Cosmic Dipole as a Symmetry Response: From the Ellis–Baldwin Formula to Correlation Function Dipoles Preprint: APS/123-QED
Tsutomu T. Takeuchi · 2026
Read Cosmic Dipole as a Symmetry Response: From the Ellis–Baldwin Formula to Correlation Function Dipoles Preprint: APS/123-QED by Tsutomu T. Takeuchi (2026) in the Tower of Babel, a free online library of public-domain works preserved in clean, searchable text. No sign-up. The full text is served as a plain-text file at a predictable URL, where text is present: https://antarctic-labs.com/tower-of-babel/arxiv-2603-16776.txt.
Introduction
Takeuchi studies how the motion of the observer distorts pair statistics in cosmology. Motion couples source selection, clustering, and how estimators are normalized. He builds a general framework that allows marked pair kernels and pair inclusion that does not factorize. In a simple limit, the leftover effect becomes a covariance between the selection score and pair excess. He shows that two galaxy populations can share a count dipole yet respond to a boost in opposite ways. He also derives contact terms and self-pair fixes for discrete harmonic estimators.
Catalog record
| CREATOR | Tsutomu T. Takeuchi |
|---|---|
| YEAR | 2026 |
| COLLECTION | PAPERS |
| FORMAT | TXT |
| FILE SIZE | 40401 |
| LICENSE | http://creativecommons.org/licenses/by/4.0/ |
| RIGHTS | CC BY |
| SOURCE | arXiv |
Source: https://arxiv.org/abs/2603.16776
Browse the full library · About the Tower of Babel
Full text
The opening of the work, from the library's hosted text:
Cosmic Dipole as a Symmetry Response: From the Ellis–Baldwin Formula to Correlation Function Dipoles Preprint: APS/123-QED Tsutomu T. Takeuchi Abstract The cosmic dipole in galaxy number counts is traditionally described by the Ellis–Baldwin (EB) formula under simplifying assumptions of power-law source counts and flux-limited selection. We reformulate the EB dipole as a symmetry response of observed counts to a Lorentz boost, leading to the general expression D = β R D=\beta R , where R = ∂ ln N / ∂ ln β R=\partial\ln N/\partial\ln\beta encodes the underlying population and selection effects. The classical EB formula is recovered as a limiting case. We show that this response framework extends beyond one-point statistics: Lorentz boosts induce a dipole component in the two-point correlation function and, more generally, a hierarchy of responses in n n -point statistics. We further clarify the relation to redshift-space distortions and relativistic galaxy clustering, and provide a unified description in which observer- and source-induced dipoles contribute to the same multipole component. This establishes the cosmic dipole as a symmetry response of finite-sample point-process statistics, offering a new perspective on dipole anisotropies and their observational interpretation. I Introduction Dipole anisotropy provides a direct probe of the global isotropy of the Universe. The dipole of the cosmic microwave background (CMB) is conventionally interpreted as a kinematic effect arising from the observer’s peculiar motion. This motion is expected to induce a corresponding dipole in the angular distribution of astrophysical sources. Motivated by this expectation, number-count dipoles in radio surveys and galaxy catalogs have long been compared with the CMB dipole. In this context, the Ellis–Baldwin (EB) formula ( 9 ) provides a standard description of the kinematic dipole in source counts. Under the assumptions of power-law number counts and a single spectral index, it yields a simple prediction for the dipole amplitude. However, realistic samples exhibit non-power-law populations and multivariate selection functions, including flux limits, color cuts, frequency bands, survey masks, and spatially varying depths. The classical EB expression should therefore be regarded as an idealized limit. Recent observations suggest that measured dipole amplitudes may exceed the CMB-based kinematic prediction, leading to the so-called cosmic dipole anomaly. A comprehensive review is given by 21 . In particular, mid-infrared quasar samples consistently show excess dipole amplitudes ( 19 ; 20 ; 8 ; 25 ; 24 ) . Although its origin remains debated, this indicates that the observed dipole cannot be treated as purely kinematic. It receives contributions from observer motion, large-scale structure, and survey-dependent systematics and selection effects. This motivates a formulation in which the dipole is treated as a statistical observable depending explicitly on population and selection. On the theoretical side, relativistic galaxy clustering shows that the observed number density contrast includes density perturbations, redshift-space distortions, Doppler terms, gravitational potentials, and lensing ( 26 ; 3 ; 1 ; 2 , e.g.,) . These generate anisotropies in the correlation function, including odd multipoles such as the dipole ( 4 ; 17 ) . Observer motion has also been studied via Lorentz boosts and multipole coupling ( 5 ; 6 ; 13 ) . These approaches derive dipole signals from dynamical and relativistic effects along the light cone. The present work provides a complementary perspective. Rather than treating the dipole as a dynamical perturbation, we formulate it as a symmetry response of finite-sample statistics under Lorentz transformations. The dipole is then defined as the response of an observable to a continuous symmetry via a logarithmic derivative with respect to the boost parameter. This formulation naturally incorporates selection effects and extends beyond idealized assumptions. While many of the dipole contributions discussed above have been derived in relativistic galaxy clustering and light-cone analyses, the present approach differs in its conceptual focus. Rather than introducing new physical effects, we provide a unifying interpretation in which these signals are understood as symmetry responses of finite-sample statistics under Lorentz transformations. This organizes existing results into a coherent framework and extends naturally to higher-order statistics. The goal of this paper is to reformulate the Ellis–Baldwin dipole as a response of number counts to a Lorentz boost and to extend this framework to finite point processes and hierarchical correlation statistics. We introduce a generalized response coefficient incorporating population and selection, and recover the classical EB expression as a limiting case. We express this response in terms of the generating functional of point processes, identifying it as a symmetry response analogous to a Ward identity. Extending the formalism to two-point statistics, we show that a Lorentz boost induces a dipole component in the correlation function, and we decompose it into observer and source contributions. The main results are as follows. First, the EB dipole is reinterpreted as a symmetry response including selection effects. Second, this extends naturally to higher-order statistics, yielding a hierarchy of responses \displaystyle R_{n}=\frac{\partial\ln\langle N_{n}\rangle}{\partial\ln\beta}. (1) Third, the dipole moment of the correlation function emerges as the two-point analogue of the EB effect, providing a unified description across statistical orders. The structure of this paper is as follows. In Sec. II , we reformulate the EB dipole including selection effects. In Sec. III , we interpret it as a symmetry response using the generating functional of point processes. In Sec. IV , we extend the framework to higher-order statistics. In Sec. V , we formulate the dipole of the correlation function and relate it to redshift-space distortions and relativistic clustering. In Sec. VI , we present a unified description of observer and source dipoles. Finally, Sec. VII summarizes our conclusions. Technical details are given in the Appendices. Appendix A evaluates the response coefficient for representative selection functions, and Appendix B derives the dipole of the correlation function. II Generalized Ellis–Baldwin Dipole as a Response to Lorentz Boost The kinematic dipole in astrophysical source samples is classically described by the Ellis–Baldwin (EB) formula derived by 9 . If the cumulative source counts follow a power law, \displaystyle N(>S)\propto S^{-x}, (2) and the spectrum is described by a single power law, \displaystyle S\propto\nu^{-\alpha}, (3) then the dipole amplitude in the number density induced by the observer velocity \beta=v/c is given by \displaystyle D=\beta[2+x(1+\alpha)]. (4) This expression has been widely used, especially in radio surveys. However, realistic observational samples generally do not follow simple power-law population distributions, and their selection functions are intrinsically multivariate. The EB formula should therefore be regarded as an idealized limit rather than a fully general description. In this paper, we reformulate the EB dipole as a response of the observed counts to a Lorentz boost . From this viewpoint, the classical EB formula appears as a special limit of a more general response formalism. In this section, we introduce only the basic definitions needed in the present paper, while the detailed derivation is deferred to the companion work ( 22 ) . II.1 Counts and selection function The total number of objects in an observed sample is written as \displaystyle N=\int f(\bm{y})\,\mathcal{W}(\bm{y})\,{\rm d}\bm{y}, (5) where f(\bm{y}) is the population distribution and \mathcal{W}(\bm{y}) is the selection function. Here \bm{y} denotes the set of observables, for example \displaystyle\bm{y}=(S,\nu,{\rm color},z,\ldots), (6) including flux, frequency, color, redshift, and other measured quantities. If the observer moves with velocity \bm{v} , the observables are transformed by