Deep Learning for Lagrangian Drift Simulation at The Sea Surface
Daria Botvynko, Carlos Granero-Belinchón, Simon Van Gennip, Abdesslam Benzinou, Ronan Fablet
Abstract
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 convolutional nature of DriftNet, we explore through a neural inversion how to diagnose model-derived velocities w.r.t. real drifter trajectories.
BibTeX
@inproceedings{icassp2023_deeplearningforl,
title = {Deep Learning for Lagrangian Drift Simulation at The Sea Surface},
author = {Daria Botvynko and Carlos Granero-Belinchón and Simon Van Gennip and Abdesslam Benzinou and Ronan Fablet},
booktitle = {ICASSP 2023},
year = {2023}
}