Mantle dynamics and convection

Mantle dynamics is the study of the motion of the earth's mantle, which is primarily generated by convection. Convection within the mantle causes the transfer of heat from one region of the earth to another. This convection facilitates the movement of the lithospheric plates of the earth, resulting in mountain building, earthquakes, volcanism, and the evolution of continents and ocean basins. Understanding mantle dynamics and convection is a major component of the framework for explaining how the earth developed, how it works, and why it is constantly changing.

Chemical and Mechanical Properties of the Mantle

The earth's interior consists of a series of shells of different compositions and mechanical properties. Based on chemical composition, the outermost layer is the crust, consisting of both continental and oceanic crust. The next major compositional layer of the earth is the mantle, which is approximately 2,890 kilometers thick and constitutes about 84 percent of the earth's volume and 68 percent of its mass. By studying fragments of the mantle that have been brought to the surface by volcanic eruptions, earth scientists have deduced that the mantle is chemically composed of silicate rocks containing primarily silicon, oxygen, iron, and magnesium.

Based upon physical, or mechanical, properties, the solid, strong, rigid outer layer of the earth is termed the lithosphere (“rock sphere”). The lithosphere includes the crust and the uppermost part of the mantle. The earth's lithosphere varies greatly in thickness, from as little as 10 kilometers in some oceanic regions to as much as 300 kilometers in some continental areas. The earth's lithosphere is broken up into a series of large fragments, or rigid plates. Seven major plates and a number of smaller ones have been distinguished, and they grind and scrape against one another as they move independently, similar to chunks of ice in water. Much of the earth's dynamic activity occurs along plate boundaries, and the global distribution of associated tectonic phenomena, particularly earthquakes and volcanism, delineates the boundaries of the plates.

Within the upper mantle, there is a major zone where the temperature and pressure are such that part of the material melts, or nearly melts. Thus the rocks in this region of the earth lose much of their strength, becoming soft and plastic like, so that they can slowly flow as a viscous liquid. This zone of easily deformed mantle is termed the asthenosphere (“weak sphere”). Seismic velocities are about 6 percent lower in the asthenosphere than in the lithosphere. Although there is no fundamental change in chemical composition between the two regions, the lithosphere and the asthenosphere are mechanically distinct.

The lithosphere rides over the plastic, partly molten asthenosphere. As the lithosphere moves, the continents split, and the large plates drift thousands of kilometers across the earth's surface. All the major structural features of the earth are the result of a system of moving lithospheric plates. Movement in the plate tectonic system is driven by the loss of internal heat energy, primarily from within the mantle. This heat-driven internal movement is responsible for the creation of ocean basins and continents, as well as deformations of the earth's solid outer layers that generate earthquakes, mountain belts, and volcanic activity at the plate boundaries. The primary source of heat within the mantle appears to be the radioactive decay of uranium, thorium, and potassium.

Below the asthenosphere, the rock becomes stronger and more rigid. The higher pressure below the asthenosphere offsets the effect of higher temperatures, making the rock stronger than in the overlying asthenosphere. The layer from the base of the earth's crust to a depth of approximately 670 kilometers is designated the upper mantle, which includes the asthenosphere and the lower part of the lithosphere. Earth scientists have had difficulty unraveling all the layers in the upper mantle. One model of the structure of the upper mantle designates a B layer, nearly 400 kilometers thick and of fairly uniform composition, and a C layer, between 200 and 300 kilometers thick, in which the chemical composition appears to be quite variable.

The transition from the upper to the lower mantle is quite gradual, with the depth of the separating, or transition, boundary usually thought to be between 600 and 670 kilometers. One of the fundamental questions about the seismic discontinuity that separates the upper mantle from the lower mantle is whether it is a barrier to lower mantle convection. One theory is that this boundary temporarily prevents the penetration of mantle material but ultimately will allow passage. As seismic wave velocities appear to be very steady throughout the DN region, a layer between 670 and 2,700 kilometers deep, the lower mantle is assumed to have a less complex structure than the upper mantle. Variations of properties inside the DN region appear to be predominantly caused by the effects of simple compression. However, inside the DO region, from 2,700 kilometers deep down to the outer core, seismic velocity falls continuously, indicating some continuous changes in physical properties and chemical composition that could produce deep mantle convection.

Evidence for Mantle Convection

Better understanding of plate tectonics from analysis and interpretation of vast amounts of seismic data, as well as the application of improved observational and experimental techniques to the study of the properties of mantle materials, has confirmed the existence of mantle convection. Apparent episodic material exchanges between the upper and lower mantle, the genesis of plumelike upwellings, and the ultimate fate of subducted slabs within the earth are all aspects of the mantle's dynamic convection system. One of the most significant questions in mantle dynamics is whether mantle convection is isolated in the upper mantle or involves the whole mantle.

The advent of seismic tomography in the mid-1980s, coupled with new laboratory and computational capabilities in the 1990s, has profoundly impacted the understanding of mantle convection. Global seismic tomography has helped to resolve the parameters that characterize mantle dynamics, particularly a viscosity increase from the upper to the lower mantle, an endothermic phase transition at the upper mantle-lower mantle boundary, heat flow across the core-mantle boundary (CMB), and the effect of the motion of rigid surface plates on convection patterns. Using seismic tomography, images of three-dimensional variations in seismic velocities have revealed the deep structure of the underlying surficial plates. Patterns of high and low seismic velocity have been identified throughout the mantle, with the strongest variations found in the upper 300 kilometers.

Unexpected features, such as deep roots of high-velocity material extending 300 to 400 kilometers below continental cratons, have greatly enhanced the understanding of plate tectonics and continental formation. Deep-seated upwellings under the ocean ridges and beneath major volcanic hot spots are indicated by low-velocity regions and by deflections of transition-zone discontinuities. While some tomographic images show descending lithospheric slabs that are apparently blocked at the upper mantle-lower mantle boundary, others show lithospheric slabs sinking nearly to the CMB. Numerical models indicate that reasonable simulations of subducted lithospheric slabs and plumes will penetrate the upper mantle-lower mantle boundary, although they may be temporarily blocked. Laboratory models of descending sheets interacting with contrasting density and viscous interfaces support this conclusion. Some slabs of subducting lithosphere have been seismically imaged as high-velocity tabular downwellings extending throughout the mantle. Studies conducted in 2004 indicated that large lithospheric slabs could be subducted to a depth of 2,900 kilometers. This supports the theory of whole-mantle convection in which convection involves all layers of the mantle.

Upper Mantle Dynamics and Convection

Because of its ability to flow, the asthenosphere figures prominently in the dynamic theories on the causes of vertical motion observed at the earth's surface, such as postglacial rebound. Periodic compensatory adjustments that take place in the interior of the earth in response to changing mass distributions at the surface that arise from erosion, sedimentation, glaciation and deglaciation, and volcanism are thought to occur through flow in the asthenosphere.

Likewise, the asthenosphere plays a prominent role in models of the large horizontal movements of the lithosphere as observed in continental drift and plate tectonics. As it slowly churns in large convection cells, the asthenosphere is the lubricating layer over which the plates glide. Thermal convection in the asthenosphere is thought to be a fundamental force in driving the tectonic plates. According to this scenario, hot mantle material rises at the mid-oceanic spreading ridges (divergent boundaries), where it escapes as magma, cools, and generates new oceanic crust. The sea floor moves in conveyor-belt fashion, ultimately to be destroyed at convergent plate boundaries, where it is subducted, or carried down, into the asthenosphere and eventually remelted. The rest of the hot mantle material spreads out sideways beneath the lithosphere, slowly cooling in the process. As it flows outward, it drags the overlying lithosphere outward with it, thus continuing to open the ridges. When the hot mantle material cools, the flowing material becomes dense enough to complete the convection cycle by sinking back deeper into the mantle, and tomographic images indicate that this is happening under subduction zones at converging boundaries.

Some debate continues as to whether convection is confined to the upper mantle in a thin asthenosphere or whether it occurs throughout the mantle. The convection cells need not be confined to the asthenosphere. Seismic velocity data indicate that oceanic lithosphere can be subducted to depths of approximately 700 kilometers. Thus convection cells may operate at least down to those depths. In addition, in some places in the asthenosphere, the temperature may reach the rock-melting temperature and produce magma, thus giving the asthenosphere another dynamical role as the source region of many types of igneous rocks.

Lower Mantle Dynamics and Convection

A serious ongoing debate among earth scientists about the size of convection cells in the mantle began in the early 1980s. Because of the different chemical compositions of the upper and lower mantle, geochemists have argued that the upper and lower mantles must have isolated convection cells with virtually no mixing between them. Thus slabs of the lithosphere that sink below the surface at the edge of tectonic plates should stay within the upper mantle, with their material being recycled there. In addition, geochemical evidence also exists for distinct reservoirs in the mantle, emphasizing the importance of plume flows from internal boundary layers as distinct from large-scale flows associated with the oceanic plates.

Opposing this view, many geophysicists have maintained that convection involves the entire mantle. In the late 1990s, numerous three-dimensional seismic tomographic studies mapped seismic speeds in the earth's mantle, and the interpretations provide strong evidence that the lithosphere is sinking well into the lower mantle. The tomographic images of seismic wave speeds at different depths are a rough indication of the temperature distribution in the mantle. Waves travel more quickly in regions that are colder, and more slowly in hotter regions. In numerous locations beneath the earth, tomographic images show cold regions to be a continuous function of depth far into the lower mantle, suggesting the descent of some slabs of oceanic lithosphere into the lower mantle at the edge of tectonic plates. One interpretation of the tomographic images shows that the mantle's heterogeneity is dominated by large-scale structures, which support a mantle convection system dominated by large flow patterns. Evidence is not conclusive as to whether the mantle is only a layered convective system, but it does suggest that significant material transport occurs across the upper mantle-lower mantle boundary.

In addition, increasing geophysical evidence supports the conjecture that the earth's core interacts with the surrounding mantle. Images from seismic tomography reveal that the lowermost 200 to 400 kilometers of the mantle is one of the most heterogeneous regions of the earth. Above the CMB, seismic images indicate the presence of two laterally variable seismic discontinuities, one at 130 to 400 kilometers and another at 5 to 50 kilometers above the CMB. According to the tomographic images, both regions have anisotropic properties, and the complexities of physical and dynamic processes are as sophisticated as those present in the lithosphere and shallow asthenosphere, which supports the idea of a dynamic lower mantle involved in convection.

In the late 1990s, the discovery that the DO layer O is associated with dramatically reduced seismic velocities at 5 to 50 kilometers above the CMB changed the persisting idea that the deep region of the lower mantle was solid. From tomographic images, some of the most unusual anomalies seen in the lower mantle are thin patches, less than 40 kilometers thick, in which seismic velocities are locally reduced by 10 percent or more. Such ultra-low-velocity zones are not seen anywhere else in the mantle. Explaining their presence requires massive local melting within the lowermost mantle, meaning that this region is most likely convecting. The significant heterogeneity, which likely involves locally hot and partially molten zones near the CMB, is indicative of the dynamical behavior of the DO layer. Some earth scientists reason that many of the volcanic plumes associated with hot spots at the earth's surface, such as the Hawaiian island chain, represent upwelling jets of hot rock in the mantle that are preferentially lined up above the ultra-low-velocity molten patches in the DO layer.

Integrated Mantle Convection

In the late 1990s, many earth scientists believed that it was necessary to make a compromise between whole-mantle convection and isolated-mantle convection. With regard to the existing geochemical evidence, the main requirement is that there exist chemically distinct reservoirs that do not have to be totally confined to the lower mantle. Three suggested models have emerged. Some earth scientists support a model of the mantle that contains isolated, discontinuous volumes of material dispersed throughout. Other earth scientists suggest that a reasonable model is one in which the mass transport in the mantle does not occur in a steady, continuous fashion but rather in an intermittent, nonsteady state. Still others believe that lithospheric slabs descending into the mantle lose geochemically monitored elements in the upper mantle as they sink to lower depths. The best model may contain aspects of all three of these alternatives.

Laboratory experiments in the mid-1990s have shown that the oxides of the earth's deep mantle react vigorously when placed in contact with liquid iron alloys, thought to exist in the outer core, at the high pressures and temperatures at the CMB. These experiments suggest that the rocky mantle is slowly dissolving, over geological time spans, into the liquid metal of the outer core. The slow dissolution appears to be related to a fundamental change in the bonding character of oxygen at high pressures. Whereas oxygen forms insulating compounds at low pressures, it can become a metal-alloying component at high pressures. Thus, when coupled with seismic tomography, experimental and theoretical investigations of high pressures point to the CMB as perhaps being the most chemically active region in the earth's interior. Numerical models of the mantle suggest that the earth may have undergone a transition from layered to whole-mantle convection caused by a combination of secular cooling and a decrease in heat production in the mantle from radioactive decay.

The products of the chemical reactions at the CMB, where insulating oxides meet metallic alloys, may well explain the seismologically observed heterogeneity of the DO layer in the mantle. In addition, piles of oceanic crust that have settled toward the bottom of the mantle may further contribute to the heterogeneity of the region. The possible occurrence of varying amounts of metal alloys at the base of the mantle is particularly important because metal conducts heat much more readily than insulating oxides do. Consequently, heat may be emerging from the CMB in a spatially variable manner that determines the pattern of convection throughout the earth's mantle.

Significance

Many of the geophysical and geological phenomena of the earth's crust are consequences of the dynamics and thermal convection within the underlying mantle. The major features of mantle convection are deduced from seismic data, laboratory investigations, and computational modeling. The emerging picture is that the tectonic plates of the lithosphere are the most active component of mantle convection, whereas mantle plumes are an important secondary component. Direct consequences of mantle dynamics and convection include the relative motions of the lithospheric plates, the spreading of the sea floor and formation of new crust, volcanism in its various tectonic settings, much of the earth's seismic activity, and the majority of the observed heat flow through the earth's surface.

Unraveling the complexities of the mantle continues to be a challenge for seismology, but the results play a crucial role in answering questions regarding the composition, dynamics, and evolution of the earth. Better understanding of the present-day mantle is providing a much more complete picture of the evolution and interaction of the earth's thermal and tectonic regimes. Plausible arguments indicate that the mantle was episodically layered in Precambrian times and that plate tectonics would not have worked when the mantle was more than about 50 degrees Celsius hotter than at present. Whether plumes were more or less important in the past is being studied. Based on the present model of the mantle's dynamic and convection patterns, plumes do not offer an alternative to plate tectonics because they are derived from a different thermal boundary layer (the DO layer).

Insight into the dynamic interactions in the lower part of the mantle near the CMB is very important for better understanding past geological phenomena of significant magnitude. In particular, there is evidence for periods of massive volcanic eruptions (superplumes) that were hundreds of times greater than anything the earth has experienced in recent geological time. Models of the deep mantle based on three-dimensional tomographic images and laboratory observations indicate that superplume events could be the surface manifestation of fluid-dynamical instabilities triggered at the CMB. Such models have generated the first glimpses of how such massive instabilities are initiated deep inside the earth.

Principal Terms

asthenosphere: the weak zone directly below the lithosphere, from 10 to 200 kilometers below the earth's surface, believed to consist of soft material that yields to viscous flow

convection cell: a pattern of movement of mantle material in which the central area is uprising and the outer area is downflowing because of density changes produced by heat variations

core-mantle boundary: the seismic discontinuity 2,890 kilometers below the earth's surface that separates the mantle from the outer core

lithosphere: the relatively rigid outer zone of the earth, which includes the continental crust, the oceanic crust, and the part of the upper mantle lying above the weaker asthenosphere

lower mantle: the seismic region of the earth between 670 and 2,890 kilometers below the surface, consisting of the DN and DO layers

mantle plume: a vertical cylindrical distribution of material in the mantle within which abnormal amounts of heat are conducted upward to form a hot spot at the earth's surface

seismic tomography: a processing technique for constructing a cross-sectional image of a slice of the subsurface from seismic data

upper mantle: the part of the mantle that lies above a depth of about 670 kilometers, consisting of the B layer and the C layer

Bibliography

Brown, G. C., and A. E. Mussett. The Inaccessible Earth. 2d ed. New York: Chapman and Hall, 1993.

Davies, Geoffrey F. Mantle Convection for Geologists. New York: Cambridge University Press, 2011.

Hamblin, W. K., and E. H. Christiansen. Earth's Dynamic Systems. 10th ed. Upper Saddle River, N.J.: Prentice Hall, 2003.

Jackson, I., ed. The Earth's Mantle. Cambridge, England: Cambridge University Press, 1998.

Jeanloz, R., and B. Romanowicz. “Geophysical Dynamics at the Center of the Earth.” Physics Today 50 (August 1997): 22.

Lay, T., and Q. Williams. “Dynamics of Earth's Interior.” Geotimes 43 (November 1998): 26.

Lillie, R. J. Whole Earth Geophysics. Upper Saddle River, N.J.: Prentice Hall, 1999.

Lowrie, William. Fundamentals of Geophysics. 2d ed. New York: Cambridge University Press, 2007.

"Mantle." National Geographic Education, 30 Apr. 2024, education.nationalgeographic.org/resource/mantle/. Accessed 10 Feb. 2025.

Murphy, J. Brendan, and Tjeerd H. van Andel. "Earth's Layers." Britannica, 14. Jan. 2025, www.britannica.com/science/plate-tectonics/Earths-layers. Accessed 10 Feb. 2025.

Schubert, Gerald, Donald L. Turcotte, and Peter Olson. Mantle Convection in the Earth and Planets. New York: Cambridge University Press, 2001.

Strahler, A. N. Plate Tectonics. Cambridge, Mass.: GeoBooks, 1998.