Ediacaran Outgassing – Evidence for a Spike in Outgassing of Carbon from the Mantle in the Ediacaran

Australia: The Land Where Time Began

It is believed that changes of the concentration of CO₂ in the atmosphere are the primary controller of long-term cycles of the climate of the Earth. An imbalance in the carbon cycle can be caused by changes in carbon emissions from volcanic activity. Proxies such as age abundance of detrital zircons have led to volcanic activity on a large scale being inferred, though the magnitude of carbon emissions depends on style of the volcanism as well as the amount of those emissions. This study compared the results of U-Pb analysis and trace element data of detrital zircons from Antarctica with the global rock record. A spike in CO₂-carbonatite and alkaline magmatism dating to the Ediacaran was identified. Secular cooling of the mantle as well as the advent of cooler subduction regimes prior to the Ediacaran promoted the sequestration of carbon that had been derived from decarbonation of oceanic slabs that were subducting in the mantle. Paulsen et al. inferred that the extensive release of carbon that resulted by subsequent magmatism that may be recorded, at least in part, in the Shuram-Wonoka carbon isotope excursion. Paulsen et al. therefore suggest that this pulse of alkaline volcanism is a reflection of profound reorganisation of deep and surface carbon cycles in the Neoproterozoic and promoted warming of the Earth prior to the Cambrian radiation.

The mantle holds the largest reservoir of carbon on Earth, but the way in which it has evolved over the history of the Earth and the way it influenced the global climate has remained a fundamental problem (Dasgupta, 2013) that has not been resolved. There have been times when, according to standard models, the exogenic carbon inputs (igneous and metamorphic degassing) outweighed the outputs (silicate weathering, burial sequestration and subduction), which resulted in global warming and rising sea levels (Berner, 1990). During the Neoproterozoic the balance of exogenic atmospheric inputs and outputs is of particular interest because this was a period that was marked by major perturbations in the global carbon cycle that have been hypothesised to be connected to changes in climate, atmospheric-oceanic oxygenation and biodiversification prior to the Cambrian explosion of life (Hoffman, Kaufman, Halverson & Schrag, 1998; Fike et al., 2006). It seems the partial pressure of CO₂ (P_CO2) rose from levels that were relatively low as the Earth transitioned from the Snowball Earth state in the Cryogenian, about 720-635 Ma, to a maximum level in the Phanerozoic during its greenhouse state in the Cambrian, though details of the partial pressure of atmospheric CO₂ throughout the Neoproterozoic are not well understood.

A worldwide detrital zircon database that has been growing was recently analysed and the results have led to the conclusion that the transitions of the Late Neoproterozoic-Phanerozoic from ice house to greenhouse conditions of the global climate with increases in the abundance of zircon (McKenzie et al., 2016). Therefore, the global warming in the long-term, has been linked to the idea that the increases in the partial pressure of CO₂ in the atmosphere are governed by widespread expansion of continental arc magmatism (McKenzie et al., 2016; Lee et al., 2013). According to Paulsen et al. significant calcareous pelagic deposition didn’t begin until the Cretaceous (Caldeira, 1992), though it has commonly been taken to mark signs of surges in carbonate subduction (Edmond & Huh, 2003) that resulted from the increasing input of exogenic CO₂. Carbon inputs into the subduction zones of the Precambrian are offered by the subduction of carbonated ocean crust and mantle, as well as the carbonate that was stripped from the continents by erosion and shed into the adjacent trenches. There are fundamental uncertainties, however, concerning palaeotemperature and palaeopressure regimes of ancient arcs and whether they met the required conditions of efficient decarbonation of slabs that are subducted to fuel substantial emissions of CO₂ (Dasgupta, 2013).

Time periods of increased carbonatite magmatism have also been noted to appear to correlate with warmer climates in the past, such as during the Cretaceous and the Eocene (Kerrick, 2001), as well as the Ediacaran (Santosh & Omori, 2008). There is a plausible causal link, as on ascent carbonatite melts can degas, thereby releasing significant amounts of CO₂ when they reach relatively shallow depths of less than 90 km (Hammouda & Keshav, 2015). Typically, carbonatites are constituted of ≥50% carbonate minerals (Rukhlov, Bell & Amelin, 2015). They therefore emit an amount of CO₂ that is anomalously large compared with the average arc magma, which typically contains less than 5,000 ppm of CO₂ (Wallace, 2005), even if the total production rate of magma is low. It is equally important that this type of magma erupts with other alkaline magmas that also have a highly enriched CO₂ content (Bailey & Hampton, 1990). Studies that have shown the potential for these magma types to emit between 10 and 50 times more CO₂ to the atmosphere, respectively, than arc magmas (Woolley & Kemp, 1989; Hudgins et al., 2015), which highlights the global impact of alkaline and carbonatite magmatism. The record of alkaline and carbonatite magmatism in the rock record, therefore, may point to significant, punctuated releases of CO₂ from the subcontinental mantle reservoir (Kerrick, 2001); a process that is potentially important operating outside the steady state within the carbon cycle that warrants investigation.

Paulsen et al. carried out a coupled U-Pb age and trace element analysis of 5,715 detrital zircons that were obtained from 46 sandstone samples, that were dated primarily to the Neoproterozoic to the Early Palaeozoic, that were distributed widely along a swath of about 3,000 km Neoproterozoic-Early Palaeozoic Pacific–Gondwana margin in Antarctica. This sandstone is known to contain high concentrations of protypical zircon dating to 700-500 Ma zircon U-Pb age population that overlaps with the global warming phase of the Ediacaran-Cambrian that have previously been linked to higher emissions of CO₂ from continental arcs (McKenzie et al., 2016). According to Paulsen et al. the sources of zircons of this age population may include exposed igneous provinces in Antarctica that are covered by ice, such as the Ross Orogen (Goodge, Fanning, Norman & Bennet, 2012; Hagen-Peter, Cottle, Smit & Cooper, 2016; Paulsen et al., 2016), e.g., as well areas of the interior of Gondwana such as the East African Orogen (Squire, Campbell, Allen & Wilson, 2006). Trace elements in the detrital zircons can provide constraints on the types of magma from which the zircons crystallised with reasonable probability, regardless of their precise provenance, which thereby provides a test for a linkage that has been hypothesised between carbonatite-alkaline magmatism and global climate changes for the time interval Ediacaran-Cambrian. In particular, zircons that are recovered from carbonatites and other alkaline magmas can be discerned confidently based on the Lu, Ta and U concentrations. Magma production by a low degree of partial melting of a subcontinental mantle that is metasomatised or carbonated or recycled oceanic crust that includes sediments (Pilet, Baker & Stolper, 2008), is a reflection of these trace elements.

Carbonatite-alkaline magmatism – evidence for a spike

A U-Pb age probability peak at 577-553 Ma was yielded by the zircons with carbonatite-alkaline signatures in all the areas of the study that contain a large population at 700-500 Ma. During the 700-500 Ma (Cryogenian-Cambrian) time interval (n=368; 17% of this U-Pb age population), though earlier time intervals are characterised by lower numbers of such carbonatite-alkaline grains. Paulsen et al. suggest that though some of the zircons may be misidentified with regard to their parent rock types (Belousova, Griffin, O’Reilly & Fisher, 2002), this does not explain the systematic change in the chemistry that is identified herein for the 700-500 Ma zircons that were recovered from the continental margin, for which the trace element signatures suggest they were derived from carbonatite and alkaline source rocks. Age peaks that have been shown by large global detrital zircon datasets have, similarly, been argued to represent a preservation bias according to which zircons are preserved selectively in sedimentary successions deposited over stable continental shields during assembly of a supercontinent (Cawood, Hawksworth & Dhuime, 2013). Statistical tests (χ²) on the cumulative zircon U-Pb age presented here confirm, however, that the proportion of carbonatite-alkaline zircons, relative to all other types of rock, is significantly higher for the time interval 900-400 Ma relative to earlier time periods. Paulsen et al. therefore concluded that the distribution of carbonatite-alkaline U-Pb age in the detrital record is not simply a preservation bias; instead it reflects a spike in the carbonatite-alkaline magmatism.

According to Paulsen et al. the results that were outlined above were interpreted to have a global significance when 3 important points are considered:

1) The length of the palaeocontinental margin, in terms of scale, along which evidence for this spike in magmatism has been identified, is itself roughly equivalent to the Nazca-South American convergent plate margin along coastal Chile.

2) In terms of regional extent, it has also been revealed by earlier analyses of detrital zircon U-Pb age and trace element data that there was a significant peak of alkaline zircons in sediments dating to the Permian-Triassic and younger over an extensive area from distal localities that include southeast Australia, southwest Australia, Antarctica, Prydz Bay and Dronning Maud Land, India and South Africa (Veevers, 2007).

3) Global in situ records of outcrops of carbonatites and Kimberlites record the same overall increase from 700-500 Ma that is indicated by the dataset of Paulsen et al. (Woolley & Bailey, 2012; Stern, Leybourne & Tsujimori, 2016), and this result lends support to the trace element-based parent rock classification applied in this study. These results, collectively, indicate a widespread enrichment of the mantle (Woolley & Bailey, 2012) prior to the Cryogen-Cambrian interval of time.

Planetary sequestration and release of carbon

Carbon in the mantle may be primordial or sourced from the crust and sediments that have been recycled by subduction (Dasgupta, 2013). Part of the arguments for crustal carbonate that has been subducted is not involved directly in the generation of alkaline-carbonatite magmas (Bell & Simonetti, 2010) is a general paucity of carbonatites along active subduction zones. Recent boron isotope analyses, however, point to enriched (crustal) signatures involved in carbonatites generation less than an age of 300 Ma, as well as during 2 earlier episodes of the assembly of supercontinents; the 1^st at 1.8 Ga and significantly, the 2^nd at 570 Ma during the spike in carbonatite-alkaline activity that is identified in this paper (Hulett, Simonetti, Rasbury & Hemming, 2016). This spike of carbonatite-alkaline magmatic activity at 700-500 Ma, importantly, also correlates with the appearance of blueschist (Tsujimori & Ernst, 2014) and ultrahigh pressure (UHP) (Liou et al., 2014) metamorphism in the rock record. A hallmark of a cooler subduction environment is blueschist facies metamorphism, which may in turn foster UHP metamorphism by deeper subduction of crust (Brown, 2007). Temperature is a dominant factor (Dasgupta, 2013) among several variables on which the decarbonation of subducting oceanic lithosphere depends. Efficient decarbonation of oceanic slabs at depths beneath forearc and arc is favoured by warmer subduction zones, whereas transport of carbon to greater depths is allowed by cooler subductions where the mantle may be enriched by the subducted crust (Dasgupta, 2013; Tsujimori & Ernst, 2014; Horton, 2015). It is suggested by this that the pulse of CO₂-rich carbonatite-alkaline magmatism evolved from secular cooling of the mantle. Greater sequestration of carbon in the mantle reservoir is fostered by the secular cooling, a process that Paulsen et al. suggest may have been ongoing during the Proterozoic to the Mesozoic (Dasgupta, 2013; Stern, Leybourne & Tsujimori, 2016) and probably increased around the time of the first appearance of blueschist and UHP metamorphic rocks in the geologic record. Carbon that had been sequestered was in turn released when tectonic conditions favoured extraction by a low degree of partial melting.

A strong affinity to continental rifts (Woolley & Bailey, 2012; Woolley, 1989) is shown by carbonatite and alkaline magmas, which implies that they are produced by processes that are related to the extension of the crust. Carbonatites have also been noted within Contractional orogenic belts (Woolley, 1989), which include those that are associated with subduction (Hagen-Peter & Cottle, 2016). The magmas in these settings are, however, also produced presumably when and where stress fields favour low degrees of partial melting in the mantle and rapid ascent of magma (Hammouda & Keshav, 2015); e.g., episodes of rollback of slabs or post-collisional/post-orogenic extension (Veevers, 2007; Hagen-Peter & Cottle, 2016). During the tectonic activity that marked the Ediacaran there was a spike in carbonatite and alkaline magmatism as carbon was released by the subcontinental mantle reservoir that had previously been sequestered during episodes of widespread crustal extension, which was probably occurring with back-arc rifting and post-orogenic extension associated with the assembly of Gondwana (Veevers, 2007, Hagen-Peter & Cottle, 2016) 700-500 Ma, as well as continental rifting (e.g., associated with Iapetus rifting (Gernon et al., 2016; Doig, 1970; Tappe et al., 2006) about 615-550 Ma).

Global carbon cycle

It has previously been found that shifts towards lower ¹³C concentrations in marine carbonates correlate with episodes of increased magmatism that are associated with greenhouse spikes that, in turn, may have been the cause of acidification of the ocean as well as mass extinctions, such as the Permian-Triassic (Payne et al., 2010) and the boundary of the Palaeocene-Eocene (Storey, Duncan & Swisher, 2007). Paulsen et al. suggest that the possibility of rapid release of significant CO₂ associated with carbonatite-alkaline magmatism that was recorded in their samples therefore warrants a comparison to the Neoproterozoic ¹³C isotope record. It has been shown by the cumulative carbonatite-alkaline zircon U-Pb age population that there was an increase after 680-670 Ma near the end of the glaciation of the Sturtian Snowball Earth and a rise to a peak at approximately 576-565 Ma. It is shown by the ¹³C isotope record of the Neoproterozoic-Cambrian progressively rising high average values, about 5‰, during the preceding 1,000-750 Ma time interval. This pattern is consistent with increased sequestration of organic carbon by burial in sediments, which may have acted in concert with reduced outgassing from continental arc that resulted in the Earth being plunged into the global Sturtian glaciation (McKenzie et al., 2016; Cox et al., 2016).

These high values of ¹³C concentration in the atmosphere in the Neoproterozoic, which may reflect in part the increased sequestration of carbon into the mantle reservoir, are in turn was punctuated by several profound negative excursions that are not well understood (Grotzinger, Fike & Fischer, 2011). The inception of the Shuram-Wonoka excursion, which was the largest negative ¹³C shift that in known of in the rock record of the Earth, ranges from about 601-556 Ma, though with full recovery of the excursion by 551 Ma (Husson et al., 2015), following which the ¹³C concentration record shows lower average values (about 0-1 ‰). Though from around 850-720 Ma, during a time interval that was marked by several ¹³C concentration shifts, the eruption of several large igneous provinces occurred, there are few significant large igneous provinces that are known to overlap in age with the Shuram-Wonoka excursion. The data resulting from this study, in contrast, show that the release of carbon that was previously sequestered from the subcontinental reservoir by way of carbonite-alkaline magmatism overlapped temporarily with (with the available age constraints) this distinct ¹³C concentration perturbation.

Recent work in Oman, Australia and the western US confirms that, in some localities, it reflects a primary (marine) signal (Husson et al., 2015; Lee et al., 2015; Minguez & Kodama, 2017), though the Shuram-Wonoka excursion has been attributed to secondary (diagenetic) processes (Grotzinger, Pike & Fischer, 2011; Derry, 2010;). Paulsen et al. noted that the mantle is heterogonous and bimodal in its ¹³C concentration composition (-6‰ and -25‰) (Deines, 2002), though the -12‰ ¹³C concentration nadir of the Shuram-Wonoka excursion exceeds the level that is typically considered by carbon cycle models to be the canonical -7‰ to -25‰ ¹³C concentration value for magmatism (Fike, Grotzinger, Pratt & Summons, 2006). Therefore assuming that the ¹³C concentration perturbation is a primary signal, mass balance suggests that the cause of the excursion may be multifactorial, though it is influenced strongly by a carbonatite-alkaline magmatic component.

The important question raised by the results of this study is whether light carbon that has been released from a hypothetical large ocean carbon reservoir (Fike, Grotzinger, 2006; Grotzinger, Fike & Fischer, 2011; Rothman, Hayes & Summons, 2003) dating to the Neoproterozoic was sourced in part, at least, from greater depths in the mantle. If the prominent increase in carbonatite-alkaline magmatism that overlaps this excursion is representative of the global pattern, then a major CO₂ concentration contribution to the atmosphere is highlighted by it. This outgassing of carbon from the mantle of the Earth is likely to have promoted global warming, given the anomalously high amount of carbon that was degassed from this type of system compared with all other types of magma. This release of significant volumes of CO₂ is consistent with the planet dodging a snowball glaciation during the short-lived, about 580 Ma, Gaskiers glacial episode (Pu et al., 2016). Also, the appearance of a significant volume of carbonatite-alkaline magmatism-volcanism that was produced by melting of mantle or oceanic crust that was metasomatised and carbon enriched marks the recycling of a similarly significant volume of crustal material, and it therefore represents the inception of the modern carbon cycle. Paulsen et al. suggest more work is required to expand the sample suite to other continental regions in order to improve rock classification schemes that are based on zircon trace element chemistry, will help to refine the models set forth in this paper further and foster the understanding of the relationship between the endogenic and exogenic components of climate change over time.

Sources & Further reading