New Expansion Rate Anomalies at Characteristic Redshifts Geometrically Determined using DESI-DR2 BAO and DES-SN5YR Observations

TL;DR

This work performs a model-independent reconstruction of cosmic distances by combining DESI-DR2 BAO and DES-SN5YR observations with two non-parametric approaches (MTGP and knot-based splines) to identify seven characteristic redshifts. At these z_i, the reconstructed Hubble parameter E(z_i) and related distances show significant deviations (up to 4–5σ) from Planck 2018 ΛCDM, especially around z ~ 0.35–0.55, suggesting potential late-time new physics or unaccounted systematics. The analysis includes rigorous cross-checks, tests for DESI DR2 BAO anomalies, and a validation of the distance-duality relation, with results indicating robustness to data subsets and consistency between SN and BAO. The findings motivate future high-precision probes (DESI-5YR, Euclid, LSST) to determine whether these anomalies reflect evolving dark energy or unknown systematics, thereby probing the late-time expansion history beyond ΛCDM.

Abstract

We perform a model-independent reconstruction of the cosmic distances using the Multi-Task Gaussian Process (MTGP) framework as well as knot-based spline techniques with DESI-DR2 BAO and DES-SN5YR datasets. We calibrate the comoving sound horizon at the baryon drag epoch $r_d$ to the Planck value, ensuring consistency with early-universe physics. With the reconstructed cosmic distances and their derivatives, we obtain seven characteristic redshifts in the range $0.3 \leq z \leq 1.7$. We derive the normalized expansion rate of the Universe $E(z)$ at these redshifts. Our findings reveal significant deviations of approximately $4$ to $5σ$ from the Planck 2018 $Λ$CDM predictions, particularly pronounced in the redshift range $z \sim 0.35-0.55$. These anomalies are consistently observed across both reconstruction methods and combined datasets, indicating robust late-time tensions in the expansion rate of the Universe and which are distinct from the existing "Hubble Tension". This could signal new physics beyond the standard cosmological framework at this redshift range. Our findings underscore the role of characteristic redshifts as sensitive indicators of expansion rate anomalies and motivate further scrutiny with forthcoming datasets from DESI-5YR BAO, Euclid, and LSST. These future surveys will tighten constraints and will confirm whether these late-time anomalies arise from new fundamental physics or unresolved systematics in the data.

Figures (5)

• Figure 1: Evolution of the cosmic distances: $d_A(z)$, $d_M(z)$ and $d_L(z)$, and their derivatives ($d_A'(z)$, $d_M'(z)$ and $d_L'(z)$), for the Planck 2018 baseline $\Lambda$CDM model. Here, $z_i$ ($i = 1 \cdots 7$) are the characteristic redshifts.
• Figure 5: Reconstruction of $D_M(z)$, $D_M'(z)$ and $D_M"(z)$, employing (i) MTGP reconstruction with RQD kernel [in upper panel], and (ii) Free-form Knot-based spline reconstruction with $k=4$ order [in lower panel] respectively.
• Figure 11: Evolution of $\eta(z)$ across the effective redshift bins ($z_{\rm eff}$) of the DESI-DR2 BAO tracers that overlap with the redshift range covered by the DES-5YR SN-Ia sample, derived from the reconstructed $d_L(z)$ from DES-5YR SN-Ia and $d_A(z)$ from DESI-DR2 BAO measurements, employing STGP (in left panel) and free-form knot reconstruction at characteristic redshift knots.
• Figure 12: Joint posterior distributions for $D_M^{(i)}$, $i = 1, \ldots, 7$, at the characteristic redshift knots $z_i$, reconstructed using a knot-based spline method. We compare three cases: I (full data), II (70% randomly selected DES-SN5YR data), and III (40% randomly selected DES-SN5YR data), all combined with DESI DR2 BAO measurements. The strong overlap in posteriors indicates statistical consistency among the cases.
• Figure 13: Reconstructed evolution of the dimensionless comoving angular diameter distance $D_M(z)$ for the three data subsets: I (full), II (70%), and III (40%) in the top panel. Residuals showing the uncertainty bands of the reconstructions with respect to the data in bottom panel. The reconstructions show excellent agreement, confirming the robustness of the method under data resampling.