{"id":"f908708a-bcf0-4a89-bb4b-711d5b0001d7","arxiv_id":"1908.01541","paper_version":1,"verdict":"CONDITIONAL","confidence":"HIGH","novelty_score":6.0,"correctness_risk":"medium","formal_verification":"none","parameter_count":1,"one_line_summary":"ALMA far-infrared measurements show two gamma-ray burst host galaxies at z~2 are ordinary main-sequence star formers, and one shows a faint [C II] line hinting at a possible [C II] deficit.","lead":"Two galaxies that hosted gamma-ray bursts at redshift 2 are forming stars at ordinary rates, about 50 and 123 solar masses per year, not the extreme rates implied by radio data. The result matters because gamma-ray burst rates are used to trace cosmic star-formation history, and it includes the first detection of a faint carbon cooling line in a high-redshift burst host, hinting at a possible [C II] deficit.","discovery_kind":"new_application","skeptic_critique":{"model":"deepseek-v4-flash","headline":"The marginal [C ii] 158 µm detection (S/N~4, 4.4±1.2 Jy km/s) is the sole support for the new 'deficit' claim; it needs independent confirmation before being treated as secure.","rationale":"The reader's weakest-assumption diagnosis is correct: the marginal [C ii] line is the load-bearing element for the most novel claim. The SFR results are well supported by high-S/N ALMA continuum detections and a standard SED code, and they do not depend on the line; even the single-FIR-point case of GRB 070521 is sufficient to rule out the radio-derived 817 M_sun/yr SFR, so the main-sequence and radio-overestimate conclusions are not threatened by the line's marginality. The [C ii] deficit, however, is a one-line, S/N~4 result, explicitly labeled marginal by the authors, and the physical interpretation that the host is a [C ii]-deficit outlier is not robust to a spurious or biased line flux. A conditional verdict is therefore appropriate: the measurements and main conclusions are credible, but the headline new claim should not be taken as established without independent confirmation. I found no reason to move the verdict to ACCEPT or REJECT based on my read; UNCHANGED is the correct output, keeping the reader's CONDITIONAL verdict.","tokens_in":24018,"tokens_out":6297,"duration_ms":69366,"concrete_test":"Re-extract the [C ii] spectrum from the calibrated visibilities in the uv-plane using only line-free channels, subtract the continuum with a uv-plane fit, and measure the line with a matched filter whose velocity offset and width are fixed a priori by the CO(1-0) detection (offset ~100 km/s; Hatsukade et al. 2019), not tuned to the [C ii] candidate. Require S/N>5 at the continuum position and a centroid within one beam of the continuum. Also recompute L[CII]/LFIR using 2-sigma extremes of both the line flux and LFIR; if the significance drops below ~3 sigma or the ratio is no longer clearly below the z~1-2 comparison sequence, the deficit claim should be presented only as a tentative upper limit.","verdict_should_be":"UNCHANGED","load_bearing_attack":"The paper's distinctive new result is the first [C ii] detection in a z>2 GRB host and the resulting [C ii] deficit interpretation (Section 5.3). That entire interpretation rests on one 8-minute band 9 observation in which the line is seen at S/N~4 after continuum subtraction and after integrating over ±125 km/s chosen around the candidate (Section 3, Table 2). The authors themselves call the detection 'marginal.' A S/N~4 line is vulnerable to correlated noise in the continuum-subtracted spectrum, to a posteriori choice of integration window and velocity offset, and to noise peaks at the known redshift being preferred. If the line is spurious, or its flux is biased high, then L[CII]/LFIR = 7.5e-4 and the 'one of the smallest ratios' and 'deficit' conclusions collapse. The luminosity ratio also lacks any propagated uncertainty. By contrast, the FIR continuum detections and the resulting SFRs are high-S/N (S/N~8, 19, 7) and are independent of the [C ii] claim, so the main-sequence and radio-overestimate conclusions survive even if the line is doubted. Minor internal issues include swapped ALMA continuum fluxes between Sections 5.1.1 and 5.1.2 (11.9 vs 3.58 mJy) and an abstract that attributes the radio excess to afterglow contamination while Section 5.1.2 favors a conversion-factor explanation for the 080207 host; these are presentation issues, not the load-bearing weakness.","agreement_with_reader":"agree"},"referee_report":{"model":"deepseek-v4-flash","summary":"The paper reports ALMA band 8/9 rest-frame far-infrared (FIR) continuum observations of two GRB host galaxies at z~2, GRB 070521 and GRB 080207, and a marginal detection of [C ii] 158 μm emission from the latter. The FIR detections are used with the public MAGPHYS high-z SED-fitting code to derive star-formation rates of 49.85(+72.33,-2.86) and 123.4(+25.19,-21.78) M_sun/yr, respectively, which place both hosts on the z~2 main sequence and well below earlier radio-derived SFRs. The paper further reports L[CII]/LFIR = 7.5e-4 for GRB 080207, interprets this as one of the smallest values at z~1-2 and as a possible '[C ii] deficit' characteristic of GRB hosts, and argues that the previous radio SFRs are overestimated by afterglow contamination. The two main scientific claims are therefore (1) that FIR-based SFRs revise these hosts from extreme starbursts to normal star-forming galaxies, and (2) that the first z>2 GRB-host [C ii] detection reveals a strong [C ii] deficit.","tokens_in":24255,"tokens_out":6864,"duration_ms":64884,"significance":"If the FIR-based SFRs hold, the paper makes a useful and timely contribution: it demonstrates that ALMA FIR photometry can correct radio-based SFR estimates that may be biased by afterglow contamination, and it strengthens the view that high-z GRB hosts are not necessarily extreme starbursts. The continuum detections are high signal-to-noise (S/N ~7-19), the SED fitting uses a standard public code, and the authors include several appropriate caveats, including a discussion of observational bias toward low L[CII]/LFIR in deep ALMA observations. The [C ii] deficit claim is novel and potentially important, but it rests on the weakest measurement in the paper, a S/N~4 line detection. The SFR results are largely independent of the [C ii] claim and should survive even if the line is questioned; the deficit interpretation, however, is not on the same footing without a robustness analysis.","major_comments":[{"comment":"The entire [C ii] deficit claim rests on a single marginal line detection: S/N ~4, velocity-integrated flux 4.4 ± 1.2 Jy km/s, integrated over ±125 km/s after continuum subtraction and at a velocity offset of 155 km/s from the optical redshift (Table 2, Fig. 2). The paper itself calls the detection 'marginal' in Section 3, yet Section 5.3 and the Conclusions treat the resulting L[CII]/LFIR = 7.5e-4 as one of the smallest values at z~1-2. Because this is the paper's most novel claim, I request a quantitative robustness analysis: noise statistics of the continuum-subtracted cube, significance as a function of the chosen integration window, checks for a posteriori selection effects, and a bootstrap or jackknife estimate of the line flux. If the line cannot be established against plausible correlated-noise or continuum-subtraction systematics, the 'first detection' and 'deficit' claims should be explicitly downgraded to tentative. The quoted L[CII]/LFIR ratio also needs a propagated uncertainty.","section":"Section 3, Table 2, Section 5.3"},{"comment":"The abstract and Conclusions state that the radio fluxes of both hosts are contaminated by long-lived afterglows, but Section 5.1.2 concludes for GRB 080207 that the radio excess is probably due to an overestimated radio SFR conversion factor, 'rather than the afterglow/AGN contamination.' These statements are contradictory. Because the paper's central message includes the origin of the radio discrepancy, the authors should harmonize the text and state explicitly which host(s) support the afterglow-contamination scenario.","section":"Section 5.1.2 vs Abstract/Conclusions"},{"comment":"For GRB 070521, the SED fit is anchored by a single rest-frame FIR photometric point, and the authors acknowledge in Section 4 that the MAGPHYS dust parameters are likely more uncertain than quoted. The derived SFR 49.85(+72.33,-2.86) M_sun/yr and its placement on the main sequence in Section 5.2 inherit this limitation. The paper should either quantify how the SFR changes under alternative dust-temperature or emissivity assumptions, or explicitly label the 070521 SFR as less secure than the formal MAGPHYS error bars, rather than presenting it in the abstract as a 'reliable' FIR-based SFR.","section":"Section 4, Table 5"}],"minor_comments":[{"comment":"The FIR flux values are swapped: the GRB 070521 continuum flux is 3.58 ± 0.45 mJy and the GRB 080207 band 9 flux is 11.9 ± 1.7 mJy (Table 2). The sentences citing 11.9 mJy in Section 5.1.1 and 3.58 mJy in Section 5.1.2 should be reversed; the conclusion that afterglow contamination is negligible is unchanged, but the text is internally inconsistent.","section":"Sections 5.1.1 and 5.1.2"},{"comment":"The ratio L[CII]/LFIR = 7.5e-4 is quoted without an uncertainty; it should be computed by propagating the errors on L[CII] and LFIR.","section":"Section 5.3.1, Table 5"},{"comment":"The phrase 'one of the smallest values' should be supported by a quantitative rank or percentile among the comparison sample shown in Fig. 5, rather than left as a visual impression.","section":"Section 5.3.1"},{"comment":"There are minor typographical issues: 'an useful' in the abstract should be 'a useful', and 'the the GRB 980425 host' appears in Section 5.3.2.","section":"Abstract and Section 5.3.2"}],"recommendation":"major_revision","confidential_remarks":"The FIR continuum results and the revised SFRs are solid and publishable after revision; the main risk is the [C ii] deficit claim, which should not be presented as secure in its current form. Please also check that the overlap with Hatsukade et al. (2019), cited as an arXiv e-print, is clearly delineated, since that paper already analyzes some of the same host galaxies with ALMA data."},"author_rebuttal":null,"desk_editor":{"model":"deepseek-v4-flash","letter":"Colleague — this one is worth a look for the continuum results, not the [C II] headline.\n\nWhat's actually new: ALMA band 8/9 continuum detections for GRB 070521 (first FIR detection) and GRB 080207 (new data covering the FIR peak). The MAGPHYS SED fits give rest-frame FIR SFRs of ~50 and ~123 M_sun/yr, placing both hosts on the z~2 main sequence and an order of magnitude below the old radio-based rates. That is a clean, useful result. The continuum detections are high-S/N (7–19), the SED fitting uses a standard public code, and the authors are careful about which photometry they exclude to avoid afterglow contamination. The main-sequence conclusion survives even if everything else in the paper is ignored.\n\nThe soft spots are exactly where the reader put them. The \"[C II] deficit\" claim rests on one 8-minute band 9 observation with the line at S/N~4 after continuum subtraction and a ±125 km/s integration window. The authors call it marginal themselves. The ratio L[CII]/LFIR = 7.5e-4 has no propagated uncertainty, and the \"one of the smallest\" claim is vulnerable to selection effects — the paper actually concedes this in §5.3.4. So treat the deficit as tentative, not established.\n\nTwo internal issues: the abstract attributes the radio excess to afterglow contamination, while §5.1.2 argues for a conversion-factor problem for 080207; and the continuum fluxes 11.9 and 3.58 mJy appear swapped between §5.1.1 and §5.1.2 — the 070521 host is not an 11.9 mJy source. These are presentation slips, not load-bearing flaws. Also, GRB 070521 has only one FIR photometric point; MAGPHYS dust parameters and the SFR uncertainties are probably understated, as the authors acknowledge.\n\nWho this is for: anyone working on GRB hosts, obscured star formation at z~2, or the [C II] deficit as a diagnostic. The FIR SFRs are a solid step. The deficit claim needs independent confirmation before it becomes part of the lore.\n\nI'd send this to a serious referee. The continuum result is valuable on its own, and the marginal line is worth refereeing rather than desk-rejecting, provided the referee is asked to check the flux accounting and the uncertainty propagation. My own verdict: accept after minor revision, with the [C II] language softened and the internal inconsistencies fixed.","headline":"Solid FIR-based SFR update for two z~2 GRB hosts; the headline [C II] deficit rests on a S/N~4 line and should be treated as tentative.","tokens_in":24969,"tokens_out":1868,"would_cite":true,"duration_ms":18241,"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":"ALMA far-infrared photometry places two z~2 GRB hosts on the galaxy main sequence and reports the first [C II] 158 micron detection in a GRB host at z>2.","keywords":["gamma-ray bursts","GRB host galaxies","star formation rate","far-infrared continuum","[C II] 158 micron","main sequence","redshift ~2","dust-obscured star formation"],"falsifier":"A longer ALMA observation of the GRB 080207 host reaching S/N > 8 in [C II] 158 micron would settle the matter: reproducing the line flux of $4.4\\,\\mathrm{Jy\\,km\\,s^{-1}}$ at the same velocity offset would confirm the deficit, while a non-detection or a much weaker line would refute it.","tokens_in":23760,"feed_emoji":"🔭","tokens_out":9939,"duration_ms":91266,"temperature":0.7,"pith_summary":"This paper reports millimetre-wavelength observations of two galaxies that hosted long gamma-ray bursts at redshift ~2. Using rest-frame far-infrared emission, which traces dust-obscured star formation without being dimmed by dust, it derives star-formation rates of $49.85^{+72.33}_{-2.86}$ and $123.4^{+25.19}_{-21.78}\\,M_\\odot\\,\\mathrm{yr}^{-1}$ for the GRB 070521 and 080207 hosts. These values put the galaxies on the \"main sequence\" of normal star-forming galaxies at $z\\sim2$, far below radio-based estimates of about $800\\,M_\\odot\\,\\mathrm{yr}^{-1}$. The paper also reports a marginal detection of the [C II] 158 micron cooling line from the 080207 host, giving a [C II]-to-FIR luminosity ratio of $7.5\\times10^{-4}$, among the lowest known at $z\\sim1$-$2$ for similar FIR luminosity. If this picture holds, high-redshift GRB hosts can be ordinary star-forming galaxies rather than extreme starbursts, and the [C II] deficit may become a new way to characterise them.","feed_headline":"Two GRB hosts are ordinary star formers, not starbursts","feed_subtitle":"Far-infrared light shows these hosts are normal star formers and flags a faint [C II] signal.","key_machinery":"The argument is carried by rest-frame far-infrared photometry that covers the peak of the dust emission, in combination with an energy-balance SED-fitting model that re-distributes energy absorbed by dust in the UV/optical into mid- and far-infrared dust templates. This yields dust-unbiased obscured star-formation rates. The second device is the [C II] 158 micron fine-structure line, the main coolant of neutral gas: after continuum subtraction, its velocity-integrated flux defines $L_{\\rm [C\\,II]}$, and the ratio $L_{\\rm [C\\,II]}/L_{\\rm FIR}$, plotted against FIR luminosity and FIR surface density, locates the host on the known [C II] deficit trend. The paper also uses the observed ratio of [N II] 205 micron to [C II] to set a metallicity upper limit of about one solar metallicity for the host.","core_discovery":"The central claim is that rest-frame far-infrared continuum detections of the GRB 070521 and 080207 host galaxies provide reliable obscured star-formation rates: $49.85^{+72.33}_{-2.86}$ and $123.4^{+25.19}_{-21.78}\\,M_\\odot\\,\\mathrm{yr}^{-1}$. These SED-based rates agree with optical estimates, place both systems on the mass--SFR main sequence at $z\\sim2$, and are far below the radio-derived rates of about 817 and 846 $M_\\odot\\,\\mathrm{yr}^{-1}$, which the paper argues are inflated by long-lived radio afterglows or by the radio SFR calibration. In addition, a marginal [C II] 158 micron line detection (S/N~4) in the 080207 host, the first for a GRB host at $z>2$, yields $L_{\\rm [C\\,II]}/L_{\\rm FIR}=7.5\\times10^{-4}$. That ratio is one of the smallest among galaxies at $z\\sim1$-$2$ with the same FIR luminosity, so the host shows a [C II] deficit; the paper identifies metallicity, a top-heavy IMF, and high gas density as possible drivers.","pith_inferences":["If these two hosts are representative of the $z\\sim2$ GRB population, GRB event rates may trace ordinary main-sequence star-forming galaxies, which would strengthen the use of long GRBs as tracers of cosmic star-formation history.","Because deep ALMA observations favor the detection of faint [C II] emitters, the low $L_{\\rm [C\\,II]}/L_{\\rm FIR}$ ratio found here may turn out to be common among $z\\sim2$ galaxies; a complete survey of [C II] in main-sequence galaxies at the same depth could show whether the \"deficit\" is specific to GRB hosts.","The relative roles of metallicity, IMF, and gas density could be separated by measuring [N II] 205 micron and CO line ratios in a larger sample of GRB hosts; the paper provides only an upper limit for one host."],"forward_implications":["The host galaxies of GRB 070521 and 080207 are consistent with the main sequence of star-forming galaxies at $z\\sim2$, so GRB hosts at this redshift do not have to be extreme starbursts.","Radio-based star-formation rates for these hosts are overestimated by factors of about 7 to 16, meaning afterglow contamination must be accounted for in GRB host studies that rely on radio continuum.","Rest-frame far-infrared observations should become a standard tool for measuring star-formation rates of GRB hosts at $z\\sim1$-$2$, because they avoid dust extinction and afterglow contamination.","The [C II] deficit, quantified by $L_{\\rm [C\\,II]}/L_{\\rm FIR}=7.5\\times10^{-4}$, is a candidate new dust-extinction-free diagnostic of GRB host environments at $z\\sim1$-$2$."],"supporting_citations":[{"why":"Supplies the radio-based SFRs and radio fluxes that the paper argues are contaminated by afterglows or inflated by calibration.","marker":"Perley & Perley 2013"},{"why":"Provides the multi-wavelength UV-to-NIR host photometry and SED-based stellar masses and SFRs used in the fits.","marker":"Perley et al. 2013"},{"why":"Supplies the H-alpha SFRs and spectroscopic redshifts for both GRB hosts.","marker":"Krühler et al. 2015"},{"why":"Provides the earlier Herschel/PACS far-infrared photometry and infrared SFR estimate for the GRB 080207 host.","marker":"Hunt et al. 2014"},{"why":"Supplies the CO line luminosities, gas mass, and dust-temperature constraints used to characterise the GRB 080207 host.","marker":"Hatsukade et al. 2019"},{"why":"Supplies the energy-balance SED-fitting model used to derive SFRs, stellar masses, dust masses, and FIR luminosities.","marker":"da Cunha et al. 2008, 2015"},{"why":"Supplies the conversion factor $\\log(L_{\\rm TIR}/L_{\\rm FIR})=0.24$ used to derive FIR luminosities.","marker":"Calzetti et al. 2000"},{"why":"Supplies the standard afterglow spectral slopes used to argue that long-lived radio afterglows contribute negligibly at the observed FIR wavelengths.","marker":"Sari et al. 1998"},{"why":"Supplies the mass--SFR main-sequence relation at $z\\sim2$ against which the GRB hosts are placed.","marker":"Whitaker et al. 2014"},{"why":"Supplies the local and high-redshift comparison samples and the $L_{\\rm [C\\,II]}/L_{\\rm FIR}$ versus $\\Sigma_{\\rm FIR}$ trend that defines the [C II] deficit.","marker":"Díaz-Santos et al. 2013, 2017"}],"fun_headline_variants":["First [CII] detection in a GRB host at z>2 reveals deficit","Radio overestimated GRB host star formation; ALMA shows normal rates","GRB hosts at z~2 are main-sequence galaxies, not starbursts","Faint [CII] in a GRB host at z>2 signals a deficit","ALMA finds GRB hosts ordinary, not the extreme starbursts radio suggested"],"cache_read_input_tokens":3200,"weakest_assumption_plain":"The load-bearing premise is that the marginal [C II] 158 micron detection in the GRB 080207 host is real; if the S/N~4 line is dominated by noise or by continuum-subtraction systematics, the claimed [C II] deficit collapses.","fun_headline_variants_meta":{"raw":{"variants":["First [CII] detection in a GRB host at z>2 reveals deficit","Radio overestimated GRB host star formation; ALMA shows normal rates","GRB hosts at z~2 are main-sequence galaxies, not starbursts","Faint [CII] in a GRB host at z>2 signals a deficit","ALMA finds GRB hosts ordinary, not the extreme starbursts radio suggested"]},"model":"deepseek-v4-flash","effort":"low","cost_usd":0.001389,"raw_usage":{"total_tokens":5767,"prompt_tokens":1239,"completion_tokens":4528,"prompt_tokens_details":{"cached_tokens":384},"prompt_cache_hit_tokens":384,"prompt_cache_miss_tokens":855,"completion_tokens_details":{"reasoning_tokens":4419}},"tokens_in":855,"tokens_out":4528,"duration_ms":31266,"temperature":1.0,"reasoning_tokens":4419,"cache_read_input_tokens":384,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-08-14T15:10:49.313816+00:00","model_set":{"reader":"deepseek-v4-flash"},"falsifier":"A longer ALMA observation of the GRB 080207 host reaching S/N > 8 in [C II] 158 micron would settle the matter: reproducing the line flux of $4.4\\,\\mathrm{Jy\\,km\\,s^{-1}}$ at the same velocity offset would confirm the deficit, while a non-detection or a much weaker line would refute it.","supporting_citations":[],"review_version":1}