News & Views | Published:

Earth science

How old roots lose their bounce

Naturevolume 417pages911913 (2002) | Download Citation


Active mountain belts have crustal 'roots' that gravitationally balance the high topography. So why do old mountains that have been worn flat by erosion still have these roots?

Some mountain ranges form along the boundaries of colliding tectonic plates, and the classic account of their rise and fall runs as follows. In these 'collisional belts', the Earth's crust is squeezed, thickened and uplifted by the forces that drive the plates together, and is also thickened by the addition of magmas from the underlying mantle. Where powerful mountain-building forces are at work today, as in the central Andes and Tibet, the mass of the mountains is gravitationally supported by a thick 'root' of buoyant low-density rock beneath them. When active uplift and crustal thickening cease, erosion begins the inexorable process of reducing once lofty mountains to low-lying plains. As the mountains are eroded, gravitational balance is maintained by continued uplift of the buoyant crustal root to compensate for mass loss at the surface.

As Fischer reports on page 933 of this issue1, however, this classic balancing act seems not to apply to many old and deeply eroded collisional belts. In these places, disproportionately thick crustal roots have often survived for hundreds of millions — even billions — of years, and Fischer provides an explanation for why that should be so (Fig. 1).

Figure 1: Mountain building at 'collisional' boundaries, and the fate of crustal roots.
Figure 1

a, The collision of tectonic plates produces a thickened crust — a mountain range — which is gravitationally supported by an underlying crustal root of buoyant, low-density rock. b, When active mountain-building ends, the erosional loss of surface mass is gravitationally balanced by rebound of the buoyant root. The principle of isostasy suggests that continued uplift and exhumation — deep erosion — of the crust should continue until the buoyant crustal root is completely consumed. c, Fischer1 finds, however, that older collisional belts tend to have disproportionately thick roots even when surface topography has been greatly reduced. She explains this as a result of reduced root buoyancy, caused by mineralogical reactions that increase root density relative to that of the surrounding mantle.

She has examined collisional mountain belts of all ages worldwide. She finds that the ratio of elevation to crustal-root thickness decreases systematically from about 0.1–0.2 for the youngest mountain belts (root thickness 5–10 times greater than surface relief) to essentially zero for mountain belts several hundred million years old (still appreciable root thickness, but minimal surface relief; see Fig. 1a on page 933). This is perplexing because, given the mass balance between crustal roots and surface load in young mountain ranges, it would seem that removal of the mass at the surface should involve proportional loss of the buoyant root. Old mountains eroded to flat-lying plains should have no remaining roots.

It was triangulation anomalies, observed in the Himalayas during the Trigonometrical Survey of India in the mid-nineteenth century, that led to the realization that mountains are not simply loads piled onto the surface but are compensated by a comparable 'mass deficit' at depth. This gravitational balancing act is termed isostatic compensation. In its most common form, isostasy is simply Archimedes' principle applied to the Earth: low-density crustal blocks 'float' in a higher-density 'fluid' mantle, much as an iceberg floats in water with its tip gravitationally balanced by a much greater volume of displaced water beneath the surface. Mountain ranges in isostatic equilibrium are held aloft by the buoyant forces of crustal roots, and the higher the mountains, the thicker the roots. Strictly speaking, the iceberg analogy is valid only if the crust and mantle are assumed to be sufficiently weak — deformable — that the crust really can 'float' in the dense 'fluid' mantle beneath. This assumption seems to be plausible for active mountain belts, but it might not be valid for older mountain ranges, for which progressive cooling may lead to an increase in rigidity.

Fischer assessed the role of isostatic balance by considering two hypotheses for how crustal roots could survive over geological time. The first was that the lithosphere — the crust and upper mantle — becomes increasingly rigid over time, impeding the buoyant rebound of low-density roots; the second was that the density contrast between crustal root and mantle decreases with time. For all but the youngest mountain belts, these two hypotheses will produce very different predicted gravity anomalies, a consequence of the differing density contrast between root and mantle implicit in the two hypotheses. Fischer modelled crustal densities to obtain a best fit to the observed gravity anomalies and shows that the data do not fit the first hypothesis: old crustal roots cannot be composed of the same low-density buoyant material that makes up the roots of active mountain belts. Instead, she shows that the density contrast between the crustal root and the mantle decreases markedly with increasing age, although in almost every case the crustal root still provides isostatic compensation for the surface features.

This analysis indicates that, over time, the crustal roots must evolve in response to cooling temperatures and to mineralogical reactions that drive the lower crust to higher density. This process does not take long: root buoyancy seems to be relatively reduced for all mountain belts in which activity ended 20 million years ago or earlier. Fischer suggests that the decrease in density contrast between deep crust and mantle with age means that crustal roots remain roughly in isostatic equilibrium even as the surface is reduced to low-lying topography. Implicit in this conclusion is that the lithosphere remains weak enough for the buoyant uplift of roots to continue over long periods of geological time.

Fischer's results have broader implications, particularly as they bear upon the widely held notion of 'gravitational collapse' of mountain belts. Gravitational collapse, also termed post-orogenic collapse, posits that the crust that is thickened during mountain building is rapidly thinned through post-orogenic 'extension'2,3 — stretching — which produces accelerated crustal thinning and gravitational collapse of the mountain belt itself. The process is widely assumed to be a normal stage in the mountain evolutionary cycle, one that should result in the virtual disappearance of crustal roots3. According to this view, active mountain building is followed relatively quickly by collapse of the mountain belt, driven largely by thermal erosion and thinning of the lithosphere (caused by heating from the deeper mantle), and by internal heating of the thickened crust itself.

Fischer's findings imply that post-orogenic crustal evolution is not always quite so simple. There are certainly examples of rapid gravitational collapse after mountain building. But the persistence of crustal roots in many old mountain belts, and their apparent continued uplift over hundreds of millions of years, mean that large-scale extension, thinning and destruction of crustal roots need not be the inevitable conclusion.


  1. 1

    Fischer, K. M. Nature 417, 933–936 (2002).

  2. 2

    Dewey, J. F. Tectonics 7, 1123–1139 (1988).

  3. 3

    Vanderhaeghe, O. & Teyssier, C. Tectonophysics 335, 211–228 (2001).

Download references

Author information


  1. the Department of Terrestrial Magnetism, Carnegie Institution of Washington, 5241 Broad Branch Road, NW, 20015, Washington DC, USA

    • David James


  1. Search for David James in:

Corresponding author

Correspondence to David James.

About this article

Publication history

Issue Date


Further reading


By submitting a comment you agree to abide by our Terms and Community Guidelines. If you find something abusive or that does not comply with our terms or guidelines please flag it as inappropriate.

Newsletter Get the most important science stories of the day, free in your inbox. Sign up for Nature Briefing