{"id":"a167c8a3-72ce-43cb-ae88-736060819760","arxiv_id":"2411.15456","paper_version":1,"verdict":"CONDITIONAL","confidence":"MODERATE","novelty_score":7.0,"correctness_risk":"medium","formal_verification":"none","parameter_count":5,"one_line_summary":"Radiative cooling by CO, H2O, OH and H3+ suppresses hydrodynamic escape of H2 from an early Earth-like atmosphere even at carbon oxide mixing ratios below 1%, extending the H2 envelope lifetime by about an order of magnitude.","lead":"Using 1D simulations, the authors show that even small amounts of carbon monoxide or carbon dioxide in an H2-rich early Earth atmosphere can strongly suppress hydrogen escape by radiating away the X-ray and ultraviolet energy that drives the outflow. If confirmed, this could extend the lifetime of the H2-rich envelope by about tenfold, with consequences for the origin of Earth's atmosphere and for exoplanet studies.","discovery_kind":"extension","skeptic_critique":{"model":"deepseek-v4-flash","headline":"The central quantitative claim rests on LTE cooling rates; if non-LTE reduces CO/H3+ cooling in the outflow, the order-of-magnitude lifetime extension shrinks.","rationale":"The reader's weakest-assumption analysis identified exactly the load-bearing point: the model's reliance on LTE radiative cooling. My independent reading of the paper confirms that this is the single most consequential uncertainty. The central claim is quantitative — one order of magnitude lifetime extension — and that number is set by the balance between XUV heating and molecular cooling. The paper itself flags the non-LTE possibility, and the dominant upper-region coolant H3+ is precisely where non-LTE corrections are expected to be largest. No other issue reaches this level of leverage on the headline: the critical-flux asymptote is a plausible physical limit, the chemical network is simplified but conservative for cooling (neglect of CH2 and larger hydrocarbons would only reduce cooling, making suppression harder, not easier), and the evolution model's extrapolation at high CO/H2 ratios is explicitly argued to have small effect. The data-availability statement is misleading for a simulation-based paper, but that affects reproducibility, not the physical correctness of the result. Therefore, a concrete non-LTE sensitivity test is the most direct way to determine whether the suppression claim holds. I agree with the reader that the verdict should remain conditional pending that test or access to code/data, and I do not recommend changing the verdict level.","tokens_in":15204,"tokens_out":6210,"duration_ms":59951,"concrete_test":"Recompute the Figure 5 escape rates with non-LTE cooling rates for CO and H3+ in place of the current LTE rates, using, for example, the escape-probability / critical-density formalism of García Muñoz et al. (2024) applied to the same temperature, density, and composition profiles, while keeping all other settings identical. Then check the basal CO/H2 ratio at which the H2 escape rate falls to within 20% of the critical flux. If that threshold rises above ~0.01, or if the implied lifetime extension in the Section 4.4 evolution model drops below a factor of ~5, the order-of-magnitude suppression claim does not survive non-LTE corrections.","verdict_should_be":"UNCHANGED","load_bearing_attack":"The paper's headline result — that trace CO/CO2 suppress H2 escape by about an order of magnitude — is quantitatively controlled by the radiative cooling rates of CO, H2O, OH, and H3+. Section 2 states the model 'assume[s] local thermal equilibrium (LTE) conditions to estimate the radiative cooling rate, although non-LTE effects can reduce the efficiency of radiative cooling, particularly in low-pressure regions.' This is not a peripheral caveat: the cooling rates enter directly into the energy equation that determines whether the outflow becomes transonic and how fast hydrogen escapes. In the simulated outflows, H3+ is the dominant coolant in the upper region (Figure 3e,f), where gas densities are lowest and LTE is most likely to break down; CO is a major coolant throughout the subsonic region. If the true LTE departure coefficients at these low densities reduce the cooling efficiency substantially, the heating efficiency in Equation (4) would rise, the H2 escape rate at a given basal CO/H2 ratio would increase, and the asymptotic approach to the critical flux (Figure 5) would require a larger mixing ratio. The order-of-magnitude lifetime extension in Section 4.4 would then shrink. Because the paper's own argument is that cooling removes the thermal energy driving escape, an overestimate of cooling directly inflates the central quantitative claim. This is a correctness risk in the modeling premise, not an internal mathematical contradiction.","agreement_with_reader":"agree"},"referee_report":{"model":"deepseek-v4-flash","summary":"This paper presents 1D hydrodynamic escape simulations of H2-rich atmospheres containing CO or CO2 on an Earth-mass planet, with a photochemical network and radiative cooling from HITRAN/ExoMol line data. The authors find that CO, H2O, OH, and H3+ cool the outflow sufficiently to reduce the H2 escape rate, which asymptotes to the Hunten critical flux once the basal CO/H2 or CO2/H2 ratio reaches about 0.001. They use these escape rates in a time-evolution model to argue that the lifetime of an H2-rich proto-atmosphere on early Earth is extended by about one order of magnitude compared with a pure H2 atmosphere, with negligible escape of heavier carbon- and nitrogen-bearing species and noble gases.","tokens_in":15490,"tokens_out":5986,"duration_ms":51441,"significance":"If robust, this result is important for early Earth atmospheric evolution and for the evolution of H2-dominated atmospheres on terrestrial exoplanets, as it identifies a relatively low threshold of carbon-oxide contamination that can switch an atmosphere from rapid hydrodynamic escape to diffusion-limited escape. The forward modeling uses standard spectroscopic databases and an existing validated framework, and the numerical asymptotic approach to the Hunten critical flux provides a useful consistency check. The central quantitative lifetime claim is conditional on the LTE cooling assumption, so the paper would be strengthened considerably by a non-LTE sensitivity test; the qualitative suppression mechanism, however, is well supported by the simulations.","major_comments":[{"comment":"Section 2 states that LTE is assumed for radiative cooling and that non-LTE effects can reduce cooling efficiency in low-pressure regions, but the paper does not quantify this. Because the energy equation directly uses these cooling rates and the heating efficiency (Equation (4)) is the key control on the escape rate, the order-of-magnitude lifetime extension in Section 4.4 depends on LTE cooling being roughly correct. In particular, H3+ is the main coolant in the upper outflow (Figure 3e,f), where densities are lowest and LTE is most suspect. A quantitative sensitivity test with reduced cooling efficiencies (e.g., a factor of 2 or 5 reduction in H3+ and CO cooling above a certain density) would be needed to show that the qualitative and quantitative conclusions are robust.","section":"Section 2 (radiative cooling; Equation (4))"},{"comment":"The abstract and conclusions claim 'negligible escape of heavier carbon- and nitrogen-bearing species and noble gases.' This goes beyond the simulations, which contain no nitrogen species in the chemical network (Section 2). The conclusion about nitrogen is an extrapolation from the crossover-mass behavior of CO and CO2. Please either remove the nitrogen claim or support it with an explicit crossover-mass argument or a simulation including N-bearing species.","section":"Section 5 (conclusions) and abstract"},{"comment":"In the evolutionary model, escape rates for basal CO/H2 or CO2/H2 above 0.5 are set to the critical flux for H2 and zero for carbon oxides. The authors state this has little effect when the initial H2 amount exceeds 10 bar, but the range of initial H2 amounts used in Figure 9 is not stated in the main text. Please specify this range and document the sensitivity of the lifetime estimate to the assumed behavior at ratios above 0.5, particularly for initial H2 amounts near or below 10 bar.","section":"Section 4.4 (evolutionary model)"}],"minor_comments":[{"comment":"Equation (7) is used twice: first for the crossover mass in Section 3 and later for the equivalent surface pressure in Section 4.4; please renumber to avoid ambiguity.","section":"Throughout"},{"comment":"The heating rate is described as a 'red dash-dotted line' in the main text and as a 'red solid line' in the figure legend at the end of the paper; please make these descriptions consistent.","section":"Figure 3(e)"},{"comment":"The abstract says suppression occurs when the basal mixing fraction of CO and CO2 is 'lower than ~0.01', while Section 3 states the escape rate becomes almost constant above ~0.001; please reconcile these thresholds to avoid confusion.","section":"Abstract and Section 3"},{"comment":"The data availability statement says 'no datasets were generated or analysed,' but Figures 2-9 are based on simulation outputs; consider making the model outputs available in a repository to support reproducibility.","section":"Data availability"}],"recommendation":"major_revision","confidential_remarks":"The paper relies extensively on the authors' prior model (Yoshida and Kuramoto 2020, 2021), which is appropriate for a series of papers, but the manuscript should ensure that the new carbon-oxide chemistry and radiative processes are sufficiently described in the Supplementary Information. I see no concerns about novelty disclosure; the work is a natural extension of previous studies. The main technical risk is the unquantified LTE assumption, which should be addressed before final acceptance."},"author_rebuttal":null,"desk_editor":{"model":"deepseek-v4-flash","letter":"Colleague, here is my read on Yoshida et al. The paper's actual new result is that trace CO/CO2 (basal C/H2 ~1e-5, not the ~1e-3 needed for CH4) suppress hydrodynamic escape of H2-rich atmospheres on an Earth-mass planet, with escape rates falling to the Hunten critical flux once the carbon oxide mixing ratio reaches ~1e-3. The order-of-magnitude lifetime extension in their evolution model follows from that. That is a real, non-obvious finding for early Earth and low-mass exoplanet work.\n\nWhat it does well: the photochemistry narrative is coherent — CO2 photolyzes to CO + O, O oxidizes H2 to OH and H2O, and those together with H3+ do the cooling. The 1D model is internally consistent, the cooling rates use HITRAN/ExoMol line lists, and the asymptotic behavior matching the Hunten critical flux gives me confidence the suppression mechanism is real. The comparison with their earlier CH4 work is useful and shows why CO/CO2 act at lower mixing ratios.\n\nSoft spots, in rough order of severity. First, the quantitative lifetime claim leans on LTE cooling rates in the outflow, and the paper itself concedes that non-LTE effects weaken cooling at low pressures — exactly where H3+ is the dominant coolant. If departure coefficients cut the cooling, the escape rates go back up and the \"order of magnitude\" shrinks. This is a modeling premise, not a mathematical error, but it is load-bearing for the headline number. The central qualitative result — that trace carbon oxides suppress escape and drive the flux toward the critical value — is more robust than the exact factor of ten. Second, the evolution model in 4.4 bakes in assumptions (fixed homopause height, no surface interactions, no loss of carbon oxides until C/H2 > 0.5) that could swing the lifetime estimate by factors of a few. That is acknowledged, but it means the \"one order of magnitude\" is a model prediction, not a measured constraint. Third, there is no archived code or simulation outputs, and the data-availability statement says data sharing not applicable. For a numerical paper of this sort, that makes the reproducibility weaker than it should be.\n\nWho it is for: escape modelers, early Earth atmospheric evolution people, and exoplanet folks thinking about H2-rich sub-Neptunes. It deserves a serious referee; the questions are about sensitivity and reproducibility, not about whether the mechanism exists.","headline":"Trace CO/CO2 suppress H2 escape on Earth-mass planets at surprisingly low mixing ratios, but the headline lifetime extension hinges on the LTE cooling assumption the authors themselves flag.","tokens_in":16038,"tokens_out":1896,"would_cite":true,"duration_ms":17409,"reading_group":"yes","serious_thinker":"yes","would_accept_peer_review":true},"rs_alignment":null,"lean_confirmation":null,"pith_extraction":{"msc":[],"pacs":[],"model":"deepseek-v4-flash","headline":"Trace carbon oxides slow hydrogen escape from early Earth tenfold.","keywords":["hydrodynamic escape","H2-rich atmosphere","carbon oxides","radiative cooling","photochemistry","early Earth","critical flux","atmospheric escape"],"falsifier":"Compute the same 1D escape flows with non-LTE level populations for CO, H2O, OH, and H3+ instead of the LTE cooling rates; if the non-LTE cooling rates in the low-density outer flow are weaker by more than the factor needed to double the escape flux, the predicted saturation at the critical flux and the order-of-magnitude lifetime extension would disappear. A simpler proxy observation is that an H2-rich atmosphere with known CO/H2 near 0.001 that keeps its escape rate rising linearly with stellar XUV flux would indicate the cooling suppression is not operating as modeled.","tokens_in":14985,"feed_emoji":"🌍","tokens_out":9706,"duration_ms":78468,"temperature":0.7,"pith_summary":"This paper asks whether trace amounts of carbon monoxide and carbon dioxide can slow the hydrodynamic escape of a hydrogen-rich atmosphere on an Earth-mass planet. Using 1D simulations with a chemical network and line-by-line radiative cooling, it finds that CO and CO2 at basal mixing fractions below about 0.01 are enough to cap the H2 escape flux near the diffusion-limited critical flux, because CO2 photolyzes to CO and atomic oxygen, the oxygen oxidizes H2 into OH and H2O, and these molecules together with CO and H3+ radiate away the XUV heat. The consequence is that an H2-rich early-Earth atmosphere survives roughly ten times longer than a pure hydrogen atmosphere, and heavy species such as carbon- and nitrogen-bearing molecules and noble gases barely escape. This matters for when Earth lost its reducing envelope and which volatiles it kept.","feed_headline":"Trace carbon oxides slow hydrogen escape from early Earth tenfold","feed_subtitle":"Even ~0.1% CO or CO2 caps H2 loss at the diffusion limit, adding ~10x to the atmosphere's lifetime","key_machinery":"The load-bearing mechanism is a radiative-cooling thermostat in the escape outflow. The model couples 1D multi-fluid hydrodynamic equations to a 287-reaction chemical network for 30 H-C-O species and to line-by-line infrared cooling rates for CO, CO2, H2O, OH, CH4, CH, CH3, H3+, OH+, and H3O+, under a young-Sun XUV spectrum 100 times the present level. In the outflow, $\\mathrm{CO_2}$ photolyzes to $\\mathrm{CO} + \\mathrm{O}$, $\\mathrm{O}$ reacts with $\\mathrm{H_2}$ to form $\\mathrm{OH}$ and $\\mathrm{H_2O}$, and these molecules radiate in the 4--15 micron region, while $\\mathrm{H_3^+}$ dominates cooling high in the flow. This cooling lowers the heating efficiency $\\eta$ (Equation 4) and pinches the H2 escape flux down to the Hunten critical flux (Equation 6), the diffusion-limited value at which H2 can barely drag heavier species out of the gravitational well.","core_discovery":"The central discovery is a saturation law: once the basal $\\mathrm{CO/H_2}$ or $\\mathrm{CO_2/H_2}$ ratio exceeds about 0.001, adding more carbon no longer lowers the escape rate, because the H2 flux asymptotes to the Hunten critical flux — the minimum hydrogen flux that can drag a heavier species out of the planet's gravitational well. In the simulations, $\\mathrm{CO_2}$ is photolyzed rapidly near the homopause into CO and atomic oxygen, and the oxygen reacts with H2 to form OH and H2O; these products, along with CO and H3+, radiate away most of the energy deposited by stellar X-ray and extreme-UV absorption, dropping the heating efficiency steeply as the carbon mixing ratio rises. The result is an H2 escape rate roughly an order of magnitude lower than in a pure hydrogen atmosphere and negligible escape of CO, CO2, heavier C/N species, and noble gases.","pith_inferences":["The paper models CO and CO2 separately from CH4; if real early atmospheres contained a mix, the escape capping would be controlled by whichever carbon carrier has the largest crossover mass and by the photochemical products it generates, so the lifetime estimate is a lower bound rather than a unique prediction.","The same cooling mechanism should operate on hydrogen-rich sub-Neptune and mini-Neptune exoplanets with trace carbon oxides, implying that their envelope-loss timescales are set by the diffusion-limited flux rather than by XUV heating alone, which could flatten the expected mass-loss trend with stellar activity.","Because the paper's cooling rates are LTE while the outer outflow is rarefied, the true early-Earth lifetime likely sits between the pure-H2 and LTE-suppressed estimates; a non-LTE version of these simulations would bracket the survival time."],"forward_implications":["At basal CO/H2 or CO2/H2 ratios below about 0.001, the H2 escape rate falls steeply as the carbon ratio rises; above about 0.001 it saturates near the critical flux and no further carbon is needed to cap escape.","An H2-rich atmosphere on early Earth containing CO or CO2 lasts about ten times longer than a pure-H2 atmosphere, exceeding 100 Myr at an initial H2 amount of roughly 200 bar equivalent.","Under the capped flux, heavier species such as CO, CO2, carbon- and nitrogen-bearing molecules, and noble gases are retained rather than dragged out, so the atmosphere's heavier inventory is largely preserved.","Radiative cooling still suppresses escape when the young-Sun XUV flux is doubled; the escape flux again asymptotes to the critical flux rather than rising linearly."],"supporting_citations":[{"why":"Supplies the 1D hydrodynamic escape model and the optically-thin LTE radiative-cooling treatment for CO that this study extends to carbon-oxide photochemistry.","marker":"Yoshida and Kuramoto (2020)"},{"why":"Establishes the earlier H2-CH4 proto-atmosphere escape results and the critical-flux interpretation that the H2-CO and H2-CO2 cases are compared against.","marker":"Yoshida and Kuramoto (2021)"},{"why":"Provides the critical flux and crossover-mass equations (Equations 6-8) that define the asymptotic escape limit and the fractionation factors.","marker":"Hunten et al. (1987)"},{"why":"Supplies the HITRAN line data used to compute the CO, CO2, H2O, OH, CH4, CH, and CH3 radiative cooling rates.","marker":"Rothman et al. (2013)"},{"why":"Supplies the ExoMol line lists used for H3+, OH+, and H3O+ cooling and for the molecular transition data.","marker":"Tennyson et al. (2016)"},{"why":"Provides the 100-Myr young-Sun X-ray and UV spectrum that sets the heating input for the escape simulations.","marker":"Claire et al. (2012)"},{"why":"Gives the radiative-transfer method used to compute the heating and photolysis rate profiles in the spherical atmosphere.","marker":"Tian et al. (2005)"},{"why":"Provides the astrochemical network that the oxygen-bearing reaction set added here is built on.","marker":"McElroy et al. (2013)"}],"fun_headline_variants":["Carbon oxides extend early Earth hydrogen atmosphere lifetime tenfold","Tiny carbon oxides give early Earth's hydrogen 10x longer life","CO and CO2 radiative cooling stretches early Earth H2 lifetime 10x","Even 0.1% carbon oxides throttle early Earth hydrogen escape"],"cache_read_input_tokens":3200,"weakest_assumption_plain":"The whole suppression rests on the assumption that the outflowing gas cools at local thermal equilibrium rates, so CO, H2O, OH, and H3+ radiate as efficiently in the low-density outer layers as in the denser thermosphere; if non-LTE effects weaken this cooling, escape speeds up and the lifetime gain shrinks.","fun_headline_variants_meta":{"raw":{"variants":["Carbon oxides extend early Earth hydrogen atmosphere lifetime tenfold","Tiny carbon oxides give early Earth's hydrogen 10x longer life","CO and CO2 radiative cooling stretches early Earth H2 lifetime 10x","Even 0.1% carbon oxides throttle early Earth hydrogen escape"]},"model":"deepseek-v4-flash","effort":"low","cost_usd":0.001092,"raw_usage":{"total_tokens":4597,"prompt_tokens":1020,"completion_tokens":3577,"prompt_tokens_details":{"cached_tokens":384},"prompt_cache_hit_tokens":384,"prompt_cache_miss_tokens":636,"completion_tokens_details":{"reasoning_tokens":3502}},"tokens_in":636,"tokens_out":3577,"duration_ms":22487,"temperature":1.0,"reasoning_tokens":3502,"cache_read_input_tokens":384,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-08-12T14:16:46.329512+00:00","model_set":{"reader":"deepseek-v4-flash"},"falsifier":"Compute the same 1D escape flows with non-LTE level populations for CO, H2O, OH, and H3+ instead of the LTE cooling rates; if the non-LTE cooling rates in the low-density outer flow are weaker by more than the factor needed to double the escape flux, the predicted saturation at the critical flux and the order-of-magnitude lifetime extension would disappear. A simpler proxy observation is that an H2-rich atmosphere with known CO/H2 near 0.001 that keeps its escape rate rising linearly with stellar XUV flux would indicate the cooling suppression is not operating as modeled.","supporting_citations":[],"review_version":1}