{"id":"8819d44d-2857-4767-87d7-91733f65c039","arxiv_id":"2411.15301","paper_version":4,"verdict":"CONDITIONAL","confidence":"MODERATE","novelty_score":5.0,"correctness_risk":"medium","formal_verification":"none","parameter_count":0,"one_line_summary":"A sharp decrease in the gravitational constant 130 Myr ago, proposed to resolve the Hubble tension, is inconsistent with solar evolution, Earth's climate history, and geological day-count records.","lead":"This paper argues that a proposed sudden drop in the gravitational constant about 130 million years ago, offered as a fix for the Hubble tension, would have caused a bright Sun, a frozen Earth, and rapid orbital changes that disagree with geological and stellar records. It compiles several independent lines of evidence that such a 'G step' is strongly disfavored.","discovery_kind":"new_application","skeptic_critique":{"model":"deepseek-v4-flash","headline":"Solar fuel-exhaustion argument depends on L⊙∝G^5.6 homology scaling not recalibrated for a recent G step; a stellar-model test is needed before excluding the GSM via helioseismology.","rationale":"The reader's weakest_assumption correctly identifies the helioseismic scaling without full recalibration and the day-count data gap. My concern sharpens the solar issue: the paper's most quantitative and impressive argument (fuel exhaustion leading to an inflated helioseismic age) depends on applying the constant-G homology exponent L⊙∝G^5.6 to a star with a recent step in G, which has not been validated by a time-dependent stellar model. This is a genuine soft spot because the exponent could vary with composition and age, and the sudden step could put the Sun out of thermal equilibrium, making the simple scaling unreliable. The paper explicitly concedes this limitation, so it is not an unacknowledged flaw, but it does mean the strongest single quantitative constraint is conditional. I do not recommend changing the verdict because: (1) the paper has several independent lines of evidence (CMB constraint in Fig. 1, day-count continuity, snowball-Earth absence) that do not rely on the precise solar scaling; (2) the reader's CONDITIONAL verdict already reflects the acknowledged caveats; and (3) the concern is testable with a concrete stellar evolution run, exactly the kind of follow-up the paper itself calls for. If the solar test weakens the argument, the paper would still stand on its other constraints, but its headline strongest claim would need qualification. Thus UNCHANGED is the appropriate verdict.","tokens_in":17541,"tokens_out":4354,"duration_ms":45781,"concrete_test":"Run a solar-calibrated stellar evolution code (e.g., MESA) with a 1 M☉ model at solar metallicity, imposing G = 1.05 G0 for t < 130 Myr ago and G = G0 afterward, and evolve to the present day. Compare the resulting central hydrogen mass fraction, present luminosity, and inferred helioseismic age against (i) a standard constant-G model and (ii) the paper's analytical estimate. If the helioseismic age remains below ~5.1 Gyr or the central hydrogen depletion differs by more than ~0.05 from the paper's 2/3 estimate, the solar argument fails; if the age exceeds 5.5 Gyr, the concern is resolved.","verdict_should_be":"UNCHANGED","load_bearing_attack":"The paper's headline helioseismic argument (Section 2.1) assumes that the luminosity scaling L⊙∝G^5.6, derived for homologous main-sequence models with constant G, applies directly to a Sun that experienced a higher G for the first ~4.4 Gyr and then a sudden drop ~130 Myr ago. This is not a trivial assumption: the exponent 5.6 comes from comparing equilibrium stellar structures at fixed mass and composition, whereas the real Sun has an evolving core composition that changes opacity, nuclear reaction rates, and the response of L to G. The paper itself flags this in Section 4: \"we did not do a full recalibration of the model for the Sun with a time-varying G. This may lead to uncharacterized theoretical uncertainties.\" The specific claim that the Sun would have exhausted about 2/3 of its fuel rather than 1/2, and hence that the helioseismic age would exceed 5.5 Gyr, rests on integrating a luminosity history that is not computed from a self-consistent stellar model with a G step. If a proper recalibration yields a less extreme luminosity enhancement or a different fuel consumption, the strongest quantitative constraint against the GSM would be substantially weakened. The day-count and CMB constraints are more independent, but the solar argument is the paper's most precise quantitative pillar, so this untested scaling assumption is the load-bearing concern.","agreement_with_reader":"partial"},"referee_report":{"model":"deepseek-v4-flash","summary":"This paper examines the G step model (GSM) proposed as a solution to the Hubble tension, in which G decreased abruptly ~130 Myr ago, and tests its implications for the Sun, the Earth, and stellar age indicators. The authors show that under the GSM the solar luminosity would have been ~30% higher in the past, which would inflate the helioseismic age and plausibly trigger a Snowball Earth; that the length of the year would jump discontinuously relative to the length of the day, in tension with geochronometry and cyclostratigraphy; and that old stars would appear ~3 Gyr younger than standard ages, conflicting with cosmic chronometers and the observed ages of the oldest stellar populations. They conclude that the GSM faces significant challenges and discuss screening-based alternatives that may avoid them.","tokens_in":17827,"tokens_out":9171,"duration_ms":86354,"significance":"The paper is a timely, readable challenge to a specific proposal for the Hubble tension. Its main value is that it confronts the GSM with several externally anchored datasets (helioseismology, meteorite ages, snowball-Earth absence, geological day counts, cosmic chronometers, and recent CMB constraints) rather than relying on the model's internal consistency alone. The CMB and day-count arguments are largely independent of the uncertain solar evolution calculation, and the paper is appropriately transparent about its limitations, explicitly stating in Section 4 that no full recalibration of solar models with time-varying G was performed and that the statistical treatment is not a full goodness-of-fit comparison. If the solar argument can be made robust by a recalibrated stellar model, the paper would constitute a strong multi-pronged exclusion; as it stands, it provides a valuable set of order-of-magnitude challenges that model-builders should address.","major_comments":[{"comment":"The helioseismic-age argument, one of the paper's main quantitative pillars, assumes that the homology scaling L⊙∝G^5.6 applies for a star whose past includes a step change in G. This is not established: a sudden change in G modifies the hydrostatic structure and nuclear burning in a time-dependent way, and the paper itself acknowledges in Section 4 that 'we did not do a full recalibration of the model for the Sun with a time-varying G'. The specific statements that the Sun would have consumed 2/3 rather than 1/2 of its fuel and would have a helioseismic age above 5.5 Gyr therefore carry an unquantified systematic uncertainty. Since this is the most precise quantitative constraint in the paper, the authors should either perform such a recalibration with a stellar evolution code or explicitly downgrade the claim to a scaling-based indication rather than a firm constraint.","section":"§2.1, §4"},{"comment":"The day-count test is weakened by the gap in the geological record at 100-200 Myr ago, which is exactly the epoch of the proposed transition. The text acknowledges this, but the statement that 'quite unusual tidal evolution of the Earth-Moon system would be required' is an extrapolation from pre-transition trends and the present lunar recession rate, not a quantitative model of LOY/LOD with a step. To make this argument a robust constraint, the authors should model the tidal evolution of the Earth-Moon system, allow for a step in G at the transition epoch, fit the model to the available geochronometric and cyclostratigraphic data, and report the posterior or likelihood for the step amplitude. Without this, the constraint is suggestive rather than conclusive.","section":"§2.3, Figure 2"},{"comment":"The stellar-age argument converts the L∝G^5.6 scaling into a global '3 Gyr age gap' for the oldest stars, but the observable stellar age distribution is not simply the constant-G age shifted by a fixed amount: stars formed over a range of redshifts, and the effect of a G step depends on formation time and subsequent evolution. The paper's connection to JWST galaxies at z>14 is qualitative. A quantitative prediction of the expected age distribution under the GSM, compared with globular cluster and field-star age determinations, is needed before this argument can serve as a firm exclusion. This does not affect the other independent arguments, but it should be framed as an indicative tension rather than a measured discrepancy.","section":"§3.1"}],"minor_comments":[{"comment":"The abstract states that the length of a year would have 'abruptly increased by about 10%', while Section 2.3 derives this value for a 'minimum plausible 5% drop' in G. Since the cosmological requirement in Equation (1) and Figure 1 corresponds to a roughly 10% higher G before the transition (and hence a ~20% longer year under LOY∝G^-2), the text should state consistently which G-drop amplitude is used for each quantitative estimate.","section":"Abstract and §2.3"},{"comment":"The derivation of T⊕∝G^1.9 is only given in a footnote; because this scaling is the basis of the glaciation argument, it would be helpful to show the steps explicitly (L⊙∝G^5.6, r∝G^-1, and T⊕∝(L/r^2)^1/4) in the main text.","section":"§2.2"},{"comment":"The caption mentions 'solid lines' representing models but does not identify which curves correspond to which model or data source; please expand the caption so the reader can distinguish data points from model curves and understand the red arrow and dashed line.","section":"Figure 2"},{"comment":"The caveats in Section 4 are welcome, but the text explicitly says the tests 'do not incorporate full goodness-of-fit comparison or hypothesis testing' and 'may be considered statistically primitive'; the numbered conclusions (i)-(vi) should be headed with this caveat so readers do not mistake them for formal exclusions.","section":"§4"}],"recommendation":"major_revision","confidential_remarks":"The paper is a challenge/commentary piece rather than a new measurement. The main risk is that its most precise quantitative argument (the solar evolution bound) rests on an un-recalibrated homology scaling that the authors themselves flag in Section 4. The day-count argument has a data gap exactly at the transition epoch. Either a stellar-model recalibration or a deliberate softening of the solar claim would substantially strengthen the paper's reliability. The discussion of screening-based alternatives is balanced and does not overclaim, and the paper would be a useful contribution after the requested revisions."},"author_rebuttal":null,"desk_editor":{"model":"deepseek-v4-flash","letter":"Colleague,\n\nQuick take on 2411.15301. The paper makes a solid case that the G step model (GSM) for the Hubble tension is in deep trouble. Its main strength is assembling several independent and mostly external constraints — solar luminosity scaling, terrestrial climate, geological day counts, stellar ages, CMB bounds — into a coherent challenge. The individual pieces are not new, but the synthesis is, and the paper uses them fairly: it cites the original GSM papers and the relevant stellar/geological literature, and it explicitly acknowledges the model's phenomenological status.\n\nWhat does it do well? It doesn't oversell. The caveats in Section 4 are unusually candid: no full solar model recalibration with time-varying G, no full goodness-of-fit. That honesty makes the arguments more credible, not less. The Figure 1 point — that the CMB already restricts G to within a few percent and the GSM requires a ~10% change — is a strong independent dagger, even before the solar-system issues.\n\nSoft spots, in proportion. The stress-test concern about L⊙ ∝ G^5.6 is real: that scaling comes from homologous main-sequence models, and the Sun with a recent G step could respond differently. The paper itself flags this in Section 4. But this isn't fatal because it only weakens the helioseismic pillar, not the whole edifice. The day-count argument also has the acknowledged 100–200 Myr data gap, but the paper argues sensibly that tidal evolution would struggle to hide a 10% jump. The statistical treatment is primitive but transparent.\n\nIs the central claim right? I think yes, with 'challenged' rather than 'ruled out'. A more fundamental theory with screening could escape some of these, as the paper notes. That's a fair framing.\n\nWho is this for? Anyone working on modified gravity, distance ladders, or the Hubble tension. It's a useful burden-shifting paper. It should definitely be sent to peer review — a good referee will tighten the solar-model caveat and maybe ask for a sensitivity analysis, but the core argument holds.\n\nMy vote: accept, with minor revisions suggested.\n\nBest.","headline":"A multi-pronged, honest critique of the G step model that lands even with its acknowledged soft spots; deserves peer review.","tokens_in":18316,"tokens_out":2126,"would_cite":true,"duration_ms":20126,"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":"A proposed 10% drop in G about 130 million years ago to resolve the Hubble tension is contradicted by the Sun's age, Earth's climate, and the geological day-count record.","keywords":["gravitation","Hubble tension","variable G","stellar evolution","helioseismology","cyclostratigraphy","Snowball Earth","Type Ia supernovae"],"falsifier":"A high-precision cyclostratigraphic record spanning the 100–200 Myr period that resolves the annual day count to better than a few percent would settle the matter: a smooth trend would falsify the GSM, while a sharp 10% drop in the reported day length at roughly 130 Myr ago would support it. Independently, a helioseismic solar age above 5.5 Gyr would confirm the faster fuel consumption the GSM demands.","tokens_in":17377,"feed_emoji":"🌍","tokens_out":9513,"duration_ms":75416,"temperature":0.7,"pith_summary":"This paper tests a proposed fix for the Hubble tension: a sudden ~10% drop in the gravitational constant G about 130 million years ago that would make distant supernovae intrinsically brighter and thus stretch inferred distances to the Hubble flow. The authors show that the same step would have altered the Sun, Earth, and Moon in ways independent observations rule out. Because stellar luminosity scales roughly as $G^{5}$.6, a slightly higher G in the past would make the Sun burn two-thirds of its fuel, pushing its helioseismic age (the age from its internal oscillations) far above the 4.567 Gyr meteorite age. The step would also drop Earth's temperature by about 10%, likely triggering a Snowball Earth, yet none occurred in the last 500 million years. And the number of days per year recorded by corals and sediment cycles shows only smooth tidal evolution, not a sudden 10% jump.","feed_headline":"A 10% drop in G 130 Myr ago fails Sun, climate, day-count tests","feed_subtitle":"The Hubble-tension fix would have made the Sun burn too fast, frozen the Earth, and left a missing 3 Gyr of stars.","key_machinery":"The argument is carried by scaling laws linking G to observables: stellar luminosity close to L ∝ $G^{5}$.6, Earth's blackbody temperature T⊕ ∝ $G^{1}$.9 after including orbital expansion with angular-momentum conservation, year length ∝ $G^{{-2}}$ while day length stays nearly constant because Earth's radius changes little, and lunar tidal stress ∝ $G^{4}$ from R ∝ 1/G and tidal force ∝ G/$R^{3}$. These relations translate the required ~10% step in G into concrete predictions for helioseismology, the geological day-count record, and the Earth–Moon tidal history, which are then checked against data.","core_discovery":"The paper claims that the G-step model cannot survive contact with Solar, terrestrial, and lunar observations. Taking the model's premise that G was about 5–10% larger until roughly 130 Myr ago and then fell abruptly, the authors derive three independent contradictions: the Sun would be too luminous over most of its history and thus appear older than 5.5 Gyr by helioseismology, whereas the measured age is below 5.1 Gyr; the Earth would have been sent into a runaway planetary glaciation by the combined drop in solar flux and orbital expansion, but the geological record shows no Snowball Earth in the past 500 Myr; and the length of a year relative to the day would have jumped by about 10%, which the cyclostratigraphic and geochronometric record of days per year does not display. The same higher-G era would make every main-sequence star burn faster, so the oldest stars would come out about 3 Gyr younger than the ages the Planck cosmology requires, leaving a gap with no stars from the first 3 Gyr of cosmic history. The paper presents these as significant challenges that any viable model of a sharp G transition would need to overcome.","pith_inferences":["The same scaling argument could be applied to any sharp-transition variant, including a step as recent as 20–40 Mpc: such a step would sit inside the Cepheid calibration zone and should show up as distance-dependent residuals in SN standardization and galaxy scaling relations.","New cyclostratigraphic data covering the 100–200 Myr gap at high resolution could turn the current day-count tension into a clean falsification or a positive detection of a 10% jump.","The CMB and BAO constraint on pre-recombination G already limits the step size to a few percent, so combining that bound with helioseismic and asteroseismic limits on other Sun-like stars (for example KIC 7970740) could sharpen the exclusion well beyond the Solar System.","A model with a gradual, secular change in G would be constrained far more tightly by the same solar and asteroseismic observations than by the CMB alone, suggesting that future asteroseismic surveys are a high-value test for any time-varying-G cosmology."],"forward_implications":["If the paper is correct, a sharp universal change in G cannot resolve the Hubble tension without also breaking the Solar System and stellar observations.","Any surviving model would need a screening mechanism that hides the change inside the Milky Way, as the paper suggests, so that stars and planets see a constant G while distant supernovae see a different value.","Distance-ladder techniques that rely on stellar luminosities would become mutually inconsistent if G changed at the required epoch, which would show up as distance-dependent offsets in Cepheid, TRGB, and surface brightness fluctuation distances.","The tightness of the radial acceleration relation, with no residual correlation with distance, already argues against a transition in stellar mass-to-light ratio inside the galaxy sample.","Cosmic chronometer reconstructions, which recover H0 within 1 km/s/Mpc of Planck, would have to be coincidentally immune to the faster stellar evolution the GSM predicts."],"supporting_citations":[{"why":"Proposes the G-step model as a Hubble-tension solution and defines the transition epoch and required change in G.","marker":"Marra & Perivolaropoulos 2021"},{"why":"Supplies the relation between G and standardized Type Ia supernova luminosity that the GSM relies on, with L ∝ G^1.46 after the Tripp correction.","marker":"Wright & Li 2018"},{"why":"Provides the L ∝ G^5.6 scaling for solar luminosity that drives the helioseismic-age and climate arguments.","marker":"degl'Innocenti et al. 1996"},{"why":"Gives the current helioseismic age of the Sun below 5.1 Gyr, which conflicts with the GSM's predicted age above 5.5 Gyr.","marker":"Bétrisey et al. 2024"},{"why":"Establishes the precise 4.567 Gyr age of the Solar System from meteorite samples, the reference point for solar evolution.","marker":"Connelly et al. 2012"},{"why":"Provides the cyclostratigraphic record of days per year used to show that the GSM's predicted 10% jump is absent.","marker":"Huang et al. 2024"},{"why":"Supplies geochronometric data on the length of day over the last 75 Myr and the present-day tidal slowdown rate.","marker":"de Winter et al. 2020"},{"why":"Shows the CMB and baryon acoustic oscillation data constrain the pre-transition G to within a few percent, already straining the GSM.","marker":"Lamine et al. 2025"}],"fun_headline_variants":["G-step model fails Sun, Earth, and star-age tests","A 10% G drop 130 Myr ago would freeze Earth and age Sun","Hubble fix G drop contradicted by Snowball-free geological record","G-step model implies missing 3 Gyr of stars; ruled out","G-step solution to Hubble tension flunks stellar and climate tests"],"cache_read_input_tokens":3200,"weakest_assumption_plain":"The argument relies on the assumption that the geological record of days per year is continuous and precise enough across the 100–200 Myr data gap to rule out a sharp 10% jump, and that the L ∝ $G^{5}$.6 solar scaling applies without a full recalibration of solar models for a time-varying G; the paper acknowledges both caveats.","fun_headline_variants_meta":{"raw":{"variants":["G-step model fails Sun, Earth, and star-age tests","A 10% G drop 130 Myr ago would freeze Earth and age Sun","Hubble fix G drop contradicted by Snowball-free geological record","G-step model implies missing 3 Gyr of stars; ruled out","G-step solution to Hubble tension flunks stellar and climate tests"]},"model":"deepseek-v4-flash","effort":"low","cost_usd":0.000318,"raw_usage":{"total_tokens":1873,"prompt_tokens":1100,"completion_tokens":773,"prompt_tokens_details":{"cached_tokens":384},"prompt_cache_hit_tokens":384,"prompt_cache_miss_tokens":716,"completion_tokens_details":{"reasoning_tokens":680}},"tokens_in":716,"tokens_out":773,"duration_ms":8200,"temperature":1.0,"reasoning_tokens":680,"cache_read_input_tokens":384,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-08-12T14:25:40.956466+00:00","model_set":{"reader":"deepseek-v4-flash"},"falsifier":"A high-precision cyclostratigraphic record spanning the 100–200 Myr period that resolves the annual day count to better than a few percent would settle the matter: a smooth trend would falsify the GSM, while a sharp 10% drop in the reported day length at roughly 130 Myr ago would support it. Independently, a helioseismic solar age above 5.5 Gyr would confirm the faster fuel consumption the GSM demands.","supporting_citations":[{"cited_title":null,"cited_arxiv_id":null,"evidence_quote":"Provides the cyclostratigraphic record of days per year used to show that the GSM's predicted 10% jump is absent."},{"cited_title":"J., et al., 2020, @doi [Paleoceanography and Paleoclimatology] 10.1029/2019PA003723 , https://ui.adsabs.harvard.edu/abs/2020PaPa...35.3723W 35, e2019PA003723","cited_arxiv_id":null,"evidence_quote":"Supplies geochronometric data on the length of day over the last 75 Myr and the present-day tidal slowdown rate."},{"cited_title":null,"cited_arxiv_id":null,"evidence_quote":"Shows the CMB and baryon acoustic oscillation data constrain the pre-transition G to within a few percent, already straining the GSM."}],"review_version":1}