ThousandWorlds: A benchmark for climate emulation of potentially habitable exoplanets

2026-06-16

AI Summary

This work addresses the challenge that existing machine learning approaches face in developing efficient exoplanet climate simulators due to the absence of multi-model exoplanetary climate datasets. To bridge this gap, the authors present the first machine learning benchmark dataset specifically designed for exoplanet climate simulation, integrating approximately 1,800 simulations from five distinct general circulation models (GCMs) that map eight planetary parameters to three-dimensional atmospheric fields. The benchmark introduces three progressively challenging evaluation tasks and a performance evaluation protocol based on inter-model discrepancies. It facilitates structured modeling and missing data handling in low-data, multi-simulator regimes. Among seven baseline methods evaluated, Gaussian processes consistently outperform others, suggesting that current general-purpose deep learning approaches do not yet hold an advantage in this domain.

Abstract

The search for life beyond Earth will depend on detecting faint signatures in the atmospheres of potentially habitable exoplanets. Interpreting those signatures requires understanding the host planet's climate: the same molecule may signal life on one planet and abiotic chemistry on another. Global climate models (GCMs) provide this understanding, but individual runs can require up to millions of core-hours and substantial domain expert time. Machine-learning emulators could remove this bottleneck, but progress has been limited by the absence of a curated, multi-model exoclimate dataset. We introduce ThousandWorlds, an ML-ready benchmark for exoclimate emulation and for the broader regime of low-data, multi-simulator, parameter-to-field regression. The dataset contains approximately 1800 simulations from five GCMs, mapping eight planet parameters to 3D atmospheric fields including temperature, humidity, winds, clouds, and radiation. Three nested subsets define progressively harder challenges: single-simulator regression, multi-simulator regression with complete observations, and multi-simulator regression with structured missingness. We propose two evaluation protocols: one for ranking methods, and one that measures performance relative to the disagreement between GCMs themselves. We evaluate seven baselines spanning simple methods, deep learning, and Gaussian processes. GP-based methods perform best, suggesting that ThousandWorlds exposes a regime where off-the-shelf deep learning does not yet succeed. Data: https://doi.org/10.57967/hf/8695. Code: https://github.com/edstevenson/ThousandWorlds.

Problem

• exoplanet climate
• climate emulation
• machine learning
• global climate models
• habitable exoplanets

Innovation

• exoclimate emulation
• machine learning benchmark
• multi-simulator regression
• parameter-to-field mapping
• low-data regime