{"id":"449ee80e-b072-42ab-a197-f3a3b3f41436","arxiv_id":"2608.13178","paper_version":1,"verdict":"CONDITIONAL","confidence":"MODERATE","novelty_score":6.0,"correctness_risk":"medium","formal_verification":"none","parameter_count":9,"one_line_summary":"Stacked DESI-LRG spectra yield a cosmographic H(z) curve from z=0.36 to 0.80 and two direct cosmic-chronometer measurements, including H(z≈0.61)=88.5^{+6.7}_{-12.6}±8.1 km/s/Mpc.","lead":"The authors measure how fast the universe was expanding between redshifts 0.3 and 1.2 using the ages of more than half a million old, passively evolving DESI galaxies. The result is a continuous H(z) curve with covariance and two direct expansion-rate measurements, useful as an independent check on the Hubble tension.","discovery_kind":"extension","skeptic_critique":{"model":"deepseek-v4-flash","headline":"Unquantified SPS-model systematic can shift the age-redshift slope; the 'comprehensive' systematic budget omits the one contribution the paper itself says is likely comparable.","rationale":"Reading in good faith: the paper's pipeline is coherent, the cosmographic derivation in Eqs. (3)-(8) is standard, and the final H(z) values are compatible with Planck-LambdaCDM. The reader's archaeological-cut concern is legitimate, but the paper does vary the t(sigma) and z(t) prescriptions in Section 4.4 and frames the cut as a window restriction; this is addressable. I find the SPS-model omission more load-bearing because every datum in the t-z plane is generated by the TMJ model, so any model-induced bias in age trends translates directly into H(z), and the paper itself concedes the magnitude. The abstract and Section 6 contradict each other on the comprehensiveness of the systematic budget. The proposed SPS re-fit is feasible and would settle the concern: if the model-induced shift is small, the central claim stands; if it is large, the quoted uncertainties are incomplete. I therefore maintain the reader's CONDITIONAL verdict, with the SPS-systematic test as the explicit condition. My agreement is partial because the reader's weakest_assumption differs, although their rationale also flags the SPS-model omission.","tokens_in":31855,"tokens_out":8197,"duration_ms":80600,"concrete_test":"Re-compute stacked-spectrum ages for the baseline stacks with an independent SPS model (e.g., EMILES, Vazdekis et al. 2015, or FSPS) using the same Lick index set, pyLick measurements, and emcee likelihood, then re-run the pivotal cosmographic fit (Section 4.3) and the two finite-difference fits (Section 5.2). Compare H(z0=0.57), H(z≈0.55), and H(z≈0.61) with the baseline values; if any shift exceeds the quoted internal systematic (11.3, 22.4, and 8.1 km/s/Mpc respectively), the systematic budget is incomplete and the paper must add an SPS contribution or narrow the claim. As a second cross-check, compare the finite-difference slope at z≈0.61 with the D4000-based H(z) of Loubser 2025 over the same redshift range.","verdict_should_be":"CONDITIONAL","load_bearing_attack":"The central claim requires that the quoted H(z) constraints are unbiased and that their uncertainties capture the dominant systematics. The paper's Section 6 explicitly admits that the choice of stellar population synthesis (SPS) model is a known source of systematic shift and that existing literature suggests this contribution is likely of the same order as the combined internal systematics, yet Section 4.4 restricts the systematic budget to data-management choices (S/N level, archaeological/cosmological prescriptions, quality flags) and does not vary the SPS model. This is internally inconsistent with the abstract's claim of a comprehensive analysis of all methodological choices. The concern is not merely a calibration offset: H(z) is extracted from the slope dt/dz, so a model whose Balmer-line responses, age-metallicity degeneracy, or high-metallicity behavior is inaccurate in a redshift-dependent way will bias the slope. The archaeological-coherence cut is itself defined from the TMJ model response, coupling sample selection to the SPS model. If the literature estimate is correct, the quoted systematic error bars (11.3 and 8.1 km/s/Mpc) are understated and the central values may shift outside them; if it is not, the paper should quantify and justify the omission.","agreement_with_reader":"partial"},"referee_report":{"model":"deepseek-v4-flash","summary":"This paper applies the cosmic chronometer method to DESI-DR1 luminous red galaxies. The authors stack ~527,000 spectroscopically selected passive galaxies in velocity-dispersion and redshift bins, measure Lick indices, fit Thomas et al. (2011) stellar population models to derive ages, and build t-z relations for five velocity-dispersion groups. They then fit a pivotal-redshift cosmographic expansion around z0=0.57 to obtain H(z0)=95.1^{+10.9}_{-6.0} (stat.) ± 11.3 (syst.) km/s/Mpc, and provide a 45-point correlated H(z) sampling with a full covariance matrix. They also derive two finite-difference H(z) measurements from disjoint sets of velocity-dispersion groups, the more precise being H(z≈0.61)=88.5^{+6.7}_{-12.6} (stat.) ± 8.1 (syst.) km/s/Mpc. The paper releases stacked spectra, spectral-feature measurements, MCMC chains, and the H(z) covariance products.","tokens_in":32213,"tokens_out":5916,"duration_ms":56916,"significance":"If the quoted constraints are unbiased, this is a valuable addition to the cosmic-chronometer literature: it extends the method to DESI-scale samples, provides a continuously sampled H(z) band over 0.36<z<0.8, and supplies the full covariance needed for cosmological use. The velocity-dispersion corrections, the stacking methodology, and the explicit treatment of data-management systematics are careful and reproducible from the released products. The recovery of a downsizing-like age scaling relation is a useful consistency check. However, the central value and error bars rest on selection and SPS-model assumptions whose quantitative impact is not fully demonstrated; the SPS-model systematic is explicitly acknowledged in Section 6 as likely comparable to the entire internal systematic budget but is not included in the quoted errors. The paper is therefore not yet at the level of support claimed in the abstract.","major_comments":[{"comment":"The systematic uncertainty budget is incomplete in a way that directly affects the central claim. Section 4.4 restricts the systematic budget to data-management choices (S/N level, archaeological/cosmological prescriptions, quality flags), while Section 6 states that the choice of SPS model is a known source of systematic shift and that 'existing literature suggests that this contribution is likely of the same order as our combined internal systematics.' Since the H(z) measurement is extracted from the slope of the t-z relation through Eqs. (10)-(11), a redshift-dependent SPS bias in the age estimates changes the slope and hence shifts H(z) by an amount potentially comparable to the quoted systematic errors. The abstract's phrase 'comprehensive analysis of all the methodological choices' is therefore not supported. The authors should either include a cross-SPS comparison (e.g., TMJ versus an independent model such as α-MC or Knowles et al.) or quantitatively propagate the literature estimate of SPS scatter into the systematic budget for both the cosmographic and finite-difference results.","section":"Section 4.4 and Section 6"},{"comment":"The exclusion of the 355<σ[km/s]<400 velocity-dispersion group is not documented as promised. The text says 'Further details on this are given in section C,' but Appendix C discusses quality flags and the z≈0.5 oscillation and contains no analysis of the 355-400 group. This is a data-driven selection applied before the cosmographic fit: one of six velocity-dispersion groups is removed because of a reported small age offset relative to the 320-355 group. The authors should show the t-z relation for this group, quantify the offset and its origin, and demonstrate that the cosmographic results are insensitive to including or excluding it. Without this, a reader cannot assess whether the exclusion is a benign redshift-window choice or a systematic that biases the fitted slope.","section":"Section 4, velocity-dispersion binning paragraph; Appendix C"},{"comment":"The 'archaeological coherence cut' couples the sample selection to the same stellar-population modeling used in the age fit. The redshift cap for each velocity-dispersion group is set by translating an age floor of 2.5 Gyr to redshift using the A25 t(σ) scaling relation and a Planck-ΛCDM z(t) prescription. The paper argues that this only limits the redshift window and does not bias the fit, and Section 4.4 does vary t(σ) and z(t) prescriptions in the systematic budget. However, all adopted t(σ) relations are empirical fits to similar data, and the cut removes exactly the low-mass, high-redshift stacks where the TMJ model response is poorest. If the scaling relation is biased, the selection is correlated with the age-redshift slope. I ask for an additional robustness test: for example, fit the t-z relations with and without the excluded low-mass high-z stacks using an independent age indicator, or show explicitly how much the fitted H(z) changes when the age floor is varied from 2.0 to 3.0 Gyr in addition to the listed t(σ) prescriptions.","section":"Section 3, Fig. 1; Section 4.4"}],"minor_comments":[{"comment":"The phrasing 'we intend apply the cosmic chronometer method' should be 'we intend to apply'.","section":"Abstract"},{"comment":"The caption says 'against redshift on the left' in both clauses; the second panel is presumably on the right.","section":"Figure C.2 caption"},{"comment":"The legend for Jimenez et al. (2023) is labeled 'Phot.' in Fig. 5 and 'Lick' in Fig. 7; the label should be consistent.","section":"Figures 5 and 7"},{"comment":"The number of alternative configurations entering the standard deviation for the systematic budget is not stated; listing the count of S/N levels, prescription combinations, and quality-flag combinations would improve transparency.","section":"Section 4.4"},{"comment":"The sentence describing the covariance contribution 'C_syst,ij = σ_syst(zi) σ_syst(zj)' assumes perfect correlation across all redshift bins; a brief justification of this fully correlated model would help users of the released covariance.","section":"Section 5.1"}],"recommendation":"major_revision","confidential_remarks":"The paper is a good fit for A&A and the released products are valuable. The main risk is that the omitted SPS-model systematic is acknowledged in the conclusions as likely comparable to the entire quoted systematic budget; this should be resolved before acceptance, either by adding the contribution or by a quantified cross-model test. The promised but missing documentation of the 355-400 exclusion should also be supplied. I would not reject the paper because the core derivation and data products are sound and the issues are addressable within the manuscript's scope."},"author_rebuttal":null,"desk_editor":{"model":"deepseek-v4-flash","letter":"The paper delivers something real: the first Lick-index cosmic-chronometer analysis of DESI-DR1 LRGs, turning roughly 500k spectra into stacked t-z relations, a 45-point H(z) curve with full covariance over 0.36<z<0.80, and two independent finite-difference H(z) values. The pivotal-redshift cosmography in Section 2 is clean and the adaptation from Fazzari et al. is sensible; the math checks out. The treatment of stacking S/N, velocity dispersion corrections, and the spread induced by the archaeological/cosmological redshift-window prescriptions and quality flags is transparent. The central values—H(z0=0.57)=95.1^{+10.9}_{-6.0}±11.3 and H(z≈0.61)=88.5^{+6.7}_{-12.6}±8.1—are consistent with ΛCDM and look plausible. This is a genuinely useful data product for a redshift range where CC data are sparse.\n\nThe main problem is the one the authors themselves state in Section 6: the SPS model choice is 'likely of the same order as our combined internal systematics,' but Section 4.4's budget covers only data-management choices—S/N level, redshift-window prescriptions, quality flags. So the abstract's 'comprehensive analysis of all methodological choices' overstates the case. This matters because H(z) is extracted from the slope of age versus redshift, not from an absolute calibration; a model whose Balmer-line responses or age-metallicity behaviour is wrong in a redshift-dependent way will tilt that slope. The archaeological coherence cut also derives from the TMJ model response, so sample selection and model are coupled. The stress-test note lands on this, and the reader's conditional verdict is right. The fix is straightforward: run the stacks through at least one independent SPS model, or remove 'comprehensive' and state plainly that SPS uncertainty is not included. Without that, the quoted systematic errors are not yet believable.\n\nMinor issues: the 355-400 km/s group is excluded with 'further details in Section C,' but Section C never provides them. And the data products (H(z) array, covariance, chains, stacks) are promised only 'in the journal version'; for a paper whose main selling point is a usable covariance matrix, that should be released at submission or, failing that, on acceptance.\n\nThe circularity burden is low and I agree with the reader that the Planck-based redshift caps only set the window, with a modest systematic spread shown when varying the prescriptions. The central cosmographic machinery holds up.\n\nWho should read it: anyone doing model-independent H(z) tests, dark-energy reconstructions, or wanting a correlated CC dataset in 0.5<z<0.8. It deserves serious peer review. I would send it out, with the SPS issue and the missing appendix explanation flagged for revision.","headline":"A genuinely new DESI cosmic-chronometer H(z) dataset with a clean cosmographic core, but the 'comprehensive' systematic budget omits the SPS-model uncertainty the authors themselves flag as comparable.","tokens_in":32725,"tokens_out":5736,"would_cite":true,"duration_ms":47729,"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":"By dating 527,000 stacked DESI galaxies, this paper measures the expansion rate from z=0.36 to z=0.80 without assuming a cosmological model.","keywords":["cosmic chronometers","Hubble parameter","DESI DR1","luminous red galaxies","Lick indices","stellar population ages","cosmography","H(z) covariance matrix"],"falsifier":"Repeat the pipeline with the archaeological redshift cap removed or shifted, for example from 2.5 Gyr to 4 Gyr, and with a different $z(t)$ conversion; if the resulting $H(z_0=0.57)$ moves by more than the quoted $\\pm 11.3$ km s$^{-1}$ Mpc$^{-1}$ systematic, the cut is biasing the slope rather than merely limiting the redshift window.","tokens_in":31688,"feed_emoji":"🌌","tokens_out":10330,"duration_ms":85003,"temperature":0.7,"pith_summary":"DESI's nearly three million luminous red galaxies can be turned into cosmic chronometers: passively evolving galaxies whose measured ages trace how long the Universe took to expand between redshifts. This paper stacks more than half a million spectroscopically clean DESI spectra in velocity-dispersion and redshift bins, extracts stellar ages from Lick absorption indices, and reads the expansion rate off the slope of the age-redshift relation. The central result is a model-independent Hubble-parameter estimate at redshift 0.57, $H=95.1^{+10.9}_{-6.0}$ (stat.) $\\pm 11.3$ (syst.) km s$^{-1}$ Mpc$^{-1}$, plus a correlated $H(z)$ curve from $z=0.36$ to $0.80$ with its covariance matrix. A classical finite-difference estimate from the reddest, most massive galaxies gives $H(z\\approx0.61)=88.5^{+6.7}_{-12.6}$ (stat.) $\\pm 8.1$ (syst.) km s$^{-1}$ Mpc$^{-1}$. If these hold, cosmologists gain a new independent check on cosmic expansion in a redshift range where chronometer data are sparse.","feed_headline":"DESI galaxy ages pin Hubble rate to 95 km/s/Mpc at z=0.57","feed_subtitle":"A model-independent cosmic-chronometer measurement from 527,000 stacked LRG spectra, with covariance for 45 redshift bins.","key_machinery":"The carrying object is the stacked age-redshift relation. DESI LRG spectra are divided into seven velocity-dispersion strata, stacked in adaptive redshift bins to a signal-to-noise near 150, and assigned ages by fitting a set of Lick indices with the TMJ stellar population synthesis model, which predicts index strengths for given age, metallicity, and $\\alpha$-enhancement. The slope of age versus redshift is then converted into $H(z)$ either by a pivotal-redshift cosmography, a Taylor expansion of $(1+z)H(z)$ around the median sample redshift $z_0=0.57$ truncated at third order, or by the classical finite-difference formula. The simplification that carries the argument is that each stack behaves as a simple stellar population, so age differences between stacks are cosmic time differences.","core_discovery":"Stacking DESI-DR1 luminous red galaxy spectra in velocity-dispersion strata and adaptive redshift bins produces age-redshift relations whose slope carries a model-independent measurement of the expansion rate. Fitting these relations with a third-order Taylor expansion of $H(z)$ about a pivotal redshift $z_0=0.57$ gives $H(z_0=0.57)=95.1^{+10.9}_{-6.0}$ (stat.) $\\pm 11.3$ (syst.) km s$^{-1}$ Mpc$^{-1}$, together with a sampled $H(z)$ from $z=0.36$ to $0.80$ and a total covariance matrix. A classical finite-difference estimate $H(z)\\approx -\\Delta z/[\\Delta t(1+z)]$ applied to the reddest, most massive galaxies gives $H(z\\approx0.61)=88.5^{+6.7}_{-12.6}$ (stat.) $\\pm 8.1$ (syst.) km s$^{-1}$ Mpc$^{-1}$. The two routes are independent, and the paper provides the full chain and covariance so the correlated nature of the cosmographic reconstruction can be used directly in model tests.","pith_inferences":["If the archaeological age floor is even mildly wrong for low-$\\sigma$ galaxies, the excluded high-redshift points may carry information about $H(z)$; re-running the fit without that cut would settle this.","Users should not treat adjacent bins of the released $H(z)$ sampling as independent; the covariance matrix encodes the fact that all bins share the global kinematic parameters, so pointwise combinations without the matrix would understate errors.","Applying the same stacking and cosmographic pipeline to future DESI releases could push the method beyond the current $z\\approx0.8$ ceiling set by the rest-frame iron-index wavelength window."],"forward_implications":["A full covariance-matrix likelihood can now be used to test any model of $H(z)$ between $z=0.36$ and $z=0.80$ without the model being assumed to build the data.","The $z\\approx0.61$ measurement, coming from the reddest envelope of the most massive galaxies, is the purest and most stringent of the two local estimates.","The two finite-difference values are independent and can be combined with existing cosmic-chronometer compilations because they come from physically different galaxy populations.","Pivotal cosmography anchored at the median redshift of a dataset is a transferable way to turn future chronometer datasets into continuous $H(z)$ reconstructions."],"supporting_citations":[{"why":"Defines the cosmic chronometer idea: $H(z)$ from the differential age of passively evolving galaxies.","marker":"Jimenez & Loeb 2002"},{"why":"Establishes the stacking method and Lick-index age fitting this work extends, and supplies the $t(\\sigma)$ scaling relation used for the redshift cap.","marker":"A25"},{"why":"Supplies the pyLick code that measures Lick indices and the stellar-population fitting approach.","marker":"Borghi et al. 2022"},{"why":"The TMJ model that translates measured indices into ages, metallicities, and $\\alpha$-enhancement.","marker":"Thomas et al. 2011"},{"why":"Provides the pivotal-redshift cosmographic expansion used for the $H(z)$ fit.","marker":"Fazzari et al. 2025"},{"why":"Gives the $z(t)$ prescription for the redshift cap and the $\\Lambda$CDM reference curve.","marker":"Planck Collaboration et al. 2020"},{"why":"Defines the DESI LRG target selection that builds the parent sample.","marker":"Zhou et al. 2023"},{"why":"Earlier DESI chronometer analysis using D4000; supplies the comparison for this independent Lick-index route.","marker":"Loubser 2025"}],"fun_headline_variants":["DESI galaxy ages measure H(z) at z=0.57 and 0.61","Model-independent H(z) from DESI LRG stellar ages: 95 and 88.5","Two independent H(z) from DESI galaxy ages: 95 and 88.5","DESI cosmic chronometers: H(z) at z≈0.6 from two routes"],"cache_read_input_tokens":3200,"weakest_assumption_plain":"The load-bearing premise is that the redshift cap imposed on each velocity-dispersion group, built from an assumed 2.5 Gyr minimum reliable age and an age-mass scaling relation translated to redshift with a Planck-$\\Lambda$CDM cosmic-age curve, removes only unreliable young galaxies and does not tilt the age-redshift slope from which $H(z)$ is read.","fun_headline_variants_meta":{"raw":{"variants":["DESI galaxy ages measure H(z) at z=0.57 and 0.61","Model-independent H(z) from DESI LRG stellar ages: 95 and 88.5","Two independent H(z) from DESI galaxy ages: 95 and 88.5","DESI cosmic chronometers: H(z) at z≈0.6 from two routes"]},"model":"deepseek-v4-flash","effort":"low","cost_usd":0.000791,"raw_usage":{"total_tokens":3626,"prompt_tokens":1223,"completion_tokens":2403,"prompt_tokens_details":{"cached_tokens":384},"prompt_cache_hit_tokens":384,"prompt_cache_miss_tokens":839,"completion_tokens_details":{"reasoning_tokens":2305}},"tokens_in":839,"tokens_out":2403,"duration_ms":15287,"temperature":1.0,"reasoning_tokens":2305,"cache_read_input_tokens":384,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-08-15T15:00:00.527416+00:00","model_set":{"reader":"deepseek-v4-flash"},"falsifier":"Repeat the pipeline with the archaeological redshift cap removed or shifted, for example from 2.5 Gyr to 4 Gyr, and with a different $z(t)$ conversion; if the resulting $H(z_0=0.57)$ moves by more than the quoted $\\pm 11.3$ km s$^{-1}$ Mpc$^{-1}$ systematic, the cut is biasing the slope rather than merely limiting the redshift window.","supporting_citations":[],"review_version":1}