Propagation of Uncertainties in Density-Driven Flow

Accurate modeling of contamination in subsurface flow and water aquifers is crucial for agriculture environmental protection. Here, we demonstrate a parallel algorithm to quantify the propagation uncertainty dispersal pollution flow. Specifically, consider density-driven estimate how from permeability porosity propagates solution. We take two-dimensional Elder-like problem as numerical benchmark, use random fields model our limited knowledge on permeability. well-known low-cost generalized polynomial chaos (gPC) expansion surrogate model, where gPC coefficients are computed by projection sparse tensor grids. The solver deterministic based multigrid method run parallel. Computation high-dimensional integrals over parametric space done too.