Roadmap | The review process | Home

Comment on:

Subduction-triggered magmatic pulses: a new class of plumes?,

by C. Faccenna, T.W. Becker, S. Lallemand, Y. Lagabrielle, F. Funiciello & C. Piromallo


Monday, March 7, 2011, Don L. Anderson

It is no surprise that upwellings and melting are associated with plate boundaries and plate processes such as subduction. A variety of volcanic phenomena are associated with normal plate tectonics. Some of these are passive athermal upwellings associated with plate divergence or delamination, and some are triggered by volatiles released from slabs and by material displaced by slabs. Upwelling and melting are associated with ridges, arcs, backarc basins, rifts, thermal contraction, fracture zones, bending of plates, slab windows, craton edges and dikes. Slabs and dense delaminated lower crustal fragments sink into the transition region and displace material out. These are all top-down or plate tectonic processes and involve the upper mantle boundary layer. The general theory of plate tectonics goes far beyond plate kinematics and explains most if not all magmatic phenomena. The question is, do any volcanoes require a heated-from-below, deep, unstable boundary layer and narrow, thermally-buoyant upwellings?

Plumes were invented as an additional mechanism for creating volcanoes that are alternative to or independent of plate tectonics and normal mantle convection. They are basically Rayleigh-Taylor instabilities, of specified dimensions and temperature, that rise by their own buoyancy from a deep boundary layer. Fertile blobs are also created by plate tectonics and these can become buoyant as they warm up to ambient mantle temperature. They can be relatively fixed if they reside below the decoupling asthenosphere or they can be entrained and blown around in the mantle wind. In contrast to plates and slabs, which are basically creatures of the upper boundary layer of the mantle, plumes are hypothetical creatures of the lower boundary layer and were originally predicted to be hotter, more stationary, deeper and more vertical than upwellings due to plate processes; they were also predicted to be chemically distinct.

Shear-driven upwellings and displacement of transition zone material by slabs are non-thermal upwellings that are not Rayleigh-Taylor (RT) instabilities of a boundary layer. Buoyant magmas can hydraulically fracture the lithosphere but this is a normal process that does not require a hot, deep mantle plume. Mafic materials in the transition zone can warm up and become buoyant but these are also not strictly RT instabilities. Although any buoyant upwelling is technically a plume by fluid dynamic convention, a mantle plume has many  other characteristics involving depth, temperature, dimensions, fixed location, and chemistry (e.g., Anderson, 2007a, b, c).  There is currently much debate as to whether mantle plumes in the Morgan sense exist (e.g., Presnall & Gudmundsson, 2011) or whether midplate volcanoes and large igneous provinces can be explained in terms of plate tectonics, shallow boundary layer sources, and abandoned mantle wedges.

It is plausible that all OIB-type signatures originate as near-surface and lower crustal materials (Salters & Sachi-Kocher, 2010; Willbold & Stracke, 2010) that are placed in the mantle at island arcs, backarc basins and mantle wedges, and become metasomatised and sheared into the subplate boundary layer (e.g., Anderson, 2010). This would explain why MORB-like signatures are generally missing from isotope mixing arrays and why these arrays represent two-component mixing lines emanating from a common non-MORB focus (Hart et al., 1992). It would also explain why continental break-up, back-arc and far-back-arc basalts are often similar to OIB. It is common to attribute any bathymetric, chemical or low-velocity seismological anomaly to high temperatures or a “plume”. MORB temperatures and compositions are often considered the reference state of ambient mantle, but this need not be the case in sub-plate boundary layers (Anderson, 2010) or in a heterogeneous mantle.

Hypothetical plume types include fossil, dying, lateral, channelled, depleted, tabular (hot-line), pulsating, subduction, fluid fluxed, refractory, zoned, cavity, diapiric, starting, impact, incubating, incipient, splash, primary, secondary, satellite, strong, weak, tilted, parasite, thermo-chemical, asymmetric, stealth, shallow, mega-, super-, mini-, cacto-, headless, petit plumes and plumelets. Most of these are simply relabeling of plate tectonic and geological phenomena.

The plate model uses sedimentary, crustal and shallow mantle components and processes in the formation and extraction of materials from the upper mantle. In the plume hypothesis these materials are relabeled “plume components” and called PM, HIMU, FOZO, C, PHEM, EM (1 & 2), LONU, RUM, and HRDM, and attributed to deep primordial-, undegassed-,  lower-mantle-, and D” reservoirs, connected to the surface by narrow tubes. To label plate tectonic, shear-driven and shallow upwellings as “plumes” or “a new class of plumes” is simply to acknowledge that mantle plumes, in the canonical sense, are unnecessary. Fixity is one of the defining characteristics of mantle plume theory. The idea that plumes are not fixed, but are tilted and blown around by the mantle wind, is inconsistent with the idea that plumes are hot and rapidly rising entities that are independent of plate teconics and mantle convection. Fertile blobs in the upper mantle, extensive mantle wedges above slabs, and material displaced out of the transition region are results of plate tectonics and do not require mantle plume theory and deep mantle reservoirs.


  • Anderson, D.L., 2001. Topside Tectonics,  Tectonics?, Science, 293, 2016-2018.
  • Anderson, D.L. (2007a). New Theory of the Earth 2nd Edition. Cambridge: Cambridge University Press, 384 pp; doi: 10.2277/0521849594.
  • Anderson, D.L. (2007b). Seismic observations of transition zone discontinuities beneath ‘hotspot’ locations; Discussion. In: Foulger, G.R. & Jurdy, D.M. (eds.) Plates, Plumes and Planetary Processes. Geological Society of America, Special Paper 430, 131–134.
  • Anderson, D.L. (2007c). The Eclogite engine: Chemical geodynamics as a Galileo thermometer. In: Foulger, G.R. & Jurdy, D.M. (eds.) Plates, Plumes and Planetary Processes. Geological Society of America, Special Paper 430, 47–64.
  • Anderson, D.L., Hawaii, Boundary Layers and Ambient Mantle—Geophysical Constraints, J. Pet., published online December 2, 2010, doi: 10.1093/petrology/egq068.
  • Hart, S.R., E.H. Hauri, L.A. Oschmann, and J.A. Whitehead (1992), Mantle plumes and entrainment—Isotopic evidence, Science, 256, 517–520.
  • Presnall, D. and G. Gudmundsson, Oceanic Volcanism from the Low-velocity Zone without Mantle Plumes, J. Pet., in press, 2011, doi:10.1093/petrology/egq093
  • Salters, V.J.M., A. Sachi-Kocher, An ancient metasomatic source for the Walvis Ridge basalts, Chemical Geology, 273, 151-167, 2010.
  • Willbold, M., A. Stracke, Formation of enriched mantle components by recycling of upper and lower continental crust, Earth Planet. Sci. Lett., 297, 188-197, 2010

last updated 8th March, 2011