Numerical simulation of the mantle convection
Please login to view abstract download link
The numerical investigation of the mantle convection plays a crucial role in understanding Earth’s dynamic processes and addressing geophysical challenges. However, such simulations suffer of two main drawbacks: i) they are computationally intensive and ii) are affected by some sources of uncertainties in the input parameters as well as in the model formulations. This study explores the state of the art of Reduced Order Models (ROMs) and Uncertainty Quantification (UQ) to address the two issues above. We consider different models transitioning from simpler incompressible Stokes flows with the Boussinesq approximation to a fully compressible Stokes formulation [1]. By analyzing the interplay between UQ and ROMs, this research establishes a robust pipeline for the numerical modelling of geophysical phenomena.