The lower continental crust, representing up to 50% of the continental mass, is largely inaccessible, making its composition difficult to constrain. Previous composite models based on geophysical evidence and geochemical data of granulite terrains and xenoliths have proposed varying results, from a mafic, relatively refractory lower crust to an intermediate-felsic, more enriched composition. Here, we investigated the mineralogy and geochemistry of predominantly mafic granulite xenoliths from eastern Australia and the Kola Peninsula, Russia, using an in situ analytical approach that minimises host magma contamination. The resulting xenolith compositions are variably and often strongly depleted in most highly incompatible trace elements, including the heat-producing elements. These xenoliths represent an extremely refractory component of the lower continental crust, likely formed after high degrees of partial melting or crystallisation from a depleted source. A lower crust composed solely of this refractory endmember would be too exhausted in heat-producing elements to satisfy heat-flow constraints. However, a volumetrically significant component of the lower crust is this mafic and refractory material, combined with undifferentiated material and a felsic or metapelitic portion. Using geophysical constraints on proportions of refractory (55%), undepleted (38%) and enriched (7%) components, a new estimate for average lower continental crust that satisfies heat flow limits was calculated, including for elements such as Be, B, Cs, W and Tl, where previous estimates relied on very few data. Finally, we show that because much of the lower continental crust is so refractory and depleted in incompatible elements, it is unlikely to be a reservoir that can balance radiogenic isotope (unradiogenic Pb) and trace element ratios (e.g. Rb/Cs, Nb/Ta) for which bulk silicate Earth departs from chondritic ratios.

During active tectonic processes the lower continental crust plays a major role in the development and evolution of the overlying crustal column. The lower crust is heavily influenced by influx of mafic magma from the mantle and extraction of more felsic magma to the shallower crust. Evidence of such interactions are preserved in residues after melting that are rarely brought to the surface as suites of granulite-facies, lower-crustal xenoliths. Here we investigate the petrology, geochemistry and petrochronology of a set of granulite xenoliths from the Pleistocene Hill 32 volcanic cone that has intruded through the geologically complex Georgetown Inlier, north Queensland, which has experienced multiple magmatic-tectonic events since the Paleoproterozoic. Petrochronology constrains the temperature in the Pleistocene lower crust to above the closure temperature of U-Pb in apatite (350-550?) and rutile (-400-500?), which define an eruption age for Hill 32 at 1.6 +/- 0.1 Ma, and below the closure temperature of titanite (up to 800?), which gives an age of 219 +/- 4 Ma. The samples contain petrographic and petrochronological evidence of recrys-tallisation and melt extraction at > 900?, conditions able to reset U-Pb systematics in zircon. The zircons give U-Pb ages consistent with the timing of the Permo-Triassic New England orogeny, indicating zircon recrystallisation/regrowth coeval with major felsic magmatism in the upper crust at that time. U-Pb in zircon records minor preservation of earlier events. While, epsilon(Hf) indicates mixing between Proterozoic to Archean precursor crust with juvenile mantle-derived magma, which provided both heat and radiogenic Hf. The samples are interpreted to be the product of melt-precursor rock hybridisation followed by melt extraction in the lower crust that represents a critical and rarely observed component of crust formation processes. (c) , 2022 International Association for Gondwana Research. Published by Elsevier B.V. All rights reserved.

Abstract Recent developments in laser-ablation Lu–Hf dating have opened a new opportunity to rapidly obtain apatite ages that are potentially more robust to isotopic resetting compared to traditional U–Pb dating. However, the robustness of the apatite Lu–Hf system has not been systematically examined. To address this knowledge gap, we conducted four case studies to determine the resistivity of the apatite Lu–Hf system compared to the zircon and apatite U–Pb system. In all cases, the apatite U–Pb system records a secondary (metamorphic or metasomatic) overprint. The apatite Lu–Hf system, however, preserves primary crystallization ages in unfoliated granitoids at temperatures of at least c. 660°C. Above c. 730°C, the Lu–Hf system records isotopic resetting by volume diffusion. Hence, in our observations for apatites of ‘typical’ grain sizes in granitoids ( c. 0.01–0.03 mm 2 ), the closure temperature of the Lu–Hf system is between c. 660 and c. 730°C, consistent with theoretical calculations. In foliated granites, the Lu–Hf system records the timing of recrystallization, while the apatite U–Pb system tends to record younger cooling ages. We also present apatite Lu–Hf dates for lower crustal xenoliths erupted with young alkali basalts, demonstrating that the Lu–Hf system can retain a memory of primary ages when exposed to magmatic temperatures for a relatively short duration. Hence, the apatite Lu–Hf system is a new insightful addition to traditional zircon (or monazite) U–Pb dating, particularly when zircons/monazites are absent or difficult to interpret due to inheritance or when U and Pb isotopes display open system behaviour. The laser-ablation-based Lu–Hf method allows campaign-style studies to be conducted at a similar rate to U–Pb studies, opening new opportunities for magmatic and metamorphic studies.

Compared to the well-studied upper continental crust, the composition of the lower crust is much more poorly constrained. Geophysical constraints and geochemical data from granulite xenoliths indicate that the lower crust is, on average, mafic and depleted in most incompatible elements, including the heat-producing elements (HPE). However, the extent of this depletion is not well known. The large uncertainties associated with lower crustal estimates have important implications for the Earth’s evolution, as the lower crust is often proposed to be a “hidden reservoir” (e.g., for unradiogenic Pb) needed to close mass balance discrepancies for the Bulk Silicate Earth. In this study, we analysed granulite xenoliths from Queensland, eastern Australia, and the Kola Peninsula, northwest Russia, using a reconstitution approach that corrects for host magma contamination. This method also provides detailed insight into which minerals control elemental distribution and concentrations of the xenoliths. The major element compositions of both suites of granulite xenoliths highlight their mafic nature, with SiO2 contents similar to previously published estimates. However, the concentrations of the most incompatible elements, including the large ion lithophile elements (LILE) and HPE, are very low. Some elements are more depleted by an order of magnitude than the most popular composites used in the literature. Zircon and monazite are rare in these mafic granulites, while apatite and rutile have relatively low Th and U concentrations. The absence of hydrous silicates (e.g., mica and amphibole) and the relatively high anorthite contents of feldspar in the xenoliths is a controlling factor in the low LILE concentrations, particularly for Rb and Cs. If this composition is representative of typical lower continental crust, then such highly refractory compositions limit the ability of the lower crust to act as a significant contributor for planetary mass balance considerations because it does not contain enough Pb, Nb, Ta, Cs and Rb to balance other inventories of the differentiated bulk silicate Earth.

Much of the continental lithosphere developed during the Archean, which was an Eon of change in terms of global geodynamics and geochemical cycles. Uncovering the causal links between crust forming processes and prevailing geodynamic mechanisms is crucial for understanding the origins and composition of the present-day continental lithosphere. Pristine Archean crust is scarce yet can be found in cratons worldwide. Many of these occurrences comprise rocks of the tonalite-trondhjemite-granodiorite (TTG) suite, which represent a prevalent component of the Archean continental crust. TTGs are generally considered to have formed by partial melting of amphibolite or eclogite source rocks that had basaltic precursors originally extracted from a depleted mantle (e.g., [1]). The age of the source rocks (i.e., the time between the basalt extraction from the mantle and TTG formation) can be determined from the initial radiogenic isotope compositions of TTGs, provided that the P/D ratio of the source can be reliably estimated and is significantly different from that of the depleted mantle. Based on this principle, we estimated the age of basaltic sources of TTGs from cratons of different age and paleogeography from initial 87Sr/86Sr compositions determined by in-situ Sr isotope analysis of primary igneous apatite (LA-MC-ICPMS). The 87Sr/86Sr of these apatites show that prior to 3.4 Ga TTGs were derived from relatively old mafic sources and that the average time between formation of basaltic material from the mantle and subsequent remelting under amphibolite to eclogite facies conditions decreased drastically during the Paleoarchean. This secular change indicates a rapid global increase in the efficiency of TTG production or the emergence of a new TTG-forming process at c. 3.4 Ga [2]. In this contribution we explore this hypothesis by comparing the 87Sr/86Sr signature of the TTGs with their trace-element compositions, as well as with 176Hf/177Hf zircon data for these rocks and contemporary TTGs from other studies. This combined geochronological, isotope and geochemical analyses will provide new constraints on the age of TTG sources during the Archean and will allow investigation into the nature and probable causes of the apparent rejuvenation at 3.4 Ga, as indicated by Sr isotopes. [1] Hoffmann, J.E. et al. (2011) Geochim. Cosmochim. Acta 75, 4157-4178. [2] Caton, S., et al., (in review) Chem. Geol.

Radiogenic isotopes provide an important means towards elucidating Archean crustal evolution. The global Hf and Nd isotope record of Archean crustal fragments has been instrumental to unveiling the history of ancient crustal growth and differentiation. The Rb-Sr system could provide valuable complementary constraints in this regard, as this system is particularly sensitive to magmatic fractionation processes, and the chemical and isotopic evolution of magma sources. Application of this system has so far been complicated, however, by its suscepti-bility to isotope re-equilibration or alteration of the Rb/Sr parent-daughter ratio. In-situ Sr isotope analysis of primary igneous minerals with very low Rb/Sr, such as apatite, provides a new means to determine the initial 87Sr/86Sr (87Sr/86Sri) values for igneous rocks directly. In this study, we apply in-situ Sr isotope analysis of apatite by LA-MC-ICPMS to tonalite-trondhjemite-granodiorite (TTG) rocks and end-member sanukitoids from Archean cratons worldwide. The 87Sr/86Sri values of sanukitoids are relatively radiogenic, supporting the model in which such rocks are formed by flux melting of a mantle strongly enriched by metasomatism, possibly by slab -derived fluids. The 87Sr/86Sri values for TTGs formed between 3.72 and 3.45 Ga are generally radiogenic, indicating aged amphibolite sources. The 87Sr/86Sri values of younger TTGs are systematically lower and were derived from mafic sources that had an average age of <= 0.2 Gyr. This evolution matches with observations from Hf isotopes for TTGs of similar age and indicates a systematic change in the nature or efficiency of TTG crust formation during the Paleoarchean. In-situ Sr isotope analysis of apatite provides a useful method to uncover the Sr record of the early continental crust, and enables constraints on local source evolution and the general two-step evolutionary process of Archean crust formation.

This study reports results from thermodynamic phase equilibrium and trace element modelling of mafic magmatic underplating and solid-liquid interaction in the lower continental crust (LCC) in intraplate settings. The arrival of underplating basalt sills into thin (∼30 km at 8 kbar) and thicker (∼45 km at 12 kbar) andesitic and basaltic LCC precursors was simulated with heating and batch-melting to yield refractory residues. Continued magmatic invasion of the LCC was then modelled at the same pressures with hybridisation between the residual solids and basaltic and picritic liquids (at 80:20 and 50:50 proportions). The first finding is that hybridisation with basalt increases the stability fields of 2-pyroxene-plagioclase and garnet-clinopyroxene-plagioclase granulites, by far the dominant LCC xenoliths found in anorogenic settings. The second finding pertains to situations when the liquid proportion of the hybrid system is lower than the fraction of the incoming mafic liquid. The resulting liquids share many characteristics with silicic volcanic rocks in continental flood basalt (CFB) provinces, including low Al and high Ti-concentrations. Stoichiometric melting reactions from 80:20 residual andesite:basalt hybridisation reproduce trace element patterns of silicic eruptives with potential for decoupling of Sr-Nd-Pb isotope systematics, as found in CFB rhyolites and LCC xenoliths. Incongruent melting of hornblende and garnet (the A in AFC) from the residual solid and dominant peritectic fractional crystallisation of plagioclase (the FC in AFC) from the liquid occur. A notable feature of LCC-picrite hybridisation is that solid residues after felsic liquid extraction, when subjected to higher pressure, are buffered to olivine and quartz-free compositions. Consequently, if these high-pressure granulite residues delaminated, they would convert to the bi-mineralic high-Mg eclogites found as xenoliths in kimberlites. The modelled residues also have the Sr-enrichment and positive Eu-anomalies of eclogite xenoliths. Solid-liquid interaction in intraplate settings has been and remains a significant process in crustal growth and evolution.

The composition of the lower continental crust is estimated via the analysis of granulite xenoliths, granulite terrains and geophysical properties. All three proxies generally agree on the lower crust's refractory nature, dominated by mafic granulites. Estimates weighted using seismic velocity reference models yield lower Th and U concentrations and higher K/U ratios than granulite xenolith averages, while terrain granulites are often much less refractory than xenoliths. Here we present new data for lower crustal xenoliths from central Queensland, an understudied part of the xenolith-bearing eastern Australian basalt provinces. The granulite chemistry was estimated using the reconstitution approach in which the modal mineralogy and in situ chemical analyses are combined. High-resolution energy-dispersive X-ray spectroscopy mapping revealed that K is enriched in anastomosing grain boundary networks and fractures in granulite and co-occurring mantle peridotite xenoliths. Laser-ablation inductively-coupled-plasma mass-spectrometry 2-D traverses show that the same networks are also significantly enriched in many highly incompatible elements. There are sharp concentration contrasts with neighbouring phases for elements with very different diffusivities (Li and U), suggesting that the networks formed during entrainment, decompression and heating within the host basalt. We propose that undetected inclusion of such late enrichment skews xenolith chemistry estimates to non-representative, overly fertile compositions. In the case of the studied xenoliths, the carrier basalts are not very strongly enriched in highly incompatible elements, and even when the K-rich networks are included in the reconstitution, the resulting granulite chemistry is very refractory with 0.52 wt% K2O, 0.07 ppm Th and 0.03 ppm U. The locally dominant lithology of lower crustal xenoliths is simple two-pyroxene, plagioclase, ilmenite granulite with few accessory phases. The granulite mineralogy and chemistry were compared with results from thermodynamic models of prograde anatexis of different metabasites, variably hydrated. The comparison shows that the granulites could be restitic calc-alkaline basalts or diorites that experienced episodic melt extraction accumulating up to 50-60% total melt loss at very high temperatures (950-1050 degrees C), implying that the temperatures recorded by two-pyroxene thermometry (750-830 degrees C) do not capture the thermal maximum. The corresponding upper crustal section of the northern New England Orogen exposes a range of Devonian to Cretaceous granitoids, some of which have complementary features to the granulites, including the low modal abundance of plagioclase; low relative abundance of Ti, very high Rb/Ba ratios, and high Th/U ratios. Together, the data suggest extensive and protracted melting of the original lower crust upon lithospheric thinning and concomitant magmatic underplating. The required high temperatures favour picritic over basaltic underplates. In such a setting, the gravity-driven delamination of more mafic garnet-rich restites and olivine-rich mafic-ultramafic underplating material is physically plausible. This delamination could help explain the long-established mass balance issue for the formation of continental crust in general.

Accurately constraining the composition of the continental crust is essential for understanding Earth’s geological evolution. However, due to the difficulty of accessing the deep crust, the composition of the lower crust is poorly constrained. This includes the heat-producing elements (HPE), Th, U and K, which are vital for investigating Earth’s thermal history. Chemical estimates of the lower crust are based on the geochemical analysis of tectonically-exhumed terrains and rare granulite xenoliths brought to the surface by explosive volcanism [1]. The databases used to create such estimates rely heavily on conventional bulk rock analysis of granulite xenoliths. However, many xenoliths are too small to be analysed with this method, creating a potential bias towards larger samples, and the effect of host magma contamination and secondary alteration could go undetected when the whole rock is homogenised to a powder. In this study, an integrated microanalytical approach is used to investigate the potential effects of secondary processes and reconstruct the primary compositions of granulite xenoliths from central Queensland, Australia. Mineral phase maps were obtained by SEM-EDS using a TESCAN Integrated Mineral Analyzer (TIMA) field emission SEM to quantify the xenoliths’ modal mineralogies (Figure 1). Elemental maps show that K is enriched in fractures, grain boundaries and secondary glass phases, and relatively low in the primary rock-forming minerals. Each xenolith sample’s modal mineralogy was combined with in situ quantitative chemical analysis of each mineral phase by electron probe micro-analysis (EPMA) for major elements and laser-ablation inductively-coupled-plasma mass spectrometry (LA-ICP-MS) for trace elements. These analyses confirm that the primary minerals in the xenoliths are depleted in the HPE

The continental crust grew and matured compositionally during the Paleo- to Neoarchean through the addition of juvenile tonalite-trondhjemite-granodiorite (TTG) crust. This change has been linked to the start of global plate tectonics, following the general interpretation that TTGs represent ancient analogs of arc magmas. However, partial melting of deep-residing mafic reservoirs, e.g., during density-driven crustal overturning, delamination, or plume activity, would likewise produce TTGs without requiring active-margin processes. Boron analysis could allow discriminating between these models and, so, provide new insight into the formation and petrogenetic setting of Archean juvenile crust. We analyzed B concentrations and isotope compositions in 3.8-2.8 Ga TTGs from different Archean terranes. Low δ 11 B values (down to -16‰)

Current estimates for the composition of the lower continental crust show significant variation for the concentrations of the highly incompatible elements, including large uncertainties for the heat-producing elements. This has consequences for models of the formation of lower crust. For example, is lower continental crust inherently poor in incompatible elements or has it become so after extraction of partial melts caused by thermal incubation? Answering these questions will require better agreement between estimates for the chemistry of the lower crust. One issue is that granulite samples may have been altered during ascent. Xenoliths often experience contamination from the entraining alkaline magma, potentially resulting in elevated concentrations of incompatible trace elements when analysed by conventional bulk rock techniques. To avoid this, we assessed an in situ approach for reconstructing whole rock compositions with granulites from the Kapuskasing Structural Zone, Superior Province, Canada. As terrain samples, they have not been affected by host magma contamination, and as subrecent glacial exposures, they show minimal modern weathering. We used scanning electron microscope electron dispersive spectroscopy (SEM-EDS) phase mapping to establish the modal mineralogy. Major and trace element concentrations of mineral phases were determined by electron microprobe and laser ablation inductively coupled plasma mass spectrometry (LA-ICPMS), respectively. These concentrations were combined with the modal mineralogies to obtain reconstructed whole rock compositions, which were compared to conventional bulk rock analyses. The reconstructed data show good reproducibility relative to the conventional analyses for samples with massive textures. However, the conventional bulk rock chemistry systematically yields higher K concentrations, which are hosted in altered feldspars. Thus, even in terrain samples, minor alteration can lead to elevated incompatible element estimates that may not represent genuine lower continental crust.

The continental crust grew and matured compositionally during the Palaeo- to Neoarchean through the addition of juvenile tonalite-trondhjemite-granodiorite (TTG) crust. This change has been linked ...

Apatite is a common accessory mineral in igneous, sedimentary and metamorphic rocks. It has potential as a provenance indicator in sedimentary systems, as it can host a wide variety of trace elements in its crystal structure and can yield thermochronological age information. However, the processes controlling the trace element and U-Pb systematics of metamorphic apatite remain poorly understood, and metamorphic apatite remains significantly under-represented in compositional provenance databases linking apatite trace-element chemistry to its corresponding parent rock type. We investigate the trace-element and U-Pb systematics of metamorphic apatite from a suite of 22 bedrock samples of diverse metamorphic grade and protolith type, sampled from a variety of metamorphic terranes. Metamorphic apatite from low- to medium-grade metapelites and metabasites can be easily distinguished from granitic apatite as it is significantly depleted in Th, REE, and Y. Depletion in Th and REE + Y is attributed to growth of co-genetic epidote, which is the dominant carrier phase of the REE + Y, Th, and U in all the low- to medium-grade samples mapped by laser ablation quadrupole inductively coupled plasma mass spectrometry (LA-Q-ICP-MS) and energy dispersive X-ray spectroscopy (EDS). Apatite U contents in low- to medium-grade metapelites and metabasites are more variable than the Th and REE + Y contents, but are typically low. Consequently, grains from these rock types are often undateable by the U-Pb LA-ICP-MS method and thus are underrepresented in apatite U-Pb detrital datasets, but can still be identified as low- to medium-grade metamorphic apatite by their trace-element characteristics. Low-grade metapelite apatite is difficult to distinguish from low-grade metabasite apatite, which is likely due to the growth of U-, Th-, and REE-rich epidote in both lithologies. Detrital (granitic) apatite is stable in very low-grade (e.g. pumpellyite-actinolite facies) metasedimentary samples, while neo- or re-crystallized metamorphic apatite is widespread by the upper-greenschist facies. LA-Q-ICP-MS imaging demonstrates that low REE + Y, Th, and U metamorphic apatite rims can nucleate on detrital igneous apatite precursors with high REE + Y, Th, and U. With increasing metamorphic grade, 1) relict detrital apatite is consumed, 2) the coherence of the U-Pb concordia systematics dating metamorphism improves, and 3) the degree of dispersion on metamorphic apatite multi-element plots decreases. The highest-grade metamorphic apatite samples investigated are paragneisses, some of which have locally undergone anatexis. Apatites from these samples yield well constrained TW concordia intercept ages and minor dispersion on multi-element plots (which is attributed to the absence of epidote) and closely resemble apatite from S-type granites in their trace element characteristics.

Current models for the properties of Hadean-Eoarchean crust encompass a full range of possibilities, involving crust that is anywhere from thick and differentiated to thin and mafic. New data are needed to test and refine these models, and, ultimately, to determine how continents were first formed. The Rb-Sr system provides a potentially powerful proxy for crustal evolution and composition. However, this system has thus far been underutilized in studies on early crustal evolution due to its susceptibility to re-equilibration. Overcoming this issue requires new analytical approaches to micro-sample ancient Sr-rich mineral relics that may retain primary Rb-Sr systematics, allowing for the precise and accurate determination of initial Sr-87/Sr-86 values. In this study, we used a novel application of laser-ablation multi-collector inductively coupled plasma mass spectrometry to determine the Sr isotope composition of apatite inclusions in >3.6 Ga zircon grains from Eoarchean granodiorite gneisses of the Acasta Gneiss Complex, Slave Province, Canada. The Rb-87-corrected Sr-87/Sr-86 values of these inclusions are largely identical and are distinctly different from values obtained from altered matrix apatite. The inclusion data provide the first direct estimate of initial Sr-87/Sr-86 for these ancient rocks. Combining this result with information on the protolith and source-extraction age yields estimates for the range of Rb/Sr values, and by extension composition, that the source of these rocks may have had. The data indicate that continental crust containing over 60 wt% of SiO2 was present in the ca. 4.2 Ga source of the Acasta Gneiss Complex. Thus vestiges of evolved crust must have existed within the primitive proto-continents that were present on the Hadean Earth. (C) 2018 Elsevier Ltd. All rights reserved.

(1) Department of Earth, Ocean and Atmospheric Sciences, University of British Columbia, Vancouver, Canada, (2) Department of Geosciences, Swedish Museum of Natural History, Stockholm, Sweden, (3) Institut für Mineralogie, Westfälische Wilhelms-Universität, Münster, Germany, (4) Institut für Geologie und Mineralogie, Universität zu Köln, Cologne, Germany, (5) Geological Survey of Canada, Ottawa, Ontario, Canada, (6) Institut für Geologie, Universität Bern, Bern, Switzerland.