Data Assimilation with Machine Learning Surrogate Models: A Case Study with FourCastNet

Modern data-driven surrogate models for weather forecasting provide accurate short-term predictions but inaccurate and nonphysical long-term forecasts. This paper investigates online weather prediction using machine learning surrogates supplemented with partial and noisy observations. We empirically...

Full description

Saved in:
Bibliographic Details
Main Authors Adrian, Melissa, Sanz-Alonso, Daniel, Willett, Rebecca
Format Journal Article
LanguageEnglish
Published 21.05.2024
Subjects
Computer Science - Learning
Physics - Atmospheric and Oceanic Physics
Physics - Chaotic Dynamics
Statistics - Applications
Online AccessGet full text

Cover

More Information
Summary:Modern data-driven surrogate models for weather forecasting provide accurate short-term predictions but inaccurate and nonphysical long-term forecasts. This paper investigates online weather prediction using machine learning surrogates supplemented with partial and noisy observations. We empirically demonstrate and theoretically justify that, despite the long-time instability of the surrogates and the sparsity of the observations, filtering estimates can remain accurate in the long-time horizon. As a case study, we integrate FourCastNet, a state-of-the-art weather surrogate model, within a variational data assimilation framework using partial, noisy ERA5 data. Our results show that filtering estimates remain accurate over a year-long assimilation window and provide effective initial conditions for forecasting tasks, including extreme event prediction.
DOI:10.48550/arxiv.2405.13180