{"id":"61a86fb2-8213-4993-9b09-78cbe1ad24cb","arxiv_id":"2412.17901","paper_version":1,"verdict":"CONDITIONAL","confidence":"MODERATE","novelty_score":6.0,"correctness_risk":"medium","formal_verification":"none","parameter_count":4,"one_line_summary":"Nine cores in the 70 μm dark clump G337.541-00.082 show infall velocities of 0.28-1.45 km/s and infall rates of 10^-4 to 10^-3 Msun/yr, higher than in low-mass regions and consistent with core growth toward high-mass star formation.","lead":"Using ALMA observations of a dark, massive cloud 4 kpc away, this paper finds gas falling into dense cores at speeds of 0.3 to 1.5 km/s, with inflow rates up to a thousand times higher than in low-mass star-forming regions. The results support the idea that low- and intermediate-mass cores grow into high-mass stars by accreting gas from their surroundings.","discovery_kind":"new_application","skeptic_critique":{"model":"deepseek-v4-flash","headline":"Hill5 infall velocities are not uniquely determined: the fixed T0=2.3 K slab model is not checked against static temperature-gradient or velocity-gradient models, so the factor-of-five velocity and order-of-magnitude infall-rate claims remain conditional.","rationale":"The paper's central claim is quantitative: infall velocities of 0.28-1.45 km/s and mass infall rates of 10^-4 to 10^-3 Msun/yr, exceeding low-mass values by more than a factor of five and an order of magnitude, respectively, thereby supporting clump-fed core growth. The paper has real supporting evidence that some infall is present: the self-absorption dip at the DCO+ peak, the virial parameters below unity, the position-velocity gradients around cores, and the explicit outflow-component removal for ALMA2 and ALMA18. However, the quantitative headline rests on a model assumption that is not independently verified for the full sample. Hill5 fixes the edge excitation temperature at 2.3 K, so the inferred v_in is conditional on a temperature structure that is not measured. In a 70-micron-dark clump with a 12 K dust temperature, a static core with a modest excitation-temperature gradient can produce a blue-bright double-peaked profile without collapse. The documented clump-scale and core-scale N2H+ velocity gradients add a second plausible alternative, since spatially averaged optically thick emission over a velocity gradient can mimic asymmetry. The MCMC error bars in Table 2 capture only statistical noise, not these systematic model uncertainties. Therefore the reader's weakest-assumption identification is correct, and the CONDITIONAL verdict is appropriate: the quantitative claims should not be accepted until the Hill5 assumption is tested against a static temperature-gradient or velocity-gradient model. The proposed concrete test would settle whether this concern actually lands, which is why the verdict should remain unchanged rather than being strengthened or relaxed.","tokens_in":20804,"tokens_out":6933,"duration_ms":76865,"concrete_test":"Pick ALMA1 (or ALMA4) and run two blinded fits of its observed HNC spectrum: (i) the published Hill5 model, and (ii) a static v_in=0 two-layer or 3D radiative-transfer model in which the excitation temperature is a free linear gradient with edge temperature allowed to vary over 2-12 K, consistent with the 12 K dust temperature, and with the observed N2H+ position-velocity gradient added as a large-scale shear. Compare Bayesian evidences or chi-square per degree of freedom; if the static model reproduces the blue asymmetry within the noise, the derived v_in cannot be uniquely attributed to infall and the quantitative claims need revision, whereas if no static model can match, the infall interpretation is strengthened. A cheaper auxiliary check is to rerun the Hill5 fits with T0 as a free parameter and report the posterior on v_in conditioned on T0 in the 2-12 K range.","verdict_should_be":"UNCHANGED","load_bearing_attack":"The quantitative headline depends on the Hill5 fits in Section 4.2, which assume that the HNC (3-2) blue asymmetry is produced by collapse in a two-layer slab with the edge excitation temperature fixed at T0=2.3 K. The paper itself notes in Section 4.2 that the asymmetry signature is viewing-angle dependent for non-spherical cores and that Hill5 'may underestimate the infall velocity in some cases', while its reliability 'improves when the line profile exhibits separate blue- and red-shifted components' - yet the abstract's 0.28-1.45 km/s range includes cores such as ALMA1, ALMA4, and ALMA5 with comparable red and blue peak brightnesses. The same double-peaked, blue-bright profile can also arise from a temperature gradient (near side warmer than far side) or from unresolved spatial/velocity structure, including the N2H+ velocity gradients reported in Section 4.1 and the outflow entrainment documented in Section 3.1. Because HNC (3-2) is optically thick with Eu/k about 26 K, the inferred infall velocity is strongly tied to the assumed excitation-temperature structure, and the quoted +/-0.01 km/s MCMC uncertainties exclude exactly this model uncertainty. If a static-core model with a plausible temperature gradient, or a velocity-gradient-only model, fits the observed spectra as well as Hill5, the inferred infall velocities - and hence the claimed factor >5 in velocity and order-of-magnitude excess in mass infall rate relative to low-mass regions - are not yet established.","agreement_with_reader":"agree"},"referee_report":{"model":"deepseek-v4-flash","summary":"This paper presents ALMA observations (12m + ACA + TP) of the 70 micron dark massive clump G337.541-00.082, using N2H+ (J=1-0), HNC (J=3-2), and HCO+ (J=3-2). From the N2H+ position-velocity diagram the authors identify a clump-scale north-south velocity gradient and smaller-scale gradients around ALMA1, ALMA2, and ALMA3, which they interpret as infall rather than rotation. From HNC spectra they report blue-asymmetric, double-peaked profiles in nine of seventeen dust cores; these are fitted with the Hill5 two-layer radiative transfer model to obtain infall velocities of 0.28-1.45 km/s and mass infall rates of roughly 3e-5 to 2.9e-3 M_sun/yr. The paper reports a strong correlation between infall velocity and nonthermal velocity dispersion, a weaker correlation with core mass, and an anti-correlation between mass infall rate and inverse mass-weighted distance. Comparing with low-mass star-forming region samples, it argues that infall speeds and rates in G337 are several times and an order of magnitude higher, respectively, supporting clump-fed core growth as a route to high-mass star formation.","tokens_in":21154,"tokens_out":8817,"duration_ms":78974,"significance":"If the quantitative results are upheld, this is a valuable pilot study: it provides one of the first multi-core, core-scale infall measurements in an early-stage, 70 micron dark massive clump, and it strengthens the observational case for clump-fed core growth. The paper's strengths include the combination of 12m, 7m, and total-power data to recover extended emission, the use of an optically thin tracer (DCO+) to locate self-absorption dips, the identification of N2H+ emission associated with outflow entrainment, and the authors' explicit acknowledgment of several model limitations. The main scientific claims, however, depend on the fidelity of the Hill5 model for the HNC line profiles and on the statistical robustness of small-sample correlations, both of which need attention before the headline infall velocities, infall rates, and comparison with low-mass regions can be accepted.","major_comments":[{"comment":"The quantitative infall velocities and infall rates rest entirely on the Hill5 model, which assumes a two-layer slab with edge excitation temperature T0=2.3 K and a prescribed linear excitation-temperature profile. The paper does not demonstrate that a static core with a near-side/far-side temperature gradient, or a model with only velocity structure, is unable to reproduce the observed blue-bright double-peaked HNC profiles; such alternatives are known to produce similar line shapes. The authors themselves note in Section 4.2 that the infall signature is viewing-angle dependent for non-spherical cores and that Hill5 'may underestimate the infall velocity in some cases', with reliability improving only when separate blue and red components are present. Nevertheless, the headline range 0.28-1.45 km/s includes ALMA1, ALMA4, and ALMA5, whose red and blue peaks are described as having comparable brightness. The MCMC uncertainties in Table 2 (+/-0.01-0.03 km/s) quantify only statistical scatter and exclude this model uncertainty. Because the factor-of-five velocity excess and the order-of-magnitude infall-rate excess in Section 5.1 and the abstract derive from these fits, these quantitative claims are not yet established. The authors should either restrict the quantitative analysis to cores with well-separated blue and red components, or add explicit tests against alternative excitation/velocity structures, or present the values as model-dependent estimates.","section":"Section 4.2, Figure 8, Table 2"},{"comment":"There is an inconsistency in the nine-core sample. Table 2 lists ALMA6 (M=0.96 M_sun, vin=0.77 km/s), but Figure 8 shows a fitted spectrum for ALMA5 with vin=0.77 km/s, and the legend of Figure 9 includes ALMA5 and not ALMA6. Section 4.2 also names ALMA1, ALMA4, and ALMA5 as the cores with comparable red and blue peak brightness. The authors need to clarify which cores constitute the nine-core blue-asymmetry sample and correct the table, figure, and legend accordingly, because the masses, radii, and infall velocities in Table 2 enter the mass infall rates and the correlation analyses in Section 5.1.","section":"Table 2 versus Figures 8 and 9"},{"comment":"The paper calls the relation between mass infall rate and inverse mass-weighted distance a 'strong anti-correlation', but the reported Spearman coefficient is rho=-0.52 with p=0.15 for n=9, which is not statistically significant at the 95% level. Similarly, the correlation between infall velocity and core mass in Figure 9(a) is rho=0.47 with p=0.21. The abstract's statement that 'the mass infall rate is larger for larger core masses and shorter distances to the clump center' therefore overstates the evidence. These should be described as tentative trends that are consistent with, but do not strongly confirm, clump-fed scenarios.","section":"Section 5.1, Figure 10, and abstract"},{"comment":"The comparison with low-mass star-forming regions combines different molecular tracers, different telescope configurations (single-dish versus interferometric), and different angular resolutions. The paper acknowledges some of these differences but still presents the factor-of-five and order-of-magnitude excesses as robust results. Given that the Hill5-derived velocities are model-dependent and the comparison samples are heterogeneous, the quantitative excess claimed in the abstract should be softened, or the comparison should be made more controlled (for example, by applying the same fitting procedure to homogenized data).","section":"Section 5.1, Figure 11"}],"minor_comments":[{"comment":"The caption refers to 'the left panel (a)', but the figure appears to be a single panel; the red square should be identified by reference to Figure 4 instead.","section":"Figure 5 caption"},{"comment":"The statement that N2H+ tracing entrained gas is 'a first report' in high-mass star-forming regions should be softened or supported by a more thorough literature search, since similar N2H+ wing or outflow features may have been reported elsewhere.","section":"Section 3.1"},{"comment":"The definition of the inverse mass-weighted distance uses a normalization by the total mass of the fitted cores, so it is not a purely geometric distance; the text should state this explicitly when interpreting Figure 10.","section":"Equation (1)"},{"comment":"The column header for the mass infall rate is incomplete; it should read 10^-4 M_sun yr^-1, not just 10^-4 M_sun.","section":"Table 2"},{"comment":"For ALMA7 and ALMA16, the fitted velocity dispersions (0.18 and 0.17 km/s) are below the HNC channel width of 0.27 km/s; the authors should comment on the reliability of fitting sub-channel-line dispersions.","section":"Figure 8"}],"recommendation":"major_revision","confidential_remarks":"The ALMA dataset is valuable and the qualitative detection of blue asymmetry in the clearest cores is solid. My main concern is that the paper's central quantitative claims, including the factor-of-five infall velocity excess and the order-of-magnitude infall rate excess relative to low-mass regions, are not yet robust against model degeneracy in the Hill5 fits and against the small-sample statistics used for the correlations. A revision that adds alternative-model checks or appropriately softens the quantitative claims would bring the paper to the level of a strong pilot study. I also note the sample inconsistency between Table 2 and Figures 8-9, which must be fixed before the paper can be considered for publication."},"author_rebuttal":null,"desk_editor":{"model":"deepseek-v4-flash","letter":"The paper is a solid pilot study with a genuinely new sample: nine cores in a 70-micron dark clump observed with ALMA 12m+ACA+TP. It makes a real first move toward multi-core infall statistics in IRDCs at core scale. The blue-asymmetry detections in ALMA2, 7, 17, and 18 look solid, and the correlation of infall velocity with nonthermal velocity dispersion, using DCO+ as an independent optically thin tracer, is a nice result and not circular.\n\nThe soft spot is the quantitative headline. The derived infall velocities and rates depend entirely on Hill5 fits to HNC 3-2, and the stress-test concern lands: the two-layer slab fixes the edge excitation temperature at 2.3 K and is not tested against static temperature-gradient or velocity-gradient-only models. The paper itself concedes the profile is viewing-angle dependent for non-spherical cores and that Hill5 may underestimate vin, yet the abstract's 0.28-1.45 km/s range includes ALMA1, ALMA4, and ALMA5, whose red and blue peaks are comparable. Since HNC 3-2 is optically thick with Eu/k ~26 K, the inferred infall velocity is strongly tied to the assumed excitation-temperature structure. The quoted +/-0.01 km/s MCMC uncertainties exclude exactly this model uncertainty, so the factor-of-five velocity excess and order-of-magnitude infall-rate excess relative to low-mass regions are conditional, not established.\n\nTwo smaller issues. The paper calls the anti-correlation between infall rate and inverse mass-weighted distance \"strong\" with Spearman rho=-0.52 and p=0.15 for n=9. That is a trend, not a strong result, and the language should be softened. And the comparison to low-mass samples mixes telescopes, lines, and fitting setups; the authors partly address this by matching masses and line widths, but the remaining systematic differences deserve more caveats.\n\nWho is this for? Observers working on high-mass star formation and the core-growth scenario. It deserves serious refereeing. A referee should ask for a robustness check of the Hill5 assumption, for example fitting a static temperature-gradient model or explicitly quantifying the degeneracy, and for a more honest treatment of the p=0.15 correlation. The qualitative conclusion that some cores in an IRDC show infall at rates higher than in low-mass regions is probably right; the exact rates are not yet.","headline":"A solid pilot study with a real new sample, but the quantitative infall rates rest on a model assumption that isn't tested; the qualitative conclusion of core-scale infall is plausible, the exact numbers are not yet.","tokens_in":21707,"tokens_out":2099,"would_cite":true,"duration_ms":22023,"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 claims that cores inside a 70-micron-dark massive clump are accreting gas fast enough to grow from intermediate to high mass within a free-fall time, supporting the clump-fed core-growth scenario for high-mass star formation.","keywords":["infrared dark clouds","star formation","star forming regions","high-mass star formation","core growth","infall","blue asymmetry","ALMA"],"falsifier":"Take the HNC ($J=3-2$) spectra of the cores whose red and blue peaks have nearly equal brightness (ALMA1, ALMA4, ALMA5) and subtract the outflow components identified in CO and SiO at the same velocities; if the blue-brighter peak disappears or the self-absorption dip no longer lines up with the optically thin DCO$^+$ centroid velocity, then the blue asymmetry is not an infall signature.","tokens_in":20640,"feed_emoji":"🌟","tokens_out":12516,"duration_ms":100196,"temperature":0.7,"pith_summary":"The paper uses ALMA observations of the massive, 70-micron-dark clump G337.541-00.082 to measure how fast gas falls into the dense cores embedded in it. Combining a north-south velocity gradient in N$_2$H$^+$ with blue-asymmetric HNC ($J=3-2$) line profiles, the authors find infall in nine of seventeen cores, with speeds from 0.28 to 1.45 km s$^{-1}$ and mass infall rates of order $10^{-4}$ to $10^{-3}$ $M_\\odot$ yr$^{-1}$. These speeds are more than a factor of five higher, and the rates an order of magnitude higher, than those measured in low-mass star-forming cores; that matters because it shows intermediate-mass cores in such clumps can accumulate enough mass within a free-fall time to become high-mass stars. The pilot study is the first in the GLASHES program, which aims to map infall across the ASHES sample of infrared dark clouds.","feed_headline":"Dark-clump cores swallow gas ten times faster than low-mass cores","feed_subtitle":"ALMA measures nine cores infalling fast enough to grow into high-mass stars.","key_machinery":"The Hill5 model, a two-layer slab radiative transfer model in which excitation temperature rises linearly toward a peak at the slab boundary and falls back to $2.3$ K at the edges, is fitted to the HNC ($J=3-2$) spectra of nine cores to extract the infall velocity from the blue-asymmetric, self-absorbed double-peaked profile. Five parameters (optical depth, systemic velocity, infall velocity, velocity dispersion, and peak excitation temperature) are explored with an affine-invariant MCMC sampler. Supporting that machinery is the position-velocity diagram of N$_2$H$^+$ ($J=1-0$), whose velocity gradients around individual cores provide independent infall timescales comparable to the free-fall time.","core_discovery":"Using the isolated hyperfine component of N$_2$H$^+$ ($J=1-0$), the paper maps a clump-scale velocity gradient along the declination axis and smaller-scale gradients around three cores, interpreting these as accretion flows rather than rotation after finding no rotation signatures in optically thin tracers and after comparison with published velocity-gradient measurements in low-mass cores. The more direct evidence is the blue asymmetry in HNC ($J=3-2$): nine of seventeen cores show the blue-brighter, self-absorbed double-peaked profile, and fitting these with the Hill5 two-layer slab model yields infall velocities of 0.28 to 1.45 km s$^{-1}$, an infall rate for the most massive core of $2.9 \\times 10^{-3}$ $M_\\odot$ yr$^{-1}$, and a strong correlation ($\\rho_s = 0.93$) between infall velocity and nonthermal velocity dispersion. The authors conclude that the nonthermal line width is contaminated by infall and that the cores are collapsing on free-fall timescales, with higher infall rates at larger core masses and closer to the clump center, matching clump-fed expectations.","pith_inferences":["If the infall rates measured here are typical of the ASHES sample, dust-continuum core masses in infrared dark clouds systematically underestimate the mass available for star formation, since cores grow significantly during the collapse phase.","Extending this pilot to the full GLASHES sample would test whether the G337 pattern (central cores infalling fastest) is universal; a direct check is whether clump-scale specific angular momentum correlates with infall rate across clumps.","The correlation between infall velocity and Mach number implies that gravitational collapse, not just turbulence, contributes to the line broadening used to compute virial parameters, so virial masses in such regions may be overestimated.","A testable extension is to compare HNC ($J=3-2$) blue-asymmetry infall velocities with independent estimates from HCO$^+$ ($J=3-2$) and from dust-kinematic measurements, to check whether the two-layer slab geometry biases the derived velocities."],"forward_implications":["If these infall rates persist for a free-fall time, the most massive core in G337 can gain about 30 $M_\\odot$ of additional gas, turning an intermediate-mass core into a high-mass star.","The blue-asymmetry infall signature appears in 53% of the cores, suggesting that core-scale infall is common in the earliest, feedback-free stages of high-mass star formation.","Because infall velocity correlates with nonthermal velocity dispersion ($\\rho_s = 0.93$), line-width-based estimates of turbulent support in such clumps must account for a substantial infall contribution.","Mass infall rate anticorrelates with inverse mass-weighted distance from the clump center, so the most massive, central cores grow fastest, as clump-fed models predict.","Comparison with low-mass star-forming regions shows that same-mass cores accrete more slowly in lower-density environments, implying the local environment sets the infall rate and the potential to form high-mass stars."],"supporting_citations":[{"why":"Supplies the Hill5 two-layer slab model used to fit the blue-asymmetry profiles and extract infall velocities.","marker":"De Vries & Myers 2005"},{"why":"The prior single-core ALMA infall measurement in an infrared dark cloud whose infall rate the paper's results match and extend.","marker":"Contreras et al. 2018"},{"why":"Provides the standard interpretation that a blue-asymmetric double-peaked profile traces infall in a collapsing core with centrally increasing density.","marker":"Evans 1999"},{"why":"Cited for the caveat that the blue-asymmetry signature depends on viewing angle in non-spherical cores.","marker":"Smith et al. 2012"},{"why":"Supplies the velocity-gradient values for low-mass cores used to argue that the observed gradients are too steep for rotation.","marker":"Goodman et al. 1993"},{"why":"Starless-core sample providing comparison infall velocities and rates from low-mass star-forming regions.","marker":"Lee et al. 2001"},{"why":"Perseus core sample providing comparison infall velocities and rates from low-mass star-forming regions.","marker":"Campbell et al. 2016"},{"why":"VeLLOs catalog providing comparison infall velocities and rates from low-mass star-forming regions.","marker":"Kim et al. 2021"},{"why":"ASHES core catalog giving the masses, radii, and densities used to compute infall rates.","marker":"Morii et al. 2023"},{"why":"DCO$^+$ data and virial parameters used for nonthermal velocity dispersions and core dynamical state.","marker":"Li et al. 2023"}],"fun_headline_variants":["Dark clump cores feed 10x faster than low-mass ones","ALMA reveals dark clump cores gulping gas at record rates","Core infall rates 10x higher in massive star nurseries","Fast infall in dark clumps may birth high-mass stars"],"cache_read_input_tokens":3200,"weakest_assumption_plain":"The blue-asymmetric HNC line profile is assumed to be produced by gas falling into a collapsing core; if the asymmetry instead comes from outflows, an unrelated temperature gradient, or a lopsided core seen from a particular angle, the quoted infall speeds and rates would not measure collapse.","fun_headline_variants_meta":{"raw":{"variants":["Dark clump cores feed 10x faster than low-mass ones","ALMA reveals dark clump cores gulping gas at record rates","Core infall rates 10x higher in massive star nurseries","Fast infall in dark clumps may birth high-mass stars"]},"model":"deepseek-v4-flash","effort":"low","cost_usd":0.000142,"raw_usage":{"total_tokens":1259,"prompt_tokens":1127,"completion_tokens":132,"prompt_tokens_details":{"cached_tokens":384},"prompt_cache_hit_tokens":384,"prompt_cache_miss_tokens":743,"completion_tokens_details":{"reasoning_tokens":58}},"tokens_in":743,"tokens_out":132,"duration_ms":2312,"temperature":1.0,"reasoning_tokens":58,"cache_read_input_tokens":384,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-08-11T05:08:16.144753+00:00","model_set":{"reader":"deepseek-v4-flash"},"falsifier":"Take the HNC ($J=3-2$) spectra of the cores whose red and blue peaks have nearly equal brightness (ALMA1, ALMA4, ALMA5) and subtract the outflow components identified in CO and SiO at the same velocities; if the blue-brighter peak disappears or the self-absorption dip no longer lines up with the optically thin DCO$^+$ centroid velocity, then the blue asymmetry is not an infall signature.","supporting_citations":[],"review_version":1}