{"id":"99e746d2-6545-4c8c-80df-f56c18072048","arxiv_id":"1908.08952","paper_version":3,"verdict":"REJECT","confidence":"HIGH","novelty_score":6.0,"correctness_risk":"high","formal_verification":"none","parameter_count":5,"one_line_summary":"Nonlinear cosmic density growth is reported to scale as the square of initial density and the square of cosmic time, but the derivation's central averaging step is unsupported.","lead":"This paper derives a power law for how cosmic density contrasts grow during nonlinear structure formation, and claims galaxy survey data confirm it. The data match well, but a key step in the derivation is not justified, so the central claim is not established as stated.","discovery_kind":"first_principles","skeptic_critique":{"model":"deepseek-v4-flash","headline":"Equation 6's conditional-averaging identity is false under linear theory, so the derivation of the quadratic growth law is unsupported; the central claim should not be accepted as analytic.","rationale":"I read the full derivation in §2 with the reader's concern in mind. The weakest point is indeed Eq 6: the replacement of a conditional integral over an initially overdense subset by the unrestricted volume integral. This step is used to drop the '1' in (1+δ), making the first term vanish by Gauss's theorem. In linear theory the velocity divergence and density contrast are perfectly correlated in the growing mode, so the conditional mean of ∇·u given δ=δp cannot equal the global mean. The phrase 'modes grow independently' refers to Fourier modes evolving linearly, not to the conditional expectation of a velocity field over a density threshold; these are unrelated. Equation 9 compounds the problem by discarding a nonzero term of order (∇·u)^2, again with no justification. Because Eq 11 follows only if both Eqs 6 and 9 hold, the claimed exact analytic law α=2, β=2 is not established. I credit the observational analysis: the CSI data, Monte Carlo completeness corrections, and the systematic binning variants in §7 are substantial, and the measured scaling might survive as a useful empirical description of how density percentiles evolve. The paper even acknowledges in §2 that the intermediate density domain remains unsolved, yet the abstract claims an exact analytic description extensible to arbitrarily low redshift. That overreach is precisely where the invalid derivation matters. The reader's verdict of REJECT is therefore appropriate, and my stress-test does not change it.","tokens_in":19318,"tokens_out":6659,"duration_ms":70378,"concrete_test":"Evaluate Eq 6 in linear perturbation theory: for a Gaussian initial density field, compute the volume average of ∇·u over the subset where δ(m,τnl)=δp using the linear relation ∇·u = -aH f δ. The result is -aH f δp, not zero, directly falsifying Eq 6. If the authors dispute the analytic check, repeat with a small N-body simulation: tag particles by initial density percentile and compare their mean velocity divergence at z=1 with the global mean; a difference beyond noise confirms the conditional average does not equal the unconditional average.","verdict_should_be":"UNCHANGED","load_bearing_attack":"The derivation's core step, Eq 6, asserts that the velocity divergence integrated over the subset of mass parcels with initial overdensity δ(m,τnl)=Qδ(p) equals the integral over the entire volume. The justification, 'modes grow independently prior to nonlinearity,' does not imply statistical independence of ∇·u and δ. In the linear growing mode, ∇·u is proportional to δ: (∇·u)/(aH) = -f δ with f≈Ω_m^0.55. Hence the conditional mean of ∇·u over parcels with initial overdensity δp is -aH f δp, not the global mean (zero for a periodic volume). The left side of Eq 6 is therefore -aH f δp V_p while the right side is 0; the equality fails at first order in δp. Consequently the '+1' term in Eq 5 is not removed, and Eq 7 does not follow. Independently, Eq 9 discards the term involving (Dδ/Dτ)(∇·u), which equals -(1+δ)(∇·u)^2. That term is nonzero and on nonlinear scales is generally comparable to the retained δ∇²φ term. Without Eqs 6 and 9, Eq 11—the origin of α=2 and β=2—is not derived from the fluid equations. The empirical fit in §7 is therefore a phenomenological scaling, not confirmation of an analytic derivation.","agreement_with_reader":"agree"},"referee_report":{"model":"deepseek-v4-flash","summary":"The paper claims a new analytic description of nonlinear structure growth: after overdense regions decouple from the Hubble expansion, fixed percentiles of the real-space matter density distribution should grow as δ(t) ∝ δ_0^α t^β with α = 2 and β = 2. This is derived in Section 2 by volume-averaging the Lagrangian fluid equations, and the prediction is tested against environmental densities measured from the Carnegie-Spitzer-IMACS Redshift Survey over 0.2 < z < 1.5, yielding α = 1.98 ± 0.04 and β = 2.01 ± 0.11.","tokens_in":19623,"tokens_out":3032,"duration_ms":33077,"significance":"If the derivation and measurement were both sound, the result would be significant: it would provide a simple analytic law for the growth of density percentiles into the nonlinear regime and would give the first direct observational confirmation of such a law. The paper also contains substantial empirical work: careful SED fitting, Monte Carlo tests of incompleteness corrections, Delaunay-based density estimation, and robustness checks over many redshift and mass binning choices. However, the theoretical core has two unproven and apparently incorrect steps, and the empirical validation is partly circular because the initial density distribution is inferred from the same data with the growth law already assumed. The paper's central claims are therefore not established.","major_comments":[{"comment":"The replacement of the conditional volume integral of ∇·u over parcels with initial overdensity Qδ(p) by the full-volume integral is asserted without support. In the linear growing mode, ∇·u/(aH) = -f δ with f ≈ Ω_m^0.55, so the conditional mean of ∇·u over parcels with δ(m, τ_nl) = δ_p is -aH f δ_p, not the global mean (zero for a periodic volume). At first order in δ_p, Eq. (6) therefore fails, and the '+1' term in Eq. (5) is not removed. Since Eq. (7) and the subsequent derivation all rest on this step, the quadratic growth law does not follow from the fluid equations as presented.","section":"§2, Eq. (6)"},{"comment":"Taking the Lagrangian time derivative of the product δ (∇·u) in Eq. (8) produces two terms, but Eq. (9) retains only the term involving δ ∇·(Du/Dτ). The discarded term is (Dδ/Dτ)(∇·u) = -(1+δ)(∇·u)^2 by Eq. (1), which is nonzero and on nonlinear scales is generally comparable to the retained term. No argument is given for its vanishing. Without this additional and unjustified truncation, Eq. (11) is not derived.","section":"§2, Eqs. (8)–(9)"},{"comment":"The empirical confirmation is partly circular. The initial lognormal A(p) = exp[σ Φ^{-1}(p)] is inferred from the same CSI percentile tracks in Section 6 after assuming β = 2, and σ is fitted. The relation B(p) ∝ [A(p)-1]^α is then compared with the data, but A(p) is constructed from those data. In addition, the free normalization γ in Eq. (17) absorbs the predicted prefactor C, so the agreement of α and β with (2, 2) tests only the shape of the growth law, not the quantitative normalization of Eq. (14). The reported posteriors therefore do not constitute an independent confirmation of the analytic derivation.","section":"§6 and §7, Eqs. (16)–(17)"},{"comment":"The low-density branch of Eq. (14) is introduced through a brief chain-rule argument involving δ′ = ⟨ρ⟩/ρ − 1, but the derivation is not shown. In particular, the factor (1+δ_p)^2 and the statement about mass conservation in the p < 0.5 ensemble require a careful accounting of how parcels leave the percentile ensemble; as written, this part of the result is not established at the same level as the high-density branch.","section":"§2, Eq. (14)"}],"minor_comments":[{"comment":"Typo: 'distrbution' should be 'distribution'.","section":"§6, first paragraph"},{"comment":"Typo: 'Eulerian' should be 'Eulerian'.","section":"§8, final paragraph"},{"comment":"A number of spelling errors appear: 'outlyers' should be 'outliers', 'correspondance' should be 'correspondence', and 'stricly' should be 'strictly'.","section":"Throughout"},{"comment":"The low-density percentiles with Φ^{-1}(p) < -1 are excluded from the fit, but the model curves are still shown for them using dashed lines; the text should state explicitly that those dashed curves are extrapolations, not fits.","section":"§7, Fig. 11 caption"}],"recommendation":"reject","confidential_remarks":"The empirical analysis is substantial and the survey work is careful, but the central theoretical derivation in Section 2 relies on two unproven identities (Eqs. 6 and 9) that appear to be false even in a linear-theory test. Because the paper's headline claim—exact analytic α = 2, β = 2 growth laws confirmed by data—depends directly on those steps, the error cannot be repaired by local revision. I recommend rejection, though the authors may be able to recast the empirical scaling as a phenomenological result if they remove the unsupported analytic claims."},"author_rebuttal":null,"desk_editor":{"model":"deepseek-v4-flash","letter":"Quick take: this paper has a nice idea and a lot of good data work, but the central theoretical claim is not established. The authors want to show that fixed percentiles of the matter density grow as delta0^2 t^2 after decoupling from the Hubble flow. The Lagrangian percentile framing is genuinely new, and the empirical measurement from CSI — 7 Gyr of density percentile tracks with careful completeness corrections — is a serious piece of work. The reported alpha=1.98±0.04 and beta=2.01±0.11 match the predicted law. That coincidence is worth taking seriously.\n\nThe problem is Section 2. Equation 6 replaces the integral of div u over the subset of parcels with initial overdensity Q_delta(p) by the integral over the whole volume. The justification (\"modes grow independently\") does not do the work. In linear theory, div u = -aH f delta, so the conditional mean over parcels with delta=delta_p is proportional to delta_p, not zero. Gauss's theorem only kills the full-volume integral. So the first term in Eq 5 does not vanish, and Eq 7 does not follow. Equation 9 has a second issue: the time derivative of the product delta (div u) produces an extra -(1+delta)(div u)^2 term, which is discarded without argument. On nonlinear scales that term is the same order as what is kept. Both steps are load-bearing for the alpha=2, beta=2 result. I don't see a way to rescue the derivation as written.\n\nThe empirical validation also carries a circularity burden. The initial lognormal A(p) is inferred from the same percentile tracks after assuming beta=2; the normalization gamma in Eq 17 is a free parameter; and the fit excludes the lowest-density percentiles. So the agreement with alpha=beta=2 is a fitted scaling, not an independent test of the analytic law. It is still a useful phenomenological measurement.\n\nBottom line: the paper deserves a serious referee because the question is important and the CSI dataset is valuable. But in current form the central claim is unsupported. I would recommend major revision at best, with the derivation either fixed or reframed as an ansatz.","headline":"The analytic derivation of the alpha=2, beta=2 growth law is not sound — two missing steps in the Lagrangian averaging undo the theory — but the CSI empirical scaling is a solid phenomenological result that deserves referee time.","tokens_in":20173,"tokens_out":3125,"would_cite":false,"duration_ms":31994,"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":"This paper derives and observationally confirms a closed-form law: fixed percentiles of the cosmic density field grow as the square of initial overdensity and time.","keywords":["nonlinear structure growth","Lagrangian fluid equations","density percentiles","lognormal density field","galaxy environment density","gravitational collapse","redshift survey","cosmic structure"],"falsifier":"Run an N-body simulation, label mass parcels by their initial overdensity at the decoupling epoch, and measure the conditional mean of $\\nabla\\cdot \\mathbf{u}$ within each initial-overdensity bin; if it is not consistent with zero, equation (6) fails and the quadratic growth law does not follow. Alternatively, measure the percentile growth exponents directly in the simulation: values of $\\alpha$ or $\\beta$ differing from 2 would falsify the claim.","tokens_in":19114,"feed_emoji":"🌌","tokens_out":9623,"duration_ms":88415,"temperature":0.7,"pith_summary":"This paper argues that the nonlinear growth of cosmic structure can be described exactly, without N-body simulations, once density fluctuations have decoupled from the Hubble expansion. Working in Lagrangian coordinates and volume-averaging the fluid equations, the authors show that the density at a fixed percentile of the real-space distribution should grow as $\\delta(t) \\propto \\delta_0^2\\,t^2$, with the same quadratic dependence for the hollowing-out of underdense regions. They test this against the evolution of galaxy stellar-mass density percentiles across roughly seven billion years in a 9.5 square degree redshift survey, measuring $\\alpha = 1.98 \\pm 0.04$ and $\\beta = 2.01 \\pm 0.11$. If the derivation and measurement hold, this supplies the first exact analytic bridge from early linear fluctuations to the fully nonlinear regime and a new, distribution-based way to interpret cross-sectional galaxy surveys.","feed_headline":"Density peaks grow as the square of their starting contrast and time","feed_subtitle":"A 7-Gyr galaxy survey measures α=1.98±0.04 and β=2.01±0.11, matching the analytic law.","key_machinery":"The central object is the volume-averaged Lagrangian continuity equation written in terms of the overdensity $\\delta(m,\\tau)$, with mass parcels labeled by a coordinate $m$, together with Gauss's divergence theorem. The pivotal step is equation (6), which asserts that the volume integral of $\\nabla\\cdot \\mathbf{u}$ over parcels selected by a fixed initial overdensity equals the unrestricted volume integral and therefore vanishes; this removes the '1' in $(1+\\delta)$ from the averaged continuity equation. What remains yields $\\langle D^2\\delta/D\\tau^2\\rangle \\propto \\delta_p^2$, the quadratic-in-initial-density law that integrates to the $\\alpha=2$, $\\beta=2$ trajectory. The companion piece is the lognormal quantile function $Q_\\delta(p)=\\exp[\\sigma\\,\\Phi^{-1}(p)]-1$, with $\\Phi^{-1}$ the probit function, which maps observed percentiles back to the inferred initial overdensity spectrum.","core_discovery":"The paper claims that after a density fluctuation decouples from the Hubble expansion, the mean evolution of its overdensity is governed by the volume-averaged Lagrangian fluid equations, and that for the ensemble of mass parcels sharing an initial overdensity $\\delta_p$ the result is $\\langle D^2\\delta/D\\tau^2\\rangle = \\frac{3}{2}\\Omega_M H_0^2 a^{-1}\\delta_p^2$. Integrating once in time gives a mean growth rate proportional to $\\delta_p^2(\\tau-\\tau_{nl})$, and integrating again gives the quadratic trajectory $\\delta(\\tau)-\\delta(\\tau_{nl}) \\propto \\delta_p^2(\\tau-\\tau_{nl})^2$, so $\\alpha=2$ and $\\beta=2$; underdense percentiles drain according to a mirrored law with an extra $(1+\\delta_p)^2$ factor. The paper reports that fixed percentiles of the local stellar-mass density distribution in the CSI survey evolve with $\\alpha = 1.98 \\pm 0.04$ and $\\beta = 2.01 \\pm 0.11$, and that extrapolating each percentile back to the start of galaxy growth recovers a lognormal initial density distribution with $\\sigma = 0.82 \\pm 0.01$. These results are presented as the first exact, analytic description of nonlinear structure growth that extends to arbitrarily low redshift and as evidence that early lognormal fluctuations grew by gravitational accretion.","pith_inferences":["If the conditional-mean step at equation (6) survives scrutiny, the same volume-averaging trick may yield analytic growth laws for velocity statistics or higher-order density moments, not just percentile trajectories.","Because the $\\alpha=2$, $\\beta=2$ law should hold for any tracer of the underlying density field, applying the same analysis to X-ray clusters, HI maps, or lensing maps would test whether baryonic tracers follow identical exponents or reveal feedback-induced deviations.","The exponent pair ($\\alpha,\\beta$) is a redshift-independent benchmark; deviations measured in a particular percentile or scale could be converted into constraints on assembly bias or on the epoch of nonlinearity $z_{nl}$.","A straightforward simulation test is to repeat the CSI percentile analysis on mock catalogs with full selection effects: recovering $\\alpha = \\beta = 2$ would close the loop, while any offset would quantify the systematic error budget of the measurement."],"forward_implications":["Percentile-by-percentile density growth in the fully nonlinear regime is predictable from a closed algebraic form, so the growth of structure no longer needs to be treated as a purely numerical problem.","Evolving observed density percentiles backwards recovers the initial lognormal spectrum with $\\sigma = 0.82 \\pm 0.01$, giving a direct empirical handle on the density field at the start of star formation.","Because the derived form extends to arbitrarily low redshift, galaxy growth and the turnover in the cosmic star formation rate density can be modeled analytically rather than only through Monte Carlo techniques.","The inferred Hurst parameter $H=1$ for accretion means the scatter around mean galaxy growth relations is itself correlated signal, so cross-sectional surveys should be analyzed distributionally rather than by tracking medians alone.","Any additional physics, such as baryonic feedback or environmental effects, should enter as modified boundary or initial conditions, making measurements of $\\alpha$ and $\\beta$ a diagnostic for such physics."],"supporting_citations":[{"why":"Supplies the survey design, near-infrared selection, spectroscopic redshifts, and completeness corrections that produce the stellar-mass density maps.","marker":"Kelson et al. 2014"},{"why":"Provides the lognormal model for the density distribution that the paper uses to map observed percentiles back to initial overdensities.","marker":"Coles & Jones 1991"},{"why":"Provides the MDPL2 N-body simulation used to estimate cosmic variance in CSI-like volumes and to check that mock catalogs show similar percentile evolution.","marker":"Klypin et al. 2016"},{"why":"Provides the mock galaxy catalogs built from MDPL2 that quantify cosmic variance and validate the observed percentile trends.","marker":"Knebe et al. 2018"},{"why":"Fixes the cosmological parameters ($\\Omega_M$, $h$, $\\Omega_\\Lambda$) that set the normalization $C=\\frac{3}{4}\\Omega_M H_0^2(1+z_{nl})$ in the predicted growth law.","marker":"Planck Collaboration et al. 2015"},{"why":"Supplies the distribution-free binomial method used to assign confidence intervals to the measured density percentiles.","marker":"Meeker, Hahn, & Escobar 2017"}],"fun_headline_variants":["Density peaks grow as δ₀²t², CSI survey confirms α≈2, β≈2","Exact nonlinear growth law: δ ∝ δ₀² t², verified by 7-Gyr data","Quadratic law for cosmic structure growth, now analytic and observed","First exact model of nonlinear growth: δ(t)=δ₀² t², survey matches"],"cache_read_input_tokens":3200,"weakest_assumption_plain":"The derivation hinges on assuming that the average expansion term within each group of mass parcels that began with the same initial density is the same as the average over all space, namely zero; if that conditional average is not zero, the predicted $\\delta_0^2 t^2$ growth law does not follow.","fun_headline_variants_meta":{"raw":{"variants":["Density peaks grow as δ₀²t², CSI survey confirms α≈2, β≈2","Exact nonlinear growth law: δ ∝ δ₀² t², verified by 7-Gyr data","Quadratic law for cosmic structure growth, now analytic and observed","First exact model of nonlinear growth: δ(t)=δ₀² t², survey matches"]},"model":"deepseek-v4-flash","effort":"low","cost_usd":0.000276,"raw_usage":{"total_tokens":1779,"prompt_tokens":1212,"completion_tokens":567,"prompt_tokens_details":{"cached_tokens":384},"prompt_cache_hit_tokens":384,"prompt_cache_miss_tokens":828,"completion_tokens_details":{"reasoning_tokens":468}},"tokens_in":828,"tokens_out":567,"duration_ms":6446,"temperature":1.0,"reasoning_tokens":468,"cache_read_input_tokens":384,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-08-14T11:26:37.314208+00:00","model_set":{"reader":"deepseek-v4-flash"},"falsifier":"Run an N-body simulation, label mass parcels by their initial overdensity at the decoupling epoch, and measure the conditional mean of $\\nabla\\cdot \\mathbf{u}$ within each initial-overdensity bin; if it is not consistent with zero, equation (6) fails and the quadratic growth law does not follow. Alternatively, measure the percentile growth exponents directly in the simulation: values of $\\alpha$ or $\\beta$ differing from 2 would falsify the claim.","supporting_citations":[{"cited_title":"2018, MNRAS, 474, 5206 Le Fevre, O., Vettolani, G., Maccagni, D., et al","cited_arxiv_id":null,"evidence_quote":"Provides the mock galaxy catalogs built from MDPL2 that quantify cosmic variance and validate the observed percentile trends."},{"cited_title":"Q., Hahn, G","cited_arxiv_id":null,"evidence_quote":"Supplies the distribution-free binomial method used to assign confidence intervals to the measured density percentiles."}],"review_version":1}