Doyen of antiplumers, Don Anderson of CalTech, deplores what has become of the Plume Hypothesis. From this month's issue of Geoscientist
It is ironic that the Earth science community views with amusement and condescension the quaint old idea of last century that continents were fixed but has now accepted so enthusiastically the idea that giant continents can drift about but small volcanic islands can not.
The plume hypothesis is unsatisfactory on many levels.1-3 First, it is not scientific. It cannot be tested. It has so many variants, exceptions, rationalizations, ad hoc adjustments, and failed predictions that it is unsatisfactory at the most basic level. It started out as a simple, elegant testable idea. It has now been tested and falsified. Falsification, in science, is usually difficult but in this case all the predictions in the original papers have been shown to be false, but, more telling, the original evidence used to support the proposal, has been discredited and abandoned. The plume hypothesis is now an inelegant, ("sophisticated"), even ugly, idea. Hotspots are not hot and they are not fixed, yet the idea survives. Occam would not be pleased.4 It is the same kind of wishful thinking and self-delusion that we associate with Ptolemy , Piltdown, phlogiston, polywater and cold fusion.
Secondly, it is not based on sound physics. Experimental geologists have always recognized the importance of proper scaling, most notably Hans Ramberg of Sweden who built large centrifuges to scale gravity properly. This is in contrast to today's modelers who inject hot fluids into a static tank of cold fluid, or do comparable computer simulations (the Boussinesq approximation). The injection experiments, upon which modern plume cartoons, and research programmes, are based, ignore the effects of pressure on physical properties, scaling, the all important dimensionless Rayleigh number, and the distribution of radioactive elements.5,6
High Prandtl number flows in large bodies, with self-consistent thermodynamic properties and free slip boundary conditions do not behave as the text-book cartoons.13 Mantle stratigraphy, petrology and mineral physics cannot be ignored in mantle geodynamics,14 but invariably are . The mantle is not a pot-on-a-stove, heated from below system, with constant properties and no pressure effects. Neither is it an isothermal homogeneous halfspace with occasional injections of hot fluid. The mantle is strongly cooled from above and gently heated from within and hardly at all from below.
Narrow plumes, comparable to the dimensions of slabs (the upper thermal boundary layer), simply cannot form at the base of a compressed mantle where most of the heating is in the upper parts and which has rapid surface cooling.6 Large sluggish upwelling are expected, and seen in global tomography.5,6 Previous conflicts between geologists and physicists, for example Lord Kelvin and Sir Harold Jeffreys ( about the age of the Earth, continental drift and the permanence of mountains) also involved over-simplified physics. The idea of a fixed reference system on a deforming, convecting planet, with mobile plates has as little basis as Ptolemy's scheme of a fixed Earth.
Thirdly, the assumptions and assertions of geochemists are not fully understood by geologists. One basic assumption is that high 3He/4He ratios imply high 3He abundances (the ratio fallacy) rather than low U and Th (a property of the lithosphere and shallow olivine cumulates). This has led to the "primordial" undegassed deep plume source idea.7,8 Another assumption is that since midocean ridge basalts come from the upper mantle, ocean island basalts must come from the lower mantle (the Modus Moron Fallacy). Isotope geochemistry cannot determine the location or size of a "reservoir", which may simply be recycled crustal and sedimentary fragments. It is fluid dynamicists, theoretical physicists and isotope geochemists who are mainly responsible for the phoenix-like rebirths of the plume dogma, since it was first proposed. With a few notable exceptions, the geological community has accepted the claims of other specialists regarding the sources and depths of their volcanoes and has not asked why there are so many geological coincidences and paradoxes; why are volcanic chains usually on pre-existing lithosphere features, why are continental flood basalts associated with cratonic sutures and the sites of closed up oceans,10 why do volcanoes repeatedly occur at the same place in lithospheric coordinates, why should plates be considered as rigid, undeformable and homogeneous?9
Why are fracture zones, abandoned ridges, incipient plate boundaries, transform faults and sutures called "hotspot tracks"? How can real rocks avoid melting under the normal thermal conditions most geologists recognize as existing in the asthenosphere ? How can the upper mantle be isothermal, dry, subsolidus and homogeneous as assumed in plume calculations? Why do the geochemical characteristics of "hotspot" magmas look so much like continental crust and sediments?
The question should be, not "why are there 20 or so "midplate" volcanoes" ("unexplained by plate tectonics") but "why aren't there volcanoes everywhere?" The mantle is certainly hot enough and fertile enough. The answer seems to be answerable by geology; the thickness, strength and stress in the crust and lithosphere, and the properties of cracks and faults on a dynamic planet. These control the properties of plates and the locations of plate boundaries and incipient plate boundaries.11 Focusing on deep thermal plumes to "explain" almost everything has diverted attention away from the really exciting problems in global geology and plate tectonics. 11,12
The roles of time, scale, melting, crustal stress, proper scaling and mantle pressure, familiar to geologists, and essential for understanding plate tectonics, volcanoes and mantle dynamics, are notably lacking from the design of experiments and calculations, and discussions from both the plume advocates and skeptics . Many of these parameters are more important in geological processes than temperature. Attaching the word plume to what used to be called volcanoes has not helped.
References cited
1. Don L. Anderson, in: Craig, G.Y. & Hull, J.H. (eds) James Hutton — Present and Future. Geological Society, London, Special Publications, v. 150, p. 13-35 (1999)
2. Don L. Anderson,International Geology Review, v. 41, p.1051-1057 (1999)
3. Don L. Anderson, Geophys. Res. Lett., v. 27, no. 22, p. 3623-3626 (2000)
4. Don L. AndersonProceedings of the American Philosophical Society, v. 146, no. 1, p. 56-76 (2002).
5. Don L. Anderson, International Geology Review, 44, p. 97-116 (2002)
6. Don L. Anderson, Science, 293, p. 2016-2018 (2001)
7. Don L. Anderson, Earth & Planetary Science Letters, 193, 77-82 (2001)
8. Don L. Anderson, Int. Res. Lett., . 42, p. 289-311 (2000)
9. J. Favela, J., and Don L. Anderson, in Problems in Geophysics for the New Millennium, E. Boschi, G. Ekström, and A. Morelli, eds., Editrice Compositori, Bologna, p. 463-498(2000)
10. Don L. Anderson, in: The Core-Mantle Boundary Region, Michael Gurnis, Michael E. Wysession, Elise Knittle and Bruce A. Buffett, eds., American Geophysical Union, Washington, D.C., p. 255-271. (1998)
11. Don L. Anderson, Geology, 30, p. 411-414 (2002)
12. Don L. Anderson, in Plate Boundary Zones, Geodynamics Series 30, 411-425 (2002)
13. D.C. Tozer, Nature, 244, 398-400 (1973).
14. D.L. Anderson,, Theory of the Earth, Blackwell Scientific Publications, Boston, 366 pp. (1989 )