Jun Korenaga
Dept. of Geology and Geophysics
Yale University
September 29, 2004
Recent global P-wave tomography by Princeton group [Montelli et al., 2004] appears to settle a long-standing debate over the depth extent of the Iceland plume. The Princeton model is based on finite-frequency tomography - so far the best methodology that can exploit the most of seismic information in a unified fashion. Now we have both P-wave and S-wave models [e.g., Ritsema and Allen, 2003] showing that the Iceland plume doesn't go any deeper than 670 km. The Iceland hotspot appears to have a shallow, upper-mantle origin.
This has a rather far-reaching implication beyond the dynamics of the currently active hotspot, because the Iceland hotspot has very close spatial and temporal connections to the flood basalt volcanism that took place during the opening of the North Atlantic (Greenland and British tertiary igneous provinces). It is most natural to consider the Iceland hotspot as some kind of a "surviver" of this large-scale continental breakup magmatism. So, if the Iceland hotspot results from a shallow upper-mantle process, how can we explain the origin of those flood basalt provinces? Can we still call for the impact of a mantle plume head? It seems very difficult to assume so. The formation of a plume head as large as several hundred kilometer wide requires sufficient "space", as a plume head grows only gradually as it rises through the mantle.
Shallow plumes have many more problems for the conventional picture of plume dynamics. Understandably, some people propose that the Iceland plume is actually deep but we just don't see it in tomography owing to some unusual interaction between a plume conduit and the surrounding mantle [e.g., Sleep, 2004]. Sticking to the conventional notion of mantle plumes, however, does not seem to be a good idea, at least for Iceland. There are a growing number of geophysical and geochemical data from the North Atlantic igneous province that grossly contradict with the prediction of a thermal plume theory (see Korenaga [2004] for summary). As far as observations are concerned, the Iceland hotspot is best explained as the result of compositional (not thermal) anomaly in the source mantle. The Icelandic mantle is intrinsically more fertile than normal; it is more Fe-rich and easier to melt.
This fertile mantle hypothesis for Iceland, originally proposed by Korenaga and Kelemen [2000], has recently gained its popularity [e.g., Foulger et al., 2004], but it has one very serious weak point: fertile mantle is denser than normal. How can it be rising? As noted by O'Hara [1975], it should be sinking instead. For the fertile mantle hypothesis to become a dynamically-plausible theory, it must be able to explain this density issue, which I expect will remain as one of major challenges in the geodynamics of LIP formation. Speculations are easy to do - what is lacking is a dynamically-consistent model. I myself have recently proposed one solution [Korenaga, 2004], which is however still too preliminary. Mantle convection with multi-scale chemical mixing has very rich dynamics with direct implications for terrestrial magmatism, in particular for large igneous provinces. Future research efforts for this complex dynamics are much needed to better understand the behavior of Earth's mantle.
References:
Foulger, G.R., J.H. Natland and D.L. Anderson, Iceland is fertile: The geochemistry of Icelandic lavas indicates extensive melting of subducted Iapetus crust in the Caledonian suture, J. Volc. Geotherm. Res., in review, 2004.
Korenaga J., Mantle mixing and continental breakup magmatism, Earth Planet. Sci. Lett., 218, 463-473, 2004.
Korenaga, J. and P.B. Kelemen, Major element heterogeneity in the mantle source of the north Atlantic igneous province, Earth Planet. Sci. Lett., 184, 251-268, 2000.
Montelli, R. et al., Finite-frequency tomography reveals a variety of plumes in the mantle, Science, 303, 338-343, 2004.
O'Hara, M. J., Is there an Icelandic mantle plume? Nature, 253, 708-710, 1975.
Ritsema, J., and R. M. Allen, The elusive mantle plumes, Earth Planet. Sci. Lett., 207, 1-12, 2003.
Sleep, N. H., Thermal haloes around plume tails, Geophys. J. Int., 156, 359-362, 2004.