Pith. sign in

REVIEW 3 major objections 5 minor 1 cited by

Stacked exoplanet transmission spectra are, under well-defined conditions, mathematically equivalent to spectra generated from the geometric mean of each planet's abundance ratios.

Reviewed by Pith at T0; open to challenge. T0 means a machine referee read the full paper against a public rubric. the ladder, T0–T4 →

Challenge this review Re-run · record.json Download PDF Read on arXiv ↗

Stacked exoplanet transmission spectra, expressed as ΔR_t/H, are approximately the logarithm of the geometric mean of the planets' abundance ratios, provided the planets share the same dominant absorbers and similar temperatures.

T0 review reviewed 2026-08-04 challenge →

load-bearing objection The core equivalence is new and holds up; the validation is partly self-referential, but the analytic derivation stands on its own and the paper honestly maps where stacking breaks. the 3 major comments →

arxiv 2510.27386 v2 pith:PFXG3S7I submitted 2025-10-31 astro-ph.EP astro-ph.IM

Stacking transmission spectra of different exoplanets

James Kirk , James E. Owen This is my paper
classification astro-ph.EP astro-ph.IM
keywords exoplanet atmospherestransmission spectroscopystacking spectrapopulation-level analysisgeometric meanabundance ratiossub-Neptunessignal-to-noise improvement
verification ladder T0 review T1 audit T2 compute T3 formal T4 reserved

The pith

A machine-rendered reading of the paper's core claim, the machinery that carries it, and where it could break.

The reading

The paper asks what you actually get when you average transmission spectra of different exoplanets. It argues that, if at each wavelength the same molecule dominates the extinction in every planet, the stacked spectrum (expressed as transit-radius differences divided by scale height) equals the logarithm of the geometric mean of the planets' abundance ratios, not an undefined average of dissimilar atmospheres. That gives population-level spectra a concrete physical meaning: they trace the typical chemistry of the sample. The authors verify the equivalence against grids of forward models and map where it breaks down, notably when planets straddle the CO/CH4 chemical transition. The result matters because stacking is the only realistic route to detecting atmospheric features on long-period planets, where repeated transits of one target are too expensive.

Core claim

On the paper's own terms, the central discovery is Eq. (20): for spectra normalized to ΔRt/H, the arithmetic mean over planets equals the natural logarithm of the geometric mean of the abundance ratios Xν1/Xν2 in the region probed by transmission spectroscopy. The derivation assumes one dominant absorber per wavelength with a power-law vertical profile; the harmonic mean of the profile indices controls the effective amplitude, and the representative planet has temperature, radius, and mass set by the geometric and harmonic means of the sample. The paper demonstrates with forward-model grids that this holds for two-species atmospheres (H2O and CO2) and for modest temperature ranges, and that

What carries the argument

The load-bearing identity is the geometric-mean equivalence (Eq. 20). The stacking quantity is ΔRt/H — the transit-depth difference between two wavelengths divided by the atmospheric scale height — which lets the per-planet cross-section ratio factor out; the arithmetic mean of logs then becomes the log of a geometric mean. Supporting it are the harmonic mean of the vertical power-law indices (controlling spectral amplitude), a weighted geometric mean for cross-sections that reduces to the unweighted one, and representative planetary parameters (T_R, R_R, M_R) defined so the stacked spectrum probes the same pressure and temperature region as the average planet.

Load-bearing premise

At every wavelength, the same single species must dominate the extinction in every stacked planet, and their per-molecule cross-sections must be similar (same temperature and pressure regime).

What would settle it

Take a sample of real exoplanet transmission spectra with precisely measured abundances, stack them after binning by temperature to avoid the CO/CH4 boundary, and check whether the stacked ΔRt/H spectrum matches the geometric-mean-abundance spectrum to within the stacked noise; a mismatch larger than the noise would falsify the equivalence.

Watch this falsifier. Get emailed when new claim-graph text bears on it.

If this is right

  • If correct, stacking can be used to measure a population's typical abundance ratio rather than an ill-defined average of heterogeneous atmospheres.
  • Long-period planets become observationally accessible: roughly ten similar planets can be observed in a few months, whereas a single target would require several years of repeated transits.
  • Stacking is only valid within parameter ranges where the same species dominates at each wavelength; crossing the CO/CH4 boundary or having a wide temperature spread biases the result.
  • Retrievals on stacked spectra should use the representative planet as the forward model, since it encodes the geometric and harmonic means of the sample.
  • For muted sub-Neptunes, the number of planets needed to rule out a flat spectrum at >5σ ranges from about 2 to 38 depending on cloud-deck pressure and per-planet precision.

Where Pith is reading between the lines

These are editorial extensions of the paper, not claims the author makes directly.

  • One consequence the paper leaves implicit: the same geometric-mean logic could apply to other observables normalized by scale height (e.g., emission spectra or phase curves) if a single absorber dominates.
  • The temperature sensitivity suggests a practical prescription: group planets by equilibrium-temperature bins of roughly 500–600 K before stacking, and test for CO/CH4 boundaries using carbon-to-oxygen ratio estimates.
  • The result gives a null hypothesis for population studies: if stacked spectra deviate from the geometric-mean prediction, that deviation is a population-level marker of chemical diversity, not just noise.
  • With a large sample (e.g., a future space-mission survey of hundreds of planets), stacking could overcome the outlier bias that plagues the small-sample limit, but a robust median or sigma-clipped stack may behave differently than the arithmetic mean treated here.
Share X Bluesky LinkedIn Reddit HN

Editorial analysis

A structured set of objections, weighed in public.

Desk editor's note, referee report, simulated authors' rebuttal, and a circularity audit.

Referee Report

3 major / 5 minor

Summary. The paper develops a formalism for stacking transmission spectra of different exoplanets. The central result, Eq. (20), states that under conditions of a single dominant absorber per wavelength and self-similar abundance profiles, the arithmetic mean of (R_t,nu1 - R_t,nu2)/H across planets equals the logarithm of the geometric mean of the abundance ratios X_nu1/X_nu2 in the transmission region. The authors derive representative planetary parameters (Eqs. 17-19) and test the approach with POSEIDON forward models over NIRSpec/G395H, mapping where stacking works (two species with self-similar ratios, moderate temperature ranges) and where it breaks down (crossing the CO/CH4 boundary, large temperature spreads). They also estimate how many sub-Neptunes must be stacked to reject a flat spectrum at >5 sigma as a function of cloud-deck pressure and per-planet precision.

Significance. If correct, this gives stacked exoplanet transmission spectra a concrete physical interpretation and supports stacking as a population-level chemical probe, particularly for long-period planets where repeated observations of a single target are expensive. The analytic derivation is transparent and the toy-model demonstrations are useful. The forward-model grid studies are extensive and provide practical quantitative guidance (e.g., the ~600 K temperature-range limit for hot Jupiters and the required number of sub-Neptunes for a >5 sigma detection as a function of cloud pressure and precision). The paper is also honest about the CO/CH4 boundary limitation. However, the printed representative-parameter equations contain a dimensional error, and Eq. (20) drops a cross-section ratio factor that is load-bearing for the abundance interpretation; these need attention before the central claim is fully supported.

major comments (3)
  1. [Section 2.6, Eqs. (18)-(19)] The representative radius and mass equations as printed are dimensionally inconsistent. With H_i proportional to T_i R_i^2 / M_i and the transmission pressure P_trans,i proportional to sqrt(T_i M_i / R_i^3), the conditions T_R R_R^2 / M_R = H({T_i R_i^2/M_i}) and sqrt(T_R M_R / R_R^3) = G({sqrt(T_i M_i / R_i^3)}) yield R_R = T_R^2 / [H({T_i R_i^2/M_i}) * G^2] and M_R = T_R^5 / [H({...})^3 * G^4] - i.e., division by the geometric-mean factors, not multiplication as printed. As written, Eqs. (18) and (19) have the wrong dimensions. Since these representative parameters are used to construct the GMA spectra in every forward-model test, this must be corrected and the affected tests re-verified.
  2. [Eq. (20) and abstract] The central equality omits the cross-section ratio. In the single-species-dominant limit the stacked quantity is log of the geometric mean of (X_nu1,j * sigma_nu1,j) / (X_nu2,j * sigma_nu2,j), not simply log G({X_nu1/X_nu2}). Eq. (20) is valid only when sigma_nu1/sigma_nu2 is approximately common across the sample (same dominant species and similar T/P conditions). The paper acknowledges this in words in Section 4.1, but the abstract and Eq. (20) should carry the same caveat. Moreover, the forward-model comparisons in Figs. 5-13 build the GMA spectrum from the same abundance inputs used to generate the individual spectra, so they test the internal consistency of the representative-parameter construction rather than independently verifying the abundance interpretation. An explicit test with individual spectra computed using opacities at each planet's T and P and a comparison spectrum comp
  3. [Section 3.2, Figure 6] The sentence 'In our assumptions, we took the opacities to be temperature insensitive; however, realistic opacities do show a temperature dependence, resulting in a difference' is ambiguous. If the opacities in this test are actually temperature-independent, then the 0.15 Delta R_t/H residual cannot be attributed to temperature-dependent opacities; if the models use realistic temperature-dependent opacities, the sentence is contradictory. Clarify which assumption is used in Figure 6 and how the residual is attributed to temperature.
minor comments (5)
  1. [Eq. (12)] The replacement of the weighted geometric mean by the unweighted geometric mean is justified heuristically and tested in toy models, but the statement that 'the weights are order unity coefficients' should be quantified. Realistic molecular opacities can vary strongly with T and P even within a single band, so the condition should be stated as an explicit assumption.
  2. [Appendix A, Eq. (A7)] The factor A is introduced but not explicitly defined in terms of the sum over species and the power-law indices l_i. An explicit expression would make the 'A is constant between planets if the temperature gradient is the same and the extinction profile is self-similar' condition easier to verify.
  3. [Section 2.2] The text says 'for the major species, where l is a positive order unity', but the toy model in Figure 2 draws l values down to -0.3. Please reconcile or clarify that negative l values are an extreme case.
  4. [Section 3.5] The statement '1 Delta R_t/H is approximately 20 ppm' is an average over the generated sub-Neptunes; the conversion depends on T, mean molecular weight, and surface gravity. Suggest rephrasing to 'typical' or quoting the range.
  5. [Figure 12] The figure is dense with nine curves. Adding a horizontal 5-sigma reference line and marking the intersection points would make the quoted numbers (2, 7, 9, 38) easier to verify.

Circularity Check

0 steps flagged

No significant circularity: the central stacked-vs-geometric-mean derivation is analytic and self-contained; the forward-model GMA comparisons are internal consistency checks rather than independent external benchmarks, and the cited self-work is not load-bearing.

full rationale

The paper's central claim (Eq. 20) is derived directly from standard slant-optical-depth formulae (Eqs. 1-4) and the identity that an arithmetic mean of logarithms equals the logarithm of a geometric mean (Eqs. 7-14). No parameter is fitted to the stacked spectra and then renamed as a prediction. The GMA spectra used in the numerical comparisons are constructed from the same abundance inputs as the individual forward models, so those comparisons verify the algebraic formalism and its assumptions with realistic opacities, but they do not independently validate the interpretation against external data. That is a limitation in evidential strength, not circularity: the derivation does not depend on the numerical comparisons. The paper also explicitly identifies the main limitation - temperature-dependent opacities and species changes such as the CO/CH4 boundary can break the abundance-ratio interpretation (Section 3.3, Discussion) - which further shows the claim is not being protected by construction. The only self-citations (e.g., Rogers et al. 2023 for a sub-Neptune mass-radius relation) are supporting inputs to illustrative calculations, not uniqueness theorems or central premises. No circularity step can be quoted because none exists; score 1 reflects only the minor, non-load-bearing self-citation and the non-independent numerical verification.

Axiom & Free-Parameter Ledger

2 free parameters · 5 axioms · 0 invented entities

The central claim rests on standard transmission-spectroscopy approximations and stated model choices, with no new particles, forces, or fitted constants. The free parameters listed affect the paper's practical guidance (planet-count numbers), not the core mathematical equivalence. The dominant physical assumption is single-species dominance with temperature-insensitive cross-section ratios, which the paper itself identifies as the breaking point.

free parameters (2)
  • per-planet spectral precision sigma = 0.5, 1.0, 2.0 ΔR_t/H (~10, 20, 40 ppm at R=200)
    Input uncertainty per spectral bin used in the sub-Neptune stacking tests (Section 3.5). The headline numbers (2, 7, 38 planets needed for >5-sigma detection) depend directly on this choice, which is selected by hand to bracket typical JWST precisions (e.g., TOI-776c at ~25 ppm).
  • grey cloud deck pressure P_cloud = 10^-4, 10^-5, 10^-6 bar
    Assumed cloud top pressure for sub-Neptune models (Section 3.5). It sets the amplitude of spectral features and therefore the planet counts in Figure 12. Not constrained by data; chosen to explore the high-metallicity/aerosol degeneracy. These inputs affect the practical guidance, not the central geometric-mean equivalence.
axioms (5)
  • domain assumption Isothermal, constant-gravity atmosphere with H/R_p << 1
    Used throughout the derivation (Eqs. 2–3, Appendix A). Standard in transmission spectroscopy; the paper extends to non-isothermal atmospheres via Eq. 16.
  • domain assumption At each wavelength a single species dominates the extinction, with a power-law abundance profile of constant index l in the transmission region
    Used to derive Eq. 3 and the stacked-spectrum result (Eqs. 7–14, Appendix A, Eq. A5–A8). The paper shows this fails at overlap regions and across the CO/CH4 boundary.
  • domain assumption Opacity scales linearly with abundance and the per-molecule cross-sections are the same across the stacked planets (temperature/pressure independent or self-similar probed regions)
    Needed for the final step to Eq. 20 (stacked spectrum = geometric mean of abundance ratios). Temperature-dependent opacities are identified as the cause of breakdown for wide temperature ranges (Sections 3.2, 4.1).
  • domain assumption Equilibrium chemistry (FastChem) with solar metallicity and C/O = 0.59 describes the atmospheric compositions
    Used for the realistic model grids in Section 3.3 onward. This sets the CO/CH4 boundary at ~1000–1300 K, which is the primary chemical transition explored in the paper.
  • domain assumption Sub-Neptune mass-radius relation from Rogers et al. (2023) and 1000x-solar metallicity with grey cloud decks
    Used for sub-Neptune tests in Section 3.5. The quantitative results (e.g., needing 7 planets at 10^-5 bar) are conditional on these assumptions, as the paper notes.

reviewed 2026-08-04 · how reviews work

0 comments
Cite this review

Pith. "Pith review of Stacking transmission spectra of different exoplanets." pith.science (2026). https://pith.science/paper/PFXG3S7I

@misc{pith2026251027386,
  author       = {Pith},
  title        = {Pith review of: Stacking transmission spectra of different exoplanets},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/PFXG3S7I}},
  note         = {Machine review of arXiv:2510.27386}
}
Share X Bluesky LinkedIn Reddit HN
abstract

In many areas of astronomy, spectra of different objects are co-added or stacked to improve signal-to-noise and reveal population-level characteristics. As the number of exoplanets with measured transmission spectra grows, it becomes important to understand when stacking spectra from different exoplanets is appropriate and what stacked spectra physically represent. Stacking will be particularly valuable for long-period planets, where repeated observations of the same planet are time-consuming. Here, we show that stacked exoplanet transmission spectra can, under well-defined conditions, be represented by spectra generated from the geometric mean of each planet's abundance ratios. We test this by comparing stacked and geometric mean spectra across grids of forward models over JWST's NIRSpec/G395H wavelength range (2.8-5.2$\mu$m). For two dominant species (e.g., H$_2$O and CO$_2$), the geometric mean accurately reflects the stacked spectrum if abundance ratios are self-similar across planets. Introducing a third species (e.g., CH$_4$) makes temperature a critical factor, with stacking becoming inappropriate across the CO/CH$_4$ boundary, which is the primary chemical transition considered in this work. Surface gravity exerts only a minor influence when stacking within comparable planetary regimes. We further assess the number of stacked, distinct sub-Neptunes with high-metallicity atmospheres and low-pressure, grey cloud decks required to rule out a flat spectrum at $>5\sigma$, as a function of both cloud deck pressure and per-planet spectral precision. These results provide guidance on when stacking is useful and on how to interpret stacked exoplanet spectra in the era of population studies of exoplanets.

Figures

Figures reproduced from arXiv: 2510.27386 by James E. Owen, James Kirk.

Figure 1
Figure 1. Figure 1: The transmission spectrum, shown as [𝑅𝑡 − 𝑅𝑡 (𝜆 = 1 𝜇m) ]/𝐻 of three planets with masses of 0.25, 1.25 & 6.25 MJ . The atmosphere contains two (arbitrary) species with Gaussian cross-sections. The left-panel shows a scenario where the abundance profiles as a function of pressure are fixed in the atmosphere. In this case because the transmission spectra of the different mass planets probe different pressure… view at source ↗
Figure 3
Figure 3. Figure 3: Demonstration that for 𝑐 = 0, the stacked spectra is given by the natural logarithm of the geometric mean of the opacity ratio. 2 4 6 8 10 Wavelength [µm] −15 −10 −5 0 5 10 15 Transmission Spectrum, ∆ Rt/H Single Planet Stacked Geometric Mean Abundances [PITH_FULL_IMAGE:figures/full_fig_p005_3.png] view at source ↗
Figure 4
Figure 4. Figure 4: Demonstration that varying all the stacked spectra is given by the harmonic mean of the scale height and geometric mean of the opacities. well described by a scenario where the effective scale height is the harmonic mean and the opacities are the geometric mean of all the individual planets (i.e. Equation 13). In order to assess the accuracy of this approach we repeat this experiment 50 times, finding a ty… view at source ↗
Figure 5
Figure 5. Figure 5: shows the results of this test. The bottom panel demon￾strates that the spectrum derived from the representative planetary parameters (𝑇𝑅, 𝑅𝑅, 𝑀𝑅 as defined in equations 17, 18 and 19) and the geometric mean of the individual planets’ abundance profiles (labelled ‘GMA’ on the figure) provides a good match to the stacked spectrum. Although there is structure within the residuals of [PITH_FULL_IMAGE:figures… view at source ↗
Figure 7
Figure 7. Figure 7: Appropriateness of stacking two planets with identical chemical abundances (H2O and CO2 as the only active species) across surface gravity (𝑔) and equilibrium temperature (Teq). The reference planet is marked by the white cross and is paired with a second planet shown as black dots. The background colour shows the residual RMS between the stacked spectra and the geometric mean abundance spectra at 𝑅 = 600.… view at source ↗
Figure 6
Figure 6. Figure 6: Testing the impact of varying a planet’s radius (top plot), mass (middle plot) and temperature (bottom plot) by ±50% while keeping the planets’ atmospheric compositions the same (solar abundances of H2O and CO2 as the only species). This demonstrates that variations in the temperature, not radius or mass, leads to the largest difference between the geometric mean abundance spectrum (orange) and the stacked… view at source ↗
Figure 8
Figure 8. Figure 8: Appropriateness of stacking two planets with different chemical compositions. Top panel: Planets with solar log 𝑍 and C/O, including H2O, CO2, CH4, CO, H2S, and SO2 as active species, varying in 𝑔 and Teq. Bottom panel: Planets with identical 𝑔 and Teq but varying log 𝑍 and C/O. In both panels, the reference planet (white cross) is paired with a second planet (black dots), and colours indicate residual RMS… view at source ↗
Figure 9
Figure 9. Figure 9: Figures showing the appropriateness of stacking two planets with different temperatures, surface gravities and different compositions. The dif￾ference to the bottom panel of [PITH_FULL_IMAGE:figures/full_fig_p010_9.png] view at source ↗
Figure 10
Figure 10. Figure 10: The result of stacking five randomly selected planets 1000 times, drawn from either a wide temperature range (500–2500 K; left column) or a narrow range (1200–1700 K; right column). For each draw, planets have randomly assigned values of surface gravity (𝑔), equilibrium temperature (𝑇eq), metallicity (log 𝑍), and C/O ratio. The x- and y-axes show the range in 𝑔 and 𝑇eq within each five-planet sample. Colo… view at source ↗
Figure 11
Figure 11. Figure 11: The result of stacking five randomly selected sub-Neptunes 1000 times. For each draw, planets have randomly assigned values of surface gravity (𝑔) and equilibrium temperature (𝑇eq), but fixed high metallicity atmospheres (1000× solar) with low pressure cloud decks (10−4 bar) to ensure small am￾plitude features in the individual spectra. The x- and y-axes show the range in 𝑔 and 𝑇eq within each five-planet… view at source ↗
Figure 12
Figure 12. Figure 12: The number of different sub-Neptunes that would need to be stacked in order to rule out a flat line at a given 𝜎 confidence with the JWST NIRSpec/G395H instrument mode. This is shown for different pres￾sures of a grey, opaque cloud deck (10−6 , blue; 10−5 , orange; 10−4 bar, green) and different uncertainties in the individual planets’ transmission spectra (Δ𝑅𝑡 /𝐻 = 0.5, dotted; 1, dashed; 2, solid; ≈ 10,… view at source ↗

discussion (0)

Sign in with ORCID, Apple, or X to comment. Anyone can read and Pith papers without signing in.

Forward citations

Cited by 1 Pith paper

Reviewed papers in the Pith corpus that reference this work. Sorted by Pith novelty score.

  1. HAT-P-70b through the Eyes of MAROON-X: Constraining Elemental Abundances of Metals and Insights on Atmosphere Dynamics

    astro-ph.EP 2026-05 conditional novelty 6.0

    New MAROON-X observations of HAT-P-70b detect multiple neutral and ionized metals with day-to-night wind signatures and demonstrate that ionization-aware retrievals yield abundance ratios closer to solar values except...

Reference graph

Works this paper leans on

61 extracted references · 3 canonical work pages · cited by 1 Pith paper

  1. [1]

    Ahrer E.-M., et al., 2025, @doi [ ] 10.1093/mnras/staf819 , https://ui.adsabs.harvard.edu/abs/2025MNRAS.540.2535A 540, 2535

    doi:10.1093/mnras/staf819 2025

  2. [2]

    Alderson L., et al., 2023, @doi [ ] 10.1038/s41586-022-05591-3 , https://ui.adsabs.harvard.edu/abs/2023Natur.614..664A 614, 664

    doi:10.1038/s41586-022-05591-3 2023

  3. [3]

    Alderson L., et al., 2024, @doi [ ] 10.3847/1538-3881/ad32c9 , https://ui.adsabs.harvard.edu/abs/2024AJ....167..216A 167, 216

    doi:10.3847/1538-3881/ad32c9 2024

  4. [4]

    J., Scott P., 2009, @doi [ ] 10.1146/annurev.astro.46.060407.145222 , https://ui.adsabs.harvard.edu/abs/2009ARA&A..47..481A 47, 481

    Asplund M., Grevesse N., Sauval A. J., Scott P., 2009, @doi [ ] 10.1146/annurev.astro.46.060407.145222 , https://ui.adsabs.harvard.edu/abs/2009ARA&A..47..481A 47, 481

    arXiv 2009

  5. [5]

    M., Grevesse N., 2021, @doi [ ] 10.1051/0004-6361/202140445 , https://ui.adsabs.harvard.edu/abs/2021A&A...653A.141A 653, A141

    Asplund M., Amarsi A. M., Grevesse N., 2021, @doi [ ] 10.1051/0004-6361/202140445 , https://ui.adsabs.harvard.edu/abs/2021A&A...653A.141A 653, A141

    doi:10.1051/0004-6361/202140445 2021

  6. [6]

    Azzam A. A. A., Tennyson J., Yurchenko S. N., Naumenko O. V., 2016, @doi [ ] 10.1093/mnras/stw1133 , https://ui.adsabs.harvard.edu/abs/2016MNRAS.460.4063A 460, 4063

    doi:10.1093/mnras/stw1133 2016

  7. [7]

    K., Heng K., 2020, @doi [ ] 10.1007/s11214-020-00666-x , https://ui.adsabs.harvard.edu/abs/2020SSRv..216...82B 216, 82

    Barstow J. K., Heng K., 2020, @doi [ ] 10.1007/s11214-020-00666-x , https://ui.adsabs.harvard.edu/abs/2020SSRv..216...82B 216, 82

    doi:10.1007/s11214-020-00666-x 2020

  8. [8]

    Benneke B., Seager S., 2012, @doi [ ] 10.1088/0004-637X/753/2/100 , https://ui.adsabs.harvard.edu/abs/2012ApJ...753..100B 753, 100

    doi:10.1088/0004-637x/753/2/100 2012

  9. [9]

    Brande J., et al., 2024, @doi [ ] 10.3847/2041-8213/ad1b5c , https://ui.adsabs.harvard.edu/abs/2024ApJ...961L..23B 961, L23

    doi:10.3847/2041-8213/ad1b5c 2024

  10. [10]

    M., 2001, @doi [ ] 10.1086/320950 , https://ui.adsabs.harvard.edu/abs/2001ApJ...553.1006B 553, 1006

    Brown T. M., 2001, @doi [ ] 10.1086/320950 , https://ui.adsabs.harvard.edu/abs/2001ApJ...553.1006B 553, 1006

    work page doi:10.1086/320950 2001

  11. [11]

    Cadieux C., et al., 2024, The Astrophysical Journal Letters, 970, L2

    2024

  12. [12]

    Crossfield I. J. M., Kreidberg L., 2017, @doi [ ] 10.3847/1538-3881/aa9279 , https://ui.adsabs.harvard.edu/abs/2017AJ....154..261C 154, 261

    doi:10.3847/1538-3881/aa9279 2017

  13. [13]

    Damiano M., Bello-Arufe A., Yang J., Hu R., 2024, The Astrophysical Journal Letters, 968, L22

    2024

  14. [14]

    J., 2005, @doi [ ] 10.1111/j.1365-2966.2005.09587.x , https://ui.adsabs.harvard.edu/abs/2005MNRAS.364..649F 364, 649

    Fortney J. J., 2005, @doi [ ] 10.1111/j.1365-2966.2005.09587.x , https://ui.adsabs.harvard.edu/abs/2005MNRAS.364..649F 364, 649

    arXiv 2005

  15. [15]

    Fu G., Deming D., Knutson H., Madhusudhan N., Mandell A., Fraine J., 2017, @doi [ ] 10.3847/2041-8213/aa8e40 , https://ui.adsabs.harvard.edu/abs/2017ApJ...847L..22F 847, L22

    work page doi:10.3847/2041-8213/aa8e40 2017

  16. [16]

    Fu G., et al., 2025, @doi [ ] 10.3847/1538-4357/ad7bb8 , https://ui.adsabs.harvard.edu/abs/2025ApJ...986....1F 986, 1

    work page doi:10.3847/1538-4357/ad7bb8 2025

  17. [17]

    Gao P., et al., 2020, @doi [Nature Astronomy] 10.1038/s41550-020-1114-3 , https://ui.adsabs.harvard.edu/abs/2020NatAs...4..951G 4, 951

    doi:10.1038/s41550-020-1114-3 2020

  18. [18]

    Gressier A., et al., 2024, @doi [ ] 10.3847/2041-8213/ad73d1 , https://ui.adsabs.harvard.edu/abs/2024ApJ...975L..10G 975, L10

    doi:10.3847/2041-8213/ad73d1 2024

  19. [19]

    A., 2014, @doi [Philosophical Transactions of the Royal Society of London Series A] 10.1098/rsta.2013.0086 , https://ui.adsabs.harvard.edu/abs/2014RSPTA.37230086G 372, 20130086

    Griffith C. A., 2014, @doi [Philosophical Transactions of the Royal Society of London Series A] 10.1098/rsta.2013.0086 , https://ui.adsabs.harvard.edu/abs/2014RSPTA.37230086G 372, 20130086

    arXiv 2014

  20. [20]

    Hansen B. M. S., 2008, @doi [ ] 10.1086/591964 , https://ui.adsabs.harvard.edu/abs/2008ApJS..179..484H 179, 484

    doi:10.1086/591964 2008

  21. [21]

    Heng K., Kitzmann D., 2017, @doi [ ] 10.1093/mnras/stx1453 , https://ui.adsabs.harvard.edu/abs/2017MNRAS.470.2972H 470, 2972

    doi:10.1093/mnras/stx1453 2017

  22. [22]

    B., Fortney J

    Hubbard W. B., Fortney J. J., Lunine J. I., Burrows A., Sudarsky D., Pinto P., 2001, @doi [ ] 10.1086/322490 , https://ui.adsabs.harvard.edu/abs/2001ApJ...560..413H 560, 413

    doi:10.1086/322490 2001

  23. [23]

    M.-R., et al., 2023, Nature, 620, 67

    Kempton E. M.-R., et al., 2023, Nature, 620, 67

    2023

  24. [24]

    Kirk J., et al., 2025, @doi [ ] 10.1093/mnras/staf208 , https://ui.adsabs.harvard.edu/abs/2025MNRAS.537.3027K 537, 3027

    doi:10.1093/mnras/staf208 2025

  25. [25]

    W., Patzer A

    Kitzmann D., Stock J. W., Patzer A. B. C., 2024, @doi [ ] 10.1093/mnras/stad3515 , https://ui.adsabs.harvard.edu/abs/2024MNRAS.527.7263K 527, 7263

    doi:10.1093/mnras/stad3515 2024

  26. [26]

    J., Belmonte J

    Kreidberg L., 2018, in Deeg H. J., Belmonte J. A., eds, , Handbook of Exoplanets. p. 100, @doi 10.1007/978-3-319-55333-7_100

    doi:10.1007/978-3-319-55333-7_100 2018

  27. [27]

    E., Rothman L

    Li G., Gordon I. E., Rothman L. S., Tan Y., Hu S.-M., Kassi S., Campargue A., Medvedev E. S., 2015, @doi [ ] 10.1088/0067-0049/216/1/15 , https://ui.adsabs.harvard.edu/abs/2015ApJS..216...15L 216, 15

    doi:10.1088/0067-0049/216/1/15 2015

  28. [28]

    Lorenz B., et al., 2023, @doi [ ] 10.3847/1538-4357/accdd1 , https://ui.adsabs.harvard.edu/abs/2023ApJ...951...29L 951, 29

    doi:10.3847/1538-4357/accdd1 2023

  29. [29]

    Lueber A., Novais A., Fisher C., Heng K., 2024, @doi [ ] 10.1051/0004-6361/202348802 , https://ui.adsabs.harvard.edu/abs/2024A&A...687A.110L 687, A110

    doi:10.1051/0004-6361/202348802 2024

  30. [30]

    J., 2023, @doi [The Journal of Open Source Software] 10.21105/joss.04873 , https://ui.adsabs.harvard.edu/abs/2023JOSS....8.4873M 8, 4873

    MacDonald R. J., 2023, @doi [The Journal of Open Source Software] 10.21105/joss.04873 , https://ui.adsabs.harvard.edu/abs/2023JOSS....8.4873M 8, 4873

    doi:10.21105/joss.04873 2023

  31. [31]

    J., Madhusudhan N., 2017, @doi [ ] 10.1093/mnras/stx804 , https://ui.adsabs.harvard.edu/abs/2017MNRAS.469.1979M 469, 1979

    MacDonald R. J., Madhusudhan N., 2017, @doi [ ] 10.1093/mnras/stx804 , https://ui.adsabs.harvard.edu/abs/2017MNRAS.469.1979M 469, 1979

    doi:10.1093/mnras/stx804 2017

  32. [32]

    Madhusudhan N., Seager S., 2009, @doi [ ] 10.1088/0004-637X/707/1/24 , https://ui.adsabs.harvard.edu/abs/2009ApJ...707...24M 707, 24

    doi:10.1088/0004-637x/707/1/24 2009

  33. [33]

    Madhusudhan N., Piette A. A. A., Constantinou S., 2021, @doi [ ] 10.3847/1538-4357/abfd9c , https://ui.adsabs.harvard.edu/abs/2021ApJ...918....1M 918, 1

    doi:10.3847/1538-4357/abfd9c 2021

  34. [34]

    M., et al., 2023, @doi [ ] 10.3847/2041-8213/ad054f , https://ui.adsabs.harvard.edu/abs/2023ApJ...959L...9M 959, L9

    May E. M., et al., 2023, @doi [ ] 10.3847/2041-8213/ad054f , https://ui.adsabs.harvard.edu/abs/2023ApJ...959L...9M 959, L9

    doi:10.3847/2041-8213/ad054f 2023

  35. [35]

    Meech A., et al., 2025, @doi [ ] 10.1093/mnras/staf530 , https://ui.adsabs.harvard.edu/abs/2025MNRAS.539.1381M 539, 1381

    doi:10.1093/mnras/staf530 2025

  36. [36]

    Miller-Ricci E., Seager S., Sasselov D., 2009, @doi [ ] 10.1088/0004-637X/690/2/1056 , https://ui.adsabs.harvard.edu/abs/2009ApJ...690.1056M 690, 1056

    doi:10.1088/0004-637x/690/2/1056 2009

  37. [37]

    Nortmann L., et al., 2018, @doi [Science] 10.1126/science.aat5348 , https://ui.adsabs.harvard.edu/abs/2018Sci...362.1388N 362, 1388

    doi:10.1126/science.aat5348 2018

  38. [38]

    Ohno K., et al., 2025, @doi [ ] 10.3847/2041-8213/ada02c , https://ui.adsabs.harvard.edu/abs/2025ApJ...979L...7O 979, L7

    doi:10.3847/2041-8213/ada02c 2025

  39. [39]

    Orell-Miquel J., et al., 2024, @doi [ ] 10.1051/0004-6361/202449411 , https://ui.adsabs.harvard.edu/abs/2024A&A...689A.179O 689, A179

    doi:10.1051/0004-6361/202449411 2024

  40. [40]

    Penzlin A. B. T., et al., 2024, @doi [ ] 10.1093/mnras/stae2362 , https://ui.adsabs.harvard.edu/abs/2024MNRAS.535..171P 535, 171

    doi:10.1093/mnras/stae2362 2024

  41. [41]

    L., Kyuberis A

    Polyansky O. L., Kyuberis A. A., Zobov N. F., Tennyson J., Yurchenko S. N., Lodi L., 2018, @doi [ ] 10.1093/mnras/sty1877 , https://ui.adsabs.harvard.edu/abs/2018MNRAS.480.2597P 480, 2597

    doi:10.1093/mnras/sty1877 2018

  42. [42]

    G., Schlichting H

    Rogers J. G., Schlichting H. E., Owen J. E., 2023, @doi [ ] 10.3847/2041-8213/acc86f , https://ui.adsabs.harvard.edu/abs/2023ApJ...947L..19R 947, L19

    doi:10.3847/2041-8213/acc86f 2023

  43. [43]

    Schlawin E., et al., 2024, The Astrophysical Journal Letters, 974, L33

    2024

  44. [44]

    D., 2000, @doi [ ] 10.1086/309088 , https://ui.adsabs.harvard.edu/abs/2000ApJ...537..916S 537, 916

    Seager S., Sasselov D. D., 2000, @doi [ ] 10.1086/309088 , https://ui.adsabs.harvard.edu/abs/2000ApJ...537..916S 537, 916

    doi:10.1086/309088 2000

  45. [45]

    E., Steidel C

    Shapley A. E., Steidel C. C., Pettini M., Adelberger K. L., 2003, @doi [ ] 10.1086/373922 , https://ui.adsabs.harvard.edu/abs/2003ApJ...588...65S 588, 65

    doi:10.1086/373922 2003

  46. [46]

    H., 1942, @doi [Journal of the American Statistical Association] 10.1080/01621459.1942.10500636 , 37, 271

    Siegel I. H., 1942, @doi [Journal of the American Statistical Association] 10.1080/01621459.1942.10500636 , 37, 271

    arXiv 1942

  47. [47]

    K., et al., 2016, @doi [ ] 10.1038/nature16068 , https://ui.adsabs.harvard.edu/abs/2016Natur.529...59S 529, 59

    Sing D. K., et al., 2016, @doi [ ] 10.1038/nature16068 , https://ui.adsabs.harvard.edu/abs/2016Natur.529...59S 529, 59

    doi:10.1038/nature16068 2016

  48. [48]

    J., et al., 2021, @doi [ ] 10.1093/mnras/staa3116 , https://ui.adsabs.harvard.edu/abs/2021MNRAS.500.4042S 500, 4042

    Spake J. J., et al., 2021, @doi [ ] 10.1093/mnras/staa3116 , https://ui.adsabs.harvard.edu/abs/2021MNRAS.500.4042S 500, 4042

    doi:10.1093/mnras/staa3116 2021

  49. [49]

    C., Strom A

    Steidel C. C., Strom A. L., Pettini M., Rudie G. C., Reddy N. A., Trainor R. F., 2016, @doi [ ] 10.3847/0004-637X/826/2/159 , https://ui.adsabs.harvard.edu/abs/2016ApJ...826..159S 826, 159

    doi:10.3847/0004-637x/826/2/159 2016

  50. [50]

    B., 2016, @doi [ ] 10.3847/2041-8205/817/2/L16 , https://ui.adsabs.harvard.edu/abs/2016ApJ...817L..16S 817, L16

    Stevenson K. B., 2016, @doi [ ] 10.3847/2041-8205/817/2/L16 , https://ui.adsabs.harvard.edu/abs/2016ApJ...817L..16S 817, L16

    doi:10.3847/2041-8205/817/2/l16 2016

  51. [51]

    W., Kitzmann D., Patzer A

    Stock J. W., Kitzmann D., Patzer A. B. C., Sedlmayr E., 2018, @doi [ ] 10.1093/mnras/sty1531 , https://ui.adsabs.harvard.edu/abs/2018MNRAS.479..865S 479, 865

    doi:10.1093/mnras/sty1531 2018

  52. [52]

    W., Kitzmann D., Patzer A

    Stock J. W., Kitzmann D., Patzer A. B. C., 2022, @doi [ ] 10.1093/mnras/stac2623 , https://ui.adsabs.harvard.edu/abs/2022MNRAS.517.4070S 517, 4070

    doi:10.1093/mnras/stac2623 2022

  53. [53]

    Teske J., et al., 2025, @doi [ ] 10.3847/1538-3881/adb975 , https://ui.adsabs.harvard.edu/abs/2025AJ....169..249T 169, 249

    doi:10.3847/1538-3881/adb975 2025

  54. [54]

    Tinetti G., et al., 2018, @doi [Experimental Astronomy] 10.1007/s10686-018-9598-x , https://ui.adsabs.harvard.edu/abs/2018ExA....46..135T 46, 135

    doi:10.1007/s10686-018-9598-x 2018

  55. [55]

    S., Tennyson J., Yurchenko S

    Underwood D. S., Tennyson J., Yurchenko S. N., Huang X., Schwenke D. W., Lee T. J., Clausen S., Fateev A., 2016, @doi [ ] 10.1093/mnras/stw849 , https://ui.adsabs.harvard.edu/abs/2016MNRAS.459.3890U 459, 3890

    doi:10.1093/mnras/stw849 2016

  56. [56]

    L., et al., 2024, @doi [arXiv e-prints] 10.48550/arXiv.2404.01264 , https://ui.adsabs.harvard.edu/abs/2024arXiv240401264W p

    Wallack N. L., et al., 2024, @doi [arXiv e-prints] 10.48550/arXiv.2404.01264 , https://ui.adsabs.harvard.edu/abs/2024arXiv240401264W p. arXiv:2404.01264

    Pith/arXiv arXiv doi:10.48550/arxiv.2404.01264 2024

  57. [57]

    F., Hubeny I., Spiegelman F., Leininger T., 2019, @doi [ ] 10.3847/2041-8213/ab5a89 , https://ui.adsabs.harvard.edu/abs/2019ApJ...887L..20W 887, L20

    Welbanks L., Madhusudhan N., Allard N. F., Hubeny I., Spiegelman F., Leininger T., 2019, @doi [ ] 10.3847/2041-8213/ab5a89 , https://ui.adsabs.harvard.edu/abs/2019ApJ...887L..20W 887, L20

    doi:10.3847/2041-8213/ab5a89 2019

  58. [58]

    N., Mellor T

    Yurchenko S. N., Mellor T. M., Freedman R. S., Tennyson J., 2020, @doi [ ] 10.1093/mnras/staa1874 , https://ui.adsabs.harvard.edu/abs/2020MNRAS.496.5282Y 496, 5282

    doi:10.1093/mnras/staa1874 2020

  59. [59]

    N., Owens A., Kefala K., Tennyson J., 2024, @doi [ ] 10.1093/mnras/stae148 , https://ui.adsabs.harvard.edu/abs/2024MNRAS.528.3719Y 528, 3719

    Yurchenko S. N., Owens A., Kefala K., Tennyson J., 2024, @doi [ ] 10.1093/mnras/stae148 , https://ui.adsabs.harvard.edu/abs/2024MNRAS.528.3719Y 528, 3719

    doi:10.1093/mnras/stae148 2024

  60. [60]

    L., Knutson H

    Zhang M., Dai F., Bean J. L., Knutson H. A., Rescigno F., 2023, @doi [ ] 10.3847/2041-8213/aced51 , https://ui.adsabs.harvard.edu/abs/2023ApJ...953L..25Z 953, L25

    doi:10.3847/2041-8213/aced51 2023

  61. [61]

    de Wit J., Seager S., 2013, @doi [Science] 10.1126/science.1245450 , https://ui.adsabs.harvard.edu/abs/2013Sci...342.1473D 342, 1473

    doi:10.1126/science.1245450 2013

This paper was first reviewed by deepseek-v4-flash on August 4, 2026.