4:00 PM - 4:15 PM
[SCG45-29] Effects of fluid chemistry on metasomatic reactions at slab-mantle interfaces within the subduction zones: Significance of Si vs. Mg and CO2
Keywords:metasomatism, Mg vs. Si, talc, chlorite, CO2, crust-mantle boundary
We modeled the fluid-rock interaction at the slab-mantle interfaces by (1) incremental addition of mantle peridotite into aqueous fluid in aqueous fluids in equilibrium with pelite and (2) incremental addition of metasediment into aqueous fluid in aqueous fluids in equilibrium with peridotite. We calculated along three subduction zones; NE Japan, Cascadia and Nankai. Geochemical modeling reveals that in the shallow part of a subduction zone, the dissolved Si content of fluids in equilibrium with pelitic schist (CSi,crust) is significantly higher than the dissolved Mg content of fluids in equilibrium with mantle peridotite (CMg,mantle); however, CMg,mantle becomes dominant at depth, resulting in the Mg-metasomatism of crustal rocks to form chlorite rocks. This Mg-metasomatism is more widespread in warmer subduction zones (e.g., the Nankai and Cascadia subduction zones) than in colder subduction zones (e.g., in Northeast Japan). Such results of Mg- and Si-metasomatism are consistent with the natural observations that chlorite rocks are developed within the crustal rocks, whereas talc is only formed within serpentinite of the high-pressure metamorphic terranes such as Sanbagawa belt. In addition, the model predicts that subduction of the sediments produces the carbonates (dolomite and magnesite) as well as talc. The preliminary results of the experiments on the metasomatic reactions at the pelitic schist and peridotite/serpentinite indicates that at 500 degreeC and 1.0 GPa, the formation of talc can more occur with intensive fracturing by CO2-metasomatism rather than Si-metasomatism, which potentially makes drastic weakening of mantle wedge.
Tarling et al., (2019) Nature Geosci, 12:1034–1042.
Collido et al., (2022) Geochem Geophys Geosys 23:e2021GC10206.
Okamoto et al., (2021) Com Earth Env, 2:151. doi:10.1038/s43247-021-00224-5