Continental growth seen through the sedimentary record

Abstract

International audienceSedimentary rocks and detrital minerals sample large areas of the continental crust, and they are increasingly seen as a reliable archive for its global evolution. This study presents two approaches to model the growth of the continental crust through the sedimentary archive. The first builds on the variations in U-Pb, Hf and O isotopes in global databases of detrital zircons. We show that uncertainty in the Hf isotope composition of the mantle reservoir from which new crust separated, in the 176Lu/177Hf ratio of that new crust, and in the contribution in the databases of zircons that experienced ancient Pb loss(es), adds some uncertainty to the individual Hf model ages, but not to the overall shape of the calculated continental growth curves. The second approach is based on the variation of Nd isotopes in 645 worldwide fine-grained continental sedimentary rocks with different deposition ages, which requires a correction of the bias induced by preferential erosion of younger rocks through an erosion parameter referred to as K. This dimensionless parameter relates the proportions of younger to older source rocks in the sediment, to the proportions of younger to older source rocks present in the crust from which the sediment was derived. We suggest that a Hadean/Archaean value of K = 1 (i.e., no preferential erosion), and that post-Archaean values of K = 4–6, may be reasonable for the global Earth system. Models built on the detrital zircon and the fine-grained sediment records independently suggest that at least 65% of the present volume of continental crust was established by 3 Ga. The continental crust has been generated continuously, but with a marked decrease in the growth rate at ~ 3 Ga. The period from > 4 Ga to ~ 3 Ga is characterised by relatively high net rates of continental growth (2.9–3.4 km3 yr− 1 on average), which are similar to the rates at which new crust is generated (and destroyed) at the present time. Net growth rates are much lower since 3 Ga (0.6–0.9 km3 yr− 1 on average), which can be attributed to higher rates of destruction of continental crust. The change in slope in the continental growth curve at ~ 3 Ga is taken to indicate a global change in the way bulk crust was generated and preserved, and this change has been linked to the onset of subduction-driven plate tectonics. At least 100% of the present volume of the continental crust has been destroyed and recycled back into the mantle since ~ 3 Ga, and this time marks a transition in the average composition of new continental crust. Continental crust generated before 3 Ga was on average mafic, dense, relatively thin (< 20 km) and therefore different from the calc-alkaline andesitic crust that dominates the continental record today. Continental crust that formed after 3 Ga gradually became more intermediate in composition, buoyant and thicker. The increase in crustal thickness is accompanied by increasing rates of crustal reworking and increasing input of sediment to the ocean. These changes may have been accommodated...

Growing primordial continental crust self-consistently in global mantle convection models

“…In a recent review on continental growth, Dhuime et al (2017) proposed that 65% of the present continental crust existed before 3 Ga. This is supported with similar results from different continental growth models built on records of detrital zircons and sedimentary rocks.…”

“…These models fall into two competing camps based on the nature of crustal growth: continuous growth with differing growth rates through Earth history (e.g., Hurley and Rand, 1969;Armstrong, 1981;Allègre and Rousseau, 1984;Taylor and McLennan, 1985;Armstrong, 1991;Taylor and McLennan, 1996;Belousova et al, 2010;Dhuime et al, 2012); versus episodic growth corresponding to supercontinent cycles or mantle plume activity (e.g., McCulloch and Bennett, 1994;Condie, 1998Condie, , 2000Condie, , 2004Rino et al, 2004;Campbell and Allen, 2008;Voice et al, 2011). Using growth models built on records of detrital zircons and sedimentary rocks, which may predominantly characterise the eroded continental crust emerged above the sea level (Flament et al, 2013), Dhuime et al (2017) proposed that 65% of the present continental crust existed by 3 Ga. They argued that there has been a continuous growth of continental crust throughout the evolution of the planet with a significant drop in net production rate from 2.9 -3.4 km 3 yr −1 on average to 0.6 -0.7 km 3 yr −1 on average at around ∼ 3 Ga. Table 4.…”

Secular change in TTG compositions: Implications for the evolution of Archaean geodynamics

“…On a global scale, statistical analysis of the chemical composition of continental igneous rocks with ages spanning 3.8 Ga to the present day reveals patterns consistent with secular cooling of the mantle since the late Archaean (Keller and Schoene, 2012). The zircon record and the median Rb/Sr ratios and SiO 2 contents of crustal rocks have been attributed to the widespread appearance of subduction at around 3.0 Ga (Dhuime et al, 2012(Dhuime et al, , 2015(Dhuime et al, , 2017Tang et al, 2016). This is consistent with the widespread survival of metamorphic rocks with contrasting thermal gradients since the Neoarchean that may indicate the beginning of subduction and consequent collisional orogenesis at that time, although there is little evidence before the Neoproterozoic for the cold collision and deep subduction of continental crust that characterises the modern plate tectonic regime (Brown and Johnson, 2018).…”

“…A growing body of petrological and geochemical (including isotopic) evidence from diverse datasets has been interpreted to indicate the widespread appearance of subduction in the Mesoarchean, ultimately leading to stabilisation of the lithosphere and a globally linked network of plate boundaries (i.e. plate tectonics) by the end of the Archaean (Bercovici and Ricard, 2014;Brown and Johnson, 2018;Dhuime et al, 2012Dhuime et al, , 2015Dhuime et al, , 2017Griffin et al, 2014;Hawkesworth and Brown, 2018;Moyen and Laurent, 2017;Naeraa et al, 2012;Van Kranendonk et al, 2015b). For the TTG dataset discussed here, we have emphasised a statistical approach that has identified clearly the importance of some fundamental change across the interval 3.3-3.0 Ga, which is particularly clear in the change point analysis (Fig.…”

Implications of erosion and bedrock composition on zircon fertility: Examples from South America and Western Australia

“…Detrital zircon age spectra have been used to address a variety of questions relating to crustal growth/evolution (Belousova et al., ; Dhuime, Hawkesworth, Cawood, & Storey, ; Dhuime, Hawkesworth, Delavault, & Cawood, ) and various strategies have been employed to optimize their representative nature. Such strategies include high‐density sampling within a single sedimentary structure (Lawrence, Cox, Mapes, & Coleman, ), sampling modern river mouths and comparing this to the known local basement (Bradley, O'Sullivan, & Bradley, ), sampling throughout a well‐defined stratigraphic succession (Ireland, Flöttmann, Fanning, Gibson, & Preiss, ) and data backfilling through geographic resampling (McKenzie et al., ; Puetz et al., ).…”