Exoplanetology: Exoplanets & Exomoons

Utilizing Machine Learning to Predict Host Stars and the Key Elemental Abundances of Small Planets

By Keith Cowing

Status Report

astro-ph.EP

March 5, 2025

Filed under AI, astro-ph.EP, astronomy, exoplanet, Habitable Worlds Observatory, Hypatia Catalog, imaging, JWST, Machine learning, NASA Exoplanet Archive, Roman Space Telescope, Stellar Cartography, sub-Neptune, super-Earth, XGBoost

Utilizing Machine Learning to Predict Host Stars and the Key Elemental Abundances of Small Planets — The [Na/Fe] vs. [Fe/H] abundance plot (Ensemble 4) showing the stars that are most likely to host a planet (green), the known planet hosts (blue), and the stars less likely to host a planet (orange). Histograms are also included that show the [Na/Fe] relative distribution to the right of the scatterplot and [Fe/H] relative distribution at the top. We note the clear sub-solar Na abundances for each experiment’s known and predicted planet hosts (see Section 5). – Small planets — astro-ph.EP

Stars and their associated planets originate from the same cloud of gas and dust, making a star’s elemental composition a valuable indicator for indirectly studying planetary compositions.

While the connection between a star’s iron (Fe) abundance and the presence of giant exoplanets is established (e.g. Gonzalez 1997; Fischer & Valenti 2005), the relationship with small planets remains unclear. The elements Mg, Si, and Fe are important in forming small planets.

Employing machine learning algorithms like XGBoost, trained on the abundances (e.g., the Hypatia Catalog, Hinkel et al. 2014) of known exoplanet-hosting stars (NASA Exoplanet Archive), allows us to determine significant “features” (abundances or molar ratios) that may indicate the presence of small planets.

We test on three groups of exoplanets: (a) all small, R_P < 3.5 R_⊕, (b) sub-Neptunes, 2.0 R_⊕ < RP < 3.5 R_⊕, and (c) super-Earths, 1.0 R_⊕ < RP < 2.0 R_⊕ — each subdivided into 7 ensembles to test different combinations of features. We created a list of stars with ≥90% probability of hosting small planets across all ensembles and experiments (“overlap stars”). We found abundance trends for stars hosting small planets, possibly indicating star-planet chemical interplay during formation.

We also found that Na and V are key features regardless of planetary radii. We expect our results to underscore the importance of elements in exoplanet formation and machine learning’s role in target selection for future NASA missions: e.g., the James Webb Space Telescope (JWST), Nancy Grace Roman Space Telescope (NGRST), Habitable Worlds Observatory (HWO) — all of which are aimed at small planet detection.

Amílcar R. Torres-Quijano, Natalie R. Hinkel, Caleb H. Wheeler III, Patrick A. Young, Luan Ghezzi, Augusto P. Baldo

Comments: 22 pages, 9 figures, 3 tables, accepted to AJ
Subjects: Earth and Planetary Astrophysics (astro-ph.EP); Instrumentation and Methods for Astrophysics (astro-ph.IM); Solar and Stellar Astrophysics (astro-ph.SR); Machine Learning (cs.LG)
Cite as: arXiv:2502.17563 [astro-ph.EP](or arXiv:2502.17563v1 [astro-ph.EP] for this version)
https://doi.org/10.48550/arXiv.2502.17563
Focus to learn more
Submission history
From: Amílcar Torres-Quijano
[v1] Mon, 24 Feb 2025 19:00:02 UTC (1,363 KB)
https://arxiv.org/abs/2502.17563
Astrobiology,

Keith Cowing

Explorers Club Fellow, ex-NASA Space Station Payload manager/space biologist, Away Teams, Journalist, Lapsed climber, Synaesthete, Na’Vi-Jedi-Freman-Buddhist-mix, ASL, Devon Island and Everest Base Camp veteran, (he/him) 🖖🏻

Follow on Twitter