The effects of local variations in conditions on carbon storage and release in the continental mantle

Abstract

Recent advances indicate that the amount of carbon released by gradual degassing from the mantle needs to be revised upwards, whereas the carbon supplied by plumes may have been overestimated in the past. Variations in rock types and oxidation state may be very local and exert strong influences on carbon storage and release mechanisms. Deep subduction may be prevented by diapirism in thick sedimentary packages, whereas carbonates in thinner sequences may be subducted. Carbonates stored in the mantle transition zone will melt when they heat up, recognized by coupled stable isotope systems (e.g. Mg, Zn, Ca). There is no single 'mantle oxygen fugacity', particularly in the thermal boundary layer (TBL) and lowermost lithosphere, where very local mixtures of rock types coexist. Carbonate-rich melts from either subduction or melting of the uppermost asthenosphere trap carbon by redox freezing or as carbonate-rich dykes in this zone. Deeply derived, reduced melts may form further diamond reservoirs, recognized as polycrystalline diamonds associated with websteritic silicate minerals. Carbon is released by either edge-driven convection, which tears sections of the TBL and lower lithosphere down so that they melt by a mixture of heating and oxidation, or by lateral advection of solids beneath rifts. Both mechanisms operate at steps in lithosphere thickness and result in carbonate-rich melts, explaining the spatial association of craton edges and carbonate-rich magmatism. High-pressure experiments on individual rock types, and increasingly on reactions between rocks and melts, are fine-tuning our understanding of processes and turning up unexpected results that are not seen in studies of single rocks. Future research should concentrate on elucidating local variations and integrating these with the interpretation of geophysical signals. Global concepts such as average sediment compositions and a uniform mantle oxidation state are not appropriate for small-scale processes; an increased focus on local variations will help to refine carbon budget models.

Keywords: carbonate melts; continental rifts; craton destruction; deep carbon cycle; lithospheric mantle.

Figure 1.

Summary of carbon transport processes in subduction and continental lithosphere. During subduction, thick carbonate sediments will rise diapirically [13], preventing deep subduction (1) as do clastic sediments in hot, wet conditions [120] (2). Thin carbonate sequences will be deeply subducted and may melt when stagnant slabs heat up in the mantle transition zone (3). Under the continents, the amount of gradual carbon degassing into the lithosphere (4) has been revised upwards, whereas plume input (5) may have been overestimated. Most carbon is trapped in the lithosphere in reduced form from melts, with carbonate-rich dykes forming only where the volume of introduced melt is high.

Figure 2.

Concepts behind plume inputs into the continental lithosphere. (a) Theoretical and numerical models invoke a narrow plume shaft, usually from the core–mantle boundary, and a large plume head [28,29], which (b) leads to a temperature anomaly of 200–300˚C or more [127]. (c) Since these models were introduced, many tomographic images have been available, which instead show more diffuse tongues of hot material with smaller temperature anomalies (based on [30]), implying that carbon inputs into the lithosphere may need to be revised downwards.

Figure 3.

Melting curves for clastic and carbonate sediments compared to subduction geotherms (grey region; [119]). Only fluid-saturated pelitic sediments will melt along modern subduction geotherms. Limestone with a 7% clastic component melts to produce silicate melt [13]; carbonate melts are produced only at high pressures (>5 GPa) from either carbonate or carbonated silicate sediments. Data summarized from [13,37,128].

Figure 4.

Behaviour of carbonate components in sediments and altered igneous oceanic crust. (1) Thick chalk or limestone sequences will rise diapirically, preventing deep subduction, and become stored in solid form in the mantle wedge beneath arcs [13]. (2) Minor carbonate components in sediments dominated by clastic materials will be subducted unless the clastic sediments rise as diapirs. (3) Predominantly carbonate sediments with minor clastic material may produce minor silicate melts, but not lose their carbon until they stall in the mantle transition zone (Fig. 3). These components and the type of sediment mixture are traceable by Mg and Zn isotopes [52,56,57]. (4) Ophicarbonates in the altered igneous sections of oceanic crust may melt only if fluxed by water from dehydration of the ultramafic rocks below (4a); otherwise they should be deeply subducted if they remain dry (4b) [40].

Figure 5.

Oxygen barometry of peridotite xenoliths indicates that the cratonic lithosphere is generally too reducing for carbonate-rich melts (xenoliths and red lines marking carbonate contents of melts from [67]; solid straight line shows average). Episodic local carbonate-rich melts from just below the lithosphere or from subducted material (arrows) cause interaction between oxidized melt and reduced rocks (circle), resulting in diamond precipitation or local carbonate-phlogopite veins and dykes [61,101]. MBL = mechanical boundary layer; TBL = thermal boundary layer.

Figure 6.

Melting curves of mantle peridotite in oxidized and reduced conditions and their relationship to likely geotherms. (a) Possible melting curves in the lithosphere: the oxidized solidus for peridotite with CO₂ and H₂O [71], has shaded areas for melt compositions (carbonatite and aillikite [10%–30% SiO₂]). The dashed line for reduced solidus is the most appropriate solidus, corresponding to reducing conditions with low water activity [68,129]. (b) Melting curves for reduced conditions; curves labelled with water activity in H₂O + CH₄ mixtures [129]. These curves are relevant for the asthenosphere, where conditions are mostly too reduced for the oxidized solidus in (a) to apply [98]. (c) Cratonic geotherms fit to an isentrope of 1315˚C (dotted line (1) [7,72]), showing that melts would be widespread in the lithosphere if conditions were oxidized, but they are not: the reduced solidus is not reached. Dotted curves (2) and (3) apply to upwelling hot material and regional downwelling, respectively, emphasizing that curve (1) is just a global average. The TBL is wider here to account for the full range of geotherms shown. (d) Summary of the most likely real-world melting curves for variation of the oxidation state with depth. No melting in the lithosphere gives way to two types of melts in the upper asthenosphere and TBL: oxidized, carbon-rich melts as in (a) are restricted to the uppermost asthenosphere and mostly solidify to deposit diamond (dark diamonds). Reduced upwelling mantle from the deeper asthenosphere will melt when it meets the solidus (dotted line) and these melts solidify on encountering the more reduced TBL/lithosphere, depositing polycrystalline diamonds (light diamonds). These melts are more oxidized than their source due to dissolution of volatiles in the melt.

Figure 7.

Precipitation mechanisms for carbon in the TBL and lower continental lithosphere. (a) Oxidized, carbonate-rich melts (red arrows) may originate from subducted materials. These initially solidify to form diamonds by redox freezing [70,61]. Both diamond and carbonate may exist in close proximity due to local variations in oxygen fugacity and melt flux. (b) Oxidized, carbonate-rich melts may be produced from peridotite [71,130] or other ultramafic rocks in oxidized patches of the uppermost asthenosphere. At higher influxes of melt, carbonates will be deposited (dark vein system; the carbonation freezing front [66]). Reactions between carbonate-rich melts and peridotite produce a zone with minerals characteristic of hydrous mantle metasomatism [99,101]. (c) Reduced melts emanating from deeper in the asthenosphere (Fig. 6d) never attain the oxidation state necessary to form carbonate, depositing diamonds as polycrystalline diamond aggregates associated with websteritic silicate minerals [84,86]. These may be easily remobilized when overprinted by carbonate melts as in (a) and (b).

Figure 8.

Competing proposed mechanisms for the erosion of cratonic roots. (a) Edge-driven convection is concentrated at step-changes in lithosphere thickness [112,114]: movement in the convecting mantle tears rheologically weak sections of the thermal boundary layer (also known as the rheological boundary layer [131]) and the lowermost lithosphere downwards and sideways. This results in the heating and oxidation of diamond causing the production of carbonate-rich melts, which migrate upwards primarily at the edge of cratons. Sketch partly after [73]. (b) Lateral advection of crumbled solid pieces of the TBL and lower lithosphere (after [107]) is promoted beneath developing rifts. Melts are produced by oxidation of reduced carbon, heating and decompression, explaining the concentration of carbonate-rich melts at the juncture of cratons and rifts [4].