{"id":"bf01d30f-58bf-4da7-bd96-8bd9a9f309ac","arxiv_id":"2506.16664","paper_version":1,"verdict":"CONDITIONAL","confidence":"MODERATE","novelty_score":5.0,"correctness_risk":"medium","formal_verification":"none","parameter_count":5,"one_line_summary":"G333 and G331 decompose into 148 hub-filament systems and 243 non-hub-filament structures, supporting a picture of molecular clouds as multi-scale gravitational networks.","lead":"This paper uses high-resolution carbon monoxide maps to break two giant molecular clouds into 148 hub-filament systems and 243 non-hub-filament structures. It proposes that molecular clouds are networks of these gravitational hubs and filaments, with dense clumps playing the role of the hubs.","discovery_kind":"extension","skeptic_critique":{"model":"deepseek-v4-flash","headline":"The inflow interpretation of filament velocity gradients is not uniquely established; the network claim's gravitational coupling rests on it and needs a directional test.","rationale":"The reader's weakest_assumption identifies the same load-bearing concern: the kinematic evidence for inflow is not uniquely established because rotation, outflow, or projection effects are not ruled out. I agree this is the most critical point for the central claim. The paper's main assertion, that molecular clouds are network structures formed by gravitational coupling of hub-filament systems, requires that the filaments be genuine accretion channels into the hubs. The morphological identification, density contrast differences, and clump associations establish that hubs are dense centers, but they do not by themselves demonstrate gravitational coupling. The velocity-gradient measurements are the only direct kinematic evidence, and they are interpreted as inflow based on magnitude agreement with free-fall models. This interpretation is load-bearing: if the gradients are due to rotation or other effects, the gravitational coupling claim is not supported. The reader's verdict of CONDITIONAL is appropriate, with the condition being a robust test of the inflow interpretation. I recommend no change to that verdict. The circularity in density contrast, while a flaw, is partly mitigated by independent physical property differences and does not directly undermine the network concept. The distance ambiguity for peak1 affects masses but is plausibly minor given the dominance of peak2/peak3. Thus the kinematic test is the single most important check to settle whether the central claim holds.","tokens_in":13757,"tokens_out":4976,"duration_ms":57575,"concrete_test":"For each of the 148 HFs, extract the velocity profile along each filfinder filament and fit the velocity gradient as in Sec. 3.4.2. Record the sign of the gradient and compute the angle between the gradient direction (in the position–velocity plane) and the spatial vector from the filament's outer end to the hub center. Test whether the hub-end velocity is systematically closer to the hub's systemic velocity than the outer-end velocity, and compare the distribution of angles to a random null. If the alignment is not significantly better than chance, or if the hub-end velocity is not closer to systemic, the inflow interpretation is unsupported. Alternatively, observe a subset of HFs in an optically thin infall tracer (e.g., HCO+ or N2H+) and search for blue-asymmetric line profiles toward the hubs.","verdict_should_be":"UNCHANGED","load_bearing_attack":"The central claim that molecular clouds are networks formed by gravitational coupling of hub-filament structures depends critically on the kinematic evidence in Sec. 3.4.2. The authors fit velocity gradients along filaments and compare their magnitudes to free-fall models (Fig. 6), concluding that gas is flowing toward the hubs. However, a linear velocity gradient along a filament can also arise from rotation, outflow, shear, or line-of-sight projection of an unrelated large-scale velocity field. The paper does not report the direction of the fitted gradient relative to the hub, nor does it verify that the gas velocity approaches the hub's systemic velocity at the hub end, as expected for inflow. Without such a check, the magnitude comparison to free-fall is insufficient to rule out alternative kinematic origins. If the gradients are not dominated by inflow, the hub-filament systems remain morphological constructs, and the 'gravitational coupling' that knits the network loses its primary empirical support. The circularity in the density contrast (C) selection, while real, is secondary because mass, virial ratio, and clump/radio associations provide independent discriminants; the kinematic interpretation is the only direct evidence for the gravitational coupling itself.","agreement_with_reader":"agree"},"referee_report":{"model":"deepseek-v4-flash","summary":"The paper uses APEX/LAsMA 13CO(3-2) and 12CO(3-2) maps of the G333 complex and G331 GMC to identify intensity peaks with astrodendro, classify them into hub-filament systems (HFs) and non-HFs using a density contrast C (Eq. 1) plus filament morphology, and derive masses, virial ratios, velocity gradients, and associations with ATLASGAL clumps and CORNISH-South radio sources. It reports 148 HFs, finds that HFs have larger masses, lower virial ratios, higher density contrast, and more frequent clump/radio associations than non-HFs, and interprets these trends as evidence for an evolutionary sequence from non-HFs to HFs. The paper concludes that molecular clouds are network structures built from gravitationally coupled multi-scale hub-filament systems, with hubs as the nodes and clumps equivalent to hubs.","tokens_in":13941,"tokens_out":7734,"duration_ms":76512,"significance":"If the central interpretation holds, this is one of the largest systematic decompositions of a GMC into hub-filament systems and strengthens the observational basis of the hub-filament paradigm at cloud-clump scales. The main strengths are the homogeneous LAsMA data, the explicit multi-wavelength cross-matching to ATLASGAL and CORNISH-South, the inclusion of independent discriminants such as virial ratio and luminosity-to-mass ratio, and the connection to prior kinematic studies of G333. However, the significance is currently limited by two load-bearing caveats: the density-contrast comparison is in part a selection effect, and the kinematic evidence for inflow has not been shown to be directional. The independent property differences and clump/radio associations are valuable regardless, but the broad network claim in Sec. 4.3 needs stronger kinematic support.","major_comments":[{"comment":"The evidence for inflow rests on fitting velocity gradients along filaments and comparing their magnitudes with free-fall predictions. A linear gradient along a filament can also be produced by rotation, shear, an outflow, or line-of-sight projection of a larger-scale velocity field. The paper does not report the sign of dv/dl relative to the hub position, does not check whether the filament velocity converges to the hub's systemic velocity at the hub end, and does not state how many of the 148 HFs pass a directional test. Please add such a test (e.g., gradient vectors pointing toward the hub, velocity reversal across the hub, or position-velocity diagrams along the spine) and report the passing fraction. This is needed to support the 'gravitational coupling' language in Sec. 4.3 and Summary point 7.","section":"Sec. 3.4.2, Figs. 5 and 6"},{"comment":"Because C is a primary criterion in the HF/non-HF classification, and because structures near C≈1.5 that were 'challenging to classify' were excluded without a reported count, the statement that HFs have higher density contrast is at least partly a selection effect rather than an independent empirical finding. This matters for Sec. 4.2, where C is interpreted as measuring the extent of gravitational collapse and the strength of the gravitational center. Please validate that interpretation independently: for example, within the HF sample alone, correlate C with L/M or with radio association, or re-derive the classification without using C and check whether the property differences persist. The independent discriminants (mass, virial ratio, clump/radio associations) should be the primary support for the evolutionary sequence.","section":"Sec. 3.3, Eq. (1), Figs. 4(a) and 7"},{"comment":"The reduction of type2 structures to their dominant velocity component is an ad hoc assumption that can remove genuine multi-component inflow, which is precisely the kind of kinematic signature expected in hub-filament systems according to the arguments listed in Sec. 3.2. Please justify this choice with position-velocity diagrams for representative type2 structures, or show that the velocity-gradient and mass estimates are robust when secondary components are retained.","section":"Sec. 3.2, type2 structures"},{"comment":"The claim that 'clumps in molecular clouds are equivalent to the hubs' equates 13CO intensity leaves with ATLASGAL dust clumps without a quantitative match of sizes, masses, or velocities, and the universal network statement is extrapolated from two complexes. Please soften or support these statements: report the fraction of hubs that coincide with ATLASGAL clumps by both position and size, and add an explicit caveat that the network hypothesis is based on the G333/G331 data and needs testing in other clouds.","section":"Sec. 4.3 and Summary point 7"}],"minor_comments":[{"comment":"The text 'As shown in Fig.3.5(e)' should read 'Fig.4(e)'.","section":"Sec. 4.2"},{"comment":"There is a typo in 'the same analysis presentend in Zhou et al. (2022)'; it should read 'presented'.","section":"Sec. 3.4.2"},{"comment":"Zhou et al. 2024b and 2024c are listed with the same A&A volume, article number, and DOI; one of these entries needs to be corrected or disambiguated.","section":"References"},{"comment":"The text describes distribution differences as 'significant' but does not report p-values or confidence intervals; please add a two-sample test (e.g., Kolmogorov-Smirnov or Mann-Whitney) for each comparison.","section":"Figs. 4 and 9"},{"comment":"The choice of min_value = 15 K km/s is said to be the best for recovering high-intensity peaks, but no quantitative criterion or sensitivity test is given; please show how the number of leaves and HFs changes with this parameter.","section":"Sec. 3.1"},{"comment":"The factor-of-two regridding is arbitrary; please show that the identified filaments and fitted gradients are stable under a different pixel scale.","section":"Sec. 3.4.1"},{"comment":"The distance ambiguity for peak1 is dismissed as not significantly affecting the statistics; please quantify the effect on masses and sizes of structures associated with peak1.","section":"Sec. 3.1"}],"recommendation":"major_revision","confidential_remarks":"This manuscript is part of a series by the same group, and the novelty relative to Zhou et al. (2023) and Zhou et al. (2024d) should be clarified in the revision; the new contribution is the systematic HF decomposition and the clump/radio statistics, while the network interpretation is largely carried over from those papers. The reference list contains a suspicious duplicate (2024b/2024c) that should be corrected before resubmission. The paper is within the scope of A&A and the data set is valuable, but the central kinematic claim needs a directional test before the network conclusion can be accepted at face value."},"author_rebuttal":null,"desk_editor":{"model":"deepseek-v4-flash","letter":"Colleague,\n\nQuick take: this is a workmanlike, useful decomposition paper with a real catalog at its core, but two soft spots you should know about: the headline density-contrast result is partly selected by the classification, and the kinematic evidence for inflow is not as strong as the abstract implies.\n\nThe new thing is concrete: 148 hub-filament systems and 243 non-HFs in the G333 complex and G331 GMC, identified from 13CO(3-2) LAsMA data, with matched ATLASGAL clumps, CORNISH-South radio sources, and MIR emission. The clump/radio association contrast (83% vs 32% for clumps; 29% vs 4% for radio) is a genuinely independent discriminant and is the most convincing evidence that HFs are more evolved. The L/M ratios and 8um peak intensity reinforce that. The authors also do a good job of cleaning overlapping velocity components and checking that the smaller scale of non-HFs is not just a resolution effect (Figure 8). That is careful work.\n\nThe density-contrast claim is partly circular. HFs are selected partly on the basis of having a distinct high-density center, quantified by C (Eq. 1), and then Figure 4a reports that HFs have higher C. So that specific comparison is not evidence. But because mass, virial ratio, and the clump/radio associations are not used in the classification, the general conclusion that HFs are more evolved survives. Just drop the C comparison as support.\n\nThe bigger soft spot is kinematics. The paper fits velocity gradients along filaments, compares their magnitudes to free-fall expectations, and concludes gas is flowing toward the hubs. A linear gradient along a filament can also come from rotation, shear, or projection of an unrelated velocity field. They don't test whether the gradient points toward the hub or whether the gas velocity approaches the hub's systemic velocity at the hub end. Without a directional test, this is weak support for inflow. The network claim — that clouds are gravitational networks of multi-scale HFs — rests largely on this inflow picture, so it should be framed as a hypothesis, not a result. The summary does use “propose,” but the abstract's “suggest” and the earlier PPV funnel evidence from Zhou et al. (2023) give it more weight than this paper alone warrants.\n\nMinor points: no error bars or significance tests on the distribution comparisons; the C=1.5 and 2.5x extension thresholds are arbitrary and untested. A sensitivity analysis would help.\n\nWho should read this: anyone working on hub-filament systems or high-mass star formation in G333. It's a solid catalog paper, worth citing for the decomposition and the clump associations, but treat the evolutionary interpretation as one plausible reading, not a proof. Worth a serious peer review; with a proper significance test and a directional velocity-gradient analysis, it could be much stronger.","headline":"A useful new catalog of hub-filament systems in G333/G331, but the headline density-contrast result is partly built into the classification and the inflow interpretation needs a directional test.","tokens_in":14550,"tokens_out":4903,"would_cite":true,"duration_ms":48090,"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 molecular clouds are network structures assembled from gravitationally coupled, multi-scale hub-filament systems, in which hubs act as local gravitational centers and the primary sites of star formation.","keywords":["ISM: structure","ISM: evolution","stars: formation","submillimeter: ISM","hub-filament systems","density contrast","velocity gradients","molecular cloud networks"],"falsifier":"Compute synthetic 13CO(3−2) maps of a cloud containing rotating or outflowing filaments with no true inflow and run the same peak-finding, density-contrast, and velocity-gradient pipeline; if hub-filament classifications and free-fall-like gradient scalings still emerge, the inflow interpretation is falsified. Observationally, a proper-motion survey of dense cores in a few dozen filaments would show directly whether the gas moves toward the hub at the free-fall speeds implied by the fitted masses.","tokens_in":13507,"feed_emoji":"🌌","tokens_out":6803,"duration_ms":65735,"temperature":0.7,"pith_summary":"Using high-resolution 13CO (3−2) maps of the G333 complex and the G331 giant molecular cloud, the authors decompose the gas into hundreds of hub-filament units and argue that these units, not the cloud itself, are the fundamental building blocks of molecular clouds. The hubs are dense local gravitational centers where star formation happens, and the filaments are gas streams feeding them. Hub-filament systems turn out to be denser, more massive, less virialized, more frequently associated with embedded star formation, and to show velocity gradients consistent with gravitational inflow, supporting an evolutionary sequence from uniform non-hub structures to hub-filament systems. If this picture is right, it reframes how cloud-scale star formation, feedback, and the structure of giant molecular clouds are understood.","feed_headline":"Molecular clouds are networks of hub-filament systems","feed_subtitle":"13CO maps of two giant complexes reveal hubs that collect inflowing gas and feed star formation.","key_machinery":"The central object is the hub-filament system (HFS): a dense hub where three or more filaments converge, operationally identified as a leaf in a dendrogram decomposition of the integrated-intensity map plus the surrounding 2.5-times-extended region. The key quantitative quantity is the density contrast, C = N_hub / N_annulus, the ratio of the mean column density in the hub to that in an elliptical ring just outside it; values around 1.5 separate HFs from non-HFs. The kinematic machinery is the fitted velocity gradient along each filament, compared with free-fall models for central masses of 100–10000 solar masses and interpreted as gas flowing toward the hub. Together this morphological and kinematic apparatus converts a catalogue of dense structures into an evolutionary and dynamical story.","core_discovery":"The paper decomposes the G333 complex and the G331 giant molecular cloud into multi-scale hub-filament systems (HFs), structures in which a dense hub collects gas along converging filaments. Applying a dendrogram-based peak identification, a filament-finding algorithm, and velocity-gradient fitting, the authors classify 148 of 438 leaf structures as HFs and 243 as non-HFs. Compared with non-HFs, HFs have significantly higher density contrast, larger masses, lower virial ratios, and higher rates of association with submillimetre clumps and radio continuum sources, as well as higher 8 µm peak intensities and clump luminosity-to-mass ratios. The velocity gradients measured along the filaments follow the free-fall scaling expected for central masses of roughly 100 to 10000 solar masses, matching the leaf mass distribution, and are interpreted as gas inflow toward the hubs. The paper concludes that molecular clouds are network structures formed by the gravitational coupling of multi-scale hub-filament systems: the hubs are the nodes, the local gravitational centers, and the main star-forming sites, and clumps in molecular clouds are equivalent to hubs. This network picture is offered as a natural explanation for why early feedback from protoclusters does not significantly change the kinematic properties of surrounding dense gas.","pith_inferences":["Extending the paper's logic, if hub-filament systems are genuinely scale-free building blocks, then the same decomposition applied to other giant molecular clouds should yield hub masses and filament lengths with scale-free distributions; testing that prediction would tell whether G333 is typical.","The network model implies that star formation efficiency in a cloud may depend more on the number and gravitational strength of its hubs than on total cloud mass, a distinction that could be tested by comparing clouds of equal mass but different hub populations.","If non-HFs evolve into HFs through gravitational focusing, then density contrast should increase monotonically with structural age; using chemical clocks such as N2H+ abundance or CO depletion as age tracers could test this proposed sequence directly.","The claim that early feedback is only a topological deformation predicts that hubs separated by more than one filament length should show statistically independent velocity dispersions; a pairwise correlation analysis of hub kinematics could falsify or support this independence."],"forward_implications":["Molecular clouds should be treated as networks of hubs connected by filaments, so the natural unit of cloud-scale star formation is the hub-filament system rather than the cloud as a monolithic object.","Clumps in molecular clouds are equivalent to hubs, meaning clump catalogs can be reinterpreted as catalogs of network nodes that undergo gravitational focusing.","Early feedback from forming stars mainly reshapes the topology of the network rather than destroying the cloud, leaving the kinematics of embedded dense gas largely governed by gravity.","Density contrast C can serve as a single diagnostic of how far gravitational collapse has progressed and how strong the gravitational center is, complementing or even superseding density itself as an evolutionary indicator.","Non-hub structures are likely earlier evolutionary states that will develop hubs and filaments, implying that the observed mix of HFs and non-HFs in a cloud maps onto a temporal sequence."],"supporting_citations":[{"why":"Supplies the LAsMA 13CO and 12CO data cubes and the large-scale kinematic evidence (velocity gradients, funnel structure) on which this paper builds.","marker":"Zhou et al. (2023)"},{"why":"Provides the procedure for identifying leaves and defining per-structure velocity ranges, plus the earlier conclusion that feedback leaves dense-gas kinematics largely unchanged.","marker":"Zhou et al. (2024d)"},{"why":"Supplies the filament-detection algorithm used to identify and characterize filaments around the hubs.","marker":"Koch & Rosolowsky (2015)"},{"why":"Established the multi-scale hub-filament model at clump-core scale and the velocity-gradient fitting method that this work applies to cloud-scale structures.","marker":"Zhou et al. (2022)"},{"why":"Provides the statistical census of hub-filament candidates around clumps that motivates the density-contrast measure and hub enhancement interpretation.","marker":"Kumar et al. (2020)"},{"why":"Supplies the submillimetre clump catalogue used to associate leaf structures with clumps and to infer the evolutionary stage of HFs versus non-HFs.","marker":"Urquhart et al. (2022)"},{"why":"Previous result that early feedback does not significantly change the physical properties of embedded dense gas structures, which the network interpretation is designed to explain.","marker":"Zhou et al. (2024e)"}],"fun_headline_variants":["Clouds are hub-filament networks","Dense hubs draw gas along filaments","Molecular clouds: gravitational networks of hubs","Hubs and filaments: the architecture of clouds"],"cache_read_input_tokens":3200,"weakest_assumption_plain":"The load-bearing premise is that the fitted velocity gradient along each filament traces gas actually flowing toward the hub; if rotation, outflow, or line-of-sight projection produces those gradients instead, the kinematic evidence for classifying the structures as hub-filament systems falls away.","fun_headline_variants_meta":{"raw":{"variants":["Clouds are hub-filament networks","Dense hubs draw gas along filaments","Molecular clouds: gravitational networks of hubs","Hubs and filaments: the architecture of clouds"]},"model":"deepseek-v4-flash","effort":"low","cost_usd":0.000363,"raw_usage":{"total_tokens":2018,"prompt_tokens":1069,"completion_tokens":949,"prompt_tokens_details":{"cached_tokens":384},"prompt_cache_hit_tokens":384,"prompt_cache_miss_tokens":685,"completion_tokens_details":{"reasoning_tokens":895}},"tokens_in":685,"tokens_out":949,"duration_ms":9381,"temperature":1.0,"reasoning_tokens":895,"cache_read_input_tokens":384,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-08-15T19:21:26.700402+00:00","model_set":{"reader":"deepseek-v4-flash"},"falsifier":"Compute synthetic 13CO(3−2) maps of a cloud containing rotating or outflowing filaments with no true inflow and run the same peak-finding, density-contrast, and velocity-gradient pipeline; if hub-filament classifications and free-fall-like gradient scalings still emerge, the inflow interpretation is falsified. Observationally, a proper-motion survey of dense cores in a few dozen filaments would show directly whether the gas moves toward the hub at the free-fall speeds implied by the fitted masses.","supporting_citations":[{"cited_title":"W., Wyrowski, F., Neupane, S., et al","cited_arxiv_id":null,"evidence_quote":"Supplies the LAsMA 13CO and 12CO data cubes and the large-scale kinematic evidence (velocity gradients, funnel structure) on which this paper builds."}],"review_version":2}