Decarbonation of subducting slabs

TY - CONF

T1 - Decarbonation of subducting slabs

T2 - insight from thermomechanical-petrological numerical modelling

AU - Gonzalez, Christopher M

AU - Gorczyk, Weronika

AU - Gerya, Taras

PY - 2015

Y1 - 2015

N2 - This work extends a numerical geodynamic modelling code (I2VIS) to simulate subduction of carbonated lithologies (altered basalts and carbonated sediments) into the mantle. Code modifications now consider devolatilisation of H2O-CO2 fluids, a CO2-melt solubility parameterisation for molten sediments, and allows for carbonation of mantle peridotites. The purpose is to better understand slab generated CO2 fluxes and consequent subduction of carbonates into the deep mantle via numerical simulation. Specifically, we vary two key model parameters: 1) slab convergence rate (1,2,3,4,5 cm y􀀀1) and 2) converging oceanic slab age (20,40,60,80 Ma) based on a half-space cooling model. The aim is to elucidate the role subduction dynamics has (i.e., spontaneoussedimentary diapirism, slab roll-back, and shear heating) with respect to slab decarbonation trends not entirelycaptured in previous experimental and thermodynamic investigations. This is accomplished within a fully coupledpetrological-thermomechanical modelling framework utilising a characteristics-based marker-in-cell techniquecapable of solving visco-plastic rheologies. The thermodynamic database is modified from its original state toreflect the addition of carbonate as CO2 added to the rock’s overall bulk composition. Modifications to original lithological units and volatile bulk compositions are as follows: GLOSS average sediments (H2O: 7.29 wt% &CO2: 3.01 wt%), altered basalts (H2O: 2.63 wt% & CO2: 2.90 wt%), and metasomatised peridotite (H2O: 1.98wt% & CO2: 1.5 wt%). We resolve stable mineralogy and extract rock properties via Perple_X at a resolutionof 5K and 25 MPa. Devolatilisation/consumption and stability of H2O-CO2 fluid is determined by accessing the thermodynamic database. When fluid is released due to unstable conditions, it is tracked via markers that freely advect within the velocity field until consumed.

AB - This work extends a numerical geodynamic modelling code (I2VIS) to simulate subduction of carbonated lithologies (altered basalts and carbonated sediments) into the mantle. Code modifications now consider devolatilisation of H2O-CO2 fluids, a CO2-melt solubility parameterisation for molten sediments, and allows for carbonation of mantle peridotites. The purpose is to better understand slab generated CO2 fluxes and consequent subduction of carbonates into the deep mantle via numerical simulation. Specifically, we vary two key model parameters: 1) slab convergence rate (1,2,3,4,5 cm y􀀀1) and 2) converging oceanic slab age (20,40,60,80 Ma) based on a half-space cooling model. The aim is to elucidate the role subduction dynamics has (i.e., spontaneoussedimentary diapirism, slab roll-back, and shear heating) with respect to slab decarbonation trends not entirelycaptured in previous experimental and thermodynamic investigations. This is accomplished within a fully coupledpetrological-thermomechanical modelling framework utilising a characteristics-based marker-in-cell techniquecapable of solving visco-plastic rheologies. The thermodynamic database is modified from its original state toreflect the addition of carbonate as CO2 added to the rock’s overall bulk composition. Modifications to original lithological units and volatile bulk compositions are as follows: GLOSS average sediments (H2O: 7.29 wt% &CO2: 3.01 wt%), altered basalts (H2O: 2.63 wt% & CO2: 2.90 wt%), and metasomatised peridotite (H2O: 1.98wt% & CO2: 1.5 wt%). We resolve stable mineralogy and extract rock properties via Perple_X at a resolutionof 5K and 25 MPa. Devolatilisation/consumption and stability of H2O-CO2 fluid is determined by accessing the thermodynamic database. When fluid is released due to unstable conditions, it is tracked via markers that freely advect within the velocity field until consumed.

M3 - Abstract

ER -