Geoscience Frontiers 9 (2018) 1009e1019

H O S T E D BY Contents lists available at ScienceDirect

China University of Geosciences (Beijing)

Geoscience Frontiers
journal homepage: www.elsevier.com/locate/gsf

Research Paper

Metamorphic consequences of secular changes in oceanic crust

composition and implications for uniformitarianism in the geological
record
Richard M. Palin a, *, Brendan Dyck b
a
Department of Geology and Geological Engineering, Colorado School of Mines, Golden, CO 80401, USA
b
Department of Earth Sciences, Simon Fraser University, Burnaby, British Columbia V5A 1S6, Canada

a r t i c l e i n f o a b s t r a c t

Article history: Cooling of the Earth’s mantle since the Meso-Archean is predicted by thermal and petrological models to
Received 14 June 2017 have induced a secular change in the composition of primary mantle-derived magmas e and thus bulk
Received in revised form oceanic crust; in particular, suggesting a decrease in maficity over time. This hypothesis underpins
24 April 2018
several recent studies that have addressed key geological questions concerning evolving plate tectonic
Accepted 27 April 2018
styles, the rates and timing of continental crust formation, comparative planetology, and the emergence
Available online 8 May 2018
of complex life on Earth. Major, minor, and trace element geochemical analyses of (meta)mafic rocks
preserved in the geological record allows exploration of this theory, although no consensus currently
Keywords:
Archean
exists about the magnitude of this change and what compositions e if anything e constitute repre-
Mantle cooling sentative examples of Paleo-, Meso-, or Neo-Archean primitive oceanic crust. In this work, we review the
Oceanic crust current state of understanding of this issue, and use phase equilibria to examine the different mineral
Basalt assemblages and rock types that would form during metamorphism of basalt of varying maficity in
Metamorphism subduction zone environments. The presence (or absence) of such metamorphic products in the
Uniformitarianism geological record is often used as evidence for (or against) the operation of modern-day subduction-
driven plate tectonics on Earth at particular time periods; however, the control that secular changes in
composition have on the stability of mineral assemblages diagnostic of subduction-zone metamorphism
weakens such uniformitarianistic approaches. Geodynamic interpretations of the Archean metamorphic
rock record must therefore employ a different set of petrological criteria for determining tectonothermal
histories than those applied to Proterozoic or Phanerozoic equivalents.
Ó 2018, China University of Geosciences (Beijing) and Peking University. Production and hosting by
Elsevier B.V. This is an open access article under the CC BY-NC-ND license (http://creativecommons.org/
licenses/by-nc-nd/4.0/).

1. Introduction Furnes et al., 2014). As oceanic crust is produced via decompres-

sion melting of mantle peridotite at divergent plate margins
“By explaining past changes by analogy with present phenomena, a (McKenzie and Bickle, 1988), its structure and composition can
limit is set to conjecture, for there is only one way in which two provide critical first-order constraints on the geochemical and
things are equal, but there is an infinity of ways in which they could petrophysical properties of the Earth’s interior (Vervoort and
be supposed different.” e Hooykaas (1963). Blichert-Toft, 1999). Even basalts produced in intraplate geo-
dynamic scenarios, such as above the heads of mantle plumes, can
Basalt is the most widespread rock on the Earth’s surface today
provide critical information about the thermal and petrological
(Nielsen and Fisk, 2010), and investigations of Phanerozoic, Prote-
properties of the Earth’s shallow interior (Campbell and Griffiths,
rozoic, and Archean terranes worldwide suggest that it has been
1990). Identifying primary magmas or otherwise relatively unal-
similarly abundant throughout geological time (Condie, 1981;
tered components of oceanic crust in the geological record is thus of
prime importance, as such discoveries would help to resolve many
major issues in solid-Earth geoscience, such as constraining the
* Corresponding author.
thermal evolutions of the mantle over time and deducing whether
E-mail address: [email protected] (R.M. Palin). or not plate tectonics operated during the Archean (e.g. Davies,
Peer-review under responsibility of China University of Geosciences (Beijing). 1992; Hamilton, 1998, 2003; Stern, 2005; Van Hunen and Moyen,

https://doi.org/10.1016/j.gsf.2018.04.004
1674-9871/Ó 2018, China University of Geosciences (Beijing) and Peking University. Production and hosting by Elsevier B.V. This is an open access article under the CC BY-NC-
ND license (http://creativecommons.org/licenses/by-nc-nd/4.0/).
1010 R.M. Palin, B. Dyck / Geoscience Frontiers 9 (2018) 1009e1019

Modern-day MgO content (wt%) Archean MgO content (wt%)

0 0 0 0
low-MgO pillow lava basaltic and
komatiitic lavas
1 1 4 4
sheeted dykes sheeted dykes

Bulk = 10.3 wt. % (WK14)

Bulk = 18–24 wt. % (H10)

2 2 (e.g. Yellowknife)

~7–14 w
8 8

~7–9 wt. %

K17
isotropic
3 gabbro 3 gabbro and

t. %
12 troctolite 12
Depth (km)

Depth (km)

Depth (km)

Depth (km)
4 4
16 16
5 layered gabbro 5

20 20
cumulates
6 14.5 wt. % 6
picritic
cumulates
24 24
7 7

depleted peridotite 28 28
8 upper mantle 8 ?
(~38 wt. %) depleted
upper mantle
9 9 32 32

6 8 10 12 14 16 5 10 15 20 25 30

Figure 1. Schematic cross sections through modern-day and (interpreted) Archean oceanic crust and upper mantle. Modern-day structure, petrology, and MgO contents are from
White and Klein (2014), and Archean structure and petrology are after Foley et al. (2003) and Nair and Chacko (2008). Bulk MgO content for Archean crust is after Herzberg et al.
(2010) (H10), and two compositional profiles discussed by Klein et al. (2017) (K17) and in this study, are shown by thick green lines.

2012; Turner et al., 2014; Ernst, 2017; Ganne and Feng, 2017; Wade parageneses that are expected to form in different mafic rocks at
et al., 2017). these variable PeT conditions, which will allow more informed
Continued improvements to quantitative chemical analysis evaluation of the rock record and can provide greater insight into
techniques in recent yearsdalongside advances in large-scale data how metamorphic processes and products may have evolved
archiving and distributiondhave heralded the rise of Big Data throughout Earth history (e.g. Grambling, 1981; Sandiford, 1989;
within the geological community (Kattge et al., 2014). Centralized, Komiya et al., 2002; Martin and Moyen, 2002; Brown, 2007;
public-access repositories containing an extensive range of petro- Bradley, 2011; Zhai and Santosh, 2013; Dyck et al., 2015; Weller
logical analyses of rocks and minerals (e.g. EarthChem1) allow and St-Onge, 2017; Nicoli and Dyck, 2018).
straightforward processing of vast amounts of information. Indeed, In this paper, we first summarize the theory behind secular
several recent studies have analyzed large databases of Proterozoic change in bulk oceanic crust and MORB composition, and discuss
and Archean metabasalt bulk-rock compositions to examine the the extent to which metabasic rocks preserved in the geological
extent of secular variations in major, minor, and trace element record validate these predictions. We then present the results of
geochemistry, and elucidate the geodynamic settings of formation phase diagram analysis and interpretation demonstrating the
of the parental mafic magmas (Keller and Schoene, 2012; Furnes petrological consequences of such compositional changes at PeT
et al., 2014; Condie et al., 2016; Ganne and Feng, 2017). These conditions characteristic of modern-day, and early and late Archean
studies, however, have produced differing opinions on the extent to subduction zones. Finally, we discuss the implications of this
which the petrology and structure of primary mafic oceanic crust theoretical secular change on uniformitarianism-based approaches
and/or mid-ocean ridge basalt (MORB) has evolved over geological to interpreting the geological record, and the onset of modern-day
time, which are predicted by models of the mantle’s thermal his- subduction-driven plate tectonics.
tory to be significant (Abbott et al., 1994; Korenaga, 2008; Herzberg
et al., 2010; Ziaja et al., 2014).
1.1. Predicted secular change in bulk oceanic crust and uppermost
Nearly all basalts in the Precambrian rock record have under-
MOR basalt compositions
gone some degree of metamorphism (Gill, 1979), sometimes in
subduction zones at high-pressure/low-temperature (HTeLP)
It has long been known that the Earth has undergone cooling
conditions, or alternatively in collisional orogens along notably
since its formation (Thomson, 1862; See, 1907; Patterson, 1956);
hotter geothermal gradients (e.g. Brown, 2007). Secular changes in
however, the absolute magnitude and rate of this temperature
primary mantle-derived magmas, bulk oceanic crust, and/or MORB
change over time is poorly constrained. Conceptual models for this
compositions will thus have distinct petrological consequences on
thermal evolution fundamentally rely upon the balance at any
the metamorphic rock type produced in each tectonic setting (e.g.
point in time between internal heat production from the decay of
Palin and White, 2016). Thus, there is a need to define the mineral
radiogenic elements, and the loss of heat by mantle convection and
surface volcanism (Breuer and Spohn, 1993). The present-day ratio
of heat production to heat lossdthe convective Urey ratiodis
1
http://www.earthchem.org/. estimated to be 0.23
0.15 (Korenaga, 2008), and can be
 R.M. Palin, B. Dyck / Geoscience Frontiers 9 (2018) 1009e1019 1011

extrapolated back in time based on constraints provided by Table 1

geophysical and geochemical models of the Earth (e.g. Turcotte, Reported secular trends in oceanic crust compositions since the early Archean. Only
elements reported as showing secular variation are noted here: if absent from this
1980). Such an exercise generally predicts a concave-upwards table, they were either reported to have remained approximately constant over
thermal-evolution curve for the Earth’s mantle potential temper- time, or were not discussed. N.B. Mg# [ ¼ 100  Mg/(Mg þ Fe2þ) in cation units],
1
ature (TP) that peaks in the Meso-Archean (w2.8e3.2 Ga), inferring basalts with depleted mantle trace element geochemistry.
an ambient upper mantle that was around 300  C hotter than to- Lithology Increase Decrease Study
day’s value of w1350  C (Korenaga, 2006; Herzberg et al., 2010).
Tholeiitic basalt Al, Ti, Zr, P Cr, Ni, Co Gill (1979)
These data imply an average mantle cooling rate of w100  C/Ga Tholeiitic basalt Al Cr, Ni, Co Condie (1985)
since this time. Basalt K, Na Mg, Cr, Ni Keller and Schoene (2012)
Primary oceanic crust produced at spreading centers on Earth Basalt/greenstone Al, Ti, Zr Mg, Ni Furnes et al. (2014)
today is w6e7 km thick (Fig. 1), has an average bulk MgO content of Depleted mantle basalt1 Na, Ti, Mg# Fe, Mn Condie et al. (2016)
Archean basalt Al Mg Ganne and Feng (2017)
10e13 wt.%, and forms from small mantle melt fractions of 8e10%
(Herzberg et al., 2010). The uppermost portion of the mafic crust
(i.e. MORB) is typically comprised of pillow basalts with an MgO
content of w7 wt.% (White and Klein, 2014). A necessary petro- ensure significance of results (e.g. Ganne and Feng, 2017). Four
logical result of a hotter Archean mantle TP up to 1650  C is deeper major studies have been performed in recent years, with all having
and more voluminous melting of peridotite (25e45% melt fraction), identified secular changes in the MgO content (and other chemical
which likely produced a thicker (w25e40 km) crust (McKenzie and components) of basalt to different degrees (Table 1).
Bickle, 1988) with a greater average bulk MgO content of Furnes et al. (2014) and Condie et al. (2016) analyzed
w18e24 wt.% (Fig. 1; Abbott et al., 1994; Van Thienen et al., 2004; geochemical databases of metabasalt compositions obtained from
Herzberg et al., 2010). Geochemical models of Archean primary over 100 Precambrian greenstone belts worldwide, and used
magmas based on theoretical mantle protolith compositions, TP incompatible trace element ratios to determine probable geo-
values, and higher melt fractions predict that the uppermost, dynamic environments of formation for each locality. From these
MORB-like portions of this primitive crust would have MgO con- data, Furnes et al. (2014) reported a relatively large, statistically
tents in the range w11e15 wt.% (Ziaja et al., 2014). Secular cooling significant decrease in bulk-rock MgO content from w15 wt.% at 3.5
of mantle TP over time would thus suggest that the maficity of Ga to w7 wt.% today, although Condie et al. (2016) reported a
mantle-derived basalts has decreased throughout geological time, smaller decrease, with a mean MgO content of w9 wt.% for rocks
although a range of non-equilibrium processes (e.g. fractional older than 3 Ga and w7 wt.% for younger samples. However,
crystallization) may act to alter the mineralogy and composition of neither study accounted for spatiotemporal data clustering within
the final crystallized product. their respective databases, indicating that calculated average
compositions and trend magnitudes may have been influenced by
1.2. Observations from the geological record sampling bias to unknown (but potentially non-trivial) degrees.
In order to account for sample bias, Keller and Schoene (2012)
The Archean rock record is represented by 35 fragments of con- and Ganne and Feng (2017) produced weighted databases of sam-
tinental lithosphere e or cratons (Bleeker, 2003) e that cover ples from the mafic magma record, which were then analyzed via a
just w4.7% of the Earth’s surface (Artemieva, 2006). Such regions Monte Carlo resampling procedure designed to minimize selection
are typically characterized by sodic granitoids of tonalitee bias in space and time. While Keller and Schoene (2012) reported a
trondhjemiteegranodiorite (TTG) composition (Jahn et al., 1981; clear secular decrease in MgO content for the global geochemical
Moyen and Martin, 2012; White et al., 2017), and mafic- average mafic rock, they did not discriminate between possible
to-ultramafic volcanic rocks and sediments, most of which are environments of formation for each sample, meaning that this re-
typically metamorphosed to greenschist-facies PeT conditions ported trend lacks key geodynamic context needed to support a
(Condie, 1981). These “greenstone” terranes have been suggested by specific change in primitive basalt/oceanic crust composition over
some workers as representing e or else containing e variably altered time. By contrast, Ganne and Feng (2017) filtered out plutonic and
remnants of primary Archean oceanic crust, supported in some cases hypabyssal units to distinguish rocks derived directly from mantle
by modern-day, ophiolite-like macrostructures including pillow melting from those affected by subsequent evolution and fraction-
lavas and sheeted dykes (e.g., Helmstaedt et al., 1986; Kusky and ation processes in the lithosphere or continental crust. Further, no
Kidd, 1992; Furnes et al., 2007; Grosch and Slama, 2017). However, preselection of data via different trace element ratios was applied, to
many greenstone terranes may be autochthonous (Bickle et al., 1994; minimize the chance of introducing unintentional selection bias
Thurston, 2002), possibly representing continental rift or flood ba- based on incorrectly translating modern-day definitions to early-
salts (Bleeker, 2002), or formed in a non-MOR marine environment Earth geodynamic environments (cf. Li et al., 2015). From this
(i.e. arc- or plume-related), as inferred by incompatible trace- reduced dataset, provided in a supplementary database, Ganne and
element ratios characteristic of different modern-day mantle Feng (2017) reported successful primary magma solutions for
source regions (Pearce, 2008). As such, not all (meta)basalts can be Archean lithologies that had an average MgO content of w20 wt.%,
used to assess the existence or not of secular compositional change and equivalent interpreted basalt compositions with a mean MgO
in primary crust composition from the Archean to the modern day. content of w13 wt.%. Although it is not possible to show that these
Early studies of geochemical variations between modern-day Archean lavas formed in an ocean-ridge setting, this high-MgO
and Archean basalts focused mostly on differences in trace content exceeds that reported for modern-day MORB (White and
element ratios, and either made comparisons when bulk-rock data Klein, 2014), and agrees with reports of strongly mafic pillow lavas
were normalized to a common Mg# [ ¼ 100  Mg/(Mg þ Fe2þ)] (e.g. (>12 wt.%) with MOR-like geochemical affinity that occur in some
Gill, 1979) or applied arbitrary compositional limits to eliminate Paleo-Archean greenstone belts (e.g. Isua; Komiya et al., 2002, 2004).
highly evolved or highly fractionated lithologies from consider-
ation, such as pyroxene- or olivine-rich cumulates (e.g. Condie, 2. Petrological implications of secular compositional changes
1985). More recent works have taken advantage of Big Data to
allow more extensive spatial and temporal coverage of metabasalts The secular changes in primary basalt composition predicted by
worldwide, and have utilized more in-depth statistical tests to thermal modeling and observed in the geological record indicate
1012 R.M. Palin, B. Dyck / Geoscience Frontiers 9 (2018) 1009e1019

Figure 2. Classical metamorphic facies diagram (modified after Maruyama et al., 1996) overlain by pressureetemperature (PeT) profiles (red lines) reported from the literature to
characterize early and late Archean subduction channels (cf. Martin and Moyen, 2002). The late Archean subduction-zone geotherm is approximately 500  C/GPa and the early
Archean subduction-zone geotherm is approximately 350  C/GPa. Dotted band represents the range of PeT conditions calculated for descending slab surfaces in all modern-day
subduction zones (Syracuse et al., 2010). Geotherm marked 150  C/GPa denotes the limit of the Forbidden Zone (Liou et al., 2000). Abbreviations for facies and minerals are as
follows: Zeo ¼ zeolite, Pre-Pump ¼ prehniteepumpellyite, UHT ¼ ultrahigh temperature, Dia ¼ diamond, Gr ¼ graphite, Qtz ¼ quartz, Coe ¼ coesite, Jd ¼ jadeite, Ab ¼ albite. Water-
saturated basalt solidus (WBS) is after Vielzeuf and Schmidt (2001) and Palin et al. (2016b).

the importance of characterizing the metamorphic products of et al., 2008; Turner et al., 2014; Maruyama et al., 2017), it is
variably magnesian rocks that occur in all greenstone terranes generally thought that subduction had begun to operate on Earth
worldwide (e.g. Kato and Nakamura, 2003). Although Earth’s first on a global scale by w3.0 Ga (cf. Shirey and Richardson, 2011;
continents likely formed in intraplate tectonic settings in a so- Korenaga, 2013; Nutman et al., 2015). We adopt this interpreta-
called plume-lid or stagnant-lid geodynamic regime (e.g. tion, as it has the advantage of providing a consistent geodynamic
Smithies, 2000; Bédard, 2006; Van Kranendonk, 2010; Debaille framework with which to investigate the petrological conse-
et al., 2013; Palin et al., 2016c), there are multiple independent quences of secular compositional changes, although increased
lines of evidence suggesting that subduction had begun to operate buoyancy of Archean oceanic lithosphere may have promoted
on a global scale by the Meso-Archean, as detailed below. As this shallower subduction than commonly occurs today (Van Hunen
point in time coincides with the reported peak in mantle TP and Moyen, 2012). Petrological arguments for and against sub-
(Herzberg et al., 2010), and so the most MgO-rich mafic crust, it is duction having operated at this time are addressed in section 4 of
pertinent to examine the petrological products of subducted and this paper within the context of our petrological calculations.
metamorphosed oceanic crust of variable maficity. As such, we The geometries and thermal structures that may have charac-
have focused our petrological analysis on the parageneses that form terized Archean subduction zones are unknown, despite both being
at PeT conditions representative of subduction zone environments critical factors for determining the metamorphic and magmatic
on the modern-day Earth and the early and late Archean. We have rock types that form there (e.g. Peacock; 1996; Van Keken et al.,
considered a diverse set of protolith bulk compositionsdboth 2002). Thermo-mechanical numerical modeling of heat transfer
natural and syntheticdthat encompasses the expected range of in subduction zones has shown that the PeT evolutions of various
maficity reported for bulk oceanic crust and MORB throughout parts of the subducting lithosphere are primarily controlled by
geological time. plate velocity, slab age, and dip angle (Kirby et al., 1991), although
additional criteria such as mantle TP, heat capacity, and the nature
2.1. Early-Earth plate velocities and subduction geotherms of phase transitions also play a role (Peacock and Wang, 1999; King,
2001; Gerya et al., 2002, 2004; Billen, 2008; Palin et al., 2017).
Although there is long-standing debate within the geological While studies have shown that calculated slab-surface PeT condi-
community regarding the timing of initiation of plate tectonics (e.g. tions are most sensitive to plate velocity, slab age, and dip angle,
Hamilton, 2003; Stern, 2005; Condie and Kröner, 2008; Hopkins which can vary widely in natural geological systems (Syracuse
 R.M. Palin, B. Dyck / Geoscience Frontiers 9 (2018) 1009e1019 1013

Table 2
Bulk compositions of lithological components discussed in this work (wt.% oxide). 1sample CHK-MGO14; 2sample 485418; 3sample 02MB256. N.B. MAOC ¼ petrological model
of Archean oceanic crust with 11, 13, and 15 wt.% MgO. In all studies, measured Fe is reported here as FeO.

Lithology Reported age SiO2 TiO2 Al2O3 FeO MgO CaO Na2O K2O Study
Bulk oceanic crust Modern day (<200 Ma) 50.10 1.10 15.70 8.30 10.30 11.80 2.21 0.11 White and Klein (2014)
“Archean” (MAOC11) 48.50 0.72 16.65 8.13 11.00 12.78 1.71 e Ziaja et al. (2014)
“Archean” (MAOC13) 48.14 0.69 15.81 8.12 13.00 12.14 1.63 e Ziaja et al. (2014)
“Archean” (MAOC15) 47.78 0.65 14.97 8.11 15.00 11.50 1.54 e Ziaja et al. (2014)
Modern MORB Modern day (<200 Ma) 50.06 1.52 15.00 10.36 7.71 11.46 2.52 0.19 White and Klein (2014)
Olivine-bearing basalt1 Paleoproterozoic (ca. 1870 Ma) 49.96 0.79 11.00 10.97 14.07 11.57 1.30 0.09 Hynes and Francis (1982)
High-MgO pillow basalt2 Mesoarchean (ca. 3075 Ma) 48.41 0.43 10.87 11.26 14.90 11.26 1.23 0.15 Polat et al. (2008)
Komatiitic basalt3 Paleoarchean (ca. 3450 Ma) 48.75 0.48 10.32 10.90 16.19 10.49 1.42 0.03 Kato and Nakamura (2003)

et al., 2010; Van Hunen and Moyen, 2012; Sizova et al., 2015), late-Archean (w300  C/GPa) subduction zone geotherms of Martin
various assumptions can be made about these criteria in the and Moyen (2002) (Fig. 2).
Archean from many independent lines of evidence.
Early studies of Archean geodynamics favored the idea that plate 2.2. Calculated metamorphic parageneses
tectonics operated at a faster rate than on Earth today (Hargraves,
1986), given that the dissipation of internal heat from a signifi- Mafic igneous rocks and their metamorphosed equivalents
cantly hotter mantle would be most efficient in a scenario involving show most compositional variability within the MnOeNa2Oe
rapid sea-floor spreading and subduction of relatively young CaOeK2OeFeOeMgOeAl2O3eSiO2eH2OeTiO2eFe2O3eCO2 (MnNC
(20e30 Myr) crust (e.g. Bickle, 1978; Abbott and Hoffman, 1984). KFMASHTOc) chemical system; however, first-order in-
Archean spreading rates up to w40 cm/yr were proposed by some terpretations of key phase changes need only consider Na2OeCaOe
workers (e.g. Dewey and Windley, 1981; Nisbet and Fowler, 1983), FeOeMgOeAl2O3eSiO2eH2O (e.g. Spear, 1995). For projection
which are significantly higher than present-day rates of 1e10 cm/yr and visualization purposes, further reduction into the ACF
(Solomon et al., 1975). While some workers continue to support (AlO3/2eCaOeFeO þ MgO) ternary compositional system was per-
this ‘fast-plate’ hypothesis (Van Kranendonk and Kirkland, 2013), formed here using Perple_X (Connolly, 2009) and the thermody-
Archean plate velocities are now thought to have been more slug- namic dataset of Holland and Powell (2011). A high activity of H2O
gish than today (Korenaga, 2013), as indicated by paleomagnetic (i.e. free fluid) was assumed for metamorphism at subsolidus
and paleogeographic data (Condie, 2015), the geochemical budget conditions (Schmidt and Poli, 1998) alongside the occurrence of
of isotope exchange between the hydrosphere and Earth’s interior minor quartz or coesite (SiO2), depending on pressure. While
(Padhi et al., 2012), calculated constraints on the Urey ratio over modern-day petrological techniques can produce bulk-rock specific
time (Korenaga, 2008), and passive margins having had longer equilibrium phase assemblage diagrams (pseudosections) for
lifespans in the Precambrian than in the Phanerozoic (Bradley, metamafic rocks (Powell et al., 1998; Green et al., 2016), these
2008). calculated equilibria are often distinctly sensitive to small varia-
Quantitative estimates of Archean plate velocities based on a tions in bulk composition oxide ratios (e.g. Stüwe, 1997; Guevara
model of time-dependent mantle heat flux were presented by and Caddick, 2016; Palin et al., 2016a) and complexities in the
Korenaga (2006). When scaled against an average present-day stabilities of minor phases can mask broader-scale e and more
value of 4 cm/yr, the ProterozoiceArchean boundary (2.5 Ga) was important e petrological features. Indeed, as no one particular bulk
characterized by a velocity of w3 cm/yr, and the ArcheaneHadean composition can be shown to be completely representative of pri-
boundary (4 Ga) by w2 cm/yr. In addition, subducted crust was mary mafic oceanic crust/MOR basalt at any point in the geological
likely to have been older and colder (Korenaga, 2013) than the past (section 1.2), we approach the problem by investigating the
w60-Myr present-day average (Bickle, 1978). The effect that slower effects that general compositional trends in the major varying
and older descending crust may have on subduction zone thermal components MgO and Al2O3 (Table 1) have on first-order changes
structure can be interpreted on a first-order basis via comparison in metamorphic mineral assemblages. We note that reported
with present-day analogues that show similar physical parameters, FeO contents of mafic crust have remained generally constant
such as the Lesser Antilles, where subducting slab surfaces have throughout geological time (e.g. Condie et al., 2016; Table 1),
PeT profiles with geotherms of w300e400  C/GPa (Syracuse et al., meaning that changes in MgO content itself lead to variation in
2010). overall bulk-rock Mg#.
Alongside these theoretical arguments, first-order evidence of All bulk-rock compositions considered herein (Table 2) are
relatively cool subduction during the MesoeArchean (ca. 3.2 Ga, shown in ACF compositional space in Fig. 3a, alongside common
Moyen et al., 2006; and ca. 2.87 Ga, Mints et al., 2010) and Paleo- minerals that form in metabasic rocks in subduction zone envi-
Proterozoic (ca. 2.2e1.8 Ga; Ganne et al., 2011; Weller and St- ronments (e.g. Rebay et al., 2010; Maldonado et al., 2016). Observed
Onge, 2017) is recorded within high-pressure metasediments and modern-day (White and Klein, 2014) and calculated Archean
metabasites preserved in the geological record. In these works, (MAOC: Ziaja et al., 2014) bulk oceanic crust compositions are
conventional thermobarometry and phase equilibrium modeling shown in Fig. 3b for reference, and a series of natural basalts of Neo-
results indicated peak subduction-related metamorphism along Proterozoic to Paleo-Archean age (Table 2) of varying maficity are
geotherms of w350e400  C/GPa, which are indeed similar to most shown in Fig. 3c, which we use to examine the petrological effects
modern-day slab-top PeT profiles (Fig. 2). As some geodynamic of reported secular change in MORB composition. This uppermost
models suggest that increased lithospheric buoyancy in the early basaltic portion of the oceanic crust e MORB e is specifically
Archean caused subduction to have operated at shallower angles examined here as it is likely to be hydrated during crust formation,
than in the late- and post-Archean (Van Hunen and Moyen, 2012; seafloor metasomatism (Peck et al., 2001), or lithospheric defor-
Sizova et al., 2014), we have considered changes in phase assem- mation at plate boundaries (Korenaga, 2017), and so undergoes
blages predicted to be stable in a range of basaltic compositions fluid-mediated equilibrium metamorphism during subduction
along both the warm early-Archean (w600  C/GPa) and cold (Guiraud et al., 2001). These natural samples were selected from
1014 R.M. Palin, B. Dyck / Geoscience Frontiers 9 (2018) 1009e1019

Figure 3. ACF (AlO3/2eCaOeFeO þ MgO) chemographic projections showing the compositions of mafic oceanic crustal lithologies of various ages. End-member mineral abbre-
viations are after Kretz (1983). (a) ACF diagram showing minerals commonly observed in metamorphosed mafic rocks alongside each bulk composition discussed in the text and
Table 2. (b) Magnification of part (a), showing the position in ACF space of estimated compositions of bulk oceanic crust today (WK14: White and Klein, 2014) and in the Archean
(Z14: Ziaja et al., 2014). N.B. MAOC ¼ petrological model of Archean oceanic crust with 11, 13, and 15 wt.% MgO. (c) Magnification of part (a) showing the position in ACF space of
basalt compositions of early Archean to modern-day age. HF82 ¼ Hynes and Francis (1982), KN03 ¼ Kato and Nakamura (2003), P08 ¼ Polat et al. (2008), WK14 ¼ White and Klein
(2014). See Table 2 for full compositions.

greenstone terranes identified by Furnes et al. (2014) as having lithologies are rare in Archean cratons (Bédard, 2017), and so e if
MOR affinity. Average modern-day MORB (White and Klein, 2014) formed on the early Earth e likely terminally descended into the
is also shown on Fig. 3c for comparative purposes. All mineral ab- aesthenospheric mantle as non-exhumable crust (Bédard, 2006).
breviations are after Kretz (1983), including Ab ¼ albite, A similar petrographic treatment is applied to theoretical sub-
Jd ¼ jadeite, Qtz ¼ quartz, and H2O ¼ water as projected phases. duction along a relatively cool geotherm (w300  C/GPa; Fig. 2),
While quartz may not always be present in metamorphosed ul- thought to be representative of late-/post-Archean slab-top PeT
tramafic rocks, examination of phase relations in ACF composi- paths (Fig. 4eeh) (Martin and Moyen, 2002). At 500  C and 1.6 GPa
tional space is valid if silica (SiO2) is a component in associated along this geotherm, representative of the earliest stages of sub-
projected phases, such as albite or jadeite, and if all phases on the duction, all considered bulk-rock compositions fall within a quad-
projection plane contain it (cf. Green et al., 1968). rilateral defined by zoisite, garnet, clinopyroxene, and sodic
A series of ACF compatibility triangles showing equilibrium amphibole (Fig. 4e). At lower blueschist-facies conditions, this four-
phase assemblage changes at PeT conditions along an early phase assemblage is divided by a zoisiteeglaucophane tie line,
Archean warm geotherm (w600  C/GPa; Fig. 2) are shown in below which all considered bulk-rock compositions lie (Fig. 4e). At
Fig. 4aed. The bulk compositions reported in Table 2 and shown in higher temperature conditions representative of the upper blues-
Fig. 3c may be separated into low-MgO modern-day MORB (Mo- chist facies, a tie-line switch occurs stabilizing garnete
MORB) and high-MgO ArcheaneProterozoic basalts (High-Mg APB). clinopyroxene assemblages in place of zoisiteeglaucophane
At relatively low-grade epidote-amphibolite facies conditions assemblages. At this point, low-MgO bulk compositions become
(500  C, 0.8 GPa), low-MgO protoliths would be characterized un- separated in ACF compositional space from high-MgO bulk com-
der metamorphic conditions by chlorite-, zoisite-, and actinolite- positions, with the former containing zoisite and the latter con-
rich assemblages, whereas basalts with higher MgO contents taining sodic amphibole (Fig. 4e). At higher-grade conditions along
would be zoisite-free. During prograde metamorphism, zoisite is this geotherm (600  C, 2 GPa), representative of the amphibole-
replaced by plagioclase feldspar and/or lawsonite during the tran- eclogite facies of Maruyama et al. (1996), all bulk compositions
sition into lower amphibolite-facies PeT conditions, with garnet considered are predicted to contain eclogitic parageneses (garnet
potentially forming in Al2O3-rich and MgO-poor lithologies and clinopyroxene), with or without zoisite, glaucophane, and talc
(Fig. 4b). With increasing metamorphism and subduction to greater (Fig. 4f). Similar equilibrium assemblages occur at higher grade
depths, transition to upper amphibolite/lower granulite-facies (750  C, 2.6 GPa; Fig. 4g), with the sole difference pertinent to the
conditions (700  C, 1.2 GPa) produces plagioclase/lawsonitee rock types considered being the stabilization of kyanite in high-
garnetehornblende assemblages in low-MgO protoliths, and gar- Al2O3/low-MgO bulk compositions. No significant changes in ACF
netehornblendeecummingtonite/orthopyroxene assemblages in phase equilibria occur during deep subduction to ultrahigh-pressure
high-MgO protoliths (Fig. 4c). The garnetehornblende tie line that “dry eclogite”-facies conditions (>800  C, >2.9 GPa; Fig. 4h), aside
forms in ACF compositional space distinctly separates these two from the transformation of quartz to coesite and likely loss of free
protolith groups, although pressures representative of this geo- H2O due to prograde devolatilization. All but the most magnesian
therm are insufficient to destabilize amphibole and thus produce protolith (komatiitic basalt) are predicted to transform to eclogite
eclogite (sensu stricto) or hornblende-free two-pyroxene granulite (sensu stricto), while these near-ultramafic precursor basalts would
in these bulk compositions. However, at granulite-facies conditions instead transform to garnet pyroxenite (Fig. 4h; Johnson et al., 2014),
of >800  C and >1.5 GPa, the garnetehornblende tie line is as also predicted by experimental petrology (Foley et al., 2003; Ziaja
replaced by the garneteclinopyroxene tie line (Fig. 4d), suggesting et al., 2014). Metamorphosed olivine-normative basalt is unlikely to
that low-MgO protoliths would produce partially melted high- form quartz/coesite at eclogite-facies conditions, with previous
pressure granulite during Archean subduction, but high-MgO pro- calculations predicting its presence during subduction up to P
toliths would instead produce garnet pyroxenite. Such pyroxenitic w22 kbar (Palin and White, 2016); however, metamorphosed
 R.M. Palin, B. Dyck / Geoscience Frontiers 9 (2018) 1009e1019 1015

Figure 4. ACF chemographic projections showing phase equilibria at pressureetemperature (PeT) conditions along the (aed) early-Archean “warm” geotherm (600  C/GPa) and
(eeh) late-/post-Archean “cool” geotherm (300  C/GPa) of Martin and Moyen (2002). End-member mineral abbreviations, including those in projection, are after Kretz (1983). Bulk
compositions of modern-day and Archean/Proterozoic basalts are shown for reference. The compositions used for ACF plots are those shown in Fig. 3.

quartz-normative tholeiitic basalts commonly contain either poly- and Feng (2017) documented significant decreases in global
morph at dry eclogite conditions (Fig. 2; St-Onge et al., 2013). average (meta)basalt maficity since the Meso-Archean, which
approached w15 wt.% MgO at 3.5 Ga, Condie et al. (2016) reported a
3. Discussion lesser decrease of just w2 wt.% over the same period. If MORB has
retained a more-or-less constant MgO content throughout
Determining whether components of oceanic crust preserved in geological time, mass-balance constraints require that the lower-
the geological record show time-dependent evolutions in bulk most portions of Archean oceanic crust must have been signifi-
composition is critical to validating predictions made by thermal cantly more magnesian than modern-day examples (Fig. 1), as the
and petrological models (e.g. Korenaga, 2006). Such prima facie overall bulk-crustal MgO content was much higher. However, no
evidence of secular trends would also have immense value in such rocks in the geological record are thought to be definitive
bounding changes in keydyet poorly constraineddwhole-Earth examples of these lower-crustal cumulate lithologies (Bickle et al.,
thermal parameters over time, such as the convective Urey ratio 1994). By contrast, some authors suggest that Archean oceanic crust
and mantle TP (Herzberg et al., 2010; Ganne and Feng, 2017). may have consisted predominantly of komatiite-like high-MgO
However, despite extensive field investigations, no coherent basalts, with no significant vertical stratification in composition or
oceanic crustal sections/ophiolites have been documented from structure (Fig. 1) (Arndt, 1983; Klein et al., 2017), with the low-MgO
Archean terranes (Bickle et al., 1994), and some reported occur- basalts found in many greenstone belts (Condie and Aster, 2009)
rences have since been disputed by other members of the com- potentially having been derived via burial metamorphism and
munity (e.g. Furnes et al., 2007; Nutman and Friend, 2007), and so partial melting of these original high-MgO picritic/komatiitic pro-
no primary constraints exist on the extent to which early-Earth toliths. If true, this supports the hypothesis that petrological in-
oceanic crust was structurally or compositionally stratified (cf. dicators of subduction e as defined by assemblages that stabilize in
Fig. 1). modern-day, low-MgO MORB e would not have formed on the
early Earth, even if identical metamorphic PeT conditions were
3.1. Secular change in MOR basalt composition: fact or fiction? reached (Palin and White, 2016).
If each of the four main studies discussed herein have used
While theoretical arguments for a secular change in the overall largely similar geochemical datasets, why is such significant vari-
maficity of bulk oceanic crust composition are robust (section 1.1), ation in compositional change over time be predicted? One
it is still uncertain whether upper-oceanic crust basalts in the prominent issue may relate to how ‘non-arc’ or ‘MOR-derived
Archean were commensurately more MgO-rich than modern-day basalt’ is defined in each case. If all mafic rocks in greenstone belts
MORB. Several workers have examined this issue and present are used for statistical analysis of secular compositional change, a
conflicting conclusions about the magnitude of change over time. major decrease in global average MgO content with time is
While Keller and Schoene (2012), Furnes et al. (2014), and Ganne observed, with a significant step occurring at w2.5 Ga (Keller and
1016 R.M. Palin, B. Dyck / Geoscience Frontiers 9 (2018) 1009e1019

Schoene, 2012), coinciding with several major transition points in today (Fig. 1). While a defining characteristic of Archean greenstone
the Earth’s geological evolution (e.g. the rise of atmospheric oxy- belts is the presence of extremely MgO rich (>25 wt.%) komatiitic
gen; Kump, 2008). While geochemical discrimination of (meta) basalts (Condie, 1981), these lavas are typically interpreted to have
basalts that formed in convergent versus divergent plate margin formed via intraplate plume-related activity at elevated mantle TP
settings should ideally be applied in order to restrict these changes (Arndt et al., 1997), and thus cannot represent these hypothesized
to MOR-derived units only (e.g. Furnes et al., 2014), there remains complementary residues to low-MgO tholeiites. Neither ultramafic
debate regarding how effectively trace element ratios used to cumulate-like lithologies nor their metamorphosed equivalents
discriminate modern-day mantle source regions can be applied in (garnet pyroxenite: Fig. 3) have been documented en-mass in Pro-
the geological past (Li et al., 2015). For example, assumptions that terozoic or Archean terranes (Bickle et al., 1994; Bédard, 2017), and
Archean basalts with high Nb/Th and Zr/Nb ratios may have had geophysical data do not suggest their presence at cratonic bases
MOR-type affinity (e.g. Condie et al., 2016) stem from observations (Abbott et al., 2013). Petrophysical arguments have been made to
made from Phanerozoic data (Pearce, 2008), and a tectonically suggest that such high-density residuum delaminated into rela-
equivalent plate margin environment may not have been charac- tively lower-density underlying aesthenospheric mantle (Vlaar
terized by the same geochemical signatures on the early Earth. et al., 1994; Van Thienen et al., 2004; Bédard, 2006; Fischer and
Many geodynamic models suggest that subduction would have Gerya, 2016), explaining their apparent absence from the geolog-
been relatively inefficient on a hotter Earth, if it operated at all, and ical record.
that the early Archean may instead have been characterized by an
active stagnant lid regime (O’Neill et al., 2007; Van Hunen and 3.2. Implications for uniformitarianism and the onset of modern-
Moyen, 2012; Sizova et al., 2014). Recently, Condie (2015) showed day plate tectonics
that Archean basalts generally differ from Proterozoic and Phan-
erozoic basalts by having very few enriched- and depleted-mantle The principle of uniformitarianism is succinctly summarized by
source signatures, as defined by Zr/Nb and Nb/Th ratios. Most data the phrase “the present is the key to the past”. Uniformitarianistic
obtained from samples with MOR-type geological characteristics arguments have been used by many studies over the past few
were tightly clustered around primitive mantle ratios, with some decades to investigate exactly when modern-day plate tectonics e
overlap into the hydrated mantle field. This pattern may indicate a defined by one-sided subduction e initiated on Earth, with
relatively unfractionated stagnant-lid mantle source, which sug- propositions spanning almost the entirety of Earth history be-
gests that mantle convection accompanying this regime would tween the Hadean (>4.2 Ga; Hopkins et al., 2008) and the late
have inefficiently recycled various enriched and depleted compo- Neo-Proterozoic (0.85 Ga; Hamilton, 2011). A wide variety of
nents in the deep mantle. If true, low-MgO basaltic rocks of Archean geological data and/or theoretical modelling results have been
age may have initially formed by fractional crystallization processes utilized to support the conclusions of each of these studies. Pre-
rather than via closed-system partial melting (e.g., Foley et al., dicted and observed changes in oceanic crust/basalt compositions
2002, 2003). However, the petrological characteristics of basalts (Table 1) may be used to shed light on first-order petrological
generated in a stagnant lid tectonic setting are very poorly under- arguments based on lithologies in the geological record that form
stood, and applying arbitrary compositional ranges to geochemical in or around subduction zone environments (Stern, 2005; Palin
databases as a predefined criterion to exclude particular lithologies and White, 2016).
(e.g. Gill, 1979; Condie, 1985; Condie et al., 2016) may inadvertently Rock types reported to be diagnostic of metamorphism along
cause a false impression of compositional quiescence through time low-temperature/high-pressure geotherms that characterize sub-
e at least in terms of overall maficity. New analytical data obtained duction zones today include, but are not limited to, glaucophane-
from studies of other planetary bodies in our solar system may and lawsonite-bearing mafic rocks (blueschists), and ultrahigh-
provide insight into this problem (Downs, 2015), as Mars, for pressure (UHP) lithologies (coesite- and diamond-bearing eclo-
example, may have been characterized by such a stagnant lid gites). The rarity of these lithologies from the rock record prior to
regime when most of its mafic crust formed. Recent investigation ca. 0.7 Ga was interpreted by Stern (2005) to argue for a Neo-
into the effects of metamorphism on such stagnant-lid versus Proterozoic onset of modern-day subduction-driven plate tec-
plate-tectonic basalts (Wade et al., 2017) indicate notable petro- tonics. However, the phase equilibria presented here for rocks of
logical and petrophysical differences between the two for equiva- variable MgO and Al2O3 contents (Fig. 3) indicate that bulk
lent metamorphic conditions. This supports our interpretations composition exerts a strong control on the mineral assemblage that
that metamorphism of primitive crust in the Archean would have should form in mafic crust descending into either warm (Fig. 4aed)
produced dramatically different rock types than observed today, or cool (Fig. 4eeh) Archean subduction zones.
even in intraplate tectonic settings. Lawsonite-bearing assemblages are shown in our calculations to
What conclusions can thus be drawn about the petrological be stable in almost all mafic bulk compositions at low-grade con-
nature of Archean oceanic crust from studying basalts in the ditions (Fig. 4e), but will only remain during continued meta-
geological record? While some greenstone belts preserve rare ex- morphism in relatively low-MgO lithologies (Fig. 4f), with more
amples of deformed and metamorphosed remnants of oceanic mafic protoliths forming garnet and clinopyroxene instead. The
lithosphere (e.g. sheeted dykes: Helmstaedt et al., 1986; Kusky and relative availability of H2O and Na2O will control whether glauco-
Kidd, 1992; Furnes et al., 2007; Grosch and Slama, 2017), no com- phane or talc additionally occurs. As such, lawsonite should not be
plete sections of juvenile crust are known (Bickle et al., 1994). expected to appear in subducted and metamorphosed MgO-rich
Evolved tholeiitic basalts with low MgO contents appear to repre- basalt that likely typified the uppermost levels of Archean
sent major volumetric constituents of both Archean and Protero- oceanic crust (Ganne and Feng, 2017), rendering its petrological
zoic greenstone belts (Condie, 1981), although these basalt significance for diagnosing the operation of subduction throughout
compositions are not in equilibrium with mantle peridotite (O’Hara Earth history dubious. Furthermore, even lawsonite that is known
and Herzberg, 2002) and so do not themselves represent primary to have formed at blueschist- and eclogite-facies conditions in
magmas. Primary mantle melts from a hotter mantle would have Phanerozoic subduction zones commonly does not survive the
been more MgO-rich than today, and the lowermost portions of the exhumation process (Clarke et al., 2006; Whitney and Davis, 2006),
Archean crust must have comprised large volumes of comple- and thus its absence in Precambrian orogens may be strongly
mentary cumulates significantly more mafic than those observed influenced by preservation bias. We further note that lawsonite’s
 R.M. Palin, B. Dyck / Geoscience Frontiers 9 (2018) 1009e1019 1017

breakdown products typically include chlorite, zoisite, paragonite, analyzed phase assemblages that would stabilize during meta-
and/or quartz (Ballevre et al., 2003; Groppo and Castelli, 2010; St- morphism of variably mafic basalts, and show these diagnostic rock
Onge et al., 2013), and so its retrograded products may superfi- types may not form in high-MgO protoliths potentially represen-
cially resemble assemblages observed in greenstones that domi- tative of the upper levels of Archean oceanic crust (Fig. 4). This
nate Archean terranes (cf. Palin and White, 2016). suggests that the modern-day metamorphic facies system, which is
Similarly, the absence of coesite- or diamond-bearing UHP mafic predicated on the mineral assemblage changes that occur in low-
rocks cannot necessarily be used as an a-priori indicator of the non- MgO mafic protoliths, must be applied to rocks in Archean ter-
operation of subduction, as many eclogite-facies rocks subducted ranes with great care. It should therefore be remembered that ‘the
during Proterozoic and Phanerozoic orogenesis record maximum absence of evidence is not evidence of absence’. Multi-disciplinary
pressures below the quartzecoesite transition (w26e28 kbar; Palin studies that are able to link multiple independent lines of evi-
et al., 2014; Möller et al., 2015; Weller et al., 2015, 2016; Maldonado dence into holistic tectonic models, such as recently demonstrated
et al., 2016). However, the lower silica content of highly magnesian by Wade et al. (2017), are required to make significant progress in
basalts (Table 1; Arndt et al., 1997) may prohibit the formation of this field in the future.
quartz/coesite within the metamorphic assemblage, even if UHP
conditions are reached (cf. Klein et al., 2017), highlighting a further Acknowledgements
limitation of applying the modern-day facies definitions to Archean
rocks. Given that both lawsonite-bearing blueschists and UHP The authors wish to thank several researchers who have taken
eclogites are absent from many convergent margins today, they the time to discuss the implications of secular change on meta-
should be viewed as sufficientdbut not necessarydindicators of morphic processes and products, including Owen Weller, Kent
subduction-related processes on the early Earth, and their absence Condie, Robert Stern, Dick White, Karen Ziaja, Steve Foley, and
may instead be accounted for by secular change in MOR-derived Stephan Buhre. We further thank Laura Morrissey and an anony-
basalt composition, or else a dramatic change in exhumation pro- mous reviewer for constructive comments that helped to signifi-
cesses at the ProterozoicePhanerozoic boundary (Korenaga, 2016). cantly improve this manuscript, and Chris Spencer for his efficient
Therefore, modern-day facies classification systems that are editorial handling.
fundamentally built around mineral assemblages that stabilize in
low-MgO metabasalt compositions cannot be confidently applied
to ancient metamorphic terranes without considering this bulk- References
rock composition effect.
Abbott, D., Burgess, L., Longhi, J., Smith, W.H.F., 1994. An empirical thermal history
of the Earth’s upper mantle. Journal of Geophysical Research 99, 13835e13850.
4. Conclusions Abbott, D.H., Hoffman, S.E., 1984. Archaean plate tectonics revisited 1. Heat flow,
spreading rate, and the age of subducting oceanic lithosphere and their effects
on the origin and evolution of continents. Tectonics 3, 429e448.
Thermal models of the Earth’s internal heat budget predict that Abbott, D.H., Mooney, W.D., VanTongeren, J.A., 2013. The character of the Moho and
the Archean mantle was notably hotter than the present day lower crust within Archean cratons and the tectonic implications. Tectono-
(Herzberg et al., 2010), which would have resulted in higher de- physics 609, 690e705.
Arndt, N.T., 1983. Role of a thin, komatiite-rich oceanic crust in the Archean plate-
grees of partial melting during mantle decompression, and the tectonic process. Geology 11, 372e375.
formation of a thicker and more magnesian oceanic crust (Van Arndt, N.T., Kerr, A.C., Tarney, J., 1997. Dynamic melting in plume heads: the for-
Thienen et al., 2004). The absence of pristine examples of obduc- mation of Gorgona komatiites and basalts. Earth and Planetary Science Letters
146, 289e301.
ted ophiolites in the geological record precludes direct examination Artemieva, I.M., 2006. Global 1  1 thermal model TC1 for the continental lith-
of the geochemistry of primary magmas or their MOR-derived osphere: implications for lithosphere secular evolution. Tectonophysics 416,
basalts, which would validate such theoretical predictions; how- 245e277.
Ballevre, M., Pitra, P., Bohn, M., 2003. Lawsonite growth in the epidote blueschists
ever, numerous recent studies have attempted to use statistical
from the Ile de Groix (Armorican Massif, France): a potential geobarometer.
methods to examine the degree of secular compositional change Journal of Metamorphic Geology 21, 723e735.
from global datasets of metamafic rocks. These works corroborate Bédard, J.H., 2006. A catalytic delamination-driven model for coupled genesis of
Archaean crust and sub-continental lithospheric mantle. Geochimica et Cos-
the theoretical predictions, and suggest that early Archean basalts
mochimica Acta 70, 1188e1214.
may have contained nearly w15 wt.% MgO, compared to just over Bédard, J.H., 2017. Stagnant lids and mantle overturns: implications for Archaean
w7 wt.% for modern-day examples (White and Klein, 2014). tectonics, magmagenesis, crustal growth, mantle evolution, and the start of
However, wide compositional variation occurs in modern-day ba- plate tectonics. Geoscience Frontiers 9, 19e49.
Bickle, M.J., 1978. Heat loss from the earth: a constraint on Archean plate tectonics
salts according to variation in tectonic setting and mantle potential from the relation between geothermal gradients and the rate of plate produc-
temperature of formation (e.g. Iceland: Winchester and Floyd, tion. Earth and Planetary Science Letters 40, 301e315.
1977). If such variation also existed in the geologic past, mean Bickle, M.J., Nisbet, E.G., Martin, A., 1994. Archean greenstone belts are not oceanic
crust. The Journal of Geology 102, 121e138.
compositions of basalts of various ages would reflect biases in Billen, M.I., 2008. Modeling the dynamics of subducting slabs. Annual Review of
sampling from various tectonic settings. Overall compositional Earth and Planetary Science 36, 325e356.
changes in basalts with time would thus be obscured. Classifying Bleeker, W., 2002. Archaean tectonics: a review, with illustrations from the Slave
craton. Geological Society of London: Special Publications 199, 151e181.
the geodynamic environments of formation of Archaean and Pro- Bleeker, W., 2003. The late Archean record: a puzzle in ca. 35 pieces. Lithos 71,
terozoic mafic igneous rocks is therefore critical, although it is 99e134.
uncertain how well trace element signatures diagnostic of partic- Bradley, D.C., 2008. Passive margins through Earth history. Earth-science Reviews
91, 1e26.
ular geodynamic environments of formation on the modern-day
Bradley, D.C., 2011. Secular trends in the geologic record and the supercontinent
Earth may translate into the geological past. cycle. Earth-science Reviews 108, 16e33.
What are the implications of a secular change in oceanic crust/ Breuer, D., Spohn, T., 1993. Cooling of the Earth, Urey ratios, and the problem of
potassium in the core. Geophysical Research Letters 20, 1655e1658.
MORB composition? A longstanding discussion exists in the geo-
Brown, M., 2007. Metamorphic conditions in orogenic belts: a record of secular
science community as to why modern-day petrological indicators change. International Geology Review 49, 193e234.
of subduction, such as lawsonite-blueschists and UHP meta- Campbell, I.H., Griffiths, R.W., 1990. Implications of mantle plume structure for the
morphic rocks, suddenly appear in the geological record at ca. 0.7 evolution of flood basalts. Earth and Planetary Science Letters 99, 79e93.
Clarke, G.L., Powell, R., Fitzherbert, J.A., 2006. The lawsonite paradox: a comparison
Ga (e.g. Stern, 2005; Brown, 2007) if subduction on a global scale of field evidence and mineral equilibria modelling: Journal of Metamorphic
has operated since at least the Meso-Archean (w3.0 Ga). We have Geology 24, 715e725.
1018 R.M. Palin, B. Dyck / Geoscience Frontiers 9 (2018) 1009e1019

Condie, K.C., 1981. Archean Greenstone Belts. Elsevier, Amsterdam, p. 434. Herzberg, C.T., Condie, K.C., Korenaga, J., 2010. Thermal history of the Earth and its
Condie, K.C., 1985. Secular variation in the composition of basalts: an index to petrological expression. Earth and Planetary Science Letters 292, 79e88.
mantle evolution. Journal of Petrology 26, 545e563. Holland, T.J.B., Powell, R., 2011. An improved and extended internally consistent
Condie, K.C., 2015. Changing tectonic settings through time: indiscriminate use of thermodynamic dataset for phases of petrological interest, involving a new
geochemical discriminant diagrams. Precambrian Research 266, 587e591. equation of state for solids. Journal of Metamorphic Geology 29, 333e383.
Condie, K.C., Kröner, A., 2008. When did plate tectonics begin? Evidence from the Hooykaas, R., 1963. The Principle of Uniformity in Geology, Biology, and Theology.
geologic record. Geological Society of America Special Papers 440, 281e294. E.J. Brill, London.
Condie, K.C., Aster, R.C., 2009. Zircon age episodicity and growth of continental Hopkins, M., Harrison, T.M., Manning, C.E., 2008. Low heat flow inferred from > 4
crust. Eos Transactions AGU 90, 364e364. Gyr zircons suggests Hadean plate boundary interactions. Nature 456,
Condie, K.C., Aster, R.C., van Hunen, J., 2016. A great thermal divergence in the 493e496.
mantle beginning 2.5 Ga: geochemical constraints from greenstone basalts and Hynes, A.J., Francis, D.M., 1982. A transect of the early proterozoic cape smith fold
komatiites. Geoscience Frontiers 7, 543e553. belt, new quebec. Tectonophysics 88, 23e59.
Connolly, J.A.D., 2009. The geodynamic equation of state: what and how. Jahn, B.M., Glikson, A.Y., Peucat, J.J., Hickman, A.H., 1981. REE geochemistry and
Geochemistry, Geophysics, Geosystems 10, Q10014. https://doi.org/10.1029/ isotopic data of Archean silicic volcanics and granitoids from the Pilbara Block,
2009GC002540. Western Australia: implications for the early crustal evolution. Geochimica et
Davies, G.F., 1992. On the emergence of plate tectonics. Geology 20, 963e966. Cosmochimica Acta 45, 1633e1652.
Debaille, V., O’Neill, C., Brandon, A.D., Haenecour, P., Yin, Q.Z., Mattielli, N., Johnson, T.E., Brown, M., Kaus, B.J., VanTongeren, J.A., 2014. Delamination and
Treiman, A.H., 2013. Stagnant-lid tectonics in early Earth revealed by 142Nd recycling of Archaean crust caused by gravitational instabilities. Nature Geo-
variations in late Archean rocks. Earth and Planetary Science Letters 373, science 7, 47e52.
83e92. Kato, Y., Nakamura, K., 2003. Origin and global tectonic significance of early archean
Dewey, J.F., Windley, B.F., 1981. Growth and differentiation of continental crust. cherts from the marble bar greenstone belt, pilbara craton, western Australia.
Philosophical Transactions of the Royal Society of London A: Mathematical, Precambrian Research 125, 191e243.
Physical and Engineering Sciences 301, 189e206. Kattge, J., Diaz, S., Wirth, C., 2014. Of carrots and sticks. Nature Geoscience 7,
Downs, R.T., 2015. Determining mineralogy on Mars with the CheMin x-ray 778e779.
diffractometer. Elements 11, 45e50. Keller, C.B., Schoene, B., 2012. Statistical geochemistry reveals disruption in secular
Dyck, B., Reno, B.L., Kokfelt, T.F., 2015. The majorqaq belt: a record of neoarchaean lithospheric evolution about 2.5 Gyr ago. Nature 485, 490e493.
orogenesis during final assembly of the north atlantic craton, southern west Kirby, S.H., Durham, W.B., Stern, L.A., 1991. Mantle phase changes and deep-
Greenland. Lithos 220, 253e271. earthquake faulting in subducting lithosphere. Science 252, 216e225.
Ernst, W.G., 2017. Earth’s thermal evolution, mantle convection, and Hadean onset King, S.D., 2001. Subduction zones: observations and geodynamic models. Physics
of plate tectonics. Journal of Asian Earth Sciences 145B, 334e348. of the Earth and Planetary Interiors 127, 9e24.
Fischer, R., Gerya, T., 2016. Early Earth plume-lid tectonics: a high-resolution 3D Klein, B.Z., Jagoutz, O., Behn, M.D., 2017. Archean crustal compositions promote full
numerical modelling approach. Journal of Geodynamics 100, 198e214. mantle convection. Earth and Planetary Science Letters 474, 516e526.
Foley, S.F., Tiepolo, M., Vannucci, R., 2002. Growth of early continental crust Korenaga, J., 2006. Archean geodynamics and the thermal evolution of Earth. In:
controlled by melting of amphibolite in subduction zones. Nature 417, 837e840. Benn, K., et al. (Eds.), Archean Geodynamics and Environments. American
Foley, S.F., Buhre, S., Jacob, D.E., 2003. Evolution of the Archean crust by delami- Geophysical Union, Washington, D.C., pp. 7e32
nation and shallow subduction. Nature 421, 249e252. Korenaga, J., 2008. Urey ratio and the structure and evolution of Earth’s mantle.
Furnes, H., de Wit, M., Staudigel, H., Rosing, M., Muehlenbachs, K., 2007. A vestige of Reviews of Geophysics 46, 1e32.
Earth’s oldest ophiolite. Science 315, 1704e1707. Korenaga, J., 2013. Initiation and evolution of plate tectonics on Earth: theories and
Furnes, H., Dilek, Y., De Wit, M., 2014. Precambrian greenstone sequences represent observations. Annual Review of Earth Sciences 41, 117e151.
different ophiolite types. Gondwana Research 27, 649e685. Korenaga, J., 2016. Plate tectonics: metamorphic myth. Nature Geoscience 9, 9e11.
Ganne, J., De Andrade, V., Weinberg, R.F., Vidal, O., Dubacq, B., Kagambega, N., Korenaga, J., 2017. On the extent of mantle hydration caused by plate bending. Earth
Naba, S., Baratoux, L., Jessell, M., Allibon, J., 2011. Modern-style plate subduction and Planetary Science Letters 457, 1e9.
preserved in the Palaeoproterozoic West African craton. Nature Geoscience 5, Komiya, T., Hayashi, M., Maruyama, S., Yurimoto, H., 2002. Intermediate-P/T type
60e65. Archean metamorphism of the Isua supracrustal belt: implications for secular
Ganne, J., Feng, X., 2017. Primary magmas and mantle temperatures through time. change of geothermal gradients at subduction zones and for Archean plate
Geochemistry, Geophysics. Geosystems 18, 872e888. tectonics. American Journal of Science 302, 806e826.
Gerya, T.V., Stöckhert, B., Perchuk, A.L., 2002. Exhumation of high-pressure meta- Komiya, T., Maruyama, S., Hirata, T., Yurimoto, H., Nohda, T., 2004. Geochemistry of
morphic rocks in a subduction channel: a numerical simulation. Tectonics 21, the oldest MORB and OIB in the Isua supracrustal belt (3.8 Ga), southern West
1e19. Greenland: implications for the composition and temperature of early Archean
Gerya, T.V., Yuen, D.A., Maresch, W.V., 2004. Thermomechanical modelling of slab upper mantle. Island Arc 13, 47e72.
detachment. Earth and Planetary Science Letters 226, 101e116. Kretz, P., 1983. Symbols for rock-forming minerals. American Mineralogist 68,
Gill, R.C.O., 1979. Comparative petrogenesis of Archean and modern low-K tholei- 277e279.
ites. A critical review of some geochemical aspects. Physics and Chemistry of Kump, L.R., 2008. The rise of atmospheric oxygen. Nature 451, 277e278.
the Earth 11, 431e447. Kusky, T.M., Kidd, W.S.F., 1992. Remnants of an Archean oceanic plateau, Belingwe
Grambling, J.A., 1981. Pressures and temperatures in Precambrian metamorphic greenstone belt, Zimbabwe. Geology 20, 43e46.
rocks. Earth and Planetary Science Letters 53, 63e68. Li, C., Arndt, N.T., Tang, Q., Ripley, E.M., 2015. Trace element indiscrimination dia-
Green, D.H., Lockwood, J.P., Kiss, E., 1968. Eclogite and almandine-jadeite-quartz grams. Lithos 232, 76e83.
rock from guajira peninsula, Colombia, south America. American Mineralogist Liou, J.G., Hacker, B.R., Zhang, R.Y., 2000. Into the forbidden zone. Science 287,
53, 1320. 1215e1216.
Green, E.C.R., White, R.W., Diener, J.F.A., Powell, R., Holland, T.J.B., Palin, R.M., 2016. Maldonado, R., Ortega-Gutiérrez, F., Hernández-Uribe, D., 2016. Garnete
Activityecomposition relations for the calculation of partial melting equilibria chloritoideparagonite metapelite from the Chuacús Complex (Central
in metabasic rocks. Journal of Metamorphic Geology 34, 845e869. Guatemala): new evidence for continental subduction in the North
Groppo, C., Castelli, D., 2010. Prograde PeT evolution of a lawsonite eclogite from AmericaeCaribbean plate boundary. European Journal of Mineralogy 28,
the Monviso meta-ophiolite (Western Alps): dehydration and redox reactions 1169e1186.
during subduction of oceanic FeTi-oxide gabbro. Journal of Petrology 51, Martin, H., Moyen, J.F., 2002. Secular changes in tonaliteetrondhjemitee
2489e2514. granodiorite composition as markers of the progressive cooling of Earth. Ge-
Grosch, E., Slama, J., 2017. Evidence for 3.3-billion-year-old oceanic crust in the ology 4, 319e322.
Barberton greenstone belt, South Africa. Geology 45, 695e698. Maruyama, S., Liou, J.G., Terabayashi, M., 1996. Blueschists and eclogites of the
Guevara, V.E., Caddick, M.J., 2016. Shooting at a moving target: phase equilibria world and their exhumation. International Geology Review 38, 485e594.
modelling of high-temperature metamorphism. Journal of Metamorphic Geol- Maruyama, S., Santosh, M., Azuma, S., 2017. Initiation of plate tectonics in the Ha-
ogy 34, 209e235. dean: eclogitization triggered by the ABEL Bombardment. Geoscience Frontiers
Guiraud, M., Powell, R., Rebay, G., 2001. H2O in metamorphism and unexpected 9, 1033e1048.
behaviour in the preservation of metamorphic mineral assemblages. Journal of McKenzie, D., Bickle, M.J., 1988. The volume and composition of melt generated by
Metamorphic Geology 19, 445e454. extension of the lithosphere. Journal of Petrology 29, 625e679.
Hamilton, W.B., 1998. Archean magmatism and deformation were not products of Mints, M.V., Belousova, E.A., Konilov, A.N., Natapov, L.M., Shchipansky, A.A.,
plate tectonics. Precambrian Research 91, 143e179. Griffin, W.L., O’Reilly, S.Y., Dokukina, K.A., Kaulina, T.V., 2010. Mesoarchean
Hamilton, W.B., 2003. An alternative earth. Geological Society of America Today 13, subduction processes: 2.87 Ga eclogites from the Kola Peninsula, Russia. Ge-
4e12. ology 38, 739e742.
Hamilton, W.B., 2011. Plate tectonics began in Neoproterozoic time, and plumes Möller, C., Andersson, J., Dyck, B., Lundin, I.A., 2015. Exhumation of an eclogite terrane
from deep mantle have never operated. Lithos 123, 1e20. as a hot migmatitic nappe, Sveconorwegian orogen. Lithos 226, 147e168.
Hargraves, R.B., 1986. Faster spreading or greater ridge length in the Archean? Moyen, J.F., Martin, H., 2012. Forty years of TTG research. Lithos 148, 312e336.
Geology 14, 750e752. Moyen, J.F., Stevens, G., Kisters, A., 2006. Record of mid-Archaean subduction from
Helmstaedt, H., Padgham, W.A., Brophy, J.A., 1986. Multiple dikes in lower kam metamorphism in the Barberton terrain, South Africa. Nature 442, 559e562.
group, yellowknife greenstone belt: evidence for archean sea-floor spreading? Nair, R., Chacko, T., 2008. Role of oceanic plateaus in the initiation of subduction and
Geology 14, 562e566. origin of continental crust. Geology 36, 583e586.
 R.M. Palin, B. Dyck / Geoscience Frontiers 9 (2018) 1009e1019 1019

Nicoli, G., Dyck, B., 2018. Exploring the metamorphic consequences of secular Sizova, E., Gerya, T., Brown, M., 2014. Contrasting styles of Phanerozoic and Pre-
change in the siliciclastic compositions of continental margins. Geoscience cambrian continental collision. Gondwana Research 25, 522e545.
Frontiers 9, 967e975. Sizova, E., Gerya, T., Stüwe, K., Brown, M., 2015. Generation of felsic crust in the
Nielsen, M.E., Fisk, M.R., 2010. Surface area measurements of marine basalts: im- Archean: a geodynamic modeling perspective. Precambrian Research 271,
plications for the sub-seafloor microbial biomass. Geophysical Research Letters 198e224.
37, L15604,. https://doi.org/10.1029/2010GL044074. Smithies, R.H., 2000. The Archaean tonaliteetrondhjemiteegranodiorite (TTG) se-
Nisbet, E.G., Fowler, C.M.R., 1983. Model for Archean plate tectonics. Geology 11, ries is not an analogue of Cenozoic adakite. Earth and Planetary Science Letters
376e379. 182, 115e125.
Nutman, A.P., Friend, C.R.L., 2007. A vestige of Earth’s oldest ophiolite: Comment. Solomon, S.C., Sleep, N.H., Richardson, R.M., 1975. On the forces driving plate tec-
Science 318, 747. tonics: inferences from absolute plate velocities and intraplate stress.
Nutman, A.P., Bennett, V.C., Friend, C.R., 2015. Proposal for a continent ‘Itsaqia’ Geophysical Journal International 42, 769e801.
amalgamated at 3.66 Ga and rifted apart from 3.53 Ga: initiation of a Wilson Spear, F.S., 1995. Metamorphic Phase Equilibria and Pressureetemperatureetime
Cycle near the start of the rock record. American Journal of Science 315, Paths. Mineralogical Society of America, Washington, p. 799.
509e536. St-Onge, M.R., Rayner, N., Palin, R.M., Searle, M.P., Waters, D.J., 2013. Integrated
O’Hara, M.J., Herzberg, C., 2002. Interpretation of trace element and isotope features pressureetemperatureetime constraints for the Tso Morari dome (Northwest
of basalts: relevance of field relations, petrology, major element data, phase India): implications for the burial and exhumation path of UHP units in the
equilibria, and magma chamber modeling in basalt petrogenesis. Geochimica et western Himalaya. Journal of Metamorphic Geology 31, 469e504. https://
Cosmochimica Acta 66, 2167e2191. doi.org/10.1111/jmg.12030.
O’Neill, C., Lenardic, A., Moresi, L., Torsvik, T., Lee, C.A., 2007. Episodic precambrian Stern, R.J., 2005. Evidence from ophiolites, blueschists, and ultrahigh-pressure
subduction. Earth and Planetary Science Letters 262, 552e562. metamorphic terranes that the modern episode of subduction tectonics
Padhi, C., Korenaga, J., Ozima, M., 2012. Thermal evolution of Earth with xenon began in Neoproterozoic time. Geology 33, 557e560.
degassing: a self-consistent approach. Earth and Planetary Science Letters 1e9, Stüwe, K., 1997. Effective bulk composition changes due to cooling: a model pre-
341e344. dicting complexities in retrograde reaction textures. Contributions to Miner-
Palin, R.M., St-Onge, M.R., Waters, D.J., Searle, M.P., Dyck, B., 2014. Phase equilibria alogy and Petrology 129, 43e52.
modelling of retrograde amphibole and clinozoisite in mafic eclogite from the Syracuse, E.M., van Keken, P.E., Abers, G.A., 2010. The global range of subduction
Tso Morari massif, northwest India: constraining the PeTeM(H2O) conditions of zone thermal models. Physics of the Earth and Planetary Interiors 183, 73e90.
exhumation. Journal of Metamorphic Geology 32, 675e693. https://doi.org/ Thomson, W., 1862. On the secular cooling of the earth. Transactions of the Royal
10.1111/jmg.12085. Society of Edinburgh 23, 157e169.
Palin, R.M., White, R.W., 2016. Emergence of blueschists on Earth linked to secular Thurston, P.C., 2002. Authochthonous development of Superior Province greenstone
changes in oceanic crust composition. Nature Geoscience 9, 60e64. https:// belts? Precambrian Research 115, 11e36.
doi.org/10.1038/ngeo2605. Turcotte, D.L., 1980. On the thermal evolution of the Earth. Earth and Planetary
Palin, R.M., Weller, O.M., Waters, D.J., Dyck, B., 2016a. Quantifying geological un- Science Letters 48, 53e58.
certainty in metamorphic phase equilibria modelling; a Monte Carlo assess- Turner, S., Rushmer, T., Reagan, M., Moyen, J.F., 2014. Heading down early on? Start
ment and implications for tectonic interpretations. Geoscience Frontiers 7, of subduction on Earth. Geology 42, 139e142.
591e607. https://doi.org/10.1016/j.gsf.2015.08.005. Van Hunen, J., Moyen, J.F., 2012. Archean subduction: fact or fiction? Annual Review
Palin, R.M., White, R.W., Green, E.C.R., Diener, J.F.A., Powell, R., Holland, T.J.B., 2016b. of Earth and Planetary Sciences 40, 195e219.
High-grade metamorphism and partial melting of basic and intermediate rocks. Van Keken, P.E., Kiefer, B., Peacock, S.M., 2002. High-resolution models of subduc-
Journal of Metamorphic Geology 34, 871e892. https://doi.org/10.1111/ tion zones: implications for mineral dehydration reactions and the transport of
jmg.12212. water into the deep mantle. Geochemistry, Geophysics. Geosystems 3, 1056.
Palin, R.M., White, R.W., Green, E.C., 2016c. Partial melting of metabasic rocks and Van Kranendonk, M.J., 2010. Two types of Archean continental crust: plume and
the generation of tonaliticetrondhjemiticegranodioritic (TTG) crust in the plate tectonics on early earth. American Journal of Science 310, 1187e1209.
Archaean: constraints from phase equilibrium modelling. Precambrian Van Kranendonk, M.J., Kirkland, C.L., 2013. Orogenic climax of earth: the 1.2e1.1 Ga
Research 287, 73e90. grenvillian superevent. Geology 7, 735e738.
Palin, R.M., Reuber, G.S., White, R.W., Kaus, B.J.P., Weller, O.M., 2017. Subduction Van Thienen, P., van den Berg, A.P., Vlaar, N.J., 2004. Production and recycling of
metamorphism in the Himalayan ultrahigh-pressure Tso Morari massif: an oceanic crust in the early Earth. Tectonophysics 368, 41e65.
integrated geodynamic and petrological modelling approach. Earth and Plan- Vervoort, J.D., Blichert-Toft, J., 1999. Evolution of the depleted mantle: Hf isotope
etary Science Letters 467, 108e119. https://doi.org/10.1016/j.epsl.2017.03.029. evidence from juvenile rocks through time. Geochimica et Cosmochimica Acta
Patterson, C., 1956. Age of meteorites and the earth. Geochimica et Cosmochimica 63, 533e556.
Acta 10, 230e237. Vielzeuf, D., Schmidt, M.W., 2001. Melting relations in hydrous systems revisited:
Pearce, J.A., 2008. Geochemical fingerprinting of oceanic basalts with implications application to metapelites, metagreywackes, and metabasalts. Contributions to
for the classification of ophiolites and search for Archean oceanic crust. Lithos Mineralogy and Petrology 141, 251e267.
100, 14e48. Vlaar, N.J., Van Keken, P.E., Van den Berg, A.P., 1994. Cooling of the Earth in the
Peacock, S.M., 1996. Thermal and petrologic structure of subduction zones. Sub- Archaean: consequences of pressure-release melting in a hotter mantle. Earth
duction top to bottom. Geophysical Monograph 96, 119e133. and Planetary Science Letters 121, 1e18.
Peacock, S.M., Wang, K., 1999. Seismic consequences of warm versus cool subduc- Wade, J., Dyck, B., Palin, R.M., Moore, J.D., Smye, A.J., 2017. The divergent fates of
tion metamorphism: examples from southwest and northeast Japan. Science primitive hydrospheric water on Earth and Mars. Nature 552, 391e394.
286, 937e939. Weller, O.M., Wallis, S.R., Aoya, M., Nagaya, T., 2015. Phase equilibria modelling of
Peck, W.H., Valley, J.W., Wilde, S.A., Graham, C.M., 2001. Oxygen isotope ratios and blueschist and eclogite from the Sanbagawa metamorphic belt of southwest
rare earth elements in 3.3 to 4.4 Ga zircons: ion microprobe evidence for high Japan reveals along-strike consistency in tectonothermal architecture. Journal
d18O continental crust and oceans in the Early Archean. Geochimica et Cos- of Metamorphic Geology 33, 579e596.
mochimica Acta 65, 4215e4229. Weller, O.M., St-Onge, M.R., Rayner, N., Waters, D.J., Searle, M.P., Palin, R.M., 2016.
Polat, A., Frei, R., Appel, P.W.U., Dilek, Y., Fryer, B., Ordóñez-Calderón, J.C., Yang, Z., UePb zircon geochronology and phase equilibria modelling of a mafic eclogite
2008. The origin and compositions of Mesoarchean oceanic crust: evidence from the Sumdo complex of south-east Tibet: insights into prograde zircon
from the 3075 Ma Ivisaartoq greenstone belt, SW Greenland. Lithos 100, growth and the assembly of the Tibetan plateau. Lithos 262, 729e741.
293e321. Weller, O.M., St-Onge, M.R., 2017. Record of modern-style plate tectonics in the
Powell, R., Holland, T.J.B., Worley, B., 1998. Calculating phase diagrams involving Palaeoproterozoic Trans-Hudson orogen. Nature Geoscience 10, 305e311.
solid solutions via non-linear equations, with examples using THERMOCALC. White, R.W., Palin, R.M., Green, E.C., 2017. High-grade metamorphism and partial
Journal of Metamorphic Geology 16, 577e588. melting in Archean composite grey gneiss complexes. Journal of Metamorphic
Rebay, G., Powell, R., Diener, J.F.A., 2010. Calculated phase equilibria for a MORB Geology 35, 181e195.
composition in a PeT range, 450e650  C and 18e28 kbar: the stability of White, W.M., Klein, E.M., 2014. 4.13dcomposition of the oceanic crust. Treatise on
eclogite. Journal of Metamorphic Geology 28, 635e645. Geochemistry (second edition) 4, 457e496.
Sandiford, M., 1989. Secular trends in the thermal evolution of metamorphic ter- Whitney, D.L., Davis, P.B., 2006. Why is lawsonite eclogite so rare? Metamorphism
rains. Earth and Planetary Science Letters 95, 85e96. and preservation of lawsonite eclogite, Sivrihisar, Turkey. Geology 34, 473e476.
Schmidt, M.W., Poli, S., 1998. Experimentally based water budgets for dehydrating Winchester, A., Floyd, P.A., 1977. Geochemical discrimination of different magma
slabs and consequences for arc magma generation. Earth and Planetary Science series and their differentiation products using immobile elements. Chemical
Letters 163, 361e379. Geology 20, 325e343.
See, T.J.J., 1907. On the temperature, secular cooling and contraction of the earth, Zhai, M., Santosh, M., 2013. Metallogeny of the North China Craton: link with
and on the theory of earthquakes held by the ancients. Proceedings of the secular changes in the evolving Earth. Gondwana Research 24, 275e297.
American Philosophical Society 46, 191e299. Ziaja, K., Foley, S.F., White, R.W., Buhre, S., 2014. Metamorphism and melting of
Shirey, S.B., Richardson, S.H., 2011. Start of the Wilson cycle at 3 Ga shown by di- picritic crust in the early Earth. Lithos 189, 173e184.
amonds from subcontinental mantle. Science 333, 434e436.