Graph-based Neural Space Weather Forecasting

Daniel Holmberg, Ivan Zaitsev, Markku Alho, Ioanna Bouri, Fanni Franssila +3 more
9/23/2025
physics.space-phcs.LGphysics.plasm-ph

No official code found. Best implementation: fmihpc/spacecast (13 stars, MIT, last commit 2026-02-14).

Abstract

Accurate space weather forecasting is crucial for protecting our increasingly digital infrastructure. Hybrid-Vlasov models, like Vlasiator, offer physical realism beyond that of current operational systems, but are too computationally expensive for real-time use. We introduce a graph-based neural emulator trained on Vlasiator data to autoregressively predict near-Earth space conditions driven by an upstream solar wind. We show how to achieve both fast deterministic forecasts and, by using a generative model, produce ensembles to capture forecast uncertainty. This work demonstrates that machine learning offers a way to add uncertainty quantification capability to existing space weather prediction systems, and make hybrid-Vlasov simulation tractable for operational use.

View on arXivView PDF

Code Implementations(1)

fmihpc/spacecastCommunity68%
MIT

Graph-based neural space weather forecasting

131PythonOct 2, 20248 months agoMIT

Similar Papers

Deterministic and probabilistic neural surrogates of global hybrid-Vlasov simulations

89% match2026

A Machine Learning-Ready Data Processing Tool for Near Real-Time Forecasting

86% match2025

Probabilistic Solar Proxy Forecasting with Neural Network Ensembles

85% match2023

Prediction of Space Weather Events through Analysis of Active Region Magnetograms using Convolutional Neural Network

84% match2024

Data driven weather forecasts trained and initialised directly from observations

82% match2024

Physics-driven machine learning for the prediction of coronal mass ejections' travel times

82% match2023

Cite this paper

@article{holmberg2025graphbased,
  title  = {Graph-based Neural Space Weather Forecasting},
  author = {Daniel Holmberg and Ivan Zaitsev and Markku Alho and Ioanna Bouri and Fanni Franssila and Haewon Jeong and Minna Palmroth and Teemu Roos},
  year   = {2025},
  eprint = {2509.19605},
  archivePrefix = {arXiv},
  url    = {http://arxiv.org/abs/2509.19605v2}
}

Add a badge to your README

Shows whether official code is linked for this paper.

Code of Paper badge
[![Code of Paper](https://codeofpaper.com/badge/2509.19605.svg)](https://codeofpaper.com/paper/2509.19605)

Discussion