Fluid production from dehydration reactions and fluid migration in the subducting slab impact various subduction processes, including intraslab and megathrust earthquakes, episodic slip and tremor, mantle wedge metasomatism, and arc-magma genesis. Quantifying those processes requires a good knowledge of the location and amount of fluid outflux at the top of the slab. Previous models of fluid migration indicate that compaction-pressure gradients induced by the dehydration reactions could drive updip intraslab fluid flow (Wilson et al., 2014). However, how the initial hydration in the oceanic mantle prior to subduction impacts the updip fluid flow has not been investigated. Here, we use a 2-D two-phase flow model to investigate this effect under various initial slab-mantle hydration states and slab thermal conditions, both of which impact the depth extent of the stability of hydrous minerals. We focus on the lateral shift between the site of dehydration reactions and the location of fluid outflux at the top of the slab due to intraslab-updip migration. Our results indicate that major updip fluid pathways form along the antigorite and chlorite dehydration fronts sub-parallel to the slab surface. This, in turn, promotes slab-fluid outflux at the slab surface as shallow as 30–40 km depths. This mechanism is more likely in young slabs (< ~30 Ma), in which the thickness of the hydrated mantle in the incoming oceanic mantle that is required to form the slab-parallel dehydration fronts is relatively small (< ~20 km) because of its warm condition and thus a relatively thin antigorite stability zone.

Mineral phase transformations significantly alter the bulk density and elastic properties of mantle rocks and consequently have profound effects on mantle dynamics and seismic wave propagation. These changes in the physical properties of mantle rocks result from evolution in the equilibrium mineralogical composition, which can be predicted by the minimization of the Gibbs Free Energy with respect to pressure (P), temperature (T), and chemical composition (X). Thus, numerical models that simulate mantle convection and/or probe the elastic structure of the Earth’s mantle must account for varying mineralogical compositions to be self-consistent. Yet coupling Gibbs Free Energy minimization (GFEM) approaches with numerical geodynamic models is currently intractable for high-resolution simulations because execution speeds of widely-used GFEM programs (100–102 ms) are impractical in many cases. As an alternative, this study introduces machine learning models (RocMLMs) that have been trained to predict thermodynamically self-consistent rock properties at arbitrary PTX conditions between 1–28 GPa, 773–2273 K, and mantle compositions ranging from fertile (lherzolitic) to refractory (harzburgitic) end-members defined with a large dataset of published mantle compositions. RocMLMs are 101–103 times faster than GFEM calculations or GFEM-based look-up table approaches with equivalent accuracy. Depth profiles of RocMLMs predictions are nearly indistinguishable from reference models PREM and STW105, demonstrating good agreement between thermodynamic-based predictions of density, Vp, and Vs and geophysical observations. RocMLMs are therefore capable, for the first time, of emulating dynamic evolution of density, Vp, and Vs in high-resolution numerical geodynamic models.