How Large a Feedback Effect Does Slab Dewatering Have on Itself ?

Arc magmas are generally believed to be produced when the mantle wedge melts as a result of fluxing of a hydrous fluid from the subducting plate. Fluids liberate from the slab at different P-T conditions; key to understanding the fate of these fluids is the knowledge of the thermal structure of the downgoing plate. Earlier works have shown that this thermal structure is a function of several variables like the age of the incoming oceanic lithosphere, the convergence rate, and the dip angle. However, the impacts of chemical reactions and heat transport by fluid flow have yet to be thoroughly explored. Fluid fluxing from the slab results from metamorphic phase transitions which consume latent heat. Latent heats for the different reactions have been quantified in experimental studies. However, little is known how this cooling effect changes the timing, location, and intensity of fluid release. One way to explore this problem is to use numerical models, as previously done by Peacock et al. Here, we present results of a new self-consistent, chemo-thermo-dynamical model for mantle flow, melting, and fluid release. To solve the governing equations of the model we use a combined finite elements, finite differences, and tracer particle advection scheme. For proper internal consistency we include the cooling effects of fluid release within the temperature solution. In this study we analyze the impact of the cooling effect of metamorphic dehydration reactions on fluid release at subduction zones and water recycling into the deeper mantle. For this analysis, we divide the incoming plate into a crustal and mantle layer consisting primarily of hydrated basalts and hydrated peridotites, respectively. We then prescribe for each layer different values for the latent heats released during dewatering. In accordance to experimentally determined values, in a series of model runs, we gradually augment the chosen values for the latent heats from a minimal to a maximal cooling effect and analyze the impact of this on the timing, location, and intensity of water release. These numerical experiments provide new insight into the interactions between fluid release and latent heat consumption.