such as mudrock or sandstone. In other words, coal accumulates in environments where other clastic rocks accumulate.

It’s worth taking a closer look at the different types of sandstone because sandstone is a common and important sedimentary rock. Typical sandstone compositions are shown in Figure 6.6. The term arenite applies to a so-called clean sandstone, meaning one with less than 15% silt and clay. Considering the sand-sized grains only, arenites with 90% or more quartz are called quartz arenites. If they have more than 10% feldspar and more feldspar than rock fragments, they are called feldspathic arenites or arkosic arenites (or just arkose). If they have more than 10% rock fragments, and more rock fragments than feldspar, they are lithic“Lithic” means “rock.” Lithic clasts are rock fragments, as opposed to mineral fragments. arenites. A sandstone with more than 15% silt or clay is called a wacke (pronounced wackie). The terms quartz wacke, lithic wacke, and feldspathic wacke are used. Another name for a lithic wacke is greywacke.

Some examples of sandstones, magnified in thin section are shown in Figure 6.7. (A thin section is rock sliced thin enough so that light can shine through.)

Clastic sedimentary rocks in which a significant proportion of the clasts are larger than 2 mm are known as conglomerate if the clasts are well rounded, and breccia if they are angular. Conglomerates form in high-energy environments where the particles can become rounded, such as fast-flowing rivers. Breccias typically form where the particles are not transported a significant distance in water, such as alluvial fans and talus slopes. Some examples of clastic sedimentary rocks are shown on Figure 6.8.

Figure 6.6 A compositional triangle for arenite sandstones, with the three most common components of sand-sized grains: quartz, feldspar, and rock fragments. Arenites have less than 15% silt or clay. Sandstones with more than 15% silt and clay are called wackes (e.g., quartz wacke, lithic wacke).

Figure 6.7 Photos of thin sections of three types of sandstone. Some of the minerals are labelled: Q=quartz, F=feldspar and L= lithic (rock fragments). The quartz arenite and arkose have relatively little silt-clay matrix, while the lithic wacke has abundant matrix.

Figure 6.8 Examples of various clastic sedimentary rocks.

Exercises

Exercise 6.2 Classifying Sandstones

The table below shows magnified thin sections of three sandstones, along with descriptions of their compositions. Using Table 6.1 and Figure 6.6, find an appropriate name for each of these rocks.

Magnified Thin Section	Description	Rock name?
	Angular sand-sized grains are approximately 85% quartz and 15% feldspar. Silt and clay make up less than 5% of the rock.
	Rounded sand-sized grains are approximately 99% quartz and 1% feldspar. Silt and clay make up less than 2% of the rock.
	Angular sand-sized grains are approximately 70% quartz, 20% lithic, and 10% feldspar. Silt and clay make up about 20% of the rock.

Attributions

Aplite Red by Rudolf Pohl is under CC BY-SA 3.0

Whereas clastic sedimentary rocks are dominated by components that have been transported as solid clasts (clay, silt, sand, etc.), chemical sedimentary rocks are dominated by components that have been transported as ions in solution (Na⁺, Ca²⁺, HCO₃^–, etc.). There is some overlap between the two because almost all clastic sedimentary rocks contain cement formed from dissolved ions, and many chemical sedimentary rocks include some clasts. Since ions can stay in solution for tens of thousands of years (some much longer), and can travel for tens of thousands of kilometres, it is virtually impossible to relate chemical sediments back to their source rocks.

The most common chemical sedimentary rock, by far, is limestone. Others include chert, banded iron formation, and a variety of rocks that form when bodies of water evaporate. Biological processes are important in the formation of some chemical sedimentary rocks, especially limestone and chert. For example, limestone is made up almost entirely of fragments of marineWe use the word marine when referring to salt water (i.e., oceanic) environments, and the word aquatic when referring to freshwater environments. organisms that manufacture calcite for their shells and other hard parts, and most chert includes at least some of the silica tests (shells) of tiny marine organisms (such as diatoms and radiolaria).

Limestone

Almost all limestone forms in the oceans, and most of that forms on the shallow continental shelves, especially in tropical regions with coral reefs. Reefs are highly productive ecosystems populated by a wide range of organisms, many of which use calcium and bicarbonate ions in seawater to make carbonate minerals (especially calcite) for their shells and other structures. These include corals, of course, but also green and red algae, urchins, sponges, molluscs, and crustaceans. Especially after they die, but even while they are still alive, these organisms are eroded by waves and currents to produce carbonate fragments that accumulate in the surrounding region, as illustrated in Figure 6.9.

Figure 6.9 Various corals and green algae on a reef at Ambergris, Belize. The light-coloured sand consists of carbonate fragments eroded from the reef organisms.

Figure 6.10 shows a cross-section through a typical reef in a tropical environment (normally between 40° N and 40° S). Reefs tend to form near the edges of steep drop-offs because the reef organisms thrive on nutrient-rich upwelling currents. As the reef builds up, it is eroded by waves and currents to produce carbonate sediments that are transported into the steep offshore fore-reef area and the shallower inshore back-reef area. These sediments are dominated by reef-type carbonate fragments of all sizes, including mud. In many such areas, carbonate-rich sediments also accumulate in quiet lagoons, where mud and mollusc-shell fragments predominate (Figure 6.11a) or in offshore areas with strong currents, where either foraminifera tests accumulate (Figure 6.11b) or calcite crystallizes inorganically to form ooids – spheres of calcite that form in shallow tropical ocean water with strong currents (Figure 6.11c).

Figure 6.10 Schematic cross-section through a typical tropical reef.

Figure 6.11 Carbonate rocks and sediments: (a) mollusc-rich limestone formed in a lagoon area at Ambergris, Belize, (b) foraminifera-rich sediment from a submerged carbonate sandbar near to Ambergris, Belize (c) ooids from a beach at Joulters Cay, Bahamas.

Limestone also accumulates in deeper water, from the steady rain of the carbonate shells of tiny organisms that lived near the ocean surface. The lower limit for limestone accumulation is around 4,000 m. Beneath that depth, calcite is soluble so limestone does not accumulate.

Calcite can also form on land in a number of environments. Tufa forms at springs (Figure 6.12) and travertine (which is less porous) forms at hot springs. Similar material precipitates within limestone caves to form stalactites, stalagmites, and a wide range of other speleothems.

Figure 6.12 Tufa formed at a spring at Johnston Creek, Alberta. The rock to the left is limestone.

Dolomite (CaMg(CO₃)₂) is another carbonate mineral, but dolomite is also the name for a rock composed of the mineral dolomite (although some geologists use the term dolostone to avoid confusion). Dolomite rock is quite common (there’s a whole Italian mountain range named after it), which is surprising since marine organisms don’t make dolomite. All of the dolomite found in ancient rocks has been formed through magnesium replacing some of the calcium in the calcite in carbonate muds and sands. This process is known as dolomitization, and it is thought to take place where magnesium-rich water percolates through the sediments in carbonate tidal flat environments.

Chert and Banded Iron Formation

As we’ve seen, not all marine organisms make their hard parts out of calcite; some, like radiolaria and diatoms, use silica, and when they die their tiny shells (or tests) settle slowly to the bottom where they accumulate as chert. In some cases, chert is deposited along with limestone in the moderately deep ocean, but the two tend to remain separate, so chert beds within limestone are quite common (Figure 6.13), as are nodules, link the flint nodules of the Cretaceous chalk of southeastern England. In other situations, and especially in very deep water, chert accumulates on its own, commonly in thin beds.

Figure 6.13 Chert (brown layers) interbedded with Triassic Quatsino Fm. limestone on Quadra Island, B.C. All of the layers have been folded, and the chert, being insoluble and harder than limestone, stands out.

Some ancient chert beds — most dating to between 1800 and 2400 Ma — are also combined with a rock known as banded iron formation (BIF), a deep sea-floor deposit of iron oxide that is a common ore of iron (Figure 6.14). BIF forms when iron dissolved in seawater is oxidized, becomes insoluble, and sinks to the bottom in the same way that silica tests do to form chert. The prevalence of BIF in rocks dating from 2400 to 1800 Ma is due to the changes in the atmosphere and oceans that took place over that time period. Photosynthetic bacteria (i.e., cyanobacteria, a.k.a. blue-green algae) consume carbon dioxide from the atmosphere and use solar energy to convert it to oxygen. These bacteria first evolved around 3500 Ma, and for the next billion years, almost all of that free oxygen was used up by chemical and biological processes, but by 2400 Ma free oxygen levels started to increase in the atmosphere and the oceans. Over a period of 600 million years, that oxygen gradually converted soluble ferrous iron (Fe²⁺) to insoluble ferric iron (Fe³⁺), which combined with oxygen to form the mineral hematite (Fe₂O₃), leading to the accumulation of BIFs. After 1800 Ma, little dissolved iron was left in the oceans and the formation of BIF essentially stopped.

Figure 6.14 Banded iron formation (red) interbedded with chert (white), Dales Gorge, Australia

Evaporites

In arid regions, lakes and inland seas typically have no stream outlet and the water that flows into them is removed only by evaporation. Under these conditions, the water becomes increasingly concentrated with dissolved salts, and eventually some of these salts reach saturation levels and start to crystallize (Figure 6.15). Although all evaporite deposits are unique because of differences in the chemistry of the water, in most cases minor amounts of carbonates start to precipitate when the solution is reduced to about 50% of its original volume. Gypsum (CaSO₄·H₂O) precipitates at about 20% of the original volume and halite (NaCl) precipitates at 10%. Other important evaporite minerals include sylvite (KCl) and borax (Na₂B₄O₇·10H₂O). Sylvite is mined at numerous locations across Saskatchewan (Figure 6.16) from evaporites that were deposited during the Devonian (~385 Ma) when an inland sea occupied much of the region.

Figure 6.15 Spotted Lake, near Osoyoos, B.C. This photo was taken in May when the water was relatively fresh because of winter rains. By the end of the summer the surface of this lake is typically fully encrusted with salt deposits.

Figure 6.16 A mining machine at the face of potash ore (sylvite) in the Lanigan Mine near Saskatoon, Saskatchewan. The mineable potash layer is about 3 m thick.

Exercises

Exercise 6.3 Making Evaporite

This is an easy experiment that you can do at home. Pour about 50 mL (just less than 1/4 cup) of very hot water into a cup and add 2 teaspoons (10 mL) of salt. Stir until all or almost all of the salt has dissolved, then pour the salty water (leaving any undissolved salt behind) into a shallow wide dish or a small plate. Leave it to evaporate for a few days and observe the result.

It may look a little like the photo here. These crystals are up to about 3 mm across.

Attributions

Figure 6.11
JoultersCayOoids By Wilson44691 under Public domain.

Figure 6.14
Dales Goge by Graeme Churchard under CC BY 2.0.

Figure 6.16
Photo courtesy of PotashCorp, used with permission

Sediments accumulate in a wide variety of environments, both on the continents and in the oceans. Some of the more important of these environments are illustrated in Figure 6.17.

Figure 6.17 Some of the important depositional environments for sediments and sedimentary rocks

Table 6.3 provides a summary of the processes and sediment types that pertain to the various depositional environments illustrated in Figure 6.17. We’ll look more closely at the types of sediments that accumulate in these environments in the last section of this chapter. The characteristics of these various environments, and the processes that take place within them, are also discussed in later chapters on glaciation, mass wasting, streams, coasts, and the sea floor.

* OM (organic matter) only accumulates in swampy parts of these environments.
Table 6.3 The important terrestrial and marine depositional environments and their characteristics

Most of the sediments that you might see around you, including talus on steep slopes, sand bars in streams, or gravel in road cuts, will never become sedimentary rocks because they have only been deposited relatively recently — perhaps a few centuries or millennia ago — and will be re-eroded before they are buried deep enough beneath other sediments to be lithified. In order for sediments to be preserved long enough to be turned into rock, a process that takes millions or tens of millions of years, they need to have been deposited in a basin that will last that long. Most such basins are formed by plate tectonic processes, and some of the more important examples are shown in Figure 6.18.

Figure 6.18 Some of the more important types of tectonically produced basins: (a) trench basin, (b) forearc basin, (c) foreland basin, and (d) rift basin

Trench basins form where a subducting oceanic plate dips beneath the overriding continental or oceanic crust. They can be several kilometres deep, and in many cases, host thick sequences of sediments from eroding coastal mountains. There is a well-developed trench basin off the west coast of Vancouver Island. A forearc basin lies between the subduction zone and the volcanic arc, and may be formed in part by friction between the subducting plate and the overriding plate, which pulls part of the overriding plate down. The Strait of Georgia is a forearc basin. A foreland basin is caused by the mass of the volcanic range depressing the crust on either side. Foreland basins are not only related to volcanic ranges, but can form adjacent to fold belt mountains like the Canadian Rockies. A rift basin forms where continental crust is being pulled apart, and the crust on both sides the rift subsides. As rifting continues this eventually becomes a narrow sea, and then an ocean basin. The East African rift basin represents an early stage in this process.

Attributions

Figure 6.17
Adaptation based on Schematic diagram showing types of depositional environment by Mike Norton under CC BY SA 3.0

Through careful observation over the past few centuries, geologists have discovered that the accumulation of sediments and sedimentary rocks takes place according to some important geological principles, as follows:

• The principle of original horizontality states that sediments accumulate in essentially horizontal layers. The implication is that tilted sedimentary layers observed to day must have been subjected to tectonic forces.
• The principle of superposition states that sedimentary layers are deposited in sequence, and that unless the entire sequence has been turned over by tectonic processes, the layers at the bottom are older than those at the top.
• The principle of inclusions states that any rock fragments in a sedimentary layer must be older than the layer. For example, the cobbles in a conglomerate must have been formed before the conglomerate.
• The principle of faunal succession states that there is a well-defined order in which organisms have evolved through geological time, and therefore the identification of specific fossils in a rock can be used to determine its age.

In addition to these principles that apply to all sedimentary rocks, a number of other important characteristics of sedimentary processes lead to the development of distinctive sedimentary features in specific sedimentary environments. By understanding the origins of these features, we can make some very useful inferences about the processes that led to deposition the rocks that we are studying.

Bedding, for example, is the separation of sediments into layers that either differ from one another in textures, composition, colour, or weathering characteristics, or are separated by partings — narrow gaps between adjacent beds (Figure 6.19). Bedding is an indication of changes in depositional processes that may be related to seasonal differences, changes in climate, changes in locations of rivers or deltas, or tectonic changes. Partings may represent periods of non-deposition that could range from a few decades to a few centuries. Bedding can form in almost any depositional environment.

Figure 6.19 The Triassic Sulphur Mt. Formation near Exshaw, Alberta. Bedding is defined by differences in colour and texture, and also by partings (gaps) between beds that may otherwise appear to be similar.

Cross-bedding is bedding that contains angled layers and forms when sediments are deposited by flowing water or wind. Some examples are shown in Figures 6.1, 6.8b, and 6.20. Cross-beds in streams tend to be on the scale of centimetres to tens of centimetres, while those in aeolian (wind deposited) sediments can be on the scale of metres to several metres.

Figure 6.20 Cross-bedded Jurassic Navajo Formation aeolian sandstone at Zion National Park, Utah. In most of the layers the cross-beds dip down toward the right, implying wind direction from right to left during deposition. One bed dips in the opposite direction, implying an abnormal wind.

Cross-beds form as sediments are deposited on the leading edge of an advancing ripple or dune. Each layer is related to a different ripple that advances in the flow direction, and is partially eroded by the following ripple (Figure 6.21). Cross-bedding is a very important sedimentary structure to recognize because it can provide information on the direction of current flows and, when analyzed in detail, on other features like the rate of flow and the amount of sediment available.

Figure 6.21 Formation of cross-beds as a series of ripples or dunes migrates with the flow. Each ripple advances forward (right to left in this view) as more sediment is deposited on its leading face.

Graded bedding is characterized by a gradation in grain size from bottom to top within a single bed. “Normal” graded beds are coarse at the bottom and become finer toward the top, a product of deposition from a slowing current (Figure 6.22). Some graded beds are reversed (coarser at the top), and this normally results from deposition by a fast-moving debris flow (see Chapter 15). Most graded beds form in a submarine-fan environment (see Figure 6.17), where sediment-rich flows descend periodically from a shallow marine shelf down a slope and onto the deeper sea floor.

Figure 6.22 A graded turbidite bed in Cretaceous Spray Formation rocks on Gabriola Island, B.C. The lower several centimetres of sand and silt probably formed over the duration of an hour. The upper few centimetres of fine clay may have accumulated over a few hundred years.

Ripples, which are associated with the formation of cross-bedding, may be preserved on the surfaces of sedimentary beds. Ripples can also help to determine flow direction as they tend to have their steepest surface facing down flow.

In a stream environment, boulders, cobbles, and pebbles can become imbricated, meaning that they are generally tilted in the same direction. Clasts in streams tend to tilt with their upper ends pointing downstream because this is the most stable position with respect to the stream flow (Figure 6.23 and Figure 6.8c).

Figure 6.23 An illustration of imbrication of clasts in a fluvial environment.

Mud cracks form when a shallow body of water (e.g., a tidal flat or pond), into which muddy sediments have been deposited, dries up and cracks (Figure 6.24). This happens because the clay in the upper mud layer tends to shrink on drying, and so it cracks because it occupies less space when it is dry.

The various structures described above are critical to understanding and interpreting the formation of sedimentary rocks. In addition to these, geologists also look very closely at sedimentary grains to determine their mineralogy or lithology (in order to make inferences about the type of source rock and the weathering processes), their degree of rounding, their sizes, and the extent to which they have been sorted by transportation and depositional processes.

Figure 6.24 Mudcracks in volcanic mud at a hot-spring area near Myvatn, Iceland [SE]

We won’t be covering fossils in any detail in this book, but they are extremely important for understanding sedimentary rocks. Of course, fossils can be used to date sedimentary rocks, but equally importantly, they tell us a great deal about the depositional environment of the sediments and the climate at the time. For example, they can help to differentiate marine, aquatic, and terrestrial environments; estimate the depth of the water; detect the existence of currents; and estimate average temperature and precipitation.

The tests of tiny marine organisms (mostly foraminifera) have been recovered from deep-ocean sediment cores from all over the world, and their isotopic signatures have been measured. As we’ll see in Chapter 19, this provides us with information about the changes in average global temperatures over the past 65 million years.

Geologists who study sedimentary rocks need ways to divide them into manageable units, and they also need to give those units names so that they can easily be referred to and compared with other rocks deposited in other places. The International Commission on Stratigraphy (ICS) (http://www.stratigraphy.org/) has established a set of conventions for grouping, describing, and naming sedimentary rock units.

The main stratigraphic unit is a formation, which according to the ICS, should be established with the following principles in mind:

The contrast in lithology between formations required to justify their establishment varies with the complexity of the geology of a region and the detail needed for geologic mapping and to work out its geologic history. No formation is considered justifiable and useful that cannot be delineated at the scale of geologic mapping practiced in the region. The thickness of formations may range from less than a meter to several thousand meters.

In other words, a formation is a series of beds that is distinct from other beds above and below, and is thick enough to be shown on the geological maps that are widely used within the area in question. In most parts of the world, geological mapping is done at a relatively coarse scale, and so most formations are in the order of a few hundred metres thick. At that thickness, a typical formation would appear on a typical geological map as an area that is at least a few millimetres thick.

A series of formations can be classified together to define a group, which could be as much as a few thousand metres thick, and represents a series of rocks that were deposited within a single basin (or a series of related and adjacent basins) over a few million to a few tens of millions of years.

In areas where detailed geological information is needed (for example, within a mining or petroleum district) a formation might be divided into members, where each member has a specific and distinctive lithology. For example, a formation that includes both shale and sandstone might be divided into members, each of which is either shale or sandstone. In some areas, where particular detail is needed, members may be divided into beds, but this is only applicable to beds that have a special geological significance. Groups, formations, and members are typically named for the area where they are found.

The sedimentary rocks of the Nanaimo Group provide a useful example for understanding groups, formations, and members. During the latter part of the Cretaceous Period, from about 90 Ma to 65 Ma, a thick sequence of clastic rocks was deposited in a foreland basin between what is now Vancouver Island and the B.C. mainland (Figure 6.25). The Nanaimo Group strata comprise a 5000 m thick sequence of conglomerate, sandstone, and mudstone layers. Coal was mined from Nanaimo Group rocks from around 1850 to 1950 in the Nanaimo region, and is still being mined in the Campbell River area.

Figure 6.25 The distribution of the Upper Cretaceous Nanaimo Group rocks on Vancouver Island, the Gulf Islands, and in the Vancouver area.

The Nanaimo Group is divided into 11 formations as described in Table 6.4. In general, the boundaries between formations are based on major lithological differences. As can be seen in the far-right column of Table 6.4, a wide range of depositional environments existed during the accumulation of the Nanaimo Group rocks, from nearshore marine for the Comox and Haslam Formation, to fluvial and deltaic with backwater swampy environments for the coal-bearing Extension, Pender, and Protection Formations, to a deep-water submarine fan environment for the upper six formations. The differences in the depositional environments are probably a product of variations in tectonic-related uplift over time.

Table 6.4 The formations of the Nanaimo Group. Formations that are predominantly fine-grained are shaded. In tables like this one the layers are always listed with the oldest at the bottom and the youngest at the top. [Based on data in Mustard, P., 1994, The Upper Cretaceous Nanaimo Group, Georgia Basin, in J. Monger (ed) Geology and Geological Hazards of the Vancouver Region, Geol. Survey of Canada, Bull. 481, p. 27-95.]

The five lower formations of the Nanaimo Group are all exposed in the Nanaimo area, and were well studied during the coal mining era between 1850 and 1950. All of these formations (except Haslam) have been divided into members, as that was useful for understanding the rocks in the areas where coal mining was taking place.

Although there is a great deal of variety in the Nanaimo Group rocks, and it would take hundreds of photographs to illustrate all of the different types of rocks, a few representative examples are provided in Figure 6.26.

Figure 6.26 Representative photos of Nanaimo Group rocks. (a) Turbidite layers in the Spray Formation on Gabriola Island. Each turbidite set consists of a lower sandstone layer (light colour) that grades upward into siltstone, and then into mudstone. (See Figure 6.21 for detail.)

(b) Two separate layers of fluvial sandstone with a thin (approx. 75 cm) coal seam in between. Pender Formation in Nanaimo.

(c) Comox Formation conglomerate at the very base of the Nanaimo Group in Nanaimo. The metal object is the end of a rock hammer that is 3 cm wide. Almost all of the clasts in this view are well-rounded basalt pebbles cobbles eroded from the Triassic Karmutsen Formation which makes up a major part of Vancouver Island.

Attributions

Figure 6.25
Redrawn based on Mustard, P., 1994, The Upper Cretaceous Nanaimo Group, Georgia Basin, in J. Monger (ed) Geology and Geological Hazards of the Vancouver Region, Geol. Survey of Canada, Bull. 481, pp. 27-95

The topics covered in this chapter can be summarized as follows:

Introduction

Metamorphism is the change that takes place within a body of rock as a result of it being subjected to conditions that are different from those in which it formed. In most cases, but not all, this involves the rock being deeply buried beneath other rocks, where it is subjected to higher temperatures and pressures than those under which it formed. Metamorphic rocks typically have different mineral assemblages and different textures from their parent rocks (Figure 7.1) but they may have the same overall composition.

Figure 7.1 Metamorphic rock (gneiss) of the Okanagan Metamorphic and Igneous Complex at Skaha Lake, B.C. The dark bands are amphibole-rich, the light bands are feldspar-rich. [SE photo]

Most metamorphism results from the burial of igneous, sedimentary, or pre-existing metamorphic to the point where they experience different pressures and temperatures than those at which they formed (Figure 7.2). Metamorphism can also take place if cold rock near the surface is intruded and heated by a hot igneous body. Although most metamorphism involves temperatures above 150°C, some metamorphism takes place at temperatures lower than those at which the parent rock formed.

Figure 7.2 The rock cycle. The processes related to metamorphic rocks are at the bottom of the cycle. [SE ]

The main factors that control metamorphic processes are:

• The mineral composition of the parent rock
• The temperature at which metamorphism takes place
• The amount and type of pressure during metamorphism
• The types of fluids (mostly water) that are present during metamorphism
• The amount of time available for metamorphism

Parent Rock

The parent rock is the rock that exists before metamorphism starts. In most cases, this is sedimentary or igneous rock, but metamorphic rock that reaches the surface and is then reburied can also be considered a parent rock. On the other hand, if, for example, a mudstone is metamorphosed to slate and then buried deeper where it is metamorphosed to schist, the parent rock of the schist is mudstone, not slate. The critical feature of the parent rock is its mineral composition because it is the stability of minerals that counts when metamorphism takes place. In other words, when a rock is subjected to increased temperatures, certain minerals may become unstable and start to recrystallize into new minerals.

Temperature

The temperature that the rock is subjected to is a key variable in controlling the type of metamorphism that takes place. As we learned in the context of igneous rocks, mineral stability is a function of temperature, pressure, and the presence of fluids (especially water). All minerals are stable over a specific range of temperatures. For example, quartz is stable from environmental temperatures (whatever the weather can throw at it) all the way up to about 1800°C. If the pressure is higher, that upper limit will be higher. If there is water present, it will be lower. On the other hand, most clay minerals are only stable up to about 150° or 200°C; above that, they transform into micas. Most other common minerals have upper limits between 150°C and 1000°C.

Some minerals will crystallize into different polymorphs (same composition, but different crystalline structure) depending on the temperature and pressure. Quartz is a good example as slightly different forms are stable between 0°C and 1800°C. The minerals kyanite, andalusite, and sillimanite are polymorphs with the composition Al2SiO5. They are stable at different pressures and temperatures, and, as we will see later, they are important indicators of pressures and temperatures in metamorphic rocks (Figure 7.3).

Figure 7.3 The temperature and pressure stability fields of the three polymorphs of Al2SiO5 (Pressure is equivalent to depth. Kyanite is stable at low to moderate temperatures and low to high pressures, andalusite at moderate temperatures and low pressures, and sillimanite at higher temperatures.) [SE]

Pressure

Pressure is important in metamorphic processes for two main reasons. First, it has implications for mineral stability (Figure 7.3). Second, it has implications for the texture of metamorphic rocks. Rocks that are subjected to very high confining pressures are typically denser than others because the mineral grains are squeezed together (Figure 7.4a), and because they may contain mineral polymorphs in which the atoms are more closely packed. Because of plate tectonics, pressures within the crust are typically not applied equally in all directions. In areas of plate convergence, the pressure in one direction (perpendicular to the direction of convergence) is typically greater than in the other directions (Figure 7.4b). In situations where different blocks of the crust are being pushed in different directions, the rocks will be subjected to sheer stress (Figure 7.4c).

One of the results of directed pressure and sheer stress is that rocks become foliated — meaning that they’ll have a directional fabric. Foliation is described in more detail later in this chapter.

Figure 7.4 An illustration of different types of pressure on rocks. (a) confining pressure, where the pressure is essentially equal in all directions, (b) directed pressure, where the pressure form the sides is greater than that from the top and bottom, and (c) sheer stress caused by different blocks of rock being pushed in different directions. (In a and b there is also pressure in and out of the page.) [SE]

Fluids

Water is the main fluid present within rocks of the crust, and the only one that we’ll consider here. The presence of water is important for two main reasons. First, water facilitates the transfer of ions between minerals and within minerals, and therefore increases the rates at which metamorphic reactions take place. So, while the water doesn’t necessarily change the outcome of a metamorphic process, it speeds the process up so metamorphism might take place over a shorter time period, or metamorphic processes that might not otherwise have had time to be completed are completed.

Secondly, water, especially hot water, can have elevated concentrations of dissolved substances, and therefore it is an important medium for moving certain elements around within the crust. So not only does water facilitate metamorphic reactions on a grain-to-grain basis, it also allows for the transportation of ions from one place to another. This is very important in hydrothermal processes, which are discussed toward the end of this chapter, and in the formation of mineral deposits.

Time

Most metamorphic reactions take place at very slow rates. For example, the growth of new minerals within a rock during metamorphism has been estimated to be about 1 mm per million years. For this reason, it is very difficult to study metamorphic processes in a lab.

While the rate of metamorphism is slow, the tectonic processes that lead to metamorphism are also very slow, so in most cases, the chance for metamorphic reactions to be completed is high. For example, one important metamorphic setting is many kilometres deep within the roots of mountain ranges. A mountain range takes tens of millions of years to form, and tens of millions of years more to be eroded to the extent that we can see the rocks that were metamorphosed deep beneath it.

There are two main types of metamorphic rocks: those that are foliated because they have formed in an environment with either directed pressure or shear stress, and those that are not foliated because they have formed in an environment without directed pressure or relatively near the surface with very little pressure at all. Some types of metamorphic rocks, such as quartzite and marble, which also form in directed-pressure situations, do not necessarily exhibit foliation because their minerals (quartz and calcite respectively) do not tend to show alignment (see Figure 7.12).

When a rock is squeezed under directed pressure during metamorphism it is likely to be deformed, and this can result in a textural change such that the minerals are elongated in the direction perpendicular to the main stress (Figure 7.5). This contributes to the formation of foliation.

Figure 7.5 The textural effects of squeezing during metamorphism. [SE]

When a rock is both heated and squeezed during metamorphism, and the temperature change is enough for new minerals to form from existing ones, there is a likelihood that the new minerals will be forced to grow with their long axes perpendicular to the direction of squeezing. This is illustrated in Figure 7.6, where the parent rock is shale, with bedding as shown. After both heating and squeezing, new minerals have formed within the rock, generally parallel to each other, and the original bedding has been largely obliterated.

Figure 7.6 The textural effects of squeezing and aligned mineral growth during metamorphism. The left-hand diagram represents shale with bedding in the direction shown. The right-hand diagram represents schist (derived from that shale), with the mica crystals orientated perpendicular to the main stress direction and the original bedding no longer easily visible. [SE]

Figure 7.7 shows an example of this effect. This large boulder has bedding still visible as dark and light bands sloping steeply down to the right. The rock also has a strong slaty foliation, which is horizontal in this view, and has developed because the rock was being squeezed during metamorphism. The rock has split from bedrock along this foliation plane, and you can see that other weaknesses are present in the same orientation.

Squeezing and heating alone (as shown in Figure 7.5) and squeezing, heating, and formation of new minerals (as shown in Figure 7.6) can contribute to foliation, but most foliation develops when new minerals are forced to grow perpendicular to the direction of greatest stress (Figure 7.6). This effect is especially strong if the new minerals are platy like mica or elongated like amphibole. The mineral crystals don’t have to be large to produce foliation. Slate, for example, is characterized by aligned flakes of mica that are too small to see.

Figure 7.7 A slate boulder on the side of Mt. Wapta in the Rockies near Field, BC. Bedding is visible as light and dark bands sloping steeply to the right. Slaty cleavage is evident from the way the rock has broken and also from lines of weakness that same trend. [SE]

The various types of foliated metamorphic rocks, listed in order of the grade or intensity of metamorphism and the type of foliation are slate, phyllite, schist, and gneiss (Figure 7.8). As already noted, slate is formed from the low-grade metamorphism of shale, and has microscopic clay and mica crystals that have grown perpendicular to the stress. Slate tends to break into flat sheets. Phyllite is similar to slate, but has typically been heated to a higher temperature; the micas have grown larger and are visible as a sheen on the surface. Where slate is typically planar, phyllite can form in wavy layers. In the formation of schist, the temperature has been hot enough so that individual mica crystals are visible, and other mineral crystals, such as quartz, feldspar, or garnet may also be visible. In gneiss, the minerals may have separated into bands of different colours. In the example shown in Figure 7.8d, the dark bands are largely amphibole while the light-coloured bands are feldspar and quartz. Most gneiss has little or no mica because it forms at temperatures higher than those under which micas are stable. Unlike slate and phyllite, which typically only form from mudrock, schist, and especially gneiss, can form from a variety of parent rocks, including mudrock, sandstone, conglomerate, and a range of both volcanic and intrusive igneous rocks.

Schist and gneiss can be named on the basis of important minerals that are present. For example a schist derived from basalt is typically rich in the mineral chlorite, so we call it chlorite schist. One derived from shale may be a muscovite-biotite schist, or just a mica schist, or if there are garnets present it might be mica-garnet schist. Similarly, a gneiss that originated as basalt and is dominated by amphibole, is an amphibole gneiss or, more accurately, an amphibolite.

Figure 7.8 Examples of foliated metamorphic rocks [a, b, and d: SE, c: Michael C. Rygel, http://en.wikipedia.org/wiki/Schist#mediaviewer/File:Schist_detail.jpg]

If a rock is buried to a great depth and encounters temperatures that are close to its melting point, it will partially melt. The resulting rock, which includes both metamorphosed and igneous material, is known as a migmatite (Figure 7.9).

Figure 7.9 Migmatite from Prague, Czech Republic

[http://commons.wikimedia.org/wiki/ File:Migmatite_in_Geopark_on_Albertov.JPG]
As already noted, the nature of the parent rock controls the types of metamorphic rocks that can form from it under differing metamorphic conditions. The kinds of rocks that can be expected to form at different metamorphic grades from various parent rocks are listed in Table 7.1. Some rocks, such as granite, do not change much at the lower metamorphic grades because their minerals are still stable up to several hundred degrees.

Table 7.1 A rough guide to the types of metamorphic rocks that form from different parent rocks at different grades of regional metamorphism

Metamorphic rocks that form under either low-pressure conditions or just confining pressure do not become foliated. In most cases, this is because they are not buried deeply, and the heat for the metamorphism comes from a body of magma that has moved into the upper part of the crust. This is contact metamorphism. Some examples of non-foliated metamorphic rocks are marble, quartzite, and hornfels.

Marble is metamorphosed limestone. When it forms, the calcite crystals tend to grow larger, and any sedimentary textures and fossils that might have been present are destroyed. If the original limestone was pure calcite, then the marble will likely be white (as in Figure 7.10), but if it had various impurities, such as clay, silica, or magnesium, the marble could be “marbled” in appearance.

Figure 7.10 Marble with visible calcite crystals (left) and an outcrop of banded marble (right) [SE (left) and http://gallery.usgs.gov/images/08_11_2010/a1Uh83Jww6_08_11_2010/large/DSCN2868.JPG (right)]

Quartzite is metamorphosed sandstone (Figure 7.11). It is dominated by quartz, and in many cases, the original quartz grains of the sandstone are welded together with additional silica. Most sandstone contains some clay minerals and may also include other minerals such as feldspar or fragments of rock, so most quartzite has some impurities with the quartz.

Figure 7.11 Quartzite from the Rocky Mountains, found in the Bow River at Cochrane, Alberta [SE]

Even if formed during regional metamorphism, quartzite does not tend to be foliated because quartz crystals don’t align with the directional pressure. On the other hand, any clay present in the original sandstone is likely to be converted to mica during metamorphism, and any such mica is likely to align with the directional pressure. An example of this is shown in Figure 7.12. The quartz crystals show no alignment, but the micas are all aligned, indicating that there was directional pressure during regional metamorphism of this rock.

Figure 7.12 Magnified thin section of quartzite in polarized light. The irregular-shaped white, grey, and black crystals are all quartz. The small, thin, brightly coloured crystals are mica. This rock is foliated, even though it might not appear to be if examined without a microscope, and so it must have formed under directed-pressure conditions.
[Photo by Sandra Johnstone, used with permission]

Hornfels is another non-foliated metamorphic rock that normally forms during contact metamorphism of fine-grained rocks like mudstone or volcanic rock (Figure 7.13). In some cases, hornfels has visible crystals of minerals like biotite or andalusite. If the hornfels formed in a situation without directed pressure, then these minerals would be randomly orientated, not foliated as they would be if formed with directed pressure.

Figure 7.13 Hornfels from the Novosibirsk region of Russia. The dark and light bands are bedding. The rock has been recrystallized during contact metamorphism and does not display foliation. (scale in cm)
[http://en.wikipedia.org/wiki/Hornfels#mediaviewer/ File:Hornfels.jpg]

All of the important processes of metamorphism that we are familiar with can be directly related to geological processes caused by plate tectonics. The relationships between plate tectonics and metamorphism are summarized in Figure 7.14, and in more detail in Figures 7.15, 7.16, 7.17, and 7.19.

Figure 7.14 Environments of metamorphism in the context of plate tectonics: (a) regional metamorphism related to mountain building at a continent-continent convergent boundary, (b) regional metamorphism of oceanic crust in the area on either side of a spreading ridge, (c) regional metamorphism of oceanic crustal rocks within a subduction zone, (d) contact metamorphism adjacent to a magma body at a high level in the crust, and (e) regional metamorphism related to mountain building at a convergent boundary. [SE]

Most regional metamorphism takes place within continental crust. While rocks can be metamorphosed at depth in most areas, the potential for metamorphism is greatest in the roots of mountain ranges where there is a strong likelihood for burial of relatively young sedimentary rock to great depths, as depicted in Figure 7.15. An example would be the Himalayan Range. At this continent-continent convergent boundary, sedimentary rocks have been both thrust up to great heights (nearly 9,000 m above sea level) and also buried to great depths. Considering that the normal geothermal gradient (the rate of increase in temperature with depth) is around 30°C per kilometre, rock buried to 9 km below sea level in this situation could be close to 18 km below the surface of the ground, and it is reasonable to expect temperatures up to 500°C. Metamorphic rocks formed there are likely to be foliated because of the strong directional pressure of converging plates.

Figure 7.15  a: Regional metamorphism beneath a mountain range related to continent-continent collision (typical geothermal gradient). (Example: Himalayan Range) [SE]

At an oceanic spreading ridge, recently formed oceanic crust of gabbro and basalt is slowly moving away from the plate boundary (Figure 7.16). Water within the crust is forced to rise in the area close to the source of volcanic heat, and this draws more water in from farther out, which eventually creates a convective system where cold seawater is drawn into the crust and then out again onto the sea floor near the ridge. The passage of this water through the oceanic crust at 200° to 300°C promotes metamorphic reactions that change the original pyroxene in the rock to chlorite and serpentine. Because this metamorphism takes place at temperatures well below the temperature at which the rock originally formed (~1200°C), it is known as retrograde metamorphism. The rock that forms in this way is known as greenstone if it isn’t foliated, or greenschist if it is. Chlorite ((Mg₅Al)(AlSi₃)O₁₀(OH)₈) and serpentine ((Mg, Fe)₃Si₂O₅(OH)₄) are both “hydrated minerals” meaning that they have water (as OH) in their chemical formulas. When metamorphosed ocean crust is later subducted, the chlorite and serpentine are converted into new non-hydrous minerals (e.g., garnet and pyroxene) and the water that is released migrates into the overlying mantle, where it contributes to flux melting (Chapter 3, section 3.2).

Figure 7.16  b: Regional metamorphism of oceanic crustal rock on either side of a spreading ridge. (Example: Juan de Fuca spreading ridge) [SE]

At a subduction zone, oceanic crust is forced down into the hot mantle. But because the oceanic crust is now relatively cool, especially along its sea-floor upper surface, it does not heat up quickly, and the subducting rock remains several hundreds of degrees cooler than the surrounding mantle (Figure 7.17). A special type of metamorphism takes place under these very high-pressure but relatively low-temperature conditions, producing an amphibole mineral known as glaucophane (Na₂(Mg₃Al₂)Si₈O₂₂(OH)₂), which is blue in colour, and is a major component of a rock known as blueschist.

If you’ve never seen or even heard of blueschist, it’s not surprising. What is surprising is that anyone has seen it! Most blueschist forms in subduction zones, continues to be subducted, turns into eclogite at about 35 km depth, and then eventually sinks deep into the mantle — never to be seen again. In only a few places in the world, where the subduction process has been interrupted by some tectonic process, has partially subducted blueschist rock returned to the surface. One such place is the area around San Francisco; the rock is known as the Franciscan Complex (Figure 7.18).

Figure 7.17  c: Regional metamorphism of oceanic crust at a subduction zone. (Example: Cascadia subduction zone. Rock of this type is exposed in the San Francisco area.) [SE]

Figure 7.18 Franciscan Complex blueschist rock exposed north of San Francisco. The blue colour of rock is due to the presence of the amphibole mineral glaucophane. [SE]

Magma is produced at convergent boundaries and rises toward the surface, where it can form magma bodies in the upper part of the crust. Such magma bodies, at temperatures of around 1000°C, heat up the surrounding rock, leading to contact metamorphism (Figure 7.19). Because this happens at relatively shallow depths, in the absence of directed pressure, the resulting rock does not normally develop foliation. The zone of contact metamorphism around an intrusion is very small (typically metres to tens of metres) compared with the extent of regional metamorphism in other settings (tens of thousands of square kilometres).

Figure 7.19  d: Contact metamorphism around a high-level crustal magma chamber (Example: the magma chamber beneath Mt. St. Helens.)  e: Regional metamorphism in a volcanic-arc related mountain range (volcanic-region temperature gradient) (Example: The southern part of the Coast Range, B.C.) [SE]

Regional metamorphism also takes place within volcanic-arc mountain ranges, and because of the extra heat associated with the volcanism, the geothermal gradient is typically a little steeper in these settings (somewhere between 40° and 50°C/km). As a result higher grades of metamorphism can take place closer to surface than is the case in other areas (Figure 7.19).

Another way to understand metamorphism is by using a diagram that shows temperature on one axis and depth (which is equivalent to pressure) on the other (Figure 7.20). The three heavy dotted lines on this diagram represent Earth’s geothermal gradients under different conditions. In most areas, the rate of increase in temperature with depth is 30°C/km. In other words, if you go 1,000 m down into a mine, the temperature will be roughly 30°C warmer than the average temperature at the surface. In most parts of southern Canada, the average surface temperature is about 10°C, so at 1,000 m depth, it will be about 40°C. That’s uncomfortably hot, so deep mines must have effective ventilation systems. This typical geothermal gradient is shown by the green dotted line in Figure 7.20. At 10 km depth, the temperature is about 300°C and at 20 km it’s about 600°C.

In volcanic areas, the geothermal gradient is more like 40° to 50°C/km, so the temperature at 10 km depth is in the 400° to 500°C range. Along subduction zones, as described above, the cold oceanic crust keeps temperatures low, so the gradient is typically less than 10°C/km. The various types of metamorphism described above are represented in Figure 7.20 with the same letters (a through e) used in Figures 7.14 to 7.17 and 7.19.

Figure 7.20 Types of metamorphism shown in the context of depth and temperature under different conditions. The metamorphic rocks formed from mudrock under regional metamorphosis with a typical geothermal gradient are listed. The letters a through e correspond with those shown in Figures 7.14 to 7.17 and 7.19. [SE]

By way of example, if we look at regional metamorphism in areas with typical geothermal gradients, we can see that burial in the 5 km to 10 km range puts us in the zeoliteZeolites are silicate minerals that typically form during low-grade metamorphism of volcanic rocks. and clay mineral zone (see Figure 7.20), which is equivalent to the formation of slate. At 10 km to 15 km, we are in the greenschist zone (where chlorite would form in mafic volcanic rock) and very fine micas form in mudrock, to produce phyllite. At 15 km to 20 km, larger micas form to produce schist, and at 20 km to 25 km amphibole, feldspar, and quartz form to produce gneiss. Beyond 25 km depth in this setting, we cross the partial melting line for granite (or gneiss) with water present, and so we can expect migmatite to form.

As described above, regional metamorphism occurs when rocks are buried deep in the crust. This is commonly associated with convergent plate boundaries and the formation of mountain ranges. Because burial to 10 km to 20 km is required, the areas affected tend to be large.

Rather than focusing on metamorphic rock textures (slate, schist, gneiss, etc.), geologists tend to look at specific minerals within the rocks that are indicative of different grades of metamorphism. Some common minerals in metamorphic rocks are shown in Figure 7.21, arranged in order of the temperature ranges within which they tend to be stable. The upper and lower limits of the ranges are intentionally vague because these limits depend on a number of different factors, such as the pressure, the amount of water present, and the overall composition of the rock.

Figure 7.21 Metamorphic index minerals and their approximate temperature ranges [SE]

The southern and southwestern parts of Nova Scotia were regionally metamorphosed during the Devonian Acadian Orogeny (around 400 Ma), when a relatively small continental block (the Meguma TerraneNo, it’s not a spelling mistake! A terrane is a distinctive block of crust that is now part of a continent, but is thought to have come from elsewhere, and was added on by plate-tectonic processes.) was pushed up against the existing eastern margin of North America. As shown in Figure 7.22, clastic sedimentary rocks within this terrane were variably metamorphosed, with the strongest metamorphism in the southwest (the sillimanite zone), and progressively weaker metamorphism toward the east and north. The rocks of the sillimanite zone were likely heated to over 700°C, and therefore must have buried to depths between 20 km and 25 km. The surrounding lower-grade rocks were not buried as deep, and the rocks within the peripheral chlorite zone were likely not buried to more than about 5 km.

Figure 7.22 Regional metamorphic zones in the Meguma Terrane of southwestern Nova Scotia [SE, after Keppie, D, and Muecke, G, 1979, Metamorphic map of Nova Scotia, N.S. Dept. of Mines and Energy, Map 1979-006., and from White, C and Barr, S., 2012, Meguma Terrane revisted, Stratigraphy, metamorphism, paleontology and provenance, Geoscience Canada, V. 39, No.1]

A probable explanation for this pattern is that the area with the highest-grade rocks was buried beneath the central part of a mountain range formed by the collision of the Meguma Terrane with North America. As is the case with all mountain ranges, the crust became thickened as the mountains grew, and it was pushed farther down into the mantle than the surrounding crust. This happens because Earth’s crust is floating on the underlying mantle. As the formation of mountains adds weight, the crust in that area sinks farther down into the mantle to compensate for the added weight. The likely pattern of metamorphism in this situation is shown in cross-section in Figure 7.23a. The mountains were eventually eroded (over tens of millions of years), allowing the crust to rebound upward and exposing the metamorphic rock (Figure 7.23b).

Figure 7.23 (a) Schematic cross-section through the Meguma Terrane during the Devonian.
The crust is thickened underneath the mountain range to compensate for the added weight of the mountains above.
Temperature contours are shown, and the metamorphic zones are depicted using colours similar to those in Figure 7.22.

Figure 7.23 (b) Schematic present-day cross-section through the Meguma Terrane.
The mountains have been eroded. As they lost mass the base of the crust gradually rebounded, pushing up the core of the metamorphosed region so that the once deeply buried metamorphic zones are now exposed at surface.

The metamorphism in Nova Scotia’s Meguma Terrane is just one example of the nature of regional metamorphism. Obviously many different patterns of regional metamorphism exist, depending on the parent rocks, the geothermal gradient, the depth of burial, the pressure regime, and the amount of time available. The important point is that regional metamorphism happens only at significant depths. The greatest likelihood of attaining those depths, and then having the once-buried rocks eventually exposed at the surface, is where mountain ranges existed and have since been largely eroded away. As this happens typically at convergent plate boundaries, directed pressures can be strong, and regionally altered rocks are almost always foliated.

Contact metamorphism takes place where a body of magma intrudes into the upper part of the crust. Any type of magma body can lead to contact metamorphism, from a thin dyke to a large stock. The type and intensity of the metamorphism, and width of the metamorphic aureole will depend on a number of factors, including the type of country rock, the temperature of the intruding body and the size of the body (Figure 7.24). A large intrusion will contain more thermal energy and will cool much more slowly than a small one, and therefore will provide a longer time and more heat for metamorphism. That will allow the heat to extend farther into the country rock, creating a larger aureole.

Figure 7.24 Schematic cross-section of the middle and upper crust showing two magma bodies.
The upper body, which has intruded into cool unmetamorphosed rock, has created a zone of contact metamorphism.
The lower body is surrounded by rock that is already hot (and probably already metamorphosed), and so it does not have a significant metamorphic aureole. [SE]

Contact metamorphic aureoles are typically quite small, from just a few centimetres around small dykes and sills, to as much as 100 m around a large stock. As was shown in Figure 7.20, contact metamorphism can take place over a wide range of temperatures — from around 300° to over 800°C — and of course the type of metamorphism, and new minerals formed, will vary accordingly. The nature of the country rock is also important. Mudrock or volcanic rock will be converted to hornfels. Limestone will be metamorphosed to marble, and sandstone to quartzite.

A hot body of magma in the upper crust can create a very dynamic situation that may have geologically interesting and economically important implications. In the simplest cases, water does not play a big role, and the main process is transfer of heat from the pluton to the surrounding rock, creating a zone of contact metamorphism (Figure 7.25a). In many cases, however, water is released from the magma body as crystallization takes place, and this water is dispersed along fractures in the country rock (Figure 7.25b). The water released from a magma chamber is typically rich in dissolved minerals. As this water cools, is chemically changed by the surrounding rocks, or boils because of a drop in pressure, minerals are deposited, forming veins within the fractures in the country rock. Quartz veins are common in this situation, and they might also include pyrite, hematite, calcite, and even silver and gold.

Figure 7.25 Depiction of metamorphism and alteration around a pluton in the upper crust.
(a) Thermal metamorphism only (within the purple zone)
(b) Thermal metamorphism plus veining (white) related to dispersal of magmatic fluids into the overlying rock
(c) Thermal metamorphism plus veining from magmatic fluids plus alteration and possible formation of metallic minerals (hatched yellow areas) from convection of groundwater

Heat from the magma body will cause surrounding groundwater to expand and then rise toward the surface. In some cases, this may initiate a convection system where groundwater circulates past the pluton. Such a system could operate for thousands of years, resulting in the circulation of millions of tonnes of groundwater from the surrounding region past the pluton. Hot water circulating through the rocks can lead to significant changes in the mineralogy of the rock, including alteration of feldspars to clays, and deposition of quartz, calcite, and other minerals in fractures and other open spaces (Figure 7.26). As with the magmatic fluids, the nature of this circulating groundwater can also change adjacent to, or above, the pluton, resulting in deposition of other minerals, including ore minerals. Metamorphism in which much of the change is derived from fluids passing through the rock is known as metasomatism. When hot water contributes to changes in rocks, including mineral alteration and formation of veins, it is known as hydrothermal alteration.

Figure 7.26 Calcite veins in limestone of the Comox Formation, Nanaimo, B.C. [SE]

A special type of metasomatism takes place where a hot pluton intrudes into carbonate rock such as limestone. When magmatic fluids rich in silica, calcium, magnesium, iron, and other elements flow through the carbonate rock, their chemistry can change dramatically, resulting in the deposition of minerals that would not normally exist in either the igneous rock or limestone. These include garnet, epidote (another silicate), magnetite, pyroxene, and a variety of copper and other minerals (Figure 7.27). This type of metamorphism is known as skarn.

Figure 7.27 A skarn rock from Mount Monzoni, Northern Italy, with recrystallized calcite (blue), garnet (brown), and pyroxene (green). The rock is 6 cm across. [by Siim Sepp, from http://commons.wikimedia.org/wiki/File:00031_6_cm_grossular_calcite_augite_skarn.jpg]

The topics covered in this chapter can be summarized as follows:

Introduction

Time is the dimension that sets geology apart from most other sciences. Geological time is vast, and Earth has changed enough over that time that some of the rock types that formed in the past could not form today. Furthermore, as we’ve discussed, even though most geological processes are very, very slow, the vast amount of time that has passed has allowed for the formation of extraordinary geological features, as shown in Figure 8.1.

Figure 8.1 Arizona’s Grand Canyon is an icon for geological time; 1,450 million years are represented by this photo. The light-coloured layered rocks at the top formed at around 250 Ma, and the dark ones at the bottom (within the steep canyon) at around 1,700 Ma. [SE]

We have numerous ways of measuring geological time. We can tell the relative ages of rocks (for example, whether one rock is older than another) based on their spatial relationships; we can use fossils to date sedimentary rocks because we have a detailed record of the evolution of life on Earth; and we can use a range of isotopic techniques to determine the actual ages (in millions of years) of igneous and metamorphic rocks.

But just because we can measure geological time doesn’t mean that we understand it. One of the biggest hurdles faced by geology students, and geologists as well, in understanding geology, is to really come to grips with the slow rates at which geological processes happen and the vast amount of time involved.

William “Strata” Smith worked as a surveyor in the coal-mining and canal-building industries in southwestern England in the late 1700s and early 1800s. While doing his work, he had many opportunities to look at the Paleozoic and Mesozoic sedimentary rocks of the region, and he did so in a way that few had done before. Smith noticed the textural similarities and differences between rocks in different locations, and more importantly, he discovered that fossils could be used to correlate rocks of the same age. Smith is credited with formulating the principle of faunal succession (the concept that specific types of organisms lived during different time intervals), and he used it to great effect in his monumental project to create a geological map of England and Wales, published in 1815. (For more on William Smith, including a large-scale digital copy of the famous map, see http://en.wikipedia.org/wiki/William_Smith_%28geologist%29.)

Inset into Smith’s great geological map is a small diagram showing a schematic geological cross-section extending from the Thames estuary of eastern England all the way to the west coast of Wales. Smith shows the sequence of rocks, from the Paleozoic rocks of Wales and western England, through the Mesozoic rocks of central England, to the Cenozoic rocks of the area around London (Figure 8.2). Although Smith did not put any dates on these — because he didn’t know them — he was aware of the principle of superposition (the idea, developed much earlier by the Danish theologian and scientist Nicholas Steno, that young sedimentary rocks form on top of older ones), and so he knew that this diagram represented a stratigraphic column. And because almost every period of the Phanerozoic is represented along that section through Wales and England, it is a primitive geological time scale.

Figure 8.2 William Smith’s “Sketch of the succession of strata and their relative altitudes,” an inset on his geological map of England and Wales (with era names added). [SE after: http://earthobservatory.nasa.gov/Features/WilliamSmith/images/sketch_of_the_succession_of_strata.jpg]

Smith’s work set the stage for the naming and ordering of the geological periods, which was initiated around 1820, first by British geologists, and later by other European geologists. Many of the periods are named for places where rocks of that age are found in Europe, such as Cambrian for Cambria (Wales), Devonian for Devon in England, Jurassic for the Jura Mountains in France and Switzerland, and Permian for the Perm region of Russia. Some are named for the type of rock that is common during that age, such as Carboniferous for the coal- and carbonate-bearing rocks of England, and Cretaceous for the chalks of England and France.

The early time scales were only relative because 19th century geologists did not know the ages of the rocks. That information was not available until the development of isotopic dating techniques early in the 20th century.

The geological time scale is currently maintained by the International Commission on Stratigraphy (ICS), which is part of the International Union of Geological Sciences. The time scale is continuously being updated as we learn more about the timing and nature of past geological events. You can view the ICS time scale at http://www.stratigraphy.org/index.php/ics-chart-timescale. It would be a good idea to print a copy (in colour) to put on your wall while you are studying geology.

Geological time has been divided into four eons: Hadean, Archean, Proterozoic, and Phanerozoic, and as shown in Figure 8.3, the first three of these represent almost 90% of Earth’s history. The last one, the Phanerozoic (meaning “visible life”), is the time that we are most familiar with because Phanerozoic rocks are the most common on Earth, and they contain evidence of the life forms that we are all somewhat familiar with.

Figure 8.3 The eons of Earth’s history [SE]

The Phanerozoic — the past 540 Ma of Earth’s history — is divided into three eras: the Paleozoic (“early life”), the Mesozoic (“middle life”), and the Cenozoic (“new life”), and each of these is divided into a number of periods (Figure 8.4). Most of the organisms that we share Earth with evolved at various times during the Phanerozoic.

Figure 8.4 The eras (middle row) and periods (bottom row) of the Phanerozoic [SE]

The Cenozoic, which represents the past 65.5 Ma, is divided into three periods: Paleogene, Neogene, and Quaternary, and seven epochs (Figure 8.5). Dinosaurs became extinct at the start of the Cenozoic, after which birds and mammals radiated to fill the available habitats. Earth was very warm during the early Eocene and has steadily cooled ever since. Glaciers first appeared on Antarctica in the Oligocene and then on Greenland in the Miocene, and covered much of North America and Europe by the Pleistocene. The most recent of the Pleistocene glaciations ended around 11,700 years ago. The current epoch is known as the Holocene. Epochs are further divided into ages (a.k.a. stages), but we won’t be going into that level of detail here.

Figure 8.5 The periods (middle row) and epochs (bottom row) of the Cenozoic [SE]

Most of the boundaries between the periods and epochs of the geological time scale have been fixed on the basis of significant changes in the fossil record. For example, as already noted, the boundary between the Cretaceous and the Paleogene coincides exactly with the extinction of the dinosaurs. That’s not a coincidence. Many other types of organisms went extinct at this time, and the boundary between the two periods marks the division between sedimentary rocks with Cretaceous organisms below, and Paleogene organisms above.

The simplest and most intuitive way of dating geological features is to look at the relationships between them. There are a few simple rules for doing this, some of which we’ve already looked at in Chapter 6. For example, the principle of superposition states that sedimentary layers are deposited in sequence, and, unless the entire sequence has been turned over by tectonic processes or disrupted by faulting, the layers at the bottom are older than those at the top. The principle of inclusions states that any rock fragments that are included in rock must be older than the rock in which they are included. For example, a xenolith in an igneous rock or a clast in sedimentary rock must be older than the rock that includes it (Figure 8.6).

Figure 8.6a A xenolith of diorite incorporated into a basalt lava flow, Mauna Kea volcano, Hawaii. The lava flow took place some time after the diorite cooled, was uplifted, and then eroded. (Hammerhead for scale) [SE]

Figure 8.6b Rip-up clasts of shale embedded in Gabriola Formation sandstone, Gabriola Island, B.C. The pieces of shale were eroded as the sandstone was deposited, so the shale is older than the sandstone. [SE]

The principle of cross-cutting relationships states that any geological feature that cuts across, or disrupts another feature must be younger than the feature that is disrupted. An example of this is given in Figure 8.7, which shows three different sedimentary layers. The lower sandstone layer is disrupted by two faults, so we can infer that the faults are younger than that layer. But the faults do not appear to continue into the coal seam, and they certainly do not continue into the upper sandstone. So we can infer that coal seam is younger than the faults (because it disrupts them), and of course the upper sandstone is youngest of all, because it lies on top of the coal seam.

Figure 8.7 Superposition and cross-cutting relationships in Cretaceous Nanaimo Group rocks in Nanaimo, B.C. The coal seam is about 50 cm thick. [SE ]

An unconformity represents an interruption in the process of deposition of sedimentary rocks. Recognizing unconformities is important for understanding time relationships in sedimentary sequences. An example of an unconformity is shown in Figure 8.8. The Proterozoic rocks of the Grand Canyon Group have been tilted and then eroded to a flat surface prior to deposition of the younger Paleozoic rocks. The difference in time between the youngest of the Proterozoic rocks and the oldest of the Paleozoic rocks is close to 300 million years. Tilting and erosion of the older rocks took place during this time, and if there was any deposition going on in this area, the evidence of it is now gone.

Figure 8.8 The great angular unconformity in the Grand Canyon, Arizona. The tilted rocks at the bottom are part of the Proterozoic Grand Canyon Group (aged 825 to 1,250 Ma). The flat-lying rocks at the top are Paleozoic (540 to 250 Ma). The boundary between the two represents a time gap of nearly 300 million years. [SE ]

There are four types of unconformities, as summarized in Table 8.1, and illustrated in Figure 8.9.

Table 8.1 The characteristics of the four types of unconformities

Figure 8.9 The four types of unconformities: (a) a nonconformity between non-sedimentary rock and sedimentary rock, (b) an angular unconformity, (c) a disconformity between layers of sedimentary rock, where the older rock has been eroded but not tilted, and (d) a paraconformity where there is a long period (millions of years) of non-deposition between two parallel layers. [SE ]

Geologists get a wide range of information from fossils. They help us to understand evolution and life in general; they provide critical information for understanding depositional environments and changes in Earth’s climate; and, of course, they can be used to date rocks.

Although the recognition of fossils goes back hundreds of years, the systematic cataloguing and assignment of relative ages to different organisms from the distant past — paleontology — only dates back to the earliest part of the 19th century. The oldest undisputed fossils are from rocks dated around 3.5 Ga, and although fossils this old are typically poorly preserved and are not useful for dating rocks, they can still provide important information about conditions at the time. The oldest well-understood fossils are from rocks dating back to around 600 Ma, and the sedimentary record from that time forward is rich in fossil remains that provide a detailed record of the history of life. However, as anyone who has gone hunting for fossils knows, that does not mean that all sedimentary rocks have visible fossils or that they are easy to find. Fossils alone cannot provide us with numerical ages of rocks, but over the past century geologists have acquired enough isotopic dates from rocks associated with fossil-bearing rocks (such as igneous dykes cutting through sedimentary layers) to be able to put specific time limits on most fossils.

A very selective history of life on Earth over the past 600 million years is provided in Figure 8.10. The major groups of organisms that we are familiar with evolved between the late Proterozoic and the Cambrian (~600 Ma to ~520 Ma). Plants, which evolved in the oceans as green algae, came onto land during the Ordovician (~450 Ma). Insects, which evolved from marine arthropods, came onto land during the Devonian (400 Ma), and amphibians (i.e., vertebrates) came onto land about 50 million years later. By the late Carboniferous, trees had evolved from earlier plants, and reptiles had evolved from amphibians. By the mid-Triassic, dinosaurs and mammals had evolved from very different branches of the reptiles; birds evolved from dinosaurs during the Jurassic. Flowering plants evolved in the late Jurassic or early Cretaceous. The earliest primates evolved from other mammals in the early Paleogene, and the genus Homo evolved during the late Neogene (~2.8 Ma).

Figure 8.10 A summary of life on Earth during the late Proterozoic and the Phanerozoic. The top row shows geological eras, and the lower row shows the periods. [SE]

If we understand the sequence of evolution on Earth, we can apply knowledge to determining the relative ages of rocks. This is William Smith’s principle of faunal succession, although of course it doesn’t just apply to “fauna” (animals); it can also apply to fossils of plants and those of simple organisms.

The Phanerozoic has seen five major extinctions, as indicated in Figure 8.10. The most significant of these was at the end of the Permian, which saw the extinction of over 80% of all species and over 90% of all marine species. Most well-known types of organisms were decimated by this event, but only a few became completely extinct, including trilobites. The second most significant extinction was at the Cretaceous-Paleogene boundary (K-Pg, a.k.a. the K-T extinction). At that time, about 75% of marine species disappeared. Again, a few well-known types of organisms disappeared altogether, including dinosaurs (but not birds) and the pterosaurs. Other types were badly decimated but survived, and then flourished in the Paleogene. The K-Pg extinction is thought to have been caused by the impact of a large extraterrestrial body (10 km to 15 km across), but it is generally agreed that the other four Phanerozoic extinctions had other causes, although their exact nature is not clearly understood.

As already stated, it is no coincidence that the major extinctions all coincide with boundaries of geological periods and even eras. Paleontologists have placed most of the divisions of the geological time scale at points in the fossil record where there are major changes in the type of organisms observed.

If we can identify a fossil to the species level, or at least to the genus level, and we know the time period when the organism lived, we can assign a range of time to the rock. That range might be several million years because some organisms survived for a very long time. If the rock we are studying has several types of fossils in it, and we can assign time ranges to those fossils, we might be able to narrow the time range for the age of the rock considerably. An example of this is given in Figure 8.11.

Figure 8.11 The application of bracketing to constrain the age of a rock based on several fossils. In this diagram, the coloured bar represents the time range during which each of the four species (A – D) existed on Earth. Although each species lived for several million years, we can narrow down the likely age of the rock to just 0.7 Ma during which all four species coexisted. [SE]

Some organisms survived for a very long time, and are not particularly useful for dating rocks. Sharks, for example, have been around for over 400 million years, and the great white shark has survived for 16 million years, so far. Organisms that lived for relatively short time periods are particularly useful for dating rocks, especially if they were distributed over a wide geographic area and so can be used to compare rocks from different regions. These are known as index fossils. There is no specific limit on how short the time span has to be to qualify as an index fossil. Some lived for millions of years, and others for much less than a million years.

Some well-studied groups of organisms qualify as biozone fossils because, although the genera and families lived over a long time, each species lived for a relatively short time and can be easily distinguished from others on the basis of specific features. For example, ammonites have a distinctive feature known as the suture line — where the internal shell layers that separate the individual chambers (septae) meet the outer shell wall, as shown in Figure 8.12. These suture lines are sufficiently variable to identify species that can be used to estimate the relative or absolute ages of the rocks in which they are found.

Figure 8.12 The septum of an ammonite (white part, left), and the suture lines where the septae meet the outer shell (right). [SE]

Foraminifera (small, carbonate-shelled marine organisms that originated during the Triassic and are still around today) are also useful biozone fossils. As shown in Figure 8.13, numerous different foraminifera lived during the Cretaceous. Some lasted for over 10 million years, but others for less than 1 million years. If the foraminifera in a rock can be identified to the species level, we can get a good idea of its age.

Figure 8.13 Time ranges for Cretaceous foraminifera (left) and modern foraminifera from the Ambergris area of Belize (right) [left: SE, from data in Scott, R, 2014, A Cretaceous chronostratigraphic database: construction and applications, Carnets de Géologie, Vol. 14., right : SE]

Originally fossils only provided us with relative ages because, although early paleontologists understood biological succession, they did not know the absolute ages of the different organisms. It was only in the early part of the 20th century, when isotopic dating methods were first applied, that it became possible to discover the absolute ages of the rocks containing fossils. In most cases, we cannot use isotopic techniques to directly date fossils or the sedimentary rocks they are found in, but we can constrain their ages by dating igneous rocks that cut across sedimentary rocks, or volcanic ash layers that lie within sedimentary layers.

Isotopic dating of rocks, or the minerals in them, is based on the fact that we know the decay rates of certain unstable isotopes of elements and that these rates have been constant over geological time. It is also based on the premise that when the atoms of an element decay within a mineral or a rock, they stay there and don’t escape to the surrounding rock, water, or air. One of the isotope pairs widely used in geology is the decay of ⁴⁰K to ⁴⁰Ar (potassium-40 to argon-40). ⁴⁰K is a radioactive isotope of potassium that is present in very small amounts in all minerals that have potassium in them. It has a half-life of 1.3 billion years, meaning that over a period of 1.3 Ga one-half of the ⁴⁰K atoms in a mineral or rock will decay to ⁴⁰Ar, and over the next 1.3 Ga one-half of the remaining atoms will decay, and so on (Figure 8.14).

Figure 8.14 The decay of 40K over time. Each half-life is 1.3 billion years, so after 3.9 billion years (three half-lives) 12.5% of the original 40K will remain. The red-blue bars represent 40K and the green-yellow bars represent 40Ar. [SE]

In order to use the K-Ar dating technique, we need to have an igneous or metamorphic rock that includes a potassium-bearing mineral. One good example is granite, which normally has some potassium feldspar (Figure 8.15). Feldspar does not have any argon in it when it forms. Over time, the ⁴⁰K in the feldspar decays to ⁴⁰Ar. Argon is a gas and the atoms of ⁴⁰Ar remain embedded within the crystal, unless the rock is subjected to high temperatures after it forms. The sample must be analyzed using a very sensitive mass-spectrometer, which can detect the differences between the masses of atoms, and can therefore distinguish between ⁴⁰K and the much more abundant ³⁹K. Biotite and hornblende are also commonly used for K-Ar dating.

Figure 8.15 Crystals of potassium feldspar (pink) in a granitic rock are candidates for isotopic dating using the K-Ar method because they contained potassium and no argon when they formed. [SE]

Why can’t we use isotopic dating techniques with sedimentary rocks?

An important assumption that we have to be able to make when using isotopic dating is that when the rock formed none of the daughter isotope was present (e.g., ⁴⁰Ar in the case of the K-Ar method). A clastic sedimentary rock is made up of older rock and mineral fragments, and when the rock forms it is almost certain that all of the fragments already have daughter isotopes in them. Furthermore, in almost all cases, the fragments have come from a range of source rocks that all formed at different times. If we dated a number of individual grains in the sedimentary rock, we would likely get a range of different dates, all older than the age of the rock. It might be possible to date some chemical sedimentary rocks isotopically, but there are no useful isotopes that can be used on old chemical sedimentary rocks. Radiocarbon dating can be used on sediments or sedimentary rocks that contain carbon, but it cannot be used on materials older than about 60 ka.

K-Ar is just one of many isotope-pairs that are useful for dating geological materials. Some of the other important pairs are listed in Table 8.2, along with the age ranges that they apply to and some comments on their applications. When radiometric techniques are applied to metamorphic rocks, the results normally tell us the date of metamorphism, not the date when the parent rock formed.

Table 8.2 A few of the isotope systems that are widely used for dating geological materials

Radiocarbon dating (using ¹⁴C) can be applied to many geological materials, including sediments and sedimentary rocks, but the materials in question must be younger than 60 ka. Fragments of wood incorporated into young sediments are good candidates for carbon dating, and this technique has been used widely in studies involving late Pleistocene glaciers and glacial sediments. An example is shown in Figure 8.16; radiocarbon dates from wood fragments in glacial sediments have been used to estimate the time of the last glacial advance along the Strait of Georgia.

Figure 8.16 Radiocarbon dates on wood fragments in glacial sediments in the Strait of Georgia [SE after Clague, J, 1976, Quadra Sand and its relation to late Wisconsin glaciation of southeast British Columbia, Can. J. Earth Sciences, V. 13, p. 803-815]

There are numerous other techniques for dating geological materials, but we will examine just two of them here: tree-ring dating (i.e., dendrochronology) and dating based on the record of reversals of Earth’s magnetic field.

Dendrochronology can be applied to dating very young geological materials based on reference records of tree-ring growth going back many millennia. The longest such records can take us back over 25 ka, to the height of the last glaciation. One of the advantages of dendrochronology is that, providing reliable reference records are available, the technique can be used to date events to the nearest year.

Dendrochronology has been used to date the last major subduction zone earthquake on the coast of B.C., Washington, and Oregon. When large earthquakes strike in this setting, there is a tendency for some coastal areas to subside by one or two metres. Seawater then rushes in, flooding coastal flats and killing trees and other vegetation within a few months. There are at least four locations along the coast of Washington that have such dead trees (and probably many more in other areas). Wood samples from these trees have been studied and the ring patterns have been compared with patterns from old living trees in the region (Figure 8.17).

Figure 8.17 Example of tree-ring dating of dead trees [SE]

At all of the locations studied, the trees were found to have died either in the year 1699, or very shortly thereafter (Figure 8.18). On the basis of these results, it was concluded that a major earthquake took place in this region sometime between the end of growing season in 1699 and the beginning of the growing season in 1700. Evidence from a major tsunami that struck Japan on January 27, 1700, narrowed the timing of the earthquake to sometime in the evening of January 26, 1700. (For more information, see https://web.viu.ca/earle/1700-quake/)

Figure 8.18 Sites in Washington where dead trees are present in coastal flats. The outermost wood of eight trees was dated using dendrochronology, and of these, seven died during the year 1699, suggesting that the land was inundated by water at that time. [SE from data in Yamaguchi, D.K., B.F. Atwater, D.E. Bunker, B.E. Benson, and M.S. Reid. 1997. Tree-ring dating the 1700 Cascadia earthquake. Nature, Vol. 389, pp. 922 – 923, 30 October 1997.]

Changes in Earth’s magnetic field can also be used to date events in geologic history. The magnetic field makes compasses point toward the North Pole, but, as we’ll see in Chapter 10, this hasn’t always been the case. At various times in the past, Earth’s magnetic field has reversed itself completely, and during those times a compass would have pointed to the South Pole. By studying magnetism in volcanic rocks that have been dated isotopically, geologists have been able to delineate the chronology of magnetic field reversals going back for some 250 Ma. About 5 Ma of this record is shown in Figure 8.19, where the black bands represent periods of normal magnetism (“normal” meaning similar to the current magnetic field) and the white bands represent periods of reversed magnetism. These periods of consistent magnetic polarity are given names to make them easier to reference. The current normal magnetic field, known as Brunhes, has lasted for the past 780,000 years. Prior to that there was a short reversed period and then a short normal period known as Jaramillo.

Figure 8.19 The last 5 Ma of magnetic field reversals. [SE after U.S. Geological Survey, http://upload.wikimedia.org/wikipedia/commons/1/13/Geomagnetic_polarity_late_Cenozoic.svg.]

Oceanic crust becomes magnetized by the magnetic field that exists as the crust forms from magma. As it cools, tiny crystals of magnetite that form within the magma become aligned with the existing magnetic field and then remain that way after all of the rock has hardened, as shown in Figure 8.20. Crust that is forming today is being magnetized in a “normal” sense, but crust that formed 780,000 to 900,000 years ago, in the interval between the Brunhes and Jaramillo normal periods, was magnetized in the “reversed” sense.

Chapter 9 has a discussion of Earth’s magnetic field, including where and how it is generated and why its polarity changes periodically.

Figure 8.20 Depiction of the formation of magnetized oceanic crust at a spreading ridge. Coloured bars represent periods of normal magnetism, and the small capital letters denote the Brunhes, Jaramillio, Olduvai, and Gauss normal magnetic periods (see Figure 8.19). [SE]

Magnetic chronology can be used as a dating technique because we can measure the magnetic field of rocks using a magnetometer in a lab, or of entire regions by towing a magnetometer behind a ship or an airplane. For example, the Juan de Fuca Plate, which lies off of the west coast of B.C., Washington, and Oregon, is being and has been formed along the Juan de Fuca spreading ridge (Figure 8.21). The parts of the plate that are still close to the ridge have normal magnetism, while parts that are farther away (and formed much earlier) have either normal or reversed magnetism, depending on when the rock formed. By carefully matching the sea-floor magnetic stripes with the known magnetic chronology, we can determine the age at any point on the plate. We can see, for example, that the oldest part of the Juan de Fuca Plate that has not subducted (off the coast of Oregon) is just over 8 million years old, while the part that is subducting underneath Vancouver Island is between 0 and about 6 million years old.

Figure 8.21 The pattern of magnetism within the area of the Juan de Fuca Plate, off the west coast of North America. The coloured shapes represent parts of the sea floor that have normal magnetism, and the magnetic time scale is shown using the same colours. The blue bands represent Brunhes, Jaramillo, and Olduvai; the green represents Gauss; and so on. (Note that in this diagram, sea-floor magnetism is only shown for the Juan de Fuca Plate, although similar patterns exist on the Pacific Plate.) [SE]

It’s one thing to know the facts about geological time — how long it is, how we measure it, how we divide it up, and what we call the various periods and epochs — but it is quite another to really understand geological time. The problem is that our lives are short and our memories are even shorter. Our experiences span only a few decades, so we really don’t have a way of knowing what 11,700 years means. What’s more, it’s hard for us to understand how 11,700 years differs from 65.5 Ma, or even from 1.8 Ga. It’s not that we can’t comprehend what the numbers mean — we can all get that figured out with a bit of practice — but even if we do know the numerical meaning of 65.5 Ma, we can’t really appreciate how long ago it was.

You may be wondering why it’s so important to really “understand” geological time. There are some very good reasons. One is so that we can fully understand how geological processes that seem impossibly slow can produce anything of consequence. For example, we are familiar with the concept of driving from one major city to another: a journey of several hours at around 100 km/h. Continents move toward each other at rates of a fraction of a millimetre per day, or something in the order of 0.00000001 km/h, and yet, at this impossibly slow rate (try walking at that speed!), they can move thousands of kilometres. Sediments typically accumulate at even slower rates — less than a millimetre per year — but still they are thick enough to be thrust up into monumental mountains and carved into breathtaking canyons.

Another reason is that for our survival on this planet, we need to understand issues like extinction of endangered species and anthropogenic (human-caused) climate change. Some people, who don’t understand geological time, are quick to say that the climate has changed in the past, and that what is happening now is no different. And it certainly has changed in the past. For example, from the Eocene (50 Ma) to the present day, Earth’s climate cooled by about 12°C. That’s a huge change that ranks up there with many of the important climate changes of the distant past, and yet the rate of change over that time was only 0.000024°C/century. Anthropogenic climate change has been 1.1°C over the past 100 years,Climate change data from NASA Goddard Institute for Space Studies: http://data.giss.nasa.gov/gistemp/tabledata_v3/GLB.Ts.txt and that is 45,800 times faster than the rate of natural climate change since the Eocene!

One way to wrap your mind around geological time is to put it into the perspective of single year, because we all know how long it is from one birthday to the next. At that rate, each hour of the year is equivalent to approximately 500,000 years, and each day is equivalent to 12.5 million years.

If all of geological time is compressed down to a single year, Earth formed on January 1, and the first life forms evolved in late March (~3,500 Ma). The first large life forms appeared on November 13 (~600 Ma), plants appeared on land around November 24, and amphibians on December 3. Reptiles evolved from amphibians during the first week of December and dinosaurs and early mammals evolved from reptiles by December 13, but the dinosaurs, which survived for 160 million years, were gone by Boxing Day (December 26). The Pleistocene Glaciation got started at around 6:30 p.m. on New Year’s Eve, and the last glacial ice left southern Canada by 11:59 p.m.

It’s worth repeating: on this time scale, the earliest ancestors of the animals and plants with which we are familiar did not appear on Earth until mid-November, the dinosaurs disappeared after Christmas, and most of Canada was periodically locked in ice from 6:30 to 11:59 p.m. on New Year’s Eve. As for people, the first to inhabit B.C. got here about one minute before midnight, and the first Europeans arrived about two seconds before midnight.

It is common for the popular press to refer to distant past events as being “prehistoric.” For example, dinosaurs are reported as being “prehistoric creatures,” even by the esteemed National Geographic Society.http://science.nationalgeographic.com/science/prehistoric-world/ The written records of our history date back to about 6,000 years ago, so anything prior to that is considered “prehistoric.” But to call the dinosaurs prehistoric is equivalent to — and about as useful as — saying that Singapore is beyond the city limits of Kamloops! If we are going to become literate about geological time, we have to do better than calling dinosaurs, or early horses (54 Ma), or even early humans (2.8 Ma), “prehistoric.”

The topics covered in this chapter can be summarized as follows:

Introduction

In order to understand how Earth works, and especially the mechanisms of plate tectonics (covered in Chapter 10), we need to know something about the inside of our planet — what it’s made of, and what goes on in there. We have a variety of ways of knowing, and these will be discussed in this chapter, but the one thing we can’t do is go down and look! Fortunately there are a few places where mantle rock is exposed on Earth’s surface, and we have some samples of material from the insides of other planetary bodies, in the form of meteorites that have landed on Earth (Figure 9.1). We also have a great deal of seismic information that can help us understand the nature of Earth’s interior.

Figure 9.1 Left: a fragment of the Tagish Lake meteorite, discovered in 2000 on the ice of Tagish Lake, B.C. It is a “stony” meteorite that is dominated by ferromagnesian silicate minerals, and is similar in composition to Earth’s mantle. Right: part of the Elbogen meteorite discovered in Germany around 1400. It is an iron meteorite, similar in composition to Earth’s core. Both samples are a few centimetres across. [left from: http://www.nasa.gov/centers/goddard/images/content/557996main_tagish-lake-meteorite.jpg] right from: http://upload.wikimedia.org/wikipedia/commons/d/dc/Elbogen_meteorite%2C_8.9g.jpg]

Earth’s interior is broadly divided by composition and depth into crust, mantle, and core (Figure 9.2). The crust is primarily (~95%) made up of igneous rock and metamorphic rock with an overall composition between intermediate and felsic. The remaining 5% is made up of sedimentary rock, which is dominated by mudstone.

The mantle includes several layers, all with the same overall ultramafic composition. The upper mantle is typically composed of peridotite, a rock dominated by olivine and pyroxene. The lower mantle has a similar chemical composition, but because of the extreme pressures, different minerals are present, including spinels and garnets. The properties of the mantle also vary with depth, as follows:

• Lithosphere: solid
• Asthenosphere: partially liquid
• Upper and lower mantle: solid but plastic (the difference between the two is in the type of minerals)
• “D” layer (the part of the mantle within 200 km of the core): partially liquid
• The core-mantle boundary (CMB) is at a depth of 2,900 km.

The core is primarily composed of iron, with lesser amounts of nickel (about 5%) and several percent oxygen. It is extremely hot (~3500° to 5000°C). The outer core is liquid while the inner core is solid — even though it is hotter — because the pressure is so much greater at that depth.

Although the CMB is a little less than half of the way to Earth’s centre, the mantle, being on the outside, is by far the major component of Earth. The mantle makes up 82.5% of the volume, the core 16.1%, and the crust only 1.4%.

In the remainder of this chapter, we’ll look first at how we know about Earth’s interior structure, and then at the properties of the different layers and the processes that take place within them.

Figure 9.2 Earth’s layers: crust is pink, mantle is green, core is blue [SE]

Seismology is the study of vibrations within Earth. These vibrations are caused by various events, including earthquakes, extraterrestrial impacts, explosions, storm waves hitting the shore, and tidal effects. Of course, seismic techniques have been most widely applied to the detection and study of earthquakes, but there are many other applications, and arguably seismic waves provide the most important information that we have concerning Earth’s interior. Before going any deeper into Earth, however, we need to take a look at the properties of seismic waves. The types of waves that are useful for understanding Earth’s interior are called body waves, meaning that, unlike the surface waves on the ocean, they are transmitted through Earth materials.

Imagine hitting a large block of strong rock (e.g., granite) with a heavy sledgehammer (Figure 9.3). At the point where the hammer strikes it, a small part of the rock will be compressed by a fraction of a millimetre. That compression will transfer to the neighbouring part of the rock, and so on through to the far side of the rock, from where it will bounce back to the top — all in a fraction of a second. This is known as a compression wave, and it can be illustrated by holding a loose spring (like a Slinky) that is attached to something (or someone) at the other end. If you give it a sharp push so the coils are compressed, the compression propagates (travels) along the length of the spring and back (Figure 9.4). You can think of a compression wave as a “push” wave — it’s called a P-wave (although the “P” stands for “primary” because P-waves arrive first at seismic stations).

Figure 9.3 Hitting a large block of rock with a heavy hammer will create seismic waves within the rock. Please don’t try this at home! [SE]

When we hit a rock with a hammer, we also create a different type of body wave, one that is characterized by back-and-forth vibrations (as opposed to compressions). This is known as a shear wave (S-wave, where the “S” stands for “secondary”), and an analogy would be what happens when you flick a length of rope with an up-and-down motion. As shown in Figure 9.4, a wave will form in the rope, which will travel to the end of the rope and back.

Compression waves and shear waves travel very quickly through geological materials. As shown in Figure 9.5, typical P-wave velocities are between 0.5 km/s and 2.5 km/s in unconsolidated sediments, and between 3.0 km/s and 6.5 km/s in solid crustal rocks. Of the common rocks of the crust, velocities are greatest in basalt and granite. S-waves are slower than P-waves, with velocities between 0.1 km/s and 0.8 km/s in soft sediments, and between 1.5 km/s and 3.8 km/s in solid rocks.

Figure 9.4 A compression wave can be illustrated by a spring (like a Slinky) that is given a sharp push at one end. A shear wave can be illustrated by a rope that is given a quick flick. [SE]

Mantle rock is generally denser and stronger than crustal rock and both P- and S-waves travel faster through the mantle than they do through the crust. Moreover, seismic-wave velocities are related to how tightly compressed a rock is, and the level of compression increases dramatically with depth. Finally, seismic waves are affected by the phase state of rock. They are slowed if there is any degree of melting in the rock. If the material is completely liquid, P-waves are slowed dramatically and S-waves are stopped altogether.

Figure 9.5 Typical velocities of P-waves (red) and S-waves (blue) in sediments and in solid crustal rocks [SE after: US Env. Prot. Agency http://www.epa.gov/esd/cmb/GeophysicsWebsite/ pages/reference/properties/Geomechanical_ (Engineering)_ Properties.htm]

Figure 9.6a P-wave and S-wave velocity variations with depth in Earth. [SE]

Figure 9.6b P-wave and S-wave velocity variations in the upper mantle and crust (This is an expanded view of the upper 600 km of the curves in Figure 9.6a)

Accurate seismometers have been used for earthquake studies since the late 1800s, and systematic use of seismic data to understand Earth’s interior started in the early 1900s. The rate of change of seismic waves with depth in Earth (as shown in Figure 9.6) has been determined over the past several decades by analyzing seismic signals from large earthquakes at seismic stations around the world. Small differences in arrival time of signals at different locations have been interpreted to show that:

• Velocities are greater in mantle rock than in the crust.
• Velocities generally increase with pressure, and therefore with depth.
• Velocities slow in the area between 100 km and 250 km depth (called the “low-velocity zone”; equivalent to the asthenosphere).
• Velocities increase dramatically at 660 km depth (because of a mineralogical transition).
• Velocities slow in the region just above the core-mantle boundary (the D” layer or “ultra-low-velocity zone”).
• S-waves do not pass through the outer part of the core.
• P-wave velocities increase dramatically at the boundary between the liquid outer core and the solid inner core.
  

One of the first discoveries about Earth’s interior made through seismology was in the early 1900s when Croatian seismologist Andrija Mohorovičić (pronounced Moho-ro-vi-chich) realized that at certain distances from an earthquake, two separate sets of seismic waves arrived at a seismic station within a few seconds of each other. He reasoned that the waves that went down into the mantle, travelled through the mantle, and then were bent upward back into the crust, reached the seismic station first because although they had farther to go, they travelled faster through mantle rock (as shown in Figure 9.7). The boundary between the crust and the mantle is known as the Mohorovičić discontinuity (or Moho). Its depth is between 60 km and 80 km beneath major mountain ranges, around 30 km to 50 km beneath most of the continental crust, and between 5 km and 10 km beneath the oceanic crust.

Figure 9.7 Depiction of seismic waves emanating from an earthquake (red star). Some waves travel through the crust to the seismic station (at about 6 km/s), while others go down into the mantle (where they travel at around 8 km/s) and are bent upward toward the surface, reaching the station before the ones that travelled only through the crust. [SE]

Our current understanding of the patterns of seismic wave transmission through Earth is summarized in Figure 9.8. Because of the gradual increase in density (and therefore rock strength) with depth, all waves are refracted (toward the lower density material) as they travel through homogenous parts of Earth and thus tend to curve outward toward the surface. Waves are also refracted at boundaries within Earth, such as at the Moho, at the core-mantle boundary (CMB), and at the outer-core/inner-core boundary.

S-waves do not travel through liquids — they are stopped at the CMB — and there is an S-wave shadow on the side of Earth opposite a seismic source. The angular distance from the seismic source to the shadow zone is 103° on either side, so the total angular distance of the shadow zone is 154°. We can use this information to infer the depth to the CMB.

P-waves do travel through liquids, so they can make it through the liquid part of the core. Because of the refraction that takes place at the CMB, waves that travel through the core are bent away from the surface, and this creates a P-wave shadow zone on either side, from 103° to 150°. This information can be used to discover the differences between the inner and outer parts of the core.

Figure 9.8 Patterns of seismic wave propagation through Earth’s mantle and core. S-waves do not travel through the liquid outer core, so they leave a shadow on Earth’s far side. P-waves do travel through the core, but because the waves that enter the core are refracted, there are also P-wave shadow zones. [SE]

Using data from many seismometers and hundreds of earthquakes, it is possible to create a two- or three-dimensional image of the seismic properties of part of the mantle. This technique is known as seismic tomography, and an example of the result is shown in Figure 9.9.

Figure 9.9 P-wave tomographic profile of area in the southern Pacific Ocean from southeast of Tonga to Fiji. Blue represents rock that has relatively high seismic velocities, while yellow and red represent rock with low velocities. Open circles are earthquakes used in the study. [from: Zhao, D., Y. Xu, D.A. Wiens, L. Dorman, J. Hildebrand, and S. Webb, Depth extent of the Lau back-arc spreading center and its relationship to the subduction process, Science, 278, 254-257, 1997, used with permission]

The Pacific Plate subducts beneath Tonga and appears in Figure 9.9 as a 100 km thick slab of cold (blue-coloured) oceanic crust that has pushed down into the surrounding hot mantle. The cold rock is more rigid than the surrounding hot mantle rock, so it is characterized by slightly faster seismic velocities. There is volcanism in the Lau spreading centre and also in the Fiji area, and the warm rock in these areas has slower seismic velocities (yellow and red colours).

As we’ve discussed in the context of metamorphism, Earth’s internal temperature increases with depth. However, as shown in Figure 9.10, that rate of increase is not linear. The temperature gradient is around 15° to 30°C/km within the upper 100 km; it then drops off dramatically through the mantle, increases more quickly at the base of the mantle, and then increases slowly through the core. The temperature is around 1000°C at the base of the crust, around 3500°C at the base of the mantle, and around 5,000°C at Earth’s centre. The temperature gradient within the lithosphere (upper 100 km) is quite variable depending on the tectonic setting. Gradients are lowest in the central parts of continents, higher in the vicinity of subduction zones, and higher still at divergent boundaries.

Figure 9.10 Generalized rate of temperature increase with depth within Earth. Temperature increases to the right, so the flatter the line, the steeper the temperature gradient. Our understanding of the temperature gradient comes from seismic wave information and knowledge of the melting points of Earth’s materials. [SE]

Figure 9.11 Rate of temperature increase with depth in Earth’s upper 500 km, compared with the dry mantle rock melting curve (red dashed line). LVZ= low-velocity zone [SE]

Figure 9.11 shows a typical temperature curve for the upper 500 km of the mantle, in comparison with the melting curve for dry mantle rock. Within the depth interval between 100 and 250 km, the temperature curve comes very close to the melting boundary for dry mantle rock. At these depths, therefore, mantle rock is either very nearly melted or partially melted. In some situations, where extra heat is present and the temperature line crosses over the melting line, or where water is present, it may be completely molten. This region of the mantle is known as the low-velocity zone because seismic waves are slowed within rock that is near its melting point, and of course it is also known as the asthenosphere. Below 250 km, the temperature stays on the left side of the melting line; in other words, the mantle is solid from here all the way down to the core-mantle boundary. 

The fact that the temperature gradient is much less in the main part of the mantle than in the lithosphere has been interpreted to indicate that the mantle is convecting, and therefore that heat from depth is being brought toward the surface faster than it would be with only heat conduction. As we’ll see in Chapter 10, a convecting mantle is an essential feature of plate tectonics.

The convection of the mantle is a product of the transfer of heat from the core to the lower mantle. As in a pot of soup on a hot stove (Figure 9.12), the material near the heat source becomes hot and expands, making it lighter than the material above. The force of buoyancy causes it to rise, and cooler material flows in from the sides. The mantle convects in this way because the heat transfer from below is not perfectly even, and also because, even though mantle material is solid rock, it is sufficiently plastic to slowly flow (at rates of centimetres per year) as long as a steady force is applied to it.

As in the soup pot example, Earth’s mantle will no longer convect once the core has cooled to the point where there is not enough heat transfer to overcome the strength of the rock. This has already happened on smaller planets like Mercury and Mars, as well as on Earth’s Moon.

Figure 9.12 Convection in a pot of soup on a hot stove (left). As long as heat is being transferred from below, the liquid will convect. If the heat is turned off (right), the liquid remains hot for a while, but convection will cease. [SE]

Why is the inside of Earth hot?

The heat of Earth’s interior comes from two main sources, each contributing about 50% of the heat. One of those is the frictional heat left over from the collisions of large and small particles that created Earth in the first place, plus the subsequent frictional heat of redistribution of material within Earth by gravitational forces (e.g., sinking of iron to form the core).

The other source is radioactivity, specifically the spontaneous radioactive decay of ²³⁵U, ²³⁸U, ⁴⁰K, and ²³²Th, which are primarily present in the mantle. As shown on this figure, the total heat produced that way has been decreasing over time (because these isotopes are getting used up), and is now roughly 25% of what it was when Earth formed. This means that Earth’s interior is slowly becoming cooler.

[Image by SE, after Arevalo, R, McDonough, W and Luong, M, 2009, The K/U ratio of Earth: insights into mantle composition, structure and thermal evolution, Earth and Planetary Science Letters, V 278, p. 361-369.]

Heat is also being transferred from the solid inner core to the liquid outer core, and this leads to convection of the liquid iron of the outer core. Because iron is a metal and conducts electricity (even when molten), its motion generates a magnetic field.

Earth’s magnetic field is defined by the North and South Poles that align generally with the axis of rotation (Figure 9.13). The lines of magnetic force flow into Earth in the northern hemisphere and out of Earth in the southern hemisphere. Because of the shape of the field lines, the magnetic force trends at different angles to the surface in different locations (red arrows of Figure 9.13). At the North and South Poles, the force is vertical. Anywhere on the equator the force is horizontal, and everywhere in between, the magnetic force is at some intermediate angle to the surface. As we’ll see in Chapter 10, the variations in these orientations provide a critical piece of evidence to the understanding of continental drift as an aspect of plate tectonics.

Earth’s magnetic field is generated within the outer core by the convective movement of liquid iron, but as we discovered in Chapter 8, the magnetic field is not stable over geological time. For reasons that are not completely understood, the magnetic field decays periodically and then becomes re-established. When it does re-establish, it may be oriented the way it was before the decay, or it may be oriented with the reversed polarity. Over the past 250 Ma, there have a few hundred magnetic field reversals, and their timing has been anything but regular. The shortest ones that geologists have been able to define lasted only a few thousand years, and the longest one was more than 30 million years, during the Cretaceous (Figure 9.14).

Figure 9.13 Depiction of Earth’s magnetic field as a bar magnet coinciding with the core. The south pole of such a magnet points to Earth’s North Pole. The red arrows represent the orientation of the magnetic field at various locations on Earth’s surface. [SE after: http://upload.wikimedia.org/wikipedia/commons/ 1/17/Earths_Magnetic_Field_ Confusion.svg]

Figure 9.14 Magnetic field reversal chronology for the past 170 Ma. The first 5 Ma of the magnetic chronology are shown in more detail in Figure 9.15. [SE after: http://upload.wikimedia.org/wikipedia/en/c/c0/Geomagnetic_polarity_0-169_Ma.svg]

Changes in Earth’s magnetic field have been studied using a mathematical model, and reversals have been shown to take place when the model was run to simulate a period of several hundred thousand years. The fact that field reversals took place shows that the model is a reasonably accurate representation of the Earth. According to the lead author of the study, Gary Glatzmaier, of University of California Santa Cruz: “Our solution shows how convection in the fluid outer core is continually trying to reverse the field but that the solid inner core inhibits magnetic reversals because the field in the inner core can only change on the much longer time scale of diffusion. Only once in many attempts is a reversal successful, which is probably the reason why the times between reversals of the Earth’s field are long and randomly distributed.” A depiction of Earth’s magnetic field lines during a stable period and during a reversal is shown in Figure 9.15. To read more about these phenomena see Glatzmaier’s Geodynamo website at: http://es.ucsc.edu/~glatz/geodynamo.html.

Figure 9.15 Depiction of Earth’s magnetic field between reversals (left) and during a reversal (right). The lines represent magnetic field lines: blue where the field points toward Earth’s centre and yellow where it points away. The rotation axis of Earth is vertical, and the outline of the core is shown as a dashed white circle. [from: http://en.wikipedia.org/wiki/Geomagnetic_reversal]

Theory holds that the mantle is able to convect because of its plasticity, and this property also allows for another very important Earth process known as isostasy. The literal meaning of the word isostasy is “equal standstill,” but the importance behind it is the principle that Earth’s crust is floating on the mantle, like a raft floating in the water, rather than resting on the mantle like a raft sitting on the ground.

The relationship between the crust and the mantle is illustrated in Figure 9.16. On the right is an example of a non-isostatic relationship between a raft and solid concrete. It’s possible to load the raft up with lots of people, and it still won’t sink into the concrete. On the left, the relationship is an isostatic one between two different rafts and a swimming pool full of peanut butter. With only one person on board, the raft floats high in the peanut butter, but with three people, it sinks dangerously low. We’re using peanut butter here, rather than water, because its viscosity more closely represents the relationship between the crust and the mantle. Although it has about the same density as water, peanut butter is much more viscous (stiff), and so although the three-person raft will sink into the peanut butter, it will do so quite slowly.

Figure 9.16 Illustration of a non-isostatic relationship between a raft and solid ground (right) and of isostatic relationships between rafts and peanut butter (left). [SE]

The relationship of Earth’s crust to the mantle is similar to the relationship of the rafts to the peanut butter. The raft with one person on it floats comfortably high. Even with three people on it the raft is less dense than the peanut butter, so it floats, but it floats uncomfortably low for those three people. The crust, with an average density of around 2.6 grams per cubic centimetre (g/cm3), is less dense than the mantle (average density of approximately 3.4 g/cm3 near the surface, but more than that at depth), and so it is floating on the “plastic” mantle. When more weight is added to the crust, through the process of mountain building, it slowly sinks deeper into the mantle and the mantle material that was there is pushed aside (Figure 9.17, left). When that weight is removed by erosion over tens of millions of years, the crust rebounds and the mantle rock flows back (Figure 9.17, right).

Figure 9.17 Illustration of the isostatic relationship between the crust and the mantle. Following a period of mountain building, mass has been added to a part of the crust, and the thickened crust has pushed down into the mantle (left). Over the following tens of millions of years, the mountain chain is eroded and the crust rebounds (right). The green arrows represent slow mantle flow. [SE]

The crust and mantle respond in a similar way to glaciation. Thick accumulations of glacial ice add weight to the crust, and as the mantle beneath is squeezed to the sides, the crust subsides. This process is illustrated for the current ice sheet on Greenland in Figure 9.18. The Greenland Ice Sheet at this location is over 2,500 m thick, and the crust beneath the thickest part has been depressed to the point where it is below sea level over a wide area. When the ice eventually melts, the crust and mantle will slowly rebound, but full rebound will likely take more than 10,000 years.

Figure 9.18a A cross-section through the crust in the northern part of Greenland (The ice thickness is based on data from NASA and the Center for Remote Sensing of Ice Sheets, but the crust thickness is less than it should be for the sake of illustration.) The maximum ice thickness is over 2,500 m. The red arrows represent downward pressure on the mantle because of the mass of the ice.

Figure 9.18b Depiction of the situation after complete melting of the ice sheet, a process that could happen within 2,000 years if people and their governments continue to ignore climate change. The isostatic rebound of the mantle would not be able to keep up with this rate of melting, so for several thousand years the central part of Greenland would remain close to sea level, in some areas even below sea level.

Figure 9.18c It is likely that complete rebound of the mantle beneath Greenland would take more than 10,000 years.

How can the mantle be both solid and plastic?

You might be wondering how it is possible that Earth’s mantle is rigid enough to break during an earthquake, and yet it convects and flows like a very viscous liquid. The explanation is that the mantle behaves as a non-Newtonian fluid, meaning that it responds differently to stresses depending on how quickly the stress is applied. A good example of this is the behaviour of the material known as Silly Putty, which can bounce and will break if you pull on it sharply, but will deform in a liquid manner if stress is applied slowly. In this photo, Silly Putty was placed over a hole in a glass tabletop, and in response to gravity, it slowly flowed into the hole. The mantle will flow when placed under the slow but steady stress of a growing (or melting) ice sheet.

[https://upload.wikimedia.org/wikipedia/commons/f/f3/Silly_putty_dripping.jpg]

Figure 9.19  The current rates of post-glacial isostatic uplift (green, blue, and purple shades) and subsidence (yellow and orange). Subsidence is taking place where the mantle is slowly flowing back toward areas that are experiencing post-glacial uplift. [From: Paulson, A., S. Zhong, and J. Wahr. Inference of mantle viscosity from GRACE and relative sea level data, Geophys. J. Int. (2007) 171, 497–508. Accessed at: http://en.wikipedia.org/wiki/Hudson_Bay#/media/File:PGR_Paulson2007_Rate_of_Lithospheric_Uplift_due_to_PGR.png]

Large parts of Canada are still rebounding as a result of the loss of glacial ice over the past 12 ka, and as shown in Figure 9.19, other parts of the world are also experiencing isostatic rebound. The highest rate of uplift is in within a large area to the west of Hudson Bay, which is where the Laurentide Ice Sheet was the thickest (over 3,000 m). Ice finally left this region around 8,000 years ago, and the crust is currently rebounding at a rate of nearly 2 cm/year. Strong isostatic rebound is also occurring in northern Europe where the Fenno-Scandian Ice Sheet was thickest, and in the eastern part of Antarctica, which also experienced significant ice loss during the Holocene.

There are also extensive areas of subsidence surrounding the former Laurentide and Fenno-Scandian Ice Sheets. During glaciation, mantle rock flowed away from the areas beneath the main ice sheets, and this material is now slowly flowing back, as illustrated in Figure 9.18b.

The topics covered in this chapter can be summarized as follows:

Introduction

As we discovered in Chapter 1, plate tectonics is the model or theory that we use to understand how our planet works. More specifically it is a model that explains the origins of continents and oceans, folded rocks and mountain ranges, igneous and metamorphic rocks, earthquakes and volcanoes, and continental drift. Plate tectonics was first proposed just over 100 years ago, but did not become an accepted part of geology until about 50 years ago. It took 50 years for this theory to become accepted for a few reasons. First, it was a true revolution in thinking about Earth, which was difficult for many established geologists. Second, there was a political gulf between the main proponent of the theory Alfred Wegener (from Germany) and the geological establishment of the day, which was mostly centred in Britain and the United States. Third, the evidence and understanding of Earth that would have supported plate tectonic theory simply didn’t exist until the middle of the 20th century.

Alfred Wegener (1880-1930) (Figure 10.1) earned a PhD in astronomy at the University of Berlin in 1904, but he had always been interested in geophysics and meteorology and spent most of his academic career working in meteorology. In 1911 he happened on a scientific publication that included a description of the existence of matching Permian-aged terrestrial fossils in various parts of South America, Africa, India, Antarctica, and Australia (Figure 10.2).

Wegener concluded that this distribution of fossils could only exist if these continents were joined together during the Permian, and he coined the term Pangea (“all land”) for the supercontinent that he thought included all of the present-day continents.

Figure 10.1 Alfred Wegener a few years before his death in 1930 [http://upload.wikimedia.org/wikipedia/ commons/6/65/Alfred_Wegener_ca.1924-30.jpg]

Figure 10.2 The distribution of several Permian terrestrial fossils that are present in various parts of continents that are now separated by oceans. During the Permian, the supercontinent Pangea included the supercontinent Gondwana, shown here, along with North America and Eurasia.

Wegener pursued his theory with determination — combing the libraries, consulting with colleagues, and making observations — looking for evidence to support it. He relied heavily on matching geological patterns across oceans, such as sedimentary strata in South America matching those in Africa (Figure 10.3), North American coalfields matching those in Europe, and the mountains of Atlantic Canada matching those of northern Britain — both in morphology and rock type. Wegener also referred to the evidence for the Carboniferous and Permian (~300 Ma) Karoo Glaciation in South America, Africa, India, Antarctica, and Australia (Figure 10.4). He argued that this could only have happened if these continents were once all connected as a single supercontinent. He also cited evidence (based on his own astronomical observations) that showed that the continents were moving with respect to each other, and determined a separation rate between Greenland and Scandinavia of 11 m per year, although he admitted that the measurements were not accurate. In fact they weren’t even close — the separation rate is actually about 2.5 cm per year!

Figure 10.3 A cross-section showing the geological similarities between parts of Brazil on the left and Angola (Africa) on the right. The pink layer is a salt deposit, which is now known to be common in areas of continental rifting. [Source: U.S. Energy Information Administration (March 2015) http://www.eia.gov/countries/analysisbriefs/Angola/angola.pdf]

Figure 10.4 The distribution of the Carboniferous and Permian Karoo Glaciation (in blue) [SE, after http://upload.wikimedia.org/wikipedia/commons/9/96/Karoo_Glaciation.png]

Wegener first published his ideas in 1912 in a short book called Die Entstehung der Kontinente (The Origin of Continents), and then in 1915 in Die Entstehung der Kontinente und Ozeane (The Origin of Continents and Oceans). He revised this book several times up to 1929. It was translated into French, English, Spanish, and Russian in 1924.

In fact the continental fits were not perfect and the geological matchups were not always consistent, but the most serious problem of all was that Wegener could not conceive of a good mechanism for moving the continents around. It was understood by this time that the continents were primarily composed of sialic material (SIAL: silicon and aluminum dominated), and that the ocean floors were primarily simatic (SIMA: silicon and magnesium dominated). Wegener proposed that the continents were like icebergs floating on the heavier SIMA crust, but the only forces that he could invoke to propel continents around were poleflucht, the effect of Earth’s rotation pushing objects toward the equator, and the lunar and solar tidal forces, which tend to push objects toward the west. It was quickly shown that these forces were far too weak to move continents, and without any reasonable mechanism to make it work, Wegener’s theory was quickly dismissed by most geologists of the day.

Alfred Wegener died in Greenland in 1930 while carrying out studies related to glaciation and climate. At the time of his death, his ideas were tentatively accepted by only a small minority of geologists, and soundly rejected by most. However, within a few decades that was all to change. For more about his extremely important contributions to Earth science, visit this NASA website: http://earthobservatory.nasa.gov/Library/Giants/Wegener/

The untimely death of Alfred Wegener didn’t solve any problems for those who opposed his ideas because they still had some inconvenient geological truths to deal with. One of those was explaining the distribution of terrestrial species across five continents that are currently separated by hundreds or thousands of kilometres of ocean water (Figure 10.2), and another was explaining the origin of extensive fold-belt mountains, such as the Appalachians, the Alps, the Himalayas, and the Canadian Rockies.

Before we go any further, it is important to know what was generally believed about global geology before plate tectonics. At the beginning of the 20th century, geologists had a good understanding of how most rocks were formed and understood their relative ages through interpretation of fossils, but there was considerable controversy regarding the origin of mountain chains, especially fold-belt mountains. At the end of the 19th century, one of the prevailing views on the origin of mountains was the theory of contractionism — the idea that since Earth is slowly cooling, it must also be shrinking. In this scenario, mountain ranges had formed like the wrinkles on a dried-up apple, and the oceans had submerged parts of former continents. While this theory helped to address the dilemma of the terrestrial fossils, it came with its own set of problems, one being that the amount of cooling couldn’t produce the necessary amount of shrinking, and the other being the principle of isostasy (which had already been around for several decades), which wouldn’t allow continents to sink. (See Section 9.4 for a review of the important principle of isostasy.)

Another widely held view was permanentism, in which it was believed that the continents and oceans have always been generally as they are today. This view incorporated a mechanism for creation of mountain chains known as the geosyncline theory. A geosyncline is a thick deposit of sediments and sedimentary rocks, typically situated along the edge of a continent (Figure 10.5).

Figure 10.5 The development of a geosyncline along a continental margin. (Note that a geosyncline is not related to a syncline, which is a downward fold in sedimentary rocks.) [SE]

The idea of geosynclines developing into fold-belt mountains originated in the middle of the 19th century, proposed first by James Hall and later elaborated by Dwight Dana, both of whom worked extensively in the Appalachian Mountains of the eastern United States. The process of converting a geosyncline into a mountain belt was never really adequately explained, although it was widely believed that mountain belts formed when geosynclines were compressed by forces pushing from either side. The problem is that, without the lateral forces related to plate tectonics, no one was able to adequately describe what would do the pushing. The sediments that accumulate within a geosyncline are derived from erosion of the adjacent continent. Geosynclinal sediments — which eventually turn into sedimentary rocks — may be many thousands of metres thick. As they accumulate, they push down the pre-existing crustal rocks. Extensive geosynclinal deposits exist around much of the coastline of most of the continents; there is a large geosyncline along the eastern edge of North America.

Proponents of the geosyncline theory of mountain formation, and there were many well into the 1960s, also had the problem of explaining the intercontinental terrestrial fossil matchups. The simple explanation was that there were “land bridges” across the Atlantic along which animals and plants could migrate back and forth. One proponent of this idea was the American naturalist Ernest Ingersoll. Referring to evidence of past climate changes, Ingersoll contributed the following to the Encyclopedia Americana in 1920: “The most interesting feature of these changes, however, is that by which, now and again, the Old World was connected with the New by necks or spaces of land, known as “land-bridges”; especially as these permitted an interchange of plants and animals, giving to us many new ones from the other side of the ocean, including, finally, man himself.”http://en.wikisource.org/wiki/The_Encyclopedia_Americana_(1920)/Land-Bridges_Across_the_Oceans

There are many problems with the land-bridge theory, one being that it is completely inconsistent with isostasy, and another that there is no evidence of the remnants of the land bridges. The Atlantic Ocean is several thousand metres deep over wide areas, and so the underwater slopes leading up to a land bridge would have to have been at least tens of kilometres wide in most places, and many times that in others. A land bridge of that size would certainly have left some trace.

Exercises

Exercise 10.1 Fitting the Continents Together

The main continents around the Atlantic Ocean are depicted here in the shapes that they might have had during the Mesozoic, including the extents of their continental shelves. Cut these shapes out and see how well you can fit them together in the positions that these areas occupied within Pangea. You can refer to a map of Pangea to help you make the fit.

As the mineral magnetite (Fe₃O₄) crystallizes from magma, it becomes magnetized with an orientation parallel to that of Earth’s magnetic field at that time. This is called remnant magnetism. Rocks like basalt, which cool from a high temperature and commonly have relatively high levels of magnetite, are particularly susceptible to being magnetized in this way, but even sediments and sedimentary rocks, as long as they have small amounts of magnetite, will take on remnant magnetism because the magnetite grains gradually become reoriented following deposition. By studying both the horizontal and vertical components of the remnant magnetism, one can tell not only the direction to magnetic north at the time of the rock’s formation, but also the latitude where the rock formed relative to magnetic north.

In the early 1950s, a group of geologists from Cambridge University, including Keith Runcorn, Ted Irving,Ted Irving later set up a paleomagnetic lab at the Geological Survey of Canada in Sidney, B.C., and did a great deal of important work on understanding the geology of western North America. and several others, started looking at the remnant magnetism of Phanerozoic British and European volcanic rocks, and collecting paleomagnetic data. They found that rocks of different ages sampled from generally the same area showed quite different apparent magnetic pole positions (Figure 10.6). They initially assumed that this meant that Earth’s magnetic field had, over time, departed significantly from its present position — which is close to the rotational pole.

Figure 10.6 Apparent polar-wandering paths (APWP) for Eurasia and North America. The view is from the North Pole (black dot) looking down. The outer circle is the equator. In the diagram to the right the curve locations have been corrected taking continental drift into account. [SE]

The curve defined by the paleomagnetic data was called a polar wandering path because Runcorn and his students initially thought that their data represented actual movement of the magnetic poles (since geophysical models of the time suggested that the magnetic poles did not need to be aligned with the rotational poles). We now know that the magnetic data define movement of continents, and not of the magnetic poles, so we call it an apparent polar wandering path (APWP).

What is a polar wandering path?

At around 500 Ma, what we now call Europe was south of the equator, and so European rocks formed then would have acquired an upward-pointing magnetic field orientation (see Figure 9.13 and the figure shown here). Between then and now, Europe gradually moved north, and the rocks forming at various times acquired steeper and steeper downward-pointing magnetic orientations.

When researchers evaluated magnetic data in this way in the 1950s, they plotted where the North Pole would have appeared to be based on the magnetic data and assumed that the continent was always where it is now. That means that the 500 Ma “apparent” north pole would have been somewhere in the South Pacific, and that over the following 500 million years it would have gradually moved north.

Of course we now know that the magnetic poles don’t move around much (although polarity reversals do take place) and that the reason Europe had a magnetic orientation characteristic of the southern hemisphere is that it was in the southern hemisphere at 500 Ma.

Runcorn and colleagues soon extended their work to North America, and this also showed apparent polar wandering, but the results were not consistent with those from Europe. For example, the 200 Ma pole for North America plotted somewhere in China, while the 200 Ma pole for Europe plotted in the Pacific Ocean. Since there could only have been one pole position at 200 Ma, this evidence strongly supported the idea that North America and Europe had moved relative to each other since 200 Ma. Subsequent paleomagnetic work showed that South America, Africa, India, and Australia also have unique polar wandering curves. In 1956, Runcorn changed his mind and became a proponent of continental drift.

This paleomagnetic work of the 1950s was the first new evidence in favour of continental drift, and it led a number of geologists to start thinking that the idea might have some merit. Nevertheless, for a majority of geologists working on global geology at the time, this type of evidence was not sufficiently convincing to get them to change their views.

During the 20th century, our knowledge and understanding of the ocean basins and their geology increased dramatically. Before 1900, we knew virtually nothing about the bathymetry and geology of the oceans. By the end of the 1960s, we had detailed maps of the topography of the ocean floors, a clear picture of the geology of ocean floor sediments and the solid rocks underneath them, and almost as much information about the geophysical nature of ocean rocks as of continental rocks.

Up until about the 1920s, ocean depths were measured using weighted lines dropped overboard. In deep water this is a painfully slow process and the number of soundings in the deep oceans was probably fewer than 1,000. That is roughly one depth sounding for every 350,000 square kilometres of the ocean. To put that in perspective, it would be like trying to describe the topography of British Columbia with elevation data for only a half a dozen points! The voyage of the Challenger in 1872 and the laying of trans-Atlantic cables had shown that there were mountains beneath the seas, but most geologists and oceanographers still believed that the oceans were essentially vast basins with flat bottoms, filled with thousands of metres of sediments.

Following development of acoustic depth sounders in the 1920s (Figure 10.7), the number of depth readings increased by many orders of magnitude, and by the 1930s, it had become apparent that there were major mountain chains in all of the world’s oceans. During and after World War II, there was a well-organized campaign to study the oceans, and by 1959, sufficient bathymetric data had been collected to produce detailed maps of all the oceans (Figure 10.8).

Figure 10.7 Depiction of a ship-borne acoustic depth sounder. The instrument emits a sound (black arcs) that bounces off the sea floor and returns to the surface (white arcs). The travel time is proportional to the water depth. [SE]

Figure 10.8 Ocean floor bathymetry (and continental topography). Inset (a): the mid-Atlantic ridge, (b): the Newfoundland continental shelf, (c): the Nazca trench adjacent to South America, and (d): the Hawaiian Island chain. [SE after NOAA, http://upload.wikimedia.org/wikipedia/commons/9/93/Elevation.jpg]

The important physical features of the ocean floor are:

• Extensive linear ridges (commonly in the central parts of the oceans) with water depths in the order of 2,000 to 3,000 m (Figure 10.8, inset a)
• Fracture zones perpendicular to the ridges (inset a)
• Deep-ocean plains at depths of 5,000 to 6,000 m (insets a and d)
• Relatively flat and shallow continental shelves with depths under 500 m (inset b)
• Deep trenches (up to 11,000 m deep), most near the continents (inset c)
• Seamounts and chains of seamounts (inset d)

Seismic reflection sounding involves transmitting high-energy sound bursts and then measuring the echos with a series of geophones towed behind a ship. The technique is related to acoustic sounding as described above; however, much more energy is transmitted and the sophistication of the data processing is much greater. As the technique evolved, and the amount of energy was increased, it became possible to see through the sea-floor sediments and map the bedrock topography and crustal thickness. Hence sediment thicknesses could be mapped, and it was soon discovered that although the sediments were up to several thousands of metres thick near the continents, they were relatively thin — or even non-existent — in the ocean ridge areas (Figure 10.9). The seismic studies also showed that the crust is relatively thin under the oceans (5 km to 6 km) compared to the continents (30 km to 60 km) and geologically very consistent, composed almost entirely of basalt.

Figure 10.9 Topographic section at an ocean ridge based on reflection seismic data. Sediments are not thick enough to be detectable near the ridge, but get thicker on either side. The diagram represents approximately 50 km width, and has a 10x vertical exaggeration. [SE]

In the early 1950s, Edward Bullard, who spent time at the University of Toronto but is mostly associated with Cambridge University, developed a probe for measuring the flow of heat from the ocean floor. Bullard and colleagues found the rate to be higher than average along the ridges, and lower than average in the trench areas. Although Bullard was a plate-tectonics sceptic, these features were interpreted to indicate that there is convection within the mantle — the areas of high heat flow being correlated with upward convection of hot mantle material, and the areas of low heat flow being correlated with downward convection.

With developments of networks of seismographic stations in the 1950s, it became possible to plot the locations and depths of both major and minor earthquakes with great accuracy. It was found that there is a remarkable correspondence between earthquakes and both the mid-ocean ridges and the deep ocean trenches. In 1954 Gutenberg and Richter showed that the ocean-ridge earthquakes were all relatively shallow, and confirmed what had first been shown by Benioff in the 1930s — that earthquakes in the vicinity of ocean trenches were both shallow and deep, but that the deeper ones were situated progressively farther inland from the trenches (Figure 10.10).

Figure 10.10 Cross-section through the Aleutian subduction zone with a depiction of the increasing depth of earthquakes “inshore” from the trench. [SE]

In the 1950s, scientists from the Scripps Oceanographic Institute in California persuaded the U.S. Coast Guard to include magnetometer readings on one of their expeditions to study ocean floor topography. The first comprehensive magnetic data set was compiled in 1958 for an area off the coast of B.C. and Washington State. This survey revealed a bewildering pattern of low and high magnetic intensity in sea-floor rocks (Figure 10.11). When the data were first plotted on a map in 1961, nobody understood them — not even the scientists who collected them. Although the patterns made even less sense than the stripes on a zebra, many thousands of kilometres of magnetic surveys were conducted over the next several years.

Figure 10.11 Pattern of sea-floor magnetism off of the west coast of British Columbia and Washington [SE after http://geomaps.wr.usgs.gov/parks/noca/nocageol4c.html, adapted from: Raff, A and Mason, R, 1961, Magnetic survey off the west coast of North America, 40˚ N to 52˚ N latitude, Geol. Soc. America Bulletin, V. 72, p. 267-270.]

The wealth of new data from the oceans began to significantly influence geological thinking in the 1960s. In 1960, Harold Hess, a widely respected geologist from Princeton University, advanced a theory with many of the elements that we now accept as plate tectonics. He maintained some uncertainty about his proposal however, and in order to deflect criticism from mainstream geologists, he labelled it geopoetry. In fact, until 1962, Hess didn’t even put his ideas in writing — except internally to the U.S. Navy (which funded his research) — but presented them mostly in lectures and seminars. Hess proposed that new sea floor was generated from mantle material at the ocean ridges, and that old sea floor was dragged down at the ocean trenches and re-incorporated into the mantle. He suggested that the process was driven by mantle convection currents, rising at the ridges and descending at the trenches (Figure 10.12). He also suggested that the less-dense continental crust did not descend with oceanic crust into trenches, but that colliding land masses were thrust up to form mountains. Hess’s theory formed the basis for our ideas on sea-floor spreading and continental drift, but it did not deal with the concept that the crust is made up of specific plates. Although the Hess model was not roundly criticized, it was not widely accepted (especially in the U.S.), partly because it was not well supported by hard evidence.

Figure 10.12 A representation of Harold Hess’s model for sea-floor spreading and subduction [SE]

Collection of magnetic data from the oceans continued in the early 1960s, but still nobody could explain the origin of the zebra-like patterns. Most assumed that they were related to variations in the composition of the rocks — such as variations in the amount of magnetite — as this is a common explanation for magnetic variations in rocks of the continental crust. The first real understanding of the significance of the striped anomalies was the interpretation by Fred Vine, a Cambridge graduate student. Vine was examining magnetic data from the Indian Ocean and, like others before, he noted the symmetry of the magnetic patterns with respect to the oceanic ridge.

At the same time, other researchers, led by groups in California and New Zealand, were studying the phenomenon of reversals in Earth’s magnetic field. They were trying to determine when such reversals had taken place over the past several million years by analyzing the magnetic characteristics of hundreds of samples from basaltic flows. As discussed in Chapter 9, it is evident that Earth’s magnetic field becomes weakened periodically and then virtually non-existent, before becoming re-established with the reverse polarity. During periods of reversed polarity, a compass would point south instead of north.

The time scale of magnetic reversals is irregular. For example, the present “normal” event, known as the Bruhnes magnetic chron, has persisted for about 780,000 years. This was preceded by a 190,000-year reversed event; a 50,000-year normal event known as Jaramillo; and then a 700,000-year reversed event (see Figure 9.15).

In a paper published in September 1963, Vine and his PhD supervisor Drummond Matthews proposed that the patterns associated with ridges were related to the magnetic reversals, and that oceanic crust created from cooling basalt during a normal event would have polarity aligned with the present magnetic field, and thus would produce a positive anomaly (a black stripe on the sea-floor magnetic map), whereas oceanic crust created during a reversed event would have polarity opposite to the present field and thus would produce a negative magnetic anomaly (a white stripe). The same idea had been put forward a few months earlier by Lawrence Morley, of the Geological Survey of Canada; however, his papers submitted earlier in 1963 to Nature and The Journal of Geophysical Research were rejected. Many people refer to the idea as the Vine-Matthews-Morley (VMM) hypothesis.

Vine, Matthews, and Morley were the first to show this type of correspondence between the relative widths of the stripes and the periods of the magnetic reversals. The VMM hypothesis was confirmed within a few years when magnetic data were compiled from spreading ridges around the world. It was shown that the same general magnetic patterns were present straddling each ridge, although the widths of the anomalies varied according to the spreading rates characteristic of the different ridges. It was also shown that the patterns corresponded with the chronology of Earth’s magnetic field reversals. This global consistency provided strong support for the VMM hypothesis and led to rejection of the other explanations for the magnetic anomalies.

In 1963, J. Tuzo Wilson of the University of Toronto proposed the idea of a mantle plume or hot spot — a place where hot mantle material rises in a stationary and semi-permanent plume, and affects the overlying crust. He based this hypothesis partly on the distribution of the Hawaiian and Emperor Seamount island chains in the Pacific Ocean (Figure 10.13). The volcanic rock making up these islands gets progressively younger toward the southeast, culminating with the island of Hawaii itself, which consists of rock that is almost all younger than 1 Ma. Wilson suggested that a stationary plume of hot upwelling mantle material is the source of the Hawaiian volcanism, and that the ocean crust of the Pacific Plate is moving toward the northwest over this hot spot. Near the Midway Islands, the chain takes a pronounced change in direction, from northwest-southeast for the Hawaiian Islands and to nearly north-south for the Emperor Seamounts. This change is widely ascribed to a change in direction of the Pacific Plate moving over the stationary mantle plume, but a more plausible explanation is that the Hawaiian mantle plume has not actually been stationary throughout its history, and in fact moved at least 2,000 km south over the period between 81 and 45 Ma.J. A. Tarduno et al., 2003, The Emperor Seamounts: Southward Motion of the Hawaiian Hotspot Plume in Earth’s Mantle, Science 301 (5636): 1064–1069.

Figure 10.13 The ages of the Hawaiian Islands and the Emperor Seamounts in relation to the location of the Hawaiian mantle plume [SE. Basemap from the National Geophysical Data Centre, accessed at: http://en.wikipedia.org/wiki/Hotspot_(geology)#/ media/File:Hawaii_hotspot.jpg.]

There is evidence of many such mantle plumes around the world (Figure 10.14). Most are within the ocean basins — including places like Hawaii, Iceland, and the Galapagos Islands — but some are under continents. One example is the Yellowstone hot spot in the west-central United States, and another is the one responsible for the Anahim Volcanic Belt in central British Columbia. It is evident that mantle plumes are very long-lived phenomena, lasting for at least tens of millions of years, possibly for hundreds of millions of years in some cases.

Figure 10.14 Mantle plume locations [Ingo Wölbern at: http://commons.wikimedia.org /wiki/File:Hotspots.jpg] Selected Mantle plumes: 1: Azores, 3: Bowie, 5: Cobb, 8: Eifel, 10: Galapagos, 12: Hawaii, 14: Iceland, 17: Cameroon, 18: Canary, 19: Cape Verde, 35: Samoa, 38: Tahiti, 42: Tristan, 44: Yellowstone, 45: Anahim

Although oceanic spreading ridges appear to be curved features on Earth’s surface, in fact the ridges are composed of a series of straight-line segments, offset at intervals by faults perpendicular to the ridge (Figure 10.15). In a paper published in 1965, Tuzo Wilson termed these features transform faults. He described the nature of the motion along them, and showed why there are earthquakes only on the section of a transform fault between two adjacent ridge segments. The San Andreas Fault in California is a very long transform fault that links the southern end of the Juan de Fuca spreading ridge to the East Pacific Rise spreading ridges situated in the Gulf of California (see Figure 10.23). The Queen Charlotte Fault, which extends north from the northern end of the Juan de Fuca spreading ridge (near the northern end of Vancouver Island) toward Alaska, is also a transform fault.

Figure 10.15 A part of the mid-Atlantic ridge near the equator. The double white lines are spreading ridges. The solid white lines are fracture zones. As shown by the yellow arrows, the relative motion of the plates on either side of the fracture zones can be similar (arrows pointing the same direction) or opposite (arrows pointing opposite directions). Transform faults (red lines) are in between the ridge segments, where the yellow arrows point in opposite directions. [SE]

In the same 1965 paper, Wilson introduced the idea that the crust can be divided into a series of rigid plates, and thus he is responsible for the term plate tectonics.

Continental drift and sea-floor spreading became widely accepted around 1965 as more and more geologists started thinking in these terms. By the end of 1967, Earth’s surface had been mapped into a series of plates (Figure 10.16). The major plates are Eurasia, Pacific, India, Australia, North America, South America, Africa, and Antarctic. There are also numerous small plates (e.g., Juan de Fuca, Nazca, Scotia, Philippine, Caribbean), and many very small plates or sub-plates. For example the Juan de Fuca Plate is actually three separate plates (Gorda, Juan de Fuca, and Explorer) that all move in the same general direction but at slightly different rates.

Figure 10.16 A map showing 15 of the Earth’s tectonic plates and the approximate rates and directions of plate motions. [SE after USGS, http://en.wikipedia.org/wiki/Plate_tectonics#/media/File:Plates_tect2_en.svg]

Rates of motions of the major plates range from less than 1 cm/y to over 10 cm/y. The Pacific Plate is the fastest at over 10 cm/y in some areas, followed by the Australian and Nazca Plates. The North American Plate is one of the slowest, averaging around 1 cm/y in the south up to almost 4 cm/y in the north.

Plates move as rigid bodies, so it may seem surprising that the North American Plate can be moving at different rates in different places. The explanation is that plates move in a rotational manner. The North American Plate, for example, rotates counter-clockwise; the Eurasian Plate rotates clockwise.

Boundaries between the plates are of three types: divergent (i.e., moving apart), convergent (i.e., moving together), and transform (moving side by side). Before we talk about processes at plate boundaries, it’s important to point out that there are never gaps between plates. The plates are made up of crust and the lithospheric part of the mantle (Figure 10.17), and even though they are moving all the time, and in different directions, there is never a significant amount of space between them. Plates are thought to move along the lithosphere-asthenosphere boundary, as the asthenosphere is the zone of partial melting. It is assumed that the relative lack of strength of the partial melting zone facilitates the sliding of the lithospheric plates.

Figure 10.17 The crust and upper mantle. Tectonic plates consist of lithosphere, which includes the crust and the lithospheric (rigid) part of the mantle. [SE]

At spreading centres, the lithospheric mantle may be very thin because the upward convective motion of hot mantle material generates temperatures that are too high for the existence of a significant thickness of rigid lithosphere (Figure 10.12). The fact that the plates include both crustal material and lithospheric mantle material makes it possible for a single plate to be made up of both oceanic and continental crust. For example, the North American Plate includes most of North America, plus half of the northern Atlantic Ocean. Similarly the South American Plate extends across the western part of the southern Atlantic Ocean, while the European and African plates each include part of the eastern Atlantic Ocean. The Pacific Plate is almost entirely oceanic, but it does include the part of California west of the San Andreas Fault.

Divergent Boundaries

Divergent boundaries are spreading boundaries, where new oceanic crust is created from magma derived from partial melting of the mantle caused by decompression as hot mantle rock from depth is moved toward the surface (Figure 10.18). The triangular zone of partial melting near the ridge crest is approximately 60 km thick and the proportion of magma is about 10% of the rock volume, thus producing crust that is about 6 km thick. Most divergent boundaries are located at the oceanic ridges (although some are on land), and the crustal material created at a spreading boundary is always oceanic in character; in other words, it is mafic igneous rock (e.g., basalt or gabbro, rich in ferromagnesian minerals). Spreading rates vary considerably, from 1 cm/y to 3 cm/y in the Atlantic, to between 6 cm/y and 10 cm/y in the Pacific. Some of the processes taking place in this setting include:

• Magma from the mantle pushing up to fill the voids left by divergence of the two plates
• Pillow lavas forming where magma is pushed out into seawater (Figure 10.19)
• Vertical sheeted dykes intruding into cracks resulting from the spreading
• Magma cooling more slowly in the lower part of the new crust and forming gabbro bodies

Figure 10.18 The general processes that take place at a divergent boundary. The area within the dashed white rectangle is shown in Figure 10.19. [SE]

Figure 10.19 Depiction of the processes and materials formed at a divergent boundary [SE after Keary and Vine, 1996, Global Tectonics (2ed), Blackwell Science Ltd., Oxford]

Spreading is hypothesized to start within a continental area with up-warping or doming related to an underlying mantle plume or series of mantle plumes. The buoyancy of the mantle plume material creates a dome within the crust, causing it to fracture in a radial pattern, with three arms spaced at approximately 120° (Figure 10.20). When a series of mantle plumes exists beneath a large continent, the resulting rifts may align and lead to the formation of a rift valley (such as the present-day Great Rift Valley in eastern Africa). It is suggested that this type of valley eventually develops into a linear sea (such as the present-day Red Sea), and finally into an ocean (such as the Atlantic). It is likely that as many as 20 mantle plumes, many of which still exist, were responsible for the initiation of the rifting of Pangea along what is now the mid-Atlantic ridge (see Figure 10.14).

Figure 10.20 Depiction of the process of dome and three-part rift formation (left) and of continental rifting between the African and South American parts of Pangea at around 200 Ma (right) [SE]

Convergent Boundaries

Convergent boundaries, where two plates are moving toward each other, are of three types, depending on the type of crust present on either side of the boundary — oceanic or continental. The types are ocean-ocean, ocean-continent, and continent-continent.

At an ocean-ocean convergent boundary, one of the plates (oceanic crust and lithospheric mantle) is pushed, or subducted, under the other. Often it is the older and colder plate that is denser and subducts beneath the younger and hotter plate. There is commonly an ocean trench along the boundary. The subducted lithosphere descends into the hot mantle at a relatively shallow angle close to the subduction zone, but at steeper angles farther down (up to about 45°). As discussed in the context of subduction-related volcanism in Chapter 4, the significant volume of water within the subducting material is released as the subducting crust is heated. This water is mostly derived from alteration of pyroxene and olivine to serpentine near the spreading ridge shortly after the rock’s formation. It mixes with the overlying mantle, and the addition of water to the hot mantle lowers the crust’s melting point and leads to the formation of magma (flux melting). The magma, which is lighter than the surrounding mantle material, rises through the mantle and the overlying oceanic crust to the ocean floor where it creates a chain of volcanic islands known as an island arc. A mature island arc develops into a chain of relatively large islands (such as Japan or Indonesia) as more and more volcanic material is extruded and sedimentary rocks accumulate around the islands.

As described above in the context of Benioff zones (Figure 10.10), earthquakes take place close to the boundary between the subducting crust and the overriding crust. The largest earthquakes occur near the surface where the subducting plate is still cold and strong.

Figure 10.21 Configuration and processes of an ocean-ocean convergent boundary [SE]

Examples of ocean-ocean convergent zones are subduction of the Pacific Plate south of Alaska (Aleutian Islands) and west of the Philippines, subduction of the India Plate south of Indonesia, and subduction of the Atlantic Plate beneath the Caribbean Plate (Figure 10.21).

At an ocean-continent convergent boundary, the oceanic plate is pushed under the continental plate in the same manner as at an ocean-ocean boundary. Sediment that has accumulated on the continental slope is thrust up into an accretionary wedge, and compression leads to thrusting within the continental plate (Figure 10.22). The mafic magma produced adjacent to the subduction zone rises to the base of the continental crust and leads to partial melting of the crustal rock. The resulting magma ascends through the crust, producing a mountain chain with many volcanoes.

Figure 10.22 Configuration and processes of an ocean-continent convergent boundary [SE]

Examples of ocean-continent convergent boundaries are subduction of the Nazca Plate under South America (which has created the Andes Range) and subduction of the Juan de Fuca Plate under North America (creating the mountains Garibaldi, Baker, St. Helens, Rainier, Hood, and Shasta, collectively known as the Cascade Range).

A continent-continent collision occurs when a continent or large island that has been moved along with subducting oceanic crust collides with another continent (Figure 10.23). The colliding continental material will not be subducted because it is too light (i.e., because it is composed largely of light continental rocks [SIAL]), but the root of the oceanic plate will eventually break off and sink into the mantle. There is tremendous deformation of the pre-existing continental rocks, and creation of mountains from that rock, from any sediments that had accumulated along the shores (i.e., within geosynclines) of both continental masses, and commonly also from some ocean crust and upper mantle material.

Figure 10.23 Configuration and processes of a continent-continent convergent boundary [SE]

Examples of continent-continent convergent boundaries are the collision of the India Plate with the Eurasian Plate, creating the Himalaya Mountains, and the collision of the African Plate with the Eurasian Plate, creating the series of ranges extending from the Alps in Europe to the Zagros Mountains in Iran. The Rocky Mountains in B.C. and Alberta are also a result of continent-continent collisions.

Transform boundaries exist where one plate slides past another without production or destruction of crustal material. As explained above, most transform faults connect segments of mid-ocean ridges and are thus ocean-ocean plate boundaries (Figure 10.15). Some transform faults connect continental parts of plates. An example is the San Andreas Fault, which connects the southern end of the Juan de Fuca Ridge with the northern end of the East Pacific Rise (ridge) in the Gulf of California (Figures 10.24 an 10.25). The part of California west of the San Andreas Fault and all of Baja California are on the Pacific Plate. Transform faults do not just connect divergent boundaries. For example, the Queen Charlotte Fault connects the north end of the Juan de Fuca Ridge, starting at the north end of Vancouver Island, to the Aleutian subduction zone.

Figure 10.24 The San Andreas Fault extends from the north end of the East Pacific Rise in the Gulf of California to the southern end of the Juan de Fuca Ridge. All of the red lines on this map are transform faults. [SE]

Figure 10.25 The San Andreas Fault at Parkfield in central California. The person with the orange shirt is standing on the Pacific Plate and the person at the far side of the bridge is on the North American Plate. The bridge is designed to slide on its foundation. [SE]

Exercises

Exercise 10.4 A Different Type of Transform Fault

This map shows the Juan de Fuca (JDF) and Explorer Plates off the coast of Vancouver Island. We know that the JDF Plate is moving toward the North American Plate at around 4 cm/y to 5 cm/y. We think that the Explorer Plate is also moving east, but we don’t know the rate, and there is evidence that it is slower than the JDF Plate.

The boundary between the two plates is the Nootka Fault, which is the location of frequent small-to-medium earthquakes (up to magnitude ~5), as depicted by the red stars. Explain why the Nootka Fault is a transform fault, and show the relative sense of motion along the fault with two small arrows.

As originally described by Wegener in 1915, the present continents were once all part of a supercontinent, which he termed Pangea (all land). More recent studies of continental matchups and the magnetic ages of ocean-floor rocks have enabled us to reconstruct the history of the break-up of Pangea.

Pangea began to rift apart along a line between Africa and Asia and between North America and South America at around 200 Ma. During the same period, the Atlantic Ocean began to open up between northern Africa and North America, and India broke away from Antarctica. Between 200 and 150 Ma, rifting started between South America and Africa and between North America and Europe, and India moved north toward Asia. By 80 Ma, Africa had separated from South America, most of Europe had separated from North America, and India had separated from Antarctica. By 50 Ma, Australia had separated from Antarctic, and shortly after that, India collided with Asia. To see the timing of these processes for yourself go to: http://barabus.tru.ca/geol1031/plates.html.

Within the past few million years, rifting has taken place in the Gulf of Aden and the Red Sea, and also within the Gulf of California. Incipient rifting has begun along the Great Rift Valley of eastern Africa, extending from Ethiopia and Djibouti on the Gulf of Aden (Red Sea) all the way south to Malawi.

Over the next 50 million years, it is likely that there will be full development of the east African rift and creation of new ocean floor. Eventually Africa will split apart. There will also be continued northerly movement of Australia and Indonesia. The western part of California (including Los Angeles and part of San Francisco) will split away from the rest of North America, and eventually sail right by the west coast of Vancouver Island, en route to Alaska. Because the oceanic crust formed by spreading on the mid-Atlantic ridge is not currently being subducted (except in the Caribbean), the Atlantic Ocean is slowly getting bigger, and the Pacific Ocean is getting smaller. If this continues without changing for another couple hundred million years, we will be back to where we started, with one supercontinent.

Pangea, which existed from about 350 to 200 Ma, was not the first supercontinent. It was preceded by Pannotia (600 to 540 Ma), by Rodinia (1,100 to 750 Ma), and by others before that.

In 1966, Tuzo Wilson proposed that there has been a continuous series of cycles of continental rifting and collision; that is, break-up of supercontinents, drifting, collision, and formation of other supercontinents. At present, North and South America, Europe, and Africa are moving with their respective portions of the Atlantic Ocean. The eastern margins of North and South America and the western margins of Europe and Africa are called passive margins because there is no subduction taking place along them.

This situation may not continue for too much longer, however. As the Atlantic Ocean floor gets weighed down around its margins by great thickness of continental sediments (i.e., geosynclines), it will be pushed farther and farther into the mantle, and eventually the oceanic lithosphere may break away from the continental lithosphere (Figure 10.26). A subduction zone will develop, and the oceanic plate will begin to descend under the continent. Once this happens, the continents will no longer continue to move apart because the spreading at the mid-Atlantic ridge will be taken up by subduction. If spreading along the mid-Atlantic ridge continues to be slower than spreading within the Pacific Ocean, the Atlantic Ocean will start to close up, and eventually (in a 100 million years or more) North and South America will collide with Europe and Africa.

Figure 10.26 Development of a subduction zone at a passive margin. Times A, B, and C are separated by tens of millions of years. Once the oceanic crust breaks off and starts to subduct the continental crust (North America in this case) will no longer be pushed to the west and will likely start to move east because the rate of spreading in the Pacific basin is faster than that in the Atlantic basin. [SE]

There is strong evidence around the margins of the Atlantic Ocean that this process has taken place before. The roots of ancient mountain belts, which are present along the eastern margin of North America, the western margin of Europe, and the northwestern margin of Africa, show that these land masses once collided with each other to form a mountain chain, possibly as big as the Himalayas. The apparent line of collision runs between Norway and Sweden, between Scotland and England, through Ireland, through Newfoundland, and the Maritimes, through the northeastern and eastern states, and across the northern end of Florida. When rifting of Pangea started at approximately 200 Ma, the fissuring was along a different line from the line of the earlier collision. This is why some of the mountain chains formed during the earlier collision can be traced from Europe to North America and from Europe to Africa.

That the Atlantic Ocean rift may have occurred in approximately the same place during two separate events several hundred million years apart is probably no coincidence. The series of hot spots that has been identified in the Atlantic Ocean may also have existed for several hundred million years, and thus may have contributed to rifting in roughly the same place on at least two separate occasions (Figure 10.27).

Figure 10.27 A scenario for the Wilson cycle. The cycle starts with continental rifting above a series of mantle plumes (A). The continents separate (B), and then re-converge some time later, forming a fold-belt mountain chain. Eventually rifting is repeated, possibly because of the same set of mantle plumes (D), but this time the rift is in a different place. [SE]

Exercises

Exercise 10.5 Getting to Know the Plates and Their Boundaries

This map shows the boundaries between the major plates. Without referring to the plate map in Figure 10.16, or any other resources, write in the names of as many of the plates as you can. Start with the major plates, and then work on the smaller ones. Don’t worry if you can’t name them all.

Once you’ve named most of the plates, draw arrows to show the general plate motions. Finally, using a highlighter or coloured pencil, label as many of the boundaries as you can as divergent, convergent, or transform. [map by SE]

It has been often repeated in this text and elsewhere that convection of the mantle is critical to plate tectonics, and while this is almost certainly so, there is still some debate about the actual forces that make the plates move. One side in the argument holds that the plates are only moved by the traction caused by mantle convection. The other side holds that traction plays only a minor role and that two other forces, ridge-push and slab-pull, are more important (Figure 10.28). Some argue that the real answer lies somewhere in between.

Figure 10.28 Models for plate motion mechanisms [SE]

Kearey and Vine (1996)Kearey and Vine , 1996, Global Tectonics (2ed), Blackwell Science Ltd., Oxford have listed some compelling arguments in favour of the ridge-push/slab-pull model, as follows: (a) plates that are attached to subducting slabs (e.g., Pacific, Australian, and Nazca Plates) move the fastest, and plates that are not (e.g., North American, South American, Eurasian, and African Plates) move significantly slower; (b) in order for the traction model to apply, the mantle would have to be moving about five times faster than the plates are moving (because the coupling between the partially liquid asthenosphere and the plates is not strong), and such high rates of convection are not supported by geophysical models; and (c) although large plates have potential for much higher convection traction, plate velocity is not related to plate area.

In the ridge-push/slab-pull model, which is the one that has been adopted by most geologists working on plate-tectonic problems, the lithosphere is the upper surface of the convection cells, as is illustrated in Figure 10.29.

Figure 10.29 The ridge-push/slab-pull model for plate motion, in which the lithosphere is the upper surface of the convective systems. [SE]

Although ridge-push/slab-pull is the favoured mechanism for plate motion, it’s important not to underestimate the role of mantle convection. Without convection, there would be no ridges to push from because upward convection brings hot buoyant rock to surface. Furthermore, many plates, including our own North American Plate, move along nicely — albeit slowly — without any slab-pull happening.

The topics covered in this chapter can be summarized as follows:

Introduction

Earthquakes scare people … a lot! That’s not surprising because time and time again earthquakes have caused massive damage and many, many casualties. Anyone who has lived through a damaging earthquake cannot forget the experience (Figure 11.1). But geoscientists and engineers are getting better at understanding earthquakes, minimizing the amount of damage they cause, and reducing the number of people affected. People living in western Canada don’t need to be frightened by earthquakes, but they do need to be prepared.

Figure 11.1 A schoolroom in Courtenay damaged by the 1946 Vancouver Island earthquake. If the earthquake had not happened on a Sunday, the casualties would have been much greater. [from Earthquakes Canada, http://www.earthquakes canada. nrcan.gc.ca/historic-historique/ events/ images/19460623_1946. school.inside.jpg ]

An earthquake is the shaking caused by the rupture (breaking) and subsequent displacement of rocks (one body of rock moving with respect to another) beneath Earth’s surface.

A body of rock that is under stress becomes deformed. When the rock can no longer withstand the deformation, it breaks and the two sides slide past each other. Most earthquakes take place near plate boundaries, but not necessarily right on a boundary, and not necessarily even on a pre-existing fault.

The engineering principle of elastic deformation, which can be used to understand earthquakes, is illustrated in Figure 11.2. The stress applied to a rock — typically because of ongoing plate movement — results in strain or deformation of the rock (Figure 11.2b). Because most rock is strong (unlike loose sand, for example), it can withstand a significant amount of deformation without breaking. But every rock has a deformation limit and will rupture (break) once that limit is reached. At that point, in the case of rocks within the crust, the rock breaks and there is displacement along the rupture surface (Figure 11.2c). The magnitude of the earthquake depends on the extent of the area that breaks (the area of the rupture surface) and the average amount of displacement (sliding).

Figure 11.2 Depiction of the concept of elastic deformation and rupture, looking down. [SE]

The concept of a rupture surface, which is critical to understanding earthquakes, is illustrated in Figure 11.3. An earthquake does not happen at a point, it happens over an area within a plane, although not necessarily a flat plane. Within the area of the rupture surface, the amount of displacement is variable (Figure 11.3), and, by definition, it decreases to zero at the edges of the rupture surface because the rock beyond that point isn’t displaced at all. The extent of a rupture surface and the amount of displacement will depend on a number of factors, including the type and strength of the rock, and the degree to which it was stressed beforehand.

Figure 11.3 A rupture surface (dark pink), on a steeply dipping fault plane (light pink). The diagram represents a part of the crust that may be up to tens or hundreds of kilometres long. The rupture surface is the part of the fault plane along which displacement occurred. In this example, the near side of the fault is moving to the left, and the lengths of the arrows within the rupture surface represent relative amounts of displacement. [SE]

Earthquake rupture doesn’t happen all at once; it starts at a single point and spreads rapidly from there. Depending on the extent of the rupture surface, the propagation of failures out from the point of initiation is typically completed within seconds to several tens of seconds (Figure 11.4). The initiation point isn’t necessarily in the centre of the rupture surface; it may be close to one end, near the top, or near the bottom.

Figure 11.4 Propagation of failure on a rupture surface. In this case, the failure starts at the dark blue heavy arrow and propagates outward, reaching the left side first (green arrows) and the right side last (yellow arrows). [SE]

Figure 11.5 shows the distribution of immediate aftershocks associated with the 1989 Loma Prieta earthquake. Panel (b) is a section along the San Andreas Fault; this view is equivalent to what is shown in Figures 11.3 and 11.4. The area of red dots is the rupture surface; each red dot is a specific aftershock that was recorded on a seismometer. The hexagon labelled “main earthquake” represents the first or main shock. When that happened, the rock at that location broke and was displaced. That released the stress on that particular part of the fault, but it resulted in an increase of the stress on other nearby parts of the fault, and contributed to a cascade of smaller ruptures (aftershocks), in this case, over an area about 60 km long and 15 km wide.

Figure 11.5 Distribution of the aftershocks of the 1989 M 6.9 Loma Prieta earthquake (a: plan view, b: section along the fault, c: section across the fault.) [from Open University under CC Sharealike, http://www.open.edu/openlearnworks/mod/page/view.php?id=45426]

So, what exactly is an aftershock then? An aftershock is an earthquake just like any other, but it is one that can be shown to have been triggered by stress transfer from a preceding earthquake. Within a few tens of seconds of the main Loma Prieta earthquake, there were hundreds of smaller aftershocks; their distribution defines the area of the rupture surface.

Aftershocks can be of any magnitude. Most are smaller than the earthquake that triggered them, but they can be bigger. The aftershocks shown in Figure 11.5 all happened within seconds or minutes of the main shock, but aftershocks can be delayed for hours, days, weeks, or even years. As already noted, aftershocks are related to stress transfer. For example, the main shock of the Loma Prieta earthquake triggered aftershocks in the immediate area, which triggered more in the surrounding area, eventually extending for 30 km along the fault in each direction and for 15 km toward the surface. But the earthquake as a whole also changed the stress on adjacent parts of the San Andreas Fault. This effect, which has been modelled for numerous earthquakes and active faults around the world, is depicted in Figure 11.6. Stress was reduced in the area of the rupture (blue), but was increased at either end of the rupture surface (red and yellow).

Figure 11.6 Depiction of stress changes related to an earthquake. Stress decreases in the area of the rupture surface, but increases on adjacent parts of the fault. [by SE based on data from 2010 Laguna Salada earthquake by Stein and Toda at: http://supersites.earthobservations.org/Baja_stress.png]

Stress transfer isn’t necessarily restricted to the fault along which an earthquake happened. It will affect the rocks in general around the site of the earthquake and may lead to increased stress on other faults in the region. The effects of stress transfer don’t necessarily show up right away. Segments of faults are typically in some state of stress, and the transfer of stress from another area is only rarely enough to push a fault segment beyond its limits to the point of rupture. The stress that is added by stress transfer accumulates along with the ongoing buildup of stress from plate motion and eventually leads to another earthquake.

Episodic tremor and slip

Episodic tremor and slip (ETS) is periodic slow sliding along part of a subduction boundary. It does not produce recognizable earthquakes, but does produce seismic tremor (rapid seismic vibrations on a seismometer). It was first discovered on the Vancouver Island part of the Cascadia subduction zone by Geological Survey of Canada geologists Herb Dragert and Gary Rogers.*

The boundary between the subducting Juan de Fuca Plate and the North America Plate can be divided into three segments, as shown below. The cold upper part of the boundary is locked. The plates are stuck and don’t move, except with very large earthquakes that happen approximately every 500 years (the last one was M8.5+ in January 26, 1700). The warm lower part of the boundary is sliding continuously because the warm rock is weaker. The central part of the boundary isn’t cold enough to be stuck, but isn’t warm enough to slide continuously. Instead it slips episodically, approximately every 14 months for about 2 weeks, moving a few centimetres each time.

You might be inclined to think that it’s a good thing that there is periodic slip on this part of the plate because it releases some of the tension and reduces the risk of a large earthquake. In fact, the opposite is likely the case. The movement along the ETS part of the plate boundary acts like a medium-sized earthquake and leads to stress transfer to the adjacent locked part of the plate. Approximately every 14 months, during the two-week ETS period, there is a transfer of stress to the shallow locked part of the Cascadia subduction zone, and therefore an increased chance of a large earthquake.

Since 2003, ETS processes have also been observed on subduction zones in Mexico and Japan. [SE drawing]

*Rogers, G. and Dragert, H., 2003, Episodic tremor and slip on the Cascadia subduction zone: the chatter of silent slip, Science, V. 300, p. 1942-1943.

The distribution of earthquakes across the globe is shown in Figure 11.7. It is relatively easy to see the relationships between earthquakes and the plate boundaries. Along divergent boundaries like the mid-Atlantic ridge and the East Pacific Rise, earthquakes are common, but restricted to a narrow zone close to the ridge, and consistently at less than 30 km depth. Shallow earthquakes are also common along transform faults, such as the San Andreas Fault. Along subduction zones, as we saw in Chapter 10, earthquakes are very abundant, and they are increasingly deep on the landward side of the subduction zone.

Figure 11.7 General distribution of global earthquakes of magnitude 4 and greater from 2004 to 2011, colour coded by depth (red: 0-33 km, orange 33-70 km, green: 70-300 km, blue: 300-700 km) [from Dale Sawyer, Rice University, http://plateboundary.rice.edu ,used with permission]

Earthquakes are also relatively common at a few intraplate locations. Some are related to the buildup of stress due to continental rifting or the transfer of stress from other regions, and some are not well understood. Examples of intraplate earthquake regions include the Great Rift Valley area of Africa, the Tibet region of China, and the Lake Baikal area of Russia.

Earthquakes at Divergent and Transform Boundaries

Figure 11.8 provides a closer look at magnitude (M) 4 and larger earthquakes in an area of divergent boundaries in the mid-Atlantic region near the equator. Here, as we saw in Chapter 10, the segments of the mid-Atlantic ridge are offset by some long transform faults. Most of the earthquakes are located along the transform faults, rather than along the spreading segments, although there are clusters of earthquakes at some of the ridge-transform boundaries. Some earthquakes do occur on spreading ridges, but they tend to be small and infrequent because of the relatively high rock temperatures in the areas where spreading is taking place.

Figure 11.8 Distribution of earthquakes of M4 and greater in the area of the mid-Atlantic ridge near the equator from 1990 to 1996. All are at a depth of 0 to 33 km [SE after Dale Sawyer, Rice University, http://plateboundary.rice.edu]

Earthquakes at Convergent Boundaries

The distribution and depths of earthquakes in the Caribbean and Central America area are shown in Figure 11.9. In this region, the Cocos Plate is subducting beneath the North America and Caribbean Plates (ocean-continent convergence), and the South and North America Plates are subducting beneath the Caribbean Plate (ocean-ocean convergence). In both cases, the earthquakes get deeper with distance from the trench. In Figure 11.9, the South America Plate is shown as being subducted beneath the Caribbean Plate in the area north of Colombia, but since there is almost no earthquake activity along this zone, it is questionable whether subduction is actually taking place.

Figure 11.9 Distribution of earthquakes of M4 and greater in the Central America region from 1990 to 1996 (red: 0-33 km, orange: 33-70 km, green: 70-300 km, blue: 300-700 km) (Spreading ridges are heavy lines, subduction zones are toothed lines, and transform faults are light lines.) [SE after Dale Sawyer, Rice University, http://plateboundary.rice.edu]

There are also various divergent and transform boundaries in the area shown in Figure 11.9, and as we’ve seen in the mid-Atlantic area, most of these earthquakes occur along the transform faults.

The distribution of earthquakes with depth in the Kuril Islands of Russia in the northwest Pacific is shown in Figure 11.10. This is an ocean-ocean convergent boundary. The small red and yellow dots show background seismicity over a number of years, while the larger white dots are individual shocks associated with a M6.9 earthquake in April 2009. The relatively large earthquake took place on the upper part of the plate boundary between 60 km and 140 km inland from the trench. As we saw for the Cascadia subduction zone, this is where large subduction earthquakes are expected to occur.

In fact, all of the very large earthquakes — M9 or higher — take place at subduction boundaries because there is the potential for a greater width of rupture zone on a gently dipping boundary than on a steep transform boundary. The largest earthquakes on transform boundaries are in the order of M8.

Figure 11.10 Distribution of earthquakes in the area of the Kuril Islands, Russia (just north of Japan) (White dots represent the April 2009 M6.9 earthquake. Red and yellow dots are from background seismicity over several years prior to 2009.) [SE after Gavin Hayes, from data at http://earthquake.usgs.gov/earthquakes/eqarchives/subduction_zone/us2009fdak/szgc/ku6_trench.pdf]

The background seismicity at this convergent boundary, and on other similar ones, is predominantly near the upper side of the subducting plate. The frequency of earthquakes is greatest near the surface and especially around the area where large subduction quakes happen, but it extends to at least 400 km depth. There is also significant seismic activity in the overriding North America Plate, again most commonly near the region of large quakes, but also extending for a few hundred kilometres away from the plate boundary.

The distribution of earthquakes in the area of the India-Eurasia plate boundary is shown in Figure 11.11. This is a continent-continent convergent boundary, and it is generally assumed that although the India Plate continues to move north toward the Asia Plate, there is no actual subduction taking place. There are transform faults on either side of the India Plate in this area.

Figure 11.11 Distribution of earthquakes in the area where the India Plate is converging with the Asia Plate (data from 1990 to 1996, red: 0-33 km, orange: 33-70 km, green: 70-300 km). (Spreading ridges are heavy lines, subduction zones are toothed lines, and transform faults are light lines. The double line along the northern edge of the India Plate indicates convergence, but not subduction. Plate motions are shown in mm/y.) [SE after Dale Sawyer, Rice University, http://plateboundary.rice.edu]

The entire northern India and southern Asia region is very seismically active. Earthquakes are common in northern India, Nepal, Bhutan, Bangladesh and adjacent parts of China, and throughout Pakistan and Afghanistan. Many of the earthquakes are related to the transform faults on either side of the India Plate, and most of the others are related to the significant tectonic squeezing caused by the continued convergence of the India and Asia Plates. That squeezing has caused the Asia Plate to be thrust over top of the India Plate, building the Himalayas and the Tibet Plateau to enormous heights. Most of the earthquakes of Figure 11.11 are related to the thrust faults shown in Figure 11.12 (and to hundreds of other similar ones that cannot be shown at this scale). The southernmost thrust fault in Figure 11.12 is equivalent to the Main Boundary Fault in Figure 11.11.

Figure 11.12 Schematic diagram of the India-Asia convergent boundary, showing examples of the types of faults along which earthquakes are focussed. The devastating Nepal earthquake of May 2015 took place along one of these thrust faults. [SE after D. Vouichard, from a United Nations University document at: http://archive.unu.edu/unupress/unupbooks/80a02e/80A02E05.htm]

There is a very significant concentration of both shallow and deep (greater than 70 km) earthquakes in the northwestern part of Figure 11.11. This is northern Afghanistan, and at depths of more than 70 km, many of these earthquakes are within the mantle as opposed to the crust. It is interpreted that these deep earthquakes are caused by northwestward subduction of part of the India Plate beneath the Asia Plate in this area.

There are two main ways to measure earthquakes. The first of these is an estimate of the energy released, and the value is referred to as magnitude. This is the number that is typically used by the press when a big earthquake happens. It is often referred to as “Richter magnitude,” but that is a misnomer, and it should be just “magnitude.” There are many ways to measure magnitude — including Charles Richter’s method developed in 1935 — but they are all ways to estimate the same number: the amount of energy released.

The other way of assessing the impact of an earthquake is to assess what people felt and how much damage was done. This is known as intensity. Intensity values are assigned to locations, rather than to the earthquake itself, and therefore intensity can vary widely, depending on the proximity to the earthquake and the types of materials and conditions of the subsurface.

Earthquake Magnitude

Before we look more closely at magnitude we need to review what we know about body waves, and look at surface waves. Body waves are of two types, P-waves, or primary or compression waves (like the compression of the coils of a spring), and S-waves, or secondary or shear waves (like the flick of a rope). An example of P and S seismic wave records is shown in Figure 11.13. The critical parameters for the measurement of Richter magnitude are labelled, including the time interval between the arrival of the P- and S-waves — which is used to determine the distance from the earthquake to the seismic station, and the amplitude of the S waves — which is used to estimate the magnitude of the earthquake.

Figure 11.13 P-waves and S-waves from a small (M4) earthquake that took place near Vancouver Island in 1997. [SE]

When body waves (P or S) reach Earth’s surface, some of their energy is transformed into surface waves, of which there are two main types, as illustrated in Figure 11.14. Rayleigh waves are characterized by vertical motion of the ground surface, like waves on water, while Love waves are characterized by horizontal motion. Both Rayleigh and Love waves are about 10% slower than S-waves (so they arrive later at a seismic station). Surface waves typically have greater amplitudes than body waves, and they do more damage.

Figure 11.14 Depiction of seismic surface waves [SE after: https://en.wikipedia.org/wiki/Rayleigh_wave#/media/File:Rayleigh_wave.jpg and https://en.wikipedia.org/wiki/Love_wave#/media/File:Love_wave.jpg]

Other important terms for describing earthquakes are hypocentre (or focus) and epicentre. The hypocentre is the actual location of an individual earthquake shock at depth in the ground, and the epicentre is the point on the land surface directly above the hypocentre (Figure 11.15).

Figure 11.15 Epicentre and hypocentre [SE]

A number of methods for estimating magnitude are listed in Table 11.1. Local magnitude (ML) was widely used until late in the 20th century, but moment magnitude (MW) is now more commonly used because it gives more accurate estimates (especially with larger earthquakes) and can be applied to earthquakes at any distance from a seismometer. Surface-wave magnitudes can also be applied to measure distant large earthquakes.

Because of the increasing size of cities in earthquake-prone areas (e.g., China, Japan, California) and the increasing sophistication of infrastructure, it is becoming important to have very rapid warnings and magnitude estimates of earthquakes that have already happened. This can be achieved by using P-wave data to determine magnitude because P-waves arrive first at seismic stations, in many cases several seconds ahead of the more damaging S-waves and surface waves. Operators of electrical grids, pipelines, trains, and other infrastructure can use the information to automatically shut down systems so that damage and casualties can be limited.

Table 11.1 A summary of some of the different methods for estimating earthquake magnitude. [SE]

The magnitude scale is logarithmic; in fact, the amount of energy released by an earthquake of M4 is 32 times higher than that released by one of M3, and this ratio applies to all intervals in the scale. If we assign an arbitrary energy level of 1 unit to a M1 earthquake the energy for quakes up to M8 will be as shown on the following chart:

In any given year, when there is a large earthquake on Earth (M8 or M9), the amount of energy released by that one event will likely exceed the energy released by all smaller earthquake events combined.

Earthquake Intensity

The intensity of earthquake shaking at any location is determined not only by the magnitude of the earthquake and its distance, but also by the type of underlying rock or unconsolidated materials. If buildings are present, the size and type of buildings (and their inherent natural vibrations) are also important.

Intensity scales were first used in the late 19th century, and then adapted in the early 20th century by Giuseppe Mercalli and modified later by others to form what we know call the modified Mercalli intensity scale (Table 11.2). Intensity estimates are important because they allow us to characterize parts of any region into areas that are especially prone to strong shaking versus those that are not. The key factor in this regard is the nature of the underlying geological materials, and the weaker those are, the more likely it is that there will be strong shaking. Areas underlain by strong solid bedrock tend to experience much less shaking than those underlain by unconsolidated river or lake sediments.

Table 11.2 The modified Mercalli intensity scale. [from http://en.wikipedia.org/wiki/Mercalli_intensity_scale]

An example of this effect is the 1985 M8 earthquake that struck the Michoacán region of western Mexico, southwest of Mexico City. There was relatively little damage in the area around the epicentre, but there was tremendous damage and about 5,000 deaths in heavily populated Mexico City some 350 km from the epicentre. The key reason for this is that Mexico City was built largely on the unconsolidated and water-saturated sediment of former Lake Texcoco. These sediments resonate at a frequency of about two seconds, which was similar to the frequency of the body waves that reached the city. For the same reason that a powerful opera singer can break a wine glass by singing the right note, the amplitude of the seismic waves was amplified by the lake sediments. Survivors of the disaster recounted that the ground in some areas moved up and down by about 20 cm every two seconds for over two minutes. Damage was greatest to buildings between 5 and 15 storeys tall, because they also resonated at around two seconds, which amplified the shaking.

An intensity map for the 1946 M7.3 Vancouver Island earthquake is shown in Figure 11.16. The intensity was greatest in the central island region where, in some communities, chimneys were damaged on more than 75% of buildings, some roads were made impassable, and a major rock slide occurred. The earthquake was felt as far north as Prince Rupert, as far south as Portland Oregon, and as far east as the Rockies

Figure 11.16 Intensity map for the 1946 M7.3 Vancouver Island earthquake. [from: http://www.earthquakescanada.nrcan.gc.ca/historic-historique/events/19460623-eng.php]

Some of the common impacts of earthquakes include structural damage to buildings, fires, damage to bridges and highways, initiation of slope failures, liquefaction, and tsunami. The types of impacts depend to a large degree on where the earthquake is located: whether it is predominantly urban or rural, densely or sparsely populated, highly developed or underdeveloped, and of course on the ability of the infrastructure to withstand shaking.

As we’ve seen from the example of the 1985 Mexico earthquake, the geological foundations on which structures are built can have a significant impact on earthquake shaking. When an earthquake happens, the seismic waves produced have a wide range of frequencies. The energy of the higher frequency waves tends to be absorbed by solid rock, while the lower frequency waves (with periods slower than one second) pass through the solid rock without being absorbed, but are eventually absorbed and amplified by soft sediments. It is therefore very common to see much worse earthquake damage in areas underlain by soft sediments than in areas of solid rock. A good example of this is in the Oakland area near San Francisco, where parts of a two-layer highway built on soft sediments collapsed during the 1989 Loma Prieta earthquake (Figure 11.17).

Figure 11.17 A part of the Cypress Freeway in Oakland California that collapsed during the 1989 Loma Prieta earthquake. [from: http://upload.wikimedia.org/wikipedia/ commons/9/91/Cypress_collapsed.jpg]

Building damage is also greatest in areas of soft sediments, and multi-storey buildings tend to be more seriously damaged than smaller ones. Buildings can be designed to withstand most earthquakes, and this practice is increasingly applied in earthquake-prone regions. Turkey is one such region, and even though Turkey had a relatively strong building code in the 1990s, adherence to the code was poor, as builders did whatever they could to save costs, including using inappropriate materials in concrete and reducing the amount of steel reinforcing. The result was that there were over 17,000 deaths in the 1999 M7.6 Izmit earthquake (Figure 11.18). After two devastating earthquakes that year, Turkish authorities strengthened the building code further, but the new code has been applied only in a few regions, and enforcement of the code is still weak, as revealed by the amount of damage from a M7.1 earthquake in eastern Turkey in 2011.

Figure 11.18 Buildings damaged by the 1999 earthquake in the Izmit area, Turkey. [from U.S. Geological Survey at: http://gallery.usgs.gov/sets/1999_Izmit,_Turkey_Earthquake/thumb/_/1]

Fires are commonly associated with earthquakes because fuel pipelines rupture and electrical lines are damaged when the ground shakes (Figure 11.19). Most of the damage in the great 1906 San Francisco earthquake was caused by massive fires in the downtown area of the city (Figure 11.20). Some 25,000 buildings were destroyed by those fires, which were fuelled by broken gas pipes. Fighting the fires was difficult because water mains had also ruptured. The risk of fires can be reduced through P-wave early warning systems if utility operators can reduce pipeline pressure and close electrical circuits.

Figure 11.19 Some of the effects of the 2011 Tohoku earthquake in the Sendai area of Japan. An oil refinery is on fire, and a vast area has been flooded by a tsunami. [from: http://en.wikipedia.org/wiki/2011_T%C5%8Dhoku_earthquake_and_tsunami#/media/File:SH-60B_helicopter_flies_over_Sendai.jpg]

Figure 11.20 Fires in San Francisco following the 1906 earthquake. [from: http://upload.wikimedia.org/wikipedia/commons/3/3e/San_francisco_fire_1906.jpg]

Earthquakes are important triggers for failures on slopes that are already weak. An example is the Las Colinas slide in the city of Santa Tecla, El Salvador, which was triggered by a M7.6 offshore earthquake in January 2001 (Figure 11.21).

Figure 11.21 The Las Colinas debris flow at Santa Tecla (a suburb of the capital San Salvador) triggered by the January 2001 El Salvador earthquake. This is just one of many hundreds of slope failures that resulted from that earthquake. Over 500 people died in the area affected by this slide. [from: http://landslides.usgs.gov/learning/ images/foreign/ElSalvadorslide.jpg]

Ground shaking during an earthquake can be enough to weaken rock and unconsolidated materials to the point of failure, but in many cases the shaking also contributes to a process known as liquefaction, in which an otherwise solid body of sediment is transformed into a liquid mass that can flow. When water-saturated sediments are shaken, the grains become rearranged to the point where they are no longer supporting one another. Instead, the water between the grains is holding them apart and the material can flow. Liquefaction can lead to the collapse of buildings and other structures that might be otherwise undamaged. A good example is the collapse of apartment buildings during the 1964 Niigata earthquake (M7.6) in Japan (Figure 11.22). Liquefaction can also contribute to slope failures and to fountains of sandy mud (sand volcanoes) in areas where there is loose saturated sand beneath a layer of more cohesive clay.

Figure 11.22 Collapsed apartment buildings in the Niigata area of Japan. The material beneath the buildings was liquefied to varying degrees by the 1964 earthquake. http://en.wikipedia.org/wiki/1964_ Niigata_earthquake#/media/File: Liquefaction_at_Niigata.JPG

Parts of the Fraser River delta are prone to liquefaction-related damage because the region is characterized by a 2 m to 3 m thick layer of fluvial silt and clay over top of at least 10 m of water-saturated fluvial sand (Figure 11.23). Under these conditions, it can be expected that seismic shaking will be amplified and, the sandy sediments will liquefy. This could lead to subsidence and tilting of buildings, and to failure and sliding of the silt and clay layer. Current building-code regulations in the Fraser delta area require that measures be taken to strengthen the ground underneath multi-storey buildings prior to construction.

Figure 11.23 Recent unconsolidated sedimentary layers in the Fraser River delta area (top) and the potential consequences in the event of a damaging earthquake. [SE]

Earthquakes that take place beneath the ocean have the potential to generate tsunami.[footnote]Tsunami is the Japanese word for harbour wave. It is the same in both singular and plural.[/footnote] The most likely situation for a significant tsunami is a large (M7 or greater) subduction-related earthquake. As shown in Figure 11.24, during the time between earthquakes the overriding plate becomes distorted by elastic deformation; it is squeezed laterally (Figure 11.24B) and pushed up.

When an earthquake happens (Figure 11.24C), the plate rebounds and there is both uplift and subsidence on the sea floor, in some cases by as much as several metres vertically over an area of thousands of square kilometres. This vertical motion is transmitted through the water column where it generates a wave that then spreads across the ocean.

Figure 11.24 Elastic deformation and rebound of overriding plate at a subduction setting (B). The release of the locked zone during an earthquake (C) results in both uplift and subsidence on the sea floor, and this is transmitted to the water overhead, resulting in a tsunami. [SE]

Subduction earthquakes with magnitude less than 7 do not typically generate significant tsunami because the amount of vertical displacement of the sea floor is minimal. Sea-floor transform earthquakes, even large ones (M7 to M8), don’t typically generate tsunami either, because the motion is mostly side to side, not vertical.

Tsunami waves travel at velocities of several hundred kilometres per hour and easily make it to the far side of an ocean in about the same time as a passenger jet. The simulated one shown in Figure 11.25 is similar to that created by the 1700 Cascadia earthquake off the coast of British Columbia, Washington, and Oregon, which was recorded in Japan nine hours later.

Figure 11.25 Model of the tsunami from the 1700 Cascadia earthquake (~M9) showing open-ocean wave heights (colours) and travel time contours. Tsunami wave amplitudes typically increase in shallow water. [from NOAA/PMEL/Center for Tsunami Research, at: http://nctr.pmel.noaa.gov/cascadia_simulated/]

Tsunami are discussed further in Chapter 17 under the topic of waves and coasts.

It has long been a dream of seismologists, geologists, and public safety officials to be able to accurately predict the location, magnitude, and timing of earthquakes on time scales that would be useful for minimizing danger to the public and damage to infrastructure (e.g., weeks, days, hours). Many different avenues of prediction have been explored, such as using observations of warning foreshocks, changes in magnetic fields, seismic tremor, changing groundwater levels, strange animal behaviour, observed periodicity, stress transfer considerations, and others. So far, none of the research into earthquake prediction has provided a reliable method. Although there are some reports of successful earthquake predictions, they are rare, and many are surrounded by doubtful circumstances.

The problem with earthquake predictions, as with any other type of prediction, is that they have to be accurate most of the time, not just some of the time. We have come to rely on weather predictions because they are generally (and increasingly) accurate. But if we try to predict earthquakes and are only accurate 10% of the time (and even that isn’t possible with the current state of knowledge), the public will lose faith in the process very quickly, and then will ignore all of the predictions. Efforts are currently focused on forecasting earthquake probabilities, rather than predicting their occurrence.

There was great hope for earthquake predictions late in the 1980s when attention was focused on part of the San Andreas Fault at Parkfield, about 200 km south of San Francisco. Between 1881 and 1965 there were five earthquakes at Parkfield, most spaced at approximately 20-year intervals, all confined to the same 20 km-long segment of the fault, and all very close to M6 (Figure 11.26). Both the 1934 and 1966 earthquakes were preceded by small foreshocks exactly 17 minutes before the main quake.

Figure 11.26 Earthquakes on the Parkfield segment of the San Andreas Fault between 1881 and 2004. [SE]

The U.S. Geological Survey recognized this as an excellent opportunity to understand earthquakes and earthquake prediction, so they armed the Parkfield area with a huge array of geophysical instruments and waited for the next quake, which was expected to happen around 1987. Nothing happened! The “1987 Parkfield earthquake” finally struck in September 2004. Fortunately all of the equipment was still there, but it was no help from the perspective of earthquake prediction. There were no significant precursors to the 2004 Parkfield earthquake in any of the parameters measured, including seismicity, harmonic tremor, strain (rock deformation), magnetic field, the conductivity of the rock, or creep, and there was no foreshock. In other words, even though every available technique was used to monitor it, the 2004 earthquake came as a complete surprise, with no warning whatsoever.

The hope for earthquake prediction is not dead, but it was hit hard by the Parkfield experiment. The current focus in earthquake-prone regions is to provide forecasts of the probability of an earthquake of a certain magnitude within a certain time period — typically a number of decades — while officials focus on ensuring that the population is educated about earthquake risks and that buildings and other infrastructure are as safe as can be. An example of this approach for the San Francisco Bay region of California is shown in Figure 11.27. Based on a wide range of information, including past earthquake history, accumulated stress from plate movement, and known stress transfer, seismologists and geologists have predicted the likelihood of a M6.7 or greater earthquake on each of eight major faults that cut through the region. The greatest probabilities are on the San Andreas, Rogers Creek, and Hayward Faults. As shown in Figure 11.27, there is a 63% chance that a major and damaging earthquake will take place somewhere in the region prior to 2036.

Figure 11.27 Probabilities of a M6.7 or larger earthquake over the period 2007 to 2036 on various faults in the San Francisco Bay region of California. [from USGS at: http://earthquake.usgs.gov/regional/nca/ucerf/]

As we’ve discussed already, it’s not sufficient to have strong building codes, they have to be enforced. Building code compliance is quite robust in most developed countries, but is sadly inadequate in many developing countries.

It’s also not enough just to focus on new buildings; we have to make sure that existing buildings — especially schools and hospitals — and other structures such as bridges and dams, are as safe as they can be. An example of how this is applied to schools in B.C. is described in Box 11.2.

Making the seismic upgrade in B.C.’s schools

British Columbia is in the middle of a multi-billion-dollar program to make schools safer for students. The program is focused on older schools, because, according to the government, those built since 1992 already comply with modern seismic codes. Some schools would require too much work to make upgrading economically feasible and they are replaced. Where upgrading is feasible, the school is assessed carefully before any upgrade work is initiated.

An example is Sangster Elementary in Colwood on southern Vancouver Island. The school was originally built in 1957, with a major addition in 1973. Ironically, the newer part of the school, built of concrete blocks, required strengthening with the addition of a steel framework, while the 1957 part, which is a wood-frame building, did not require seismic upgrading. The work was completed in 2014.

[Sangster Elementary image from Google Maps – street view]

As of January 2015, upgrades had been completed at 145 B.C. schools, 11 were underway, and an additional 57 were ready to proceed with funding identified.* Another 129 schools were listed as needing upgrades. In May 2015, the provincial government announced that the target date for completion of the upgrades, originally set for 2020, had been delayed to 2030.* http://www2.gov.bc.ca/gov/topic.page?id=00C5FFBE51C44325A845819C007A01E7

The final part of earthquake preparedness involves the formulation of public emergency plans, including escape routes, medical facilities, shelters, and food and water supplies. It also includes personal planning, such as emergency supplies (food, water, shelter, and warmth), escape routes from houses and offices, and communication strategies (with a focus on ones that don’t involve the cellular network).

The topics covered in this chapter can be summarized as follows:

Introduction

Figure 12.1  Folds in sedimentary rocks near Golden and the Kickinghorse River, B.C. [SE]

Observing and understanding geological structures helps us to determine the kinds of stresses that have existed within Earth in the past.  This type of information is critical to our understanding of plate tectonics, earthquakes, the formation of mountains, metamorphism, and Earth resources.  Some of the types of geological structures that are important to study include fractures, faults, and folds.  Structural geologists make careful observations of the orientations of these structures and the amount and direction of offset along faults.

Rocks are subject to stress —mostly related to plate tectonics but also to the weight of overlying rocks—and their response to that stress is strain (deformation).  In regions close to where plates are converging stress is typically compressive—the rocks are being squeezed.  Where plates are diverging the stress is extensive—rocks are being pulled apart.  At transform plate boundaries, where plates are moving side by side there is sideways or shear stress—meaning that there are forces in opposite directions parallel to a plane. Rocks have highly varying strain responses to stress because of their different compositions and physical properties, and because temperature is a big factor and rock temperatures within the crust can vary greatly.

We can describe the stress applied to a rock by breaking it down into three dimensions—all at right angles to one-another (Figure 12.2). If the rock is subject only to the pressure of burial, the stresses in all three directions will likely be the same.  If it is subject to both burial and tectonic forces, the pressures will be different in different directions.

Figure 12.2 Depiction of the stress applied to rocks within the crust. The stress can be broken down into three components. Assuming that we’re looking down in this case, the green arrows represent north-south stress, the red arrows represent east-west stress, and the blue arrows (the one underneath is not visible) represent up-down stress. On the left, all of the stress components are the same. On the right, the north-south stress is least and the up-down stress is greatest. [SE]

Rock can respond to stress in three ways: it can deform elastically, it can deform plastically, and it can break or fracture.  Elastic strain is reversible; if the stress is removed, the rock will return to its original shape just like a rubber band that is stretched and released. Plastic strain is not reversible. As already noted, different rocks at different temperatures will behave in different ways to stress. Higher temperatures lead to more plastic behaviour. Some rocks or sediments are also more plastic when they are wet.  Another factor is the rate at which the stress is applied.  If the stress is applied quickly (for example, because of an extraterrestrial impact or an earthquake), there will be an increased tendency for the rock to fracture. Some different types of strain response are illustrated in Figure 12.3.

Figure 12.3 The varying types of response of geological materials to stress. The straight dashed parts are elastic strain and the curved parts are plastic strain. In each case the X marks where the material fractures. A, the strongest material, deforms relatively little and breaks at a high stress level. B, strong but brittle, shows no plastic deformation and breaks after relatively little elastic deformation. C, the most deformable, breaks only after significant elastic and plastic strain.  The three deformation diagrams on the right show A and C before breaking and B after breaking. [SE]

The outcomes of placing rock under stress are highly variable, but they include fracturing, tilting and folding, stretching and squeezing, and faulting. A fracture is a simple break that does not involve significant movement of the rock on either side. Fracturing is particularly common in volcanic rock, which shrinks as it cools. The basalt columns in Figure 12.4a are a good example of fracture. Beds are sometimes tilted by tectonic forces, as shown in Figure 12.4b, or folded as shown in Figure 12.1.

Figure 12.4 Rock structures caused by various types of strain within rocks that have been stressed [all by SE]

When a body of rock is compressed in one direction it is typically extended (or stretched) in another.  This is an important concept because some geological structures only form under compression, while others only form under tension. Most of the rock in Figure 12.4c is limestone, which is relatively easily deformed when heated. The dark rock is chert, which remains brittle. As the limestone stretched (parallel to the hammer handle) the brittle chert was forced to break into fragments to accommodate the change in shape of the body of rock. A fault is a rock boundary along which the rocks on either side have been displaced relative to each other (Figure 12.4d).

When a body of rock, especially sedimentary rock, is squeezed from the sides by tectonic forces, it is likely to fracture and/or become faulted if it is cold and brittle, or become folded if it is warm enough to behave in a plastic manner.

The nomenclature and geometry of folds are summarized on Figure 12.5.  An upward fold is called an anticline, while a downward fold is called a syncline. In many areas it’s common to find a series of anticlines and synclines (as in Figure 12.5), although some sequences of rocks are folded into a single anticline or syncline. A plane drawn through the crest of a fold in a series of beds is called the axial plane of the fold. The sloping beds on either side of an axial plane are limbs. An anticline or syncline is described as symmetrical if the angles between each of limb and the axial plane are generally similar, and asymmetrical if they are not. If the axial plane is sufficiently tilted that the beds on one side have been tilted past vertical, the fold is known as an overturned anticline or syncline.

Figure 12.5 Examples of different types of folds and fold nomenclature. Axial planes are only shown for the anticlines, but synclines also have axial planes. [SE]

A very tight fold, in which the limbs are parallel or nearly parallel to one another is called an isoclinal fold (Figure 12.6). Isoclinal folds that have been overturned to the extent that their limbs are nearly horizontal are called recumbent folds.

Figure 12.6 An isoclinal recumbent fold [SE]

Folds can be of any size, and it’s very common to have smaller folds within larger folds (Figure 12.7).  Large folds can have wavelengths of tens of kilometres, and very small ones might be visible only under a microscope. Anticlines are not necessarily, or even typically, expressed as ridges in the terrain, nor synclines as valleys. Folded rocks get eroded just like all other rocks and the topography that results is typically controlled mostly by the resistance of different layers to erosion (Figure 12.8).

Figure 12.7 Folded limestone (grey) and chert (rust-coloured) in Triassic Quatsino Formation rocks on Quadra Island, B.C.  The image is about 1 m across. [SE]

Figure 12.8 Example of the topography in an area of folded rocks that has been eroded. In this case the green and grey rocks are most resistant to erosion, and are represented by hills. [SE]

Exercises

Exercise 12.1  Folding Style

This photograph shows folding in the same area of the Rocky Mountains as Figure 12.1.  Describe the types of folds using the appropriate terms from above (symmetrical, asymmetrical, isoclinal, overturned, recumbent etc.).  You might find it useful to first sketch in the axial planes.

A body of rock that is brittle—either because it is cold or because of its composition, or both— is likely to break rather than fold when subjected to stress, and the result is fracturing or faulting.

Fracturing

Fracturing is common in rocks near the surface, either in volcanic rocks that have shrunk on cooling (Figure 12.4a), or in other rocks that have been exposed by erosion and have expanded (Figure 12.9).

Figure 12.9 Granite in the Coquihalla Creek area, B.C. (left) and sandstone at Nanoose, B.C. (right), both showing fracturing that has resulted from expansion due to removal of overlying rock. [SE]

A fracture in a rock is also called a joint. There is no side-to-side movement of the rock on either side of a joint. Most joints form where a body of rock is expanding because of reduced pressure, as shown by the two examples in Figure 12.9, or where the rock itself is contracting but the body of rock remains the same size (the cooling volcanic rock in Figure 12.4a). In all of these cases, the pressure regime is one of tension as opposed to compression. Joints can also develop where rock is being folded because, while folding typically happens during compression, there may be some parts of the fold that are in tension (Figure 12.10).

Figure 12.10  A depiction of joints developed in the hinge area of folded rocks. Note that in this situation some rock types are more likely to fracture than others.  [SE]

Finally joints can also develop when rock is under compression as shown on Figure 12.11, where there is differential stress on the rock, and joint sets develop at angles to the compression directions.

Figure 12.11  A depiction of joints developed in a rock that is under stress.  [SE]

Faulting

A fault is boundary between two bodies of rock along which there has been relative motion (Figure 12.4d). As we discussed in Chapter 11, an earthquake involves the sliding of one body of rock past another. Earthquakes don’t necessarily happen on existing faults, but once an earthquake takes place a fault will exist in the rock at that location. Some large faults, like the San Andreas Fault in California or the Tintina Fault, which extends from northern B.C. through central Yukon and into Alaska, show evidence of hundreds of kilometres of motion, while others show less than a millimetre. In order to estimate the amount of motion on a fault, we need to find some geological feature that shows up on both sides and has been offset (Figure 12.12).

Figure 12.12  A fault (white dashed line) in intrusive rocks on Quadra Island, B.C. The pink dyke has been offset by the fault and the extent of the offset is shown by the white arrow (approximately 10 cm). Because the far side of the fault has moved to the right, this is a right-lateral fault. If the photo were taken from the other side, the fault would still appear to have a right-lateral offset.   [SE]

There are several kinds of faults, as illustrated on Figure 12.13, and they develop under different stress conditions. The terms hanging wall and footwall in the diagrams apply to situations where the fault is not vertical. The body of rock above the fault is called the hanging wall, and the body of rock below it is called the footwall. If the fault develops in a situation of compression, then it will be a reverse fault because the compression causes the hanging wall to be pushed up relative to the footwall. If the fault develops in a situation of extension, then it will be a normal fault, because the extension allows the hanging wall to slide down relative to the footwall in response to gravity.

The third situation is where the bodies of rock are sliding sideways with respect to each other, as is the case along a transform fault (see Chapter 10). This is known as a strike-slip fault because the displacement is along the “strike” or the length of the fault. On strike-slip faults the motion is typically only horizontal, or with a very small vertical component, and as discussed above the sense of motion can be right lateral (the far side moves to the right), as in Figures 12.12 and 12.13, or it can be left lateral (the far side moves to the left). Transform faults are strike-slip faults.

Figure 12.13  Depiction of reverse, normal, and strike-slip faults. Reverse faults happen during compression while normal faults happen during extension. Most strike-slip faults are related to transform boundaries. [SE after: http://www.nature.nps.gov/geology/education/images/GRAPHICS/fault_types_2.jpg]

In areas that are characterized by extensional tectonics, it is not uncommon for a part of the upper crust to subside with respect to neighbouring parts. This is typical along areas of continental rifting, such as the Great Rift Valley of East Africa or in parts of Iceland, but it is also seen elsewhere. In such situations a down-dropped block is known as a graben (German for ditch), while an adjacent block that doesn’t subside is called a horst (German for heap) (Figure 12.14). There are many horsts and grabens in the Basin and Range area of the western United States, especially in Nevada. Part of the Fraser Valley region of B.C., in the area around Sumas Prairie is a graben.

Figure 12.14  Depiction of graben and horst structures that form in extensional situations. All of the faults are normal faults.  [SE]

A special type of reverse fault, with a very low-angle fault plane, is known as a thrust fault. Thrust faults are relatively common in areas where fold-belt mountains have been created during continent-continent collision. Some represent tens of kilometres of thrusting, where thick sheets of sedimentary rock have been pushed up and over top of other rock (Figure 12.15).

Figure 12.15 Depiction a thrust fault. Top: prior to faulting. Bottom: after significant fault offset. [SE]

There are numerous thrust faults in the Rocky Mountains, and a well-known example is the McConnell Thrust, along which a sequence of sedimentary rocks about 800 m thick has been pushed for about 40 km from west to east (Figure 12.16). The thrusted rocks range in age from Cambrian to Cretaceous, so in the area around Mt. Yamnuska Cambrian-aged rock (around 500 Ma) has been thrust over, and now lies on top of Cretaceous-aged rock (around 75 Ma) (Figure 12.17).

Figure 12.16  Depiction of the McConnell Thrust in the eastern part of the Rockies. The rock within the faded area has been eroded. [SE]

Figure 12.17  The McConnell Thrust at Mt. Yamnuska near Exshaw, Alberta. Carbonate rocks (limestone) of Cambrian age have been thrust over top of Cretaceous mudstone.  [SE]

Geologists take great pains to measure and record geological structures because they are critically important to understanding the geological history of a region. One of the key features to measure is the orientation, or attitude, of bedding. We know that sedimentary beds are deposited in horizontal layers, so if the layers are no longer horizontal, then we can infer that they have been affected by tectonic forces and have become either tilted, or folded. We can express the orientation of a bed (or any other planar feature) with two values: first, the compass orientation of a horizontal line on the surface—the strike—and second, the angle at which the surface dips from the horizontal, (perpendicular to the strike)—the dip (Figure 12.18).

It may help to imagine a vertical surface, such as a wall in your house. The strike is the compass orientation of the wall and the dip is 90˚ from horizontal. If you could push the wall so it’s leaning over, but still attached to the floor, the strike direction would be the same, but the dip angle would be less than 90˚. If you pushed the wall over completely so it was lying on the floor, it would no longer have a strike direction and its dip would be 0˚. When describing the dip it is important to include the direction. In other words. if the strike is 0˚ (i.e., north) and the dip is 30˚, it would be necessary to say “to the west” or “to the east.”  Similarly if the strike is 45˚ (i.e., northeast) and the dip is 60˚, it would be necessary to say “to the northwest” or “to the southeast.”

Measurement of geological features is done with a special compass that has a built-in clinometer, which is a device for measuring vertical angles. An example of how this is done is shown on Figure 12.19.

Figure 12.18  A depiction of the strike and dip of some tilted sedimentary beds partially covered with water. The notation for expressing strike and dip on a map is shown.  [SE]

Figure 12.19 Measurement of strike (left) and dip (right) using a geological compass with a clinometer.  [SE]

Strike and dip are also used to describe any other planar features, including joints, faults, dykes, sills, and even the foliation planes in metamorphic rocks. Figure 12.20 shows an example of how we would depict the beds that make up an anticline on a map.

Figure 12.20 A depiction of an anticline and a dyke in cross-section (looking from the side) and in map view (a.k.a. plan view) with the appropriate strike-dip and anticline symbols. [SE]

The beds on the west (left) side of the map are dipping at various angles to the west. The beds on the east side are dipping to the east. The middle bed (light grey) is horizontal; this is denoted by a cross within a circle. The dyke is dipping at 80˚ to the west. The hinge of the fold is denoted with a dashed line with two arrows point away from it.  If it were a syncline, the arrows would point towards the line.

The topics covered in this chapter can be summarized as follows:

Introduction

Figure 13.1 A small waterfall on Johnston Creek in Johnston Canyon, Banff National Park, AB [SE]

Streams are the most important agents of erosion and transportation of sediments on Earth’s surface. They are responsible for the creation of much of the topography that we see around us. They are also places of great beauty and tranquility, and of course, they provide much of the water that is essential to our existence. But streams are not always peaceful and soothing. During large storms and rapid snowmelts, they can become raging torrents capable of moving cars and houses and destroying roads and bridges. When they spill over their banks, they can flood huge areas, devastating populations and infrastructure. Over the past century, many of the most damaging natural disasters in Canada have been floods, and we can expect them to become even more severe as the climate changes.

Water is constantly on the move. It is evaporated from the oceans, lakes, streams, the surface of the land, and plants (transpiration) by solar energy (Figure 13.2). It is moved through the atmosphere by winds and condenses to form clouds of water droplets or ice crystals. It comes back down as rain or snow and then flows through streams, into lakes, and eventually back to the oceans. Water on the surface and in streams and lakes infiltrates the ground to become groundwater. Groundwater slowly moves through the rock and surficial materials. Some groundwater returns to other streams and lakes, and some goes directly back to the oceans.

Figure 13.2 The various components of the water cycle. Black or white text indicates the movement or transfer of water from one reservoir to another. Yellow text indicates the storage of water. [SE after Wikipedia: http://upload.wikimedia.org/wikipedia/commons/5/54/Water_cycle_blank.svg]

Even while it’s moving around, water is stored in various reservoirs. The largest, by far, is the oceans, accounting for 97% of the volume (Figure 13.3). Of course, that water is salty. The remaining 3% is fresh water. Two-thirds of our fresh water is stored in the ground and one-third is stored in ice. The remaining fresh water — about 0.03% of the total — is stored in lakes, streams, vegetation, and the atmosphere.

Figure 13.3a The storage reservoirs for water on Earth. Glacial ice is represented by the white band, groundwater the red band, and surface water the very thin blue band at the top. The 0.001% stored in the atmosphere is not shown. [SE using data from: https://water.usgs.gov/edu/ watercyclefreshstorage.html]

To put that in perspective, let’s think about putting all of Earth’s water into a 1 L jug (Figure 13.3b). We start by almost filling the jug with 970 ml of water and 34 g of salt. Then we add one regular-sized (~20 mL) ice cube (representing glacial ice) and two teaspoons (~10 mL) of groundwater. All of the water that we see around us in lakes and streams and up in the sky can be represented by adding three more drops from an eyedropper.

Figure 13.3b Representation of the Earth’s water as a 1 L jug. The three drops represent all of the fresh water in lakes, streams, and wetlands, plus all of the water in the atmosphere. [SE]

Although the proportion of Earth’s water that is in the atmosphere is tiny, the actual volume is huge. At any given time, there is the equivalent of approximately 13,000 km³ of water in the air in the form of water vapour and water droplets in clouds. Water is evaporated from the oceans, vegetation, and lakes at a rate of 1,580 km³ per day, and just about exactly the same volume falls as rain and snow every day — over both the oceans and land. The precipitation that falls on land goes back to the ocean in the form of stream flow (117 km³/day) and groundwater flow (6 km³/day). Most of the rest of this chapter is about that 117 km³/day of streamflow. The average discharge of the Fraser River into the ocean is approximately 0.31 km³/day, or 0.26% of the world’s total.

A stream is a body of flowing surface water of any size, ranging from a tiny trickle to a mighty river. The area from which the water flows to form a stream is known as its drainage basin. All of the precipitation (rain or snow) that falls within a drainage basin eventually flows into its stream, unless some of that water is able to cross into an adjacent drainage basin via groundwater flow. An example of a drainage basin is shown in Figure 13.4.

Figure 13.4 Cawston Creek near Keremeos, B.C. The blue line shows the extent of the drainage basin. The dashed red line is the drainage basin of one of its tributaries. [SE]

Figure 13.5 Profile of the main stem of Cawston Creek near Keremeos, B.C. The maximum elevation of the drainage basin is about 1,840 m, near Mount Kobau. The base level is 275 m, at the Similkameen River. As shown, the gradient of the stream can be determined by dividing the change in elevation between any two points (rise) by the distance between those two points (run). [SE]

Cawston Creek is a typical small drainage basin (approximately 25 km²) within a very steep glaciated valley. As shown in Figure 13.5, the upper and middle parts of the creek have steep gradients (averaging about 200 m/km but ranging from 100 to 350 m/km), and the lower part, within the valley of the Similkameen River, is relatively flat (<5 m/km). The shape of the valley has been controlled first by tectonic uplift (related to plate convergence), then by pre-glacial stream erosion and mass wasting, then by several episodes of glacial erosion, and finally by post-glacial stream erosion. The lowest elevation of Cawston Creek (275 m at the Similkameen River) is its base level. Cawston Creek cannot erode below that level unless the Similkameen River erodes deeper into its flood plain (the area that is inundated during a flood).

Metro Vancouver’s water supply comes from three large drainage basins on the north shore of Burrard Inlet, as shown in Figure 13.6. This map illustrates the concept of a drainage basin divide. The boundary between two drainage basins is the height of land between them. A drop of water falling on the boundary between the Capilano and Seymour drainage basins (a.k.a., watersheds), for example, could flow into either one of them.

Figure 13.6 The three drainage basins that are used for the Metro Vancouver water supply. [Used with permission of Metro Vancouver]

The pattern of tributaries within a drainage basin depends largely on the type of rock beneath, and on structures within that rock (folds, fractures, faults, etc.). The three main types of drainage patterns are illustrated in Figure 13.7. Dendritic patterns, which are by far the most common, develop in areas where the rock (or unconsolidated material) beneath the stream has no particular fabric or structure and can be eroded equally easily in all directions. Examples would be granite, gneiss, volcanic rock, and sedimentary rock that has not been folded. Most areas of British Columbia have dendritic patterns, as do most areas of the prairies and the Canadian Shield. Trellis drainage patterns typically develop where sedimentary rocks have been folded or tilted and then eroded to varying degrees depending on their strength. The Rocky Mountains of B.C. and Alberta are a good example of this, and many of the drainage systems within the Rockies have trellis patterns. Rectangular patterns develop in areas that have very little topography and a system of bedding planes, fractures, or faults that form a rectangular network. Rectangular drainage patterns are rare in Canada.

In many parts of Canada, especially relatively flat areas with thick glacial sediments, and throughout much of Canadian Shield in eastern and central Canada, drainage patterns are chaotic, or what is known as deranged (Figure 13.8, left). Lakes and wetlands are common in this type of environment.

Figure 13.7 Typical dendritic, trellis, and rectangular stream drainage patterns. [SE]

A fourth type of drainage pattern, which is not specific to a drainage basin, is known as radial (Figure 13.8, right). Radial patterns form around isolated mountains (such as volcanoes) or hills, and the individual streams typically have dendritic drainage patterns.

Figure 13.8 Left: a typical deranged pattern; right: a typical radial drainage pattern developed around a mountain or hill. [SE]

Over geological time, a stream will erode its drainage basin into a smooth profile similar to that shown in Figure 13.9. If we compare this with an ungraded stream like Cawston Creek (Figure 13.5), we can see that graded streams are steepest in their headwaters and their gradient gradually decreases toward their mouths. Ungraded streams have steep sections at various points, and typically have rapids and waterfalls at numerous locations along their lengths.

Figure 13.9 The topographic profile of a typical graded stream. [SE]

A graded stream can become ungraded if there is renewed tectonic uplift, or if there is a change in the base level, either because of tectonic uplift or some other reason. As stated earlier, the base level of Cawston Creek is defined by the level of the Similkameen River, but this can change, and has done so in the past. Figure 13.10 shows the valley of the Similkameen River in the Keremeos area. The river channel is just beyond the row of trees. The green field in the distance is underlain by material eroded from the hills behind and deposited by a small creek (not Cawston Creek) adjacent to the Similkameen River when its level was higher than it is now. Sometime in the past several centuries, the Similkameen River eroded down through these deposits (forming the steep bank on the other side of the river), and the base level of the small creek was lowered by about 10 m. Over the next few centuries, this creek will seek to become graded again by eroding down through its own alluvial fan.

Figure 13.10 An example of a change in the base level of a small stream that flows into the Similkameen river near Keremeos. The previous base level was near the top of the sandy bank. The current base level is the river. [SE]

Another example of a change in base level can be seen along the Juan de Fuca Trail on southwestern Vancouver Island. As shown in Figure 13.11, many of the small streams along this part of the coast flow into the ocean as waterfalls. It is evident that the land in this area has risen by about 5 m in the past few thousand years, probably in response to deglaciation. The streams that used to flow directly into the ocean now have a lot of down-cutting to do to become regraded.

Figure 13.11 Two streams with a lowered base level on the Juan de Fuca Trail, southwestern Vancouver Island. [SE]

The ocean is the ultimate base level, but lakes and other rivers act as base levels for many smaller streams. We can create an artificial base level on a stream by constructing a dam.

Exercises

Exercise 13.2 The Effect of a Dam on Base Level

Revelstoke Dam and Revelstoke Lake on the Columbia River at Revelstoke, BC [SE]

When a dam is built on a stream, a reservoir (artificial lake) forms behind the dam, and this temporarily (for many decades at least) creates a new base level for the part of the stream above the reservoir. How does the formation of a reservoir affect the stream where it enters the reservoir, and what happens to the sediment it was carrying? The water leaving the dam has no sediment in it. How does this affect the stream below the dam?

Sediments accumulate within the flood plain of a stream, and then, if the base level changes, or if there is less sediment to deposit, the stream may cut down through those existing sediments to form terraces. A terrace on the Similkameen River is shown in Figure 13.10 and some on the Fraser River are shown in Figure 13.12. The Fraser River photo shows at least two levels of terraces.

Figure 13.12 Terraces on the Fraser River at High Bar. [Marie Betcher photo, used with permission]

In the late 19th century, American geologist William Davis proposed that streams and the surrounding terrain develop in a cycle of erosion (Figure 13.13). Following tectonic uplift, streams erode quickly, developing deep V-shaped valleys that tend to follow relatively straight paths. Gradients are high, and profiles are ungraded. Rapids and waterfalls are common. During the mature stage, streams erode wider valleys and start to deposit thick sediment layers. Gradients are slowly reduced and grading increases. In old age, streams are surrounded by rolling hills, and they occupy wide sediment-filled valleys. Meandering patterns are common.

Davis’s work was done long before the idea of plate tectonics, and he was not familiar with the impacts of glacial erosion on streams and their environments. While some parts of his theory are out of date, it is still a useful way to understand streams and their evolution.

Figure 13.13 A depiction of the Davis cycle of erosion: a: initial stage, b: youthful stage, c: mature stage, and d: old age. [SE]

As we discussed in Chapter 6, flowing water is a very important mechanism for both erosion and deposition. Water flow in a stream is primarily related to the stream’s gradient, but it is also controlled by the geometry of the stream channel. As shown in Figure 13.14, water flow velocity is decreased by friction along the stream bed, so it is slowest at the bottom and edges and fastest near the surface and in the middle. In fact, the velocity just below the surface is typically a little higher than right at the surface because of friction between the water and the air. On a curved section of a stream, flow is fastest on the outside and slowest on the inside.

Figure 13.14 The relative velocity of stream flow depending on whether the stream channel is straight or curved (left), and with respect to the water depth (right). [SE]

Other factors that affect stream-water velocity are the size of sediments on the stream bed — because large particles tend to slow the flow more than small ones — and the discharge, or volume of water passing a point in a unit of time (e.g., m³/second). During a flood, the water level always rises, so there is more cross-sectional area for the water to flow in; however, as long as a river remains confined to its channel, the velocity of the water flow also increases.

Figure 13.15 shows the nature of sediment transportation in a stream. Large particles rest on the bottom — bedload — and may only be moved during rapid flows under flood conditions. They can be moved by saltation (bouncing) and by traction (being pushed along by the force of the flow).

Smaller particles may rest on the bottom some of the time, where they can be moved by saltation and traction, but they can also be held in suspension in the flowing water, especially at higher velocities. As you know from intuition and from experience, streams that flow fast tend to be turbulent (flow paths are chaotic and the water surface appears rough) and the water may be muddy, while those that flow more slowly tend to have laminar flow (straight-line flow and a smooth water surface) and clear water. Turbulent flow is more effective than laminar flow at keeping sediments in suspension.

Stream water also has a dissolved load, which represents (on average) about 15% of the mass of material transported, and includes ions such as calcium (Ca⁺²) and chloride (Cl-) in solution. The solubility of these ions is not affected by flow velocity.

Figure 13.15 Modes of transportation of sediments and dissolved ions (represented by red dots with + and – signs) in a stream. [SE]

The faster the water is flowing, the larger the particles that can be kept in suspension and transported within the flowing water. However, as Swedish geographer Filip Hjulström discovered in the 1940s, the relationship between grain size and the likelihood of a grain being eroded, transported, or deposited is not as simple as one might imagine (Figure 13.16). Consider, for example, a 1 mm grain of sand. If it is resting on the bottom, it will remain there until the velocity is high enough to erode it, around 20 cm/s. But once it is in suspension, that same 1 mm particle will remain in suspension as long as the velocity doesn’t drop below 10 cm/s. For a 10 mm gravel grain, the velocity is 105 cm/s to be eroded from the bed but only 80 cm/s to remain in suspension.

Figure 13.16 The Hjulström-Sundborg diagram showing the relationships between particle size and the tendency to be eroded, transported, or deposited at different current velocities

On the other hand, a 0.01 mm silt particle only needs a velocity of 0.1 cm/s to remain in suspension, but requires 60 cm/s to be eroded. In other words, a tiny silt grain requires a greater velocity to be eroded than a grain of sand that is 100 times larger! For clay-sized particles, the discrepancy is even greater. In a stream, the most easily eroded particles are small sand grains between 0.2 mm and 0.5 mm. Anything smaller or larger requires a higher water velocity to be eroded and entrained in the flow. The main reason for this is that small particles, and especially the tiny grains of clay, have a strong tendency to stick together, and so are difficult to erode from the stream bed.

It is important to be aware that a stream can both erode and deposit sediments at the same time. At 100 cm/s, for example, silt, sand, and medium gravel will be eroded from the stream bed and transported in suspension, coarse gravel will be held in suspension, pebbles will be both transported and deposited, and cobbles and boulders will remain stationary on the stream bed.

A stream typically reaches its greatest velocity when it is close to flooding over its banks. This is known as the bank-full stage, as shown in Figure 13.17. As soon as the flooding stream overtops its banks and occupies the wide area of its flood plain, the water has a much larger area to flow through and the velocity drops significantly. At this point, sediment that was being carried by the high-velocity water is deposited near the edge of the channel, forming a natural bank or levée.

Figure 13.17 The development of natural levées during flooding of a stream. The sediments of the levée become increasingly fine away from the stream channel, and even finer sediments — clay, silt, and fine sand — are deposited across most of the flood plain. [SE]

Stream channels can be straight or curved, deep and slow, or rapid and choked with coarse sediments. The cycle of erosion has some influence on the nature of a stream, but there are several other factors that are important.

Youthful streams that are actively down-cutting their channels tend to be relatively straight and are typically ungraded (meaning that rapids and falls are common). As shown in Figures 13.1 and 13.18, youthful streams commonly have a step-pool morphology, meaning that the stream consists of a series of pools connected by rapids and waterfalls. They also have steep gradients and steep and narrow V-shaped valleys — in some cases steep enough to be called canyons.

Figure 13.18 The Cascade Falls area of the Kettle River, near Christina Lake, B.C. This stream has a step-pool morphology and a deep bedrock channel. [SE]

In mountainous terrain, such as that in western Alberta and B.C., steep youthful streams typically flow into wide and relatively low-gradient U-shaped glaciated valleys. The youthful streams have high sediment loads, and when they flow into the lower-gradient glacial valleys where the velocity isn’t high enough to carry all of the sediment, braided patterns develop, characterized by a series of narrow channels separated by gravel bars (Figure 13.19).

Figure 13.19 The braided channel of the Kicking Horse River at Field, B.C. [SE]

Braided streams can develop anywhere there is more sediment than a stream is able to transport. One such environment is in volcanic regions, where explosive eruptions produce large amounts of unconsolidated material that gets washed into streams. The Coldwater River next to Mt. St. Helens in Washington State is a good example of this (Figure 13.20).

Figure 13.20 The braided Coldwater River, Mt. St. Helens, Washington. [SE]

A stream that occupies a wide, flat flood plain with a low gradient typically carries only sand-sized and finer sediments and develops a sinuous flow pattern. As you saw in Figure 13.14, when a stream flows around a corner, the water on the outside has farther to go and tends to flow faster. This leads to erosion of the banks on the outside of the curve, deposition on the inside, and formation of a point bar (Figure 13.21). Over time, the sinuosity of the stream becomes increasingly exaggerated, and the channel migrates around within its flood plain, forming a meandering pattern.

Figure 13.21 The meandering channel of the Bonnell Creek, Nanoose, B.C. The stream is flowing toward the viewer. The sand and gravel point bar must have formed when the creek was higher and the flow faster than it was when the photo was taken. [SE]

A well-developed meandering river is shown in Figure 13.22. The meander in the middle of the photo has reached the point where the thin neck of land between two parts of the channel is about to be eroded through. When this happens, another oxbow lake will form like the others in the photo.

Figure 13.22 The meandering channel of the Nowitna River, Alaska. Numerous oxbow lakes are present and another meander cutoff will soon take place. [Oliver Kumis, http://commons.wikimedia.org/wiki/File:Nowitna_river.jpg]

Exercises

Exercise 13.4 Determining Stream Gradients

Gradient is the key factor controlling stream velocity, and of course, velocity controls sediment erosion and deposition. This map shows the elevations of Priest Creek in the Kelowna area. The length of the creek between 1,600 m and 1,300 m elevation is 2.4 km, so the gradient is 300/2.4 = 125 m/km.

1. Use the scale bar to estimate the distance between 1,300 m and 600 m and then calculate that gradient.

2. Estimate the gradient between 600 and 400 m.

3. Estimate the gradient between 400 m on Priest Creek and the point where Mission Creek enters Okanagan Lake.

At the point where a stream enters a still body of water — a lake or the ocean — sediment is deposited and a delta forms. The Fraser River has created a large delta, which extends out into the Strait of Georgia (Figure 13.23). Much of the Fraser delta is very young in geological terms. Shortly after the end of the last glaciation (10,000 years ago), the delta did not extend past New Westminster. Since that time, all of the land that makes up Richmond, Delta, and parts of New Westminster and south Surrey has formed from sediment from the Fraser River. (You can see this in more detail at Geoscape Vancouver http://www.cgenarchive.org/vancouver-fraserdelta.html.)

Figure 13.23 The delta of the Fraser River and the plume of sediment that extends across the Strait of Georgia. The land outlined in red has formed over the past 10,000 years. [September 2011, SE after NASA: http://earthobservatory.nasa.gov/IOTD/view.php?id=77368]

The discharge levels of streams are highly variable depending on the time of year and on specific variations in the weather from one year to the next. In Canada, most streams show discharge variability similar to that of the Stikine River in northwestern B.C., as illustrated in Figure 13.24. The Stikine River has its lowest discharge levels in the depths of winter when freezing conditions persist throughout most of its drainage basin. Discharge starts to rise slowly in May, and then rises dramatically through the late spring and early summer as a winter’s worth of snow melts. For the year shown, the minimum discharge on the Stikine River was 56 m3/s in March, and the maximum was 37 times higher, 2,470 m3/s, in May.

Figure 13.24 Variations in discharge of the Stikine River during 2013. [SE from data at Water Survey of Canada, Environment Canada, http://www.ec.gc.ca/rhc-wsc/]

Streams in coastal areas of southern British Columbia show a very different pattern from those in most of the rest of the country because their drainage basins do not remain entirely frozen and because they receive a lot of rain (rather than snow) during the winter. The Qualicum River on Vancouver Island typically has its highest discharge levels in January or February and its lowest levels in late summer (Figure 13.25). In 2013, the minimum discharge was 1.6 m3/s, in August, and the maximum was 34 times higher, 53 m3/s, in March.

Figure 13.25 Variations in discharge of the Qualicum River during 2013. [SE from data at Water Survey of Canada, Environment Canada, http://www.ec.gc.ca/rhc-wsc/]

When a stream’s discharge increases, both the water level (stage) and the velocity increase as well. Rapidly flowing streams become muddy and large volumes of sediment are transported both in suspension and along the stream bed. In extreme situations, the water level reaches the top of the stream’s banks (the bank-full stage, see Figure 13.17), and if it rises any more, it floods the surrounding terrain. In the case of mature or old-age streams, this could include a vast area of relatively flat ground known as a flood plain, which is the area that is typically covered with water during a major flood. Because fine river sediments are deposited on flood plains, they are ideally suited for agriculture, and thus are typically occupied by farms and residences, and in many cases, by towns or cities. Such infrastructure is highly vulnerable to damage from flooding, and the people that live and work there are at risk.

Most streams in Canada have the greatest risk of flooding in the late spring and early summer when stream discharges rise in response to melting snow. In some cases, this is exacerbated by spring storms. In years when melting is especially fast and/or spring storms are particularly intense, flooding can be very severe.

One of the worst floods in Canadian history took place in the Fraser Valley in late May and early June of 1948. The early spring of that year had been cold, and a large snow pack in the interior was slow to melt. In mid-May, temperatures rose quickly and melting was accelerated by rainfall. Fraser River discharge levels rose rapidly over several days during late May, and the dykes built to protect the valley were breached in a dozen places. Approximately one-third of the flood plain was inundated and many homes and other buildings were destroyed, but there were no deaths. The Fraser River flood of 1948, which was the highest in the past century, was followed by very high river levels in 1950 and 1972 and by relatively high levels several times since then, the most recent being 2007 (Table 13.1). In the years following 1948, millions of dollars were spent repairing and raising the existing dykes and building new ones; since then damage from flooding in the Fraser Valley has been relatively limited.

Table13.1 Ranking of the maximum stage and discharge values for the Fraser River at Hope between 1948 and 2008. Typical discharge levels are around 1,000 m3/s. [From date in Mannerstrom, 2008Mannerström, M, 2008, Comprehensive Review of Fraser River at Hope Flood Hydrology and Flows Scoping Study, Report prepared for the B.C. Ministry of the Environment. Available at: http://www.env.gov.bc.ca/wsd/public_safety/flood/pdfs_word/review_fraser_flood_flows_hope.pdf]

Serious flooding happened in July in 1996 in the Saguenay-Lac St. Jean region of Quebec. In this case, the floods were caused by two weeks of heavy rainfall followed by one day of exceptional rainfall. July 19 saw 270 mm of rain, equivalent to the region’s normal rainfall for the entire month of July. Ten deaths were attributed to the Saguenay floods, and the economic toll was estimated at $1.5 billion.

Just a year after the Saguenay floods, the Red River in Minnesota, North Dakota, and Manitoba reached its highest level since 1826. As is typical for the Red River, the 1997 flooding was due to rapid snowmelt. Because of the south to north flow of the river, the flooding starts in Minnesota and North Dakota, where melting starts earlier, and builds toward the north. The residents of Manitoba had plenty of warning that the 1997 flood was coming because there was severe flooding at several locations on the U.S. side of the border.

After the 1950 Red River flood, the Manitoba government built a channel around the city of Winnipeg to reduce the potential of flooding in the city (Figure 13.26). Known as the Red River Floodway, the channel was completed in 1964 at a cost of $63 million. Since then it has been used many times to alleviate flooding in Winnipeg, and is estimated to have saved many billions of dollars in flood damage. The massive 1997 flood was almost too much for the floodway; in fact the amount of water diverted was greater than the designed capacity. The floodway has recently been expanded so that it can be used to divert more of the Red River’s flow away from Winnipeg.

Figure 13.26 Map of the Red River Floodway around Winnipeg, Manitoba (left), and aerial view of the southern (inlet) end of the floodway (right). [Map from http://en.wikipedia.org/wiki/1997_Red_River_Flood#/media/File:Rednorthfloodwaymap.png and photo from Natural Resources Canada 2012, courtesy of the Geological Survey of Canada (Photo 2000-118 by G.R. Brooks).]

Canada’s most costly flood ever was the June 2013 flood in southern Alberta. The flooding was initiated by snowmelt and worsened by heavy rains in the Rockies due to an anomalous flow of moist air from the Pacific and the Caribbean. At Canmore, rainfall amounts exceeded 200 mm in 36 hours, and at High River, 325 mm of rain fell in 48 hours.

Figure 13.27 Map of the communities most affected by the 2013 Alberta floods (in orange) [SE]

In late June and early July, the discharges of several rivers in the area, including the Bow River in Banff, Canmore, and Exshaw, the Bow and Elbow Rivers in Calgary, the Sheep River in Okotoks, and the Highwood River in High River, reached levels that were 5 to 10 times higher than normal for the time of year (see Exercise 13.5). Large areas of Calgary, Okotoks, and High River were flooded and five people died (see Figures 13.27 and 13.28). The cost of the 2013 flood is estimated to be approximately $5 billion. For more about Alberta’s flood of the century, visit: http://www.ec.gc.ca/meteo-weather/default.asp?lang=En&n=5BA5EAFC-1&offset=2&toc=hide.

Figure 13.28 Flooding in Calgary (June 21, left) and Okotoks (June 20, right) during the 2013 southern Alberta flood [http://upload.wikimedia.org/wikipedia/commons/6/6a/Riverfront_Ave_Calgary_Flood_2013.jpg http://upload.wikimedia.org/wikipedia/en/9/9b/Okotoks_-_June_20%2C_2013_-_Flood_waters_in_local_campground_playground-03.JPG]

Exercises

Exercise 13.5 Flood Probability on the Bow River

The graph below shows the highest discharge per year between 1915 and 2014 on the Bow River at Calgary. Using this data set, we can calculate the recurrence interval (Ri) for any particular flood magnitude using the equation: Ri = (n+1)/r (where n is the number of floods in the record being considered, and r is the rank of the particular flood). There are a few years missing in this record, and the actual number of data points is 95.

The largest flood recorded on the Bow River over that period was the one in 2013, 1,840 m³/s on June 21. Ri for that flood is (95+1)/1 = 96 years. The probability of such a flood in any future year is 1/Ri, which is 1%. The fifth largest flood was just a few years earlier in 2005, at 791 m³/s. Ri for that flood is (95+1)/5 = 19.2 years. The recurrence probability is 5%.

1. Calculate the recurrence interval for the second largest flood (1932, 1,520 m³/s).

2. What is the probability that a flood of 1,520 m³/s will happen next year?

3. Examine the 100-year trend for floods on the Bow River. If you ignore the major floods (the labelled ones), what is the general trend of peak discharges over that time?

[SE, from data at Water Surveys of Canada, Environment Canada, http://wateroffice.ec.gc.ca/search/searchDownload_e.html]

One of the things that the 2013 flood on the Bow River teaches us is that we can’t predict when a flood will occur or how big it will be, so in order to minimize damage and casualties we need to be prepared. Some of the ways of doing that are as follows:

• Mapping flood plains and not building within them
• Building dykes or dams where necessary
• Monitoring the winter snowpack, the weather, and stream discharges
• Creating emergency plans
• Educating the public

The topics covered in this chapter can be summarized as follows:

Introduction

Figure 14.1 A spring flowing from a limestone cave on Quadra Island, B.C. [SE]

As we saw in Chapter 13, fresh water makes up only 3% of the water on Earth. Approximately two-thirds of that is glacial ice and most of the rest is groundwater. We can’t live without water, and it’s easy to see that groundwater represents a critically important component of our water supply. Groundwater is not as easily accessed as surface water, but it is also not as easily contaminated as surface water. If more than 7 billion of us want to continue living comfortably here on Earth, we have to take great care of our groundwater and learn how to use it sustainably.

Groundwater is stored in the open spaces within rocks and within unconsolidated sediments. Rocks and sediments near the surface are under less pressure than those at significant depth and therefore tend to have more open space. For this reason, and because it’s expensive to drill deep wells, most of the groundwater that is accessed by individual users is within the first 100 m of the surface. Some municipal, agricultural, and industrial groundwater users get their water from greater depth, but deeper groundwater tends to be of lower quality than shallow groundwater, so there is a limit as to how deep we can go.

Porosity is the percentage of open space within an unconsolidated sediment or a rock. Primary porosity is represented by the spaces between grains in a sediment or sedimentary rock. Secondary porosity is porosity that has developed after the rock has formed. It can include fracture porosity — space within fractures in any kind of rock. Some volcanic rock has a special type of porosity related to vesicles, and some limestone has extra porosity related to cavities within fossils.

Porosity is expressed as a percentage calculated from the volume of open space in a rock compared with the total volume of rock. The typical ranges in porosity of a number of different geological materials are shown in Figure 14.2. Unconsolidated sediments tend to have higher porosity than consolidated ones because they have no cement, and most have not been strongly compressed. Finer-grained materials (e.g., silt and clay) tend to have greater porosity — some as high as 70% — than coarser materials (e.g., gravel). Primary porosity tends to be higher in well-sorted sediments compared to poorly sorted sediments, where there is a range of smaller particles to fill the spaces made by the larger particles. Glacial till, which has a wide range of grain sizes and is typically formed under compression beneath glacial ice, has relatively low porosity.

Consolidation and cementation during the process of lithification of unconsolidated sediments into sedimentary rocks reduces primary porosity. Sedimentary rocks generally have porosities in the range of 10% to 30%, some of which may be secondary (fracture) porosity. The grain size, sorting, compaction, and degree of cementation of the rocks all influence primary porosity. For example, poorly sorted and well-cemented sandstone and well-compressed mudstone can have very low porosity. Igneous or metamorphic rocks have the lowest primary porosity because they commonly form at depth and have interlocking crystals. Most of their porosity comes in the form of secondary porosity in fractures. Of the consolidated rocks, well-fractured volcanic rocks and limestone that has cavernous openings produced by dissolution have the highest potential porosity, while intrusive igneous and metamorphic rocks, which formed under great pressure, have the lowest.

Figure 14.2 Variations in porosity of unconsolidated materials (in red) and rocks (in blue) [SE]

Porosity is a measure of how much water can be stored in geological materials. Almost all rocks contain some porosity and therefore contain groundwater. Groundwater is found under your feet and everywhere on the planet. Considering that sedimentary rocks and unconsolidated sediments cover about 75% of the continental crust with an average thickness of a few hundred metres, and that they are likely to have around 20% porosity on average, it is easy to see that a huge volume of water can be stored in the ground.

Porosity is a description of how much space there could be to hold water under the ground, and permeability describes how those pores are shaped and interconnected. This determines how easy it is for water to flow from one pore to the next. Larger pores mean there is less friction between flowing water and the sides of the pores. Smaller pores mean more friction along pore walls, but also more twists and turns for the water to have to flow-through. A permeable material has a greater number of larger, well-connected pores spaces, whereas an impermeable material has fewer, smaller pores that are poorly connected. Permeability is the most important variable in groundwater. Permeability describes how easily water can flow through the rock or unconsolidated sediment and how easy it will be to extract the water for our purposes. The characteristic of permeability of a geological material is quantified by geoscientists and engineers using a number of different units, but the most common is the hydraulic conductivity. The symbol used for hydraulic conductivity is K. Although hydraulic conductivity can be expressed in a range of different units, in this book, we will always use m/s.

The materials in Figure 14.3 show that there is a wide range of permeability in geological materials from 10-12 m/s (0.000000000001 m/s) to around 1 m/s. Unconsolidated materials are generally more permeable than the corresponding rocks (compare sand with sandstone, for example), and the coarser materials are much more permeable than the finer ones. The least permeable rocks are unfractured intrusive igneous and metamorphic rocks, followed by unfractured mudstone, sandstone, and limestone. The permeability of sandstone can vary widely depending on the degree of sorting and the amount of cement that is present. Fractured igneous and metamorphic rocks, and especially fractured volcanic rocks, can be highly permeable, as can limestone that has been dissolved along fractures and bedding planes to create solutional openings.

Figure 14.3 Variations in hydraulic conductivity (in metres/second) of unconsolidated materials (in red) and of rocks (in blue) [SE]

Why is clay porous but not permeable?

Both sand and clay deposits (and sandstone and mudstone) are quite porous (30% to 50% for sand and 40% to 70% for silt and clay), but while sand can be quite permeable, clay and mudstone are not.

The surface of most silicate mineral grains has a slight negative charge due to imperfections in the mineral structure. Water (H2O) is a polar molecule. This means that while it has no overall electrical charge, one side of the molecule has a slight positive charge (the side with the two hydrogens), compared to a slight negative charge on the other side. Water is strongly attracted to all mineral grains and water within that bound water layer (a few microns around each grain) is not able to move and flow along with the rest of the groundwater. In the lower diagrams shown here, the bound water is represented by dark blue lines around each grain and the water that can move is light blue. In the sand, there is still a lot of water that is able to move through the sediment, but in the clay/silt almost all of the water is held tightly to the grains and this reduces the permeability. [SE]

We have now seen that there is a wide range of porosity in geological materials and an even wider range of permeability. Groundwater exists everywhere there is porosity. However, whether that groundwater is able to flow in significant quantities depends on the permeability. An aquifer is defined as a body of rock or unconsolidated sediment that has sufficient permeability to allow water to flow through it. Unconsolidated materials like gravel, sand, and even silt make relatively good aquifers, as do rocks like sandstone. Other rocks can be good aquifers if they are well fractured. An aquitard is a body that does not allow transmission of a significant amount of water, such as a clay, a till, or a poorly fractured igneous or metamorphic rock. These are relative terms, not absolute, and are usually defined based on someone’s desire to pump groundwater; what is an aquifer to someone who does not need a lot of water, may be an aquitard to someone else who does. An aquifer that is exposed at the ground surface is called an unconfined aquifer. An aquifer where there is a lower permeability material between the aquifer and the ground surface is known as a confined aquifer, and the aquitard separating ground surface and the aquifer is known as the confining layer.

Figure 14.4 shows a cross-section of a series of rocks and unconsolidated materials, some of which might serve as aquifers and others as aquitards or confining layers. The granite is much less permeable than the other materials, and so is an aquitard in this context. The yellow layer is very permeable and would make an ideal aquifer. The overlying grey layer is a confining layer.

The upper buff-coloured layer (K = 10-2 m/s) does not have a confining layer and is an unconfined aquifer. The yellow layer (K = 10-1 m/s) is “confined” by the confining layer (K = 10-4 m/s), and is a confined aquifer. The confined aquifer gets most of its water from the upper part of the hill where it is exposed at the surface, and relatively little by seepage through the fine silt layer.

Figure 14.4 A cross-section showing materials that might serve as aquifers and confining layers. The relative permeabilities are denoted by hydraulic conductivity (K = m/s). The pink rock is granite; the other layers are various sedimentary layers. [SE]

If you go out into your garden or into a forest or a park and start digging, you will find that the soil is moist (unless you’re in a desert), but it’s not saturated with water. This means that some of the pore space in the soil is occupied by water, and some of the pore space is occupied by air (unless you’re in a swamp). This is known as the unsaturated zone. If you could dig down far enough, you would get to the point where all of the pore spaces are 100% filled with water (saturated) and the bottom of your hole would fill up with water. The level of water in the hole represents the water table, which is the surface of the saturated zone. In most parts of British Columbia, the water table is several metres below the surface.

Water falling on the ground surface as precipitation (rain, snow, hail, fog, etc.) may flow off a hill slope directly to a stream in the form of runoff, or it may infiltrate the ground, where it is stored in the unsaturated zone. The water in the unsaturated zone may be used by plants (transpiration), evaporate from the soil (evaporation), or continue past the root zone and flow downward to the water table, where it recharges the groundwater.

A cross-section of a typical hillside with an unconfined aquifer is illustrated in Figure 14.5. In areas with topographic relief, the water table generally follows the land surface, but tends to come closer to surface in valleys, and intersects the surface where there are streams or lakes. The water table can be determined from the depth of water in a well that isn’t being pumped, although, as described below, that only applies if the well is within an unconfined aquifer. In this case, most of the hillside forms the recharge area, where water from precipitation flows downward through the unsaturated zone to reach the water table. The area at the stream or lake to which the groundwater is flowing is a discharge area.

What makes water flow from the recharge areas to the discharge areas? Recall that water is flowing in pores where there is friction, which means it takes work to move the water. There is also some friction between water molecules themselves, which is determined by the viscosity. Water has a low viscosity, but friction is still a factor. All flowing fluids are always losing energy to friction with their surroundings. Water will flow from areas with high energy to those with low energy. Recharge areas are at higher elevations, where the water has high gravitational energy. It was energy from the sun that evaporated the water into the atmosphere and lifted it up to the recharge area. The water loses this gravitational energy as it flows from the recharge area to the discharge area.

In Figure 14.5, the water table is sloping; that slope represents the change in gravitational potential energy of the water at the water table. The water table is higher under the recharge area (90 m) and lower at the discharge area (82 m). Imagine how much work it would be to lift water 8 m high in the air. That is the energy that was lost to friction as the groundwater flowed from the top of the hill to the stream.

Figure 14.5 A depiction of the water table in cross-section, with the saturated zone below and the unsaturated zone above. The water table is denoted with a small upside-down triangle. [SE]

The situation gets a lot more complicated in the case of confined aquifers, but they are important sources of water so we need to understand how they work. As shown in Figure 14.6, there is always a water table, and that applies even if the geological materials at the surface have very low permeability. Where there is a confined aquifer — meaning one that is separated from the surface by a confining layer — this aquifer will have its own “water table,” which is actually called a potentiometric surface, as it is a measure of the total potential energy of the water. The red dashed line in Figure 14.6 is the potentiometric surface for the confined aquifer, and it describes the total energy that water is under within the confined aquifer. If we drill a well into the unconfined aquifer, the water will rise to the level of the water table (well A in Figure 14.6). But if we drill a well through both the unconfined aquifer and the confining layer and into the confined aquifer, the water will rise above the top of the confined aquifer to the level of its potentiometric surface (well B in Figure 14.6). This is known as an artesian well, because the water rises above the top of the aquifer. In some situations, the potentiometric surface may be above the ground level. The water in a well drilled into the confined aquifer in this situation would rise above ground level, and flow out, if it’s not capped (well C in Figure 14.6). This is known as a flowing artesian well.

Figure 14.6 A depiction of the water table and the potentiometric surface of a confined aquifer. [SE]

In situations where there is an aquitard of limited extent, it is possible for a perched aquifer to exist as shown in Figure 14.7. Although perched aquifers may be good water sources at some times of the year, they tend to be relatively thin and small, and so can easily be depleted with over-pumping.

Figure 14.7 A perched aquifer above a regular unconfined aquifer. [SE]

In 1856, French engineer Henri Darcy carried out some experiments from which he derived a method for estimating the rate of groundwater flow based on the hydraulic gradient and the permeability of an aquifer, expressed using K, the hydraulic conductivity. Darcy’s equation, which has been used widely by hydrogeologists ever since, looks like this:

V = K * i

(where V is the velocity of the groundwater flow, K is the hydraulic conductivity, and i is the hydraulic gradient).

We can apply this equation to the scenario in Figure 14.5. If we assume that the permeability is 0.00001 m/s we get: V = 0.00001 * 0.08 = 0.0000008 m/s. That is equivalent to 0.000048 m/min, 0.0029 m/hour or 0.069 m/day. That means it would take 1,450 days (nearly four years) for water to travel the 100 m from the vicinity of the well to the stream. Groundwater moves slowly, and that is a reasonable amount of time for water to move that distance. In fact it would likely take longer than that, because it doesn’t travel in a straight line.

It’s critical to understand that groundwater does not flow in underground streams, nor does it form underground lakes. With the exception of karst areas, with caves in limestone, groundwater flows very slowly through granular sediments, or through solid rock that has fractures in it. Flow velocities of several centimetres per day are possible in significantly permeable sediments with significant hydraulic gradients. But in many cases, permeabilities are lower than the ones we’ve used as examples here, and in many areas, gradients are much lower. It is not uncommon for groundwater to flow at velocities of a few millimetres to a few centimetres per year. 

As already noted, groundwater does not flow in straight lines. It flows from areas of higher hydraulic head to areas of lower hydraulic head, and this means that it can flow “uphill” in many situations. This is illustrated in Figure 14.8. The dashed orange lines are equipotential,  meaning lines of equal pressure. The blue lines are the predicted groundwater flow paths. The dashed lines red lines are no-flow boundaries, meaning that water cannot flow across these lines. That’s not because there is something there to stop it, but because there’s no pressure gradient that will cause water to flow in that direction.

Groundwater flows at right angles to the equipotential lines in the same way that water flowing down a slope would flow at right angles to the contour lines. The stream in this scenario is the location with the lowest hydraulic potential, so the groundwater that flows to the lower parts of the aquifer has to flow upward to reach this location. It is forced upward by the pressure differences, for example, the difference between the 112 and 110 equipotential lines.

Figure 14.8 Predicted equipotential lines (orange) and groundwater flow paths (blue) in an unconfined aquifer. The orange numbers are the elevations of the water table at the locations shown, and therefore they represent the pressure along the equipotential lines. [SE]

Groundwater that flows through caves, including those in karst areas — where caves have been formed in limestone because of dissolution — behaves differently from groundwater in other situations. Caves above the water table are air-filled conduits, and the water that flows within these conduits is not under pressure; it responds only to gravity. In other words, it flows downhill along the gradient of the cave floor (Figure 14.9). Many limestone caves also extend below the water table and into the saturated zone. Here water behaves in a similar way to any other groundwater, and it flows according to the hydraulic gradient and Darcy’s law.

Figure 14.9 Groundwater in a limestone karst region. The water in the caves above the water table does not behave like true groundwater because its flow is not controlled by water pressure, only by gravity. The water below the water table does behave like true groundwater. [SE]

Except in areas where groundwater comes naturally to the surface at a spring (a place where the water table intersects the ground surface), we have to construct wells in order to extract it. If the water table is relatively close to the surface, a well can be dug by hand or with an excavator, but in most cases we need to use a drill to go down deep enough. There are many types of drills that can be used; an example is shown in Figure 14.10. A well has to be drilled at least as deep as the water table, but in fact must go much deeper; first, because the water table may change from season to season and from year to year, and second, because when water is being pumped, the water level will drop, at least temporarily.

Figure 14.10 A water-well drilling rig in operation in the Cassidy area, near Nanaimo, B.C. In the photo on the right the well is being test-pumped with air pressure. The casing (yellow arrow) is about 40 cm in diameter. [SE]

Where a well is drilled in unconsolidated sediments or relatively weak rock, it has to be lined with casing (steel pipe in most cases) in order to ensure that it doesn’t cave in. A specially designed well screen is installed at the bottom of the casing. The size of the holes in the screen is carefully chosen to make sure that it allows the water to move into the well freely, but prevents aquifer particles from entering the well. A submersible pump is typically used to lift water from within the well up to the where it is needed. The well shown in Figure 14.10 has casing that is about 40 cm in diameter, which might be typical for a municipal water supply well, or a very large well for irrigation. Most domestic wells have 15 cm casing.

Pumping water from the well removes water from inside the well at first. That lowers the water level inside the well. This means that water will flow from the surrounding aquifer (higher groundwater head) toward the pumping well where the groundwater head is now lower. That is how a well gets water from the ground. The water table, or potentiometric surface, will slope in toward the well where the water is being withdrawn. That indicates the energy gradient that is allowing water to flow toward the well. This creates a shape known as a cone of depression surrounding the well, as illustrated in Figure 14.11. If pumping from a well continues for hours to days, the cone of depression may result in a loss of water in nearby wells. As shown in Figure 14.12, pumping of well C has contributed to well B going dry. If pumping continues in well C, it too may go dry.

Figure 14.11 Three wells in an unconfined aquifer. Well A is not being pumped. Well B is being pumped at a slow rate and well C, which has a larger cone of depression, is being pumped at a faster rate. [SE]

Figure 14.12 A similar scenario to that in Figure 14.11, but in this case, wells B and C have been pumped unsustainably for a long time. The cone of depression from well C has reached well B and has contributed to it going dry. [SE]

Like other provinces in Canada, British Columbia has a network of observation wells administered by the Ministry of the Environment. These are wells that are installed to measure water levels; they are not pumped. There are 145 active observation wells in B.C. (in 2015), most equipped with automatic recorders that monitor water levels continuously. The main purpose of the observation wells is to monitor water table levels so that we can see if there are long-term natural fluctuations in groundwater quantity, and shorter-term fluctuations related to overuse of the resource. They are also sampled regularly to monitor groundwater chemistry and quality.

An example of an observation well is illustrated in Figure 14.13. This one is situated at Cassidy on Vancouver Island and is used to monitor an unconsolidated aquifer that is widely used by residents with private wells.

Figure 14.13 B.C. observation well 232 near Cassidy Airport, Vancouver Island. The installation also has a solar panel, which is not visible in this view. [SE]

The water-level data from B.C.’s observation wells are available to the public, and an example data set is illustrated in Figure 14.14. The water level in Ministry of Environment observation well 232 (OW-232), situated in Lantzville on Vancouver Island, dropped significantly from 1979 (average depth ~1.5 m), to 2010 (average depth ~5.5 m), but has recovered a little since then.

Figure 14.14 Water level data for B.C. observation well 232 on Harby Rd., Lantzville, Vancouver Island. From 1979 to 2003, depths were recorded monthly. Automated equipment was installed in 2003, and the depths were recorded hourly since that time. [SE from data at: http://www.env.gov.bc.ca/wsd/data_searches/obswell/map/]

The short-term variations in the level of well 232 are at a period of one year and are related to annual cycles of recharge and discharge governed by the wet winter climate and drier summers. The data for part of the period are shown in more detail in Figure 14.15. On Vancouver Island, most wells drop to their lowest levels in September or October after the long dry summer period. Levels increase rapidly from October through February as high winter precipitation adds recharge to the aquifer, and water is stored. The water table reaches a peak in March or April. Most wells then drop over the summer as groundwater continues to flow, but no new recharge is added. The water is drained from storage into streams or lakes and eventually into the ocean, and as a result, the water table decreases, reaching its lowest level again in September or October. Similar fluctuations are observed at most observations wells around the province, although the timing is slightly different from region to region.

Figure 14.15 Water level data for B.C. observation well 232 for the period 1996 to 2000, showing seasonal variations [SE from data at: http://www.env.gov.bc.ca/wsd/data_searches/obswell/map/]

The long-term fluctuations in levels in observation wells around the province are also quite variable. Long-term changes in climate can lead to gradual natural changes in water levels. These long-term cycles lasting years or decades are mixed with the effects of well pumping. Some observation wells show consistent decreases in water level that may indicate long-term over-extraction. Many others show generally consistent levels over several decades, and some show increases in water level. One of the important jobs performed by hydrogeologists working for different government ministries is to examine these long-term records of water levels for indications of how sustainable the groundwater use might be.

In 2014, the B.C. government introduced the Water Sustainability Act, which will require licensing groundwater extraction for the first time. This comes into effect in January 2016. The new Act also includes provisions for determining “environmental flow needs” — the amount of water that must be in surface water streams at different times of the year to meet the needs of the ecosystem that depends on the streamflow. For example, many streams in B.C. support populations of salmon that live in the stream for part of their life cycle or return to their home stream for spawning. Groundwater forms a part of the baseflow in a watershed, and is therefore an important part of the environmental flow needs. Careful work is needed in the coming years to ensure that the amount of water licensed to be extracted from surface water and groundwater for human use does not interfere with the amount of water needed for the natural water-dependent ecosystems to function.

The situation in California, where groundwater extraction over large areas is leading to declining water levels, is quite different from that in B.C. According to the state Department of Water Resources, 80% of groundwater wells showed drops in water level of 0 m to 7.5 m between 2011 and 2013, another 6% dropped by 7.5 m to 15 m, and 3% dropped by more than 15 m (Figure 14.16). Over the same time period, only 10% of well levels increased by 0 m to 7.5 m, and 1% increased by more than 7.5 m. The drought that gripped California in 2013 had worsened significantly by 2015, and California farmers — and the people across North America that eat the food they produce — continue to have a prodigious appetite for irrigation water. California, like B.C., is introducing new groundwater regulations to try to control water usage and halt water table declines.

Figure 14.16 Changes in water levels in wells in California from 2011 to 2013 [SE from State of California Department of Water Resources data]

Impermeable Surfaces

Even if groundwater supplies are not being depleted by overuse, or by a changing climate, we are continuing to put stress on aquifers by covering vast areas with impermeable surfaces that don’t allow rain and snowmelt to infiltrate and become groundwater. Instead, water that falls on these surfaces is channelled into drainage systems, then into storm sewers, and then directly into rivers and the ocean. In cities and their suburbs, the biggest culprits are parking lots, roads, and highways. While it would great if we didn’t dedicate such huge areas to cars, that’s not about to change quickly, so we need to think about ways that we can improve surface water infiltration in cities. One way is to use road and parking surfaces that will allow water to seep through, although this is not practical in many cases. Another way is to ensure that runoff from pavement is channelled into existing or constructed wetlands that serve to decontaminate the water, and then allow it to infiltrate into the ground.

As was noted at the very beginning of this chapter, one of the good things about groundwater as a source of water is that it is not as easily contaminated as surface water is. But there are two caveats to that: one is that groundwater can become naturally contaminated because of its very close connection to the materials of its aquifer, and the second is that once contaminated by human activities, groundwater is very difficult to clean up.

Natural Contamination of Groundwater

Groundwater moves slowly through an aquifer, and unlike the surface water of a stream, it has a lot of contact with the surrounding rock or sediment. In most aquifers, the geological materials that make up the aquifer are relatively inert, or are made up of minerals that dissolve very slowly into the groundwater. Over time, however, all groundwater gradually has more and more material dissolved within it as it remains in contact with the aquifer. In some areas, that rock or sediment includes some minerals that could potentially contaminate the water with elements that might make the water less than ideal for human consumption or agricultural use. Examples include copper, arsenic, mercury, fluorine, sodium, and boron. In some cases, contamination may occur because the aquifer material has particularly high levels of the element in question. In other cases, the aquifer material is just normal rock or sediment, but some particular feature of the water or the aquifer allows the contaminant to build up to significant levels.

An example of natural contamination takes place in the bedrock aquifers of the east coast of Vancouver Island and the adjacent Gulf Islands. The aquifer is the Cretaceous (90 Ma to 65 Ma) Nanaimo Group, which is made up of sandstone, mudstone, and conglomerate (Figure 14.17).

Figure 14.17 Cretaceous Nanaimo Group sandstone exposed in a Nanaimo parking lot [SE]

The rocks of the Nanaimo Group are not particularly enriched in any trace elements, but the submarine-fan sandstone that makes up much of the group is a lithic wacke, and therefore has relatively high levels of clay (for a sandstone). This clay is good at adsorbing“Adsorb” (with a “d”) is not the same as “absorb” (with a “b”). Water can be absorbed by a sponge. Ions dissolved in water can be adsorbed onto — or desorbed from — the surfaces of clay minerals. some elements from the water and desorbing others, and in the process, its pH goes up (it becomes alkaline). At high pH levels (some as high as 9 in the Nanaimo Group), the element fluorine that is present naturally in the rock (as it is in almost any rock) has an increased tendency to dissolve in the water. In some areas, groundwater in the Nanaimo Group has fluorine levels that are well above recommended levels for drinking water. The World Health Organization (WHO) maximum acceptable concentration (MAC) for fluorine is 1.5 mg/L (milligrams per litre). Between 5% and 10% of the domestic wells around Nanaimo and adjacent Gabriola Island have more than that, some as much as 10 mg/L. A small amount of fluorine in the human diet is considered important for maintaining dental health, but high levels can lead to malformation and discolouration of teeth, and long-term exposure can lead to other more serious health effects such as skeletal problems.

Nanaimo Group groundwater can also have elevated levels of boron, again related to pH and adsorption from clay minerals. While boron at the levels found there is not toxic to humans, there is enough boron in some wells to be toxic to plants, and the water cannot be used for irrigation.

Rural residents in the densely populated country of Bangladesh (over 1,000 residents/km2, compared with 3.4/km2 in Canada) used to rely mostly on surface supplies for their drinking water, and many of these were subject to bacterial contamination. Infant mortality rates were among the highest in the world and other illnesses such as diarrhea, dysentery, typhoid, cholera, and hepatitis were common. In the 1970s, international agencies, including UNICEF, started a program of drilling wells to access abundant groundwater supplies at depths of 20 m to 100 m. Eventually over 8 million such wells were drilled. Infant mortality and illness rates dropped dramatically, but it was later discovered that the water from a high proportion of these wells has arsenic above safe levels (Figure 14.18).

Figure 14.18 The distribution of arsenic in groundwater in Bangladesh. The WHO recommended safe level for arsenic is 10 μg/L. All of the green, orange, and red areas on the map exceed that limit.
[From: BGS and DPHE. 2001. Arsenic contamination of groundwater in Bangladesh. Kinniburgh, D G and Smedley, P L (Editors). British Geological Survey Technical Report WC/00/19. British Geological Survey: Keyworth. (http://www.bgs.ac.uk/arsenic/bangladesh/) ]

Most of the wells in the affected areas are drilled into relatively recent sediments of the vast delta of the Ganges and Brahmaputra Rivers. While these sediments are not particularly enriched in arsenic, they have enough organic matter in them to use up any oxygen present. This leads to water with a naturally low oxidation potential (anoxic conditions); arsenic is highly soluble under these conditions, and so any arsenic present in the sediments easily gets dissolved into the groundwater. Arsenic poisoning leads to headaches, confusion, and diarrhea, and eventually to vomiting, stomach pain, and convulsions. If not treated, the final outcomes are heart disease, stroke, cancer, diabetes, coma, and death. There are ways to treat arsenic-rich groundwater, but it is a challenge in Bangladesh to implement the simple and effective technology that is available.

Anthropogenic Contamination of Groundwater

Groundwater can become contaminated by pollution at the surface (or at depth), and there are many different anthropogenic (human-caused) sources of contamination.

The vulnerability of aquifers to pollution depends on several factors, including the depth to the water table, the permeability of the material between the surface and the aquifer, the permeability of the aquifer, the slope of the surface, and the amount of precipitation. Confined aquifers tend to be much less vulnerable than unconfined ones, and deeper aquifers are less vulnerable than shallow ones. Steeper slopes mean that surface water tends to run off rather than infiltrate (and this can reduce the possibility of contamination). Contamination risk is also less in dry areas than in areas with heavy rainfall.

Studies of groundwater vulnerability have been completed for various regions of British Columbia. A groundwater vulnerability map for southern Vancouver Island is shown in Figure 14.19. The yellow to red areas are considered to have high vulnerability to pollution from surface sources, and most of these are where the aquifers are unconfined in quite permeable unconsolidated sediments of either glacial or fluvial origin, where the water table is relatively shallow and the terrain is relatively flat.

Figure 14.19 The vulnerability to anthropogenic contamination of aquifers on southern Vancouver Island. Much of the island is not mapped (shown as white) because of a lack of aquifer information in areas without wells. [From: Newton, P. and Gilchrist, A. 2010. Technical summary of intrinsic vulnerability mapping methods of Vancouver Island, Vancouver Island Water Resources Vulnerability Mapping Project, Vancouver Island University, 45pp. Used with permission. https://web.viu.ca/groundwater/PDF/VI_DRASTIC_Summary_Phase2_2010.pdf]

The important sources of anthropogenic groundwater contamination include the following:

• Chemicals and animal waste related to agriculture, and chemicals applied to golf courses and domestic gardens
• Landfills
• Industrial operations
• Mines, quarries, and other rock excavations
• Leaking fuel storage tanks (especially those at gas stations)
• Septic systems
• Runoff from roads (e.g., winter salting) or chemical spills of materials being transported

Agriculture

Intensive agricultural operations and golf courses can have a significant impact on the environment, especially where chemicals and other materials are used to enhance growth or control pests. An example of agricultural contamination is in the Abbotsford area of the Fraser Valley, where nitrate levels above the 44 mg/L maximum acceptable level (expressed as nitrate) in the Abbotsford-Sumas aquifer have been observed since the 1950s; however, the problem became much worse as agriculture intensity increased in the 1980s. By 2004, groundwater with nitrate levels in excess of 44 mg/L was reported over an area of about 75 km2 around Abbotsford, and the problem extended across the border into the Sumas area of Washington State.

This region is intensively used for berry crops (especially raspberries and blueberries) and large poultry operations, as well as lesser amounts of grazing and forage crops. Chicken manure is typically stored in fields adjacent to chicken barns, and may release nitrogen to the environment from runoff water, and from releases of ammonia gas. Over decades, both chemical fertilizers and chicken manure and other manures have been applied to the berry crops to provide extra nitrogen to help maximize berry growth. If the fertilizer added is in excess of what the plants need, or is poorly timed compared to when it is needed, then the extra nitrogen may be leached into the groundwater below. Berry crops are irrigated over the summer to help the crops grow. Summer irrigation and winter rainfall may carry excess nitrate from the near surface to the aquifer below.

Since the 1990s, agricultural practices have been tightened up to reduce the rate of groundwater contamination, but it will take decades for nitrate levels to drop in the Abbotsford-Sumas aquifer. Agriculture and Agri-Food Canada and many others are conducting research on better irrigation and nitrate management techniques to reduce the amount of nitrogen that leaches to groundwater.

Landfills

In the past, domestic and commercial refuse was commonly trucked to a “dump” (typically a hole in the ground), and when the hole was filled, it was covered with soil and forgotten. In situations like this, rain and melting snow can easily pass through the soil used to cover the refuse. This water passes into the waste itself, and the resulting landfill leachate that flows from the bottom of the landfill can seriously contaminate the surrounding groundwater and surface water. In the past few decades, regulations around refuse disposal have been significantly strengthened, and important steps have been taken to reduce the amount of landfill waste by diverting recyclable and compostable materials to other locations.

A modern engineered landfill has an impermeable liner (typically heavy plastic, although engineered clay liners or natural clay may be adequate in some cases), a plumbing system for draining leachate (the rainwater that flows through the refuse and becomes contaminated), and a network of monitoring wells both within and around the landfill (Figure 14.20). Once part or all of a landfill site is full, it is sealed over with a plastic cover, and a system is put in place to extract landfill gas (typically a mixture of carbon dioxide and methane). That gas can be sent to a nearby location where it is burned to create heat or used to generate electricity. The leachate must be treated, and that can be done in a normal sewage treatment plant.

Figure 14.20 A cross-section of a typical modern landfill [SE]

The monitoring wells are used to assess the level of the water table around the landfill and to collect groundwater samples so that any leakage can be detected. Because some leakage is almost inevitable, the ideal placement for landfills is in areas where the depth to the water table is significant (tens of metres if possible) and where the aquifer material is relatively impermeable. Landfills should also be situated far from streams, lakes, or wetlands so that contamination of aquatic habitats can be avoided.

Today there are hundreds of abandoned dumps scattered across the country; most have been left to contaminate groundwater that we might wish to use sometime in the future. In many cases, it’s unlikely that we’ll be able to do so.

Industrial Operations

Although western Canada doesn’t have the same extent of industrial pollution as other parts of the country, there are still seriously contaminated sites in the west, most with the potential to contaminate groundwater. One example is the lead and zinc smelter at Trail, B.C. The largest in the world, it has been operating for over 100 years and has left a residue of metal contamination around the region (Figure 14.21). In some parts of Trail, the contamination is serious enough that existing soil has been removed from residential properties and replaced with clean soil brought in from elsewhere. This contaminated soil has contributed to contamination of groundwater in the Trail area. Groundwater beneath the actual smelter site is contaminated, and the operator (Teck Resources) is currently working on plans to prevent that water from reaching the nearby Columbia River.

Figure 14.21 The Trail lead-zinc smelter in 1929
[http://upload.wikimedia.org /wikipedia/commons/2/20 /Trail_Smelter_in_Year_1929.png]

Mines, Quarries, and Rock Excavations

Mines and other operations that involve the excavation of large amounts of rock (e.g., highway construction) have the potential to create serious environmental damage. The exposure of rock that has previously not been exposed to air and water can lead to the oxidation of sulphide-bearing minerals, such a pyrite (FeS2), within the rock. The combination of pyrite, water, oxygen, and a special type of bacteria (Acidithiobacillus ferrooxidans) that thrives in acidic conditions leads to the generation of acidity, in some cases to pH less than 2. Water that acidic is hazardous by itself, but the low pH also has the property of increasing the solubility of certain heavy metals. The water that is generated by this process is known as acid rock drainage (ARD). ARD can occur naturally where sulphide-bearing rocks are near the surface. The issue of ARD is a major environmental concern at both operating mines and abandoned mines (see Chapter 20). In streams around the Mt. Washington Mine on Vancouver Island (Figure 14.22), copper levels are high enough to be toxic to fish. Groundwater adjacent to the contaminated streams in the area is very likely contaminated as well.

Figure 14.22 Acidic runoff at the abandoned Mt. Washington Mine near Courtenay, B.C. [SE]

Leaking Fuel Tanks

Underground storage tanks (USTs) are used to store fuel at gas stations, industrial sites, airports, and anywhere that large volumes of fuel are used. They do not last forever, and eventually they start to leak their contents into the ground. This is a particular problem at older gas stations — although it may also become a future problem at newer gas stations. You may have noticed gas stations that have been closed and then surrounded by chain-link fence (Figure 14.23). In virtually all such cases the closure has been triggered by the discovery of leaking USTs and the requirement to cease operations and remediate the site.

Figure 14.23 A closed and fenced gas station site in Nanaimo, B.C. The white pipes in the background are wells for monitoring groundwater contamination on the site. [SE]

Petroleum fuels are complex mixtures of hydrocarbon compounds and the properties of their components — such as density, viscosity, solubility in water, and volatility — tend to vary widely. As a result, a petroleum spill is like several spills for the price of one. The petroleum liquid slowly settles through the unsaturated zone and then tends to float on the surface of the groundwater (Figure 14.24). The more readily soluble components of the spill dissolve in the groundwater and are dispersed along with the normal groundwater flow, and the more volatile components of the spill rise toward the surface, potentially contaminating buildings.

Figure 14.24 A depiction of the fate of different components of a petroleum spill from an underground storage tank. [SE]

Septic Systems 

In areas that are not served by sewage networks leading to a central sewage treatment plant, most homeowners rely on septic systems for disposal of sewage. There are two primary components to a simple septic system, the septic tank and the drainage field (Figure 14.25). A typical septic tank is constructed of either concrete or plastic and has a volume of 5,000 L to 10,000 L (5 m3 to 10 m3). This forms the first treatment and is designed to be anaerobic (without oxygen). That promotes the activity of certain bacteria that help break down the waste. As the waste is degraded, some portions tend to sink to form sludge at the base of the tank, and others float to the surface, forming a scum layer. A septic tank may be divided into two parts to keep the sludge at the bottom and the scum on the top from draining out. The water then moves to the drainage field, which provides the right conditions for a different set of bacteria that operate in aerobic conditions. The drainage field includes an array of plastic pipes that are perforated to allow the effluent to drain out over a large area and seep slowly into the ground. In order to install a drainage field, it is first necessary to test the soil below, as it must be sufficiently permeable to allow the effluent to percolate away, but not so permeable that it flows too quickly and the soil is not able to filter out the pathogenic bacteria.

If they are properly installed and used, and if the sludge is periodically removed from the tank, a septic system should be effective in treating the sewage for decades. The anaerobic and aerobic bacteria should be able to break down the incoming waste and there should be little risk to the surface environment or groundwater. But many things can go wrong with a septic system, including the following:

• If inappropriate chemicals are added to the waste stream, they may interfere with the natural breakdown of the sewage.
• If the tank is not periodically pumped out, solids can get into the drainage field and compromise the drainage, resulting in the flow of effluent toward the surface.
• If the soil is either not sufficiently permeable or too permeable, the effluent will not drain away (and will start to pool at the surface) or it will drain too quickly.
• If the drainage field is constructed in an area where the water table is close to surface, some of the effluent is likely to flow into the groundwater without being treated.

Figure 14.25 A typical septic system. [SE]

Prevention and Mitigation of Groundwater Contamination

As illustrated in the landfill example above, there are two fairly simple ways to significantly reduce the chance and degree of groundwater contamination from surface sources. One is to prevent rainwater from infiltrating down to the water table and picking up contaminants; this can be achieved by simply capping or roofing over the landfill, mine tailings, or spill site. The second is to provide an impermeable barrier beneath the contaminant. Modern landfills and mine tailings impoundments are all built using some combination of clay and engineered plastic barriers. Both of these solutions — caps and liners — are subject to failure due to leaks.

Once contaminants are in the groundwater, the main form of remediation is to pump out the contaminated water and treat it at the surface. This can be a slow process, and preventing the contaminant from travelling significantly during this process can be accomplished by manipulating local groundwater flow through the extraction or injection of water at certain locations. Consider this in the exercise below.