Learning to Advect: A Neural Semi-Lagrangian Architecture for Weather Forecasting

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Pereira, Carlos A., Gaudreault, Stéphane, Dallerit, Valentin, Subich, Christopher, Panday, Shoyon, Wei, Siqi, Zhang, Sasa, Rout, Siddharth, Haber, Eldad, Spiteri, Raymond J., Millard, David, Diaconescu, Emilia
Format: Preprint
Published: 2026
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866913128196866048
author Pereira, Carlos A.
Gaudreault, Stéphane
Dallerit, Valentin
Subich, Christopher
Panday, Shoyon
Wei, Siqi
Zhang, Sasa
Rout, Siddharth
Haber, Eldad
Spiteri, Raymond J.
Millard, David
Diaconescu, Emilia
author_facet Pereira, Carlos A.
Gaudreault, Stéphane
Dallerit, Valentin
Subich, Christopher
Panday, Shoyon
Wei, Siqi
Zhang, Sasa
Rout, Siddharth
Haber, Eldad
Spiteri, Raymond J.
Millard, David
Diaconescu, Emilia
contents Recent machine-learning approaches to weather forecasting often employ a monolithic architecture in which distinct physical mechanisms-advection (long-range transport), diffusion-like mixing, thermodynamic processes, and forcing-are represented implicitly within a single large network. This is particularly problematic for advection, where long-range transport typically requires expensive global interaction mechanisms or deep stacks of local convolutional layers. To mitigate this, we present PARADIS, a physics-inspired global weather prediction model that enforces inductive biases on network behavior through a functional decomposition into advection, diffusion, and reaction blocks acting on latent variables. We implement advection through a Neural Semi-Lagrangian operator that performs trajectory-based transport via differentiable interpolation on the sphere, enabling end-to-end learning of both the latent modes to be transported and their characteristic trajectories. Diffusion-like processes are modeled by depthwise-separable spatial mixing, whereas local source terms and vertical interactions are handled via pointwise channel interactions, yielding a physically structured operator decomposition. Evaluated on ERA5 benchmarks, PARADIS achieves competitive deterministic forecast skill, with particularly strong short-lead performance, while preserving substantially better spectral fidelity and forecast activity during medium-range rollouts.
format Preprint
id arxiv_https___arxiv_org_abs_2601_21151
institution arXiv
publishDate 2026
record_format arxiv
spellingShingle Learning to Advect: A Neural Semi-Lagrangian Architecture for Weather Forecasting
Pereira, Carlos A.
Gaudreault, Stéphane
Dallerit, Valentin
Subich, Christopher
Panday, Shoyon
Wei, Siqi
Zhang, Sasa
Rout, Siddharth
Haber, Eldad
Spiteri, Raymond J.
Millard, David
Diaconescu, Emilia
Machine Learning
Atmospheric and Oceanic Physics
Recent machine-learning approaches to weather forecasting often employ a monolithic architecture in which distinct physical mechanisms-advection (long-range transport), diffusion-like mixing, thermodynamic processes, and forcing-are represented implicitly within a single large network. This is particularly problematic for advection, where long-range transport typically requires expensive global interaction mechanisms or deep stacks of local convolutional layers. To mitigate this, we present PARADIS, a physics-inspired global weather prediction model that enforces inductive biases on network behavior through a functional decomposition into advection, diffusion, and reaction blocks acting on latent variables. We implement advection through a Neural Semi-Lagrangian operator that performs trajectory-based transport via differentiable interpolation on the sphere, enabling end-to-end learning of both the latent modes to be transported and their characteristic trajectories. Diffusion-like processes are modeled by depthwise-separable spatial mixing, whereas local source terms and vertical interactions are handled via pointwise channel interactions, yielding a physically structured operator decomposition. Evaluated on ERA5 benchmarks, PARADIS achieves competitive deterministic forecast skill, with particularly strong short-lead performance, while preserving substantially better spectral fidelity and forecast activity during medium-range rollouts.
title Learning to Advect: A Neural Semi-Lagrangian Architecture for Weather Forecasting
topic Machine Learning
Atmospheric and Oceanic Physics
url https://arxiv.org/abs/2601.21151