{"id":"668d87fb-ee44-4d5b-958f-76f556fcf1d5","arxiv_id":"2412.07834","paper_version":2,"verdict":"CONDITIONAL","confidence":"MODERATE","novelty_score":6.0,"correctness_risk":"medium","formal_verification":"none","parameter_count":2,"one_line_summary":"Using 11 CLASH clusters, this paper finds that satellite quenching is anisotropic out to at least 2.5 R200, peaking near 1.25 R200, and concludes pre-processing in large-scale structure, not AGN feedback, drives the signal.","lead":"Satellite galaxies around massive galaxy clusters are more likely to have stopped forming stars along the cluster's long axis than along its short axis, and this pattern persists far beyond the cluster's edge. The authors argue this 'anisotropic quenching' is caused by galaxies being pre-processed in cosmic filaments, not by outflows from the central black hole.","discovery_kind":"extension","skeptic_critique":{"model":"deepseek-v4-flash","headline":"The extended signal at 1.25–2.5R200 may be dominated by photo-z field contamination: membership uses a single broad Δz cut with no background subtraction, and the same large-scale structure that aligns with the BCG major axis can produce an anisotropic interloper population.","rationale":"The central claim has three novel parts: anisotropic quenching out to 3R200, significance to 2.5R200, and a radial peak near 1.25R200, followed by a pre-processing-over-AGN interpretation. All three rest on membership assignment at large radius, which is done with a single photo-z cut, Δz = 0.03(1+z)^3.26, and no background subtraction. The line-of-sight depth of this window at the sample's median redshift is hundreds of Mpc, so beyond R200 the sample is likely dominated by galaxies that are not cluster members. Because BCG major axes align with filaments, the interlopers are not isotropically distributed in angle; the same filament that feeds the cluster will populate the foreground and background of the photo-z window preferentially along the major axis. These filament galaxies are plausibly more passive from pre-processing, so they can produce or enhance exactly the measured anisotropic passive-fraction signal. This is a structural selection effect, not a disagreement with prior work; it threatens the headline new claims while leaving the inner result (within R200) largely intact. I considered the weak 1.25R200 peak (only ~2σ in one magnitude-limited sample) and the speculative backsplash interpretation as alternative concerns, but those are downstream: if the extended sample is contaminated, both the peak and the extended significance disappear regardless of their internal statistics. The paper deserves credit for a clear annulus methodology, two magnitude-limited samples, and consistency with Stott (2022) and Ando et al. (2023) inside R200. Still, a background-subtracted reanalysis or a spectroscopic-member check in the outskirts is required before the extended claims can be accepted. The reader's CONDITIONAL verdict is appropriate; I keep it unchanged pending this test.","tokens_in":27179,"tokens_out":11344,"duration_ms":106639,"concrete_test":"Recompute the annulus amplitudes in Figures 3 and 6 after statistical field subtraction: from the same Subaru catalog, select a field sample in photo-z windows shifted by ±0.04(1+z)^3.26 beyond the membership window (non-overlapping), normalize by area, and subtract its position-angle and colour distributions from the member sample in each 0.5R200 annulus; then refit Equation 1. If the amplitude at 2–2.5R200 falls below ~2σ or the 1.25R200 peak disappears, the extended signal is not robust to field contamination.","verdict_should_be":"UNCHANGED","load_bearing_attack":"The claim that anisotropic quenching remains significant out to 2.5R200 and peaks at 1.25R200 rests entirely on the photometric member catalogue. Membership is defined by a single colour-independent cut, Δz = 0.03(1+z)^3.26, calibrated using spectroscopic redshifts for a few clusters (Section 2; slope uncertainty ±1.05). At the median z ≈ 0.36, Δz ≈ 0.08, a line-of-sight depth of roughly 250–300 Mpc, i.e., hundreds of R200. No background or foreground subtraction is performed anywhere. At projected radii beyond R200, the surface density of true cluster members drops steeply while the photo-z window admits a large interloper population. Because BCG major axes are aligned with the surrounding cosmic web, those interlopers are not angle-random around the cluster: the same filament that feeds the cluster passes in front of and behind it, populating the broad Δz window preferentially along the BCG major axis, and those filament galaxies are plausibly more passive due to pre-processing. This can create or boost exactly the anisotropic passive-fraction signal claimed at 1.25–2.5R200. The inner signal below R200 is more secure and consistent with prior work, but the extended claims and the pre-processing-over-AGN inference built on them are not established until the selection effect is quantified.","agreement_with_reader":"agree"},"referee_report":{"model":"deepseek-v4-flash","summary":"This paper analyses satellite galaxy colour (B−R) and passive galaxy fraction as functions of the angle from the BCG major axis in 11 CLASH clusters at z ≈ 0.2–0.5, using Subaru/Suprime-Cam data. The authors report a significant sinusoidal anisotropic quenching signal within 1.5R200, with colour amplitude 0.14±0.01 and f_pass amplitude 0.063±0.006, both with periods consistent with 180°. They extend the analysis to 3R200, claim the signal remains significant out to at least 2.5R200, and report a radial peak of the amplitude at ≈1.25R200. They also show that f_pass is higher along the major axis for fixed local surface density at low densities, and conclude that pre-processing in large-scale structure, not AGN outflows, drives anisotropic quenching in massive clusters.","tokens_in":27416,"tokens_out":6921,"duration_ms":64068,"significance":"If the extended-radius claims hold, this would be an important result: it would be the first direct measurement of a radial peak in anisotropic quenching amplitude and the first detection of the signal at 2.5–3R200, with implications for the relative roles of pre-processing versus AGN feedback. The paper is clearly written and uses publicly available data, and the central inner signal (within 1.5R200) appears robust, with a fitted period consistent with 180° and f_pass amplitude in excess of 5σ. The main value of the paper is therefore in the extended radial and density analyses, but these rest on a photometric membership selection whose contamination is not quantified.","major_comments":[{"comment":"The cluster membership selection is a single colour-independent photometric redshift cut, Δz = 0.03(1+z_cluster)^3.26, with no background or foreground subtraction. At the median cluster redshift z ≈ 0.36 this window is Δz ≈ 0.08, corresponding to a line-of-sight depth of roughly 300 Mpc, i.e., several hundred R200. Because BCG major axes are aligned with the surrounding cosmic web (as cited in Section 4.2, e.g., Smith et al. 2023), the same filaments that feed the cluster can contribute an angle-dependent interloper population whose galaxies are already pre-processed and hence redder or more passive. This can produce or boost precisely the anisotropic signal reported at 1.25–2.5R200. The inner signal below R200 is less exposed to this effect and is consistent with previous work, but the abstract claims of a radial peak at ≈1.25R200 and significance out to 2.5R200 are not established until this selection effect is quantified, e.g., by comparing with an equal-area background sample or by reweighting by photo-z probability.","section":"Section 2"},{"comment":"The claim that the anisotropic quenching amplitude has a radial peak at ≈1.25R200 is supported only by a ~2σ drop to the adjacent annulus in the −16.8 mag sample, while in the −18.6 mag sample the peak amplitude is consistent with the surrounding annuli (amplitude 0.16±0.03 at 1–1.5R200 versus 0.15±0.04 at 1.5–2R200). Moreover, no formal significance test is reported for the individual annulus amplitudes, so the statement that the signal 'remains significant out to at least 2.5R200' is not quantified. The abstract's claim that this is the first direct measurement of a radial peak is therefore overstated; the data are consistent with a plateau as well as with a peak.","section":"Section 3.1, Figures 3 and 6"},{"comment":"The passive-fraction measurement is not an independent probe of the colour measurement: the sample is split at (B−R)_corr = 1.45, a cut chosen from the same colour distribution that defines the colour signal, and the f_pass angular signal is a non-linear transformation of the same colours. Consequently, the >5σ amplitude quoted for f_pass in Section 3.2 should not be presented as an independent confirmation of the colour signal. This does not weaken the inner detection, but the two analyses should be described as correlated tests rather than independent ones.","section":"Sections 3.1 and 3.2"},{"comment":"The conclusion that pre-processing in large-scale structure, rather than AGN outflows, drives anisotropic quenching rests on the density profiles shown in Figures 9 and 10, which are computed from the same photometric member sample. Given the broad photo-z window and the absence of background subtraction (see first major comment), the density contrast between major and minor axes at fixed cluster-centric radius could be inflated by anisotropic interlopers. A direct test with spectroscopically confirmed members in the outer annuli, or with a field-subtracted density estimate, is needed before the causal conclusion can be accepted.","section":"Section 4.2"}],"minor_comments":[{"comment":"The text states that galaxies are defined as being along an axis if within ±15° of that axis, but the captions of Figures 7 and 8 say ±30°; please reconcile the opening angle used.","section":"Section 3.3 and Figures 7–8 captions"},{"comment":"The text refers to spectroscopic redshifts for cluster members in 'MACS2129', but Table 1 lists the cluster as RXJ2129; this appears to be a typo that should be corrected.","section":"Section 2"},{"comment":"The surface density formula is written as log10(4/π d^2_n), which is ambiguous; it should be log10(4/(π d_n^2)) (and similarly for the 5th-nearest-neighbour term).","section":"Equation 2"},{"comment":"The statement that MACS1311 'only had R_C data available from Subaru' conflicts with Table 1, which lists B842, V843, and z'_IMACS from other instruments; please clarify that only the R_C band is from Subaru.","section":"Section 2"},{"comment":"The k-and-evolution correction uses an SSP model for quiescent galaxies, which may be less appropriate for star-forming galaxies; a brief sensitivity test using a different template set would help assess this systematic.","section":"Section 2, k-correction"}],"recommendation":"major_revision","confidential_remarks":"The central detection within 1.5R200 is likely robust and the paper fits the journal's scope. The main issue is the unquantified interloper contamination in the photometric member sample at large radii, which directly affects the abstract's flagship claims (radial peak, significance out to 2.5R200, and the pre-processing conclusion). This should be fixable with standard background-subtraction or photo-z weighting tests applied to the outer annuli, so I recommend major revision rather than rejection."},"author_rebuttal":null,"desk_editor":{"model":"deepseek-v4-flash","letter":"Colleague,\n\nThe paper confirms anisotropic quenching in 11 CLASH clusters with strong significance (f_pass amplitude 0.063±0.006, period consistent with 180°) and extends the measurement to ~3R200 using Subaru's wider field. The inner signal, within ~R200, is robust and consistent with prior work by Martín-Navarro, Stott, and Ando. The genuinely new items are the annulus-resolved amplitude measurements and the fixed-surface-density comparison; those are useful and worth having.\n\nThe soft spots are real, and they sit on the headline claims. Membership beyond R200 rests entirely on the photometric-redshift cut Δz=0.03(1+z)^3.26, calibrated on a few clusters, with no background subtraction. At z~0.36 that is a line-of-sight window of roughly 250–300 Mpc. Because BCG major axes are known to align with the surrounding filament network (the paper itself cites Smith et al. 2023 out to 10R200), the interloper population in that broad window is plausibly anisotropic and more passive along the major axis—exactly the pattern claimed at 1.25–2.5R200. So the radial peak at 1–1.5R200 and the “significant out to 2.5R200” assertion are not established until this selection effect is quantified. Section 4.2 acknowledges Ando et al. used background subtraction but does not apply one here; that is a gap.\n\nTwo smaller points. First, the f_pass and colour signals are not independent, since the passive/SF split uses the same (B−R) colour; that the two reinforce each other is partially circular. Second, the pre-processing-over-AGN conclusion is a reasonable inference from the cavity-size argument and the agreement with Karp et al. and Zakharova et al., but it inherits the membership uncertainty—if the extended signal is largely filament interlopers, the conclusion is about large-scale structure quenching rather than cluster satellites.\n\nWho should read this: anyone working on environmental quenching, BCG alignment, or cluster outskirts. It deserves a serious referee: the central result is solid, the questions raised are important, and the extended claims are addressable with better membership control—spectroscopic calibration, statistical background subtraction, or simulation-based contamination estimates. I would send it to review, expecting the extended claims to be softened or re-analyzed before publication.","headline":"Solid confirmation of anisotropic quenching in CLASH clusters, but the new claims about the radial peak and the signal out to 2.5R200 rest on a photo-z membership cut that could be contaminated by anisotropic interlopers.","tokens_in":28007,"tokens_out":3531,"would_cite":true,"duration_ms":36612,"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":"Across 11 massive CLASH clusters at $z\\approx0.36$, satellite galaxies aligned with the brightest cluster galaxy's major axis are significantly more likely to be quenched than those along the minor axis, a signal that extends to at least…","keywords":["anisotropic quenching","galaxy clusters","satellite galaxies","pre-processing","large-scale structure","brightest cluster galaxy","passive galaxy fraction","CLASH"],"falsifier":"A spectroscopic redshift survey of satellites at $1$\\textendash$2.5R_{200}$ in the same CLASH clusters would settle the question: if the colour excess along the major axis disappears once only confirmed members are used, the extended signal is a membership artifact rather than true anisotropic quenching.","tokens_in":26934,"feed_emoji":"🔭","tokens_out":7253,"duration_ms":57153,"temperature":0.7,"pith_summary":"The paper analyses 11 massive galaxy clusters from the CLASH survey at $z\\approx 0.2$\\textendash$0.5$ to test whether satellite galaxies along the brightest cluster galaxy's (BCG's) major axis are quenched more often than those along the minor axis. A clear anisotropic signal appears in both galaxy colour $(B-R)$ and passive fraction, with peaks along the major axis and a fitted period consistent with $180^\\circ$. The signal remains significant out to at least $2.5R_{200}$ (the radius enclosing 200 times the cosmic critical density) and its amplitude peaks near $1.25R_{200}$, the first direct measurement of a radial peak. Because the passive fraction is higher along the major axis even at fixed local surface density, the authors conclude that pre-processing of galaxies as they fall in along cosmic filaments, rather than AGN-driven outflows from the BCG, drives anisotropic quenching in massive clusters.","feed_headline":"Major-axis satellites quench first, out to 2.5R200","feed_subtitle":"First radial map of anisotropic quenching peaks at 1.25 R200, pointing to filament pre-processing over AGN feedback.","key_machinery":"The argument is carried by a simple angular decomposition: median colour and passive fraction are binned in angle from the BCG major axis and fitted with $y = A\\cos(f x) + c$, where the amplitude $A$ quantifies the anisotropic quenching strength. Radial structure is mapped by repeating the fit in $0.5R_{200}$-wide circular annuli out to $3R_{200}$, which reveals the peak at $1$\\textendash$1.5R_{200}$. To separate density from mechanism, the local surface density of each satellite is computed from the 4th- and 5th-nearest-neighbour distances, and colours and passive fractions are compared along the two axes at fixed density.","core_discovery":"The central discovery is that anisotropic quenching in massive clusters is a large-scale structure phenomenon, not an AGN feedback effect. The paper reports the first anisotropic quenching measurement out to $3R_{200}$, significant to at least $2.5R_{200}$, with an amplitude peak at approximately $1.25R_{200}$ that the authors attribute to a build-up of backsplash galaxies at that radius. At fixed local surface density, the passive fraction is higher along the major axis, ruling out a simple density artifact and showing that satellites along the major axis have spent more time in dense, pre-processing environments. The paper reconciles earlier discrepant results by arguing that the same filament-fed pre-processing explains the anisotropic signal seen in both low- and high-mass systems.","pith_inferences":["If filament pre-processing is the cause, the anisotropic signal should correlate with the actual filament orientation, not just the BCG major axis; stacking the CLASH clusters with external filament tracers would test this.","The backsplash interpretation predicts a stronger $1.25R_{200}$ peak in relaxed clusters, where more satellites have completed an orbit; splitting the sample by relaxation state would test it.","A similar analysis in galaxy groups ($M_h \\sim 10^{13} M_\\odot$) should show a weaker anisotropic signal, since coherent filamentary infall is less established in lower-mass halos.","Measuring green-valley (post-starburst) fractions along both axes could reveal whether the quenching timescale is genuinely shorter along the major axis, as the fixed-density offset implies."],"forward_implications":["Anisotropic quenching is not confined to the cluster core; it persists to at least $2.5R_{200}$, so any model of satellite quenching must act before galaxies cross the virial boundary.","The radial peak at $\\approx1.25R_{200}$ implicates backsplash galaxies, which pile up at this radius after orbiting through the cluster, as an important contributor to the signal.","The fixed-density passive fraction offset shows the effect cannot be explained away as a trivial density difference along the two axes.","AGN-fuelled X-ray cavities, which typically extend at most $\\sim0.2R_{200}$, cannot be the primary cause of the signal in massive clusters; pre-processing in filaments is the preferred explanation.","The similar signal strength in the two magnitude-limited samples argues against ram-pressure stripping as the dominant driver, since low-mass satellites would be affected more strongly."],"supporting_citations":[{"why":"first described anisotropic quenching and attributed it to AGN outflows; the paper directly tests this hypothesis.","marker":"Martín-Navarro et al. (2021)"},{"why":"previous CLASH-based detection of anisotropic quenching at $z\\sim0.4$\\textendash$0.5$; the present work extends it to larger radii with Subaru data.","marker":"Stott (2022)"},{"why":"found anisotropic quenching within $R_{200}$ in HSC-SSP clusters and no signal beyond; the paper contrasts its own detection out to $2.5R_{200}$.","marker":"Ando et al. (2023)"},{"why":"provided the Subaru Suprime-Cam photometry and photometric redshift catalogues that define the galaxy sample.","marker":"Umetsu et al. (2014)"},{"why":"CLASH photometric catalogues used to determine the BCG position angles for the angular analysis.","marker":"Molino et al. (2017)"},{"why":"supplied the cluster masses and $R_{200}$ values used to normalise cluster-centric radii.","marker":"Merten et al. (2015)"},{"why":"simulations showing AGN outflows are not required to explain anisotropic quenching; used to support the pre-processing interpretation.","marker":"Karp et al. (2023)"},{"why":"recent IllustrisTNG simulations finding anisotropic quenching out to $5R_{200}$ via filament infall; the paper cites it as independent agreement.","marker":"Zakharova et al. (2025)"},{"why":"demonstrates that BCGs align with large-scale structure out to $10R_{200}$, underpinning the filament-alignment argument.","marker":"Smith et al. (2023)"}],"fun_headline_variants":["Filaments, not black holes, drive galaxy quenching","First radial peak of anisotropic quenching at 1.25R200","Pre-processing in filaments, not AGN, drives anisotropic quenching","Backsplash galaxies drive quenching peak at 1.25R200"],"cache_read_input_tokens":3200,"weakest_assumption_plain":"Cluster membership at radii beyond $R_{200}$ rests entirely on the photometric-redshift cut $\\Delta z = 0.03(1+z)^{3.26}$, with no background subtraction; if the photo-$z$ scatter is underestimated or the cut admits an angle-dependent foreground/background population, the extended anisotropic signal could be artificially enhanced or mimicked.","fun_headline_variants_meta":{"raw":{"variants":["Filaments, not black holes, drive galaxy quenching","First radial peak of anisotropic quenching at 1.25R200","Pre-processing in filaments, not AGN, drives anisotropic quenching","Backsplash galaxies drive quenching peak at 1.25R200"]},"model":"deepseek-v4-flash","effort":"low","cost_usd":0.001764,"raw_usage":{"total_tokens":7025,"prompt_tokens":1073,"completion_tokens":5952,"prompt_tokens_details":{"cached_tokens":384},"prompt_cache_hit_tokens":384,"prompt_cache_miss_tokens":689,"completion_tokens_details":{"reasoning_tokens":5880}},"tokens_in":689,"tokens_out":5952,"duration_ms":36359,"temperature":1.0,"reasoning_tokens":5880,"cache_read_input_tokens":384,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-08-11T18:29:23.482184+00:00","model_set":{"reader":"deepseek-v4-flash"},"falsifier":"A spectroscopic redshift survey of satellites at $1$\\textendash$2.5R_{200}$ in the same CLASH clusters would settle the question: if the colour excess along the major axis disappears once only confirmed members are used, the extended signal is a membership artifact rather than true anisotropic quenching.","supporting_citations":[],"review_version":1}