Spherical Harmonic Models of Observed Dynamic Topography

Abstract:

Mantle convective simulations are often used to predict present-day dynamic topography on Earth. Most of these models suggest that dynamic topography is dominated by degree 2 and 3 patterns which have peak amplitudes of 1–2 km. In order to test the applicability of these results, it is useful to construct an accurate global observational database. We have collated ~1000 seismic reflection profiles and ~500 wide-angle refraction experiments from the oceanic realm. This dataset can be used to calculate residual depth anomalies with respect to the well-known age-depth cooling relationship by carefully taking sedimentary and crustal loading effects into account. Resulting anomalies have wavelengths of 10³–10⁴ km with typical amplitudes of ±1 km. Average uncertainties are ±150 meters. We have combined these oceanic residual depths with onshore estimates from GRACE gravity anomalies to generate a spherical harmonic map of present-day dynamic topography. The resultant power spectrum is significantly less red than most predictive models. In other words, there is significantly less power at degrees 2 and 3 and much greater power at degrees 20 to 30 (i.e. wavelengths of 2000–1300 km). This mismatch implies that predictive models are currently compromised by limited resolution of mantle density structure and weak constraints on viscosity variation. Inclusion of accurate shallow mantle structure, which is the likely source of dynamic topography at degrees 20–30, is particularly important for future convective simulations.