{"id":"5f59c97a-076b-4935-ab56-1e0399cbf208","arxiv_id":"2507.05590","paper_version":1,"verdict":"CONDITIONAL","confidence":"HIGH","novelty_score":5.0,"correctness_risk":"medium","formal_verification":"none","parameter_count":7,"one_line_summary":"Extended tidal structure is identified around 22 Milky Way globular clusters using Pristine photometric metallicities and Gaia astrometry, including six new tentative detections.","lead":"The authors use a synthetic photometric metallicity catalogue plus Gaia astrometry to search for stars stripped from 30 Milky Way globular clusters, finding likely extra-tidal stars around 22 clusters, six of which are new tentative detections. The work tests whether adding a chemistry dimension to proper-motion searches improves detection of faint cluster debris.","discovery_kind":"new_application","skeptic_critique":{"model":"deepseek-v4-flash","headline":"The six new GC extended-structure detections rest on an uncalibrated significance: adaptive smoothing and in-sample polynomial subtraction make the reported sigma levels untested against the null hypothesis. A null-field control experiment is needed before these detections can be trusted.","rationale":"The paper's central claim has two pillars: the density-map overdensities around 22 GCs (six new) and the metallicity consistency of high-probability outer stars. The metallicity consistency is acknowledged by the authors to be partly self-consistent because the membership selection uses the CaHK color-color space, which is metallicity-sensitive. The density maps are therefore the primary independent evidence for extended structure. Those maps are produced by a pipeline whose significance estimates are not calibrated against a null hypothesis. The adaptive smoothing in Eq. (8) introduces spatial correlations; fitting a first-order polynomial to the same map and computing sigma from the residuals makes the significance dependent on the fitting procedure and the effective number of independent knots. The paper's own hedge in Section 5.3 concedes that the overdensities could be residual MW fluctuations. This is precisely the load-bearing weakness: if the null-field false-positive rate is high, the six new detections evaporate, and even the 22 classifications become unreliable. A null-field control is a concrete, inexpensive experiment that would settle the question. It does not require new observations—only running the existing code on control fields. The reader's weakest assumption (smooth field / first-order polynomial) points in the same direction, but the deeper issue is that the significance metric itself is uncalibrated; a null-field test addresses both. Because the authors provide the full membership probability catalogue and appropriately use 'tentative' language, the appropriate verdict remains conditional acceptance pending this control. No change to the reader's verdict is required.","tokens_in":32096,"tokens_out":7742,"duration_ms":93311,"concrete_test":"For each of the 30 GCs, construct a control field centered ~10 degrees away from the cluster at similar Galactic latitude (avoiding known streams and other clusters), and run the identical PGS quality cuts, reference samples (using the target GC's cluster and field references), membership probability assignment, adaptive density map (Eq. 8), first-order polynomial subtraction, and sigma calculation. Record how many control fields produce a residual overdensity above 2 sigma with a position-angle uncertainty sigma_theta < 8 deg (the paper's tidal-tail criterion). If this false-positive rate approaches the claimed detection fraction (22/30, or especially 6/30 new), the detections are not statistically supported.","verdict_should_be":"UNCHANGED","load_bearing_attack":"The central claim—22 GCs with extended structure, six new tentative—depends entirely on the significance of residual overdensities in the 2D density maps. In Section 3, the map is built with an adaptive-knot estimator (Eq. 8) in which Rloc is set by the distance to the P_mem-weighted stars; this produces spatially correlated knot densities. A first-order polynomial is then fitted to those knots (excluding the inner 2 deg) and subtracted, and sigma levels are computed from the RMS of the residuals of that same fit. Because the fit absorbs large-scale field gradients and the knots are not independent, the resulting sigma values do not measure the probability that a residual overdensity arises from a smooth Milky Way field. The authors themselves concede in Section 5.3 that 'regions could just as easily be residual fluctuations in the MW field around the GCs observed.' The tidal-tail/envelope classification also depends on sigma_theta < 8 deg, which is derived from the same uncalibrated maps. No null-field control is run: the pipeline is never applied to a field without a GC. Without such a control, the six new detections—and, to a lesser extent, all 22 classifications—could be false positives from field curvature, halo substructure, or the statistical fluctuations of the adaptive estimator.","agreement_with_reader":"partial"},"referee_report":{"model":"deepseek-v4-flash","summary":"This paper presents a search for extra-tidal structure around 30 Milky Way globular clusters using the Pristine-Gaia-Synthetic catalogue, which provides photometric [Fe/H] estimates from synthetic CaHK photometry combined with Gaia astrometry and photometry. The authors define a probabilistic membership model in proper-motion, colour-magnitude, and colour-colour spaces, using cluster stars inside the tidal radius as a template and field stars outside two degrees as a contrast. They build 2D surface density maps with adaptive smoothing, subtract a first-order polynomial to remove the Milky Way field gradient, and identify 22 clusters with extended structure within 5 degrees, six of which are new tentative detections. They also measure mean [Fe/H] in three radial ranges and find broad consistency, and they compare position angles of the extended structure with eTidals N-body models, particle-spray models, and Galstreams tracks.","tokens_in":32402,"tokens_out":6940,"duration_ms":74145,"significance":"If the detections are secure, this would be one of the largest homogeneous censuses of extra-tidal structure in GCs and would demonstrate the power of adding CaHK-based photometric metallicities to Gaia astrometry for this purpose. The paper is careful about photometric depth and contamination, releases membership probabilities as online material, and makes a useful comparison to dynamical models. However, the two main quantitative claims—the detection statistics of the overdensities and the metallicity consistency of the outer populations—are weakened by an uncalibrated significance estimator and by a selection that uses a metallicity-sensitive dimension, respectively. The six new detections are explicitly tentative, but the paper's headline count of 22 clusters with extended structure depends on the same uncalibrated maps.","major_comments":[{"comment":"The significance of the residual overdensities that underlie the claim of 22 GCs with extended structure (six of them new) is not calibrated against the null hypothesis. The adaptive-knot estimator in Eq. (8) sets Rloc from the distance to the nearest P_mem-weighted stars, producing spatially correlated knot densities; the first-order polynomial is fitted to and subtracted from these same knots, and the reported σ levels are computed from the RMS of the residuals of that fit. No null-field control is run, so the σ values do not measure the probability that a residual overdensity arises from a smooth Milky Way field. The authors' own statement in Section 5.3 that \"regions could just as easily be residual fluctuations in the MW field around the GCs observed\" applies directly to the six new tentative detections (NGC 6101, NGC 6205, NGC 6218, NGC 6541, NGC 6752, NGC 6934). I request that the authors run the full pipeline on null fields (e.g., off-cluster positions with the same footprint and quality cuts) or on randomized P_mem assignments, and report the false-positive rate as a function of the claimed σ threshold.","section":"Section 3, Eq. (8) and Section 4.2, Fig. 4"},{"comment":"The finding that high-probability stars outside the Jacobi radius have metallicities consistent with the cluster is partly circular. The membership likelihood includes a colour-colour space built on the CaHK index, which is a direct photometric metallicity proxy, and the GC reference sample defines the locus in that space. Stars with high P_mem are therefore selected because they lie near the cluster locus in a metallicity-sensitive dimension, so the agreement between [Fe/H]_PGS outside r_J and inside r_t is not an independent test that these stars are stripped cluster members. To support this claim, the authors should either demonstrate that the outer [Fe/H] distribution is narrower or more peaked than what the selection function alone would produce (e.g., by applying the same P_mem cut to a field-only control sample), or explicitly re-frame the metallicity comparison as a consistency check rather than a confirmation of membership.","section":"Section 3, Eqs. (3)-(6); Section 4.1, Table 2"},{"comment":"The split between tidal-tail-like and diffuse-envelope structures is based on the threshold σθ < 8 deg, where σθ is the weighted uncertainty of position angles measured from the same uncalibrated 2D density maps. For the six new tentative detections, this classification is provisional at best; reporting fractions such as \"17 GCs have tidal tail structure and 4 have envelope structure\" as quantitative results gives these provisional classifications more weight than the data support. The authors should either tie the classification to a calibrated detection significance or present it as a morphological description of the candidate maps only.","section":"Section 4.2.1, Fig. 5"}],"minor_comments":[{"comment":"The word \"Specifcally\" is a typo; the sentence should read \"Specifically.\"","section":"Section 2"},{"comment":"The text \"theeTidalsGCs9 project\" contains a typo and should be \"the eTidals GCs project\" or \"the eTidalsGCs project.\"","section":"Section 5, COMPARISON TO MODELS"},{"comment":"The reference list contains a duplicated entry for Massari, Koppelman, and Helmi (2019), with one version garbled as \"r 1/4ap\"; the duplicate should be removed.","section":"References"},{"comment":"The contamination estimate at P_mem > 0.5 is described as the ratio of stars with [Fe/H] consistent with the cluster to all stars present; this does not measure contamination in the usual sense (the fraction of apparent members that are field stars) because both the numerator and denominator are already filtered by a metallicity-sensitive selection, and the wording should be adjusted accordingly.","section":"Section 4.1"},{"comment":"In the criterion for whether a cluster is suitable for 2D density map analysis, \"P Pmem > 5\" should be written as \"Σ P_mem > 5\" to avoid confusion.","section":"Appendix A"}],"recommendation":"major_revision","confidential_remarks":"The paper's main value is as a homogeneous catalogue of candidate extra-tidal stars and as a demonstration of the Pristine-Gaia-Synthetic combination for this science case. The six new detections, however, are not yet statistically supported without a null-field calibration, and the metallicity-consistency claim is partly built into the selection. The scope fits the journal; the authors should either add the requested control experiments or substantially soften the quantitative claims in the abstract and conclusions."},"author_rebuttal":null,"desk_editor":{"model":"deepseek-v4-flash","letter":"The paper does two useful things: it applies a published machine-learning method to a larger, homogeneous sample of 30 globular clusters, and it releases the resulting membership probabilities. The six tentative new detections (NGC 6101, 6205, 6218, 6541, 6752, 6934) are genuinely new in the sense that they are not in the cited prior literature, and the catalogue itself is a valuable community resource.\n\nThe writing is careful and self-aware. The authors repeatedly flag the photometric depth limitation (G~17), they explicitly say that residual overdensities \"could just as easily be residual fluctuations in the MW field,\" and they avoid overclaiming the detections as bona fide. The comparison with eTidals and particle-spray models is thorough and balanced, and the literature engagement is solid without being uncritical.\n\nThe main weakness is exactly where the reader's stress-test lands: the significance of the overdensities is not calibrated against a null hypothesis. The adaptive-knot density estimator, followed by an in-sample first-order polynomial subtraction and sigma computed from the RMS of the same residuals, does not yield a meaningful false-positive probability. A null-field control — running the same pipeline on fields without a cluster — is an obvious and cheap experiment that would substantially strengthen the six new claims. The tail-versus-envelope split using sigma_theta < 8 deg is also arbitrary, though the qualitative dichotomy between tightly and loosely constrained position angles does seem to exist.\n\nA second soft spot is the partly circular metallicity comparison. Since membership itself uses the CaHK index in colour-colour space, finding that extra-tidal high-probability stars have cluster-like metallicities is partly by construction. But the paper does not lean hard on this — it presents the consistency as supporting, not decisive, and it separately quantifies contamination at P_mem > 0.5.\n\nNone of these flaws are fatal, and the authors are not trying to hide them. The work would benefit from a null-field test and a more rigorous significance threshold, but even as-is it is a solid, honest catalogue paper that will be useful to anyone working on GC tidal structure or planning spectroscopic follow-up. I would send it to a serious referee with the expectation of major revisions rather than desk-reject it.","headline":"A careful, honest extension of the KI25 method to 30 globular clusters, but the six new tentative detections rest on uncalibrated significance levels and need a null-field control before they can be fully trusted.","tokens_in":32943,"tokens_out":1701,"would_cite":true,"duration_ms":23457,"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":"Adding photometric metallicities from the Pristine-Gaia-Synthetic catalogue to Gaia astrometry reveals extended structure around 22 of 30 Milky Way globular clusters, six of them new tentative detections.","keywords":["globular clusters","tidal tails","extra-tidal stars","photometric metallicities","Pristine survey","Gaia astrometry","unsupervised machine learning","Milky Way stellar halo"],"falsifier":"Obtain moderate-resolution spectra of the conservative-sample stars beyond the Jacobi radius for the six new tentative detections: if the measured line-of-sight velocities do not cluster around each cluster's systemic velocity, or the spectroscopic metallicities disagree with the photometric values, the claimed extra-tidal structure would be revealed as field contamination rather than stripped cluster stars.","tokens_in":31920,"feed_emoji":"🌌","tokens_out":14303,"duration_ms":129181,"temperature":0.7,"pith_summary":"This paper claims that adding a photometric-metallicity dimension to Gaia astrometry can expose stars tidally stripped from Milky Way globular clusters, even at large angular separations. Applying an unsupervised machine-learning membership classifier to the Pristine-Gaia-Synthetic catalogue around 30 clusters, the authors find 22 clusters with extended structure within five degrees of their centres, six of which are new tentative detections. The high-probability stars beyond the Jacobi radius carry photometric metallicities consistent with the cluster values, supporting the interpretation that they were stripped from the cluster. The result matters because it demonstrates a homogeneous, all-sky route to building candidate lists of extra-tidal stars that upcoming multi-object spectrographs can confirm.","feed_headline":"Photometric metallicities map tidal debris around 22 globular clusters","feed_subtitle":"Pristine's CaHK-based [Fe/H] joins Gaia proper motions to expose stripped stars beyond the Jacobi radius.","key_machinery":"The load-bearing object is the Pristine-Gaia-Synthetic (PGS) catalogue, which supplies all-sky synthetic CaHK-band photometry and photometric $[\\mathrm{Fe/H}]$ for Gaia sources, giving a metallicity-sensitive colour index $(\\mathrm{CaHK}-G_0)-2.5(G_{\\mathrm{BP}}-G_{\\mathrm{RP}})_0$. The membership machinery is an unsupervised k-nearest-neighbour likelihood that combines three parameter spaces — Gaia proper motion, Gaia colour–magnitude, and PGS colour–colour — to assign each star a probability of belonging to the cluster rather than the Milky Way field. A conservative sample ($P_{\\mathrm{mem}} > 0.99$) provides the cleanest extra-tidal candidates, and a variable smoothing radius plus first-order field subtraction turns those candidates into density maps whose position angles are then fit as a function of radius and compared with orbits and N-body models.","core_discovery":"Within five degrees of each of 30 Milky Way globular clusters, the authors combine Gaia proper motions and photometry with synthetic CaHK-band photometric metallicities ($[\\mathrm{Fe/H}]$) from the Pristine-Gaia-Synthetic catalogue. A k-nearest-neighbour likelihood assigns each star a membership probability in proper-motion, colour-magnitude, and colour-colour spaces, and adaptive smoothing of the high-probability stars produces two-dimensional density maps. After subtracting a first-order Milky Way field gradient and excluding eight clusters with incomplete coverage or too few high-probability stars, 22 clusters show extended structure; six of these are new tentative detections. Four clusters present diffuse envelope-like features and the rest show tidal tail-like structure, with position angles broadly consistent with cluster orbits and N-body disruption models. The mean photometric metallicities of conservative high-probability samples outside the Jacobi radius agree with their clusters' metallicities, indicating the stars are likely genuinely stripped.","pith_inferences":["Because PGS reaches only about $G = 17$, the detected tails are likely the bright tips of longer structures; deeper CaHK photometry should reveal that several new tentative detections are longer streams than the current maps show.","If spectroscopic follow-up confirms the six tentative detections at the predicted metallicities, the same classifier could be run on future all-sky narrowband surveys to build a complete census of dissolving globular clusters in the Milky Way.","The disagreement between observed debris orientation and N-body predictions for NGC 1261, NGC 1851, NGC 2808 and NGC 6934 suggests the comparison itself is a probe of the Milky Way potential; fitting these data with flexible potentials might constrain the bar or halo shape more sharply than the paper's fixed-potential comparison does.","The roughly 20 percent contamination fraction at $P_{\\mathrm{mem}} > 0.5$ outside the Jacobi radius implies that membership-probability thresholds alone cannot separate true members from field stars that mimic cluster properties; radial velocities are required before individual stars can be claimed as members."],"forward_implications":["If the central claim holds, photometric metallicities from narrowband CaHK surveys become a general tool for finding stripped stars around globular clusters, not just for Omega Centauri.","The candidate lists for the six new tentative detections (NGC 6101, NGC 6205, NGC 6218, NGC 6541, NGC 6752, NGC 6934) are concrete targets for spectroscopic follow-up that can confirm membership with line-of-sight velocities.","Broad agreement between measured position angles and cluster orbits implies that the morphology of extended structure can be mapped from density maps alone in many clusters, enabling population-wide studies of globular cluster disruption.","The consistency of metallicities inside and outside the Jacobi radius suggests that extra-tidal stars trace the same stellar populations as their progenitors, so chemistry can be used to link debris to parent clusters."],"supporting_citations":[{"why":"Develops the unsupervised membership technique on Omega Centauri that this paper applies to 30 clusters, supplying the likelihood model and the conservative-sample definition.","marker":"KI25"},{"why":"Releases the Pristine first data release and the Pristine-Gaia-Synthetic catalogue of all-sky synthetic CaHK photometry and photometric metallicities.","marker":"N. F. Martin et al. 2024"},{"why":"Defines the Pristine survey and establishes the CaHK narrowband index as a photometric [Fe/H] proxy.","marker":"E. Starkenburg et al. 2017"},{"why":"Provides the Gaia astrometry and photometry that anchor the proper-motion and colour-magnitude dimensions of the membership model.","marker":"Gaia Collaboration et al. 2016"},{"why":"Supplies the cluster proper-motion catalogue used to build reference samples and to integrate orbits for position-angle comparison.","marker":"E. Vasiliev & H. Baumgardt 2021"},{"why":"Supplies the heliocentric distances used to correct proper motions for solar reflex motion and to interpret radial ranges.","marker":"H. Baumgardt & E. Vasiliev 2021"},{"why":"Provides the tidal radii that define the inner reference samples and the radial metallicity bins.","marker":"T. J. L. de Boer et al. 2019"},{"why":"Provides the Jacobi radii used to separate bound, pre-stripping, and stripped star populations.","marker":"E. Balbinot & M. Gieles 2018"},{"why":"Provides the eTidals N-body models whose position-angle profiles are compared with the observed extended structure.","marker":"S. Ferrone et al. 2023"},{"why":"Supplies the photometric-depth selection technique and earlier tidal-tail detections for several clusters, used as method and comparison.","marker":"A. Sollima 2020"}],"fun_headline_variants":["Tidal tails and envelopes found for 22 globular clusters","Photometric metals trace stripped stars around 22 clusters","Six new globular clusters with extended outskirts discovered","Metal-poor stars map tidal debris of 22 Milky Way clusters","Pristine survey spots extra-tidal stars in 22 globular clusters"],"cache_read_input_tokens":3200,"weakest_assumption_plain":"The analysis assumes that the Milky Way field density across each five-degree field is smooth enough to be removed by subtracting a fitted first-order polynomial, so any residual overdensity is attributed to cluster debris; if the field contains real substructure such as other streams or halo clumps, some of the detected excess could be spurious.","fun_headline_variants_meta":{"raw":{"variants":["Tidal tails and envelopes found for 22 globular clusters","Photometric metals trace stripped stars around 22 clusters","Six new globular clusters with extended outskirts discovered","Metal-poor stars map tidal debris of 22 Milky Way clusters","Pristine survey spots extra-tidal stars in 22 globular clusters"]},"model":"deepseek-v4-flash","effort":"low","cost_usd":0.001388,"raw_usage":{"total_tokens":5648,"prompt_tokens":1005,"completion_tokens":4643,"prompt_tokens_details":{"cached_tokens":384},"prompt_cache_hit_tokens":384,"prompt_cache_miss_tokens":621,"completion_tokens_details":{"reasoning_tokens":4558}},"tokens_in":621,"tokens_out":4643,"duration_ms":36440,"temperature":1.0,"reasoning_tokens":4558,"cache_read_input_tokens":384,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-08-06T19:22:25.210825+00:00","model_set":{"reader":"deepseek-v4-flash"},"falsifier":"Obtain moderate-resolution spectra of the conservative-sample stars beyond the Jacobi radius for the six new tentative detections: if the measured line-of-sight velocities do not cluster around each cluster's systemic velocity, or the spectroscopic metallicities disagree with the photometric values, the claimed extra-tidal structure would be revealed as field contamination rather than stripped cluster stars.","supporting_citations":[{"cited_title":null,"cited_arxiv_id":null,"evidence_quote":"Provides the tidal radii that define the inner reference samples and the radial metallicity bins."}],"review_version":1}