{"id":"a1e7d650-ef36-4a29-ba0a-756304ac3d64","arxiv_id":"2608.03441","paper_version":1,"verdict":"CONDITIONAL","confidence":"MODERATE","novelty_score":6.0,"correctness_risk":"medium","formal_verification":"none","parameter_count":13,"one_line_summary":"The 21-cm power spectrum, not the sky-averaged signal alone, can distinguish a redshift-evolving galaxy population from a static one during the cosmic dawn.","lead":"Using simulations of the early universe, this paper shows the sky-averaged 21-centimeter radio signal alone cannot tell apart galaxies whose star formation changes with time from galaxies whose star formation does not. Adding measurements of how patchy that signal is breaks the ambiguity, which will help interpret JWST's surprising abundance of bright galaxies at early times.","discovery_kind":"new_method","skeptic_critique":{"model":"deepseek-v4-flash","headline":"The 'definitively broken' claim rests on a single hand-tuned static model; the static parameter space (beta_L, alpha, SFE shape) is not systematically searched, so a static population matching both global signal and power spectrum may exist.","rationale":"The paper's demonstration is instructive: a hand-tuned static model reproduces the Fiducial global absorption trough while producing a lower power spectrum, suggesting that the power spectrum carries information beyond the global signal. The numerical framework is carefully validated (two independent RT schemes, convergence tests in appendices). However, the headline claim that the degeneracy is 'definitively broken' is stronger than the evidence. The static model family is defined by redshift-independent power-law efficiencies (eqs. 2.6, 2.17), but within that family the tuning explores only a subspace: f_L and f_X normalizations and beta_X, with beta_L = 0 and alpha = 1.5 fixed. Figure 2 shows beta_L and beta_X substantially change the power spectrum, and positive beta_L would increase the bias of Ly-alpha sources, potentially raising the Ly-alpha peak. Similarly, varying alpha or the static SFE slope could alter the X-ray peak. Without a systematic search, the existence of a static model matching both global and power spectrum cannot be ruled out. This is the single most load-bearing assumption: if such a model exists, the central conclusion fails. The single-k comparison (k = 0.05 h/Mpc) compounds the issue, since the scale dependence of the difference is not shown. A systematic scan, feasible given SCRIPT's speed, would settle this. Thus the appropriate verdict is conditional acceptance: the method and qualitative insight are publishable, but the claim should be qualified or supported by a fuller search.","tokens_in":22296,"tokens_out":7682,"duration_ms":65653,"concrete_test":"Run a systematic grid or MCMC scan over the Static model's radiation parameters (f_L, beta_L, f_X, beta_X, alpha) and optionally the static SFE parameters, using SCRIPT, with the constraint that the global delta_Tb(z) matches the Fiducial model within a tolerance (e.g., 2 mK) over z = 6-25. For each accepted model, compute the 21-cm power spectrum at multiple k (e.g., 0.01, 0.05, 0.1, 0.5 h/Mpc) and test whether any model matches the Fiducial power spectrum within, say, 20% at all k. If at least one static model matches both, the 'definitively broken' claim is falsified; if none, it is supported.","verdict_should_be":"UNCHANGED","load_bearing_attack":"The central claim that the 21-cm power spectrum definitively breaks the degeneracy between redshift-evolving and static galaxy populations (abstract; §3.3, Fig. 3; §4) depends on the assertion that no static population can reproduce both the Fiducial global signal and its power spectrum. The evidence is a single 'simple trial-and-error search' (§3.3) that tunes f_L (normalization only, beta_L fixed to 0), f_X, and beta_X while leaving the SED index alpha = 1.5 and the static star-formation parameters fixed to their z~6 UVLF calibration (Table 1). This ignores physically relevant degrees of freedom that the paper itself shows control the power spectrum: eq. (2.6) allows beta_L to shift Ly-alpha emission to more clustered high-mass halos, and Fig. 2 (bottom panels) demonstrates that beta_L and beta_X strongly modulate the peak amplitudes. A static model with positive beta_L and a shallower beta_X (or a different alpha) could plausibly match the global signal while raising the Ly-alpha and X-ray peaks toward the Fiducial amplitudes. Because the search space is not exhaustively explored, 'definitively broken' is not established. Additionally, the comparison is shown only at k = 0.05 h/Mpc (Fig. 3); even if the Static-Tuned model is unique, the conclusion would need to hold over the observable k range. The load-bearing assumption is that the adopted power-law, redshift-independent parameterization is complete enough to represent any static population.","agreement_with_reader":"agree"},"referee_report":{"model":"deepseek-v4-flash","summary":"This paper extends the SCRIPT semi-numerical framework to include 21-cm spin-temperature fluctuations driven by inhomogeneous Lyman-alpha coupling and X-ray heating during the Cosmic Dawn. The authors compare a JWST-calibrated Fiducial galaxy model with a redshift-evolving star-formation efficiency against a Static, redshift-independent baseline model. They show that a tuned Static model (f_L=10, f_X=5e-4, beta_X=-3) can reproduce the Fiducial global 21-cm absorption trough, but that the corresponding 21-cm power spectrum at k=0.05 h/Mpc is substantially lower, and conclude that combining global and power-spectrum measurements definitively breaks the degeneracy between evolving and static galaxy populations.","tokens_in":22670,"tokens_out":4262,"duration_ms":40207,"significance":"If the central claim is made robust, the paper is valuable: it demonstrates a concrete route toward distinguishing evolving from static source populations with 21-cm observables, and it extends SCRIPT into a fast, photon-conserving Cosmic Dawn tool. The paper's strengths are its two independently implemented radiative-transfer schemes (Source and Sink) that cross-validate each other, convergence tests over spatial resolution and redshift step, the explicit treatment of sub-grid Lyman-shell weighting, and the calibration of source models against JWST UV luminosity functions and reionization constraints. The quantitative prediction that a static-population fit to the global signal underproduces the 21-cm power spectrum is falsifiable with upcoming interferometers, which makes the framework a useful basis for future parameter estimation.","major_comments":[{"comment":"The central claim that the degeneracy is 'definitively broken' rests on a single hand-tuned Static-Tuned model found by a 'simple trial-and-error search'. That search varies f_L, f_X, and beta_X while fixing beta_L=0, alpha=1.5, and the static SFE shape. This is not a sufficient exploration of the static-model parameter space. The paper's own Fig. 2 bottom-left panel shows that positive beta_L increases the Lyman-alpha peak amplitude, and Eq. (2.6) permits beta_L to shift Lyman-alpha emission toward more clustered high-mass halos. A static model with beta_L>0 and a less negative beta_X could plausibly match the global signal while raising the power-spectrum peaks toward Fiducial values. To support 'definitively broken', the authors should either scan the static parameter space, including beta_L and alpha, and show that no model reproduces both the global signal and the power spectrum, or rephrase the conclusion as a proof-of-concept demonstration. As written, the load-bearing assumption of parameterization completeness is unverified.","section":"§3.3, Table 1, Fig. 3"},{"comment":"The 21-cm power-spectrum comparison is shown only at k = 0.05 h/Mpc. The abstract and Section 4 state that the tuned static model predicts lower amplitude 'across all characteristic peaks', but this is a statement about one wavenumber across redshift. The conclusion would be much more robust if the power spectra were compared over the observable k range, for example k = 0.01-1 h/Mpc, or if the authors justified why k = 0.05 h/Mpc is representative. Without this, the 'definitively broken' claim is not established for future interferometric measurements that will probe a range of scales.","section":"Fig. 3, §4"},{"comment":"The Static-Tuned global signal is described as 'nearly indistinguishable' from the Fiducial model, but no quantitative metric is defined. Since the subsequent argument relies on the two global signals being consistent within observational uncertainty, the authors should specify a tolerance, such as a residual rms in mK or a chi-square relative to the expected noise of upcoming global-signal experiments. A quantitative criterion would make the degeneracy claim concrete and reproducible.","section":"§3.3"}],"minor_comments":[{"comment":"The phrase 'definitively broken' is stronger than the currently presented evidence; consider 'can be broken' or 'strongly disfavored' for the specific Static-Tuned model until the parameter-space exploration is broadened.","section":"Abstract, §4"},{"comment":"The redshift-integrator convergence test with Delta z=1 is described as 'not shown here'. Since convergence claims are an important part of the paper's methodological contribution, this test should either be shown in a figure or summarized with concrete numbers.","section":"Appendix B.2"},{"comment":"The paper notes that thermal effects of Lyman-alpha photons are neglected, citing the authors' own earlier work on this effect. A brief quantitative statement of the expected magnitude at the redshifts of interest would help readers assess the impact of this approximation.","section":"§2.2"},{"comment":"The bottom panels of Fig. 2 show power spectra at a single wavenumber k=0.05 h/Mpc; this should be stated in the caption explicitly, as the same convention is used throughout.","section":"Fig. 2 caption"}],"recommendation":"major_revision","confidential_remarks":"The paper is within scope and the code extension appears credible. My main concern is the gap between the single Static-Tuned example and the 'definitively broken' conclusion. I believe a major revision that either adds a systematic exploration of the static parameter space or rephrases the result as a proof-of-concept would make the central claim defensible. The convergence tests and dual radiative-transfer schemes are genuine strengths."},"author_rebuttal":null,"desk_editor":{"model":"deepseek-v4-flash","letter":"First thing you should know: this is a solid technical paper with an overclaimed headline. The SCRIPT extension to spin-temperature fluctuations is carefully built, and the demonstration that a static galaxy population can mimic a JWST-motivated evolving population's global trough while producing a different power spectrum is genuinely instructive. The two independent source/sink algorithms and the convergence checks in the appendices give me real confidence that the numerical results are not grid artifacts.\n\nWhat's new: combining a JWST-calibrated evolving SFE model with a static counter-model tuned to the same global signal, then comparing the power spectra, is not in the literature in this form. The qualitative point that fluctuations retain clustering-bias memory is established in the broader 21-cm literature, but this concrete quantitative example is a useful addition. The paper is also honest about what the Static model can and cannot do.\n\nThe soft spot is exactly where the stress-test note points: the 'definitively broken' claim rests on one hand-tuned Static-Tuned model (f_L = 10, f_X = 5e-4, beta_X = -3). The search was trial-and-error and covered only f_L, f_X, and beta_X, with beta_L = 0, alpha = 1.5, and the SFE slope fixed to the z~6 calibration. Since the paper's own Fig. 2 shows that beta_L and beta_X strongly modulate the power-spectrum peaks, a static model with positive beta_L or a different SED shape could plausibly raise the Ly-alpha and X-ray peaks toward the Fiducial amplitudes. So 'definitively' is not established. The comparison is also shown only at k = 0.05 h/Mpc, and there is no noise or foreground forecast, so the practical claim that current or near-term experiments could actually see this difference is untested.\n\nNone of this kills the paper. The qualitative conclusion that the power spectrum carries structural memory and can break degeneracies the global signal cannot is well supported. What needs to change is the wording and ideally the evidence: a small grid search over beta_L, alpha, and the SFE shape would either strengthen or temper the claim, and a multi-k plot would make the k-dependence clear.\n\nCitation pattern looks fine; prior SCRIPT and Raste-Sethi work is cited appropriately. Data availability is 'upon request,' which is weaker than ideal for a semi-numerical code paper, but not disqualifying.\n\nWho gets value: anyone working on 21-cm Cosmic Dawn theory, JWST-galaxy synthesis, or forecasting SKA-era constraints. I'd send it to a referee, with conditional acceptance as the likely outcome, but I'd ask for the strong claim to be qualified or supported by a broader search and an observational feasibility estimate.\n\nRecommendation: engage with it. The method deserves refereeing; the headline should be softened.","headline":"Solid SCRIPT extension with a clear global-signal degeneracy demo, but the 'definitively broken' claim outruns the evidence; referee it with a request to qualify or broaden the static-model search.","tokens_in":23206,"tokens_out":2394,"would_cite":true,"duration_ms":21908,"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 static galaxy population can mimic the 21-cm global signal but not its power spectrum.","keywords":["21-cm cosmology","Cosmic Dawn","spin temperature fluctuations","Lyman-alpha coupling","X-ray heating","JWST high-redshift galaxies","UV luminosity functions","power spectrum degeneracy"],"falsifier":"Run the same SCRIPT comparison with a static star-formation model but allow the X-ray efficiency to vary with redshift, or adopt a harder X-ray spectral energy distribution, and check whether a static population can match both the Fiducial global absorption trough and the Fiducial power-spectrum amplitudes at all three peaks; if such a model exists, the structural-memory degeneracy breaking is not definitive.","tokens_in":22059,"feed_emoji":"📡","tokens_out":8295,"duration_ms":72567,"temperature":0.7,"pith_summary":"The paper tackles a practical question: if JWST is right that galaxies evolved rapidly at z>10, can 21-cm observations of neutral hydrogen prove it? It argues that the sky-averaged global 21-cm signal alone cannot, because a galaxy population with no redshift evolution can be tuned to reproduce the absorption trough of an evolving population. The spatial power spectrum can, however, because the fluctuations remember the clustering bias of the halos that emitted the radiation: the tuned static model predicts substantially lower power-spectrum amplitudes at all characteristic peaks. If correct, this means the combination of global-signal and power-spectrum measurements can robustly constrain the redshift evolution of high-redshift galaxies.","feed_headline":"Static galaxies fake the 21-cm trough but fail the power spectrum","feed_subtitle":"Fluctuations remember halo clustering bias, so global-plus-power-spectrum measurements can pin down galaxy evolution.","key_machinery":"The central object is the 21-cm power spectrum, evaluated at k=0.05 h/Mpc, together with the halo population that produces the radiation. The framework SCRIPT computes inhomogeneous spin-temperature fluctuations from Lyman-alpha coupling and X-ray heating using two independent photon-conserving schemes, Source and Sink, cross-validated for resolution convergence. The argument-carrying identity is that the amplitude of the power spectrum depends on the square of the global signal times the spatial clustering of the emitting halos; low-mass halos are more abundant but less clustered, so shifting emission to them suppresses the power spectrum even when the photon budget is identical. This is the structural memory of the source halo bias.","core_discovery":"The paper demonstrates a degeneracy and its resolution. Using the photon-conserving SCRIPT framework extended to spin-temperature fluctuations, it builds a Fiducial galaxy model with redshift-evolving star-formation efficiency that matches JWST UV luminosity functions, and a Static model with redshift-independent efficiency. It then finds a Static-Tuned model with f_L=10, f_X=5e-4, beta_X=-3 whose global absorption trough nearly exactly matches the Fiducial one: the steep negative X-ray mass scaling shifts heating to abundant, low-mass halos, masking the absence of evolution. Yet the 21-cm power spectrum of Static-Tuned has systematically lower amplitudes at the Lyman-alpha, X-ray heating, and reionization peaks, because the spatial fluctuations retain a structural memory of the clustering bias of the source halos. This breaks the degeneracy that global-signal measurements alone cannot resolve, and the paper concludes that combining global and fluctuation measurements is essential to robustly constrain the redshift evolution of high-redshift galaxies.","pith_inferences":["Inference: the structural-memory argument generalizes beyond this specific pair of models: any two source populations with the same photon budget but different halo-mass weighting should be separable by the 21-cm power spectrum, since the fluctuation amplitude encodes the bias of the dominant emitters.","Inference: a wider static parameter space could still hide a model that matches both the global signal and the power spectrum; the paper's trial-and-error search does not rule that out, so the 'definitively broken' claim should be read as a demonstration within this parameterization.","Inference: the same Fiducial-versus-Static comparison could be run for the reionization-era part of the signal around z~7, where UV ionizing photons dominate, to test whether the bias memory persists when the driver is ionizing radiation rather than Lyman-alpha and X-ray photons.","Inference: if a real global-signal detection is accompanied by low power-spectrum amplitudes at all peaks, that would point toward radiation dominated by abundant, weakly clustered low-mass halos — a distinctive signature of the static-type population rather than an evolving one."],"forward_implications":["A static galaxy population can reproduce the global absorption trough of an evolving population only by shifting X-ray heating to low-mass halos, so global-signal fits by themselves cannot identify the galaxy population.","The 21-cm power spectrum separates the two populations: the tuned static model has systematically lower amplitudes at the Lyman-alpha, X-ray heating, and reionization peaks even when the global signals match.","Combining global and fluctuation measurements is the route to robustly constraining the redshift evolution of high-redshift galaxies, complementing bright-end UV luminosity function constraints.","The upgraded SCRIPT framework, with its resolution-converged dual radiative-transfer schemes, is suited to fast parameter-space exploration and emulator-based inference for upcoming low-frequency arrays."],"supporting_citations":[{"why":"Provides the explicitly photon-conserving SCRIPT foundation that this work extends to spin-temperature fluctuations.","marker":"[84]"},{"why":"Supplies the star-formation efficiency parameterization and UVLF calibration on which the Fiducial and Static galaxy models are built.","marker":"[85]"},{"why":"The excursion-set semi-numerical framework whose filtering approach the Sink method parallels; establishes a comparison baseline for fast 21-cm simulations.","marker":"[72]"},{"why":"Provides the Lyman-series cascade probabilities used to compute the injected Lyman-alpha contribution to spin-temperature coupling.","marker":"[89]"},{"why":"Gives the X-ray photon flux and heating formulation used to compute inhomogeneous heating before reionization.","marker":"[96]"},{"why":"Supplies the secondary ionization and heating fractions used in the X-ray heating and ionization rates.","marker":"[99]"},{"why":"Provides JWST UV luminosity function measurements at z=8-15 used to calibrate the Fiducial model's redshift-evolving star-formation efficiency.","marker":"[13]"},{"why":"Provides JWST UV luminosity function measurements at z=9-16 that anchor the high-redshift galaxy abundance the Fiducial model is designed to match.","marker":"[14]"}],"fun_headline_variants":["Power spectrum breaks 21-cm global-signal degeneracy","Static galaxies fake the trough, not the power spectrum","21-cm fluctuations remember halo bias, break degeneracy","Global 21-cm signal fools, power spectrum does not","Cosmic Dawn galaxy evolution sussed by 21-cm power spectrum"],"cache_read_input_tokens":3200,"weakest_assumption_plain":"The conclusion assumes that the simple redshift-independent power-law parameterization used for the static galaxy population covers every plausible static population; if a wider class of static models were allowed and one of them matched both the global signal and the power spectrum, the claimed degeneracy breaking would fail.","fun_headline_variants_meta":{"raw":{"variants":["Power spectrum breaks 21-cm global-signal degeneracy","Static galaxies fake the trough, not the power spectrum","21-cm fluctuations remember halo bias, break degeneracy","Global 21-cm signal fools, power spectrum does not","Cosmic Dawn galaxy evolution sussed by 21-cm power spectrum"]},"model":"deepseek-v4-flash","effort":"low","cost_usd":0.000838,"raw_usage":{"total_tokens":3683,"prompt_tokens":1004,"completion_tokens":2679,"prompt_tokens_details":{"cached_tokens":384},"prompt_cache_hit_tokens":384,"prompt_cache_miss_tokens":620,"completion_tokens_details":{"reasoning_tokens":2594}},"tokens_in":620,"tokens_out":2679,"duration_ms":17164,"temperature":1.0,"reasoning_tokens":2594,"cache_read_input_tokens":384,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-08-15T14:50:21.127890+00:00","model_set":{"reader":"deepseek-v4-flash"},"falsifier":"Run the same SCRIPT comparison with a static star-formation model but allow the X-ray efficiency to vary with redshift, or adopt a harder X-ray spectral energy distribution, and check whether a static population can match both the Fiducial global absorption trough and the Fiducial power-spectrum amplitudes at all three peaks; if such a model exists, the structural-memory degeneracy breaking is not definitive.","supporting_citations":[],"review_version":2}