{"id":"73afa9de-52cf-479c-aad0-57f6fa081f40","arxiv_id":"2602.05575","paper_version":1,"verdict":"CONDITIONAL","confidence":"MODERATE","novelty_score":4.0,"correctness_risk":"medium","formal_verification":"none","parameter_count":5,"one_line_summary":"Fisher forecasts show BINGO and SKA1-MID 21-cm intensity mapping, combined with Planck priors, could constrain the f(R) gravity parameter B0 down to ~10⁻⁶–10⁻⁸.","lead":"This paper forecasts how well future radio surveys (BINGO and SKA1-MID) will constrain f(R) modified gravity using 21-cm hydrogen intensity mapping together with Planck CMB priors. Its headline numbers say SKA Band 2 could measure the f(R) Compton-wavelength parameter B0 to σ≈6×10⁻⁸.","discovery_kind":"extension","skeptic_critique":{"model":"deepseek-v4-flash","headline":"The claimed σ(B0) is set by the highest ℓ in the analysis: Fig. 5 shows B0 does not saturate by ℓmax=400, yet those multipoles correspond to k≳0.3 Mpc⁻¹ where linear theory and unscreened f(R) are invalid. The headline uncertainty is therefore cutoff-dependent and likely too small.","rationale":"The central claim is that SKA Band 2 can constrain B0 to 6×10⁻⁸, three orders of magnitude below current limits. The argument is a Fisher forecast; the key condition for this claim to hold is that the model for C_l and its derivative with respect to B0 is accurate over the range of ℓ that actually drives the Fisher information. I examined whether this condition is met. The forecast uses linear perturbation theory and the quasi-static μ,γ parameterization (Eqs. 12-13) out to ℓmax=400. At z≈0.3, this reaches k≈0.3–0.4 Mpc⁻¹, where nonlinear corrections to the matter power spectrum are significant and the f(R) fifth force is screened by the chameleon mechanism. Because the f(R) scale-dependent signal grows with k², the Fisher derivative is weighted toward the highest k; hence any error in the high-k model translates directly into an error in σ(B0). This is confirmed by Fig. 5, where B0 is the slowest-converging parameter and does not plateau by ℓmax=400. The authors' own conclusion admits that nonlinear effects are omitted. I considered other possible weaknesses—the fixed Ω_HI, the boundary at B0=0—but they are secondary: Ω_HI uncertainty is partly degenerate with As and b_HI, and the boundary issue affects interpretation of the interval more than the headline sensitivity order-of-magnitude. The nonlinearity/cutoff issue is thus the single most load-bearing concern. It does not, however, invalidate the paper's methodology as a forecast; it means the numbers should be conditioned on the robustness of the small-scale model. A concrete re-computation with a k-cut or a larger ℓmax would settle whether the concern lands. Since the reader already judged the paper CONDITIONAL with moderate confidence, and this concern matches that judgment, I leave the verdict unchanged.","tokens_in":14982,"tokens_out":21658,"duration_ms":224734,"concrete_test":"Recompute the Fisher matrix for SKA Band 2 (and BINGO) using the same code with three choices: (i) ℓmax=400 (published), (ii) a conservative per-redshift k_max defined by σ(R)=0.5 (roughly k_NL≈0.2–0.3 Mpc⁻¹ at z≲0.5, corresponding to ℓmax≈200–250), and (iii) ℓmax=800 with the same linear theory. If σ(B0) from (ii) is more than ~3× larger than from (i), or if (iii) yields >10% further improvement over (i), the headline constraint is not converged and is inflated by scales where the model is invalid. If (ii) and (iii) bracket (i) tightly, the nonlinearity concern is resolved.","verdict_should_be":"UNCHANGED","load_bearing_attack":"Section V reports σ(B0)≃6.37×10⁻⁸ for SKA Band 2 and 3.75×10⁻⁸ with Planck priors. These numbers are produced by a Fisher matrix whose signal is evaluated with linear perturbation theory and the quasi-static μ,γ parameterization of Eqs. (12)–(13) with no cutoff up to ℓmax=400. At the redshifts probed (z≲0.5), ℓ=400 corresponds to k≈ℓ/χ(z)≈0.3–0.4 Mpc⁻¹, firmly in the mildly nonlinear regime, where nonlinear structure formation and the chameleon screening mechanism of f(R) gravity modify the matter power spectrum in ways not captured by Eqs. (12)–(13). The concern is not stylistic: because μ−1 grows like k² at small B0, the Fisher derivative ∂C_l/∂B0 is dominated by the largest k included, so the constraint is set by the highest multipoles. Fig. 5 is consistent with this: the B0 curves (right axes) are the slowest to converge and do not show a clear plateau at ℓmax=400; e.g., for SKA Band 2 the ratio σ(B0;ℓmax)/σ(B0;400) remains very large at ℓmax≈100 and is still dropping at 400. Thus the quoted σ(B0) is an artifact of the arbitrary ℓmax=400 cutoff and of the omitted nonlinear/screening physics. The authors themselves acknowledge in the Conclusion that nonlinear effects are not included, but the headline claim is presented without this caveat. If a physically motivated k-cut (k_NL~0.2–0.3 Mpc⁻¹) is imposed, σ(B0) could increase substantially, undermining the central claim of probing f(R) orders of magnitude below current constraints.","agreement_with_reader":"agree"},"referee_report":{"model":"deepseek-v4-flash","summary":"This paper presents Fisher-matrix forecasts for the f(R) gravity Compton-wavelength parameter B0 using HI 21-cm intensity mapping, for BINGO, SKA1-MID Band 1, and SKA1-MID Band 2, both alone and combined with Planck priors. The signal is modeled with angular power spectra based on density and redshift-space distortion terms, using a quasi-static mu,gamma parameterization of f(R) gravity; thermal and shot noise are included. The headline results are sigma(B0) = 2.27e-6 for BINGO, 4.48e-6 for SKA Band 1, 6.37e-8 for SKA Band 2, and 3.75e-8 for SKA Band 2 + Planck, which the paper interprets as showing that future 21-cm IM surveys can probe f(R) gravity orders of magnitude below current bounds.","tokens_in":15494,"tokens_out":5696,"duration_ms":61464,"significance":"If the forecasts are robust, the paper would demonstrate a genuinely powerful new probe of modified gravity: SKA Band 2 constraints on B0 would be roughly three to four orders of magnitude tighter than current limits. The paper is clearly structured, uses a standard Fisher formalism, and includes a convergence test in Fig. 5. However, the central numbers rest on linear perturbation theory and quasi-static screening-free f(R) modeling up to multipoles where the signal is not converged, and a key input, Omega_HI, is never specified. These issues must be addressed before the quantitative claims can be accepted.","major_comments":[{"comment":"The headline sigma(B0) values are set by multipoles where the linear quasi-static treatment is not valid. In the f(R) parameterization, mu-1 grows approximately as lambda^2 k^2 for small B0, so the Fisher derivative is strongly weighted toward the largest k included. At z≈0.27, ell_max=400 corresponds to k≈0.3-0.4 Mpc^-1, well inside the mildly nonlinear regime, and the chameleon screening that suppresses f(R) modifications in collapsed regions is absent from Eqs. (12)-(13). Fig. 5's right axes show that sigma(B0;ell_max)/sigma(B0;400) is very large at low ell_max and still decreasing at 400, so the constraint is dominated by scales where the model is least trustworthy. The Conclusion acknowledges that nonlinear effects are not included. Please recompute with a physically motivated k_max (e.g., k≈0.2-0.3 Mpc^-1) and report how sigma(B0) changes; this is a necessary robustness check for t","section":"§V, Fig. 5, Eqs. (12)-(13), Eq. (38)"},{"comment":"The signal amplitude, and therefore all Fisher errors, depend on Omega_HI(z), but no fiducial value is given anywhere. Equation (15) defines T_bar_b proportional to h Omega_HI, and Eq. (38) uses this T_bar_b for both signal and shot noise, yet Table II lists fiducial values only for the seven parameters in Eq. (34). Without Omega_HI, the forecast is not reproducible and the absolute scale of sigma(B0) cannot be checked. Please state the assumed value and redshift dependence, if any, and provide a sensitivity test over a plausible range of Omega_HI.","section":"§III, Eq. (15)"}],"minor_comments":[{"comment":"The text and Table I adopt a 10 MHz channel bandwidth, but the captions of Figs. 1 and 2 quote 9.33 MHz. Since thermal noise in Eq. (29) depends on delta_nu, these must be harmonized.","section":"§IV A, Table I, Figs. 1-2"},{"comment":"The noise term N_l(zi,zj) B_l(zi,zj) is not specified for i != j. If N_l is diagonal as stated, the expression should be written with the diagonal restriction made explicit.","section":"§IV B, Eq. (38)"},{"comment":"The definition of B is typeset ambiguously; the factors H, dR/dtau, and (dH/dtau - H^2) are not clearly attached. Please clarify the notation.","section":"§II, Eq. (5)"},{"comment":"The integral for the source density has no lower limit and uses chi^2(z) without defining the variable; please specify the integration range and all symbols.","section":"§IV B, Eq. (33)"},{"comment":"The manuscript contains numerous typographical errors and missing spaces (e.g., 'FLR W metrcds', 'opetaion', 'Figrue', 'decipted', 'tunned', 'examlpe'), and the title formatting is inconsistent. A careful language and proofreading pass is recommended.","section":"General"}],"recommendation":"major_revision","confidential_remarks":null},"author_rebuttal":null,"desk_editor":{"model":"deepseek-v4-flash","letter":"Short version: this is a competent but fairly standard Fisher forecast for the f(R) parameter B0 from HI 21cm intensity mapping. The specific numbers for BINGO and SKA1-MID Band 1/2 are new, and the pipeline is the same as the group's earlier paper [40]. The headline σ(B0) ~ 6×10^-8 for SKA Band 2 is not robust as it stands, and the paper contains a couple of internal inconsistencies that need attention.\n\nWhat it does well: the calculation is transparent, uses the standard MGCAMB quasi-static parameterization, includes RSD, thermal and shot noise, and shows how Planck priors break degeneracies. The convergence study in Fig. 5 is good practice, even if it undercuts the headline.\n\nWhere it is soft: the main issue is the ℓmax=400 cutoff. At z~0.5, ℓ=400 corresponds to k~0.3–0.4 Mpc^-1, where linear theory and the unscreened quasi-static f(R) approximation are not reliable. f(R) deviations grow like k^2, so the Fisher derivative is dominated by the highest k. Fig. 5 shows exactly that: the B0 ratio is still falling at ℓmax=400. The text says B0 is driven by large-scale modes, but the figure implies the opposite. So the quoted σ(B0) is effectively set by an arbitrary cutoff and by physics that is not modeled. That needs to be fixed, either with a physically motivated k-cut or a proper nonlinear/screening treatment.\n\nAlso, the paper never gives the value of Ω_HI, which sets the signal amplitude and is fixed in the Fisher matrix. That makes the forecast unreproducible. There is also a 10 MHz vs 9.33 MHz inconsistency between Table I and Figs. 1–2. And Fisher errors at B0=0 are quoted as Gaussian intervals despite the physical boundary; that is a minor but real overstatement.\n\nIs the central idea wrong? No. The method is standard, and the forecast is plausible in spirit. But the headline orders-of-magnitude claim depends on scales that the model does not treat. The paper is useful for readers who want a worked example of a 21cm IM Fisher forecast, but the specific numbers should be treated with caution.\n\nVerdict: this deserves peer review because it is a new, concrete forecast from a known pipeline, but it needs major revision. I would send it to a referee rather than desk reject it.","headline":"A workmanlike Fisher forecast for f(R) gravity with HI intensity mapping, but the headline σ(B0) is set by mildly nonlinear scales beyond the model's validity and needs a physically motivated cutoff before being taken at face value.","tokens_in":15986,"tokens_out":3791,"would_cite":false,"duration_ms":40914,"reading_group":"maybe","serious_thinker":"yes","would_accept_peer_review":true},"rs_alignment":null,"lean_confirmation":null,"pith_extraction":{"msc":[],"pacs":["98.80.-k","04.50.Kd"],"model":"deepseek-v4-flash","headline":"Forecast: HI 21-cm surveys can pin the f(R) gravity parameter B0 down to 3.75×10⁻⁸","keywords":["f(R) gravity","21-cm intensity mapping","B0 parameter","Fisher matrix forecast","SKA1-MID","BINGO","angular power spectrum","modified gravity"],"falsifier":"Recompute the same forecasts with a nonlinear prescription or with a conservative small-scale cutoff (e.g., k_max ≈ 0.2 Mpc⁻¹) and compare the resulting σ(B0); if the constraint degrades by more than a factor of a few, the linear-only claim is falsified. Observationally, a measurement of the 21-cm angular power spectrum from SKA1-MID Band 2 that does not show the predicted scale-dependent growth enhancement at ℓ≈100–400 would also rule out the claimed sensitivity.","tokens_in":14877,"feed_emoji":"📡","tokens_out":7665,"duration_ms":71863,"temperature":0.7,"pith_summary":"This paper forecasts how well future neutral-hydrogen 21-cm intensity mapping surveys — BINGO and the two low-frequency bands of SKA1-MID — can measure B0, the parameter that sets the Compton wavelength of the scalaron in f(R) gravity. Using a Fisher-matrix analysis of the 21-cm angular power spectrum, the authors argue that even the pathfinder BINGO telescope would reach σ(B0) ≈ 2.3×10⁻⁶, and that SKA1-MID Band 2, especially when combined with CMB priors, would reach σ(B0) ≈ 3.75×10⁻⁸. That would test deviations from General Relativity several orders of magnitude more stringently than current cosmological constraints, which sit near 10⁻⁴. The gain comes from the scale-dependent growth of structure that f(R) imprints on the 21-cm signal at small angular scales.","feed_headline":"21-cm maps could constrain f(R) gravity to 10⁻⁸","feed_subtitle":"Forecast says SKA1-MID Band 2 plus CMB priors reaches σ(B0)=3.75×10⁻⁸, far below today's limits.","key_machinery":"The forecasts are carried by the quasi-static μ(a,k) and γ(a,k) functions that encode how f(R) modifies the Poisson equation and the metric anisotropy ratio, together with the 21-cm angular power spectrum C_l built from the density and redshift-space-distortion terms of the brightness temperature. The Fisher matrix then converts these spectra and the survey's thermal and shot noise into marginalized uncertainties. The named central object is B0, the present-day Compton wavelength of the scalaron in units of the Hubble length; B0=0 recovers ΛCDM.","core_discovery":"The paper's central claim is that low-redshift 21-cm intensity mapping measures not just the background expansion but the growth of structure, and that this gives it strong leverage on f(R) gravity. Building the brightness-temperature angular power spectrum from the density and redshift-space distortion terms, and using the quasi-static μ,γ parameterization for f(R) perturbations, the forecast finds σ(B0) ≈ 6.37×10⁻⁸ for SKA1-MID Band 2 alone and σ(B0) ≈ 3.75×10⁻⁸ when CMB priors are added. The authors also argue that most of the constraining power for B0 comes from multipoles up to a few hundred, where the linear approximation is expected to hold.","pith_inferences":["If nonlinear structure growth or chameleon screening suppresses the small-scale enhancement on k≳0.3 Mpc⁻¹, the real SKA Band 2 constraint on B0 will be weaker than 3.75×10⁻⁸; a simulation-based forecast with a nonlinear cutoff would test this directly.","The B0–bHI degeneracy shown in the paper is only partially broken by CMB priors; cross-correlating the 21-cm maps with a galaxy catalog that independently measures the HI bias could tighten B0 further.","The same μ,γ formalism is specific to f(R), but the scale-dependent 21-cm power spectrum would also carry signatures of other modified-gravity theories, so the survey will yield general growth-of-structure tests.","A practical next step is to verify whether the predicted scale-dependent bump in C_l at ℓ≈100–400 survives realistic foreground subtraction; the forecast assumes this cleanly, making it the riskiest observational step."],"forward_implications":["BINGO alone would reach σ(B0) ≈ 2.27×10⁻⁶, already surpassing current cosmological limits near 10⁻⁴.","SKA1-MID Band 2 alone yields σ(B0) ≈ 6.37×10⁻⁸, and combining with CMB priors tightens this to ≈ 3.75×10⁻⁸.","Adding CMB priors breaks the strong degeneracy between B0 and the Hubble parameter, shrinking the allowed region dramatically.","Constraints on most parameters saturate by ℓ_max ≈ 300–400, and B0 is dominated by large-scale modes, where linear theory is safest."],"fun_headline_variants":["SKA1-MID 21-cm maps forecast f(R) gravity to 10⁻⁸","21-cm intensity mapping forecast tightens f(R) bounds to 10⁻⁸","BINGO and SKA1-MID: 21-cm surveys as gravity probes","Future 21-cm surveys could pin f(R) gravity to 10⁻⁸","SKA1-MID 21-cm forecast: σ(B0)=6×10⁻⁸"],"cache_read_input_tokens":2304,"weakest_assumption_plain":"The forecast assumes the linear, quasi-static f(R) approximation holds on the small scales (ℓ up to 400, k≳0.3 Mpc⁻¹) that provide most of the B0 signal; if nonlinearities or screening become important there, the quoted σ(B0) values are too optimistic.","fun_headline_variants_meta":{"raw":{"variants":["SKA1-MID 21-cm maps forecast f(R) gravity to 10⁻⁸","21-cm intensity mapping forecast tightens f(R) bounds to 10⁻⁸","BINGO and SKA1-MID: 21-cm surveys as gravity probes","Future 21-cm surveys could pin f(R) gravity to 10⁻⁸","SKA1-MID 21-cm forecast: σ(B0)=6×10⁻⁸"]},"model":"deepseek-v4-flash","effort":"low","cost_usd":0.000869,"raw_usage":{"total_tokens":3654,"prompt_tokens":852,"completion_tokens":2802,"prompt_tokens_details":{"cached_tokens":256},"prompt_cache_hit_tokens":256,"prompt_cache_miss_tokens":596,"completion_tokens_details":{"reasoning_tokens":2680}},"tokens_in":596,"tokens_out":2802,"duration_ms":19395,"temperature":1.0,"reasoning_tokens":2680,"cache_read_input_tokens":256,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-08-03T04:11:46.466777+00:00","model_set":{"reader":"deepseek-v4-flash"},"falsifier":"Recompute the same forecasts with a nonlinear prescription or with a conservative small-scale cutoff (e.g., k_max ≈ 0.2 Mpc⁻¹) and compare the resulting σ(B0); if the constraint degrades by more than a factor of a few, the linear-only claim is falsified. Observationally, a measurement of the 21-cm angular power spectrum from SKA1-MID Band 2 that does not show the predicted scale-dependent growth enhancement at ℓ≈100–400 would also rule out the claimed sensitivity.","supporting_citations":[],"review_version":1}