Multi-level Monte Carlo Finite Volume methods for non-linear systems of conservation laws in multi-dimensions with random initial data

Abstract

We extend the Multi-Level Monte Carlo (MLMC) algorithm of [19] in order to quantify uncertainty in the solutions of multi-dimensional hyperbolic systems of conservation laws with uncertain initial data. The algorithm is presented and several issues arising in the massively parallel numerical implementation are addressed. In particular, we present a novel load balancing procedure that ensures scalability of the MLMC algorithm on massively parallel hardware. A new code ALSVID-UQ is described and applied to simulate uncertain solutions of the Euler equations and ideal MHD equations. Numerical experiments showing the robustness, eciency and scalability of the proposed algorithm are presented.