Mass-Conserving Thermal Structure for Slabs in Instantaneous Models of Subduction

Subduction is driven by difference in mass between the sinking plate and the surrounding mantle. The deformation calculated in numerical models of subduction is strongly dependent on the magnitude of the mass difference. The mass difference depends on the temperature of the slab. As the tectonic plate sinks it heats up, but it also cools down the surrounding mantle. The amount of heating and cooling is determined by conservation of thermal energy. Because the temperature also determines the thermal mass, conversing thermal energy also leads to conserving mass. For some studies, models of subduction are made to match the present day structure of a sinking plate. In this case, the temperature is defined to follow the observed geometry. In some previous studies, the temperature structure did not explicitly enforce conservation of energy or mass, and thus the density of the slab was not physically consistent, which is added source of uncertainty when analyzing the resulting flow and sensitivity of model results to mantle and slab rheology. Here we present a mass-conserving thermal structure for slabs that also creates a smoothly varying temperature structure. The thermal structure is based on a 1-D half-space cooling model (bottom) and an infinite space cooling model (top). It uses the age of the plate at the trench to determine the initial mass anomaly of the slab. The sinking velocity modifies the rate of heating and migration of the minimum temperature into the slab interior. The thermal model is calibrated against simple 2D subduction models in which the age and subduction velocity are held fixed. The new thermal structure has been implemented in the Geodynamic WorldBuilder (1), which can be used with different mantle convection software and is distributed as a plugin for ASPECT (2). Comparison of model results with the mass conserving slab thermal structure to the "plate" model from McKenzie (1970) is used to illustrate the differences in modeled results. References: 1. Fraters, M. R. T. et al., Solid earth, 2019. 2. Bangerth, W. et al., https://doi.org/10.5281/ZENODO.5131909, 2021.