Stochastic Ocean Energy Backscatter via Pressure and Momentum Perturbation

Oceanic dynamics, with their wide range of interacting scales, present challenges for accurate numerical modeling. This work focuses on improving the representation of mesoscale eddies at eddy-permitting resolutions by incorporating novel stochastic perturbations. Eddy viscosity, which leads to numerical over-dissipation near the grid scale, requires methods like dynamic backscatter to mitigate this effect. Stochastic perturbations help capture small-scale processes and uncertainties in ocean flows.

Using a double-gyre configuration, we apply linear inverse modeling with high-resolution reference simulations to generate stochastic perturbation patterns. We explore two stochastic forcing implementations based on different balanced flow constraints. Results show that stochastic forcing improves simulated dynamics, particularly heat distribution at intermediate and small scales, while previously implemented dynamic backscatter injects energy at larger scales. A scale analysis of production, advection, dissipation, and total energy supports these findings.