Graph neural networks for exoplanet atmospheres
Abstract
Calculating disequilibrium chemistry in exoplanet atmospheres remains a significant computational bottleneck in atmospheric retrievals. The increasing observational precision from facilities such as JWST and the Ariel mission requires including disequilibrium chemistry in these analyses. Previous studies have demonstrated that neural networks can emulate kinetic chemistry, although their spatial inductive bias does not align with the topology of chemical reaction networks. This study introduces a graph neural network surrogate that represents chemical species as nodes and temperature-dependent reaction rates as edges, thereby enabling information propagation along physically meaningful chemical pathways. The model is trained on atmospheres generated using the Venot+2020 chemical scheme and Guillot temperature-pressure profiles. The GNN accurately reconstructs disequilibrium abundances across the sampled parameter space and reduces the mean abundance error by a factor of approximately 3 compared to the previous U-Net model. When applied to transmission spectra, most predictions fall within the observational precision expected for JWST and Ariel, with only about 7% of test atmospheres exceeding a 20 ppm mean spectral error. Performance variations are primarily observed in chemically transitional regimes near a carbon-to-oxygen ratio of one and at low temperatures. An evaluation of the boundary-case planet WASP-39b demonstrates effective performance under a moderate domain shift. Perturbation analysis indicates that disturbances propagate along chemical connectivity rather than spatial adjacency, confirming that the architecture captures the structure of reaction networks. These results suggest that GNN surrogates provide accurate, computationally efficient predictions of disequilibrium chemistry, facilitating integration into the atmospheric retrieval pipeline.
Related
- Accelerating Chemical Kinetics for Exoplanet Atmospheres using Neural Networks
- Exploration of Exoplanet Atmospheres with the James Webb Space Telescope
- Using High-Resolution Spectroscopy to Study the Composition, Temperature, and Dynamics of Exoplanet Atmospheres
- Interplanetary Dust Ablation Seeds Silicate Clouds in Exoplanet Atmospheres
- Interplanetary Dust Ablation Seeds Silicate Clouds in Exoplanet Atmospheres
BibTeX
@misc{vojtekova2026graph,
title = {{Graph neural networks for exoplanet atmospheres}},
author = {Antonia Vojtekova and Kai Hou Yip and Ingo P. Waldmann and Nikolaos Nikolaou and Olivia Venot and Ahmed Faris Al-Refaie and Bruno Merín},
howpublished = {arXiv},
year = {2026},
month = sep,
doi = {10.48550/arxiv.2609.17894},
eprint = {2609.17894},
archivePrefix = {arXiv},
url = {https://arxiv.org/abs/2609.17894},
note = {CC BY 4.0},
}
RIS
TY - UNPB
TI - Graph neural networks for exoplanet atmospheres
AU - Antonia Vojtekova
AU - Kai Hou Yip
AU - Ingo P. Waldmann
AU - Nikolaos Nikolaou
AU - Olivia Venot
AU - Ahmed Faris Al-Refaie
AU - Bruno Merín
PY - 2026
DA - 2026/09/15
PB - arXiv
DO - 10.48550/arxiv.2609.17894
UR - https://arxiv.org/abs/2609.17894
AB - Calculating disequilibrium chemistry in exoplanet atmospheres remains a significant computational bottleneck in atmospheric retrievals. The increasing observational precision from facilities such as JWST and the Ariel mission requires including disequilibrium chemistry in these analyses. Previous studies have demonstrated that neural networks can emulate kinetic chemistry, although their spatial inductive bias does not align with the topology of chemical reaction networks. This study introduces a graph neural network surrogate that represents chemical species as nodes and temperature-dependent reaction rates as edges, thereby enabling information propagation along physically meaningful chemical pathways. The model is trained on atmospheres generated using the Venot+2020 chemical scheme and Guillot temperature-pressure profiles. The GNN accurately reconstructs disequilibrium abundances across the sampled parameter space and reduces the mean abundance error by a factor of approximately 3 compared to the previous U-Net model. When applied to transmission spectra, most predictions fall within the observational precision expected for JWST and Ariel, with only about 7% of test atmospheres exceeding a 20 ppm mean spectral error. Performance variations are primarily observed in chemically transitional regimes near a carbon-to-oxygen ratio of one and at low temperatures. An evaluation of the boundary-case planet WASP-39b demonstrates effective performance under a moderate domain shift. Perturbation analysis indicates that disturbances propagate along chemical connectivity rather than spatial adjacency, confirming that the architecture captures the structure of reaction networks. These results suggest that GNN surrogates provide accurate, computationally efficient predictions of disequilibrium chemistry, facilitating integration into the atmospheric retrieval pipeline.
ER -