S-to-P Heterogeneity Ratio in the Lower Mantle and Thermo-Chemical Implications

Abstract:

Seismic models show high values of the heterogeneity ratio R_S/P=δlnV_S /δlnV_P in the lower mantle. Based on the analysis of our new V_S−V_P model SPani, we show that estimated values are biased by data resolution, model parameterization and regularization choices. Spherical-average R_S/P reduces from around 2.7 to 2.0, at 2500 km depth, if we apply different roughness damping to V_P which does not compromise data fit and maintains unchanged the main structural features of the model. In agreement with previous studies, our model has relatively high R_S/P in large low shear-velocity provinces, but a similar increasing trend is also observed in anomalously positive regions. We are unable to find any plausible physical explanation for this trend. Boosting the V_P anomalies of 20% drastically reduces R_S/P without significantly degrading the data fit. We conclude that lateral compositional variations are not required to explain seismic data. Our result also highlights the limitations of tomography images to directly extract important diagnostic compositional parameters as R_S/P.