{"id":"61e45a8f-b170-439f-b51e-4e59e47343d5","arxiv_id":"2506.21859","paper_version":1,"verdict":"CONDITIONAL","confidence":"MODERATE","novelty_score":5.0,"correctness_risk":"medium","formal_verification":"none","parameter_count":5,"one_line_summary":"Applying X-arithmetic to 15 deeply observed Chandra halos, the authors classify AGN feedback structures and report multiple shocks in groups versus mainly isobaric structures around cavities in massive clusters.","lead":"Astronomers applied X-arithmetic, an X-ray image-combination technique introduced in 2016, to 15 galaxy clusters, groups, and massive galaxies to sort AGN feedback features into shocks, bubbles, and cooling gas. They report that smaller systems show multiple shocks while massive clusters mostly show structures linked to cooling, which could change how feedback is implemented in simulations.","discovery_kind":"new_application","skeptic_critique":{"model":"deepseek-v4-flash","headline":"Headline mass-dependent shock counts rest on visual classification built on linear, single-effective-EOS and midplane-projection assumptions; without a quantitative classifier and mass-matched resolution test the dichotomy is not yet secured.","rationale":"The reader's verdict is CONDITIONAL, and my read supports that rather than moving it. I agree that the linear, single-effective-equation-of-state decomposition and the z=0 projection correction are load-bearing; the method's own mix language shows that projected mixtures produce labels that are not cleanly defined. However, I would place the decisive weight slightly differently: even granting the method's assumptions, the headline trend is a comparison of visually counted features in a sample with only three low-mass objects, no quantitative classification threshold, and an admitted resolution and selection mismatch between groups and clusters. The A2052 spectroscopic checks are important independent support, but they validate three labels, not the mass-dichotomy trend. The concrete test I propose would settle whether the trend survives an objective re-measurement and whether the recovery rate of shocks is mass-independent in simulations; until then, CONDITIONAL is the right verdict, so I recommend no change.","tokens_in":811,"tokens_out":895,"duration_ms":88920,"concrete_test":"Use the released Zenodo X-arithmetic maps to build an objective, threshold-based classifier: for each pixel, compute the significance of the residual in the no-adiabatic and no-isobaric images, assign the type whose suppression map most suppresses the feature, and count connected components per object above a fixed signal-to-noise threshold. Recompute shock counts per mass bin and compare with the paper's visual counts. Separately, run the identical pipeline on mock Chandra/AXIS observations of TNG300 clusters posed at the redshifts and angular resolutions of the group and cluster subsamples, and compare recovered shock counts with the true 3D shock surfaces. If objective counts reproduce the trend and mock recovery is mass-independent, the concern is resolved; if not, the headline dichotomy should be retracted or heavily qualified.","verdict_should_be":"UNCHANGED","load_bearing_attack":"The headline claim - multiple shocks in groups and massive galaxies but only one to two in clusters, and isobaric structures preferentially around cluster cavities - depends on the reliability of every feature label assigned by X-arithmetic. That reliability inherits the assumptions in Eqs. (7)-(9): perturbations are a linear sum of three pure types, density fluctuations are small, and the projection correction in Section 4.3 places each perturbation at z=0. A line-of-sight mixture of adiabatic and isobaric gas produces an intermediate hard-to-soft ratio that is not fully suppressed in any of the three combination images; the paper explicitly sees this in many features (called mix of shocked and isobaric perturbations) but has no quantitative decomposition or criterion for assigning a label. The three spectroscopic checks in A2052 are genuine, but they cover one cluster, not the full sample. The central trend is then a comparison of visual shock counts across mass bins containing only three low-mass systems, with the resolution and selection confound admitted in Section 6.4 but never quantified. Unless the visual labels are shown to be unbiased across mass, the observed dichotomy cannot be distinguished from a detection-threshold or resolution effect.","agreement_with_reader":"partial"},"referee_report":{"model":"deepseek-v4-flash","summary":"The paper applies the 'X-arithmetic' image-manipulation technique, previously introduced by Churazov et al. (2016), to a sample of 15 galaxy clusters, groups, and massive galaxies observed deeply with Chandra. By combining soft- and hard-band residual images, the method classifies X-ray surface brightness perturbations into three physical types: adiabatic (weak shocks and sound waves), isothermal (bubbles), and isobaric (subsonic motions and cooling). The authors report a mass-dependent dichotomy: multiple shocks appear in groups and massive galaxies, but only one to two shocks in clusters, while prominent isobaric structures are abundant around inner cavities in clusters. They present spectroscopic follow-up of three A2052 features, which confirms the X-arithmetic classification for one isobaric feature, one shock, and one 'isothermal shock' that is better described as a mix of adiabatic and isobaric perturbations. They also demonstrate application of the method to a mock AXIS observation of a TNG300 cluster and discuss future observational prospects.","tokens_in":29611,"tokens_out":3759,"duration_ms":48877,"significance":"If the methodological claims are robust, this paper offers an efficient, largely non-spectroscopic route from X-ray morphology to physical perturbation type, and it proposes a potentially important mass dependence of AGN feedback signatures. The paper has notable strengths: the hard-to-soft ratio in Eq. (9) is derived from APEC/AtomDB emissivities without fitting the data to force a classification; three A2052 features receive independent deprojected spectroscopic checks in Section 6.1; the application to a TNG300 mock observation is a useful feasibility demonstration; and the X-arithmetic maps and mock data are publicly released on Zenodo. However, the headline sample-wide trend rests on visual classification of features under linear, small-amplitude assumptions and a midplane-projection correction, and the manuscript itself acknowledges in Section 6.4 that resolution differences between clusters and groups are a possible confound. Because the central claim depends on the reliability of every feature label, the trend is not yet secured without a quantitative classifier and a resolution-matched detection-efficiency test.","major_comments":[{"comment":"The sample-wide shock and isobaric counts in the Abstract and Conclusions rely on assigning each feature a single type from three suppressed images, but the underlying decomposition in Eqs. (7)-(9) is only defined for pure perturbation modes. A line-of-sight or physically mixed feature produces an intermediate hard-to-soft ratio that is not fully removed from any combination image; the paper repeatedly labels such cases as 'mix' but does not define a quantitative criterion for when a feature is called a shock, isobaric, or mixed. I request a per-feature catalog reporting measured w_H/w_S or combination-image residuals, a blinded classification protocol, and an accuracy test on the TNG300 mock where the true perturbation types are known from the 3D density and temperature fields. Without this, the visual labels cannot be shown to be unbiased across mass, which is load-bearing for the central dichotomy claim.","section":"Sections 4, 5 and 6.4; Eq. (9)"},{"comment":"The mass-dependent trend is based on only three low-mass systems (NGC 5813, NGC 5044, M84) compared with twelve clusters, and the physical scales resolved in the inner regions differ substantially. The text explicitly acknowledges that 'we simply lack the resolution to capture the diverse structures within the innermost regions of the most massive systems,' but it does not quantify this detection-threshold or resolution effect. The manuscript should include a resolution-matching exercise: for example, degrade and rebin the group images to cluster-like PSF and pixel scales and re-run the X-arithmetic classification, or inject simulated shock/isobaric features of known amplitude into cluster images to measure recovery fractions. Until this is done, the statement that groups show multiple shocks while clusters show only one to two shocks cannot be distinguished from a selection or resolution bias.","section":"Section 6.4 and Conclusions"},{"comment":"The projection correction in Eq. (14) assumes the perturbation is located at the cluster midplane z=0 and uses a single beta-model atmosphere. This assumption is violated by the strong cocoon shock in Cygnus A, deep cavities, and filamentary structures that extend along the line of sight, as noted qualitatively in Section 5.3. The paper reports stability against model choices but does not test sensitivity to the assumed z0 or to finite line-of-sight extent. I ask for a quantitative sensitivity test: vary z0 over a plausible range for a representative feature, or use simulation slices at different line-of-sight positions to quantify the induced error in the hard-to-soft ratio and the resulting misclassification rate.","section":"Section 4.3, Eqs. (11)-(14)"},{"comment":"The linear small-amplitude expansion stated in assumptions 1-4 after Eq. (7) is the backbone of the method, yet the TNG300 demonstration in Section 6.2 is only qualitative: it shows that X-arithmetic maps resemble density and temperature slices, but it does not measure the actual amplitude of density fluctuations or compare the classified perturbation types against the true 3D decomposition. A quantitative validation using the simulation would strengthen the method considerably: compute the true per-pixel or per-voxel adiabatic/isobaric/isothermal content from the simulated density and temperature fields, and compare with the X-arithmetic labels, especially near strong shocks and cavities where the small-amplitude and single-effective-EOS assumptions are most stressed.","section":"Section 4, assumptions 1-4, and Section 6.2"}],"minor_comments":[{"comment":"The title contains a typo ('F eedback'), and there are numerous small spacing and capitalization errors throughout (e.g., 'T able' in Table 1 and 'T able 2' in the Appendix). These should be corrected in a final proofreading pass.","section":"Title and general text"},{"comment":"The text says 'we assume that the observed perturbation is located at the midline of the gaseous halo z = 0'; the intended term is likely 'midplane' rather than 'midline'.","section":"Section 4.3"},{"comment":"The three-color image construction relies on object-dependent thresholds, and the caption notes that some detections are spurious because only default maps are used. The threshold choice is not described quantitatively; a brief statement of how thresholds are set and how robust the resulting colors are to threshold variation would help reproducibility.","section":"Appendix B, Fig. 16"},{"comment":"The text states that 'many regions appear to be a mix of all perturbations,' but the corresponding figure labels are not always listed in the caption. A consistent labeling convention across all figure captions would make the visual classifications easier to audit.","section":"Section 5.14, NGC 5044"}],"recommendation":"major_revision","confidential_remarks":"The paper builds on the authors' prior work and appropriately cites Churazov et al. (2016). The three A2052 spectroscopic checks are a genuine strength and give me confidence that the method has real diagnostic power. My main concern is that the headline mass-dependent trend is not yet backed by a quantitative classifier or a resolution-bias test; without those, the abstract's strongest claim will be hard for the community to trust. The requested validation is well within the scope of the current simulation and public-data products, so I do not view this as a fatal flaw, but it is essential before publication."},"author_rebuttal":null,"desk_editor":{"model":"deepseek-v4-flash","letter":"Here's my take. The X-arithmetic method is from Churazov et al. (2016), but this paper does new work: it applies the method to 15 halos (including groups and massive galaxies), gives per-object classifications, and then backs up three A2052 features with genuine deprojected spectroscopy. The internal math in Section 4 is transparent and correct, the assumptions are stated rather than buried, and the authors explicitly note in Section 6.4 that the group-versus-cluster trend could be a resolution effect. That level of candor is rare and earns credit.\n\nThe headline result—multiple shocks in groups and massive galaxies, one or two in clusters, with isobaric structures common in clusters and rare in groups—is genuinely interesting and would be a useful target for simulators if it holds. The stress-test note has it right, though: the shock counts come from visual classification of combination images, and the low-mass bin is three nearby objects (M84, NGC 5813, NGC 5044) observed at much better physical resolution than the more distant clusters. The authors admit the resolution confound but do not quantify it. A quantitative classification rule, or a matched-resolution test, would be needed to separate a real dichotomy from a detection-threshold effect. The three spectroscopic confirmations in A2052 are solid but cover a single cluster.\n\nThe method's assumptions—linear superposition of three EOS types, small density fluctuations, a single effective EOS per feature, and the z=0 projection correction—are all stated and are standard for this kind of work. They get pushed hardest for strong shocks (Cygnus A) and for line-of-sight mixtures, where the paper sees 'mixes' but lacks a quantitative decomposition. These are caveats, not fatal flaws.\n\nBottom line: the per-object classifications and the A2052 reclassifications are credible and worth citing, and the simulation and AXIS feasibility demonstration is a nice addition. The mass-dependent trend is a promising hypothesis, not a secure result. This deserves a serious referee: send it out, ask for a less selected and resolution-matched sample or a quantitative classifier, and it should be publishable.","headline":"A solid extension of X-arithmetic to 15 halos with real spectroscopic checks, but the mass-dependent dichotomy rests on visual labels and a three-object low-mass sample, so it is a promising hypothesis rather than a secure result.","tokens_in":30310,"tokens_out":3843,"would_cite":true,"duration_ms":42595,"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 paper claims that AGN feedback leaves a mass-dependent fingerprint in X-ray images: groups and massive galaxies show multiple shocks, while clusters show few shocks and abundant pressure-balanced (isobaric) structures, and that…","keywords":["AGN feedback","X-ray cavities","galaxy clusters","galaxy groups","intracluster medium","X-arithmetic","shocks","isobaric perturbations"],"falsifier":"A concrete check is to take a simulated cluster whose true gas state is known, generate mock X-ray images with the same band choices and exposure as a Chandra observation, run X-arithmetic on them, and compare every assigned feature type to the truth; if known adiabatic shocks with Mach number above about 2 or known line-of-sight mixtures are systematically assigned to the wrong effective equation of state, the small-amplitude linear decomposition is the failing premise.","tokens_in":29180,"feed_emoji":"🌌","tokens_out":8513,"duration_ms":85449,"temperature":0.7,"pith_summary":"AGN feedback from supermassive black holes heats and redistributes gas in the hot halos around galaxies, groups, and clusters, but how that energy is deposited is hard to read from X-ray images alone. This paper applies X-arithmetic, a technique that combines soft- and hard-band X-ray images to suppress one perturbation type at a time, to 15 deeply observed systems and classifies bright features into adiabatic (shocks and sound waves), isothermal (bubbles and cavities), and isobaric (cooling and subsonic motions). The main claim is that the feedback pattern depends on halo mass: groups and massive galaxies show multiple shocks in their inner regions, while clusters show only one or two shocks and abundant isobaric structures around their cavities. If true, this suggests that AGN feedback is more violent or more effective in shallower-potential systems, and it provides a fast, spectroscopy-free way to classify features in current and future X-ray images. The paper also reinterprets some previously identified \"isothermal shocks\" as mixtures of isobaric and adiabatic structures, supported by a spectroscopic deprojection of one feature in A2052.","feed_headline":"X-ray arithmetic reveals AGN feedback is more violent in galaxy groups","feed_subtitle":"Applying image arithmetic to 15 Chandra halos, groups show multiple shocks while clusters show isobaric structures.","key_machinery":"The central object is X-arithmetic: for each X-ray feature, the temperature and density fluctuations are assumed to obey an effective equation of state $\\delta T/T = \\alpha\\,\\delta n/n$, with $\\alpha=2/3$ for adiabatic perturbations (weak shocks and sound waves), $\\alpha=0$ for isothermal perturbations (bubbles and cavities), and $\\alpha=-1$ for isobaric perturbations (cooling and subsonic motions). The method computes the expected ratio of hard-band to soft-band surface brightness fluctuations $w_H/w_S$ for each type using plasma emissivity curves, a ratio that is independent of the density fluctuation amplitude, and then linearly combines the residual images to suppress each type in turn. A projection correction factor, derived assuming the perturbation sits at the cluster midplane in a single $\\beta$-model atmosphere, is applied to the hard-band image before combination. The work this machinery does is the paper's entire evidentiary base: visually identifying which features disappear when each component is removed is what turns morphology into physical classification.","core_discovery":"The paper's central discovery is a mass-dependent dichotomy in the physical nature of AGN feedback signatures. Analyzing 15 galaxy clusters, groups, and massive galaxies observed by Chandra, the authors find that lower-mass systems such as M84, NGC 5044, and NGC 5813 consistently show multiple shocks within about 20 kpc and clear bubbles, whereas massive cool-core clusters show at most one to two shocks and are dominated near their cavities by isobaric perturbations associated with gas cooling and subsonic motions. This is established by applying X-arithmetic, a linear decomposition of X-ray surface brightness fluctuations into adiabatic, isothermal, and isobaric components, and by confirming one central reclassification with spectroscopic deprojection: the A2052 northeast feature, previously a candidate second shock, has a continuous pressure profile across its edge, making it isobaric and likely a cold front from gas sloshing. The paper interprets the trend as evidence that feedback effects are stronger in smaller-mass systems, either because their shallower gravitational potentials make a given outburst more disruptive or because feedback is inherently more violent there, while noting resolution limits as an alternative explanation.","pith_inferences":["If the mass-dependent dichotomy holds in larger samples, it implies that AGN feedback models must reproduce not only global cluster properties but also the spatial mix of shock versus isobaric features, which is a stricter test than matching average profiles.","The projection correction assumes a feature sits at the cluster midplane, so for real systems with filaments or multiple outbursts along the line of sight, classification confidence could be improved by quantifying a confusion matrix using simulations viewed at various inclinations.","The reinterpretation of \"isothermal shocks\" as mixtures suggests that some published temperature-jump measurements may have been diluted by overlapping isobaric gas, so a systematic re-analysis of those features could change estimates of shock heating."],"forward_implications":["If the trend is real, galaxy groups and massive galaxies experience multiple AGN-driven shocks within their inner regions, while massive clusters are dominated by isobaric structures around their cavities.","Earlier identifications of \"isothermal shocks\" in clusters are likely mixtures of isobaric and adiabatic perturbations, so some published shock counts and Mach numbers may need revision.","X-arithmetic gives a fast, spectroscopy-free way to map perturbation types and can guide follow-up spectroscopy by locating the edges of structures.","The same method applied to mock X-ray images of simulations can test whether feedback implementations reproduce the observed shock and isobaric patterns.","With future high-throughput X-ray observatories, the method can be extended to larger samples beyond the brightest cluster cores."],"supporting_citations":[{"why":"Introduced the X-arithmetic method and applied it to Virgo and Perseus, providing the technique and previous results that this paper extends to a larger sample.","marker":"Churazov et al. (2016)"},{"why":"Identified the A2052 first shock and the candidate northeast second shock, supplying the features the paper reclassifies and the earlier spectroscopic comparison.","marker":"Blanton et al. (2011)"},{"why":"Identified M87's arms, shocks, and the soft-band density tracing behavior, providing prior structure identifications and motivation for the two-band approach.","marker":"Forman et al. (2007)"},{"why":"Found three AGN outbursts and shock fronts in NGC 5813, giving the group comparison that anchors the paper's mass-dependent trend.","marker":"Randall et al. (2015)"},{"why":"Cataloged Centaurus structures such as plumes, hooks, and sloshing features, against which the X-arithmetic classifications are compared.","marker":"Sanders et al. (2016)"},{"why":"Discovered the giant cavities and shock in MS 0735, providing the baseline for classifying that cluster's large-scale feedback structures.","marker":"McNamara et al. (2005)"},{"why":"Supplies the standard Chandra reprocessing, flare filtering, and background subtraction steps used to prepare the images.","marker":"Vikhlinin et al. (2005)"},{"why":"Provides XSPEC, which is used to simulate emissivity curves and derive the hard-to-soft ratio predictions for each perturbation type.","marker":"Arnaud (1996)"}],"fun_headline_variants":["X-arithmetic reveals shock-rich feedback in galaxy groups","Galaxy groups host more shocks from AGN feedback","Mass-dependent shock patterns seen in AGN feedback","Chandra data: groups shockier than clusters","Feedback's shock signature splits by mass"],"cache_read_input_tokens":3200,"weakest_assumption_plain":"Every classification inherits the assumption that each observed feature is a small-amplitude, clean superposition of just three perturbation types (adiabatic, isobaric, and isothermal), with second-order terms and projection mixing ignored; a strong shock, a deep cavity, or several structures overlapping along the line of sight could therefore be labeled wrongly even when the processed images look clean.","fun_headline_variants_meta":{"raw":{"variants":["X-arithmetic reveals shock-rich feedback in galaxy groups","Galaxy groups host more shocks from AGN feedback","Mass-dependent shock patterns seen in AGN feedback","Chandra data: groups shockier than clusters","Feedback's shock signature splits by mass"]},"model":"deepseek-v4-flash","effort":"low","cost_usd":0.000222,"raw_usage":{"total_tokens":1521,"prompt_tokens":1078,"completion_tokens":443,"prompt_tokens_details":{"cached_tokens":384},"prompt_cache_hit_tokens":384,"prompt_cache_miss_tokens":694,"completion_tokens_details":{"reasoning_tokens":371}},"tokens_in":694,"tokens_out":443,"duration_ms":5321,"temperature":1.0,"reasoning_tokens":371,"cache_read_input_tokens":384,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-08-06T22:19:48.944143+00:00","model_set":{"reader":"deepseek-v4-flash"},"falsifier":"A concrete check is to take a simulated cluster whose true gas state is known, generate mock X-ray images with the same band choices and exposure as a Chandra observation, run X-arithmetic on them, and compare every assigned feature type to the truth; if known adiabatic shocks with Mach number above about 2 or known line-of-sight mixtures are systematically assigned to the wrong effective equation of state, the small-amplitude linear decomposition is the failing premise.","supporting_citations":[],"review_version":1}