{"id":"d8b6eb3c-aea4-414b-afad-a4767b2794f4","arxiv_id":"1908.08775","paper_version":3,"verdict":"CONDITIONAL","confidence":"MODERATE","novelty_score":6.0,"correctness_risk":"medium","formal_verification":"none","parameter_count":4,"one_line_summary":"A population synthesis census finds about 1.2e8 single black holes and 9.3e6 binary black holes currently in the Milky Way, with BH-BH merger rates of 3 to 81 per million years.","lead":"This paper simulates the entire population of stellar black holes in the Milky Way, counting about 120 million isolated black holes and 9 million black holes in binary systems. The accompanying open-access catalog is a reference for future gravitational-wave and electromagnetic searches for black holes.","discovery_kind":"extension","skeptic_critique":{"model":"deepseek-v4-flash","headline":"Headline BH counts inherit a scaling factor that appears inconsistent with the stated IMF sample sizes in Sec. 2.1; the quoted simulated masses imply an implausible average mass per generated star and a factor ~3 shift in the census.","rationale":"The reader identified the SFR/metallicity model as the weakest assumption. That is a legitimate external-model concern, but it is at least clearly stated and could be explored with sensitivity runs. The more load-bearing issue I find is internal: the denominator of the scaling that converts simulated events into Galactic numbers appears inconsistent with the stated sample sizes and IMF. If the quoted simulated masses are wrong, every headline number changes proportionally, independent of astrophysical modeling choices. This makes the central claim less secure than the reader's rationale suggests, although the paper's tables do sum consistently to the quoted totals, and the catalog is openly available for checking. The right outcome remains conditional acceptance, with the added condition that the authors verify the simulated-mass normalization and either correct it or demonstrate that the published tables were produced with the correct masses. I do not see grounds for rejection without first running this arithmetic check, because a simple typo in Sec. 2.1 could in principle explain the discrepancy.","tokens_in":25300,"tokens_out":16885,"duration_ms":166017,"concrete_test":"Compute the total IMF mass represented by the stated samples using the Kroupa IMF of Sec. 2.1: for singles, M_sim = N_single * (integral_0.08^150 m xi(m) dm) / (integral_5^150 xi(m) dm), with N_single=5e6; for binaries, convolve the primary mass distribution with the secondary mass ratio q in [0.08/M1,1] and the adopted period/eccentricity distributions. Compare the resulting M_sim values with 7.6e8 Msun and 3.1e8 Msun. Then re-run the Sec. 2.1 linear scaling using the recomputed simulated masses and report the new total single and binary BH counts. If the totals shift by more than about 10%, the catalog normalization must be corrected before the census is used as a quantitative reference.","verdict_should_be":"UNCHANGED","load_bearing_attack":"The central census is obtained by scaling simulated BH counts from the 18 discrete stellar populations to the adopted Galactic stellar masses. The scale factor is the ratio of the real mass to the 'total simulated mass of systems generated for each stellar population' (Sec. 2.1). The text quotes 7.6e8 Msun for single stars and 3.1e8 Msun for binaries, from 5e6 single stars with M>5 Msun and 2.5e6 binaries with primary M>5 Msun. Under the stated Kroupa IMF (alpha1=-1.3, alpha2=-2.2, alpha3=-2.3), one star above 5 Msun represents roughly 48 Msun of total IMF mass including lower-mass stars, so 5e6 single stars correspond to about 2.4e8 Msun, not 7.6e8; the binary estimate is also too high by roughly a factor of 1.7. Equivalently, the quoted single-star simulated mass implies an average of about 152 Msun per generated star, which is not achievable with the stated IMF cutoff at 150 Msun and a steep slope. Since the final numbers scale linearly with the simulated masses, the headline 1.2e8 single BHs and 9.3e6 binary BHs would shift by roughly this factor if the quoted simulated masses are used in the scaling. This is an internal consistency problem, independent of the choice of SFR model, and it directly controls the central claim.","agreement_with_reader":"partial"},"referee_report":{"model":"deepseek-v4-flash","summary":"The paper presents a forward population-synthesis census of stellar-origin black holes in the Milky Way, using the StarTrack code with 18 discrete stellar populations representing the bulge, thin disk, thick disk, and halo under a newly constructed star formation history and metallicity model. It reports about 1.2e8 single BHs with average mass ~14 Msun and 9.3e6 BHs in binaries with average mass ~19 Msun, with component-by-component tables, mass and velocity distributions, orbital parameters, a dark-matter constraint for halo BHs, and current Galactic merger rates for BH-BH, BH-NS, and NS-NS systems. Two common-envelope treatments (A and B) and three merger mass-retention prescriptions are explored.","tokens_in":25635,"tokens_out":12512,"duration_ms":108030,"significance":"If the numerical results are correct, this is a useful reference catalog and a clear step beyond single-component constant-SFR models: the SFR/metallicity model is more realistic, the forward-modeling approach does not fit the headline numbers, the two CE treatments bracket an important uncertainty, and the sensitivity of merger-mass prescriptions is at least partially explored. The public database is a valuable community resource. However, the quantitative census currently rests on a scaling step in Sec. 2.1 that appears internally inconsistent, so the headline numbers cannot be taken at face value until that normalization is corrected and the impact on all tables is assessed.","major_comments":[{"comment":"The total simulated masses quoted in Sec. 2.1 are inconsistent with the stated sample sizes and the adopted Kroupa IMF. For the given slopes (alpha=-1.3, -2.2, -2.3), the mean mass of stars drawn from 5-150 Msun is about 461 Msun; hence 5.0e6 single stars alone have a total initial mass of roughly 2.3e9 Msun, and 2.5e6 binary primaries alone contribute about 1.15e9 Msun before adding secondaries. The text instead gives 7.6e8 Msun and 3.1e8 Msun for the total simulated masses. Because every entry in Tables 2-9 is obtained by scaling simulated BH counts linearly with the ratio of real stellar mass to this simulated mass, the headline numbers (1.2e8 single BHs and 9.3e6 binary BHs) and all component counts are shifted by roughly a factor of three if the quoted simulated masses are used. Please specify how the total simulated masses were computed and re-derive the scale factors and all resulting numbers.","section":"2.1"},{"comment":"The new SFR and metallicity model is a clear improvement over a single-component constant-SFR disk, but the construction relies on hand-drawn discrete episodes with round-number SFRs and metallicities, and the final census scales linearly with the adopted stellar masses. The paper should propagate at least the quoted uncertainties from Licquia & Newman (2015) (disk 5.17 +/- 1.11e10 Msun, bulge 0.91 +/- 0.07e10 Msun) and bracket the effect of the SFR episode decomposition, for example by shifting the 10-12 Gyr bulge formation peak or varying the ten 1 Gyr thin-disk episodes. Without such a bracket, the abstract's 'about 1.2e8' cannot be distinguished from a range that may span 30-50% or more.","section":"2.5"},{"comment":"The merger mass-retention fractions fMS=fHe=0.8 and fG=0.5 are acknowledged as ad hoc, and Sec. 3.5 shows that the number of BHs in the first and second mass gaps depends strongly on them. The highlighted 113 Msun halo BH is produced in the CO+He merger channel and therefore inherits fHe=0.8. Since the f=0 and f=0.5 tests in Sec. 3.5 do not report the maximum single-BH mass, the paper should state the maximum mass and the number of objects above, say, 70 Msun in the alternative retention models; otherwise the abstract's emphasis on 113 Msun overstates the robustness of that endpoint.","section":"2.4/3.5"}],"minor_comments":[{"comment":"The formation channel of the 113 Msun BH is described as 'BH + He' in Sec. 3.3 and in the Abstract, but Conclusion item 3 says 'BH-MS system coalescence'; please correct the inconsistent statement.","section":"Sec. 3.3 / Conclusions"},{"comment":"The model-B total for thick-disk single BHs is printed as 8.5e7, while the row entries sum to about 8.5e6; in the same table, the MS-He row has an empty average mass for model B (19.9/()).","section":"Table 5"},{"comment":"The text states that the average single BH mass in the halo is ~17 Msun, while Table 8 gives 21.0 Msun (model A) and 19.9 Msun (model B); please align text and table.","section":"Sec. 3.5"},{"comment":"The velocity and coordinate model is very simplified, for example a uniform disk height of 0.3 kpc and a flat rotation curve used for both disk and halo; this is acceptable for a rough catalog, but the paper should state explicitly that the reported velocities are not dynamical predictions and should not be used for kinematic studies without caveats.","section":"Sec. 2.6"},{"comment":"The sentence noting 'full agreement' with LIGO/Virgo estimates via Belczynski et al. (2017) is not a comparison performed in this paper and uses the same StarTrack model family; either remove the claim or show the comparison directly.","section":"Sec. 3.10"},{"comment":"There are numerous typographical problems in figure captions and text ('surrvive', 'botton', 'wavelenghts', 'distrupted', 'unenvolved'); a careful proofreading pass is needed.","section":"General"}],"recommendation":"major_revision","confidential_remarks":"The Sec. 2.1 normalization issue is serious and should be checked by the authors before any numerical result is quoted; if the quoted simulated masses are wrong, the census changes by roughly a factor of three and the abstract, conclusions, and all component tables need to be revised. The conceptual framework of the paper is sound and the sensitivity treatments are a strength, so I would not reject, but the scaling arithmetic must be corrected and the resulting numbers re-reported."},"author_rebuttal":null,"desk_editor":{"model":"deepseek-v4-flash","letter":"Useful paper, with an important flaw.\n\nThe genuinely new piece is the three-component SFR and chemical evolution model for disk, bulge, and halo. That is a real improvement over the old single-constant-SFR disk. The open database and the split of results by component and by two CE models give observers something concrete to use, and the paper is transparent about the physical choices. The merger-rate history plots and the BH mass distributions are worth having.\n\nThe soft spot is not the SFR model; it is Section 2.1. They state 5e6 single stars with M>5 Msun per population and a total simulated mass of 7.6e8 Msun. With the stated Kroupa IMF, one star above 5 Msun corresponds to roughly 46 Msun of total mass down to 0.08 Msun. That makes 5e6 stars correspond to about 2.3e8 Msun, not 7.6e8. The binary number has the same problem. If the linear scaling uses the quoted simulated masses, every final count is off by roughly a factor of three. This is independent of the SFR choices and it directly controls the headline 1.2e8 single BHs and 9.3e6 binary BHs. The authors need to clarify what the quoted mass means: either it is the mass of the evolved M>5 stars (then 7.6e8 implies 5e7 stars, not 5e6), or it is the full IMF mass (then it should be about 2.3e8). Either way the catalog needs re-scaling.\n\nOther issues are minor by comparison: no error bars on the headline numbers; the merger mass retention fractions fMS=fHe=0.8 and fG=0.5 are ad hoc, though the paper does test alternatives; the BH-BH merger rate spans a factor of about 25 across the two CE models; Table 5's thick disk total looks like a typo. The self-cited LIGO/Virgo agreement is reasonable but should not be treated as independent validation. Their own removal of MS+He merger BHs above 70 Msun is an honest caveat.\n\nThe central argument—that the Milky Way holds of order 1e8 stellar-origin BHs, mostly single—is probably right and consistent with older estimates. But the specific catalog numbers should not be used as a quantitative reference until the scaling inconsistency is resolved.\n\nI would send it to review: the SFR model and the catalog are worth publishing, and the flaw is fixable. The referee should insist on the re-scaling and a clear description of the normalization, ideally with the code or a precise enumeration of what was simulated.","headline":"Open catalog and new three-component SFR model are genuinely useful, but the headline BH census has an internal IMF scaling inconsistency that shifts the numbers by a factor of a few.","tokens_in":26164,"tokens_out":9087,"would_cite":false,"duration_ms":83387,"reading_group":"yes","serious_thinker":"yes","would_accept_peer_review":true},"rs_alignment":null,"lean_confirmation":null,"pith_extraction":{"msc":[],"pacs":[],"model":"deepseek-v4-flash","headline":"The Milky Way likely hosts about 130 million stellar-origin black holes, mostly single, with the most massive one formed from a low-metallicity binary merger.","keywords":["black holes","Milky Way","population synthesis","star formation history","binary stars","gravitational waves","microlensing","dark matter"],"falsifier":"A microlensing survey sensitive to lenses of 10-100 solar masses toward the Galactic bulge could count dark compact objects; if the measured event rate is far below the predicted billion-scale single-BH population, the census overcounts, and if far above, the census undercounts.","tokens_in":2017,"feed_emoji":"🕳️","tokens_out":3120,"duration_ms":99276,"temperature":0.7,"pith_summary":"This paper tries to establish a census of every stellar-origin black hole in the Milky Way by simulating how single and binary stars evolve in each Galactic component. It reports that the present-day Galaxy contains about 1.2e8 single black holes (average mass ~14 solar masses) and 9.3e6 black holes in binaries (average mass ~19 solar masses), for a total near 130 million. Roughly half of single black holes come from single-star evolution, about 30% from binary mergers, and about 20% from disrupted binaries. The authors also derive current double-compact-object merger rates and bound the halo dark matter that could hide as stellar black holes.","feed_headline":"Milky Way likely holds ~130 million black holes","feed_subtitle":"Simulation: 120 million single black holes plus 9 million in binaries, mostly in BH-BH pairs.","key_machinery":"The central machinery is the StarTrack population-synthesis code, run on 18 discrete stellar populations representing the bulge, thin and thick disk, and halo, each with an assigned star formation rate and metallicity. The new Galactic star formation history and chemical evolution model (Section 2.5) is what makes this census different from older single-component, constant-metallicity estimates. The code combines a Kroupa IMF, binary orbital-parameter draws from observed massive-star binaries, a rapid supernova remnant model, pulsational pair-instability and pair-instability limits, and an explicit merger-product prescription with mass-retention fractions f_MS=0.8, f_He=0.8, and f_G=0.5. Two common-envelope models A and B bracket the uncertain fate of Hertzsprung-gap donors, and all population weights are scaled linearly to the adopted total stellar masses of the Galactic components.","core_discovery":"The paper claims that at the current moment the Milky Way, including disk, bulge, and halo, contains about 1.2e8 single black holes with average mass about 14 solar masses and about 9.3e6 black holes in binary systems with average mass about 19 solar masses. Most binary black holes reside in BH-BH pairs, and the three main single-BH formation channels are single-star evolution (~50%), binary-system mergers (~30%), and disrupted binaries (~20%). The most massive single black hole in the simulation, about 113 solar masses, forms from a BH+helium-star merger in the low-metallicity halo, while the most massive black hole in a binary reaches about 60 solar masses. The paper further limits stellar-origin black holes in the 20-100 solar mass range to at most about 0.006% of the total Galactic halo mass, and estimates present-day Galactic merger rates of roughly 3-81 per Myr for BH-BH, 1-9 per Myr for BH-NS, and 14-59 per Myr for NS-NS systems.","pith_inferences":["Inference (editorial): The same simulation machinery could be reused to predict the Galactic neutron-star and white-dwarf census, and the catalog's spatial and velocity distributions could be tested against future astrometric surveys of dark companions.","Inference (editorial): A microlensing survey sensitive to 10-100 solar mass lenses would provide a direct counting experiment for the predicted single-BH population, since these objects otherwise emit nothing.","Inference (editorial): The gap between mean binary BH mass (~19 solar masses) and mean single BH mass (~14 solar masses) implies that mass-selected search strategies will over-represent binaries, which matters for interpreting future space-based gravitational-wave detections of Galactic binaries.","Inference (editorial): The claimed 0.006% halo bound is conditional on the modeled halo stellar mass of about 2e9 solar masses; a heavier stellar halo would raise the hidden BH mass fraction proportionally."],"forward_implications":["Most Galactic black holes are invisible to current surveys; the census implies that the roughly 20 detected systems form a tiny and likely unrepresentative sample.","If the numbers are right, gravitational-wave detectors should see present-day Milky Way merger rates of about 3-81 BH-BH, 1-9 BH-NS, and 14-59 NS-NS events per Myr.","Stellar-origin black holes cannot make up a significant fraction of the halo's dark matter: at most about 0.006% of total halo mass is hidden in 20-100 solar mass black holes.","About 5% of single black holes and well under 0.001% of binary black holes are moving fast enough to escape the Galaxy, giving an upper bound on ejected black holes.","The most massive local black holes should be found in old, low-metallicity halo environments, with masses above the usual pair-instability gap, up to about 113 solar masses."],"supporting_citations":[{"why":"Supplies the adopted total stellar masses of the bulge and disk, which set the overall scale of the BH census.","marker":"Licquia & Newman (2015)"},{"why":"Provides the age and age-metallicity relations used to define the bulge and halo stellar populations.","marker":"Kobayashi & Nakasato (2011)"},{"why":"Provides the thin-disk star formation history that the authors approximate with ten 1-Gyr episodes.","marker":"Casagrande et al. (2011)"},{"why":"Supplies the thin-disk age-metallicity relation used to raise disk metallicity by 0.1 solar metallicity per Gyr.","marker":"Haywood et al. (2013)"},{"why":"Calibrates the high-mass IMF exponent used to generate single and binary primary masses.","marker":"Kroupa (2002)"},{"why":"Gives the initial orbital period and eccentricity distributions for binary systems, with the period extrapolated by de Mink & Belczynski (2015).","marker":"Sana et al. (2012)"},{"why":"Defines the rapid supernova remnant-mass model that sets which cores collapse to black holes and their masses.","marker":"Fryer et al. (2012a)"},{"why":"Sets the pair-instability and pulsational pair-instability mass limits used to cap black-hole formation and to shift the limit for H-rich envelopes.","marker":"Woosley (2017)"},{"why":"Supplies the merger-product type table used to classify the outcome of every stellar coalescence.","marker":"Hurley et al. (2002)"},{"why":"Establishes the metallicity dependence of stellar-wind mass loss that drives the higher black-hole masses in low-metallicity halo populations.","marker":"Belczynski et al. (2010)"}],"fun_headline_variants":["Simulation: 120M single + 9M binary BHs in Milky Way","Galaxy census: ~130M black holes, mostly in binaries","New mock catalog: 130M black holes in Milky Way, 10% in binaries","Map of 130M black holes: 120M solo, 9M paired in Milky Way"],"cache_read_input_tokens":28288,"weakest_assumption_plain":"The hand-approximated star formation history and metallicity values for each Galactic component—especially the thin disk's ten discrete episodes and the halo's two low-metallicity bursts—are the load-bearing premise; if the real assembly history differs, every black-hole count scales with it.","fun_headline_variants_meta":{"raw":{"variants":["Simulation: 120M single + 9M binary BHs in Milky Way","Galaxy census: ~130M black holes, mostly in binaries","New mock catalog: 130M black holes in Milky Way, 10% in binaries","Map of 130M black holes: 120M solo, 9M paired in Milky Way"]},"model":"deepseek-v4-flash","effort":"low","cost_usd":0.000656,"raw_usage":{"total_tokens":3102,"prompt_tokens":1139,"completion_tokens":1963,"prompt_tokens_details":{"cached_tokens":384},"prompt_cache_hit_tokens":384,"prompt_cache_miss_tokens":755,"completion_tokens_details":{"reasoning_tokens":1872}},"tokens_in":755,"tokens_out":1963,"duration_ms":13340,"temperature":1.0,"reasoning_tokens":1872,"cache_read_input_tokens":384,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-08-14T11:29:20.266559+00:00","model_set":{"reader":"deepseek-v4-flash"},"falsifier":"A microlensing survey sensitive to lenses of 10-100 solar masses toward the Galactic bulge could count dark compact objects; if the measured event rate is far below the predicted billion-scale single-BH population, the census overcounts, and if far above, the census undercounts.","supporting_citations":[{"cited_title":"R., Tout, C","cited_arxiv_id":null,"evidence_quote":"Supplies the merger-product type table used to classify the outcome of every stellar coalescence."}],"review_version":1}