Cosmological constraints from a joint DESI DR1 Full-Shape and DR2 BAO

Abstract

We present a cosmological analysis combining full-shape (FS) clustering measurements from the Dark Energy Spectroscopic Instrument (DESI) DR1 with baryon acoustic oscillation (BAO) measurements from DESI DR2. To achieve a robust combination that accounts for the correlation between the two data releases, we employ the ShapeFit compression method and estimate the joint covariance using EZmocks. This compressed approach inherently mitigates the prior volume effects that have previously dominated Bayesian constraints from DESI data with minimal external priors. Consequently, we obtain--for the first time within a Bayesian framework--reliable DESI-only constraints on extensions to $Λ$CDM using only a Big Bang Nucleosynthesis prior on the baryon density and a wide prior on the spectral index. In flat $Λ$CDM, we find $Ω_m = 0.3035 \pm 0.0085$, $h = 0.6876 \pm 0.0059$, and $σ_8 = 0.822 \pm 0.034$. For the $w_0 w_a$CDM dynamical dark energy model, we measure $w_0 = 0.49 \pm 0.25$ and $w_a = -1.52 \pm 0.77$, improving constraints by $\sim 30\%$ relative to the analogous DR1 measurement and reducing the discrepancy with $Λ$CDM to $1.4σ$ when compared to BAO only analyses. We also report competitive limits on the sum of neutrino masses and spatial curvature. This work demonstrates that the ShapeFit compression provides a prior-robust and computationally efficient pathway to constrain beyond-$Λ$CDM physics with large-scale structure.

Figures (14)

• Figure 1: Correlation matrices of the compressed Full-Shape $\times$ BAO data vector. Top triangle: Matrix elements estimated from DR1(2) ShapeFit (BAO) posterior samples, i.e. without the cross-correlation terms ($\mathbf{r}^{\rm BAO+ SF}_{\rm B}$). Bottom triangle: Matrix elements estimated in this work using mock ensemble, thus including the cross correlations ($\mathbf{r}^{\rm BAO+ SF}_{\rm A}$).
• Figure 2: Distribution of errors for each of the six compressed parameters measured from the 100 EZmocks (histograms) compared to the actual data errors (vertical lines), for different tracers. The quantity shown corresponds to the $Z$-score component, $g(\theta) = \qty(\sigma_\theta - \bar{\sigma}_\theta) / \mathrm{std}(\sigma_\theta)$, where $\bar{\sigma}_\theta$ denotes the mean error on parameter $\theta$ over the EZmocks and $\mathrm{std}(\sigma_\theta)$ the corresponding standard deviation. Each panel corresponds to one parameter-sample combination, the shaded area shows the $2\sigma$ region. Given that the combined ELG1+LRG3 sample was not used in the FS analyses, these panels are empty.
• Figure 3: Comparison of different marginalised cosmological posteriors for different covariance choices (see section \ref{['sec:methods_combination']}) within the $\Lambda$CDM model. We show our baseline covariance matrix choice (C) at the bottom row and show its corresponding 1 and 2$\sigma$ limits as shaded grey areas and the mean as a vertical line, to facilitate the comparison with the other two approaches.
• Figure 4: Comparison of our results (DESI1.5) to previous DESI1 and BAO2 analyses in the context of $\Lambda$CDM. Constraints from DESI1 using FS constraints, both from FM and SF, are also shown for reference.
• Figure 5: Marginalized 1D posteriors under $w_0w_a$CDM. We also show the corresponding MAP values for each (FS) likelihood as a triangular symbol. The systematic shift between the MAP and the posterior in DESI1 (FM) shows the presence of significant prior volume effects in combinations with BBN and CMB. Our results, i.e. DESI1 (SF) and DESI1.5, do not show these effects as the MAP overlap the posterior mass. We plot the corresponding BAO constraints for reference.