Archean Cratons Time Capsules of the Early Earth

1811-5209/24/0020-071$2.50  DOI: 10.2138/gselements.20.3.162

Keywords: Archean; Hadean; craton; crustal evolution

Archean Cratons: Time Capsules of the Early Earth

The first 2 billion years of Earth history were an unparalleled period of dynamic change for our planet. As the Earth formed by accretion of planetesimals, energy from impacts and the radioactive decay of short-lived and longlived radioisotopes strongly heated the early Earth. One result was segregation of the core, a process that released gravitational energy in the form of heat when metal sank to the center of the Earth. These processes combined to heat the Earth sufficiently that magma oceans formed at both regional and planetary scales. Collision with a Mars-sized impactor (Theia) led to formation of the Moon as debris from the collision coalesced in near-Earth orbit. All this took place within about the first 100 million years of our Solar System’s history (Nimmo and Kleine 2015). But rapid change did not stop there as the rate of impacts decreased and early magma oceans solidified. By 4.4 billion years ago (Ga), grains of the mineral zircon (ZrSiO4) crystallized in magmas that solidified to become part of an early silicate crust (Wilde et al. 2001; Valley et al. 2014). The oceans may have been present by this time (Wilde et al. 2001), certainly by 3.8–3.7 Ga when the first pillow basalts appear in the geologic record (Nutman et al. 2013), and the fossilized remains of early microbial life forms are also present in the rock record by 3.7–3.5 Ga (Hickman-Lewis et al. 2023). This early history of our planet has been pieced together over many years from studies of Archean (4.0–2.5 Ga) rocks that preserve the oldest record of early Earth evolution and from meteorites, which provide the strongest evidence for the age and overall chemical composition of the Earth. We now know that by the end of the Hadean (4.57–4.03 Ga), silicate crust began aggregating to form what became cratons. These cratons are the oldest physical components of our planet that can be sampled and studied. Using various techniques, we continue to tease out the early history of the Earth and how it became the habitable planet it is today.

THE SURVIVORS: ARCHEAN CRATONS

Cratons are parts of the Earth’s crust and lithospheric mantle that have not experienced significant penetrative deformation or calc-alkalic magmatism for hundreds of millions to billions of years. Their long-term stability is related to their 150–250 km thick, cool, and viscous lithospheric mantle roots (Bedle et al. 2021; Cooper and Miller 2024 this issue). Archean cratons are found on every continent(Fig. 1).

Archean cratons typically are composed of two distinct rock assemblages: high-grade gneiss terranes and greenstone belts. High-grade gneiss terranes are composed of complex suites of metamorphosed intrusive rocks sometimes referred to as “grey gneiss.” These terranes, which typically include rocks of multiple ages, are commonly metamorphosed at amphibolite- to granulite-facies and show evidence of partial melting at the higher grades. Although these terranes are dominated by tonalite-trondhjemitegranodiorite (TTG) gneisses, calc-alkalic intermediate to felsic rocks, tholeiitic and calc-alkalic mafic rocks, and ultramafic rocks such as komatiites and  pyroxenites are commonly present in limited volumes. 

Larger intrusions of calc-alkalic granodiorite and granite tend to be younger and to have formed by partial melting of the sodic TTG suite (Moyen 2011; White et al. 2017). High-grade gneiss terranes and their TTG suites are typical of the time period spanning the Eoarchean to the Mesoarchean (4.03–2.8 Ga) crust (Frost et al. 2023; Laurent et al. 2024 this issue). In contrast, greenstone belts are characterized by a wide variety of supracrustal rocks, both volcanic and sedimentary. Overall, most are elongate, synformal structures and appear on all cratons (Anhaeusser 2014). Their name derives from the greenish color imparted to metabasalts by greenschist metamorphism. In addition to greenstones, these belts also contain ultramafic volcanic rocks with minor intermediate to felsic intrusive, volcanic, and volcaniclastic rocks. The volcanic rocks are typically overlain by sedimentary rocks, including greywacke, tuff, pelite, chert, minor carbonate, and iron formations. Spherule layers in some greenstone belts have been interpreted as fallout from impact events (Anhaeusser 2014). Some greenstone belts also contain >3.5 Ga metacarbonate rocks with stromatolites and filamentous microfossils that may record primitive life. Archean greenstone belts host globally important mineral deposits, including massive sulfide Ni–Cu–Zn deposits, magmatic cumulate Cr deposits, iron-formations, and placer and magmatic lode gold. Structurally, many greenstone belts are characterized by a “dome and keel” structure (Fig. 2), in which shallowly-emplaced granitic plutons intrude the metasupracrustal rocks such as typically found in the Pilbara, Superior, and Zimbabwe greenstone belts (Anhaeusser 2014).

CHALLENGES IN STUDYING ARCHEAN CRATONS

Studying Earth’s oldest rocks can be like trying to learn one’s family history by consulting the oldest living relative with a failing memory: some facts may be reliable, but others are difficult to untangle and place in correct chronological order and geographic location, some reminiscences may be misleading, and for some significant events, there may be no recollection at all. Like geoscientists studying younger rocks, geologists studying Earth’s oldest rocks use a combination of analytical approaches that include field studies, petrology, geochemistry, geophysics, and numerical modeling to try to reconstruct the family history of the planet. However, unlike those who study the Phanerozoic (541 Ma to present), geologists who study the Archean encounter severe challenges that make this task especially difficult, as discussed throughout this issue’s thematic articles and Triple Point.

The Non-uniformitarian Earth

Geologic processes are commonly assumed to be uniformitarian, that is, we can infer how rocks formed and were modified in the past by looking at modern processes that affect the Earth today. This assumption cannot, however, be applied to the first 1.5 to 2 billion years of Earth history. For example, the formation of the core, silicate magma oceans, and the Moon were early, one-time events. Exponential, secular cooling of the planet occurred as the energy of accretion dissipated and radioactive heat production decreased. This cooling led to a continually evolving temperature regime within the early Earth, which likely included a 100–250 °C hotter mantle. This higher temperature regime surely affected the nature and rates of geologic processes, including global tectonics. Instead of rigid plates of crust moving horizontally above a convecting mantle in a modern “mobile lid” regime, Hadean and Archean tectonics may have been driven by mantle upwellings beneath a stationary crust, in a “stagnant lid” scenario similar to that thought to exist on Mars (see Toolkit and Perspective in this issue). Instead of subduction, the early crust may have responded to mantle upwelling and downwelling, including a process called sagduction, and a range of other proposed forms of “drip tectonics” (see Laurent et al. 2024 this issue). In addition, studies of the lunar surface indicate by analogy that extraterrestrial impacts strongly affected the early Earth, but tailed off by the end of the Archean (see Toolkit). The lesson here is that we cannot allow our interpretations of Archean environments or processes to be constrained by our experience with the modern Earth. Rather, we must keep an open and creative mind when trying to imagine and understand its early history. It is also important to remember that although the geodynamic environment of the Hadean-Archean Earth may have been different than today, basic thermodynamic principles still applied. For example, the temperature and pressure at which hydrated basalt melts would be the same in the past as it is today. How and where these conditions were met in the early Earth could have varied: instead of by subduction of oceanic crust, melting may have occurred at the base of thick oceanic plateaus. Although such melts do not form at the base of these plateaus today, the higher geotherms in the Archean suggest that melting was possible at that time. In addition, impacts of large bolides (extraterrestrial bodies of unspecified composition—stony, metallic, gaseous, or a combination) could lead to decompression melting, much as basaltic magmas filled impact craters on the Moon (see Perspective). In short, the tectonic environments in which rocks reach their melting temperature change over time, but the basic laws of physics and chemistry do not—they are timeless and universal. 

Preservation Bias

Another important aspect of placing early Earth evolution in context is to keep in mind that we only have the “leftovers” to examine, which is referred to as preservation bias. Whereas many models of crustal evolution suggest that >50% of the continental crust formed in the Archean (see Toolkit), Archean and older crust available for study make up only ~3% of Earth’s exposed surface. Just as most of the Phanerozoic oceanic crust has been subducted and is no longer available for direct study, tectonic processes may have preferentially destroyed certain Archean rock assemblages, leaving those that survived over-represented in the rock record. We cannot assume, therefore, that what is left of the Archean crustal record is representative of what was present more than 2.5 billion years ago. The fact that cratons are associated with thick lithospheric mantle keels may have helped them survive at the expense of other crust that was more easily recycled or removed by erosion. Consequently, we cannot take for granted that the TTG suite that dominates the high-grade gneiss terranes typical of many cratons was as prevalent in the Archean as the rock record suggests.

Overprinting by Subsequent Events

The older the craton, the more likely it is that subsequent events have obscured its origin and evolution. For example, consider the world’s oldest known rocks of the Acasta gneiss complex in the Slave craton of northwestern Canada. The gneiss that yields the 4.03 Ga zircons for which the gneiss is famous occupies a relatively small (~25 km2) area. Even a cursory look at these exposures shows that this area has experienced a complex history of intrusion, metamorphism, and deformation (Fig. 3). It is little wonder that it took a generation of studies to sort out the complex history of the Slave craton, which would not have been possible without the robustness of the mineral zircon and its ability to yield precise U-Pb geochronologic and isotopic information about magma sources and metamorphic events (see Toolkit; Reimink et al. 2016).

CRATON CLANS AND SUPERCRATONS

As indicated in Fig. 1, Archean cratons vary considerably in size, from relatively small, e.g., Tanzania, Wyoming, Rio de la Plata, and Kola, to larger ones like Superior and Yilgarn. These cratons are found worldwide, embedded within younger continental collages. For example, Laurentia (ancestral North America) is a continental landmass that was assembled around 10 or more Archean cratonic nuclei (Fig. 4). Most surviving cratons have Proterozoic rifted or faulted margins, suggesting they are fragments of what were once larger landmasses. Present-day cratons have been grouped into “clans” based on common geologic histories and isotopic characteristics (Bleeker 2003). These clans may identify cratons that originally composed “supercratons,” which, in turn, may be precursors to the well-known supercontinents Gondwana, Rodinia, and Nuna/Columbia. These supercontinents aggregated, broke up, and reformed in different configurations in repeated cycles over geologic time, resulting in the dispersed locations of these Archean cratons today.

APPROACHES FOR IDENTYFYING ARCHEAN CRATON CLANS

Geologic Histories

Similar geologic histories have long been used to propose correlations among cratons. Bleeker (2003) used a detailed examination of Archean timelines of igneous, metamorphic, and sedimentary events to suggest specific correlations of cratonic fragments with similar geologic histories. He proposed possible correlations between the Slave, Dharwar, Zimbabwe, and Wyoming cratons and suggested they might together make up the Slave “clan” that comprised a supercraton he named Sclavia. The oldest rocks in cratons may also help identify members of a clan. Nutman et al. (2015), for example, proposed that 10 cratons, each containing >3.6 Ga rocks, could have been part of an ancient continent that they called Itsaqia. In contrast, the Superior province is made up of terranes that preserve a geologic history dominated by Neoarchean (2.8–2.5 Ga) crust formed from juvenile mantle reservoirs (Percival et al. 2012). It represents the archetypal craton of Bleeker’s Superia clan. Another distinguishing feature of craton clans is the time of cratonization, which is the time after which the cratons became tectonically stable with minimal, if any, further penetrative deformation or magmatism. According to Bleeker (2003), the Kaapvaal and Pilbara cratons cratonized around 3 Ga, the Superior craton around 2.65 Ga, and the Slave clan cratons by 2.5 Ga. Similar geologic histories only suggest, rather than prove, which cratons were once together in one landmass. In fact, the plethora of different proposals for “supercratons” composed of various combinations of cratons suggests that geologic history alone is not sufficient to classify cratons into clans. 

Evidence from Common Pb

In addition to geologic histories, the initial Pb isotopic composition of cratons can also be used to help identify clans. The initial Pb isotopic composition of the Earth is taken to be that of iron meteorites (see Toolkit). The Pb isotope evolution of the mantle begins with an initial composition well constrained by data from meteorites. Data from feldspars and sulfide ores, such as galena, feed into forward models of crustal evolution (see Mueller et al. 2014). Because continental crust has higher 238U/204Pb than the mantle, it evolves to higher 206Pb/204Pb and 207Pb/204Pb than coeval mantle (Fig. 5). Because 235U decays to 207Pb much faster than 238U decays to 206Pb, growth curves are steep early in Earth history (see Toolkit Fig. 6A), and early-formed reservoirs can quickly evolve to elevated 207Pb/204Pb for a given 206Pb/204Pb. Significantly, the initial Pb isotopic compositions of Archean igneous rocks from some cratons require that these rocks formed from a reservoir with a substantially higher 238U/204Pb than coeval mantle and model continental crust. Thus, the initial Pb isotopic composition can be the basis for grouping cratons into those from high-μ sources (where μ = 238U/204Pb) and those derived from low-μ reservoirs such as contemporary depleted mantle. This is possible because feldspar, particularly potassium feldspar, is characterized by relatively high Pb contents (tens of ppm) and very low U/Pb ratios. These low U/Pb ratios mean that feldspar Pb isotopic compositions do not change much, if at all, over time from radiogenic ingrowth and can hence preserve their initial Pb isotopic compositions. Initial Pb isotopic compositions can also be calculated for rocks of known age from their U and Pb contents and present-day Pb isotopic compositions, assuming no loss or gain of either element.

An example of how Pb isotopic compositions discriminate cratons formed from reservoirs with different μ is shown in Fig. 5. Feldspars and Neoarchean TTG rocks from the high-grade gneiss terrane of the northern Wyoming craton define an array that corresponds to an age of 2.8 Ga, which matches the U-Pb zircon ages of 2.79–2.83 Ga. This agreement means that the measured U, Pb, and Pb isotopic compositions are primary. In the case of the Wyoming craton, first-order modeling suggests that a Hadean reservoir with at least an average crustal μ value (μ ~ 10–11) is required. This in turn suggests that the northern Wyoming craton was built on Hadean lithosphere. This hypothesis is supported by the presence of 4.0 Ga detrital zircons in the metasedimentary rocks intruded by the Neoarchean batholiths (Mueller et al. 2014). The high-μ character of the Wyoming craton contrasts strongly with the low-μ character of the Superior craton (e.g., Mueller and Wooden 1988), but is compatible with the high-μ character of the Slave craton (Thorpe 1972). These observations support a Wyoming-Slave connection reaching back to the Hadean, but do not support a connection with the Superior or other low-μ cratons. In terms of clans, other cratons with high-μ terranes in addition to Wyoming include the western Slave, Kaapvaal, and North China cratons.

Evidence for Hadean Crust from Detrital Zircons

Beyond the high- and low-μ characteristics suggested by Pb isotopes, we can consider the possibility of linking cratons that contain evidence of Hadean crust using the U-Pb age and Hf isotopic compositions of detrital zircons (see Toolkit; Laurent et al. 2024 this issue; O’Neil et al. 2024 this issue). All Hadean zircon grains are detrital, i.e., they have been removed from their host-rock and hence their original geologic context, and have been transported unknown distances from their sources. They provide an interesting, but biased, sample, revealing information only about magmas that crystallized zircon. Nevertheless, they and their inclusions provide important information about the early Earth (Harrison et al. 2017). To date, Hadean zircons have been reported from Australia, India, North America, South America, and Asia. The best-known suite is from the Jack Hills region of the Yilgarn craton of Western Australia, which contains grains up to 4.4 Ga (Harrison 2020). Even so, Hadean grains have not been reported from every craton, including the Superior craton, despite thousands of analyses. Although the Yilgarn is the Pilbara craton’s nearest cratonic neighbor (Fig. 1) and both cratons contain Paleoarchean crust (3.6–3.5 Ga), Hadean zircons have not been reported from Pilbara.

Evidence from Mafic Dike Swarms and Paleomagnetic Data 

Radiating mafic dike swarms have been interpreted as originating from mantle plume centers that may be associated with the break-up of supercratons (Ernst and Bleeker 2010). If so, correlation of well-dated dikes across two cratons may indicate that they once were contiguous. The lateral motion of cratons can then be quantified from information about the direction of the Earth’s magnetic field recorded in igneous and sedimentary rocks at the time they acquired their magnetization. Because magnetic poles rarely lie far from the geographic poles, the changes in orientation of the craton in relation to the poles track the motion of the craton across the Earth’s surface. Unfortunately, highquality paleomagnetic data are sparse in the Archean compared with the Proterozoic and Phanerozoic. Extant data have, however, been used to suggest a Vaalbara supercraton between 2.8 and 2.2 Ga, that break-up of Superia did not occur until 2.1 Ga, that break-up of Sclavia occurred at 2.0 Ga, and that these two supercratons were not likely adjacent from 2.4 to 2.2 Ga (Salminen et al. 2021).

CRATON CLANS AND SHIFTING ALLEGIANCES: WYOMING, SLAVE, AND SUPERIOR

The Wyoming craton has been grouped with the Sclavia clan because of the similar geologic histories between the northern part of the Wyoming craton and the western half of the Slave craton (Bleeker 2003). Both expose extensive 3.5–3.0 Ga TTG gneisses and their detrital zircon populations extend back to 4.0 Ga. They share high-μ common Pb isotopic signatures, and the negative initial εNd of the Paleoarchean gneisses in both cratons indicate derivation from an ancient evolved source (Wooden and Mueller 1988; Thorpe 1972; Davis et al. 2003; Frost et al. 2006; Mueller et al. 2014).

At the end of the Mesoarchean, sedimentary sequences interpreted to be rift-related were deposited on both the northern Wyoming and western Slave cratons. The southwestern Slave craton was covered by a supracrustal sequence composed of quartz arenite, sandstone-argillite, banded iron formation (BIF), quartz pebble conglomerate, and minor felsic volcaniclastic rocks and mafic volcanic flows (Corcoran 2012). Along the southern margin of the Wyoming province, similar supracrustal rocks (quartzite, BIF, metapelite, metabasalt, and minor felsic volcanic rocks) were deposited in the Neoarchean (Mogk et al. 2023). Subsequently, their geologic histories diverge, suggesting that if the two cratons were originally adjacent, they had moved apart by 2.85 Ga (Fig. 6). The later geologic histories

of the two cratons are distinct: the Eastern Slave accreted to the western Slave craton between 2.7 and 2.6 Ga (Davis et al. 2003), and in the Wyoming craton, the southern accreted terranes docked at ~2.65 Ga (Mogk et al. 2023).

What happened to Wyoming after it rifted from Slave? Correlated mafic dike swarms and paleomagnetic data suggest that the Wyoming craton may have changed allegiances and joined Superia by the end of the Archean (Fig. 6). The collision of accreted terranes along Wyoming’s southern margin at ~2.65 Ga has been interpreted to be related to closure of an ocean between Superior and Wyoming (Ernst and Bleeker 2010). High-quality paleomagnetic data place Wyoming adjacent to Superia by 2.45 Ga (Fig. 6; Salminen et al. 2021).

IMPLICATIONS

Archean cratons represent an incomplete record of early Earth history. The surviving cratons almost certainly represent only a fraction of the original Archean crust. They may not be representative of Archean crust as a whole if cratons were preserved preferentially because of their thick, strong, lithospheric roots. Their geologic, petrologic, geochemical, and isotopic characteristics are complicated by a long history of deformation, partial melting, and metamorphism before cratonization. Consequently, detailed study using multiple approaches provides the best insight into early Earth processes and the protracted journey from craton to continent. Key insights include:

• It appears that crust formation on Earth was not a global, synchronous event. Instead, different craton clans formed at different times from different reservoirs. In the example we discussed, the Slave and Wyoming cratons formed in the Paleoarchean from a similar, high-U/Pb reservoir, incorporating detrital and xenocrystic zircon as old as 4 Ga. By contrast, most of the Superior craton is characterized by greenstone terranes formed from juvenile mantle sources in the Neoarchean.
• By the end of the Hadean (if not earlier), the Earth had differentiated into a number of geochemical and isotopic reservoirs. We discussed how common Pb isotope data suggest highand low-μ reservoirs formed in the Hadean. Using excesses in 142Nd, O’Neil et al. (2024 this issue) present evidence for multiple depletion events in the Hadean mantle. They note that these excesses are not a feature shared by all cratons, suggesting that early crustal genesis and mantle depletion may have been local, not global.
• Rifted and faulted cratonic margins and paleomagnetic evidence for lateral movement of cratons suggest that mobile lid tectonics was operating by the Neoarchean. Although paleomagnetic data indicate the possibility of far traveled cratons, they do not necessarily imply that plate tectonics was widespread in the Archean. As suggested by Rey et al. (2024 this issue), both mobile lid and stagnant lid tectonics may have operated simultaneously in the early Archean. As the planet cooled and rigid plates formed, this temporary dual-mode tectonic scenario eventually was replaced by plate tectonics at the expense of the stagnant lid regime.
• The small volume of surviving Archean crust may mean that preservation bias affected the remaining array of cratons. It is likely significant that these survivors are characterized by thick mantle roots, which may have protected them from modification and destruction from the Archean to the present. As pointed out by Cooper

and Miller (2024 this issue), sharp lateral changes in temperature between thick cratonic lithosphere and surrounding mantle can induce edge-driven convective instabilities that weaken, modify, and deform cratonic lithosphere. The lithosphere beneath the eastern part of the North China craton appears to have thinned as a consequence of Mesozoic flat-slab subduction (Bedle et al. 2021; Cooper and Miller 2024 this issue), suggesting that other cratons also may have been weakened and even destroyed by similar processes.

Understanding the evolution of the Earth’s earliest crust and mantle is a challenging area of research for geoscientists at large. Using a variety of geologic, geochemical, isotopic, geophysical, and geodynamical tools and models, we continue to gain increasingly detailed insight into our planet’s earliest history. This issue of Elements presents aspects of our current understanding that we hope will stimulate interest in the early Earth and spur future investigations.

ACKNOWLEDGMENTS

We acknowledge the many thought-provoking conversations and exchanges with numerous colleagues over the years. We thank the authors and reviewers for their critical contributions to this special issue of Elements and extend special thanks to Principal Editor Janne Blichert-Toft for her expert guidance.

REFERENCES

Anhaeusser CR (2014) Archaean greenstone belts and associated granitic rocks – a review. Journal of African Earth Sciences 100: 684-732, doi: 10.1016/j.jafrearsci.2014.07.019

Bedle H, Cooper CM, Frost CD (2021) Nature versus nurture: preservation and destruction of Archean cratons. Tectonics 40: e2021TC006714, doi: 10.1029/2021TC006714

Bleeker W (2003) The late Archean record: a puzzle in ca. 35 pieces. Lithos 71: 99-134, doi: 10.1016/j.lithos.2003.07.003

Cooper CM, Miller MS (2023) Embracing craton complexity at depth. Elements 20: 187-192

Corcoran PL (2012) Archaean sedimentary sequences of the Slave craton. Marine and Petroleum Geology 33: 80-91, doi: 10.1016/j.marpetgeo.2011.08.006

Davis WJ, Canil D, MacKenzie JM, Carbno GB (2003) Petrology and U-Pb geochronology of lower crustal xenoliths and the development of a craton, Slave Province, Canada. Lithos 71: 541-573, doi: 10.1016/S0024-4937(03)00130-0

Ernst R, Bleeker W (2010) Large igneous provinces (LIPs), giant dyke swarms, and mantle plumes: significance for breakup events within Canada and adjacent regions from 2.5 Ga to the present. Canadian Journal of Earth Sciences 47: 695-739, doi: 10.1139/E10-025

Frost CD, Frost BR, Kirkwood R, Chamberlain KR (2006) The tonalite–trondhjemite–granodiorite (TTG) to granodiorite–granite (GG) transition in the Late Archean plutonic rocks of the central Wyoming province. Canadian Journal of Earth Sciences 43: 1419-1444, doi: 10.1139/e06-082

Frost CD, Mueller PA, Mogk DW, Frost BR, Henry DJ (2023) Creating continents: Archean cratons tell the story. GSA Today 33: 4-10, doi: 10.1130/GSATG541A.1

Harrison TM, Bell EA, Boehnke P (2017) Hadean zircon petrochronology. Reviews in Mineralogy and Geochemistry 83: 329-363, doi: 10.2138/rmg.2017.83.11

Harrison TM (2020) Hadean Earth. Springer International Publishing, 291 pp, doi: 10.1007/978-3-030-46687-9

Hickman-Lewis K and 11 coauthors (2023) Advanced two- and three-dimensional insights into Earth’s oldest stromatolites (ca. 3.5 Ga): prospects for the search for life on Mars. Geology 51: 33-38, doi: 10.1130/G50390.1

Laurent O, Guitreau M, Bruand E, Moyen J-F (2023) At the dawn of continents: Archean tonalite-trondhjemite-granodiorite suites. Elements 20: 174-179

Mogk DW, Frost CD, Mueller PA, Frost BR, Henry DJ (2023) Crustal genesis and evolution of the Archean Wyoming Province: continental growth through vertical magmatic and horizontal tectonic processes. In: Whitmeyer SJ, Williams ML, Kellett DA, Tikoff B (eds) Laurentia: Turning Points in the Evolution of a Continent. Geological Society of America Memoir 220: 1-24, doi: 10.1130/2022.1220(01)

Moyen J-F (2011) The composite Archaean grey gneisses: petrologic significance, and evidence for a non-unique tectonic setting for Archaean crustal growth. Lithos 123: 21-36, doi: 10.1016/j.lithos.2010.09.015

Mueller PA, Wooden JL (1988) Evidence for Archean subduction and crustal recycling, Wyoming Province. Geology 16: 871-874, doi: 10.1130/0091-7613(1988)016%3C0871:EFASAC%3E2.3.CO;2

Mueller PA, Mogk DW, Henry DJ, Wooden JL, Foster DA (2014) The plume to plate transition: Hadean and Archean crustal evolution in the northern Wyoming Province, U.S.A. In: Dilek Y, Furnes H (eds) Evolution of Archean Crust and Early Life. Modern Approaches to Solid Earth Sciences, Volume 7. Springer, Dordrecht, pp 23-54, doi: 10.1007/978-94-007-7615-9_2

Nimmo F, Kleine T (2015) Early differentiation and core formation: processes and timescales. In: Badro J, Walter MJ (eds) The Early Earth: Accretion and Differentiation. AGU Geophysical Monograph 212, Washington DC, pp 83-102, doi: 10.1002/9781118860359.ch5

Nutman AP and 6 coauthors (2013) The Itsaq gneiss complex of Greenland: episodic 3900 to 3660 Ma juvenile crust formation and recycling in the 3660 to 3600 Ma Isukasian orogeny. American Journal of Science 313: 877-911, doi: 10.2475/09.2013.03

Nutman AP, Bennett VC, Friend CRL (2015) Proposal for a continent ‘Itsaqia’ amalgamated at 3.66 Ga and rifted apart from 3.53 Ga: initiation of a Wilson Cycle near the start of the rock record. American Journal of Science 315: 509-536, doi: 10.2475/06.2015.01

O’Neil J, Rizo H, Reimink J, Garçon M, Carlson RW (2024) Earth’s earliest crust. Elements 20: 168-173

Percival JA and 6 coauthors (2012) Geology and tectonic evolution of the Superior Province, Canada. In: Percival JA, Cook FA, Clowes RM (eds) Tectonic Styles in Canada: The Lithoprobe Perspective. Geological Association of Canada Special Paper 49, 321-378

Reimink JR and 8 coauthors (2016) No evidence for Hadean continental crust within Earth’s oldest evolved rock unit. Nature Geoscience 9: 777-780, doi: 10.1038/NGEO2786

Rey PF, Coltice N, Flament N (2024) Archean geodynamics underneath weak, flat, and flooded continents. Elements 20: 180-186

Salminen J, Pehrsson S, Evans DAD, Wang C (2021) Neoarchean–Paleoproterozoic supercycles. In: Pesonen LJ, Salminen J, Elming S-Å, Evans DAD, Veikkolainen T (eds) Ancient Supercontinents and the Paleogeography of Earth. Elsevier, Amsterdam, pp 465-498, doi: 10.1016/B978-0-12-818533-9.00014-X

St Onge MR, van Gool J, Garde AA, Scott D (2009) Correlation of Archaean and Palaeoproterozoic units between northeastern Canada and western Greenland: constraining the pre-collisional upper late accretionary history of the Trans-Hudson orogen. In: Cawood PA, Kröner A (eds) Earth Accretionary Systems in Space and Time. Geological Society, London, Special Publication 318: 193-235, doi: 10.1144/SP318.7

Stacey JS, Kramers JD (1975) Approximation of terrestrial lead isotope evolution by a two-stage model. Earth and Planetary Science Letters 26: 207-221, doi: 10.1016/0012-821X(75)90088-6

Thorpe RI (1972) Interpretation of lead isotope data for base metal and gold deposits, Slave Province, Northwest Territories. In: Report of Activities, Pt. B, Nov. 1971 to March 1972, Geological Survey of Canada Paper 72-1B: 72-77, doi: 10.4095/105276

Valley JW and 10 coauthors (2014) Hadean age for a post-magma ocean zircon confirmed by atom-probe tomography. Nature Geoscience 7: 219-223, doi: 10.1038/ngeo2075

White RW, Palin RM, Green ECR (2017) High-grade metamorphism and partial melting in Archean composite grey gneiss complexes. Journal of Metamorphic Geology 35: 181-195, doi: 10.1111/jmg.12227

Wilde SA, Valley JW, Peck WH, Graham CM (2001) Evidence from detrital zircons for the existence of continental crust and oceans on the Earth 4.4 Gyr ago. Nature 409: 175-178, doi: 10.1038/35051550

Wooden JL, Mueller PA (1988) Pb, Sr and Nd isotopic compositions of a suite of Late Archean, igneous rocks, eastern Beartooth Mountains: implications for crust-mantle evolution. Earth and Planetary Science Letters 87: 59-72, doi: 10.1016/0012-821X(88)90064-7