Low-redshift-agnostic BAO Constraints on Binned Dark-energy Density Evolution from DESI DR1 and DR2

Qian-Mo Liu, Gong-Bo Zhao · 2026

Read Low-redshift-agnostic BAO Constraints on Binned Dark-energy Density Evolution from DESI DR1 and DR2 by Qian-Mo Liu, Gong-Bo Zhao (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-2604-06888.txt.

Introduction

Liu and Zhao build a low-redshift-agnostic squeeze of DESI sound-wave spacings for dark energy density growth. Sound-wave numbers mix low-redshift growth history into higher-redshift signals through transverse spacings. The authors swap those spacings for step-to-step increments while keeping radial spacings, so errors stay exact and growth bins come out nearly uncorrelated. Dropping one absolute transverse mode strips low-redshift bias from the fit, and on DESI DR1 and DR2 data, every growth bin matches a constant density within errors.

Catalog record

CREATORQian-Mo Liu, Gong-Bo Zhao
YEAR2026
COLLECTIONPAPERS
FORMATTXT
FILE SIZE24125
LICENSEhttp://creativecommons.org/licenses/by/4.0/
RIGHTSCC BY
SOURCEarXiv

Source: https://arxiv.org/abs/2604.06888

Browse the full library · About the Tower of Babel

Full text

The opening of the work, from the library's hosted text:

Low-redshift-agnostic BAO Constraints on Binned Dark-energy Density Evolution from DESI DR1 and DR2 Qian-Mo Liu, Gong-Bo Zhao Abstract We present a low-redshift-agnostic compression of anisotropic baryon acoustic oscillation (BAO) distances to constrain the normalized dark-energy density evolution, X ⁡ ( z ) ≡ ρ DE ​ ( z ) / ρ DE ​ ( 0 ) X(z)\equiv\rho_{\rm DE}(z)/\rho_{\rm DE}(0) , above the lowest BAO redshift node z 1 z_{1} . Standard BAO summaries include the transverse comoving distance D M ​ ( z ) / r d D_{\!M}(z)/r_{d} , which depends on the integral of H − 1 ​ ( z ) H^{-1}(z) from z = 0 z=0 to z z and therefore mixes the expansion history at z < z 1 z<z_{1} with the higher-redshift signal. We instead replace the set { D M ​ ( z i ) / r d } \{D_{\!M}(z_{i})/r_{d}\} by adjacent increments { Δ ​ D M ​ ( z i , z i + 1 ) / r d } \{\Delta D_{\!M}(z_{i},z_{i+1})/r_{d}\} while retaining the radial distances { D H ​ ( z i ) / r d } \{D_{\!H}(z_{i})/r_{d}\} . The mapping is linear, so the covariance propagates exactly. This compression intentionally removes one absolute transverse-distance mode, namely the additive contribution to D M / r d D_{\!M}/r_{d} below the first BAO node, and preserves the remaining information relevant to reconstructing the expansion history above z 1 z_{1} . Applied to DESI DR1 and DR2 anisotropic BAO measurements, the method yields almost uncorrelated constraints on piecewise-constant interval parameters X j X_{j} . In this sense, the compressed likelihood provides a conservative band-power-like estimate of dark-energy evolution: each interval is constrained mainly by BAO information from its own redshift range, while one nonlocal transverse mode and stronger global assumptions are deliberately projected out or marginalized over. Because our baseline analysis also marginalizes over bin-local matter-density and distance-scale parameters with broad external priors, the resulting X j X_{j} constraints should be interpreted as a low-redshift-agnostic BAO baseline rather than as a fully prior-free reconstruction. All bins are consistent with X = 1 X=1 within current uncertainties. \volnopage 20XX Vol. XX No. XX , 000–000 keywords dark energy — cosmological parameters — large-scale structure of Universe — baryon acoustic oscillations 1 Introduction The physical origin of cosmic acceleration remains one of the central open questions in modern cosmology. In the standard \Lambda CDM model the late-time expansion is driven by a cosmological constant, but a wide class of alternatives predict departures from a constant dark-energy density. A natural strategy is therefore to reconstruct the expansion history as flexibly as possible and to identify which redshift modes are actually constrained by the data ( Huterer & Starkman 2003 ; Wang & Tegmark 2004 ; Huterer & Cooray 2005 , e.g.,) . Recent DESI DR2 dark-energy studies have pushed this program further with both parametric and non-parametric reconstructions combined with supernova and CMB information ( Lodha et al. 2025 ; Gu et al. 2025 ) , providing an important broader context for the deliberately conservative BAO-based baseline developed here. Rather than adopting a specific functional form for w(z) , we focus on the normalized dark-energy density evolution, X(z)\equiv\frac{\rho_{\rm DE}(z)}{\rho_{\rm DE}(0)}, (1) and describe it in redshift bins. Such reconstructions are intrinsically challenging because distance measurements are nonlocal integrals of the expansion rate. As a result, naive binned reconstructions often develop strong bin-to-bin correlations even when the underlying data appear local in redshift. Baryon acoustic oscillations (BAO) provide a particularly robust geometric probe. The sound horizon at baryon drag, r_{d} , acts as a standard ruler calibrated by early-universe physics ( Eisenstein & Hu 1998 ; Eisenstein et al. 2005 ) . Anisotropic BAO analyses constrain both the transverse comoving distance D_{\!M}(z)/r_{d} and the radial distance D_{\!H}(z)/r_{d} . The standard transverse-distance summary, however, depends on the integral of H^{-1}(z) from 0 to z , so every D_{\!M}(z)/r_{d} point inherits an additive contribution from the poorly isolated low-redshift interval below the first BAO node. The goal of this paper is to construct a transparent BAO-based baseline that is agnostic to that low-redshift contribution. Our key step is to replace the set of transverse distances at BAO nodes by adjacent increments, \Delta D_{\!M}(z_{i},z_{i+1})\equiv D_{\!M}(z_{i+1})-D_{\!M}(z_{i}), (2) while retaining the measured D_{\!H}(z_{i})/r_{d} nodes. This linear transformation propagates the covariance exactly. It does not preserve the full original BAO summary vector: instead, it intentionally projects out one absolute transverse-distance mode, namely the additive contribution to D_{\!M}/r_{d} below the first BAO node. The remaining compressed data vector is the information relevant to reconstructing the expansion history above z_{1} . We combine this low-redshift-free compression with a piecewise-constant description of X(z) matched to the BAO redshift nodes. In our baseline implementation, each redshift interval is also assigned its own matter-density and distance-scale nuisance parameters. This choice is useful for localizing the inference, but it means that the inferred X_{j} should be interpreted as effective interval-wise dark-energy-density parameters rather than as direct samples of a single globally tied function X(z) . The advantage of this localized construction is that the resulting bins are nearly uncorrelated, so the final constraints behave much like band powers in redshift. They are also deliberately conservative: by projecting out the absolute low-redshift transverse mode and by not enforcing a globally tied cosmological model in the baseline fit, we sacrifice some constraining power in exchange for a cleaner, more weakly model-dependent interpretation. Likewise, the analysis is not fully prior-free, because the local nuisance sector is regularized by broad external priors. Our aim here is therefore modest and explicit: to provide a low-redshift-agnostic BAO baseline, clarify what information is being removed and retained, and quantify the resulting interval-wise constraints from current DESI BAO measurements. This paper is organized as follows. Section 2 introduces the compressed BAO observables, the node-matched binning for X(z) , and the likelihood used in our baseline analysis. Section 3 summarizes the DESI DR1 and DR2 BAO inputs. Section 4 presents the interval-wise constraints and their correlation matrices. Section 5 discusses the interpretation, limitations, and future extensions of the method. 2 Methodology 2.1 Background expansion and BAO observables We work in a spatially flat background and neglect radiation over the redshift range relevant to the BAO measurements used here. The dimensionless expansion history is E(z)\equiv\frac{H(z)}{H_{0}}, (3) and we write E^{2}(z)=\Omega_{m}(1+z)^{3}+(1-\Omega_{m})X(z), (4) where \Omega_{m} is the present-day matter density parameter and X(z) is the normalized dark-energy density evolution. Because BAO distances are reported in units of the sound horizon at baryon drag, it is convenient to define the nuisance combination S\equiv H_{0}\,r_{d}, (5) so that the overall distance scale appears as c/S . The anisotropic BAO observables are then \displaystyle\frac{D_{\!H}(z)}{r_{d}} \displaystyle=\frac{c}{S}\,\frac{1}{E(z)}, (6) \displaystyle\frac{D_{\!M}(z)}{r_{d}} \displaystyle=\frac{c}{S}\,\int_{0}^{z}\frac{dz^{\prime}}{E(z^{\prime})}. (7) Equation ( 7 ) makes the key nonlocality explicit: each transverse-distance point contains the full integral from z=0 to z . 2.2 Low-redshift-free compression Assume that anisotropic BAO measurements are available at a set of effective redshifts (“nodes”) \{z_{i}\}_{i=1}^{N},\qquad z_{1}<z_{2}<\cdots<z_{N}. (8) The standard BAO summary vector is \bm{x}=\left(\frac{D_{\!M}(z_{1})}{r_{d}},\frac{D_{\!H}(z_{1})}{r_{d}},\frac{D_{\!M}(z_{2})}{r_{d}},\frac{D_{\!H}(z_{2})}{r_{d}},\ldots,\frac{D_{\!M}(z_{N})}{r_{d}},\frac{D_{\!H}(z_{N})}{r_{d}}\right)^{\mathsf{T}}, (9) with covariance \bm{C}_{x} . To remove the additive low-redshift contribution below z_{1} , we replace the set \{D_{\!M}(z_{i})/r_{d}\} by adjacent increments, \Delta D_{\!M}(z_{j},z_{j+1})\equiv D_{\!M}(z_{j+1})-D_{\!M}(z_{j}),\qquad j=1,\ldots,N-1, (10) and define the compressed vector \bm{y}=\left(\frac{D_{\!H}(z_{1})}{r_{d}},\ldots,\frac{D_{\!H}(z_{N})}{r_{d}},\frac{\Delta D_{\!M}(z_{1},z_{2})}{r_{d}},\ldots,\frac{\Delta D_{\!M}(z_{N-1},z_{N})}{r_{d}}\right)^{\mathsf{T}}. (11) The dimension

Read the complete text