5.3. Do Isua ultramafic rocks require formation in mantle?
In this section, we compare our findings of Isua and Pilbara ultramafic rocks with those of similarly altered compiled and modelled cumulates and mantle peridotites to establish whether any feature of Isua ultramafic rocks needs to be explained uniquely via plate tectonic-related mantle slices. First, many geochemical parameters commonly used to study modern mantle rocks (e.g., Mg/Si-Al/Si plots) cannot be used to differentiate Eoarchean olivine-rich cumulates from depleted mantle residues (see also Waterton et al., 2022). This is because olivine-rich cumulates, which can be modelled via <10% fractional crystallization of basaltic melts (Fig. 6 ; Mallik et al. 2020), are similar to variably depleted and altered mantle rocks in terms of many whole-rock major element systematics (e.g., MgO, SiO2, Al2O3, FeOt; Figs. 5, 6 ). Although the Isua and Pilbara ultramafic rocks, and other Eoarchean ultramafic cumulates generally have systematically lower CaO contents in comparison to fertile mantle peridotites due to absence or very low clinopyroxene abundance (Fig. 5 ), these characteristics can be also related to alteration effects (see section 5.1). In fact, small clinopyroxene inclusions occur in olivine grains of some Isua ultramafic rocks from lens A. This has been explained to indicate olivine growths coupled with clinopyroxene dissolution during reactions between mantle peridotites and ascending melts (Nutman et al., 2021a). However, clinopyroxene undersaturation and olivine saturation is possible across a range of pressure-temperature-composition combinations (Chen and Zhang, 2009 and references therein) and could happen under crustal conditions during magma crystallization in the presence of water, crustal assimilation and/or magma recharge (e.g., Kelemen, 1990; Gordeychik et al., 2018).
Similarly, primitive mantle-normalized trace element patterns cannot distinguish mantle rocks from ultramafic cumulates. Isua and Pilbara ultramafic rocks show flat or mildly fractionated trace element abundances 0.1 to 10 times of primitive mantle values with ~1.1 to ~3.8 (La/Sm)PMand ~0.3 to ~1.7 (Gd/Yb)PM (Fig. 7 ). In contrast, depleted mantle rocks that did not undergone fluid and melt metasomatism have highly depleted trace element patterns (e.g., LREE lower than 0.01 times PM) and systematically low (Gd/Yb)PM (generally <0.3; Fig. 7A ). The level of trace element enrichment in Isua and Pilbara ultramafic rocks, especially as shown by Th-Gd/Yb systematics (Fig. 7B ), can only be explained by cumulates or depleted mantle experienced melt-rock interactions. Previous studies (Friend and Nutman, 2011; Van de Löcht et al. 2020; Waterton et al. 2022) show that Isua ultramafic rocks could have interacted with basaltic melts, although the melts cannot be represented by the Isua high Al2O3/TiO2 basalts (boninitic) due to their largely radiogenic Os isotopic signatures in comparison to the Isua ultramafic rocks (Waterton et al. 2022). As noted above, melt-rock interaction is common in both crustal magma chambers and upper mantle; therefore, the existence of such processes neither discriminate petrogenetic origins nor tectonic models (cf. Friend and Nutman, 2011; Van de Löcht et al. 2020).
The Nb depletion relative to Th and La, which exists in both Isua ultramafic rocks and Isua basalts (Fig. 7 ; Polat and Hofmann, 2003), has been extensively cited as indicating arc-like signatures for these rocks. In this Eoarchean subduction interpretation, Nb (and Ta) depletion would indicate fractionation effects of fluids and rutile associated with eclogitized slabs (e.g., Münker, 1998; Keppler, 1996). However, this signature is not unique to volcanic arcs, particular with respect to the early Earth. For example, basalts, felsic volcanics, TTGs, and ultramafic rocks (Fig. 7C) of the East Pilbara Terrane also have strong Nb depletion (e.g., Martin et al. 2005; Smithies et al. 2007), but this terrane is widely thought to be plume-generated in a non-plate tectonic setting (e.g., Van Kranendonk et al. 2007). Furthermore, rutile as well as fluids form readily via metamorphic dehydration in the lower parts of the thickened lithosphere of non-plate tectonic settings via metamorphism (Johnson et al., 2017), particularly in a heat-pipe lithosphere featuring cold geotherm (a la Moore and Webb, 2013). Recycling of such lower crust materials (which is possible in plate and non-plate tectonic settings via mechanisms like delamination, sagduction and/or downwards advection) and subsequent fluid fluxing and melting could generate igneous rocks with Nb-Ta depletion. Alternatively, Nb depletion may be a secondary signature formed via fluid metasomatism under amphibolite facies conditions (Guice et al. 2018). Vigorous fluid activities and material exchanges between mantle and crust in non-plate tectonic settings can also explain the mantle-like oxygen isotopes found in some Isua olivines (Nutman et al. 2021a). Indeed, mantle-like oxygen isotopes are observed in zircons from some TTGs (originally lower crust partial melts) of the East Pilbara Terrane (Smithies et al., 2021). This finding implies a fluid-rich early mantle, buffered by fluxing from the recycled crust, that was capable of introducing mantle-like oxygen isotope signatures to early crust and magmas (Smithies et al., 2021).
In terms of HSE patterns, we agree with Waterton et al. (2022) and Szilas et al. (2015) that HSE signatures of Isua ultramafic are consistent with cumulate origins. This argument is further strengthened by the HSE patterns of our cumulate-textured Pilbara ultramafic rocks, which are similar to those of Isua lenses A and B samples in terms of positive Ru anomalies relative to Ir and Pt and relative depletion of Pt and Pd versus Ir (Fig. 8 ; Waterton et al. 2022). Similar patterns can be found in other Eoarchean cumulates dominated by olivine and chromite (e.g., Coggon et al., 2015; McIntyre et al., 2019).
Some spinel crystals with ~100 Cr# and ~0 Mg# values in the new and compiled Isua ultramafic rocks reflect metamorphic modifications of primary chromite into magnetite (Fig. 3b ; Barnes and Roeder, 2001). However, igneous petrogenesis can be interpreted from primary chromite grains of both Isua and Pilbara ultramafic samples. New and compiled spinel data of these rocks match the Fe Ti trend in the Mg# Cr# space (Fig. 9b ). Such a trend can be produced by equilibration of spinel phases during fractional crystallization (Barnes and Roeder, 2001), and thus can be found in cumulates (Fig. 9b ). Chromite crystals of Isua and Pilbara samples also have variable TiO2 (up to ~2 and ~5 wt.%, respectively) In contrast, due to equilibration with olivine, mantle spinel typically has high Mg# and varied Cr# (i.e., the Cr Al trend in Fig. 9b , Barnes and Roeder, 2001) as well as low TiO2 (typically <1 wt.%;Fig. 9a ) (e.g., Tamura and Arai, 2006). Although fluid/melt assisted alterations could impact spinel geochemistry in mantle rocks, expected changes include Cr# reduction and Mg# increase along with the Cr Al trend (El Dien et al., 2019), which are not consistent with the observed spinel geochemistry. Therefore, we conclude that some chromite spinel crystals from Isua (Szilas et al., 2015) and Pilbara ultramafic rocks (new data; Fig. 9 ) are not similar to spinel hosted in mantle rocks, but rather indicate cumulate origins (cf. Nutman et al., 2021a).
Although the B-type olivine fabrics (Kaczmarek et al., 2016) have been interpreted to reflect mantle environments, they are also consistent with cumulate origins. Waterton et al. (2022) pointed out that B-type fabrics in Isua dunites may be formed via magmatic or metamorphic processes (e.g., Chin et al., 2020; Holtzman et al., 2003; Nagaya et al., 2014; Yao et al., 2019) rather than via deformation in mantle wedge (cf. Kaczmarek et al. 2016). We found additional evidence that may support this interpretation: olivine in Isua lens B samples are considered to be dehydration products of antigorite-breakdown (e.g., Guotana et al. 2022). Alignment of olivine shape long-axes in lens B (see Fig. 1D of Nutman et al. 2021a) generally parallel to the regional lineation directions (mostly trending southeast; Zuo et al. 2021). If olivine long-axes correspond to their [001] crystal directions as suggested by Kaczmerak et al. (2016), then olivine [001] is generally parallel with lineation directions, which are also antigorite (010) directions in deformed serpentinites (e.g., Nagaya et al. 2017). Such crystal axis relationships are consistent with a metamorphic origin of olivine B-type fabrics, in which topotactic growth of olivine occurred with olivine [001] axes parallel to antigorite (010) (Nagaya et al. 2014). Therefore, with current rock and mineral textural data from Isua ultramafic rocks, mantle wedge conditions are not required, and cumulate origins are viable.
Finally, the presence of Ti-humite in Isua ultramafic rocks have been interpreted to reflect low-temperature, UHP (i.e., <500 °C, >2.6 GPa) metamorphism (Friend and Nutman, 2011; Nutman et al., 2020; Guotana et al., 2022) primarily using the petrogenetic grid generated from experiments (i.e., Shen et al. 2015). However, our complementary work (Mueller et al., pre-print) shows that the results of Shen et al. (2015) cannot be directly applied to Isua ultramafic rocks. This is because Shen et al. (2015) experimented on a CO2-free chemical system, but Isua ultramafic rocks preserve carbonate phases (Fig. 2a ) that appear to be a reaction product of an olivine-breakdown reaction, equally producing antigorite and Ti-humite (Mueller et al., pre-print). Conversely, Mueller et al. (pre-print) show that Ti-humite could have been formed under much lower pressures, such as the amphibolite facies conditions recorded by the other parts of the belt (Ramírez-Salazar et al. 2021). Therefore, the Isua supracrustal belt may not have experienced (U)HP metamorphism, obviating the need for plate tectonic subduction (Waterton et al., 2022; cf. Friend and Nutman, 2011; Nutman et al., 2020; Guotana et al., 2022).
In summary, although several features of Isua or Pilbara ultramafic samples are commonly associated with plate tectonic processes (e.g., the B-type olivine fabrics, mantle-like oxygen isotopes, Nb depletion, and Ti-humite preserved in Isua ultramafic samples), these features are not inconsistent with rock formation and metamorphism under crustal conditions. In addition, the cumulate textures of Pilbara ultramafic samples and the spinel geochemical characteristics of both Isua and Pilbara ultramafic samples are inconsistent with tectonically-emplaced depleted mantle, but instead are compatible with cumulate origins (Figs. 4–9 ). As such, both Isua and Pilbara ultramafic samples can be interpreted as crustal cumulates that experienced alterations under crustal conditions. Because crustal cumulates are produced by fractional crystallization of melts, these rocks are consistent with both plate tectonics and hot stagnant-lid tectonics. Thus, plate tectonics is not required to explain the petrogenesis of Isua and Pilbara ultramafic rocks (cf. Nutman et al., 2020, 2021a).