11:45 AM - 12:00 PM
[SEM12-10] Electrical feature of old oceanic mantle in the northwestern Pacific: Quantitative constraints on the thermal and compositional states
Keywords:magnetotellurics, electrical conductivity, oceanic lithosphere and asthenosphere, seafloor flattening
Oceanic upper mantle beneath the northwestern Pacific has large-scale lateral heterogeneity that is impossible to attribute to an age-dependency of the thermal structure based on a simple cooling of the lithosphere. This fact was revealed from seafloor magnetotelluric (MT) data collected in two areas, northwest (Area A) and southeast (Area B) of the Shatsky Rise, through Normal Oceanic Mantle Project (Baba et al., 2017). Electrical conductivity structure obtained by the MT data can be interpreted in terms of geothermal structure, direct (enhancing the conductivity) and indirect (reducing the mantle solidus and leading partial melting) effects of volatile components such as H2O and CO2 dissolved in the mantle rock. The consideration of the reduction of peridotite solidus due to H2O and CO2 and partition of them in solid mineral phases and melt can couple the parameters mutually and they can be constrained more by self-consistent manner although there are trade-off relations between the parameters.
We considered one-dimensional thermal structures based on three different cooling scenarios: 1) a plate cooling with age (PC), 2) a half-space cooling with apparent age representing rejuvenation (RJ), and 3) a cooling with small-scale convection (SSC), respectively, with a reasonable mantle adiabat. Then, for the PC scenario, five parameters controlling the electrical conductivity structure, which are the mantle potential temperature (TP), the thickness of thermally conductive plate (h), H2O and CO2 contents in the bulk mantle (CH2O and CCO2, respectively), and the crustal conductivity (σc), were investigated by a grid search inversion to reconstruct the MT responses in the two areas. For the RJ and SSC scenarios, the apparent age (ta) and Frank-Kamenetskii parameter (θ) that is related to the temperature dependency of the mantle viscosity are respectively introduced instead of h. h, ta, and θ are key parameters to reproduce the thickness of thermally conductive layer and corresponding electrically resistive layer. The grid search inversion approach guarantees that the preferred model parameters can explain the observed MT data within a statistical confidence limit and enable us to investigate the quantitative trade-off relations between the parameters.
From the inversion analyses, we found that the MT data for Area B can be explained by any cooling scenarios considered in the present study but the MT data for Area A can be explained only by the PC scenario. We quantitatively verified that the thickness of the thermally conductive layer significantly thinner for Area A than for Area B as qualitatively suggested by the previous study (Baba et al., 2017). With reasonable potential temperature (1350±30 degree C), small amount (<160 ppm) H2O is necessary, small amount (several tens ppm) of CO2 likely exists, and possible melt fraction is very small (<0.06 vol%) in the mantle beneath the areas, which is compatible with the depleted source mantle that is the origin of the normal oceanic crust.
Seafloor subsidence was reproduced for the cooling scenarios with the parameters acceptable in terms of MT observations and the predicted water depth was compared with the observed water depth in Area A and Area B. However, the consistency between the MT data and the water depths needs more careful examinations because there are many additional parameters to be assumed for reproducing the seafloor subsidence.
We considered one-dimensional thermal structures based on three different cooling scenarios: 1) a plate cooling with age (PC), 2) a half-space cooling with apparent age representing rejuvenation (RJ), and 3) a cooling with small-scale convection (SSC), respectively, with a reasonable mantle adiabat. Then, for the PC scenario, five parameters controlling the electrical conductivity structure, which are the mantle potential temperature (TP), the thickness of thermally conductive plate (h), H2O and CO2 contents in the bulk mantle (CH2O and CCO2, respectively), and the crustal conductivity (σc), were investigated by a grid search inversion to reconstruct the MT responses in the two areas. For the RJ and SSC scenarios, the apparent age (ta) and Frank-Kamenetskii parameter (θ) that is related to the temperature dependency of the mantle viscosity are respectively introduced instead of h. h, ta, and θ are key parameters to reproduce the thickness of thermally conductive layer and corresponding electrically resistive layer. The grid search inversion approach guarantees that the preferred model parameters can explain the observed MT data within a statistical confidence limit and enable us to investigate the quantitative trade-off relations between the parameters.
From the inversion analyses, we found that the MT data for Area B can be explained by any cooling scenarios considered in the present study but the MT data for Area A can be explained only by the PC scenario. We quantitatively verified that the thickness of the thermally conductive layer significantly thinner for Area A than for Area B as qualitatively suggested by the previous study (Baba et al., 2017). With reasonable potential temperature (1350±30 degree C), small amount (<160 ppm) H2O is necessary, small amount (several tens ppm) of CO2 likely exists, and possible melt fraction is very small (<0.06 vol%) in the mantle beneath the areas, which is compatible with the depleted source mantle that is the origin of the normal oceanic crust.
Seafloor subsidence was reproduced for the cooling scenarios with the parameters acceptable in terms of MT observations and the predicted water depth was compared with the observed water depth in Area A and Area B. However, the consistency between the MT data and the water depths needs more careful examinations because there are many additional parameters to be assumed for reproducing the seafloor subsidence.