{"id":"eba23b5a-d29f-4a31-a595-95107966f52f","arxiv_id":"1908.02301","paper_version":1,"verdict":"CONDITIONAL","confidence":"HIGH","novelty_score":4.0,"correctness_risk":"medium","formal_verification":"none","parameter_count":0,"one_line_summary":"A review arguing that three widely used assumptions about star clusters, the YMC-globular analogy, resolved cluster mass loss, and a universal CFR-SFR link, remain unproven or misleading in current practice.","lead":"This conference review argues that three common shortcuts in star cluster research are not supported by current evidence: young massive clusters may not stand in for ancient globular clusters, galaxy simulations do not resolve the tidal shocks that destroy clusters, and the cluster formation rate does not always track the star formation rate. It is a cautionary piece aimed at researchers using simulations and upcoming surveys to interpret cluster formation and evolution.","discovery_kind":"review","skeptic_critique":{"model":"deepseek-v4-flash","headline":"Central no-simulation claim rests on an unquantified sub-resolution noise assertion; a controlled convergence test could settle it.","rationale":"The reader identifies the Section 2 footnote on resolution elements and numerical noise as the load-bearing premise, and I reach the same conclusion. The paper's strongest claim is a strong negative about all current cosmological simulations, and its force depends on the assertion that sub-resolution tidal measurements are unphysical. That assertion is plausible but not demonstrated here; it is cited to the author's own PhD thesis, and the footnote's dismissal of E-MOSAICS's convergence on mass-loss is a conjecture rather than a proof. The paper is a clearly written, useful cautionary review, but the central claim is stronger than the evidence presented. Since the reader already assigned a CONDITIONAL verdict based on exactly this soft spot, my stress test does not move the verdict: UNCHANGED is appropriate, with the condition that the resolution/noise assertion be either demonstrated or softened before the paper is used as a definitive methodological warning.","tokens_in":7350,"tokens_out":3315,"duration_ms":39874,"concrete_test":"Build a controlled numerical experiment: take a high-resolution galaxy simulation that resolves GMC-scale gas structures (or an idealized disk plus molecular-cloud model), smooth the density/potential to roughly 100 pc cells, and recompute tidal tensors with the E-MOSAICS derivative scheme. Feed both the smoothed and the true tidal histories into the same N-body cluster evolution code and compare the resulting cluster mass-loss distributions. If the distributions agree within the E-MOSAICS convergence range, the numerical-noise objection fails; if they differ by more than the claimed uncertainty, the objection is confirmed.","verdict_should_be":"UNCHANGED","load_bearing_attack":"The Section 2 conclusion that 'to date no cosmological simulation captures the tidal field with sufficient precision to predict the mass-loss of clusters' is the paper's most consequential claim, but it depends on the footnote assertion that any tidal quantity measured below a few resolution elements is 'unphysical, and likely relates to numerical noise induced by the derivation,' supported only by a citation to the author's PhD thesis. The paper does not quantify how much cluster mass-loss is controlled by unresolved, rapidly varying tidal shocks compared with the resolved secular tide, nor does it show why the E-MOSAICS convergence on mass-loss across smoothing scales is invalid. The rebuttal that resolving smaller structures 'would add their contribution to the net tidal field' assumes both that such structures exist in the simulated galaxies and that their contribution dominates the mass-loss budget. Without quantitative evidence for these assumptions, the sweeping negative claim is an assertion about required resolution rather than a demonstrated failure of existing simulations.","agreement_with_reader":"agree"},"referee_report":{"model":"deepseek-v4-flash","summary":"This invited proceedings contribution (IAU Symposium 351) presents three cautionary points about the current state of star cluster research. First, present-day young massive clusters (YMCs) cannot currently be established as local analogues of young globular clusters, because the formation conditions at high redshift differ in gas fraction, turbulence, metallicity, and likely in the stellar IMF, and because the observed universality of the present-day globular cluster mass function is difficult to reconcile with a universal initial CMF under environment-dependent evolution. Second, the author argues that no current cosmological simulation resolves the small-scale, rapidly varying tidal field component (disk crossings, cloud encounters, sub-10 pc ISM structure) that drives cluster mass-loss, with E-MOSAICS singled out as an example of tidal shocks measured below the resolution limit; the paper concludes that cluster mass-loss and the evolution of the CMF remain out of reach of current simulations. Third, the cluster formation rate (CFR) need not trace the star formation rate (SFR), because star formation can occur in pre-existing clusters through the formation of multiple stellar populations and gas accretion onto nuclear clusters. The body of the paper is carefully hedged, with explicit statements such as \"we still don't know\", while the Section 2 conclusion is stated categorically.","tokens_in":7456,"tokens_out":15424,"duration_ms":148291,"significance":"If the Section 2 claim is correct, it has broad consequences: inferences about cluster destruction, survival timescales, and the evolution of the cluster mass function that are drawn from galaxy-scale and cosmological simulations (including E-MOSAICS) would rest on an unverified basis, and the paper sets a concrete, falsifiable resolution target (effective sub-parsec resolution with several resolution elements covering cold, <10 pc ISM features). The paper is a perspective rather than a new calculation, but it makes its claims testable: the Section 2 resolution requirement can be checked by controlled convergence studies, and the Section 3 prediction of a CFR-SFR mismatch in mergers can be confronted with observations. I credit the author for engaging directly with the E-MOSAICS convergence argument in a footnote rather than ignoring counter-evidence, and for repeatedly flagging the limits of the community's knowledge. The main weakness is that the paper's most consequential claim rests on an unquantified assertion about numerical noise below the resolution limit, which makes the strength of the conclusion disproportionate to the evidence presented.","major_comments":[{"comment":"The section's conclusion — \"to date no cosmological simulation captures the tidal field with sufficient precision to predict the mass-loss of clusters\" — rests on the footnote assertion that \"any quantity measured with less than a few resolution elements (and all the more below resolution limit, like in E-MOSAICS) is unphysical, and likely relates to numerical noise induced by the derivation\", which is supported by a citation to the author's PhD thesis (Renaud 2010) rather than demonstrated here. This premise is load-bearing: if it is false, the critique of E-MOSAICS and the global negative conclusion lose their force. The author should either provide a quantitative justification (e.g., a convergence test of the tidal field measurement itself, or an estimate of the numerical noise floor relative to the physical amplitudes of disk-crossing and cloud-encounter shocks) or soften the claim to state that the required resolution has not yet been demonstrated, rather than that existing measurements are unphysical. The rest of the paper is carefully hedged, and this categorical step stands out.","section":"§2, footnote and concluding paragraph (p. 4-5)"},{"comment":"The rebuttal to the E-MOSAICS convergence test states that resolving smaller structures \"would add their contribution to the net tidal field\". This requires two quantitative inputs that the paper does not supply: that structures at the relevant scales (<10 pc clouds and finer) exist with the assumed properties in the simulated galaxies, and that their contribution dominates the cluster mass-loss budget relative to the resolved secular tide. Without such an estimate (e.g., from analytic shock-driven mass-loss prescriptions or from the higher-resolution Li et al. 2017 simulations), the interpretation of mass-loss convergence across smoothing scales as evidence of uncaptured physics, rather than of a physically converged treatment, cannot be evaluated. I recommend adding a short quantitative discussion of the relative mass-loss contributions from the resolved versus unresolved tidal components, or explicitly framing the claim as a resolution requirement for future work.","section":"§2, E-MOSAICS convergence footnote (p. 4)"}],"minor_comments":[{"comment":"The section titles \"Young massive clusters are not local analogues of young globular clusters\" and \"The cluster formation rate does not correlate with the star formation rate\" are stronger than the corresponding body text, which explicitly states \"we still don't know\" and \"quantifying the importance of these mechanisms is necessary before reaching definite conclusions\". I recommend rewording the titles to match the hedged content (e.g., \"... are not established as local analogues ...\" and \"... need not correlate ...\").","section":"Section 1 and Section 3 titles"},{"comment":"There are several typos in Section 2: \"E-MOASAIC\" should be \"E-MOSAICS\" (in both the text and the footnote), \"a effectiv e sub-parsec resolution\" should be \"an effective sub-parsec resolution\", and the phrase \"It is needless to say\" is awkward and could simply be removed.","section":"§2"},{"comment":"The proceedings title contains \"W ay\" and the abstract contains \"V era Rubin Survey Telescope\", which appear to be compilation or OCR artifacts and should be corrected during editing. In addition, the acronym VRST is non-standard; the facility is normally referred to as LSST or the Vera C. Rubin Observatory.","section":"Title and Abstract"},{"comment":"In the bullet list, \"An holistic understanding of the IMF is still lacking\" should read \"A holistic understanding\", and the phrase \"the stellar winds (of which energy depends on the metallicity)\" is awkward and should be rephrased.","section":"§1, bullet list"},{"comment":"The claim that \"simulations suggest that the CFR significantly deviates from the SFR at different degrees at different stages of a galactic interaction (see e.g. Fig. 11 of Renaud et al. 2015)\" is not supported by a reproduced figure; since this is a proceedings paper, citing an internal figure is acceptable, but including the relevant panel or an independent confirmation would help the reader assess the claim.","section":"§3"},{"comment":"The paper relies heavily on the author's own prior work (Renaud 2010, 2014, 2015, 2017; Gieles & Renaud 2016) for several of the load-bearing physical and numerical points; this is not improper, but independent references would strengthen the arguments for readers outside the author's immediate collaborations, notably for the resolution/noise criterion and the CFR-SFR mismatch.","section":"References"}],"recommendation":"major_revision","confidential_remarks":"This is a proceedings-style perspective contribution, so I do not expect a new calculation; however, the paper's most consequential claim (Section 2) is currently supported by an assertion about numerical resolution that is cited to the author's own PhD thesis. If the proceedings' editorial policy permits programmatic statements without full quantitative support, a minor revision could suffice; under standard research-journal criteria, the load-bearing premise needs work, hence my recommendation of major_revision. I would also note for the editor that the Section 2 critique names a specific, widely cited project (E-MOSAICS); the critique is made in good faith and cites the relevant papers, but the categorical tone of the conclusion could be softened for a proceedings volume."},"author_rebuttal":null,"desk_editor":{"model":"deepseek-v4-flash","letter":"This is a conference proceedings piece, not a new measurement or simulation, and the reader's conditional verdict is about right. What the paper does well: it is clearly written, honest about what we don't know, and it bundles three genuine caveats that often get lost in the literature. The YMC/globular analogy point is carefully argued and the 'we still don't know' refrain is a fair brake on overreach. The CFR/SFR discussion is a useful reminder that star formation in pre-existing clusters breaks a naive proportionality. The real value, though, is the Section 2 critique of E-MOSAICS: deriving tidal fields from potential differences at scales 200 times below resolution does look dubious, and this deserves a public airing.\n\nNow the soft spots. The section titles overstate the body: 'are not local analogues' and 'does not correlate' are stronger than the carefully hedged text underneath. That mismatch matters because the titles are what people will remember. More importantly, the central claim of Section 2 — that no cosmological simulation can predict cluster mass-loss because unresolved structures dominate the tidal field — rests on a footnote assertion that any quantity measured below a few resolution elements is 'unphysical, likely numerical noise,' supported only by a citation to the author's PhD thesis. That is a plausible rule of thumb, but it is not demonstrated here, and the stress-test note is right that the paper never quantifies how much of the mass-loss budget comes from sub-resolution shocks versus the resolved secular tide. The rebuttal that E-MOSAICS's convergence on mass-loss across smoothing scales is itself evidence that shocks are not captured assumes that resolving smaller structures would necessarily add to the tidal field; that might be true, but it is not proven. So the strongest conclusion should be 'we still don't know whether current simulations get the mass-loss right,' not 'they definitely don't.'\n\nI would send this to a referee. It is a well-written, opinionated review that identifies a real methodological concern, and the E-MOSAICS critique, even if overstated, is sharp enough to warrant careful response from the simulators. The referee's main job would be to force the author to either soften the Section 2 claim or provide a quantitative resolution test. For a reading group, it would spark a good discussion about resolution and interpretation in galaxy-scale simulations, but it is not a must-read. I would not cite it in my own work, though I might mention it as a cautionary flag.\n\nRecommendation: accept it as a proceedings contribution after minor revision, with the request that the author qualify the no-simulation claim and add a sentence about what a convergence test would look like. The paper is honest and well-intentioned, but the central assertion needs a better foundation.","headline":"A lucid, opinionated conference review that earns a serious referee, but its most important claim about simulation tides rests on an unquantified resolution criterion.","tokens_in":8034,"tokens_out":1292,"would_cite":false,"duration_ms":15550,"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":"No current simulation can predict how star clusters dissolve.","keywords":["star clusters","globular clusters","tidal field","tidal shocks","cluster mass function","cosmological simulations","star formation rate","cluster formation rate"],"falsifier":"Take one galaxy formation simulation and run it at both several-hundred-parsec and sub-parsec gas resolution with identical initial conditions; if the coarse run reproduces the same cluster mass-loss statistics as the fine run, the central claim fails, while a difference in tidal shocks would support it.","tokens_in":7095,"feed_emoji":"🌌","tokens_out":8586,"duration_ms":83674,"temperature":0.7,"pith_summary":"This paper argues that three widely used shortcuts in star cluster research are misleading. It first contends that young massive clusters are not established local analogues of the globular clusters formed in the early universe, because the present-day globular cluster mass function is universal while the young massive cluster mass function is a power law, implying the initial mass function must depend on environment. It then makes the stronger claim that no cosmological simulation currently resolves the rapidly varying tidal field that drives cluster mass-loss, so predictions of cluster survival and dissolution from galaxy-scale simulations are unreliable. Finally, it argues that the cluster formation rate does not track the star formation rate, since stars can form in pre-existing clusters and in nuclear clusters fed by gas inflows. If these points hold, interpreting star cluster populations without a resolved tidal field and an environment-dependent initial mass function is premature.","feed_headline":"No simulation can yet predict star cluster dissolution","feed_subtitle":"Three unchecked assumptions about tides, analogues, and formation rates underlie cluster evolution claims.","key_machinery":"The central object is the tidal field around a cluster, split into a slowly varying secular component and a rapidly varying shock component. The argument's load-bearing mechanism is a resolution criterion: physical structures that generate shocks—disk crossings at about 100 parsecs, molecular clouds at about 10 parsecs, and filaments below a parsec—must be covered by several resolution elements, requiring sub-parsec effective resolution to be captured at all. Applying this criterion separates physical tides from numerical noise in simulation derivatives.","core_discovery":"The paper's most consequential assertion is that the rapidly varying component of the tidal field—the shocks a cluster experiences when crossing a galactic disk or encountering molecular clouds—cannot be resolved in any current cosmological simulation, including zoom-in runs. Since these shocks inject the energy that allows stars to escape, a simulation that cannot resolve structures down to roughly a parsec cannot predict cluster mass-loss or dissolution. The paper therefore concludes that the evolution from the unknown initial cluster mass function to the present-day one remains out of reach of current simulations, and that apparent tidal shocks measured at several-hundred-parsec resolution are numerical noise from differentiating a coarse potential. Alongside this, the paper argues that young massive clusters are not proven local analogues of young globular clusters and that the cluster formation rate is not tied to the star formation rate.","pith_inferences":["A direct corollary the paper leaves implicit is that comparing observed cluster populations to cosmological simulations yields conclusions only as strong as the resolution of the tidal field; until convergence is shown, apparent agreements may be coincidental.","The resolution argument suggests that improving gas resolution in simulations should change predicted cluster mass-loss rates, and likely increase them as more small-scale shocks are captured; this is testable by resolution studies.","The CFR-SFR decoupling, if real at high redshift, has implications beyond clusters: the environments that produce dense stellar systems may not trace star formation, so using star formation to predict gravitational-wave merger rates from cluster cores could be biased.","The paper's logic could be extended to observational surveys: the universality of the globular cluster mass function is consistent with an equilibrium, not a universal formation process, so survey data should be interpreted with an explicit evolution model before claiming a common origin."],"forward_implications":["Predictions of globular cluster survival, destruction rates, and the evolution of the cluster mass function from current galaxy-scale simulations should be treated as provisional until the tidal field is resolved to sub-parsec scales.","If the initial cluster mass function is environment-dependent rather than a universal power law, then present-day globular cluster mass functions record a balance between formation conditions and evolutionary losses, not a single universal initial condition.","Using star formation rates or galaxy interaction stage to infer cluster formation rates can mislead, particularly in gas-rich, high-redshift or merging systems where the two rates decouple.","A cluster evolution model that includes a properly resolved time-varying tidal field could be used to test formation hypotheses across cosmic time, but none exists yet."],"supporting_citations":[{"why":"This is the galaxy-scale simulation the paper argues cannot resolve tidal shocks at its coarse resolution.","marker":"Pfeffer et al. 2018"},{"why":"This companion paper makes the same cluster evolution claims and is cited as the target of the resolution critique.","marker":"Kruijssen et al. 2019"},{"why":"This work is cited as the best current effort, reaching about 10 parsec resolution and capturing some shock components while still missing the finest structures.","marker":"Li et al. 2017"},{"why":"This thesis is cited as the basis for the rule that quantities derived below the resolution limit are numerical noise.","marker":"Renaud 2010"},{"why":"This work establishes that neglecting shock-like tidal events leads to large errors in cluster mass-loss and survival.","marker":"Gnedin & Ostriker 1999"},{"why":"This work is cited to moderate the claim, noting repeated shocks have a milder effect than single strong shocks.","marker":"Gieles & Renaud 2016"},{"why":"This provides the observed power-law or Schechter form of the young massive cluster mass function used in the globular-analogue argument.","marker":"Adamo et al. 2017"},{"why":"This review is cited for the observed universality of the present-day globular cluster mass function.","marker":"Vesperini 2001"},{"why":"This work is cited for the idea that the initial and evolutionary dependences balance toward an equilibrium universal globular cluster mass function.","marker":"Parmentier & Gilmore 2007"},{"why":"This work is cited for the premise that all stars form in clusters, which underlies the cluster formation rate discussion.","marker":"Lada & Lada 2003"}],"fun_headline_variants":["Simulations can't resolve cluster-killing tides","Cluster formation rate decoupled from star formation","Young clusters aren't ancient analogues","Star cluster simulations hit a parsec wall","Unseen tidal shocks break cluster models"],"cache_read_input_tokens":3200,"weakest_assumption_plain":"The entire resolution critique rests on the rule that a quantity measured with less than a few resolution elements is unphysical numerical noise; if that rule is too strict, coarse simulations could still capture real tidal shocks.","fun_headline_variants_meta":{"raw":{"variants":["Simulations can't resolve cluster-killing tides","Cluster formation rate decoupled from star formation","Young clusters aren't ancient analogues","Star cluster simulations hit a parsec wall","Unseen tidal shocks break cluster models"]},"model":"deepseek-v4-flash","effort":"low","cost_usd":0.000226,"raw_usage":{"total_tokens":1452,"prompt_tokens":911,"completion_tokens":541,"prompt_tokens_details":{"cached_tokens":384},"prompt_cache_hit_tokens":384,"prompt_cache_miss_tokens":527,"completion_tokens_details":{"reasoning_tokens":478}},"tokens_in":527,"tokens_out":541,"duration_ms":6338,"temperature":1.0,"reasoning_tokens":478,"cache_read_input_tokens":384,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-08-14T14:48:14.852204+00:00","model_set":{"reader":"deepseek-v4-flash"},"falsifier":"Take one galaxy formation simulation and run it at both several-hundred-parsec and sub-parsec gas resolution with identical initial conditions; if the coarse run reproduces the same cluster mass-loss statistics as the fine run, the central claim fails, while a difference in tidal shocks would support it.","supporting_citations":[{"cited_title":null,"cited_arxiv_id":null,"evidence_quote":"This companion paper makes the same cluster evolution claims and is cited as the target of the resolution critique."},{"cited_title":"Dynamics of the Tidal Fields and Formation of Star Clusters in Galaxy Mergers","cited_arxiv_id":"1008.0331","evidence_quote":"This thesis is cited as the basis for the rule that quantities derived below the resolution limit are numerical noise."},{"cited_title":null,"cited_arxiv_id":null,"evidence_quote":"This work establishes that neglecting shock-like tidal events leads to large errors in cluster mass-loss and survival."},{"cited_title":"& Renaud, F","cited_arxiv_id":null,"evidence_quote":"This work is cited to moderate the claim, noting repeated shocks have a milder effect than single strong shocks."},{"cited_title":"& Gilmore, G","cited_arxiv_id":null,"evidence_quote":"This work is cited for the idea that the initial and evolutionary dependences balance toward an equilibrium universal globular cluster mass function."}],"review_version":1}