{"id":"dd8f4b5b-aa40-4104-b6e4-33f017751648","arxiv_id":"2608.05530","paper_version":1,"verdict":"CONDITIONAL","confidence":"MODERATE","novelty_score":5.0,"correctness_risk":"medium","formal_verification":"none","parameter_count":11,"one_line_summary":"A joint analysis of clustering, galaxy-shear, and shear measured from one HSC-Y3 photometric catalog gives S8 = 0.76 to 0.81, about 25 percent tighter than HSC-Y3 cosmic shear alone.","lead":"This paper combines galaxy clustering, galaxy-shear cross-correlation, and cosmic shear measured from a single photometric galaxy catalog from the Hyper Suprime-Cam survey to constrain cosmic structure. The analysis finds S8 between 0.76 and 0.81, consistent with earlier HSC cosmic shear results but about 25 percent tighter, a method that could help future surveys like Euclid and Rubin extract more cosmological information.","discovery_kind":"new_application","skeptic_critique":{"model":"deepseek-v4-flash","headline":"The 29.5% data-driven area cut and template deprojection for the galaxy density field are never validated end-to-end; this untested systematics-mitigation step could bias the clustering spectra that the S8 tightening depends on.","rationale":"The reader's weakest_assumption identifies the same load-bearing gap: the systematics mitigation applied to the galaxy density maps—the 29.5% area cuts and template deprojection—is not covered by the mock validation, and a bias in the clustering spectra would propagate into S8 through the self-calibration of Delta_z3, Delta_z4 and b_g. I agree with this assessment. The paper is otherwise careful: the theoretical model includes IA, magnification, baryonic feedback, and redshift-shift parameters; the B-mode tests are consistent with zero; the mock inference with 64 realizations checks the statistical pipeline; and the Section 6.3 robustness tests show the S8 interval is stable against several modeling choices. Those are genuine supporting checks. However, they do not address the one step where real data and mocks differ: the mocks contain no observational-condition variations and therefore no area cuts, deprojection, or their interaction. A clustering amplitude bias of even a few percent is enough to matter at the claimed ~0.025 S8 error. The proposed test—injecting realistic systematics into the mocks and applying the full masking/deprojection pipeline—would settle whether the concern lands. Because the reader already conditioned the verdict on this same issue, my stress-test pass does not move the verdict: it remains CONDITIONAL, and no new objection beyond the existing conditional concern is identified.","tokens_in":26596,"tokens_out":5436,"duration_ms":55684,"concrete_test":"Using the existing 64 mock realizations, inject realistic observational systematics into the mock galaxy density maps: assign the HSC-Y3 seeing, sky level, survey depth, star density, and pixel-coverage maps with amplitudes calibrated to the observed nbar-versus-systematic correlations, then apply the exact Appendix 2 area cuts and template deprojection and rerun the full photo-3x2-pt likelihood pipeline. If the ensemble-mean S8 or Delta_z3/Delta_z4 shifts by more than roughly 0.3 sigma relative to the current no-cut mocks, the area-cut/deprojection chain is biasing the central result; if no significant shift appears, the concern is resolved.","verdict_should_be":"UNCHANGED","load_bearing_attack":"The central claim—that adding galaxy clustering and galaxy-shear spectra to cosmic shear yields an unbiased S8 interval about 25% tighter than HSC-Y3 cosmic shear—rests on the galaxy clustering spectra carrying clean cosmological information. In the real analysis, those spectra are produced after two aggressive systematics-mitigation steps: (i) data-driven area cuts removing 29.5% of the survey area (Table 3, Appendix 2), and (ii) template deprojection of six contaminant maps (Section 3.2.3). The mock validation does not exercise either step: Appendix A.1.3 states, 'No area cut by observational conditions was made for galaxy density map as any variations of observational conditions were not considered in creating mock data.' Thus no end-to-end test verifies that the mitigation preserves the cosmological clustering signal. The mechanism for bias is concrete: the Table 3 thresholds are chosen by inspecting the same data's nbar-versus-systematic relations; e.g., the seeing cut removes 10.2% of the area and includes a lower threshold set because 'the slope of the correlation fluctuated erratically,' and the pixel-coverage cut removes 12.9%. If these cuts preferentially remove over- or underdense pixels, or if deprojection subtracts modes correlated with true density fluctuations, the clustering auto-spectra are biased. Because Section 6.1 uses these clustering spectra to self-calibrate Delta_z3, Delta_z4 and b_g, a clustering bias propagates directly into S8, not merely into an uninteresting nuisance parameter. The internal robustness tests in Section 6.3 vary IA, baryonic, and bias models but never vary the mask or deprojection, and the B-mode null tests do not probe galaxy-density systematics (a scalar density field has no B-mode). This is a fixable validation gap, but it is load-bearing for the claimed 25% tightening.","agreement_with_reader":"agree"},"referee_report":{"model":"deepseek-v4-flash","summary":"This paper presents a joint \"photo-3x2-pt\" cosmological analysis using the HSC-Y3 weak lensing shape catalog: it measures shear-shear, galaxy-shear, and galaxy clustering angular power spectra with the pseudo-C_ell method, mitigates systematics in the galaxy density maps via area cuts and template deprojection, and performs a Bayesian parameter inference with nuisance parameters for intrinsic alignments, galaxy bias, lensing magnification, baryonic feedback, and redshift distribution shifts. For a flat LCDM model the paper reports a 68% credible interval 0.76 <= S8 <= 0.81, which it states is consistent with the HSC-Y3 cosmic shear results of Dalal et al. (2023) and Li et al. (2023) but about 25% tighter. The analysis is supported by mock catalogs built from N-body and ray-tracing simulations, with 64 realizations, which are used to test unbiased recovery of S8, IA amplitudes, galaxy bias, and redshift shift parameters.","tokens_in":27019,"tokens_out":5502,"duration_ms":54623,"significance":"If the central claim holds, the paper demonstrates an important practical result: adding galaxy clustering and galaxy-shear spectra from the same photometric catalog used for cosmic shear can self-calibrate nuisance parameters such as intrinsic alignment amplitudes and redshift distribution shifts, yielding a tighter and apparently unbiased S8 constraint than cosmic shear alone. This is relevant for current Stage-III analyses and for the design of Stage-IV 3x2-pt and 6x2-pt analyses. The paper has clear strengths: it uses public, widely used software (NaMaster, CAMB, MultiNest, OneCovariance); it reports B-mode null tests for the new galaxy-shear and shear-shear combinations; it performs a non-trivial mock validation with 64 realizations; and it checks robustness of the S8 result against several modeling choices (IA model variants, baryonic feedback, galaxy bias redshift dependence). The main weakness is that the most aggressive systematics-mitigation steps applied to the real galaxy density maps, namely the 29.5% area cut and the template deprojection, are not exercised by the mock validation, which leaves a gap in the evidence for the central claim.","major_comments":[{"comment":"The end-to-end validation of the galaxy clustering systematics mitigation is missing, and this is load-bearing for the central claim. The real clustering spectra are produced after two steps: data-driven area cuts that remove 29.5% of the survey area (Table 3) and template deprojection of six contaminant maps (Section 3.2.3). The thresholds in Table 3 are chosen by inspecting the relation between galaxy number density and the systematics maps in the same data; for example, the seeing cut includes a lower threshold set because \"the slope of the correlation fluctuated erratically\" (Appendix 2). The mock validation explicitly does not exercise these steps: Appendix A.1.3 states \"No area cut by observational conditions was made for galaxy density map as any variations of observational conditions were not considered in creating mock data.\" The template deprojection is likewise not tested in the mocks. If the area cuts preferentially remove over- or underdense regions, or if the deprojection subtracts modes that are correlated with the true density field, the galaxy clustering auto-spectra are biased. Because Section 6.1 uses these clustering spectra to self-calibrate Delta_z3, Delta_z4 and the galaxy biases, such a bias would propagate directly into S8, not merely into an uninteresting nuisance parameter. I recommend that the authors add a validation test in which mock density maps are populated with realistic observational-condition templates (depth, seeing, sky level, coverage) and the same area-cut and deprojection pipeline is applied, checking that the recovered S8 and Delta_z values remain unbiased. At minimum, the real-data analysis should report how the S8 result shifts when the Table 3 thresholds are varied within plausible ranges.","section":"Appendix 2 and Section 3.2.3"},{"comment":"The headline comparison that the result is \"~25% tighter than the HSC-Y3 cosmic shear studies\" is not an apples-to-apples comparison, and the paper should either qualify it or provide a matched comparison. The HSC-Y3 analyses of Dalal et al. (2023) and Li et al. (2023) use different scale cuts (their cosmic shear extends to ell ~ 1800), different treatment of PSF systematics and shear calibration (marginalized over, as stated in Sections 5.3.2 of this paper), and different covariance choices. This paper's own \"gamma-gamma-only\" reference analysis uses the more conservative scale cut 317 <= ell <= 1000, so the relative tightening shown in Figure 8 conflates the effect of adding the new probes with the effect of changing scale cuts and systematics treatments. The paper does note some of these caveats, but the abstract and Section 6.4 nevertheless present the 25% figure as a headline. I suggest reporting the fractional width improvement from a matched analysis, e.g., the ratio of the S8 interval from the full photo-3x2-pt data vector to that from the same pipeline applied to the shear-shear spectra alone with identical scale cuts and systematics treatment.","section":"Section 6.4 and Figure 8"},{"comment":"The goodness-of-fit statistic chi2 = 70.5 for \"effective\" 85 degrees of freedom, with p = 0.87, is based on a post-hoc selection of the 11 parameters that the data happen to constrain more tightly than their priors. This procedure is not a standard chi2 test: the full model has 26 varied parameters, and selecting the effective number of degrees of freedom from the width of the posteriors makes the reported p-value optimistic. The unusually high p-value could also indicate that the covariance is somewhat overestimated. This does not directly invalidate the S8 constraint, but the paper should either report the chi2 and p-value for the full 96-26 = 70 degrees of freedom, or present the effective-dof calculation as an approximate diagnostic rather than a formal goodness-of-fit test.","section":"Section 6.2 and footnote 5"}],"minor_comments":[{"comment":"The table reports a cut of \"Seeing >= 0.5 and <= 0.75\" removing 10.2% of the area, but the lower threshold motivation is only described qualitatively. Since this is one of the largest area cuts, a quantitative stability test (e.g., how the inferred clustering amplitudes change when the lower threshold is varied by +/- 0.05 arcsec) would be valuable.","section":"Appendix 2, Table 3"},{"comment":"In the third row of Table 2, the dumped-redshift list \"0.449, 0.923, 0.538, ...\" contains a value 0.923 that is not in monotonic order with the neighboring entries; this is presumably a typo for something like 0.492 and should be corrected.","section":"Appendix 2, Table 2"},{"comment":"The text says \"in summery we cut ~30 percent of the survey area\"; \"summery\" should be \"summary\".","section":"Section 3.2.3"},{"comment":"The sentence \"the constraints on alpha_mu for all the bins are very week\" contains a typo: \"week\" should be \"weak\".","section":"Section 6.2"},{"comment":"The sentence \"our credible intervals are ~30% tiger\" contains a typo: \"tiger\" should be \"tighter\".","section":"Section 6.4"}],"recommendation":"major_revision","confidential_remarks":"The paper is a solid methodology demonstration with a plausible and interesting result, and the mock validation is a genuine strength. The central issue is that the systematics-mitigation pipeline for the galaxy clustering field, which carries the new information, is not validated end-to-end; this is the main reason I cannot recommend acceptance in the current form. If the authors add the recommended mock-injection test or a threshold-variation robustness test, and if they qualify the 25% comparison properly, the paper would be suitable for publication in PASJ. The effective-dof issue is secondary but should be fixed for statistical rigor."},"author_rebuttal":null,"desk_editor":{"model":"deepseek-v4-flash","letter":"Hamana reports the first HSC-Y3 photo-3x2-pt measurement: galaxy clustering, galaxy-shear, and cosmic shear power spectra from a single photometric catalog, giving S8 = 0.76–0.81, about 25% tighter than the HSC-Y3 cosmic shear analyses. The genuine advance is using the clustering auto-spectra to self-calibrate source redshift shifts Δz3 and Δz4, which removes a known error inflation in cosmic shear-only analyses.\n\nWhat is done well: the measurement pipeline is careful and standard (NaMaster pseudo-Cℓ, template deprojection, NLA + 1-loop IA, HMcode), the fit is good (χ² = 70.5 for 85 dof), and the mock validation is substantial—64 realizations with unbiased recovery of S8, IA amplitudes, bias, and Δz. The paper is also honest; it explicitly says the mock does not include observing-condition variations, so no area cut is applied to the mock density maps, and it acknowledges that fixing PSF and multiplicative-bias parameters slightly underestimates errors.\n\nThe load-bearing weakness is exactly that mock gap. The real clustering spectra are produced after cutting 29.5% of the survey area using thresholds chosen by inspecting the same data, then deprojecting six contaminant templates. The mocks do not exercise either step, so the central claim—that the added clustering information is clean enough to tighten S8 by 25%—is not tested end-to-end. The internal robustness tests vary IA, baryonic, and bias models but never the mask or deprojection, and the B-mode nulls don't probe scalar density-field systematics. This is fixable, not fatal, but the quoted tightening should be treated as conditional until the area cut and deprojection are validated on masked mocks or via a systematics-only null test on real data.\n\nThe 'first application of photo-3x2-pt' framing is also slightly strong; photometric 3x2-pt analyses such as DES Y3 exist in the literature and are not cited. The theoretical modeling and data handling otherwise look solid, and the limitations that are stated are stated clearly.\n\nBottom line: this deserves serious peer review. The method and result are worth engaging, and the validation gap is precisely what referees should ask to be closed. I would bring it to the reading group and cite it for the method demonstration, flagging the S8 tightening as conditional.","headline":"A careful first HSC-Y3 photo-3x2-pt measurement whose 25% S8 tightening depends on clustering systematics not exercised by the mocks; worth refereeing with a request for end-to-end validation.","tokens_in":27621,"tokens_out":3701,"would_cite":true,"duration_ms":31472,"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":"Combining galaxy clustering, galaxy-shear cross-correlation, and cosmic shear from the same photometric galaxy catalog tightens the S8 constraint by about 25% while remaining consistent with shear-only results.","keywords":["cosmology: observations","dark matter","cosmological parameters","large-scale structure of universe","cosmic shear","galaxy clustering","galaxy-galaxy lensing","intrinsic alignment"],"falsifier":"Apply the same photo-3x2pt pipeline to mock catalogs that include realistic spatial maps of seeing, depth, extinction, and sky level, impose the same area cuts and deprojection, and check whether the mean recovered $S_8$ over many realizations equals the input value within the quoted error; a shift larger than the statistical uncertainty would show the mitigation is biased.","tokens_in":26382,"feed_emoji":"🔭","tokens_out":10783,"duration_ms":84095,"temperature":0.7,"pith_summary":"This paper seeks to establish that a joint analysis of cosmic shear, galaxy-shear cross-correlation, and galaxy clustering measured from a single photometric galaxy catalog recovers the growth-of-structure parameter $S_8 = \\sigma_8\\sqrt{\\Omega_m/0.3}$ more tightly than cosmic shear alone. Using the HSC-Y3 weak lensing shape catalog, the author measures the three angular power spectra with the pseudo-$C_\\ell$ method and removes systematics from the galaxy density maps through area cuts and template deprojection. The reported 68% credible interval is $0.76 \\le S_8 \\le 0.81$, consistent with the HSC-Y3 cosmic shear analyses but about 25% tighter. A mock-catalog performance test is presented to show that the key cosmological and nuisance parameters are recovered without bias.","feed_headline":"Adding galaxy clustering to cosmic shear tightens S8 by 25 percent","feed_subtitle":"Combining clustering and galaxy-shear spectra from the same HSC-Y3 catalog reduces S8 uncertainty by a quarter.","key_machinery":"The data vector consists of 96 band-power measurements: ten shear-shear, ten galaxy-shear, and four galaxy clustering auto-spectra, with four band powers each; the scale cuts are $317 \\le \\ell \\le 1000$ for shear-shear and $178 \\le \\ell \\le 562$ for the other two. The power spectra are computed with the pseudo-$C_\\ell$ formalism, and the galaxy density maps are cleaned by cutting about 29.5% of the survey area where observing conditions correlate with galaxy density and then deprojecting contaminant template maps. The load-bearing mechanism for the improved $S_8$ is the self-calibration: because the galaxy clustering kernel is directly proportional to the redshift distribution, the clustering auto-spectra constrain the $\\Delta z_i$ shift parameters that otherwise inflate shear-only errors, while the galaxy-shear spectra constrain the intrinsic alignment amplitudes.","core_discovery":"The central claim is that the photo-3x2-pt data vector, built from the same photometric galaxy catalog that supplies the shear sample, constrains $S_8$ to $0.76 \\le S_8 \\le 0.81$ at 68% credibility. This interval overlaps the results of the HSC-Y3 cosmic shear studies but is about 25% narrower. The improvement is attributed to the added sensitivity to the two nuisance parameters that dominate shear-only errors: the intrinsic alignment amplitude, which the galaxy-shear spectra help pin down, and the shift parameters of the source redshift distributions, which the galaxy clustering auto-spectra effectively self-calibrate. The paper also reports that the measured B-mode power spectra are consistent with zero, and that mock-catalog inferences recover unbiased values of $S_8$, intrinsic alignment parameters, galaxy clustering bias, and redshift shift parameters.","pith_inferences":["If this result holds, photo-3x2pt analyses could reduce the reliance of Stage-IV weak lensing surveys on external photometric redshift calibration, since the clustering spectra self-calibrate the relevant shifts.","Beyond the paper, the same single-catalog strategy could be combined with a spectroscopic lens sample in a 6x2pt analysis; the paper's mock validation gives partial empirical support for that extension.","A testable prediction of the mechanism is that adding clustering and galaxy-shear spectra should shrink the $S_8$ posterior mainly by narrowing the $\\Delta z_3$ and $\\Delta z_4$ and intrinsic alignment directions, which could be verified by inspecting posterior covariances.","The area-cut thresholds in this paper are chosen from the data; a stronger validation would be to inject simulated systematics into mocks with realistic observing conditions and confirm that the recovery remains unbiased."],"forward_implications":["The $S_8$ credible interval narrows by roughly 25% relative to the HSC-Y3 cosmic shear-only analyses, with a consistent central value.","The shift parameters $\\Delta z_3$ and $\\Delta z_4$ for the two highest redshift bins are constrained about 30% more tightly, removing the main nuisance-driven error of shear-only analyses.","No significant B-mode signal is found in the galaxy-shear or shear-shear spectra, supporting the adopted scale cuts.","Because the extra information comes from the same galaxies already used for the shear catalog, the method requires no separate spectroscopic lens sample and can be applied to other photometric surveys."],"supporting_citations":[{"why":"It supplies the HSC-Y3 cosmic shear measurement and systematics treatment that set the baseline comparison and provide the PSF and multiplicative-bias corrections adopted here.","marker":"Dalal et al. 2023"},{"why":"It is the companion HSC-Y3 cosmic shear analysis whose $S_8$ interval and redshift shift constraints are the direct comparison targets.","marker":"Li et al. 2023"},{"why":"It is the public HSC-Y3 shape catalog from which all measured power spectra are constructed.","marker":"Li et al. 2022"},{"why":"It provides the redshift distributions of the four tomographic samples used in the theoretical models.","marker":"Rau et al. 2023"},{"why":"It establishes the clustering measurement and systematics-mitigation pipeline, including masks, systematics maps, and deprojection, that this analysis adapts.","marker":"Nicola et al. 2020"},{"why":"It introduces the template deprojection method used to subtract contaminants from the galaxy density maps.","marker":"Elsner et al. 2016"},{"why":"It implements the pseudo-$C_\\ell$ power spectrum estimation and mode-coupling correction used for all three spectra.","marker":"Alonso et al. 2019"},{"why":"It provides the full-sky lensing ray-tracing simulation framework and N-body settings on which the mock catalogs are based.","marker":"Takahashi et al. 2017"},{"why":"It supplies the 1-loop intrinsic alignment model terms used in the power spectrum predictions.","marker":"Blazek et al. 2015"},{"why":"It supplies the nonlinear matter power spectrum model used in the theoretical predictions.","marker":"Mead et al. 2016"}],"fun_headline_variants":["Same-catalog shear plus clustering cuts S8 error by 25%","HSC-Y3: clustering tightens cosmic shear S8 by a quarter","Photo-3x2-pt: 25% tighter S8 from one catalog","Joint shear-clustering analysis narrows S8 uncertainty 25%"],"cache_read_input_tokens":3200,"weakest_assumption_plain":"The result stands or falls on whether the area cuts and template deprojection remove observing-condition contaminants from the galaxy density maps without biasing the cosmological clustering signal, a step the mock test does not exercise because it omits observational condition variations and area cuts.","fun_headline_variants_meta":{"raw":{"variants":["Same-catalog shear plus clustering cuts S8 error by 25%","HSC-Y3: clustering tightens cosmic shear S8 by a quarter","Photo-3x2-pt: 25% tighter S8 from one catalog","Joint shear-clustering analysis narrows S8 uncertainty 25%"]},"model":"deepseek-v4-flash","effort":"low","cost_usd":0.000654,"raw_usage":{"total_tokens":3030,"prompt_tokens":1009,"completion_tokens":2021,"prompt_tokens_details":{"cached_tokens":384},"prompt_cache_hit_tokens":384,"prompt_cache_miss_tokens":625,"completion_tokens_details":{"reasoning_tokens":1939}},"tokens_in":625,"tokens_out":2021,"duration_ms":11873,"temperature":1.0,"reasoning_tokens":1939,"cache_read_input_tokens":384,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-08-08T11:25:40.794782+00:00","model_set":{"reader":"deepseek-v4-flash"},"falsifier":"Apply the same photo-3x2pt pipeline to mock catalogs that include realistic spatial maps of seeing, depth, extinction, and sky level, impose the same area cuts and deprojection, and check whether the mean recovered $S_8$ over many realizations equals the input value within the quoted error; a shift larger than the statistical uncertainty would show the mitigation is biased.","supporting_citations":[{"cited_title":"M., Dalal, R., Zhang, T., et al","cited_arxiv_id":null,"evidence_quote":"It provides the redshift distributions of the four tomographic samples used in the theoretical models."},{"cited_title":"2020, Journal of C osmology and Astroparticle Physics, 2020, 044","cited_arxiv_id":null,"evidence_quote":"It establishes the clustering measurement and systematics-mitigation pipeline, including masks, systematics maps, and deprojection, that this analysis adapts."},{"cited_title":null,"cited_arxiv_id":null,"evidence_quote":"It introduces the template deprojection method used to subtract contaminants from the galaxy density maps."},{"cited_title":null,"cited_arxiv_id":null,"evidence_quote":"It implements the pseudo-$C_\\ell$ power spectrum estimation and mode-coupling correction used for all three spectra."},{"cited_title":"2017, ApJ, 8 50, 24","cited_arxiv_id":null,"evidence_quote":"It provides the full-sky lensing ray-tracing simulation framework and N-body settings on which the mock catalogs are based."},{"cited_title":"2015, Journal of Cosmology and Astroparticle Physics, 2015, 015","cited_arxiv_id":null,"evidence_quote":"It supplies the 1-loop intrinsic alignment model terms used in the power spectrum predictions."},{"cited_title":"J., Heymans, C., Lombriser, L., et al","cited_arxiv_id":null,"evidence_quote":"It supplies the nonlinear matter power spectrum model used in the theoretical predictions."}],"review_version":1}