Beruflich Dokumente
Kultur Dokumente
When Galileo let his balls run down it precisely a sphere or even an ellipsoid of rev-
an inclined plane with a gravity which olution. Although mountains, ocean basins and
variations in crustal thickness contribute to the
he had chosen himself ... then a light observed irregular shape and gravity field of the
dawned upon all natural philosophers. Earth, they cannot explain the long-wavelength
f. Kant departures from a hydrostatic figure.
The distribution of elevations on the Earth is
Terrestrial planets are almost spherical because distinctly bimodal, with a peak near + 0.1 krn
of gravity and the weakness of rock in large representing the mean elevation of continents
masses . The largest departures from sphericity and a peak near -4.7 krn corresponding to the
are due to rotation and variations in buoyancy mean depth of the oceans lsee Coogle Images
of the surface and interior shells. Otherwise, h ypsorne try ]. This bimodal character contrasts
the overall shape of the Earth and its heat flow with that of the other terrestrial planets. The
are manifestations of convection in the interior spherical harmonic spectrum of the Earth's
and conductive cooling of the outer layers . The topography shows a strong peak for l = 1, cor-
style of convection is uncertain. There are var- responding to the distribution of continents in
ious hypotheses in this field that parallel those one hemisphere, and a regular decrease with
in petrology and geochemistry. The end-members increasing n. The topography spectrum is simi-
are whole-mantle convection in a chemically lar to that of the other terrestrial planets. There
uniform mantle, layered convection with inter- are small peaks in the spectrum at l = 3 and
change and overturns, and irreversible chemi- I = 9 - 10, the latter corresponding to the dis-
cal stratification with little or no interchange of tribution of subduction zones and large oceanic
material between layers. Layered schemes have swells.
several variants involving a primitive lower man- The wavelength , in kilometers, is related to
tle or a depleted (in U and Th) lower mantle. the spherical harmonic degree I and the circum-
In a convecting Earth we lose all of our refer- ference of the Earth (in km) approximately by
ence systems. The mantle is heated from within,
cooled from above and experiences secular cool- Wavelength = 40 040/ (l + 0.5)
ing. Global topography and gravity provide
Thus, a wavelength of 10 degrees or 1100 Ian
constraints on mantle dynamics .
corresponds to a spherical harmonic degree of
about 40.
Topography Active orogenic belts such as the Alpine and
Himalayan are associated with thick crust, and
Although the Earth is not flat or egg-shaped, as high relief, up to 5 krn. Older orogenic belts such
previously believed at various times, neither is as the Appalachian and Caledonian, because of
TOPOGRAPHY 63
erosion and lower crustal delamination, are asso- 0 to 70 Myr, topography are described by
ciated with low relief, less than 1 km, and thin-
d(t) = 2500 + 350t112
ner crusts. Regional changes in the topography
of the continents are generally accompanied by where t is crustal age in Myr and d(t) is the
changes in mean crustal thickness . Continents depth in meters . Older seafloor does not follow
stand high because of thick, low-density crust, this simple relationship, being shallower than
compared with oceans. TI1ere is a sharp cut-off predicted, and there is much scatter at all ages .
in crustal thickness at about 50 km, probably Slightly different relations hold if the seafloor
due to delamination of over-thickened crust at is subdivided into tectonic corridors. There are
the gabbro-eclogite phase change boundary. As large portions of the ocean floor where depth
the dense root grows, the surface subsides, form- cannot be explained by simple thermal models;
ing sedimentary basins. Upon delamination, the these include oceanic islands, swells, aseismic
surface pops up, forming a swell, often accom- ridges and oceanic plateaus as well as other areas
panied by magmatism. Many continental flood where the effects of surface tectonics and crustal
basalt provinces (CFB) erupt on top of sedimen- structure are not readily apparent. Simple cool-
tary basins and the underlying crust is thinner ing models assume that the underlying mantle
than average for the continents. is uniform and isothermal and that all of the
The long-wavelength topography of the ocean variation in bathymetry is due to cooling of a
floor exhibits a simple relationship to crustal thermal boundary layer (TBL). TI1e North Atlantic
age, after averaging and smoothing. The system- is generally too shallow for its age, and the
atic increase in the depth of the ocean floor Indian Ocean between Australia and Antarctica
away from the midocean ridges can be explained is too deep. Continental insulation, a chemically
by simple cooling models for the evolution of heterogenous mantle and accumulated slabs at
the oceanic lithosphere. The mean depth of depth may explain these anomalies. There is no
ocean ridges is 2.5 Ian below sealevel although evidence that shallow regions are caused by par-
regional variations off 1 lm1 around the mean ticularly hot mantle. In fact, there is evidence for
are observed. Thermal subsidence of the seafloor moderate mantle temperature anomalies
is well approximated by an empirical relationship associated with hotspot volcanism.
of the form Residual depth anomalies, the depar-
ture of the depth of the ocean from the value
d(t) = d + At 112
0
expected for its age, in the ocean basins h ave
where d is seafloor depth referred to sea-level dimensions of order 2000 km and amplitudes
and positive downward, do is mean depth of mid- greater than 1 km . Part of the residual anoma-
ocean ridges and t is crustal age. TI1e value of A lies are due to regional changes in crustal thick-
is around 350 m/(my) 112 if d and d0 are expressed ness. This cannot explain all of the anomalies.
in meters and t in my. Depth anomalies or resid- Positive (shallow) depth anomalies - or swells -
ual depth anomalies refer to oceanfloor topog- are often associated with volcanic regions such as
raphy minus the expected thermal subsidence. Bermuda, Hawaii, the Azores and the Cape Verde
Although there is a large literature on the inter- Islands. These might be due to thinning of the
pretation of positive depth anomalies - swells plate, chemically buoyant material in the shal-
- it should be kept in mind that in a convect- low mantle, or the presence of abnormally hot
ing Earth, with normal variations in temperature upper mantle. Patches of eclogite in the man-
and composition, the depth of the seafloor is not tle are dense when they are colder than ambi-
expected to be a simple function of time or age. ent mantle, but they melt at temperatures some
Geophysical anomalies, both positive and nega- 200 °C colder than peridotite and can therefore be
tive, are well outside the normal expected varia- responsible for elevation and melting anomalies.
tions for a uniform isothermal mantle. Shallow areas often exceed 1200 m in height
Data from the western North Atlantic and above the expected depth and occupy almost the
central Pacific Oceans , for seafloor ages from entire North Atlantic and most of the western
64 THE SHAPE OF THE EARTH
Pacific. Almost every volcanic island , seamount or underlying and sinking piece of subducted slab
seamount chain surmounts a broad topographic or delaminated lower crust.
swell. The swells generally occur directly beneath
the volcanic centers and extend along fracture
zones. Small regions of anom.alously shallow Dynamic topography
depth occur in the northwestern Indian Ocean The long-wavelength topography is a dynamic
south of Pakistan, in the western North Atlantic effect of a convecting mantle. It is difficult to
near the Caribbean, in the Labrador Sea and in determine because of other effects such as crustal
the southernmost South Pacific. They are not thiclmess . Density and thermal variations in a
associated with volcanism but are slow regions convecting mantle deform the surface, and this
of the upper mantle as determined from seismic is known as the mantle dynamic topography.
tomography. The long -wave length geoid of the Earth is
Shallow regions probably associated with controlled by density variations in the deep man-
plate flexure border the Kurile Trench, the tle and has been explained by circulation models
Aleutian Trench and the Chile Trench. Major involving whole mantle flow. However, the rela-
volcanic lineaments without swells include the tionship of long-wavelength topography to man-
northern end of the Emperor Seamount chain, tle circulation has been a puzzling problem in
the Cobb Seamounts off the west coast of North geodynamics . Dynamic topography is mainly due
America and the Easter Island trace on the East to density variations in the upper mantle. Lay-
Pacific Rise. Bermuda and Vema, in the south- ered mantle convection, with a shallow origin for
east Atlantic, are isolated swells with no associ- surface dynamic topography, is consistent with
ated volcanic trace. For most of the swells expla- the spectrum, small amplitude and pattern of the
nations based on sediment or crustal thickness topography. Layered mantle convection,
and plate flexure can be ruled out. They seem with a barrier near 1000 km depth provides a
instead to be due to variations in lithospheric self-consistent geodynamic model for the amplitude and
composition or thickness, or abnormal upper pattern of both the long-wavelength geoid and surface
mantle. Dike and sill intrusion , underplating of topography.
the lithosphere by basalt or depleted peridotite ,
serpentinization of the lithosphere, delamina-
tion, or reheating and thinning the lithosphere
are mechanisms that can decrease the density The geoid
or thiclmess of the lithosphere and cause uplift
of the seafloor. A higher temperature astheno- The centrifugal effect of the Earth's rotation
sphere, greater amounts of partial melt, chem- causes an equatorial bulge, the principal depar-
ical inhomogeneity of the asthenosphere and ture of the Earth's surface from a spherical shape.
upwelling of the asthenosphere are possible sub- If the Earth were covered by oceans then, apart
lithospheric mechanisms. from winds and internal currents, the surface
A few places are markedly deep , notably the would reflect the forces due to rotation and the
seafloor between Australia and Antarctica - the gravitational attraction of external bodies, such
Australian-Antarctic Discordance or AAD - and as the Sun and the Moon, and effects arising fi·om
the Argentine Basin of the South Atlantic. Other the interior. When tidal effects are removed, the
deep regions occur in the central Atlantic and shape of the surface is due to density variations
the eastern Pacific and others, most notably in the interior. Mean sea level is an equipotential
south of India, are not so obvious because of surface called the geoid or figure of the Earth.
deep sedimentary fill. Most of the negative areas Crustal features, continents, mountain ranges
are less than 400 m below the expected depth, and midoceanic ridges rep resent departures of
and they comprise a relatively small fraction of the actual surface from the geoid, but mass com-
the seafloor area. They represent cold mantle, pensation at depth, isostasy, minimizes the influ-
lower melt contents, dense lower crust or an ence of surface features on the geoid. To first
THE GEOID 65
PACIFIC PLATE
lows are found south of India, near Antarctica
(south of New Zealand) and south of Australia.
The locations of the mass anomalies responsible
for these lows are probably in the lower man-
tle. Many shield areas are in or near geoid lows,
some of which are the result of deglaciation and
incomplete rebound. The thick continental crust
would, by itself, raise the center of gravity of con-
tinents relative to oceans and cause slight geoid
~~lilill· Geoid lows are concentrated in a narrow polar
highs . The thick lithosphere ( ~ 150 km) under
band passing through Antarctica , the Canadian Shield and continental shields is cold, but the seismic veloc-
Siberia. Most of the continents and smaller tectonic plates are ities and xenoliths from kimberlite pipes suggest
in this band . Long-wavelength geoid highs and the larger that it is olivine-rich and garnet-poor; the temper-
plates (Africa, Pacific) are antipodal and are centered on the ature and petrology have compensating effects
equator. The geoid highs control the location of the axis of on density. TI1e longterm stability of shields indi-
rotation. Large-scale mass anomalies in the deep mantle
cates that, on average, the crust plus its under-
control the long-wavelength geoid. These in turn can affect
lying lithosphere is buoyant. Midocean ridges
the stress in the surface plates.
show mild intermediate-wavelength geoid highs ,
but they occur on the edges of long-wavelength
order, near-surface mass anomalies that are com- highs. Hotspots, too, are associated with geoid
pensated at shallow depth have no effect on the highs. The long-wavelength features of the geoid
geoid. are probably due to density variations in the
The shape of the geoid is now known fairly lower mantle and the resulting deformations of
well, particularly in oceanic regions, because the core-mantle boundary and other boundaries
of the contributions from satellite geodesy [see in the mantle (Richards and Hager, 1984).
geoid images]. Apart from the geoid highs Geoid anomalies are expressed as the differ-
associated with subduction zones, there is little ence in elevation between the measured geoid
correlation of the long-wavelength geoid with and some reference shape. The reference shape is
such features as continents and 1nidocean ridges. usually either a spheroid with the observed flat-
The geoid reflects temperature and density vari- tening or the theoretical hydrostatic flattening
ations in the interior, but these are not simply associated with the Earth's rotation, the equilib-
related to the surface expressions of plate rium form of a rotating Earth. TI1e latter, used
tectonics. in Figure 6.2, is the appropriate geoid for geo-
The largest departures of the geoid from physical purposes and is known as the nonhy-
a radially symmetric rotating spheroid are the drostatic geoid. TI1e geometric flattening of the
equatorial and antipodal geoid highs centered on Earth is 1/298.26. The hydrostatic flattening is
the central Pacific and Africa (Figures 6.1 and 6.2). 1/299.64.
The complementary pattern of geoid lows lie in a The maximum geoid anomalies are of the
polar band that contains most of the large shield order of 100 m. This can be compared with the
regions of the world. The largest geoid highs of 21 Jan difference between the equatorial and
intermediate scale are associated with subduc- polar radii. To a good approximation the net
tion zones. The most notable geoid high is cen- mass of all columns of the crust and mantle
tered on the subduction zones of the southwest are equal when averaged over dimensions of a
Pacific near New Guinea, again near the equa- few hundred kilometers. This is one definition
tor. The equatorial location of geoid highs is not of isostasy. Smaller-scale anomalies can be sup-
accidental; mass anomalies in the mantle control ported by the strength of the crust and litho-
the moments of inertia of the Earth and, there- sphere. TI1e geoid anomaly is nonzero in such
fore, the location of the spin axis and the equa- cases and depends on the distribution of mass.
tor. TI1e largest intermediate-wavelength geoid A negative t,p, caused for example by thermal
66 THE SHAPE OF THE EARTH
from the sinker itself. For an increase in viscos- Basalt chemistry exhibits lateral variations on
ity with depth, the deformation of the upper length scales of 150 and 400 lm1 that may be
boundary is less and the net geoid anomaly is related to intrinsic heterogeneity of the man-
positive. tle. Large variations in magma output along vol-
canic chains occur over dista nces of hundreds
to thousands of km; most chains - often called
Shorter wavelength features hotspot tracks - are less than 1000 lm1 long. These
There is a broad range of dominant wavelengths- dimensions may be the characteristic scales of
or peaks in the spectrum - in the geoid and mantle chemical and fertility variations . This
bathymetry, ranging in wavelength from 160 km provides a straightforward explanation of the
to 1400 km. Although these have been inter- order of magnitude variations in volcanic output
preted as the scales of convection and thermal along long volcanic chains and along spreading
variations they could also be caused by density ridges .
variations due to chemistry and, perhaps, par-
tial melt content, in the upper mantle . Several Interpreting the geoid
of the spectral peaks are similar in wavelength to Quantitative interpretations of the geoid are
chemical variations along the ridges . The shorter often based on relations such as wavelength
wavelengths may be related to thermal contrac- vs . sph erical harmonic degree ; the geoid
tion and bending of the lithosphere. The longer bears little relation to global tectonic maps
wavelengths probably correspond to lithologic or to present tectonic features of the Earth other
(major element) variations in the asthenosphere than trenches. The Earth's largest pos-
and, possibly, fertility and melting point vari- itive geoid anomalies have no simple
ations. Intermediate-wavelength (400-600 km) re l ation s h ip to continents and ridges.
geoid undulations are continuous across fracture The Mesozoic supercontinent of Pangea , however,
zones and some have linear volcanic seamount apparently occupied a central position in the
chains at their crests. Atlantic-African geoid high. This and the equato-
Profiles of gravity and topography along the rial Pacific geoid high contain most of the world's
zero age contour of oceanic crust are perhaps hotspots although there is little evidence that the
the best indicators of mantle heterogeneity. mantle in these regions is particularly hot. The
These show some very long wavelength varia- plateaus and rises in the western Pacific formed
tions , ~ sooo km and ~ 1000 km. and also abrupt in the Pacific geoid high, and this may have been
changes. Ridges are not uniform in depth, grav- the early Mesozoic position of a subduction com-
ity or chemical properties. Complex ridge-plume plex, the fragments of which are now the Pacific
interactions have been proposed, the assumption rim portions of the continents. Geoid highs that
being that normal ridges should have uniform are unrelated to present subduction zones may
properties. The basalts along midocean ridges are be the former sites of continental aggregations,
fairly uniform in composition but nevertheless the centers of large long-lived plates - which
show variations in major oxide and isotopic com- cause mantle insulation and , therefore, hotter
positions. Major and minor element chemistry shows than normal mantle. The pent-up heat causes
spectral peaks with wavelengths of 225 and 575 km. In uplift, magmatism, fragmentation , and the sub-
general, one cannot pick out the ridge-centered sequent formation of plateaus, aseismic ridges
and near-ridge hotspots from profiles of grav- and seamount chains. However, the effect must
ity, geoid , chemistry and seismic velocity. This be deep in order to also affect the long wave-
suggests that short-wavelength elevation anoma- length geoid .
lies, e.g. hotspots, do not h ave deep roots or deep When the subduction-related geoid highs are
causes. Some hotspots have low seismic velocities removed from the observed field, the residual
at shallow depths, shallower than 200 km, consis- geoid shows broad highs over the central Pacific
tent with low-melting-point constituents in the and the eastern Atlantic-African regions. Like the
asthenosphere. total geoid, the residual geoid does not reflect
68 THE SHAPE OF THE EARTH
The approximate locations of subducted slabs or is important for these problems. By contrast the
subduction zones during the last two supercontinent cycles. upper mantle has only a weak correlation with
Note the similarity with the long wavelength geoid. This the l = 2 and 3 geoid. The effects of pressure on
pattern also matches that of seismic tomography at a depth of
viscosity and thermal expansion are such that
about 800-1000 km depth , suggesting that slabs bottom out
we expect the lower mantle to convect very slug-
at ~ I 000 km depth.
gishly. The large feat ures in the lower mantle are
probably ancient, and not caused by recent plate
advancing continents. These are labeled 'fast' tectonic cycles. The implication is that the geoid
because these are seismically fast regions of the and the rotation axis are relatively stable. The
transition region, where cold lithosphere may plates and the upper 1000 km of the mantle are
have cooled off the mantle. The arrows repre- the active layers.
sent the motions of the continents over the
past 110 Myr. Most of the hatched regions are Polar wander
also geoid lows. Figure 6.5 shows the inferred
locations of subduction zones over the past two Because the Earth is a dynamic body, it is impos-
supercontinent cycles. Slabs and delaminated sible to define a permanent internal reference
lower crust have entered the mantle in these frame. There are three reference frames in com-
regions, both cooling it and fertilizing it. mon use: the rotation axis, the geomagnetic ref-
erence frame and t h e h otspot reference frame.
The rotational frame is controlled by the size of
Involvement of the lower mantle the mass anomalies and their distance from the
Tomographic techniques can be applied to the axis of rotation . Upper-mantle effects are impor-
problem of lateral heterogeneity of the lower tant because lateral heterogeneity is greater than
mantle [tomography geoid lower mantle, lower-mantle or core heterogeneity and because
Hager 0 • Connell). Long-wavelength velocity they are far from the center of the Earth. The
anomalies in the lower mantle correlate well lower mantle is important because of its large
with the l ~ 2, 3 geoid. Phenomena such as volume, but a given mass anomaly has a greater
tides. Chandler wobble, polar wander and the effect in the upper mantle. The location of the
orientation of the Earth's spin axis depend on the magnetic pole is controlled by convection in the
l = 2 component of the geoid; the lower mantle core, which in turn is influenced by the rotation
POLAR WANDER 71
would expect that this configuration would be extensive volcanism and enriched mag1nas, pre-
consistent with a stationary or a southern migra- sumably from the shallow mantle.
tion of Gondwana , unless a geoid high centered The largest known positive gravity anomaly
on or near Africa was rotating the whole assem- on any planet is associated with the Tharsis vol-
blage toward the equator. The areas of very low canic province on Mars . Both geologic and grav-
upper-mantle velocities in northeast Africa and ity data suggest that the positive mass anomaly
the western Indian Ocean may be the former site associated with the Tharsis volcanoes reoriented
of the center of Gondwana. the planet with respect to the spin axis. placing
Thus, expanding the paradigm of continental the Tharsis region on the equator. There is also
drift and plate tectonics to include continental evidence tha t magmatisn1 associated with large
insulation and true-polar wandering may explain impacts reoriented the Moon. The largest mass
the paradoxes of synchronous global tectonic and anomaly on Ea rth is centered over New Guinea,
magmatic activity, rapid breakup and dispersal and it is also almost precisely on the equator.
of continents following long periods of conti- The long-wavelength part of the geoid correlates
nental stability, periods of static pole positions well with subduction zones , a nd these appear to
separated by periods of rapid polar wandering, control the orientation of the mantle relative to
sudden changes in the paths of the wandering the spin axis. Thus, we have the possibility of a
poles, the migration of rifting and subduction, feedback relation between geologic processes and
initiation of melting, the symmetry of ridges the rotational dynamics of a planet. Volcanism
and fracture zones with respect to the rotation and continental collisions cause mass excesses to
axis, and correlation of tectonic activity and be placed near the surface. These reorient the
pola r wandering with magnetic reversals . Tum- planet, causing large stresses that initiate rifting
bling of the mantle presumably affects convec- and faulting, which in turn affect volcanism and
tion in the core and orientation of the inner core subduction. Curiously, Earth scientists have been
and offers a link between tectonic and magnetic more reluctant to accept the inevitability of true-
field variations. Global plate reorganizations are polar wandering than to accept continental drift,
a necessary part of plate tectonics on a sphere. even though the physics of the former is better
New plate boundaries are often accompanied by understood.