{"id":"14361011-7c35-4b49-9201-b10385ff574e","arxiv_id":"2506.17400","paper_version":2,"verdict":"CONDITIONAL","confidence":"MODERATE","novelty_score":6.0,"correctness_risk":"medium","formal_verification":"none","parameter_count":5,"one_line_summary":"Population III galaxies can stay detectable through metal emission lines in a brief self-polluted phase, and new UV diagnostics may reveal them in JWST surveys.","lead":"This paper models how galaxies containing the first metal-free stars evolve as those stars pollute their surroundings with supernova metals. It proposes new ultraviolet line-ratio diagnostics and applies them to JWST data, identifying nine possible Population III galaxy candidates.","discovery_kind":"new_application","skeptic_critique":{"model":"deepseek-v4-flash","headline":"Self-polluted phase lasts a single 1-Myr timestep in NEFERTITI, so the candidate pool depends on an unmodeled star-formation delay.","rationale":"The reader's conditional verdict already flags the short self-polluted phase and U-dependence. I agree but place the timescale at the center: it is an internal limitation that the paper itself acknowledges, and it directly controls whether the proposed diagnostics have any practical detection window. The nine JADES candidates are selected using lines whose predicted ratios come from snapshots in this 1-Myr phase, so if the true duration is much shorter—or if the phase does not exist at finer time resolution—the candidate claim loses its foundation. The U-dependence is a separate external parameter uncertainty; the timescale is a more fundamental internal model property. I therefore do not change the reader's CONDITIONAL verdict: the paper is a useful model-guided search strategy, but the quantitative candidate claims should be treated as conditional on a delay that is not modeled.","tokens_in":11960,"tokens_out":7842,"duration_ms":79682,"concrete_test":"Run a controlled NEFERTITI experiment on a subset of halos with a time resolution of 0.1 Myr (instead of 1 Myr) and measure the duration of the self-polluted phase, defined by Z_gas > Z_cr with no PopII stars yet formed. If the measured duration is not robustly >3 Myr—or collapses below 1 Myr—the self-polluted phase is a timestep artifact, and the JADES candidate selection should be re-evaluated using models with an explicit star-formation delay parameter t_delay = 0, 1, 3, 10 Myr. If t_delay=0 yields no self-polluted phase longer than one timestep, the paper's candidate pool rests entirely on the unmodeled delay.","verdict_should_be":"UNCHANGED","load_bearing_attack":"Section 3.1 states that in NEFERTITI the self-polluted phase—pure PopIII stars with gas enriched by PopIII SNe—\"lasts only for our time resolution of 1 Myr,\" because the enriched gas immediately reaches Z_cr and PopII stars form. Yet the paper's central claim that this phase opens a pool of more easily detectable candidates, and the reported nine JADES candidates with >25% PopIII stellar mass, depend on this phase being observable. The Discussion acknowledges that the phase can be extended only \"if we consider a temporal delay between PopIII SNe and the subsequent star formation,\" citing Katz et al. 2023. That delay is not implemented in NEFERTITI; it is an external, unmodeled assumption. Without it, the self-polluted phase is a single 1-Myr timestep, so the [OIII]/Hbeta approximately 1 signature is essentially instantaneous in the fiducial simulation. The candidate diagnostics (Eqs. 1-4) are calibrated on snapshots from this phase, so the predicted candidate count is not a robust model prediction. The ionization-parameter dependence (Appendix A) compounds this: at logU < -1, the oxygen line weakens further, making the phase even harder to catch. The paper is honest about the 1-Myr duration, but the abstract and conclusions present the self-polluted phase as a general property rather than a resolution-limited, short-lived state.","agreement_with_reader":"partial"},"referee_report":{"model":"deepseek-v4-flash","summary":"This paper uses the NEFERTITI semi-analytical galaxy formation model coupled to Cloudy photoionization calculations to synthesize emission-line spectra of about 1128 PopIII galaxies from z = 17 to z = 6. It identifies a 'self-polluted' phase in which the stellar population is purely metal-free while the gas has been chemically enriched by the first PopIII supernovae, and claims that in this phase [OIII]/Hbeta is approximately 1 at the fiducial logU = -1. The paper further predicts that strong HeII emission can persist up to about 20 Myr, including during the hybrid phase when PopIII and PopII stars coexist, and proposes UV metal-line diagnostic relations (Eqs. 1-4) to select PopIII candidates with more than 25% of stellar mass in metal-free stars. Applying these diagnostics to JADES DR3 data yields nine candidate galaxies. The paper is transparent about some limitations, notably the 1-Myr duration of the self-polluted phase in NEFERTITI.","tokens_in":12256,"tokens_out":6571,"duration_ms":65824,"significance":"If the main claims are robust, the paper would broaden the JWST search for PopIII galaxies beyond the classic criterion of HeII emission without metal lines, by showing that metal lines can actually be present in galaxies hosting only metal-free stars. The modeling chain is a strength: NEFERTITI self-consistently follows chemical enrichment and stochastic IMF sampling, and the synthetic spectra are compared with external AGN and star-forming galaxy models. The proposed diagnostics are falsifiable and the JADES candidate list is a concrete, testable output. However, the quantitative claims are not yet robust: the self-polluted phase is a single 1-Myr timestep in the fiducial model, the ionization parameter is a free parameter, and the candidate selection relies on low-significance detections. These issues affect the central 'pool of candidates' claim and need to be addressed before the results can be considered secure.","major_comments":[{"comment":"The self-polluted phase, defined as purely PopIII stars with gas already enriched by PopIII SNe, lasts only one 1-Myr timestep in NEFERTITI because the enriched gas immediately reaches the critical metallicity and PopII stars form. The paper's central claim that this phase 'opens the pool of candidates' therefore depends on an unmodeled temporal delay between the PopIII supernovae and the subsequent star formation. The Discussion mentions this delay and cites Katz et al. (2023), but it is not implemented in the fiducial model. Without it, the [OIII]/Hbeta approximately 1 signature is essentially instantaneous, and the predicted candidate rate is not a robust model prediction. Please either implement such a delay in the model or quantify the detection probability implied by a 1-Myr duration.","section":"Section 3.1 and Discussion"},{"comment":"The ionization parameter U is a free parameter in the modeling, and Appendix A shows that [OIII]/Hbeta decreases with decreasing U. Thus the claimed [OIII]/Hbeta approximately 1 signature, obtained at the fiducial logU = -1, is not robust across the explored range logU = [-3, -0.5, 0]. The paper does not derive a physically motivated value of U for PopIII galaxies; the restriction to logU > -2 in Section 3.2 is an assumption. Because the location of the self-polluted phase and the diagnostic boundaries depend on U, the quantitative predictions should be marginalized over U or justified with a physical argument.","section":"Appendix A and Fig. 5"},{"comment":"The diagnostic boundaries in Eqs. (1)-(4) are defined by the authors' own NEFERTITI+Cloudy model grid; the external AGN and star-forming galaxy models are used only to show that they do not populate the selected regions, not to validate the boundaries independently. In addition, the JADES candidate selection uses tentative detections with S/N > 2, and the caption of Fig. 3 states that all HeII/Hbeta measurements are 5-sigma upper limits. The claim of nine PopIII candidates is therefore not yet a secure observational result. The paper should present the expected completeness and false-positive rate of the selection, or explicitly label the candidate number as an upper limit pending deeper observations.","section":"Section 3.2 and Fig. 3"}],"minor_comments":[{"comment":"In the second paragraph of the Discussion, '25% M⋆,PopIIII' contains an extra 'I' in 'PopIIII'; it should read 'PopIII'.","section":"Discussion"},{"comment":"The notation 'Z gas' for the gas metallicity should be made consistent, for example as Z_gas, to avoid confusion with the stellar metallicity and with the critical metallicity Z_cr.","section":"Section 2.1"},{"comment":"The meaning of 'S/N > 2' should be defined: is it the signal-to-noise ratio of the integrated line flux, and which lines are required to meet this threshold?","section":"Fig. 3"},{"comment":"The quadratic fit for b(U) in Eq. (B5) is presented without a goodness-of-fit statistic or a statement of the range of U over which the fit is valid; please add this information.","section":"Appendix B"},{"comment":"The left schematic in Fig. 1 groups phases into pure PopIII and hybrid, but the color coding of the density distributions in the right panel is not described in the caption; please clarify which colors correspond to which evolutionary stage.","section":"Fig. 1"}],"recommendation":"major_revision","confidential_remarks":"The paper is suitable in scope for a Letters journal and the authors are transparent about the 1-Myr duration of the self-polluted phase. The main issues are model-dependent and can, in principle, be addressed within the manuscript's scope by adding a treatment or quantitative discussion of the star-formation delay, by marginalizing over the ionization parameter, and by presenting the candidate selection with appropriate caveats. I therefore recommend major revision rather than rejection."},"author_rebuttal":null,"desk_editor":{"model":"deepseek-v4-flash","letter":"Colleague, here's the take: engage with the core idea, but treat the headline candidate count with caution. The genuinely new result is that a galaxy hosting only metal-free stars can still show metal lines, because the gas is already enriched by the first supernovae. That reverses the usual \"no metal lines = PopIII\" logic and gives a concrete reason to look for [OIII]/Hbeta ~ 1 sources. The HeII lifetime extension to ~20 Myr in the hybrid phase is a nice model result, and the UV diagnostic diagrams are a useful organizing framework. The modeling itself is careful: NEFERTITI is locally calibrated, they run ~28,000 Cloudy models, and they are transparent about the free parameters.\n\nThe soft spots are real. The self-polluted phase lasts one 1-Myr timestep in NEFERTITI; the enriched gas immediately reaches Z_cr and PopII forms. The paper says this honestly in Section 3.1, and the Discussion acknowledges that a longer phase requires an unmodeled delay between SNe and subsequent star formation. But the abstract and conclusions present the \"pool of candidates\" as a general property. The nine JADES candidates are selected with S/N > 2 detections, the diagnostic boundaries are fitted to their own models, and at least one source is a previously identified AGN. The ionization parameter dependence in Appendix A compounds this: at logU < -1 the oxygen line weakens, so the [OIII]/Hbeta ~ 1 signature is not robust across the explored range. No code or data are shipped, so the candidate list is not independently reproducible at this stage.\n\nNone of this kills the central argument. The physical point—metal lines do not rule out PopIII—holds as a logical consequence of self-pollution. What is fragile is the quantitative claim that this opens a large pool of easily detectable candidates. That depends on a delay mechanism that is not in the model. If real systems have a short delay, the self-polluted phase is essentially instantaneous and the JADES candidates become model-guided targets rather than confirmed PopIII galaxies.\n\nFor peer review: send it. It deserves a serious referee. The referee should push on the delay timescale, the U prior, and the S/N threshold for the candidates, but the paper is coherent, honest about its main limitation, and likely to stimulate follow-up work. I'd bring it to reading group.","headline":"The physical idea that metal lines don't rule out PopIII is worth taking seriously, but the candidate pool is model-dependent and the self-polluted phase is a single timestep without an invoked delay.","tokens_in":12813,"tokens_out":2062,"would_cite":true,"duration_ms":20691,"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":"Metal lines betray galaxies that host only first stars","keywords":["Population III stars","high-redshift galaxies","chemical enrichment","James Webb Space Telescope","emission-line diagnostics","first stars","galaxy formation","photoionization modeling"],"falsifier":"Observe a sample of high-redshift galaxies selected to lie inside the proposed UV diagnostic regions with JWST/NIRSpec deep spectroscopy; if the predicted HeII λ1640 emission (or EW(HeII λ4686) ≳ 4 Å for the optical cases) is not detected, the diagnostics are selecting ordinary star-forming galaxies or AGN rather than PopIII-dominated systems. Alternatively, measure CIV/CIII] to infer the ionization parameter: if the candidates cluster at log U ≤ −2, the claimed selection region does not apply to real PopIII galaxies.","tokens_in":11734,"feed_emoji":"🔭","tokens_out":6597,"duration_ms":59903,"temperature":0.7,"pith_summary":"The paper argues that galaxies hosting only metal-free Population III stars are not necessarily metal-line-free: once the first supernovae explode, the gas is chemically enriched while the stars remain pristine, and in that self-polluted phase the galaxies show [OIII]/Hβ ≈ 1. If true, the standard practice of rejecting high-[OIII] sources as PopIII candidates is too restrictive, and the pool of detectable first-star galaxies is larger than thought. The paper predicts the bright HeII signature of PopIII stars can persist up to roughly 20 Myr, including the hybrid phase in which later-generation stars already coexist. It also proposes UV metal-line ratio diagnostics that isolate galaxies with more than 25% of stellar mass in metal-free stars, and it applies them to JADES data to identify nine such candidates.","feed_headline":"Metal lines betray galaxies that host only first stars","feed_subtitle":"New UV line-ratio diagnostics catch pristine-star galaxies in their self-polluted phase; JWST data already hold 9 candidates.","key_machinery":"The argument runs on a semi-analytical galaxy-formation model (NEFERTITI) that follows individual PopIII stars, their supernova yields, and the chemical enrichment of gas within Milky Way–like progenitors, coupled to the Cloudy photoionization code to synthesize emission-line spectra for about 28,000 model galaxies. The load-bearing output is the evolutionary staging of a PopIII galaxy—pristine, self-polluted, hybrid, and PopIII-poor—based on the fraction of stellar mass in metal-free stars. The new candidate-selection tool is the set of four UV metal-line ratio inequalities, valid for log U > −2, which define regions populated only by galaxies with more than 25% of stellar mass in metal-free stars.","core_discovery":"The central claim is that a pure PopIII galaxy—one whose stellar population contains no metals—can display metal emission lines because the interstellar gas has been enriched by supernovae from those same first stars. In this self-polluted phase the oxygen line ratio sits near [OIII]/Hβ ≈ 1, so metal emission does not rule out a PopIII nature. The same models show strong HeII emission (EW(HeII λ4686) ≳ 4 Å) can last up to roughly 20 Myr, part of it in the hybrid phase where PopII stars are already forming. From the modeled line ratios, the paper derives simple inequalities in CIII]/HeII versus CIV/CIII], OIII]/HeII, CIV/HeII, and SiIII]/HeII that select galaxies with M⋆,PopIII > 25% M⋆ while avoiding AGN and normal star-forming contamination. Applying these to the JADES DR3 catalog yields nine PopIII candidates.","pith_inferences":["Editorial inference: If real PopIII star formation is delayed relative to the first supernovae longer than the single 1-Myr timestep assumed here, the self-polluted phase would last longer and the number of detectable PopIII galaxies could be substantially larger than the nine JADES candidates.","Editorial inference: The strong dependence of CIV and OIII] on the ionization parameter means the same line ratios may be used to measure U in candidate sources; a target with log U ≤ −2 would fall outside the claimed diagnostic region, so such measurements provide an independent test.","Editorial inference: The same UV-line selection could be applied to other JWST surveys and to gravitationally lensed fields, where brighter fluxes would allow direct confirmation of the predicted HeII equivalent widths.","Editorial inference: Because the diagnostics select on stellar-mass fraction rather than pure metal-free stars, some of the nine candidates may turn out to be hybrids with only a minority PopIII component; deep HeII EW measurements will separate those cases."],"forward_implications":["High-[OIII] galaxies at high redshift should no longer be excluded from PopIII candidacy on the basis of metal lines alone; self-polluted pure PopIII galaxies naturally have [OIII]/Hβ ≈ 1.","Strong HeII emission remains a viable PopIII indicator for up to ~20 Myr, including the early hybrid phase, extending the observable window well beyond the few-Myr pristine phase.","The four UV line-ratio diagnostics (Eqs. 1–4) can select PopIII-dominated galaxies with >25% PopIII stellar mass with no AGN or ordinary star-forming contamination for log U > −2.","Nine galaxies in the JADES DR3 catalog satisfy the new diagnostics; they are the systems most worth deep follow-up to confirm HeII.","The known candidates LAP1-B and RX J2129–z8HeII are consistent with the models, including pure self-polluted PopIII galaxies, though AGN contamination is not excluded for RX J2129–z8HeII."],"supporting_citations":[{"why":"Supplies the NEFERTITI galaxy-formation model that tracks individual PopIII stars and chemical enrichment.","marker":"I. Koutsouridou et al. 2023"},{"why":"Provides PopIII stellar spectra and evolutionary tracks used to build the synthetic stellar emission and the HeII timescale.","marker":"D. Schaerer 2002"},{"why":"Supplies the Cloudy photoionization code used to compute the nebular emission-line spectra for the model galaxies.","marker":"M. Chatzikos et al. 2023"},{"why":"Defines the EW(HeII) threshold and AGN-contaminated region that frame the proposed PopIII diagnostics.","marker":"K. Nakajima & R. Maiolino 2022"},{"why":"Provides the AGN photoionization models used to check that the proposed UV diagnostics are not contaminated by AGN.","marker":"A. Feltre et al. 2016"},{"why":"Provides the star-forming galaxy photoionization models used to show that ordinary PopII/I galaxies fall outside the PopIII selection regions.","marker":"J. Gutkin et al. 2016"},{"why":"Supplies the JADES DR3 line-flux catalog from which the nine PopIII candidates are drawn.","marker":"F. D'Eugenio et al. 2025"},{"why":"Sets the critical gas metallicity threshold that triggers PopII formation, controlling the duration of the self-polluted phase.","marker":"M. de Bennassuti et al. 2017"}],"fun_headline_variants":["Metal lines fingerprint pure PopIII galaxies, adding 9 JWST candidates","Self-polluted metal lines reveal pristine-star galaxies: 9 JWST picks","Metal lines don't rule out PopIII stars: 9 candidates from JADES","First stars pollute their own gas: metal lines now a PopIII signpost","Self-polluted first stars: new diagnostics find 9 PopIII galaxies in JWST"],"cache_read_input_tokens":3200,"weakest_assumption_plain":"The whole search window depends on the assumed ionization parameter (fiducial log U = −1) and on the assumption that the gas stays enriched for at least about a million years before second-generation stars form; if real first-star systems have lower U or form PopII stars immediately, the self-polluted phase and the proposed diagnostics would be far rarer.","fun_headline_variants_meta":{"raw":{"variants":["Metal lines fingerprint pure PopIII galaxies, adding 9 JWST candidates","Self-polluted metal lines reveal pristine-star galaxies: 9 JWST picks","Metal lines don't rule out PopIII stars: 9 candidates from JADES","First stars pollute their own gas: metal lines now a PopIII signpost","Self-polluted first stars: new diagnostics find 9 PopIII galaxies in JWST"]},"model":"deepseek-v4-flash","effort":"low","cost_usd":0.0007,"raw_usage":{"total_tokens":3208,"prompt_tokens":1040,"completion_tokens":2168,"prompt_tokens_details":{"cached_tokens":384},"prompt_cache_hit_tokens":384,"prompt_cache_miss_tokens":656,"completion_tokens_details":{"reasoning_tokens":2063}},"tokens_in":656,"tokens_out":2168,"duration_ms":16321,"temperature":1.0,"reasoning_tokens":2063,"cache_read_input_tokens":384,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-08-15T19:09:07.912427+00:00","model_set":{"reader":"deepseek-v4-flash"},"falsifier":"Observe a sample of high-redshift galaxies selected to lie inside the proposed UV diagnostic regions with JWST/NIRSpec deep spectroscopy; if the predicted HeII λ1640 emission (or EW(HeII λ4686) ≳ 4 Å for the optical cases) is not detected, the diagnostics are selecting ordinary star-forming galaxies or AGN rather than PopIII-dominated systems. Alternatively, measure CIV/CIII] to infer the ionization parameter: if the candidates cluster at log U ≤ −2, the claimed selection region does not apply to real PopIII galaxies.","supporting_citations":[],"review_version":2}