Deep learning for Lagrangian drift simulation at the sea surface
This work addresses computational and accuracy limitations in simulating ocean surface drift, which is important for oceanography and environmental monitoring, but it appears incremental as it builds on existing deep learning and Fokker-Planck representations.
The paper tackled Lagrangian drift simulation in geophysical dynamics by introducing DriftNet, a deep learning architecture that reduces computational complexity and error propagation compared to state-of-the-art methods, demonstrating its relevance in numerical experiments for sea surface simulations.
We address Lagrangian drift simulation in geophysical dynamics and explore deep learning approaches to overcome known limitations of state-of-the-art model-based and Markovian approaches in terms of computational complexity and error propagation. We introduce a novel architecture, referred to as DriftNet, inspired from the Eulerian Fokker-Planck representation of Lagrangian dynamics. Numerical experiments for Lagrangian drift simulation at the sea surface demonstrates the relevance of DriftNet w.r.t. state-of-the-art schemes. Benefiting from the fully-convolutional nature of Drift-Net, we explore through a neural inversion how to diagnose modelderived velocities w.r.t. real drifter trajectories.