Fig. 5

Stress-versus-strain data and best-fitting predictions of the ISV and nonlinear Burgers models for dunites in anhydrous conditions. a–c) The model fits to the data used by Sherburn et al. (2011). All the constants from \(C_1\) to \(C_{20}\) are from Table 1 of Sherburn et al. (2011), except for \(C_{13}=0.83\), which corrects a typographic error. The reduced chi-square value for the ISV and nonlinear Burgers models are 41.1 and 13.9, respectively, indicating that the nonlinear Burgers model performs better than ISV. d–f Comparison with data not used in the optimization. Reduced chi-square values for the ISV and nonlinear Burgers models on these data are 411.2 and 105.1, respectively. The nonlinear Burgers model, which uses fewer parameters compared to the ISV model, performs better