Figure 6. Major element geochemical characteristics of the Isua
and Pilbara ultramafic samples in comparison with those of Phanerozoic
cumulates, arc peridotites, >3.2 Ga ultramafic rocks (seeFig. 3 for data sources), and modelled liquid lines of descent.
All data are presented using anhydrous values (i.e., all major element
abundances are normalized to zero LOI and 100 wt.% total). The data in
this figure show that Isua and Pilbara ultramafic rocks, Mg-rich
cumulates and mantle peridotites have similar major element geochemical
systematics. Data sources for the cumulates and MELTS modelling curves
are from Chin et al. (2018), Mallik et al. (2020), and references
therein. Specifically, cumulates from oceanic island settings (panel b)
cannot be modelled due to limitations of MELTs programs on modelling
ultrahigh pressure (>3 GPa) melting and enriched mantle
sources, which are necessary for generating oceanic island basalts.
Mineral geochemistry
Olivine grains in Isua sample AW17724-2C (lens B) have extraordinarily
high Mg# values of ~95– 98 and NiO of
~0.39–0.63 wt.%. In contrast, olivine grains in Isua
sample AW17725-4 (lens A) have Mg# values of ~87 and
NiO of ~0.52– 0.61 wt% (Table S2 ).
Ti-humite phases in sample AW17724-2C have variable TiO2abundances of ~3.0– 8.1 wt.%. All analyzed
spinel grains in the Isua samples contain a high magnetite component
(i.e., FeOt of ~90 wt.%) (Table S2 ).
Spinel of both chromite or magnetite compositions occur in the Pilbara
samples. Specifically, chromite spinel grains have
Cr2O3 of
~40– 50 wt.%, TiO2 of
0.6– 4.3 wt.%, and MgO of 5– 12 wt.%. The Cr#
[Cr/(Cr+Al)] values and Mg# values of chromite spinel grains are
~65– 75 and ~17– 46,
respectively (Fig. 9 ; Table S2 ).
DiscussionWe analyzed phaneritic ultramafic rocks in the Eoarchean Isua
supracrustal belt and the East Pilbara Terrane to explore their
petrogenesis as a means of testing the viability of existing tectonic
models. Specifically, we explore whether these rocks need to be
explained as mantle peridotites that were emplaced in the crust in a
subduction setting. Our new petrological and geochemical data from six
ultramafic samples from the Isua supracrustal belt and three
ultramafic samples from the East Pilbara Terrane show that (1) Isua
and Pilbara samples have been variably altered and now contain several
alteration minerals (e.g., serpentine, talc, carbonate) that replaced
igneous ferromagnesian silicates (Figs. 2–3 ) ; (2) Pilbara
ultramafic samples preserve poikilitic textures and polygonal textures
(Fig. 3 ); one Isua sample (AW17725-4 from lens A) also
preserves relict polygonal textures (Fig. 2b ); (3) trace
element abundances in both Isua and Pilbara ultramafic samples range
from depleted with respect to the primitive mantle values (0.1 times
primitive mantle values) to enriched (10 times primitive mantle
values) (Fig. 6a–b ); (4) two out of three Pilbara ultramafic
samples show fractionated, relatively high concentrations of Os and Ir
versus Pt, Pd, and Re in the primitive mantle-normalized diagram
(Fig. 7c ), which are similar to those of Isua meta-peridotite
lens samples (Waterton et al., 2022); and (5) chromite spinel in
Pilbara ultramafic samples feature Cr# of
~65– 75, and Mg# of
~17– 46 (Fig. 9 ). In the following
sections, we first discuss the potential impacts of alterations on
petrology and geochemistry. Then, we show that new and compiled
petrology, geochemistry, and microstructures of Isua and Pilbara
ultramafic rocks are consistent with a cumulate origin, whereas an
origin as thrust-emplaced mantle slices is not required. We then
discuss the implications for testing early Earth tectonic models and
the initiation of plate tectonics.