Biosignatures & Paleobiology

A Complex Systems Approach to Exoplanet Atmospheric Chemistry: New Prospects for Ruling Out the Possibility of Alien Life-As-We-Know-It

By Keith Cowing
Status Report
astro-ph.EP
October 10, 2023
Filed under , , , ,
A Complex Systems Approach to Exoplanet Atmospheric Chemistry: New Prospects for Ruling Out the Possibility of Alien Life-As-We-Know-It
Bayesian analysis of the posterior likelihood of life, P(life|observation), based on assuming an observed value for network metrics and methane gas abundance. Given that the probability for life to exist on an exoplanet is unknown, we assess the posterior likelihood over a range of prior probabilities for life, P(life). Grey Regions highlight metric values that are associated with a zero probability of life and are therefore anti-biosignatures for the sets of models compared herein. In general, network metrics can provide greater constraints on the probability of life given a set of observations than CH4 abundance alone, allowing distinguishing cases where Earth-like methanogenesis can be ruled out even when CH4 abundance alone is not conclusive. — astro-ph.EP

The near-term capability to characterize terrestrial exoplanet atmospheres may bring us closer to discovering alien life through atmospheric data.

However, remotely detectable candidate biosignature gases are subject to possible false positive signals as they can also be produced abiotically. To distinguish biological, abiotic and anomalous sources of these atmospheric gases, we take a complex systems approach using chemical reaction network analysis of planetary atmospheres.

We simulated 30,000 terrestrial atmospheres, organized in two datasets: Archean Earth-like worlds and modern Earth-like worlds. For Archean Earth-like worlds we study cases where CH4 is produced abiotically via serpentinization, biologically via methanogenesis, or from anomalous sources.

We also simulate modern Earth-like atmospheres with and without industrial CFC-12. Network properties like mean degree and average shortest path length effectively distinguish scenarios where CH4 is produced from methanogenesis and serpentinization, with biologically driven networks exhibiting higher connectivity and efficiency.

Network analysis also distinguishes modern Earth atmospheres with CFC-12 from those without, with industrially polluted networks showing increased mean degree. Using Bayesian analysis, we demonstrate how atmospheric network property statistics can provide stronger confidence for ruling out biological explanations compared to gas abundance statistics alone.

Our results confirm how a network theoretic approach allows distinguishing biological, abiotic and anomalous atmospheric drivers, including ruling out life-as-we-know-it as a possible explanation. Developing statistical inference methods for spectral data that incorporate network properties could significantly strengthen future biosignature detection efforts.

Theresa Fisher, Estelle Janin, Sara Imari Walker

Comments: 22 pages (including references), 4 figures, 1 table
Subjects: Earth and Planetary Astrophysics (astro-ph.EP)
Cite as: arXiv:2310.05359 [astro-ph.EP] (or arXiv:2310.05359v1 [astro-ph.EP] for this version)
Submission history
From: Sara Walker
[v1] Mon, 9 Oct 2023 02:37:31 UTC (1,043 KB)
https://arxiv.org/abs/2310.05359
Astrobiology

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) 🖖🏻