13:45 〜 14:00
[SGC35-01] Toward more complete (and realistic) assessment of deep carbon subduction; The devil (could be) in the details
★Invited Papers
キーワード：subduction, carbon cycling, metamorphism, New Zealand, UHP metamorphic rocks
We have undertaken studies of HP/UHP metamorphic rocks, sediments entering trenches, and volcanic gases aimed at determining the key factors governing the extents of deep carbon subduction, integrating field, petrologic, geochemical, and theoretical approaches in studies of ancient and modern subduction margins.
The blueschist to eclogite facies Schistes Lustrés, exposed in the French/Italian Alps, contains units of carbonate-rich sediment underplated at 40-70 km depths and variably metamorphosed along a P-T gradient similar to that experienced in modern subduction margins. Integrated field, petrologic, and theoretical study of this suite indicates that, at smaller scales, C in carbonate is removed and redistributed through decarbonation and dissolution; however, as for other studies of metamorphic rocks, it is difficult to scale such effects observed at outcrop scales to consideration of whole-margin cycling. It appears that, despite some remobilization of C in such rocks, they can retain large fractions of their protolithic C to depths approaching those beneath volcanic fronts (Cook-Kollars et al., 2014, Chemical Geology; Epstein et al., 2019, Lithos).
Study of along-margin C input flux at the Sunda margin (House, Bebout, and Hilton, 2019; Geology) indicates the influence of sedimentological environment, sediment sources, and degrees of sediment accretion, on the nature and size of the deeply subducted C reservoir. At this margin, proportions of organic and carbonate C vary dramatically along the strike of the margin, and extensive accretion in some areas removes large fractions of the C entering the trench. Shortfall in the calculated delivery of sedimentary C to the subarc, compared with volcanic outputs, seemingly requires significant additions of C via decarbonation in subducting altered oceanic crust.
We have been evaluating the cycling of C, N, and noble gases at the Hikurangi margin, with analyses of gases from across the forearc-arc-backarc (Epstein, Bebout, Christenson, Sumino, Wada, Werner, and Hilton, 2021, G-cubed). We have obtained C-N concentrations and isotope compositions of sediments outboard of the trench and wall-rock metasediment in the Taupo Volcanic Zone (TVZ). We compare these data with noble gas and C-N data for gases from fumaroles and thermal springs. This work integrates thermal modeling, thermodynamic calculations of prograde devolatilization, and estimation of TVZ CO2 output flux. Devolatilization reactions and volatile loss were calculated using the thermodynamic software Perple_X. We modeled subduction of a normal oceanic crust + serpentinized mantle peridotite as well as three idealized plateau structures containing variably altered, thickened oceanic crust. Seismic studies indicate significant sediment accretion along the Hikurangi margin and we include only 500m of pelagic carbonate in the model input. This modeling shows diffuse CO2 loss in the North extending from subarc depths to >220km depth. Towards the South, there is an increase in subarc CO2 flux but a significant decrease in backarc release. Total whole-margin flux estimates of slab-derived CO2 for the four models range from 3.9–6.2 Tg/yr or 1.3–2.1% of the global flux. Analyzed arc gases have δ13CVPDB from -11.2 to -1.4‰, and CO2/3He from 2x109 to 2.6x1011, indicating 20-87% C contribution from subducted carbonate (median=64.8%; using the approach of Sano and Marty, 1995, Chemical Geology). Overlap in δ13Cred and δ15N of the incoming sediments and Torlesse wall rocks complicates differentiation of C sources, but seismic evidence as well as geographic distributions of gas chemistry are consistent with the reduced C component being derived largely from assimilation of Torlesse metasedimentary material.
Reconciling oceanic inputs with volcanic outputs will require assessment of this range of factors, in order to understand individual-margin cycling and thus ultimately to understand the whole-Earth C balance.
The blueschist to eclogite facies Schistes Lustrés, exposed in the French/Italian Alps, contains units of carbonate-rich sediment underplated at 40-70 km depths and variably metamorphosed along a P-T gradient similar to that experienced in modern subduction margins. Integrated field, petrologic, and theoretical study of this suite indicates that, at smaller scales, C in carbonate is removed and redistributed through decarbonation and dissolution; however, as for other studies of metamorphic rocks, it is difficult to scale such effects observed at outcrop scales to consideration of whole-margin cycling. It appears that, despite some remobilization of C in such rocks, they can retain large fractions of their protolithic C to depths approaching those beneath volcanic fronts (Cook-Kollars et al., 2014, Chemical Geology; Epstein et al., 2019, Lithos).
Study of along-margin C input flux at the Sunda margin (House, Bebout, and Hilton, 2019; Geology) indicates the influence of sedimentological environment, sediment sources, and degrees of sediment accretion, on the nature and size of the deeply subducted C reservoir. At this margin, proportions of organic and carbonate C vary dramatically along the strike of the margin, and extensive accretion in some areas removes large fractions of the C entering the trench. Shortfall in the calculated delivery of sedimentary C to the subarc, compared with volcanic outputs, seemingly requires significant additions of C via decarbonation in subducting altered oceanic crust.
We have been evaluating the cycling of C, N, and noble gases at the Hikurangi margin, with analyses of gases from across the forearc-arc-backarc (Epstein, Bebout, Christenson, Sumino, Wada, Werner, and Hilton, 2021, G-cubed). We have obtained C-N concentrations and isotope compositions of sediments outboard of the trench and wall-rock metasediment in the Taupo Volcanic Zone (TVZ). We compare these data with noble gas and C-N data for gases from fumaroles and thermal springs. This work integrates thermal modeling, thermodynamic calculations of prograde devolatilization, and estimation of TVZ CO2 output flux. Devolatilization reactions and volatile loss were calculated using the thermodynamic software Perple_X. We modeled subduction of a normal oceanic crust + serpentinized mantle peridotite as well as three idealized plateau structures containing variably altered, thickened oceanic crust. Seismic studies indicate significant sediment accretion along the Hikurangi margin and we include only 500m of pelagic carbonate in the model input. This modeling shows diffuse CO2 loss in the North extending from subarc depths to >220km depth. Towards the South, there is an increase in subarc CO2 flux but a significant decrease in backarc release. Total whole-margin flux estimates of slab-derived CO2 for the four models range from 3.9–6.2 Tg/yr or 1.3–2.1% of the global flux. Analyzed arc gases have δ13CVPDB from -11.2 to -1.4‰, and CO2/3He from 2x109 to 2.6x1011, indicating 20-87% C contribution from subducted carbonate (median=64.8%; using the approach of Sano and Marty, 1995, Chemical Geology). Overlap in δ13Cred and δ15N of the incoming sediments and Torlesse wall rocks complicates differentiation of C sources, but seismic evidence as well as geographic distributions of gas chemistry are consistent with the reduced C component being derived largely from assimilation of Torlesse metasedimentary material.
Reconciling oceanic inputs with volcanic outputs will require assessment of this range of factors, in order to understand individual-margin cycling and thus ultimately to understand the whole-Earth C balance.