Download Earth`s heterogeneous mantle: A product of convection

Chemical Geology 330–331 (2012) 274–299
Contents lists available at SciVerse ScienceDirect
Chemical Geology
journal homepage: www.elsevier.com/locate/chemgeo
Review article
Earth's heterogeneous mantle: A product of convection-driven interaction between
crust and mantle
Andreas Stracke ⁎
Max-Planck-Institut für Chemie, Postfach 3060, D-55020 Mainz, Germany
ETH Zürich, Institute of Geochemistry and Petrology, Clausiusstr. 25, 8092 Zürich, Switzerland
a r t i c l e
i n f o
Article history:
Received 23 February 2012
Received in revised form 8 August 2012
Accepted 10 August 2012
Available online 18 August 2012
Editor: K. Mezger
Keywords:
Mantle
Continental crust
Isotope geochemistry
a b s t r a c t
Ubiquitous heterogeneity in the Earth's mantle has been documented by numerous chemical and isotopic
analyses of oceanic basalts. Despite the ever-increasing amount of data, the way in which compositional heterogeneity is manifest in the Earth's mantle, as well as the processes leading to mantle heterogeneity remain
fundamental questions. The large amount of available isotope data in oceanic basalts shows that, statistically,
only two principal compositional vectors capture the essential features of the data. Care must be taken, however, when estimating the isotopic composition of mantle from basalt samples. This is because partial melting, and melt mixing during melt extraction leads to a biased representation and subdued compositional
variability in the basalts relative to their mantle sources. In both ridge and ocean island settings, for example,
erupted lavas are expected to be isotopically less depleted than the most depleted source components. Abyssal peridotites indeed range to much more depleted isotope compositions than mid ocean ridge basalts
(MORB). The extent of heterogeneity of the MORB mantle source, the depleted mantle, therefore depends
on the proportion, as well as differences in composition, age, and sampling of its various depleted and
enriched source components. While MORB data thus do not reflect the full extent of mantle heterogeneity,
the large amount of trace element and isotope data in ocean island basalts (OIB) suggests that enriched isotope signatures in OIB closely correspond to those of their average enriched mantle components. OIB can
therefore be used to trace the geologic reservoirs that exchange mass with the mantle and to identify the geological processes that introduce enriched material into the Earth's mantle. The generation and subduction of
oceanic plates into the deeper mantle, together with small amounts of lower and upper continental crust, appears to be the main process for mantle enrichment. Thereby, erosion and subduction of the lower continental crust accounts for a large part of the enriched isotope signatures in oceanic basalts. Recycling of the upper
continental crust, on the other hand, is inferred to be only a minor process, but required to explain the entire
spectrum of enriched OIB signatures. Hence a first order geologic process – the generation and subduction of
oceanic plates – accounts for the first-order heterogeneity of the Earth's mantle. Moreover, one of the main
processes for establishing the composition of the continental crust – erosion and recycling of the lower continental crust – is also one of the main processes for the generation of mantle heterogeneity. Overall,
large-scale chemical cycling between Earth's two major lithophile element reservoirs, the mantle and
the oceanic and continental crust, is responsible for mantle enrichment. Once introduced into the mantle,
the heterogeneous materials become stretched, reduced in size and distributed by mantle convection. The isotopic heterogeneity observed in melt inclusions and abyssal peridotites suggests that eventually, the heterogeneity of the mantle sources of oceanic basalts will exist at relatively small scales, certainly on the kilometer
scale of the melting region but perhaps even smaller. The way in which mantle heterogeneity is manifest
in the source of oceanic basalts is therefore directly related to the fluid dynamics of mantle convection,
whereas the timing, nature, and extent of crust–mantle interaction govern the differentiation and compositional evolution of the silicate Earth.
© 2012 Elsevier B.V. All rights reserved.
Contents
1.
2.
Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Oceanic basalts as tracers of mantle composition . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
275
276
⁎ Corresponding author at: Westfälische Wilhelms Universität, Institut für Mineralogie, Corrensstr. 24, 48149 Münster, Germany. Tel.: +49 251 83 33487; fax: +49 251 83 38397.
E-mail address: [email protected].
0009-2541/$ – see front matter © 2012 Elsevier B.V. All rights reserved.
http://dx.doi.org/10.1016/j.chemgeo.2012.08.007
A. Stracke / Chemical Geology 330–331 (2012) 274–299
2.1.
Systematics of MORB and OIB isotope arrays . . . . . . . . . . . . . .
2.1.1.
Phenomenological description . . . . . . . . . . . . . . . .
2.1.2.
Principal component analysis (PCA) . . . . . . . . . . . . . .
2.1.3.
Implications for distribution and sampling of mantle components
2.2.
Inherent ambiguity in oceanic basalt isotope systematics . . . . . . . .
2.3.
The influence of melt mixing on oceanic basalt isotope arrays . . . . . .
3.
The origin of mantle heterogeneity . . . . . . . . . . . . . . . . . . . . . .
3.1.
The depleted mantle (DM) . . . . . . . . . . . . . . . . . . . . . .
3.1.1.
How depleted is the DM? . . . . . . . . . . . . . . . . . . .
3.1.2.
How enriched is the DM? . . . . . . . . . . . . . . . . . . .
3.2.
The enriched mantle (EM) . . . . . . . . . . . . . . . . . . . . . .
3.2.1.
The correspondence of basalt and mantle EM components . . .
3.2.2.
A common continental heritage for EM-type mantle sources . .
3.2.3.
Quantitative recycling models . . . . . . . . . . . . . . . . .
3.2.4.
Mode and geodynamics of continental crust–mantle recycling . .
3.3.
Implications for continental crust evolution . . . . . . . . . . . . . .
3.4.
Alternative hypothesis for generating EM sources . . . . . . . . . . . .
3.4.1.
Selective recycling of different marine sediments . . . . . . . .
3.4.2.
Large-scale metasomatism of the oceanic mantle? . . . . . . .
3.5.
HIMU . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
3.5.1.
The distribution and sampling of HIMU mantle components . .
3.5.2.
Origin of the HIMU mantle components . . . . . . . . . . . .
3.6.
The age of mantle heterogeneity . . . . . . . . . . . . . . . . . . . .
4.
Synthesis and outlook . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Acknowledgments . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
1. Introduction
The main driving force for the chemical differentiation of the silicate
Earth is partial melting of the Earth's upper mantle. This process generates the oceanic, but ultimately also the continental crust, and Earth's
atmosphere (Gast et al., 1964; Allègre et al., 1987b; Hofmann, 1988).
Left behind is a residual mantle that is preferentially depleted in those
elements that make up the crust and atmosphere, the so-called incompatible elements which are difficult to incorporate into the mineral
structure of common mantle minerals (e.g., Hofmann, 1988). This depleted mantle is, however, continuously re-enriched in incompatible elements and the chemical composition and overall heterogeneity of the
Earth's mantle have evolved progressively, possibly from the earliest
stages of the Earth ~4.5 Ga ago to the present day (e.g., Harper and
Jacobsen, 1992; Caro et al., 2003; Boyet and Carlson, 2005; Harrison et
al., 2005).
As of yet, no consensus has been reached about what the most important processes for re-enrichment of the mantle are. Among the great diversity of persisting models, the most often invoked are recycling of
oceanic and continental crust (Hawkesworth et al., 1979; White and
Hofmann, 1982; Zindler and Hart, 1986; Weaver, 1991; Chauvel et al.,
1992; Stracke et al., 2003a; Willbold and Stracke, 2006, 2010), delamination and foundering of the subcontinental lithosphere (e.g., McKenzie and
O'Nions, 1983; Mahoney et al., 1989, 1996; Milner and LeRoex, 1996;
Douglass et al., 1999; Douglass and Schilling, 2000; Lustrino et al., 2000;
Doucelance et al., 2003; Hanan et al., 2004), and various types of mantle
metasomatism (e.g., Zindler et al., 1979; Roden et al., 1984; Hart, 1988;
Niu and O'Hara, 2003; Donnelly et al., 2004; Workman et al., 2004; Pilet
et al., 2005; Salters and Sachi-Kocher, 2010).
The degree to which these processes influence the compositional
evolution of the silicate Earth remains a matter of active debate. It has
become clear, however, that with the possible exception of intramantle metasomatism, re-enrichment of the mantle principally occurs
by large-scale chemical cycling between different silicate reservoirs on
Earth: the continental and oceanic crust and associated lithospheres,
and the Earth's mantle. Element cycling between these different global
reservoirs links the continuing compositional evolution of the Earth's
mantle to the global differentiation and compositional evolution of
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
275
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
276
276
278
279
280
282
283
283
283
284
285
285
285
285
287
288
289
289
289
291
291
291
292
293
294
294
our planet. Understanding the driving forces responsible for the current
state of mantle heterogeneity is therefore key for improving our basic
understanding of Earth's workings.
Owing to the paucity of direct mantle samples, the composition of the
mantle is mostly studied through basalts erupted on the ocean floor.
Oceanic basalts are large-scale partial melts from the Earth's mantle that
escape contamination by the continental crust and hence indirectly
record mantle composition. Numerous isotope and trace element studies of oceanic basalts show that the Earth's mantle is heterogeneous
(e.g., Gast et al., 1964; Tatsumoto et al., 1965; Hedge, 1966; Tatsumoto,
1966; Hart et al., 1973; White and Schilling, 1978; Allègre, 1982;
White, 1985; Zindler and Hart, 1986; Hofmann, 1997; Stracke et al.,
2005) and that most mantle sources consist of complex assemblages
of two or more components with isolated long-term chemical evolution,
on both global and local scales. Traditionally, the observed range of isotopic compositions in oceanic basalts is assumed to directly reflect those
of the underlying mantle. A long-prevailing view is that the Earth's mantle consists of a restricted number of large-scale reservoirs: a depleted
mantle and about 3–4 enriched reservoirs that mix to produce the isotopic variability observed in oceanic basalts (e.g., Zindler and Hart, 1986).
Recent studies show, however, that isotopic heterogeneity in oceanic
basalts is observed even at the μm-scale of melt inclusions (Saal et al.,
1998; Shimizu et al., 1998; Saal et al., 2005; Jackson and Hart, 2006;
Maclennan, 2008b; Paul et al., 2011; Sobolev et al., 2011). This smallscale heterogeneity is evidence for a highly heterogeneous mantle, perhaps even on a sub kilometer scale. Furthermore, comparison of melt inclusions with erupted basalts shows that the elemental and isotopic
heterogeneity in the melt inclusions is greater than in the erupted basalts (see references above and below). This observation indicates that
melts are mixed to different extents during the melting and melt aggregation process. Erupted melts are therefore generally incomplete mixtures of small degree partial melts produced at variable depths in the
melting region (Sobolev and Shimizu, 1993; Nielsen et al., 1995;
Shimizu, 1998; Morgan and Morgan, 1999; Sours-Page et al., 1999; Slater et al., 2001; Maclennan et al., 2003a,b; Stracke et al., 2003b; Ito and
Mahoney, 2005; Laubier et al., 2007; Maclennan, 2008a,b; Stracke and
Bourdon, 2009; Shorttle and Maclennan, 2011; Waters et al., 2011;
Koornneef et al., 2012).
276
A. Stracke / Chemical Geology 330–331 (2012) 274–299
Hence before attributing the isotopic signatures of the basalts directly to those of the mantle, it needs to be understood how melting
samples and averages heterogeneous source components. The key
challenge thereby is to disentangle the interrelated effects of source
composition, pressure and temperature of melting, as well as the
style of melting and melt mixing during melt extraction. Only then
can we infer the distribution of heterogeneous source components,
their size, mineralogical, chemical and isotopic composition, which
is a prerequisite for understanding the origin of the various heterogeneous materials in the Earth's mantle. Ultimately this requires
combining observations from a range of geochemical parameters
(isotope and major/trace element data) with information from quantitative models, petrology, geodynamics and geophysics. Only by
attempting to fuse all these observations into a self-consistent
model can we expect to gain a better understanding about the origin
of mantle heterogeneity and thus about the large-scale mass cycles
that govern the differentiation of the silicate Earth.
In the following, the extent and underlying systematics of the isotopic
variation observed in oceanic basalts will be discussed by defining principal directions that account for the observed variability (Section 2.1).
While this exercise provides some first-order information about mantle
heterogeneity, the isotope systematics of oceanic basalts can certainly
be interpreted in more than one way (Section 2.2). This ambiguity inherent in oceanic basalt isotope systematics, results, at least in part, because
partial melting introduces a sampling bias between the isotopic composition of the mantle and those of the erupted lavas (Section 2.3). A first step
in the attempt to infer mantle from basalt composition thus has to assess
the extent to which the basalts represent source signatures, for example
by exploiting the isotope-trace element relationships as demonstrated
in Section 2.3. Especially for unraveling the extent of depletion in the
Earth's mantle by investigating mid ocean ridge basalts (MORB), this sampling bias results in differences between the isotopic heterogeneity present in the source and that observed in MORB (Section 3.1). The full extent
of source heterogeneity is thus sometimes obscured, depending on how
partial melts from the various source components mix during melting
and melt extraction. Only the enriched isotope and incompatible element
ratios in ocean island basalts (OIB) correspond closely to those of their
average enriched source components (Sections 2.3 and 3.2). The OIB isotope signatures can thus be utilized to deduce possible processes that
account for both the composition, variability and covariance between
the different enriched isotope and trace element parameters in OIB
(Sections 3.2–3.4). A model that explains mantle heterogeneity as a result
of large-scale chemical cycling between Earth's two major lithophile element reservoirs – the oceanic and continental crust and mantle – will
be presented in Section 3.2, before discussing alternative, or additional
processes for generating the entire spectrum of isotopic heterogeneity
in oceanic basalts (Section 3.4). The model advocated here presents a
simple conceptual framework that tightly links continental crust and
mantle evolution. Hence advancing our understanding about the processes responsible for mantle heterogeneity is crucial for improving
our knowledge about the large-scale differentiation of our planet.
2. Oceanic basalts as tracers of mantle composition
2.1. Systematics of MORB and OIB isotope arrays
2.1.1. Phenomenological description
The isotopic composition of oceanic basalts is highly heterogeneous (e.g., Gast et al., 1964; Tatsumoto et al., 1965; Tatsumoto,
1966; Hart et al., 1973; White and Schilling, 1978; Allègre, 1982;
White, 1985; Zindler and Hart, 1986; Hofmann, 1997; Stracke et al.,
2005). The observed heterogeneity, however, is not random, but rather varies in a systematic fashion. In 2D and 3D isotope ratio diagrams,
the isotopic variation in oceanic basalts on a local scale, for example
for basalts from one ocean island or ridge segment, is usually in
form of approximately linear trends. These trends point to a limited
number of directions. For descriptive purposes, the entire isotope
space can thus be defined with a few points corresponding to the
most extreme isotopic signatures of the major directions observed
(e.g., Zindler et al., 1982; Zindler and Hart, 1986). It is convenient to
think of these extreme compositions in terms of components with a
characteristic isotopic composition. Therefore, unless clearly stated
as “mantle component”, the term component is used here for describing basalts with a specific isotopic composition. This distinction is motivated by the observation that it is not straightforward to equate
basalt and mantle composition (see discussion in the following sections). Whenever the term “mantle component” is used in the following, this is done in a purely descriptive sense to denote mantle that is
homogeneous on the scale of partial melting (Zindler and Hart, 1986),
but more importantly, without any further connotation regarding its
physical size and properties, mineralogic composition, genetic origin,
age, or location and distribution in the mantle.
Depleted mantle (DM) and FOZO versus PREMA. First-order isotopic
variation in oceanic basalts is observed between the isotopically
most depleted compositions found exclusively in MORB (the Depleted Mantle component = DM) and a more enriched component characterized by high Pb but intermediate Sr, Nd, and Hf isotope ratios
(Fig. 1). This component is similar to “FOZO”, a mantle component
postulated by Hart et al. (1992; FOcal ZOne) to explain the apparent
convergence of ocean island basalt (OIB) trends in two or three
dimensional isotope ratio space. According to Hart et al. (1992),
“FOZO” is an internal component of the oceanic basalt trends, which
could represent a common lower mantle component. It is debatable,
however, whether or not OIB trends really do converge into a restricted region in two or three dimensional isotope ratio space
(e.g., Stracke et al., 2005). Whether there is any justification for
postulating a component internal to the oceanic basalt arrays
(Zindler and Hart, 1986; Farley et al., 1992; Hart et al., 1992; Hanan
and Graham, 1996), therefore, is a matter of debate.
In contrast to Hart et al. (1992), Stracke et al. (2005), used “FOZO”
with different connotations. These authors assigned no unique composition to “FOZO”, but rather assigned a range of compositions
extending from MORB with high 206Pb/ 204Pb ratios to even more
enriched compositions (e.g., Fig. 1e, g). Some OIB form trends parallel
to the array between DM and “FOZO”. Hence systematic first-order
variation along the DM-“FOZO” array is observed in basalts independent of their tectonic setting and “FOZO” appears to be present in
many MORB and OIB sources (see also Section 2.1.3). Stracke et al.
(2005) therefore interpreted the “FOZO” signature to represent a
range of ubiquitously present and isotopically enriched mantle components in a variously depleted mantle.
The above reasoning, however, renders the acronym “FOZO”
inappropriate. Note that Zindler and Hart (1986) used “PREMA” for
PREvalent MAntle to describe the “high frequency of Nd and Sr isotopic compositions at about 0.5130 and 0.7033”. Hence it appears more
feasible to revive the dormant, but suitable acronym “PREMA” than to
continue using the term “FOZO”. In the following, “PREMA” refers to a
range of isotope compositions at, or in extension of the enriched end
of the global MORB array, as highlighted in Fig. 1. In Section 3.1, the
discussion returns to the possible origin of the “PREMA” signatures
as defined here and previously by Zindler and Hart (1986).
Enriched mantle (EM). Most OIB trends originate from intermediate
locations along the DM-PREMA array and define a series of trends,
directed toward more enriched isotopic compositions with higher
87
Sr/ 86Sr and lower 143Nd/ 144Nd and 176Hf/ 177Hf isotope ratios and
a range of Pb isotope ratios within the limits of the DM-PREMA
range (Enriched Mantle components = EM; Fig. 1e, g, h, i). These
EM-type OIB account for most of the observed Sr–Nd isotopic variability in oceanic basalts (Fig. 1a). OIB with EM affinity are further
characterized by higher 208Pb/ 204Pb and 87Sr/ 86Sr, but lower 143Nd/
A. Stracke / Chemical Geology 330–331 (2012) 274–299
144
Nd and 176Hf/ 177Hf ratios for given 206Pb/ 204Pb relative to basalts
on the DM-PREMA array (White, 1985; Hart et al., 1986; Zindler and
Hart, 1986; Hart, 1988; Stracke et al., 2003a; Stracke et al., 2005). In
plots of 143Nd/ 144Nd versus 87Sr/ 86Sr and 87Sr/ 86Sr ( 143Nd/ 144Nd)
versus 206Pb/ 204Pb (Fig. 1a, e, g) the Pitcairn and Samoa trends define
the extremes of an array of trends directed toward different EM components. In both the Pitcairn and Samoan basalts, Sr and Nd isotope
ratios are inversely correlated, but for a given 87Sr/ 86Sr ratio, the
143
Nd/ 144Nd ratios in Samoan basalts are markedly higher than
those in basalts from Pitcairn (Fig. 1a). In addition, 87Sr/86Sr ratios in
Samoan basalts increase steadily for relatively constant 206Pb/204Pb ratios, while in the Pitcairn basalts, 206Pb/204Pb isotope ratios decrease
systematically with increasing 87Sr/ 86Sr ratios, i.e., Sr and Pb ratios are
negatively coupled (Fig. 1e). There are a range of OIB trends that project
from endpoints intermediate between the enriched ends of the Samoa
and Pitcairn trends to different locations along the DM-PREMA array
(e.g., basalts from the islands Tristan da Cunha and Gough). Note,
277
however, that that there are no trends transverse to those of Pitcairn
and Samoa (e.g., Zindler and Hart, 1986; Hart, 1988).
Zindler and Hart (1986) suggested on the basis of a more limited
data set that the entire range of distinct EM components observed
in the basalts represents only two different EM-type mantle components, EM-1 and EM-2, whose compositions directly correspond to
the most enriched signatures observed in the Pitcairn and Samoa
basalts (e.g., White, 1985; Zindler and Hart, 1986; Hart, 1988; Hart
et al., 1992). According to Zindler and Hart (1986) each individual
endpoint of the EM-type OIB trends between those of Pitcairn and
Samoa (e.g., Tristan da Cunha, Fig. 1) represents a specific mixture
of the EM-1 and EM-2 mantle components. With the large amount
of isotope data now available, it has become apparent that all individual EM trends together form a continuous array that covers the entire
area between the Pitcairn and Samoa trends (e.g., Fig. 1e, g, h;
Willbold and Stracke, 2006, 2010; Class et al., 2009). Thus, each EM
basalt trend is representative of an EM mantle component with a
Fig. 1. Diagrams showing the systematic isotopic variability in global mid ocean ridge (MORB) and ocean island basalts (OIB). a) 87Sr/86Sr versus 143Nd/144Nd, b) 143Nd/144Nd
versus 176Hf/177Hf, c) 206Pb/204Pb versus 207Pb/204Pb, d) 206Pb/204Pb versus 208Pb/204Pb, e) 206Pb/204Pb versus 87Sr/86Sr, f) 87Sr/86Sr versus 176Hf/177Hf, g) 206Pb/
204Pb versus 143Nd/144Nd, h) 206Pb/204Pb versus 176Hf/177Hf, i) (208Pb/206Pb)* versus 87Sr/86Sr, and k) (208Pb/206Pb)* versus 206Pb/204Pb. On a local scale, for example
on the scale of one ocean island or ridge segment, the basalts form linear trends. The range of compositions indicated for “PREMA” (“FOZO”) is approximate and may extend to both
more depleted and enriched compositions. (208Pb/206Pb)* is the time-integrated Th/U ratio defined by Galer and O'Nions (1985) as (208Pb/206Pb)* = (208Pb/204Pbmeasured−
29.476) / (206Pb/204Pbmeasured − 9.307). The data compilation is an updated version of the one given in Stracke et al. (2003a) and is provided in the supplementary materials
(Supplementary Table 1).
278
A. Stracke / Chemical Geology 330–331 (2012) 274–299
Fig. 1. (continued).
unique and characteristic composition and a more conservative interpretation thus is to treat all EM components as part of one family with
a range of EM components. Whether all these EM components could
ultimately derive from only two different mantle components, however, remains to be determined. Hence although the long-standing
EM-1 and EM-2 dichotomy is useful for describing the distribution
of EM basalt data in multi-dimensional isotope ratio space, it may
be misleading for gaining information about the origin of EM sources
(Willbold and Stracke, 2006, 2010). Note, in this respect, the diversity
of models advocated to date (e.g., McKenzie and O'Nions, 1983; Hart,
1988; Weaver, 1991; Chauvel et al., 1992; Gasperini et al., 2000;
Tatsumi, 2000; Workman et al., 2004; Lustrino, 2005; Willbold and
Stracke, 2006, 2010; Jackson et al., 2007).
2.1.2. Principal component analysis (PCA)
Purely on a phenomenological basis (Figs. 1 and 2), it was argued
above that the systematic isotope variation in oceanic basalts is
captured by a limited number of directions in 2D and 3D isotope
ratio diagrams: one parallel to the DM-PREMA array and an array of
trends toward the EM components in between the Pitcairn and
Samoa trends. In the following, it will be tested whether these main
directions identified on a descriptive basis, with somewhat subjective
criteria, coincide with those resulting from a rigorous statistical treatment of the data.
Principal component analysis (PCA) is a mathematical method
that reduces the dimensionality of the data while retaining most of
the variation in the data. It accomplishes this reduction by identifying
A. Stracke / Chemical Geology 330–331 (2012) 274–299
vectors, the principal components, along which the variation in the
data is maximal (Ringner, 2008). Mathematically, the principal components are the eigenvectors of the variance–covariance matrix
formed by a given data set, in this case a large set of Sr, Nd and Pb isotope data of MORB and OIB (n = 4013; Figs. 1 and 2; Supplementary
materials). PCA reveals the covariance structure of this multivariate
data set (5 variables and n = 4013 samples) by reducing the data
set to a smaller number of linearly independent vectors, the principal
components; in this case 5 principal components for 5 variables. PCA
therefore captures the important features inherent in the isotopic
variability of oceanic basalts, but with only those few variables that
are necessary to describe the entire isotopic range. Hence PCA serves
to recognize patterns in the isotope systematics of oceanic basalts
(e.g., Allègre et al., 1987a; Hart et al., 1992; Albarède, 1995).
Applying PCA to a large set of Sr, Nd and Pb isotope data from MORB
and OIB (n= 4013), shows that 94.3, 97.2, and 99.2% of the variance in
the oceanic basalts isotope data is explained by the first two, three or
four principal components, respectively (PC1–4). Tables 1a and 1b as
well as Fig. 3 show that the first principal component (PC1) is dominated by the Pb isotopes and that the second and third, PC2 and PC3, are
controlled by the inverse correlation between the Sr and Nd isotope ratios. 87Sr/86Sr is negatively correlated with 206Pb/204Pb for PC2, whereas
87
Sr/ 86Sr is not correlated with 206Pb/204Pb for PC3. Fig. 3 shows that
PC1 corresponds to the DM-PREMA direction and that PC2 and PC3
are similar to the directions defined by the Pitcairn and Samoa EM
trends. PC4 is more ambiguous, but may account for HIMU signatures.
Hence, PCA confirms that two main vectors, one along the DM-PREMA
array and one similar to the Pitcairn EM trend, account for most
(94.4%) of the isotopic variability in oceanic basalts. Additional, but
overall minor, variation is introduced by Samoa-like EM and by the
HIMU components (PC3 and PC4).
Previous studies using PCA, or related techniques, have reached
similar conclusions with respect to the number and statistical significance of the main components necessary to explain the isotopic variability in oceanic basalts (e.g., Allègre et al., 1987a; Hart et al., 1992;
Albarède, 1995; Iwamori and Albarède, 2008; Iwamori et al., 2010).
For further details of the statistical methods, the influence of using a
different combination of Sr–Nd–Pb isotope ratios, or of adding another isotope ratio such as 176Hf/ 177Hf, the reader is referred to these
previous studies.
0.706
0.708
0.705
0.707
0.704
0.706
0.703
0.705
0.704
22
0.51 24
0.51
14
126
3
Nd 0.5 5128
.
0
/ 14
30
0.51
4
Nd 0.5132
34
0.51
Sr / 86Sr
0.707
87
87
Sr / 86Sr
0.708
0.703
17
18
19
206
21
20
Pb /
204
279
Table 1a
Results of principal component analysis of MORB and OIB data.
Principal component
Score [%]
Cumulative score [%]
PC1
PC2
PC3
PC4
62.8
31.5
2.9
2.0
62.8
94.3
97.2
99.2
2.1.3. Implications for distribution and sampling of mantle components
Association and distribution of mantle components. Most MORB fall on
the DM-PREMA array, although some MORB, mostly from the Indian
and South Atlantic Oceans, depart from the DM-PREMA array toward
EM components similar to that of the Pitcairn trend. So far, MORB
with EM signatures similar to Samoa have not been observed (see
Figs. 1 and 2), and this EM component therefore seems indigenous
to a few oceanic hot spots (e.g., Samoa, Marquesas, Society). Some
OIB trends either follow the DM-PREMA array, but most depart
from it toward a variety of EM components (see discussion above).
Most continental volcanic rocks also follow trends from PREMA toward the EM space (Fig. 4; see also, Wörner et al., 1986; Lustrino
and Wilson, 2007). HIMU basalts appear to be a special group of
OIB, which mainly occur at two intra-oceanic localities, St. Helena in
the Atlantic Ocean and the Cook–Austral Island chain in the South Pacific Ocean (HIMU signatures have also been observed in basalts from
the Chatham islands in the SW Pacific) (Panter et al., 2006) and Mt.
Erebus in Antarctica (Sims et al., 2008).
At the Cook–Austral Islands, basalts with HIMU signatures occur in
close association with EM-type basalts. Overall, OIB from the Cook–
Austral Islands form a broad trend from HIMU to Pitcairn-like EM
compositions in Fig. 1e (samples labeled “Austral Cook: EM”). This
trend is transverse to other EM OIB trends and suggests that HIMU
and EM components may be part of the same mantle source
(Chauvel et al., 1992; Hauri and Hart, 1993; Hémond et al., 1994;
Chauvel et al., 1997; Dostal et al., 1998). However, Fig. 1a shows
that basalts from the Cook–Austral Islands either form trends
between EM and the enriched end of the DM-PREMA array (i.e.,
PREMA; samples labeled “Austral Cook: EM”), or between HIMU and
PREMA (samples labeled “Austral Cook: HIMU”), but do not form
trends between HIMU and EM (Fig. 1a). Thus, HIMU appears together
with (DM-)PREMA and in fact, both HIMU and DM-PREMA isotope
signatures can occur in basalts from the same island during different
episodes of volcanism (e.g. Rurutu, Chauvel et al., 1997). HIMU and
EM components, on the other hand, do not appear to be directly associated, which makes HIMU different from all other components.
In summary, the DM-PREMA and Pitcairn-like EM components
(PC1 and PC2) are observed in basalts from different tectonic settings
(MORB, OIB, continental volcanics). These components account for
most (94.3%) of the isotopic variability in oceanic and continental basalts and hence appear to represent an ubiquitous assemblage of
mantle components. Others, as for example the Samoa-like EM component and the HIMU component (PC3 and PC4) are responsible for
the remaining variability (4.9%) and seem representative of mantle
components with a more limited distribution.
22
Pb
Fig. 2. Three-dimensional diagram of 206Pb/204Pb versus 143Nd/144Nd and 87Sr/86Sr in
global mid ocean ridge (MORB) and ocean island basalts (OIB) show that the isotopic
variation in oceanic basalts is systematic. On a local scale, for example on the scale of
one ocean island or ridge segment, the basalts form of linear trends or vectors. The
data compilation is an updated version of the one given in Stracke et al. (2003a) and
is provided in the supplementary materials (Supplementary Table 1). Symbols as defined in Fig. 1.
Table 1b
Eigenvalues of the variance covariance matrix of MORB and OIB data.
86
86
Sr/ Sr
Nd/144Nd
Pb/204Pb
207
Pb/204Pb
208
Pb/204Pb
143
206
PC1
PC2
PC3
PC4
0.28
−0.37
0.47
0.52
0.55
−0.66
0.56
0.41
0.25
0.13
−0.68
−0.69
−0.10
−0.17
0.13
0.12
0.05
0.51
−0.80
0.28
280
A. Stracke / Chemical Geology 330–331 (2012) 274–299
-0.4
-0.2
0
0.2
0.4
0.6
0.70 9
PC3
0.6
Sr/ 86Sr
PC4
87
Sr / 86Sr
87
0.2
0
0.70 6
0.70 5
0.70 4
PC1
17
18
19
20
21
22
0.5134
PC1
0.2
0.5132
DM
b
PC4
0
-0.2
143
-0.4
-0.6
PREMA
0.5130
Nd/ 144Nd
Nd / 144Nd
HIMU
DM
0.70 2
143
PREMA
0.70 3
-0.4
0.4
EM
0.70 7
0.4
-0.2
a
0.70 8
PC2
0.5128
HIMU
0.5126
PC2
PC3
0.5124
EM
0.5122
0.6
17
18
19
20
21
22
PC1
206
Pb/ 204Pb
PC4
0.2
PC3
0
208
Pb / 204Pb
0.4
PC2
-0.2
-0.4
-0.4
-0.2
0
0.2
0.4
0.6
206
Pb / 204Pb
Fig. 3. Principal components depicted in diagrams of 206Pb/204Pb versus 87Sr/86Sr,
143
Nd/144Nd and 208Pb/204Pb. Note that isotope ratios are expressed as the deviation
from the mean scaled to the standard deviation of the mean. The data compilation is
an updated version of the one given in Stracke et al. (2003a) and is provided in the
supplementary materials (Supplementary Table 1). For the PCA only data with all 5
isotope ratios have been selected (n = 4013). Symbols as defined in Fig. 1.
Sampling of mantle components by partial melting. Fig. 5a shows that
MORB with enriched isotope signatures, i.e., either PREMA-like with
intermediate 143Nd/ 144Nd and high 206Pb/ 204Pb up to about 19.5, or
EM-like with low 143Nd/ 144Nd and 206Pb/ 204Pb, also have systematically higher La/Yb ratios. Note that owing to the large contrast in incompatibility between La and Yb, the observed La/Yb variability is
primarily due to variable degrees of partial melting (see also
discussion in Section 2.3; data sources for MORB are given in Supplementary Table 2). Hence enriched components in the depleted mantle (DM) are sampled preferentially at low degrees of partial
melting (high La/Yb ratios).
Further support for this assertion comes from the isotope systematics of OIB and continental volcanic rocks (Fig. 4). Both OIB and continental volcanics form trends from the PREMA end of the DM-PREMA
array (e.g., at 206Pb/ 204Pb > 18.5) toward the Pitcairn-like EM space
Eifel
Siebengebirge
Vogelsberg
Westerwald
Hessian Depression
Rhön
Urach/Hegau
Iblean basin
Etna
Tyrrhenian Sea
Southern Sardinia
Pantelleria
Northeast Spain
Southeast Spain
Massif Central
Poland
Pannonian Basin
Carpathians
Fig. 4. Tertiary continental European volcanics in diagrams of a) 206Pb/204Pb versus
87
Sr/86Sr and b) 206Pb/204Pb versus 143Nd/144Nd. Also shown in gray are global
MORB and OIB data from Figs. 1 and 2. All continental European volcanics from trends
from the enriched (PREMA) end of the DM-PREMA array and depart toward enriched
isotope compositions similar to those observed in Pitcairn-like OIB (compare to
Figs. 1 and 2). Data for European continental volcanics are taken from the compilation
provided by Lustrino and Wilson (2007).
(Figs. 1, 2, and 4; Wörner et al., 1986; Lustrino and Wilson, 2007).
Owing to the thick lithospheric lid under continents, continental volcanics are often generated at higher average pressure and are similar
or smaller degree melts than OIB. Note that isotope trends of continental volcanic rocks rarely progress from PREMA toward more
DM-like compositions with increasing degree of partial melting.
Hence MORB formed by low degrees of partial melting, OIB, and continental volcanics all preferentially sample the PREMA components of the
depleted mantle (DM). These ubiquitous PREMA mantle component(s)
are thus likely to have different physico-chemical properties compared
to the isotopically more depleted DM component(s).
2.2. Inherent ambiguity in oceanic basalt isotope systematics
In the preceding sections, the main patterns in the basalt isotope
data could be identified. It should be noted, however, that this characterization has not equated basalt and mantle components and has
refrained from speculating about the possible physical size and
A. Stracke / Chemical Geology 330–331 (2012) 274–299
41
0
PREMA
DM
PREMA
20
17.0
15
20
PREMA
3
Step 2 : solid-state mixing of
EM with PREMA component
3
EM-1
Step 3 : solid-state mixing with DM and
subsequent melting of (PREMA-EM-DM)
FOZO
b
39
38
0.51
34
0.51
32
5
.
0
7
.
5130
1
0
.
8
0.51
1
28
18.5 9.0
0.51
1
d
206
26
144 N
19.5 0.0 0.512
Pb/ 204
/
4
2
3
4 Nd
P
EM
2
3
208
EM
a
37
DM
17
18
19
20
21
22
1
Fig. 5. Three-dimensional diagram of 143Nd/144Nd versus 206Pb/204Pb and La/Yb in
global MORB. MORB with enriched isotope signatures, i.e., either PREMA-like or
EM-like have systematically higher La/Yb ratios suggesting that enriched components
in the depleted mantle (DM) are sampled preferentially at low degrees of partial melting
(high La/Yb ratios). Data sources for MORB are compiled from the PetDB database and are
provided given in Supplementary Table 2.
properties, mineralogic composition, genetic origin, age, or location
and distribution of potential mantle components. It also has neither
attempted to define mantle reservoirs nor to infer what the governing
processes leading to mantle heterogeneity are. Rather, the discussion
has been restricted to describe (1) how certain signatures are associated in oceanic basalts, (2) perceive the observed isotopic variability
of the basalts as somehow representative of an essentially unknown
range of isotopic variability in the mantle, and (3) to gain some
first-order insight about how partial melting samples mantle components with different isotopic compositions. The rationale behind this
approach is that the isotope systematics of oceanic basalts can certainly be interpreted in more than one way, and that sampling of
mantle components by partial melting biases the isotopic composition of the erupted lavas (basalts) relative to those of the mantle.
This ambiguity inherent in applying oceanic basalt isotope systematics directly to mantle heterogeneity is perhaps best illustrated
on the example of EM-type OIB. Their continuous array of trends in
isotope (Figs. 1 and 2) and trace element spaces (Willbold and
Stracke, 2006, 2010), the systematic fanning-out of EM trends from
the DM-PREMA array, and the fact that there are many different EM
trends can be interpreted in at least two different ways (Fig. 6):
(1a) Following Zindler and Hart (1986), there are only two isotopically enriched mantle components, EM-1 and EM-2, which
may or may not be genetically related. These homogeneous,
enriched components evolve chemically isolated from other
parts of the mantle, but at some point they mix to form the discrete intermediate enriched mantle component of each individual EM source: step 1 in Fig. 6a (e.g., Zindler and Hart, 1986; Hart,
1988; Hart et al., 1992). One way to mix the homogeneous EM-1
and EM-2 mantle components is in the solid state by progressive
stretching and convective stirring (e.g., Tackley, 2003 and references therein). The observation that all EM isotope trends display little spread transverse to their axes implies that mixing of
the two putative EM-1 and EM-2 mantle components must
have been completed before mixing with the DM-PREMA mantle
components occurred: step 2 in Fig. 6a. In this case, mantle convection must not only bring two EM components together,
which are initially neither genetically nor spatially related (for
example the EM-1 and EM-2 of Zindler and Hart, 1986), it must
also mix the EM-1 and EM-2 mantle components before mixing
with the DM-PREMA components: step 3 in Fig. 6a. Each basalt
along a particular EM trend thus corresponds to melting a discrete EM-PREMA-DM source mixture with no subsequent
41
40
Pb / 204Pb
15
EM
EM-2
1
Step 1 : melt from
heterogeneous continental
crust (EM) component(s) 1
1
2
EMmix
PREMA
3
39
3
208
La / Yb
10
10
Step 1 : solid-state mixing
of two EM components
40
Pb / 204Pb
5
5
La / Yb
0
281
Step 2 :
EM and PREMA melts mix
3
38
1
Step 3 :
Variable dilution of (EM-PREMA) melts
with DM melts
b
37
DM
17
18
19
20
21
22
206
Pb / 204Pb
Society Islands
Samoa
Marquesas
Kerguelen Heard
Pitcairn
Walvis Ridge
Tristan da Cunha
Gough
Fig. 6. Diagrams of 206Pb/204Pb versus 208Pb/204Pb showing schematically how melting
heterogeneous mantle sources with depleted mantle and various enriched mantle
components can produce the different trends of the EM-type basalts. See text for further details.
mixing of melts from the different sources during melting and
melt extraction (step 3 in Fig. 6a). Note that this is an endmember type model that requires selective mixing of different
mantle components in the solid state prior to melting, which appears problematic from a fluid dynamical point of view, unless
the different source components have rather different mechanical behaviors. Furthermore, the EM trends are approximately
linear in multi-dimensional isotope and trace element space
(Figs. 1a, b, e–k, and 2). Mixing in the solid state is expected to result in non-linear, hyperbolic, trends, whereas melt mixing is more
likely to produce approximately linear trends (e.g., Vollmer, 1976;
Langmuir et al., 1978).
(1b) Another way to create the individual EM trends is therefore by
mixing melts from the various mantle components during melting and melt extraction. This requires that the EM-1 and EM-2
mantle components must have been brought together by mantle
convection, but does not necessarily require selective mixing of
mantle components in the solid state; it only requires that the
size of these mantle components is similar to or smaller than
the mantle volume subject to melting. If the solidus of both EM
mantle components is lower than that of the DM-PREMA mantle
components, mixing of partial melts can form a unique EM
component: step 1 in Fig. 6a, which subsequently mixes with
the DM-PREMA melts upon further melting to produce the
observed approximately linear trends: steps 2 and 3 in Fig. 6a.
282
A. Stracke / Chemical Geology 330–331 (2012) 274–299
(2) Alternatively, there may be as many EM mantle components as
EM-type ocean islands (Willbold and Stracke, 2006, 2010).
Hence, one or several geological processes could introduce individual, albeit highly heterogeneous EM-type materials with compositions between the observed extremes (Pitcairn and Samoa),
which remain chemically isolated during residence in the mantle.
This scenario implies that the geologic reservoir from which the
EM mantle components derive must be compositionally heterogeneous. Sampling and transferring random parts of this EM reservoir into the mantle then creates many different heterogeneous
EM-type mantle components, which become stretched and stirred
with other mantle components, for example DM. The presence of
a range of apparently homogeneous EM components between
Pitcairn and Samoa can be explained if, at each locality, partial
melts from heterogeneous EM components blend before mixing
with melts from the DM-PREMA mantle components: step 1 in
Fig. 6b. Scenario (2) is thus similar to scenario (1b) in that mixing
is exclusively in the melt and not solid state, but with the difference that there are many intrinsically heterogeneous and not
only two EM-type mantle components. Note that the extreme
EM components defined by the endpoints of the Pitcairn and
Samoa vectors may represent rather fortuitous mantle components, i.e., there may be even more isotopically extreme mantle
components that are not (yet) sampled by EM-type basalts
(White and Hofmann, 1982; Wright and White, 1987; Hauri et
al., 1993; Workman et al., 2004; Jackson and Hart, 2006; Jackson
et al., 2007).
The two scenarios discussed above are not mutually exclusive, and
it is obvious that stirring and stretching during mantle convection as
well as mixing of melts from different mantle components during
melt extraction occurs. It has to be pointed out, however, that in scenario two there is no requirement for mantle convection to produce
physical mixtures of mantle components before melting. Rather, heterogeneous components are introduced together into the mantle. It
further implies that all EM sources are produced by a limited number
of underlying geological processes, i.e., the range of EM mantle components is directly related to the geological process, the composition
of the sampled reservoir, and the geologic conditions during sampling, transfer and residence in the mantle (e.g., White, 1985;
Willbold and Stracke, 2006, 2010; Class et al., 2009). Hence a geologic
process produces a range of materials, possibly at different times,
resulting in a family of different mantle components. At the other extreme, the concept of only two EM mantle components implies that
each component is generated at a single time in the geologic past by
two geologic processes, which are possibly, but not necessarily, different. In consequence, the generation of EM mantle signatures is related to only two episodes in Earth history, and is not a continuous
process. Alternatively, one has to postulate that there are processes
that produce, at different times, EM components that with time develop only two distinct isotopic signatures, which seems highly unlikely from both a geochemical, geodynamic, and statistical point of
view. For example, how can mantle convection isolate EM components for just the right amount of time to produce only two different
EM isotope signatures today and eventually distribute and mix them
selectively?
2.3. The influence of melt mixing on oceanic basalt isotope arrays
The bottom line of the preceding discussion is that the isotopic
composition of oceanic basalts alone provides an ambiguous view
about mantle heterogeneity. The reason for this is that partial melts
from heterogeneous mantle sources, for both ridge and ocean island
settings, are mixed over a range of depths during melt extraction
and subsequent evolution in magma chambers. A detailed understanding of these mechanisms is consequently required to constrain
to what extent the observed variation in the basalts corresponds to
that of their mantle source.
One possibility is to compare the observed trace element with the
isotopic variability. The rationale behind this approach is that the isotopic variability is a parameter that depends only on the relative contribution from enriched and depleted source components, whereas
the trace element variability is additionally influenced by fractionation during partial melting. Hence comparison of the trace element
variability to the isotopic variability allows assessing the relative importance of partial melting and melt mixing processes on the compositional heterogeneity of the erupted melts (see Fig. 7 and detailed
discussion in Stracke and Bourdon, 2009).
This approach takes advantage of the fact that the relationship between source and melt heterogeneity, for any given melting and melt
mixing scenario, depends strongly on the absolute and relative incompatibility of the elements investigated (Fig. 7). For example, the
more incompatible the elements of a certain trace element ratio are
and the smaller their difference in incompatibility is, the more effective is the melting and melt mixing process for homogenizing source
heterogeneity (compare, for example, La/Nb, La/Sm, and La/Yb in
Fig. 7a–c). Ratios between the highly incompatible elements tend to
be homogenized effectively (e.g., La/Nb, La/Th, Ba/Th), whereas
source heterogeneity is preserved or even magnified for trace element ratios between a highly to moderately incompatible element
(e.g., La/Sm, La/Yb). Source heterogeneity can thus be erased or
amplified by the melting process, depending on the parameters
considered.
It follows from Fig. 7 that the isotopic composition of mantle components is likely to be more extreme than the range observed in
erupted melts from heterogeneous sources. While the most enriched
melts may correspond closely to the composition of the enriched
mantle component, even the most depleted melts are likely to be
less depleted than the most depleted mantle component (Fig. 7). During progressive melting (i.e., with decreasing pressure), the isotopes
of the moderately incompatible elements (Sr, Nd, Hf, Pb) continue
to change in the mixed melts, even when highly incompatible trace
element ratios have already reached constant values after a minimal
amount of melt mixing (Fig. 7a). In other words, isotopic variability
can be preserved over a larger depth range of melting and melt
mixing than for the highly incompatible trace element ratios. Thus,
whether or not correlations between isotope and trace element ratios
are expected depends strongly on the trace element ratio investigated
and on the depth range from which erupted melts are extracted.
Whenever melts are extracted from a large range of depths with
incomplete melt mixing, as in the case of MORB, and especially
ridge centered hot-spots such as Iceland, large trace element variations are expected. In this case, the quality of the correlation between different trace elements and isotope ratios is expected to
decrease systematically with increasing absolute incompatibility
and decreasing relative incompatibility contrast between the two
elements of a given trace element ratio (e.g., from La/Yb over La/Sm to
La/Nb; Fig. 7). If melts from only a small depth interval are extracted, as
for alkaline OIB, a less systematic behavior results (Stracke and Bourdon,
2009). Overall fewer correlations between trace element and isotope ratios are expected owing to the smaller overall variability in the generated
melts. Complete compositional homogenization, for both trace element
and isotope ratios can occur in both ridge (e.g., Sims et al., 2002) and
ocean island settings (e.g., Woodhead, 1996). In contrast, large variability
in the erupted melts is expected only in those settings where either the
absence of a thick lithosphere or high excess mantle temperatures lead
to a large depth range of melting and a limited amount of melt mixing.
In summary, the properties of the isotope-trace element relationships are useful for assessing the extent to which the basalt signatures
represent source signatures (see Section 3.2.1). The relative amount
of melt mixing during melting, melt extraction, and evolution in
magma chambers is reflected in the correlations between isotope
A. Stracke / Chemical Geology 330–331 (2012) 274–299
Peridotite
2
La/Nb
1.5
decreasing P
increasing mixing
>
>
Pyroxenite
1
mixed melts
a
0.5
8
Pyroxenite
b
4
2
Peridotite
mixed melts
decreasing P
increasing mixing
>
c
mixed melts
10
5
0
0.5128
>
Peridotite
La/Yb
Pyroxenite
0
20
15
quality of the correlations between isotope ratios and La/Th, La/Sm,
and La/Yb ratios is expected. The average enriched signatures in OIB
should thus correspond closely to those of the average enriched mantle
component (Section 3.2.1). In ridge-related settings where the depleted
mantle components melt to a larger extent than in ocean island settings,
the enriched lavas are expected to deviate to a larger extent from the
composition of the enriched mantle components (Stracke et al.,
2003b; Stracke and Bourdon, 2009). In both ridge and ocean island settings, however, the most depleted lavas are unlikely to reflect the true
isotopic composition of the depleted mantle component, which is
expected to be isotopically more depleted than even the most depleted
erupted lavas (Salters and Dick, 2002; Stracke et al., 2003b; Stracke,
2008; Salters et al., 2011; Stracke et al., 2011).
3. The origin of mantle heterogeneity
>
La/Sm
6
283
decreasing P
increasing mixing
>
0.5130
0.5132
143
Nb/ 144Nb
Fig. 7. Variations in trace element and isotope ratios in melts from a heterogeneous
mantle source consisting of an isotopically enriched pyroxenite and a depleted peridotite. Shown are the pyroxenite and peridotite and mixed melt compositions as a function of depth of melting (GPa; red, blue and gray curves, respectively). The calculations
assume a solidus temperature for the peridotite of 1480 °C at 3 GPa (Walter, 1998;
Hirschmann, 2000) and pyroxenites with solidus temperatures of 1300 and 1470 °C,
respectively (solid, and dashed lines, respectively). For a 1400 °C mantle adiabat,
therefore, peridotite melting starts at b2.5 GPa and is assumed to occur in the spinel
stability field. The pyroxenites start melting at 4 and 2.5 GPa, respectively. Melts
from both lithologies are pooled at any given depth from the entire underlying melting
column and mixed according to their relative mass proportions in the peridotite melting interval. The pyroxenite/peridotite source ratio is 7:93. For further details of the
calculations and input parameters see Stracke and Bourdon (2009).
and different trace element parameters. For OIB, which represent an
overall low degree of partial melting, and melt extraction over a
short depth interval, little melting of depleted mantle components
is expected. This has the result that melts from the enriched mantle
components do not become diluted to a large extent by melts from
the depleted components. This limited amount of dilution by depleted
melts is expected to result in few correlations between isotope and trace
element ratios in general, and specifically, little improvement in the
In the following, possible processes that account for both the composition, variability and covariance between the different enriched
isotopes and trace element parameters in OIB will be discussed. In addition, several other fundamental questions remain or have arisen
from the previous discussion. The depleted components in the Earth's
mantle, for example, are likely to be more depleted than even the
most depleted MORB. Thus, how depleted and how heterogeneous,
is the Earth's mantle?
3.1. The depleted mantle (DM)
3.1.1. How depleted is the DM?
Owing to the large variability in MORB chemistry and isotope
composition, it has become obvious that their mantle source – the depleted mantle (DM) – consists of a range of enriched and depleted
components (e.g., Tatsumoto et al., 1965; Tatsumoto, 1966; Hart et
al., 1973; White and Schilling, 1978; Dupré and Allègre, 1980, 1983;
Schilling et al., 1983; Allègre et al., 1984; Hamelin et al., 1984, 1986;
Ito et al., 1987; Le Roex et al., 1989; Mahoney et al., 1989; Schiano
et al., 1997; Niu et al., 1999; Vlastelic et al., 1999; Niu et al., 2002;
Hofmann, 2003; Hanan et al., 2004; Agranier et al., 2005; Meyzen et
al., 2005; Sun et al., 2008; Arevalo and McDonough, 2010; Paulick et
al., 2010; Jenner and O'Neill, 2012). In the simplest case it could be
a mixture between just one depleted and one enriched component,
for example depleted mantle peridotite and PREMA (Fig. 1). More
conservatively, it could be perceived as a statistical assemblage of
an unknown number of mantle components, each with their own statistical distribution (e.g., Allègre et al., 1984, 1987a; Meibom and
Anderson, 2003; Rudge et al., 2005; Kellogg et al., 2007).
Arguably, the most abundant and most depleted component of the
DM is residual peridotitic mantle that has experienced one or more
episodes of partial melting. As discussed in Section 1, enriched components with various possible origin counteract the progressive depletion of the DM. The overall extent of heterogeneity of the DM
therefore depends on the proportion, as well as the differences in
composition and age of the various depleted and enriched components. Owing to their different origins, enriched components may
also differ in their mineralogical composition, and hence their melting
behavior. A compositional or sampling bias may thus be introduced
between source (DM) and melt (MORB) depending on how partial
melts from the various source components mix during melting and
melt extraction (see discussion in Section 2.3 and, e.g., Allègre et al.,
1984; Spiegelman and Kelemen, 2003; Maclennan, 2008b; Rubin et
al., 2009; Stracke and Bourdon, 2009; Shorttle and Maclennan,
2011; Waters et al., 2011; Koornneef et al., 2012).
For illustrative purposes, let us assume a two component source,
consisting of a depleted peridotite and an enriched pyroxenite component that corresponds to an ancient recycled oceanic basalt. For
the simplest, but unrealistic case of complete homogenization of partial melts from both source components over the entire depth range
284
A. Stracke / Chemical Geology 330–331 (2012) 274–299
Nd of depleted source component
of melting, inferring the composition of the mantle source components reduces to a simple two component mixing problem. Estimating the composition and abundance of the average melt from the
enriched source component thus allows estimating the average composition of the melts from the depleted mantle source component.
Fig. 8 shows that, in this case, there is an unavoidable discrepancy between the average isotopic composition of MORB and those of the depleted and enriched DM components. If, for example, enriched melts
with εNd = 0 contribute 10% to an average MORB with a typical εNd of
11, the average εNd of the melts from the depleted source component
must be 17.5 (Fig. 8). Although this scenario is too simplistic to realistically simulate MORB generation, it illustrates why MORB isotope
compositions may generally underestimate the extent of heterogeneity of the DM. On a theoretical basis, this notion is confirmed by more
elaborate models for melting heterogeneous sources (Morgan, 2001;
Spiegelman and Kelemen, 2003; Ito and Mahoney, 2005; Stracke
and Bourdon, 2009; Shorttle and Maclennan, 2011; Koornneef et al.,
2012) and statistical models for the isotope evolution of the depleted
mantle (Meibom and Anderson, 2003; Rudge et al., 2005; Rudge,
2006; Kellogg et al., 2007). But what is the observational evidence
to support this notion?
First, the trace element compositions in melt inclusions invariably
range to more depleted compositions than those observed in the associated basalts. While this observation could simply be explained
by incomplete aggregation of partial melts from either homogeneous
or heterogeneous mantle sources (e.g., Sobolev and Shimizu, 1993;
Nielsen et al., 1995; Shimizu, 1998; Sours-Page et al., 1999; Slater et
al., 2001; Maclennan et al., 2003a; Laubier et al., 2007), a similar observation is made by comparing the Pb and Sr isotope composition
in melt inclusions and their basaltic hosts from ridges (Shimizu et
al., 1998; Maclennan, 2008b) and oceanic islands (Saal et al., 1998,
2005; Jackson and Hart, 2006; Paul et al., 2011; Sobolev et al.,
2011). The greater isotopic variability in the melt inclusions compared to their host rocks confirms that the isotopic variability in the
mantle is greater than that observed in the basalts.
Moreover, the neodymium isotopic ratios of abyssal peridotites extend to more depleted values than those of the associated basalts
(Salters and Dick, 2002; Cipriani et al., 2004). There is also a difference
in Os isotope composition between MORB (Schiano et al., 1997;
Gannoun et al., 2003; Escrig et al., 2004; Escrig et al., 2005b; Gannoun et
24
20%
t
20
8%
16
2%
4%
6%
Nd
12
d
ete
epl
of d
om
ec
10%
c
sou
Nd
en
pon
of erupted melt (MORB)
8
4
Nd
0
0
0.05
0.10
of enriched souce component
0.15
0.20
0.25
Fraction of enriched melt
Fig. 8. A diagram showing that the εNd of erupted MORB underestimates the extent of
heterogeneity in the DM. For simplicity, a two-component mantle source is assumed
consisting of a depleted peridotite and an enriched pyroxenite component (εNd = 0),
which both melt and mix to produce an erupted MORB with a typical εNd value of
11. Both the enriched and depleted source components melt to a large extent (e.g.
56 and 20%) and the enriched and depleted melts mix to form the erupted MORB.
The Nd concentrations in the enriched and depleted melt are 19.2 and 3.6 ppm, respectively. The red line show how the εNd value of the depleted source component must
vary as a function of the relative proportion of enriched and depleted melt to keep
the εNd of the erupted MORB constant (εNd = 11).
al., 2007) and peridotites (e.g., Martin, 1991; Roy-Barman and Allègre,
1994; Reisberg and Lorand, 1995; Snow and Reisberg, 1995; Parkinson
et al., 1998; Burton et al., 1999; Brandon et al., 2000; Standish et al.,
2002; Alard et al., 2005; Harvey et al., 2006; Liu et al., 2008; Harvey et
al., 2011). In addition, the 187Os/188Os isotope ratios in peridotites from
the Gakkel ridge cover a large range: from super-chondritic to the lowest,
or most depleted Os isotope ratios in abyssal peridotites measured to date
(Liu et al., 2008). These highly depleted Os isotope signatures suggest
that portions of the Arctic DM range to more depleted values than previously inferred from MORB, and that their depletion is ancient, dating
back to about 2.2 Ga. Although the Os isotopes could selectively reflect
the isotopic composition of sulfide minerals in mantle rocks, rather than
the silicate mantle evolution (Hart and Ravizza, 1996; Burton et al.,
1999), the excellent correlation between the Hf and Os isotope ratios
in the Gakkel Ridge peridotites (Stracke et al., 2011) reaffirms the conclusions by Liu et al. (2008).
The example of the Gakkel ridge (Liu et al., 2008; Stracke et al.,
2011) and other abyssal peridotite isotope studies (Snow et al.,
1994; Salters and Dick, 2002; Cipriani et al., 2004; Warren et al.,
2009) further shows that the residual peridotite components of the
DM have often, but not always, experienced a complex, multi-stage
history of depletion by partial melting and re-enrichment by melts
during ancient or recent melt extraction. Thereby, the Nd isotope system is much more susceptible to resetting by melt–rock interaction
than the Hf and Os isotope systems (Stracke et al., 2011). Owing to
their relative immunity to resetting by melt–rock reaction, the Hf
and Os isotope ratios in abyssal peridotites preserve a record of ancient mantle depletion that reveals a much greater extent of mantle
depletion than observed in the peridotite Nd isotopes or the isotopic
composition of MORB. The highest εHf value in MORB, for example, is
about 28.4 (Hanan et al., 2004), whereas the highest εHf value in abyssal peridotites published so far is 104.2 (Stracke et al., 2011). Moreover, recent Hf and Nd isotope analyses in MORB and OIB suggest
that ultra-depleted isotopic domains, similar to those identified in
the Gakkel Ridge peridotites, are likely to be a common feature of
the oceanic mantle (Salters et al., 2011).
Hence it has become apparent that the DM, at least in part, has a
more extreme and more heterogeneous isotopic composition than indicated by the range of MORB isotope compositions. The observation
that both abyssal peridotites and enriched isotope signals in MORB
are highly variable further suggests that both the depleted and the
enriched components of the DM are isotopically heterogeneous.
MORB may therefore generally underestimate the average isotopic
composition of the DM, depending on how enriched and depleted
DM components are transposed from source to melt.
3.1.2. How enriched is the DM?
Several recent studies concluded that the isotopic variability in the
DM (Fig. 1) could mainly result from continuous partial melting and
recycling of the generated oceanic lithosphere (e.g., Christensen and
Hofmann, 1994; Donnelly et al., 2004; Rudge et al., 2005; Stracke et
al., 2005; Rudge, 2006; Kellogg et al., 2007). In these models a range
of variably depleted peridotite components are formed continuously
as residues of partial melting. Recycling of the melting products – the
oceanic crust – back into the mantle generates a range of variably
enriched mantle components. Isotopic heterogeneity therefore results
from variable enrichment and depletion as well as a range of differentiation times for the recycled oceanic crust and residual peridotitic
mantle.
One key variable for establishing the composition of the DM is
consequently the turn-around time of the mantle, i.e., how often a
given parcel of mantle undergoes melting. This determines the overall extent of depletion and variability of the residual peridotite DM
components, as well as the rate of generation and recycling of oceanic
crust. An upper bound of 1.8–2.0 Ga for the turn-around time is given
by the pseudo-age inferred from the 206Pb/ 204Pb– 207Pb/ 204Pb isotope
A. Stracke / Chemical Geology 330–331 (2012) 274–299
correlation defined by the DM-PREMA trend (Tatsumoto, 1978), and
a lower bound of about 0.5 Ga is given by recent statistical models
(Rudge et al., 2005; Rudge, 2006). The geodynamics of the Earth's
mantle thus clearly has a large impact on its compositional evolution
(e.g., Allègre, 1982; Zindler and Hart, 1986).
Whereas the compositional and isotopic heterogeneity of the depleted DM components is established by repeated cycles of partial
melting, the enrichment by recycled oceanic lithosphere extents the
isotopic variability toward enriched, mostly PREMA-like, compositions (Fig. 1). Stracke et al. (2005) demonstrated that the parent–
daughter and Sr, Nd and Pb isotope ratios measured in present-day
MORB are appropriate for producing the DM-PREMA isotope array
in the future. The time-integrated evolution with positively correlated Rb/Sr and U/Pb ratios, which is required for establishing the
DM-PREMA array, can therefore result mostly from igneous fractionation during generation of the oceanic crust, where both Rb and U behave significantly more incompatible than Sr and Pb, respectively. A
similar conclusion is reached by several recent statistical models (e.g.,
Christensen and Hofmann, 1994; Rudge et al., 2005; Rudge, 2006;
Kellogg et al., 2007). Hence subduction of oceanic plates accounts for
much of the first-order isotopic variation observed in the Earth's mantle (the DM-PREMA vector, Section 2.1). In order to fully account for
the Pb isotopic variability, however, some additional removal of Pb
from the oceanic crust, for example during subduction, is required
(e.g., Chauvel et al., 1995; Stracke et al., 2005; Kellogg et al., 2007).
Note that Zindler and Hart (1986) have originally attributed “PREMA”
to “a quasi-continuous separation of a crustal component from the
mantle over time”. The latter interpretation appears quite similar to
the origin by continuous recycling of oceanic crust attributed to
PREMA here – which is equivalent to “FOZO” as used by Stracke et al.
(2005) – but differs in detail from the original interpretation of
Zindler and Hart (1986).
In addition to enriched components created by recycled oceanic
crust (i.e., PREMA), it is obvious that enriched, EM-like components
are required to account for the isotope signatures in the South Atlantic
and Indian ocean MORB (Dupré and Allègre, 1983; Hart, 1984; Hamelin
and Allègre, 1985; Hamelin et al., 1986; Hanan et al., 1986; Michard et
al., 1986; Price et al., 1986; Dosso et al., 1988; Klein et al., 1988;
Mahoney et al., 1989; Mahoney et al., 1992; Pyle et al., 1992; 1995;
Fontignie and Schilling, 1996; Mahoney et al., 1996; Mahoney et al.,
1998; Douglass et al., 1999; Andres et al., 2002a,b; Kempton et al.,
2002; Le Roux et al., 2002a,b; Mahoney et al., 2002; Meyzen et al.,
2003; Escrig et al., 2004; Hanan et al., 2004; Escrig et al., 2005b;
Meyzen et al., 2005; Nauret et al., 2006; Meyzen et al., 2007; Cordier
et al., 2010). Therefore, a family of enriched components with different
compositions and age account for the enriched signatures of the DM
(i.e., PREMA, EM). Especially the EM signatures, however, are most
prominent in OIB and will be discussed in detail in the following.
3.2. The enriched mantle (EM)
For deciphering the processes that formed EM sources it is important to find a process, or several processes, that account for the entire
array of EM trends in isotope and trace element space, and not just its
most extreme branches, Samoa and Pitcairn (Fig. 1e). A model that
tightly links continental crust and mantle evolution will be presented
in the following whereby the entire range of EM-type mantle sources
is created by elemental cycling between the heterogeneous continental crust and mantle. Alternative models for the origin of EM-type
mantle sources will be discussed, before presenting potential mechanisms for generating more exotic (HIMU-type) mantle signatures.
3.2.1. The correspondence of basalt and mantle EM components
In Section 2.3, it was argued that enriched isotope and incompatible element ratios in OIB correspond closely to those of their enriched
source components. This will be re-affirmed in the following using
285
the approach outlined in Section 2.3, and provides justification for
using the enriched isotope signatures in OIB directly to infer the origin of and processes leading to EM-type mantle components in the
following.
Fig. 9 shows that, relative to the Nd isotope ratios, there is only a
limited increase in the variability from La/Nb, to La/Sm, and La/Yb, for
each island or island group, suggesting that for each given island or island
group, melting occurs over a limited depth interval in the garnet stability field. In addition, little improvement in the quality of the correlations
between isotope ratios and La/Nb, La/Sm, and La/Yb ratios is generally
observed (one exception are the basalts from Kerguelen; Fig. 9a–c).
This observation indicates that the extent of melting is limited to levels
where little dilution of the enriched melts by the depleted source components occurs (see also discussion in Section 2.3). Hence, the isotope
and incompatible trace element ratios of the EM-type (and HIMU) OIB
investigated in Fig. 9 should correspond closely to those of their average
enriched source components, although some damping of the isotope
and trace element ratios due to mixing with melts from the depleted
source components is expected.
3.2.2. A common continental heritage for EM-type mantle sources
EM basalts are enriched in the alkali elements (Rb, K), Th, and Pb and
are depleted in Nb relative to similarly incompatible elements (Dupuy
et al., 1988; Weaver, 1991; Chauvel et al., 1992; White and Duncan,
1995; Dostal et al., 1998; Workman et al., 2004; Willbold and Stracke,
2006); features that are also diagnostic of the continental crust. These
relative enrichments and depletions are coupled to more radiogenic Sr
isotope signatures (Devey et al., 1990; Eisele et al., 2002; Jackson et
al., 2007), suggesting that the variable enrichments and depletions in
these elements are a long-term feature indicative of continental crust
components in EM-type mantle sources. This is in good agreement
with several previous studies, which suggested that recycled upper continental crust is part of the Samoa-like EM sources (Hawkesworth et al.,
1979; Cohen and O'Nions, 1982; White and Hofmann, 1982; Wright and
White, 1987; Weaver, 1991; Chauvel et al., 1992; White and Duncan,
1995; Rehkämper and Hofmann, 1997; Dostal et al., 1998; Stracke et
al., 2003a; Willbold and Stracke, 2006; Jackson et al., 2007; Rapp et al.,
2008; Workman et al., 2008).
Willbold and Stracke (2006, 2010) further showed that typical
compositional differences between the upper and lower continental
crusts correspond to those observed between the most extreme
branches of EM, that is, basalts from the Samoan and Pitcairn
island(s). Most notably, the EM basalts with the highest Sr isotope
ratios have Eu/Eu* b 1 (e.g., Samoa), whereas those with less extreme
Sr isotope have Eu/Eu* ratios ≥ 1 (e.g., Pitcairn, Tristan da Cunha,
Gough Island). Eu/Eu* ratios ≥ 1 are diagnostic features of the lower
and Eu/Eu* ratios b 1 of the upper continental crust (Rudnick and
Fountain, 1995; Rudnick and Gao, 2003). The differences in Eu/Eu* ratios observed in EM basalts (Willbold and Stracke, 2010) therefore
suggest that both upper and lower continental crusts are present
in EM-type mantle sources. Moreover, the time-integrated Th/U ratios
(as expressed by the 208Pb/206Pb ratios) in all EM basalts are variable,
but are higher in basalts with no or positive Eu anomalies compared
to those with negative ones (Willbold and Stracke, 2006, 2010). The
latter is consistent with the complementary low and high Th/U in
the upper and lower continental crusts, respectively (Rudnick and
Fountain, 1995; Rudnick and Gao, 2003). The entire spectrum of
EM-type sources could thus result from melting heterogeneous
mantle components consisting of recycled oceanic lithosphere plus
different proportions of recycled lower and upper continental crusts
(see also, Arndt and Goldstein, 1989; Paul et al., 2002; Willbold and
Stracke, 2006, 2010; White, 2010).
3.2.3. Quantitative recycling models
The calculated trace element and isotope composition of mixtures of
subduction-modified oceanic lithosphere and different types of lower
286
A. Stracke / Chemical Geology 330–331 (2012) 274–299
2.0
1.8
1.6
La / Nb
1.4
1.2
1.0
0.8
0.6
a
0.4
9
8
La / Sm
7
6
5
4
b
3
2
50
La / Yb
40
30
20
10
c
0
0.702
0.703
0.704
0.705
0.706
0.707
0.708
0.709
87
Sr / 86Sr
St. Helena
Rurutu old
Mangaia
Rurutu young
Gough
Tristan da Cunha
Pitcairn
Kerguelen
Samoa
Society
Marquesas
Fig. 9. Diagrams of 87Sr/86Sr versus La/Nb, La/Sm, and La/Yb ratios for EM- and
HIMU-type OIB. Relative to the Nd isotope ratios, there is only a limited increase in
the variability from La/Nb, to La/Sm, and La/Yb, for each given island or island group.
The isotope trace-element relationships can be used to assess to what extent the trace
element and isotope ratios in OIB correspond to those of their source values (Fig. 9a–c).
For further explanation see Section 2.3 and Figs. 7 and 8.
Data sources as given in Willbold and Stracke (2010).
and upper continental crust corroborate this hypothesis (Figs. 10 and
11). Several previous studies presented detailed quantitative models
for calculating the trace element and isotopic composition of ancient
recycled oceanic and continental crust (e.g., Hart and Staudigel,
1989; Chauvel et al., 1992; Hauri and Hart, 1993; Rehkämper and
Hofmann, 1997; Stracke et al., 2003a; Chauvel et al., 2008; Porter
and White, 2009). The modeled trends in Figs. 10 and 11 show the
average isotopic composition of ancient recycled oceanic lithosphere
including various amounts of mafic lower or felsic upper continental crust. The oceanic lithosphere consists of 2 Ga old
subduction-modified oceanic lithosphere (altered MORB, unaltered
MORB and oceanic gabbro together with the attached residual mantle) that has undergone chemical modification during subduction,
i.e., loss of fluid-mobile trace elements such as U, Pb and Rb, for example, that is isotopically similar to PREMA (see detailed discussions
in Stracke et al., 2003a, 2005; Willbold and Stracke, 2006; Beier et al.,
2007). For simplicity, average compositions for the upper and lower
continental crusts given by Rudnick and Gao (2003) are taken with
only minor modifications. For the lower crust, higher Th and U concentrations compared to Rudnick and Gao (2003), but with the
same Th/U ratio of 6 are taken (Th = 2 ppm and U = 0.33 ppm, versus Th = 1.2 and U = 0.2 ppm). For the upper crust, slightly lower
Pb (15 versus 17 ppm and 2.6 versus 2.7 ppm) are used, which are
considered to be within the uncertainty of these estimates. For a further in-depth discussion of model calculations and parameters see
Stracke et al. (2003a) and Willbold and Stracke (2006, 2010). Note that
the trace element composition of the upper continental crust and average subducted marine sediments (GLOSS; Plank and Langmuir, 1998)
is similar. Hence, at least qualitatively, similar trends result if average
upper continental crust is replaced by GLOSS (e.g., Stracke et al., 2003a).
The modeled trends in Fig. 10 are parallel to those of the most extreme branches of the EM basalt array: the Pitcairn-OIB trend is consistent with the vector formed by recycling of oceanic lithosphere and
lower continental crust, whereas the Samoa-OIB trend mirrors the vector formed by recycling of oceanic lithosphere and upper continental
crust. Hence recycling of pure lower or pure upper continental crust together with subducted oceanic lithosphere is consistent with the principal vectors required to explain the isotopic variability of EM-type
oceanic basalts (PC2 and PC3, Section 2.1.2). Recycling of different proportions of upper and lower continental crusts and subducted oceanic
lithosphere therefore produces EM components intermediate between
the extremes (Pitcairn and Samoa) and can account for the entire
range of EM signatures observed.
Furthermore, the general feature of the trace element pattern of
these OIB is also approximated well by partial melting of sources
consisting of depleted mantle (DM), recycled oceanic crust and upper
or lower continental crust, respectively (Fig. 11). Note that the trace element modeling assumes fractional melting of a homogeneous mixture
of depleted mantle and recycled components (see Stracke et al. (2003a)
for details of the isotope and trace element modeling). This is clearly
over-simplified and does not capture the complexity of the melting
and melt extraction process of heterogeneous mantle sources. The general features of the modeled and observed trace element patterns, however, agree well and generally support the results of the isotope models.
As summarized by Willbold and Stracke (2010), therefore, “one common process, the recycling of upper and lower continental crusts and oceanic lithosphere at destructive plate margins and their subsequent
re-melting as part of the mantle sources of ocean island basalts, can account for the entire range of chemical and isotopic signatures in
EM-type oceanic basalts. Each individual EM source is envisaged as a
collection of subducted oceanic crust and lithosphere along with components from the upper and lower continental crusts in various proportions, that has been processed in similar, but not identical, manner at
different times and locations. If sampled by partial melting, each EM
source will therefore have a unique, but perhaps variable, isotopic composition arising from the different proportions of the various continental and oceanic crust and lithosphere components.” The tectonic setting,
the availability and type of continental crust, and the characteristics of
plate subduction thereby determine the individual mixture of
subducted materials. The entire range of enriched and depleted components required to produce the observed isotope and trace element variation in a given EM-type basalt suite (i.e., DM-PREMA-EM) could
therefore be related to a single subduction event (Willbold and
Stracke, 2006, 2010).
A. Stracke / Chemical Geology 330–331 (2012) 274–299
287
0.708
Oceanic lithosphere +
Upper continental crust
20%
2%
87
10%
8%
6%
4%
0.704
143
Sr / 86Sr
0.5132
4%
0.706
0.705
d
a
Nd / 144Nd
0.707
0.5134
6%
Oceanic lithosphere +
Lower continental crust
2%
0.703
0.5130
Oceanic lithosphere +
Upper continental crust
2%
0.5128
2%
8%
10%
0.5126
4%
0.5124
Oceanic lithosphere +
Lower continental crust
6%
20%
0.5122
0.702
0.702
0.703
0.704
0.705
87
0.5134
0.706
0.707
0.708
Sr / 86Sr
0.2836
0.2834
0.5130
0.5124
2%
6%
8%
10%
0.5126
Hf / 177Hf
0.5128
2%
4%
6%
20%
Oceanic lithosphere +
Lower continental crust
Oceanic lithosphere +
8% Upper continental crust
b
Oceanic lithosphere +
Upper continental crust
Pb / 204Pb
4%
10%
6%
2%
6%4%
8%
Oceanic lithosphere +
Lower continental crust
20%
143
40
208
2%
0.2828
e
0.2824
0.5122 0.5124 0.5126 0.5128 0.5130 0.5132 0.5134
41
20%
Oceanic lithosphere +
Upper continental crust
0.2830
0.2826
0.5122
39
0.2832
176
4%
143
Nd / 144Nd
0.5132
Nd / 144Nd
2%
2%
4%
8%6%
10%
20%
38
Oceanic lithosphere +
Lower continental crust
37
17
18
19
206
20
21
c
22
Pb / 204Pb
Fig. 10. a–c) Diagrams of 206Pb/204Pb versus 87Sr/86Sr, 143Nd/144Nd and 208Pb/204Pb, and d, e) 87Sr/86Sr versus 143Nd/144Nd and e) 143Nd/144Nd versus 176Hf/177Hf
showing global MORB and OIB data and vectors of the calculated composition of recycled oceanic lithosphere (i.e., altered and fresh MORB and oceanic gabbro plus the attached
residual mantle) and various proportions of average upper and lower continental crusts. Ticks indicate the proportion of upper or lower continental crust relative to that of the
oceanic crust (MORB and gabbro). The recycling age is assumed to be 2 Ga, the oceanic crust composition as given in Stracke et al. (2003a), Willbold and Stracke (2006) and Stracke
and Bourdon (2009), and the continental crust compositions are taken from Rudnick and Gao (2003) with the modifications as given in the text. Symbols as defined in Fig. 1.
3.2.4. Mode and geodynamics of continental crust–mantle recycling
What are the mechanisms for recycling both upper and lower continental crust components into the Earth's mantle? From the chemical
composition of island arc volcanics, in particular the 10Be abundances
and the Th/Rb and Th/La ratios, there is ample evidence that marine
sediments – which are largely similar in composition to average upper
continental crust – are recycled back into the mantle (e.g., White and
Dupré, 1986; Morris et al., 1990; Elliott et al., 1997; Plank, 2005). At
the same time, dehydration and partial melting of the subducted sediment occurs during subduction processing and transfers parts of the
subducted sediment to the island arc volcanics (e.g., White and Dupré,
1986; Nichols et al., 1994; Stalder et al., 1998; Johnson and Plank,
1999; Hermann and Spandler, 2008; Rapp et al., 2008; Porter and
White, 2009; Behn et al., 2011). The extent to which the original
sediment cover survives sub-arc processing and makes its way into
the deeper mantle remains debated, however. Nevertheless, it has
often been suggested that recycling of marine sediment accompanies
oceanic crust subduction (Hawkesworth et al., 1979; Armstrong and
Harmon, 1981; Cohen and O'Nions, 1982; White and Hofmann, 1982;
Wright and White, 1987; Weaver, 1991; Chauvel et al., 1992; White
and Duncan, 1995; Rehkämper and Hofmann, 1997; Dostal et al.,
1998; Stracke et al., 2003a; Willbold and Stracke, 2006; Jackson et al.,
2007; Rapp et al., 2008; Workman et al., 2008; Porter and White,
2009). Recycling of marine sediments thus appears a viable process
for recycling upper continental crust into the Earth's mantle. Owing to
the lack of large exposed areas of lower continental crust (Rudnick
and Fountain, 1995), however, recycling the lower continental crust requires a different mechanism.
288
A. Stracke / Chemical Geology 330–331 (2012) 274–299
PM normalized
100
average Tristan da Cunha
average Pitcairn
10
Oceanic lithosphere +
Lower continental crust
1
PM normalized
100
average Society (Tahaa)
average Samoa (Malumalu)
10
Oceanic lithosphere +
Upper continental crust
1
Rb Th Nb La Pb Sr Hf Eu Gd Y Yb
Cs Ba U Ta Ce Nd Zr Sm Ti Dy Er Lu
Fig. 11. Trace element composition of melts from sources containing depleted mantle
(Salters and Stracke, 2004), recycled oceanic lithosphere (see Fig. 10) and upper or
lower continental crust. In the upper panel an assemblage consisting of 90% depleted mantle, 8% recycled oceanic crust and 2% lower continental crust (i.e., an
oceanic-continental crust proportion of 80–20%, see Fig. 10) is melted to 1% using partition coefficients appropriate for garnet–peridotite. In the lower panel an assemblage
consisting of 90% depleted mantle, 9.8% recycled oceanic crust and 0.2% lower continental
crust (i.e. an oceanic-continental crust proportion of 98–2%, see Fig. 10) is melted to 0.8%
using partition coefficients appropriate for garnet–peridotite. Model calculations
performed using the model described in detail in Stracke et al. (2003a). Note that the
trace element modeling assumes a homogeneous mixture of depleted mantle and
recycled components. This is clearly over-simplified and does not capture the complexity
of the melting and melt extraction process of heterogeneous mantle sources. The general
features of the modeled and observed trace element patterns, however, agree well and
generally support the results of the isotope models.
Data sources as given in Willbold and Stracke (2010).
Subduction erosion is an important means for mass transfer between the continental crust and mantle and the volume of continental material is significant and might exceed the volume of subducted
sediments (von Huene and Scholl, 1991; Ranero and von Huene,
2000; Clift and Vannucchi, 2004; Willbold and Stracke, 2006, 2010;
Keppie et al., 2009; White, 2010; Stern, 2011). Several studies also
highlighted the importance of delamination of the continental lithosphere for recycling of the continental lithosphere and attached lower
continental crust (Houseman et al., 1981; McKenzie and O'Nions,
1983; Arndt and Goldstein, 1989; Kay and Kay, 1991, 1993; Rudnick,
1995; Jull and Kelemen, 2001). Present-day delamination of the subcontinental lithosphere is inferred or observed in both intra-continental
and convergent margin settings (e.g., Ducea and Saleeby, 1998; Boyd
et al., 2004; Gao et al., 2004; Zandt et al., 2004; Behn and Kelemen,
2006; Behn et al., 2007; Fillerup et al., 2010; Gutiérrez-Alonso et al.,
2011; Levander et al., 2011). In consequence, delaminated and foundered fragments of subcontinental lithosphere have often been invoked
for explaining the trace element and isotopic features of oceanic and
continental basalts (Dupré and Allègre, 1983; McKenzie and O'Nions,
1983; Hawkesworth et al., 1986; Shirey et al., 1987; Mahoney et al.,
1989; Hoernle et al., 1991; Mahoney et al., 1992; Fontignie and
Schilling, 1996; Milner and LeRoex, 1996; Widom and Shirey, 1996;
Douglass et al., 1999; Widom et al., 1999; Douglass and Schilling,
2000; Tatsumi, 2000; Kamenetsky et al., 2001; Andres et al., 2002a;
Doucelance et al., 2003; Escrig et al., 2004; Hanan et al., 2004; Escrig
et al., 2005a,b; Gibson et al., 2005; Lustrino, 2005; Meyzen et al.,
2007; Geldmacher et al., 2008; Goldstein et al., 2008; Coltorti et al.,
2010). The distinct sub-chondritic Os isotopes signature of the continental lithosphere (Walker et al., 1989; Pearson and Nowell, 2002;
Pearson et al., 2004) could be a tracer for delaminated subcontinental mantle (plus lower crust) if the residual lithosphere contributes
significantly to melt production. So far, however, Os isotope data in
oceanic basalts provide ambiguous evidence (e.g., Eisele et al.,
2002; Escrig et al., 2004; and references above).
Geochemical and isotopic data in EM basalts alone are unlikely to
elucidate the physical process by which lower continental crust is
recycled into the mantle. Both processes, i.e., subduction erosion
and foundering of subcontinental lithosphere and lower continental
crust, are by no means mutually exclusive and could take place contemporaneously. Plate subduction is likely to have occurred over
much of Earth's history (e.g., de Wit, 1998; Smithies et al., 2005;
Nutman et al., 2007; Van Kranendonk et al., 2007; Shirey and
Richardson, 2011), but there is also observational evidence for foundering of the subcontinental lithosphere and lower crust (see references above). From a geodynamic perspective, delamination and
foundering of the subcontinental lithosphere and lower continental
crust imply that the EM component is introduced into the mantle at
different times and by different geologic processes than other components required to explain the EM trends (e.g., PREMA; Figs. 1, 2, and
10). In this scenario, therefore, convective stirring is needed to mingle
the different materials in the mantle. This is unlike the case of subduction erosion, where the EM (continental crust) and PREMA (oceanic crust) components are transferred into the mantle at the same
time and by the same geologic process, i.e., transported into the mantle as a physically coherent body. No matter which of these two processes dominates, large-scale chemical cycling between Earth's two
major lithophile element reservoirs – the continental crust and
mantle – appears to be a major process for enrichment of the Earth's
mantle.
3.3. Implications for continental crust evolution
Ultimately, constraining the mass exchange between the mantle and
continental crust is not only important for constraining mantle evolution
but also for understanding the composition and evolution of the continental crust. Although deciphering the nature of the continents has
been a major focus of Earth science, fundamental questions such as how
and when the continental crust formed, and whether it grew continuously or episodically remain open (e.g., McCulloch and Bennett, 1994; Elliott
et al., 1999; Kemp and Hawkesworth, 2003; Hawkesworth and Kemp,
2006a,b; Harrison, 2009; Taylor and MacLennan, 2009). However, if
recycling of continental crust is an important process for explaining EM
signatures in oceanic basalts, it is clear that this must have a major impact
on both mantle heterogeneity and the chemical evolution and growth of
the continental crust over geological history.
Several authors suggested that the relative volume of oceans and
continents did not change significantly since about 2.5 Ga ago (“continental freeboard constraint”, e.g., McLennan and Taylor, 1983;
Schubert and Reymer, 1985; Galer, 1991; Kasting and Holm, 1992).
If this notion is accepted and it is further assumed that the bulk
chemical composition of the continental crust did not change over
this period (e.g., Taylor and McLennan, 1995, 2009), the relative
rates of recycling of upper and lower continental crusts must have
been constant and balanced by their respective growth rates. Models
that invoke sediment (upper continental crust) subduction as the
only mechanism of crustal recycling (see references above) do not
meet these constraints. They are also at odds with the inferred
A. Stracke / Chemical Geology 330–331 (2012) 274–299
3.4. Alternative hypothesis for generating EM sources
3.4.1. Selective recycling of different marine sediments
Many authors previously suggested that the isotopic differences in
EM signatures are due to selective recycling of either marine pelagic
or clastic (“terrigenous”) sediment (e.g., Zindler and Hart, 1986;
Wright and White, 1987; Hart, 1988; Woodhead and McCulloch,
1989; Barling and Goldstein, 1990; Le Roex et al., 1990; Weaver,
1991; Chauvel et al., 1992; Hauri and Hart, 1993; Weis et al., 1993;
Woodhead and Devey, 1993; Hémond et al., 1994; Roy-Barman and
Allègre, 1995; White and Duncan, 1995; Hauri et al., 1996; Hofmann,
1997; Rehkämper and Hofmann, 1997; Blichert-Toft et al., 1999;
Eisele et al., 2002; Class and Le Roex, 2008).
Pelagic and terrigeneous sediments both cover a large range of the
total compositional variability in marine sediments, which is governed
by the abundance of biogenic and detrital phases, the nature of the
source of the detrital components, and the sedimentation rate (see
Plank and Langmuir (1998) for a detailed discussion). Recycling of pelagic sediments several billions years ago could indeed explain a range
of Pb isotope ratios in OIB (e.g., Stracke et al., 2003a; Porter and
White, 2009). All marine sediments, including pelagic sediments, however, have Eu/Eu* ratios b 1 (Ben Othman et al., 1989; Plank and
Langmuir, 1998; McLennan et al., 2005). Since EM-type mantle components have Eu/Eu* ratios both greater and less than one, this observation is difficult to reconcile with recycled marine sediments as the
only source for EM-type mantle components.
On a local scale, however, pelagic sediments could be part of individual OIB sources (see references above). Pelagic sediments produced in an oxidizing environment are characterized by Ce/Ce*
ratios b 1. Class and Le Roex (2008) interpreted the range of Ce/Ce*
ratios from 0.92 to 1.04 in lavas from Gough Island to be a source feature resulting from selective recycling of pelagic sediments (Fig. 12;
Ce/Ce* denotes the deviation from the Ce concentration resulting
from the interpolation of the La and Nd chondrite-normalized concentrations). In this case, however, a correlation between Ce/Ce*
and 87Sr/ 86Sr ratios would be expected, but is not observed (Class
and Le Roex, 2008). Trace element and isotope data on a large set of
samples from Gough Island and the adjacent Tristan da Cunha
(Willbold and Stracke, 2006, 2010) suggest that the only two samples
from Gough with Ce/Ce* ratios b 0.95 (Ce/Ce* = 0.83 and 0.91) are altered, because they show a deficiency of fluid mobile incompatible
trace elements (e.g., U, Rb, K) relative to similarly incompatible
fluid-immobile elements (e.g., Th, Nb). The other, fresh samples
have Ce/Ce* from 0.95 to 1.04 for Gough and Ce/Ce* = 0.99–1.04 for
Tristan da Cunha (according to Ce/Ce* as defined by Class and Le
Roex (2008); Fig. 12).
Hence Ce/Ce* ratios in lavas from Tristan da Cunha and Gough are
ambiguous tracers for recycled sediments. In general, however, Ce/Ce*
ratios are a viable tool for identification of recycled marine sediments
formed under oxidizing surface conditions, i.e., sediments formed
after 2.5 Ga, if subduction processes do not appreciably fractionate La
from Ce and Nd. The latter is supported by recent experimental studies
(e.g., Hermann and Rubatto, 2009).
3.4.2. Large-scale metasomatism of the oceanic mantle?
Another alternative is that EM-type OIB derive from almost primitive or depleted mantle sources that have been metasomatized several
billion years ago (e.g., Richardson et al., 1982; Hawkesworth et al.,
1984; Roden et al., 1984; Hart et al., 1986; Hart, 1988; McKenzie and
O'Nions, 1991; Niu and O'Hara, 2003; Workman et al., 2004; Pilet et
al., 2005; Salters and Sachi-Kocher, 2010). Ancient metasomatism of
the Earth's mantle is investigated here by migrating small-degree partial
melts through the ambient mantle in a mid ocean ridge setting, similar
to Workman et al. (2004). More complex scenarios involving volatilerich melt or unusual residual phases can be envisioned for the oceanic
upper mantle (e.g., Hauri et al., 1993; Chauvel et al., 1997; Class and
Goldstein, 1997; Grégoire et al., 2000; Dasgupta et al., 2006; Pilet et al.,
2011), but it remains essentially unknown whether these are local or
global phenomena. Hence the following discussion is restricted to simple
melt–rock interaction.
Thereby, an initial depleted mantle peridotite (Salters and Stracke,
2004; Workman and Hart, 2005) is created by extraction of a 2% melt
from a primitive mantle (McDonough and Sun, 1995). Metasomatism of
1.1
1.0
Ce / Ce*
andesitic composition of the bulk continental crust (Weaver and
Tarney, 1984; Rudnick, 1995; Rudnick and Fountain, 1995; Taylor
and McLennan, 1995; Wedepohl, 1995; Gao et al., 1998; Rudnick
and Gao, 2003). This is because an andesitic bulk continental crust
composition requires removal of some fraction of the lower
crust to balance the mafic lower and felsic upper continental crust
(e.g., Rudnick, 1995; Hawkesworth and Kemp, 2006b). Continental
crust formation since the Proterozoic (2.5 Ga) is assumed to occur
mainly by island-arc processes (e.g., Taylor, 1967; Kelemen, 1995;
Rudnick, 1995; Rudnick and Gao, 2003; Taylor and MacLennan,
2009). It seems likely then that the return flux of lower continental
crust into the mantle in this period is also controlled by arc-related
processes, i.e., either subduction erosion (von Huene and Scholl,
1991; Ranero and von Huene, 2000; Stern, 2011) or sub-arc
delamination/foundering of lower continental crust (Kay and Kay,
1988; Kay et al., 1994; Kelemen et al., 2003; Behn and Kelemen,
2006; Behn et al., 2007). In contrast, maintaining a constant balance
between sediment recycling at subduction zones and foundering of
lower continental crust outside convergent margin settings appears
rather unlikely over time scales of ~ 2.5 billion years (e.g., Taylor
and MacLennan, 2009). At least since the early Proterozoic
(2.5 Ga), the evolution of the continental crust and mantle heterogeneity consequently appears to be controlled by plate tectonic processes at convergent plate margins.
Note, however, that major plate subduction events, and thus times
of major mantle enrichment, need not be contemporaneous to prominent episodes of (upper) continental crust formation. The reason is
that upper crust formation is at least a two-step process involving partial melting of the mantle to form a basalt and its further igneous differentiation to produce the felsic upper continental crust (e.g., Rudnick,
1995; Hawkesworth and Kemp, 2006a,b). Hence, although enrichment
and depletion of the Earth's mantle is intimately linked to (upper) continental crust formation, the timing of major crust formation events
could be decoupled from the age of enriched components in the Earth's
mantle.
289
0.9
Gough (Class & LeRoex, 2008)
Gough (Willbold & Stracke, 2006, 2010)
Tristan (Willbold & Stracke, 2006, 2010)
0.8
0.7046
0.7048
0.7050
87
0.7052
0.7054
0.7056
86
Sr/ Sr
143
144
Fig. 12.
Nd/ Nd versus Ce/Ce* in lavas from Gough Island and Tristan da Cunha,
showing that Ce/Ce* ratios do not correlate with 87Sr/86Sr (Ce/Ce* is the deviation
from the Ce concentration resulting from the interpolation of the La and Nd
chondrite-normalized concentrations). This lack of correlation argues against a pelagic
sediment source component in the mantle source of these OIB.
290
A. Stracke / Chemical Geology 330–331 (2012) 274–299
this depleted mantle by a percolating low-degree melt (1%) re-enriches
the peridotite. This metasomatized mantle is either slightly enriched or
depleted compared to primitive mantle, depending on the assumed
melt–rock ratio and the initial melt fraction to create depleted peridotite.
The larger the melt–rock ratio and the smaller the melt fraction of the
metasomatizing melt is, the more enriched is the metasomatized source.
If metasomatism is ancient enough (e.g., 3 Ga, Fig. 13), this process can
generate sources with isotopic compositions similar to those of the EM
basalts (Fig. 13, see also references above). Melting such ancient
metasomatized mantle sources and mixing of the partial melts with
DM-PREMA derived melts can then yield trends similar to those formed
by EM-type OIB (Fig. 13).
During partial melting, Rb, Th, U, Nd and Hf behave more incompatible than Sr, Pb, Sm, and Lu. Hence the percolating melt has higher
Th, U/Pb, Rb/Sr, and lower Sm/Nd and Lu/Hf ratios than its mantle
source. The metasomatized sources therefore develop high Sr and
Pb and low Nd and Hf ratios, with increasing Sr for increasing Pb,
but decreasing Nd and Hf isotope ratios for increasing Pb and Sr isotope ratios (Fig. 13). In multi-dimensional isotope space, such
metasomatized sources therefore form trends unlike those observed
in EM-type basalts (see Section 2.1 and Fig. 13). In particular, it is impossible to generate sources that develop high Sr but low Pb isotope
ratios that also have a decoupled 206Pb/ 204Pb and 208Pb/ 204Pb isotope
evolution, which is required to explain the Pitcairn-like Sr–Nd–Hf–Pb
isotope signatures (Section 2.1, Fig. 13c). Although for low melt–rock
ratios, isotope compositions on intermediate positions along the
trends formed by Pitcairn-like OIB can result, it is impossible to
generate sources with higher 87Sr/ 86Sr, but lower 206Pb/ 204Pb ratios
than bulk Earth by ancient melt rock interaction of an originally primitive mantle source (Fig. 13c). Ancient metasomatized mantle sources
can therefore develop Sr–Nd–Hf–Pb isotope ratios appropriate to explain the EM vectors of the Samoa-like OIB, but cannot generate mantle sources suitable for the Pitcairn-like OIB in a simple two-stage
process.
Moreover, metasomatized mantle sources appropriate for explaining
the supra-chondritic Sr and Pb and sub-chondritic Nd and Hf isotope ratios of the Samoa-like EM basalts are characterized by smoothly increasing incompatible element concentrations from La to Cs in a multielement diagram (see Fig. 22 in Workman et al. (2004) and Halliday et
al., 1995; Salters and Sachi-Kocher, 2010). Such incompatible trace element patterns, however, are difficult to reconcile with the irregular and
shallowly sloped incompatible trace element pattern of the isotopically
most extreme Samoan basalts from Malumalu (low (Cs, Ba, U, K)/La and
high (Th, Rb)/La ratios; Fig. 11) and other EM basalts (Willbold and
Stracke, 2006). Any deviation in the trace element patterns of EM basalts
relative to the smoothly increasing incompatible element pattern of the
metasomatized mantle source must consequently either be explained
by unusual residual phases during partial melting, by admixing of melts
from other sources with the appropriate trace element characteristics,
or by more complex, multi-stage metasomatic processes. Although a
metasomatic origin by ancient melt–rock interaction of EM sources
appears possible from an isotopic perspective, it is difficult to reconcile
with the resulting trace element patterns (e.g., Halliday et al., 1995;
Workman et al., 2004; Salters and Sachi-Kocher, 2010).
0.708
3%
0.5134
a
0.707
c
2%
Sr / 86Sr
0.706
0.5130
1%
0.5128
87
143
Nd / 144Nd
0.5132
0.5126
2%
0.5124
1%
0.705
0.704
0.703
0.5122
0.702
0.702
0.703
0.704
0.705
0.706
0.707
0.708
17
18
Sr / 86Sr
5%
21
22
0.5132
Nd / 144Nd
4%
3%
40
2%
39
1%
0.5130
1%
0.5128
143
Pb / 204Pb
20
Pb / 204Pb
0.5134
41
208
19
206
87
0.5126
38
b
37
0.5124
2%
d
EM
0.5122
17
18
19
206
20
Pb / 204Pb
21
22
17
18
19
206
20
21
22
Pb / 204Pb
Fig. 13. Diagrams of a) 87Sr/86Sr versus 143Nd/144Nd, b) 206Pb/204Pb versus 208Pb/204Pb, c) 206Pb/204Pb versus 87Sr/86Sr, and d) 206Pb/204Pb versus 143Nd/144Nd including trends for a
depleted mantle source that has been metasomatized by reaction with a small degree partial melt 3 Ga ago (see text for details of the calculation). Although sources similar to those
of the most enriched Samoa basalts can result, the general trend is toward high 87Sr/86Sr and 206Pb/204Pb isotope ratios and therefore forms a vector oblique to those observed in
EM-type basalts. See text for further details of the modeling. Tick marks indicate mixing proportion between small-degree metasomatizing melt and depleted mantle. Symbols as
defined in Fig. 1. See text for further details.
A. Stracke / Chemical Geology 330–331 (2012) 274–299
10
3.5.1. The distribution and sampling of HIMU mantle components
HIMU is stands for “high-μ”, with μ being defined as μ =
( 238U/ 204Pb)t = 0 (Houtermans, 1953; Zindler and Hart, 1986).
Basalts with HIMU signatures, i.e., with 206Pb/ 204Pb > 20.5 and
87
Sr/ 86Sr b 0.703 occur at only a few spatially widely separated
localities: St. Helena (Atlantic ocean), the Cook Austral and
Chatham Islands in the Southwest Pacific Ocean, and Mt. Erebus
in Antarctica.
As discussed in Section 2.1, HIMU is the only component observed
in the basalts that is not associated with EM components. It does,
however, occur together with the PREMA-DM components. Another
peculiarity of HIMU basalts is that they are by far the most homogeneous of all isotopic families of oceanic basalts (MORB, EM-type
OIB), both isotopically and in terms of their trace element composition (Figs. 14, 15, and 16).
Assuming that the HIMU signatures in the basalts are close to their
source values (see Fig. 9 and discussion Section 3.2.1), the HIMU mantle components must have evolved with remarkably uniform composition for similar time periods, irrespective of their inferred origin
(Stracke et al., 2005). Consequently, the HIMU mantle components
may originate from an isotopically homogeneous part of the mantle.
It is therefore likely that a single process in the geologic past generated
the HIMU mantle components. Hence the HIMU mantle components
must have started out as a physical entity that became stretched, stirred
and randomly distributed by mantle convection. Is it geodynamically
possible that HIMU components at the present time are sampled at
only a few localities by partial melting in the shallow mantle?
Stegman et al. (2002; their Fig. 1) shows that an initially concentrated
anomaly is spread out over 3 Ga into a “ribbon” with a large lateral extent, at least in a mantle with a depth-dependent viscosity profile where
the lower mantle is about 30 times more viscous than the upper mantle.
An originally coherent body that became dispersed by mantle convection could thus indeed, owing to the randomness of the sampling process by partial melting, occur at only a few geographically dispersed
locations similar to what is observed for HIMU.
3.5.2. Origin of the HIMU mantle components
As inferred from the basalts, the HIMU mantle source is characterized
by depletion of the very incompatible elements (Cs, Rb, Ba, Th, Nb, Ta
and U) relative to the REE and enrichment of most other trace elements
N (Number of data)
20
HIMU
MORB
Pitcairn-like EM
Samoa-like EM
80
120
40
60
80
10
20
40
40
5
20
0
0.702
0.703
0.704
0.705
0.706
0.707
0.708
0.709
87
Sr/ 86Sr
Fig. 14. Histogram showing the Sr isotope variation in different groups of oceanic basalts. HIMU-OIB are those OIB with 206Pb/204Pb > 20.5 and 87Sr/86Sr b 0.703 (Stracke
et al., 2005), the Samoa-like OIB are basalts from Samoa, and the Society and
Marquesas islands, the Pitcairn-like OB are OIB from Pitcairn, Tristan da Cunha, and
Gough Islands as well as basalts from the Walvis Ridge and Kerguelen. Data compilation is an updated version of the one given in Stracke et al. (2003a) and is provided
in the supplementary materials (Supplementary Table 1).
STD error of the mean
3.5. HIMU
15
291
1
Sr isotope data
0.1
Nd isotope data
0.01
0.001
HIMU
MORB
EM-1
EM-2
87
Fig. 15. Diagram showing the standard deviation of the mean of the Sr/86Sr and 143Nd/
144
Nd ratios in different groups of oceanic basalts. HIMU-OIB are those OIB with 206Pb/
204
Pb>20.5 and 87Sr/86Srb 0.703 (Stracke et al., 2005), the Samoa-like OIB are basalts
from Samoa, and the Society and Marquesas islands, the Pitcairn-like OB are OIB from
Pitcairn, Tristan da Cunha, and Gough Island as well as basalts from the Walvis ridge
and Kerguelen. Data compilation is an updated version of the one given in Stracke et al.
(2003a) and is provided in the supplementary materials (Supplementary Table 1).
with respect to Bulk Earth (Fig. 16). It also has (Rb, Ba, Th, U)/REE and
Rb/Sr ratios lower than, and (U, Th)/Pb, Sm/Nd, and (Nb, Ta)/La ratios
higher than those in bulk earth (Willbold and Stracke, 2006). This distinct depletion of very incompatible trace elements (Fig. 16) relative to
La, enrichment of Nb and Ta relative to U and La, as well as its depletion
of Pb relative to Nd, provide a strong case that HIMU sources are not only
depleted in the very incompatible elements, but also modified by preferential loss of fluid-mobile trace elements (e.g., Dupuy et al., 1988; Hart
and Staudigel, 1989; Weaver, 1991; Chauvel et al., 1992; Hémond et
al., 1994; Woodhead, 1996; Kogiso et al., 1997b; Stracke et al., 2003a;
Willbold and Stracke, 2006; Kawabata et al., 2011). Hence recycling of
oceanic crust that has been modified by fluid extraction in the sub-arc
environment, i.e., recycling of subduction-modified oceanic crust, has
often been invoked (e.g., Vidal et al., 1984; Zindler and Hart, 1986;
Nakamura and Tatsumoto, 1988; Hart and Staudigel, 1989; Weaver,
1991; Chauvel et al., 1992; Hauri and Hart, 1993; Reisberg et al., 1993;
Hémond et al., 1994; Hauri et al., 1996; Woodhead, 1996; Kogiso et al.,
1997b; Lassiter and Hauri, 1998; Salters and White, 1998; Stracke et
al., 2003a, 2005; Willbold and Stracke, 2006; Chan et al., 2009; Parai et
al., 2009; Vlastélic et al., 2009; Day et al., 2010; Hanyu et al., 2011;
Kawabata et al., 2011). This scenario requires, however, that the recycled
oceanic crust is devoid of the overlying marine sediment cover, because
even minute quantities of additionally subducted sediment would lead
to development of radiogenic isotope signatures quite unlike HIMU
(e.g., Stracke et al., 2003a; Chauvel et al., 2008; Porter and White, 2009).
As discussed in the preceding paragraph and Section 3.5.1, the apparent uniformity of the HIMU mantle components implies that they
must have evolved with remarkably similar and overall depleted incompatible trace element composition (Willbold and Stracke, 2006)
for similar time periods, irrespective of their inferred origin. If it is assumed that the HIMU mantle components indeed represent ancient
subduction-modified oceanic crust, the observed narrow range of isotopic compositions can only result if similar material is processed
through the “subduction factory” in a similar way and at about the
same time. In other words, the origin of the HIMU components
must be related to a single subduction event in the past. Otherwise,
oceanic crust with different initial compositions and age would have
to be processed differently as a function of time and composition,
but always in such a way that the end-product has similar isotopic
signatures today. Given that subduction is an ubiquitous process accompanied by transfer of elements to the island arc volcanics, the apparent rarity and uniformity of the HIMU mantle components appears
difficult to reconcile with their inferred origin from subduction-
292
A. Stracke / Chemical Geology 330–331 (2012) 274–299
100
PM normalized
St. Helena
10
average St. Helena
PM normalized
100
Mangaia
10
average St. Helena
1
PM normalized
100
Rurutu
Rurutu young lavas
10
Rurutu old lavas
average St. Helena
1
Rb Th Nb La Pb Sr Hf Eu Tb Ho Er Yb
Cs Ba U Ta Ce Nd Zr Sm Gd Dy Y Tm Lu
Fig. 16. Trace element patterns of the different HIMU-type OIB from St. Helena in the
Atlantic Ocean and the Mangaia and Rurutu of the Cook–Austral Islands in the South
Pacific Ocean. Note the similarity of the trace element patterns from these different
islands by comparison to the average trace element pattern of the St. Helena basalts.
Data for St. Helena from Willbold and Stracke (2006), for Mangaia from Woodhead
(1996), and for Rurutu from Chauvel et al. (1997).
modified oceanic crust. Why are the HIMU components not more
ubiquitous? Why should such a unique signature be related to a
such an ubiquitous process? How is the marine sediment cover removed during subduction and sub-arc processing? Alternatively, if
an origin by subduction-modified oceanic crust is taken for granted,
one could ask why did this process, in this particular way, apparently
happened only once and what special circumstances prevented it
from happening more often? Currently, these are open questions
that ultimately have to be reconciled with the often-invoked origin
of HIMU by subduction-modified recycled oceanic crust (see references above).
Alternatively, the HIMU mantle components may have a different
origin. One proposed alternative is a metasomatic origin for the HIMU
source (e.g., Menzies and Murthy, 1980; Hart et al., 1986; Zindler and
Hart, 1986; Hart, 1988; Halliday et al., 1995; Pilet et al., 2011). The
specific case of melt–rock interaction, however, leads to Sr–Pb isotope
characteristics of HIMU basalts quite unlike those of HIMU lavas
(Fig. 13). Note that the time-integrated Rb/Sr and U, Th/Pb ratios and
Sr–Pb isotope systematics of the HIMU mantle components are apparently decoupled (almost invariable 87Sr/ 86Sr for a range of 206Pb/204Pb
ratios; Fig. 1e). As shown in Section 3.4.2, enrichment of mantle sources
that are metasomatized by small-degree partial melts is governed by
the behavior of incompatible trace elements during partial melting,
that is, they are characterized by a coupled enrichment in Rb and U relative to Sr and Pb and thus develop high Pb and high Sr isotope ratios
(Fig. 13). Moreover, Nb, Ta and U are hardly fractionated during partial
mantle melting as witnessed by the small range of Nb/(U, Ta) ratios in
oceanic basalts (e.g., Hofmann et al., 1986; Sims and DePaolo, 1997;
Hofmann, 2003; Willbold and Stracke, 2006; Pfänder et al., 2007). The
enrichment of Nb and Ta relative to La and U in HIMU basalts is therefore difficult to produce if generation of HIMU mantle components is
solely related to partial melting processes. Hence the combined isotope
and trace element composition of HIMU basalts and their sources suggests that processes other than partial melting must be involved in
their genesis.
Alkali elements and Pb are considerably more fluid-mobile than
the REE and the fluid-immobile HFSE (e.g., Nb and Ta; see references
below). The enrichment of Nb–Ta relative to the similarly incompatible trace elements Cs, Rb, K, Ba and U, and the depletion of Pb relative
to Nd (Fig. 16), consequently provide a strong case that HIMU sources
were influenced by fluid-rock alteration processes at some stage. This
selective depletion of fluid mobile elements is consistent with loss of
fluid-mobile trace elements in the sub-arc environment, where dehydration of subducted MORB leads to net loss of Cs, Rb, Ba, K, the LREE,
Pb and Sr but relative enrichment of Nb and Ta (e.g., McCulloch and
Gamble, 1991; Brenan et al., 1994; Kogiso et al., 1997a; Stalder et
al., 1998; Foley et al., 2000; Klemme et al., 2002; John et al., 2004;
Kessel et al., 2005; Klemme et al., 2005; Klimm et al., 2008). Fluid–
rock interaction processes, however, may also occur in a number of
other crustal or shallow mantle environments. Provided that the
fluid-metasomatized rocks can be readily transferred or produced in
the mantle, scenarios other than recycling of subduction modified oceanic crust are possible (Hart et al., 1986; Pilet et al., 2011). Given that
currently no such suitable source rocks have been identified, however,
and considering the problems inherent to a subduction-related origin
(see discussion above), the quest for finding a suitable HIMU reservoir
certainly continues.
3.6. The age of mantle heterogeneity
Another unresolved and often neglected question is to what extent different isotopic signatures observed in oceanic basalts result
from compositional differences and/or different residence or storage
times in the mantle. Os isotope studies in peridotites (e.g., Martin,
1991; Roy-Barman and Allègre, 1994; Reisberg and Lorand, 1995;
Snow and Reisberg, 1995; Parkinson et al., 1998; Burton et al., 1999;
Brandon et al., 2000; Standish et al., 2002; Alard et al., 2005; Harvey
et al., 2006; Liu et al., 2008; Harvey et al., 2011) and mantle xenoliths
(Walker et al., 1989; Martin, 1991; Walker et al., 1996; Parkinson et
al., 1998; Burton et al., 1999; Meisel et al., 2001; Alard et al., 2002;
Bizimis et al., 2005; Bizimis et al., 2007; Harvey et al., 2010) not only document the extent of mantle depletion and heterogeneity, but also
suggest that mantle depletions are ancient, often >2 Ga old. However, Os is controlled by accessory sulfides and metals (Morgan, 1986;
Hart and Ravizza, 1996; Burton et al., 1999; Brandon et al., 2000;
Luguet et al., 2001; Alard et al., 2002; Standish et al., 2002; Walker et
al., 2002; Alard et al., 2005; Brandon et al., 2006; Luguet et al., 2008;
Harvey et al., 2010; Harvey et al., 2011) and may therefore trace a different history than the lithophile isotopes that are hosted by the main
mantle minerals (garnet, pyroxene, olivine, spinels).
Although the 206Pb/204Pb versus 207Pb/204Pb mantle array has often
been interpreted as an average mantle depletion age of 1.8–2 Ga
A. Stracke / Chemical Geology 330–331 (2012) 274–299
(Tatsumoto, 1978), it is only a pseudo-isochron without direct age significance, which most likely reflects different mantle turn-over (stirring)
times (Christensen and Hofmann, 1994; Allègre and Lewin, 1995;
Albarède, 2001; Xie and Tackley, 2004; Rudge et al., 2005; Rudge,
2006). Furthermore, calculation of model ages for HIMU sources, the by
far most homogenous of all identified mantle components, does not result
in consistent age information, because each isotope system gives a different model age, irrespective of whether measured or partial melting
corrected parent–daughter ratios are used.
Why is it apparently so difficult to extract an age of mantle heterogeneity from the isotopic composition of mantle rocks and oceanic basalts?
With respect to the basalts one reason certainly is that basalt compositions do not correspond directly to mantle compositions (Section 2.3).
Even for homogeneous mantle sources with a simple one or two stage
source evolution, the different behaviors of the lithophile trace elements
and isotopes (U–Th–Pb, Rb–Sr, Sm–Nd, Lu–Hf) during partial melting
and melt mixing tend to obscure any intrinsic age information
(Section 2.3). Source heterogeneity or a more complicated evolution of
the different source components only magnifies the problem. With respect to the mantle rocks (peridotites), similar problems apply, because
different isotope systems may trace a different history (e.g., Stracke et
al., 2011; Rampone and Hofmann, 2012) and a multi-stage source evolution tends to obscure the age information inherent in the isotopic composition of mantle-derived rocks.
4. Synthesis and outlook
The Earth's mantle is highly heterogeneous (e.g., Gast et al., 1964;
Tatsumoto et al., 1965; Hedge, 1966; Tatsumoto, 1966; Hart et al.,
1973; White and Schilling, 1978; Allègre, 1982; Zindler and Hart,
1986; Hofmann, 1997; Stracke et al., 2005). Oceanic basalts (MORB
and OIB) have traditionally been regarded as faithful recorders of mantle heterogeneity because they are large-scale partial melts from the
Earth's mantle that escape contamination by the continental crust.
One way to study mantle heterogeneity has therefore been to identify extreme components in the basalts and to associate these directly
with mantle components, or even large-scale mantle reservoirs (e.g.,
Zindler and Hart, 1986).
The approach taken here is to identify the minimum number of
principal directions in multi-dimensional isotope and trace element
space that capture all important features of the data. Once identified,
understanding how these principal directions represent mantle heterogeneity is considered a crucial step for accurately inferring mantle
from basalt composition. Here, we stress that melt mixing during partial melting and melt extraction can bias the composition of the basalts compared to that of the mantle source (Section 2.3). Although
it is therefore problematic to equate basalt and mantle isotopic composition, the properties of the isotope-trace element relationships in
oceanic basalts serve to assess to what extent the basalts represent
source signatures (Section 2.3).
Using the combined trace element and isotope signatures in a large
number of OIB, it has been confirmed that enriched signatures in OIB
closely correspond to those of their average enriched source components (Sections 2.3 and 3.2.1). Enriched signatures in OIB can therefore
be used to trace the geologic reservoirs that exchange mass with the
mantle and to identify the geological processes that introduce heterogeneous material into the Earth's mantle. Whenever melting is over a
greater range of pressures than in OIB settings, for example at mid
ocean ridges, the depleted mantle components melt to a large extent
and the enriched lava compositions are expected to deviate from
those of the enriched mantle components (Sections 2.3 and 3.1). In
both ridge and ocean island settings, however, the most depleted
lavas are unlikely to reflect the true isotopic composition of the depleted mantle components, which is expected to be isotopically more extreme than even the most depleted erupted lavas. Hence the depleted
mantle (DM) is isotopically more extreme than the most depleted
293
MORB (Section 3.1). Peridotites are thus more likely to record the true
depletion of the Earth's mantle, but a representative estimate of the average depleted mantle composition from peridotites is prevented by
their scarcity, often altered state, and the limited amount of available
(isotope) data (e.g., Rampone and Hofmann, 2012).
A number of recent studies have demonstrated that the isotopic variability along the DM-PREMA (see Section 2.1) is likely to be caused by
continuous production and recycling oceanic crust (e.g., Christensen
and Hofmann, 1994; Donnelly et al., 2004; Rudge et al., 2005; Stracke
et al., 2005; Rudge, 2006; Kellogg et al., 2007). Thereby, a family of
enriched components – ancient recycled oceanic crust with different
compositions and age lumped together under the term PREMA –
account for the enriched signatures of the DM-PREMA array, whereas
partial melting is expected to produce a range of variably depleted residual peridotites. Hence a first order geologic process, the generation
and subduction of oceanic plates, accounts for the first-order heterogeneity of the Earth's mantle. The second-most important process appears
to be large-scale element cycling between the continental crust and
mantle, via recycling of upper and lower continental crusts into the
Earth's mantle. Recycling of lower continental crust is proposed to be responsible for most of the isotopic heterogeneity outside the DM-PREMA
range (EM-type MORB and OIB). Recycling of the upper continental
crust, on the other hand, is inferred to be only a minor process, but is required to explain the entire spectrum of enriched signatures (EM). Note
that removal of some fraction of the lower crust is also required
for maintaining an andesitic bulk continental crust composition
(e.g., Rudnick, 1995; Hawkesworth and Kemp, 2006b). Mass exchange between the lower continental crust and mantle is therefore
not only consistent with the most abundant enriched mantle (EM) signatures, it is also of major importance for continental crust evolution.
Hence one of the main processes responsible for establishing the formation and composition of the continental crust is also one of the main
processes for the generation of mantle heterogeneity. Other, more elusive processes create exotic signatures (HIMU), but overall account for
only a small part of the observed mantle heterogeneity.
The model advocated here presents a simple conceptual framework that explains mantle heterogeneity as a result of large-scale
chemical cycling between Earth's two major lithophile element reservoirs, the oceanic and continental crust and mantle. Mass exchange
between the crust and mantle introduces heterogeneous materials
into and thereby enriches the Earth's mantle. To what degree this initial heterogeneity is preserved depends on the degree to which these
materials retain their physico-chemical integrity during residence in
the mantle. In this way, mantle geochemistry is linked to the physics
and fluid dynamics of the mantle. On the basis of the isotopic heterogeneity observed in melt inclusions, heterogeneity in the mantle is of
small scale; certainly on the kilometer scale of the melting region but
perhaps even down to the millimeter scale of individual minerals.
This implies that heterogeneous components introduced into the
mantle become stretched, reduced in size and are more or less statistically distributed (e.g., Morgan, 1999; Meibom and Anderson, 2003;
Stracke et al., 2003b; Ito and Mahoney, 2005; Rudge et al., 2005;
Stracke et al., 2005; Kellogg et al., 2007). Large-scale statistical differences caused by differences in the relative abundance of the distinct
mantle components may thus occur, for example between ocean basins. Whether such potential large-scale “domains” in the Earth's mantle
(e.g. DUPAL (Dupré and Allègre, 1983; Hart, 1984) or SOPITA (Staudigel et
al., 1991)) actually represent a different abundance of mantle components or are simply an artifact from different sampling of a similar
distribution of mantle components (e.g., Morgan, 1999; Meibom and
Anderson, 2003; Ito and Mahoney, 2005) is a matter of debate. At
mid-ocean ridges, for example, the observed degree of isotopic variability
decreases as the scale of melting, or rate of processing of mantle material
(as measured by spreading rate), increases (e.g., Allègre et al., 1984;
Stracke et al., 2003b). These observations confirm that the scale of mantle
components is small compared to the maximum dimension over which
294
A. Stracke / Chemical Geology 330–331 (2012) 274–299
melts are produced and mixed beneath ridges. They also show that the
manner in which heterogeneous mantle components are represented in
the erupted melts could be influenced by large-scale tectonic or
geodynamic differences of the melting regions (e.g., spreading rate).
Supplementary data to this article can be found online at http://
dx.doi.org/10.1016/j.chemgeo.2012.08.007.
Acknowledgments
The editor Klaus Mezger is thanked for the invitation to write a review article about mantle heterogeneity. Many of the ideas and concepts presented in this review article have been developed through
a fruitful collaboration with Matthias Willbold. John Rudge is thanked
for his help with plotting principal components and actual data in one
diagram. Comments on an earlier version if this manuscript by Bill
White are much appreciated and, in addition to constructive formal
comments by Catherine Chauvel, Cornelia Class, and the editor
Klaus Mezger resulted in significant improvements and clarifications.
The author also thanks the GeoRoc and PetDB database teams for
their efforts and the DFG for support through grant STR 853/2-1.
References
Agranier, A., Blichert-Toft, J., Graham, D., Debaille, V., Schiano, P., Albarède, F., 2005. The
spectra of isotopic heterogeneities along the Mid-Atlantic ridge. Earth and Planetary Science Letters 238, 36–109.
Alard, O., Griffin, W.L., Pearson, N.J., Lorand, J.P., O'Reilly, S.Y., 2002. New insights into
the Re–Os systematics of sub-continental lithospheric mantle from in situ analysis
of sulphides. Earth and Planetary Science Letters 203, 651–663.
Alard, O., Luguet, A., Pearson, N.J., Griffin, W.L., Lorand, J.-P., Gannoun, A., Burton, K.W.,
O'Reilly, S.Y., 2005. In situ Os isotopes in abyssal peridotites bridge the isotopic gap
between MORBs and their source mantle. Nature 436, 1005–1008.
Albarède, F., 1995. Introduction to Geochemical Modeling. Cambridge University Press.
Albarède, F., 2001. Radiogenic ingrowth in systems with multiple reservoirs: applications to the differentiation of the mantle–crust system. Earth and Planetary Science
Letters 189, 59–73.
Allègre, C.J., 1982. Chemical geodynamics. Tectonophysics 81, 109–132.
Allègre, C.J., Lewin, E., 1995. Isotopic systems and stirring times of the Earth's mantle.
Earth and Planetary Science Letters 136, 629–646.
Allègre, C.J., Hamelin, B., Dupré, B., 1984. Statistical analysis of isotopic ratios in MORB:
the mantle blob cluster model and the convective regime of the mantle. Earth and
Planetary Science Letters 71, 71–84.
Allègre, C.J., Hamelin, B., Provost, A., Dupré, B., 1987a. Topology in isotopic multi-space
and origin of the mantle chemical heterogeneities. Earth and Planetary Science Letters 81, 319–337.
Allègre, C.J., Staudacher, T., Sarda, P., 1987b. Rare gas systematics: formation of the atmosphere, evolution and structure of the Earth's mantle. Earth and Planetary Science Letters 81, 127–150.
Andres, M., Blichert-Toft, J., Schilling, J.G., 2002a. Hafnium isotopes in basalts from the
southern Mid-Atlantic ridge from 40°S to 55°S: Discovery and Shona plume–ridge
interactions and the role of recycled sediment. Geochemistry, Geophysics,
Geosystems 3. http://dx.doi.org/10.1029/2002GC000324.
Andres, M., Blichert-Toft, J., Schilling, J.G., 2002b. Nature of the depleted upper mantle
beneath the Atlantic: evidence from Hf isotopes in normal mid-ocean ridge basalts
from 79°N to 55°S. Earth and Planetary Science Letters 225, 89–103.
Arevalo Jr., R., McDonough, W.F., 2010. Chemical variations and regional diversity observed in MORB. Chemical Geology 271, 70–85.
Armstrong, R.L., Harmon, R.S., 1981. Radiogenic isotopes: the case for crustal recycling
on a near-steady-state no-continental-growth earth. Philosophical Transactions of
the Royal Society 301, 443–472.
Arndt, N.T., Goldstein, S.L., 1989. An open boundary between lower continental crust
and mantle: its role in crust formation and crustal recycling. Tectonophysics 161,
201–212.
Barling, J., Goldstein, S.L., 1990. Extreme isotopic variations in Heard Island lavas and
the nature of mantle reservoirs. Nature 348, 59–62.
Behn, M.D., Kelemen, P.B., 2006. Stability of arc lower crust: insights from the Talkeetna
arc section, south central Alaska, and the seismic structure of modern arcs. Journal
of Geophysical Research 111, B11207. http://dx.doi.org/10.1029/2006JB004327.
Behn, M.D., Hirth, G., Kelemen, P.B., 2007. Trench-parallel anisotropy produced by
foundering of arc lower crust. Science 317, 108–111.
Behn, M.D., Kelemen, P.B., Hirth, G., Hacker, B.R., Massonne, H.-J., 2011. Diapirs as the
source of the sediment signature in arc lavas. Nature Geoscience 4, 1–6.
Beier, C., Stracke, A., Haase, K.M., 2007. The peculiar geochemical signatures of Sao
Miguel (Azores) lavas: metasomatised or recycled mantle sources? Earth and Planetary Science Letters 259, 186–199.
Ben Othman, D., White, W.M., Patchett, J., 1989. The geochemistry of marine sediments,
island arc magma genesis, and crust–mantle recycling. Earth and Planetary Science
Letters 94, 1–21.
Bizimis, M., Sen, G., Salters, V.J.M., Keshav, S., 2005. Hf–Nd–Sr isotope systematics of garnet pyroxenites from Salt Lake Crater, Oahu, Hawaii: evidence for a depleted component in Hawaiian volcanism. Geochimica et Cosmochimica Acta 69, 2629–2646.
Bizimis, M., Griselin, M., Lassiter, J.C., Salters, V.J.M., Sen, G., 2007. Ancient recycled
mantle lithosphere in the Hawaiian plume: osmium–hafnium isotope evidence
from peridotite mantle xenoliths. Earth and Planetary Science Letters 257, 259–
273.
Blichert-Toft, J., Frey, F.A., Albarede, F., 1999. Hf isotope evidence for pelagic sediments
in the source of Hawaiian basalts. Science 285, 879–882.
Boyd, O.S., Jones, C.H., Sheehan, A.F., 2004. Foundering lithosphere imaged beneath the
southern Sierra Nevada, California, USA. Science 305, 660–662.
Boyet, M., Carlson, R.W., 2005. 142Nd evidence for early (>4.53 Ga) global differentiation of the silicate Earth. Science 309, 576–581.
Brandon, A.D., Snow, J.E., Walker, R.J., Morgan, J.W., Mock, T.D., 2000. 190Pt–186Os and
187
Re–187Os systematics of abyssal peridotites. Earth and Planetary Science Letters
177, 319–335.
Brandon, A.D., Walker, R.J., Puchtel, I.S., 2006. Platinum–osmium isotope evolution of
the Earth's mantle: constraints from chondrites and Os-rich alloys. Geochimica et
Cosmochimica Acta 70, 2093–2103.
Brenan, J.M., Shaw, H.F., Phinney, D.L., Ryerson, F.J., 1994. Rutile-aqueous fluid
partitioning of Nb, Ta, Hf, Zr, U and Th—implications for high-field strength element depletions in island-arc basalts. Earth and Planetary Science Letters 128,
327–339.
Burton, K.W., Schiano, P., Birck, J.-L., Allègre, C.J., 1999. Osmium isotope disequilibrium
between mantle minerals in a spinel-lherzolite. Earth and Planetary Science Letters
172, 311–322.
Caro, G., Bourdon, B., Birck, J.L., Moorbath, S., 2003. 146Sm–142Nd evidence from Isua
metamorphosed sediments for early differentiation of the Earth's mantle. Nature
423, 428–432.
Chan, L.-H., Lassiter, J.C., Hauri, E.H., Hart, S.R., Blusztajn, J., 2009. Lithium isotope systematics of lavas from the Cook–Austral Islands: constraints on the origin of
HIMU mantle. Earth and Planetary Science Letters 277, 433–442.
Chauvel, C., Hofmann, A.W., Vidal, P., 1992. HIMU-EM: the French–Polynesian connection. Earth and Planetary Science Letters 110, 99–119.
Chauvel, C., Goldstein, S.J., Hofmann, A.W., 1995. Hydration and dehydration of oceanic
crust controls Pb evolution of the mantle. Chemical Geology 126, 65–75.
Chauvel, C., McDonough, W., Guille, G., Maury, R., Duncan, R., 1997. Contrasting old and
young volcanism in Rurutu Island, Austral chain. Chemical Geology 139, 125–143.
Chauvel, C., Lewin, E., Carpentier, M., Arndt, N.T., Marini, J.-C., 2008. Role of recycled
oceanic basalt and sediment in generating the Hf–Nd mantle array. Nature Geoscience 1, 64–67.
Christensen, U.R., Hofmann, A.W., 1994. Segregation of subducted oceanic crust in the
convecting mantle. Journal of Geophysical Research 99, 19,867–19,884.
Cipriani, A., Brueckner, H.K., Bonatti, E., Brunelli, D., 2004. Oceanic crust generated by
elusive parents: Sr and Nd isotopes in basalt-peridotite pairs from the MidAtlantic ridge. Geology 32, 657–660.
Class, C., Goldstein, S.L., 1997. Plume–lithosphere interactions in the ocean basins: constraints from the source mineralogy. Earth and Planetary Science Letters 150,
245–260.
Class, C., Le Roex, A.P., 2008. Ce anomalies in Gough Island lavas—trace element characteristics of a recycled sediment component. Earth and Planetary Science Letters
265, 475–486.
Class, C., Goldstein, S.L., Shirey, S.B., 2009. Osmium isotopes in Grande Comore lavas: a
new extreme among a spectrum of EM-type mantle endmembers. Earth and Planetary Science Letters 284, 219–227.
Clift, P., Vannucchi, P., 2004. Controls on tectonic accretion versus erosion in subduction zones—implications for the origin and recycling of the continental crust. Reviews of Geophysics 42, RG2001. http://dx.doi.org/10.1029/2003RG000127.
Cohen, R.S., O'Nions, R.K., 1982. Identification of recycled continental material in the
mantle from Sr, Nd and Pb isotope investigations. Earth and Planetary Science Letters 61, 73–84.
Coltorti, M., Bonadiman, C., O'Reilly, S.Y., Griffin, W.L., Pearson, N.J., 2010. Buoyant ancient continental mantle embedded in oceanic lithosphere (Sal Island, Cape Verde
Archipelago). Lithos 120, 223–233.
Cordier, C., Benoit, M., Hémond, C., Dyment, J., Le Gall, B., Briais, A., Kitazawa, M., 2010.
Time scales of melt extraction revealed by distribution of lava composition across a
ridge axis. Geochemistry, Geophysics, Geosystems 11, Q0AC06. http://dx.doi.org/
10.1029/2010GC003074.
Dasgupta, R., Hirschmann, M.M., Stalker, K., 2006. Immiscible transition from
carbonate-rich to silicate-rich melts in the 3 GPa melting interval of eclogite + CO2
and genesis of silica-undersaturated ocean island lavas. Journal of Petrology 47,
647–671.
Day, J.M.D., Pearson, D.G., Macpherson, C.G., Lowry, D., Carracedo, J.C., 2010. Evidence for
distinct proportions of subducted oceanic crust and lithosphere in HIMU-type mantle
beneath El Hierro and La Palma, Canary Islands. Geochimica et Cosmochimica Acta 74,
6565–6589.
de Wit, M.J., 1998. On Archean granites, greenstones, cratons and tectonics: does the
evidence demand a verdict? Precambrian Research 91, 181–226.
Devey, C.W., Albarède, F., Cheminee, J.-L., Michard, A., Muhe, R., Stoffers, P., 1990. Active submarine volcanism on the Society hotspot swell (West Pacific): a geochemical study. Journal of Geophysical Research 95, 5049–5066.
Donnelly, K.E., Goldstein, S.L., Langmuir, C.H., Spiegelman, M., 2004. Origin of enriched
ocean ridge basalts and implications for mantle dynamics. Earth and Planetary Science Letters 226, 347–366.
Dosso, L., Bougault, H., Beuzart, P., Calvez, J.Y., Joron, J.L., 1988. The geochemical structure of the south-east Indian ridge. Earth and Planetary Science Letters 88, 47–59.
A. Stracke / Chemical Geology 330–331 (2012) 274–299
Dostal, J., Cousens, B., Dupuy, C., 1998. The incompatible element characteristics of an
ancient subducted sedimentary component in ocean island basalts from French
Polynesia. Journal of Petrology 39, 937–952.
Doucelance, R., Escrig, S., Moreira, M., Gariepy, C., Kurz, M., 2003. Pb–Sr–He and trace element geochemistry of the Cape Verdes Archipelago. Geochimica et Cosmochimica
Acta 67, 3717–3733.
Douglass, J., Schilling, J.G., 2000. Systematics of three-component, pseudo-binary
mixing lines in 2D isotope ratio space representations and implications for mantle
plume–ridge interaction. Chemical Geology 163, 1–23.
Douglass, J., Schilling, J.G., Fontignie, D., 1999. Plume–ridge interactions of the Discovery and Shona mantle plumes with the southern Mid-Atlantic ridge (40°–55°S).
Journal of Geophysical Research 104, 2941–2962.
Ducea, M., Saleeby, J., 1998. A case for delamination of the deep batholitic crust beneath
the Sierra Nevada, California. International Geology Review 40, 78–93.
Dupré, B., Allègre, C.J., 1980. Pb–Sr–Nd isotopic correlation and the chemistry of the
North Atlantic mantle. Nature 286, 17–22.
Dupré, B., Allègre, C.J., 1983. Pb–Sr isotope variation in Indian Ocean and mixing phenomena. Nature 303, 142–146.
Dupuy, C., Barsczus, H.G., Liotard, J.M., Dostal, J., 1988. Trace element evidence for the
origin of ocean island basalts: an example from the Austral Islands (French Polynesia). Contributions to Mineralogy and Petrology 98, 293–302.
Eisele, J., Sharma, M., Galer, J.G., Blichert-Toft, J., Devey, C.W., Hofmann, A.W., 2002. The
role of sediment recycling in EMI inferred from Os, Pb, Hf, Nd, Sr isotope and trace
element systematics in the Pitcairn hotspot. Earth and Planetary Science Letters
196, 197–212.
Elliott, T., Plank, T., Zindler, A., White, W., Bourdon, B., 1997. Element transport from
slab to volcanic front at the Mariana arc. Journal of Geophysical Research 102,
14991–15019.
Elliott, T., Zindler, A., Bourdon, B., 1999. Exploring the Kappa Conundrum: the role of
recycling in the lead isotope evolution of the mantle. Earth and Planetary Science
Letters 169, 129–145.
Escrig, S., Capmas, F., Dupré, B., Allègre, C.J., 2004. Osmium isotopic constraints on the nature of the DUPAL anomaly from Indian mid-ocean-ridge basalts. Nature 431, 59–63.
Escrig, S., Doucelance, R., Moreira, M., Allègre, C.J., 2005a. Os isotope systematics in
Fogo Island: evidence for lower continental crust fragments under the Cape
Verde Southern Islands. Chemical Geology 219, 93–113.
Escrig, S., Schiano, P., Schilling, J.-G., Allègre, C.J., 2005b. Rhenium–osmium isotope systematics in MORB from the southern Mid Atlantic ridge (40°–50°S). Earth and
Planetary Science Letters 235, 528–548.
Farley, K.A., Natland, J.H., Craig, H., 1992. Binary mixing of enriched and undegassed
(primitive?) mantle components (He, Sr, Nd, Pb) in Samoan lavas. Earth and Planetary Science Letters 111, 183–199.
Fillerup, M.A., Knapp, J.H., Knapp, C.C., Raileanu, V., 2010. Mantle earthquakes in the absence of subduction? Continental delamination in the Romanian Carpathians. Lithosphere 2, 333–340.
Foley, S., Barth, M.G., Jenner, G.A., 2000. Rutile/melt partition coefficients for trace
elements and an assessment of the influence of rutile on the trace element characteristics of subduction zone magmas. Geochimica et Cosmochimica Acta 64, 933–938.
Fontignie, D., Schilling, J.-G., 1996. Mantle heterogeneities beneath the South Atlantic:
a Nd–Sr–Pb isotope study along the Mid-Atlantic ridge (3°S–46°S). Earth and Planetary Science Letters 142, 209–221.
Galer, S.J.G., 1991. Interrelationships between continental freeboard, tectonics and
mantle temperature. Earth and Planetary Science Letters 105, 214–228.
Galer, S.J.G., O'Nions, R.K., 1985. Residence time of thorium, uranium and lead in the
mantle with implications for mantle convection. Nature 316, 778–782.
Gannoun, A., Burton, K.W., Thomas, L.E., Parkinson, I.J., vanCalsteren, P., Schiano, P.,
2003. Osmium isotope heterogeneity in the constituent phases of mid-ocean
ridge basalts. Science 303, 70–72.
Gannoun, A., Burton, K.W., Parkinson, I.J., Alard, O., Schiano, P., Thomas, L.E., 2007. The
scale and origin of the osmium isotope variations in mid-ocean ridge basalts. Earth
and Planetary Science Letters 259, 541–556.
Gao, S., Luo, T.-C., Zhang, B.-R., Zhang, H.-F., Han, Y.-w, Zhao, Z.-D., Hu, Y.-K., 1998.
Chemical composition of the continental crust as revealed by studies in East
China. Geochimica et Cosmochimica Acta 62, 1959–1975.
Gao, S., Rudnick, R.L., Yuan, H.-L., Liu, X.-M., Liu, Y.-S., Xu, W.-L., Ling, W.-L., Ayers, J.,
Wang, X.-C., Wang, Q.-H., 2004. Recycling lower continental crust in the North
China craton. Nature 432, 892–897.
Gasperini, D., Blichert-Toft, J., Bosch, D., Del Moro, A., Macera, P., Telouk, P., Albarède, F.,
2000. Evidence from Sardinian basalt geochemistry for recycling of plume heads
into the Earth's mantle. Nature 408, 701–704.
Gast, P.W., Tilton, G.R., Hedge, C., 1964. Isotopic composition of lead and strontium
from Ascension and Gough Islands. Science 145, 1181–1185.
Geldmacher, J., Hoernle, K., Klügel, A., van den Bogaard, P., Bindeman, I., 2008. Geochemistry of a new enriched mantle type locality in the northern hemisphere: implications for the origin of the EM-I source. Earth and Planetary Science Letters 265,
167–182.
Gibson, S.A., Thompson, G.M., Day, J.A., Humphris, S.E., Dickin, A.P., 2005. Melt generation processes associated with the Tristan mantle plume: constraints on the origin
of EM-1. Earth and Planetary Science Letters 237, 744–767.
Goldstein, S.L., Soffer, G., Langmuir, C.H., Lehnert, K.A., Graham, D.W., Michael, P.J.,
2008. Origin of a “Southern Hemisphere” geochemical signature in the Arctic
upper mantle. Nature 453, 89–93.
Grégoire, M., Lorand, J.P., O'Reilly, S.Y., Cottin, J.Y., 2000. Armalcolite-bearing, Ti-rich
metasomatic assemblages in harzburgitic xenoliths from the Kerguelen Islands:
implications for the oceanic mantle budget of high-field strength elements.
Geochimica et Cosmochimica Acta 64, 673–694.
295
Gutiérrez-Alonso, G., Murphy, J.B., Frenández-Suárez, J., Weil, A.B., Franco, M.P.,
Gonzalo, J.C., 2011. Lithospheric delamination in the core of Pangea: Sm–Nd insights from the Iberian mantle. Geology 39, 155–158.
Halliday, A.N., Lee, D.C., Tommasini, S., Davies, G.R., Paslick, C.R., Fitton, J.G., James, D.E.,
1995. Incompatible trace-elements in OIB and MORB and source enrichment in the
sub-oceanic mantle. Earth and Planetary Science Letters 133, 379–395.
Hamelin, B., Allègre, C.J., 1985. Large-scale regional units in the depleted upper mantle
revealed by an isotope study of the South-West Indian Ridge. Nature 315, 196–199.
Hamelin, B., Dupré, B., Allègre, C.J., 1984. Lead–strontium isotopic variations along the
east Pacific Rise and the Mid-Atlantic ridge: a comparative study. Earth and Planetary Science Letters 67, 340–350.
Hamelin, B., Dupré, B., Allègre, C.J., 1986. Pb–Sr–Nd isotopic data of Indian Ocean
Ridges: new evidence of large scale mapping of mantle heterogeneities. Earth
and Planetary Science Letters 76, 288–298.
Hanan, B.B., Graham, D.W., 1996. Lead and helium isotope evidence from oceanic basalts for a common deep source of mantle plumes. Science 272, 991–995.
Hanan, B.B., Kingsley, R.H., Schilling, J.-G., 1986. Pb isotope evidence in the South Atlantic for migrating ridge-hotspot interactions. Nature 322, 137–144.
Hanan, B.B., Blichert-Toft, J., Pyle, D.G., Christie, D.M., 2004. Contrasting origins of the
upper mantle MORB source as revealed by Hf and Pb isotopes from the Australian–Antarctic discordance. Nature 432, 91–94.
Hanyu, T., Tatsumi, Y., Senda, R., Miyazaki, T., Chang, Q., Hirahara, Y., Takahashi, T.,
Kawabata, H., Suzuki, K., Kimura, J.-I., Nakai, S. i, 2011. Geochemical characteristics
and origin of the HIMU reservoir: a possible mantle plume source in the lower
mantle. Geochemistry, Geophysics, Geosystems 12, Q0AC09. http://dx.doi.org/
10.1029/2010gc003252.
Harper, C.L., Jacobsen, S.B., 1992. Evidence from coupled 147Sm–143Nd and
146
Sm–142Nd systematics for very early (4.5-Gyr) differentiation of the Earth's
mantle. Nature 360, 728–732.
Harrison, T.M., 2009. The Hadean crust: evidence from >4 Ga zircons. Annual Review
of Earth and Planetary Sciences 37, 479–505.
Harrison, T.M., Blichert-Toft, J., Muller, W., Albarede, F., Holden, P., Mojzsis, S.J., 2005.
Heterogeneous Hadean hafnium: evidence of continental crust at 4.4 to 4.5 Ga.
Science 310, 1947–1950.
Hart, S.R., 1984. A large-scale isotope anomaly in the southern-hemisphere mantle. Nature 309, 753–757.
Hart, S.R., 1988. Heterogeneous mantle domains—signatures, genesis and mixing chronologies. Earth and Planetary Science Letters 90, 273–296.
Hart, S.R., Ravizza, G., 1996. Os partitioning between phases in lherzolite and basalt. In:
Basu, A., Hart, S.R. (Eds.), Earth processes: reading the isotopic code. Geophysical
Monograph Series. AGU, Washington, pp. 123–134.
Hart, S.R., Staudigel, H., 1989. Isotopic characterization and identification of recycled
components. In: Hart, S.R., Gulen, L. (Eds.), Crust/mantle Recycling at Convergence
Zones. Kluwer Academic Publishers, pp. 15–28.
Hart, S.R., Schilling, J.G., Powell, J.L., 1973. Basalts from Iceland and along the Reykjanes
Ridge: Sr isotope geochemistry. Nature 246, 104–107.
Hart, S.R., Gerlach, D.C., White, W.M., 1986. A possible new Sr–Nd–Pb mantle array and
consequences for mantle mixing. Geochimica et Cosmochimica Acta 50, 1551–1557.
Hart, S.R., Hauri, E.H., Oschmann, L.A., Whitehead, J.A., 1992. Mantle plumes and
entrainment—isotopic evidence. Science 256, 517–520.
Harvey, J., Gannoun, A., Burton, K.W., Rogers, N.W., Alard, O., Parkinson, I.J., 2006. Ancient melt extraction from the oceanic upper mantle revealed by Re–Os isotopes
in abyssal peridotites from the Mid-Atlantic ridge. Earth and Planetary Science Letters 244, 606–621.
Harvey, J., Gannoun, A., Burton, K.W., Schiano, P., Rogers, N.W., Alard, O., 2010.
Unravelling the effects of melt depletion and secondary infiltration on mantle
Re–Os isotopes beneath the French Massif Central. Geochimica et Cosmochimica
Acta 74, 293–320.
Harvey, J., Dale, C.W., Gannoun, A., Burton, K.W., 2011. Osmium mass balance in peridotite and the effects of mantle-derived sulphides on basalt petrogenesis.
Geochimica et Cosmochimica Acta 75, 5574–5596.
Hauri, E.H., Hart, S.R., 1993. Re–Os isotope systematics of HIMU and EMII oceanic island basalts from the South-Pacific Ocean. Earth and Planetary Science Letters 114, 353–371.
Hauri, E., Shimizu, N., Dieu, J., Hart, S.R., 1993. Evidence for hotspot-related carbonatite
metasomatism in the oceanic upper mantle. Nature 365, 221–227.
Hauri, E.H., Lassiter, J.C., DePaolo, D.J., 1996. Osmium isotope systematics of drilled
lavas from Mauna Loa, Hawaii. Journal of Geophysical Research 101, 11793–11806.
Hawkesworth, C.J., Kemp, A.I.S., 2006a. The differentiation and rates of generation of
the continental crust. Chemical Geology 226, 134–143.
Hawkesworth, C.J., Kemp, A.I.S., 2006b. Evolution of the continental crust. Nature 443,
811–817.
Hawkesworth, C.J., Norry, M.J., Roddick, J.C., Vollmer, R., 1979. 143Nd/144Nd and 87Sr/86Sr
ratios from the Azores and their significance in LIL-element enriched mantle. Nature
280, 28–31.
Hawkesworth, C.J., Rogers, N.W., van Calsteren, P.W.C., Menzies, M.A., 1984. Mantle enrichment processes. Nature 311, 331–335.
Hawkesworth, C.J., Mantovani, M.S.M., Taylor, P.N., Palacz, Z., 1986. Evidence from the
Parana of south Brazil for a continental contribution to Dupal basalts. Nature 322,
356–359.
Hedge, C.E., 1966. Variations in radiogenic strontium found in volcanic rocks. Journal of
Geophysical Research 71, 6119–6126.
Hémond, C., Devey, C.W., Chauvel, C., 1994. Source compositions and melting processes
in the Society and Austral plumes (South-Pacific ocean)—element and isotope (Sr,
Nd, Pb, Th) geochemistry. Chemical Geology 115, 7–45.
Hermann, J., Rubatto, D., 2009. Accessory phase control on the trace element signature
of sediment melts in subduction zones. Chemical Geology 265, 512–526.
296
A. Stracke / Chemical Geology 330–331 (2012) 274–299
Hermann, J., Spandler, C.J., 2008. Sediment melts at sub-arc depths: an experimental
study. Journal of Petrology 49, 717–740.
Hirschmann, M.M., 2000. Mantle solidus: experimental constraints and the effects of
peridotite composition. Geochemistry, Geophysics, Geosystems 1, 1024. http://
dx.doi.org/10.1029/2000GC000070.
Hoernle, K., Tilton, G., Schmincke, H.-U., 1991. Sr–Nd Pb isotopic evolution of Gran
Canaria: evidence for shallow enriched mantle beneath the Canary Islands. Earth
and Planetary Science Letters 106, 44–63.
Hofmann, A.W., 1988. Chemical differentiation of the Earth—the relationship between
mantle, continental-crust, and oceanic-crust. Earth and Planetary Science Letters
90, 297–314.
Hofmann, A.W., 1997. Mantle geochemistry: the message from oceanic volcanism. Nature 385, 219–229.
Hofmann, A.W., 2003. Sampling mantle heterogeneity through oceanic basalts: isotopes and trace elements. In: Carlson, R.W., Holland, H.D., Turekian, K.K. (Eds.),
Treatise on Geochemistry: The Mantle and Core. Elsevier, pp. 61–101.
Hofmann, A.W., Jochum, K.P., Seufert, M., White, W.M., 1986. Nb and Pb in oceanic basalts—
new constraints on mantle evolution. Earth and Planetary Science Letters 79, 33–45.
Houseman, G.A., McKenzie, D.P., Molnar, P., 1981. Convective instability of a thickened
boundary layer and its relevance for the thermal evolution of continental convergent belts. Journal of Geophysical Research 86, 6115–6132.
Houtermans, F.G., 1953. Determination of the age of the earth from the isotopic composition of meteoritic lead. Nuovo Cimento, Series 9 10 (2), 1623–1633.
Ito, G., Mahoney, J.J., 2005. Flow and melting of a heterogeneous mantle: 1. Method and
importance to the geochemistry of ocean island and mid-ocean ridge basalts. Earth
and Planetary Science Letters 230, 29–46.
Ito, E., White, W.M., Gopel, C., 1987. The O, Sr, Nd and Pb isotope geochemistry of
MORB. Chemical Geology 62, 157–176.
Iwamori, H., Albarède, F., 2008. Decoupled isotopic record of ridge and subduction zone processes in oceanic basalts by independent component analysis. Geochemistry, Geophysics, Geosystems 9, Q04033. http://dx.doi.org/10.1029/2007gc001753.
Iwamori, H., Albarède, F., Nakamura, H., 2010. Global structure of mantle isotopic heterogeneity and its implications for mantle differentiation and convection. Earth
and Planetary Science Letters 299, 339–351.
Jackson, M.G., Hart, S.R., 2006. Strontium isotopes in melt inclusions from Samoan
basalts: implications for heterogeneity in the Samoan plume. Earth and Planetary
Science Letters 245, 260–277.
Jackson, M.G., Hart, S.R., Koppers, A.A.P., Staudigel, H., Konter, J., Blusztajn, J., Kurz, M.,
Russell, J.A., 2007. The return of subducted continental crust in Samoan lavas. Nature 448, 684–687.
Jenner, F.E., O'Neill, H.S.C., 2012. Analysis of 60 elements in 616 ocean floor basaltic
glasses. Geochemistry, Geophysics, Geosystems 13, Q02005. http://dx.doi.org/
10.1029/2011gc004009.
John, T., Scherer, E., Haase, K.M., Schenk, V., 2004. Trace element fractionation during
fluid-induced eclogitization in a subducting slab: trace element and Lu–Hf–Sm–
Nd isotope systematics. Earth and Planetary Science Letters 227, 441–456.
Johnson, M.C., Plank, T., 1999. Dehydration and melting experiments constrain the fate
of subducted sediments. Geochemistry, Geophysics, Geosystems 1, 1007. http://
dx.doi.org/10.1029/1999GC000014.
Jull, M., Kelemen, P.B., 2001. On the conditions for lower crustal convective instability.
Journal of Geophysical Research 106, 6423–6446.
Kamenetsky, V.S., Maas, R., Sushchevskaya, N.M., Norman, M.D., Cartwright, I., Peyve,
A.A., 2001. Remnants of Gondwanan continental lithosphere in oceanic upper
mantle: evidence from the South Atlantic Ridge. Geology 29, 243–246.
Kasting, J.F., Holm, N.G., 1992. What determines the volume of the oceans? Earth and
Planetary Science Letters 109, 507–515.
Kawabata, H., Hanyu, T., Chang, Q., Kimura, J.-I., Nichols, A.R.L., Tatsumi, Y., 2011. The
petrology and geochemistry of St. Helena alkali basalts: evaluation of the oceanic
crust-recycling model for HIMU OIB. Journal of Petrology 52, 791–838.
Kay, R.W., Kay, S.M., 1988. Crustal recycling and the Aleutian arc. Geochimica et
Cosmochimica Acta 52, 1351–1359.
Kay, R.W., Kay, S.M., 1991. Creation and destruction of lower continental crust.
Geologische Rundschau 80, 259–278.
Kay, R.W., Kay, S.M., 1993. Delamination and delamination magmatism. Tectonophysics
219, 177–189.
Kay, S.M., Coira, B., Viramonte, J., 1994. Young mafic back arc volcanicrocksas indicators
of continental lithospheric delamination beneath the Argentine Puna plateau, central Andes. Journal of Geophysical Research 99, 24,323–24,339.
Kelemen, P.B., 1995. Genesis of high Mg# andesites and the continental crust. Contributions to Mineralogy and Petrology 120, 1–19.
Kelemen, P.B., Hanghøj, K., Greene, A.R., 2003. One view of the geochemistry of
subduction-related magmatic arcs, with an emphasis on primitive andesite and
lower crust. In: Rudnick, R.L. (Ed.), The Crust. Elsevier, New York, pp. 593–659.
Kellogg, J.B., Jacobsen, S.B., O'Connell, R.J., 2007. Modeling lead isotopic heterogeneity in mid-ocean ridge basalts. Earth and Planetary Science Letters 262,
328–342.
Kemp, A.I.S., Hawkesworth, C.J., 2003. Granitic perspectives on the generation and secular evolution of the continental crust. In: Heinrich, D.H., Karl, K.T. (Eds.), Treatise
on Geochemistry. Pergamon, Oxford, pp. 349–410.
Kempton, P.D., Pearce, J.A., Barry, T.L., Fitton, G.M., Langmuir, C.H., Christie, D.M., 2002. Sr–
Nd–Pb–Hf isotope results from ODP Leg 187: evidence for mantle dynamics of the
Australian–Antarctic discordance and origin of the Indian MORB source. Geochemistry, Geophysics, Geosystems 3, 1074. http://dx.doi.org/10.1029/2002GC000320.
Keppie, D.F., Currie, C.A., Warren, C., 2009. Subduction erosion modes: comparing finite
element numerical models with the geological record. Earth and Planetary Science
Letters 287, 241–254.
Kessel, R., Schmidt, M.W., Ulmer, P., Pettke, T., 2005. Trace element signature of
subduction-zone fluids, melts and supercritical liquids at 120–180 km depth.
Nature 437, 724–727.
Klein, E.M., Langmuir, C.H., Zindler, A., Staudigel, H., Hamelin, B., 1988. Isotope evidence of a mantle convection boundary at the Australian–Antarctic discordance.
Nature 333, 623–629.
Klemme, S., Blundy, J.D., Wood, B.J., 2002. Experimental constraints on major and trace element partitioning during partial melting of eclogite. Geochimica et Cosmochimica
Acta 66, 3109–3123.
Klemme, S., Prowatke, S., Hameter, K., Günther, D., 2005. Partitioning of trace elements
between rutile and silicate melts: implications for subduction zones. Geochimica et
Cosmochimica Acta 69, 2361–2371.
Klimm, K., Blundy, J.D., Green, T.H., 2008. Trace element partitioning and accessory
phase saturation during H2O-saturated melting of basalt with implications for subduction zone chemical fluxes. Journal of Petrology 49, 523–553.
Kogiso, T., Tatsumi, Y., Nakano, S., 1997a. Trace element transport during dehydration processes in the subducted oceanic crust: 1. Experiments and implications for the origin of
ocean island basalts. Earth and Planetary Science Letters 148, 193–205.
Kogiso, T., Tatsumi, Y., Shimoda, G., Barsczus, H.G., 1997b. High μ (HIMU) ocean island
basalts in southern Polynesia: new evidence for whole mantle scale recycling of
subducted oceanic crust. Journal of Geophysical Research 102, 8085–8103.
Koornneef, J.M., Stracke, A., Bourdon, B., Meier, M.A., Jochum, K.P., Stoll, B., Grönvold, K., 2012.
Melting of a two-component source beneath Iceland. Journal of Petrology 53, 127–157.
Langmuir, C.H., Vocke, R.D., Hanson, G.N., Hart, S.R., 1978. A general mixing equation with
applications to Icelandic basalts. Earth and Planetary Science Letters 37, 380–392.
Lassiter, J., Hauri, E., 1998. Osmium-isotope variations in Hawaiian lavas: evidence for
recycled oceanic lithosphere in the Hawaiian plume. Earth and Planetary Science
Letters 164, 483–496.
Laubier, M., Schiano, P., Doucelance, R., Ottolini, L., Laporte, D., 2007. Olivine-hosted
melt inclusions and melting processes beneath the FAMOUS zone (Mid-Atlantic
ridge). Chemical Geology 240, 129–150.
Le Roex, A.P., Dick, H.J.B., Fisher, R.L., 1989. Petrology and geochemistry of MORB from
25°E to 46°E along the southwest Indian ridge: evidence for contrasting styles of
mantle enrichment. Journal of Petrology 30, 947–986.
Le Roex, A.P., Cliff, R.A., Adair, B.J.I., 1990. Tristan da Cunha, South Atlantic—geochemistry
and petrogenesis of a basanite–phonolite lava series. Journal of Petrology 31,
779–812.
Le Roux, P.J., Le Roex, A.P., Schilling, J.G., 2002a. MORB melting processes beneath the
southern Mid-Atlantic ridge (40–55°S): a role for mantle plume-derived pyroxenite. Contributions to Mineralogy and Petrology 144, 206–229.
Le Roux, P.J., Le Roex, A.P., Schilling, J.G., Shimizu, N., Perkins, W.W., Pearson, N.J.G.,
2002b. Mantle heterogeneity beneath the southern Mid-Atlantic ridge: trace element evidence for contamination of ambient asthenospheric mantle. Earth and
Planetary Science Letters 203, 479–498.
Levander, A., Schmandt, B., Miller, M.S., Liu, K., Karlstrom, K.E., Crow, R.S., Lee, C.T.A.,
Humphreys, E.D., 2011. Continuing Colorado plateau uplift by delamination-style
convective lithospheric downwelling. Nature 472, 461–465.
Liu, C.Z., Snow, J.E., Hellebrand, E., Brügmann, G., von der Handt, A., Büchl, A., Hofmann,
A.W., 2008. Ancient, highly depleted heterogeneous mantle beneath Gakkel ridge,
Arctic Ocean. Nature 452, 311–316.
Luguet, A., Alard, O., Lorand, J.P., Pearson, N.J., Ryan, C., O'Reilly, S.Y., 2001. Laserablation microprobe (LAM)-ICPMS unravels the highly siderophile element geochemistry of the oceanic mantle. Earth and Planetary Science Letters 189, 285–294.
Luguet, A., Graham Pearson, D., Nowell, G.M., Dreher, S.T., Coggon, J.A., Spetsius, Z.V.,
Parman, S.W., 2008. Enriched Pt–Re–Os isotope systematics in plume lavas
explained by metasomatic sulfides. Science 319, 453–456.
Lustrino, M., 2005. How the delamination and detachment of lower crust can influence
basaltic magmatism. Earth-Science Reviews 72, 21–38.
Lustrino, M., Wilson, M., 2007. The circum-Mediterranean anorogenic Cenozoic igneous province. Earth-Science Reviews 81, 1–65.
Lustrino, M., Melluso, L., Morra, V., 2000. The role of lower continental crust and lithospheric mantle in the genesis of Plio-Pleistocene volcanic rocks from Sardinia
(Italy). Earth and Planetary Science Letters 180, 259–270.
Maclennan, J., 2008a. Concurrent mixing and cooling of melts under Iceland. Journal of
Petrology 49, 1931–1953.
Maclennan, J., 2008b. Lead isotope variability in olivine-hosted melt inclusions from
Iceland. Geochimica et Cosmochimica Acta 72, 4159–4176.
Maclennan, J., McKenzie, D., Grönvold, K., Shimizu, N., Eiler, J., Kitchen, N., 2003a. Melt
mixing and crystallization under Theistareykir, northeast Iceland. Geochemistry,
Geophysics, Geosystems 4, 8624. http://dx.doi.org/10.1029/2003GC000558.
Maclennan, J., MCKenzie, D., Hilton, F., Grönvold, K., Shimizu, N., 2003b. Geochemical
variability from a single flow from northern Iceland. Journal of Geophysical Research 108. http://dx.doi.org/10.1029/2000JB000142.
Mahoney, J.J., Natland, J.H., White, W.M., Poreda, R., Bloomer, S.H., Fisher, R.L., Baxter,
A.N., 1989. Isotopic and geochemical provinces of the western Indian Ocean
spreading centers. Journal of Geophysical Research 94, 4033–4052.
Mahoney, J., Le Roex, A.P., Peng, Z., Fisher, R.L., Natland, J.H., 1992. Southwestern limits
of Indian Ocean ridge mantle and the origin of low 206Pb/204Pb mid-ocean ridge basalt: isotope systematic of the central southwest Indian ridge (17-50E). Journal of
Geophysical Research 97, 19,771–19,790.
Mahoney, J.J., White, W.M., Upton, B.G.J., Neal, C.R., Scrutton, R.A., 1996. Beyond EM-1:
lavas from Afanasy–Nikitin Rise and the Crozet Archipelago, Indian Ocean. Geology
24, 615–618.
Mahoney, J.J., Frei, R., Tejada, M.L.G., Mo, X.X., Leat, P.T., Nägler, T.F., 1998. Tracing the
Indian ocean mantle domain through time: isotopic results from old west Indian,
east Tethyan, and south Pacific seafloor. Journal of Petrology 39, 1285–1306.
A. Stracke / Chemical Geology 330–331 (2012) 274–299
Mahoney, J.J., Graham, D.W., Christie, D.M., Johnson, K.T.M., Hall, L.S., Vonderhaar, D.L.,
2002. Between a hotspot and a coldspot: isotopic variation in the southeast Indian
ridge asthenosphere, 86°E–118°E. Journal of Petrology 43, 1155–1176.
Martin, C.E., 1991. Osmium isotopic characteristics of mantle derived rocks. Geochimica et
Cosmochimica Acta 55, 1421–1434.
McCulloch, M.T., Bennett, V.C., 1994. Progressive growth of the Earth's continental
crust and depleted mantle: geochemical constraints. Geochimica et Cosmochimica
Acta 58, 4717–4738.
McCulloch, M.T., Gamble, J.A., 1991. Geochemical and geodynamical constraints on
subduction zone magmatism. Earth and Planetary Science Letters 102, 358–374.
McDonough, W.F., Sun, S.-s, 1995. The composition of the Earth. Chemical Geology 120,
223–253.
McKenzie, D.P., O'Nions, R.K., 1983. Mantle reservoirs and ocean island basalts. Nature
301, 229–231.
McKenzie, D., O'Nions, R.K., 1991. Partial melt distributions from inversion of rare earth
element concentrations. Journal of Petrology 32, 1021–1091.
McLennan, S.M., Taylor, S.R., 1983. Continental freeboard, sedimentation rates and
growth of continental crust. Nature 306, 169–172.
McLennan, S.M., Taylor, S.R., Hemming, S.R., 2005. Composition, differentiation, and evolution of continental crust: constraints from sedimentary rocks and heat flow. In:
Rushmer, T. (Ed.), Brown, M. Cambridge University Press, Cambridge, pp. 92–134.
Meibom, A., Anderson, D.L., 2003. The statistical upper mantle assemblage. Earth and
Planetary Science Letters 217, 123–139.
Meisel, T., Walker, R.J., Irving, A.J., Lorand, J.P., 2001. Osmium isotopic compositions of
mantle xenoliths: a global perspective. Geochimica et Cosmochimica Acta 65,
1311–1323.
Menzies, M.A., Murthy, V.R., 1980. Nd and Sr isotope geochemistry of hydrous mantle
nodules and their host alkali basalts; implications for local heterogeneities in
metasomatically veined mantle. Earth and Planetary Science Letters 46, 323–334.
Meyzen, C.M., Toplis, M.J., Humler, E., Ludden, J.N., Mével, C., 2003. A discontinuity in
mantle composition beneath the southwest Indian ridge. Nature 421, 731–733.
Meyzen, C.M., Ludden, J.N., Humler, E., Luais, B., Toplis, M.J., Mével, C., Storey, M., 2005.
New insights into the origin and distribution of the DUPAL isotope anomaly in the
Indian Ocean mantle from MORB of the Southwest Indian Ridge. Geochemistry,
Geophysics, Geosystems 6, Q11K11. http://dx.doi.org/10.1029/2005GC000979.
Meyzen, C.M., Blichert-Toft, J., Ludden, J.N., Humler, E., Mevel, C., Albarede, F., 2007. Isotopic portrayal of the Earth's upper mantle flow field. Nature 447, 1069–1074.
Michard, A., Montigny, R., Schlich, R., 1986. Geochemistry of the mantle below the
Rodriques Triple junction and the South East Indian Ridge. Earth and Planetary Science Letters 78, 104–114.
Milner, S.C., LeRoex, A.P., 1996. Isotope characteristics of the Okenyenya igneous complex, northwestern Namibia: constraints on the composition of the early Tristan
plume and the origin of the EM1 mantle component. Earth and Planetary Science
Letters 141, 277–291.
Morgan, J.W., 1986. Ultramafic xenoliths' clues to Earth's late accretionary history.
Journal of Geophysical Research 91, 12,375–12,387.
Morgan, J.P., 1999. Isotope topology of individual hotspot basalt arrays: mixing curves
or melt extraction trajectories. Geochemistry, Geophysics, Geosystems 1. http://
dx.doi.org/10.1029/1999GC000004.
Morgan, J.P., 2001. Thermodynamics of pressure release melting of a veined plum pudding mantle. Geochemistry, Geophysics, Geosystems 2. http://dx.doi.org/10.1029/
2000GC000049.
Morgan, J.P., Morgan, W.J., 1999. Two-stage melting and the geochemical evolution of
the mantle: a recipe for mantle plum-pudding. Earth and Planetary Science Letters
170, 215–239.
Morris, J.D., Leeman, W.P., Tera, F., 1990. The subducted component in island arc lavas:
constraints from Be isotopes and B–Be systematics. Nature 344, 31–36.
Nakamura, Y., Tatsumoto, M., 1988. Pb, Nd, and Sr isotopic evidence for a
multicomponent source for rocks of Cook–Austral Islands and heterogeneities of
mantle plumes. Geochimica et Cosmochimica Acta 52, 2909–2924.
Nauret, F., Abouchami, W., Galer, S.J.G., Hofmann, A.W., Hemond, C., Chauvel, C.,
Dyment, J., 2006. Correlated trace element-Pb isotope enrichments in Indian
MORB along 18–20°S, Central Indian Ridge. Earth and Planetary Science Letters
245, 137–152.
Nichols, G.T., Wyllie, P.J., Stern, C.R., 1994. Subduction zone melting of pelagic sediments constrained by melting experiments. Nature 371, 785–788.
Nielsen, R.L., Crum, J., Bourgeois, R., Hascall, K., Forsythe, L.M., Fisk, M.R., Christie, D.M.,
1995. Melt inclusions in high-An plagioclase from the Gorda ridge: an example of
the local diversity of MORB parent magmas. Contributions to Mineralogy and
Petrology 122, 34–50.
Niu, Y.L., O'Hara, M.J., 2003. Origin of ocean island basalts: a new perspective from petrology, geochemistry, and mineral physics considerations. Journal of Geophysical
Research 108, 19. http://dx.doi.org/10.1029/2002JB002048.
Niu, Y.L., Collerson, K., Batiza, R., Wendt, J., Regelous, M., 1999. Origin of enriched-type
mid-ocean ridge basalt at ridges far from mantle plumes: the East Pacific Rise at
11°20′N. Journal of Geophysical Research 104, 7067–7087.
Niu, Y.L., Regelous, M., Wendt, I.J., Batiza, R., O'Hara, M.J., 2002. Geochemistry of nearEPR seamounts: importance of source vs. process and the origin of enriched mantle
component. Earth and Planetary Science Letters 1–23.
Nutman, A.P., Friend, C.R.L., Horie, K., Hidaka, H., Martin, J., van Kranendonk, R.H.S.,
Vickie, C.B., 2007. The Itsaq Gneiss complex of Southern West Greenland and the
construction of Eoarchaean crust at convergent plate boundaries. Developments
in Precambrian Geology. Elsevier, pp. 187–218.
Panter, K.S., Blusztajn, J., Hart, S.R., Kyle, P.R., Esser, R., McIntosh, W.C., 2006. The origin
of HIMU in the SW Pacific: evidence from intra-plate Volcanism in Southern New
Zealand and SubAntarctic Islands. Journal of Petrology 47, 1673–1704.
297
Parai, R., Mukhopadhyay, S., Lassiter, J.C., 2009. New constraints on the HIMU mantle
from neon and helium isotopic compositions of basalts from the Cook–Austral
Islands. Earth and Planetary Science Letters 277, 253–261.
Parkinson, I.J., Hawkesworth, C.J., Cohen, A.S., 1998. Ancient mantle in a modern arc: osmium isotopes in Izu–Bonin–Mariana forearc peridotites. Science 281, 2011–2013.
Paul, D., White, W.M., Turcotte, D.L., 2002. Modeling the isotopic evolution of the Earth.
Philosophical Transactions of the Royal Society 360, 2433–2474.
Paul, B., Woodhead, J.D., Hergt, J., Danyushevsky, L., Kunihiro, T., Nakamura, E., 2011.
Melt inclusion Pb-isotope analysis by LA–MC–ICPMS: assessment of analytical performance and application to OIB genesis. Chemical Geology 289, 210–223.
Paulick, H., Münker, C., Schuth, S., 2010. The influence of small-scale mantle heterogeneities on Mid-Ocean Ridge volcanism: evidence from the southern Mid-Atlantic
ridge (7°30′S to 11°30′S) and Ascension Island. Earth and Planetary Science Letters
296, 299–310.
Pearson, D.G., Nowell, G.M., 2002. The continental lithospheric mantle: characteristics
and significance as a mantle reservoir. Philosophical Transactions of the Royal
Society 360, 2383–2410.
Pearson, D.G., Irvine, G.J., Ionov, D.A., Boyd, F.R., Dreibus, G.E., 2004. Re–Os isotope systematics and platinum group element fractionation during mantle melt extraction:
a study of massif and xenolith peridotite suites. Chemical Geology 208, 29–59.
Pfänder, J.A., Münker, C., Stracke, A., Mezger, K., 2007. Nb/Ta and Zr/Hf in ocean island
basalts—implications for crust–mantle differentiation and the fate of Niobium.
Earth and Planetary Science Letters 254, 158–172.
Pilet, S., Hernandez, J., Sylvester, P., Poujol, M., 2005. The metasomatic alternative for
ocean island basalt chemical heterogeneity. Earth and Planetary Science Letters
236, 148–166.
Pilet, S., Baker, M.B., Müntener, O., Stolper, E.M., 2011. Monte Carlo simulations of
metasomatic enrichment in the lithosphere and implications for the source of alkaline basalts. Journal of Petrology 52, 1415–1442.
Plank, T., 2005. Constraints from Thorium/Lanthanum on sediment recycling at subduction zones and the evolution of the continents. Journal of Petrology 46,
921–944.
Plank, T., Langmuir, C.H., 1998. The chemical composition of subducting sediment and
its consequences for the crust and mantle. Chemical Geology 145, 325–394.
Porter, K.A., White, W.M., 2009. Deep mantle subduction flux. Geochemistry, Geophysics,
Geosystems 10, Q12016. http://dx.doi.org/10.1029/2009gc002656.
Price, R.C., Kennedy, A.K., Riggs-Sneeringer, M., Frey, F.A., 1986. Geochemistry of basalts
from the Indian Ocean triple junction: implication for the generation and evolution
of Indian Ocean ridge basalts. Earth and Planetary Science Letters 78, 279–296.
Pyle, P.G., Christie, D.M., Mahoney, J.M., 1992. Resolving an isotopic boundary within
the Australian–Antarctic discordance. Earth and Planetary Science Letters 112,
161–178.
Pyle, D.G., Christie, D.M., Mahoney, J.J., Duncan, R.A., 1995. Geochemistry and geochronology of ancient southeast Indian and southwest Pacific seafloor. Journal of
Geophysical Research 100, 22, 261–22, 282.
Rampone, E., Hofmann, A.W., 2012. A global overview of isotopic heterogeneities in the
oceanic mantle. Lithos 148, 247–261.
Ranero, C.R., von Huene, R., 2000. Subduction erosion along the Middle America convergent margin. Nature 404, 748–752.
Rapp, R.P., Irifune, T., Shimizu, N., Nishiyama, N., Norman, M.D., Inoue, T., 2008. Subduction recycling of continental sediments and the origin of geochemically enriched
reservoirs in the deep mantle. Earth and Planetary Science Letters 271, 14–23.
Rehkämper, M., Hofmann, A.W., 1997. Recycled ocean crust and sediment in Indian
Ocean MORB. Earth and Planetary Science Letters 147, 93–106.
Reisberg, L., Lorand, J.P., 1995. Longevity of sub-continental mantle lithosphere
from osmium isotope systematics in orogenic peridotite massifs. Nature 376,
159–162.
Reisberg, L., Zindler, A., Marcantonio, F., White, W., Wyman, D., Weaver, B., 1993. Os
isotope systematics in ocean island basalts. Earth and Planetary Science Letters
120, 149–167.
Richardson, S.H., Erlank, A.J., Duncan, A.R., Reid, D.L., 1982. Correlated Nd, Sr, and Pb
isotopic variation in Walvis Ridge basalts and implications for the evolution of
their mantle source. Earth and Planetary Science Letters 59, 327–342.
Ringner, M., 2008. What is principal component analysis? Nature Biotechnology 26,
303–304.
Roden, M.K., Hart, S.R., Frey, F.A., Melson, W.G., 1984. Sr, Nd and Pb isotopic and REE
geochemistry of St. Paul's Rocks—the metamorphic and metasomatic development
of an alkali basalt mantle source. Contributions to Mineralogy and Petrology 85,
376–390.
Roy-Barman, M., Allègre, C.J., 1994. Os-187/Os-186 ratios of Mid Ocean ridge basalts
and abyssal peridotites. Geochimica et Cosmochimica Acta 58, 5043–5054.
Roy-Barman, M., Allègre, C.J., 1995. 187Os/186Os in oceanic island basalts: tracing oceanic crust recycling in the mantle. Earth and Planetary Science Letters 129, 145–161.
Rubin, K.H., Sinton, J.M., Maclennan, J., Hellebrand, E., 2009. Magmatic filtering of mantle compositions at mid-ocean-ridge volcanoes. Nature Geoscience 2, 321–328.
Rudge, J.F., 2006. Mantle pseudo-isochrons revisited. Earth and Planetary Science Letters 249, 494–513.
Rudge, J., McKenzie, D., Haynes, P.H., 2005. A theoretical approach to understanding the
isotopic heterogeneity of mid-ocean ridge basalt. Geochimica et Cosmochimica
Acta 69, 3873–3887.
Rudnick, R.L., 1995. Making continental crust. Nature 378, 571–578.
Rudnick, R.L., Fountain, D.M., 1995. Nature and composition of the continental crust: a
lower crustal perspective. Reviews of Geophysics 33, 267–309.
Rudnick, R.L., Gao, S., 2003. Composition of the continental crust. In: Carlson, R.W.,
Holland, H.D., Turekian, K.K. (Eds.), Treatise on Geochemistry: The Crust. Elsevier,
pp. 1–64.
298
A. Stracke / Chemical Geology 330–331 (2012) 274–299
Saal, A.E., Hart, S.R., Shimizu, N., Hauri, E.H., Layne, G.D., 1998. Pb isotopic variability in
melt inclusions from oceanic island basalts, Polynesia. Science 282, 1481–1484.
Saal, A.E., Hart, S.R., Shimizu, N., Hauri, E.H., Layne, G.D., Eiler, J.M., 2005. Pb isotopic
variability in melt inclusions from the EMI-EMII-HIMU mantle end-members and
the role of the oceanic lithosphere. Earth and Planetary Science Letters 240,
605–620.
Salters, V.J.M., Dick, H.J.B., 2002. Mineralogy of the mid-ocean-ridge basalt source from
neodymium isotopic composition of abyssal peridotites. Nature 418, 68–72.
Salters, V.J.M., Sachi-Kocher, A., 2010. An ancient metasomatic source for the Walvis
Ridge basalts. Chemical Geology 273, 151–167.
Salters, V.J.M., Stracke, A., 2004. Composition of the depleted mantle. Geochemistry,
Geophysics, Geosystems 5, 27, Q05004. http://dx.doi.org/10.1029/2003GC000597.
Salters, V.J.M., White, W.M., 1998. Hf isotope constraints on mantle evolution. Chemical
Geology 145, 447–460.
Salters, V.J.M., Mallick, S., Hart, S.R., Langmuir, C.H., Stracke, A., 2011. Domains of depleted
mantle; new evidence from hafnium and neodymium isotopes. Geochemistry, Geophysics, Geosystems 12, Q10017. http://dx.doi.org/10.1029/2011GC003874.
Schiano, P., Birck, J., Allègre, C.J., 1997. Osmium–strontium–neodymium–lead isotopic
covariations in mid-ocean ridge basalt glasses and the heterogeneity of the
upper mantle. Earth and Planetary Science Letters 150, 363–379.
Schilling, J.G., Zajac, M., Evans, R., Johnston, T., White, W., Devine, J.D., Kingsley, R.,
1983. Petrologic and geochemical variations along the Mid-Atlantic ridge from
29°N to 73°N. American Journal of Science 283, 510–586.
Schubert, G., Reymer, A.P.S., 1985. Continental volume and freeboard through geological time. Nature 316, 336–339.
Shimizu, N., 1998. The geochemistry of olivine-hosted melt inclusions in a FAMOUS basalt ALV519-4-1. Physics of the Earth and Planetary Interiors 107, 183–201.
Shimizu, N., Sobolev, A., Layne, G.D., 1998. In-situ Pb isotope analysis of olivine-hosted
melt inclusions from Mid-Ocean ridges. Eos Trans. AGU, 79. Fall Meet. Suppl., Abstract F950.
Shirey, S.B., Richardson, S.H., 2011. Start of the Wilson Cycle at 3 Ga shown by diamonds from subcontinental mantle. Science 333, 434–436.
Shirey, S.B., Bender, J.F., Langmuir, C.H., 1987. Three-component isotopic heterogeneity
near the Oceanographer Transform, Mid-Atlantic ridge. Nature 325, 217–223.
Shorttle, O., Maclennan, J., 2011. Compositional trends of Icelandic basalts: implications
for short length scale lithological heterogeneity in mantle plumes. Geochemistry,
Geophysics, Geosystems 12, Q11008. http://dx.doi.org/10.1029/2011gc003748.
Sims, K.W.W., DePaolo, D.J., 1997. Inferences about mantle magma sources from incompatible element concentration ratios in oceanic basalts. Geochimica et Cosmochimica
Acta 61, 765–784.
Sims, K.W.W., Goldstein, S.J., Blichert-Toft, J., Perfit, M.R., Kelemen, P., Fornari, D.J.,
Michael, P., Hart, S.R., DePaolo, D.J., Layne, G., Ball, L., Jull, M., Bender, J., 2002.
Chemical and isotopic constraints on the generation and transport of magma beneath the East Pacific Rise. Geochimica et Cosmochimica Acta 66, 3481–3504.
Sims, K.W.W., Blichert-Toft, J., Kyle, P.R., Pichat, S., Gauthier, P.-J., Blusztajn, J., Kelly, P.,
Ball, L., Layne, G., 2008. A Sr, Nd, Hf, and Pb isotope perspective on the genesis and
long-term evolution of alkaline magmas from Erebus volcano, Antarctica. Journal
of Volcanology and Geothermal Research 177, 606–618.
Slater, L., McKenzie, D., Grönvold, K., Shimizu, N., 2001. Melt generation and movement
beneath Theistareykir, NE Iceland. Journal of Petrology 42, 321–354.
Smithies, R.H., Champion, D.C., Van Kranendonk, M.J., Howard, H.M., Hickman, A.H.,
2005. Modern-style subduction processes in the Mesoarchaean: Geochemical evidence from the 3.12 Ga Whundo intra-oceanic arc. Earth and Planetary Science
Letters 231, 221–237.
Snow, J.E., Reisberg, L., 1995. Os Isotopic Systematics of the MORB mantle—results from
altered abyssal peridotites. Earth and Planetary Science Letters 133, 411–421.
Snow, J.E., Hart, S.R., Dick, H.J.B., 1994. Nd and Sr isotope evidence linking Mid-OceanRidge basalts and Abyssal Peridotites. Nature 371, 57–60.
Sobolev, A.V., Shimizu, N., 1993. Ultra-depleted primary melt included in an olivine
from the Mid-Atlantic ridge. Nature 363, 151–154.
Sobolev, A.V., Hofmann, A.W., Jochum, K.P., Kuzmin, D.V., Stoll, B., 2011. A young source
for the Hawaiian plume. Nature 476, 434–437.
Sours-Page, R., Johnson, K.T.M., Nielsen, R.L., Karsten, J.L., 1999. Local and regional variation of MORB parent magmas: evidence from melt inclusions from the Endeavour
segment of the Juan da Fuca ridge. Contributions to Mineralogy and Petrology 134,
342–363.
Spiegelman, M., Kelemen, P.B., 2003. Extreme chemical variability as a consequence of
channelized melt transport. Geochemistry, Geophysics, Geosystems 4, 1055.
http://dx.doi.org/10.1029/2002GC000336.
Stalder, R., Foley, S.F., Brey, G.P., Horn, I., 1998. Mineral-aqueous fluid partitioning of
trace elements at 900–1200 °C and 3.0–5.7 GPa: new experimental data for garnet,
clinopyroxene, and rutile, and implications for mantle metasomatism. Geochimica
et Cosmochimica Acta 62, 1781–1801.
Standish, J.J., Hart, S.R., Blusztjan, J., Dick, H.J.B., Lee, K.L., 2002. Abyssal peridotite osmium isotopic compositions from Cr-spinel. Geochemistry, Geophysics, Geosystems
3, 1004. http://dx.doi.org/10.1029/2001GC000161.
Staudigel, H., Park, K.H., Pringle, M., Rubenstone, J.L., Smith, W.H.F., Zindler, A., 1991.
The longevity of the South-Pacific isotopic and thermal anomaly. Earth and Planetary Science Letters 102, 24–44.
Stegman, D.R., Richards, M.A., Baumgardner, J.R., 2002. Effects of depth-dependent viscosity and plate motions on maintaining a relatively uniform mid-ocean ridge basalts reservoir in whole mantle flow. Journal of Geophysical Research 107, 2116.
http://dx.doi.org/10.1029/2001JB000192.
Stern, C.R., 2011. Subduction erosion: rates, mechanisms, and its role in arc magmatism
and the evolution of the continental crust and mantle. Gondwana Research 20,
284–308.
Stracke, A., 2008. Chemical geodynamics: tracking mantle depletion. Nature Geoscience 1, 215–216.
Stracke, A., Bourdon, B., 2009. The importance of melt extraction for tracing mantle
heterogeneity. Geochimica et Cosmochimica Acta 73, 218–238.
Stracke, A., Bizimis, M., Salters, V.J.M., 2003a. Recycling of oceanic crust: quantitative
constraints. Geochemistry, Geophysics, Geosystems 4, 8003. http://dx.doi.org/
10.1029/2001GC000223.
Stracke, A., Zindler, A., Salters, V.J.M., McKenzie, D., Blichert-Toft, J., Albarède, F., Grönvold,
K., 2003b. Theistareykir revisited. Geochemistry, Geophysics, Geosystems 4, 8507.
http://dx.doi.org/10.1029/2001GC000201.
Stracke, A., Hofmann, A.W., Hart, S.R., 2005. FOZO, HIMU and the rest of the mantle zoo.
Geochemistry, Geophysics, Geosystems 6, Q05007. http://dx.doi.org/10.1029/
2004GC000824.
Stracke, A., Snow, J.E., Hellebrand, E., von der Handt, A., Bourdon, B., Birbaum, K.,
Günther, D., 2011. Abyssal peridotite Hf isotopes identify extreme mantle depletion. Earth and Planetary Science Letters 308, 359–368.
Sun, W., Hu, Y., Kamenetsky, V.S., Eggins, S.M., Chen, M., Arculus, R.J., 2008. Constancy of Nb/U
in the mantle revisited. Geochimica et Cosmochimica Acta 72, 3542–3549.
Tackley, P.J., 2003. Mantle geochemical geodynamics. In: Schubert, G. (Ed.), Treatise on
Geophysics. Elsevier, Amsterdam, pp. 438–505.
Tatsumi, Y., 2000. Continental crust formation by crustal delamination in subduction
zones and complementary accumulation of the enriched mantle I component in
the mantle. Geochemistry, Geophysics, Geosystems 1, 1053. http://dx.doi.org/
10.1029/2000GC000094.
Tatsumoto, M., 1966. Genetic relations of oceanic basalts as indicated by lead isotopes.
Science 153, 1094–1101.
Tatsumoto, M., 1978. Isotopic composition of lead in oceanic basalt and its implication
to mantle evolution. Earth and Planetary Science Letters 38, 63–87.
Tatsumoto, M., Hedge, C.E., Engel, A.E.J., 1965. Potassium, rubidium, strontium, thorium,
uranium, and the ratio of strontium-87 to strontium-86 in oceanic tholeiitic basalt.
Science 150, 886–888.
Taylor, S.R., 1967. The origin and growth of continents. Tectonophysics 4, 17–34.
Taylor, S.R., MacLennan, S.M., 2009. Planetary Crusts: Their Composition, Origin and
Evolution. Cambridge University Press, p. 378.
Taylor, S.R., McLennan, S.M., 1995. The geochemical evolution of the continental crust.
Reviews of Geophysics 33, 241–265.
Van Kranendonk, M.J., Smithies, R.H., Hickman, A.H., Champion, D.C., Martin, J., van
Kranendonk, R.H.S., Vickie, C.B., 2007. Paleoarchean Development of a Continental
Nucleus: The East Pilbara Terrane of the Pilbara Craton, Western Australia, Developments in Precambrian Geology. Elsevier, pp. 307–337.
Vidal, P., Chauvel, C., Brousse, R., 1984. Large mantle heterogeneity beneath French
Polynesia. Nature 307, 536–538.
Vlastélic, I., Aslanian, D., Dosso, L., Bougault, H., Olivet, J.L., Géli, L., 1999. Large-scale
chemical and thermal division of the Pacific mantle. Nature 399, 345–350.
Vlastélic, I., Koga, K., Chauvel, C., Jacques, G., Télouk, P., 2009. Survival of lithium isotopic heterogeneities in the mantle supported by HIMU-lavas from Rurutu Island,
Austral Chain. Earth and Planetary Science Letters 286, 456–466.
Vollmer, R., 1976. Rb–Sr and U–Th–Pb systematics of alkaline rocks: the alkaline rocks
from Italy. Geochimica et Cosmochimica Acta 40, 283–295.
von Huene, R., Scholl, D.W., 1991. Observations at convergent margins concerning sediment subduction, subduction erosion, and the growth of continental crust. Reviews of Geophysics 29, 279–316.
Walker, R.J., Carlson, R.W., Shirey, S.B., Boyd, F.R., 1989. Os, Sr, Nd, and Pb isotope systematics of southern African peridotite xenoliths: implications for the chemical evolution
of subcontinental mantle. Geochimica et Cosmochimica Acta 53, 1583–1595.
Walker, R.J., Hanski, E., Vuollo, J., Liipo, J., 1996. The Os isotopic composition of Proterozoic upper mantle: evidence for chondritic upper mantle from the Outokumpu
ophiolite, Finland. Earth and Planetary Science Letters 141, 161–173.
Walker, R.J., Prichard, H.M., Ishiwatari, A., Pimentel, M., 2002. The osmium isotopic
composition of convecting upper mantle deduced from ophiolite chromites.
Geochimica et Cosmochimica Acta 66, 329–345.
Walter, M.J., 1998. Melting of garnet peridotite and the origin of komatiite and depleted lithosphere. Journal of Petrology 39, 29–60.
Warren, J.M., Shimizu, N., Sakaguchi, C., Dick, H.J.B., Nakamura, E., 2009. An assessment of
upper mantle heterogeneity based on abyssal peridotite isotopic compositions. Journal of Geophysical Research 114, B12203. http://dx.doi.org/10.1029/2008JB006186.
Waters, C.L., Sims, K.W.W., Perfit, M.R., Blichert-Toft, J., Blusztajn, J., 2011. Perspective
on the genesis of E-MORB from chemical and isotopic heterogeneity at 9°–10°N
East Pacific Rise. Journal of Petrology 52, 565–602.
Weaver, B.L., 1991. The origin of ocean island basalt end-member compositions: trace element and isotopic constraints. Earth and Planetary Science Letters 104, 381–397.
Weaver, B.L., Tarney, J., 1984. Empirical approach to estimating the composition of the
continental crust. Nature 310, 575–577.
Wedepohl, K.H., 1995. The composition of the continental crust. Geochimica et
Cosmochimica Acta 59, 1217–1232.
Weis, D., Frey, F.A., Leyrit, H., Gautier, I., 1993. Kerguelen Archipelago: geochemical and
isotopic study of the Southeast Province lavas. Earth and Planetary Science Letters
118, 101–119.
White, W.M., 1985. Sources of oceanic basalts—radiogenic isotopic evidence. Geology
13, 115–118.
White, W.M., 2010. Oceanic island basalts and mantle plumes: the geochemical perspective. Annual Review of Earth and Planetary Sciences 38, 133–160.
White, W., Duncan, R., 1995. Geochemistry and geochronology of the Society Islands:
new evidence for deep mantle recycling. In: Hart, S.R., Basu, A. (Eds.), Isotope Studies of Crust Mantle Evolution. AGU Monograph Series. American Geophysical
Union, Washington DC, pp. 183–206.
A. Stracke / Chemical Geology 330–331 (2012) 274–299
White, W.M., Dupré, B., 1986. Sediment subduction and magma genesis in the Lesser
Antilles—isotopic and trace-element constraints. Journal of Geophysical Research
91, 5927–5941.
White, W.M., Hofmann, A.W., 1982. Sr and Nd isotope geochemistry of oceanic basalts
and mantle evolution. Nature 296, 821–825.
White, W.M., Schilling, J.G., 1978. Nature and origin of geochemical variation in MidAtlantic ridge basalts from central North-Atlantic. Geochimica et Cosmochimica
Acta 42, 1501–1516.
Widom, E., Shirey, S.B., 1996. Os isotope systematics in the Azores: implications for
mantle plume sources. Earth and Planetary Science Letters 142, 451–465.
Widom, E., Hoernle, K.A., Shirey, S.B., Schmincke, H.U., 1999. Os isotope systematics in
the Canary Islands and Madeira: lithospheric contamination and mantle plume signatures. Journal of Petrology 40, 279–296.
Willbold, M., Stracke, A., 2006. Trace element composition of mantle end-members: implications for recycling of oceanic and upper and lower continental crust. Geochemistry,
Geophysics, Geosystems 7, Q04004. http://dx.doi.org/10.1029/2005GC001005.
Willbold, M., Stracke, A., 2010. Formation of enriched mantle components by recycling
of upper and lower continental crust. Chemical Geology 276, 188–197.
Woodhead, J.D., 1996. Extreme HIMU in an oceanic setting: the geochemistry of Mangaia
Island (Polynesia), and temporal evolution of the Cook–Austral hot spot. Journal of
Volcanology and Geothermal Research 72, 1–19.
Woodhead, J.D., Devey, C.W., 1993. Geochemistry of the Pitcairn Seamounts. 1. Source
character and temporal trends. Earth and Planetary Science Letters 116, 81–99.
Woodhead, J.D., McCulloch, M.T., 1989. Ancient seafloor signals in Pitcairn-Island lavas
and evidence for large-amplitude, small length-scale mantle heterogeneities. Earth
and Planetary Science Letters 94, 257–273.
Workman, R.K., Hart, S.R., 2005. Major and trace element composition of the depleted
mantle. Earth and Planetary Science Letters 231, 53–72.
299
Workman, R.K., Hart, S.R., Jackson, M., Regelous, M., Farley, K.A., Blusztajn, J., Kurz, M.,
Staudigel, H., 2004. Recycled metasomatized lithosphere as the origin of the Enriched
Mantle II (EM2) end-member: evidence from the Samoan volcanic chain. Geochemistry, Geophysics, Geosystems 5, Q04008. http://dx.doi.org/10.1029/2003GC000623.
Workman, R.K., Eiler, J.M., Hart, S.R., Jackson, M.G., 2008. Oxygen isotopes in Samoan
lavas: confirmation of continent recycling. Geology 36, 551–554.
Wörner, G., Zindler, A., Staudigel, H., Schmincke, H.U., 1986. Sr, Nd, and Pb isotope geochemistry of Tertiary and Quaternary alkaline volcanics from West-Germany.
Earth and Planetary Science Letters 79, 107–119.
Wright, E., White, W.M., 1987. The origin of Samoa—new evidence from Sr, Nd, and Pb
isotopes. Earth and Planetary Science Letters 81, 151–162.
Xie, S., Tackley, P.J., 2004. Evolution of U–Pb and Sm–Nd systems in numerical models
of mantle convection and plate tectonics. Journal of Geophysical Research 109,
B11204. http://dx.doi.org/10.1029/2004JB003176.
Zandt, G., Gilbert, H., Owens, T.J., Ducea, M., Saleeby, J., Jones, C.H., 2004. Active foundering of a continental arc root beneath the southern Sierra Nevada in California.
Nature 431, 41–46.
Zindler, A., Hart, S., 1986. Chemical geodynamics. Annual Review of Earth and Planetary Sciences 14, 493–571.
Zindler, A., Hart, S.R., Frey, F.A., Jakobsson, S.P., 1979. Nd and Sr isotope ratios and
rare-earth element abundances in Reykjanes Peninsula basalts—evidence for
mantle heterogeneity beneath Iceland. Earth and Planetary Science Letters 45,
249–262.
Zindler, A., Jagoutz, E., Goldstein, S., 1982. Nd, Sr and Pb isotopic systematics in a threecomponent mantle—a new perspective. Nature 298, 519–523.