2 Interior Structure of EARTH

Download as pdf or txt
Download as pdf or txt
You are on page 1of 6

2 THE EARTH’S INTERIOR

As mentioned above, virtually all igneous rocks originate by melting of material at


some depth within the Earth. All terrestrial rocks that we now find at the Earth’s
surface were derived initially from the mantle, although some have since gone
through one or more cycles of subsequent sedimentary, metamorphic, and/or
igneous processes. If these rocks have an ultimate origin at depth, it follows that
we need to know what makes up the Earth if we want to understand their origin
more fully. From a compositional perspective, the Earth’s interior is subdivided
into three major units:
1. The crust,
2. The mantle, and
3. The core (Figure 1).
These units were recognized decades ago, during the early days of seismology,
because they were separated by major discontinuities in the velocities of P
(compressional) and S (shear) waves as they propagate through those layers in the
Earth (Figure 2).

The crust comprises about 1% of the volume of the Earth. There are two basic
types of crust—oceanic and continental— and both are too thin to represent
accurately on Figure 1 (even the thickest continental crust would be thinner
than the uppermost line).

The thinner of the two, the oceanic crust (about 10 km thick), has an essentially
basaltic composition. Because plate tectonics is creating oceanic crust at mid-ocean
ridges and consuming it at subduction zones, the oceanic crust is continually being
renewed and recycled. The oldest oceanic crust is in the southwest Pacific and is
about 160 Ma old. The continental crust is thicker: typically 30 to 45 km beneath
stable areas but generally 50 to 60 km thick in orogenic areas and extending up to
90 km at a few localities. It is also more heterogeneous, including all sorts of
sedimentary, igneous, and metamorphic rocks.

A very crude average composition of the continental crust would be represented by


a granodiorite. Continental crust covers about 40% of the Earth’s surface. Unlike
the oceanic crust, it is too buoyant to subduct far. The amount of continental crust
has been increasing over the past 4 Ga. Some continental crust is thus very old,
whereas some is quite new.

The stable continents (cratons) consist of more ancient crystalline shields and
stable platforms. Platforms typically have basement crystalline rocks (typically
younger than shields, but not necessarily so) overlain by a few kilometers of
sedimentary rocks.

Several marginal orogenic belts may also flank cratons, typically associated with
subduction, which add to the continental crust over time. The lower crust is
believed to be depleted in many of the more mobile elements and to have a more
mafic character than the upper crust. A seismic discontinuity in the lower
continental crust (the Conrad discontinuity) is recognized in some areas and may
mark the transition between this deeper crust and the shallower sialic portion. It
was once believed that the Conrad discontinuity was continuous and separated a
more granitic shallow crust from a basaltic lower crust (similar to the oceanic
crust) that formed the continental base, but we now realize that the upper and lower
continental crust is much more heterogeneous. The base of the continents is not
always sharply defined seismically and may locally be gradual and even have a
layered transition into the sub- continental mantle. There will be lots more to say
about the nature of the continental crust and the igneous and metamorphic
processes associated with it throughout this text.

For a good summary of continental structure, see Sleep (2005), and for the lower
portion, see Fountain et al. (1992).

FIGURE 2 Variation in P and S wave velocities with depth. Compositional subdivisions of the
Earth are on the left, and rheological subdivisions are on the right. After Kearey and Vine (1990)
2. Mantle

Immediately beneath the crust, and extending to nearly 3000 km, is the mantle,
comprising about 83% of the Earth’s volume. The boundary, or discontinuity,
between the crust and mantle is called the Moho, or M discontinuity (shortened
from Mohorovici´c, the name of the Balkan seismologist who discovered it in
1909). At this discontinuity, most readily observed beneath oceanic crust, the
velocity of P waves increases abruptly from about 7 to over 8 km/sec.
This results in refraction, as well as reflection, of seismic waves as they encounter
the discontinuity, making it relatively simple to determine the depth. The mantle is
composed predominantly of Fe- and Mg-rich silicate minerals.

Within the mantle, several other seismic discontinuities separate layers that are
distinguished more by physical than chemical differences. The shallowest such
layer, between 80 and 220 km, is called the low-velocity layer because within it,
seismic waves slow down slightly, as compared to the velocity both above and
below the layer (Figure 2). The slowness of seismic waves is unusual because
seismic velocities generally increase with depth because they propagate more
readily through more compacted (hence more incompressible and
rigid) materials.

The reason seismic waves slow down in the low velocity layer is attributed to 1 to
5% partial melting of the mantle. The melt probably forms a thin discontinuous
film along mineral grain boundaries, which retards the seismic waves. The melt
also weakens the mantle in this layer, making it behave in a more ductile fashion.
The low-velocity layer varies in thickness, depending on the local pressure,
temperature, melting point, and availability of H2O.

Below the low-velocity layer we encounter two more seismic discontinuities


within the mantle. The 410-km discontinuity is believed to result from a phase
transformation in which olivine (the major mineral constituent of the upper mantle)
changes from the well-known (“α-phase”) structure to wadsleyite (“β-phase”) and
then to ringwoodite (“γ-phase”) with an isometric spinel-type structure. At 660 km,
the coordination of Si in mantle silicates changes from the familiar IV-fold to VI-
fold, and the dominant silicate becomes an (Mg,Fe)SiO3 magnesium silicate with a
perovskite-like structure, and the excess Mg and Fe form an (Mg,Fe)O oxide called
either magnesiowüstite or ferropericlase. This latter transition, of course, is not a
simple a →b phase transformation, but an a →b + c reaction. Both the 410-km and
660-km transitions result in an abrupt increase in the density of the mantle,
accompanied by a jump in seismic velocities. Below the 660-km discontinuity, the
velocities of seismic waves increase fairly uniformly with depth (Figure 2).

At the very base of the mantle is a ~200 km thick heterogeneous layer of


anomalously low seismic velocity called the D" layer. A thin (~40km), apparently
discontinuous layer with even lower velocities has also recently been resolved at
the mantle–core boundary, most clearly beneath the central Pacific (Garnero and
Helmberger, 1995, 1996). The nature of the D" and 40-km sublayer are not entirely
clear, but their properties are sufficiently anomalous to require more than a thermal
boundary perturbation, and they probably represent a layer of different
composition (and hence greater density) than the overlying mantle. A popular
proposition is that they represent an accumulation of dense “dregs” of subducted
oceanic crust that has settled to the base of the mantle (Christensen and Hofmann,
1994).

3 Core

Beneath the mantle is the core. The mantle–core boundary is a profound chemical
discontinuity at which the silicates of the mantle give way to a much denser Fe rich
metallic alloy with minor amounts of Ni, S, Si, O, etc.

The outer core is in the liquid/molten state, whereas the inner core is solid.
The composition of the inner core and outer core is probably similar. The transition
to a solid is a response to increased pressure with depth, which favors the solid
state. S-waves cannot propagate through a liquid because liquids cannot resist
shear.

Although S-waves are only slowed by the thin liquid films in the low-velocity
layer, they disappear entirely as they reach the outer core (Figure 2). P-waves slow
in the liquid core and refract downward, resulting in the seismic “shadow zone,” a
ringlike zone in which earthquake P-waves don’t reach the surface of the Earth on
the side away from which they originated. The two types of crust, the mantle, and
the core are distinguished on the basis of composition.

An alternative way to consider the subdivisions of the Earth is based on


rheological properties (right side of Figure 2). Using these criteria (how materials
respond to deformation), we can consider the crust plus the more rigid portion of
the uppermost mantle above the low-velocity layer to behave as a strong, coherent
unit, collectively called the lithosphere. Oceanic lithosphere is thin (~50 km) near
warm mid-ocean ridges and thickens to about 110 km when cool and mature. The
lithosphere is 200 to 250 km thick under the stable continental shields (McKenzie
and Priestly, 2008).

The more ductile mantle immediately below the lithosphere is called the
asthenosphere (from the Greek asthenes, “without strength”). The asthenosphere
is important to plate tectonics because the ductility is thought to provide the zone
of dislocation upon which the rigid lithospheric plates may differentially move.

From a rheological standpoint, the mantle below the asthenosphere is called the
mesosphere. The lithosphere– asthenosphere–mesosphere boundaries are all
within the mantle and correspond to the transition from rigid to ductile and back to
less ductile material with depth. The transitions are somewhat gradual and difficult
to resolve seismically, particularly the bottom of the ductile layer.

The asthenosphere is probably about 150 km thick. The rheological nature of the
mesosphere is not well known, but seismic waves are not greatly attenuated,
suggesting that this layer is relatively strong. The liquid outer core and solid inner
core are of course distinguishable on a mechanical basis.

You might also like