Motivation & Scope:
Mantle convection is the principal control on Earth's thermal, chemical and geological evolution. It is central to our understanding of the origin and evolution of tectonic deformation, the thermal and compositional evolution of the mantle and, ultimately, the evolution of Earth as a whole. Plate tectonics and volcanism are surface manifestations of mantle convection, which, in turn, control multiple surface processes, such as mountain building and sea level change. By transferring heat to the surface, mantle convection dictates the cooling of Earth's core and has a direct impact on the geodynamo. Understanding convection within Earth's mantle has recently been designated one of ten 'Grand Research Questions in the Earth sciences', by a US national academies report (De Paolo et al. 2008).
One of the most fundamental gaps in our understanding of mantle convection lies in the dynamics and geochemical expression of mantle plumes: buoyant upwellings that bring hot material from Earth's deep-mantle to the surface, generating large volcanic provinces and volcanic island chains, such as Hawaii. Mantle plumes play an important role in a range of processes, including continental rifting, super-volcanic eruptions, global mass extinctions, changes in oceanic gateways, ore and diamond genesis, and hydrocarbon generation (e.g. White & Lovell, 1997; Poore et al. 2006; Torsvik et al. 2010). Although our awareness of these phenomena is increasing, it remains unclear how their variable geochemical expression at Earth's surface (i.e. geochemical variations recorded in volcanic hotspot lavas) relates to the heterogeneous structure of their deep-mantle source. For example, at Hawaii, distinct isotopic compositions are observed at the Kea (northern) and Loa (southern) volcanic tracks: Kea volcanism exhibits 'average' mantle compositions, whereas Loa volcanism is more `enriched' (see Fig. 1a). Such systematic variations are attributed to an internal zonation of the plume conduit, which, ultimately, is believed to reflect much larger-scale chemical heterogeneities in the deep-mantle. However, relating surface hotspot observations to the nature of deep-mantle heterogeneity remains a challenge.
Fig 1: (a) Schematic Nd-Pb isotopic diagram of oceanic island basalts from Hawaii and Samoa - at Hawaii (Samoa), Loa (Malu) track volcanism is enriched, whereas Kea (Vai) track volcanism exhibits a more average composition; (b) These systematic variations are attributed to internal zonation within the underlying plume conduits, which, ultimately, reflect much larger-scale heterogeneity in the deep-mantle. As such, Weis et al. (2011) hypothesize that the southern side of the Hawaiian plume preferentially samples isotopically enriched material from a distinct reservoir (LLSVP) in the deep Pacific mantle. Such heterogeneity would then be preserved within the conduit during plume ascent, yielding the observed surface geochemical expression; (c) Although this hypothesis may be valid for Hawaii, it is inconsistent with observations from Samoa: Samoa lies on the southern margin of the Pacific LLSVP. The northern side of the Samoan plume should therefore preferentially sample enriched material, yielding an enriched Vai track. However, the opposite trend is observed. This project will utilize cutting-edge numerical models to decipher the message that plumes carry from Earth's lowermost mantle, by quantifying how deep-mantle heterogeneities are transported into a plume, and how these heterogeneities are mixed during plume ascent. This will allow us to relate, for the first time, geochemical observations from volcanic hotspots at Earth's surface to the thermo-chemical structure of Earth's lower mantle, under an Earth-like, fluid- dynamical framework.
Weis et al. (2011) hypothesize that the observed variations at Hawaii occur because the southern side of the Hawaiian plume preferentially samples isotopically enriched material, from a distinct, large-scale geochemical reservoir in the deep-mantle, beneath the Pacific (see Fig. 1b). This hypothesis makes two central assumptions: (i) that the large low shear-wave velocity province (LLSVP), observed by several seismological studies of the deep Pacific mantle (e.g. Ritsema et al. 2011), represents a chemically distinct structure; and (ii) that source-region heterogeneities are transported into and preserved (i.e. not mixed) within a plume conduit, as the plume rises from the deep-mantle to Earth's surface. We have recently demonstrated that the Pacific LLSVP is unlikely to represent a coherent, distinct chemical reservoir (Davies et al. 2012), whilst it remains unclear whether or not mantle plumes can preserve the heterogeneous structure of their source region. Furthermore, recent observations from the Samoan hotspot appear inconsistent with the Weis et al. (2011) hypothesis (see Fig. 1a/c): Samoa lies on the southern margin of the Pacific LLSVP and, accordingly, the northern side of the Samoan plume would be expected to preferentially sample isotopically enriched material. However, the opposite trend is observed, with the Malu (southern) track exhibiting more enriched compositions, when compared to the Vai (northern) track (Huang et al. 2011).
There is no doubt that plumes carry a message from Earth's lowermost mantle (e.g. Griffiths & Campbell, 1992). However, as illustrated by conflicting observations for Samoa and Hawaii, deciphering this message is challenging, since we do not yet have a realistic quantitative framework that explains: (i) how deep- mantle heterogeneities are transported into a mantle plume; and (ii) the extent to which such heterogeneities are mixed as a plume rises from the deep-mantle towards Earth's surface. This project will use a state-of-the-art computational approach to develop a quantitative model for how deep-mantle heterogeneities are incorporated into a plume and how these are mixed during plume ascent. This will allow us to relate geochemical observations from hotspot lavas at Earth's surface to the thermo-chemical structure of Earth's lower mantle, under a self-consistent, Earth-like, fluid-dynamical framework.
High-resolution and high-precision geochemical data, from hotspot locations around the globe, have already identified the problem. The proposed research will provide constraints from geodynamical models, which will allow us to understand what the geochemical data means. As such, the project will yield unrivaled insights into the nature of lower mantle heterogeneity and will play a key role in reconciling geophysical and geochemical constraints on mantle structure, which has been a long-standing goal for the Earth sciences.