{"id":"5204462d-657b-46cd-96f0-624fa139a212","arxiv_id":"2504.15086","paper_version":2,"verdict":"CONDITIONAL","confidence":"MODERATE","novelty_score":5.0,"correctness_risk":"medium","formal_verification":"none","parameter_count":4,"one_line_summary":"A Moon-based radio array with roughly SKA-like collecting area could detect enough high-redshift radio-loud quasars at z≈10 to enable 21-cm forest studies, while smaller arrays reach only z≈5-6.","lead":"This paper calculates what a future radio telescope on the far side of the Moon would need to detect distant quasars that serve as background lights for studying the early universe's gas. It finds that a relatively small array could find some, while a Moon-sized version of the SKA telescope could find enough to map the era when the first stars formed.","discovery_kind":"extension","skeptic_critique":{"model":"deepseek-v4-flash","headline":"The z~10-to-16 detection limits rest on a 10% radio-loud fraction and a QLF fitted at z<=6.8, extrapolated without uncertainty bands; a plausible decline in the radio-loud fraction would shift the N=10 crossing to lower z.","rationale":"The reader's weakest_assumption identifies exactly this extrapolation issue: the QLF at z=6-6.8 is extrapolated to z=10-16, and the radio-loud and obscuration fractions are assumed redshift-independent. This is the most load-bearing condition for the abstract's central claim, because the predicted HzRLQ counts enter linearly in Eq. (3) and directly set the N>10 redshift limits. The paper explicitly flags the obscuration uncertainty but does not quantify it, and it never propagates uncertainties from the fitted model parameters into Figures 2-5. A concrete parameter-variation test is therefore the right way to decide whether the headline numbers hold. The confusion-limit formula in Eq. (11) is also questionable, but it affects the baseline-length requirement rather than the collecting-area scaling, so it is secondary for the abstract's detection-limit claims. Since the reader already marked the paper CONDITIONAL and asked for uncertainty quantification, this stress-test does not change the verdict; it sharpens the specific test that should be required before the point-value redshifts are accepted.","tokens_in":12931,"tokens_out":18892,"duration_ms":188994,"concrete_test":"Re-run the forecast pipeline of Section IV with two modifications: (i) replace the fixed 10% radio-loud fraction with the value and 1-sigma upper/lower bounds measured from the z>6 quasar radio-detection samples in Liu et al. (2021), and (ii) propagate the covariance of the fitted parameters tq and A from the z=6 and z=6.8 least-squares fits through Eq. (3). Record the N=10 crossing redshift for Nst=512 in 21-cm forest mode for the nominal, lower, and upper cases. If the crossing redshift drops by more than Delta-z ~= 1 relative to Figure 4, the headline z~10 claim is not robust to the model extrapolation and should be reported as a range rather than a point value.","verdict_should_be":"UNCHANGED","load_bearing_attack":"The paper's central forecast---that the Nst=512 configuration detects HzRLQs to z~10 in 21-cm forest mode and z~16 in continuum mode---is computed from Eq. (3), which multiplies a halo-mass-function integral by a fixed 10% radio-loud fraction and by the obscuration correction f_obsc of Eq. (4). The QLF is fitted only at z=6 and z=6.8 using optical data (Section II.B), then extrapolated to z=10-16. The text itself acknowledges (Section II.B) that f_obsc is calibrated from X-ray samples at z<5 and is assumed redshift-independent, stating that 'this may introduce some uncertainties in our predictions.' The 10% radio-loud fraction is likewise adopted from low-redshift 'observational experience.' No uncertainty propagation from tq, A, f_obsc, or the radio-loud fraction appears in Figures 2-5. Because the number counts in Eq. (3) scale linearly with the radio-loud fraction and with (1-f_obsc), a decline from 10% to 3%---within the range allowed by the sparse z>6 radio-loud quasar sample---reduces predicted counts by a factor of about 3, moving the N=10 detection limit to lower redshift and weakening the abstract's point-value claims.","agreement_with_reader":"agree"},"referee_report":{"model":"deepseek-v4-flash","summary":"The paper investigates the array configuration requirements for a future Moon-based low-frequency interferometer to detect high-redshift radio-loud quasars (HzRLQs) that could serve as background sources for 21-cm forest observations. The authors extend a previously developed quasar luminosity function (QLF) by including an X-ray-based obscuration correction, then combine this with a standard interferometer sensitivity formula to forecast the number of detectable HzRLQs for various configurations. Two survey modes are considered: a continuum survey with 10 MHz bandwidth and a 21-cm forest survey with 5 kHz spectral resolution, assuming 10^4 deg^2 of sky coverage and one year of observation. The main quantitative results are that a collecting area of about 6500 m^2 enables detection at z~6, an SKA-like array with N_st=512 extends the detection limit to z~10 for the 21-cm forest mode and z~16 for the continuum mode, and arrays with N_st=2048 reach z~11 in the 21-cm forest mode. The paper also estimates the confusion limit and concludes that a maximum baseline of about 25 km is sufficient to make confusion negligible.","tokens_in":13250,"tokens_out":8295,"duration_ms":74192,"significance":"If the forecasts hold, the paper provides useful first-order engineering guidance for a lunar low-frequency array and identifies a plausible path to obtaining the high-redshift radio-loud quasar sample needed for 21-cm forest studies during the Epoch of Reionization. The sensitivity calculation is transparent and standard, the QLF fitting procedure is clearly described, and the authors explicitly acknowledge several limitations, including the redshift-independent obscuration fraction and the extrapolation of the QLF to z>6. The paper is a forecast, not a measurement, so its value depends on the realism of the input assumptions; nevertheless, the configuration comparison is a sensible and timely contribution.","major_comments":[{"comment":"The detection counts in Eq. (3) are linear in the assumed 10% radio-loud fraction and in (1−f_obsc), and the QLF is fitted only at z=6 and 6.8 before being extrapolated to z=10–16, with f_obsc calibrated from z≲5 X-ray samples and assumed redshift-independent. No uncertainty or sensitivity analysis is presented for these inputs, so the headline redshift limits (z∼10 for 21-cm forest, z∼16 for continuum) are conditional point values. For example, decreasing the radio-loud fraction from 10% to 3%, a change within the range allowed by current sparse high-z samples, reduces predicted counts by a factor of about 3 and shifts the N=10 crossing to lower redshift. Please add a robustness test or clearly qualify all point-value claims in the abstract and conclusions.","section":"§II.B, Eq. (3), Figs. 2–5"},{"comment":"The abstract states that 'a minimum collecting area of ~6500 m^2 enables detection at z~6,' but §IV.B reports that for the same eight-station configuration (A_eff≈6547 m²) the 21-cm forest survey becomes nearly impossible beyond z>5. The abstract does not specify which survey mode this statement refers to; please reconcile or clarify the mode and detection criterion so that the abstract is not internally inconsistent.","section":"Abstract and §IV.B"},{"comment":"The confusion-limit estimate uses a single power-law fit (C=3.94, β=−1.07) with R²=0.997 but no quoted uncertainties, and adopts m=30 from Hogg (2001) without discussion of its applicability to the synthesized-beam and brightness-temperature regime considered here. Since the required baseline length is a headline engineering requirement (≳25 km), please propagate the fit errors and show the dependence of the required baseline on m, or justify m for this context.","section":"§IV.C, Eq. (11)"},{"comment":"The calculation does not state how the fixed total observing time (1 yr) is allocated among pointings for each configuration. Because the field of view changes with station diameter, the per-pointing integration time Δt in Eq. (7) differs among the scenarios in Figs. 3 and 5; without this information the comparison between configurations mixes sensitivity with survey speed. Please give the tiling formula and any survey-efficiency factor used.","section":"§III, Eq. (7)"}],"minor_comments":[{"comment":"The section heading reads 'C. onfusion limit'; it should be 'C. Confusion limit.'","section":"§IV.C"},{"comment":"The text introduces 'φ0 = 0.73' but the equation uses ψ0; please define the relationship between φ0 and ψ0 and include units for the parameters.","section":"§II.B, Eq. (4)"},{"comment":"The phrase 'least-square values' should be 'chi-square values' or 'sum of squared residuals' with the degrees of freedom stated; as written, the values 7.79 and 3.78 are not interpretable.","section":"§II.B"},{"comment":"The text refers to a 'gray histogram' for the N_st=8 configuration, but Figure 2 displays lines, not histograms; please correct the wording.","section":"§IV.A"},{"comment":"The caption should state explicitly that the first two faint-end data points are omitted from the fit; currently this information appears only in the main text.","section":"Fig. 1 caption"},{"comment":"The power-law fit expression N=3.94 S^{−1.07} lacks units for N and S; please specify per square degree and Jy, respectively.","section":"§IV.C"},{"comment":"The conclusion that the obscuration-corrected QLF yields 'more accurate predictions' overstates what can be claimed given the redshift-independent obscuration assumption; consider softening this to 'partially alleviates,' consistent with the caveat in §II.B.","section":"§V"}],"recommendation":"major_revision","confidential_remarks":"The paper is within scope for a cosmology/astrophysics journal, and the authors do acknowledge the key modeling limitations in the text, which is a positive sign. The main concern is the gap between those acknowledged limitations and the precision implied by the abstract's point-value redshift limits; a sensitivity analysis or a thorough qualification of the headline claims should resolve this. No issues with novelty or citation practice."},"author_rebuttal":null,"desk_editor":{"model":"deepseek-v4-flash","letter":"Colleague,\n\nThe paper is a straightforward design study: it takes Niu et al.'s quasar luminosity function, adds an X-ray-based obscuration correction, and runs that through a standard interferometer sensitivity calculation to predict how many high-redshift radio-loud quasars a Moon-based array could see. The genuinely new piece is the obscuration-corrected QLF and its application to a lunar survey geometry. The sensitivity treatment is transparent, the configuration comparisons are clear, and the confusion-limit analysis in Eq. (11) checks out—the reader's worry about that formula is unfounded.\n\nThe real soft spot is the extrapolation. The QLF is calibrated at z=6 and 6.8, then pushed to z=16 with a fixed 10% radio-loud fraction and a redshift-independent obscuration correction. The authors acknowledge these assumptions in Section II.B, but they never propagate uncertainties into the detection limits, and the abstract presents point values without caveats. A plausible decline in the radio-loud fraction to 3% would reduce predicted counts by a factor of about three and shift the N=10 crossing to noticeably lower redshift. For a design study that is not fatal, but the headline numbers need error bars or at least a clear sensitivity statement.\n\nThere is also a genuine abstract-level problem: the abstract says a ~6500 m^2 array 'enables detection at z~6,' but Figure 2 shows the Nst=8 configuration in continuum mode detects N>10 out to z~10, while in 21-cm forest mode it fails beyond z~5. The abstract undersells the continuum reach and muddles which observing mode it refers to. That should be fixed.\n\nWho is this for? People planning lunar low-frequency arrays and anyone who needs a quick estimate of what a Moon-based SKA-like array could do for the 21-cm forest. It is not a breakthrough, but it is a legitimate, reproducible input to design discussions, and the construction/data challenges are treated sensibly.\n\nRecommendation: accept for peer review. The core calculation is sound, the extrapolation caveat is manageable, and the abstract needs revision. A serious referee should ask for uncertainty propagation on the QLF and a clarified abstract.","headline":"A useful design forecast for a lunar low-frequency array, but the headline redshift limits rest on an unquantified QLF extrapolation and the abstract overstates the small-array reach.","tokens_in":13749,"tokens_out":3036,"would_cite":false,"duration_ms":27007,"reading_group":"maybe","serious_thinker":"yes","would_accept_peer_review":true},"rs_alignment":null,"lean_confirmation":null,"pith_extraction":{"msc":[],"pacs":[],"model":"deepseek-v4-flash","headline":"This paper forecasts that a Moon-based radio interferometer with about 6,500 m² of collecting area can detect the high-redshift radio-loud quasars needed for 21-cm forest studies at z≈6, and that an SKA-scale lunar array pushes those…","keywords":["21-cm forest","high-redshift radio-loud quasars","Moon-based interferometer","Epoch of Reionization","quasar luminosity function","radio array configuration","low-frequency radio surveys","SKA-scale array"],"falsifier":"A deep 150 MHz survey covering 10,000 square degrees down to roughly ten microjansky would count radio-loud quasars at z≈8–10; if those counts fall well below the predicted curves in the paper's Figures 2 and 4, the extrapolated luminosity function is falsified. A second, independent check is to measure the obscured quasar fraction at z>5 from X-ray-selected active galactic nuclei; a steep rise with redshift would break the flat-obscuration assumption.","tokens_in":12760,"feed_emoji":"📡","tokens_out":7473,"duration_ms":70088,"temperature":0.7,"pith_summary":"The paper works out how large a low-frequency radio interferometer on the far side of the Moon must be to find the high-redshift radio-loud quasars that 21-cm forest observations need as background light sources. Using a physical quasar luminosity function extended with a dust-obscuration correction, it forecasts detectable source counts in a 10,000 square-degree, one-year survey for both broad-band continuum mode and high-resolution 21-cm forest mode. The central numbers are that an array of eight 40 m stations, roughly 6,500 m², already produces statistically usable quasar samples in continuum mode out to z≈10, while an SKA-scale array with 512 stations reaches z≈10 in 21-cm forest mode and z≈16 in continuum mode. A larger 2048-station array, exceeding one square kilometer, extends the forest-mode limit to z≈11. The paper also shows that splitting the same collecting area into more, smaller stations improves survey speed, at the cost of sharply higher data and processing demands.","feed_headline":"6500 m² on the Moon opens the 21-cm forest","feed_subtitle":"Eight 40 m stations reach z≈6; an SKA-scale array reaches z≈10 for background quasars.","key_machinery":"The machinery is a physical-driven quasar luminosity function folded through the radio interferometer noise equation. The luminosity function starts from a halo mass function, assigns each halo a central black hole through a mass relation, converts black hole mass to quasar activity with an Eddington-limited duty cycle fitted to observed luminosity functions at z≈6 and z≈6.8, assumes that 10 percent of quasars are radio-loud, and uses an observed radio-loudness distribution to integrate the fraction whose radio flux exceeds a given threshold. That threshold is set at ten times the array noise variance, which is computed from system temperature, effective station area, number of stations, bandwidth, and integration time, so each array configuration maps directly onto predicted quasar counts. An X-ray-based obscuration correction modifies the optical luminosity function before the radio-loud fraction is applied.","core_discovery":"On its own terms, the paper establishes configuration thresholds for a lunar low-frequency array: for a 10,000 deg² survey with one year of integration, detection of high-redshift radio-loud quasars at ten times the noise level requires roughly 6,500 m² of collecting area at z≈6, about 419,000 m² in an SKA-scale layout to reach z≈10 in 21-cm forest mode and z≈16 in continuum mode, and more than one square kilometer to reach z≈11 in forest mode. It also finds that holding the total collecting area fixed while increasing the station count and shrinking station diameter, for example 8192 stations of 10 m instead of 512 stations of 40 m, raises the number of detected quasars and the redshift limit by widening the field of view, with the trade-off appearing in data transmission and computation. An additional result is that a maximum baseline of at least about 25 km keeps source confusion below thermal noise for the configurations considered. The obscuration-corrected luminosity function produces a flatter faint end and partially closes the gap between the model and observed luminosity functions at z≈6 and z≈6.8.","pith_inferences":["If future X-ray or infrared surveys show that the obscured quasar fraction grows toward z>6, the faint-end counts here are optimistic and the collecting-area thresholds should be read as lower bounds.","The same forecasting machinery could be applied directly to a ground-based SKA-scale array, so SKA-Low could empirically test the predicted quasar counts at z≈6–10 before any lunar construction, and those counts would anchor the lunar design.","The paper counts quasars per redshift bin but does not simulate the 21-cm absorption spectra themselves; the actual constraining power on the intergalactic medium will also depend on line widths and optical depths, so a detected sample of roughly ten sources per bin may yield fewer usable forest sightlines.","The engineering result on station diameter suggests an optimization frontier: for a fixed collecting area, the scientifically best layout is many small stations, but the data-rate limit may ultimately set the practical station number for a lunar observatory."],"forward_implications":["A lunar array with only eight 40 m stations can already provide a statistically meaningful sample of radio-loud quasars out to z≈10 in continuum mode, enough to begin selecting 21-cm forest background sources at z≈6.","At SKA scale, the 21-cm forest survey itself detects background quasars out to z≈10, meaning the reionization era is reachable without first requiring a separate ultra-deep continuum survey.","Continuum mode with an SKA-scale array reaches z≈16, extending quasar counts to the edge of Cosmic Dawn.","Going from 512 to 2048 stations buys roughly one additional redshift unit in 21-cm forest mode, from z≈10 to z≈11.","Reconfiguring a fixed collecting area into more, smaller stations increases the number of detectable quasars and the redshift limit, but data rates and processing load grow with the number of baselines.","A maximum baseline of roughly 25 km is sufficient to keep source confusion below thermal noise at 200 MHz for all array sizes considered."],"supporting_citations":[{"why":"Builds the physical-driven quasar luminosity function and the baseline high-redshift radio-loud quasar abundance model that this work extends.","marker":"[18]"},{"why":"Supplies the z≈6 quasar luminosity function data used to fit the quasar duty cycle.","marker":"[30]"},{"why":"Supplies the z≈6.8 quasar luminosity function data used to fit the duty cycle and test the obscuration-corrected model.","marker":"[31]"},{"why":"Provides the observed radio-loudness distribution used to set what fraction of quasars exceeds a given radio flux threshold.","marker":"[35]"},{"why":"Provides the X-ray-based obscuration fraction model used to correct the optical quasar luminosity function.","marker":"[38]"},{"why":"Provide the parameters for the obscuration fraction as a function of X-ray luminosity.","marker":"[39, 40]"},{"why":"Supplies the bolometric correction used to convert bolometric luminosity to hard X-ray luminosity in the obscuration correction.","marker":"[41]"},{"why":"Sets the receiver temperature assumption in the system-noise model.","marker":"[42]"}],"fun_headline_variants":["Moon array needs 6,500 m² for 21-cm forest at z~6","SKA-scale lunar array detects quasar forest to z~10","More stations, same area: more quasars but heavier data","25 km baseline keeps lunar 21-cm view clear"],"cache_read_input_tokens":3200,"weakest_assumption_plain":"The calculation leans on extrapolating the abundance of radio-loud quasars from redshifts 6 to 7 out to redshifts 10 to 16, assuming the same 10 percent radio-loud share and the same dust obscuration at every redshift; if radio-loud quasars become rarer at high redshift than this model says, the required array sizes are too small and the redshift limits are too high.","fun_headline_variants_meta":{"raw":{"variants":["Moon array needs 6,500 m² for 21-cm forest at z~6","SKA-scale lunar array detects quasar forest to z~10","More stations, same area: more quasars but heavier data","25 km baseline keeps lunar 21-cm view clear"]},"model":"deepseek-v4-flash","effort":"low","cost_usd":0.000427,"raw_usage":{"total_tokens":2275,"prompt_tokens":1125,"completion_tokens":1150,"prompt_tokens_details":{"cached_tokens":384},"prompt_cache_hit_tokens":384,"prompt_cache_miss_tokens":741,"completion_tokens_details":{"reasoning_tokens":1072}},"tokens_in":741,"tokens_out":1150,"duration_ms":10555,"temperature":1.0,"reasoning_tokens":1072,"cache_read_input_tokens":384,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-08-16T11:34:46.699253+00:00","model_set":{"reader":"deepseek-v4-flash"},"falsifier":"A deep 150 MHz survey covering 10,000 square degrees down to roughly ten microjansky would count radio-loud quasars at z≈8–10; if those counts fall well below the predicted curves in the paper's Figures 2 and 4, the extrapolated luminosity function is falsified. A second, independent check is to measure the obscured quasar fraction at z>5 from X-ray-selected active galactic nuclei; a steep rise with redshift would break the flat-obscuration assumption.","supporting_citations":[{"cited_title":"Iguchi, T","cited_arxiv_id":null,"evidence_quote":"Supplies the z≈6.8 quasar luminosity function data used to fit the duty cycle and test the obscuration-corrected model."},{"cited_title":"The Assembly of Black Hole Mass and Luminosity Functions of High-redshift Quasars via Multiple Accretion Episodes","cited_arxiv_id":"2210.02308","evidence_quote":"Sets the receiver temperature assumption in the system-noise model."}],"review_version":1}