L. Fery | Searcharxiv

arXiv2025

A Statistical Learning Approach to Mediterranean Cyclones

Mediterranean cyclones are extreme meteorological events of which much less is known compared to their tropical, oceanic counterparts. The raising interest in such phenomena is due to their impact on a region increasingly more affected by climate change, but a precise characterization remains a non trivial task. In this work we showcase how a Bayesian algorithm (Latent Dirichlet Allocation) can classify Mediterranean cyclones relying on wind velocity data, leading to a drastic dimensional reduction that allows the use of supervised statistical learning techniques for detecting and tracking new cyclones.

physics.ao-ph↗

arXiv2021

A model of interacting Navier-Stokes singularities

We introduce a model of interacting singularities of Navier-Stokes, named pin\,cons. They follow a Hamiltonian dynamics, obtained by the condition that the velocity field around these singularities obeys locally Navier-Stokes equations. This model can be seen of a generalization of the vorton model of Novikov, that was derived for the Euler equations. When immersed in a regular field, the pin\,cons are further transported and sheared by the regular field, while applying a stress onto the regular field, that becomes dominant at a scale that is smaller than the Kolmogorov length. We apply this model to compute the motion of a dipole of pin\,cons. When the initial relative orientation of the dipole is inside the interval (0, pi/2), a dipole made of pin\,con of same intensity exhibits a transient collapse stage, following a scaling with dipole radius tending to 0 like (tc - t) power 0.63. For long time, the dynamics of the dipole is however repulsive, with both components running away from each other to infinity.

physics.flu-dyn↗