{"id":"7031eafe-d72b-4318-94f3-8c0f74685109","arxiv_id":"2411.12693","paper_version":2,"verdict":"CONDITIONAL","confidence":"MODERATE","novelty_score":6.0,"correctness_risk":"medium","formal_verification":"none","parameter_count":2,"one_line_summary":"New NOEMA and ALMA-ACA observations of six Orion cores confirm that dust opacity indices from 2.9 to 3.6 mm are much shallower (beta around -0.16 to 1.45) than the beta > 1.3 values measured on larger scales.","lead":"The paper measures the millimeter brightness of six young star-forming cores in the Orion Molecular Cloud with two radio interferometers and finds that the dust emission stays unexpectedly bright at long wavelengths, with shallower opacity indices than standard models predict. It confirms earlier hints of a flattened dust spectrum beyond 2 mm and suggests that embedded disks may contaminate core-scale measurements.","discovery_kind":"extension","skeptic_critique":{"model":"deepseek-v4-flash","headline":"The NOEMA-only β values may be biased by an unquantified frequency-dependent amplitude calibration error, since a 10% relative flux error across the 20 GHz band changes α by ~0.45 and Section 2.1 quotes up to 21.6% amplitude loss.","rationale":"The reader's weakest assumption concerned spatial-filtering matching between ALMA and NOEMA, which mostly affects the secondary comparison of slopes, not the primary NOEMA-only flattening claim. The frequency-dependent amplitude calibration issue is more load-bearing because a realistic 10% relative error across the 20 GHz NOEMA band changes α by ~0.45, directly reshaping the reported β values and their comparison to single-dish β>1.3. The paper's own calibration uncertainty estimates in Section 2.1 make this concern concrete, and no cross-check of the relative bandpass is provided. Nevertheless, the qualitative flattening is supported by prior single-dish work and by the large persistent excess at 3 mm, so the concern is a condition to be resolved rather than a demonstrated fatal flaw. The verdict should remain CONDITIONAL, but with the added requirement of a quantitative calibration test before the measured β values are taken at face value.","tokens_in":25184,"tokens_out":18428,"duration_ms":180011,"concrete_test":"Use the 3C84 and phase-calibrator observations from the same track to derive the relative bandpass response in all four NOEMA basebands. Compare the measured flux ratios to the adopted calibrator spectral index; if they deviate by more than 5%, simulate the expected frequency-dependent amplitude loss as a function of baseline and time, correct the target visibilities, and re-fit the NOEMA slopes. If corrected β values shift by more than 0.3 from Table 4, the flattening claim needs to be weakened accordingly.","verdict_should_be":"UNCHANGED","load_bearing_attack":"The central claim of flattened opacity indices at 2.9-3.6 mm rests entirely on NOEMA slopes measured over 82.7-102.3 GHz. Section 2.1 quotes calibration uncertainties up to 21.6% amplitude loss, <30% pointing error, and <30% focus error. If these losses are not strictly frequency-independent, the ratio of fluxes between the 102.3 GHz and 82.7 GHz basebands is systematically biased. Phase decorrelation from atmospheric noise scales roughly as frequency, and pointing errors are more damaging at the higher-frequency, smaller-beam basebands. A relative amplitude error of 10% across the band changes α by about 0.45, comparable to or larger than the reported statistical uncertainties on β. The statement in Section 3.2 that simultaneous flux calibration makes NOEMA slopes insensitive to calibration errors is valid only for a constant gain error, and no test for frequency dependence is presented. This is the most load-bearing unvalidated assumption because it directly affects the measured β values, not just the ALMA/NOEMA comparison.","agreement_with_reader":"partial"},"referee_report":{"model":"deepseek-v4-flash","summary":"The paper presents NOEMA 2.9-3.6 mm and ALMA-ACA Band 4/5 (1.6-2.2 mm) continuum observations of six protostellar cores in the OMC 2/3 filament. The authors fit power-law SEDs separately to the NOEMA and ALMA data, derive dust opacity indices beta = alpha - 2, and compare them with single-dish modified-blackbody beta values from Sadavoy et al. (2016). They report flattened beta values between 2.9 and 3.6 mm (beta approximately -0.16 to 1.45), agreement between ALMA and NOEMA slopes for four sources, and disagreement for FIR2 and MMS6. They discuss free-free emission, anomalous microwave emission, disk contamination, and unusual dust properties as possible explanations, and conclude that disk contamination may be significant for some sources and that multi-scale SED analyses require caution.","tokens_in":25412,"tokens_out":7743,"duration_ms":70875,"significance":"If the flattened opacity indices are real, the result is significant: it would confirm that the dust SED in OMC 2/3 cores cannot be described by a single power law from 0.16 to 3.6 mm, with implications for mass estimates and dust grain models. The paper's strengths include the use of independent NOEMA and ALMA-ACA datasets, explicit line-channel flagging, a uv-cut to isolate comparable spatial scales, source flux extraction with background fitting, a Monte Carlo treatment of the ALMA 10% calibration uncertainty, and a direct comparison with GBT/MUSTANG data to test line contamination and missing short-spacing flux. These are appropriate steps that go beyond earlier single-dish studies. However, the central claim rests on the NOEMA-only slopes, and two methodological issues - differing resolution across the NOEMA basebands and unquantified frequency-dependent calibration errors - need to be addressed before the flattened beta values can be considered secure.","major_comments":[{"comment":"The four NOEMA basebands are smoothed to different final beams rather than to a common resolution. Table 2 gives final beam sizes of 14.2x8.8 arcsec at 82.7 GHz, 13.2x8.4 arcsec at 86.8 GHz, 11.4x7.9 arcsec at 98.2 GHz, and 10.7x7.6 arcsec at 102.3 GHz. Because several targets show extended emission (e.g., MMS9 and FIR2, Section 4 and Figures 1-2), the lower-frequency points may include more extended flux than the higher-frequency points, which would flatten the NOEMA-only slope in exactly the direction claimed. The paper should smooth all NOEMA basebands to a common beam (e.g., the largest synthesized beam) and re-fit the slopes.","section":"Section 3.1, Table 2"},{"comment":"The NOEMA calibration uncertainties are not propagated into the slope errors. Section 2.1 quotes up to 21.6% amplitude loss, <30% pointing error, and <30% focus error, and states that a spectral index of -0.38 was adopted for the RF calibrator. Section 3.2 claims that because the NOEMA data were taken simultaneously with the same flux calibration, the flux calibration errors will not affect the slope; this holds only for a constant gain error across the band. Any frequency dependence in the amplitude loss or an error in the adopted calibrator spectral index changes the relative flux between the 82.7 and 102.3 GHz basebands. A 10% relative amplitude error changes alpha by about 0.45, which is comparable to the reported 1-sigma uncertainties in Table 4. The authors should quantify this effect, for example with a Monte Carlo that draws per-baseband amplitude gains and by testing the sensitivity to the RF calibrator spectral index, before claiming the flattened beta values are robust.","section":"Section 2.1, Section 3.2"},{"comment":"The comparison between ALMA and NOEMA slopes uses heterogeneous error bars. The ALMA slopes in Table 4 include the Monte Carlo calibration uncertainty, while the NOEMA slopes are fit-only errors from emcee (Section 3.2). Therefore, the 'consistent within 1-sigma' classification for FIR6B, MMS7, MMS9, and NW167 does not include NOEMA calibration systematics. For MMS7 the slopes differ by only 0.41 with quoted errors of 0.39 and 0.42; adding a plausible calibration uncertainty of order 0.45 to the NOEMA slope would make the agreement test inconclusive. Similarly, the two 'discrepant' sources (FIR2 and MMS6) could change classification. The paper should report a single error budget for each slope that includes both fit and calibration terms, or explicitly state which error bars are used in the comparison.","section":"Section 4, Table 4"},{"comment":"The statement that the ALMA and NOEMA data 'should be consistently tracing emission from the envelope and core over the same spatial scales' is not demonstrated. The uv>5 klambda cut and beam smoothing are necessary, but the two datasets have different baseline distributions and the paper does not show the overlapping uv coverage. A quantitative test, such as fitting the ALMA visibilities at the same uv range as NOEMA or re-imaging with different uv cuts, would show whether residual spatial-filtering differences can explain the ALMA/NOEMA slope discrepancies. Without this, the conclusion that FIR2 and MMS6 trace different emission components is not fully supported.","section":"Section 3.1"}],"minor_comments":[{"comment":"The phrase 'and < in 30% focus error' should read 'and <30% in focus error'.","section":"Section 2.1"},{"comment":"The phrase 'we used a random selector to generate two sets of 5000 samples' should say 'a random number generator' or similar.","section":"Section 3.2"},{"comment":"The sentence 'We find that only for FIR2 (α = 2− 3) and MMS6 (α = 3, disk contamination is likely minimal' is missing a closing parenthesis after the α = 3 and should be rephrased for clarity.","section":"Section 6.4"},{"comment":"The table note refers to 'β-21 values' but should be 'β-B21 values' to match the citation to Bouvier et al. (2021).","section":"Table 5"},{"comment":"The phrase 'This result is not unsurprising' is a double negative and should be revised.","section":"Section 6.1"},{"comment":"The source is sometimes abbreviated 'MM6' instead of 'MMS6'; please use the same abbreviation throughout.","section":"Section 5.3"}],"recommendation":"major_revision","confidential_remarks":"The paper addresses an interesting and timely question, and the authors have assembled a valuable dataset. However, the abstract's central claim of flattened opacity indices at 2.9-3.6 mm rests entirely on the NOEMA slopes, which currently have two unaddressed systematics: differing resolution across the four NOEMA basebands and unquantified frequency-dependent calibration errors. Both are fixable with re-analysis, but they are load-bearing. I would encourage the editor to request a revision that addresses these points before considering the paper for publication."},"author_rebuttal":null,"desk_editor":{"model":"deepseek-v4-flash","letter":"Here's my take. The paper confirms with new interferometric data what the single-dish work already suggested: the SEDs of these six OMC 2/3 cores are flatter between 2.9 and 3.6 mm than a single power-law extrapolation from shorter wavelengths. The measurements are real contributions: NOEMA basebands imaged separately, ALMA-ACA Band 4/5 data cut to a similar uv range, per-source beta values with uncertainties, and a direct ALMA/NOEMA slope comparison for six targets. The authors also check their power-law fitting against a modified blackbody and find only a 10-15% bias in beta, and they show the NOEMA fluxes agree with GBT/MUSTANG, which argues against line contamination or short-spacing problems. That is honest, useful work.\n\nThe main soft spot is the one the stress-test flags. The NOEMA slopes depend on calibration errors being frequency-independent across the 20 GHz band. The paper quotes up to 21.6% amplitude loss, <30% pointing errors, and <30% focus errors, and the claim that simultaneous flux calibration makes the slopes immune is only valid for a constant gain error. A 10% relative error across the band changes alpha by about 0.45, which is comparable to the reported statistical errors. No test for frequency dependence is presented. This needs to be addressed before I would trust the exact beta values. It is not fatal to the central claim, because the ALMA slopes independently show low beta for most sources, and the flattening at lambda>2 mm is already seen in single-dish data. But the paper should either quantify the frequency-dependent component or soften the \"confirm\" language.\n\nThe spatial filtering match between ALMA and NOEMA is also not quantitatively validated; the uv cut and beam smoothing are reasonable, but residual differences in missing short-spacing or background subtraction could affect the slope comparison. The sample is selected from cores known to have elevated 3 mm emission, so the result shows the flattening exists in these cases, not that it is common. The disk contamination estimate depends on assumed alpha=2 or 3, which the authors acknowledge.\n\nOverall, this is a competent, honest observational paper that strengthens the case for a non-standard dust opacity or an extra emission component at long wavelengths. I'd send it to peer review. A good referee will push on the NOEMA calibration and the spatial filtering, but the data and analysis deserve publication with revisions. If you work on dust SEDs or star formation mass estimates, this is worth your time. I'd bring it to a reading group, and I'd cite it.","headline":"Solid new data confirm flattened long-wavelength dust opacity in Orion cores; the NOEMA slope calibration needs scrutiny, but the result holds up.","tokens_in":26038,"tokens_out":3838,"would_cite":true,"duration_ms":35088,"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":"Millimeter observations of six Orion protostellar cores confirm that the 3 mm dust emission is too bright for a single opacity power law.","keywords":["dust opacity index","spectral energy distribution","Orion Molecular Cloud","protostellar cores","millimeter interferometry","disk contamination","dust grain growth"],"falsifier":"A matched-resolution, matched-uv-coverage comparison of single-dish and interferometric data at 2.9-3.6 mm that included all short spacings and recovered beta > 1.3 would falsify the flattening claim; alternatively, resolving the embedded disks in FIR6B and MMS7 at 3 mm and finding their flux below the extrapolated disk contribution would falsify the proposed disk-contamination explanation.","tokens_in":24986,"feed_emoji":"🌌","tokens_out":5701,"duration_ms":52143,"temperature":0.7,"pith_summary":"The paper targets a puzzle in the Orion Molecular Cloud: dust emission at wavelengths beyond 2 mm is brighter than a single power-law opacity law predicts from shorter wavelengths. Using interferometric measurements of six protostellar cores from 1.7 to 3.6 mm, matched to the same spatial scales of about 0.02-0.08 pc, it confirms that the opacity index beta between 2.9 and 3.6 mm is flat (beta approximately -0.16 to 1.45), well below the beta > 1.3 inferred from single-dish data on 0.08 pc scales. If right, a single dust-opacity power law cannot describe these cores across 1.6-3.6 mm, so masses and dust properties derived from extrapolating one band would be biased. The paper argues that in four cores the long-wavelength excess may come largely from large grains in embedded disks, while two cores show different behavior.","feed_headline":"Orion dust shines too bright at 3 mm for one power law","feed_subtitle":"Six protostellar cores show flat opacity slopes from 2.9 to 3.6 mm, below single-dish values.","key_machinery":"The load-bearing object is the spectral index $\\alpha$ defined by S_nu proportional to nu^$\\alpha$ on the Rayleigh-Jeans tail, with opacity index $\\beta$ = $\\alpha$ - 2. The argument works by measuring $\\alpha$ separately from interferometric data at 83-102 GHz (2.9-3.6 mm) and 137-177 GHz (1.6-2.2 mm), after cutting visibilities at uv > 5 kilolambda and smoothing beams so that both datasets trace 0.02-0.08 pc scales, then comparing with modified-blackbody fits to single-dish data from 0.16 to 2 mm.","core_discovery":"On the paper's own terms, the discovery is that the 2.9-3.6 mm continuum of six OMC 2/3 protostellar cores is systematically elevated relative to a modified blackbody fitted to single-dish data from 0.16 to 2 mm, giving power-law opacity indices beta between approximately -0.16 and 1.45. Four of the six sources have interferometric spectral slopes consistent within one sigma across the two wavelength regimes, indicating a common emission mechanism from 1.7 to 3.6 mm; the other two (FIR2 and MMS6) have slopes differing by more than two sigma. The paper proposes that embedded disks with large grains can bias longer-wavelength fluxes for the consistent sources, while free-free emission and anomalous microwave emission are insufficient to explain the flattening. The conclusion is that combining multi-scale observations or extrapolating single-band observations requires care.","pith_inferences":["Inference: if disk contamination is as large as suggested for FIR6B and MMS7, envelope masses and column densities estimated from 3 mm continuum in similar protostars may be systematically overestimated.","Inference: the broken-power-law interpretation predicts that higher-resolution observations that resolve out the disks would recover steeper envelope slopes; this is testable at roughly 0.01 pc resolution.","Inference: the same flat-slope signature seen elsewhere in Orion and Serpens suggests the disk-contamination bias may affect cloud-wide surveys, not just individual cores."],"forward_implications":["If the flattened slopes are real, dust opacity cannot be a single power law across 1.6-3.6 mm for these cores.","Four of the six cores have matching interferometric slopes, so their long-wavelength excess can be explained by a common emission component, likely embedded disks with large grains.","Disk contamination can account for up to roughly 70 percent of the 3 mm flux in FIR6B and MMS7, meaning protostellar disk mass can bias core-scale measurements.","Free-free emission and anomalous microwave emission cannot explain the flattening.","Extrapolating single-band observations to other wavelengths is unsafe at these scales."],"supporting_citations":[{"why":"Provides the earlier low-beta measurements at 3 mm and 1 mm on 0.02-0.1 pc scales that this paper extends and tests.","marker":"Schnee et al. (2014)"},{"why":"Supplies the single-dish SED fits (Herschel and IRAM 0.16-2 mm) whose beta values the interferometric slopes are compared against.","marker":"Sadavoy et al. (2016)"},{"why":"Confirmed the flattened spectral index at wavelengths longer than 2 mm in OMC 2/3 and ruled out free-free and anomalous microwave emission as the dominant cause.","marker":"Mason et al. (2020)"},{"why":"Provides the high-resolution multi-wavelength measurements of MMS6 that the paper compares with to identify scale-dependent slope changes.","marker":"Takahashi et al. (2009)"},{"why":"Supplies the 0.87 mm disk fluxes and disk masses used to estimate the fraction of long-wavelength emission that could come from embedded disks.","marker":"Tobin et al. (2020)"},{"why":"Quantifies free-free contamination at 32.9 GHz for FIR2 and MMS9, which the paper uses to assess whether free-free can explain the flat NOEMA slopes.","marker":"Bouvier et al. (2021)"},{"why":"Shows that similar flattened slopes appear across Orion A, Orion B, and Serpens, placing the OMC 2/3 result in a broader context.","marker":"Lowe et al. (2022)"}],"fun_headline_variants":["Six Orion cores flatten dust opacity at millimeter waves","Dust emission in Orion cores resists one-size-fits-all power law","Flat dust slopes in OMC 2/3 from 2.9 to 3.6 mm","Orion protostars defy single dust opacity index"],"cache_read_input_tokens":3200,"weakest_assumption_plain":"The analysis assumes that, after matching spatial filtering and resolution, the interferometric and single-dish measurements trace the same dust emission components and that a single power law describes the spectrum across 1.7-3.6 mm.","fun_headline_variants_meta":{"raw":{"variants":["Six Orion cores flatten dust opacity at millimeter waves","Dust emission in Orion cores resists one-size-fits-all power law","Flat dust slopes in OMC 2/3 from 2.9 to 3.6 mm","Orion protostars defy single dust opacity index"]},"model":"deepseek-v4-flash","effort":"low","cost_usd":0.000231,"raw_usage":{"total_tokens":1551,"prompt_tokens":1076,"completion_tokens":475,"prompt_tokens_details":{"cached_tokens":384},"prompt_cache_hit_tokens":384,"prompt_cache_miss_tokens":692,"completion_tokens_details":{"reasoning_tokens":397}},"tokens_in":692,"tokens_out":475,"duration_ms":5246,"temperature":1.0,"reasoning_tokens":397,"cache_read_input_tokens":384,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-08-12T17:14:41.256189+00:00","model_set":{"reader":"deepseek-v4-flash"},"falsifier":"A matched-resolution, matched-uv-coverage comparison of single-dish and interferometric data at 2.9-3.6 mm that included all short spacings and recovered beta > 1.3 would falsify the flattening claim; alternatively, resolving the embedded disks in FIR6B and MMS7 at 3 mm and finding their flux below the extrapolated disk contribution would falsify the proposed disk-contamination explanation.","supporting_citations":[{"cited_title":null,"cited_arxiv_id":null,"evidence_quote":"Provides the high-resolution multi-wavelength measurements of MMS6 that the paper compares with to identify scale-dependent slope changes."},{"cited_title":"2022, ApJ, 929, 102, doi: 10.3847/1538-4357/ac5d4f 18 N ozari et al","cited_arxiv_id":null,"evidence_quote":"Shows that similar flattened slopes appear across Orion A, Orion B, and Serpens, placing the OMC 2/3 result in a broader context."}],"review_version":1}