Search for a project

Constructing geochemical analogues of the asthenosphere: Earth's largest geochemical reservoir

Dr Jason Harvey (SEE), Dr Dan Morgan (SEE), Dr Thomas Mueller (SEE)

Contact email:

Project synopsis

Figure 1. Serpentinized oceanic lithospheric mantle. But how representative is abyssal peridotite like this of the convecting mantle as a whole, before it is hydrated? (from Bach et al., 2004)

It is becoming increasingly clear that the study of basalt composition does not always lead to an accurate assessment of the isotopic and chemical composition of its mantle source. Therefore, in order to understand mantle composition better, it is more prudent to collect direct evidence from the mantle itself. However, obtaining representative samples of the modern-day convecting mantle is problematic because it is overlain by the lithosphere, yet lithospheric mantle samples are the only examples of the mantle available. Drilled or dredged abyssal peridotite is often heavily altered by interaction with seawater and hydrothermal fluids. On the other hand, peridotite xenoliths, while exposed at the surface of the Earth, are often affected by metasomatism, melt-rock interaction and / or refertilization processes. This project will examine the various processes that have affected the composition of peridotites recovered from the Mid-Atlantic ridge, the Iberian passive margin and the Cape Verde Islands in order to (i) better understand the processes that can overprint the chemical and isotopic signatures preserved in the upper mantle at different tectonic settings, and (ii) assess peridotite from each of these provinces for its suitability as a chemical and isotopic proxy for the more voluminous, yet unreachable, asthenospheric mantle. The project will employ a combination of petrology and isotope geochemistry in order to achieve these objectives. Critically, lithophile (Sr-Nd-Pb) and siderophile (Re-Os) element isotope systems will be used to untangle the effects of metasomatism, melt-rock interaction and refertilization.


Two key questions can be posed: (i) Which of the chemical and isotopic heterogeneities observed in lithospheric mantle samples are representative of all of the mantle (i.e. the lithosphere and asthenosphere) and which are only present in the lithosphere (i.e. are unlikely to contribute to the composition of asthenosphere-derived basalts) and (ii) in which source of lithospheric peridotite (abyssal peridotite vs. peridotite xenoliths) is it easiest to “see through”  secondary chemical and isotopic “smudging” and provide a realistic representation of the composition of the asthenospheric mantle and the heterogeneities that are preserved within it?


Figure 2. Peridotite from oceanic basins (abyssal peridotite) while only recently separated from the asthenosphere is often extensively altered by seawater – peridotite interaction, obscuring the composition of primitive melt depleted mantle (from Bach et al., 2004)

For over half a century the composition of the Earth’s mantle has been determined largely by indirect methods. These range from geophysical measurements, which identify large volumes of the mantle that have distinctly different physical properties to their immediate surroundings, to the composition of basalts, which traditionally have been assumed to be representative of the mantle domains that produced them through partial melting. Apart from fortuitous instances where the mantle is exposed, for example in orogenic massifs, in ophiolites and as peridotite xenoliths (accidently transported to the surface in alkali basalts), the majority of what we think we know about the composition of the mantle has, to date, been derived indirectly. Experimental petrology and geochemical modelling tell us about the behaviour of incompatible trace elements during partial melting in the mantle, allowing the prediction that peridotite and the lavas that it produces have complimentary elemental compositions. Moreover, because isotope ratios of elements are not fractionated during high temperature processes, it is assumed that the isotopic fingerprint of a mantle reservoir is reproduced in the melts that it produces. However, there is increasing evidence to suggest that it is not necessarily safe to make the assumption that these fingerprints are faithfully transferred from mantle source to basalt – the fingerprint is somewhat “smudged” as there are domains of varying sizes (cm to km) in the mantle that do not correspond to the composition of basalts they are inferred to have produced. Unfortunately, the mantle samples that we have to work with are exclusively from the lithospheric mantle – the asthenosphere, which contributes to the majority of intra-plate basalt petrogenesis is not accessible for direct sampling. Dredged or drilled abyssal peridotite is often heavily hydrated and its composition drastically altered through interaction with seawater and / or hydrothermal fluids, whereas peridotite xenoliths can be affected by cryptic metasomatism, melt infiltration and refertilization – all of which potentially obscure the primary nature of the original peridotite that it represented when melt was first extracted from it.


Figure 3. Melt-rock interaction in peridotite from the sub-continental lithospheric mantle  overprints primary geochemical and isotopic melt depletion signatures. Only by “looking through” these secondary processes can we gain insight into the composition of Earth’s asthenosphere (from Harvey et al. (2010).

For this project Mid-Atlantic Ridge peridotites will be sampled from the extensive collections available to Harvey at Woods Hole Oceanographic Institution, Massachusetts and Scripps Institution of Oceanography, California. In addition, peridotite samples from the Iberian passive margin (Ocean Drilling Program Leg 173) will be also be selected from the IODP core repository in Bremen. These will be supplemented by field sampling of peridotite xenoliths from the Cape Verde Islands (in which the student will be involved), where primitive basalts have erupted through similar oceanic mantle to the other sample suites. The majority of the project will concern high-precision isotope geochemistry (Sr-Nd-Pb, Re-Os) to be performed in a clean laboratory environment at the University of Leeds. We are developing cutting edge techniques for the measurement of mantle sulphide grains and mantle minerals for e.g. Os and Pb isotopes in individual sulphide grains and high precision Sr and Nd isotope measurements on single clinopyroxene grains, and measurements of this nature will be a key element of the project. This work will complement ongoing work in this field already being undertaken at the University of Leeds by Harvey, Morgan and Mueller who have a strong track record in publishing high quality geochemistry research.

The ideal candidate

You will have a good degree in geology or a closely related area and have a strong interest in “hard-rock” sub-disciplines (igneous and metamorphic petrology, mineralogy, geochemistry) and preferably an interest in analytical techniques. Previous analytical experience would be an advantage, but is not essential. The project will mostly be laboratory based, but you will be expected to participate in fieldwork in the Cape Verde Islands


The successful applicant will be trained in a wide range of cutting edge geochemical techniques involving sample preparation, purification and analytical methods. The measurements made will make full use of the facilities available within the School of Earth and Environment (secondary electron microscope, electron probe microanalyser, inductively coupled plasma mass spectrometry (both with and without laser ablation) and thermal ionization mass spectrometry

Potential for high impact outputs

Challenging the fundamental tenet that mantle composition cannot be accurately assessed through the composition of the basalts that it produces has enormous potential to yield high impact outputs. Over the last few years the lead supervisor and co-workers have demonstrated that the Earth’s mantle is chemically and isotopically heterogeneous at cm to km scales and that this variability is not always expressed in the composition of basalt. The next stage of this research is to unequivocally separate the processes that introduce heterogeneity into lithospheric mantle from those that are likely affecting the inaccessible asthenospheric mantle i.e. which processes are the real control on geochemical and isotopic signatures during basalt petrogenesis and which, although clearly present in the lithospheric mantle, simply mask asthenosphere composition.

Further reading

Bach et al. (2004) Geochem, Geophys, Geosys 5 (9).

Burton et al. (2012) Nat Geosci 5, 570-573.

Harvey et al. (2011) Geochim Cosmochim Acta 75, 5574-5596.

Harvey et al. (2012) J. Petrol 53, 1709-1742.

Harvey et al. (2015) Geochim Cosmochim Acta 166, 210-233.

Related undergraduate subjects:

  • Geochemistry
  • Geoscience