Fig. 6: Illustration of method to choose flow law parameters.
From: Rapid subduction initiation and magmatism in the Western Pacific driven by internal vertical forces
a The flow laws used in this study, for dislocation creep and diffusion creep, are plotted in red. Experimentally determined flow laws for hydrated peridotite, commonly used for mantle material in geodynamic numerical models, are plotted in black for reference. The studies that determined these diffusion creep flow laws are (listed from strongest to weakest at 1000 °C): Hirth and Kohlstedt43, Jaoul et al.44, and Houlier et al.45. Similarly, the studies for these dislocation creep flow laws: Kirby and Kronenberg46 and Hirth and Kohlstedt43 Karato and Wu47. b We use an approximate Arrhenius formulation of the exponential flow law for Peierl’s creep of hydrated peridotite, determined by Katayama and Karato36 (as our modelling technique does not allow for exponential flow laws). We used a parameter grid search to minimise the total misfit between our flow law and the flow law of Katayama and Karato36 in temperature-stress space (using our yield stress, or 8 × 108 Pa, as the maximum stress). These panels show the viscosity as given by Katayama and Karato36; the viscosity used in our study; and the minimised misfit.
