{"id":"dd2f5bea-a485-411a-bd0f-aff5d949d287","arxiv_id":"2505.14791","paper_version":2,"verdict":"CONDITIONAL","confidence":"MODERATE","novelty_score":6.0,"correctness_risk":"medium","formal_verification":"none","parameter_count":5,"one_line_summary":"An alternative richness based on cylindrical line-of-sight galaxy counts, applied to Magneticum hydrodynamical simulations, yields no significant optical cluster selection bias on pairwise kSZ, pairwise velocity, or optical depth at current precision.","lead":"Using hydrodynamical simulations, this paper tests whether optical cluster selection biases the pairwise kinematic Sunyaev-Zel'dovich (kSZ) signal, pairwise velocities, and optical depth. It finds no significant bias within roughly 16%, 10%, and 8% uncertainties, suggesting upcoming kSZ measurements from DESI and CMB surveys may not need large optical selection corrections.","discovery_kind":"new_application","skeptic_critique":{"model":"deepseek-v4-flash","headline":"Random-slice lightcone construction may dilute richness–environment correlations and artificially push the measured selection bias toward unity; the authors' symmetry argument in Sec. 5.2 does not hold.","rationale":"The reader's weakest assumption is that the random-slice lightcone suppresses line-of-sight large-scale structure correlations, weakening the test for selection bias. I agree this is the single most load-bearing concern. The paper's central claim—no significant optical-selection bias on pairwise kSZ, pairwise velocities, or optical depth within ~16%, 10%, and 8%—depends entirely on the mock richness capturing the physical correlation between richness scatter and kSZ-related observables at fixed mass. If the cylinder used to measure richness extends beyond a slice boundary, the galaxies in the adjacent slice are drawn from an independent random realization and are uncorrelated with the cluster. This adds pure noise to the richness, diluting any true selection bias. The reconstructed sample is mass-selected and does not suffer this noise, so the bias ratio is biased toward one. The authors' written defense that the limitation affects both samples equally is not correct, and I could not find a quantitative statement in the paper that bounds the dilution factor. A true ~20% bias—comparable to the weak-lensing selection bias found by Wu et al. (2022) and Salcedo et al. (2024) using similar cylindrical richness—could be masked. I also considered the acknowledged low-mass-tail deficit relative to HOD-based mass distributions (Sec. 5.2), which would similarly reduce the selection bias potential; however, that effect is more model-dependent and less directly testable than the slice-boundary artifact. The proposed continuous-snapshot test cleanly isolates the slice construction as a cause of the null result. Because the paper is otherwise carefully executed, with the alternative richness calibrated to the DES-Y1 mass-richness relation and a matched-filter analysis that is robust to template details, the appropriate verdict remains CONDITIONAL pending this test, and I do not change the reader's judgment.","tokens_in":15737,"tokens_out":11388,"duration_ms":109074,"concrete_test":"Run the identical cylindrical-richness pipeline on a single continuous snapshot from Magneticum at z≈0.3 (a cubic volume with periodic boundaries), treating one axis as the line of sight, and recompute the bias ratio for pairwise kSZ, v12, and τ with the same λ≥5 cut and mass weighting. Because no slice boundaries exist, the full line-of-sight large-scale structure correlations are preserved. If the bias ratio moves away from unity by more than the lightcone result (e.g., from ~1.0 to >1.16), the null result is an artifact of the random-slice construction. As a cross-check within the existing data, split clusters by whether their cylinder is fully contained in one slice, and compare the recovered bias; a systematically higher bias in the contained subset would confirm the dilution.","verdict_should_be":"UNCHANGED","load_bearing_attack":"The most load-bearing weakness is the random-slice lightcone construction, acknowledged in Sec. 5.2: 'the slice sizes are comparable with the depth of the cylinder ... which could reduce large-scale structure correlations with the clusters.' The cylinder depth is defined as the maximum galaxy-cluster separation (40–100 cMpc), so the cylinder extends ±depth along the line of sight. With slice thicknesses of 151–161 cMpc (Table 1), for depth=60 cMpc a cylinder has total length 120 cMpc; roughly 80% of clusters are within 60 cMpc of a slice boundary and thus have part of their cylinder in an adjacent slice where the galaxy distribution is uncorrelated with the cluster. This adds uncorrelated Poisson noise to the alternative richness, reducing the correlation between richness residual and the pairwise-velocity/kSZ residual at fixed mass. The reconstructed (mass-selected) sample does not use richness, so it is unaffected. The authors' defense that 'these limitations affect both the richness-selected and the reconstructed samples in the same way' is therefore incorrect: the numerator is diluted toward unity by selection noise while the denominator is not. A true bias of ~20% could be reduced to ~10% and hidden inside the quoted 16% uncertainty.","agreement_with_reader":"agree"},"referee_report":{"model":"deepseek-v4-flash","summary":"This paper tests whether optical cluster selection, modeled by an alternative richness defined as galaxy counts in cylindrical volumes along the line of sight, biases the pairwise kinematic Sunyaev-Zel'dovich (kSZ) signal, the pairwise velocity, and the mean optical depth. Using the Magneticum Box2 5x5 deg^2 lightcone and a grid of galaxy-selection criteria that produce M-lambda relations consistent with DES-Y1, the authors assign mock richness to roughly 23,000 mock clusters at z=0.2-0.6, select a lambda>=5 sample, and compare its pairwise signals with a mass-selected sample reweighted to the same mass distribution. The bias ratio is estimated with a matched filter whose normalization cancels. The central finding is that all measured bias ratios are consistent with unity within statistical uncertainties of approximately 16%, 10%, and 8% for pairwise kSZ, pairwise velocity, and optical depth, respectively.","tokens_in":16017,"tokens_out":8945,"duration_ms":81534,"significance":"If the null result is robust, it is a timely and useful result for pairwise kSZ cosmology with optically selected cluster samples from upcoming surveys such as DESI, ACT, SO, and CMB-S4, since it would indicate that no large selection-bias correction is needed at current precision. The paper usefully extends the cylindrical-count selection-bias methodology of Wu et al. (2022) from weak lensing to kSZ and makes appropriate use of hydrodynamical simulations with ICM physics, which is necessary for modeling kSZ. The calibration of the mock M-lambda relation against DES-Y1 in the full sample and in four redshift bins is a strength, as is the systematic exploration of a wide range of galaxy-selection criteria. The matched-filter construction is carefully designed so that the filter normalization cancels in the ratio, and the bootstrap error estimation is appropriate. The main caveat is the random-slice construction of the lightcone, which the authors acknowledge but whose impact on the bias ratio they do not quantify; this is the primary reason the central claim is not yet fully supported.","major_comments":[{"comment":"The random-slice lightcone construction can artificially dilute the selection bias and therefore undermine the main null claim. The slice widths in Table 1 are 151-161 cMpc while the cylinder depths are 40-100 cMpc; because the cylinder extends +/-depth along the line of sight, a depth of 60 cMpc gives a total cylinder length of 120 cMpc, so for roughly 80% of clusters (2*60/151) part of the cylinder lies in an adjacent redshift slice where the galaxy distribution is uncorrelated with the cluster. This adds Poisson-like noise to the alternative richness, weakening the correlation between the richness residual and the kSZ/velocity residual at fixed mass and biasing the measured bias ratio toward unity. The reconstructed, mass-weighted sample does not contain this extra noise, so the statement in Sec. 5.2 that 'these limitations affect both the richness-selected and the reconstructed samples in the same way' is not correct for the bias ratio. I ask the authors to quantify this dilution, for example by comparing cylindrical richness measured in the full periodic box with richness measured in the sliced lightcone, or by validating against a continuously constructed lightcone, and to either correct the reported biases or present the result as an upper limit rather than a null detection.","section":"Sec. 5.2, Table 1"},{"comment":"The bias ratio is estimated with a matched filter whose template is the total pairwise velocity profile. If the optical-selection bias is scale-dependent or changes sign across the pair-separation range, the filtered ratio can remain consistent with unity even when the unfiltered profiles differ. Figure 3 shows the profile comparison for only one galaxy selection; I recommend reporting the unfiltered bias ratio, or a binned-in-separation version, for at least the DES-Y1-consistent selections, so that the reader can verify that no scale-dependent bias is hidden by the matched filter.","section":"Sec. 4.3, Appendix A"}],"minor_comments":[{"comment":"The abstract reports uncertainty limits of approximately 16%, 10%, and 8%, while Sec. 5.1 states that for R_theta=2.7 arcmin the biases are consistent with unity above the levels of 19%, 11%, and 9% and that the median uncertainties across selections are 16%, 10%, and 8%. Please use a consistent definition of the quoted uncertainty.","section":"Abstract, Sec. 5.1"},{"comment":"Units are inconsistent: Table 1 lists slice depths and widths in cMpc, while the cylinder depth in Sec. 4.1 and Fig. 1 is given in h^-1 cMpc. Please specify whether the Table 1 values are h^-1 cMpc and use consistent notation throughout.","section":"Table 1, Sec. 4.1"},{"comment":"The matched-filter template is derived from the 35x35 deg^2 lightcone based on Box0, which has different resolution and halo selection than the main Box2 lightcone. The authors state that the main result is insensitive to the template shape, but a quantitative test, such as recomputing the bias ratio with a different template shape, would strengthen this claim.","section":"Sec. 4.3, Appendix A"},{"comment":"The phrase 'violin areas include both statistical errors and systematic errors within a set of proposed galaxy selections' is unclear; the violins appear to show the distribution of bias values across galaxy selections rather than a formal error budget. Please clarify the definition of the displayed width.","section":"Fig. 5 caption"},{"comment":"The sentence about probing the tau-v independence 'down to a much lower mass' would benefit from a quantitative test or a reference, since the validity of the tau-v independence assumption is important for interpreting the kSZ bias as the product of the pairwise velocity and optical-depth biases.","section":"Sec. 5.2"}],"recommendation":"major_revision","confidential_remarks":"The paper is within the scope of A&A and the methodology is generally sound, but the main risk is the slice-boundary dilution of the selection bias in the mock lightcone. I would be willing to accept the paper after the authors provide a quantitative demonstration that this dilution is small, or after they correct the measured biases for it. I do not see a citation-pattern concern; the self-citations are to the methods being extended."},"author_rebuttal":null,"desk_editor":{"model":"deepseek-v4-flash","letter":"Short version: this is a solid, well-scoped simulation study that finds no significant optical selection bias on pairwise kSZ, pairwise velocity, or optical depth for DES-Y1-consistent richness in Magneticum, with median uncertainties around 16/10/8%. The main caveat is that the lightcone construction may dilute the line-of-sight correlations that generate the bias, so the null is provisional, not closed.\n\nWhat's new: applying the cylindrical-count selection-bias framework from Wu et al. (2022) to kSZ observables in hydrodynamical simulations, with richness matched to DES-Y1, is genuinely new. The paper is careful about mass-matching, uses a sensible matched-filter approach for the pairwise profile, and checks a broad grid of galaxy selection criteria. The M-lambda compatibility filter is a good idea. The bias ratios are consistent with unity with bootstrap errors, and the authors are honest about their limitations.\n\nSoft spots: the random-slice lightcone is the real issue. The authors acknowledge that slice sizes (151–161 cMpc) are comparable to cylinder depths (40–100 cMpc), which can reduce large-scale structure correlations with clusters. The stress-test note makes a sharper point: their defense that 'these limitations affect both samples in the same way' does not hold. The reconstructed (mass-selected) sample does not use richness, so projection noise only enters the numerator. That asymmetry could push a true bias toward unity. Given the quoted 16% uncertainty, a ~20% bias would be hidden inside the error bar. This does not kill the paper, but it means the headline claim is an upper bound rather than a demonstrated null.\n\nMinor: no code or data release, so independent verification is limited. The matched filter is fitted to the same simulation family, but the normalization cancels in the bias ratio, so that is not a serious issue.\n\nWho this is for: people working on pairwise kSZ with upcoming spectroscopic surveys (DESI, ACT, SO, CMB-S4) and cluster cosmology. It is a useful reference point even with the caveat. I would send it to peer review; the issue is an acknowledged limitation rather than an internal contradiction. The paper should be published with the caveat prominently stated, and ideally with a quantitative estimate of the projection dilution.\n\nRecommendation: engage with it, but treat the null as provisional.","headline":"Clean simulation study with an honest null result on kSZ selection bias, but the random-slice lightcone may dilute the very projection effect it aims to test.","tokens_in":16604,"tokens_out":1757,"would_cite":true,"duration_ms":16244,"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":"This paper finds no significant optical-selection bias in pairwise kSZ measurements from hydrodynamical simulations, within roughly 16% for the kSZ signal, 10% for pairwise velocity, and 8% for optical depth.","keywords":["pairwise kinematic Sunyaev-Zel'dovich effect","optical cluster selection bias","richness","pairwise velocity","optical depth","hydrodynamical simulations","cylindrical galaxy counts","matched filter"],"falsifier":"Run the same cylindrical-count selection on a continuous, non-sliced hydrodynamical simulation volume or a much larger lightcone, and check whether the bias ratios for pairwise kSZ, pairwise velocity, and optical depth move away from unity beyond the 16%, 10%, and 8% uncertainties.","tokens_in":15517,"feed_emoji":"🌌","tokens_out":9016,"duration_ms":80166,"temperature":0.7,"pith_summary":"The paper asks whether optically selecting galaxy clusters—via a richness-like count of galaxies along the line of sight—biases the pairwise kinematic Sunyaev-Zel'dovich (kSZ) signal, and it finds no significant bias at current precision. The authors build an alternative richness from galaxy counts in cylinders that mimic photometric redshift projection, calibrate it to the observed mass-richness relation, and compare kSZ observables for richness-selected clusters against a mass-selected sample reweighted to the same mass distribution. Across a wide grid of galaxy selection criteria, the bias ratios for pairwise kSZ, pairwise velocity, and optical depth are consistent with unity within uncertainties of about 16%, 10%, and 8%, respectively. If correct, pairwise kSZ measurements from optically selected cluster samples—central to upcoming spectroscopic and CMB surveys—can be interpreted without a large selection-bias correction.","feed_headline":"No optical-selection bias found for pairwise kSZ","feed_subtitle":"Optically selected mock clusters show pairwise kSZ, velocities, and optical depth consistent with mass-selected samples.","key_machinery":"The machinery is an alternative richness built from cylindrical galaxy counts. For every halo, the paper counts galaxies selected by stellar mass, age, and specific star formation rate inside a cylinder whose radius is about R200m and whose depth, 40-100 comoving Mpc/h, mimics photometric redshift uncertainty; this count is the richness, and the resulting mass-richness relation is required to match the observed survey-calibrated relation. The unbiased comparison is constructed by reweighting a mass-selected halo catalog to the richness-selected sample's mass probability distribution, and the two pairwise signals are compared with a matched filter that weights pair separations to optimize signal-to-noise, using a template and covariance from a larger simulation lightcone. Optical depth is handled separately by averaging binned per-cluster estimates with the same mass weights.","core_discovery":"In a hydrodynamical simulation of a 5x5 degree sky patch, the paper constructs mock galaxy and cluster catalogs and assigns each halo an alternative richness equal to the number of bright and red galaxies inside a cylinder of radius about R200m and depth 40-100 comoving Mpc/h, matching the observed survey-calibrated mass-richness relation. It then measures pairwise kSZ, pairwise velocity, and optical depth for the richness-selected subsample and compares them with a reconstructed unbiased signal: a mass-selected catalog weighted to reproduce the richness-selected mass distribution. The ratios are consistent with one; the quoted median uncertainties are roughly 16% for the kSZ amplitude, 10% for pairwise velocity, and 8% for optical depth. The paper concludes that optical cluster selection does not create a detectable kSZ equivalent of the selection bias that affects weak lensing.","pith_inferences":["Beyond the paper: the simulation lightcone is assembled from redshift slices whose thickness is comparable to the cylinder depth, so a continuous simulation volume might preserve more line-of-sight clustering and reveal a bias larger than the one reported.","Beyond the paper: at the higher precision expected from future surveys, a small residual bias below the 16% level could still matter for cosmological parameter constraints, so rerunning the same pipeline on larger simulations would set a tighter upper limit.","Beyond the paper: because pairwise kSZ is roughly a product of optical depth and pairwise velocity, biases in the two components might partially cancel; measuring their correlation directly in a larger sample would sharpen the interpretation."],"forward_implications":["Pairwise kSZ analyses using optically selected clusters can proceed without applying a selection-bias correction at the precision of current and near-future measurements.","The cylindrical-count method with a survey-calibrated mass-richness relation is a workable mock for selection-bias studies across different galaxy-selection assumptions.","Selection bias does not affect kSZ, velocity, and optical depth in the same way it affects weak lensing, so the two probes may be combined without assuming a common projection bias.","The results hold for two different photometric aperture radii and across a broad grid of galaxy selection criteria, reinforcing the null conclusion."],"supporting_citations":[{"why":"Supplies the cylindrical galaxy-count method for assigning alternative richness and the weighting scheme used to reconstruct an unbiased signal.","marker":"Wu et al. (2022)"},{"why":"Provides the observed mass-richness relation that the mock clusters must match.","marker":"To et al. (2021a)"},{"why":"Establishes that such cylindrical selection creates a significant bias in weak lensing, the comparison case.","marker":"Sunayama et al. (2020)"},{"why":"Shows how optical selection bias could affect cluster cosmology, motivating the kSZ test.","marker":"Salcedo et al. (2024)"},{"why":"Provides the hydrodynamical-simulation lightcone and the empirical pairwise-velocity template used by the matched filter.","marker":"Soergel et al. (2018)"},{"why":"Defines the pairwise kSZ estimator and the redshift-evolution correction the paper chooses not to apply.","marker":"Hand et al. (2012)"},{"why":"Gives the pairwise estimator form used to measure the kSZ signal from aperture photometry.","marker":"Ferreira et al. (1999)"},{"why":"Defines redMaPPer richness, the observable whose selection effect the cylindrical count mimics.","marker":"Rykoff et al. (2014)"}],"fun_headline_variants":["Simulation: no optical-selection bias in pairwise kSZ","Pairwise kSZ robust to optical cluster selection","Optical selection does not bias pairwise kSZ in mocks","Mock clusters show no kSZ bias from optical selection","Pairwise kSZ unaffected by optical selection bias"],"cache_read_input_tokens":3200,"weakest_assumption_plain":"The mock sky is stitched from separate depth slices, so it may break the long-range line-of-sight correlations that connect projected galaxy counts to cluster gas and velocities; if those correlations matter, the null result could be artificially clean.","fun_headline_variants_meta":{"raw":{"variants":["Simulation: no optical-selection bias in pairwise kSZ","Pairwise kSZ robust to optical cluster selection","Optical selection does not bias pairwise kSZ in mocks","Mock clusters show no kSZ bias from optical selection","Pairwise kSZ unaffected by optical selection bias"]},"model":"deepseek-v4-flash","effort":"low","cost_usd":0.000582,"raw_usage":{"total_tokens":2750,"prompt_tokens":969,"completion_tokens":1781,"prompt_tokens_details":{"cached_tokens":384},"prompt_cache_hit_tokens":384,"prompt_cache_miss_tokens":585,"completion_tokens_details":{"reasoning_tokens":1703}},"tokens_in":585,"tokens_out":1781,"duration_ms":13171,"temperature":1.0,"reasoning_tokens":1703,"cache_read_input_tokens":384,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-08-07T15:29:19.362521+00:00","model_set":{"reader":"deepseek-v4-flash"},"falsifier":"Run the same cylindrical-count selection on a continuous, non-sliced hydrodynamical simulation volume or a much larger lightcone, and check whether the bias ratios for pairwise kSZ, pairwise velocity, and optical depth move away from unity beyond the 16%, 10%, and 8% uncertainties.","supporting_citations":[{"cited_title":"2020, Monthly Notices of the Royal Astronomical Society, 496, 4468 Article number, page 10 Y .-H","cited_arxiv_id":null,"evidence_quote":"Establishes that such cylindrical selection creates a significant bias in weak lensing, the comparison case."},{"cited_title":"N., Wu, H.-Y ., Rozo, E., et al","cited_arxiv_id":null,"evidence_quote":"Shows how optical selection bias could affect cluster cosmology, motivating the kSZ test."},{"cited_title":"2018, Monthly Notices of the Royal Astronomical Society, 478, 5320","cited_arxiv_id":null,"evidence_quote":"Provides the hydrodynamical-simulation lightcone and the empirical pairwise-velocity template used by the matched filter."},{"cited_title":"E., Aubourg, E., et al","cited_arxiv_id":null,"evidence_quote":"Defines the pairwise kSZ estimator and the redshift-evolution correction the paper chooses not to apply."},{"cited_title":"G., Juszkiewicz, R., Feldman, H","cited_arxiv_id":null,"evidence_quote":"Gives the pairwise estimator form used to measure the kSZ signal from aperture photometry."},{"cited_title":"S., Rozo, E., Busha, M","cited_arxiv_id":null,"evidence_quote":"Defines redMaPPer richness, the observable whose selection effect the cylindrical count mimics."}],"review_version":1}