_version_ 1866908497277353984
author Roy, Sujit
Schmude, Johannes
Lal, Rohit
Gaur, Vishal
Freitag, Marcus
Kuehnert, Julian
van Kessel, Theodore
Hegde, Dinesha V.
Muñoz-Jaramillo, Andrés
Jakubik, Johannes
Vos, Etienne
Mandal, Kshitiz
Asanjan, Ata Akbari
Almeida, Joao Lucas de Sousa
Lin, Amy
Singh, Talwinder
Yang, Kang
Pandey, Chetraj
Hong, Jinsu
Aydin, Berkay
Kurth, Thorsten
McGranaghan, Ryan
Kasapis, Spiridon
Upendran, Vishal
Bahauddin, Shah
da Silva, Daniel
Pogorelov, Nikolai V.
Spalding, Anne
Watson, Campbell
Maskey, Manil
Guhathakurta, Madhulika
Bernabe-Moreno, Juan
Ramachandran, Rahul
author_facet Roy, Sujit
Schmude, Johannes
Lal, Rohit
Gaur, Vishal
Freitag, Marcus
Kuehnert, Julian
van Kessel, Theodore
Hegde, Dinesha V.
Muñoz-Jaramillo, Andrés
Jakubik, Johannes
Vos, Etienne
Mandal, Kshitiz
Asanjan, Ata Akbari
Almeida, Joao Lucas de Sousa
Lin, Amy
Singh, Talwinder
Yang, Kang
Pandey, Chetraj
Hong, Jinsu
Aydin, Berkay
Kurth, Thorsten
McGranaghan, Ryan
Kasapis, Spiridon
Upendran, Vishal
Bahauddin, Shah
da Silva, Daniel
Pogorelov, Nikolai V.
Spalding, Anne
Watson, Campbell
Maskey, Manil
Guhathakurta, Madhulika
Bernabe-Moreno, Juan
Ramachandran, Rahul
contents Heliophysics is central to understanding and forecasting space weather events and solar activity. Despite decades of high-resolution observations from the Solar Dynamics Observatory (SDO), most models remain task-specific and constrained by scarce labeled data, limiting their capacity to generalize across solar phenomena. We introduce Surya, a 366M parameter foundation model for heliophysics designed to learn general-purpose solar representations from multi-instrument SDO observations, including eight Atmospheric Imaging Assembly (AIA) channels and five Helioseismic and Magnetic Imager (HMI) products. Surya employs a spatiotemporal transformer architecture with spectral gating and long--short range attention, pretrained on high-resolution solar image forecasting tasks and further optimized through autoregressive rollout tuning. Zero-shot evaluations demonstrate its ability to forecast solar dynamics and flare events, while downstream fine-tuning with parameter-efficient Low-Rank Adaptation (LoRA) shows strong performance on solar wind forecasting, active region segmentation, solar flare forecasting, and EUV spectra. Surya is the first foundation model in heliophysics that uses time advancement as a pretext task on full-resolution SDO data. Its novel architecture and performance suggest that the model is able to learn the underlying physics behind solar evolution.
format Preprint
id arxiv_https___arxiv_org_abs_2508_14112
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle Surya: Foundation Model for Heliophysics
Roy, Sujit
Schmude, Johannes
Lal, Rohit
Gaur, Vishal
Freitag, Marcus
Kuehnert, Julian
van Kessel, Theodore
Hegde, Dinesha V.
Muñoz-Jaramillo, Andrés
Jakubik, Johannes
Vos, Etienne
Mandal, Kshitiz
Asanjan, Ata Akbari
Almeida, Joao Lucas de Sousa
Lin, Amy
Singh, Talwinder
Yang, Kang
Pandey, Chetraj
Hong, Jinsu
Aydin, Berkay
Kurth, Thorsten
McGranaghan, Ryan
Kasapis, Spiridon
Upendran, Vishal
Bahauddin, Shah
da Silva, Daniel
Pogorelov, Nikolai V.
Spalding, Anne
Watson, Campbell
Maskey, Manil
Guhathakurta, Madhulika
Bernabe-Moreno, Juan
Ramachandran, Rahul
Solar and Stellar Astrophysics
Instrumentation and Methods for Astrophysics
Artificial Intelligence
Heliophysics is central to understanding and forecasting space weather events and solar activity. Despite decades of high-resolution observations from the Solar Dynamics Observatory (SDO), most models remain task-specific and constrained by scarce labeled data, limiting their capacity to generalize across solar phenomena. We introduce Surya, a 366M parameter foundation model for heliophysics designed to learn general-purpose solar representations from multi-instrument SDO observations, including eight Atmospheric Imaging Assembly (AIA) channels and five Helioseismic and Magnetic Imager (HMI) products. Surya employs a spatiotemporal transformer architecture with spectral gating and long--short range attention, pretrained on high-resolution solar image forecasting tasks and further optimized through autoregressive rollout tuning. Zero-shot evaluations demonstrate its ability to forecast solar dynamics and flare events, while downstream fine-tuning with parameter-efficient Low-Rank Adaptation (LoRA) shows strong performance on solar wind forecasting, active region segmentation, solar flare forecasting, and EUV spectra. Surya is the first foundation model in heliophysics that uses time advancement as a pretext task on full-resolution SDO data. Its novel architecture and performance suggest that the model is able to learn the underlying physics behind solar evolution.
title Surya: Foundation Model for Heliophysics
topic Solar and Stellar Astrophysics
Instrumentation and Methods for Astrophysics
Artificial Intelligence
url https://arxiv.org/abs/2508.14112