lithosphere

lithosphere Plate tectonics relies on the concept of a rigid lithosphere. Although the term ‘crust’ was used originally in geology to denote such a rigid layer, the modern concept of the lithosphere generally includes the crust and the uppermost part of the upper mantle (Fig. 1). Whereas the base of the crust is essentially a compositional boundary between mafic and ultramafic rocks, characterized by a rapid increase in seismic P-wave velocity to more than 8 km s⁻¹ (the Mohorovicicć discontinuity), the lithosphere is defined rheologically as the strong, elastic layer at the surface of the Earth, overlying the weak, ductile asthenosphere. The base of the lithosphere is marked by ťhe transition from brittle to ductile deformation, and is probably gradational. The lithosphere, as defined in this way, is sometimes referred to as a ‘mechanical boundary layer’.

Because the lithosphere is rigid, it cannot convect as the deeper mantle does. Heat therefore passes through it mainly by conduction, which is a less effective mechanism than convection. In consequence, the geothermal gradient in the lithosphere is considerably higher than that in the rest of the mantle, and there is a large temperature difference across it. One can therefore also think of the lithosphere as a thermal boundary layer.

There are several ways of estimating the thickness of the lithosphere. Because it is a rheological boundary, and rheological properties depend, inter alia, on strain rate, the thickness also depends on strain rate.

One of the most common ways of estimating lithospheric thickness is to use seismology. Since this depends on the elastic deformation of rocks at frequencies of about 0.01–1 Hz, it is effectively a very high strain-rate method. Seismology shows that the velocity of seismic waves generally increases with depth in the upper mantle. However, in most places there is a small decrease in velocity starting at depths of around 100 km, which is thought to mark the transition from lithosphere to asthenosphere. The thickness of the lithosphere defined on this basis varies between different geological provinces. It is greatest under the cratonic centres of continents, where it may possibly reach around 300 km, and it is least, probably about 4 km, under mid-ocean ridges where new lithosphere is formed.

Lithospheric thickness generally increases with age since its last major tectonic reactivation. The lithosphere–asthenosphere transition should roughly follow the isotherm corresponding to the brittle–ductile transition. For oceanic lithosphere the depths of isotherms are determined by conductive cooling since its creation at a mid-ocean ridge. These depths can be computed and are approximately proportional to the square root of the age, as therefore is the lithospheric thickness. The seismically defined oceanic lithosphere increases in thickness from just a few kilometres at zero age to about 100 km at ages of over 100 Ma (million years). These depths correspond approximately to the 1000 °C isotherm.

Another method of estimating lithospheric thickness is by measuring how much it flexes under applied loads. Geophysicists have shown that the lithosphere can be modelled successfully as a thin elastic plate overlying a viscous substratum. The magnitude and wavelength of the flexure depend on the applied load and on the thickness of the plate. A variety of loads can be modelled, including sea mounts, ocean islands, subducting slabs, and the isostatic forces acting on the dipping boundaries of fault blocks. Such studies also show an increase of lithospheric thickness with age, but the thicknesses inferred are approximately one-third of those determined seismically. This is because flexural adjustments take place over long timescales of around 10⁵ years or more, and at these low strain rates the brittle–ductile transition occurs at lower temperatures. In fact, the elastic thickness estimates from flexural studies fall between the 350 °C and 650 °C isotherms (Fig. 2).

The fact that the lithosphere comprises a significant part of the upper mantle has an important effect on its rheology. Oceanic lithosphere is typically over 90 per cent mantle; its strength, dominated by the mineral olivine, is high, and its density is also relatively high. Continental lithosphere may be only about 60 per cent mantle, and with significant quantities of typically crustal minerals such as feldspar and quartz it is much weaker, and relatively light. Thus, if two lithospheric plates are in collision and one is oceanic, the oceanic one can resist major deformation and be relatively easily overridden to form a subduction zone. However, if two continental plates are in collision, both are too light to sink easily, but they are also weak enough to be relatively easily folded or faulted. Thus continental collision zones tend to be wide and complex, containing many individual fault blocks as well as large fold mountains.

Roger Searle

Bibliography

Kearey, P. and and Vine, F. J. (1996) Global tectonics. Blackwell Science, Oxford.