T306 Subduction, Fluids, Accessory Minerals, and Trace Elements: A Celebration of Sorena Sorensen's Career (Posters)
Wednesday, 4 November 2015
9:00 AM - 6:30 PM
Exhibit Hall
P-T PATHS FROM SYROS, GREECE, AND CONSTRAINTS ON SUBDUCTION ZONE FLUID GENERATION
Paper number: T306-12
Program: Technical Sessions
Time: 9:00 AM – 6:30 PM
Authors
-
M.J. CaddickVirginia Tech, Blacksburg, VA
-
Ethan BaxterActon, MA
-
Kyle T. AshleyVirginia Tech, Blacksburg, VA
-
Jamie A. KendallBoston College, Chestnut Hill, MA
-
Besim DragovicVirginia Polytechnic Institute and State University, Blacksburg, VA
-
Hanna L. BrooksVirginia Polytechnic Institute and State University, Blacksburg, VA
-
Evan J. RamosBoston University, Boston, MA
Abstract
The purpose of this work is to build detailed P-T paths for blueschist and eclogite facies rocks from Syros, using P-T pseudosections and quartz in garnet Raman barometry when applicable. Phase fractionation models constructed stepwise along these calculated P-T paths are then used to predict the volume and composition of fluid evolved during subduction, subject to uncertainties about the thermodynamic properties of high-P fluids. We constructed P-T paths for (i) a garnet bearing blueschist with abundant lawsonite pseudomorphs and (ii) a micaceous eclogite with millimeter sized garnet crystals. Results suggest that garnet first grew in the blueschist at ~ 500˚C and 1.8–2.2 GPa, with peak metamorphism at ~ 550˚C and 2.4 GPa. In the eclogite, garnets grew from ~550˚C and 2.2GPa with peak metamorphism at about 600˚C and above 2.4GPa. Models that progressively fractionate garnet and fluid during prograde evolution predict substantial fluid release with chlorite breakdown in the blueschist and small changes in the fluid H2O/CO2 ratio over the prograde history. In the eclogite, fluid release is dominated by the breakdown of lawsonite and amphibole. Differences between the location of thermodynamically-predicted garnet-in reactions and the P-T of initial garnet growth implied by preserved crystal compositions suggest reaction overstepping of 40-100˚C, implying that these samples may have spent a sizable part of their history removed from sample-wide equilibrium.