Predominantly recycled carbon in Earth's upper mantle revealed by He-CO2-Ba systematics in ultradepleted ocean ridge basalts
Introduction
The degree to which carbon has been recycled to Earth's deep and shallow mantle is debated. An early view, comparing the concentration and isotope composition of C in peridotites and basalts, was that much of the carbon throughout Earth's mantle was recycled over geologic time given the apparent large flux of carbon from the mantle compared to the mass budget of carbon in the so-called “exosphere”, i.e., the crust+oceans+atmosphere (Javoy et al., 1982). Although the magnitude of that flux was probably overestimated (Walker, 1983; Javoy et al., 1983), the amount of C recycled may still have been significant. Given that the mantle has cooled over geologic time, and the thermal structure of subduction zones has allowed crustal material to be transported to depths greater than the generation depth of volcanic arc magma, recycling of carbon to the mantle seems quite likely to have occurred since at least the Proterozoic (Dasgupta, 2013).
Recent studies of carbon at the Hawaiian hotspot indicate that there is a relatively larger concentration of C (by at least a factor 3 to 8) in its deep mantle source compared to Earth's shallow mantle source of mid-ocean ridge basalts (Tucker et al., 2019). Given that other radiogenic isotope and noble gas evidence shows that the Hawaiian mantle source contains both recycled and primitive mantle components, its enrichment in C may be due either to an abundance of C-rich primitive material or to a significant recycling of surficial C (Tucker et al., 2019). In a study of xenoliths from high-3He/4He ocean island plume localities (Loihi-Hawaii, Réunion, and Kerguelen), Trull et al. (1993) showed that the similar C isotope and C/3He systematics for ocean island and MORB sources were consistent with zoned mantle convection and significant surficial C recycling to the deep mantle, with limited or no C recycling to the upper mantle. Recently, Kelemen and Manning (2015) showed that there seems to be a significant accumulation of C in the continental lithosphere during subduction and plate collision, and argued that the amount of C recycling to the asthenosphere may actually be negligible. But in another recent study, Hirschmann (2018) estimated that >25% of the C that has been outgassed to the exosphere has been recycled to the mantle over geologic time based on the observed difference in CO2/Ba between the mantle and the exosphere.
Mid-ocean ridge basalts are important for assessing the compositional variability of Earth's upper mantle and for providing insight into the deep carbon cycle. Notably, MORBs that are ultradepleted in incompatible elements (UD-MORBs) are sometimes also volatile-undersaturated; that is, their total volatile pressure (determined from the CO2 and H2O content of the glass plus any bubbles that might be present) is less than the sample collection pressure. Olivine-hosted melt inclusions (OHMI) may also retain undegassed magmatic volatile concentrations. Under these circumstances, basalt or OHMI compositions may be used to estimate the C concentration of the mantle source and the volcanic C flux to the exosphere (Saal et al., 2002; Rosenthal et al., 2015; Michael and Graham, 2015; Le Voyer et al., 2017, 2018; Hirschmann, 2018). Although volatile-undersaturated basalt glasses are rare, they are significant because they can represent magma that has not lost any gas since its formation, and aid in the development of chemical proxies for volatile elements such as C that are lost to an unknown amount in other volcanic rocks.
There are at least two complications when utilizing undersaturated basalt glasses to characterize the upper mantle. First, their rare occurrence can make it difficult to obtain a geologic context and understand their relationship to nearby basalts that are not undersaturated. Michael and Graham (2015) suggested from trace element abundances in their study of UD-MORBs that they could be derived by partial melting of an upper mantle source that had undergone an earlier and relatively recent melting event during which 1-2% of melt was removed. About a dozen separate localities along the global mid-ocean ridge system were found to erupt undersaturated basalts; these localities are sometimes associated with atypical tectonic settings, such as near the end of ridge segments or, in a number of cases, they occur at intra-transform spreading centers. A second complication arises from the nearly ubiquitous occurrence of concurrent mixing and crystal fractionation in magma as it ascends and aggregates beneath mid-ocean ridges (Shorttle, 2015). This phenomenon could lead to the eruption of volatile-undersaturated magmas (or OHMI) having ratios of volatile/lithophile elements, such as CO2/Ba or CO2/Nb, that are lower than their mantle source values (Matthews et al., 2017).
The present study investigates a suite of 25 submarine basalts collected from well surveyed regions of the southern East Pacific Rise and within the Garrett Transform Fault. We find that CO2 is highly correlated with Ba and Nb in the transform domain basalts (n = 13, including 5 undersaturated samples). These GTF lavas have CO2/Ba and CO2/Nb ratios (ppm/ppm) of 106 ± 8 and 308 ± 27, respectively, and are indistinguishable from the global UD-MORB suite of Michael and Graham (2015). Furthermore, 3He/4He shows no overlap between EPR and GTF basalt populations. The GTF basalts are derived by partial melting of mantle that underwent a small degree of prior melting beneath the EPR (Wendt et al., 1999). The ultradepleted mantle sampled by the GTF basalts has 3He/4He >9.2 RA and molar CO2/3He . We propose that the helium isotope and CO2/3He ratios of typical MORB (∼7-9 RA and ∼2, respectively) are significantly affected by the presence of tectonically recycled material in their source, and that most of the carbon present in Earth's upper mantle is not juvenile, but has been tectonically recycled from Earth's surface. (The term juvenile is used here to indicate primordial material that has never been present at Earth's surface.)
Section snippets
Background
The Garrett Transform Fault is 130 km long and offsets the superfast spreading East Pacific Rise near 13°30'S (145 mm/y full spreading rate). The transform domain is volcanically active, and basalt, gabbro, and peridotite have all been recovered there (Hébert et al., 1983, 1997; Hékinian et al., 1992, 1995; Cannat et al., 1990; Niu and Hékinian, 1997). Basalts from both the nearby East Pacific Rise and within the Garrett Transform Fault are geochemically well-characterized (Sinton et al., 1991;
Methods and results
Analyses of 3He/4He and He concentrations in vesicles and glass were performed through a series of crushing and melting experiments followed by noble gas mass spectrometry. The CO2 content of vesicles was determined by capacitance manometry, and CO2 and H2O concentrations dissolved in the glasses were measured by FTIR spectroscopy. The basalt glasses were also analyzed for trace element concentrations by laser ablation ICP-MS. Methods follow those described previously (Michael and Graham, 2015;
CO2-Ba and CO2-Nb systematics and the origin of GTF undersaturated basalts
Michael and Graham (2015) refined the CO2/Nb proxy of Saal et al. (2002) and Cartigny et al. (2008) by studying a global suite of undersaturated mid-ocean ridge basalts. They found that another useful proxy element was Ba, and suggested that the depleted upper mantle was characterized by a CO2/Ba ratio of 105 ± 9. Using a simplistic melting model (mean extent of partial melting of 12%) and the mean global MORB Ba concentration of 29.2 ppm (Gale et al., 2013), they estimated the upper mantle [CO2
Summary
He and CO2 were measured in vesicles and glass of East Pacific Rise and Garrett Transform basalts. Total volatile pressures in the basalts reveal the presence of 5 volatile-undersaturated basalts collected from 3 distinct locations within the GTF (a leaky strike-slip fault, Beta Ridge, and adjacent to Alpha Ridge). The CO2-Ba and CO2-Nb systematics in the GTF undersaturated basalts indicate an origin by variable source depletion, with limited or no mixing of melts. The mantle source has ratios
CRediT authorship contribution statement
David W. Graham: Conceptualization, Data curation, Formal analysis, Investigation, Methodology, Visualization, Writing - original draft, Writing - review & editing. Peter J. Michael: Formal analysis, Investigation, Methodology, Visualization, Writing - review & editing.
Declaration of Competing Interest
The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.
Acknowledgments
Thanks to John Sinton for providing the EPR samples and Craig Lundstrom and Frank Tepley for the Garrett samples. We thank Jackie Dixon and Jonathan Tucker for very constructive and helpful reviews. Fig. 6 was produced using GeoMapApp (http://www.geomapapp.org/), which is supported by the National Science Foundation and internal funding from Lamont-Doherty Earth Observatory. This work was supported by The National Science Foundation through grants OCE15-58798 (DG) and OCE15-58802 (PM).
References (71)
- et al.
Plastic deformation and magmatic impregnation in serpentinized ultramafic rocks from the Garrett transform fault (East Pacific Rise)
Earth Planet. Sci. Lett.
(1990) - et al.
Towards a consistent mantle carbon flux estimate: insights from volatile systematics (H2O/Ce, ∂D, CO2/Nb) in the North Atlantic mantle
Earth Planet. Sci. Lett.
(2008) - et al.
Influence of magma ascent rate on degassing of carbon dioxide at oceanic ridges: message in a bubble
Earth Planet. Sci. Lett.
(2012) - et al.
Constraints on the noble gas composition of the deep mantle by bubble-by-bubble analysis of a volcanic glass samples from Iceland
Chem. Geol.
(2015) Fractionation of U and Th during mantle melting: a reprise
Chem. Geol.
(1997)- et al.
Constraints on the DUPAL anomaly from helium isotope systematics in the Southwest Indian mid-ocean ridge basalts
Chem. Geol.
(2015) - et al.
Helium isotope geochemistry of mid-ocean ridge basalts from the South Atlantic
Earth Planet. Sci. Lett.
(1992) - et al.
Extreme incompatibility of helium during mantle melting: evidence from undegassed mid-ocean ridge basalts
Earth Planet. Sci. Lett.
(2016) - et al.
Variations in magmatic He, CO2 and C during a mid-ocean ridge eruption at 9°50'N on the East Pacific Rise
Earth Planet. Sci. Lett.
(2018) - et al.
Ancient melt extraction from the oceanic upper mantle revealed by Re-Os isotopes in abyssal peridotites from the Mid-Atlantic Ridge
Earth Planet. Sci. Lett.
(2006)
Ultramafic and mafic rocks from the Garrett Transform Fault near 13°30'S on the East Pacific Rise: igneous petrology
Earth Planet. Sci. Lett.
Volcanic activity and crust-mantle exposure in the ultrafast Garrett transform fault near 13°28'S in the Pacific
Earth Planet. Sci. Lett.
Comparative deep Earth volatile cycles: the case for C recycling from exosphere/mantle fractionation of major (H2O, C, N) volatiles and from H2O/Ce, CO2/Ba and CO2/Nb exosphere ratios
Earth Planet. Sci. Lett.
Subduction cycling of U, Th, and Pb
Earth Planet. Sci. Lett.
Correlated helium, neon, and melt production on the super-fast spreading East Pacific Rise near 17°S
Earth Planet. Sci. Lett.
Negligible sulfur isotope fractionation during partial melting: evidence from Garrett transform fault basalts, implications for the late-veneer and the hadean matte
Earth Planet. Sci. Lett.
Isotope and trace element characteristics of a super-fast spreading ridge: East Pacific Rise, 13-23°S
Earth Planet. Sci. Lett.
Constraints on the mantle mineralogy of an ultra-slow ridge: hafnium isotopes in abyssal peridotites and basalts from the 9-25°E Southwest Indian Ridge
Earth Planet. Sci. Lett.
The origins and concentrations of water, carbon, nitrogen and noble gases on Earth
Earth Planet. Sci. Lett.
C/3He in volatile fluxes from the solid Earth: implications for carbon geodynamics
Earth Planet. Sci. Lett.
Volatiles (He, C, N, Ar) in mid-ocean ridge basalts: assessment of shallow-level fractionation and characterization of source composition
Geochim. Cosmochim. Acta
Constraining mantle carbon: CO2-trace element systematics in basalts and the roles of magma mixing and degassing
Earth Planet. Sci. Lett.
The behavior and concentration of CO2 in the suboceanic mantle: inferences from undegassed ocean ridge and ocean island basalts
Lithos
Estimating the carbon content of the deep mantle with Icelandic melt inclusions
Earth Planet. Sci. Lett.
VolatileCalc: a silicate melt-H2O-CO2 solution model written in Visual Basic for Excel
Comput. Geosci.
Basaltic liquids and harzburgitic residues in the Garrett Transform: a case study at fast-spreading ridges
Earth Planet. Sci. Lett.
The evolution of helium isotopes in the upper mantle and the generation of isotopic anomalies
Earth Planet. Sci. Lett.
Stochastic melting of the marble cake mantle: evidence from local study of the East Pacific Rise at 12°50'N
Earth Planet. Sci. Lett.
Experimental determination of C, F, and H partitioning between mantle minerals and carbonated basalt, CO2/Ba and CO2/Nb systematics of partial melting, and the CO2 contents of basaltic source regions
Earth Planet. Sci. Lett.
Mantle melting and magma supply to the Southeast Indian Ridge: the roles of lithology and melting conditions from U-series disequilibria
Earth Planet. Sci. Lett.
Two-component mantle melting-mixing model for the generation of mid-ocean ridge basalts: implications for the volatile content of the Pacific upper mantle
Geochim. Cosmochim. Acta
Evaluating the roles of melt-rock interaction and partial melting on the CO2/Ba ratios of MORB: implications for the CO2 budget in Earth's depleted mantle
Geochim. Cosmochim. Acta
Geochemical variability in MORB controlled by concurrent mixing and fractional crystallisation
Earth Planet. Sci. Lett.
He, Ne and Ar isotope signatures of mid-ocean ridge basalts and their implications for upper mantle structure: a case study from the Mid-Atlantic Ridge at 4-12°S
Geochim. Cosmochim. Acta
U-series disequilibria in MORB from the Garrett Transform and implications for mantle melting
Earth Planet. Sci. Lett.
Cited by (9)
A carbon, nitrogen, and multi-isotope study of basalt glasses near 14°N on the Mid-Atlantic Ridge. Part B: Mantle source heterogeneities
2024, Geochimica et Cosmochimica ActaCarbon isotope composition of basalts from Loihi Seamount: Primordial or recycled carbon in the Hawaiian mantle plume?
2023, Earth and Planetary Science LettersHigh-C content and CO<inf>2</inf>/Ba ratio of the Earth's enriched upper mantle
2023, Geochimica et Cosmochimica ActaCitation Excerpt :Given that the CO2/Ba ratio of the flux from the mantle into the exosphere (at mid-ocean ridge and ocean island) is higher than the CO2/Ba ratio in the exosphere, this suggests that the material that is returned into the mantle at subduction zones must have a CO2/Ba ratio that is even higher than that of the mantle (i.e., more efficiently recycling of C back into the convecting mantle relative to Ba). This is consistent with studies that have argued for significant recycling of subducted C into the convecting mantle (e.g., Javoy et al., 1982; Marty and Zimmermann, 1999; Sleep and Zahnle, 2001; Dasgupta and Hirschmann, 2010; Farsang et al., 2021; Graham and Michael, 2021). If this is the case, we speculate that our observation of high CO2/Ba ratio and C content in the Pacific E-MORB mantle source would be consistent with greater recycling of C relative to Ba through subduction (e.g., Hirschmann and Stolper, 1996; Donnelly et al., 2004).
Carbon stable isotope constraints on CO<inf>2</inf> degassing models of ridge, hotspot and arc magmas
2022, Chemical GeologyCitation Excerpt :For hotspot samples, the SiO2 content has been used to take into account the increase in CO2 solubility for alkalic melts undersaturated in SiO2 (Dixon, 1997). Carbon concentration and isotope data have been compiled from the following references: Alt et al. (1993), Aubaud (2002), Aubaud et al. (2004, 2005, 2006), Barry et al. (2014), Blank et al. (1993), Cartigny et al. (2001, 2008), Cushman et al. (2004), Des Marais and Moore (1984), Dixon et al. (1988, 1991, 1997, 2002, 2017), Dixon and Clague (2001), Exley et al. (1986), Fine and Stolper, (1985/1986), Gardner et al. (2016), Graham et al. (2018), Graham and Michael (2020), Gribble et al. (1996), Helo et al. (2011), Hochstaedter et al. (1990), Jackson et al. (2015), Javoy and Pineau (1991), Jendrzejewski (1994), Jendrzejewski et al. (1996a, 1996b), Jones et al. (2018, 2019), Kendrick et al. (2014, 2015), Le Roux et al. (2006), Le Voyer et al. (2015, 2017, 2019), Macpherson and Mattey (1994), Macpherson et al. (1999, 2005), Marty and Jambon (1987), Marty and Zimmermann (1999), Mattey et al. (1984, 1989), Michael and Graham (2015), Moore et al. (1977), Newman et al. (2000), Nishio et al. (1998, 1999), Pineau et al. (1976, 2004), Pineau and Javoy (1983), Pineau and Javoy (1994), Ribeiro et al. (2015), Saal et al. (2002), Sakai et al. (1984), Shaw et al. (2004, 2010), Shimizu et al. (2016), Simons et al. (2002), Soule et al. (2012), Standish et al. (2008), Stolper and Newman (1994), Wanless and Shaw (2012) and Workman et al. (2006). The compiled CO2 and H2O concentration data are presented in Fig. 4.
The origin of nitrogen in Earth's mantle: Constraints from basalts <sup>15</sup>N/<sup>14</sup>N and N<inf>2</inf>/<sup>3</sup>He ratios
2022, Chemical GeologyCitation Excerpt :The CO2/3He ratios of depleted MORB is ~5.0 × 108 (Marty and Zimmermann, 1999, Graham and Michael, 2021). Enriched MORB have higher CO2/3He, on average ~ 6.0 × 109 (Graham and Michael, 2021, Marty and Zimmermann, 1999, Péron et al., 2019). The relationship between CO2/3He and enrichment likely results from carbon recycling as an illustration of the marble-cake mantle (Marty and Zimmermann, 1999).