{"id":"fe2e784f-e4fe-4d9a-8548-4b57dcdf788e","arxiv_id":"2412.13162","paper_version":2,"verdict":"CONDITIONAL","confidence":"HIGH","novelty_score":6.0,"correctness_risk":"medium","formal_verification":"none","parameter_count":1,"one_line_summary":"Redshift bin edges create a Doppler-boost boundary term in the kinematic matter dipole that can match or reverse the standard Ellis-Baldwin signal.","lead":"Galaxy surveys that measure the cosmic matter dipole in redshift slices must add a new boundary correction caused by the Doppler boost of galaxy redshifts. The correction can rival or exceed the standard dipole amplitude and even flip its direction, so ignoring it would distort future tomographic tests of cosmology.","discovery_kind":"new_method","skeptic_critique":{"model":"deepseek-v4-flash","headline":"Eq. (20) assumes W_b independent of S_*; photometric forecasts violate this, so the general-selection boundary-term formula is not yet validated for flux-limited photo-z surveys.","rationale":"The paper's central formal result is the boundary-term correction. Equation (13) for sharp spectroscopic cuts is a clean integration by parts and is not affected by the S_*-dependence concern. The general photometric formula Eq. (20) is the part that supports the survey forecasts. Its derivation in Appendix B assumes W_b(z) is independent of S_*; this is not a minor technicality because in a flux-limited photometric survey the training sample used to infer P(z',z) is itself flux-limited, so the inferred selection function inherits an S_* dependence. Section III.C acknowledges this but leaves the full computation to future work. The forecasts in Sec. IV use a Gaussian P with no flux dependence, so they represent a best-case assumption rather than a validated prediction. A quantitative mock test with flux-dependent photo-z scatter would settle whether the missing terms are negligible. This does not change the reader's conditional verdict: the core mechanism is likely correct, but the general applicability to photometric surveys remains open.","tokens_in":27840,"tokens_out":10216,"duration_ms":107724,"concrete_test":"Construct a mock Euclid-like catalog with flux S and true z drawn from Eq. (28), assign flux-dependent photo-z scatter σ_z(z,S) = σ_0(1+z)(S/S_*)^-γ to mimic a flux-limited calibration sample, boost redshifts and fluxes, apply top-hat z' bins, and measure the per-bin number-count dipole. Compare the measured boundary terms with Eq. (20) computed using the S_*-averaged W_b. If |ΔB| is ≳ 0.5 in any bin (comparable to the standard Ellis-Baldwin amplitude ~4β), the flux-cut independence assumption is numerically important and Eq. (20) needs generalization.","verdict_should_be":"UNCHANGED","load_bearing_attack":"Section III.C (Eqs. 25-27) explicitly concedes that W_b can depend on S_* in photometric surveys, because the photo-z conditional P(z',z;S_*) is calibrated on flux-limited spectroscopic subsamples. The derivation of Eq. (20), and the definitions of x̃_b and α̃_b in Appendix B (Eqs. B3-B6 and B16-B19), assume ∂W_b/∂S_* = 0. If this fails, N_b(S_*) = ∫ n(z,S_*) W_b(z;S_*) dz contains an extra term from ∂W_b/∂S_* in x̃_b, and the integration-by-parts step no longer reduces to the simple log-derivative of W_b. The paper does not quantify this term, explicitly leaving it to future work. Consequently, the boundary-term predictions for Euclid and Rubin-LSST in Sec. IV are computed with an S_*-independent Gaussian P and need not apply to real flux-limited photo-z samples, where photo-z scatter varies strongly with magnitude. The sharp-cut spectroscopic result Eq. (13) does not rely on this assumption, so the central boundary-term mechanism itself is not in doubt; but the claim to handle arbitrary redshift selection functions is overbroad.","agreement_with_reader":"agree"},"referee_report":{"model":"deepseek-v4-flash","summary":"The paper computes the kinematic dipole in the angular distribution of matter tracers when the sample is split into redshift bins. Its central claim is that, in addition to the standard Ellis-Baldwin terms from aberration and flux boosting, the Doppler boosting of observed redshifts produces a boundary term at the edges of the redshift selection. For sharp cuts the term is (1+z) f_b(z) evaluated at the bin edges (Eq. 13); for general redshift selection functions it is given by the logarithmic derivative of the selection function W_b(z) (Eq. 20). The authors show with model redshift distributions for Euclid, Rubin-LSST, and SKA that these boundary terms can be comparable to or larger than the standard amplitude and can even reverse the dipole direction. They also discuss redshift uncertainties, biased photo-zs, and color cuts.","tokens_in":28072,"tokens_out":10613,"duration_ms":102862,"significance":"If correct, the result is important: tomographic measurements of the kinematic matter dipole must include the redshift-boost boundary term, otherwise the predicted amplitude is off by O(1) multiples of beta and can even have the wrong sign. The central identity is derived, not fitted; the sharp-cut result Eq. (13) is exact and depends only on observables, and the mock checks in App. C2 confirm that Eq. (21) reproduces the measured dipole in simulated samples. The forecasts are useful but conditional on the assumed analytic redshift distributions and on the explicitly stated assumption that selection functions are independent of the flux cut. The main weakness is that the general-selection formula is applied to flux-limited photometric surveys even though the paper itself concedes that real photo-z selection functions can depend on the flux cut.","major_comments":[{"comment":"The integration by parts in Eq. (B19) drops the term -∫ n(z) W_b(z) dz, which is exactly the -1 that converts the 3 in Eq. (B15) into the 2 in Eq. (20). As printed, the chain (B15)->(B19) yields D_b = [3 + x̃_b(1+α̃_b) - ∫ f_b dlog W_b/dlog(1+z)] β, not Eq. (20). The correct identity is ∫ f_b dlog n/dlog(1+z) dz = f_b(z)(1+z)|_0^∞ - 1 - ∫ f_b dlog W_b/dlog(1+z) dz, in direct analogy with Eq. (12). Eq. (20) itself appears to be the correct final result, but the appendix derivation must be corrected.","section":"Appendix B, Eqs. (B15)-(B19)"},{"comment":"Eq. (20) is derived under the assumption ∂W_b/∂S_* = 0, stated immediately before the equation. Section III.C concedes that for photometric surveys calibrated on flux-limited spectroscopic subsamples, P(z',z;S_*) and hence W_b depend on the flux cut, so the integration-by-parts step no longer reduces to the simple log-derivative of W_b, and the extra ∂W_b/∂S_* contribution to x̃_b in Eqs. (B3)-(B6) is not quantified. Nevertheless, Sec. IV.A and IV.B compute boundary-term forecasts for Euclid, Rubin-LSST, and SKA using an S*-independent Gaussian P (Eq. 29). These forecasts are therefore not directly applicable to real flux-limited photo-z samples, where photo-z scatter varies with magnitude. The sharp-cut spectroscopic result Eq. (13) is not affected, but the claim to handle arbitrary redshift selection functions is overbroad; the paper should either quantify the omitted term, restrict the photometric forecasts, or qualify the abstract and conclusion accordingly.","section":"Sec. III.C and Sec. IV.A, Eq. (20)"}],"minor_comments":[{"comment":"The distributional derivative of log W_b for a discontinuous top-hat selection function is not well defined because W_b vanishes outside the bin and 1/W_b is singular at the jumps; the final result Eq. (23) is correct, but it should be derived as a limit of smooth W_b rather than through the formal expression in Eq. (22).","section":"Eqs. (22)-(23)"},{"comment":"The notation would be clearer if the authors stated explicitly that n(z) in Eqs. (15) and (6) denotes the boosted observed-redshift monopole distribution, not the comoving or cosmological redshift distribution, since the distinction is central to the boundary-term argument.","section":"Sec. III, around Eq. (15)"},{"comment":"The quantities D_obs and B_obs appearing in the last panel of Fig. 8 are not defined in the caption or in the surrounding text; please define them explicitly where the mock analysis is described.","section":"Fig. 8"},{"comment":"The phrase 'arbitrary redshift selection functions' should be qualified to make clear that the formula in Eq. (20) assumes no flux-cut dependence of W_b; the discussion in Sec. III.C is honest about this, but the abstract overstates the generality.","section":"Abstract and Sec. III"}],"recommendation":"major_revision","confidential_remarks":"The core mechanism is sound and the paper will be a useful contribution once the appendix derivation is fixed and the photometric-forecast caveat is addressed. I do not see grounds for rejection: the sharp-cut result is exact, the mock checks validate the boundary term, and the flux-cut limitation is explicitly acknowledged in §III.C. The main tasks for revision are local but load-bearing: correct Eq. (B19) and either quantify or clearly delimit the photometric forecasts."},"author_rebuttal":null,"desk_editor":{"model":"deepseek-v4-flash","letter":"The thing to know: this paper identifies a correction that anyone measuring the kinematic matter dipole in redshift bins will have to include. The boundary term from boosting redshifts across bin edges is not exotic—it is a straightforward integration by parts applied to an existing per-redshift dipole expression, and it can be as large as or larger than the standard Ellis–Baldwin amplitude, even flipping the dipole direction. The sharp-cut result, Eq. (13), is clean and I do not doubt it. The mocks in Appendix C2 confirm that the general-selection formula, Eq. (21), reproduces the dipole measured in simulated samples, which is real evidence.\n\nWhat is genuinely new: the general selection-function formula, Eq. (20), the treatment of photometric redshift uncertainties with smooth boundary terms, the color-cut discussion, and the survey-specific forecasts for Euclid, Rubin-LSST, and SKA. The paper is honest about its limitations—Section III.C explicitly flags that the simple boundary-term formula assumes the selection function is independent of the flux cut, and that photometric surveys whose photo-z scatter is calibrated on flux-limited spectroscopic subsamples may violate this. The color-cut analysis likewise rests on a conditional-distribution ansatz and leaves a running spectral index to future work. These are not hidden flaws; they are stated caveats.\n\nThat said, the stress-test concern lands on a real soft spot. The forecasts for Euclid and Rubin-LSST use a Gaussian photo-z scatter that is independent of flux, but real photo-z scatter varies with magnitude. So the quantitative boundary-term predictions for those surveys should be read as illustrative until the flux-dependent case is worked out. The phrase \"arbitrary redshift selection functions\" in the abstract is a bit strong; it should say selection functions independent of the flux cut. This does not undermine the central mechanism—the sharp-cut spectroscopic result does not depend on that assumption—but it means the general-selection formula is not yet the last word for photometric surveys.\n\nWho is this for: anyone doing redshift-tomographic measurements of the number-count dipole with Euclid, LSST, or SKA, and anyone comparing such measurements to the CMB dipole. The paper deserves a serious referee. I would send it to review rather than desk reject it, and I would ask the referee to scrutinize the flux-dependence caveat and the applicability of the photo-z forecasts, while expecting the sharp-cut result to stand.","headline":"A clean integration-by-parts result that exposes a real, often large boundary correction to the tomographic matter dipole; the sharp-cut version is solid, and the photo-z generalization is useful but still conditional on the flux-cut independence assumption.","tokens_in":28584,"tokens_out":1675,"would_cite":true,"duration_ms":18555,"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":"Redshift bin edges add a boundary term to the kinematic matter dipole that can rival or reverse the standard Ellis-Baldwin dipole.","keywords":["kinematic matter dipole","redshift tomography","Ellis-Baldwin dipole","boundary terms","Doppler boost","redshift selection function","photometric redshifts","source number counts"],"falsifier":"Simulate a full-sky catalogue with a known boost $\\beta$ and no intrinsic clustering dipole, bin it in observed redshift with a known selection function, and compare the measured per-bin dipole amplitudes with the prediction of Eq. (20) using the true $\\tilde{x}_b$ and $\\tilde{\\alpha}_b$; disagreement in the bin-to-bin pattern of amplitudes would falsify the boundary-term claim. The paper's own 1000-mock comparison performs this same check.","tokens_in":27636,"feed_emoji":"🌌","tokens_out":13221,"duration_ms":105663,"temperature":0.7,"pith_summary":"The kinematic matter dipole is the faint dipolar asymmetry that an observer's peculiar motion imprints on the sky distribution of distant galaxies, and the Ellis-Baldwin formula predicts its amplitude from the flux slope of source counts. This paper shows that once a galaxy sample is split into observed-redshift bins, the Doppler boost of redshifts themselves adds a boundary term at the edges of each bin. For sharp cuts the term is $(1+z)\\,f_b(z)$ evaluated at the two bin boundaries, and for photometric selections it becomes an integral over the logarithmic redshift derivative of the selection function $W_b(z)$. The correction can rival or exceed the Ellis-Baldwin amplitude and can even flip the dipole direction, so it must be included in redshift-tomographic tests of the cosmological principle.","feed_headline":"Redshift bin edges can reverse the cosmic matter dipole","feed_subtitle":"Doppler-boosted redshifts add a boundary term that can rival or outweigh the standard dipole signal.","key_machinery":"The central object is the boundary term $B(z_b)$ that appears when the redshift-dependent dipole $D_{\\mathrm{kin}}(z) = [3 + x(z)(1+\\alpha(z)) + d\\log n(z)/d\\log(1+z)]\\,\\beta$ is averaged over a bin and integrated by parts. The surface term $(1+z)f_b(z)|_{z_1}^{z_2}$, or its smooth-selection generalisation $-\\int dz\\, f_b(z)\\, d\\log W_b(z)/d\\log(1+z)$, carries the new physics: it counts how sources are Doppler-shifted into or out of the bin, exactly as the flux threshold $S_*$ does in the Ellis-Baldwin effect. The selection function $W_b(z)$, built by convolving a top hat on observed redshift with the conditional photo-$z$ distribution $P(z',z)$, turns the boundary term into a quantity computable from photometric survey data.","core_discovery":"The claim is that a sample selected in observed redshift carries an additional contribution to the kinematic matter dipole from the boosting of redshifts, on top of the usual angular-aberration and flux-boosting terms. In a bin with normalised redshift distribution $f_b(z)$, the dipole amplitude becomes $D_{\\mathrm{kin}}(z_b) = [2 + \\tilde{x}_b(1+\\tilde{\\alpha}_b) + (1+z)f_b(z)|_{z_1}^{z_2}]\\,\\beta$ for sharp cuts, and $D_{\\mathrm{kin}}(z_b) = [2 + \\tilde{x}_b(1+\\tilde{\\alpha}_b) - \\int_0^\\infty dz\\, f_b(z)\\, \\frac{d\\log W_b(z)}{d\\log(1+z)}]\\,\\beta$ for a general selection function $W_b(z)$. The new term is non-zero whenever the selection function is asymmetric around the bin's effective redshift, and it can dominate the Ellis-Baldwin terms or reverse the dipole's sign. The authors verify the expressions against full-sky mock maps and forecast the boundary terms for Euclid, Rubin-LSST and SKA redshift distributions, finding values of order a few times $\\beta$ that change sign near the peak of the source distribution.","pith_inferences":["If the boundary-term picture holds, previously published redshift-integrated dipole measurements remain valid, but future tomographic analyses must report their per-bin selection functions; otherwise the predicted dipole in each bin cannot be reconstructed from the published amplitude alone.","Because sources cross bin boundaries under the boost, dipole measurements in neighbouring redshift bins should be anti-correlated at the level of the boundary terms; searching for that cross-bin correlation would isolate the effect from other systematics.","The same formalism could be applied to the kinematic quadrupole, where the prefactor may lift the second-order signal above the clustering quadrupole, or to redshift-weighted number count estimators, whose weights are themselves boost-affected.","The boundary terms depend sensitively on the shape of $n(z)$, so redshift-distribution uncertainty, not just shot noise, is likely to dominate the error budget of future tomographic dipole measurements."],"forward_implications":["Tomographic measurements by Euclid, Rubin-LSST and SKA will see boundary terms of order a few times $\\beta$ that change sign near the peak of the source redshift distribution, so the expected dipole in each bin differs substantially from the Ellis-Baldwin value.","The common practice of removing low-redshift sources changes the kinematic dipole expectation for the remaining sample by tens of percent, and the removed low-redshift subsample carries boundary terms of order $+7.5\\beta$ for SKA-like specifications.","Photometric redshift uncertainties do not erase the boundary terms; for Gaussian photo-$z$ scatter they merely reshape the selection function, and the correction is controlled by the shape of $W_b(z)$ rather than by the sharpness of the bin.","Since the boundary terms are proportional to $\\beta$, they can aid a measurement of the kinematic boost, provided the redshift distribution and selection function are estimated accurately.","Cuts on color generate the same class of correction through a redshift selection function when source spectra are not perfect power laws; the effect cancels for exact power-law spectra."],"supporting_citations":[{"why":"Supplies the baseline Ellis-Baldwin dipole formula that the boundary terms modify.","marker":"[17]"},{"why":"Gives the redshift-dependent kinematic dipole expression used as the starting point.","marker":"[29]"},{"why":"Shows the alternative redshift-dipole form whose integration by parts exposes the boundary terms.","marker":"[30]"},{"why":"Establishes the general equivalence between the luminosity-function and observed-quantity approaches; dropping its boundary-vanishing assumption opens the present derivation.","marker":"[33]"},{"why":"Supplies the forecast galaxy redshift distribution parameters and photometric redshift bias model used in the worked examples.","marker":"[45]"},{"why":"Supplies the photometric survey redshift distribution parameters and binning choices used in the forecasts.","marker":"[46]"},{"why":"Supplies the radio survey redshift distribution parameters used for the forecasts.","marker":"[49]"},{"why":"Motivates the low-redshift cut whose boundary-term impact the paper computes.","marker":"[52]"}],"fun_headline_variants":["Redshift bin edges can flip the matter dipole direction","Doppler boosting of redshifts can reverse the kinematic dipole","Selection functions can dominate and reverse the cosmic dipole","Bin selection adds a term that can flip the dipole's sign","Redshift cuts can reverse the direction of the matter dipole"],"cache_read_input_tokens":3200,"weakest_assumption_plain":"The derivation assumes that the redshift selection function $W_b(z)$ is independent of the flux cut $S_*$; if that fails, as it can for photometric redshifts calibrated from flux-limited spectroscopic subsamples, the simple boundary-term formula no longer applies and a much harder computation is needed.","fun_headline_variants_meta":{"raw":{"variants":["Redshift bin edges can flip the matter dipole direction","Doppler boosting of redshifts can reverse the kinematic dipole","Selection functions can dominate and reverse the cosmic dipole","Bin selection adds a term that can flip the dipole's sign","Redshift cuts can reverse the direction of the matter dipole"]},"model":"deepseek-v4-flash","effort":"low","cost_usd":0.0002,"raw_usage":{"total_tokens":1383,"prompt_tokens":960,"completion_tokens":423,"prompt_tokens_details":{"cached_tokens":384},"prompt_cache_hit_tokens":384,"prompt_cache_miss_tokens":576,"completion_tokens_details":{"reasoning_tokens":345}},"tokens_in":576,"tokens_out":423,"duration_ms":4324,"temperature":1.0,"reasoning_tokens":345,"cache_read_input_tokens":384,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-08-11T13:22:28.201821+00:00","model_set":{"reader":"deepseek-v4-flash"},"falsifier":"Simulate a full-sky catalogue with a known boost $\\beta$ and no intrinsic clustering dipole, bin it in observed redshift with a known selection function, and compare the measured per-bin dipole amplitudes with the prediction of Eq. (20) using the true $\\tilde{x}_b$ and $\\tilde{\\alpha}_b$; disagreement in the bin-to-bin pattern of amplitudes would falsify the boundary-term claim. The paper's own 1000-mock comparison performs this same check.","supporting_citations":[],"review_version":1}