{"id":"cfb17fc4-fb8b-4d58-82da-de599c3874a7","arxiv_id":"2411.11837","paper_version":3,"verdict":"CONDITIONAL","confidence":"MODERATE","novelty_score":5.0,"correctness_risk":"medium","formal_verification":"none","parameter_count":3,"one_line_summary":"JWST measurements of O, Ar, and Ne in eight galaxies at redshift 1.8 to 5.3 find an average sub-solar Ar/O ratio while Ne/O stays solar, implying that high-redshift galaxies are still dominated by core-collapse supernova enrichment.","lead":"Eight galaxies seen by JWST, around 12 billion years back in time, show less argon relative to oxygen than the Sun does. This suggests that the gas in these young galaxies was enriched mainly by the explosions of massive stars, called core-collapse supernovae, with the delayed type Ia supernovae having contributed little argon yet.","discovery_kind":"new_application","skeptic_critique":{"model":"deepseek-v4-flash","headline":"The Ar/O deficit depends on an unverified T_e[S III] scaling relation; a systematic offset of ~3000 K would shift log(Ar/O) by ~0.2-0.3 dex, comparable to the 0.19 dex deficit.","rationale":"I read the paper as a careful, honest abundance analysis. The sample is small and SFR-biased, but the authors state this. The treatment of ICFs, atomic data, and dust depletion is explicit, and those alternatives move log(Ar/O) by ~0.1 dex, which does not rescue or destroy the result. The Ne/O consistency with solar is a useful control and strengthens the differential argument. The remaining and most consequential assumption is the temperature of the Ar2+ zone. Because [Ar III] is a high-excitation line and only T_e[O III] is directly measured, the derived Ar/O is exponentially sensitive to T_e[S III]. A systematic temperature offset of 2000-3000 K, of the kind seen in the Sunburst Arc, is plausible at z~4, where ionizing spectra and density structure may differ from the local H II regions used to calibrate Eq. (1). Since the claimed offset is only 0.19 dex, a 0.2-0.3 dex systematic could fully account for it. The paper's own Section 5.3 makes exactly this point, and even shows the Welch et al. result flips when the temperature assumption is changed, demonstrating the assumption's leverage. This does not disprove the paper's scenario, but it means the central claim is not yet established at the quoted significance. Conditional acceptance is the right call, pending a direct T_e[S III] constraint at z~4.","tokens_in":24793,"tokens_out":7764,"duration_ms":78461,"concrete_test":"Stack the G395M spectra of the eight EXCELS galaxies and measure (or set a strong limit on) the [S III] λ6312 auroral line to obtain a direct T_e[S III] for the stack. Recompute all eight Ar/H and Ar/O values using this stacked T_e[S III] in place of Eq. (1) and re-derive the inverse-variance weighted mean. If the mean stays below log(Ar/O) = -2.40 (≳2σ from solar), the temperature assumption is not the driver; if it moves above -2.40 or within 2σ of solar, the paper's central claim fails. An immediate intermediate check is to recompute the mean assuming the Sunburst Arc's T_e[S III]/T_e[O III] ratio; if the mean shifts by more than ~0.15 dex, the concern is quantitatively material.","verdict_should_be":"UNCHANGED","load_bearing_attack":"The headline result is a differential statement: log(Ar/O) is 0.19 dex below solar while log(Ne/O) is solar. But Ar2+, the only observed Ar ion, is assigned to the intermediate-ionization zone, whose temperature is not measured. Section 3.1 adopts T_e[S III] = 0.83 T_e[O III] + 1700 K (Garnett 1992, Eq. 1). The [Ar III] λ7135 emissivity varies roughly as T^-1/2 exp(-E/kT), so lowering T_e[S III] from 13000 K to 10000 K raises the inferred Ar/H by ~0.2 dex, and to 8000 K by ~0.4 dex. The 1300 K scatter added to Eq. 1 only covers random scatter around a local/photoionization-calibrated mean, not a systematic offset at z~4. Welch et al. (2024) measured T_e[S III] << T_e[O III] in the Sunburst Arc; if the same relation applied to EXCELS, the weighted mean -2.50±0.07 could move above -2.31, eliminating the claimed deficit. Ne/O is measured in the high-ionization zone with the directly measured T_e[O III], so it is insensitive to this assumption; thus the clean Ne/O control does not certify Ar/O. The authors acknowledge the issue in Section 5.3 but do not fold it into the quoted significance.","agreement_with_reader":"agree"},"referee_report":{"model":"deepseek-v4-flash","summary":"The paper presents direct-method electron-temperature abundance measurements of O, Ar, and Ne for eight JWST EXCELS star-forming galaxies with redshifts 1.8 < z < 5.3 (median z ~ 4.0). Using [O III] lambda4363/lambda5007 for T_e, and the Garnett (1992) and Campbell et al. (1986) scaling relations for the intermediate and low-ionization zones, it derives total O/H and ionic Ar++/H and Ne++/H, applying ICFs from Izotov et al. (2006). The sample-averaged inverse-variance weighted log(Ar/O) = -2.50 +/- 0.07 (Ar/O = 0.65 +/- 0.10 solar) is claimed as a ~3-sigma sub-solar offset, while log(Ne/O) = -0.60 +/- 0.05 is consistent with solar. The authors interpret this as evidence that z~4 ISMs are dominated by CCSNe products with a delayed SNe Ia contribution to Ar, compare the measurements to the Kobayashi et al. (2020b) chemical evolution model and M31 PNe, and discuss systematic uncertainties from atomic data, ICFs, dust depletion, and the assumed temperature structure.","tokens_in":25269,"tokens_out":5963,"duration_ms":57337,"significance":"If correct, the result provides one of the first multi-element (O, Ar, Ne) direct-method abundance sets at z~4 and strengthens the case that alpha-element ratios can trace the onset of SNe Ia enrichment. The paper is careful: it uses PyNeb with modern atomic data, propagates Monte Carlo uncertainties, tests alternative ICFs and atomic data (Section 5.1), discusses dust depletion (Section 5.2), and explicitly acknowledges the T_e-structure caveat (Section 5.3). The Ne/O ratio serves as a sensible control, and the comparison with local CLASSY galaxies and M31 PNe is useful. The sample, though small, is a meaningful increase over previous direct Ar/O measurements at z>2.","major_comments":[{"comment":"The headline claim of a ~3-sigma sub-solar Ar/O ratio does not include the systematic uncertainty in the assumed T_e[S III] - T_e[O III] scaling relation. The Ar2+ abundance, which drives Ar/O, is derived using T_e[S III] from Garnett (1992) with only 1300 K of random scatter added in quadrature. As the authors note, Welch et al. (2024) measured T_e[S III] << T_e[O III] in the Sunburst Arc, and an alternative temperature structure could shift the weighted mean log(Ar/O) by roughly 0.2-0.3 dex, which would bring it into agreement with the solar value. Because the Ne/O ratio is measured in the high-ionization zone with the directly observed T_e[O III], the Ne/O control does not certify Ar/O. The paper should propagate an explicit systematic term (e.g., a plausible range of T_e[S III] relations) through the weighted average and quote the resulting significance, or clearly condition the headline result on the validity of the local T_e scaling relations.","section":"Section 3.1, Eq. (1); Section 5.3; Table 3"},{"comment":"The quoted uncertainty of the headline result (log(Ar/O) = -2.50 +/- 0.07; Ar/O = 0.65 +/- 0.10 solar) reflects only the internal Monte Carlo scatter; systematic contributions from atomic data, ICF choice and dispersion, and temperature scaling are discussed qualitatively but are not folded into the reported significance. Section 5.1 shows that alternative atomic data and ICFs move the average to log(Ar/O) < -2.40, a ~2-sigma offset, but this conditional statement does not appear in the abstract or conclusions. The authors should provide a combined systematic error budget and state the resulting significance, so that the '3-sigma' claim is traceable to a well-defined error model.","section":"Abstract; Section 4.1.1; Section 5.1"}],"minor_comments":[{"comment":"The abstract refers to 'measurements of O/Ar at z>2' while the rest of the paper consistently uses 'Ar/O'; the notation should be made uniform.","section":"Abstract"},{"comment":"There is a duplicated word in the sentence 'We run the default level 1 configuration except for turning on advanced snowball rejection and make use of of the CRDS_CTX...'; 'of of' should be corrected.","section":"Section 2.1"},{"comment":"The offset from solar is quoted as '~3.5 sigma' in Section 4.1.1 and as '~3 sigma' in the conclusions; these numbers should be reconciled.","section":"Section 4.1.1 and Section 6"},{"comment":"To facilitate reproducibility, the paper would benefit from a table listing the measured emission-line fluxes (or at least [Ar III] lambda7135 and [Ne III] lambda3869 fluxes with uncertainties and S/N); the zoom-in figures in Appendix A are useful but do not provide numerical values.","section":"Table 2 and Appendix A"},{"comment":"The phrase 'these affects can alter' should be 'these effects can alter' (appearing both in Section 5.2 and in point (v) of the conclusions).","section":"Section 5.2 and Section 6"}],"recommendation":"major_revision","confidential_remarks":"The manuscript is within the scope of MNRAS and the dataset is valuable. The main issue is that the quoted significance of the central Ar/O deficit does not include the acknowledged systematic uncertainty in the T_e[S III] scaling relation; this is addressable with additional propagation and revised presentation, so I recommend major revision rather than rejection."},"author_rebuttal":null,"desk_editor":{"model":"deepseek-v4-flash","letter":"The short version: this is a careful, honest abundance study of eight z~4 star-forming galaxies from JWST EXCELS, roughly doubling the sample of direct Ar/O measurements at z>2. The headline result—sub-solar Ar/O with solar Ne/O—is plausible and consistent with delayed SNe Ia enrichment of Ar, but it rests on an unverified temperature scaling relation for the intermediate-ionization zone, and the potential systematic is large enough to erase the signal.\n\nWhat's actually new: eight galaxies with direct T_e-based O, Ar, Ne abundances, adding to ~13 literature points. The analysis is standard practice done well: Monte Carlo error propagation, explicit discussion of ICFs, atomic data, dust depletion, and the comparison to Kobayashi et al. (2020b) models and M31 PNe is appropriately framed. The Ne/O control is clean and useful. The paper is also honest about its limitations—Section 5.3 flags the temperature structure issue clearly.\n\nThe soft spot is the temperature structure. Ar2+ is the only observed Ar ion, and its abundance is assigned to T_e[S III] via Garnett (1992) scaling from T_e[O III]. The 1300 K scatter added in quadrature covers random scatter around a local/photoionization-calibrated mean, not a systematic offset at z~4. Welch et al. (2024) measured T_e[S III] << T_e[O III] in the Sunburst Arc; if the same relation held for EXCELS, the weighted mean log(Ar/O) of -2.50 could move above -2.31, removing the deficit. The paper acknowledges this but does not fold it into the quoted significance. That is the difference between a 3-sigma result and a 2-sigma hint, and it is the main thing a referee should push on.\n\nThe sample is small and SFR-biased, but that is acknowledged. Data availability is 'upon reasonable request'—adequate, though not open.\n\nBottom line: this is a useful observational contribution and deserves a serious referee. The right outcome is conditional acceptance with a request for an explicit quantitative systematic budget for the temperature scaling—e.g., recompute the weighted average under alternative T_e[S III] prescriptions and report how the significance changes. I would cite this paper in my own work on high-redshift abundances, and I would bring it to reading group as a case study in how temperature systematics propagate.","headline":"A careful JWST abundance study that roughly doubles the z>2 Ar/O sample, but the headline sub-solar Ar/O deficit hinges on an unverified T_e[S III] scaling relation and could vanish under a plausible systematic offset.","tokens_in":25765,"tokens_out":2448,"would_cite":true,"duration_ms":23836,"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 finds that eight star-forming galaxies at z≈4 have an average Ar/O ratio 0.65±0.10 times solar—a ~3 sigma deficit—while Ne/O stays solar, indicating argon is not yet built up because of delayed Type Ia supernova enrichment.","keywords":["galaxy abundances","high-redshift galaxies","JWST NIRSpec spectroscopy","Type Ia supernovae","core-collapse supernovae","chemical enrichment","argon-to-oxygen ratio","electron temperature"],"falsifier":"A deep JWST spectrum of a $z \\simeq 4$ star-forming galaxy that detects the [S III] $\\lambda$6312 auroral line would give an independent intermediate-zone temperature; if $T_e[\\mathrm{S\\,III}]$ comes out substantially below $0.83\\,T_e[\\mathrm{O\\,III}] + 1700$ K, as it does for the Sunburst Arc, the inferred Ar$^{2+}$ abundance rises and the reported Ar/O deficit would weaken or disappear.","tokens_in":24596,"feed_emoji":"🌌","tokens_out":11916,"duration_ms":101001,"temperature":0.7,"pith_summary":"This paper uses JWST/NIRSpec spectra of eight star-forming galaxies at redshift $\\langle z \\rangle = 4.0$ to measure oxygen, argon, and neon abundances from faint emission lines. It finds that the average argon-to-oxygen ratio is $0.65 \\pm 0.10$ times the solar value, a roughly $3\\sigma$ deficit, while the neon-to-oxygen ratio is $1.07 \\pm 0.12$ times solar. The authors interpret the argon shortfall as evidence that the interstellar medium of these young galaxies is still dominated by core-collapse supernova products, with argon not yet built up because a substantial fraction of its production is delayed through Type Ia supernovae. The result matters because Ar/O can trace the delayed enrichment from Type Ia supernovae using emission lines that JWST can observe, without the deep ultraviolet continuum required for iron abundances.","feed_headline":"z≈4 galaxies carry 35 percent less argon than the Sun","feed_subtitle":"Argon builds late via Type Ia supernovae, so its shortfall signals young interstellar gas.","key_machinery":"The machinery is the direct-method abundance analysis built on electron temperatures: the [O III] $\\lambda$4363/$\\lambda$5007 ratio fixes the high-ionization-zone temperature, locally calibrated scaling relations assign the low- and intermediate-zone temperatures ($T_e[\\mathrm{O\\,II}] = 0.7\\,T_e[\\mathrm{O\\,III}] + 3000$ K and $T_e[\\mathrm{S\\,III}] = 0.83\\,T_e[\\mathrm{O\\,III}] + 1700$ K), and the [Ar III] $\\lambda$7135 and [Ne III] $\\lambda$3870 lines yield the observed argon and neon ions. Ionization correction factors from Izotov et al. (2006) complete the unseen Ar$^{+}$ and Ar$^{3+}$ and Ne$^{3+}$ stages, converting the line ratios into total abundances. This turns the single measurable ratio Ar/O into a tracer of the relative contribution of core-collapse versus Type Ia supernovae to the interstellar medium, with Ne/O serving as a control ratio expected to track oxygen.","core_discovery":"The central discovery is that the inverse-variance weighted average of the eight EXCELS galaxies is $\\log(\\mathrm{Ar/O}) = -2.50 \\pm 0.07$, corresponding to $\\mathrm{Ar/O} = 0.65 \\pm 0.10\\,(\\mathrm{Ar/O})_\\odot$, a roughly $3\\sigma$ offset below solar; in contrast, the average neon-to-oxygen ratio is consistent with the solar ratio ($1.07 \\pm 0.12$ times solar). On the paper's interpretation this is exactly the pattern expected if argon is an $\\alpha$-element with a significant Type Ia supernova production channel, about $34$ per cent in current yield models, while neon tracks oxygen as a core-collapse product. Young galaxies at $z \\simeq 4$ have not yet accumulated enough Type Ia enrichment, so their interstellar medium is dominated by core-collapse supernova yields. The sample average falls at the knee of the Milky Way chemical evolution track and at the extension of the M31 planetary-nebula sequence in the $\\log(\\mathrm{O/Ar})$--$\\log(\\mathrm{Ar/H})$ plane, connecting these early systems to the delayed-enrichment pathway seen locally.","pith_inferences":["If the Ar deficit holds up, the knee in the $\\log(\\mathrm{O/Ar})$--$\\log(\\mathrm{Ar/H})$ plane could serve as a star-formation-history clock in the same way the O/Fe knee does; galaxies with older stellar populations at fixed metallicity should show higher Ar/O, a prediction the paper does not test.","The outliers with near-solar Ar/O, EXCELS-121806 and the Sunburst Arc, could share a common temperature-structure cause rather than exotic enrichment; comparing their [S III] $\\lambda$6312 temperatures would discriminate between these explanations.","Because oxygen depletes onto dust while argon and neon do not, an independent oxygen tracer that sidesteps depletion could move the Ar/O offset to larger significance and strengthen the SNe Ia delay interpretation.","Selecting galaxies by specific star-formation rate at fixed redshift would provide a sharper test: if the Ar deficit comes from delayed SNe Ia, galaxies with lower specific star-formation rates should already show Ar/O closer to solar."],"forward_implications":["If the deficit is real, the interstellar medium of $z \\simeq 4$ star-forming galaxies is predominantly enriched by core-collapse supernova products, with Type Ia supernova products still sub-dominant.","Ar/O becomes a practical high-redshift analogue of O/Fe that can be measured from JWST emission lines rather than from deep far-ultraviolet continuum, greatly increasing the number of galaxies in which SNe Ia enrichment can be tracked.","The solar Ne/O ratio in the same galaxies confirms neon follows oxygen, validating the use of Ne/O as a control element and ruling out a general anomaly in all $\\alpha$-element ratios.","Agreement with the Milky Way chemical evolution model and the M31 planetary-nebula sequence implies the same delayed-enrichment pathway operating locally is already visible at $z \\simeq 4$.","Future moderate-depth JWST spectroscopy can extend this measurement to larger samples and higher redshifts, testing whether the Ar/O deficit evolves as Type Ia supernovae begin to contribute."],"supporting_citations":[{"why":"Supplies the Milky Way chemical evolution model and yield calculations used as the main theoretical track for interpreting Ar/O versus Ar/H.","marker":"Kobayashi et al. (2020b)"},{"why":"Provides nucleosynthesis yields giving about 34 per cent of argon from Type Ia supernovae, the quantitative expectation behind the deficit.","marker":"Kobayashi et al. (2020a)"},{"why":"Supplies the Te[O II]-Te[O III] scaling relation used to set the low-ionization-zone temperature for the O+ abundance.","marker":"Campbell et al. (1986)"},{"why":"Supplies the Te[S III]-Te[O III] scaling relation used to set the intermediate-zone temperature that controls the Ar2+ abundance.","marker":"Garnett (1992)"},{"why":"Supplies the ionization correction factors used to convert the observed Ar2+ and Ne2+ ionic abundances into total abundances.","marker":"Izotov et al. (2006)"},{"why":"Provides the earlier z~3 CECILIA galaxy measurement of sub-solar Ar/O that the EXCELS average is consistent with.","marker":"Rogers et al. (2024)"},{"why":"Provides the Sunburst Arc with solar-like Ar/O, which the paper explains through its different measured temperature structure.","marker":"Welch et al. (2024)"},{"why":"Provides the local CLASSY comparison sample and tests of the argon ionization correction factors used to validate the abundances.","marker":"Arellano-Córdova et al. (2024b)"},{"why":"Provides the M31 planetary-nebula O/Ar versus Ar/H sequence onto which the high-redshift average is placed.","marker":"Arnaboldi et al. (2022)"}],"fun_headline_variants":["JWST finds 35% argon deficit in early galaxies","Argon-poor galaxies at z=4 reveal young interstellar gas","Webb's EXCELS survey spots missing argon in ancient starbursts","Early galaxy argon shortfall traces Type Ia supernovae","z=4 galaxies show delayed argon enrichment, JWST shows"],"cache_read_input_tokens":3200,"weakest_assumption_plain":"The load-bearing premise is that the locally calibrated electron-temperature scaling relations remain valid at $z \\simeq 4$, because the paper only measures the [O III] temperature directly; if the intermediate-zone temperature is actually colder than assumed, the derived argon abundance rises and the Ar/O deficit shrinks toward solar.","fun_headline_variants_meta":{"raw":{"variants":["JWST finds 35% argon deficit in early galaxies","Argon-poor galaxies at z=4 reveal young interstellar gas","Webb's EXCELS survey spots missing argon in ancient starbursts","Early galaxy argon shortfall traces Type Ia supernovae","z=4 galaxies show delayed argon enrichment, JWST shows"]},"model":"deepseek-v4-flash","effort":"low","cost_usd":0.000389,"raw_usage":{"total_tokens":2138,"prompt_tokens":1122,"completion_tokens":1016,"prompt_tokens_details":{"cached_tokens":384},"prompt_cache_hit_tokens":384,"prompt_cache_miss_tokens":738,"completion_tokens_details":{"reasoning_tokens":927}},"tokens_in":738,"tokens_out":1016,"duration_ms":9869,"temperature":1.0,"reasoning_tokens":927,"cache_read_input_tokens":384,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-08-12T18:03:52.229549+00:00","model_set":{"reader":"deepseek-v4-flash"},"falsifier":"A deep JWST spectrum of a $z \\simeq 4$ star-forming galaxy that detects the [S III] $\\lambda$6312 auroral line would give an independent intermediate-zone temperature; if $T_e[\\mathrm{S\\,III}]$ comes out substantially below $0.83\\,T_e[\\mathrm{O\\,III}] + 1700$ K, as it does for the Sunburst Arc, the inferred Ar$^{2+}$ abundance rises and the reported Ar/O deficit would weaken or disappear.","supporting_citations":[],"review_version":1}