{"id":"e25e9135-91e8-4fa5-8130-b4effb5539d7","arxiv_id":"2501.15117","paper_version":1,"verdict":"REJECT","confidence":"MODERATE","novelty_score":3.0,"correctness_risk":"high","formal_verification":"none","parameter_count":5,"one_line_summary":"A forecast study claims SKA with 0.001 Hz spectral resolution could measure cosmic redshift drift at millimeter-per-second precision over 0.5 years, yielding tight dark energy constraints.","lead":"This paper forecasts how the Square Kilometre Array could measure cosmic redshift drift using hydrogen gas in galaxies and quasar absorption systems over just six months. If the forecast holds, the method would test dark energy models in real time, much faster than decade-long telescope programs.","discovery_kind":"extension","skeptic_critique":{"model":"deepseek-v4-flash","headline":"Eq. (9), the linchpin error forecast, is unverifiable and numerically inconsistent: stated inputs give ~10^-5 cm/s, not the plotted 0.01–0.03 cm/s.","rationale":"I read the paper as a forecasting exercise, not a real measurement; the Sandage-Loeb concept and the comparisons with Kloeckner et al. (2015) are legitimate. But a forecast is only useful if the error model is reproducible and internally consistent. The weakest point is indeed Eq. (9). The reader correctly flags the unpublished σ_n normalization; I find a more specific numerical inconsistency: substituting the paper's own stated numbers into Eq. (9) yields error bars orders of magnitude smaller than those plotted and used in the likelihood. Because every parameter constraint in the paper is driven by these σ_i, the central precision claim is not supported even on the paper's own terms. I therefore leave the reader's reject verdict unchanged. The objection is to the argument, not to the authors.","tokens_in":17755,"tokens_out":5818,"duration_ms":48847,"concrete_test":"Recompute the Fig. 5 error bars directly from Eq. (9) using the stated inputs: N=10^7, Δν=0.001 and 0.002 Hz, λ=1.09 and 1.52, and σ_n=1–5 cm/s, at each redshift bin z=0.1–1.0. If the resulting σ_i differ from the plotted 0.01–0.03 cm/s by more than a factor of ~3, the forecast is internally inconsistent. Then independently derive σ_n from first principles for a 100 mJy source with 0.5-year integration on SKA, and compare with the value required to reproduce the plotted error bars.","verdict_should_be":"UNCHANGED","load_bearing_attack":"The central claim — sub-mm/s redshift-drift sensitivity with 0.001/0.002 Hz SKA observations and the resulting dark-energy constraints — depends entirely on the per-bin velocity error σ_i from Eq. (9). That equation is calibrated by an unpublished 'Kang 2024, in prep' normalization σ_n ∈ [1,5] cm/s, with no derivation. More decisively, Eq. (9) is not consistent with the numbers plotted. Direct substitution of the values stated in §4 for the 0.001 Hz case (N=10^7 per 0.1 bin, λ=1.09, Δν=0.001 Hz) at z≈0.5 gives Δv_obs ≈ (1–5)×3.16e-4×1.56×0.0316 ≈ 1.6e-5 to 7.8e-5 cm/s. The paper's Fig. 5 and text quote measured uncertainties 0.01–0.03 cm/s, roughly three orders of magnitude larger. The quoted source density and normalization cannot produce those errors; conversely, producing 0.01 cm/s would require σ_n ~ 10^3 cm/s. Because the χ² likelihood (Eq. 6) and all contours in Figs. 6–11 are built from these σ_i, the forecast is internally inconsistent and not reproducible. The claimed precision may still be conceivable, but this manuscript does not establish it.","agreement_with_reader":"partial"},"referee_report":{"model":"deepseek-v4-flash","summary":"The paper argues that with the SKA's high spectral resolution modes (0.001 and 0.002 Hz), the Sandage-Loeb redshift drift can be measured from HI 21cm emission and absorption over a 0.5-year observing campaign, reaching sub-mm/s precision near z~1. The authors construct mock 'observed' drift velocities using an error formula (Eq. 9), fit them within ΛCDM, wCDM, and CPL models, and report tight constraints on H0, Ωm, w, w0, and wa. The central claim is that these SKA configurations can provide real-time cosmological measurements of cosmic acceleration.","tokens_in":18160,"tokens_out":8869,"duration_ms":74719,"significance":"If fully substantiated, the paper would be a valuable forecast for a highly challenging observable, extending earlier Sandage-Loeb analyses to the HI 21cm emission and absorption channels with SKA. The comparative treatment of emission galaxies versus DLA absorption systems and of three dark-energy parametrizations is a useful framework. The paper also makes a clear falsifiable prediction about the required spectral resolution (0.001-0.002 Hz) for a 0.5-year experiment. However, the central detectability claim rests on an unvalidated and internally inconsistent error model, and the parameter constraints are derived from mock data without an explicit forecast framing. As it stands, the numerical results do not support the claimed precision, and the paper needs substantial reanalysis before its conclusions can be accepted.","major_comments":[{"comment":"The stated inputs to Eq. (9) do not reproduce the plotted uncertainties. For the 0.001 Hz case, using N=10^7 per 0.1 bin as stated after Eq. (9), λ=1.09, Δν=0.001 Hz, and (1+z)=1.5 at z≈0.5, one obtains Δv_obs ≈ (1-5) × 1.6×10^-5 cm/s, i.e., roughly 1.6×10^-5 to 8×10^-5 cm/s. Even taking the larger counts N~10^8 shown in Fig. 4 gives values near 10^-5 cm/s. Fig. 5 and the text report observed uncertainties of 0.01-0.03 cm/s, about three orders of magnitude larger. Because the χ² statistic in Eq. (6) is built directly from these σ_i, the confidence contours in Figs. 6-11 are not reproducible from the stated equations and inputs, and the claimed constraints are not internally supported.","section":"Section 4, Eq. (9) and Fig. 5"},{"comment":"The normalization constant σ_n, which controls the overall scale of every error bar in the forecast, is cited solely to an unpublished work ('Kang 2024, in prep'). No derivation, fitting procedure, or independent calibration is provided, and the scaling (1+z)^λ Δν^{1/2} is asserted without a reference or derivation. Since the entire detectability claim—sub-mm/s precision in 0.5 years—depends on this constant, the paper does not currently offer a verifiable basis for its central result.","section":"Section 4, Eq. (9)"},{"comment":"The 'SKA data' points are mock data generated from the fiducial ΛCDM model using Eq. (9), but the text repeatedly refers to them as 'observed', 'SKA data', and 'empirically determined'. Fitting these mock points with the same models that generated them (ΛCDM, wCDM, CPL) guarantees recovery of the input parameters, so the reported confidence intervals are a property of the assumed noise model rather than an independent measurement. The paper must explicitly identify the analysis as a forecast, specify exactly how the mock data were generated, and discuss the role of the priors in Table 2 in shaping the contours.","section":"Section 4, Figs. 5-11"},{"comment":"The headline numerical claims are internally contradictory. The abstract quotes drift rates of 0.01-0.21 mm/s and 0.031-0.17 mm/s; Section 4 and Fig. 5 quote signals of 0.05-0.15 cm/s with uncertainties of 0.01-0.03 cm/s; and the conclusion states uncertainties of 2-5 mm/s. These are not equivalent under unit conversion. In addition, Table 3 reports Ω_m = 0.311^{+0.304}_{-2.214} for ΛCDM, whose lower 1σ bound is unphysical, and Table 4 contains similarly malformed intervals. The numerical results as presented are therefore not internally consistent.","section":"Abstract, Section 4, Tables 3-4"}],"minor_comments":[{"comment":"Eq. (5) as written, Δv = kh [1 + E(z)/(1+z)], is inconsistent with Eq. (2); substituting Eq. (2) into Δv = ˙v Δt gives kh [1+z - E(z)]/(1+z). If this is a typographical error, it should be corrected; if it is not, the sign/structure error changes the predicted signal.","section":"Eq. (5)"},{"comment":"There are several typos: 'Proir' should be 'Prior' in Table 2; 'Burerau' in the affiliations should be 'Bureau'; and 'constrainted' in Section 4 should be 'constrained'.","section":"Table 2 and affiliations"},{"comment":"The reference list contains duplicate entries for Alves et al. (2019), Kanekar et al. (2001), and Rawlings & Schilizzi (2011); these should be consolidated.","section":"References"},{"comment":"The notation S_z and S_v is used without precise definitions of units; Fig. 3's caption calls the velocity drift 'dimensionless' even though it is expressed in cm/s. Please clarify the definitions and units.","section":"Section 2, Figs. 2-3"},{"comment":"The sentence about peculiar motion states it 'will be attenuated to 10^-14' without specifying the units or the quantity; please clarify what is 10^-14 (e.g., a velocity, a fractional shift, a redshift).","section":"Section 3"},{"comment":"The y-axis label 'N' should specify that the counts are per 0.1 redshift bin, since the text alternates between 'N' and 'N per 0.1 redshift interval'.","section":"Fig. 4"}],"recommendation":"reject","confidential_remarks":"The central error model is not only unpublished but numerically inconsistent with the plotted uncertainties; even with a revised normalization, the paper would need to be repositioned as a forecast using mock data, with the noise model fully derived and validated. Given that Eq. (9) is load-bearing for every quantitative result, I do not see a path to acceptance within a normal revision cycle. The editor may wish to consider whether a fully reworked version—one that presents a reproducible noise model and clearly labels the mock-data forecast—could be suitable as a future submission."},"author_rebuttal":null,"desk_editor":{"model":"deepseek-v4-flash","letter":"Bottom line: the paper targets something worth asking—whether SKA HI 21cm observations could detect redshift drift in six months rather than ELT's decade—but the forecast that would make that news is built on an unverified normalization constant, and the numbers do not add up internally. I would desk reject, not because the idea is bad, but because the central claim cannot be checked as written.\n\nWhat is genuinely useful: it extends the earlier Kloeckner et al. 2015 and Alves et al. 2019 formalism to very high spectral resolution (0.001/0.002 Hz), includes both galaxy emission and DLA absorption, and produces mock-data contours for H0, Omega_m, w, w0, and wa across LambdaCDM, wCDM, and CPL. The DLA side is a reasonable complement to the emission forecasts. If the error model were solid, the 0.5-year timescale would be a striking result.\n\nThe problem is the error model. Equation (9) is the linchpin; its normalization sigma_n is cited only to 'Kang 2024, in prep', with no derivation or external check. The stress-test note is right: plugging the paper's own Section 4 values into Eq. (9) at z approx 0.5 (N=10^7, lambda=1.09, Delta_nu=0.001 Hz, sigma_n=1-5 cm/s) gives per-bin errors around 1.6e-5 to 7.8e-5 cm/s. Figure 5 and the text quote 0.01-0.03 cm/s. That is a factor of roughly a thousand, not a rounding error. The conclusion then quotes 2-5 mm/s (0.2-0.5 cm/s), a third set of numbers. Since Figures 6-11 are built from the chi-square in Eq. (6) using these sigma_i, the derived constraints are not reproducible.\n\nThere is also a framing issue: the 'observed' SKA data are mock points generated from a fiducial model and then fitted with the same model. That is acceptable as a Fisher-style exercise if clearly labeled, but the paper calls them 'observed'/'SKA data' and presents the recovered parameters as empirical constraints. The abstract's drift rates (0.01-0.21 mm/s and 0.031-0.17 mm/s) also do not match the 0.05-0.15 cm/s in Figure 5, and the relation to the quoted '1.28 mm/s theoretical accuracy limit' is muddled.\n\nWho is this for? People working on SKA HI surveys and Sandage-Loeb forecasts might use it as a pointer, but I would not trust any quantitative conclusion in the current version. It deserves a serious rewrite, not a serious referee as-is. If the authors can derive sigma_n, make the units and numbers consistent, and relabel the mock data, there may be a useful paper underneath.","headline":"The SKA 0.5-year redshift-drift forecast is a nice question, but the error model is unverifiable and the paper's own numbers disagree by orders of magnitude.","tokens_in":18633,"tokens_out":3941,"would_cite":false,"duration_ms":34623,"reading_group":"no","serious_thinker":"no","would_accept_peer_review":false},"rs_alignment":null,"lean_confirmation":null,"pith_extraction":{"msc":[],"pacs":[],"model":"deepseek-v4-flash","headline":"SKA 0.001 Hz mode can clock cosmic drift at sub-mm/s precision over half a year.","keywords":["redshift drift","Sandage-Loeb effect","HI 21cm line","Square Kilometre Array","dark energy constraints","cosmic acceleration","spectral resolution","cosmological parameter forecasting"],"falsifier":"Measure sigma_n directly from SKA1 or SKA2 commissioning spectra by comparing observed channel-to-channel velocity residuals against the $N^{{-1/2}}$(1+z)^$\\lambda$ $Delta_nu^{{1/2}}$ scaling in Eq (9) at 0.001 and 0.002 Hz; if the fitted sigma_n falls outside 1 to 5 cm/s, the claimed sub-mm/s precision and the H0, Omega_m, w, w0, and wa constraints do not hold. An independent check is to re-derive sigma_n from receiver temperature, integration time, and bandpass stability without invoking the in-prep normalization.","tokens_in":17595,"feed_emoji":"📡","tokens_out":8046,"duration_ms":66465,"temperature":0.7,"pith_summary":"Using a half-year observing campaign, this paper argues that the Square Kilometre Array (SKA) can resolve the Sandage-Loeb redshift drift—the tiny change in a source's redshift caused by cosmic expansion—through redshifted HI 21cm line emission and absorption. The argument turns on two ultra-high spectral resolutions, 0.001 Hz and 0.002 Hz, which the paper shows are the SKA configurations whose frequency resolution beats the ~1.28 mm/s velocity-drift accuracy limit across z=0 to z=1. With roughly a billion HI galaxies and about 1,800 Damped Lyman-alpha systems, the forecast reaches sub-mm/s per-half-year precision, enough to constrain H0 near 70 km/s/Mpc, Omega_m near 0.3, and the dark-energy equation of state near w=-1. If correct, this makes the SL effect a practical, model-independent probe of cosmic acceleration in the SKA era rather than a decades-distant prospect.","feed_headline":"SKA 0.001 Hz mode can clock cosmic drift at sub-mm/s","feed_subtitle":"A 0.5-year survey would pin H0 near 70 km/s/Mpc and dark energy near w=-1.","key_machinery":"The load-bearing object is the Sandage-Loeb velocity-drift relation, $\\frac{dv}{dt} = \\frac{c H_0}{1+z}\\\\left[1+z - \\frac{H(z)}{H_0}\\\\right]$, together with the error-forecasting formula $\\\\Delta v_{\\\\rm obs} = \\\\sigma_n N^{-1/2}(1+z)^{\\\\lambda} \\\\Delta\\\\nu^{1/2}$ in cm/s, where $\\\\sigma_n$ is a normalization constant set to 1 to 5 cm/s, $N$ is the number of detected 21cm sources per redshift bin, and $\\\\lambda$ is 1.09 for 0.001 Hz or 1.52 for 0.002 Hz. The forecast works by comparing this predicted precision against the theoretical drift signal: at 0.001 and 0.002 Hz the velocity drift is 0.01 to 0.21 mm/s, exceeding the 1.28 mm/s detection limit, so the measurement is signal-dominated rather than noise-dominated. The other machinery is the source census: fitted dN/dz coefficients for HI galaxy counts and a DLA incidence function, which convert the chosen spectral resolution into the number statistics that drive the error formula.","core_discovery":"The central claim is that the SKA, observing redshifted HI 21cm emission from face-on galaxies and 21cm absorption from Damped Lyman-$\\alpha$ systems over a 0.5-year interval, can detect the cosmological redshift drift at z around 1 with velocity uncertainties of 0.01 to 0.21 mm/s, provided the spectral resolution is set to 0.001 or 0.002 Hz. The paper derives the required frequency resolution from the Lambda-CDM (Planck 2018) prediction that the frequency shift stays below 0.1 Hz, then uses source-count statistics with roughly $10^{7}$ galaxies per redshift bin to forecast error bars. The resulting simulated data recover best-fit parameters H0 near 67 to 70 km/s/Mpc, Omega_m near 0.29 to 0.33, w near -0.95, w0 near -1.0, and wa near -0.1 across the Lambda-CDM, wCDM, and CPL models. The paper concludes that the SL effect observed through HI 21cm lines is a viable real-time cosmological probe of dark energy.","pith_inferences":["Beyond the paper's claims: if the normalization constant sigma_n in the error formula comes out higher than 5 cm/s in SKA commissioning data, the claimed sub-mm/s precision and all quoted parameter constraints degrade roughly linearly; a useful early test is to measure sigma_n with a single bright calibrator before full surveys begin.","Beyond the paper's claims: combining the emission and absorption channels in a joint likelihood could break some of the H0-Omega_m degeneracy visible in the paper's separate contours, since the two channels have different redshift weightings.","Beyond the paper's claims: the 0.001 and 0.002 Hz criterion suggests a natural target-selection strategy, prioritizing face-on galaxies with S/N at least 100 and single Gaussian profiles while using DLA systems mainly as cross-checks rather than primary drift anchors.","Beyond the paper's claims: extending the baseline to 1 or 5 years would not only shrink the 1/sqrt(N) error but also change the redshift dependence of the drift signal, offering a consistency check on dark-energy models that a single 0.5-year snapshot cannot provide."],"forward_implications":["At 0.001 Hz spectral resolution, the projected HI 21cm emission sample reaches about 1.36 billion galaxies by z=1, giving per-bin velocity uncertainties of 0.01 to 0.03 cm/s per half-year.","Only the 0.001 and 0.002 Hz configurations achieve the required precision across the full z=0 to 1 range; 0.005 Hz and 0.01 Hz do not, so the experiment's feasibility rests on SKA delivering those two spectral modes.","Emission-line forecasts constrain H0 to sub-1% precision with best fits near 70 km/s/Mpc and Omega_m near 0.3, and tighten w, w0, and wa relative to current SN Ia and BAO constraints.","DLA absorption-line data, though yielding about 1,800 systems and weaker constraints with H0 uncertainties above 2.8, still recover the fiducial parameters and independently support the same acceleration signal.","A 0.5-year observing window, not a decade-long campaign, suffices for the SL signal if the spectral resolution and source counts are as assumed."],"supporting_citations":[{"why":"This reference defines the SKA HI 21cm redshift drift experiment and supplies the precision formula and spectral-resolution requirements that the forecast is built on.","marker":"Kloeckner et al. 2015"},{"why":"This reference establishes the SL signal error scaling with 1/sqrt(N) and the observational requirements that motivate the error-forecasting approach.","marker":"Liske et al. 2008"},{"why":"This unpublished work provides the normalization constant sigma_n = 1-5 cm/s used in Eq (9), which is the single unvalidated input to the error forecast.","marker":"Kang 2024, in prep"},{"why":"This reference supplies the coefficient set c1 through c5 for the dN/dz function that determines the number of detectable HI galaxies per redshift bin.","marker":"Yahya et al. 2015"},{"why":"This reference supplies the DLA incidence function and 21cm absorption detection rates that set the absorption-line source counts.","marker":"Kanekar & Briggs 2004"},{"why":"This reference provides the fiducial Lambda-CDM cosmology used to compute theoretical drift curves and define the signal the SKA must beat.","marker":"Planck Collaboration et al. 2020"},{"why":"This reference gives the dimensionless drift definitions and the systematic-error treatment adopted for target selection.","marker":"Alves et al. 2019"}],"fun_headline_variants":["SKA 21cm survey measures cosmic acceleration to sub-mm/s in 6 months","Redshift drift via HI 21cm: SKA pins dark energy in half-year","SKA's 0.001 Hz mode detects redshift drift at 0.1 mm/s precision","Half-year SKA campaign to clock universal expansion to mm/s accuracy","SKA 21cm lines promise direct measurement of cosmic acceleration"],"cache_read_input_tokens":3200,"weakest_assumption_plain":"All the quoted precision depends on the normalization constant sigma_n in the error formula, which the paper sets to 1 to 5 cm/s citing only unpublished work; if the real SKA noise statistics differ, every velocity error and every recovered cosmological parameter changes with it.","fun_headline_variants_meta":{"raw":{"variants":["SKA 21cm survey measures cosmic acceleration to sub-mm/s in 6 months","Redshift drift via HI 21cm: SKA pins dark energy in half-year","SKA's 0.001 Hz mode detects redshift drift at 0.1 mm/s precision","Half-year SKA campaign to clock universal expansion to mm/s accuracy","SKA 21cm lines promise direct measurement of cosmic acceleration"]},"model":"deepseek-v4-flash","effort":"low","cost_usd":0.000846,"raw_usage":{"total_tokens":3775,"prompt_tokens":1134,"completion_tokens":2641,"prompt_tokens_details":{"cached_tokens":384},"prompt_cache_hit_tokens":384,"prompt_cache_miss_tokens":750,"completion_tokens_details":{"reasoning_tokens":2536}},"tokens_in":750,"tokens_out":2641,"duration_ms":16729,"temperature":1.0,"reasoning_tokens":2536,"cache_read_input_tokens":384,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-08-10T14:36:10.830313+00:00","model_set":{"reader":"deepseek-v4-flash"},"falsifier":"Measure sigma_n directly from SKA1 or SKA2 commissioning spectra by comparing observed channel-to-channel velocity residuals against the $N^{{-1/2}}$(1+z)^$\\lambda$ $Delta_nu^{{1/2}}$ scaling in Eq (9) at 0.001 and 0.002 Hz; if the fitted sigma_n falls outside 1 to 5 cm/s, the claimed sub-mm/s precision and the H0, Omega_m, w, w0, and wa constraints do not hold. An independent check is to re-derive sigma_n from receiver temperature, integration time, and bandpass stability without invoking the in-prep normalization.","supporting_citations":[{"cited_title":"R., Obreschkow, D., Martins, C., et al","cited_arxiv_id":null,"evidence_quote":"This reference defines the SKA HI 21cm redshift drift experiment and supplies the precision formula and spectral-resolution requirements that the forecast is built on."},{"cited_title":"2024, Research in Astronomy and Astrophysics, 24, 075002, doi: 10.1088/1674-4527/ad48d1","cited_arxiv_id":null,"evidence_quote":"This unpublished work provides the normalization constant sigma_n = 1-5 cm/s used in Eq (9), which is the single unvalidated input to the error forecast."},{"cited_title":"S., Leite, A","cited_arxiv_id":null,"evidence_quote":"This reference gives the dimensionless drift definitions and the systematic-error treatment adopted for target selection."}],"review_version":1}