Towards the Habitable Worlds Observatory: 1D CNN Retrieval of Reflection Spectra from Evolving Earth Analogs

However, traditional atmospheric retrieval frameworks are too computationally intensive to explore the high-dimensional parameter spaces such missions will generate.

Here, we present a one-dimensional convolutional neural network (1D CNN), trained on over one million synthetic, noise-injected spectra simulating Archean, Proterozoic, and Modern Earth analogs, as observed by LUVOIR-B (0.2-2.0 μm) and HabEx/SS (0.2-1.8 μm).

Our model simultaneously infers six molecular abundances (including biosignatures O₂ and O₃ ) along with radius, gravity, surface pressure, and temperature. Inference on unseen test data is performed via Monte Carlo Dropout, enabling uncertainty estimation across thousands of realizations within seconds.

The network performs best where spectral features are prominent, accurately recovering CH₄ and CO₂ in Archean atmospheres and O₂ and O₃ in Modern cases, while avoiding false positives and outputting near-zero abundances in scenarios of true absence such as Archean O₂ and O₃ . Interpretation via Integrated Gradients confirms that the model bases its predictions on physically meaningful features, including the Fraunhofer A band for O₂, and the Hartley-Huggins band for O₃ .

Credibility curve analysis indicates that O₃ remains retrievable across a wide range of stellar types and distances, while O₂ is detectable out to 12 pc around FG stars. These results elevate the CNN from proof of concept to a mission-ready retrieval engine, capable of processing direct-imaging spectra with HWO on an operational cadence.

Sarah G. A. Barbosa, Raissa Estrela, Paulo C. F. da Silva Filho, Daniel B. de Freitas

Comments: 19 pages, 9 figures, submitted to MNRAS
Subjects: Earth and Planetary Astrophysics (astro-ph.EP)
Cite as: arXiv:2508.00076 [astro-ph.EP](or arXiv:2508.00076v1 [astro-ph.EP] for this version)
https://doi.org/10.48550/arXiv.2508.00076
Focus to learn more
Submission history
From: Sarah Barbosa
[v1] Thu, 31 Jul 2025 18:10:12 UTC (4,445 KB)
https://arxiv.org/abs/2508.00076
Astrobiology, exoplanet,