LARGE-SCALE OPEN-SYSTEM BEHAVIOR OF CARBON DIOXIDE IN THE CONTINENTAL LITHOSPHERE DEDUCED FROM CLOSED-SYSTEM MODELING OF METAMORPHIC PHASE EQUILIBRIA IN THE WEPAWAUG SCHIST, CT

The impact of the global carbon cycle on Earth’s climate has long been recognized, yet an important flux – the addition of metamorphic CO₂ to the atmosphere – remains poorly constrained. Decarbonation of metamorphic rocks is greatly increased if water-rich fluids infiltrate during prograde reaction. Thus, the extent and significance of open-system processes is fundamental to constraining large scale orogenic CO₂ fluxes. One of John Ferry’s seminal contributions was demonstrating that fluid infiltration was essential to drive decarbonation and reaction progress in northern New England’s metasedimentary sequences. We seek to build on these results with a new computational approach.

We present results of thermodynamic modeling of metamorphosed (Acadian orogeny) carbonate rocks from the Wepawaug Schist, CT. The ~10 km thick unit consists of metapelites ranging in grade from chlorite up to the staurolite-kyanite zone, and metacarbonates characterized by ankerite, biotite, amphibole, and diopside zones with increasing grade. We calculate two kinds of pseudosections from bulk compositions measured in low-grade precursor rocks: (1) closed-system (no infiltration) metamorphism and (2) metamorphism buffered by an H₂O-CO₂ fluid. By varying the X_CO2 of this fluid, we explore the effect of infiltration on phase relations. Thus we utilize closed-system modeling to test whether open-system processes were necessary to produce the observed mineral assemblages.

Our results indicate that the purely closed-system approach consistently makes grossly inaccurate predictions of CO₂ loss and mineral assemblages, and that an H₂O-rich fluid – derived from dewatering metaclastic rocks or degassing intrusions – must have been present during metamorphism. For example, closed-system models predict that ankerite will be stable up to ~620 ˚C at 8 kbar, yet the ankerite-out isograd formed between ~500 and ~530 ˚C in nature. On the other hand, models buffered by H₂O-rich fluids of X_CO2 ~0.05–0.1 are in excellent agreement with observed field relationships, including the succession of prograde mineral zones. Closed-system models of devolatilization will likely severely underestimate orogenic CO₂ fluxes in metaclastic-metacarbonate sequences, necessitating consideration of open-system volatile transport and reaction.