{"id":"decf6573-d6ce-432d-806d-cff01639ff54","arxiv_id":"2608.07453","paper_version":1,"verdict":"CONDITIONAL","confidence":"MODERATE","novelty_score":6.0,"correctness_risk":"medium","formal_verification":"none","parameter_count":4,"one_line_summary":"CMB-S4 forecasts: the two-site design reaches sigma(Neff)=0.029 in seven years, the Chile-only revised design reaches 0.030 in nine, and a cosmic-variance-limited survey would reach 0.0073.","lead":"This paper forecasts how precisely future CMB surveys could measure Neff, the effective number of relativistic particles in the early universe, using a public pipeline called DRAFT. The original two-site CMB-S4 design would reach the target of sigma(Neff)=0.029, the redesigned Chile-only plan would reach it in about nine years, and an idealized survey would reach 0.0073.","discovery_kind":"extension","skeptic_critique":{"model":"deepseek-v4-flash","headline":"The 0.029 forecast assumes exact foreground-shape knowledge; a modest residual or bias increase would push the design past the 0.030 target, so the headline claim is conditional on the unresolved foreground-model validation.","rationale":"The reader's weakest-assumption identification matches the most load-bearing concern: the forecast's reliance on a foreground model and MV-ILC whose exact-shape assumption is acknowledged as optimistic. The paper has real independent support—the DRAFT tool is public, the analysis uses standard and validated codes (CLASS, FisherLens, PySM), and the pipeline is end-to-end reproducible. The concern is not that the machinery is wrong, but that the headline σ(Neff)=0.029 is a statistical error bar computed under an idealized assumption that the foreground shapes are exactly known; the paper itself defers the bias and robustness quantification to a companion paper [72] that is not yet available. Since 0.029 is only 0.001 below the target, even a small increase in the effective residual noise or a small unmodeled bias would invalidate the central claim. This justifies the conditional verdict: the forecast should be treated as design guidance pending the foreground-model validation in [72], not as a guaranteed experimental outcome. The reader's verdict already captures this, so no change is needed.","tokens_in":49829,"tokens_out":7962,"duration_ms":75922,"concrete_test":"Run the public DRAFT tool for the two-site conceptual design with galactic foregrounds generated from PySM 3 spatially varying SED models (e.g., d3/s3) instead of the uniform-SED d1/s1 baseline, keeping all instrumental inputs identical. Recompute the ILC residuals and the Fisher matrix; if σ(Neff) exceeds 0.030, or if the extended-Fisher bias in Neff estimated by FisherLens exceeds about 0.015 (0.5σ), the headline claim that the conceptual design meets the CMB-S4 target is not robust to the foreground-modeling assumption.","verdict_should_be":"UNCHANGED","load_bearing_attack":"The central forecast (Table 6: σ(Neff)=0.029 for the two-site conceptual design, meeting the CMB-S4 target of 0.030) is computed by inserting post-ILC residual spectra C_res^XY into the Fisher covariance (3.5). These residuals are derived from the same foreground model that the analysis assumes to be exact. Section 3.1 states the forecasts 'assume the adopted model to be accurate and do not account for residual uncertainties in the foreground modeling itself,' and Section 3.4 concedes the MV-ILC is 'somewhat optimistic because it assumes that the shapes of the foreground signals are exactly known, which can lead to biased estimates of cosmological parameters.' Because the headline number sits only 3% below the target (0.029 vs 0.030), any degradation of the residual spectra—or, more seriously, an unmodeled bias in the component-separated CMB from spatially varying dust and synchrotron SEDs—would move the design past the target. The paper argues the impact is limited because the Neff constraint is driven by TE and EE, where foregrounds are subdominant, but it does not quantify the bias from mis-specified SED shapes in polarization. The quantitative assessment of these foreground-induced biases is deferred to the unpublished companion paper [72], so the headline claim is not yet independently verified against the very assumption the paper flags as optimistic.","agreement_with_reader":"agree"},"referee_report":{"model":"deepseek-v4-flash","summary":"This paper presents Fisher-matrix forecasts of the sensitivity to the effective number of relativistic species, N_eff, for several CMB-S4 survey configurations considered during the project's design phase. The forecasting pipeline, implemented in the publicly released DRAFT tool, combines map-based Galactic foreground simulations and template-based extragalactic foregrounds, minimum-variance internal linear combination component separation, iterative delensing with CLASS delens, and a Gaussian Fisher covariance. The main quantitative results are that the two-site conceptual design reaches sigma(N_eff)=0.029 over seven years, thereby meeting the original CMB-S4 target of 0.030, the Chile-only revised configuration reaches 0.030 after nine years of combined CMB-S4 and SO-like operations, a South-Pole-only alternative achieves 0.037, and a cosmic-variance-limited survey over the same multipole range gives sigma(N_eff)=0.0073. The paper also discusses the implications of these sensitivities for thermal relics, radiation-sector properties, BBN, and other early-universe physics.","tokens_in":50207,"tokens_out":6265,"duration_ms":59977,"significance":"The forecasts are directly relevant to the planning of current and future ground-based CMB experiments and to the interpretation of their N_eff science goals. The paper has several concrete strengths: the DRAFT tool is publicly released; the pipeline uses established codes (CLASS delens, FisherLens) and validates the Gaussian Fisher assumption with DALI for the fiducial cosmology; intermediate products such as post-ILC residual spectra are made available; and the physics discussion in Section 5 is broad and well referenced. The central threshold claim, however, rests on optimistic or idealized assumptions that are acknowledged in the text but not quantitatively bounded, and the margin by which the conceptual design meets the science target is only a few percent. If the headline claim is to be used for design decisions, the robustness of sigma(N_eff)=0.029 to foreground-model uncertainty and to SO goal-versus-baseline noise should be demonstrated or the claim should be explicitly conditioned on those assumptions.","major_comments":[{"comment":"The headline claim that the two-site conceptual design meets the CMB-S4 target with sigma(N_eff)=0.029 is computed from post-ILC residual spectra obtained under the assumption that the adopted foreground model is exactly correct. Section 3.1 states this explicitly, and Section 3.4 concedes that the MV-ILC is 'somewhat optimistic because it assumes that the shapes of the foreground signals are exactly known, which can lead to biased estimates of cosmological parameters.' Because 0.029 is only about 3% below the 0.030 target, even a modest increase in residual foreground power or a small SED mis-specification bias could push the forecast above the threshold. The argument that N_eff constraints are driven by TE and EE, where foregrounds are subdominant, addresses residual variance but does not quantify bias from mis-specified SED shapes in polarization. The paper does not provide this robustness quantification, despite noting that FisherLens includes an extended-Fisher bias estimator, and defers the analysis to the unpublished companion paper [72]. The manuscript should include a quantitative sensitivity test (for example, rescaling foreground amplitudes, varying dust and synchrotron SED parameters, or using a constrained ILC) or should explicitly frame the headline as conditional on the foreground model and remove the unqualified 'meets the target' language.","section":"§3.1, §3.4, Table 6"},{"comment":"The forecast that the Chile-only revised configuration reaches sigma(N_eff)=0.030 after nine years assumes that the SO-like large-aperture telescope reaches its 'goal' noise depth rather than its more conservative 'baseline' target, as stated in Section 2.4. No forecast using the baseline depth is presented, and no sensitivity of sigma(N_eff) to this assumption is given. Since the revised configuration reaches the threshold only after a long combined observing period, and the SO-like LAT contributes a substantial fraction of the sky area and depth, the goal-versus-baseline choice is material to the revised-configuration claim. The authors should present the baseline-depth result or quantitatively demonstrate that the difference is negligible.","section":"§2.4, Table 6"}],"minor_comments":[{"comment":"The abstract and several discussion sections quote sigma(N_eff)=0.030 as the experiment's sensitivity without recalling that this holds when the primordial helium abundance Y_p is fixed by its BBN-consistent relation to omega_b and N_eff; Table 7 shows that marginalizing over Y_p degrades the conceptual-design constraint to 0.072. This model assumption should be restated wherever the headline sensitivity is quoted.","section":"Abstract, §5, Table 7"},{"comment":"The figure caption states that the negative-time range [-4,0] years corresponds to calendar years 2029-2033 for a nominal 2033 CMB-S4 start, while the text of Section 2.4 says the SO-like LAT begins observing in 2028. This inconsistency should be reconciled.","section":"Figure 6 caption and §2.4"},{"comment":"The statement that reaching sigma(N_eff)=0.030 from the South Pole alone would require 'roughly nine CMB-S4 Ultra-deep-like LATs and observe f_sky=0.25 over a period of 20 years' is not derived in the text; either show the scaling argument behind this estimate or cite the analysis in which it was obtained.","section":"§4.3"}],"recommendation":"major_revision","confidential_remarks":"The paper is a collaboration forecast paper whose main threshold claim is conditional on two optimistic assumptions that are acknowledged but not quantified: exact foreground-shape knowledge in the MV-ILC, and SO goal-depth noise for the revised configuration. Both assumptions are load-bearing because the headline margins are small (0.029 vs. 0.030, and a nine-year time to threshold for the revised configuration). The companion paper [72] is unpublished, so the present manuscript cannot rely on it for the required robustness demonstration. The public DRAFT tool and the use of established, validated codes are strong assets that make the requested additional numerical tests straightforward to carry out."},"author_rebuttal":null,"desk_editor":{"model":"deepseek-v4-flash","letter":"The useful new content here is concrete: which CMB-S4-style configuration reaches sigma(Neff)=0.030, how long it takes, and how far a cosmic-variance-limited survey sits below that. The Chile-only revised configuration reaching 0.030 after nine years and the CVL floor of 0.0073 are genuinely new numbers. The method is standard Fisher plus ILC plus delensing, but the implementation is careful: it checks the Gaussian likelihood assumption with DALI, uses established codes, and ships the DRAFT pipeline and precomputed residual spectra as public products. That is real, reproducible work and should count as such.\n\nThe central forecast is plausible but not robust to its own stated assumptions. The paper says in Section 3.1 that the foreground model is taken to be accurate, and in Section 3.4 that the minimum-variance ILC is somewhat optimistic because foreground shapes are assumed known. Since 0.029 sits only about 3 percent below the 0.030 target, a modest degradation in foreground residuals or an unmodeled bias from spatially varying dust SEDs could push the design past target. The stress-test note is right about this. It is not fatal: Neff constraints are driven mostly by TE and EE, where polarized foregrounds are subdominant, and the paper defends that with reference to prior ILC results. But the quantitative bias assessment is deferred to the unpublished companion paper [72], so the headline number should be read as the statistical sensitivity under exact foreground-shape knowledge, not as a guaranteed outcome.\n\nMinor soft spots: the South-Pole-only noise model is a simple rescaling and the authors admit it; dropping patches 3 and 4 of the hybrid survey is reasonable but deserves a robustness check; and the Yp-degeneracy material is useful but secondary. The citation pattern is appropriate, heavy on CMB-S4 documents and SPT/Planck foreground papers, with nothing inflated. I also checked for circularity: no parameter is fitted to the 0.030 target; the forecast follows from the pipeline.\n\nThis paper is for CMB experiment planners, forecasters, and neutrino phenomenologists. It deserves a serious referee. I would send it to peer review, ask the referee to scrutinize the Chile-only timeline and the foreground-residual treatment, and suggest the authors soften the abstract until the companion systematic-bias study is available. As it stands, the paper is an honest and useful design reference, not a guaranteed prediction.","headline":"A careful, well-documented CMB-S4 Neff forecast with a public pipeline; the headline 0.029 meets the target only narrowly, so the foreground-model caveats matter more than usual, but this deserves a serious referee.","tokens_in":772,"tokens_out":794,"would_cite":true,"duration_ms":27400,"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":"The paper forecasts that a CMB-S4-scale survey can measure the effective number of relativistic species with $\\sigma(N_\\mathrm{eff}) = 0.029$, meeting the original science target of $0.030$, and that a cosmic-variance-limited successor…","keywords":["cosmic microwave background","effective number of relativistic species Neff","light thermal relics","neutrino cosmology","Fisher-matrix forecasts","internal linear combination","delensing","CMB survey design"],"falsifier":"Rerun the DRAFT pipeline for the two-site conceptual design with spatially varying dust and synchrotron spectral indices, or with the measured tSZ-CIB correlation included, and check whether $\\sigma(N_\\mathrm{eff})$ rises above 0.030; likewise, the authors' own companion systematic-bias study, flagged as in preparation, settles the question if it shows foreground-induced biases at or above $\\sigma(N_\\mathrm{eff}) \\approx 0.03$.","tokens_in":49655,"feed_emoji":"🌌","tokens_out":9552,"duration_ms":73158,"temperature":0.7,"pith_summary":"The paper forecasts how precisely next-generation ground-based cosmic-microwave-background surveys could measure $N_\\mathrm{eff}$, the effective number of relativistic species, which encodes the radiation content of the early universe beyond photons. Its central result is that the CMB-S4 two-site conceptual design, a wide Chilean survey combined with an ultra-deep South Pole patch, reaches $\\sigma(N_\\mathrm{eff}) = 0.029$ over seven years, meeting the project's original science target of $\\sigma(N_\\mathrm{eff}) = 0.030$, while the redesigned Chile-only configuration (a CMB-S4 hybrid LAT plus a Simons-Observatory-like telescope) reaches the same precision after nine years. A cosmic-variance-limited survey over the same multipole range would reach $\\sigma(N_\\mathrm{eff}) \\approx 0.0073$, a factor of three to four better, showing how much primary-CMB information remains untapped. If these numbers hold, a CMB-S4-scale survey would for the first time be sensitive to any thermal relic of spin $\\geq 1/2$ that ever equilibrated with the Standard Model, down to $\\Delta N_\\mathrm{eff} \\approx 0.047$ for a Weyl fermion, and would test the physics of neutrino decoupling at the $1.5\\sigma$ level.","feed_headline":"σ(Neff)=0.029 forecast meets CMB-S4's neutrino target","feed_subtitle":"Two-site design hits σ(Neff)=0.029; the Chile-only plan reaches 0.030 in nine years, probing spin-1/2 relics.","key_machinery":"The load-bearing object is the DRAFT (Dark Radiation Anisotropy Flowdown Team) pipeline, an end-to-end forecasting tool that turns an experimental specification into a projected parameter error bar. It assembles a multi-frequency covariance from map-based PySM 3 galactic dust and synchrotron simulations plus SPT-calibrated templates for the thermal and kinematic Sunyaev-Zel'dovich effects, cosmic infrared background, and radio galaxies, applies the Planck GAL090 galactic mask, extracts the CMB with a minimum-variance internal linear combination (ILC), computes iteratively delensed TT/TE/EE spectra with CLASS delens, and feeds them into a Fisher-matrix calculation via FisherLens. The post-ILC residual spectra act as the effective noise entering the Gaussian, diagonal-in-multipole covariance, and the final $\\sigma(N_\\mathrm{eff})$ is read from the inverse Fisher matrix. The ILC step is where the paper's central optimism lives: minimum-variance ILC assumes the foreground spectral shapes are exactly known.","core_discovery":"The paper's claim is a sensitivity projection: with the DRAFT forecasting pipeline, the CMB-S4 two-site conceptual design achieves $\\sigma(N_\\mathrm{eff}) = 0.029$ in its nominal seven years, meeting the original CMB-S4 science target of $\\sigma(N_\\mathrm{eff}) = 0.030$, and the Chile-only revised configuration, jointly analyzing the CMB-S4 hybrid LAT with an SO-like wide survey over $f_\\mathrm{sky} \\approx 0.62$, reaches $\\sigma(N_\\mathrm{eff}) = 0.030$ after roughly nine years. The South-Pole-only alternative, with 20% of the sky after masking, falls about 22% short of the target at comparable cost, demonstrating that sky coverage rather than depth drives $N_\\mathrm{eff}$ sensitivity because the information is spread across many acoustic peaks in the damping tail. The authors also show that polarization carries most of the small-scale constraining power, that the constraint saturates by $\\ell_\\mathrm{max} \\approx 3750$ for the realistic survey, and that marginalizing over the primordial helium abundance $Y_p$ degrades $\\sigma(N_\\mathrm{eff})$ by a factor of 2.5, a degeneracy partially broken by the free-streaming neutrino phase shift.","pith_inferences":["My reading: the headline numbers are statistical floors, not end-to-end systematics budgets; the paper's own caveats imply the real sensitivity will be set by how well foreground templates are validated, so an early empirical check is to compare DRAFT's post-ILC residuals against actual SO and SPT-3G small-scale polarization data.","The sky-coverage-over-depth result suggests that pushing below $\\sigma(N_\\mathrm{eff}) \\approx 0.02$ naturally points to full-sky space surveys or northern-hemisphere wide patches, a direction the authors note but do not quantify, and the released machinery makes this a direct rescaling exercise.","The $\\Delta N_\\mathrm{eff} = 0.027$ scalar threshold is only about $3\\sigma$ even at the cosmic-variance floor, implying that conclusively ruling out the lightest thermal scalar will require combining primary-CMB constraints with large-scale-structure or BBN information rather than CMB anisotropies alone.","If the $N_\\mathrm{eff} \\approx 2.8$ central value favored by current ACT+SPT+Planck data persists, the same forecasts imply a deviation well past $5\\sigma$ at CMB-S4 precision, turning what is framed as a design-forecast paper into a discovery template for non-standard radiation content."],"forward_implications":["At $\\sigma(N_\\mathrm{eff}) = 0.030$, the survey would detect or exclude a Weyl-fermion thermal relic at $1.6\\sigma$ and a massless vector boson at $1.8\\sigma$; a cosmic-variance-limited survey near $\\sigma(N_\\mathrm{eff}) \\approx 0.01$ would reach almost $3\\sigma$ for the lightest scalar, about $5\\sigma$ for a Weyl fermion, and more than $5\\sigma$ for a vector boson.","Because the $N_\\mathrm{eff}$ information saturates by $\\ell_\\mathrm{max} \\approx 3750$ for the realistic survey and lives mostly in E-mode polarization, the headline precision is robust to cutting small-scale temperature data, making the forecast resilient against many temperature systematics.","Marginalizing over the primordial helium abundance raises $\\sigma(N_\\mathrm{eff})$ from 0.029 to 0.072 for the conceptual design, so claims about beyond-Standard-Model physics at the 0.03 level hold only under the standard BBN consistency relation between $Y_p$, $\\omega_b$, and $N_\\mathrm{eff}$.","The gap between the CMB-S4-level forecasts and the cosmic-variance floor of $\\sigma(N_\\mathrm{eff}) = 0.0073$ quantifies the primary-CMB information still accessible to future full-sky or low-foreground missions.","The same observations would tighten constraints on the neutrino free-streaming phase shift, raising the lower bound on the free-streaming fraction of the radiation sector from about 82% today to above 96% at CMB-S4 precision."],"supporting_citations":[{"why":"Sets the $\\sigma(N_\\mathrm{eff}) = 0.030$ science target that the forecasted sensitivities are judged against.","marker":"[69]"},{"why":"Supplies the telescope counts, noise levels, and scan-strategy footprints of the two-site conceptual design.","marker":"[70]"},{"why":"Defines the Chile-only revised configuration, including the hybrid LAT survey and joint analysis with an SO-like telescope.","marker":"[79]"},{"why":"Provides the goal-depth noise levels of the SO-like LAT included in the revised-configuration forecast.","marker":"[67]"},{"why":"Provides the internal-linear-combination implementation (minimum-variance, constrained, partial, and cross variants) used for component separation.","marker":"[102]"},{"why":"Supplies CLASS delens for iteratively delensed CMB spectra and the FisherLens Fisher-matrix computation.","marker":"[105]"},{"why":"Provides Planck noise parameters, large-scale temperature information, the optical-depth prior, and the current $N_\\mathrm{eff}$ constraint baseline.","marker":"[58]"},{"why":"Calibrates the extragalactic foreground templates (tSZ, kSZ, CIB, radio) that enter the multi-frequency covariance.","marker":"[87]"},{"why":"Supplies the spin-dependent thermal-relic $\\Delta N_\\mathrm{eff}$ thresholds (0.027, 0.047, 0.054) that set the physics targets.","marker":"[41]"},{"why":"The companion foreground-bias study, in preparation, that the authors acknowledge must accompany the statistical forecasts.","marker":"[72]"}],"fun_headline_variants":["Two-site CMB-S4 design hits σ(Neff)=0.029","CMB-S4 forecasts: σ(Neff)=0.029 in seven years","For Neff, sky coverage beats depth in CMB-S4","DRAFT forecasts confirm CMB-S4 Neff target"],"cache_read_input_tokens":3200,"weakest_assumption_plain":"The forecasts assume the adopted foreground model is accurate and that the minimum-variance ILC knows the foreground shapes exactly; if real foregrounds deviate from the PySM 'model 0' templates and the SPT-based extragalactic parameterization, the projected $\\sigma(N_\\mathrm{eff})$ values are underestimated.","fun_headline_variants_meta":{"raw":{"variants":["Two-site CMB-S4 design hits σ(Neff)=0.029","CMB-S4 forecasts: σ(Neff)=0.029 in seven years","For Neff, sky coverage beats depth in CMB-S4","DRAFT forecasts confirm CMB-S4 Neff target"]},"model":"deepseek-v4-flash","effort":"low","cost_usd":0.000348,"raw_usage":{"total_tokens":1994,"prompt_tokens":1123,"completion_tokens":871,"prompt_tokens_details":{"cached_tokens":384},"prompt_cache_hit_tokens":384,"prompt_cache_miss_tokens":739,"completion_tokens_details":{"reasoning_tokens":791}},"tokens_in":739,"tokens_out":871,"duration_ms":7724,"temperature":1.0,"reasoning_tokens":791,"cache_read_input_tokens":384,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-08-15T14:26:50.706892+00:00","model_set":{"reader":"deepseek-v4-flash"},"falsifier":"Rerun the DRAFT pipeline for the two-site conceptual design with spatially varying dust and synchrotron spectral indices, or with the measured tSZ-CIB correlation included, and check whether $\\sigma(N_\\mathrm{eff})$ rises above 0.030; likewise, the authors' own companion systematic-bias study, flagged as in preparation, settles the question if it shows foreground-induced biases at or above $\\sigma(N_\\mathrm{eff}) \\approx 0.03$.","supporting_citations":[],"review_version":2}