olelog

Thermally driven upwellings in the mantle - so-called mantle plumes - are often envisioned to originate at the core-mantle boundary, from which they rise to create hot-spot volcanism. Clear geochemical signatures of the outer core in hot-spot lavas would prove the existence of deep-rooted mantle plumes, with important implications for large-scale mantle dynamics, and for decades mantle geochemists have tried to find such clear isotopic or chemical signatures. Different isotopes have been used in such studies. In my post on Volcanoes and Isotopes I concentrated on Helium isotopes. The platinum-osmium (Pt-Os) and rhenium-osmium (Re-Os) isotopic systems have, however, played a dominant role in these studies.

Earth's mantle was strongly depleted in osmium, platinum, and rhenium during core formation, because these elements tend to move into metallic phases. If the osmium concentration is much higher in the liquid outer core than in the surrounding solid mantle, mixing a little bit of outer-core material back into the mantle at the core-mantle boundary will change the local osmium isotopic composition to resemble that of the outer core. And if the osmium isotopic composition of the outer core is very different from that of the upper mantle, it might be detected in lavas brought to Earth's surface.

Osmium isotopic analyses of mantle-derived materials has, however, showed that the upper mantle is very heterogeneous. According to a report by Luguet et al. published in Science of 25 January 2008 (Enriched Pt-Re-Os Isotope Systematics in Plume Lavas Explained by Metasomatic Sulfides) there is no longer a need to invoke an outer-core input to explain the osmium isotopic compositions of mantle-derived materials. Their results may radically change the basis on which osmium isotopic compositions from mantle-derived materials are evaluated. That there is evidence for a very high degree of geochemical heterogeneity in the upper mantle is not only true for the osmium isotopic systems; it seems to be generally true and has important consequences for explaining the origin of isotopic "anomalies" in mantle-derived materials. The minerals in question are highly mobile in the mantle.

Geochemical heterogeneity is introduced into the mantle, for example, by subduction of sediments, oceanic crust, and lithosphere, and by melt extraction. Other intramantle mixing processes can also contribute to the creation of a range of geochemical components in the upper mantle. All these processes redistribute (fractionate) major and trace elements among different minerals, fluids, and melts, which in turn allows different components to evolve along divergent isotopic trajectories.

With an upper mantle as heterogeneous as the data by Luguet et al. suggest, it is difficult to imagine that isotopic signatures in oceanic basalts can be uniquely tied to the outer core. The debate about the existence and possible origins of deep-rooted mantle plumes will most likely have to be settled with geophysical methods.

With such a heterogenous mantle I think that we are still left with two fundamental questions. We can follow cold dense sinking subducted crust for a few hundred km down into the mantle, but what really happens with this material at greater depth (in the lower mantle)? And where (and how) do mantle plumes (if any) arise? Further speculations over these two questions may lead to theories like the one I wrote about yesterday in my post on Tectonic Plate Recycling.

If you want my personal opinion (but who cares), I still believe in mantle plumes, but find the theory I described yesterday too far fetched. Further studies are needed - as we so often read in the conclusions of (geological) scientific papers.

• http://www.sciencemag.org/cgi/content/full/319/5862/418
• http://www.sciencemag.org/cgi/content/full/319/5862/453

Everything on Earth can be explained in terms of 4 states/phases of matter - solid, liquid, gas, and plasma. In this sense the mantle is solid. From the way seismic waves travel through the Earth we know that the mantle is solid. P waves will travel and refract through both fluid and solid materials. S waves, however, cannot travel through fluids. S waves travel through the mantle, but not through the outer core, so the mantle is solid.

Chris at Highly Allochthonous had an excellent post on the solid state of the mantle titled Annoying misconceptions in Geology. But how solid is it after all. It is often described as plastic - the plastic mantle. Other relevant words may be elastic or soft. I find it a telling fact, that you don’t get earthquakes beneath something like 700 km, so down there the mantle cannot be that rigid.

According to an article published in Science of 25 January 2008 laboratory experiments suggest that the lower mantle is softer than previously thought. The lower mantle extends from about 660 km to 2900 km into Earth and sits atop the liquid outer core. Knowledge of this deep and inaccessible region is derived largely from seismic data. Pressures and temperatures are so brutal there that materials are changed into forms that don’t exist in rocks at the planet’s surface and must be studied under carefully controlled conditions in the laboratory. The pressures range from 230,000 times the atmospheric pressure at sea level (23 GPa), to 1.35 million times sea-level pressure (135 GPa). And the heat is equally extreme - from about 1,500° to 3,700° Celsius.

Changes in the electronic configuration of iron at these pressures and temperatures seem to alter the elastic behaviour, making the lower mantle "softer" than previously estimated. The results suggest that scientists may have to go back to the drawing board to model this region of the Earth.

Crowhurst et al. Elasticity of (Mg,Fe)O Through the Spin Transition of Iron in the Lower Mantle http://www.sciencemag.org/cgi/content/abstract/319/5862/451
• http://www.eurekalert.org/pub_releases/2008-01/ci-eg012208.php
• http://www.scientificblogging.com/news_releases/earth_getting_soft_in_the_middle
• http://www.livescience.com/environment/080124-soft-middle.html