Comparative Biosignatures (2505.01512v1)

Abstract: The discovery of inhabited exoplanets hinges on identifying biosignature gases. JWST is revealing potential biosignatures in exoplanet atmospheres, though their presence is yet to provide strong evidence for life. The central challenge is attribution: how to confidently identify biogenic sources while ruling out, or deeming unlikely, abiotic explanations? Attribution is particularly difficult for individual planets, especially regarding system-scale stochastic processes that could set atmospheric conditions. To address this, we here propose a comparative multi-planet approach: comparing atmospheric compositions across multiple planets within a system to empirically define the 'abiotic baseline'. This baseline serves as a reference point for biosignatures, and enables marginalisation over inaccessible, shared abiotic parameters. This is possible because planets within a system are linked by their birth in the same natal disk, having been irradiated by the same evolving star, and having a related dynamical history. Observations aligning with the abiotic baseline, where the locally informed abiotic models demonstrate high out-of-sample predictive accuracy, are likely non-biological. Deviations from the baseline -- potentially biotic anomalies -- suggest an alternative origin. We present the application of Bayesian leave-one-out cross-validation to evaluate the performance of geochemical- and biogeochemical-climate models in explaining these anomalies, using the expected log pointwise predictive density as a diagnostic. When biogeochemical models outperform their abiotic counterparts, the anomaly may be shaped by life, and constitutes a comparative biosignature. If both models perform poorly, the anomaly is flagged as an "unknown unknown" -- a signature of either unrecognised abiotic chemistry, or life as we don't yet know it.