{"id":"76b7cf27-5b6b-4856-aa0b-9b93df1ccba4","arxiv_id":"2412.13483","paper_version":1,"verdict":"CONDITIONAL","confidence":"MODERATE","novelty_score":6.0,"correctness_risk":"medium","formal_verification":"none","parameter_count":4,"one_line_summary":"Using the CIELO simulations, the authors find that stellar halos are dominated by stars accreted from satellites, with metallicity rising with halo mass and the alpha-to-iron ratio tracing satellite star formation histories.","lead":"This paper uses 28 simulated galaxies to show that stellar halos are built mostly from shredded smaller galaxies, and that the chemistry of halo stars records how and when those galaxies were swallowed. A generalist reader might care because it offers a way to read a galaxy's violent assembly history from the light of its faintest stars.","discovery_kind":"extension","skeptic_critique":{"model":"deepseek-v4-flash","headline":"Population classification depends on snapshot cadence and SUBFIND; the claimed channel fractions and chemical offsets may partly reflect misassignment at infall or disruption.","rationale":"The reader's weakest assumption is the same one I identify: the particle-level classification into in-situ, endo-debris, and ex-situ is the foundation for nearly every headline result, and it is not quantified. I would add that the error is likely directional rather than symmetric noise: coarse snapshots and SUBFIND failures preferentially blur ex-situ into endo-debris and stripped gas into in-situ, which makes the concern harder to dismiss as scatter. The paper's central quantitative statements—median fractions near 50/40/10%, the claimed mass-independence of those fractions, and the chemical ordering with ex-situ stars older and more alpha-rich than endo-debris stars—are all measured on the classified populations, so an unquantified misassignment rate propagates into the mass-metallicity and [O/Fe]-[Fe/H] relations as well. Section 3.2 discloses the difficulty but provides no test of its magnitude. I do not see an internal inconsistency; the issue is a robustness gap rather than a demonstrated error. A targeted high-cadence rerun, or even an analytic bound on the stellar mass formed within one snapshot cadence of a crossing/disruption event, would settle whether the concern lands. Absent such a check, CONDITIONAL remains the appropriate verdict: the claims are plausible and consistent with earlier Auriga and Milky Way work, but the key classification assumption is not yet secured.","tokens_in":26593,"tokens_out":5599,"duration_ms":54363,"concrete_test":"Rerun one low-mass and one high-mass CIELO halo with snapshots saved at ~10 Myr cadence over the redshift intervals containing their main infall and disruption events, then reapply the exact Sec. 3.2 definitions to the same particles. If the ex-situ/endo-debris/in-situ mass fractions shift by more than ~10% (relative) or the median [O/Fe] separation between ex-situ and endo-debris stars shifts by more than ~0.05 dex, the snapshot-cadence/SUBFIND assumption is driving the central chemical-channel claims. Comparing the original and high-cadence classifications directly isolates the misassignment rate.","verdict_should_be":"UNCHANGED","load_bearing_attack":"Central claims—dominant accreted material, mass-independent population fractions, and distinct [O/Fe]-[Fe/H] for ex-situ vs endo-debris—rest entirely on assigning each halo star to a formation channel in Sec. 3.2. The assignment uses the closest snapshot after birth (median 1.4e8 yr) and SUBFIND to decide whether the birth gas/star was bound to the central galaxy or a subhalo. The paper itself notes (Sec. 2, Fig. 2) that SUBFIND fails to identify substructures when they are 'too fluffy' or close to the central galaxy, and that merger stages are hard to disentangle. For a star formed inside a satellite just before it crosses rvir, the next snapshot may already place it inside rvir, turning an ex-situ star into endo-debris; for gas stripped just before a snapshot, the first snapshot may show it bound to the central, turning an endo-debris star into in-situ. The claimed median fractions (ex-situ ~50%, endo-debris ~40%, in-situ ~10%) and the age/[O/Fe] ordering of populations are exactly the quantities these errors would bias. The uncertainty is acknowledged qualitatively but never quantified, so the load-bearing assumption is unsecured.","agreement_with_reader":"agree"},"referee_report":{"model":"deepseek-v4-flash","summary":"The paper uses 28 zoom-in cosmological hydrodynamical simulations from the CIELO project to study the formation channels of outer stellar halos (defined as the region between 1.5 r_opt and r_vir) in galaxies with stellar masses 10^9–10^11 M_sun. Stellar particles in the halo are classified as in-situ, endo-debris, or ex-situ by tracing each particle to the closest snapshot after its birth and determining whether it is bound to the central galaxy or to a SUBFIND subhalo. The authors report that accreted material dominates the halos (~50% ex-situ, ~40% endo-debris, ~10% in-situ), that these fractions do not correlate with halo mass, and that the [Fe/H] of each population increases linearly with halo mass (Eq. 1). They further analyze satellite accretion histories, showing that more massive halos require more satellites to build 90% of their accreted mass, and they examine the [O/Fe]-[Fe/H] plane (Eq. 2), arguing that it is more sensitive to the star formation histories of contributing satellites than the mass-metallicity relation.","tokens_in":26947,"tokens_out":10536,"duration_ms":85103,"significance":"If confirmed, the paper's main results would be useful for interpreting stellar halo observations in the era of large spectroscopic surveys. The CIELO sample spans a decade and a half in galaxy mass, and the analysis resolves the stellar halo into ex-situ, endo-debris, and in-situ components with distinct ages, metallicities, and alpha abundances. The claim that the [O/Fe]-[Fe/H] plane is more sensitive to the SFH of contributing satellites than the stellar halo mass-metallicity relation is a falsifiable prediction that can be tested with future data. The paper is transparent about the MZhR being inherited from the galaxy MZR, and it provides detailed tables and a clear account of the definitions used. Its principal limitations at this stage are the unquantified sensitivity of the population decomposition to snapshot cadence and SUBFIND completeness, the small number of high-mass halos, and the lack of uncertainties on the fitting relations.","major_comments":[{"comment":"The central population classification is not robustly established. Section 3.2 assigns each stellar particle to a formation channel using the closest snapshot after birth (median cadence 1.4e8 yr, Sec. 3.1) and SUBFIND membership, yet Section 2 documents that SUBFIND misses substructures that are 'too fluffy' or close to the central galaxy. A star formed in a satellite just before infall can therefore be classified as endo-debris rather than ex-situ, and stars born from recently stripped gas can be classified as in-situ rather than endo-debris. The reported median fractions (~50% ex-situ, ~40% endo-debris, ~10% in-situ; Sec. 3.2, Fig. 5) and the age and [O/Fe] ordering of populations (Figs. 6 and 11) are exactly the quantities this misassignment would bias. The paper acknowledges the difficulty qualitatively, but it does not quantify it. Please add a sensitivity test: for example, re-run the classification using the previous snapshot instead of the closest one, or restrict to stars whose birth gas is unambiguously bound to a subhalo in both adjacent snapshots, and report the resulting range of population fractions and chemical offsets. Also test the sensitivity to the adopted 100-stellar-particle satellite cut (Sec. 4.2) by varying it by a factor of a few and quoting the change in the median chemical trends.","section":"Sections 2 and 3.2"},{"comment":"The high stellar halo mass bin (M_halo > 10^9.5 M_sun) contains only three galaxies in Table 1: LG1-4337, P7-2389, and P7-7805. The statements in Sec. 4.2 that massive halos need a median of five additional satellites to build 90% of their accreted mass, and in Sec. 5.2 that high-mass halos have higher [O/Fe], rest on medians over these three systems. The paper already notes caution for the [O/Fe] trend, but the quoted numbers in Table 2 and the text are still presented without uncertainty. Provide bootstrap or jackknife confidence intervals for the high-mass medians, and either restrict the mass-dependence claims to the range adequately sampled (M_halo < 10^9.5 M_sun) or clearly flag the high-mass results as tentative.","section":"Table 1 and Fig. 5"},{"comment":"The four linear regressions in Eq. (1) are quoted to two decimal places without slope/intercept uncertainties, scatter, or a statement of the fitting method. The slopes (0.23–0.29) are mutually consistent within plausible errors, so the claim that each stellar population defines a distinct MZhR is not yet supported. Report the best-fit parameters with uncertainties and test whether the slopes differ significantly across populations. The same applies to the regressions in Eq. (2) and to the SHMC mass fit in Sec. 4.2 (log10 M_sat = 1.10 log10 M_halo − 1.26).","section":"Sec. 5.1, Eq. (1)"},{"comment":"Eq. (2) writes [O/Fe]_pop = a [Fe/H]_halo + b for each population, using the halo median [Fe/H] as the independent variable, while Fig. 14 is described as showing the medians of [O/Fe] versus [Fe/H] per galaxy for the whole stellar halo and each population. If the x-axis in Fig. 14 is the population's own [Fe/H], then Eq. (2) is mislabeled and the fitted slopes do not correspond to the plotted relation; if the x-axis is the halo's [Fe/H], then the slopes do not measure population-specific [O/Fe]-[Fe/H] trends and the claim that the endo-debris relation is steeper than the ex-situ one (Sec. 5.2) is unsupported. Please correct this inconsistency and provide uncertainties for the fits.","section":"Sec. 5.2, Eq. (2) and Fig. 14"}],"minor_comments":[{"comment":"The text contains a typo: 'suchs as' should be 'such as' in the sentence introducing LAMOST and the other surveys.","section":"Section 1"},{"comment":"The disk definition includes the condition 'r ≤ 2rropt'; this should read 'r ≤ 2 ropt' since r_opt is defined just above.","section":"Section 3.1"},{"comment":"The cross-reference 'In Table 4.2, we summarize...' points to a table that is actually labeled 'Table 2.' Please correct the numbering.","section":"Section 4.2"},{"comment":"The acronym 'SMHCs' is used in the sentence about the stellar mass fraction with [O/Fe] < 0.1 dex; the paper earlier defines the acronym as 'SHMC' (stellar halo main contributor), so the spelling should be made consistent.","section":"Section 5.2"},{"comment":"The caption states that SHMC1 and SHMC2 are shown as black and gray contours, while the text mentions 'dashed-black line' and 'dashed-dot gray line'; please make the line styles consistent between the caption and the text.","section":"Figure 11"},{"comment":"The percentile superscript/subscript notation in phrases such as 'median age of 12 12.44 11.77 Gyr' is typeset incorrectly in the arXiv version; please format these as median with 25th–75th percentiles in a clear way.","section":"Section 3.2.1"},{"comment":"Because the galaxy MZR is part of the model that CIELO was built to reproduce (Sec. 2), the stellar halo MZhR should be framed explicitly as an expected consistency check rather than an independent prediction; the last paragraph of Sec. 5.1 makes this point, but the abstract and conclusions should not present the MZhR as an unqualified new result.","section":"Sec. 5.1 and Abstract"},{"comment":"The statement that the mass fractions of in-situ, endo-debris, and ex-situ stars are 'independent of stellar halo mass' is stronger than the evidence supports; with N=28, non-significant Spearman p-values only show that no correlation is detected, so the wording should be softened to something like 'no significant trend is found within this sample.'","section":"Sec. 3.2, Fig. 5"}],"recommendation":"major_revision","confidential_remarks":"The manuscript depends on a companion simulation paper (Tissera et al. 2024, submitted) for the description of CIELO and its reproduction of the galaxy MZR; the editor should ensure that the companion is available for review so that the MZhR inheritance claim can be checked. The paper fits the scope of A&A. The main scientific value lies in the population-resolved chemical analysis; the mass-independence claim and the high-mass-bin trends need to be hardened before publication."},"author_rebuttal":null,"desk_editor":{"model":"deepseek-v4-flash","letter":"Here's my take. The paper does something concrete: it takes the established AM-E decomposition and population classification from the Tissera/Monachesi lineage and applies it to 28 CIELO zoom-in galaxies spanning 1e9–1e11 Msun, explicitly pulling out endo-debris as an independent channel. That combination is new. The headline results—accreted material dominates, population mass fractions are roughly flat with halo mass, and the [O/Fe]-[Fe/H] plane is more sensitive to satellite SFH than the mass-metallicity relation—are clearly presented and consistent with earlier Auriga and Milky Way work. I believe them.\n\nThe soft spots are real but mostly minor. The MZhR is partly inherited from the galaxy MZR the simulations were tuned to reproduce, so it is not a new prediction. The linear fits come without slope/intercept uncertainties, and the high-mass bin has three galaxies. The population classification depends on snapshot cadence (median 1.4e8 yr) and SUBFIND performance; the paper acknowledges this in Section 3.2 but does not quantify the impact on the reported fractions or chemical offsets. The stress-test concern about misassignment at infall/disruption is plausible. I would like to see a resolution or cadence test, or at least a conservative error estimate, before using the endo-debris/ex-situ split quantitatively. But the qualitative ordering—ex-situ oldest and most alpha-rich, endo-debris younger and less alpha-rich—is robust to modest misclassification and matches physical expectations.\n\nThe authors are honest about limitations, explicitly discussing SUBFIND's difficulty with fluffy structures and merger stages. No code or data is released, and the simulation suite is described in a companion paper still under review, so external reproducibility is limited. That is typical for this subfield, but it does slow down verification.\n\nWho is this for? People building empirical stellar halo models or interpreting resolved-star surveys (GALAH, 4MOST) will find the scaling relations and satellite-count statistics useful. I would cite it. It deserves a serious referee; the main asks would be to quantify classification systematics, publish fits with errors, and ideally release or share the relevant simulation data. I would send it to review.","headline":"A useful, honest extension of stellar halo scaling relations across a broad mass range; the endo-debris split is new and the paper deserves serious review despite unquantified classification systematics.","tokens_in":696,"tokens_out":890,"would_cite":true,"duration_ms":32836,"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":"Outer stellar halos are built mostly from the shredded remains of infalling satellite galaxies, and their chemical abundances record the assembly history of the host galaxy.","keywords":["Galaxies: abundances","Galaxies: formation","Galaxies: halos","Galaxies: star formation","stellar halo assembly","mass-metallicity relation","alpha enhancement"],"falsifier":"Re-running one CIELO galaxy with snapshots saved ten times more frequently, and comparing the birth-site label of every stellar particle, would settle the classification question: if more than a few percent of particles switch categories, the reported population fractions and their mass independence would not hold.","tokens_in":26441,"feed_emoji":"🌌","tokens_out":12127,"duration_ms":101683,"temperature":0.7,"pith_summary":"This paper argues that the outer stellar halos of galaxies are assembled almost entirely from material that fell in from smaller galaxies, and that the chemical composition of that material records how the assembly happened. Using 28 galaxies from the CIELO cosmological simulations, with stellar masses between $10^9$ and $10^{11}\\,M_\\odot$, the authors trace every halo star back to its birth site and sort it into three channels: stars born in the main galaxy, stars born from gas that belonged to a satellite already inside the host's halo, and stars born in a satellite before it fell in. They find that accreted stars of the last two types together make up more than 80 percent of the outer halo mass, that the mass fractions of the three channels do not depend on halo mass, and that the median metallicities of all three channels rise linearly with halo mass. The paper's central proposal is that the $[\\mathrm{O/Fe}]$-$[\\mathrm{Fe/H}]$ plane carries more information about the star formation histories of the contributing satellites than the mass-metallicity relation alone, making halo chemistry a practical archive of galaxy assembly.","feed_headline":"Stellar halos are mostly shredded satellite galaxies","feed_subtitle":"28 simulated galaxies show halo chemistry can reveal how each galaxy assembled.","key_machinery":"The argument rests on tracing each stellar particle back to the closest simulation snapshot after its birth and labeling it by formation site: in-situ stars were bound to the central galaxy, ex-situ stars were born in a subhalo outside the virial radius, and endo-debris stars formed from gas bound to a subhalo inside the virial radius and later stripped. These labels are assigned within halos identified by the AM-E method, which combines angular momentum and binding energy to separate bulge, disk, and halo. The chemical probe that carries the interpretation is the $[\\mathrm{O/Fe}]$-$[\\mathrm{Fe/H}]$ plane, where oxygen is produced mainly by short-lived type II supernovae and iron mostly by delayed type Ia supernovae, so a population's position encodes how long and how intensely its host satellite formed stars before being accreted.","core_discovery":"Across 28 simulated galaxies spanning $M_{\\star,\\rm gal}\\in[10^9,10^{11}]\\,M_\\odot$, the outer stellar halo (the region between $1.5\\,r_{\\rm opt}$ and the virial radius) is predominantly accreted: ex-situ stars, born in satellites before infall, and endo-debris stars, born from satellite gas after infall, together provide more than 80 percent of the halo mass, with in-situ stars contributing a median of roughly 10 percent. The mass fractions of the three populations show no significant Spearman correlation with stellar halo mass, whereas the median $[\\mathrm{Fe/H}]$ of each population follows a linear mass-metallicity relation with slopes of 0.23-0.29 dex per decade in halo mass. The paper finds that massive halos need more and more massive contributor satellites (a median of eight satellites to reach 90 percent of the accreted mass, versus 2.5 for low-mass halos), and that the $[\\mathrm{O/Fe}]$-$[\\mathrm{Fe/H}]$ plane separates the populations according to the star formation histories of their source satellites, including the presence or absence of the $\\alpha$ knee. For the authors, this makes the $\\alpha$ plane a more informative assembly-history diagnostic than the mass-metallicity relation.","pith_inferences":["If the mass fractions are truly mass-independent, the scatter in the $[\\mathrm{O/Fe}]$-$[\\mathrm{Fe/H}]$ plane at fixed halo mass becomes a direct observational stand-in for the diversity of assembly histories, more informative than halo mass itself.","The same machinery could be applied to the inner halo, which the paper leaves for a forthcoming study; the prediction would be that the alpha plane there is shaped more by the host galaxy's own disk and bulge evolution than by satellite histories.","For nearby galaxies where individual halo stars can be resolved, the predicted slopes and zero points could be tested with deep imaging of red-giant populations across the mass range, connecting the simulation's chemical fingerprints to observations before large spectroscopic surveys arrive."],"forward_implications":["Observed outer stellar halos should be treated as predominantly accreted material at all masses probed here, so inferences about a galaxy's merger history can be drawn from halo abundances without assuming a dominant in-situ component.","A halo's median metallicity is predictable from its mass via the fitted relations, giving observers a quantitative target with slopes of roughly 0.23-0.29 dex per decade in halo mass for comparing halos.","The number and mass of dominant building blocks scale with halo mass: reaching 90 percent of the accreted mass takes a median of 8 satellites for the most massive halos versus 2.5 for the least massive, and the main contributor mass correlates with halo mass as $\\log M_{\\rm sat} = 1.10\\log M_{\\star,\\rm halo} - 1.26$.","The $[\\mathrm{O/Fe}]$-$[\\mathrm{Fe/H}]$ plane can reveal the star formation history of the building blocks: halos whose satellites quenched before type Ia supernovae enriched them show no alpha knee, while halos built from massive starbursting satellites reach higher $[\\mathrm{O/Fe}]$ at fixed $[\\mathrm{Fe/H}]$."],"supporting_citations":[{"why":"Provides the AM-E decomposition method used to identify stellar halos and their components.","marker":"Tissera et al. (2012)"},{"why":"Origin of the clumpy in-situ channel of halo star formation that the in-situ definition builds on.","marker":"Zolotov et al. (2009)"},{"why":"Identified three stellar halo subpopulations with distinct locations on the [O/Fe]-[Fe/H] plane, the direct precedent for the chemical diagnostic.","marker":"Tissera et al. (2013)"},{"why":"Defines endo-debris stars as formed from gas-rich satellites inside the virial radius, a category used throughout the population analysis.","marker":"Tissera et al. (2014)"},{"why":"Observational stellar halo mass-metallicity relation that the simulated relation is compared against.","marker":"Harmsen et al. (2017)"},{"why":"Simulation baseline for Milky Way-mass halos, including the median number of satellites needed to reach 90 percent of the accreted stellar mass.","marker":"Monachesi et al. (2019)"},{"why":"Attributes the mass-metallicity relation to a dominant accretion event and its scatter to varied accretion histories, the interpretation the paper adopts.","marker":"D'Souza & Bell (2018)"},{"why":"Chemical evolution model linking star formation history to [O/Fe] through supernova enrichment timescales, grounding the alpha-plane sensitivity claim.","marker":"Matteucci & Brocato (1990)"},{"why":"Establishes the locations of dwarf galaxy stellar populations on the [O/Fe]-[Fe/H] plane, used to interpret stars from low-mass accreted satellites.","marker":"Tolstoy et al. (2009)"}],"fun_headline_variants":["Stellar halos are mostly shredded satellite galaxies","Halo chemistry traces stolen stars back to their homes","Simulations show halos built from tidally torn dwarfs","Metal fingerprints expose how galaxies assembled their halos"],"cache_read_input_tokens":3200,"weakest_assumption_plain":"The whole classification depends on correctly deciding, at the moment each star is born, whether the gas it formed from belonged to the main galaxy, to a satellite already inside the host's halo, or to a galaxy outside; when satellites are close together or being torn apart, those decisions can be wrong, and the reported population fractions and chemical trends would shift.","fun_headline_variants_meta":{"raw":{"variants":["Stellar halos are mostly shredded satellite galaxies","Halo chemistry traces stolen stars back to their homes","Simulations show halos built from tidally torn dwarfs","Metal fingerprints expose how galaxies assembled their halos"]},"model":"deepseek-v4-flash","effort":"low","cost_usd":0.000322,"raw_usage":{"total_tokens":1917,"prompt_tokens":1161,"completion_tokens":756,"prompt_tokens_details":{"cached_tokens":384},"prompt_cache_hit_tokens":384,"prompt_cache_miss_tokens":777,"completion_tokens_details":{"reasoning_tokens":693}},"tokens_in":777,"tokens_out":756,"duration_ms":8003,"temperature":1.0,"reasoning_tokens":693,"cache_read_input_tokens":384,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-08-11T13:05:32.374105+00:00","model_set":{"reader":"deepseek-v4-flash"},"falsifier":"Re-running one CIELO galaxy with snapshots saved ten times more frequently, and comparing the birth-site label of every stellar particle, would settle the classification question: if more than a few percent of particles switch categories, the reported population fractions and their mass independence would not hold.","supporting_citations":[{"cited_title":"B., White, S","cited_arxiv_id":null,"evidence_quote":"Provides the AM-E decomposition method used to identify stellar halos and their components."},{"cited_title":"M., et al","cited_arxiv_id":null,"evidence_quote":"Origin of the clumpy in-situ channel of halo star formation that the in-situ definition builds on."}],"review_version":1}