Skip Navigation | ANU Home | Search ANU | Directories
The Australian National University
Research School of Earth Sciences
Printer Friendly Version of this Document
Untitled Document

Mantle Evolution, Dynamical and Chemical, Earth and Venus

Geoffrey F. Davies and Andreea Papuc

Research School of Earth Sciences, Australian National University, Canberra, ACT 0200, Australia


Melting under a mid-ocean ridge in a heterogeneous mantle.

Numerical modelling of mantle dynamics is leading to important insights into the history of Earth and Venus.

A series of studies of how mantle convection stirs chemical tracers has established a quantitative basis for a new hypothesis regarding the abundance of trace elements in the mantle, including the enigmatic noble gases.  Because the mantle is heterogeneous, being a mixture of subducted oceanic crust and peridotitic mantle, the extraction of melt at mid-ocean ridges is expected to be inefficient.  A cartoon of melting under ridges is shown in the Figure.  The inefficient extraction implies that the abundance of incompatible trace elements in the mantle is higher than has been estimated in the past.  Geophysical constraints indicate the abundance is 2-3 times previous estimates.  This removes the need for a 'hidden' reservoir, clarifies the relationship between continental crust and the mantle, and helps to resolve a discrepancy between estimates of radioactive heating and models of the thermal evolution of the mantle.

The source of 'unradiogenic' helium, i.e. helium that has a low 4He/3He ratio, from Hawaii and other hostpots has been an enduring puzzle for which the new hypothesis offers a resolution.  Melting of the heterogeneous mantle is expected to produce a 'hybrid pyroxenite' that contains much of the mantle's complement of incompatible elements, including the noble gases.  It is also likely to be denser than average mantle, like subducted oceanic crust.  Numerical models have shown that such denser components tend to partially settle to the bottom, plausibly explaining the seismological D" region at the base of the mantle.  Whereas subducted oceanic crust is expected to contain little noble gas, the hybrid pyroxenite should contain substantial noble gas.  Furthermore the material in D" has a longer residence time, according to the numerical models, so it will degas more slowly, meaning the content of primordial 3He will be higher.  D" is already believed to be the source of mantle plumes, so the new hypothesis offers a straightforward explanation of mantle helium observations.  A simple quantitative model, based on results from numerical models, then successfully explains the helium, neon and argon observations from mid-ocean ridge basalts and oceanic island basalts.

These results, if substantiated, go far to reconciling mantle geochemistry with the dynamical picture of the mantle based on geophysical evidence and numerical modelling.  It has been unclear for at least three decades how this could be achieved.

Work reported last year on numerical models that yield episodic layering and overturns in Earth's early mantle is now being extended to Venus.  Venus was volcanically resurfaced about 500 Myr ago.  PhD student Andreea Papuc's project is to investigate whether conditions in Venus' mantle today are conducive to layering and overturn, as Earth's mantle was early in Earth history.  Layering occurs because of the 'basalt barrier' mechanism, in which subducted oceanic crust tends to accumulate at 660 km depth because it is buoyant between 660 and 750 km depth, but negatively buoyant at other depths.  Initial results show that overturns are indeed still likely in Venus, mainly because of its hotter surface and lack of plate tectonics.  Venus' slightly smaller size, gravity and mantle density also seem to favour layering, though it is not yet clear why.  The effects of different lithosphere strengths are yet to be investigated, and the goal is to do evolutionary models of Venus.  It will be important for understanding Venus' atmosphere and geochemistry to know whether and how often earlier resurfacing events might have occurred.

Davies, G. F. (2008) Inefficient melt extraction from a heterogeneous, mildly depleted mantle:  no hidden reservoir, Earth Planet. Sci. Lett., submitted.

Davies, G. F. (2008) Noble gases in the dynamic mantle, Earth Planet. Sci. Lett., submitted.