A whole-lithosphere view of continental growth
Affiliations | Corresponding Author | Cite as | Funding information
Published by the European Association of Geochemistry
under Creative Commons License CC BY 4.0
Keywords: continental crust, Zircon, crustal reworking, Hf-isotope, O-isotope
- Share this article
-
Article views:917Cumulative count of HTML views and PDF downloads.
- Download Citation
- Rights & Permissions
top
Abstract
Figures
 Figure 1 (a) Various crustal growth rate curves shown by dashed lines compared to the cratonic mantle age distribution. Vertical coloured arrows show the amount of continental recycling predicted by various estimates. The black curve shows the age distribution of unmodified cratonic mantle ages (Pearson et al., 2021), while the green field shows the percentage of preserved mature sedimentary packages (Reimink et al., 2021). (b) Shows the continental growth curve predicted in this work. The grey curve shows U–Pb ages of the preserved rock record (Puetz et al., 2018). The black curve is the age distribution of unmodified cratonic mantle roots (Pearson et al., 2021). The green curve is the crustal growth curve calculated here using bulk rock major element chemistry. The dashed green curve uses the same calculations as the solid green curve except for the utilisation of a synthetic Mesoarchean–Hadean rock record, to evaluate potential for sampling bias. |  Figure 2 The distribution of zircon oxygen isotope compositions through time, highlighting the tendency of zircon oxygen isotope ratios to become progressively more extreme through time, making them a poor discriminant of continental reworking. The orange band shows a typical field for ‘juvenile’ zircon isotope compositions. |  Figure 3 The fraction of rocks classified as reworked continental crust from Figure 1 using the bulk geochemistry approach adopted here (green curve with grey band showing 2SE uncertainty) compared to the reworking estimates based on zircon oxygen isotope ratios. Curves calculated in 200 Ma moving windows in 10 Ma time steps. The zircon record dramatically under-estimates reworking in the Archean. |  Figure 4 Reworking through time as seen through the whole-rock record. (a) The whole-rock geochemical data used to calculate a ‘theta’ value, with experimental melts shown coloured by starting composition. (b) Our classifier for data shown in panel b, showing the probability density for each category across a range of Theta values. (c) The distribution of Theta values across geological time. A probability density estimator is shown for the theta values for rocks split into 100 Ma age bins and coloured according to the number of rock samples in each age bin. Rocks that likely represent reworked crust fall into the blue field and have been present since the early Archean. |
| Figure 1 | Figure 2 | Figure 3 | Figure 4 |
top
Introduction
Earth’s intermediate to felsic composition continental crust is thicker and more buoyant than mafic oceanic crust, and represents an excellent archive of fundamental processes such as regulating the long-term carbon cycle, concentrating and hosting valuable mineral deposits, and providing unique habitats for biological development and diversification. Despite the importance of continental crust to humanity, there is little consensus on the timing of its formation and its stabilisation. Many widely used models, some of which are underpinned by zircon U–Pb and Hf isotope systematics (Belousova et al., 2010
Belousova, E.A., Kostitsyn, Y.A., Griffin, W.L., Begg, G.C., O’Reilly, S.Y., Pearson, N.J. (2010) The growth of the continental crust: Constraints from zircon Hf-isotope data. Lithos 119, 457–466. https://doi.org/10.1016/j.lithos.2010.07.024
; Dhuime et al., 2012Dhuime, B., Hawkesworth, C.J., Cawood, P.A., Storey, C.D. (2012) A Change in the Geodynamics of Continental Growth 3 Billion Years Ago. Science 335, 1334–1336. https://doi.org/10.1126/science.1216066
), emphasize the potential importance of voluminous continental growth in the Archean. These models produce a dramatic inflection point at ∼3 Ga, where continental growth rate purportedly subsided and large tracts of crust were stabilised. However, there is little preserved evidence of this hypothesised voluminous ancient continental crust. Likewise, the continental lithospheric mantle—thick roots that stabilised ancient continental crust—have little extensive record prior to ∼3.0 Ga (Pearson et al., 2021Pearson, D.G., Scott, J.M., Liu, J., Schaeffer, A., Wang, L.H., van Hunen, J., Szilas, K., Chacko, T., Kelemen, P.B. (2021) Deep continental roots and cratons. Nature 596, 199–210. https://doi.org/10.1038/s41586-021-03600-5
).Determining the relationship between the continental crust and the cratonic lithospheric mantle roots (CLMR) is critical in deciphering the growth of continental crust. In the modern Earth, deep lithospheric roots stabilise and appear to preserve extant continental crust (Lee et al., 2017
Lee, C.-T.A., Caves, J., Jiang, H., Cao, W., Lenardic, A., McKenzie, N.R., Shorttle, O., Yin, Q., Dyer, B. (2017) Deep mantle roots and continental emergence: implications for whole-Earth elemental cycling, long-term climate, and the Cambrian explosion. International Geology Review 60, 1–18. https://doi.org/10.1080/00206814.2017.1340853
), so the discrepancy between preserved lithospheric mantle ages and previous crustal growth models (Fig. 1) suggests a genetic disconnect. Recent studies of the continental crust-CLMR relationship suggest that, while very ancient continental crust was formed and rapidly destroyed, the formation of lithospheric roots stabilised existing continental crust, thereby slowing the destruction, and growth rate, of continents (Hawkesworth et al., 2017Hawkesworth, C.J., Cawood, P.A., Dhuime, B., Kemp, T.I.S. (2017) Earth’s Continental Lithosphere Through Time. Annual Review of Earth and Planetary Sciences 45, 169–198. https://doi.org/10.1146/annurev-earth-063016-020525
; Pearson et al., 2021Pearson, D.G., Scott, J.M., Liu, J., Schaeffer, A., Wang, L.H., van Hunen, J., Szilas, K., Chacko, T., Kelemen, P.B. (2021) Deep continental roots and cratons. Nature 596, 199–210. https://doi.org/10.1038/s41586-021-03600-5
). This has been taken by some authors to indicate that continental crust and continental lithospheric mantle are related by selective preservation—and that their formation mechanisms were not related (Fig. 1). However, there are considerable uncertainties in models for continental growth rates.Figure 1 (a) Various crustal growth rate curves shown by dashed lines compared to the cratonic mantle age distribution. Vertical coloured arrows show the amount of continental recycling predicted by various estimates. The black curve shows the age distribution of unmodified cratonic mantle ages (Pearson et al., 2021
Pearson, D.G., Scott, J.M., Liu, J., Schaeffer, A., Wang, L.H., van Hunen, J., Szilas, K., Chacko, T., Kelemen, P.B. (2021) Deep continental roots and cratons. Nature 596, 199–210. https://doi.org/10.1038/s41586-021-03600-5
), while the green field shows the percentage of preserved mature sedimentary packages (Reimink et al., 2021Reimink, J.R., Davies, J.H.F.L., Ielpi, A. (2021) Global zircon analysis records a gradual rise of continental crust throughout the Neoarchean. Earth and Planetary Science Letters 554, 116654. https://doi.org/10.1016/j.epsl.2020.116654
). (b) Shows the continental growth curve predicted in this work. The grey curve shows U–Pb ages of the preserved rock record (Puetz et al., 2018). The black curve is the age distribution of unmodified cratonic mantle roots (Pearson et al., 2021Pearson, D.G., Scott, J.M., Liu, J., Schaeffer, A., Wang, L.H., van Hunen, J., Szilas, K., Chacko, T., Kelemen, P.B. (2021) Deep continental roots and cratons. Nature 596, 199–210. https://doi.org/10.1038/s41586-021-03600-5
). The green curve is the crustal growth curve calculated here using bulk rock major element chemistry. The dashed green curve uses the same calculations as the solid green curve except for the utilisation of a synthetic Mesoarchean–Hadean rock record, to evaluate potential for sampling bias.Destruction of an ancient crustal record can take several forms, but is typically separated into two categories (Cawood et al., 2013
Cawood, P.A., Hawkesworth, C.J., Dhuime, B. (2013) The continental record and the generation of continental crust. Geological Society of America Bulletin 125, 14–32. https://doi.org/10.1130/B30722.1
): 1) Reworking—processes that overprint the radiometric ages of the crustal record, but do not remove mass from the continents, such as partial melting and sedimentary erosion, and 2) Recycling—full scale removal of continental mass back into the mantle by delamination or sediment and continental subduction.While there is isotopic evidence for some amount of continental crust recycled into the modern mantle (Jackson et al., 2007
Jackson, M.G., Hart, S.R., Koppers, A.A.P., Staudigel, H., Konter, J., Blusztajn, J., Kurz, M., Russell, J.A. (2007) The return of subducted continental crust in Samoan lavas. Nature 448, 684–687. https://doi.org/10.1038/nature06048
), there is little evidence for vast volumes of ancient continental material residing in the mantle. For instance, volatile-element isotopic measurements typically indicate the onset of detectable crustal recycling near ∼2.5 Ga (Coltice et al., 2000Coltice, N., Albarède, F., Gillet, P. (2000) 40K-40Ar Constraints on Recycling Continental Crust into the Mantle. Science 288, 845–847. https://doi.org/10.1126/science.288.5467.845
; Parai and Mukhopadhyay, 2018Parai, R., Mukhopadhyay, S. (2018) Xenon isotopic constraints on the history of volatile recycling into the mantle. Nature 560, 223–227. https://doi.org/10.1038/s41586-018-0388-4
; Labidi et al., 2020Labidi, J., Barry, P.H., Bekaert, D.V., Broadley, M.W., Marty, B., Giunta, T., Warr, O., Sherwood Lollar, B., Fischer, T.P., Avice, G. and Caracausi, A. (2020) Hydrothermal 15 N 15 N abundances constrain the origins of mantle nitrogen. Nature 580, 367–371. https://doi.org/10.1038/s41586-020-2173-4
). Thus, recycling is unlikely to be a major factor in destroying large volumes of primary continental crust before 2.5 Ga. Yet, the most ancient of crust—older than ∼3.8 Ga—is restricted in its exposure at the surface of the Earth to the parts per million level. Thus, if large volumes of ancient continental crust did exist on Earth, reworking must be primarily responsible for overprinting the ancient geochronological signatures of that crust. Therefore, accurate quantification of reworking is fundamentally important to model the primary formation age of continental landmasses.top
Estimates of Continental Growth Rate
Recent estimates of the volumes of continental crust throughout Earth’s history have relied on zircon Hf and O isotopes (Belousova et al., 2010
Belousova, E.A., Kostitsyn, Y.A., Griffin, W.L., Begg, G.C., O’Reilly, S.Y., Pearson, N.J. (2010) The growth of the continental crust: Constraints from zircon Hf-isotope data. Lithos 119, 457–466. https://doi.org/10.1016/j.lithos.2010.07.024
; Dhuime et al., 2012Dhuime, B., Hawkesworth, C.J., Cawood, P.A., Storey, C.D. (2012) A Change in the Geodynamics of Continental Growth 3 Billion Years Ago. Science 335, 1334–1336. https://doi.org/10.1126/science.1216066
; Korenaga, 2018Korenaga, J. (2018) Estimating the formation age distribution of continental crust by unmixing zircon ages. Earth and Planetary Science Letters 482, 388–395. https://doi.org/10.1016/j.epsl.2017.11.039
). Hafnium isotopes in zircon can be used to both calculate the age of crust formation and the time the source material was extracted from the mantle, while oxygen isotope ratios have been used as a filter (Dhuime et al., 2012Dhuime, B., Hawkesworth, C.J., Cawood, P.A., Storey, C.D. (2012) A Change in the Geodynamics of Continental Growth 3 Billion Years Ago. Science 335, 1334–1336. https://doi.org/10.1126/science.1216066
). Zircon oxygen isotope ratios reflect the oxygen isotope composition of the source to the zircon-forming magmas. Oxygen isotope ratios in igneous rocks can be shifted from the unaltered mantle value by incorporation of material that has interacted with surface waters, i.e. sediments. Such incorporation of sedimentary rocks into the igneous system, e.g., continental recycling, can generate zircon with high oxygen isotope ratios (Valley et al., 2005Valley, J.W., Lackey, J.S., Cavosie, A.J., Clechenko, C.C., Spicuzza, M.J., Basei, M.A.S., Bindeman, I.N., Ferreira, V.P., Sial, A.N., King, E.M., Peck, W.H., Sinha, A.K. and Wei, C.S. (2005) 4.4 billion years of crustal maturation: oxygen isotope ratios of magmatic zircon. Contributions to Mineralogy and Petrology 150, 561–580. https://doi.org/10.1007/s00410-005-0025-8
), at least on the modern Earth.Zircon U–Pb, Hf, and O isotope data have been combined to calculate widely used crustal growth estimates (Dhuime et al., 2012
Dhuime, B., Hawkesworth, C.J., Cawood, P.A., Storey, C.D. (2012) A Change in the Geodynamics of Continental Growth 3 Billion Years Ago. Science 335, 1334–1336. https://doi.org/10.1126/science.1216066
) that indicate rapid crustal growth in the Archean and a shift in crustal growth rate near 3.0 Ga to slower growth. Despite the popularity of this approach, drawbacks have been pointed out (Korenaga, 2018Korenaga, J. (2018) Estimating the formation age distribution of continental crust by unmixing zircon ages. Earth and Planetary Science Letters 482, 388–395. https://doi.org/10.1016/j.epsl.2017.11.039
). A fundamental issue with the U–Pb/Hf/O approaches is that it is solely reliant on zircon geochemistry to accurately track continental recycling throughout geological time. While the combination of the U–Pb and Hf isotope systems may be reliable (Korenaga, 2018Korenaga, J. (2018) Estimating the formation age distribution of continental crust by unmixing zircon ages. Earth and Planetary Science Letters 482, 388–395. https://doi.org/10.1016/j.epsl.2017.11.039
), the use of oxygen isotopes to track recycling is prone to uncertainty. For instance, recent studies have shown that zircon oxygen isotope ratios do not accurately identify sediment recycling in the Neoarchean (Bucholz and Spencer, 2019Bucholz, C.E., Spencer, C.J. (2019) Strongly Peraluminous Granites across the Archean–Proterozoic Transition. Journal of Petrology 60, 1299–1348. https://doi.org/10.1093/petrology/egz033
), a critical time period for constraining continental growth estimates. Additionally, the maximum oxygen isotope ratio of igneous zircons and shales (mature sedimentary rocks) continues to increase over time (Valley et al., 2005Valley, J.W., Lackey, J.S., Cavosie, A.J., Clechenko, C.C., Spicuzza, M.J., Basei, M.A.S., Bindeman, I.N., Ferreira, V.P., Sial, A.N., King, E.M., Peck, W.H., Sinha, A.K. and Wei, C.S. (2005) 4.4 billion years of crustal maturation: oxygen isotope ratios of magmatic zircon. Contributions to Mineralogy and Petrology 150, 561–580. https://doi.org/10.1007/s00410-005-0025-8
; Bindeman et al., 2018Bindeman, I.N., Zakharov, D.O., Palandri, J., Greber, N.D., Dauphas, N., Retallack, G.J., Hofmann, A., Lackey, J.S., Bekker, A. (2018) Rapid emergence of subaerial landmasses and onset of a modern hydrologic cycle 2.5 billion years ago. Nature 557, 545–548. https://doi.org/10.1038/s41586-018-0131-1
). This means that the sensitivity of the O-isotope reworking metric also changes throughout geological time (Fig. 2), making the proxy significantly less sensitive in the Archean than today. If continental reworking in the Neoarchean is under- or overestimated, it will impose a dramatic bias on any derivative continental growth curve.Figure 2 The distribution of zircon oxygen isotope compositions through time, highlighting the tendency of zircon oxygen isotope ratios to become progressively more extreme through time, making them a poor discriminant of continental reworking. The orange band shows a typical field for ‘juvenile’ zircon isotope compositions.
To circumvent these issues, we take an approach that integrates the detrital zircon Hf isotope record—the record of the mantle extraction age of continental crust—with the bulk composition of the preserved continental rock record, to identify and correct for crustal reworking. We adopt this approach because the major element composition of igneous rocks can accurately quantify the extent of reworking of previous continental crust, whether reworking occurs via incorporation of sediments or direct melting of pre-existing continental crust (Frost and Frost, 2008
Frost, B.R., Frost, C.D. (2008) A Geochemical Classification for Feldspathic Igneous Rocks. Journal of Petrology 49, 1955–1969. https://doi.org/10.1093/petrology/egn054
; Moyen et al., 2017Moyen, J.-F., Laurent, O., Chelle-Michou, C., Couzinié, S., Vanderhaeghe, O., Zeh, A., Villaros, A., Gardien, V. (2017) Collision vs. subduction-related magmatism: Two contrasting ways of granite formation and implications for crustal growth. Lithos 277, 154–177. https://doi.org/10.1016/j.lithos.2016.09.018
). The major-element bulk composition approach is not inherently biased towards specific rock types. Contrary to oxygen isotope ratios, the range of major element compositions of igneous rocks is limited by the thermodynamics of partial melting: by mantle melting on one side and eutectic granite melting on the other. Thus, igneous rocks have strict limits to their composition irrespective of their age, rendering them accurate and consistent tracers of continental reworking through geological time.top
Crustal Reworking
Our estimate of the reworking rate of continental crust (Fig. 3) uses input from the classic ACNK/ANK diagram of Shand, (1943)
Shand, S.J. (1943) Eruptive rocks: their genesis, composition, and classification, with a chapter on meteorites. J. Wiley & sons, Incorporated.
, further developed to isolate source composition from fractionation and assimilation trends in magmatic rocks by Moyen et al. (2017)Moyen, J.-F., Laurent, O., Chelle-Michou, C., Couzinié, S., Vanderhaeghe, O., Zeh, A., Villaros, A., Gardien, V. (2017) Collision vs. subduction-related magmatism: Two contrasting ways of granite formation and implications for crustal growth. Lithos 277, 154–177. https://doi.org/10.1016/j.lithos.2016.09.018
. This method is explained in detail in the Supplementary Materials. In this projection, a theta value of 10–30 degrees reflects crustal and peraluminous melt sources, i.e. a rock formed by continental reworking, whereas a theta value of less than 10 represents juvenile primitive magma with minimal crustal input (Fig. 4). An added benefit of this metric is that it classifies melts of continental crustal rocks in a similar way to melts of sedimentary rocks—useful for our purposes as both origins reflect continental reworking. This is an improvement on the commonly used aluminum saturation index, another metric used to classify whole rock geochemical data that successfully discriminates pure sediment derived melts but does not identify evolved (fractionated) compositions formed from igneous sources—a composition we must accurately identify when considering continental recycling through time.Figure 3 The fraction of rocks classified as reworked continental crust from Figure 1 using the bulk geochemistry approach adopted here (green curve with grey band showing 2SE uncertainty) compared to the reworking estimates based on zircon oxygen isotope ratios. Curves calculated in 200 Ma moving windows in 10 Ma time steps. The zircon record dramatically under-estimates reworking in the Archean.
Figure 4 Reworking through time as seen through the whole-rock record. (a) The whole-rock geochemical data used to calculate a ‘theta’ value, with experimental melts shown coloured by starting composition. (b) Our classifier for data shown in panel b, showing the probability density for each category across a range of Theta values. (c) The distribution of Theta values across geological time. A probability density estimator is shown for the theta values for rocks split into 100 Ma age bins and coloured according to the number of rock samples in each age bin. Rocks that likely represent reworked crust fall into the blue field and have been present since the early Archean.
The theta value calculation does not divide rock compositions into a binary ‘reworked’ or ‘juvenile’ category. Instead, we have employed a naïve Bayesian classifier to calculate five probabilities for each whole-rock composition, one for each class of source materials ranging from ultramafic to sedimentary. These probabilities are divided into two groups, reworked and juvenile (see Supplemental Methods for further explanation). The sum total of each reworking and juvenile probability, across all individual whole-rock measurements in any particular age bin, were then totalled to determine the reworking fraction in that age bin.
The resulting trace of crustal reworking through geological time, as viewed by the whole-rock elemental record, primarily differs from the zircon oxygen isotope record in that it is relatively constant through time. For instance, using the major element-based temporal trace in Fig. 3, the fraction of reworked crust varies by less than a factor of two (only between 0.5 and 0.3 post- 3 Ga), whereas the O-isotope based trace varies by a factor of seven, with significant swings in magnitude over short time intervals (Fig. 3).
In the Neoarchean, whole-rock data indicate significantly more reworking than the zircon oxygen isotope model. The causes of this difference are not readily apparent, but may be due to anoxic weathering conditions that adversely affect the ability of oxygen isotope ratios to accurately track continental reworking (Bucholz and Spencer, 2019
Bucholz, C.E., Spencer, C.J. (2019) Strongly Peraluminous Granites across the Archean–Proterozoic Transition. Journal of Petrology 60, 1299–1348. https://doi.org/10.1093/petrology/egz033
). Many sedimentary rocks and their derivative melts are known from the Neoarchean period (Donaldson and de Kemp, 1998Donaldson, J.A., de Kemp, E.A. (1998) Archaean quartz arenites in the Canadian Shield: examples from the Superior and Churchill Provinces. Sedimentary Geology 120, 153–176. https://doi.org/10.1016/S0037-0738(98)00031-1
; Laurent et al., 2014Laurent, O., Martin, H., Martin, H., Moyen, J.F., Moyen, J.-F., Doucelance, R., Doucelance, R. (2014) The diversity and evolution of late-Archean granitoids: Evidence for the onset of “modern-style” plate tectonics between 3.0 and 2.5Ga. Lithos 205, 1–82. https://doi.org/10.1016/j.lithos.2014.06.012
), indicating that continental reworking took place; yet, the existence of significant crustal reworking is not clearly captured by the zircon oxygen isotope record. The whole-rock bulk geochemical compositional record appears to be a more reliable index of crustal reworking than the zircon oxygen isotope tracer for a combination of reasons. The angular projection employed in our “reworking index”, based on major elements, can identify the recycling of continental igneous rocks, as well as the products of melting of sedimentary rocks—a phenomenon that zircon oxygen isotope ratios are only well-suited to detect in the post-NeoArchean rock record (Fig. 2) after large volumes of sedimentary rocks began to be deposited following continental emergence (Valley et al., 2005Valley, J.W., Lackey, J.S., Cavosie, A.J., Clechenko, C.C., Spicuzza, M.J., Basei, M.A.S., Bindeman, I.N., Ferreira, V.P., Sial, A.N., King, E.M., Peck, W.H., Sinha, A.K. and Wei, C.S. (2005) 4.4 billion years of crustal maturation: oxygen isotope ratios of magmatic zircon. Contributions to Mineralogy and Petrology 150, 561–580. https://doi.org/10.1007/s00410-005-0025-8
; Reimink et al., 2021Reimink, J.R., Davies, J.H.F.L., Ielpi, A. (2021) Global zircon analysis records a gradual rise of continental crust throughout the Neoarchean. Earth and Planetary Science Letters 554, 116654. https://doi.org/10.1016/j.epsl.2020.116654
).We leverage our improved reworking metric to calculate a continental growth rate curve (Fig. 1; see supplementary methods for details) starting from a widely used crustal growth estimate based on zircon U–Pb & Hf isotopes (Dhuime et al., 2012
Dhuime, B., Hawkesworth, C.J., Cawood, P.A., Storey, C.D. (2012) A Change in the Geodynamics of Continental Growth 3 Billion Years Ago. Science 335, 1334–1336. https://doi.org/10.1126/science.1216066
) following some refinement (Korenaga, 2018Korenaga, J. (2018) Estimating the formation age distribution of continental crust by unmixing zircon ages. Earth and Planetary Science Letters 482, 388–395. https://doi.org/10.1016/j.epsl.2017.11.039
) and an updated zircon Hf dataset (Roberts and Spencer, 2015Roberts, N.M.W., Spencer, C.J. (2015) The zircon archive of continent formation through time. Geological Society, London, Special Publications 389, 197–225. https://doi.org/10.1144/SP389.14
) (green curve Fig. 1b). In contrast to approaches based solely on zircon, our new bulk rock-based continental growth curve shows no slowing of continental growth at 3.0 Ga but instead indicates the onset of significant continental growth at ∼3.5 Ga and reduction in crustal growth rate just before ∼1.0 Ga, almost 2.0 Ga later than the zircon-based method. It has been noted that zircon Hf-isotopes have a tendency to over-estimate the mass of reworking in the source of a given rock, so the curve presented in this work (Fig. 1) likely represents a maximum crustal growth curve, as decoupling of rock mass from Hf-isotope systematics would bias the curve to artificially old ages. This over-estimation may also be a source of offset between the crust and mantle growth curves in the Mesoarchean (Fig. 1a).The appearance of >10 % of continental crust at ca. 3.5 Ga coincides with a time on Earth when continental rocks were first preserved in significant volumes in the rock record. For instance, many cratonic nuclei contain rock samples that formed ∼3.4–3.6 Ga (Bauer et al., 2020
Bauer, A.B., Reimink, J.R., Chacko, T., Foley, B.J., Shirey, S.B., Pearson, D.G. (2020) Hafnium isotopes in zircons document the gradual onset of mobile-lid tectonics. Geochemical Perspectives Letters 14, 1–6. https://doi.org/10.7185/geochemlet.2015
). Thus, our calculations broadly agree with first order observations from the preserved rock record—an important test. Also, our new crustal growth curve indicates that continental growth has stagnated since ∼1.1 Ga, a time on Earth marked by the appearance of preserved paired metamorphic belts (Holder et al., 2019Holder, R.M., Viete, D.R., Brown, M., Johnson, T.E. (2019) Metamorphism and the evolution of plate tectonics. Nature 572, 378–381. https://doi.org/10.1038/s41586-019-1462-2
), possibly indicating the progressive evolution of plate tectonics to a modern style of colder and steeper subduction. This timing coincides with a dramatic slowing of new additions to cratonic mantle lithosphere (Pearson et al., 2021Pearson, D.G., Scott, J.M., Liu, J., Schaeffer, A., Wang, L.H., van Hunen, J., Szilas, K., Chacko, T., Kelemen, P.B. (2021) Deep continental roots and cratons. Nature 596, 199–210. https://doi.org/10.1038/s41586-021-03600-5
), pointing to a shared lineage between stable crust and mantle lithosphere. Note that we have performed key sensitivity tests on our modelling results (preservational bias, chemical biases, etc) that show that our crustal growth rate curves are immune to systematic biases (Figs. S7–9).top
Crustal Growth Rates
The continental growth rate reflected in global bulk rock data suggests a temporal relationship between the evolution of continental crust and the formation of deep, stable sub-continental lithospheric mantle roots that are key to defining the cratons. This relationship is very different to that previously proposed based on alternative continental growth curves (Dhuime et al., 2012
Dhuime, B., Hawkesworth, C.J., Cawood, P.A., Storey, C.D. (2012) A Change in the Geodynamics of Continental Growth 3 Billion Years Ago. Science 335, 1334–1336. https://doi.org/10.1126/science.1216066
; Korenaga, 2018Korenaga, J. (2018) Estimating the formation age distribution of continental crust by unmixing zircon ages. Earth and Planetary Science Letters 482, 388–395. https://doi.org/10.1016/j.epsl.2017.11.039
) which argue for a preservational relationship between continental crust and deep mantle roots (Hawkesworth et al., 2017Hawkesworth, C.J., Cawood, P.A., Dhuime, B., Kemp, T.I.S. (2017) Earth’s Continental Lithosphere Through Time. Annual Review of Earth and Planetary Sciences 45, 169–198. https://doi.org/10.1146/annurev-earth-063016-020525
; Pearson et al., 2021Pearson, D.G., Scott, J.M., Liu, J., Schaeffer, A., Wang, L.H., van Hunen, J., Szilas, K., Chacko, T., Kelemen, P.B. (2021) Deep continental roots and cratons. Nature 596, 199–210. https://doi.org/10.1038/s41586-021-03600-5
), whereby large volumes of deep mantle keels stabilised and preserved extant continental crust, whereas older, unstabilised continental crust was preferentially reworked. Instead, our bulk rock-based estimate of continental growth indicates that continental crust and deep mantle keels may have been formed in a similar time window, with the key inflection point being at ∼1 Ga, a defining point for cratons (Pearson et al., 2021Pearson, D.G., Scott, J.M., Liu, J., Schaeffer, A., Wang, L.H., van Hunen, J., Szilas, K., Chacko, T., Kelemen, P.B. (2021) Deep continental roots and cratons. Nature 596, 199–210. https://doi.org/10.1038/s41586-021-03600-5
). Continental crust begins to grow prior to the mantle lithosphere as recorded in mantle xenoliths on the modern Earth. The difference in growth rate could be due to mantle lithosphere being overprinted by younger magmatic events (Pearson et al., 2002Pearson, N.J., Alard, O., Griffin, W.L., Jackson, S.E., O’Reilly, S.Y. (2002) In situ measurement of Re-Os isotopes in mantle sulfides by laser ablation multicollector-inductively coupled plasma mass spectrometry: analytical methods and preliminary results. Geochimica et Cosmochimica Acta 66, 1037–1050. https://doi.org/10.1016/S0016-7037(01)00823-7
; Alard et al., 2005Alard, O., Luguet, A., Pearson, N.J., Griffin, W.L., Lorand, J.-P., Gannoun, A., Burton, K.W., O’Reilly, S.Y. (2005) In situ Os isotopes in abyssal peridotites bridge the isotopic gap between MORBs and their source mantle. Nature 436, 1005–1008. https://doi.org/10.1038/nature03902
; Liu et al., 2021Liu, J., Pearson, D.G., Wang, L.H., Mather, K.A., Kjarsgaard, B.A., Schaeffer, A.J., Irvine, G.J., Kopylova, M.G., Armstrong, J.P. (2021) Plume-driven recratonization of deep continental lithospheric mantle. Nature 592, 732–736. https://doi.org/10.1038/s41586-021-03395-5
). This may be likely as >2.85 Ga diamond-bearing lithosphere clearly exists in several cratonic regions (Smart et al., 2016Smart, K.A., Tappe, S., Stern, R.A., Webb, S.J., Ashwal, L.D. (2016) Early Archaean tectonics and mantle redox recorded in Witwatersrand diamonds. Nature Geoscience 9, 1–6. https://doi.org/10.1038/ngeo2628
; Timmerman et al., 2022Timmerman, S., Reimink, J.R., Vezinet, A., Nestola, F., Kublik, K., Banas, A., Stachel, T., Stern, R.A., Luo, Y., Sarkar, C. and Ielpi, A. (2022) Mesoarchean diamonds formed in thickened lithosphere, caused by slab-stacking. Earth and Planetary Science Letters 592, 117633. https://doi.org/10.1016/j.epsl.2022.117633
). Thus, we emphasise that any mechanism proposed to explain either the formation of ancient continental crust or their underlying deep mantle keels, features that collectively define the cratons (Pearson et al., 2021Pearson, D.G., Scott, J.M., Liu, J., Schaeffer, A., Wang, L.H., van Hunen, J., Szilas, K., Chacko, T., Kelemen, P.B. (2021) Deep continental roots and cratons. Nature 596, 199–210. https://doi.org/10.1038/s41586-021-03600-5
), must account for the formation of both features at nearly the same time (Pearson et al., 2007Pearson, D.G., Parman, S.W., Nowell, G.M. (2007) A link between large mantle melting events and continent growth seen in osmium isotopes. Nature 449, 202–205. https://doi.org/10.1038/nature06122
). The inflection in continental growth began near 3.5 Ga and has substantially slowed since ∼1.0 Ga (Fig. 1), a feature mirrored by continental roots (Pearson et al., 2021Pearson, D.G., Scott, J.M., Liu, J., Schaeffer, A., Wang, L.H., van Hunen, J., Szilas, K., Chacko, T., Kelemen, P.B. (2021) Deep continental roots and cratons. Nature 596, 199–210. https://doi.org/10.1038/s41586-021-03600-5
). Though felsic crust clearly cannot be derived directly from peridotite, this broad temporal link points to the possibility of mechanistic links in the formation of continental crust and the rapid docking of thick lithospheric keels beneath them, perhaps by lateral accretion and slab imbrication—a process that has been separately invoked for the production of ancient continental crust (Bauer et al., 2020Bauer, A.B., Reimink, J.R., Chacko, T., Foley, B.J., Shirey, S.B., Pearson, D.G. (2020) Hafnium isotopes in zircons document the gradual onset of mobile-lid tectonics. Geochemical Perspectives Letters 14, 1–6. https://doi.org/10.7185/geochemlet.2015
) and ancient lithospheric mantle (Timmerman et al., 2022Timmerman, S., Reimink, J.R., Vezinet, A., Nestola, F., Kublik, K., Banas, A., Stachel, T., Stern, R.A., Luo, Y., Sarkar, C. and Ielpi, A. (2022) Mesoarchean diamonds formed in thickened lithosphere, caused by slab-stacking. Earth and Planetary Science Letters 592, 117633. https://doi.org/10.1016/j.epsl.2022.117633
).Our continental growth curve indicates that Earth’s volumes of continental crust grew progressively over a 2.5 Gyr period in the middle of Earth history. There is no evidence for either large volumes of Hadean continental crust, nor signs of a decrease in crustal growth rate near 3.0 Ga, removing a key constraint used to argue for a geodynamic shift in Earth’s tectonic regime near that time. Instead, our analysis indicates that most continental crust grew between 3.5 Ga and ca. 1.0 Ga in a relatively consistent manner (Condie et al., 2018
Condie, K.C., Puetz, S.J., Davaille, A. (2018) Episodic crustal production before 2.7 Ga. Precambrian Research 312, 16–22. https://doi.org/10.1016/j.precamres.2018.05.005
; Garçon, 2021Garçon, M. (2021) Episodic growth of felsic continents in the past 3.7 Ga. Science Advances 7, eabj180. https://doi.org/10.1126/sciadv.abj1807
), occurring over the same time period that cratonic mantle roots formed (Pearson et al., 2021Pearson, D.G., Scott, J.M., Liu, J., Schaeffer, A., Wang, L.H., van Hunen, J., Szilas, K., Chacko, T., Kelemen, P.B. (2021) Deep continental roots and cratons. Nature 596, 199–210. https://doi.org/10.1038/s41586-021-03600-5
). Though the links between continental growth, craton root development, and the emergence of freeboard remain to be fully understood, our analysis suggests that they may be unrelated to a distinct change in the geodynamics of the solid Earth in the Neoarchean. Instead, our analysis places emphasis on the change in lithosphere evolution in the Mesoarchean, and may suggest that continental freeboard on Earth formed simply due to continent formation in large volumes (Reimink et al., 2021Reimink, J.R., Davies, J.H.F.L., Ielpi, A. (2021) Global zircon analysis records a gradual rise of continental crust throughout the Neoarchean. Earth and Planetary Science Letters 554, 116654. https://doi.org/10.1016/j.epsl.2020.116654
). Thus, continental emergence and the rise of subaerial weathering cycles may have been caused simply by the formation and stabilisation of the continents themselves.top
Data Availability
The data reduction code used to process this data set can be found in the Supplementary Information. Zircon U–Pb data shown in Figure 1 is from Puetz and Condie
Puetz, S.J., Condie, K.C. (2019) Time series analysis of mantle cycles Part I: Periodicities and correlations among seven global isotopic databases. Geoscience Frontiers 10, 1305–1326. https://doi.org/10.1016/j.gsf.2019.04.002
, (2019Puetz, S.J., Condie, K.C. (2019) Time series analysis of mantle cycles Part I: Periodicities and correlations among seven global isotopic databases. Geoscience Frontiers 10, 1305–1326. https://doi.org/10.1016/j.gsf.2019.04.002
).top
Acknowledgements
Roberta Rudnick and an anonymous reviewer are thanked for their comments that led to substantial improvements in this manuscript. Ambre Luguet and Raúl Fonseca are thanked for their comments and editorial handling in the production of this publication.
Editor: Ambre Luguet and Raúl Fonseca
top
References
Alard, O., Luguet, A., Pearson, N.J., Griffin, W.L., Lorand, J.-P., Gannoun, A., Burton, K.W., O’Reilly, S.Y. (2005) In situ Os isotopes in abyssal peridotites bridge the isotopic gap between MORBs and their source mantle. Nature 436, 1005–1008. https://doi.org/10.1038/nature03902
Show in context The difference in growth rate could be due to mantle lithosphere being overprinted by younger magmatic events (Pearson et al., 2002; Alard et al., 2005; Liu et al., 2021).
View in article
Bauer, A.B., Reimink, J.R., Chacko, T., Foley, B.J., Shirey, S.B., Pearson, D.G. (2020) Hafnium isotopes in zircons document the gradual onset of mobile-lid tectonics. Geochemical Perspectives Letters 14, 1–6. https://doi.org/10.7185/geochemlet.2015
Show in context For instance, many cratonic nuclei contain rock samples that formed ∼3.4–3.6 Ga (Bauer et al., 2020).
View in article
Though felsic crust clearly cannot be derived directly from peridotite, this broad temporal link points to the possibility of mechanistic links in the formation of continental crust and the rapid docking of thick lithospheric keels beneath them, perhaps by lateral accretion and slab imbrication—a process that has been separately invoked for the production of ancient continental crust (Bauer et al., 2020) and ancient lithospheric mantle (Timmerman et al., 2022).
View in article
Belousova, E.A., Kostitsyn, Y.A., Griffin, W.L., Begg, G.C., O’Reilly, S.Y., Pearson, N.J. (2010) The growth of the continental crust: Constraints from zircon Hf-isotope data. Lithos 119, 457–466. https://doi.org/10.1016/j.lithos.2010.07.024
Show in context Many widely used models, some of which are underpinned by zircon U–Pb and Hf isotope systematics (Belousova et al., 2010; Dhuime et al., 2012), emphasize the potential importance of voluminous continental growth in the Archean.
View in article
Recent estimates of the volumes of continental crust throughout Earth’s history have relied on zircon Hf and O isotopes (Belousova et al., 2010; Dhuime et al., 2012; Korenaga, 2018).
View in article
Bindeman, I.N., Zakharov, D.O., Palandri, J., Greber, N.D., Dauphas, N., Retallack, G.J., Hofmann, A., Lackey, J.S., Bekker, A. (2018) Rapid emergence of subaerial landmasses and onset of a modern hydrologic cycle 2.5 billion years ago. Nature 557, 545–548. https://doi.org/10.1038/s41586-018-0131-1
Show in context Additionally, the maximum oxygen isotope ratio of igneous zircons and shales (mature sedimentary rocks) continues to increase over time (Valley et al., 2005; Bindeman et al., 2018).
View in article
Bucholz, C.E., Spencer, C.J. (2019) Strongly Peraluminous Granites across the Archean–Proterozoic Transition. Journal of Petrology 60, 1299–1348. https://doi.org/10.1093/petrology/egz033
Show in context For instance, recent studies have shown that zircon oxygen isotope ratios do not accurately identify sediment recycling in the Neoarchean (Bucholz and Spencer, 2019), a critical time period for constraining continental growth estimates.
View in article
The causes of this difference are not readily apparent, but may be due to anoxic weathering conditions that adversely affect the ability of oxygen isotope ratios to accurately track continental reworking (Bucholz and Spencer, 2019).
View in article
Cawood, P.A., Hawkesworth, C.J., Dhuime, B. (2013) The continental record and the generation of continental crust. Geological Society of America Bulletin 125, 14–32. https://doi.org/10.1130/B30722.1
Show in context Destruction of an ancient crustal record can take several forms, but is typically separated into two categories (Cawood et al., 2013): 1) Reworking—processes that overprint the radiometric ages of the crustal record, but do not remove mass from the continents, such as partial melting and sedimentary erosion, and 2) Recycling—full scale removal of continental mass back into the mantle by delamination or sediment and continental subduction.
View in article
Coltice, N., Albarède, F., Gillet, P. (2000) 40K-40Ar Constraints on Recycling Continental Crust into the Mantle. Science 288, 845–847. https://doi.org/10.1126/science.288.5467.845
Show in context For instance, volatile-element isotopic measurements typically indicate the onset of detectable crustal recycling near ∼2.5 Ga (Coltice et al., 2000; Parai and Mukhopadhyay, 2018; Labidi et al., 2020).
View in article
Condie, K.C., Puetz, S.J., Davaille, A. (2018) Episodic crustal production before 2.7 Ga. Precambrian Research 312, 16–22. https://doi.org/10.1016/j.precamres.2018.05.005
Show in context Instead, our analysis indicates that most continental crust grew between 3.5 Ga and ca. 1.0 Ga in a relatively consistent manner (Condie et al., 2018; Garçon, 2021), occurring over the same time period that cratonic mantle roots formed (Pearson et al., 2021).
View in article
Dhuime, B., Hawkesworth, C.J., Cawood, P.A., Storey, C.D. (2012) A Change in the Geodynamics of Continental Growth 3 Billion Years Ago. Science 335, 1334–1336. https://doi.org/10.1126/science.1216066
Show in context Many widely used models, some of which are underpinned by zircon U–Pb and Hf isotope systematics (Belousova et al., 2010; Dhuime et al., 2012), emphasize the potential importance of voluminous continental growth in the Archean.
View in article
Recent estimates of the volumes of continental crust throughout Earth’s history have relied on zircon Hf and O isotopes (Belousova et al., 2010; Dhuime et al., 2012; Korenaga, 2018).
View in article
Hafnium isotopes in zircon can be used to both calculate the age of crust formation and the time the source material was extracted from the mantle, while oxygen isotope ratios have been used as a filter (Dhuime et al., 2012).
View in article
Zircon U–Pb, Hf, and O isotope data have been combined to calculate widely used crustal growth estimates (Dhuime et al., 2012) that indicate rapid crustal growth in the Archean and a shift in crustal growth rate near 3.0 Ga to slower growth. Despite the popularity of this approach, drawbacks have been pointed out (Korenaga, 2018).
View in article
We leverage our improved reworking metric to calculate a continental growth rate curve (Fig. 1; see supplementary methods for details) starting from a widely used crustal growth estimate based on zircon U–Pb & Hf isotopes (Dhuime et al., 2012) following some refinement (Korenaga, 2018) and an updated zircon Hf dataset (Roberts and Spencer, 2015) (green curve Fig. 1b).
View in article
This relationship is very different to that previously proposed based on alternative continental growth curves (Dhuime et al., 2012; Korenaga, 2018) which argue for a preservational relationship between continental crust and deep mantle roots (Hawkesworth et al., 2017; Pearson et al., 2021), whereby large volumes of deep mantle keels stabilised and preserved extant continental crust, whereas older, unstabilised continental crust was preferentially reworked.
View in article
Donaldson, J.A., de Kemp, E.A. (1998) Archaean quartz arenites in the Canadian Shield: examples from the Superior and Churchill Provinces. Sedimentary Geology 120, 153–176. https://doi.org/10.1016/S0037-0738(98)00031-1
Show in context Many sedimentary rocks and their derivative melts are known from the Neoarchean period (Donaldson and de Kemp, 1998; Laurent et al., 2014), indicating that continental reworking took place; yet, the existence of significant crustal reworking is not clearly captured by the zircon oxygen isotope record.
View in article
Frost, B.R., Frost, C.D. (2008) A Geochemical Classification for Feldspathic Igneous Rocks. Journal of Petrology 49, 1955–1969. https://doi.org/10.1093/petrology/egn054
Show in context We adopt this approach because the major element composition of igneous rocks can accurately quantify the extent of reworking of previous continental crust, whether reworking occurs via incorporation of sediments or direct melting of pre-existing continental crust (Frost and Frost, 2008; Moyen et al., 2017).
View in article
Garçon, M. (2021) Episodic growth of felsic continents in the past 3.7 Ga. Science Advances 7, eabj180. https://doi.org/10.1126/sciadv.abj1807
Show in context Instead, our analysis indicates that most continental crust grew between 3.5 Ga and ca. 1.0 Ga in a relatively consistent manner (Condie et al., 2018; Garçon, 2021), occurring over the same time period that cratonic mantle roots formed (Pearson et al., 2021).
View in article
Hawkesworth, C.J., Cawood, P.A., Dhuime, B., Kemp, T.I.S. (2017) Earth’s Continental Lithosphere Through Time. Annual Review of Earth and Planetary Sciences 45, 169–198. https://doi.org/10.1146/annurev-earth-063016-020525
Show in context Recent studies of the continental crust-CLMR relationship suggest that, while very ancient continental crust was formed and rapidly destroyed, the formation of lithospheric roots stabilised existing continental crust, thereby slowing the destruction, and growth rate, of continents (Hawkesworth et al., 2017; Pearson et al., 2021).
View in article
This relationship is very different to that previously proposed based on alternative continental growth curves (Dhuime et al., 2012; Korenaga, 2018) which argue for a preservational relationship between continental crust and deep mantle roots (Hawkesworth et al., 2017; Pearson et al., 2021), whereby large volumes of deep mantle keels stabilised and preserved extant continental crust, whereas older, unstabilised continental crust was preferentially reworked.
View in article
Holder, R.M., Viete, D.R., Brown, M., Johnson, T.E. (2019) Metamorphism and the evolution of plate tectonics. Nature 572, 378–381. https://doi.org/10.1038/s41586-019-1462-2
Show in context Also, our new crustal growth curve indicates that continental growth has stagnated since ∼1.1 Ga, a time on Earth marked by the appearance of preserved paired metamorphic belts (Holder et al., 2019), possibly indicating the progressive evolution of plate tectonics to a modern style of colder and steeper subduction.
View in article
Jackson, M.G., Hart, S.R., Koppers, A.A.P., Staudigel, H., Konter, J., Blusztajn, J., Kurz, M., Russell, J.A. (2007) The return of subducted continental crust in Samoan lavas. Nature 448, 684–687. https://doi.org/10.1038/nature06048
Show in context While there is isotopic evidence for some amount of continental crust recycled into the modern mantle (Jackson et al., 2007), there is little evidence for vast volumes of ancient continental material residing in the mantle.
View in article
Korenaga, J. (2018) Estimating the formation age distribution of continental crust by unmixing zircon ages. Earth and Planetary Science Letters 482, 388–395. https://doi.org/10.1016/j.epsl.2017.11.039
Show in context Recent estimates of the volumes of continental crust throughout Earth’s history have relied on zircon Hf and O isotopes (Belousova et al., 2010; Dhuime et al., 2012; Korenaga, 2018).
View in article
Zircon U–Pb, Hf, and O isotope data have been combined to calculate widely used crustal growth estimates (Dhuime et al., 2012) that indicate rapid crustal growth in the Archean and a shift in crustal growth rate near 3.0 Ga to slower growth. Despite the popularity of this approach, drawbacks have been pointed out (Korenaga, 2018).
View in article
While the combination of the U–Pb and Hf isotope systems may be reliable (Korenaga, 2018), the use of oxygen isotopes to track recycling is prone to uncertainty.
View in article
This relationship is very different to that previously proposed based on alternative continental growth curves (Dhuime et al., 2012; Korenaga, 2018) which argue for a preservational relationship between continental crust and deep mantle roots (Hawkesworth et al., 2017; Pearson et al., 2021), whereby large volumes of deep mantle keels stabilised and preserved extant continental crust, whereas older, unstabilised continental crust was preferentially reworked.
View in article
Labidi, J., Barry, P.H., Bekaert, D.V., Broadley, M.W., Marty, B., Giunta, T., Warr, O., Sherwood Lollar, B., Fischer, T.P., Avice, G. and Caracausi, A. (2020) Hydrothermal 15 N 15 N abundances constrain the origins of mantle nitrogen. Nature 580, 367–371. https://doi.org/10.1038/s41586-020-2173-4
Show in context For instance, volatile-element isotopic measurements typically indicate the onset of detectable crustal recycling near ∼2.5 Ga (Coltice et al., 2000; Parai and Mukhopadhyay, 2018; Labidi et al., 2020).
View in article
Laurent, O., Martin, H., Martin, H., Moyen, J.F., Moyen, J.-F., Doucelance, R., Doucelance, R. (2014) The diversity and evolution of late-Archean granitoids: Evidence for the onset of “modern-style” plate tectonics between 3.0 and 2.5Ga. Lithos 205, 1–82. https://doi.org/10.1016/j.lithos.2014.06.012
Show in context Many sedimentary rocks and their derivative melts are known from the Neoarchean period (Donaldson and de Kemp, 1998; Laurent et al., 2014), indicating that continental reworking took place; yet, the existence of significant crustal reworking is not clearly captured by the zircon oxygen isotope record.
View in article
Lee, C.-T.A., Caves, J., Jiang, H., Cao, W., Lenardic, A., McKenzie, N.R., Shorttle, O., Yin, Q., Dyer, B. (2017) Deep mantle roots and continental emergence: implications for whole-Earth elemental cycling, long-term climate, and the Cambrian explosion. International Geology Review 60, 1–18. https://doi.org/10.1080/00206814.2017.1340853
Show in context In the modern Earth, deep lithospheric roots stabilise and appear to preserve extant continental crust (Lee et al., 2017), so the discrepancy between preserved lithospheric mantle ages and previous crustal growth models (Fig. 1) suggests a genetic disconnect.
View in article
Liu, J., Pearson, D.G., Wang, L.H., Mather, K.A., Kjarsgaard, B.A., Schaeffer, A.J., Irvine, G.J., Kopylova, M.G., Armstrong, J.P. (2021) Plume-driven recratonization of deep continental lithospheric mantle. Nature 592, 732–736. https://doi.org/10.1038/s41586-021-03395-5
Show in context The difference in growth rate could be due to mantle lithosphere being overprinted by younger magmatic events (Pearson et al., 2002; Alard et al., 2005; Liu et al., 2021).
View in article
Moyen, J.-F., Laurent, O., Chelle-Michou, C., Couzinié, S., Vanderhaeghe, O., Zeh, A., Villaros, A., Gardien, V. (2017) Collision vs. subduction-related magmatism: Two contrasting ways of granite formation and implications for crustal growth. Lithos 277, 154–177. https://doi.org/10.1016/j.lithos.2016.09.018
Show in context We adopt this approach because the major element composition of igneous rocks can accurately quantify the extent of reworking of previous continental crust, whether reworking occurs via incorporation of sediments or direct melting of pre-existing continental crust (Frost and Frost, 2008; Moyen et al., 2017).
View in article
Our estimate of the reworking rate of continental crust (Fig. 3) uses input from the classic ACNK/ANK diagram of Shand, (1943), further developed to isolate source composition from fractionation and assimilation trends in magmatic rocks by Moyen et al. (2017).
View in article
Parai, R., Mukhopadhyay, S. (2018) Xenon isotopic constraints on the history of volatile recycling into the mantle. Nature 560, 223–227. https://doi.org/10.1038/s41586-018-0388-4
Show in context For instance, volatile-element isotopic measurements typically indicate the onset of detectable crustal recycling near ∼2.5 Ga (Coltice et al., 2000; Parai and Mukhopadhyay, 2018; Labidi et al., 2020).
View in article
Pearson, D.G., Parman, S.W., Nowell, G.M. (2007) A link between large mantle melting events and continent growth seen in osmium isotopes. Nature 449, 202–205. https://doi.org/10.1038/nature06122
Show in context Thus, we emphasise that any mechanism proposed to explain either the formation of ancient continental crust or their underlying deep mantle keels, features that collectively define the cratons (Pearson et al., 2021), must account for the formation of both features at nearly the same time (Pearson et al., 2007).
View in article
Pearson, D.G., Scott, J.M., Liu, J., Schaeffer, A., Wang, L.H., van Hunen, J., Szilas, K., Chacko, T., Kelemen, P.B. (2021) Deep continental roots and cratons. Nature 596, 199–210. https://doi.org/10.1038/s41586-021-03600-5
Show in context Likewise, the continental lithospheric mantle—thick roots that stabilised ancient continental crust—have little extensive record prior to ∼3.0 Ga (Pearson et al., 2021).
View in article
Recent studies of the continental crust-CLMR relationship suggest that, while very ancient continental crust was formed and rapidly destroyed, the formation of lithospheric roots stabilised existing continental crust, thereby slowing the destruction, and growth rate, of continents (Hawkesworth et al., 2017; Pearson et al., 2021).
View in article
The black curve shows the age distribution of unmodified cratonic mantle ages (Pearson et al., 2021), while the green field shows the percentage of preserved mature sedimentary packages (Reimink et al., 2021).
View in article
The black curve is the age distribution of unmodified cratonic mantle roots (Pearson et al., 2021).
View in article
This timing coincides with a dramatic slowing of new additions to cratonic mantle lithosphere (Pearson et al., 2021), pointing to a shared lineage between stable crust and mantle lithosphere.
View in article
This relationship is very different to that previously proposed based on alternative continental growth curves (Dhuime et al., 2012; Korenaga, 2018) which argue for a preservational relationship between continental crust and deep mantle roots (Hawkesworth et al., 2017; Pearson et al., 2021), whereby large volumes of deep mantle keels stabilised and preserved extant continental crust, whereas older, unstabilised continental crust was preferentially reworked.
View in article
Instead, our bulk rock-based estimate of continental growth indicates that continental crust and deep mantle keels may have been formed in a similar time window, with the key inflection point being at ∼1 Ga, a defining point for cratons (Pearson et al., 2021).
View in article
Thus, we emphasise that any mechanism proposed to explain either the formation of ancient continental crust or their underlying deep mantle keels, features that collectively define the cratons (Pearson et al., 2021), must account for the formation of both features at nearly the same time (Pearson et al., 2007).
View in article
The inflection in continental growth began near 3.5 Ga and has substantially slowed since ∼1.0 Ga (Fig. 1), a feature mirrored by continental roots (Pearson et al., 2021).
View in article
Instead, our analysis indicates that most continental crust grew between 3.5 Ga and ca. 1.0 Ga in a relatively consistent manner (Condie et al., 2018; Garçon, 2021), occurring over the same time period that cratonic mantle roots formed (Pearson et al., 2021).
View in article
Pearson, N.J., Alard, O., Griffin, W.L., Jackson, S.E., O’Reilly, S.Y. (2002) In situ measurement of Re-Os isotopes in mantle sulfides by laser ablation multicollector-inductively coupled plasma mass spectrometry: analytical methods and preliminary results. Geochimica et Cosmochimica Acta 66, 1037–1050. https://doi.org/10.1016/S0016-7037(01)00823-7
Show in context The difference in growth rate could be due to mantle lithosphere being overprinted by younger magmatic events (Pearson et al., 2002; Alard et al., 2005; Liu et al., 2021).
View in article
Puetz, S.J., Condie, K.C. (2019) Time series analysis of mantle cycles Part I: Periodicities and correlations among seven global isotopic databases. Geoscience Frontiers 10, 1305–1326. https://doi.org/10.1016/j.gsf.2019.04.002
Show in context Zircon U‒Pb data shown in Figure 1 is from Puetz and Condie, (2019).
View in article
Reimink, J.R., Davies, J.H.F.L., Ielpi, A. (2021) Global zircon analysis records a gradual rise of continental crust throughout the Neoarchean. Earth and Planetary Science Letters 554, 116654. https://doi.org/10.1016/j.epsl.2020.116654
Show in context The black curve shows the age distribution of unmodified cratonic mantle ages (Pearson et al., 2021), while the green field shows the percentage of preserved mature sedimentary packages (Reimink et al., 2021).
View in article
The angular projection employed in our “reworking index”, based on major elements, can identify the recycling of continental igneous rocks, as well as the products of melting of sedimentary rocks—a phenomenon that zircon oxygen isotope ratios are only well-suited to detect in the post-NeoArchean rock record (Fig. 2) after large volumes of sedimentary rocks began to be deposited following continental emergence (Valley et al., 2005; Reimink et al., 2021).
View in article
Instead, our analysis places emphasis on the change in lithosphere evolution in the Mesoarchean, and may suggest that continental freeboard on Earth formed simply due to continent formation in large volumes (Reimink et al., 2021).
View in article
Roberts, N.M.W., Spencer, C.J. (2015) The zircon archive of continent formation through time. Geological Society, London, Special Publications 389, 197–225. https://doi.org/10.1144/SP389.14
Show in context We leverage our improved reworking metric to calculate a continental growth rate curve (Fig. 1; see supplementary methods for details) starting from a widely used crustal growth estimate based on zircon U–Pb & Hf isotopes (Dhuime et al., 2012) following some refinement (Korenaga, 2018) and an updated zircon Hf dataset (Roberts and Spencer, 2015) (green curve Fig. 1b).
View in article
Shand, S.J. (1943) Eruptive rocks: their genesis, composition, and classification, with a chapter on meteorites. J. Wiley & sons, Incorporated.
Show in context Our estimate of the reworking rate of continental crust (Fig. 3) uses input from the classic ACNK/ANK diagram of Shand, (1943), further developed to isolate source composition from fractionation and assimilation trends in magmatic rocks by Moyen et al. (2017).
View in article
Smart, K.A., Tappe, S., Stern, R.A., Webb, S.J., Ashwal, L.D. (2016) Early Archaean tectonics and mantle redox recorded in Witwatersrand diamonds. Nature Geoscience 9, 1–6. https://doi.org/10.1038/ngeo2628
Show in context This may be likely as >2.85 Ga diamond-bearing lithosphere clearly exists in several cratonic regions (Smart et al., 2016; Timmerman et al., 2022).
View in article
Timmerman, S., Reimink, J.R., Vezinet, A., Nestola, F., Kublik, K., Banas, A., Stachel, T., Stern, R.A., Luo, Y., Sarkar, C. and Ielpi, A. (2022) Mesoarchean diamonds formed in thickened lithosphere, caused by slab-stacking. Earth and Planetary Science Letters 592, 117633. https://doi.org/10.1016/j.epsl.2022.117633
Show in context This may be likely as >2.85 Ga diamond-bearing lithosphere clearly exists in several cratonic regions (Smart et al., 2016; Timmerman et al., 2022).
View in article
Though felsic crust clearly cannot be derived directly from peridotite, this broad temporal link points to the possibility of mechanistic links in the formation of continental crust and the rapid docking of thick lithospheric keels beneath them, perhaps by lateral accretion and slab imbrication—a process that has been separately invoked for the production of ancient continental crust (Bauer et al., 2020) and ancient lithospheric mantle (Timmerman et al., 2022).
View in article
Valley, J.W., Lackey, J.S., Cavosie, A.J., Clechenko, C.C., Spicuzza, M.J., Basei, M.A.S., Bindeman, I.N., Ferreira, V.P., Sial, A.N., King, E.M., Peck, W.H., Sinha, A.K. and Wei, C.S. (2005) 4.4 billion years of crustal maturation: oxygen isotope ratios of magmatic zircon. Contributions to Mineralogy and Petrology 150, 561–580. https://doi.org/10.1007/s00410-005-0025-8
Show in context Such incorporation of sedimentary rocks into the igneous system, e.g., continental recycling, can generate zircon with high oxygen isotope ratios (Valley et al., 2005), at least on the modern Earth.
View in article
Additionally, the maximum oxygen isotope ratio of igneous zircons and shales (mature sedimentary rocks) continues to increase over time (Valley et al., 2005; Bindeman et al., 2018).
View in article
The angular projection employed in our “reworking index”, based on major elements, can identify the recycling of continental igneous rocks, as well as the products of melting of sedimentary rocks—a phenomenon that zircon oxygen isotope ratios are only well-suited to detect in the post-NeoArchean rock record (Fig. 2) after large volumes of sedimentary rocks began to be deposited following continental emergence (Valley et al., 2005; Reimink et al., 2021).
View in article
top
Supplementary Information
The Supplementary Information includes:
Download Tables S-1 to S-4 (xlsx)
Download the Supplementary Information (PDF)
Figures
Figure 1 (a) Various crustal growth rate curves shown by dashed lines compared to the cratonic mantle age distribution. Vertical coloured arrows show the amount of continental recycling predicted by various estimates. The black curve shows the age distribution of unmodified cratonic mantle ages (Pearson et al., 2021
Pearson, D.G., Scott, J.M., Liu, J., Schaeffer, A., Wang, L.H., van Hunen, J., Szilas, K., Chacko, T., Kelemen, P.B. (2021) Deep continental roots and cratons. Nature 596, 199–210. https://doi.org/10.1038/s41586-021-03600-5
), while the green field shows the percentage of preserved mature sedimentary packages (Reimink et al., 2021Reimink, J.R., Davies, J.H.F.L., Ielpi, A. (2021) Global zircon analysis records a gradual rise of continental crust throughout the Neoarchean. Earth and Planetary Science Letters 554, 116654. https://doi.org/10.1016/j.epsl.2020.116654
). (b) Shows the continental growth curve predicted in this work. The grey curve shows U–Pb ages of the preserved rock record (Puetz et al., 2018). The black curve is the age distribution of unmodified cratonic mantle roots (Pearson et al., 2021Pearson, D.G., Scott, J.M., Liu, J., Schaeffer, A., Wang, L.H., van Hunen, J., Szilas, K., Chacko, T., Kelemen, P.B. (2021) Deep continental roots and cratons. Nature 596, 199–210. https://doi.org/10.1038/s41586-021-03600-5
). The green curve is the crustal growth curve calculated here using bulk rock major element chemistry. The dashed green curve uses the same calculations as the solid green curve except for the utilisation of a synthetic Mesoarchean–Hadean rock record, to evaluate potential for sampling bias.
Figure 2 The distribution of zircon oxygen isotope compositions through time, highlighting the tendency of zircon oxygen isotope ratios to become progressively more extreme through time, making them a poor discriminant of continental reworking. The orange band shows a typical field for ‘juvenile’ zircon isotope compositions.
Figure 3 The fraction of rocks classified as reworked continental crust from Figure 1 using the bulk geochemistry approach adopted here (green curve with grey band showing 2SE uncertainty) compared to the reworking estimates based on zircon oxygen isotope ratios. Curves calculated in 200 Ma moving windows in 10 Ma time steps. The zircon record dramatically under-estimates reworking in the Archean.
Figure 4 Reworking through time as seen through the whole-rock record. (a) The whole-rock geochemical data used to calculate a ‘theta’ value, with experimental melts shown coloured by starting composition. (b) Our classifier for data shown in panel b, showing the probability density for each category across a range of Theta values. (c) The distribution of Theta values across geological time. A probability density estimator is shown for the theta values for rocks split into 100 Ma age bins and coloured according to the number of rock samples in each age bin. Rocks that likely represent reworked crust fall into the blue field and have been present since the early Archean.