{"id":"cdc0f48d-f730-41b1-b7e9-7523fec91895","arxiv_id":"2501.13317","paper_version":2,"verdict":"CONDITIONAL","confidence":"MODERATE","novelty_score":6.0,"correctness_risk":"medium","formal_verification":"none","parameter_count":1,"one_line_summary":"For field galaxies below Milky Way mass, accreted stellar halos are predicted to be diffuse, often compact, and generally subdominant, with the accreted component overtaking in-situ stars only far outside the galaxy.","lead":"This paper predicts how many stars in small galaxies were stolen from shredded companion galaxies, and where those stolen stars end up. The results give upcoming ultra-deep surveys a concrete target: M33-like galaxies should show a transition to accreted stars at about 15 kiloparsecs, and some Milky-Way-sized dark halos should hide surprisingly dim galaxies.","discovery_kind":"new_application","skeptic_critique":{"model":"deepseek-v4-flash","headline":"The predicted ~15 kpc transition radius in conclusion (v) depends on the unmodeled spherical-exponential shape of the in-situ component; replacing it with a realistic disc may shift the crossover and the quoted surface-density range.","rationale":"The paper is a careful extension of particle tagging to low-mass field galaxies, and its central value is the concrete, testable prediction about where accreted stars dominate the projected density. The model's primary approximation is the f_mb tagging prescription, which sets both the in-situ sizes and the phase-space distribution of new stars. The in-situ component is spherical exponential by construction, as the paper openly states. The transition radius in conclusion (v) is a direct comparison of the projected in-situ and accreted profiles, so the assumed morphology of the in-situ component is not a minor detail: it sets the baseline against which the accreted density is measured. Section 5.1.2 concedes that the model may be too simplistic to draw firm conclusions about accretion-related structure, yet the paper does not quantify how the ~15 kpc crossover would change for a disc-like in-situ component. This is the weakest link for the headline number. Other concerns, such as baryonic effects on satellite disruption, are mitigated by the paper's comparisons to Auriga and TNG (Appendix D) and by the resolution of COCO. The f_mb parameter itself is calibrated to the size-mass relation, and varying it within the allowed range does not change the accreted profile (Section 5.3), so it is less critical for the transition. The failed Milky Way population is model-dependent but affects selection rather than the profile at fixed virial mass. The proposed test directly addresses the unquantified morphology dependence and would settle whether the headline prediction survives a more realistic in-situ disc. Since the reader already assigned CONDITIONAL, this concern does not change the verdict, but it sharpens the condition: the prediction should be accompanied by a quantitative assessment of the disc-morphology assumption.","tokens_in":51573,"tokens_out":7715,"duration_ms":73635,"concrete_test":"Using the public stellar particle data (Appendix A), select M33 analogues (3×10^9 < Mstar/Msun < 7×10^9) and compute the median projected accreted density profile. Replace the in-situ component with a thin exponential disc of the same total stellar mass and scale length matched to M33 (h_R ≈ 2 kpc), projected both face-on and edge-on, and recompute the radius where accreted density exceeds in-situ density (and where it is 10× higher). If the crossover radius shifts by more than ~3 kpc from 15 kpc, or the quoted log10 surface-density range shifts by more than ~0.5 dex, conclusion (v) is not robust to the in-situ morphology assumption.","verdict_should_be":"UNCHANGED","load_bearing_attack":"The quantitative prediction in conclusion (v)—accreted surface density exceeds in-situ at ≈15 kpc with log10 Sigma in [3.5,6]—is computed from in-situ profiles that are spherical exponentials, not discs (Section 2.3.3). Section 5.1.2 explicitly acknowledges this is too simplistic to draw firm conclusions about central accretion structure. The transition radius is defined by comparing projected in-situ and accreted densities (Section 5.1.1); a real disc in-situ component has a different projected surface density at 15 kpc (lower by factors of ~2–3 in the plane for a scale length h_R≈2 kpc), so the crossover radius and the surface density at crossover will shift. The paper does not quantify this sensitivity for the M33-like sample. Because the headline claim is a precise radius and density range for observers, the spherical-vs-disc assumption is the most load-bearing unquantified simplification.","agreement_with_reader":"partial"},"referee_report":{"model":"deepseek-v4-flash","summary":"This paper uses the GALFORM semi-analytic galaxy formation model applied to the COCO zoom N-body simulation, together with the STINGS particle-tagging technique, to predict the properties of accreted stellar halos in field galaxies with present-day virial masses between 10^8 and 10^12 solar masses. After calibrating the single free tagging parameter f_mb = 3 per cent against the observed size-mass relation, the authors compare their model to the SMHM relation, TNG, Auriga, and observational data for M33 and other low-mass galaxies. The main quantitative claims are that accreted mass fractions are typically below 10 per cent in the least massive star-forming halos, that the radial scale lengths of accreted halos are comparable to or smaller than the in situ components at low masses, and that at the M33 stellar-mass scale the accreted surface density exceeds the in situ surface density at about 15 kpc, with surface densities in the range 3.5 < log10(M_star/Msun kpc^-2) < 6. The paper also highlights a population of 'failed Milky Ways' and makes a subset of the stellar particle data publicly available.","tokens_in":51839,"tokens_out":4709,"duration_ms":59102,"significance":"If the quantitative predictions hold, this paper provides a useful and testable extension of stellar-halo modelling to the dwarf-galaxy regime, a range that has been comparatively little explored with self-consistent cosmological methods. The headline prediction of a transition radius around 15 kpc and the associated surface-density range is falsifiable with current and upcoming ultra-low-surface-brightness surveys, and the comparison with M33 data is a genuinely useful synthesis. The paper's strengths include its explicit validation against the SMHM relation, size-mass relation, Auriga, TNG, and Dragonfly data; its transparent discussion of resolution limits and model dependence; and the public release of stellar particle data. The principal weakness is that the quantitative transition-radius prediction rests on a single value of f_mb calibrated to sizes and on a spherical exponential in-situ component that is acknowledged to be an oversimplification; the sensitivity of the quoted numbers to these assumptions is not quantified.","major_comments":[{"comment":"The headline prediction of a transition radius near 15 kpc and the quoted surface-density range is computed using in-situ profiles that are spherical exponentials by construction (Section 2.3.3), and the paper itself acknowledges in Section 5.1.2 that this is too simplistic to draw firm conclusions about central accretion structure. The transition radius in Fig. 7 is defined as the radius where the projected accreted density exceeds the projected in-situ density, so replacing the spherical exponential with a realistic thin exponential disc changes the projected in-situ density at 15 kpc by a factor of order 2-3 for plausible scale lengths, and therefore shifts the crossover radius and the surface density at crossover. The paper does not quantify this sensitivity for the M33-analogue sample. Because the abstract and conclusion quote these numbers as precise predictions for observers, this sensitivity should be quantified, for example by applying a simple disc-projection correction or by varying f_mb, before the headline numbers can be taken at face value.","section":"Sections 2.3.3, 5.1.1, 5.1.2; conclusion (v)"},{"comment":"The single free parameter f_mb = 3 per cent is calibrated against the size-mass relation in the regime around M33 and Milky Way masses (Figs. 3 and C1), and it directly sets the in-situ scale length. The conclusions about low-mass halos, in particular the claim that accreted fractions are below 10 per cent and that the accreted and in-situ profiles have similar shapes at M200 < 10^11 Msun, rely on extrapolating this calibration to lower masses. Fig. 3 shows that the predicted size relation flattens near the force-softening scale, and the text cautions that galaxies below M* ~ 10^7 Msun may have artificially large sizes. A demonstration that these conclusions are stable to a small grid of f_mb values, or a comparison against resolved dwarf galaxies with reliable sizes in this regime, would substantially strengthen the low-mass portion of the paper's central claim.","section":"Section 3.2; Appendix C; conclusions (i)-(iv)"},{"comment":"The 'failed Milky Way' population is an acknowledged peculiarity of the L16 galform model that is not present to the same degree in TNG, Eagle, or Auriga, and the paper shows that this population has a significant effect on stellar-mass-selected samples. Since the M33-analogue sample in Figs. 7 and 8 is selected by stellar mass, it includes these failed Milky Ways at high virial mass, and the transition-radius statistics and the conclusion (v) are therefore affected by objects whose existence is disputed. The paper does not show how the transition-radius distribution or the quoted density range changes when failed Milky Ways are removed or when the sample is reweighted to an empirical SMHM relation. Given that the existence of this population is one of the paper's more model-dependent results, this is a relevant robustness test for the quantitative claims.","section":"Sections 3.1 and 4.1; Appendix B1; Figs. 7 and 8"}],"minor_comments":[{"comment":"The phrase 'Himass function' should read 'HI mass function'.","section":"Section 2.1"},{"comment":"'halo with viral mass' should read 'halo with virial mass'.","section":"Section 2, first paragraph"},{"comment":"The sentence referring to 'the much steeper locus of S03' is clear, but the earlier reference to 'early types from M15' appears to be a typo; the text in context suggests M13.","section":"Section 3.2"},{"comment":"The legend entry 'Fiducal model (3%)' should be 'Fiducial model (3%)'.","section":"Fig. 3 caption and legend"},{"comment":"The phrase 'the surface density of accreted stars exceeds that of the in situ component at ~15 kpc, corresponding to surface densities 3.5 < log10 ... < 6' is ambiguous because the quoted density range could be read as referring to the accreted component alone rather than the total or the in-situ component at the transition; the text should specify which surface density is being quoted.","section":"Abstract and conclusion (v)"}],"recommendation":"major_revision","confidential_remarks":"The paper is well within the scope of MNRAS and the underlying simulation work is careful and transparent. My main concern is that the quantitative headline numbers, especially the ~15 kpc transition radius and the associated surface-density range, are presented as robust predictions despite resting on the spherical-exponential in-situ approximation and a single f_mb value. This is fixable within the manuscript's scope by adding a sensitivity analysis, so I do not recommend rejection. The 'failed Milky Way' population is also a model-specific feature that should be more carefully separated from the M33-analogue statistics if the quoted numbers are to be used by observers."},"author_rebuttal":null,"desk_editor":{"model":"deepseek-v4-flash","letter":"Colleague,\n\nBottom line: this is a careful, honest extension of STINGS particle tagging to low-mass halos (10^8–10^12 Msun). If you work on stellar halos or ultra-low-surface-brightness surveys, it's worth a serious read. The headline prediction—accreted surface density overtaking in-situ at ~15 kpc for M33-like galaxies—is genuinely testable with ARRAKIHS/Euclid, but I would not hang a proposal on the exact radius without reading Section 5.1.2 twice.\n\nWhat's new: previous STINGS work (C13) stopped at ~10^12 Msun; here they push down to 10^8, resolve star-forming progenitors with 1000+ particles, and make the first systematic predictions for field dwarf accreted halos. The 'failed Milky Ways'—~10^12 Msun halos with M* ~ 10^8.5–9.5—are a sharp, falsifiable prediction. The paper validates the pipeline against the SMHM, size–mass, Auriga, TNG, and the M33 data; Appendix D comparing to Auriga and Dragonfly is genuinely informative. They also ship stellar particle data, which is more than most papers do.\n\nSoft spots: the f_mb = 3% parameter is fit to the size–mass relation and directly sets the in-situ scale lengths, so the 'predictions' about in-situ profiles are partly circular. More importantly, the in-situ component is a spherical exponential, not a disc. The paper admits in Section 5.1.2 that this is too simplistic to draw firm conclusions about central accretion structure—yet the abstract's transition radius and density range come straight from those spherical profiles. A realistic disc would shift the crossover radius and density; the paper doesn't quantify that for the M33 sample. The failed Milky Way population is also specific to the L16 GALFORM model; TNG and Eagle don't produce it, as the paper's own Appendix B1 shows. That doesn't make it wrong—it makes it a constraint on feedback physics—but it shouldn't be sold as a generic ΛCDM prediction.\n\nOverall the accreted component itself is governed by orbital dynamics and matches Auriga well, so the main trends (accreted fractions <10% in dwarfs, scale-length reversal, single-progenitor dominance) are on firmer ground than the transition-radius headline.\n\nRecommendation: send it to a serious referee. Ask for a robustness test of the transition radius under a disc-like in-situ geometry and an explicit statement of how f_mb uncertainty propagates into the quoted surface-density range. The paper is honest about its limits, but the abstract overstates one number.","headline":"Solid extension of particle tagging to low-mass halos; testable predictions, but the transition-radius headline rides on a spherical-in-situ assumption the paper itself flags.","tokens_in":52340,"tokens_out":3184,"would_cite":true,"duration_ms":34534,"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":"Low-mass field galaxies are predicted to have faint, compact accreted stellar halos that overtake the in-situ stars only around 15 kiloparsecs, at surface densities of $10^{3.5}$ to $10^6$ solar masses per square kiloparsec.","keywords":["stellar halos","dwarf galaxies","accreted stellar mass","particle tagging","semi-analytic galaxy formation","N-body cosmological simulation","low surface brightness","M33 analogues"],"falsifier":"Image a few dozen isolated M33-like field galaxies at surface brightnesses near 30 magnitudes per square arcsecond (roughly $10^3$ solar masses per square kiloparsec). If the radius at which an extended, accreted component overtakes the in-situ body is systematically outside 10 to 25 kiloparsecs, or the crossover surface density lies outside the predicted range $10^{3.5}$ to $10^6$ solar masses per square kiloparsec at matched virial mass, the tagging prescription fails at this mass scale. An independent check is the 'failed Milky Way' population: deep surveys of field galaxies around $10^{12}$-solar-mass halos should uncover an excess of M33-stellar-mass galaxies with anomalously bright accreted halos, and not finding that excess would falsify the model's star-formation efficiency at this halo mass.","tokens_in":51413,"feed_emoji":"🌌","tokens_out":19346,"duration_ms":167040,"temperature":0.7,"pith_summary":"This paper sets out quantitative predictions for the faint, diffuse outer stellar halos of field dwarf galaxies — the component built from stars ripped out of smaller satellite galaxies as they merge — across host dark-matter masses from $10^8$ to $10^{12}$ solar masses. The authors combine the GALFORM semi-analytic model of galaxy formation with the COCO cosmological N-body simulation and the STINGS particle-tagging technique, which lets each stellar generation inherit the phase-space distribution of the most-bound dark matter in its halo. The claims are concrete: in the least massive halos that can form stars the accreted fraction is below ten per cent; accreted stars follow profiles as compact as, or more compact than, the in-situ body, so accretion rarely builds bulges in dwarfs; and in M33-like galaxies the accreted surface density crosses that of the in-situ stars at about 15 kiloparsecs, at $10^{3.5}$ to $10^6$ solar masses per square kiloparsec. If these numbers are right, ultra-deep imaging surveys know how faint a dwarf stellar halo must be to be detectable, and a predicted population of 'failed Milky Ways' — $10^{12}$-solar-mass halos with exceptionally inefficient star formation — becomes a count-up target among M33-stellar-mass field galaxies.","feed_headline":"At 15 kpc, accreted stars take over in M33-like galaxies","feed_subtitle":"Deep imaging can verify the predicted crossover brightness and spot a new population of 'failed Milky Ways'.","key_machinery":"The central mechanism is the STINGS particle-tagging method, in which every single-age stellar population formed by the semi-analytic model receives its own set of $N$-body particles drawn from the most-bound fraction $f_{\\rm mb}=3$ per cent of its host dark-matter halo, ranked by binding energy and given equal stellar weight. Because tagged particles diffuse in phase space over time, each population relaxes into a near-exponential surface-density profile whose scale is set by the mass and concentration of the halo at formation, with no baryonic back-reaction on the potential. This one-parameter construction converts the GALFORM merger trees in the COCO simulation into roughly 6400 resolved central galaxies above $10^{10}$-solar-mass halos, and tags every star with its formation time and its branch of the merger tree, so in-situ and accreted stars are separated cleanly. The value of $f_{\\rm mb}$ is calibrated against the observed size-mass relation of low-mass galaxies, and the paper shows the average profile of the accreted component is insensitive to its value.","core_discovery":"The paper claims that at fixed stellar mass the stellar halos of low-mass field galaxies are far more diverse than at fixed virial mass, and that the properties of the accreted component are governed by the virial mass of the host. Its headline result is at M33-like stellar masses: the surface density of accreted stars exceeds that of the in-situ component at roughly 15 kiloparsecs, at a level $3.5 \\lesssim \\log_{10} M_\\star/{\\rm M}_\\odot\\,{\\rm kpc}^{-2} \\lesssim 6$ that varies systematically with virial mass. Around this scale the paper further claims that the scale lengths of accreted halos are smaller than those of the in-situ disk yet the accreted component remains diffuse rather than bulge-like; that in halos below about $10^{11}$ solar masses the accreted density is lower than the in-situ density at every radius accessible to integrated-light observations; and that below roughly $10^8$ solar masses of stars the fraction of galaxies with no significant accreted component rises toward half of the population. It also advances a population statement: a small but significant subset of $\\sim 10^{12}$-solar-mass halos hosts 'failed Milky Ways', galaxies with M33-like stellar mass and exceptionally inefficient star formation whose comparatively bright accreted halos make their abundance measurable by low-surface-brightness surveys.","pith_inferences":["A scale-free version of the paper's central test is implied by its own figures: the radius where accreted stars overtake the in-situ body, divided by the in-situ half-mass radius, varies strongly across the simulated population, so measuring that ratio — which needs no virial masses — would discriminate this model from prescriptions that tie the halo transition to the disk scale length.","The model's neglect of baryonic gravity cuts asymmetrically at this mass scale: the claim that accreted halos are invisible at observable radii in the lowest-mass halos is the fragile half (a disk potential would shred satellites more aggressively than the collisionless treatment), whereas the M33-scale crossover numbers are the robust half, because they are set by dark-matter assembly that the ta","The 'failed Milky Way' population converts a galaxy-formation question into a counting experiment: with a few hundred ultra-deep fields of M33-like dwarfs, the measured space density of galaxies with anomalously bright accreted halos would either confirm or rule out the star-formation efficiency at fixed halo mass assumed by the semi-analytic model used here.","Passing the released star-particle data through a mock-image pipeline that applies a real survey's surface-brightness limits, pixel scale, and sky noise would upgrade the paper's qualitative 'observable halo' statements into countable detection fractions — a direct route from these predictions to a survey strategy."],"forward_implications":["Ultra-deep surveys of M33-mass field galaxies should find the accreted component overtaking the in-situ component near 15 kiloparsecs, with the crossover surface density tied systematically to virial mass.","Below virial masses of about $10^{11}$ solar masses the accreted halo should be effectively undetectable in integrated light at every radius accessible to current surveys.","Classical bulges built by accretion should be rare below Milky Way mass, so observed spheroids in dwarf galaxies should trace in-situ formation (pseudo-bulges) rather than merger-built components.","Near $10^9$ solar masses of stars, halos assembled from several comparable progenitors become rare, and below $10^8$ solar masses the fraction of galaxies with no significant accreted component rises to roughly half at $10^7$ solar masses.","The average metallicity measured at about 30 kiloparsecs should recover the mass of the most massive accreted progenitor to within roughly a factor of ten, with scatter growing toward lower galaxy mass."],"supporting_citations":[{"why":"The earlier application of the same tagging method to Milky Way-mass halos, whose accreted-fraction statistics and density-profile techniques this paper extends to lower mass.","marker":"C13"},{"why":"The validation study that benchmarks the tagging method against hydrodynamical simulations and justifies continuous tagging of every stellar population.","marker":"C17"},{"why":"The semi-analytic galaxy formation model that provides the star formation histories, feedback, and the population of exceptionally inefficient 'failed Milky Ways'.","marker":"L16"},{"why":"The COCO zoom N-body simulation whose halos and merger trees the tagging runs on, setting the resolution floor for the lowest-mass predictions.","marker":"Hellwing et al. (2016)"},{"why":"Establishes the critical virial mass below which halos do not form stars in situ, fixing the low-mass regime of the accreted-halo predictions.","marker":"Benitez-Llambay & Frenk (2020)"},{"why":"The observed size-mass relation used to calibrate the tagging parameter f_mb = 3 per cent.","marker":"Mosleh et al. (2013)"},{"why":"The M33 stellar halo model whose surface-density profile anchors the observational comparison at the ~15 kiloparsec crossover scale.","marker":"Smercina et al. (2023)"},{"why":"The EAGLE simulation's accreted-mass fractions, used to cross-check the model's predictions against a hydrodynamical treatment.","marker":"Proctor et al. (2024b)"}],"fun_headline_variants":["Failed Milky Ways: a new target for low-surface-brightness surveys","Stellar halo crossover at 15 kpc predicted for M33-like galaxies","M33-like galaxies: accreted stars dominate at 15 kpc","Accreted halos show wider diversity at fixed stellar mass","Stellar halos in low-mass galaxies: more diverse than expected"],"cache_read_input_tokens":3200,"weakest_assumption_plain":"The load-bearing assumption is that each new generation of stars simply copies the orbits of the most-bound 3 per cent of its halo's dark-matter particles at birth, with no extra gravity from the galaxy itself, so the in-situ component comes out spherical rather than disk-like and satellite disruption is driven by dark matter alone.","fun_headline_variants_meta":{"raw":{"variants":["Failed Milky Ways: a new target for low-surface-brightness surveys","Stellar halo crossover at 15 kpc predicted for M33-like galaxies","M33-like galaxies: accreted stars dominate at 15 kpc","Accreted halos show wider diversity at fixed stellar mass","Stellar halos in low-mass galaxies: more diverse than expected"]},"model":"deepseek-v4-flash","effort":"low","cost_usd":0.001732,"raw_usage":{"total_tokens":6954,"prompt_tokens":1162,"completion_tokens":5792,"prompt_tokens_details":{"cached_tokens":384},"prompt_cache_hit_tokens":384,"prompt_cache_miss_tokens":778,"completion_tokens_details":{"reasoning_tokens":5698}},"tokens_in":778,"tokens_out":5792,"duration_ms":40834,"temperature":1.0,"reasoning_tokens":5698,"cache_read_input_tokens":384,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-08-10T16:15:45.029163+00:00","model_set":{"reader":"deepseek-v4-flash"},"falsifier":"Image a few dozen isolated M33-like field galaxies at surface brightnesses near 30 magnitudes per square arcsecond (roughly $10^3$ solar masses per square kiloparsec). If the radius at which an extended, accreted component overtakes the in-situ body is systematically outside 10 to 25 kiloparsecs, or the crossover surface density lies outside the predicted range $10^{3.5}$ to $10^6$ solar masses per square kiloparsec at matched virial mass, the tagging prescription fails at this mass scale. An independent check is the 'failed Milky Way' population: deep surveys of field galaxies around $10^{12}$-solar-mass halos should uncover an excess of M33-stellar-mass galaxies with anomalously bright accreted halos, and not finding that excess would falsify the model's star-formation efficiency at this halo mass.","supporting_citations":[],"review_version":1}