{"id":"9ba7c69b-63b3-4bd6-bd7c-e229b7a831bd","arxiv_id":"2501.02225","paper_version":1,"verdict":"CONDITIONAL","confidence":"MODERATE","novelty_score":5.0,"correctness_risk":"medium","formal_verification":"none","parameter_count":3,"one_line_summary":"Using passive tracer particles in simulations, the authors show that star-forming cores in clustered regions have lower filling factors, larger masses and sizes, and more protostars than isolated cores.","lead":"This paper uses computer simulations of turbulent gas clouds to trace which gas will fall onto each newborn star, defining each star's mass reservoir. It finds that cores with many nearby stars are clumpy and have low filling factors, and proposes filling factor as a way to distinguish isolated from clustered star-forming regions.","discovery_kind":"new_method","skeptic_critique":{"model":"deepseek-v4-flash","headline":"The 0.3 Myr look-ahead window in Section 2.4(i) that selects which tracer particles define a star-forming core is arbitrary and untested; the filling-factor correlations forming the central claim could depend on this choice.","rationale":"The paper's central claim is that the filling factor of convex hull cores robustly distinguishes isolated from clustered star-forming regions, independent of turbulence strength. Every quantity in this claim — filling factor, mass, radius, number of protostars, and bound fraction — is derived from the star-forming cores, which are defined by the set of tracer particles that accrete within 0.3 Myr of protostar formation (Section 2.4, step i). This look-ahead window is a free parameter with no physical justification beyond an arbitrary choice, and the paper provides no sensitivity analysis. A shorter window would select only gas already close to the protostar, likely producing compact, high-filling-factor cores; a longer window would include gas farther away, potentially spanning multiple accretion streams or competing protostars and lowering the filling factor. The observed inverse correlation between filling factor and protostar count could therefore be a direct consequence of choosing a window long enough to encompass competitive-accretion zones. The same applies to the mass and radius trends and to the unbound-core fraction attributed to the inertial-inflow model. This concern is the most load-bearing because it sits at the root of the analysis chain: if the window changes core membership, all downstream conclusions can change. The reader's weakest assumption identifies exactly this point, and I agree. A secondary but real issue is the swapped bound fractions between Section 6 and Section 4.2, which should be corrected but does not undermine the filling-factor correlation itself. The proposed test — recomputing the analysis with windows of 0.1, 0.2, 0.5, and 1.0 Myr — would settle whether the correlation is robust or an artifact of the chosen timescale. Because the reader already issued a conditional verdict, my read does not move the verdict.","tokens_in":18810,"tokens_out":7550,"duration_ms":72083,"concrete_test":"Re-run the core identification on the existing tracer data with look-ahead windows of 0.1, 0.2, 0.5, and 1.0 Myr, keeping all other steps unchanged. Recompute filling factors, masses, radii, and bound fractions for each window, and test whether the monotonic trends in Figures 7–10 (e.g., the Spearman rank correlation between filling factor and number of protostars) remain statistically significant and have the same sign for all windows. If the correlation reverses or disappears for any alternative window, the central claim is window-dependent and must be qualified.","verdict_should_be":"UNCHANGED","load_bearing_attack":"The central claim requires that the star-forming cores defined in Section 2.4 are robust to the chosen 0.3 Myr look-ahead window. This window determines which gas is assigned to each protostar: shorter windows should yield smaller, more compact cores with higher filling factors, while longer windows will include more distant gas, potentially creating elongated or multi-clump footprints with lower filling factors. All downstream quantities — convex hull geometry, filling factor, mass, radius, and bound fraction — are computed from these cores. Without a sensitivity study, the observed correlations in Figures 7–10 (lower filling factor corresponding to more protostars, larger mass, and larger size) may be an artifact of this one timescale rather than a robust physical indicator. For example, in a clustered region the 0.3 Myr window might be just long enough to capture several competing accretion streams from different protostars, artificially lowering the filling factor and inflating the apparent number of protostars; a shorter window might miss this competition and yield a smooth, high-filling-factor core. The paper provides no test of how the results vary with this parameter. A secondary issue is the swapped bound fractions in Section 6 relative to Section 4.2, but that is a proofreading error rather than a threat to the central filling-factor correlation. The load-bearing assumption is the untested 0.3 Myr window.","agreement_with_reader":"agree"},"referee_report":{"model":"deepseek-v4-flash","summary":"The paper uses three-dimensional adaptive-mesh-refinement hydrodynamics simulations with passive tracer particles to identify 'star-forming cores' as the gas that will accrete onto each protostar within 0.3 Myr after protostar formation. From two runs with initial turbulent Mach numbers 2 and 10, the authors identify 260 star-forming cores, enclose them in convex hulls, and define a filling factor as the ratio of core volume to hull volume. They report that low-filling-factor hulls contain more protostars, have larger masses and sizes, and that most hulls are gravitationally bound (97% at Mach 2 and 84% at Mach 10), with 16% unbound in the Mach 10 model. The filling factor is proposed as an observational indicator for distinguishing isolated from clustered star-forming regions, and the unbound cores are interpreted in the context of the inertial-inflow model.","tokens_in":19129,"tokens_out":5876,"duration_ms":56554,"significance":"The tracer-based core definition is a genuine methodological advance: it targets mass reservoirs directly instead of relying on density-threshold clump finding, and the 260-core sample across two turbulence strengths is substantial for a clump-scale study. The filling factor provides a simple geometric diagnostic that could, if robust, connect simulation-based mass reservoirs to observed core morphology and to the isolated-versus-clustered distinction. The paper also engages usefully with prior work, including Collins et al. (2023) on core filling fractions and Pelkonen et al. (2021) on progenitor cores. However, the central correlations rest on an untested lookahead timescale and on single realizations per Mach number, so at present the quantitative claims are suggestive rather than established.","major_comments":[{"comment":"The 0.3 Myr accretion lookahead window is load-bearing and is not tested. The tracer particles selected in step i define the entire star-forming core, and all downstream quantities (convex hull geometry, filling factor, mass, R90, and virial ratio) are computed from that selection. A shorter window would select gas closer to the protostar and likely produce more compact, higher-filling-factor cores, while a longer window would include more distant gas and could stitch together multiple accretion streams in clustered regions, lowering the filling factor and increasing the number of protostars inside the hull. The authors should provide a sensitivity study over the window (e.g., 0.1, 0.2, 0.3, and 0.5 Myr) and show explicitly that the correlations in Figures 7-10, especially the low-filling-factor/multi-protostar trend, are unchanged. Without this test, the central claim may be an artifact of one timescale.","section":"§2.4, step i"},{"comment":"The bound fractions are internally inconsistent. Section 4.2 and Figure 10 state that 97% of Mach 2 convex hull cores and 84% of Mach 10 cores are gravitationally bound, with 16% unbound at Mach 10. Summary item iv reverses these numbers, reporting '97% of the cores in the Mrms = 10 model and 84% in the Mrms = 2 model.' Because the abstract and discussion rely on 16% unbound at Mach 10, the Summary sentence is incorrect and must be corrected. The reversal is not merely typographical: Summary item iv is presented as the paper's conclusion and directly affects the interpretation of turbulence strength on core stability.","section":"§4.2 vs. Summary item iv"},{"comment":"The paper reports no error bars or statistical uncertainties, and each Mach number is represented by one simulation with one initial turbulent realization. Claims such as 'regardless of turbulence strength' in the abstract and Section 5.2 are supported only by two realizations. The authors should report bootstrap or jackknife uncertainties on the binned averages in Figures 7-10, or run additional turbulent realizations, to show that the filling-factor correlations are not produced by chance. This is especially important because the number of cores per filling-factor bin is small at the extremes (e.g., very low filling factors in Figures 7 and 8).","section":"§4, Figures 7-10"},{"comment":"There is a potential circularity in using the filling factor as an indicator of clustering. The convex hull encloses the star-forming core plus surrounding non-accreting gas and other protostars; when another protostar lies inside the hull, the hull volume can be inflated and the star-forming core can be geometrically fragmented by competitive accretion. Thus low filling factor and a large number of embedded protostars are partly generated by the same construction. A concrete control test would be to compute a filling factor using hulls built only from the target protostar's tracer particles, or to compare against a null distribution obtained by randomly placing the same number of protostars inside a cluster. Without such a control, the manuscript does not fully establish that low filling factor carries information beyond being a direct consequence of how the hull and core were defined.","section":"§4.1-4.2, filling factor definition"},{"comment":"The mass and radius of the convex hull core include other protostars and ambient gas inside the hull, so the stronger correlations in Figures 8 and 9 for low-filling-factor hulls are partly mechanical: a hull that contains more protostars will tend to have larger total mass and larger R90. The authors should separate the gas mass and protostar mass contributions, and report whether the filling-factor correlation with, e.g., gas mass alone or with protostar mass alone still holds. This would strengthen the physical interpretation that low filling factor traces clustered mass supply rather than merely hull size.","section":"§3.1 and §5.2"}],"minor_comments":[{"comment":"The caption lists 'Mach2Mach5Mach10' but only Mrms = 2 and Mrms = 10 are presented in the paper; remove the stray 'Mach5'.","section":"Figure 2 caption"},{"comment":"The caption contains the typo 'represemt' for 'represent'.","section":"Figure 10 caption"},{"comment":"There is a typo, 'particluarly', that should read 'particularly'.","section":"Summary item iv"},{"comment":"The word 'inertia' is misspelled as 'intertia' in the sentence describing the moment of inertia tensor.","section":"§3.2"},{"comment":"Equation (8) for tracer advection is dimensionally awkward as written; please clarify the interpolation notation or refer explicitly to the corresponding equations in Koga et al. (2022).","section":"Equation (8)"},{"comment":"The reference 'Smith et al. 2009' is cited for the virial-parameter definition, but the listed reference is to 'Environment and Planning A: Economy and Space', which appears to be a mismatched citation likely intended for a star-formation simulation paper; please correct it.","section":"References"},{"comment":"The y-axis label '10N' should likely read 'N' (number of protostars).","section":"Appendix A, Figure A1"},{"comment":"The exclusion of five convex hull cores with energy ratios below 0.05 is mentioned only in the caption; the text should describe this selection and state the affected sample sizes.","section":"Figure 10"}],"recommendation":"major_revision","confidential_remarks":"The paper is potentially suitable for this journal after major revision. The main blockers are the untested 0.3 Myr lookahead window, the lack of statistical uncertainty or multiple realizations, and the bound-fraction reversal in the Summary. I would also ask the authors to run the suggested control analyses on the filling-factor definition, since the central claim currently risks being partly definitional. The tracer-particle methodology itself is sound and worth publishing once these points are addressed."},"author_rebuttal":null,"desk_editor":{"model":"deepseek-v4-flash","letter":"The paper delivers a clean, incremental methods advance: it defines star-forming cores as the tracer-particle footprints at protostar formation for gas that will accrete within 0.3 Myr, wraps them in convex hulls, and computes filling factors. Across Mach 2 and 10, lower filling factor correlates with more protostars, higher mass, and larger size. That is a plausible diagnostic with observational relevance, and it builds sensibly on Pelkonen et al. (2021) and Collins et al. (2023). The bound-fraction analysis is standard but honest, and the authors are upfront about missing magnetic fields, feedback, and turbulence driving.\n\nThe load-bearing soft spot is the 0.3 Myr lookahead window. It selects which tracers define a core, and every downstream quantity—filling factor, mass, radius, bound fraction—depends on it. A shorter window will shrink cores and raise filling factors; longer will do the opposite. The paper shows no sensitivity test, and with only one realization per Mach number you cannot separate the trend from stochastic variation. The multiplicity correlation is also partly geometric: when other protostars sit inside the convex hull, the hull is large and the filling factor low, so that correlation is not fully independent. The summary's reversal of the bound fractions (97% vs 84% swapped between Mach 2 and 10) is a proofreading error, but it appears in the abstract-level summary and should be caught. No code or data release accompanies the paper, so the tracer implementation and core catalog are not independently reproducible.\n\nThis is a paper for star formation theorists and observers working on core catalogs and the CMF/IMF connection. The filling factor could become a useful morphological indicator, but it needs the sensitivity test before I would lean on it. It deserves serious refereeing: the method is sound in outline and the correlations are worth publishing after revision. Ask for at least one alternative lookahead timescale (e.g., 0.1 and 0.5 Myr), a second realization if feasible, and a fix for the swapped numbers.","headline":"A useful tracer-based filling-factor diagnostic, but the 0.3 Myr core-definition window is untested and the summary swaps the bound fractions.","tokens_in":19609,"tokens_out":4881,"would_cite":false,"duration_ms":44788,"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 argues that the filling factor of a convex hull around a protostar's accreted gas reservoir distinguishes isolated from clustered star-forming cores, with lower filling factors implying more protostars, larger mass, and larger…","keywords":["star formation","molecular cloud cores","passive tracer particles","filling factor","convex hull","turbulence","protostars","core mass function"],"falsifier":"Re-run the core identification with lookahead windows of, say, 0.1 and 0.5 million years and check whether the filling-factor anticorrelations with protostar number, mass, and radius persist; if they weaken or reverse, the reported indicator is an artifact of the chosen window rather than a property of the mass reservoirs.","tokens_in":18650,"feed_emoji":"⭐","tokens_out":6054,"duration_ms":58574,"temperature":0.7,"pith_summary":"This paper aims to establish that the gas which will actually feed a protostar can be identified in simulations with passive tracer particles, and that the ``filling factor'' of the region enclosing that gas tells whether the protostar is forming in isolation or in a cluster. Using hydrodynamic simulations with two turbulence strengths, the authors identify 260 star-forming cores by tracing gas that falls onto a protostar within 0.3 million years. They find that lower filling factors always go with more protostars, larger masses, and larger sizes, so the filling factor is proposed as an indicator of clustered versus isolated star formation. They also find that most cores are gravitationally bound, but the stronger-turbulence run contains more low-mass unbound cores, consistent with the inertial-inflow picture. The result matters because observed core catalogs may miss the diffuse, fragmented gas that actually feeds stars in clustered regions.","feed_headline":"Low filling factor marks clustered star-forming cores","feed_subtitle":"Gas-tracing simulations show that fragmented, low-filling-factor cores hold more protostars and more mass.","key_machinery":"The central object is the star-forming core defined by passive tracer particles: three million tracer particles, each assigned a gas mass of $0.001\\,M_\\odot$, are advected with the flow and followed until they accrete onto a sink particle, and the gas traced back to the moment of protostar formation defines the mass reservoir. Around each star-forming core the authors build a convex hull, the smallest convex polyhedron enclosing the core, and define the filling factor $\\phi_{\\rm core}=V_{\\rm core}/V_{\\rm hull}$, the fraction of the hull volume that is actually accreting onto the protostar. This construction converts a dynamical accretion history into a geometric, porosity-like statistic that can be correlated with core mass, size, stellar content, and gravitational binding.","core_discovery":"The central claim is that star-forming cores, defined as the actual mass reservoirs of protostars rather than as high-density blobs, do not coincide with the dense regions selected by observational core-finding algorithms once nearby protostars are present. In clustered environments, gas selectively accretes onto several protostars, so a single star's reservoir is clumpy and fragmented. When each reservoir is enclosed in a convex hull, the ratio of reservoir volume to hull volume, the filling factor $\\phi_{\\rm core}$, is lower in cores that contain more protostars and have larger mass and radius, in both the $\\mathcal{M}_{\\rm rms}=2$ and $\\mathcal{M}_{\\rm rms}=10$ runs. No massive convex-hull core has a high filling factor, implying that massive cores feed multiple stars rather than a single star. Finally, 97% of the Mach 2 and 84% of the Mach 10 convex-hull cores are gravitationally bound, with the extra unbound cores in the Mach 10 run being low-mass and attributed to the inertial-inflow model, in which protostars grow by accretion from gas that is not itself self-gravitating.","pith_inferences":["The paper leaves implicit that the filling factor could be estimated from synthetic observations, for example from dendrogram leaf volumes or column-density filling within a bounding polygon, and used to classify observed cores without needing tracer particles.","Because the 0.3 million year lookahead window sets the reservoir size, the reported core masses are trajectory-based rather than instantaneous; an observed core mass function built from density-threshold cores may therefore differ systematically from the true stellar mass reservoir, especially in clusters.","A testable extension is to vary the lookahead window and check whether the filling-factor anticorrelations persist, and to test whether final stellar mass correlates more tightly with star-forming-core mass than with dense-core mass.","If the filling-factor trend survives in runs with magnetic fields or feedback, it could become a practical bridge between numerical reservoir definitions and observed core morphologies."],"forward_implications":["Filling factor can serve as a classification diagnostic: convex-hull cores with low filling factors are associated with clustered regions, while high-filling-factor cores are isolated.","Massive cores always have low filling factors and host multiple protostars, supporting competitive accretion or clump-fed star formation rather than one massive core forming one massive star.","Observed dense cores identified by density thresholds in clustered regions likely underestimate the true mass reservoir, because the reservoir includes diffuse gas and fragmented clumps that observations may miss.","Turbulence strength changes the core population: the Mach 10 run produces more low-mass and more unbound cores than the Mach 2 run, so stronger turbulence yields a higher fraction of gravitationally unbound reservoirs.","Most convex-hull cores are gravitationally bound as whole regions, meaning the reservoir plus surrounding gas is typically collapsing as a unit containing several forming stars."],"supporting_citations":[{"why":"Provides the passive-tracer approach for linking progenitor gas to protostars and the finding that much of the mass reservoir lies outside the identified dense core.","marker":"Pelkonen et al. (2021)"},{"why":"Supplies the SFUMATO adaptive-mesh code, the sink-particle criteria, and the turbulent initial conditions with variable rms Mach number.","marker":"Matsumoto et al. (2015)"},{"why":"Reports overlapping cores and a core filling-fraction distribution peaking near 25%, serving as the comparison for the filling-factor result.","marker":"Collins et al. (2023)"},{"why":"Offers the inertial-inflow model that the paper invokes to explain gravitationally unbound convex-hull cores.","marker":"Padoan et al. (2020)"},{"why":"Shows cores becoming gravitationally bound as they evolve, providing a comparison for the bound-core result.","marker":"Smullen et al. (2020)"},{"why":"Identifies core phases in Mach 2 turbulence simulations and shows cores becoming bound as turbulence weakens, supporting the bound-core interpretation.","marker":"Offner et al. (2022)"},{"why":"Provides the energy-ratio definition and the clump-fed star formation scenario used in the stability analysis.","marker":"Smith et al. (2009)"}],"fun_headline_variants":["Filling factor separates isolated and clustered star cores","Clustered cores are fragmented and hold more protostars","Low filling factor marks multiple-star cores in simulations","Turbulence creates more unbound low-mass cores","Fragmented cores signal clustered star formation"],"cache_read_input_tokens":3200,"weakest_assumption_plain":"The load-bearing premise is the choice of 0.3 million years as the lookahead window for deciding which tracer particles will accrete onto a protostar; changing this window changes which gas is assigned to each core and would directly affect the computed filling factors, masses, sizes, and bound fractions.","fun_headline_variants_meta":{"raw":{"variants":["Filling factor separates isolated and clustered star cores","Clustered cores are fragmented and hold more protostars","Low filling factor marks multiple-star cores in simulations","Turbulence creates more unbound low-mass cores","Fragmented cores signal clustered star formation"]},"model":"deepseek-v4-flash","effort":"low","cost_usd":0.00059,"raw_usage":{"total_tokens":2843,"prompt_tokens":1096,"completion_tokens":1747,"prompt_tokens_details":{"cached_tokens":384},"prompt_cache_hit_tokens":384,"prompt_cache_miss_tokens":712,"completion_tokens_details":{"reasoning_tokens":1673}},"tokens_in":712,"tokens_out":1747,"duration_ms":12775,"temperature":1.0,"reasoning_tokens":1673,"cache_read_input_tokens":384,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-08-10T22:13:18.920891+00:00","model_set":{"reader":"deepseek-v4-flash"},"falsifier":"Re-run the core identification with lookahead windows of, say, 0.1 and 0.5 million years and check whether the filling-factor anticorrelations with protostar number, mass, and radius persist; if they weaken or reverse, the reported indicator is an artifact of the chosen window rather than a property of the mass reservoirs.","supporting_citations":[{"cited_title":"M., Padoan, P., Haugbølle, T., & Nordlund, ˚A","cited_arxiv_id":null,"evidence_quote":"Provides the passive-tracer approach for linking progenitor gas to protostars and the finding that much of the mass reservoir lies outside the identified dense core."},{"cited_title":"C., Le, D., & Jimenez Vela, L","cited_arxiv_id":null,"evidence_quote":"Reports overlapping cores and a core filling-fraction distribution peaking near 25%, serving as the comparison for the filling-factor result."},{"cited_title":null,"cited_arxiv_id":null,"evidence_quote":"Identifies core phases in Mach 2 turbulence simulations and shows cores becoming bound as turbulence weakens, supporting the bound-core interpretation."},{"cited_title":"M., Clarke, G","cited_arxiv_id":null,"evidence_quote":"Provides the energy-ratio definition and the clump-fed star formation scenario used in the stability analysis."}],"review_version":1}