{"id":"092c7436-4e1b-488d-add5-85d1e00a22b8","arxiv_id":"1908.05909","paper_version":1,"verdict":"CONDITIONAL","confidence":"MODERATE","novelty_score":6.0,"correctness_risk":"medium","formal_verification":"none","parameter_count":5,"one_line_summary":"The runaway greenhouse threshold of an Earth-sized planet varies with surface water distribution from about 130% to 180% of Earth's insolation, so the inner edge of the habitable zone depends on the planet's ocean coverage.","lead":"This paper uses 3D climate simulations to show that the amount of sunlight a rocky planet can survive before losing its ocean depends on where its oceans and continents sit, not just on the star's brightness. A dry, land-covered planet can keep liquid water at up to about 155-180% of Earth's current sunlight, while an ocean-covered planet loses it near 130%.","discovery_kind":"extension","skeptic_critique":{"model":"deepseek-v4-flash","headline":"The 10% ocean boundary and threshold curves depend on the bucket reset scheme; a sensitivity test is needed.","rationale":"The reader's weakest assumption hit the same spot. I agree that the bucket hydrology and reset are the main soft point. The paper's internal comparisons and qualitative trend are well supported, and the existence of a range is not in doubt. But the specific numbers that make the paper new—the 0.4 land-fraction boundary and the 10% VOCE water amount—are derived from a model component that is not independently validated and is not subjected to sensitivity testing. Since the paper claims a 'strong constraint' on water amounts from future observations, this quantitative boundary must be shown to be insensitive to the bucket scheme. A simple parameter sweep is feasible and would settle it. I would keep the reader's conditional recommendation rather than reject the paper, because the qualitative central claim is credible and the requested tests are straightforward.","tokens_in":71,"tokens_out":7582,"duration_ms":86563,"concrete_test":"Re-run the Earth-topography cases from Section 4 with Wg,crit = 1 cm and 50 cm (and optionally with the reset disabled so non-pool soil moisture evolves continuously without the 10 cm cap) for the water amounts 5e16, 1e17, and 5e17 m^3. Compute the runaway thresholds and diagnosed land fractions. If the land fraction at the aqua/land transition changes by more than 0.1 or the boundary water amount moves outside the 1e17-1e18 m^3 band, the central 'around 10% Earth ocean' claim is shown to be an artifact of the bucket parameters. If the thresholds and boundary remain within ~5% S0 and a factor of 2 in water amount, the concern is resolved.","verdict_should_be":"UNCHANGED","load_bearing_attack":"The central quantitative results—the collapse of the runaway threshold onto a land-fraction curve and the ~10% Earth-ocean boundary—are not controlled by the prescribed water distribution alone. In Section 3.1, after each insolation increment, surface water is reset to 100 m in the prescribed pools and to exactly 10 cm in every non-pool region that currently has beta=1, with other regions left unchanged. Because Wg,crit is set to 10 cm, any excess precipitation in non-pool regions is discarded at every reset, and the only memory retained is whether each cell was wet or dry at the end of the previous step. Thus the diagnosed land fraction (beta<1) and the threshold curves in Figures 7 and 9 are products of this binary wet/dry reset and the arbitrary choice Wg,crit=10 cm. The paper provides no sensitivity tests for these parameters. The qualitative claim that the runaway threshold depends on water distribution is robust and matches prior work, but the quantitative boundary 'around 10% of Earth's ocean' would shift if a different bucket capacity, a different critical soil moisture, or a scheme with lateral runoff were used. This is the load-bearing weakness.","agreement_with_reader":"agree"},"referee_report":{"model":"deepseek-v4-flash","summary":"This paper investigates how the surface water distribution of an Earth-sized planet affects the insolation threshold for the runaway greenhouse effect, using the CCSR/NIES AGCM5.4g three-dimensional general circulation model. The authors consider idealized meridionally uniform water distributions (dispersed and concentrated cases with 1-9 wet grid columns), distributions generated by pouring water into the topographies of Earth, Mars, and Venus, and a pseudo-Earth setup with an ocean-like surface albedo of 0.07 in water regions. They report that the runaway threshold rises from about 130% S0 for aqua planets to about 155% S0 for meridionally uniform land-planet cases, while prior zonally uniform cases reach about 180% S0. For the topographic cases, the threshold collapses onto a curve in terms of diagnosed land fraction, with an aqua/land climate boundary near a land fraction of 0.4 and a corresponding surface water amount around 10% of Earth's ocean. The paper concludes that the inner edge of the habitable zone is not uniquely determined by stellar luminosity but depends on the planet's surface water distribution.","tokens_in":20209,"tokens_out":2546,"duration_ms":26163,"significance":"If the quantitative results hold, the paper provides a useful mapping between surface water inventories and the inner edge of the habitable zone for terrestrial planets, with observational consequences: a planet with less than roughly 10% of an Earth ocean should behave climatically as a land planet, extending the habitable zone inward. The study's strengths include a systematic parameter sweep over water distributions, explicit comparisons between idealized and topographic cases, a pseudo-Earth benchmark that reproduces the previous 3-D runaway threshold near 113% S0 (comparable to Leconte et al. 2013b), and straightforward data availability via the Zenodo repository. The central qualitative conclusion, that the runaway threshold depends on the spatial arrangement of surface water and not only on insolation, is robust and consistent with prior work. The main weakness is that the quantitative boundary — the land fraction near 0.4 and the 10% ocean volume — is produced by a bucket hydrology scheme with a binary reset and a fixed critical soil moisture, with no sensitivity tests reported.","major_comments":[{"comment":"The reset procedure described in Section 3.1 is load-bearing for the quantitative claims, but its sensitivity is not assessed. After each insolation increment, all non-pool cells with beta=1 are reset to exactly 10 cm of water, while Wg,crit is set to 10 cm. This means that excess precipitation in non-pool regions is discarded at every reset, and the only memory retained by the bucket is the binary wet/dry state at the end of the previous step. The diagnosed land fraction (beta<1) in Figures 7 and 9, and the derived boundary of about 10% of Earth's ocean, are therefore outputs of this specific reset rule and the choice Wg,crit=10 cm. A different bucket capacity, a different critical soil moisture, or a scheme with lateral surface water transport could shift the threshold curves and the aqua/land boundary. The manuscript reports no sensitivity experiments for these parameters, so the central quantitative claims are not yet supported to the precision claimed.","section":"Section 3.1, reset procedure"},{"comment":"The claim that the aqua/land climate boundary lies at around 10% of Earth's ocean is based on a small number of discrete water amounts for each topography: for Earth, the boundary is bracketed between 5e16 m3 and 1e17 m3, and for Mars and Venus it lies in different ranges. The text in Section 5.1 generalizes this to 'a water planet with an amount of water less than 1e17 m3 should behave as if it has the climate of a land planet,' but this generalization rests on the same bucket-reset scheme criticized above and on only a few bracketing simulations. A quantitative statement of the uncertainty on this boundary, or an explicit statement that it is model-dependent, is needed.","section":"Section 5.1, amount of water at the boundary"},{"comment":"The runaway threshold is defined as the highest insolation for which a thermal equilibrium is maintained, with insolation incremented by 1% S0. The manuscript reports threshold values to integer precision but provides no estimate of the variability of the radiation budget or the sensitivity of the diagnosed threshold to the spin-up length or the 10-year equilibration time. Given that the differences between key cases are sometimes only 1-2% S0 (for example, the dispersed 9-grid case approaching the aqua value near 130% S0), a quantitative statement about the numerical uncertainty of the thresholds is necessary to support comparisons at this resolution.","section":"Section 3.2 and Figure 6, runaway threshold definition"},{"comment":"The topographic cases set all planetary radii to Earth's value and do not include altitude variations in the GCM, even though the water pool regions are derived from topography. This flattening removes any direct effect of elevated terrain on atmospheric temperature and circulation, and it implicitly assumes that topography only matters through the initial placement of water. The authors acknowledge this choice, and it is reasonable for isolating the effect of water distribution, but the implications for the Mars and Venus cases should be stated more explicitly: with real topography and different gravity, the diagnosed land fractions and thresholds could differ.","section":"Section 4.2, topographic water distributions"}],"minor_comments":[{"comment":"There are several typographical and reference errors: 'Natute' should be 'Nature', 'Meterology' should be 'Meteorology', 'Jhon & Sons' should likely be 'John Wiley & Sons', and 'Radiant' in the Nakajima and Tanaka reference should be 'Radiative'. These should be corrected before publication.","section":"Throughout"},{"comment":"The abbreviation VOCE is used in the text and figures but is never defined explicitly; the paper should state that VOCE denotes the volume of the present Earth's ocean (1.37e18 m3).","section":"Section 4.2"},{"comment":"The phrase 'the water poor region' appears in the text near Figure 6; this should be 'water pool region' for consistency with the rest of the manuscript.","section":"Section 3.2, Figure 6"},{"comment":"The text in Section 4.2 refers to supporting Figures S1-S3, while Section 3.1 refers to Figure S1. The numbering should be checked so that the reader can locate the water distribution figures and the stability criterion figure unambiguously.","section":"Supporting Information"},{"comment":"The pseudo-Earth experiment with 1 VOCE gives a runaway threshold of 113% S0, which the authors compare with Leconte et al. 2013b. It would be helpful to state the nominal 1% S0 increment and any interannual variability for this benchmark case, since the comparison is used to validate the model's ocean-albedo setup.","section":"Section 5.2"}],"recommendation":"major_revision","confidential_remarks":"The manuscript presents a worthwhile systematic study and the qualitative conclusion is well supported. My main concern is that the headline quantitative boundary (land fraction ~0.4, water amount ~10% of Earth's ocean) is produced by a single bucket-hydrology setup with no sensitivity analysis. I would support publication after the authors either add sensitivity tests for Wg,crit and the reset depth, or substantially soften the quantitative claims and clearly scope them to the chosen bucket scheme."},"author_rebuttal":null,"desk_editor":{"model":"deepseek-v4-flash","letter":"First thing to know: this paper shows the runaway greenhouse threshold—and hence the inner edge of the habitable zone—is not a unique function of stellar luminosity. It ranges from about 130% S0 for aqua planets to about 155% S0 for meridionally uniform land-planet cases, with prior zonal cases up to about 180%. The genuinely new pieces are the meridionally uniform dispersed/concentrated experiments and the Earth/Mars/Venus topography cases, which collapse onto a common land-fraction curve with an aqua/land boundary near 0.4. The paper does well by that claim: the threshold procedure is described clearly, the dispersed and concentrated cases bracket the range, and the pseudo-Earth check lands close to Leconte et al. 2013b (113% vs 110% S0). Data for typical figures are on Zenodo, which helps. Citation pattern is reasonable—comparisons to Kodama et al. 2018 are bracketing cases, not fitted inputs, and the Leconte benchmark is independent.\n\nThe soft spot is the bucket hydrology. The reset scheme in Section 3.1 sets non-pool wet cells back to exactly Wg,crit=10 cm at each insolation step, discarding any excess precipitation, and there is no lateral surface water transport. The diagnosed land fraction, and therefore the ~0.4 boundary and the ~10% Earth-ocean statement, are outputs of these choices. No sensitivity tests are reported for Wg,crit, field capacity, or the reset rule, and all thresholds come from one GCM without uncertainty bars. That is a genuine robustness gap, not a refutation: the qualitative dependence of the threshold on water distribution is consistent with prior 3D work and is internally coherent.\n\nOne smaller point: the abstract's \"around 10% of Earth's ocean\" is tighter than the topography results, which the discussion actually places between about 1e17 and 1e18 m3 (roughly 10–100% VOCE) depending on the planet. I wouldn't call that load-bearing, but the headline number overstates the specificity.\n\nThis paper is for people modeling exoplanet habitability and thinking about observational priors on surface water. It deserves a serious referee. I'd send it to review with a request for sensitivity tests on Wg,crit and the reset scheme, or at minimum an explicit caveat that the quantitative boundaries are bucket-scheme-dependent.","headline":"Runaway greenhouse threshold depends on surface water distribution; the new meridional and topography cases bracket the range, though the quantitative boundaries rest on bucket-hydrology choices that need sensitivity tests.","tokens_in":20757,"tokens_out":2341,"would_cite":true,"duration_ms":21610,"reading_group":"maybe","serious_thinker":"yes","would_accept_peer_review":true},"rs_alignment":null,"lean_confirmation":null,"pith_extraction":{"msc":[],"pacs":[],"model":"deepseek-v4-flash","headline":"A planet's surface water layout, not just stellar brightness, sets where its habitable zone begins.","keywords":["runaway greenhouse effect","habitable zone inner edge","surface water distribution","land planet","aqua planet","general circulation model","Simpson-Nakajima limit","exoplanet habitability"],"falsifier":"Run the same protocol with a model that includes lateral surface water transport—runoff, river flow, and ocean circulation—for a planet with about 10% of Earth's ocean at 140% S0; if it enters a runaway state like an aqua planet instead of remaining stable, or if the aqua/land boundary moves far from a land fraction of 0.4, the central claim would be contradicted.","tokens_in":19798,"feed_emoji":"🌍","tokens_out":8342,"duration_ms":76181,"temperature":0.7,"pith_summary":"This paper argues that the inner edge of the habitable zone—the distance from a star inside which all surface water vaporizes—is not fixed by stellar luminosity alone. Using a three-dimensional atmospheric general circulation model, it shows that the runaway greenhouse threshold climbs from about 130% of Earth's insolation for aqua planets to roughly 155% for land planets with meridionally spread surface water, and up to about 180% when water is confined to narrow latitude bands. The study further finds that a land fraction near 0.4, corresponding to roughly 10% of Earth's ocean water, separates aqua-like from land-like climates, and that the same boundary holds when water is distributed by Earth, Mars, or Venus topography. This matters because future observations of exoplanet climate could then constrain the amount of surface water, a quantity that is nearly impossible to measure directly from mass and radius alone.","feed_headline":"Water distribution sets where a planet's ocean boils away","feed_subtitle":"A planet with less than ~10% of Earth's ocean flips to a dry-tropics climate and can stay habitable closer to its star.","key_machinery":"The central tool is a three-dimensional atmospheric general circulation model with a bucket hydrology scheme, where evaporation efficiency rises linearly with soil moisture up to a critical value of 10 cm and stays at unity beyond it. Prescribed 'water pool' regions remain always wet, while wetness elsewhere is computed from local precipitation minus evaporation, producing the land fraction that organizes the results. The runaway threshold is located by stepping up insolation until the climate can no longer reach a balanced radiation budget. The physical mechanism is the width of the wet tropical atmosphere: a moist equator caps outgoing longwave radiation at the Simpson-Nakajima limit, lowering the threshold, whereas dry tropical regions let the planet radiate more and push the threshold outward.","core_discovery":"The runaway greenhouse threshold—the insolation beyond which a planet with liquid surface water can no longer maintain thermal equilibrium and all water evaporates—ranges from about 130% S0 for aqua planets to about 155% S0 for meridionally uniform land planets, and can be as high as about 180% S0 when the water distribution is zonally confined. The threshold is set by how much of the low-latitude atmosphere stays wet: dry tropical regions allow outgoing longwave radiation to exceed the Simpson-Nakajima limit, while wet tropics cap the radiation and force runaway at lower insolation. Across idealized and topography-based water distributions, a land fraction of about 0.4 marks the boundary between aqua-planet and land-planet climate regimes, corresponding to a surface water amount near 10% of Earth's ocean. Therefore the inner edge of the habitable zone is not uniquely determined by the central star's luminosity; it depends on the planet's own surface water distribution.","pith_inferences":["If the land-fraction boundary near 0.4 holds across different models, then exoplanet mass-radius measurements—which cannot detect water inventories below roughly 1% of Earth's ocean—could be complemented by photometric or spectral signatures of a dry-tropics climate to estimate water amount.","The meridionally dispersed case being the lower limit suggests that the commonly quoted 180% S0 land-planet threshold from zonal studies may be an upper bound; real planets with any tropical wet regions will likely have lower runaway thresholds.","A natural next test is to include lateral surface water transport and a dynamic ocean in the same protocol; if the aqua/land boundary shifts significantly away from 0.4, the 10%-ocean threshold would need revision.","The paper's idealized 1-bar N2 atmosphere omits CO2 and the moist greenhouse limit, so coupling the runaway threshold with atmospheric chemistry could alter the habitable-zone width, although the authors argue pCO2 has little effect on the runaway threshold itself."],"forward_implications":["A water-poor planet with a dry tropics region can remain stable at insolation levels up to roughly 155% S0 (meridional spread) or about 180% S0 (zonal confinement), while an aqua planet runs away near 130% S0.","The aqua/land climate boundary sits at a land fraction near 0.4, which for realistic topography corresponds to about 10% of Earth's ocean, so small changes in water content produce large, observable changes in climate.","For a Sun-like star, the inner edge of the habitable zone can extend inward to roughly 0.75 AU when the runaway threshold is 180% S0, meaning a low-water planet could stay habitable at Venus's orbit.","When ocean albedo is set to a realistic low value (0.07), the runaway threshold for an Earth-like ocean is about 113% S0, and it rises to about 178% S0 at 1% of Earth's ocean, so surface reflectivity strongly influences the habitable edge.","Because the climate boundary falls at a water amount that is difficult to measure from mass-radius relations, atmospheric observations could serve as a proxy for surface water inventory on exoplanets."],"supporting_citations":[{"why":"Provides the zonally uniform surface water distribution results, the water-flow-limit concept, and the upper-limit runaway thresholds up to about 180% S0 that this paper brackets.","marker":"Kodama et al. [2018]"},{"why":"Introduced land planets and showed that dry tropical regions allow high outgoing longwave radiation, the physical mechanism behind higher runaway thresholds.","marker":"Abe et al. [2011]"},{"why":"Supplies a three-dimensional GCM aqua-planet runaway threshold near 110% S0 and the unsaturated Hadley-cell mechanism for comparison.","marker":"Leconte et al. [2013b]"},{"why":"Provides another three-dimensional GCM aqua-planet runaway threshold near 121% S0 and evidence of stable climates under strong insolation.","marker":"Wolf and Toon [2015]"},{"why":"Defines the Simpson-Nakajima radiation limit from one-dimensional radiative-convective equilibrium, the cap on outgoing radiation from a wet atmosphere.","marker":"Nakajima et al. [1992]"},{"why":"Supplies the bucket model for soil moisture and evaporation efficiency used to compute surface wetness and land fraction.","marker":"Manabe [1969]"},{"why":"Gives the runaway greenhouse limit value and the terminology 'Simpson-Nakajima limit' used to interpret the thresholds.","marker":"Goldblatt et al. [2013]"},{"why":"Provides the one-dimensional estimate (about 104% S0) that frames the older, lower runaway threshold against which the three-dimensional results are compared.","marker":"Kopparapu et al. [2013]"},{"why":"Supplies the spherical-harmonic topography coefficients used to construct Earth, Mars, and Venus surface water distributions.","marker":"Hirt et al. [2012]"}],"fun_headline_variants":["Water map decides when a planet's ocean boils away","Dry tropics let planets hug their star longer","Habitable zone inner edge depends on water layout","Ocean coverage shifts runaway greenhouse limit","10% of Earth's ocean flips planet climate regime"],"cache_read_input_tokens":3200,"weakest_assumption_plain":"The assumed rule for how wet the ground outside the prescribed water pools stays only uses local rain minus evaporation, with a fixed critical soil moisture of 10 cm and no sideways flow of surface water, so if real moisture recycling or runoff behaves differently the computed land fractions and threshold curves would shift.","fun_headline_variants_meta":{"raw":{"variants":["Water map decides when a planet's ocean boils away","Dry tropics let planets hug their star longer","Habitable zone inner edge depends on water layout","Ocean coverage shifts runaway greenhouse limit","10% of Earth's ocean flips planet climate regime"]},"model":"deepseek-v4-flash","effort":"low","cost_usd":0.000223,"raw_usage":{"total_tokens":1498,"prompt_tokens":1029,"completion_tokens":469,"prompt_tokens_details":{"cached_tokens":384},"prompt_cache_hit_tokens":384,"prompt_cache_miss_tokens":645,"completion_tokens_details":{"reasoning_tokens":399}},"tokens_in":645,"tokens_out":469,"duration_ms":4877,"temperature":1.0,"reasoning_tokens":399,"cache_read_input_tokens":384,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-08-14T13:01:28.730239+00:00","model_set":{"reader":"deepseek-v4-flash"},"falsifier":"Run the same protocol with a model that includes lateral surface water transport—runoff, river flow, and ocean circulation—for a planet with about 10% of Earth's ocean at 140% S0; if it enters a runaway state like an aqua planet instead of remaining stable, or if the aqua/land boundary moves far from a land fraction of 0.4, the central claim would be contradicted.","supporting_citations":[{"cited_title":"This GCM was developed for the modeling of the present Earth’s climate [Numaguchi, 1999] and was applied to the paleoclimate of Earth [Abe-Ouchi et al., 2013]","cited_arxiv_id":null,"evidence_quote":"Introduced land planets and showed that dry tropical regions allow high outgoing longwave radiation, the physical mechanism behind higher runaway thresholds."},{"cited_title":null,"cited_arxiv_id":null,"evidence_quote":"Supplies the spherical-harmonic topography coefficients used to construct Earth, Mars, and Venus surface water distributions."}],"review_version":1}