Detectability of Atmospheric Biosignatures in Earth Analogs with Varying Surface Boundary Conditions: Prospects for Characterization in the UV, Visible, Near-Infrared, and Mid-Infrared Regions

TL;DR

The paper tackles the challenge of detecting biosignature gases in Earth-like exoplanet atmospheres under surface boundary conditions that reflect biological and geological activity. It combines NWP-derived temperature–pressure profiles with a 1D photochemical model (VULCAN) and forward spectral synthesis (PSG for HWO; LIFEsim for LIFE) to produce UV–VIS–NIR and mid-IR spectra for a modern Earth analog at 10 pc, and assesses detectability using band-integrated SNRs defined as $\mathrm{SNR} = \sqrt{\sum_{i=1}^{n} (\Delta y_i / \sigma(y_i))^2}$. Key findings show that O3 is detectable with both HWO and LIFE across boundary conditions; CO2 is detectable with LIFE; H2O detectability depends strongly on surface humidity and fixed surface abundances, while CH4 and N2O require continuous surface outgassing for LIFE detectability (and CH4 is further masked by water features in some cases). The work demonstrates the complementary value of UV–VIS–NIR and mid-IR observations for constraining Earth-like atmospheres under disequilibrium chemistry, and provides practical guidance for mission design and interpretation of future exoplanet spectra.

Abstract

The search for potentially habitable exoplanets centers on detecting biosignature molecules in Earth-like atmospheres, which makes it essential to understand their detectability under biologically and geologically influenced conditions. In this study, we model the reflection and thermal emission spectra of such atmospheres across the UV/VIS/NIR and mid-IR regions and simulate their detectability with future mission concepts such as the Habitable Worlds Observatory (HWO) and the Large Interferometer for Exoplanets (LIFE). We employ Numerical Weather Prediction (NWP) model data, based on Earth's atmosphere, to derive temperature pressure profiles and couple them with a 1D photochemical model to assess the detectability of these molecules in Earth analogs located 10 parsecs away. We investigate the dominant reaction pathways and their contributions to the atmospheric composition of an Earth analog, with a focus on how they shape the resulting molecular signatures. We also examine the role of surface boundary conditions, which indirectly trace the effects of biological and geological processes, on the detectability of these molecules using HWO- and LIFE-type mission concepts. Our findings indicate that O3 is detectable with both mission concepts, while H2O requires specific surface humidity levels for detection with LIFE and shows only potential detectability with HWO. CO2 is detectable with LIFE. Both N2O and CH4 require continuous surface outgassing for potential detection with LIFE, and CH4 further requires low surface humidity to prevent masking by water features. Our work highlights the feasibility of characterizing the atmospheres of Earth analogs in the UV/VIS/NIR and mid-IR domains using HWO- and LIFE-type mission concepts and offers guidance for the development of future missions operating in these spectral regions.

Figures (10)

• Figure 1: Flowchart representing our methodical pipeline. The model types are indicated in the large boxes: green for climate model, peach for chemistry model, yellow for radiative transfer model, cyan for observables simulator, and blue for noise output. These models are interconnected. The reflection spectra and thermal emission spectra represent the overall outputs of the pipeline, leading to the spectra that will be observed by future HWO and LIFE missions.
• Figure 2: 5000 temperature-pressure (T-P) profiles derived from a numerical weather prediction (NWP) model (left), an average T-P profile considering all five types of sampling techniques (middle), and the mean temperature-pressure profile representing the global average conditions of modern Earth (right). The leftmost panel illustrates the variability in T-P profiles due to different atmospheric conditions captured by the NWP model, while the far right panel provides a single, averaged T-P profile that can serve as a reference for global atmospheric studies of an Earth analog. The global average T-P profile remains the same across all our models: M1, M2, M3, M4, and M5.
• Figure 3: Mixing ratio profiles obtained from VULCAN for the temperature sampled average T-P profile. The molecules are $\mathrm{H_2O}$, $\mathrm{O_3}$, $\mathrm{CH_4}$, $\mathrm{CO}$, $\mathrm{CO_2}$, $\mathrm{OH}$, $\mathrm{HCN}$, $\mathrm{O_2}$, $\mathrm{H_2}$, $\mathrm{NO}$, $\mathrm{NO_2}$, $\mathrm{N_2O}$, $\mathrm{SO_2}$, $\mathrm{H_2SO_4}$. The molecules are selected considering their average mixing ratios $> 10^{-18}$ between 1 bar and 0.01 mbar.
• Figure 4: Top: simulated reflection spectra observation of modern Earth analog for the HWO mission concept. The simulated flux is considered for a planetary system at 10 pc distance. The wavelength is between 0.15 and 2 µ m. The $1 \sigma$ and $2 \sigma$ error bars are represented with cyan and light grey shaded area with solid edges respectively. The black curve is the spectrum from PSG. The cyan scatter plot represents the simulated observation points with corresponding uncertainties. The simulated reflection spectra is considered at a phase angle of $77.8^\circ$. Molecular features are labeled in black, and CIA features are marked in grey; Bottom: the associated SNR vs wavelength for the reflection spectra. The SNR is assumed to be 10 at three reference wavelengths: 0.35 µ m, 0.55 µ m and 1.2 µ m
• Figure 5: Top: simulated thermal emission spectra observation of modern Earth analog by the LIFE concept assuming a simplified noise. The simulated flux is considered for a planetary system at 10 pc distance. The wavelength is between 4 and 18.4 µ m. The $1 \sigma$ and $2 \sigma$ error bars are represented with cyan and light grey shaded area with solid edges respectively. The black curve is the spectrum from PSG fed into LIFEsim to simulate the observation (cyan scatter plot) with error bars. The simulated reflection spectrum is considered at a phase angle of $77.8^\circ$. Molecular features are labeled in black, and CIA features are marked in grey; Bottom: the associated SNR vs wavelength for the thermal emission spectra. The SNR is assumed to be 10 at the reference wavelength 11.2 µ m